Viruses in Water: The Imaginative In Pursuit of the Fugitive
Water-borne enteric viruses are probably not the first microorganisms which spring to mind when thinking of polluted water. Cholera, typhoid and cryptosporidiosis are more prominent in the public mind, though viruses are likely to have been the cause of many outbreaks of water-borne disease. The difficulty has, until comparatively recently, been proving the link between the water and the sick person.
The aquatic environment contains an ever-changing kaleidoscope of microorganisms, including those present as a result of pollution by sewage and agricultural run-off. The threat of cholera and typhoid as the principal diseases transmitted by water in the cities of the 19th and early 20th centuries, and eventually the relative ease of detecting salmonellae and vibrios in water all eclipsed the idea that other infections, prevalent in the communities but not well characterized, might also be water-borne. In the first half of the 20th century epidemics of poliomyelitis were common in Europe and the US and many were associated, albeit anecdotally, with ingestion of polluted water. The most notable victim was President Franklin Roosevelt, who developed symptoms after falling in the water during a boating holiday. Though it is now thought that the disease might not have been polio, Roosevelt supported campaigns for research into the disease, which culminated in the Salk and Sabin vaccines.
Today, pollution of bathing waters by sewage produces a perceived public health problem internationally. Many countries depend on tourism for much of their national income and a large part of their GDP is derived from seasonal tourism around their coastlines. Pressure group activity and public health concerns about recreational water quality have driven governments, water undertakings and the European Union to continue improvements in bathing water quality begun in the 1970s with the first Bathing Water Directive, and there are at present nearly 28,000 bathing waters covered by the legislation.
Viruses are more environmentally stable than bacteria and so will persist in waters where only low levels of bacteria may be found. Studies done in the 1990s suggested that any gastrointestinal symptoms experienced by bathers following immersion in water were due to viral, rather than bacterial, infection.
Everyone carries their own population of enteric viruses. Like enteric bacteria, they generally do no harm and are shed in the faeces. Since they are very common and therefore shed by most of the population they will be found in any water polluted by domestic sewage. In addition to those harmless viruses, some enteric viruses cause disease, usually gastroenteritis. They also will be shed by those unfortunate enough to catch them, but for only short periods. They will therefore be present less often in sewage, and less often in polluted water. However, they may be just as robust as the less harmful types and so will survive for extended periods of time.
There are many groups of enteric viruses. The enteroviruses and adenoviruses are common; most cause few symptoms in healthy adults, and are often present in sewage-polluted water. Conversely, hepatitis A virus and norovirus are examples of pathogenic viruses which are shed by infected individuals and which are usually associated only with illness. Study of these viruses in the local sewage thus gives us a picture of the virological state of the population served by the local treatment plants. Recognition of the occurrence of enteric viruses in the environment is therefore of use in epidemiological surveys where exposure to microbial agents is considered. The microbiological quality of bathing waters continues to be measured in terms of bacteria, E.coli and enterococci in the new Bathing Water Directive, and methods for measuring the levels of these organisms are straightforward. However, few people become ill through ingestion of waterborne enterococci and there remains a case for the detection of (a) pathogens rather than indicators and (b) viral indicators which more closely reflect the pathogenic viruses in polluted water.
The association of the presence of the causative agent in water with its presence in an affected individual has been one of the main stumbling blocks in proving that water-borne viruses cause disease in recreational waters or in drinking water. There have been a number of outbreaks which illustrate the progress made in the last few years in solving this problem.
Drinking water. Bramham is a village in North Yorkshire, UK. In 1980 about 3,000 of its 12,000 population became ill with diarrhea and vomiting. Investigations showed that sewage had leaked from a cracked pipe and contaminated the local drinking water supply at the same time as a chlorination failure occurred. In retrospect, the epidemiological picture strongly suggests a norovirus outbreak, but at the time no virology was done on stool samples, nor on water nor sewage, so the exact cause remained unsolved.
In Riding Mill, Northumberland, UK, in 1990, 229 people were made ill by the drinking water supply being inadvertently connected to the untreated (and polluted) river water. Again, the epidemiology suggested a viral cause, but no environmental investigations were done.
By the mid-1990s it was possible to detect water-borne enteropathogenic viruses much more reliably. In Heinavesi, Finland, in 1998, almost 3,000 people became ill through their drinking water supply becoming polluted with sewage. Finland is a land of over 18,000 lakes and many small communities exist on their shores. Water fl owing from one lake to another has historically been regarded as sufficient dilution for the relatively small quantities of sewage allowed to enter the water, and chlorination was not widely practiced until recently. But the 1998 outbreak showed that consumption of polluted water could lead to viral diseases, and cause and effect were confirmed by finding the same virus (norovirus genogroup II) in the drinking water and in some of the affected individuals. Recreational waters.
Recreational waters have also been implicated in norovirus outbreaks; in Vermont, in 2004, 53 people (mostly children) became ill after swimming club activities at a local pool; although the virus was not recovered from the water, the children all had the same strain of virus. The cause was a faulty chlorination unit coupled with poor management practices. In the Netherlands, in 2002, 90 children had diarrhoea and vomiting after playing in a recreational water fountain. No E.coli or enterococci were found in drinking water samples at the playground site, but samples from the fountain had bacterial counts that exceeded the EU limits, and the same norovirus was found in the water fountain samples as in faecal samples taken from the affected children.
Detection of water-borne viruses is technically much more demanding than finding bacteria. Viruses in faeces are normally present in lower concentrations than bacteria, even in disease states. The detection process consists of at least two stages: concentration of virus in the water sample and detection of viruses in the concentrate. The sample volume will vary according to the likelihood of finding virus, so for drinking water it is common to concentrate 100”“1,000 liters, while for potentially polluted river water or treated sewage effluent 10 liters will be sufficient.
There are many methods for concentration of viruses in water; it may be done by filtration through membranes which retain virus by electrostatic adsorption, through resinbased membranes, or through other filtration matrices. Glass wool, packed into a column, is a very cheap and effective filtration matrix. Developed in France in connection with drinking water analysis, the technique has found widespread use in concentration of many virus types from a range of waters. Virus is eluted from filters into a small volume (usually 100”“ 200 ml) of high pH buffer, sometimes containing skimmed milk or beef extract, and is then further concentrated to between 1 and 10 ml, depending on the detection procedure.
Other concentration methods include ultrafiltration and ultracentrifugation. The former is useful for pH-intolerant viruses and is currently being trialled as a way of concentrating avian orthomyxoviruses in water, even though these viruses may be more pH-resistant than their human counterparts. Ultracentrifugation is a catch-all method and has revealed the presence of viruses in relatively small volumes of water. It is not a technique likely to find widespread use in routine environmental monitoring laboratories however, owing to the high cost of the equipment.
Most virus detection is done by molecular biological methods, principally using the reverse transcriptase polymerase chain reaction (RT-PCR), though nucleic acid sequence-based amplifi cation (NASBA) and loop-mediated isothermal amplifi cation (LAMP) techniques are also used. Cell culture, used to detect water-borne enteroviruses since the 1970s, is used less since it is restricted to very few virus types and takes a long time to produce results. Whichever approach is used however, the main problem is one of inhibitors in the concentrate. These are usually soil-derived humic or fulvic acids which inhibit the RT-PCR (usually the RT) and which are difficult to remove. Their presence may therefore lead to a false-negative result, and suitable amplification controls need to be used to guard against this. Substances toxic to cell cultures may be removed by treating the concentrate with chelating agents in chloroform, but this may have an adverse effect on the virus too.
Viruses and water quality
Methodology has now advanced sufficiently to make routine virus detection in water a practicable proposition, and it has moved from the possible to the feasible in respect of establishing a virus standard for use in regulatory contexts. The microbiological parameters were reduced to E.coli and enterococci in the 2006 revision of the Bathing Water Directive and the enterovirus standard of the 1976 Directive was removed, which was scientifically sensible since it had been based on precarious assumptions. The new Directive contains Article (14) which requires the European Commission to report by 2008 on (interalia) scientific progress relating to viruses in water. In pursuit of this task the Commission funded the FP6 Project Virobathe (www.virobathe.org) which involved 16 laboratories across Europe and which has just finished. The aim was to devise a routine method for detecting adenoviruses and noroviruses in recreational waters. The study compared different approaches and produced protocols for use in routine environmental virology laboratories. It clearly demonstrated that routine virus analysis is a feasible proposition. It is therefore possible to begin the task of linking enterococci levels (which are the only reliable indicators of health effects in bathing waters) to virus levels (since viruses are the most likely causes of disease acquired from ingesting polluted water while bathing) and so to increase confidence in monitoring programmes designed to improve water quality and protect the public health.
Peter Wyn-Jones Senior Research Fellow, IGES, University of Wales Aberystwyth, Aberystwyth SY23 2DB, UK (e firstname.lastname@example.org)
Percival, S.L., Chalmers, R.M.,
Embrey, M., Hunter, P.R., Sellwood,
J. & Wyn-Jones, P. (2004). Microbiology
of Waterborne Diseases, Elsevier AcademicPress.