Skip to main content
NCBI home page
Applied and Environmental Microbiology logoLink to Applied and Environmental Microbiology
. 2010 Sep 17;76(21):7068–7075. doi: 10.1128/AEM.00185-10

Effect of Host Cells on Low- and Medium-Pressure UV Inactivation of Adenoviruses ▿

Huiling Guo 1, Xiaona Chu 1, Jiangyong Hu 1,*
PMCID: PMC2976258  PMID: 20851976

Abstract

UV disinfection is highly effective against most pathogens, with the exception of the adenoviruses (AD). To date, many studies have focused on low-pressure (LP) UV inactivation of AD, but little is known about the effect of medium-pressure (MP) UV inactivation of AD. Despite numerous studies of LP UV inactivation of AD, extreme variabilities in the LP UV dose requirements of AD had been observed because of differing experimental conditions used, such as the types of cell lines used for AD enumeration. This study therefore investigates the effect of three different host cell lines (PLC/PRF/5, human embryonic kidney 293 [HEK293], and XP17BE) on the LP and MP UV dose requirements of AD serotype 5 (AD5), AD40, and AD41 under similar experimental settings. Results showed that for 4-log inactivation of AD, LP UV and MP UV doses needed to be in the ranges of 123 to 182 mJ/cm2 and 65 to 90 mJ/cm2, respectively, when HEK293 and PLC/PRF/5 cells were used for enumeration. The UV doses required for MP UV inactivation of AD were significantly lower than those required for LP UV inactivation (P value < 0.05). When different cell lines were used for enumeration, UV dose requirements for AD differed. AD were portrayed to be most susceptible to UV (LP UV doses of <57 mJ/cm2 and MP UV doses of <42 mJ/cm2 for 4-log AD inactivation) when the XP17BE cells were used as the host cell. The use of different cell lines for AD enumeration affected LP UV dose results more significantly than MP UV dose results (P value < 0.05). Cell line variability factors for LP UV disinfection (CLLP) and MP UV disinfection (CLMP) for AD5, AD40, and AD41 enumerated with HEK293, PLC/PRF/5, and XP17BE cells were in the ranges of 1.0 to 3.2 and 1.0 to 2.5, respectively.

UV disinfection has increasingly been adopted as a favorable technology for disinfection worldwide. It does not involve any disinfection by-product (DBP) formation and is highly effective against protozoan, bacterial, and most viral pathogens at low doses (40 to 60 mJ/cm2) commonly rendered in water treatment plants. Despite numerous advantages over other disinfectants, a major concern pertaining to low-pressure (LP) UV disinfection is the low germicidal effect on adenoviruses (AD).

AD are present in various environments, such as rivers, coastal waters, swimming pool waters, tap waters, and treated drinking waters worldwide (11, 19, 32, 34, 35). There are 52 different human serotypes (enteric or nonenteric) which are responsible for respiratory illnesses, conjunctivitis, and death in persons with compromised immune systems (15). AD are categorized into 6 subcategories, from species A to F. Of these, species F is known to be enteric (AD serotype 40 [AD40] and AD41). Recent studies have shown that enteric AD are not limited to causing only gastrointestinal problems. There is a possibility for all AD serotypes to be waterborne, regardless of whether they exist as enteric or nonenteric strains (37). Similarly, some nonenteric AD which are known to cause respiratory diseases (e.g., AD2 and -5) can also infect the gastrointestinal tract, leading to diarrhea (37). This emphasizes the importance of targeting both enteric and nonenteric AD serotypes during drinking water disinfection. The UV disinfection guidance manual (UVDGM) has also listed AD as a benchmark for UV reactor validation. UV reactors must be capable of delivering UV doses of 186 mJ/cm2 for 4-log virus inactivation (33). Such a high UV dose requirement is more than 4 times that provided in most drinking water treatment plants.

To date, many studies have been conducted to investigate the effect of low-pressure (LP) UV light on AD inactivation (12, 18, 23, 25, 30), but only five such studies have been published on medium-pressure (MP) UV inactivation (10, 20, 21, 22, 27). Despite numerous studies on LP UV inactivation, the UV dose requirements obtained from different studies were highly variable. Based on a review of results from the literature, it was observed that in order to achieve 4-log inactivation of AD40, LP UV dose requirements ranged from 124 to 226 mJ/cm2 (21, 23, 30). Likewise, LP UV dose requirements for 4-log inactivation of AD41 ranged from 112 to 222 mJ/cm2 (4, 18, 23). These differences in LP UV dose requirements have been attributed to the use of different cell lines and culturing methods, history of the viral stocks, virus preparation methods, virus serotypes, the subjective nature of cell culture infectivity assays, cell/virus storage time, and variability in experimental conditions, such as UV exposure setups, UV dosimetry measurements, etc. (2, 10, 18). To date, it is not known which factor/methodology, if any, contributes directly to these extreme variabilities in UV dose results (4) and whether these variations are due to actual differences in UV resistances or other experimental factors.

Previous studies have demonstrated that slight differences in the absorbances of the water matrices (laboratory-buffered water, groundwater, or treated drinking water) did not significantly affect AD resistance to LP UV inactivation (21, 30). On the cellular level, studies have been conducted to investigate the susceptibility of different cell lines for AD detection/enumeration. These include the BGMK, Caco-2, HeLa, Hep-2, A549, PLC/PRF/5, and human embryonic kidney 293 (HEK293) cell lines (7, 13, 36). Such studies are especially crucial for the enteric AD (AD40 and AD41) because they can infect cells without any cytopathogenic effect (CPE) and are more difficult to cultivate than other AD (1). Despite numerous studies of cell lines, there has yet to be an established method/cell line for the propagation or assay of different AD serotypes (4). This arises due to the conflicting viewpoints on the susceptibility of different cell lines to AD infection (7, 13, 16, 36). To add on to the complexity, extreme variations in the growth of enteric AD within the same cell line have also been observed (9). Not only do these variations exist for the same AD serotypes from different laboratories, but different batches of cell lines from the same origin/source also resulted in different susceptibilities to AD growth (1). Hence, this may lead to a compromise in results when the UV doses of AD are compared between different laboratories using differing batches and types of cell lines for the enumeration and tabulation of the UV dose response of AD. Differing results may also be obtained when different cell lines are used for the assaying of UV-irradiated AD due to different availabilities of repair enzymes in different cell lines (12). These factors can inevitably distort UV dose results to a certain extent.

In view of the aforementioned issues with AD, we henceforth provide a comprehensive study on the LP and MP UV dose requirements of AD5, AD40, and AD41 under similar experimental settings. Many studies have been conducted using AD2 because it provides a good representation of various AD of public health concern and it can be easily grown to high titers. In this study, respiratory AD5 and enteric AD40 and AD41 were chosen. We also address the question of whether the use of three different cell lines (PLC/PRF/5, HEK293, and XP17BE) for AD enumeration will influence the UV dose response of AD. PLC/PRF/5 and HEK293 cells were chosen because they had been frequently used for AD propagation and assays in previous studies (4, 13, 16, 18, 21, 22, 23). This would therefore provide a good comparison of results with those from past studies for AD5, -40, and -41 enumeration, using HEK293 and PLC/PRF/5 as host cells. The repair-deficient cells (XP17BE) were obtained from a xeroderma pigmentosum patient. This cell line, which was known to display UV-induced unscheduled DNA synthesis at 30 to 60% of normal cells, was also chosen as a control for comparison. This study will establish the importance of the types of host cells on AD resistance to UV light and the extent to which MP UV irradiation provides a more efficient alternative to LP UV irradiation for AD inactivation.

MATERIALS AND METHODS

Virus and host cells.

All viruses and cell lines were purchased from American Type Culture Collection (ATCC). AD used in this study included AD5 (ATCC VR-5), AD40 strain Dugan (ATCC VR-931), and AD41 strain Tak (ATCC VR-930). Primary liver carcinoma (PLC/PRF/5; ATCC CRL-8024), HEK293 (ATCC CRL-1573; transformed with AD5 DNA), and xeroderma pigmentosum (XP17BE; ATCC CRL-1360) cells were used for AD enumeration prior to and following UV inactivation. The XP17BE cells were known to be 30 to 60% deficient in repair.

PLC/PRF/5, HEK293, and XP17BE cells were grown in growth medium comprised of Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS). Cells were grown in 75-cm2 culture flasks at 37°C in an incubator (Sanyo Electric Biomedical MCO-17AIC) supplied with 5% CO2. All cells were subcultured (1:3 to 1:4 dilution) when 70 to 80% confluence was attained.

Virus propagation.

AD5, -40, and -41 were propagated using the HEK293 cell line. HEK293 cells were grown in growth medium comprised of DMEM supplemented with 10% FBS. When cells attained 80% confluence, growth media was aspirated from 75-cm2 culture flask, and the cell monolayer was washed with 1× phosphate-buffered saline (PBS). One milliliter of virus stock was applied to the cell monolayer for virus infection. A control flask was also included, with 1 ml of PBS applied rather than virus. Cells and virus were incubated in a CO2 incubator with 5% CO2 at a temperature of 37°C for 1 h. The flask was subjected to frequent rocking at 15-min intervals to aid virus adsorption onto cells. Thereafter, maintenance medium comprising DMEM supplemented with 2% FBS was added prior to further incubation at 37°C in 5% CO2 until the cytopathogenic effect (CPE) was complete. The CPE was characterized by cell rounding, shrinkage, and detachment from the surfaces of the wells. Following infection, viruses were separated from cell debris by centrifugation at 7,000 × g for 10 min (Hettich Universal 32 R). A second round of harvesting was then conducted by adding 1 ml of maintenance medium to the cell pellet, followed by three freeze-thaw cycles to release viruses from infected host cells. Centrifugation was conducted at 7,000 × g for 10 min. The aqueous phase containing viruses was stored in cryogenic tubes at −80°C for further use.

Collimated beam setup.

LP and MP UV irradiations were performed using a collimated beam setup (Calgon Carbon Corporation). The monochromatic 10-W LP mercury lamp emitted UV light at 253.7 nm, and the polychromatic 1,000-W MP lamp emitted light in the wavelength range of 200 to 400 nm. The UV lamp was enclosed in a black containment to which a black-colored collimated beam was attached. These compartments were black anodized to minimize any internal scattering. Compressed air was used to pneumatically regulate the shutter to control the UV irradiation time. Incident UV irradiance was measured using a radiometer (International Light IL1400A) and a SED240 detector calibrated to 254 nm, according to the National Institute of Standards and Technology (NIST). The average germicidal irradiance values for LP and MP were approximately 33 to 36 μW/cm2 and 64 to 79 μW/cm2, respectively.

For LP UV, the petri factor, reflection factor (0.975), water factor (0.687), and divergence factor (0.993) were applied to the incident irradiance to obtain the average germicidal irradiance. The LP UV dose was computed using the product of the average germicidal irradiance and exposure time (5). To obtain the average germicidal MP UV irradiance, two additional factors were applied. Because the detector was calibrated to 254 nm, a sensor factor (1.206) was applied to correct for the variation in the detector sensitivity to different wavelengths of the MP UV emission spectrum. The germicidal factor was also applied by taking into consideration the relative DNA absorbance efficiency across the MP UV emission spectrum. Only the weighted MP UV dose was considered in this study. UV absorbances and average germicidal irradiances are presented in Table 1.

TABLE 1.

UV absorbances and average germicidal irradiances

UV lamp type AD serotype Avg germicidal irradiance (mW/cm2) UV absorbances (A254)
LP 5 0.033 0.198-0.206
40 0.036 0.241-0.268
41 0.033 0.252-0.264
MP 5 0.073 0.190-0.202
40 0.079 0.220-0.249
41 0.064 0.252-0.268
Open in a new tab

UV irradiation procedures.

Stock cultures of AD5, -40, and -41 were diluted in 1× PBS solution prior to UV disinfection. The monochromatic and polychromatic absorbances of the virus solution were measured using a UV-visible (UV-Vis) spectrophotometer (UV-1700; Shimadzu, Japan) for LP and MP UV irradiation, respectively. Absorbance values were then utilized to obtain the UV irradiation time required for each specified UV dose. Ten milliliters of virus suspension (initial virus titer [N0] = ∼106 50% tissue culture infectious dose [TCID50]/ml) was irradiated in 60-mm by 15-mm sterile polystyrene petri dishes while being subjected to continuous stirring at 300 rpm with a magnetic stirrer. Ten milliliters of the same virus suspension, subjected to continuous stirring without UV light exposure, was used as a control. UV-irradiated viruses and their controls were then enumerated. UV irradiation experiments were performed in triplicate.

Virus enumeration.

The 50% tissue culture infectious dose (TCID50) method was used to enumerate AD5, -40, and -41, with PLC/PRF/5, HEK293, and XP17BE as host cells. Briefly, the cell lines, grown to 70% confluence, were rinsed with 1× PBS. Two milliliters of trypsin-EDTA was added to detach the cell monolayer from the flask surface. Maintenance medium was added to the cell culture and mixed thoroughly before cells were transferred into the 96-well microtiter plates at approximately 2 × 104 cells per well. Cells were incubated overnight at 37°C in 5% CO2 prior to virus infection. Virus controls and UV-irradiated viruses were serially diluted in maintenance medium, and 100 μl of viruses was added into each well. The microtiter plates were then incubated at 37°C with 5% CO2 and observed for the CPE over the course of 10 days for AD5 and 21 days for AD40 and AD41. Since each AD serotype was enumerated with three different cell lines, an AD serotype enumerated with a specific cell line is denoted “AD serotype (host cell)” from this point forth for easy interpretation. For example, AD5 (HEK293) refers to AD5 enumerated with the HEK293 cell line.

Data analyses.

AD inactivation kinetics were described using Chick's law. The UV dose response of AD was fitted according to the following: regression of log10 (N0/N) = KIt, where N0 is the initial concentration of virus, N is the final concentration of virus after UV inactivation, K is the inactivation rate, I is the average germicidal irradiance value, t is the UV exposure time. The linear regression fits were obtained from the UV dose response based on the LP UV doses of 0 to 120 mJ/cm2 and MP UV doses of 0 to 80 mJ/cm2. At least three data points were used for the construction of each linear fit. Based on the regression fit, the inactivation rate K was obtained for LP UV (KLP) and MP UV (KMP). UV dose requirements for 4-log inactivation of AD were then tabulated using these K values. Statistical analyses were conducted using Microsoft Excel (Microsoft Office 2007), and P values were tabulated at a 95% confidence level.

The effectiveness levels of LP and MP UV irradiation were compared in terms of their UV dose requirements using the LP/MP ratio. The LP/MP ratio of each AD (host cell) combination was obtained by using the following equation: LP/MP ratio = LP UV dose required for 4-log AD inactivation/MP UV dose required for 4-log AD inactivation.

To evaluate the susceptibility of different cell lines for AD enumeration and the effect of cell lines on the UV dose requirement of AD, cell line variability factors were developed for both LP UV disinfection (CLLP) and MP UV disinfection (CLMP). For LP UV disinfection, this was done by dividing the inactivation rate K pertaining to an AD serotype enumerated with XP17BE cells by the inactivation rate K pertaining to an AD serotype enumerated with either HEK293 or PLC/PRF/5 cells (CLLP,i = KXP17BE/Ki, where i is either HEK293 or PLC/PRF/5 cells; e.g., if the HEK293 cell line was being considered, then CLLP,HEK293 = KXP17BE/KHEK293). The same was done for MP UV results. An AD serotype with CL of 1.0 indicated that the inactivation rate K of the particular AD when enumerated with that cell line is similar to that when the same AD was enumerated with the repair-deficient cell line XP17BE (i.e., both cells are equally susceptible to AD infection, and no bias in UV results will arise from the use of either cell lines during AD enumeration).

RESULTS AND DISCUSSION

The LP UV inactivation of various AD serotypes had been widely studied, but some existing concerns are still yet to be addressed. It has been observed that variations in LP UV dose results exist across different studies. Different studies have reported different LP UV dose requirements for the same AD serotype. Likewise, in general, LP UV doses in the range of 122 to 222 mJ/cm2 were observed to be required for 4-log inactivation of AD (4, 18, 21, 23, 25, 30). A definitive UV dose requirement for the different AD serotypes could not be established because of various inevitable conditions. These included UV irradiation procedures, water matrices, cell line or virus propagation/culturing procedures, etc. (2, 4, 10, 18). To date, some have clarified that different water matrices could not be the leading factor contributing to differences in UV dose requirements despite the difference in absorbances (21, 30). Since the cell line and virus propagation/culturing procedures affect the inherent properties of AD, they are seen as likely causative agents for the disparities in AD resistance to LP and MP UV inactivation. In this study, similar experimental conditions (e.g., propagation, harvesting, and UV irradiation procedures) were maintained for AD5, -40, and -41, so as to provide for impartial comparison between their LP and MP UV dose responses. Prior to and following LP and MP UV inactivation, AD were enumerated using the HEK293, PLC/PRF/5, and XP17BE cell lines to understand the susceptibility of all three cell types to the infection of these three AD strains.

Low-pressure (LP) and medium-pressure (MP) UV inactivation of AD5, AD40, and AD41.

The absorption spectra of AD5, -40, and -41 in the test water (i.e., 1× PBS) and the emission spectra of LP and MP UV lamps are illustrated in Fig. 1. The MP UV lamp has a peak at 266.4 nm, which coincided closely with the absorbance peak of AD (in 1× PBS) at approximately 267 nm. LP UV and full-spectrum MP UV irradiations were performed for AD5, -40, and -41. LP and MP UV dose results for 4-log AD inactivation are shown in Fig. 2. Based on the first-order inactivation kinetics, KLP and KMP and their corresponding R2 values are presented in Table 2. Results indicated that LP UV dose requirements for 4-log inactivation of AD were comparable to those stated in other literature when the effect of the host cells was not considered (Fig. 2) (4, 18, 21, 23, 25, 30). For 4-log inactivation of AD5, -40, and -41, LP UV dose requirements were in the range of 123 to 182 mJ/cm2, and those for MP UV were in the range of 65 to 90 mJ/cm2, with HEK293 and PLC/PRF/5 as host cells. However, by observing the CPE in different cell lines following UV inactivation, the UV dose response of each AD serotype was different when different cell lines were used for enumeration.

FIG. 1.

FIG. 1.

Open in a new tab

Absorption spectra of AD5, -40, and -41 in 1× PBS and emission spectra of LP and MP UV lamps used in this study.

FIG. 2.

FIG. 2.

Open in a new tab

LP and MP UV dose requirements for 4-log inactivation of AD5, -40, and -41. The LP and MP UV dose requirements were tabulated based on the inactivation rates.

TABLE 2.

Summary of K and R2 values of AD with LP and MP UV irradiation and enumeration with HEK293, PLC/PRF/5, and XP17BE cells

Virus (host cell) UV irradiation
LP
 MP
K R2 K R2
AD5 (HEK293) 0.0264 0.97 0.0442 0.99
AD5 (PLC/PRF/5) 0.0324 0.99 0.0460 0.99
AD5 (XP17BE) 0.0765 0.98 0.1103 0.96
AD40 (HEK293) 0.0287 0.98 0.0602 0.99
AD40 (PLC/PRF/5) 0.0298 0.96 0.0612 0.99
AD41 (HEK293) 0.0220 0.97 0.0510 0.98
AD41 (PLC/PRF/5) 0.0292 0.99 0.0561 0.96
AD41 (XP17BE) 0.0697 0.98 0.0944 0.99
Open in a new tab

The AD5 results in this study were comparable to previous results with the PLC/PRF/5 cell line (25). Nwachuku et al. (25) observed a LP UV dose requirement of 90 mJ/cm2 for 2.8-log inactivation of AD5 (PLC/PRF/5). A requirement of 93 mJ/cm2 was observed for 3-log inactivation of AD5 (PLC/PRF/5) in this study. In contrast, AD5 (HEK293) was observed to require a LP UV dose of approximately 120 mJ/cm2 for 3-log inactivation (4; this study). The variation in AD5 results in comparison to those in other studies was considered insignificant (P values > 0.05) despite the use of different cell lines and, possibly, different UV irradiation procedures. Unlike enteric AD40 and AD41, the use of different cell types for AD5 enumeration in UV inactivation studies is probably not as crucial. This is based on the knowledge that AD5 could be easily grown in cell cultures and with most cell types (4). Therefore, the bias and differences in UV dose requirements arising from the use of different cell lines for the enumeration of UV-irradiated AD5 could be greatly minimized. Furthermore, this observation was also evident in our study when the CPE by AD5 appeared during the first few days of virus infection of the PLC/PRF/5 and HEK293 cell lines, and the CPE that formed was clear and distinct, whereas the CPEs formed by enteric AD were slow to appear and hard to distinguish.

The LP UV dose requirements for AD40 and AD41 were different, except the difference posed by HEK293 and PLC/PRF/5 cells for AD40 was not as significant as those for AD5 and AD41 (P value > 0.05). For AD5 and AD41, the LP UV dose requirement for 4-log inactivation was higher when HEK293 (152 and 182 mJ/cm2 for AD5 and AD41, respectively) rather than PLC/PRF/5 (123 and 137 mJ/cm2 for AD5 and AD41, respectively) cells were used for enumeration. With AD41 (PLC/PRF/5), LP UV dose requirements were approximately <137 mJ/cm2 (23; this study). A higher LP UV dose requirement with AD41 (HEK293) (182 mJ/cm2) than with AD41 (PLC/PRF/5) (P value < 0.05) was observed. Ko et al. observed a LP UV dose requirement of 222 mJ/cm2 for 4-log inactivation of AD41 when HEK293 cells were used for AD41 propagation and mRNA reverse transcriptase PCR (RT-PCR) was used for quantitation of the virus titer (18). In contrast, for AD40, the type of cell line appeared to have minor effects on AD susceptibility, with AD40 (HEK293) and AD40 (PLC/PRF/5) LP UV dose requirements of 139 and 134 mJ/cm2, respectively, for 4-log inactivation. To our knowledge, no documentation on HEK293 cells for AD40 enumeration in UV-related studies was available.

With repair-deficient host cells (XP17BE), LP and MP UV results were distinctively different from those with HEK293 and PLC/PRF/5 cell lines. AD5 (XP17BE) and AD41 (XP17BE) were observed to be the most susceptible to LP and MP UV irradiation. With LP UV doses of <60 mJ/cm2, 4-log inactivation of AD5 and AD41 could be attained, with AD5 (XP17BE) and AD41 (XP17BE) requiring LP UV doses of 52 and 57 mJ/cm2, respectively. Because no CPE could be observed when AD40 was enumerated with XP17BE cells, no UV results for AD40 (XP17BE) could be obtained. This phenomenon was not unanticipated for enteric AD, which can infect cells without demonstrating any CPE (1). The choice of cell line used for enumeration is therefore an important aspect of AD enumeration.

When MP UV light was used, differences in UV doses required for 4-log inactivation of different AD serotypes were not as variable as those when LP UV light was used. Differences in LP UV results were more pronounced with the use of different host cells (P value < 0.05), compared to those in MP UV results. The MP UV doses required for 4-log inactivation of AD5 (HEK293) and AD5 (PLC/PRF/5) were 90 and 87 mJ/cm2, respectively (Fig. 2). The MP UV doses required for 4-log inactivation of AD40 (HEK293 and PLC/PRF/5) and AD41 (HEK293 and PLC/PRF/5) were in the ranges of 65 to 78 mJ/cm2 and 71 to 78 mJ/cm2, respectively. However, with XP17BE cells, MP UV doses required for 4-log inactivation of AD5 (XP17BE) and AD41 (XP17BE) were almost half of those required when the same AD serotypes were enumerated with HEK293 and PLC/PRF/5 cells. The LP UV doses required for AD5 (XP17BE) and AD41 (XP17BE) were only slightly higher than the MP UV dose requirements (P values > 0.05), where MP UV dose requirements for 4-log inactivation of AD5 (XP17BE) and AD41 (XP17BE) were 36 and 42 mJ/cm2, respectively.

Effectiveness of LP versus MP UV inactivation.

Few studies on MP UV inactivation have been conducted, and they are limited to the inactivation of AD2 and AD40 (10, 20, 21, 22, 27). In this study, differences in LP UV dose requirements for 4-log inactivation of AD5, -40, and -41 were larger than those in MP UV dose requirements. These results were contrary to those for AD5 and -41 enumerated with repair-deficient XP17BE cells.

Thus far, no known study have been conducted on MP UV inactivation of AD5. However, based on knowledge gathered from previous studies, AD5 and AD2 were almost equally resistant to LP UV inactivation. This was probably due to their similar nature, as both are respiratory disease-causing AD. An LP UV dose of 120 mJ/cm2 (HEK293 as the host cell lines) was required for 3-log inactivation of AD2 (4), and an LP UV dose of 160 mJ/cm2 (HEK293 and PLC/PRF/5 as host cells) was required for 4-log inactivation of AD2 (12). Similarly, in this study, the LP UV doses required for 3- and 4-log inactivation of AD5 (HEK293) inactivation were 114 and 152 mJ/cm2, respectively. With MP UV dosing, approximately 90 mJ/cm2 was required for 4-log inactivation of AD2 (A549) (21), which was comparable to that result for AD5 in this study. Shin et al. (27) also observed that 2-log inactivation of AD2 (A549) was achieved with LP and MP UV doses of 80 and 32 mJ/cm2, respectively. To achieve 2-log inactivation of AD5 with HEK293 and PLC/PRF/5 cells, the LP UV doses required were 76 and 62 mJ/cm2, respectively, and the MP UV doses required were 45 and 43 mJ/cm2, respectively (this study). Therefore, with LP and MP UV inactivation, AD5 was as resistant as AD2. However, because no study was conducted with AD5, -40, and -41 using A549 as the host cell, it was not possible to ascertain the actual effect of A549 cells on the extent of repair for these three AD serotypes following LP and MP UV inactivation. Based on inspection, the LP UV dose requirements for AD2 (A549) and, hence, the extent of AD2 repair by A549 appeared to resemble those for AD5 (HEK293), but with MP UV disinfection, the UV dose required by AD2 (A549) was lower than those required by both AD5 (HEK293) and AD5 (PLC/PRF/5).

In the development of the LP/MP ratio, the LP UV dose requirement of an AD serotype enumerated with HEK293 cells was compared with the MP UV dose requirement of the same AD serotype enumerated using the same cell type. This was to eliminate any bias arising from the differences in the types of cell lines used for AD enumeration, which hence might influence the tabulation of the UV dose requirements. LP/MP ratios for the combinations of AD (host cell) are summarized in Table 3. AD5 (PLC/PRF/5), AD5 (XP17BE), and AD41 (XP17BE) had the lowest LP/MP ratio values at 1.4. The LP/MP ratio values for the other variants (i.e., AD enumerated with HEK293 and PLC/PRF/5 cells) were in the range of 1.7 to 2.1. This indicated that for those AD (host cell) combinations, the MP UV dose requirements for 4-log AD inactivation were 48 to 59% of the LP UV dose requirements. MP UV dose requirements are lower than those for LP UV regardless of the host cell used for enumeration (P values < 0.05). However, based on observations of AD5 (PLC/PRF/5), AD5 (XP17BE), and AD41 (XP17BE), the advantage of using MP UV rather than LP UV was not as significant compared to using the other AD (host cell) combinations evaluated in this study (P values > 0.05).

TABLE 3.

LP/MP ratios corresponding to AD enumerated with three different host cell lines

Virus (host cell)a LP/MP ratiob
AD5 (HEK293) 1.7
AD5 (PLC/PRF/5) 1.4
AD5 (XP17BE) 1.4
AD40 (HEK293) 2.1
AD40 (PLC/PRF/5) 2.1
AD41 (HEK293) 2.3
AD41 (PLC/PRF/5) 1.9
AD41 (XP17BE) 1.4
Open in a new tab
a

No CPE observed for AD40 (XP17BE).

b

The LP/MP ratio was calculated as follows: LP UV dose required for 4-log inactivation of each AD serotype/MP UV dose required for 4-log inactivation of each AD serotype.

The reduction in the UV dose requirement for AD5 resulting from a switch from LP UV to MP UV inactivation was not as significant as that observed in a previous study of AD2 and AD40 (21). LP/MP ratios for AD5 were also distinctively lower than those for AD40 and AD41 with both PLC/PRF/5 and HEK293 as host cells (P values < 0.05). LP/MP ratios for AD40 and AD41, on the other hand, were comparable to those obtained previously for AD2, for which the relative effectiveness of MP UV irradiation in comparison to that of LP UV was 2.2 (22). Linden et al. observed that the switch from LP to MP UV irradiation resulted in a UV dose reduction from 120 mJ/cm2 to 40 mJ/cm2 for 4-log inactivation of AD40 (21). This reduction in the UV dose requirement arising from the switch from LP to MP inactivation arises from the different germicidal effects of both LP and MP UV. LP UV light (at 253.7 nm) damages the virus's ability to function/multiply by creating dimers in its genome (26). Lately, it has also been suggested that LP UV light also possesses the ability to cause slight conformational changes in viral proteins and, hence, the loss of the virus's ability to attach to cell receptors (24). MP UV, which comprises wavelengths in the UV-A (320 to 400 nm), UV-B (290 to 320 nm), and UV-C (190 to 290 nm) range, may inflict additional damage on both DNA and AD proteins, disrupting the virion structure (3, 14, 17). The low LP/MP ratio values for AD5 and AD41 enumerated with repair-deficient XP17BE cells, with both being 1.4, were probably due either to the limited repair capability of XP17BE cells under both LP and MP UV irradiation or to the damage inflicted on the virion structure, hence inhibiting viral infection into host cells. Differences in UV dose response posed by the use of different cell lines are further understood using the CLLP and CLMP values, which will be discussed below.

Effect of host cells on AD susceptibility to LP and MP UV inactivation.

The types of cell lines used are a major determining factor for the growth/infection and, hence, extent of CPE formation by AD. Many studies which focused solely on cell culture had differing viewpoints regarding the susceptibility of various cell lines to AD propagation. Some studies have demonstrated that A549 cells were effective for cultivating AD40 and AD41 (7, 36), whereas others reported that PLC/PRF/5 cells were the most efficient (13). This indicated that the variability in AD UV dose results, particularly the enteric AD, could have arisen due to the use of different cell lines for AD enumeration.

Results obtained from both LP and MP UV inactivation of AD, as well as results from previous studies, indicated the impending effect of the choice of host cells on the UV dose response of AD. When a particular AD serotype (before and after UV irradiation) was enumerated using the three different cell lines, the tabulated UV dose responses differed when different cell lines were used. The CLLP and CLMP of the different combinations of AD (host cells) are summarized in Table 4. Despite the same experimental procedures being adopted for all cell lines, AD5 and AD41 enumerated with HEK293 cells appeared more resistant, with higher CLLP and CLMP values, than those enumerated with PLC/PRF/5 cells (P value > 0.05 for AD5; P value < 0.05 for AD41). The CLLP,HEK293 values for AD5 and AD41 were 2.9 and 3.2, respectively. This implied that the inactivation rates (K) of AD5 (XP17BE) and AD41 (XP17BE) were 2.9 and 3.2 times higher than those of AD5 (HEK293) and AD41 (HEK293), respectively. With PLC/PRF/5 as the host cell, the CLLP,PLC/PRF/5 values for AD5 (PLC/PRF/5) and AD41 (PLC/PRF/5) equaled 2.4.

TABLE 4.

Cell line variability factorsa

Virus (host cell) CL value
LP UV MP UV
AD5 (HEK293) 2.9 2.5
AD5 (PLC/PRF/5) 2.4 2.4
AD5 (XP17BE) 1.0 1.0
AD41 (HEK293) 3.2 1.9
AD41 (PLC/PRF/5) 2.4 1.7
AD41 (XP17BE) 1.0 1.0
Open in a new tab
a

The cell line variability factor values for LP UV (CLLP, host cell) and MP UV (CLMP, host cell) were calculated as follows: the inactivation rate K of AD enumerated with XP17BE cells/that of AD enumerated with either HEK293 or PLC/PRF/5 cells. For example, CLLP,i = KXP/Ki, where KXP is the LP UV inactivation rate K of AD enumerated with XP17BE cells and Ki is the LP UV inactivation rate K of AD enumerated with either HEK293 or PLC/PRF/5 cells. i, HEK293 or PLC/PRF/5 cells.

Unlike those for CLLP, the CLMP values were not as extreme for both AD5 and -40. The CLMP,HEK293 and CLMP,PLC/PRF/5 values for AD5 and AD40 were ≤2.5. The lower CLMP values indicated that the type of cell line used for AD enumeration was not as highly influential over the inactivation rates of AD as that of LP UV. Analogous to the trend observed with CLLP, the CLMP values for both AD5 (HEK293) and AD41 (HEK293) were also higher than those obtained when PLC/PRF/5 was used as the host cell.

The results suggested that because of the differences in cell line susceptibility to infection by different AD serotypes, UV dose responses obtained with different cell lines were inherently different. The differences in the inactivation rates (K) of each AD, enumerated with three different host cells in this study, clearly spells out the differences in host cell susceptibility to the infection of different AD serotypes (Table 2) (P values of <0.05 for all LP and MP UV results, except for the comparison of KMP values between AD5 (HEK293) and AD5 (PLC/PRF/5), where the P value was >0.05). However, with the enteric AD, the UV results could be complicated by a series of factors, such as the limited occurrence of the CPE in different cell lines and the extent of repair by different cell lines. Taken together, one cannot adequately conclude which AD serotype is the most resistant to LP UV inactivation without considering the type of cell line used during enumeration. This is because of the existence of disparities in cell lines and also the detection methods used to quantitate the cell titers.

Thus far, results from this study and those from recent studies on AD (10, 20, 21, 22, 27) have provided evidence regarding the superiority of MP UV disinfection rather than LP UV disinfection. MP UV results have provided a positive indication regarding its ability to inactivate AD, even at much lower UV doses than those used with LP UV inactivation. Nevertheless, regardless of whether MP or LP UV was used, the choice of cell line does affect AD enumeration following UV irradiation to a certain extent. The different AD resistances of an AD strain enumerated with different cell lines were probably a result of (i) the selectivity of different host cells for different virus types (1, 37), (ii) the different availabilities of repair enzymes present in different cell types (12), and (iii) the type of lamp used for UV disinfection (27). Grabow et al. (13) observed that the PLC/PRF/5 cell line was 100 times more sensitive to a laboratory strain of AD41 and 10 times more sensitive to a laboratory strain of AD40 than the HEK293 cell line. Both of these enteric AD have been known to be more difficult to culture than other AD, and they can infect some cells without causing a CPE (7, 31). Some studies had observed that long periods of incubation are required with HEK293 cells before the appearance of a CPE. The reason that such a phenomenon is occurring in certain cell lines is probably due to the existence of an early replicative block in the AD growth cycle (28). In contrast, some have noted that the HEK293 cell line has been successfully used for enteric AD growth because this cell line contains sequences from the AD5 E1 region, which is complementary to the defective region in this E1 function of the enteric AD. Hence, this assisted in AD replication within the HEK293 cell line (1, 6, 28, 29).

Besides the selectivity of the cell line, another possible reason could be the existence of different amounts of repair enzymes in different host cells (12). Prior to UV irradiation, the use of different cell lines could result in differences in initial AD titers due to the selectivity of different cell types for AD. It is known that damaged DNA of UV-irradiated AD can be repaired by the nucleotide excision repair (NER) mechanism of the host cell, thus restoring the ability of AD to replicate (26). During enumeration, the same titer of AD surviving immediately after UV irradiation undergoes different extents of repair when infected in different cell lines, which may contain differing amounts of repair enzymes. Depending on the extent of repair after they are assayed in different cell types, different amounts of the CPE will be inflicted. These differences will hence be reflected during the tabulation of the UV dose response curves. Furthermore, certain AD serotypes may repair more quickly in one cell line than in other cell lines, hence leading to disparities in results when plaque assay or TCID50 results are analyzed on the same day. Conversely, Day studied the effect of cell types and repair enzymes by assaying UV-irradiated AD2 in cells obtained from xeroderma pigmentosum patients, which are deficient in the NER mechanism, and demonstrated that AD2 was as sensitive to UV as other microorganisms (8).

The type of lamp used for inactivation also affected the extents of repair (27). By using XP17BE cells as the benchmark in our study, CLLP,host cell was in the range of 2.4 to 3.2, which was reduced to the range of 1.7 to 2.5 for CLMP,host cell. Slightly higher UV dose requirements were required for AD5 and AD41 enumerated with XP17BE cells when LP UV inactivation rather than MP UV inactivation was used. Though it is known that XP17BE cells are repair deficient, it has not yet been demonstrated that completely no repair takes place. In a recent study by Shin et al. (27), AD2 inactivated by LP UV appeared to restore its infectivity over a 14-day assay period, whereas AD2 inactivated by MP UV irradiation demonstrated no restoration of infectivity with A549 as the host cell line. Restoration/repair of AD2 appeared to be significant with LP UV irradiation but not with MP UV irradiation following infection of A549 cells. Based on XP17BE as a host cell line in this study, the slightly lower UV doses required for MP UV rather than LP UV disinfection arise because of two possible scenarios, as follows. (i) No repair takes place following MP UV disinfection, but slight repair takes place after that of LP UV. (ii) Assuming that the XP17BE cell line is used as the benchmark, whereby no repair is rendered to AD, reduced infection of XP17BE cells by AD following MP UV disinfection could be a result of virion protein damage, and hence, virus adsorption onto cells could be affected. Notably, it is likely that the extent of physical damage to the virion protein is limited during UV irradiation and does not vary between AD serotypes.

Since XP17BE cells are known to be repair deficient, the extent of protein damage could be gauged from that which occurred when XP17BE cells were infected with AD following LP and MP UV irradiation, because no/minimal repair takes place within these cells. Henceforth, the difference in LP and MP UV dose requirements based on XP17BE cells accounts for the damage inflicted on the proteins and probably is minimal due to the repair mechanism. Further comparison of AD survival rates following LP and MP UV disinfection and infection in HEK293 and PLC/PRF/5 cells suggested that these differences in UV dose requirements could be due dominantly to the differences in AD repair and, to a minor extent, the effects of protein damage, which limit viral adsorption onto host cells. Because CLMP values for all three AD were close to 1.0, this indicated that MP UV light was significantly more effective than LP UV light (22) in minimizing the effect of the host cell's selectivity of AD growth/infection (P value < 0.05). MP UV irradiation could have affected the ability of AD to induce NER in the HEK293 and PLC/PRF/5 cells, and to a minor extent, it could have altered structural proteins on the surfaces of AD, hence hindering AD attachment to different host cells. These therefore lead to smaller differences and bias arising from the use of different cell lines for enumeration following MP UV inactivation.

In the context of drinking water, there is a need to determine which cell line is reflective of actual case scenarios of AD consumption by the human body. There is no doubt that AD enumerated with repair-deficient XP17BE cells are portrayed to be highly susceptible to LP UV inactivation and can be seen to be inactivated at low UV doses, but this may pose as an inadequate measure. AD (XP17BE) are portrayed to be inactivated at low LP UV doses, but in actual fact, when consumed by humans, AD may still be able to repair themselves. In typical human cells, repair is predicted to occur. Human kidney and/or liver cells appeared to be possible options for host cells in the context of AD ingestion, since they are capable of repair. The issue now should probably depend on which cell line could most accurately represent normal human cells. Adoption of the UV dose response of AD (HEK293) could pose as a conservative measure for the estimation of the extent of kill by LP or MP UV light during drinking water disinfection. In vivo studies could probably be proposed, using animals to ascertain the type of cells which closely resemble human cells and the repair rendered to AD following UV disinfection.

Conclusion.

This study demonstrated that UV dose requirements of AD5, -40, and -41 depended highly on the type of cell line used for AD enumeration. This applies especially to LP UV inactivation of AD. Based on CLLP and CLMP, LP and MP UV dose requirements for the inactivation of each AD strain enumerated with HEK293 cells were distinctively higher than those enumerated with PLC/PRF/5 cells for a specified level of log inactivation. In contrast, with repair-deficient cell line XP17BE, AD5 and AD41 were portrayed to be very susceptible to LP and MP UV disinfection. With the type of cell line contributing to the variability in the UV dose response of AD, the comparison of UV dose results for a particular AD serotype obtained by different studies, in which the same AD serotype is enumerated using different cell types, may create a certain extent of bias. This applies, in particular, to fastidious enteric AD40 and AD41, which possess multiple defects in their DNA. Therefore, comparison of UV dose results for AD must be made with consideration to the type of cell line used during enumeration. There may also be a need to establish a standard cell line to be used for the enumeration of each AD serotype due to the selectivity of different cell types for different AD serotypes. The types of cell lines used for enumeration, however, were not as influential on MP UV-irradiated AD. The MP UV dose requirements of AD enumerated with the PLC/PRF/5 and HEK293 cell lines were not significantly different (P value > 0.05). Overall, MP UV inactivation, which was proven to be approximately twice as effective as LP UV light for AD inactivation, proves to be a good alternative to LP UV disinfection of the AD serotypes evaluated.

Footnotes

▿

Published ahead of print on 17 September 2010.

REFERENCES

  • 1.Albert, M. J. 1986. Enteric adenoviruses. Arch. Virol. 88:1-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Ballester, N. A., and J. P. Malley. 2004. Sequential disinfection of adenovirus type 2 with UV-chlorine-chloramine. J. Am. Water Works Assoc. 96:97-103. [Google Scholar]
  • 3.Battigelli, D. A., M. D. Sobsey, and D. C. Lobe. 1993. The inactivation of hepatitis A virus and other model viruses by UV irradiation. Water Sci. Technol. 27:339-342. [Google Scholar]
  • 4.Baxter, C. S., R. Hofmann, M. R. Templeton, M. Brown, and R. C. Andrews. 2007. Inactivation of adenovirus types 2, 5, and 41 in drinking water by UV light, free chlorine, and monochloramine. J. Environ. Eng. 133:95-103. [Google Scholar]
  • 5.Bolton, J. R., and K. G. Linden. 2003. Standardization of methods for fluence (UV dose) determination in bench-scale UV experiments. J. Environ. Eng. 129:209-215. [Google Scholar]
  • 6.Brown, M., M. Petric, and P. J. Middleton. 1984. Diagnosis of fastidious enteric adenoviruses 40 and 41 in stool specimens. J. Clin. Microbiol. 20:334-338. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Cromeans, T. L., X. Lu, D. D. Erdman, C. D. Humphrey, and V. R. Hill. 2008. Development of plaque assays for adenoviruses 40 and 41. J. Virol. Methods 151:140-145. [DOI] [PubMed] [Google Scholar]
  • 8.Day, R. S., III. 1974. Cellular reactivation of ultraviolet-irradiated human adenovirus 2 in normal and xeroderma pigmentosum fibroblasts. Photochem. Photobiol. 19:9-13. [DOI] [PubMed] [Google Scholar]
  • 9.de Jong, J. C., R. Wigand, A. H. Kidd, G. Wadell, J. G. Kapsenberg, C. J. Muzerie, A. G. Wermenbol, and R.-G. Firtzlaff. 1983. Candidate adenoviruses 40 and 41: fastidious adenoviruses from human infant stool. J. Med. Virol. 11:215-231. [DOI] [PubMed] [Google Scholar]
  • 10.Eischeid, A. C., J. N. Meyer, and K. G. Linden. 2009. UV disinfection of adenoviruses: molecular indications of DNA damage efficiency. Appl. Environ. Microbiol. 75:23-28. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Enriquez, C. E., C. J. Hurst, and C. P. Gerba. 1995. Survival of the enteric adenoviruses 40 and 41 in tap, sea, and waste water. Water Res. 29:2548-2553. [Google Scholar]
  • 12.Gerba, C. P., D. M. Gramos, and N. Nwachuku. 2002. Comparative inactivation of enteroviruses and adenovirus 2 by UV light. Appl. Environ. Microbiol. 68:5167-5169. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Grabow, W. O. K., D. L. Puttergill, and A. Bosch. 1992. Propagation of adenovirus types 40 and 41 in the PLC/PRF/5 primary liver carcinoma cell line. J. Virol. Methods 37:201-207. [DOI] [PubMed] [Google Scholar]
  • 14.Harm, W. 1970. Analysis of photoenzymatic repair of UV lesions in DNA by single light flashes. V. Determination of the reaction-rate constant in E. coli cells. Mutat. Res. 10:277-290. [DOI] [PubMed] [Google Scholar]
  • 15.Hilborn, E. D., M. O. Royster, and D. J. Drabkowski. 2002. Survey of U. S. public health laboratories: microbial pathogens on the CCL. J. Am. Water Works Assoc. 94:88-96. [Google Scholar]
  • 16.Jiang, S. C. 2006. Human adenoviruses in water: occurrence and health implications: a critical review. Environ. Sci. Technol. 40:7132-7140. [DOI] [PubMed] [Google Scholar]
  • 17.Kallsvaart, B. F. 2001. Photobiology effects of polychromatic medium pressure UV lamps. Water Sci. Technol. 32:191-197. [PubMed] [Google Scholar]
  • 18.Ko, G., T. L. Cromeans, and M. D. Sobsey. 2005. UV inactivation of adenovirus type 41 measured by cell culture mRNA RT-PCR. Water Res. 39:3643-3649. [DOI] [PubMed] [Google Scholar]
  • 19.Lee, S.-H., and S.-J. Kim. 2002. Detection of infectious enteroviruses and adenoviruses in tap water in urban areas in Korea. Water Res. 36:248-256. [DOI] [PubMed] [Google Scholar]
  • 20.Linden, K. G., G.-A. Shin, J.-K. Lee, K. Scheible, C. Shen, and P. Posy. 2009. Demonstrating 4-log adenovirus inactivation in a medium-pressure ultraviolet disinfection reactor. J. Am. Water Works Assoc. 101:90-99. [Google Scholar]
  • 21.Linden, K. G., J. Thurston, R. Schaefer, and J. P. Malley, Jr. 2007. Enhanced UV inactivation of adenoviruses under polychromatic UV lamps. Appl. Environ. Microbiol. 73:7571-7574. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Malley, J. P., Jr., N. A. Ballester, A. B. Margolin, K. G. Linden, A. Modfidi, J. R. Bolton, G. Crozes, B. Cushing, E. Mackey, J. M. Laine, and M.-L. Janex. 2004. Inactivation of pathogens with innovative UV technologies. AWWA Research Foundation, Denver, CO.
  • 23.Meng, Q. S., and C. P. Gerba. 1996. Comparative inactivation of enteric adenoviruses, poliovirus and coliphages by ultraviolet irradiation. Water Res. 30:2665-2668. [Google Scholar]
  • 24.Nuanualsuwan, S., and D. O. Cliver. 2003. Capsid functions of inactivated human picornaviruses and feline calicivirus. Appl. Environ. Microbiol. 69:350-357. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Nwachuku, N., C. P. Gerba, A. Oswald, and F. D. Mashadi. 2005. Comparative inactivation of adenovirus serotypes by UV light disinfection. Appl. Environ. Microbiol. 71:5633-5636. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Rainbow, A. J., and S. Mak. 1973. DNA damage and biological function of human adenovirus after UV irradiation. Int. J. Radiat. Biol. Relat. Stud. Phys. Chem. Med. 24:59-72. [DOI] [PubMed] [Google Scholar]
  • 27.Shin, G.-A., J.-K. Lee, and K. G. Linden. 2009. Enhanced effectiveness of medium-pressure ultraviolet lamps on human adenovirus 2 and its possible mechanism. Water Sci. Technol. 60:851-857. [DOI] [PubMed] [Google Scholar]
  • 28.Takiff, H. E., and S. E. Straus. 1982. Early replicative block prevents the efficient growth of fastidious diarrhea-associated adenoviruses in cell culture. J. Med. Virol. 9:93-100. [DOI] [PubMed] [Google Scholar]
  • 29.Takiff, H. E., S. E. Straus, and C. F. Garon. 1981. Propagation and in vitro studies of previously non-cultivable enteral adenoviruses in 293 cells. Lancet ii:832-834. [DOI] [PubMed] [Google Scholar]
  • 30.Thurston-Enriquez, J. A., C. N. Haas, J. Jacangelo, K. Riley, and C. P. Gerba. 2003. Inactivation of feline calicivirus and adenovirus type 40 by UV radiation. Appl. Environ. Microbiol. 69:577-582. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Tiemessen, C. T., and A. H. Kidd. 1994. Adenovirus type 40 and 41 growth in vitro: host range diversity reflected by differences in patterns of DNA replication. J. Virol. 68:1239-1244. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Uhnoo, I., G. Wadell, L. Svensson, and M. E. Johansson. 1984. Importance of enteric adenoviruses 40 and 41 in acute gastroenteritis in infants and young children. J. Clin. Microbiol. 20:365-372. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.U.S. Environmental Protection Agency. 2006. Ultraviolet disinfection guidance manual, proposal draft EPA 815-D-03-007. U.S. Environmental Protection Agency, Washington, DC.
  • 34.van Heerden, J., M. M. Ehlers, W. B. van Zyl, and W. O. K. Grabow. 2003. Incidence of adenoviruses in raw and treated water. Water Res. 37:3704-3708. [DOI] [PubMed] [Google Scholar]
  • 35.van Heerden, J., M. M. Ehlers, W. B. van Zyl, and W. O. K. Grabow. 2005. Prevalence, quantification and typing of adenoviruses detected in river and treated drinking water in South Africa. J. Appl. Microbiol. 99:234-242. [DOI] [PubMed] [Google Scholar]
  • 36.Witt, D. J., and E. B. Bousquet. 1988. Comparison of enteric adenovirus infection in various human cell lines. J. Virol. Methods 20:295-308. [DOI] [PubMed] [Google Scholar]
  • 37.Yates, M. V. 2008. Adenoviruses and ultraviolet light: an introduction. Ozone Sci. Eng. 30:70-72. [Google Scholar]

Articles from Applied and Environmental Microbiology are provided here courtesy of American Society for Microbiology (ASM)

RESOURCES