Enteric viruses in the aquatic environment

Introduction, 945 2. The matrix, 946 3. Enteric viruses in water, 946 3.1 Enteroviruses, 946 3.2 Norwalk-like viruses (NLV), 947 3.3 Astroviruses, 947 3.4 Rotavirus, 948 3.5 Enteric adenoviruses, 948 3.6 Hepatitis A and E viruses, 948 4. Concentration of enteric viruses, 948 4.1 Adsorption/elution, 950 4.1.1 Gauze pads, 950 4.1.2 Electronegative membranes and cartridges, 950 4.1.3 Electropositive cartridges and membranes, 951 4.1.4 Concentration using glass powder, 952 4.1.5 Concentration using glass wool, 952 4.1.6 Secondary concentration, 952 4.2 Entrapment, 952 4.2.1 Ultrafiltration, 953 4.2.2 Hydroextraction, 953 4.3 Ultracentrifugation, 953 4.4 Other methods, 954 4.4.1 Two-phase separation, 954 4.4.2 Immunoaffinity columns and magnetic beads, 954 5. Detection, enumeration and identification, 954 5.1 Detection by cell culture, 954 5.1.1 Liquid culture assay, 955 5.1.2 Plaque assays, 955 5.2 Immunological detection, 955 5.3 Detection by molecular biology, 955 6. Summary, 956 7. Issues that must be addressed, 957 8. Acknowledgement, 957 9. References, 957 Water quality, and therefore human health, may be significantly affected by the presence of pathogenic enteric microorganisms derived from sewage discharged to surface waters. Knowledge of the role of water in the ecology and transmission of these agents in the environment varies; the part played by enteric viruses is less well understood than that of many bacteria and protozoa, mainly due to the difficulties associated with detecting the agents in the aquatic environment. Outbreaks of enteric virus disease have been linked to water at various times and to different causes. Pipeline failures account for most incidents, such as those at Bramham, UK, in 1980 when 3000 individuals were affected (135). Waterborne disease outbreaks with other origins have been reported from many countries; consumption of contaminated ice, borehole or well pollution (78; 11; 28), and contamination of a municipal supply with sewage (71) are typical documented causes. Viruses in the water supply may also cause disease by foodborne transmission through individuals involved with food preparation becoming contaminated (18). The list of potentially pathogenic microorganisms transmitted by water increases every year (1). Waterborne disease may be transmitted by consumption of polluted drinking water, by immersion in recreational water or by contact through skin or inhalation. The health effects caused by microorganisms have been reviewed extensively; 45 and 142 documented drinking water outbreaks in the UK and 63, 22 and 117 reviewed disease associated with recreational water contact. Though explosive outbreaks may be well documented epidemiologically, with the exception of hepatitis A virus (HAV), disease transmission through ingesting water has been conclusively demonstrated in very few cases. Virus presence in water is underestimated because those agents which cause most enteric disease cannot be grown by normal culture methods. The driving force for obtaining more information on viruses in the aquatic environment is thus one of a need for increased knowledge contributing to better public health. Greater knowledge of the behaviour of these agents in water will facilitate the provision of appropriate advice to the public and responsible authorities on use and treatment of water. The 9 and the 32 are EU instruments which aim to ensure that water is fit for purpose. Compliance with the parameters of the Directives implies that water is monitored regularly and that no indicator organisms are found above the set limits. However, technological advances now permit the direct detection of many enteric pathogens. Accumulation of information also permits better sampling strategies to be devised, and more reliable data to be obtained. This in turn allows more confident risk assessment to be done and the development of models for predicting the risks associated with use of different kinds of water (87). Analysis of water for enteric viruses is usually a two-stage process. Viruses are too few to be detected directly so the sample is processed to concentrate them into smaller volumes, usually less than 10 ml. A range of detection methods is then available, and if used in a quantitative fashion any virus present may be enumerated. Where necessary, virus may be isolated and identified. This review will focus on the methods for both concentration and detection of enteric virus from water and similar materials. It should be noted that, despite repeated efforts to improve recovery rates for many enteric viruses from water, there is no method which gives consistently high recovery. Indeed, improvements in the detection of enteric viruses, driven by clinical diagnostic needs, have not been matched by advances in concentration techniques to provide reproducible high recovery for different virus types. The aquatic environment comprises many water types (matrices), each characterized by its concentration of organic matter and salts. Good quality drinking water, by definition, contains few impurities and no pathogens. Other matrices may contain human viruses if they contain human faecal material; untreated sewage will contain the highest quantity. Treated sewage effluent contains a variable number of viruses depending on the season, the viruses circulating in the population at any one time, and the degree of virus removal during sewage treatment. When sewage effluent is discharged into fresh or marine water, human viruses may therefore be present. Methods to concentrate viruses in sewage and other waters are well established and discussed below. However, methods for processing the highly variable and potentially toxic sewage sludge or biosolids are not yet optimized or of proven efficiency. In view of the increased land disposal of sludge, much current research is directed at improving recovery methods for this matrix (95; 147). Many groups of viruses may inhabit the human intestine and replicate in the epithelial cells which cover the villi. Only some cause sufficient local damage for gastroenteritis or other illness to result. The term 'enteric viruses' encompasses all those groups of viruses which may be present in the gastrointestinal tract; they may cause disease, or infection may be asymptomatic. Being shed in the faeces they will be present in sewage, but as they are obligate intracellular parasites they can not multiply in the environment. They may be divided into two groups according to their growth in cell culture. Most enteroviruses, which form a distinct taxonomic group (including poliovirus, Coxsackievirus and echovirus), grow well in readily available cell cultures of primate origin and are thus well characterized. Other groups that may be grown in cell culture such as some adenovirus serotypes and reoviruses do not cause recognizable gastrointestinal disease. The remaining enteric viruses form a heterogeneous group which includes the rotaviruses, astroviruses, adenovirus 40/41, human caliciviruses (Norwalk-like viruses and Sapporo-like viruses) and hepatitis A and E. These cannot be grown easily, if at all, in culture. Their importance lies in their causing distinct gastrointestinal disease or hepatitis. The important viruses in the context of water pollution are discussed below. Enteroviruses are Picornaviridae and are naked icosahedral particles approximately 27 nm in diameter with no obvious surface structure. It is of relevance to their environmental survival that the virions are stable within the pH range 3–10, and sensitive to chlorine and UV radiation (113). They are classified by biomolecular and biochemical characterization, and serological typing which also relates to the species in which they are found. Enteroviruses are present in the gastrointestinal tract of clinically healthy individuals, either as vaccine poliovirus or as nonpolio enteroviruses, usually of low pathogenicity. Although most infections are asymptomatic, enteroviruses may cause a variety of symptoms, including paralysis, meningitis and cardiomyopathy, and less severe conditions such as colds and fever. They are not associated with gastroenteritis in the absence of wider systemic illness but they are nevertheless a potential public health hazard. More information is available on the occurrence of enteroviruses in the environment than any other virus group because they have been the most readily detectable in the laboratory for many years. They can be concentrated from river water or seawater and can be grown and identified in standard cell culture assays. The numbers shed in faeces are not as high as bacterial indicators; the number of enteroviruses in untreated sewage may reach 105 plaque-forming units (pfu) l–1 whereas the count of total coliforms is often 108/100 ml. Neither are viruses shed consistently throughout the year. Isolations of Coxsackieviruses from river water, for example, showed a seasonal trend which peaked in mid-summer and which matched reports to clinical laboratories (131). The majority of studies on enteric viruses in water and associated materials took place between 1975 and 1985 in the UK and the US, and most involved enteroviruses. The consensus of findings was and remains that in the case of polioviruses, serotype 2 predominates, type 3 being the least numerous across all water types (109; 97). Wild-type strains continue to be identified (115; 31) although in the UK and more recently in other parts of the world only vaccine strains have been reported (133). Coxsackievirus B serotypes 1–6 have all been detected in water, although B3, B4 and B5 are the most frequently identified (references as for poliovirus, and 60). The predominant serotype has been shown to change over time and to reflect the serotype most frequently found in clinical material. Echoviruses have been least frequently identified (97) unless liquid culture was used when the combined serotypes of echoviruses were the most numerous strain (132). Geographically the presence of enteroviruses in water has been shown to follow a similar pattern in temperate countries across the world. In addition to surveillance and determination of water quality, data has been used in attempts to assess health risks associated with use of water for drinking or recreational purposes (e.g. 49). Human caliciviruses are naked single-stranded positive-sense RNA viruses with a single capsid protein and, following sequencing of the genome, are classified into two groups (67; 77). Norwalk-like viruses (NLVs, formerly termed SRSVs in the UK) are divided into genogroups I and II (116; 5). The Sapporo-like viruses (SLV), formerly termed 'classic caliciviruses', are also currently divided into two genogroups (13; 59; 152). Norwalk-like viruses are the principal cause of viral gastroenteritis in adults and were initially defined by a ragged particle outline and poorly defined surface morphology (41). Up to six antigenic types have been described by immune electron microscopy (81). Multiple genotypes and strains are recognized within the two genogroups and are identified by the geographical area where they were initially identified. Norwalk, Southampton and Desert Shield viruses are classified as genogroup I. Lordsdale virus, Mexico virus and Snow Mountain agent are classified within genogroup II (103). Being shed in faeces, NLVs will be transported through sewerage systems and may reach receiving waters. The agent is robust and will therefore survive in the environment, where it may present a hazard to public health, as evidence of shellfish-associated gastroenteritis testifies (e.g. 80). Waterborne outbreaks of NLV infection due to sewage-contaminated drinking water are well documented (e.g. 78; 75, 76), but reports of illness associated with recreational contact are less common and are based on epidemiological or circumstantial evidence (55). Sapporo-like viruses ('classic' human caliciviruses) cause gastroenteritis, mainly in babies under 1 year old, during which time they are excreted in the faeces (152). Morphologically they display a well-defined surface structure. Since young children are commonly infected and adults are immune, their public health significance in relation to potential for waterborne disease remains uncertain. Astroviruses (AV), the only genus in the Astroviridae, are 28 nm diameter particles which often bear a five- or six-pointed stellate surface structure and are associated with sporadic cases and occasionally outbreaks of diarrhoea, principally in children, though adults may also be affected (158). They do not grow in commonly used cell cultures, though propagation has been reported in cell lines derived from intestinal tumours (e.g. 157; 160). Few reports exist on the incidence of astroviruses in the aquatic environment. Detection of astrovirus in environmental samples has been described by 92, and 112 demonstrated the occurrence of infectious astrovirus in water from an area where a concurrent gastroenteritis outbreak was reported. 2 studied the survival of astroviruses in different water types. The survival characteristics found in these studies were comparable to those determined for rotavirus and enteric adenoviruses, though astrovirus decay was more pronounced at the higher temperatures. Rotaviruses are the major viral cause of infantile gastroenteritis. They are 75 nm diameter icosahedral virus particles with a triple capsid protein layer (38); the virion consists of a core and a shell of intermediate and outer layers from which 60 spikes protrude. The genome consists of 11 segments of double-stranded RNA (dsRNA) that are capable of genetic reassortment. The complete, infectious particle has a smooth appearance by EM which becomes a rough outline when the outer shell is lost. Faeces contains many more rough noninfectious particles, which are environmentally robust, than complete infectious ones. Rotaviruses are classified on the basis of serogroups, serotypes and, most recently, genogroups. Six serogroups, termed Groups A–F (plus a possible seventh, G), are recognized and between each of these groups no cross-reaction occurs using assays based on the inner capsid antigen, VP6. Group A is the most common human rotavirus. Serotypes are based on neutralization assays for outer capsid proteins VP7 and VP4. Genogroups are based on gene 4 (P-types, VP4) and gene 9 (G-types, VP7). Strains circulating in the community can thus be monitored (65). Limited numbers of human or monkey kidney cell culture types, such as CaCo-2 and MA-104 will support rotavirus replication. Limited numbers of reports of waterborne transmission have been published and those that have are as a result of sewage contaminated drinking water (reviewed by 49). Enteric adenoviruses (types 40 and 41) are important aetiological agents of paediatric gastroenteritis, principally in temperate climates (146). Their role in transmission of disease in water is undetermined. 36 found that they were more stable in water than poliovirus 1 or hepatitis A virus. 47 first reported detection of the genome in a high proportion of water samples in South Africa, and similar results have been found in other countries. Other adenovirus serotypes are found in sewage but they are not associated with gastrointestinal disease (132). Because of its well-defined association with waterborne and, particularly, shellfish-associated disease (88), there is much information on the role of the environment in the transmission of HAV, and though it is relatively uncommon in countries which have a good standard of hygiene (89), raised numbers of cases are still reported every 8–10 years. It is a picornavirus and the only member of the Hepatovirus genus. One serotype is recognized, with strain differences. Humans are the only natural host. HAV is relatively resistant to low pH and to heating, so is environmentally robust. Outbreaks have occurred in many residential settings, especially schools where hygiene is poor. Outbreaks in the US have been associated with drinking water drawn from wells (14) and after contact with recreational water (90). Contamination of borehole water in Ireland in 1991 (42) resulted in a major outbreak of gastroenteritis but a follow-up study identified no cases of hepatitis. HAV has been detected in well water (28), sewage (70) and in sewage sludge (54). Hepatitis E virus is a member of the Caliciviridae and is a 30-nm iscosahedral particle without obvious structure by electron microscopy. Hepatitis E virus is not endemic in the UK but cases are imported from Asia, China and Africa. The infection is similar to HAV hepatitis but can be more severe, especially in pregnant women. Large scale waterborne outbreaks have been reported (69). Virus particles exhibit polarity and can adsorb to a wide variety of charged matrices which may be immobilized (such as membranes) or fluid (such as glass powder). Considered as protein, virus particles have a high relative molecular mass (Mr > 106) and lend themselves to concentration by ultrafiltration and by ultracentrifugation. Based on these general properties, numerous methods have been devised for the concentration of viruses from aquatic matrices. To be of optimum practical use, any method must fulfil the following criteria (after 15): (i) be technically easy to accomplish in a short time; (ii) have a high virus recovery rate; (iii) concentrate a large range of viruses; (iv) provide a small volume of concentrate; (v) not be costly; (vi) be capable of processing large volumes of water; (vii) be repeatable (within a laboratory) and be reproducible (between laboratories). There is no single method which fulfils all these requirements. There are four principal approaches to virus concentration based on different properties of virus particles; they have been in place for many years, and each technique has numerous variations (Table 1). Most procedures can be used to concentrate viruses in sample volumes of 1–100 l, and some (mainly those based on cartridges) are suitable for volumes of 1000 l or more. For immobilized filter matrices the maximum volume that can be filtered will be dependent upon the turbidity of the sample. Groundwater and drinking water will contain very few viruses and l or more will need to be recreational and river waters may contain many more viruses so processing 10 l samples is effluent that has treatment will have higher of virus so one samples are Viruses in sewage may be detectable directly or samples may be and some techniques the first which the volume to between and in a two-stage concentration process. The the virus usually by and to the which is in The development of virus methods from the of and in (e.g. In the sample is into contact with a matrix to which virus will adsorb under conditions of pH and virus is the water in which it was is Virus is then from the matrix into a smaller though this is still too large to be of matrix, fluid and processing conditions will be by the of the sample and by of virus from the matrix is commonly done using one of at pH at pH or at pH In EU or in is often the virus is concentrated in the by and the to which the virus is concentration method has and and it is important to that there is no single method which fulfils all the above The is to virus recovery virus due to the concentration enteroviruses are robust to following to pH and pH this is not the case for other rotaviruses, where is pH The principal variations of the are as particles as water through of virus is very They cannot be used in waters low in and they are therefore to sewage and for which they were (e.g. Concentration of viruses in water using charged has been for many and there are many variations of the though change in the process. The of membranes lies in their in various and For waters a is used of the The of filter will on the for of diameter and are commonly and are of and it is possible to good of virus from waters by good rates and a of filter In many virus can be Virus is to the filter by and is using or This method is currently the of viruses from sewage and sludge as well as recreational and surface waters in most water laboratories in the and is the UK method It is also a for the recovery of viruses from waters and as published by the The concentrate volume is a practical for cell culture but is a large volume for molecular Enteroviruses and NLVs can be concentrated by this the few reports of this to are and The water sample is to facilitate of virus to filter matrix by to pH and addition of though is divided as to are when using charged have also been used in They were used for concentration of viruses from and have been for concentration of viruses from river water (e.g. and other waters These also need the sample to be for virus particles to their at least as good as they are less and can be complete in They to and cannot be used with water according to though the as used by and no in this Because of of of or the processing of seawater samples in this is to a maximum of l have to be nevertheless they for processing of 10 l EU Water were by The increased surface area for higher and were better than by they are though they may be to times by in Many to the have been but have on or other recovery rates are as variable with charged filter as with any other membranes relatively as they recovery of the a on but recovery with river water. using reported recovery of in to 1000 l Few studies have been done on recovery from marine though they have been done to the of the method using drinking water. reported in the of virus following processing of l water samples Though the recovery was from to in two in laboratories and in one were higher and less variable where a higher of virus was but it must be noted that the of is more than is found in water charged adsorb virus from water and other materials without the need for of the sample with or They adsorb virus in the pH range though at pH above the so the pH still to be These properties the use of charged not only for the of not to the sample but also because it possible the concentration of viruses such as rotavirus and which are sensitive to the low pH conditions for to charged reported that type 1 poliovirus and rotavirus at least on at so this may them for or transmission through Other than not to the water are used in the as materials. Virus is from the filter and concentration is as for the types. from charged are similar to those from charged reported recovery using a two-stage in the concentration of poliovirus from drinking water, and enteroviruses and HAV from river water and sewage using 112 infectious astrovirus using the using cartridges, detected infectious and adenovirus type 40 when surface water samples were processed and by cell culture The charged is of a these with and for recovery of poliovirus from 1000 l and of with both types. Though especially in these have an in the large volumes they can without the need for the sample. from the filter still to be at pH 9 or which their use to viruses stable that and concentrated using charged materials in the especially have been reported to more with virus recovery from cartridges than from materials A different was which was but were reported to be not as good as in charged filter materials are in the US but are not so used in in have resulted in membranes being available for concentration of viruses from water. used charged membranes with ultrafiltration for the concentration of a variety of enteric viruses to detection by membranes are in and a surface over the pH range 3–10, which of charged virus filter is also in which the of Few studies have been done on this in of viruses from and its low and of use that research should be membranes and cartridges have been used in for the removal of and viruses from though the virus can be from the filter is not glass of diameter form a good for viruses under conditions similar to those used for They form a and so have the that the filter matrix cannot as with first this which was by For low sample volumes the method a low volume which not need concentration to into cell culture. However, in common with the for the recovery with the type of from with water to with using