Antigenic Relationships among Autonomous Parvoviruses

J. gen. Virol. (1986), 67, 2839-2844. Printed in Great Britain 2839 Key words: antigenic relationships/autonomous parvoviruses Antigenic Relationships among Autonomous Parvoviruses ByW. L. MENGELING, l* P. S. PAUL, 2 T. O. BUNN 3 AND J. F. RIDPATH 1 1USDA, ARS, National Animal Disease Center, P.O. Box 70, Ames, Iowa 50010, 2Iowa State University, Veterinary Medical Research Institute, Ames, Iowa 50011 and 3 USDA, APHIS, VS, National Veterinary Services Laboratories, P.O. Box 844, Ames, Iowa 50010, U.S.A. (Accepted 7 August 1986) SUMMARY The antigenic relatedness of minute virus of mice (MVM), Kilham rat virus (KR), H- 1 virus (H- 1), haemorrhagic encephalopathy of rats virus (HER), porcine parvovirus (PPV), canine parvovirus (CPV), feline panleukopenia virus (FPV), goose parvovirus (GPV) and bovine parvovirus (BPV) was studied by immunofluorescence microscopy (FA) and by serum neutralization (SN). An antigenically related group comprising MVM, KR, HER, PPV, CPV and FPV was recognized by FA and most reactions within the group were reciprocal. Antigenic relatedness was less evident when the same viruses and antisera were tested by SN. Only CPV and FPV were closely and reciprocally related. Other cross-reactions by SN were quantitatively minor and included neutralization of CPV and FPV by pig anti-ppv serum and neutralization of H-1 and HER by rat anti-kr serum. Neither FA nor SN revealed any antigenic relationship of BPV and GPV either with each other or with any of the other viruses tested. Although most of the autonomous parvoviruses have a single natural host species, several have been shown to be antigenically related. For a few, namely, canine parvovirus (CPV), feline panleukopenia virus (FPV) and mink enteritis virus (MEV), the relationship is extensive and has been demonstrated by a variety of serological tests, including those with a high degree of stringency (Parrish & Carmichael, 1983). For others, however, relationships are more limited, at least with regard to the types of serological tests by which they can be clearly demonstrated. For example, cross-reactivity between CPV and porcine parvovirus (PPV) and their respective antisera is evident by immunofluorescence microscopy (FA) but it is weak and variable by serum neutralization (SN) and haemagglutination inhibition (HI) (Mengeling et al., 1983). Similarly, minute virus of mice (MVM) and two rat parvoviruses, namely Kilham rat virus (KR) and H-I virus (H-I), appear unrelated by SN, HI and complement fixation tests, but a one-way cross between antisera for the rat viruses and antigens of MVM has been demonstrated by indirect FA (Cross & Parker, 1972). The involvement of non-structural proteins in such reactions is suggested by the finding that non-structural proteins of MVM are precipitated by antisera for several autonomous parvoviruses, whereas structural proteins are precipitated only by anti-mvm serum (Cotmore et al., 1983). The more recent of the findings described above (Cotmore et al., 1983; Mengeling et al., 1983) have raised a question as to whether antigenic relationships among the autonomous parvoviruses are more common than suggested by earlier reports (Siegl, 1976). In this study, antisera prepared in the natural host of each of a selected group of autonomous parvoviruses were tested qualitatively and quantitatively by FA and SN for both homologous and heterotogous reactivity. For two of the selected viruses (PPV, CPV) the reactivity of antisera prepared in their natural hosts was compared to that of antisera prepared in mice by immunization with purified virus. 0000-7192 1986 SGM

2840 Short communication Viruses and their sources were as follows: CPV and FPV were from stocks maintained at the National Veterinary Services Laboratories, Ames, Iowa, U.S.A. ; MVM, KR, H-I, haemorrhagic encephalopathy of rats virus (HER) and goose parvovirus (GPV) were from the American Type Culture Collection, Rockville, Md., U.S.A.; PPV and bovine parvovirus (BPV) were from stocks maintained at this Center. Viruses were propagated in cell types selected for their known susceptibility, namely, foetal rat fibroblasts for MVM, KR, H-1 and HER, Crandall feline kidney cells for CPV and FPV, goose fibroblasts for GPV, foetal porcine kidney cells for PPV, and foetal bovine spleen cells for BPV. Antisera for MVM, KR, PPV, CPV, FPV, GPV and BPV were prepared in the natural host of the respective virus. With one exception, all animals were either gnotobiotic (pig, calf) or from a known or reportedly parvovirus-free colony (mouse, rat, dog, cat, goose). The exception was that an additional antiserum for BPV was prepared in a calf from a conventionally maintained herd. Pre-exposure serum samples were collected just before one (mouse, rat) dose, or the first of two (pig, dog, cat, calves) or three (goose) doses, of at least 104 median cell culture infective doses (CCIDso) of virus were administered oronasally (pig, dog), oronasally and intraperitoneally (mouse, rat, cat), intravenously (goose), or oronasally, intravenously and subcutaneously (calves). Multiple exposures were mostly at 2-week intervals. In general, the immunization schedules were based on previous experience in producing high titred antisera for each of the viruses. Each of the antisera was prepared at a different time and all animals were kept in isolation during immunization. Antisera were collected 1 to 4 weeks after the last (or only) administration of virus. Additional antisera for PPV and CPV were prepared in mice. About 1010 CCIDso of concentrated and purified virus (Hallauer & Kronauer, 1965; Tattersall et al., 1976) were administered intraperitoneally and again, 18 days later, intravenously. Sera were collected just before the first exposure to virus and 3 days after the last exposure to virus. To determine antiviral FA activity, sera were reacted with acetone-fixed, virus-infected cell cultures that had been prepared on 10 x 35 mm coverslips in Leighton tubes. All sera were screened at a 1 : 5 dilution and all positive sera were subsequently titrated in twofold increments starting with an initial 1 : 5 dilution. Goose sera were tested by direct FA by labelling them with fluorescein isothiocyanate. All other sera were tested by indirect FA with appropriate fluorescein isothiocyanate-labelled antispecies immunoglobulin sera (Cappel Laboratories, West Chester, Pa., U.S.A.). The FA titre was recorded as the reciprocal of the maximum dilution of serum that resulted in unequivocal specific antiviral fluorescence. To determine neutralization titres, sera were diluted in twofold increments starting with an initial 1 : 10 dilution. Each dilution was mixed with an equal volume of virus containing about 500 CCIDs0/0.1 ml and the mixture was incubated at 37 C for 2 h. After incubation, 0-2 ml of the virus-serum mixture was added to the nutrient medium of each of two freshly seeded cell cultures on 10 35 mm coverslips in Leighton tubes. Controls included two replicate cultures to which an equal amount of virus without serum was added and two replicate cultures to which neither virus nor serum was added. All cultures were subsequently incubated at 37 C. Between 72 and 96 h later, when the uninfected control cultures had reached confluence, all cultures were fixed for subsequent examination by either direct or indirect FA. The SN titre was recorded as the reciprocal of the maximum dilution of serum that reduced the number of infected cells by more than 99 ~o when compared to infected controls. Many of the autonomous parvoviruses studied were found to have one or more common antigens that were evident by FA. The related group included MVM, KR, H-l, HER, PPV, CPV and FPV (Table 1, Fig. 1). Antisera produced in the natural hosts for KR, PPV, CPV and FPV reacted with int.racellular antigen(s) of all members of the related group, whereas mouse anti-mvm serum reacted only with MVM and the other rodent parvoviruses, namely, KR, H-1 and HER. In general, higher FA titres were associated with homologous reactions. There was no evidence that GPV or BPV were antigenically related either to each other or to any of the other viruses tested. The FA and SN titres of the anti-bpv serum produced in the calf from a conventionally maintained herd (Tables 1 and 2) were about twofold higher than those of anti- BPV serum produced in a gnotobiotic calf (data not shown) but the results were otherwise similar. In contrast to the extensive cross-reactivity of pig anti-ppv serum and dog anti-cpv

Short communication 2841 Fig. 1. Reactivity of selected antiviral sera and autonomous parvoviruses by indirect immunofluorescence microscopy. Cell cultures were infected with PPV (a to c) or MVM (d to f). Infected cultures were then tested against pig anti-ppv serum (a, d), mouse anti-ppv serum (b, e) or mouse anti-mvm serum (c, f). All antiviral sera were run at a 1:5 dilution. serum, mouse anti-ppv and mouse anti-cpv sera reacted with only PPV, and CPV and FPV, respectively. Antigenic relatedness was less evident when the same antisera and viruses were tested by SN (Table 2). Only CPV and FPV were closely related. Additional cross-reactions were confined to neutralization of CPV and F P V by pig anti-ppv serum and neutralization of H-I and H E R by rat a n t i - K R serum. None of the pre-exposure sera was reactive by either F A or SN with any of the viruses included in the study. Results of this study clearly establish the existence of common antigens among m a n y of the autonomous parvoviruses. Although our observations may appear contrary to those of several previous reports (Siegl, 1976), it is important to note that most of the earlier work focused on species differentiation and involved relatively stringent serological tests such as SN and HI.

2842 Short communication Table 1. Antigenic relatedness of selected autonomous parvoviruses tested by FA* t" ~k Antiserum MVM KR H-1 HER PPV CPV FPV BPV GPV Mouse anti-mvm 2560 20 5. 5 < 5 < 5 < 5 < 5 < 5 Rat anti-kr 320 5120 640 5120 160 160 80 <5 <5 Pig anti-ppv 80 80 40 80 2560 40 40 < 5 < 5 Mouse anti-ppv < 5 < 5 < 5 < 5 640 < 5 < 5 < 5 < 5 Dog anti-cpv 160 160 320 320 160 1280 640 <5 <5 Mouse anti-cpv <5 <5 <5 <5 <5 80 40 <5 <5 Cat anti-fpv 80 160 80 80 40 640 320 <5 <5 Calf anti-bpv <5 <5 <5 <5 <5 <5 <5 160 <5 Goose anti-gpv <5 <5 <5 <5 <5 <5 <5 <5 640 * Titres are expressed as the reciprocal of the maximum dilution of serum that resulted in unequivocal, specific antiviral fluorescence; < 5, no reaction at lowest dilution of serum tested. All of the pre-exposure sera were < 5 for all of the autonomous parvoviruses included in the study. Virus Table 2. Antigenic relatedness of selected autonomous parvoviruses tested by SN* Virus A Antiserum MVM KR H-1 HER PPV CPV FPV BPV GPV Mouse anti-mvm 1280 < 10 < 10 < 10 < 10 < 10 < 10 < 10 < 10 Rat anti-kr < 10 640 10 10 < 10 < 10 < 10 < 10 < 10 Pig anti-ppv < 10 < 10 < 10 < 10 160 40 40 < 10 < 10 Mouse anti-ppv < 10 < 10 < 10 < 10 10 < 10 < 10 < 10 < 10 Dog anti-cpv < 10 < 10 < 10 < 10 < 10 10240 1260 < 10 < 10 Mouse anti-cpv < 10 < 10 < 10 < 10 < 10 320 320 < 10 < 10 Cat anti-fpv <10 <10 <10 <10 <10 5120 10240 <10 <10 Calf anti-bpv <10 <10 <10 <10 <10 <10 <10 40 <10 Goose anti-gpv < 10 < 10 < 10 < 10 < 10 < 10 < 10 < 10 2560 * Titres are expressed as the reciprocal of the maximum serum dilution that reduced infectivity by more than 99~ when compared to infected controls; < 10, infectivity not reduced more than 99~ with the lowest dilution of serum tested. All of the pre-exposure sera were < 10 for all of the autonomous parvoviruses included in the study. Both of these tests recognize epitopes associated with the final configuration of viral proteins presented on the surface of the virion. Virus-coded non-structural proteins have been recognized for several of the autonomous parvoviruses and are probably a common feature of the genus (Bloom et al., 1982; Cotmore et al., 1983; Lederman et al., 1984; Mitra et al., 1983; Matsunaga & Matsuno, 1983; Molitor et al., 1985; Paradiso, 1984; Pintel et al., 1983). It is possible that for at least some of the autonomous parvoviruses, common antigens are associated mainly with non-structural proteins. The most supportive evidence is the study of Cotmore et al. (1983) in which non-structural proteins of MVM were precipitated by anti-mvm serum and antisera for several other autonomous parvoviruses, whereas structural proteins of the virus were precipitated only by anti-mvm serum. An association of common antigens with non-structural proteins may also explain why antisera for PPV and CPV raised in their natural hosts were broadly cross-reactive by indirect FA whereas those raised in mice to purified virions had little or no cross-reactivity when tested similarly (Table 1). We assume that mice do not support replication of either PPV or CPV and that their immune system was therefore exposed only to structural proteins of the virion. Exposure may even be limited to epitopes on the surface of the virion since in an earlier study it was found that all of 90 monoclonal antibodies produced to purified PPV had some degree of neutralizing activity (Mengeling et al., 1984). Conversely, replication of virus in its natural host

Short communication 2843 should provide exposure to the full complement of viral antigens as well as a longer and perhaps enhanced response to relatively minor epitopes. An augmented range and magnitude of immune responses in the natural host could also explain why mouse (Table 1), but not rat (Cross & Parker, 1972) anti-mvm serum reacted with rat parvoviruses KR and H-1. Based on the above observations, it is tempting to speculate that with the exception of CPV, FPV and MEV, common antigens are limited mainly to non-structural proteins. However, we cannot as yet exclude the existence of epitopes associated with the structural protein pool in an infected cell that are either lost or no longer accessible in the final configuration of the virion or are simply not functional in serological tests such as SN. Moreover, our earlier finding with CPV and PPV (Mengeling et al., 1983) and the present findings (Table 2) indicate that there are weak but definite antigenic relationships among some other autonomous parvoviruses that can be demonstrated by SN under appropriate conditions. In such cases, responsible epitopes may be only a minor component of the virion or only marginally related. From reciprocal testing, it appears that BPV and GPV are antigenically distinct from each other and from all of the other autonomous parvoviruses of lower animals included in our study. However, a suggestion that they may be members of a second antigenically related group, which may also include a human parvovirus, has been provided by the results of a separate study in which some human sera were found to have antibody for both BPV and GPV (Mengeling & Paul, 1986). The combined presence or absence of antibody to BPV and GPV in individual human sera suggested that such antibody had been raised to a single agent. The collective results of this and earlier studies emphasize the complex nature of the antigenic relationships among the autonomous parvoviruses. In at least some cases, it appears that the relationship is due to multiple epitopes which may be shared to a variable extent among different members of a related group. We are currently attempting to produce monoclonal antibodies as probes to characterize such epitopes further. The authors acknowledge Ms Ann Vorwald for technical assistance. REFERENCES BLOOM, M. E., RACE, R. E. & WOLFINBARGER, J. B. (1982). Identification of a nonvirion protein of Aleutian disease virus: mink with Aleutian disease have antibody to both virion and nonvirion proteins. Journalof Virology 43, 6084i16. COTMORE, S. F., STURZENBECKER, L. J. & TATTERSALL, P. (1983). The autonomous parvovirus MVM encodes two nonstructural proteins in addition to its capsid polypeptides. Virology 129, 333-343. CROSS, S. S. & PARKER, J. C. (1972). Some antigenic relationships of murine parvoviruses: minute virus of mice, rat virus, and H-1 virus. Proceedings of the Society for Experimental Biology and Medicine 139, 105-108. HALLAUER, C. & KRONAUER, G. (1965). Extraction of cell-associated virus without damage of the culture. ArchivJ~r die gesamte Virusforschung 15, 433-440. LEDERMAN, M., PATTON, J. Y., STOUT, E. R. & BATES, R. C. (1984). Virally coded noncapsid protein associated with bovine parvovirus infection. Journal of Virology 49, 315-318. tca~tsunaga, Y. & MATSUNO, S. (1983). Structural and nonstructurai proteins of a rabbit parvovirus. Journal of Virology 45, 627-633. MENGELING, W. L. & PAUL, P. S. (1986). Antibodies for autonomous parvoviruses of lower animals detected in human serum. Archives of Virology 88, 127-133. MENGELING, W. L., BUNN, T. O. & PAUL, P. S. (1983). Antigenic relationship between porcine and canine parvoviruses. American Journal of Veterinary Research 44, 865-867. MENGELING, W. L., WHITEFORD, R. A., VAN DEUSEN, R. A., KRAMER, T. T. & PAUL, P. S. (1984). Potential of monoclonal antibodies for systemic immunoprophylaxis in the pig. Proceedings of the International Pig Veterinary Society, 8th LP.V.S. Congress, p. 15. MITRA, R., WALl, T., VALDEZ, V., FABISCH, P. & SALZMAN, L. A. (1983). Transcription and translation in the autonomous parvovirus KRV. Virology 125, 349-360. MOLITOR, T. W., JOO, H. S. & COLLETT, M. C. (1985). Identification and characterization of a porcine parvovirus nonstructural polypeptide. Journal of Virology 55, 554-559. PMtADISO, P. R. (1984). Identification of multiple forms of the noncapsid parvovirus protein NCVP 1 in H-1 parvovirus-infected cells. Journal of Virology 52, 82-87. PARRISH, C. R. & CAXMICHAEL, L. E. (1983). Antigenic structure and variation of canine parvovirus type-2, feline panleukopenia virus, and mink enteritis virus. Virology 129, 401-414.

2844 Short communication P1NTEL, D., DADACHAN/, D., ASTELL, C. R. & WARD, D. C. ( 983). The genome of minute virus of mice, an autonomous parvovirus, encodes two overlapping transcription units. Nucleic Acids Research 11, 1019-1038. SIEGL, G. (1976). The parvoviruses. Virology Monographs, vot. 15. Wien & New York: Springer-Verlag. TATTEgSALL, P., CAWTE, P. J., S~L~TKIN, A. J. R WARD, D. C. (1976). Three structural polypeptides coded for by minute virus of mice, a parvovirus. Journal of Virology 20, 273-289. (Received 10 April 1986)