Stability of human enteric viruses in seawater samples from mollusc depuration tanks coupled with ultraviolet irradiation

Authors

  • A. de Abreu Corrêa,

    Corresponding author
    • Laboratório de Virologia Aplicada, Departamento de Microbiologia, Imunologia e Parasitologia, Centro de Ciências Biológicas, UFSC, Florianópolis, Brazil
    Search for more papers by this author
  • D.S.M. Souza,

    1. Laboratório de Virologia Aplicada, Departamento de Microbiologia, Imunologia e Parasitologia, Centro de Ciências Biológicas, UFSC, Florianópolis, Brazil
    Search for more papers by this author
  • V. Moresco,

    1. Laboratório de Virologia Aplicada, Departamento de Microbiologia, Imunologia e Parasitologia, Centro de Ciências Biológicas, UFSC, Florianópolis, Brazil
    Search for more papers by this author
  • C.R. Kleemann,

    1. Laboratório de Virologia Aplicada, Departamento de Microbiologia, Imunologia e Parasitologia, Centro de Ciências Biológicas, UFSC, Florianópolis, Brazil
    Search for more papers by this author
  • L.A.T. Garcia,

    1. Laboratório de Virologia Aplicada, Departamento de Microbiologia, Imunologia e Parasitologia, Centro de Ciências Biológicas, UFSC, Florianópolis, Brazil
    Search for more papers by this author
  • C.R.M. Barardi

    1. Laboratório de Virologia Aplicada, Departamento de Microbiologia, Imunologia e Parasitologia, Centro de Ciências Biológicas, UFSC, Florianópolis, Brazil
    Search for more papers by this author

Correspondence

Adriana de Abreu Corrêa, Laboratório de Virologia Aplicada, Departamento de Microbiologia, Imunologia e Parasitologia, Centro de Ciências Biológicas, UFSC, Florianópolis 88040-970, Santa Catarina, Brazil.E-mail: adriabreucorrea@gmail.com

Abstract

Aims

To evaluate the stability in seawater of human adenovirus (HAdV2), murine norovirus (MNV-1) and hepatitis A virus (HAV) in a shellfish depuration system with and without ultraviolet (UV) treatment.

Methods and Results

Seawater was seeded with viruses and disinfected using a 36 W lamp. Samples were collected at 24, 48, 72, 96 and 120 h; viruses were concentrated and the viral decay was evaluated using molecular and cell culture methods. Based on the molecular results, at 120 h of disinfection, there was a reduction of more than 3 log10 for HAdV2 and HAV; MNV-1, a 4·5 log10 reduction was observed at 72 h. Infectious MNV-1 was not detected after 72 h of treatment; while HAdV2 remained infectious. Seawater not treated demonstrated a progressive viral reduction for the three viruses tested.

Conclusions

The UV reduced the number of viral particles, and the results indicate there is natural and gradual decrease of viral load and viability in seawater.

Significance and Impact of the Study

UV irradiation is the method of choice for shellfish depuration in many countries; this work showed useful information about the viral stability in seawater and application of UV to water disinfection to be used in shellfish depuration tanks.

Introduction

Ultraviolet (UV) irradiation has increasingly been adopted as a favourable 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 the low doses (40–60 mJ cm−2) that are commonly rendered in water treatment plants (Guo et al. 2010). UV is known to act on nucleic acids because the main mechanism of UV action is to break the DNA double-helix structure, resulting in progressive damage to micro-organisms (Upadhyaya et al. 2004). When a micro-organism absorbs a high irradiation dose, biological death occurs; at lower irradiation doses, there is a photochemical transformation of pyrimidine bases, which causes dimer formation and affects the replication of the micro-organism, thus making it noninfectious (Hijnen et al. 2006).

UV irradiation is the method of choice for shellfish depuration in many countries and is advantageous over other methods of seawater disinfection because it does not interfere with the shellfish filtering activity and does not change the physicochemical properties of the water (Food and Agriculture Organization of the United Nations 2008).

The limitations of UV treatment in seawater are turbidity and dissolved salts. These factors affect the UV irradiation transmission through the water column, thereby reducing the biocide action (Richards et al. 2010). A concern is that pathogens associated with particulate matter can be protected from UV disinfection, resulting in a lower inactivation rate than expected. However, the use of a sand filter to remove solids in seawater has been shown to decrease the water turbidity (Rodrick and Schneider 2003).

The recognized effect of environmental parameters on viral stability and decay in the marine environment varies greatly with the type of virus studied. Temperature and UV irradiation are environmental parameters that mostly affect viral particle stability in seawater, while salinity is considered to be a secondary factor in viral inactivation (Le Guyader et al. 1994; Gantzer et al. 1998; Bitton 1999). However, salinity has been indicated as an important factor in the increase of aggregation of viral particles, contributing to a reduction in viral titre in seawater samples (Gerba and Schaiberger 1975). According to Tree et al. (2004), a 99·9% decay of poliovirus in seawater usually occurs between 2 and 5 days at temperatures from 22 to 30°C. However, little is known about the viral stability and disinfection of enteric viruses. In addition to the environmental factors mentioned above, other factors must also be taken into account in studies of viral stability in the marine environment such as the association of viral particles in particulate organic matter, nutrient concentration and the presence of other micro-organisms (Lipp et al. 2001; Fong and Lipp 2005).

In studies undertaken to develop a depuration shellfish system in which natural seawater is used, each component of the system must be tested for its effectiveness. In this context, the goal of this study was to evaluate the viral disinfection and stability of DNA and RNA viruses in natural seawater and the applicability of UV irradiation to inactivate selected viruses to obtain useful information about the method of water disinfection to be used in shellfish depuration tanks.

Material and methods

Virus and cells

The HAdV2 strain (NCPV# 00213; National Collection of Pathogenic Viruses) was grown on A549 cells derived from human lung adenocarcinoma (European Collection of Cell Cultures), both kindly donated by Dr Rosina Gironès from the University of Barcelona, Spain. The HAV strain HM175 was grown on FRhK-4 cells (rhesus kidney-derived cells), both donated from Macquarie University, Sydney, Australia. Cell lines were cultured in Eagle's Minimal Medium (MEM; Sigma-Aldrich, St Louis, MO, USA) supplemented with 10% (v/v) foetal bovine serum (Gibco-BRL; Life Technologies do Brazil Ltda, São Paulo, SP, Brazil), streptomycin (100 μg ml−1), penicillin G (100 U ml−1) and amphotericin (0.025 μg ml−1) (Gibco-BRL).

MNV-1 was propagated in RAW 264·7 cells (a macrophage-like Abelson leukaemia virus-transformed cell line derived from BALB/c mice). This cell line and MNV-1 were also donated by Dr Rosina Gironès from the University of Barcelona, Spain.

Raw cells were cultured in 1X Dulbecco MEM (DMEM; Gibco) supplemented with 10% FBS (low endotoxin serum), 1·5% HEPES, 1% penicillin–treptomycin, 1% nonessential amino acids and 1% l-glutamine.

For viral titre determination, an indirect immunofluorescence assay was used for the HAdV2 and HAV viruses as previously described (Barardi et al. 1999; Sincero et al. 2006). MNV-1 was quantified using a plaque assay according to Bae and Schwab (2008).

Depuration system and natural seawater

The shellfish depuration system tested in this work was designed by Blue Water Aquaculture Company (Florianópolis, Santa Catarina, Brazil) and located in our laboratory. This system was composed of a plastic tank (300 l), a water recirculation pump (flow rate of 1800 l h−1) and a 254 nm UV filter (36 W, monochromatic and low-pressure lamp, ATMAN II) with an estimated intensity of 43·85 mJ cm−2. The water flow into the tank allows the total volume of seawater contact to the UV filter six times per hour, resulting in an accumulated UV dose of 263 mJ cm−2. This tank has a capacity to accommodate 50 dozen oysters divided into four perforated plastic boxes supported by a bracket to prevent direct contact with material excreted by the shellfish. Natural seawater was collected from Barra da Lagoa, a beach eastern of Florianopolis. Prior to use, the water was filtered through a sand filter to remove suspended solids.

Artificial contamination, disinfection and stability assays

Studies of viral disinfection and stability in seawater were performed in triplicate, and each experiment used two purification tanks filled with 300 l of seawater. One tank was connected to the 36 W UV system (Tank 1), and the other tank did not have UV irradiation applied (Tank 2). For seawater spiking, 1000 ml of seawater was inoculated simultaneously with concentrations of HAV, HAdV2 and MNV-1 that would allow the detection of 4 log10 inactivation, and this was used to contaminate 299 l of seawater (to fill 300 l into the depuration tank). After this dilution, the final concentration of each virus was approximately 2 × 102 FFU, 2 × 103 FFU and 3 × 103 PFU per ml of seawater of HAV, HAdV2 and MNV-1, respectively. The seawater in both tanks was subjected to the recirculation process for 120 h; samples of 1 l of seawater were collected at 0, 24, 48, 72, 96 and 120 h from each tank. Immediately after each sampling time, the seawater samples were concentrated by the organic flocculation method described below. The temperature and pH were monitored every sampling period.

Another viral stability assay was also performed in a bench-scale format to confirm the results obtained for seawater without UV treatment in the depuration tank. The assay consisted of 10 l of seawater spiked with MNV-1 and HAdV2 at final concentrations of 107 FFU ml−1 and 108 PFU ml−1, respectively. The seawater was stirred at room temperature, and after 0, 24, 48, 72, 96 and 120 h, 1000 ml was collected and submitted to the same concentration process described to disinfection assays.

Viral concentration and nucleic acid isolation

The viral concentration from seawater was performed according to Calgua et al. (2008) with minor modifications. Briefly, the samples were acidified with 0·1 mol l−1 HCl to reach pH 3·5, and preflocculated skimmed milk solution (1%, w/v) was added (pH 3·5) to a final concentration of 0·01%. The samples were stirred for 8 h at room temperature, and flocculants were allowed to sediment by gravity sedimentation for another 8 h. After this period, the supernatant was aspirated, and the final volume (approximately 100 ml) containing the sediment was transferred to a centrifuge tube and centrifuged at 7000 g for 30 min at 4°C. The supernatant was carefully removed, and the pellet was dissolved in 1 ml of 0·2 mol l−1 phosphate buffer at pH 7·5 (1 : 2 v/v of 0·2 mol l−1 Na2HPO4 and 0·2 mol l−1 NaH2PO4). The concentrate was stored at −80°C. Before the use of the concentrate to infectivity studies, cytotoxicity tests were performed (data not shown) according to Rigotto et al. (2005), and samples were also treated with an antibiotic/antifungal (100 U ml−1 penicillin G, 100 μg ml−1 streptomycin and amphotericin B 0·25 mg ml−1).

The method chosen to isolate the viral genomes from the samples utilized the QIAmp Viral Mini kit (Qiagen, Valencia, CA, USA), which was used according to the manufacturer's instructions. The nucleic acids were eluted in a final volume of 60 μl and stored at −80°C until employed in genome quantification assays.

Quantitative real-time PCR (qPCR and RT-qPCR)

The sets of primers and probes for each virus are described in Table 1 with respective references. The assays were performed with the StepOne Plus Real-Time PCR System (Applied Biosystems, Foster City, CA, USA).

Table 1. Primers and probe sets used to detect HAV, HAdV2 and MNV-1 by qPCR and RT-qPCR
Primers and probesSequence (5′–3′)Reference
HAV
HAV-F ATGTTTGCCAGTGATGATTGAAAAJothikumar et al. (2005)
HAV-RGGAAAGTCTTTAGGGTCTTCTACCTTT
HAV-P 6-FAM AGGATCCCAACACTCTACCCCACCTAAAAAGA-TAMRA
HAdV2
ADFCWTACATGCACATCKCSGGHernroth et al. (2002)
ADRCRCGGGCRAAYTGCACCAG
ADP1 6-FAM–CCGGGCTCAGGTACTCCGAGGCGTCCT-TAMRA
MNV-1
Fw-ORF1/ORF2CAC GCC ACC GAT CTG TTC TGBaert et al. (2008)
Rv-ORF1/ORF2 GCG CTG CGC CAT CAC TC
MGB-ORF1/ORF2 6FAM-CGC TTT GGA ACA ATG-MGBNFQ

For the generation of standards to be used in the real-time PCR assays, three plasmid constructs were employed. For HAdV2, plasmid pBR22 containing the hexon region of HAdV41 was used. For MNV-1 and HAV, the pGEM-T-Easy vector (Promega, Madison, WI, USA) containing the ORF1/ORF2 junction gene and the 5′UTR region, respectively, was used. These plasmids were kindly donated by Dr Rosina Gironès from the University of Barcelona, Spain.

Escherichia coli JM109 cells (Promega) were transformed with the plasmids. Plasmid DNA was purified from bacteria using the Qiagen Plasmid Midi Kit (Qiagen GmbH, Inc., Hilden, Germany) following the manufacturer's instructions, and the DNA obtained was quantified by spectrophotometry. Dilutions of 10−2 to 107 viral DNA molecules per 1 ml were made in 1× TE buffer, pH 8. The standard dilutions were then aliquoted and stored at −80°C until use.

For HAdV2, amplification by qPCR was performed in a 25-μl reaction mixture with PCR master mix (Applied Biosystems). The samples consisted of 10 μl of either the DNA sample (both pure and a 10-fold dilution) or a quantified plasmid DNA, 1× Taq-Man universal master mix, the corresponding primers and the Taq-Man probe. The amplification was performed following the activation of the uracil N-glycosylase (2 min at 50°C), activation of AmpliTaq Gold for 10 min at 95°C and 40 cycles of amplification (15 s at 95°C and 1 min at 60°C).

For detection and quantitation of the MNV-1 and HAV genomes by RT-qPCR, 5 μl of the RNA extraction and a 10-fold RNA dilution were assayed. Amplification was performed in a 20-μl reaction mixture with the RNA UltraSense One-Step Quantitative RT-PCR System (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. The amplification for both viruses was performed after the reverse transcriptase reaction (30 min at 50°C) and activation of the HotStart Taq (15 min at 95°C); 45 cycles of amplification (10 s at 95°C, 20 s at 55°C, 15 s at 72°C) were performed.

The analysis of each standard curves used showed that the three qPCR assays (HAdV2, MNV-1 and HAV) were sensitive (detection limits were around 100 genomic copies per reaction). The Slope values were –3·2, –3·15 and –3·3 for HAdV2, MNV-1 and HAV, respectively; the square regression coefficient (R2) value was ≥0·998 for all three assays.

Infectivity assays

Plaque assay for MNV-1

In this work, the plaque assay was used to evaluate the infectivity of MNV-1 based on the procedure described by Bae and Schwab (2008) with some modifications. Briefly, RAW 264·7 cells were seeded into 60-mm plates at a density of 2 × 106 cells per well and allowed to adhere for 48 h at 37°C in the presence of 5% CO2. To quantify infectious MNV-1, the seawater samples were diluted in PBS at noncytotoxic dilutions. The cell culture medium was removed, and cells were inoculated with 0·5 ml of each seawater dilution. After incubation for 1 h at 37°C in the presence of 5% CO2, the inocula were aspirated and replaced with 2 ml of a solution containing 1·5% SeaPlaque Agarose (Lonza, Basel, Switzerland), supplemented with MEM (according to section 2·1), allowed to solidify and incubated at 37°C for up to 48 h until plaques were visible. To better visualize the plaques, a neutral red solution (final concentration, 0·1%) was added to 2 ml of the agarose–MEM mixture after the solidification step, and the mixture was incubated at 37°C for 6–8 h. The plaques were counted, and the virus titre in the samples was expressed as plaque-forming units per litre (PFU l−1).

Integrated Cell Culture-RT-PCR for HAdV2

For this assay, the mRNA of viable viruses was extracted as previously described for the amplification reactions. The procedures for the infection and viral RNA isolation from A549 cells were performed based on the procedure described by Ko et al. (2003). Briefly, cells were seeded into 24-well plates at a density of 2 × 105 cells ml−1 and allowed to adhere for 24 h at 37°C in the presence of 5% CO2. The samples were inoculated (200 μl) on confluent A549 monolayers for a period of 1 h at 37°C in 5% CO2 atmosphere to allow for viral adsorption; then, the samples were removed and replaced with 1 ml of MEM 1X (2% FSB). The plates were incubated at 37°C at 5% CO2 for 48 h.

Subsequent to this incubation period, the culture medium was removed, and the infected cells were subjected to nucleic acid extraction using Trizol® reagent (Invitrogen, Carlsbad, CA) according to manufacturer's instructions to obtain the viral RNA. To eliminate DNA contamination before gene amplification, 1 μg of RNA extracted from the samples was treated with 4 U of DNAse I (deoxyribonuclease I; Qiagen) and was then added to 1X reaction buffer (provided with the enzyme), 20 U RNAse inhibitor (RNAse Out; Invitrogen) and RNAse-free water to a final volume of 20 μl. This reaction mixture was incubated for 15 min at room temperature. To stop the enzymatic activity, the samples were treated with 0·0025 mol l−1 EDTA followed by heat treatment at 65°C for 10 min. For cDNA synthesis, a 5-μl aliquot of RNA isolated from HAdV2-infected cells was heated to 99°C for 5 min and was then chilled quickly on ice for 2 min. The RT reaction was performed using a random primer (hexamer pd(N)6 - 50 A260 units; Amersham Biosciences, Buckinghamshire, UK) and MMLV reverse transcriptase (Invitrogen) following the manufacturer's recommendations. Subsequent amplification of the synthesized cDNA by PCR and nested PCR was performed using the oligonucleotide primer pairs (Invitrogen Brazil–Life Technologies, São Paulo, SP, Brazil) hexAA 1885/hexAA 1913 (conventional PCR) and nehexAA 1893/nehexAA 1905 (nested PCR) and the PCR conditions as described by Allard et al. (1992). The expected size of the PCR product was 300 bp and 142 bp for PCR and nested PCR, respectively. Amplified fragments were visualized by standard gel electrophoresis with 10 μl of the final reaction mixture in 1% agarose gels stained with 1 μg ml−1 ethidium bromide.

Flow cytometry

The methodology for A549 cell infection and HAdV2 detection by flow cytometry was based on the procedure described by Barardi et al. (1999) with minor modifications. Briefly, the cells were seeded into 24-well plates, and samples were inoculated on A549 monolayers according to describe at the subsection of Integrated Cell Culture-RT-PCR. The plates were kept at 37°C and 5% CO2 for 72 h. After this incubation period, the maintenance medium was removed. To harvest the cells, 150 μl of trypsin (2·5 mg ml−1) was added to each well. After 5–10 min at 37°C, the detached cells were recovered, resuspended in 1 ml of blocking solution (BS) containing 1X PBS, 1% bovine serum albumin, 0·05% Tween-20 and 0·005 mol l−1 EDTA transferred to a microtube and counted using the Countess® Automated Cell Counter (Invitrogen), according to the manufacturer's instructions. The cells were centrifuged at 3000 g for 4 min at room temperature, and the pellet was resuspended in 100 μl of BS. After this step, the cells were fixed with 900 μl of chilled methanol for 5 min. The cells were incubated with 100 μl of HAdV monoclonal antibody (MAb8092 Nihon Millipore, Tokyo, Japan) in a 1 : 400 dilution in BS and then incubated for 1 h at 37°C on a rotator. The cells were washed three times with BS and incubated for 15 min with antimouse IgG conjugated with FITC (Sigma-Aldrich®) that was diluted 1 : 100 in BS. After this period, the cells were washed three times in BS, centrifuged and resuspended in 1 ml of BS. Cells were then analysed using a Becton Dickinson FACSCanto II flow cytometer (BD Biosciences, San Jose, CA, USA); a total of 10 000 cells were analysed for each sample.

Data analysis

The number of genome copies per litre (GC l−1) obtained in the qPCR and RT-qPCR, which is the measure of viral decay, is presented as the log10 reduction value obtained in each of the performed assays. To evaluate the log10 decrease for each experiment, the viral concentration in the untreated control samples at time 0 were expressed as ‘N0’ (initial virus concentration). For each test sample, the average concentration of virus (in triplicate assays) was determined. The percentage of initial viruses remaining at each time (t) was computed as the virus concentration at each test time (Nt)/initial virus concentration (N0). These values were then log10 transformed (log10 (Nt/N0)). The mean data for Nt/N0 from three independent experiments were then paired with its corresponding sampling time (t) and plotted. Graphics and Statistical analyses were performed using GraphPad Prism 5.00 software (GraphPad Software, La Jolla, CA). The data were tested using parametric tests (anova, post-test Tukey's multiple comparison test). All statistical analyses used a significance level of 5%.

Results

Disinfection of HAdV2

Based on the qPCR results (Fig. 1), HAdV2 showed a 2·4 log10 reduction in 24 h of contact time with UV irradiation. At the end of the assay (120 h), a reduction of more than 5 log10 was observed. Samples from Tank 2 (without UV) were assayed in the same way and showed a reduction in viral genome copy number at each sampling time. Within 24  h of seawater recirculation, a 0·25 log10 reduction was detected; at end of the assay, a 2·5 log10 reduction was observed. The anova showed statistically significant differences among the results obtained in the HAdV2 disinfection (P ≤ 0·0001). These analyses demonstrated that there were differences observed between recirculation times (P = 0·0001), which were responsible for 48·83% of the total variance, and between the seawater treatment groups with or without UV (P < 0·0001), which were responsible for 19·79% of the total variance.

Figure 1.

Log10 reduction curves for seeded HAdV2 in seawater treated and not treated by UV irradiation (36 W). Each data point is an average of three independent experiments and was evaluated by qPCR. (image) No UV; (image) UV.

The HAdV2 infectivity was assessed by two different methods. ICC-RT-PCR (Table 2) revealed that the virus remained infective after up to 72 h of seawater recirculation in both tanks tested. The analysis by flow cytometry did not provide satisfactory results; the percentage of infected cells in samples treated or not treated with UV showed no difference when compared with the negative control (uninfected cells). For this reason, the other sampling times were not analysed.

Table 2. Infectivity assays to assess HAdV2 and MNV-1 viability in seawater with and without UV treatment. NA: not analysed
Seawater treatmentTimeICC-RT-PCR for HAdV2Plaque assay for MNV-1 (PFU l−1)
UV0 h3/33·6 × 103
24 h1/31·3 × 103
48 h0/37 × 102
72 h1/34 × 102
96 h0/3NA
120 h0/3NA
No UV0 h3/35·1 × 103
24 h3/34·9 × 103
48 h2/33·8 × 103
72 h1/33·2 × 103
96 h0/3NA
120 h0/3NA

Disinfection of MNV-1

Based on RT-qPCR results (Fig. 2), MNV-1 had a 3·1 log10 reduction after 24 h of UV treatment. At the end of the assay, viral genomes were not detected in the samples; after 72 h of UV contact time, a 4·7 log10 reduction was observed. Similar to the data collected for HAdV2, MNV-1 samples from Tank 2 showed viral instability. There were 2 log10 and 3 log10 reductions of viral load after 24 and 120 h of contact time, respectively. These differences were statistically significant as measured by an anova analysis, with P < 0·0001 for both recirculation (57·94% of total variance) and seawater treatment groups (26·58% of total variance).

Figure 2.

Log10 reduction curves for seeded MNV-1 in seawater treated and not treated by UV irradiation (36 W). Each data point is an average of three independent experiments and was evaluated by RT-qPCR. (image) No UV; (image) UV.

A plaque assay analysis was applied to quantify infectious MNV-1, and the presence of MNV-1 was observed for all samples analysed (Table 2) despite a large reduction in the viral particles concentration relative to the genomic copy number for samples from both tanks.

Disinfection of HAV

Due to problems with the antibody used for detection, the assays to evaluate HAV infectivity were not performed. The results by RT-qPCR (Fig. 3) indicated the presence of the HAV genome in all of the samples with a 2·1 log10 and 3·5 log10 reduction after 24 and 120 h of UV contact time, respectively. Similar to the results obtained for the other tested viruses, the reduction in HAV genome copy number was also observed in samples not subjected to UV treatment. Within 24 h of seawater recirculation, the genome copy number quantitation showed a 1·5 log10 reduction; at 120 h of contact time, a 1·7 log10 reduction was observed. An anova analysis showed that the results from the both seawater treatments (P = 0·0005) and the recirculation times (P < 0·0001) were statistically significant and showed 66·37% of the total variance.

Figure 3.

Log10 reduction curves for seeded HAV (HM175) in seawater treated and not treated by UV irradiation (36 W). Each data point is an average of three independent experiments and was evaluated by RT-qPCR. (image) No UV; (image) UV.

Stability tests

To confirm the results obtained from seawater samples from Tank 2, an additional experiment was conducted in a bench scale (10 l of seawater), with higher concentrations of HAdV2 and MNV-1, by a period of 120 h. The result (Fig. 4) confirmed that both viruses have instability in seawater, showing a genomic copy number reduction after 48 h of contact time. For HAdV2, a 1·7 log10 decrease was observed at 120 h; MNV-1 showed a 1·5 log10 reduction at the end of the assay. When these samples were subjected to infectivity assays, a flow cytometry revealed a low sensitivity to detect viruses in seawater samples for HAdV2. However, the plaque assay results indicated the presence of infectious MNV-1 until 96 h of contact time with seawater (data not shown).

Figure 4.

Log10 reduction curves for seeded HAdV2 and MNV-1 in seawater not treated by UV irradiation (36 W) to evaluate the genome copy numbers. Each value corresponds to a unique experiment that was analysed by qPCR (HAdV2) and RT-qPCR (MNV-1). (image) HAdV2; (image) MNV-1.

Discussion

The use of UV for water disinfection is widely used in the treatment of water for human consumption and recreation (Hijnen et al. 2006). The evaluation of UV treatment efficiency in seawater is directly related to the application of UV treatment in aquaculture, specifically in depuration shellfish plants, which require a supply of seawater with good microbiological quality. However, few studies have been published about seawater disinfection in depuration shellfish plants (Hill et al. 1967, 1970; Chang et al. 1998).

The purification system used in this study has the capacity for 50-dozen oysters with 300 l of recirculating seawater. These characteristics indicate that this system can be used in locations far from a seawater source, such as offshore restaurants; thus, the UV efficiency should be evaluated. The results obtained indicate that there are differential rates of HAdV2, MNV-1 and HAV disinfection, as well as instability of viral particles in seawater after 120 h of contact time with UV HAdV2 showed a higher resistance to UV irradiation compared with the other two viruses evaluated.

Previously published works on the disinfection of different HAdV serotypes have shown a significant resistance of this virus to standard protocols of UV disinfection (monochromatic low-pressure lamps) when compared with other viruses such as coxsackie virus, rotavirus and calicivirus (Ballester and Malley 2004; Ko et al. 2005; Eischeid et al. 2009). These studies showed that a 4 log10 reduction in viral concentration of HAdV could be reached with UV doses ranging from 80 to 200 mJ cm−2, depending on the serotype analysed. In contrast, doses from 30 to 40 mJ cm−2 were necessary to inactivate other viruses such as poliovirus (Eischeid et al. 2009). In this study, we observed a strong reduction of the initial HAdV2 concentration and a loss of infectivity after 72 h of UV irradiation with accumulated doses of more than 200 mJ cm−2.

Many works have been published on HAdV inactivation by UV irradiation, and cell culture assays have been used to analyse the viral infectivity (Gerba et al. 2002; Ko et al. 2003; Baxter et al. 2007; Eischeid et al. 2009). With this in mind, two methods were applied in this study to detect infectious HAdV2 in seawater and compare the results with those obtained by qPCR (total genomes). The ICC-RT-PCR method was more effective than flow cytometry in detecting viable HAdV2, showing infectious virus from both tanks for up to 72 h. This method has already been reported in the literature, achieving promising results in environmental samples (Ko et al. 2005; Li et al. 2010)

Analysis by ICC-PCR combines cell culture and molecular detection of the viral genome. The cell culture prior to PCR increases either the infectious virus concentration or the sensitivity of viral detection mainly due to the viruses that do not have a cytopathic effect (Ko et al. 2005). According to Ko et al. (2003), the detection of viral mRNA in this assay would be an indication of the presence of infectious viral particles in the samples because only viable virus can penetrate cells and transcribe mRNA during the replication process.

Another method to study the viral infectivity that was used in this study was based on flow cytometry. This method allows the counting and sorting of a cell suspension in one or more groups according to granularity, size and fluorescence characteristics of each cell type (Quiros et al. 2007). The flow cytometry has been used as an strategy for evaluating the different viral populations in the marine environment (Duhamel and Jacquet 2006), and some studies have employed this method in environmental samples (Barardi et al. 1999; Caballero et al. 2004; Li et al. 2010). The protocol used in this study showed a poor sensitivity (data not shown) to detect infectious HAdV2 in the seawater samples. Similarly, viral detection was not effective even in samples not treated with UV, possibly due to the composition of the seawater and the possible aggregation and reduction of viral load due to the salt concentration. Flow cytometry applications to analyse viral infectivity in seawater samples after a disinfection process are not well reported in the literature. Thus, further studies are needed to optimize this method for seawater samples.

The results observed in this study indicated effective HAdV2 disinfection in seawater treated with UV irradiation at a dose of approximately 263 mJ cm−2. At this dose, a high log10 reduction (>5 log10) in genomic copy number was achieved at the end of 120 h, and a total inactivation of HAdV2 was observed after 72 h. Because these assays were performed in a closed-system depuration and because the seawater passes the filter six times an hour, the cumulative effect of the UV irradiation must be taken into account, helping to explain this finding.

Several previously published studies show conflicting results on the UV dose necessary to inactivate HAdV (Thurston-Enriquez et al. 2003; Ko et al. 2005). The reasons are not well explained but may be related to differences in the stability of virus particles among serotypes, the variability of the infectivity assays used, the differences in the viral stock production (such as number of freezing and thawing cycles or extraction with organic solvents) and the differences in the experimental conditions, especially the method of exposure to UV irradiation as well as the pH and ionic strength of the matrix in which the viruses are diluted.

Another factor that is not usually mentioned but may cause different results in viral inactivation studies is the cell line used for infectivity assays (Sirikanchana et al. 2008). The data observed in this work are similar to results obtained by Eischeid et al. (2009), which showed a 4 log10 reduction for HAdV2 at a dose of 80 mJ cm−2 of UV irradiation when using A549 cells for the infectivity assay. In other studies, PLC/PRF/5 cells (Thompson et al. 2003), HeLa cells (Thurston-Enriquez et al. 2003) and HEK-293 cells were used (Ko et al. 2005). For all of these studies, the results indicated that HAdV are more resistant to UV irradiation than other human enteric pathogens studied, requiring doses of 150 m cm−2 for a 4 log10 reduction. There is evidence that the evaluation of viral infectivity in different cell lines may give different results due the DNA repair system present in each cell line (Eischeid et al. 2009).

A more recent approach to viral disinfection by UV is the use of medium pressure (MP) and polychromatic lamps, which emit different UV wavelengths, including wavelengths that are absorbed by proteins. MP and polychromatic can damage the viral capsid proteins and nucleic acids and is thus more effective in viral inactivation. According to Shin and Lee (2010), MP UV irradiation was more effective than low-pressure lamp (LP) UV irradiation when used in water disinfection for HAdV2. A 3 log10 inactivation was achieved with a dose of 40 mJ cm−2 compared with a 1 log10 reduction at the same UV dose with the LP lamp. In another study from this same research group, HAdV2 inactivation with a dose of 90 mJ cm−2 achieved a 5·3 log10 reduction (Shin et al. 2009). These results indicate that MP lamps may be more effective against HAdV2 than LP lamps and that the efficacy is associated with greater damage to the viral particle.

Despite the significance of noroviruses in food and water contamination, their persistence in different environmental conditions and after UV irradiation has not yet been investigated. In this context, this study evaluated the stability and inactivation of MNV-1, as a human norovírus surrogate, in seawater. The results for the UV disinfection of MNV-1 showed viral susceptibility by both the genome quantification and infectivity assays. The MNV-1 genome was detected in samples treated with UV after 72 h of water circulation with a 4·7 log10 reduction in viral load. In samples with no UV treatment, it was possible to detect the genome after 120 h of contact time, indicating that UV irradiation was capable of damaging the viral RNA.

To our knowledge, this is the first study that describes the MNV-1 stability and disinfection in natural seawater. However, the results can be compared with other matrices. According to Lee et al. (2008), MNV-1 had more UV resistance than other caliciviruses that were also used as human norovírus surrogates. UV doses of 10, 20 and 25 mJ cm−2 led to a 1, 2·8 and 3·3 log10 reduction, respectively, when the viral suspension was prepared in PBS. Previous studies have reported a 4 log10 reduction of feline calicivirus (FCV) after UV treatment at a dose of 19·4 mJ cm−2 (Tree et al. 2004) and a 3 log10 reduction at a dose of 12 mJ cm−2 (Duizer et al. 2004). However, MNV-1 is more susceptible to UV irradiation than other enteric viruses such as HAdV2, as observed in this study.

The MNV-1 stability in seawater without UV treatment may have been influenced by salinity. According to Lee et al. (2008), a high salt concentration leads to the inactivation of MNV-1, indicating that noroviruses can be inactivated more quickly in seawater than in fresh water because seawater has a NaCl concentration of approximately 3% (0·5 mol l−1) NaCl. The results obtained in this study showed a 2 log10 reduction at a concentration of 0·5 mol l−1 NaCl after 72 h of incubation at room temperature. These data corroborate those obtained in our work because we observed a 2·3 log10 reduction after 72 h of water recirculation and a 31og10 reduction after 120 h at 25°C.

Regarding the results observed for HAV disinfection, we detected the viral genome at all sampling times. A 3·5 log10 and 1·7 log10 reduction was observed for UV treated and untreated water, respectively. These log10 reduction values were lower than the values observed for other viruses assayed; however, this result was most likely found because the seeded HAV concentration in seawater was lower than MNV-1 and HAdV2 concentrations, thus preventing a large log10 reduction. According Battigelli et al. (1993), HAV (strain HM 175) has a high sensitivity to UV treatment. The dose of UV required to inactivate more than 4 log10 of virus suspension in PBS was 16 mJ cm−2; in buffered water, a 3 log10 reduction was obtained at a dose of 20 mJ cm−2

In conclusion, overall the data presented indicate there is a natural and gradual decrease of viral viability and a reduction in the viral load due to environmental factors after 120 h in seawater. Treatment with UV irradiation was effective in the viral disinfection of seawater, and the different kinetics of inactivation are related to the physical and chemical characteristics of the virus particles, especially the genome structure, which is primarily affected by the absorption of UV irradiation.

Acknowledgements

Financial support was from CNPq (470808/2009-8), MAPA/CNPq (578200/2008-2) and AECID (Agencia Española de Cooperación Internacional para el Desarrollo). A.A. Corrêa, D.S.M. Souza, L.A.T, Garcia, C.R. Kleeman, V. Moresco and C.R.M. Barardi were supported by scholarships from CNPq. We thank Dr Rosina Gironés and Dr Sílvia Bofill at the University of Barcelona, Spain, for their kind donation of biological material (cells, viruses and plasmids for the establishment of qPCR standard curves) used in this work.

Ancillary