Previous Article | Next Article ![]()
Applied and Environmental Microbiology, October 2005, p. 6049-6053, Vol. 71, No. 10
0099-2240/05/$08.00+0 doi:10.1128/AEM.71.10.6049-6053.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Laboratoire de Microbiologie, IFREMER, BP 21 105, 44311 Nantes cedex 03, France,1 Department of Molecular Virology and Microbiology, Baylor College of Medicine, 1 Baylor Plaza, Houston, Texas 77030,2 Virologie Moleculaire et Structurale, Unité mixte CNRS-INRA, 1 av. de la Terrasse, 91198 Gif-sur-Yvette, France3
Received 10 March 2005/ Accepted 25 May 2005
| ABSTRACT |
|---|
|
|
|---|
| INTRODUCTION |
|---|
|
|
|---|
Working with live bivalve mollusks under laboratory conditions may introduce artificial parameters that do not adequately account for environmental factors (i.e., nutriments, temperature, or suspended matter, which can modify filtration rates) important for interpretation of results (8, 17, 27, 31, 33, 35). The use of large depuration tanks and professional equipment allows better reliability and reproducibility of different environmental parameters (aeration, UV, water quality, speed of water recycling, temperature, disinfection of wastewater). However, under such conditions, the potential environmental impact of live enteric viruses precludes their use, even when the viruses are of animal origin.
Coexpression of viral capsid proteins in the baculovirus expression systems results in the assembly of virus-like particles (VLPs) that maintain the structural and functional characteristics of the native particles: they resemble a real virus, but they are noninfectious (9). These VLPs have been helpful in fundamental research (7, 13) but also in environmental persistence and inactivation studies (5, 22) and in monitoring the fate and transport of viruses in subsurface water (28, 29).
The purpose of the study was to inoculate rotavirus VLPs containing full-length VP2 and VP6 (VLP2/6) and MS2 bacteriophage into bivalve mollusks and subsequently evaluate the effectiveness of seeded rotavirus VLPs to act as a surrogate in oysters over a prolonged period of time in an oyster farm.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Phage.
To produce the F+ RNA bacteriophage strain MS2, 0.25 ml of an overnight culture of Escherichia coli K-12 Hfr (NCTC 12486) was inoculated into 25 ml of a tryptone-yeast extract-glucose medium and incubated for 1.5 h at 37°C. The phage MS2 (ATCC 15597-BI) was then added to a final concentration of 107 PFU/ml and incubated for 5 h at 37°C before addition of 2.5 ml of chloroform (Sigma). After centrifugation for 20 min at 3,000 x g, the supernatant was collected and used directly or frozen at 80°C.
Depuration pilot.
A pilot tank was built on the Prat-Ar-Coum premises (Etablissements Y. Madec, Lannilis, France), as previously described (27). Briefly, each independent calibrated tank was equipped with pumps to provide effective aeration and seawater recycling (5 to 30 m3/h) and a heater/cooler to work at constant temperatures (8 to 40°C). Each tank had a volume capacity of 6 m3 and a maximum oyster weight of 600 kg. The pilot used seawater that was disinfected through a large sand filter and with UV. This pilot system provided all facilities necessary to assess shellfish depuration according to professional practices.
Oyster contamination.
A homogeneous group of oysters (Crassostrea gigas) was selected for study to limit the effects of shellfish size and age on other physiological parameters (filtration rate, growth status, risk of death) that could impact the success of depuration.
Contamination with VLP2/6.
Three different concentrations of VLP2/6 were tested: suspensions of 5 x 1014 (experiment A), 5 x 1012 (experiment B), or 5 x 108 (experiment C) VLP2/6 were diluted in 1 liter of seawater and then added to a tank containing 3 m3 of seawater. The final concentrations were 1.7 x 1011, 1.7 x 109, and 1.7 x 105 particles per liter, respectively, for experiments A, B, and C.
Contamination with VLP2/6 and phages.
A suspension of VLP2/6 (5 x 108 particles) and a suspension of MS2 phages (2 x 109 PFU) were diluted in 1 liter of seawater and then added to a tank containing 3 m3 of seawater (experiment D). Final concentrations were 1.7 x 105 particles and 6.7 x 105 PFU per liter.
For all contamination experiments, VLPs were dispersed in the water of the tank for 20 min by pumping air through the water, and then 10 kg of oysters (representing 100 to 120 individuals) was immersed into the contaminated water for 24 h. After 24 h, 1 liter of water and 10 oysters were collected for assay (day 0). The water in the tank was evacuated, and remaining oysters were placed under depuration conditions.
Oyster depuration and relaying.
Depuration experiments were conducted in identical clean tanks with clean seawater. Seawater (final volume = 3 m3) was pumped from the middle of the estuary and was cleaned by passage through a sand filter and exposure to UV-C at a 25.75-mJ/cm2 minimal dose as used by professional producers. The seawater was changed twice a day for 7 days at 22°C and was under aeration. Twelve oysters were sampled at days 0, 1, 2, 3, 5, and 7 and kept at 4°C during transportation to the laboratory and until analysis.
Relaying experiments were performed using the remaining oysters that had been depurated for 7 days under the conditions described above. The oysters were placed in the estuary in front of the shellfish farm for a natural depuration under the influence of the tides. Depending on the tide, oysters were out of the water twice a day for 2 hours, 20 days per month.
Oysters were sampled every week until the VLP2/6 analysis gave negative results.
Water analysis.
Ten milliliters of a polyethylene glycol 6000 (Sigma, St. Quentin, France; 50% [wt/vol])-sodium chloride (1.2 M) solution was added to 40 ml of seawater and rocked for 1 hour at 4°C. After centrifugation for 90 min at 11,000 x g, the pellet was suspended in 200 µl of sterile water and analyzed in duplicate by enzyme-linked immunosorbent assay (ELISA).
Oyster analysis.
For analysis, the oysters were washed, shucked, and weighed. The stomach and digestive diverticula were removed by dissection. All analyses were performed in duplicate. VLP2/6 analysis was performed using 1.5 g of stomach and digestive diverticulum (representing about 1.5 oysters), as previously described (20). Briefly, the dissected tissues were homogenized with glycine buffer, pH 9.6, and the VLPs were eluted using chloroform-butanol and Cat-FLoc (Calgon, Elwood City, PA), followed by polyethylene glycol 6000 (Sigma, St. Quentin, France) precipitation. The pellet was then suspended in 500 µl of sterile water, and 100 µl was used to determine the VLP2/6 concentration by ELISA. The sensitivity limit of the ELISA methods was assessed by adding 100 µl of serial dilutions of VLP to 1.5 g of stomach and digestive diverticulum. Then the tissues were processed as described above.
ELISA.
Falcon 3915 (Probind assay plate; Becton-Dickinson, France) microtiter plates were coated with 200 µl of rabbit antirotavirus serum 8148 diluted 1:2,500 in carbonate-bicarbonate buffer (pH 9). The plates were incubated overnight at 37°C and then washed five times. All washes were performed with phosphate-buffered saline (PBS) containing 0.1% Tween 20 (Sigma Aldrich, France). Then the plates were blocked with 200 µl of PBS-4% nonfat dry milk for 1 h 30 min at 37°C and washed five times. Serial dilutions of VLP2/6 ranging from 103 to 108 particles/ml of shellfish extracts were made in PBS (100 µl) and then incubated in the microtiter plates for 1 h 30 min at 37°C. The plates were washed five times, and 100 µl of peroxidase-conjugated remi-2 antibody (Argène Biosoft, France) diluted 1:100 in PBS was added and incubated for 1 h at 37°C. After five additional washes, 100 µl of 3,3',5,5'-tetramethylbenzidine (TMB; Sigma Aldrich, France) was added and incubated at room temperature in the dark for 15 min. The reaction was stopped by the addition of 100 µl of 1 M H2SO4, and optical density was read at 450 nm with a spectrophotometer (Safire, Tecan, France) (22).
MS2 phage analysis was performed following homogenization of the dissected tissues with a Waring blender (Bioblock) in the presence of 2 volumes of sterile peptone water and centrifugation for 15 min at 1,500 x g at 4°C. Phages were counted, as previously described, according to the protocol ISO 10705-1 (2). Briefly, 2.5 ml of molten 1% tryptone-yeast extract-glucose agar, held at 45°C, was added to 1-ml volumes of appropriately diluted shellfish tissue homogenates and 1 ml of WG49 host culture. This mixture was stirred and poured onto a previously prepared 2% tryptone-yeast extract-glucose agar base in a 90-mm-diameter petri dish. The overlays were inverted and incubated overnight at 37°C.
RT-PCR.
To check for rotavirus contamination from the estuary during the relaying experiments performed with experiments A, B, and C, nucleic acids were extracted as previously described (20). Briefly, viral nucleic acid was extracted and purified from the suspended polyethylene glycol pellet by digestion with 0.2 mg of proteinase K (Amresco, Solon, Ohio) per milliliter, phenol-chloroform (Applied Biosystems, Foster City, CA) extraction, ethanol precipitation, 1.4% (wt/vol) cetyltrimethylamonium bromide (Sigma) precipitation, and a final ethanol precipitation. Viral nucleic acid was suspended in 100 µl of RNase-free H2O. Reverse transcription-PCR (RT-PCR) was performed using primers targeting the VP6 gene and consisted of 40 cycles of amplification. Negative results were confirmed to be negative by hybridization (20, 36).
Statistical analysis.
The rate of VLP depuration was assumed to be linear (single-compartment model). Linear regression analysis was performed using the StatGraphic software. The probability for a difference between the slopes of the linear regression curves for experiments A, B, and C was calculated and tested for significance at the 0.05 confidence level.
| RESULTS |
|---|
|
|
|---|
Contamination of oysters with VLPs and phage.
Oysters were exposed to aerated seawater contaminated with VLPs (experiments A to D) and phage (experiment D only) for 1 day, and samples were taken to evaluate the contamination. Almost all (>99%) of the VLPs added to the tank were concentrated in the oysters for each of the VLP2/6 quantities seeded into the seawater. For experiment A, the concentration in the oyster was 1012 particles/g of dissected tissue and in the water was 2.5 x 108 particles per liter. For experiment B, the concentration in the oyster was 1010 particles/g of dissected tissue and in the water was 2.5 x 106 particles per liter. For experiment C, the concentration in the oyster was 106 particles/g of dissected tissue and could not be detected in water (below the sensitivity limit of 106 particles per liter). For experiment D, the concentrations in the oyster were 106 particles/g of dissected tissue and 2.6 x 103 PFU/g of dissected tissue. For this experiment, no detection was attempted in the water.
Oyster depuration.
The salinity during the experiments was equivalent to the seawater salinity in the estuary (33 to 34.5
), and oxygen was up to 90% of saturation in all experiments.
As shown by the mean value calculated per day, the initial titer of 1012 particles/oyster at day 0 (experiment A) was reduced by 1.5 log10 units after 7 days at 22°C; for experiments B and C, a reduction of 1 log10 unit was observed after 7 days (Fig. 1). The rates of decline for experiments A, B, and C were similar, as determined by the linear regression analysis (0.236 [95% confidence interval {CI}, 0.317 to 0.154], 0.147 [95% CI, 0.236 to 0.06], and 0.161 [95% CI, 0.240 to 0.08], respectively, for experiments A, B, and C) using a single-compartment model.
|
|
. The results of natural relaying for the three experiments are summarized in Table 1. For experiments A and B, with high concentrations of VLP2/6, the last positive results (more than 104 particles/oyster, the sensitivity threshold of our ELISA method) were obtained at days 82 and 70, respectively. When the initial concentration was lower (105 VLP per oyster in experiment C), VLP2/6 could be detected for up to 37 days. Controls examined for rotavirus by RT-PCR were negative, excluding the possibility of external virus contamination during relaying.
|
| DISCUSSION |
|---|
|
|
|---|
In the natural environment (i.e., the estuary), the VLP concentration decreased slowly to reach undetectable levels after a few weeks for the low initial concentration in shellfish and after a few months for the high initial concentration. Recent quantitative data obtained using real-time RT-PCR and most-probable-number RT-PCR have shown virus concentrations of about 102 to 103 genome copies per oyster (21, 23, 26), which is lower than the sensitivity threshold of the procedure for VLP detection employed in this study. Moreover, to be able to follow a 4 log10 decrease in VLP concentration (as suggested for water treatment by Gerba et al. [14]), we had to increase the tracer concentration to be able to follow the relaying, and it is not clear whether the use of higher levels of VLPs may have affected our results. But this may be representative of contamination found in shellfish from prohibited areas close to sewage discharge.
This is the first study that clearly demonstrates that a virus surrogate (virus-like particles) persists in shellfish for several weeks in a natural environment. However, this persistence was suggested as early as 1973 when an outbreak of hepatitis A was linked to shellfish consumption (25). The shellfish were likely to have been contaminated following a flood in the spring and retained infectious virus for 1 to 2 months before the outbreak. Virus carriage for an extended period of time has been suspected previously to have an important impact on public health (8, 16, 25), after detection of infectious hepatitis A virus for at least 3 weeks and for 6 weeks by RT-PCR in an oyster laboratory experiment (17). After a norovirus outbreak following oyster consumption, the follow-up of the harvest area for approximately 2 months showed the persistence of the virus strain contaminating shellfish at levels up to 1,000 RT-PCR units per oyster prior to depuration of the shellfish (21). However, in this case, the possibility of continued exposure of the shellfish could not be excluded. One advantage of the VLPs, as they contain no viral RNA, is that the possibility of a recurrent contamination by rotavirus from the environment can be evaluated by RT-PCR. Although we did not directly assess the physical integrity of the VLPs over the course of the study, previous studies have shown that the rates of decline of live rotavirus and rotavirus VLP2/6 are similar in seawater (22), and it is likely that nonparticulate antigen would be digested by the shellfish.
In summary, this study demonstrated that VLPs can be used in field experiments and that this virus surrogate can persist in shellfish for long periods of time. The data presented here also suggest that after contamination by human enteric viruses, shellfish may be unsafe for human consumption for quite a long period of time. Virus persistence may explain the number of outbreaks linked to depurated shellfish meeting coliform standards but still contaminated by viruses, and virus particles may be detected outside the digestive tract (11, 32, 33). Such a long persistence needs to be confirmed and validated for other enteric viruses such as norovirus and for shellfish exposed to contaminants over prolonged periods. For such studies, VLPs are very good surrogates, as they are available now for most enteric viruses.
| ACKNOWLEDGMENTS |
|---|
We acknowledge the helpful assistance of Annie Charpilienne (CNRS-INRA) in the production of rotavirus VLPs and Cecile Le Mennec for oyster experiments. We acknowledge Y. Madec, Prat-Ar-Coum Farm, for housing the depuration and relaying experiments. We are indebted to M. K. Estes (Baylor College of Medicine, Houston) for helpful advice and discussion.
| FOOTNOTES |
|---|
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| J. Bacteriol. | Microbiol. Mol. Biol. Rev. | Eukaryot. Cell | All ASM Journals |
|---|