Previous Article | Next Article ![]()
Applied and Environmental Microbiology, March 2006, p. 1800-1809, Vol. 72, No. 3
0099-2240/06/$08.00+0 doi:10.1128/AEM.72.3.1800-1809.2006
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
Food Animal Health Research Program, Department of Veterinary Preventive Medicine, Ohio Agricultural Research and Development Center, The Ohio State University, Wooster, Ohio 44691-4096,1 Laboratoire de Microbiologie, IFREMER, 44311 Nantes Cedex 03, France,2 Veterinary Science and Microbiology Department, University of Arizona, Tucson, Arizona 857213
Received 15 September 2005/ Accepted 14 December 2005
|
|
|---|
|
|
|---|
The percentage of outbreaks associated with seafood is around 10 to 20% in countries such as the United States and Australia, but this percentage increases to 70% in countries in which seafood consumption is greater, such as Japan, or wherever seafood is eaten raw (7, 33). Seafood includes both fish and shellfish (mollusks and oysters). Mollusks are filter feeders that can concentrate more than fourfold in their tissue particles present in the surrounding water, including fecal coliforms and viruses (6). An etiological agent has been confirmed in only 44% of the outbreaks associated with seafood, and 47% of the outbreaks associated with seafood and with confirmed etiology are caused by viruses (7, 64).
Viruses in the family Caliciviridae are divided into four genera, Norovirus, Sapovirus, Vesivirus, and Lagovirus. Human and animal caliciviruses associated with gastroenteritis belong to the genera Norovirus and Sapovirus. Human noroviruses (HuNoVs) cause illness in people of all ages, whereas human sapoviruses (HuSaVs) cause illness primarily in children (19). The genus Norovirus is divided into five genogroups, genogroups I, II, III, IV, and V, which can be subdivided into 8, 19, 2, 1, and 1 genotypes, respectively, based on phylogenetic tree topology and distance analysis of the capsid gene (62). Human noroviruses belong to genogroups I, II, and IV. The genus Sapovirus is also divided in five genogroups that can be subdivided into three, three, one, one, and one genotypes based on a similar analysis of the capsid sequence (14). Human sapoviruses belong to genogroups I, II, IV, and V.
Animal enteric caliciviruses cause gastroenteritis in calves and pigs and have also been isolated from healthy pigs. Porcine norovirus (PoNoV) has been detected in Japan and Europe and was recently found in the United States (55, 57, 62). The PoNoVs detected in Japan and the United States belong to three different genotypes in genogroup II, the most widely detected norovirus genogroup in humans. Bovine noroviruses (BoNoVs) have been detected in Germany, England, and the United States, and they belong exclusively to genogroup III (9, 39, 53). Porcine sapovirus (PoSaV) has emerged as an important pathogen associated with diarrhea and subclinical infections in pigs of all ages since it was discovered in 1980 (15, 21).
Low infectious doses (19), prolonged asymptomatic shedding (17), environmental stability (49), and great strain diversity (1, 14) increase the risk of infection by members of this virus family. Moreover, the identification of closely related animal enteric caliciviruses in cattle and pigs and the existence of recombinants of BoNoVs (23, 46), PoNoVs, HuNoVs, and also HuSaVs (26, 29, 30) raise concerns about possible human infections or coinfection of animals or humans with human and animal enteric caliciviruses.
In the United States, HuNoVs accounted for 93 to 96% of the outbreaks of nonbacterial acute gastroenteritis submitted to the CDC from 1997 to 2000 (13). It is estimated that 50 to 66% of all food-borne illness whose etiology is known is due to HuNoVs, and 52% of the gastroenteritis cases associated with consumption of raw or partially cooked shellfish are attributable to HuNoVs, which makes these viruses the leading cause of seafood-associated (especially oysters) food-borne illness (7). The first oyster- and NoV-associated gastroenteritis outbreak involved about 2,000 persons, and it occurred during the summer of 1978 in Australia (43). Since then, several reports worldwide have indicated that NoV-contaminated oysters were associated with outbreaks of gastroenteritis (4, 11-13, 31, 37, 50, 51).
The objectives of this study were to conduct a survey of regional markets in the United States in order to determine the presence of human and animal enteric caliciviruses in market oysters and to compare the calicivirus strains detected in these market oysters with strains previously detected and associated with human gastroenteritis.
|
|
|---|
|
View this table: [in a new window] |
TABLE 1. States and number of bays sampled in each state between the summer of 2002 and the winter of 2002/2003
|
Virus elution and concentration.
A 1.5-g aliquot of homogenate was thawed on ice and processed as previously described by Atmar et al. (3), with minor modifications. The homogenate was homogenized in phosphate-buffered saline (pH 7.4) with a grinder and transferred to a centrifuge tube. Chloroform-butanol was added to remove tissues, and Cat-Floc T (20% polydiallyldimethyl ammonium chloride) (Calgon Corp.) was added to coagulate proteins. After centrifugation the supernatant was recovered, and polyethylene glycol 6000 was added to concentrate the virus. After centrifugation, the pellet was resuspended in 3 ml of diethyl pyrocarbonate (DEPC)-treated water for nucleic acid extraction.
Nucleic acid extraction.
Nucleic acids were purified from concentrated virus as previously described by Atmar et al. (3), with minor modifications. Briefly, after digestion of a pellet with proteinase K (20 mg/ml) (AMRESCO), double extraction was performed with 70% phenol-chloroform-water (Applied Biosystems) and chloroform-isoamyl alcohol (24:1). The aqueous phase was precipitated with ethanol, the pellet was resuspended in DEPC-treated water, and 5% cetyltrimethylammonium bromide (Sigma) was added to remove PCR inhibitors. After 15 min of incubation at room temperature, samples were centrifuged for 30 min at 14,000 x g, and the supernatant was discarded. The pellet was resuspended in 1 M NaCl and reprecipitated with ethanol and 3 M sodium acetate (pH 5.2) for 30 min at 80°C. The precipitated nucleic acids were resuspended in 50 µl of DEPC-treated water with 0.8 U/µl RNasin (Promega, Madison, WI). The presence of natural PCR inhibitors was tested by using an internal control (IC) and performing reverse transcription (RT)-PCR with primers NVp35 and NVp36, which were specific for this IC, as described below.
Detection of PCR and RT-PCR inhibitors in oysters.
An IC was used in the RT-PCR to assess the presence of RT-PCR inhibitors. The IC (kindly provided by M. K. Estes and R. L. Atmar) was generated from a DNA clone containing the region that was amplified by primers NVp35 and NVp36 (4,487 to 4,956 bp) with a 123-bp deletion that yielded, after RNA synthesis by SP6 polymerase, a 347-bp RNA amplicon (50). One microliter of IC was added to each sample during the RT-PCR, and primers NVp35 and NVp36, which were specific for this IC, were used (Table 2). Briefly, for RT, 3 µl of RNA was added to 22 µl of an RT mixture containing 2.5 µl of 10x PCR buffer (100 mM Tris-HCl [pH 9.0 at 25°C], 500 mM KCl, 1% Triton X-100), 2.5 µl of a 25 mM MgCl2 solution, 1 µl of a solution containing each deoxynucleoside triphosphate (dNTP) at a concentration of 10 mM, 1 µl of a 50 µM reverse primer NVp35 solution, 2 U of avian myeloblastosis virus (AMV) reverse transcriptase (Promega), and 8 U of RNasin (Promega), and the mixture was incubated at 42°C for 1 h; this was followed by heat inactivation at 94°C for 3 min. For PCR, another 25-µl PCR mixture containing 2.5 µl of 10x PCR buffer, 2.5 µl of a 25 mM MgCl2 solution, 1 U of Taq DNA polymerase, and 1 µl of a 50 µM forward primer NVp36 solution was added to the RT reaction mixture. After initial denaturation at 94°C for 3 min, 40 cycles of 94°C for 30 s, 50°C for 30 s, and 72°C for 30 s were performed, followed by a final extension at 72°C for 10 min. Amplified products were analyzed by 9% acrylamide gel electrophoresis, stained with ethidium bromide, and visualized under UV light. If inhibition was detected, the RNA template was diluted 1:10 to overcome the inhibition, and the RT-PCR was repeated.
|
View this table: [in a new window] |
TABLE 2. Primers and probes used to detect human and animal noroviruses and sapoviruses in oysters
|
RT-PCR with primers JV12Y and JV13I was also performed in two steps, as previously described by Vennema et al., with minor modifications (58). For RT, 4 µl of reverse primer JV13I (50 µM) was added to 5 µl of RNA, incubated at 94°C for 2 min, and transferred to ice. Six microliters of an RT mixture containing 2.3 µl of DEPC-treated water, 1.5 µl of 10x PCR buffer, 1.5 µl of a 25 mM MgCl2 solution, 0.3 µl of a solution containing each dNTP at a concentration of 10 mM, 2 U of AMV reverse transcriptase, and 8 U of RNasin was added and incubated at 42°C for 1 h, and this was followed by heat inactivation at 94°C for 3 min. For PCR, a 5-µl aliquot of the RT product was added to 45 µl of a PCR mixture, and the PCR was performed as previously described (58).
For animal enteric caliciviruses, RT-PCR was performed with primers CBEC-UF and CBEC-UR for BoNoVs as previously described by Smiley et al. (53) and with primers PEC66-Bio and PEC65 for PoSaVs as previously described by Guo et al. (21), and for PoNoVs RT-PCR was performed with primers PNV7-Bio and PNV8 as described by Wang et al. (62).
Seminested PCR was performed by using the product of RT-PCR obtained with primers NVp110 and NV-4611 as the template for a reaction with primers NVp110 and NI. Briefly, 3 µl of the RT-PCR product was added to 47 µl of a PCR mixture containing 5 µl of 10x PCR buffer, 5 µl of a 25 mM MgCl2 solution, 1 µl of a solution containing each dNTP at a concentration of 10 mM, 1 µl of a 50 µM forward primer NI solution, 1 µl of a 50 µM reverse primer NVp110 solution, and 1 U of Taq DNA polymerase. After initial denaturation at 94°C for 3 min, 40 cycles of 94°C for 30 s, 50°C for 30 s, and 72°C for 30 s were performed, followed by a final extension at 72°C for 10 min. All amplified products were analyzed by electrophoresis using 9% acrylamide gels with ethidium bromide staining and were visualized under UV light.
Hybridization assay for HuNoVs.
Dot blot analysis was used to confirm RT-PCR products, as previously described by Le Guyader et al. (36). Probes GGIa, GGIb, and GGII (59) (Table 2) were labeled with digoxigenin using a 3'DIG oligonucleotide tailing kit (Boehringer Mannheim). The RT-PCR products (obtained with NVp110 and p290 or with JV12Y and JV13I) were diluted in buffer (10 mM Tris-HCl [pH 8.0], 1 mM EDTA [pH 8.0]) and denatured at 95°C. The RT-PCR products were blotted onto a positively charged nylon membrane (Sigma) under a vacuum and fixed by UV light. Positive controls were introduced onto each membrane. After prehybridization at 50°C for 30 min, hybridization was performed for 2 h at 50°C. Hybridized probes were detected by chemiluminescence using a CDP-Star ready-to-use kit (Roche) according to the manufacturer's protocol.
Microwell probe capture hybridization assay for PoNoV and PoSaV.
For PoNoV and PoSaV a microplate probe capture hybridization assay with specific probes was used to confirm RT-PCR products, as previously described by Wang et al. (61). Briefly, EIA/RIA eight-well strips (Corning Inc., Corning, NY) were coated with 100 ng/well of probes PoNoro 1A, PoNoro 1B, and PoNoro 1C to detect PoNoVs or with 100 ng/well of probes PoSapo1A, PoSapo 1B, and PoSapo 1C to detect PoSaVs. The RT-PCR products (obtained with PNV7-Bio and PNV8 or with PEC66-Bio and PEC65) were diluted 1:1 in denaturation buffer and transferred to the strips. Positive and negative RT-PCR controls were included. After incubation, hybridized products were detected with Neutravidin-horseradish peroxidase conjugate (Pierce Biotechnology, Rockford, IL) and tetramethylbenzidine as the substrate (KPL, Gaithersburg, MD). The absorbance at 450 nm was determined with a spectrophotometer.
Sequence analysis.
RT-PCR products were purified from polyacrylamide gels with a Gebaflex kit used as described by the manufacturer (Bioworld, Ohio). The DNA was sequenced directly (when the amount of DNA was enough) or after cloning into the pCR2.1-TOPO (T/A) vector (Invitrogen), using BigDye terminator cycle chemistry and a 3700 DNA analyzer (Applied Biosystems, Foster City, CA). Sequence data were aligned using the Lasergene software package (DNASTAR Inc., Madison, WI) and were compared with the previously published sequences using the Basic Local Alignment Search Tool (BLAST) and Clustal methods.
|
|
|---|
Crude shellfish extracts often contain low virus concentrations compared with the concentrations in feces, in addition to natural RT-PCR inhibitors. To overcome these problems, virus was eluted from the oyster tissue and concentrated. In addition, cetyltrimethylammonium bromide was included in the solution used for extraction of RNA to remove inhibitors, and the presence of inhibitors after RNA extraction was assessed by using an IC. The use of the IC showed that 32 of 45 samples (71%) displayed inhibition (data not shown). To overcome this problem, RNAs were diluted 1:10 in DEPC-treated water before analysis.
Primarily genogroup II HuNoVs were detected in oysters, as determined by RT-PCR and hybridization.
The HuNoVs were detected with one or two genogroup-specific primer sets in 44% (20/45) of the samples (Table 3). In addition, human caliciviruses were detected with broadly reactive primers but not with NoV-specific primers in 3/45 samples (7%). No positive samples were detected with primers NVp110 and NVp36 for genogroup I HuNoVs.
|
View this table: [in a new window] |
TABLE 3. Number of bays positive for human caliciviruses as determined by RT-PCR, nested PCR, and hybridization during the summer of 2002 and the winter of 2002/2003
|
Animal enteric caliciviruses were detected in oysters mainly from the West Coast, as determined by RT-PCR or hybridization.
Fifteen of 45 samples (33%) were positive with specific primer sets for animal enteric caliciviruses (Table 2); 2 of these samples were positive with the BoNoV-specific primer set, 1 was positive with the PoSaV-specific primer set, and 12 were positive with the PoNoV primer set. The microwell probe capture hybridization assay was used to confirm the RT-PCR results for PoNoV and PoSaV. Seven of 12 samples that were positive for PoNoV as determined by RT-PCR were positive as determined by hybridization, and the viruses were confirmed to be genogroup II PoNoVs. Six of these seven samples (86%) were from the West Coast (three samples from Oregon, one sample from California, and two samples from Washington), and one sample was from the East Coast (Delaware). In the case of PoSaV, the RT-PCR product amplified with primers PEC66-Bio and PEC65 was also confirmed by hybridization to be genogroup III PoSaV. The only sample positive for PoSaV was from Oregon. Two samples were positive with primers for BoNoV; one of these samples was from Washington, and the other was from Oregon (Table 4).
|
View this table: [in a new window] |
TABLE 4. Number of bays positive for animal enteric caliciviruses as determined by RT-PCR and microplate hybridization during the summer of 2002 and the winter of 2002/2003
|
|
View this table: [in a new window] |
TABLE 5. Geographic and seasonal distribution of oysters positive for human and/or animal enteric calicivirus from the East, West, and Gulf coasts during the summer of 2002 and the winter of 2002/2003
|
Human noroviruses were detected in 10 states (Washington, Oregon, Alaska, Maine, Virginia, New York, Delaware, Florida, Louisiana, and Mississippi) and were not detected in New Jersey, California, and South Carolina. Although it was clear that when HuNoV was detected in a state, it was present in both seasons (Washington, New York, and Maine), only in the summer (Oregon), or only in the winter (Virginia and Florida), it was also clear that within each state, some bays were positive and other bays were negative at different times (Table 5). In Louisiana, samples from three bays (bays 6G, 7G, and 12G) were positive for human caliciviruses, whereas the other two bays (bays 5G and 11G) were negative. Four of six bays in Florida were negative. In New York, three of four bays were positive (bays 4E, 5E, and 11E). In the case of Alaska and Washington, four bays in each state were sampled. In both states, two bays were negative, but HuNoVs were detected in the other two bays (bays 3W, 4W, 8W, and 11W). All samples from Oregon were positive for HuNoVs (bays 1W, 2W, and 10W).
Animal enteric caliciviruses were detected in three of the four West Coast states (Oregon, Washington, and California) and in only one East Coast state (Delaware) (Table 5). In Oregon, PoNoV was detected in the three bays studied, PoSaV was detected in one bay (bay 1W), and BoNoV was detected in one bay (bay 2W). Two of four bays in Washington were positive for PoNoVs (bays 3W and 6W), whereas BoNoV was detected in a third bay (bay 4W) and one bay was negative for all of the viruses. One bay was sampled in California, and it was positive for PoNoV, as was one of two bays sampled in Delaware.
Human and animal enteric caliciviruses were detected in both seasons in the same or different bays.
Human caliciviruses were detected in both seasons by RT-PCR with different primer sets, but only samples obtained in the summer were also positive as determined by hybridization with probe GGII (HuNoV genogroup II) (3). Human noroviruses were detected only in the summer in Oregon (bays 1W, 2W, and 10W) and only in the winter in Virginia and Florida (bays 3E and 1G). However, in some states, such as Washington, New York, and Maine, HuNoVs were detected in both seasons, in the same or different bays. During the summer in Florida, samples from one bay (bay 8G) were positive for human caliciviruses, whereas oysters from the other two bays were negative, including oysters from bay 1G, from which oysters that were later positive for HuNoV in the winter were obtained. Three other samples collected during the winter in Florida were negative. In New York, three of four bays were positive during the summer (bays 4E, 5E, and 11E), and the bay that was sampled in both seasons was also positive in the winter (bay 4E). In Washington, four bays were sampled during the summer; two of these bays were positive for HuNoVs and two were negative. However, the two positive bays were also sampled in the winter, but only oysters from bay 3W were positive for HuNoVs. In Oregon, HuNoV genogroup II was detected in all bays sampled during the summer. However, in the winter samples collected from two of these bays (bays 1W and 2W) were negative with all of the human calicivirus primers used (including genogroup II-specific primers that showed positive reactions with summer samples from these bays).
Animal enteric caliciviruses were detected in both seasons. However, whereas the seven samples that were positive for PoNoVs were from Oregon, Washington, California, and Delaware bays sampled during the summer, the two samples that were positive for BoNoVs and the one sample that was positive for PoSaVs were obtained from two bays (bays 2W and 4W) and one bay (bay 1W), respectively, in the winter in Oregon and Washington.
Human and animal enteric calicivirus were simultaneously detected in oysters collected from the same bay during the same season.
In the states surveyed with high cattle or swine livestock production, such as Oregon and Washington, HuNoVs and different animal enteric caliciviruses were detected in the same bay in the same or different seasons (Table 5). In Oregon, HuNoVs and PoNoVs were simultaneously detected in oysters collected from all bays sampled during the summer (bays 1W, 2W, and 11W), whereas in the winter PoSaVs and BoNoVs were detected in different bays (bays 1W and 2W) and no PoNoVs or HuNoVs were detected. In Washington, HuNoVs were detected in one bay (bay 4W), PoNoVs were detected in another bay (bay 6W), and both were detected in bay 3W during the summer. In the winter only HuNoVs were detected in bay 3W, and BoNoV, but not HuNoV, was detected in bay 4W.
|
|
|---|
Most food-borne disease outbreaks are due to direct contamination of food or water by a food handler during food distribution at the end of the chain, but contamination may also occur at any point from harvest to table. In the case of seafood, the association of shellfish-transmitted infectious diseases with sewage pollution, symptomatic or asymptomatic cases, or illegal overboard sewage discharges into harvest areas has been well established worldwide, and human enteric viruses are the most common etiological agents transmitted by bivalve shellfish (48). In United States, the first seafood-associated NoV outbreaks occurred in 1980, and they were due to oysters that were harvested in Florida and were contaminated with NoVs (20). Since then, several outbreaks have been traced back to the harvest location in the same state or different states (4, 11). Oyster-related outbreaks will continue unless frequent monitoring and more stringent control measures are established.
In this study, oysters from oyster beds were purchased from different markets in the coastal United States; these oysters were approved for human consumption and were supplied by licensed shippers that harvest oysters from bays on the East, West, and Gulf coasts. To assess the presence of human or animal enteric caliciviruses, 33 bays (12 bays on the East Coast, 12 bays on the West Coast, and 9 bays on the Gulf Coast) during the summer and 12 bays (4 bays on each coast) during the winter (a total of 45 bays) were randomly selected. Different methodologies have been used to assess viral contamination in oysters. Here we used a method based on analysis of dissected tissue (3), which allowed us to obtain a more representative sample of the overall contamination, because oysters from the same bay and harvested at the same time were processed together and pooled.
Enteric caliciviruses (human and/or animal) were detected by RT-PCR with broadly reactive and genogroup-specific primers in oysters from 28 of the 45 bays (62%) during the summer or winter (Tables 3 and 4). Human and animal enteric caliciviruses were detected simultaneously in five samples (11%) (three samples from Oregon, one sample from Washington, and one sample from Delaware); only human enteric caliciviruses were detected in 18 samples (40%), and only animal enteric caliciviruses were detected in five samples (11%). Other reports have shown lower levels of detection in areas that were authorized or not authorized for harvesting. Human noroviruses were detected in 0 to 16% of samples collected from 20 different areas authorized for harvesting (European Community category A, less than 230 Escherichia coli cells per 100 g of shellfish flesh in 90% of the samples; European Community category B, less than 4,600 E. coli cells per 100 g of shellfish flesh in 90% of the samples) in northern and southern Europe (16). In France HuNoVs were detected in 23% of samples collected from authorized harvesting areas (European Community category B) during a 3-year study, and in England NoVs were detected in 27% of samples collected from a highly polluted area (34, 36). In a study performed in the United States in 2003, 52% of samples that were previously depurated and relocated to a prohibited shellfish-growing area were positive for HuNoVs (52). Our results showed a higher percentage of positive samples than the percentages detected in other studies, although the samples came from approved shellfish-growing areas based on the criteria used for bacterial contamination (fecal coliform level). The difference could be a consequence of the sampling protocol, because in our case samples were collected at only two times and in the other studies samples were collected during consecutive months and included many negative samples that decreased the percentage of positive samples. Our results confirm that fecal coliforms are not an appropriate marker for viral contamination and that contamination by virus should be directly assessed until a good marker is found. Other studies have suggested that F-RNA bacteriophage could be used as a viral indicator. However, this proposal is questionable because in some reports these phages correlate well with the presence of human enteric viruses, whereas in other reports the correlation is poor (10, 44).
Hybridization assays with HuNoV genogroup I or II probes were used to confirm RT-PCR products and to group the strains into genogroup I or II. The HuNoV strains detected belonged to genogroup II. However, hybridization did not occur with 11 of 20 RT-PCR samples that were positive for HuNoVs, presumably because of differences in the genome segment used for the probes, as previously reported (60). The samples positive for genogroup II HuNoVs after hybridization included two samples from the Gulf Coast (bays 10G and 12G), two samples from the East Coast (bays 5E and 11E), and five samples from the West Coast (bays 1W, 2W, 3W, 10W, and 11W). Sequence analysis of the 116-bp segment obtained with primers NVp110 and NI for samples from Washington (bay 3W), Oregon (bay 1W), Louisiana (bay 12G), New York (bay 5E), and Mississippi (bay 10G) revealed the highest levels of identity with the norovirus GII/4 strain, particularly with the NoV 95/96-US subset and other Lordsdale strains previously detected in oysters, implicated in diarrhea outbreaks in the United States, and distributed worldwide (13, 45, 51). However, because the sequence analyzed was short, further studies are required to assess strain identity.
Interestingly, three samples (from bays 7E, 7G, and 8G) were positive as determined by RT-PCR with the broadly reactive primers NVp110 and p290 but were negative with all NoV genogroup-specific primers (specific for genogroups I and II). These three samples were also negative with primers designed for animal enteric caliciviruses. These results could be explained by the fact that NVp110 and p290 can also detect SaVs (27). Although SaVs are not frequently associated with food-borne outbreaks (19), the possibility of water contaminated with enteric caliciviruses other than genogroup II NoVs cannot be dismissed. Most of the surveillance systems used so far have tested only for NoVs because of their close association with food-borne outbreaks, but other members of the Caliciviridae family (i.e., SaVs) could also be present.
To our knowledge, animal enteric caliciviruses have not previously been found in shellfish. In our survey, animal enteric caliciviruses were detected by RT-PCR, and the results were confirmed by hybridization or sequence analysis with oysters obtained from 10 of 45 bays distributed on the three coasts. Seven of 45 samples (16%) were confirmed to be PoNoV positive after hybridization. The positive samples included six samples from the West Coast (California, Washington, and Oregon) and one sample from the East Coast (Delaware). Sequence analysis of the 210-bp segment of these samples obtained with primers PNV7-Bio and PNV8 revealed the presence of genogroup II NoVs. In our study PoSaV was also detected in Oregon and BoNoV was detected in samples from Oregon and Washington. BoNoVs, PoNoVs, and PoSaVs have been detected and characterized directly from cattle and swine samples in the United States, as well as in Europe and Japan (55, 57, 62). Phylogenetic analysis has indicated that the BoNoVs and PoNoVs differ from HuNoVs, but they cluster in the NoV genus with BoNoVs in genogroup III and with PoNoVs in genogroup II; the latter is the most frequently detected genogroup in humans. There is little data regarding interspecies transmission of enteric caliciviruses. Animal caliciviruses in the genus Vesivirus have wide host ranges, and interspecies transmission has been documented (54). Although animal enteric caliciviruses have not been isolated from humans, human infection with NoVs related to genogroup III BoNoV was suggested by the presence of antibodies against BoNoV GIII/2 in veterinarians in The Netherlands (63). On the other hand, Oliver et al. (47) and Han et al. (23) have reported that bovine strains are unlikely to be a risk for humans, because they form a third genogroup distinct from the HuNoVs. However in the case of PoNoVs, phylogenetic analysis has indicated that PoNoV strains belong to two distinct PoNoV clusters, clusters 11 and 19, and also to cluster 18, which includes human genogroup II strains (62), the most widely detected genogroup in humans, suggesting that under appropriate conditions interspecies transmission could occur (24).
Human and animal enteric caliciviruses were simultaneously detected in samples collected during the summer in Oregon, Washington, and Delaware. The fact that two or more human viruses can be detected in the same sample has been observed previously. In France simultaneous detection of at least four different human viruses (NoVs, rotavirus, enterovirus, and astrovirus) was described for 10% of samples collected during 3 years (36). The coexistence of several genogroup I and II NoVs and different genotypes of each genogroup was reported for oysters associated with food-borne outbreaks in Japan between December 1998 and January 2001 (28). In this study we focused not only on enteric caliciviruses but also on both human and animal strains. The confirmed presence of two genogroup-related strains (human and/or animal) in the same sample indicates that there is a potential risk of recombination among similar genogroup members in the human host. The detection of HuNoVs and HuSaVs whose genomes were derived from naturally occurring recombinants between members of the same (26, 29, 30) or different (5, 24) genogroups (intragenogroup and intergenogroup recombination, respectively), the existence of chimeric BoNoVs within NoV genogroup III (23, 46), and the recent detection of PoNoVs that belong to genogroup II of the NoV genus suggest that coinfection with genetically closely related human and animal enteric caliciviruses could result in genomic recombination between the viruses (as part of the natural evolution of NoVs) and emergence of new strains relevant to the control of NoV outbreaks in humans.
The seasonal and geographic distribution of the positive samples showed that there was a dynamic pattern. Human caliciviruses were not detected in three states (California, South Carolina, and New Jersey), whereas samples from the other 10 states were positive in the summer (54%, 18/33 samples) and/or the winter (42%, 5/12 samples). Two studies performed in Europe suggested that there was seasonal variation in the number of NoV-positive samples detected and that a larger number of positive samples was obtained during the winter. Formiga-Cruz et al. (16) indicated that there was increased detection of NoVs in samples collected during January and February, whereas Le Guyader et al. (36) showed that in France there was increased detection of HuNoVs in samples collected from November to January. We found a higher prevalence during the summer, but there was not a significant correlation with season, perhaps because a large geographic area was covered by our sampling protocol. Similar to our results, in studies of shellfish from the Norwegian Coast Myrmel et al. (44) did not find a correlation between the presence of HuNoVs in shellfish and season.
Within each state, when calicivirus was detected, it was present in both seasons (Washington, New York, and Maine), only in the summer (Oregon), or only in the winter (Virginia). In addition, some bays were positive and others were negative in the same state at different times. As reported for France, HuNoVs were not always detected at the same site in every season. This observation may be partially explained by the presence of fecal waste from human-associated recreation in the water, illegal dumping, accidental contamination of water with human waste with or without treatment (floods, etc.), or contamination during harvesting by infected but asymptomatic workers.
In contrast to the observations for HuNoVs, animal enteric caliciviruses were present in only four states (Washington, Oregon, California, and Delaware). Also, the seasonal distribution was less variable. Whereas PoNoVs were detected only in the summer, samples positive for BoNoVs and PoSaVs were obtained in the winter. It is not surprising that most of these samples were from Washington and Oregon, because these states have the highest livestock production among the states included in this study. What is noteworthy is the distribution of positive samples in these states. In Washington, HuNoVs were detected in samples from bays 6W and 3W, in which PoNoV was also detected. However, in the winter HuNoVs were detected only in bay 3W and BoNoVs were detected in bay 6W. Also notable was the detection of enteric caliciviruses in Oregon, where HuNoVs and PoNoVs were detected by RT-PCR in the three bays studied, as confirmed by hybridization and sequencing; however, in the winter no HuNoVs or PoNoVs were detected, but PoSaVs and BoNoVs were present.
To our knowledge, no data have been published previously confirming viral contamination in commercial shellfish areas that are representative of a range of geographic locations along the U.S. coasts and during different seasons. Our results support previous reports that confirmed that there was direct enteric calicivirus contamination of harvested seafood (38). Also, the presence of enteric caliciviruses of animal origin in shellfish with potential risk for humans has not been reported previously. To our knowledge, this study is the first study to demonstrate that both human and animal enteric caliciviruses are present in oysters from approved harvesting areas. Because these areas have been approved based on fecal coliform levels, a different marker needs to be used to determine contamination by enteric viruses. The fact that the oysters were purchased from farmers, wholesalers, and retailers suggests that there is a potential disease risk that requires confirmation by outbreak investigation. The simultaneous detection of human and animal enteric caliciviruses raises concerns about possible human infection or coinfection of humans with human and animal enteric caliciviruses, resulting in recombination and the emergence of new strains relevant for the control of disease.
Salaries and research support were provided by state and federal funds appropriated to the Ohio Agricultural Research and Development Center, The Ohio State University, and by NRICGP/USDA Epidemiologic Approaches to Food Safety award 2001-35212-10876.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»