key: cord-0041394-90g0461f
authors: Hurst, Christon J.; Benton, William H.; Stetler, Ronald E.
title: Detecting Viruses in Water
date: 1989-09-01
journal: J Am Water Works Assoc
DOI: 10.1002/j.1551-8833.1989.tb03273.x
sha: a9fbce0bceadaeed380c3c7a5c0490ba9a8587a4
doc_id: 41394
cord_uid: 90g0461f

Various and divergent approaches that have been used to concentrate and assay viruses from tap water and environmental freshwaters are summarized and briefly explained. The basic principles behind the different methodologies and descriptions of the most recent developments are emphasized. Comparisons help demonstrate the relative sensitivities of different concentration and assay techniques.

Viruses that replicate within and are subsequently shed from the gastrointestinal tract of animals are referred to as enteric viruses. These viruses are principally transmitted by the fecal-oral route, meaning that they cause illness after being ingested by an uninfected individual. The known human enteric viruses are listed in Table 1 . Not all of these viruses, however, have been studied with regard to their presence in drinking water. Although some of the viruses listed in Table 1 can also replicate in animals other than humans, their presence in raw waters is presumably related to human fecal pollution.

The human enteric viruses are presumed to present the greatest concern with regard to waterborne viral disease. Although raw waters may also contain related enteric viruses originating from wild and domesticated animals, most of these viruses are considered to have relatively limited pathogenicity for humans. Infectious human enteric viruses can be found in surface freshwaters and groundwatersr-4where they can originate from septic effluents and wastewater.5-R Viral contamination of surface water could also result from human recreational activities either in or around bodies of water." Enteric viruses can survive improperly operated conventional water treatment trains and may then be found in potable water.1-4J.i".rr This cause for public health concern has resulted in the development of many methods for recovering both human enteric viruses and bacteriophages from water. Bacteriophages (viruses that infect bacteria) have been studied in the hope that they may prove useful as indicators of the removal and SEPTEMBER 1989 destruction of human enteric viruses during water treatment.

The remainder of this article describes and categorizes the various Viruses were concentrated from river water by directed adsorption onto a pleated membrane cartridge filter.

techniques that have been used for detecting viruses in environmental freshwaters and in treated drinking waters. These techniques fall into two main categories. The first is virus concentration techniques, in which viruses recovered from large volumes of water are concentrated into smaller, more manageable volumes; the second is the viral assay procedures used to examine the concentrated samples.

Classification of techniques. Five major categories of techniques are used for concentrating viruses contained in water samples (Table 2) . Before examining thesedifferent concentration techniques, it is helpful to briefly mention the structure of what are termed classical viruses, a category that encompasses all of theviruses listed in Table 1 . Classical viruses consist of a nucleic acid core, or genome, which provides their genetic information and is surrounded by a shell of proteins termed a capsid. In some types of viruses this capsid is, in turn, surrounded by a lipid bilayer membrane, or envelope. Of the viruses listed in Table 1 , only coronaviruses possess such envelopes. Nonenveloped viruses essentially present themselves to the environment as large protein structures. For this reason, studies like Mizutani's,"" which addresses the adsorption of pro teins to glass surfaces, help explain the mechanics of virus concentration techniques. The only study found that addressed concentrating enveloped viruses from water samples used the herpes simplex virus and the measles virus,"" neither of which is considered enteric or otherwise of concern. with regard to the waterborne transmission of human illness.

Passiveadsorption.

The first type of concentration technique mentioned in Table 2 is passive adsorption. This technique, originated by MacCallum et aLs7 involves the use of pads consisting of either gauze alone or a combination of gauze and cotton batting for the purpose of absorbing a portion of the water sample into which the pads are placed. A small portion of the viruses suspended in the water sample are taken into these pads during the passive absorption process. Any actual adsorption of viruses onto the surface of the gauze or cotton fibers is incidental. The pads may be used in either a stationary or flow-through configuration.r"."i Viruses are recovered from the pads by expression of the absorbed fluid. Additional fluids may be used in kneading the pads to help achieve recovery of any adsorbed viruses. Fluids used for desorbing viruses from surfaces are termed eluants, and the process is termed elution. Following exposure to an adsorbing surface, it is hoped that the fluid, or eluate, will contain any viruses that had been on the surface. The relatively low efficiency of gauze pads for concentrating viruses from water was demonstrated, however, by a study in which such pads were used for prefiltering a water sample to remove large particulate solids, before concentrating the viruses from the same water sample by passage through a layer of virus-adsorbing granular material.:{' Directed adsorption involves the use of either filters or granular solids as surfaces onto which viruses are deliberately adsorbed and then ~~~~~~~~~~l~.~~.27.~0.~4.~~7.44,4~,~~-~~ Ad. sorption is generally facilitated by modifying the water sample, the adsorbent material, or both. The adsorbent material can be modified either duringor after its manufacture. Subsequent recovery of the adsorbed viruses can occur either by dissolving the filter material, if it is composed of alginate,"" or by exposing the adsorbent to a volume of eluant that is smaller than the original water sample and facilitates a reversal of the virus adsorption process. Most techniques now used for concentrating viruses from large volumes of either tap water or environmental freshwaters belong to the category of directed adsorption. These techniques are discussed in greater detail later in this article. The general mechanisms involved in concentration methods of this type have been presented in publications by Gartner, "' England, "" Cookson, fifi Mix, 67 and Sweet et Viruses are adsorbed onto the surface of filter materials or granular solids, facilitated in many instances by pretreatment of the water. Processes evaluated as pretreatments include removing soluble organics from the water using charged resins, adjustment of the water sample pH, and addition of salts of metal cations-particularly those that are multivalent such as aluminum and magnesium. Efforts have also been made to pretreat the adsorbent material itself to enhance the efficiency of viral adsorption. Examples of the latter approach include binding metal precipitates or charged polymers to the matrix of microporous filters. Subsequent recovery of the bound viruses from the filter materials or granular solids used as adsorbents is achieved by exposing the adsorbents to an eluant. Viruses are retained in the original water sample during a reduction of its volume achieved by pore size exclusion. This generally is accomplished by either recirculating the water under pressure through hollow fiber filters or over the surface of flat sheet filters in either recirculating or nonrecirculating systems. In some cases an eluant is subsequently passed through the same concentration unit to facilitate virus recovery. Viruses are concentrated either by association with a generated precipitate or by selective partitioning of the water sample during processes that yield a polymeric phase separation, or they are retajned in the original water sample during a reduction in its volume through hydroextraction.

Ultrafiltration.

Ultrafiltration involves selectively removing small molecules from a water sample, including the water molecules themselves, and leaving behind large solutes or suspended particles such as the viruses. This technique uses filters that may be configured either as hollow fibers through which the water sample is passed or sheet filters against which the water is kept in motion by means of a recirculating pump or stirring apparatus.4".4X.b9 The general mechanism used in this type of concentration technique has been discussed by Mix.fi7 The filters may be pretreated before they are used on a water sample to suppress virus adsorption.@ Alternatively, a second fluid, that either may or may not serve as an eluant, can be used to flush viruses from the surface of the filters after they have been used to concentrate a water sample.46 Ultrafiltration is seldom used for directly concentrating viruses from water samples *Secondary concentration is also frequently reterred to as second-step concentration or reconcentration. +May be supplemented with materials such as specific amino acids: other protein-derived products have ken evaluated as substitutes for beef extract (references for the latter include 76 and 103) $These types of rluants may include compounds such as alanine. EDTA. glycine. imidazole. lysine. and polysorbate 80.

." X.0 8.5 3.5 7.0 8.0 8. Several techniques fit into the category of direct physicochemical treatment for concentrating viruses from water samples. One general approach is to concentrate viruses through physical association with a flocculant precipitate that is induced by chemical supplementation of the water sample. The types of precipitates produced include aluminum hydroxide, aluminum phosphate, and calcium phosphate.53.65 Alternatively, proteinaceous materials may be added to the water sample and then precipitated by means of processes such as adding protamine sulfate50or lowering the PH.~O Viruses in the water are concentrated through association with these proteinaceous precipitates.

Polyethylene glycol has been used in two different ways for concentrating viruses from water samples. The first of these, two-phase separation, consists of adding sodium chloride, dextran sulfate, and polyethylene glycol to the water. Ionic strength contributed by the sodium chloride causes polymeric incompatability, resulting in the formation of two distinct aqueous phases, one containing most of the dextran sulfate and the other containing most of the polyethylene glycol. The densities of these two phases differ sufficiently that they can then separate gravimetrically. Viruses in the water sample undergo a partitioning during the generation and separation of these two phases and are largely concentrated into only one of them.5',54.65.@The second technique involves placing the water sample into a dialysis bag and packing the whole unit in solid polyethylene glycol. The polymer then concentrates the sample by drawing water out of the dialysis bag, a process called hydroextraction.51.52

Likeultrafiltration, thedifferent physicochemical concentration techniques are used seldom, if at all, for directly concentrating viruses from water samples much larger than 20 L. The reason is that the time expended processing water samples by these techniques increases dramatically with sample size. Both techniques are often used, however, in a secondary mode to reduce the volume of eluates generated during the processing of water samples by directed adsorption. This process is variously referred to as secondary concentration, second-step concentration, or reconcentration. Affinity chromatography.

Affinity chromatography consists of passing an aqueous solution through a column of polysaccharide gel material bearing highly specific, purified antibodies that can selectively bind viruses or some other target antigen from the fluid. This binding process is reversible, and the bound antigens can subsequently be released by passing a smaller volume of a second aqueous solution through the column. The second solution differs from the first in terms of ionic strength and is formulated to reduce the favorability of antigen-antibody binding reactions. Although this technique certainly merits mentioning, its use is limited with regard to recovering viruses from tap water or environmental freshwaters because it is costly and requires a large amount of preparation time and operational skill. In the literature, only one mention could be found of its use for recovering viruses from water samples.'s Directed adsorption combined with secondaryconcentration. Directed adsorption has generally replaced the other types of concentration techniques listed in Table  2 for use in recovering viruses from large volumes of water. The different solid adsorbents used successfully for this purpose are listed in Table 3 . These solids are divided into two groups, filter materials and granular solids. The types of filters that have been tested are yarn fiber that is wound around a hollow core to form a depth filtration cartridge,"'J4Ri sheet filter materials used as flat layers,l7.2"."1.32.34.71.7~-79cartridges prepared as hollow tubesof filter material,"2,77and cartridges composed of pleated sheets of filter material.28Jl Viruses adsorb onto the filter matrix during the passage of viral-containing water samples through the filters. Recovery of the adsorbed viruses is normally achieved by subsequently passing an eluant through the filters or by dissolving the filter if it is made of alginate.2Y Granular solids are generally used in one of three modes: (1) batch utilization by which thegranules are mixed into the water sample and then recovered using either filtration"" or magnetic attractionx" the latter requiring that the granules be magnetic; (2) layers of granules either supported on or sandwiched between sheets of nonadsorbing flat filter material through which the water sample is passed;rRJ8or columns'",fi0 or fluidized bedsZ7 of granules exposed to the water sample using a flow-through configuration. Subsequent desorption of viruses occurs during exposure of the granules to an eluant using one of these same three modes.

Of the many different types and configurations of virus adsorbents, those now preferred for use in recovering viruses from large volumes of water are wound cartridge filters*' and pleated cartridge filter@ based on either glass fiber or nylon, including types that are positively charged, and columns of glass powder.26 Also of interest is the use of sheet filter material that has been modified in situ either by a precipitation of metal hydroxides within the filter ma-trix2" or by coating the filter with cationic 74 RESEARCHANDTECHNOLOGY Cultures of mammalian cells were prepared using a laminar-flow filtered air hood.

polymersJ2 The largest volumes of sample water that have been concentrated by means of cartridge filters and glass powder appear to be, respectively, 1,900 L"i and 500 L.'7 In comparison, the technique of using a granular adsorbent sandwiched between layers of sheet filter material has been used for concentrating volumes of water as large as 1,000 L."x In the choice between filters and glass powder columns, the filters seem to offer greater advantage because they are easier to transport and operate in the field. This capability eliminates the need for transporting awkward high-volume water samples back to a laboratory. Following the adsorption of viruses onto the filters, the filters can be transported, including postal shipment, back to the laboratory with little concern about harming either the filter or the adsorbed viruses.qO-92 Elution of the filters and subsequent processing of the eluates can then be performed under the more manageable conditions that exist in the laboratory.

Theory of directed adsorption, elution, and reconcentration.

The history of virus adsorption to filters dates back to a study published in 1931 by Elford"" who determined that during passage of virus suspensions through collodion (nitrocellulose) filters, viruses were removed from the solution by adsorption onto the surface of the filter matrix as well as by simple sieving action. The adsorption of viruses onto filters, and presumably other solid adsorbents as well, is governed both by electrostatic interactions, which predominate at lower pH levels (14) and by hydrophobic interactions, which predominate at higher pH levels (?9)."4-'J" The chemical composition of the filter matrix and the fluid in which theviruses are suspended influence the extent of attraction or repulsion between the viruses and the filter. Other important factors that relate to virus adsorption onto filters include the chemical composition of the filter matrix,7rWhe rateof water flow through the filter material,"7 and the ratio of filter pore diameter to virus particle diameter."" These factors are important because if the distance between a virus particle and the filter matrix is toogreat (i.e., the pore size is too large) or the flow rate of the sample through the filter is too great, then the extent of attraction between the virus particle and filter matrix may be insufficient for adsorption to occur.

Cliver"" seems to deserve credit for first presenting the application of this adsorption phenomenon toconcentrating viruses from water samples. Cliver also discovered that the adsorption of viruses to the membrane matrix was strongly inhibited by the presence of added proteins in the input virus .suspension.4".""."*He suggested that this inhibition resulted from a competition of the proteins and viruses for adsorption sites extant on the surface of the filter matrix."x This is indeed a reasonable conclusion in light of the structure of viruses, particularly those that are nonenveloped. Other types of soluble organics, such as humic and fulvic acids, can also interfere with virus adsorption.zOJzJ3.99 Means of facilitating the adsorption of viruses onto solid matrixes include first removing dissolved organic materials from the water samples by passage through a resin colurnn"4.44."" and adding salts to the virus suspension.lOoThese salts may includechlorides of sodium,lOO magnesiurn,'*.2".44.fin.s2 or, more effectively, aluminum. I~.~1.HJ.R:~.92 Virus adsorption can also be facilitated by adjusting the pH of a water sample, a process that can be performed either in a batching operation or in placing an inline injector ahead of the filter.2JJ".1"1 A pH of approximately 3.5 seems preferable for use with negatively charged filters,22J~7fl~X:~ versus the near neutrality (pH = 7) used for the more positively charged filters.r",zx Cliveralsodiscovered that theadsorption process was reversible and was able to recover viruses from filters by eluting them with a proteinaceous solution in the form of diluted blood serum.'" Evidence collected from several studies suggests that eluantscan bedivided into at least two categories based on their mode of action.'",""~l~' The first category is proteinaceous materials, which simply compete with the proteins of the virus particles for binding sites on the adsorbent."" Beef extract is now the most predominantly used eluant material in this category, although other proteinaceous products may prove to be suitable substitutes.~~~i"~l(':~ The second category consists of compounds that alter the favorability of adsorption."","".'"' These eluants include solutions that contain various active substances. Among them are chaotropic agents like glycine or trichloroacetic acid,"'detergents,*~l~2and EDTA serving as a chelating agent for metal cations.1"2 XVolumes of dechlorinated tap water simultaneously containing all three of the indicated viruses were adjusted to the desired pH (3.5 for type A and 7.0 for types Band C) and then passed through a sterile 47.mm-diameter filter of the indicated type. Adsorbed viruses were recovered by passing 10 mL of the indicated eluant through the filter. Calculation of viral recovery was based on the amount 3f virus present in the eluant. expressed as a percent of that contained in the water sample before filtration. Values given represent the average of three independently conducted trials. SAverage value for all three virus types precipitation.lo9 This mechanism differs from true coprecipitation, in which neither of the involved materials can precipitate on its own. A pH of approximately 3.5 appears optimal for recovering viruses from beef extract eluates by organic flocculation.~~~~l0~

With the exception of organic flocculation, the various techniques referenced in Table 4 for secondary concentration of eluates have been discussed earlier in this article. Organic flocculation, introduced by Katzenelson and his coworkers,"'" is performed by lowering the pH of the eluate, during which the proteinaceous material supplied by the beef extract spontaneously precipitates. Viruses contained in the beef extract become associated with this precipitate, which can subsequently be collected by centrifugation and dissolved in a small amount of higher-pH buffer. The process by which virusesareconcentrated during this technique may result from a binding of the naturally precipitating material onto the surface of the viruses at higher pH levels where the material is in solution. This bound material would then cause the viruses to aggregate as the pH of the solution is lowered to induce Comparison of techniques for concentrating viruses using membrane filters. The efficiencies of virus adsorption by different filter materials are compared in Table 5 . The efficiencies of adsorption in combination with subsequent elution are compared in Table 6 . The viruses f2 and +X 174 are bacteriophages that infect Escherichia coli. The procedures for their propagation and assay have been described by Ward and Mahler.l'O The virus designated Pl is human poliovirus type 1, a human enteric virus. The procedure for propagation and assay of this virus has been described by Benton and Hurst."' conducted using water samples that simultaneously contained all three viruses. The viruses were then differentiated on the basis of selective assay procedures.l' These results demonstrate that the adsorption process is often far more efficient than the subsequent desorption. The use of these two processes in conjunction is often referred to as adsorption-elution. This process is sometimes called membrane chromatography when filters are used as the adsorbent.-'" Although the incorporation of EDTA into filter eluants can increase their efficiency, as demonstrated by the results shown in Table 6 , a note of caution is important regarding its use for rotaviruses or reoviruses: EDTA can destroy the infectivity of these two virus groups by removing theouter shell layer of their protein capsids.

Virus assay techniques Of the different filter types mentioned in Tables 5 and 6, type At is composed of glass fiber plus an epoxy resin binder, type B$ is composed of charged glass fiber, and type Cg is composed of charged nylon. The type A material is considered electronegative, whereas the other two types of filter material are considered electropositive. The evaluations represented in these two tables were Stages in the viral assay process. The various procedures that have been used for detecting viruses in concentrates prepared from environmental samples areoutlined in Table 7 exception of direct electron microscopy, each of thesecan be separated into three parts or stages. The first stage consists of the target material that is recognized by a given technique. In the case of electron microscopy, the target material is virus particles, termed virions, that appear intact. Some assay techniques detect only infectious virions, a category that may differ from those virions recognizable by electron microscopy as visually intact but somehow damaged or deficient so that they lack infectiousness. Still other techniques detect only either viral proteins or nucleic acids. The second stage of the various detection methods consists of target recognition; this is considered a nonvisual step. Target recognition, in turn, facilitates the third stage of the detection methods, visualization of the target material. Target visualization may be direct, as in observing virions by electron microscopy, or indirect, as in observing changes induced within host cells as a result of successful viral infection. Indirect visualization may consist of detectingalterations in cellular morphology, including outright death, or the production of progeny viral proteins or nucleic acids.

In general the infectivity of animal viruses contained in environmental samples is examined using cultures of animal cells that are prepared in the laboratory, rather than inoculating live animals. Figure 1 compares the relative sensitivities and required completion times for the assay techniques used in titrating viruses recovered from environmental samples.

Electron microscopy.

The two different approaches for electron microscopy, direct and immune, are presented in Figure 1 in terms of the quantitation of physical viral particles. In contrast, all other viral assay techniques mentioned are represented in terms of their relative sensitivity for detecting viral infectious units. This distinction is important because for many different types of viruses, not all particles in a given preparation may be capable of producing an infection in host cells. For some types of bacteriophages, the ratio of viral particles to infectious units can be as little as between 1 and 2.5. For the various types of animal viruses, the ratioof viral particles to infectious units can range from 2.6 to more than 100.llZ Both categories of electron microscopy involve the examination of virus preparations after they have been negatively stained using solutions of electronopaque metals. In the case of immune electron microscopy, the viruses are incubated in preparations of diluted antibodies before they are negatively stained. The virions in the sample normally aggregate during this incubation if a serological identity exists between them and the antibodies. Neither direct examination nor immune electron microscopy is practical for use with environmental samples because both require that extremely large numbers of viral particles be processed into volumes of, typically, less than 1 mL before they are stained for observation. Furthermore, bacterial viruses and other natural materials present in environmental concentrates could interfere with visualizing animal viruses by electron microscopy.

Cytopathogenicity and plaque assays. Cytopathogenicity assays are based on the appearance of visibly observable, characteristically recognizable changes in the morphology of living host cells that result from the process of viral infection. Such changes, termed "cytopathogenic effects," can be assessed by microscopic examination of the infected cells. Plaque assays detect the production of focal areas of death within an infected culture of cells. Detecting these areas of cell death is often facilitated by exposing the culture of cells to a vital stain such as neutral red. At modest concentrations neutral red does not interfere with cellular viability, yet it causes live cells to appear pink and dead cells to appear colorless. Plaques, which represent the focal areas of cell death, are normally detected by visual examination of the cultures without the aid of a microscope. Explanations of the procedures used in assaying virus preparations by either cytopathogenicity or plaque production, along with specific information on their use with environmental samples and comparisons of their relative effectiveness for this purpose, can be found in references 111 and 113-117. Recent advances for these types of assays include the deliberate use of mixed cell types within single cultures that are to be inoculated with virus samples. Another example is the treatment of cell cultures with nucleotideanalogs such as iododeoxyuridine before the introduction of sample material, both of which can increase assay sensitivity."' Immunological assays. Immunoassays are based on an identity binding between antibodies, used as "probes," and specific "target" antigens of interest to the observer. This identity binding is a natural recognition process and serves as an aid in determining the presence of the target antigens. In thecaseof viruses, such target antigens may be serologically distinctive portions of either proteins, glycoproteins, or glycolipids. Depending on the laboratory protocol followed, these antigens may be suspended in aqueous solutions, trapped on filters, or associated with infected cells. The antibodies used as probes may first be "tagged" or "labeled" to facilitate subsequently identifying the presence of target antigens with which theantibodies bind. Methods used for the taggingof antibodies include chemically incorporating radioactive isotopes into the molecular structure of the antibodies, which can then be detected by the isotope emissions. Alternatively, the antibodies may be covalently linked tocompounds such as fluorescent dyes and functional enzymes. The dyes fluoresce when illuminated at the proper wavelength, normally in the near ultraviolet portion of the spectrum.

When dealing with environmental samples, antibodies conjugated to fluorescent dyes are often used to detect the production of progeny viral proteins within individual cells of an infected host culture. This approach is termed cell immunofluorescence. During performance of the immunofluorescence assay, virally infected cultures of host cells are first exposed to a solution containing fluorescently tagged antibodies, termed probes, whose immunological reactivities are known. During this exposure the probe antibodies bind to the cells wherever the target viral antigen is present. The same cells are later examined using a specially adapted microscope that provides illumination at the proper wavelength for inducing the dye to fluoresce. The resulting fluo- Immunological detection assays that use antibodies tagged with enzymes are broadly termed enzyme immunoassays. These enzyme-tagged antibodies can also be said to serve as probes. Horseradish peroxidase is one example of the enzymes used as tags for this purpose. Enzyme-tagged antibodies can be used in assays for the detection of viral antigens in solution, blotted onto filters, or associated with infected cells. Evidence of the desired antigen-antibody binding reaction is subsequently provided by measuring a chemical reaction facilitated by the enzyme. Very often the observed reaction involves conversion of a colorless substrate into a colored product.

The use of enzyme-tagged antibodies for detecting antigens in solution is termed ELISA, an acronym for enzymelinked immunosorbent assay. References 119 and 120 discuss the use of ELISA to detect viral antigens present in environmental samples. One notable variation of the ELISA technique for detecting viruses in environmental samples relies on first trapping the viral antigens onto filter paper, then performing the ELISA test in situ on that filter.rZ1 This approach can be termed a blot enzyme immunoassay. The use of enzyme immunoassays for detecting viral antigens directly associated with the infected cells of a host culture is similar to cell immunofluorescence and is, likewise, studied microscopically. In this article, this assay technique is referred to as a cell enzyme immunoassay. Payment and Trudel i"i.122 have provided examples of the use of cell enzyme immunoassays for detecting viruses in environmental samples and compared its sensitivity relative to cytopathogenicity assays. Kurstak et alIZ have prepared an excellent review on the use of enzyme immunoassays and other related procedures in virology.

Nucleic acid hybridization.

Nucleic acid hybridization assays are based on the use of homologous nucleic acid materials as probes to bind with, or hybridize to, specific target nucleic acids and thereby facilitate the detection of those target nucleic acids. Hybridization, in this sense, is a physical pairing by means of hydrogen bonding between corresponding nucleotide sequences on those nucleic acid strands that respectively represent the probe and target. Depending on the assay protocol, the target viral nucleic acid sequences may variously be freely suspended in an aqueous solution, bound on filter materials, or associated with virally infected cells. Evidence of the desired hybridization (binding) reaction between the probe and targets may be provided by one of SEPTEMBER 1989 Rslatlve Sa"*itl"lty In'ecllo"s unar assays, normally expressed in terms of infectious units per millilitre as a 50 percent endpoint, were divided by a factor of 2 to obtain endpoints expressed in terms of infectious units per millilitre. +Adapted from reference 117 several methods. The first of these involves the detection of emissions from radioactive isotopes incorporated within the probe nucleic acid material.124,1~5 A second, two-step method involves the use of probe nucleic acids that contain nucleotides tagged, or joined by covalent linkage, to markers such as biotin. In this case, the biotin moieties attached to the probe nucleic acids can, in turn, serve as binding sites for molecules of avidin, which are themselves covalently linked to an enzyme such as horseradish peroxidase. Avidin is a natural compound that binds to biotin with high affinity. In this type of assay, the bound nucleic acid probe material is subsequently detected, as is done for enzyme immunoassays, by measuring the enzyme-mediated conversion of added substrate into products. The use of nucleic acid probes to directly detect viral genomic nucleic acid produced within infected cells is termed in situ nucleic acid hybridization. This process is conceptually similar to using antibodies as probes for detecting viral antigens associated with infected cells. The major difference is whether the target material consists of viral antigens or viral genomic nucleic acids. The use of nucleic acid probes for detecting viral nucleic acid that has been trapped on filter material is termed blot nucleic acid hybridization. A more detailed discussion on the use of nucleic acid probe hybridization assays for virus detection has been published by Norval and Bingham.i2" Comparison of viral assay methods. One important point to consider regarding the use of different assay techniques for identifying viruses in environmental samples, and particularly for viral particles recovered from samples subjected to drinking water treatment, is whether the assay result represents infectivity. None of the two types of electron microscopy procedures, the solution ELISA, nor the blot styles of immunoassay and nucleic acid hybridization assay can, by themselves, reveal this most important piece of information. Their results are instead qualitative or semiquantitative. The other methods compared in Figure 1 are directly usable for quantitating viral infectivity because they are based on detecting changes within host cells that are specifically caused by viral replication. The various changes measured are the disruption of normal cytomorphology (cytopathogenicity), overt cell death (plaque assay), and the production of either viral-specific antigens (immunofluorescence and enzyme immunoassay) or nucleic acid (in situ hybridization). The electron microscopy, solution ELISA, or blotting techniques could be used for estimating the infectivity of viruses present in samples if a part of the sample is examined directly as collected and another part of the same sample is then inoculated into a host cell system supporting the production of viral products that could later be measured by the same type of assay.

Viruses are most suitably concentrated by directed adsorption onto the surface of either a filter material or granular solid. Recovery from the surface is achieved during an elution treatment that reverses the adsorption process. As a result, the viruses are contained in a smaller volume of eluant fluid, which may or may not be further processed by a secondary concentration technique. The choice of a technique for quantitating viruses recovered by the concentration process often involves balancing the relative merits of detection sensitivity versus the time required for assay completion. This article presents information on the actual target materials detected by the various assay techniques and information on evaluating whether a particular type of assay procedure provides a direct measure of viral infectivity. Variations of the nucleic acid hybridization assay may represent the optimum balance for viral assay techniques by simultaneously combining sensitivity, determination of viral infectivity, and rapidity with which the assay can be completed. Cytopathogenicity, cell immunofluorescence, and in situ hybridization for the detection of indigenous adenoviruses present in concentrates produced from environmental samples are compared in Table 8 . These data indicate that the overall numbers of infectious viral units detected by either cytopathogenicity or immunofluorescence were approximately equivalent. The number of viruses detected by in situ hybridization was approximately 40 percent greater than that revealed by the other two techniques. Analysis of these results by means of the paired two-tailed T-test revealed that the differences in mean values between the titers obtained by cytopathogenicity and immunofluorescence were not significantly different (p = 0.67). The mean value of the titers obtained by in situ hybridization were, however, significantly different from those obtained by either cytopathogenicity (p = 0.010) or immunofluorescence (p = 0.003). This information suggests that in situ hybridization is superior toeither cytopathogenicity or immunofluorescence because of both its greater sensitivity (Table 8 ) and the speed with which the assay can be completed (Figure 1 ). 14.

16.

This document has been reviewed and approved in accordance with US Environmental Protection Agency policy. Approval does not signify that the contents reflect the views or policies of the agency, and mention of trade names or products does not constitute endorsement or recommendation for use.

19. References 20.

1.

A recent improvement in virus detection methodology, suitable for use in these types of viral assays, relies on the incorporation of selectively acting antiviral compounds such as guanidine into the medium used for maintaining cell culture viability.1'7 Gmnidine suppresses enterovirus replication while not deleteriously affecting the replication of adenoviruses127 and thus imparts a measure of specificity into the assays. Microbial., 26:518 (1980 This article, which reviews the subject of detecting viruses in water, encompasses two topics. The first topic consists of methods used for concentrating viruses from large volumes of water into smaller, more manageable volumes. The second topic consists of assay methods used for examining viruses contained in the concentrated samples. Many of the references cited in this article contain flow diagrams that outline the techniques discussed. Bull. Acad. Natl. Med., 163:668 (1979) .

Occurrence of Viruses in Treated Drinking Water in the United States

Presence of Enteric Viruses in the Drinking Water of Cochabamba iBolivia)

Comparison of Adsorption-Elution Methods for Concentration and Detection of Viruses in Water

Evaluation of Gauze Pad Method to Recover Viruses From Water

Virus in Water, I. A Preliminary Study on a Flow-Through Gauze Sampler for Recovering Virus From Waters

Evaluation of Methods for Concentrating Hepatitis A Virus From Drinking Water

Concentration of Poliovirus From Tap Water Onto Membrane Filters With Aluminum Chloride at Ambient pH Levels

Concentration of Poliovirus From Tap Water Using Positively Charged Microporous Filters

Use of Bituminous Coal as an Alternative Technique for Field Concentration of Waterboine Viruses

Poliovirus Concentration From Tao Water With Electropositive Adsorbent Filters

Influence of Water Quality on Enteric Virus Concentration by Microporous Filter Methods

Simultaneous Concentration of Four Enteroviruses From Tap, Waste. and Natural Waters

Evaluating Adsorbent Filter Performance for Enteri; Virus Concentrations in Tap Water

Concentration of Viruses From Water by Using Cellulose Filters Modified bv In Situ Precipitation of Ferric and Aluminum Hydroxides

Concentration of Enteroviruses From Large Volumes of Tap Water, Treated Sewage, and Seawater

Isolation of Enteroviruses From Water

A New and Simple Method for Recuperation of Enterobiruses From Wate

Detection d'Enter-3virus par Concentration sur Poudre de Verre en lit Fluidise a Partir

Round-Robin Investigation of Methods for the Recovery of P&ovirus From Drinking Water

Conception d'Une Membrane Filtrante a Base d'Alginate Pour la Concentration des Virus Hydriques

Virus Isolations From Sewage and From a Stream Receiving Effluents of Sewage Treatment Plants

Concentration of Viruses From Large Volumes of Tap Water Using-Pleated Membrane Filters

Effects of Humic Materials on Virus Recovery From Water

Effects of Humic and Fulvic Acids on Poliovirus Concentration From Water by Microporous Filtration

Concentration of Enteroviruses From Large Volumes of Water

Simultaneous Concentration of Salmonella and Enterovirus From Surface Water bv Using Micro-Fiber Glass Filters

Rapid Concentration of Bacteriophages Frbm Aquatic Habitats

Comparison of Talc-Celite and Polyelectrolyte 60 in Virus Recovery From Sewage: Development of Technique and Experiments With Poliovirus (Type 1, Sabin)-Contaminated Multilitre Samples

Use of Talc-Celite Layers in the Concentration of Enteroviruses From Large Volumes of Potable Waters

Effects of Bentonite Clay Solids on Poliovirus Concentration From Water by Microporous Filter

Concentration of Coliphage From Water and Sewage With Charge-Modified Filter Aid

Recent Advances in the Detection of Human Viruses in Drinking Water

Novel Approach for Modifying Microporous Filters for Virus Concentration From Water

Concentration of Enteroviruses on Membrane Filters

Enterovirus Detection by Membrane Chromatography. Transmission of Viruses by the Water Route

Concentration of Poliovirus in Water by Molecular Filtration

Ein Zweistufenverfahren zur Virusanreicherungaus Losungen mit Geringem Virustiter

Protamine Precipitation of Two Reovirus Particle Types From Polluted Waters

Concentration of Enteric Viruses in Water by Hydro-Extraction and Two-Phase Separation. Transmission of Viruses by the Water Route

Detection of Enteric Viruses by Concentration With Polyethylene Glvcol. Transmission of Viruses by the Wilter Route

Concentration of Viruses on Aluminum Phosphate and Aluminum Hydroxide Precipitates

Transmission i,f Viruses by the Water Route

The Phase-Separation Method for the Concentration and Detection of Viruses in Water

Studies on the Isolation and Identification of Hepatitis Viruses in Water

Protein Adsorption on Glass Surfaces Using Porous Glass

The Use of Gauze Swabs for the Detection of Poliomyelitis Virus in Sewers. Poliomyelitis

Concentration of Viruses From Water by Membrane Chromatography

Membrane Filters in Virology

A Simple Method for Concentrating and Detecting Viruses in Water

Comparative Study of Four Microporous Filters for Concentrating Viruses From Drinking Water

Developments in Methods for the Detection of Viruses in Large Volumes of Water

Concentration of Bacteriophages From Natural Water

Virus Detection in Water by Ultrafiltration

Retention and Recovery of Polioviruses on a Soluble Ultrafilter. Transmission of Viruses by the Water Route

Recovery of Viruses From Waste and Other Waters by Chemical Methods

The Chemistry of Virus Concentration by Chemical Methods

MIX, T.W. The Physical Chemistry of Membrane-Virus Interaction

Recovery and Removal of Viruses From Water-Utilizing Membrane Techniques

Rapid Concentration of Bacteriophages From Large Volumes of Freshwater: Evaluation of Positively Charged, Microporous Filters

Detection of Human Rotavirus in Sewage Through Two Concentration Procedures

Membrane Filter Evaluations Using Poliovirus

Concentration of Virus From Tap Water at Ambient Salt and pH Levels Using Positively Charged Filter Media

Concentration of Viruses in Beef Extract by Flocculation With Ammonium Sulfate

Rotavirus Concentration From Raw Water Using Positively Charged Filters

A Comparison of Current Methods of Poliovirus Concentration From Tap Water

Concentration of Simian Rotavirus SA-11 From Tap Water by Membrane Filtration and Organic Flocculation

Evaluation of Procedures for Recovery of Viruses From Water-I. Concentration Systems

Efficiency of Several Micro-Fiber Glass Filto= fnr Recovery of Poliovirus From Tap Water

Concentration of Heptatitis A Virus

Recovery of Small Quantities of Viruses From Clean Waters on Cellulose Nitrate Membrane Filters

Wound Fiberglass Depth Filters as a Less Expensive Approach for the Concentration of Viruses From Water

A Portable Device for Concentrating

Bacteriophages From Large Volumes ot Freshwater

Modifications of the Tentative Standard Method for Improved Virus Recovery Efficiency

Modified Membrane-Filter Procedure for Concentration of Enteroviruses From Tap Water

Effectiveness of the Organic Flocculation Method in Concentrating Echovirus 7 and Coxsackirvirus A9 From Water

Preformed Magnesium Hydroxide Precipitate for Second-Step Concentration of Enteroviruses From Drinking and Surface Waters

Detection of Viruses in Drinking Water bv Concentration on Magnetic Iron Oxide

Concentration of Seeded Simian Rotavirus SA-11 From Potable Waters by Using Talc-Celite Layers and Hydroextraction

Adsorption of Virusrs on Magnetic Particles. Australian Water Resources Council

Survie de Virus Enteriques Adsorbes sur Microfibre de Verreau Cours d'un Transport Postal. Can. lour. Microbial

Processing and Transport of Environmental Virus Samples

A New Series of Graded Collodion Membranes Suitable for General Bacteriological Use, Especially in Filterable Virus Studies

Influence of Salts on Electrostatic Interactions Between Poliovirus and Membrane Filters

Effects of Chaotropic and Antichaotropic Agents on Elution of Poliovirus Adsorbed

Membrane Filters

Virus Interactions With Membrane Filters

Virus Retention by Membrane Filters

Factors in the Membrane Filtration of Enteroviruses

Humic Acid Interference With Virus -Recovery by Electropositive Microporous Filters

Efficient Filtration and Sizingof Viruses With Membrane Filters

Reduction of InterferingCytotoxicity Associated With Wastewater Sludge Concentrates Assayed for Indigen&s Enteric Viruses

Induction of Cytopathogenicity in Mammalian Cell Lines Challenged With Culturable Enteric Viruses and its Enhancement by 5-Iododeoxyuridine

Evaluation of Procedures for Recovery of Viruses From Water-II. Detection Systems

Influence of Inoculum Size, Incubation Temperature, and Cell Culture Densityon Virus Detection in Environmental Samples

Improved Method for the Use of Proportioning Injectors to Condition Large Volumes of Water for Virological Analysis

Comparison of Positively Charged Membrane Filters and Their [Jse in Concentrating Bacteriophages in Water

Comparison of Commercial Beef Extracts and Similar Materials for Recovering Viruses From Environmental Samples

Second-Step Reconcentration of Environmental Samples by Ammonium Sulfate Flocculation of Beef Extract

Organic Flocculation: An Efficient Second-Step Concentration Method for the Detection of

Recovery of Viruses From Water by a Modified Flocculation Procedure for Second-Step Concentration

Immuno-peroxidaseMethod With Human Immune Serum Globulin for Broad-Spectrum Detection of Cultivable Human Enteric Viruses: Application to Enumeration of Cultivable Viruses in Environmental Samples

Hepatitis A Virus Concentration on Cellulose Membranes

Imoroved Method for Recovery of Enteric Viruses From Wastewater Sludges

Uptake of f2 Through Plant Roots

Evaluation of Mixed Cell Types and 5-Iodo-2' Deoxyuridine Treatment Upon Plaque Assav Titers of Human Enteric Viruses

Hybridization as Methods for the Detection of Adenoviruses

Clinical and Research Application of an Enterovirus Group

Comparison of Methods for Rotavirus Detection in Water and Results of a Survey of Jerusalem Wastewater

An A-ELISA to Detect Heptatis A Virus iii Estuarine Samples

lJse of the Nitrocellulose-Enzvme Immunosorbent Assay for Rapid, Sensitive, and Quantitative Detection of Human Enteroviruses

Detection and Quantitation of Human Enteric Viruses in Waste Waters: Increased Sensitivity IJsing a Human Immune Serum Globulin-Immunoperoxidase Assay on MA-104 Cells

Enzyme Immunoassays and Related Procedures in Diagnostic

Detection of Hepatitis AVirus by Hybridization With ~Single-Stranded -RNA Probes

Detection of Hepatitis A Virus in Seeded Estuarine Samules bv Hybridization With cDNA Probes

Advances in the Use of Nucleic Acid Probes in Diagnosis of Viral Diseases of Man

Suppression of Viral Replication by Guanidine: A Comparison of Human Adenoviruses and Enteroviruses