Comparative Sensitivity of Various Cell Culture Systems for Isolation of Viruses from Wastewater and Fecal Samples

Transcription

1 APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Sept. 1978, p /78/ $02.00/0 Copyright i) 1978 American Society for Microbiology Vol. 36, No. 3 Printed in U.S.A. Comparative Sensitivity of Various Cell Culture Systems for Isolation of Viruses from Wastewater and Fecal Samples NATHALIE J. SCHMIDT,* HELEN H. HO, JOHN L. RIGGS, AND EDWIN H. LENNETTE Viral and Rickettsial Disease Laboratory, State of California Department of Health, Berkeley, California Received for publication 26 June 1978 In efforts to define the most sensitive cell culture systems for recovery of viruses from wastewaters, 181 samples were inoculated in parallel into tube cultures of various cell types and were plaqued in bottle and petri dish cultures of three types of monkey kidney cells. Polioviruses were recovered most frequently in the RD line of human rhabdomyosarcoma cells, group A coxsackieviruses in RD and human fetal diploid kidney (HFDK) cells, group B coxsackieviruses in the BGM line of African green monkey kidney cells, echoviruses in RD and primary rhesus monkey kidney (RhMK) cells, and reoviruses in RhMK cells. BGM cells were unsatisfactory for recovery ofviruses other than polioviruses and group B coxsackieviruses, and a line of fetal rhesus monkey kidney (MFK) was not a satisfactory substitute for primary RhMK. With RhMK cells, comparable numbers of virus isolations were made in tube cultures and in plaque assays conducted in bottle cultures, but with BGM and MFK cells, fewer isolations were made by plaquing than by inoculation of tube cultures. In comparative plaque assays on fecal samples under three different overlays in bottle and plate cultures of RhMK, BGM, and MFK cells, it was found that plaquing in the most sensitive system, RhMK, was less efficient for virus recovery than was inoculation of tube cultures of RhMK or HFDK cells. Overall, plaque assays performed in petri dishes in a CO2 incubator yielded fewer virus isolates than did parallel plaque assays performed in closed bottle cultures. Other limitations of plaque assays for recovery of human enteric viruses are discussed. 480 Comparative evaluation of the sensitivity of various cell culture systems for detection of naturally occurring viruses in water and wastewater has been assigned a high priority for research by workers involved in wastewater reclamation. At present it is recognized that no single cell culture system is suitable for recovery of all of the common human enteric viruses which may be present in wastewaters, and that virus isolation attempts with plaque assays or tube cultures under a fluid medium each have limitations (8, 9). In efforts to better define satisfactory and sensitive host cell culture systems for isolation of viruses from wastewater samples, comparative studies were conducted in which wastewater samples from Water Factory 21 in Orange County, Calif., were inoculated in parallel into a variety of cell types in tube cultures under fluid medium and were also plaqued in both bottle and petri dish cultures of three different types of monkey kidney cells. A second phase of the studies consisted of examining the effect of different media and culture vessels on the sensitivity of plaque assays for enteroviruses likely to be recovered from water and sewage samples. MATERLALS AND METHODS Cell cultures. Primary rhesus monkey kidney cells (RhMK) were dispersed from kidney tissue and grown into monolayer cultures by the standard procedures of this laboratory (N. J. Schmidt, in E. H. Lennette and N. J. Schmidt, ed., Diagnostic Procedures for Viral, Rickettsial and Chiamydial Infections, in press). A line of fetal rhesus monkey kidney (MFK) cells was established in this laboratory by procedures described elsewhere (Schmidt, in press); the cells were used in the present study at the 69th to 75th passage levels, at which time they were no longer diploid in their chromosome configuration. The BGM line of African green monkey kidney cells (1) was obtained from Theodore Metcalf and was used in the present studies at passage levels of 127 to 140. Human fetal diploid kidney (HFDK) cell lines were established and propagated in this laboratory by standard procedures (Schmidt, in press). The RD line of human rhabdomyosarcoma cells (3) was propagated as previously described (6). Fluid maintenance medium used on tube cultures of cells other than RD consisted of 2 mil of 2% fetal bovine serum and 98% Eagle minimal essential me-

2 VOL. 36, 1978 dium (MEM) prepared in Earle balanced salt solution. RD cells were maintained on 2% fetal bovine serum and 98% Leibovitz medium no. 15, and the medium was replaced at 7 days. Maintenance media contained 200 U of penicillin, 200 lag of streptomycin, 50 ug of gentamicin, and 1 ilg of amphotericin B per ml. Plaque assays. For plaque assay of wastewater samples in bottle or plate cultures, the overlay medium described by Dahling et al. (2) was used; this consisted of Eagle MEM in Hanks salts supplemented with 2% fetal bovine serum, 1% of a 10Ox solution of nonessential amino acids (Grand Island Biological Co.), 1% sterile, whole milk, 3% of a 7.5% solution of NaHCO3, 1% of a 1% solution of MgCl2 6H20, 1.5% of a 1:1,000 dilution of neutral red, and 1.8% of lonagar no. 2 (Oxoid). In comparative studies on plaquing fecal specimens, the three overlay media used in parallel were (A) the medium of Dahling et al. (2); (B) a modification of the enterovirus plaquing medium of Wallis and Melnick (10) consisting of Eagle MEM in Earle salts supplemented with 1.8% fetal bovine serum, 1.5% of a 1:1,000 dilution of neutral red, and 1.5% Difco purified agar, and (C) Eagle MEM in Earle salts supplemented with 5% fetal bovine serum, 1.5% of a 1:1,000 dilution of neutral red, and 1.2% SeaPlaque agarose. Overlay media contained the same antibiotics used for maintenance media on tube cultures. Cultures in 2-ounce (ca. 60-ml) bottles received 5 ml of overlay medium, and those in 30-mm cups (Linbro FB-6 plates) received 3 ml. Closed bottles were incubated at 36 C in a dry incubator, and plates were incubated at the same temperature in a CO2 incubator. Plaque counts were recorded at 7 and 14 days. The virus content of plaques was confirmed by subpassaging into tube cultures of the same cell type and observing for a cytopathic effect. Materials eamined A total of 181 wastewater samples of various sorts was obtained from Water Factory 21 in Orange County, Calif. (Table 1). For the most part these were concentrates from various filter absorption-elution procedures and also a few grab samples. The plant influent was not raw sewage, but rather secondary effluent; the concentrates represented 50-gallon (ca liter) samples reduced to a volume of 50 to 100 ml, or 1- to 2-gallon samples concentrated by the same factor by flocculation. Because the purpose of this study was not to evaluate various sampling and concentration procedures or the efficacy of sewage treatment methods for inactivation of vinrs, the various types of wastewater samples were grouped together for evaluation of the host cell systems for virus recovery. Some preliminary studies were performed on wastewater samples which had previously been examined by the Virus Laboratory at the Los Angeles County Health Department. These consisted almost entirely of ultrafiltrates of untreated or chlorinated primary or secondary effluents. Wastewater samples were treated with gentamicin at a concentration of 100 plg/ml and amphotericin B at a concentration of 1 gtg/ml before they were inoculated into cell cultures. For studies on the effect of composition of the overlay medium and type of culture vessel on plaquing efficiency of enterovinue, fecal samples which had yielded enterovirues on routine testing were used. VIRUS RECOVERY FROM WASTEWATERS 481 Suspensions were prepared by our standard procedure (Schmidt, in press). As a control on the virus content of the specimens at the time of plaquing, the suspensions were also inoculated into tube cultures of RhMK and HFDK cells. Isolation and identification of viruses. Each specimen was inoculated in a volume of 0.25 ml into duplicate tube, bottle, or plate cultures containing 1 ml of Eagle MEM with 5% fetal bovine serum. After incubation at 36 C for 2 h, the medium was removed and replaced with maintenance medium or plaquing overlay medium. Cultures were observed for cytopathic effects or plaques over a period of 14 days. Vinrses were identified by neutralization tests against enterovirus immune serum pools performed in microtiter plates (Schmidt, in press). Reoviruses were grouped by neutralization tests with type 1 antiserum and typed by hemagglutination inhibition tests against antisera produced in geese. RESULTS Toxicity of wastewater samples for cell cultures. Samples previously examined by the Los Angeles County Health Department were used for preliminary studies on the development of cell culture procedures for examination of wastewaters. Of 60 specimens, 16 had been toxic to cell cultures in the Los Angeles laboratory, but none was toxic in tube cultures of RhMK, MFK, BGM, or HFDK cells in our examinations. The possibility was considered that absorption of the samples in the presence of 5% fetal bovine serum followed by removal of the inoculum had reduced the toxicity of the samples. Further testing confirmed that when wastewater samples were absorbed in the absence of fetal bovine serum or when the inoculum was not removed before the addition of maintenance medium, the samples were frequently toxic to the cells. With the procedure described above for inoculating cell cultures, none of the 181 wastewater samples from Water Factory 21 was toxic for tube cultures. Certain wastewater samples appeared to contain small particulate toxic material which was localized by plaquing overlays and produced necrotic areas resembling virus plaques in the cell monolayers, but virus could not be recovered from these foci of necrosis. This toxic effect occurred more frequently in BGM cells than in primary RhMK; for example, of 18 samples from the Los Angeles County laboratory which were negative for virus in all host systems, 6 produced "false plaques" in BGM cells, and only 1 of these also produced necrotic foci in RhMK cells. Recovery of viruses from wastewater samples. Of the 181 wastewater samples from Water Factory 21, 64 yielded virus in a cell culture system. Table 1 summarzes the virus recoveries from the various types of samples. Of

3 482 SCHMIDT ET AL. APPL. ENVIRON. MICROBIOL. the 61 positive influent samples, a single virus type was isolated from 34, two virus types were isolated from 15, three virus types were isolated from 6, four virus types were isolated from 2, five virus types were isolated from 1, six virus types were isolated from 2, and seven types were recovered from 1 specimen. Each of the three positive specimens taken after lime treatment TABLE 1. Summary of virus recoveries from various types of wastewater samples from Water Factory 21, Orange County, Calif. Virus recovery Type of sample No. tested No. of No. of samples samples positive negative Plant influent Sample after lime treatment Sample after NH stripping Sample after chlo rination Virus~' TABLE 2. yielded a single virus type. Although the positive specimens contained a variety of virus types, and frequently more than one type, the concentration of each virus appeared to be low, as evidenced by the fact that each virus was usually recovered in only a single set of tube or plaque cultures and sometimes in only a single culture of the pair. Only 3 of the samples yielded a given virus in three host systems, and 19 yielded a virus in two host systems; the remaining isolations were made only in a single host system. Specimens from which virus was isolated by plaquing produced very few plaques. Twenty-eight of the positive specimens produced only a single virus plaque, nine produced two plaques, three produced three plaques, and one produced five plaques. Table 2 shows the virus types isolated from the wastewater samples in each of the host cell culture systems. RhMK cells in tubes and in plaquing bottles yielded comparable numbers of isolates, but with the monkey kidney cell lines, tube cultures were more sensitive than plaque assays. It is noteworthy that it was only for a Viruses recovered in various host cell systems from wastewater samples No. of specimens positive in: No. of RhMK MFK BGM specimens Plaque Plaque Plaque HFDK RD positive (Tubes) (Tubes) Tubeue ue Bot- Plates Bot- Plates Btl- Plates P P P CA CA CA CA CA CB CB CB CB CB6 2 2 El E3 1 1 E E E9 1 1 Ell 4 4 E E E E E Reo Reo Unidenti- 1 1 fied Totals ' P, Poliovirus; CA, group A coxsackievirus; CB, group B coxsackievirus; E, echovirus; Reo 1, reovirus type 1; Reo, reovirus type 2.

4 VOL. 36, 1978 few echovirus types, namely 1, 7 and 8, that RhMK plaque assays in bottles recovered more viruses than were isolated in RhMK tube cultures. Plaque assays conducted in plates in a CO2 incubator were markedly less sensitive for virus isolation than were assays performed in closed bottle cultures, except in the case of tests performed in BGM cells. The viruses isolated by plaquing in BGM cells consisted solely of polioviruses and group B coxsackieviruses. Table 3 compares the isolation rates of the major groups of enteric viruses in each type of host cell. Polioviruses were isolated most frequently in RD cells, group A coxsackieviruses were isolated most frequently in HFDK and RD cells, group B coxsackieviruses were isolated most frequently in BGM cells, echoviruses were isolated most frequently in RhMK and RD cells, and reoviruses were isolated most frequently in RhMK and MFK cells. Effect of overlay media and culture vessels on the sensitivity of enterovirus isolation by plaque assay. Because adequate volumes of wastewater samples were not available for extensive studies comparing the efficiency of TABLE 3. Comparative sensitivity of various host cell cultures for recovery of viruses from wastewater samples No. of No. of % Posi- Virus group Cell isolation positive tive type isoat- isolation lation attempts" attempts tempts Polioviruses RhMK MFK BGM HFDK RD Group A RhMK coxsackie- MFK viruses BGM HFDK RD Group B RhMK coxsackie- MFK viruses BGM HFDK RD Echoviruses RhMK MFK BGM HFDK RD Reoviruses RhMK MFK BGM HFDK RD "Each isolation attempt consisted of inoculating a sample into two tube, bottle, or plate cultures of the cell type. VIRUS RECOVERY FROM WASTEWATERS 483 enterovirus isolation under different overlay media and in bottle and plate cultures, the studies were done on fecal samples from which enteroviruses had been isolated by our diagnostic laboratory. Each specimen, as a 20% suspension, was inoculated in parallel into 2-ounce bottle cultures and 30-mm plate cultures of RhMK, MFK, and BGM cells, and duplicate cultures of each cell type were overlaid with the three different media described above. Table 4 shows the isolations of each virus type in bottle cultures under the three overlay media. Of 21 strains of polioviruses and coxsackieviruses, 20 were reisolated in tube cultures of RhMK, 8 were also recovered in HFDK, and 1 (a poliovirus) was reisolated only in HFDK. However, in the most efficient plaquing systems for these virus types (RhMK and BGM), only 11 to 16 of the isolation attempts were positive. Of 27 echovirus strains, 20 were reisolated in tube cultures of RhMK, and all 27 were reisolated in HFDK tubes. Only 16 or 17 of these specimens yielded virus in the most sensitive plaquing system, RhMK. MFK and BGM were very insensitive for recovery of echoviruses by plaquing. These studies furnished additional evidence that virus isolation attempts performed by plaquing are less efficient for recovery of enteroviruses, particularly echoviruses, than are tests done in tube cultures. Only 32 of the 40 specimens which yielded isolates in tube cultures of RhMK were positive by plaquing in this cell type. Virus isolation rates were similar under each of the three overlay media; 56 isolations were made with overlay A, 52 were made with overlay B, and 55 were made with overlay C. In Table 5 the plaquing efficiency of enteroviruses in fecal samples in each cell culture system is further compared by showing the average plaque counts obtained for those positive specimens which produced plaques sufficiently dispersed to permit enumeration. Some of the specimens included in the preceding table produced plaques too numerous to count. Although the isolation rates for polioviruses and group B coxsackieviruses were similar in RhMK and BGM (Table 4), it is seen that average plaque counts of samples containing polioviruses were about twice as high in BGM cells as in RhMK cells, and samples containing group B coxsackieviruses produced five to nine times as many plaques in BGM cells as in RhMK. On the other hand, coxsackievirus type A9 failed to plaque in BGM cells, and only a single specimen containing an echovirus produced plaques in BGM. Table 6 compares the isolation of enteroviruses from fecal samples in bottle cultures and

5 484 SCHMIDT ET AL. TABLE 4. Comparison of the recovery of enteroviruses from fecal samples by inoculation into tube cultures and by plaquing in bottle cultures under three different overlay media No. of specimens positive in tube No. of positive isolations by plaque assay in bottle cultures of: No. of cultures Virus type specimens RhMK, overlay': MFK, overlay: BGM, overlay: RhMK HFDK A B C A B C A B C Poliovirus 1, 2, or Coxsackievirus A ** Coxsackievirus B Coxsackievirus B Coxsackievirus B Coxsackievirus B Subtotal Echovirus Echovirus Echovirus Echovirus Echovirus Echovirus Echovirus Echovirus Echovirus Echovirus Echovirus Echovirus Subtotal Grand total "See text for composition of overlay media. Blank space indicates negative results. in plate cultures for the 22 specimens which were positive by plaque assay and which were tested in parallel in the two types of culture vessels. For each of the virus groups and each of the cell types, fewer isolations were made in plate culture than in bottle cultures. DISCUSSION Results of these studies further emphasize that no single cell culture system is satisfactory for recovery of naturally occurring viruses from wastewater and fecal samples, and combinations of cell culture systems must be employed for maximum efficiency in monitoring wastewaters. RhMK, or other primary monkey kidney cells of comparable sensitivity, are clearly needed for recovery of reoviruses and many of the echoviruses. RhMK recovered certain echoviruses, notably types 1, 7, 8 and 14, which were not isolated in HFDK and RD cells. The MFK cell line was not a satisfactory substitute for RhMK in virus isolation, but it can be used to good advantage for identification of isolates from RhMK cells, and thus can be of great value in conserving RhMK, which is becoming increasingly expensive and difficult to obtain. APPL. ENVIRON. MICROBIOL. The BGM line of continuous African green monkey kidney cells, which has been highly recommended by some workers for recovery of viruses from water and sewage samples (2), was found in our comparative studies to be completely unsatisfactory for isolation of naturally occurring echoviruses and group A coxsackieviruses and less sensitive than RhMK for recovery of reoviruses. A previous study (7) from this laboratory using BGM cells from a commercial source found them to be markedly less sensitive than RhMK and HFDK cells for isolation of echoviruses from clinical specimens. Cells for the present study were obtained from another laboratory where they were found to be highly sensitive for polioviruses (T. G. Metcalf, personal communication). In our studies these BGM cells were as satisfactory as RhMK cells for isolation of polioviruses and, based upon the recovery of group B coxsackieviruses from wastewater samples and the higher plaque counts obtained with group B coxsackieviruses from fecal specimens, they would appear to be somewhat more sensitive than RhMK for isolating viruses of this group. Although HFDK cells did not recover a large

6 VOL. 36, 1978 number of viruses from the wastewater samples, this may have been a reflection of the virus types present in the community at the time, because previous (4, 5) and continuing experience with these cells has clearly shown that they are more sensitive than RhMK for isolating certain echovirus and group A coxsackievirus types. The RD cell line was included in our compar- TABLE 5. Comparison of the plaquing efficiency of enteroviruses from fecal samples in bottle cultures of RhMK, MFK, and BGM under three different overlay media No. of Avg PFU/sample under over- Virus type mens- Cells lay': tested A B C Poliovirus 1, 2,or 4 RhMK MFK BGM Coxsackievirus 6 RhMK B2, B4, or B5 MFK BGM Coxsackievirus 3 RhMK A9 MFK BGM Echovirus type 4, 21 RhMK , 8, 9, 11, 18, MFK ,21,30,31,or BGM <lh 'See text for composition of overlay media. PFU, plaque-forming units. ' One plaque containing echovirus type 18 was produced. TABLE 6. VIRUS RECOVERY FROM WASTEWATERS 485 ative studies because of its ability to support the replication of certain group A coxsackieviruses which previously could be grown only in suckling mice (6). Although only a few group A coxsackieviruses were isolated, the RD cells recovered a surprisingly wide variety of other enteroviruses, including a number of echovirus types which previously have not been recovered in continuous cell lines. Overall, the isolation rate in RD cells was as high as that in RhMK, and thus they were shown to be a valuable addition to the spectrum of host cells used for recovery of viruses from wastewater samples. The use of plaquing has been fairly widely recommended for isolation of viruses from wastewater samples, because plaque assays have the potential of permitting enumeration of infectious virus particles and of separating mixtures of viruses. However, in our comparative studies on wastewater samples, and also on fecal samples containing enteroviruses, plaque assays were much less sensitive than inoculation oftube cultures under fluid medium for recovery of enteric viruses. A number of the echoviruses and reoviruses which produced a cytopathic effect in cell cultures with a fluid medium failed to produce plaques in the same type of cells under a solid overlay. Other problems also limited the reliability of plaque assays for isolation of viruses. Some wastewater samples apparently contained small particulate toxic material which was localized by the solid overlay and produced necrotic areas in the cell monolayers which resembled virus plaques, but from which virus was not recoverable. This effect occurred most frequently in Comparison of the recovery of enteroviruses from fecal specimens by plaquing in bottle cultures and in plate cultures No. of positive isolations in: Viru types No. of positive Cell type Bottle cultures, overlay": Plate cultures, overlay: specimens A B C A B C Polioviruses 4 RhMK MFK BGM Coxsackievirus A9 2 RhMK MFK BGM Group B coxsackie- 7 RhMK viruses MFK 2 BGM Echoviruses 9 RhMK MFK BGM 1 "See text for composition of overlay media.

7 486 SCHMIDT ET AL. BGM cells, and emphasized the need to confirm the viral identity of plaques seen in isolation attempts. Otherwise, the erroneous interpretation might be made that BGM cells give higher virus plaque counts than other cells less susceptible to this toxic effect. Another problem with plaque assays for virus isolation lies in the fact that with heavily infected samples, such as fecal specimens, the rapid and extensive viral destruction of the cells may prevent recovery of infectious virus from the plaques, presumably due to the localization of virus in areas without viable cells for continued viral replication and the destruction of viral infectivity at incubator temperatures. The wastewater samples which we tested did not require plaquing for separation of different naturally occurring viruses. Although a wide variety of viruses was present in some samples, the concentration of each was low, as evidenced by the fact that mixtures of viruses were seldom recovered from a single culture, that virus was often isolated from only a single culture of a set, and that when plaquing was successful for virus isolation, the positive cultures showed very few plaques. Rather than attempting to isolate viruses from wastewater samples by plaque assay, it might be more rational to extend the number and types of tube cultures employed. In comparing different overlay media for plaquing enteroviruses which occur in fecal and wastewater samples, it was found that comparable viral isolation rates were obtained with the three overlay media tested and that a relatively simple medium without additives and containing agarose as a solidifying agent was as satisfactory as a more complex medium. In comparing plaque assays performed in parallel in bottle cultures and in plate cultures in a CO2 incubator, the surprising finding was made that plaque assays in plate cultures yielded far fewer viruses from wastewater and fecal samples than did assays in closed bottle cultures, particularly in the case of echovirus isolations. The reasons for this are not apparent, but the results APPL. ENVIRON. MICROBIOL. would indicate that with the methods employed the simpler plaque assays in plates in a CO2 atmosphere are not adequately sensitive for some field strains of human enteroviruses. ACKNOWLEDGMENTS We are indebted to Koichi Nakamura and Kent Dupuis for assistance in these studies and to Lawrence Y. C. Leong, Aiko Butsumyo, and Richard Barnes for providing us with wastewater samples. These studies were conducted under Interagency Agreement no. W between the California Water Resources Board and the California Department of Health. LITERATURE CITED 1. Barron, A. L., C. Olshevsky, and M. M. Cohen Characteristics of the BGM line of cells from African green monkey kidney. Arch. Virusforsch. 32: Dahllng, D. R., G. Berg, and D. Berman BGM, a continuous cell line more sensitive than primary rhesus and African green kidney cells for the recovery of viruses from water. Health Lab. Sci. 11: McAllister, R. M., J. Melnyk, J. Z. Finklestein, E. C. Adams, Jr., and M. B. Gardner Cultivation in vitro of cells derived from a human rhabdomyosarcoma. Cancer 24: Schmidt, N. J Tissue culture in the laboratory diagnosis of viral infections. Am. J. Clin. Pathol. 57: Schmidt, N. J., H. H. Ho, and E. H. Lennette Comparative sensitivity of human fetal diploid kidney cell strains and monkey kidney cell cultures for isolation of certain human viruses. Am. J. Clin. Pathol. 43: Schmidt, N. J., H. H. Ho, and E. H. Lennette Propagation and isolation of group A coxsackieviruses in RD cells. J. Clin. Microbiol. 2: Schmidt, N. J., H. H. Ho, and E. H. Lennette Comparative sensitivity of the BGM cell line for isolation of enteric viruses. Health Lab. Sci. 13: Sigel, M. M., D. F. Rippe, A. R. Beasley, and M. Dorsey, Jr Systems for detecting viruses and viral activity, p In G. Berg, H. L. Bodily, E. H. Lennette, J. L. Melnick, and T. G. Metcalf (ed.), Viruses in water. American Public Health Association, Inc., Washington, D.C. 9. Sobsey, M. D Methods for detecting enteric viruses in water and wastewater, p In G. Berg, H. L. Bodily, E. H. Lennette, J. L. Melnick, and T. G. Metcalf (ed.), Viruses in water. American Public Health Association, Inc., Washington, D.C. 10. Wallis, C., and J. L. Melnick Mechanism of enhancement of virus plaques by cationic polymers. J. Virol. 2: