2013 phd thesis king’s college london andrea corrado profeta by Andrea Corrado Profeta

King’s College London!

Thesis submitted for the degree of PhD Andrea Corrado Profeta BDS Hons Department of Restorative Dentistry Biomaterials Science, Biomimetics and Biophotonics (B3) Research Group King’s College London Dental Institute at Guy’s, King’s College and St Thomas’ Hospitals

MMXIII

This electronic theses or dissertation has been downloaded from the Kingâ&#x20AC;&#x2122;s Research Portal at https://kclpure.kcl.ac.uk/portal/

Title:Hybridisation of dental hard tissues with modified adhesive systems: therapeutic impact of bioactive silicate compounds on bonding to dentine

Author:Andrea Corrado Profeta The copyright of this thesis rests with the author and no quotation from it or information derived from it may be published without proper acknowledgement.

END USER LICENSE AGREEMENT This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License. http://creativecommons.org/licenses/by-nc-nd/3.0/ You are free to: Share: to copy, distribute and transmit the work Under the following conditions: Attribution: You must attribute the work in the manner specified by the author (but not in any way that suggests that they endorse you or your use of the work). Non Commercial: You may not use this work for commercial purposes. No Derivative Works - You may not alter, transform, or build upon this work.

Any of these conditions can be waived if you receive permission from the author. Your fair dealings and other rights are in no way affected by the above.

Take down policy If you believe that this document breaches copyright please contact librarypure@kcl.ac.uk providing details, and we will remove access to the work immediately and investigate your claim.

The copyright of this thesis rests with the Author and no quotation from it or information derived from it may be published without proper acknowledgement. Copies (by any process) either in full, or of extracts, may be made only in accordance with instructions given by the Author and lodged in the Maughan Library of King’s College London. Details may be obtained from the Librarian. This page must form part of any such copies made. Further copies (by any process) of copies made in accordance with such instructions may not be made without the permission (in writing) of the Author. The ownership of any intellectual property rights which may be described in this thesis is vested in King’s College London, subject to any prior agreement to the contrary, and may not be made available for use by third parties without the written permission of the University, which will prescribe the terms and conditions of any such agreement. Further information on the conditions under which disclosures and exploitation may take place is available online at the College institutional repository: http://www.kcl.ac.uk/library/visiting/maughan.aspx

Recommended Citation: Hybridisation of dental hard tissues with modified adhesive systems: therapeutic impact of bioactive silicate compounds on bonding to dentine. Profeta AC. PhD Thesis 2013. King’s College London, Strand, London WC2R 2LS, England, United Kingdom.

Hybridisation of dental hard tissues with modified adhesive systems: therapeutic impact of bioactive silicate compounds on bonding to dentine

Andrea Corrado Profeta Bachelor of Dental Surgery BDS Hons Università Cattolica del Sacro Cuore (UCSC) Class of 2006

Thesis submitted for the degree of Doctor of Philosophy PhD in Clinical Dentistry King’s College London (KCL)

Department of Restorative Dentistry Biomaterials Science, Biomimetics and Biophotonics (B3) Research Group KCL Dental Institute at Guy’s, King’s College and St Thomas’ Hospitals 2013

I dedicate this work to the adversities that made it so worthwhile

Structure of the thesis, objectives and working plan

The first section of this work is a review of the literature necessary to understand the objectives of the project; it includes general information about dental adhesive technology as well as adhesion testing, about dentine hybridisation and about the drawbacks of contemporary bonding systems. Several studies revealed excellent immediate and short-term bonding effectiveness of etch-and-rinse adhesives, yet substantial reductions in resindentine bond strength occur after ageing. Degenerative phenomena involve hydrolysis of suboptimally polymerised hydrophilic resin components and degradation of mineral-deprived water-rich resin-sparse collagen matrices by matrix metalloproteinases and cysteine cathepsins. Silicate compounds, including calcium/sodium phosphosilicates, such as commercially available bioactive glass, and calcium-silicate Portland-derived cements are known to promote the formation of apatite in aqueous environments that contain calcium and phosphate (e.g. saliva); thus, we have raised questions about whether their presence at the bonded interface could increase the in vitro durability of resin-dentine bonds through crystal formation and self-sealing, in the presence of phosphate buffered saline or simulated body fluid solutions. In answering these questions, the objectives were accomplished by employing Bioglass® 45S5 in etch-and-rinse bonding procedures either (i) included within the composition of a resin adhesive as a tailored micro-ﬁller, or (ii) applied directly onto acid-etched wetted dentine. Alternative light-curable methacrylate-

based agents containing (iii) three modified calcium-silicates derived from ordinary Portland cement were also tested. Confirming the relative success of bioactive materials incorporated in the dentine bonding procedures required assessment of the potential to reduce nano-leakage, as well as their effect upon the strength of the bond over time. In order to explore these possibilities, which have not been previously investigated, a combination of methods were applied in the second experimental section. Bond strength variations were quantified using the microtensile test while scanning electron microscopy, confocal laser scanning microscopy and Knoop micro-indentation analysis were used to evaluate optically and mechanically adjustments to mineral and water content within the resin bonded-dentine interface. Initially, high microtensile values were achieved in each tested group. All the resin-dentine interfaces created with bonding agents containing micro-fillers showed an evident reduction of nano-leakage and mineral deposition after the ageing period. However, only adhesive systems containing Bioglass and two modified Portland cement-based microfillers were found to reduce nano-leakage with no negative effects on bond strength. Furthermore, specimens created with the same experimental adhesives did not restore micro-hardness to the level of sound dentine but were able to maintain statistically unaltered Knoop values. The second section is also composed of a set of preliminary studies that involved the use of up-to-date spectroscopic (attenuated total reflection Fourier transform infrared spectroscopy) and thermoanalytical (differential scanning calorimetry) techniques to predict the chemical-physical properties and apatiteforming ability of the novel ion-leachable hybrid materials. Lastly, the overall

conclusions of the present work and directions for future research are discussed.

Acknowledgements According to Merriam-Webster's dictionary, adversity means “a state, condition, or instance of serious or continued difficulty or adverse fortune” while triumph denotes “a great victory or success.” In any case, it is impossible to experience a sense of triumph over adversity unless you have first stared the possibility of disaster in the face. The taste of success means little unless you have a hint of the flavour of failure to compare it to. Acts of great courage are only taken after terrifying fears have been acknowledged and understood. Against almost everyone’s predictions, this thesis is respectfully submitted to Professor Dianne Rekow, Dean of the Dental Institute at King’s College London (KCL), and to Professor Tim Watson, Head of the Institute’s Biomaterials, Biomimetics and Biophotonics (B3) Research Group. The Dental Institute at KCL is full of talented, masterful and honourable people. I am proud to have been part of the B3 team and lucky to know so many brilliant clinicians and scientists. I wish past and present staff members who interacted with me throughout this project all the best; most especially, I would like to place on record my thanks to Professors Alistair Lax and Gordon Proctor for their direct involvement in bringing it to a successful conclusion. Also, I would like to extend my appreciation to Dr. Richard Foxton for his assistance in the academic and administrative requirements involved in my candidacy. Of course I am grateful to my family for their unconditional support in everything I choose to do and obsess over. Special mention to Agnė for helping me going through all those years, and for so much more... She knows the kind of pandemonium I endured in my life and that completing this work was a pretty big deal for me. Something I am glad I experienced, but would never welcome back again. Should somebody else ask me now, ‘Did you enjoy your PhD?’ ‘Did you use your time wisely?’ I will not hand over a piece of paper with the CV and other achievements on it to use up most of the alphabet after my name, or give an explanation of why I might be better than others. It is not, at least for me, about looking back or looking down, about titles, honorifics and status. I am simply going to stand up and smile a smile which lets people know I have no regrets at all. I was eager to be faced with all this experience had to offer, the intensity and unique opportunity to do things at the highest level, and discover what it might show me about myself. Unexpectedly my world was turned upside-down, my trust tested and my ego crushed. I had to be twice as good, three times as sharp, four times as focused than all the other PhD candidates. I had to prove myself ten times over but I never gave up and I succeeded where others failed. I can look at this record now and think how far I have come, and how far I have grown and also how grateful I am for all those experiences, regardless of how difficult they were at the time. Things I can take with me wherever I go, essential ingredients in a better me which can never be taken away, not just material goods I own briefly. The latin saying NIL DIFFICILE VOLENTI has certainly proved true for me and I am sure it will hold true for anyone who believes it.

List of contents

Structure of the thesis, objectives and working plan.............................

Acknowledgements...................................................................................

List of Figures............................................................................................

List.of Tables..............................................................................................

Section I - A review of the literature.........................................................

Chapter 1: Adhesive technology and dentine bonding limitations................................................................................

1.1 Introduction............................................................................................

1.1.1 Coupling resin monomers to enamel...........................................

1.1.2 Adhesion to dentinal substrates...................................................

1.2 Development of dentine-resin bonding technology................................

1.2.1 Early dentine bonding agents.......................................................

1.2.2 Smear-layer removal and acid conditioning……………………….

1.2.3 Dentine hybridisation and resin-infiltrated smear-layer................

1.3 Physico-mechanical considerations of resin-bonded dentine................ 1.3.1 Wettability of dentinal surfaces and contact angle....................... 1.3.2 Solubility of adhesive monomers.................................................

35 36 39

1.3.3 Permeability of the collagen network and monomers diffusivity....................................................................

1.3.4 Permeability of adhesive resins and water sorption.....................

1.4 Mechanisms responsible for loss of mechanical stability.......................

1.4.1 Hydrolytic degradation of dental adhesive resins.........................

1.4.2 Endogenous collagenolytic activity..............................................

1.5 Adhesion testing.....................................................................................

1.5.1 Assessment of sealing ability.......................................................

1.5.1.1 Micro-leakage and micro-permeability............................

1.5.1.2 Nano-leakage..................................................................

1.5.2 Bond strength measurement........................................................

1.5.2.1 Macro-bond strength test................................................

1.5.2.2 Micro-bond strength test.................................................

1.6 Classification of contemporary bonding systems................................... 1.6.1 Etch-and-rinse.............................................................................. 1.6.2 Self-etch....................................................................................... 1.6.3 Self-adhesive...............................................................................

71 72 75 82

Chapter 2: Strategies for preventing resin-dentine bond degradation..............................................................................

2.1 Introduction............................................................................................

2.1.1 Improvement of degree of conversion and esterase resistance......................................................................................

2.1.2 Inhibition of enzyme-catalysed hydrolytic cleavage of collagen..................................................................................... 2.1.3 Use of collagen cross-linking agents............................................. 2.1.4 Ethanol-wet bonding technique.....................................................

90 96 102

2.1.5 Restoring the mineral phase of the collagen matrix…………………...…………………………………………….

105

2.1.5.1 Guided tissue remineralisation.........................................

108

2.1.5.2 Top-down remineralisation via epitaxial growth…….……

114

2.1.5.3 Key objectives in the design of bioactive dentine bonding systems..............................................................

122

2.2 Development of ion-releasing adhesives comprising bioactive fillers........................................................................................ 2.2.1 Calcium/sodium phosphate-phyllosilicates fillers.......................... 2.2.2 Filler phase consisting of calcium silicate cements.......................

124

128 133

2.2.3 Dye-assisted confocal microscopy imaging of remineralised hard tissues............................................................

137

2.2.4 Aims of the study...........................................................................

141

Section II - Experimental projects............................................................

143

Chapter 3: Chemical-physical properties and apatite-forming ability of experimental dental resin cements containing bioactive fillers..................................................... 3.1 Introduction............................................................................................ 3.2 Materials and methods...........................................................................

144

145 147

3.2.1 Experimental micro-fillers and resin blends formulation.................................................................................... 3.2.2 Specimen preparation...................................................................

147 150

3.2.3 Water sorption and solubility evaluation........................................

151

3.2.4 Differential scanning calorimetry (DSC)........................................

152

3.2.5 Statistics........................................................................................

153

3.2.6 ATR-FTIR spectroscopy................................................................

153

3.3 Results...................................................................................................

154

3.3.1 Water sorption and solubility evaluation.......................................

154

3.3.2 Differential scanning calorimetry (DSC).......................................

157

3.3.3 ATR-FTIR spectroscopy...............................................................

159

3.4 Discussion.............................................................................................. 3.5 Conclusion.............................................................................................

164 169

Chapter 4: Bioactive effects of a calcium/sodium phosphosilicate on the resin-dentine interface: a microtensile bond strength, scanning electron microscopy, and confocal microscopy study...................................................................

170

4.1 Introduction............................................................................................

171

4.2 Materials and methods...........................................................................

172

4.2.1 Specimen preparation..................................................................

172

4.2.2 Experimental bonding procedures and formulation of resin adhesives......................................................................... 4.2.3 ÎźTBS and SEM fractography and failure analysis.........................

173

178

4.2.4 Confocal microscopy ultramorphology and nano-leakage evaluation...................................................................................... 4.3 Results...................................................................................................

179 182

4.3.1 μTBS and SEM fractography and failure analysis……..………….

182

4.3.2 Confocal microscopy ultramorphology and nano-leakage evaluation.......................................................................................

186

4.4 Discussion..............................................................................................

189

4.5 Conclusion.............................................................................................

195

Chapter 5: Experimental etch-and-rinse adhesives doped with calcium silicate-based micro-fillers to generate therapeutic bioactivity within resin-dentine interfaces................................................................................. 5.1 Introduction............................................................................................ 5.2 Materials and methods...........................................................................

196 197 199

5.2.1 Preparation of the experimental bioactive resin-base bonding agents............................................................

199

5.2.2 Specimen preparation and bonding procedures...........................

203

5.2.3 μTBS and SEM observations of the failed bonds..........................

205

5.2.4 Dye-assisted CLSM evaluation.....................................................

206

5.3 Results...................................................................................................

207

5.3.1 μTBS and SEM observations of the failed bonds..........................

207

5.3.2 Dye-assisted CLSM evaluation..................................................... 5.4 Discussion.............................................................................................. 5.5 Conclusion.............................................................................................

211 216 222

Chapter 6: In vitro micro-hardness of resin-dentine interfaces created by etch-and-rinse adhesives comprising bioactive fillers........................................................................

223

6.1 Introduction............................................................................................

224

6.2 Materials and methods...........................................................................

226

6.2.1 Teeth collection and preparation...................................................

226

6.2.2 Formulation of the comonomer resin adhesive blend………………………………………………………..

226

6.2.3 Bioactive fillers and experimental bonding systems......................................................................................... 6.2.4 Bonding procedures...................................................................... 6.2.5 Knoop micro-hardness (KHN) analysis......................................... 6.3 Results...................................................................................................

229

230 231 234

6.3.1 Knoop micro-hardness (KHN) analysis.........................................

234

6.4 Discussion..............................................................................................

237

6.5 Conclusion.............................................................................................

242

Chapter 7: General discussion and conclusion......................................

243

7.1 Summary................................................................................................

244

7.2 Research contributions..........................................................................

249

7.3 Recommendations for future research...................................................

Bibliography...............................................................................................

251

254

List of publications in international peer-reviewed journals as a result of this work..............................................................................

325

List of abstracts in international conferences of dental research from this work…............…….….….….….….…...........………....................

326

Appendix.....................................................................................................

327

List of Figures

Figure 1.1 - Crystal structure of biogenic hydroxyapatite.…………................ 24

Figure 3.1 - ATR-FTIR spectra of the unmilled comonomer blend, of Bioglass® 45S5, HOPC, HPCTO and HPCMM powders and of the hybrid experimental adhesives immediately after curing and following 60 days in DPBS……….. 162

Figure 4.1 - Schematic illustrating the experimental study design................. 176

Figure 4.2 - Schematic illustrating the composite-tooth matchsticks (1 mm) prepared using a water-cooled diamond saw, stored in PBS for 24 h or 6 months, and then subjected to microtensile bond strength (μTBS) testing and scanning electron microscopy failure analysis. This schematic also illustrates how composite-tooth slabs were prepared, stored in PBS for 24 h or 6 months, and evaluated by confocal laser scanning microscopy................................... 181

Figure 4.3 - Scanning electron microscopy images of failure modes of the resinbonded specimens created using the three different bonding approaches tested.............................................................................................................. 185

Figure 4.4 - Confocal laser scanning microscopy (CLSM) images showing the interfacial characterisation and nanoleakage, after 24 h of storage in PBS, of

the resin-dentine interfaces created using the three different bonding approaches tested......................................................................................... 187

Figure 5.1 - Chemical structures of the methacrylate monomers used in the tested resin blends.......................................................................................... 201

Figure 5.2 - Schematic illustrating the resin-dentine match-sticks prepared using a water-cooled diamond saw, stored in SBS for 24 h or 6 months, and then subjected to microtensile bond strength (ÂľTBS) testing and scanning electron microscopy fractography. This schematic also illustrates how composite-tooth slabs were prepared, stored in SBS for 24 h or 6 months, immersed in fluorescein (nanoleakage) or Xylenol Orange (Calcium-binding dye)

and

finally

evaluated

confocal

laser

scanning

microscopy

(CLSM)............................................................................................................ 204

Figure 5.3 - SEM failure analysis of debonded specimens............................ 210

Figure 5.4 - Confocal laser scanning microscopy (CLSM) single-projection images showing the interfacial characterisation and nanoleakage, after 24 h of storage in SBS................................................................................................ 213

Figure 5.5 - CLSM single-projection images disclosing the fluorescent calciumchelators dye xylenol orange.......................................................................... 214

Figure 5.6 - Confocal laser scanning microscopy (CLSM) single-projection images showing the interfacial characterisation and nanoleakage after 6 months of SBS storage................................................................................................ 215

Figure 6.1 - Optical images obtained during the micro-hardness test along the resin-dentine interface.................................................................................... 233

List of Tables

Table 3.1 - Chemical structures of the constituent monomers and composition (wt%) of the experimental adhesives used in this study................................ 149

Table 3.2 - Summary of maximum water uptake, solubility and net water uptake data................................................................................................................ 156

Table 3.3 - Means and standard deviations for Tg initially, after the ageing period

and

percentage

change

determined

DSC

analysis.......................................................................................................... 158

Table 4.1 - Composition of the experimental bonding procedures/adhesive systems used in this study............................................................................. 177

Table 5.1 - Chemical composition (wt%) and application mode of the experimental adhesive system used in this study.......................................... 202

Table 5.2 - Mean and standard deviation (SD) of the ÎźTBS (MPa) to dentine........................................................................................................... 209 17

Table 6.1 - Chemical composition (wt%) of the experimental adhesive systems used in this study........................................................................................... 228

Table 6.2 - The results of the micro-hardness measurements for each bonding system after 24 hours and 6 months of PBS storage.................................... 236

Section I - A review of the literature

Chapter 1: Adhesive technology and dentine bonding limitations

1.1 Introduction Adhesion or bonding is the process of forming an adhesive junction, which consists of two materials joined together. Any event described as adhesion is really an assembly involving a substrate (or ‘adherend’) with an applied ‘adhesive’ that creates an intervening ‘interface’. In reparative dentistry (Small, 2008), the adherends are enamel and dentine to which the adhesive is applied. Dental adhesives are solutions of resin monomers that join a restorative material with the tooth structure after their polymerisation is completed. While most adhesive joints involve only two interfaces, dental adhesive joints may be more complex such as the dentine-adhesive-composite interface of a bonded composite direct restoration. The aim is to create a close relationship between the dental substrate and restorative material, reproducing the natural relationship of the dental tissues, and to protect the pulp. Biomimetics, or imitating nature, is concerned with not only the natural appearance and aesthetic aspects of the restorations but the way they work. To copy nature is to understand the mechanics of the tooth, the way it looks and functions, and the way every stress is distributed. Ideally, the interface should provide a secure marginal seal and have the ability to withstand the stresses that have an effect on the bonding integrity of the adhesives, in order to keep the restoration adherent to the cavity walls. There are several sequential events that are necessary to form an effective adhesive joint. Bonding between hard tissues of the tooth and dental adhesive involves potential contributions from chemical (e.g., ionic bonds), physical (e.g., van der Waals) and mechanical sources but primarily relies on micro-mechanical interaction for success. For the development of strong adhesion, good wetting and intimate contact between the

adhesive and substrate, which must be clean and therefore in a high energy state, are required. Films of water, organic debris, and/or biofilms are always present in the clinical situation as dental surfaces that are prepared for restorative procedures, which present contaminants that remain on them. Consequently the steps for interface formation are the creation of a clean surface and the generation of a rough surface for interfacial interlocking.

1.1.1 Coupling resin monomers to enamel In 1955, Michael G. Buonocore reported the use of 85% phosphoric acid for 60 seconds in order to improve the retention of an acrylic resin on enamel (Buonocore, 1955, Simonsen, 2002). Still today adhesion to enamel is achieved by means of etching with 32-37% phosphoric acid, ideally in a gel form, for 1520 seconds. Acid etching creates microporosities, produces surface roughness, reduces the surface tension of the prepared surface and forms facets on the mineral crystals (Baier, 1992); this permits the hydrophilic monomers of the fluid adhesive resin to penetrate into the micro-retentive spaces in-between or within the enamel crystals. Accordingly, the micromechanical nature of the interaction of dental adhesives with enamel is a result of the infiltration of resin monomers into the microporosities left by the acid dissolution of enamel and subsequent enveloping of the exposed hydroxyapatite (HAp) crystals with the polymerised monomers (Swift et al., 1995). This makes it possible to obtain an adhesivecomposite-enamel bond strength able to resist a shearing force of more than 20 MPa, which is clinically remarkably effective (Swift et al., 1998).

1.1.2 Adhesion to dentinal substrates Whilst the adhesion to enamel, thanks to the etching technique, promptly demonstrated its efficacy, convincing in the following years researchers and clinicians, the same cannot be said for the adhesion to dentinal surfaces. The quest for an adequate dentine bonding agent has been longer and even today there is no confirmation of having attained the effectiveness which the adhesion to enamel has demonstrated. Enamel is composed of 96% HAp (mineral) by weight, whereas dentine contains a large percentage of organic material and water. It has been found that its bulk chemical composition is about 50% in volume made of mineral substance, 30% in volume formed of organic material, and the residue 20 vol% represented by water (Marshall et al., 1997). This tissue can be considered a biological composite that consists of a highly crosslinked and insoluble in acids collagen matrix filled with mineral crystallites located both within (intrafibrillar) and between (interfibrillar) collagen fibrils (Kinney et al., 2003). The mineral component is primarily a carbonated nanocrystalline hydroxyapatite whose structure is far different from stoichiometric hydroxyapatite, represented by the formula: Me10(XO4)6Y2 Where Me is a divalent metal (Ca2+, Sr2+, Ba2+, Pb2+ …), XO4 is a trivalent anion (PO43-, AsO43-, VO43- …), and Y is a monovalent anion (F-, Cl-, Br-, I-, OH-…). Given the unique mechanism involved in apatite crystal formation in biology, biogenic apatite varies in several ways from the corresponding geologically produced mineral.

First, biogenic apatite has a hexagonal lattice structure, having a strong ability to form solid solutions, and to accept numerous substitutions (Figure 1.1).

Figure 1.1 - Crystal structure of biogenic hydroxyapatite.

These substitutions affect the apatitic lattice parameters: the crystal size is decreased, and thereby the surface area is increased compared to stoichiometric HAp, thus permitting additional adsorption of ions and molecules on the apatite surface (LeGeros, 1991). Biological apatite contains in fact various trace elements from intrinsic or extrinsic origins, namely significant carbonate substitutions, OH - deficiencies, and imperfections in the crystal lattice (Boskey, 2007). This phenomenon

provides certain physico-chemical, biological, functional, and chemical features important in the formation and dissolution of the crystals in dental tissues. For example, F- ions are readily incorporated into dental apatite, forming fluoroapatite, a less soluble phase of calcium phosphate as compared to HAp, confering to enamel its low dissolution properties to resist acidic attacks. Likewise, trace elements present in extracellular fluids may have a specific role on mineral quality and condition. With respect to dentine apatite structure, this is represented by numerous substitutions (i) by hydrogenophosphate (HPO42-) of XO4 groups and (ii) by carbonate (CO32-) of Y2 and XO4 groups. Finally, biological minerals tend to attain high crystallinity and a more organised structure on the time scale of days or months rather than years (Verdelis et al., 2007). The dentine matrix is mainly composed of type-I collagen fibrils with associated noncollagenous proteins, to form a three-dimensional matrix that is reinforced by the apatite crystals (Marshall et al., 1997). Collagen microfibrils are described as those strands of collagen that are 5-10 nm in diameter, collagen fibrils are bundles of microfibrils that are 50-100 nm in diameter, and collagen fibres are bundles or networks of fibrils that are approximately 0.5-1 Â Îźm Â thick (Eick et al., 1997). This mineral-reinforced fibril composite is described by Weiner and Wagner as containing parallel platelike HAp crystals with their c-axis aligned with the long axis of the fibril (Weiner and Wagner, 1998). The location of these crystals in the fibril was demonstrated in a study by Traub and co-workers that showed

that mineralised collagen fibrils had the same banded pattern as negatively stained collagen fibrils (Traub et al., 1989). This indicated that mineral is concentrated in the hole zones of the fibril. It was proposed that these mineral platelets were arranged in parallel like a stack of cards within the interstices of the fibril (Palmer et al., 2008). The quality of dentine is dependent upon the total sum of characteristics of the tissue that influence its competence: microstructure, mineral density and especially the particular location of the mineral with respect to organic structures of the tissue. From a microstructural perspective, the collagen fibrils in dentine serve as a scaffold for mineral crystallites that reinforce the matrix, supporting the surrounding enamel. This microstructure suggests the necessity of a hierarchical approach to the understanding of its mechanical properties (Kinney et al., 2003). The mineral component incorporates oriented tubules that run continuously from the dentine-enamel junction (DEJ) to the pulp in coronal dentine, and from the cementum-dentine junction (CEJ) to the pulp canal in the root. Each tubule is encased in a collar of highly mineralised dentine, called peritubular dentine, embedded in the intertubular matrix (Marshall, 1993). These tubes are elongated cones with their largest diameters (ca. 3.0 Âľm) at the pulp and their smallest diameters (ca. 0.8 Âľm) at the DEJ, and are filled with a liquid that flows inside, with a pressure of about 20 mm of Hg (Van Hassel, 1971). The quantity of the tubules decreases from about 45,000 per mm2 in the proximity of the pulp to about 20,000 per mm2 near the dentino-enamel junction. As they converge on the pulp chamber, the surface area of the intertubular dentine diminishes while the tubule density augments, from about 1.9 x 106

tubules/cm2 at the DEJ to between 4.5 x 106 and 6.5 x 106 tubules/cm 2 at the dentine-pulp edge (Garberoglio and Brännström, 1976). This humid and organic nature of dentine makes it very challenging to bond to and have an effect on the integrity of the tooth-adhesive side of the interface. A peculiarity of dentine is the presence of the dentinal fluid in the tubular constitution that couples the pulp with the enamel-dentinal junction (EDJ). As stated by the hydrodynamic theory (Neill, 1838, Gysi, 1900, Brannström and Aström, 1972), when the enamel is lost and the dentine is exposed, external stimuli cause fluid shifts across the dentine which activate pulpal nerves and cause pain. This fluid flux within tubules, accountable for dentine sensitivity (Pashley et al., 1993), is also responsible for the persistent wetness of exposed dentine surfaces due to the outward fluid movement from the pulp, which may influence the quality of the adhesive-dentine interface and may decrease the bond strength between resins and dentine (Sauro et al., 2007). Furthermore, the increasing number of tubules with depth and, consequently, the increment in dentine wetness, can make bonding to deeper dentine even more difficult than to superficial dentine. The fluid movement in the dentinal tubules under the influence of pulpal pressure may in fact interfere with the penetration of the adhesive into the conditioned dentine surface (Chersoni et al., 2004), as well as causing deterioration of the adhesive interface with time. Another characteristic of dentine is the presence of a coating of debris produced with mechanical preparation, called smear-layer, consisting of shattered and crushed HAp, as well as fragmented and denatured collagen that is contaminated by bacteria and saliva (Brännström et al., 1981). It is revealed by scanning electron microscopy (SEM) as a 1-2 μm adherent surface with a mainly granular

substructure that varies in roughness, density and degree of attachment to the underlying tooth structure according to the surface preparation (Pashley et al., 1988). While cutting dentine, the heat and shear forces produced by the rotary movement of the bur cause this debris to compact and aggregate. The orifices of the dentinal tubules are obstructed by debris tags, called smear-plugs, that are contiguous with the smear-layer and may extend into the tubules to a depth of 1-10 Îźm (Prati et al., 1993). The application of acidic agents opens the pathway for the diffusion of monomers into the collagen network, but it also facilitates the outward seepage of tubular fluid from the pulp to the dentine surface, leading to a deterioration in the bonding effectiveness of some of the current adhesives. After the HAp crystals have been removed, it is quite challenging to also maintain the spaces created between collagen fibrils to allow monomers to diffuse into the substrate. The demineralised dentinal matrix can actually easily collapse if the matrix peptides, including collagen, are denatured during the conditioning, causing a decrease in the interfibril spacing and a loss of permeability to resin monomers (Nakabayashi et al., 1982).

1.2 Development of dentine-resin bonding technology In the developments of dental adhesives several attempts have been made to provide a stronger and more reliable bond as well as simplifying the clinical procedures. These attempts have resulted in the introduction of different generations of bonding systems which are different in chemistry, mechanism, number of bottles, application techniques and clinical effectiveness. In general, dentine bonding agents all contain similar ingredients, namely cross-

linking agents, bifunctional monomers, organic solvents, curing initiators, inhibitors or stabilisers, and sometimes inorganic filler particles. Whereas cross-linkers have two polymerisable groups (vinyl-groups or -CQC-) or more, functional monomers commonly have only one polymerisable group and a functional group, which can serve different purposes, such as enhancing wetting of dentine. Bifunctional monomers have in fact (meth)acrylate functions at one end, in order to provide covalent bonds with the composite monomers, and the so-called functional group, usually carboxyl, phosphate, or phosphonate at the other end which will impart monomer-specific functions (Van Landuyt et al., 2008).

1.2.1 Early dentine bonding agents Since calcium is abundant in dentine, the earliest dentine bonding formulas attempted to chemically bond to dentine by ionic bonds to this alkaline metal. The first adhesive resin system was created and manufactured at the Amalgamated Dental Company, England, UK, by a Swiss chemist called Oscar Hagger: it was composed of glycerophosphoric acid dimethacrylate (GPDM) and it was made available on the market as Sevriton Cavity Seal (The Amalgamated Dental Company, Ltd, London, UK) (Haggar, 1951). Kramer and McLean (1952) were among the first to investigate the bonding ability of this material to dentine (Kramer and Mc Lean, 1952). The dentine bonded with this adhesive system was observed by light microscopy: during the histologic examination they demonstrated altered staining of the bonded subsurface which took up haematoxylin more readily than did the control surfaces. It was supposed that the resin-primer had altered the dentine. This study was followed

by a work of Buonocore and co-workers (Buonocore and Quigley, 1958), who etched dentine with 7% hydrochloric acid and then applied GPDM bonding resin. These attempts were unsuccessful because of limitations in the adhesive monomer formulations and a general lack of knowledge of dentine as a bonding substrate. It was demonstrated that there was little evidence for the formation of chemical bonds between resins and dentine. A few years later, coincident with an expansion of knowledge in this area, considerable advances were made in adhesive monomer formulations to improve resin penetration into the tissue matrix. The development of a cross-linking dimethacrylate 2,2-bis[4(2-hydroxy3-methacryloyloxy-propyloxy)-phenyl] propane (Bis-GMA) reinvigorated the research on adhesion to dentine (Bowen, 1963). Dentine adhesive products such as Dentin Adhesit (Vivadent, Schaan, Liechtenstein), Scotchbond DualCure (3M Dental Products, St. Paul, MN, USA), Prisma Universal Bond (Caulk/Dentsply, Milford, DE, USA) and Bondlite (Kerr, Danburry, CT, USA) did not remove the smear-layer prior to resin application but were applied directly to smear-layer covered dentine. The presence of smear-layer on ground dentinal surfaces greatly reduced the permeability of tubular (Pashley, 1991) and intertubular dentine (Watanabe et al., 1994); for this reason the resin was unable to penetrate profoundly enough to establish a bond with intact dentine, and hence gave very low bond strength values (ca. 3-7 MPa) (Eick et al.). Examination of both parts of the failed resin-dentine bonds employing scanning electron microscopy (SEM) revealed smear-layer that split into upper and lower halves on each side. Thus, the “bond strength” was not really a measure of bonding, but measured the strength of the cohesive forces holding smear-layer particles together (Pashley, 1991). The actual interfacial bond strength between

the resin and the uppermost part of the smear-layer was higher by an unknown amount, because it did not fail.

1.2.2 Smear-layer removal and acid conditioning Despite widespread scepticism among the dental academic community, T. Fusayama proposed in 1980 to remove the smear-layer along with the underlying smear-plugs that prevented resin tag formation (Fusayama, 1980). Acid etching permitted demineralisation of the top 5-10 µm of the underlying sound dentine allowing dentinal tubules to receive micro-tags of resin and represented an important innovation which paved the way for the modern concepts of dentine bonding. However, even smear-layer removal was insufficient for high resin-dentine bond strength. The technique fell short of expectations in practice because of the intratubular fluid’s pressure that makes the dentine extremely humid, especially in deep cavities where a large number of wider dentinal tubules are exposed. This phenomenon inhibited the hydrophobic resin to efficiently adhere to the dentinal substrate and water was regarded as a contaminant. As a result, these bonding agents required severe air-drying of the dentine surface before application. The outcome of this manoeuvre was frequently a layer so thin that atmospheric oxygen inhibited their polymerisation (Erickson, 1989). Air-drying led to the evaporation of water maintaining the collagen network expanded and its collapse due to surface tension forces. The spaces between the collapsed collagen fibrils were therefore greatly reduced and with this the permeability of intertubular dentine to adhesive resins (Pashley et al., 1995a). What was also required was stripping the mineral phase from the collagen fibrillar matrix of dentine and keeping it as

expanded as possible in order to produce a large increase in surface and subsurface porosity. If monomers could have infiltrated this mesh-work and coated the fibrils with polymerised resin to improve micromechanical retention, they would have produced high bond strength. For the mentioned reasons and in view of the fact that a water-free environment is unachievable during clinical procedures, dentine bonding agents were reformulated in a more hydrophilic blend (Nakabayashi and Takarada, 1992). The introduction of low molecular weight monomers called primers, such as 4-methacryloxyethyl-trimellitic anhydride (4-META) and 2-hydroxyethylmethacrylate (HEMA), containing bifunctional groups - a hydrophilic functional group with a high affinity for the aqueous dentinal substrate, and a hydrophobic functional group having high affinity for the bonding adhesive - along with the etch-and-rinse technique enhanced the strength of the adhesion to dentine and provided reliable resin/dentine bond strengths (Barkmeier and Cooley, 1992). Furthermore, leaving the dentine wet made it possible to preserve porosity necessary for primer penetration in the demineralised spaces (Tay et al., 1995) and led to the formation of a acid resistant layer consisting of polymerisable hydrophilic monomers and exposed collagen fibrils (Nakabayashi and Takarada, 1992).

1.2.3 Dentine hybridisation and resin-infiltrated smear-layer Nakabayashi and his colleagues were the first to use transmission electron microscopy (TEM) with sufficient resolution to show the penetration of resin nano-tags into the demineralised dentine matrix to create an entirely new biomaterial that was half collagen fibrils and half resin. It was neither resin nor dentine but a hybrid of the two, and so was called hybrid layer (Nakabayashi

and Pashley, 1998). Hybridisation thoroughly modified the physico-chemical properties of tooth surfaces and subsurfaces and was considered a form of tissue engineering. With the introduction of the hybrid layer, many clinicians believed that the mechanism of bonding had been solved. Instead, the complexities of bonding became apparent when the same adhesive agents produced hybrid layers of different thicknesses depending on dentine depth and dentine condition. Hybrid layer formation was the major bonding mechanism in superficial dentine, which incorporates fewer tubules than deep dentine, due to the amount of intertubular dentine present in this area with little contribution from resin tags. Whereas, in deep dentine, resin micro-tag formation remained accountable for the bond strength, with a reduced contribution of the hybrid layer due to the limited amount of intertubular dentine available, as the tubules become larger and closer together. Nano-tags seemed to be much more important to overall retention (Marshall et al., 2010) increasing the bond strengths to 32 MPa, concurring for a better marginal seal and acting as an elastic cushion that, thanks to its elasticity (modulus of elasticity 3.4 GPa), was able to moderate the polymerisation shrinkage stress of the restorative composite (Wang and Spencer, 2003). Although the smear-layer is regarded a limiting factor in achieving high bond strengths, nowadays it can also be considered as a bonding substrate thanks to the development of smear-layer incorporating systems called self-etch. This was realised by raising the amount of acidic monomers and adding 20-30% acidic methacrylates (pH 1.9-2.8) to 20% water, 20% ethanol, 30% HEMA or dimethacrylates. Self-etch adhesives contain high concentrations of water and

acidic monomers (Watanabe et al., 1994). Water is a necessary ingredient required to ionise the acidic monomers so that they can etch through smearlayers into the underlying hard tissues (Tay et al., 2002b). Its presence entailed the use of water-miscible hydrophilic comonomers (e.g. HEMA) and/or acetone or ethanol as a solvent to prevent phase changes from occurring (Van Landuyt et al., 2005). Smear-layer covered dentine is substantially drier than acid-etched dentine. Smear-layer and smear-plugs being present, the transdentinal permeability is greatly reduced and no significant wetness is present on the dentine surface (Pashley, 1989). All the same, self-etching bonding systems are applied to smear-layer covered dentine under dry conditions, since they contain their own water. Combining acidic conditioners and resin primers did not require a separate etch-and-rinse phase and made these agents able to simultaneously condition and prime enamel and dentine (Chigira et al., 1994). Self-etching systems interact very superficially with the smear-layer and the underlying dentine. They can easily penetrate 1-2 Îźm of smear-layers but their penetration is restricted just to 0.5 Îźm into the top portion of the underlying intact dentinal matrix (Watanabe et al., 1994). This is due, in part, to the fact that the acidity is partially buffered by the smear-layer during comonomer penetration (Reis et al., 2004), and because the underlying mineralised dentine is less porous, and hence less permeable, than smear-layers. Water is also useful for solubilising the calcium and phosphate ions that are liberated by the etching. These ions, released from apatite crystallites during self-etching, get incorporated into the water of the adhesive blend or precipitate as calcium phosphates, which become dispersed within the comonomers in the interfibrillar spaces. Some of the calcium ions may also associate with the acidic monomers as calcium salts.

This mixture fills the interfibrillar spaces and some free ions may still diffuse up into the overlying adhesive layer (Bayle et al., 2007). The primed surfaces are not rinsed with water, leaving the dissolved smear-layer and demineralisation products to reprecipitate within the diffusion channels created by the acid primers. Compared with etch-and-rinse adhesives, many advantages have been attributed to self-etch adhesives. It has been suggested that they improve the efficiency of clinical procedures by omitting the obligatory rinse phase in etch-and-rinse adhesives and thus reducing the chairside time. Conditioning, rinsing and drying steps, which may be critical and difficult to standardise in clinical conditions, are eliminated in self-etch adhesives. Technique sensitivity correlated with bonding to dehydrated demineralised dentine is eliminated, as rinsing and drying phases are no longer needed. Since monomers infiltrate concomitantly as they demineralise, the collapse of the collagen network is prevented (Peumans et al., 2005). For the same reason, incomplete resin infiltration should be avoided. As the smear-layer and smear-plugs are not removed before the actual bonding procedure, rewetting of dentine by dentinal fluid should be disallowed too (Van Meerbeek et al., 2005). However, some leakage observations in the hybrid layer, and especially beyond the hybrid layer, have shed doubt on the concept that self-etch adhesives guarantee complete resin infiltration (Carvalho et al., 2005).

1.3 Physico-mechanical considerations of resin-bonded dentine One factor that could be easily overlooked is the requirement for the bonding system to act as a means of transferring load from one part of a structure to another. This generates stresses and strains within the resin-bonded dentine

and it is important that the adhesive has the necessary physico-mechanical properties to withstand these stresses and strains. Thus, the assessment of a bonding system should be based on its ability to carry load and contribute to the structural integrity of the whole unit. The durability of the resin-dentine bond is related to the depth of demineralisation versus the depth of monomer penetration, and the ability of the polymer not only to envelope each fibril but also to do so without leaving any gap or space between the resin and the fibril. That is, the resin-infiltrated layer must be free of any porosity or defects that can act as stress raisers under function or permit hydrolysis of collagen fibrils (Nakabayashi et al., 1982).

1.3.1 Wettability of dentinal surfaces and contact angle Wetting is a general term used to indicate the ability of a liquid to come into intimate contact with a solid substrate and to maintain contact with it. The balance between adhesive and cohesive forces dictates the degree of wetting (wettability). Adhesive forces between the liquid and the solid cause a drop to spread across, whereas cohesive forces within the liquid cause the drop to ball up and avoid contact with the surface. If a liquid can spread across a surface, it is said to “wet the surface”. This wetting ability of a liquid for a surface is usually characterised by measuring the contact angle (resultant between adhesive and cohesive forces) of a droplet on the surface. Resin contact angle measurement on dentine provides information on the interaction between adhesives and dentine, and it also indicates the affinity of dentine for the adhesive resin (Rosales-Leal et al., 2001).

Low contact angles imply good wetting while a contact angle greater than 90Â° usually means that wetting of the surface is unfavourable: the fluid does not spread over a large area of the surface but tends to minimise contact with it forming a compact liquid droplet. The tendency of a drop to spread out over a flat, solid surface hence increases as the contact angle decreases. For water, a wettable surface may also be termed hydrophilic and a non-wettable surface hydrophobic. Superhydrophobic surfaces have contact angles greater than 150Â°, showing almost no contact between the liquid drop and the surface (Feng et al., 2002). Wettability of dentine is an important topic to take into consideration as good spreading of monomers on this tissue is very important for successful bonding. For a liquid to spread uniformly across a solid surface, the surface tension of the liquid must be less than the free surface energy of the substrate. Substrates for bonding may present low or high surface energy. HAp is a high-energy substrate while collagen has a low-energy surface (Akinmade and Nicholson, 1993). Accordingly, acid etching increases the surface energy of enamel but decreases that of dentine. Unlike enamel, acid-etched dentine does not increase its surface energy to facilitate spreading of adhesive resins (Attal et al., 1994). Thus, for hybridisation of demineralised dentine with resin to occur, it is necessary to match the surface tension of the primer with that of the demineralised dentinal surface, depending on whether it is wet or dry. Commonly used bonding monomers such as HEMA have excellent spreading properties (Bowen et al., 1996) and could be considered to be surface-active comonomers (Rosales-Leal et al., 2001). That is, they are considered to

improve the ability of the monomers to wet the surface of acid-etched dentinal substrate. Wetting of the surface of dentine by monomers is a necessary initial step in bonding, but it alone is not sufficient to establish a successful bond, because it does not guarantee monomer penetration into the subsurface. The permeability of the demineralised intertubular dentinal network to monomers is a critical variable in dentine bonding (Nakabayashi and Takarada, 1992). To attain intimate association between resin monomers and collagen fibrils, the primers and bonding agents must be able to “wet” the collagen fibrils. If the fibril is enveloped by water, the monomers must be able to successfully compete with water for the fibril surface. Barbosa and collaborators found that dentine permeability was also intensified by the removal of organic materials (Barbosa et al., 1994). Sodium hypochlorite (NaOCl) is a well-known nonspecific proteolytic agent and its collagen removal ability after acid conditioning has been evaluated (Wakabayashi et al., 1994). After NaOCl treatment, the extent to which the primer wets the dentine surface is increased because the interactions between the primer and the deproteinised dentine are greater than before (Toledano et al., 2002). Deproteinisation leads to a hydrophilic surface (Attal et al., 1994) and eliminates the exposed collagen fibres. Besides, dentine becomes a porous structure with multiple irregularities which allows good mechanical retention (Vargas et al., 1997). However, complete removal of the collagen matrix with NaOCl as an adjunctive step of restorative and adhesive dentistry is still a subject for debate. Sauro et al. (Sauro et al., 2009a) evaluated the efficacy of a 12% w/v NaOCl solution for complete removal of exposed collagen matrices from acid-etched dentine

surfaces within a maximum clinically possible period of 120 seconds and a longer period of application (10 minutes) using confocal reflection/immunofluorescence microscopy and ESEM. An extended period (45 minutes) of NaOCl application was also performed as a negative control. This study demonstrated that complete removal of the exposed collagen matrix from the etched dentine surface can be achieved by applying a 12% w/v NaOCl solution, but at this concentration, it required a far longer reaction time than is clinically acceptable.

1.3.2 Solubility of adhesive monomers Solubility is the property of a substance called solute to dissolve in a liquid solvent to form a homogeneous solution. It is measured as the saturation concentration where adding more solute does not increase the concentration of the solution. The term “solubility parameter” was first used in dentistry by Asmussen (Asmussen et al., 1991). They regarded demineralised collagen as a porous solid polymer and reasoned that for primers to penetrate demineralised dentine, the primer should have a solubility parameter that is similar to the polymeric substrate, as is generally true in polymer chemistry. The concept was extended to Hansen’s triple solubility parameters so as to calculate the relative contribution of dispersive force (δd), polar force (δp), hydrogen bonding force (δh), and the total cohesive energy density of adhesive (δt). As Hoy’s triple solubility parameters are more widely used on dentine bonds, chemical structures modify the calculated Hoy’s triple solubility parameters for δd, δp, δh and δt (Mai et al., 2009).

Solubility parameter calculations have been used to quantify the degrees of hydrophilicity of polymers, important for the adhesive penetration into exposed collagen fibrils, and predict dentine-adhesive bond strengths (Asmussen and Uno, 1993). When a primer that has a low solubility in water is applied to moist demineralised dentine, the result is a limited distribution of the monomers into the water-filled three-dimensional network between the collagen fibrils, with a consequent low bond strength. Some hydrophilic monomers, such as HEMA, are very solubile in either water or acetone. Replacing water in the spaces around collagen fibrils, HEMA acts like a polymerisable solvent for the adhesive monomers placed thereafter. The uptake of adhesive monomers into these nano-spaces is contingent on their solubility in the solvent that occupies the spaces, hence this theory is very useful in predicting how miscible monomers should be in demineralised matrices saturated with various solvents (Sadek et al., 2007). Furthermore, the diffusion of the monomers is also determined by the size of the spaces between collagen fibrils and by the depth that they must reach from the surface. Wet demineralised dentine exhibits a fully expanded collagen network that offers maximal volumes between its fibrils. Under similar conditions, the bonding substrate has high permeability. At the other extreme, when there are no spaces between the collagen fibrils, as in air-dried, fully collapsed dentine (Carvalho et al., 1996), the permeability to monomer is extremely low. The ideal condition exists when there is both high permeability of the substrate (dentine) and high diffusivity of the solute (resin monomer) (Nakabayashi and Takarada, 1992).

Unfortunately, many adhesive monomers are not very soluble in water. That is why marketed adhesives are generally solvated in ethanol or acetone. When solvated adhesives are placed on water-saturated acid-etched dentine, their solvents attempt to penetrate into the water-filled spaces and some of the water in these spaces diffuses into the solvent. This culminates in too little solvent remaining in the infiltrating adhesive with the capacity to keep hydrophobic dimethacrylates like BisGMA (2,2-bis [4-(2-hydroxy-3-methacryloyloxypropoxy)] - phenyl propane) in solution. The net result is partial penetration of BisGMA into water-saturated matrices (Spencer and Wang, 2002). When BisGMAHEMA

mixtures

are

placed

water-saturated

dentine,

the

applied

concentrations changes as the much more water-soluble HEMA diffuses to the base of the demineralised zone. This can result in final molar ratios of BisGMA and HEMA in the hybrid layer that are very different from the applied molar ratio.

1.3.3 Permeability of the collagen network and monomers diffusivity Permeability quantifies the effort with which a substance can penetrate a membrane or diffusion barrier. The permeability of dentinal substrate to monomers and their diffusivity are extremely important for the creation of the hybrid layer. After the dentinal surface is acid etched and subsequently rinsed, intertubular spaces are filled with water and are presumed to be still as wide as when they were occupied by apatite crystallites (Van Meerbeek et al., 1996). Maintaining the permeability of the substrate as high as possible allows the achievement of good monomer infiltration because it is through these 15 to 20

nm wide diffusion pathways that adhesive monomer must move to fill the demineralised dentinal matrix and envelop every fibril. As these molecules diffuse into demineralised dentine, they may encounter some very small or narrow constrictions within the interfibrillar spaces, especially if the permeability of the collagen network has not been maintained. This reduces the rate of inward diffusion of adhesive monomers. If the strength of the bond is proportional to the sum of the cross-sectional areas of the resin-infiltrated interfibrillar spaces, then reductions in the size of these spaces should lead to lower bond strengths. Therefore it is essential to increase monomer concentration in demineralised dentine and to ensure that it becomes fully polymerised, to produce strong, durable hybrid layers. The ability of resins to infiltrate the exposed collagen mesh of dentine and to create a molecular-level intertwining within the fibril network depends upon their concentration and uniformity of penetration (Eick et al., 1996), their degree of polymerisation and cross-linking, and the amount of water that should be replaced in the demineralised dentinal substrate (Jacobsen and Soderholm, 1995). The mechanism available for resin infiltration involves the diffusion of the monomer into the solvent present in the spaces of the substrate and along collagen fibrils. That is the reason why this zone is also known as the resin interdiffusion zone (Van Meerbeek et al., 1996). The rate of diffusion depends on the affinity of the monomer for the substrate and is proportional to the concentration, temperature and viscosity of the solution (Cussler, 1976). The intrinsic diffusivity of the molecule, namely, its intrinsic free diffusion coefficient in the solvent, which is inversely related to its molecular weight or size, is also

an important variable. As the diffusion rate is proportional to the square root of the molecular weight, the smaller molecules diffuse faster and deeper than the larger ones (Nakabayashi and Pashley, 1998). On this account, whenever a blend of monomers of widely differing molecular weights is used in a primer or bonding agent, the rate of diffusion into the underlying substrate may vary to a considerable extent. This can result in final molar ratios of monomers in the hybrid layer that are very different from the initial applied concentrations (Eick et al., 1997). It has been mentioned how the presence of water during bonding procedures may come from several sources (i.e. tubular fluid, relative humidity, rinsing procedures). Post-etching rinsing thoroughly sponged out the dissolved dentine minerals and left approximately 70% of the demineralised dentine occupied by water (Nakabayashi et al., 2004). One of the assumptions with the 'wet-bonding' technique is that exposed collagen is not dried out thoroughly after etching to prevent its collapse to a thinner less permeable layer and the consequent restriction of the spaces around fibrils through which resins had to diffuse (Nakaoki et al., 2000). One way to avoid more than necessary and desirable air drying of dentine is to add water-miscible solvents in the primer solutions to chemically remove water from demineralised dentine (Suh, 1991). During the priming phase, the solvent (which exceeds the water) diffuses through the spaces between the collagen fibrils to reach the bottom of the demineralised zone in conjunction with the monomers that therefore have less water to challenge with (Eick et al., 1996). After evaporation of the solvent, the resin infiltration is thought to take the place of all the water present between the collagen fibrils.

However, when it was demonstrated that acid-etching lowered the stiffness of dentine from 18000 MPa to 1-5 MPa (Eddleston et al., 2003), also the susceptibility of the demineralised matrix to collapse became evident. It was discovered that even after primer infiltration (35% HEMA in 65% water) into the matrix, this was still so compliant that evaporation of the solvent was enough to cause it to collapse and extrude much of the monomers it had taken up (Eddleston et al., 2003). Solvents such as ethanol or acetone have much higher vapour pressures and generate less surface tension forces on the collagen fibrils network compared with aqueous primers while they evaporate (Maciel et al., 1996). Despite this, the use of ethanol-solvated primer mixtures also seems to stiffen the matrix enough to lower, but not to completely prevent, matrix collapse (Agee et al., 2006).

1.3.4 Permeability of adhesive resins and water sorption Ideally, polymer networks should be insoluble materials with relatively high chemical and thermal stability. Unfortunately, very few polymers are absolutely impermeable to water. Water movement in a polymer system is related to the availability of molecular-sized pores in its structure, and the affinity of the polymer components with water (Van Landingham et al., 1999). The availability of nanopores depends on the polymer microstructure, morphology and crosslink density, which are functions of degree of cure, relationship between the relative quantities of substances forming the compound, molecular chain stiffness and the cohesive energy density of the polymer (Soles and Yee, 2000). The affinity of the polymer to water is related to the presence of hydrogen

bonding sites along the polymer chains which create attractive forces between the polymer and water molecules (Soles and Yee, 2000). Incorporation of high concentrations of hydrophilic functional groups and methacrylate-based resin monomers in contemporary bonding systems, to achieve immediate bond strength to an intrinsically wet substrate such as dentine, also increased their attraction of water (Nishitani et al., 2007). The more hydrophilic the polymer, the greater is also the likelihood of formation of micro-cavities of different sizes in the polymeric network (Van Landingham et al., 1999). Many in vivo and in vitro studies have shown that resin-dentine interfaces become much weaker over time (Hashimoto et al., 2003). Sauro and collaborators (Sauro et al., 2007) showed that continued water flow under simulated pulpal pressure increased convective

fluid

movement

through

polymerised

resins.

was

also

demonstrated that the higher is the dentine permeability, the lower is the tensile bond strengths of simplified adhesives. The presence of hydroxyl, carboxyl and phosphate groups in monomers and their resultant polymers make them more hydrophilic and, as a result, more prone to water sorption. In the manners now being exemplified, when water sorption is sufficiently high, macromolecular polymer chains undergo a relaxation process as they swell to absorb the water. Most of the unreacted methacrylate groups trapped in the polymer network should not be released into aqueous environments, because they are still part of dimethacrylate molecules that have reacted and therefore are covalently bonded to the main polymer chain. Despite this, significant amounts of unreacted monomer or small chain polymer are released to the surrounding environment at a rate that is controlled by the swelling and relaxation capacities of the polymer (Santerre et al., 2001).

A number of studies have shown that elution for resin-based materials ranged from 0.05% to 2.0% of the weight of the specimen into aqueous media, with elution into alcohol and other organic solvents being higher in most cases (26%) (Ferracane, 1994, Hume and Gerzina, 1996, Pelka, 1999, Munksgaard et al., 2000, Tanaka et al., 1991). It has been demonstrated that the movement of water from hydrated dentine may cause the formation of water filled channels within the polymer matrices of contemporary hydrophilic dentine adhesives (Tay et al., 2004b). More hydrophilic polymer networks permit a faster release of unreacted monomers through nanovoids in the material (Brazel and Peppas, 1999). Accordingly, these water filled channels may accelerate elution of unreacted monomers from polymerised resins (Ito et al., 2005), as well as further the progress of weakening of the polymers by plasticisation (Wang and Spencer, 2003). This phenomenon decreases the stiffness of the polymers (Ito et al., 2005), produces stresses on the interface with the cavity wall and reduces bond strengths (Carrilho et al., 2005b). Water sorption/solubility investigations of hydrophilic adhesives in common use demonstrated that these systems have much higher water sorption than the more hydrophobic BisGMA/TEGDMA resins employed to seal multi-step adhesives (Ito et al., 2005). The hybrid layer created by simplified adhesives, containing high percentages of hydrophilic monomers, resulted in the formation of a porous interface (Wang and Spencer, 2003). This interface behaved as a permeable membrane (Tay et al., 2002a) that allowed water sorption, polymer swelling, resin hydrolysis and elution of unreacted monomers (Malacarne et al., 2006). When 3-step etch-and-rinse and 2-step self-etch adhesives were

challenged with thermomechanical loading between 5 and 55Â°C and up to 100 000 cycles, their microtensile bond strengths fell 25-30%. Conversely, the microtensile bond strengths of 1-step self-etch adhesives fell 50-80% after thermomechanical loading (Frankenberger et al., 2005). When dentine, respectively bonded with 3-step etch-and-rinse, 2-step etch-and-rinse, 2-step self-etch and 1-step self-etch adhesives, was directly exposed to water using miniature specimens that accelerate water sorption, the microtensile bond strengths of 3-step etch-and-rinse and 2-step self-etch adhesives did not lessen remarkably after one year of direct water storage. In contrast, the bonding effectiveness values of the 2-step etch-and-rinse and 1-step self-etch adhesives were reduced to almost zero after the same period of direct water exposure (De Munck et al., 2006). Clearly, the more hydrophilic the resins, the more water the polymers absorb, the more the polymers become plasticised and the more they lose their mechanical properties. Thus, water plasticisation of resins contributes to a reduction in resin-dentine bond strength durability.

1.4 Mechanisms responsible for loss of mechanical stability Despite successful immediate bonding, the longevity of

resin-bonded

restorations remains questionable due to physical (occlusal forces, expansion and contraction stresses related to temperature changes) and chemical factors challenging the adhesive interface (Breschi et al., 2008). Today, the most difficult task in adhesive dentistry is to make the adhesive-tooth interface more resistant against ageing, thereby rendering the restorative treatment more predictable in terms of clinical performance in the long term. Despite the enormous advances made in adhesive technology during the last 50 years, the

bonded interface itself remains the weakest area of composite restorations and none of the current adhesives or techniques is able to produce an interface that is absolutely resistant to degradation (Breschi et al., 2008). The degradation of the adhesive interface, which may occur in a relatively short term, depends on the way the adhesive has been manipulated, on the actual adhesive approach and on the adhesive composition. Hydrolysis of interface components, such as dentinal collagen and resin, due to water sorption, potentially enhanced by enzymatic degradation, and subsequent elution of the break-down products are the major factors thought to destabilise the adhesive-dentine bond (De Munck et al., 2009).

1.4.1 Hydrolytic degradation of dental adhesive resins Dental polymer networks have been shown to be susceptible to hygroscopic and hydrolytic effects to varying extents dependent upon their chemistry and structure (Ferracane, 2006). In the evolution of dentine adhesives, manufacturers have incorporated increasing concentrations of hydrophilic and ionic monomers to make these adhesives more compatible for bonding to intrinsically moist, acid-etched dentine (Van Landuyt et al., 2007). Increasing the hydrophilic nature of the adhesive-dentine interface has several disadvantages (Tay and Pashley, 2003a) and affects the integrity and durability of the adhesive/dentine interfacial bond (Spencer et al., 2010). Hydrophilic and ionic resin monomers are vulnerable to hydrolysis, due to the presence of ester linkages, typical of all methacrylates (Ferracane, 2006).

These ester linkages are theoretically susceptible to several esterases in body fluids (Soderholm et al., 1984). Adhesive hydrophilicity, water sorption, and subsequent hydrolytic degradation have been considered as highly correlative, because hydrolytic degradation occurs only in the presence of water (Carrilho et al., 2005a). Several studies have established a direct relationship between the presence of hydrophilic and acidic resin monomers in adhesive blends with decreased longevity of resin-dentine bonds (Peumans et al., 2005), owing to the fact that resin composition and hydrophilicity expedite water sorption in hydrophilic resins (Malacarne et al., 2006). Even the inclusion of small amounts of water may culminate in nano-phase separation of the adhesive components in the form of nanoscopic worm-like structures between the polymerised hydrophilic and hydrophobic resin phases (Ye et al., 2009b). Nano-phase separation reduces the dynamic mechanical properties of the polymerised adhesives (Park et al., 2010) and increases their susceptibility to esterase-catalysed hydrolysis (Kostoryz et al., 2009). Esterases known to activate ester hydrolysis include salivary esterase, cholesterol esterase, pseudocholinesterase, porcine liver esterase, and acetylcholinesterase. In contrast to HEMA, Bis-GMA has greater susceptibility to hydrolysis by cholesterol esterase and acetylcholinesterase. Biodegradation of HEMA/Bis-GMA adhesives in the presence of either enzyme appear to be more clinically relevant, since they simulate salivary enzyme activity (Yourtee et al., 2001). Previous work has shown that human saliva contains sufficient esterase activity to attack resin composites (Lin et al., 2005).

Nevertheless, it is not known whether there are similar esterases in dentinal fluid and how they could reach resin-dentine interfaces. Hydrolysis of methacrylate ester bonds caused either by the increase in acidity of monomer components (Aida et al., 2009) or by salivary esterases (Shokati et al., 2010) can break covalent bonds between the polymers by the addition of water to the ester bonds. Apart from water, the interfibrillar spaces in acid-etched dentine also include highly hydrated negatively charged proteoglycans that constitute a hydrogel within that space (Scott and Thomlinson, 1998). If these hydrogels continue to be hydrated in interfibrillar spaces, they may be responsible for “molecular sieving” of larger dimethacrylates like BisGMA, allowing only smaller molecules such as HEMA to infiltrate the base of the hybrid layers. Since HEMA forms a linear polymer that does not cross-link, HEMA-rich regions of hybrid layers may undergo large strains during function that prompt further degradation and compromise the longevity of resin-dentine bonds (Liu et al., 2011c).

1.4.2 Endogenous collagenolytic activity Collagen serves as a structural barrier between tissues, and thus collagen catabolism (collagenolysis) is required to be a tightly regulated process in normal physiology. The turnover of connective tissue and degradation of nearly all extracellular matrix components has been ascribed to different members of the matrix metalloproteinase (MMP) family, due to their ability to catalyse the hydrolysis of type I collagen triple helical structure. MMPs are a group of zincand calcium-dependent enzymes operating in homeostatic and reparative

processes, but unregulated catalysis by these extracellular proteinases leads to the pathological destruction of the tissues to which they are bound. In soft tissues, these collagenases are either secreted in a latent form or inhibited by tissue inhibitors or metalloproteinases (TIMPs). In mineralised tissues, these enzymes may be active, secreted in a latent form or inhibited by TIMPs as well as being incorporated by apatite crystallites that fossilise them and enable their activity. It has been mentioned that resin-dentine bonding could be considered a unique form of tissue engineering in which dentists utilise the natural collagen fibril matrix of demineralised dentine, which is continuous with the underlying mineralised matrix, as a scaffold for resin infiltration. The collagen fibrils of the hybrid layer, by being anchored into the underlying mineralised matrix, provide micromechanical retention of adhesive resins that, in turn, retain resin composites. The only continuity between adhesively retained restorations and the hybrid layer are the resin tags in the tubules, along with the nanometre-wide resin extensions that pass around and between collagen fibrils. Nevertheless, unprotected type I collagen fibrils situated at the bottom of the hybrid layer are subjected to deterioration over time due to the activation of endogenous collagenolytic enzymes (Mazzoni et al., 2006). Several studies reported that mineralised dentine contains in fact bound MMPs such as MMP-2, -3, -8, -9 and -20 (Toledano et al., 2010). Even though the quantitative analysis of different MMPs in dentine remains to be completed, the currently available data indicate that MMP- 2 may be the prevalent MMPs in human dentine matrix (Mazzoni et al., 2007). Although classified as a gelatinase

(gelatinase A), MMP- 2 is also an effective collagenase (Aimes and Quigley, 1995). These host-derived proteases contribute to the breakdown of collagen matrices in the pathogenesis of dentinal caries (Chaussain-Miller et al., 2006) and periodontal disease (Hannas et al., 2007). In addition, non-collagen-bound MMPs are also present in saliva (Sulkala et al., 2001), in dentinal tubules, and, presumably, in dentinal fluid (Boushell et al., 2008). Proof of degenerative modifications in hybrid layers was offered by De Munck and collaborators (2003) with long-term in vitro TEM studies that indicated loss of staining and loss of cross-banded collagen after 4-5 years of water storage (De Munck et al., 2003). The degradation was irregular and variable but also extensive. The high resolution provided by TEM examination suggested that collagen had been converted into gelatin. That is, the hybrid layer was not empty but still contained organic material not pigmented with heavy metal stains which are typically taken up by native cross-banded collagen fibrils (GarcĂaGodoy et al., 2007). When normal hybrid layers receive tensile stressing, the collagen fibrils share the stress with the resin network by being loaded in parallel. Subsequent to cleavage of collagen and its conversion to weaker gelatin (i.e. loss of crossbanded collagen), the stresses applied to the weakened hybrid layer are carried only by the stiffest surviving material. In this way the resin meshworks pull out of the "gelatinised" hybrid layer, producing lower bond strengths (De Munck et al., 2003). To demonstrate the degradation of dentine matrices by endogenous MMPs, Pashley and collaborators (2004) acid-etched disks of dentine with 37%

phosphoric acid for 15 s, then placed them in buffered calcium- and phosphatecontaining media with or without four protease inhibitors, normally utilised in biochemistry to prevent MMPs during collagen extraction and purification (Pashley et al., 2004). Since MMPs are technically hydrolases, that is to say they catalyse specific peptide bonds in presence of water, half of the etched specimens were incubated in mineral oil. Specimens were removed and processed for TEM observation of the quality of the collagen after 24h, 90 days and 250 days. The naked collagen fibrils had degraded down to the mineralised base after the period of incubation in the absence of protease inhibitors. By contrast, in specimens incubated in the presence of protease inhibitors, the collagen fibrils appeared normal. Similarly, specimens incubated in oil looked normal over the 250 days, as in the absence of water MMPs could not cleave collagen. Mazzoni and collaborators (2006) reported that when etch-and-rinse systems were applied on dentine their intrinsic acidity (i.e. pHs between 2.6 and 4.7) was enough to demineralise dentine but not to denature the collagenases. Hence, the pH of the adhesives was sufficient to expose and set in motion dentinal MMPs, initiating autolytic phenomena that ultimately affected the hybrid layer (Mazzoni et al., 2006). Such results were consistent with a previous study showing that exposure of MMPs to an acidic pH (c. pH 4.5) activates MMPs in carious dentine (TjĂ¤derhane et al., 1998). Furthermore, when normal mineralised human dentine powder was mixed with different self-etch adhesives with pHs between 1.5 and 2.7, the gelatinolytic and collagenolytic activity of dentine increased more than 10-fold (Nishitani et al.,

2006). Following application of self-etching primers, increases of collagenolytic activity were also reported for root canal dentine shavings produced during rotary instrumentation with Gates-Glidden burs (Tay et al., 2006). This body of increasing evidence indicates that endogenous MMPs are uncovered and/or activated by many, if not all, dentine bonding procedures. It was also suggested that mildly acidic resin monomers can activate MMPs by inhibiting TIMPs (Ishiguro et al., 1994) in TIMP-MMP complexes, thereby producing active MMPs (TjĂ¤derhane et al., 1998, Sulkala et al., 2001). Alternatively, acidic resin monomers may set in motion latent forms of MMPs (pro-MMPs) via the cysteine-switch mechanism that uncovers the catalytic domain of these enzymes that were blocked by propeptides (Tallant et al., 2010). Cysteine cathepsins are papain-like endopeptidases having a vital role in mammalian cellular turnover, e.g. bone remodelling and resorption. Most of these peptidases become activated at the low pH found in lysosomes. Thus, their activities occur almost entirely within those organelles, playing a part in intracellular proteolysis within the lysosomal compartments of living cells (Dickinson, 2002). However, they also exist as exopeptidases and participate in extracellular matrix degradation through the breakdown of type I collagen and proteoglycans (Obermajer et al., 2008). For example, cathepsin K, highly expressed in type I collagen degradation, works extracellularly after secretion by osteoclasts during bone homeostasis. The different members of this family of proteases are distinguished by their structure, catalytic mechanism, and which proteins they cleave. Cathepsins B,

L, and S cleave the non-helical telopeptide extensions of collagen molecules, while cathepsin K cleaves the collagen molecules along their triple helix region (Liu et al., 2011c). Unlike the collagenolytic MMPs (MMP-1, -2, -8, and -13) that cleave type I collagen into a Âž N-terminal fragment and Âź C-terminal fragment at a single site within the triple helix (between amino acids 775 and 776 from the first GXY triplet of the triple helix domain), cathepsin K cleaves collagen molecules at multiple sites within the triple helix, thereby giving rise to fragments of various sizes (Garnero et al., 1998). Tersariol et al. reported for the first time the presence of cysteine cathepsins in dentine demonstrating their expression by mature human odontoblasts (Tersariol et al., 2010). However, these collagen-degrading enzymes are thought to be more abundant (approximately 10-fold) in carious dentine (Liu et al., 2011c). Like MMPs, cysteine cathepsins may be activated in mildly acidic environments. Acid activation of dentine-bound cathepsins may also coincide with the conversion of matrix-bound MMPs into their reactive form. On top of that, glycosaminoglycans (GAGs) can promote further conversion of the latent forms of the cathepsin enzyme family into their mature forms at neutral pH (Obermajer et al., 2008). Consequently, GAG-cathepsin activation allows active cathepsins to be functional even in neutral pH environments. The existence of cysteine cathepsins in dentinal tubules (Tersariol et al., 2010) indicates that they are derived from the dental pulp via the dentinal fluid and may be activated by mildly acidic resin monomers. They may subsequently

interact with GAGs and assist salivary MMPs in the degradation of incompletely infiltrated collagen fibrils within the hybrid layer.

1.5 Adhesion testing Several aspects should be considered when testing the strength and durability of the bond to dentine. These include the heterogeneity of its structure and composition, the features of the dentinal surface exposed after cavity preparation, and the characteristics of the adhesive itself, such as its strategy of interaction and basic physicochemical properties. Laboratory experiments conducted on dental adhesives can be classified into two types, namely behavioural tests and structural integrity tests. In the behavioural tests the focus is on understanding how the material behaves and how one might be able to change the properties of the material by changing such things as its composition. These experiments are not designed to assess the clinical performance of the material used. Examples of the sorts of things one might measure are tensile/shear bond-strength, thus enabling bond strength to be measured as a material property, elastic modulus, fracture toughness, coefficient of thermal expansion and translucency. However, all sorts of chemical and mechanical challenges that are inherent to the oral environment should also be taken into account, such as moisture, masticatory stresses, changes in temperature and pH, and dietary and chewing related habits (Mjรถr and Gordan, 2002). Structural integrity tests aim to provide an experimental arrangement that mimic the performance of the material during function. In other words, the material is being applied in a situation in an attempt to provide some insight into how the material might respond to a clinical environment and

to learn what makes the structure fail. This will be a complex interaction between material, design and environment. Thus, the structural integrity test is seeking to establish a link between the material and its performance in a clinical situation. Typical examples of such tests are represented by fatigue tests. Besides static bond-strength tests, theoretically clinically more relevant is in fact to test adhesive interfaces dynamically, as in the clinical situation toothcomposite bonds are seldom subjected to acute tensile/shear stresses. It is, however, exposed to cyclic sub-critical loadings produced during chewing (De Munck et al., 2005). Although fatigue tests are more labour intensive and timeconsuming than static bond-strength tests, a steadily growing, but still only low number of fatigue tests have been tried out throughout recent years with regard to their potential to predict clinical effectiveness. In the literature, six different fatigue tests have been reported on, as there are, chronologically: (i) a macropush-out fatigue test (Frankenberger et al., 1999); (ii) a macro-shear fatigue test (Erickson et al., 2009); (iii) a micro-rotary fatigue test (Van Meerbeek et al., 2003); (iv) a micro-shear fatigue test (Braem, 2007); (v) a micro-4-point-bend fatigue test (Staninec et al., 2008); and (vi) a micro-tensile fatigue test (Poitevin et al., 2010). Despite the alleged need for more fatigue testing of adhesives and even though several typical fatigue phenomena can be observed, little new information on bonding effectiveness is provided than that revealed by the easier and faster static bond-strength tests (Van Meerbeek et al., 2010). For example, micro-rotary as well as micro-tensile fatigue testing revealed a similar superior bonding effectiveness of the 3-step ‘gold-standard’ etch-and-rinse adhesive OptiBond FL (Kerr, West Collins Orange, CA) over the 2-step ‘goldstandard’ self-etch adhesive Clearfil SE Bond (Kuraray, Tokyo, Japan), that in

turn bonds significantly better than the 1-step adhesive G-Bond (GC, Tokyo, Japan). In addition, these fatigue tests have largely been applied to dentine with bonding to enamel being much more difficult to assess in fatigue (Van Meerbeek et al., 2010). The longevity of the bond upon ageing of the specimens is another aspect of the performance of dental adhesives that requires particular attention. Several studies

highlighted

very

good

instantaneous

and

short-term

bonding

effectiveness either to enamel or dentine (Inoue et al., 2001), but durability and stability of the resin-dentine bonded interfaces created by current adhesive systems still remain unconvincing (De Munck et al., 2005). This shifted the focus of researchers’ investigations to the evaluation of ageing mechanisms. Accordingly, besides determining ‘immediate’ bond strength values, measuring the ‘aged’ bond strength was decisive in order to estimate the clinical effectiveness of this type of material (Breschi et al., 2008). In vivo studies are ideally suited to assess both the performance and the longevity of restorative materials (Hebling et al., 2005, Carrilho et al., 2007b), but their feasibility is complicated or even precluded by the associated bureaucratic requirements, they also require much more time to collect significant information and a higher cost is involved in the procedure (Reinke et al., 2012). Laboratory studies, on the other hand, offer the advantages of lower costs, shorter duration, greater standardisation due to the possibility of isolation of variables and have been widely used to predict the performance and longevity of adhesive materials (De Munck et al., 2005, Van Noort, 1994, Amaral et al., 2007).

Most of the knowledge we have about the longevity of dentine bonds are based on in vitro studies, in which some kind of ‘ageing’ factor is added to the investigation design (De Munck et al., 2005). This could range from examining the effects of long-term storage in water, or some more aggressive solutions (Lee et al., 1994, Yamauti et al., 2003, De Munck et al., 2007, Toledano et al., 2006) along with the use of pH (Peris et al., 2007, Passalini et al., 2010), thermal (Price et al., 2003, Nikaido et al., 2002, Bedran-de-Castro et al., 2004, Lodovici et al., 2009), and mechanical loading cycling (Bedran-de-Castro et al., 2004, Lodovici et al., 2009, Li et al., 2002, Osorio et al., 2005) as well as their combinations (Grande et al., 2005, Bedran-de-Castro et al., 2004, Lodovici et al., 2009) in order to recreate some of the challenges that these restorations are prone to under clinical service for prolonged periods of time. The immersion of micro-specimens in water is a well-validated method to assess resin-dentine bond strength durability (De Munck et al., 2006). It usually requires 6 months to detect drops on the μTBS values (De Munck et al., 2005), but this period of time may be even shorter when daily water exchange is performed (Skovron et al., 2010). Doing so, it was reported that all classes of adhesives exhibited mechanical and morphological evidence of degradation that resembled in vivo ageing (Shono et al., 1999). Other water-storage studies confirmed that immediate resin-dentine bond strength values do not always correlate with long term bond stability since deterioration throughout the dentine bonded interface occurs at a fast pace (Carrilho et al., 2005b, Garcia-Godoy et al., 2010, Hashimoto et al., 2010a). The introduction of pH, thermal, and mechanical loading cycling are attempts to simulate clinically relevant conditions; however, they still lack standardisation in

the number of cycles, temperature, dwell time, immersion time, load and load frequency and this may hinder comparison of study results and lead to contradictory findings (Amaral et al., 2007, Reinke et al., 2012). Recently, an in situ model has been used for the evaluation of ageing mechanisms involved in the degradation of resin-dentine bonded interfaces created

with two simplified etch-and-rinse adhesives [Adper Single Bond 2

(3MESPE, St. Paul, MN, USA) and Optibond Solo Plus (Kerr, Danburry, CT, USA)] under more realistic conditions (Reinke et al., 2012). Compared to the immediate results, where no restorations were included in the intra-oral appliances used by volunteers and no ageing method was performed, rapid deterioration in resin-dentine bond strength were observed after the 14day simulated cariogenic challenge accountable for a more intense and rapid degradation rate of the collagen. However, the findings of the present investigation could not be compared to other durability studies since this was the first one that employed an in situ model to investigate the degradation of resin-dentine bonds that occurs with etch-and-rinse adhesives.

1.5.1 Assessment of sealing ability The seal of a restorative material against the tooth structure, and the quality and durability of the seal, are major considerations for the longevity of adhesive composite restorations. Since the longevity of an adhesive composite restoration is mainly affected by the leakage of oral fluids along the interface between the restorative material and the tooth substrate (De Almeida et al., 2003), it is very important to evaluate the capacity of a bonding system to

maintain the seal of the tooth-restoration interface. In case of bonding failure, it would be desirable that the hybrid layer and resin tags remained in the dentine surface rather than being pulled away, thereby maintaining a surface seal which will continue to protect the pulp (Griffiths and Watson, 1995).

1.5.1.1 Micro-leakage and micro-permeability The study of resistance to the diffusion of a substance into a fluid-filled gap or a defect between filling material and tooth structure, has been of great concern in restorative dentistry. The ability of a bonding system to maintain the toothrestoration interface seal can be evaluated using high-resolution leakage studies on the μm scale (micro-leakage) or, alternatively, by micro-permeability studies (Griffiths et al., 1998). Micro-leakage is a typical example of structural integrity test, often assessed in vitro on cross-sections using a tracer or dye that is able to infiltrate the composite-tooth interface, such as silver nitrate. Silver nitrate has been employed to assess the porosity of the hybrid layer because it is very soluble (a 50% aqueous solution is usually applied), the silver ion is very small, and, as soon as it has diffused into a region and has been reduced to metallic silver, it remains at that site and cannot diffuse away or fade as is frequent with watersoluble dyes. Micro-permeability tests also make use of fluorescent dyes, that are ‘loaded’ to the pulp chamber in order to investigate at higher resolution the sealing ability of adhesives at the interface itself. Employing confocal laser scanning microscopy, fluorescent dyes have been shown to penetrate through dentine towards the

interfacial region thus indicating the intimacy of the hybrid layer to the bonded interface (Sauro et al., 2009a). In this method, the fluorophore presence can show with clarity the existence of possible routes for micro-permeability either around resin tags, through the porous base of the hybrid layer, and along the interface between the hybrid layer and the adhesive agent. The term 'micro-permeability' was first used by Sidhu and Watson (Sidhu and Watson, 1998, Sauro et al., 2009b) in the evaluation of the interfacial characteristics of resin-modified glass ionomers. Water movement from the pulp chamber towards the resin-bonded dentine interface through dentinal tubules was identified by using a solution of rhodamine B deposited in the pulp chamber. This study revealed much information regarding porosity of the interfacial bonded-layer, particularly in samples seemingly free of interfacial gaps. The extent of permeability is contingent on the penetration of the adhesive components into etched dentine and on the development of gaps or porosities in the bonded interface resulting from polymerisation shrinkage of the primer, adhesive, or resin components.

1.5.1.2 Nano-leakage In principle, resin hybridisation of dentine should protect hermetically the collagen

fibrils

against

the

subsequent

exogenous

and

endogenous

denaturation challenges, in a manner that is analogous to the protective function of the apatite phases in mineralised dentine. This would ensure the

absence of spaces between the collagen fibrils and resins, making the hybridised dentine long-lasting and resistant to degradation. However Griffiths et al., using a control adhesive system that retained a modified smear-layer, observed that fluorescent dye could penetrate not only the porosities within the bonded interface but also the smear-layer itself (Griffiths et al., 1999). The term 'nano-leakage' was introduced by Sano et al. (1995) to distinguish these findings from micro-leakage phenomena and indicate possible pathways for permeability through the hybrid zone (Sano et al., 1995a). Nano-leakage occurs in the absence of visibly discernible, 10-20 Îźm wide gaps between restorative materials and cavity walls, through nanometre-sized spaces of approximately 0.02 Îźm at the bottom of the hybrid layer, where resin monomers interface with decalcified dentine. These sites of incomplete interfibrillar resin infiltration were detected for the first time by Sano et al. in hybrid layers created with both etch-and-rinse and selfetch adhesives when these interfaces were immersed in an acidic (pH 4.5) silver nitrate solution (Sano et al., 1995b). The silver nitrate tracer was reduced into silver granules that were deposited as reticular patterns (so-called 'water trees') within the interfibrillar spaces of the hybrid layer, which were then observed using the back-scattered mode of a SEM or in thin sections prepared for TEM. These were considered sites of incomplete water removal and subsequent suboptimally polymerised resins. If the resin had perfectly filled all the empty spaces between the hybrid layer and the underlying demineralised dentine, there should have not been room available for silver ion penetration. According to the theory of 'water tree' (Tay et al., 2004a), these reticular

patterns expand as a result of water movement from the underlying dentine substrate into the partially polymerised adhesive resin matrix, in which the polymer chains are not sufficiently cross-linked to resist their displacement by the bulk free water. Thus water sorption fosters the transformation from the original isolated silver grains to water-filled channels in the adhesive resin matrices. Since nano-leakage occurs in the deepest part of the hybrid layer, and spreads throughout this structure, it has been identified as a special type of intertubular dentine permeability within the defective nanometre-wide zones around the collagen fibrils that were not completely enveloped by the resin. Nano-leakage has also been observed using confocal light microscopy when a fluorescent dye such as rhodamine B was used as a tracer (Pioch et al., 2001). Uninfiltrated interfibrillar spaces are very small to permit bacterial penetration, but are large enough to serve as a pathway for water movement within the adhesive-dentine interface, leaving the hybrid layer with a large amount of porosity as a predictable site of enzymatic and hydrolytic degradation over time (Tay and Pashley, 2003b). When a 50% ammoniacal silver nitrate tracer solution was employed instead of regular aqueous solutions, two different modes of nano-leakage expression were also seen. The first mode was made up of reticular interfibrillar deposits, already characterised using the traditional silver nitrate solution, while the second mode consisted of isolated spotted silver grains (Tay et al., 2002a). Nano-leakage patterns can change over time. Tay and Pashley (2003) hypothesised that water sorption and, afterward, hydrolytic degradation may be evidenced by the variations in uptake of ammoniacal silver nitrate within resin-

dentine interfaces (Tay and Pashley, 2003b). Some of these variations are the initial reduction of the reticular silver tracer patterns within the hybrid layer, and the increment in size and density of the isolated spotted silver grains within the adhesive. In fact, these represent sites where the cationic diamine silver ion complexes interact with the anionic functional groups of the hydrophilic resin monomers, and probably where subsequent water sorption within the polymer matrix occurs via hydrogen bonding. Interconnecting silver-filled channels have been reported within the adhesive layers of 2-step etch-and-rinse and 1-step self-etch adhesives (Tay and Pashley, 2003b). Water trees were predominantly located along, and perpendicular to, the surface of the hybrid layer, extending into the overlying adhesive layers. In their mildest forms, they seemingly expanded from the surface of the bonded dentine into the adhesive layer. In their most severe forms, commonly seen with 1-step self-etch adhesives, more than 50% of the adhesive layer was filled with heavy silver deposits.

1.5.2 Bond strength measurement While

adhesive-enamel/dentine

interfacial

characterisation

with

electron

microscopy (possibly supplemented by chemical interfacial analysis) certainly discloses a deeper insight into the underlying mechanisms of adhesion, the actual bonding effectiveness of today’s adhesive approaches should be determined using a mechanical bond-strength test. The rationale behind bond strength measurements is that the stronger the adhesion, the better the material will endure any stress imposed by resin polymerisation and oral function. By definition, the ideal bond-strength test should be simple and reasonably fast. In

general, advantages of ‘laboratory testing’ are, among others, the relative simplicity of the test methodologies commonly used, the relatively quick collection of data on a specific parameter/property, to be able to test simultaneously many experimental groups within one study set-up, to be able to directly

compare

the

performance

new

and/or

experimental

material/technique with that of the current ‘gold-standard’, and the possibility to measure one specific parameter, while keeping all other variables constant. The final objective of a laboratory test should obviously be to gather data in prediction of the eventual clinical outcome. In order to measure the bonding effectiveness of adhesives, diverse methodologies can today be used (Burke et al., 2008). The bond strength can be measured statically using a macro- or micro-test set-up, basically depending upon the size of the bond area.

1.5.2.1 Macro-bond strength test The macro-bond strength, with a bond area larger than 1 mm 2, can be measured using either ‘tensile’ or ‘push-out’ protocols and in ‘shear’ manner. The macro-tensile bond-strength approach is the less popular and can be used for instance to measure the bond strength of cements to hard materials such as ceramics and metal alloys (Abreu et al., 2009, Kern et al., 2009). A push-out approach has also been employed, in particular to dynamically test the fatigue resistance of adhesive-dentine bonds (Drummond et al., 1996, Zicari et al., 2008). It has however never been adopted as a universal bond-strength test method, most likely because of the more laborious specimen preparation involved as well as the more time-consuming methodology. This method appeared however very useful to test the retention of posts luted in root canals

(Goracci et al., 2004, Zicari et al., 2008). Of all the adhesive tests used in dentistry, the shear bond strength test has been one of the most popular bonding experiments ever devised (Burke et al., 2008); it was found to have been used in 26% of scientific papers reporting on bond strength (Van Meerbeek et al., 2010). Its popularity has much to do with the simplicity with which this experiment can be conducted as no further specimen processing is required after the bonding procedure. When this experimental design was first introduced the quality of the dentine bonding agents was quite poor relative to the bonding agents available today. In those initial experiments the bond strength was so poor that failure would generally occur at the adhesive interface such that the bond strength, reported as a shear stress based on the ratio of load to bonding area, could be used as a reliable measure of the quality of the adhesion achieved. In this way different adhesives could be compared by calculating the nominal shear bond strength as long as the experimental design used was consistent in terms of size, shape and load application. As the quality of dentine bonding agents improved, fractures no longer occurred at the adhesive interface but would tend to be cohesive in nature, more often regarding the dentine, and without an increase in the shear bond strength. The reason for this is that the stresses generated predispose any occurring crack to deviate into the dentine when confronted with a strong adhesive bond. In shear bond strength tests this deviation of the fracture mode from the interface into the dentine tends to occur at a nominal shear stress of some 20 MPa. This has led some investigators to mistakenly suppose that the bond strength exceeds the cohesive strength of the dentine and that the maximum bond strength that can be achieved to dentine is of the order of 20 MPa, which is not in agreement

with the evidence that the ultimate tensile strength of dentine may be as high as 100 MPa (Bouillaguet et al., 2001).

1.5.2.2 Micro-bond strength test Bond strength is typically measured in tensile by the micro-tensile bond strength (μTBS) test, according to the procedure developed in 1994 by Sano (Sano et al., 1994). It has been demonstrated that smaller specimens are stronger and fail at higher stress than larger bonds, as a consequence of the reduction and removal of inherent flaws in the system (Pashley et al., 1995b). Defects in bonded interphases, such as air bubbles, water blisters or regions of resin-solvent phase separations, act in fact as stress concentrators during bond testing. These flaws permit initiation of local stresses that exceed the cohesive strength of one of the components of the bonded interphase (resin compositeadhesive layer; adhesive layer-top of hybrid layer; infiltrated resin-collagen; or non infiltrated dentine-mineralised dentine junctions) and result in cracks that propagate rapidly to cause catastrophic failure. With smaller bonded specimens, the stress distribution throughout the resindentine interface is more uniform and it is more likely for the bonds to fail adhesively rather than cohesively in dentine (Sano et al., 1994). In specimens used for micro-tensile bond strength tests, the bond area tested is much smaller compared to that of the ‘macro’ tests, being about 1 mm2 or less. After the bonding procedure, some further specimen processing or the actual preparation of the micro-specimens is required, rendering the test more laborious and technique-sensitive. Nevertheless, a long list of advantages is typically ascribed to μTBS when compared to macro-bond-strength testing, of

which the most important are the better economic use of teeth (multiple microspecimens obtained from a single tooth enable large study set-ups), the better control of regional differences as well as substrate variables (e.g. peripheral versus central dentine) and the better stress distribution at the true interface (Sano et al., 1994). In fact, all measurements are taken in the central part of the specimen, well away from the clamping site, such that a uniform, uniaxial stress is generated and homogenously distributed in the cross-sectional bonded area (Soares et al., 2008). The maximum tensile stress is then calculated simply dividing the load by the cross-sectional area. In other words, the microtensile bond strength is calculated as the tensile load at failure divided by the crosssectional area of the bonded interface. As the micro-tensile bond strength test may sample heterogeneous regions of dentine, it is recommended that the multiple values for any tooth be averaged to provide a mean and standard deviation around that mean. The potential ability of the micro-tensile bond strength to calculate the average tensile stress at the adhesive interface is a significant difference with the shear bond strength test. This means that it is then possible to assess the quality of the adhesion by comparing the tensile load at failure for different bonding agents and evaluate the mode of failure. Another notable difference between the micro-tensile test and the shear test is that the cohesive failure in tooth substrate or composite occurs less frequently (Van Meerbeek et al., 2010). In addition, given that the failure in shear occurs at lower stresses than failure in tension, the 20 MPa ceiling, observed for the shear bond strength test, ceases to exist and tensile bond strengths in excess of 40 MPa can be achieved. Thus, a micro-tensile protocol appears to be able to discriminate adhesives better on

their fundamental behavioural characteristics and bonding performances than a traditional shear bond-strength approach. This is the most likely reason for up to 60% of current scientific papers reporting on bond strengths having used the μTBS approach (Van Meerbeek et al., 2010). Van Meerbeek et al. reviewed the literature with regard to the relation between bond strength testing and clinical effectiveness of adhesives in terms of retention rates of Class-V restorations. Significant differences were found in the ‘pooled’ mean bond strength, as the weighted bond strength means of individual adhesives ranged from about 12 MPa (for Absolute, Dentsply-Sankin, Tokyo, Japan) to 49 MPa (for OptiBond FL, Kerr, West Collins Orange, CA, USA) and the weighted bond-strength means per adhesive class ranged from 31 MPa for 3-step etch-and-rinse adhesives, to 29 MPa for 2-step self-etch adhesives, 26 MPa for 2-step etch-and-rinse adhesives, and 20 MPa for 1-step self-etch adhesives (Van Meerbeek et al., 2010). The weighted mean and large confidence interval for the (2-step) glassionomers was less reliable, since only a few products were able to be included. Glass-ionomer adhesives perform as well as the two-step self-etch adhesives, yet the bond strength of glass-ionomers is scarcely tested, which could be due to the well-known fact that during the test they tend to fail cohesively within the material itself, rather than de-bonding from the tooth surface, so that the actual bond strength to tooth tissue can hardly be determined. The poorer mechanical properties of glass-ionomers also explain the lower scores achieved in bondstrength tests when compared to those of resin-based adhesives. Hence, 3-step etch-and-rinse adhesives bond more strongly to dentine than all other adhesives that use simplified application procedures, even if some 2-step

self-etch adhesives may come close to the bonding effectiveness of etch-andrinse adhesives. Also in order to generate as many specimens as possible from a single tooth, a micro-shear bond-strength test (ÎźSBS) was introduced in 2002 (Shimada et al., 2002). This test combines the ease of manipulation with the ability to test several specimens per tooth. The very fine composite build up (cylinder) with a typical diameter of 0.7 mm, in combination with a relatively thick adhesive layer, may however result in considerable bending and variable and non-uniform loading conditions. This non-uniform stress distribution is probably even more pronounced as compared to macro-shear bond testing. Furthermore, it is impossible to confine the adhesive to the area tested, as required by ISO Technical Specification No. 11405 (ISO/TS, 2003). Basically due to these major shortcomings, the ÎźSBS test has not been adopted very often, since only 7% of recent bond-strength studies have used this protocol (Van Meerbeek et al., 2010). In a recent study comparing both micro-bond methodologies, it was shown that the micro-shear values were about 1/3 of the micro-tensile values, while no difference in failure analysis was observed (Yildirim et al., 2008).

1.6 Classification of contemporary bonding systems A good classification of adhesives is indispensable for maintaining an overview of the current field. It has been mentioned how the main bonding mechanism of current bonding systems can be regarded as an exchange process involving substitution of inorganic tooth material by resin monomers which, with in situ polymerisation, become micro-mechanically interlocked in the micro-porosities created. Diffusion is the principal way to obtain such micro-mechanical

retention. Recently, more evidence has corroborated the potentially important role of additional chemical interactions at the biomaterial-tooth interface, especially with regard to bond stability (Yoshida et al., 2004b). Accordingly, modern bonding strategies can be divided into (1) an etch-and-rinse [or totaletch (Kanca, 1992)], (2) a self-etch (or etch&dry), and (3) nowadays also a selfadhesive approach (Sarr et al., 2010). The strength of this classification lies in its simplicity and its scientific basis. Each category is characterised by a specific bonding mechanism, a specific and distinct application protocol and by a specific interfacial ultrastructure as best imaged using TEM.

1.6.1 Etch-and-rinse The multi-step etch-and-rinse approach requires a phosphoric acid conditioning step that at enamel creates deep etch-pits in the HAp-rich substrate, and at dentine demineralises up to a depth of 3-5 Îźm to expose a HAp-deprived collagen mesh (Peumans et al., 2005). The next step includes either two separate primer and adhesive resin steps, according to a 3-step procedure, or a single priming step consisting in the application/curing of a combined primer/adhesive resin, in accordance with a simplified 2-step procedure. In other words, the latter approach combines the priming and the bonding steps into one; these adhesives are frequently referred to as 'one-bottle adhesives' and misleadingly suggest a single application step. The final purpose is the micro-mechanical interlocking upon diffusion and in situ polymerisation of monomers into the enamel etch-pits, the opened dentinal tubules and the mainly organic substance remaining at acid-etched dentine. Ideally, resin tags should bond to the tubule walls strongly enough to exceed the cohesive

strength of their resin components. This not only effectively seals the restoration margins in the long term, but also safeguards the more vulnerable bond to dentine against degradation (Peumans et al., 2005). On the contrary, etching dentine is a rather aggressive procedure as it dissolves and removes (through rinsing) the natural protection of collagen in preparation for creating a resincollagen complex that is susceptible of degradation. True chemical adhesion between collagen and the methacrylate monomers is unlikely, because of the inert nature of collagen fibrils and the low affinity of the monomers for HApdepleted collagen (Van Meerbeek et al., 1998). Consequently, the rather poor adaptation of resin to the collagen fibrils leaves the nanometre-sized gaps accountable for water sorption and degenerative processes (De Munck et al., 2003). This is regarded as the major shortcoming of today’s etch-and-rinse approach. Nevertheless, traditional 3-step etch-and-rinse adhesives are still today considered as ‘gold-standard’ among adhesives, in spite of the rather elaborate and lengthy working procedure. In fact, after ageing procedures in durability studies, the bonding integrity of these adhesives is better maintained (De Munck et al., 2003). For this reason, they are usually employed as control in order to compare the performance of new-generation adhesives in many of today’s bond-strength studies. Likewise, the clinical durability of 3-step etchand-rinse adhesives confirms their generally superior laboratory results (Van Meerbeek et al., 2010). It has been extensively reported that 2-step etch-andrinse adhesives performed clinically less favourably than conventional 3-step etch-and-rinse adhesives (Peumans et al., 2005). Laboratory studies have corroborated these results, ascribing their poorer performance to their higher hydrophilicity and reduced hybridisation potential. In fact, primed dentine is

covered with a layer of non-solvated hydrophobic adhesives in the final phase of 3-step etch-and-rinse approach and the combination behaves as if it were hydrophobic (Brackett et al., 2005, King et al., 2005). This measure reduces the incorporation of water within the interface when the adhesive is applied to the wet dentinal substrate, and at the same time minimises the susceptibility of the bonded interfaces to water sorption in the long run. Conversely 2-step etch-andrinse formulations are somewhat hydrophilic (Tay and Pashley, 2003a). That is, they contain too much HEMA and other hydrophilic monomers that create hydrophilic copolymers. The hydrophilic polymers absorb too much water which causes swelling of resins that are plasticised by water uptake, thereby reducing their mechanical properties. Besides, 2-step etch-and-rinse formulations are associated with greater technique sensitivity than their three steps counterparts, which is understandable as a single solution combines the two separate functions of primer and bonding resin. The primer solvent within etch-and-rinse adhesives is a major factor affecting the handling and performance properties of these materials. Water-based adhesives are believed to be the most forgiving regarding application errors, but the water content in the resultant interface jeopardises the durability. Acetonebased adhesives, on the other hand, have water-free formulations, but require the challenging ‘wet-bonding‘ technique (Tay et al., 1996). Ethanol-based adhesives are thought to be an acceptable compromise with regard to user-handling and performance (Van Meerbeek et al., 2003). It is noteworthy that, irrespective of the number of application steps, acetone-based etch-and-rinse adhesives have generally performed less satisfactorily than their water/ethanol-based alternatives (Van Meerbeek et al., 2010). The use of

acetone-based primers presents monomer diffusion problems. One benefit of acetone is that adhesive monomers are very soluble in this solvent. However, during bonding procedures, the acetone present in the first layer of monomers that is placed on water-saturated dentine mixes with the water (acetone and water being very miscible), this blocks monomer penetration and makes it come out of solution before it has a chance to diffuse far into the interfibrillar spaces (Nakabayashi and Pashley, 1998). With each subsequent application of primer, the acetone may re-dissolve the monomer and permit it to diffuse further into the demineralised dentine. This high technique-sensitivity of acetone-based adhesives must be the reason for their compromised long term clinical data. One of the more recent developments is the use in a 2-step etch-and-rinse adhesive of tert-butanol solvent. This tertiary alcohol has similar vapor pressure to ethanol, but also a better ability for chemical reaction with monomers. Tert-butanol-based XP Bond (Dentsply DeTrey GmbH, Konstanz, Germany) has shown good micro-tensile bond strength data, as well as performed well when tested following a conventional shear bond strength (Manhart and Trumm, 2010).

1.6.2 Self-etch Self-etching primers contain acidic monomers that only dissolve the smearlayer, but do not remove the dissolved calcium phosphates, as there is no rinse phase but only air-drying. The self-etch approach can be further subdivided into a ‘strong’ (pH<1), an ‘intermediately strong’ (pH≈1.5), a ‘mild’ (pH≈2), and an ‘ultra-mild’ (pH≥2.5) self-etch

approach

depending

the

etching

aggressiveness or demineralisation intensity (Van Meerbeek et al., 2011).

Consequently, the morphological features of the hybrid layer produced by selfetch adhesives depend a great deal on the aggressiveness of the functional monomers (Van Meerbeek et al., 2003). TEM images of 'strong' self-etch adhesives applied on dentine strongly resemble the morphological aspect of an etch-and-rinse adhesive with a thick hybrid layer, which is completely devoid of HAp crystals, and with resin tags. The more aggressive the self-etching, the deeper the hybridisation, the more calcium phosphates are dissolved (exposing collagen) and embedded within the interfacial transition zone (Koshiro et al., 2004). Such resin-encapsulated calcium phosphates within the exposed collagen mesh are, however, not very hydrolytically stable and rather soluble, thereby seriously jeopardising the bond longevity. This fact may account for the significantly worse laboratory and clinical bonding efficiency, especially to dentine, offered by 'strong' self-etch adhesives in comparison with the welldocumented and consistently good-performing 3-step etch-and-rinse (Brackett et al., 2002). ‘Intermediately strong’ self-etch adhesives exhibit morphological features that lie between the ‘strong’ and ‘mild’ self-etch adhesives. The latter demineralise dentine only very superficially, giving rise to a rather shallow hybrid layer with submicron dimensions, while still leaving substantial HApcrystals to protect the collagen fibrils (Van Meerbeek et al., 2010). Hence, the resultant hybrid layer consists of a partially HAp-deprived collagen mesh infiltrated by resin. The less intense the self-etching, the more bur-smear occurs. This interferes with the eventual bonding performance (Ermis et al., 2008, Cardoso et al., 2008);; nonetheless ‘mild’ self-etch adhesives seem to cope relatively well with the smear-layer, even when produced by various surface preparation methods such as diamond burs with different grit sizes, air-

abrasion and sono-abrasion (Van Meerbeek et al., 2011). The resin must penetrate through it and engage intact dentine, but how deep through the sound tissue is unknown. TEM revealed that ‘ultra-mild’ self-etch adhesives, such as Clearfil S3 Bond (Kuraray, Tokyo, Japan), decalcified the dentine surface only very superficially, for up to a few hundreds of nanometres, with few collagen fibrils exposed at the interface (Fukuoka et al., 2011). This particular interfacial feature has therefore been termed nano-interaction zone. The relationship between hybrid layer of various thickness and bond strength has been investigated, and neither the thickness of the hybrid layer, nor the length of the resin tags seemed to play an important role regarding the bond strength (Inoue et al., 2001). Similar bond strengths were obtained regardless of hybrid layer depth (Finger et al., 1994, Yoshiyama et al., 1996) and this may be due to the fact that resin retention, as measured by conventional bonding tests, is related to the cohesive strength of adhesive resin that engages the very top of the hybrid layer. Deeper penetration of resin may not increase the cross-sectional area of resin engagement of collagen fibrils, although it may improve the durability of the demineralised collagen. Be that as it may, in spite of the small hybrid layer and the absence of resin tags (little micro-mechanical retention), 'mild' self-etch adhesives such as Clearfil SE Bond (Kuraray, Tokyo, Japan) can reach satisfactory results in terms of bond strength to dentine (Inoue et al., 2001). In two clinical studies, this system was reported to have high retention rates in non-carious class V cavities after 2 and 5 years (Türkün, 2003, Peumans et al., 2007). Together with the finding that the thickness of the hybrid layer and the presence of resin tags do not overly affect the bonding performance (Inoue et al., 2001), additional chemical interaction between

polycarboxyl- and phosphate-based monomers with residual HAp has been proposed as a plausible explanation for the good performance of some smearlayer incorporating adhesives (Yoshida et al., 2004a). The carboxylic and phosphate groups that render these monomers hydrophilic and that function as proton donors, have been proven to bond ionically with calcium in HAp (Yoshida et

al.,

2000).

Some

functional

monomers,

the

10-MDP

(10-

methacryloyloxydecyl dihydrogen phosphate), have been shown to interact with this residual HAp through primary ionic binding (Yoshida et al., 2004a). The chemical bonding promoted by 10-MDP turned up to be not only more effective, but also more stable in an aqueous environment than that produced by other functional monomers like 4-MET (4-methacryloyloxyethyl trimellitic acid) and phenyl-P (2-methacryloyloxyethyl phenyl phosphoric acid), in this order, as revealed

by AAS

(atomic absorption

spectroscopy) and

XPS

(X-ray

photoelectron spectroscopy) (Yoshida et al., 2004a). The hydrolytic stability of the monomer itself is also important, especially with regard to bond durability. Whereas micro-mechanical retention is thought to provide resistance to 'acute' de-bonding stresses, the relevance of additional chemical bonding is suggested to lie in durability and survival of adhesion (Van Meerbeek et al., 2003). This two-fold micro-mechanical/chemical bonding mechanism closely resembles that of glass-ionomers (Coutinho et al., 2006) and gives an explanation for the actual bonding effectiveness of ‘mild’ self-etch adhesives. With regard to the actual bonding effectiveness, it is now very clear that the in vitro and in vivo performance of an adhesive greatly depends on its specific ingredient composition. Since the ability to bond chemically is monomerspecific, the interaction with tooth substrate is dependent to a great extent on

the kind of acidic functional monomer and eventually the overall composition of the adhesive. For this reason, not all the self-etching adhesives available today are equally effective, and there is a certain variation in bonding performance between these agents. This variation is appreciably larger than that of etch-andrinse adhesives, all relying on the use of phosphoric acid and the subsequent infiltration of monomers. There are two major classes of self-etching adhesives: self-etching primer adhesives where the primer-treated dentine is covered with a separate, solvent-free, more hydrophobic adhesive; and self-etching adhesives that are not covered with more hydrophobic adhesive resins. 2-step self-etch adhesives are those requiring the application of a different, more hydrophobic adhesive resin to cover the dried hydrophilic primer. Likewise conventional 3-step etch-and-rinse adhesives. In these bonding systems hydrophilic and ionic resin monomers are only contained in the primers. Making hydrophobic coatings upon hydrophilic resins better seals the interface, improves the bond durability and the blend results hydrophobic like it happens for 3-step etch-and-rinse adhesives (Brackett et al., 2005, King et al., 2005). As a result, these adhesives tend to approach the bonding laboratory effectiveness and the clinical performances of 3-step etch-and-rinse adhesives in terms of low annual failure rates (Peumans et al., 2005). Similar to the combined primer-adhesive resin solution of 2-step etch-and-rinse adhesives, the most rapid 1-step self-etch formulations (all-in-one) are complex mixtures of both hydrophilic and hydrophobic components. When applied to dentine, the resulting homogenised resin films behave as hydrophilic resins and consistently achieved lower bond strengths, compared with the multi-step selfetch and etch-and-rinse versions (Inoue et al., 2001). Due to polymerisation

shrinkage of resin-based composites, a high configuration factor (C-factor) in deep class I cavities leads to a certain amount of stress when the material is bonded (Nikolaenko et al., 2004). Many 1-step self-etch adhesives might not be able to withstand the resultant high polymerisation shrinkage stress, giving rise to early failures and postoperative sensitivity (De Munck et al., 2005). The lower bonding efficiency has been attributed to the fact that hydrophilic phases or domains extended across the entire adhesive resin layer and allowed water movement through it (Sauro et al., 2007). Additionally, quantitative fluid permeability was consistent with the number of water droplets per unit area. These hydrophilic domains may be visualised as fine, discrete silver grains (i.e. the spotted mode of nano-leakage) in the adhesive after immersion in ammoniacal silver nitrate (Tay et al., 2002a). Therefore, accumulating evidence indicates that the use of solvated mixtures of hydrophilic and hydrophobic monomers creates polymerised resin coatings permeable to water from the outside environment as well as from the host dentine. In addition, these agents are quite sensitive to application mistakes. In particular, the air-drying step subsequent to their application is critical to minimise the amount of solvent and water in the adhesive layer as much as possible. It is known that thick adhesive layers of simplified adhesives decrease bond strengths massively (Zheng et al., 2001). On the other hand, this air-drying operation should also be performed in such a way that an adequate amount of monomers is kept at the surface to provide satisfactory mechanical properties to the adhesive layer. In narrow and complex cavities, the right balance between too much and not enough air-drying is difficult to achieve; while the adhesive may still pool in the cavity corners, the

amount of monomers on the cavity walls may already be too low (Van Landuyt et al., 2005).

In light of all the drawbacks attributed to simplified adhesives, conventional 3step etch-and-rinse adhesives and ‘mild’ 2-step self-etch adhesives are today the benchmarks for high-quality adhesion to both enamel and dentine in routine clinical practice. In spite of the improved ease-of-use and faster application, a simplified application procedure so far seems to entail a reduced bonding effectiveness, and the benefits of these adhesives should therefore be trade off against their major shortcomings. In particular, when bonding only to enamel, the etch-and-rinse approach is definitely preferred, indicating that simple micromechanical interaction appears sufficient to achieve a durable bond to this tissue. Altogether, when bonding to both enamel and dentine, selective etching of enamel followed by the application of the 2-step self-etch adhesive to both enamel and dentine currently appears the best choice. This is a consequence of the fact that whenever bonding solely to dentine is required, a mild 2-step self-etch approach is superior, as the additional ionic binding with residual HAp enhances bond durability (Van Meerbeek et al., 2010). In this case, the better durability may also be attributed to partial dissolution of the apatite minerals which appear to exert a protective effect on collagen degradation. The latter was corroborated by the minimal involvement of endogenous MMP-2 and MMP-9 in interfaces bonded by the mild 2-step selfetch adhesives (De Munck et al., 2010).

Nevertheless, mild 2-step self-etch adhesives are not totally immune to bond degradation since they are hydrophilic and enzymatic hydrolysis of ester bonds may still occur over time. Failure of water to be completely removed is probably another reason why bond degradation may occur in this class of adhesives, as shown by the presence of silver nano-leakage (Liu et al., 2011c).

1.6.3 Self-adhesive The third adhesive approach is based on the technology of glass-ionomers and their auto-adhesive capacity. These restorative materials have a specific composition, containing polyacrylic acid (PAA), alkenoic copolymers, glass-filler particles and water. Diverse formulas of glass-ionomers are on the market, varying in their uses. Conventional glass-ionomer cements (GICs) are two-component systems, consisting of an ion-leachable flouroaluminosilicate glass powder and an aqueous solution of polycarboxylic acid. GICs are set by means of an acid-base neutralisation reaction, with the product of a hydrogel salt acting as a binding matrix. On mixing of the two-components, hydrogen ions liberated from the polymeric acid attack the glass, causing the release of metal ions such Al 3+, Ca2+ etc. In turn, these cations cross-link the acid to form an insoluble matrix which subsequently bonds to the residual silicate structure formed on the glass surface. Polymeric materials are being increasingly introduced and used in dentistry as cements, filling materials, dentine substitutes and treatment of early carious lesions. Various efforts have therefore been directed also to the combination of methacrylate technology and conventional glass-ionomer chemistry. When resin

components are added to GICs, these materials are referred to as hybrid-GICs, resin-ionomers or resin-modified glass-ionomer cements (RMGICs) (Tyas and Burrow, 2004). The composition of RMGICs is variable but typically consists of vinyl-modified polyalkenoic acid, water soluble methacrylates such as HEMA, an ion-leachable glass and water (Aranha et al., 2006, Bertacchini et al., 1999, Nakabayashi and Takarada, 1992). Other ‘self-adhesive’ materials are the so-called self-adhesive luting composites that have been introduced some years ago to adhesively lute indirect ceramic restorations (Hikita et al., 2007, Radovic et al., 2008, Monticelli et al., 2008). They are often mistakenly termed as ‘self-etching’, while they interact only very superficially with dentine without clear signs of demineralisation (Van Meerbeek et al., 2010). GICs are the only true ‘self-adhesive’ materials as they can bond dentine micromechanically, through infiltration of the collagen network which is exposed by a short PAA pre-treatment, in combination with chemical bonding obtained by ionic interaction of carboxyl groups from the acid with calcium ions of remaining HAp crystals (Sidhu and Watson, 1998). In detail, the weak acidic molecule removes the smear-layer and exposes collagen fibrils up to a depth of about 0.5-1 µm. The abundant functional carboxylic groups of PAA ‘grab’ HAp simultaneously at different and remote sites, before the other glass-ionomer components diffuse and establish a micromechanical bond following the principle of hybridisation. The ability of GICs to provide a shallow but uniform hybrid layer, along with their additional capability

to chemically bond to the dentinal substrate, are considered convenient in terms of resistance to long-term hydrolytic degradation. Owing to the mild and partial demineralisation, HAp crystals can be micromorphologically distinguished around the collagen fibrils within the hybrid layer. Typically, a 'gel phase' closely attached to the hybrid layer can be observed with some glass-ionomers. This amorphous phase on top of the interface has been reported to represent a salt of calcium polycarboxylate (Yoshida et al., 2001). The basic difference with the resin based self-etch approach is that glassionomers are self-etching through the use of a relatively high-molecular-weight (from 8000 to 15,000) polycarboxyl-base polymer. This limits their infiltration capacity and is the reason why only hybrid layers of little depth are created. Moreover, because of this high molecular weight, they cannot penetrate phosphoric-acid-decalcified dentine. In this case the etchant would demineralise dentine to a depth greater than that penetrated by the adhesive, leaving the collagen network unshielded by mineral or polymer and thus exposed to oral fluids. Consequently, such aggressive conditioners should not be used in conjunction with glass-ionomers (Van Meerbeek et al., 2003). The chemical bonding and good adhesion to dental tissues is undeniably a positive aspect of glass ionomers, in addition to the fact that the glass itself may be used as a fluoride reservoir. Other advantages offered by conventional GICs are biocompatibility and thermal expansion coefficients matched to tooth structure. Also, the surface of these cements resists mild acid attack and staining by certain agents, such as those that can occur in the mouth (Wilson, 1991).

Nonetheless, conventional GICs are far from ideal as restorative materials. Glass ionomers lack the aesthetic benefits of resin - while they can attain a match to tooth colour, the characteristic translucency of natural tooth structure has been difficult to achieve with these materials. They offer low wear resistance when placed on chewing surfaces, low tensile strength and fracture toughness since they are brittle and tend to fracture relatively easily. In addition, they are susceptible to attack by moisture in the initial stages of setting period and this leads to crazing, lack of resistance to abrasion and susceptibility to fracture under high shear stresses. Into the bargain, conventional GICs have short working time, are porous and difficult to finish to a smooth surface (Attin et al., 1996, Bell and Barkmeier, 1994, Croll et al., 1993, de Gee et al., 1998, Kerby et al., 1997, Leevailoj et al., 1998, Mitchell et al., 1999). In order to address the latter of these issues, metals have been added to the cement to reinforce their structure (McLean, 1990); however, this practice is diminishing in use (Tyas and Burrow, 2004). The introduction of resin-modified glass-ionomer cements (RMGIC) has resolved some of the problems inherent with GICs. Like conventional glassionomer cements, RMGICs have a setting reaction including an acid-base reaction between the ion-leachable glass and the polyalkenoic acid, but also a photoactivated polymerisation reaction involving unsaturated side-chains on the modified polyacid takes place. The two networks of polyacid and ionically crosslinked polyalkenoate chains provides the structural integrity of the cement and can be cross-linked through pendant methacrylate groups on the polyalkenoate molecules (Attin et al., 1996, Allen et al., 1999).

Compared to conventional GICs, advantages of these materials include a shortened setting time, early strength development, decreased early moisture sensitivity and command set. However, RMGICs require many improvements before they can be considered to be superior restorative material, as they have been found to leach cytotoxic compounds (Geurtsen, 2000, Geurtsen, 1998). Several in vitro studies demonstrated that most of the commercial resinmodified glass-ionomer cements present more intense cytotoxic effects than conventional glass-ionomer cements (Aranha et al., 2006). The high cytotoxicity of resin-modified glass-ionomer cements is probably caused by leachable resin components, such as HEMA, which has frequently been added to their chemical composition. Leached residual monomer can easily diffuse through the dentinal tubules due to its hydrophilic property and low molecular weight, and reach dental pulp cells (Bouillaguet et al., 1996, Gerzina and Hume, 1996, Hamid et al., 1998, Kan et al., 1997, Souza et al., 2006). All light-cured systems suffer from the limited depth penetration of visible light. Hence, layering techniques are required, despite the fact of being timeconsuming. Another significant disadvantage of resin ionomer is the hydrophilic nature of poly- hydroxyethyl methacrylate, which results in increased water absorption and subsequent plasticity and hygroscopic expansion (Pashley et al., 1998, Yap and Lee, 1997).

Chapter 2: Strategies for preventing resindentine bond degradation

2.1 Introduction The ultimate goal in the design and development of dental adhesives is to render a stronger and more durable adhesion to hard dental tissues - despite the severe conditions in the oral environment. Regarding the different mechanisms of degradation, corresponding strategies to preserve the intact hybrid layers have been proposed and practiced in vitro and in vivo. They are as follows: (i) increasing the degree of conversion and esterase resistance of hydrophilic adhesive (Marchesi et al., 2010); (ii) the use of broad-spectrum inhibitors of collagenolytic enzymes, including novel inhibitor functional groups grafted to methacrylate resins monomers to produce anti-MMP adhesives (Breschi et al., 2010a); (iii) the use of cross-linking agents for improving the resistance of uncross-linked or mildly cross-linked collagen matrices to degradation by MMP and cathepsins (Castellan et al., 2010b); (iv) ethanol wetbonding to completely replace water from the extrafibrillar and intrafibrillar collagen compartments, increase resin uptake and produce better sealing of the collagen matrix, using hydrophobic monomers that absorb much less water over time and lead to more durable bonds because of improved resistance to hydrolytic attack (Sauro et al., 2010); and (v) progressive water replacement from the resin-sparse regions of the hybrid layer with hierarchical deposition of intrafibrillar and extrafibrillar apatite crystallites to exclude exogenous collagenolytic enzymes and fossilise endogenous collagenolytic enzymes (Liu et al., 2011a).

2.1.1 Improvement of degree of conversion and esterase resistance Resin degradation is directly related to the absorption of water, given that hydrolytic attack on ester linkages by increased acidity/basicity of resin components or by salivary esterases can both take place in presence of water. Human saliva incorporates sufficient amounts of pseudocholinesterase and cholesterol esterase, which operate synergistically to degrade dimethacrylates (Finer et al., 2004). Beside, the use of hydrophobic photoinitiators such as camphorquinone for hydrophilic adhesives in the presence of water does not lead to an optimal polymerisation of these adhesives (Ye et al., 2009a). These challenges provided the rationale for the development of watercompatible and esterase-resistant dentine adhesives (Spencer et al., 2010). It has been suggested that the degree of conversion of hydrophilic adhesive components

could

enhanced

using hydrophilic photoinitiators and

compatible accelerators, as well as experimental bulky/branched esteraseresistant hydrophilic urethane-modified resin monomers (Hayakawa et al., 2005, Ye et al., 2009a, Park et al., 2010, Park et al., 2008). Other groups have also employed water-soluble photoinitiators (Cadenaro et al., 2010, Ikemura et al., 2009) to improve the polymerisation of hydrophilic adhesives within the confines of water-rich dentine substrates. Concomitant cross-linking of polar functional groups on methacrylate side-chains may augment the hydrophobic character of the hybrid layer after preliminary infiltration of the hydrophilic resin monomers into the partially or completely demineralised collagen matrix. Increasing the degree of conversion of monoand dimethacrylate resins, with the associated reduction in the number of unreacted pendant functional groups, may also diminish the susceptibility to

esterase hydrolysis of the polymerised adhesive layer and resins that infiltrated the hybrid layer. The application of these new photoinitiators, accelerators, and cross-linking resin monomers is likely to intensify the dynamic mechanical properties of the resin-dentine interface immediately after polymerisation, as well as decreasing the heterogeneity and nanophase separation of the polymer matrices in the presence of water (Park et al., 2010). These strategies, however, will not result in a quantum increase in the ability of adhesive resin monomers to infiltrate the demineralised collagen matrix entirely, particularly deep inside the aqueous compartments of the collagen fibrils (Liu et al., 2011c).

2.1.2 Inhibition of enzyme-catalysed hydrolytic cleavage of collagen Several authors have attempted to determine the benefits of employing synthetic MMP inhibitors during bonding procedures (Liu et al., 2011c) in relation to the dentine collagenolytic and gelatinolytic processes, responsible for the degradation of collagen fibrils within incompletely resin-infiltrated hybrid layers (Zhang and Kern, 2009) and the loss of quasi-static mechanical properties of the collagen matrix (Tezvergil-Mutluay et al., 2010a). The quest for developing specific MMP inhibitors, capable of potently and selectively inhibiting or blocking the uncontrolled activity of individual MMPs, is a nearly three-decade endeavour and only few selective and effective drugs with the desired properties have emerged (Li and Wu, 2010). A number of rationally designed MMP inhibitors have shown some promise in the treatment of pathological conditions in which MMPs are suspected to be involved, such as

cancer, arthritis, and other diseases associated with tissue remodelling. Most of these inhibitors, however, have performed poorly in clinical trials. The reasons for this failure include shortcomings in the chemistry of these compounds, namely, broad MMP sub-type selectivity, toxicity and metabolic lability. With respect to the selection of appropriate non-specific inhibitors for dentine matrixbound MMPs, the structural homology of the catalytic domains among members of the MMP family represented a favourable characteristic because the application of such inhibitors to acid-etched dentine was analogous to a topical application method and involved nanogram quantities, making systemic toxicity a less critical issue, albeit still important. Chlorhexidine (CHX), a widely used antimicrobial agent, has been successfully employed to suppress the activities of MMP-2, MMP-8, and MMP-9 (Gendron et al., 1999). Several investigators have used this cationic bis-biguanide as a nonspecific MMP-inhibitor, demonstrating CHX-related improvement in terms of bond strength preservation and hybrid layer stability, revealed

as a lower

interfacial nano-leakage expression compared to control specimens after different periods of in vivo ageing. In split-mouth design experiments (Carrilho et al., 2007b, Hebling et al., 2005), teeth of a quadrant were treated with 2% CHX between acid-etching and bonding procedures; control teeth on the other side had no CHX treatment. After 6 to 14 months, control hybrid layers revealed extensive loss of cross-banded collagen with the formation of voids in the hybrid layer, while the CHX-treated teeth had normal hybrid layers without any signs of degradation. Fourteen-month results of microtensile bond strength revealed that the mean (Âąstandard deviation) bond strength had decreased from 29.3Âą9.0 to 19.0Âą5.2 MPas in the control group (Carrilho et al., 2007b).

In contrast, the experimental CHX-treated teeth showed a mean 1-day bond strength of 32.7Âą7.6 MPa, while the 14-month value was 32.2Âą7.2 MPa. Thus, there was virtually no loss of bond strength in the CHX-treated specimens. In a fluorescein-labelled functional assay, collagenolytic activity of mineralised dentine powder was suppressed for the 73% by protease inhibitors, and almost 100% by 0.2% concentrations of CHX (Moszner et al., 2005). Another in vitro study confirmed the protective role of 0.2% CHX after 1 year of storage of bonded specimens in artificial saliva, suggesting that lower concentration of CHX can be equally effective compared to 2% concentrations (Breschi et al., 2009). One of the advantages of using CHX is represented by its substantivity since it is able to bind to mineralised dentine for at least 12 weeks (Mohammadi and Abbott, 2009). Additionally, it has been shown that demineralised dentine can bind more CHX than mineralised dentine and that it may remain attached to demineralised dentine after bonding (Kim et al., 2010d). The same study proved that CHX is not de-bound by HEMA and this may be the reason for the longterm effectiveness of CHX as a MMP inhibitor in resin dentine bonds. Relatively large amounts of CHX remained bound to partially and completely demineralised dentine incubated in phosphate-buffered saline for at least 8 weeks, and no de-binding occurred after the first half-hour of incubation (Carrilho et al., 2010). Since the binding mechanism is electrostatic in nature and is reversible (Blackburn et al., 2007), CHX molecules could be eventually displaced by competing cations derived from dentinal fluid or saliva and leach out of the denuded collagen matrix. There is a prevailing notion that its binding to demineralised dentine simply postpones instead of permanently arresting

bond degradation (Mai et al., 2009). This may account for the discovery that bonds made to CHX pre-treated acid-etched dentine with commercial adhesives and water-wet bonding techniques were conserved after 9 months but not after 18 months, with severe hybrid layer degradation at the 18th month observation (Sadek et al., 2010a). Such a concern provided the rationale for chemically grafting CHX to resin monomers to produce CHX-methacrylates (Luthra and Sandhu, 2005) and including these resin monomers with anti-MMP potentials into dentine adhesives. The introduction of quaternary ammonium methacrylate resin monomers with assorted anti-MMP properties (Liu et al., 2011c) has recently attracted the attention of investigators. A study employing one of the best-known commercially

available

quaternary

methacryloyloxydodecylpyridinium

ammonium

bromide

(MDPB),

methacrylates,

12-

confirmed

the

that

experimental antibacterial adhesive systems employing MDPB-containing primer or/and bonding-resin could produce an effective bond under in vivo conditions (Imazato et al., 2007). This could have accounted, in hindsight, for the in vitro and in vivo observations that resin-dentine bonds degraded after one year when Clearfil SE Bond (Kuraray Medical Inc., Tokyo, Japan) was used as the self-etching primer, while bonds created in the same study with the MDPBcontaining self-etching primer Clearfil Protect Bond (Kuraray) were well preserved after one year, with more pronounced water treeing observed in the former adhesive under both ageing conditions (Donmez et al., 2005). These resin monomers were included into dentine adhesives to avoid the displacement of electrostatically bound CHX and other potential non-resin-

conjugated MMP inhibitors from denuded collagen matrices of the hybrid layers during

ageing.

Copolymerisation

CHX-methacrylates

quaternary

ammonium methacrylates with other dimethacrylate resin monomers made CHX unable to de-bind and leach out from incompletely resin-infiltrated hybrid layers. With respect to self-etch adhesives, chlorhexidine has been included directly into primers (De Munck et al., 2009) (Zhou et al., 2010). However, there were critical restrictions when CHX was directly incorporated into polymerisable resin monomer formulations. Although incorporation of 1-2% CHX directly into resin blends did not have any significant effect on their degree of conversion, such an approach adversely influenced the mechanical properties of the polymerised resins. It was demonstrated that the addition of even 1% CHX to a variety of resin blends with different hydrophilicity reduced the modulus of elasticity (i.e., stiffness) of the polymerised resins by 27-48% (Cadenaro et al., 2009a). This decreasing effect continued over time as CHX release from the resins was related to water-induced swelling (Hiraishi et al., 2008). In addition, release of CHX from a polymer matrix was pH-dependent, with more being released at lower pH values (Anusavice et al., 2006). More important still, CHX proved to be beneficial only for etch-and-rinse adhesives, as it could not bind to the collagen matrix in the presence of an acidic environment (Curtis and Watson, 2008). Another non-specific MMP inhibitor, GM 6001 (galardin), is often used as a reference inhibitor in generic MMP assay kits. Breschi et al. aimed to determine the effect of a synthetic MMPs inhibitor (galardin) used as an experimental primer on acid-etched dentine prior to the application of an etch-and-rinse adhesive (Breschi et al., 2010a). The inhibitory effect of galardin on dentinal

MMPs was confirmed by zymographic analysis, as complete inhibition of both MMP-2 and -9 was observed. Interfacial nano-leakage expression after ageing revealed reduced silver deposits in galardin-treated specimens compared to controls. Furthermore, the use of galardin had no effect on immediate bond strength, while it significantly decreased bond degradation after 1 year. Others have shown that polyvinylphosphonic acid (PVPA) (Tezvergil-Mutluay et al., 2010b) and benzalkonium chloride (BAC) (Tezvergil-Mutluay et al., 2011) also possess generic anti-MMP activities and could be used as an alternative to CHX to prevent collagen degradation within hybrid layers. Chemically modified tetracyclines (CMTs) (i.e., tetracyclines that lack antibiotic activities but retain their anti-MMP activities) are effective non-specific MMP inhibitors and have been used as MMP inhibitors in experimental caries (Sulkala et al., 2001). This study suggested that MMPs could have an important role in dentine caries pathogenesis, and that MMP inhibitors could be useful in the prevention of caries progression. However, CMTs have not been used to prevent the degradation of hybrid layers, since they may stain teeth with a purple hue after photo-oxidation of the tetracycline. Much work has also been done on designing cathepsin K inhibitors (Teno and Masuya, 2010). Selective inhibitors of cathepsin K could be in fact promising therapeutic agents for the treatment of diseases characterised by excessive bone loss, such as osteoporosis. Representative inhibitors have demonstrated antiresorptive activity both in vitro and in vivo and therefore are promising leads for therapeutic agents (Thompson et al., 1997). Expansion of these inhibitor concepts can be envisioned for the many other cysteine proteases implicated for therapeutic intervention. However, little is known about whether these drugs

that are targeted at inhibiting specific cathepsins are useful for direct application to acid-etched dentine or incorporation into self-etch adhesives for preventing the degradation of type I dentine collagen scaffolds. In contrast with this, CHX has been reported to be effective against cysteine proteinases produced by Porphyromonas gingivalis (Houle et al., 2003) and human recombinant cysteine cathepsins (Scaffa et al., 2012). Thus, it is important to determine if CHX could be used as a non-specific inhibitor for both MMPs and cysteine cathepsins associated with intact and carious dentine.

2.1.3 Use of collagen cross-linking agents As the major component of the dentine organic matrix, fibrillar type I collagen plays a number of structural roles, such as to provide the tissue with viscoelasticity, forming a rigid, strong space-filling biomaterial (Cheng et al., 1996). Type I collagen is a heterotrimeric molecule composed of two Îą1 chains and one Îą2 chain that is comprised of 3 domains: the NH2-terminal non-triple helical (N-telopeptide), the central triple helical, and the COOH-terminal non-triple helical (C-telopeptide) domains (Bedran-Russo et al., 2009). Dentine collagen fibrils are stabilised by lysyloxidase-mediated covalent interand intramolecular cross-links (Yamauchi and Shiiba, 2002) which increase the tissue resistance to thermal denaturing and enzymatic degradation (Kuboki and Mechanic, 1982). In addition, intrinsic collagen cross-links provide the tensile properties of collagen molecules (Yamauchi, 2000). On this account several in vitro studies suggested that introducing supplementary cross-links to acid-demineralised dentine collagen could help to

reduce the susceptibility of additionally cross-linked dentine collagen to enzymatic degradation by collagenases, improve its short-term mechanical properties and increase the stability of the resin-dentine interface. The effects of selective collagen cross-linkers, such as glutaraldehyde, genipin, carbodiimide and proanthrocyanidin, applied during adhesive procedures has been investigated in several studies over the last few years (Al-Ammar et al., 2009, Macedo et al., 2009, Bedran-Russo et al., 2009, Bedran-Russo et al., 2010, Castellan et al., 2010a, Castellan et al., 2010b). Glutaraldehyde, a synthetic cross-linking agent, is widely used as fixative agent (Nimni et al., 1998) and has been reported to improve mechanical properties of various collagen-based tissues (Bedran-Russo et al., 2008, Bedran-Russo et al., 2007, Charulatha and Rajaram, 2003, Ritter et al., 2001, Yannas, 1992, Silver, 1994). However, despite its ability to induce cross-links in collagen, glutaraldehyde is also known for its cytotoxicity (Sung et al., 1999). Genipin, a naturally occurring cross-linking agent, not only has shown to improve the mechanical properties of various protein-based biomaterials (Bedran-Russo et al., 2007, Tsai et al., 2002), but also presented low toxicity when compared with glutaraldehyde (Frujikawa et al., 1987). Increased resistance to collagenase challenge and mechanical properties of collagen-based materials have been reported following treatment with carbodiimide, a cyanamide isomer, able to assemble amino acids into peptides. It presents very low cytotoxicity when compared to glutaraldehyde as the urea derivative, released when the cross-link is generated, is easily rinsed from the collagen, leaving no residual chemicals (Khor, 1997).

Proanthocyanidin, widely present in fruits, vegetables, nuts, seeds and flowers is a potent antioxidant cross-linking agent with vast biological activities, including inhibition of MMP-2 and MMP-9 (Matchett et al., 2005). Recently, the use of a grape seed extract, mainly composed of proanthocyanidin, has been shown to improve the mechanical properties of demineralised dentine (BedranRusso et al., 2008, Bedran-Russo et al., 2007, Frujikawa et al., 1987). Since dentine collagen is already highly cross-linked, it is questionable if the increase in resin-dentine bond longevity can be explained only by augmentation in cross-linking density. Current literature suggests that this MMP resistance may be attributed to silencing of MMPs and probably other exogenous collagen degradation enzymes via conformational changes in the enzyme 3-D structure (Busenlehner and Armstrong, 2005). Theoretically, this may be achieved via irreversible changes induced within the catalytic domain or allosteric inhibition of other modular domains that co-participate in collagen degradation (SelaPasswell et al., 2010). Undeniably, the use of cross-linking agents increases the resistance of uncrosslinked or mildly cross-linked collagen matrices to degradation by bacterial collagenases (Avila and Navia, 2010, Ma et al., 2010). Even though there is no evidence that the catalytic domain of collagenolytic MMPs can be cross-linked to inactivate their functions, oxidative cross-linking of adjacent tryptophan and glycine residues in the catalytic domain of MMP-7 by hypochlorous acid, a potent oxidant produced by the myeloperoxidase system of phagocytes, resulted in inactivation of the catabolic activity of this enzyme (Fu et al., 2004). These observations indicated that specific structural motifs are

important for controlling protein modification by oxidants and suggested that pericellular oxidant production by phagocytes might limit MMP activity. The use of cross-linking agents may also play a part in MMP silencing via allosteric control of non-catalytic domains. Lauer-Fields et al. indicated that hemopexin-like domains collaborate with catalytic domains in collagen catabolism by properly aligning the triple helix and coupling conformational states to facilitate hydrolysis. Therefore, the catalytic domains in collagenolytic MMPs can cleave non-collagen substrates, but the hemopexin-like domain of these enzymes is necessary for them to first unwind and then cleave the three triple-helical fibrillar elements of the collagen molecule in succession (LauerFields et al., 2009). The MMP family has developed at least two distinct mechanisms for collagen unwinding and cleavage. These distinctive mechanisms underly a drastically different mode of interaction with triple helical fibrillar collagen I, according to which the MMP domain is involved in binding (Gioia et al., 2007). One mechanism is characterised by binding (likely through the hemopexin-like domain) and cleavage of alpha-1 and/or alpha-2 chains without distinguishing between them and keeping the gross conformation of the triple helix (at least during the first cleavage step). The other instead involves preferentially binding of the alpha-1 chains (likely through the fibronectin-like domain, grossly altering the whole triple helical arrangement of the collagen molecule and cleaving preferentially the alpha-2 chain. Regardless of which of the two collagen-binding mechanisms is involved, crosslinking of either the hemopexin-like or fibronectin-like domains may contribute to

inactivation of the associated MMPs and reduction in their collagenolytic efficacy. This hypothesis appears to be supported by the results of a study that analysed the tissue properties of pericardium from young calves and pigs after Crosslinking with glutaraldehyde or carbodiimide. Cross-linking with glutaraldehyde completely abolished gelatinase activities, while the use of carbodiimide was less effective; but, interestingly, a relative reduction of MMP-9 versus MMP-2 was detected (Calero et al., 2002). Cathepsin K is also allosterically regulated by modifiers, such as sulphated glycosaminoglycans (GAGs), that bind outside of its active catalytic site (Novinec et al., 2010). It has been shown that at physiological plasma pH the enzyme fluctuates between

multiple

conformations

which

are

differently

susceptible

macromolecular inhibitors and can be manipulated by varying the ionic strength of the medium. Thus, GAGs may act as natural allosteric modifiers of cathepsin K, exploiting the conformational flexibility of the enzyme to regulate its activity and stability against autoproteolysis. It is possible that the use of cross-linking agents may alter the GAG-cathepsin allosteric interaction and “trap” the enzyme in a certain conformation that inactivates its collagenolytic activity. Cross-linking may also affect MMP activities that are usually modified by noncollagenous proteins (Malla et al., 2008). MMPs, like other proteinases, can undergo autolytic degradation once activated in vivo. In dentine, MMP activities and resistance to degradation may be regulated by serum glycoprotein fetuin-A, a member of the cystatin superfamily

100

(Ray et al., 2003), SIBLINGs Bone Sialoprotein and Dentine Matrix Protein-1 (Fedarko et al., 2004). Taken together, these data suggest that cross-linking of these non-collagenous proteins may indirectly silence MMPs via inactivation of the functional domains of these glycoproteins. The mechanical properties of dentine are a fundamental aspect of restorative procedures, since dentine constitutes the greatest volume of tooth structure. Similarly to the utilisation of non-specific inhibitors, the major limitation in the use of cross-linking agents is that a water-rich, resin-sparse collagen matrix with poor mechanical properties is retained within the hybrid layer. Demineralised dentine collagen has a modulus of elasticity of less than 8 MPa. Even if these demineralised collagen fibrils can be stiffened 50X with cross-linking agents, the resulting modulus of elasticity (ca. 0.4 GPa) is still far inferior to that of resininfiltrated dentine (ca. 3-5 GPa) (Ito et al., 2005, Chiaraputt et al., 2008) and mineralised dentine (ca. 20 GPa) (Kinney et al., 2003). Besides, chemical cross-linking of collagen does not alter the intrinsic collagen molecular stiffness (Liao et al., 2005). Thus, these very flaccid collagen fibrils are susceptible to creep and subsequent fatigue rupture after prolonged function (Fung et al., 2009). This accentuated the need for alternative strategies that enable the dynamic mechanical properties of denuded collagen matrices to be improved or regained as a mechanism to prevent the degradation of resin-dentine bonds (Liu et al., 2011c).

101

2.1.4 Ethanol-wet bonding technique Two main reasons have been advocated to interpret the relatively poor infiltration of etch-and-rinse adhesives into water wet-dentine: (1) substantial discrepancies between molar concentrations of water (55.6 moles/L) and that of typical dentine adhesives comonomer blends (2.3-3.9 moles/L), making it nearly impossible for comonomers to expel and replace all of the water from collagen fibrils (Pashley et al., 2007); (2) presence of hydrophobic dimethacrylates in dental adhesives with low solubility in water, that can give rise to macro, micro or nanophase separation when applied on acid-etched water wet-dentine (Eliades et al., 2001). After rinsing the acid-etched dentine, the water is replaced with absolute ethyl alcohol. Being completely miscible with water, and since the volume of ethanol applied is far in excess of the amount of water in the collagen matrix, ethanol replaces all the water in the interfibrillar spaces and at the top of the dentinal tubules. As a result, ethanol-solvated hydrophobic resin blend consisting of BisGMA and TEGDMA should penetrate deep into ethanol-saturated collagen without undergoing phase changes (Sadek et al., 2008), and any residual layer of ethanol would allow the infiltrating monomers to dissolve and create a more intimate and resistant association between collagen and resin (Tay et al., 2007). Whereas etch-and-rinse adhesives applied to water-saturated dentine invariably resulted in a diffusion gradient of resin infiltration within the collagen matrix, studies with two-photon laser confocal microscopy (Sauro et al., 2009b) and micro-Raman spectral analysis (Shin et al., 2009) indicated that a relatively

102

homogeneous distribution of hydrophobic resins within the hybrid layer could be achieved with ethanol-wet bonding. Replacement of water by ethanol increased resin uptake of hydrophobic BisGMA/TEGDMA mixtures and produced better sealing of the collagen matrix by bringing the Hoy solubility parameter for Î´t of the ethanol-saturated matrix very close to that of ethanol-solvated bis-GMA/TEGDMA mixtures. The Î´t value of water-saturated mixtures is in fact too far away from the Î´t of BisGMA/TEGDMA, indicating that the solvated comonomers are not miscible with water-wet bonding (Pashley et al., 2007). Two variants of the ethanol-wet bonding technique have been described. In the simplified version, 100% ethanol was applied to water-saturated acid-etched dentine for 1 min preceding the application of ethanol-solvated hydrophobic resin comonomer blends (Nishitani et al., 2006, Sauro et al., 2010). The rationale was to provide a method of application of hydrophobic resin comonomers to acid-etched dentine within a clinically relevant time frame. However, this approach proved to be extremely susceptible to operator-specific handling and did not completely reduce dentine permeability without the use of adjunctive tubular occlusion agents (Sadek et al., 2007, Cadenaro et al., 2009b, Sauro et al., 2009c), even after three absolute ethanol applications (Sadek et al., 2010b). In the progressive ethanol replacement adaptation of the technique, water was gradually removed from the collagen matrix via a series of ascending ethanol concentrations (Sadek et al., 2007, Sadek et al., 2010a). Nevertheless, this technique version was time-consuming and impractical for clinical application.

103

In both cases, when ethanol replacement is not meticulously performed to prevent water-saturated collagen from exposure to air, the surface tension present along the air-collagen interface can easily result in collapse of the collagen matrix and prevent optimal infiltration of the adhesive monomers (Osorio et al., 2010). Also, the technique is sensitive to moisture contamination (Sadek et al., 2007, Sadek et al., 2008). Since hydrophobic monomers are immiscible with water, contamination of these resin monomers with as little as 5% of water on the ethanol-saturated dentine substrate resulted in a 25% reduction in tensile strength of the experimental hydrophobic adhesive to dentine (Sadek et al., 2008). As a result, whether experimental ethanol-wet bonding can become a clinically applicable technique for bonding to deep, vital dentine remains uncertain. Moreover, ethanol-wet bonding is not suitable for self-etch adhesives, with being water a prerequisite for the ionisation of their acidic resin monomer components (Hiraishi et al., 2005). Ethanol-wet bonding is now considered a philosophy rather than a proper bonding technique, due to its clinical impracticality. However, it represented a major contribution to adhesive technology, since the reasoning behind it revealed the critical barrier to overcome in dentine bonding with contemporary etch-and-rinse and self-etch adhesives. Ethanol replacement of water-saturated dentine provided an opportunity for higher resin uptake and made it possible to obtain hybrid layers characterised by wider interfibrillar spaces alongside collagen fibrils with reduced fibrillar diameter (Tay et al., 2007, Hosaka et al., 2009).

104

Given that these intrafibrillar spaces were penetrated by resin during infiltration, no space was left empty after polymerisation (Liu et al., 2011c). As a matter of fact, when ethanol-wet bonding was meticulously performed, neither nanoleakage (Sadek et al., 2008) nor intrafibrillar remineralisation (Kim et al., 2010a) could be detected. Such a phenomenon has never been observed in hybrid layers created with hydrophilic etch-and-rinse adhesives bonded to water-wet dentine, or with water-containing self-etch adhesives (Liu et al., 2011c). Generally, milder versions of self-etch adhesives have the ability to infiltrate the interfibrillar spaces of a partially demineralised collagen matrix via the mechanism of simultaneous etching and resin infiltration. Nevertheless, intrafibrillar spaces created by both self-etch and etch-and-rinse adhesives are amenable to remineralisation by apatite crystallites (Kim et al., 2010b). Such an observation provided indirect evidence that both classes of commercially available adhesives are incapable of completely replacing water from the intrafibrillar collagen compartment, and that resin monomers can entirely fill this compartment if it is saturated with ethanol, but not if it contains water.

2.1.5 Restoring the mineral phase of the collagen matrix Sadek et al. proved that bonds made to ethanol-saturated dentine with an experimental hydrophobic adhesive did not degrade over an 18-month ageing period with preservation of hybrid layer integrity (Sadek et al., 2010a). This was an important observation, since it addressed the issue that MMPs are not capable of collagenolysis in the absence of water as a functional medium.

105

Even if a decreasing gradient of resin monomer diffusion within acid-etched dentine was not present, with a resultant phase of demineralised collagen matrix at the base of the hybrid layer (Pashley et al., 2011), and hydrophilic adhesives could infiltrate to the full extent immobilising collagenolytic enzymes, resin hydrolysis via esterases and water sorption would subsequently result in reactivation of those immobilised enzymes. Encapsulating MMPs and cathepsins with hydrophobic resins represented another innovative way of immobilising the catalytic and allosteric domains of these enzymes to inactivate their functional activities, and highlighted the concept of water replacement from the collagen intrafibrillar compartments as the ultimate goal in extending the longevity of resin-dentine bonds (Liu et al., 2011c). Such a method is conceptually similar to the technique of molecular imprinting to produce abiotic polymers with enzymatic functions (Takeuchi and Hishiya, 2008), with the exception that the enzyme substrate is not removed to expose their functional sites. During collagen mineralisation, bulk water and loosely bound water are progressively removed from the internal compartments of the collagen fibrils and replaced by apatite crystallites (Chesnick et al., 2008). Collagen fibrils that are stabilised by intrafibrillar and interfibrillar apatite crystallites in mineralised tissues do not degrade over time. Hard tissue fossils from the late Cretaceous era (65 million years ago) were well preserved at the microstructural and molecular levels and responded to collagenase digestion (Avci et al., 2005). Analysis of these data suggested that MMP-bound collagen

106

must have been functionally immobilised by apatite crystallites for the integrity of those collagen fibrils to be preserved. Apart from increases in mechanical properties (Kinney et al., 2003, Balooch et al., 2008), a major role played by the hierarchical deposition of apatite in mineralised collagen is the exclusion of molecules larger than water (ca. 18 Da) from the mineral-protein biocomposite (Lees and Page, 1992). Conversely, when apatites are replaced by water in clinical dentine bonding, the life span of these resin-dentine bonds has seldom been shown to last for more than 10 years (Van Dijken et al., 2007). For these reasons, replacement of water within incompletely resin-infiltrated hybrid layers, accompanied by inactivation or silencing of collagenolytic enzymes via remineralisation, emerged as a viable strategy to overcome the critical barriers currently restraining the longevity of resin-dentine interfaces (Liu et al., 2011c). The physical exclusion of exogenous collagenolytic enzymes (activated cathepsin K, 27 kDa; activated MMP-2, 67 kDa; bacterial collagenase, 68-130 kDa; activated MMP-9, 85 kDa) by apatite represents the postulate of the “enzyme exclusion” mechanism that preserved archeological collagen from degradation (Nielsen-Marsh et al., 2000). Studies on collagenase hydrolysis of dentine have corroborated the protective role played by the mineral phase on collagen degradation (Klont and ten Cate, 1991). Similar to a host of growth factors and signaling molecules, endogenous MMPs and cathepsins become “fossilised” (Smith, 2003) but retain inside the mineralised dentine matrix their biologic characteristics. These are restored upon removal of the mineral phase, provided that the demineralisation agent is not strong enough to denature these molecules. As

107

collagen mineralises, free and loosely bound water is progressively replaced by apatite. This physiologic dehydration mechanism (Chesnick et al., 2008) guarantees that the internal environment of the mineralised fibril continues to be relatively dry to protect the integrity of the entrapped bioactive molecules.

2.1.5.1 Guided tissue remineralisation It has been recently suggested that the evolutional mechanism of molecular immobilisation of the functional activity of collagenolytic enzymes could be recapitulated in man-made resin-dentine bonds with a guided tissue remineralisation strategy. To this end, Tay and coworkers (2011) have proposed that acid-etched dentine should be treated with a biomimetic remineralising primer just before bonding. This primer would contain nanoparticles of amorphous calcium phosphate and other reagents that would remineralise any water-rich, resin-poor regions in hybrid layers (Kim et al., 2010c, Kim et al., 2010f, Liu et al., 2011b, Liu et al., 2011c, Ito et al., 2012). Biomimetic mineralisation (bioremineralisation) is a strategy that utilises nanotechnology principles to mimic what occurs in biomineralisation (Tay and Pashley, 2008). This method replaces water from resin-sparse regions of the hybrid layer with apatite crystallites that are small enough to occupy the extrafibrillar and intrafibrillar compartments of the collagen matrix (Tay and Pashley, 2009). By restoring the enzyme exclusion and fossilisation properties of mineralised dentine, it should be possible to preserve the longevity of resindentine bonds (Kim et al., 2010a). Enormous progress has been made over the last few decades in understanding the biomineralisation processes in mammalian tissues such as bone, enamel

108

and dentine. Some knowledge has been acquired about the role of proteins in achieving a morphologically controlled deposition of mineral as opposed to precipitation of unstructured agglomerates of crystals. Biologically mineralised tissues have remarkable hierarchical structures that have evolved over time in order to achieve great functions in a large variety of organisms. Mineralised crystals are typically formed in an organic matrix with precise regulation of synthetic mechanisms through proteins. These proteins are in dynamic equilibrium with their environment, thus resulting in fluctuations and tissue remodelling. In consequence, several conditions are required to restore lost mineral constituents in demineralised dentine. The internal consistency of the collagen structure is the primary requisite. Organic phases play in fact a key role in templating the structure of mineralised tissues; therefore, their matrices should be sound as a scaffold for the mineral crystals to grow (Kusboki et al., 1977, Gonzalez-Cabezas, 2010). Second, there should be residual mineral crystals to serve as growth centres, or, at the very least, there should be newly formed nucleation sites in case of complete demineralisation (Xu et al., 2010). Last but not least, mineral sources containing calcium and phosphorous should be supplied to the lesion (Peters, 2010). In the biomineralising scheme, there are no apatite seed crystallites present in the organic scaffold. In other words, this approach cannot rely on the presence of remnant dentine matrix proteins within a demineralised collagen matrix and is compelled to mineralise reconstituted collagen that is devoid of mineralisationpromoting proteins (Kim et al., 2010g).

109

As a result, biomineralisation has to proceed through an unconventional sequence of chemical reactions that includes homogeneous nucleation. The latter is not a thermodynamically favourable process and necessitates alternative kinetically driven protein/polymer-modulated pathways in order to lower the activation energy barrier for crystal nucleation (Wang and Nancollas, 2009). For most vertebrates, collagen fibrils alone cannot initiate tissue mineralisation. These controls, imposed by matrix and soluble proteins leading to the sequential steps of phase transformation, entail sequestration of the amorphous mineral phase by poly(anionic) extracellular matrix proteins (ECM). Noncollagenous ECM derived from the secretory calcium-binding phosphoprotein family are necessary for regulating bone and dentine mineralisation. They are also needed for controlling the dimension, order and hierarchy of carbonated apatite apposition within mineralised hard tissues. Collagen is an active scaffold in the formation of oriented hydroxyapatite platelets, with domains of charged amino acids at the border of the gap and overlap zones acting as nucleation sites. Modelling of collagen fibrils shows that these nanosized, positively charged regions are used for mineral infiltration as well as charge-charge attraction. This leads to the deposition of a dense network of pre-nucleation clusters bound by polyanions within any nanosized region, and their subsequent transformation into amorphous calcium phosphate and, finally, oriented crystalline hydroxyapatite inside the fibrils (BarrĂ¨re et al., 2006). Over the last two decades, there has been intensive research into isolating ECM and examining their role in biomineralisation (He and George, 2004).

110

However, the complexity of biological proteins limits our understanding of the functional groups. Meanwhile, use of native or recombinant matrix proteins as nucleating agents for in situ biomineralisation is not yet economically viable in routine restorative dentistry therapies. Thus, scientists have resorted to the use of poly(anionic) acid molecules to mimic the functional domains of naturally occurring mineralisation-promoting proteins (Liu et al., 2011c). Additional acidic matrix phosphoproteins are employed as templates to promote crystal nucleation and growth within the collagen structure (George and Veis, 2008). In nanotechnology terminology, such pathways are examples of a bottom-up synthesis (Wong et al., 2009), wherein nanoscale materials are created by a particle-based self-assembly process (Gower, 2008). Bioremineralisation of resin-dentine bonds adopted the guided bottom-up assembly by using two poly(anionic) analogs of acidic matrix proteins to separately mimic the sequestration and templating functional motifs that are present in naturally occurring matrix protein molecules (Liu et al., 2011c). Blocks from different silicate-based materials (as a source of calcium and hydroxyl ions) were immersed in a biomimetic analog of matrix proteins made of simulated body fluid containing a polycarboxylic acid such as polyacrylic acid (Girija et al., 2004, Tay and Pashley, 2008, Kim et al., 2010e) and a phosphorus-based

analog

matrix

phosphoproteins

such

polyvinylphosphonic acid (George and Veis, 2008, Gu et al., 2011). The role of polyacrylic acid is to mimic aspartic acid as well as serine-rich mineralisation-promoting C-terminal domain (ASARM) of cleaved DMP-1

111

(Gericke et al., 2004) and stabilise amorphous calcium phosphate into fluidic nanoprecursor particles (Gower, 2008, Nudelman et al., 2010). This makes possible the inflitration of the microfibrillar spaces by amorphous calcium phosphate and the formation of initially amorphous, intertwining mineral strands within the collagen fibrils (Kim et al., 2010g). Instead, polyvinylphosphonic acid is used as a collagen-binding template to induce the nucleation and growth of apatite from the initial amorphous mineral phase (Gajjeraman et al., 2008). In addition, this molecule has been shown to possess MMP-inhibiting properties that prevented collagen degradation during remineralisation (Tezvergil-Mutluay et al., 2010b). The two biomimetic analogs (sequestration and templating) must be utilised in concert to reproduce the dimension and hierarchy of the apatite crystallites that are found in natural mineralised dentine. In fact, when only polyaspartic acid was used as the sequestration analog, infiltration of amorphous calcium phosphate

nanoprecursors

into

reconstituted

collagen

fibrils

produced

intrafibrillar mineralisation that apparently lacked the hierarchical order of apatite arrangement in natural mineralised collagen (Deshpande and Beniash, 2008). Conversely, large extrafibrillar mineral spheres were deposited around the collagen matrix when only a templating analog was employed (Li and Chang, 2008). Hybrid layers created by both etch-and-rinse and self-etch adhesives have shown to be remineralisable with the bottom-up biomimetic approach. For etchand-rinse adhesives, Mai and co-workers detected, after one month, apatite crystallites in both extrafibrillar and intrafibrillar spaces of the denuded collagen

112

matrices, with hybrid layers remineralised to 80-90% of their thickness after 2-4 months (Mai et al., 2010). Apatite deposition was almost exclusively identified from the intrafibrillar spaces in case of self-etch bonding systems, as interfibrillar spaces were filled with adhesive resin (Kim et al., 2010b, Kim et al., 2010c). Unfortunately, guided bottom-up mineral assembly using biomimetic analogs of dentine matrix proteins is a slow process, since it involves at least two kinetically driven pathways and usually takes 3-4 months to complete. In experimental studies, long times were also required because remineralisation was performed via a lateral diffusion mechanism by the immersion of specimen slabs in the medium containing dissolved biomimetic analogs without direct dentine-material contact (Liu et al., 2011c). Thus, while the mineral constituents is being restored, the region of the exposed collagen fibrils is still susceptible to hydrolytic degradation by MMPs. Furthermore, the water-filled denuded collagen fibrils remain flaccid and exhibit weak mechanical properties (Bertassoni et al., 2009). The modulus of elasticity of resin-infiltrated dentine beams increased hundredfold after biomimetic remineralisation as a result of intrafibrillar remineralisation of the collagen matrices (Gu et al., 2010). A limitation of that study, however, was that the resin-infiltrated dentine beams (macro-hybrid layers) were evaluated en masse by three-point bending. Since remineralisation does not occur in locations of the collagen matrices that are occupied by resins, three-point bending is insensitive for evaluating the localised increases in modulus of elasticity of the remineralised collagen.

113

Although the reincorporation of mineral into the demineralised dentine matrix does not represent a full recovery of its functionality, it still plays a very important role, since the remineralised remnant crystallites in the subsurface of the tissue may be much more resistant to subsequent acid attacks (Featherstone, 1996). An experimental calcium-silicate-based hydrophilic composite has been developed for the sustained release of calcium and hydroxyl ions (Kim et al., 2010g). The inclusion of reactive calcium-silicate mineral powder as filler in hydrophylic resin may be an innovative method for the biomimetic remineralisation of apatite-depleted dentine surfaces and to prevent the demineralisation of hypomineralised/carious dentine, with potentially great advantage in clinical applications. However, while this method proved to be useful for remineralising resin-dentine interfaces with thin adhesive layers, remineralisation was hampered in presence of thick adhesive layers (Dickens and Flaim, 2008). Thus, alternative strategies are being developed to deliver calcium, hydroxyl, and phosphate ions to the base of the hybrid layers created by etch-and-rinse adhesives, where remineralisation is most needed.

2.1.5.2 Top-down remineralisation via epitaxial growth Remineralisation of apatite-depleted, partially demineralised dentine is not new. Reports on remineralisation of carious dentine appeared in the dental literature more than half a century ago. Currently, dental literature abounds with reports on the use of bioactive glass particles (Vollenweider et al., 2007), calciumphosphate-based composites (Peters et al., 2010) and fluoride-releasing glass-

114

ionomer cements (Ngo, 2010) for remineralising partially demineralised carious dentine. Fluoride, which is not a functional motif for biomineralisation, has been shown to enhance conventional remineralisation over existing seed crystallites (Liu et al., 2011c). In nanotechnology terminology, remineralisation techniques currently used in dentistry represent a top-down approach (Wong et al., 2009). This approach creates materials using scaled down versions of a bulk material that incorporates nanoscale details of the original material. Partial demineralisation of mineralised collagen matrix by acids derived from bacteria or dentine bonding procedures creates the seed crystallites necessary for this top-down remineralisation approach. The orientation of those remineralised crystalline lattices is pre-determined by the lattice of the original seed crystallites. However, remineralisation does not occur in locations where nuclei for crystallisation are completely absent, as demonstrated with the use of a strontium-based glass-ionomer cement (Kim et al., 2010f). Non-denatured intact collagen is remineralisable as long as seed crystallites are present as nidi for heterogeneous nucleation of calcium phosphate phases (Koutsoukos and Nancollas, 1981). In fact, partially demineralised dentine is considered to have a better capacity to restore its original mineralised state because it contains remnant mineral crystals and noncollagenous phosphoproteins that can act as nucleation sites for remineralisation (Clarkson et al., 1991, Bertassoni et al., 2011). Detailed electron-microscopic analysis of crystallites in various zones of caries lesions has also confirmed that remineralisation occurs by growth of existing crystals (Featherstone, 1996).

115

Moreover, the mineral precipitated may work as a constant site for further nucleation of mineral ions present in the oral cavity, facilitating continuous remineralisation over time. Unlike the biomimetic bottom-up remineralisation approach that takes months of protracted chemical reactions to complete, top-down mineral fabrication of apatite crystals proceeds rapidly via epitaxial growth over existing seed crystallites (Liu et al., 2011c). Furthermore,

traditional

remineralisation

epitaxial

growth

thermodynamically favourable process that overcomes the energy barrier of homogeneous nucleation (Jiang and Liu, 2004). Despite their complex hierarchical structures, the basic building blocks of teeth and bones during both initial and later formation stages are nanosized mineral particles. Initial contacts between the organic matrix and mineral nuclei is a key factor for the control of crystallisation pathways and the organisation of hierarchical structures of teeth and bones at different length scales (Wang and Nancollas, 2008a, Wang and Nancollas, 2008b). The earliest heterogeneous nucleation events in the presence of the organic template and subsequent growth involve various possible precursor phases (amorphous or crystalline) to the final mineral phase. Classic crystallisation theory assumes that crystals nucleate and grow from elementary building blocks (ions, molecules) in a supersaturated solution, although phase transformations may also occur in the later stages. The association of solution species to form “metastable intermediate precursors”

116

(Onuma and Ito, 1998), or “growth units”, that subsequently dissolve as the precipitation reactions proceed, is a crucial initial step (Eanes et al., 1965). The nature of the primary mineral phase leading up to mature bone and dentine mineral apatite remains controversial. Both brushite and octacalcium phosphate have been implicated as possible precursors to the formation of biogenic hydroxyapatite with trace of carbonate and fluoride. Moreover, in vivo calcifications have also suggested the involvement of an initial amorphous calcium phosphate phase (ACP) followed by transformation to the final HAp product (Mahamid et al., 2008). According to Posner, mineral apatite derives from calcium phosphate clusters [Ca9(PO4)6] packing randomly with interfacial water to form ACP precursors (Posner, 1985). This theory is supported by the presence of several calcium phosphate growth inhibitors such as magnesium that stabilise the amorphous state (Barrére et al., 1999, Root, 1990). Although extensive investigations of calcium phosphate crystallisation have been performed, many have studied the final structures and morphologies and have not emphasised the need to consider the molecular contacts between mineral and matrix that drive nucleation or the thermodynamic and kinetic controls imposed by matrix and soluble proteins during the nucleation stage. Current results and concepts of crystal nucleation and growth at the molecular level, and the role of site-specific interactions in crystallisation, provide possible mechanisms of calcium phosphate crystallisation that are related to the mineralisation of teeth and bones (Wang and Nancollas, 2008a, Wang and Nancollas, 2008b).

117

However, the detailed physical and chemical processes by which nucleation control is established and the thermodynamic and kinetic parameters that define those processes remain largely unknown. These precursor nanoparticle phases grow further in size via ion-by-ion attachment (Niederberger and Cölfen, 2006) and aggregation, with local loss of solvent,

undergoing

amorphous-crystalline

transformations

phase

transformations en route to a thermodynamically stable macro-crystal. According to Ostwald’s rule (Ostwald, 1897), normally the first occurring phase in polymorphism is the least stable and closest in free energy to the motherphase, followed by other phases in order of increasing stability. The formation, dissolution, and transformation of calcium phosphates depend on the nature of the calcium phosphate body (particle size, crystallographic features, density) and the nature of the solution (composition, pH, temperature). Most calcium phosphates are sparingly soluble in water, and some are very insoluble, but all dissolve in acids. Their solubility, defined as the amount of dissolved solute contained in a saturated solution when particles of solute are continually passing into solution (dissolving) while other particles are returning to the solid solute phase (growth) at exactly the same rate (Wu and Nancollas, 1998), decreases with the increase in temperature and in pH (de Groot, 1983). Each calcium phosphate phase possesses its own thermodynamical solubility. For example, at pH=7 and 37ºC, HAp is the most stable phase (Barrère et al., 2006). However, these thermodynamical considerations are under equilibrium conditions, and therefore they do not take into account kinetics that dictate the formation of one or the other phase under dynamic conditions.

118

In vivo, the interactions between calcium phosphates and their “biological surroundings” are highly complex due to the non-equilibrium conditions and due to the undefined amount of compounds playing a role in these interactions. The second important factor in the stability of the calcium phosphates is the characteristics of the solution in which these salts are formed or placed, namely the solution supersaturation in free calcium and phosphates ions (Tang et al., 2001). At a given pH and temperature, a free calcium and phosphate ion containing solution can be categorised in three different states: (i) the stable (undersaturated) condition, when crystallisation is impossible; (ii) the metastable condition, when spontaneous crystallisation of calcium phosphate salt is improbable,

although

the

concentrations

are

higher

than

the

ones

corresponding to the salt solubility. If a crystal seed were placed in such a metastable solution, growth would occur on the seed; (iii) the unstable condition, when spontaneous crystallisation of calcium phosphate is probable, but not inevitable (Barrère et al., 2006). Extracellular fluids that are supersaturated for calcium and phosphate may induce the nucleation and growth of new calcium phosphate crystals (Barrère et al., 2006). Almost all mineralised tissues are highly hierarchical at many different length scales. At the lowest level they often consist of crystals formed of thin plates of irregular shapes. Their sizes range in length from 20 Å for the smallest particles, to 1100 Å for the largest particles (Kim et al., 1995).

119

This size is not arbitrary; rather, it seems to give biominerals such as bone and tooth remarkable physical characteristics (Wang and Nancollas, 2008a, Wang and Nancollas, 2008b). These mineral crystals expose a very large surface area to the extracellular fluids, which is critically important for the rapid exchange of ions with these fluids. Minerals start to nucleate into the holes and pores present in the collagen fibrils (Glimcher, 1987). This heterogeneous nucleation is catalysed by the presence of phosphated esters groups (Glimcher et al., 1984) and carboxylate groups (Rhee et al., 2000) present in the collagen fibrils. Subsequently, the growth, or mineralisation, takes place along the collagen fibrils, eventually interconnecting all of the collagen fibrils. There is uncertainty over the specific morphology of the mineral crystallites. The shape of each inorganic crystal is often related to the intrinsic unit cell structure, correspondingly the same material can exhibit diverse crystal morphologies due to different surface energies of the faces and the external growth environment. Kinney et al. (Kinney et al., 2001) used small-angle X-ray scattering to indirectly assess the micromorphology of the apatite crystallites in dentine and suggested that minerals have a rod-like shape near the pulp while they are more plate-like, with approximately 5 nm thickness, nearer the dentine-enamel junction. Similarly, Nalla et al. (Nalla et al., 2005) using transmission electron microscopy confirmed early observations from Boyde (Boyde, 1974) and suggested the presence of needle-like crystallites in the intertubular dentine region. Lowenstam and Weiner (Lowenstam and Weiner, 1980), also using TEM, evaluated the ultrastructure of the crystallites in bone (which has a similar model of mineralisation) after removal of its organic structures and found that

120

the average length and width of the crystallites was 50 and 25 nm instead, with an approximate thickness of 2-3 nm, resembling plate-like structures. The mineral phase is classified as intrafibrillar and extrafibrillar, according to its location with respect to the collagen fibrils. The former is confined within or immediately adjacent to the gap zones of collagen fibrils, extending amid tropocollagen molecules, while the latter lies in the interstitial spaces between the fibrils. Structurally intact collagen fibrils that retain their banded characteristics

and

intermolecular

crosslinks

are

considered

physiologically remineralisable (Landis et al., 1993, Landis, 1996). The concept of remineralisation through the growth of existing apatite crystals is based on the incorporation of ions (calcium, hydroxyl, and phosphate) spontaneously from the oral fluid or, alternatively, from external sources with tailored treatments that deliver the same ions into the surrounding fluids and elicit a positive response at the remnant crystallites within the subsurface (Featherstone, 1990, ten Cate and Featherstone, 1991). The newly formed minerals, in their turn, may act as sites for further nucleation promoting a continuous remineralisation over time when in presence of environmental mineral ions. The capability of a material to induce the formation of apatite on demineralised dentine (remineralisation ability) is strictly related to the biointeractivity and bioactivity, i.e. the ability to evoke a positive response from the biological environment. Bioactive materials have been proven to promote bone formation and to form a stable bond to bone (Hench, 1997) and prediction of a material's bioactivity can be made both in vivo and in vitro (Kokubo and Takadama, 2006).

121

The spontaneous formation of HAp on a material surface upon storage in a simulated body fluid (SBF) is taken as evidence of bioactivity (Piskounova et al., 2009). The control of metallic ion release from dentine bonding agents must be considered as an attractive approach to enhance the biological capability of adhesives for dental tissue engineering. A bioactive dentine bonding agent which forms HAp on the surface would have several benefits including remineralisation of adjacent tooth substance, possible closure of gaps between material and tooth and potentially better bond strength over time (less degradation of bond). In general, today’s dental restorative materials are not considered to be bioactive and no dentine bonding systems with proven capability to induce dentine remineralisation are available on the market (Gandolfi et al., 2011b). Currently, only a few attempts to develop bioactive dental materials have been published based on either calcium aluminate (ceramics) or via addition of calcium phosphates to resin based materials (Engqvist et al., 2004, Skrtic et al., 1996).

2.1.5.3 Key objectives in the design of bioactive dentine bonding systems There is a necessity for restorative materials that bond strongly and durably to dentine; at the same time, there is the demand for these materials to be aesthetically pleasing. Resin-based restoratives meet the latter qualification, but there is much room for refinement in the case of the former. It is possible that the incorporation of additional materials into the resin-bonding procedure can produce this enhancement - specifically, incorporating bioactive agents. Because of the ability of such materials to promote the formation of apatite in

122

aqueous environments that contain calcium and phosphate (e.g. saliva), their presence at the bonded interface could ameliorate the quality of the resindentine bond through mineral deposition throughout the course of nano-leakage events. Restorative compounds capable of filling voids by crystal deposition can be categorised as self-healing materials and may protect resin-tooth bonds because they heal nanometre- and micrometre-sized voids before and after ageing. The ability of restorative materials to fill in voids, making them selfrepairing, has great potential to enable restorations to have longer lifetimes due to this self-healing ability. In view of the clinical demand on engineered dental tissue, new biologically active adhesive/primer formulations have to exhibit: I.

light-curable characteristics with controlled water sorption and solubility behaviour of the cross-linked network in water and oral fluids.

II.

A hydrophilic nature to interact with oral fluids and absorb water for the necessary ion movement.

III.

Ability to release calcium, hydroxyl, and phosphate ions in a sustained manner (bioavailability of remineralising ions).

IV.

Potential to completely replace water from the resin-sparse regions of the hybrid layer with redeposition of thermodynamically stable, apatitic tooth mineral (bioactivity).

Alkalinising activity within the resin-dentine interface to prevent MMPs activity. Hydroxyl ions released during the hydration reaction may also create unfavourable conditions for bacterial survival and proliferation. Antibacterial properties are primarily required at the dentine-restoration interfacial region. The presence of residual bacteria within dentine further

123

increases the risk of reinfection and secondary caries, in particular when using dental composites lacking any antimicrobial activity. These said features thus form the requisites for a successful, advanced ionreleasing dentine bonding system.

2.2 Development of ion-releasing adhesives comprising bioactive fillers Many modifications and improvements have been made to resin bonding products in the years since their inception. Several “generations” of bonding agents have been developed in the last few decades, but the advancements have mainly increased ease of use and reduced technique sensitivity. To date, however, little progress has been made with regard to reducing the propensity of the adhesive towards nano-leakage, though the achievement of improved bond strengths has significantly reduced gross gaps from occurring (Pashley et al., 2011). One possible modification that may significantly reduce the occurrence and extent of nano-leakage within the hybrid layers is represented by the incorporation of biologically active agents into the bonding process; materials which may create a chemical bond in addition to the micromechanical one, or potentially precipitate mineral into any open space that remains after polymerisation of the adhesive. The development of a bioactive ion-releasing dentine bonding system with therapeutic ability to remineralise mineral-depleted sites within the bondeddentine interface is currently one of the main targets of the dental biomaterial research (Tay and Pashley, 2008, Liu et al., 2011b, Peters et al., 2010, Bresciani et al., 2010, Moshaverinia et al., 2011).

124

In the remineralising process the bioavailability of mineral ions from restorative materials is the basic requirement to enhance the apatite formation and the mineralisation of the dentinal tissue in the presence of mineral ions (calcium, phosphate, fluoride) from the oral fluid. Gaining full understanding of the remineralisation process may be difficult, since both physical and chemical processes are relevant. The site and the amount of mineral deposition are probably determined by the physical condition (mineral distribution profile and transport mechanism) and by the chemical process (deposition). Notwithstanding this, there is little doubt that significant remineralisation of dentine lesions can occur under both in vitro and in vivo conditions. In vitro (Dickens and Flaim, 2008, Dickens et al., 2003, Ryou et al., 2011) and in vivo (Peters et al., 2010, Bresciani et al., 2010) studies suggested in fact that the amount of minerals and the mechanical properties (i.e. hardness and modules of elasticity) of dental hard tissues may be favourably increased. In particular, Bertassoni et al. (Bertassoni et al., 2009) proved that a continuous delivery of calcium (Ca2+) and phosphates (PO4-3) ions may induce remineralisation and mechanical recovery of mineral-deficient dentine due to a fine association of the minerals with the organic matrix. Agents that enhance and/or promote remineralisation of dentine lesions are part of

new

era

dentistry

aimed

controlling

the

demineralisation/remineralisation cycle, depending upon the microenvironment around the tooth (Rao and Malhotra, 2011). Today a variety of agents that aid in remineralisation of dental hard tissues are available commercially. Remineralising agents can be incorporated into different

125

products

and

restorative

materials could

become

vehicles for these

remineralising agents as it occurs with pit-and-fissure sealants, dentifrices, chewing gums, and rinses. Peters et al. (Peters et al., 2010) demonstrated by means of electron probe elemental microanalysis (EPMA) techniques an increased mineral content in caries affected dentine treated with resin-based materials containing calcium phosphate cements. Furthermore, a number of studies (Moshaverinia et al., 2011, Ryou et al., 2011, ten Cate and van Duinen, 1995) have recently demonstrated that mineral precipitation in residual caries affected dentine can also be encouraged by GICs or RMGICs. Conventional GICs offer adhesion to tooth structure and therefore they are not as prone to leakage as are resin-bonded restorations, when used in appropriate situations (Masih et al., 2011). They have been used as liner/base materials, gaining popularity because of beneficial properties such as biocompatibility and ability to release fluoride, available for the formation of a less soluble fluorapatite (Wiegand et al., 2007). Despite the great mass of information on the positive effects of fluoride on enamel, no data have demonstrated the effectiveness of fluoride ions to induce new mineralisation of demineralised dentine and no nucleation of new apatite crystallites within an apatite-free dentine has been identified in the demineralised dentine immersed in a calcium- and -phosphate containing remineralisation media in presence of a glass ionomer cement (Kim et al., 2010g).

126

RMGICs do possess positive qualities, but in this case there is much room for improvement where leaching of resin components, cytotoxicity and durability are concerned (Pashley et al., 1998, Yap and Lee, 1997). Hydroxyapatite itself and some other Ca-containing materials exhibit excellent biocompatibility manifested in minimal tissue toxicity and foreign-body reaction, osteoinductivity, and osteogenicity (Hench, 1999, Kokubo, 1990, Kokubo, 1991, Yuan et al., 2000). The reason for this characteristic may be their ability to release calcium and phosphate ions, which are critical factors in bone metabolism. It has been reported that HAp particles may biomineralise on the surface of chitosan/gelatin network films through hydrogen bonding between COOH, OH -, and NH2 groups of the film and OH- group of the HAp crystals (Li et al., 2009). Therefore, inducing the same interactions between HAp nanoparticles and collagen fibres of demineralised dentine could be an especially promising prospect for mineral replenishment of incompletely resin-infiltrated hybrid layers. Although the remineralisation process would strongly depend on the concentration of the HAp particles adjacent to the collagen fibres and pH of the media, Sadat-Shojai et al. suggested that hydroxyapatite nanorods may be regarded as alternative fillers for dentine bonding systems (Sadat-Shojai et al., 2010). They found that the use of HAp nanorods in low percentages provided adhesives with improved properties. The increase in bond strength might be explained by the fact that the nanofillers can reinforce the adhesive layer at the resin-dentine interface (Atai et al., 2009). However, even if HAp nanoparticles are currently used as coatings in orthopedic and dental implants (Domingo et al., 2003, Ong and Chan, 2000),

127

their poor mechanical properties along with the high water uptake of hydroxyapatite remain a matter of concern and have limited its application for load-bearing materials (Labella et al., 1994, Santos et al., 2001, Domingo et al., 2003).

2.2.1 Calcium/sodium phosphate-phyllosilicates fillers Recently, surface-active glasses have been introduced in many fields of dentistry (Hench, 2006, Margonar et al., 2012, Sauro et al., 2012b). Bioactive glasses have numerous novel features, most important of which are their ability to act as biomimetic mineralisers, matching the body’s own mineralising traits, while also affecting cell signals in a way that benefits the restoration of tissue structure and function (Hench and Paschall, 1974). Bioactive glasses, as opposed to most technical glasses, are characterised by the materials’ reactivity in aqueous environments that contain calcium and phosphate. This bioactivity is derived from their reactions with tissue fluids, resulting in the formation of a carbonated hydroxyapatite layer on the glass/tissue interface, which makes it possible to bond bone and soft tissue without toxicological consequences (Hench and Paschall, 1974). The best-studied and -characterised bioactive glass is the commercially available Bioglass® (formula 45S5), a high biocompatible calcium/sodium phosphate-phyllosilicate originally developed as osteoconductive material but chemically similar to natural tooth mineral (Hench and Andersson, 1993). It is one of the foremost bioactive compounds with an excellent ability to promote calcium phosphate (Ca/P) precipitation and subsequent crystallisation into HAp, when immersed in simulated body fluid, a protein-free solution with ion

128

concentrations similar to those of human blood plasma, or saliva (Andersson and Kangasniemi, 1991). The components in calcium/sodium phosphate-phyllosilicates are basically oxides of calcium, sodium, phosphorus, and silicon at certain weight ratios. The chemical composition is significant. Silica maintains the glass structure, while sodium, soluble in an aqueous environment, aids in maintenance of physiological ionic balance and pH (Hench et al., 1971). Calcium and phosphate are required, as they are the principal constituents of apatite mineral. The proportions of all these constituents have been varied with differing results (Hench et al., 1971). BioglassÂŽ 45S5 is endowed with unique compositional characteristics, namely 45 wt% SiO2, 24.5 wt% Na2O and CaO, and 6 wt% P2O5. Bioactive glasses have traditionally kept the P2O5 fraction constant while varying the SiO2 content. In fact, the network breakdown of silica by OH- was found to be time dependant upon the concentration of SiO2. It is now understood that keeping the silica content below 60 wt% and maintaining a high CaO/P 2O5 ratio guarantees a highly reactive surface. So far, most studies on bioactive glasses have been focused on orthopedic research, due to their ability to form a bone-like apatite layer on their surfaces in the body environment (Gatti et al., 1994, Heikkila et al., 1993, Turunen et al., 1994, Heikkila et al., 1995). The material is reported to have a long record of safety and efficacy as as bone grafting material and has been cleared by the FDA for use in orthopedic surgery (Bauer and Smith, 2002).

129

More than three decades of study have revealed a distinct, though as yet not completely understood, series of chemical reactions that takes place on the material's surface when it is brought into contact with body fluids. These reactions do not depend on the presence of tissue and, within minutes of exposure to the calcium-phosphate-rich environment, begin with leaching and exchange of cations followed by loss of soluble silica. Both steps are characterised by formation of silanols (SiOH). Polycondensation of SiOH to form a hydrated silica gel precedes the formation of an amorphous calcium phosphate layer. The reaction is concluded by the crystallisation of a hydroxycarbonate apatite (HCAp) layer (Hench and Andersson, 1993, Andersson et al., 1988, Kokubo, 1992). More in detail, leaching is characterised by release of alkali or alkaline earth elements (Na+ or K+), usually through cation exchange with H+ or H3O+ ions, and de-alkalinisation of the glass surface layer. Ion exchange is quite rapid because these cations are not part of the glass network, they only impart minor changes to the network by forming non-bridging oxygen bonds. This ion exchange process leads to an increase in interfacial pH, to values > 7.4. Network dissolution occurs concurrently by breaking of -S-O-Si-O-Si- bonds of the glass structure through the action of hydroxyl ions (base catalysed hydrolysis of -S-O-Si-O-Si- bonds). Breakdown of the network is localised and releases silica into the solution in the form of silicic acid (SiOH 4). Last stages involve the development of silica rich and amorphous calcium phosphate layers respectively. Hydrated silica formed on the glass surface by these reactions undergoes

rearrangement

polycondensation

culminating in a silica-rich gel layer.

130

neighboring

SiOH,

During the precipitation reaction, calcium and phosphate ions liberated from the glass along with those of the solution combine to create a calcia-phosphate-rich (CaP) layer on the silica gel. The calcium phosphate phase that stratifies in the gel surface is initially amorphous (a-CaP). Within 3-6 h in vitro, the a-CaP will crystallise into a HCAp structure by incorporating carbonate anions from the solution within the a-CaP phase. Chemically and structurally, this apatite is nearly identical to bone and tooth mineral, thus allowing the body to attach directly to it. These surface reactions from contact with a calcium-phosphaterich fluid to the 100-150 μm formation of a HCAp layer takes in total 12 to 24 h (Hench and Andersson, 1993). Although Bioglass® 45S5 has also been successfully used for dentine remineralisation when directly applied on the dentinal tissue (Curtis et al., 2010), the development of mineralising dentine bonding systems containing bioactive glass ceramic micro-fillers remains an important target to accomplish. When calcium/sodium phosphate-phyllosilicates are utilised for pulp-capping procedures and to reduce tooth sensitivity, the particle size can be as large as 300 μm (Oonishi et al., 1997). Such large particles would be inadequate for incorporation into the resin bonding process, being the dentine tubule diameters on the order of 1-2.5 μm (Marshall et al., 1997). The spaces in the three-dimensional mesh of the collagen network exposed through demineralisation are smaller still. The kinetics of dissolution of small (~5 μm) particles of Bioglass® 45S5 have been studied (Sepulveda et al., 2002), and the material retains its ability to precipitate apatite even at that small size (Vollenweider et al., 2007).

131

Including bioactive glasses in dentine bonding procedures, or in case of methacrylate resins containing bioactive filler phases for potential application as adhesive materials, it is expected that precipitated apatite would bond to specific amino acids within exposed dentine collagen as occurs in bone. Evidence has emerged suggesting that bonds to bone are mechanically strong and believed to be chemical in nature; force is required to separate the tissue from the glass (Wilson et al., 1981). The exact character of the bonds is not known, though it has been suggested that the forming mineral bonds to specific amino acids within the collagen, that is the main structural component of bone (Hench et al., 1971, Hench and Paschall, 1973). Bioactive glasses have already been shown to react favourably with dentine, creating a mechanical bond (Efflandt et al., 2002), forming HCAp similar to tooth mineral (Aoba et al., 1992, Yoshiyama et al., 1996) and displaying antibacterial properties that would be beneficial to the prevention of secondary caries (Zehnder et al., 2004). Filling the adhesive with glass is also expected to minimise polymerisation shrinkage (Say et al., 2006), which is likely to improve marginal integrity and diminish leakage phenomena. To date, the glasses used in these applications have been conventional silica glass or fluoride-releasing, but the use of calcium/sodium phosphate-phyllosilicates as fillers in dentine bonding systems is yet to be attempted. Previously, two types of bioactive glass have been shown to reduce leakage without detrimental effects on the bond strength (Zeiger, 2008). However, the adopted method of application of the glass -vacuum deposition- is not possible in the clinic.

132

2.2.2 Filler phase consisting of calcium silicate cements Ion-releasing adhesives comprising bioactive fillers are meant to be in intimate contact with biological fluid, soft and hard human tissues, hence, they should principally present high biocompatibility as well as potential osteogenic induction. Based on the same principle, experimental remineralising resin-based calcium silicate cements (ion-leaching composites) have been proposed as dental materials with tailored remineralising properties, to be used as restorative baseliner materials in sandwich restorations (Gandolfi et al., 2011b). Interestingly, Gandolfi et al. demonstrated that the presence of experimental composites in contact

with

demineralised

dentine

surfaces

induced

significant

remineralisation of the phosphorous-depleted demineralised dentine surface (Gandolfi et al., 2011b). These findings contributed to and endorse the development strategy for an advanced dentine bonding system containing biointeractive calcium-silicate cements, and thus able to improve the longevity of hybrid layers through selfsealing caused by the formation of apatite in the presence of nano-leakage. Moreover, hydroxyl ions released during their hydration reaction may create unfavourable conditions for bacterial survival and proliferation at the dentinerestoration interfacial region. Calcium silicate cements, such as the Portland cement, have been introduced in dentistry as materials for different endodontic clinical applications due to their biocompatibility and bioactive properties (Parirokh and Torabinejad, 2010). In particular, mineral trioxide aggregate (MTA), a mechanical mixture of Portland cement (75%), bismuth oxide (20%), gypsum (5%) with trace amounts of silica,

133

calcium oxide, magnesium oxide, potassium sulfate and sodium sulfate, was introduced as a root-end filling material (Torabinejad et al., 1995), subsequently expanding to many other applications of root repair and bone healing. These applications included direct pulp capping, repair of root and furcation perforations, and apexification (Torabinejad and Chivian, 1999, Schwartz et al., 1999, Torabinejad et al., 1993). Collectively, calcium-silicate cements are hydrophylic materials able to tolerate moisture (hydraulic materials) and to polymerise and harden (setting) also in the presence of biological fluids. They are in fact porous media, partially or completely saturated with a pore solution. Due to this porosity, materialenvironment mass exchanges occur, in particular ion diffusion through the porosity. The apatite-forming ability (i.e. bioactivity) of these cements have been adequately documented. Their biological behaviour is mainly related to calcium release, to the presence of silicon ions on their surface and to the formation of bonelike apatite. Portland cements are mixtures of dicalcium silicate, tricalcium silicate, tricalcium aluminate, and tetracalcium aluminoferrite (Sarkar et al., 2005). They are ionleaching materials able to release calcium and hydroxyl ions (alkalinising activity) into the surrounding fluids, stimulating the formation of new apatitecontaining tissues (Gandolfi et al., 2010a, Gandolfi et al., 2010b, Gandolfi et al., 2010c, Taddei et al., 2011). However, since calcium-silicate cements lack phosphate, they are able to induce apatite formation only in presence of phosphate-containing fluids (e.g. blood, plasma, saliva, dentinal fluid) (Qi et al., 2012). In a similar environment,

134

they dissolve, releasing all of their major cationic components. Of all the ions released, Ca2+ is the most dominant. Because it is sparingly soluble in biologic fluids, it leads to the precipitation of HAp (Sarkar et al., 2005). The hydration process of Portland cements represents one of the most important factors for the setting and maturation of the cements (Taylor, 1997). Notably, while calcium-silicate particles react with water, a solid-liquid interface forms on the mineral particles and ion dissolution occurs almost immediately. Ca2+ ions are rapidly released (calcium hydroxide formation) and migrate into the solution. A high-pH solution containing Ca2+, OH- and silicate ions is established. Silicates are attacked by OH- ions (hydrolysis of SiO44- groups in alkaline environment) and a CSH phase (calcium-silicates hydrates) forms on mineral particles (Parirokh and Torabinejad, 2010). CSH is a porous, fine-grained and highly disorganised hydrated silicate gel layer containing Si-OH silanol groups and negative surface charges. Actually, at alkaline pH, the deprotonation of silanol groups should predominate (Sanchez and Zhang, 2008) with the consequent formation of SiO- negative groups. The attraction between CSH particles has been reported as a consequence of the very high negative charge density of the CSH particles and the presence of Ca2+ ions (Plassard et al., 2005). The SiO- negative groups induce heterogeneous nucleation of apatite by bonding calcium ions from the mineral particles on the silica-rich CSH surface. While the calcium silicate hydrate gel sets over time to form a solid network, the release of calcium hydroxide increases the alkalinity of the surrounding medium (Pellenq et al., 2009).

135

It was demonstrated how a material covered with an apatitic layer in proximity to calcified tissues forms a chemical bond with the latter (Hench, 1999, Kokubo, 1990, Kokubo, 1992, Yuan et al., 2000) Calcium silicates appeared to bond chemically to dentine when placed against it, possibly via a diffusion-controlled reaction between their apatitic surface and dentine (Sarkar et al., 2005). Sarkar et al. proposed that after the placement of MTA in root canals and its gradual dissolution, with the release all of its major cationic components, HAp crystals nucleate and grow, filling the microscopic space between MTA and the dentinal wall. Initially, this seal is mechanical. With time, they speculated that a diffusion-controlled reaction between the apatite layer and dentine could lead to their chemical bonding. The result is the creation of a seal at the MTA-dentine interface (Sarkar et al., 2005). The first step in the development of alternative Portland-based filling/adhesive materials requires the analysis of the physical characteristics of basic Portland cements used as a principal constituent. In fact, other components may be added for their beneficial effects either on manufacturing processes or on biologic responses and mode of action. The chemical composition and the hydration process may therefore play an important role in the development of a new generation of biomedical materials which would influence physical, chemical and structural characteristics (Gombia et al., 2010). Although different Portland cements (clinkers) are classified under the same UNI EN 197/1 specification (UNI EN, 2007), they may nevertheless present different physical characteristics and this may influence their basic properties and behaviour. For instance, variable amounts of calcium sulfate are usually

136

added to regulate the setting time of Portland cements. Indeed, the presence of calcium sulfate increases the setting time up to a minimum of 45 minutes (Gombia et al., 2010) by reacting with tricalcium aluminate to form hexacalcium aluminate trisulfate, a reactive salt known as ettringite (Gombia et al., 2010). This is normally formed during the setting reaction of Portland cement. It sets in contact with water and undergoes a hydration process in which Ca 2+ and OHions are released from tricalcium silicate into the surrounding environment. At supersaturation levels, it forms calcium hydroxide (portlandite) precipitate and amorphous calcium silicate hydrate gel (Camilleri, 2008). This reaction is initially extremely fast with a subsequent decrease in velocity due to the formation of an excessive layer of calcium sulfate dihydrate (gypsum) which covers and protects the calcium aluminates. On the contrary, the absence of gypsum would leave the tricalcium aluminates to react faster with water forming hydrated calcium aluminates with a cubic structure (Taylor, 1997).

2.2.3 Dye-assisted confocal microscopy imaging of remineralised hard tissues As with many imaging techniques in optical microscopy, the main function of a confocal imaging system is to improve image contrast: to delineate structures that would otherwise be difficult to see. Confocal microscopy offers several advantages

over

conventional

optical

microscopy

evaluate

the

ultramorphology of resin-dentine interfaces, including controllable depth of field, elimination of image degrading out-of-focus information, and the ability to collect serial optical sections from thick specimens. The key to the confocal approach

137

is the use of spacial filtering to eliminate out-of-focus light or flare that occurs in thick fluorescently labelled specimens. In a conventional widefield microscope, the entire specimen is flooded evenly in light from a light source. All parts of the specimen in the optical path are excited at the same time and the resulting fluorescence is detected by an image capture device or photographic film including a large unfocused background part. In other words, secondary fluorescence emitted by the specimen that appears away from the region of interest often interferes with the resolution of those features that are in focus. This is especially problematic for specimens having a thickness greater than circa 2 μm. In contrast, a confocal microscope eliminates out-of-focus signal as only light produced by fluorescence very close to the focal plane can be detected, and the image's optical resolution, particularly in the sample depth direction, results much better than that of wide-field microscopes (Watson, 1997). The word 'confocal' derives from the use of an aperture in the optically conjugate focal plane in front of the detector, in both the illuminating and imaging pathways of the microscope. The area surrounding the aperture rejects stray light returning from areas that are not in the focal plane of the lens. Furthermore, this microscope presents the optical sectioning characteristic, which enables the detection of subsurfaces components. The terminology refers to the noninvasive method by which the instrument collects images, using focused light rather than physical means to section the specimen (Watson, 1997). Consequently, high-resolution confocal microscopic images may derive from either the surface of a sample or beneath the surface. These images are thin (>0.35 μm) optical slices up to 200 μm in depth. With microscopes running under "normal" conditions, the optical section thickness can be >1 μm and the

138

effective penetration into dentine a maximum of 100 Îźm. At this distance the sharpness and contrast of the image may be poor, the best images being derived from structures just below the surface (<20 Îźm). Imaging samples from their surfaces without the need for thin section preparation has also significant advantages when it comes to biomaterials/tooth interfaces as relatively large intact tooth samples can be placed on the microscope stage. All that is required is to section the sample once and observe directly the subsurface structures, taking advantage of the optical sectioning feature. All microscopic techniques involve the interaction of the sample under examination with the interrogating system being used to probe it. The simplest technique for using confocal microscopy in dental materials research is to highlight the distribution of components within the bonded interface with fluorescent

labels.

Fluorescent

dyes,

also

called

fluorophores

or/and

fluorochromes, are useful as tracers to identify the path or the current location of a compound in clinical as well as laboratory investigations (Watson, 1997). The principle of fluorescence involves the absorption by the dye molecule of energy in the form of light at one wavelength range (excitation), which causes electrons to move into higher energy shells away from the nucleus. When these electrons return to more stable, lower-energy shells the energy is dissipated as both heat and light. Because the emitted photon has less energy than the excitation photon, the wavelength of the emission is longer by Stokes' Law, and the difference between the wavelengths of the excitation and emission maxima is known as the Stokes' Shift (Rost, 1995). Most of the dyes used in dental research are water-soluble and easily detectable, even in a low concentration. They are made of very small particles

139

(0.60-0.80 nm in diameter) that may erratically permeate throughout dentine and hybrid layer. Rhodamine B, a water-soluble molecule with a molar mass of 479 g mol-1, is one of the most frequently used fluorochromes. This compound is excited using green light (540 nm) and emits red in colour (590 nm). Rhodamine B is effective in very low concentration, fairly labile, moves freely across the bonded interface, and is easily detected microscopically with appropriate filters. The compound is also stable under various pH conditions. In the early 1970s, Rahn & Perren introduced xylenol orange (Rahn and Perren, 1971), calcein blue (Rahn and Perren, 1970) and alizarin complexone (Rahn and Perren, 1972). These chemicals fluoresce at different wavelengths and colours, and all possess adjacent donor sites for calcium chelation. This means that the label can indicate any area of mineralisation in the body, including a site of bone formation or dentine deposition (Rahn and Perren, 1971). Tetracyclines are also strong chelating agents, able to sequester a metallic ion such as calcium and firmly bind it into a ring (Lee et al., 2003). The use of fluorochromes in remineralisation studies is a widely accepted technique that dates back to the 1950s. Several pioneers, such as Harold Frost (Frost, 1963), have thoroughly investigated the potential of labelling with fluorescent calcium chelators for the study of bone formation and bone remodelling dynamics. Since the development of bone tissue engineering, a renewed interest in the benefits of fluorochrome use was perceived. Fluorochrome use in animal models makes it possible to determine the onset time and location of osteogenesis, which are the fundamental parameters in bone tissue engineering studies (Van Gaalen et al., 2010).

140

Since most of the fluorescent labels present one or another disadvantage, the possibility of selection from a larger number increases the probability of finding the one best suited for a special purpose. Important requirements are that (i) the label should be slow to bleach and stable in an aqueous environment, remaining deposited at sites of new mineralisation; (ii) the fluorescence should be clearly outlined and produce a strong signal; (iii) the fluorescent dye should be non-destructive to the substrate or material to which it comes into contact, namely, there should be little interference with the experiment from local effects on calcification; and (iv) the fluorescence should be resistant to chemicals and to the light used for its excitation. Easy photographic recording and reasonable cost of the dye are further considerations for selection.

2.2.4 Aims of the study The aims of this study were: To evaluate the chemical-physical properties, bioactivity and adhesive effectiveness of novel light-curable methacrylate-based dentine bonding agents either incorporating calcium/sodium phosphosilicate (BioglassÂŽ 45S5) or three distinct hydrated blends of experimental calcium-silicate cements. To implement a prospective adhesive procedure involving the preliminary application of a bioactive surface reactive glass-ceramic material (BioglassÂŽ 45S5). The following investigations were planned:

141

A measurement of the changes in dimension and weight resulting from water sorption in the experimental light-cured resin-based materials.

II.

A thermoanalytical study (differential scanning calorimetry) to assess the relationship between the degree of hydration and thermal properties.

III.

An attenuated total reflection Fourier transform infrared spectroscopy study of the bioactive micro-filled materials as a function of soaking time in phosphate-containing solutions.

IV.

A microtensile bond strength test and scanning electron microscopic study of resin-dentine specimens created with the experimental protocol and adhesives bonded to flat midcoronal dentine after immersion in phosphate buffered saline or simulated body fluid solutions for 24 h or 6 months.

A morphological and nano-leakage confocal laser scanning microscopic observation of the experimental resin-dentine interfaces after ageing in phosphate buffered saline or simulated body fluid solutions for 24 h or 6 months.

VI.

A micro-indentation hardness analysis (micro-hardness testing) of the experimental resin-dentine interfaces (hybrid layer and its surroundings) after ageing in phosphate buffered saline or simulated body fluid solutions for 24 h or 6 months.

142

Section II - Experimental projects

143

Chapter 3: Chemical-physical properties and apatite-forming ability of experimental dental resin cements containing bioactive fillers

144

3.1 Introduction The clinical success of composite restorations is compromised by the incomplete

infiltration

currently

used

resinous

materials

into

the

demineralised dentine that leaves unprotected collagen fibrils below and within the resultant hybrid layer (Spencer et al., 2010). In an attempt to improve the durability of resin-dentine interfaces, it was speculated that dentine bonding agents (DBAs) containing BioglassÂŽ 45S5 (BAG) as ion-releasing micro-filler could be considered a promising approach for a possible therapeutic/protective effect associated with the precipitation of mineral compounds (i.e apatite) within the collagen matrix (Profeta et al., 2012). Likewise, light-curable resins containing calcium-silicate Portland cements have been proposed to deliver calcium and hydroxyl ions (alkalinising activity) into the surrounding fluids and elicit a positive response at the interface from the biological environment (Tay and Pashley, 2009). It is common knowledge that release of ions from fillers in dentine adhesives and restoratives depends on the rate of water sorption and the segmental mobility of the polymer chains within the copolymerised, highly cross-linked resin matrix (Ikemura et al., 2003). Their ion-leaching potential thus relies on the hydrophilic nature of the resin phase to absorb water for the necessary ion movement (Cattani-Lorente et al., 1999). An important property in polymeric materials science is the glass transition temperature (Tg) of the cured matrix which indicates the degree of cross-linking, physical state and final mechanical properties of synthetic organic materials used as plastics and resins. It has been suggested that water absorbed into the adhesive polymers would result in a reduction of the Tg and a weakening of the

145

polymer network (Ito et al., 2005, Tay et al., 2002a, Dhanpal et al., 2009). The Tg reached after the incorporation of fillers into the adhesive would also depend on the amount of filler particles contained in its composition (Moraes and Grandini, 2011). However, because of the novelty of these adhesives, the influence of the amount of water sorption on their thermal properties is still unknown. As such, the relationship between Tg and degree of hydration needs to be determined. Recent studies selected vibrational spectroscopy to investigate and identify a large range of components associated with dental materials (Wang et al., 2012, Kim et al., 2010e). Attenuated Total Reflection Fourier Transform Infrared Spectroscopy (ATR-FTIR) is often used to evaluate the in vitro apatite-forming ability of bioactive materials as a function of soaking time in phosphatecontaining solutions (Dulbecco’s Phosphate Buffered Saline, DPBS) (Gandolfi et al., 2011a). Thus, the objective of this study was to investigate the performance of four new light-curable methacrylate-based DBAs either incorporating BAG or three different experimental calcium-silicate cements, with particular emphasis on the water sorption and solubility behaviour of the cross-linked networks under a simulated wet oral environment. For each material, the Tg was characterised by using differential scanning calorimetry (DSC) directly after curing and following 60 days of storage in deionised water. The in vitro apatite-forming ability was assessed by ATR-FTIR technique after soaking in DPBS for 60 days. The null hypotheses to be tested in this study were: (i) there is no difference between DBAs with respect to water sorption, solubility and water uptake. (ii) There is no effect of water uptake and micro-filler content on the thermal

146

properties (Tg) of the tested adhesives. (iii) Milled DBAs possess bioactivity with the ability to form hydroxycarbonate apatite (HCA) after immersion in DPBS.

3.2 Materials and methods

3.2.1 Experimental micro-fillers and resin blends formulation A comonomer blend was formulated from commercially available monomers by using 28.75 wt% of hydrophilic 2-hydroxyethyl methacrylate (HEMA: Aldrich Chemical, Gillingham, UK) and 40 wt% of cross-linking dimethacrylate 2, 2bis[4(2-hydroxy-3-methacryloyloxy-propyloxy)-phenyl] Esstech

Essington,

PA,

USA).

propane wt%

dimethacryloyloxyethyloxycarbonyl-1,4-benzenedicarboxylic

(Bis-GMA: of

acid

2,5(PMDM:

Esstech Essington) was included to obtain a dental bonding system with chemical affinity to Calcium (Ca2+) (Table 3.1). To make the resins light-curable, 0.25 wt% camphorquinone (CQ: Aldrich Chemical) and 1.0 wt% 2-ethyldimethyl-4-aminobenzoate (ETDA: Aldrich Chemical) were also added. A calcium sodium phosphosilicate (BAG: BioglassÂŽ 45S5, SYLC, OSspray Ltd, London, UK) and three modified calcium-silicate cements were used as calcium and hydroxyl ions releasing micro-fillers (20-30 Îźm-sized particles). The first calcium-silicate filler (HOPC) was created by mixing 82.5 wt% of a type I ordinary Portland (OPC: Italcementi Group, Cesena, Italy), mainly constituted by tri-calcium silicate (Alite: 3CaO x SiO2), di-calcium silicate (Belite: 2CaO x SiO2), tri-calcium aluminate (3CaO x Al2O3) and gypsum (CaSO4 x 2H2O), with 7.5 wt% of phyllosilicate consisting of sodium-calcium-aluminum-magnesium silicate

hydroxide

hydrate

[(Na,Ca)(Al,Mg)6(Si4010)3(OH)6-nH2O;

147

Acros

Organics, Fair Lawn, NJ, USA] in deionised water (Ratio 2:1). The last two fillers were formulated as following: i) HPCMM: 90 wt% OPC with the addition of 7.5 wt% phyllosilicate and 2.5 wt% hydrotalcite containing aluminummagnesium-carbonate hydroxide hydrate [(Mg6Al2(CO3)(OH)16·4(H2O); SigmaAldrich] in deionised water (Ratio 2:1); ii) HPCTO: 80 wt% OPC plus 7.5 wt% phyllosilicate, 2.5 wt% hydrotalcite and 10 wt% titanium oxide (TiO 2: SigmaAldrich, Gillingham, UK)] in deionised water (Ratio 2:1). The three experimental Portland-base silicates were allowed to set in incubator at 37°C for 24 h, ground in an agate jar and finally sieved to obtain < 30 μm-sized micro-particles. Hybrid photopolymerisable adhesive agents were prepared by mixing with a spatula for 30 s on a glass plate each individual filler (40 wt%) and the neat resin (60 wt%) (Hashimoto et al., 2010b) in order to form a homogeneous paste. A generic label for every experimental material has been proposed to refer to the main components (Res-Contr, Res-BAG, Res-HOPC, Res-HPCMM and ResHPCTO). This made the accounting easier but it is non-presumptive about molecular formula and structure.

148

Table 3.1 - Chemical structures of the constituent monomers and composition (wt%) of the experimental adhesives used in this study. Abbreviations. Bis-GMA: 2, 2-bis[4(2-hydroxy-3-methacryloyloxy-propyloxy)phenyl] propane; HEMA: hydrophilic 2-hydroxyethyl methacrylate; PMDM: 2,5dimethacryloyloxyethyloxycarbonyl-1,4-benzenedicarboxylic acid; CQ: camphoroquinone; EDAB: 2-ethyl-dimethyl-4-aminobenzoate; BAG: BioglassÂŽ 45S5; HOPC: set Portland cement and smectite; HPCMM: Portland cement, Smectite and Hydrotalcite; HPCTO: set Portland cement, Smectite, Hydrotalcite and Titanium Oxide.

149

3.2.2 Specimen preparation Each co-monomer mixture was directly dispensed drop-by-drop into a Teflon split ring mould (7 mm in diameter and 1 mm thick) in an ordinary laboratory environment at ~23Â°C taking care to make the adhesive bubble-free. The Teflon mold was then sandwiched between two glass slides covered with Mylar sheets to exclude atmospheric oxygen and the complete assembly was clamped. The resin was light-cured for 20 s on each side using a quartz-tungsten-halogen light-curing unit (Optilux VLC, Demetron Research Co., CT, USA). The lightcuring unit had an exit-window diameter of 5 mm and was operated at 600mWcm-2. The output intensity of the curing unit was verified before testing with a halogen radiometer (Optilux Radiometer Model 100P/N - 10503; Demetron Research Co.). Standardisation of the distance between the light source and the specimen was provided by the thickness of the glass slide (1 mm), and the glass slides also provided a smooth surface for the testing process. Selection of curing time was determined in a pilot experiment by measuring a baseline micro-hardness of the surface of the resin disks (unpublished data). With the adopted total curing time (40 s), resins exhibited a mean Knoop micro-hardness of 20 Âą 2 KHN that was sufficient to allow specimens to be removed from the Teflon mold without undergoing permanent deformation. Any specimen with visible voids was discarded. The excess material around the disks was removed using a scalpel blade and the margins were rounded and finished using 1000-grit silicon carbide grinding paper. The ultimate specimens (n = 10 per group) were flat and had very smooth surfaces. The thickness and diameter of the specimens were measured at four points using a digital caliper (Mitutoyo Corporation, Tokyo, Japan), rounded to the

150

nearest 0.01 mm, and these measurements were used to calculate the volume (V) of each disk (in mm3). The resin disks were then stored in a desiccator with silica gel at 37°C to ensure dryness for measurement of the initial mass. Each disk was weighed in an electronic analytical balance (Model AD6, PerkinElmer, Shelton, CT, USA) with a reproducibility of ± 0.1 mg until a constant mass (M0) was obtained (i.e. variation lower than 0.2 mg in 24 h), which was the baseline weight for the absorption-desorption cycle.

3.2.3 Water sorption and solubility evaluation The protocol for this study was determined according to the ISO Standard Specification No. 4049 (ISO Standard, 2000) except for the dimension of the specimen disks and the period of water storage that had been extended up to 60 days. The specimens were individually immersed in Paraffin-sealed glass vials containing 20 ml of deionised water (pH 7.2) at 37 ± 0.5°C. At noted intervals (3h, 5h, 10h, 24h, 2d, 3d, 6d, 9d, 12d, 15d, 18d, 21d), each specimen was taken out of the glass vial using tweezers, blotted on a Whatman's filter paper to remove excess fluid, weighed and restored in fresh deionised water. The uptake of water was recorded until there was no significant change in weight, i.e. maximum wet mass of the surface-blotted water-equilibrated specimens (Ms) was attained. The specimens then underwent desorption in a desiccator, as previously described, and weighed daily until a dried constant mass (MD) was obtained. The values for water sorption (WS) and solubility (SL), (μg/mm3), were calculated for each specimen using the following equations:

151

WS = (Ms − M0) / V SL = (M0 − MD) / V (Ito et al., 2005) Because the mass variation of resin disks is the net result of both an increase in mass owing to water penetration and the decrease in mass owing to elution of low-molecular-weight monomers, and of oligomers, the net water uptake was calculated as the sum of maximum water sorption and solubility (Unemori et al., 2003).

3.2.4 Differential scanning calorimetry (DSC) The Tg of the experimental bonding systems were characterised by the use of a Perkin-Elmer Jade DSC system (Perkin-Elmer Corp., Waltham, MA, USA). The system was calibrated with zinc/indium. Additional specimens (n = 4 per group) were produced under the same photoactivation conditions used in the water sorption kinetics experiments and divided into two groups. The Tg values of the first groups were measured immediately after curing, and the Tg values of the second groups were measured after 60 days. The second groups of specimens were stored in lightproof boxes after the curing procedure to prevent further exposure to light and in distilled water at 37 ± 0.5°C. Two crimpled aluminium pans with perforations (diameter = 4mm, 1.2 thick) were placed in the sample holder of the DSC furnace. Adhesive films (10-13 mg) were obtained from each specimen and these were sealed in one of two aluminum pans while the other pan was left empty as reference. Each experimental adhesive was tested five times at a rate of 10°C min -1 under

152

nitrogen gas (20 ml min-1) over a two step heating/cooling cycle temperature range of -20 to 200Â°C, to eliminate any residual water. The analysis of DSC curves was carried out for the second heating run data in order to exclude the effect of any possible impurity, which might influence the thermal properties of the materials during the first heating cycle. The glass transition temperature was determined by using the inflection midpoint of the initial S-shaped transition slope, and determined from the onset with the aid of Perkin-Elmer Spectrum computer software. For each thermograph the Tg was calculated three times and the average value was used as the result and for the purpose of comparison.

3.2.5 Statistics Kruskal-Wallis analysis of variance (ANOVA) was used to evaluate whether there were any differences between groups for maximum water uptake, solubility, net water uptake and the percentage change in Tg values. The percentage change in Tg values was used to compensate for some differences in initial Tg values. Where a statistically significant p value was observed (p < 0.05), post-ANOVA pair-wise comparisons were conducted using MannWhitney-U tests with p < 0.01 regarded as statistically significant to compensate for multiple comparisons.

3.2.6 ATR-FTIR spectroscopy Four disks (n = 4) were made per group according to the procedure previously described. In vitro bioactivity was determined by immersing the materials in sealed cylindrical polystyrene holder (3 cm high and 4 cm in diameter)

153

containing 5 mL of DPBS fluid. DPBS is a physiological-like buffered (pH 7.4) Ca- and Mg-free solution with the following composition (mM): K + (4.18), Na+ (152.9), Cl- (139.5), PO43- (9.56, sum of H2PO4- 1.5mM and HPO42- 8.06 mM). The solution was replaced every 72 h. Infrared spectra were obtained immediately after curing and after 60 days of DPBS ageing at 37°C (Gandolfi et al., 2010a), using a Spectrum One FTIR spectrophotometer (Perkin-Elmer Corp., Norwalk, Conn., USA) equipped with a diamond crystal attenuated total reﬂection (ATR) accessory. Prior to the spectrophotometric analysis, each sample was rinsed with distilled water for 30 s to stop the exchange reactions and then completely air dried. ATR/FTIR spectra were acquired from Bioglass® 45S5 powder as well as the three hydrated blends of calcium-silicate cements alone used for preparing the filled adhesive. The ATR area had a 2 mm diameter. The IR radiation penetration was about 2 μm. Spectra were collected in the range of 650-4000 cm-1 at 4 cm-1 resolution for a total of 88 scans for each spectrum, and processed by smoothing, baseline correction, and normalisation with Spectrum One Software Version 5.0.1 (Perkin-Elmer Corp., Norwalk, Conn., USA). To avoid complications deriving from potential lack of homogeneity of the samples, five spectra were recorded on each specimen. The reported IR spectra were the average of the spectra recorded on five different points.

3.3 Results

3.3.1 Water sorption and solubility evaluation All the specimens remained intact after the absorption and desorption cycles.

154

No visible signs of discoloration, crazes or cracks were observed in the resin disks. Table 3.2 shows a summary of mean (± standard deviation) data for mass gain (maximum water uptake) and mass loss (solubility). Since these processes occur simultaneously (Tay and Pashley, 2003b), they were added together to provide an estimate of the net water uptake. Kruskal-Wallis analysis of variance indicated that maximum and net water uptake showed potential differences between groups (p < 0.001). Post-ANOVA contrasts indicated that all groups were different from each other with the exception that the Res-HPCTO group was not different from the groups ResHOPC and Res-HPCMM. Overall, the neat resin (Res-Contr) exhibited values of maximum water uptake, solubility and net water uptake that were significantly lower when compared to its corresponding filled versions. With respect to water sorption, the Res-Contr group absorbed the least amount of water (Res-Contr < Res-HPCMM = ResHPCTO = Res-HOPC < Res-BAG), while Res-BAG showed the highest values and both differed statistically from all the other DBAs (p < 0.05). Res-HPCTO presented intermediate values and did not differ statistically from Res-HPCMM and Res-HOPC groups (p > 0.05). Following the same trend, the lowest value of net water uptake (89.5 μg/mm3) was observed for the comonomer blend with no filler (Res-Contr < Res-HPCMM = Res-HPCTO = Res-HOPC < Res-BAG), which was significantly different from the values obtained with the other DBAs (p < 0.05). No differences on net water uptake were detected when the value obtained for Res-HPCTO (212.5 μg/mm3) was compared with the results of Res-HPCMM (185.5 μg/mm3) and Res-HOPC

155

groups (218.9 μg/mm3) (p > 0.05), respectively. The loss of dry mass following water sorption was defined as solubilisation of resin. No statistically significant differences between all the DBSs were observed for this parameter (p > 0.05).

Table 3.2 - Summary of maximum water uptake, solubility and net water uptake data. Values are mean (± standard deviation) in relative terms (1 μg/mm3 = 0.001 mg/mm3 x 100 = 0.1 mg/100 mm3 = 0.1%) and in absolute terms (μg/mm3) to provide comparisons to literature values which include both expressions. For each parameter investigated, same superscript lower case letters (analysis in columns) indicate no statistically significant differences (p > 0.05). Negative solubility values indicate that the dried constant mass obtained after final desiccation (MD) was higher than the initial adhesive polymer mass before water immersion (M0), suggesting that the absorbed water may have not been completely eliminated.

156

3.3.2 Differential scanning calorimetry (DSC) DSC thermograms displayed two heating/cooling cycles for each specimen during the heating programme. The upper part of the DSC curve represented the heating cycles whereas, the lower part represented the cooling cycles. The thermal heat-capacity changes of the Tg obtained for the control and the four experimental DBAs during the second heating run are reported in Table 3.3. This shows a summary of the mean data for initial Tg, Tg after ageing and the percentage change in Tg. For the resin control, the initial glass transition temperature was 115.3°C whereas initial Tg values of Res-BAG, Res-HOPC, Res-HPCMM and ResHPCTO were 119.7°C, 117.2°C, 114.5°C and 114.4°C, respectively. After the ageing period, the Tg for the Res-Contr group increased to 121.3°C. The Tg also increased at the end of the period of water storage for all the other experimental DBAs, ranging between 131.2°C and 147.8°C. Kruskal-Wallis analysis of variance indicated that the percentage change in Tg showed potential differences between groups (p < 0.001). Post-ANOVA contrasts indicated that all groups were different from each other except that the lowest percentage change in Tg presented by Res-Contr group was not statistically different from the percentage change of the Res-HOPC group and that the percentage change in Tg for Res-BAG group was not statistically different from the Res-HOPC and Res-HPCMM groups, respectively. Lastly, the Res-HPCTO group exhibited the highest percentage change in Tg.

157

Table 3.3 - Means and standard deviations for Tg initially, after the ageing period and percentage change as determined by DSC analysis. Numbers in parentheses are standard deviations. Same letter indicates no differences in columns. * Five measurements

158

3.3.3 ATR-FTIR spectroscopy The results of ATR-FTIR analyses are shown in Figure 3.1. Band assignments have been given according to the literature (Rueggeberg et al., 1990, Wang et al., 2011). In the unmilled comonomer blend, the group assignments of the IR absorption bands confirmed prominent bands at 1716, 1637, 1454, 1162-1078 cm-1 attributed to the C=O, C=C, CH2, and C-O-C vibrations (HEMA), respectively. Absorption bands at 1260, 1385, 1520-1610, 3000 and 3440 cm-1 were attributed to C-O-C6H4, CH2, C6H4, C-H and OH vibrations (Bis-GMA). The IR spectra of the hydrated powders presented similarities for the calciumsilicate cements. Calcium carbonate bands at about 1457-1460 cm-1 (HOPC, HPCTO and HPCMM) and 1420 cm-1 (HOPC) were present. The band at 1110 cm-1 can be attributed to the SO42- group. The 870-850 cm-1

bands are

attributable to belite. The HPCTO and HPCMM powders were characterised by a higher relative intensity of the bands at about 1460 cm -1; in the group constituted by type I ordinary Portland cement, the band at 1460 cm-1 presented lower intensity than in the other cements (i.e. lower carbonate content). IR spectra of BioglassÂŽ 45S5 powder were also obtained to characterise its chemical structure. BioglassÂŽ 45S5 powder presented vibrational bands at 731, 920, 1030 and 1457 cm-1 corresponding to CO32-, Si-O, Si-O-Si and CO32stretches, respectively (Kim et al., 1989). Figure 3.1 also shows the FTIR-ATR spectra recorded on the outer surfaces of freshly prepared materials immediately after curing and of the samples soaked

159

in DPBS for 60 days, revealing the presence of their respective constitutive elements. With the exception of samples made of pure comonomer blend with 0 wt% micro-filler content (Res-Contr), used as references to evaluate the structural evolution and the bioactivities of the prepared biomaterials, FTIR analyses proved the presence of carbonated apatite on all the experimental adhesives after the ageing period. It was observed the beginning of apatite deposition (1030-1465 cm-1) for Res-HPCTO and Res-HPCMM groups, with the latter exhibiting a more crystalline apatite phase at 1465 cm -1. A more crystalline apatite phase was also observed on the adhesives incorporating type I ordinary Portland cement (Res-HOPC), as primarily revealed by the higher resolution of the phosphate bending bands at 960, 1025 and 1465 cm-1. Following the ageing period, the band present in the BioglassÂŽ 45S5 powder at 731 and 920 cm-1 disappeared on the Res-BAG adhesive. A small carbonate (HCA) band was present at 870 cm-1 (Lusvardi et al., 2009); this carbonate band is usually taken as an indication for carbonate being incorporated into apatite, resulting in HCA, rather than stoichiometric HAp (Lu and Leng, 2005). It is difficult to distinguish whether this band is split, in which case it would be Btype substitution (i.e. carbonate replacing a phosphate group). However, broad CO32- bands are present in the region starting from 1410-1440 cm-1 indicating B-type substitution (Brauer et al., 2010). Spectra of Res-BAG sample after the soaking period also showed a shoulder at 1080-1090 cm-1 due to the P-O stretch which is observed in B-type substituted HCA (LeGeros et al., 1969). In the same samples, the presence of

160

orthophosphate at 60 days was confirmed by a clearly pronounced orthophosphate band at 1027 cm-1, superimposed on the Si-O-Si stretch band at 1030 cm-1 which increased but did not markedly sharpen. The absorption band at 1716 cm-1, attributed to C=O stretching, can clearly be seen in the IR spectra of all the milled adhesives. This indicates that the C=O of the amide groups along the chains of HEMA was actively involved in the formation of the polymeric networks.

161

162

Figure 3.1 - ATR-FTIR spectra of the unmilled comonomer blend, of BioglassÂŽ 45S5, HOPC, HPCTO and HPCMM powders and of the hybrid experimental adhesives immediately after curing and following 60 days in DPBS.

163

3.4 Discussion The water absorption characteristics of any polymeric material, whether filled or not, is of great importance for dental applications. Water is known to swell the resin polymer network and hydrolytically cleave methacrylate ester bonds (Salz et al., 2005). This enables water percolation through the hybrid layer and may favour the rapid and catastrophic degradation of resin-dentine bonds (Pashley et al., 2004). However, some water absorption may not always be entirely disadvantageous (Wei et al., 2011) and this study explored the inclusion of four different ionreleasing fillers in an experimental dental adhesive as a way to benefit from the resin's hydrophilic behaviour and open up the potential to create favourably biointeractive dental adhesives. Hence a precise evaluation of the water absorption characteristics of the polymeric biomaterials was clearly important. By using the data obtained with the neat resin (Res-Contr) as a parameter for the relationship between water sorption and type of micro-filler, maximum water uptake and net water uptake were found to increase by adding BioglassÂŽ 45S5 or calcium-silicate cements to the polymer network, leading the anticipated first null hypothesis to be partially rejected and corroborating the results of previous studies (Yiu et al., 2004). After 60 days of storage in deionised water, all the experimental adhesives provided statistically significant increments relative to their initial glass transition temperature that compelled us to reject also the second null hypothesis. Conversely, the third null hypothesis was accepted, since ATR-FTIR analyses proved the presence of carbonated apatite on all the experimental adhesives after prolonged storage in DPBS.

164

In ISO 4049 (ISO standard, 2000), resin solubility is calculated as a loss of dry mass in specimens that have been immersed in water over time. Unreacted monomers or oligomers can leach out of the polymer during water sorption and subsequent polymer expansion. Generally, increases in water sorption are associated with increases in solubility. This is in contrast with the results of the present study in which no significant differences were observed for the latter between all the tested materials (Table 3.2). A possible explanation for this controversy might be that the dried constant mass obtained after final desiccation was higher than the initial adhesive polymer mass before water immersion, suggesting that the absorbed water may have not been completely eliminated and thus giving either poor (Res-BAG, Res-HOPC and Res-HPCTO) or negative solubility values (Res-Contr, Res-HPCMM). However, this does not mean that no solubility occurred but rather that water sorption was greater than solubility. Since calcium and phosphate ion release is slow at later times (Abou Neel et al., 2010), this could in part provide a further reason of the slight reduction in material mass and volume after 60 days. Water diffuses into polymers at different rates depending on the polarity of their molecular structure, the degree of cross-linking and the presence of residual monomers and/or other water-attracting species, e.g. glass surfaces (Fabre et al., 2007). According to these molecular and microstructural factors, the mechanism of water diffusion can be summarised in two main theories: (i) free volume theory, according to which water diffuses through nanometre spaces within the polymer and (ii) interaction theory, according to which water binds to specific ionic groups of the polymer chain. In this case, water diffusion occurs

165

according to water-affinity of these groups (Malacarne et al., 2006). Water absorbed into resin polymers has been demonstrated to decrease the glass transition temperature of dental adhesives. However, the damage is reversible at the outset with removal of the sorbed water (Dhanpal et al., 2009). After light-curing, the presence of Bis-GMA monomers in the experimental resin created a polymeric network able to stabilise the outer surface of the resin disks. Nonetheless, water-filled regions and/or hydrophilic polymer domains were also present in the cured resin (Tay et al., 2002b), and these channels were thought to be responsible for water movement within the adhesives. Due to the hydrophilic pendant groups of HEMA, once immersed in aqueous medium the designed resin matrix was permeable enough to absorb water that was reused for hydrating the ionic products (PO43- and Ca2+) released from the water-soluble micro-fillers. This may explain why the addition of the bioactive particles to the control resin showed no reduction of the Tg for the experimental adhesives. As already pointed out in a previous work (Min et al., 1993), Tg variation has also been attributed to various molecular parameters, such as molecular weight and stiffness of the cross-linked chains. Along with a sustained and effective assimilation of water, the reactive fillers introduced phosphate and silicate groups that had a hindering effect on chain mobility; furthermore, secondary interactions were generated by the possible formation of oxygen bridges between close proximity phosphate and silicate groups within the copolymeric network, requiring a greater amount of energy to free the chains and thus raising the thermal heat-capacity of the material (Kemal et al., 2011). Intraoral temperatures that exceed the Tg may result in softening of the material

166

and, consequently, in failure of the clinical procedure. It is equally important to observe that all the Tg values obtained are well above the range of oral environment temperature. The ability to release biologically active ions (biointeractivity) is a prerequisite for a material to be bioactive and trigger the formation of apatite. Hashimoto et al. demonstrated that crystal formation was a common behavior of adhesives prepared with 40 mass% bioglass and 60 mass% of light-cured resin containing Bis-GMA and HEMA during long-term water storage (Hashimoto et al., 2010b). The occurrence of similar chemical-physical events was recently reported in a light-curable MTA-based material containing an amphiphilic resin immersed in phosphate-containing solutions (Gandolfi et al., 2011a). In this study, ATR-FTIR spectroscopy technique yielded informations on the chemical composition change which occurred on the surface of all the resin disks during the ageing period. As showed in Figure 3.1, IR spectra of the methacrylate-based filled resin disks soaked in DPBS for 60 days confirmed the presence of carbonate ions in different chemical phases, mainly as apatite deposits at 960, 1025 and 1465 cm-1 (Res-HOPC); 1030 and 1465 cm-1 (ResHPCTO and Res-HPCMM groups); 870, 1027, 1080 and 1440 cm-1 (Res-BAG) It appears from the preceding discussion that the similarity in mode of biologic action of the micro-fillers employed in this study stems from one common characteristic they all possess: their propensity to release Ca 2+ and ability to form hydroxycarbonate apatite (HCA) on the polymerised methacrylate. It is important to consider that BioglassÂŽ 45S5 and the silicate phases of Portland cements when hydrated underwent a series of physicochemical reactions with subsequent precipitation of a polycondensated silica-rich layer

167

“Si-gel” (Res-BAG) (Andersson and Kangasniemi, 1991) or resulting in the formation of a nanoporous matrix/gel of calcium silicate hydrates “C-S-H phases” together with a soluble fraction of calcium hydroxide Ca(OH) 2 and portlandite (Camilleri, 2008) (Res-HOPC, Res-HPCMM and Res-HPCTO). Further incorporation of various mineral ions from the surrounding fluid helped the amorphous Si-gel and C-S-H phases to finally evolve into apatite crystals. This process is governed mainly by a tailored surface charge; it has been previously demonstrated that negatively charged polar groups must be present for a catalytic effect on apatite nucleation (Welch et al., 2010). Presumably, crystal formation also depends on the ion product being greater than the solubility product constants of the solid phase formed on the material surface (Filgueiras et al., 1993). Indeed, the Si-gel and C-S-H phases provided the negatively charged sites for the migration of Ca2+ ions, which in turn led to an over-saturated solution that exceeded the solubility product constants for a number of mineral forms, inducing crystal growth (Kokubo et al., 2003, Kim et al., 2005). There is morphological evidence that the extent of permeability through the hybrid layer (nano-leakage) increases after 1 year of water storage (Tay et al., 2003, Hashimoto et al., 2002), in consequence of the adsorption of water, hydrolysis of the ester bonds and component release processes in methacrylate materials (Spencer et al., 2010). As the crystallisation process can bind water and encourage reprecipitation of less soluble species in material regions from which components have been released, it can also be anticipated that this might limit nano-leakage and reduction in bond strength during long-term function.

168

3.5 Conclusion Within the limitations of this study, it can be concluded that experimental methacrylate-based DBAs either incorporating BAG or calcium-silicate cements are not inert materials in a simulated oral environment, and the precipitation of apatite deposits in the surrounding fluid or contiguous dental tissues may occur in the intra-oral conditions. A bioactive dental material which forms HCA on the surface would have several benefits including closure of gaps forming at the resin-dentine interface and potentially better bond strength over time (less degradation of bond). Future experiments will explore the functional consequences of the water sorption values observed in this study on microtensile bond strength when the same experimental adhesives are applied to acid-etched water-saturated dentine. Microscopy ultra-morphological analysis needs to be also performed.

169

Chapter 4: Bioactive effects of a calcium/sodium phosphosilicate on the resin-dentine interface: a microtensile bond strength, scanning electron microscopy, and confocal microscopy study

170

4.1 Introduction Bioactive materials are often used in operative dentistry due to their ability to interact actively with dental hard tissues, inducing calcium-phosphates (Ca/P) deposition in the presence of body fluids or saliva (Kim et al., 2010e, Ryou et al., 2011, Hench and Andersson, 1993, Sauro et al., 2011a). Whereas remineralisation of enamel lesions can be achieved predictably (Shen et al., 2001, Lagerweij and ten Cate, 2002), there is little information on whether it is possible to remineralise specific mineral-deficient areas within the resindentine interface (i.e. hybrid layers) (Ryou et al., 2011). Some glass ionomer based, fluoride-releasing adhesive resins may induce crystal growth within gaps in the bonded interface after long-term storage in water (Hashimoto et al., 2008). Furthermore, bioactive, ion-releasing materials, such as calciumphosphate (Ca/P) cements, have the potential to encourage dentine remineralisation by mineral precipitations (Ngo et al., 2006, Dickens et al., 2003, Dickens and Flaim, 2008, Arends et al., 1997). Peters et al. (Peters et al., 2010) showed the presence of a higher mineral content [determined by electron probe elemental micro-analysis (EPMA) techniques] and an increase in micro-hardness along the interface of resinbonded caries-affected dentine, following the application of materials containing Ca/P cements. Bioactive calcium/sodium (Ca/Na) phosphosilicates, such as BioglassÂŽ 45S5 (BAG), are able to induce deposition of hydroxycarbonate apatite (Sauro et al., 2011a, Vollenweider et al., 2007, Efflandt et al., 2002, Hench and Paschall, 1973). Although bioactive glasses have previously been used for dentine remineralisation by direct application onto demineralised dentinal tissue when dispersed in water solutions (Sauro et al., 2011a, Efflandt

171

et al., 2002), there is little information about the potential therapeutic effects of BAG on the resin-dentine interface when used during etch-and-rinse bonding procedures. Therefore, this study was devised to assess the bioactive effects of BAG during etch-and-rinse dentine-bonding procedures on the resin-dentine interface. This aim was accomplished by evaluating the microtensile bond strength (μTBS) of specimens after 24 h and 6 months of storage in PBS. Fractographic analysis was also performed through scanning electron microscopy (SEM). The ultramorphology and nano-leakage analysis of the resin-bonded dentine was executed using confocal laser scanning microscopy (CLSM). The null hypotheses to be tested in this study were: (i) the use of BAG employed during bonding procedures has no effect on the bond strength, and (ii) the presence of BAG does not reduce nano-leakage within the demineralised ‘poorly-infiltrated’ areas within the resin-dentine interface.

4.2 Materials and methods

4.2.1 Specimen preparation Caries-free human third molars, extracted for surgical reasons from 20 to 40 yr. old patients under a protocol approved by the institutional review board of Guy's Hospital (South East London ethical approval 10/H0804/056), were used in this study. The treatment plan of any of the involved patients, who had given informed consent for use of their extracted teeth for research purposes, was not altered by this study.

172

The study was conducted in accordance with the ethical guidelines of the Research Ethics Committee (REC) for medical investigations. The teeth were stored in deionised water (pH 7.1) at 4Â°C and used within 1 month after extraction. The coronal dentine specimens were prepared by sectioning the roots 1 mm beneath the cemento-enamel junction (CEJ) with a hard tissue microtome (Accutom-50; Struers, Copenhagen, Denmark) using a slow-speed, water-cooled diamond wafering saw (330-CA RS-70300; Struers). A 180-grit silicon carbide (SiC) abrasive paper mounted on a water-cooled rotating polishing machine (Buehler Meta-Serv 3000 Grinder-Polisher; Buehler, DĂźsseldorf, Germany) was used (30 s) to remove the diamond saw smear layer and to replace it with a standard and more clinically relevant smear layer (Oliveira et al., 2003).

4.2.2 Experimental bonding procedures and formulation of resin adhesives A resin co-monomer blend was formulated by using a hydrophobic, crosslinking dimethacrylate monomer - bisphenyl-A-glycidyl methacrylate (Bis-GMA; Esstech, Essington, PA, USA) - and a hydrophilic monomer - 2-hydroxyethyl methacrylate (HEMA; Sigma-Aldrich, Gillingham, UK). In order to obtain a dental bonding system with chemical affinity to calcium (present in dentine and BAG), an acidic functional monomer - 2,5-dimethacryloyloxyethyloxycarbonyl1,4-benzenedicarboxylic acid (PMDM; Esstech Essington) - was also included within the composition of the resin blend. Subsequently, the resin blend was made

light-curable

binary

photoinitiator

system

based

camphoroquinone (CQ; Sigma-Aldrich) and 1,2-ethyl-dimethyl-4-aminobenzoate

173

(EDAB; Sigma-Aldrich). This resin co-monomer blend was used to formulate the experimental primer and the bonds used in this study (Table 4.1). A BAG (Sylc; OSspray, London, UK) with particle size < 10 Îźm was employed in the etch-and-rinse bonding procedures using

two different experimental

approaches: (i) BAG-AD (30 wt% BAG included within the composition of a resin adhesive as a bioactive microfiller), and (ii) BAG-PR (BAG applied directly onto H3PO4-etched/wetted dentine before bonding procedures). The neat adhesive, with no BAG, served as the control (RES-Ctr) (Fig. 4.1). In detail, a water wet-bonding dentine substrate was achieved by water-rinsing, for 15 s, the dentine surfaces acid-etched with 37% phosphoric acid solution (H3PO4) (Sigma-Aldrich) and gently blowing off (for 2 s) excess water to leave a wet reflective-surface. The control bonding procedure (RES-Ctr) was accomplished by applying two consecutive coats of an ethanol-solvated resin primer [50 wt% absolute ethanol (Sigma-Aldrich) and 50 wt% of neat co-monomer resin blend] and a layer of the neat co-monomer resin blend (Table 4.1) within a period of 20 s. Light-curing was immediately performed for 30 s (> 600 mW/cm-2, Optilux VLC; Demetron, Danbury, CT, USA). The first experimental bonding procedure (BAG-AD) was performed by applying the same ethanol-solvated resin primer onto H3PO4-etched dentine, as previously described, followed by a layer of bonding resin containing BAG (Table 4.1; Fig. 4.1). The bonding and the light-curing procedures were executed as previously described for the RES-Ctr group. The second experimental bonding procedure (BAG-PR) was performed as follows. The 37% H3PO4 solution (Sigma-Aldrich) was applied onto the dentine

174

surface for 15 s. Then, 0.05 g of BAG powder was placed onto the H 3PO4etched wet dentine surface, spread immediately for 10 s using a cotton pellet, and finally rinsed with copious amounts of deionised water for 15 s (Fig. 4.1). The primer/bond application and the light-curing procedures were performed as previously described for the RES-Ctr group. A final composite build-up (5 mm) was constructed on each specimen using a light-cured resin composite (Filtek Z250; 3M-ESPE, St Paul, MN, USA) in five incremental layers (of 1 mm thickness). Each layer of composite was individually light cured for 20 s. The resin-bonded dentine specimens were stored in PBS for 24 h or 6 months at 37Â°C. The PBS was composed of (in g/l) CaCl2 (0.103), MgCl2.6H2O (0.019), KH2PO4 (0.544), KCl (17), and HEPES (acid) buffer (4.77), and the pH was 7.4.

175

Figure 4.1 - Schematic illustrating the experimental study design. Human third molars were used to prepare standardised dentine surfaces. The three different bonding approaches were performed using specific components and application procedures. Abbreviations. Bis-GMA: bisphenyl-A-glycidyl methacrylate; HEMA: 2-hydroxyethyl methacrylate; PMDM: 2,5-dimethacryloyloxyethyloxycarbonyl-1,4-benzenedicarboxylic acid.

176

Table 4.1 - Composition of the experimental bonding procedures/adhesive systems used in this study. Abbreviations. Bis-GMA: bisphenyl A glycidyl methacrylate; HEMA: hydrophilic 2-hydroxyethyl methacrylate; PMDM: 2,5-dimethacryloyloxyethyloxycarbonyl-1,4-benzenedicarboxylic acid. At the end of the formulation of the resins, 0.25 wt% camphoroquinone (CQ) and 1.0 wt% 2-ethyl-dimethyl-4-aminobenzoate (EDAB) were added to the resin mixture.

177

4.2.3 μTBS and SEM fractography and failure analysis Twenty dentine-bonded specimens from each group were sectioned using a slow-speed water-cooled diamond wafering blade (Struers) mounted on a hardtissue microtome (Isomet 11/1180; Buehler) in both x and y directions across the adhesive interface to obtain matchsticks with cross-sectional areas of 0.9 mm2. By excluding peripheral beams showing the presence of residual enamel, only the remaining matchsticks (n = 10-15) were selected to create three groups with the same total number of resin-dentine specimens in each group (n = 280). The exact width of each matchstick was checked using a calliper (Mitutoyo CD15; Mitutoyo, Kawasaki, Japan) and half of them (n = 140) were tested after 24 h of storage in PBS and the remaining half (n = 140) were tested after 6 months of storage in PBS at 37°C. The μTBS test was performed using a microtensile jig in a LAL300 linear actuator (SMAC Europe; Horsham, UK) with a LAC-1 high-speed controller single axis with a built-in amplifier and at the following settings: stroke length = 50 mm, peak force = 250 N, displacement resolution = 0.5 mm, and crosshead speed = 1 mm min -1. Bond-strength data were calculated and expressed in MPa, the μTBS values of sticks from the same restored teeth were averaged, and the mean bond strength was used as one statistical unit for the statistical analysis. The μTBS (mean-MPa) data for each group were analysed using a repeated-measures ANOVA and Tukey’s post-hoc test for pairwise comparisons (α = 0.05). The mode of failure was classified as percentage of adhesive, mixed, or cohesive. The failed bonds were examined at x30 magnification using a stereoscopic

microscope

(Leica

M205A;

Germany).

178

Leica

Microsystems, Wetzlar,

Five representative debonded specimens for each group that failed in mixed or adhesive modes were selected for ultramorphology analysis of the fractured surface (SEM Fractography). They were dried overnight and mounted on aluminium stubs with carbon cement, then sputter-coated with gold (SCD 004 Sputter Coater; Bal-Tec, Vaduz, Liechtenstein) and examined using SEM (S3500; Hitachi, Wokingham, UK) with an accelerating voltage of 15 kV and a working distance of 25 mm at increasing magnifications.

4.2.4 Confocal microscopy ultramorphology and nano-leakage evaluation A further three dentine specimens from each group were bonded, as previously described, with the primer/bond resins doped with 0.1 wt% rhodamine-B (Rh-B: Sigma-Aldrich, St Louis, MO, USA) and employed for the confocal microscopy analysis (Sauro et al., 2012a, Sauro et al., 2012b). The specimens were serially sectioned across the adhesive interface to obtain resin-dentine slabs (of 1 mm thickness). The resin-dentine slabs (n = 10 per group) were then divided into two subgroups based on the period of storage in PBS (24 h or 6 months) (Fig. 4.2). Subsequent to the storage period, the specimens were coated with two layers of fast-setting nail varnish applied 1 mm from the resin-dentine interfaces and immersed in 1 wt% aqueous fluorescein (Sigma-Aldrich) solution for 24 h. The specimens were then treated in an ultrasonic water bath for 2 min and polished using SiC abrasive papers of ascending grit (#1200 to #4000) (Versocit; Struers) on a water-cooled rotating polishing machine (Buehler MetaServ 3000 Grinder-Polisher; Buehler). A final treatment in an ultrasonic water bath (5 min) completed the specimen preparation for the confocal microscopy evaluation (Fig. 4.2).

179

The microscopy examination was performed using a confocal laser scanning microscope (Leica SP2 CLSM; Leica, Heidelberg, Germany) equipped with a 63 X /1.4 NA oil-immersion lens and using 488-nm argon/helium (fluorescein excitation) or 568-nm krypton (rhodamine excitation) laser illumination. The reflection imaging was performed using the argon/helium laser. Confocal laser scanning microscopy reflection and fluorescence images were obtained with a 1-Îźm z-step to section optically the specimens to a depth up to 20 Îźm below the surface (Sauro et al., 2012a). The z-axis scans of the interface surface were arbitrarily pseudo-coloured by two selected operators and compiled into single projections using the Leica image-processing software (Leica). The configuration of the system was standardised and used at the same settings for the entire investigation. Each resin-dentine interface was completely investigated and then five optical images were randomly captured. Micrographs representing the most common features of nano-leakage observed along the bonded interfaces were captured and recorded (Sauro et al., 2012b).

180

Figure 4.2 - Schematic illustrating the composite-tooth matchsticks (1 mm) prepared using a water-cooled diamond saw, stored in PBS for 24 h or 6 months, and then subjected to microtensile bond strength (ÎźTBS) testing and scanning electron microscopy failure analysis. This schematic also illustrates how composite-tooth slabs were prepared, stored in PBS for 24 h or 6 months, and evaluated by confocal laser scanning microscopy.

181

4.3 Results

4.3.1 μTBS and SEM fractography and failure analysis The BAG-bonding technique versus storage time was statistically significant only for the BAG-AD group (P = 0.001); no significant reduction of the μTBS values was observed after 6 months of storage in PBS (P > 0.05). On the other hand, significant μTBS reductions were observed in both the BAG-PR and RES-Ctr groups (P < 0.05) after prolonged storage in PBS (6 months). The μTBS results (expressed as Mean and SD) are presented in Table 4.2. High μTBS values were achieved in all groups after 24 h of storage in PBS, with failures occurring mainly in cohesive mode in all groups; in contrast, important changes in the μTBS were observed after 6 months of storage in PBS. For instance, the μTBS of specimens in the RES-Ctr group (no BAG) showed a significant (P < 0.05) decrease after 6 months of storage in PBS and failed mostly in adhesive mode (66%). The specimens stored for 24 h in PBS that fractured in mixed mode were characterised by the presence of exposed dentinal tubules with spare extruded resin tags (Fig. 4.3A2). Conversely, the surface of the specimens that failed in adhesive mode after 6 months of storage in PBS presented several ‘funnelled’ dentinal tubules with no exposed collagen fibrils (Fig. 4.3A3). The resin-dentine specimens of the BAG-AD group maintained a high μTBS (P > 0.05) after 6 months of storage in PBS (23.89 ± 7.74 MPa). In these specimens the failure was prevalent in cohesive (43%) and mixed (40%) modes (Fig. 4.3B1) and the SEM analysis of the fractured surface revealed a dentine surface predominantly covered by residual resin and mineral crystals embedded within a resin/collagen network (Fig. 4.3B3).

182

The specimens of the BAG-PR group, where the BAG powder was applied onto acid-etched/wetted dentine before application of the adhesive system, showed a significant decrease in ÎźTBS values (P < 0.05) after prolonged storage in PBS (Table 4.2). These specimens failed mainly in adhesive mode (56%) after 6 months of storage in PBS, and the SEM fractographic analysis showed that the fracture during ÎźTBS testing occurred along the intertubular dentine, leaving an intact peritubular dentine and a consistent precipitation of mineral inside the dentinal tubules (Fig. 4.3C3).

183

Table 4.2 - Means and standard deviations (SD) of the microtensile bond strength values (MPa) obtained for the different experimental groups and percentage distribution of failure modes after microtensile bond strength testing; total number of beams (tested stick/pre-load failure). For each horizontal row: values with identical numbers indicate no significant difference. For each vertical column: values with identical letters indicate no significant difference using Student-Newman-Keuls test (P > 0.05).

184

Figure 4.3 - Scanning electron microscopy images of failure modes of the resinbonded specimens created using the three different bonding approaches tested. (A) Micrograph of the failure mode (cohesive) of the resin control bonded to etched dentine (37% H3PO4) after 24 h of storage in PBS (A1). At higher magnification (A2) it was possible to observe the presence of some exposed dentinal tubules (t), but most remained obliterated by resin tags (rt). No exposed collagen fibrils were visible on the dentine surface, and a well resin-hybridised hybrid layer was present (pointer). At 6 months (A3), the resin-dentine interfaces created with the control resin (RES-Ctr; containing no bioactive filler) showed only a few resin tags inside the dentinal tubules and no collagen fibrils were visible on a dentine surface characterised by funnelled dentinal tubules (pointer). (B) Micrograph of the failure mode (mixed) of the calcium/sodium phosphosilicatecontaining adhesive (BAG-AD) bonded to dentine, after 6 months of storage in PBS (B1). At higher magnification (B2) no exposed dentinal tubules or exposed collagen fibrils were observed; the dentine surface was well resin-hybridised (pointer). After 6 months of storage in PBS (B3), the debonded resin-dentine interface showed the presence of resin tags remaining inside the dentinal tubules (rt) and mineral crystals embedded within a preserved collagen network (pointer). (C) Micrograph of the failure mode (adhesive) of BAG applied directly onto H3PO4etched/wetted dentine before bonding (BAG-PR) after 24 h of storage in PBS (C1). At higher magnification (C2) it was possible to observe the presence of some exposed dentinal tubules (t), while most remained obliterated by resin tags (rt) containing few BAG particles. No exposed collagen fibrils were present on the dentine surface (pointer). At 6 months testing (C3), the resin-dentine interface created with the BAGPR showed a dentine surface characterised by the presence of remineralised dentinal tubules (t) obliterated by mineral crystals (cr). It is interesting to note how the fracture occurred along the intertubular dentine leaving an intact peritubular dentine around the mineral-obliterated dentinal tubule (pointer).

185

4.3.2 Confocal microscopy ultramorphology and nano-leakage evaluation The CLSM investigation showed that all the bonding procedures used in this study were able to create a resin diffusion within the demineralised dentine (hybrid layer 7-9 Îźm) and several resin tags into the dentinal tubules (Fig. 4.4A). Nevertheless, the resin-dentine interfaces of the specimens created in the three groups showed evident fluorescein penetration (nano-leakage) within the hybrid layer and along the dentinal tubules after 24 h of storage in PBS (Fig. 4.4B,C). The experimental bonding approach used to bond the specimens of the BAGPR group created resin-dentine interfaces characterised by the presence of mineral deposits inside the dentinal tubules and within the hybrid layer (Fig. 4.4D). The prolonged storage in PBS induced important changes in terms of ultramorphology and nano-leakage. For instance, the resin-dentine interface of the RES-Ctr group specimens was affected by severe nano-leakage within the hybrid layer and the presence of a continuous gap between dentine and composite (Fig. 4.5A). Conversely, the specimens of the BAG-AD group showed the presence of a strong reflective mineral material and partial dye penetration within the hybrid layer (Fig. 4.5B). The resin-dentine interface of specimens in the BAG-PR group was affected by partial dye penetration within a crystallised hybrid layer. However, gaps were also observed between the hybrid and adhesive layers (Fig. 4.5C), probably caused by the sample preparation procedure before the CLSM analysis.

186

Figure 4.4 - Confocal laser scanning microscopy (CLSM) images showing the interfacial characterisation and nanoleakage, after 24 h of storage in PBS, of the resindentine interfaces created using the three different bonding approaches tested. (A) Confocal laser scanning microscopy three-dimensional (3D) single-projection (fluorescence mode) image exemplifying the interfacial characteristics of the resindentine interface created using the control adhesive system (RES-Ctr) applied onto H3PO4-etched dentine. It is possible to observe a clear hybrid layer (hl) (approximate thickness 9 Îźm) (pointer) located underneath a thick adhesive layer (a) and long resin tags (rt). (B) This CLSM 3D single-projection (fluorescence/reflection mode) image of the resindentine interface created using the bioactive calcium/sodium phosphosilicatecontaining adhesive (BAG-AD) shows an intense nanoleakage signal from the hybrid layer (pointer) located underneath a thick adhesive layer (a) characterised by the presence of BAG microfiller (fl). The presence of long resin tags (rt) is also evident. (C) The resin-dentine interface created using the bonding procedure where the BAG is applied directly onto H3PO4-etched/wetted dentine (BAG-PR) shows evident dye penetration within the hybrid layer (pointer). Short resin tags (rt) are visible underneath a thick adhesive layer. The reason why only short resin tags could be created during this type of bonding procedure is shown in (D) where it is possible to observe a strong reflective signal from the demineralised dentine layer (pointer) and inside the dentinal tubules (t), indicating the presence of mineral particles.

187

Figure 4.5 - Confocal laser scanning microscopy (CLSM) images showing the interfacial characterisation and nanoleakage, after 6 months of storage in PBS, of the resin-dentine interfaces. (A) Confocal laser scanning microscopy three-dimensional single-projection (fluorescence/reflection mode) image of the resin-dentine interface created using the control adhesive system (RES-Ctr) applied onto H3PO4-etched dentine. It is possible to note the presence of evident dye diffusion (nanoleakage) within the hybrid layer and inside the dentinal tubules (pointer). A gap (g) is present between the dentine and the composite (c). (B) The resin-dentine interface created using the bonding approach where the bioactive calcium/sodium phosphosilicate-containing adhesive (BAG-AD) is applied onto H3PO4-etched dentine shows partial dye diffusion within a hybrid layer characterised by a strong reflective signal (pointer). (C) The resin-dentine interface created using the bonding procedure where the BAG is applied directly onto H3PO4-etched/wet dentine (BAG-PR) shows a crystallised reflective layer (pointer) characterised by low dye penetration (nanoleakage). A pronounced gap (g) can be seen between the adhesive layer (a) and the composite (c). It is also possible to observe the remaining reflective mineral materials on the fractured edge of the adhesive layer (arrows).

188

4.4 Discussion Hybrid layers created using etch-and-rinse adhesives include water-rich, resinsparse regions that account for 2-3% of their entire volume, which increase subsequent to prolonged ageing in fluids (Reis et al., 2007). The water-rich, resin-sparse regions represent essentially the nanoporosities within the demineralised collagen fibrils, created during adhesive application as a result of incomplete replacement of water by resin infiltration (Pashley et al., 2011). This undisplaced water may act as a functional medium for the hydrolysis of suboptimally polymerised resin matrices by esterases and denaturation of collagen via the activation of host-derived matrix metalloproteinases (MMPs), jeopardising the durability of the resin-dentine interfaces (Pashley et al., 2011, Pashley et al., 2004, Breschi et al., 2010b). Several methods have been advocated to increase the longevity of these resindentine interfaces, including the inhibition of the MMPs within the hybrid layer (Pashley et al., 2004, Breschi et al., 2010b) and enhancement of the resin infiltration within the demineralised collagen fibril using more hydrophobic resin monomers and ethanol wet-bonding (Pashley et al., 2011). Based on the results obtained in this study, the first null hypothesis must be partially rejected because the use of BAG produced bioactive/protective effects on the bond strength only when used as resin microfiller within the adhesive composition. The second null hypothesis must be totally rejected as both the experimental bonding approaches based on the use of BAG were able to reduce the nano-leakage within the demineralised ‘poorly infiltrated’ areas within the resin-dentine interface.

189

In detail, the control bonding procedure (RES-Ctr) and the two experimental bonding approaches used (BAG-AD and BAG-PR) to bond the acid-etched dentine produced comparably high μTBS values after 24 h of storage in PBS (Table 4.2). Conversely, a significant decrease in μTBS (P > 0.05) occurred in all groups after storage in PBS for 6 months, except for the specimens bonded using the resin adhesive containing BAG microfiller (BAG-AD). The SEM analysis of the fractured specimens of the RES-Ctr group showed, after 24 h of storage in PBS, a dentine surface characterised by a hybrid layer that was highly hybridised with resin and no presence of demineralised collagen fibrils exposed (Fig. 4.3A2). Conversely, these resin-dentine specimens stored for 6 months in PBS had a fractured surface characterised by ‘funnelled’ dentinal tubules, indicating degradation of the demineralised peritubular dentine (Fig. 4.3A3). In contrast, the bonded-dentine specimens of the BAG-AD group immersed in PBS for 6 months had a fractured (adhesive mode) dentine surface, with mineral crystals embedded within a preserved collagen network and no evidence of ‘funnelled’ dentinal tubules (Fig. 4.3B3). The SEM ultramorphology analysis of the fractured specimens (adhesive mode) of the RES-PR group stored for 24 h in PBS demonstrated the presence of dentinal tubules obliterated by resin tags and no exposed collagen fibrils (Fig. 4.3C2). Interestingly, when this type of dentine-bonded specimen was immersed in PBS for 6 months it was possible to detect a fractured dentine surface characterised by dentinal tubules obliterated by mineral crystals and a distinctive fracture along the intertubular dentine, which left an intact peritubular dentine (Fig. 4.3C3).

190

Possible explanations for such longevity attained in dentine-bonded specimens created using BAG-AD after prolonged storage in PBS may be as follows: I.

The presence of BAG within the resin-dentine interface may have induced the release of a silicic acid, such as Si(OH)4, and a subsequent polycondensation reaction between the silanols compounds and the demineralised collagen via electrostatic, ionic, and/or hydrogen bonding (Vollenweider et al., 2007, Välimäki et al., 2005, Zhong et al., 1994), which interfered with the ability of MMPs - although BAG is not a direct MMP inhibitor - to execute their collagenolytic and gelatinolytic activities. A study by Osorio et al. (Osorio et al., 2011) showed that it is possible to reduce the collagen-degradation process by using specific chemical compounds, such as zinc oxide, which interfere with the zinc-binding and calcium-binding catalytic domains of MMPs.

II.

The precipitation of an amorphous calcium phosphate (ACP) on the polycondensate SiO2-rich template of nucleation (Hench and Andersson, 1993, Shen et al., 2001, Vollenweider et al., 2007, Sauro et al., 2012a) induced by the dissolution and immediate reaction between Ca 2+ and PO43- species from BAG may have also favoured the formation of a highmolecular-weight complex (Ca/P-MMPs), which restricted the activities of MMP-2 and MMP-9 within the hybrid layer (Kremer et al., 1998). However, the ability of specific bioactive glass, such as Bioglass® 45S5, to modulate and/or reduce the presence of collagens I, II, and III, osteocalcin, osteonectin, and osteopontin, has also been demonstrated in bone-regeneration studies (Välimäki et al., 2005).

191

III.

The release of Na+ and Ca2+ ions from BAG, and the incorporation of H3O+ protons into the glass particles, may have created an optimal alkaline environment (Shen et al., 2001, Sauro et al., 2012a) within the resin-dentine interface that interfered with the activity of MMPs, which are very acidic-pH dependent (Pashley et al., 2004, Breschi et al., 2010b).

IV.

The bioactive remineralisation induced by BAG may have decreased the distribution of the water-rich, resin-sparse regions within the hybrid layer (Ryou et al., 2011, Sauro et al., 2012a) via silanols polycondensation and subsequent ACP/AH remineralisation, which probably interfered with the water-dependent hygroscopic and hydrolytic degradation of the polymer network (Ferracane, 2006).

The confocal microscopy evaluation performed after 6 months of storage in PBS indicated that both the experimental bonding approaches used in this study (BAG-AD and BAG-PR) created a resin-dentine interface affected only by partial dye penetration (nano-leakage) within a hybrid layer characterised by the deposition of a strong reflective mineral (Fig. 4.5B,C). Whereas it is reasonable to believe that the hybrid layer of the specimens created using the BAG-AD approach remineralised as a result of the bioactive/biomimetic activity of BioglassÂŽ 45S5 after prolonged (6 months) storage in PBS (Bakry et al., 2011a, Sauro et al., 2012a, Zhong et al., 1994), a completely different bioactive phenomenon may have occurred within the resindentine interface created by directly applying the BAG on the demineralised

192

H3PO4-wetted dentine, as a significant decrease of μTBS was attained after prolonged storage in PBS (Table 4.2). In this case, a possible explanation for the reduced confocal nano-leakage may be due to the chemical nature of mineral precipitation that occurred within the resin-dentine interface created as a result of the experimental bonding procedures (BAG-PR). Our hypothesis is that the chemical reaction between BAG and H3PO4 solution (Fig. 4.4D) may have induced the precipitation of dicalcium-phosphate salts (i.e. brushite and monetite). Bakry et al. (Bakry et al., 2011a, Bakry et al., 2011b) showed that the acid-base chemical reaction between BAG and H3PO4 may induce the formation of brushite via combination of the phosphate (released from the BAG and H3PO4) and calcium ions (released from BAG and etched dentine). The precipitation reaction of the brushite may be responsible for the creation of an acidic environment (Mandel and TAS, 2010), which may have evoked the activation of MMPs (Pashley et al., 2004, Breschi et al., 2010b); this situation is also aggravated by the fact that BAG no longer has the ability to create a localised ‘protective’ alkaline pH within the resin-dentine interface. Moreover, it is also possible that the BAG/ H3PO4 reaction may have altered the chemical and/or physical characteristics of BAG, in particular those responsible for the polycondensation of silanols and ACP/HA precipitation (Vollenweider et al., 2007, Välimäki et al., 2005, Zhong et al., 1994), which may be fundamental in altering the activity of MMPs (Kremer et al., 1998), as previously described. However, even supposing that the reaction between hydroxyl ions and Si(OH) 4 formed the silanols compounds and induced the polycondensation reaction,

193

they may have been washed out by application of the air-water jet before application of the primer and bond (Bunker et al., 1988). Furthermore, a slightly acidic environment may have remained in loco within the resin-dentine interface during the prolonged storage in PBS as a result of the release of H+ from the acidic monomer (2,5-dimethacryloyloxyethyloxycarbonyl1,4-benzenedicarboxylic acid) contained within the resin adhesives (Wang and Spencer, 2005, Sauro et al., 2011b, Bayle et al., 2007), causing a longstanding, MMP-mediated degradation of collagen in both the RES-Ctr and BAGPR groups. In addition, a durable acidic environment may have induced supplementary

precipitation

dicalcium

octocalcium

phosphates

(Jayaraman and Subramanian, 2002, Mandel and TAS, 2010) during buffered condition (replacement of PBS) within the microporosities generated by the degradation of the dentine collagen fibrils (Fig. 4.5C). Indeed, as a result of this probable additional precipitation of mineral over time, the interface created using the BAG-PR bonding technique may have achieved mechanical characteristics similar to those created using glass ionomer cements (GICs) applied onto polyacrylic acid-treated dentine (Sauro et al., 2012a, Liu et al., 2009, Hewlett et al., 1991). This is probably why bond strength reduction and gap formation were observed in the BAG-PR specimens. The GIC-bonded interfaces can reach a tensile or shear bond strength of approximately 5 MPa and frequently prefail during specimen preparation (Hewlett et al., 1991, Berry and Powers, 1994). Yip et al. (Yip et al., 2001) affirmed that the results obtained from tensile testing of GICs bonded to dentine do not represent the actual strength of such stiff bonded interfaces and that only an accurate ultramorphology analysis using

194

electron microscopy may reveal the proper bonding ability of such restorative materials. However, it is also important to consider that the hydrophilic characteristics conferred by specific resin monomers, such as HEMA and PMDM, within the tested adhesives (Fig. 4.1) may have compromised the mechanical properties (i.e. modulus of elasticity) of the hybrid layers (Giannini et al., 2004, BedranRusso et al., 2007) as a result of polymer hydrolysis and swelling tensions generated within the polymer chains. In contrast, the BAG microfiller contained within the adhesive used in the BAG-AD group may have absorbed and used the water not required by the hydrophilic/acid monomers for the bioactive processes of conversion into apatite (Sauro et al., 2012a), thus preventing the polymer

network

from

considerable

hygroscopic/hydrolytic

degradation

(Ferracane, 2006).

4.5 Conclusion In conclusion, this study provided preliminary evidence for the use of bioactive Ca/Na

phosphosilicate,

such

BioglassÂŽ

45S5,

dentine-bonding

procedures in order to enhance the durability of the resin-dentine interfaces. However, further in vitro (i.e. transmission electron microscopy as well as computer-controlled indentation techniques) and long-term clinical studies are required to confirm the protective/therapeutic effects of BAG on the resindentine interface. Confocal Raman analysis will be also necessary to confirm the chemical nature of the mineral precipitates observed within the bondeddentine interfaces created with the two experimental BAG-bonding procedures.

195

Chapter 5: Experimental etch-and-rinse adhesives doped with calcium silicate-based micro-fillers to generate therapeutic bioactivity within resin-dentine interfaces

196

5.1 Introduction The durability of resin-dentine interfaces represents one of the main concerns in the contemporary adhesive dentistry as they are affected by severe degradation processes associated with water contact. Bond degradation occurs via water sorption (Ito et al., 2005), hydrolysis of monomer methacrylates ester bonds caused by salivary esterases (Moszner et al., 2005), and hydrolysis of collagen fibrils which may be enhanced by activation of endogenous dentine matrix metalloproteinases (MMPs) (Pashley et al., 2004). Regarding the different mechanisms of degradation, strategies to preserve the intact hybrid layers such as ethanol-wet bonding (Sadek et al., 2007, Tay et al., 2007) and the use of MMP inhibitors (Carrilho et al., 2007a) have been proposed. Nevertheless, current attempts to extend the longevity of resin-dentine bonds via incorporation of more hydrolytically stable resin monomers (Moszner et al., 2006) and/or the use of matrix metalloproteinase inhibitors (Carrilho et al., 2010) fail to address two fundamental issues in the degradation of resin-dentine bonds: 1) replacement of the mineral phase within the demineralised dentine collagen; 2) protection of the collagen from biodegradation (Bakry et al., 2011b). The use of bioactive materials which promptly interact with dental hard tissues through therapeutic/protective effects may provide a feasible means to extend the longevity of resin-dentine bonds (Profeta et al., 2012). Furthermore, experimental resin-based calcium-phosphate cements have been advocated as potential therapeutic restorative base-liner materials due to their ability to induce remineralisation of caries-affected dentine-bonded interfaces (Peters et al., 2010). Nevertheless, alternative strategies are being developed in order to enhance calcium, hydroxyl, and phosphate ions delivery within and beneath the

197

hybrid layers. Calcium-silicate cements are Portland-derived cements able to release calcium and hydroxyl

ions (alkalinising activity) (Huan et al., 2008,

Coleman et al., 2009), so creating favourable conditions for the remineralisation of dental hard tissues (i.e. dentine and enamel). These materials possess a bioactive behavior since they are able to develop apatite on their surface in a short induction period (Taddei et al., 2011) eliciting a positive response at the interface from the biological environment (Parirokh and Torabinejad, 2010). However, the use of the Portland cement-based materials in operative dentistry is still debated due to clinical limitations related to their long setting time (Taddei et al., 2011, Mandel and TAS, 2010), high dissolution rate and poor mechanical properties (Bunker et al., 1988). In contrast, the incorporation of resin specific monomers such as 2-hydroxyethyl methacrylate (HEMA), triethyleneglycol dimethacrylates (TEGDMA) and urethane dimethacrylates (UDMA) in silicatebased materials has been proposed to improve the mechanical properties, bond strength to dental tissues and reduce the setting time (Taddei et al., 2011, Wang and Spencer, 2005). Since there is little information concerning the use of such “hybrid” resin-base light-curable adhesive materials, this study was purposed to assess the therapeutic/bioactive effects of three newly developed experimental bonding agents containing modified Portland cement-based micro-fillers on resin-dentine interfaces. This aim was accomplished by evaluating the microtensile bond strength (µTBS) after simulated body fluid solution (SBS) storage (24 h or 6 months). Scanning electron microscopy (SEM) fractography on the de-bonded specimens and confocal microscopy (CLSM) analysis of the ultramorphology and nano-leakage of the resin-dentine interface were executed. The null

198

hypotheses to be tested were that the inclusion of tested micro-fillers within the composition of the experimental bonding agent induces: (i) no effect on the bond strength durability; (ii) no mineral precipitation and nano-leakage reduction within the demineralised ‘poorly resin-infiltrated’ areas within the resin-dentine interface.

5.2 Materials and methods

5.2.1 Preparation of the experimental bioactive resin-base bonding agents A type I ordinary Portland cement (82.5 wt%), (OPC: Italcementi Group, Cesena, Italy) mainly consisting of tri-calcium silicate (Alite: 3CaO x SiO2), dicalcium silicate (Belite: 2CaO x SiO2), tri-calcium aluminate (3CaO x Al2O3) and gypsum (CaSO4 x 2H2O) was mixed with 7.5 wt% of phyllosilicate consisting of sodium-calcium-aluminum-magnesium

silicate

hydroxide

hydrate

[(Na,Ca)(Al,Mg)6(Si4010)3(OH)6-nH2O; Acros Organics, Fair Lawn, NJ, USA] in deionised water (Ratio 2:1) to create the first experimental filler (HOPC). The second experimental filler (HPCMM) was created by mixing 90 wt% of type I OPC, 7.5 wt% phyllosilicate and 2.5 wt% of hydrotalcite consisting of aluminummagnesium-carbonate hydroxide hydrate [(Mg6Al2 (CO3)(OH)16·4(H2O); SigmaAldrich]. The third filler (HPCTO) used in this study was created by mixing OPC (80 wt%), phyllosilicate (7.5 wt%), hydrotalcite (2.5 wt%) and 10 wt% titanium oxide (TiO2: Sigma-Aldrich, Gillingham, UK). The three modified Portland-base silicates were mixed with deionised water (Ratio 2:1) and allowed to set in incubator at 37°C for 24 h. Subsequently, they were ground in an agate jar and sieved to obtain < 30 μm-sized micro-filler particles.

199

A resin co-monomer blend was prepared as a typical three-step, etch-and-rinse bonding agent including a neat resin blend as bond and a 50 wt% ethanolsolvated resin mixture as primer (Res-Ctr - no filler). The neat resin blend was formulated by using 40 wt% of a hydrophobic crosslinking

dimethacrylate

2,2-bis[4(2-hydroxy-3-methacryloyloxy-propyloxy)-

phenyl]-propane (Bis-GMA; Esstech, Essington, PA, USA) and 28.75 wt% of hydrophilic 2-hydroxyethyl methacrylate (HEMA; Aldrich Chemical, Gillingham, UK). An acidic functional monomer 2,5-dimethacryloyloxyethyloxycarbonyl-1,4benzenedicarboxylic acid (PMDM: Esstech Essington) was also added (30 wt%) to the blend solution to obtain a dental bonding system with chemical affinity to Calcium (Ca2+) present in dentine and in the micro-fillers (Fig. 5.1). The neat resin was made light-curable by adding 0.25 wt% of camphoroquinone (CQ; Aldrich), 0.5 wt% of 2-ethyl-dimethyl-4-aminobenzoate (EDAB; Aldrich) and 0.5% diphenyliodonium hexafluorophosphate. The resin co-monomer blend was used as control filler-free or mixed with each micro-filler in order to formulate three experimental resin-based agents (GB patent application No 1118138.5 - filed on 20th October 2011): i) Res-HOPC: 60 wt% of neat resin and 40 wt% of HOPC; ii) Res-HPCMM: 60 wt% of neat resin and 40 wt% of HPCMM; iii) Res-HPCTO: 60 wt% of neat resin and 40 wt% of HPCTO filler (Table 5.1). The hybrid calcium silicate-based adhesives systems were prepared by mixing the neat resin and the fillers for 30 s on a glass plate to form a homogeneous paste prior the bonding procedures.

200

Figure 5.1 - Chemical structures of the methacrylate monomers used in the tested resin blends. Abbreviations. Bis-GMA: Bisphenol A diglycidyl methacrylate; HEMA: 2-hydroxyethyl methacrylate; TEGDMA: triethylene-glycol dimethacrylate; PDMD: Bis(2-Methacryloyloxyethyl) Pyromellitate.

201

Table 5.1 - Chemical composition (wt%) and application mode of the experimental adhesive system used in this study. Abbreviations. Bis-GMA: bisphenyl A glycidyl methacrylate; HEMA: hydrophilic 2-hydroxyethyl methacrylate; PMDM: 2,5-dimethacryloyloxyethyloxycarbonyl-1,4-benzenedicarboxylic acid; HOPC: set Portland cement and smectite; HPCMM: Portland cement, Smectite and Hydrotalcite; HPCTO: set Portland cement, Smectite, Hydrotalcite and Titanium Oxide. *Three discs for each experimental resin-base material (6 mm in diameter and 1 mm thick) were light-cured for 30 s, immersed in 25 ml of H2O (pH 6.7) at 37Â°C and maintained for 30 days; the pH/alkalinising activity was evaluated using a professional pH electrode (Mettler-Toledo, Leicester, UK) at room temperature (~ 24Â° C)

202

5.2.2 Specimen preparation and bonding procedures Caries-free human molars (age 20-40 yr.), extracted for periodontal reasons under a protocol approved by the institutional review board of Guy's Hospital (South East London ethical approval 10/H0804/056), were used in this study. The treatment plan of any of the involved patients, who had given informed consent that their extracted teeth could be used for research purposes, was not altered by this investigation. This study was conducted in accordance with the ethical guidelines of the Research Ethics Committee (REC) for medical investigations. The teeth were stored in deionised water (pH 7.1) at 4Â°C and used within 1 month after extraction. A flat mid-coronal dentine surface was exposed using a hard

tissue microtome

(Accutom-50;

Struers, Copenhagem, Denmark)

equipped with a slow-speed, water-cooled diamond wafering saw (330-CA RS70300; Struers). The roots were sectioned 1 mm beneath the cemento-enamel junction (CEJ) using the slow-speed diamond saw. A 180-grit silicon carbide (SiC) abrasive paper mounted on a water-cooled rotating polishing machine (Buehler Meta-Serv 3000 Grinder-Polisher, DĂźsseldorf, Germany) was used (30 s) to remove the diamond saw smear layer and to replace it with a standard and more clinically related smear layer (Koibuchi et al., 2001). The specimens were divided into four groups (n = 5/group) based on the tested materials (Table 5.1). The specimens were etched using a 37% phosphoric acid solution (H 3PO4; Aldrich Chemical) for 15 s followed by a copious water rinse. The etcheddentine surfaces were gently air-dried for 2 s to remove the excess of water and leave a wet reflective-surface. The control and experimental adhesives (ResCtr; Res-HOPC; Res-HPCMM; Res-HPCTO) were applied within a period of 20

203

s. The specimens were immediately light-cured for 30 s using a quartztungsten-halogen (QTH) lamp (600mW/cm-2, Optilux VLC; Demetron, CT, USA). Five 1-mm-thick incremental build-up were performed using a resin composite (Filtek Z250; 3M-ESPE, St Paul, MN, US) light-activated for 20 s each step with a final curing of 60 s (Figure 5.2). The specimens were finally stored in SBS solutions (Oxoid, Basingstoke, Hampshire, UK) for 24 h and 6 months at 37Â°C.

5.2.3 μTBS and SEM observations of the failed bonds The specimen of each group were sectioned perpendicular to the adhesive interface with a slow speed water-cooled diamond wafering blade (Accutom-50; Struers) mounted on a hard tissue microtome (Isomet 11/1180; Buehler). Subsequently, match-sticks with cross-sectional adhesive area of 0.9 mm2 were created (Fig. 5.2). As each tooth yielded 16 beams, there were 80 match-sticks in total per bonding material. For each group, half of the match-sticks (n = 40) were tested after 24 h and the remaining half (n = 40) after 6 months of SBS storage at 37°C. Each resin-dentine match-stick was attached to a testing apparatus with a cyanoacrylate adhesive (Zapit, Dental Ventures, CA, USA). A tensile load was applied with a customised micro-tensile jig in a LAL300 linear actuator (SMAC Europe, Horsham, West Sussex, UK) with LAC-1 high speed controller single axis with built-in amplifier, that has a stroke length of 50 mm, peak force of 250 N, displacement resolution of 0.5 mm and crosshead speed of 1 mm-1 (Sauro et al., 2012b). The load (N) at failure and the cross-sectional area of each failed beam (measured with a digital micrometer: Mitutoyo CD15; Mitutoyo, Kawasaki, Japan) permitted calculation of the μTBS that was expressed in MPa. The μTBS (mean-MPa) data for each group were subjected to a repeated measures ANOVA and Tukey's post-hoc test for pair-wise comparisons (α = 0.05). Fisher's least significant difference (LSD) test was used to isolate and compare the significant differences (P < 0.05) between the groups. Premature failures were included in the statistical analysis as zero values. Modes of failure were classified as percentage of adhesive (A), mixed (M), or cohesive (C) when the failed bonds were examined at x 30 magnification with a

205

stereoscopic

microscope

(Leica

M205A;

Leica

Microsystems, Wetzlar,

Germany). For each group, five representative de-bonded specimens, depicting the most frequent failure modes, were chosen for SEM ultra-morphology analysis of the fractured surfaces. They were dried overnight and mounted on aluminum stubs with carbon cement. They were sputter-coated with gold (SCD 004 Sputter Coater; Bal-Tec, Vaduz, Liechtenstein) and examined using a scanning electron microscope (SEM) (S-3500; Hitachi, Wokingham, UK) with an accelerating voltage of 15 kV and a working distance of 25 mm at increasing magnifications from x 60 to x 5000.

5.2.4 Dye-assisted CLSM evaluation Three further dentine-bonded specimens were prepared as previously described for each group with the primer/bond resins doped with 0.1 wt% Rhodamine-B (Rh-B: Sigma-Aldrich, Munich, Germany) and then serially sectioned across the adhesive interface to obtain resin-dentine slabs (n = 12 per group) with a thickness of approx. 1 mm (Fig. 5.2). The resin-dentine slabs were then allocated to two subgroups (n = 6/group) based on the period of storage in SBS (24 h or 6 months). Following each ageing period, the specimens were coated with two layers of fast-setting nail varnish applied 1 mm away from the resin-dentine interfaces. Three specimens from each subgroup were immersed in 1 wt% aqueous fluorescein (Sigma-Aldrich) and the other three specimens in 0.5 wt% Xylenol Orange solution (XO: Sigma-Aldrich) for 24 h at 37Â°C. The latter is a calcium-chelator fluorophore commonly used in mammals bone remineralisation studies (Rahn and Perren, 1971), due to its ability to form complexes with divalent Ca2+ ions. The specimens were then

206

treated in an ultrasonic water bath for 2 min and polished using ascending (#1200 to #4000) grit SiC abrasive papers (Versocit; Struers) on a water-cooled polishing device (Buehler Meta-Serv 3000 Grinder-Polisher; Buehler). A final ultrasonic treatment (5 min) concluded the specimen preparation for the confocal microscopy analysis which was immediately performed using a confocal laser scanning microscope (Leica SP2 CLSM; Leica, Heidelberg, Germany) equipped with a 63x / 1.4 NA oil immersion lens. The xylenol orange and the fluorescein were excited at 488-nm using an argon/helium laser. The ultramorphology evaluation (resin-diffusion) was executed using a 568-nm krypton (rhodamine excitation) laser. CLSM images were obtained with a 1 μm z-step to optically section the specimens to a depth up to 20 μm below the surface (Sauro et al., 2012a). The z-axis scans of the interface surface were arbitrarily pseudo-colored by the same operator for better exposure and compiled into single projections using the Leica image-processing software (Leica). The configuration of the system was standardised and used at the same settings for the entire investigation. Each resin-dentine interface was completely investigated and then five optical images were randomly captured. Micrographs representing the most common features of nano-leakage observed along the bonded interfaces were captured and recorded (Profeta et al., 2012).

5.3 Results

5.3.1 μTBS and SEM observations of the failed bonds The interaction between bonding system versus SBS storage was statistically significant only for the Res-HOPC and Res-HPCMM groups (P = 0.001); no

207

significant reduction of the µTBS values was observed after 6 months of SBS ageing (P > 0.05). Conversely, significant drops in µTBS values were observed in both Res-HPCTO and Res-Ctr groups (P < 0.05) after prolonged storage in SBS (6 months). The µTBS results (expressed as Mean and SD) and modes of failures obtained for each group are summarised in Table 5.2. All the tested materials showed high μTBS values after 24 h of SBS storage with failures occurring mainly in the cohesive mode. However, only the resindentine specimens of the Res-HOPC and Res-HPCMM groups maintained high μTBS values (P > 0.05) after 6 months of storage in SBS (31.3 ± 11.5 and 24 ± 12.7 MPa, respectively) and they debonded prevalently in cohesive mode (57% and 54%, respectively). The SEM analysis of the fractured surfaces at 24h of SBS storage revealed the absence of both exposed dentine tubules and collagen fibrils indicating a good hybridisation of dentine (Figures 5.3A and 5.3B, respectively). After 6 months of SBS storage, the dentine surfaces were characterised by embedded mineral crystals and remnant resin presenting filler lacunas (Res-HOPC: Figures 5.3A1 and 5.3A2; Res-HPCMM: Figures 5.3B1 and 5.3B2). In contrast, a significant drop (P < 0.05) in μTBS was observed after 6 months of storage in SBS, with the specimens created using the Res-Ctr group (filler-free) and with those created with the Res-HTCPO. These latter specimens showed, after 24 h, a well-hybridised de-bonded surface embedding micro-fillers (Fig. 5.3C). On the contrary, the specimens de-bonded after 6 months of SBS ageing showed a de-bonded surface with a few dentinal tubules but with no sign of clear degradation and well-hybridised peritubular dentine with resin tags (Figures 5.3C1 and 5.3C2). The Res-Ctr specimens tested after 24 h of SBS storage and analysed with SEM presented few exposed dentinal

208

tubules but mostly were obliterated by resin tags or covered by resin remnants (Fig. 5.3D). Conversely, the surface of the specimens de-bonded after 6 months of SBS exhibited no collagen fibrils on the dentine surface with rare resin tags and degraded funneled dentinal tubules (Figures 5.3D1 and 5.3D2).

Table 5.2 - Mean and standard deviation (SD) of the ÎźTBS (MPa) to dentine. Values are mean Âą SD in MPa. In each row, same numbers indicate no differences (p > 0.05) after 24 h and 6 m of SBS storage. In columns, same capital letter indicates no statistically significant differences between each group (p > 0.05). Premature failures were included in the statistical analysis as zero values and are indicated in parentheses (for instance 5/35 means that there were 5 premature failures and 35 testable beams). The modes of failure are expressed in percentage in the brackets [adhesive/mix/cohesive].

209

Figure 5.3 - SEM failure analysis of debonded specimens. (A): SEM micrograph (1000x magnification) of an adhesively fractured stick bonded with Res-HOPC after 24 h of SBS storage. Observe the dentine entirely covered adhesive resin (ar) with some fillers’ lacunas (pointer) and rarely found opened dentinal tubules (dt). (A1) In another specimen, note the presence of resin adhesive (ar) onto dentine without unprotected collagen fibrils. Some fillers were detached during SEM preparation (pointer) and initial mineral precipitation may be observed (white asterisk). (A2)After 6 months, the debonded dentine surface at higher magnification (2.500X) kept covered with adhesive resin (ar).Mineral crystals (asterisk) embedded within a preserved collagen network were vastly encountered albeit some fillers were detached (pointer). (B): SEM micrograph of a specimen bonded with Res-HPCMM after 24h. Note the presence of adhesive resin (ar) mostly covering the dentine and dentinal tubules (dt);; some fillers’ lacunas are also observed (pointer). (B1)At another region of the same specimen, more lacunas are observed due to filler detaching, but also initial mineral crystallisation was depicted (asterisk). (B2) After 6 months storage, a failure mode observed under higher magnification (2.500X) similar to that found in B1 is disclosed showing the very slow bonding degradation. More mineral precipitation was observed (asterisk) and the fillers’ lacunas (pointer) are wider due to expansion of the fillers when exposed to water. (C): Micrograph of a debonded stick from group Res-HPCTO after 24h showing dentine completely covered with rare filler detachment (pointer). (C1) In other region, note the presence of few exposed dentinal tubules as well as intact resin tags (rt). (C2) After 6 months, the fractured dentine surface bonded with Res-HPCTO showed a de-bonding at hybrid layer and/or at its bottom. A lot of resin tags (rt) were observed well-hybridised with peritubular dentine (black pointer) suggesting potential remineralisation albeit some funneled dentinal tubules were encountered without resin tags (black asterisk). (D): SEM micrograph from a specimen bonded with Res-Ctr and presenting a perfectly hybridised and resin covered dentine. (D1) Other fractured stick from the same group showed an adhesive failure at the bottom of hybrid layer with remnants of resin adhesive (ar) and some resin tags (rt). (D2) The control adhesive after 6 months showed many signs of degradation since most collagen fibrils were degraded, the funneled dentinal tubules (asterisk) were often found along with poorly hybridised resin tags (pointer). Symbols. White finger: filler lacuna; white asterisk: mineral precipitation; black asterisk: funneled degraded peritubular dentine; black finger: hybridisation between tags and peritubular.

210

5.3.2 Dye-assisted CLSM evaluation CLSM imaging of the bonded-dentine interfaces subsequent to 24 h of SBS storage revealed relevant ultramorphology and nano-leakage information for all groups. It was observed that all tested materials were able to diffuse within the demineralised dentine, creating an hybrid layer 7-10 μm thick, with a multitude of resin tags penetrating the dentinal tubules (Fig. 5.4). Nevertheless, all these interfaces were affected by conspicuous fluorescein penetration (nano-leakage) within porous hybrid layers and through dentinal tubules (Fig. 5.4). Furthermore, the resin-dentine interface created using the experimental bonding agents containing the micro-fillers and the acidic functional monomer PMDM showed presence of calcium-chelator dye (XO: Xylenol orange) also within the hybrid layer and inside the dentinal tubules (Fig. 5.5). On the contrary, the acid-etched dentine bonded using the resin control (Res-Ctr, filler-free) showed no presence of XO along the interface. Significant ultramorphological changes were observed subsequent to prolonged SBS storage. For instance, the CLSM analysis revealed no gap and limited fluorescein penetration (nano-leakage) within the resin-dentine interfaces created using the Res-HOPC and Res-HPCMM (Figures 5.6A and 5.6B). In addition, XO produced a clearly outlined fluorescence due to a consistent Caminerals deposition within the resin-dentine interface and inside the dentinal tubules (Figures 5.5A and 5.5B). The resin-dentine interfaces created using Res-HPCTO showed less nano-leakage within the hybrid layer; resin degradation of the adhesive layer was also observed (Fig. 5.6C). Intense nanoleakage and constant gaps affected the resin-dentine interfaces created using the Res-Ctr (Fig. 5.6D). When the same interfaces were investigated employing

211

xylenol orange, only the walls of the dentinal tubules were stained by the fluorescent calcium-chelator dye (Fig. 5.5E).

212

Figure 5.4 - Confocal laser scanning microscopy (CLSM) single-projection images showing the interfacial characterisation and nanoleakage, after 24 h of storage in SBS. Images (1) indicate the projection of fluorescein dye whereas the images (2) disclose the projection of rhodamine B dye. The images (3) are depicting the projections of both dyes. (A1, A2, A3): CLSM images showing the interfacial characteristics of the bonded-dentine interface created using ResHOPC. It is possible to observe a clear hybrid layer (hl) with long resin tags (rt) penetrating the dentinal tubules (dt) underneath an adhesive layer (ad) characterised by evident mineral fillers (FL). Intense fluorescein uptake was observed within the entire resin-dentine interface as well as the adhesive layer. (B1, B2, B3): Micrographs showing the interfacial characteristics of the bonded-dentine interface created using ResHPCMM. Similarly to images from Res-HOPC, these images presented high dye uptake throughout the entire resindentine interface as well as the adhesive layer. (C1, C2, C3): The resin-dentine interface created using the Res-HPCTO bonding system was characterised by a clear hybrid layer (hl) located underneath the adhesive layer (ad) containing the experimental filler (FL). Long resin tags (rt) penetrating the dentinal tubules (dt) were observed as well as evident nanoleakage and dye uptake along the entire interface and adhesive layer. (D1, D2, D3): CLSM images showing the bonded-dentine control interface (RES-Ctr) characterised by a thick and fluorescent hybrid layer (hl) (approximately 8 Îźm thickness) located underneath an adhesive layer (ad) devoid of fillers.

213

Figure 5.5 - CLSM single-projection images disclosing the fluorescent calcium-chelators dye xylenol orange. All images were obtained from specimens immersed in simulated body-fluid solution for 24 h or 6 months. (A): CLSM image of the resin-dentine interface created with Res-HOPC after 24 h of SBS storage. Mineral deposition can be visualised within the adhesive layer (ad), the hybrid layer (hl) along the walls of dentinal tubules (dt) and the filler inside the resin tags (rt). (B): CLMS image of the resin-dentine interface created with Res-HPCMM and immersed in SBS for 6 months where it is possible to observe a clear fluorescence signal due to a consistent presence of Cadeposits within the adhesive layer (ad), hybrid layer, walls of the dentinal tubules (dt) and resin tags (rt). (C): Image of the resin-dentine interface created with Res-HPCTO and immersed in SBS for 24h. Xylenol Orange was able to stain the Ca-minerals within adhesive layer, hybrid layer and dentinal tubule (dt). Note the intense calcium deposition at bottom of hybrid layer. (D): Image of the resin-dentine interface created with Res-HPCTO and immersed in SBS for 6 months showing also in this case Ca-mineral presence at the bottom and within the hybrid layer, dt and rt. (E): Image of the resin-dentine interface created with Res-Ctr (no filler) in which one may note absence of calcium deposition both within the hybrid (hl) and adhesive layer (ad). Only the walls of the dentinal tubule tubules (dt) were stained by the fluorescent dye.

214

Figure 5.6 - Confocal laser scanning microscopy (CLSM) single-projection images showing the interfacial characterisation and nanoleakage after 6 months of SBS storage. (A): Image showing the resin-dentine interface created using Res-HOPC characterised by reduced nanoleakage within the hybrid layer (hl). Note the absence of fluorescein uptake within the adhesive layer (ad). (B): Image showing the interfacial features of the bonded-dentine interface created using the ResHPCMM. Note the low overall nanoleakage with very little fluorescein uptake in hybrid (hl) and adhesive layers (ad). (C): Image of the resin-dentine interface created using Res-HPCTO. Despite the mineral deposition and reduced nanoleakage within adhesive layer (ad), the weak bond strength created gaps (gap) in which the fluorescein was deposited. The gaps may be induced by the cutting procedures as the resin degradation was replaced by mineral precipitation creating an interface with low elasticity (high stiffness properties). The nanoleakage was only observed within the hybrid layer (hl). (D): Micrograph of the resin-dentine interface created using the control adhesive system (RES-Ctr). Note the presence of intense dye uptake (nanoleakage) within the hybrid layer (hl) and at the bottom of adhesive layer (ad). In this case the presence of gaps frequently observed between hybrid and adhesive was very likely due to hybrid layer degradation (reduced thickness). Other symbols. dt: dentinal tubules; rt: resin tags; c: composite.

215

5.4 Discussion The resin-dentine interfaces created using contemporary “simplified” etch-andrinse bonding agents are affected by bond strength reduction subsequent to prolonged water ageing (Breschi et al., 2008). This phenomenon occurs due to the inability of such materials to completely replace loosely bound and bulk-free water from the apatite-depleted dentine collagen matrix during bonding procedures

which

degradation

cause

polymer

hygroscopic networks

and

swelling favours

effects dentinal

and

hydrolytic

collagenolytic

metalloproteinases (MMPs)-mediated activity upon water ageing (Breschi et al., 2008, Kim et al., 2010d, Pashley et al., 2011, Breschi et al., 2010b, Liu et al., 2011a). Nevertheless, the presence of water may be essential to facilitate apatite nucleation within the gap zones of collagen fibrils and fossilisation of the host-derived, collagen-bound enzymes (MMPs) (Kim et al., 2010d, Pashley et al., 2011, Tay and Pashley, 2009). Recent investigations have demonstrated that it is possible to reduce the nano-leakage and micropermeability within the resin-dentine interface and maintain the bond strength (Profeta et al., 2012) of bioactive

resin-base

materials

applied

H3PO4

acid-etched

dentine

subsequent to simulated body fluid storage for 3-6 months (Wang and Spencer, 2005, Sauro et al., 2012a). Ryou et al. (Ryou et al., 2011) demonstrated that using a biomimetic remineralisation approach it is feasible to remineralise the dentine collagen within the resin-dentine interface via slow release of calcium ions from set white Portland cement and subsequent interaction of these ions with phosphate species from SBS or dentine substrate. Portland cements designed for dental applications, also called hydraulic silicate cements or MTA, mainly constituted by Alite (3CaO x SiO2), Belite (2CaO x SiO2) and tri-calcium

216

aluminate (3CaO x Al2O3), exhibit outstanding biological properties and high bioactivity when immersed in SBF (Wang and Spencer, 2005, Ryou et al., 2011, Darvell and Wu, 2011). In the present study, modified Portland cement-based micro-fillers (< 20 μm) were included within the composition of a representative three-step/etch-andrinse bonding agent in order to create a material with therapeutic remineralising effects on the mineral-deficient areas along the resin-dentine interface. Based on the results obtained in this study, the first null hypothesis that the inclusion of tested micro-fillers within the composition of the experimental bonding agent has no effect on the bond strength durability must be rejected as only the use of Res-HOPC and Res-HPCMM bonding agents preserved the bond strength durability. The second null hypothesis that no mineral precipitation and nanoleakage reduction will be observed within the demineralised

‘poorly resin-

infiltrated’ areas within the resin-dentine interface must be also rejected. In detail, the three experimental bonding agents containing experimental microfillers (Res-HOPC, Res-HPCMM and Res-HPCTO) and the control co-monomer blend (RES-Ctr) used to bond the acid-etched dentine produced comparably high μTBS values (P > 0.05) following 24 h of storage in SBS (Table 5.2). Conversely, after 6 months of storage in SBS a significant decrease in μTBS (P < 0.05) was observed for the RES-Ctr and Res-HPCTO groups, while the specimens bonded using Res-HOPC or Res-HPCMM maintained consistent long-term bond strength values (P > 0.05) compared to the control group (24 h SBS storage). The specimens of the Res-HOPC and Res-HPCMM groups debonded after 6 months of SBS storage showed, during SEM fractography examination, residual resin presence and newly formed mineral-bodies (Figures

217

5.3A1 and 5.3A2, 5.3B1 and 5.3B2). Several morphological differences were observed in the specimens of the Res-HPCTO group which presented a debonded surface characterised by very few partially exposed dentinal tubules and an important precipitation of mineral crystals after 6 months of SBS storage (Figures 5.3C1 and 5.3C2). The SEM analysis revealed that the de-bonded dentine surface of the specimens in RES-Ctr group was well resin-hybridised and characterised by no exposed collagen fibrils after 24 h of SBS storage (Fig. 5.3D). In contrast, the prolonged SBS storage (6 months) induced radical changes; the dentine surface presented funneled dentinal tubules as a sign of degradation of the “poorly resin-infiltrated” demineralised peritubular dentine (Figures 5.3D1 and 5.3D2). These results were also supported by the CLSM analysis performed to evaluate the nano-leakage and the presence of calciumcompounds within the resin-dentine interface subsequent to SBS storage (24 h or 6 months). Indeed, further evidence of the therapeutic bioactivity of the experimental bonding agents containing the modified Portland cement-based micro-fillers were attained; reduced fluorescent-dye uptake (nano-leakage) was observed along the entire resin-dentine interface after 6 months of storage in SBS (Figures 5.5A, 5.5B and 5.5C). These latter observations along with the strong xylenol orange signal from the hybrid layer and the dentinal tubules (Fig. 5.5), clearly indicated the remineralisation of those areas which were previously detected within the resin-dentine interface as mineral-deficient/poor-resin infiltrated zones. It is hypothesised that the therapeutic remineralising effects observed within the mineral-depleted resin-dentine interface were essentially due to the bioactivity of the experimental micro-fillers. Indeed, the reaction mechanism of the

218

Portland cement-based micro-fillers involved the reaction of the polymerised calcium-silicate hydrate gel with water to release calcium hydroxide and to the consequent increase of the alkalinity of the surrounding environment (Lawrence, 1998); this increase of pH was confirmed in this study (Table 5.2). This localised increase in pH within the resin-dentine interface may have interfered with the activity of MMPs (Pashley et al., 2004, Breschi et al., 2008). Furthermore, the interaction between the phosphate ions present in the ageing solution (SBF) or in the dentine substrate, and the calcium released from the experimental Portland-based micro-fillers may have enhanced the formation of new apatite deposits upon existing mineral constituents within the dentine matrix (biocatalysation) (Gandolfi et al., 2010c). However, it is well known that the increase in environmental pH and the presence of free OH â&#x2C6;&#x2019; may facilitate apatite nucleation and reduce the solubility of intermediate Ca/P species formed during apatite formation (Sauro et al., 2011b). The most appropriate pH to support the formation of stoichiometric hydroxyapatite (HA) in vitro (Bayle et al., 2007) and in vivo (Jayaraman and Subramanian, 2002) falls in a range between 8 and 9. At higher pH it is common to obtain a Ca-deficient HA (lower solubility then stoichiometric hydroxyapatite) characterised by higher concentrations of PO43- and lower Ca2+ ions (Liu et al., 2009). Furthermore, the presence of carboxylic species (R-COO-) within the acidic functional monomer (PMDM) used in this study may have acted as sequestering agent for Ca/P cluster favouring the precipitation of nano-apatite within the polymer network and dentine collagen (Tay and Pashley, 2009, Ryou et al., 2011). Moreover, the RCOO- species of PMDM may have interacted with the remnant calcium present along the front of demineralisation at the bottom of the hybrid layer acting as a

219

sort of biomimetic template primer which promoted precipitation of Cacompounds (Wang and Spencer, 2005, Sauro et al., 2012a). During all these processes for Ca/P nucleation, apatite precipitation may have reduced the distribution of water-rich regions within the resin-dentine interface (Wang and Spencer, 2005, Qi et al., 2012), interfering with the hydrolytic and hygroscopic mechanisms involved in the degradation of dental polymers (Ferracane, 2006). In addition to the formation of apatite crystals, the nanostructure of the calcium silicate hydrate may also have contributed to seal the dentinal tubules due the small-scale volume of the forming gels, along with a slight expansion of the calcium silicate-based materials once immersed in SBS (Skinner et al., 2010). In particular, the phyllosilicates (i.e. smectite) and hydrotalcite, which were contained in the micro-fillers used in this study, have the ability to expand considerably following water sorption into the interlayer molecular spaces (Bhattacharyya and Gupta, 2008). The amount of expansion is due largely to the type of exchangeable cation contained in the micro-filler; the uptake kinetics of cation exchange is fast and the presence of Na +, as the predominant exchangeable cation, can result in material swelling. In this condition, the exceeding water is removed, thereby preventing hygroscopic effects and hydrolytic degradation of the polymer chains (Malachovรก et al., 2009). Also, it is reasonable to expect that the metallic ions intercalated on phyllosilicate were easily released by ion-exchange with cations present in the surrounding solutions and acted as effective antibacterial substances in the long term (Ferracane, 2006). In contrast, the bond strength reduction observed in the resin-dentine interfaces created using the Res-HPCTO bonding agent after prolonged storage in SBS (Table 5.2) may be due to the high hydrophilicity of

220

the TiO2. Micro-fine titanium oxide (TiO2) have been used as an inorganic additive of resin composites to match the opaque properties of teeth (Yu et al., 2009) and as nano-particles to increase the micro-hardness and flexural strength of dental composites (Bhattacharyya and Gupta, 2008). However, TiO2 has been advocated as a super-hydrophilic component, in particular under ultraviolet (UV) light irradiation (Xia et al., 2008, Chadwick et al., 1994, Bonding et al., 1987). Therefore, a possible explanation for the μTBS reduction may be attributed to this high hydrophilicity which may have permitted excessive water adsorption which induced severe hydrolytic resin and collagen degradation as well as the extraction of water-soluble un-reacted monomers or oligomers from the resin-matrix (Karuppuchamy and Jeong, 2005). Moreover, the replacement of the degraded resin by mineral crystallisation within the Res-HPCTO bondeddentine interface (Figures 5.4 and 5.5) over prolonged SBS storage may have conferred mechanical characteristics related to bond strength comparable to those created by conventional glass-ionomer cements (GICs) applied onto polyacrylic acid-etched dentine and submitted to tensile tests (Spencer et al., 2010, Hewlett et al., 1991). Indeed, several studies indicated that the bond strength of GICs when tested using tensile or shear methods was approximately 5 MPa; these values results do not reflect the true adhesive strength to dentine (Berry and Powers, 1994). These factors may have been also responsible for the formation of gaps within the resin-dentine interface created by the ResHPCTO during the cutting/sample preparation (Profeta et al., 2012).

221

5.5 Conclusion In conclusion, as the results of this study demonstrated that the resin-dentine bond may be maintained over time by inducing a therapeutic remineralisation of the bonding interface, specific experimental resin bonding systems containing bioactive micro-fillers, such as Res-HOPC, Res-HPCMM or Res-HPCTO may offer the possibility to improve the durability of the resin-dentine interfaces. The characteristic of promoting bioactivity should also open up the potential to create therapeutic restorative materials able to reduce the incidence of secondary caries. Indeed, it is important to consider that restorative materials containing bioactive fillers may be effective in killing a wide selection of aerobic bacteria due to the increase of the local pH and concentration of alkaline ions (Yip et al., 2001, Hewlett et al., 1991). The antibacterial properties are potentially of great importance as the infiltration of microorganisms may cause secondary caries which jeopardise the longevity of resin-dentine interface leading to the replacement of dental restorations (Giannini et al., 2004, Sauro et al., 2006). Further studies are ongoing in order to evaluate the species-specific antibacterial effects and biocompatibility of the materials tested in this study.

222

Chapter 6: In vitro micro-hardness of resindentine interfaces created by etch-and-rinse adhesives comprising bioactive fillers

223

6.1 Introduction Current concepts of resin/dentine adhesion imply that chemicals are applied before bonding to alter the structure of dentine and favour resin infiltration (Nakabayashi and Pashley, 1998). Subsequently, resin hybridisation should restore the biological and mechanical properties of the partially demineralised dentine to approximate those of the original undemineralised dentine. Unfortunately, despite significant improvements in adhesive systems, the resin/dentine bonded interface formed by a mixture of collagen organic matrix, residual hydroxyapatite crystallites and resin monomers still remains the weakest area of adhesive restorations (Breschi et al., 2008). The discrepancy between the etching depth and the adhesive system penetrating capacity make the collagen-rich zone underlying the hybrid layer susceptible to nano-infiltration (Sano et al., 1995a). This may lead to the formation of pathways in which oral fluid and endogenous proteolytic enzymes concur to degrade each component of the resin-dentine bonds (Hashimoto et al., 2003). Recently, progressive removal of water by apatite deposition emerged as a viable strategy to address the fundamental issue of replacing mineral that is iatrogenically depleted during the acid etching phase (Kim et al., 2010e, Ryou et al., 2011). This should lead to a more durable form of tissue engineered dentine that is capable of preserving its organic components and maintain adhesive strength after ageing. Remineralisation of incompletely resin-infiltrated collagen matrices can be promoted by the use of bioactive, ion-releasing materials, e.g., glass-ionomer cements (Endo et al., 2010) or resin-based calcium-phosphate (CaP) cements

224

(Ngo et al., 2006). Bresciani et al. (Bresciani et al., 2010) revealed significantly increased Knoop hardness along the interface of resin-bonded dentine, confirming the capacity of materials containing CaP cements to promote remineralisation of cariesaffected residual dentine. Silicate compounds, including calcium/sodium phosphosilicates, such as BioglassÂŽ 45S5 (BAG), and certain calcium-silicate cements (often referred to as Cal-Sil) are known to release ions in aqueous solution and induce deposition of carbonated hydroxyapatite (HCA). Whereas BAG has been successfully used for dentine hypersensitivity (Greenspan, 2010, Salian et al., 2010) and Cal-Sil cements are already employed in dentistry for different endodontic clinical applications (Taddei et al., 2011), the development of resin-based restorative materials containing bioactive micro-fillers with remineralising effects on the mineral-depleted areas within the bonded-dentine interface remains an important target to accomplish. The present study further investigated the performance of four experimental adhesives either incorporating BAG (Chapter 3) or three distinct hydrated blends of Cal-Sil cements (Chapter 4), for their potential benefits with respect to the improvement of micro-hardness in resin-dentine interfaces created with etch-and-rinse bonding techniques. This aim was accomplished by comparing micro-hardness values of the area that was considered to represent the resindentine interface (the hybrid layer and its surroundings) after 24 h and 6 months of storage in phosphate buffered solution (PBS). Our null hypothesis was that treatment with adhesives containing BAG or Cal-Sil cements has a preserving effect on the micro-hardness of resin infiltrated dentine surfaces.

225

6.2 Materials and methods

6.2.1 Teeth collection and preparation Intact fresh human third molars extracted for surgical reasons were collected for this study with the informed consent of the donors (20 to 40 yr. old), following a protocol approved by the ethical guidelines of the Research Ethics Committee (REC) for medical investigations. The teeth were stored in deionised water (pH 7.1) at 4°C, and they were used within 1 month following extraction. Each tooth was prepared by exposing a flat mid-coronal dentine surface and removing the roots 1 mm beneath the cemento-enamel junction (CEJ) at an angle of 90° to their longitudinal axis using a slow-speed, water-cooled diamond wafering saw (330-CA RS-70300; Struers) mounted on a hard tissue microtome (Accutom-50; Struers, Copenhagem, Denmark). A 180-grit silicon carbide (SiC) abrasive paper installed on a water-cooled rotating polishing machine (Buehler Meta-Serv 3000 Grinder-Polisher; Buehler, Düsseldorf, Germany) was used (30 s) to remove the diamond saw smear layer and to replace it with a standard and more clinically relevant smear layer (Oliveira et al., 2003).

6.2.2 Formulation of the comonomer resin adhesive blend The resin co-monomer blend used in this study as dentine bonding agent represents the formulation of a typical three-step, etch-and-rinse adhesive. It was prepared from commercially available monomers - 2, 2-bis[4(2-hydroxy-3methacryloyloxy-propyloxy)-phenyl] propane (Bis-GMA; Esstech, Essington, PA, USA) and 2-hydroxyethyl methacrylate (HEMA; Aldrich Chemical, Gillingham, UK) - and included a 50 wt% ethanol-solvated resin mixture used as

226

primer [Bis-GMA, HEMA, PMDM, 50% absolute ethanol (Sigma-Aldrich)]. To obtain a dental bonding system with chemical affinity to Calcium (Ca 2+) available in dentine and in each micro-filler, we included the acidic functional monomer 2,5-dimethacryloyloxyethyloxycarbonyl-1,4-benzenedicarboxylic acid (PMDM: Esstech Essington). A binary photoinitiator system based on camphoroquinone (CQ; Aldrich) and 2-ethyl-dimethyl-4-aminobenzoate (EDAB; Aldrich) made the neat resin light-curable (Table 6.1).

227

Table 6.1 - Chemical composition (wt%) of the experimental adhesive systems used in this study. Abbreviations. Bis-GMA: bisphenyl A glycidyl methacrylate; HEMA: hydrophilic 2-hydroxyethyl methacrylate; PMDM: 2,5-dimethacryloyloxyethyloxycarbonyl-1,4-benzenedicarboxylic acid; CQ: camphoroquinone; EDAB: 2-ethyl-dimethyl-4-aminobenzoate; BAG: BioglassÂŽ 45S5; HOPC: set Portland cement and smectite; HPCMM: Portland cement, Smectite and Hydrotalcite; HPCTO: set Portland cement, Smectite, Hydrotalcite and Titanium Oxide.

228

6.2.3 Bioactive fillers and experimental bonding systems Bioglass® 45S5 (Sylc; OSspray, London, UK) with an average particle size of 20 μm was included within the composition of the neat resin as bioactive microfiller (60 wt% resin blend/40 wt% BAG) to create the first experimental dentine bonding system which was called Res-BAG (Table 6.1). Subsequently, three calcium-silicate micro-fillers were designed and prepared. A type I ordinary Portland cement (82.5 wt%), (identified as OPC: Italcementi Group, Cesena, Italy) constituted by tri-calcium silicate (Alite: 3CaO x SiO2), di-calcium silicate (Belite: 2CaO x SiO2), tri-calcium aluminate (3CaO x Al2O3) and gypsum (CaSO4 x 2H2O) was mixed with 7.5 wt% of phyllosilicate consisting of sodiumcalcium-aluminum-magnesium

silicate

hydroxide

hydrate

[(Na,Ca)(Al,Mg)6(Si4010)3(OH)6-nH2O; Acros Organics, Fair Lawn, NJ, USA] in deionised water (Ratio 2:1) to create the first Portland-base experimental filler (HOPC). The second experimental filler (HPCMM) was created by mixing 90 wt% of type I OPC, 7.5 wt% phyllosilicate and 2.5 wt% of hydrotalcite consisting of

aluminum-magnesium-carbonate

hydroxide

hydrate

[(Mg6Al2

(CO3)(OH)16·4(H2O); Sigma-Aldrich]. The third calcium silicate-based filler (HPCTO) used in this study was created by mixing OPC (80 wt%), phyllosilicate (7.5 wt%), hydrotalcite (2.5 wt%) and 10 wt% titanium oxide (TiO 2: SigmaAldrich, Gillingham, UK). Each cement was combined with deionised water (Ratio 2:1), allowed to set in an incubator at 37°C for 24 h and finally grinded in an agate ball mill as well as being sieved to obtain 20-30 μm-sized hydrated silicate fillers (HOPC, HPCMM, HPCTO). Ultimately, three Cal-Sil cements-based experimental adhesive systems were also prepared (GB patent application No 1118138.5 - filed on 20th October

229

2011): Res-HOPC (60 wt% resin blend/40 wt% HOPC); Res-HPCMM (60 wt% resin blend/40 wt% HPCMM); Res-HPCTO (60 wt% resin blend/40 wt% HPCTO) (Table 6.1). Each hybrid etch-and-rinse adhesive was prepared by mixing the neat resin and the fillers for 30 s on a glass plate to form a homogeneous paste prior to the bonding procedures. Overall, four experimental bonding systems were created in this study, while the application of the neat resin adhesive with no filler served as a control group (Res-Contr).

6.2.4 Bonding procedures The specimens were etched using 37% phosphoric acid solution (H3PO4; Aldrich Chemical) for 15 s followed by copious water-rinsing. The etcheddentine surfaces were gently air-dried for 2 s to remove the excess of water and leave a wet reflective substrate. Each time the bonding procedure was accomplished by applying two consecutive coats of the ethanol-solvated primer and a layer of the control or experimental bonding resin (Res-Contr, Res-BAG, Res-HOPC, Res-HPCMM, Res-HPCTO) within a period of 20 s. Light-curing was immediately performed for 30 s using a quartz-tungsten-halogen (QTH) lamp (600mWcm-2, Optilux VLC; Demetron, CT, USA). Following adhesive treatment, five 1-mm-thick increments of resin composite (Filtek Z250; 3MESPE, St Paul, MN, US) were built up and individually light-activated for 20 s. The resin bonded specimens were stored in PBS solutions (Oxoid, Basingstoke, Hampshire, UK) for 24 h and 6 months at 37Â°C.

230

6.2.5 Knoop micro-hardness (KHN) analysis Three dentine-bonded specimens were created for each group and every single sample was subsequently sectioned across the adhesive interface using a sectioning machine (Accutom-50; Struers) with a water-cooled diamond wafering saw (330-CA RS-70300; Struers) to obtain two 4 mm-thick resindentine slabs (n = 6 per group). Following 24 h of storage in PBS, each slab was treated in an ultrasonic water bath for 2 min and polished using ascending #500, #1200, #2400 and #4000 grit SiC abrasive papers (Versocit, Struers A/S, Copenhagen, Denmark) on a watercooled rotating polishing machine (Buehler Meta-Serv 3000 Grinder-Polisher; Buehler). Between each polishing step the slabs were cleansed in an ultrasonic bath containing deionized water for 5 min. The Knoop micro-hardness evaluation (Duramin-5, Struers A/S, DK-2750 Ballerup, Denmark) was performed using a 25 g load and 15 s dwell time to produce indentations with long diagonals of a suitable size for accurate measurements in relation to the thickness of the hybrid layer, while minimising surface damage. To minimise errors caused by tilting and to avoid the introduction of stresses during the micro-hardness testing, each section was mounted and supported on a glass slide using green-stick compound and a paralleling device. The metal chuck containing the dentine slices was clamped onto the stage of the testing machine and the surfaces were oriented perpendicular to the diamond indenter axis. Each polished surface received 15 indentations performed immediately after polishing to provide a more uniform surface for reading and to improve the

231

precision of the indentations. These were arranged in three widely separated straight lines starting from hybrid layer and placed perpendicular to the resindentine interface. As a result, any interference of the deformation areas caused by neighboring marks was avoided. Five measurements were executed along each line, every 30 µm up to 115 µm in depth for sufficient hardness data to be subjected to statistical analysis (Bresciani et al., 2010, Reinke et al., 2012) (Figure 6.1). The dentine surface was covered with a wet tissue paper for 1 min after each indentation to avoid dehydration of the surface (Xu et al., 1998). The length of the long diagonal of each indentation was determined immediately to avoid possible shrinkage caused by mechanical recovery of the tooth surfaces with a resolution of 0.1 μm (Duramin-5 software; Struers). The main criteria for accepting an indentation were clarity of outline and visibility of its apices. The values obtained were converted into Knoop Hardness Numbers according to the following formula: KHN = 14,229 P/d2 where P = applied load in g, and d = length of the longest diagonal in μm. After 6 months of PBS storage, hardness measurements were executed again from the same sections not far away from the first indentations. In total, 900 data points were obtained, 450 after 24 h and 450 after the prolonged ageing period, respectively. All micro-hardness values of the slabs obtained from the same tooth were averaged, and just one value per tooth was used in the statistical analysis. Mean (±SD) KHN numbers were treated with two-way analysis of variance (ANOVA) to determine differences between materials and the effect of ageing. Subsequent one-way ANOVA was performed to assess differences between

232

materials for the two different storage times separately. Post-ANOVA contrasts were performed using a Bonferroni test for multiple comparisons.

Figure 6.1 - Optical images obtained during the micro-hardness test along the resin-dentine interface. A): Picture illustrating how the five measurements (indentations) were taken along each line every 30 Âľm up to 115 Âľm in depth. B): At high magnification it is possible to observe that the first indentation was performed exactly on a hybrid layer (arrow) located between the adhesive resins (a) and the dentine surface (d).

233

6.3 Results 6.3.1 Knoop micro-hardness (KHN) analysis The indenter load produced micro-indentations with long diagonals that enabled accurate Knoop micro-hardness measurement. There was no major surface or subsurface damage evident on all the examined specimens, and microscopic inspection revealed no evidence of cracks radiating from the apices of the indentations. The biomechanical properties (surface micro-hardness) were influenced by dentine treatment, position along the resin-dentine interface and storage time. Two-way analysis of variance indicated that there was a statistically significant interaction between materials and the effect of ageing. Statistical comparisons of mean KHN values (±SD) at different depths obtained after 24 h and 6 months of PBS storage are shown in Table 6.2. Analysis of the present data showed a statistical reduction of KHN values after prolonged PBS storage (p < 0.001) within the hybrid resin-dentine zones created with the Res-Contr adhesive containing no filler (24 h: 24.4 ± 0.5 KHN; 6 m: 15.9 ± 2.9 KHN) and the experimental bonding system Res-HPCTO (24 h: 21.3 ± 3 KHN; 6 m: 14.1 ± 2.5 KHN). On the contrary, no significant KHN decrease of the average superficial micro-hardness (p > 0.001) was observed after 6 m in all the other indentations taken away from the composite resin layer for both groups (Table 6.2, Group 6 m). The experimental bonding agents of the group Res-BAG, Res-HOPC and ResHPCMM maintained high KHN values with no statistical difference after the ageing period in all the tested locations (p > 0.001). The mean surface microhardness values of the resin-dentine regions were 17.5 ± 0.8 at 24 h and 17.6 ± 0.3 after 6 months for Res-BAG; 18.8 ± 3.6 at 24 h and 24.2 ± 1.7 after 6 234

months for Res-HOPC; 17.5 ± 3 at 24 h and 13.1 ± 2.4 after 6 months for ResHPCMM, respectively. Likewise, all the other KHN values corresponding to 25, 55, 85 and 115 µm in depth did not show a statistical decrease in the mean hardness of the surface after 6 m of PBS storage (Table 6.2, Group 6 m).

235

Table 6.2 - The results of the micro-hardness measurements for each bonding system after 24 hours and 6 months of PBS storage. In each row, same capital letter indicate no statistical difference between mean KHN values (SD). Asterisk symbol (*) indicates a statistically significant reduction (p < 0.001) of mean KHN values (SD) subsequent to 6 m of PBS storage. I : Mean KHN values (SD) at the resin-dentine bonded interface. II : Mean KHN values (SD) at the points 25 μm distant from the composite resin layer in the direction of the dentine. III : Mean KHN values (SD) at the points 55 μm distant from the composite resin layer in the direction of the dentine. IV : Mean KHN values (SD) at the points 85 μm distant from the composite resin layer in the direction of the dentine. V : Mean KHN values (SD) at the points 115 μm distant from the composite resin layer in the direction of the dentine.

236

6.4 Discussion There are enduring challenges in adhesive dentistry due to the incomplete infiltration of wet dentine with resin monomers that yield resin-dentine bonds prone to degradation (Breschi et al., 2008). Major concerns have been expressed regarding interfacial ageing caused by absorption of water, hydrolysis of the resin and disruption of the collagen network (Spencer et al., 2010). At the same time, in vitro studies have suggested possible strategies to reduce the flaws inherent in the dentine bonding systems, each one having its own merits and limitations (Liu et al., 2011c). The development of bioactive ion-releasing restorative materials with a therapeutic ability to fill micro- and nano-sized voids by crystal deposition is currently one of the main targets of dental biomaterial research (Tay and Pashley, 2008, Liu et al., 2011b, Peters et al., 2010, Bresciani et al., 2010). In our previous work, the bonding durability and morphological changes induced on resin-infiltrated dentine were evaluated using four novel etch-and-rinse adhesives, either incorporating BAG (Chapter 4) or three distinct hydrated blends of calcium-silicate cements (Chapter 5), to gain perspective on their potential clinical use. In view of the mechanism of its formation, the resin-dentine interface represents a continuous structure from the hybrid layer (in which decalcified dentine impregnated by resin and that not impregnated by resin are considered to be mixed) to the healthy dentine (Nakazawa et al., 1999). Due to the complexity of these regions, we have resorted to the use of a indentation technique for appraising regional differences in micro-hardness associated with mineral

237

deposition over time (Van Meerbeek et al., 1993). The results of this study indicated that incorporation of the aforementioned bioactive fillers had an effect on the superficial hardness profile of water-rich, resin-sparse regions within the bonded interface after prolonged storage in PBS. However, it appeared that only treatment with Res-BAG, Res-HOPC and Res-HPCMM did not result in a statistically significant change in KHN with ageing. On the other hand, the reduction of resistance to local deformation in the hybrid resin-dentine zones created with the Res-Contr and Res-HPCTO bonding systems (p < 0.001) requires the null hypothesis of this study to be partially rejected. Various methods have been suggested for evaluating the extent of remineralisation in dental tissues, mostly based on the determination of changes in mineral content. Although several of these experimental methodologies are able to analyse mineral levels in detail [transverse microradiography (TMR), micro-computer tomography (CT), X-ray micro-tomography (XMT), etc.], comparable, objective relative dentine hardness measurements are often undertaken in vitro using micro-hardness indenters, resulting in reproducible data sets with minimal damage to the sample (Banerjee et al., 1999, Ogawa et al., 1983). Micro-hardness measurements can be correlated with mechanical properties such as modulus of elasticity, fracture resistance (Perinka et al., 1992), and yield strength (Currey and Brear, 1990, Mahoney et al., 2000). A positive correlation also exists between Knoop micro-hardness and bond strength and it was proposed that adhesion mechanisms for both enamel and dentine are controlled, to a major extent, by the mineral content of the tooth (Panighi and

238

G’Sell, 1993, Yoshiyama et al., 2002). In addition, Bresciani et al. (Bresciani et al., 2010) hypothesised that it may be possible to evaluate the therapeutic ability of bioactive calcium phosphatebased materials in repairing the demineralised intertubular and intrafibrillar dentine collagen by re-establishment of the superficial micro-hardness. In our study, cross-sectional Knoop micro-indentation measurements provided an average hardness of each surface and gave valuable additional information regarding the behaviour of dentine/restoration interfaces because any variation observed reflected a quantitative difference in mineral content (Hu and Featherstone, 2005, Owens and Miller, 2000, Iijima et al., 2012). However, many factors may influence the hardness results such as the dentine depth and the relative quantities of the tubular, peritubular or intertubular areas which vary considerably with location (Fuentes et al., 2003, Hosoya et al., 2000, Pashley et al., 1985). Consequently, Knoop test measurements were obtained from the same specimens and close to previously assessed locations after 6 m of PBS storage. This was done in order to minimise the effect of the structural variations within the same tooth, and to establish a reasonable baseline for evaluation. In bonding systems using phosphoric acid, the area 10 μm away from the interface in the direction of healthy dentine might be within the decalcified layer not impregnated by resin, resulting in lower hardness (Nakazawa et al., 1999). Overall mean KHN values calculated in sound, mineralised dentine adjacent to restorations (25, 55, 85 and 115 μm from the composite resin layer) were in agreement with those reported by other authors (Craig and Peyton, 1958, Fusayama et al., 1966, Meredith et al., 1996) .The similarity of reported values

239

is certainly due to the reproducible micro-indentation technique employed (Fuentes et al., 2003). For Knoop hardness, upon unloading, elastic recovery occurs mainly along the shortest diagonal, but the longest diagonal remains relatively unaffected (Shannon and Keuper, 1976, Marshall et al., 1982). Therefore, the hardness measurements obtained by this method are virtually insensitive to the elastic recovery of the material. Another chief characteristic of the Knoop hardness test is its sensitivity to surface effects and textures (Lysaght and DeBellis, 1969, Knoop et al., 1939). Accordingly, wider impressions of the tool mark were found on all the resindentine interfaces, these regions being more elastic and softer than the healthy dentine, and lower KHN values were obtained. Even these observations correlate with previous findings indicating that, in the absence of intrafibrillar mineralisation, the hardness and modulus of elasticity are inferior to those of mineralised dentine (Balooch et al., 2008). In fact, the modulus of elasticity of wet demineralised dentinal matrix is only about 5 MPa (Bedran-Russo et al., 2008) which is more than 1000 times lower than that of mineralised dentine. This different histological configuration is held accountable for hardness reduction in the resin-dentine interface and therefore for the little resistance offered to the testing indenter. Interestingly, specimens created with the experimental adhesives Res-BAG, Res-HOPC and Res-HPCMM did not restore micro-hardness to the level of sound dentine in these zones but maintained the same KHN values and no statistical difference reduction was found following 6 m of PBS storage. The only statistically significant change occurred in the resin-dentine interfaces

240

bonded with either Res-Contr or Res-HPCTO that were subjected to a reduction of KHN values with ageing. These results are in agreement with our previous findings (Chapter 4-5) where apatite precipitation and concomitant reduction of water-rich regions within the resin-dentine interface appeared to restrict the collagenolytic and hydrolytic mechanisms responsible for loss of mechanical stability. Remineralisation, defined as restoration of lost mineral content (Bresciani et al., 2010), should redevelop the mechanical properties to approximate those of original undemineralised dentine. Even so, the stiffness of resin-dentine interfaces cannot be compared with natural mineralised dentine, as the adhesive resins used to infiltrate the collagen matrices are much more viscoplastic than bioapatites. It is equally suggested that the lack of mechanical reconstitution may be attributed to an heterogeneous arrangement of the newly deposited mineral within the demineralised organic network. In fact, mineralisation patterns that differ from the usual organisation and orientation of minerals might lead to different mechanical properties of the resultant substrates (Bresciani et al., 2010, Bertassoni et al., 2011). Strengths of this study include the fact that every effort was made to standardise the experimental process. A thoughtfully planned pilot study was carried out to develop an effective experimental design. Nonetheless, the relationship between hardness and mineral content remains complex, not yet fully understood and necessitates further detailed investigation (Bresciani et al., 2010). Additional long-term studies and examination of the resin-dentine interface at narrower intervals (i.e. atomic force microscopy nanoindentation) will give us better insight into remineralisation dynamics, rate of

241

mineral uptake and hardness modifications induced by novel dentine bonding systems containing bioactive fillers.

6.5 Conclusions Within the limitations of this in vitro study the following conclusions were made: I.

In terms of micro-hardness, the hybrid layer of the resin-dentine interfaces created in the Res-BAG, Res-HOPC and Res-HPCMM groups resisted water degradation showing no statistical change in KHN values after 6 months of PBS ageing. Conversely, the use of Res-Contr and Res-HPCTO bonding systems showed a statistically significant microhardness drop in the hybrid layer after 6 months of PBS storage.

II.

Remineralisation of imperfect mineral-depleted sites where resin monomers border on decalcified dentine may be a potential means for preserving the micro-mechanical properties of resin-dentine bonds. This particular approach may be suitable for the contemporary concept of Conservative Dentistry where minimally invasive treatments are followed by therapeutic restorations which stabilise the carious lesion and/or create an optimal environment for the protection of the remaining dental hard tissues.

242

Chapter 7: General discussion and conclusion

243

7.1 Summary Over the past three decades, bonding of resin-based composite restorations has been revolutionised by continuing advances in dental adhesive technology. Improvements have been made in the areas of aesthetic appeal, ease of use and reduction of technique sensitivity. However, the process was, and still is, not without its faults, and chief among those is the reduced durability of resindentine bonds compared with resin-enamel bonds, owing to the fact that dentine bonding relies on organic components (Marshall et al., 1997). The infiltration of hydrophilic resin monomers into demineralised collagen matrix, to produce a hybrid layer that couples adhesives/resin composites to the underlying mineralised dentine, provides ample opportunity for nano-leakage to occur beneath the restoration: oral fluids penetrate any poorly infiltrated area and serve as a functional medium for esterases and collagenolytic enzymes that act synergistically to increase the biodegradation of both polymer matrices and exposed collagen (Spencer et al., 2010). While each experimental strategy that attempted to overcome these problems has its own benefits and reciprocal limitations (Liu et al., 2011c), progressive water replacement by apatite generated during dentine remineralisation may be a suitable strategy for extending the service life of resin-based dentine bonding procedures and its actualisation has been a source of conjecture until now (Liu et al., 2011a). In this case, nano-leakage might only be a temporary phenomenon that could be solved by new hard-tissue formation (Tay and Pashley, 2009). Water may provide the aqueous environment for the infiltration of amorphous calcium phosphate precursors into the gap zones of collagen fibrils to initiate nucleation and growth of newly formed mineral crystals. As remineralisation proceeds,

244

water then becomes less readily available for the functioning of collagen degrading matrix metalloproteinases (Sadek et al., 2010a). Moreover, these host-derived, collagen-bound proteases (Mazzoni et al., 2007) may be fossilised by the intrafibrillar apatite crystallites that are deposited within and on the collagen fibrils (Liu et al., 2011a). Previous research has concentrated on nanotechnology principles and the use of biomimetic analogs of matrix phosphoproteins to mimic what occurs in biomineralisation, particularly in areas devoid of seed crystallites (Zhang et al., 2012).

The work presented in this thesis set out to answer questions about whether the presence of designed reactive silicate mineral powders at the bonded interface could have a significant impact on the processes occurring in their vicinity, and the formation of hydroxycarbonate apatite deposits on hypomineralised adjacent dentine surfaces might take place in presence of phosphate buffered saline or simulated body fluid solutions. This hypothesis was based on the established fact that the experimental methacrylate-based adhesives employed in

this

study,

either

when

incorporating

calcium/sodium

phosphate-

phyllosilicates or calcium silicate cements, demonstrated to possess bioactive characteristics in an aqueous environment that contained calcium and phosphate ions (Section II - Experimental projects, Chapter 3).

At this stage, the first objective was to establish if the bioactive glass powder of the well-characterised 45S5 formulation and of an average particle size of < 10 μm could be included in the resin-to-dentine bonding process. When 0.05 g of Bioglass® 45S5 were applied onto H3PO4-etched wet dentine surfaces before

245

the bonding procedures (Chapter 4), a commercially available adhesive was able to form an apparently normal hybrid layer. Notably, after 6 months of storage in phosphate buffered saline, the occurrence and extent of nanometresized voids within the hybrid layers was reduced due to the chemical nature of the mineral precipitation (dicalcium-phosphate salts) within the resin-dentine interface. Conversely, severe water uptake was observed within the resinbonded dentine interfaces created with the control resin both after 24 h and six months of storage. However, a concomitant decrease in microtensile bond strength over time was also observed. On this account, the crystallised bondeddentine interfaces may have attained mechanical characteristics comparable to those created by conventional glass-ionomer cements applied onto polyacrylic acid-etched dentine when submitted to tensile tests (Yip et al., 2001), not reflecting their true adhesive strength. Following this promising start, it was decided to include 30 wt% of BioglassÂŽ 45S5 within the composition of a resin adhesive as bioactive micro-filler. This approach produced bioactive/protective effects not only in terms of nano-leakage reduction but also preserving high adhesive strength and joint integrity. Analysis of the failure modes displayed a pronounced tendency for these samples to fail cohesively, while the negative controls tended towards adhesive failure. Another important trend was confirmed in this study: the ultramorphology analysis of the fractured specimens demonstrated the formation of mineral crystals embedded within a wellpreserved collagen network over the course of the ageing period.

Recently, hydraulic calcium silicate cements have been investigated with respect to their potential use for the biomimetic remineralisation of apatite-

246

depleted

dentine

surfaces

and

prevent

the

demineralisation

hypomineralised/carious dentine (Gandolfi et al., 2011b). Thus, it remained to include even these materials in an adhesive system, and to see if the subsequent bonds to dentine would be of long duration or not. In chapter 5, small amounts of three derivatives and proprietary formulations based on the composition of an ordinary Portland cement were added to a resin representing a typical 3-step etch-and-rinse adhesive, and teeth were bonded for the purpose of carrying out bond strength tests, ultramorphology and nano-leakage studies. The most favourable results were obtained with the addition of set typeI Portland cement modified using either sodium-calcium-aluminum-magnesium silicate hydroxide or aluminum-magnesium-carbonate hydroxide hydrates. The use of resin bonding agents containing these two tailored micro-fillers promoted a therapeutic mineral deposition mechanism within the hybrid layer: nanoleakage did indeed appear to be reduced after soaking in simulated body fluid for 6 months, and consistent bond strength values were maintained when compared

negative

controls.

The

reduced

fluorescent-dye

uptake

observations, along with the strong calcium chelating fluorophore (xylenol orange) signal from the hybrid layer and the dentinal tubules, clearly indicated the remineralisation of those areas which were previously detected as mineraldeficient/porous zones. A further supporting result was found in the scanning electron microscopy: the ultramorphological analysis performed on fractured match-sticks demonstrated the presence of mineral crystals within the resindentine matrix subsequent to the storage period. Despite no specific biomimetic analogue agents being used in this study, there may have been a sort of biomimetic activity evoked by the main products of the two silicate-based fillers

247

hydration. This process caused an alkaline condition which may have interfered with the activity of metalloproteinases within the demineralised collagen matrix and may have played a supplementary role in the remineralisation kinetics enhancing the growth of apatite crystals (Somasundaran et al., 1985). Addition of a Portland cement-based micro-filler containing titanium oxide to the same adhesive was less successful. Again, it was shown that an apparently normal hybrid layer was able to form, and the strength of the bonds formed was comparably high following 24 h of storage in simulated body fluid. However, long-term evaluation of the micro-mechanical strength was less definitive than in the other two etch-and-rinse adhesives doped with calcium silicates. Although nano-leakage was substantially reduced, specimens of this experimental adhesive group showed diminution of bond strength durability due to the high hydrophilicity of titanium oxide.

The mechanical properties of resin-dentine bonds are a fundamental aspect of restorative procedures. The study in chapter 6 found that the same experimental bonding agents, which appeared to restrict the collagenolytic and hydrolytic mechanisms responsible for reduction of microtensile bond strength to dentine, were also able to maintain unchanged micro-hardness values within the resin-dentine interfaces submitted to prolonged storage (6 months) in phosphate buffered solution. These results suggested that a positive correlation exists between Knoop micro-hardness and bond strength and that adhesion mechanisms for both enamel and dentine may be controlled, to some extent, by the mineral content of the tooth.

248

7.2 Research contributions The current ethos in minimally invasive operative treatment requires the execution of therapeutic restorations that may combat the carious process and remineralise the dental hard tissues (Peters and McLean, 2001). The bioavailability and crystallographic inclusion of foreign ions is the basic requirement for apatite formation (biocatalysation) in presence of aqueous environments that contain calcium and phosphate (e.g. saliva). The concept of a therapeutic restoration may be satisfied only if innovative bioactive materials with the ability to release specific ions within the bonding interface are employed, so evoking a positive response from the biological environment and inducing protection and/or remineralisation of the mineral-depleted dental tissues (Qi et al., 2012).

This series of experiments provided preliminary evidence that the durability of the resin-dentine interfaces may be enhanced by modified bonding agents in a clinically relevant manner. The inclusion of highly reactive silicate compounds, such as commercially available bioactive glass and calcium-silicate Portlandderived cements, within the composition of a representative etch-and-rinse bonding system conferred on the hybrid adhesives attractive basic properties, such as: I.

light-curing ability with controlled solubility in water and oral fluids as detailed in Chapter 3.

II.

A tendency to tolerate humidity during placement and to interact with oral fluids and wet tooth structures (hydrophylic nature).

249

III.

High bioavailability and propensity to release various remineralising species; when activated with water, each micro-filler releases the predominant ions of its composition enhancing mineral ion delivery within and beneath the hybrid layers (mineral enrichment effect).

IV.

A potential to displace water from the resin-sparse regions of the hybrid layer with redeposition of apatitic tooth minerals (bioactivity).

Alkalinising activity to buffer the environmental acids as well as to interfere with the activity of matrix metalloproteinases, and antibacterial properties.

VI.

Providing nano-leakage reduction and lower rates of enzymatic degradation, two of the major causes of restoration failure, while producing no negative effects on bond strength over time (Chapter 4-5).

VII.

A capacity to preserve the mechanical properties and stability of resindentine bonded interfaces (Chapter 6).

The initial evaluation of bioactive silicate compounds as a possible addition to the resin-dentine bonding process has shown their value; a new generation of biologically active dental materials able to induce physicochemical reactions yielding apatite formation in demineralised dentine has been obtained as promising adhesives to be tested in future studies and clinical trials. These results support the possibility of prospective applications in clinical practice for direct resin-composite restorations, dentine hypersensitivity treatment and as cavity liners. This material may also have a potential use in apical root, root canal and root perforation treatments.

250

7.3 Recommendations for future research While the first steps toward the improvement of dentine bonding technology with regard to nano-leakage have been made, much work remains to be done. Continuing studies should be performed in order to optimise the appropriate solvent for the adhesive along with the amount of micro-filler to include, its particle size and composition. Particle size, or fineness, affects hydration rate: the smaller the particle size, the greater the surface-area-to-volume ratio, and thus, the more area available for water-cement interaction per unit volume. The addition of specific chemical compounds, such as zinc oxide, might increase the therapeutic/protective effects against the breakdown of collagen matrices mediated by matrix metalloproteinases within the aged resin-bonded dentine (Osorio et al., 2011). A further protective effect might as well be due to the release of Zn2+ ions exerting an antibacterial action within the bonded-dentine interface (Swetha et al., 2012). Reactions and bioactivity mechanisms of bioactive glasses depend on the glass composition. It was recently shown that fluoride-containing bioactive glasses are able to form fluorapatite, which is much less vulnerable to acid attack than is carbonated apatite (Lynch et al., 2012). In addition to the traditional bioactive glass components (silica, sodium, calcium, and phosphate), magnesium can also be added for its synergistic effects on the crystallinity and solubility of apatites: slowdown in the deposition of calcium phosphate, which is believed to lead to better-controlled (and better quality) mineralisation (Diba et al., 2012).

The relationship between hardness and mineral content necessitates further detailed investigation. Examination of the resin-dentine interface at narrower

251

intervals will give us better insight into remineralisation dynamics, rate of mineral uptake and hardness modifications induced by novel dentine bonding systems containing bioactive fillers. Probing biomaterial properties on a nanoscale requires comparatively low indentation loads and an ability to examine materials in a hydrated state. It has been advocated that atomic force microscopy nano-indentation performed in hydrated dentine may be a suitable method for the determination of the visco-elasticity of the demineralised dentine and its effective remineralisation (Bertassoni et al., 2009).

Similarly, confocal micro-Raman spectroscopy can be considered as a valuable method for future in vitro investigations of the dentine/adhesive interface. This technique would offer distinct advantages, including minimal sample preparation and both qualitative and quantitative analysis at ∼ 1 μm spatial resolution. This would enable spatially resolved chemical analysis of the interface areas between the modified adhesives and the underlying dentine, via direct examination of specimens without compromising their integrity. The nondestructive nature of this analysis would also allow investigation of the same specimen using complementary techniques. In combination with StreamLine™ imaging techniques, changes in the mineral composition of the hybrid layer associated with ageing of the specimens can be investigated.

Additionally, it would be important to learn the durability of these bonds over a greater time span; bond strengths and nano-leakage should be examined after storage of the samples for more than six months, or even years. It would also be useful to investigate interface-degradation patterns under more realistic

252

conditions using an in situ model. This method, considered as an intermediate stage between in vitro and in vivo studies, would permit to ageing the resinbonded interfaces in a relatively short period of time mimicking most of the challenging conditions adhesive restorations are submitted to in the oral environment (Reinke et al., 2012). Finally, clinical studies should be conducted to support laboratory data, in order to point out if modified adhesive systems may really provide restoration longevity.

253

Bibliography

254

ABOU NEEL, E. A., PALMER, G., KNOWLES, J. C., SALIH, V. & YOUNG, A. M. 2010. Chemical, modulus and cell attachment studies of reactive calcium phosphate filler-containing fast photo-curing, surface-degrading, polymeric bone adhesives. Acta Biomater, 6, 2695-703. ABREU, A., LOZA, M. A., ELIAS, A., MUKHOPADHYAY, S., LOONEY, S. & RUEGGEBERG, F. A. 2009. Tensile bond strength of an adhesive resin cement to different alloys having various surface treatments. J Prosthet Dent, 101, 10718. AGEE, K. A., BECKER, T. D., JOYCE, A. P., RUEGGEBERG, F. A., BORKE, J. L., WALLER, J. L., TAY, F. R. & PASHLEY, D. H. 2006. Net Expansion of dried demineralized dentin matrix produced by monomer/alcohol infiltration and solvent evaporation. J Biomed Mater Res A, 79, 349-58. AIDA, M., ODAKI, M., FUJITA, K., KITAGAWA, T., TESHIMA, I. & SUZUKI, K. 2009. Degradation-stage effect of self-etching primer on dentin bond durability. J Dent Res, 88, 443-8. AIMES, R. T. & QUIGLEY, J. P. 1995. Matrix metalloproteinase-2 is an interstitial collagenase. Inhibitor-free enzyme catalyzes the cleavage of collagen fibrils and soluble native type I collagen generating the specific 3/4- and 1/4-length fragments. J Biol Chem, 270, 5872-6. AKINMADE, A. O. & NICHOLSON, J. W. 1993. Glass-ionomer cements as adhesives. Part I. Fundamental aspects and their clinical relevance. J Mater Sci Mater Med, 4, 95-101. AL-AMMAR, A., DRUMMOND, J. L. & BEDRAN-RUSSO, A. K. 2009. The use of collagen cross-linking agents to enhance dentin bond strength. J Biomed Mater Res B Appl Biomater, 91, 419-24.

255

ALLEN, E. P., BAYNE, S. C., BECKER, I. M., DONOVAN, T. E., HUME, W. R. & KOIS, J. C. 1999. Annual review of selected dental literature: report of the Committee on Scientific Investigation of the American Academy of Restorative Dentistry. J Prosthet Dent, 82, 27-66. AMARAL, F. L., COLUCCI, V., PALMA-DIBB, R. G. & CORONA, S. A. 2007. Assessment of in vitro methods used to promote adhesive interface degradation: a critical review. J Esthet Restor Dent, 19, 340-53. ANDERSSON, O. & KANGASNIEMI, I. 1991. Calcium phosphate formation at the surface of bioactive glass in vitro. J Biomed Mater Res A, 25, 1019-30. ANDERSSON, O., KARLSSON, K. H., KANGASNIEMI, K. & YLI-URPO, A. 1988. Model for physical properties and bioacticity of phosphate opal glasses. Glastech Ber, 61, 300-5. ANUSAVICE, K. J., ZHANG, N. Z. & SHEN, C. 2006. Controlled release of chlorhexidine from UDMA-TEGDMA resin. J Dent Res, 85, 950-4. AOBA, T., MORENO, E. C. & SHIMODA, S. 1992. Competitive Adsorption of Magnesium and Calcium-Ions onto Synthetic and Biological Apatites. Calcif Tissue Int, 51, 143-50. ARANHA, A. M., GIRO, E. M., SOUZA, P. P., HEBLING, J. & DE SOUZA COSTA, C. A. 2006. Effect of curing regime on the cytotoxicity of resin-modified glassionomer lining cements applied to an odontoblast-cell line. Dent Mater, 22, 8649. ARENDS, J., RUBEN, J. L. & INABA, D. 1997. Major Topics in Quantitative Microradiography of Enamel and Dentin: R Parameter, Mineral Distribution Visualization, and Hyper-Remineralization. Adv Dent Res, 11, 403-14.

256

ASMUSSEN, E., HANSEN, E. K. & PEUTZFELDT, A. 1991. Influence of the solubility parameter of intermediary resin on the effectiveness of the gluma bonding system. J Dent Res, 70, 1290-3. ASMUSSEN, E. & UNO, S. 1993. Solubility parameters, fractional polarities, and bond strengths of some intermediary resins used in dentin bonding. J Dent Res, 72, 558-65. ATAI, M., L., S., NODEHI, A., MIRABEDINI, S. M., KASRAEI, S. & AKBARI, K. 2009. PMMA-grafted nanoclay as novel filler for dental adhesives. Dent Mater, 25, 339-47. ATTAL, J. P., ASMUSSEN, E. & DEGRANGE, M. 1994. Effects of surface treatment on the free energy of dentin. Dent Mater, 10, 259-64. ATTIN, T., VATASCHKI, M. & HELLWIG, E. 1996. Properties of resin-modified glass-ionomer restorative materials and two poly-acid-modified resin composite materials. Quintessence Int, 27, 203-9. AVCI, R., SCHWEITZER, M. H., BOYD, R. D., WITTMEYER, J. L., TERĂ N ARCE, F. & CALVO, J. O. 2005. Preservation of bone collagen from the late Cretaceous period studied by immunological techniques and atomic force microscopy. Langmuir, 21, 3584-90. AVILA, M. Y. & NAVIA, J. L. 2010. Effect of genipin collagen crosslinking on porcine corneas. J Cataract Refract Surg, 36, 659-64. BAIER, R. E. 1992. Principles of adhesion. Oper Dent Suppl, 5, 1-9. BAKRY, A. S., TAKAHASHI, H., OTSUKI, M., SADR, A., YAMASHITA, K. & TAGAMI, J. 2011a. CO2 laser improves 45S5 bioglass interaction with dentin. J Dent Res, 90, 246-50.

257

BAKRY, A. S., TAMURA, Y., OTSUKI, M., KASUGAI, S., OHYA, K. & TAGAMI, J. 2011b. Cytotoxicity of 45S5 bioglass paste used for dentine hypersensitivity treatment. J Dent, 39, 599-603. BALOOCH, M., HABELITZ, S., KINNEY, J. H., MARSHALL, S. J. & MARSHALL, G. W. 2008. Mechanical properties of mineralized collagen fibrils as influenced by demineralization. J Struct Biol, 162, 404-10. BANERJEE, A., SHERRIFF, M., KIDD, E. A. & WATSON, T. F. 1999. A confocal microscopic study relating the autofluorescence of carious dentine to its microhardness. Br Dent J, 187, 206-10. BARBOSA, S. V., SAFAVI, K. E. & SPÅNGBERG, S. W. 1994. Influence of sodium hypochlorite on the permeability and structure of cervical human dentine. Int Endod J, 27, 309-12. BARKMEIER, W. W. & COOLEY, R. L. 1992. Laboratory evaluation of adhesive systems. Oper Dent, 5, 50-61. BARRÉRE, F., LAYROLLE, P., VAN BLITTERSWIJK, C. A. & DE GROOT, K. 1999. Biomimetic calcium phosphate coatings on Ti6AI4V: a crystal growth study of octacalcium phosphate and inhibition by Mg2+ and HCO3-. Bone, 25, 107S-11S. BARRÈRE, F., VAN BLITTERSWIJK, C. A. & DE GROOT, K. 2006. Bone regeneration: molecular and cellular interactions with calcium phosphate ceramics. Int J Nanomedicine, 1, 317-32. BAUER, T. W. & SMITH, S. T. 2002. Bioactive materials in orthopaedic surgery: overview and regulatory considerations. Clin Orthop Relat Res, 395, 11-22.

258

BAYLE, M. A., GRÃ&#x2030;GOIRE, G. & SHARROCK, P. 2007. The role of acrylophosphonic acid monomers in the formation of hybrid layers based on self-etch adhesives. J Dent, 35, 302-8. BEDRAN-DE-CASTRO, A. K., PEREIRA, P. N., PIMENTA, L. A. & THOMPSON, J. Y. 2004. Effect of thermal and mechanical load cycling on microtensile bond strength of a total-etch adhesive system. Oper Dent, 29, 150-6. BEDRAN-RUSSO, A. K., PASHLEY, D. H., AGEE, K., DRUMMOND, J. L. & MIESCKE, K. J. 2008. Changes in stiffness of demineralized dentin following application of collagen crosslinkers. J Biomed Mater Res B Appl Biomater, 86, 330-4. BEDRAN-RUSSO, A. K., PEREIRA, P. N., DUARTE, W. R., DRUMMOND, J. L. & YAMAUCHI, M. 2007. Application of crosslinkers to dentin collagen enhances the ultimate tensile strength. J Biomed Mater Res B Appl Biomater, 80, 268-72. BEDRAN-RUSSO, A. K., VIDAL, C. M., DOS SANTOS, P. H. & CASTELLAN, C. S. 2010. Long-term effect of carbodiimide on dentin matrix and resin-dentin bonds. J Biomed Mater Res B Appl Biomater, 94, 250-5. BEDRAN-RUSSO, A. K., YOO, K. J., EMA, K. C. & PASHLEY, D. H. 2009. Mechanical properties of tannic-acid-treated dentin matrix. J Dent Res, 88, 80711. BELL, R. B. & BARKMEIER, W. W. 1994. Glass-ionomer restoratives and liners: shear bond strength to dentin. J Esthet Dent, 6, 129-34. BERRY, E. A. & POWERS, J. M. 1994. Bond strength of glass ionomers to coronal and radicular dentin. Oper Dent, 19, 122-6.

259

BERTACCHINI, S. M., ABATE, P. F., BLANK, A., BAGLIETO, M. F. & MACCHI, R. L. 1999. Solubility and fluoride release in ionomers and compomers. Quintessence Int, 30, 193-7. BERTASSONI, L. E., HABELITZ, S., KINNEY, J. H., MARSHALL, S. J. & MARSHALL, G. W. 2009. Biomechanical perspective on the remineralization of dentin. Caries Res, 43, 70-7. BERTASSONI, L. E., HABELITZ, S., MARSHALL, S. J. & MARSHALL, G. W. 2011. Mechanical recovery of dentin following remineralization in vitro--an indentation study. J Biomech, 44, 176-81. BHATTACHARYYA, K. G. & GUPTA, S. S. 2008. Adsorption of a few heavy metals on natural and modified kaolinite and montmorillonite: a review. Adv Colloid Interface Sci, 140, 114-31. BLACKBURN, R. S., HARVEY, A., KETTLE, L. L., MANIAN, A. P., PAYNE, J. D. & RUSSELL, S. J. 2007. Sorption of chlorhexidine on cellulose: mechanism of binding and molecular recognition. J Phys Chem B, 111, 8775-84. BONDING, N., GRAEM, N., RYGAARD, J. & DABELSTEEN, E. 1987. Effects of irradiation with dental light curing units on Langerhans cells in human stratified epithelium in heterotransplanted skin. Scan J Dent Res, 95, 463-6. BOSKEY, A. L. 2007. Mineralization of Bones and Teeth. Elements Magazine, 3, 38591. BOUILLAGUET, S., CIUCCHI, B., JACOBY, T., WATAHA, J. C. & PASHLEY, D. 2001. Bonding characteristics to dentin walls of class II cavities, in vitro. Dent Mater, 17, 316-21.

260

BOUILLAGUET, S., WATAHA, J. C., HANKS, C. T., CIUCCHI, B. & HOLZ, J. 1996. In vitro cytotoxicity and dentin permeability of HEMA. J Endod, 22, 2448. BOUSHELL, L. W., KAKU, M., MOCHIDA, Y., BAGNELL, R. & YAMAUCHI, M. 2008. Immunohistochemical localization of matrixmetalloproteinase-2 in human coronal dentin. Arch Oral Biol, 53, 109-16. BOWEN, R. L. 1963. Properties of a silica-reinforced polymer for dental restorations. J Am Dent Assoc, 66, 57-64. BOWEN, R. L., BENNETT, P. S., GROH, R. J., FARAHANI, M. & EICHMILLER, F. C. 1996. New surface-active comonomer for adhesive bonding. J Dent Res, 75, 606-10. BOYDE, A. 1974. Transmission electron microscopy of ion beam thinned dentine. Cell Tissue Res, 152, 543-50. BRACKETT, W. W., COVEY, D. A. & ST-GERMAIN, H. A. 2002. One-year clinical performance of a self-etching adhesive in class V resin composites cured by two methods. Oper Dent, 27, 218-22. BRACKETT, W. W., ITO, S., TAY, F. R., HAISCH, L. D. & PASHLEY, D. H. 2005. Microtensile dentin bond strength of self-etching resins: Effect of a hydrophobic layer. Oper Dent, 30, 734-9. BRAEM, M. 2007. Microshear fatigue testing of tooth/adhesive interfaces. J Adhes Dent, 9, 249-53. BRANNSTRÖM, M. & ASTRÖM, A. 1972. The hydrodynamics of the dentine; its possible relationship to dentinal pain. Int Dent J, 22, 219-27. BRÄNNSTRÖM, M., JOHNSON, G. & NORDENVALL, K. J. 1981. Resin impregnation of sensitive dentin surfaces. Odontoiatr Prat, 16, 5-18.

261

BRAUER, D. S., KARPUKHINA, N., O’DONNELL, M. D., LAW, R. V. & HILL, R. G. 2010. Fluoride-containing bioactive glasses: effect of glass design and structure on degradation, pH and apatite formation in simulated body fluid. Acta Biomater, 6, 3275-82. BRAZEL, C. S. & PEPPAS, N. A. 1999. Dimensionless analysis of swelling of hydrophilic glassy polymers with subsequent drug release from relaxing structures. Biomaterials, 20, 721-32. BRESCHI, L., CAMMELLI, F., VISINTINI, E., MAZZONI, A., VITA, F., CARRILHO, M., CADENARO, M., FOULGER, S., MAZZOTI, G., TAY, F., DI LENARDA, R. & PASHLEY, D. 2009. Influence of chlorhexidine concentration on the durability of etch-and-rinse dentin bonds: a 12-month in vitro study. J Adhes Dent, 11, 191-8. BRESCHI, L., MARTIN, P., MAZZONI, A., NATO, F., CARRILHO, M., TJÄDERHANE, L., VISINTINI, E., CADENARO, M., TAY, F. R., DE STEFANO DORIGO, E. & PASHLEY, D. H. 2010a. Use of a specific MMPinhibitor (galardin) for preservation of hybrid layer. Dent Mater, 26, 571-8. BRESCHI, L., MAZZONI, A., NATO, F., CARRILHO, M., VISINTINI, E. T., L., RUGGERI, A. J. R., TAY, F. R., S., D. & PASHLEY, D. H. 2010b. Chlorhexidine stabilizes the adhesive interface: a 2-year in vitro study. Dent Mater, 26, 320-5. BRESCHI, L., MAZZONI, A., RUGGERI, A., CADENARO, M., DI LENARDA, R. & DORIGO, E. 2008. Dental adhesion review: aging and stability of the bonded interface. Dent Mater, 24, 90-101.

262

BRESCIANI, E., WAGNER, W. C., NAVARRO, M. F., DICKENS, S. H. & PETERS, M. C. 2010. In vivo dentin microhardness beneath calcium-phosphate cement. J Dent Res, 89, 836-41. BUNKER, B. C., TALLANT, D. R. & HEADLEY, T. J. 1988. The structure of leached sodium borosilicate glass. Phys Chem Class, 29, 106-20. BUONOCORE, M. G. 1955. A simple method of increasing the adhesion of acrylic filling materials to enamel surfaces. J Dent Res, 34, 849-53. BUONOCORE, M. G. & QUIGLEY, M. 1958. Bonding of synthetic resin material to human dentin: preliminary histological study of the bond area. J Am Dent Assoc, 57, 807-11. BURKE, F. J. T., HUSSAIN, A., NOLAN, L. & FLEMING, G. J. P. 2008. Methods used in dentine bonding tests: an analysis of 102 investigations on bond strength. Eur J Prosthodont Rest Dent, 16, 158-65. BUSENLEHNER, L. S. & ARMSTRONG, R. N. 2005. Insights into enzyme structure and dynamics elucidated by amide H/D exchange mass spectrometry. Arch Biochem Biophys, 433, 34-46. CADENARO, M., ANTONIOLLI, F., CODAN, B., AGEE, K., TAY, F. R., DORIGO, E., PASHLEY, D. H. & BRESCHI, L. 2010. Influence of different initiators on the degree of conversion of experimental adhesive blends in relation to their hydrophilicity and solvent content. Dent Mater, 26, 288-94. CADENARO, M., BRESCHI, L., RUEGGEBERG, F. A., AGEE, K., DI LENARDA, R., CARRILHO, M., TAY, F. R. & PASHLEY, D. H. 2009a. Effect of adhesive hydrophilicity and curing time on the permeability of resins bonded to water vs. ethanol-saturated acid-etched dentin. Dent Mater, 25, 39-47.

263

CADENARO, M., PASHLEY, D. H., MARCHESI, G., CARRILHO, M., ANTONIOLLI, F., MAZZONI, A., TAY, F. R., DI LENARDA, R. & BRESCHI, L. 2009b. Influence of chlorhexidine on the degree of conversion and E-modulus of experimental adhesive blends. Dent Mater, 25, 1269-74. CALERO, P., JORGE-HERRERO, E., TURNAY, J., OLMO, N., LÓPEZ DE SILANES, I., LIZARBE, M. A., MAESTRO, M. M., ARENAZ, B. & CASTILLO-OLIVARES, J. L. 2002. Gelatinases in soft tissue biomaterials. Analysis of different crosslinking agents. Biomaterials. Biomaterials, 23, 34738. CAMILLERI, J. 2008. Characterization of hydration products of mineral trioxide aggregate. Int Endod J, 41, 408-17. CARDOSO, M. V., COUTINHO, E., ERMIS, R. B., POITEVIN, A., VAN LANDUYT, K. & DE MUNCK, J. 2008. Influence of dentin cavity surface finishing on micro-tensile bond strength of adhesives. Dent Mater, 24, 492-501. CARRILHO, M. R., CARVALHO, R. M., DE GOES, M. F., DI HIPÓLITO, V., GERALDELI, S., TAY, F. R., PASHLEY, D. H. & TJÄDERHANE, L. 2007a. Chlorhexidine preserves dentin bond in vitro. J Dent Res, 86, 90-4. CARRILHO, M. R., CARVALHO, R. M., SOUSA, E. N., NICOLAU, J., BRESCHI, L., MAZZONI, A., TJÄDERHANE, L., TAY, F. R., AGEE, K. & PASHLEY, D. H. 2010. Substantivity of chlorhexidine to human dentin. Dent Mater, 26, 779-85. CARRILHO, M. R., CARVALHO, R. M., TAY, F. R., YIU, C. & PASHLEY, D. H. 2005a. Durability of resin-dentin bonds related to water and oil storage. Am J Dent, 18, 315-9.

264

CARRILHO, M. R., GERALDELI, S., TAY, F., DE GOES, M. F., CARVALHO, R. M., TJÃ&#x201E;DERHANE, L., REIS, A. F., HEBLING, J., MAZZONI, A., BRESCHI, L. & PASHLEY, D. H. 2007b. In vivo preservation of the hybrid layer by chlorhexidine. J Dent Res, 86, 529-33. CARRILHO, M. R. O., F.R., T., PASHLEY, D. H., TJADERHANE, L. & CARVALHO, R. M. 2005b. Mechanical degradation of resin dentin bond components. Dent Mater, 21, 232-41. CARVALHO, R. M., CHERSONI, S., FRANKENBERGER, R., PASHLEY, D. H., PRATI, C. & TAY, F. R. 2005. A challenge to the conventional wisdom that simultaneous etching and resin infiltration always occurs in self-etch adhesives. Biomaterials, 26, 1035-42. CARVALHO, R. M., SANO, H., HOMER, J. A., BREWER, P. D. & PASHLEY, D. H. 1996. Comparison of interfacial morphology and strength of bonds made to superficial and deep dentin. Am J Dent, 8, 297-302. CASTELLAN, C. S., PEREIRA, P. N., GRANDE, R. H. & BEDRAN-RUSSO, A. K. 2010a.

Mechanical

characterization

proanthocyanidin-dentin

matrix

interaction. Dent Mater, 26, 968-73. CASTELLAN, C. S., PEREIRA, P. N., VIANA, G., CHEN, S. N., PAULI, G. F. & BEDRAN-RUSSO, A. K. 2010b. Solubility study of phytochemical crosslinking agents on dentin stiffness. J Dent, 38, 431-6. CATTANI-LORENTE, M. A., DUPUIS, V., MOYA, F., PAYAN, J. & MEYER, J. M. 1999. Comparative study of the physical properties of a polyacid-modified composite resin and a resin-modified glass ionomer cement. Dent Mater, 15, 2132.

265

CHADWICK, R. G., TRAYNOR, N., MOSELEY, H. & GIBBS, N. 1994. Blue light curing units â&#x20AC;&#x201C; a dermatological hazard. Br Dent J, 176, 17-31. CHARULATHA, V. & RAJARAM, A. 2003. Influence of different crosslinking treatments on the physical properties of collagen membranes. Biomaterials, 24, 759-67. CHAUSSAIN-MILLER, C., FIORETTI, F., GOLDBERG, M. & MENASHI, S. 2006. The role of matrix metalloproteinases (MMPs) in human caries. J Dent Res, 85, 22-32. CHENG, H., CATERSON, B., NEAME, P. J., LESTER, G. E. & YAMAUCHI, M. 1996. Differential distribution of lumican and fibromodulin in tooth cementum. Connect Tissue Res, 34, 87-96. CHERSONI, S., SUPPA, P., GRANDINI, S., GORACCI, C., MONTICELLI, F., YIU, C., HUANG, C., PRATI, C., BRESCHI, L., FERRARI, M., PASHLEY, D. H. & TAY, F. R. 2004. In vivo and in vitro permeability of one-step self-etch adhesives. J Dent Res, 83, 459-64. CHESNICK, I. E., MASON, J. T., GIUSEPPETTI, A. A., EIDELMAN, N. & POTTER, K. 2008. Magnetic resonance microscopy of collagen mineralization. Biophys J, 95, 2017-26. CHIARAPUTT, S., MAI, S., HUFFMAN, B. P., KAPUR, R., AGEE, K. A., YIU, C. K., CHAN, D. C., HARNIRATTISAI, C., AROLA, D. D., RUEGGEBERG, F. A., PASHLEY, D. H. & TAY, F. R. 2008. Changes in resin-infiltrated dentin stiffness after water storage. J Dent Res, 87, 655-60. CHIGIRA, H., YUKITANI, W. & HASEGAWA, T. 1994. Self-etching dentin primers containing Phenyl-P. J Dent Res, 73, 1088-95.

266

CLARKSON, B. H., FEAGIN, F. F., MCCURDY, S. P., SHEETZ, J. H. & SPEIRS, R. 1991. Effects of phosphoprotein moieties on the remineralization of human root caries. Caries Res, 25, 166-73. COLEMAN, N. J., AWOSANYA, K. & NICHOLSON, J. W. 2009. Aspects of the in vitro bioactivity of hydraulic calcium (alumino)silicate cement. J Biomed Mater Res A, 90, 166-74. COUTINHO, E., VAN LANDUYT, K., DE MUNCK, J., POITEVIN, A., YOSHIDA, Y. & INOUE, S. 2006. Development of a self-etch adhesive for resin-modified glass ionomers. J Dent Res, 85, 349-53. CRAIG, R. G. & PEYTON, F. A. 1958. The micro-hardness of enamel and dentin. J Dent Res, 37, 661-8. CROLL, T. P., KILLIAN, C. M. & HELPIN, M. L. 1993. A restorative dentistry renaissance for children: light-hardened glass ionomer/resin cement. ASDC J Dent Child, 60, 89-94. CURREY, J. D. & BREAR, K. 1990. Young’s modulus and yield strength in mammalian mineralized tissues. J Mater Sci Mater Med, 1, 14-20. CURTIS, A. R., WEST, N. X. & SU, B. 2010. Synthesis of nanobioglass and formation of apatite

rods to

occlude

exposed dentine tubules

and

eliminate

hypersensitivity. Acta Biomater, 6, 3740-6. CURTIS, R. V. & WATSON, T. F. 2008. Dental biomaterials: imaging, testing and modeling. Abington Hall, England: Woodhead Publishing Ltd. CUSSLER, E. L. 1976. “Multicomponent Diffusion”. Elsevier Scientific Pub., Amsterdam, Netherlands. DARVELL, B. W. & WU, R. C. 2011. "MTA"-an hydraulic silicate cement: review update and setting reaction. Dent Mater, 27, 407-22.

267

DE ALMEIDA, J. B., PLATT, J. A., OSHIDA, Y., MOORE, B. K., COCHRAN, M. A. & ECKERT, G. J. 2003. Three different methods to evaluate microleakage of packable composites in Class II restorations. Oper Dent, 28, 453-60. DE GEE, A. J., LELOUP, G., WERNER, A., VREVEN, J. & DAVIDSON, C. L. 1998. Structural integrity of resin-modified glass ionomers as affected by the delay or omission of light activation. J Dent Res, 77, 1658-63. DE GROOT, K. 1983. Ceramics of calcium phosphates: preparation and properties. Bioceramics of calcium phosphate (de Groot K. ed.) CRC Press Inc, 100-11. DE MUNCK, J., ERNIS, R. B., KOSHIRO, K., INOUE, S., IKEDA, T., SANO, H., VAN LANDUYT, K. L. & VAN MEERBEEK, B. 2007. NaOCl degradation of a HEMA-free all-in-one adhesive bonded to enamel and dentin following two air-blowing techniques. J Dent, 35, 74-83. DE MUNCK, J., MINE, A., VAN DEN STEEN, P. E., VAN LANDUYT, K. L., POITEVIN, A. & OPDENAKKER, G. 2010. Enzymatic degradation of adhesive-dentin interfaces produced by mild self-etch adhesives. Eur J Oral Sci, 118, 494-501. DE MUNCK, J., SHIRAI, K., YOSHIDA, Y., INOUE, S., VAN LANDUYT, K., LAMBRECHTS, P., SUZUKI, K., SHINTANI, H. & VAN MEERBEEK, B. 2006. Effect of water storage on the bonding effectiveness of 6 adhesives to Class I cavity dentin. Oper Dent, 31, 456-65. DE MUNCK, J., VAN DEN STEEN, P. E., MINE, A., VAN LANDUYT, K. L., POITEVIN, A., OPDENAKKER, G. & VAN MEERBEEK, B. 2009. Inhibition of enzymatic degradation of adhesive-dentin interfaces. J Dent Res, 88, 1101-6.

268

DE MUNCK, J., VAN LANDUYT, K., PEUMANS, M., POITEVIN, A., LAMBRECHTS, P. & BRAEM, M. 2005. A critical review of the durability of adhesion to tooth tissue: methods and results. J Dent Res, 84, 118-32. DE MUNCK, J., VAN MEERBEEK, B., YOSHIDA, Y., INOUE, S., VARGAS, M., SUZUKI, K., LAMBRECHTS, P. & VANHERLE, G. 2003. Four-year water degradation of total-etch adhesives bonded to dentin. J Dent Res, 82, 136-40. DESHPANDE, A. S. & BENIASH, E. 2008. Bio-inspired Synthesis of Mineralized Collagen Fibrils. Cryst Growth Des, 8, 3084-90. DHANPAL, P., YIU, C. K., KING, N. M., TAY, F. R. & HIRAISHI, N. 2009. Effect of temperature on water sorption and solubility of dental adhesive resins. J Dent, 37, 122-32. DIBA, M., TAPIA, F., BOCCACCINI, A. R. & STROBEL, L. A. 2012. Magnesiumcontaining bioactive glasses for biomedical applications. IJAGS, 2, 221-53. DICKENS, S. H. & FLAIM, G. M. 2008. Effect of a bonding agent on in vitro biochemical activities of remineralizing resin-based calcium phosphate cements. Dent Mater, 24, 1273-80. DICKENS, S. H., FLAIM, G. M. & TAKAGI, S. 2003. Mechanical properties and biochemical activity of remineralizing resin-based Ca-PO4 cements. Dent Mater, 19, 558-66. DICKINSON, D. P. 2002. Cysteine peptidases of mammals: their biological roles and potential effects in the oral cavity and other tissues in health and disease. Crit Rev Oral Biol Med, 13, 238-75. DOMINGO, C., ARCIS, R. W., OSORIO, E., OSORIO, R., FANOVICH, M. A. & RODRIGUEZ-CLEMENTE, R. 2003. Hydrolytic stability of experimental hydroxyapatite-filled dental composite materials. Dent Mater, 19, 478-86.

269

DONMEZ, N., BELLI, S., PASHLEY, D. H. & TAY, F. R. 2005. Ultrastructural correlates of in vivo/in vitro bond degradation in self-etch adhesives. J Dent Res, 84, 355-9. DRUMMOND, J. L., SAKAGUCHI, R. L., RACEAN, D. C., WOZNY, J. & STEINBERG, A. D. 1996. Testing mode and surface treatment effects on dentin bonding. J Biomed Mater Res, 32, 533-41. EANES, E. D., GILLESSEN, I. H. & POSNER, A. S. 1965. Intermediate states in the precipitation of hydroxyapatite. Nature, 208, 365-7. EDDLESTON, C., HINDLE, A., RUEGGEBERG, F. A., AGEE, K., CARVALHO, R., TAY, F. R. & PASHLEY, D. H. 2003. Dimensional changes in aciddemineralized matrices following the use of HEMA-water vs. HEMA-alcohol primers. J Biomed Mater Res A, 67, 267-82. EFFLANDT, S. E., MAGNE, P., DOUGLAS, W. H. & FRANCIS, L. F. 2002. Interaction between bioactive glasses and human dentin. J Mater Sci Mater Med, 13, 557-65. EICK, J. D., COBB, C. M., CHAPELL, R. P., SPENCER, P. & ROBINSON, S. J. 1991. The dentinal surface: its influence on dentinal adhesion. Part I. Quintessence Int, 22, 967-77. EICK, J. D., GWINNETT, A. J., PASHLEY, D. H. & ROBINSON, S. J. 1997. Current Concepts On Adhesion To Dentin. Crit Rev Oral Biol Med, 8, 306-35. EICK, J. D., MILLER, R. G., ROBINSON, S. I., BOWLES, C. O., GUTSHALL, P. L. & CHAPPELOW, C. C. 1996. Quantitative analysis of the dentin adhesive interface by Auger spectroscopy. J Dent Res, 75, 1027-33.

270

ELIADES, G., VOUGIOUKLAKIS, G. & PALAGHIAS, G. 2001. Heterogeneous distribution of single-bottle adhesive monomers in the resin-dentin interdiffusion zone. Dent Mater, 17, 277-83. ENDO, K., HASHIMOTO, M., HARAGUCHI, K. & OHNO, H. 2010. Crystal growth by restorative filling materials. Eur J Oral Sci, 118, 489-93. ENGQVIST, H., SCHULTZ-WALZ, J. E., LOOF, J., BOTTON, G. A., MAYER, D. & PHANEUF, M. W. 2004. Chemical and biological integration of a mouldable bioactive ceramic material capable of forming apatite in vivo in teeth. Biomaterials, 25, 2781-7. ERICKSON, R. L. 1989. Mechanism and clinical implication of bond formation for two dentin bonding agents. Am J Dent, 2, 117-23. ERICKSON, R. L., BARKMEIER, W. W. & KIMMES, N. S. 2009. Fatigue of enamel bonds with self-etch adhesives. Dent Mater, 25, 716-20. ERMIS, R. B., DE MUNCK, J., CARDOSO, M. V., COUTINHO, E., VAN LANDUYT, K. L. & POITEVIN, A. 2008. Bond strength of self-etch adhesives to dentin prepared with three different diamond burs. Dent Mater, 24, 978-85. FABRE, H. S., FABRE, S., CEFALY, D. F., DE OLIVEIRA CARRILHO, M. R., GARCIA, F. C. & WANG, L. 2007. Water sorption and solubility of dentin bonding agents light-cured with different light sources. J Dent, 35, 253-8. FEATHERSTONE, J. D. 1990. An updated understanding of the mechanism of dental decay and its prevention. Nutr Q, 14, 5-11. FEATHERSTONE, J. D. 1996. Modeling the caries-inhibitory effects of dental materials. Dent Mater, 12, 194-7.

271

FEDARKO, N. S., JAIN, A., KARADAG, A. & FISHER, L. W. 2004. Three small integrin binding ligand N-linked glycoproteins (SIBLINGs) bind and activate specific matrix metalloproteinases. FASEB J, 18, 734-6. FENG, L., LI, S., LI, Y., LI, H., ZHANG, L., ZHAI, J., SONG, Y., LIU, B., JIANG, L. & ZHU, D. 2002. Super-Hydrophobic Surfaces: From Natural to Artificial. Adv Mater, 14, 1857-60. FERRACANE, J. L. 1994. Review Elution of leachable components from composites. J Oral Rehabil, 21, 441-52. FERRACANE, J. L. 2006. Hygroscopic and hydrolytic effects in dental polymer networks. Dent Mater, 22, 211-22. FILGUEIRAS, M. R., LATORRE, G. P. & HENCH, L. L. 1993. Solution effects on the surface reactions of a bioactive glass. J Biomed Mater Res, 27, 445-53. FINER, Y., JAFFER, F. & SANTERRE, J. P. 2004. Mutual influence of cholesterol esterase and pseudocholinesterase on the biodegradation of dental composites. Biomaterials, 25, 1787-93. FINGER, W. J., INOUE, M. & ASMUSSEN, E. 1994. Effect of wettability of adhesive resins on bonding to dentin. Am J Dent, 7, 35-8. FRANKENBERGER, R., KRÃ&#x201E;MER, N. & PETSCHELT, A. 1999. Fatigue behaviour of different dentin adhesives. Clin Oral Investig, 3, 11-7. FRANKENBERGER, R., PASHLEY, D. H., REICH, S. M., LOHBAUER, U., PETSCHELT, A. & TAY, F. R. 2005. Characterisation of resin-dentine interfaces by compressive cyclic loading. Biomaterials, 26, 2043-52. FROST, H. M. 1963. Measurement of human bone formation by means of tetracycline labelling. Can J Biochem Physiol, 41, 31.

272

FRUJIKAWA, S., YOKOTA, T., KONGA, K. & KUMADA, J. 1987. The continuous hydrolysis of geniposide to genipin using immobilized betaglucosidase on calcium alginate gel. J Biotechnol Lett, 9, 697-702. FU, X., J.L., K., BERGT, C., KASSIM, S. Y., HUQ, N. P., D'AVIGNON, A., PARKS, W. C., MECHAM, R. P. & HEINECKE, J. W. 2004. Oxidative cross-linking of tryptophan to glycine restrains matrix metalloproteinase activity: specific structural motifs control protein oxidation. J Biol Chem, 279, 6209-12 FUENTES, V., TOLEDANO, M., OSORIO, R. & CARVALHO, R. M. 2003. Microhardness of superficial and deep sound human dentin. J Biomed Mater Res A, 66A, 850-3. FUKUOKA, A., KOSHIRO, K., INOUE, S., YOSHIDA, Y., TANAKA, T., IKEDA, T., SUZUKI, K., SANO, H. & MEERBEEK, B. V. 2011. Hydrolytic stability of one-step self-etching adhesives bonded to dentin. J Adhes Dent, 13, 243-8. FUNG, D. T., WANG, V. M., LAUDIER, D. M., SHINE, J. H., BASTA-PLJAKIC, J., JEPSEN, K. J., SCHAFFLER, M. B. & FLATOW, E. L. 2009. Subrupture tendon fatigue damage. J Orthop Res, 27, 264-73. FUSAYAMA, T. 1980. Concepts in Operative Dentistry. Tokio: Quintessence. FUSAYAMA, T., OKUSE, K. & HOSODA, H. 1966. Relationship between hardness, discoloration, and microbial invasion in carious dentin. J Dent Res, 15, 1033-46. GAJJERAMAN, S., HE, G., NARAYANAN, K. & GEORGE, A. 2008. Biological assemblies provide novel templates for the synthesis of hierarchical structures and facilitate cell adhesion. Adv Funct Mater, 18, 3972-80. GANDOLFI, M. G., TADDEI, P., SIBONI, F., MODENA, E., CIAPETTI, G. & PRATI, C. 2011a. Development of the foremost light-curable calcium-silicate

273

MTA cement as root-end in oral surgery. Chemical-physical properties, bioactivity and biological behavior. Dent Mater, 27, e134-57. GANDOLFI, M. G., TADDEI, P., SIBONI, F., MODENA, E., DE STEFANO, E. D. & PRATI, C. 2011b. Biomimetic remineralization of human dentin using promising innovative calcium-silicate hybrid "smart" materials. Dent Mater, 27, 1055-69. GANDOLFI, M. G., TADDEI, P., TINTI, A., DORIGO DE STEFANO, E., ROSSI, P. L. & PRATI, C. 2010a. Kinetics of apatite formation on a calcium-silicate cement for root-end filling during ageing in physiological-like phosphate solutions. Clin Oral Investig, 14, 659-68. GANDOLFI, M. G., TADDEI, P., TINTI, A. & PRATI, C. 2010b. Apatite-forming ability of ProRoot MTA. Int Endod J, 43, 917-29. GANDOLFI, M. G., VAN LANDUYT, K., TADDEI, P., MODENA, E., VAN MEERBEEK, B. & PRATI, C. 2010c. Environmental scanning electron microscopy connected with energy dispersive x-ray analysis and Raman techniques to study ProRoot mineral trioxide aggregate and calcium silicate cements in wet conditions and in real time. J Endod, 36, 851-7. GARBEROGLIO, R. & BRÄNNSTRÖM, M. 1976. Scanning electron microscopic investigation of human dentinal tubules. Arch Oral Biol, 21, 355-62. GARCIA-GODOY, F., KRAMER, N., FEILZER, A. J. & FRANKENBERGER, R. 2010. Longterm degradation of enamel and dentin bonds: 6-year results in vitro vs. in vivo. Dent Mater, 26, 1113-8. GARCÍA-GODOY, F., TAY, F. R., PASHLEY, D. H., FEILZER, A., TJÄDERHANE, L. & PASHLEY, E. L. 2007. Degradation of resin-bonded human dentin after 3 years of storage. Am J Dent, 20, 109-13.

274

GARNERO, P., BOREL, O., BYRJALSEN, I., FERRERAS, M., DRAKE, F. H. & MCQUENEY, M. S. 1998. The collagenolytic activity of cathepsin K is unique among mammalian proteinases. J Biol Chem, 273, 32347-52. GATTI, A., VALDRE, G. & ANDERSSON, O. 1994. Analysis of the in vivo reactions of a bioactive glass in soft and hard tissue. Biomaterials, 15, 208-12. GENDRON, R., GRENIER, D., SORSA, T. & MAYRAND, D. 1999. Inhibition of the activities of matrix metalloproteinases 2, 8, and 9 by chlorhexidine. Clin Diagn Lab Immunol, 6, 437-9. GEORGE, A. & VEIS, A. 2008. Phosphorylated proteins and control over apatite nucleation, crystal growth, and inhibition. Chem Rev, 108, 4670-93. GERICKE, A., QIN, C., SUN, Y., REDFERN, R., REDFERN, D., FUJIMOTO, Y., TALEB, H., BUTLER, W. T. & BOSKEY, A. L. 2004. Different forms of DMP1 play distinct roles in mineralization. J Dent Res, 89, 355-9. GERZINA, T. M. & HUME, W. R. 1996. Diffusion of monomers from bonding resinresin composite combinations through dentine in vitro. J Dent, 24, 125-8. GEURTSEN, W. 1998. Substances released from dental resin composites and glass ionomer cements. Eur J Oral Sci, 106, 687-95. GEURTSEN, W. 2000. Biocompatibility of resin-modified filling materials. Crit Rev Oral Biol Med, 11, 333-55. GIANNINI, M., SOARES, C. J. & DE CARVALHO, R. M. 2004. Ultimate tensile strength of tooth structures. Dent Mater, 20, 322-9. GIOIA, M., MONACO, S., FASCIGLIONE, G. F., COLETTI, A., MODESTI, A., MARINI, S. & COLETTA, M. 2007. Characterization of the mechanisms by which

gelatinase

neutrophil

275

collagenase,

and

membrane-type

metalloproteinase MMP-14 recognize collagen I and enzymatically process the two alpha-chains. J Mol Biol, 368, 1101-13. GIRIJA, E. K., YOKOGAWA, Y. & NAGATA, F. 2004. Apatite formation on collagen fibrils in the presence of polyacrylic acid. J Mater Sci Mater Med, 15, 593-9. GLIMCHER, M. J. 1987. The nature of the mineral component of bone and the mechanism of calcification. Instr Course Lect, 36, 49-69. GLIMCHER,

J.,

KOSSIVA,

BRICKLEY-PARSONS,

1984.

Phosphoproteins of chicken bone matrix. Proof of synthesis in bone tissue. J Biol Chem, 259, 290-3. GOMBIA, M., BORTOLOTTI, V., DE CARLO, B., MONGIORGI, R., ZANNA, S. & FANTAZZINI, J. 2010. Nanopore structure buildup during endodontic cement hydration studied by time-domain nuclear magnetic resonance of lower and higher mobility 1H. J Phys Chem B, 114, 1767-74. GONZALEZ-CABEZAS, C. 2010. The chemistry of caries: remineralization and demineralization events with direct clinical relevance. Dent Clin North Am, 54, 469-78. GORACCI, C., TAVARES, A. U., FABIANELLI, A., MONTICELLI, F., RAFFAELLI, O. & CARDOSO, P. C. 2004. The adhesion between fiber posts and root canal walls: comparison between microtensile and push-out bond strength measurements. Eur J Oral Sci, 112, 353-61. GOWER, L. B. 2008. Biomimetic model systems for investigating the amorphous precursor pathway and its role in biomineralization. Chem Rev, 108, 4551-627. GRANDE, R. H., REIS, A., LOGUERCIO, A. D., SINGER JDA, M., SHELLARD, E. & NETO, P. C. 2005. Adhesive systems used for sealing contaminated surfaces: a microleakage evaluation. Braz Oral Res, 19, 17-22.

276

GREENSPAN, D. C. 2010. NovaMin and tooth sensitivity--an overview. J Clin Dent, 21, 61-5. GRIFFITHS, B. M. & WATSON, T. F. 1995. Resinâ&#x20AC;&#x201C;dentin interface of scotchbond multi-purpose dentin adhesive. Am J Dent, 8, 212-6. GRIFFITHS, B. M., WATSON, T. F. & SHERRIFF, M. 1998. The influence of dentine bonding systems and their handling characteristics on the morphology and micropermeability of dentine adhesive systems. J Dent, 27, 63-71. GRIFFITHS, B. M., WATSON, T. F. & SHERRIFF, M. 1999. Variable polymerisation shrinkage and the interfacial micropermeability of a dentin bonding system. J Adhes Dent, 2, 119-31. GU, L., KIM, Y. K., LIU, Y., TAKAHASHI, K., ARUN, S. & WIMMER, C. E. 2011. Immobilization of a phosphonated analog of matrix phosphoproteins within cross-linked collagen as a templating mechanism for biomimetic mineralization. Acta Biomater, 7, 268-77. GU, L. S., HUFFMAN, B. P., AROLA, D. D., KIM, Y. K., MAI, S. & ELSALANTY, M. E. 2010. Changes in stiffness of resin-infiltrated demineralized dentin after remineralization by a bottom-up biomimetic approach. Acta Biomater, 6, 145361 GYSI, A. 1900. An attempt to explain the sensitiveness of dentin. Brit J Dent Sci, 43, 865-8. HAGGAR, O. 1951. Sevitron cavity seal. British patent 687 299. HAMID, A., OKAMOTO, A., IWAKU, M. & HUME, W. R. 1998. Component release from light-activated glass ionomer and compomer cements. J Oral Rehabil, 25, 94-9.

277

HANNAS, A. R., PEREIRA, J. C., GRANJEIRO, J. M. & TJÄDERHANE, L. 2007. The role of matrix metalloproteinases in the oral environment. Acta Odontol Scand, 65, 1-13. HASHIMOTO, M., FUJITA, S., NAGANO, F., OHNO, H. & ENDO, K. 2010a. Tenyear degradation of resin-dentin bonds. Eur J Oral Sci, 118, 404-10. HASHIMOTO, M., IIJIMA, M., NAGANO, F., OHNO, H. & ENDO, K. 2010b. Effect of monomer composition on crystal growth by resin containing bioglass. J Biomed Mater Res B Appl Biomater, 94, 127-33. HASHIMOTO, M., NAKAMURA, K., KAGA, M. & YAWAKA, Y. 2008. Crystal growth by fluoridated adhesive resins. Dent Mater, 24, 457-63. HASHIMOTO, M., OHNO, H., SANO, H., KAGA, M. & OGUCHI, H. 2003. In vitro degradation of resin-dentin bonds analysed by microtensile bond test, scanning and transmission electron microscopy. Biomaterials, 24, 3795-803. HASHIMOTO, M., OHNO, H., SANO, H., TAY, F. R., KAGA, M., KUDOU, Y., OGUCHI, H., ARAKI, Y. & KUBOTA, M. 2002. Micromorphological changes in resin-dentin bonds after 1 year of water storage. J Biomed Mater Res, 63, 30611. HAYAKAWA, T., KIKUTAKE-SUGIYAMA, K. & NEMOTO, K. 2005. Efficacy of water-soluble photoinitiator on the adhesion of composite resin to bovine teeth in all-in-one bonding system. Dent Mater J, 24, 213-8. HE, G. & GEORGE, A. 2004. Dentin matrix protein 1 immobilized on type I collagen fibrils facilitates apatite deposition in vitro. J Biol Chem, 279, 11649-56. HEBLING, J., PASHLEY, D. H., TJÄDERHANE, L. & TAY, F. R. 2005. Chlorhexidine arrests subclinical degradation of dentin hybrid layers in vivo. J Dent Res, 84, 741-6.

278

HEIKKILA, J. T., AHO, H. J., YLI-URPO, A., ANDERSSON, O. H., AHO, A. J. & HAPPONEN, R. P. 1993. Bioactive glass is as good as hydroxylapatite in reconstruction of experimental osteochondral defects. Acta Orthop Scand, 64, 672-8. HEIKKILA, J. T., AHO, H. J., YLI-URPO, A., HAPPONEN, R. P. & AHO, A. J. 1995. Bone formation in rabbit cancellous bone defects filled with bioactive glass granules. Acta Orthop Scand, 66, 463-7. HENCH, L.L. 1997. Biomaterials: a forecast for the future. European Materials Research Society 1997 meeting, symposium G - Biomaterials - perspectives for research and industry at the century change 1419-23. HENCH, L.L. 1999. Bioactive glasses and glass-ceramics. Shackelford JF, Ed. Bioceramics: applications of ceramic and glass materials in medicine. Enfield, NH: Trans Tech Publications, 37-64. HENCH, L.L. 2006. The story of Bioglass. J Mater Sci Mater Med, 17, 967-78. HENCH, L.L. & ANDERSSON, O. 1993. Bioactive glasses. Advanced Series in Ceramics - Vol. 1. An introduction to Bioceramics. World Scientific Singapore Hench LL, Wilson J eds, 45-7. HENCH, L.L. & PASCHALL, H. 1974. Histological responses at a biomaterial's interface. J Biomed Mater Res, 8, 49-64. HENCH, L.L. & PASCHALL, H. A. 1973. Direct chemical bond of bioactive glassceramic materials to bone and muscle. J Biomed Mater Res, 7, 25-42. HENCH, L.L., SPLINTER, R. J., ALLEN, W. C. & GREENLEE, T. K. 1971. Bonding Mechanisms at the Interface of Ceramic Prosthetic Materials. J Biomed Mater Res, 5, 117-41.

279

HEWLETT, E. R., CAPUTO, A. A. & WROBET, D. C. 1991. Glass ionomer bond strength and treatment of dentin with polyacrylic acid. J Prosthet Dent, 66, 76772. HIKITA, K., VAN MEERBEEK, B., DE MUNCK, J., IKEDA, T., VAN LANDUYT, K. & MAIDA, T. 2007. Bonding effectiveness of adhesive luting agents to enamel and dentin. Dent Mater, 23, 71-80. HIRAISHI, N., NISHIYAMA, N., IKEMURA, K., YAU, J. Y., KING, N. M., TAGAMI, J., PASHLEY, D. H. & TAY, F. R. 2005. Water concentration in self-etching primers affects their aggressiveness and bonding efficacy to dentin. J Dent Res, 84, 653-8. HIRAISHI, N., YIU, C. K., KING, N. M., TAY, F. R. & PASHLEY, D. H. 2008. Chlorhexidine release and water sorption characteristics of chlorhexidineincorporated hydrophobic/hydrophilic resins. Dent Mater, 24, 1391-9. HOSAKA, K., NISHITANI, Y., TAGAMI, J., YOSHIYAMA, M., BRACKETT, W. W., AGEE, K. A., TAY, F. R. & PASHLEY, D. H. 2009. Durability of resindentin bonds to water- vs. ethanol-saturated dentin. J Dent Res, 88, 146-51. HOSOYA, Y., MARSHALL, S. J., WATANABE, L. G. & MARSHALL, G. W. 2000. Microhardness of carious deciduous dentin. Oper Dent, 25, 81-9. HOULE, M. A., GRENIER, D., PLAMONDON, P. & NAKAYAMA, K. 2003. The collagenase activity of Porphyromonas gingivalis is due to Arg-gingipain. FEMS Microbiol Lett, 221, 181-5. HU, W. & FEATHERSTONE, J. D. 2005. Prevention of enamel demineralization: an in-vitro study using light-cured filled sealant. Am J Orthod Dentofacial Orthop, 128, 592-600.

280

HUAN, Z., CHANG, J. & HUANG, X. H. 2008. Self-setting properties and in vitro bioactivity of Ca2SiO4/CaSO4.1/2H2O composite bone cement. J Biomed Mater Res B Appl Biomater, 87, 387-94. HUME, W. R. & GERZINA, T. M. 1996. Review Bioavailability of components of resin-based materials which are applied to teeth. Crit Rev Oral Biol Med, 7, 1729. IIJIMA, M., HASHIMOTO, M., NAKAGAKI, S., MUGURUMA, T., KOHDA, N., ENDO, K. & MIZOGUCHI, I. 2012. Bracket bond strength and cariostatic potential of an experimental resin adhesive system containing Portland cement. Angle Orthod, 82, 900-6. IKEMURA, K., ICHIZAWA, K., FUCHIGAMI, K., ITO, S. & ENDO, T. 2009. Design of a new dental adhesive--effect of a water-soluble sodium acylphosphine oxide with crown ether on adhesion to dental hard tissues. Dent Mater J, 28, 267-76. IKEMURA, K., TAY, F. R., KOURO, Y., ENDO, T., YOSHIYAMA, M., MIYAI, K. & PASHLEY, D. H. 2003. Optimizing filler content in an adhesive system containing pre-reacted glass-ionomer fillers. Dent Mater, 19, 137-46. IMAZATO, S., TAY, F. R., KANESHIRO, A. V., TAKAHASHI, Y. & EBISU, S. 2007. An in vivo evaluation of bonding ability of comprehensive antibacterial adhesive system incorporating MDPB. Dent Mater, 23, 170-6. INOUE, S., VARGAS, M. A., ABE, Y., YOSHIDA, Y., LAMBRECHTS, P. V., G. , SANO, H. & VAN MEERBEEK, B. 2001. Microtensile bond strength of eleven contemporary adhesives to dentin. J Adhes Dent, 3, 237-45. ISHIGURO, K., YAMASHITA, K., NAKAGAKI, H., IWATA, K. & HAYAKAWA, T. 1994. Identification of tissue inhibitor of metalloproteinases-1 (TIMP-1) in

281

human teeth and its distribution in cementum and dentine. Arch Oral Biol, 39, 345-9. ISO Standard 4049: 2000 (E). Dentistry-polymer-based filling, restorative and luting materials. Geneva: International Organization for Standardization; 2000, 18-21. ISO/TS 11405: 2003. Dental materials - testing of adhesion to tooth structure. Geneva: International Organization for Standardization/Technical Specification; 2003. ITO, S., HASHIMOTO, M., WADGAONKAR, B., SVIZERO, N., CARVALHO, R. M., YIU, C., RUEGGEBERG, F. A., FOULGER, S., SAITO, T., NISHITANI, Y., YOSHIYAMA, M., TAY, F. R. & PASHLEY, D. H. 2005. Effects of resin hydrophilicity on water sorption and changes in modulus of elasticity. Biomaterials, 26, 6449-59. ITO, S., IIJIMA, M., MOTAI, F., MIZOGUCHI, I. & SAITO, T. 2012. Effects of calcium salts of acidic monomers on mineral induction of phosphoprotein immobilized to agarose beads. J Biomed Mater Res A, 100, 2760-5. JACOBSEN, T. & SODERHOLM, K. J. 1995. Some effects of water on dentin bonding. Dent Mater, 11, 132-6. JAYARAMAN, M. & SUBRAMANIAN, M. V. 2002. Preparation and characterization of two new composites: collagen-brushite and collagen octa-calcium phosphate. Med Sci Monit, 8, 481-7. JIANG, H. & LIU, X. Y. 2004. Principles of mimicking and engineering the selforganized structure of hard tissues. J Biol Chem, 297, 41286-93. KAN, K. C., MESSER, L. B. & MESSER, H. H. 1997. Variability in cytotoxicity and fluoride release of resin-modified glass-ionomer cements. J Dent Res, 76, 15027.

282

KANCA, J. R. 1992. Resin bonding to wet substrate. 1. Bonding to dentin. Quintessence Int, 23, 39-41. KARUPPUCHAMY, S. & JEONG, J. M. 2005. Super-hydrophilic amorphous titanium dioxide thin film deposited by cathodic electrodeposition. Mater Chem Phys, 93, 251-4. KEMAL, E., ADESANYA, K. O. & DEB, S. 2011. Phosphate based 2-hydroxyethyl methacrylate hydrogels for biomedical applications. J Mater Chem, 21, 2237-45. KERBY, R. E., KNOBLOCH, L. & THAKUR, A. 1997. Strength properties of visiblelight-cured resin-modified glass-ionomer cements. Oper Dent, 22, 79-83. KERN, M., BARLOI, A. & YANG, B. 2009. Surface conditioning influences zirconia ceramic bonding. J Dent Res, 88, 817-22. KHOR, E. 1997. Methods for the treatment of collagenous tissues for bioprostheses. Biomaterials, 18, 95-105. KIM, C. Y., CLARK, A. E. & HENCH, L. L. 1989. Early stages of calcium-phosphate layer formation in bioglasses. J Non-Cryst Solids, 113, 195-202. KIM, H. M., HIMENO, T., KOKUBO, T. & NAKAMURA, T. 2005. Process and kinetics of bonelike apatite formation on sintered hydroxyapatite in a simulated body fluid. Biomaterials, 26, 4366-73. KIM, H. M., REY, C. & GLIMCHER, M. J. 1995. Isolation of calcium-phosphate crystals of bone by nonaqueous methods at low-temperature. J Bone Miner Res, 10, 1589-601. KIM, J., AROLA, D. D., GU, L., KIM, Y. K., MAI, S., LIU, Y., PASHLEY, D. H. & TAY, F. R. 2010a. Functional biomimetic analogs help remineralize apatitedepleted demineralized resin-infiltrated dentin via a bottom-up approach. Acta Biomater, 6, 2740-50.

283

KIM, J., UCHIYAMA, T., CARRILHO, M., AGEE, K. A., MAZZONI, A., BRESCHI, L., CARVALHO, R. M., TJÄDERHANE, L., LOONEY, S., WIMMER, C., TEZVERGIL-MUTLUAY, A., TAY, F. R. & PASHLEY, D. H. 2010b. Chlorhexidine binding to mineralized versus demineralized dentin powder. Dent Mater, 26, 771-8. KIM, J., VAUGHN, R. M., GU, L., ROCKMAN, R. A., AROLA, D. D., SCHAFER, T. E., CHOI, K. K., PASHLEY, D. H. & TAY, F. R. 2010c. Imperfect hybrid layers created by an aggressive one-step, self-etch adhesive in primary dentin are amendable to biomimetic remineralization in vitro. J Biomed Mater Res A, 93, 1225-34. KIM, J. R., GU, L., BRESCHI, L., TJÄDERHANE, L., CHOI, K. K., PASHLEY, D. H. & TAY, F. R. 2010d. Implication of ethanol wet bonding in hybrid layer remineralization. J Dent Res, 89, 575-80. KIM, Y. K., GU, L. S., BRYAN, T. E., KIM, J. R., CHEN, L. & LIU, Y. 2010e. Mineralization

reconstituted

collagen

using

polyvinylphosphonic

acid/polyacrylic acid templating matrix protein analogues in the presence of calcium, phosphate and hydroxyl ions. Biomaterials, 31, 6618-27. KIM, Y. K., MAI, S., MAZZONI, A., LIU, Y., TEZVERGIL-MUTLUAY, A., TAKAHASHI, K., ZHANG, K., PASHLEY, D. H. & TAY, F. R. 2010f. Biomimetic remineralization as a progressive dehydration mechanism of collagen matrices—Implications in the aging of resin-dentin bonds. Acta Biomater, 6, 3729-39. KIM, Y. K., YIU, C. K. Y., KIM, J. R., GU, L., KIM, S. K. & WELLER, R. N. 2010g. Failure of a glass ionomer to remineralize apatite-depleted dentin. J Dent Res, 89, 230-5.

284

KING, N. M., TAY, F. R., PASHLEY, D. H., HASHIMOTO, M., ITO, S., BRACKETT, W. W., GARCIA-GODOY, F. & SUNICO, M. 2005. Conversion of one-step to two-step self-etch adhesives for improved efficacy and extended application. Am J Dent, 18, 126-34. KINNEY, J. H., HABELITZ, S., MARSHALL, S. J. & MARSHALL, G. W. 2003. The importance of intrafibrillar mineralization of collagen on the mechanical properties of dentin. J Dent Res, 82, 957-61. KINNEY, J. H., POPLE, J. A., MARSHALL, G. W. & MARSHALL, S. J. 2001. Collagen orientation and crystallite size in human dentin: a small angle X-ray scattering study. Calcif Tissue Int, 69, 31-7. KLONT, B. & TEN CATE, J. M. 1991. Susceptibility of the collagenous matrix from bovine incisor roots to proteolysis after in vitro lesion formation. Caries Res, 25, 46-50. KNOOP, F., PETERS, C. G. & EMERSOON, W. B. 1939. A sensitive pyramidaldiamond tool for indentation measurements. J Res Nat Bur Stand, 23, 39. KOIBUCHI, H., YASUDA, N. & NAKABAYASHI, N. 2001. Bonding to dentin with a self-etching primer: the effect of smear layers. Dent Mater, 17, 122-6. KOKUBO, N., KIM, H. M. & KAWASHIMA, M. 2003. Novel bioactive materials with different mechanical properties. Biomaterials, 24, 2161-75. KOKUBO, T. 1990. Surface chemistry of bioactive glass-ceramics. J Non-Cryst Solids, 120, 138-51. KOKUBO, T. 1991. Bioactive glass ceramics: properties and applications. Biomaterials, 12, 155-63.

285

KOKUBO, T. 1992. Bioactivity of glasses and glass ceramics. Ducheyne P, Kokubo T, van Blitterswijk C, editors. Bone bonding biomaterials. Leiderdrop: Reed Healthcare Communications, 31-46. KOKUBO, T. & TAKADAMA, H. 2006. How useful is SBF in predicting in vivo bone bioactivity? Biomaterials, 27, 2907-15. KOSHIRO, K., INOUE, S., TANAKA, T., KOASE, K., FUJITA, M., HASHIMOTO, M. & SANO, H. 2004. In vivo degradation of resin-dentin bonds produced by a self-etch vs. a total-etch adhesive system. Eur J Oral Sci, 112, 368-75. KOSTORYZ, E. L., DHARMALA, K., YE, Q., WANG, Y., HUBER, J. & PARK, J. G. 2009. Enzymatic biodegradation of HEMA/bisGMA adhesives formulated with different water content. J Biomed Mater Res B Appl Biomater, 88, 394-401. KOUTSOUKOS, P. G. & NANCOLLAS, G. H. 1981. Crystal growth of calcium phosphatesâ&#x20AC;&#x201D;epitaxial considerations. J Cryst Growth, 53, 10-9. KRAMER, I. R. H. & MC LEAN, J. W. 1952. Alterations in the staining reactions of dentine resulting from a constituent of a new self-polymerizing resin. Br Dent J, 92, 150-3. KREMER, E. A., CHEN, Y., SUZUKI, K., NAGASE, H. & GORSKI, J. P. 1998. Hydroxyapatite induces autolytic degradation and inactivation of matrix metalloproteinase-1 and -3. J Bone Miner Res, 13, 1890-902. KUBOKI, Y. & MECHANIC, G. L. 1982. Comparative molecular distribution of crosslink in bone and dentin collagen. Structure-function relationships. Calcif Tissue Int, 34, 306-8. KUSBOKI, Y., OHGUSHI, K. & FUSAYAMA, T. 1977. Collagen biochemistry of the two layers of carious dentin. J Dent Res, 56, 1233-7.

286

LABELLA, R., BRADEN, M. & DEB, S. 1994. Novel hydroxyapatite-based dental composites. Biomaterials, 15, 1197-200. LAGERWEIJ, M. D. & TEN CATE, J. M. 2002. Remineralisation of enamel lesions with daily applications of a high-concentration fluoride gel and a fluoridated toothpaste: an in situ study. Caries Res, 36, 270-4. LANDIS, W. J. 1996. Mineral characterization in calcifying tissues: atomic, molecular and macromolecular perspectives. Connect Tissue Res, 34, 239-46. LANDIS, W. J., SONG, M. J., LEITH, A., MCEWEN, L. & MC-EWEN, B. F. 1993. Mineral and organic matrix interaction in normally calcifying tendon visualized in three dimensions by high-voltage electron microscopic tomography and graphic image reconstruction. J Struct Biol, 110, 39-54. LAUER-FIELDS, J. L., CHALMERS, M. J., BUSBY, S. A., MINOND, D., GRIFFIN, P. R. & FIELDS, G. B. 2009. Identification of specific hemopexin-like domain residues that facilitate matrix metalloproteinase collagenolytic activity. J Biol Chem, 284, 24017-24. LAWRENCE, C. D. 1998. Physicochemical and mechanical properties of Portland cements. LEA’s Chemistry of Cement and Concrete, P. C.Hewlett, Ed.,Woburn,Mass, USA, Butterworth & Heinemann, 4th edition, 343-420. LEE, S. Y., GREENER, E. H., MUELLER, H. J. & CHIU, C. H. 1994. Effect of food and oral simulating fluids on dentine bond and composite strength. J Dent, 22, 352-9. LEE, T. C., MOHSIN, S., TAYLOR, D., PARKESH, R., GUNNLAUGSSON, T., O'BRIEN, F. J., GIEHL, M. & GOWIN, W. 2003. Detecting microdamage in bone. J Anat, 203, 161-72.

287

LEES, S. & PAGE, E. A. 1992. A study of some properties of mineralized turkey leg tendon. Connect Tissue Res, 28, 263-87. LEEVAILOJ, C., PLATT, J. A., COCHRAN, M. A. & MOORE, B. K. 1998. In vitro study of fracture incidence and compressive fracture load of all-ceramic crowns cemented with resin-modified glass ionomer and other luting agents. J Prosthet Dent, 80, 699-707. LEGEROS, R. Z. 1991. Calcium phosphates in oral biology and medicine. Myers HM, ed. Monographs in Oral Science. Basel: Karger. LEGEROS, R. Z., TRAUTZ, O. R., KLEIN, E. & LEGEROS, J. P. 1969. 2 Types of carbonate substitution in apatite structure. Experientia, 25, 5-7. LI, H., BURROW, M. F. & TYAS, M. J. 2002. The effect of load cycling on the nanoleakage of dentin bonding systems. Dent Mater, 18, 111-9. LI, J., DOU, Y., YANG, J., YIN, Y., ZHANG, H. & YAO, F. 2009. Surface characterization

and

biocompatibility

micro-

and

nano-

hydroxyapatite/chitosanâ&#x20AC;&#x201C;gelatin network films. Mater Sci Eng C, 29, 1207-15. LI, X. & CHANG, J. 2008. Preparation of bone-like apatite-collagen nanocomposites by a biomimetic process with phosphorylated collagen. J Biomed Mater Res A, 85, 293-300. LI, X. & WU, J. F. 2010. Recent developments in patent anti-cancer agents targeting the matrix metalloproteinases (MMPs). Recent Pat Anticancer Drug Discov, 5, 10941. LIAO, J., YANG, L., GRASHOW, J. & SACKS, M. S. 2005. Molecular orientation of collagen in intact planar connective tissues under biaxial stretch. Acta Biomater, 1, 45-54.

288

LIN, B. A., JAFFER, F., DUFF, M. D., TANG, Y. W. & SANTERRE, J. P. 2005. Identifying enzyme activities within human saliva which are relevant to dental resin composite biodegradation. Biomaterials, 26, 4259-64. LIU, G., ZHAO, D., TOMSIA, A. P., MINOR, A. M., SONG, X. & SAIZ, E. 2009. Three-dimensional biomimetic mineralization of dense hydrogel templates. J Am Chem Soc, 29, 9937-9. LIU, Y., LI, N., QI, Y., NIU, L. N., ELSHAFIY, S., MAO, J., BRESCHI, L., PASHLEY, D. H. & TAY, F. R. 2011a. The use of sodium trimetaphosphate as a biomimetic analog of matrix phosphoproteins for remineralization of artificial caries-like dentin. Dent Mater, 27, 465-77. LIU, Y., MAI, S., LI, N., YIU, C. K. Y., MAO, J., PASHLEY, D. H. & TAY, F. R. 2011b. Differences between top-down and bottom-up approaches in mineralizing thick, partially demineralized collagen scaffolds. Acta Biomater, 7, 1742-51. LIU, Y., TJÃ&#x201E;DERHANE, L., BRESCHI, L., MAZZONI, A., LI, N., MAO, J., PASHLEY, D. H. & TAY, F. R. 2011c. Limitations in bonding to dentin and experimental strategies to prevent bond degradation. J Dent Res, 90, 953-68. LODOVICI, E., REIS, A., GERALDELI, S., FERRACANE, J. L., BALLESTER, R. Y. & RODRIGUES FILHO, L. E. 2009. Does adhesive thickness affect resindentin bond strength after thermal/load cycling? Oper Dent, 34, 58-64. LOWENSTAM, H. A. & WEINER, S. 1980. On biomineralization New York, Oxford University Press. LU, X. & LENG, Y. 2005. Theoretical analysis of calcium phosphate precipitation in simulated body fluid. Biomaterials, 26, 1097-108.

289

LUSVARDI, G., MALAVASI, G., MENABUE, L., AINA, V. & MORTERRA, C. 2009. Fluoride-containing bioactive glasses: surface reactivity in simulated body fluids solutions. Acta Biomater, 5, 3548-62. LUTHRA, A. K. & SANDHU, S. S. 2005. Methods and clinical devices for the inhibition or prevention of mammalian cell growth. United States Patent Application: 6,929,818. LYNCH, E., BRAUER, D. S., KARPUKHINA, N., GILLAM, D. G. & HILL, R. G. 2012. Multi-component bioactive glasses of varying fluoride content for treating dentin hypersensitivity. Dent Mater, 28, 168-78. LYSAGHT, V. E. & DEBELLIS, A. 1969. Microhardness testing. Wilkes B, editor. Hardness testing handbook. New York: American Chain and Cable Co., 76-105. MA, D. H., LAI, J. Y., CHENG, H. Y., TSAI, C. C. & YEH, L. K. 2010. Carbodiimide cross-linked amniotic membranes for cultivation of limbal epithelial cells. Biomaterials, 31, 6647-58. MACEDO, G. V., YAMAUCHI, M. & BEDRAN-RUSSO, A. K. 2009. Effects of chemical cross-linkers on caries-affected dentin bonding. J Dent Res, 88, 1096100. MACIEL, K. T., CARVALHO, R. M., RINGLE, R. D., PRESTON, C. D., RUSSELL, C. M. & PASHLEY, D. H. 1996. The effects of acetone, ethanol, HEMA, and air on the stiffness of human decalcified dentin matrix. J Dent Res, 75, 1851-8. MAHAMID, J., SHARIR, A., ADDADI, L. & WEINER, S. 2008. Amorphous calcium phosphate is a major component of the forming fin bones of zebrafish: Indications for an amorphous precursor phase. Proc Natl Acad Sci USA, 105, 12748-53.

290

MAHONEY, E., HOLT, A., SWAIN, M. & KILPATRICK, N. 2000. The hardness and modulus of elasticity of primary molar teeth: An ultra-microindentation study. J Dent Res, 28, 589-94. MAI, S., GU, L. & LING, J. Q. 2009. Current methods for preventing degradation of resin-dentin bonds. Hong Kong Dent J, 6, 83-92. MAI, S., KIM, Y. K., KIM, J., YIU, C. K., LING, J., PASHLEY, D. H. & TAY, F. R. 2010. In vitro remineralization of severely compromised bonded dentin. J Dent Res, 89, 405-10. MALACARNE, J., CARVALHO, R. M., DE GOES, M. F., SVIZERO, N., PASHLEY, D. H., TAY, F. R., YIU, C. K. & CARRILHO, M. R. 2006. Water sorption/solubility of dental adhesive resins. Dent Mater, 22, 973-80. MALACHOVÁ, K., PRAUS, P., PAVLÍČKOVÁ, Z. & TURICOVÁ, M. 2009. Activity of antibacterial compounds immobilised on montmorillonite. Applied Clay Science, 43, 364-8. MALLA, N., S., S., WINBERG, J. O., HADLER-OLSEN, E. & UHLIN-HANSEN, L. 2008. Biological and pathobiological functions of gelatinase dimers and complexes. Connect Tissue Res, 49, 180-4. MANDEL, S. & TAS, A. C. 2010. Brushite (CaHPO4×2H2O) to Octacalcium Phosphate (Ca8(HPO4)2(PO4)4×5H2O) Transformation in DMEM (Dulbecco’s Modified Eagle Medium) Solutions at 36.5°C. Mater Sci Eng C, 30, 245-54. MANHART, J. & TRUMM, C. 2010. Marginal adaptation of an etch-and-rinse adhesive with a new type of solvent in class II cavities after artificial aging. Clin Oral Investig, 14, 699-705. MARCHESI, G., NAVARRA, C. O., CADENARO, M., CARRILHO, M. R., CODAN, B., SERGO, V., DI LENARDA, R. & BRESCHI, L. 2010. The effect of ageing

291

on the elastic modulus and degree of conversion of two multistep adhesive systems. Eur J Oral Sci, 3, 304-10. MARGONAR, R., QUEIROZ, T. P., LUVIZUTO, E. R., MARCANTONIO, É., LIA, R. C., HOLZHAUSEN, M. & MARCANTONIO-JÚNIOR, É. 2012. Bioactive glass for alveolar ridge augmentation. J Craniofac Surg, 23, e220-2. MARSHALL, D. B., NOMA, T. & EVANS, A. G. 1982. A simple method for determining elastic-modulus to hardness ratios using Knoop indentation measurements. Commun ACS, 65, 175-6. MARSHALL, G. W. 1993. Dentin: Microstructure and characterization. Quintessence Int, 24, 606-17. MARSHALL, G. W., MARSHALL, S. J., KINNEY, J. H. & BALOOCH, M. 1997. The dentin substrate: structure and properties related to bonding. J Dent Res, 25, 441-58. MARSHALL, S. J., BAYNE, S. C., BAIER, R., TOMSIA, A. P. & MARSHALL, G. W. 2010. A review of adhesion science. Dent Mater, 26, e11-6. MASIH, S., THOMAS, A. M., KOSHY, G. & JOSHI, J. L. 2011. Comparative evaluation of the microleakage of two modified glass ionomer cements on primary molars. An in vivo study. J Indian Soc Pedod Prev Dent, 29, 135-9. MATCHETT, M. D., MACKINNON, S. L., SWEENEY, M. I., GOTTSCHALL-PASS, K. T. & HURTA, R. A. 2005. Blueberry flavonoids inhibit matrix metalloproteinase activity in DU145 human prostate cancer cells. Biochem Cell Biol, 83, 637-43. MAZZONI, A., MANNELLO, F., TAY, F. R., TONTI, G. A., PAPA, S., MAZZOTTI, G., DI LENARDA, R., PASHLEY, D. H. & BRESCHI, L. 2007. Zymographic

292

analysis and characterization of MMP-2 and -9 forms in human sound dentin. J Dent Res, 86, 436-40. MAZZONI, A., PASHLEY, D. H. & NISHITANI, Y. 2006. Reactivation of inactivated endogenous proteolytic activities in phosphoric acid-etched dentine by etch-andrinse adhesives. Biomaterials, 27, 4470-6. MCLEAN, J. W. 1990. Cermet cements. J Am Dent Assoc, 120, 43-7. MEREDITH, N., SHERIFF, M., SETCHELL, D. J. & SWANSON, S. A. V. 1996. Measurement of the microhardness and Young’s modulus of human enamel and dentine using an indentation technique. Arch Oral Biol, 41, 539-45. MIN, B. G., STACHURSKI, Z. H. & HODGKIN, J. H. 1993. Cure kinetics of elementary reactions of a DGEBA/DDS epoxy resin: 1. Glass transition temperature versus conversion. Polymer, 34, 4908-12. MITCHELL, C. A., DOUGLAS, W. H. & CHENG, Y. S. 1999. Fracture toughness of conventional, resin-modified glass-ionomer and composite luting cements. Dent Mater, 15, 7-13. MJÖR, I. A. & GORDAN, V. V. 2002. Failure, repair, refurbishing and longevity of restorations. Oper Dent, 27, 528-34. MOHAMMADI, Z. & ABBOTT, P. V. 2009. Antimicrobial substantivity of root canal irrigants and medicaments: a review. Aust Endod J, 35, 131-9. MONTICELLI, F., OSORIO, R., MAZZITELLI, C., FERRARI, M. & TOLEDANO, M. 2008. Limited decalcification/diffusion of self-adhesive cements into dentin. J Dent Res, 87, 974-9. MORAES, J. C. S. & GRANDINI, C. R. 2011. The glass transition temperature in dental composites. Metal, ceramic and polymeric composites for various uses, 7, 669-84.

293

MOSHAVERINIA, A., ANSARI, S., MOSHAVERINIA, M., SCHRICKER, S. R. & CHEE, W. W. 2011. Ultrasonically set novel NVC-containing glass-ionomer cements for applications in restorative dentistry. J Mater Sci Mater Med, 22, 2029-34. MOSZNER, N., FISCHER, U. K., ANGERMANN, J. & RHEINBERGER, V. 2006. Bis-(acrylamide)s as new cross-linkers for resin-based composite restoratives. Dent Mater, 22, 1157-62. MOSZNER, N., SALZ, U. & ZIMMERMANN, J. 2005. Chemical aspects of selfetching enamel-dentin adhesives: a systematic review. Dent Mater, 21, 895-910. MUNKSGAARD, E. C., PEUTZFELDT, A. & ASMUSSEN, E. 2000. Elution of TEGDMA and BisGMA from a resin and a resin composite cured with halogen or plasma light. Eur J Oral Sci, 108, 341-5. NAKABAYASHI, N., KOJIMA, K. & MASUHARA, E. 1982. Promotion of adhesion by the infiltration of monomers into tooth substrates. J Biomed Mater Res, 16, 265-73. NAKABAYASHI, N. & PASHLEY, D. H. 1998. Hybridization of Dental Hard Tissues. Quintessence Publishing, Tokyo. NAKABAYASHI, N. & TAKARADA, K. 1992. Effect of HEMA on bonding to dentin. Dent Mater, 8, 125-30. NAKABAYASHI, N., WATANABE, A. & IGARASHI, K. 2004. AFM observation of collapse and expansion of phosphoric acid-demineralized dentin. J Biomed Mater Res A, 68, 558-65. NAKAOKI, Y., NIKAIDO, T., PEREIRA, P. N., INOKOSHI, S. & TAGAMI, J. 2000. Dimensional changes of demineralized dentin treated with HEMA primers. Dent Mater, 16, 441-6.

294

NAKAZAWA, Y., SEINO, E., USHIKI, T., OGATA, T., HIRAI, Y., KAWADA, E. & ODA, Y. 1999. Microhardness evaluations of resin-dentin bonding areas by nano-indentation. Bull Tokyo Dent Coll, 40, 47-54. NALLA, R. K., PORTER, A. E., DARAIO, C., MINOR, A. M., RADMILOVIC, V., STACH, E. A., TOMSIA, A. P. & RITCHIE, R. O. 2005. Ultrastructural examination of dentin using focused ion-beam cross-sectioning and transmission electron microscopy. Micron, 36, 672-80. NEILL, J. 1838. The Hydrostatic Pressure Doctrine. Transactions of the College of Physicians of Philadelphia, 111. NGO, H. 2010. Glass-ionomer cements as restorative and preventive materials. Dent Clin North Am, 54, 551-63. NGO, H. C., MOUNT, G., MC INTYRE, J., TUISUVA, J. & VON DOUSSA, R. J. 2006. Chemical exchange between glass-ionomer restorations and residual carious dentine in permanent molars: An in vivo study. J Dent, 34, 608-13. NIEDERBERGER, M. & CÃ&#x2013;LFEN, H. 2006. Oriented attachment and mesocrystals: non-classical crystallization mechanisms based on nanoparticle assembly. Phys Chem Chem Phys, 8, 3271-87. NIELSEN-MARSH, C. M., HEDGES, R. E. M., MANN, T. & COLLINS, M. J. 2000. A preliminary investigation of the application of differential scanning calorimetry to the study of collagen degradation in archaeological bone. Thermochim Acta, 365, 129-39. NIKAIDO, T., KUNZELMANN, K. H., CHEN, H., OGATA, M., HARADA, N., YAMAGUCHI, S., COX, C. F., HICKEL, R. & TAGAMI, J. 2002. Evaluation of thermal cycling and mechanical loading on bond strength of a self-etching primer system to dentin. Dent Mater, 18, 269-75.

295

NIKOLAENKO, S. A., LOHBAUER, U., ROGGENDORF, M., PETSCHELT, A., DASCH, W. & FRANKENBERGER, R. 2004. Influence of c-factor and layering technique on microtensile bond strength to dentin. Dent Mater, 20, 57985. NIMNI, M. E., CHEUNG, D., STARTES, B., KODAMA, M. & SHEIKH, K. 1998. Bioprosthesis derived from crosslinked and chemically modified collagenous tissues. Nimni ME, editor. Collagen, Vol. III. Boca Raton, FL: CRC Press, 1-38. NISHITANI, Y., YOSHIYAMA, M., HOSAKA, K., TAGAMI, J., DONNELLY, A., CARRILHO, M., TAY, F. R. & PASHLEY, D. H. 2007. Use of Hoy's solubility parameters to rank water sorption/solubility of experimental self-etching primers and adhesives. Eur J Oral Sci, 115, 81-6. NISHITANI, Y., YOSHIYAMA, M. & WADGAONKAR, B. 2006. Activation of gelatinolytic/collagenolytic activity in dentin by self-etching adhesives. Eur J Oral Sci, 114, 160-6. NOVINEC, M., KOVACIC, L., LENARCIC, B. & BAICI, A. 2010. Conformational flexibility and allosteric regulation of cathepsin K. Biochem J, 429, 379-89. NUDELMAN, F., PIETERSE, K., GEORGE, A., BOMANS, P. H., FRIEDRICH, H., BRYLKA, L. J., HILBERS, P. A., DE WITH, G. & SOMMERDIJK, N. A. 2010. The role of collagen in bone apatite formation in the presence of hydroxyapatite nucleation inhibitors. Nat Mater, 9, 1004-9. OBERMAJER, N., JEVNIKAR, Z., DOLJAK, B. & KOS, J. 2008. Role of cysteine cathepsins in matrix degradation and cell signalling. Connect Tissue Res, 49, 193-6.

296

OGAWA, K., YAMASHITA, Y., ICHIJO, T. & FUSAYAMA, T. 1983. The ultrastructure and hardness of the transparent layer of human carious dentin. J Dent Res, 62, 7-10. OLIVEIRA, S. S., PUGACH, M. K., STROTHER, J., DARLING, C. L., FRIED, D., GANSKY, S. A., MARSHALL, S. J. & MARSHALL JR, G. W. 2003. The influence of the dentin smear layer on adhesion: a self-etching primer vs. a totaletch system. Dent Mater, 19, 758-67. ONG, J. L. & CHAN, D. C. N. 2000. Hydroxyapatite and their use as coatings in dental implants: a review. Crit Rev Biomed Eng, 2000, 667a-707a. ONUMA, K. & ITO, A. 1998. Cluster growth model for hydroxyapatite. Chem Mater, 10, 3346-51. OONISHI, H., KUSHITANI, S., YASUKAWA, E., IWAKI, H., HENCH, L. L., WILSON, J., TSUJI, E. & SUGIHARA, T. 1997. Particulate bioglass compared with hydroxyapatite as a bone graft substitute. Clin Orthop Relat Res, 334, 31625. OSORIO, E., TOLEDANO, M., AGUILERA, F. S., TAY, F. R. & OSORIO, R. 2010. Ethanol wet-bonding technique sensitivity assessed by AFM. J Dent Res, 89, 1264-9. OSORIO, R., TOLEDANO, M., OSORIO, E., AGUILERA, F. S. & TAY, F. R. 2005. Effect of load cycling and in vitro degradation on resin-dentin bonds using a self-etching primer. J Biomed Mater Res A, 72, 399-408. OSORIO, R., YAMAUTI, M., OSORIO, E., ROMĂ N, J. S. & TOLEDANO, M. 2011. Zinc-doped dentin adhesive for collagen protection at the hybrid layer. Eur J Oral Sci, 119, 401-10.

297

OSTWALD, W. F. 1897. Studies on formation and transformation of solid materials. Z Phys Chem, 22, 289-302. OWENS, S. E. J. & MILLER, B. H. 2000. A comparison of shear bond strengths of three visible light-cured orthodontic adhesives. Angle Orthod, 70, 352-6. PALMER, L. C., NEWCOMB, C. J., KALTZ, S. R., SPOERKE, E. D. & STUPP, S. I. 2008. Biomimetic systems for hydroxyapatite mineralization inspired by bone and enamel. Chem Rev, 108, 4754-83. PANIGHI, M. & G’SELL, C. 1993. Effect of the tooth microstructure on the shear bond strength of a dental composite. J Biomed Mater Res, 27, 975-81. PARIROKH, M. & TORABINEJAD, M. 2010. Mineral trioxide aggregate: a comprehensive literature review—part III clinical applications, drawbacks and mechanism of action. J Endod, 36, 400-13. PARK, J., YE, Q., TOPP, E. M., MISRA, A., KIEWEG, S. L. & SPENCER, P. 2010. Effect of photoinitiator system and water content on dynamic mechanical properties of a light-cured bisGMA/HEMA dental resin. J Biomed Mater Res A, 93, 1245-51. PARK, J. G., YE, Q., TOPP, E. M., KOSTORYZ, E. L., WANG, Y., KIEWEG, S. L. & SPENCER, P. 2008. Preparation and properties of novel dentin adhesives with esterase resistance. J Appl Polym Sci, 107, 3588-97. PASHLEY, D., OKABE, A. & PARHAM, P. 1985. The relationship between dentin microhardness and tubule density. Endod Dent Traumatol, 1, 176-9. PASHLEY, D. H. 1989. Dentin, a dynamic substrate - a review. Scanning Microsc, 3, 161-76. PASHLEY, D. H. 1991. Dentin bonding: overview of the substrate with respect to adhesive material. J Esthet Dent, 3, 46-50.

298

PASHLEY, D. H., CIUCCHI, B., SANO, H. & HOMER, J. A. 1993. Permeability of dentin to adhesive agents. Quintessence Int, 24, 618-31. PASHLEY, D. H., SANO, H., CIUCCHI, B., CARVALHO, R. M. & RUSSELL, C. M. 1995a. Bond strength versus dentin structures: a modeling approach. Arch Oral Biol, 40, 1109-18. PASHLEY, D. H., SANO, H., CIUCCHI, B., YOSHIYAMA, M. & CARVALHO, R. M. 1995b. Adhesion testing of dentin bonding agents: a review. Dent Mater, 11, 117-25. PASHLEY, D. H., TAO, L., BOYD, L., KING, G. E. & HORNER, J. A. 1988. Scanning electron microscopy of the substructure of smear layers in human dentine. Arch Oral Biol, 33, 265-70. PASHLEY, D. H., TAY, F. R., BRESCHI, L., TJÄDERHANE, L., CARVALHO, R. M., CARRILHO, M. & TEZVERGIL-MUTLUAY, A. 2011. State of the art etch-and-rinse adhesives. Dent Mater, 27, 1-16. PASHLEY, D. H., TAY, F. R., CARVALHO, R. M., RUEGGEBERG, F. A., AGEE, K. A., CARRILHO, M., DONNELLY, A. & GARCÍA-GODOY, F. 2007. A review of the interactions between dentin matrix and solvated resins using a macromodel of the hybrid layer. Am J Dent, 20, 7-20. PASHLEY, D. H., TAY, F. R., YIU, C., HASHIMOTO, M., BRESCHI, L., CARVALHO, R. M. & ITO, S. 2004. Collagen degradation by host-derived enzymes during aging. J Dent Res, 83, 216-21. PASHLEY, E. L., ZHANG, Y., LOCKWOOD, P. E., RUEGGEBERG, F. A. & PASHLEY, D. H. 1998. Effects of HEMA on water evaporation from waterHEMA mixtures. Dent Mater, 14, 6-10.

299

PASSALINI, P., FIDALGO, T. K., CALDEIRA, E. M., GLEISER, R., NOJIMA, M. D. C. & MAIA, L. C. 2010. Mechanical properties of one and two-step fluoridated orthodontic resins submitted to different pH cycling regimes. Braz Oral Res, 24, 197-203. PELKA, M. 1999. Elution parameters and HPLC-detection of single components from resin composite. Clin Oral Investig, 3, 194-200. PELLENQ, R. J. M., KUSHIMA, A. & SHAHSAVARI, R. 2009. A realistic molecular model of cement hydrates. Proc Natl Acad Sci USA, 106, 16102-7. PERINKA, L., SANO, H. & HOSODA, H. 1992. Dentin thickness, hardness, and Caconcentration vs bond strength of dentin adhesives. Dent Mater, 8, 229-33. PERIS, A. R., MITSUI, F. H., LOBO, M. M., BEDRAN-RUSSO, A. K. & MARCHI, G. M. 2007. Adhesive systems and secondary caries formation: Assessment of dentin bond strength, caries lesions depth and fluoride release. Dent Mater, 23, 308-16. PETERS, M. C. 2010. Strategies for noninvasive demineralized tissue repair. Dent Clin North Am, 54, 507-25. PETERS, M. C., BRESCIANI, E., BARATA, T. J. E., FAGUNDES, T. C., NAVARRO, R. L., NAVARRO, M. F. L. & DICKENS, S. H. 2010. In vivo dentin remineralization by calcium-phosphate cement. J Dent Res, 89, 286-91. PETERS, M. C. & MCLEAN, M. E. 2001. Minimally invasive operative care. I. Minimal intervention and concepts for minimally invasive cavity preparations. J Adhes Dent, 3, 7-16. PEUMANS, M., DE MUNCK, J., VAN LANDUYT, K., LAMBRECHTS, P. & VAN MEERBEEK, B. 2007. Five-year clinical effectiveness of a two-step selfetching adhesive. J Adhes Dent, 9, 7-10.

300

PEUMANS, M., KANUMILLI, P., DE MUNCK, J., VAN LANDUYT, K., LAMBRECHTS, P. & VAN MEERBEEK, B. 2005. Clinical effectiveness of contemporary adhesives: a systematic review of current clinical trials. Dent Mater, 21, 864-81. PIOCH, T., STAEHLE, H. J., DUSCHNER, H. & GARCIA-GODOY, F. 2001. Nanoleakage at the composite-dentin interface: a review. Am J Dent, 14, 252-8. PISKOUNOVA, S., FORSGREN, J., BROHEDE, U., ENGQVIST, H. & STRÃ&#x2DC;MME, M. 2009. In vitro characterization of bioactive titanium dioxide/hydroxyapatite surfaces functionalized with BMP-2. J Biomed Mater Res B Appl Biomater, 91, 780-7. PLASSARD, C., LESNIEWSKA, E., POCHARD, I. & NONAT, A. 2005. Nanoscale experimental investigation of particle interactions at the origin of the cohesion of cement. Langmuir, 21, 7263-70. POITEVIN, A., DE MUNCK, J., CARDOSO, M. V., MINE, A., PEUMANS, M., LAMBRECHTS, P. & VAN MEERBEEK, B. 2010. Dynamic versus static bond-strength testing of adhesive interfaces. Dent Mater, 26, 1068-76. POSNER, A. S. 1985. The mineral of bone. Clin Orthop Relat Res, 200, 87-99. PRATI, C., FAVA, F., DI GIOIA, D., SELIGHINI, M. & PASHLEY, D. H. 1993. Antibacterial effectiveness of dentin bonding systems. Dent Mater, 9, 338-43. PRICE, R. B., DERAND, T., ANDREOU, P. & MURPHY, D. 2003. The effect of two configuration factors, time, and thermal cycling on resin to dentin bond strengths. Biomaterials, 24, 1013-21. PROFETA, A. C., MANNOCCI, F., FOXTON, R. M., THOMPSON, I., WATSON, T. F. & SAURO, S. 2012. Bioactive effects of a calcium/sodium phosphosilicate

301

on the resin-dentine interface: a microtensile bond strength, scanning electron microscopy, and confocal microscopy study. Eur J Oral Sci, 120, 353-62. QI, Y. P., LI, N., NIU, L. N., PRIMUS, C. M., LING, J. Q., PASHLEY, D. H. & TAY, F. R. 2012. Remineralization of artificial dentinal caries lesions by biomimetically modified mineral trioxide aggregate. Acta Biomater, 8, 836-42. RADOVIC, I., MONTICELLI, F., GORACCI, C., VULICEVIC, Z. R. & FERRARI, M. 2008. Self-adhesive resin cements: a literature review. J Adhes Dent, 10, 251-8. RAHN, B. A. & PERREN, S. M. 1970. Calcein blue as a fluorescent label in bone. Experientia, 26, 519. RAHN, B. A. & PERREN, S. M. 1971. Xylenol orange, a fluorochrome useful in polychrome sequential labelling of calcifying tissues. Stain Technol, 46, 125-9. RAHN, B. A. & PERREN, S. M. 1972. Alizarinkomplexon, Fluorochromzur Markierung von Knocken und Dentinanbau. Experientia, 28, 180. RAO, A. & MALHOTRA, N. 2011. The role of remineralizing agents in dentistry: a review. Compend Contin Educ Dent, 32, 26-33. RAY, S., LUKYANOV, P. & OCHIENG, J. 2003. Members of the cystatin superfamily interact with MMP-9 and protect it from autolytic degradation without affecting its gelatinolytic activities. Biochim Biophys Acta, 1652, 91-102. REINKE, S. M., LAWDER, J. A., DIVARDIN, S., RAGGIO, D., REIS, A. & LOGUERCIO, A. D. 2012. Degradation of the resin-dentin bonds after simulated and inhibited cariogenic challenge in an in situ model. J Biomed Mater Res B Appl Biomater, 100, 1466-71.

302

REIS, A., GRANDE, R. H., OLIVEIRA, G. M., LOPES, G. C. & LOGUERCIO, A. D. 2007. A 2-year evaluation of moisture on microtensile bond strength and nanoleakage. Dent Mater, 23, 862-70. REIS, A. F., ARRAIS, C. A., NOVAES, P. D., CARVALHO, R. M., DE GOES, M. F. & GIANNINI, M. 2004. Ultramorphological analysis of resin-dentin interfaces produced with water-based single-step and two-step adhesives: nanoleakage expression. J Biomed Mater Res B Appl Biomater, 71, 90-8. RHEE, S. H., LEE, J. D. & TANAKA, J. 2000. Nucleation of hydroxyapatite crystal through chemical interaction with collagen. J Am Ceram Soc, 83, 2890-2. RITTER, A. V., SWIFT, E. J. & YAMAUCHI, M. 2001. Effects of phosphoric acid and glutaraldehyde-HEMA on dentin collagen. Eur J Oral Sci, 109, 348-53. ROOT, M. J. 1990. Inhibition of the amorphous calcium phosphate phase transformation reaction by polyphosphates and metal ions. Calcif Tissue Int, 47, 112-6. ROSALES-LEAL, J. I., OSORIO, R., HOLGADO-TERRIZA, J. A., CABRERIZOVĂ?LCHEZ, M. A. & TOLEDANO, M. 2001. Dentin wetting by four adhesive systems. Dent Mater, 17, 526-32. ROST, F. W. D. 1995. Fluorescence Microscopy II. Cambridge: Cambridge University Press. RUEGGEBERG, F. A., HASHINGER, D. T. & FAIRHURST, C. W. 1990. Calibration of FTIR conversion analysis of contemporary dental resin composites. Dent Mater, 6, 241-9. RYOU, H., NIU, L. N., DAI, L., PUCCI, C. R., AROLA, D. D., PASHLEY, D. H. & TAY, F. R. 2011. Effect of biomimetic remineralization on the dynamic nanomechanical properties of dentin hybrid layers. J Dent Res, 90, 1122-8.

303

SADAT-SHOJAI, M., ATAI, M., NODEHI, A. & KHANLAR, L. N. 2010. Hydroxyapatite nanorods as novel fillers for improving the properties of dental adhesives: Synthesis and application. Dent Mater, 26, 471-82. SADEK, F. T., BRAGA, R. R., MUENCH, A., LIU, Y., PASHLEY, D. H. & TAY, F. 2010a. Ethanol wet-bonding challenges current anti-degradation strategy. J Dent Res, 89, 1499-504. SADEK, F. T., MAZZONI, A., BRESCHI, L., TAY, F. R. & BRAGA, R. R. 2010b. Six-month evaluation of adhesives interface created by a hydrophobic adhesive to acid-etched ethanol-wet bonded dentine with simplified dehydration protocols. J Dent, 38, 276-83. SADEK, F. T., PASHLEY, D. H., FERRARI, M. & TAY, F. R. 2007. Tubular occlusion optimizes bonding of hydrophobic resins to dentin. J Dent Res, 86, 524-8. SADEK, F. T., PASHLEY, D. H., NISHITANI, Y., CARRILHO, M. R., DONNELLY, A., FERRARI, M. & TAY, F. R. 2008. Application of hydrophobic resin adhesives to acid-etched dentin with an alternative wet bonding technique. J Biomed Mater Res A, 84, 19-29. SALIAN, S., THAKUR, S., KULKARNI, S. & LATORRE, G. 2010. A randomized controlled clinical study evaluating the efficacy of two desensitizing dentifrices. J Clin Dent, 21, 82-7. SALZ, U., ZIMMERMANN, J., ZEUNER, F. & MOSZNER, N. 2005. Hydrolytic stability of self-etching adhesive systems. J Adhes Dent, 7, 107-16. SANCHEZ, F. & ZHANG, L. 2008. Molecular dynamics modeling of the interface between surface functionalized graphitic structures and calcium–silicate–

304

hydrate: interaction energies, structure, and dynamics. J Colloid Interf Sci, 323, 349-58. SANO, H., CIUCCHI, B., MATTHEWS, W. G. & PASHLEY, D. H. 1994. Tensile properties of mineralized and demineralized human and bovine dentin. J Dent Res, 73, 1205-11. SANO, H., TAKATSU, T., CIUCCHI, B., HORNER, J. A., MATTEWS, W. G. & PASHLEY, D. H. 1995a. Nanoleakage: leakage within the hybrid layer. Oper Dent, 20, 18-35. SANO, H., YOSHIYAMA, M., EBISU, S., BURROW, M. F., TAKATSU, T. & CIUCCHI, B. 1995b. Comparative SEM and TEM observations of nanoleakage within the hybrid layer. Oper Dent, 20, 160-7. SANTERRE, J. P., SHAJII, L. & LEUNG, B. W. 2001. Relation of dental composite formulations to their degradation and the release of hydrolyzed polymeric-resinderived products. Crit Rev Oral Biol Med, 12, 136-51. SANTOS, C., LUKLINSKA, Z. B., CLARKE, R. L. & DAVY, K. W. M. 2001. Hydroxyapatite as a filler for dental composite materials: mechanical properties and in vitro bioactivity of composites. J Mater Sci Mater Med, 12, 565-73. SARKAR, N. K., CAICEDO, R., RITWIK, P., MOISEYEVA, R. & I., K. 2005. Physicochemical basis of the biologic properties of mineral trioxide aggregate. J Endod, 31, 97-100. SARR, M., MINE, A., DE MUNCK, J., CARDOSO, M. V., KANE, A. W., VREVEN, J., VAN MEERBEEK, B. & VAN LANDUYT, K. L. 2010. Immediate bonding effectiveness of contemporary composite cements to dentin. Clin Oral Investig, 14, 569-77.

305

SAURO, S., MANNOCCI, F., TAY, T., PASHLEY, D. H., COOK, R., CARPENTER, G. H. & WATSON, T. F. 2009a. Deproteinization effects of NaOCl on acidetched dentin in clinically-relevant vs prolonged periods of application. A confocal and environmental scanning electron microscopy study. Oper Dent, 34, 166-73. SAURO, S., OSORIO, R., WATSON, T. F. & TOLEDANO, M. 2012a. Therapeutic effects of novel resin bonding systems containing bioactive glasses on mineraldepleted areas within the bonded-dentine interface. J Mater Sci Mater Med, 23, 1521-32. SAURO, S., PASHLEY, D. H., MONTANARI, M., CHERSONI, S., CARVALHO, R. M., TOLEDANO, M., OSORIO, R., TAY, F. R. & PRATI, C. 2007. Effect of simulated pulpal pressure on dentin permeability and adhesion of self-etch adhesives. Dent Mater, 23, 705-13. SAURO, S., THOMPSON, I. & WATSON, T. F. 2011a. Effects of common dental materials used in preventive or operative dentistry on dentin permeability and remineralization. Oper Dent, 36, 222-30. SAURO, S., TOLEDANO, M., AGUILERA, F. S., MANNOCCI, F., PASHLEY, D. H., TAY, F. R., WATSON, T. F. & OSORIO, R. 2010. Resin-dentin bonds to EDTA-treated vs. acid-etched dentin using ethanol wet-bonding. Dent Mater, 26, 368-79. SAURO, S., TOLEDANO, M., AGUILERA, F. S., MANNOCCI, F., PASHLEY, D. H., TAY, F. R., WATSON, T. F. & OSORIO, R. 2011b. Resin-dentin bonds to EDTA-treated vs. acid-etched dentin using ethanol wet-bonding. Part II: Effects of mechanical cycling load on microtensile bond strengths. Dent Mater, 27, 56372.

306

SAURO, S., WATSON, T. F., MANNOCCI, F., MIYAKE, K., HUFFMAN, B. P., TAY, F. R. & PASHLEY, D. H. 2009b. Two-photon laser confocal microscopy of micropermeability of resin-dentin bonds made with water or ethanol wet bonding. J Biomed Mater Res B Appl Biomater, 90, 327-37. SAURO, S., WATSON, T. F., MANNOCCI, F., TAY, F. R. & PASHLEY, D. H. 2009c. Prevention of water contamination of ethanol-saturated dentin and hydrophobic hybrid layers. J Adhes Dent, 11, 271-8. SAURO, S., WATSON, T. F., TAY, F. R., CHERSONI, S., BRESCHI, L. & BERNARDI, F. 2006. Water uptake of bonding systems applied on root dentin surfaces: a SEM and confocal microscopic study. Dent Mater, 22, 671-80. SAURO, S., WATSON, T. F., THOMPSON, I., TOLEDANO, M., NUCCI, C. & BANERJEE, A. 2012b. Influence of air-abrasion executed with polyacrylic acid-Bioglass 45S5 on the bonding performance of a resin-modified glass ionomer cement. Eur J Oral Sci, 120, 168-77. SAY, E. C., NAKAJIMA, M., SENAWONGSE, P., SOYMAN, M., OZER, F., OGATA, M. & TAGAMI, J. 2006. Microtensile bond strength of a filled vs unfilled adhesive to dentin using self-etch and total-etch technique. J Dent, 34, 283-91. SCAFFA, P. M., VIDAL, C. M., BARROS, N., GESTEIRA, T. F., CARMONA, A. K., BRESCHI, L., PASHLEY, D. H., TJÃ&#x201E;DERHANE, L., TERSARIOL, I. L., NASCIMENTO, F. D. & CARRILHO, M. R. 2012. Chlorhexidine inhibits the activity of dental cysteine cathepsins. J Dent Res, 91, 420-5. SCHWARTZ, R. S., MAUGER, M., CLEMENT, D. J. & WALKER, W. A. 1999. Mineral trioxide aggregate: a new material for endodontics. J Am Dent Assoc, 130, 967-75.

307

SCOTT, J. E. & THOMLINSON, A. M. 1998. The structure of interfibrillar proteoglycan bridges (‘shape modules’) in extracellular matrix of fibrous connective tissues and their stability in various chemical environments. J Anat, 192, 391-405. SELA-PASSWELL, N., ROSENBLUM, G., SHOHAM, T. & SAGI, I. 2010. Structural and functional bases for allosteric control of MMP activities: Can it pave the path for selective inhibition? Biochim Biophys Acta, 1803, 29-38. SEPULVEDA, P., JONES, J. R. & HENCH, L. L. 2002. In vitro dissolution of meltderived 45S5 and sol-gel derived 58S bioactive glasses. J Biomed Mater Res, 61, 301-11. SHANNON, I. L. & KEUPER, J. B. 1976. Microhardness of human dentin: Baseline values and effects of fluorides. J Miss Dent Assoc, 56, 11-8. SHEN, P., CAI, F., NOWICKI, A., VINCENT, J. & REYNOLDS, E. C. 2001. Remineralization of enamel subsurface lesions by sugar-free chewing gum containing casein phosphopeptide-amorphous calcium phosphate. J Dent Res, 80, 2066-70. SHIMADA, Y., YAMAGUCHI, S. & TAGAMI, J. 2002. Micro-shear bond strength of dual-cured resin cement to glass ceramics. Dent Mater, 18, 380-8. SHIN, T. P., YAO, X., HUENERGARDT, R., WALKER, M. P. & WANG, Y. 2009. Morphological and chemical characterization of bonding hydrophobic adhesive to dentin using ethanol wet bonding technique. Dent Mater, 25, 1050-7. SHOKATI, B., TAM, L. E., SANTERRE, J. P. & FINER, Y. 2010. Effect of salivary esterase on the integrity and fracture toughness of the dentin-resin interface. J Biomed Mater Res B Appl Biomater, 94, 230-7.

308

SHONO, Y., TERASHITA, M., SHIMADA, J., KOZONO, Y., CARVALHO, R. M. & RUSSELL, C. M. 1999. Durability of resinâ&#x20AC;&#x201C;dentin bonds. J Adhes Dent, 1, 2118. SIDHU, S. K. & WATSON, T. F. 1998. Interfacial characteristics of resin-modified glass-ionomer materials: a study on fluid permeability using confocal fluorescence microscopy. J Dent Res, 77, 1749-59. SILVER, F. H. 1994. Wound dressings and skin replacement. Silver FH, editor. Biomaterials medical devices and tissue engineering an integral approach. London: Chapman & Hall, 46-91. SIMONSEN, R. J. 2002. Pit and fissure sealant: review of the literature. Pediatr Dent, 24, 393-414. SKINNER, L. B., CHAE, S., BENMORE, C. J., WENK, H. R. & MONTEIRO, P. J. M. 2010. Nanostructure of calcium silicate hydrates in cements. Phys Rev Lett, 104, Article ID 195502. SKOVRON, L., KOGEO, D., GORDILLO, L. A., MEIER, M. M., GOMES, O. M., REIS, A. & LOGUERCIO, A. D. 2010. Effects of immersion time and frequency of water exchange on durability of etch-and-rinse adhesive. J Biomed Mater Res B Appl Biomater, 95, 339-46. SKRTIC, D., HAILER, A. W., TAKAGI, S., ANTONUCCI, J. M. & EANES, E. D. 1996. Quantitative assessment of the efficacy of amorphous calcium phosphate/methacrylate composites in remineralizing caries-like lesions artificially produced in bovine enamel. J Dent Res, 75, 1679-86. SMALL, B. W. 2008. Reparative dentistry or restorative dentistry? Gen Dent, 56, 1268, 130.

309

SMITH, A. J. 2003. Vitality of the dentin-pulp complex in health and disease: growth factors as key mediators. J Dent Educ, 67, 678-89. SOARES, C. J., SOARES, P. V., SANTOS-FILHO, P. C. & ARMSTRONG, S. R. 2008. Microtensile specimen attachment and shape-finite element analysis. J Dent Res, 87, 89-93. SODERHOLM, K. J., ZIGAN, M., RAGAN, M., FISCHLSCHWEIGER, W. & BERGMAN, M. 1984. Hydrolytic degradation of dental composites. J Dent Res, 63, 1248-54. SOLES, C. F. & YEE, A. F. 2000. A discussion of the molecular mechanisms of moisture transport in epoxy resins. J Polym Sci Part B Polym Phys, 38, 792-802. SOMASUNDARAN, P., OFORI AMANKONAH, J. & ANANTHAPADMABHAN, K. P. 1985. Mineral---solution equilibria in sparingly soluble mineral systems. Colloids Surf, 15, 309-33. SOUZA, P. P., ARANHA, A. M., HEBLING, J., GIRO, E. M. & COSTA, C. A. 2006. In vitro cytotoxicity and in vivo biocompatibility of contemporary resinmodified glass-ionomer cements. Dent Mater, 22, 838-44. SPENCER, P. & WANG, Y. 2002. Adhesive phase separation at the dentin interface under wet bonding conditions. J Biomed Mater Res, 62, 447-56. SPENCER, P., YE, Q., PARK, J., TOPP, E. M., MISRA, A., MARANGOS, O., WANG, Y., BOHATY, B. S., SINGH, V., SENE, F., ESLICK, J., CAMARDA, K. & KATZ, J. L. 2010. Adhesive/Dentin interface: the weak link in the composite restoration. Ann Biomed Eng, 38, 1989-2003. STANINEC, M., KIM, P., MARSHALL, G. W., RITCHIE, R. O. & MARSHALL, S. J. 2008. Fatigue of dentinâ&#x20AC;&#x201C;composite interfaces with four-point bend. Dent Mater, 24, 799-803.

310

SUH, B. I. 1991. All-Bond: fourth-generation dentin bonding system. J Esthet Dent, 3, 139-47. SULKALA, M., WAHLGREN, J., LARMAS, M., SORSA, T., TERONEN, O., SALO, T. & TJÃ&#x201E;DERHANE, L. 2001. The effects of MMP inhibitors on human salivary MMP activity and caries progression in rats. J Dent Res, 80, 1545-9. SUNG, H.-W., HUANG, D.-M., CHANG, W.-H., HUANG, R.-N. & HSU, J.-C. 1999. Evaluation of gelatin hydrogel crosslinked with various crosslinking agents as bioadhevises: In vitro study. J Biomed Mater Res, 46, 520-30. SWETHA, M., SAHITHI, K., MOORTHI, A., SARANYA, N., SARAVANAN, S., RAMASAMY, K., SRINIVASAN, N. & SELVAMURUGAN, N. 2012. Synthesis, characterization, and antimicrobial activity of nano-hydroxyapatitezinc for bone tissue engineering applications. J Nanosci Nanotechnol, 12, 16772. SWIFT, E. J., PERDIGAO, J. & HEYMANN, H. O. 1995. Bonding to enamel and dentin: a brief story and state of the art. Quintessence Int, 26, 95-110. SWIFT, E. J., PERDIGAO, J. & HEYMANN, H. O. 1998. Enamel bond strengths of "one-bottle" adhesives. Pediatr Dent, 20, 259-62. TADDEI, P., MODENA, E., TINTI, A., SIBONI, F., PRATI, C. & GANDOLFI, M. G. 2011. Vibrational investigation on the in vitro bioactivity of commercial and experimental calcium-silicate cements for root-end endodontic therapy. J Mol Struct, 993, 367-75. TAKEUCHI, T. & HISHIYA, T. 2008. Molecular imprinting of proteins emerging as a tool for protein recognition. Org Biomol Chem, 6, 2459-67.

311

TALLANT,

C.,

MARRERO,

GOMIS-RÜTH,

2010.

Matrix

metalloproteinases: fold and function of their catalytic domains. Biochim Biophys Acta, 1803, 20-8. TANAKA, K., TAIRA, M., SHINTANI, H., WAKASA, K. & YAMAKI, M. 1991. Residual monomers (TEGDMA and Bis-GMA) of a set visible-light-cured dental composite resin when immersed in water. J Oral Rehabil, 18, 353-62. TANG, R. K., NANCOLLAS, G. H. & ORME, C. A. 2001. Mechanism of dissolution of sparingly soluble electrolytes. J Am Chem Soc, 123, 5437-43. TAY, F. R., CARVALHO, R. M. & PASHLEY, D. H. 2004a. Water movement across bonded dentin - too much of a good thing. J Appl Oral Sci, 12, 12-25. TAY, F. R., GWINNETT, A. I., PONG, K. U. & WEI, S. H. Y. 1995. The overwet phenomenon, an optical, micromorphological study of surface moisture in the acid etched, resin-dentin interface. Am J Dent, 9, 43-8. TAY, F. R., GWINNETT, A. J., PANG, K. M. & WEI, S. H. 1996. Resin permeation into acid-conditioned, moist, and dry dentin: a paradigm using water-free adhesive primers. J Dent Res, 75, 1034-44. TAY, F. R., HASHIMOTO, M., PASHLEY, D. H., PETERS, M. C., LAI, S. C. N., YIU, C. K. Y. & CHEONG, C. 2003. Aging affects two modes of nanoleakage expression in bonded dentin. J Dent Res, 82, 537-41. TAY, F. R. & PASHLEY, D. H. 2003a. Have dentin adhesives become too hydrophilic? J Can Dent Assoc, 69, 726-31. TAY, F. R. & PASHLEY, D. H. 2003b. Water-treeing – a potential mechanism for degradation of dentine adhesives. Am J Dent, 16, 6-12. TAY, F. R. & PASHLEY, D. H. 2008. Guided tissue remineralisation of partially demineralised human dentine. Biomaterials, 29, 1127-37.

312

TAY, F. R. & PASHLEY, D. H. 2009. Biomimetic remineralization of resin-bonded acid-etched dentin. J Dent Res, 88, 719-24. TAY, F. R., PASHLEY, D. H., GARCIA-GODOY, F. & YIU, C. K. 2004b. Singlestep, self-etch adhesives behave as permeable membranes after polymerization. Part II. Silver tracer penetration evidence. Am J Dent, 17, 315-22. TAY, F. R., PASHLEY, D. H., KAPUR, R. R., CARRILHO, M. R., HUR, Y. B., GARRETT, L. V. & TAY, K. C. 2007. Bonding BisGMA to dentin--a proof of concept for hydrophobic dentin bonding. J Dent Res, 86, 1034-9. TAY, F. R., PASHLEY, D. H., LOUSHINE, R. J., WELLER, R. N., MONTICELLI, F. & OSORIO, R. 2006. Self-etching adhesives increase collagenolytic activity in radicular dentin. J Endod, 32, 862-8. TAY, F. R., PASHLEY, D. H., SUH, B. I., CARVALHO, R. M. & ITTHAGARUN, A. 2002a. Single-step adhesives are permeable membranes. J Dent, 30, 371-82. TAY, F. R., PASHLEY, D. H. & YOSHIYAMA, M. 2002b. Two modes of nanoleakage expression in single-step adhesives. J Dent Res, 81, 472-6. TAYLOR, H. F. W. 1997. Cement Chemistry, 2nd edn. Thomas Telford Ltd., London (1997). Thomas Telford Ltd., London. TEN CATE, J. M. & FEATHERSTONE, J. D. 1991. Mechanistic aspects of the interactions between fluoride and dental enamel. Crit Rev Oral Biol Med, 2, 283-96. TEN CATE, J. M. & VAN DUINEN, R. N. B. 1995. Hypermineralization of dental lesions adjacent to glass-ionomer cement restorations. J Dent Res, 74, 1266-71. TENO, N. & MASUYA, K. 2010. Orally bioavailable cathepsin K inhibitors with pyrrolopyrimidine scaffold. Curr Top Med Chem, 10, 752-66.

313

TERSARIOL, I. L., GERALDELI, S., MINCIOTTI, C. L., NASCIMENTO, F. D., PÄÄKKÖNEN, V., MARTINS, M. T., CARRILHO, M. R., PASHLEY, D. H., TAY, F. R., SALO, T. & TJÄDERHANE, L. 2010. Cysteine cathepsins in human dentin-pulp complex. J Endod, 36, 475-81. TEZVERGIL-MUTLUAY, A., AGEE, K. A., HOSHIKA, T., CARRILHO, M., BRESCHI, L., TJÄDERHANE, L., NISHITANI, Y., CARVALHO, R. M., LOONEY, S., TAY, F. R. & PASHLEY, D. H. 2010a. The requirement of zinc and calcium ions for functional MMP activity in demineralized dentin matrices. Dent Mater, 26, 1059-67. TEZVERGIL-MUTLUAY, A., AGEE, K. A., HOSHIKA, T., TAY, F. R. & PASHLEY, D. H. 2010b. The inhibitory effect of polyvinylphosphonic acid on functional matrix metalloproteinase activities in human demineralized dentin. Acta Biomater, 6, 4136-42. TEZVERGIL-MUTLUAY, A., MUTLUAY, M. M., GU, L. S., ZHANG, K., AGEE, K. A., CARVALHO, R. M., MANSO, A., CARRILHO, M., TAY, F. R., BRESCHI, L., SUH, B. I. & PASHLEY, D. H. 2011. The anti-MMP activity of benzalkonium chloride. J Dent, 39, 57-64. THOMPSON, S. K., HALBERT, S. M., BOSSARD, M. J., TOMASZEK, T. A., LEVY, M. A., ZHAO, B., SMITH, W. W., ABDEL-MEGUID, S. S., JANSON, C. A., D'ALESSIO, K. J., MCQUENEY, M. S., AMEGADZIE, B. Y., HANNING, C. R., DESJARLAIS, R. L., BRIAND, J., SARKAR, S. K., HUDDLESTON, M. J., LJAMES, C. F., CARR, S. A., GARNES, K. T., SHU, A., HEYS, J. R., BRADBEER, J., ZEMBRYKI, D., LEE-RYKACZEWSKI, L., JAMES, I. E., LARK, M. W., DRAKE, F. H., GOWEN, M., GLEASON, J. G.

314

& VEBER, D. F. 1997. Design of potent and selective human cathepsin K inhibitors that span the active site. Proc Natl Acad Sci USA, 94, 14249-54. TJÄDERHANE, L., LARJAVA, H., SORSA, T., UITTO, V. J., LARMAS, M. & SALO, T. 1998. The activation and function of host matrix metalloproteinases in dentin matrix breakdown in caries lesions. J Dent Res, 77, 1622-9. TOLEDANO, M., NIETO-AGUILAR, R., OSORIO, R., CAMPOS, A., OSORIO, E. & TAY, F. R. 2010. Differential expression of matrix metalloproteinase-2 in human coronal and radicular sound and carious dentine. J Dent, 38, 635-40. TOLEDANO, M., OSORIO, R., ALBALADEJO, A., AGUILERA, F. S. & OSORIO, E. 2006. Differential effect of in vitro degradation on resin-dentin bonds produced by self-etch versus total-etch adhesives. J Biomed Mater Res A, 77, 128-35. TOLEDANO, M., PERDIGÃO, J., OSORIO, E. & OSORIO, R. 2002. Influence of NaOCl deproteinization on shear bond strength in function of dentin depth. Am J Dent, 15, 252-5. TORABINEJAD, M. & CHIVIAN, N. 1999. Clinical applications of mineral trioxide aggregate. J Endod, 25, 197-205. TORABINEJAD, M., HONG, C. U., MCDONALD, F. & PITT FORD, T. 1995. Physical and chemical properties of a new root-end filling material. J Endod, 21, 349-53. TORABINEJAD, M., WATSON, T. F. & PITT FORD, T. R. 1993. Sealing ability of a mineral trioxide aggregate when used as a root-end filling material. J Endod, 19, 591-5.

315

TRAUB, W., ARAD, T. & WEINER, S. 1989. Three-dimensional ordered distribution of crystals in turkey tendon collagen fibers. Proc Natl Acad Sci USA, 86, 98226. TSAI, T. R., TSENG, T. Y., CHEN, C. F. & TSAI, T. H. 2002. Identification and determination of geniposide contained in gardenia jasminoides and in two preparations of mixed traditional chinese medicines. J Chromatogr A, 961, 83-8. TÜRKÜN, S. L. 2003. Clinical evaluation of a self-etching and a one-bottle adhesive system at two years. J Dent, 31, 527-34. TURUNEN, T., PELTOLA, J., YLI-URPO, A. & HAPPONEN, R. P. 1994. Effect of bioactive glass granules and polytetrafluoroethylene membrane on repair of cortical defects of rabbit tibia. J Mater Sci Mater Med, 6, 639-41. TYAS, M. J. & BURROW, M. F. 2004. Adhesive restorative materials: a review. Aust Dent J, 49, 112-21. UNEMORI, M., MATSUYA, Y., MATSUYA, S., AKASHI, A. & AKAMINE, A. 2003.

Water

absorption

poly(methyl

methacrylate)

containing

methacryloxyethyl trimellitic anhydride. Biomaterials, 24, 1381-7. UNI EN - 2007. Cement - Part 1: Composition, Specifications And Conformity Criteria For Common Cements. VÄLIMÄKI, V. V., YRJANS, J. J., VUORIO, E. I. & ARO, H. T. 2005. Molecular biological evaluation of bioactive glass microspheres and adjunct bone morphogenetic protein 2 gene transfer in the enhancement of new bone formation. Tissue Eng, 11, 387-94. VAN DIJKEN, J. W., SUNNEGÅRDH-GRÖNBERG, K. & LINDBERG, A. 2007. Clinical long-term retention of etch-and-rinse and self-etch adhesive systems in non-carious cervical lesions. A 13 years evaluation. Dent Mater, 23, 1101-7.

316

VAN GAALEN, S. M., KRUYT, M. C., GEUZE, R. E., DE BRUIJN, J. D., ALBLAS, J. & DHERT, W. J. 2010. Use of fluorochrome labels in in vivo bone tissue engineering research. Tissue Eng Part B Rev, 16, 209-17. VAN HASSEL, H. J. 1971. Physiology of the human dental pulp. Oral Surg Oral Med Oral Pathol, 32, 126-34. VAN LANDINGHAM, M. R., EDULJEE, R. F. & GILLESPIE, J. W. 1999. Moisture diffusion in epoxy systems. J Appl Polym Sci, 71, 787-98. VAN LANDUYT, K. L., DE MUNCK, J., SNAUWAERT, J., COUTINHO, E., POITEVIN, A., YOSHIDA, Y., INOUE, S., PEUMANS, M., SUZUKI, K., LAMBRECHTS, P. & VAN MEERBEEK, B. 2005. Monomer-solvent phase separation in one-step self-etch adhesives. J Dent Res, 84, 183-8. VAN LANDUYT, K. L., SNAUWAERT, J., DE MUNCK, J., PEUMANS, M., YOSHIDA, Y. & POITEVIN, A. 2007. Systematic review of the chemical composition of contemporary dental adhesives. Biomaterials, 28, 3757-85. VAN LANDUYT, K. L., YOSHIDA, Y., HIRATA, I., SNAUWAERT, J., DE MUNCK, J. & OKAZAKI, M. 2008. Influence of the chemical structure of functional monomers on their adhesive performance. J Dent Res, 87, 757-61. VAN MEERBEEK, B., CONN, L. J., DUKE, E. S., EICK, J. D., ROBINSON, S. J. & GUERRERO,

1996.

Correlative

transmission

electron

microscopy

examination of nondemineralized and demineralized resin-dentin interfaces formed by two dentin adhesive systems. J Dent Res, 75, 879-88. VAN MEERBEEK, B., DE MUNCK, J., YOSHIDA, Y., INOUE, S., VARGAS, M., VIJAY, P., VAN LANDUYT, K., LAMBRECHTS, P. & VANHERLE, G. 2003. Buonocore memorial lecture. Adhesion to enamel and dentin: current status and future challenges. Oper Dent, 28, 215-35.

317

VAN MEERBEEK, B., KANUMILLI, P., DE MUNCK, J., VAN LANDUYT, K., LAMBRECHTS, P. & PEUMANS, M. 2005. A randomized controlled study evaluating the effectiveness of a two-step self-etch adhesive with and without selective phosphoric-acid etching of enamel. Dent Mater, 21, 375-83. VAN MEERBEEK, B., PERDIGÃ&#x192;O, J., LAMBRECHTS, P. & VANHERLE, G. 1998. The clinical performance of adhesives. J Dent, 26, 1-20. VAN MEERBEEK, B., PEUMANS, M., POITEVIN, A., MINE, A., VAN ENDE, A., NEVES, A. & DE MUNCK, J. 2010. Relationship between bond-strength tests and clinical outcomes. Dent Mater, 26, 100-21. VAN MEERBEEK, B., WILLEMS, G., CELIS, J. P., ROOS, J. R., BRAEM, M., LAMBRECHTS, P. & VANHERLE, G. 1993. Assessment by nanoindentation of the hardness and elasticity of the resin-dentin bonding area. J Dent Res, 72, 1434-42. VAN MEERBEEK, B., YOSHIHARA, K., YOSHIDA, Y., MINE, A., DE MUNCK, J. & VAN LANDUYT, K. L. 2011. State of the art of self-etch adhesives. Dent Mater, 27, 17-28. VAN NOORT, R. 1994. Clinical relevance of laboratory studies on dental materials: strength determination--a personal view. J Dent, 22, S4-8. VARGAS, M. A., COBB, D. S. & ARMSTRONG, S. R. 1997. Resin-dentin shear bond strength and interfacial ultrastructure with and without a hybrid layer. Oper Dent, 22, 159-66. VERDELIS, K., LUKASHOVA, L., WRIGHT, J. T., MENDELSOHN, R., PETERSON, M. G. E., DOTY, S. & BOSKEY, A. L. 2007. Maturational changes in dentin mineral properties. Bone, 40, 1399-407.

318

VOLLENWEIDER, M., BRUNNER, T. J., KNECHT, S., GRASS, R. N., ZEHNDER, M., IMFELD, T. & STARK, W. J. 2007. Remineralization of human dentin using ultrafine bioactive glass particles. Acta Biomater, 3, 936-43. WAKABAYASHI, Y., KONDOU, Y., SUZUKI, K., YATANI, H. & YAMASHITA, A. 1994. Effect of dissolution of collagen on adhesion to dentin. Int J Prosthodont, 7, 302-6. WANG, L. & NANCOLLAS, G. H. 2009. Pathways to biomineralization and biodemineralization of calcium phosphates: the thermodynamic and kinetic controls. Dalton Trans, 15, 2665-72. WANG, L. J. & NANCOLLAS, G. H. 2008a. Biomineralization. From Nature to Application. Sigel A, Sigel H, Sigel RKO, editors. Metal Ions in Life Sciences Series 4. John Wiley & Sons; Chichester, 413-56. WANG, L. J. & NANCOLLAS, G. H. 2008b. Calcium orthophosphates: crystallization and dissolution. Chem Rev, 108, 4628-69. WANG, Y. & SPENCER, P. 2003. Hybridization efficiency of the adhesive/dentin interface with wet bonding. J Dent Res, 82, 141-5. WANG, Y. & SPENCER, P. 2005. Continuing etching of an all-in-one adhesive in wet dentin tubules. J Dent Res, 84, 350-4. WANG, Z., JIANG, T., SAURO, S., WANG, Y., THOMPSON, I., WATSON, T. F., SA, Y., XING, W., SHEN, Y. & HAAPASALO, M. 2011. Dentine remineralization induced by two bioactive glasses developed for air abrasion purposes. J Dent, 39, 746-56. WANG, Z., JIANG, T., SAURO, S., WANG, Y., XING, W., LIANG, S., SA, Y., ZHANG, C., SHEN, Y. & HAAPASALO, M. 2012. Nerve-targeted

319

desensitizing toothpastes occlude dentin tubules and induce mineral precipitation. Am J Dent, 25, 26-30. WATANABE, I., NAKABAYASHI, N. & PASHLEY, D. H. 1994. Bonding to ground dentin by a phenyl-P self-etching primer. J Dent Res, 73, 1212-20. WATSON, T. F. 1997. Fact and artefact in confocal microscopy. Adv Dent Res, 11, 433-41. WEI, Y. J., SILIKAS, N., ZHANG, Z. T. & WATTS, D. C. 2011. Hygroscopic dimensional changes of self-adhering and new resin-matrix composites during water sorption/desorption cycles. Dent Mater, 27, 259-66. WEINER, S. & WAGNER, H. D. 1998. The material bone: Structure-Mechanical Function Relations. Annu Rev Mater Sci, 28, 271-98. WELCH, K., CAI, Y., ENGQVIST, H. & STRĂ&#x2DC;MME, M. 2010. Dental adhesives with bioactive and on-demand bactericidal properties. Dent Mater, 26, 491-9. WIEGAND, A., BUCHALLA, W. & ATTIN, T. 2007. Review on fluoride-releasing restorative materials--fluoride release and uptake characteristics, antibacterial activity and influence on caries formation. Dent Mater, 23, 343-62. WILSON, A. D. 1991. Glass-ionomer cement--origins, development and future. Clin Mater, 7, 275-82. WILSON, J., PIGOTT, G. H., SCHOEN, F. J. & HENCH, L. L. 1981. Toxicology and biocompatibility of bioglasses. J Biomed Mater Res, 15, 805-17. WONG, T. S., BROUGH, B. & HO, C. M. 2009. Creation of functional micro/nano systems through top-down and bottom-up approaches. Mol Cell Biomech, 1, 155.

320

WU, W. J. & NANCOLLAS, G. H. 1998. The dissolution and growth of sparingly soluble inorganic salts: A kinetics and surface energy approach. Pure Appl Chem, 70, 1867-72. XIA, Y., ZHANG, F. M., XIE, H. F. & GU, N. 2008. Nanoparticle-reinforced resinbased dental composites. J Dent, 36, 450-5. XU, H. H. K., SMITH, D. T., JAHANMIR, S., ROMBERG, E., KELLY, J. R., THOMPSON, V. P. & REKOW, E. D. 1998. Indentation damage and mechanical properties of human enamel and dentin. J Dent Res, 77, 472-80. XU, Z., NEOH, K. G. & KISHEN, A. 2010. A biomimetic strategy to form calcium phosphate crystals on type I collagen substrate. Mater Sci Eng C, 30, 822-6. YAMAUCHI, M. 2000. Collagen biochemistry: an overview. Advances in tissue banking. Vol. 6. Phillips GO, editor. New Jersey: World Scientific Publishing Co, 455-500. YAMAUCHI, M. & SHIIBA, M. 2002. Lysine hydroxylation and crosslinking of collagen. Methods Mol Biol, 194, 277-90 YAMAUTI, M., HASHIMOTO, M., SANO, H., OHNO, H., CARVALHO, R. M., KAGA, M., TAGAMI, J., OGUCHI, H. & KUBOTA, M. 2003. Degradation of resin-dentin bonds using NaOCl storage. Dent Mater, 19, 399-405. YANNAS, I. V. 1992. Tissue regeneration by use of collagen-glycosaminoglycan copolymers. Clin Mater, 9, 179-87. YAP, A. & LEE, C. M. 1997. Water sorption and solubility of resin-modified polyalkenoate cements. J Oral Rehabil, 24, 310-4. YE, Q., PARK, J., TOPP, E. & SPENCER, P. 2009a. Effect of photoinitiators on the in vitro performance of a dentin adhesive exposed to simulated oral environment. Dent Mater, 25, 452-8.

321

YE, Q., WANG, Y. & SPENCER, P. 2009b. Nanophase separation of polymers exposed to simulated bonding conditions. J Biomed Mater Res B Appl Biomater, 88, 339-48. YILDIRIM, S., TOSUN, G., KOYUTÜRK, A. E., SENER, Y., SENGÜN, A. & OZER, F. 2008. Microtensile and microshear bond strength of an antibacterial selfetching system to primary tooth dentin. Eur J Dent, 2, 11-7. YIP, H. K., TAY, F. R., NGO, H. C., SMALES, R. G. & PASHLEY, D. H. 2001. Bonding of contemporary glass ionomer cements to dentin. Dent Mater, 17, 456-70. YIU, C. K., KING, N. M., PASHLEY, D. H., SUH, B. I., CARVALHO, R. M., CARRILHO, M. R. & TAY, F. R. 2004. Effect of resin hydrophilicity and water storage on resin strength. Biomaterials, 25, 5789-96. YOSHIDA, E., HASHIMOTO, M., HORI, M., KAGA, M., SANO, H. & OGUSHI, H. 2004a. Deproteinizing effects on resin-tooth bond structures. J Biomed Mater Res B Appl Biomater, 68, 29-35. YOSHIDA, Y., NAGAKANE, K., FUKUDA, R., NAKAYAMA, Y., OKAZAKI, M. & SHINTANI, H. 2004b. Comparative study on adhesive performance of functional monomers. J Dent Res, 83, 454-8. YOSHIDA, Y., VAN MEERBEEK, B., NAKAYAMA, Y., SNAUWAERT, J., HELLEMANS, L., LAMBRECHTS, P., VANHERLE, G., WAKASA, K. & YOSHIYAMA, M. 2000. Evidence of chemical bonding at biomaterial-hard tissue interfaces. J Dent Res, 79, 709-14. YOSHIDA, Y., VAN MEERBEEK, B., NAKAYAMA, Y., YOSHIOKA, M., SNAUWAERT, J., ABE, Y., LAMBRECHTS, P., VANHERLE, G. &

322

OKAZAKI, M. 2001. Adhesion to and decalcification of hydroxyapatite by carboxylic acids. J Dent Res, 80, 1565-9. YOSHIYAMA, M., SANO, H., EBISU, S., TAGAMI, I., CIUCCHI, B. & CARVALHO, R. M. 1996. Regional tensile bond strengths of cervical sclerotic root dentin. J Dent Res, 75, 1404-13. YOSHIYAMA, M., TAY, F. R., DOI, J., NISHITANI, Y., YAMADA, T., ITOU, K., CARVALHO, R. M., NAKAJIMA, M. & PASHLEY, D. H. 2002. Bonding of self-etch and total-etch adhesives to carious dentin. J Dent Res, 81, 556-60. YOURTEE, D. M., SMITH, R. E., RUSSO, K. A., BURMASTER, S., CANNON, J. M., EICK, J. D. & E.L., K. 2001. The stability of methacrylate biomaterials when enzyme challenged: kinetic and systematic evaluations. J Biomed Mater Res, 57, 522-31. YU, B., AHN, J. S., LIM, J. I. & LEE, Y. K. 2009. Influence of TiO2 nanoparticles on the optical properties of resin composites. Dent Mater, 25, 1142-73. YUAN, H., LI, Y., DE BRUIJN, J. D., DE GROOT, K. & ZHANG, X. 2000. Tissue responses of calcium phosphate cement: a study in dogs. Biomaterials, 21, 128390. ZEHNDER, M., SODERLING, E., SALONEN, J. & WALTIMO, T. 2004. Preliminary evaluation of bioactive glass S53P4 as an endodontic medication in vitro. J Endod, 30, 220-4. ZEIGER, D. N. 2008. The effects of bioactive glass on bonding to dentin. Doctoral Thesis / Dissertation, Chapters 2, 3, and 4, UMI Dissertation Publishing ProQuest University of California, San Francisco.

323

ZHANG, S. C. & KERN, M. 2009. The role of host-derived dentinal matrix metalloproteinases in reducing dentin bonding of resin adhesives. Int J Oral Sci, 1, 163-76. ZHANG, X., NEOH, K. G., LIN, C. C. & KISHEN, A. 2012. Remineralization of partially demineralized dentine substrate based on a biomimetic strategy. J Mater Sci Mater Med, 23, 733-42. ZHENG, L., PEREIRA, P. N., NAKAJIMA, M., SANO, H. & TAGAMI, J. 2001. Relationship between adhesive thickness and microtensile bond strength. Oper Dent, 26, 97-104. ZHONG, J. P., LA TORRE, G. P. & HENCH, L. L. 1994. The kinetics of bioactive ceramics part VII: Binding of collagen to hydroxyapatite and bioactive glass. Bioceramics, 7, 61-6. ZHOU, J., TAN, J., YANG, X., CHENG, C., WANG, X. & CHEN, L. 2010. Effect of chlorhexidine application in a self-etching adhesive on the immediate resindentin bond strength. J Adhes Dent, 12, 27-31. ZICARI, F., COUTHINO, E., DE MUNCK, J., POITEVIN, A., SCOTTI, R. & NAERT, I. 2008. Bonding effectiveness and sealing ability of fiber-post bonding. Dent Mater, 24, 967-77.

324

List of publications in international peer-reviewed journals as a result of this work

Bioactive effects of a calcium/sodium phosphosilicate on the resin-dentine interface: a microtensile bond strength, scanning electron microscopy, and confocal microscopy study. Profeta AC, Mannocci F, Foxton RM, Thompson I, Watson TF, Sauro S. Eur

Oral

Sci.

2012

Aug;120(4):353-62.

doi:

10.1111/j.1600-

0722.2012.00974.x. Epub 2012 Jul 3. PMID: 22813227 [PubMed - indexed for MEDLINE]

Experimental etch-and-rinse adhesives doped with calcium silicate-base microfillers to generate therapeutic bioactive resin-dentin interfaces. Profeta AC, Mannocci F, Foxton RM, Watson TF, Feitosa VP, De Carlo B, Mongiorgi R, ValdrĂŠ G , Sauro S. Dent Mater. 2013 Jul;29(7):729-41. doi: 10.1016/j.dental.2013.04.001. Epub 2013 Apr 29. PMID: 23639454 [PubMed - in process]

325

List of abstracts in international conferences of dental research from this work

Bioactivity and adhesion of an experimental bioactive glass-containing bonding system. Profeta AC, Sauro S, Mannocci F, Foxton RM, Festy F, Watson TF. 89th General Session & Exhibition of the International Association for Dental Research (IADR), San Diego, Calif., USA; March 16-19, 2011

Bioactivity and adhesion of an experimental bioactive glass-containing bonding agent. Profeta AC, Sauro S, Mannocci F, Foxton RM, Thompson I, Watson TF. British Society for Oral and Dental Research (BSODR) Annual Meeting 2011, Sheffield, England, September 13-15, 2011

Therapeutic effects of two experimental etch-and-rinse adhesives containing bioactive micro-fillers. Profeta AC, Mannocci F, Foxton RM, Thompson I, Watson TF, Sauro S. Sixth International Association for Dental Research Pan-European Region Meeting (IADR/PER), Helsinki, Finland, September 12 -15, 2012

326

Appendix

327

! 2012 Eur J Oral Sci

European Journal of Oral Sciences

Bioactive effects of a calcium/sodium phosphosilicate on the resin–dentine interface: a microtensile bond strength, scanning electron microscopy, and confocal microscopy study

Andrea C. Profeta, Francesco Mannocci, Richard M. Foxton Ian Thompson, Timothy F. Watson, Salvatore Sauro* Biomaterials, Biomimetics and Biophotonics Research Group (B3), King’s College London Dental Institute, Guy’s Hospital, London, UK

Profeta AC, Mannocci F, Foxton RM, Thompson I, Watson TF, Sauro S. Bioactive effects of a calcium/sodium phosphosilicate on the resin–dentine interface: a microtensile bond strength, scanning electron and confocal microscopy study. Eur J Oral Sci 2012; 120: 353–362. © 2012 Eur J Oral Sci This study evaluated, through microtensile bond strength (lTBS) testing, the bioactive effects of a calcium/sodium phosphosilicate (BAG) at the resin–dentine interface after 6 months of storage in phosphate buffer solution (PBS). Confocal laser scanning microscopy (CLSM) and scanning electron microscopy (SEM) were also performed. Three bonding protocols were evaluated: (i) RES-Ctr (no use of BAG), (ii) BAG containing adhesive (BAG-AD), and (iii) BAG/H3PO4 before adhesive (BAG-PR). The dentin-bonded specimens were prepared for lTBS testing, which was carried out after 24 h or 6 months of storage in PBS. Scanning electron microscopy ultramorphology analysis was performed after debonding. Confocal laser scanning microscopy was used to evaluate the morphological and nanoleakage changes induced by PBS storage. High lTBS values were achieved in all groups after 24 h of storage in PBS. Subsequent to 6 months of storage in PBS the specimens created using the BAG-AD bonding approach still showed no significant reduction in lTBS. Moreover, specimens created using the BAG-AD or the BAG-PR approach showed an evident reduction of nanoleakage after prolonged storage in PBS. The use of BAG-containing adhesive may enhance the durability of the resin–dentine bonds through therapeutic/protective effects associated with mineral deposition within the bonding interface and a possible interference with collagenolytic enzyme activity (matrix metalloproteinases) responsible for the degradation of the hybrid layer.

Bioactive materials are often used in operative dentistry due to their ability to interact actively with dental hard tissues, inducing calcium-phosphates (Ca/ P) deposition in the presence of body fluids or saliva (1–4). Whereas remineralisation of enamel lesions can be achieved predictably (5, 6), there is little information on whether it is possible to remineralise specific mineraldeficient areas within the resin–dentine interface (i.e. hybrid layers) (2). Some polyalkanoate cements may induce crystal growth within gaps in the bonded interface after long-term storage in water (7). Furthermore, bioactive, ion-releasing materials, such as calcium-phosphate (Ca/P) cements, have the potential to encourage dentine remineralisation by mineral precipitations (8–11).

Salvatore Sauro, Dental Biomaterials Science, King’s College London Dental Institute, Floor 17 Guy’s Tower, London SE1 9RT, UK Telefax: +44-207-1881823 E-mail: salvatore.sauro@kcl.ac.uk Key words: adhesion durability; bioactive glass; bonded-dentine interface Accepted for publication May 2012

PETERS et al. (12) showed the presence of a higher mineral content [determined by electron probe elemental micro-analysis (EPMA) techniques] and an increase in microhardness along the interface of resin-bonded caries-aﬀected dentine, following the application of materials containing Ca/P cements. Bioactive calcium/ sodium (Ca/Na) phosphosilicates, such as Bioglass 45S5 (BAG), are able to induce deposition of hydroxycarbonate apatite (4, 13–15). Although bioactive glasses have previously been used for dentine remineralisation by direct application onto demineralised dentinal tissue when dispersed in water solutions (4, 14), there is little information about the potential therapeutic eﬀects of BAG on the resin–dentine interface when used during etch-and-rinse bonding procedures.

354

Profeta et al.

Therefore, this study was devised to assess the bioactive eﬀects of BAG during etch-and-rinse dentine-bonding procedures on the resin–dentine interface. This aim was accomplished by evaluating the microtensile bond strength (lTBS) of specimens after 24 h and 6 months of storage in PBS. Fractographic analysis was also performed through scanning electron microscopy (SEM). The ultramorphology and nanoleakage analysis of the resin-bonded dentine was executed using confocal laser scanning microscopy (CLSM). The null hypotheses to be tested in this study were: (i) the use of BAG employed during bonding procedures has no eﬀect on the bond strength, and (ii) the presence of BAG does not reduce nanoleakage within the demineralised ‘poorly-infiltrated’ areas within the resin-dentine interface.

Material and methods Specimen preparation Caries-free human third molars, extracted for surgical reasons from 20- to 40-yr-old patients, were used in this study. The treatment plan of any of the involved patients, who had given informed consent for use of their extracted teeth for research purposes, was not altered by this study. The study was conducted in accordance with the ethical guidelines of the Research Ethics Committee (REC) for medical investigations. The teeth were stored in deionised water (pH 7.1) at 4° C and used within 1 month after extraction. The coronal dentine specimens were prepared by sectioning the roots 1 mm beneath the cemento–enamel junction (CEJ) with a hard tissue microtome (Accutom-50; Struers, Copenhagen, Denmark) using a slow-speed, water-cooled diamond wafering saw (330-CA RS-70300; Struers) (Fig. 1). A 180grit silicon carbide (SiC) abrasive paper mounted on a water-cooled rotating polishing machine (Buehler MetaServ 3000 Grinder-Polisher; Buehler, Du¨sseldorf, Germany) was used (30 s) to remove the diamond saw smear layer and to replace it with a standard and more clinically relevant smear layer (16). Experimental bonding procedures and formulation of resin adhesives A resin co-monomer blend was formulated by using a hydrophobic, cross-linking dimethacrylate monomer – bisphenyl-A-glycidyl methacrylate (Bis-GMA; Esstech, Essington, PA, USA) – and a hydrophilic monomer – 2-hydroxyethyl methacrylate (HEMA; Sigma-Aldrich, Gillingham, UK). In order to obtain a dental bonding system with chemical aﬃnity to calcium (present in dentine and BAG), an acidic functional monomer – 2,5dimethacryloyloxyethyloxycarbonyl-1,4-benzenedicarboxylic acid (PMDM; Esstech Essington) – was also included within the composition of the resin blend. Subsequently, the resin blend was made light-curable by a binary photoinitiator system based on camphoroquinone (CQ; Sigma-Aldrich) and 1,2-ethyl-dimethyl-4-aminobenzoate (EDAB; Sigma-Aldrich). This resin co-monomer blend was used to formulate the experimental primer and the bonds used in this study (Table 1).

Fig. 1. Schematic illustrating the experimental study design. Human third molars were used to prepare standardised dentine surfaces. The three diﬀerent bonding approaches were performed using specific components and application procedures. Bis-GMA, bisphenyl-A-glycidyl methacrylate; HEMA, 2-hydroxyethyl methacrylate; PMDM, 2,5-dimethacryloyloxyethyloxycarbonyl-1,4-benzenedicarboxylic acid.

A BAG (Sylc; OSspray, London, UK) with particle size < 10 lm was employed in the etch-and-rinse bonding procedures using two diﬀerent experimental approaches: (i) BAG-AD (30wt% BAG included within the composition of a resin adhesive as a bioactive microfiller), and (ii) BAG-PR (BAG applied directly onto H3PO4-etched/ wetted dentine before bonding procedures). The neat adhesive, with no BAG, served as the control (RES-Ctr) (Fig. 1). In detail, a water wet-bonding dentine substrate was achieved by water-rinsing, for 15 s, the dentine surfaces acid-etched with 37% phosphoric acid solution (H3PO4) (Sigma-Aldrich) and gently blowing oﬀ (for 2 s) excess water to leave a wet reflective-surface. The control bonding procedure (RES-Ctr) was accomplished by applying two consecutive coats of an ethanolsolvated resin primer [50 wt% absolute ethanol (SigmaAldrich) and 50 wt% of neat co-monomer resin blend] and a layer of the neat co-monomer resin blend (Table 1) within a period of 20 s. Light-curing was immediately performed for 30 s (>600 mW/cm!2, Optilux VLC; Demetron, Danbury, CT, USA). The first experimental bonding procedure (BAG-AD) was performed by applying the same ethanol-solvated resin primer onto H3PO4-etched dentine, as previously described, followed by a layer of bonding resin containing BAG (Table 1; Fig. 1). The bonding and the light-curing procedures were executed as previously described for the RES-Ctr group. The second experimental bonding procedure (BAG-PR) was performed as follows. The 37% H3PO4 solution (Sigma-Aldrich). was applied onto the dentine surface for 15 s. Then, 0.05 g of BAG powder was placed onto the H3PO4-etched wet dentine surface, spread immediately for 10 s using a cotton pellet, and finally rinsed with copious amounts of deionised water for 15 s (Fig. 1). The primer/ bond application and the light-curing procedures were performed as previously described for the RES-Ctr group. A final composite build-up (5 mm) was constructed on each specimen using a light-cured resin composite (Filtek

Eﬀects of a bioactive glass on the resin–dentine interface

355

Table 1 Composition of the experimental bonding procedures/adhesive systems used in this study

Experimental bonding procedures

Chemical composition (wt%) of the adhesive systems Dentine conditioning

Primer

Bond

BAG-AD

37% H3PO4 solution, 15 s immediately followed by the application of a 0.05 g of BAG on H3PO4 wet dentine

20.21 wt% BisGMA 15.54 wt% PMDM 14.25 wt% HEMA 50.00 wt% Absolute ethanol

37.50 wt% BisGMA 16.80 wt% PMDM 15.70 wt% HEMA 30.00 wt% BAG

BAG-PR

37% phosphoric acid solution, 15 s – H3PO4:

20.21 wt% BisGMA 15.54 wt% PMDM 14.25 wt% HEMA 50.00 wt% Absolute ethanol

40.00 wt% Bis-GMA 31.50 wt% PMDM 28.50 wt% HEMA

RES-Ctr

37% phosphoric acid solution, 15 s – H3PO4:

20.21 wt% BisGMA 15.54 wt% PMDM 14.25 wt% HEMA 50.00 wt% Absolute ethanol

40.00 wt% Bis-GMA 31.50 wt% PMDM 28.50 wt% HEMA

Bis-GMA, bisphenyl A glycidyl methacrylate; HEMA, hydrophilic 2-hydroxyethyl methacrylate; PMDM, 2,5-dimethacryloyloxyethyloxycarbonyl-1,4-benzenedicarboxylic acid. At the end of the formulation of the resins, 0.25 wt% camphoroquinone (CQ) and 1.0 wt% 2-ethyl-dimethyl-4-aminobenzoate (EDAB) were added to the resin mixture.

Z250; 3M-ESPE, St Paul, MN, USA) in five incremental layers (of 1 mm thickness). Each layer of composite was individually light cured for 20 s. The resin-bonded dentine specimens were stored in PBS for 24 h or 6 months at 37° C. The PBS was composed of (in g/l) CaCl2 (0.103), MgCl2.6H2O (0.019), KH2PO4 (0.544), KCl (17), and HEPES (acid) buﬀer (4.77), and the pH was 7.4. lTBS and SEM fractography and failure analysis Twenty dentine-bonded specimens from each group were sectioned using a slow-speed water-cooled diamond wafering blade (Struers) mounted on a hard-tissue microtome (Isomet 11/1180; Buehler) in both x and y directions across the adhesive interface to obtain matchsticks with cross-sectional areas of 0.9 mm2. By excluding peripheral beams showing the presence of residual enamel, only the remaining matchsticks (n = 10–15) were selected to create three groups with the same total number of resin–dentine specimens in each group (n = 280). The exact width of each matchstick was checked using a calliper (Mitutoyo CD15; Mitutoyo, Kawasaki, Japan) and half of them (n = 140) were tested after 24 h of storage in PBS and the remaining half (n = 140) were tested after 6 months of storage in PBS at 37°C. The lTBS test was performed using a microtensile jig in a LAL300 linear actuator (SMAC Europe; Horsham, UK) with a LAC-1 high-speed controller single axis with a built-in amplifier and at the following settings: stroke length = 50 mm, peak force = 250 N, displacement resolution = 0.5 mm, and crosshead speed = 1 mm min!1. Bond-strength data were calculated and expressed in MPa, the lTBS values of sticks from the same restored teeth were averaged, and the mean bond strength was used as one statistical unit for the statistical analysis. The lTBS (mean-MPa) data for

each group were analysed using a repeated-measures ANOand Tukey’s post-hoc test for pairwise comparisons (a = 0.05). The mode of failure was classified as percentage of adhesive, mixed, or cohesive. The failed bonds were examined at 930 magnification using a stereoscopic microscope (Leica M205A; Leica Microsystems, Wetzlar, Germany). Five representative debonded specimens for each group that failed in mixed or adhesive modes were selected for ultramorphology analysis of the fractured surface (SEM Fractography). They were dried overnight and mounted on aluminium stubs with carbon cement, then sputter-coated with gold (SCD 004 Sputter Coater; Bal-Tec, Vaduz, Liechtenstein) and examined using SEM (S-3500; Hitachi, Wokingham, UK) with an accelerating voltage of 15 kV and a working distance of 25 mm at increasing magnifications. VA

Confocal microscopy ultramorphology and nanoleakage evaluation A further three dentine specimens from each group were bonded, as previously described, with the primer/bond resins doped with 0.1 wt% rhodamine-B (Rh-B: SigmaAldrich, St Louis, MO, USA) and employed for the confocal microscopy analysis (18, 19). The specimens were serially sectioned across the adhesive interface to obtain resin–dentine slabs (of 1 mm thickness). The resin–dentine slabs (n = 10 per group) were then divided into two subgroups based on the period of storage in PBS (24 h or 6 months) (Fig. 2). Subsequent to the storage period, the specimens were coated with two layers of fast-setting nail varnish applied 1 mm from the resin–dentine interfaces and immersed in 1 wt% aqueous fluorescein (SigmaAldrich) solution for 24 h. The specimens were then treated in an ultrasonic water bath for 2 min and polished

356

Profeta et al.

using SiC abrasive papers of ascending grit (#1200 to #4000) (Versocit; Struers) on a water-cooled rotating polishing machine (Buehler Meta-Serv 3000 Grinder-Polisher; Buehler). A final treatment in an ultrasonic water bath (5 min) completed the specimen preparation for the confocal microscopy evaluation (Fig. 2). The microscopy examination was performed using a confocal laser scanning microscope (Leica SP2 CLSM; Leica, Heidelberg, Germany) equipped with a 63 9 /1.4 NA oil-immersion lens and using 488-nm argon/helium (fluorescein excitation) or 568-nm krypton (rhodamine excitation) laser illumination. The reflection imaging was performed using the argon/helium laser. Confocal laser scanning microscopy reflection and fluorescence images were obtained with a 1-lm z-step to section optically the specimens to a depth up to 20 lm below the surface (18). The z-axis scans of the interface surface were arbitrarily pseudo-coloured by two selected operators and compiled into single projections using the Leica image-processing software (Leica). The configuration of the system was standardised and used at the same settings for the entire investigation. Each resin–dentine interface was completely investigated and then five optical images were randomly captured. Micrographs representing the most common features of nanoleakage observed along the bonded interfaces were captured and recorded (19).

Results lTBS and SEM fractography and failure analysis

The BAG-bonding technique vs. storage time was statistically significant only for the BAG-AD group (P = 0.001); no significant reduction of the lTBS values was observed after 6 months of storage in PBS (P > 0.05). On the other hand, significant lTBS reductions were observed in both the BAG-PR and RES-Ctr groups (P < 0.05) after prolonged storage in PBS (6 months). The lTBS results (expressed as Mean and SD) are presented in Table 2. High lTBS values were achieved in all groups after 24 h of storage in PBS, with failures occurring mainly in cohesive mode in all groups; in contrast, important changes in the lTBS were observed after 6 months of storage in PBS. For instance, the lTBS of specimens in the RES-Ctr group (no BAG) showed a significant (P < 0.05) decrease after 6 months of storage in PBS and failed mostly in adhesive mode (66%). The specimens stored for 24 h in PBS that fractured in mixed mode were characterised by the presence of exposed dentinal tubules with spare extruded resin tags (Fig. 3A2). Conversely, the surface of the specimens that failed in adhesive mode after 6 months of storage in PBS presented several ‘funnelled’ dentinal tubules with no exposed collagen fibrils (Fig. 3A3). The resin– dentine specimens of the BAG-AD group maintained a high lTBS (P > 0.05) after 6 months of storage in PBS (23.89 ± 7.74 MPa). In these specimens the failure was prevalent in cohesive (43%) and mixed (40%) modes (Fig. 3B1) and the SEM analysis of the fractured surface revealed a dentine surface predominantly covered by residual resin and mineral crystals embedded within a resin/collagen network (Fig. 3B3). The specimens of the BAG-PR group, where the BAG powder was applied onto acid-etched/wetted Table 2 Mean and standard deviation (SD) of microtensile bond strength values (MPa) obtained for the different experimental groups and percentage distribution of failure mode after microtensile bond strength testing; total number of beams (tested stick/pre-load failure) lTBS – mean ± SD (N of tested/pre-failed beams) % Failure [A/M/C] 24 h test 6 month test BAG-AD

BAG-PR

Fig. 2. Schematic illustrating the composite-tooth matchsticks (1 mm) prepared using a water-cooled diamond saw, stored in PBS for 24 h or 6 months, and then subjected to microtensile bond strength (lTBS) testing and scanning electron microscopy failure analysis. This schematic also illustrates how composite-tooth slabs were prepared, stored in PBS for 24 h or 6 months, and evaluated by confocal laser scanning microscopy.

RES-Ctr

26.91 ± 3.43 (140/0) [0/10/90] A1 27.20 ± 3.92 (140/0) [0/9/91] A1 29.12 ± 4.75 (140/0) [0/4/96]

B1 23.89 ± 7.75 (135/5) [17/40/43] A2 13.35 ± 5.32 (134/6) [56/11/33] A2 18.18 ± 5.66 (133/7) [66/10/24]

For each horizontal row: values with identical numbers indicate no significant diﬀerence. For each vertical column: values with identical letters indicate no significant diﬀerence using Student-Newman–Keuls test (P > 0.05).

Eﬀects of a bioactive glass on the resin–dentine interface A1

357

Fig. 3. Scanning electron microscopy images of failure modes of the resin-bonded specimens created using the three diﬀerent bonding approaches tested. (A) Micrograph of the failure mode (cohesive) of the resin control bonded to etched dentine (37% H3PO4) after 24 h of storage in PBS (A1). At higher magnification (A2) it was possible to observe the presence of some exposed dentinal tubules, but most remained obliterated by resin tags. No exposed collagen fibrils were visible on the dentine surface, and a well resin-hybridised hybrid layer was present (pointer). At 6 months (A3), the resin–dentine interfaces created with the control resin (RES-Ctr; containing no bioactive filler) showed only a few resin tags inside the dentinal tubules and no collagen fibrils were visible on a dentine surface characterised by funnelled dentinal tubules (pointer). (B) Micrograph of the failure mode (mixed) of the calcium/sodium phosphosilicate-containing adhesive (BAG-AD) bonded to dentine, after 6 months of storage in PBS (B1). At higher magnification (B2) no exposed dentinal tubules or exposed collagen fibrils were observed; the dentine surface was well resin-hybridised (pointer). After 6 months of storage in PBS (B3), the debonded resin–dentine interface showed the presence of resin tags remaining inside the dentinal tubules and mineral crystals embedded within a preserved collagen network (pointer). (C) Micrograph of the failure mode (adhesive) of BAG applied directly onto H3PO4-etched/wetted dentine before bonding (BAGPR) after 24 h of storage in PBS (C1). At higher magnification (C2) it was possible to observe the presence of some exposed dentinal tubules, while most remained obliterated by resin tags containing few BAG particles. No exposed collagen fibrils were present on the dentine surface (pointer). At 6 months testing (C3), the resin–dentine interface created with the BAG-PR showed a dentine surface characterised by the presence of remineralised dentinal tubules obliterated by mineral crystals. It is interesting to note how the fracture occurred along the intertubular dentine leaving an intact peritubular dentine around the mineral-obliterated dentinal tubule (pointer). rt, resin tags; t, dentinal tubules.

dentine before application of the adhesive system, showed a significant decrease in lTBS values (P < 0.05) after prolonged storage in PBS (table 2). These specimens failed mainly in adhesive mode (56%) after 6 months of storage in PBS, and the SEM fractographic analysis showed that the fracture during lTBS testing occurred along the intertubular dentine, leaving an intact peritubular dentine and a consistent precipitation of mineral inside the dentinal tubules (Fig. 3C3). Confocal microscopy ultramorphology and nanoleakage evaluation

The CLSM investigation showed that all the bonding procedures used in this study were able to create a

resin diﬀusion within the demineralised dentine (hybrid layer 7–9 lm) and several resin tags into the dentinal tubules (Fig. 4A). Nevertheless, the resin–dentine interfaces of the specimens created in the three groups showed evident fluorescein penetration (nanoleakage) within the hybrid layer and along the dentinal tubules after 24 h of storage in PBS (Fig. 4B,C). The experimental bonding approach used to bond the specimens of the BAG-PR group created resin–dentine interfaces characterised by the presence of mineral deposits inside the dentinal tubules and within the hybrid layer (Fig. 4D). The prolonged storage in PBS induced important changes in terms of ultramorphology and nanoleakage. For instance, the resin–dentine interface of the RES-

358

Profeta et al.

Ctr group specimens was aﬀected by severe nanoleakage within the hybrid layer and the presence of a continuous gap between dentine and composite (Fig. 5A). Conversely, the specimens of the BAG-AD group showed the presence of a strong reflective mineral material and partial dye penetration within the hybrid layer (Fig. 5B). The resin–dentine interface of specimens in the BAG-PR group was aﬀected by partial dye penetration within a crystallised hybrid layer. However, gaps were also observed between the hybrid and adhesive layers (Fig. 5C), probably caused by the sample preparation procedure before the CLSM analysis.

Discussion Hybrid layers created using etch-and-rinse adhesives include water-rich, resin-sparse regions that account for 2–3% of their entire volume, which increase subsequent

to prolonged aging in fluids (20). The water-rich, resinsparse regions represent essentially the nanoporosities within the demineralised collagen fibrils, created during adhesive application as a result of incomplete replacement of water by resin infiltration (21). This undisplaced water may act as a functional medium for the hydrolysis of suboptimally polymerised resin matrices by esterases and denaturation of collagen via the activation of host-derived matrix metalloproteinases (MMPs), jeopardizing the durability of the resin– dentine interfaces (21–23). Several methods have been advocated to increase the longevity of these resin–dentine interfaces, including the inhibition of the MMPs within the hybrid layer (22, 23) and enhancement of the resin infiltration within the demineralised collagen fibril using more hydrophobic resin monomers and ethanol wet-bonding (21). Based on the results obtained in this study, the first null hypothesis must be partially rejected because the

Fig. 4. Confocal laser scanning microscopy (CLSM) images showing the interfacial characterisation and nanoleakage, after 24 h of storage in PBS, of the resin–dentine interfaces created using the three diﬀerent bonding approaches tested. (A) Confocal laser scanning microscopy three-dimensional (3D) single-projection (fluorescence mode) image exemplifying the interfacial characteristics of the resin–dentine interface created using the control adhesive system (RES-Ctr) applied onto H3PO4-etched dentine. It is possible to observe a clear hybrid layer (approximate thickness 9 lm) located underneath a thick adhesive layer and long resin tags. (B) This CLSM 3D single-projection (fluorescence/reflection mode) image of the resin–dentine interface created using the bioactive calcium/sodium phosphosilicate-containing adhesive (BAG-AD) shows an intense nanoleakage signal from the hybrid layer (pointer) located underneath a thick adhesive layer characterised by the presence of BAG microfiller. The presence of long resin tags is also evident. (C) The resin–dentine interface created using the bonding procedure where the BAG is applied directly onto H3PO4-etched/wetted dentine (BAG-PR) shows evident dye penetration within the hybrid layer (pointer). Short resin tags are visible underneath a thick adhesive layer. The reason why only short resin tags could be created during this type of bonding procedure is shown in (D) where it is possible to observe a strong reflective signal from the demineralised dentine layer (pointer) and inside the dentinal tubules, indicating the presence of mineral particles. a, adhesive layer; c, composite; fl, BAG microfiller; rt, resin tags; t, dentinal tubules.

Eﬀects of a bioactive glass on the resin–dentine interface A

359

Fig. 5. Confocal laser scanning microscopy (CLSM) images showing the interfacial characterisation and nanoleakage, after 6 months of storage in PBS, of the resin–dentine interfaces. (A) Confocal laser scanning microscopy three-dimensional single-projection (fluorescence/reflection mode) image of the resin–dentine interface created using the control adhesive system (RES-Ctr) applied onto H3PO4-etched dentine. It is possible to note the presence of evident dye diﬀusion (nanoleakage) within the hybrid layer and inside the dentinal tubules (pointer). A gap is present between the dentine and the composite. (B) The resin–dentine interface created using the bonding approach where the bioactive calcium/sodium phosphosilicate-containing adhesive (BAG-AD) is applied onto H3PO4-etched dentine shows partial dye diﬀusion within a hybrid layer characterised by a strong reflective signal (pointer). (C) The resin–dentine interface created using the bonding procedure where the BAG is applied directly onto H3PO4etched/wet dentine (BAG-PR) shows a crystallised reflective layer (pointer) characterised by low dye penetration (nanoleakage). A pronounced gap can be seen between the adhesive layer and the composite. It is also possible to observe the remaining reflective mineral materials on the fractured edge of the adhesive layer (arrows). a, adhesive layer; c, composite; g, gap.

use of BAG produced bioactive/protective eﬀects on the bond strength only when used as resin microfiller within the adhesive composition. The second null hypothesis must be totally rejected as both the experimental bonding approaches based on the use of BAG were able to reduce the nanoleakage within the demineralised ‘poorly infiltrated’ areas within the resin–dentine interface. In detail, the control bonding procedure (RES-Ctr) and the two experimental bonding approaches used (BAG-AD and BAG-PR) to bond the acid-etched dentine produced comparably high lTBS values after 24 h of storage in PBS (Table 2). Conversely, a significant decrease in lTBS (P > 0.05) occurred in all groups after storage in PBS for 6 months, except for the specimens bonded using the resin adhesive containing BAG microfiller (BAG-AD). The SEM analysis of the fractured specimens of the RES-Ctr group showed, after 24 h of storage in PBS, a dentine surface characterised by a hybrid layer that was highly hybridised with resin and no presence of demineralised collagen fibrils exposed (Fig. 3A2). Conversely, these resin–dentine specimens stored for 6 months in PBS had a fractured surface characterised by ‘funnelled’ dentinal tubules, indicating degradation of the demineralised peritubular dentine (Fig. 3A3). In contrast, the bonded-dentine specimens of the BAG-AD group immersed in PBS for 6 months had a fractured (adhesive mode) dentine surface, with mineral crystals embedded within a preserved collagen network and no evidence of ‘funnelled’ dentinal tubules (Fig. 3B3). The SEM ultramorphology analysis of the fractured specimens (adhesive mode) of the RES-PR group stored for 24 h in PBS demonstrated the presence of dentinal tubules obliterated by resin tags and no exposed collagen fibrils (Fig. 3C2). Interestingly, when

this type of dentine-bonded specimen was immersed in PBS for 6 months it was possible to detect a fractured dentine surface characterised by dentinal tubules obliterated by mineral crystals and a distinctive fracture along the intertubular dentine, which left an intact peritubular dentine (Fig. 3C3). Possible explanations for such longevity attained in dentine-bonded specimens created using BAG-AD after prolonged storage in PBS may be as follows: (i) The presence of BAG within the resin–dentine interface may have induced the release of a silicic acid, such as Si(OH)4, and a subsequent polycondensation reaction between the silanols compounds and the demineralised collagen via electrostatic, ionic, and/or hydrogen bonding (13, 24, 25), which interfered with the ability of MMPs – although BAG is not a direct MMP inhibitor – to execute their collagenolytic and gelatinolytic activities. A study by OSORIO et al. (26) showed that it is possible to reduce the collagen-degradation process by using specific chemical compounds, such as zinc oxide, which interfere with the zinc-binding and calcium-binding catalytic domains of MMPs. (ii) The precipitation of an amorphous calcium phosphate (ACP) on the polycondensate SiO2-rich template of nucleation (3, 5, 13, 18) induced by the dissolution and immediate reaction between Ca2+ and PO43! species from BAG may have also favoured the formation of a high-molecular-weight complex (Ca/P–MMPs), which restricted the activities of MMP-2 and MMP-9 within the hybrid layer (27). However, the ability of specific bioactive glass, such as Bioglass 45S5, to modulate and/or reduce the presence of collagens I, II, and III, osteocalcin, osteonectin, and osteopontin, has

360

Profeta et al.

also been demonstrated in bone-regeneration studies (24). (iii) The release of Na+ and Ca2+ ions from BAG, and the incorporation of H3O+ protons into the glass particles, may have created an optimal alkaline environment (5, 18) within the resin–dentine interface that interfered with the activity of MMPs, which are very acidic-pH dependent (22, 23). (iv) The bioactive remineralisation induced by BAG may have decreased the distribution of the waterrich, resin-sparse regions within the hybrid layer (2, 18) via silanols polycondensation and subsequent ACP/AH remineralisation, which probably interfered with the water-dependent hygroscopic and hydrolytic degradation of the polymer network (28). The confocal microscopy evaluation performed after 6 months of storage in PBS indicated that both the experimental bonding approaches used in this study (BAG-AD and BAG-PR) created a resin–dentine interface aﬀected only by partial dye penetration (nanoleakage) within a hybrid layer characterised by the deposition of a strong reflective mineral (Fig. 5B,C). Whereas it is reasonable to believe that the hybrid layer of the specimens created using the BAG-AD approach remineralised as a result of the bioactive/biomimetic activity of Bioglass 45S5 after prolonged (6 months) storage in PBS (17, 18, 25), a completely diﬀerent bioactive phenomenon may have occurred within the resin–dentine interface created by directly applying the BAG on the demineralised H3PO4-wetted dentine, as a significant decrease of lTBS was attained after prolonged storage in PBS (Table 2). In this case, a possible explanation for the reduced confocal nanoleakage may be due to the chemical nature of mineral precipitation that occurred within the resin–dentine interface created as a result of the experimental bonding procedures (BAG-PR). Our hypothesis is that the chemical reaction between BAG and H3PO4 solution (Fig. 4D) may have induced the precipitation of dicalcium-phosphate salts (i.e. brushite and monetite). BAKRY et al. (17, 29) showed that the acid–base chemical reaction between BAG and H3PO4 may induce the formation of brushite via combination of the phosphate (released from the BAG and H3PO4) and calcium ions (released from BAG and etched dentine). The precipitation reaction of the brushite may be responsible for the creation of an acidic environment (30), which may have evoked the activation of MMPs (22, 23); this situation is also aggravated by the fact that BAG no longer has the ability to create a localised ‘protective’ alkaline pH within the resin–dentine interface. Moreover, it is also possible that the BAG/H3PO4 reaction may have altered the chemical and/or physical characteristics of BAG, in particular those responsible for the polycondensation of silanols and ACP/HA precipitation (13, 24, 25), which may be fundamental in altering the activity of MMPs (27), as previously

described. However, even supposing that the reaction between hydroxyl ions and Si(OH)4 formed the silanols compounds and induced the polycondensation reaction, they may have been washed out by application of the air-water jet before application of the primer and bond (31). Furthermore, a slightly acidic environment may have remained in loco within the resin–dentine interface during the prolonged storage in PBS as a result of the release of H+ from the acidic monomer (2,5-dimethacryloyloxyethyloxycarbonyl-1,4-benzenedicarboxylic acid) contained within the resin adhesives (32–34), causing a long-standing, MMP-mediated degradation of collagen in both the RES-Ctr and BAG-PR groups. In addition, a durable acidic environment may have induced supplementary precipitation of dicalcium or octocalcium phosphates (30, 35) during buffered condition (replacement of PBS) within the microporosities generated by the degradation of the dentine collagen fibrils (Fig. 5C). Indeed, as a result of this probable additional precipitation of mineral over time, the interface created using the BAG-PR bonding technique may have achieved mechanical characteristics similar to those created using glass ionomer cements (GICs) applied onto polyacrylic acid-etched dentine (18, 36, 37). This is probably why bond strength reduction and gap formation were observed in the BAG-PR specimens. The GIC-bonded interfaces can reach a tensile or shear bond strength of approximately 5 MPa and frequently prefail during specimen preparation (37, 38). YIP et al. (39) affirmed that the results obtained from tensile testing of GICs bonded to dentine do not represent the actual strength of such stiff bonded interfaces and that only an accurate ultramorphology analysis using electron microscopy may reveal the proper bonding ability of such restorative materials. However, it is also important to consider that the hydrophilic characteristics conferred by specific resin monomers, such as HEMA and PMDM, within the tested adhesives (Fig. 1) may have compromised the mechanical properties (i.e. modulus of elasticity) of the hybrid layers (40, 41) as a result of polymer hydrolysis and swelling tensions generated within the polymer chains. In contrast, the BAG microfiller contained within the adhesive used in the BAG-AD group may have absorbed and used the water not required by the hydrophilic/acid monomers for the bioactive processes of conversion into apatite (18), thus preventing the polymer network from considerable hygroscopic/hydrolytic degradation (28). In conclusion, this study provided preliminary evidence for the use of bioactive Ca/Na phosphosilicate, such as Bioglass 45S5, in dentine-bonding procedures in order to enhance the durability of the resin–dentine interfaces. However, further in vitro (i.e. transmission electron microscopy and atomic force microscopynanoindentation examination) and long-term clinical studies are required to confirm the protective/therapeutic effects of BAG on the resin-dentine interface. Confocal Raman analysis will be also necessary to confirm the chemical nature of the mineral precipitates observed

Eﬀects of a bioactive glass on the resin–dentine interface

within the bonded-dentine interfaces created with the two experimental BAG-bonding procedures. Acknowledgements – This article presents independent research commissioned by the National Institute for Health Research (NIHR) under the i4i programme and the Comprehensive Biomedical Research Centre at Guy’s & St Thomas’ Trust. The views expressed in this publication are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health. The authors also acknowledge support from the Centre of Excellence in Medical Engineering funded by the Wellcome Trust. Conflicts of interest – None.

References 1. KIM YK, GU LS, BRYAN TE, KIM JR, CHEN L, LIU Y, YOON JC, BRESCHI L, PASHLEY DH, TAY FR. Mineralisation of reconstituted collagen using polyvinylphosphonic acid/polyacrylic acid templating matrix protein analogues in the presence of calcium, phosphate and hydroxyl ions. Biomaterials 2010; 31: 6618–6627. 2. RYOU H, NIU LN, DAI L, PUCCI CR, AROLA DD, PASHLEY DH, TAY FR. Effect of biomimetic remineralization on the dynamic nanomechanical properties of dentin hybrid layers. J Dent Res 2011; 90: 1122–1128. 3. HENCH LL, ANDERSSON O¨. Bioactive glasses. In: Wilson J, ed. Introduction to bioceramics. Singapore: World Scientific;1993; 45–47. 4. SAURO S, THOMPSON I, WATSON TF. Effects of common dental materials used in preventive or operative dentistry on dentin permeability and remineralization. Oper Dent 2011; 36: 222–230. 5. SHEN P, CAI F, NOWICKI A, VINCENT J, REYNOLDS EC. Remineralization of enamel subsurface lesions by sugar-free chewing gum containing casein phosphopeptide-amorphous calcium phosphate. J Dent Res 2001; 80: 2066–2070. 6. LAGERWEIJ MD, TEN CATE JM. Remineralisation of enamel lesions with daily applications of a high-concentration fluoride gel and a fluoridated toothpaste: an in situ study. Caries Res 2002; 36: 270–274. 7. HASHIMOTO M, NAKAMURA K, KAGA M, YAWAKA Y. Crystal growth by fluoridated adhesive resins. Dent Mater 2008; 24: 457–463. 8. NGO HC, MOUNT G, MCINTYRE J, TUISUVA J, VON DOUSSA RJ. Chemical exchange between glass-ionomer restorations and residual carious dentine in permanent molars: an in vivo study. J Dent 2006; 34: 608–613. 9. DICKENS SH, FLAIM GM, TAKAGI S. Mechanical properties and biochemical activity of remineralizing resin-based Ca-PO4 cements. Dent Mater 2003; 19: 558–566. 10. DICKENS SH, FLAIM GM. Effect of a bonding agent on in vitro biochemical activities of remineralizing resin-based calcium phosphate cements. Dent Mater 2008; 24: 1273– 1280. 11. ARENDS J, RUBEN JL, INABA D. Major topics in quantitative microradiography of enamel and dentin: R parameter, mineral distribution visualization, and hyper-remineralization. Adv Dent Res 1997; 11: 403–414. 12. PETERS MC, BRESCIANI E, BARATA TJ, FAGUNDES TC, NAVARRO RL, NAVARRO MF, DICKENS SH. In vivo dentin remineralization by calcium-phosphate cement. J Dent Res 2010; 89: 286–291. 13. VOLLENWEIDER M, BRUNNER TJ, KNECHT S, GRASS RN, ZEHNDER M, IMFELD T, STARK WJ. Remineralization of human dentin using ultrafine bioactive glass particles. Acta Biomater 2007; 3: 936–943. 14. EFFLANDT SE, MAGNE P, DOUGLAS WH, FRANCIS LF. Interaction between bioactive glasses and human dentin. J Mater Sci Mater Med 2002; 13: 557–565.

361

15. HENCH LL, PASCHALL HA. Direct chemical bond of bioactive glass-ceramic materials to bone and muscle. J Biomed Mater Res 1973; 7: 25–42. 16. OLIVEIRA SS, PUGACH MK, HILTON JF, WATANABE LG, MARSHALL SJ, MARSHALL GW Jr. The influence of the dentin smear layer on adhesion: a self-etching primer vs. a total-etch system. Dent Mater 2003; 19: 758–767. 17. BAKRY AS, TAKAHASHI H, OTSUKI M, SADR A, YAMASHITA K, TAGAMI J. CO2 laser improves 45S5 bioglass interaction with dentin. J Dent Res 2011; 90: 246–250. 18. SAURO S, OSORIO R, WATSON TF, TOLEDANO M. Therapeutic effects of novel resin bonding systems containing bioactive glasses on mineral-depleted areas within the bonded-dentine interface. J Mater Sci Mater Med 2012; 23: 1521–1532. 19. SAURO S, WATSON TF, THOMPSON I, TOLEDANO M, NUCCI C, BANERJEE A. Influence of air-abrasion executed with polyacrylic acid-Bioglass 45S5 on the bonding performance of a resinmodified glass ionomer cement. Eur J Oral Sci 2012; 120: 168 –177. 20. REIS A, GRANDE RH, OLIVEIRA GM, LOPES GC, LOGUERCIO AD. A 2-year evaluation of moisture on microtensile bond strength and nanoleakage. Dent Mater 2007; 23: 862–870. 21. PASHLEY DH, TAY FR, BRESCHI L, TJA¨DERHANE L, CARVALHO RM, CARRILHO M, TEZVERGIL-MUTLUAY A. State of the art etch-and-rinse adhesives. Dent Mater 2011; 27: 1–16. 22. PASHLEY DH, TAY FR, YIU C, HASHIMOTO M, BRESCHI L, CARVALHO RM, ITO S. Collagen degradation by host-derived enzymes during aging. J Dent Res 2004; 83: 216–221. 23. BRESCHI L, MAZZONI A, NATO F, CARRILHO M, VISINTINI E, TJA¨DERHANE L, RUGGERI AJR, TAY FR, DORIGO S, PASHLEY DH. Chlorhexidine stabilizes the adhesive interface: a 2-year in vitro study. Dent Mater 2010; 26: 320–325. 24. VA¨LIMA¨KI VV, YRJANS JJ, VUORIO EI, ARO HT. Molecular biological evaluation of bioactive glass microspheres and adjunct bone morphogenetic protein 2 gene transfer in the enhancement of new bone formation. Tissue Eng 2005; 11: 387–394. 25. ZHONG JP, LATORRE GP, HENCH LL. The kinetics of bioactive ceramics part VII: binding of collagen to hydroxyapatite and bioactive glass. In: Andersson OH, Yli-Urpo A, eds. Bioceramics 7. Oxford: Butterworth-Heinemann, 1994; 61–66. 26. OSORIO R, YAMAUTI M, OSORIO E, ROMAŃ JS, TOLEDANO M. Zinc-doped dentin adhesive for collagen protection at the hybrid layer. Eur J Oral Sci 2011; 119: 401–410. 27. KREMER EA, CHEN Y, SUZUKI K, NAGASE H, GORSKI JP. Hydroxyapatite induces autolytic degradation and inactivation of matrix metalloproteinase-1 and -3. J Bone Miner Res 1998; 13: 1890–1892. 28. FERRACANE JL. Hygroscopic and hydrolytic effects in dental polymer networks. Dent Mater 2006; 22: 211–222. 29. BAKRY AS, TAMURA Y, OTSUKI M, KASUGAI S, OHYA K, TAGAMI J. Cytotoxicity of 45S5 bioglass paste used for dentine hypersensitivity treatment. J Dent 2011; 39: 599–603. 30. MANDEL S, TAS AC. Brushite (CaHPO4·2H2O) to octacalcium phosphate (Ca8(HPO4)2(PO4)4·5H2O) transformation in DMEM solutions at 36.5°C. Mat Sci Mat Sci Eng C-Bio S 2010; 30: 245–254. 31. BUNKER BC, TALLANT DR, HEADLEY TJ, TURNER GL, KIRKPATRICK RJ. The structure of leached sodium borosilicate glass. Phys Chem Glasses 1988; 29: 106–120. 32. WANG Y, SPENCER P. Continuing etching of an all-in-one adhesive in wet dentin tubules. J Dent Res 2005; 84: 50– 354. 33. SAURO S, TOLEDANO M, AGUILERA FS, MANNOCCI F, PASHLEY DH, TAY FR, WATSON TF, OSORIO R. Resin-dentin bonds to EDTA-treated vs. acid etched dentin using ethanol wet-bonding. Part II: effects of mechanical cycling load on microtensile bond strengths. Dent Mater 2011; 27: 563–572. 34. BAYLE MA, GRE´GOIRE G, SHARROCK P. The role of acrylophosphonic acid monomers in the formation of hybrid layers based on self-etch adhesives. J Dent 2007; 35: 302–308. 35. JAYARAMAN M, SUBRAMANIAN MV. Preparation and characterization of two new composites: collagen-brushite

362

Profeta et al.

and collagen octa-calcium phosphate. Med Sci Monit 2002; 8: 481–487. 36. LIU G, ZHAO D, TOMSIA AP, MINOR AM, SONG X, SAIZ E. Three-dimensional biomimetic mineralization of dense hydrogel templates. J Am Chem Soc 2009; 29: 9937–9939. 37. HEWLETT ER, CAPUTO AA, WROBET DC. Glass ionomer bond strength and treatment of dentin with polyacrylic acid. J Prosthet Dent 1991; 66: 767–772. 38. BERRY EA, POWERS JM. Bond strength of glass ionomers to coronal and radicular dentin. Oper Dent 1994; 19: 122–126.

39. YIP HK, TAY FR, NGO HC, SMALES RG, PASHLEY DH. Bonding of contemporary glass ionomer cements to dentin. Dent Mater 2001; 17: 456–470. 40. GIANNINI M, SOARES CJ, DE CARVALHO RM. Ultimate tensile strength of tooth structures. Dent Mater 2004; 20: 322 –329. 41. BEDRAN-RUSSO AK, PEREIRA PN, DUARTE WR, DRUMMOND JL, YAMAUCHI M. Application of crosslinkers to dentin collagen enhances the ultimate tensile strength. J Biomed Mater Res B Appl Biomater 2007; 80: 268–272.

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741

Available online at www.sciencedirect.com

journal homepage: www.intl.elsevierhealth.com/journals/dema

Experimental etch-and-rinse adhesives doped with bioactive calcium silicate-based micro-fillers to generate therapeutic resin–dentin interfaces A.C. Profeta a , F. Mannocci b , R. Foxton b , T.F. Watson a , V.P. Feitosa a,c , B. De Carlo d , R. Mongiorgi d , G. Valdré d , S. Sauro a,∗ a

Biomaterials, Biomimetics and Biophotonics, King’s College London Dental Institute, Guy’s, King’s College and St. Thomas’ Hospital, London SE1 9RT, UK b Department of Conservative Dentistry, King’s College London Dental Institute, London, UK c Department of Restorative Dentistry, Dental Materials Division, Piracicaba Dental School, State University of Campinas, Limeira Av. 901, 13414-903 Piracicaba, Brazil d Department of Scienze della Terra e Geologico-Ambientali, University of Bologna, Bologna, Italy

a r t i c l e

i n f o

a b s t r a c t

Article history:

Objectives. This study aimed at evaluating the therapeutic bioactive effects on the bond

Received 2 October 2012

strength of three experimental bonding agents containing modified Portland cement-based

Received in revised form

micro-fillers applied to acid-etched dentin and submitted to aging in simulated body fluid

17 March 2013

solution (SBS). Confocal laser (CLSM) and scanning electron microscopy (SEM) were also

Accepted 4 April 2013

performed. Methods. A type-I ordinary Portland cement was tailored using different compounds such as sodium–calcium–aluminum–magnesium silicate hydroxide (HOPC),

Keywords:

aluminum–magnesium–carbonate hydroxide hydrates (HCPMM) and titanium oxide

Bioactive micro-fillers

(HPCTO) to create three bioactive micro-fillers. A resin blend mainly constituted by Bis-GMA,

Resin–dentin interface

PMDM and HEMA was used as control (RES-Ctr) or mixed with each micro-filler to create

Bond strength

three experimental bonding agents: (i) Res-HOPC, (ii) Res-HCPMM and (iii) Res-HPCTO. The

Durability

bonding agents were applied onto 37% H3 PO4 -etched dentin and light-cured for 30 s. After

Dentin remineralization

build-ups, they were prepared for micro-tensile bond strength (!TBS) and tested after 24 h or 6 months of SBS storage. SEM analysis was performed after de-bonding, while CLSM was used to evaluate the ultra-morphology/nanoleakage and the mineral deposition at the resin–dentin interface. Results. High !TBS values were achieved in all groups after 24 h. Only Res-HOPC and ResHCPMM showed stable !TBS after SBS storage (6 months). All the resin–dentin interfaces created using the bonding agents containing the bioactive micro-fillers tested in this study showed an evident reduction of nanoleakage and mineral deposition after SBS storage. Conclusion. Resin bonding systems containing specifically tailored Portland cement microfillers may promote a therapeutic mineral deposition within the hybrid layer and increase the durability of the resin–dentin bond. © 2013 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.

Corresponding author. Tel.: +44 207 188 3874; fax: +44 020 71881823. E-mail address: salvatore.sauro@kcl.ac.uk (S. Sauro). 0109-5641/$ – see front matter © 2013 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.dental.2013.04.001 ∗

730

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741

Introduction

The durability of resin–dentin interface represents one of the main concerns in adhesive dentistry as it is affected by severe degradation processes. Bond degradation occurs mainly via water sorption [1], hydrolysis of monomer methacrylates ester bonds caused by salivary esterases [2], and hydrolysis of collagen fibrils, which may be enhanced by activation of endogenous dentin matrix metalloproteinases (MMPs) [3]. Regarding these different mechanisms of degradation, experimental strategies to preserve the hybrid layer such as ethanol-wet bonding [4,5] and the use of MMP inhibitors [6] have been proposed. Nevertheless, current attempts to extend the longevity of resin–dentin bonds via incorporation of more hydrolytically stable resin monomers [7] and/or the use of matrix metalloproteinase inhibitors [8] fail to address two fundamental issues: (1) replacement of the mineral phase within the demineralized dentin collagen; and (2) protection of the collagen from biodegradation through fossilization of MMPs [9]. The use of bioactive materials which promptly interact with dental hard tissues through therapeutic/protective effects may provide a feasible means to extend the longevity of resin–dentin interface [10]. Experimental resin-based calcium-phosphate cements have been advocated as potential therapeutic restorative base-liner materials due to their ability to induce remineralization of caries-affected dentin [11]. Nonetheless, alternative strategies are being developed in order to enhance calcium (Ca2+ ), hydroxyl (OH− ), and phosphate (PO4 −3 ) ions delivery within and beneath the resin–dentin hybrid layer. Calcium-silicate Portland-derived cements are able to release Ca2+ and OH− , so creating favorable conditions for the remineralization of dental hard tissues (i.e. dentin and enamel) [12,13]. These materials possess a bioactive activity since they are able to induce the formation of apatite-like crystals on their surface in a short induction period [14] eliciting a positive response at the interface from the biological environment [15]. However, the use of the Portland cements in operative dentistry is still debated due to clinical limitations related to their long setting time [14,16], high dissolution rate and “specific” mechanical properties [17]. In contrast, the incorporation of resin specific monomers such as 2-hydroxyethyl methacrylate (HEMA), triethyleneglycol dimethacrylates (TEGDMA) and urethane dimethacrylates (UDMA) in silicate-based materials has been proposed to improve the mechanical properties, bond strength to dental tissues and reduce the setting time (light-curable systems) [14,18]. Since there is little information concerning the use of such “hybrid” resin-based photo-polymerizable dental adhesives, this study was purposed to assess the therapeutic/bioactive effects of three innovative bonding agents containing tailored Portland cement-based micro-fillers on the resin–dentin interface. This aim was accomplished by evaluating the micro-tensile bond strength (!TBS) after simulated body fluid solution (SBS) storage (24 h or 6 months). Fractography scanning electron microscopy (SEM) of the debonded specimens, ultra-morphology confocal microscopy (CLSM) and nanoleakage of the resin–dentin interface were

also executed. The null hypotheses to be tested were that the inclusion of the tested micro-fillers within the composition of the experimental bonding agents induces: (i) no effect on the bond strength durability; and (ii) no mineral precipitation and nanoleakage reduction within the demineralized ‘poorly resin-infiltrated’ areas within the resin–dentin interface.

Materials and methods

2.1. Preparation of the experimental bioactive resin-base bonding agents A type I ordinary Portland cement (82.5 wt%) (OPC: Italcementi Group, Cesena, Italy) mainly consisting of tricalcium silicate (Alite: 3CaO × SiO2 ), di-calcium silicate (Belite: 2CaO × SiO2 ), tri-calcium aluminate (3CaO × Al2 O3 ) and gypsum (CaSO4 × 2H2 O) was mixed with 7.5 wt% of phyllosilicate consisting of sodium–calcium–aluminum–magnesium silicate hydroxide hydrate [(Na,Ca)(Al,Mg)6 (Si4 O10 )3 (OH)·6H2 O; Acros Organics, Fair Lawn, NJ, USA] in deionized water (ratio 2:1) to create the first experimental filler (HOPC). The second filler (HCPMM) was created by mixing 90 wt% of type I OPC, 7.5 wt% phyllosilicate and 2.5 wt% of hydrotalcite consisting of aluminum–magnesium–carbonate hydroxide hydrate [Mg6 Al2 (CO3 )(OH)16 ·4(H2 O); Sigma–Aldrich, Gillingham, UK]. The third filler (HPCTO) used in this study was created by mixing OPC (80 wt%), phyllosilicate (7.5 wt%), hydrotalcite (2.5 wt%) and 10 wt% titanium oxide (TiO2 : Sigma–Aldrich). The three modified Portland-based silicates were mixed with deionized water (ratio 2:1) and allowed to set in incubator at 37 ◦ C for 24 h. Subsequently, they were ground in an agate jar and sieved to obtain <30 !m micro-filler particles. A resin co-monomer blend was prepared as a typical three-step, etch-and-rinse bonding agent including a neat resin blend as bond and a 50 wt% ethanol–solvated resin mixture as primer (Res-Ctr – no filler). The neat resin blend was formulated by using 40 wt% of a hydrophobic cross-linking dimethacrylate 2,2-bis[4(2-hydroxy-3-methacryloyloxypropyloxy)-phenyl]-propane (Bis-GMA; Esstech, Essington, PA, USA) and 28.75 wt% of hydrophilic 2-hydroxyethyl methacrylate (HEMA; Sigma–Aldrich). An acidic functional monomer Bis(2-Methacryloyloxyethyl) Pyromellitate (PMDM; Esstech Essington) was also added (30 wt%) to the blend solution to obtain a dental bonding system with chemical affinity to the calcium present in the micro-fillers (Fig. 1). The neat resin was made light-curable by adding 0.25 wt% camphoroquinone (CQ; Sigma–Aldrich), 0.5 wt% 2-ethyl-dimethyl-4-aminobenzoate (EDAB; Sigma–Aldrich) and 0.5% diphenyliodonium hexafluorophosphate (PIHF; Sigma–Aldrich). The resin co-monomer blend was used as control filler-free or mixed with each micro-filler in order to formulate three experimental resin-base bonding agents (GB patent application no. 1118138.5 – filed on 20th October 2011): (i) Res-HOPC: 60 wt% of neat resin and 40 wt% of HOPC; (ii) Res-HCPMM: 60 wt% of neat resin and 40 wt% of HCPMM; and (iii) ResHPCTO: 60 wt% of neat resin and 40 wt% of HPCTO filler (Table 1). The hybrid calcium silicate-based bonding agents were prepared by mixing the neat resin and the fillers for

731

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741

Fig. 1 – Chemical structures of the methacrylate monomers used in the tested resin blends. Abbreviations: BisGMA: 2,2-bis[4(2-hydroxy-3-methacryloyloxy-propyloxy)-phenyl]-propane; HEMA: 2-hydroxyethyl methacrylate; TEGDMA: triethylene-glycoldimethacrylate; PDMD: Bis(2-Methacryloyloxyethyl) Pyromellitate. Table 1 – Chemical composition (wt%) and application mode of the experimental adhesive system used in this study. Group

Primer

Res-Ctr pH (4.6)a

20 wt% Bis-GMA 14.35 wt% HEMA 14.4 wt% PMDM 50 wt% ethanol

Res-HOPC pH (8.4)a

20 wt% Bis-GMA 14.35 wt% HEMA 14.4 wt% PMDM 50 wt% ethanol

Res-HCPMM pH (8.1)a

20 wt% Bis-GMA 14.35 wt% HEMA 14.4 wt% PMDM 50 wt% ethanol

Res-HPCTO pH (8.3)a

20 wt% Bis-GMA 14.35 wt% HEMA 14.4 wt% PMDM 50 wt% ethanol

Bond 40 wt% Bis-GMA 28.75 wt% HEMA 30 wt% PMDM 0.25 wt% camphoroquinone 0.5 wt% 2-ethyl-dimethyl-4-aminobenzoate 0.5% diphenyliodonium hexafluorophosphate 24 wt% Bis-GMA 17.25 wt% HEMA 18 wt% PMDM 0.15 wt% camphoroquinone 0.3 wt% 2-ethyl-dimethyl-4-aminobenzoate 0.3 wt% diphenyliodonium hexafluorophosphate 40 wt% HOPC 24 wt% Bis-GMA 17.25 wt% HEMA 18 wt% PMDM 0.15 wt% camphoroquinone 0.3 wt% 2-ethyl-dimethyl-4-aminobenzoate 0.3 wt% diphenyliodonium hexafluorophosphate 40 wt% HCPMM 24 wt% Bis-GMA 17.25 wt% HEMA 18 wt% PMDM 0.15 wt% camphoroquinone 0.3 wt% 2-ethyl-dimethyl-4-aminobenzoate 0.3 wt% diphenyliodonium hexafluorophosphate 40 wt% HPCTO

Bonding procedures (1) Dentin conditioning with 37% H3 PO4 for 15 s (2) Copious rinse with deionized water (3) Air-drying for 2 s (4) Application of a first layer of each experimental primer for 20 s (5) Air-drying for 5 s at maximum stream power (6) Application of a second layer of each experimental adhesive for 20 s (7) Gently air-drying for 2 s (8) Light-curing for 30 s (9) Resin composite application and light-curing

Bis-GMA: bisphenyl A glycidyl methacrylate; HEMA: hydrophilic 2-hydroxyethyl methacrylate; PMDM: 2,5-dimethacryloyloxyethyloxycarbonyl1,4-benzenedicarboxylic acid; HOPC: set Portland cement and smectite; HPCMM: Portland cement, smectite and hydrotalcite; HPTCO: set Portland cement, smectite, hydrotalcite and titanium oxide. a Three discs for each experimental resin-base material (6 mm in diameter and 1 mm thick) and were light-cured for 30 s immersed in 25 ml of H2 O (pH 6.7) at 37 ◦ C and maintained for 30 days; the pH/alkalinizing activity was evaluated using a professional pH electrode (Mettler-Toledo, Leicester, UK) at room temperature (∼24 ◦ C).

732

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741

Fig. 2 – Schematic illustrating the resin–dentin match-sticks prepared using a water-cooled diamond saw, stored in SBS for 24 h or 6 months, and then subjected to micro-tensile bond strength (!TBS) testing and scanning electron microscopy/fractography. This schematic also illustrates how composite-tooth slabs were prepared, stored in SBS for 24 h or 6 months, immersed in fluorescein (nanoleakage) or xylenol orange (calcium-binding dye) and finally analyzed using confocal laser scanning microscopy (CLSM).

30 s on a glass plate to form a homogeneous paste prior the bonding procedures.

2.2.

Specimen preparation and bonding procedures

Caries-free human molars (age 20–40 years), extracted for periodontal reasons were used in this study. The treatment plan of any of the involved patients, who had given informed consent that their extracted teeth could be used for research purposes, was not altered by this investigation. This study was conducted in accordance with the ethical guidelines of the Research Ethics Committee (REC) for medical investigations. The teeth were stored in deionized water (pH 7.1) at 4 ◦ C and used within 1 month after extraction. A flat midcoronal dentin surface was exposed using a hard tissue microtome (Accutom-50; Struers, Copenhagem, Denmark) equipped with a slow-speed, water-cooled diamond wafering saw (330-CA/RS-70300; Struers). A 180-grit silicon carbide (SiC) abrasive paper mounted on a water-cooled rotating polishing machine (Buehler Meta-Serv 3000; Grinder-Polisher, Düsseldorf, Germany) was used (30 s) to remove the diamond saw smear layer and to replace it with a standard and more clinically related smear layer [19]. The specimens were divided into four groups (n = 5/group) based on the tested materials (Table 1). The specimens were etched using a 37% phosphoric acid solution (H3 PO4 ; Aldrich Chemical) for 15 s followed by copious water rinse. The etched-dentin surfaces were air-dried for 2 s to remove the excess of water. The control (Res-Ctr) and experimental adhesives (Res-HOPC; Res-HCPMM; Res-HPCTO) were applied within a period of 20 s. The specimens were immediately light-cured for 30 s using a quartz–tungsten–halogen (QTH) system (>600 mW/cm2 , Optilux VLC; Demetron, CT,

USA). Five 1-mm-thick incremental build-up were performed using a resin composite (Filtek Z250; 3M-ESPE, St. Paul, MN, USA), light-activated for 20 s each step with a final curing of 60 s (Fig. 2). The specimens were finally stored in SBS solutions (Oxoid, Basingstoke, Hampshire, UK) for 24 h and 6 months at 37 ◦ C.

2.3.

!TBS test and SEM analysis of the failed bonds

The specimens were sectioned perpendicular to the adhesive interface with a slow speed water-cooled diamond wafering blade (Accutom-50; Struers) mounted on a hard tissue microtome (Isomet 11/1180; Buehler). Subsequently, match-sticks with cross-sectional adhesive area of 0.9 mm2 were created (Fig. 2). As each tooth yielded 16 beams, there were a total of 80 match-sticks in each group. Half of these match-sticks (n = 40) were tested after 24 h and the remaining half (n = 40) after 6 months of static SBS storage (37 ◦ C). Each resin–dentin matchstick was attached to a testing apparatus with a cyanoacrylate adhesive (Zapit; Dental Ventures, CA, USA). A tensile load was applied with a customized micro-tensile jig in a LAL300 linear actuator (SMAC Europe; Horsham, West Sussex, UK) with LAC-1 high speed controller single axis with built-in amplifier, that has a stroke length of 50 mm, peak force of 250 N, displacement resolution of 0.5 mm and crosshead speed of 1 mm/min [20]. The load (N) at failure and the cross-sectional area of each failed beam (Digital micrometer Mitutoyo CD15; Mitutoyo, Kawasaki, Japan) permitted calculation of the !TBS in MPa. The !TBS (mean-MPa) data for each group were subjected to the repeated measures ANOVA and Tukey’s post hoc test for pair-wise comparisons (˛ = 0.05). Fisher’s least significant difference (LSD) test was used to isolate and compare the significant differences (P < 0.05) between the groups.

733

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741

Premature failures were included in the statistical analysis as zero values. Modes of failure were classified as percentage of adhesive (A), mixed (M), or cohesive (C) when the failed bonds were examined at 30× using a stereoscopic microscope (Leica M205A; Leica Microsystems, Wetzlar, Germany). For each group, five representative de-bonded specimens, depicting the most frequent failure modes, were chosen for SEM ultramorphology analysis of the fractured surfaces. They were dried overnight and mounted on aluminum stubs with carbon cement. They were sputter-coated with gold (SCD 004 Sputter Coater; Bal-Tec, Vaduz, Liechtenstein) and examined using an SEM (S3500; Hitachi, Wokingham, UK) with an accelerating voltage of 15 kV and a working distance of 25 mm at increasing magnifications from 60× to 5000×.

2.4.

Dye-assisted CLSM evaluation

Three further dentin-bonded specimens were prepared as previously described for each group with the primer/bond resins doped with 0.05 wt% Rhodamine B (Rh-B: Sigma–Aldrich) and then serially sectioned across the adhesive interface to obtain resin–dentin slabs (n = 12 per group) with a thickness of approx. 1 mm (Fig. 2). The resin–dentin slabs were then allocated to two subgroups (n = 6/group) based on the period of static storage in SBS (24 h or 6 months). Following each aging period, the specimens were coated with two layers of fast-setting nail varnish applied 1 mm away from the resin–dentin interfaces. Three specimens from each subgroup were immersed in 1 wt% aqueous fluorescein (Sigma–Aldrich) and the other three specimens in 0.5 wt% xylenol orange solution (XO: Sigma–Aldrich) for 24 h at 37 ◦ C (pH 7.2). The latter is a calcium-chelator fluorophore commonly used in bone remineralization studies [21], due to its ability to form complexes with divalent calcium ions. The specimens were then treated in an ultrasonic water bath for 2 min and polished using ascending (#1200–4000) grit SiC abrasive papers (Versocit; Struers) on a water-cooled polishing device (Buehler Meta-Serv 3000 Grinder-Polisher; Buehler). A final ultrasonic treatment (5 min) concluded the specimen preparation for the confocal microscopy analysis which was performed using a confocal laser scanning microscope (DM-IRE2 CLSM; Leica, Heidelberg, Germany) equipped with a 63×/1.4 NA oil immersion lens. The fluorescein was excited at 488-nm, while XO at 514-nm using an argon laser. The ultra-morphology evaluation (resin-diffusion) was executed using a 568-nm krypton (rhodamine excitation) laser. CLSM images were obtained with a 1 !m z-step to optically section the specimens to a depth up to 20 !m below the surface [22]. The z-axis scans of the interface surface were arbitrarily pseudo-colored by the same operator for better exposure and compiled into single projections using the Leica image-processing software (Leica). The configuration of the system was standardized and used at the same settings for the entire investigation. Each resin–dentin interface was completely investigated and then five optical images were randomly captured. Micrographs representing the most common features observed along the bonded interfaces were captured and recorded [10].

Table 2 – Mean and standard deviation (SD) of the !TBS (MPa) to dentin. Group

24 h

Res-Ctr

Res-HOPC

Res-HCPMM

Res-HPCTO

29.2 ± 9.9 A1 (0/40) [5/15/80]

18.5 ± 10.4 B2 (5/35) [40/43/17]

32.2 ± 9.4 A1 (0/40) [2/36/62]

30.3 ± 11.5 A1 (3/37) [36/57/7]

28.2 ± 11.4 A1 (0/40) [3/20/77]

25 ± 12.7 A1 (0/40) [40/54/6]

29 ± 11.1 A1 (0/40) [3/25/72]

5.7 ± 8.1 C2 (23/17) [31/31/38]

Values are mean ± SD in MPa. In each row, same numbers indicate no differences (p > 0.05) after 24 h and 6 m of SBS storage. In columns, same capital letter indicates no statistically significant differences between each group (p > 0.05). Premature failures were included in the statistical analysis as zero values and are indicated in parentheses (for instance 5/35 means that there were 5 premature failures and 35 testable beams). The modes of failure are expressed in percentage in the brackets [adhesive/mix/cohesive].

Results

3.1.

!TBS test and SEM analysis of the failed bonds

The interaction bonding system vs SBS storage was statistically significant only for the Res-HOPC and Res-HCPMM groups (P = 0.001); no significant reduction of the !TBS values was observed after 6 months of SBS aging (P > 0.05). Conversely, significant drops in the !TBS value were observed in both ResHPCTO and Res-Ctr groups (P < 0.05) after prolonged storage in SBS (6 months). The !TBS results (expressed as mean and SD) and modes of failures obtained for each group are summarized in Table 2. In details, all the tested materials showed high !TBS values after 24 h of SBS storage with failures occurring mainly in cohesive mode. However, only the resin–dentin specimens of the Res-HOPC and Res-HCPMM groups maintained high !TBS values (P > 0.05) after 6 months of storage in SBS (31.3 ± 11.5 and 24 ± 12.7 MPa, respectively); the failure mode was prevalently mixed (57% and 54%, respectively). The SEM analysis of the fractured surfaces at 24 h of SBS storage revealed the absence of both patent dentinal tubules and collagen fibrils, so indicating a good hybridization of acid-etched dentin (Fig. 3A and B, respectively). After 6 months of SBS storage, the de-bonded dentin surfaces were characterized by embedded mineral crystals and remnant resin presenting filler lacunas (Res-HOPC: Fig. 3A1 and A2; Res-HCPMM: Fig. 3B1 and B2). In contrast, a significant drop (P < 0.05) in !TBS was observed after 6 months of storage in SBS, with the specimens created using the Res-Ctr group (filler-free) and with those created with the Res-HTCPO. These latter specimens showed, after 24 h, a well-hybridized de-bonded surface embedding several micro-fillers (Fig. 3C). On the contrary, specimens de-bonded after 6 months SBS aging showed a de-bonded surface with a few dentinal tubules but with no sign of clear degradation and

734

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741

Fig. 3 – SEM failure analysis of de-bonded specimens. (A) SEM micrograph (1000×) of an adhesively fractured stick bonded with Res-HOPC after 24 h of SBS storage. Note the dentin entirely covered with adhesive resin (ra) with some fillers’ lacunas (pointer) and few patent dentinal tubules (dt). (A1) It was also possible to see the presence of resin adhesive (ra) and no unprotected collagen fibrils. Some filler particles were detached during testing procedures (pointer) but initial mineral precipitation may be observed (white asterisk). (A2) After 6 months, the de-bonded dentin surface at higher magnification (2500×) was covered with adhesive resin (ra). Mineral crystals (asterisk) embedded within a preserved collagen network were vastly encountered albeit some fillers were detached (pointer). (B) SEM micrograph of a specimen bonded with Res-HCPMM after 24 h. Note the presence of adhesive resin (ra) covering the dentin and dentinal tubules (dt); some fillers’ lacunas are also observed (pointer). (B1) In another portion of the same specimen, more lacunas are observed, but also initial mineral crystallization was depicted (asterisk). (B2) After 6 months storage, at higher magnification (2500×), very low bonding degradation can be observed. More mineral precipitation was detected (asterisk) and the fillers’ lacunas (pointer) are wider probably due to the expansion of the fillers when exposed to water. (C) Micrograph of a de-bonded stick from group Res-HPCTO after 24 h showing dentin completely covered with rare filler detachment (pointer). (C1) In other region, note the presence of few exposed dentinal tubules as well as intact resin tags (rt). (C2) After 6 months, the fractured dentin surface bonded with Res-HPCTO showed a de-bonding within the hybrid layer and/or at its bottom. Several resin tags (rt) were observed well-hybridized with peritubular dentin (black pointer); some funneled dentinal tubules were also encountered (black asterisk). (D) SEM micrograph from a specimen bonded with Res-Ctr and presenting a well-hybridized resin-covered dentin surface. (D1) A fractured stick from the same group showed an adhesive failure at the bottom of hybrid layer with remnants of resin adhesive (ar) and some resin tags (rt). (D2) The control adhesive after 6 months showed many signs of degradation since most collagen fibrils were degraded, the funneled dentinal tubules (asterisk) were often found along with poorly hybridized resin tags (pointer). White finger: filler lacuna; white asterisk: mineral precipitation; black asterisk: funneled degraded peritubular dentin; black finger: hybridization between tags and peritubular.

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741

well-hybridized peritubular dentin with resin tags (Fig. 3C1 and C2). The Res-Ctr specimens tested after 24 h of SBS storage and analyzed with SEM presented few exposed dentinal tubules but mostly were obliterated by resin tags or covered by resin remnants (Fig. 3D). Conversely, the surface of the specimens de-bonded after 6 months of SBS exhibited no collagen fibrils on the dentin surface with rare resin tags and degraded funneled dentinal tubules (Fig. 3D1 and D2).

3.2.

Dye-assisted CLSM evaluation

CLSM imaging of the bonded-dentin interfaces subsequent to 24 h of SBS storage showed relevant ultra-morphology and nanoleakage information for all groups. It was observed that all tested materials were able to diffuse within the demineralized dentin, creating a hybrid layer 8–14 !m thick, with a multitude of resin tags penetrating the dentinal tubules (Fig. 4). Nevertheless, all these interfaces were affected by conspicuous fluorescein penetration (nanoleakage) through dentinal tubules into a porous hybrid layer (Fig. 4). Furthermore, the resin–dentin interface created using the experimental bonding agents containing the bioactive microfillers showed the presence of XO dye within the hybrid, adhesive layers and inside the dentinal tubules (Fig. 5). On the contrary, the acid-etched dentin bonded using the resin control (Res-Ctr, filler-free) showed no presence of XO along the interface. Significant ultra-morphology changes were observed subsequent prolonged SBS storage. For instance, the CLSM analysis revealed no gap and limited fluorescein penetration (nanoleakage) within the resin–dentin interfaces created using the Res-HOPC and Res-HCPMM (Fig. 6A and B). In addition, OX-dye produced a clearly outlined fluorescence due to a consistent Ca-minerals deposited within the bonding interface and inside the dentinal tubules (Fig. 5A and B). The resin–dentin interfaces created using Res-HPCTO showed less nanoleakage within the hybrid layer; evident porosities were also observed (Fig. 6C). Intense nanoleakage and constant gaps affected the resin–dentin interfaces created using the ResCtr (Fig. 6D). When the same interfaces were investigated employing OX, only the walls of the dentinal tubules were highlighted (Fig. 5E).

Discussion

The resin–dentin interfaces created during bonding procedures when using contemporary “simplified” etch-and-rinse bonding agents are affected by bond strength reduction subsequent to prolonged (3–6 months) water aging [23]. This phenomenon occurs mainly due to the inability of such materials to completely replace loosely bound and bulk-free water from the apatite-depleted dentin collagen matrix during bonding procedures. This residual water within the resin–dentin interface causes hygroscopic swelling effects and hydrolytic degradation of polymer networks and favors metalloproteinases (MMPs)-mediated collagenolytic degradation [23–27]. In specific circumstances, the presence of water within the hybrid layer may be essential to facilitate a biomimetic apatite nucleation within the gap zones of collagen fibrils

735

and to fossilize host-derived, collagen-bound MMPs [24,25,28]. Indeed, recent investigations have demonstrated that it is possible to reduce the nanoleakage and micropermeability within the resin–dentin interface and maintain the bond strength [10] of bioactive resin-base materials applied to H3 PO4 acidetched dentin subsequent to simulated body fluid storage for 3–6 months [18,22]. Ryou et al. [29] demonstrated that using a biomimetic remineralization approach it is feasible to remineralize the dentin collagen within the resin–dentin interface via slow release of calcium ions from set white Portland cement and subsequent interaction of these ions with phosphate species from SBS or dentin substrate. Portland cements designed for dental applications, also known as hydraulic silicate cements or MTA, exhibit outstanding biological properties and high bioactivity when immersed in SBF [18,29,30]. In the present study, modified Portland cement-based micro-fillers were created and included within the composition of a representative three-step/etch-and-rinse bonding agent in order to create a material with therapeutic remineralizing effects on the mineral-deficient areas along the bonding interface. Based on the results obtained in this study, the first null hypothesis that the inclusion of bioactive micro-fillers within the composition of the experimental bonding agent has no effect on the bond strength durability must be must be rejected as the use of the experimental Res-HOPC and Res-HCPMM systems preserved the bond strength over prolonged SBS storage. The second null that no mineral precipitation and nanoleakage reduction would be observed within the demineralized ‘poorly resin-infiltrated’ areas within the resin–dentin interface must be also rejected. In detail, the three experimental bonding agents containing the bioactive micro-fillers (Res-HOPC, Res-HCPMM, Res-HPCTO) and the control co-monomer blend (RES-Ctr) used in this study to bond the acid-etched dentin produced comparably high !TBS values (P > 0.05) following 24 h of storage in SBS (Table 2). Conversely, after 6 months of storage in SBS a significant decrease in !TBS (P < 0.05) was observed for the RES-Ctr and Res-HPCTO groups, while the specimens bonded using Res-HOPC or Res-HCPMM maintained consistent long-term bond strength values (P > 0.05). The specimens of the Res-HOPC and Res-HCPMM groups de-bonded after 6 months of SBS storage showed, during SEM fractography examination, residual resin presence and mineral bodies on the fractured surface (Fig. 3A1, A2, B1 and B2). Important morphological differences were observed in the specimens of the Res-HPCTO group which presented a debonded surface characterized by very few partially exposed dentinal tubules and an important presence of mineral crystals after 6 months of SBS storage (Fig. 3C1 and C2). The SEM analysis revealed that the de-bonded dentin surface of the specimens in RES-Ctr group was well resin-hybridized after 24 h of SBS storage (Fig. 3D). In contrast, the prolonged SBS storage (6 months) induced radical changes; the dentin surface presented funneled dentinal tubules as an essential sign of degradation of the “poorly resin-infiltrated” demineralized and peritubular dentin (Fig. 3D1 and D2). These results were also supported by the CLSM analysis performed to evaluate the nanoleakage and the presence of calciumcompounds within the resin–dentin interface subsequent to

736

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741

Fig. 4 – Confocal laser scanning microscopy (CLSM) single-projection images showing the interfacial characterization and nanoleakage, after 24 h of storage in SBS. Images (1) indicate the fluorescein dye (nanoleakage) within the resin dentin interface, whereas the images (2) disclose the diffusion (Rhodamine B) of the bonding systems within the demineralized (acid-etched) dentin. The images (3) are the combined projections of both dyes. (A1–A3) CLSM images showing the interfacial characteristics of the bonded-dentin interface created using Res-HOPC. It is possible to observe a clear hybrid layer (hl) with long resin tags (rt) penetrating the dentinal tubules (dt) underneath the adhesive layer (ad) presenting evident mineral fillers (FL). Intense fluorescein uptake was observed within the entire resin–dentin interface as well as the adhesive layer. (B1–B3) Characteristics of the bonded-dentin interface created using Res-HCPMM. Similarly to images from Res-HOPC, these images presented high dye uptake (nanoleakage throughout the entire resin–dentin interface as well as the adhesive layer). (C1–C3) The resin–dentin interface created using the Res-HPCTO bonding system was characterized by a clear hybrid layer (hl) located underneath the adhesive layer (ad) containing the experimental micro-filler (FL). Long resin tags (rt) penetrating the dentinal tubules (dt) were observed as well as evident nanoleakage and dye uptake along the entire interface and adhesive layer. (D1–D3) These images show the control bonded-dentin interface (RES-Ctr) characterized by a thick hybrid layer (hl) (approximately 8 !m thickness) located underneath an adhesive layer (ad) devoid of fillers.

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741

737

Fig. 5 – CLSM single-projection image disclosing the fluorescent calcium-chelators dye xylenol orange. All images were obtained from specimens immersed in simulated body-fluid solution for 24 h or 6 months. (A) Resin–dentin interface created with Res-HOPC after 24 h of SBS storage. Mineral deposition can be visualized within the adhesive layer (ad), the hybrid layer (hl) along the walls of dentinal tubules (dt) and the filler inside the resin tags (rt). (B) Resin–dentin interface created with Res-HCPMM and immersed in SBS for 6 months showing a clear fluorescence signal due to a consistent presence of Ca-deposits within the adhesive layer (ad), hybrid layer, walls of the dentinal tubules (dt) and resin tags (rt). (C) Resin–dentin interface created with Res-HPCTO and immersed in SBS for 24 h. Xylenol orange was able to stain the Ca-minerals within adhesive layer, hybrid layer and dentinal tubule (dt). Note the intense calcium deposition at bottom of hybrid layer. (D) Resin–dentin interface created with Res-HPCTO and immersed in SBS for 6 months showing also in this case Ca-mineral presence at the bottom and within the hybrid layer, dt and rt. (E) Resin–dentin interface created with Res-Ctr (no filler) in which it is possible to note absence of calcium deposition both within the hybrid (hl) and adhesive layer (ad). Only the walls of the dentinal tubule tubules (dt) were stained by the fluorescent dye.

738

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741

Fig. 6 – Confocal laser scanning microscopy (CLSM) single-projection images showing the interfacial characterization and nanoleakage after 6 months of SBS storage. (A) Resin–dentin interface created using Res-HOPC characterized by reduced nanoleakage within the hybrid layer (hl). Note the absence of fluorescein uptake within the adhesive layer. (B) Image showing the interfacial features of the bonded-dentin interface created using the Res-HCPMM. Note the low overall nanoleakage with very little fluorescein uptake in hybrid and adhesive layers. (C) Resin–dentin interface created using Res-HPCTO. Despite the mineral deposition and reduced nanoleakage within adhesive layer, a gap was probably created by the cutting procedures. As the low presence of fluorescein within the interface, it is possible to assume that the resin degradation was replaced by mineral precipitation creating an interface with low elasticity (high stiffness properties). (D) Resin–dentin interface created using the control adhesive system (RES-Ctr). Note the presence of intense dye uptake (nanoleakage) within the hybrid layer and at the bottom of adhesive layer. In this case the presence of gaps frequently observed between hybrid and adhesive was very likely due to hybrid layer degradation but with no clear mineral precipitation within the interface. Dt, dentinal tubules; rt, resin tags; ad, adhesive layer; c, composite; hl, hybrid layer.

SBS storage (24 h or 6 months). Indeed, further evidences of the therapeutic bioactivity of the experimental bonding agents containing the tailored Portland cement-based micro-fillers were attained; reduced fluorescent-dye uptake (nanoleakage) was observed along the entire resin–dentin interface after 6 months of storage in SBS (Fig. 5A–C). These latter observations along with the strong xylenol orange (XO-dye) signal from the hybrid layer and the dentinal tubules (Fig. 5), clearly indicated the remineralization of those areas which were previously detected as mineraldeficient/poor-resin infiltrated zones of the resin–dentin interface. We hypothesize that the therapeutic remineralizing effects observed within the mineral-depleted resin–dentin interface were essentially due to the bioactivity of the experimental micro-fillers. Indeed, the reaction mechanism of the Portland

cement-based micro-fillers may have involved the reaction of the polymerized calcium-silicate hydrate gel with water to release calcium hydroxide and to the consequent increase of the alkalinity of the surrounding environment [31]; this increase of pH was confirmed in this study (Table 2). This localized increase in pH within the resin–dentin interface may have also interfered with the activity of MMPs [3,23]. Furthermore, the interaction between the phosphate ions present in the aging solution (SBF) or in the dentin substrate, and the calcium released from the Portland-based micro-fillers may have enhanced the formation of new mineral deposits upon existing mineral constituents within the dentin matrix (biocatalyzation) [32]. However, it is well known that the increase in environmental pH and the presence of free OH− may facilitate apatite nucleation and reduce the solubility of intermediate Ca/P species formed during the aging period [33]. The

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741

most appropriate pH to support the formation of stoichiometric hydroxyapatite (HA) in vitro [34] and in vivo [35] falls in a range between 8 and 9. At higher pH it is common to obtain a Ca-deficient HA (lower solubility then stoichiometric hydroxyapatite) characterized by higher concentrations of PO4 3− and lower Ca2+ ions [36]. Moreover, we hypothesize the release of carboxylic species (R-COO− ) from the PMDM which may have interacted with the remnant calcium present along the front of demineralization at the bottom of the hybrid layer acting as a sort of sequestering agent for Ca/P cluster, promoting the precipitation of Ca-compounds [18,22]. All these mineralizing processes just mentioned may also have reduced the distribution of water-rich regions within the hybrid layer at the resin–dentin interface [18,37]. However, the water absorbed overtime during SBS storage may have been responsible for the hydrolytic and hygroscopic mechanisms involved in the degradation of dental polymers [38]. In addition to the formation of minerals precipitants within the interface, the nanostructure of the calcium-silicate hydrate may also have contributed to seal the dentinal tubules due to the small-scale volume of the forming gels, along with a slight expansion of the calcium silicate-based materials once immersed in SBS [39]. In particular, the phyllosilicates (i.e. smectite) and hydrotalcite, which were contained in the micro-fillers used in this study, have the ability to expand considerably following water sorption into the interlayer molecular spaces [40]. The amount of expansion is due largely to the type of exchangeable cation contained in the micro-filler; the uptake kinetics of cation exchange is fast and the presence of Na+ , as the predominant exchangeable cation, can result in material swelling. In this condition, the exceeding water is removed, thereby preventing hygroscopic effects and hydrolytic degradation of the polymer chains [41]. Also, it is reasonable to expect that the metallic ions intercalated on phyllosilicate were easily released by ionexchange with cations present in the surrounding solutions and acted as effective antibacterial substances in the long term [38]; further studies are necessary to confirm this latter hypothesis. In contrast, the bond strength reduction observed in the resin–dentin interfaces created using the Res-HPCTO bonding agent after prolonged immersion in SBS (Table 2) may be due to the high hydrophilicity of the TiO2 . Micro-fine titanium oxide (TiO2 ) has been used as inorganic additive of resin composites to match the opaque properties of teeth [42] and as nano-particles to increase the microhardness and flexural strength of dental composites [40]. However, TiO2 has been advocated as a super-hydrophilic component, in particular under ultraviolet (UV) light irradiation [43–45]. Therefore, a possible explanation for the !TBS reduction may be attributed to this high hydrophilicity which may have permitted excessive water adsorption and induced severe resin degradation as well as the extraction of water-soluble unreacted monomers or oligomers from the resin-matrix [46]; also in this case further studies are necessary to clarify these particular processes of degradation. Moreover, the replacement of the degraded resin by “over” mineral crystallization within the Res-HPCTO bonded-dentin interface (Figs. 4 and 5) during prolonged SBS storage may have conferred mechanical characteristics related to bond strength comparable to those

739

created by conventional glass-ionomer cements (GICs) applied onto polyacrylic acid-etched dentin and submitted to tensile tests [47,48]. Indeed, several studies indicated that the bond strength of GICs when tested using tensile or shear methods was approximately 5 MPa; these values results do not reflect the true adhesive strength to dentin [49]. These factors may have been also responsible for the formation of gaps within the resin–dentin interface created by the Res-HPCTO during the sample preparation [10]. Finally, a further important issue regarding the microfillers used in this study is their biocompatibility. Several studies showed the adequate biocompatibility of Portland cement-based fillers [50–52]. The cytotoxicity of degradation products from methacrylate-based resins is far worse than those from the micro-fillers. Therefore, the addition of 40 w% micro-fillers reduced the resin percentage and, consequently, the negative effects of its degradation products. Furthermore, the products of hydrolysis of the therapeutic micro-fillers are essentially mineral phases of calcium compounds which precipitate within the resin–dentin interface (Figs. 4 and 5) promoting remineralization. As the results of this study demonstrated that the resin–dentin bond may be maintained overtime by inducing a therapeutic remineralization of the bonding interface, specific experimental resin bonding systems containing bioactive micro-fillers, such as Res-HOPC, Res-HCPMM or Res-HPCTO may offer the possibility to improve the durability of the resin–dentin interfaces. The characteristic of promoting bioactivity should also open up the potential to create therapeutic restorative materials able to reduce the incidence of secondary caries. Indeed, it is important to consider that restorative materials containing bioactive fillers may be effective in killing a wide selection of aerobic bacteria due to the increase of the local pH and concentration of alkali ions [53,54]. The antibacterial properties are potentially of great importance as the infiltration of microorganisms may cause secondary caries which jeopardize the longevity of resin–dentin interface leading to the replacement of dental restorations [55,56]. Further studies are ongoing in order to evaluate the species-specific antibacterial effects and biocompatibility of the materials tested in this study.

Conclusion

The present outcomes showed that the inclusion of calcium silicate-based micro-fillers within the composition of resin bonding agent may promote a therapeutic mineral deposition within the resin–dentin interface. However, only the resin–dentin interfaces created using Res-HOPC and Res-HCPMM showed prolonged bond durability. Therefore, these two micro-fillers may represent potential bioactive components to be included during the formulation of new innovative and smart therapeutic adhesive resin systems. The addition of titanium oxide within the composition of bioactive micro-fillers for resin-based systems may enhance the remineralization effects. Thus, this type of micro-filler should be avoided for bonding purposes but considered for healing treatments in step-wise restorative procedures [57].

740

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741

Acknowledgments [13]

This work was partially supported by the Centre of Excellence in Medical Engineering funded by the Wellcome Trust and by the independent research commissioned by the National Institute for Health Research under the Comprehensive Biomedical Research Centre at Guy’s & St. Thomas’ Trust. The views expressed in this publication are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health. The authors also acknowledge the laboratory support offered by Dr. Silvano Zanna in the formulation of the modified Portland-based micro-fillers used in this study. The authors have no financial affiliation or involvement with any commercial organization with direct financial interest in the materials discussed in this manuscript. Any other potential conflict of interest is disclosed.

references

[1] Ito S, Hashimoto M, Wadgaonkar B, Svizero N, Carvalho RM, Yiu CK, et al. Effects of resin hydrophilicity on water sorption and changes in modulus of elasticity. Biomaterials 2005;26:6449–59. [2] Moszner N, Salz U, Zimmermann J. Chemical aspects of self-etching enamel–dentin adhesives: a systematic review. Dental Materials 2005;21:895–910. [3] Pashley DH, Tay FR, Yiu CK, Hashimoto M, Breschi L, Carvalho RM, et al. Collagen degradation by host-derived enzymes during aging. Journal of Dental Research 2004;83:216–21. [4] Sadek FT, Pashley DH, Ferrari M, Tay FR. Tubular occlusion optimizes bonding of hydrophobic resins to dentin. Journal of Dental Research 2007;86:524–8. [5] Tay FR, Pashley DH, Kapur RR, Carrilho MR, Hur YB, Garrett LV, et al. Bonding BisGMA to dentin – a proof of concept for hydrophobic dentin bonding. Journal of Dental Research 2007;86:1034–9. [6] Carrilho MR, Carvalho RM, de Goes MF, di Hipolito V, Geraldeli S, Tay FR, et al. Chlorhexidine preserves dentin bond in vitro. Journal of Dental Research 2007;86:90–4. [7] Moszner N, Fischer UK, Angermann J, Rheinberger V. Bis-(acrylamide)s as new cross-linkers for resin-based composite restoratives. Dental Materials 2006;22:1157–62. [8] Carrilho MR, Carvalho RM, Sousa EN, Nicolau J, Breschi L, Mazzoni A, et al. Substantivity of chlorhexidine to human dentin. Dental Materials 2010;26:779–85. [9] Brackett MG, Li N, Brackett WW, Sword RJ, Qi YP, Niu LN, et al. The critical barrier to progress in dentine bonding with the etch-and-rinse technique. Journal of Dentistry 2011;39:238–48. [10] Profeta AC, Mannocci F, Foxton RM, Thompson I, Watson TF, Sauro S. Bioactive effects of a calcium/sodium phosphosilicate on the resin–dentin interface: a microtensile bond strength, scanning electron microscopy, and confocal microscopy study. European Journal of Oral Sciences 2012;120:353–62. [11] Peters MC, Bresciani E, Barata TJ, Fagundes TC, Navarro RL, Navarro MF, et al. In vivo dentin remineralization by calcium-phosphate cement. Journal of Dental Research 2010;89:286–91. [12] Huan Z, Chang J, Huang XH. Self-setting properties and in vitro bioactivity of Ca2 SiO4 /CaSO4 ·1/2H2 O composite bone

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

cement. Journal of Biomedical Materials Research Part B: Applied Biomaterials 2008;87:387–94. Coleman NJ, Awosanya K, Nicholson JW. Aspects of the in vitro bioactivity of hydraulic calcium (alumino)silicate cement. Journal of Biomedical Materials Research Part A 2009;90:166–74. Taddei P, Modena E, Tinti A, Siboni F, Prati C, Gandolfi MG. Vibrational investigation on the in vitro bioactivity of commercial and experimental calcium-silicate cements for root-end endodontic therapy. Journal of Molecular Structure 2011;993:367–75. Parirokh M, Torabinejad M. Mineral trioxide aggregate: a comprehensive literature review. Part III: clinical applications, drawbacks and mechanism of action. Journal of Endodontics 2010;36:400–13. Camilleri J. Evaluation of selected properties of mineral trioxide aggregate sealer cement. Journal of Endodontics 2009;35:1412–7. Bogen G, Kuttler S. Mineral trioxide aggregate obturation: a review and a case series. Journal of Endodontics 2009;35:777–90. Sauro S, Osorio R, Watson TF, Osorio E, Toledano M. Novel light-curable materials containing experimental bioactive micro-fillers remineralise mineral-depleted bonded-dentine interfaces. Journal of Biomaterials Science, Polymer Edition. http://dx.doi.org/10.1080/09205063.2012.727377 Koibuchi H, Yasuda N, Nakabayashi N. Bonding to dentin with a self-etching primer: the effect of smear layers. Dental Materials 2001;17:122–6. Sauro S, Watson TF, Thompson I, Toledano M, Nucci C, Banerjee A. Influence of air-abrasion executed with polyacrylic acid-Bioglass 45S5 on the bonding performance of a resin-modified glass ionomer cement. European Journal of Oral Sciences 2012;120:168–77. Rahn BA, Perren SM. Xylenol orange, a fluorochrome useful in polychrome sequential labelling of calcifying tissues. Stain Technology 1971;46:125–9. Sauro S, Osorio R, Watson TF, Toledano M. Therapeutic effects of novel resin bonding systems containing bioactive glasses on mineral-depleted areas within the bonded-dentin interface. Journal of Materials Science Materials in Medicine 2012;23:1521–32. Breschi L, Mazzoni A, Ruggeri A, Cadenaro M, Di Lenarda R, De Stefano Dorigo E. Dental adhesion review: aging and stability of the bonded interface. Dental Materials 2008;24:90–101. Kim J, Gu L, Breschi L, Tjaderhane L, Choi KK, Pashley DH, et al. Implication of ethanol wet bonding in hybrid layer remineralization. Journal of Dental Research 2010;89: 575–80. Pashley DH, Tay FR, Breschi L, Tjaderhane L, Carvalho RM, Carrilho MR, et al. State of the art etch-and-rinse adhesives. Dental Materials 2011;27:1–16. Breschi L, Mazzoni A, Nato F, Carrilho MR, Visintini E, Tjaderhane L, et al. Chlorhexidine stabilizes the adhesive interface: a 2-year in vitro study. Dental Materials 2010;26:320–5. Liu Y, Li N, Qi Y, Niu LN, Elshafiy S, Mao J, et al. The use of sodium trimetaphosphate as a biomimetic analog of matrix phosphoproteins for remineralization of artificial caries-like dentin. Dental Materials 2011;27:465–77. Tay FR, Pashley DH. Biomimetic remineralization of resin-bonded acid-etched dentin. Journal of Dental Research 2009;88:719–24. Ryou H, Niu LN, Dai L, Pucci CR, Arola DD, Pashley DH, et al. Effect of biomimetic remineralization on the dynamic nanomechanical properties of dentin hybrid layers. Journal of Dental Research 2011;90:1122–8.

d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 729–741

[30] Darvell BW, Wu RC. MTA – an hydraulic silicate cement: review update and setting reaction. Dental Materials 2011;27:407–22. [31] Lawrence CD. Physicochemical and mechanical properties of Portland cements. In: Hewlett PC, editor. LEA’s chemistry of cement and concrete. 4th edition Woburn, MA, USA: Butterworth & Heinemann; 1998. p. 343–420. [32] Gandolfi MG, Van Landuyt KL, Taddei P, Modena E, Van Meerbeek B, Prati C. Environmental scanning electron microscopy connected with energy dispersive X-ray analysis and Raman techniques to study ProRoot mineral trioxide aggregate and calcium silicate cements in wet conditions and in real time. Journal of Endodontics 2010;36: 851–7. [33] Somasundaran P, Ofori Amankonah J, Ananthapadmabhan KP. Mineral-solution equilibria in sparingly soluble mineral systems. Colloids and Surfaces 1985;15:309–33. [34] Kikuchi M, Itoh S, Ichinose S, Shinomiya K, Tanaka J. Self-organization mechanism in a bone-like hydroxyapatite/collagen nanocomposite synthesized in vitro and its biological reaction in vivo. Biomaterials 2001;22:1705–11. [35] Recillas S, Rodríguez-Lugo V, Montero ML, Viquez-Cano S, ˜ VM. Studies on the precipitation Hernandez L, Castano behavior of calcium phosphate solutions. Journal of Ceramic Processing Research 2012;13:5–10. [36] Durucan C, Brown PW. Alpha-tricalcium phosphate hydrolysis to hydroxyapatite at and near physiological temperature. Journal of Materials Science Materials in Medicine 2000;11:365–71. [37] Qi YP, Li N, Niu LN, Primus CM, Ling JQ, Pashley DH, et al. Remineralization of artificial dentinal caries lesions by biomimetically modified mineral trioxide aggregate. Acta Biomaterialia 2012;8:836–42. [38] Ferracane JL. Hygroscopic and hydrolytic effects in dental polymer networks. Dental Materials 2006;22: 211–22. [39] Skinner LB, Chae SR, Benmore CJ, Wenk HR, Monteiro PJM. Nanostructure of calcium silicate hydrates in cements. Physical Review Letters 2010;104:1955021–4. [40] Bhattacharyya KG, Gupta SS. Adsorption of a few heavy metals on natural and modified kaolinite and montmorillonite: a review. Advances in Colloid and Interface Science 2008;140:114–31. [41] Malachová K, Praus P, Pavlickova Z, Turicova M. Activity of antibacterial compounds immobilised on montmorillonite. Applied Clay Science 2009;43:364–8. [42] Yu B, Ahn JS, Lim JI, Lee YK. Influence of TiO2 nanoparticles on the optical properties of resin composites. Dental Materials 2009;25:1142–7. [43] Xia Y, Zhang F, Xie H, Gu N. Nanoparticle-reinforced resin-based dental composites. Journal of Dentistry 2008;36:450–5.

741

[44] Chadwick RG, Traynor N, Moseley H, Gibbs N. Blue light curing units – a dermatological hazard? British Dental Journal 1994;176:17–21. [45] Bonding N, Graem N, Rygaard J, Dabelsteen E. Effects of irradiation with dental light curing units on Langerhans cells in human stratified epithelium in heterotransplanted skin. Scandinavian Journal of Dental Research 1987;95: 463–6. [46] Karuppuchamy S, Jeong JM. Super-hydrophilic amorphous titanium dioxide thin film deposited by cathodic electrodeposition. Materials Chemistry and Physics 2005;93:251–4. [47] Spencer P, Ye Q, Park J, Topp EM, Misra A, Marangos O, et al. Adhesive/dentin interface: the weak link in the composite restoration. Annals of Biomedical Engineering 2010;38:1989–2003. [48] Hewlett ER, Caputo AA, Wrobel DC. Glass ionomer bond strength and treatment of dentin with polyacrylic acid. Journal of Prosthetic Dentistry 1991;66:767–72. [49] Berry 3rd EA, Powers JM. Bond strength of glass ionomers to coronal and radicular dentin. Operative Dentistry 1994;19:122–6. [50] Camilleri J, Montesin FE, Di Silvio L, Pitt Ford TR. The chemical constitution and biocompatibility of accelerated Portland cement for endodontic use. International Endodontic Journal 2005;38:834–42. [51] Hwang YC, Kim DH, Hwang IN, Song SJ, Park YJ, Koh JT, et al. Chemical constitution, physical properties, and biocompatibility of experimentally manufactured Portland cement. Journal of Endodontics 2011;37:58–62. [52] Ribeiro DA, Duarte MA, Matsumoto MA, Marques ME, Salvadori DM. Biocompatibility in vitro tests of mineral trioxide aggregate and regular and white Portland cements. Journal of Endodontics 2005;31:605–7. [53] Yip HK, Tay FR, Ngo HC, Smales RJ, Pashley DH. Bonding of contemporary glass ionomer cements to dentin. Dental Materials 2001;17:456–70. [54] Hu S, Chang J, Liu M, Ning C. Study on antibacterial effect of 45S5 Bioglass. Journal of Materials Science Materials in Medicine 2009;20:281–6. [55] Boeckh C, Schumacher E, Podbielski A, Haller B. Antibacterial activity of restorative dental biomaterials in vitro. Caries Research 2002;36:101–7. [56] Sauro S, Watson TF, Tay FR, Chersoni S, Breschi L, Bernardi F, et al. Water uptake of bonding systems applied on root dentin surfaces: a SEM and confocal microscopic study. Dental Materials 2006;22:671–80. [57] Bjørndal L, Larsen T, Thylstrup A. A clinical and microbiological study of deep carious lesions during stepwise excavation using long treatment intervals. Caries Research 1997;31:411–7.