w a t e r r e s e a r c h 5 6 ( 2 0 1 4 ) 1 8 1 e1 8 9
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/watres
Evaluation of DNA damage reversal during medium-pressure UV disinfection Christopher Poepping a, Sara E. Beck a, Harold Wright b, Karl G. Linden a,* a b
Department of Civil, Environmental, and Architectural Engineering, University of Colorado Boulder, CO, USA Carollo Engineers, Boise, ID, USA
article info
abstract
Article history:
Ultraviolet (UV) disinfection relies on the principal that DNA exposure to UV irradiation
Received 25 November 2013
leads to the formation of cytotoxic lesions resulting in the inactivation of microorganisms.
Received in revised form
Cyclobutane pyrimdine dimers (CPDs) account for the majority of DNA lesions upon UV
20 February 2014
exposure. Past research has demonstrated reversal of CPDs in extracted DNA formed at
Accepted 21 February 2014
high UV-C wavelength irradiation (280 nm) upon subsequent irradiation at lower UVC
Available online 5 March 2014
wavelengths (230e240 nm). Medium-pressure (MP) UV lamps produce a polychromatic emission giving rise to the possibility that cellular DNA in a target pathogen may undergo
Keywords:
simultaneous damage and repair when exposed to multiple wavelengths during the
Ultraviolet light
disinfection process, decreasing the efficiency of MP UV lamp disinfection. Culture tech-
Medium-pressure UV
niques and a quantitative polymerase chain reaction (qPCR) assay were used to examine
DNA damage
cell viability and DNA damage reversal. qPCR results indicated direct photoreversal of UV-
Cyclobutane pyrimidine dimer
induced DNA damage through sequential irradiations of 280 nm followed by 228 nm in
Photoreversal
Escherichia coli DNA. However, significant photoreversal was only observed after high initial doses and secondary doses of UV light. The doses where significant photoreversal took place were more than 10 times higher than those typically used in UV disinfection. Despite evidence of CPD photoreversal, bacterial growth assays showed no indication that sequential-wavelength irradiations result in higher survival rates than single-wavelength irradiations. ª 2014 Elsevier Ltd. All rights reserved.
1.
Introduction
Ultraviolet (UV) light is a proven and effective method for the inactivation of microorganisms. Consequently, it has become a widely used technology in water and wastewater treatment facilities. The benefits of UV are well known throughout the water industry; however, increased use of UV disinfection and technological advancement has spurred an interest for further research concerning process challenges and
* Corresponding author. Tel.: þ1 3034924798; fax: þ1 3034927317. E-mail address: [email protected] (K.G. Linden). http://dx.doi.org/10.1016/j.watres.2014.02.043 0043-1354/ª 2014 Elsevier Ltd. All rights reserved.
optimization. The foundation of UV disinfection lies in the ability of UV light to induce damage to DNA leading to inhibition of vital cellular processes such as transcription and replication and ultimately lead to the inactivation of the organism (Mone´ et al., 2011; Sinha and Ha¨der, 2002; Rodrı´guez et al., 2013). DNA strongly absorbs UV-C (200e280 nm) with a relative peak at 260 nm and the absorption can lead to the formation of lesions between neighboring nucleobases, primarily pyrimidines (Douki, 2013; Eischeid and Linden, 2007; Rastogi et al., 2010). Two primary types of lesions may form:
182
w a t e r r e s e a r c h 5 6 ( 2 0 1 4 ) 1 8 1 e1 8 9
cyclobutane pyrimidine dimers (CPDs) and 6-4 photoproducts. CPDs are characterized as four-membered ring structures formed between the C5 and C6 atoms on each pyrimidine base and 6e4 photoproducts are characterized by an interaction between one carbon on each base (Douki, 2013; Rastogi et al., 2010). CPDs represent the majority of lesions induced by UV-C light accounting for about 75% of lesions formed while the remaining 25% are 6e4 photoproducts (Sinha and Ha¨der, 2002). Many microbes have evolved mechanisms to repair certain UV-induced lesions, including CPDs and 6-4 photoproducts, creating a potential concern for UV disinfection processes (Thoma, 1999). These mechanisms have been studied extensively and are well understood from a molecular point-of-view. Briefly, there are two principal categories of DNA damage repair: photoreactivation and excision repair. Photoreactivation is an enzyme-mediated mechanism stimulated by the exposure to visible, near-UV light. Excision repair is also an enzyme-mediated mechanism, but does not require visible light (Thoma, 1999; Sinha and Ha¨der, 2002). In addition to enzyme-mediated DNA repair mechanisms, there has been evidence that UV-induced lesions can be directly cleaved through subsequent UV-C exposure (Setlow and Setlow, 1962; Pan et al., 2012). Past studies utilizing isolated DNA observed that CPDs formed via irradiation at 280 nm can be directly cleaved, or split into their original components, with subsequent irradiation from lower wavelength UV-C irradiation near 230e240 nm. Setlow and Setlow (1962) used this concept to investigate the role of thymine dimers in the biological damage to microbes by UV light. Other studies used direct photocleavage to further investigate the prevalence of thymine dimers in UV-induced damage or reported on the phenomenon using absorbance measurements or paper chromatogaraphy (Johns et al., 1962; Setlow and Setlow, 1962; Setlow and Carrier, 1963; Johns et al., 1964). Recently, Pan et al. (2012) used the concept of direct photocleavage from sequential irradiations to investigate the effect of neighboring purines on CPD formation with the principal method of detection being absorption. It is important to note that direct photocleavage of CPDs is generally observed at higher than typical UV disinfection doses. For example, the minimum initial dose at 280 nm required to observe DNA damage reversal through subsequent irradiation at wavelengths below 240 nm is at least 100 mJ/cm2 (Setlow and Setlow, 1962). Furthermore, the degree of photoreversal increases with increasing lower wavelength irradiation (Setlow and Setlow, 1962). Medium-pressure (MP) UV lamps produce an emission spectrum spanning the 200e400 nm UV range and into the visible range, resulting in simultaneous exposure to multiple wavelengths. In contrast, low-pressure (LP) UV lamps have a monochromatic UV emission at 253.7 nm. Fig. 1 illustrates the polychromatic and monochromatic emission of MP and LP lamps, respectively. For UV systems utilizing LP UV lamps, direct photocleavage of CPDs would not be a concern; however, the polychromatic emission of MP lamps simultaneously exposes DNA to multiple wavelengths, including those below 240 nm. Therefore, there exists the opportunity for reversal of DNA damage through direct photocleavage of UV-induced CPDs during MP UV irradiation. MP UV lamps are becoming increasingly popular, as they exhibit increased effectiveness
Fig. 1 e Emission spectra of LP and MP lamps.
for the inactivation of adenovirus (Linden et al., 2007). Furthermore, MP UV irradiation has been shown to limit the degree of photoreactivation seen after UV exposure in bacteria compared to LP UV irradiation (Zimmer and Slawson, 2002). The broad band emission characteristic of the MP UV lamp has been shown to cause damage to DNA other than CPDs and 6-4 photoproducts, including Dewar isomers and single-strand breaks (Cadet et al., 2005). While there have been numerous studies examining the phenomenon of direct photocleavage of CPDs, there is a lack of research regarding this concept in the framework of UV disinfection which is founded on the capability of UV light to induce CPDs, primarily, in microbial DNA. Herein, direct photocleavage of UV-induced CPDs is examined in the context of the MP UV disinfection process. Specifically, the research aimed to 1) determine if direct photoreversal occurs in cellular DNA (previous research has used isolated DNA in the irradiations) and to 2) determine if the amount of high and lowwavelength UV-C typically present in MP UV doses results in direct photocleavage. These questions are addressed using conventional culture techniques to examine cell viability and quantitative polymerase chain reaction (qPCR) to quantify damage to the bacterial genome. Since cell viability can be affected by any UV-induced damage, this dual approach provides fundamental insight into whether or not direct CPD photocleavage can account for enough DNA damage reversal to return DNA to a viable state in Escherichia coli.
2.
Materials and methods
2.1.
UV Exposures and sample preparation
A bench-scale, collimated beam apparatus equipped with a 1kW MP UV lamp (Calgon Carbon Corporation Inc., Pittsburgh, PA, USA) was used for the exposures. For the wavelengthspecific exposures, the lamp was outfitted with 280 and 228 nm bandpass filters (Andover Corporation, Salem, NH, USA), which had a full width at half maximum (FWHM) of 10e12 nm and emission spectra as illustrated in Fig. 2. The emission spectra for the LP and MP sources (Fig. 1) and the bandpass filters (Fig. 2) were measured with an Ocean Optics
183
w a t e r r e s e a r c h 5 6 ( 2 0 1 4 ) 1 8 1 e1 8 9
irradiations, resulting in a sample depth of w0.5 cm. The samples were continuously stirred using sterile magnetic stir bars throughout the entire exposure. After exposures, 200 mL of each sample was collected for DNA damage analysis.
2.2.2.
Design 2
Samples were exposed initially to UV light at 280 nm up to 100 mJ/cm2 in increments of 10 mJ/cm2. Sequentialwavelength exposures were followed by 10 mJ/cm2 of UV light at 228 nm (Table 2). After the initial 280 nm wavelength exposure, 200 mL was collected for molecular analysis and the remaining sample was subsequently exposed to 10 mJ/cm2 of 228 nm light. 200 mL was again collected for analysis following completion of the exposure. Fig. 2 e Emission spectra of MP UV light through 228 nm and 280 nm bandpass filters.
Maya 2000 Pro spectrometer (Dunedin, FL). Incident irradiance at the sample surface was measured using a calibrated IL-1700 radiometer complete with a SED 240 detector and W-diffuser (International Light, Peabody, MA, USA) with calibration factors specific to each bandpass filter emission. Exposure time for each UV dose was calculated as described in Bolton and Linden (2003)). Briefly, sample absorbance was measured using a Cary 100 Bio Spectrophotometer (Agilent Technology, Santa Clara, CA, USA) and used to calculate average irradiance for a continuously stirred batch system. Average irradiance was divided by target UV dose to obtain specific exposure times. Nonhomogeneous lamp emission was accounted for using the Petri factor (Bolton and Linden, 2003). Ampicillin/streptomycin resistant E. coli (GAP Environmental Services, London, Ontario, Canada) was grown to late log phase in Tryptic Soy Broth (TSB) containing ampicillin/ streptomycin. One mL of the solution was washed 3 times with 1 phosphate-buffered saline (PBS) and diluted in PBS to a target concentration of 106 colony forming units (CFU)/mL. The exposure procedure varied slightly depending on the specific experimental design as discussed below.
2.2.
Design 3
For Design 3, UV doses for 280 nm and 228 nm were calculated to realistically represent the amount of UV irradiation during MP UV irradiation that is attributed to the ranges 275e285 nm and 225e235 nm, respectively (Table 3). The percent of the total average irradiance that is attributed to the ranges 275e285 nm and 225e235 nm was determined. The range of 275e285 nm accounted for approximately 10% of the total average irradiance and the range of 225e235 nm accounted for 6e7% of the average irradiance. The percent values were then used to find the amount of the total MP dose in mJ/cm2 these ranges accounted for by multiplying the percent values by the representative MP dose. For clarification, the representative MP dose was multiplied by w10% to get the 280 nm and w6e7% to get the 228 nm dose. MP UV doses up to 1000 mJ/cm2 were considered in increments of 100 mJ/cm2. Note that as described in Section 3.3, it was not possible to simultaneously expose the samples to both 228 and 280 nm irradiation, as would be the case for a full spectrum MP UV lamp. The percent of total dose for each wavelength range varied slightly (0.05%) for each exposure as the sample absorbance varied slightly between experiments. Collection process followed same pattern as Design 2: 200 mL was collected after initial 280 nm exposure and after subsequent exposure at 228 nm.
Experimental designs
Experiments were performed using MP UV light with 280 and 228 nm bandpass filters. UV doses of sequential irradiations were chosen to examine a variety of initial 280 nm doses and subsequent 228 nm doses.
2.2.1.
2.2.3.
Design 1
UV doses for Design 1 followed a general pattern of an initial dose of 120 mJ/cm2 at either 280 nm or 228 nm (Table 1). For single-wavelength irradiations, the sample was then exposed to UV fluences of 30, 60, 90, and 120 mJ/cm2 at the same wavelength as the initial dose. For the sequential-wavelength irradiations, the samples were initially exposed to 280 nm light followed by 30, 60, 90, and 120 mJ/cm2 of 228 nm light. During the initial dose, 25 mL of sample was exposed in a 60 mm diameter petri dish at 120 mJ/cm2, resulting in sample depth of w0.9 cm. Of the initial sample, 5 mL volumes in a 35 mm diameter petri dish were used for subsequent
Table 1 e UV dose designations for Design 1. Sample
1 2 3 4 5 280 280 280 280 280 230 230 230 230 230
Control Control Control Control Control Control Control Control Control Control
Initial 280 nm dose (mJ/cm2)
1 2 3 4 5 1 2 3 4 5
120 120 120 120 120 120 120 120 120 120 0 0 0 0 0
Subsequent Subsequent 280 nm 228 nm dose (mJ/cm2) dose (mJ/cm2) 0 0 0 0 0 0 30 60 90 120 0 0 0 0 0
0 30 60 90 120 0 0 0 0 0 120 150 180 210 240
184
w a t e r r e s e a r c h 5 6 ( 2 0 1 4 ) 1 8 1 e1 8 9
Table 2 e UV dose designations for Design 2. Sample
Initial 280 nm dose (mJ/cm2)
228 nm Dose (mJ/cm2)
10 20 30 40 50 60 70 80 90 100 10 20 30 40 50 60 70 80 90 100
0 0 0 0 0 0 0 0 0 0 10 10 10 10 10 10 10 10 10 10
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
2.3.
E. coli plate counting exposures and enumeration
UV doses were determined via the same method as Design 3 to represent the amount of radiation the sample would receive from 280 nm to 228 nm in MP exposures but used lower target MP UV doses of 50, 100, and 150 mJ/cm2. Briefly, 20 mL of sample in a 60 mm diameter petri dish was initially exposed to 280 nm dose and 1 mL of the sample was collected for analysis. The remaining sample was subsequently exposed to 228 nm light for the specified dose and 1 mL was collected upon completion. 20 mL of sample in a 60 mm diameter petri
Table 3 e UV dose designations for Design 3. Sample
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Representative MP dose (mJ/cm2)
Corresponding 280 nm dose (mJ/cm2)
Corresponding 228 nm dose (mJ/cm2)
100 200 300 400 500 600 700 800 900 1000 100 200 300 400 500 600 700 800 900 1000
10 20 30 40 50 60 70 80 90 100 10 20 30 40 50 60 70 80 90 100
0 0 0 0 0 0 0 0 0 0 6.5 13.0 19.5 26.0 32.5 39.0 45.5 52.0 58.5 65.0
dish was also used for single-wavelength 228 nm exposures with 1 mL of sample collected for analysis. Tryptic Soy Agar (TSA) plates containing streptomycin/ ampicillin were prepared and allowed to solidify overnight under refrigeration before plating. Samples were diluted in sterile 1 PBS and 100 mL of sample was pipetted to surface of TSA plates and spread using a sterile L-shaped glass stick. All plating was completed the same day as the exposure, with less than 1 h between exposure and plating. Each sample was plated in triplicate. Plates were allowed to incubate at 37 C overnight for 18e24 h. The volume of sample used and the dilution factor were taken into account when determining CFU counts. Log inactivation was calculated as Log (N0/N) where N is CFU/mL for the sample and N0 is CFU/mL for the untreated control.
2.4. Real-time quantitative polymerase chain reaction (qPCR) DNA was extracted using DNeasy Blood and Tissue Kit (Qiagen, Valencia, CA). Primers and probe (Table 4) were developed as described in Rudi et al. (2010) and synthesized by Integrated DNA Technologies (Coralville, IA). The fragment size was 1504 base pairs. Primers and probe were synthesized by International DNA Technologies. Each sample was prepared with 7.5 mM of forward and reverse primer, PrimeTime probe, nuclease-free water, master mix (Promega, Madison, WI) consisting of 2 units of GoTaq polymerase and 1 GoTaq Buffer, and 2 mL of DNA solution for a total volume of 25 mL. The qPCR samples were performed in duplicate per the program outlined in Table 5 using a MJ MiniOpticon Real-Time PCR machine (Bio-Rad, Hercules, CA). The thermocycle for the PCR assay is shown in Table 5. Similar qPCR methods have been utilized in previous research for the purpose of UV-induced DNA damage quantitation (Rudi et al., 2010). Briefly, the flourogenic probe is designed to attach to a specific DNA sequence within the amplified gene bracketed by the two primers. When not attached, the probe exhibits fluorescence quenching; however, upon attachment to the DNA, the quenching properties of the probe are degraded (Frahm and Obst, 2003). An online sensor in the PCR machine monitors fluorescence levels and the quantitation is based on the number of cycles necessary to reach a fluorescence threshold. DNA quantification was determined using a dilution curve. A standard curve representative of the amount of amplifiable DNA was calculated from the number of PCR cycles required to reach the fluorescence threshold for the dilution series. Log
Table 4 e Primer and probe used for qPCR. Primer
Sequence
Forward AAGAGTTTGATCATGGCTCA Primer Reverse CGGTTACCTTGTTACGACTT Primer Probe CGTATTACCGCGGCTGCTGGCAC a
Genome Fragment Regiona size (bp) 42
1504
1546
Position relative to E. coli 16S rRNA with respect to 50 end.
w a t e r r e s e a r c h 5 6 ( 2 0 1 4 ) 1 8 1 e1 8 9
Table 5 e qPCR Thermocycle Program. Step Denaturation Annealing Synthesis
Temp ( C)
Time (s)
95 55 72
30 30 90
reduction in DNA was calculated as log (N0/N), where N is amount of amplifiable DNA in the treated sample and N0 is the amount of DNA in the untreated control.
2.5.
Statistical analysis
Quantification analyses and calculations were performed in Microsoft Excel. Statistical analyses including Analysis of Variance (ANOVA) and confidence intervals were conducted using MiniTab. Data points indicate average values from experimental trials; however, regressions were conducted using data from all samples.
3.
Results and discussion
3.1.
PCR Assays
3.1.1.
Design 1
Fig. 3 shows the results of sequential-wavelength irradiations and the single-wavelength control irradiations. For the 280 nm 9
Log Reduction in Amplifiable DNA
8 7 6 5 4 3 280 nm/228 nm
2
280 Controls 228 Controls
1 0
0
30 60 90 Dose Following Initial 120 mJ/cm Irradiation (mJ/cm )
120
Fig. 3 e Log reduction of amplifiable DNA extracted from E. coli at various dosing conditions. The x-axis shows the subsequent dose of UV light applied after an initial dose of 120 mJ/cm2 at the specific wavelength. The y-axis is the log reduction values for amplifiable DNA. Each data point represents the average of the sample size (n) for each condition and error bars are shown to demonstrate 80% confidence intervals. Closed circle: sample initially exposed to 120 mJ/cm2 at 280 nm followed by doses of 228 nm shown on x-axis (n [ 4). Open square: samples were initially exposed to 120 mJ/cm2 at 280 nm followed by 280 nm doses shown on x-axis (n [ 2). Open triangle: samples were initially exposed to 120 mJ/cm2 at 228 nm followed by more 228 nm light at the doses shown on the x-axis (n [ 2).
185
and 228 nm single-wavelength exposures, amplifiable DNA decreases with increasing dose for both wavelengths. The 280 nm irradiation demonstrated greater DNA damage after 120 mJ/cm2 than an equivalent dose at 228 nm, which follows the observation that w280 nm is more germicidally effective than w230 nm (Wang et al., 2005). For sequential-wavelength exposures, 280 nm followed by 228 nm irradiation resulted in increased amplifiable DNA following an initial 120 mJ/cm2 dose of 280 nm irradiation. An increase in amplifiable DNA indicates photocleavage of the CPD. The subsequent dose of 120 mJ/cm2 of 228 nm light following an initial dose of 120 mJ/ cm2 at 280 nm provided photoreversal amounting to 1.63 log. PCR results obtained from Design 1 exposures clearly demonstrated DNA damage reversal in samples exposed sequentially to irradiation from 280 nm followed by 228 nm. There was a significant decrease in the log reduction of amplifiable DNA observed from the initial dose 120 mJ/cm2 at 280 nm with each increased dose of 228 nm irradiation. Linear regression for sequential-wavelength irradiations showed a negative slope of 0.012 between log reduction and increasing 228 nm exposure (p-value of slope:
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/watres
Evaluation of DNA damage reversal during medium-pressure UV disinfection Christopher Poepping a, Sara E. Beck a, Harold Wright b, Karl G. Linden a,* a b
Department of Civil, Environmental, and Architectural Engineering, University of Colorado Boulder, CO, USA Carollo Engineers, Boise, ID, USA
article info
abstract
Article history:
Ultraviolet (UV) disinfection relies on the principal that DNA exposure to UV irradiation
Received 25 November 2013
leads to the formation of cytotoxic lesions resulting in the inactivation of microorganisms.
Received in revised form
Cyclobutane pyrimdine dimers (CPDs) account for the majority of DNA lesions upon UV
20 February 2014
exposure. Past research has demonstrated reversal of CPDs in extracted DNA formed at
Accepted 21 February 2014
high UV-C wavelength irradiation (280 nm) upon subsequent irradiation at lower UVC
Available online 5 March 2014
wavelengths (230e240 nm). Medium-pressure (MP) UV lamps produce a polychromatic emission giving rise to the possibility that cellular DNA in a target pathogen may undergo
Keywords:
simultaneous damage and repair when exposed to multiple wavelengths during the
Ultraviolet light
disinfection process, decreasing the efficiency of MP UV lamp disinfection. Culture tech-
Medium-pressure UV
niques and a quantitative polymerase chain reaction (qPCR) assay were used to examine
DNA damage
cell viability and DNA damage reversal. qPCR results indicated direct photoreversal of UV-
Cyclobutane pyrimidine dimer
induced DNA damage through sequential irradiations of 280 nm followed by 228 nm in
Photoreversal
Escherichia coli DNA. However, significant photoreversal was only observed after high initial doses and secondary doses of UV light. The doses where significant photoreversal took place were more than 10 times higher than those typically used in UV disinfection. Despite evidence of CPD photoreversal, bacterial growth assays showed no indication that sequential-wavelength irradiations result in higher survival rates than single-wavelength irradiations. ª 2014 Elsevier Ltd. All rights reserved.
1.
Introduction
Ultraviolet (UV) light is a proven and effective method for the inactivation of microorganisms. Consequently, it has become a widely used technology in water and wastewater treatment facilities. The benefits of UV are well known throughout the water industry; however, increased use of UV disinfection and technological advancement has spurred an interest for further research concerning process challenges and
* Corresponding author. Tel.: þ1 3034924798; fax: þ1 3034927317. E-mail address: [email protected] (K.G. Linden). http://dx.doi.org/10.1016/j.watres.2014.02.043 0043-1354/ª 2014 Elsevier Ltd. All rights reserved.
optimization. The foundation of UV disinfection lies in the ability of UV light to induce damage to DNA leading to inhibition of vital cellular processes such as transcription and replication and ultimately lead to the inactivation of the organism (Mone´ et al., 2011; Sinha and Ha¨der, 2002; Rodrı´guez et al., 2013). DNA strongly absorbs UV-C (200e280 nm) with a relative peak at 260 nm and the absorption can lead to the formation of lesions between neighboring nucleobases, primarily pyrimidines (Douki, 2013; Eischeid and Linden, 2007; Rastogi et al., 2010). Two primary types of lesions may form:
182
w a t e r r e s e a r c h 5 6 ( 2 0 1 4 ) 1 8 1 e1 8 9
cyclobutane pyrimidine dimers (CPDs) and 6-4 photoproducts. CPDs are characterized as four-membered ring structures formed between the C5 and C6 atoms on each pyrimidine base and 6e4 photoproducts are characterized by an interaction between one carbon on each base (Douki, 2013; Rastogi et al., 2010). CPDs represent the majority of lesions induced by UV-C light accounting for about 75% of lesions formed while the remaining 25% are 6e4 photoproducts (Sinha and Ha¨der, 2002). Many microbes have evolved mechanisms to repair certain UV-induced lesions, including CPDs and 6-4 photoproducts, creating a potential concern for UV disinfection processes (Thoma, 1999). These mechanisms have been studied extensively and are well understood from a molecular point-of-view. Briefly, there are two principal categories of DNA damage repair: photoreactivation and excision repair. Photoreactivation is an enzyme-mediated mechanism stimulated by the exposure to visible, near-UV light. Excision repair is also an enzyme-mediated mechanism, but does not require visible light (Thoma, 1999; Sinha and Ha¨der, 2002). In addition to enzyme-mediated DNA repair mechanisms, there has been evidence that UV-induced lesions can be directly cleaved through subsequent UV-C exposure (Setlow and Setlow, 1962; Pan et al., 2012). Past studies utilizing isolated DNA observed that CPDs formed via irradiation at 280 nm can be directly cleaved, or split into their original components, with subsequent irradiation from lower wavelength UV-C irradiation near 230e240 nm. Setlow and Setlow (1962) used this concept to investigate the role of thymine dimers in the biological damage to microbes by UV light. Other studies used direct photocleavage to further investigate the prevalence of thymine dimers in UV-induced damage or reported on the phenomenon using absorbance measurements or paper chromatogaraphy (Johns et al., 1962; Setlow and Setlow, 1962; Setlow and Carrier, 1963; Johns et al., 1964). Recently, Pan et al. (2012) used the concept of direct photocleavage from sequential irradiations to investigate the effect of neighboring purines on CPD formation with the principal method of detection being absorption. It is important to note that direct photocleavage of CPDs is generally observed at higher than typical UV disinfection doses. For example, the minimum initial dose at 280 nm required to observe DNA damage reversal through subsequent irradiation at wavelengths below 240 nm is at least 100 mJ/cm2 (Setlow and Setlow, 1962). Furthermore, the degree of photoreversal increases with increasing lower wavelength irradiation (Setlow and Setlow, 1962). Medium-pressure (MP) UV lamps produce an emission spectrum spanning the 200e400 nm UV range and into the visible range, resulting in simultaneous exposure to multiple wavelengths. In contrast, low-pressure (LP) UV lamps have a monochromatic UV emission at 253.7 nm. Fig. 1 illustrates the polychromatic and monochromatic emission of MP and LP lamps, respectively. For UV systems utilizing LP UV lamps, direct photocleavage of CPDs would not be a concern; however, the polychromatic emission of MP lamps simultaneously exposes DNA to multiple wavelengths, including those below 240 nm. Therefore, there exists the opportunity for reversal of DNA damage through direct photocleavage of UV-induced CPDs during MP UV irradiation. MP UV lamps are becoming increasingly popular, as they exhibit increased effectiveness
Fig. 1 e Emission spectra of LP and MP lamps.
for the inactivation of adenovirus (Linden et al., 2007). Furthermore, MP UV irradiation has been shown to limit the degree of photoreactivation seen after UV exposure in bacteria compared to LP UV irradiation (Zimmer and Slawson, 2002). The broad band emission characteristic of the MP UV lamp has been shown to cause damage to DNA other than CPDs and 6-4 photoproducts, including Dewar isomers and single-strand breaks (Cadet et al., 2005). While there have been numerous studies examining the phenomenon of direct photocleavage of CPDs, there is a lack of research regarding this concept in the framework of UV disinfection which is founded on the capability of UV light to induce CPDs, primarily, in microbial DNA. Herein, direct photocleavage of UV-induced CPDs is examined in the context of the MP UV disinfection process. Specifically, the research aimed to 1) determine if direct photoreversal occurs in cellular DNA (previous research has used isolated DNA in the irradiations) and to 2) determine if the amount of high and lowwavelength UV-C typically present in MP UV doses results in direct photocleavage. These questions are addressed using conventional culture techniques to examine cell viability and quantitative polymerase chain reaction (qPCR) to quantify damage to the bacterial genome. Since cell viability can be affected by any UV-induced damage, this dual approach provides fundamental insight into whether or not direct CPD photocleavage can account for enough DNA damage reversal to return DNA to a viable state in Escherichia coli.
2.
Materials and methods
2.1.
UV Exposures and sample preparation
A bench-scale, collimated beam apparatus equipped with a 1kW MP UV lamp (Calgon Carbon Corporation Inc., Pittsburgh, PA, USA) was used for the exposures. For the wavelengthspecific exposures, the lamp was outfitted with 280 and 228 nm bandpass filters (Andover Corporation, Salem, NH, USA), which had a full width at half maximum (FWHM) of 10e12 nm and emission spectra as illustrated in Fig. 2. The emission spectra for the LP and MP sources (Fig. 1) and the bandpass filters (Fig. 2) were measured with an Ocean Optics
183
w a t e r r e s e a r c h 5 6 ( 2 0 1 4 ) 1 8 1 e1 8 9
irradiations, resulting in a sample depth of w0.5 cm. The samples were continuously stirred using sterile magnetic stir bars throughout the entire exposure. After exposures, 200 mL of each sample was collected for DNA damage analysis.
2.2.2.
Design 2
Samples were exposed initially to UV light at 280 nm up to 100 mJ/cm2 in increments of 10 mJ/cm2. Sequentialwavelength exposures were followed by 10 mJ/cm2 of UV light at 228 nm (Table 2). After the initial 280 nm wavelength exposure, 200 mL was collected for molecular analysis and the remaining sample was subsequently exposed to 10 mJ/cm2 of 228 nm light. 200 mL was again collected for analysis following completion of the exposure. Fig. 2 e Emission spectra of MP UV light through 228 nm and 280 nm bandpass filters.
Maya 2000 Pro spectrometer (Dunedin, FL). Incident irradiance at the sample surface was measured using a calibrated IL-1700 radiometer complete with a SED 240 detector and W-diffuser (International Light, Peabody, MA, USA) with calibration factors specific to each bandpass filter emission. Exposure time for each UV dose was calculated as described in Bolton and Linden (2003)). Briefly, sample absorbance was measured using a Cary 100 Bio Spectrophotometer (Agilent Technology, Santa Clara, CA, USA) and used to calculate average irradiance for a continuously stirred batch system. Average irradiance was divided by target UV dose to obtain specific exposure times. Nonhomogeneous lamp emission was accounted for using the Petri factor (Bolton and Linden, 2003). Ampicillin/streptomycin resistant E. coli (GAP Environmental Services, London, Ontario, Canada) was grown to late log phase in Tryptic Soy Broth (TSB) containing ampicillin/ streptomycin. One mL of the solution was washed 3 times with 1 phosphate-buffered saline (PBS) and diluted in PBS to a target concentration of 106 colony forming units (CFU)/mL. The exposure procedure varied slightly depending on the specific experimental design as discussed below.
2.2.
Design 3
For Design 3, UV doses for 280 nm and 228 nm were calculated to realistically represent the amount of UV irradiation during MP UV irradiation that is attributed to the ranges 275e285 nm and 225e235 nm, respectively (Table 3). The percent of the total average irradiance that is attributed to the ranges 275e285 nm and 225e235 nm was determined. The range of 275e285 nm accounted for approximately 10% of the total average irradiance and the range of 225e235 nm accounted for 6e7% of the average irradiance. The percent values were then used to find the amount of the total MP dose in mJ/cm2 these ranges accounted for by multiplying the percent values by the representative MP dose. For clarification, the representative MP dose was multiplied by w10% to get the 280 nm and w6e7% to get the 228 nm dose. MP UV doses up to 1000 mJ/cm2 were considered in increments of 100 mJ/cm2. Note that as described in Section 3.3, it was not possible to simultaneously expose the samples to both 228 and 280 nm irradiation, as would be the case for a full spectrum MP UV lamp. The percent of total dose for each wavelength range varied slightly (0.05%) for each exposure as the sample absorbance varied slightly between experiments. Collection process followed same pattern as Design 2: 200 mL was collected after initial 280 nm exposure and after subsequent exposure at 228 nm.
Experimental designs
Experiments were performed using MP UV light with 280 and 228 nm bandpass filters. UV doses of sequential irradiations were chosen to examine a variety of initial 280 nm doses and subsequent 228 nm doses.
2.2.1.
2.2.3.
Design 1
UV doses for Design 1 followed a general pattern of an initial dose of 120 mJ/cm2 at either 280 nm or 228 nm (Table 1). For single-wavelength irradiations, the sample was then exposed to UV fluences of 30, 60, 90, and 120 mJ/cm2 at the same wavelength as the initial dose. For the sequential-wavelength irradiations, the samples were initially exposed to 280 nm light followed by 30, 60, 90, and 120 mJ/cm2 of 228 nm light. During the initial dose, 25 mL of sample was exposed in a 60 mm diameter petri dish at 120 mJ/cm2, resulting in sample depth of w0.9 cm. Of the initial sample, 5 mL volumes in a 35 mm diameter petri dish were used for subsequent
Table 1 e UV dose designations for Design 1. Sample
1 2 3 4 5 280 280 280 280 280 230 230 230 230 230
Control Control Control Control Control Control Control Control Control Control
Initial 280 nm dose (mJ/cm2)
1 2 3 4 5 1 2 3 4 5
120 120 120 120 120 120 120 120 120 120 0 0 0 0 0
Subsequent Subsequent 280 nm 228 nm dose (mJ/cm2) dose (mJ/cm2) 0 0 0 0 0 0 30 60 90 120 0 0 0 0 0
0 30 60 90 120 0 0 0 0 0 120 150 180 210 240
184
w a t e r r e s e a r c h 5 6 ( 2 0 1 4 ) 1 8 1 e1 8 9
Table 2 e UV dose designations for Design 2. Sample
Initial 280 nm dose (mJ/cm2)
228 nm Dose (mJ/cm2)
10 20 30 40 50 60 70 80 90 100 10 20 30 40 50 60 70 80 90 100
0 0 0 0 0 0 0 0 0 0 10 10 10 10 10 10 10 10 10 10
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
2.3.
E. coli plate counting exposures and enumeration
UV doses were determined via the same method as Design 3 to represent the amount of radiation the sample would receive from 280 nm to 228 nm in MP exposures but used lower target MP UV doses of 50, 100, and 150 mJ/cm2. Briefly, 20 mL of sample in a 60 mm diameter petri dish was initially exposed to 280 nm dose and 1 mL of the sample was collected for analysis. The remaining sample was subsequently exposed to 228 nm light for the specified dose and 1 mL was collected upon completion. 20 mL of sample in a 60 mm diameter petri
Table 3 e UV dose designations for Design 3. Sample
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Representative MP dose (mJ/cm2)
Corresponding 280 nm dose (mJ/cm2)
Corresponding 228 nm dose (mJ/cm2)
100 200 300 400 500 600 700 800 900 1000 100 200 300 400 500 600 700 800 900 1000
10 20 30 40 50 60 70 80 90 100 10 20 30 40 50 60 70 80 90 100
0 0 0 0 0 0 0 0 0 0 6.5 13.0 19.5 26.0 32.5 39.0 45.5 52.0 58.5 65.0
dish was also used for single-wavelength 228 nm exposures with 1 mL of sample collected for analysis. Tryptic Soy Agar (TSA) plates containing streptomycin/ ampicillin were prepared and allowed to solidify overnight under refrigeration before plating. Samples were diluted in sterile 1 PBS and 100 mL of sample was pipetted to surface of TSA plates and spread using a sterile L-shaped glass stick. All plating was completed the same day as the exposure, with less than 1 h between exposure and plating. Each sample was plated in triplicate. Plates were allowed to incubate at 37 C overnight for 18e24 h. The volume of sample used and the dilution factor were taken into account when determining CFU counts. Log inactivation was calculated as Log (N0/N) where N is CFU/mL for the sample and N0 is CFU/mL for the untreated control.
2.4. Real-time quantitative polymerase chain reaction (qPCR) DNA was extracted using DNeasy Blood and Tissue Kit (Qiagen, Valencia, CA). Primers and probe (Table 4) were developed as described in Rudi et al. (2010) and synthesized by Integrated DNA Technologies (Coralville, IA). The fragment size was 1504 base pairs. Primers and probe were synthesized by International DNA Technologies. Each sample was prepared with 7.5 mM of forward and reverse primer, PrimeTime probe, nuclease-free water, master mix (Promega, Madison, WI) consisting of 2 units of GoTaq polymerase and 1 GoTaq Buffer, and 2 mL of DNA solution for a total volume of 25 mL. The qPCR samples were performed in duplicate per the program outlined in Table 5 using a MJ MiniOpticon Real-Time PCR machine (Bio-Rad, Hercules, CA). The thermocycle for the PCR assay is shown in Table 5. Similar qPCR methods have been utilized in previous research for the purpose of UV-induced DNA damage quantitation (Rudi et al., 2010). Briefly, the flourogenic probe is designed to attach to a specific DNA sequence within the amplified gene bracketed by the two primers. When not attached, the probe exhibits fluorescence quenching; however, upon attachment to the DNA, the quenching properties of the probe are degraded (Frahm and Obst, 2003). An online sensor in the PCR machine monitors fluorescence levels and the quantitation is based on the number of cycles necessary to reach a fluorescence threshold. DNA quantification was determined using a dilution curve. A standard curve representative of the amount of amplifiable DNA was calculated from the number of PCR cycles required to reach the fluorescence threshold for the dilution series. Log
Table 4 e Primer and probe used for qPCR. Primer
Sequence
Forward AAGAGTTTGATCATGGCTCA Primer Reverse CGGTTACCTTGTTACGACTT Primer Probe CGTATTACCGCGGCTGCTGGCAC a
Genome Fragment Regiona size (bp) 42
1504
1546
Position relative to E. coli 16S rRNA with respect to 50 end.
w a t e r r e s e a r c h 5 6 ( 2 0 1 4 ) 1 8 1 e1 8 9
Table 5 e qPCR Thermocycle Program. Step Denaturation Annealing Synthesis
Temp ( C)
Time (s)
95 55 72
30 30 90
reduction in DNA was calculated as log (N0/N), where N is amount of amplifiable DNA in the treated sample and N0 is the amount of DNA in the untreated control.
2.5.
Statistical analysis
Quantification analyses and calculations were performed in Microsoft Excel. Statistical analyses including Analysis of Variance (ANOVA) and confidence intervals were conducted using MiniTab. Data points indicate average values from experimental trials; however, regressions were conducted using data from all samples.
3.
Results and discussion
3.1.
PCR Assays
3.1.1.
Design 1
Fig. 3 shows the results of sequential-wavelength irradiations and the single-wavelength control irradiations. For the 280 nm 9
Log Reduction in Amplifiable DNA
8 7 6 5 4 3 280 nm/228 nm
2
280 Controls 228 Controls
1 0
0
30 60 90 Dose Following Initial 120 mJ/cm Irradiation (mJ/cm )
120
Fig. 3 e Log reduction of amplifiable DNA extracted from E. coli at various dosing conditions. The x-axis shows the subsequent dose of UV light applied after an initial dose of 120 mJ/cm2 at the specific wavelength. The y-axis is the log reduction values for amplifiable DNA. Each data point represents the average of the sample size (n) for each condition and error bars are shown to demonstrate 80% confidence intervals. Closed circle: sample initially exposed to 120 mJ/cm2 at 280 nm followed by doses of 228 nm shown on x-axis (n [ 4). Open square: samples were initially exposed to 120 mJ/cm2 at 280 nm followed by 280 nm doses shown on x-axis (n [ 2). Open triangle: samples were initially exposed to 120 mJ/cm2 at 228 nm followed by more 228 nm light at the doses shown on the x-axis (n [ 2).
185
and 228 nm single-wavelength exposures, amplifiable DNA decreases with increasing dose for both wavelengths. The 280 nm irradiation demonstrated greater DNA damage after 120 mJ/cm2 than an equivalent dose at 228 nm, which follows the observation that w280 nm is more germicidally effective than w230 nm (Wang et al., 2005). For sequential-wavelength exposures, 280 nm followed by 228 nm irradiation resulted in increased amplifiable DNA following an initial 120 mJ/cm2 dose of 280 nm irradiation. An increase in amplifiable DNA indicates photocleavage of the CPD. The subsequent dose of 120 mJ/cm2 of 228 nm light following an initial dose of 120 mJ/ cm2 at 280 nm provided photoreversal amounting to 1.63 log. PCR results obtained from Design 1 exposures clearly demonstrated DNA damage reversal in samples exposed sequentially to irradiation from 280 nm followed by 228 nm. There was a significant decrease in the log reduction of amplifiable DNA observed from the initial dose 120 mJ/cm2 at 280 nm with each increased dose of 228 nm irradiation. Linear regression for sequential-wavelength irradiations showed a negative slope of 0.012 between log reduction and increasing 228 nm exposure (p-value of slope:
