Infectivity-based assays are the assays of choice for the detection of pathogenic mammalian viruses. While it is intuitively appropriate to conduct testing and validation studies with the known viral burden or a closely related mammalian species, logistic considerations often dictate otherwise. Consequently, bacteriophages have served as suitable surrogates for mammalian viruses in both medical and environmental virology applications. The wide range of bacteriophages available offers a powerful analytical tool amenable to several different applications: filter validation studies (where removal is based on size exclusion), investigations into virus contamination control issues, evaluation of barrier materials, etc. There is a considerable body of evidence to suggest and support the use of bacteriophages as surrogates for mammalian viruses. Use of appropriately sized bacteriophages provides an innocuous, efficacious and expeditious method for economical testing and validation of viral clearance capabilities of virus removal filters, thus facilitating performance of filter validation studies in biopharmaceuticals under product- and process- specific conditions in an overall effort towards ensuring the virological safety of biologicals. This paper discusses the limitations associated with mammalian virus assays and provides a rationale for the use of bacteriophages as surrogates for mammalian viruses. Data from published literature documenting applicability of bacteriophages in filter validation studies, especially when removal is based on size exclusion, is reviewed along with examples of studies from the fields of medical and environmental
virology.

Prodigious advances in biotechnology in recent years have led to a dramatic increase in the number of commercially produced biologicals with therapeutic and prophylactic applications. A considerable range of both conventional and recombinant DNA-derived products are currently available; these include vaccines, plasma-derived coagulation factor concentrates, monoclonal antibodies, recombinant proteins and hormones, etc. When biologicals are either sourced from or produced in biological systems, there is a possibility of product contamination by viruses associated with the system, i.e. endogenous viruses associated with continuous cell lines, e.g. murine retroviruses, or pathogenic viruses that may be associated with source materials such as blood, e.g. Hepatitis viruses, Human Immunodefic-iency Virus (HIV). Alternatively, viruses may gain access to the biological product extraneously, i.e. through raw materials/supplements used in the bioreaction, reagents used in purification, e.g. monoclonal-antibody affinity chromatography columns, from manufacturing environments/personnel, etc. Contamination of biologicals by adventitious viruses such as bovine viral diarrhea virus (Erickson et al., 1991), epizootic haemorrhagic disease virus (Rabenau et al., 1993) and minute virus of mice (Garnick, 1996) has been reported. The inherent risks concomitant with the use of biologicals must be recognized, and clinical acceptability must be guided by risk-benefit analysis.

Historically, one of the factors contributing to the occurrence of iatrogenic viral infections has been the inability of the available methodologies to detect the viral contaminant. Thus, for example, the yellow fever vaccine administered during World War II was contaminated with hepatitis B virus (HBV) (Fox et al., 1942) which was adventitiously introduced into the vaccine along with the human serum used as the excipient. Though the serum was implicated in the hepatitis infection, actual demonstration of HBV presence was not possible, as neither infectivity assays nor alternative detection methods (e.g. molecular diagnostic techniques such as polymerase chain reaction) were available at the time. Similarly, endogenous contaminants have been associated with biological systems used in vaccine manufacture, as, for example, presence of SV40 in the primary Rhesus monkey kidney cell line used for polio vaccine production (Shah and Nathanson, 1976) and the avian leukosis virus contamination of the hens eggs used for cultivation of the yellow fever virus (Harris et al., 1966).

Recognizing the potential for viral contamination of biologicals by both known and potential or putative viruses (i.e. ones that may exist and are potentially infectious but which remain undetectable by presently available detection methodologies), current regulatory guidelines require that an adequate safety margin be incorporated during the manufacturing process to provide assurance of virological safety. Manufacturing processes must be validated for their ability to clear both real and potential viruses that could possibly contaminate the final product (ICH harmonized tripartite guideline, 1997). Validation studies are currently being conducted using a panel of viruses which may include relevant viruses (i.e. viruses known to be actual contaminants like HIV-1), specific model viruses (i.e., viruses that are closely related to the purported contaminant, as, for example, Murine Leukemia Virus (MuLV) which is used as a model for endogenous retroviral contaminants), and also non-specific model viruses which include viruses with different physicochemical properties. Inclusion of non-specific model viruses is for the purpose of evaluation of the robustness of the manufacturing process to ensure clearance of a broad range of viral contaminants (ICH harmonized tripartite guideline, 1997).

This paper discusses the limitations associated with mammalian virus assays and provides a rationale for the use of bacteriophages as surrogates for mammalian viruses. Data from published literature documenting applicability of bacteriophages in filter validation studies, especially when removal is based on size exclusion, is reviewed along with examples of studies from the fields of medical and environmental virology.

a. Limitations with mammalian virus assays: The only true test of viral infectivity is a biological assay. Routinely, infectivity assays for mammalian viruses are undertaken by in vitro inoculation of a susceptible cell line with the virus of interest, followed by monitoring for appearance of cytopathic effects (CPEs) such as formation of plaques, foci of infection, or induction of abnormal cellular morphology (Brando, 1995). This approach is feasible primarily when the identity of the viral etiological agent is known, as, for example, in the diagnosis of specific viral infections. Infectivity-based mammalian virus assays have several limitations: (i) they have limited sensitivity; i.e., the small sample volumes used in viral infectivity assays may result in low concentrations of virus being undetected; yet, these low levels of virus may be able to initiate infection in a susceptible host. (ii) Viral assays are highly specific; the great diversity of viruses, therefore, dictates that different assay systems be used for the detection of each specific virus. (iii) Furthermore, if unknown viral variants arise, they may have altered amphitropism, and would then remain undetected. (iv) The lengthy analysis time (up to 6 weeks) for infectivity assays, makes the results of questionable value, especially when the testing is undertaken for routine monitoring for viral contaminants, as, for example, in the manufacture of biologicals or the monitoring of water quality. (v) Some viruses are not culturable in vitro due to lack of susceptible cell lines, e.g. certain blood-borne (e.g. Hepatitis C virus, HCV) and water-borne (e.g. Norwalk virus) viruses. While diagnostic procedures based on electron microscopy, immunoassays, or molecular techniques are of assistance in the detection of viruses especially those not readily culturable in vitro (DeLeon et al., 1992; Jiang et al., 1987), they fail to distinguish between infectious and non-infectious particles, which are of critical relevance in public health virology.

b. Applicability of surrogate species: While it seems intuitively appropriate and preferable to use the specific entity implicated as the contaminant/etiological agent to evaluate the efficacy of processes/conduct validation studies, this approach is not always feasible. Therefore, use of surrogate species to serve as indicators of contamination is often advocated and accepted. In the case of bacteria, indicator bacterial species have been identified and their applicability as indices of contamination has been well documented. For example, Brevundimonas (Pseudomonas) diminuta is accepted as the indicator organism in sterile filtration applications (Guidelines for Sterile Drug Products Produced by Aseptic Processing, 1987). For sterilization by heat, radiation or gas (ethylene oxide), spores of Bacillus stearothermophilus, Bacillus pumilus and Bacillus subtilis, respectively, are the recommended indicator species (Meltzer, 1987). Coliform bacteria are the most commonly used indicators of fecal pollution (Grabow, 1986); reportedly, detection of Escherichia coli yields almost conclusive evidence of fecal pollution (Grabow and DuPreez, 1979).

Bacteriophages have been used as indicators of viral pathogens both in medical and environmental virology applications (Aranha-Creado et al., 1997, 1998; Brown, 1993; Kott et al., 1974; Lytle et al., 1991a, 1991b, Stetler, 1984). Bacteriophages are favored as surrogates for mammalian viruses for several reasons: (i) They are innocuous and allow for expeditious and economical screening for the presence of pathogenic mammalian viruses. (ii) They can be grown to higher titers than most mammalian viruses and, therefore, can constitute a more sensitive assay. (iii) Results from bacteriophage assays are available within several hours post inoculation compared with the days to weeks required to obtain results from mammalian virus infectivity-based assays. (iv) Also, unlike testing for mammalian viruses, phage assays do not require specialized testing facilities or specific expertise. All these considerations make bacteriophage assays a very convenient analytical tool. Table I lists several of the advantages of using bacteriophages as surrogates for mammalian viruses.

Table I
Use of Bacteriophages as Surrogates for Mammalian Viruses in Viral Clearance Assays – Advantages

• Innocuous

• Cultivatable to high titers (> 108 / ml)

• Detectable by rapid and inexpensive infectivity assays

• High degree of sensitivity and specificity

Filtration processes, in general, operate via the following two mechanisms: (i) size exclusion, which occurs due to geometric or spatial constraint, and, (ii) adsorptive retention of particulates. Size exclusion provides a predictable method of particle removal as it is not directly influenced by process and filtration conditions; consequently, it constitutes a ‘robust’ method of viral clearance. Factors affecting adsorptive retention include electrokinetic or hydrophobic interactions, filtration process conditions and characteristics of the product to be filtered. While both of these mechanisms operate concomitant with each other, the relative importance and role of each may vary.

When the specific viral clearance mechanism is based on size exclusion, removal of the particle, albeit, biological, is effected by spatial constraint. Therefore, it is logical that a virus (mammalian or bacterial) in the same size range should be applicable as a surrogate. The following considerations are important in the selection of an appropriate model virus for filter validation: (i) the size of the virus (which should simulate the virus of interest) and minimal aggregation of the model virus under the test conditions, (ii) shape, i.e. spherical; (iii) ability to cultivate the virus to a high concentration, to ensure adequate challenge levels during testing; (iv) sensitivity and ease of detection with an appropriate infectivity assay;
and, (v) stability of the virus under the test conditions.

a. Filter validation for documentation of viral clearance: In the manufacture of biologicals, the production process has to be validated for its bacterial/viral clearance capability. For example, to document bacterial retention, Brevundimonas (Pseudomonas) diminuta is accepted as the indicator organism in sterile filtration applications (Guidelines for Sterile Drug Products Produced by Aseptic Processing, 1987). Validation documentation requirements with regard to virus reduction by filtration are as follows: (a) viral clearance documentation by the filter manufacturer that the (model) virus-removal performance claim can be correlated to a test applicable during product manufacture, such as a filter integrity test; (b) product- and process- specific validation studies conducted with one or more viruses. Bacteriophages fulfill the criteria for an appropriate model and have been demonstrated to be suitable surrogates for mammalian viruses in filter validation studies.

Thus, for example, since bacteriophage ø6 and mammalian retroviruses are in the same size range (Retroviruses: 80-100 nm; bacteriophage ø6: 75 nm), bacteriophage ø6 has been used as a surrogate for retroviruses such as Murine Leukemia Virus (MuLV) and HIV (Aranha-Creado et al., 1998; Lytle et al., 1992). Aranha-Creado et al.(1998) challenged the grade DV50 membrane with MuLV in a monoclonal antibody product; no MuLV was detected in effluent aliquots assayed and these investigators reported LTR values of > 6.7 for this virus. The greater than (>) sign indicates that this is, in fact, a worst case estimate of filter viral clearance and higher clearance capabilities could possibly be demonstrated if higher input challenge concentrations could be achieved. These results corroborate LTR values reported for ø6 (LTR: > 8.7) in gelatin phosphate as the carrier fluid (Aranha-Creado et al. 1997). Similar LTR values, i.e. 4.9 and 4.7 for MuLV and ø6, respectively, were reported on challenge of the Polyether sulfone (PES) Omega 300 K VR ultrafiltration membrane cassettes (Pall Filtron, Northborough, MA; Aranha-Creado and Herczeg, 1998). It must be noted that MuLV is employed as a specific model virus for endogenous retroviruses (ICH Tripartite Guideline, 1997) and is routinely included in the panel of challenge viruses in validation studies with continuous cell culture-derived products. Representative data from the above cited studies are summarized in Table II. Table III lists specific physicochemical properties of bacteriophages used in filter validation studies.

Table II
Viral Clearance Studies With Mammalian and Bacterial Viruses in Filtration Applications

Filter Type1	Challenge Mammalian Virus and Size (nm)	Log Titer Reduction (LTR)	Challenge Bacteriophage and Size (nm)	Log Titer Reduction (LTR)	Reference
PVDF: Ultipor® VF grade DV50	• Murine Leukemia Virus (80-120)	> 6.7	ø6 (75)	> 8.7	Aranha-Creado et al., 1998, 1997
	• Poliovirus (25-30)	> 2.3	PP7 (25)	< 1	Oshima et al., 1996
PES: Omega® 300 K VR	• Murine Leukemia Virus (80-120)	4.9	ø6 (75)	4.7	Aranha-Creado and Herczeg, 1998
PAN: MWCO, 50K	• Poliovirus (25-30)	4.6	PP7 (25)	4.3	Oshima et al., 1995
PS: MWCO, 6K	• Poliovirus (25-30)	7.4	PP7 (25)	7.6	Oshima et al., 1995
PVDF: Viresolve/ 70™	• Murine Leukemia Virus (80-120)	> 6.7	ø6 (75)	> 6.8	DiLeo et al., 1993
	• Poliovirus (25-30)	3.5	øX-174 (25-27)	2.93	DiLeo et al., 1993

¹: Abbreviations:PVDF: Polyvinylidene Difluoride; PES: Polyethersulfone; PAN: Polyacrylonitrile; PS: Polysulfone.

Similarly, Oshima et al. (1996) used several different bacterial and mammalian viruses in their viral retention studies for evaluation of a PVDF microfiltration membrane (Ultipor® VF grade DV50 membrane; Pall Biopharmaceuticals, East Hills, NY). The grade DV50 membrane is applicable for virus removal from products with molecular weight in the immmunoglobulin G (IgG) size range, i.e. approximately 160 kD. These investigators conducted challenges using several viruses in different carrier fluids (water, Dulbecco’s Modified Eagle Medium-MEM, and MEM+10% serum); for small viruses such as poliovirus (sized at 25-30 nm) and bacteriophage PP7 (sized at 25 nm) they reported log titer reduction values (LTR) of 2.3 and < 1, respectively, in MEM+10% serum as the carrier fluid (Table II), suggesting similar retention characteristics. In other studies evaluating the virus retentive capabilities of hollow-fiber ultrafilters, Oshima et al. (1995) reported that polyacrylonitrile (PAN) hollow fiber modules (Microza®, Pall Corporation, East Hills, NY), with molecular weight cut-off (MWCO) of 50 K, provided similar viral clearance for both poliovirus and bacteriophage PP7, i.e. LTR of 4.6 and 4.3 (in MEM+10% serum) for poliovirus and bacteriophage PP7, respectively (Table II). Polysulfone (PS) hollow fiber modules (Microza®, Pall Corporation, East Hills, NY) with MWCO: 6 K provided LTR of 7.4 and 7.6, for poliovirus and bacteriophage PP7, respectively (Oshima et al., 1995). Based on their data, Oshima and coworkers suggested that bacteriophage PP7 can serve as an appropriate surrogate for poliovirus; they indicated that in view of the slightly smaller size of PP7 (25 nm) as compared with poliovirus (28-30 nm), bacteriophage PP7, in fact, constitutes a ‘worst case’ model in size-exclusion filter validation studies. Other bacteriophages in the similar size range (25 nm), e.g. øX-174 have also been used as surrogates for poliovirus. Thus, for example, DiLeo et al. (1993) challenged a PVDF ultrafilter (Viresolve/70™; Millipore Corporation, Bedford, MA), a membrane applicable for filtration of albumin and similar sized molecules, i.e. ~ 66 kD with poliovirus and bacteriophage øX-174 (in phosphate buffered saline as the challenge fluid), and reported LTR values of 3.5 and 2.93, respectively (Table II).

b. Filter validation and use - practical considerations: Logistic considerations dictate that filter validation testing by the filter user be conducted in scaled-down studies. The pathogenic potential of viruses, the high viral concentrations needed, the considerable harvest of viral stock required and especially introduction of viruses into the production environment (which is against good manufacturing practices (GMPs)) all make virus validation studies at process scale impractical, if not impossible. While it is, therefore, acceptable from a validation principle and regulatory standpoint to provide product-specific virus-removal validation data conducted with scaled-down tests, the filter’s virus-retention performance under process conditions must also be confirmed. Filter manufacturers, therefore, provide filter validation data correlating virus-retention performance, using an appropriate model virus, to a non-destructive physical test such as an integrity test.

One type of non-destructive integrity test is the forward flow test which is based on the measurement of air flow through a wetted filter at a defined test pressure (Pall, 1975). Thus, for example, in the case of the Ultipor® VF grade DV50 filter, the manufacturer-specified forward flow integrity test limit value is12.5 cc/min per 10 inch element (P/N: AB1UDV50); filters that test at or below this limit value will assuredly provide a log titer reduction of 6 or greater for any virus larger than 50 nm (Aranha-Creado et al., 1997). This virus retention claim was established and documented using bacteriophage PR772, which is sized at 53 nm (Validation Guide for Pall Ultipor® VF™ Grade DV50 Ultipleat™ AB Style Virus Removal Filter Cartridges, 1995). Figure 1 provides the validation data for the AB1UDV50 elements (Aranha-Creado et al., 1997). The validity of the virus-removal claim, i.e. removal of viruses larger than 50 nm has been documented in published (Anazawa et al., 1997, Aranha-Creado et al., 1997, 1998; Oshima et al., 1996; Roberts, 1997) and unpublished data (personal communications). Thus, for example, in studies with the grade DV50 filter greater than 6 log removal has been demonstrated for the following mammalian viruses: Vaccinia virus (250 nm x 300-450 nm); Herpes Simplex Virus (120-300 nm); Pseudorabies virus (120-300 nm); Influenza virus (80-120 nm); Murine Leukemia Virus (80-120 nm); Human Immunodeficiency virus (80-100 nm); Adenovirus (65-80 nm); Reovirus (60-80 nm); Sindbis virus (40-70 nm), and SV40 (40-55 nm).

Figure I
Correlation of forward flow limit (isopropyl alcohol/water, 30:70 wet at 85 lb/in²) with log titer reduction (for the model virus, bacteriophage PR772) for Ultipor VF AB1UDV50 filter elements (Aranha-Creado et al., 1997)

Table III
Characteristics of Bacteriophages That Have Been Used as Size-Based Models in Medical Virology Applications

Virus	Family (-viridae)	Genus	Host	Genome	Envelope	Size (nm)	Shape
ø6	Cysto-	Cystovirus	Pseudomonas phaseolicoli	RNA	Yes	75-86	Spherical
PR772	Tecti-	Tectivirus	E. coli K 12	DNA	No	53	Icosahedral
PP7	Levi-	Levivirus	Pseudomonas aeruginosa	RNA	No	25	Icosahedral
MS-2	Levi-	Levivirus	E. coli K12	RNA	No	25	Icosahedral
øx174	Micro-	Microvirus	E. coli C	DNA	No	25-27	Icosahedral

Table IV
Examples of Studies with Bacteriophages as Surrogates for Mammalian Viruses in Medical Virology-Related Applications

Bacteriophage Model	Study Type	Comment (Used As...)	Reference
ø6	• Filter validation	• Size-based model for Murine Leukemia virus	• Aranha-Creado et al., 1998
	• Evaluation of Gloves	• Size-based model for HIV	• Lytle et al., 1991b
	• Evaluation of virucidal activity of lipophilic dyes	• Model for enveloped viruses	• Lytle et al., 1991a
PR772	• Filter validation	• Size-based model for viruses > 50 nm	• Aranha-Creado et al., 1997
PP7	• Filter validation	• Size-based model for polio virus	• Oshima et al., 1995; 1996
øX-174	• Filter validation	• Size-based model for polio virus	• DiLeo et al., 1993
	• Evaluation of protective clothing in healthcare environments	• ‘Worst-case model’ for blood-borne pathogens such as HIV, HBV, HCV	• Brown, 1993
	• Evaluation of condoms	• ‘Worst-case model’ for sexually- transmitted viruses	• Lytle et al., 1990; Voeller et al., 1994
MS2	• Evaluation of virucidal disinfectants	• Model for polio virus	• Jones et al., 1991; Maillard et al., 1994
	• Evaluation of membrane gas filters	• Model to document effective virus control of bioreactor gas inlets/ exhausts	• Bradburne et al., 1994

Table V
Examples of Studies woth Bacteriophages as Surrogates for Mammalian Viruses in Environmental Virology - Related Applications

Bacteriophage Model	Study Type	Comment (Used As...)	Reference
Escherichia coli phages (coliphages)	• Water quality surveillance	• Models for enteroviruses	• Grabow, 1986; Stetler, 1984
	• Virucidal treatments of sewage effluents	• Model for enteroviruses	• Kott et al., 1974
Bacteroides fragilis phages	• Water pollution	• Specific indicator for fecal pollution	• Chung and Sobsey,1993; Jofre et al., 1986
	• Effectiveness of drinking water treatments	• Model for water quality monitoring	• Jofre et al., 1995
MS2	• Sewage effluent treatment	• Model for enteric viruses	• Kott et al., 1974

Literature is replete with studies in the field of medical and environmental virology documenting the applicability of bacteriophages as surrogates for mammalian viruses (Tables IV and V). The following is a brief review of the reported studies.

a. Bacteriophages as surrogates for mammalian viruses in Medical Virology Applications:
Evaluation of the efficacy of barrier materials: Due to the significant risks associated with occupational exposure to pathogens, barrier materials are currently used for nosocomial infection control in addition to personal protection. Concern about virus transmission through gloves, condoms and other barrier materials has necessitated the evaluation of the integrity of these materials. Liquid penetration tests such as the 1000-ml water leak test detect only gross defects, i.e. holes larger than 10 microns (Carey et al., 1989), and therefore, it is generally recognized that virus penetration tests are essential (Kotilainen et al., 1992). The significant risks associated with laboratory testing conducted with pathogens that may compromise safety in the health care environment, e.g. HBV, HCV, HIV, or viruses implicated in sexually transmitted diseases, e.g. HIV; Herpes Simplex Virus (HSV) necessitates that studies be undertaken with non-virulent species such as bacteriophages. For example, in studies on viral penetration of barrier materials such as gloves, Lytle et al. (1991b) evaluated several bacteriophages (øX-174, T7, PRD1, ø6) as possible surrogates for pathogenic human viruses. In other studies, Lytle and coworkers (Lytle et al., 1992) conducted experiments on sizing by filtration to determine the effective diameter of HIV-1, HSV-1 and several bacteriophages. These investigators concluded that bacteriophage ø6 models HIV-1 in filtration size (Lytle et al, 1992) and bacteriophage ø6 is currently being used as a surrogate for HIV-1. Similarly, for evaluation of protective clothing for health care workers, Brown (1993) recommended use of bacteriophage øX-174 as a pathogen model for HBV, HCV and HIV for the following reasons: øX-174 is icosahedral or nearly spherical (similar to HBV, HCV and HIV), non-enveloped (similar to HCV) and considerably smaller (25 nm) than the pathogens listed and, therefore constitutes a 'worst-case' challenge of the test material. Studies (Lytle et al., 1990; Voeller et al., 1994) to evaluate the penetrability of condoms made from natural and artificial membranes also used bacteriophage øX-174 as a ‘worst-case’ challenge virus for sexually-transmitted viruses such as Human papilloma virus (45-55 nm), HIV (80-100 nm), Herpes Simplex Virus (120-300 nm) and Cytomegalovirus (120-300 nm).

Evaluation of virucidal compounds: For determining efficacy of photoinactivation by lipophilic dyes, Lytle et al. (1991a) evaluated HSV-1 along with several lipid-containing (e.g. ø6, PRD1) and non-lipid-containing (øX-174, T7) bacteriophages. Their studies demonstrated the suitability of using bacteriophage f6 as a surrogate for enveloped viruses to evaluate photoinactivation by lipophilic dyes. Bacteriophage ø6 possesses a lipid envelope (Vidaver et al., 1973) whereas bacteriophage PRD1 has an internal lipid layer, similar to bacteriophage PR772 (Olsen et al., 1974). Bacteriophage MS2 has been evaluated as a model in studies to determine viral sensitivity to disinfectants (Lehman and Bansemir, 1987; Maillard et al., 1994) and to help establish viral models for human pathogens (Figueroa et al., 1978). In studies on the evaluation of virucidal activity of biocides, Jones et al. (1991) report that bacteriophage MS2 behaves similar to poliovirus; they, therefore, advocate replacing human enterovirus assays with bacteriophage assays using MS2.

b. Bacteriophages as surrogates in Environmental Virology Applications: The introduction of pathogenic viruses into the environment as a consequence of sewage disposal practices, i.e. disposal over land or discharge into bodies of water, raises public health issues. Reliable and practical methods of assessment of water quality from a virological safety standpoint are essential because a variety of pathogenic viruses, commonly referred to as the enteric viruses, are transmitted by water. Enteric viruses may cause severe diseases, such as paralysis (polioviruses), meningitis (echovirus), myocarditis (coxsackievirus), or infectious hepatitis (hepatitis A and E viruses). Enteric viruses, which are transmitted by the fecal-oral route of transmission belong to several taxonomic groups; some of the families include: Picornaviridae (Enteroviruses: polio, Coxsackie A and B, ECHO, enteroviruses 68-71 and enterovirus 72 or hepatitis A virus), Adenoviridae (adenoviruses 40 and 41), Caliciviridae, Coronaviridae and Reoviridae (reoviruses and rotaviruses). Several viruses such as the Norwalk-like viruses, Astroviruses, Small Round Viruses and enteric non-A, non-B hepatitis viruses are poorly characterized.

Evaluation of water quality: Potable and recreational waters have been routinely evaluated for sewage/fecal contamination using Escherichia coli as the indicator of fecal pollution. However, it has become increasingly clear that commonly used bacterial indicators of fecal contamination (e.g. coliforms) have limited predictive value for human pathogenic viruses. Enteroviruses have been detected in recreational waters considered acceptable as judged by total and fecal coliform standards (Gerba et al., 1979). The uptake/bioaccumulation of viral and bacterial pathogens by shellfish (e.g. oysters, mussels, clams), especially since many of them are eaten either raw or partially cooked, must be addressed (Lucena et al., 1994). Viral disease outbreaks have been linked to ingestion of shellfish harvested from contaminated environments (Desenclos et al., 1991; Truman et al., 1987). Enteroviruses have been detected in waters that met standards for shellfish-harvesting (Gerba et al., 1979). The failure to correlate occurrence of enteroviruses with the detection of indicator bacteria and the frequent occurrence of enteroviruses in water that meets current bacteriological standards, highlights the necessity for additional standards in water quality safety.

Presently available virological techniques are impractical in water quality surveillance; consequently, the suitability of other indicator organisms has been evaluated by several investigators. Bacteriophages, particularly coliphages, meet most of the requirements as indicators for human pathogenic viruses, and they have been suggested as possible models for enteroviruses (Grabow, 1986; Kott et al., 1974; Stetler, 1984). Coliphages are closely associated with wastewater pollution and they generally outnumber enteric viruses in water by a factor of 1000 or more (Grabow and Coubrough, 1986; Grabow et al., 1984). Grabow and Coubrough (1986) reported a variety of somatic and F (male) specific coli-phages in water environments; F specific coli-phages are a more homogenous group than somatic coliphages and have generally a greater resistance. Havelaar and coworkers (Havelaar et al., 1984, 1986) indicate that the F specific coliphages constitute suitable surrogates for enteric viruses because their morphology and composition closely resembles that of picornaviruses e.g. poliovirus and Hepatitis A virus; they are specific for sewage pollution and are exceptionally resistant. Phages of Bacteroides fragilis have the advantage of being highly specific for fecal pollution (Jofre et al., 1986); also, their survival in inimical environments is comparable to or better than most enteric viruses, and thus they would constitute a ‘worst case’ situation, if present (Chung and Sobsey, 1993).

Evaluation of treatment of drinking and wastewater: Drinking water treatment plants generally include combinations of treatments such as alum flocculation, sand filtration, ozonation and disinfection with chlorine or chlorine dioxide to ensure conformation to the potable water quality standards. Coliphages have been used to evaluate the efficacy of drinking water treatments to inactivate/remove human enteric viruses (Jofre et al., 1995; Payment, 1991; Stetler, 1984). Reports suggest that coliphages and enteroviruses are removed at comparable rates during treatment processes and coli-phages have been suggested as indicators of water pollution and as possible models for enteroviruses (Stetler, 1984). Jofre et al. (1995) evaluated somatic and F-specific coliphages, and phages infecting Bacteroides fragilis as indicators to assess the effectiveness of drinking water treatment for pathogenic virus removal; these authors recommend bacteriophages of Bacteroides fragilis as models for monitoring water quality in drinking water plants. Similarly, bacteriophages (bacteriophages MS2 and ø2) have been used to evaluate virucidal treatments of sewage effluents (Kott et al., 1974).

Bacteriophages have been widely used as surrogates for mammalian viruses in medical and environmental virology applications. This is because the assays of choice for virus detection are infectivity-based assays and conventional mammalian virus assays have several limitations associated with them. For example, often, the viral pathogen of prime importance may remain undetected as it is present in low numbers (compared with indigenous or adventitious species) and may be difficult to detect, or alternatively, the pathogen may be difficult to cultivate in vitro. Other limitations include the complexity and expense of mammalian assays and the need for specialized equipment; also, they may be too slow for routine quality surveillance.

Bacteriophages offer several advantages: assays are infectivity-based and simple to perform; they do not require specialized equipment. Their use as surrogates circumvents the need for handling of pathogenic viruses and allows for economical and expeditious screening for the presence of pathogenic species. The wide range of bacteriophages available offers a powerful analytical tool amenable to several different applications: filter validation studies (where removal is based on size exclusion), investigations into virus contamination control issues, evaluation of barrier materials, etc. While it is intuitively appropriate to conduct validation studies with the known viral burden or a closely related mammalian species, logistic considerations often dictate otherwise. There is a considerable body of evidence to suggest and support the use of bacteriophages as surrogates for mammalian viruses especially in size-exclusion filter validation studies. Use of appropriately sized bacteriophages provides an innocuous, efficacious and expeditious method for economical testing and validation of viral clearance capabilities of virus removal filters, thus facilitating performance of filter validation studies in biopharmaceuticals under product- and process- specific conditions in an overall effort towards ensuring the virological safety of biologicals.

H. Anazawa, I.E. Lee, and K. Sakai, “HIV removal efficiency of Pall Ultipor VF DV50 Virus Removal Filter,” Pharm Tech Japan 13: 951-955 (1997).

H. Aranha-Creado and A. Herczeg, “Demonstration of clearance of murine leukemia virus by a tangential flow membrane filtration system - Pall Filtron Omega™ 300 K VR,” Scientific and Technical Report, STR-FIL (manuscript in preparation) Pall Ultrafine Filtration Co., East Hills, NY (1998).

H. Aranha-Creado, K. Oshima, S. Jafari, G. Howard, Jr., and H. Brandwein, “Virus retention by a hydrophilic triple-layer PVDF microporous membrane filter,” PDA J. Pharm. Sci. Technol. 51: 119-124 (1997).

Aranha-Creado, J. Peterson, and P. Y. Huang, “Clearance of Murine Leukemia Virus from monoclonal antibody solutions by a hydrophilic PVDF microporous membrane filter,” Biologicals, June (1998).

A.F. Bradburne, P.R. Ball, and A.J. Hunter, “Evaluation of virus removal efficiency of membrane gas filters,” Proceedings of the International Congress: Advanced Technologies for Manufacturing of Aseptic and Terminally Sterilized Pharmaceuticals and Biopharmaceuticals, Basel, Switzerland (1992).

L. V. J. Brando, “Cell culture systems,” In P.R. Murray (Editor in chief), E. J. Baron, M. A. Pfaller, F. C. Tenover, R. H. Yolken, Eds., Manual of clinical microbiology: Sixth edition, pp. 158-165, ASM Press, Washington, DC (1995).

P. L. Brown, “Protective clothing for health care workers: liquid proofness versus microbiological resistance,” In J. P. McBriarty and N. W. Henry, Eds., Performance of protective clothing: Fourth volume, ASTM STP 1133, pp.65-82, American Society for Testing and Materials, Philadelphia, PA, (1992).

R. Carey, W. Herman, B. Herman, B. Krop, and J. Casamento, “A laboratory evaluation of standard leakage tests for surgical and examination gloves,” J. Clin. Engineering 14: 133-143 (1989).

H. Chung and M. D. Sobsey, “Comparative survival of indicator viruses and enteric viruses in seawater and sediment,” Wat. Sci. Tech. 27: 425-428 (1993).

R. DeLeon, S. M. Matsui, R. S. Baric, J. E. Hermann, N. R. Blacklow, H. B. Greenberg, and M. D. Sobsey, “Detection of Norwalk virus in stool specimens by reverse transcriptase-polymerase chain reaction and nonradioactive oligoprobes, J. Clin. Microbiol. 30: 3151-3157 (1992).

J. C. Desenclos, K. C. Klontz, M. H. Wilder, O. V. Naiman, H. S. Margolis, and R. A. Gunn, “A multistate outbreak of hepatitis A caused by the consumption of raw oysters,” Am. J. Public Health 81: 1268-1272 (1991).

A. J. DiLeo, D. A. Vacante, and E. Deane, “Size exclusion removal of model mammalian viruses using a unique membrane system, Part I: Membrane qualification,” Biologicals 21: 275-286 (1993).

G. A. Erickson, S. R. Bolin and J. G. Lundgraf, “Viral contamination of fetal bovine serum used for tissue culture: risks and concerns,” Dev. Biol. Stand. 75:173-175 (1991).

J. P. Fox, C. Manso, H. A. Penna and M. Pare, “Observation on the occurrence of icterus in Brazil following vaccination against yellow fever,” Am. J. Hyg. 36: 68-116 (1942).

R. Figueroa, A. Sopulveda, M. A. Soto, and J. Teha, “Informational analysis of MS2 and øX-174 virus genomes,” J. Theor. Biol. 74: 203-207.

R.L. Garnick, “Experience with viral contamination in cell culture,” Dev. Biol. Stand. 88: 49-56 (1996).

C. P. Gerba, S. M. Goyal, R. L. LaBelle, I. Cech, and G. F. Bodgan, “Failure of indicator bacteria to reflect the occurrence of enteroviruses in marine waters,” Am. J. Public Health 69: 1116-1119 (1979).

W. O. K. Grabow, “Indicator systems for assessment of the virological safety of treated drinking water,” Wat. Sci. Tech. 18: 159-165 (1986).

W. O. K. Grabow, and P. Coubrough, “A practical direct plaque assay for coliphages in 100 ml samples of drinking water,” Appl. Environ. Microbiol. 52: 430-433 (1986).

W. O. K. Grabow, and M. Du Preez, “Comparison of m-Endo, LES, MacConkey, and Teepol media for membrane filtration counting of total coliform bacteria in water,” Appl. Environ. Microbiol. 38: 351-358 (1979).

W. O. K. Grabow, P. Coubrough, E. M. Nupen, and B.W. Bateman, “Evaluation of coliphages as indicators of the virological quality of sewage-polluted water,” Wat. S.A. 10: 7-14 (1984).

Guidelines for Sterile Drug Products Produced by Aseptic Processing, Center for Drugs and Biologics and Office of Regulatory Affairs, U.S., FDA, Washington, D.C., June 1987.

R. J. C. Harris, R. M. Dougherty, P. M. Biggs, L. N. Payne, A. P. Goffe, and A. E. Churchill, “Contaminant viruses in two live virus vaccines produced in chick cells,” J. Hyg. (Cambridge) 64: 1-6 (1966).

A. H. Havelaar, K. Furuse, and W. M. Hogeboom, “Bacteriophages and indicator bacteria in human and animal faeces,” J. Appl. Bacteriol. 60: 255-262 (1986).

A. H. Havelaar, W. M. Hogeboom, and R. Pot, “F specific RNA bacteriophages in sewage: methodology and occurrence,” Wat. Sci. Tech. 17: 645-655 (1984).

ICH harmonized tripartite guideline (1997) Viral safety evaluation of biotechnology products derived from cell lines of human or animal origin, Step 4 Adoption Recommendation, pp. 1-27, IFPMA, Geneva, Switzerland.

X. Jiang, M. K. Estes, and T. G. Metcalf, “Detection of hepatitis A virus by hybridization with single-stranded RNA probes,” Appl. Environ. Microbiol. 53: 2487-2495 (1987).

J. Jofre, A. Bosch, F. Lucena, R. Girones, and C. Tartera, “Evaluation of Bacteroides fragilis bacteriophages as indicators of the virological quality of water,” Wat. Sci. Tech. 18: 167-173 (1986).

J. Jofre, E. Olle, F. Ribas, A. Vidal, and F. Lucena, “Potential usefulness of bacteriophages that infect Bacteroides fragilis as model organisms for monitoring virus removal in drinking water treatment plants,” Appl. Environ. Microbiol. 61: 3227-3231 (1995).

M. V. K. Jones, K. Bellamy, R. Alcock, and R. Hudson, “The use of bacteriophage MS2 and a model system to evaluate virucidal hand disinfectants,” J. Hosp. Infect. 17: 279-285 (1991).

H. R. Kotilainen, W. H. Cyr, W. Truscott, N. M. Gantz, L. B. Routson, and C. D. Lytle, “Ability of 100 ml water leak test for medical gloves to detect gloves with potential for virus penetration,” In J. P. McBriarty and N. W. Henry, Eds., Performance of protective clothing: Fourth volume, ASTM STP 1133, pp. 38-48, American Society for Testing and Materials, Philadelphia, PA, (1992).

Y. Kott, N. Roze, S. Sperber, and N. Betzer, “Bacteriophages as viral pollution indicators,” Wat. Res. 8: 165-171 (1974).

R.H. Lehman, and K. P. Bansemir, “The testing of disinfectants,” Hyg. Med. 12: 9-12 (1987).

F. Lucena, J. Lasobras, D. McIntosh, M. Forcadell, and J. Jofre, “Effect of distance from the polluting focus on relative concentrations of Bacteroides fragilis phages and coliphages in mussels,” Appl. Environ. Microbiol. 60: 2272-2277 (1994).

C. D. Lytle, P. G. Carney, S. Vohra, W. H. Cyr, and L. E. Bockstahler, "Virus leakage through natural membrane condoms," Sex. Transm. Dis. 17: 58-62 (1990).

C. D. Lytle, A. P. Budacz, E. Keville, S. A. Miller, and K. N. Prodouz, "Differential Inactivation of Surrogate Viruses with Merocyanine 540," Photochemistry and Photobiol. 54: 489-493 (1991a).

C. D. Lytle, S. C. Tondreau, W. Truscott, A. P. Budacz, R. K. Kuester, L. Venegas, R. E. Schmukler, and W. H. Cyr, "Filtration Sizes of Human Immunodeficiency Virus Type 1 and Surrogate Viruses Used to Test Barrier Materials," Appl. Environ. Microbiol. 58: 747-749 (1992).

C. D. Lytle, W. Truscott, A. P. Budacz, L. Venegas, L. B. Rouston, and W. H. Cyr , "Important Factors for Testing Barrier Materials with Surrogate Viruses," Appl. Environ. Microbiol. 57: 2549-2554 (1991b).

J. Y. Maillard, T. S. Beggs, M. J. Day, R. A. Hudson, and A. D. Russell, “Effect of biocides on MS2 and K coliphages,” Appl. Environ. Microbiol. 60: 2205-2206 (1994).

T. Meltzer, “Viable and non-viable particle and pyrogen retention,” In T. Meltzer, Ed., Filtration in the pharmaceutical industry,” pp. 493-544, Marcel Dekker Inc., N.Y. (1987).

R. H. Olsen, J. S. Siak, and R. H. Gray, “Characteristics of PRD1, a plasmid-dependent broad host range DNA bacteriophage,” J. Virol. 14: 689-699 (1974).

K. H. Oshima, T. T. Evans-Strickfaden, A. K. Highsmith, and E. W. Ades," The removal of phage T1 and PP7, poliovirus from fluids with 50,000, 13,000, and 6,000 molecular weight, hollow fiber ultrafilters," Can. J. Microbiol. 41: 316-322 (1995).

H. Oshima, T. T. Evans-Strickfaden, A. K. Highsmith, and E. W. Ades, "The use of a microporous polyvinylidene fluoride (PVDF) membrane filter to separate contaminating viral particles from biologically important proteins, Biologicals 24: 137-145 (1996).

D.B. Pall, “Quality control of absolute bacteria removal filters,” Bull. PDA 29: 192-204 (1975).

P. Payment, “Elimination of coliphages, Clostridium perfringens and human enteric viruses during drinking water treatment: results of large volume sampling,” Wat. Sci. Tech. 24: 213-215 (1991).

H. Rabenau, V. Ohlinger, J. Anderson, B. Selb, J. Cinatl, W. Wolf, J. Frost, P. Mellor, and H. W. Doerr, “Contamination of genetically engineered CHO-cells by epizootic haemorrhagic disease virus (EHDV),” Biologicals 21: 207-214 (1993).

P. Roberts, “Efficient removal of viruses by a novel polyvinylidene fluoride membrane filter,” J. Virol. Methods 65: 27-31 (1997).

K. Shah and N. Nathanson, “Human exposure to SV40: review and comment,” Am. J. Epidemiol. 103: 1-12 (1976).

R.E. Stetler, “Coliphages as indicators of Enteroviruses,” Appl. Environ. Microbiol. 48: 668-670 (1984).

B. I. Truman, H. P. Madore, M. A. Menegus, J. L. Nitzkin, and R. Dolin, “Snow mountain agent gastroenteritis from clams,” Am. J. Epidemiol. 126: 516-525 (1987).

“Validation Guide for Pall Ultipor® VF™ Grade DV50 Ultipleat™ AB Style Virus Removal Filter Cartridges,” Publication TR-UDV50, Pall Ultrafine Filtration Company, Pall Corporation, East Hills, N.Y. (1995).

A. K. Vidaver, R. K. Koski, and J. L. Van Etten, “Bacteriophage f6: a lipid-containing virus of Pseudomonas phaseolicoli,” J. Virol. 11: 799-805 (1973).

B. Voeller, J. Nelson, and C. Day, “Viral leakage risk differences in latex condoms,” AIDS research and human retroviruses 10: 701-710 (1994).