Viruses and bivalve shellfish

International Journal of Food Microbiology 59 (2000) 81–116www.elsevier.nl / locate / ijfoodmicro

Viruses and bivalve shellfish

*David Lees

European Community Reference Laboratory for Bacterial and Viral Contamination of Bivalve Molluscs, Centre for Environment,Fisheries and Aquaculture Science, Weymouth, Dorset DT4 8UB, UK

Received 20 August 1999; accepted 10 February 2000

Keywords: Shellfish; Virus; Public health; Pollution

1. Introduction rman, 1999), only contaminating human viruses haveever been associated with illness in seafood consum-

Two fundamental properties of viruses limit the ers.Viral problems are therefore limited to the role ofextent to which they present a seafood safety prob- food in recycling human viruses back to humans.lem. Firstly, viruses are obligate intracellular para- The viruses most adapted, and likely, to be carried insites depending entirely on a living host cell for the this way are those transmitted by the faecal–oralessential elements of replication. Although viruses route. These include viral agents causing gastro-can be extremely hardy in the external world, for intestinal disease in humans but also agents such asinstance hepatitis A virus has been shown to survive hepatitis A virus and polio virus which althoughfor more than a month in seawater (Callahan et al., being transmitted by the faecal–oral route, and often1995), they are inherently unable to multiply in the having a growth phase in the gut, exhibit theirenvironment or in dead cells. Viruses are not there- classical clinical symptoms elsewhere in the body.fore associated with food spoilage issues. They may Such viruses can contaminate seafood either throughhowever survive well on, or in, food following a contamination at source, principally through sewagecontamination event. Indeed, the common seafood pollution of the marine environment, or in associa-processing procedures of icing and freezing are tion with seafood processing through inadequatelikely to enhance survival of viruses as these are hygiene practices of operatives or systems. Thiswidely used laboratory preservation techniques for review concentrates on seafood contaminationviruses. Secondly, the obligate intracellular nature of through pollution of the marine environment. Con-viruses tends to make them species specific for their tamination during food processing or handling shareshost cell. Although viruses can jump species, or can common features with contamination of other foodhave evolved a life cycle involving multiplication in stuffs and is covered elsewhere.different species, infection of widely divergent Many viruses transmitted by the faecal–oral routespecies is uncommon. So although indigenous are widely prevalent in the community and infectedmarine viruses are the most abundant life form in the individuals can shed millions of virus particles insea, typically numbering ten billion per litre (Fuh- their faeces. Consequently viruses, of many types,

occur in large numbers in sewage. Sewage treatmentprocesses, if present, are only partially effective at*Tel.: 1 44-1305-206-600; fax: 1 44-1305-206-601.

E-mail address: [email protected] (D. Lees) virus removal (Sorber, 1983) therefore coastal dis-

0168-1605/00/$ – see front matter 2000 Elsevier Science B.V. All rights reserved.PI I : S0168-1605( 00 )00248-8

82 D. Lees / International Journal of Food Microbiology 59 (2000) 81 –116

charges constantly release human viruses into the such as mussels and oysters is supplemented bymarine environment. Following shed into the en- breeding and farming introduced species such asvironment viruses can survive for weeks to months pacific oysters and manila clams. Bivalve species(Nasser, 1994; Callahan et al., 1995; Gantzer et al., vary greatly in their characteristics and habitats.1998) either in the water column or by attaching to Those adapted to drying conditions close their shellsparticulate matter and accumulating in sediments. tightly when out of the water to retain a marineThus seafood species harvested from sewage pol- environment around their fleshy internal parts. Suchluted areas may potentially be contaminated with species (oysters, mussels and clams) can survive forhuman enteric viruses. However, a number of factors extended periods out of the water and are historicallydetermine whether such potential contamination con- traded, and consumed, as whole live animals. Otherstitutes a health risk. Major factors include whether species such as cockles are less hardy and areviral contamination remains on the surface or be- normally processed soon after harvest, however, theycomes internalised, and if so whether such contami- may also be traded as live animals if carefullynated organs are consumed or are removed during handled. Scallops and other species not adapted tofood preparation, and how thoroughly the seafood is drying conditions soon die out of water and arecooked before consumption. Of the various seafood normal handled as chilled or processed fisheryspecies although some, such as shrimp (Botero et al., product.1996), have occasionally been shown to be contami- Contamination of bivalve shellfish with virusesnated with enteric viruses only the bivalve molluscan and other agents occurs principally because theseshellfish have consistently proven to be an effective animals obtain their food by filtering small particles,vehicle for the transmission of viral disease. such as algae, from their surrounding water. Many of

the commercial species are common in inshoreestuaries or similar shallow or drying areas wherenutrient levels are high and the waters are sheltered.

2. Bivalve molluscan shellfish Unfortunately such shallow, in-shore, growing wa-ters are also frequently contaminated with human

Bivalve molluscs are a type of shellfish that have sewage. In the process of filter-feeding bivalvetwo shell halves which hinge together. Species shellfish may also concentrate and retain humancommonly commercially exploited in Europe include pathogens derived from such sewage contamination.the native or flat oyster (Ostrea edulis), pacific oyster Bivalve shellfish can also accumulate naturally oc-(Crassostrea gigas), common blue mussel (Mytilus curring toxic algae and pathogenic bacteria throughedulis) and Mediterranean blue mussel (Mytilus filter-feeding activity. The hazards posed by bioac-galloprovincialis), cockles (Cerastoderma edule), cumulation of harmful micro-organisms areking scallops (Pecten maximus) and queen scallops compounded by the traditional consumption of cer-(Chlamys opercularis), and various clams including tain shellfish species raw, or only lightly cooked, andthe native clam or palourde (Tapes descussatus), the by the consumption of the whole animal includinghard shell clam (Mercenaria mercenaria), the manila the viscera. These circumstances are largely uniqueclam (Tapes philippinarum), and the razor shell clam to bivalve shellfish and they therefore represent a(Ensis spp.). With the exception of scallops these are special case among the microbial hazards associatednormally static animals that attach themselves to, or with food. Human health problems associated withbury themselves in the sea bed or other submerged bivalve shellfish are well recognised internationallysurface. Dense beds of the animals can occur in and have been recorded since medieval times. Theproductive areas and have been an important source association of shellfish-transmitted infectious diseaseof food since prehistoric times. with sewage pollution became well documented in

Indigenous species such as cockles, mussels and the late 19th and early 20th century with numerousthe native oyster continue to be harvested from outbreaks of typhoid fever in several Europeannatural populations, however, the characteristics of countries, the USA and elsewhere (Allen, 1899). Inbivalve molluscs also make them suitable for cultiva- more recent years the epidemiological evidencetion. Nowadays the cultivation of indigenous species (reviewed below) suggests that human enteric viruses

D. Lees / International Journal of Food Microbiology 59 (2000) 81 –116 83

are the most common aetiological agents transmitted morphology caliciviruses have characteristic cupby bivalve shellfish. shaped depressions on their surface known as the

Among bivalves the oyster predominates as a ‘‘Star of David’’ morphology. These surface featuresdisease vector in clinical statistics for the UK and are much less distinct in a large number of relatedAustralia. In the USA both the oyster and the clam strains with a generally amorphous surface with aare important vehicles. A large variety of oyster and ragged outline described collectively as Small Roundclam species have been associated with transmission Structured Viruses (SRSV) (Caul and Appleton,of viral disease (Jaykus et al., 1994). The cockle and 1982). This group contains the prototype Norwalkmussel have also been responsible for some large virus, other characterised strains such as Southam-outbreaks both in the UK and elsewhere (see Section pton virus, Hawaii agent, Snow Mountain agent, and4). The association of infectious disease incidents many other less well characterised strains. Recentwith these species probably reflects their traditional taxonomic proposals classify the human entericconsumption raw, or only lightly cooked, and by the caliciviruses into two genus, Norwalk-like virusesconsumption of the whole animal including the (NLV) having a generally undefined SRSV typeviscera. Scallops which are both harvested from less surface morphology and Sapporo-like viruses withpolluted off-shore waters, and are generally eviscer- the more classical caliciviruses morphology (Pringle,ated and cooked, do not present the same infectious 1998). Hepatitis E virus, which like hepatitis A virusdisease hazard as oysters, mussels, cockles and is transmitted by the faecal–oral route and is char-clams. The filter-feeding nature of scallops however acterised by causing liver damage, also hasmeans they may still be associated with health calicivirus like features. However recent taxonomicincidents relating to contamination with marine proposals remove hepatitis E virus from thebiotoxins and naturally occurring bacterial patho- Caliciviridae and leave it not formally assigned togens. Other bivalve species such as razor clams any virus group (Pringle, 1998). These human(Ensis spp.) are not commonly associated with caliviruses are difficult to work with in the laboratoryoutbreaks possibly because they currently occupy a and cannot be cultured by conventional virologicalspecialised, and small, share of the commercial techniques. This has impeded progress in under-market. Gastropods (e.g. winkles and whelks) are not standing of these agents until fairly recently with thefilter feeders, are extensively cooked before con- advent of molecular techniques. Both NLV (see 5.2),sumption, and therefore do not present the same Sapporo virus (Liu et al., 1995) and hepatitis E virusinfectious disease hazards as the bivalve shellfish. (Bradley, 1995) have now been cloned and se-However, like scallops, these species may present a quenced. These molecular advances have paved thehazard from bioaccumulation of marine biotoxins. way for virus characterisation (Clarke and Lambden,

1997) and for important diagnostic advances throughdeployment of molecular techniques such as thepolymerase chain reaction (PCR) and have lead to a

3. Viruses associated with bivalve shellfishbetter understanding of taxonomic relationships(Berke et al., 1997).

A number of human viruses transmitted by theSapporo viruses are genetically distinct from the

faecal–oral route have been associated with shellfishNLVs (Liu et al., 1995) and, although studies are at

either by the isolation of virus from shellfish samplesan early stage, may comprise several distinct strains

or from epidemiological evidence linking them with(Jiang et al., 1997). Seroepidemiological studies have

disease in shellfish consumers.shown a world-wide distribution for these viruseswith a high seropositivity rate in adults (Nakata et

3.1. Caliciviruses al., 1996). Sapporo viruses cause sporadic individualcases and occasional outbreaks of diarrhoea illness

Human enteric caliviruses are a group of related mainly in infants and young children less than 5viruses with a positive sense single stranded nonseg- years old, and also in the elderly. Clinical symptomsmented RNA genome, a generally round morphology are similar to those following NLV infection (de-and small at about 30–35 nm diameter. Classical scribed below). Sapporo viruses have not been


documented as causing infectious disease following demic spread of diarrhoea and vomiting in closedseafood consumption. communities such as hospitals, cruise and military

NLV (Kapikian, 1994) comprises a genetically ships and old people’s homes (Khan et al., 1994;diverse group of virus strains separated into genog- Sharp et al., 1995; Jiang et al., 1996b). The generallyroup I, containing the prototype Norwalk virus, and susceptible nature of the adult population becomesgenogroup II, containing Snow Mountain and other particularly noticeable in outbreaks where attackstrains (Green et al., 1994; Wang et al., 1994; Ando rates can be as high as 90–100%. NLV has beenet al., 1994). Over recent years NLV genogroup II frequently linked with diarrhoea and vomiting (gas-strains have tended to be more prevalent (Fankhauser troenteritis) following seafood, and in particularet al., 1998; Maguire et al., 1999). Until recently bivalve molluscan shellfish, consumption. Patterns ofNLVs had only been recognised as human viruses. immunity may influence this association since adultsHowever, similar diarrhoea causing viruses, closely are generally susceptible and also tend to be therelated by morphological and molecular criteria, primary consumers of seafoods. This is particularlyhave now been recognised in cattle (Dastjerdi et al., so for species such as the bivalve molluscs (oysters,1999; Liu et al., 1999) and also possibly in pigs mussels etc) known to be most at risk from viral(Sugieda et al., 1998). Human NLV is now recog- contamination. The epidemiological features associ-nised to be a major cause of epidemic gastrointesti- ated with NLV associated seafood outbreaks arenal illness occurring in families or in community- described in more detail below (4.2).wide outbreaks. Infections occur in all age groups Hepatitis E virus (Bradley, 1995) shares mor-including older children and adults. The evidence phological and biophysical characteristics withclearly suggests that NLV is the most common cause caliciviruses however there are significant genomicof non-bacterial gastroenteritis in adults (Kaplan et differences. Recent taxonomic proposals leave hepa-al., 1982b; Kapikian, 1996; Vinje and Koopmans, titis E virus not formally assigned to a virus group1996; Caul, 1996a; Fankhauser et al., 1998). Further- (Pringle, 1998). Like NLV, characterisation of hepati-more recent data from the UK suggests that NLV is tis E virus has been achieved largely through molec-the most significant cause of outbreaks of infectious ular methods. Currently a number of aspects of theintestinal disease, causing 43% of outbreaks during natural history of the virus are not well defined1995 and 1996 compared with 15% for salmonellas however it does appear to have a wide distribution(Evans et al., 1998), and also appears to be the most and to infect all age groups. Although a number ofimportant cause of infectious intestinal disease in the genotypes can be differentiated, linked to geographi-community (Tompkins et al., 1999). Clinical symp- cal isolation, there may only be a single serotypetoms include nausea, vomiting, diarrhoea, fever, and (Tsarev et al., 1999). The clinical presentation ofabdominal pain with an incubation period of 1–4 hepatitis E as an acute self-limiting liver diseasedays generally followed by complete recovery with shares many features with hepatitis A (see 3.6) withno complications. As far as is known these viruses some differences notably the high mortality rate inare prevalent throughout the world. Unlike common pregnant women. Hepatitis E is endemic in manycauses of viral gastroenteritis in children, such as developing countries, particularly in Asia, causingrotavirus, astrovirus and enteric adenovirus, NLV both outbreaks and sporadic cases and has beenalso frequently causes symptomatic infection in recognised as a major cause of enterically trans-adults. This may be linked to the immunological mitted non-A, non-B hepatitis. In developed coun-response to these agents which is complex, and not tries, such as the UK, cases are usually associatedwell understood, but appears to be short term with with a history of travel to endemic areas. Mostindividuals becoming susceptible to reinfection with- hepatitis E outbreaks are associated with the con-in six months to a year (Greenberg and Matsui, sumption of contaminated drinking water (Cromeans1992). Consequently much of the adult population et al., 1994). The propensity for water borne trans-appears susceptible to infection with these agents and mission suggests shellfish would be at risk ofepidemic spread can ensue. NLV is highly trans- contamination and could act as a vector for transmis-missible and often becomes noticeable through epi- sion of hepatitis E. Although direct evidence is not


currently available the association of shellfish with and water (Kurtz and Lee, 1987; Oishi et al., 1994)transmission of non-A, non B hepatitis (see 4.1) is and seafood (oyster) (Yamashita et al., 1991; Anon,highly suggestive. 1998) were vehicles of infection. However, direct

epidemiological links with food consumption are not3.2. Astroviruses well documented and there are few such reports

compared with NLV. The importance of astrovirusesAstroviruses (Kurtz and Lee, 1987) have a posi- as causative agents of gastroenteritis following

tive sense single stranded (ss) non-segmented RNA seafood consumption is therefore currently unclear.genome. Virus particles are small, at 28 nm diameter, The increasing availability of molecular diagnosticand round. A proportion of virus particles display a techniques for both outbreak investigation and foodcharacteristic five or six pointed star-shape in their surveillance (see 5.2) should help elucidate the truecentre (hence astro). Astroviruses do not grow prevalence of astroviruses as a cause of diarrhoeareadily in cell culture however some strains have illness following seafood consumption.been adopted to grow in a continuous cell culture(Lee and Kurtz, 1981) and more recently specialised 3.3. Rotavirusescell lines have been used to grow astrovirusesdirectly from stool samples (Willcocks et al., 1990; Rotaviruses (Mattion et al., 1994; Bishop, 1996),Taylor et al., 1997). These advances have helped the established as a genus of the family Reoviridae, aremolecular characterisation of astroviruses (Carter spherical viruses with cubic symmetry. Completeand Willcocks, 1996). Immunological and molecular virions are about 72 nm diameter with a smoothtechniques for diagnosis and characterisation are outer edge. They are named after their wheel likeavailable in specialised laboratories (Cubitt, 1996). morphology in which the double icosahedral shell ofThere are seven recognised human serotypes. As- protein subunits forms a rim round the nucleic acidtroviruses cause sporadic individual cases and occa- core. Rotaviruses have a double-stranded RNAsional outbreaks of diarrhoea illness mainly in genome consisting of 11 separate strands. Theseinfants, young children and the elderly. Astrovirus strands can be readily demonstrated by poly-infections occur throughout the world and in temper- acrylamide gel electrophoresis and provide charac-ate zones are most common during the winter teristic information on virus strains. The majority ofmonths. Until recently astroviruses had been consid- human rotavirus isolates share a common groupered as fairly rare causes of enteric illness however it antigen, with at least seven serotypes existing withinis now becoming clearer that astroviruses are com- this group, and are known as group A rotaviruses.mon agents of childhood diarrhoea (Glass et al., However other viruses have been isolated that are1996b). Clinical symptoms are similar to those morphologically identical but serologically unrelatedfollowing NLV infection and include vomiting, diar- to group A. These are known as group B to Grhoea, fever, and abdominal pain. The incubation rotaviruses. Only groups A, B and C have beenperiod may be longer than for NLV at 2–4 days with associated with disease in humans. Rotavirus strainsillness typically lasting another 2–3 days but occa- can be adapted to grow in cell culture which hassionally 10–14 days (Kurtz and Lee, 1987; Caul, considerably facilitated characterisation. The cultur-1996a). Recovery is normally complete with no able properties of rotavirus, and the presence of largecomplications. Immunity following natural infection amounts of virus in stool, has lead to the develop-is solid, however, it may wane with age and the ment of a variety of diagnostic tests and assayselderly are also a high risk group for astrovirus (Yolken and Wilde, 1994) and consequently rotavir-infection. It has been suggested that the use of uses have been thoroughly investigated. Group Adiagnostic electron microscopy may tend to under- rotaviruses are ubiquitous infections with antibodiesestimate astrovirus prevalence in outbreaks since the detectable by the age of 5 years in virtually allclassical astroviruses morphological features are not children. Group A rotaviruses have been frequentlyalways readily visible in stool specimens. Astrovir- identified as the most important viral pathogen inuses have been implicated in outbreaks where food diarrhoea disease requiring treatment or hospitalisa-


tion in children under 2 years of age. In developing number of serotypes are grouped into six maincountries they are very important pathogens account- subgenera. Adenoviruses are generally readily cultur-ing for some 20% of all diarrhoreal-associated deaths able except for the fastidious enteric adenovirus,in children under five years of age. Rotaviruses are serotypes 40 and 41, predominantly associated withalso important causes of diarrhoea requiring hos- gastroenteritis (Wadell et al., 1994). Enteric adeno-pitalisation of children in industrialised countries. viruses are reported to cause about 10% of infantileHowever, group A rotaviruses are not considered to gastroenteritis and are second only to rotavirus as acause any significant degree of disease in adults. By cause of diarrhoea related hospitalisation in children.contrast non-group A rotaviruses (Bridger, 1994) are Clinical symptoms in children include diarrhoea,much less common and tend to cause sporadic family vomiting and fever. In comparison with rotavirusand community outbreaks of, sometimes severe, infections enteric adenoviruses generally cause a lessgastroenteritis in all age groups. Antibody prevalence severe disease but with a more prolonged course.levels in the community are correspondingly low and Diarrhoea may persist for several weeks infectionadults can get symptomatic infections. among children seems common since serological

The group A rotaviruses therefore normally pres- studies show antibody in about 50% of older childrenent as paediatric infections occurring in infants or and young adults. Both enteric adenoviruses causingearly childhood. Group A rotaviruses are common diarrhoea, and non-enteric adenoviruses associatedinfections and viruses are shed in large numbers in with infection elsewhere such as in the respiratory

12stools ( . 10 particles /g faeces) leading to readily system, may be shed into the gut and be isolatabledetectable virus presence in sewage effluents and from faeces. Although adenoviruses can be detectedpolluted receiving waters (Gajardo et al., 1995; in polluted sewage effluents, in seawater and inDubois et al., 1997). Consequently rotavirus pres- shellfish (Girones et al., 1995; Vantarakis andence has also been demonstrated in bivalve shellfish Papapetropoulou, 1998; Pina et al., 1998b) nogrown in contaminated waters (see 5.2). However seafood related outbreaks have been reported. This isrotaviruses have not been linked with infectious presumably because enteric adenoviruses are notdisease following seafood consumption. The general generally associated with gastroenteric disease inabsence of symptomatic disease in the adult popula- adults and, for the most part, young children are nottion probably influences this lack of association. primary consumers of seafood.Adults, rather than children, tend to be the primaryconsumers of seafoods and adults tend not to demon- 3.5. Enterovirusesstrate clinical disease following rotavirus infection.Age related resistance to severe infection may be due The enteroviruses (Minor et al., 1990) are a genusto active immunity reinforced by repeated infection of the family Picornaviridae which comprises a largethroughout life (Bishop, 1996). Non-group A rotavir- family of viruses. Enteroviruses have a positiveuses could pose a potential threat via contamination sense single stranded nonsegmented RNA genome.of seafood, particularly bivalve shellfish, however The virion is a largely featureless smooth roundthe author is not aware of any such documented nonenveloped particle of about 27 nm diameter.incidents. About 66 immunologically distinct serotypes are

known to cause infections in humans including the3.4. Adenoviruses polioviruses, group A and B coxsackieviruses, ech-

oviruses, and the more recently designated enterovir-Adenoviruses (Wadell et al., 1994) have double uses serotypes 68 to 71 (Muir et al., 1998). All age

stranded nonsegmented DNA genome. The virus groups can become infected. The normal site ofparticle is a symmetrical nonenveloped icosahedron replication is the intestinal tract where infection maywith a diameter of 80 nm. Adenoviruses were often be mild or clinically inapparent. However in aoriginally isolated from adenoids, hence the name, proportion of cases the virus can spread to otherand are common viruses among both humans, birds organs and can cause serious or even fatal disease.and other animals causing respiratory, ocular, gas- The clinical characteristics of which, as for exampletroenteric and other infections. In humans a large with poliomyelitis, may be characteristic of par-


ticular enterovirus types. In addition to poliomyelitis illness whereas overt hepatitis develops in the ma-enteroviruses can cause acute myocarditis, aseptic jority of infected adults. Recovery is complete andmeningitis, hemorrhagic conjunctivitis, congenital leads to long term immunity from reinfection. Hepa-infection of neonates and other non-specific febrile titis A is a common endemic infection in developingillnesses. There is also concern that enteroviruses countries with most children being seropositive by 6may cause or contribute to a range of common years of age. However improving sanitary conditionschronic diseases such as diabetes mellitus. Common- in developed countries have lead to declining preval-ly such acute symptoms occur only in a small ence and resulted in large sectors of the populationproportion of those suffering an enterovirus infec- being susceptible to infection (see 4.1).tion. It should be noted that although enteroviruses During infection virus is released from liver cellsreplicate in the intestinal tract, and are transmitted by into the bile and is then excreted in faeces. Althoughthe faecal–oral route, they do not commonly cause laboratory adapted hepatitis A virus strains can beclassical gastroenteritic symptoms such as diarrhoea grown in culture, wild type hepatitis A virus is moreand vomiting. They are however shed in large fastidious. However hepatitis A virus can be readilynumbers in faeces and most types can be readily demonstrated in stools by molecular techniquescultured from stool samples. Because of their preval- (Yotsuyanagi et al., 1996) and has also been demon-ence and ease of culture enteroviruses have a long strated in sewage effluents and polluted receivinghistory of use as markers of viral pollution of the waters (Tsai et al., 1994) and in bivalve shellfishenvironment. Many reports have documented their using PCR (see 5.2). Bivalve shellfish have frequent-isolation in sewage effluents, polluted receiving ly been implicated in outbreaks of hepatitis Awaters and in bivalve shellfish (reviewed by Jaykus infection the epidemiological features of which areet al. 1994). However despite numerous demonstra- described in more detail below.tions of contamination bivalve shellfish have notbeen linked to transmission of enterovirus disease.The epidemiological considerations are discussed in 4. Epidemiology of shellfish associated illnessmore detail below (4.3).

Despite the variety of viruses transmitted by the3.6. Hepatitis A virus faecal–oral route, many of which can be shown to

contaminate shellfish or their harvesting areas, only aHepatitis A virus (Cromeans et al., 1994) was few have been epidemiologically linked to disease

formerly a member of the genus Enterovirus following consumption of bivalve shellfish or other(serotype 72) and shares morphological and general seafood. Illness has been related to viruses causinggenomic properties, and transmissibility by the faec- gastroenteritis and viruses causing hepatitis. Theseal–oral route, with the enteroviruses as described are further discussed below.above. However it is atypical with regard to itsnucleic acid sequence, its extreme stability and its 4.1. Infectious hepatitissite of replication, currently thought to be the liver,and is now placed in a separate genus, the Hepato- Hepatitis A is the most serious virus infectionvirus (Minor, 1991), within the Picornaviridae. Com- linked to shellfish consumption causing a seriouspared to other enteric viruses hepatitis A virus has an debilitating disease and even, occasionally, death.extended incubation period of about 4 weeks (range The first documented infectious hepatitis outbreak2–6 weeks) and causes a serious debilitating disease occurred in Sweden in 1955 when 629 cases wereprogressing from a non-specific illness with fever, associated with raw oyster consumption (Roos,headache, nausea and malaise to vomiting, diarrhoea, 1956). Since this time many hepatitis A virusabdominal pain and jaundice. Hepatitis A is self- outbreaks world-wide have been linked to bivalvelimiting and rarely causes death but patients may be mollusc consumption and have been reviewed byincapacitated for several months. Age has an im- several authors (Richards, 1985; Rippey, 1994;portant bearing on the severity of the infection with Jaykus et al., 1994). Occasionally such outbreaksyoung children frequently experiencing only mild assume an epidemic scale and are worthy of special


note. The most graphic illustration being an outbreak numbers of cases. The fairly protracted (average 4of Hepatitis A in Shanghai, China in 1988 when weeks) incubation period of hepatitis A virus makesalmost 300 000 cases were traced to the consumption association with a particular food vehicle in in-of clams harvested from a sewage polluted area dividual or sporadic cases very difficult. Normally(Halliday et al., 1991; Tang et al., 1991). This the food is not available for testing and consumptionoutbreak ranks as the largest viral food poisoning histories inconclusive. Generally therefore reports ofoutbreak reported. In the outbreak the estimated shellfish associated hepatitis A virus disease fall intoattack rates in those who had, and had not, eaten two groups: the large outbreaks where a commonclams were about 12% and 0.5% respectively with a vehicle of infection is more obvious, and the smallervery high overall hepatitis attack rate during the clusters or sporadic infections where a food vehicleepidemic of 4083 per 100 000 population (1 person involvement may only become evident throughin 25) (Halliday et al., 1991). The USA has also diligent epidemiological investigation. Since suchreported a number of shellfish associated hepatitis A investigation is the exception rather than the rule theoutbreaks (Richards, 1985; Rippey, 1994) as has the association of shellfish with sporadic hepatitis A isUK (Bostock et al., 1979; Sockett et al., 1985) and probably considerably underreported (Rippey, 1994).other countries such as Italy (Mele et al., 1989; However, retrospective seroepidemiological inves-Malfait et al., 1996) and Japan (Fujiyama et al., tigation can help identify the respective risk factors1985). A hepatitis A epidemic in western France in for hepatitis A virus infection and place shellfish1991/92 involving around 800 cases was considered consumption in the larger disease context. Severalto be associated with shellfish consumption although such studies have been published. In an early studydefinitive evidence was not available (Apairemar- in the USA among infectious hepatitis patients in 10chais et al., 1995). Over 1000 cases of oyster and Boston hospitals consumption of raw shellfish orclam associated hepatitis A occurred in several US steamed clams was as significant a risk factor as wasStates during several major outbreaks from 1961 to exposure to a jaundiced person (Koff et al., 1967). In1964 (Richards, 1985). Since then further sizeable a German study between 1968 and 1971 shellfishepisodes occurred in 1973 due to Louisiana oysters consumption was linked to infectious hepatitis in(Glass et al., 1996a) and in a multistate outbreak in 19% of patients attending a clinic in Frankfurt (Stille1988 due to oyster from Florida (Desenclos et al., et al., 1972). Similar data was reported in a UK1991). The Louisiana outbreak was attributed to study which showed a significant correlation betweensewage pollution of harvesting areas through flood- infectious hepatitis and cockle consumption with asing of the Mississippi and the Florida outbreak to many as 25% of cases attributed to shellfishillegal harvesting from contaminated areas. In the (O’Mahony et al., 1983). In a Japanese study 651Louisiana outbreak hepatitis A virus was suspected patients with acute viral hepatitis identified serologi-to have been retained in shellfish for at least 6 weeks cally between 1976 and 1985 were examined for riskfollowing the contamination event (Glass et al., factors occurring within 6 months of the onset of1996a). At the time of harvesting oysters were fully infection (Kiyosawa et al., 1987). Of the patientscompliant with the USA sanitation program stan- with hepatitis A, ingestion of raw shellfish was thedards. These reported hepatitis A outbreaks have highest identified risk factor (11%) slightly exceed-generally been linked to shellfish consumption by ing familial contact with a patient (10%), the otherepidemiological data although hepatitis A virus was main risk factor identified. Studies in Italy havedetected by molecular techniques in the harvesting consistently identified shellfish consumption as oneareas associated with the Florida outbreak (Desen- of the most important risk factors (Mele et al., 1990;clos et al., 1991) and by isolation in primates in the Mele et al., 1991). Recent estimates suggest thatShanghai outbreak (Li et al., 1992). shellfish may be responsible for 70% of all hepatitis

Despite the incidents described above hepatitis A cases in Italy (Salamina and D’Argenio, 1998).outbreaks are less frequently reported than outbreaks These various studies suggest that shellfish may bewith other clinical presentations, particularly gas- significant vehicles of hepatitis A virus in sporadictroenteritis. It is also notable that many of the cases as well as identified outbreak cases. However ahepatitis A virus outbreaks reported involve large large case control study of sporadic hepatitis A in


England carried out from July 1990 to June 1991 did linked to shellfish consumption (Jaykus et al., 1994;not support these findings (Maguire et al., 1995). Torne et al., 1988) and shellfish have been possiblyAlthough shellfish consumption appeared to be sig- implicated in transmission through seroepidemiologi-nificant at the 5% level in a simple model it failed to cal studies (Mele et al., 1986). It is probable thatretain significance when confounders were ex- such cases are caused by faecal–orally transmittedamined. It may be significant that during this study hepatitis E virus although this has not yet beenperiod there was only a single reported outbreak of formally established. Hepatitis E is not endemic toshellfish associated hepatitis A virus in England the UK and other developed countries and most(Sockett et al., 1993). Possibly recognisable out- cases are travel associated. However, shellfish maybreaks reflect a larger burden of sporadic shellfish- be vectors for hepatitis E in endemic areas such asborne disease in the community. It is noticeable that India and it is possible that globally traded shellfishearlier seroepidemiological studies in the UK were could cause infections in non-indigenous areas.performed in years when several large outbreaks Recently hepatitis E has been demonstrated in sew-were reported. age samples from a non-endemic area (Barcelona,

A common feature of hepatitis A in many de- Spain) raising the concern that shellfish could act asveloped countries is the generally declining inci- vectors for this virus even in non-endemic areasdence of infection as shown by seroepidemiological (Pina et al., 1998a).surveys (Mele et al., 1990; Koff, 1995; Gdalevich etal., 1998). Seropositivity rates in the older population 4.2. Viral gastroenteritismay now be much higher than in young adults. Inendemic areas hepatitis A is generally acquired in Gastroenteritis has been recognised as a clinicalchildhood when the disease may be mild or subclini- consequence of the consumption of contaminatedcal and generally acquired by social contact. In many shellfish for many years (Richards, 1985; Rippey,developed countries declining endemic exposure of 1994). The clinical symptoms in such cases are oftenchildren leaves the adult population exposed to the similar and are typically described as relativelypotential for hepatitis A virus epidemics from vectors ‘mild’ gastroenteritis, often including diarrhoea andsuch as food or water or to acquiring infection from vomiting, with an onset of 24–28 h, a duration oftravel to an endemic area (Armigliato et al., 1986; about 2 days and rarely requiring care from aPebody et al., 1998). Although shellfish associated physician. Since the last cases of typhoid fever in thehepatitis A virus infections may currently be un- fifties bacteria have rarely been identified as thecommon in some developed countries, such as the cause of shellfish-vectored enteric illness and certain-UK, continuing vigilance is important as shellfish ly the majority of cases of gastroenteritis appeared tohave demonstrated on many previous occasions their be non-bacterial in nature. Indeed until the 1970spotential to act as vectors for hepatitis A virus laboratory investigation generally failed to reveal aepidemics and the adult population is now largely causative agent in the large majority of such inci-susceptible. Australia suffered such an explosive dents. However the clinical symptoms were charac-outbreak of oyster mediated hepatitis A during 1997. teristic of viral gastroenteritis, such as caused byAlthough official reports are not yet available reports NLV, and it is now widely recognised that these viralin the media and from subsequent court proceedings agents are the major cause of shellfish-vectoredsuggest this was the first such hepatitis A virus gastroenteritis.epidemic for a number of years and that over 400 The first linkage of viruses with shellfish-bornepeople were infected (Anon, 1999b). The impact of gastroenteritis infection was made in the winter ofsuch an epidemic on both the oyster industry, the 1976/77 in the UK when cooked cockles werehealth services and the regulatory authorities should epidemiologically linked to 33 incidents effectingnot be underestimated. In Australia expanding coast- nearly 800 people (Appleton and Pereira, 1977).al populations in fragile estuarine locations with Small round virus like particles, like those seen ininadequate sewage systems were highlighted as the outbreaks of ‘winter vomiting disease’, were ob-probable cause. served by EM in a high proportion of patients faeces.

Cases of non-A non-B hepatitis have also been Subsequently the same author examined clinical


samples associated with 9 separate shellfish-vectored agent based on the clinical features of disease, ongastroenteritis outbreaks and showed similar small seroconversion to Norwalk virus and by demonstra-round virus like particles in about 90% of samples. tion of Norwalk virus contamination of shellfish byThis evidence indicated for the first time the prob- radioimmunoassay. Shellfish originated from severalable broad significance of viruses in shellfish trans- north-eastern states with faecal contamination causedmitted infections in the UK (Appleton et al., 1981). by runoff due to heavy spring rains the most likelyMeanwhile in Australia a large oyster associated cause. A further large incident suspected to begastroenteritis outbreak involving some 2000 persons caused by Norwalk virus occurred in Louisiana inoccurred during the summer of 1978 (Murphy et al., 1982 with 472 cases linked to oysters (Richards,1979). Laboratory investigations failed to show a 1985). These and other outbreaks during the earlybacterial aetiology but ‘parvovirus-like’ virus par- 1980s prompted dealers to import purified (depu-ticles were seen by EM in patients faeces and rated) English clams during 1983. These were re-development of a related antibody response shown sponsible for a further wave of at least 14 separateby immune electron microscopy. Further examina- suspected viral gastroenteritis outbreaks involvingtion identified the virus particles as Norwalk virus some 2000 consumers in New York and New Jerseydemonstrating for the first time the association of this over a 3 month period during 1983 (Richards, 1985).virus with shellfish-vectored disease. This outbreak In one outbreak alone over 1100 persons reported illwas linked to sewage contamination of the oyster after consumption of clams at a picnic. In the UKharvesting area near Sydney caused by heavy rain- oysters were implicated in an outbreak in London infall. Frozen oysters from the same harvesting area 1983 causing illness in over 300 people (Gill et al.,caused a related outbreak in Darwin, Australia in 1983). The attack rate was 79% among oyster eatersDecember 1978 and Norwalk virus was again con- and NLV was confirmed as the aetiological agent byfirmed by both serology and by EM in patients stools detection in stools by electron microscopy and by a(Linco and Grohmann, 1980). Subsequent efforts in rise in antibody to small round viruses. Oysters hadAustralia to improve the food safety of oysters again been depurated for 72 h prior to consumption.demonstrated the importance of Norwalk virus as a Further gastroenteritis outbreaks associated withcause of gastroenteritis when a number of volunteers depurated shellfish (oysters) were reported in the UKin consumption trials became ill with this agent in 1986 (Heller et al., 1986), 1993 (Chalmers and(Grohmann et al., 1981). Again illness was associ- Mcmillan, 1995), 1994 (Perrett and Kudesia, 1995)ated with heavy rainfall, and hence probable sewage and 1997 (Ang, 1998). Investigation of two incidentscontamination, in the oyster harvesting area. in New York State during 1983 provided the first

Since these early reports numerous publications documented shellfish associated outbreak of Snowhave documented shellfish associated outbreaks of Mountain agent, another NLV strain (Truman et al.,gastroenteritis caused by NLV and incidents up to 1987). Attack rates in those who eat clams were 55%about 1990 have been reviewed by several authors and clam eaters were more likely to seroconvert to(Richards, 1985; Rippey, 1994; Jaykus et al., 1994). Snow Mountain Agent than controls. Clams wereIn the USA the first cases of shellfish-associated harvested in Massachusetts during a period of heavygastroenteritis shown to be caused by NLV occurred rainfall and high runoff. In 1984 and 1985 furtherin 1980 among individuals consuming oysters har- outbreaks of shellfish-associated probable viral gas-vested in Florida (Gunn et al., 1982). Since this time troenteritis were reported in New York State (Rich-a succession of suspected or confirmed shellfish ards, 1985; Guzewich and Morse, 1986). In 1990associated viral gastroenteritis incidents have been another major incident involving oysters from thereported in the USA (listed up to 1985 by (Richards, same area as previous outbreaks occurred in Aus-1985). During 1982 the situation was judged to have tralia (Bird and Kraa, 1995). Fifty-seven separatetaken on epidemic proportions with 103 well docu- outbreaks of oyster-associated viral gastro-enteritismented outbreaks in which more than 1000 persons occurred over an 18 day period: the incident fol-became ill from eating clams or oysters in the State lowed a period of heavy rainfall with the resultantof New York alone (Morse et al., 1986). Norwalk flooding of the sewage system and discharge of largevirus was implicated as the predominant aetiological amounts of crude sewage. In late 1991 an outbreak


of gastroenteritis following the consumption of raw and by seroconversion to Norwalk virus in patientsoysters provided the first documented shellfish asso- (Kohn et al., 1995b). The oysters were harvested byciated NLV outbreak in Canada (Pontefract et al., boat from a remote bed thought to be free of any1993). Some 200 persons were effected. NLV was sewage pollution. However the investigation showeddemonstrated by electron microscopy in stools and oyster harvesters routinely disposed of faeces over-an antibody response to these virus particles demon- board and furthermore that one harvester, with a highstrated by immune electron microscopy. The source level of antibodies to Norwalk virus, had ex-of oyster contamination could not be determined. perienced gastroenteritis just prior to the responsibleGastroenteritis outbreaks connected to oysters have harvesting period. It is remarkable that faeces from abeen reported from a number of districts in Japan. In single individual may have been responsible forKyushu district four of five oyster related outbreaks contaminating such a large quantity of oysters andbetween 1987 and 1992 were shown to be caused by powerfully illustrates the bioaccumulative potentialNLV (Otsu, 1999). In Gifu prefecture 2 oyster of bivalve shellfish. Molecular epidemiological ap-outbreaks occurring in 1989 and 1991 were shown to proaches were applied, possibly for the first time, tobe caused by NLV (Kawamoto et al., 1993). Attack the investigation of this NLV outbreak. PCR fol-rates among oyster eaters were 68% and 92% for the lowed by sequencing was applied to stools from aoutbreaks. Western blotting immunoassay provided number of cases from different clusters of illness andadditional information and showed the NLV strain this clearly showed that clusters were linked anddetected to be antigenically related to Hawaii agent caused by a single unique NLV strain. This approach(genogroup II) but not to Norwalk virus (genogroup was also able to distinguish this large outbreak fromI). Investigation of a further four outbreaks were a smaller co-incident cluster of cases associated withreported in Shizuoka district between 1987 and 1994 Florida oysters, an outbreak in a nursing home, and awith both genogroup I and genogroup II NLV strains case in an oyster packer, which were all shown to bedemonstrated by molecular techniques (Sugieda et caused by unrelated NLV strains (Ando et al., 1995a).al., 1996). During recent years this application of The cluster of cases in Florida were separatedmolecular techniques to NLV (see 5.2) has raised investigated and shown to be caused by locallynew possibilities for the investigation of outbreaks. harvested oysters and to be unrelated to the mainVirus genome characterisation by sequencing enables Louisiana outbreak (Davis et al., 1994). A furtherthe definitive linkage of patient and vector, the large gastroenteritis outbreak also linked to over-linkage of isolated episodes in a large single-source board dumping of faeces by oyster harvestersoutbreak, and potentially the tracing back of con- occurred in Florida in January 1995 (Mcdonnell ettaminated product to the responsible harvesting area. al., 1997). Illness was reported in 38 gatheringsA good example of the application of these new state-wide and was reported to be the largest oyster-molecular epidemiological tools occurred in 1993 borne outbreak yet identified in Florida. A cohortwhen a large multistate outbreak of gastroenteritis epidemiological study among 223 people exposedoccurred in the USA in Louisiana (Kohn et al., confirmed oysters as the vehicle and showed an1995b), Maryland, North Carolina, Mississippi, attack rate of 58% among oyster eaters. TheTexas and Pennsylvania (Dowell et al., 1995). The aetiological agent was confirmed as NLV by electronoutbreak was linked to oysters harvested in a small microscopy and by seroconversion of patients toarea in Louisiana over a 4 day period in November Norwalk virus. However during this outbreak PCR1993. Twenty five separate clusters of illness were was not found useful possibly because the primersidentified with some 200 cases identified probably used were not appropriate for the outbreak strain.representing only a small proportion of the total The majority of oysters were restaurant or homenumber of persons effected. With over 4 million cooked prior to consumption however this appearedoysters harvested and an attack rate among oyster to offer little protection. During the investigation iteaters of 62% the authors calculate that as many as became apparent that an outbreak of diarrhoea illness186 000 people may have become ill (Dowell et al., had occurred in two bayside communities just prior1995). NLV was confirmed as the causative agent by to oyster harvesting for the New Year celebrations.both electron microscopy and PCR on stool samples This resulted in an increase in the reportedly com-


mon practice of overboard dumping of faeces by all countries, is the compliance of both harvestingharvesters with the consequential contamination of areas, treatment processes, and products sold to thethe beds in December 1996 through to January 1997 consumer, with all regulatory requirements. It seemsyet further outbreaks of viral gastroenteritis were therefore to be increasing widely recognised thatlinked to oysters harvested in Louisiana (Farley et current regulations are apparently unable to guaran-al., 1998). Sixty clusters of cases comprising 493 tee protection of the oyster consumer or to preventpersons were reported from 5 USA States. NLV was further outbreaks. In the UK, Australia, and somefound in stools by electron microscopy and by PCR. other countries, purification (depuration) is widelyHowever unlike previous outbreaks sequencing iden- practised for oysters and other shellfish consumedtified 3 different NLV strains occurring in the clusters raw. However a number of reports show that gas-examined which implicated 3 separate harvesting troenteritis outbreaks can occur following consump-areas in Louisiana. Again overboard dumping of tion of oysters purified according to all legal require-sewage by oyster boats was implicated as the ments (Grohmann et al., 1981; Gill et al., 1983;probable cause of oyster contamination. Molecular Chalmers and Mcmillan, 1995; Perrett and Kudesia,techniques have also been applied to investigation of 1995; Ang, 1998) and therefore that this process as4 outbreaks of shellfish-associated gastroenteritis currently practised also cannot guarantee consumeroccurring in Japan between 1987 and 1994. Both protection. The need for measures to more adequate-genogroup I and genogroup II NLV strains were ly protect the consumer against viral infection aredemonstrated in both stool and shellfish samples by widely noted in outbreak reports.PCR and sequencing confirming NLV as the causa- Recent molecular data has shown that patientstive agent in these outbreaks (Sugieda et al., 1996). may suffer from a mixed infection of both genogroupRecently PCR has been used to document NLV I and genogroup II NLV strains following shellfishcontamination of oysters associated with viral gas- consumption (Ando et al., 1995a; Sugieda et al.,troenteritis outbreaks in the UK (Lees et al., 1995b; 1996). Recent studies in this laboratory concur withGreen et al., 1998) and the USA (Le Guyader et al., these findings and suggest that mixtures of NLV1996b) illustrating the increasing potential of this strains appear fairly common in shellfish harvestedtechnique for determination of virus contamination in from contaminated sites. Likewise it is interesting toshellfish as well as stool samples (see 6.2). In Europe note that in some outbreaks patients may experiencea major outbreak, involving some 350 identified gastroenteritis followed by hepatitis (Richards, 1985;cases, occurred in Denmark and neighbouring Scan- Halliday et al., 1991) suggesting a mixed NLV/dinavian countries in January 1997 (Christensen et hepatitis A virus contamination in shellfish causingal., 1998). The outbreak was caused by imported such outbreaks. In a similar fashion in an oysteroysters. NLV was identified as the causative agent of associated outbreak involving navel officers twothe gastroenteritis by electron microscopy and PCR episodes of gastroenteritis occurred the first causedin stool samples and by PCR in shellfish. During this by NLV and the second seemingly by astrovirusoutbreak unusual secondary complications were seen (Caul, 1996b). It seems possible therefore thatwhich may have indicated infection or intoxication shellfish harvested from contaminated areas couldby an agent additional to NLV. Enterovirus was often contain a cocktail of viruses and that patientsdetected by PCR in both stools and shellfish and low, may consequently be infected simultaneously with asub-regulatory, levels of domoic acid (a naturally number of virus strains. This could contribute to theoccurring marine algal biotoxin) were also found. high attack rates, sometimes as high as 100%,

Suspect or confirmed viral gastroenteritis out- noticed in some shellfish mediated NLV outbreaks.breaks associated with oysters and other bivalve NLV immunity is generally regarded as short termshellfish continue to be reported in the UK and many and given the emerging data showing antigenicother countries up to the present date (epidemiologi- diversity between strains (Jiang et al., 1996a) maycal surveillance data is reviewed below). Incidents also prove to be fairly strain specific. In a mixedvary in size and significance however when appro- infection even where short term immunity is presentpriate laboratory tests are conducted NLV is almost it is unlikely to protect against a flux of contaminat-invariably identified as the responsible pathogen. An ing strains. These findings may also emphasis theincreasingly frequent aspect highlighted in reports, in potential significance of shellfish as a vehicle for


spreading NLV strains particularly given the increas- techniques such as PCR (see 5.1 and 5.2). Indeed foringly global trade in food commodities such as many years methods for detection of ‘viruses’ inshellfish. shellfish targeted the culturable enteroviruses. En-

Retrospective seroepidemiological studies can help teroviruses detected include poliovirus, originatingestablish high risk factors, and hence the importance from the live oral vaccine, coxsackievirus, echovirusof particular vehicles of transmission, and have been and other strains. It is now clear that enterovirusesused to investigate the significance of food-borne are not generally responsible for clinically apparenthepatitis A virus. As immunological reagents begastroenteritis following shellfish consumption. En-come available for the NLVs it will be interesting to teroviruses transmitted by the faecal–oral route arecarry out similar studies to help establish the contri- however clinically associated with a wide variety ofbution of shellfish as a food vector to the community other illness ranging from aseptic meningitis todisease burden with NLVs. Such data is an important chronic diseases such as myocarditis. Such systemicaspect of modern risk assessment techniques. Cur- illness often has a protracted incubation periodrently however NLV proteins expressed by molecular following initial faecal–oral transmission and maytechniques appear too narrow in their serological only occur in a percentage of those infected. En-reactivity to be particularly useful reagents for such teroviruses have only rarely been associated withseroepidemiological studies (Noel et al., 1997). food-borne infections (Cliver, 1994b) and have not

Although NLV is clearly the predominate cause of generally been associated with shellfish (or seafood)gastroenteritis linked to shellfish consumption other vectored infection. This is intriguing given the readyviruses have also been identified in stool specimens. demonstration of enterovirus contamination in shell-Virus particles morphologically similar to parvovir- fish harvested from polluted, and even relativelyuses (small round and featureless) have been reported clean, sites (reviewed by Jaykus et al., 1994) andin stool samples from shellfish and other outbreaks hence the probable exposure of shellfish consumers.(Appleton, 1994). However establishing their clinical Possibly the epidemiological presentation of manysignificance is complicated by their observation also enterovirus infections (i.e. their long incubationin patients with no gastroenteric symptoms and by period and unpredictable appearance of clinicaltheir occurrence in stool specimens alongside more symptoms) does not favour their identification andestablished gastroenteritis agents such as NLV and linkage to a particular food vehicle. Other authorsrotavirus. As EM is currently the only detection have commented that the public health impact ofmethod for these agents progress has been slow and enteroviruses transmitted by shellfish is probably nottheir significance as agents of foodborne disease fully appreciated (Lipp and Rose, 1997). In a recentremains to be established. A few reports link as- oyster associated outbreak of NLV gastroenteritis introvirus to shellfish consumption (Yamashita et al., Scandinavia unusual secondary symptoms, possibly1991; Caul, 1996b; Anon, 1998) however this seems reminiscent of enterovirus infection, occurred incomparatively rare. Such incidents may be associated some infected consumers (Christensen et al., 1998).with less common serotypes as astroviruses are now Enterovirus was detected by PCR in both shellfishgenerally regarded as ubiquitous infections of the and patients stool with isolations from stool showingyoung with consequently high levels of adult im- a common sequence identity suggestive of a commonmunity (see 3.2). Rotaviruses are a similarly com- source infection. It is not clear whether an en-mon infection of the young and although they have terovirus infection caused the secondary symptomsbeen demonstrated as contaminants of polluted shell- however further consideration of the possible role offish they have not been reported as a cause of enteroviruses in shellfish transmitted infection isshellfish-vectored disease outbreaks (see 3.3). probably warranted.

Two epidemiological case control studies have4.3. Other viral diseases reported, amongst other significant factors, statisti-

cally significant or nearly statistically significant,Human enteroviruses are commonly isolated from relationships between consumption of raw bivalve

sewage and sewage polluted waters and can readily shellfish and Creutzfeldt-Jakob disease (Bobowick etbe demonstrated as contaminants of bivalve shellfish al., 1973; Davanipour et al., 1985). This is aby conventional virus isolation and by molecular surprising linkage given that a route of transmission


is not obvious. However these studies were fairly occasionally in England and Wales however over thesmall with evaluation of multiple risk factors there- last decade such outbreaks have become less fre-fore this association remains speculative at present. quent now constituting , 5% of reported outbreaks.

It is interesting to note that several large gastroen-4.4. Disease surveillance data teritis and hepatitis outbreaks reported in the 1970s

and 1980s were associated with commerciallyInvestigations of specific disease outbreaks linked cooked cockles (Appleton and Pereira, 1977; Sockett

to shellfish consumption have been reported in the et al., 1985). Following review and strengthening ofscientific literature in many countries however few UK controls on commercial cooking procedures (seecountries systematically collate and report such data 6.2) cockle associated outbreaks markedly decreased.through a disease surveillance system. In the UK Outbreaks associated with authorised commercialepidemiological data is collected for England and processors of cooked cockles or mussels have notWales by the Public Health Laboratory Service been reported for a number of years suggesting theseCommunicable Disease Surveillance Centre and improvements were effective. Bacterial causes ofpublished periodically (Sockett et al., 1985, 1993; infection associated with bivalve shellfish have onlyAnon, 1992, 1993c, 1998; PHLS Viral Gastroen- occasionally been reported during the last decadeteritis Sub-Committee, 1993). Between the years (Anon, 1998) probably reflecting the success of1941 to 1970 more than 80% of reported shellfish- depuration processes (see 6.3) at removal of theseassociated food poisoning outbreaks were of un- contaminants. Of the bivalve species oysters nowknown cause. However since the late 1970s with the predominate as the shellfish vehicle of most concernrecognition of viruses as significant agents of inin the UK. Non-filter feeding shellfish have not beenfection increasing numbers of outbreaks were recog- reported as outbreak vectors for viral gastroenteritisnised as viral in origin. Between the years 1965 and in the UK. Incidents of food poisoning associated1983, of 60 reported outbreaks associated with with these species are less common than thosemolluscan shellfish, 22 (37%) were caused by ‘Small associated with filter-feeders and often of a bacterialRound Viruses’, 10 (17%) by hepatitis A virus and aetiology suggesting post-processing contamination26 (43%) were of unknown cause (Sockett et al., rather than contamination at source (Sockett et al.,1985). No outbreaks were identified as bacterial in 1985).origin. During the decade 1981–90 a further 99 Data on infectious diseases associated with mol-outbreaks of gastroenteritis related to molluscan luscan shellfish consumption has also been collated,shellfish were reported (PHLS Viral Gastroenteritis from a variety of sources, for the USA (Richards,Sub-Committee, 1993) and this trend has continued 1985; Rippey, 1994). Until the 1980s up to about 50with the number of reported outbreaks in subsequent disease outbreaks per decade were documented in theyears ranging between 5 and 15 a year. The latest USA. This increased dramatically to 217 outbreaksavailable data shows that 6 shellfish-associated gas- in the decade 1980 to 1990 with nearly 7000troenteritis outbreaks were reported in 1996 and 12 associated cases (Rippey, 1994). Oysters and clamsin 1997 (Anon, 1998). Of these incidents 13 were were responsible for most of these reported casesdue to oysters with NLV identified as the causative with soft clams, mussels and scallops only in-agent in 5 outbreaks. The number of individual cases frequently implicated. Most illness outbreaks wereassociated with each outbreak varies considerably associated with organisms derived from faecal pollu-from , 10 to hundreds depending on the particular tion of shellfish harvesting areas. Over the periodcircumstances. Gastroenteritis accounts for the large 1905 to 1990 some 80% of reports were ascribed tomajority of clinical symptoms reported with NLV the gastroenteritis with no causative agent isolated ormost frequently identified cause. In the UK during identified. Excluding typhoid, which last occurred inrecent years NLV has accounted for up to 40% of 1954, of outbreaks with an identified cause 17%reported incidents. Other viral causes of gastroen- were of viral origin with only 4% bacterial in nature.teritis, such as astroviruses, also feature occasionally Among identified viral causes hepatitis A virusin disease reports. Shellfish associated outbreaks predominated being responsible for 42 reportedcaused by hepatitis A virus continue to be reported outbreaks, with some 1800 cases, between the turn of


the century and 1990. However, shellfish associated Gastroenteritis Sub-Committee, 1993), the USAhepatitis A virus outbreaks were not recognised in (Richards, 1985; Rippey, 1994) and elsewhere. Inthe USA until the early 1960s and over 1000 of these many outbreaks the failure to identify a causecases occurred during several major outbreaks from probably relates mainly to the absence of appropriate1961 to 1964 (Richards, 1985). A further major virological investigation. Statutory hygiene controlsoutbreak occurred in 1973 and shellfish vectored or shellfish are based on bacterial indicators of faecalhepatitis A virus cases continue to be reported in the pollution, such as E. coli or faecal coliforms. TheUSA although at reduced levels (Rippey, 1994). relative paucity of outbreaks confirmed as bacterialUntil recently NLV was positively identified in only a in origin is generally attributed to the success ofrelatively small percentage of outbreaks in the USA. these controls for containing contamination of shell-However, the symptoms of disease reported in most fish by bacterial pathogens. By contrast many reportsoutbreaks of unknown cause were similar and were have shown that viral contamination of shellfish isdescribed as ‘mild’ gastroenteritis with clinical fea- not readily identified or controlled using bacterialtures consistent with NLV gastroenteritis as described indicators of faecal pollution.below. During the last decade with increased aware- It is interesting to note that UK epidemiologicalness and improved diagnostic techniques many more data shows a pronounced seasonal effect for illnessoutbreaks in the USA have been ascribed to NLV associated with shellfish consumption with outbreaks(Dowell et al., 1995; Mcdonnell et al., 1997; Farley occurring predominantly in the winter monthset al., 1998). (PHLS Viral Gastroenteritis Sub-Committee, 1993).

Outbreaks caused by NLV are characterised by Data for the USA also shows seasonal variation butmean incubation periods of 24–48 h, mean illness here disease peaks occur in both late spring and latedurations of 12–60 h, and a high percentage of fall (Rippey, 1994). In the UK NLV circulation in thepatients with diarrhoea, nausea, abdominal cramps community has historically been associated withand vomiting and a lower percentage with fever winter months being known in early literature as(Kaplan et al., 1982a). This combination of clinical ‘winter vomiting disease’. However circulation in thesymptoms is highly characteristic for viral gastroen- community is not always consistent with peaks ofteritis and unlike those reported for enteric infections shellfish associated gastroenteritis. As an example inwith bacteria or parasites or for chemical intoxica- the UK during 1995 greatest numbers of NLVtion. In particular outbreaks of Salmonella are char- infection were reported during spring and summeracterised by longer duration of illness and more fever months with the fourth quarter being unusually lowthan vomiting (Hedberg and Osterholm, 1993). (Anon, 1996). In the UK the large majority ofThese characteristics, along with stool cultures nega- oysters are purified (depurated) prior to sale to thetive for bacterial pathogens, have been proposed as consumer. Recent data from this laboratory suggestsepidemiological criteria for considering an outbreak that the seasonal pattern of shellfish associatedto be viral in origin (Kaplan et al., 1982a). The illness in the UK may also be influenced by failureavailable epidemiological data for both the UK and of oysters to clear viral contamination in purificationthe USA show that an aetiological agent is not systems during the winter months when shellfishidentified in the majority of molluscan shellfish metabolic activity is reduced (Dore et al., 1998).associated outbreaks. Of the causative agents iden- Disease surveillance authorities in the UK (PHLStified viruses predominate. However, unlike other Viral Gastroenteritis Sub-Committee, 1993) andforms of food poisoning, bacterial causes of gas- elsewhere (Rippey, 1994) recognise that officiallytroenteritis are rarely linked to molluscan shellfish reported cases considerably underestimate the actualconsumption. It is now widely accepted that the burden of disease from eating shellfish. There are aclinical symptoms reported in the large majority of number of reasons for this including the absence ofoutbreaks of unknown cause are consistent with the mandatory requirements for reporting gastroenteritisepidemiological criteria for viral gastroenteritis, such and the relatively mild nature of the illness which,as caused by NLV, and therefore that these agents although unpleasant, is of short duration and fol-probably constitute the bulk of the disease problem lowed by complete recovery with no longer termboth in the UK (Sockett et al., 1985; PHLS Viral sequel. Patients frequently do not present to their


doctor and even when they do the illness is not taminants of shellfish they are not responsible forfurther investigated or reported. Reported incidents disease symptoms, such as diarrhoea and vomiting,therefore relate principally to large functions where following consumption of contaminated shellfish.multiple illness is more apparent and where classical With the recognition of NLV as the main aetiologicalstatistical correlations between illness and a food agent responsible for shellfish associated viral gas-vehicle can be made. Individual cases are rarely if troenteritis more recent developments have focusedever investigated and do not feature in the disease on developing methods for these and other agentsstatistics of either the UK or the USA. It is likely causing gastrointestinal illness and for hepatitis Ahowever that the majority of oysters and other virus.shellfish are consumed in the individual or smallgroup setting. It is clear that public awareness 5.1. Detection of culturable enteric virusesthrough media reports (Davis et al., 1994) andaggressive investigation and case finding by au- Until recently methods for detecting human entericthorities (Rippey, 1994) can substantially effect both viruses from shellfish had as their goal the pro-the number of outbreaks identified and the volume of duction of extracts suitable for inoculation into cellillness reported. This highlights the probable major cultures. Crude shellfish extracts are highly cytotoxicunderreporting by passive surveillance systems. and therefore not suitable for direct inoculation ontoWhen immunological assays become available case cells. Simple dilution produces unfeasibly largecontrolled seroepidemiological studies may help volumes of extract therefore methodologies wereestablish more accurately the true incidence of required for the separation of virus from shellfishshellfish transmitted viral gastroenteritis. Such data is tissue and their subsequent reconcentration. Duringan important aspect of modern risk assessment the seventies and eighties numerous workers tackledtechniques and would help inform the debate over this difficult and complex task. Such structures haveexpenditure on remediation of sewage pollution in been comprehensively reviewed (Gerba and Goyal,shellfish harvesting areas. USA FDA risk assess- 1978; Williams and Fout, 1992; Jaykus et al., 1994)ments estimate cases of Norwalk viral gastroenteritis and a detailed treatise is outside the scope of thisrelated to seafood consumption at some 100 000 per review. Key objectives were to develop processingyear (Williams and Zorn, 1997). Considering the procedures that resulted in a low volume, non-cyto-numbers that are potentially at risk during major toxic extract with good virus recovery. The plethoraepisodes (Dowell et al., 1995) this estimate may be of reported methods, and the lack of consensus on anrealistic. Similar estimates have not been performed ideal candidate, testifies to the difficulty of the task.in the UK. Methods are generally separated into two alternative

approaches, the extraction-concentration methodsand the adsorption-elution-concentration methods

5. Detection of viruses in shellfish (Sobsey, 1987; Jaykus et al., 1994). However, withinthe literature there are many variations on detail

The development of methods for investigation of within these general schemes. Both extraction-con-seafood contamination with enteric viruses has concentration (Landry et al., 1980; Richards and Gol-centrated on the bivalve shellfish because of their dmintz, 1982; Lewis and Metcalf, 1988) and ad-proven association with disease outbreaks. Until this sorption-elution-concentration (Sobsey et al., 1975,decade such developments focused almost entirely 1978; Johnson et al., 1981; Sullivan et al., 1984;on detection of the enteroviruses. This reflected the Jaykus et al., 1994) methods have been reported foravailability of conventional culture techniques for a variety of bivalve species including oysters, clams,these viruses and their recognition as contaminants mussels and cockles. Following extraction and re-of the marine environment (Sellwood and Dadswell, concentration final elluents produced by either ap-1991; Sellwood, 1992). Early literature is sometimes proach are normally treated with antibiotics (toconfused regarding the epidemiological significance reduce bacterial and fungal overgrowth) prior toof enterovirus detection in shellfish. It is now clear inoculation into cell culture. Cell lines commonlythat although enteroviruses are common viral con- used for culturable enteric viruses include buffalo


green monkey kidney (BGM), Vero, MA-104, HeLa, they are currently regarded as non-culturable. ForBSC-1, LLC-MK2 and others. No one cell line both of these viruses therefore methods for theirsupports the growth of all enteric viruses however detection in shellfish cannot be based on growth inBGM cells have proved useful for the growth of a cell culture. Animal tests are also not generallyvariety of enteroviruses from environmental samples, available for detection of these viruses, with thesuch as water and shellfish, and are used by a possible exception of hepatitis A virus. Following anumber of workers (Morris and Waite, 1980). The large epidemic in China hepatitis A was shown tolack of standardisation in method development and contaminate clams by experimental exposure ofreporting makes direct comparison of method effica- marmosets (Li et al., 1992). Although this provedcy difficult. Comparative evaluations of different useful in this major epidemic the use of primates formethods have been performed (Cole et al., 1986; diagnostic tests is not generally feasible or desirable.Riordan, 1991; Speirs et al., 1987; Millard et al., For many years electron microscopy was the only1987) however such studies have often reported technique available for study of the nonculturableconflicting results. It seems clear that no one pro- gastroenteritis viruses and this remains a key tech-cedure is clearly superior in all attributes and certain- nique for examination of clinical samples. Howeverly a universal consensus on methodology has not electron microscopy is relatively insensitive requir-been reached. Given these difficulties the extraction ing in the order of a million particles for virusand culture of enteric viruses from shellfish is likely visualisation (Hedberg and Osterholm, 1993). Whilstto remain a task performed only in specialised such numbers are obtainable in clinical samples,laboratories. such as stools, they are probably rarely present in

The extraction methods described above for de- contaminated foods. There are only a few reports oftection of the culturable enteroviruses have also been visualisation of enteric viruses in shellfish by elec-adapted for the detection of rotavirus (Speirs et al., tron microscopy (Murphy et al., 1979; Appleton et1987; Lewis and Metcalf, 1988; Boher et al., 1991) al., 1981) and certainly this method has not provedand hepatitis A virus (Millard et al., 1987; Lewis and practical for routine application.Metcalf, 1988; Alouini and Sobsey, 1995) in shell- During recent years non-culture detection meth-fish. These viruses however require more specialised ods, such as immunoassays and molecular basedmethods for virus growth and titration following techniques, have become available for a number ofextraction. To date most studies have been per- gastroenteritis viruses and for hepatitis A virus.formed on shellfish experimentally contaminated in Laboratory adapted hepatitis A strains grown inthe laboratory with virus strains adapted for growth culture have provided antigen preparations for thein culture. Astroviruses and enteric adenoviruses are development of diagnostic immunoassays (Coulepisalso difficult to culture although specialised or et al., 1985). Immunoassays are also available for theprimary cell culture systems are available. However, astroviruses (Moe et al., 1991; Herrmann et al.,being predominantly paediatric infections these vi- 1990), the enteric adenoviruses (Vizzi et al., 1996),ruses are rarely linked to seafood consumption. and for NLV (Herrmann et al., 1985, 1995; MonroeConsequently no work has been reported on the et al., 1993). The development of immunoassays forapplication of culture methods for their detection in NLV in particular has been greatly aided by theshellfish. advent of molecular cloning and in-vitro protein

expression (see below). Immunoassays have proved5.2. Detection of non-culturable enteric viruses valuable for clinical investigations, including food-

borne outbreaks (Fleissner et al., 1989), and areNLV and hepatitis A virus are the most significant helping researchers understand the epidemiological

viral pathogens associated with shellfish consump- significance of these viruses (Moe et al., 1991; Graytion. Although experimental work can be performed et al., 1993; Smit et al., 1997). However althoughon laboratory adapted strains of hepatitis A virus, attempts have been made to apply immunoassayswild type strains can only be cultured with great (such as enzyme linked immunosorbant assay) todifficulty. NLV is even more problematical. Culture detection of viruses in water (Grabow et al., 1993),methods for this virus have never been reported and food (Hernandez et al., 1997) and shellfish (Giachino


et al., 1985), reports are very limited and not always been utilised for clinical investigations and, par-successful (Pietri et al., 1988). The limited success ticularly for NLV, have provided data of publicof this approach is probably because immunoassay, health significance that could not have been providedlike electron microscopy, is relatively insensitive by classical epidemiological methods alone (Ando etrequiring a thousand or more virus particles for a al., 1995a; Green et al., 1995a,b; Beller et al., 1997).positive reaction (Kogawa et al., 1996). Whilst such However, such studies have also demonstrated thetitres are common in clinical samples, contaminated diversity of NLV strains (Moe et al., 1994; Norcott etfood will contain much lower numbers since con- al., 1994; Jiang et al., 1995) which has madetamination is passive with no virus multiplication. construction of universal PCR primer sets difficult.Viral infections generally have a low infectious dose Recent studies have divided strains into two broadwith NLV thought to be as low as 10–100 virions genogroups (Ando et al., 1994; Wang et al., 1994)(Caul, 1996b). Clearly therefore low virus titres in and proposed regions with sufficient commonality tofood may constitute an infectious hazard and de- allow broadly reactive primer design (Green et al.,tection methods need to be correspondingly sensi- 1995a,b; Ando et al., 1995b; Le Guyader et al.,tive. 1996a). The two genogroups reflect antigenic clusters

The cloning and sequencing of hepatitis A (Cohen of strains with related, but distinguishable, propertieset al., 1987) and E (Bradley, 1995), Norwalk (Jiang (Noel et al., 1997). Currently combinations of PCRet al., 1993) and related NLVs (Lambden et al., 1993; primers seem able to detect most NLV strains inLew et al., 1994; Dingle et al., 1995), and other clinical samples but expressed antigen based im-gastroenteritis viruses, revolutionised possibilities for munoassays lag behind (Noel et al., 1997). Bythe development of sensitive virus detection meth- contrast hepatitis A virus has a relatively conservedods. Initial studies utilising molecular methods em- genome with a single serotype. PCR primer designployed gene probing and hybridisation techniques has therefore been more straight forward and aand were applied to detection of hepatitis A, number of sets have been proposed (Birkenmeyerrotavirus and enterovirus in shellfish (Zhou et al., and Mushahwar, 1994).1991; Le Guyader et al., 1993; Van Cuyck-Gandre et Progress with clinical PCR assays for NLV andal., 1994). However probe hybridisation suffered, other enteric viruses prompted the exploration of thislike previous methods, from sensitivity problems technology to detection of viruses in food. Severalrequiring in the region often thousand or more virus initial studies utilised culturable viruses, such asparticles for a positive reaction (Zhou et al., 1991; poliovirus, simian rotavirus or laboratory adaptedKogawa et al., 1996). Like many other areas of strains of hepatitis A virus, in laboratory seedingbiological science the major breakthrough came with studies to develop methods and assess the feasibilitythe development of the powerful and sensitive gene of PCR based assays for detection of enteric virus inamplification polymerase chain reaction (PCR) pro- shellfish (Atmar et al., 1993; Goswami et al., 1993;cedure. A further major benefit of PCR is the added Jaykus et al., 1993, 1995, 1996; Yang and Xu, 1993;potential for virus strain characterisation by PCR Le Guyader et al., 1994; Lees et al., 1994, 1995a;product probe hybridisation or sequencing. The Cromeans et al., 1997). These studies generallyavailability of genomic sequence for NLV quickly showed that the approach is feasible but also thatlead to the development of sensitive PCR based crude shellfish extracts are inhibitory to the PCR.assays for clinical samples (De Leon et al., 1992; Method development has concentrated on refiningJiang et al., 1992; Moe et al., 1994; Green et al., virus extraction and or nucleic acid extraction and1995a,b). purification techniques to overcome this inhibition

Similarly clinical diagnostic PCR assays have problem. A common approach has been to basebeen reported for hepatitis A (Jansen et al., 1990; methods for virus extraction on those previouslyApairemarchais et al., 1995) and E viruses developed for detection of culturable enteroviruses in(McCaustland et al., 1991; Turkoglu et al., 1996), shellfish (Lees et al., 1994; Jaykus et al., 1995). Aastroviruses (Saito et al., 1995; Jonassen et al., later modification targeted extraction of the shellfish1995), rotavirus (Gouvea et al., 1990; Xu et al., digestive organ (Atmar et al., 1995), since these are1990) and the enteric adenoviruses (Tiemessen and known to harbour most of the contaminating virusNel, 1996). These powerful new techniques have (Romalde et al., 1994), thus minimising contamina-


tion by other shellfish tissues. Following virus ex- not completely remove viruses it does seem totraction a variety of subsequent nucleic acid ex- reduce virus levels (see 6.3). Residual low virustraction and purification protocols have been em- levels whilst still a health hazard may be difficult toployed. Since PCR reaction volumes are small detect by conventional single round PCR (Lees et al.,(typically , 100 ul) these protocols generally need 1995b). We and co-workers have recently described ato incorporate concentration steps. Commonly used nested PCR for NLV that helps overcome theseapproaches include the purification of nucleic acid by sensitivity problems and has allowed more reliablevirus lysis with guanidine and nucleic acid recovery detection of NLV in shellfish associated with out-with a silica matrix (Lees et al., 1994; Hafliger et al., breaks and sold for consumption (Green et al., 1998;1997), and purification with organic solvents fol- Dore et al., 1998, 2000) and has allowed reliablelowed by selective precipitation of nucleic acid using monitoring of NLV contamination in polluted har-the cationic detergent cetyltrimethyl ammonium bro- vesting areas (Henshilwood et al., 1998). Thismide (Atmar et al., 1993; Jaykus et al., 1995). With development has also facilitated post PCR characteri-the advent of useable PCR primers for NLV similar sation by sequencing which provides major benefit,shellfish seeding studies were also performed for for epidemiological investigation (see 4.2). Otherthese viruses using stool preparations as source of workers have reported failure of PCR to detect NLVvirus (Gouvea et al., 1994; Atmar et al., 1995; Lees in oysters associated with outbreaks of NLV gas-et al., 1995b; Jaykus et al., 1996; Hafliger et al., troenteritis in the USA (Kohn et al., 1995; Mcdon-1997). These studies showed that virus extraction nell et al., 1997). However, further work did detectand nucleic acid purification methods developed for NLV in oysters using PCR primers specific to theother enteric viruses were generally equally applic- virus strain associated with one of these outbreaksable to NLV. A multi-centre collaborative study also (Le Guyader et al., 1996b). This report confirmeddemonstrated that NLV detection techniques could be NLV presence by nucleotide sequencing of clonedreliably applied in a number of laboratories (Atmar PCR amplicons. However the sequence was notet al., 1996). found to be identical to that obtained from the stools

Seeding studies have demonstrated the successful of patients from the outbreak. A recent report fromapplication of PCR to detection of virus in shellfish Japan has documented the coexistence of two NLVin the laboratory. However, fewer studies have genogroup strains in both faecal samples and shell-reported the application of these methods to de- fish associated with an outbreak (Sugieda et al.,tection of virus in naturally contaminated samples. 1996). Recent experience in our laboratory hasContamination of shellfish obtained from polluted confirmed this finding in outbreaks in Europe andharvesting areas with hepatitis A, enterovirus and indeed suggests that mixed cocktails of NLV strainsrotavirus has been documented (Desenclos et al., are common in shellfish from polluted harvesting1991; Yang and Xu, 1993; Le Guyader et al., 1994; areas (Henshilwood et al., 1998). These findingsLees et al., 1994, 1995a; Chung et al., 1996) and in suggest that definitive linking of shellfish and clinicalone small survey rotavirus was detected in three samples through sequence comparison may proveoyster samples from a market (Hafliger et al., 1997). problematical in some cases. Following similar de-Detection of NLV in naturally contaminated shellfish velopments PCR based methods have also beenfrom polluted harvesting areas, or in shellfish associ- described for the detection of astroviruses and adeno-ated with disease outbreaks, has currently been viruses in environmental samples such as polluteddocumented in only a few reports (Lees et al., water and shellfish (Marx et al., 1997; Girones et al.,1995b; Le Guyader et al., 1996b; Sugieda et al., 1995; Pina et al., 1998b). A complicating factor for1996; Green et al., 1998; Henshilwood et al., 1998). NLVs which has emerged recently is the existence ofExperience in our laboratory has been that although closely related bovine and porcine strains cross-presence of NLV can be fairly readily demonstrated reactive with PCR primers used for human NLVin polluted harvesting areas reliable detection in detection (Sugieda et al., 1998; Dastjerdi et al., 1999;processed products sold to the consumer, or associ- Liu et al., 1999). If these veterinary pathogens areated with outbreaks, is more difficult. In the UK wide spread it could complicate the interpretation ofdepuration is routinely performed on shellfish, such NLV positive results obtained from environmentalas oysters, sold live. Although this processing may samples subject to agricultural run-off. It remains to


be seen whether this is a significant problem and if cate the presence of viruses capable of causingso whether primer design can be refined accordingly. illness.

An uncertainty with the use of PCR is whether test Since removal of PCR amplification inhibitors is aresults necessarily indicate the presence of viable major factor effecting successful application of PCRinfectious virus. PCR amplifies nucleic acid which to shellfish most authors agree the need for goodcould originate from either viable virus or damaged documentation of method capability in this respect.non-infectious virus. It is not clear whether this is Inadequate removal of inhibitors can lead to falselikely to be a significant problem effecting detection negative reactions. The use of internal virus RNAof viruses in shellfish. The enteric viruses of concern standards has been proposed (Atmar et al., 1995;all have RNA genomes. Although DNA is remark- Nairn et al., 1995) and these should assist resultably stable RNA is not, being inherently susceptible interpretation. Internal standards may also proveto digestion by widely prevalent cellular enzymes useful for tackling another problematical aspect of(RNAses). It is debatable whether free virus genome, PCR, the lack of quantitation, through the develop-or virus genome unprotected by a complete protein ment of a competitive PCR approach. However suchcoat, would remain intact for long in the RNAse rich standards are not yet readily available for all entericenvironment of sewage or in the hostile environment viruses, and strains, of interest. Good quality controlof the shellfish digestive tract. Very little data is procedures for amplification inhibition, and otheravailable however a laboratory seeding study on aspects, remain an important area to be addressed.cooked shellfish showed that in some samples feline Despite these uncertainties PCR has proved to be acalicivirus, a possible model for NLV, was detectable major step forward in the development of methodsby PCR but not by culture (Slomka and Appleton, for detection of enteric viruses in shellfish. However,1998). Clearly therefore this may be a significant most data has currently been generated from labora-problem in applying PCR to the quality assessment tory experiments with only a few applied studies yetof cooked shellfish. It is not clear whether similar reported. Further developments can be anticipated inproblems would arise when testing live shellfish. To this area over the next few years. PCR methods areaddress these possible concerns some workers have clearly beneficial for the epidemiological investiga-proposed an initial round of culture in cells followed tion of outbreaks, for investigation of commercialby detection of propagated virus with molecular shellfish processing procedures (such as depuration)methods. This approach has been reported for de- for virus removal (see 6.3), and for investigation oftection of viable astrovirus (Abad et al., 1997b) and the mechanisms of virus uptake and elimination inenterovirus (Murrin and Slade, 1997) in water shellfish (Romalde et al., 1994; Schwab et al., 1998).however it is not applicable to the unculturable NLV. They may also prove useful for investigation andAn alternative approach has combined the selective surveillance of virus contamination of shellfish har-properties of antibodies with the sensitivity of PCR vesting areas and for surveys of virus contaminationin an antibody capture PCR. This has been applied to in shellfish. Such applications are beginning to bedetection of hepatitis A virus in seeded shellfish reported (Dore et al., 1998; Henshilwood et al.,samples and shown to be both sensitive and to 1998). However, further work and improvement isremove PCR inhibitors (Deng et al., 1994; Lopez- required on method simplification and standardisa-sabater et al., 1997). An advantage of this approach tion, internal and quality controls, cost, quantitationis that whole virions are recovered which may help and method availability before PCR can be consid-ensure that the PCR is detecting viable virus. This ered for more routine applications.approach may prove useful for other enteric viruseshowever immunological reagents for the most im-portant group, the NLVs, are not yet sufficiently 6. Controls on the production and processing ofadvanced or available for general use. In the absence bivalve shellfishof definitive human volunteer studies more extensivecorrelation of NLV detection in shellfish sold for The infectious disease hazards associated withconsumption with illness in consumers should help consumption of bivalve shellfish have been recog-establish whether PCR positive results always indi- nised for many years. Consequentially most coun-


tries have enacted sanitary controls on the production It is generally accepted that the most effective andof bivalve shellfish. In the European Union these reliable approach to controlling the contamination ofwere drawn together into a European Directive 91/ shellfish is to harvest from areas with good water492/EEC (Anon, 1991a) to enable operation of the quality. Control of contamination through molluscsingle European market in 1993. In the United States processing procedures tends to be less effectiveby interstate trading agreements set out in the FDA although providing a practical option for manyNational Shellfish Sanitation Program Manual of countries where waters may be subject to sewageOperations (Anon, 1993b). These regulations cover contamination. There are, conventionally, two differ-similar ground on the requirements for clean growing ent forms of commercial processing available forareas, the controls and processing requirements for reducing microbial contamination of shellfish. Con-more contaminated areas, the hygiene conditions for tamination may be reduced by heat treatment (cook-processing and dispatch establishments, requirements ing) or by extending the natural filter-feeding pro-for marketing documentation etc. Third country cesses in clean seawater to purge out microbialimports into the EU and USA have to be produced to contaminants. This can be performed in tanks (depu-the same standard as domestic products. Major ration) or in the natural environment (relaying).exporting nations have therefore developed programs Additionally, shellfish harvesting may be prohibitedfor compliance with the regulations of their target in areas with unacceptable levels of pollution. Theexport markets A major feature of these controls is EU and USA sanitary legislation controls thesethe use of traditional bacterial indicators of faecal various options primarily through the use of faecalcontamination, such as the faecal coliforms or E. indicator monitoring which determines the appro-coli, to assess contamination and hence implement priate treatment in accordance with the level ofthe appropriate control measures. Faecal indicators contamination.are either measured in the shellfish themselves (EUapproach) or in the shellfish growing waters (US 6.1. Classification of shellfish harvesting areasFDA approach). Historically it has been widelyaccepted internationally that harvested shellfish Both EU and USA legislation employ a grading orwhich meet a microbiological standard of less than classification approach with standards set for230 E. coli, or 300 faecal coliforms, in 100 g of categories varying from waters with very low con-shellfish flesh can be placed on the market for human tamination levels to those where harvesting is prohi-consumption. This standard, together with standards bited. A synopsis of the EU and USA legislativefor specific pathogens (such as salmonella), chemi- standards for bivalve shellfish is set out in Table 1.cal; and algal biotoxins, has been adopted as an In the EU faecal indicators are measured in shellfish‘end-product’ standard in EU Directive 91/492. flesh whereas in the USA indicators are measured in

Table 1aSynopsis of EU and USA legislative standards for live shellfish

Shellfish treatment US FDA Microbiological EU Microbiologicalrequired classification standard per classification standard per 100 g

100 ml water shellfish

Non required Approved GM , 14 FCs Category A All samples , 230 E. coliand or90% , 43 FCs All samples , 300 FCs

Depuration or Restricted GM , 88 FCs Category B 90% , 4600 E. colirelaying and or

90% , 260 FCs 90% , 6000 FCsProtected relaying Category C All samples , 60 000( . 2 months) FCsHarvesting – Above levels – Above levelsprohibited exceeded exceeded

a FCs 5 faecal coliforms, GM 5 geometric mean, 90% 5 90%-ile compliance.


the shellfish growing waters. Both the EU and the minimum period of 2 months is specified) or com-USA systems base standards on a 5-tube 3-dilution mercial heat treatment by an approved method.most probable number (MPN) test. Point source Shellfish from category C areas must contain lesspollution inputs are generally identified through a than 60 000 faecal coliforms per 100 g of shellfishshore line survey, or similar identification, and flesh (see Table 1). Relayed shellfish may, if neces-quality monitoring programs designed accordingly. sary, be depurated before being placed on theUSA FDA ‘approved’ and EU ‘category A’ stan- market. USA FDA controls do not incorporate andards describe the cleanest growing areas from equivalent to EU category C. Shellfish growing areaswhich shellfish can be taken for direct human exceeding these prescribed levels of contamination,consumption without further processing. All shellfish or areas for which harvesting area survey andfrom EU category A areas must contain less than 30 classification has not been conducted, cannot beE. coli, or 300 faecal coliforms in 100 g of shellfish harvested for human consumption in either USA orflesh. The USA FDA program gives a choice of EU legislation.using either total or faecal coliforms to establish a If areas meet the above standards only for certainclassification. It further expresses standards in two periods because of predictable pollution events au-components, a geometric mean (GM) count of thorities may classify them for a restricted period. Inresults and an upper standard which not more than the USA FDA program such areas are defined as10% of results can exceed. Approved areas must ‘conditionally approved’ or ‘conditionally restricted’comply with a total coliform GM of 70 per 100 ml and may be harvested during periods when they meetwater with not more than 10% of samples exceeding the standards subject to a management plan. In the230 per 100 ml. Alternatively they can comply with UK such areas are described as ‘seasonally classi-a faecal coliform GM of 14 per 100 ml water with fied’ with the period when the classification appliesnot more than 10% of samples exceeding 43 per 100 defined. In addition to criteria for harvesting areaml (see Table 1). classification both EU and USA legislation set out

In both EU and USA legislation shellfish cannot requirements for other aspects such as bivalve trans-be harvested for direct human consumption from port, wet storage, depuration, relaying, analyticalgrowing areas exceeding the above levels of con- methods, movement documentation and provisionstamination. They may however be placed on the for suspension of harvesting from classified areasmarket following commercial depuration, relaying or following a pollution or public health emergency.heat processing. However because these processesare known not to be completely effective an upper 6.2. Heat treatment (cooking)threshold is placed on the degree of contaminationbeyond which such procedures may not be used. For shellfish sold as a processed product commer-Shellfish from EU ‘category B’ and USA FDA cial heat treatment (cooking) is used to reduce levels‘restricted’ classifications may be placed in the of microbiological contamination. Various heat treat-market following depuration or relaying (see 6.3). ment processes have been described for shellfishShellfish from EU category B may also be heat varying from pasteurisation through to sterilisationtreated by an approved method (see 6.2). EU cate- by canning. However it is important to ensure thatgory B areas must contain less than 4600 E. coli, or cooking processes are properly regulated and con-6000 faecal coliforms per 100 g of shellfish flesh in trolled. In the UK during the 1970s and 1980s a90% of samples. USA FDA restricted areas must number of outbreaks of gastroenteritis and hepatitiscomply with a total coliform GM of 700 per 100 ml were linked to commercially cooked cockles (Ap-water with not more than 10% of samples exceeding pleton and Pereira, 1977; Sockett et al., 1985).2300 per 100 ml. Alternatively they can comply with Investigation suggested that the batch cooking pro-a faecal coliform GM of 88 per 100 ml water with cedures in use at the time were probably undercook-not more than 10% of samples exceeding 260 per ing shellfish particularly when environmental tem-100 ml (see Table 1). peratures were low and shellfish were insufficiently

In EU legislation shellfish from harvesting areas warmed prior to cooking. Research following theseexceeding the limits of category B may only be outbreaks showed that hepatitis A virus could beplaced on the market following protracted relaying (a inactivated by more than 4 log infectious units by10


raising the internal temperature of shellfish (cockle) uct with low consumer acceptance. In the UKmeats to 85–908C for 1 min (Millard et al., 1987). mussels are not eaten raw, but are frequently lightly

`Subsequent recommendation by the UK Ministry of cooked in dishes such as moules mariniere. MusselsAgriculture for commercial cooking operations was sold live have to conform to hygiene standards setthat internal shellfish meat temperatures should be for species, such as oysters, consumed raw. Despiteraised to 908C and held at that temperature for 1.5 this, home or restaurant cooked mussels continue tomin. However such heat cook parameters may be feature occasionally in the UK disease statistics as adifficult to reliably achieve for shellfish cooked in cause of outbreaks of gastroenteritis (Anon, 1993c;large batches without rendering some shellfish un- Anon, 1998). However, the low incidence of suchpalatable. Consequently continuous flow machinery reporting compared to species commonly eaten raw,was designed for high throughput operations capable such as oysters, suggests home or restaurant cookingof reliably delivering the above heat cook parameters may be having at least some protective effect. Theto all shellfish. Since this review and strengthening success of commercial heat processing for control-of UK controls on commercial cooking procedures, ling health risks in cockles and mussels shows thatcockle associated illness outbreaks markedly de- commercially achievable heat cook parameters willcreased. In the UK viral outbreaks associated with reliably inactivate enteric viruses. It seems likelyauthorised commercial processors of cooked cockles therefore that viral problems associated with homeor mussels have now not been reported for a number and restaurant cooked mussels are a consequence ofof years suggesting these improvements were effec- under or inconsistent cooking. Only a few scientifictive. EU Directive 91/493/EEC (Anon, 1991c) studies relevant to this are available. An epi-requires that commercial heat treatment methods demiological investigation into an epidemic of hepa-undertaken directly on EU category B or C shellfish titis A in China associated with clam consumptionmust be approved. Several commercial heat treat- showed that within the population investigated attackment processes used in the EU have been officially rates were highest in those who eat raw clamsapproved including the UK heat cook parameters of (18%), but also higher among those who ate cookedraising the internal temperature of shellfish meats to clams (7%) than among those who did not eat clams908C for 1.5 min (Anon, 1993a). These heat cooked (1%) (Wang et al., 1990). Similarly a recent study inparameters were based on data obtained for hepatitis the USA following a well researched multi-stateA virus in cockles. However, the most frequently outbreak of NLV gastroenteritis associated withreported illness associated with consumption of oysters showed that home or restaurant cookingbivalve molluscs is gastroenteritis caused by NLV. offered little or no protection. In this study theSince NLV cannot be cultivated heat inactivation data authors suggest that the degree of cooking requiredfor this virus is not available. However similar to reliably inactivate NLV would probably renderstudies carried out on feline calicivirus, a possible oysters unpalatable to consumers (Mcdonnell et al.,model for NLV, showed this virus to be more readily 1997). These epidemiological findings are supportedinactivated than hepatitis A virus (Slomka and by a recent laboratory study showing that rotavirusesAppleton, 1998). It is probable therefore that current- and hepatitis A virus could still be recovered inly approved commercial heat treatment processes steamed mussels 5 min after the opening of theused in the UK are effective for NLV. This view is mussel valves (Abad et al., 1997a). It seems likelycertainly supported by the available epidemiological therefore that home and restaurant cooking as cur-evidence. rently performed is, at best, only a partially effective

Although modern carefully regulated commercial control measure. Cooking may possibly offer moreheat treatment processes appear to have proved protection for smaller species, such as mussels, thaneffective in controlling enteric virus continuation in for larger species such as oysters and clams.shellfish sold processed, such as cockles, the samecannot be said for shellfish sold live and cooked in 6.3. Purification (depuration) and relayingthe home or restaurant. A major factor limiting theeffectiveness of such home or restaurant cooking Although commercial heat processing seems effec-appears to be the limited nature of the processing tive it is not however applicable for shellfish soldapplied. Overcooking results in an unpalatable prod- live which constitute the bulk of the infectious


disease hazard. Purification (depuration) procedures tralia during volunteer trials to assess the safety ofwere first developed in the UK during the 1920s depurated shellfish (Grohmann et al., 1981) and has(Dodgson, 1928) as a means of extending the natural also been documented in outbreaks in the UK (Gillbivalve filter-feeding processes in clean seawater to et al., 1983; Heller et al., 1986; Chalmers andpurge out microbial contaminants. Tank-based depu- Mcmillan, 1995; Perrett and Kudesia, 1995; Ang,ration is now widely practised in many countries 1998) and the USA (Richards, 1985). The epi-including Australia, the UK, France, Italy, Spain and demiological evidence therefore strongly suggest thatelsewhere, it is however less widely used in the USA depuration may fail to eliminate enteric viruses from(Otwell et al., 1991). Depuration periods may vary contaminated shellfish and further that compliancefrom 1 to 7 days, however around 2 days is probably with E. coli (or faecal coliform) end-product stan-the most widely used period. Depuration systems dards does not provide a guarantee of virus absence.also vary and include processes where water is static The success of depuration in removing bacterialor changed in batches through to systems where contaminants may account for the very low incidenceseawater is flushed through continually or recycled of such infections following shellfish consumptionthrough a steriliser. Water sterilisation processes (see 4.4) however it is clear that this process, asinclude ozone, chlorination, UV irradiation and currently undertaken, is considerably less effectiveiodophores (Otwell et al., 1991; Roderick and for virus elimination. This observation is supportedSchneider, 1994; Poggi and Le Gall, 1995). Depura- by numerous laboratory studies which have ex-tion has been applied to most bivalve shellfish amined elimination rates of human enteric virusesspecies sold live including oysters, clams, mussels (such as poliovirus), or possible models for theand cockles. behaviour of human enteric viruses (such as various

Commercial depuration, when used as a treatment bacteriophage species), from shellfish during theprocess to reduce microbial contaminants, is subject depuration process (Richards, 1988; Power andto legal control in both the EU, the USA and in other Collins, 1989, 1990a,b; Sobsey and Jaykus, 1991;countries. The National Shellfish Sanitation Program Jaykus et al., 1994; Dore and Lees, 1995). AlthoughManual of Operations (Anon, 1993b) sets out FDA elimination rates in individual studies vary signifi-requirements in the USA. In the EU Directive 91/ cantly the overwhelming finding from these studies492/EEC (Anon, 1991a) details requirements for is that viruses are eliminated from bivalve shellfish atapproval of shellfish purification centres covering a slower rate than faecal coliforms. These findingssuch aspects as tank construction and operation, are confirmed by recent laboratory seeding studieslaboratory testing, packaging, labelling and trans- using NLV detected by PCR. NLV was found toportation. In addition purified shellfish are required efficiently accumulate in shellfish (oysters and clams)to comply with the end-product standard for shellfish however it was only poorly removed by depurationsold live which includes the faecal coliform parame- compared to E. coli (Schwab et al., 1998). Thister of less than 230 E. coli, or 300 faecal coliforms, laboratory has recently evaluated virus elimination inin 100 g of shellfish flesh. The regulatory principles commercially depurated shellfish (oysters) as judgedrelating to purification plant construction and opera- by both NLV and male specific RNA (F 1 ) bac-tion set out in the EU Directive are implemented by teriophage (a potential viral indicator) content (Dorethe ‘competent authority’ in each member state. In et al., 1998, 2000). Processed shellfish were found topractice compliance with the end-product faecal be routinely compliant with faecal coliform end-coliform standard is frequently seen as evidence of product standards however significant numbers weresatisfactory design and operation of depuration contaminated with both NLV and F 1 bacteriophage.plants. However evidence from various sources Viral contamination was found to be highly corre-suggests that depuration plants functioning satisfac- lated with the degree of harvesting area pollution andtorily by faecal coliform criteria may still fail to fully to the known incidence of disease outbreaks linkedremove human enteric viruses. Perhaps most signifi- to the site. This data supports previous laboratorycant is the epidemiological evidence demonstrating findings and confirms that compliance with faecalthat infection can occur following consumption of coliform end-product standards does not provide adepurated shellfish. This was documented in Aus- guarantee of the absence of enteric viruses in depu-


rated shellfish. A further important finding was that ing. Combinations of these processes may also bevirus contamination, as judged by both NLV and F 1 used, for instance in France traditional practicesbacteriophage content, in commercially depurated include holding molluscs in ‘claires’ (man-madeoysters was much more prevalent during colder tidally submersible ponds) for several months andwinter months. The dramatic effect of season on then in ‘degorgeoirs’ (wet storage ponds) for twoviral, but not bacterial content suggests that physio- days. Relatively little information exists on thelogical requirements for elimination of viruses during removal of viruses during shellfish relaying althoughshellfish depuration may be significantly different to again factors such as seawater temperature and initialthose required for effective elimination of faecal contamination levels appear critical (Cook and Ellen-coliforms. Laboratory studies have suggested that der, 1986; Jaykus et al., 1994). Data from thisprocess temperature (Power and Collins, 1990a,b; laboratory on NLV and F 1 bacteriophage (Dore etJaykus et al., 1994; Dore et al., 1998) may be an al., 1998) and unpublished data suggests that remov-important factor in virus removal during depuration. al of viruses from heavily contaminated shellfish byThis is supported by these seasonality findings since a combination of relaying for 4–6 weeks followed byseawater used in commercial shellfish depuration depuration can be effective but again is criticallyplants is generally not heated except in extreme dependant on seawater temperature. In these studiesconditions. The effect of temperature on virus elimi- differences were also seen between virus clearance innation during depuration requires further study as do native (O. edulis) and cultured (C. gigas) oystersother physiologically important parameters such as suggesting that species specific factors should also befood availability, salinity, dissolved oxygen, and considered. Other workers have reported similar datashellfish condition. Recent studies in this laboratory using bacteriophage studies (Humphrey et al., 1995).suggest that a seawater temperature of 18–208C isoptimal for removal of F 1 bacteriophage from 6.4. The prospects for viral standardsoysters (C. gigas) but also that successful eliminationwithin a 2–3 day period is critically dependent on Viral disease outbreaks associated with oystersinitial contaminant on level (Dore et al., 1998). and other bivalve shellfish continue to be reported inHeavily contaminated shellfish failed to clear bac- many countries (see Section 4). An aspect high-teriophage within 7 days even at elevated tempera- lighted in many outbreak reports is the compliance oftures. It is probably the case that many other aspects both harvesting areas, treatment processes, an prod-of depuration plant design and operation, optimised ucts sold to the consumer, with the various regula-to ensure faecal coliform removal, require reevalua- tory requirements summarised above. It seems there-tion in the light of methods for detection of viruses. fore to be increasing recognised that current regula-

Relaying involves the transfer of harvested ani- tions are apparently unable to guarantee protection ofmals to cleaner estuaries or inlets for self-purification the shellfish consumer or to prevent further out-in the natural environment. This process can be used breaks. In many countries depuration is widelyas an alternative to depuration for lightly polluted practised for oysters and other shellfish consumedshellfish. Shellfish can only be held for relatively raw. However a number of reports have shown thatshort periods in depuration tanks but can obviously gastroenteritis outbreaks can occur following con-be maintained for much longer periods in the natural sumption of oysters purified according to all legalenvironment. This makes relaying also suitable for requirements (see 6.3) and therefore that this processtreating more heavily polluted shellfish where longer as currently practised also cannot guarantee con-periods (a minimum of two months is specified in sumer protection. The consequential requirement forEU Directive 9 1/492 for category C shellfish) are measures to better protect the consumer from virusrequired to remove heavy contaminant loads. The contamination are widely noted in outbreak reports.main disadvantages of relaying are the limited During a recent outbreak in Scandinavia oystersavailability of suitable unutilised clean coastal areas complying with the relevant European faecaland of obtaining ownership rights to those areas the coliform standards (Anon, 1991a) were shown todifficulty of controlling water quality and other water cause NLV gastroenteritis in at least 356 patientsparameters and the susceptibility of stock to poach- (Christensen et al., 1998). This and previous similar


outbreaks have prompted the formal establishment are more hardy than bacteria and therefore survivewithin the EU of a network of reference laboratories better in the marine environment (see 6.5) and aretasked with the introduction of better systems for the more resilient to inactivation or removal duringmonitoring and control of viral contamination of depuration or relaying (see 6.3). In this context it isbivalve molluscs (Anon, 1999a). This laboratory has likely that other possible bacterial indicators, such asbeen designated as the co-ordinating European Com- the faecal streptococci, would share similar disad-munity reference laboratory. Recent outbreaks in vantages to the faecal coliforms. For these reasons aother countries (Anon, 1999b) have also prompted number of workers have proposed alternative in-reevaluation of approaches for the control of viral dicators for better assessment of viral contaminationcontamination of shellfish. In the USA a proposed in the marine environment. Some of the mostextensive reevaluation of approaches to shellfish promising candidates are various species of bac-sanitation, the National Indicator Study, has pub- teriophage because of their physical and genomiclished a comprehensive literature review (Hackney similarity to human enteric viruses, their abundanceand Pierson, 1994) prior to further initiatives. USA in sewage effluents and their ease of assay (IAWPRCworkers have also proposed a novel quantitative risk Study group on health related water microbiology,assessment approach (Rose and Sobsey, 1993). IAWPRC, 1991). Male specific RNA (F 1 ) bac-

Clearly PCR based detection methods for viral teriophage have been proposed as candidate viralpathogens (see 5.2) could form an element of indicators for water pollution (Havelaar, 1987) andimproved regulatory controls for viral contamination specialised hosts have been developed for theirof shellfish. This laboratory has recently demon- specific assay (Debartolomeis and Cabelli, 1991;strated that NLV monitoring by RT-PCR in a pol- Havelaar et al., 1993). F 1 bacteriophage shareluted harvesting area is possible and that virus many physical and genomic properties with thecontamination can both be readily detected and was enteric viruses of concern however, like the faecalconsistent with subsequent outbreaks in products coliforms, their distribution is not restricted toharvested from the study area (Henshilwood et al., human effluents. To address this issue other workers1998). French workers have also recently applied have proposed the use of bacteriophage of theRT-PCR to detection of enteric viral (NLV, hepatitis obligate anaerobe Bacteroides fragilis as a potentialA, rotavirus and enterovirus) contamination in shell- indicator of human specific pollution (Tartera andfish harvesting areas and have similarly shown that Jofre, 1987). It may also be possible to speciate F 1

these techniques can readily detect virus contamina- bacteriophage in order to ascribe contamination totion (Le Guyader et al., 1998). Both studies have animal or human sources (Hsu et al., 1995; Beekwil-demonstrated that enteric virus contamination can be der et al., 1995). A number of studies have beendetected in areas judged suitable for commercialisa- performed using these bacteriophage and in generaltion by faecal oliform criteria. However currently they have been shown to hold promise as indicatorspublished methods are complex, not yet standardised, of enteric virus presence in sewage effluents andand not widely available. It is probable therefore that polluted receiving waters (Lucena et al., 1994; Lee etfurther work on these aspects is required before their al., 1997). General somatic bacteriophage ofintroduction on a widespread basis for regulatory coliforms have also been suggested (Humphrey etcontrol could be considered practical. These methods al., 1995), however since this constitutes a diversedo however have important current applications for grouping standardisation and reproducibility areresearch and for epidemiological investigations. problematical (Havelaar, 1987; Vaughn and Metcalf,

The current regulatory controls in most countries 1975). An alternative to bacteriophage may be therely heavily on the concept of faecal pollution detection of other human viruses which are eitherindicator organisms (as described above) to assess more widely prevalent in polluted waters, or easier tomicrobiological hazard. These methods are cheap, detect, than the viral pathogens of concern. Humanstandardised and widely available. However the enteroviruses have a long history in this context (seeorganisms currently used, the faecal coliforms, have 5.1) however extraction and culture of enterovirusbeen shown to inadequately reflect the presence of from shellfish is a specialised and expensive task andviral contaminants. This is probably because viruses most workers have now abandoned this approach.


Recently human adenoviruses, detected by a molecu- tion methods, for the further development of publiclar approach, were suggested as a possible index of health controls for bivalve shellfish. Given the farhuman specific pollution (Pina et al., 1998b). Adeno- reaching implications of adoption into legislation it isviruses are widely prevalent in sewage effluents and important that potential viral controls for shellfish areeasily detectable by PCR. This is an interesting well researched and shown to perform adequately inproposal however further evaluation is required both practice. Adoption of appropriate measures will alsoof adenovirus prevalence compared with the enteric be facilitated by a scientific consensus on the mostpathogens of concern and of the current practicality appropriate methods and approaches. Issues to beof molecular assays in a regulatory context. considered are the respective roles of alternative

The majority of studies on alternative indicators indicators versus pathogen detection, on the merits ofhave been performed in the context of water pollu- the various candidate indicator organisms, on thetion. However during recent years studies relating to selection of methodological procedures, and onbivalve shellfish have also been performed. This numerical standards that might be applied. Currentlylaboratory has investigated the potential of F 1 however little has been published on this topic. Thisbacteriophage in commercially depurated market laboratory has recently suggested a possible manage-ready oysters and demonstrated that F 1 bac- ment strategy for bringing the various elementsteriophage is a much better indicator of virus con- together (Dore et al., 1998). However much furthertamination (as judged by degree of harvesting area work remains to be done before the most appropriatepollution, association of the harvesting area with approach can be identified and the implications fullygastroenteritis outbreaks, and presence of NLVs in understood.marketed shellfish) than is E.coli (Dore et al., 1998,2000). F 1 bacteriophage may have particular ad- 6.5. Control of harvesting area pollutionvantages as a ‘viral’ indicator for depurated shellfishas several studies have shown its elimination kinetics It may seem obvious to say that the most effectiveduring this process appear to reflect those of enteric way to tackle shellfish vectored virus disease is toviruses (Power and Collins, 1989; Dore and Lees, prevent sewage pollution of shellfish harvesting1995). Recently other workers have evaluated the areas, prevention generally proving better than cure.performance of F 1 bacteriophage and phages of B. However growing coastal populations and the highfragilis as indicators of virus pollution in shellfish investment cost of sewage treatment processes haveharvesting areas and similarly concluded that F 1 proven difficult obstacles to overcome. In somebacteriophage is a promising indicator of human countries, such as in the UK, the shellfish industry isenteric virus pollution in oysters (Chung et al., widely dispersed whereas in others, such is The1998). By comparison conventional bacterial in- Netherlands, it is concentrated into a few geographi-dicators did not adequately reflect virus presence and cal areas. Clearly focusing on the difficult andphages of B. fragilis were present in insufficient expensive task of achieving and maintaining highnumbers to be useful. Other workers have suggested standards of water quality is easier where the shell-phages of B. fragilis may prove useful for indication fish industry is recognised as a major factor in theof remote pollution effecting shellfish (Lucena et al., local economy. In other situations expenditure on1994). Perhaps most persuasively, analysis in this adequate sewage treatment can seem disproportion-laboratory of shellfish associated with gastroenteritis ate to the value of the shellfish industry and this hashas shown that of 14 outbreak samples 10 were hampered investment. However it is not clear thatnegative for E. coli with only 2 above the required agencies have ever attempted to factor in the widerbacteriological standard ( , 230 E. coli per 100 g), public health and other costs of shellfish transmittedwhereas all samples were positive for F 1 bac- infections. Where assessments have been performedteriophage with average levels of 2500 pfu per 100 g they suggest that such costs may not be trivial(range 60–17 500) (unpublished data). (Dalton et al., 1996). However in recent years the

Given these various advances it is now becoming economic arguments have become balanced, in manypertinent to consider how best to utilise potential countries, by growing environmental concerns. In thealternative indicators, and advances in virus detec- EU environmental quality legislation has become a


major factor effecting expenditure on sewage infra- maximise public health gains from expenditure onstructure. Water quality standards for bathing beaches sewage infrastructure it is therefore wise for agencies(Anon, 1976) and for minimum levels of sewage responsible for water quality to adopt a holistictreatment prior to marine discharge (Anon, 1991b) approach considering, in addition to the details of thehave dictated high levels of expenditure in many sewage treatment scheme, the appropriate dischargeEuropean countries. Of direct relevance to location and the adequacy of arrangements for stormshellfisheries Directive 79/923/EEC (Anon, 1979) water storage and treatment. Over reliance on simpleon the quality required of shellfish waters stipulates a numerical compliance with bacterial water qualityguideline faecal coliform standard approximately standards at a set monitoring point is unlikely toequivalent to category A (quality suitable for direct yield the optimum protection from viral contamina-consumption) under the sanitary controls Directive tion for vulnerable bivalve shellfisheries. Clearly91/492. In many EU countries this has become an meaningful tests for virological quality of seawaterimportant driver for maintenance or improvement of would bring similar benefits in this area as it wouldwater quality in molluscan shellfisheries. Whilst to sanitary assessments of shellfish quality. A num-these various initiatives can bring demonstrable, and ber of workers are actively pursuing such goalswelcome, improvements it should be recognised that (Metcalf et al., 1995).they also rely heavily on conventional bacterialpollution indicators for measuring performanceagainst set water quality standards. Like sanitary 7. Summarycontrols for shellfish such water quality standardscannot necessarily be relied upon to deliver the The epidemiological data clearly demonstrates thatnecessary improvements to the virological quality of filter feeding bivalve shellfish can, and do, act aswater. Periodic monitoring programs provide better efficient vehicles for the transmission of entericprotection against continuous discharges than against viruses transmitted by the faecal–oral route. Thisintermittent spills associated with storm water dis- identified hazard has been documented as a cause forcharges in combined sewer and rainfall systems. concern by various international agencies and has aHowever the latter may be heavily contaminated long history. Disease outbreaks can occur on anwith untreated effluent, particularly during the ‘first epidemic scale as graphically illustrated by an out-flush’. Rainfall associated outbreaks of shellfish break of Hepatitis A in Shanghai, China in 1988vectored disease have indeed been reported on many involving about 300 000 cases. Improvement ofoccasions (see 4.2) demonstrating the importance of harvesting area water quality offers the most sustain-this factor. Indeed although the antiviral efficiency of able route to improvement in the virological qualitysewage treatment can be questioned it is interesting of bivalve shellfish sold live. However there isto note that fully treated effluents have not yet been growing awareness, and concern, that current regula-shown to have contributed to foodborne or water- tory standards based on faecal coliform monitoringborne viral disease (Cliver, 1994a). An additional do not fully protect the shellfish consumer from viralproblem is that virus can survive for weeks to infection. New viral test methods based on PCR, andmonths in the marine environment (Nasser, 1994; the development of alternative more reliable faecalCallahan et al., 1995; Gantzer et al., 1998) which is pollution indicators, offer new approaches for theappreciably longer than for bacterial indicators (Solic further development of public health controls. How-and Krstulovic, 1992). Sewage schemes designed for ever, further work is required to build a scientificcompliance with bacterial standards are generally consensus and to understand the implications of theirmodelled using standard bacterial T90 seawater introduction into legislation.decay rates. It remains to be seen whether suchdesigns will prove adequate for protection againstviral contamination, In a similar manner disinfection Referencesof sewage effluents designed for simple compliancewith bacterial standards may distort bacterial in- Abad, F.X., Pinto, R.M., Gajardo, R., Bosch, A., 1997a. Viruses indicator to virus ratios and achieve at best cosmetic mussels – public health implications and depuration. J. Foodimprovements to water quality (Tree et al., 1997). To Prot. 60, 677–681.


Abad, F.X., Pinto, R.M., Villena, C., Gajardo, R., Bosch, A., laboratories for monitoring bacteriological and viral contamina-1997b. Astrovirus survival in drinking water. Appl. Environ. tion of bivalve molluscs (1999/313/BC). Off. J. Eur. Com-Microbiol. 63, 3119–3122. munities. 120, 40–41.

Allen, K., 1899. The transmission of typhoid by sewage polluted Anon, 1999b. Wilcox, J. Ryan v Great Lakes Council, 1999.oysters. Rep. Am. Public Health Assoc. 25, 154–165. Federal Court of Australia. FCA 177. p. 1–110

Alouini, S., Sobsey, M.D., 1995. Evaluation of an extraction- Apairemarchais, V., Robertson, B.H., Aubineauferre, V., Leroux,precipitation method for recovering hepatitis A virus and M.G., Leveque, F., Schwartzbrod, L., Billaudel, S., 1995.poliovirus from hardshell clams (Mercenaria mercenaria). Direct sequencing of hepatitis A virus strains isolated during anWater Sci. Technol. 31, 465–469. epidemic in France. Appl. Environ. Microbiol. 61, 3977–3980.

Ando, T., Jin, Q., Gentsch, J.R. et al., 1995a. Epidemiologic Appleton, H., 1994. Norwalk virus and the Small Round Virusesapplications of novel molecular methods to detect and differen- causing foodborne gastroenteritis. In: Hui, Y.H., Gorham, J.R.,tiate small round structured viruses (Norwalk-like viruses). J. Murrell, K.D., Cliver, D.O. (Eds.), Foodborne Disease Hand-Med. Virol. 47, 145–152. book. Marcel Dekker, New York, pp. 57–79.

Ando, T., Monroe, S.S., Gentsch, J.R., Jin, Q., Lewis, D.C., Glass, Appleton, H., Palmer, S.R., Gilbert, R.J., 1981. FoodborneR.I., 1995b. Detection and differentiation of antigenically gastroenteritis of unknown aetiology: A virus infection? Br.distinct small round-structured viruses (Norwalk-like viruses) Med. J. 282, 1801–1802.by reverse transcription PCR and southern hybridization. J. Appleton, H., Pereira, M.S., 1977. A possible virus aetiology inClin. Microbiol. 33, 64–71. outbreaks of food poisoning from cockles. Lancet. i, 780–781.

Ando, T., Mulders, M.N., Lewis, D.C., Estes, M.K., Monroe, S.S., Armigliato, M., Bortolotti, F., Bertaggia, A. et al., 1986. Epi-Glass, R.L., 1994. Comparison of the polymerase region of demiology of hepatitis A in northern Italy a seven-year survey.small round structured virus strains previously classified in Infection 14, 283–285.three antigenic types by solid-phase immune electron micro- Atmar, R.L., Metcalf, T.G., Neill, F.H., Estes, M.K., 1993.scopy. Arch.Virol. 135, 217–226. Detection of enteric viruses in oysters by using the polymerase

Ang, L.H., 1998. An outbreak of viral gastroenteritis associated chain reaction. Appl. Environ. Microbiol. 59, 631–635.with eating raw oysters. Comm. Dis. Pub. Health 1, 38–40. Atmar, R.L., Neill, F.H., Romalde, J.L. et al., 1995. Detection of

Anon, 1976. Council Directive of 8th December 1975 concerning Norwalk virus and hepatitis A virus in shellfish tissues with thethe quality of bathing water V(76/160/EEC). Off. J. Eur. PCR. Appl. Environ. Microbiol. 61, 3014–3018.Communities. L31, 1–7. Atmar, R.L., Neill, F.H., Woodley, C.M. et al., 1996. Collabora-

Anon, 1979. Council Directive of 30 October 1979 on the quality tive evaluation of a method for the detection of Norwalk virusrequired of shellfish waters (79/923/EEC). Off. J. Eur. Com- in shellfish tissues by PCR. Appl. Environ. Microbiol. 62,munities. 281, 47–52. 254–258.

Anon, 1991a. Council Directive of 15th July 1991 laying down Beekwilder, J., Nieuwenhuizen, R., Havelaar, A.H., Duin, J.v.,the health conditions for the production and placing on the 1995. An oligonucleotide hybridization assay for the identifica-market of live bivalve molluscs (91/492/EEC). Off. J. Eur. tion and enumeration of F-specific RNA phages in surfaceCommunities. water. J. App. Bact. 80, 179–186.

Anon, 1991b. Council Directive of 21st May 1991 concerning Beller, M., Ellis, A., Lee, S.H. et al., 1997. Outbreak of viralurban waste water treatment (91/271/EEC). Off. J. Eur. gastroenteritis due to a contaminated well – internationalCommunities. 135, 40–45. consequences. J. Am. Med. Assoc. 278, 563–568.

Anon, 1991c. Council Directive of 22nd July 1991 laying down Berke, T., Golding, B., Jiang, X. et al., 1997. Phylogeneticthe health conditions for the production and placing on the analysis of the caliciviruses. J. Med. Virol. 52, 419–424.market of fishery products (91/493/EEC). Off. J. Eur. Com- Bird, P., Kraa, E., 1995. Overview of the 1990 viral gastroenteritismunities. 268, 15–34. outbreak from oysters. In: Poggi, R., Le Gall, J.Y. (Eds.),

Anon, 1992. Shellfish-associated illness. Comm. Dis. Report. 2, Purification des Coquillages (Shellfish Depuration). Proceed-39. ings of the Second International Conference on Shellfish

Anon, 1993a. Council Decision of 11th December 1992 approving Depuration, Rennes, France, 1992. IFREMER-Centre de Brest,certain heat treatments to inhibit the development of patho- Plouzane, pp. 31–36.genic micro-organisms in bivalve molluscs and marine gas- Birkenmeyer, L.G., Mushahwar, I.K., 1994. Detection of hepatitistropods (93/25/EEC). Off J. Eur. Communities. 16, 22–23. A, B and D virus by the polymerase chain reaction. J. Virol.

Anon, 1993b. National Shellfish Sanitation Program, Manual of Methods 49, 101–112.Operations, 1993 Revision. USA Department of Health and Bishop, R.F., 1996. Natural history of human rotavirus infection.Human Services, Public Health Service, Food and Drug Arch. Virol. 12, 119–128.Administration. Bobowick, A.R., Brody, J.A., Matthews, M.R., Roos, R., Gaj-

Anon, 1993c. Shellfish-associated illness. Comm. Dis. Report. 3, dusek, D.C., 1973. Creutzfeldt-Jakob disease: a case-control29. study. Am. J. Epidemiol. 98, 381–394.

Anon, 1996. Small round structured viruses. Comm. Dis. Report. Boher, S., Beril, C., Terver, D., Schwartzbrod, L., 1991. Com-6, 207. parison of two methods for the recovery of rotavirus from

Anon, 1998. Outbreaks of gastroenteritis in England and Wales mussels and oysters. Water Sci. Technol. 24, 423–426.associated with shellfish: 1996 and 1997. Comm. Dis. Report. Bostock, A.D., Mepham, P., Phillips, S., 1979. Hepatitis A8, 21–24. infection associated with the consumption of mussels. J. Infect.

Anon, 1999a. Council Decision of 29th April 1999 on reference 1, 171–177.


Botero, L., Montiel, M., Porto, L., 1996. Enteroviruses in shrimp sorbent assay with monoclonal antibodies. J. Clin. Microbiol.harvested from contaminated marine waters. Int. J. Environ. 22, 119–124.Health Res. 6, 103–108. Cromeans, T., Nainan, O.V., Fields, H.A., Favorov, M.O., Mar-

Bradley, D.W., 1995. Hepatitis B virus–a brief review of the golis, H.S., 1994. Hepatitis A and B viruses. In: Hui, Y.H.,biology, molecular virology, and immunology of a novel virus. Gorham, J.R., Murrell, K.D., Cliver, D.O. (Eds.), FoodborneJ. Hepatol. 22, 140–145. Disease Handbook. Marcel Dekker, New York, pp. 1–55.

Bridger, J.C., 1994. Non-group A rotaviruses. In: Kapikian, A.Z. Cromeans, T.L., Nainan, O.V., Margolis, H.S., 1997. Detection of(Ed.), Viral Infections of the Gastrointestinal Tract. Marcel hepatitis A virus RNA in oyster meat. Appl. Environ. Mi-Dekker, New York, pp. 369–407. crobiol. 63, 2460–2463.

Callahan, K.M., Taylor, D.J., Sobsey, M.D., 1995. Comparative Cubitt, W.D., 1996. Historical background and classification ofsurvival of hepatitis A virus, poliovirus and indicator viruses in

caliciviruses and astroviruses. Arch. Virol. 12, 225–235.geographically diverse seawaters. Water Sci. Technol. 31, 189–

Dalton, C.B., Haddix, A., Hoffman, R.E., Mast, E.E., 1996. The193.

cost of a food-borne outbreak of hepatitis A in Denver,Carter, M.J., Willcocks, M.M., 1996. The molecular biology of

Colorado. Arch. Intern. Med. 156, 1013–1016.astroviruses. Arch. Virol. 12, 277–285.

Dastjerdi, A.M., Green, J., Gallimore, C.I., Brown, D.G., Bridger,Caul, E.O., 1996a. Viral gastroenteritis–small round structured

J.C., 1999. The bovine Newbury agent-2 is genetically moreviruses, caliciviruses and astroviruses. 1. The clinical andclosely related to human SRSVs than to animal caliciviruses.diagnostic perspective. J. Clin. Pathol. 49, 874–880.Virology 254, 1–5.Caul, E.O., 1996b. Viral gastroenteritis – small round structured

Davanipour, Z., Alter, M., Sobel, E., Asher, D.M., Gajdusek, D.C.,viruses, caliciviruses and astroviruses. 2. the epidemiological1985. A case-control study of Creutzfeldt-Jakob disease.perspective. J. Clin. Pathol. 49, 959–964.Dietary risk factors. Am. J. Epidemiol. 122, 443–451.Caul, E.O., Appleton, H., 1982. The electron microscopic and

Davis, C., Smith, A., Walden, R. et al., 1994. Viral gastroenteritisphysical characteristics of small round human fecal viruses anassociated with consumption of raw oysters – Florida, 1993. J.interim scheme for classification. J. Med. Virol. 9, 257–266.Am. Med. Assoc. 272, 510–511.Chalmers, J.W.T., Mcmillan, J.H., 1995. An outbreak of viral

De Leon, R., Matsui, S.M., Baric, R.S. et al., 1992. Detection ofgastroenteritis associated with adequately prepared oysters.Norwalk virus in stool specimens by reverse transcriptase-Epidemiol. Infect. 115, 163–167.polymerase chain reaction and nonradioactive oligoprobes. J.Christensen, B.F., Lees, D., Wood, K.H., Bjergskov, T., Green, J.,Clin. Microbiol. 30, 3151–3157.1998. Human enteric viruses in oysters causing a large

Debartolomeis, J., Cabelli, V.J., 1991. Evaluation of an escherich-outbreak of human food borne infection in 1996/97. J.ia-coli host strain for enumeration of F male-specific bac-Shellfish Res. 17, 1633–1635.teriophages. Appl. Environ. Microbiol. 57, 1301–1305.Chung, H., Jaykus, L.A., Lovelace, G., Sobsey, M.D., 1998.

Deng, M.Y., Day, S.P., Oliver, D.O., 1994. Detection of hepatitisBacteriophages and bacteria as indicators of enteric viruses inoysters and their harvest waters. Water Sci. Technol. 38, 37– A virus in environmental samples by antigen-capture PCR.44. Appl. Environ. Microbiol. 60, 1927–1933.

Chung, H.M., Jaykus, L.A., Sobsey, M.D., 1996. Detection of Desenclos, J.-C.A., Klontz, K.C., Wilder, M.H., Nainan, O.V.,human enteric viruses in oysters by in vivo and in vitro Margolis, H.S., Gunn, R.A., 1991. A multistate outbreak ofamplification of nucleic acids. Appl. Environ. Microbiol. 62, hepatitis A caused by the consumption of raw oysters. Am. J.3772–3778. Public Health 81, 1268–1272.

Clarke, I.N., Lambden, P.R., 1997. The molecular biology of Dingle, K.E., Lambden, P.R., Caul, E.O., Clarke, I.N., 1995.caliciviruses. J. Gen. Virol. 78, 291–301. Human enteric caliciviridae – the complete genome sequence

Cliver, D.O., 1994a. Epidemiology of viral foodborne disease. J. and expression of virus-like particles from a genetic group IIFood Prot. 57, 263–266. small round structured virus. J. Gen. Virol. 76, 2349–2355.

Cliver, D.O., 1994b. Other foodborne viral diseases. In: Hui, Y.H., Dodgson, R.W., 1928. Report on mussel purification. FisheriesGorham, J.R., Murrell, K.D., Cliver, D.O. (Eds.), Foodborne Investigations Series II. 10, 1-498. Ministry of Agriculture andDisease Handbook. Marcel Dekker, New York, pp. 137–158. Fisheries, HMSO, London.

Cohen, J.I., Ticehurst, J.R., Purcell, R.H., Buckler-White, A., Dore, W.J., Henshilwood, K., Lees, D.N., 1998. The developmentBaroudy, M., 1987. Complete nucleotide sequence of wild-type of management strategies for control of virological quality inhepatitis A virus: comparison with different strains of hepatitis oysters. Water Sci. Technol. 38, 29–35.A virus and other picornaviruses. J. Virol. 61, 50–59. Dore, W.J., Henshilwood, K., Lees, D.N., 2000. Evaluation of

Cole, M.T., Kilgen, M.B., Hackney, C.R., 1986. Evaluation of F-specific RNA bacteriophage as a candidate human entericmethods for extraction of enteric virus from Louisiana oysters. virus indicator for bivalve molluscan shellfish. Appl. Environ.J. Food Prot. 49, 592–595. Microbiol. 66, 1280–1285.

Cook, D.W., Ellender, R.D., 1986. Relaying to decrease the Dore, W.J., Lees, D.N., 1995. Behavior of Escherichia coli andconcentration of oyster-associated pathogens. J. Food Prot. 49, male-specific bacteriophage in environmentally contaminated196–202. bivalve molluscs before and after depuration. Appl. Environ.

Coulepis, A.G., Veale, M.F., MacGregor, A., Kornitschuk, M., Microbiol. 61, 2830–2834.Gust, I., 1985. Detection of hepatitis A virus and antibody by Dowell, S.F., Groves, C., Kirkland, K.B. et al., 1995. A multistatesolid-phase radioimmunoassay and enzyme-linked immuno- outbreak of oyster-associated gastroenteritis-implications for


interstate tracing of contaminated shellfish. J. Infect. Dis. 171, Gouvea, V., Glass, R.I., Woods, P. et al., 1990. Polymerase chain1497–1503. reaction amplification and typing of rotavirus nucleic acid from

Dubois, E., Leguyader, F., Haugarreau, L., Kopecka, H., Cormier, stool specimens. J. Clin. Microbiol. 28, 276–282.M., Pommepuy, M., 1997. Molecular epidemiological survey Gouvea, V., Santos, N., Timenetsky, M.D., Estes, M.K., 1994.of rotaviruses in sewage by reverse transcriptase seminested Identification of Norwalk virus in artificially seeded shellfishPCR and restriction fragment length polymorphism assay. and selected foods. J. Virol. Methods 48, 177–187.Appl. Environ. Microbiol. 63, 1794–1800. Grabow, W.O.K., Kfir, R., Genthe, B., 1993. Advantages and

Evans, H.S., Madden, P., Douglas, C. et al., 1998. General disadvantages of the use of immunodetection techniques for theoutbreaks of infectious intestinal disease in England and Wales: enumeration of microorganisms and toxins in water. Water Sci.1995 and 1996. Comm. Dis. Pub. Health 1, 165–171. Technol. 27, 243–252.

Fankhauser, R.L., Noel, J.S., Monroe, S.S., Ando, T., Glass, R.I.,Gray, J.J., Jiang, X., Morgan-Capner, P., Desselberg, U., Estes,

1998. Molecular epidemiology of ‘‘Norwalk-like viruses’’ inM.K., 1993. Prevalence of antibodies to Norwalk virus in

outbreaks of gastroenteritis in the United States. J. Infect. Dis.England, detection by enzyme-linked immunosorbent assay

178, 1571–1578.using baculovirus-expressed Norwalk virus capsid antigen. J.

Farley, T.A., Mcfarland, L., Estes, M., Schwab, K., 1998. ViralClin. Microbiol. 31, 1022–1025.

gastroenteritis associated with eating oysters – Louisiana,Green, J., Gallimore, C.I., Norcott, J.P., Lewis, D., Brown, D.W.G.,December 1996 to January 1997. J. Am. Med. Assoc. 279,

1995a. Broadly reactive reverse transcriptase polymerase chain10–11.reaction for the diagnosis of SRS V-associated gastroenteritis.Fleissner, M.L., Herrmann, I.E., Booth, J.W., Blacklow, N.R.,J. Med. Virol. 47, 392–398.Nowak, N.A., 1989. Role of Norwalk virus in two foodborne

Green, J., Henshilwood, K., Gallimore, C.I., Brown, D.W.G., Lees,outbreaks of gastroenteritis definitive virus association. Am. J.D.N., 1998. A nested reverse transcriptase PCR assay forEpidemiol. 129, 165–172.detection of small round-structured viruses in environmentallyFuhrman, J.A., 1999. Marine viruses and their biogeochemical andcontaminated molluscan shellfish. Appl. Environ. Microbiol.ecological effects. Lancet 399, 541–548.64, 858–863.Fujiyama, S., Akahoshi, M., Sagara, K., Sato, T., Tsurusaki, R.,

Green, S.M., Dingle, K.E., Lambden, P.R., Caul, E.O., Ashley,1985. An epidemic of hepatitis A related to ingestion of rawC.R., Clarke, I.N., 1994. Human enteric Caliciviridae: A newoysters. Gastroenterol. Jpn. 20, 6–13.prevalent small round-structured virus group defined by RNA–Gajardo, R., Bouchriti, N., Pinto, R.M., Bosch, A., 1995.dependent RNA polymerase and capsid diversity. J. Gen. Virol.Genotyping of rotaviruses isolated from sewage. Appl. En-75, 1883–1888.viron. Microbiol. 61, 3460–3462.

Green, S.M., Lambden, P.R., Deng, Y. et al., 1995b. PolymeraseGantzer, C., Dubois, E., Crance, J.M. et al., 1998. Influence ofchain reaction detection of small round-structured viruses fromenvironmental factors on the survival of enteric viruses intwo related hospital outbreaks of gastroenteritis using inosine-seawater. Oceanol. Acta 21, 983–992.

Gdalevich, M., Grotto, I., Mandel, Y., Mimouni, D., Shemer, J., containing primers. J. Med. Virol. 45, 197–202.Ashkenazi, I., 1998. Hepatitis A antibody prevalence among Greenberg, H.B., Matsui, S.M., 1992. Astrovirus and calicicivir-young adults in Israel – the decline continues. Epidemiol. uses: emerging enteric pathogens. Infect. Agents Dis. 1, 71–91.Infect. 121, 477–479. Grohmann, G.S., Murphy, A.M., Christopher, P.J., Auty, E.,

Gerba, C.P., Goyal, S.M., 1978. Detection and occurrence of Greenberg, H.B., 1981. Norwalk virus gastro enteritis inenteric viruses in shellfish: A review. J. Food Prot. 41, 743– volunteers consuming depurated oysters crassostrea-commer-745. cialis. Aust. J. Exp. Biol. Med. Sci. 59, 219–228.

Giachino, N., Costa, L., Livoti, V., Onofaro, V., Pipito, S., 1985. Gunn, R.A., Janowski, H.T., Lieb, S., Prather, E.C., Greenberg,Identification of the hepatitis A virus in mussels by ELISA. Ig. H.B., 1982. Norwalk virus gastroenteritis following raw oysterMod. 83, 975–978. consumption. Am. J. Epidemiol. 115, 348–351.

Gill, O.N., Cubitt, W.D., Mcswiggan, D.A., Watney, B.M., Bartlett, Guzewich, J.J., Morse, D.L., 1986. Sources of shellfish in out-C.L.R., 1983. Epidemic of gastro enteritis caused by oysters breaks of probable viral gastroenteritis: implications for con-contaminated with small round structured viruses. Br. Med. J. trol. J. Food Prot. 49, 389–394.287, 1532–1534. Hackney, C.R., Pierson, M.D., 1994. Environmental Indicators

Girones, R., Puig, M., Allard, A., Lucena, F., Wadell, G., Jofre, J., and Shellfish Safety. Chapman and Hall, New York, pp. 1–523.1995. Detection of adenovirus and enterovirus by PCR amplifi- Hafliger, D., Gilgen, M., Luthy, J., Hubner, P., 1997. Seminestedcation in polluted waters. Water Sci. Technol. 31, 351–357. RT-PCR systems for small round structured viruses and

Glass, R.I., Gentsch, J.R., Ivanoff, B., 1996a. New lessons for detection of enteric viruses in seafood. Int. J. Food Microbiol.rotavirus vaccines. Science 272, 46–48. 37, 27–36.

Glass, R.I., Noel, J., Mitchell, D. et al., 1996b. The changing Halliday, M.L., Kang, L.-Y., Zhou, T.-K. et al., 1991. An epidemicepidemiology of astrovirus-associated gastroenteritis – a re- of hepatitis A attributable to the ingestion of raw clams inview. Arch. Virol. 12, 287–300. Shanghai China. J. Infect. Dis. 164, 852–859.

Goswami, B.B., Koch, W.H., Cebula, T.A., 1993. Detection of Havelaar, A.H., 1987. Bacteriophages as model organisms inhepatitis A virus in mercenaria–mercenaria by coupled reverse water treatment. Microbiol. Sci. 4, 362–364.transcription and polymerase chain reaction. Appl. Environ. Havelaar, A.H., Van Olphen, M., Drost, Y.C., 1993. F-specificMicrobiol. 59, 2765–2770. RNA bacteriophages are adequate model organisms for enteric


viruses in fresh water. Appl. Environ. Microbiol. 59, 2956– and antigenic diversity of human caliciviruses (HUCVs) using2962. RT-PCR and new EIAs. Arch. Virol. 12, 251–262.

Hedberg, C.W., Osterholm, M.T., 1993. Outbreaks of food-borne Jiang, X., Turf, E., Hu, J. et al., 1996b. Outbreaks of gastroen-and waterborne viral gastroenteritis. Clin. Microbiol. Rev. 6, teritis in elderly nursing homes and retirement facilities associ-199–210. ated with human caliciviruses. J. Med. Virol. 50, 335–341.

Heller, D., Gill, O.N., Raynham, E., Kirkland, T., Zadick, P.M., Jiang, X., Wang, J., Estes, M.K., 1995. Characterization of SRSVsStanwell-Smith, R., 1986. An outbreak of gastrointestinal using RT-PCR and a new antigen ELISA. Arch. Virol. 140,illness associated with consumption of raw depurated oysters. 363–374.Br. Med. J. 292, 1726–1727. Jiang, X., Wang, J., Graham, D.Y., Estes, M.K., 1992. Detection of

Henshilwood, K., Green, J., Lees, D.N., 1998. Monitoring the Norwalk virus in stool by polymerase chain reaction. J. Clin.marine environment for small round structured viruses Microbiol. 30, 2529–2534.(SRSVs): a new approach to combating the transmission of Jiang, X., Wang, M., Wang, K., Estes, M.K., 1993. Sequence andthese viruses by molluscan shellfish. Water Sci. Technol. 38, genomic organization of Norwalk virus. Virology 195, 51–61.51–56. Johnson, K.M., Cooper, R.C., Straube, D.C., 1981. Procedure for

Hernandez, F., Monge, R., Jimenez, C., Taylor, L., 1997. recovery of enteroviruses from the Japanese cockle TapesRotavirus and hepatitis A virus in market lettuce (latuca sativa) japonica. Appl. Environ. Microbiol. 41, 932–935.in Costa Rica. Int. J. Food Microbiol. 37, 221–223. Jonassen, T.O., Monceyron, C., Lee, T.W., Kurtz, J.B., Grinde, B.,

Herrmann, J.E., Blacklow, N.R., Matsui, S.M. et al., 1995. 1995. Detection of all serotypes of human astrovirus by theMonoclonal antibodies for detection of Norwalk virus antigen polymerase chain reaction. J. Virol. Methods 52, 327–334.in stools. J. Clin. Microbiol. 33, 2511–2513. Kapikian, A.Z., 1994. Norwalk and Norwalk-like viruses. In:

Herrmann, J.E., Nowak, N.A., Blacklow, N.R., 1985. Detection of Kapikian, A.Z. (Ed.), Viral Infections of the GastrointestinalNorwalk virus in stools by enzyme immunoassay. J. Med. Tract. Marcel Dekker, New York, pp. 471–518.Virol. 17, 127–134. Kapikian, A.Z., 1996. Overview of viral gastroenteritis. Arch.

Herrmann, J.E., Nowak, N.A., Perron-Henry, D.M., Hudson, R.W., Virol. 12, 7–19.Cubitt, W.D., Blacklow, N.R., 1990. Diagnosis of astrovirus Kaplan, J.E., Feldman, R., Campbell, D.S., Lookabaugh, C., Gary,gastroenteritis by antigen detection with monoclonal antibo- G.W., 1982a. The frequency of a Norwalk-like pattern of illnessdies. J. Infect. Dis. 161, 226–229. in outbreaks of acute gastro enteritis. Am. J. Public Health 72,

Hsu, F.C., Shieh, Y.S.C., Vanduin, J., Beekwilder, M.J., Sobsey, 1329–1332.M.D., 1995. Genotyping male–specific RNA coliphages by Kaplan, I.E., Gary, G.W., Singh, R.C.B.N., Schonberger, L.B.,hybridization with oligonucleotide probes. Appl. Environ. Feldman, R., Greenberg, H.B., 1982b. Epidemiology of Nor-Microbiol. 61, 3960–3966. walk gastro enteritis and the role of Norwalk virus in outbreaks

Humphrey, T.J., Martin, K.W., Walker, D., 1995. Bacteriophage – of acute non bacterial gastro enteritis. Ann. Intern. Med. 96,a possible indicator for SRSV in oysters. Microbiol. Digest. 12, 756–761.205–207. Kawamoto, H., Hasegawa, S., Sawatari, S. et al., 1993. Small,

IAWPRC Study group on health related water microbiology, 1991. round-structured viruses (SRSVs) associated with acute gas-Bacteriophages as model viruses in water quality control. Water troenteritis outbreaks in Gifu. Jpn. Microbiol. Immunol. 37,Res. 25, 529–545. 991–997.

Jansen, R.W., Siegl, G., Lemon, S.M., 1990. Molecular epidemiol- Khan, A.S., Moe, C.L., Glass, R.I. et al., 1994. Norwalk virus-ogy of human hepatitis A virus defined by an antigen-capture associated gastroenteritis traced to ice consumption aboard apolymerase chain reaction method. Proc. Natl. Acad. Sci. USA cruise ship in Hawaii: comparison and application of molecular87, 2867–2871. method-based assay. J. Clin. Microbiol. 32, 318–322.

Jaykus, L.A., De Leon, R., Sobsey, M.S., 1993. Application of Kiyosawa, K., Gibo, Y., Sodeyama, T. et al., 1987. PossibleRT-PCR for the detection of enteric viruses in oysters. Water infectious causes in 651 patients with acute viral hepatitisSci. Technol. 27, 49–53. during a 10-year period 1976–1985. Liver 7, 163–168.

Jaykus, L.A., Deleon, R., Sobsey, M.D., 1995. Development of a Koff, R.S., 1995. Seroepidemiology of hepatitis A in the Unitedmolecular method for the detection of enteric viruses in States. J. Infect. Dis. 171, 19–23.oysters. J. Food Prot. 58, 1357–1362. Koff, R.S., Grady, G.F., Chalmers, T.C., Mosley, J.W., Swartz,

Jaykus, L.A., Deleon, R., Sobsey, M.D., 1996. A virion con- B.L., 1967. Viral hepatitis in a group of Boston hospitals. III.centration method for detection of human enteric viruses in Importance of exposure to shellfish in a non-epidemic period.oysters by PCR and oligoprobe hybridization. Appl. Environ. New Engl. J. Med. 276, 703–710.Microbiol. 62, 2074–2080. Kogawa, K., Nakata, S., Ukae, S. et al., 1996. Dot blot hybridiza-

Jaykus, L.A., Hemard, M.T., Sobsey, M.D., 1994. Human enteric tion with a cDNA probe derived from the human caliciviruspathogenic viruses. In: Hackney, C.R., Pierson, M.D. (Eds.), sapporo 1982 strain. Arch. Virol. 141, 1949–1959.Environmental Indicators and Shellfish Safety. Chapman and Kohn, M.A., Farley, T.A., Ando, T. et al., 1995. An outbreak ofHall, New York, pp. 92–153. Norwalk virus gastroenteritis associated with eating raw oys-

Jiang, X., Cubitt, W.D., Berke, T. et al., 1997. Sapporo-like human ters: Implications maintaining safe oyster beds. J. Am. Med.caliciviruses are genetically and antigenically diverse. Arch. Assoc. 273, 466–471.Virol. 142, 1813–1827. Kurtz, J.B., Lee, T.W., 1987. Astroviruses: Human animal. In:

Jiang, X., Matson, D.O., Cubitt, W.D., Estes, M.K., 1996a. Genetic Bock, G., Whelan, J. (Eds.), Novel Diarrhoea Viruses. Founda-


tion Symp, Vol. No 128. John Wiley, Chichester, UK, pp. Shanghai 1988 by contaminated clams. Acta. Acad. Med.92–107. Shanghai 19, 129–134.

Lambden, P.R., Caul, E.O., Ashley, C.R., Clarke, I.N., 1993. Linco, S.J., Grohmann, G.S., 1980. The Darwin outbreak ofSequence and genome organization of a human small round- oyster-associated viral gastroenteritis. Med. J. Aust. 1, 211–structured Norwalk-like virus. Science 259, 516–519. 213.

Landry, E.F., Vaughn, J.M., Vicale, T.J., 1980. Modified procedure Lipp, E.K., Rose, J.B., 1997. The role of seafood in foodbornefor extraction of poliovirus from naturally infected oysters disease in the United States of America. Rev. Sci. Tech. Off.using CatFloc and beef extract. J. Food Prot. 43, 91–94. Int. Epiz. 16, 620–640.

Le Guyader, F., Apaire-Marchais, V., Brillet, J., Billaudel, S., Liu, B.L., Clarke, I.N., Caul, E.O., Lambden, P.R., 1995. Human1993. Use of genomic probes to detect hepatitis A virus and enteric caliciviruses have a unique genome structure and areenterovirus RNAs in wild shellfish and relationship of viral

distinct from the Norwalk-like viruses. Arch. Virol. 140, 1345–contamination to bacterial contamination. Appl. Environ. Mi-

1356.crobiol. 59, 3963–3968.

Liu, B.L., Lambden, P.R., Gunther, H., Otto, P., Elschner, M.,Le Guyader, F., Dubois, E., Menard, D., Pommepuy, M., 1994.

Clarke, I.N., 1999. Molecular characterization of a bovineDetection of hepatitis A virus, rotavirus, and enterovirus in

enteric calicivirus: relationship to the Norwalk-like viruses. J.naturally contaminated shellfish and sediment by reverse

Virol. 73, 819–825.transcription-seminested PCR. Appl. Environ. Microbiol. 60,Lopezsabater, E.I., Deng, M.Y., Cliver, D.O., 1997. Magnetic3665–3671.

immunoseparation PCR assay (MIPA) for detection of hepatitisLe Guyader, F., Estes, M.K., Hardy, M.E. et al., 1996a. EvaluationA virus (HAV) in American oyster (Crassostrea virginica).of a degenerate primer for the PCR detection of humanLett. Appl. Microbiol. 24, 101–104.caliciviruses. Arch. Virol. 141, 2225–2235.

Lucena, F., Lasobras, J., Mcintosh, D., Forcadell, M., Jofre, J.,Le Guyader, F., Neill, F.H., Estes, M.K., Monroe, S.S., Ando, T.,1994. Effect of distance from the polluting focus on relativeAtmar, R.L., 1996b. Detection and analysis of a small roundconcentrations of Bacteroides fragilis phages and coliphages instructured virus strain in oysters implicated in an outbreak ofmussels. Appl. Environ. Microbiol. 60, 2272–2277.acute gastroenteritis. Appl. Environ. Microbiol. 62, 4268–

Maguire, A.J., Green, J., Brown, D.G., Desselberger, U., Gray,4272.J.J., 1999. Molecular epidemiology of outbreaks of gastroen-Le Guyader, F., Miossec, L., Haugarreau, L., Dubois, E.,teritis associated with small round-structured viruses in EastKopecka, H., Pommepuy, M., 1998. RT–PCR evaluation ofAnglia, United Kingdom, during the 1996–1997 season. J.viral contamination in five shellfish beds over a 21-monthClin. Microbiol. 37, 81–89.period. Water Sci. Technol. 38, 45–50.

Maguire, H.C., Handford, S., Perry, K.R. et al., 1995. A collabora-Lee, J.V., Dawson, S.R., Ward, S., Surman, S.B., Neal, K.R., 1997.tive case control study of sporadic hepatitis A in England.Bacteriophages are a better indicator of illness rates thanComm. Dis. Report 5, 33–40.bacteria amongst users of a white water course fed by a

lowland river. Water Sci. Technol. 35, 165–170. Malfait, P., Lopalco, P.L., Salmaso, S. et al., 1996. An outbreak ofLee, T.W., Kurtz, J.B., 1981. Serial propagation of astrovirus in hepatitis A in Puglia, Italy. Eurosurveillance 1, 33–35.

tissue culture with the aid of trypsin. J. Gen. Virol. 57, 421– Marx, F.E., Taylor, M.B., Grabow, W.O.K., 1997. A comparison424. of two sets of primers for the RT–PCR detection of astrovir-

Lees, D.N., Henshilwood, K., Butcher, S., 1995a. Development of uses in environmental samples. Water South Africa 23, 257–a PCR-based method for the detection of enteroviruses and 262.hepatitis virus in molluscan shellfish and its application to Mattion, N.M., Cohen, J., Estes, M.K., 1994. The rotaviruspolluted field samples. Water Sci. Technol. 31, 457–464. proteins. In: Kapikian, A.Z. (Ed.), Viral Infections of the

Lees, D.N., Henshilwood, K., Dore, W.J., 1994. Development of a Gastrointestinal Tract. Marcel Dekker, New York, pp. 169–method for detection of enteroviruses in shellfish by PCR with 249.poliovirus as a model. Appl. Environ. Microbiol. 60, 2999– McCaustland, K.A., Bi, S., Purdy, M.A., Bradley, D.W., 1991.3005. Application of two RNA extraction methods prior to amplifica-

Lees, D.N., Henshilwood, K., Green, J., Gallimore, C.I., Brown, tion of hepatitis E virus nucleic acid by the polymerase chainD.W.G., 1995b. Detection of small round structured viruses in reaction. J. Virol. Methods 35, 331–342.shellfish by reverse transcription PCR. Appl. Environ. Mi- Mcdonnell, S., Kirkland, K.B., Hlady, W.G. et al., 1997. Failure ofcrobiol. 61, 4418–4424. cooking to prevent shellfish-associated viral gastroenteritis.

Lew, J.F., Kapikian, A.Z., Valdesuso, J., Green, K.Y., 1994. Arch. Intern. Med. 157, 111–116.Molecular characterization of Hawaii virus and other Norwalk- Mele, A., Pasquini, P., Pana, A., 1991. Hepatitis A in italylike viruses: Evidence for genetic polymorphism among human epidemiology and suggestions for control. Ital. J. Gastroenterol.caliciviruses. J. Infect. Dis. 170, 535–542. 23, 341–343.

Lewis, G.D., Metcalf, T.G., 1988. Polyethylene glycol precipi- Mele, A., Rastelli, M.G., Gill, O.N. et al., 1989. Recurrenttation for recovery of pathogenic viruses including hepatitis A epidemic hepatitis A associated with consumption of rawvirus and human rotavirus from oyster water and sediment shellfish probably controlled through public health measures.samples. Appl. Environ. Microbiol. 54, 1983–1988. Am. J. Epidemiol. 130, 540–546.

Li, Z., Wang, Y., Wang, J., Ju, L., Zhan, Y., Xu, Z., 1992. Mele, A., Rosmini, F., Zampieri, A., Gill, O.N., 1986. IntegratedExperimental HAV infection of marmoset an epidemic in epidemiological system for acute viral hepatitis in Italy


(SEIEVA): description and preliminary results. Eur. J. Epi- gastroenteritis in the United States, 1990 to 1995. J. Med.Virol.demiol. 2, 300–304. 53, 372–383.

Mele, A., Stazi, M.A., Corona, R. et al., 1990. Decline of Norcott, J.P., Green, J., Lewis, D., Estes, M.K., Barlow, K.I.,incidence of A B and non-A non-B hepatitis in Italy results of Brown, D.W.G., 1994. Genomic diversity of small roundfour years surveillance 1985–88. Ital. J. Gastroenterol. 22, structured viruses in the United Kingdom. J. Med. Virol. 44,274–280. 280–286.

Metcalf, T.G., Melnick, J.L., Estes, M.K., 1995. Environmental O’Mahony, M.C., Gooch, C.D., Smyth, D.A., Thrussell, A.J.,virology: From detection of virus in sewage and water by Bartlett, C.L.R., Noah, N.D., 1983. Epidemic hepatitis A fromisolation to identification by molecular biology – A trip of over cockles. Lancet. 8323 518–520.50 years. Annu. Rev. Microbiol. 49, 461–487. Oishi, I., Yamazaki, K., Kimoto, T. et al., 1994. A large outbreak

Millard, J., Appleton, H., Parry, J.V., 1987. Studies on heatof acute gastroenteritis associated with astrovirus among

inactivation of hepatitis A virus with special reference tostudents and teachers in Osaka, Japan. J. Infect. Dis. 170,

shellfish. Epidemiol. Infect. 98, 397–414.439–443.

Minor, P.D., 1991. Picornaviridae. In: Franki, R.I.B., Fauquet,Otsu, R., 1999. Outbreaks of gastroenteritis caused by SRSVs

C.M., Knudson, D.L., Brown, F. (Eds.), Classification andfrom 1987 to 1992 in Kyushu, Japan: Four outbreaks associ-

Nomenclature of Viruses: the Fifth Report of the Internationalated with oyster consumption. Eur. J. Epidemiol. 15, 175–180.Committee on Taxonomy of Viruses. Springer Verlag, Vienna,

Otwell, W.S., Rodrick, G.E., Martin, R.E. (Eds.), 1991. Molluscanp. 320.Shellfish Depuration. CRC Press, Boca Raton, Florida.Minor, P.D., Morgan-Capner, P., Schild, G.C., 1990. The En-

Pebody, R.G., Leino, T., Ruutu, P. et al., 1998. Foodborneteroviruses. In: Zuckerman, A.J., Banatvala, J.E., Pattison, J.R.outbreaks of Hepatitis A in a low endemic country: an(Eds.), Principles and Practice of Clinical Virology, 2ndemerging problem? Epidemiol. Infect. 120, 55–59.Edition. John Wiley, Chichester, pp. 389–409.

Perrett, K., Kudesia, G., 1995. Gastroenteritis associated withMoe, C.L., Allen, J.R., Monroe, S.S. et al., 1991. Detection ofoysters. Comm. Dis. Rep. 5, 153–154.astrovirus in pediatric stool samples by immunoassay and RNA

PHLS Viral Gastroenteritis Sub-Committee, 1993. Outbreaks ofprobe. J. Clin. Microbiol. 29, 2390–2395.gastroenteritis associated with SRSVs. PHLS Microbiol. DigestMoe, C.L., Gentsch, T.J., Ando, T. et al., 1994. Application of10, 2–8.PCR to detect Norwalk virus in fecal specimens from outbreaks

Pietri, C., Hugues, B., Crance, J.M., Puel, D., Cmi, C., Deloince,of gastroenteritis. J. Clin. Microbiol. 32, 642–648.R., 1988. Hepatitis A virus levels in shellfish exposed in aMonroe, S.S., Stine, S.E., Jiang, X., Ester, M.K., Glass, R.I., 1993.natural marine environment to the effluent from a treatedDetection of antibody to recombinant Norwalk virus antigen insewage outfall. Water Sci. Techno. 20, 229–234.specimens from outbreaks of gastroenteritis. J. Clin. Microbiol.

Pina, S., Jofre, J., Emerson, S.U., Purcell, R.H., Girones, R.,31, 2866–2872.1998a. Characterization of a strain of infectious hepatitis EMorris, R., Waite, W.M., 1980. Evaluation of procedures for

recovery of viruses from water. II. Detection systems. Water virus isolated from sewage in an area where hepatitis B is notRes. 14, 795–798. endemic. Appl. Environ. Microbiol. 64, 4485–4488.

Morse, D.L., Guzewich, J.J., Hanrahan, J.P. et al., 1986. Wide- Pina, S., Puig, M., Lucena, F., Jofre, J., Girones, R., 1998b. Viralspread outbreaks of clam- and oyster-associated gastroenteritis: pollution in the environment and in shellfish – human adeno-Role of Norwalk virus. New Engl. J. Med. 314, 678–681. virus detection by PCR as an index of human viruses. Appl.

Muir, P., Kammerer, U., Korn, K. et al., 1998. Molecular typing of Environ. Microbiol. 64, 3376–3382.enteroviruses – current status and future requirements. Clin. Poggi, R., Le Gall, J.Y. (Eds.), 1995. Purification des CoquillagesMicrobiol. Rev. 11, 202–227. (Shellfish Depuration). Proc. 2nd Int. Conf. Shellfish Depura-

Murphy, A.M., Grohmann, G.S., Christopher, R.J., Lopez, W.A., tion, Rennes, France, April 1992. IFREMER – Centre de Brest,Davey, G.R., Millsom, R.H., 1979. An Australia-wide outbreak Plouzane.of gastroenteritis from oysters caused by Norwalk virus. Med. Pontefract, R.D., Bishai, F.R., Hockin, J., Bergeron, G., Parent, R.,J. Aust. 2, 329–333. 1993. Norwalk-like viruses associated with a gastroenteritis

Murrin, K., Slade, J., 1997. Rapid detection of viable enterovir- outbreak following oyster consumption. J. Food Prot. 56, 604–uses in water by tissue culture and semi-nested polymerase 607.chain reaction. Water Sci. Technol. 35, 429–432. Power, U.F., Collins, J.K., 1989. Differential depuration of

Nairn, C., Galbraith, D.N., Clements, G.B., 1995. Comparison of poliovirus Escherichia-coli and a coliphage by the commonCoxsackie B neutralisation and enteroviral PCR in chronic mussel Mytilus-edulis. Appl. Environ. Microbiol. 55, 1386–fatigue patients. J. Med. Virol. 46, 310–313. 1390.

Nakata, S., Kogawa, K., Numata, K. et al., 1996. The epidemiolo- Power, U.F., Collins, J.K., 1990a. Elimination of coliphages andgy of human calicivirus / sapporo /82/ japan. Arch. Virol. 12, Escherichia-coli from mussels during depuration under varying263–270. conditions of temperature salinity and food availability. J. Food

Nasser, A.M., 1994. Prevalence and fate of hepatitis A virus in Prot. 53, 208–212.water. Crit. Rev. Environ. Sci. Technol. 24, 281–323. Power, U.F., Collins, J.K., 1990b. Elimination of coliphages and

Noel, J.S., Ando, T., Leite, J.P. et al., 1997. Correlation of patient Escherichia-coli from mussels during depuration under varyingimmune responses with genetically characterized small round- conditions of temperature salinity and food availability. J. Foodstructured viruses involved in outbreaks of nonbacterial acute Prot. 53, 226.


Pringle, C.R., 1998. Virus taxonomy – San Diego 1998. Arch. fish, seawater, and sediments. In: Berg, G. (Ed.), Methods forVirol. 143, 1449–1459. Recovering Viruses from the Environment. CRC Press, Boca

Richards, G.P., 1985. Outbreaks of shellfish-associated enteric Raton, FL, pp. 77–108.virus illness in the United States: requisite for development of Sobsey, M.D., Carrick, R.J., Jensen, H.R., 1978. Improved meth-viral guidelines. J. Food Prot. 48, 815–823. ods for detecting enteric viruses in oysters. Appl. Environ.

Richards, G.P., 1988. Microbial purification of shellfish a review Microbiol. 36, 121–128.of depuration and relaying. J. Food Prot. 51, 218–251. Sobsey, M.D., Jaykus, L.A., 1991. Human enteric viruses and

Richards, G.P., Goldmintz, D., 1982. Rapid methods for extraction depuration of bivalve mollusks. In: Otwell, W.S., Rodrick,and concentration of poliovirus from oyster tissues. J. Virol. G.E., Martin, R.E. (Eds.), Molluscan Shellfish Depuration.Methods 5, 285–291. CRC Press, Boca Raton, FL, pp. 71–114.

Riordan, T., 1991. Norwalk-like viruses and winter vomiting Sockett, P.N., Cowden, J.M., LeBaigue, S., Ross, D., Adak, G.,disease. In: P. Morgan-Capner (Ed.), Current Topics in Clinical Evans, H., 1993. Foodborne disease surveillance in EnglandVirology, Public Health Laboratory Service, London, pp. 61– and Wales: 1989–1991. Comm. Dis. Rep. 3, 159–174.93. Sockett, P.N., West, P.A., Jacob, M., 1985. Shellfish and public

Rippey, S.R., 1994. Infectious diseases associated with molluscan health. PHLS Microbiol. Digest 2, 29–35.shellfish consumption. Clin. Microbiol. Rev. 7, 419–419. Solic, M., Krstulovic, N., 1992. Separate and combined effects of

Roderick, G.E., Schneider, K.R., 1994. Depuration and relaying of solar radiation, temperature, salinity, and pH on the survival ofmolluscan shellfish. In: Hackney, C.R., Pierson, M.D. (Eds.), faecal coliforms in seawater. Mar. Pollut. Bull. 24, 411–416.Environmental Indicators and Shellfish Safety. Chapman & Sorber, C.A., 1983. Removal of viruses from wastewater andHall, New York, pp. 331–363. effluent by treatment processes. In: Berg, G. (Ed.), Viral

Romalde, J.L., Estes, M.K., Szucs, G., Atmar, R.L., Woodley, Pollution of the Environment. CRC Press, Boca Raton, FL, pp.C.M., Metcalf, T.G., 1994. In situ detection of hepatitis A virus 39–52.in cell cultures and shellfish tissues. Appl. Environ. Microbiol. Speirs, J.I., Pontefract, R.D., Harwig, J., 1987. Methods for60, 1921–1926. recovering poliovirus and rotavirus from oysters. Appl. En-

Roos, R., 1956. Hepatitis epidemic conveyed by oysters. Sven viron. Microbiol. 53, 2666–2670.Lakartidningen 53, 989–1003. Stille, W.B., Kunkel, B., Nerger, K., 1972. Oyster-transmitted

Rose, J.B., Sobsey, M.D., 1993. Quantitative risk assessment for hepatitis. Dtsch. Med. Wochenschr. 97, 145–147.viral contamination of shellfish and coastal waters. J. Food Sugieda, M., Nagaoka, H., Kakishima, Y., Ohshita, T., Nakamura,Prot. 56, 1043–1050. S., Nakajima, S., 1998. Detection of Norwalk-like virus genes

Saito, K., Ushijima, H., Nishio, O. et al., 1995. Detection of in the caecum contents of pigs. Arch. Virol. 143, 1215–1221.astroviruses from stool samples in Japan using reverse tran- Sugieda, M., Nakajima, K., Nakajima, S., 1996. Outbreaks ofscription and polymerase chain reaction amplification. Mi- Norwalk-like virus-associated gastroenteritis traced to shellfishcrobiol. Immunol. 39, 825–828. – coexistence of two genotypes in one specimen. Epidemiol.

Salamina, G., D’Argenio, P., 1998. Shellfish consumption and Infect. 116, 339–346.awareness of risk of acquiring hepatitis A among Neapolitan Sullivan, R., Peeler, J.T., Tierney, J.T., Larkin, E.P., 1984.families – Italy, 1997. Eurosurveillance 3, 97–98. Evaluation of a method for recovering poliovirus 1 from 100

Schwab, K.J., Neill, F.H., Estes, M.K., Metcalf, T.G., Atmar, grain oyster samples. J. Food Prot. 47, 108–110.R.L., 1998. Distribution of Norwalk virus within shellfish Tang, Y.W., Wang, J.X., Xu, Z.Y., Guo, Y.F., Qian, W.H., Xu, J.X.,following bioaccumulation and subsequent depuration by de- 1991. A serologically confirmed case-control study of a largetection using RT–PCR. J. Food Prot. 61, 1674–1680. outbreak of hepatitis A in China associated with consumption

Sellwood, J., 1992. The virology of the water cycle. PHLS of clams. Epidemiol. Infect. 107, 651–658.Microbiol. Digest. 9, 72–75. Tartera, C., Jofre, J., 1987. Bacteriophages active against bac-

Sellwood, J., Dadswell, J.V., 1991. Human viruses and water. In: teroides-fragilis in sewage-polluted waters. Appl. Environ.Morgan-Capner, P. (Ed.), Current Topics in Clinical Virology, Microbiol. 53, 1632–1637.Public Health Laboratory Service, London, pp. 29–45. Taylor, M.B., Grabow, W.O.K., Cubitt, W.D., 1997. Propagation of

Sharp, T.W., Hyams, K.C., Watts, D. et al., 1995. Epidemiology of human astrovirus in the plc /prf /5 hepatoma cell line. J. Virol.Norwalk virus during an outbreak of acute gastroenteritis Methods 67, 13–18.aboard a USA aircraft carrier. J. Med. Virol. 45, 61–67. Tiemessen, C.T., Nel, M.J., 1996. Detection and typing of

Slomka, M.J., Appleton, H., 1998. Feline calicivirus as a model subgroup F adenoviruses using the polymerase chain reaction.system for heat inactivation studies of small round structured J. Virol. Methods 59, 73–82.viruses in shellfish. Epidemiol. Infect. 121, 401–407. Tompkins, D.S., Hudson, M.J., Smith, H.R. et al., 1999. A study

Smit, T.K., Steele, A.D., Peenze, I., Jiang, X., Estes, M.K., 1997. of infectious intestinal disease in England: microbiologicalStudy of Norwalk virus and Mexico virus infections at Ga- findings in cases and controls. Comm. Dis. Pub. Health 2,Rankuwa hospital, Ga-Rankuwa, South Africa. J. Clin. Mi- 108–113.crobiol. 35, 2381–2385. Torne, J., Miralles, R., Tomas, S., Saballs, P., 1988. Typhoid fever

Sobsey, M.D., Wallis, C., Melnick, J.L., 1975. Development of a and acute non-A non-B hepatitis after shellfish consumption.simple method for concentrating enteroviruses from oysters. Eur. J. Clin. Microbiol. Infect. Dis. 7, 581–582.Appl. Microbiol. 29, 21–26. Tree, J.A., Adams, M.R., Lees, D.N., 1997. Virus inactivation

Sobsey, M.D., 1987. Methods for recovering viruses from shell- during disinfection of wastewater by chlorination and UV


irradiation and the efficacy of F 1 bacteriophage as a viral 1990. Risk factors analysis of an epidemic of hepatitis A in aindicator. Water Sci. Technol. 35, 227–232. factory in Shanghai. Int. J. Epidemiol. 19, 435–438.

Truman, B.I., Madore, H.P., Menegus, M.A., Nitzkin, J.L., Dolin, Wang, J.X., Jiang, X., Madore, H.P. et al., 1994. SequenceR., 1987. Snow mountain agent gastroenteritis from clams. diversity of small, round-structured viruses in the NorwalkAm. J. Epidemiol. 126, 516–525. virus group. J. Virol. 68, 5982–5990.

Tsai, Y.L., Tran, B., Sangermano, L.R., Palmer, C.J., 1994. Willcocks, M.M., Carter, M.J., Laidler, F.R., Madeley, C.R.,Detection of poliovirus, hepatitis A virus, and rotavirus from 1990. Growth and characterization of human fecal astrovirus insewage and ocean water by triplex reverse transcriptase PCR. a continuous cell line. Arch. Virol. 113, 73–82.Appl. Environ. Microbiol. 60, 2400–2407. Williams, Jr. F.P., Fout, G.S., 1992. Contamination of shellfish by

Tsarev, S.A., Binn, L.N., Gomatos, P.J. et al., 1999. Phylogenetic stool-shed viruses methods of detection. Environ. Sci. Technol.analysis of hepatitis E virus isolates from Egypt. J. Med. Virol. 26, 689–696.57, 68–74. Williams, R.A., Zorn, D.J., 1997. Hazard analysis and critical

Turkoglu, S., Lazizi, Y., Meng, J.H. et al., 1996. Detection of control point systems applied to public health risks: thehepatitis E virus RNA in stools and serum by reverse tran- example of seafood. Rev. Sci. Tech. Off. Int. Epiz. 16, 349–scription-PCR. J. Clin. Microbiol. 34, 1568–1571. 358.

Van Cuyck-Gandre, H., Gratier, D., Burckhart, M.F., Crance, J.M., Xu, L., Harbour, D., McCrae, M.A., 1990. The application ofSchwartzbrod, L., 1994. Detection of hepatitis-A virus in polymerase chain reaction to the detection of rotaviruses inoysters. Int. J. Food Sci. Technol. 29, 185–193. faeces. J. Virol. Methods 27, 29–38.

Vantarakis, A.C., Papapetropoulou, M., 1998. Detection of en- Yamashita, T., Kobayashi, S., Sakae, K. et al., 1991. Isolation ofteroviruses and adenoviruses in coastal waters of SW Greece cytopathic small round viruses with BSC-l cells from patientsby nested polymerase chain reaction. Water Res. 32, 2365– with gastroenteritis. J. Infect. Dis. 164, 954–957.2372. Yang, F., Xu, X., 1993. A new method of RNA preparation for

Vaughn, J.M., Metcalf, T.G., 1975. Coliphages as indicators of detection of hepatitis A virus in environmental samples by theenteric viruses in shellfish and shellfish raising estuarine polymerase chain reaction. J. Virol Methods 43, 77–84.waters. Water Res. 9, 613–616. Yolken, R.H., Wilde, J.A., 1994. Assays for detecting human

Vinje, J., Koopmans, P.G., 1996. Molecular detection and epi- rotavirus. In: Kapikian, A.Z. (Ed.), Viral Infections of thedemiology of small round-structured viruses in outbreaks of Gastrointestinal Tract. Marcel Dekker, New York, pp. 251–gastroenteritis in the Netherlands. J. Infect. Dis. 174, 610–615. 276.

Vizzi, E., Ferraro, D., Cascio, A., Distefano, R., Arista, S., 1996. Yotsuyanagi, H., Koike, L., Yasuda, K. et al., 1996. ProlongedDetection of enteric adenoviruses 40 and 41 in stool specimens fecal excretion of hepatitis A virus in adult patients withby monoclonal antibody-based enzyme immunoassays. Res. hepatitis A as determined by polymerase chain reaction.Virol. 147, 333–339. Hepatology 24, 10–13.

Wadell, G., Allard, A., Johansson, M., Svensson, L., Uhnoo, I., Zhou, Y.-J., Estes, M.K., Jiang, X., Metcalf, T.G., 1991. Con-1994. Enteric adenoviruses. In: Kapikian, A.Z. (Ed.), Viral centration and detection of hepatitis A virus and rotavirus fromInfections of the Gastrointestinal Tract. Marcel Dekker, New shellfish by hybridization tests. Appl. Environ. Microbiol. 57,York, pp. 519–547. 2963–2968.

Wang, J.Y., Hu, S.L., Liu, H.Y., Hong, Y.L., Cao, S.Z., Wu, L.F.,

Viruses and bivalve shellfish

Documents

Transcript of Viruses and bivalve shellfish