Arhiv 66 2

ISSN 0004-1254

ARHIV ZA HIGIJENU RADA I TOKSIKOLOGIJU

ARCHIVES OF INDUSTRIAL HYGIENE AND TOXICOLOGY

Arh Hig Rada Toksikol • Vol. 66 • No. 2 • pp. 97-180 • ZAGREB, CROATIA 2015

CONTENTS Mini-review Nanoparticle interaction with the immune system

Veno Kononenko, Mojca Narat, and Damjana Drobne

97

Mirta Milić, Ružica Rozgaj, Vilena Kašuba, Ana-Marija Jazbec, Boris Starčević, Barnaba Lyzbicki, Gloria Ravegnini, Corrado Zenesini, Muriel Musti, Patrizia Hrelia, and Sabrina Angelini

109

Sofia Grabovska and Yuriу Salyha

121

ADHD-like behaviour in the offspring of female rats exposed to low chlorpyrifos doses before pregnancy

Suzana Žunec, Božica Radić, Kamil Kuča, Kamil Musilek, and Ana Lucić Vrdoljak

129

Comparative determination of the efficacy of bispyridinium oximes in paraoxon poisoning

Neda Mažuran, Vladimir Hršak, and Goran Kovačević

135

The effects of CaCl2 and CaBr2 on the reproduction of Daphnia magna Straus

Željka Vidaković-Cifrek, Mirta Tkalec, Sandra Šikić, Sonja Tolić, Hrvoje Lepeduš, and Branka Pevalek-Kozlina

141

Growth and photosynthetic responses of Lemna minor L. exposed to cadmium in combination with zinc or copper

Fulya Dilek Gökalp Muranli, Martin Kanev, and Kezban Ozdemir

153

Genotoxic effects of diazinon on human peripheral blood lymphocytes

Karolina Gromadzka, Agnieszka Waśkiewicz, Joanna Świetlik, Jan Bocianowski, and Piotr Goliński

159

The role of wastewater treatment in reducing pollution of surface waters with zearalenone

Ante Miličević and Nenad Raos

165

Modelling of copper(II) binding to pentapeptides related to atrial natriuretic factor using the 3χv connectivity index

Silvije Davila, Jadranka Pečar Ilić, and Ivan Bešlić

171

Real-time dissemination of air quality information using data streams and Web technologies: linking air quality to health risks in urban areas

A11

Report (in English)

Original articles Polymorphisms in DNA repair genes: link with biomarkers of the CBMN cytome assay in hospital workers chronically exposed to low doses of ionising radiation

Cover page: The cover page shows an image of duckweed (Lemna minor L.) and was provided by Professor Željka Vidaković-Cifrek from the Faculty of Science, University of Zagreb. Due to its high sensibility to the presence of various compounds in growth medium as well as simple cultivation in laboratory conditions, duckweed is frequently used as a test organism.

Disclaimer: This photo is intended to evoke the content of this issue of the journal. It is not intended for instructional or scientific purposes.

97

Kononenko V, et al. Nanoparticle interaction with the immune system Arh Hig Rada Toksikol 2015;66:97-108

Mini-review

DOI: 10.1515/aiht-2015-66-2582

Nanoparticle interaction with the immune system Veno Kononenko1, Mojca Narat2, and Damjana Drobne1 Department of Biology1, Department for Animal Science2, Biotechnical Faculty, University of Ljubljana, Slovenia [Received in October 2014; CrossChecked in October 2014; Accepted in April 2015] When nanoparticles enter the body, their interactions with cells are almost unavoidable. Unintended nanoparticle interaction with immune cells may elicit a molecular response that can have toxic effects and lead to greater susceptibility to infectious diseases, autoimmune disorders, and cancer development. As evidenced by several studies, nanoparticle interactions with biological systems can stimulate inflammatory or allergic reactions and activate the complement system.

Nanoparticles can also stimulate immune response by acting as adjuvants or as haptens. Immunosuppressive effects have also been reported. This article gives a brief review of in vitro and in vivo research evidencing stimulatory or suppressive effects of nanoparticles on the immune system of mammals. In order to ensure safe use of nanosized particles, future research should focus on how their physical and chemical properties influence their behaviour in the biological environment, as they not only greatly affect nanoparticle-immune system interactions but can also interfere with experimental assays.

KEY WORDS: immune response; immunomodulation; immunotoxicity; nanomaterials; nanoparticle properties; nanosafety

We are currently witnessing a rapid progress of nanotechnology and an increasing manufacture and use of engineered nanoparticles. Nanoparticles are defined as particles that have at least one dimension smaller than 100Â nm (1). Their small size means an increased proportion of surface atoms and therefore changed physicochemical properties (2). These properties can be used beneficially for many applications, from electronics, cosmetics, and textile industry to drug delivery and bioimaging (3). However, the same properties can make nanoparticles more harmful to living organisms due to increased reactivity and easy penetration into organisms and cells (4). Several studies have shown that particles of the same chemical composition but different size pose different risk; smaller particles are more harmful (5-7). Numerous nanotoxicological studies have focused on cytotoxicity (8-10), which occurs at a relatively high nanoparticle concentration/dose. At a lower concentration/dose, the sub-lethal and long-term effects on cells can occur (11-14). Studying the immunomodulatory effects of nanoparticles is particularly important, because immunocompromised organisms are susceptible to infections and cancer development (15). The primary function of the immune system is to detect and recognise foreign substances in order to protect the host. Nanoparticles can interfere with this function or can themselves be recognised as foreign antigens and thus elicit immune response (16). A disturbance in the immune system can Correspondence to: Veno Kononenko, Department of Biology, Biotechnical Faculty, University of Ljubljana, VeÄ?na pot 111, SI-1000 Ljubljana, Slovenia, E-mail: veno.kononenko@bf.uni-lj.si

lead to severe medical conditions (17) and understanding how different factors influence the host defence mechanisms is an important part of toxicological research.

Nanoparticles can enter the body unintentionally through the gastrointestinal tract, skin, and airways or can be intentionally administered to the body with biomedical applications (18). Inside the body, there is a high probability that nanoparticles will come into contact with immune cells, which can lead to nanoparticleimmune system interactions (15, 19). These interactions have an immunomodulatory potential, as they can activate or suppress immune function (Figure 1) and lead to inflammation, increased susceptibility to infectious diseases, or even to autoimmune diseases or cancer (15).

However, in some biomedical applications, for example in vaccine delivery (19, 20), we can design nanoparticles for targeted modulation of immune response.

Stimulation of immune response Depending on their physicochemical properties nanoparticles can stimulate innate and adaptive immune response (Table 1). However, it is still unclear how individual nanoparticles affect it.

Activation of innate immune response When nanoparticles enter the body, they can interact with immune cells and trigger inflammatory response.

Inflammatory response is accompanied by the secretion of signalling molecules (cytokines, chemokines) that provide communication between immune cells and coordinate

98 molecular events. Positively charged nanoparticles usually possess a higher inflammatory potential than negatively charged or neutral nanoparticles (20). This can be explained by the fact that macrophages have a negatively charged sialic acid on their surface and readily interact with cationic substances (21). Macrophages recognise foreign antigens with their toll-like receptors (TLRs), which bind to corresponding antigens and activate the signal transduction pathway and inflammatory response (21). In their in vitro experiment Lucarelli et al. (22), exposed human macrophages to non-toxic concentrations of different SiO2, TiO2, ZrO2, and Co nanoparticles and observed increased expression of TLR receptors and production of inflammatory cytokines. The experiment showed that different nanoparticles triggered inflammatory response in different ways. SiO 2 nanoparticles induced the production of inflammatory cytokines IL-1β and TNF-α, and Co nanoparticles inhibited anti-inflammatory IL-1RA and induced inflammatory TNF-α (22). Another in vitro study (23) showed a cytotoxic and inflammatory effect of Ag nanoparticles on rat brain microvascular endothelial cells (RBMEC), with an increased release of proinflammatory mediators IL-1β, TNF, and PGE-2. The effect of Ag nanoparticles was significantly stronger with smaller (25 nm) than with larger particles (40 and 80 nm). Chuang et al. (24) recently showed that the intensity of inflammatory response induced by carbon black nanoparticles of different size correlated with their surface area. Xia et al. (25) observed a cytotoxic effect of ZnO nanoparticles and the induction of inflammatory response in RAW 264.7 and BEAS-2B cell lines. In contrast, CeO2 and TiO2 nanoparticles did not elicit any such effect.

The results of several in vivo studies have also shown how nanoparticles can affect inflammatory response. Park et al. (26) treated mice with Fe 3O 4 nanoparticles by intratracheal instillation and noticed increased production of pro-inflammatory cytokines IL-1, TNF-α, and IL-6. They also reported increased production of Th0-type cytokine IL-2, Th1-type cytokine IL-12, Th2-type cytokines IL-4 and IL-5, TGF-β, and an increased production of IgE. Kaewamatawong et al. (27) have found that intratracheally instilled SiO 2 nanoparticles can cause pulmonary inflammation in mice. Nishimori et al. (28) observed that i.v. injected SiO2 particles in mice had size-dependent hepatotoxic effects. Only smaller particles (<100 nm) caused higher serum markers of liver injury, serum aminotransferase, and inflammatory cytokines IL-6 and TNF-α. Cho et al. (29) noticed a gene expression pattern typical of apoptosis and inflammation in mice liver after i.v. administration of Au nanoparticles coated with polyethylene glycol (PEG). Single-walled carbon nanotubes were also hypothesized to cause inflammation (30). Some studies however suggest that the main cause of inflammation are impurities resulting from nanoparticle synthesis (31).

Kononenko V, et al. Nanoparticle interaction with the immune system Arh Hig Rada Toksikol 2015;66:97-108

Exposure to nanoparticles can also interfere with response to infection. Mice exposed to carbon nanotubes before infection with Listeria monocytogenes had an enhanced acute pulmonary inflammation and delayed bacterial clearance (decreased phagocytosis and nitric oxide production) (32).

Activation of the complement system The complement system is an important part of the innate immune system that helps antibodies and phagocytic cells to remove pathogens from the host. There are a number of reports claiming that nanoparticles activate the complement system via different pathways (33-40).

Furthermore, altering nanoparticle surface properties can increase or decrease complement activation (33, 35, 40).

Pondman et al. (38) have shown that complement opsonisation of carbon nanotubes enhances their uptake by U937 cells without inflammatory response. Pedersen et al.

(36) have shown that dextran-coated Fe3O4 particles can activate the complement system. In another study, Au nanoparticles did not activate the complement system - even though complement proteins prevailed in the corona - nor did they affect complement activation by a known activator (41).

Activation of adaptive immune response Unlike the innate immune system, the adaptive immune system is antigen-specific, requires some time to achieve its maximum effect, and typically generates an immunological memory. It consists of humoral and cellular antigen-specific responses, and nanoparticles can stimulate both. Liu et al. (42) found that polyhydroxylated fullerenes [C60(OH)20] stimulate the production of Th1 cytokines and decrease the production of Th2 cytokines (42). C60(OH)20 nanoparticles show a low cytotoxic effect on immune cells, but significantly stimulate TNF-α release, which has an important role in the removal of abnormal cells. In addition, they seem to suppress tumours in vivo, as they increase the CD4+/CD8+ lymphocyte ratio.

Some nanoparticles have an epitope structure to which specific antibodies bind. Being small molecules by definition however, most nanoparticles probably act as haptens, which are immunogenic only when attached to a larger carrier molecule. Chen et al. (43) demonstrated that the immune system can generate antibodies specific to nanoparticles. After the immunisation of mice with a C60 fullerene derivate conjugated to bovine thyroglobulin, they produced IgG antibodies specific to fullerenes. Other researchers were not able to detect fullerene-specific antibodies, even when they used a carrier molecule (44).

This inconsistency in results could be explained by the use of different fullerene derivatives or differences between the animal models and immunisation protocols employed (20). For some biomedical applications, nanoparticles are functionalised by growth factors,

99

Kononenko V, et al. Nanoparticle interaction with the immune system Arh Hig Rada Toksikol 2015;66:97-108

receptors, and other biomolecules that may induce autoantibodies (20).

Several studies have shown that nanoparticles can also act as adjuvants, i.e. as substances that are added to the antigen in order to stimulate immune response (34, 44-50).

Polymethylmethacrylate (PMMA) nanoparticles used as adjuvants for HIV-2 virus vaccine in mice induced up to a 100 times higher antibody response than the conventional adjuvant aluminium hydroxide [Al(OH)3] (49). How exactly nanoparticles function as adjuvants is poorly understood, but some studies suggest that nanoparticles can promote the uptake of antigens or can stimulate antigen-presenting cells (20). The adjuvant-like properties of nanoparticles depend on their physicochemical properties. Sun et al.

(51) found a correlation between the shape and crystallinity of AlOOH nanoparticles and their adjuvant capacity both in vitro (activation of dendritic cells) and in vivo (production of IgG and IgE against ovalbumin) (51). Li et al. (50) showed that Al(OH)3 nanoparticles induced a stronger humoral response than microparticles of the same chemical composition. Cao et al. (52) also found that ultra-small graphene oxide-supported gold nanoparticles (usGO-Au) used as an adjuvant stimulated humoral and cellular immune responses.

Some studies have associated exposure to nanoparticles with allergic reactions. Nanoparticles can increase (53-55) or inhibit allergic reactions (56). Chen et al. (57) reported that TiO2 nanoparticles directly stimulated histamine release from the mast cells. Mast cells can contribute to inflammation and the toxic effect of some nanoparticles (19). There is increasing evidence that mast cells have an important role in the biological events following nanoparticle exposure (58-61).

Suppression of immune response Nanoparticles can also suppress the immune system (Table 1), which can weaken immune response against infections and cancerous cells. These immunosuppressive properties, on the other hand, can make nanoparticles useful in preventing transplant rejection, in treating inflammatory and autoimmune diseases, and in delivering immunosuppressive drugs (62-64). However, we still do not know which nanoparticle properties are responsible for immunosuppressive effects. While some nanoparticles are used to deliver immunosuppressive drugs, others have their own immunosuppressive properties. Shen et al. (65) have shown that Fe3O4 nanoparticles weaken the antigen-specific humoral response and T cell cytokine expression in ovalbumin-challenged mice. Mitchell et al.

(66, 67) reported that multi-walled carbon nanotubes (MWCNTs) suppressed systemic humoral immunity in mice. Some nanoparticles have been shown to possess antiinflammatory properties. CeO2 nanoparticles were reported to reduce ROS and the level of inflammatory cytokines IL-6 and TNF-Îą in murine macrophages (68). Shaunak et al.

(69) reported that polyvalent dendrimer glucosamine conjugates inhibited the induction of inflammatory cytokines and chemokines in human macrophages and dendritic cells exposed to bacterial endotoxin. John et al.

(70) have designed polymerised lipid nanoparticles that bind to specific selectins on inflammation-activated lung endothelial cells and reduce inflammation in the allergic airway disease. Ryan et al. (56) report that fullerene inhibits hypersensitivity reaction to allergens in vitro and in vivo.

Nanoparticle physicochemical properties affecting immune response The effect of nanoparticles on the immune system is determined by their physicochemical properties (15, 20).

For a proper interpretation of the biological effects of nanoparticles it is therefore important to know their physicochemical properties (21, 71). Warheit (72) suggests that a nanotoxicological experiment should be preceded by the characterisation of at least the following nanoparticle properties: size, size distribution, surface area and reactivity, crystallinity, aggregation in relevant medium, composition and surface coating, method of synthesis, and impurities.

The effect of nanoparticles can also depend on surface ion dissolution (73); more soluble particles such as ZnO and FeO are more toxic than the less soluble ones such as CeO2 and TiO2 (74). Therefore, it is advisable to check their solubility in relevant media before testing. Nevertheless, some studies have shown that nanoparticle effects on the immune system are different from the effects of their ions (75-77).

Several studies have demonstrated that size significantly determines nanoparticle biological effects (5-7, 78-83). The smaller the size, the higher the relative surface area, and therefore the higher the dissolution of toxic ions and reactive oxygen species (ROS) production (71).

Nanoparticle shape is also important for biological effects (84). For example, fullerenes and carbon nanotubes have the same chemical composition, but different shape, which influences their toxicological properties (85). The surface properties of nanoparticles affect their behaviour in suspensions and interactions with cell membranes. The surface charge correlates with nanoparticle aggregation/ agglomeration in media and with the ability to cross biological barriers (86). Sonication, which is often used to disperse nanoparticle aggregates/agglomerates in suspension, can accelerate ion dissolution and ROS production on the surface of nanoparticles (87) and increase cytotoxicity.

Biological effects can also be altered by impurities, generated as by-products in nanoparticle synthesis (31, 88), or by endotoxins (89). We also have to take into consideration that the properties of nanoparticles can change in biological environments such as cell culture media in vitro or bloodstream in vivo, which can influence biological response to nanoparticle exposure. Several studies have

100

Kononenko V, et al. Nanoparticle interaction with the immune system Arh Hig Rada Toksikol 2015;66:97-108

Table 1 Stimulation and suppression of immune response by nanoparticles Immunostimulatory effects of nanoparticles Observed immune effect Pro-inflammatory effects

Study design

Tested particles (size)

Test system

Reference

In vitro

SiO2 (4-40 nm), TiO2 (20-160 nm), ZrO2 (5-30 nm), Co (50-200 nm)

Human myelomonocytes (U-937)

22

Ag (25, 40, 80 nm)

Rat brain microvessel endothelial cells (RBMEC)

23

SiO2 (10, 100 nm)

Human peripheral blood mononuclear cells (PBMC)

108

ZnO (13 nm)

Murine macrophages (RAW 264.7) and human bronchial epithelial cells (BEAS-2B)

25

NiO (<50 nm)

Human bronchial epithelial cells (BEAS-2B) and human lung carcinoma cells (A549)

109

Au (3, 6, 40 nm)

Murine macrophages (J774 A1)

110

Fe3O4 (~5 nm)

ICR mice

26

SiO2 (14 nm)

ICR mice

27

SiO2 (70 nm)

BALB/c mice

28

SiO2 (50, 500 nm)

Tuck-Ordinary mice

111

Au coated with PEG (4, 100 nm)

BALB/c mice

29

SWCNT (1-4 nm × 1-3 µm)

C57BL/6 mice

32

SWCNT (1-2 nm × 10 nm to several µm)

ICR mice

112

TiO2 nanorods (diameter of 4-6 nm)

Wistar rats

113

Latex nanomaterial (25, 50, 100 nm)

ICR mice

80

Ag (20 nm)

Brown Norway rats

114

Ag (15 nm)

Fischer rats

115

Carbon black (15, 51, 95 nm)

SH rats

24

SWCNT (different sizes)

Human serum

33

SWCNT coated with PEG (1-5 nm × 50-300 nm)

Human serum

35

CNTs (different sizes)

Human serum

38

In vivo

Activation of the complement system

In vitro

101

Kononenko V, et al. Nanoparticle interaction with the immune system Arh Hig Rada Toksikol 2015;66:97-108

Table 1 continued

Nanoparticles as haptens Nanoparticles as adjuvants

CNTs (different sizes)

Human serum

39

Functionalized MWCNT (10-20 nm × 10-50 nm)

Human serum

40

Dextran-coated Fe3O4 (50, 250, 600 nm)

Human serum

36

In vivo

C60 fullerene

BALB/c mice

43

In vitro

AlOOH (different shapes and sizes) usGO-Au (5-10 nm)

51 52

AlOOH (different shapes and sizes)

C57BL/6 mice

51

usGO-Au (5-10 nm)

C57BL/6 mice

52

Al(OH)3 (112 nm)

BALB/c mice

50

In vitro

TiO2 (60 ± 10 nm)

Rat mast cells (RBL-2H3)

57

In vivo

TiO2 (15, 50, 100 nm)

NC/Nga mice

116

Ag (10 nm)

BALB/c mice

117

ZnO (20, 240 nm)

BALB/c mice

118

In vivo

Stimulation of allergic reactions

Human leukemic monocyte (THP-1 ) and murine bone marrow derived dendritic cells (BMDCs) Murine macrophages (RAW 264.7)

Immunosuppressive effects of nanoparticles Observed immune effect

Study design

Tested particles (size)

Test system

Reference

Anti-inflammatory effect

In vitro

CeO2

Murine macrophages (J774)

68

Peripheral blood mononuclear cells (PBMCs)

69

Human lung carcinoma cells (A549) and human bronchial epithelial cells (NHBE)

119

Polyvalent dendrimer glucosamine conjugates SWCNT (0,8-1,2 nm × 800 nm)

Suppression of hypersensitivity, reaction to allergens Suppression of the humoral immune response

In vivo

ZnO (20, 240 nm)

BALB/c mice

118

In vitro

C60 fullerenes

Human mast cells and peripheral blood basophils

56

In vivo

C60 fullerenes

C57BL/6 mice

56

In vivo

Fe2O3

BALB/c mice

65

MWCNT (10-20 nm × 5-15 µm)

C57BL/6 mice

66

MWCNT

C57BL/6 mice

67

102

Kononenko V, et al. Nanoparticle interaction with the immune system Arh Hig Rada Toksikol 2015;66:97-108

Figure 1 Nanoparticle interaction with the immune system. The primary function of the immune system is to protect the host from foreign substances. When nanoparticles enter the body (I), they get in contact with different immune cells (II). Nanoparticle interactions with immune cells can activate immune response (IIIa). Nanoparticles can also interfere with the immune systemâ&#x20AC;&#x2122;s recognition of other immunogenic substances and can stimulate or suppress immune response (IIIb). Normally, immune response gradually leads to the removal of foreign matter from the body, but nanoparticle interaction with immune response can have toxic consequences (IV)

103

Kononenko V, et al. Nanoparticle interaction with the immune system Arh Hig Rada Toksikol 2015;66:97-108

evaluated how biomolecules bound to nanoparticle surface (the so-called biomolecular corona) affect nanoparticle effects on cells (90-92). Nanoparticles that enter the bloodstream can bind with opsonins, which makes them more visible to phagocytic cells, which in turn remove them from the circulation (93). But, even phagocytes can be affected by nanoparticle toxicity (94). Therefore, the surface of nanoparticles that need to enter the bloodstream should be modified to avoid the opsonin binding.

Adjustment and validation of standard methods for testing nanoparticle interaction with the immune system In vitro evaluation of nanoparticle effects on immune cells and the immune system is essential for comprehensive understanding of nanoparticle effects on living organisms in order to make their use safe. Although common cytotoxicity tests may be useful in identifying acute toxicity risks for host cells, including the immune cells, they do not detect the sublethal effects and the dysregulation of the immune system function. Therefore, researchers studying immunotoxicity have established a set of methods for testing immune function (95-99).

Due to their specific physicochemical properties nanoparticles can interfere with the established tests, which were originally developed for testing the biological effects of conventional chemicals. Interactions between nanoparticles and the test method can lead to false positive or false negative results (100-104).

Because of that and because of different mechanisms through which nanoparticles can interact with the immune system, it is necessary to use a battery of broadrange methods. There are several in vitro and in vivo assays for testing nanoparticle effects on the immune system, which have been reviewed elsewhere (105-107).

Their protocols have to be properly adjusted and validated.

When studying the effects of nanoparticles on the immune system, we should also consider the type of the selected biological system as well as time and route of exposure. Different immune cells have different functions in immune response, as they have different receptors and uptake mechanisms.

Furthermore, when testing the long-term and chronic effects of nanoparticles we have to avoid the use of high nanoparticle concentrations that can result in acute toxicity and cell death.

CONCLUSIONS Studies that have been done to date have shown that nanoparticles can interact with different components of the immune system. These interactions are diverse, complex, and not well understood, yet. They may result in unforeseen changes in the functioning of different immune

cells, leading to unpredictable outcomes. The diversity and specific properties of nanoparticles make their risk assessment difficult. To date, the correlation between the properties of nanoparticles and their biological effects, including the effect on the immune system, are poorly understood. Since nanoparticles can interfere with the traditional testing methods developed for testing the biological effects of chemicals, additional attention should be given to the selection of appropriate methods.

Identifying the effects of nanoparticles on the immune system is crucial for their safe use. Nanoparticles for biomedical applications can be designed to interact with the immune system in an intended way or not to react at all. However, we are still a long way from being able to design nanoparticles that will have only a desirable biological effect.

Future research should focus on which nanoparticle property contributes to which effect. This means more in vitro and in vivo studies with detailed nanoparticle characterisation. More attention should also be given to determining the mechanisms of interaction between nanoparticles and different components of the immune system to understand why the same nanoparticles stimulate certain immune functions and suppress others.

With new findings about the interactions between nanoparticles and the immune system we will be able to make better and safer nanotechnological products.

Acknowledgements This study is part of Veno Kononenko’s PhD dissertation and was supported by the Slovenian Research Agency (ARRS), grant no. 1000-14-0510.

REFERENCES 1. Gartman A, Findlay AJ, Luther GW. Nanoparticulate pyrite and other nanoparticles are a widespread component of hydrothermal vent black smoker emissions. Chem Geol 2014;336:32-41. doi: 10.1016/j.chemgeo.2013.12.013 2. Wise JP, Goodale BC, Wise SS, Craig GA, Pongan AF, Walter RB, Thompson WD, Ng AK, Aboueissa AM, Mitani H, Spalding MJ, Mason MD. Silver nanospheres are cytotoxic and genotoxic to fish cells. Aquat Toxicol 2010;97:34-41. doi: 10.1016/j.aquatox.2009.11.016 3. Ngô C, Van de Voorde MH. Nanotechnology in a Nutshell: From Simple to Complex Systems. Paris: Atlantis Press; 2014.

4. Donaldson K, Poland CA, Schins RPF. Possible genotoxic mechanisms of nanoparticles: Criteria for improved test strategies. Nanotoxicology 2010;4:414-20. doi: 10.3109/17435390.2010.482751 5. Coradeghini R, Gioria S, Garcia CP, Nativo P, Franchini F, Gilliland D, Ponti J, Rossi F. Size-dependent toxicity and cell interaction mechanisms of gold nanoparticles on mouse fibroblasts. Toxicol Lett 2013;217:205-16. doi: 10.1016/j. toxlet.2012.11.022

104 6. Shang L, Nienhaus K, Nienhaus GU. Engineered nanoparticles interacting with cells: size matters. J Nanobiotechnol 2014;12:5. doi: 10.1186/1477-3155-12-5 7. Toyooka T, Amano T, Ibuki Y. Titanium dioxide particles phos-phorylate histone H2AX independent of ROS production. Mutat Res 2012;742:84-91. doi: 10.1016/j. mrgentox.2011.12.015 8. Ingle AP, Duran N, Rai M. Bioactivity, mechanism of action, and cytotoxicity of copper-based nanoparticles: a review.

Appl Microbiol Biotechnol 2014;98:1001-9. doi: 10.1007/ s00253-013-5422-8 9. Kroll A, Dierker C, Rommel C, Hahn D, Wohlleben W, Schulze-Isfort C, Göbbert C, Voetz M, Hardinghaus F, Schnekenburger J. Cytotoxicity screening of 23 engineered nanomaterials using a test matrix of ten cell lines and three different assays. Part Fibre Toxicol 2011;8:9. doi: 10.1186/1743-8977-8-9 10. Lewinski N, Colvin V, Drezek R. Cytotoxicity of nanoparticles.

Small 2008;4:26-49. doi: 10.1002/smll.200700595 11. Przybytkowski E, Behrendt M, Dubois D, Maysinger D. Nanoparticles can induce changes in the intracellular metabolism of lipids without compromising cellular viability.

FEBS J 2009;276:6204-17. doi: 10.1111/j.1742- 4658.2009.

07324.x 12. Saptarshi SR, Feltis BN, Wright PF, Lopata AL. Investigating the immunomodulatory nature of zinc oxide nanoparticles at sub-cytotoxic levels in vitro and after intranasal instillation in vivo. J Nanobiotechnol 2015;13:6. doi:10.1186/s12951015-0067-7 13. Kawata K, Osawa M, Okabe S. In vitro toxicity of silver nanoparticles at noncytotoxic doses to HepG2 human hepatoma cells. Environ Sci Technol 2009;43:6046-51. doi: 10.1021/es900754q 14. Dworak N, Wnuk M, Zebrowski J, Bartosz G, Lewinska A. Genotoxic and mutagenic activity of diamond nanoparticles in human peripheral lymphocytes in vitro. Carbon 2014;68:763-76 doi:10.1016/j.carbon.2013.11.067 15. Zolnik BS, González-Fernández A, Sadrieh N, Dobrovolskaia MA. Nanoparticles and the immune system. Endocrinology 2010;151:458-65. doi: 10.1210/en.2009-1082 16. Klippstein R, Fernandez-Montesinos R, Castillo PM, Zaderenko AP, Pozo D. Silver nanoparticles interactions with the immune system: implications for health and disease. In: Pozo Perez D, editor. Silver nanoparticles. 1st ed. Rijeka: In TechOpen; 2010. p. 309-24.

17. Glaser R, Kiecolt-Glaser JK. Stress-induced immune dysfunction: implications for health. Nat Rev Immunol 2005;5:243-51. doi: 10.1038/nri1571 18. Cupaioli FA, Zucca FA, Boraschi D, Zecca L. Engineered nanoparticles. How brain friendly is this new guest? Prog Neuroboil 2014;119-120:20-38. doi: 10.1016/j. pneurobio.2014.05.002 19. Smith MJ, Brown JM, Zamboni WC, Walker NJ. From immunotoxicity to nanotherapy: the effects of nanomaterials on the immune system. Toxicol Sci 2014;138:249-55. doi: 10.1093/toxsci/kfu005 20. Dobrovolskaia MA, McNeil SE. Immunological properties of engineered nanomaterials. Nat Nanotechnol 2007;2:46978. doi: 10.1038/nnano.2007.223 21. Dwivedi PD, Misra A, Shanker R, Das M. Are nanomaterials a threat to the immune system? Nanotoxicology 2009;3:1926. doi: 10.1080/17435390802604276

Kononenko V, et al. Nanoparticle interaction with the immune system Arh Hig Rada Toksikol 2015;66:97-108

22. Lucarelli M, Gatti AM, Savarino G, Quattroni P, Martinelli L, Monari E, Boraschi D. Innate defence functions of macrophages can be biased by nano-sized ceramic and metallic particles. Eur Cytokine Netw 2004;15:339-46.

PMID: 15627643 23. Trickler WJ, Lantz SM, Murdock RC, Schrand AM, Robinson BL, Newport GD, Schlager JJ, Oldenburg SJ, Paule MG, Slikker W Jr, Hussain SM, Ali SF. Silver nanoparticle induced blood-brain barrier inflammation and increased permeability in primary rat brain microvessel endothelial cells. Toxicol Sci 2010;118:160-70. doi: 10.1093/toxsci/kfq244 24. Chuang HC, Chenc LC, Leic YC, Wuc KY, Fengb PH, Chengc TJ. Surface area as a dose metric for carbon black nanoparticles: A study of oxidative stress, DNA singlestrand breakage and inflammation in rats. Atmos Environ 2015;106:329-34. doi:10.1016/j.atmosenv.2015.02.014 25. Xia T, Kovochich M, Liong M, Mädler L, Gilbert B, Shi H, Yeh JI, Zink JI, Nel AE. Comparison of the mechanism of toxicity of zinc oxide and cerium oxide nanoparticles based on dissolution and oxidative stress properties. ACS Nano 2008;2:2121-34. doi: 10.1021/nn800511k 26. Park EJ, Kim H, Kim Y, Yi J, Choi K, Park K. Inflammatory responses may be induced by a single intratracheal instillation of iron nanoparticles in mice. Toxicology 2010;275:65-71. doi: 10.1016/j.tox.2010.06.002 27. Kaewamatawong T, Shimada A, Okajima M, Inoue H, Morita T, Inoue K, Takano H. Acute and subacute pulmonary toxicity of low dose of ultrafine colloidal silica particles in mice after intratracheal instillation. Toxicol Pathol 2006;34:958-65. doi: 10.1080/01926230601094552 28. Nishimori H, Kondoh M, Isoda K, Tsunoda S, Tsutsumi Y, Yagi K. Silica nanoparticles as hepatotoxicants. Eur J Pharm Biopharm 2009;72:496-501. doi: 10.1016/j.ejpb.2009.02.005 29. Cho WS, Kim S, Han BS, Son WC, Jeong J. Comparison of gene expression profiles in mice liver following intravenous injection of 4 and 100 nm-sized PEG-coated gold nanoparticles. Toxicol Lett 2009;191:96-102. doi: 10.1016/j. toxlet.2009.08.010 30. Shvedova AA, Kisin ER, Mercer R, Murray AR, Johnson VJ, Potapovich AI, Tyurina YY, Gorelik O, Arepalli S, SchweglerBerry D, Hubbs AF, Antonini J, Evans DE, Ku BK, Ramsey D, Maynard A, Kagan VE, Castranova V, Baron P. Unusual inflammatory and fibrogenic pulmonary responses to singlewalled carbon nanotubes in mice. Am J Physiol Lung Cell Mol Physiol 2005;289:L698-708. doi: 10.1152/ ajplung.00084.2005 31. Pulskamp K, Diabaté S, Krug HF. Carbon nanotubes show no sign of acute toxicity but induce intracellular reactive oxygen species in dependence on contaminants. Toxicol Lett 2007;168:58-74. doi: 10.1016/j.toxlet.2006.11.001 32. Shvedova AA, Fabisiak JP, Kisin ER, Murray AR, Roberts JR, Tyurina YY, Antonini JM, Feng WH, Kommineni C, Reynolds J, Barchowsky A, Castranova V, Kagan VE. Sequential exposure to carbon nanotubes and bacteria enhances pulmonary inflammation and infectivity. Am J Respir Cell Mol Biol 2008;38:579-90. doi: 10.1165/ rcmb.2007-0255OC 33. Andersen AJ, Robinson JT, Dai HJ, Hunter AC, Andresen TL, Moghimi SM. Single-walled carbon nanotube surface control of complement recognition and activation. ACS Nano 2013;7:1108-19. doi: 10.1021/nn3055175

Kononenko V, et al. Nanoparticle interaction with the immune system Arh Hig Rada Toksikol 2015;66:97-108

34. Dykman LA, Sumaroka MV, Staroverov SA, Zaitseva IS, Bogatyrev VA. Immunogenic properties of the colloidal gold.

Biol Bull 2004;31:75-9. doi: 10.1023/B:BIBU.0000014358.

98422.9c 35. Hamad I, Christy Hunter A, Rutt KJ, Liu Z, Dai H, Moein Moghimi S. Complement activation by PEGylated singlewalled carbon nanotubes is independent of C1q and alternative pathway turnover. Mol Immunol 2008;45:3797803. doi: 10.1016/j.molimm.2008.05.020 36. Pedersen MB, Zhou X, Larsen EK, Sorensen US, Kjems J, Nygaard JV, Nyengaard JR, Meyer RL, Boesen T, VorupJensen T. Curvature of synthetic and natural surfaces is an important target feature in classical pathway complement activation. J Immunol 2010;184:1931-45. doi: 10.4049/ jimmunol.0902214 37. Pham CT, Mitchell LM, Huang JL, Lubniewski CM, Schall OF, Killgore JK, Pan D, Wickline SA, Lanza GM, Hourcade DE. Variable antibody-dependent activation of complement by functionalized phospholipid nanoparticle surfaces. J Biol Chem 2011;286:123-30. doi: 10.1074/jbc.M110.180760 38. Pondman KM, Sobik M, Nayak A, Tsolaki AG, Jäkel A, Flahaut E, Hampel S, Ten Haken B, Sim RB, Kishore U. Complement activation by carbon nanotubes and its influence on the phagocytosis and cytokine response by macrophages.

Nanomedicine 2014;10:1287-99. doi: 10.1016/j. nano.2014.02.010 39. Salvador-Morales C, Flahaut E, Sim E, Sloan J, Green ML, Sim RB. Complement activation and protein adsorption by carbon nanotubes. Mol Immunol 2006;43:193-201. doi: 10.1016/j.molimm.2005.02.006 40. Salvador-Morales C, Basiuk EV, Basiuk VA, Green ML, Sim RB. Effects of covalent functionalization on the biocompatibility characteristics of multi-walled carbon nanotubes. J Nanosci Nanotechnol 2008;8:2347-56. doi: 10.1166/jnn.2008.090 41. Dobrovolskaia MA, Neun BW, Man S, Ye X, Hansen M, Patri AK, Crist RM, McNeil SE. Protein corona composition does not accurately predict hematocompatibility of colloidal gold nanoparticles. Nanomedicine 2014;10:1453-63. doi: 10.1016/j.nano.2014.01.009 42. Liu Y, Jiao F, Qiu Y, Li W, Qu Y, Tian C, Li Y, Bai R, Lao F, Zhao Y, Chai Z, Chen C. Immunostimulatory properties and enhanced TNF- alpha mediated cellular immunity for tumor therapy by C 60(OH) 20 nanoparticles. Nanotechnology 2009;20:415102. doi: 10.1088/0957-4484/20/41/415102 43. Chen BX, Wilson SR, Das N, Coughlin DJ, Erlanger BF. Antigenicity of fullerenes: Antibodies specific for fullerenes and their characteristics. Proc Natl Acad Sci USA 1998;95:10809-13. doi: 10.1073/pnas.95.18.10809 44. Andreev SM, Babakhin AA, Petrukhina AO, Romanova VS, Parnes ZN, Petrov RV. Immunogenic and allergenic properties of fulleren conjugates with aminoacids and proteins. Dokl Biochem 2000;370:4-7. PMID: 11977250 45. Castignolles N, Morgeaux S, Gontier-Jallet C, Samain D, Betbeder D, Perrin P. A new family of carriers (biovectors) enhances the immunogenicity of rabies antigens. Vaccine 1996;14:1353-60. doi: 10.1016/S0264-410X(96)00043-6 46. de Haar C, Hassing I, Bol M, Bleumink R, Pieters R. Ultrafine but not fine particulate matter causes airway inflammation and allergic airway sensitization to co-administered antigen in mice. Clin Exp Allergy 2006;36:1469-79. doi: 10.1111/j.1365-2222.2006.02586.x

105 47. Niikura K, Matsunaga T, Suzuki T, Kobayashi S, Yamaguchi H, Orba Y, Kawaguchi A, Haegawa H, Kajino K, Ninomiya T, Ijiro K, Sawa H. Gold nanoparticles as a vaccine platform: Influence of size and shape on immunological responses in vitro and in vivo. ACS Nano 2013;7:3926-38. doi: 10.1021/ nn3057005 48. Rajananthanan P, Attard GS, Sheikh NA, Morrow WJ. Evaluation of novel aggregate structures as adjuvants: composition, toxicity studies and humoral responses. Vaccine 1999;17:715-30. doi: 10.1016/S0264-410X(98)00256-4 49. Stieneker F, Kreuter J, Löwer J. 1991. High antibody titres in mice with polymethylmethacrylate nanoparticles as adjuvant for HIV vaccines. AIDS 1991;5:431-5. doi: 10.1097/00002030-199104000-00012 50. Li X, Aldayel AM, Cui Z. Aluminum hydroxide nanoparticles show a stronger vaccine adjuvant activity than traditional aluminum hydroxide microparticles. J Control Release 2014;173:148-57. doi: 10.1016/j.jconrel.2013.10.032 51. Sun B, Ji Z, Liao YP, Wang M, Wang X, Dong J, Chang CH, Li R, Zhang H, Nel AE, Xia T. Engineering an effective immune adjuvant by designed control of shape and crystallinity of aluminum oxyhydroxide nanoparticles. ACS Nano 2013;7:10834-49. doi: 10.1021/nn404211j 52. Cao Y, Ma Y, Zhang M, Wang H, Tu X, Shen H, Dai J, Guo H, Zhang Z. Ultrasmall graphene oxide supported gold nanoparticles as adjuvants improve humoral and cellular immunity in mice. Adv Funct Mater 2014;24:6963-71. doi: 10.1002/adfm.201401358 53. Al-Humadi NH, Siegel PD, Lewis DM, Barger MW, Ma JY, Weissman DN, Ma JK. The effect of diesel exhaust particles (DEP) and carbon black (CB) on thiol changes in pulmonary ovalbumin allergic sensitized Brown Norway rats. Exp Lung Res 2002;28:333-49. doi: 10.1080/01902140290091976 54. Nel AE, Diaz-Sanchez D, Ng D, Hiura T, Saxon A. Enhancement of allergic inflammation by the interaction between diesel exhaust particles and the immune system. J Allergy Clin Immunol 1998; 102:539-54. doi: 10.1016/ S0091-6749(98)70269-6 55. Nygaard UC, Hansen JS, Samuelsen M, Alberg T, Marioara CD, Løvik M. Single-walled and multi-walled carbon nanotubes promote allergic immune responses in mice.

Toxicol Sci 2009;109:113-23. doi: 10.1093/toxsci/kfp057 56. Ryan JJ, Bateman HR, Stover A, Gomez G, Norton SK, Zhao W, Schwartz LB, Lenk R, Kepley CL. Fullerene nanomaterials inhibit the allergic response. J Immunol 2007;179:665-72.

PMID: 17579089 57. Chen EY, Garnica M, Wang YC, Mintz AJ, Chen CS, Chin WC. A mixture of anatase and rutile TiO2 nanoparticles induces histamine secretion in mast cells. Part Fibre Toxicol 2012;9:2. doi: 10.1186/1743-8977-9-2 58. Katwa P, Wang X, Urankar RN, Podila R, Hilderbrand SC, Fick RB, Rao AM, Ke PC, Wingard CJ, Brown JM. A carbon nanotube toxicity paradigm driven by mast cells and the IL-33/ST2 axis. Small 2012;8:2904-12. doi: 10.1002/ smll.201200873 59. Murray AR, Kisin E, Leonard SS, Young SH, Kommineni C, Kagan VE, Castranova V, Shvedova AA. Oxidative stress and inflammatory response in dermal toxicity of singlewalled carbon nanotubes. Toxicology 2009;257:161-71. doi: 10.1016/j.tox.2008.12.023 60. Shannahan JH, Kodavanti UP, Brown JM. Manufactured and airborne nanoparticle cardiopulmonary interactions: A

106

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71.

72.

73.

review of mechanisms and the possible contribution of mast c e l l s . I n h a l To x i c o l 2 0 1 2 ; 2 4 : 3 2 0 - 3 9 . d o i : 10.3109/08958378.2012.668229 Wang X, Katwa P, Podila R, Chen P, Ke PC, Rao AM, Walters DM, Wingard CJ, Brown JM. Multi-walled carbon nanotube instillation impairs pulmonary function in C57BL/6 mice.

Part Fibre Toxicol 2011;8:24. doi: 10.1186/1743-8977-8-24 Azzi J, Tang L, Moore R, Tong R, Haddad NE, Akiyoshi T, Mfarrej B, Yang S, Jurewicz M, Ichimura T, Lindeman N, Cheng J, Abdi R. Polylactide-cyclosporin A nanoparticles for targeted immunosuppression. FASEB J 2010;24:3927-38. doi: 10.1096/fj.10-154690 Higaki M, Ishihara T, Izumo N, Takatsu M, Mizushima Y. Treatment of experimental arthritis with poly(D, L-lactic/ glycolic acid) nanoparticles encapsulating betamethasone sodium phosphate. Ann Rheum Dis 2005;64:1132-6. doi: 10.1136/ard.2004.030759 Xu W, Ling P, Zhang T. Toward immunosuppressive effects on liver transplantation in rat model: tacrolimus loaded poly(ethylene glycol)-poly(D,L-lactide) nanoparticle with longer survival time. Int J Pharm 2014;460:173-80. doi: 10.1016/j.ijpharm.2013.10.035 Shen CC, Wang CC, Liao MH, Jan TR. A single exposure to iron oxide nanoparticles attenuates antigen-specific antibody production and T-cell reactivity in ovalbumin-sensitized BALB/c mice. Int J Nanomedicine 2011;6:1229-35. doi: 10.2147/IJN.S21019 Mitchell LA, Gao J, Wal RV, Gigliotti A, Burchiel SW, McDonald JD. Pulmonary and systemic immune response to inhaled multiwalled carbon nanotubes. Toxicol Sci 2007;100:203-14. doi: 10.1093/toxsci/kfm196 Mitchell LA, Lauer FT, Burchiel SW, McDonald JD. Mechanisms for how inhaled multiwalled carbon nanotubes suppress systemic immune function in mice. Nat Nanotechnol 2009;4:451-6. doi: 10.1038/NNANO.2009.151 Hirst SM, Peairs AD, Gogal R, Seal S, Reilly CM. Cerium oxide nanoparticles decrease inflammation in J774 cells.

FASEB J 2008;22(Meeting Abstracts):758.2.

Shaunak S, Thomas S, Gianasi E, Godwin A, Jones E, Teo I, Mireskandari K, Luthert P, Duncan R, Patterson S, Khaw P, Brocchini S. Polyvalent dendrimer glucosamine conjugates prevent scar tissue formation. Nat Biotechnol 2004;22:97784. doi: 10.1038/nbt995 John AE, Lukacs NW, Berlin AA, Palecanda A, Bargatze RF, Stoolman LM, Nagy JO. Discovery of a potent nanoparticle P-selectin antagonist with anti-inflammatory effects in allergic airway disease. FASEB J 2003;17:2296-8. doi: 10.1096/fj.03-0166fje Magdolenova Z, Collins A, Kumar A, Dhawan A, Stone V, Dusinska M. Mechanisms of genotoxicity. A review of in vitro and in vivo studies with engineered nanoparticles.

Nanotoxicology 2014;8:233-78. doi: 10.3109/17435390.2013.773464 Warheit DB. How meaningful are the results of nanotoxicity studies in the absence of adequate material characterization?

Toxicol Sci 2008:101;183-5. doi: 10.1093/toxsci/kfm279 Brun NR, Lenz M, Wehrli B, Fent K. Comparative effects of zinc oxide nanoparticles and dissolved zinc on zebrafish embryos and eleuthero-embryos: importance of zinc ions.

Sci Total Environ 2014;476-477:657-66. doi: 10.1016/j. scitotenv.2014.01.053

Kononenko V, et al. Nanoparticle interaction with the immune system Arh Hig Rada Toksikol 2015;66:97-108

74. Brunner TJ, Wick P, Manser P, Spohn P, Grass RN, Limbach LK, Bruinink A, Stark WJ. In vitro cytotoxicity of oxide nanoparticles: comparison to asbestos, silica, and the effect of particle solubility. Environ Sci Technol 2006;40:4374-81. doi: 10.1021/es052069i 75. Lin W, Xu Y, Huang CC, Ma Y, Shannon KB, Chen DR, Huang YW. Toxicity of nano- and micro-sized ZnO particles in human lung epithelial cells. J Nanopart Res 2009;11:25-39. doi: 10.1007/s11051-008-9419-7 76. Petrarca C, Perrone A, Verna N, Verginelli F, Ponti J, Sabbioni E, Di Giampaolo L, Dadorante V, Schiavone C, Boscolo P, Mariani Costantini R, Di Gioacchino M. Cobalt nanoparticles modulate cytokine in vitro release by human mononuclear cells mimicking autoimmune disease. Int J Immunopathol Pharmacol 2006;19(4 Suppl):11-4. PMID: 17291400 77. Reale M, Vianale G, Lotti LV, Mariani-Costantini R, Perconti S, Cristaudo A, Leopold K, Antonucci A, Di Giampaolo L, Iavicoli I, Di Gioacchino M, Boscolo P. Effects of palladium nanoparticles on the cytokine release from peripheral blood mononuclear cells of palladium-sensitized women. J Occup E n v i r o n M e d 2 0 11 ; 5 3 : 1 0 5 4 - 6 0 . d o i : 1 0 . 1 0 9 7 / JOM.0b013e318228115e 78. Ariano P, Zamburlin P, Gilardino A, Mortera R, Onida B, Tomatis M, Ghiazza M, Fubini B, Lovisolo D. Interaction of spherical silica nanoparticles with neuronal cells: sizedependent toxicity and perturbation of calcium homeostasis.

Small 2011;7:766-74. doi: 10.1002/smll.201002287 79. Fifis T, Gamvrellis A, Crimeen-Irwin B, Pietersz GA, Li J, Mottram PL, McKenzie IF, Plebanski M. Size-dependent immunogenicity: therapeutic and protective properties of nano-vaccines against tumors. J Immunol 2004;173:3148-54. doi: 10.4049/jimmunol.173.5.3148 80. Inoue K, Takano H, Yanagisawa R, Koike E, Shimada A. Size effects of latex nanomaterials on lung inflammation in mice. Toxicol Appl Pharmacol 2009;234:68-76. doi: 10.1016/j.taap.2008.09.012 81. Mottram PL, Leong D, Crimeen-Irwin B, Gloster S, Xiang SD, Meanger J, Ghildyal R, Vardaxis N, Plebanski M. Type 1 and 2 immunity following vaccination is influenced by nanoparticle size: formulation of a model vaccine for respiratory syncytial virus. Mol Pharm 2007;4:73-84. doi: 10.1021/mp060096p 82. Pan Y, Neuss S, Leifert A, Fischler M, Wen F, Simon U, Schmid G, Brandau W, Jahnen-Dechent W. Size-dependent cytotoxicity of gold nanoparticles. Small 2007;3:1941-9. doi: 10.1002/smll.200700378 83. SchĂśler N, Hahn H, MĂźller RH, Liesenfeld O. Effect of lipid matrix and size of solid lipid nanoparticles (SLN) on the viability and cytokine production of macrophages. Int J Pharm 2002;231:167-76. doi: 10.1016/S03785173(01)00882-1 84. Yang H, Liu C, Yang D, Zhang H, Xi Z. Comparative study of cytotoxicity, oxidative stress and genotoxicity induced by four typical nanomaterials: the role of particle size, shape and composition. J Appl Toxicol 2009;29:69-78. doi: 10.1002/jat.1385 85. Radomski A, Jurasz P, Alonso-Escolano D, Drews M, Morandi M, Malinski T, Radomski MW. Nanoparticleinduced platelet aggregation and vascular thrombosis. Br J Pharmacol 2005;146:882-93. doi: 10.1038/sj.bjp.0706386

Kononenko V, et al. Nanoparticle interaction with the immune system Arh Hig Rada Toksikol 2015;66:97-108

86. Ahamed M, Karns M, Goodson M, Rowe J, Hussain SM, Schlager JJ, Hong Y. DNA damage response to different surface chemistry of silver nanoparticles in mammalian cells.

Toxicol Appl Pharmacol 2008;233:404-10. doi: 10.1016/j. taap.2008.09.015 87. Lee WM, An YJ,Yoon H, Kweon HS. Toxicity and bioavailability of copper nanoparticles to the terrestrial plants mung bean (Phaseolus radiatus) and wheat (Triticum aestivum): plant agar test for water-insoluble nanoparticles.

Environ Toxicol Chem 2008;27:1915-21. doi: 10.1897/07481.1 88. Schöler N, Olbrich C, Tabatt K, Müller RH, Hahn H, Liesenfeld O. Surfactant, but not the size of solid lipid nanoparticles (SLN) influences viability and cytokine production of macrophages. Int J Pharm 2001;221:57-67.

PMID: 11397567 89. Vallhov H, Qin J, Johansson SM, Ahlborg N, Muhammed MA, Scheynius A, Gabrielsson S. The importance of an endotoxin-free environment during the production of nanoparticles used in medical applications. Nano Lett 2006;6:1682-6. doi: 10.1021/nl060860z 90. Fedeli C, Segat D, Tavano R, De Franceschi G, de Laureto PP, Lubian E, Selvestrel F, Mancin F, Papini E. Variations of the corona HDL:albumin ratio determine distinct effects of amorphous SiO 2 nanoparticles on monocytes and macrophages in serum. Nanomedicine (Lond) 2014; 9:248197. PMID: 24661258 91. Tenzer S, Docter D, Kuharev J, Musyanovych A, Fetz V, Hecht R, Schlenk F, Fischer D, Kiouptsi K, Reinhardt C, Landfester K, Schild H, Maskos M, Knauer SK, Stauber RH. Rapid formation of plasma protein corona critically affects nanoparticle pathophysiology. Nat Nanotechnol 2013;8:77281. doi: 10.1038/nnano.2013.181 92. Wang F, Yu L, Monopoli MP, Sandin P, Mahon E, Salvati A, Dawson KA. The biomolecular corona is retained during nanoparticle uptake and protects the cells from the damage induced by cationic nanoparticles until degraded in the lysosomes. Nanomedicine 2013;9:1159-68. doi: 10.1016/j. nano.2013.04.010 93. Gref R, Minamitake Y, Peracchia MT, Trubetskoy V, Torchilin V, Langer R. Biodegradable long-circulating polymeric nanospheres. Science 1994;263:1600-3. doi: 10.1126/ science.8128245 94. Costantini LM, Gilberti RM, Knecht DA. The phagocytosis and toxicity of amorphous silica. PLoS One 2011;6:e14647. doi: 10.1371/journal.pone.0014647 95. Dujmović IH. Comparison of two guidelines on immunotoxicity testing of medicinal products. Arh Hig Rada Toksikol 2005;56:265-8. PMID: 16180612 96. Kawabata TT, Evans EW. Development of immunotoxicity testing strategies for immunomodulatory drugs. Toxicol Pathol 2012;40:288-93. doi: 10.1177/0192623311430238 97. Lankveld DP, Van Loveren H, Baken KA, Vandebriel RJ. In vitro testing for direct immunotoxicity: state of the art.

Methods Mol Biol 2010;598:401-23. doi: 10.1007/978-160761-401-2_26 98. Luebke R. Immunotoxicant screening and prioritization in the twenty-first century. Toxicol Pathol 2012;40:294-9. doi: 10.1177/0192623311427572 99. Luster MI, Portier C, Pait DG, White KL Jr, Gennings C, Munson AE, Rosenthal GJ. Risk assessment in immunotoxicology. I. Sensitivity and predictability of

107 immune tests. Fundam Appl Toxicol 1992;18:200-10. doi: 10.1016/0272-0590(92)90047-L 100. Ong KJ, MacCormack TJ, Clark RJ, Ede JD, Ortega VA, Felix LC, Dang MKM, Ma G, Fenniri H, Veinot JGC, Goss GG. Widespread nanoparticle-assay interference: implications for nanotoxicity testing. PLoS ONE 2014;9:e90650. doi: 10.1371/journal.pone.0090650 101. Wörle-Knirsch JM, Pulskamp K, Krug HF. Oops they did it again! Carbon nanotubes hoax scientists in viability assays.

Nano Lett 2006;6:1261-8. doi: 10.1021/nl060177c 102. Oostingh GJ, Casals E, Italiani P, Colognato R, Stritzinger R, Ponti J, Pfaller T, Kohl Y, Ooms D, Favilli F, Leppens H, Lucchesi D, Rossi F, Nelissen I, Thielecke H, Puntes VF, Duschl A, Boraschi D. Problems and challenges in the development and validation of human cell-based assays to determine nanoparticle-induced immunomo dulatory effects. Part Fibre Toxicol 2011;8:8. doi: 10.1186/1743-8977-8-8 103. Kroll A, Pillukat MH, Hahn D, Schnekenburger J. Interference of engineered nanoparticles with in vitro toxicity assays. Arch Toxicol 2012;86:1123-36. doi: 10.1007/s00204012-0837-z 104. Rajapakse K, Drobne D, Kastelec D, Marinsek-Logar R. Experimental evidence of false-positive Comet test results due to TiO2 particle-assay interactions. Nanotoxicology 2013;7:1043-51. doi: 10.3109/17435390.2012.696735 105. Dobrovolskaia MA, Aggarwal P, Hall JB, McNeil SE. Preclinical studies to understand nanoparticle interaction with the immune system and its potential effects on nanoparticle biodistribution. Mol Pharm 2008;5:487-95. doi: 10.1021/mp800032f 106. Dobrovolskaia MA, Germolec DR, Weaver JL. Evaluation of nanoparticle immunotoxicity. Nat Nanotechnol 2009;4:411-4. doi: 10.1038/nnano.2009.175 107. Dobrovolskaia MA, McNeil SE. Understanding the correlation between in vitro and in vivo immunotoxicity tests for nanomedicines. J Control Release 2013;172:456-66. doi: 10.1016/j.jconrel.2013.05.025 108. Mendoza A, Torres-Hernandez JA, Ault JG, Pedersen-Lane JH, Gao D, Lawrence DA. Silica nanoparticles induce oxidative stress and inflammation of human peripheral blood mononuclear cells. Cell Stress Chaperones 2014;19:777-90. doi: 10.1007/s12192-014-0502-y 109. Capasso L, Camatini M, Gualtieri M. Nickel oxide nanoparticles induce inflammation and genotoxic effect in lung epithelial cells. Toxicol Lett 2014;226:28-34. doi: 10.1016/j.toxlet.2014.01.040 110. Yen HJ, Hsu SH, Tsai CL. Cytotoxicity and immunological response of gold and silver nanoparticles of different sizes.

Small 2009;5:1553-61. doi: 10.1002/smll.200900126 111. Nemmar A, Albarwani S, Beegam S, Yuvaraju P, Yasin J, Attoub S, Ali BH. Amorphous silica nanoparticles impair vascular homeostasis and induce systemic inflammation. Int J Nanomedicine 2014;9:2779-89. doi: 10.2147/IJN.S52818 112. Chou CC, Hsiao HY, Hong QS, Chen CH, Peng YW, Chen HW, Yang PC. Single-walled carbon nanotubes can induce pulmonary injury in mouse model. Nano Lett 2008;8:437-45. doi: 10.1021/nl0723634 113. Nemmar A, Melghit K, Ali BH. The acute proinflammatory and prothrombotic effects of pulmonary exposure to rutile TiO 2 nanorods in rats. Exp Biol Med (Maywood) 2008;233:610-9. doi: 10.3181/0706-RM-165

108

Kononenko V, et al. Nanoparticle interaction with the immune system Arh Hig Rada Toksikol 2015;66:97-108

114. Seiffert J, Hussain F, Guo C, Chang Y, Zhang J, Smith R, Tetley T, Chung F. Inhaled silver nanoparticles induce pulmonary oxidative injury and inflammation: Differential effects between rat strains. Eur Respir J 2014;44(Suppl 58):P3939.

115. Braakhuis HM, Gosens I, Krystek P, Boere J, Cassee FR, Fokkens P, Post J, van Loveren H, Park M. Particle size dependent deposition and pulmonary inflammation after short-term inhalation of silver nanoparticles. Part Fibre Toxicol 2014;11:49. doi: 10.1186/s12989-014-0049-1 116. Yanagisawa R, Takano H, Inoue K, Koike E, Kamachi T, Sadakane K, Ichinose T. Titanium dioxide nanoparticles aggravate atopic dermatitis-like skin lesions in NC/Nga mice.

Exp Biol Med (Maywood) 2009;234:314-22. doi: 10.3181/0810-RM-304

117. Hirai T, Yoshioka Y, Ichihashi K, Mori T, Nishijima N, Handa T, Takahashi H, Tsunoda S, Higashisaka K, Tsutsumi Y. Silver nanoparticles induce silver nanoparticle-specific allergic responses. J Immunol 2014;192(Suppl 1):118.19.

118. Ilves M, Palomäki J, Vippola M, Lehto M, Savolainen K, Savinko T, Alenius H. Topically applied ZnO nanoparticles suppress allergen induced skin inflammation but induce vigorous IgE production in the atopic dermatitis mouse model. Part Fibre Toxicol 2014;11:38. doi: 10.1186/s12989014-0038-4 119. Herzog E, Byrne HJ, Casey A, Davoren M, Lenz AG, Maier KL, Duschl A, Oostingh GJ. SWCNT suppress inflammatory mediator responses in human lung epithelium in vitro.

Toxicol Appl Pharmacol 2009;234:378-90. doi: 10.1016/j. taap.2008.10.015

Interakcije nanodelcev z imunskim sistemom Ko nanodelci vstopijo v organizem, pridejo v kontakt s celicami imunskega sistema. Nezaželene interakcije nanodelcev z imunskim sistemom lahko sprožijo molekularni odziv, ki lahko pripelje do toksičnih učinkov in povečane dovzetnosti organizma za okužbe, avtoimunska obolenja ter razvoj raka. Dosedanje raziskave so pokazale, da nanodelci lahko sprožijo vnetne in alergijske reakcije, lahko pa tudi aktivirajo sistem komplementa. Nanodelci lahko delujejo kot adjuvansi ali kot hapteni. Obstajajo pa tudi poročila, ki kažejo na sposobnost nanodelcev, da zavrejo imunski odziv. V članku bomo povzeli ugotovitve dosedanjih raziskav in vitro ter in vivo, ki so bile narejene na področju proučevanja vplivov nanodelcev na stimulacijo ali supresijo imunskega sistema sesalcev. Za zagotovitev varne uporabe nanodelcev moramo razumeti kako fizikalno-kemijske lastnosti nanodelcev vplivajo na njihovo obnašanje v biološkem okolju. Lastnosti nanodelcev moramo upoštevati tudi ob izvajanju poskusov, da se izognemo lažnim rezultatom zaradi potencialne interference nanodelcev z dejavniki v eksperimentalnem okolju. Čeprav je bilo do sedaj narejenih že več nanotoksikoloških raziskav, je vpliv nanodelcev na imunski sistem še vedno slabo razumljen. Sposobnost nanodelcev za modulacijo imunskega odziva narekuje potrebo po nadaljnjih raziskavah interakcij nanodelcev z imunskim sistemom.

KLJUČNE BESEDE: imunomodulacija; imunotoksičnost; imunski odziv; lastnosti nanodelcev; nanomateriali; nanovarnost

Milić M, et al. Polymorphisms in DNA repair genes: CBMN cytome assay in workers chronically exposed to low doses of ionising radiation Arh Hig Rada Toksikol 2015;66:109-120

Original article

109

DOI: 10.1515/aiht-2015-66-2655

Polymorphisms in DNA repair genes: link with biomarkers of the CBMN cytome assay in hospital workers chronically exposed to low doses of ionising radiation Mirta Milić1,4, Ružica Rozgaj1, Vilena Kašuba1, Ana-Marija Jazbec2, Boris Starčević3, Barnaba Lyzbicki4, Gloria Ravegnini4, Corrado Zenesini5, Muriel Musti5, Patrizia Hrelia4, and Sabrina Angelini4 Institute for Medical Research and Occupational Health1, Faculty of Forestry, University of Zagreb2, University Hospital Dubrava, Department of invasive cardiology3, Zagreb, Croatia, Department of Pharmacy and Biotechnology, University of Bologna4, Department of Public Health, Epidemiological Service, Local Health Authority of Bologna5, Bologna, Italy [Received in April 2015; CrossChecked in April 2015; Accepted in June 2015] Individual sensitivity to ionising radiation (IR) is the result of interaction between exposure, DNA damage, and its repair, which is why polymorphisms in DNA repair genes could play an important role. We examined the association between DNA damage, expressed as micronuclei (MNi), nuclear buds (NBs), and nucleoplasmic bridges (NPBs) and single nucleotide polymorphisms in selected DNA repair genes (APE1, hOGG1, XRCC1, XRCC3, XPD, PARP1, MGMT genes; representative of the different DNA repair pathways operating in mammals) in 77 hospital workers chronically exposed to low doses of IR, and 70 matched controls. A significantly higher MNi frequency was found in the exposed group (16.2±10.4 vs. 11.5±9.4; P=0.003) and the effect appeared to be independent from the principal confounding factor.

Exposed individuals with hOGG1, XRCC1, PARP1, and MGMT wild-type alleles or APEX1, as well as XPD (rs13181) heterozygous showed a significantly higher MNi frequency than controls with the same genotypes. Genetic polymorphism analysis and cytogenetic dosimetry have proven to be a powerful tool complementary to physical dosimetry in regular health surveillance programmes.

KEY WORDS: genotype analysis; micronucleus; nuclear bud; nucleoplasmic bridge; occupational exposure The health consequences of continuous exposure to low doses of ionising radiation (IR) are still a topic of great scientific interest. Medical workers are the most commonly studied group among chronically exposed professionals, with regular medical surveillance and obligatory dosimetry.

The duration and amount of received radiation have significantly decreased over the last decades, with received doses well below the allowable limits of 20 mSv per year.

However, most studies on occupationally exposed subjects have shown an increase in genetic damage after chronic exposure to IR low doses, without evidence of any doseeffect relationship. Nevertheless, recent literature, including a paper on as many as 400.000 nuclear power plant workers, shows significant correlation between accumulated doses and risk of tumour development (1-4).

The cytokinesis-blocked micronucleus (CBMN) assay has been widely used to evaluate DNA damage after occupational, therapeutic or accidental exposure to IR, as well as to assess in vitro radiosensitivity and cancer susceptibility. For instance, in Belgium and Croatia, the Correspondence to: Mirta Milić, PhD, Institute for Medical Research and Occupational Health, Mutagenesis Unit, Zagreb, Croatia; Phone: +385 1 4682 633, Fax: +385 1 46 73 303; E-mail address: mmilic@imi.hr

CBMN assay is regularly applied in the biomonitoring of workers exposed to IR higher than or expected to reach 20 mSv (5). In recent years, the CBMN assay has evolved into the novel cytome assay, where every cell is scored for its damage and mitotic status (6). Originally, the CBMN assay was developed to measure micronuclei (MNi) - whole chromosomes or acentric chromosome fragments that lag behind during anaphase and are not distributed to the main nuclei. Subsequently it was observed that the CBMN assay may also measure other forms of damage, such as nuclear buds (NBs) and nucleoplasmic bridges (NPBs). NBs have been proposed as markers of gene amplification and/or altered gene dosage; NPBs provide a measure of chromosome rearrangement, or DNA misrepair, and may break to form MNi (6). The CBMN assay also allows one to calculate the nuclear division index (NDI), providing information on a cell cycle’s delay with regard to exposure.

Individual sensitivity to IR exposure is the result of a close interaction between DNA damage and DNA repair.

Several authors have already described the association between genetic damage, IR exposure, and polymorphisms in DNA repair genes (7-10).

to low doses of ionising radiation Arh Hig Rada Toksikol 2015;66:109-120 110 Milić M, et al. Polymorphisms in DNA repair genes: CBMN cytome assay in workers chronically exposed

To the best of our knowledge, this is the first study to evaluate if biomarkers of the CBMN cytome assay (MNi, NPBs, NBs, and NDI) could be sensitive enough to evaluate the impact of occupational exposure to IR low doses.

Moreover, we investigated the possible influence on biomarkers of the CBMN cytome assay of a comprehensive panel of polymorphisms in DNA repair genes. Included in the analysis were polymorphisms in the APE1, hOGG1, XRCC1, XRCC3, XPD, PARP1, and MGMT genes, representative of the different DNA repair pathways operating in mammals.

MATERIALS AND METHODS Study population The study included 77 medical workers occupationally exposed to low doses of IR, and 70 controls who had never been occupationally exposed to IR or other known carcinogenic agents. All of the subjects gave their written consent after being informed on the study scope and experimental details. The study followed the guidelines of the Declaration of Helsinki regarding medical research involving human subjects, and was approved by an ethics committee. Standardised questionnaires were administered to all of the participants to determine their sociodemographic characteristics, medical history (e.g. history of medical treatments, radiography, recent vaccination, severe infections, or viral diseases over the past six months, presence of known inherited genetic disorders and chronic diseases, family history of cancer), and individual life styles [e.g. smoking, alcohol consumption, dietary habits, including deficient or peculiar habits (e.g. vegetarian or vegans), or intake of multi-vitamins supplements, and use of contraceptive]. Exclusion criteria included the use of any therapeutic drugs, radiotherapy, diagnostic X-rays undergone 12 months prior to sampling, which could have significantly contributed to the received dose and/or genetic damage. For medical workers, the questionnaires covered the duration of occupational exposure to IR. Selected demographic characteristics of the study population are reported in Table 1. Among the medical workers, we distinguished between seven different working tasks: gastroenterologist, interventional cardiologist, anaesthesiologists, surgeons, radiologists, and engineers of radiology. All of the IR exposed workers were under medical surveillance and regular film dosimetry, however according to the written consent provide to the subjects, we only know the annual dose received did not exceed the limit of 20 mSv per year.

Venous blood was obtained from each subject and transferred to the laboratory within a few hours for subsequent CBMN assay and genotype analysis.

Table 1 Demographic characteristics of the study population Exposed

Controls

Total

77

70

147

46 (59.7) 31 (40.3)

26 (37.1) 44 (62.9)

72 (49.0) 75 (51.0)

42.2 ± 10.6 23-69

40.8 ± 10.4 20-60

41.5 ± 10.5 20-69

Sample size (n) Sex Female n (%) Male n (%) Age mean±SD range

Smoking status Never (%) 47 (61.0) 50 (71.4) 97 (66.0) Current (%) 30 (39.0) 20 (28.6) 50 (34.0) Years of exposure mean±SD 13.7±8.9 range 1-38 *Significantly lower compared to controls P=0.032 (Wilcoxon test)

CBMN cytome assay Cultures for the CBMN assay were set up in triplicate.

Lymphocytes were cultured in RPMI 1640 medium (Gibco, Paisley, UK), supplemented with 1 % of phytohaemagglutinin, (Apogent, USA), 20 % of foetal calf serum (Gibco), and 1 % penicillin-streptomycin solution (Sigma-Aldrich, St Louis, MO, USA) and incubated at 37 °C in humidified 5 % CO2 atmosphere for 72 h. After 44 h, all cultures were supplemented with cytochalasin B (Sigma; final concentration 6 µg mL-1). At the end of the incubation period, lymphocytes were subjected to a mild hypotonic treatment, fixed, and stained according to Kapka et al. (11).

One thousand binucleated lymphocytes with well-preserved cytoplasm per subject were analysed. The criteria for analysis of MNi, NPBs, and NBs were as described by Fenech et al. (12). NDI was calculated according to the formula NDI=(1M1+2M2+3M3+4M4)/1000 cells where M1-M4 indicates lymphocytes with 1 to 4 nuclei (13).

DNA isolation Genomic DNA was isolated from EDTA-anticoagulated whole blood using standard sodium perchlorate/chloroform extraction procedures or the QIAamp DNA Blood kit (Qiagen, Hilden, Germany) as recommended by the manufacturer. For genotype analysis, DNA samples were diluted and stored as 10 ng μL -1 aliquots at -20 °C. Genotyping was performed by PCR-based assays: RFLP and/or real-time (Table 2) (9, 14, 15). Negative controls were included in each reaction as quality control.

Genotyping by real-time PCR was performed by the 5′-nuclease allelic discrimination assay (TaqMan®, Applied Biosystems, Foster City, CA, USA), according to the manufacturer’s instruction. Genotype screening was carried out simultaneously in a blinded manner to work allocation (exposed, non-exposed). Genotype results were regularly confirmed by repetition of 90 % of the samples.

Milić M, et al. Polymorphisms in DNA repair genes: CBMN cytome assay in workers chronically exposed to low doses of ionising radiation Arh Hig Rada Toksikol 2015;66:109-120

111

Table 2 Details on investigated SNPs in DNA repair genes GENE rs unique code [base and amino acid change]

METHOD[A]

APEX1 rs1130409 [gaT > gaG; Asp148Glu] hOGG1 rs1052133 [tCc > tGc; Ser326Cys]

RFLP [Hu et al., 2002] (14) RT TaqMan assay C_8921503_10 RFLP [Godderis et al., 2006] (15) RT TaqMan assay C_3095552_1

XPD rs1799793 [Gac > Aag; Asp312Asn] rs13181 [Aag > Cag; Lys751Gln] XRCC1 rs861539 [cAg > cGg; Gln399Arg]

RT TaqMan assay C_3145050_10 RFLP [Angelini et al., 2005] (8) RT TaqMan assay C_3145033_10 RFLP [Angelini et al., 2005] (8) RT TaqMan assay C_622564_10

XRCC3 rs861539 [aCg > aTg; Thr241Met] PARP1 rs1136410 [gTg > gCg; Val762Gly]

RFLP [Angelini et al., 2005] (8) RT TaqMan assay C_3145033_10 RT TaqMan assay C_1515368_1_

MGMT rs12917 [Ctt > Ttt; Leu115Phe]

RT TaqMan assay C_3157955_10

RFLP=PCR-RFLP analysis carried out according to published methods (reference parenthetically); RT=Real-Time PCR with TaqMan allelic discrimination assay (Applera, Foster City, USA) [A]

Statistical analysis The Wilcoxon rank-sum test was used to test the difference in MNi, NB, and NPB frequency between exposed workers and controls. The association between MNi, NB, and NPB frequency and the various genotypes was tested by Kruskal-Wallis test. The distribution of genotypes was tested for Hardy-Weinberg (HW) equilibrium using the online HW test tool offered by the Institute for Human Genetics, Technical University Munich, Germany.

Linear regression analysis was applied to assess the correlation between years of IR exposure and MNi, NBs, and NPBs in the exposed workers. The Poisson regression analysis was applied to evaluate the influence of age, sex, smoking status, and working task on MNi, NPBs, and NBs in the overall population and in both groups separately. The level of significance was set at P<0.05; statistical analysis was conducted using Stata Intercooled version 11.0 (16).

RESULTS The principal demographic characteristics of the study population, both overall and by group are reported in Table 1. In summary, age distribution was similar in the two groups (P=0.212); while sex was significantly different, as females were over-represented in the IR-exposed individuals than controls (59.7 vs. 37.1 %, P=0.006). Regarding

smoking status, 39.0 % of the exposed workers and 28.6 % controls were smokers at the time of sampling. No difference was observed between the two groups in terms of years of smoking or daily cigarette consumption, all being mild smokers (less than 10 cigarettes per day).

CBMN assay Results of CBMN cytome assay (MNi, NPB, NB frequencies and NDI) are presented in Table 3. Of the four analysed parameters, only MNi frequency was significantly higher in the exposed workers than in controls (P=0.003).

The frequencies of the other parameters were similar between the exposed and controls. The range of IR exposure duration in radiological workers, i.e. years of employment, was 1-38 (Table 1). Linear regression analysis revealed a significant association between years of employment and MNi (β=0.403, P=0.003; Figure 1A) and NBs (β=0.075, P=0.027; Figure 1B), whereas no association emerged with NPBs (β=0.024, P=0.230).

The results of Poisson regression analysis reporting the influence of confounding factors, which included age, sex, and smoking status on MNi, NPBs, and NBs, are reported in Table 3 to Table 6 respectively. With regard to MNi (Table 4), an increase in age resulted in a significant increase in MNi frequency both in controls and exposed workers (P<0.0001 for both). Sex exerted a significant influence on the yield of MNi only in controls, with the frequency being

to low doses of ionising radiation 112 Milić M, et al. Polymorphisms in DNA repair genes: CBMN cytome assay in workers chronically exposed Arh Hig Rada Toksikol 2015;66:109-120

Table 3 Results of micronucleus assay in the study groups Exposed

Controls

Micronuclei Mean±SD 16.2±10.4* 11.5±9.4 (Range) 1-47 0-37 Nucleoplasmatic bridges Mean±SD 0.9±1.5 1.7±4.0 (Range) 0-8 0-30 Nuclear buds Mean±SD 1.7±2.6 2.2±3.5 (Range) 0-15 0-23 Nuclear division index Mean±SD 1.8±0.3 1.9±0.4 (Range) 1.2-2.3 0-2.7 * Significantly higher compared to controls P=0.003 (Wilcoxon test)

higher in females as compared to males (P=0.002). Smoking status significantly influenced MNi frequency in both study groups, however at an opposite trend: being higher in current smokers compared to non-smokers in the IR exposed group, whereas in controls current smokers were characterized by lower MNi frequency than those that had never smoked (P<0.0001 for both). Regarding NPB frequency (Table 5), only sex exerted a significant influence in both study groups. In particular, NPB frequency was significantly higher in males compared to females (P=0.004 in IR exposed; P=0.015 in controls). In the control group, we also observed a significant influence of age (P<0.0001).

As regard to NBs (Table 6), in the IR exposed group we observed a significant age-dependent effect (P=0.001).

Non-smokers were characterised by a significantly higher NB frequency as compared to current smokers (P=0.002).

In the control group, we observed the influence of sex, with NBs being significantly higher in males compared to females (P=0.008), and a significant age-dependent effect (P=0.008).

Poisson regression analysis applied to the overall study population highlighted the significant influence of age on MNi, NPBs, and NBs. In all three instances, an increase of age was associated with a significant increase of the observed frequency of DNA damage (P<0.0001 for all).

With regard to sex, MNi frequency was lower in males, although not significantly (P=0.064), whereas the opposite results were observed for NPB and NB frequency, being significantly higher in males than female (P<0.0001 and P=0.013 respectively). We observed a significant effect of smoking status only in NBs; surprisingly this biomarker was higher in non-smokers compared to active smokers (P<0.0001). Interestingly, when dividing smokers into two classes (0-10 and >10 years) of smoking habits - a significant effect was observed only in the second group.

In particular, individuals who smoked for more than 10 years exhibited a significantly lower NBs frequency (IRR 0.579, 95 % CI 0.403-0.833; P=0.003), whereas no effect was observed in those who smoked less than 10 years.

Table 4 Poisson regression analysis of confounding factors on MNi frequencies Confounding factorsa IRR

P

95 % CI

All Age (years) 1.020 <0.0001 1.016-1.025 Sex (0,1) 0.918 0.064 0.839-1.004 Smoking status (0,1) 1.047 0.335 0.954-1.149 Exposure (years) 1.332 <0.0001 1.214-1.460 Controls Age (years) 1.018 <0.0001 1.011-1.026 Sex (0,1) 0.801 0.002 0.696-0.923 Smoking status (0,1) 0.729 <0.0001 0.612-0.867 Exposed workers Age (years) 1.020 <0.0001 1.015-1.026 Sex (0,1) 0.971 0.621 0.864-1.091 Smoking status (0,1) 1.227 <0.0001 1.094-1.376 IRR: Incidence Rate Ratio a Sex: 0-Female, 1-Male; Smoking status: 0-Never, 1-Current Table 5 Poisson regression analysis of confounding factors on NPBs frequencies Confounding factorsa IRR

P

95 % CI

All Age (years) 1.031 <0.0001 1.016-1.046 Sex (0,1) 1.771 <0.0001 1.290-2.433 Smoking status (0,1) 0.863 0.377 0.623-1.196 Exposure (years) 0.621 0.002 0.458-0.843 Controls Age (years) 1.049 <0.0001 1.029-1.070 Sex (0,1) 1.700 0.015 1.111-2.603 Smoking status (0,1) 1.013 0.951 0.660-1.557 Exposed workers Age (years) 1.008 0.457 0.987-1.030 Sex (0,1) 2.021 0.004 1.258-3.246 Smoking status (0,1) 0.718 0.189 0.439-1.176 IRR: Incidence Rate Ratio a Sex: 0-Female, 1-Male; Smoking status: 0-Never, 1-Current Table 6 Poisson regression analysis of confounding factors on NBs frequencies Confounding factorsa

IRR

P

95 % CI

All Age (years) 1.025 <0.0001 1.013-1.037 Sex (0,1) 1.368 0.013 1.069-1.751 Smoking status (0,1) 0.597 <0.0001 0.448-0.795 Exposure (years) 0.841 0.164 0.660-1.073 Controls Age (years) 1.023 0.008 1.006-1.040 Sex (0,1) 1.635 0.008 1.136-2.353 Smoking status (0,1) 0.703 0.088 0.469-1.054 Exposed workers Age (years) 1.029 0.001 1.012-1.047 Sex (0,1) 1.166 0.396 0.817-1.663 Smoking status (0,1) 0.530 0.002 0.354-0.795 IRR: Incidence Rate Ratio a Sex: 0-Female, 1-Male; Smoking status: 0-Never, 1-Current

Milić M, et al. Polymorphisms in DNA repair genes: CBMN cytome assay in workers chronically exposed to low doses of ionising radiation Arh Hig Rada Toksikol 2015;66:109-120

113

Figure 1 Relationship between DNA damage, assessed as micronuclei [MNi (A)] and nuclear buds [NBs (B)] in peripheral blood lymphocytes, and years of exposure to ionising radiation. The thick line represents the linear regression analysis of the data

With regard to exposure, Poisson analysis revealed that an increase in the years of IR exposure was associated with a significant increase in MNi frequency (P<0.0001) and surprisingly a significant decrease of NPBs (P=0.002). No effect of years of exposure on NBs was observed.

Genotype distribution and influence on biomarkers of the cytome assay Details on the investigated SNPs in DNA repair genes are represented in Table 2. Genotype distribution of the eight studied DNA repair genes among the different study groups and in the overall population is presented in Table 7. Frequencies of the variant allele observed in our study were consistent with those reported in the publicly available database NCBI (dbSNP) for Caucasians. Deviation from the HW equilibrium was observed for one SNP in the IR exposed group (XPD rs13181); departure from HW equilibrium was not observed for any other SNPs in any of the studied group, or in the overall population. Allele frequencies were similar between the IR exposed and controls, with the exception of hOGG1 rs1052133 and

PARP1 rs1136410 SNPs. To be more precise, the hOGG1 variant allele and PARP1 wild-type allele were significantly underrepresented in controls compared to IR exposed (P=0.0017, and P=0.039 respectively).

Distribution of MNi, NPBs and NBs by DNA repair genotypes and exposure status are shown in Table 8 to Table 10 respectively. For several genes, due to the small number of individuals homozygous for the variant allele, the approach was to group them together with heterozygous in order to increase statistical power. Among the controls, PARP1 wild type allele was associated with significantly higher NBs compared to the combined homozygous SNP plus heterozygous genotype (2.5±3.0 vs. 1.0±1.4, P=0.044; Table 9). In the IR exposed group, the same genotype was also significantly associated with higher NPBs (1.2±1.7 vs.

0.4±0.8, P=0.027; Table 10). No detectable influence of other genotypes on MNi, NBs or NPBs was observed within the two groups. When we compared IR exposed workers with controls, a significantly higher MNi frequency was found in radiological workers homozygous wild-type for the hOGG1, XRCC1, XRCC3, MGMT1, and PARP1 compared to controls with the same genotypes (Table 8).

to low doses of ionising radiation 114 MiliÄ&#x2021; M, et al. Polymorphisms in DNA repair genes: CBMN cytome assay in workers chronically exposed Arh Hig Rada Toksikol 2015;66:109-120

Table 7 Genotype distribution, allele frequency and Hardy-Weinberg equilibrium of the investigated SNPs in DNA repair genes Genotype*

Exposed (n=77)

Controls (n=56)**

Total (n=133)

0

26 (33.8)

19 (33.9)

45 (33.8)

1

35 (45.5)

27 (48.2)

62 (46.6)

2

16 (20.7) q=0.44; PHWE=0.509

10 (17.9) q=0.42; PHWE=0.939 Pafd=0.802

26 (19.6) q=0.43; PHWE=0.578

0

38 (49.3)

42 (75.0)

80 (60.1)

1

34 (44.1)

14 (25.0)

48 (36.1)

2

5 (6.6) q=0.29; PHWE=0.285

0 (0.0) q=0.12; PHWE=0.473 Pafd=0.0017

5 (3.8) q=0.22; PHWE=0.501

0

31 (40.3)

17 (30.4)

48 (36.1)

1

33 (42.8)

33 (58.9)

66 (49.6)

2

13 (16.9) q=0.38; PHWE=0.413

6 (10.7) q=0.40; PHWE=0.091 Pafd=0.758

19 (14.3) q=0.39; PHWE=0.628

0

32 (41.6)

15 (27.2)

47 (35.3)

1

27 (35.1)

33 (58.9)

60 (45.1)

2

18 (23.3) q=0.41; PHWE=0.016

8 (13.9) q=0.44; PHWE=0.140 Pafd=0.643

26 (19.6) q=0.42; PHWE=0.389

0

31 (40.2)

22 (39.3)

53 (39.9)

1

41 (53.2)

25 (44.6)

66 (49.6)

2

5 (6.6) q=0.33; PHWE=0.076

9 (16.1) q=0.38; PHWE=0.674 Pafd=0.374

14 (10.5) q=0.35; PHWE=0.322

0

23 (29.9)

23 (42.6)

46 (35.1)

1

38 (49.3)

25 (46.3)

63 (48.1)

2

16 (20.8) q=0.45; PHWE=0.967

6 (11.1) q=0.34; PHWE=0.838 Pafd=0.070

22 (16.8) q=0.41; PHWE=0.956

0

55 (71.4)

33 (58.9)

88 (66.2)

1

21 (27.3)

18 (32.1)

39 (29.3)

2

1 (1.3) q=0.15; PHWE=0.520

5 (9.0) q=0.25; PHWE=0.285 Pafd=0.039

6 (4.5) q=0.19; PHWE=0.534

0

56 (72.7)

39 (69.6)

95 (71.4)

1

18 (23.4)

16 (28.6)

34 (25.6)

2

3 (3.9) q=0.16; PHWE=0.328

APEX1 rs1130409

hOGG1 rs1052133

XPD rs1799793

XPD rs13181

XRCC1 rs861539

XRCC3 rs861539

PARP1 rs1136410

MGMT rs12917

1 (1.8) q=0.16; PHWE=0.658 Pafd=0.914 *0=homozygous wild-type (wt); 1=Heterozygous (HE); 2=homozygous polymorphic (SNP) q=frequency of the SNP PHWE=P-value of the Hardy-Weinberg Equilibrium Pafd=P-value of the allele frequency difference between controls and IR exposed **14 samples are missing due to lack of biological material

4 (3.0) q=0.16; PHWE=0.655

Milić M, et al. Polymorphisms in DNA repair genes: CBMN cytome assay in workers chronically exposed to low doses of ionising radiation Arh Hig Rada Toksikol 2015;66:109-120

115

Table 8 Mean MNi frequencies in the study population subdivided by exposure status and genotype distribution Exposed (n=77) Gene*

Subject (n)

Controls (n=56)

Mean±SD

95 % CI

Subject (n)

Mean±SD

95 % CI

15.1±10.2 17.3±11.2a 15.5±9.5

11.0 -19.2 13.5-21.1 10.4-20.6

19 27 10

11.0±10.3 10.3±8.8 12.2±7.1

6.0-16.0 6.8-13.8 7.1-17.3

12.4-19.1 13.1-20.1

42 14

10.6±9.2 11.7±8.6

7.7-13.4 6.7-16.7

APEX1 rs1130409 0 1 2

26 35 16

hOGG1 rs1052133 0 1+2

38 39

15.8±10.2 b 16.6±10.8

0 1+2

31 46

16.5±11.1 16.0±10.1 c

12.4-20.6 16.0-19.0

17 39

11.4±7.3 10.6±9.7

7.7-15.2 7.5-13.8

0 1 2

32 27 18

16.7±10.8 16.4±10.7 d 15.0±9.8

12.8-20.6 12.1-20.6 10.1-19.9

15 33 8

11.6±7.7 9.8±8.5 14.0±12.8

7.3-15.8 6.7-12.8 3.3-24.7

XPD rs1799793 XPD rs13181

XRCC1 rs861539 0 1+2

31 46

17.3±12.3 e 15.4±9.0

12.8-21.8 12.8±18.1

22 34

9.6±8.8 11.6±9.1

5.7-13.5 8.5-14.8

0 1+2

23 54

18.3±9.9 f 15.3±10.6 g

14.0-22.6 12.4-18.2

23 31

11.8±9.8 9.7±8.4

7.6-16.0 6.6-12.7

XRCC3 rs861539

PARP1 rs1136410 0 1+2

55 22

17.3±10.5 h 13.5±9.9

14.4-20.1 9.1-17.9

33 23

11.3±8.9 10.4±8.8

8.2-14.5 6.2-14.2

0 1+2

56 21

18.2±10.6 i 11.0±8.0

15.3-21.0 7-3-14.6

39 17

11.1±9.2 10.2±8.7

8.1-14.1 5.8-14.7

MGMT rs12917

*0=homozygous wild-type (wt); 1=Heterozygous (HE); 2=homozygous polymorphic (single nucleotid polymorphism, SNP) Significantly different from controls with the same genotypes (Wilcoxon test): aP=0.005; bP=0.012; cP=0.004; dP=0.003; eP=0.010; f P=0.017; gP=0.007; hP=0.008; iP=0.001

With regard to XRCC3, MNi frequency was also significantly higher in IR exposed homozygous SNP pooled with heterozygous, compared to controls with the same pooled genotype. In addition, MNi frequency was also significantly higher in IR exposed heterozygous for APEX1 and XPD rs13181. Regarding XPD rs1799793, we also observed a significantly higher MNi frequency in homozygous SNP pooled with heterozygous, compared to controls with the same pooled genotype. Concerning the APEX1 genotype, we also observed an influence on NB frequency. In particular, significantly lower NBs frequency was observed in IR exposed workers homozygous SNP compared to controls with the same genotype (Table 9).

DISCUSSION CBMN assay has already shown to be a reliable biomarker in the evaluation of IR exposure in different settings, including radiotherapy and occupational and accidental environmental exposure (17). The strength and novelty of the present study lies in its investigation of biomarkers of the CBMN cytome assay, which includes NPBs and NBs in addition to MNi. To the best of our knowledge, this is the first study to analyse NPBs and NBs in individuals occupationally exposed to IR-low doses. The

analysis of NPBs and NBs in lymphocytes has become increasingly important for their sensitivity in revealing chromosomal damage in humans. The results of our study showed that MNi frequency was significantly higher in the IR exposed workers compared to controls. The significance was unlikely to be related to differences in age, sex, and smoking within the two groups, as the multivariate analysis showed that MNi frequency was significantly influenced

by IR exposure. This finding confirms the reliability of the MNi frequency as biological dosimetry in population occupationally exposed to low doses of IR, as shown in several earlier cytogenetic studies (9, 10, 18-23). To date, none of the studies on IR occupationally exposed workers have included the analysis of NPBs and NBs. Surprisingly, in our study these biomarkers were higher in controls compared to IR exposed workers, however the differences were small and statistically not significant. None of the occupationally exposed subjects studied here had ever exceeded the permitted radiation limit for occupational exposure, recommended by the International Commission on Radiological Protection (ICRP). The lack of knowledge of the IR dose equivalent to the whole body (external wholebody dose equivalent, Hwb), accumulated over the entire working-life period may represent a limitation of the present study. On the other hand, most studies to date have failed

to low doses of ionising radiation Arh Hig Rada Toksikol 2015;66:109-120 116 Milić M, et al. Polymorphisms in DNA repair genes: CBMN cytome assay in workers chronically exposed Table 9 Mean NB frequencies in the study population subdivided by exposure status and genotype distribution Exposed (n=77) Gene*

Subject (n)

Controls (n=56)

Mean±SD

95 % CI

Subject (n)

Mean±SD

95 % CI

26 35 16

1.5±2.0 1.9±2.5 1.6±3.8 a

0.7-2.3 1.1-2.8 0.6-3.8

19 27 10

1.2±1.9 2.0±2.6 3.0±3.1

0.3-2.1 1.0-3.1 0.8-5.2

0 1+2

38 39

1.4±2.3 1.9±2.9

0.7-2.2 1.0-2.9

42 14

1.8±2.6 2.3±2.5

1.0-2.6 0.8-3.7

0 1+2

31 46

1.7±1.7 1.7±3.1

1.1-2.3 0.8-2.6

17 39

3.0±3.4 1.4±1.9

1.2-4.8 0.8-2.0

0 1 2

32 27 18

1.9±1.7 1.6±3.3 1.5±3.0

1.2-25 0.3-2.9 0.1-3.0

15 33 8

2.6±3.8 1.7±1.8 1.6±2.1

0.5-4.7 1.0-2.3 0.1-3.4

0 1+2

31 46

1.3±1.9 2.0±3.0

0.6-2.0 0.7-1.8

22 34

2.1±3.2 1.8±2.0

0.7-3.5 1.1-2.5

0 1+2

23 54

1.6±3.2 1.8±2.4

0.2-3.0 1.1-2.4

23 31

2.0±2.3 2.0±2.8

1.0-2.9 0.9-3.0

55 22

1.9±2.8 1.2±2.0

1.1-2.7 0.4-2.1

33 23

2.5±3.0 1.0±1.4b

1.5-3.6 0.5-1.6

0 56 1.9±2.9 1.1-2.7 39 1+2 21 1.2±1.6 0.5-2.0 17 *0=homozygous wild-type (wt); 1=Heterozygous (HE); 2=homozygous polymorphic (SNP) a Significantly different from controls with the same genotypes (Wilcoxon test, P=0.041) b Significantly different from homozygous wild-type controls (Kruskal-Wallis test, P=0.044)

1.7±2.4 2.4±2.9

0.9-2.5 0.3-2.8

APEX1 rs1130409 0 1 2

hOGG1 rs1052133 XPD rs1799793 XPD rs13181

XRCC1 rs861539 XRCC3 rs861539 PARP1 rs1136410 0 1+2

MGMT rs12917

Table 10 Mean NPB frequencies in the study population subdivided by exposure status and genotype distribution Gene* Subject (n) APEX1 rs1130409

Exposed (n=77) Mean±SD

95 % CI

Subject (n)

Controls (n=56) Mean±SD

95 % CI

0 1 2

26 35 16

1.0±1.8 1.0±1.4 0.8±1.5

0.3-1.7 0.5-1.4 0.1-1.6

19 27 10

0.7±1.4 1.7±2.2 1.8±2.3

0.1-1.4 0.8-2.5 0.1-3.5

0 1+2

38 39

0.7±1.1 1.2±1.9

0.4-1.1 0.6-1.8

42 14

1.1±1.8 2.1±2.6

0.6-1.7 0.6-3.6

31 46

0.7±1.0 1.1±1.8

0.3-1.1 0.6-1.7

17 39

1.4±2.3 1.4±1.9

0.2-2.6 0.7-2.0

0 1 2

32 27 18

1.0±1.8 0.6±0.9 1.3±1.8

0.4-1.7 0.3-1.0 0.4-2.2

15 33 8

1.5±2.4 1.3±1.8 1.5±2.5

0.1-2.8 0.7-1.9 0.1-3.5

0 1+2

31 46

0.7±1.0 1.1±1.8

0.3-2.1 0.6-1.6

22 34

1.6±2.6 1.2±1.6

0.5-2.8 0.7-1.8

0 1+2

23 54

1.3±2.0 0.8±1.2

0.4-2.1 0.5-1.2

23 31

1.5±2.3 1.3±1.9

0.5-2.5 0.6-2.0

0 1+2

55 22

1.2±1.7 0.4±0.8a

0.7-1.6 0.1-0.8

33 23

1.8±2.5 0.7±0.8

1.0-2.7 0.4-1.1

1.3±1.9 1.5±2.4

0.7-1.9 0.3-2.8

hOGG1 rs1052133 XPD rs1799793 0 1+2

XPD rs13181

XRCC1 rs861539 XRCC3 rs861539 PARP1 rs1136410 MGMT rs12917

0 56 1.1±1.5 0.6-1.5 39 1+2 21 0.7±1.5 0.1-1.3 17 *0=homozygous wild-type (wt); 1=Heterozygous (HE); 2=homozygous polymorphic (SNP) a Significantly different from homozygous wild-type IR exposed (Kruskal-Wallis test, P=0.027)

Milić M, et al. Polymorphisms in DNA repair genes: CBMN cytome assay in workers chronically exposed to low doses of ionising radiation Arh Hig Rada Toksikol 2015;66:109-120

to find any association between chromosome damage and Hwb accumulated after chronic exposure to low doses (9, 19, 20, 22, 24, 25). From this point of view, the years the workers were occupationally exposed to IR may represent an attempt to overcome the limits associated with the use of dosimeters. Positive correlation of MNi frequency and exposure duration has been confirmed in previous studies (23, 26), with a reported elevated high risk of cancer in medical X-ray workers as their service duration increased (2). Our IR exposed group demonstrated a clear relation between MNi and NB frequencies, that is, an increase of 0.4 and 0.07 in the number of MNi and NBs respectively, per 1 year of IR occupational exposure.

Age, sex, and smoking status are common confounding factors known to affect MNi frequency, whereas information on NBs and NPBs are limited. Despite the relatively small sample size, our data confirm the effect of aging on MNi frequency in all of the studied groups – i.e. exposed, controls and overall population. The same age dependent effect was seen for NBs, a finding that might not be surprising as the nuclear budding process has emerged as another unique mechanisms of MNi formation. NBs are also associated with the alteration of DNA stability, with evidence that these structures contain entire or fragments of chromosomes (6, 27). Our study is in agreement with an international collaborative study on pooled data from 25 laboratories that confirmed the impact of sex on MNi frequencies among subjects involved in occupational and environmental surveys (28), although in our case female subjects differed significantly from the male only in the control group. In the entire studied population, men had a higher NPB frequency, as in another study on healthy volunteers (29), but the knowledge of the effect of confounding factor on NPBs frequencies is not as extensive as on MNi. Generally, the effects of smoking on DNA damage in individuals exposed to IR are still unclear, with opposite results (9, 20, 24, 26, 30, 31). In IR exposed subjects, smoking habits were associated with increased MNi frequency, which is in agreement with previous studies (20, 30), but there was a significantly decreased MNi frequency in the control group.

Interestingly, NB frequency was significantly decreased both in the IR exposed workers and in the overall population.

In our study population, none of the individuals reported being heavy smokers (≥30 cigarettes per day), which made it impossible to further study the potential increase of genetic damage, primarily MNi. It has been shown that MNi frequency among occupationally and environmentally exposed individuals is influenced only in non-exposed heavy smokers (32) and slightly reduced in smokers exposed to genotoxic agents, with two possible explanations: appearance of apoptotic/necrotic cells due to cigarette smoke damage that would not be detected in the CBMN assay (20,29) and possible adaptive response stimulation caused by the intake of a few cigarettes per day, causing a lowering in MN frequency (29).

117

Another critical consideration is the cellular response to IR low doses. Exposure is a complex mechanism, leading to the activation of multiple signal transduction pathways, which besides DNA repair include apoptosis, proliferation, inflammation, and genomic instability. Therefore, different genes, belonging to these different pathways, and characterised by several polymorphisms, may contribute to the individual genome sensitivity in IR exposed subjects (8-10, 33-36). Genotype analysis revealed an association between all of the investigated polymorphisms and MNi frequency, while only the APEX1 polymorphism was associated with NB frequency. With regard to the XRCC3 polymorphism, the effect on MNi is unclear. Both genotype groups in the IR exposed group had significantly higher MNi frequencies compared to controls with the same genotypes, with no difference within the group. This hypothesis corroborates in vitro studies that reported an association between increased sensitivity to IR in human lymphocytes and SNPs in DNA repair genes (7, 36). It is feasible that gene-gene interactions may influence DNA damage in response to dose; however, the reduction in sample sizes, as a consequence of successive categorization, even further limits the already weak statistical power (due to small sample size). It is undoubtedly possible that some of the associations become significant by chance, due to the inadequate statistical power. However, this does not rule out an association between DNA repair gene polymorphisms and DNA damage frequency (36). Achieving a complete understanding of this interaction may be of great importance for implementing radiation protection and radiotherapy programmes (37, 38). In view of this, concomitant analysis of DNA damage and interindividual differences in DNA repair genes, due to the presence of polymorphisms, may represent a valuable multi-biomarker approach (39).

In conclusion, our group was the first to score MNi, NBs, and NPBs in IR exposed workers. Clearly, the results obtained confirmed the genotoxic implication resulting from the occupational exposure to IR low doses. Moreover, we believe we have confirmed the value of the MNi frequency as a standard and powerful cytogenetic method for studying genotoxicity in populations exposed to IR low doses.

Whether the cytome assay, meaning NBs and NPBs that were not found to be higher, could improve the predictive capacity of the assay to reveal genetic damage after chronic IR low dose exposure, remains to be determined. Obviously, before ruling out the usefulness of NB and NPB analysis in IR exposed individuals, it would be desirable to replicate the study on a larger population. Nevertheless, it might be interesting to apply the cytome assay in populations exhibiting different IR exposure levels. In particular, interventional cardiologists are currently exposed to a significantly radiation risk compared to other occupational IR exposed individuals.

to low doses of ionising radiation 118 Milić M, et al. Polymorphisms in DNA repair genes: CBMN cytome assay in workers chronically exposed Arh Hig Rada Toksikol 2015;66:109-120

Acknowledgement This study was partly supported by the Ministry of Science, Education and Sports of the Republic of Croatia (grant number 0022-0222148-2137). Dr Mirta Milić’s visit to the University of Bologna was made possible thanks to a fellowship awarded by the Ministry of Science, Education and Sports of the Republic of Croatia. Corrado Zenesini and Muriel Musti were supported by a grant from Fondazione del Monte di Bologna e Ravenna. Gloria Ravegnini was supported by a grant from Fondazione Umberto Veronesi – Post-doctoral fellowship 2014. The funding agencies had no role in the study design, data collection and analysis, or preparation of the manuscript.

8.

9.

10.

Conflict of interest The authors declare that there is no conflict of interests regarding the publication of this paper.

11.

REFERENCES 1. Berrington A, Darby SC, Weiss HA, Doll R. 100 years of observation on British radiologists: mortality from cancer and other causes 1897–1997. Br J Radiol 2001;74:507-19. doi: 10.1259/bjr.74.882.740507 2. Wang JX, Zhang LA, Li BX, Zhao YC, Wang ZQ, Zhang JY, Aoyama T. Cancer incidence and risk estimation among medical X-ray workers in China. Health Phys 2002;82:45566. PMID: 11906134 3. Cardis E, Vrijheid M, Blettner M, Gilbert E, Hakama M, Hill C, Howe G, Kaldor D, Muirhead CR, Schubauer-Berigan M, Yoshimura T, Bermann F, Cowper G, Fix J, Hacker C, Heinmiller B, Marshall M, Thierry-Chef I, Utterback D, Ahn YO, Amoros E, Ashmore P, Auvinen A, Bae JM, Solano JB, Biau A, Combalot E, Deboodt P, Diez Sacristan A, Eklof M, Engels H, Engholm G, Gulis G, Habib R, Holan K, Hyvonen H, Kerekes A, Kurtinaitis J, Malker H, Martuzzi M, Mastauskas A, Monnet A, Moser M, Pearce MS, Richardson DB, Rodriguez-Artalejo F, Rogel A, Tardy H, TelleLamberton M, Turai I, Usel M, Veress K. Risk of cancer after low doses of ionising radiation: retrospective cohort study in 15 countries. Br Med J 2005;331:77-80. doi: 10.1136/ bmj.38499.599861.E0 4. Metz-Flamant C, Samson E, Caër-Lorho S, Acker A, Laurier D. Leukemia risk associated with chronic external exposure to ionizing radiation in a French cohort of nuclear workers.

Radiat Res 2012;178:489-98. doi: 10.1667/RR2822.1 5. Kopjar N, Kašuba V, Milić M, Rozgaj R, Želježić D, Gajski G, Mladinić M, Garaj-Vrhovac V. Normalne i granične vrijednosti mikronukleus-testa na limfocitima periferne krvi u ispitanika opće populacije Republike Hrvatske [Normal and cut-off values of the cytokinesis-block micronucleus assay on peripheral blood lymphocytes in the Croatian general population, in Croatian]. Arh Hig Rada Toksikol 2010:61:219-34. doi: 10.2478/10004-1254-61-2010-2027 6. Fenech M. Cytokinesis-block micronucleus cytome assay.

Nat Protoc 2007;2:1084-104. doi: 10.1038/nprot.2007.77 7. Hu JJ, Smith TR, Miller MS, Mohrenweiser HW, Golden A, Case LD. Amino acid substitution variants of APE1 and

12.

13.

14.

15.

16.

17.

18.

19.

20.

XRCC1 genes associated with ionizing radiation sensitivity.

Carcinogenesis 2001;22:917-22. doi: 10.1093/ carcin/22.6.917 Aka P, Mateuca R, Buchet JP, Thierens H, Kirsch-Volders M. Are genetic polymorphisms in OGG1, XRCC1 and XRCC3 genes predictive for the DNA strand break repair phenotype and genotoxicity in workers exposed to low dose ionising radiations? Mutat Res 2004;556:169-81. PMID: 15491645 Angelini S, Kumar R, Carbone F, Maffei F, Cantelli-Forti G, Violante FS, Lodi V, Curti S, Hemminki K, Hrelia P. Micronuclei in humans induced by exposure to low level of ionizing radiation: influence of polymorphims in DNA repair genes. Mutat Res 2005;570:105-17. PMID: 15680408 Andreassi MG, Foffa I, Manfredi S, Botto N, Cioppa A, Picano E. Genetic polymorphisms in XRCC1, OGG1, APE1 and XRCC3 DNA repair genes, ionizing radiation exposure and chromosomal DNA damage in interventional cardiologists. Mutat Res 2009;666:57-63. doi: 10.1016/j. mrfmmm.2009.04.003 Kapka L, Baumgartner A, Siwińska E, Knudsen LE, Anderson D, Mielzyńska D Environmental lead exposure increases micronuclei in children. Mutagenesis 2007;22:2017. doi: 10.1093/mutage/gem004 Fenech M, Chang WP, Kirsch-Volders M, Holland N, Bonassi S, Zeiger E. HUMN project: detailed description of the scoring criteria for the cytokinesis-block micronucleus assay using isolated human lymphocyte cultures. Mutat Res 2003;534:65-75. PMID: 12504755 Eastmond DA, Tucker JD. Kinetochore localization in micronucleated cytokinesis-blocked Chinese hamster ovary cells: a new and rapid assay for identifying aneuploidyinducing agents. Mutat Res 1989;224:517-25. PMID: 2685592 Hu JJ, Mohrenweiser HW, Bell DA, Leadon SA, Miller MS. Symposium overview: genetic polymorphisms in DNA repair and cancer risk. Toxicol Appl Pharmacol 2002;185:64-73.

PMID: 12460738 Godderis L, Aka P, Mateuca R, Kirsch-Volders M, Lison D, Veulemans H. Dose dependent influence of genetic polymorphisms on DNA damage induced by styrene oxide, ethylene oxide and gamma-radiation. Toxicology 2006;219:220-9. doi: 10.1016/j.tox.2005.11.021 Stata Corporation. Stata Statistical software, release 11.

College Station, TX: stata Corporation; 2011.

Vral A, Fenech M, Thierens H. The micronucleus assay as a biological dosimeter of in vivo ionizing radiation exposure.

Mutagenesis 2011;26:11-7. doi: 10.1093/mutage/geq078 Fučić A, Markučič D, Mijić A, Jazbec AM. Estimation of genome damage after exposure to ionizing radiation and ultrasound used in industry. Environ Mol Mutagen 2000;36:47-51. PMID: 10918359 Cardoso RS, Takahashi-Hyodo S, Peitl P Jr, Ghilardi-Neto T, Sakamoto-Hojo ET. Evaluation of chromosomal aberrations, micronuclei, and sister chromatid exchanges in hospital workers chronically exposed to ionizing radiation.

Teratog Carcinog Mutagen 2001;21:431-9. PMID: 11746256 Maffei F, Angelini S, Cantelli-Forti G, Lodi V, Violante FS, Mattioli S, Hrelia P. Micronuclei frequency in hospital workers occupationally exposed to low levels of ionizing radiation: influence of smoking status and other factors.

Mutagenesis 2002;17:405-9. doi: 10.1093/mutage/17.5.405

Milić M, et al. Polymorphisms in DNA repair genes: CBMN cytome assay in workers chronically exposed to low doses of ionising radiation Arh Hig Rada Toksikol 2015;66:109-120

21. Hadjidekova VB, Bulanova M, Bonassi S, Neri M. Micronucleus frequency is increased in peripheral blood lymphocytes of nuclear power plant workers. Radiat Res 2003;160:684-90. PMID: 14640795 22. Zakeri F, Hirobe T. A cytogenetic approach to the effects of low levels of ionizing radiations on occupationally exposed individuals. Eur J Radiol 2010;73:191-5. doi: 10.1016/j. ejrad.2008.10.015 23. Sakly A, Gaspar JF, Kerkeni E, Silva S, Teixeira JP, Chaari N, Cheikh HB. Genotoxic damage in hospital workers exposed to ionizing radiation and metabolic gene polymorphisms. J Toxicol Environ Health A 2012;75:934-46. doi: 10.1080/15287394.2012.690710 24. Maluf SW, Passos DF, Bacelar A, Speit G, Erdtmann B. Assessment of DNA damage in lymphocytes of workers exposed to X-radiation using the micronucleus test and the comet assay. Environ Mol Mutagen 2001;38:311-5. doi: 10.1002/em.10029 25. Thierens H, Vral A, Barbé M, Meijlaers M, Baeyens A. Chromosomal radiosensitivity study of temporary nuclear workers and the support of the adaptive response induced by occupational exposure. Int J Rad Biol 2002;78:1117-26. doi: 10.1080/0955300021000034710 26. Sari-Minodier I, Orsiere T, Auquier P, Martin F, Botta A. Cytogenetic monitoring by use of the micronucleus assay among hospital workers exposed to low doses of ionising radiation. Mutat Res 2007;629:111-21. PMID: 17428723 27. Dutra A, Pak E, Wincovitch S, John K, Poirier MC, Olivero OA. Nuclear bud formation: a novel manifestation of Zidovudine genotoxicity. Cytogen Genome Res 2010;128:105-10. doi: 10.1159/000298794 28. Bonassi S, Fenech M, Lando C, Lin YP, Ceppi M, Chang WP, Holland N, Kirsch-Volders M, Zeiger E, Ban S, Barale R, Bigatti MP, Bolognesi C, Jia C, Di Giorgio M, Ferguson LR, Fucic A, Lima OG, Hrelia P, Krishnaja AP, Lee TK, Migliore L, Mikhalevich L, Mirkova E, Mosesso P, Müller WU, Odagiri Y, Scarffi MR, Szabova E, Vorobtsova I, Vral A, Zijno A. HUman MicroNucleus project: international database comparison for results with the cytokinesis-block micronucleus assay in human lymphocytes: I effects of laboratory protocol, scoring criteria, and host factors on the frequency of micronuclei. Environ Mol Mutagen 2001;37:3145. PMID: 11170240 29. Donmez-Altuntas H, Bitgen N. Evaluation of the genotoxicity and cytotoxicity in the general population in Turkey by use of the cytokinesis-block micronucleus cytome assay. Mutat Res 2012;748:1-7. doi: 10.1016/j.mrgentox.2012.05.013 30. Dias FL, Antunes LM, Rezende PA, Carvalho FE, Silva CM, Matheus JM, Oliveira JV Jr, Lopes GP, Pereira GA, Balarin MA. Cytogenetic analysis in lymphocytes from workers occupationally exposed to low levels of ionizing radiation.

31.

32.

33.

34.

35.

36.

37.

38.

39.

119

Environ Toxicol Pharmacol 2007;23:228-33. doi: 10.1016/j. etap.2006.10.003 Eken A, Aydin A, Erdem O, Akay C, Sanal HT, Soykut B, Sayal A, Somuncu I. Cytogenetic analysis of peripheral blood lymphocytes of hospital staff occupationally exposed to low doses of ionizing radiation. Toxicol Ind Health 2010;26:27380. doi: 10.1177/0748233710365693 Bonassi S, Neri M, Lando C, Ceppi M, Lin Y, Chang WP, Holland N, Kirsch-Volders M, Zeiger E, Fenech M; The HUMN collaborative group. Effect of smoking habits on the frequency of micronuclei in human lymphocytes: results from the Human MicroNucleus project. Mutat Res 2003;543:155-66. PMID: 12644185 Damiola F, Byrnes G, Moissonnier M, Pertesi M, Deltour I, Fillon A, Le Calvez-Kelm F, Tenet V, McKay-Chopin S, McKay JD, Malakhova I, Masyakin V, Cardis E, Lesueur F, Kesminiene A. Contribution of ATM and FOXE1 (TTF2) to risk of papillary thyroid carcinoma in Belarusian children exposed to radiation. Int J Cancer 2014;134:1659-68. doi: 10.1002/ijc.28483 Salnikova L, Chumachenko A, Belopolskaya O, Rubanovich A. Correlations between DNA polymorphism and frequencies of gamma-radiation induced and spontaneous cytogenetic damage. Health Phys 2012;103:37-41. doi: 10.1097/HP.0b013e3182231a9d Cho YH, Kim YJ, An YS, Woo HD, Choi SY, Kang CM, Chung HW. Micronucleus-centromere assay and DNA repair gene polymorphisms in lymphocytes of industrial radiographers. Mutat Res 2009;680:17-24. doi: 10.1016/j. mrgentox.2009.08.007 Rzeszowska-Wolny J, Polanska J, Pietrowska M, Palyvoda O, Jaworska J, Butkiewicz D, Hancock R. Influence of polymorphisms in DNa repair genes XPD, XRCC1 and MGMT on DNA damage induced by gamma radiation and its repair in lymphocytes in vitro. Radiat Res 2005;164:13240. PMID: 16038584 Mohrenweiser HW, Wilson DM, Jones IM. Challenges and complexities in estimating both the functional impact and the disease risk associated with the extensive genetic variation in human DNA repair genes. Mutat Res 2003;526:93-125. PMID: 12714187 Roos WP, Binder A, Böhm L. Determination of the initial DNA damage and residual DNA damage remaining after 12 hours of repair in eleven cell lines at low doses of irradiation.

Int J Radiat Biol 2000;76:1493-500. PMID: 11098852 Touil N, Aka PV, Buchet JP, Thierens H, Kirsch-Volders M. Assessment of genotoxic effects related to chronic low level exposure to ionizing radiation using biomarkers for DNA damage and repair. Mutagenesis 2002;17:223-32. PMID: 11971994

to low doses of ionising radiation 120 Milić M, et al. Polymorphisms in DNA repair genes: CBMN cytome assay in workers chronically exposed Arh Hig Rada Toksikol 2015;66:109-120

Polimorfizmi u genima za popravak DNA: poveznica s biomarkerima mikronukleus-testa u medicinskih radnika kronično izloženih niskim dozama ionizirajućeg zračenja Individualna osjetljivost na ionizirajuće zračenje rezultat je međudjelovanja samog izlaganja zračenju, oštećenja DNA nastalog prilikom tog izlaganja te samog popravka nastalog oštećenja. Veliki doprinos razlikama čine i polimorfizmi u genima za popravak DNA. U ovom radu istražili smo povezanost nastalih oštećenja DNA u obliku mikronukleusa (MN), jezgrinih pupova (NB) i nukleoplazmatskih mostova (NPB) s polimorfizmima jednog nukleotida (SNP) u genima za popravak DNK (APE1, hOGG1, XRCC1, XRCC3, XPD, PARP1, MGMT) koji sudjeluju u različitim mehanizmima popravka. Rezultati skupine od 77 medicinskih radnika kronično izloženih niskim dozama ionizirajućeg zračenja uspoređeni su s rezultatima skupine od 70 odgovarajućih kontrola. Izložena skupina imala je značajno veću učestalost MN-a (16,2±10,4 vs. 11.5±9.4; P=0,003), a sama pojavnost oštećenja bila je neovisna o medijatornoj varijabli (kovarijati). Značajno više učestalosti MN nađene su u izloženoj skupini u homozigotnih nositelja divljeg tipa gena hOGG1, XRCC1, PARP1 i MGMT i u heterozigotnih nositelja gena APEX1 i XPD (rs13181) u odnosu na kontrolnu skupinu istoga genotipa. Analiza genskih polimorfizama i citogenetička dozimetrija važna su dopuna osobnom dozimetrijskom nadzoru izloženih radnika.

KLJUČNE RIJEČI: genotipska analiza; jezgrini pupovi; mikronukleus; nukleoplazmatski mostovi; profesionalna izloženost

Grabovska S, Salyha Y. ADHD-like behaviour in the offspring of female rats exposed to low chlorpyrifos doses before pregnancy Arh Hig Rada Toksikol 2015;66:121-127

Original article

121

DOI: 10.1515/aiht-2015-66-2624

ADHD-like behaviour in the offspring of female rats exposed to low chlorpyrifos doses before pregnancy Sofia Grabovska and Yuriу Salyha Institute of Animal Biology NAAS, Lviv, Ukraine [Received in February 2015; CrossChecked in February 2015; Accepted in May 2015] The aim of this study was to investigate how chronic low-dose chlorpyrifos exposure of female Wistar rats before and during pregnancy affects behavioural parameters in their offspring. Four months before pregnancy, we exposed three groups of rats to chlorpyrifos doses of 5, 10, and 15 mg kg-1 body weight every day for 30 days, whereas one group received a single 30 mg kg-1 dose on gestational day 6. When the offspring of the exposed rats grew up, we studied their anxiety rate, motor activity, and cognitive abilities using the respective behavioural tests: open field test, dark/light box, and the extrapolation escape test. The offspring of rats exposed before pregnancy had significantly higher activity rate than controls, and even showed motor agitation and hyperactivity signs. The offspring of rats exposed to the single dose had difficulties solving the extrapolation escape test and showed poorer short- and long-term memory performance. This confirmed that even pre-pregnancy chlorpyrifos exposure can cause neurobehavioral consequences in offspring. Even though the mechanisms of the observed changes remain unclear and need further investigation, these data seem alarming and may serve as an important argument for revising the terms of safe pesticide use.

KEY WORDS: neurotoxicity; organophosphate pesticides; anxiety; memory; motor activity Due to its neurotoxic properties chlorpyrifos (CPF) is one of the most common and most dangerous organophosphate pesticides used in agriculture, industry, and even at homes (1). It inhibits cholinesterase enzymes, causing acetylcholine overload in the cholinergic brain regions, over-excitation, and then disruption of synaptic transmission (2).

Another toxic effect it causes is oxidative stress (2).

This effect can induce the death of isolated hippocampal neurons in vitro (3). One more recently discovered neurotoxic effect of CPF is that it affects the endocannabynoid system in the brain (4); even at low doses and without detectable AChE inhibition, CPF can lead to the accumulation of anandamide and a decrease in amidase activity. Moreover, CPF has been reported to affect the immune system; in a study by Nakamura et al. (5) perinatal exposure to CPF led to irreversible changes in T-lymphocytic response in mice. However, despite intense research, the mechanisms of CPF toxicity remain unclear.

Acute poisoning with CPF can lead to severe nerve damage, hypoxia, blood pressure decline, seizures, and even death (1). Behavioural manifestations of acute CPF poisoning observed in animal studies include a decrease in anxiety-related behaviour, motor hyperactivity or, conversely, reduced activity (6).

Correspondence to: Sofia Grabovska, PhD student, Institute of Animal Biology NAAS, Lviv, Stusa str., 38, 79034, Ukraine; E-mail: sofia_v@i.ua

Chronic exposure to low doses of CPF may not cause any acute symptoms, but the subtle changes in CNS functioning induced by small amounts of the pesticide may lead to neurological damage. Subchronic exposure studies suggest a number of behavioural effects. MiddlemoreRisher et al. (7) reported increased impulsivity and attention problems in rats. We too studied how subchronic low-dose CPF exposure affected rat behaviour (8). The exposed rats showed a decrease in long- and short-term spatial memory, and the anxiety rate increased rapidly on day 1 of exposure, compared to controls, but it dropped to control values within two weeks, showing adaptation to the toxicant.

In humans, chronic exposure to CPF may have severe neurological consequences, including Parkinson’s disease (9). Recently, Rauh et al. (10) published alarming information that exposure to CPF in women (such as those working with pesticides on farms) may impair cognitive and behavioural development in their children. Bouchard et al. (11) found an IQ decline in children prenatally exposed to CPF, whereas other researchers associated it with the attention deficit hyperactivity disorder (ADHD) syndrome (12) and autism (13).

However, we found no information on the risks of CPF exposure before pregnancy for the neurological development of children, and it is still unclear whether CPF affects the developing foetus directly or through changes in the mother’s body. We therefore designed an animal study whose aim was to compare behavioural and cognitive effects of CPF in the offspring of female rats receiving CPF

122

Grabovska S, Salyha Y. ADHD-like behaviour in the offspring of female rats exposed to low chlorpyrifos doses before pregnancy Arh Hig Rada Toksikol 2015;66:121-127

subchronically before pregnancy and in the offspring of female rats that received a single CPF dose during pregnancy.

MATERIALS AND METHODS Chlorpyrifos preparation Chlorpyrifos (CAS No. 2921-88-2) (O,O-diethyl-O3,5,6-trichloro-2-pyridyl phosphorothioate, purity 99.9 %) was purchased from Sigma Chemical Co. (St. Louis, MO, USA). As CPF is poorly soluble in water, we dissolved it in refined sunflower oil at concentrations corresponding to doses given to each animal according to its body weight (b.w.) and group using oral probe.

Experimental design Figure 1 summarises the design of our experiment.

Three groups of adult female Wistar rats (three in each group) were receiving CPF at the doses of 5 (group 1), 10 (group 2), or 15 mg kg-1 b.w. (group 3) every day for 30 days. As the LD50 range for rats varies between 90 and 270 mg kg-1 (1), our dose selection was based on literature data from similar research and on our previous work. The highest chronic dose of 15 mg kg-1 is the threshold for the visible effects of toxicity (such as porphyrine marks around the eyes). These rats were then kept under standard vivarium conditions without further exposure to CPF or other adverse factors for four months before pregnancy. Group 4 (three animals of the same age and weight) received a single dose of 30 mg kg-1 CPF on gestational day 6. This dose has been used in a number of studies as a sub-toxic acute dose that does not cause immediate symptoms of poisoning but can lead to some adverse effects (14, 15). Control females (group 5) received pure oil.

Each rat was weighed and numbered, and all were housed in standard conditions with the 12-hour light/dark cycle and free access to standard feed and water.

After weaning on postnatal day 21, rat pups were divided by experimental groups; rats of each group were housed in separate cages and numbered. When they reached maturity at two months of age, the offspring rats were put to the following behavioural tests: open field, dark/light box, and extrapolation escape (see Figure 2). The open field and extrapolation escape tests were repeated after 10 days.

The dark/light box test was performed once, as its results clearly confirmed the ones of the open field test, and it seemed unnecessary to repeat it.

To avoid artefacts, the animals were kept in silence under dim light for two hours before the tests. No feeding, grouping, or other manipulations were performed at that time. All testing was carried out at the same time of the day, under the same lighting and temperature conditions, without disturbing odours and noise.

All procedures followed the European Convention for the Protection of Vertebrate Animals used for Experimental and Other Scientific Purposes (16) and General Ethical Principles of Experiments using Animals (17).

Open field test The open field test was introduced by Hall and Ballachey (18) in 1932 and is commonly used to assess animal emotionality, exploratory activity, and anxiety rate.

It is based on the natural exploration drive of animals, rodents in particular, when placed in a new environment.

We used the test apparatus shown in Figure 2. One rat at a time was placed in the test arena for 3 min, and its behaviour was video-recorded. We looked for inner and outer horizontal activity, vertical activity (free and wallsupported), long and short grooming, number and total time of freezing reactions, number of defecations, and number of times the rats sniffed the holes. After each test, the arena

Group

CPF dose (mg kg-1 b.w.)

Time and duration of treatment

Number of pregnancies

Number of offspring

1

5

every day for 30 days, 4 months before pregnancy

1

14

2

10

every day for 30 days, 4 months before pregnancy

1

12

3

15

every day for 30 days, 4 months before pregnancy

1

10

4

30

a single dose on gestational day 6

2

21 (12+9)

Control

0

none

1

13

Figure 1 Scheme of the experimental design and animal groups

Grabovska S, Salyha Y. ADHD-like behaviour in the offspring of female rats exposed to low chlorpyrifos doses before pregnancy Arh Hig Rada Toksikol 2015;66:121-127

123

dark and a light chamber by a partition with a hole in it.

Each animal was put into the light chamber for 3 min and its behaviour observed. As rodents prefer dark and enclosed places, they tend to hide in the dark chamber, but also like to peek out and sometimes return to the light chamber to investigate it. We measured the time spent in the light chamber before entering the dark one, the number of times the animal peeked out from the hole, the number of times the animal returned to the light chamber, and total time spent there.

Extrapolation escape test The extrapolation escape test serves to assess the cognitive function in rodents (20). The animals were placed under acute stress and had to find a way to escape and then to recall the successful strategy when they faced the same stress again 30 minutes later. The test apparatus consisted of a Plexiglas water container and a Plexiglas cylinder in its centre (Figure 2). The lower edge of the cylinder was immersed 2.5 cm below the water surface. The rats were put in the cylinder, and the only way to escape was to dive below the lower edge of the cylinder. If a rat failed to do so, we took it out after 2 min. We calculated the ratio of successful and failed escape attempts and measured the time before diving and the duration of initial immobility.

Also, we calculated the “learning effectiveness” parameter that is the difference between the time before diving in first and second trials of each animal in each test session.

Statistical analysis The obtained data were analysed with multifactorial ANOVA and ANOVA for repeated measures using Statistica 12 software (StatSoft Inc., Tulsa, OK, USA). The value of p<0.05 was considered statistically significant. ANOVA has been used in many studies with behavioural tests similar to our own (21-23).

RESULTS AND DISCUSSION

Figure 2 Test apparatuses was cleaned up to avoid the influence of other animals’ odour marks.

Dark/light box test The dark/light box test is another behavioural test to assess animal emotionality, exploratory activity, and anxiety rate, especially in laboratory rodents. We used the test apparatus as described by Bourin and Hascoe (19) (for details see Figure 2), which was a plastic box divided to a

The results obtained by the open field and dark/light box tests had similar tendencies. In the open field test group 3 showed significantly higher horizontal and vertical activity and the number of hole sniffing actions than control and other experimental groups (Table 1).

In the repeated test, group 3 also showed significantly lower anxiety rate than controls (long and short grooming, defecation, and freezing). Forty to sixty crossed squares (vs 10-20 in controls), no freezing, no defecation, and no short grooming at all constitute abnormal behaviour. In fact, these results suggest hyperactivity disorder, and confirm that exposure of mothers to OP pesticides may lead to ADHD in children (24).

Table 2 shows the results of the dark/light box test.

Group 3 significantly differed from control in the number of peeks out of the hole, once more demonstrating a decrease

22.83±4.37

14.88±4.51

2

4

18.38±5.14

1

42.50±6.50**

16.13±3.27

Control

3

29.75±3.29

4

21.63±3.57

2

49.00±4.69**

28.50±2.73

1

3

Outer horizontal activity

Group

19.48±5.18 35.6±24.15 13.70±3.57 34.13±13.89

1

2

3

4

Control 45.70±19.75 * significantly different from control (p≤0.05)

Hiding time (s)

5.63±1.89

9.08±2.25

12.75±2.06**

4.50±1.38

5.25±1.51

10.00±2.21

10.00±1.33

14.75±0.85**

7.75±1.88

10.50±0.82

Vertical activity

Group

Table 2 Dark/light box test results

0.25±0.16

2.67±1.28

1.00±0.71

1.13±0.88

0.38±0.26

1.63±0.80

1.08±0.23

4.00±1.35

0.75±0.49

1.50±0.68

Inner horizontal activity

Control 11.75±4.46 *significantly different from control (p≤0.05) **significantly different from control (p≤0.01)

2

1

Test session

Table 1 Open field test results

1.75±0.67

5.00±0.77*

6.25±0.63*

4.14±1.24

5.63±0.56*

Peeks out

0.63±0.26

0.42±0.14

0.25±0.25*

0.50±0.19

0.25±0.16

1.00±0.46

0.25±0.13

0

0.25±0.16

0.38±0.26

Long grooming

0.63±0.26

0.67±0.26

0*

0*

0.25±0.16

1.38±0.38

0.75±0.25

0.50±0.29

0.25±0.16

0.50±0.19

Short grooming

0

21.13±9.23

10.63±6.13

0.97±0.97

11.38±6.72

Returns time (s)

1.88±0.77

1.08±0.57

0*

1.25±0.45

0.50±0.50

2.75±0.59

2.25±0.75

0.50±0.50

2.25±0.84

0*

Defecation

66.13±19.45

27.17±15.04

0.14±0.14

0.63±0.32

0

0.83±0.30

0.5±0.29

0**

60.38±22.14

48.38±15.20

2.63±2.63

5.83±5.83

0

30.25±18.50

0

Freezing time (s)

No. of returns

1.00±0.38

1.83±0.39

2.50±1.04**

0.88±0.64

1.38±0.63

0.25±0.16

1.33±0.28

2.25±0.75**

0.75±0.25

1.13±0.27

Hole sniffing

1.13±0.30

0.42±0.19

0**

0.88±0.35

0.88±0.30

0.13±0.13

0.16±0.16

0

0.38±0.18

0

Freezing quantity

124 Grabovska S, Salyha Y. ADHD-like behaviour in the offspring of female rats exposed to low chlorpyrifos doses before pregnancy Arh Hig Rada Toksikol 2015;66:121-127

Trial 1

38.65±17.89

23.66±9.11

66.67±21.40

Control 1.99±1.32 *significantly different from control (p≤0.05) **significantly different from control (p≤0.01)

1.76±1.31

2

40.46±17.23

38.65±17.89

69.45±13.98

2.95±1.53

1

2.64±1.22

1.99±1.32

Control

67.98±13.19

4

1.70±1.15

4

75.64±15.12

45.96±13.86

61.76±14.50

Diving time (s)

0

0

0.64±0.64

2

3

4.20±2.30

Immobility time (s)

1

Group

3

2

1

Test session

Table 3 Extrapolation escape test results

6.64±2.12

5.84±2.25

3.62±2.09

25.37±15.87

5.04±2.05

4.19±2.21

14.97±9,74

3.07±1.62

5.59±2.00

4.07±1.42

Immobility time (s)

Trial 2

50.95±17.39

75±27±15.97

5.91±1.53

76.23±20.85

23.20±13.90

49.65±16.01

78.99±14.67

36.79±14.05

80.20±19.25

30.35±11.89

Diving time (s)

0.75±0.16

0.41±0.15*

1.00±0

0.43±0.20*

0.88±0.13

0.75±0.16

0.67±0.14

0.75±0.25

0.86±0.14

0.75±0.16

Successful attempts ratio

-1.29±15.11

3.72±1.48

5.28±2.94

3.97±2.84

5.27±1.50

7.61±23.77

-1.47±13.35

51.98±19.85

-20.70±25.07

21.30±7.83

Learning effectiveness

Grabovska S, Salyha Y. ADHD-like behaviour in the offspring of female rats exposed to low chlorpyrifos doses before pregnancy Arh Hig Rada Toksikol 2015;66:121-127

125

126

Grabovska S, Salyha Y. ADHD-like behaviour in the offspring of female rats exposed to low chlorpyrifos doses before pregnancy Arh Hig Rada Toksikol 2015;66:121-127

in anxiety-like behaviour and increased motor activity.

Other differences between the groups were not statistically significant.

In contrast to the open field, the differences from control in the extrapolation escape test were less prominent (Table 3). Groups 2 and 4 showed significantly fewer successful attempts than controls, and those who were successful spent longer time in the cylinder before diving, which points to poorer cognitive performance. In contrast, Group 3 showed significantly better diving time in the repeated test than all other groups, including control, and all escape attempts were successful. Although the learning effectiveness in group 3 may seem higher than in other groups, the difference was not significant due to statistical error caused by a wide variation of data within the group. It is also worth noting that in the first test, group 3 did not show any immobility at all, which points to motor hyperactivity.

Our findings suggest that chronic CPF exposure of mothers before pregnancy could cause hyperactivity and reduce anxiety in their offspring, but does not seem to affect the cognitive function, which points to symptoms similar to the ADHD syndrome, as reported elsewhere (12). Acute prenatal exposure, however, seems to affect the cognitive function.

The limitation of our study is the number of animals we used. Due to high mortality of pups (particularly in group 3, as only four animals reached maturity and were tested), data within the groups varied strongly and caused high statistical error, which rendered some of the observed differences (mostly in the extrapolation escape test) nonsignificant. This is why we plan to use more animals in future research. Moreover, future research should benefit from including mice and other species as well as from including other behavioural tests, such as T-maze, Vogel conflict test, and beam-walking.

Our study has produced intriguing data on the adverse effects of maternal exposure to CPF even before fertilisation.

It opens a number of questions about the underlying mechanisms and tissue accumulation of CPF (being a nonpersistent pesticide), as it remains unclear how the pesticide caused the observed changes in young rats that have not been exposed to it either pre- or postnatally. Future research should address these questions, specifically ADHD and other behavioural abnormality risks in children exposed to chronic low doses of pesticides through their mothers and the mechanisms that increase these risks.

REFERENCES 1. The Dow Chemical Company. Chlorpyrifos and Responsible Use [displayed 5 May 2015]. Available at http://www. chlorpyrifos.com/benefits-and-use/use/responsible-use.htm 2. Salyha Y. Biological effects assessment of chlorpyrifos and some aspects of its neurotoxicity. Visnyk of Lviv University.

Biology series 2010;54:3-14.

3. Salyha Y. Chlorpyrifos leads to oxidative stress-induced death of hippocampal cells in vitro. Neurophysiology 2013;45:1939. doi: 10.1007/s11062-013-9356-7 4. Carr RL, Graves CA, Mangum LC, Nail CA, Ross MK. Low level chlorpyrifos exposure increases anandamide accumulation in juvenile rat brain in the absence of brain cholinesterase inhibition. Neurotoxicology 2014;43:82-9. doi: 10.1016/j.neuro.2013.12.0094 5. Nakamura R, Kimura Y, Matsuoka H, Hachisuka A, Nakamura R, Nakamura A, Shibutani M, Teshima R. [Effects of transplacental and trans-breast milk exposure to the organophosphate compound chlorpyrifos on the developing immune system of mice, in Japanese]. Kokuritsu Iyakuhin Shokuhin Eisei Kenkyusho Hokoku 2011;129:105-10.

PMID: 222598505 6. Chen WQ, Yuan L, Xue R, Li YF, Su RB, Zhang YZ, Li J. Repeated exposure to chlorpyrifos alters the performance of adolescent male rats in animal models of depression and anxiety. Neurotoxicology 2011;32:355-61. doi: 10.1016/j. neuro.2011.03.0086 7. Middlemore-Risher ML, Buccafusco JJ, Terry AV, Jr. Repeated exposures to low-level chlorpyrifos results in impairments in sustained attention and increased impulsivity in rats. Neurotoxicol Teratol 2010;32:415-24. doi: 10.1016/j. ntt.2010.03.008 8. Rosalovsky V, Salyha Y. New biochemical and physiological aspects of chlorpyrifos neurotoxicity. Toxicol Lett 2013;221(Supplement):S200. doi: 10.1016/j.toxlet.2013.

05.468 9. Brown TP, Rumsby PC, Capleton AC, Rushton L, Levy LS. Рesticides and parkinson’s disease – is there a link? Environ Health Perspect 2006;114:156-64. doi: 10.1289/ehp.8095 10. Rauh V, Arunajadai S, Horton M, Perera F, Hoepner L, Barr DB, Whyatt R. Seven-year neurodevelopmental scores and prenatal exposure to chlorpyrifos, a common agricultural pesticide. Environ Health Perspect 2011;119:1196-201. doi: 10.1289/ehp.1003160 11. Bouchard MF, Chevrier J, Harley KG, Kogut K, Vedar M, Calderon N, Trujillo C, Johnson C, Bradman A, Barr DB, Eskenazi B. Prenatal exposure to organophosphate pesticides and IQ in 7-year-old children. Environ Health Perspect 2011;119:1189-95. doi: 10.1289/ehp.1003185 12. Rauh VA, Garfinkel R, Perera FP, Andrews HF, Hoepner L, Barr DB, Whitehead R, Tang D, Whyatt RW. Impact of prenatal chlorpyrifos exposure on neurodevelopment in the first 3 years of life among inner-city children. Pediatrics 2006;118:1845-59. PMID: 17116700 13. Terry AV Jr, Beck WD, Warner S, Vandenhuerk L, Callahan PM. Chronic impairments in spatial learning and memory in rats previously exposed to chlorpyrifos or diisopropyl fluorophosphates. Neurotoxicol Teratol 2012;34:1-8. doi: 10.1016/j.ntt.2011.08.015 14. Yang YL, Gordon CJ. Possible role of vasopressin in the thermoregulatory response to chlorpyrifos in the rat.

Pharmacol Toxicol. 2002;90(6):311-6. PMID: 12403052 15. Quistad GB, Nomura DK, Sparks SE, Segall Y, Casida JE. Cannabinoid CB1 receptor as a target for chlorpyrifos oxon and other organophosphorus pesticides. Toxicol Lett 2002;135:89-93. doi: 10.1016/S0378-4274(02)00251-5 16. European Convention for the Protection of Vertebrate Animals used for Experimental and Other Scientific PurposesStrasbourg, 1986.

Grabovska S, Salyha Y. ADHD-like behaviour in the offspring of female rats exposed to low chlorpyrifos doses before pregnancy Arh Hig Rada Toksikol 2015;66:121-127

17. General Ethical Principles of Experiments using Animals.

First National Congress of Bioethics, Kyiv, 2001 18. Hall CS, Ballachey EL. A study of the rat’s behavior in a field: a contribution to method in comparative psychology.

Univ Calif Publ Psychol 1932;6:1-12.

19. Bourin M., Hascoe M. The mouse light/dark box test. Eur J Pharmacol 2003;463:55-65. doi: 10.1016/S00142999(03)01274-3 20. Extrapolation Escape Task. OpenScience, Moscow, Russia. [displayed 13 February 2015]. Available at http://www. openscience.ru/index.php?page=ts&item=004 21. Cardinal RN, Aitken MRF. ANOVA for the Behavioral Sciences Researcher. Mahwah (NJ): Lawrence Erlbaum Associates; 2006.

127

22. Mariano MO, Esteves AM, Frank MK, Caperuto LC, Manconi M, Tufik S, De Mello MT. Changes in motor behavior during pregnancy in rats: the basis for a possible animal model of restless legs syndrome. Rev Bras Ginecol Obstet 2014;36:436-41. PMID: 25317821 23. Kulesskaya N, Voikar V. Assessment of mouse anxiety-like behavior in the light-dark box and open-field arena: role of equipment and procedure. Physiol Behav 2014;133:30-8. doi: 10.1016/j.physbeh.2014.05.006 24. Yolton K, Cornelius M, Ornoy A, McGough J, Makris S, Schantz S. Exposure to neurotoxicants and the development of attention deficit hyperactivity disorder and its related behaviors in childhood. Neurotoxicol Teratol 2014;44:30-45. doi: 10.1016/j.ntt.2014.05.00322

Ponašanje nalik ADHD-u u potomaka ženki štakora izloženih niskim dozama klorpirifosa prije trudnoće Cilj je ovog istraživanja bio ispitati kako izloženost ženki Wistar štakora niskim, kroničnim dozama klorpirifosa prije i tijekom trudnoće utječe na parametre ponašanja njihovih potomaka. Četiri mjeseca prije trudnoće tri su skupine štakorica 30 dana primale klorpirifos u dnevnim dozama od 5, 10 i 15 mg kg-1 tjelesne mase, a jedna je skupina primila jednokratnu dozu od 30 mg kg-1 šestog dana gestacije. Kad je mladunčad odrasla, bihevioralnim testovima otvorenog polja, testom tamne/svijetle komore i ekstrapolacijskim testom bijega izmjerili smo njihovu razinu tjeskobe, motoričke aktivnosti i kognitivne sposobnosti. Mladunci ženki izloženih prije trudnoće iskazali su značajno više razine aktivnosti od kontrolne skupine, napose motoričku agitaciju i znakove hiperaktivnosti. Mladunci ženki izloženih jednokratnoj dozi imali su poteškoća u rješavanju ekstrapolacijskoga testa bijega te su iskazali slabije kratkoročno i dugoročno pamćenje. Naši su rezultati pokazali da izloženost klorpirifosu prije trudnoće može uzrokovati neurobihevioralne poremećaje u mladunčadi.

Premda istraživanjem nismo uspjeli utvrditi mehanizme uočenih promjena, ova su saznanja uznemirujuća i mogu poslužiti kao snažan argument za ponovno promišljanje o ograničenjima u primjeni pesticida.

KLJUČNE RIJEČI: motorička aktivnost; neurotoksičnost; organofosforni pesticidi; pamćenje; tjeskoba

Žunec S, et al. Comparative determination of the efficacy of bispyridinium oximes in paraoxon poisoning Arh Hig Rada Toksikol 2015;66:129-134

Original article

129

DOI: 10.1515/aiht-2015-66-2623

Comparative determination of the efficacy of bispyridinium oximes in paraoxon poisoning Suzana Žunec1, Božica Radić1, Kamil Kuča2,3, Kamil Musilek2,3, and Ana Lucić Vrdoljak1 Institute for Medical Research and Occupational Health, Zagreb, Croatia1, Biomedical Research Center, University Hospital Hradec Kralove2, University of Hradec Kralove, Faculty of Science, Department of Chemistry3, Hradec Kralove, Czech Republic [Received in February 2015; CrossChecked in February 2015; Accepted in May 2015] The inability of standard therapy to provide adequate protection against poisoning by organophosphorus compounds (pesticides and nerve agents) motivated us to search for new, more effective oximes. We investigated the pharmacotoxicological properties of six experimental K-oximes (K027, K033, K048, K074, K075, and K203) in vivo.

The therapeutic efficacy of K-oximes (at doses of 5 or 25 % of their LD50) combined with atropine was assessed in paraoxon-poisoned mice and compared with conventionally used oximes HI-6 and TMB-4. The bisoxime K074 was the most toxic (LD50=21.4 mg kg-1) to mice, while monoxime K027 was the least toxic (LD50=672.8 mg kg-1). With the exception of K033, all of the tested K-oximes showed better therapeutic efficiency than HI-6 and TMB-4. K027 and K048 stood out by demonstrating low acute toxicities and ensuring protective indices ranging from 60.0 to 100.0 LD50 of paraoxon. Taking into account that these two oximes showed a similar therapeutic efficacy regardless of the applied doses, our results suggest that K027 and K048 could be antidotes for paraoxon intoxication.

KEY WORDS: acute toxicity; mice; therapeutic efficacy Organophosphorus (OP) compounds are a heterogeneous group of organic compounds that represent a serious toxicological problem and therapeutic challenge (1). The introduction of synthetic OP compounds half a century ago has resulted in the repeated misuse of nerve agents during military conflicts and terrorist attacks as well as 3 million cases of poisoning by pesticides, all leading to nearly 260,000 human fatalities per year (2-4). The main toxic mechanism of OP compounds involves the inhibition of esterase enzymes; acetylcholinesterase (AChE) in synapses and red blood cell membranes, and butyrylcholinesterase (BChE) in plasma (5). Although acute BChE inhibition does not seem to cause clinical manifestations, AChE inhibition results in the accumulation of the neurotransmitter acetylcholine (ACh) at cholinergic synapses, with an overstimulation of cholinergic receptors of the muscarinic and nicotinic type (5). As these receptors are localised in most organs, a “cholinergic syndrome” may ensue increased sweating and salivation, profound bronchial secretion, bronchoconstriction, miosis, increased gastrointestinal motility, diarrhoea, tremors, muscular twitching, and various negative effects on the central nervous system.

Death is believed to be due to respiratory failure (5).

Standard post-exposure antidotal treatment of OP poisoning includes administrating an anticholinergic drug Correspondence to: Suzana Žunec, PhD, Toxicology Unit, Institute for Medical Research and Occupational Health, Ksaverska c. 2, HR-10001, Zagreb, Croatia, Tel: +385 1 4682 641, Fax: +385 1 4673 303, E-mail: suzana@imi.hr

such as atropine and AChE reactivators called oximes in accordance with the functional oxime group. Atropine decreases the effects of excess ACh primarily by blocking peripheral muscarinic receptor sites, resulting in reduced secretion and reversing the constriction of smooth muscles.

Since it has little effect on nicotinic sites, skeletal muscle fasciculation continues, resulting in the paralysis of respiratory muscles, i.e., peripheral respiratory failure (6).

Oximes break the OP-AChE bond and restore the activity of inhibited AChE (7). Thus far, only four oximes of a pyridinium aldoxime structure (2-PAM, TMB-4, HI-6, and obidoxime) have found clinical application, but none is sufficiently effective against all of the known OP compounds (7). Due to this fact, laboratories worldwide are searching for a broad spectrum AChE reactivator.

K-oximes appear to be among the most promising compounds developed (8-10).

Therefore, the present study was undertaken to assess and compare the therapeutic efficacy of six experimental K-oximes (K027, K033, K048, K074, K075, and K203) combined with atropine in OP pesticide paraoxon poisoned mice.

MATERIALS AND METHODS Chemicals K-series oximes K027, K033, K048, K074, K075 and K203 were prepared as described earlier (11-15). TMB-4

130

Žunec S, et al. Comparative determination of the efficacy of bispyridinium oximes in paraoxon poisoning Arh Hig Rada Toksikol 2015;66:129-134

[1,3-bis(4-hydroxyiminomethylpyridinium) propane dichloride] was synthesised in Bosnalijek, Sarajevo, Bosnia and Herzegovina, while HI-6 [(1-(2- hydroxyiminomethylpyridinium)-3-(4-carbamoylpyridinium)-2-oxapropane dichloride)] was synthesised at the University of Defence, Hradec Kralove, Czech Republic (16). Oximes were kept at room temperature and dissolved in distilled water or atropine immediately before use. Paraoxon (diethyl p-nitrophenyl phosphate) was purchased from SigmaAldrich, Steinheim, Germany. The stock solution of 50 mg mL-1 of paraoxon was prepared in isopropanol.

Further dilutions were made in saline, shortly before use.

Atropine sulphate was purchased from Kemika, Zagreb, Croatia. The solution of 5 mg mL-1 of atropine was prepared in distilled water.

Animals Male NIH/Ola Hsd mice were purchased from the Institute of Immunology, Inc., Department of Experimental Animals and Antisera, Zagreb, Croatia. The mice were kept in Macrolone cages at 21 °C maintained by a thermostat with exchanging light and dark cycles every 12 h. The animals were fed a standard diet (4RF21, Mucedola, Milano, Italy) with free access to water. Selection was made by body weight (18-25 g) following random distribution into groups of four animals. This study was performed with the approval of the Ethics Committee of the Institute for Medical Research and Occupational Health in Zagreb, Croatia.

Acute toxicity Acute toxicity (LD50) was based upon 24 h-mortality rates calculated according to Thompson and Weil (17-18).

Each LD50 was evaluated from the results obtained with four to six doses of a given compound and four animals were injected per dose. Whenever the results of the experiment allowed, the 95 % confidence limits were estimated from tables described elsewhere (17-18).

Therapeutic efficacy The therapeutic effect against paraoxon poisoning was tested by administering the studied oximes (5 or 25 % of their LD50) together with atropine sulphate (10 mg kg-1), immediately after paraoxon. The OP compound was given subcutaneously (s.c.) while therapy was administered intraperitoneally (i.p.). Mice were observed for 24 h and the antidotal efficacy of the tested oximes was expressed as protective index (PI) and maximal dose of poison (MDP).

PI was the ratio of LD50 between OP with antidote and OP given alone. MDP was the highest multiple of the LD50 of OP, which was fully counteracted (survival of all animals) by the antidote.

RESULTS AND DISCUSSION The inadequacy of standard therapy to provide protection against OP poisoning is a matter of continuous public and scientific concern. Within the last few years, a new generation of oximes has been developed in the Czech Republic (8). Structurally, they are bispyridinium oximes that differ in the length of the connection chain between two pyridinium rings, the position of the oxime group, and the number of oxime groups in a molecule. Promising results using several of the previously mentioned oximes were obtained for poisoning by the nerve agent tabun, as well as in the case of pesticide poisoning (8, 14, 19-29).

Although the use of OP pesticides is under restriction in most parts of the world, especially in developed countries, some, such as parathion, are still widely (mis)used (30).

The active metabolite of the insecticide parathion, paraoxon, is one of the most potent acetylcholinesterase-inhibiting compounds available (30). Therefore, it is important to determine the antidotal potency of new oximes in paraoxon poisoning. Previous studies have shown good potency of several K-oximes in vitro to reactivate AChE inhibited by paraoxon (22-23, 25, 27, 29). In the present study, we used mice as an experimental model to broaden our knowledge on the pharmacotoxicological properties of six K-oximes (K027, K033, K048, K074, K075, and K203). The conventional oximes HI-6 and TMB-4 were included for comparison.

The acute toxicity (LD50; i.p.) of the tested oximes is shown in Table 1. Bisoximes K033, K074, and K075 were more toxic than monoximes K027 and K048, which is not surprising considering that oximes with two oxime groups have a higher affinity for native AChE (31). For example, bisoxime TMB-4 is the most toxic of the four mentioned conventional oximes (7) and the novel oximes K027 and K048 had nearly an eight and three-times lower acute toxicity, respectively, than TMB-4. Moreover, with an LD50 of 672.8 mg kg-1 body weight, K027 was the least toxic.

Our result agrees with in vitro studies performed by Petroianu and Lorke (26) and Lorke et al. (32), which singled out K027 as the least toxic in comparison with the other oximes tested. An exception was noticed in the case of monoxime K203. Although K203 has one oxime and one carbamoyl moiety just like K027 and K048, its different stereoelectronic profile is probably why its LD50 was so low (33).

Tables 2 and 3 show the therapeutic effects of the tested oximes combined with atropine on paraoxon toxicity in male mice. Exposure to paraoxon led to classical signs of cholinergic toxicity, but severe signs of toxicity were also observed in all of the animals despite antidotal treatment.

Muscle fasciculation and tremor generally occurred within 1-2 min after poisoning. Convulsions appeared with a latency of 3-4 min. During the acute phase, all of the animals exhibited dyspnoea and cyanosis. The animals that survived remained active for 24 h.

131

Žunec S, et al. Comparative determination of the efficacy of bispyridinium oximes in paraoxon poisoning Arh Hig Rada Toksikol 2015;66:129-134

Table 1 Chemical name, structure, and LD50 of the tested oximes in male mice Compound

1-(4-carbamoylpyridinium)-3-(4hydroxyiminomethylpyridinium)propane dibromide

bis-1,3-(2hydroxyiminomethylpyridinium)butane dibromide

Code

O

K027

HO

N N

N

K033

+

N

+

NH 2

+

OH

HO

672.8a (599.0-755.3)

N+

33.4a (29.7-37.5)

N+

224.9a (154.2-328.0)

N

HO

1-(4-carbamoylpyridinium)-4-(4hydroxyiminomethylpyridinium)butane dibromide

LD50 (mg kg-1) (95Â % confidence limits)

Structure

N

N N+

K048

NH 2 O

1,4-bis(4hydroxyiminomethylpyridinium)butane dibromide

(E)-1,4-bis(4hydroxyiminomethylpyridinium) -but2-ene dibromide

HO

N+

K074

HO

N

N+

K075

N

N+

K203

OH

37.5 (31.4-44.7)

+

N

N

21.4b (19.0-24.0)

N+ N

HO

(E)-1-(4-carbamoylpyridinium)-4-(4hydroxyiminomethylpyridinium)-but2-ene dibromide

N

N

OH

+

NH 2

89.1c (75.7-104.9)

O O

1-(2-hydroxyiminomethylpyridinium)3-(4-carbamoylpyridinium)-2oxapropane dichloride

1,3-bis(4hydroxyiminomethylpyridinium) propane dichloride From ref. 19 From ref. 20 c From ref. 21 a b

N+

HI-6

O

NH 2

N+

635.2 (555.3-725.2)

N OH

TMB-4

HO

N

N N+

N+

OH

89.8 (-)

132

Žunec S, et al. Comparative determination of the efficacy of bispyridinium oximes in paraoxon poisoning Arh Hig Rada Toksikol 2015;66:129-134

Table 2 Therapeutic effect of the tested oximes (5 % of their LD50) combined with atropine upon (s.c.) paraoxon toxicity* in male mice LD50 (μg kg-1)

95 % confidence limits (μg kg-1)

PI

MDP

atropine

4191.8

3563.3-4931.2

5.9

5.0

K027 + atropine

64032.4

50321.0-81479.8

74.1

50.4

K033 + atropine

13721.4

10356.2-18180.3

15.9

12.6

K048 + atropine

86457.0

56689.2-131856.0

100.0

63.0

K074 + atropine

23528.8

18491.0-29939.2

27.2

20.0

K075 + atropine

29648.3

23302.4-37722.5

34.3

25.2

K203 + atropine

24451.2

16966.8-35237.1

28.3

15.8

HI-6 + atropine

17290.2

8991.5-33248.3

20.0

12.6

14282.4

10936.3-18652.2

20.0

20.0

Treatment

TMB-4 + atropine LD50 (paraoxon)=864.3 μg kg PI-protective index MDP–maximal dose of poison *

-1

The administration of a single atropine dose of 10 mg kg-1 one minute after paraoxon resulted in a protective index (PI) of 5.9, although in most cases of nerve agent poisoning, therapy with atropine alone results in a PI below 2 (34). Even though it seemed that atropine reduced the signs of cholinergic toxicity elicited by paraoxon more efficiently than that caused by nerve agents, survival was improved and symptoms reduced only following the coadministration of oximes. A marked improvement in the therapy of paraoxon poisoning was noticed with all of the tested oximes compared to atropine alone (Tables 2 and 3).

When they were applied at a dose of 5 % of their LD50, conventional oximes HI-6 and TMB-4 ensured a PI of 20.0.

The PI of the equitoxic dose of K-oximes ranged from 15.9 to 100.0 LD50 of paraoxon (Table 2). The best results were obtained with oximes K027 and K048. The PI of the therapy composed of K027 or K048 and atropine was about 15 times better than that obtained by atropine alone. Moreover, these combinations ensured the survival of all animals at up to 63.0 LD50 of paraoxon. A dose-response relationship was observed for K074, K075, and K203, as an increase of the therapeutic dose from 5 to 25 % of their LD50 resulted in a 2 to 5 times higher PI (max. was 171.5). The highest PI was obtained using 25 % of K075 and K203 LD50 together with atropine, where all animals survived up to 100.0 LD50 of paraoxon (Table 3). These results confirmed the hypothesis

Table 3 Therapeutic effect of the tested oximes (25 % of their LD50) combined with atropine upon (s.c.) paraoxon toxicity* in male mice LD50 (μg kg-1)

95 % confidence limits (μg kg-1)

PI

MDP

atropine

4191.8

3563.3-4931.2

5.9

5.0

K027 + atropine

54992.1

44222.2-68384.9

63.6

50.4

K033 + atropine

21782.1

14282.4-33220.0

25.2

15.8

K048 + atropine

51903.4

43503.0-61927.0

60.0

40.0

K074 + atropine

47141.1

37051.0-59979.1

54.5

40.0

K075 + atropine

148217.1

116493.0-188582.0

171.5

100.0

K203 + atropine

108918.1

82205.3-144311.2

126

63.0

HI-6 + atropine

20594.4

17261.2-24571.4

23.8

15.9

TMB-4 + atropine

28523.3

21527.8-37792.0

40.0

25.2

Treatment

LD50 (paraoxon)=864.3 μg kg-1 PI-protective index MDP–maximal dose of poison *

Žunec S, et al. Comparative determination of the efficacy of bispyridinium oximes in paraoxon poisoning Arh Hig Rada Toksikol 2015;66:129-134

that oxime reactivation is a very important treatment modality for OP compound poisoning.

Oximes exhibit their potency by enabling the recovery of an active AChE in contrast to the symptomatic treatment of excessive cholinergic stimulation with atropine largely in the periphery. By testing the therapeutic efficacy of newer oximes in paraoxon-poisoned mice, we aimed to single out compound(s) that would be more effective than the oximes used currently. With the exception of K033, all of the oximes showed better antidotal activity than HI-6 and TMB-4.

Unfortunately, the high acute toxicity of these oximes is a limiting factor for their usage. The protective index of the two less toxic oximes, K027 and K048, ranged from 60.0 to 100.0. Moreover, therapy with both 5 and 25 % LD50 doses of K027 plus atropine resulted in the survival of all animals at a 50.4 LD50 dose of paraoxon. A similar result was obtained for K048, with the exception that a higher therapeutic efficiency was achieved with a lower dose of this oxime (5 % of the respective LD50). Results of our in vivo experiments on mice showed a relatively good correlation with in vitro results obtained by other authors.

To be more precise, oximes K027 and K048 were found to be potent reactivators of the erythrocyte AChE in in vitro studies with methyl-, ethyl-paraoxon and DFP (23, 25-27, 29). Thus, it seems that these compounds have a pharmacological effect indeed related to the reactivation of paraoxon-inhibited AChE, and - what is even more important - these oximes can be used at lower doses, applicable for human use.

In summary, this study indicates a higher potency of the majority of the tested K-oximes to protect against high lethal doses of paraoxon when compared to the conventional oximes HI-6 and TMB-4. Among the tested oximes, K027 and K048 stood out with low acute toxicities and very good antidotal effects. We can conclude that K027 and K048 might be antidotes for paraoxon intoxication therapy.

Acknowledgements The authors thank Jasna Mileković, Marija Kramarić, and Mirjana Matašin for technical assistance. This work was supported by the Ministry of Science, Education and Sports of the Republic of Croatia (Grant No. 022-02221482139), the Grant Agency of the Czech Republic (no.

GAP15-16701S), and the Ministry of Education, Youth and Sports of the Czech Republic (no. LD13009).

Conflicts of interest The authors declare no conflict of interest.

REFERENCES 1. Kwong TC. Organophosphate pesticides: biochemistry and clinical toxicology. Ther Drug Monit 2002;24:144-9. PMID: 11805735

133

2. Macllwain C. Study proves Iraq used nerve gas. Nature 1993;363:3. doi: 10.1038/363003b0 3. Nagao M, Takatori T, Matsuda Y, Nakajima M, Iwase H, Iwadate K. Definitive evidence for the acute sarin poisoning diagnosis in the Tokyo subway. Toxicol Appl Pharmacol 1997;144:198-203. PMID: 9169085 4. Gunnell D, Eddleston M, Phillips MR, Konradsen F. The global distribution of fatal pesticide self-poisoning: systematic review. BMC Public Health 2007;7:357. doi: 10.1186/1471-2458-7-357 5. Lotti M. Clinical toxicology of anticholinesterase agents in humans. In: Krieger R, editor. Handbook of pesticide toxicology. San Diego: Academic Press; 2001. p. 1043-85.

6. Somani SM, Husain K. Low-level nerve agent toxicity under normal and stressful conditions. In: Somani SM, Romano Jr. JA, editors. Chemical warfare agents: toxicity at low levels. Boca Raton (FL): CRS Press LLC; 2001. p. 83-120.

7. Antonijević B, Stojiljković MP. Unequal efficacy of pyridinium oximes in acute organophosphate poisoning. Clin Med Res 2007;5:71-82. doi: 10.3121/cmr.2007.701 8. Kuča K, Musilek K, Jun D, Bajgar J, Kassa J. Novel oximes.

In: Gupta RC, editor. Handbook of toxicology of chemical warfare agents. London: Elsevier; 2009. p. 997-1021.

9. Barelli A, Soave PM, Del Vicario M, Barelli R. New experimental oximes in the management of organophosphorus pesticides poisoning. Minerva Anestesiol 2011;77:1197-203.

PMID: 21799476 10. Masson P. Evolution of and perspectives on therapeutic approaches to nerve agent poisoning. Toxicol Lett 2011;206:5-13. doi: 10.1016/j.toxlet.2011.04.006 11. Kuča K, Bielavský J, Cabal J, Bielavská M. Synthesis of a potential reactivator of acetylcholinesterase 1-(4hydroxyiminomethylpyridinium)-3 -(carbamoylpyridinium)propane dibromide. Tetrahedron Lett 2003;44:3123-5. doi: 10.1016/S0040-4039(03)00538-0 12. Kuča K, Bielavský J, Cabal J, Kassa J. Synthesis of a new reactivator of tabun-inhibited acetylcholinesterase. Bioorg Med Chem Lett 2003;13:3545-7. doi: 10.1016/S0960894X(03)00751-0 13. Musilek K, Holas O, Kuča K, Jun D, Dohnal V, Doležal M. Synthesis of asymmetrical bispyridinium compounds bearing cyano moiety and evaluation of their reactivation activity against tabun and paraoxon-inhibited acetylcholinesterase.

Bioorg Med Chem Lett 2006;16:5673-6. doi: 10.1016/j. bmcl.2006.08.011 14. Musilek K, Kuča K, Jun D, Dohnal V, Doležal M. Synthesis of the novel series of bispyridinium compounds bearing (E)-but-2-ene linker and evaluation of their reactivation activity against chlorpyrifos-inhibited acetylcholinesterase.

Bioorg. Med Chem Lett 2006;16:622-7. doi: 10.1016/j. bmcl.2005.10.059 15. Musilek K, Jun D, Cabal J, Kassa J, Gunn-Moore F, Kuča K. Design of a potent reactivator of tabun-inhibited acetylcholinesterase-synthesis and evaluation of (E)-1- (4 -carbamoylpyridinium)-4-(4-hydroxyiminomethylpyridi nium)-but-2-ene dibromide (K203). J Med Chem 2007;50:5514-8. doi: 10.1021/jm070653r 16. Kuča K, Stodulka P, Hrabinova M, Hanusova P, Jun D, Doležal B. Convenient preparation of the oxime HI-6 (dichloride and DMS) - antidote against nerve agents. Def Sci J 2008;58:399-404.

134

Žunec S, et al. Comparative determination of the efficacy of bispyridinium oximes in paraoxon poisoning Arh Hig Rada Toksikol 2015;66:129-134

17. Thompson WR. Use of moving averages and interpolation to estimate median effective dose. Bacteriol Rev 1947;11:11545. PMCID: PMC440915 18. Weil CS. Tables for convenient calculation of medianeffective dose (LD50 or ED50) and instruction in their use.

Biometrics 1952;8:249-63.

19. Čalić M, Lucić Vrdoljak A, Radić B, Jelić D, Jun D, Kuča K, Kovarik Z. In vitro and in vivo evaluation of pyridinium oximes: Mode of interaction with acetylcholinesterase, effect on tabun- and soman-poisoned mice and their cytotoxicity.

Toxicology 2006;219:85-96. doi: 10.1016/j.tox.2005.11.003 20. Berend S, Lucić Vrdoljak A, Radić B, Kuča K. New bispyridinium oximes: In vitro and in vivo evaluation of their biological efficiency in soman and tabun poisoning. Chem Biol Interact 2008;175:413-6. doi: 10.1016/j.cbi.2008.04.031 21. Kovarik Z, Lucić Vrdoljak A, Berend S, Katalinić M, Kuča K, Musilek K, Radić B. Evaluation of oxime K203 as antidote in tabun poisoning. Arh Hig Rada Toksikol 2009;60:19-26. doi: 10.2478/10004-1254-60-2009-1890 22. Park NJ, Jung YS, Musilek K, Jun D, Kuča K. Potency of several structurally different acetylcholinesterase reactivators to reactivate house fly and bovine acetylcholinesterases inhibited by paraoxon and DFP. Bull Korean Chem Soc 2006;27:1401-4.

23. Petroianu GA, Arafat K, Kuča K, Kassa J. Five oximes (K-27, K-33, K-48, BI-6 and methoxime) in comparison with pralidoxime: in vitro reactivation of red blood cell acetylcholinesterase inhibited by paraoxon. J Appl Toxicol 2006;26:64-71. doi: 10.1002/jat.1108 24. Petroianu GA, Nurulain SM, Nagelkerke N, Al-Sultan MA, Kuča K, Kassa J. Five oximes (K-27, K-33, K-48, BI-6 and methoxime) in comparison with pralidoxime: survival in rats exposed to the organophosphate paraoxon. J Appl Toxicol 2006;26:262-8. doi: 10.1002/jat.1143 25. Petroianu GA, Kalasz H. Comparison of the ability of pyridinium aldoximes to reactivate human RBC cholinesterases inhibited by ethyl- and methyl-paraoxon.

Curr Org Chem 2007;11:1624-34. doi: 10.2174/ 138527212800564277

26. Lorke DE, Hasan MY, Arafat K, Kuča K, Musilek K, Schmitt A, Petroianu GA. In vitro oxime reactivation of red blood cell acetylcholinesterase inhibitied by diisopropylfluorophosphate (DFP). J Appl Toxicol 2008;28:422-9. doi: 10.1002/jat.1344 27. Musilova L, Kuča K, Jung YS, Jun D. In vitro oxime assisted reactivation of paraoxon-inhibited human acetylcholinesterase and butyrylcholinesterase. Clin Toxicol 2009;47:545-50. doi: 10.1080/15563650903058914 28. Kuča K, Musilek K, Jun D, Pohanka M, Kumar Ghosh K, Hrabinova M. Oxime K027: novel low-toxic candidate for the universal reactivator of nerve agent- and pesticideinhibited acetylcholinesterase. J Enzyme Inhib Med Chem 2010;25:509-12. doi: 10.3109/14756360903357569 29. Gupta B, Sharma R, Singh N, Kuča K, Acharya JR, Ghosh KK. In vitro reactivation kinetics of paraoxon- and DFPinhibited electric eel AChE using mono- and bis-pyridinium oximes. Arch Toxicol 2014;88:381-90. doi: 10.1007/s00204013-1136-z 30. Yousefpour M, Bahrami F, Behboodi BS, Khoshbaten A, Asgari A. Paraoxon-induced ultrastructural growth changes of rat cultured hippocampal cells in neurobasal/B27.

Toxicology 2006;217:221-7. doi:10.1016/j.tox.2005.09.018 31. Kovarik Z, Čalić M, Bosak A, Šinko G, Jelić D. In vitro evaluation of aldoxime interactions with human acetylcholinesterase. Croat Chem Acta 2008;81:47-57.

32. Petroianu GA, Lorke DE. Pyridinium oxime reactivators of cholinesterase inhibited by diisopropylfluorophosphate (DFP): Predictive value of in vitro testing for in vivo efficacy.

Mini Rev Med Chem 2008;8:1328-42. doi: 10.2174/ 138955708786369555 33. Bhattacharjee AK, Kuča K, Musilek K, Gordon RK. In silico pharmacophore model for tabun-inhibited acetylcholinesterase reactivators: a study of their stereoelectronic properties.

Chem Res Toxicol 2010;23:26-36. doi: 10.1021/tx900192u 34. Dawson RM. Review of oximes available for the treatment of nerve agent poisoning. J Appl Toxicol 1994;14:317-31. doi: 10.1002/jat.2550140502

Usporedno određivanje učinkovitosti bispiridinijevih oksima pri trovanju paraoksonom Činjenica da standardna terapija ne omogućuje dovoljnu zaštitu pri otrovanju organofosfornim spojevima (pesticidima i živčanim bojnim otrovima) potaknula nas je na istraživanje novih, učinkovitijih oksima. U uvjetima in vivo ispitali smo farmakotoksikološka svojstva šest eksperimentalnih K-oksima (K027, K033, K048, K074, K075 i K203). Terapijski učinak kombinacije K-oksima (primjenjenih u dozi 5 ili 25 % njihove LD50) i atropina testiran je na miševima otrovanim paraoksonom i uspoređen s konvencionalnim oksimima HI-6 i TMB-4. Bisoksim K074 je bio najtoksičniji (LD50=21.4 mg kg-1) za miševe, dok je monooksim K027 bio najmanje toksičan (LD50=672.8 mg kg-1). Osim K033, svi K-oksimi pokazali su bolji terapijski učinak u miševa trovanih paraoksonom u odnosu na HI-6 i TMB-4. Iz skupine testiranih oksima istaknuli su se K027 i K048 koji su pokazali nisku akutnu toksičnost i osigurali protektivne indekse u rasponu od 60.0 do 100.0 LD50 paraoksona.

Uzmemo li u obzir da su ta dva oksima pokazala sličan terapijski učinak bez obzira na primjenjenu dozu, prikazani rezultati upućuju na K027 i K048 kao perspektivne antidote u terapiji trovanja paraoksonom.

KLJUČNE RIJEČI: akutna toksičnost; miševi; terapijska učinkovitost

135

Mažuran N, et al. Effects of CaCl2 and CaBr2 on the reproduction of Daphnia magna Straus Arh Hig Rada Toksikol 2015;66:135-140

Original article

DOI: 10.1515/aiht-2015-66-2516

The effects of CaCl2 and CaBr2 on the reproduction of Daphnia magna Straus Neda Mažuran1, Vladimir Hršak2, and Goran Kovačević3 Salopekova 2b1, Division of Botany2, Division of Zoology3, Department of Biology, Faculty of Science, University of Zagreb, Zagreb, Croatia [Received in March 2014; CrossChecked in March 2014; Accepted in May 2015] Concentrated CaCl2 and CaBr2 salt solutions of densities up to 2.3 kg L-1 are regularly used to control hydrostatic pressure in oil wells during special operations in the exploration and production of natural gas and crude oil. Various concentrations of high density salts are frequently left in mud pits near the drilling site as waste, polluting fresh and ground waters by spillage and drainage. The toxic effects of these salts have already been observed. This study investigated the effects of CaCl2 and CaBr2 on water flea Daphnia magna Straus in a 21-day reproduction test. The three tested concentrations of CaCl2 (240, 481, and 1925 mg L-1) caused a significant dose-response decrease of reproduction (p<0.001). With CaBr2 (533 and 1066 mg L-1), only aborted eggs were produced, demonstrating the embryotoxicity of the substance. The results suggest that high concentrations of the tested chemicals are harmful to Daphnia’s reproduction and could reduce its abundance.

KEY WORDS: acute immobilisation test; bromide embryotoxicity; chronic reproduction test; high density salts Concentrated solutions of calcium chloride (CaCl2) and calcium bromide (CaBr2), densities from 1.39 to 2.30 kg L-1, and their mixtures commonly known as high density salts or clear brines, are extensively used as completion or workover fluids in oil and natural gas exploration and production industry. Their main function is to provide the required density that controls subsurface pressure in the formations with abnormally high pressure, as well as to minimize formation damage, maintain borehole stability, transport moveable solids, and suspend solids, all of which maximise the recovery of hydrocarbons from the production reservoir (1-3). After operations, high density brines are often left in the mud pits near the well site as waste and may pollute fresh and groundwater by spilling over and draining through the ground.

Several studies investigating the toxic effects of concentrated solutions of CaCl2 and CaBr2 have been performed using plant test systems and an animal test organism, the freshwater snail Planorbarius corneus (Linnaeus, 1758) (4-8). Studies on plants (4-7) showed a significant inhibition of duckweed (Lemna minor) growth at higher concentrations of CaCl 2 (5550, 8325, and 9624 mg L-1) and CaBr2 (14931 and 21320 mg L-1).

Photosynthetic pigments chlorophyll a and b increased and the stress indicator anthocyanin were also increased in greater duckweed (Spirodella polyrrhiza) (4-7). Growth inhibition was observed in the green algae Chlorella kessleri Correspondence to: Vladimir Hršak, Department of Botany, Faculty of Science, University of Zagreb, Marulićev trg 20, E-mail: vladimir.hrsak.@biol.pmf.hr

at 21320 mg L-1 CaBr2 (5). The study on P. corneus showed a significant dose response decrease of survival and fecundity (egg-mass number) in chronic toxicity test at concentrations ranging from 1203 to 4813 mg L-1 CaCl2, 1066 to 5329 mg L-1 CaBr2, and 774 to 5801 mg L-1 of their mixture 1:1 (8).

To further explore the toxic effects of CaCl2 and CaBr2, observed in our previous study on P. corneus (8), we selected Daphnia magna Straus as a model system to perform a semistatic 21-day reproduction test. This species frequently occupies habitats in north-western Croatia, where gas and oil exploration and production are performed. Furthermore, water fleas as primary grazers and primary forage for invertebrates and vertebrates represent ecologically important constituents of food webs and are also an established model commonly used in toxicology (9, 10).

MATERIALS AND METHODS Daphnia laboratory culture Daphnids originating from a local wild population and an unpolluted area were reared in the laboratory culture in 1000 mL glass vessels at a density of 50 individuals per L in dechlorinated tap water passed through activated charcoal. They were fed daily with a concentrated suspension of fresh cultured green algae Chlorella vulgaris and Scenedesmus obliquus alternately with an addition of suspended dry baking yeast once a week and kept at

136

Mažuran N, et al. Effects of CaCl2 and CaBr2 on the reproduction of Daphnia magna Straus Arh Hig Rada Toksikol 2015;66:135-140

20±2 °C at 12/12 photoperiod (550 lux). Once a week, daphnids were transferred to fresh water and neonates were regularly removed.

Acute and chronic semi-static toxicity test procedure Concentrations of the chemicals for chronic reproduction test were chosen based on 24 and 48 h preliminary exposure.

Acute and chronic tests were conducted according to the OECD standard procedures (11, 12). For the acute exposure, four replicates for each treatment and control were used with five neonates for each replicate. There were five concentrations for both tested chemicals. Animals were kept under the same laboratory conditions as those in the chronic test but were not fed. Acute exposure to CaCl2 showed 100 % mobility after 48 h in 240 and 481 mg L-1 and 60 % mobility in 1925 mg L-1 and for CaBr2 90 and 80 % mobility for 533 and 1066 mg L-1 respectively, so the chronic reproduction test was set with the same chemical concentrations.

For the chronic experiments, neonates from 3rd-5th brood ≤24 h old (from mothers previously acclimated in dilution water) were placed individually in 50 mL of test solution in glass beakers covered with watch glass. Ten replicates per test solution and ten replicates as a control were kept at room temperature (20±2 °C), at 12:12 photoperiod.

Animals were fed daily with 4×105 cell mL-1 of the algae species Chlorella vulgaris and Scenedesmus obliquus 2:1.

Three times a week. the animals were transferred to a fresh test solution and test beakers were positioned in a random manner within the testing area, neonates and aborted eggs were recorded and discarded, survival was recorded daily.

Dilution water used for preparing test solutions and the controls (ISO test water) contained 11.76 g CaCl2x2H2O, 4.93 g MgSO4x7H2O, 2.59 g NaHCO3, and 0.23 g KCl.

Each salt was diluted in deionised water to 1 L and 25 mL of each solution was added to prepare 1 L of dilution water.

The dilution water had a conductivity of <10 μS cm-1, pH 7.8 and hardness 250 mg L-1 CaCO3 (11). ISO test water contains 294 mg L -1 CaCl 2 so each exposure CaCl 2 concentration was an addition to that content. All chemicals used in the experiments were produced by Merck (Darmstadt, Germany) and of p.a. grade, except CaBr2 which was extra pure grade. The actual concentrations of the chemicals in the test solutions were measured once a week after preparing fresh solutions and they varied from the nominal concentrations ±0.1 to 5 %. Calcium was measured with a Varian Techron AA5 atomic absorption spectrophotometer (13), bromide was measured by the spectrophotometric method (14), and chloride was measured by silver nitrate volumetric method (15).

Statistical analysis The experiments were statistically analysed with onefactorial ANOVA and Tukey post hoc test to reveal the statistical differences between treatments (16), using Statistica 6 (StatSoft, Tulsa, AZ, USA). The half maximal effective concentration EC50 values were calculated using regression analysis and Bonferroni test and the significance of females’ mortality was determined using CochranArmitage trend test (12).

RESULTS The results are given in Table 1 and Figures 1 and 2.

Our results of the acute immobilisation tests showed that the EC50 48-hour value for CaCl2 was 2020 mg L-1 (p=0.01) and for CaBr2 2238.6 mg L-1 (p=0.003).

Figure 1 Effect of CaCl2 on reproduction of Daphnia magna: mean number of living neonates per female (number of female at the start of the test which did not inadvertently or accidentally die during the test, determined by Cochran-Armitage trend test)

137

Mažuran N, et al. Effects of CaCl2 and CaBr2 on the reproduction of Daphnia magna Straus Arh Hig Rada Toksikol 2015;66:135-140

Table 1 Reproduction and survival of Daphnia magna exposed to CaCl2 and CaBr2 for 21 days Group

No. of neonates

No. of broods

Brood size

Days to 1st brood

Viable neonates/ surviving female/day

No. of aborted eggs

Females’ mortality (n)

Longevity (Days)

Control

85.5±7.72

4.5±0.4

20.7±3.31

9.7±0.34

4.1±0.34

0

1

20.9±032

2

20.8 ±0.42

3

20.6±0.70

8

16.5±3.84

6

-

5

-

CaCl2 23.4±2.41* 4.1±0.3 5.7±0.53* 9.8±1.03 1.1±0.12* 0 240 mg L-1 CaCl2 17.7±1.77* 4.3±0.4 4.8±0.81* 10.2±0.79 0.8±0.08* 0 481 mg L-1 CaCl2 2.1±3.00*# 0.9±0.7*# 1.6±3.02*# 11.0±1.15 0.26±0.14*# 0 1925 mg L-1 CaBr2 3.5±1.74 5.4±2.54* 10.2±0.53 21.9±14.2 533 mg L-1 CaBr2 2.7±1.62 3.2±1.80* 12.5±5.41*$ 10.9±7.22 1066 mg L-1 The number of females for each group was 10 Results are expressed as mean±SD SD=standard deviation * Significant difference between groups and the control (p<0.001) # Significant difference compared to other groups (p<0.001) $ Significant difference compared to other groups (p<0.01) Statistical differences between groups were determined using one-factorial ANOVA and Tukey post hoc test

The results of chronic reproduction tests showed a significant dose-related reduction of neonate production in all test solutions with CaCl2 compared to the control (p<0.001). At 240 and 481 mg L-1 CaCl2, the neonate production was significantly higher than at 1925 mg L-1 (p<0.001) and mortality was lower by 20 and 30 %, respectively. The greatest reduction appeared with the highest concentration of 1925 mg L-1 CaCl2 which caused 80 % mortality (Figure 1, Table 1). Parental mortality in exposed replicates followed a concentration–response pattern as determined by Cochran-Armitage trend test. The EC50 value for reproduction was 900 mg L-1.

Both concentrations of CaBr2 were highly toxic to adults (50 and 60 % mortality) and embryos (100 %) so only aborted eggs were found. There were twice as many produced eggs in the lower concentration of CaBr2 than in the higher concentration (Figure 2, Table 1).

As shown in Table 1, the first-brood day was significantly delayed only in 1066 mg L-1 CaBr2. The average number of neonates per brood decreased in all three CaCl 2 concentrations and the average number of broods significantly decreased only in the most toxic solution of CaCl2, 1925 mg L-1 (Table 1).

Figure 2 Effect of CaBr2 on reproduction of Daphnia magna: mean number of aborted eggs per female

138

Mažuran N, et al. Effects of CaCl2 and CaBr2 on the reproduction of Daphnia magna Straus Arh Hig Rada Toksikol 2015;66:135-140

DISCUSSION A research study on the toxicity of CaCl2 reported that the EC50–48 h for Daphnia magna was 2190 mg L-1 (17).

The EC50–48 h value of 2020 mg L-1 determined in this study is very similar to the reported value for the acute toxicity of CaCl2. The results of the reproduction test on daphnids in sublethal concentrations of CaCl2 from the same study (17) showed a significantly lower reproduction compared to the control at the concentration of 1173 mg L-1, but not at lower concentrations (235 and 496 mg L-1). The highest CaCl2 concentration investigated in our present study (1925 mg L-1 CaCl2, corresponding to 1230 mg L-1 Cl-) caused 80 % mortality and a significant loss of reproduction in the 21-day reproduction test. The two lower concentrations (240 and 481 mg L-1) were also significantly toxic so the calculated EC50 amounted to 900 mg L-1.The difference between the reported results and ours could be attributed to the different sensitivity between the D. magna clones or culture conditions (18, 19).

A report on the toxicity of chloride to high sensitive cladocerans (20) stated that the chronic toxicity threshold value for reproduction inhibition is 421 mg L-1 Cl- (Daphnia magna 21-day IC25) and 454 mg L-1 Cl- (Ceratodaphnia dubia 7-day IC25); toxicity tests measuring chloride toxicity were based on using NaCl to minimize the toxicity contributed by the more toxic counter cation such as K+, Mg2+, Ca2+. The lowest concentration investigated here (240 mg L-1 CaCl2, corresponding to 153.3 mg L-1 Cl-) that reduced neonate production to 27 % of the control indicated that CaCl2 could be more toxic than NaCl. There is a lack of information on the chronic toxicity of CaBr2 to Daphnia magna but data on chronic and acute toxicity of NaBr are widely distributed in the literature. The NOEC (No Observed Effect Concentration) on fecundity generally ranges from <3 to >117 mg L-1 NaBr (19). The differences in results may depend on the clone chosen to run the test and laboratory environment (18, 19, 21). Some of the reported acute toxic values are: LC50 of 7900 and EC50 for reproduction of 29 mg L-1 NaBr (22), EC50 of 5700 to 10800 mg L-1 NaBr (23), EC50 24 h of 37.37 mmol L-1 NaBr (24), etc. Hermens et al. (25) reported an LC50 48 h of 13500 and an EC50 (16-day reproduction) of 29 mg L-1 NaBr and concluded that the effect on reproduction for compounds like NaBr is at much lower levels than for acute mortality. Sloof and Canton (26) reported that NaBr strongly affected daphnid reproduction in a semi-chronic test. Their NOLC (No Observed Lethal Concentration) was 3200 mg L-1 and NOEC (reproduction) 10 mg L-1 NaBr. Based on the results of the long term toxicity tests on NaBr, Canton et al. (27) found that bromide had a marked effect on the reproduction of fresh water organisms (D. magna, Poecilia reticulata, and Lymnaea sp., NOEC was 7.8 mg L-1 Br-) - the bromide ion was proven to have an embryotoxic effect. Investigating chronic bromide stress on D. magna reproduction in chronic toxicity experiments, Leeuwen et

al. (28) found that bromide did not delay the onset of reproduction nor the brood frequency, but disturbed embryonic development so that at 100 mg L -1 NaBr reproduction was completely inhibited and only aborted eggs were found, but survival was unaffected at up to 10000 mg L-1. Therefore, the authors concluded that bromide inhibited the reproduction process.

The results of the present study on CaBr2 are in good agreement with the former as there were only aborted eggs released by females, significant delay to the onset of reproduction was observed only at 1066 mg L-1 CaBr2 but not at 533 mg L-1. The mean brood frequency in CaBr2 was lowered to 3.5 and 2.7 due to the high mortality of the females before the end of the test. Our results indicate that CaBr2 could be more toxic than NaBr, as it caused 50 and 60 % mortality at much lower concentrations than reported for NaBr (26, 28); however, these differences could also be explained with clonal and culturing differences (18, 19, 21).

Many toxicants can cause partial or complete abortion of clutches of eggs and embryo abnormalities e.g. NaBr and 3,4-dichloroaniline (29), ibuprofen (30), and insecticides (31). An increased number of aborted eggs and a reduction in the number of hatched individuals have been documented in NaCl solutions at concentrations of Na ranging from 25 to 249 mg L-1 Na+ and 41 to 249 mg L-1 Na+, respectively, in chronic exposure of the freshwater cladoceran Pseudosida ramosa (32, 33). Baird et al. (29) reported that the chronic inhibition of reproduction in D. magna at 3 to 17 mg L-1 NaBr was due solely to the acute lethal effect on eggs developing in the brood chamber. The decreased number of aborted eggs in higher concentration of CaBr2 compared to the lower one in the present study could indicate that such a high concentration of bromide (1066 mg L-1 CaBr2, 852 mg L-1 Br-) could also affect the production of eggs in the ovarium and not only development in the brood chamber.

Comparing the results of this investigation with those from our previous study on the freshwater snail Planorbarius corneus, it can be concluded that Daphnia magna is more sensitive to the toxic effect of the chemicals than Planorbarius corneus.

The results of this study have also shown that high concentrations of CaCl2 are toxic to Daphnia magna reproduction, reducing offspring production. CaBr2 showed an embryotoxic effect preventing neonate production. The contamination of freshwater with the investigated chemicals near drilling sites could have a harmful effect on Daphnia magna reproduction and reduce the population by reducing its abundance.

Taken together and generally speaking, we can conclude that our findings suggest the need for a more detailed clarification of the toxicity profiles of CaCl2 and CaBr2, primarily because they have demonstrated significant negative impacts on daphnid reproduction. Future studies should focus on examining the effects of these two compounds on other freshwater species.

Mažuran N, et al. Effects of CaCl2 and CaBr2 on the reproduction of Daphnia magna Straus Arh Hig Rada Toksikol 2015;66:135-140

REFERENCES 1. Schmidt DD, Hudson TE, Harris TM. Introduction on brine completion and workover fuids. Part 1 - Chemical and physical properties of clear completion brines. Petrol Eng Int 1983:80-96.

2. Meyer RL, Vargas RH. Process of selecting completion of workover fluids requires series of tradeoffs. Oil Gas J 1984;82:144-8.

3. Hemeida AM, Gawish A. Evaluation the potassium bromide and zinc bromide brines for workover operations. Oil Gas Business 2008;1:1-10.

4. Tkalec M, Vidaković-Cifrek Ž, Regula I. The effect of oil industry “High density brines“ on duckweed Lemna minor L. Chemosphere 1998;37:2703-15. doi: 10.1016/S00456535(98)00156-8 5. Vidaković-Cifrek Ž, Tkalec M, Horvatić J, Regula I. Effects of oil industry high density brines in miniaturized algal growth bioassay and Lemna Test. Phyton; Special issue: “Plant Physiology” 1999;39:193-7.

6. Vidaković-Cifrek Ž, Wonisch A, Tausz M, Grill D. Effects of CaCl2 and CaBr2 on growth, photosynthetic pigments and ion accumulation in duckweed. Phyton (Austria) 2005;45:183-96.

7. Vujević M, Vidaković-Cifrek Ž, Tkalec M, Tomić M, Regula I. Calcium chloride and calcium bromide aqueous solutions of technical and analytical grade in Lemna bioassay.

Chemosphere 2000;41:1535-42. doi: 10.1016/S00456535(00)00070-9 8. Mažuran N, Hršak V, Tomić M, Papeš D. Effects of CaCl2 and CaBr2 on the fecundity of Planorbarius corneus L. Chemosphere 1999;38:2345-55.

9. Oda S, Tatarazako N, Watanabe H, Morita M, Iguchi T. Production of male neonates in Daphnia magna (Cladocera, Crustacea) exposed to juvenile hormones and their analogs.

Chemosphere 2005;61:1168-74. doi: 10.1016/j. chemosphere.2005.02.075 10. Shaw JR, Pfrender ME, Eads BD, Klaper R, Callaghan A, Sibly RM, Colson I, Jansen B, Gilbert, D, Colbourne JK. Daphnia as an emerging model for toxicological genomics.

Adv Exp Biol 2008;2:165-219. doi: 10.1016/s18722423(08)00005-7 11. Organization of Economic Co-operation and Development (OECD). Guide- lines for the testing of chemicals no. 202.

Daphnia sp., acute immobilisation test. Paris: OECD; 2004.

12. Organization of Economic Co-operation and Development (OECD). Guidelines for the testing of chemicals no. 211.

Daphnia magna reproduction test. Paris: OECD; 2012.

13. ASTM D 511-93, 1995. Standard test methods for calcium and magnesium in water, Test method B, Atomic Absorption spectrophotometric. Annal Book of ASTM standards, Section 11, Water and Environmental Technology, Vol 11.01 Water (1). Philadelphia (PA): American Society for Testing and Materials; 1995.

14. Collins AG, Watkins JW. Spectrophotometric determination of iodides and bromides in oil field brines. Anal Chem 1959;31:1182-4. doi: 10.1021/ac60151a032 15. ASTM D 512-89, 1995. Standard test methods for chloride ion in water. Test method B, Silver nitrate titration. Annual Book of ASTM standards, Section II, Water and Environmental Technology, Vol. 11.01 Water (1). Philadelphia (PA): American Society for Testing and Materials; 1995.

16. Conover WJ. Practical Nonoparametric Statistics. 2nd ed.

New York (NY): John Wiley; 1980.

17. Bervoets L, Baillieul M, Blust R, Verheyen R. Evaluation of effluent toxicity and ambient toxicity in a polluted lowland

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

139

river. Environ Pollut 1996;91:333-41. doi: 10.1016/02697491(96)80915-8 Baird DJ, Barber I, Bradley M, Soares AMVM, Calow P. A comparative study of genotype sensitivity to acute toxic stress using clones of Daphnia magna Straus. Ecotoxicol Environ Saf 1991;21:257-65. PMID: 1868782 Baird DJ, Barber I, Bradley M, Calow P, Soares AMVM. The Daphnia bioassay: a critique. Hydrobiologia 1989;188/189:403-6. doi: 10.1007/BF00027806 Elphick JRF, Bergh KD, Bailey HC. Chronic toxicity of chloride to freshwater species: effects of hardness and implications for water quality guidelines. Environ Toxicol Chem 2011;30:239-46. doi: 10.1002/etc.365 Haap T, Köhler HR. Cadmium tolerance in seven Daphnia magna clones is associated with reduced hsp70 baseline levels and induction. Aquat Toxicol 2009;94:131-7. doi: 10.1016/j.aquatox.2009.06.006 Canton JH, Sloof W. A Proposal to classify compounds and to establish water quality criteria based on laboratory data.

Ecotoxicol Environ Saf 1979;3:126-32. PMID: 540555 Naylor C, Cox EJ, Bradley MC, Calow P. Effect of differing maternal food ration on susceptibility of Daphnia magna Straus neonates to toxic substances. Aquat Toxicol 1992;24:75-82. doi: 10.1016/0166-445X(92)90017-H Lilius H, Isomaa B, Holmström T. A comparison of the toxicity of 50 reference chemicals to freshly isolated rainbow trout hepatocytes and Daphnia magna. Aquat Toxicol 1994;30:47-60. doi: 10.1016/0166-445X(94)90005-1 Hermens J, Canton H, Steyger N, Wegman R. Joint effects of a mixture of 14 chemicals on mortality and inhibition of reproduction of Daphnia magna. Aquat Toxicol 1984;5:31522. doi: 10.1016/0166-445X(84)90012-2 Sloof W, Canton JH. Comparison of the susceptibility of 11 freshwater species to 8 chemical compounds. II. (semi) chronic toxicity tests. Aquat Toxicol 1983;4:271-82.

Canton JH, Wester PW, Mathijssen-Spiekman EAM. Study on the toxicity of sodium bromide to different freshwater organisms. Food Chem Toxicol 1983;21:369-78. PMID: 6684619 van Leeuwen CJ, Rijkeboer M, Niebeek G .Population dynamics of Daphnia magna as modified by chronic bromide stress. Hydrobiologia 1986;133:277-85. doi: 10.1007/ BF00005599 Baird DJ, Barber I, Bradley M, Soares AMVM, Calow P. An early life-stage test with Daphnia magna Straus: an alternative to the 21-day chronic test? Ecotoxicol Environ Saf 1991;22:1-7. PMID: 1914991 Heckman LH, Callaghan A, Hooper HL, Connon R, Hutchinson TH, Maund SJ, Sibly RM. Chronic toxicity of ibuprofen to Daphnia magna: Effects on life history traits and population dynamics. Toxicol Lett 2007;172:137-45. doi: 10.1016/j.toxlet.2007.06.001 Palma P, Palma VL, Fernandes RM, Bohn A, Soares AMVM, Barbosa IR. Embryo-toxic effects of environmental concentrations of chlorpyrifos on the crustacean Daphnia magna. Ecotoxicol Environ Saf 2009;72:1714-8. doi: 10.1016/j.ecoenv.2009.04.026 Freitas EC, Rocha O. Acute and chronic effects of sodium and potassium on the tropical freshwater cladoceran Pseudosida ramosa. Ecotoxicology 2011;20:88-96. doi: 10.1007/s10646-010-0559-z Freitas EC, Rocha O. Effects of sodium and potassium on life history parameters of freshwater cladoceran Pseudosida ramosa. J Braz Soc Ecotoxicol 2012;7:85-91. doi: 10.5132/ jbse.2012.02.013

140

Mažuran N, et al. Effects of CaCl2 and CaBr2 on the reproduction of Daphnia magna Straus Arh Hig Rada Toksikol 2015;66:135-140

Učinak CaCl2 i CaBr2 na razmnožavanje vodenbuhe (Daphnia magna Straus) Koncentrirane otopine soli CaCl2 i CaBr2 gustoće do 2,3 kg L-1 redovito se koriste za kontrolu hidrostatskoga tlaka u bušotinama tijekom posebnih operacija u istraživanju i proizvodnji prirodnoga plina i sirove nafte. Različite koncentracije soli visoke gustoće često su bile ostavljene u isplačnim jamama u blizini područja bušenja te su prelijevajući se ili ocjeđujući u podzemlje zagađivale slatke i podzemne vode. Iako je dosad već provedeno nekoliko istraživanja toksičnoga učinka koncentriranih otopina soli CaCl2 i CaBr2, u ovom radu istraživan je učinak CaCl2 i CaBr2 na vodenbuhu Daphnia magna u dvadesetjednodnevnom testu reprodukcije. Tri ispitane koncentracije CaCl2 (1925, 481 i 240 mg L-1) prouzročile su značajan, o dozi ovisan pad reprodukcije (p<0,001). U otopinama CaBr2 (533 i 1066 mg L-1) proizvedena su samo abortirana jaja, što pokazuje embriotoksičnost te tvari. Rezultati upućuju na to da su visoke koncentracije ispitivanih tvari štetne za razmnožavanje vodenbuhe te mogu smanjiti brojnost tih organizama.

KLJUČNE RIJEČI: akutni test imobilizacije; embriotoksičnost bromida; kronični test reprodukcije; soli visoke gustoće

Vidaković-Cifrek Ž, et al. Growth and photosynthetic responses of Lemna minor L. exposed to Cd in combination with Zn or Cu Arh Hig Rada Toksikol 2015;66:141-152

Original article

141

DOI: 10.1515/aiht-2015-66-2618

Growth and photosynthetic responses of Lemna minor L. exposed to cadmium in combination with zinc or copper Željka Vidaković-Cifrek1, Mirta Tkalec1, Sandra Šikić2, Sonja Tolić2, Hrvoje Lepeduš3, and Branka Pevalek-Kozlina1 Division of Botany, Department of Biology, Faculty of Science, University of Zagreb1, Department of Environmental Protection and Health Ecology, Andrija Štampar Teaching Institute of Public Health2, Zagreb, Faculty of Humanities and Social Sciences, Josip Juraj Strossmayer University of Osijek, Osijek3, Croatia [Received in January 2015; CrossChecked in January 2015; Accepted in May 2015] Metals have a variety of negative outcomes on plants, essential components of any ecosystem. The effects of CdCl2 (5 µmol L-1), ZnCl2 (25 or 50 µmol L-1), and CuCl2 (2.5 or 5 µmol L-1) and combinations of CdCl2 with either ZnCl2 or CuCl2 on the growth, photosynthetic pigments, and photosystem II (PSII) efficiency of duckweed (Lemna minor L.) were investigated. All of the treatments caused growth inhibition and remarkable metal accumulation in plant tissue after 4 and 7 days. In the combined treatments, the accumulation of each metal applied was lesser in comparison to treatments with single metals. After 4 days, all of the treatments generally diminished chlorophyll a content and decreased the maximum quantum yield (Fv/Fm) and effective quantum yield (ΔF/F’m) of PSII. However, after 7 days of exposure to a combination of Cd and Zn, pigment content and PSII activity recovered to control levels. A higher concentration of Cu (5 µmol L-1) as well as Cd in combination with Cu had a prolonged inhibitory effect on photosynthetic features. Our results suggest that growth inhibition was due to the toxic effect of absolute metal quantity in plant tissue. Zn counteracted Cd uptake, as seen from the recovery of pigment content and PSII efficiency in plants exposed for 7 days to the Cd and Zn combination. Cu-induced oxidative stress led to a prolonged inhibitory effect in plants treated both with a higher concentration of Cu (5 µmol L-1) and simultaneously with Cd and Cu. Our findings could contribute to general knowledge on anthropogenic and environmental contaminants that endanger plant communities and significantly disrupt the sensitive balance of an ecosystem by influencing photosynthetic mechanisms.

KEY WORDS: chlorophyll fluorescence; duckweed; metal uptake; photosynthetic pigments; PSII efficiency Excess amounts of heavy metals are present in the environment due to natural conditions or industry, mining, and agricultural and other human activities. They influence not only plant growth and reproduction but also human health (1). The impact of heavy metals on plant growth and metabolism has been studied extensively; however, the effect of plant exposure to a single heavy metal has received much more attention than the simultaneous effects of multiple metals, although contaminated terrestrial and aquatic environments usually contain a mixture of substances.

Cadmium (Cd) is a non-essential toxic heavy metal. Its salts are dangerous due to their high water solubility and relatively high soil mobility. It is easily taken up by plants and introduced into the food chain, which is why it can have adverse effect not only on plants, but also on animal and human health (2-4). Apart from growth inhibition, leaf chlorosis, and a disturbed uptake and distribution of water and essential elements, Cd has also been reported to bind Correspondence to: Željka Vidaković-Cifrek, PhD, Associate Professor, University of Zagreb, Faculty of Science, Department of Biology, Division of Botany, Rooseveltov trg 6, HR-10000 Zagreb, Croatia Tel. +385 1 48 77 739, Fax +385 1 48 26 260, E-mail: zcifrek@biol.pmf.hr

to functional groups (sulfhydryl, carboxyl, imidazole, etc.) and replace essential elements (e.g. Zn) in active sites resulting in the inhibition of enzyme activity (5-7). Cd also inhibits photosynthesis by impairing chlorophyll (Chl) biosynthesis, the activity of photosystem II (PSII), photosynthetic electron transport, the Calvin cycle, and chloroplast organization (8-11). Furthermore, Cd can cause oxidative stress by interference with the components of antioxidant defence system resulting in a disturbed cellular equilibrium between the generation and the neutralization of reactive oxygen species (ROS) (11-13).

Zinc (Zn) is an essential plant nutrient involved in the catalytic function of many enzymes, structural stability of cell membranes and proteins, DNA-binding proteins (Znfingers) as well as protection of biomembranes against oxidative damage (14, 15). It carries out biochemical functions as a divalent cation that is redox-stable under physiological conditions in biological medium, and cannot be either further reduced or oxidised (13, 15). For optimal growth of most plant species, Zn ion activity in nutrient media has to fall within the range of 1x10-5-1x10-4 µmol L‑1 (16). Zn toxicity symptoms include reduced and stunted growth, chlorosis induced by Fe-deficiency followed by

142

Vidaković-Cifrek Ž, et al. Growth and photosynthetic responses of Lemna minor L. exposed to Cd in combination with Zn or Cu Arh Hig Rada Toksikol 2015;66:141-152

decreased Chl biosynthesis, chloroplast degradation as well as interference with phosphorous, magnesium, and manganese uptake (15).

Copper (Cu) is an also essential nutrient, involved in a wide range of biochemical and physiological processes in plant cells. It participates in electron-transfer reactions of photosynthesis in the form of plastocyanin, acts as a cofactor of Cu-Zn superoxide dismutase, polyphenoloxidase, and other enzymes and participates in the structural stability of chromosomes (17). Redox cycling between its two ion forms (Cu2+ and Cu+) can lead to ROS production. Activity of Cu ions in nutrient media for optimal growth of most plant species ranges from 10-8 µmol L-1 to 10-10 µmol L-1 (16). At high levels, Cu is strongly phytotoxic and can cause growth inhibition, chlorosis, necrosis, and leaf depigmentation. Further modes of Cu toxicity at the molecular level include binding onto sulfhydryl groups of proteins, impairing essential element uptake and cell transport processes (18-21).

It has long been established that the toxic effect of Cd can be modified by essential elements such as Zn, Ca, Fe, Cu, and Mn (4, 5, 22). The effects of Cd, Zn and Cu, separately and in combination, were studied in the hydroponically grown bean Phaseolus vulgaris (23), duckweeds L. minor (20), L. gibba (24, 25) and L. trisulca (26), rice Oryza sativa (27), and cucumber Cucumis sativus (28). An extensive study of the combined effect of Cd and Zn on metal accumulation (29), oxidative stress (30), and photosynthesis (8) in freshwater macrophyte Ceratophyllum demersum showed an alleviating effect of Zn on Cd-induced toxicity.

In our previous studies (31, 32) we investigated the effect of Cd on the induction of oxidative stress and DNA damage in duckweed (L. minor L.), a rapidly growing aquatic monocotyledon commonly used in ecotoxicological studies (33). To further elucidate the effects of Cd (5 µmol L-1) alone and in combination with Zn (25 and 50 µmol L-1) or Cu (2.5 and 5 µmol L-1) on photosynthetic features, the present work evaluated growth, chlorophyll and carotenoid content as well as photosynthetic efficiency by chlorophyll fluorescence. As an added value, principal component analysis (PCA) was applied to discriminate the responses to individual metals and mixtures as well as their time dependence.

MATERIALS AND METHODS Plant material and growth conditions Duckweeds (L. minor L.) were collected from the Botanical Garden of the Faculty of Science, University of Zagreb, and sterilised according to Krajnčič and Devidé (34). The stock cultures have been maintained in chamber conditions (24±2 °C) on Pirson–Seidel nutrient medium (35) since 1996, by transferring the plants to a fresh medium

every two weeks. Illumination (16:8 hours light:dark cycle) was provided by cool white fluorescent light with an intensity of 40 μmol m‑2 s‑1 at plant level.

Cd, Zn, and Cu treatment The chemicals used for the nutrient media were purchased from Sigma-Aldrich (St. Louis, MO, USA) or Kemika (Zagreb, Croatia), CdCl2 from Fluka (Buchs, Switzerland), ZnCl2 and CuCl2 from Kemika (Zagreb, Croatia).

The experiment was carried out under static conditions after pre-cultivation (2×7-days) of plants on modified Steinberg medium prepared according to ISO 20079 protocol (36): 3.46 mmol L-1 KNO3, 1.25 mmol L-1 Ca(NO3)2 x4 H2O, 0.66 mmol L-1 KH2PO4, 0.072 mmol L-1 K2HPO4, 0.41 mmol L-1 MgSO4x7 H2O, 0.91 µmol L-1 MnCl2x4 H2O, 1.94 µmol L -1 H 3BO 3, 0.63 µmol L -1 ZnSO 4x7 H 2O, 0.18 µmol L -1 Na 2 MoO 4 x2 H 2 O, 4.03 µmol L -1 Na2EDTAx2 H2O, 2.81 µmol L-1 FeCl3x6 H2O; pH value was adjusted to 5.5 by using KOH. For the growth experiment, individual colonies comprising 2–3 fronds were inoculated to 100 mL Erlenmeyer flasks containing 60 mL of Steinberg medium.

For metal analysis, determination of pigment content as well as chlorophyll fluorescence measurement, 5-10 colonies were inoculated into 300 mL Erlenmeyer flasks containing 150 mL of medium.

For single metal treatments, Steinberg nutrient medium was supplemented with 5 µmol L-1 CdCl2, 25 µmol L-1 or 50 µmol L-1 ZnCl2 and 2.5 µmol L-1 or 5 µmol L-1 CuCl2.

The same concentrations were applied for the combined treatments, i.e. CdCl2+ZnCl2 and CdCl2+CuCl2. Control plants were grown on Steinberg medium without the addition of tested salts.

Treatments lasted for seven days in the same growth conditions as for stock cultures. Every treatment group as well as control was prepared in six replicates.

Cd, Zn, and Cu determination Plants were harvested on the 4th and 7th day of the experiment, carefully washed with distilled water, ovendried at 80 °C for 24 h, and subjected to microwave digestion in two steps. The first step was digestion of 2030 mg of dry tissue in 10 mL of 16 mol L-1 HNO3 (Kemika, Zagreb, Croatia) at 70 °C for 5 min, then 5 min at 130 °C and 4 min at 150 °C. The second step was carried out after the addition of 1 mL H2O2 (30 % v/v, Kemika, Zagreb, Croatia) at 85 °C for 5 min and additionally for 4 min at 130 °C. After cooling, the samples were diluted with 1 % (v/v) HNO 3 up to the total volume of 50 mL. Metal concentrations were determined using inductively coupled plasma-optical emission spectroscopy (ICP-OES, IRIS INTREPID II XSP – Thermo Fisher Scientific, Waltham, MA, USA) according to the ISO 11885 standard (37). The obtained results were processed using TEVA software.

Vidaković-Cifrek Ž, et al. Growth and photosynthetic responses of Lemna minor L. exposed to Cd in combination with Zn or Cu Arh Hig Rada Toksikol 2015;66:141-152

Metal concentrations were calculated according to the calibration curve obtained with a set of standards of known concentrations (Merck, Darmstadt, Germany). A concentration range of 1 to 50 μg L -1 for Cd, and a concentration range of 50 to 5000 μg L-1 for Zn and Cu were used. The detection limits were 0.5 μg L-1 for Cd and 10 μg L -1 for Zn and Cu, respectively. The limit of quantification (LOQ) was <1 μg L-1 for Cd and <20 μg L-1, for Zn and Cu. Metal contents in plant material were expressed as µg g-1 dry weight.

Bioconcentration factor (BCF) for Cd, Zn or Cu was calculated by dividing metal concentration in plant tissue (µg g-1 dry weight) measured at harvest with the initial concentration of the metal in nutrient solution (µg mL-1) (38).

Growth assessment Plant growth was estimated by counting fronds on the 4th and 7th day of the experiment and expressed as relative plant growth (RG) according to the equation (39): RG=

N t − N0 N0

143

followed by Newman-Keul’s test to determine the significant difference between treatments (P<0.05).

Results were obtained from two individual experiments.

For metal content in plant tissue, values were means of at least four replicates while for all other measurements each data point is the mean value of at least six replicates.

Data obtained for day 4 and day 7 of the experiment were used for principal component analysis (PCA) in order to evaluate the most important responses of L. minor exposed to metals, discriminate the responses to individual metals and mixtures as well as to determine their timedependence. Data sets used for PCA were comprised of 10 variables including growth, Cd, Zn, and Cu accumulation, Chl a,Chl b, total chlorophylls and carotenoid content, maximum quantum yield (Fv/Fm) and effective quantum yield (∆F/F’m). Also, malondialdehyde (MDA) content, reported previously (31, 32), was added as a variable representing oxidative stress parameter. PCA was applied to the standardised data set and the factor loadings were classified as “strong”, “moderate”, and “weak” corresponding to absolute loading values of >0.75, 0.75-0.50, and 0.500.30, respectively (42).

where Nt is the number of fronds at day t (day 4 or 7) and N0 is the number of fronds at the beginning of the experiment (day 0).

RESULTS

Photosynthetic pigments

Toxicity symptoms

Fresh samples were homogenised with 80 % (v/v) cold acetone, centrifuged at 5000 g for 10 min and the absorbance of the supernatant measured at 663, 646, and 470 nm. The photosynthetic pigments, Chl a, Chl b, and carotenoids (Cars) were calculated according to Lichtenthaler (40) and expressed as mg g-1 fresh weight.

The applied concentrations of the tested salts caused visible morphological changes. Plants treated with CdCl2 started to show signs of chlorosis and rejected roots on the 3rd day of the exposure. Fronds were smaller in comparison to control plants and induction of frond abscission was noted. Zn caused an appearance of light green daughter plants, which remained connected to mother plants thus creating larger colonies. Plants exposed to a combination of CdCl2 and ZnCl2 were smaller and also showed delayed frond abscission. In comparison to treatment with CdCl2 only, chlorosis was not so pronounced. Plants exposed to 5 µmol L-1 of CuCl2 showed the signs of chlorosis and were smaller than control plants and those treated with 2.5 µmol L-1 CuCl2. A mixture of CdCl2 and CuCl2 caused inhibition of separation of daughter plants resulting in formation of colonies with overlapping plants. In plants treated with CdCl2 combined with CuCl2 (5 µmol L-1), considerable chlorosis was noticed.

In vivo measurement of chlorophyll fluorescence In vivo chlorophyll fluorescence was measured from the upper face of duckweed fronds by pulse-modulated chlorophyll fluorometer (Qubit Systems Inc., Kingston, Canada). Prior to the measurements, plants were darkadapted for 30 min. After minimal fluorescence (F0) determination, a 0.7 s pulse of saturating light (>5000 µmol quanta m ‑2 s -1) was applied to induce maximum fluorescence (Fm). Steady-state fluorescence (F) and maximum fluorescence (F’m) of the sample under illuminated conditions (150 µmol m-2 s-1 actinic light) were also measured. From the obtained parameters (F0, Fm, F’m, and F), maximum quantum yield of PSII photochemistry (Fv/Fm) and effective quantum yield of PSII (ΔF/F’m) were calculated according to Maxwell and Johnson (41).

Data analysis Statistical analysis was performed using STATISTICA 12.0 (StatSoft Inc., Tulsa, OK, USA) software package.

Data were compared by analysis of variance (ANOVA)

Metal content in plant tissue and BCF Duckweed showed a remarkable accumulation of all of the three metals tested (Table 1). Zn, as an essential element, was detected in plants from all treatment groups, including controls. After four days of exposure to 25 µmol L-1 ZnCl2, the Zn content in plants was approximately eight times above control level, and in plants exposed to 50 µmol L-1 nine times. After seven days of exposure, Zn content

144

Vidaković-Cifrek Ž, et al. Growth and photosynthetic responses of Lemna minor L. exposed to Cd in combination with Zn or Cu Arh Hig Rada Toksikol 2015;66:141-152

Table 1 Cadmium, zinc and copper content and bioconcentration factor (BCF) values in duckweed (Lemna minor L.) after four and seven days of exposure to single Cd, Zn, and Cu salts and their combinations Treatments

Content (µg g‑1 dry weight) Cd

Zn

Cu

<LOQ

285.25±21.6d (BCF 6923.5)

<LOQ

5 µmol L‑1 Cd

834.32±33.66a (BCF 1484.5)

164.73±13.47d (BCF 3998.3)

<LOQ

25 µmol L‑1 Zn

<LOQ

2219.66±68.52b (BCF 1324.6)

<LOQ

50 µmol L‑1 Zn

<LOQ

2571.88±107.76a (BCF 777.2)

<LOQ

25 µmol L‑1 Zn+5 µmol L‑1 Cd

396.98±9.72c (BCF 706.4)

1508.88±12.24c (BCF 900.4)

<LOQ

50 µmol L‑1 Zn+5 µmol L‑1 Cd

342.82±28.30c (BCF 610.0)

2293.99±137.42b (BCF 693.2)

<LOQ

2.5 µmol L‑1 Cu

<LOQ

117.71±5.17d (BCF 2857.0)

244.71±25.53c (BCF 1540.9)

5 µmol L‑1 Cu

<LOQ

115.25±6.62d (BCF 2797.3)

363.81±14.71a (BCF 1144.9)

2.5 µmol L‑1 Cu+5 µmol L‑1 Cd

536.1±65.86b (BCF 953.9)

126.98±16.52d (BCF 3082.0)

232.22±17.71c (BCF 1462.3)

5 µmol L‑1 Cu+5 µmol L‑1 Cd

431.54±11.65c (BCF 767.9)

76.47±2.25d (BCF 1856.0)

310.9±9.76b (BCF 978.4)

<LOQ

265.58±23.82D (BCF 6446.1)

<LOQ

5 µmol L‑1 Cd

1055.64±37.94A (BCF 1878.4)

174.01±12.38D (BCF 4223.5)

<LOQ

25 µmol L‑1 Zn

<LOQ

3715.60±136.74B (BCF 2217.3)

<LOQ

50 µmol L‑1 Zn

<LOQ

4517.66±270.54A (BCF 1365.1)

<LOQ

25 µmol L‑1 Zn+5 µmol L‑1 Cd

720.54±51.14B (BCF 1282.1)

2608.82±94.14C (BCF 1556.8)

<LOQ

50 µmol L‑1 Zn+5 µmol L‑1 Cd

539.54±44.47C (BCF 960.0)

3595.83±132.63B (BCF 1086.6)

<LOQ

2.5 µmol L‑1 Cu

<LOQ

111.48±7.79D (BCF 2705.8)

342.91±36.36C (BCF 2159.4)

5 µmol L‑1 Cu

<LOQ

111.17±2.66D (BCF 2698.3)

480.98±27.82A (BCF 1513.7)

808.84±52.44B (BCF 1439.2)

96.96±5.20D (BCF 2353.4)

264.51±12.11D (BCF 1665.7)

Control

Day 4

Control

Day 7

2.5 µmol L‑1 Cu+5 µmol L‑1 Cd

698.81±60.3B 94.44±14.65D 405.87±18.43B (BCF 1243.4) (BCF 2292.3) (BCF 1277.3) The values are expressed as mean±SE of at least four replicates from two individual experiments.

Values obtained for each treatment period (day 4 and day 7) are compared separately and marked with lowercase letters for day 4 and uppercase letter for day 7. Values followed by the same letter are not significantly different (Newman-Keuls test).

Instrument’s limit of quantification (LOQ) was <1 μg L-1 for Cd and <20 μg L-1 for Cu Metal content in plants was reported also in our previous studies of Zn and Cd (31) as well as Cu and Cd (32) effects in L. minor.

BCF for corresponding metal is shown in the second row in parentheses and expressed in µg g-1 dry weight 5 µmol L‑1 Cu+5 µmol L‑1 Cd

Vidaković-Cifrek Ž, et al. Growth and photosynthetic responses of Lemna minor L. exposed to Cd in combination with Zn or Cu Arh Hig Rada Toksikol 2015;66:141-152

increased up to approximately 14 times and 17 times above control level, respectively.

Cu was expected to be present, but in much lower quantities; however, it was detectable only in plants cultivated on media supplemented with excess amounts of CuCl2. Its content in plants was also dependent on Cu concentration in culture medium and duration of treatments, being the highest after seven days of treatment with 5 µmol L-1 CuCl2.

Cd was detected only in plants cultivated on media supplemented with CdCl2. Plants exposed to 5 µmol L-1 CdCl2 for seven days contained a 1.25 times higher amount of Cd than plants exposed to the same Cd concentration for four days. Comparison of Cd and Cu accumulation in plants during single metals exposures revealed a much higher accumulation of Cd. In plants treated with the mixtures of CdCl2 and ZnCl2 or CuCl2 the concentration of both essential metals was decreased in comparison to plants treated with only ZnCl2 or CuCl2. The only exception was the treatment with CdCl2 and 2.5 µmol L-1 CuCl2, where addition of CdCl2 did not decrease the Cu content in plant tissue. The analysis of Cd content in plants exposed to the mixture of CdCl2 and ZnCl2, as well as to the mixture of CdCl2 and CuCl2, showed also a lower content of Cd in plant tissue after combined treatments. After the 4-day treatment period, plants treated with CdCl2 and 25 µmol L-1 ZnCl2 contained only 47 % and plants treated with CdCl2 and 50 µmol L-1 ZnCl2 only 41 % of total amount of Cd accumulated in plants treated with CdCl 2 only. After seven days, Cd content was also

145

significantly lower and contained 68.2 and 51.1 %, respectively, of Cd content in plants treated with CdCl2 only. In cultures exposed to CdCl2 in combination with 2.5 or 5 µmol L-1 CuCl2, content of Cd was 64 and 51.7 %, respectively, of Cd-content in plants treated with CdCl2 alone. After the 7-day treatment, Cd content was also lowered to 75 % in plants treated with 2.5 µmol L-1 CuCl2, and to 65 % in plants treated with 5 µmol L-1 CuCl2, in comparison to plants treated seven days with CdCl2 alone.

The accumulation of all three metals was dependent on time, which means that accumulation of metals after seven days was greater than after four days.

BCF values after the separate addition of either of the three tested metals decreased with the increase of metal concentrations in the nutrient medium (Table 1).

Furthermore, the BCF for the corresponding metal was even lower in plants simultaneously exposed to excess amounts of two metals. BCF values for Cd and Cu increased with time, while BCF for Zn was the highest at the end of the experiment only in plants treated with ZnCl2.

Effect on growth In comparison to control plants, all treatments significantly reduced duckweed growth rate on day 4 and day 7 of the experiment (Figure 1). Analysis of the relative growth of plants treated with CdCl2 alone and plants exposed to the combinations of salts showed no significant effects on Cd-toxicity.

Figure 1 Relative growth of duckweed (Lemna minor L.) after four and seven days of exposure to single Cd, Zn and Cu salts and their combinations Values (expressed as means±SE of six replicates) for each treatment period are compared separately and marked with lowercase letter for day 4 and uppercase letter for day 7 Values followed by the same letter are not significantly different (Newman-Keuls test)

146

Vidaković-Cifrek Ž, et al. Growth and photosynthetic responses of Lemna minor L. exposed to Cd in combination with Zn or Cu Arh Hig Rada Toksikol 2015;66:141-152

Effect on photosynthetic pigments content On day 4, all treatments, except 50 µmol L-1 ZnCl2, caused a decrease in Chl a content and total Chls (Table 2).

Chl b was affected by treatments with 5 µmol L-1 CdCl2 and combination of CdCl2 with 2.5 µmol L-1 CuCl2, while total Cars was affected by 5 µmol L-1 CuCl2. Treatment with 50 µmol L-1 ZnCl2 caused an increase of Chl a, Chl b, and Cars in comparison to control.

After seven days, in plants exposed to combinations of CdCl2 and ZnCl2 (25 or 50 µmol L-1) and in those treated with 2.5 µmol L-1 CuCl2, content of Chl a, Chl b, total Chls and Cars did not change in comparison to control levels.

Higher concentrations of CuCl2 and both combinations of CdCl2 with CuCl2 lowered the pigment content.

Effect on chlorophyll fluorescence In plants exposed to the tested metals for four days, all of the applied treatments caused a decrease of maximum quantum yield (Fv/Fm) of PSII (Figure 2a). The most prominent decrease was noticed in plants treated with both of CdCl2 and CuCl2 combinations and in plants treated with 5 µmol L-1 CuCl2.

The effective quantum yield of PSII (ΔF/F’m) after four days was also significantly decreased by the applied treatments (Figure 2b). The most prominent effect was noticed in plants exposed to CdCl2 in combination with higher concentration of ZnCl2, as well as to CdCl2 in combination with higher concentrations of CuCl2.

After seven days, in most of the treated plants, values of both fluorescence parameters were at control levels. The exceptions were treatments with CdCl2, both concentrations of CuCl2 and combination of CdCl2 with 5 µmol L-1 CuCl2.

These treatments caused lower values of Fv/Fm, while exposure to 5 µmol L-1 CuCl2 showed a decrease of ΔF/F’m.

PCA analysis PCA applied to the combined data sets obtained for day 4 and day 7 of the experiment yielded three significant PCs capturing almost 80 % of the total data variance. The first two PCA components (PC1 and PC2) explained 69 % of the total variance (Figure 3a). PC1 (45 %) was largely determined by pigment content and fluorescence parameters with strong negative loadings, as well as MDA and Cu content with moderate positive loading. PC2 (24 %) had a moderate positive loading on MDA and Cu content and moderate negative loading on growth and Zn accumulation.

The scores plot (Figure 3b) revealed a grouping of plants exposed to CuCl2, CdCl2 and combinations of these salts in one cluster based on the induction of oxidative stress (increased MDA content), decreased photosynthetic features, and/or Cu content. Most plants exposed to ZnCl2 for four days and combinations of ZnCl2 and CdCl2 as well as control plants grouped together based on responses of all photosynthetic parameters. Plants exposed seven days to both concentrations of ZnCl 2, CdCl 2 as well as

combination of CdCl2 and 2.5 µmol L-1 CuCl2 grouped together corresponding mostly in terms of decreased pigments content. PC3 representing 10 % of the total variance correlated strongly with Cd accumulation in the negative part and moderately with plant growth in the positive part but did not provide any additional information and was not included into further analyses.

DISCUSSION All of the three tested metal salts influenced duckweed growth and development, as indicated by morphological changes and reduced growth rates based on frond number.

This was to be expected, since 5 µmol L-1, and an even lower concentration of Cd, have already been confirmed to affect plant metabolism (20, 23, 43, 44). It is also wellknown that Zn and Cu added in amounts above optimal level have adverse effects on plant growth (20, 45). In our experiment, contrary to the studies on the water plant C. demersum (29) and rice (27), Zn or Cu in combination with Cd did not alleviate the inhibitory effect of Cd on growth.

When added in excess amount, the tested metals tended to accumulate with time but also, when in combination, influenced their mutual uptake. Lower Zn and Cd uptake in plants treated with a mixture of metals indicates their strong competition for transport proteins because they are both taken up into plant cells by Fe, Ca, and Zn channels/ transporters of low specificity (4, 46). There are data about the participation of transporters for other divalent cations (e.g. Fe and Zn) in Cu uptake (47) so competition between Cu and Cd is also possible, as shown after simultaneous treatment with Cd and Cu (5 µmol L-1 Cu). Therefore, the effects of Cd on growth and photosynthesis in our experiment could be due to the interference of Cd with the uptake of other nutrients (Ca, Mg, Fe, Mn, S, and P) and not only to the presence of Cd alone (4, 48).

The calculated BCF values for Cd, Zn and Cu were lower in plants exposed to combined treatments in comparison to single metal treatments, thus confirming the competition of metals for uptake mechanisms. Surprisingly, the reduced uptake of metals in combined treatments did not alleviate the effect on growth, which suggests that growth inhibition was due to the toxic effect of absolute metal quantity. Contrary, Aravind and Prasad (14) reported a suppression in Cd uptake due to increased Zn accumulation leading to an alleviated effect of Zn on Cd-induced toxicity in C. demersum. However, comparing the metal content in L. minor and. C. demersum, we can see that Cd as well as Zn content were much higher in L. minor. In our study, treatment with 50 µmol L-1 ZnCl2 affected L. minor, while in C. demersum this concentration was not toxic (14).

Therefore, it could be concluded that these two water plant species differ in metal uptake, accumulation, and/or response to metals.

0.343±0.029b

0.424±0.020b

0.639±0.035a

0.408±0.034b

0.419±0.027b

0.413±0.010b

0.383±0.032b

0.401±0.025b

0.413±0.018b

5 µmol L‑1 Cd

25 µmol L‑1 Zn

50 µmol L‑1 Zn

25 µmol L‑1 Zn + 5 µmol L‑1 Cd

50 µmol L‑1 Zn + 5 µmol L‑1 Cd

2.5 µmol L‑1 Cu

5 µmol L‑1 Cu

2.5 µmol L‑1 Cu + 5 µmol L‑1 Cd

5 µmol L‑1 Cu + 5 µmol L‑1 Cd 0.305±0.013BC

0.187±0.009E

0.341±0.012B

0.442±0.013A

0.428±0.029A

0.467±0.017A

0.251±0.018CD

0.171±0.018E

0.266±0.017CD

0.453±0.032A

7th day

0.136±0.005bc

0.120±0.010c

0.133±0.009bc

0.138±0.005bc

0.147±0.009bc

0.143±0.012bc

0.226±0.015a

0.178±0.007b

0.122±0.009c

0.178±0.014b

4th day

7th day

0.105±0.004C

0.067±0.003D

0.122±0.003B

0.155±0.006A

0.160±0.011A

0.169±0.006A

0.095±0.005C

0.063±0.005D

0.090±0.006C

0.160±0.008A

Chlorophyll b

0.589±0.026b

0.493±0.046b

0.513±0.042b

0.572±0.016b

0.596±0.038b

0.573±0.049b

0.900±0.052a

0.566±0.018b

0.458±0.035b

0.776±0.056a

4th day

0.426±0.017C

0.264±0.013D

0.498±0.021B

0.620±0.020A

0.612±0.040A

0.663±0.024A

0.360±0.023C

0.244±0.024D

0.369±0.024C

0.662±0.036A

7th day

Total chlorophylls

0.129±0.004bc

0.119±0.014bc

0.113±0.003c

0.118±0.005bc

0.123±0.004bc

0.120±0.009bc

0.169±0.008a

0.121±0.005bc

0.127±0.004bc

0.144±0.009b

4th day

7th day

0.100±0.002B

0.083±0.003C

0.112±0.003B

0.127±0.005A

0.132±0.006A

0.136±0.003A

0.104±0.003B

0.085±0.003C

0.108±0.005B

0.133±0.005A

Total carotenoids

Values (expressed as means±SE of six replicates) for each treatment period (day 4 and day 7) are compared separately and marked with lowercase letter for day 4 and uppercase letter for day 7 Values followed by the same letter are not significantly different (Newman-Keuls test)

0.569±0.039a

4th day

Chlorophyll a

Control

Treatments

Content (mg g-1 fresh weight)

Table 2 The content of photosynthetic pigments (expressed as mg g-1 fresh weight) in duckweed (Lemna minor L.) after four and seven days of metal treatments

Vidaković-Cifrek Ž, et al. Growth and photosynthetic responses of Lemna minor L. exposed to Cd in combination with Zn or Cu Arh Hig Rada Toksikol 2015;66:141-152

147

148

Vidaković-Cifrek Ž, et al. Growth and photosynthetic responses of Lemna minor L. exposed to Cd in combination with Zn or Cu Arh Hig Rada Toksikol 2015;66:141-152

a

b

Figure 2 a) Maximum quantum yield – Fv/Fm and b) effective quantum yield – ∆F/F’m of PSII in Lemna minor L. after four and seven days of exposure to single Cd, Zn, and Cu salts and their combinations Values (expressed as means±SE of six replicates) for each treatment period are compared separately and marked with lowercase letter for day 4 and uppercase letter for day 7 Values followed by the same letter are not significantly different (Newman-Keuls Test)

Vidaković-Cifrek Ž, et al. Growth and photosynthetic responses of Lemna minor L. exposed to Cd in combination with Zn or Cu Arh Hig Rada Toksikol 2015;66:141-152

Reduced photosynthetic pigments content and related chlorosis due to Cd and excess amounts of Zn and Cu as well as some other metals (Pb, Ni, and Hg) has been well documented in different plant species (44, 49). Therefore, chlorosis accompanied with reduced pigments content observed in Cd- and Cu-treated plants was not surprising.

Cd inhibits Chl biosynthesis through δ-aminolevulinic acid dehydratase and protochlorophyllide reductase due to interaction with –SH groups which leads to a diminished production of δ-aminolevulinic acid - the first common precursor for all tetrapyrroles (29). The observed effect of CdCl2 on pigments could partly be caused by Cd-induced oxidative stress (31). Contrary to redox active metals, Cd is involved in the production of ROS indirectly - by interactions with antioxidative defence system, substitution of essential elements at enzyme active sites, and disruption of electron transport chain (29, 50).

An unexpected response in pigments content was found after treatment with ZnCl2, where the lower concentration (25 µmol L-1) decreased Chl a and total Chls content on day 4 and day 7 while the higher concentration (50 µmol L-1) increased Chl b and Cars on day 4 but decreased their content on day 7. Ralph and Burchett (51) suggested that a greater impact of the lower metal concentration could be correlated to the Zn accumulation capacity of the plant. In their work on Halophila ovalis, low levels of Zn in cells were tolerated. However, the concentration of Zn exceeding the tolerance level triggered a mechanism that limits Zn uptake into the cell, therefore reducing the damage of the photosynthetic pigments. However, there are reports that Zn could maintain Chl synthesis through protecting –SH group of the oxidation-prone δ-aminolevulinic acid dehydratase and protochlorophyllide reductase (49).

Moreover, δ-aminolevulinic acid dehydratase requires Mg2+

149

or Zn2+, so Zn possibly plays a role in activating this enzyme, facilitating the formation of the Chl molecule (8). This role of Zn could also explain the higher pigment content and alleviated symptoms of chlorosis on day 7 of our experiment in plants exposed to the combination of CdCl2 and ZnCl2 in comparison to CdCl2-treated plants. Our results are in agreement with experiments on C. demersum (8, 29, 30), which proved that Zn-addition to Cd-treated plants restored Chls and Cars levels and protected chloroplasts from Cd toxicity. In rice, the addition of Zn to the nutrient solution also alleviated photosynthesis inhibition and Chl content reduction caused by Cd (27). Considering the effect of CuCl2, the lower concentration (2.5 µmol L-1) decreased photosynthetic pigment content after four days but after prolonged treatment pigment content reached control levels, suggesting a tolerance to the lower concentration of Cu. Higher Cu concentration (5 µmol L-1) affected the pigment content during the whole experiment.

Contrary to the biologically inert Zn and Cd, Cu is highly reactive due to the existence of two interconvertible oxidation states (52). It can catalyse the formation of free radicals through the Haber-Weiss reaction, and consequently cause oxidative stress in plant cells leading to the breakdown of membrane lipids, degradation of proteins, and other macromolecules, including pigments. Therefore, it was not surprising that CuCl2 enhanced CdCl2-induced chlorosis and reduction of pigments content in combined treatments.

Moreover, Cvjetko et al. (32) confirmed that Cu applied simultaneously with Cd increased Cd-induced oxidative stress remarkably.

Changes in duckweed photosynthetic functions under stress caused by the heavy metals were determined by chlorophyll fluorescence analysis. After four days of exposure, the maximum quantum yield of PSII (Fv/Fm) and

Figure 3 Principal component analysis - a) loadings and b) scores. Growth, accumulation of Cd, Zn and Cu content of Chl a, Chl b, total chlorophylls (Chl a+b) and carotenoids (Cars), maximum quantum yield (Fv/Fm) of PSII, effective quantum yield (∆F/F’m) of PSII and malondialdehyde (MDA) content [reported previously, (31, 32)], represent variables Treatments: C-control, A-Cd, B1-25 µmol L‑1 Zn, B2-50 µmol L‑1 Zn, D1-2.5 µmol L‑1 Cu, D2-5 µmol L‑1 Cu Numbers 4 and 7 represent days of treatment

150

Vidaković-Cifrek Ž, et al. Growth and photosynthetic responses of Lemna minor L. exposed to Cd in combination with Zn or Cu Arh Hig Rada Toksikol 2015;66:141-152

the effective quantum yield of PSII (ΔF/F’ m) were diminished by all of the applied metal treatments. Lower value for Fv/Fm and ΔF/F’m can indicate a certain proportion of inactive PSII reaction centres. Reduced pigment content could also decrease these parameters, which is consistent with the observed lower Chls content in plants exposed to heavy metals for four days. Diminished PSII efficiency can also indicate the inhibition of electron transport at other sites in the electron transport chain downstream of PSII and can be explained by the toxic effects of the tested metals, as both Cd and Cu are well-known for such an effect (21, 49). Also, PSII is highly sensitive to Cd and Cu because they can bind to various parts of the acceptor and donor side of PSII (9, 17, 21, 49).

Excess Zn can also disturb photosynthesis. It inhibits O2 evolution and interacts with the donor side of PSII and may also affect its acceptor side (53). Cd as well as Zn and Cu can affect enzymes of the Calvin cycle through inactivation of sulfhydryl groups (50) leading to downregulation of electron transport chain due to excessive amounts of ATP and NADPH. Recovery of PSII activity after 7-day treatment with ZnCl2 and its combination with CdCl2 could be explained by subcellular compartmentalisation and extrusion of Cd and Zn, mediated by the activation of specific transport processes (4, 54). Moreover, other physiological and biochemical processes such as increased antioxidant activity and synthesis of phytochelatins (4) could also contribute to Cd and Zn tolerance. In plants exposed for 7 days to the higher concentration of CuCl2 (5 µmol L-1) both fluorescence parameters (Fv/Fm and ΔF/F’m) were decreased. The more prominent inhibitory effect of Cu in comparison to Zn on PSII efficiency is probably due to its ability to cause oxidative stress (18). The results of our study are consistent with those by Frankart et al. (55) who found a sensitivity of L. minor photosynthetic efficiency to Cu at 3.15 and 6.3 µmol L-1.

The PCA analysis performed in this study contributed to knowledge on the differences in duckweed responses after exposure to single metals and their mixtures The results obtained imply that, unlike Zn, Cu could not alleviate Cdinduced toxicity. Therefore, our findings support previous observations by Cvjetko et al. (32) on Cu ability to induce oxidative stress and affect the photosynthetic apparatus in duckweed. Reduction of Cd-induced stress by Zn is probably mediated through lowering Cd uptake, which is in agreement with the results obtained for C. demersum (29) and tobacco (56). Interestingly, PCA did not show substantial subgrouping of the plant responses based on the treatment duration, except for the plants exposed to single Zn treatment. Furthermore, longer Zn exposure correlated with a decrease in pigment content. Such results suggest that longer exposure to Zn results in the accumulation of Zn, the excessive amounts of which exert adverse effect on plants.

The results of our study reveal significant growthmodulatory effects of Cd, Zn, and Cu in duckweed, accompanied by adverse effects on photosynthetic efficiency and pigment content, which calls for further research. Nevertheless, the findings presented here contribute to general knowledge on the specific toxicity mechanisms of metals, which both as anthropogenic and environmental contaminants may endanger plant communities and significantly disrupt the sensitive balance of an ecosystem.

Acknowledgements This work was supported by the Ministry of Science, Education and Sports of the Republic of Croatia, projects no. 119-1191196-1202 and 073-0731674-1673.

REFERENCES 1. Felix-Henningsen P, Urushadze T, Steffens D, Kalandadze B, Narimanidze E. Uptake of heavy metals by food crops from highly-polluted Chernozem-like soils in an irrigation district south of Tbilisi, eastern Georgia. Agronomy Research 2010;8:781-95.

2. Benavides MP, Gallego SM, Tomaro ML. Cadmium toxicity in plants. Braz J Plant Physiol 2005;17:21-34. doi: 10.1590/ S1677-04202005000100003 3. Prince WSPM, Senthil Kumar P, Doberschutz KD, Subburam V. Cadmium toxicity in mulberry plants with special reference to the nutritional quality of leaves. J Plant Nutr 2002;25:689-700. doi: 10.1081/PLN-120002952 4. Nazar R, Iqbal N, Masood A, Khan MlR, Syeed S, Khan NA. Cadmium toxicity in plants and role of mineral nutrients in its alleviation. Am J Plant Sci 2012;3:1476-89. doi: 10.4236/ ajps.2012.310178 5. Das P, Samantaray S, Rout GR. Studies on cadmium toxicity in plants: a review. Environ Pollut 1997;98:29-36. doi: 10.1016/S0269-7491(97)00110-3 6. Siedlecka A. Some aspects of interactions between heavy metals and plant mineral nutrients. Acta Soc Bot Pol 1995;64:265-72. doi: 10.5586/asbp.1995.035 7. Tran TA, Popova LP. Functions and toxicity of cadmium in plants: recent advances and future prospects. Turk J Bot 2013;37:1-13. doi: 10.3906/bot-1112-16 8. Aravind P, Prasad MNV. Zinc protects chloroplasts and associated photochemical functions in cadmium exposed Ceratophyllum demersum L., a freshwater macrophyte. Plant Sci 2004;166:1321-7. doi: 10.1016/j.plantsci.2004.01.011 9. Parmar P, Kumari N, Sharma V. Structural and functional alterations in photosynthetic apparatus of plants under cadmium stress. Bot Stud 2013;54:45. doi: 10.1186/19993110-54-45 10. Prasad MNV. Cadmium toxicity and tolerance in vascular plants. Environ Exp Bot 1995;35:525-45. doi: 10.1016/00988472(95)00024-0 11. Tkalec M, Prebeg T, Roje V, Pevalek-Kozlina B, Ljubešić N. Cadmium-induced responses in duckweed Lemna minor L. Acta Physiol Plant 2008;30:881-90. doi: 10.1007/s11738008-0194-y

Vidaković-Cifrek Ž, et al. Growth and photosynthetic responses of Lemna minor L. exposed to Cd in combination with Zn or Cu Arh Hig Rada Toksikol 2015;66:141-152

12. Rodríguez-Serrano M, Romero-Puertas MC, Pazmiño DM, Testillano PS, Risueño MC, del Río LA, Sandalio LM. Cellular response of pea plants to cadmium toxicity: cross talk between reactive oxygen species, nitric oxide, and calcium. Plant Physiol 2009;150:229-43. doi: 10.1104/ pp.108.131524 13. Schützendübel A, Polle A. Plant responses to abiotic stresses: heavy metal induced oxidative stress and protection by mycorrhization. J Exp Bot 2002;53:1351-65. doi: 10.1093/ jexbot/53.372.1351 14. Aravind P, Prasad MNV. Zinc alleviates cadmium-induced oxidative stress in Ceratophyllum demersum L.: a free floating freshwater macrophyte. Plant Physiol Biochem 2003;41:391-7. doi: 10.1016/S0981-9428(03)00035-4 15. Broadley MR, White PJ, Hammond JP, Zelko I, Lux A. Zinc in plants. New Phytol 2007;173:677-702. doi: 10.1111/j.1469-8137.2007.01996.x 16. Welch RM. Micronutrient nutrition of plants. Crit Rev Plant Sci 1995;14:49-82. doi: 10.1080/07352689509701922 17. Maksymiec W. Effect of copper on cellular processes in higher plants. Photosynthetica 1997;34:321-42. doi: 10.1023/A:1006818815528 18. Babu TS, Marder JB, Tripuranthakam S, Dixon DG, Greenberg BM. Synergistic effects of a photooxidized polycyclic aromatic hydrocarbon and copper on photosynthesis and plant growth: evidence that in vivo formation of reactive oxygen species is a mechanism of copper toxicity. Environ Toxicol Chem 2001;20:1351-8. doi: 10.1002/etc.5620200626 19. Kanoun-Boulé M, Vicente JAF, Nabais C, Prasad MNV, Freitas H. Ecophysiological tolerance of duckweeds exposed to copper. Aquat Toxicol 2009;91:1-9. doi: 10.1016/j. aquatox.2008.09.009 20. Khellaf N, Zerdaoui M. Growth response of the duckweed Lemna minor to heavy metal pollution. Iran J Environ Health Sci Eng 2009;6:161-6.

21. Yruela I. Copper in plants. Braz J Plant Physiol 2005;17:14556. doi: 10.1590/S1677-04202005000100012 22. Hewitt EJ. Sand and Water Culture Method Used in the Study of Plant Nutrition, 2nd ed. Technical Communication No 22.

Farnham Royal (UK): Commonwealth Agricultural Bureaux; 1966.

23. Chaoui A, Ghorbal MH, El Ferjani E. Effects of cadmiumzinc interactions on hydroponically grown bean (Phaseolus vulgaris L.). Plant Sci 1997;126:21-8. doi: 10.1016/S01689452(97)00090-3 24. Megateli S, Semsari S, Couderchet M. Toxicity and removal of heavy metals (cadmium, copper, and zinc) by Lemna gibba. Ecotoxicol Environ Saf 2009;72:1774-80. doi: 10.1016/j.ecoenv.2009.05.004 25. Khellaf N, Zerdaoui M. Growth response of L. gibba (duckweed) to copper and nickel phytoaccumulation.

Ecotoxicology 2010;19:1363-8. doi: 10.1007/s10646-0100522-z 26. Prasad MNV, Malec P, Waloszek A, Bojko M, Strzałka K. Physiological responses of Lemna trisulca L. (duckweed) to cadmium and copper bioaccumulation. Plant Sci 2001;161:881-9. doi: 10.1016/S0168-9452(01)00478-2 27. Hassan MJ, Zhang G, Wu F, Wei K, Chen Z. Zinc alleviates growth inhibition and oxidative stress caused by cadmium in rice. J Plant Nutr Soil Sci 2005;168:255-61. doi: 10.1002/ jpln.200420403

151

28. An YJ, Kim YM, Kwon TI, Jeong SW. Combined effect of copper, cadmium, and lead upon Cucumis sativus growth and bioaccumulation. Sci Total Environ 2004;326:85-93. doi: 10.1016/j.scitotenv.2004.01.002 29. Aravind P, Prasad MNV. Cadmium-zinc interactions in a hydroponic system using Ceratophyllum demersum L.: adaptive ecophysiology, biochemistry and molecular toxicology. Braz J Plant Physiol 2005;17:3-20. doi: 10.1590/ S1677-04202005000100002 30. Aravind P, Prasad MNV, Malec P, Waloszek A, Strzałka K. Zinc protects Ceratophyllum demersum L. (free-floating hydrophyte) against reactive oxygen species induced by cadmium. J Trace Elem Med Biol 2009;23:50-60. doi: 10.1016/j.jtemb.2008.10.002 31. Balen B, Tkalec M, Šikić S, Tolić S, Cvjetko P, Pavlica M, Vidaković-Cifrek Ž. Biochemical responses of Lemna minor experimentally exposed to cadmium and zinc. Ecotoxicology 2011;20:815-26. doi: 10.1007/s10646-011-0633-1 32. Cvjetko P, Tolić S, Šikić S, Balen B, Tkalec M, VidakovićCifrek Ž, Pavlica M. Effect of copper on the toxicity and genotoxicity of cadmium in duckweed (Lemna minor L.) Arh Hig Rada Toksikol 2010;61:287-96. doi: 10.2478/100041254-61-2010-2059 33. Lewis MA. Use of freshwater plants for phytotoxicity testing: a review. Environ Pollut 1995;87:319-36. doi: 10.1016/02697491(94)P4164-J 34. Krajnčič B, Devidé Z. Report on photoperiodic responses in Lemnaceae from Slovenia. Berichte des Geobot Inst ETH Stiftung Rübel (Zürich) 1980;47:75-86.

35. Pirson A, Seidel F. Zell- und stoffwechselphysiologiche Untersuchungen an der Wurzel von Lemna minor unter besonderer Berucksichtigung von Kalium- und Calciummangel [Cell metabolism and physiology in Lemna minor root deprived of potassium and calcium, in German].

Planta 1950;38:431-73.

36. ISO 20079;2005 - Water quality - Determination of the toxic effect of water constituents and waste water to duckweed (Lemna minor) - Duckweed growth inhibition test. Geneva: International Organization for Standardization; 2005.

37. ISO 11885;2009 - Water quality - Determination of selected elements by inductively coupled plasma optical emission spectrometry (ICP-OES). Geneva: International Organization for Standardization; 2009.

38. Rahmani GNH, Sternberg SPK. Bioremoval of lead from water using Lemna minor. Bioresour Technol 1999;70:22530. doi: 10.1016/S0960-8524(99)00050-4 39. Ensley HE, Barber JT, Polito MA, Oliver AI. Toxicity and metabolism of 2,4-dichlorophenol by the aquatic angiosperm Lemna gibba. Environ Toxicol Chem 1994;13:325-31. doi: 10.1002/etc.5620130217 40. Lichtenthaler HK. Chlorophylls and carotenoids: pigments of photosynthetic biomembranes. Methods Enzymol 1987;148:350-82. doi: 10.1016/0076-6879(87)48036-1 41. Maxwell K, Johnson GN. Chlorophyll fluorescence - a practical guide. J Exp Bot 2000;51:659-68. doi: 10.1093/ jexbot/51.345.659 42. Liu C-W, Lin K-H, Kuo Y-M. Application of factor analysis in the assessment of groundwater quality in a blackfoot disease area in Taiwan. Sci Total Environ 2003;313:77-89. doi: 10.1016/S0048-9697(02)00683-6 43. Kwan KHM, Smith S. Some aspects of the kinetics of cadmium and thallium uptake by fronds of Lemna minor L.

152 44.

45.

46.

47.

48.

49.

Vidaković-Cifrek Ž, et al. Growth and photosynthetic responses of Lemna minor L. exposed to Cd in combination with Zn or Cu Arh Hig Rada Toksikol 2015;66:141-152

New Phytol 1991;117:91-102. doi: 10.1111/j.1469-8137.1991. tb00948.x Dias MC, Monteiro C, Moutinho-Pereira J, Correia C, Gonçalves B, Santos C. Cadmium toxicity affects photosynthesis and plant growth at different levels. Acta Physiol Plant 2013;35:1281-9. doi: 10.1007/s11738-0121167-8 Shaw BP, Sahu SK, Mishra RK. Heavy metal induced oxidative damage in terrestrial plants. In: Prasad MNV, editor. Heavy metal stress in plants: from biomolecules to ecosystems. Berlin, Heidelberg: Springer; 2004. p. 84-126.

Clemens S. Toxic metal accumulation, responses to exposure and mechanisms of tolerance in plants. Biochimie 2006;88:1707-19. doi: 10.1016/j.biochi.2006.07.003 Palmer CM, Guerinot ML. Facing the challenges of Cu, Fe and Zn homeostasis in plants. Nat Chem Biol 2009;5:333-40. doi: 10.1038/nchembio.166 Sun JY, Shen ZG. [Effects of Cd stress on photosynthetic characteristics and nutrient uptake of cabbages with different Cd-tolerance, in Chinese]. Ying Yong Sheng Tai Xue Bao 2007;18:2605-10. PMID: 18260471 Myśliva-Kurdziel B, Prasad MNV, Strzalka K. Photosynthesis in heavy metal stressed plants. In: Prasad MNV, editor. Heavy metal stress in plants: from biomolecules to ecosystems.

Berlin, Heidelberg: Springer; 2004. p. 146-81.

50. Van Assche F, Clijsters H. Effects of metals on enzyme activity in plants. Plant Cell Environ 1990;13:195-206. doi: 10.1111/j.1365-3040.1990.tb01304.x 51. Ralph PJ, Burchett MD. Photosynthetic response of Halophila ovalis to heavy metal stress. Environ Pollut 1998;103:91-101. doi: 10.1016/S0269-7491(98)00121-3 52. Clemens S. Molecular mechanisms of plant metal tolerance and homeostasis. Planta 2001;212:475-86. doi: 10.1007/ s004250000458 53. Rashid A, Bernier M, Pazdernick L, Carpentier R. Interaction of Zn2+ with the donor side of Photosystem II. Photosynth Res 1991;30:123-30. doi: 10.1007/BF00042010 54. Krämer U, Talke IN, Hanikenne M. Transition metal transport. FEBS Lett 2007;581:2263-72. doi:10.1016/j. febslet.2007.04.010 55. Frankart C, Eullaffroy P, Vernet G. Photosynthetic responses of Lemna minor exposed to xenobiotics, copper, and their combinations. Ecotoxicol Environ Saf 2002;53:439-45. doi: 10.1016/S0147-6513(02)00003-9 56. Tkalec M, Peharec Štefanić P, Cvjetko P, Šikić S, Pavlica M, Balen B. The effects of cadmium-zinc interactions on biochemical responses in tobacco seedlings and adult plants.

PLoS ONE 2014;9:e87582. doi: 10.1371/journal. pone.0087582

Rast i fotosinteza u vodene leće (Lemna minor L.) izložene kadmiju u kombinaciji s cinkom ili bakrom Izloženost metalima može izazvati različite štetne učinke u biljaka. Vodene leće izložili smo solima teških metala CdCl2 (5 µmol L‑1), ZnCl2 (25 µmol L‑1 ili 50 µmol L‑1) i CuCl2 (2,5 µmol L‑1 ili 5 µmol L‑1) te kombinaciji CdCl2 sa svakom od navedenih koncentracija ZnCl2 i CuCl2. Rast biljaka, količina fotosintetskih pigmenata i učinkovitost fotosistema II (PSII) mjereni su nakon četiri i sedam dana tretmana. Utvrđeno je da su svi tretmani uzrokovali značajnu inhibiciju rasta te akumulaciju metala u biljci. U biljaka koje su bile izložene kombinacijama teških metala količina pojedinog metala u tkivu bila je niža u odnosu na količinu istog metala u biljaka izloženih samo tom metalu. Nakon četiri dana tretmana sva su tri metala, neovisno o tome jesu li bila primijenjena zasebno ili u kombinacijama, uzrokovala smanjenje količine klorofila a i pad vrijednosti maksimalnog (Fv/Fm) i efektivnog (∆F/F’m) prinosa PSII. Međutim, u biljaka koje su bile istovremeno izlagane kadmiju i cinku, vrijednosti količine pigmenata i učinkovitost PSII vratile su se nakon sedam dana na kontrolnu razinu, a bakar u koncentraciji 5 µmol L‑1 te kombinacija kadmija i bakra i dalje su imali inhibitorni učinak.

Budući da smanjeno primanje pojedinog metala uočeno u biljaka izloženih kombiniranim tretmanima nije ublažilo inhibitorni učinak na rast, možemo zaključiti da je inhibicija rasta uzrokovana apsolutnom količinom metala primljenog u tkivo. Povećanje količine fotosintetskih pigmenata i učinkovitosti PSII nakon sedam dana tretmana kadmijem i cinkom upućuje na oporavak biljaka, što se može objasniti ublažavajućim djelovanjem cinka na učinak kadmija uslijed smanjenog primanja kadmija u biljku. Suprotno tome, dugotrajni inhibitorni učinak istovremenog tretmana biljaka kadmijem i bakrom te samim bakrom u koncentraciji 5 µmol L‑1 može se objasniti oksidacijskim stresom uzrokovanim bakrom. Rezultati ovoga istraživanja pridonose saznanjima o štetnim učincima antropogenih i okolišnih onečišćivača, koji narušavanjem mehanizama fotosinteze ugrožavaju ne samo biljke i njihove zajednice nego i osjetljivu ravnotežu ekosustava.

KLJUČNE RIJEČI: fluorescencija klorofila; fotosintetski pigmenti; primanje metala; učinkovitost PSII; vodena leća

153

Gökalp Muranli FD, et al. Genotoxic effects of diazinon on human peripheral blood lymphocytes Arh Hig Rada Toksikol 2015;66:153-158

Original article

DOI: 10.1515/aiht-2015-66-2584

Genotoxic effects of diazinon on human peripheral blood lymphocytes Fulya Dilek Gökalp Muranli, Martin Kanev, and Kezban Ozdemir Trakya University Faculty of Science, Department of Biology, Edirne, Turkey [Received in October 2014; CrossChecked in October 2014; Accepted in May 2015] The aim of this study was to evaluate the genetic damage in human peripheral blood lymphocytes following 24 and 48hour exposure to a commercial diazinon formulation Basudin 60EM® at concentrations between 0.01 and 40 µg mL-1.

For this purpose we used the micronucleus (MN), fluorescence in situ hybridization (FISH), and alkaline single cell gel electrophoresis (comet) assay. Diazinon significantly increased the frequency of micronucleated cells compared to control.

Forty-eight-hour exposure increased this frequency even at lower concentrations (0.01-10 µg mL-1). The FISH results revealed aneugenic effects at 10 µg mL-1. The comet assay also confirmed DNA damage at concentrations between 10 and 40 µg mL-1. Our findings have confirmed the genotoxic potential of diazinon and its cytotoxic effect on human lymphocytes. The increased DNA damage in our study raises concern about the current assessment of the health risk posed by this pesticide and calls for a high level of caution in agricultural and household use.

KEY WORDS: comet assay; FISH; genotoxicity; micronucleus; organophosphate pesticides Organophosphorus pesticides, which are commonly used in agriculture, are known neurotoxic agents. Being cholinesterase inhibitors, they are less persistent but also more toxic to mammals than organochlorine pesticides (1).

Diazinon (0,0-diethyl-0-(2-isopropyl-4-methyl-6pyrimidinyl phosphorothionate) is a thionophosphorous organophosphate pesticide used to control a variety of insects in agriculture and domestic settings (2). Most studies have confirmed the genotoxic and cytotoxic effects of diazinon on animals. For example, it had a cytogenetic effect on bone marrow cells of mice (3) and a clastogenic and probably genotoxic effect on humans exposed to the insecticide (4).

Some in vivo and in vitro studies of diazinon reported controversial results. It increased the frequency of sister chromatid exchange and decreased replicative indices, suggesting toxic and genotoxic effects in vitro (5, 6). In a Chinese hamster cell line V79, it had cytotoxic effects but did not induce sister chromatid exchange (7). It also failed to increase chromosome aberrations or sister chromatid exchange in cell cultures (1, 8). In cultured human lymphocytes diazinon (0.04 µg mL-1) induced a weak increase in the number of micronucleated cells (9).

However, in another study of cultured human blood cells, it increased the frequency of micronuclei and inhibited cell proliferation (10). Contradictory findings obtained so far highlight the need to use different assays and different test materials for diazinon genotoxicity studies.

Correspondence to: Fulya Dilek Gökalp Muranli,Trakya University Faculty of Science, Department of Biology, 22030 Edirne, Turkey. E-mail: fulyadilek@trakya.edu.tr

Commercial formulations are often more toxic than the pure pesticide compound, as they contain surface active ingredients, dyes, stabilisers, activity enhancers, and organic solvents with unknown or poorly characterised toxicity.

Yet, only a few studies have investigated the effects of commercial formulations of diazinon. One has found that commercial formulations of insecticides could be toxic and harmful to the developing embryo and in vitro fertilisation (11). Another has shown that diazinon is toxic to mammalian spermatogenic cells (12).

The genotoxic potential of commercial formulations of diazinon in cultured human lymphocytes has not yet been investigated. With our study we wanted to fill this gap and learn more about the mechanisms involved in the cytotoxic and genotoxic action of commercial diazinon in peripheral blood lymphocytes of healthy human volunteers. To do that, we used the micronucleus (MN), comet, and the fluorescence in situ hybridisation (FISH) assay.

MATERIALS AND METHODS Test chemicals We used a commercial formulation obtained from a local market (Basudin 60EM®, Syngenta, Basel, Switzerland) containing 630 g of diazinon per litre of the product. The test substance was prepared in sterile double-distilled water.

The concentrations were selected based on our preliminary study whose aim was to determine mitotic index inhibition concentrations.

154

Gökalp Muranli FD, et al. Genotoxic effects of diazinon on human peripheral blood lymphocytes Arh Hig Rada Toksikol 2015;66:153-158

Mitomycin C (cat. no.: 50-07-7; Sigma, Steinheim, Germany) was used at a concentration of 0.1 μg mL-1 as positive control in the MN assay. Vinblastine sulphate (0.1 μmol L -1, cat. no.: 143-67-9; MP Biomedicals, California, USA) was used as aneugenic agent in the FISH assay, and hydrogen peroxide (H2O2; 20 µg mL-1, cat. no.: 7722-84-1; Sigma) was used as positive control in the comet assay.

Blood sampling and lymphocyte culturing Samples of venous blood were collected from four healthy, non-smoking volunteers, two women and two men, aged between 20 and 25 years, with no history of pesticide exposure. The donors were informed about the study and signed the consent form allowing the use of their blood samples. The investigation observed the ethical standards laid down by the Declaration of Helsinki.

Blood lymphocytes were cultured at 37 °C for 70 h by adding 0.3 mL of whole blood to 4.7 mL of Ham’s F-10 medium (Sigma) supplemented with 20 % foetal bovine serum (Sigma), 2 % phytohaemagglutinin (Gibco, Paisley, UK), and antibiotics (penicillin at the concentration of 100 IU mL-1 and streptomycin at 100 µg mL-1) (IE Ulagay, Istanbul, Turkey.)

MN assay The cytokinesis-block micronucleus (CBMN) assay is used to detect early biological effects of DNA-damaging and spindle-damaging compounds (13). Combined with hybridisation and general or chromosome-specific centromeric/telomeric probes, CBMN can identify the mechanisms most responsible for micronucleation.

To block cytokinesis, we added cytochalasin B (cat. no.

14930-96-2, Sigma) at the final concentration of 6 µg mL-1 44 h after the cultures were started. Diazinon was added to the cultures in the concentrations of 0.01, 0.1, 1, 5, or 10 µg mL-1 on hours 22 and 46 for 48 and 24 h, respectively.

The CBMN assay followed the method described by Fenech (14). We scored the number of micronuclei in 1500 binucleated cells per donor (a total of 6000 binucleated cells per concentration and treatment period). To determine lymphocyte proliferation, we analysed 500 cells for each concentration per donor (a total of 2000 cells per concentration and treatment period). The nuclear division index (NDI) was calculated as follows: NDI= (M1+2M2+3M3+4M4)/n. M1-M4 indicates the number of nuclei per cell from, and n is the total number of cells recorded (15).

FISH assay We used the FISH assay with a centromere-specific α-satellite DNA probe (PanCentromeric, DiaGen, Ankara, Turkey) to distinguish the micronuclei produced by clastogens from those produced by aneugens. The FISH protocol was based on a procedure described elsewhere

(16). The cells used for FISH were treated with diazinon at the concentrations of 5 and 10 µg mL-1 for 48 h. The slides were prepared in line with the protocol for the CBMN assay and stored at -20 °C for later analysis. After pre-treatment with RNase, HCl, and pepsin, the slides and the probe were denatured by baking on a hot block at 70 °C, then hybridised at 37 °C overnight, and counterstained with 4’,6-diamidino2-phenylindole (DAPI) (DiaGen). For analysis we used a fluorescence microscope (Olympus BX51, Tokyo, Japan) with 360 nm, 460 nm, and 510 nm excitation filters.

Micronuclei were scored as centromere-positive when the brightness of the pancentromere probe signal in the micronucleus was comparable with the spots in the nucleus.

The scores are expressed as percentage. Only the cells with micronuclei showing bright fluorescent spots were classified as centromere-positive, and the rest were classified as centromere-negative micronuclei. Fifty micronuclei were scored for positive or negative centromeres for each diazinon concentration and centromere-positive micronuclei expressed as percentage.

Comet assay Lymphocytes were isolated from 5 mL of heparinised blood using the Ficoll-Hypaque (Sigma) density gradients and washed with the RPMI 1640 medium (Sigma). The concentration of cells in the medium was about 2×105 mL-1.

Before the experiment started, we checked the lymphocytes for viability using the Trypan blue dye (Sigma). The test was continued if cell viability was more than 90 %. As in our preliminary study we found no DNA damage at concentrations below 5 µg mL-1, the lymphocytes were exposed to the final diazinon concentrations of 5, 10, 20, and 40 µg mL-1 at 37 °C for 0.5 h. The cells were washed twice with RPMI 1640 and re-suspended in the medium.

The comet assay followed the procedure described by Singh et al. (17) and the slides were prepared as described

by Bajpayee et al. (18). They were stained with ethidium bromide (20 µg mL-1) for 5 min and observed with a fluorescence microscope (Olympus BX51) using a 510 nm excitation filter. Images of 100 randomly selected comets (50 cells from each of two replicate slides) were analysed from each sample, and the tail length, tail intensity, and tail moment measured on screen using the Comet Score computer-based image analysis system (TriTek Corp., Sumerduck, VA, USA).

Statistics The statistical analyses (IBM SPSS Statistics for Windows, version 22.0. IBM, Armonk, NY, USA) included all data from the four donors. The frequency of binucleated micronuclei in the MN assay (BNMN‰) and of the percentage of centromere-positive (C+) micronuclei in the FISH assay were compared to control using Fisher’s exact test. NDI was analysed using the chi-square test. Pearson correlation was used to correlate NDI, BNMN‰, and

155

Gökalp Muranli FD, et al. Genotoxic effects of diazinon on human peripheral blood lymphocytes Arh Hig Rada Toksikol 2015;66:153-158

concentrations. The results of the comet assay were analysed with the one-way analysis of variance (ANOVA) and the post hoc analysis of differences between the groups with the least significant difference test. All differences equalling and below 0.05 were considered significant.

RESULTS AND DISCUSSION Compared to controls, 24-hour exposure to diazinon significantly increased micronucleus frequency at concentrations of 1, 5, and 10 µg mL-1, and 48-hour exposure significantly increased the frequency at the concentrations of 0.01, 0.1, 1, 5, and 10 µg mL-1 (Table 1).

The insecticide also significantly decreased NDI as the concentrations increased (Pearson correlation, p=0.038) (Figure 2). Correlations were also observed between BNMN‰ and NDI following 24 (Pearson correlation, p=0.006) and 48-hour exposure (p=0.024) (Figure 1).

Diazinon concentrations and BNMN‰ did not correlate.

These correlations point to the genotoxic and cytotoxic activity of the pesticide. An earlier study of human lymphocyte chromosomes in cell culture (19) reported that diazinon did not significantly increase the percentage of chromosomal aberrations but that it was clearly cytotoxic, as indicated by reduced mitotic indices. This result might be explained by the cytotoxic effect masking chromosome aberrations. Another study indicated that the percentage of metaphase cells with structural aberrations decreased in a dose-dependent manner in human lymphocyte cultures (1).

All these findings are in line with ours. The cytotoxic effects

of diazinon did not necessarily cause chromosome breaks but could lead to spindle disturbance.

The novelty of our study are the FISH assay findings.

Diazinon exposure at the concentrations of 5 and 10 µg mL-1 has demonstrated that the insecticide has aneugenic effects (centromere-positive micronuclei; see Table 2). Some organophosphates with mechanisms of action similar to diazinon such as trichlorfon have been shown to induce aneuploidy in vivo (20). Sun et al. (21) have reported that trichlorfon is a potent spindle poison in V79 cells and that it induces aneuploidy in mouse spermatocytes. Some organophosphate pesticides also induced aneuploidy in in vivo studies with pesticide sprayers (22, 23). Using the comet assay, Karzmer reported carbofuran-induced DNA damage in human peripheral blood lymphocytes (24), whereas Želježić et al. (25) reported aneugenic effects, established by the FISH assay, as the frequency of micronuclei and C+ positive micronuclei increased.

Diazinon-induced damage observed in our study might be related to its cytotoxic and apoptotic action. Table 3 shows the amount of DNA breakage (tail length), percentage of DNA (tail intensity), and DNA migrated in the tail (tail moment) determined with the comet assay. Tail length and tail moment significantly increased at the concentrations of 10 and 40 µg mL-1. Tail intensities significantly increased at all concentrations compared to negative control (Figure 2), demonstrating the DNA-damaging potential of diazinon.

These findings support earlier reports on diazinon-related DNA effects (26-28) and point to two main mechanisms of diazinon action on the cell level: genotoxicity, which was established by the comet assay and aneugenicity, which was

Table 1 Micronucleus (MN) frequency and the nuclear division index (NDI) in human lymphocytes exposed to diazinon (total values of four donors) Exposure period Concentration (µg mL-1) (-)Control 24 h Distilled water (+) Control MMC 0.01

48 h

BNMN‰±SE

BN%

NDI±SE

4.98±0.28

24.3

1.35±0.035

***65.5±3.38

8.3

***1.08±0.01

8.44±1.76

21.9

1.31±0.02

0.1

8.85±1.15

21.4

**1.26±0.03

1

*13.34±1.53

18.0

***1.23±0.03

5

*13.78±0.88

12.6

***1.16±0.04

10 (-)Control Distilled water (+) Control MMC 0.01

*13.76±1.75

12.4

***1.15±0.04

2.88±0.58

22.8

1.30±0.01

***196.69±14.1

2.4

***1.02±0.01

*11.74±2.85

21.0

1.26±0.01

0.1

*10.44±0.76

18.4

*1.23±0.01

1

*11.50±2.09

14.7

***1.18±0.02

5

**14.45±1.74

10.4

***1.14±0.04

10 **15.46±0.76 9.6 ***1.12±0.02 BNMN-binucleated cells with a micronucleus; BN% - percentage of binucleated cells; BNMN‰ - mean number of micronuclei per 1000 binucleated cells; SE - standard error

*p≤0.05; **p≤0.01; ***p≤0.001 (Fisher’s exact test for BNMN‰ and chi-square test for NDI)

156

Gökalp Muranli FD, et al. Genotoxic effects of diazinon on human peripheral blood lymphocytes Arh Hig Rada Toksikol 2015;66:153-158

A B

Figure 1 Correlations between the frequency of binucleated micronucleated lymphocytes (BNMN‰) and the nuclear division index (NDI) after 24 h (A) and 48 h (B) of exposure (Pearson correlation, p=0.006 and p=0.024, respectively). The figure also shows the correlations between NDI and diazinon concentrations after 24 h (A) and 48 h (B) of exposure (Pearson correlation, p=0.038) Table 2 The number and percentage of centromere-positive (C+) and centromere-negative (C-) micronuclei after fluorescence in situ hybridization (FISH) Chemicals and concentrations

MN (C-)

MN (C+)

% MN (C+)

Negative control

12.6±2.3

25.6±2.3

69.3±0.6

0.1 µmol L vinblastin sulphate

9±1.4

41±1.4

**82±2.8

14.3±2.6

31.3±5.7

68.7±1.6

7.6±3.6

29±3.5

*81±5.5

-1

5 µg mL-1 diazinon 10 µg mL diazinon *p≤0.05; **p≤0.01 data are given as mean±SE -1

Table 3 Tail length, intensity, and moment comparison between the treated groups and negative control Sample Negative control Positive control (20 µg mL-1 H2O2)

Tail length Tail intensity Tail moment 0.997

1.000

1.000

***0.001

**0.002

**0.002

5 µg mL-1 diazinon

-0.357

-0.234

-0.592

10 µg mL-1 diazinon

***0.000

***0.001

**0.004

20 µg mL-1 diazinon 40 µg mL-1 diazinon *p≤0.05; **p≤0.01; ***p≤0.001

-0.993

**0.003

-0.329

***0.000

***0.000

***0.000

Figure 2 Tail intensities by groups. All treated groups significantly differed from negative control

Gökalp Muranli FD, et al. Genotoxic effects of diazinon on human peripheral blood lymphocytes Arh Hig Rada Toksikol 2015;66:153-158

established by the MN and FISH assays. Both mechanisms contribute to the diazinon’s cytotoxic activity, possibly by triggering apoptosis. Further research should investigate the exact mechanism of diazinon action using other test methods.

The increased DNA damage in peripheral lymphocytes in our study raises concern about the current assessment of the health risk posed by this pesticide and calls for a high level of caution in agricultural and household use.

Conflict of interest statement We declare that no conflict of interest exists in relation to this study or this article.

REFERENCES 1. Lopez D, Aleixandre C, Merchan M, Carrascal E. In vitro induction of alterations in peripheral blood lymphocytes by different doses of diazinon. Bull Environ Contam Toxicol 1986;37:517-22. doi: 10.1007/BF01607798 2. Cox C. Diazinon: toxicology. J Pest Reform 2000;20:15-21.

3. Altamirano-Lozano MA, Camacho-Manzanilla M.del C, Loyola-Alvarez R, Roldan-Reyes E. Mutagenic and teratogenic effects of diazinon. Rev Int Contam Ambient 1989;5:49-58.

4. Satar S, Kayraldiz A, Rencuzogullari E, Karakoc E, Sebe A, Avci A, Yesilagac H, Topaktas M. The genotoxicity and cytotoxicity among patients diagnosed with organophosphate poisoning. Bratisl Lek Listy 2009;110:476-9. PMID: 19750985 5. Hatjian BA, Mutch E, Williams FM, Blain PG, Edwards JW. Cytogenetic response without changes in peripheral cholinesterase enzymes following exposure to a sheep dip containing diazinon in vivo and in vitro. Mutat Res 2000;472:85-92. doi: 10.1016/S1383-5718(00)00131-5 6. Sobti RC, Krishan A, Pfaffenberger CD. Cytokinetic and cytogenetic effects of some agricultural chemicals on human lymphoid cells in vitro: organophosphates. Mutat Res 1982;102:89-102. PMID: 6981766 7. Kuroda K, Yamaguchi Y, Endo G. Mitotic toxicity, sister chromatid exchange, and rec assay of pesticides. Arch Environ Contam Toxicol 1992;23:13-8. doi: 10.1007/BF00225990 8. Chen HH, Hsueh JL, Sirianni SR, Huang CC. Induction of sister-chromatid exchanges and cell cycle delay in cultured mammalian cells treated with eight organophosphorus pesticides. Mutat Res 1981;88:307-16. PMID: 7254224 9. Bianchi-Santamaria A, Gobbi M, Cembran M, Arnaboldi A. Human lymphocyte micronucleus genotoxicity test with mixtures of phytochemicals in environmental concentrations.

Mutat Res 1997;388:27-32. doi: 10.1016/S13835718(96)00128-3 10. Colovic M, Krstic D, Petrovic S, Leskovac A, Joksic G, Savic J, Franko M, Trebse P, Vasic V. Toxic effects of diazinon and its photodegradation products. Toxicol Lett 2010;193:9-18. doi: 10.1016/j.toxlet.2009.11.022 11. Ducolomb Y, Casas E, Valdez A, Gonzalez G, AltamiranoLozano M, Betancourt M. In vitro effect of malathion and diazinon on oocytes fertilization and embryo development in

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

157

porcine. Cell Biol Toxicol 2009;25:623-33. doi: 10.1007/ s10565-008-9117-3 Fattahi E, Parivar K, Jorsaraei SGA, Moghadamnia AA. The effects of diazinon on testosterone, FSH and LH levels and testicular tissue in mice. Iran J Reprod Med 2009;7:56-64.

Carrano AV, Natarajan AT. International Commission for Protection Against Environmental Mutagens and Carcinogens.

ICPEMC publication no. 14. Considerations for population monitoring using cytogenetic techniques. Mutat Res 1988;204:379-406. PMID: 3347212 Fenech M. Cytokinesis-block micronucleus cytome assay.

Nature Protoc 2007;2:1084-104. doi: 10.1038/nprot.2007.77 Eastmond DA, Tucker JD. Identification of aneuploidyinducing agents using cytokinesis-blocked human lymphocytes and an antikinetochore antibody. Environ Mol Mutagen 1989;13:34-43. PMID: 2783409 Durm M, Haar FM, Hausmann M, Ludwig H, Cremer C. Optimization of fast-fluorescence in situ hybridization with repetitive alpha-satellite probes. Z Naturforsch C 1996;51:25361. PMID: 8639232 Singh NP, McCoy MT, Tice RR, Schneider EL. A simple technique for quantitation of low levels of DNA damage in individual cells. Exp Cell Res 1988;175:184-91. doi: 10.1016/0014-4827(88)90265-0 Bajpayee M, Dhawan A, Parmar D, Pandey AK, Mathur N, Seth PK. Gender-related differences in basal DNA damage in lymphocytes of a healthy Indian population using the alkaline Comet assay. Mutat Res 2002;520:83-91. PMID: 12297147 Lopez DE, Carrascal E. Sensitivity of human lymphocyte chromosome to diazinon at different times during cell culture.

Bull Environ Contam Toxicol 1987;38:125-30. doi: 10.1007/ BF01606569 Doherty AT, Ellard S, Parry EM, Parry JM. A study of the aneugenic activity of trichlorfon detected by centromerespecific probes in human lymphoblastoid cell lines. Mutat Res 1996;372:221-31. doi: 10.1016/S0027-5107(96)00142-X Sun FY, Schmid TE, Schmid E, Baumgartner A, Adler ID. Trichlorfon induces spindle disturbances in V79 cells and aneuploidy in male mouse germ cells. Mutagenesis 2000;15:17-24. PMID: 10640526 Padungtod C, Hassold TJ, Millie E, Ryan LM, Savitz DA, Christiani DC, Xu X. Sperm aneuploidy among Chinese pesticide factory workers: scoring by the FISH method. Am J Ind Med 1999;36:230-8. PMID: 10398931 Recio R, Robbins WA, Ocampo-Gomez G, Borja-Aburto V, Morán-Martinez J, Froines JR, Hernandez RMG, Cebrian ME. Organophosphorous pesticide exposure increases the frequency of sperm sex null aneuploidy. Environ Health Persp 2001;109:1237-40. PMCID: PMC1240505 Karczmar A. Anticholinesterases: dramatic aspects of their use and misuse. Neurochem Int 1998;32:401-11. PMID: 9676738 Želježić D, Lucić Vrdoljak A, Radić B, Fuchs N, Berend S, Oreščanin V, Kopjar N. Comparative evaluation of acetylcholinesterase status and genome damage in blood cells of industrial workers exposed to carbofuran. Food Chem Toxicol 2007;45:2488-98. doi: 10.1016/j.fct.2007.05.025 Tsitsimpikou C, Tzatzarakis M, Fragkiadaki P, Kovatsi L, Stivaktakis P, Kalogeraki A, Kouretas D, Tsatsakis AM. Histopathological lesions, oxidative stress and genotoxic effects in liver and kidneys following long term exposure of

158

Gökalp Muranli FD, et al. Genotoxic effects of diazinon on human peripheral blood lymphocytes Arh Hig Rada Toksikol 2015;66:153-158

rabbits to diazinon and propoxur. Toxicology 2013;307:10914. doi: 10.1016/j.tox.2012.11.002 27. Tisch M, Faulde M, Maier H. Genotoxische Effekte aktuell gebräuchlicher Insektizide auf humane Tonsillenschleimhautepithelien [Genotoxic effects of insecticides in current use on mucosal epithelial cells from

human tonsil tissue, in German]. HNO 2007;55(Suppl 1):E1522. doi: 10.1007/s00106-006-1481-9 28. Yassa VF, Girgis SM, Abumourad IMK. Potential protective effects of vitamin E on diazinon-induced DNA damageand some haematological and biochemical alterations in rats. J Mediterr Ecol 2011;11:31-9.

Genotoksično djelovanje diazinona na limfocite ljudske periferne krvi Cilj je ovog istraživanja bio ocijeniti genetička oštećenja u limfocitima ljudske periferne krvi nakon 24-satne odnosno 48-satne izloženosti komercijalnom diazinonu (Basudin 60EM®) u rasponu koncentracija od 0,01 do 40 µg mL-1. U tu smo svrhu rabili mikronukleus (MN)-test, fluorescencijsku in situ hibridizaciju (FISH) i metodu elektroforeze pojedinačnih stanica u agaroznom gelu (tzv. komet-test). Diazinon je značajno povećao učestalost stanica s mikronukleusima u odnosu na kontrolu. Taj je učinak bio još izraženiji nakon 48-satne izloženosti, gdje je značajno povećanje zamijećeno već pri koncentracijama diazinona od 0,01 do 10 µg mL-1. FISH je pokazao aneugeno djelovanje diazinona u koncentranciji od 10 µg mL-1. Naši rezultati potvrdili su ranija saznanja o genotoksičnom i citotoksičnom djelovanju diazonona na ljudske limfocite. Oštećenje DNA koje smo zamijetili u našem istraživanju dovodi u pitanje trenutačne procjene zdravstvenih rizika povezanih s tim pesticidom te poziva na izrazit oprez pri njegovoj primjeni u poljoprivredi i kućanstvima.

KLJUČNE RIJEČI: FISH; genotoksičnost; komet-test; mikronukleus; organofosforni pesticidi

Gromadzka K, et al. The role of wastewater treatment in reducing pollution of surface waters with zearalenone Arh Hig Rada Toksikol 2015;66:159-164

Original article

159

DOI: 10.1515/aiht-2015-66-2606

The role of wastewater treatment in reducing pollution of surface waters with zearalenone Karolina Gromadzka1, Agnieszka Waśkiewicz1, Joanna Świetlik2, Jan Bocianowski3, and Piotr Goliński1 Department of Chemistry, Poznań University of Life Sciences1, Department of Water Treatment Technology, Adam Mickiewicz University2, Department of Mathematical and Statistical Methods, Poznań University of Life Sciences3, Poznań, Poland [Received in January 2015; CrossChecked in January 2015; Accepted in June 2015] Zearalenone (ZEA) is a mycotoxin produced by some Fusarium species in food and feed. The toxicity of ZEA and its metabolites is related to the chemical structure of the mycotoxin, which is similar to naturally occurring oestrogens.

Currently, there is increasing awareness of the presence of fungi and their toxic metabolites in the aquatic environment.

One of the sources of these compounds are the effluents from wastewater treatment plants. The average annual efficiency of zearalenone reduction in the Łęczyca plant in our three-year study was in the range from 51.35 to 69.70 %. The threeway analysis of variance (year, month, and kind of wastewater) shows that the main effects of all factors and all interactions between them were significant for zearalenone and dissolved organic carbon content. Our findings suggest that wastewater is not the main source of surface water pollution with zearalenone. Future research should investigate the means to reduce ZEA and its migration from the fields through prevention strategies such as breeding for crops, plant debris management (crop rotation, tillage), and/or chemical and biological control.

KEY WORDS: aquatic environment; dissolved organic carbon; HPLC; mycotoxins; water quality Since recently, large-scale environmental monitoring of natural toxic compounds has included the products of fungal biosynthesis - the so called mycotoxins. Mycotoxins are produced by three genera: Aspergillus, Penicillium, and Fusarium. The first two contaminate food during drying and storage, whereas certain Fusarium species produce mycotoxins before or immediately after the harvest. The most important sources of mycotoxins for humans are food of plant origin naturally contaminated with these toxins and food of animal origin contaminated with mycotoxin metabolites (1, 2).

Unfortunately, little is known about the distribution of the Fusarium genus and their mycotoxins in the environment.

Recent efforts in the European Union to address this issue have resulted in the inclusion of surface waters in mycotoxin monitoring (3). So far, reports have only included data on the prevalence of zearalenone (ZEA) and deoxynivalenol (DON) in the aquatic environment.

According to Hartmann et al. (3), the presence of mycotoxins in the aquatic environment is the result of runoff from agricultural fields. However, Criado et al. (4) have shown that the fungi Alternaria, Penicillium, and Cladosporium can grow and synthesise mycotoxins in bottled mineral water, which is a serious threat to consumer Correspondence to: Agnieszka Waśkiewicz, Department of Chemistry, Poznań University of Life Sciences, Wojska Polskiego 75, 60-625 Poznań, Poland. E-mail: agat@up.poznan.pl

health. Russell and Paterson (5), in turn, confirmed the ability of Fusarium graminearum to produce ZEA in drinking water.

There are a number of controversies about the entry routes of mycotoxins into the aquatic environment, and therefore it is necessary to conduct additional studies, which will clearly establish how mycotoxins migrate in the environment.

Studies on the prevalence of mycotoxins in the aquatic environment are mainly focused on ZEA due to its strong oestrogenic activity. Several recent publications reported the occurrence of ZEA in surface waters. Its concentrations ranged from below the detection limit to 65.2 ng L-1 (6-8).

Information about mycotoxins in wastewater and their removal efficiency is still scarce. Lagana et al. (9, 10) reported that ZEA concentrations in untreated and treated wastewater reached 18.0 and 10.0 ng L-1, respectively, while Spengler et al. (11) reported as high 36.0 ng L-1 of ZEA in a wastewater treatment plant.

Even though the reported ZEA concentrations in water are not high, their accumulation in water used for food production may present a health risk for humans and animals (5, 7, 8). This risk increases with the presence of other endocrine disruptors such as natural oestrogenic steroids in water (6).

The aim of our study was to investigate the efficiency of ZEA removal during regular wastewater treatment and

160

Gromadzka K, et al. The role of wastewater treatment in reducing pollution of surface waters with zearalenone Arh Hig Rada Toksikol 2015;66:159-164

the correlation between ZEA levels in treated wastewater and the nearby river into which the treated wastewater is released.

noise ratio of 3:1. Method linearity, recovery, and precision have been described earlier (7).

MATERIALS AND METHODS

Dissolved organic carbon (DOC) was analysed with a TOC 1030 total organic carbon analyser (IO Analytical, College Station, TX, USA) using the persulphate wet oxidation method at 100 °C. The amount of carbon dioxide was measured with an IR detector and relayed to a computer.

The method’s detection limit was 0.1 mg L-1.

Reagents Zearalenone (ZEA), acetonitrile, and methanol (HPLC grade) were purchased from Sigma-Aldrich (Steinheim, Germany). ZearalaTest WB immunoaffinity columns were purchased from Vicam (Milford, MA, USA). Water (HPLC grade) was obtained using the Millipore water purification system (Millipore, Bedford, MA, USA).

Sample collection River water and wastewater samples were collected once a month between March 2010 and December 2012.

Wastewater samples were taken from the wastewater treatment plant Łęczyca primary sedimentation basin, receiving 500,000 m3 of wastewater per year from the Komorniki community, and from the treated effluents, which are directly released into the Warta tributary Wirynka.

River water samples were collected from the Warta downstream of the treatment plant. All samples were taken in triplicate. The Łęczyca wastewater treatment plant uses standard Polish treatment infrastructure with mechanicalbiological treatment, precipitation, nitrification, and denitrification.

Zearalenone analysis All water samples were stored in the dark at 4 °C until analysis. ZEA was extracted within 24-36 h in order to keep microbiological degradation to a minimum and to avoid the addition of chemical preservatives. Water samples were filtered and analysed according to the method described by Gromadzka et al. (7, 8). In short, 1000 mL of water was filtered through four filters in a sequence and then purified using ZearalaTest affinity columns. The obtained ZEA was eluted with 3 mL of methanol, the mixture collected in a vial, and the content evaporated to dryness. Followed highperformance liquid chromatography (HPLC), in which the vial residue was dissolved in 250 µL of a water:methanol:acetonitrile mixture in the ratio of 70:20:10, respectively. Then, 20 μL of the solution was injected in a Waters (Waters Corporation, Milford, MA, USA) reversedphase C18 column (150x3.9 mm, 4 mm particle size).

For analysis we used a Waters 2695 HPLC with a Waters 2475 fluorescence detector and a Waters 2996 photodiode detector. Millennium software was used for data processing and calculation (Edison, NJ, USA). The wavelengths of excitation and emission were 274 and 440 nm, respectively.

The limit of detection (LOD) for ZEA was 0.3 ng L-1, which corresponded to the concentration that gave a signal-to-

Analysis of dissolved organic carbon

Statistics We used three-way analysis of variance (ANOVA) to determine the effects of years, months, and kind of wastewater (treated or untreated) as well as of the interactions between years×months, years×kind of wastewater, months×kind of wastewater, and years×months ×kind of wastewater on the concentrations of ZEA and DOC. Variability of ZEA and DOC concentrations was determined with coefficients of variation (CV) (12). The relationships between the concentrations of ZEA and DOC in untreated wastewater, treated wastewater, and river water were estimated using Pearson correlation.

RESULTS AND DISCUSSION The main objective of this study was to determine ZEA levels in the untreated and treated wastewater and to evaluate ZEA removal efficiency.

The results of the three-way ANOVA show significant (P<0.001) effect of all three parameters and their interactions on ZEA and DOC levels (Table 1).

Table 2 shows ZEA and DOC levels in untreated and treated wastewater from 2010 to 2012. ZEA levels in Table 1 Mean squares from three-way analysis of variance (ANOVA) for the observed factors DF

ZEA content

DOC content

Year

2

419.8753*

1372.47*

Month

9

188.9871*

1540.01*

Kind of wastewater

2

172.1069*

31594.72*

Year×Month

18

196.1838*

1786.82*

Year×Kind of wastewater

4

58.1694*

2844.39*

Month×Kind of wastewater

18

87.643*

1133.71*

Year×Month×Kind of wastewater

36

79.4798*

1324.09*

Residual

180

0.8156

23.11

Source of variation

DF-degrees of freedom P<0.001

*

ZEA-zearalenone; DOC-dissolved organic carbon

161

Gromadzka K, et al. The role of wastewater treatment in reducing pollution of surface waters with zearalenone Arh Hig Rada Toksikol 2015;66:159-164

Table 2 Zearalenone and dissolved organic carbon content in untreated and treated wastewater Month (2010)

Untreated wastewater

Treated wastewater

Removal efficiency (%)

ZEA (ng L-1)

DOC (mg L-1)

ZEA (ng L-1)

DOC (mg L-1)

ZEA

DOC

March

18.21

29.87

9.18

14.18

49.59

52.53

April

1.77

90.51

1.77

90.51

0.00

0.00

May

10.80

26.20

0.66

10.66

93.89

59.33

June

6.93

22.09

1.97

16.00

71.59

27.57

July

0.33

62.85

nd

14.58

100.00

76.80

August

19.80

18.83

5.10

10.61

74.24

43.65

September

12.69

20.06

0.96

13.07

92.43

34.85

October

0.69

26.01

0.60

11.90

13.04

54.25

November

0.43

24.53

0.13

10.65

69.77

56.58

December

0.71

28.32

0.27

9.87

62.64

65.15

63.04

136.86

116.40

CV

100.61

Mean value Month (2011)

Untreated wastewater

Treated wastewater

62.72

47.07

Removal efficiency (%)

ZEA (ng L )

DOC (mg L )

ZEA (ng L )

DOC (mg L )

ZEA

DOC

2.85

32.48

2.25

14.86

21.05

54.25

April

1.11

36.47

0.25

13.40

77.39

63.25

May

2.03

85.13

0.68

14.69

66.36

82.74

June

2.02

48.93

nd

13.76

100.00

71.88

July

1.02

27.42

0.15

10.55

85.29

61.53

August

0.43

36.20

0.32

11.93

25.44

67.03

September

6.65

58.82

1.50

11.43

77.46

80.57

October

0.51

26.72

nd

12.84

100.00

51.95

November

0.89

28.35

0.318

12.07

64.47

57.43

December

0.59

23.65

0.121

11.06

79.49

53.23

CV

98.21

44.88

128.98

11.22

69.70

64.39

March

-1

-1

-1

-1

Mean value Month (2012)

Untreated wastewater

Treated wastewater

Removal efficiency (%)

ZEA (ng L )

DOC (mg L )

ZEA (ng L )

DOC (mg L )

ZEA

DOC

March

3.37

43.75

2.61

14.87

22.34

66.01

April

2.29

37.50

1.29

14.23

43.85

62.06

May

2.37

59.78

1.26

14.90

46.84

75.07

June

2.58

205.37

0.36

12.28

86.05

94.02

July

3.42

39.41

2.76

9.81

19.30

75.11

August

2.91

42.83

1.20

12.59

58.76

70.61

September

2.01

37.27

1.02

16.57

49.35

55.54

October

7.44

70.71

1.80

18.45

75.81

73.91

November

2.22

41.02

0.84

17.89

62.16

56.37

December

3.10

38.09

1.58

18.12

49.00

52.43

CV

47.20

79.68

48.39

18.08

-1

-1

Mean value

-1

-1

CV-coefficients of variation; ZEA-zearalenone; DOC-dissolved organic carbon

51.35

68.11

162

Gromadzka K, et al. The role of wastewater treatment in reducing pollution of surface waters with zearalenone Arh Hig Rada Toksikol 2015;66:159-164

a

b

c Figure 1 Zearalenone levels in treated and untreated wastewater and Warta River water in: a) 2010, b) 2011, c) 2012

Gromadzka K, et al. The role of wastewater treatment in reducing pollution of surface waters with zearalenone Arh Hig Rada Toksikol 2015;66:159-164

untreated wastewater peaked in 2010 and dropped in the subsequent years, which may reflect Fusarium contamination of the nearby fields. This contamination may be associated with the prevailing weather conditions of the year before.

Earlier studies (7, 8, 13, 14) have shown that leaching from the fields to the aquatic ecosystem is not instantaneous. The growing season of 2009 was characterised by substantial rainfall from early June to mid-August and by high prevalence of Fusarium head blight (FHB). Precipitation in 2010 was lower and FHB was not observed. In 2011, the temperature was high and there was substantial rainfall in the second half of June, when spring wheat flowers, but relative humidity was lower than in 2009, and FSB was much less prevalent than in 2009 (15).

However, our findings also point to a curious behaviour of ZEA. In 2010, its levels were the highest in the Warta, while in 2011 and 2012 they were the highest in the untreated wastewater (Figure 1). This suggests that the circulation of mycotoxins in the environment is a complex process, and that it is very important to determine the contribution of individual components if we want to reduce contamination of the aquatic ecosystem. What Figure 1 also shows is a strong correlation between ZEA levels in untreated wastewater, treated wastewater, and river water (r=0.7526 for untreated vs. treated wastewater; r=0.5729 for untreated vs. river water; and r=0.3768 for treated vs. river water; P<0.001 for all correlations).

Wastewater treatment in our study proved relatively efficient (Table 2) compared to reports from Italy (9, 10) and Germany (11), although not as efficient as in Switzerland (13). These studies have shown that the efficiency of the toxin’s removal does not depend on the initial ZEA concentration in the wastewater. An interesting alternative would be a combination of traditional wastewater treatment and an integrated photocatalysis-microfiltrationnanofiltration process. Dudziak (16) reported an over 90 % removal efficiency with this process.

Although ZEA was present in the treated water discharged into the river, its levels were a negligible source of surface water contamination (Figure 1). Even so, longterm exposure even to small doses of ZEA may affect animal and human health and its levels in aqueous environment should be monitored.

CONCLUSION Our monitoring study has shown a seasonal pattern in ZEA levels in wastewater and river water, reaching its peak in the autumn. This is probably related to ZEA leaching from crops into the wastewater treatment basin as well as the river. This also explains a strong correlation between ZEA levels in all three waters.

Our findings suggest that ZEA levels in wastewater are not the best indicator of the toxin’s presence in the environment, as wastewater is not the main source of surface

163

water pollution. Future research should investigate the means to reduce ZEA levels and their migration from the fields through prevention strategies such as breeding for crops, plant debris management (crop rotation, tillage), and/ or chemical and biological control.

Acknowledgments This study was partly supported by the Polish Ministry of Science and Higher Education (Project no: NN 305 1655 37). The authors would like to thank Mrs Ewa Rymaniak for technical support during the implementation of the study.

REFERENCES 1. Bennett JW, Klich M. Mycotoxins. Clin Microbiol Rev 2003;16:497-16. doi: 10.1128/CMR.16.3.497-516.2003 2. Goliński P, Waśkiewicz A, Gromadzka K. Mycotoxins and mycotoxicoses under climatic conditions of Poland. Polish J Vet Sci 2009;12:581-8. PMID: 20169938 3. Hartmann N, Erbs M, Wettstein FE, Schwarzenbach RP, Bucheli TD. Quantification of estrogenic mycotoxins at the ng/L level in aqueous environmental samples using deuterated internal standards. J Chromatogr A 2007;1138:13240. doi: 10.1016/j.chroma.2006.10.045 4. Criado MV, Fernández PVE, Badessari A, Cabral D. Conditions that regulate the growth of moulds inoculated into bottled mineral water. Int J Food Microbiol 2005;99:3439. doi: 10.1016/j.ijfoodmicro.2004.10.036 5. Russell R, Paterson M. Zearalenone production and growth in drinking water inoculated with Fusarium graminearum.

Mycol Progress 2007;6:109-13. doi: 10.1007/s11557-0070529-x 6. Bucheli TD, Erbs M, Hartmann N, Vogelgsang S, Wettstein FE, Forrer HR. Estrogenic mycotoxins in the environment.

Mitt Lebensm Hyg 2005;96:386-403.

7. Gromadzka K, Waśkiewicz A, Goliński P, Świetlik J, Bocianowski J. Dissolved organic carbon as an indicator of the presence of zearalenone in the aquatic environment.

World Mycotoxin J 2012;5:357-64. doi: 10.3920/ WMJ2011.1355 8. Gromadzka K, Waśkiewicz A, Golinski P, Świetlik J. Occurrence of estrogenic mycotoxin – zearalenone in aqueous environmental samples with various NOM content.

Water Res 2009;43:1051-9. doi: 10.1016/j.watres.2008.11.042 9. Laganá A, Fago G, Marino A, Santarelli D. Development of an analytical system for the simultaneous determination of anabolic macrocyclic lactones in aquatic environmental samples. Rapid Commun Mass Spectrom 2001;15:304-10. doi: 10.1002/rcm.223 10. Laganá A, Bacaloni A, De Leva I, Faberi A, Fago G, Marino A. Analytical methodologies for determining the occurrence of endocrine disrupting chemicals in sewage treatment plants and natural waters. Anal Chim Acta 2004;501:79-88. doi: 10.1016/j.aca.2003.09.020 11. Spengler P, Körner W, Metzger JW. Substances with estrogenic activity in effluents of sewage treatment plants in southwestern Germany. 1. Chemical analysis. Environ Toxicol Chem 2001;20:2133-41. doi: 10.1002/ etc.5620201001

164

Gromadzka K, et al. The role of wastewater treatment in reducing pollution of surface waters with zearalenone Arh Hig Rada Toksikol 2015;66:159-164

12. Kozak M, Bocianowski J, Rybiński W. Note on the use of coefficient of variation for data from agricultural factorial experiments. Bulg J Agric Sci 2013;19:644-6.

13. Hartmann N, Erbs M, Wettstein FE, Hörger CC, Vogelgsang S, Forrer HR, Schwarzenbach RP, Bucheli TD. Environmental exposure to estrogenic and other myco- and phytotoxins.

Chimia 2008;62:364-7. doi: 10.2533/chimia.2008.364 14. Waśkiewicz A, Gromadzka K, Bocianowski J, Pluta P, Goliński P. Zearalenone contamination of the aquatic environment as a result of its presence in crops. Arh Hig Rada Toksikol 2012;63:429-35. doi: 10.2478/10004-125463-2012-2229 15. Góral T, Ochodzki P, Walentyn-Góral D, Nielsen LK, Justesen AF, Jorgensen LN. Wplyw przedplonu oraz

warunków pogodowych na porazenie klosów pszenicy jarej przez grzyby z rodzaju Fusarium oraz zawartosc mikotoksyn w ziarnien [Effect of pre-crop and weather conditions on infection of heads of spring wheat with Fusarium fungi and content of mycotoxins in grain, in Polish]. Biuletyn Instytutu Hodowli i Aklimatyzacji Roślin 2012;265:11-21.

16. Dudziak M. Usuwanie mykoestrogenów z roztworów wodnych w zintegrowanym procesie fotokatalizamikrofiltracja-nanofiltracja [Removal of mycoestrogens from aqueous solutions in the integrated photocatalysismicrofiltration-nanofiltration process, in Polish]. Ochrona Środowiska 2012;34:29-32.

Uloga pročišćavanja otpadnih voda u smanjenju onečišćenja površinskih voda zearalenonom Zearalenon (ZEA) je mikotoksin koji u hrani proizvode neke vrste gljivica roda Fusarium. Njegova toksičnost i toksičnost njegovih metabolita ovisi o kemijskoj strukturi mikotoksina, a djelovanje mu je slično onome prirodnoga estrogena. Sve smo svjesniji važnosti gljivica i njihovih toksičnih metabolita u vodenom okolišu. Jedan od izvora spoja u površinskim vodama jesu i otpadne vode. Naše je trogodišnje praćenje pokazalo da se uspješnost pročišćenja zearalenona iz otpadnih voda kreće u rasponu od 51,35 do 69,70 % na godišnjoj razini. Trostrana analiza varijance (godina, mjesec, vrsta otpadne vode - nepročišćena/pročišćena) upućuje na to da je djelovanje svih čimbenika i svih njihovih međusobnih interakcija značajno utjecalo na razine zearalenona i otopljenog organskog ugljika.

Istraživanje je pokazalo da otpadne vode nisu glavni izvor onečišćenja površinskih voda zearalenonom. Buduća bi istraživanja trebala utvrditi preventivne strategije uzgoja, upravljanja ostacima biljke (rotacijom, obradom zemljišta), odnosno tretiranje kemijskim i biološkim sredstvima kojima bi se smanjile razine zearalenona i njegova migracija s polja u vodeni okoliš.

KLJUČNE RIJEČI: HPLC; kakvoća vode; mikotoksini; otopljeni organski ugljik; vodeni okoliš

Miličević A, Raos N. Modelling of copper(II) binding to pentapeptides related to atrial natriuretic factor using the 3χv connectivity index Arh Hig Rada Toksikol 2015;66:165-170

Original article

165

DOI: 10.1515/aiht-2015-66-2631

Modelling of copper(II) binding to pentapeptides related to atrial natriuretic factor using the 3χv connectivity index Ante Miličević and Nenad Raos Institute for Medical Research and Occupational Health, Zagreb, Croatia [Received in February 2015; CrossChecked in February 2015; Accepted in May 2015] Using molecular graph theory we studied the binding of NSFRY-NH2 and 12 related pentapeptide amides to Cu(II) as a model system for atrial natriuretic factor (ANF) peptide interactions with copper. Linear regression models based on the valence connectivity index of the 3rd order (3χv) reproduced experimental stability constants (log β) for 1N, 2N, 3N, and 4N coordinated complexes with the standard error of 0.30-0.39 log β units. We developed separate models for seven tyrosinic (N=28) and five non-tyrosinic peptides (N=20), and a common model for both kinds of peptides (N=48) with an indicator (dummy) variable. The results indicate additional aromatic stabilisation in 4N complexes due to metal cation-π interactions with tyrosine but not with the phenylalanine residue. We have also amended the log K and log K* values to correct miscalculations published by Janicka-Klos et al. in 2013.

KEY WORDS: coordination compounds; peptides; stability constants; topological indices Atrial natriuretic factor (ANF) is a peptide hormone secreted by the heart to control extracellular fluid volume and blood pressure by maintaining water and salt balance (1, 2). All ANF peptides have the same C-terminal sequence, NSFRY, and the same 17-residue disulphide-bonded core, which is essential for their function (1). As copper(II) deficiency leads to heart hypertrophy (3) and overproduction of ANF in the left ventricle (4), copper(II) complexes with ANF may have a protective role in heart diseases.

Using a variety of methods, Janicka-Klos et al. (5) systematically analysed Cu(II) binding to pentapeptides mimicking the N-terminal sequence of ANF (5). Beside NSFRY-NH2, the peptide with the same sequence as the terminal part of ANF, they also investigated its analogues in order to see how residues of the model peptide affected Cu(II) binding. The most interesting question, however, is how copper interacts with aromatic residues, phenylalanine and tyrosine in particular. We know that cation-π interactions enhance the stability of coordination compounds such as Cu(II) chelates with amino acids (6, 7), but interactions with aromatic side chains may also include π-π stacking as well as other hydrophobic interactions (8-13).

The aim of our study was to further elucidate the behaviour of these pentapeptides. To do that, we had earlier developed a simple and efficient method to predict stability constants using regression models based on graph theory (14-25). We had also applied graph theory models on Correspondence to: Ante Miličević, Institute for Medical Research and Occupational Health, Ksaverska c. 2, POB 291, HR-10001 Zagreb, Croatia.

Tel.:+385 14682524; fax: +385 14673303. E-mail: antem@imi.hr

oligopeptides (n=2-5), namely on their Cu(II) and Ni(II) complexes with copolymers of glycine, aliphatic (Ala, Val, Leu, norVal, norLeu), and aromatic (Phe, Tyr) amino acid (26, 27), as well as Cu(II) complexes with peptides containing a cysteinic disulphide bridge (28).

These models proved suitable for peptides, giving the standard error of prediction of 0.2-0.3 log K units (26, 27).

METHODS Calculation of topological indices We calculated the topological indices using the E-DRAGON online system developed by R. Todeschini et al. (29, 30), capable of yielding 119 topological indices in a single run along with many other molecular descriptors.

Connectivity matrices were constructed with the aid of Online SMILES Translator and Structure File Generator (31).

All of the models were developed using the 3χv index (the valence molecular connectivity index of the 3rd order), which was defined as (32-35): 3 v χ = Σ [δ(i) δ(j) δ(k) δ(l)]–0.5 [1] path

where δ(i), δ(j), δ(k), and δ(l) are weights (valence values) of vertices (atoms) i, j, k, and l making a path of length 3 (three consecutive chemical bonds) in a vertexweighted molecular graph. The valence value, d(i), of vertex i is defined as: d(i) = [Zv(i) – H(i)]/[Z(i) - Zv(i) - 1] [2]

166

Miličević A, Raos N. Modelling of copper(II) binding to pentapeptides related to atrial natriuretic factor using the 3χv connectivity index Arh Hig Rada Toksikol 2015;66:165-170

HN O

H2 N

H2 O H2 O

NH2

HN

NH

NH H2 N H2 O

OH

H2 N

O

Cu

O

H N

O

NH2

NH2

N H

Cu

O

O

H N

HO

N H

H N

NH2

O

N H

O

[CuLH]

O

N

H2 O

H N

O

NH2

O

OH

2+

O

[CuL]

OH

+

OH HN NH

H2 N

NH2

O

N H

N H HN

O

HN

NH2

O NH2

O

O

O

O

HO N HO

N

Cu

H2 N N

NH

OH2

H2 N

Cu

OH

N

O

NH2

N

O

O

O O NH2

[CuLH-1]

[CuLH-2]

-

Figure 1 Tetracoordinated Cu(II) complexes with NSFRY-NH2 in its protonated and deprotonated forms

where Z (i) is the number of valence electrons belonging to the atom corresponding to vertex i, Z(i) is its atomic number, and H(i) is the number of hydrogen atoms attached to it.

The 3χv index was calculated for all complexes from graph representations of aqua-complexes under the assumption that Cu(II) is tetracoordinated (Figure 1). v

Regression calculations Regression calculations, including the leave-one-out procedure (LOO) of cross validation (cv), were done using the CROMRsel program (36). The standard error of the cross validation estimate is defined as:

corrections (marked in bold), and reports log β and 3χv values.

We plotted the dependence of log β on 3χv (Figure 2) and found that the initial set of peptide complexes could be divided in two subsets; the subset with tyrosine residue (Subset 1, N=32) and the subset without it (Subset 2, N=20).

The former subset is much more stable than the latter.

Figure 2 shows that the dependence of log β on 3χv is linear for the complexes of each peptide, (CuP)n: log β = a1[3χv(CuP)n] + b [4] where n is the charge of a complex (from +2 to -1).

20

∑ i

∆X i [3] N

where ΔX and N denote cv residuals and the number of reference points, respectively.

15

[CuLH]2+

10

[CuL]+

5

log β

S.E.cv =

2

0

[CuLH-1]

[CuL]2+ +

[CuLH-1]

[CuLH-2]-

-5

RESULTS AND DISCUSSION In order to model the stability constant (β) of Cu(II) complexes with the NSFRY-NH2 peptide and its substitute analogues using 3χv, we relied on experimental stability values reported recently by Janicka-Klos et al. (5). However, some of the values for log K and log K* were calculated erroneously. Table 1 amends the original values with our

[CuLH-2]

-10 -15 -20

[CuLH-3]5

6

7

8

9

10

11

12

13

3

χv

Figure 2 Dependence of experimental log β stability constants on 3 v χ index for Cu(II) complexes with NSFRY-NH2 and its analogues.

Each line represents linear dependence of log β on 3χv for complexes of a single peptide (CuP)n

Miličević A, Raos N. Modelling of copper(II) binding to pentapeptides related to atrial natriuretic factor using the 3χv connectivity index Arh Hig Rada Toksikol 2015;66:165-170

167

Table 1 Experimental log β and amended values for log K and log K* (bolded) of Cu(II) complexes with NSFRY-NH2 and its analogues from Janicka-Klos et al. (5) and their 3χv index Coordination 3 v Peptide Species log β log K log K* χ mode -1.89 NSFRY-NH2 (1) [CuLH]2+ 1N 14.13 4.98 8.901 -6.87 [CuL]+ 2N 9.15 6.44 9.773 -13.31 [CuLH-1] 3N 2.71 6.73 11.218 -20.04 [CuLH-2]4N -4.02 9.48 12.596 -3.25 ASFRY-NH2 (2) [CuLH]2+ 1N 14.05 4.90 8.678 -8.15 [CuL]+ 2N 9.15 7.29 9.545 -15.44 [CuLH-1] 3N 1.86 5.97 10.991 -21.41 [CuLH-2]4N -4.11 10.49 12.369 [CuLH]2+ 1N 15.88 5.16 -1.49 8.678 (R)-ASFRY-NH2 (3) [CuL]+ 2N 10.72 5.64 -6.65 9.545 [CuLH-1] 3N 5.08 6.80 -12.29 10.991 [CuLH-2]4N -1.72 6.43 -19.09 12.369 -2.31 NAFRY-NH2 (4) [CuLH]2+ 1N 13.8 5.17 8.837 -7.48 [CuL]+ 2N 8.63 6.84 9.809 -14.32 [CuLH-1] 3N 1.79 6.35 11.25 -20.67 [CuLH-2]4N -4.56 10.26 12.627 4.59 NShFRY-NH2 (5) [CuLH]2+ 1N 13.45 -2.6 9.182 7.26 [CuL]+ 2N 8.86 -7.19 10.055 6.28 [CuLH-1] 3N 1.60 -14.45 11.496 10.33 [CuLH-2]4N -4.68 -20.73 12.873 -1.92 NSARY-NH2 (6) [CuLH]2+ 1N 13.98 4.90 7.737 -6.82 [CuL]+ 2N 9.08 6.96 8.609 -13.78 [CuLH-1] 3N 2.12 6.63 10.155 -20.41 [CuLH-2]4N -4.51 10.54 11.528 -2.10 NSFAY-NH2 (7) [CuLH]2+ 1N 14.22 4.91 8.035 -7.01 [CuL]+ 2N 9.31 7.07 8.907 -14.08 [CuLH-1] 3N 2.24 6.57 10.352 -20.65 [CuLH-2]4N -4.33 10.69 11.835 -2.57 NSFRA-NH2 (8) [CuL]2+ 1N 3.74 4.72 7.65 -7.29 [CuLH-1]+ 2N -0.98 7.08 8.522 -14.37 [CuLH-2] 3N -8.06 7.14 9.968 -21.51 [CuLH-3]4N -15.2 11.346 NSAAA-NH2 (9) [CuL]2+ 1N 3.87 4.33 -2.73 5.62 [CuLH-1]+ 2N -0.46 7.11 -7.06 6.493 [CuLH-2] 3N -7.57 7.70 -14.17 8.039 [CuLH-3]4N -15.27 8.63 -21.87 9.517 4.54 NSFAA-NH2 (10) [CuL]2+ 1N 3.90 -2.63 6.784 [CuLH-1]+ 2N -0.64 7.08 -7.17 7.656 [CuLH-2] 3N -7.72 7.34 -14.25 9.102 [CuLH-3]4N -15.06 8.64 -21.59 10.584 NSARA-NH2 (11) [CuL]2+ 1N 3.93 4.58 -2.57 6.486 [CuLH-1]+ 2N -0.65 7.2 -7.15 7.359 [CuLH-2] 3N -7.85 7.33 -14.35 8.904 [CuLH-3]4N -15.18 9.02 -21.68 10.278 NSAAY-NH2 (12) [CuLH]2+ 1N 13.80 4.8 -2.41 6.871 [CuL]+ 2N 9.00 7.27 -7.21 7.743 [CuLH-1] 3N 1.73 6.72 -14.48 9.289 [CuLH-2]4N -4.99 10.25 -21.2 10.767 AAAAA-NH2 (13) [CuL]2+ 1N 4.56 5.13 -3.28 5.334 [CuLH-1]+ 2N -0.57 7.50 -8.41 6.301 [CuLH-2] 3N -8.07 8.33 -15.91 7.842 [CuLH-3]4N -16.4 -24.24 9.32

168

Miličević A, Raos N. Modelling of copper(II) binding to pentapeptides related to atrial natriuretic factor using the 3χv connectivity index Arh Hig Rada Toksikol 2015;66:165-170

Table 2 Regression models for the estimation of log β for Cu(II) complexes with NSFRY-NH2 and its analogues Regression coefficients Eq. N Peptides r S.E. a1 (S.E.) a2 (S.E.) a3 (S.E.) b (S.E.) 5 28 with Tyr -4.849(49) 4.75(10) 14.47(78) 0.999 0.35 5

20

without Tyr

-5.030(41)

4.728(83)

6

48

all

-4.926(36)

4.733(75)

6

36

without CuP-

-4.979(54)

4.780(85)

15 10

log β (exp)

5 0 -5 -10 -15 -15

-10

-5

0

5

10

15

S.E.cv 0.39

5.90(48)

0.999

0.25

0.30

10.35(17)

5.02(44)

0.999

0.35

0.38

10.14(17)

5.22(44)

0.999

0.30

0.34

The model yielded S.E.cv=0.39 and S.E.cv=0.30 for Subset 1 and 2, respectively (Table 2). From Subset 1 we omitted complexes with the (R)-alanine residue [(R)ASFRY-NH2] because it is the only residue with unnatural configuration and the model gave much worse results with those complexes: r=0.993 and S.E.cv=0.89 (N=32).

As there was no significant difference between regression parameters for the two subsets (Table 2) except in their intercepts (b), we developed a model that included both subsets (N=48) by introducing an indicator (dummy) variable (In=1 and In=0, for Subsets 1 and 2, respectively) to Equation 5:

log β (calc)

Figure 3 Experimental vs. calculated values of log β for complexes of Cu(II) complexes with NSFRY-NH2 and its analogues (N=48, Table 2) using Equation 6

Figure 4 Residual plot of log β values of CuP- (test set, N=12) calculated from the regression model parameterised on all other complexes (training set, N=36; Table 2) using Equation 6. F, hF, and Y denote complexes of peptides with respective residues

As the slopes for complexes with different peptides are almost the same – from -4.64 to -4.92 and from -4.88 to -5.21, for Subset 1 and 2, respectively – we used a single model to estimate all log βs in a subset: log β = a1[3χv(CuP)n] + a2[3χv(CuP)2+] + b

[5]

The variable 3χv(CuP)2+, i.e. the 3χv of a peptide complex in its fully protonated form, was added to Equation 4 because it determines the distances between the regression lines for the complexes of each peptide (Figure 2).

log β = a1[3χv(CuP)n] + a2[3χv(CuP)2+] + a3In + b

[6]

The model yielded r=0.999 and S.E.cv=0.38 (Table 2, Figure 3). This is worse than the previous model for Subset 2, but slightly better than for Subset 1.

In order to investigate additional stabilisation of CuP(4N) complexes with tyrosine and phenylalanine residue (Y and F) caused by metal-π interactions (5), we calculated log β values of CuP- (test set, N=12) from the model parameterised on all other complexes (training set, N=36) using Equation 6 (Table 2). Differences between experimental and calculated log β values show that residuals for the CuP- complexes with tyrosine peptides differ significantly from others (Figure 4), as all calculated values for log β of complexes belonging to Subset 1 are lower than the experimental ones, whereas values calculated for Subset 2 are higher than the experimental, except for NSAAA-NH2.

Furthermore, the S.E. of the predicted log β for tyrosine peptide complexes was twice as high as the S.E. for other complexes (0.59 vs. 0.28).

As our model cannot explicitly calculate electronic interactions, systematically lower and poorer (S.E.=0.59) predicted stabilities for complexes that are capable of aromatic stabilisation may also evidence additional stabilisation of CuP - complexes. In addition, of all 4N-complexes with tyrosine peptides, error was the smallest for NShFRY-NH2 (hFY, Figure 4), which is in accordance with the report by Janicka-Klos et al. (5), suggesting that π-interactions with Cu(II) are less favourable for stability due to the elongation of phenylalanine side chain. However, peptides with phenylalanine residue (F, Figure 4) do not differ from other non-tyrosinic peptides, indicating that phenylalanine residue does not participate in cation-π interactions, but seems to enhance tyrosine interactions.

Miličević A, Raos N. Modelling of copper(II) binding to pentapeptides related to atrial natriuretic factor using the 3χv connectivity index Arh Hig Rada Toksikol 2015;66:165-170

Finally, we would like to say a few words about the possible influence of Cu(II) on ANF function. Even though NSFRY-NH2 (1) showed very high affinity to Cu(II), it did not have the highest stability constants among the modelled pentapeptides (Table 1). It is therefore possible that Cu(II) binding to the C-terminal sequence of ANF is influenced by the neighbouring disulphide loop and we believe that the graph theory method may help to prove or disprove this hypothesis.

Acknowledgement This study was supported by the Croatian Ministry of Science, Technology, Education and Sports (grant no. 0221770495-2901).

REFERENCES 1. Flynn TG, Davies PL. The biochemistry and molecular biology of atrial natriuretic factor. Biochem J 1985;232:31321. PMCID: PMC1152881 2. Rosenzweig A, Seidman CE. Atrial natriuretic factor and related peptide hormones. Annu Rev Biochem 1991;60:22955. doi: 10.1146/annurev.bi.60.070191.001305 3. Prohaska JR. Biochemical changes in copper deficiency. J Nutr Biochem 1990;1:452-61. doi: 10.1016/09552863(90)90080-5 4. Kang YJ, Zhou ZX, Wu H, Wang GW, Saari JT, Klein JB. Metallothionein inhibits myocardial apoptosis in copperdeficient mice: role of atrial natriuretic peptide. Lab Invest 2000;80:745-57. PMID: 10830785 5. Janicka-Klos A, Porciatti E, Valensin D, Conato C, Remelli M, Oldziej S, Valensin G, Kozlowski H. The unusual stabilization of the Ni2+ and Cu2+ complexes with NSFRY. Dalton Trans 2013;42:448-58. doi: 10.1039/c2dt31959d 6. Odani A, Yamauchi O. Preferential formation of ternary copper(II) complexes involving substituted ethylenediamines and amino acids with an aromatic side chain. Inorg Chim Acta 1984;93:13-8.

7. Yamauchi O, Odani A, Takani M. Metal-amino acid chemistry. Weak interactions and related functions of side chain groups. J Chem Soc, Dalton Trans 2002;3411-21. doi: 10.1039/B202385G 8. Sugimori T, Masuda H, Ohata N, Koiwai K, Odani A, Yamauchi O. Structural dependence of aromatic ring stacking and related weak interactions in ternary amino acidcopper(II) complexes and its biological implication. Inorg Chem 1997;36:576-83. doi: 10.1021/ic9608556 9. Siegel H, Tribolet R, Scheller KH. Solvent effects on intramolecular hydrophobic ligand – ligand interactions in binary and ternary complexes. Inorg Chim Acta 1985;100:15164. doi: 10.1016/S0020-1693(00)88303-6 10. Aoki K, Yamazaki H. A model for coenzyme-metal ionapoenzyme interactions: crystal structure of the ternary complex [(thiamine pyrophosphate)(1,10-phenanthroline) aquacopper]-dinitrate-water. J Am Chem Soc 1980;102:687880. doi: 10.1021/ja00542a051 11. Yamauchi O, Odani A. Structure-stability relationship in ternary copper(II) complexes involving aromatic amines and tyrosine or related amino acids. Intramolecular aromatic ring

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

169

stacking and its regulation through tyrosine phosphorylation.

J Am Chem Soc 1985;107:5938-45. doi: 10.1021/ ja00307a019 Sigel H. Intramolecular equilibria in metal ion complexes of artificial nucleotide analogues with antiviral properties. A case study. Coord Chem Rev 1995;144:287-319. doi: 10.1016/0010-8545(95)01158-L Yajima T, Takamido R, Shimazaki Y, Odani A, Nakabayashi Y, Yamauchi O. π–π Stacking assisted binding of aromatic amino acids by copper(II) – aromatic diimine complexes.

Effects of ring substituents on ternary complex stability.

Dalton Trans 2007;21:299-307. doi: 10.1039/B612394E Rouvray DE. The modeling of chemical phenomena using topological indices. J Comput Chem 1987;8:470-80. doi: 10.1002/jcc.540080427 Trinajstić N. Chemical Graph Theory. 2nd ed. Boca Raton (FL): CRC Press; 1992.

Todeschini R, Consonni V. Handbook of Molecular Descriptors. Weinheim: Wiley-VCH; 2000.

Gutman I. Degree-based topological indices. Croat Chem Acta 2013;86:351-61. doi: 10.5562/cca2294 Raos N, Miličević A. Estimation of stability constants of coordination compounds using models based on topological indices. Arh Hig Rada Toksikol 2009;60:123-8. doi: 10.2478/10004-1254-60-2009-1923 Miličević A, Raos N. Estimation of stability of coordination compounds by using topological indices. Polyhedron 2006;25:2800-8. doi: 10.1016/j.poly.2006.04.012 Miličević A, Raos N. Influence of chelate ring interactions on copper(II) chelate stability studied by connectivity index functions. J Phys Chem A 2008;112:7745-9. doi: 10.1021/ jp802018m Miličević A, Raos N. A model to estimate stability constants of amino acid chelates with Cu(II) and Ni(II) at different ionic strengths. J Mol Liq 2012;165:139-42. doi: 0.1016/j. molliq.2011.11.001 Miličević A, Raos N. Estimation of stability constants of mixed copper(II) chelates using valence connectivity index of the 3rd order derived from two molecular graph representations. Acta Chim Slov 2009;56:373-8.

Miličević A, Raos N. Empirical model for the stability constants of acetate mono-complexes with La3+, Nd3+, Gd3+, and Yb3+ at different temperatures and ionic strengths. J Mol Liq 2013;177:60-2. doi: 10.1016/j.molliq.2012.09.003 Miličević A, Raos N. Stability prediction of Cu2+, Ni2+ and Zn2+ N-salicylidene-aminoacidato complexes by models based on connectivity index 3χv. Cent Eur J Chem 2014;12:749. doi: 10.2478/s11532-013-0345-x Miličević A, Raos N. Comparison of two methods for the estimation of stability of copper(II) bis-complexes with aromatic ligands relevant to Alzheimer’s disease. Arh Hig Rada Toksikol 2013;64:539-45. doi: 10.2478/10004-125464-2013-2418 Miličević A, Raos N. Estimation of stability constants of copper(II) and nickel(II) chelates with dipeptides by using topological indices. Polyhedron 2008;27:887-92. doi: 10.1016/j.poly.2007.11.017 Miličević A, Raos N. Estimation of stability constants with connectivity index: development of bivariate and multivariate linear models for copper(II) chelates with oligopeptides.

Croat Chem Acta 2009;82:633-9.

170

Miličević A, Raos N. Modelling of copper(II) binding to pentapeptides related to atrial natriuretic factor using the 3χv connectivity index Arh Hig Rada Toksikol 2015;66:165-170

28. Miličević A, Raos N. Stability prediction of copper(II) complexes with peptides containing cysteinic disulfide bridge by models based on the connectivity index 3χv. J Coord Chem 2014;67:623-9. doi: 10.1080/00958972.2014.888716 29. Tetko IV, Gasteiger J, Todeschini R, Mauri A, Livingstone D, Ertl P, Palyulin VA, Radchenko EV, Zefirov NS, Makarenko AS, Tanchuk VY, Prokopenko VV. Virtual computational chemistry laboratory-design and description.

J Comput Aid Mol Des 2005;19:453-63. doi: 10.1007/ s10822-005-8694-y 30. VCCLAB, Virtual Computational Chemistry Laboratory [displayed 23 April 2015]. Available at http://www.vcclab. org 31. National Cancer Institute. Online SMILES Translator and Structure File Generator [displayed 23 April 2015]. Available at http://cactus.nci.nih.gov/services/translate/

32. Kier LB, Hall LH. Molecular connectivity VII: Specific treatment to heteroatoms. J Pharm Sci 1976;65:1806-9.

PMID: 1032667 33. Kier LB, Hall LH. Molecular Connectivity in Chemistry and Drug Research. New York: Academic Press; 1976.

34. Kier LB, Hall LH. Molecular Connectivity in StructureActivity Analysis. New York: Willey; 1986.

35. Randić M. On history of the Randic index and emerging hostility toward chemical graph theory. MATCH Commun Math Comput Chem 2008;59:5-124.

36. Lučić B, Trinajstić N. Multivariate regression outperforms several robust architectures of neural networks in QSAR modeling. J Chem Inf Comput Sci 1999;39:121-32. doi: 10.1021/ci980090f

Modeliranje vezivanja bakra(II) za pentapeptide povezane s atrijalnim natriuretičkim faktorom pomoću indeksa povezanosti 3χv Služeći se teorijom molekularnih grafova istraživali smo vezivanje NSFRY-NH2 i 12 srodnih pentapetidnih amida za Cu(II) kao modelnog sustava za interakciju bakra s peptidnom molekulom atrijalnog natriuretičkog faktora (ANF). Modeli linarne regresije temeljeni na valencijskom indeksu povezanosti trećega reda (3χv) reproducirali su eksperimentalnu konstantu stabilnosti (log β) za komplekse koordinacije 1N, 2N, 3N i 4N sa standardnom pogreškom u rasponu od 0,30 do 0,39 log β jedinica. Razvili smo odvojene modele za sedam tirozinskih (N=28) i pet netirozinskih (N=20) peptida te skupni model s indikatorskom varijablom za obje vrste peptida (N=48). Rezultati upućuju na dodatnu aromatsku stabilizaciju u kompleksima vrste 4N zbog interakcija kationa s π-orbitalama tirozinskog ostatka, ali ne i fenilalaninskoga. Ispravili smo i pogrešne vrijednosti log K i log K* nastale omaškom u radu Anne Janicka-Klos i sur. 2013.

KLJUČNE RIJEČI: kompleksni spojevi; peptidi; konstante stabilnosti; topološki indeksi

Davila S, et al. Real-time dissemination of air quality information using data streams and Web technologies Arh Hig Rada Toksikol 2015;66:171-180

Original article

171

DOI: 10.1515/aiht-2015-66-2633

Real-time dissemination of air quality information using data streams and Web technologies: linking air quality to health risks in urban areas Silvije Davila1, Jadranka Pečar Ilić2, and Ivan Bešlić1 Institute for Medical Research and Occupational Health1, Ruđer Bošković Institute2, Zagreb, Croatia [Received in March 2015; CrossChecked in March 2015; Accepted in June 2015] This article presents a new, original application of modern information and communication technology to provide effective real-time dissemination of air quality information and related health risks to the general public. Our on-line subsystem for urban real-time air quality monitoring is a crucial component of a more comprehensive integrated information system, which has been developed by the Institute for Medical Research and Occupational Health. It relies on a StreamInsight data stream management system and service-oriented architecture to process data streamed from seven monitoring stations across Zagreb. Parameters that are monitored include gases (NO, NO2, CO, O3, H2S, SO2, benzene, NH3), particulate matter (PM10 and PM2.5), and meteorological data (wind speed and direction, temperature and pressure). Streamed data are processed in real-time using complex continuous queries. They first go through automated validation, then hourly air quality index is calculated for every station, and a report sent to the Croatian Environment Agency. If the parameter values exceed the corresponding regulation limits for three consecutive hours, the web service generates an alert for population groups at risk. Coupled with the Common Air Quality Index model, our web application brings air pollution information closer to the general population and raises awareness about environmental and health issues. Soon we intend to expand the service to a mobile application that is being developed.

KEY WORDS: air pollution; air quality index; monitoring; integrated information system; public health; web services Urban air pollution is increasing in major world cities where there are continuous or large emissions of air pollutants. Currently, the most polluted cities are in developing countries such as China, India, and Pakistan with high population density, recent economic boost, and increased consumption (1, 2). The main sources of urban air pollution are transportation, commerce, and industry releasing criteria pollutants (CO, SO2, NOx, PM10, PM2.5, Pb, and O3) and hazardous air pollutants (heavy metals, volatile organic compounds, etc.). Exposure to air pollutants, both short and long-term, has been associated with health effects in particularly vulnerable population groups such as children, the elderly or outdoor workers.

High pollutant levels increase the risk of respiratory infections, heart disease, and even stroke and lung cancer (1, 3, 4).

In Croatia, local air quality monitoring networks have a 50-year-long tradition of basic pollutant monitoring and are organised across the country. Air quality data are delivered to the Croatian Environment Agency (CEA), which collects data and enters them into the Air Quality Information System (AQIS) as an integral part of the National Environmental Information System (5).

Correspondence to: Silvije Davila, Institute for Medical Research and Occupational Health, Ksaverska cesta 2, 10000 Zagreb, Croatia, E-mail: sdavila@imi.hr

The Institute for Medical Research and Occupational Health (IMROH) is one of the collaborating institutions responsible for the local network of the City of Zagreb, which consists of seven monitoring stations. IMROH has a long tradition in multi-disciplinary research of environmental effects on human health as well as in providing reliable environmental monitoring services (including urban air quality) as an accredited national reference laboratory for particulate matter (6).

Effective urban air quality management requires an integrated approach, in line with the best practice and standards of environmental management and environmental informatics (7). In this sense, information and communication technology (ICT) is crucial for an effective air quality management information system (AQMIS), in line with the recommendations of the European Environment Agency (EEA), principles of the European Sharing Environmental Information System (SEIS) (8), and standards for geospatial information technology (EU INSPIRE Directive 2007/2/ EC) (9).

In order to fulfil its roles at the national and local level, IMROH has developed Web applications that transfer air quality data to CEA’s AQIS. Furthermore, IMROH has been working on a more comprehensive integrated information system and geo-portal with functionalities such as (10, 11): • on-line subsystem for urban air quality monitoring in real time (data streams from automatic stations);

172

Davila S, et al. Real-time dissemination of air quality information using data streams and Web technologies Arh Hig Rada Toksikol 2015;66:171-180

•

assessment of air quality (based on air quality index); • prediction of air pollution and back-trajectory analyses of particulate matter (based on dispersion and statistical models); • geographic information system (GIS) to integrate data and models and to estimate the contribution of pollution; • dynamic spatial-temporal reporting and informing the public; • decision-making support for authorised users.

This article presents our newly developed on-line subsystem for urban air quality monitoring and assessment as a crucial component of the integrated information system.

Its architecture is based on Web services technology, which has been applied in many fields, but complex streaming and querying design and application in real-time air quality information reporting based on air quality indices are new.

Data stream management systems vs. traditional Database management systems Traditional database management systems (DBMS) are designed to process datasets that do not change continuously with large data inputs. In addition, today’s system databases are ill-equipped for the execution of any specialised data storage and management or data stream searches. Early data stream applications and systems completely ignored DBMSs or used them for offline data warehousing, but today they use queries that do not differ much from traditional database queries (12). There are two types of data stream queries: classic and continuous. Classic queries execute a dataset and return a response. Continuous queries execute data as they arrive and either save the response or convert it into a data stream (13). Queries can also be predefined or ad hoc. Predefined queries precede data input and are generally continuous, although in some situations they can be classic. Ad hoc queries are made online after

the data have already started streaming. They can be classic or continuous (14-16).

Data streams differ from the conventional relational databases in as much as data are fed online; the system does not control the order in which data are fed or processed; data streams are not limited by data size; and once the data have been processed by a query, they can either be rejected or stored. Rejected data cannot be recovered.

Data stream management systems in air quality monitoring Since it can handle large amounts of constantly changing data, data streaming seems to better fit the requirements of air quality monitoring than a DBMS, yet to the best of our knowledge none of the institutions monitoring air quality in Europe and the world use it. One of the reasons could be that data streams require complex programming, which has so far been mainly in the domain of open-source and limited to scientific research of stream semantics and data flows, stream processing languages, models, and architectures for data stream management databases (13-17). The other reason may be that most air quality monitoring applications are not fit for data transfer from a terminal or a device to the server/information system.

There are a number of stream databases and data stream management systems that enable real-time processing of multiple continuous queries over data streams, such as STREAM, AURORA, TelegraphCQ, StreamGlobe, StreamInsight, and InfoSphere Streams (16-18).

We opted for Microsoft StreamInsight (19, 20) as one of the most stable systems with strong customer support that can handle our SQL Server database. StreamInsight uses the programming language C# and LanguageIntegrated Query (LINQ). C# is an object-oriented language developed by Microsoft and is fully integrated into the .

NET Framework. It has become popular among programmers and the scientific community because of support and

Figure 1 Example of basic components in StreamInsight package according to (30)

Davila S, et al. Real-time dissemination of air quality information using data streams and Web technologies Arh Hig Rada Toksikol 2015;66:171-180

173

Figure 2 Automatic measuring stations of Zagreb City

development potential in terms of more advanced programming. LINQ is a set of features that adds querying capacities to C#. Its greatest advantage is the flexibility of use in any kind of database and form of data storage.

StreamInsight consists of three key components (Figure 1). The first is input adapter, which defines and decides where and at which rate data are collected (starting from 1 ms on). The second component is reserved for queries.

The amount, type and course of executing queries solely depend on what we want to do with the flow of data collected on a data sheet. The third component is the output adapter, responsible for the way in which data will be stored, sent, or displayed.

Real-time air quality monitoring in the city of Zagreb In order to enable real-time air quality monitoring in the City of Zagreb we combine several types of automatic monitoring stations to cover the whole city. For now, this includes three automatic stations within the national network monitoring traffic emissions, one local network station monitoring urban background pollution, and three private stations monitoring industrial emissions (Figure 2).

With time we intend to increase the number of local network stations. The amount of streamed data ranges from 2,688 to 3,696 rows per day. Continuous data streams that keep arriving from each automatic monitoring station have a common structure: stream ID, station ID, date, time, parameter ID, and value.

Parameters that are monitored include gases (NO, NO2, CO, O3, H2S, SO2, benzene, NH3), particulate matter (PM10

and PM2.5), and meteorological data (wind speed and direction, temperature, and pressure).

Architecture of THE On-line subsystem Our on-line subsystem for urban air quality monitoring and assessment relies on the so called Service Oriented Architecture (SOA), which is independent of specific technologies but can be implemented using a wide range of technologies (20). The benefit of implementing SOA with Web services is that it is platform-neutral.

Figure 3 shows the architecture of the on-line subsystem we developed for urban air quality monitoring and assessment. Automated stations feed the information system with data through a predefined query. They first go through automated validation, then air quality index (AQI) is calculated for every station, and a report sent to CEA for the local monitoring station. The last step is to save processed data in a database that shares data with our Web service application monitoring NO2, O3, and SO2 emissions on all automatic stations in Zagreb. If their values exceed national regulatory limits (O3 - 240 μg m-3, SO2 - 500 μg m-3, and NO2 - 400 μg m-3) (21) for three consecutive hours, the web service generates an alert for population groups at risk (children, pregnant women, the elderly, the chronically ill, and people in poor health) not to leave their homes.

Continuous queries over data streams Figure 4 shows sequential execution of multiple continuous queries over data streams coming from the automatic stations. The first complex continuous query validates concentration values in data streams with respect

174

Davila S, et al. Real-time dissemination of air quality information using data streams and Web technologies Arh Hig Rada Toksikol 2015;66:171-180

Figure 3 UML component diagram of on-line subsystem for urban air quality monitoring and assessment

Figure 4 Multiple continuous queries over data streams

Davila S, et al. Real-time dissemination of air quality information using data streams and Web technologies Arh Hig Rada Toksikol 2015;66:171-180

to a given set of criteria. The second query calculates air quality index, and the third query uploads an automatic hourly report to the CEA. The data obtained with each of the three queries are stored in the database and made accessible online by our web application. After 30 days, the data are archived.

Automatic data validation Streamed data are validated in real-time using complex continuous queries (see Query 1 in Figure 4). In the first stage, the pollutant value is matched to criteria taking into account outliers and pollutant limits (21, 22). In this way, the query discards outliers caused by incorrect measurement.

These values are not deleted but are marked not to have passed a specific criterion, so that missing data for a specific hour or parameter are all accounted for.

In the second stage, concentration values of pollutants that are known to be related (such as O3 and NO) are rechecked and compared to remove errors due to a failure or faulty measurement (e.g., span and zero readings due to a non-sealing valve). Data that do not pass the second stage of validation are also marked and stored in the database.

Data that pass both validation stages are marked as validated and ready for further processing.

Air quality index calculations Presently air quality information in Croatia (and the City of Zagreb) consists of raw monitoring data or corresponding categories of air quality for the past year.

These categories are based on pollution levels, limit and target values, and long-term goals. According to the national Air Protection Act (22) the first category denotes clean or slightly polluted air and the second category polluted air.

Such air quality information is more suitable for air quality experts rather than general public.

For the first time in Croatia, IMROH has introduced an air quality index based on the Common Air Quality Index (CAQI) model in accordance with the EU CAFE directive (23). The great advantage of AQI is that it allows comparison with a number of European cities in real time (24). The model uses a scale of 1 to 100, where 1 denotes best air quality and 100 the worst. The index is calculated from pollutant hourly mass concentrations, using a calculation grid (see Table 1) and the following formula (25, 26):

(1) where I is air quality index, C pollutant concentration, Clow concentration breakpoint is â&#x2030;¤C, Chigh concentration breakpoint is â&#x2030;ĽC, Ilow air quality index corresponding to Clow, and Ihigh air quality index corresponding to Chigh.

To adjust reporting to Croatian regulations (21, 22), air quality indices have been split at the AQI threshold of 75

175

to correspond to the two national air quality categories: I - clean to moderately polluted air and II - highly polluted air (see the grid in Table 1).

Table 2 shows the number of days in 2012 in which a prevailing pollutant (one with the highest index on a particular day at a particular station) exceeded its limit values. This calculation also relies on national legislation as well as CAQI methodology based on air quality indices.

For example, PM10 exceeded its national limit value on 38 days at the IMROH station and was the prevailing pollutant on 301 days.

Assessment of air quality based on air quality indices The second type of continuous query is used to assess air quality based on the AQIs (see Query 2 in Figure 4). It calculates the hourly index for each pollutant in real-time and the hourly index for the measuring site in real-time.

The hourly index for Zagreb in real-time and the mean index for the whole city are calculated by the web service later, after the data are stored in the database. The formula shown in Equation 1 and the AQI calculation grid (Table 1) are directly integrated into the continuous query over data streams to avoid communication with the database, as it would slow down data stream processing by 15Â %.

The hourly index for a measuring site is the highest index of a pollutant measured at the station for that hour.

The first step is to set zero as the default index, which will change as pollutants are being measured and indices calculated for a measurement site. The calculated index for each pollutant is later stored in a database and displayed online on a digital map of Zagreb.

Once air quality has been indexed, the third continuous query generates an automatic report for the CEA (see Query 3 in Figure 4) in the format of Air Quality-DEM and XML files (as required by the EEA).

Air quality database Streamed data are saved in a relational database. Figure 5 shows a segment of the database which is used to store information from automatic stations. AQ_DATA entity contains the basic attributes of the stream. In addition to the primary key, DataID, and foreign keys StationID, ComponentID, and AQIID, AQ_DATA has two binary-type attributes, AQDEM and AQXML, defining the format of the generated files. All other entities are related to AQ_ DATA. The COMPONENT entity contains attributes related to pollutants and meteorological data measured at the automatic stations. The AQI entity contains attributes related to air quality index. The STATION entity contains attributes with information on the automatic stations.

Air quality Web application Zagreb air quality monitoring information is displayed on the web page http://kvaliteta-zraka.imi.hr/ thanks to the web application based on CAQI methodology and

100-200 200-400

50-75

75-100

>100

Medium

High

Very high >180

90-180

50-90

24 h

>100

50-100

25-50

12.5-25

0-12.5

PM10

25-50

0-25

1h

Mandatory

>20000

10001-20000

7501-10000

5001-7500

0-5000

CO

Auxiliary

>400

200-400

100-200

50-100

0-50

NO2

301

0

4

0

60

PM10

SO2

NO2

CO

O3

Highest AQI

11

0

0

0

38

≥LV

Station IMROH

0

0

6

0

353

Highest AQI

0

0

0

28

≥LV

Station Zagreb1

>180

90-180

50-90

0

0

1

0

364

Highest AQI

0

0

0

28

≥LV

24 h

>100

>20000

10001-20000

7501-10000

5001-7500

71

0

0

0

294

0

0

0

0

31

≥LV

0-50

SO2

>500

300-500

100-300

50-100

Auxiliary

0-5000

CO

Station Zagreb3

>240

180-240

120-180

60-120

Highest AQI

50-100

25-50

12.5-25

0-60

O3

City background

0-12.5

PM10

25-50

0-25

1h

Mandatory

Station Zagreb2

Table 2 Number of days in 2012 when prevailing pollutants had the highest AQI and exceeded Croatian limit values (LV)

>400

50-100

25-50

Low

0-50

NO2

0-25

Index range (calculation grid)

Very low

Pollution class

Traffic

Table 1 Pollutants and calculation grid for the CAQI. Source: project CITEAIR II (35, 38)

176 Davila S, et al. Real-time dissemination of air quality information using data streams and Web technologies Arh Hig Rada Toksikol 2015;66:171-180

Davila S, et al. Real-time dissemination of air quality information using data streams and Web technologies Arh Hig Rada Toksikol 2015;66:171-180

177

Figure 5 UML diagram of relational database for storage of data from automated stations developed in accordance with European directives. Figure 6a and 6b shows the interactive dynamic map of the City of Zagreb with location markers of the automatic monitoring stations. Their colour changes with the air quality index.

Placing a cursor or clicking on location marker selects a monitoring station and displays air quality index details.

This allows a person who is not familiar with the Air Protection Act to understand the AQI in plain terms (11).

The application also associates AQIs with health risks for sensitive and general population and the corresponding notifications/alerts (Table 3). These notifications are reported for each station with the calculated index.

Every hour for each automated station our Web service application monitors whether pollutant levels have exceeded their limits. If the levels are exceeded for three hours in a row, the Web site displays an alert for the corresponding population group.

After a three-month trial, our web application was launched in the early 2014. Initially, it only processed the data obtained from the automatic station at the IMROH site.

In the early September of 2014, our on-line information system at IMROH started receiving and processing data

streams from all public automatic stations in Zagreb.

Judging by the server CPU and RAM usage of only 2-5 %, it can support a much larger number of stations. Furthermore, the amount of data streamed to the server (250 MB per hour or 6 GB a day) does not constitute a burden on the Internet connection at IMROH, whose bandwidth is currently 100 Mbps but will soon increase to 1 Gbps.

CONCLUSION Until now, data streaming has not been used in air quality monitoring. One of the reasons is that it requires complex streaming and querying designs that need to validate continuous measurements from automatic monitoring stations. Another reason could be that most research in the field of air quality ignores data transfer and communication from a terminal or device to the server or information system.

Our solution provides great benefits compared to traditional approaches to collecting, processing, and disseminating information. Continuous queries over data

Table 3 Air quality index and health risk notifications for sensitive and general population groups Air quality index

Notifications for sensitive population groups (old, children, ill, pregnant women, asthmatic patients, outdoor workers)

Notifications for the general, healthy population

Very low

Enjoy your daily outdoor activities.

Air quality is ideal for outdoor activities.

Low

Enjoy your daily outdoor activities.

Air quality is ideal for outdoor activities.

Medium

Consider reducing strenuous outdoor activities if symptoms occur.

High

Reduce or postpone strenuous outdoor activities. Children and the elderly should spend less time outdoors.

Very high

Avoid strenuous outdoor activities. Children and the elderly should also avoid strenuous outdoor activities.

No need to modify your usual outdoor activities unless you experience symptoms such as coughs and throat irritation.

Consider reducing or postponing strenuous outdoor activities if you experience symptoms such as cough and throat irritation.

Reduce or postpone strenuous outdoor activities, especially if you have symptoms such as coughing and throat irritation.

178

Davila S, et al. Real-time dissemination of air quality information using data streams and Web technologies Arh Hig Rada Toksikol 2015;66:171-180

a

b Figure 6 Web application of air quality in Zagreb a) map of the city; b) details of monitoring stations and information for the general public on air quality

Davila S, et al. Real-time dissemination of air quality information using data streams and Web technologies Arh Hig Rada Toksikol 2015;66:171-180

streams enable high-speed processing of a large number of measurement data and easily connect to automatic stations.

Furthermore it has proven to be stable and effective for streamed data collection and processing.

Coupled with the CAQI model, our web service application brings air pollution information closer to the general population and raises general awareness about environmental and health issues.

In the future we intend to develop a mobile application to expand the use of these web services. A future system could also use sound signals or messages for the benefit of the visually impaired or blind population. The system could also be extended to other air quality measuring stations across the country to get a better insight into the air quality in Croatia. With small adjustments, the system could also collect data from other non-air-quality measuring stations such as water quality stations, radiation stations, weather stations, and other automated stations.

9.

10.

11.

REFERENCES 1. World Health Organization and United Nations Environmental Programme. Health environment: managing the linkages for sustainable development: a toolkit for decision-makers: synthesis report, 2008 [displayed 23 January 2015]. Available at http://whqlibdoc.who.int/publications/2008/97 89241563727_eng.pdf?ua=1 2. Kura B, Verma S, Ajdari E, Iyer A. Growing public health concerns from poor urban air quality: strategies for sustainable urban living. CWEEE 2013;3:170-81. doi: 10.4236/cweee.2013.22B001 3. Kukec A, Farkas J, Erzen I, Zaletel-Kragelj L. A prevalence study on outdoor air pollution and respiratory diseases in children in Zasavje, Slovenia, as lever to trigger evidencebased environmental health activities. Arh Hig Rada Toksikol 2013;64:9-22. doi: 10.2478/10004-1254-64-2013-2165 4. Šega K, Bešlić I. Health impacts assessment of black smoke and airborne particulates on Zagreb Population. WHO Newsletter 2006;38:2-6.

5. Agencija za zaštitu okoliša. ISZO - Informacijski sustav zaštite okoliša [Croatian Environment Agency - Air Quality Information System, in Croatian] [displayed 23 January 2015]. Available at http://iszo.azo.hr/ 6. Croatian Accreditation Agency. Accreditation Certificate for Determination of air quality and pollutants in the air, No. 1288, Ref. No. 383-02/09-30/026, Id. No. 569-02/1-14-31, Zagreb, 2014-11-13. [displayed 23 January 2015]. Available at http://www.akreditacija.hr/registry 7. Hilty LM, Page B, Hrebicek J, editors. Environmental Informatics.

Environ Modelling Software 2006;21(11):1517-630.

8. Hrebicek J, Pillmann W. Shared environmental information system and single information space in Europe for the environment: antipodes or associates? In: Hrebicek J, Hradec J, Pelikán P, Mírovský O, Pillmann W, Holoubek I, Bandholtz T, editors. Proceedings of the European conference of the Czech Presidency of the Council of the EU TOWARDS eENVIRONMENT. Opportunities of SEIS and SISE: Integrating Environmental Knowledge in Europe; 25-27

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

179

March 2009. Prague, Czech Republic. Brno: Masaryk University in Brno; 2009. p. 447-58.

The European Parliament and the Council of the European Union. Directive 2007/2/EC of the European Parliament and of the Council of 14 March 2007 establishing an Infrastructure for Spatial Information in the European Community (INSPIRE) [displayed 21 May 2015]. Available at http:// e u r - l e x . e u r o p a . e u / L e x U r i S e r v / L e x U r i S e r v. do?uri=OJ:L:2007:108:0001:0014:en:PDF Davila S, Pečar-Ilić J, Bešlić I, Šega K. Towards real-time air quality monitoring in the city of Zagreb using stream database and web technologies. In: Kornelius G, editor.

Proceedings of the 16th IUAPPA World Clean Air Congress “Many Nations - One Atmosphere: Plotting the Path to Sustainability”; 29. September – 4 October 2013; Cape Town, Republic of South Africa. Cape Town: National Association for Clean Air; 2013. p. 1-6.

Davila S, Bešlić I, Pečar-Ilić J, Šega K. ICT activities for air quality monitoring: an example of network stations of the city of Zagreb. In: Arndt H-K, Knetsch G, Pillmann W, editors. Proceedings of the 26th International Conference on Informatics – Informatics for Environmental Protection, Sustainable Development and Risk Management “EnviroInfo 2012: Man Environment Bauhaus: Light up the Ideas of Environmental Informatics”; 29-31 August 2012; Dessau, Germany. Aachen: Shaker Verlag; 2012. p. 111-8.

Pawluk P. Stream Database [Master thesis]. Ronneby: School of Engineering: Blekinge Institute of Technology; 2006.

Kumar A. Event Processing Engine for Stream Databases.

Guwahati, India: Department of computer science and engineering, Indian Institute of Technology Guwahati; 2007.

Babu S, Widom J. Continuous queries over data streams.

ACM SIGMOD Record 2001;30:109-20.

Carney D, Centintemel U, Cherniack M, Convey C, Lee S, Seidman G, Stonebraker M, Tatbul N, Zdonik S. Monitoring streams – A new class of data management applications. In: Proceedings of the 28th VLDB Conference 2002; Hong Kong, China 2002. p. 215-26.

Chandrasekaran S, Franklin MJ. Streaming queries over streaming data. In: Proceedings of the 28th VLDB Conference 2002; Hong Kong, China 2002. p. 203-14.

Chandrasekaran S, Franklin MJ. PSoup: a system for streaming queries over streaming data. VLDB J 2003;12:14056. doi: 10.1007/s00778-003-0096-y Krishnan R, Goldstein J, Raizman A. Hitchhiker’s Guide to Microsoft StreamInsight Queries [displayed 23 January 2 0 1 5 ] . Av a i l a b l e a t h t t p : / / g o . m i c r o s o f t . c o m / fwlink/?LinkId=256236 Microsoft Cooperation. Microsoft StreamInsight™ 2.1.

Developer’s Guide; USA 2012 [displayed 23 January 2015].

Available at https://msdn.microsoft.com/en-us/library/ ee391564(v=sql.111).aspx Microsoft Cooperation. Understanding service-oriented architecture [displayed 23 January 2015]. Available at https:// msdn.microsoft.com/en-us/library/aa480021.aspx Uredba o razinama onečišćujućih tvari u zraku [Regulation on air pollutant levels, in Croatian]. Narodne novine 117/2012.

Zakon o zaštiti zraka [Air Protection Act, in Croatian].

Narodne novine 130/2011.

The European Parliament and the Council of the European Union. Directive 2008/50/EC of the European Parliament

180

Davila S, et al. Real-time dissemination of air quality information using data streams and Web technologies Arh Hig Rada Toksikol 2015;66:171-180

and of the Council of 21 May 2008 on ambient air quality and cleaner air for Europe [displayed 21 May 2015].

Available at http://eur-lex.europa.eu/LexUriServ/ LexUriServ.do?uri=OJ:L:2008:152:0001:0044:EN:PDF 24. van den Elshout S, Léger K, Nussio F. Comparing urban air quality in Europe in real time a review of existing air quality indices and the proposal of a common alternative. Environ Int 2008;34:720-6. doi: 10.1016/j.envint.2007.12.011 25. Hodges N, van den Elshout S, Heich H-J, Lad C, Léger K, Nussio F. CITEAIR – Common information to European air.

In: Hrebicek J, Racek J, editors. Proceedings of the 19th International Conference Informatics for Environmental Protection - EnviroInfo: Networking Environmental Information, Part 2; 7-9 Sept 2005; Brno, Czech Republic.

Brno: Masaryk University in Brno; 2005. p. 518-22.

26. van den Elshout S, Léger K, Heich H. CAQI Common Air Quality Index — Update with PM2.5 and sensitivity analysis.

Sci Total Environ 2014;488-489:461-8. doi: 10.1016/j. scitotenv.2013.10.060.

Izvješćivanje o kvaliteti zraka u stvarnom vremenu kontinuiranim prijenosom podataka i web tehnologijama povezivanje kvalitete zraka sa zdravstvenim rizicima u urbanim sredinama U ovom se članku predstavlja nova, originalna primjena suvremene informacijske i komunikacijske tehnologije radi učinkovitoga izvješćivanja opće populacije o kvaliteti zraka i s njom povezanih zdravstvenih rizika. Naš online podsustav praćenja kvalitete zraka u gradovima ključan je dio složenijega integriranoga informacijskoga sustava koji je razvio Institut za medicinska istraživanja i medicinu rada. Oslanja se na sustav upravljanja kontinuiranim prijenosom informacija (engl. data stream management system) razvijen pomoću StreamInsighta i SOA arhitekture radi obrade podataka koji neprekidno dolaze sa sedam automatskih postaja za praćenje kvalitete zraka diljem Zagreba. Prate se sljedeći parametri: NO, NO2, CO, O3, H2S, SO2, benzen, NH3, čestice u zraku (PM10 i PM2.5), brzina i smjer vjetra, temperatura i tlak zraka.

Zbog stalnih složenih upita (engl. continuous query) podaci se obrađuju u stvarnom vremenu. Prvi je korak automatska validacija pristiglih podataka, zatim se izračunava indeks kvalitete zraka za svaki sat, a potom se izvještaj šalje Agenciji za zaštitu okoliša. Ako tri sata za redom vrijednosti pojedinih parametara nadilaze granične vrijednosti utvrđene zakonom, web usluga šalje upozorenje osjetljivim populacijskim skupinama (bolesnicima, trudnicama, djeci, radnicima na otvorenom i dr.). Oslanjajući se na model europskoga indeksa kvalitete zraka (Common Air Quality Index, CAQI), naša web aplikacija približava općoj populaciji aktualne podatke o onečišćenju zraka te podiže svijest o problemima vezanima uz okoliš i zdravlje. Uskoro namjeravamo proširiti ovu uslugu na mobilnu aplikaciju, koja je u izradi.

KLJUČNE RIJEČI: indeks kvalitete zraka; integrirani informacijski sustav; javno zdravstvo; onečišćenje zraka; praćenje kvalitete zraka; web usluge

A11

REPORT Arh Hig Rada Toksikol 2015;66:A11-A12

REPORT 10th SYMPOSIUM OF THE CROATIAN RADIATION PROTECTION ASSOCIATION with international participation Solaris-Hotel Jure, Šibenik, Croatia, 15-17 April 2015 The Tenth Symposium of the Croatian Radiation Protection Association (CRPA) with international participation was held from 15-17 April 2015 in Šibenik, Croatia. The symposium was organized by the CRPA in collaboration with the Institute for Medical Research and Occupational Health (IMROH), Ruđer Bošković Institute (RBI) and State Office for Radiological and Nuclear Safety (SORNS). The presidents of the Organizing and Scientific Committee were Tomislav Bituh and Branko Petrinec from IMROH, respectively.

There were 90 participants in total: 70 from national scientific and research institutions, universities, health facilities, and private companies and 20 international participants (from Bosnia and Hercegovina, Hungary, Slovenia, Serbia, Macedonia, and Qatar). This year, for the first time, the venue for the Symposium was the pleasant and comfortable hotel “Jure” in the Solaris Beach Resort in Šibenik.

note that all of the oral presentations were presented in a timely manner and every presentation resulted in a discussion. K. Marušić et al. (RBI) were voted first for best poster presentation for: “Protection of cultural heritage objects by ionizing radiation”. The second prize went to two poster presentations: “On-site assessment methods for environmental radioactivity” by B. Petrinec et al. (IMROH) and “226Ra in trickling waters from the vicinity of a phosphogypsum deposition site” by T. Bituh et al.

The Cathedral of St. James in Šibenik (view from the fortress of St. Michael)

Poster section (28 presentations)

The scientific part of the symposium was divided into eight sections: General topics in radiation science and radiation protection; Radiation dosimetry; Biological effects of radiation; Radiation protection of the public; Radiation protection in medicine; Radioecology; Non-ionising radiation and Instrumentation and measuring techniques.

This year, there was a total of 65 papers, all printed in the Proceedings edited by Branko Petrinec, Tomislav Bituh, Nevenka Kopjar, and Mirta Milić. There were 28 poster presentations and 37 oral presentations. It is important to

(IMROH). The third best poster was authored by I. Franulović et al. (IMROH) and was entitled “90Sr activity levels in the Croatian part of the Sava River course”.

This year, we had only one exhibitor: LKB Vertriebs Ges.m.b.H. with two participants: Dušan Djurdjević (Vienna) and Draško Petrović (Zagreb).

The program comprised two introductory lectures: Ines Krajcar Bronić (CRPA President) “Ten symposia of the Croatian Radiation Protection Association 1992-2015” and

A12

REPORT Arh Hig Rada Toksikol 2015;66:A11-A12

Presentations were held in rooms Murter I+II of the Hotel Jure (Solaris-Šibenik)

CRPA 10: the participants during their visit to the fortress of St. Michael

Dubravko Pevec (Faculty of Electrical Engineering and Computing, University of Zagreb) “The potential of fission nuclear energy in resolving global climate change”.

The CRPA Assembly was also held during the Symposium.

For all of the participants, two excursions were organized. The Šibenik city guided tour with a visit to the

fortress of St. Michael, and after the closing of the symposium a visit to Etnoland Dalmati in Pakovo Selo.

The success of the symposium should also at least in part be attributed to our sponsors: ALARA uređaji, BISS, Jadransko osiguranje, LKB Vertriebs Ges.m.b.H., Medicem servis, Mettler Toledo, Podzemno skladište plina, Selmet, the Ivanić-Grad Fire department, and VERN. Tomislav Bituh