Comparison%20of%20the%20action%20spectra%20and%20relative%20dna%20absorbance%20spectra

water research 43 (2009) 5087–5096

Available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/watres

Comparison of the action spectra and relative DNA absorbance spectra of microorganisms: Information important for the determination of germicidal fluence (UV dose) in an ultraviolet disinfection of water Ren Zhuo Chen 1, Stephen A. Craik 2, James R. Bolton* Department of Civil and Environmental Engineering, University of Alberta, Edmonton, AB, Canada T6G 2W2

article info

abstract

Article history:

The action spectra of Bacillus subtilis spores (ATCC6633) and Salmonella typhimurium LT2

Received 29 June 2009

were characterized using physical radiometry for irradiance measurements and a multiple

Received in revised form

target model to interpret the inactivation kinetics. The observed action spectrum of

21 August 2009

B. subtilis spores deviated significantly from the relative absorbance spectrum of the DNA

Accepted 24 August 2009

purified from the spores, but matched quite well with the relative absorbance spectrum of

Available online 28 August 2009

decoated spores. The action spectrum of B. subtilis spores determined in this study was statistically different from those reported in previous studies. On the other hand, the

Keywords:

action spectrum of S. typhimurium bacteria matched quite well with the relative absorbance

Bacillus subtilis spores

spectrum of DNA extracted from vegetative cells, except in the region below 240 nm. It is

Salmonella typhimurium

concluded that the common use of the relative DNA absorbance spectrum as a surrogate

Ultraviolet

for the germicidal action spectrum can result in systematic errors when evaluating the

Action spectrum

performance of a polychromatic UV light reactors using bioassays. For example, if the

DNA absorbance spectrum

weighted germicidal fluence (UV dose) calculated using the relative DNA absorbance spectrum as the germicidal weighting factor is found to be 40 mJ cm 2 for a medium pressure lamp UV reactor, that calculated using the relative action spectrum of B. subtilis spores, as determined in this study, would be 66 mJ cm 2. ª 2009 Elsevier Ltd. All rights reserved.

1.

Introduction

Ultraviolet (UV) treatment is an increasingly popular technology for the disinfection of drinking water and secondary or tertiary wastewater effluents. Although low pressure (LP) UV lamps are commonly used for small UV disinfection plants, medium pressure (MP) mercury discharge lamps, which produce polychromatic germicidal UV light, are found in many large-scale UV disinfection facilities for drinking water

and wastewater treatment. Many UV disinfection studies have been carried out using MP lamps in bench-scale batch UV reactors (Bolton et al., 1998; Craik et al., 2000; MamaneGravetz et al., 2005; Linden et al., 2005). The computation of the fluence (UV dose) in large continuous-flow UV reactors, used for the disinfection of drinking water, is complex. The fluence (UV dose) delivery for a given UV reactor design must be validated. Several approaches (biodosimetry, computational fluid dynamic (CFD) modeling and dyed microsphere

* Corresponding author. Tel.: þ1 780 439 4709; fax: þ1 780 439 7792.

E-mail address: jb3@ualberta.ca (J.R. Bolton).

1 Present address: Beijing Onyx Technologies, 11 Xingxi Road, ShangDi, Haidian District, Beijing 100085, China.

2 Present address: EPCOR Water Services, 10065 Jasper Avenue, Edmonton, AB, Canada T5 J 3B1.

0043-1354/$ – see front matter ª 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2009.08.032

5088

water research 43 (2009) 5087–5096

Nomenclature d BPF CFU DPA F IL LP k krel

Shoulder inactivation coefficient Band-Pass Filter Colony Forming Units Dipicolinic acid Fluence (UV dose) International Light Low Pressure mercury discharge lamp Inactivation rate constant (cm2 mJ 1) Inactivation rate constant relative to 254 nm

actinometry) are available or potentially available for the validation of UV reactors (USEPA, 2006). Biodosimetry, in which the inactivation of a surrogate microorganism added to the water is measured, has been accepted by regulatory agencies. For reactors equipped with MP lamps, the evaluation of the reactor performance using a bioassay is affected by the assumption made concerning the relative action spectra of the surrogate microorganism and target pathogen.

A standardized method, which uses a germicidal factor to calculate the germicidal fluence (UV dose) from a polychromatic source in bench-scale UV experiment, has been proposed (Bolton and Linden, 2003). When polychromatic lamps are used for UV disinfection studies and for interpreting the results of MP UV reactor bioassays, careful determination of the relative action spectrum (relative to a ‘germicidal action’ of 1.00 at 254 nm) of a microorganism is crucial for the accurate determination of the germicidal factors and hence the germicidal fluence (UV dose). The experimental methods to determine action spectra, however, are not standardized.

Different methods have been reported in the literature for the determination of the germicidally weighted fluence (UV dose) for polychromatic UV sources. Moreover, studies have reported that the germicidal efficacy of UV light is highly dependent on the wavelength and might vary from microorganism to microorganism (Gates, 1929; Rauth, 1965; Linden et al., 2001; Linden et al., 2005).

The DNA absorbance spectrum reported by von Sonntag and Schuchmann (1992) has been conventionally used as a surrogate when the inactivation action spectrum of a microorganism is not available. However, it is not clear from their paper as to what the source of the DNA was. Since the nucleotide composition of DNA varies from one organism to another, the DNA absorbance spectrum might vary. In this study, action spectra are compared with the relative absorption spectra of DNA and other components from the same microorganism. Two microorganisms were chosen for the study: Bacillus subtilis spores, which are widely used as a nonpathogenic surrogate in biodosimetry testing of UV reactors and Salmonella typhimurium LT2, an attenuated strain of an important pathogenic bacterium.

B. subtilis spores have been used commonly as a surrogate microorganism to validate the performance of continuousflow UV reactors used in drinking water treatment plants ¨ NORM, 2001, 2003) and for (USEPA, 2006; DVGW, 2006; O fundamental UV disinfection studies (Sommer and Cabaj, 1993; Uvbiama and Craik, 2005). The action spectra for

MP nc N N0 Ni PBDI REF SASP

Medium Pressure mercury discharge lamp Number of critical targets of multiple target model Concentration of viable microorganisms after UV exposure Concentration of viable microorganisms before UV exposure Relative photon flux at wavelength li Phosphate Buffered Deionized Reduction Equivalent Fluence Small Acid Soluble Proteins

B. subtilis have been determined for spores produced from liquid-cultivated media (Cabaj et al., 2002) and from surfacecultivated sporulation media (Mamane-Gravetz et al., 2005).

Both studies reported similar first-order spore UV inactivation rate constants. However, the reported wavelength dependence of the ‘shoulder’ in the fluence (UV dose)–response curve for B. subtilis spores, differed between the two studies.

The shoulder region at low fluence (UV dose), where there is little or no inactivation, must be considered carefully when evaluating UV reactors using bioassays (Cabaj et al., 2002).

Riesenman and Nicholson (2000) concluded that the spore coat does not contribute to the resistance of spores to 254 nm UV light. It has been reported that B. subtilis spores have their own unique photochemistry and repair mechanisms, and this may account for why spores are more resistant to UV light than are vegetative B. subtilis cells (Setlow, 2001).

Salmonella spp. is an important waterborne pathogen in drinking water. Most UV disinfection studies of Salmonella spp. have been carried out using monochromatic UV light (Chang et al., 1985; Tosa and Hirata, 1998; Koivunen and HeinonenTanski, 2005). However, the inactivation action spectrum of Salmonella has not yet been reported. Moreover, Salmonella DNA might have its own unique absorbance spectrum. The strain of S. typhimurium LT2 used in this study has a highly attenuated virulence and has been commonly used as a surrogate for virulent Salmonella spp. in laboratory studies.

In this paper not only are the relative action spectra of B. subtilis spores and S. typhimurium bacteria determined, but also, for the first time, the action spectra are compared to the relative DNA absorbance spectra extracted from the same microorganism. Also, in the case of B. subtilis spores, a comparison is made between the action spectrum and the relative absorbance spectrum of decoated spores. Where available, the relative action spectra are compared to previously published action spectra. Finally, the potential effect of different action spectra on the determination of the germicidal fluence (UV dose) in a MP UV reactor validation was evaluated.

2.

Materials and methods

2.1.

B. subtilis spore production, preparation of test suspensions, and enumeration All sterile media, reagents and materials were pre-sterilized.

Procedures for the preparation of B. subtilis spores (ATCC 6633)

5089

water research 43 (2009) 5087–5096

stock solution by growth and sporulation in liquid culture and for the enumeration of the concentration of spore samples have been described in detail elsewhere (Uvbiama and Craik, 2005; Craik and Uvbiama, 2005); these procedures were used here with slight modifications. The water matrix used to prepare a test suspension or series dilution for enumeration was sterile phosphate-buffered (0.05 M, pH 7.0) deionized phosphate-buffered (PBDI) water. The experimental spore suspensions were prepared by adding an aliquot of the stock spore suspension to a volume of sterile PBDI water to yield a spore concentration of approximate 106 CFU mL 1. A nutrient agar pour plate technique was used to determine the viable spore concentration (in CFU mL 1) in the stock solution, test suspensions and experimental samples.

2.2.

S. typhimurium LT2 production, preparation of test suspensions, and enumeration The strain of S. typhimurium LT2 (SL3770 from the Salmonella Genetic Stock Centre at University of Calgary) was kindly provided by Dr. Diane Taylor’s laboratory, at the Department of Medical Microbiology, University of Alberta. The Salmonella culture medium was prepared by dissolving 30 g of Tryptone Soya Broth (TSB, Oxid CM129) per liter of Milli-Q water. The medium was incubated at 35 C in an incubator shaker (Innova 4080, New Brunswick Instruments Co. Inc., Edison, NJ) at 180 rpm for 24 h. After incubation, Salmonella was harvested by centrifugation at 7500 g for 20 min at 4 C. The supernatant was decanted and the pellet was re-suspended in sterile PBDI water and then homogenized with a homogenizer (PowerGen 700, Fisher Scientific). The centrifugation and wash steps were repeated more than three times. The final pellet was then suspended in sterile 0.05 M PBDI water and homogenized. The homogenized solution was diluted in sterile PBDI water to produce a stock solution. A test solution with approximately 107 CFU mL 1 of Salmonella was prepared by diluting the stock solution into sterile PBDI water with a Teflon coated magnetic stir bar.

A spread plate method was employed to enumerate the viable bacteria.. The detailed procedures are described elsewhere (Maier et al., 2000; Pepper et al., 1995). Decimal dilutions were prepared and three replicates of a selected dilution were were spread on solidified MacConkey agar (Difco , Becton Dickson and Company, MD) using a sterile glass rod. The plates with seeded agar were incubated at 32 C for 24 h and plates yielding approximately 30–300 colonies were counted and recorded.

2.3.

UV exposures

The protocol for bench-scale UV exposure experiments described by Bolton and Linden (2003) was used for this study.

A collimated beam apparatus (Calgon Carbon Corporation, USA) was used for all UV exposures. To generate monochromatic UV radiation at 254 nm, spore suspensions were exposed to a LP mercury arc lamp (Ster-L-Ray Germicidal Lamp, 10 W, model G12T6L 15114, Atlantic Ultraviolet Corporation, USA). A set of ten band-pass filters (BPF) (Andover Corporation, USA) was used to isolate narrow bands of UV radiation produced from a 1 kW MP lamp (HNG, Germany).

Each wavelength band was produced through the combination of the MP lamp output spectrum and the spectral transmission spectrum of the appropriate filter. The filter was inserted at the top of the collimating tube, about 15 cm beneath the MP lamp. The relative spectral irradiance of each wavelength band was measured directly using a spectroradiometer (Model USB2000, Ocean Optics, USA).

The weighted average wavelength of each band was calculated from equation (1) n P

Ni li Weighted average wavelength ¼ i¼1n P Ni

(1)

i¼1

where Ni is the relative photon flux at wavelength li. The characteristics of each wavelength band are presented in Table 1.

2.4.

Fluence (UV dose) determination

Calculations of the unweighted volume-averaged fluence rate at each wavelength band were based on the protocol proposed

by Bolton and Linden (2003). For the LP lamp, the spreadsheet proposed by Bolton (IUVA, 2005) was used for fluence rate calculations. A modified version of the Excel spreadsheet developed by Bolton (IUVA, 2004) was used for the wavelength bands produced from the MP lamp using BPFs. The modification was that a 0.38 nm wavelength computation interval was used instead of a 5 nm interval. Also, for each wavelength band, the total irradiance incident at the center of the surface of the suspension was first measured using a calibrated International Light radiometer (Model IL 1400A) with a model SEL 240 detector (International Light, Peabody, MA). The detector sensor factor, which accounts for the wavelength dependence of the sensor response, was determined using calibration data provided by International Light. The absorbance spectrum between 200 and 320 nm of each test suspension was measured before each UV exposure experiment using a spectrophotometer (UV-2401PC, Shimadzu Corp., Columbia, MD). For each wavelength band, the unweighted volume-averaged fluence rate was then calculated and the fluence (UV dose) was determined by multiplying the volume-averaged fluence rate by the UV exposure time.

Table 1 – Characteristics of the narrow band UV light produced by the MP lamp and the band-pass filters (BPFs).

Nominal BPF Weighted average Half peak wavelength/nm wavelength/nm height bandwidth/nm 214 220 228 239 248 260 270 280 290 300

222 225 232 243 251 264 268 279 292 303

10.3 9.1 9.9 13.3 6.8 9.1 7.2 3.4 9.7 9.7

5090

water research 43 (2009) 5087–5096

2.5.

Comparison of radiometry calibration with ferrioxalate actinometry Ferrioxalate actinometry is accepted as the ‘gold standard’ for determining absolute irradiances in photochemical studies (Kuhn et al., 2004). Recently, Goldstein and Rabani (2008) have updated the quantum yield versus wavelength for this actinometer. Ferrioxalate actinometry was carried out for each of the BPFs plus the LP UV lamp as a check on the calibration of the International Light radiometer detector, as provided by the manufacturer. Table 2 shows the comparison of irradiances determined by the two methods. The results agree within 10%.

2.6.

Data analysis and modeling

The concentrations N0 and N of viable B. subtilis spores for the unexposed and UV exposed test suspensions, respectively, were determined by the enumeration protocol described above. For each wavelength band, the fluence (UV dose) inactivation response curve was mathematically modeled using the multitarget model (Severin et al., 1983; Cabaj et al., 2002) given by:

nc N ¼ 1 1 10 kF N0

(2)

where F is the fluence (UV dose) at a given wavelength, nc is the number of critical targets and k is the first-order inactivation rate constant. For each wavelength band, k and nc, were estimated using non-linear regression with least squares criteria.

Calculations were carried out with the Solver tool in Microsoft Excel 2003.

To facilitate comparison between spectra determined in this and in previous studies, a normalized action spectrum was then obtained by plotting the ratio krel ¼ kl/k254 where k254 is the rate constant determined at 254 nm, and kl is the rate constant determined for the particular wavelength band as described in Table 1.

2.7.

Measurement of the spectral absorbance of B. subtilis DNA and decoated spores B. subtilis spores in the stock solution were suspended in MilliQ water, and the suspension was centrifuged at 7500 g for

10 min at 4 C. The supernatant was removed and the pellet was re-suspended in clean Milli-Q water. The wash step was repeated twice more. The detailed spore decoating process is described elsewhere (Slieman and Nicholson, 2000). The suspension of decoated spores was then treated with 0.5 mL lysozyme (Sigma , USA), incubated at 37 C for 30 min, and then treated with 0.1 mL proteinase K (DNeasy , Germany).

The lysed spores were incubated at 75 C for 30 min. Chromosomal DNA in the lysed spores solution was extracted and purified by phenol–chloroform extraction and ethanol precipitation according to the method proposed by Ausuvel et al.

(1992). The DNA precipitate was dried in air to evaporate the ethanol for more than 12 h and then dissolved in TE buffer (10 mM pH 8.0 Tris–HCl buffer and 1 mM EDTA). The absorbance spectrum of the spore DNA was measured using the UV-2401PC spectrophotometer in dual beam mode. A solution of TE buffer served as the reference for the spectrophotometry.

To measure the absorbance spectrum of decoated spores, the decoated spore suspension was centrifuged at 7500 g for 5 min at 4 C. The supernatant was removed, and the pellet was re-suspended in Milli-Q water to a final volume of approximately 10 mL. This wash step was repeated twice more. The pellet was then dissolved by the addition of 1 mL of 1 M HCl followed by incubation at 75 C for 30 min (Tsuboi, 1950). The dissolved spore solution was neutralized by adding 1 mL of 1 M NaOH and was buffered to pH 7.0 by addition of 8 mL 0.05 M PBDI water. The absorbance spectrum of decoated spores was measured as described above. A solution prepared using the same procedure, excluding the decoated spore pellet, served as the spectrophotometry reference.

2.8.

Measurement of the spectral absorbance of DNA from S. typhimurium LT2 Salmonella cells (<2 109) were harvested in a microcentrifuge tube by centrifugation at 5000 g for 10 min and the supernatant was decanted. The pellet was resuspended in Milli-Q water and mixed by vortex. The cells were washed and purified by centrifugation at 5000 g at 4 C for 10 min three times.

Salmonella genomic DNA was extracted and purified using a DNeasy kit (QIAGEN, Germany). The detailed procedure provided by the manufacturer for the DNA extraction and

Table 2 – Comparison between the Ferrioxalate Actinometry and Radiometry.

Wavelength (nm)

222 225 232 243 251 LP 253.7 264 268 279 292 303

Radiometer reading (W/m2)

IL Sensor factor

0.228 0.308 0.678 0.755 0.798 0.943 1.800 1.250 1.100 0.655 0.930

1.72 1.61 1.33 1.10 1.01 1.00 1.02 1.03 1.13 1.37 1.98

True irradiance (W/m2) Physical radiometry

Ferrioxalate actinometry

0.391 0.495 0.902 0.831 0.805 0.943 1.836 1.391 1.243 0.897 1.839

0.437 0.546 0.932 0.788 0.738 0.854 1.771 1.363 1.128 0.843 1.599

Actinometry/ Radiometry ratio

1.12 1.10 1.03 0.95 0.92 0.91 0.97 0.98 0.91 0.94 0.90

5091

water research 43 (2009) 5087–5096

purification was followed. The absorbance spectrum of Salmonella DNA was measured using the UV 2401PC spectrophotometer in dual beam mode AE buffer served as the reference.

tailing behavior, data in the tailing regions were omitted from the least-squares estimates of the parameters k and nc. The best fits to the inactivation curves shown in Fig. 1 were derived from the least-squares estimates of k and nc.

3.

3.2.

Comparison of action spectrum of B. subtilis spores and the absorbance spectra of B. subtilis DNA and decoated spores

Results and discussion

3.1.

Fluence (UV dose)–inactivation response curves for B. subtilis spores The inactivation responses of B. subtilis spores versus fluence for the various wavelength bands are presented in Fig. 1 using semi-log plots. Each response curve exhibited a shoulder effect at low fluence (UV dose) that is characteristic of B. subtilis spores. The resistance of the spores to low fluences has been attributed to the self-repair capability of their doublestranded DNA (Harm, 1980). When the log inactivation was greater than 4, tailing effects were observed in some of the response curves (e.g., at 225, 232 and 264 nm), probably due to experimental artifacts such as spore clumping (Craik et al., 2001). Since the multi-target model does not accommodate

a

b

Fluence (UV Dose) / mJ cm-2 0

0

50

100

150

200

Fig. 2 compares the relative action spectrum for B. subtilis spores and the absorbance spectra of purified spore DNA and of decoated spores. The inactivation action spectrum indicates the presence of a maximum at 222 nm, a minimum at around 250 nm and a smaller peak around 270 nm. The spectrum then declines to nearly zero at 303 nm. The action spectrum of B. subtilis spores reveals that the spores are much more sensitive to UV at wavelengths below 240 nm and around 270 nm than would be predicted by the DNA absorbance spectrum.

As shown clearly in Fig. 2, the B. subtilis spore action spectrum deviates significantly from the absorbance spectrum of purified spore DNA. Previous studies also found that

250

0

1

0

500

1000

1500

1

292 nm

-lo g (N /N0 )

-lo g (N /N 0 )

Fluence (UV Dose) / mJ cm-2

2 3

2000

2500

303 nm

2 3

4

4

5

5

Fluence (UV Dose) / mJ cm-2

c 0

0

10

20

30

40

50

60

70

80

1

-log(N /N0 )

2

3

4

5

6

Fig. 1 – Fluence-inactivation response of B. subtilis spores (ATCC 6633) in 0.05 M PBDI water for different wavelengths and curves derived from a best-fit of the multiple target model to the experimental data. (a) 292 nm; (b) 303 nm; (c) A 222 nm, :225 nm, - 232 nm, 3 243 nm, B 251 nm, 6 254 nm (LP lamp), D 264 nm, , 268 nm, > 279 nm.

water research 43 (2009) 5087–5096

mechanisms explain, at least partly, why the relative action spectrum of B. subtillis spores matches the relative absorbance spectrum of the decoated spore solution better than it does to that of the spore DNA. The former contain DPA, SASP and other proteins such as SP lyase, while the latter does not.

3.0 2.5 2.0 1.5

3.3.

Comparison to B. subtilis inactivation action spectra reported in previous studies

1.0 0.5 0.0 200 210 220 230 240 250 260 270 280 290 300 310

Wavelength / nm

Fig. 2 – Comparison of the absorbance spectrum of DNA purified from B. subtilis spores, the absorbance spectrum of decoated B. subtilis spores and the spore action spectrum: absorbance spectrum of decoated spores; absorption spectrum of DNA isolated from spores; action spectrum of B. subtilis spores.

the relative action spectra of B. subtilis spores did not match the relative DNA absorbance spectrum (Cabaj et al., 2002; Mamane-Gravetz et al., 2005). They suggested that other factors may be involved in the inactivation of B. subtilis spores

by UV light.

On the other hand, the action spectrum of B. subtilis spores matches very well with the relative absorbance spectrum of decoated spores. It has been reported that the UV resistance of spores is elevated by a group of a/b type SASPs that bind to the DNA (Setlow et al., 1992). This binding alters the spore DNA photochemistry and affects DNA repair. Indeed, when these proteins were removed from B. subtilis spores in vitro, it was observed that the photochemistry of DNA and the UV sensitivity were similar to those of the vegetative cells (Mason and Setlow, 1986). Also, certain aromatic amino acids in proteins, such as phenylalanine, tyrosine, and tryptophan, show a strong absorbance peak below 240 nm and a weaker peak in the 260–280 nm region. In addition, two DNA repair mechanisms, the nucleotide excision repair and the unique spore photoproduct repair, can efficiently remove spore photoproducts in the early phase of spore germination and outgrowth. These two repair mechanisms may also account for the higher UV resistance of B. subtilis spores compared to that of vegetative cells (Setlow, 2001).

A recent study reported that the unique photochemistry of B. subtilis spores may be associated with DPA (Douki et al., 2005). Thus, one may hypothesize that the novel photochemistry of DNA in B. subtilis spores may be attributed to the combined effects of SASP and DPA in the spore core. These two components bind spore DNA tightly and change the DNA conformation (Setlow, 2001). The mechanism for the effects of SASP and DPA on the photochemistry of B. subtilis spores can be explained by a photosensitized triplet energy transfer (Lamola, 1970; Charlier and Helene, 1972). The photon energy absorbed by either SASP or DPA in the spore core may be transferred to the DNA strands and thus cause DNA damage.

The combined effects of proteins, DPA and/or other spore components are thus a possible cause of the photochemistry that leads to inactivation of UV exposed spores. These

Fig. 3 compares the relative action spectrum for B. subtilis spores to the relative action spectra reported previously in the literature for the same species of microorganism. Clearly, the relative action spectrum determined in this study is distinctly different from those reported by Cabaj et al. (2002) and by Mamane-Gravetz et al. (2005), particularly in the wavelength region below 240 nm and in the region of the peak at 270 nm. To compare the action spectrum from this study to other two, the krel values at each 5 nm interval were interpolated from 220 to 290 nm for all three action spectra.

A paired t-test on the relative k values indicated that the action spectrum determined in this study was statistically different from the other two at a 95% confidence level.

Possible explanations for the differences between studies include differences in the methods used to interpret and model the inactivation curves. In this study a non-linear multi-target model that accounts for the shoulder effect was used to model the inactivation curves of Fig. 1, while data from the shoulder regions were not included in the linear model of Mamane-Gravetz et al. (2005) or the non-linear model of Cabaj et al. (2002). In addition, data in the tailing region of the inactivation curves in Fig. 1 was omitted from the modeling analysis in this study. In previous studies, data in the tailing region might have been included, thus resulting in bias and less accurate estimates of the first-order rate constant. If the data in the tailing region were excluded from the analysis of the fluence (UV dose) response at 265 nm in the Mamane-Gravetz et al. (2005) dataset, the value of krel generated by linear regression would better match the value determined in the present study in the region above 250 nm.

Cabaj et al. (2002) used the k values from at least two previous papers in the literature to generate their action spectrum.

This might be problematic because the UV sensitivity of B. subtilis spores could vary from batch to batch (Uvbiama, Inactivation Rate Constant (krel) relative to 254 nm

Inactivation Rate Constant (krel) or Absorbance (Arel) relative to 254 nm

5092

3.0 2.5 2.0 1.5 1.0 0.5 0.0 200 210 220 230 240 250 260 270 280 290 300 310

Wavelength / nm

Fig. 3 – Comparison of action spectra of B. subtilis spores this study; Cabaj et al., 2002; (ATCC 6633).

Mamane-Gravetz et al. 2005.

5093

water research 43 (2009) 5087–5096

2006). Finally, we point out that Cabaj et al. (2002) and Mamane-Gravetz et al. (2005) did not have any data points near 222 and 225 nm in their studies. It may be that, as a result, they missed the low-wavelength peak.

3.4.

UV fluence–inactivation response curves of Salmonella Curves of log UV inactivation responses of S. typhimurium LT2 versus UV fluence for each wavelength band are presented in Fig. 4. Each fluence response also exhibited a shoulder effect at low UV fluence that is characteristic of S. typhimurium LT2.

These fluence response curves were plotted using the parameters estimated from the multiple target model. The initial resistance to low UV fluence has been attributed to the self-repair capability of double-stranded DNA in these microorganisms as discussed previously. When the log inactivation was greater than 4, slight tailing effects were also

a

Fluence (UV Dose) / mJ cm-2 50

100

150

200

250

3.5.

Comparison of the action spectrum to the DNA absorbance spectrum of S. typhimurium LT2 As shown in Fig. 5, the overall shape of the action spectrum of S. typhimurium LT2 decreases from 222 nm with a low valley around 240 nm, then increases up to a peak around 265 nm, and declines to nearly zero at 302 nm. For comparison, the absorbance spectrum of purified Salmonella DNA relative to 254 nm is also presented in Fig. 5. The action spectrum of S. typhimurium LT2 is fairly close to its own DNA absorbance spectrum over most of the wavelength range. This indicates that the absorbance of photons by DNA is the primary mechanism of Salmonella spp. inactivation by UV light. An appreciable difference between the two spectra, however, was observed below 230 nm, where the inactivation was greater

Fluence (UV Dose) / mJ cm-2

b 300

350

0

0.0

0.0

1.0

1.0

2.0

2.0

-log(N/N0 )

–log(N/N0 )

0

observed in some of the response curves (i.e., at 225, 292 and 303 nm).

3.0

10

5

15

20

25

30

35

40

45

50

3.0

4.0

4.0

5.0

5.0

Fluence (UV Dose) / mJ cm-2

c

0

2

4

6

8

10

12

14

16

18

20

0.0

1.0

– log(N/N0 )

2.0

3.0

4.0

5.0

6.0

Fig. 4 – Fluence-inactivation response of S. typhimurium LT2 in 0.05 M PBDI water for different wavelengths and curves derived from a best-fit of the multiple target model to the experimental data. (a) 292 nm; (b) 303 nm; (c) A 222 nm, :225 nm, 268 nm, , 279 nm.

6 232 nm, 3 243 nm, 251 nm, C 254 nm (LP lamp), D 264 nm,

*

Relative inactivation rate krel or Absorbance Arel to 254 nm

5094

water research 43 (2009) 5087–5096

1.80 1.60 1.40 1.20 1.00 0.80 0.60 0.40 0.20 0.00 210 220 230 240 250 260 270 280 290 300 310

Wavelength (nm) Fig. 5 – Comparison of the absorbance spectra of S. typhimurium LT2 DNA and the action spectra of absorption spectrum of DNA S. typhimurium LT2: action spectrum isolated from S. typhimurium LT2; of S. typhimurium LT2.

than that predicted based on DNA absorbance alone. This suggests that proteins may also be involved and may play a minor role in the inactivation mechanism, since proteins absorb much more strongly in this region. One possible explanation is the formation of DNA-protein crosslinks induced by UV light (Smith et al. 1966), although these researchers used monochromatic UV light (253.7 nm) in which the absorbance of proteins was much lower than that of nucleic acids. However, the intrinsic sensitivity of Escherichia coli cells was directly correlated to the amount of DNA crosslinked to protein.

the standard DNA absorbance spectrum is determined to be 40 mJ cm 2, the fluence (UV dose) calculated using the relative action spectrum of B. subtilis spores and that of S. typhimurium LT2 determined in this study would be 66 mJ cm 2 and 42.5 mJ/m2 respectively.

Regulatory agencies have proposed using B. subtilis spores as a surrogate microorganism in reactor bioassays (USEPA, ¨ NORM, 2001, 2003). The bioassays are 2006; DVGW, 2006; O used to evaluate the inactivation of target pathogens by the UV reactor under flow-through test conditions using a surrogate microorganism because the direct measurement of pathogen inactivation is not practical. Evaluation of reactor performance involves determination of the Reduction Equivalent Fluence (REF) in a bioassay experiment. The determination of REF for MP lamp reactors includes calculating a polychromatic bias and is thus more complex than for LP lamp reactors. The polychromatic bias accounts for the spectral differences in the lamp output, lamp sleeve UV transmittance, water UV transmittance, and differences between the action spectra of the target pathogen and the surrogate microorganism. USEPA (2006) used germicidal outputs of a target pathogen and a surrogate microorganism to determine a correction factor because of the difference between the action spectra. Using the DNA absorbance spectrum instead of the action spectrum of B. subtilis spores may result in inaccurate calculation of the REF for MP lamp UV reactors. An accurate knowledge of the inactivation action spectrum of B. subtilis spores should improve the prediction of MP lamp UV reactor performance, and consequently help in the design and optimization of UV reactors equipped with MP lamps. However, using the ‘standard’ DNA absorbance instead of the action spectrum of some microorganisms such as S. typhimurium LT2 may be acceptable.

3.6.

Effect of inactivation spectra on germicidal fluence (UV dose) determination

4.

In a standard collimated beam UV inactivation study using a polychromatic light source, knowledge of the action spectrum of the microorganism under study is required in order to calculate the germicidal fluence (UV dose) accurately (Bolton and Linden, 2003). Conventionally, the DNA absorbance spectrum has been used as a substitute for the actual microorganism action spectrum, when the latter is not known. The effect of using the DNA absorbance spectrum instead of the actual B. subtilis spore action spectrum for calculating germicidal dose delivery from a MP lamp could be significant.

The relative action spectrum of B. subtilis spores determined in this study was also used to calculate the germicidal fluence (UV dose) from a polychromatic MP UV light source.

Although Mamane-Gravetz et al. (2005) used both k and shoulder coefficients to determine a germicidal fluence (UV dose), the value of k alone can be used to represent the spectral sensitivity of B. subtilis spores because log(nc) was found to be fairly constant and essentially independent of wavelength in this study. For a collimated beam exposure, the determination of germicidal fluence (UV dose) has been standardized based on DNA absorbance (Bolton and Linden, 2003). Using the standardized spreadsheet provided by Bolton (IUVA, 2004), if the germicidally weighted fluence (UV dose) calculated using

Conclusions

The relative inactivation action spectrum of B. subtilis spores determined in this study was found to be statistically different from those published in two previous studies. The difference may have arisen from differences in the source of data or from differences in data analysis procedures. The inactivation action spectrum was significantly different from the relative absorbance spectrum of purified spore DNA, but matched the relative absorbance spectrum of decoated spores fairly well.

This strongly indicates that the photons absorbed by components other than DNA in spores may be involved in the inactivation effect by germicidal UV light, when the absorbance of these components is not negligible. The use of a relative DNA absorbance spectrum as an alternative to the relative action spectrum of B. subtilis spores will result in an inaccurate determination of the weighted germicidal fluence (UV dose) in UV disinfection studies. This, in turn, may result in a systematic error when evaluating the performance of polychromatic UV light reactors using bioassays.

In contrast, the action spectrum of S. typhinium matched very well to the relative absorbance of DNA extracted from that organism. Clearly, in this case, the ‘action’ arises primarily from UV photons absorbed by DNA.

water research 43 (2009) 5087–5096

Acknowledgments The authors gratefully acknowledge the Natural Sciences and Engineering Research Council of Canada (NSERC) and the Canadian Foundation for Innovation (CFI) for providing the necessary funding of this research.

references

Ausuvel, F.M., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A., Struhl, K., 1992. Current Protocols in Molecular Biology. Wiley, New York.

Bolton, J.R., Linden, K.G., 2003. Standardization of methods for fluence (UV dose) determination in bench-scale UV experiments. J. Environ. Eng. 129 (3), 209–215.

Bolton, J.R., Dussert, B., Bukhari, Z., Hargy, T., Clancy, J.L., 1998.

Inactivation of Cryptosporidium parvum by medium-pressure ultraviolet light in finished drinking water. In: Proc. AWWA 1998 Annual Conference, Dallas, TX, vol. A. American Water Works Association, Denver, CO, pp. 389–403.

Cabaj, A., Sommer, R., Pribil, W., Haider, T., 2002. The spectral UV sensitivity of microorganisms used in biodosimetry. Water Sci. Tech.: Water Supply 2 (3), 175–181.

Chang, J.C.H., Ossoff, S.F., Lobe, D.C., Dorfman, M.H., Dumais, C.M., Qualls, R.G., Johnson, J.D., 1985. UV Inactivation of pathogenic and indicator microorganisms. Appl. Environ. Microbiol 49 (6), 1361–1365.

Charlier, M., Helene, C., 1972. Photochemical reactions of aromatic ketones with nucleic acids and their components-III. Chain breakage and thymine dimerization in benzophenone photosensitized DNA. Photochem. Photobiol. Sci. 51, 527–536.

Craik, S.A., Finch, G.R., Bolton, J.R., Belosevic, M., 2000.

Inactivation of Giardia muris cysts using medium-pressure ultraviolet radiation in filtered drinking water. Water Res. 34 (18), 4325–4332.

Craik, S.A., Weldon, D., Finch, G.R., Bolton, J.R., Belosevic, M., 2001. Inactivation of Cryptosporidium parvum oocysts using medium- and low-pressure ultraviolet radiation. Water Res. 35 (6), 1387–1398.

Craik, S.A., Uvbiama, R.D., 2005. Effect of aggregation on UV inactivation of microorganisms in filtered drinking water. In: Proc. Water Quality Technology Conference, Quebec City, QC, Canada; American Water Works Association, Denver, CO. Douki, T., Setlow, B., Setlow, P., 2005. Photosensitization of DNA by dipicolinic acid, a major component of spores of Bacillus species. Photoch. Photobiol. Sci. 4 (8), 591–597.

DVGW, 2006. UV Devices for Disinfection in Drinking Water Supply: Part 1: Requirements on Properties, Function and Operation; Part 2: Testing of Properties, Function and Disinfection Effectiveness; Part 3: Ports and Sensors for the Radiometric Monitoring of UV Disinfection devices – Requirements, Testing and Calibration [Online] Available only in German at. Deutsche Vereinigung des Gas- und Wasserfaches (German Association for Gas and Water), Bonn, Germany. http://www.wvgw.de/index.php?id¼451&submit_ gesamtsuche¼&tx_indexedsearch%5Bsword%5D¼Wþ294.

Gates, F.L., 1929. A study of the bactericidal action of ultra violet high – II. The effect of various environmental factors and conditions. J. Gen. Physiol. 13 (2), 249–260.

Goldstein, S., Rabani, J., 2008. The ferrioxalate and iodideiodate actinometers in the UV region. J. Photochem.

Photobiol. – A – Chem. 193 (1), 50–55.

Harm, M., 1980. Biological Effects of Ultraviolet Radiation.

Cambridge University Press, Westford, MA, USA.

5095

IUVA, 2004. Fluence (UV Dose) calculations for a medium pressure UV lamp. Excel spreadsheet developed by J.R. Bolton and available in the Member Zone of the IUVA Web Site (www. iuva.org).

IUVA, 2005. Fluence (UV Dose) calculations for a low pressure UV lamp, Excel spreadsheet developed by J.R. Bolton and available in the Member Zone of the IUVA Web Site (www.iuva.org).

Koivunen, J., Heinonen-Tanski, H., 2005. Inactivation of enteric microorganisms with chemicla disinfectants, UV irradiation and combined chemical/UV treatments. Water Res. 39, pp.

1519–1526.

Kuhn, H.J., Braslavsky, S.E., Schmidt, R., 2004. Chemical Actinometry. IUPAC Technical Report. Pure Appl. Chem. 76 (12), 2105–2146.

Lamola, A.A., 1970. Triplet photosensitization and the photobiology of thymine dimers in DNA. Pure Appl. Chem. 24, 599–610.

Linden, K.G., Shin, G., Sobsey, M.D., 2001. Comparative effectiveness of UV wavelengths for the inactivation of Cryptosporidium parvum oocysts in water. Water Sci. Technol.

43 (12), 171–174.

Linden, K.G., Johnson, S., Malley, J.P., 2005. Importance of wavelength for UV inactivation of Adenovirus in water.

CDROM. In: Proc. 3rd International Congress on Ultraviolet Technologies, Whistler, BC, Canada, International Ultraviolet Association, P.O. Box 28154, Scottsdale, AZ, 85255.

Mamane-Gravetz, H., Linden, K.G., Cabaj, A., Sommer, R., 2005.

Spectral sensitivity of Bacillus subtilis spores and MS2 Coliphage for validation testing of ultraviolet reactors for water disinfection. Environ. Sci. Technol. 39 (20), 7845–7852.

Mason, J.M., Setlow, P., 1986. Essential role of small, acid-soluble spore proteins in resistance of Bacillus subtilis spores to UVLight. J. Bacteriol. 167 (1), 174–178.

Maier, R.M., Pepper, J.L., Gerba, C.P., 2000. Environmental Microbiology. Academic, San Diego, CA. ¨ NORM, 2001. Austrian National Standard: O ¨ NORM M 5873-1 E. O 2001. Plants for Disinfection of Water using Ultraviolet Radiation: Requirements and Testing, Part 1: Low Pressure Mercury Lamp Plants. Available from:. Austrian Standards Institute, Vienna, Austria www.on-norm.at. ¨ NORM, 2003. Austrian National Standard: O ¨ NORM M 5873-2 E. O 2003. Plants for disinfection of water using ultraviolet radiation: requirements and testing, Part 2: Medium pressure mercury lamp plants. Available from:. Austrian Standards Institute, Vienna, Austria www.on-norm.at.

Pepper, I.L., Gerba, C.P., Brendecke, J.W., 1995. Environmental Microbiology: a Laboratory Manual. Academic, San Diego, CA. Rauth, A.M., 1965. Physical state of viral nucleic acid and sensitivity of viruses to ultraviolet light. Biophys. J. 5 (3), 257–268.

Riesenman, P.J., Nicholson, W.L., 2000. Role of the spore coat layers in Bacillus subtilis spore resistance to hydrogen peroxide, artificial UV-C, UV-B, and solar UV radiation. Appl.

Environ. Microbiol. 66 (2), 620–626.

Setlow, P., 2001. Resistance of spores of Bacillus species to ultraviolet light. Environ. Mol. Mutagen. 38 (2–3), 97–104.

Setlow, B., Xu, D., Setlow, P., 1992. Interaction between DNA and a/ b type small, acid-soluble spore proteins: a new class of DNA-binding protein. J. Bacteriol. 174 (7), 2312–2322.

Severin, B.F., Suidan, M.T., Engelbrecht, R.S., 1983. Kinetic modeling of UV disinfection of water. Water Res. 17 (11), 1669–1678.

Slieman, T.A., Nicholson, W.L., 2000. Artificial and solar UV radiation induces strand breaks and cyclobutane pyrimidine dimers in Bacillus subtilis spore DNA. Appl. Environ. Microbiol.

66 (1), 199–205.

Smith, K.C., Hodgkins, B., Oleary, M.E., 1966. Biological importance of ultraviolet light induced DNA-protein crosslinks in Escherichia coli 15 Tau. Biochim. Biophys. Acta 114 (1), 1–8.

5096

water research 43 (2009) 5087–5096

Sommer, R., Cabaj, A., 1993. Evaluation of the efficiency of a UV plant for drinking water disinfection. Water Sci. Technol. 27 (3/4), 357–362.

Tosa, K. and Hirata, T., 1998. HRWM-39: Photoreactivation of Salmonella following UV disinfection. In: IAWQ 19th Biennial International Conference, Vancouver, Canada.

Tsuboi, K.K., 1950. Mouse liver nucleic acids. II. Ultraviolet absorption studies. Biochim. Biophys. Acta 6, 202–209.

USEPA, 2006. Ultraviolet Disinfection Guidance Manual, EPA 815D-03–007. Available online at. Office of Water, Washington DC, USA. http://www.epa.gov/safewater/disinfection/lt2/pdfs/ guide_lt2_uvguidance.pdf.

Uvbiama, R.D., 2006. The effect of upstream treatment processes on UV inactivation of microorganisms in filtered drinking water. MSc. thesis, Department of Civil and Environmental Engineering, University of Alberta, Edmonton, AB, Canada.

Uvbiama, R.D., Craik, S.A., 2005. The effect of upstream treatment processes on UV inactivation of microorganisms in filtered water. CDROM. In: Proc. Third International Congress on Ultraviolet Technologies, Whistler BC, Canada, International Ultraviolet Association, P.O. Box 28154, Scottsdale, AZ, 85255. von Sonntag, C., Schuchmann, H.-P., 1992. UV disinfection of drinking water and by-product formation – some basic considerations. J. Water SRT-Aqua 41 (2), 67–74.