Raccoon Poxvirus Feline Panleukopenia Virus VP2 Recombinant Provided by Elsevier - Publisher Connector Protects Cats Against FPV Challenge

VIROLOGY 218, 248–252 (1996) ARTICLE NO. 0186

SHORT COMMUNICATION

View metadata, citation and similar papers at core.ac.uk brought to you by CORE Raccoon Poxvirus Feline Panleukopenia Virus VP2 Recombinant provided by Elsevier - Publisher Connector Protects Cats against FPV Challenge

LIANGBIAO HU,*,1 JOSEPH J. ESPOSITO,† and FRED W. SCOTT*,2

*Cornell Feline Health Center, Department of Microbiology and Immunology, College of Veterinary Medicine, Cornell University Ithaca, New York 14853; and †Poxvirus Section, Division of Viral and Rickettsial Diseases, National Center for Infectious Diseases, Centers for Disease Control and Prevention, Atlanta, Georgia 30333 Received October 11, 1995; accepted February 1, 1996

An infectious raccoon poxvirus (RCNV) was used to express the feline panleukopenia virus (FPV) open reading frame VP2. The recombinant, RCNV/FPV, was constructed by homologous recombination with a chimeric plasmid for inserting the expression cassette into the thymidine kinase (TK) locus of RCNV. Expression of the VP2 DNA was regulated by the

vaccinia virus late promoter P11 . Southern blot and polymerase chain reaction (PCR) analyses confirmed the cassette was in the TK gene of the RCNV genome. An immunofluorescent antibody assay using feline anti-FPV polyclonal serum showed the expressed viral antigen in the cytoplasm of infected cells. Radioimmunoprecipitation with the same antiserum detected a 67-kDa VP2 protein which exactly matched the migration of the authentic FPV VP2 protein by SDS–polyacrylamide gel electrophoresis. Nine five-month-old cats were vaccinated and 21 days later were boosted with the recombinant virus. Peroral FPV challenge 2 weeks after the booster showed that the cats were fully protected as measured by examining clinical signs and total white blood cell counts in peripheral blood. Cats not immunized developed low to very low leukocyte counts following peroral FPV challenge. The nine vaccinated cats showed high FPV neutralization antibody prior to challenge, whereas nonvaccinated cats formed anti-FPV antibodies only after challenge. ᭧ 1996 Academic Press, Inc.

Feline panleukopenia virus (FPV) causes severe clinical VP2 protein which is converted to a 60- to 64-kDa poly- illness in young kittens with high morbidity and mortality peptide (VP3) by proteolytic cleavage to remove 15 to 20 (1–3). Because the replication of FPV requires rapidly multi- amino acids. Empty viral particles propagated in vitro or plying cells, the lesions caused by FPV are mainly found in vivo contain mostly VP2, a few VP1 subunits, and no in tissues with the greatest rate of mitotic activity, such as VP3 (9, 10). Intact virions contain the three proteins in bone marrow (1, 2). Clinically, a marked drop in total nested to include the viral DNA (11, 12). Infectious CPV leukocyte count by Days 4 to 6 after infection is the predom- particles contain 60 subunits of protein that is predomi- inant indicator of FPV infection (2–4). nantly VP2. FPV is closely related to canine parvovirus (CPV) and Structure analysis of CPV particles has revealed that mink enteritis virus at the protein and nucleic acid levels part of the amino terminal region of VP2 is on the exterior (5, 6). The complete genomic DNA sequences of FPV of the virion and is associated with immunodominant and CPV have been determined. The sequences reveal antigenic epitopes (13). Certain residues on the extreme that there are only 50 nucleotide differences between outside of the CPV particles are in common with residues the two viruses (7). The host range differences between on FPV particles (13). Because the amino terminal of VP1 the two viruses were found in the DNA region of 59–73 is unusually basic, it might help neutralize the DNA (14). map units (7, 8). The genome of FPV is a linear, single- Antisera against FPV and CPV have shown very similar stranded DNA of about 5,000 nucleotides that encodes antibody titers against either virus by virus neutralization three structural proteins: a large 80- to 85-kDa protein (VN) and hemagglutination-inhibition tests (5, 8, 15). (VP1) composing 10 to 15% of the virion protein, a me- Based on the close similarities between CPV and FPV, dium size 64- to 67-kDa protein (VP2), and a part of the it has been speculated that the vaccines developed with either virus will protect against both feline and canine parvovirus infections (16). Finally, the structural proteins, 1 Present address: Division of Comparative Medicine, Johns Hopkins VP1 and VP2 of Aleutian mink disease parvovirus (ADV), University, School of Medicine, Baltimore, MD 21205. have been produced in a baculovirus expression system; 2 To whom correspondence and reprint requests should be ad- the expressed proteins self-assembled as empty parti- dressed. Fax: (607) 253-3419. cles (17).

0042-6822/96 $18.00 248 Copyright ᭧ 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

AID VY 7820 / 6a13$$$321 03-01-96 21:25:15 viras AP: Virology SHORT COMMUNICATION 249

Raccoon poxvirus (RCNV) is an orthopoxvirus; two iso- cleotides 516–535) and GATCACCATCTGCTGGTTGA lates were derived from upper respiratory tissue of two (VP2 ORF nucleotides 1627–1608) were utilized to iden- apparently healthy raccoons trapped in Maryland in 1964 tify satisfactory recombinant viral plaques. The VP2 ORF (18). A comparative study with several different poxviruses sequences were detected in lysates of virus-infected has indicated that RCNV inoculation of cats induces higher cells by dot-blot hybridizations, and in recombinant viral autologous VN antibody titer than autologous antibodies genome DNA by detecting PCR amplified VP2 ORF inter- induced by other poxviruses tested (19). Laboratory animal nal sequences in thrice plaque-purified, phenol-chloro- studies with a RCNV rabies virus glycoprotein recombinant form-extracted, recombinant viral genome DNA prepara- (RCN-G) have indicated that oral administration of RCN-G tions (data not shown). The recombinant virus isolate induces rabies VN antibodies and protection against rabies replicated in Rat-2 cells under BUDR and formed typical challenge in raccoons and in a variety of other animals RCNV plaques. The results suggested to us that the in- tested (20). In cats vaccinated perorally, RCN-G has been serted VP2 ORF was located in the TK locus of the RCNV shown to induce rabies virus VN antibodies and not spread genome. Production of the recombinant virus confirmed to cats in contact with vaccinated cats (16). Earlier, it had also that VV TK insertion vectors such as pTKgptF3S can been shown that the thymidine kinase (TK) gene fragment be directly used for foreign gene insertion into the RCNV of RCNV was able to cross-hybridize with that of vaccinia TK locus (20, 21). virus (VV), and this enabled construction of RCNV recombi- The in vitro expression of VP2 protein by recombinant nants with various chimeric plasmids designed for foreign RCNV/FPV was demonstrated by indirect immunofluores- gene expression by marker rescue into the VV TK locus cent antibody (IIFA) and radioimmunoprecipitation (RIP) (20–22). tests as shown in Figs. 1 and 2. In the IIFA test, RCNV- In the present study, an infectious RCNV FPV VP2 re- infected CrFK cells (Fig. 1a) and mock-infected CrFK cells combinant (RCNV/FPV) was developed by recombining (not shown) did not react with anti-FPV antibodies. FPV- a FPV VP2 expression cassette into the TK locus of the infected CrFK cells showed intranuclear fluorescence RCNV genome. To generate the recombinant, the FPV (Fig. 1b); however, RCNV/FPV-infected CrFK cell mono- VP2 open reading frame (ORF) in 2304 bp excised from layers showed VP2 antigen in the cytoplasm (Fig. 1c). a FPV infectious clone was inserted into the plasmid The FPV polyclonal antiserum used in our experiments pTKgptF3S (23). The first 33 bases of the ORF were de- is a standard diagnostic serum that does not cross-react leted by HincII digestion and the remaining DNA frag- with RCNV or feline viruses generally tested for in veteri- ment was inserted under control of the VV P11 late pro- nary clinics. moter in pTKgptF3S. Southern blot analysis followed by The infection and replication cycle of orthopoxviruses, sequencing with a CTACTTGCATAGATAGGT primer, de- including raccoon poxvirus, occurs completely in the cy- signed to anneal 167 bases up from a start codon at the toplasm (27). Consistent with this, the VP2 protein ex- end of the P11 late promoter, enabled selection of an pressed by RCNV/FPV under promoter P11 appears in appropriate chimeric plasmid (not shown) for developing the cytoplasm during the late infection stage. Interest- recombinant RCNV. Thus, pTKgptF3S containing the left- ingly, ADV structural proteins VP1 and VP2 coexpressed end of the VV TK followed by TAAAAATATAGTAGAATT- in a baculovirus expression system have been shown to TCATTTTGTTTTTTTCTATGCTATAAATGAATTCCTGC- be empty particles in the nucleus of baculovirus recombi- AGGTCAACCTGCTGTCA and the remainder of VP2 nant-infected insect cells (17). During authentic FPV in- sequences to the stop codon was selected for further fection, VP1 and VP2 also appear in the nucleus, mainly analysis (note: nucleotides to the left of the ATG (bold in virus particles. Because we expressed only FPV VP2, font) are P11 late promoter sequences; nucleotides to the we did not examine RCNV/FPV-infected cells for empty right of the ATG are the in-frame bases of the VP2 ORF FPV particles. The reason why the VP2 transcription from a HincII site in a polylinker sequence that remained product in RCNV/FPV-infected cells is not transported to after cloning). the nucleus is unclear. However, the marked shut down Infectious RCNV recombinants carrying the VP2 ORF of host cell protein synthesis by the late time of RCNV/ were obtained by transfecting the chimeric plasmid into FPV infection and/or lack of other coexpressed FPV com- CV-1 cell monolayers that had been infected for 2 hr with ponents, e.g., VP1, might be responsible. RCNV by methods described elsewhere (20, 22, 24–26). RIP of [ 35S]methionine pulse-labeled proteins in ly- Rat-2 TK minus cells were then used to select RCNV sates of cells infected with RCNV/FPV or RCNV also recombinant viruses by thrice plaque-purifying in the showed that VP2 protein is expressed by RCNV/FPV and presence of 50 mg 5-bromo-2؅-deoxyuridine (BUDR)/ml that VP2 is recognized by FPV antibodies (Fig. 2). Four of medium. Dot-blot hybridizations with a VP2 DNA probe plaque-purified RCNV/FPV recombinant preparations and PCR amplifications of VP2 ORF internal sequences that we made expressed VP2 protein when examined by with primers AGTTCAACCAGACGGTGGTC (VP2 ORF nu- 8 to 25% gradient gel SDS–PAGE. The expressed protein

AID VY 7820 / 6a13$$$321 03-01-96 21:25:15 viras AP: Virology 250 SHORT COMMUNICATION migrated at 67 kDa (lanes 1–4), which is the same as reported (9, 10, 13) for mature VP2 protein from FPV- infected cells. The 67-kDa protein did not appear in lysates of cells infected with RCNV (lane 5) or mock-infected cell lysates (not shown). Specific recognition of VP2 protein by FPV polyclonal antiserum in IIFA and RIP assays suggested to us that

FIG. 2. Radioimmunoprecipitation detection of FPV VP2 expressed by RCNV/FPV. CV-1 cell monolayers were infected with the RCNV/FPV or RCNV and then incubated at 37Њ for 1 hr in methionine-free medium followed by pulse-labeling for 16 hr at 37Њ in medium containing 500 mCi [ 35S]methionine. The cells were washed and then lysed in a buffer

containing 100 mM NaCl, 50 mM Tris–HCl, pH 8.0, 5 mM Na2EDTA, 1% Triton X-100, 1% deoxycholic acid, and 50 mM phenylmethylsulfonyl fluoride. The lysate was then incubated overnight at 4Њ with cat anti- FPV serum. Protein A–Sepharose was then added to the antibody– lysate mixture and incubated and agitated for 2 hr at room temperature. The complex was washed by centrifugation and finally dissolved and heated for 3 min at 100Њ in loading buffer for a 8–25% SDS–PAGE gradient PhastGel (Pharmacia Biotech, Sweden). After electrophoresis in the Phastsystem Separation and Control Unit, the gel was fluorogra- phed and then exposed to the X-ray film to detect 35S-labeled proteins. Lanes 1–4 show precipitates from four separate plaque-purified recombinant-infected cell lysates of CV-1 cells infected with RCNV/FPV; protein migrating at 67 kDa is apparent; lane 5 shows lack of a precipitated homolog from a lysate of CV-1 cells infected with RCNV.

the protein expressed by RCNV/FPV is structurally satisfactory for appropriate immunogenic determinants of VP2. Consistent with the data here, in a related study, an FPV VP2 monoclonal antibody identified the same expressed VP2 protein in another RCNV construct that coexpresses FPV VP2 and the rabies virus glycoprotein (Hu et al., unpublished data). The immune response of nine 5-month-old cats to vaccination with the RCNV/FPV was examined by subcuta- neous (SQ) inoculation with the recombinant followed by a booster 21 days later. The vaccinated cats and two unvaccinated cats were challenged by using the USDA standard FPV challenge method (28) on Day 34 after primary vaccination. In Fig. 3 we show that following the initial vaccination with RCN/FPV, the geometric mean serum VN antibody titer of the nine vaccinated cats had FIG. 1. Indirect immunofluorescent FPV antibody reactivity of CrFK reached 1:1,000 by Day 21, the day on which the cats cells infected with RCNV, FPV, or RCNV/FPV virus. CrFK cell monolayers were boosted. Seven days after the booster, the mean in 8-well LabTek chamber slides (Nunc, Inc., IL) were infected with 10 PFU per cell of RCNV, FPV, or RCNV/FPV. After 24 hr incubation, the serum VN antibody titer was 1:5,000. Thirteen days after cells were acetone-fixed, blocked with normal rabbit serum, and then booster, the mean VN titer was still 1:5000; however, by incubated with feline FPV polyclonal antiserum followed by washing 22 days after the booster the titer was determined to be and incubation with fluorescein-conjugated rabbit anti-cat IgG. (a) This 1:17,500. No FPV VN antibodies were detected in cats panel shows CrFK cell monolayer infected with RCNV does not react not vaccinated. with the anti-FPV serum; (b) shows a CrFK cell infected with FPV reacted with anti-FPV serum and immunofluorescence was mainly in the Challenge of the vaccinated cats with FPV on Day 13 nucleus; (c) shows a CrFK cell infected with RCNV/FPV reacted with after the booster did not raise the VN antibody level, anti-FPV serum and immunofluorescence was mainly in the cytoplasm. suggesting that challenge virus replication was stopped

AID VY 7820 / 6a13$$$321 03-01-96 21:25:15 viras AP: Virology SHORT COMMUNICATION 251 by vaccination with RCNV/FPV. On the other hand, unvaccinated animals showed detectable VN titers at Day 5 after FPV challenge. The appearance of circulatory FPV antibodies is consistent with development of a humoral response to infection during challenge. In a separate study (16) with RCNV/FPV at a higher dose in older cats less susceptible to FPV, a single vaccination also showed a significant VN antibody response and protection of cats from FPV challenge. We have not yet examined the cell-mediated immune response after RCNV/FPV vaccination; however, we speculate that such a response is also playing an active role in providing protection from FPV challenge. In other candidate parvovirus vaccine systems that do not use a recombinant that replicates in the vaccinated animal, e.g., peptide vaccines and baculovirus systems, it is likely that cell-mediated immunity levels are low or not produced. Total white blood cell and differential counts were FIG. 4. Geometric mean of the total white blood cell (WBC) counts measured on Days 0, 3, 5, 9, and 20 following FPV chal- in peripheral blood samples taken after FPV challenge of vaccinated and nonvaccinated cats. The nine RCNV/FPV vaccinated cats and the lenge. Fig. 4 shows that the nine vaccinated cats had 4.7 two mock-vaccinated cats challenged with 10 TCID50 FPV perorally total white blood cell counts within a normal range of 13 days after booster according to a USDA standard procedure (28). 3 6.1–21.1 1 10 /ml throughout the vaccination and chal- The total and differential WBC were determined from blood samples lenge. Also, counts of lymphocytes, monocytes, eosino- taken on Days 0, 3, 5, 9, and 20 after challenge. Total WBCs on Day phils, and basophils remained within normal ranges 0 were considered to be the normal level for the individual cat. The WBC counts were determined using a quantitative automated hematology analyzer (Coulter Counter Model S Plus IV, Coulter Corporation, Hia- leah, FL). The dark squares with standard deviation indicate the geometric mean counts of WBCs in the group of nine vaccinated cats. The stippled squares with standard deviation represent the white cell counts in the two mock-vaccinated cats (WBC counts for unvaccinated cats were are similar for cats given standard RCNV in another study not shown). The average of the total WBCs in peripheral blood samples from the nine vaccinated cats and the two unvaccinated cats on the challenge day were considered the normal WBCs.

throughout the experimental course. In contrast, the total white cell counts of cats not vaccinated were less than 5 1 103/ml on Days 3 and 9 and were 2.2 1 103/mlon Day 5. Thus, the unvaccinated cats clearly developed panleukopenia, severe lymphopenia, and eosinopenia following FPV challenge. As a typical course of feline panleukopenia, unvaccinated cats had cell counts in the FIG. 3. Geometric mean of FPV VN titers of sera of nine 5-month- normal range by Day 20 after challenge. old specific-pathogen-free cats vaccinated and boosted subcutane- In conclusion, the results in this report show in vivo ously with RCNV/FPV. Nine cats were subcutaneously vaccinated with and in vitro expression of FPV VP2 by a recombinant 1 1 107 PFU of RCNV/FPV and then boosted subcutaneously on Day RCNV/FPV. The recombinant induced a rapid, vigorous, 21. Two cats were mock-infected and boosted with PBS at the same FPV VN antibody response and the response protected time as the vaccinated group. The 11 cats were challenged 13 days after the booster. Serum samples were collected on Days 0, 21, 28, cats against FPV challenge administered 13 days after 34, 37, 39, 43, and 54 after the primary vaccination. The FPV VN test a booster immunization. Of interest, studies on human procedures and antibody titer interpretations were done essentially as parvovirus B19 have indicated that B19 virus capsids described previously (29). The reciprocal of the highest dilution of se- composed of only VP2 elicit weak neutralizing antibody rum at which no FPV inclusion bodies were detected by microscopy activity and capsids containing both VP1 and VP2 induce of infected cells was interpreted as the VN titer. The solid bars with indicated standard deviations represent VN responses of the vacci- a strong VN antibody activity (30). Here we show that nated group and the hatched bars with standard deviations represent FPV VP2 alone is able to induce sufficient antibodies VN response of unvaccinated cats. protecting against FPV challenge.

AID VY 7820 / 6a13$$$321 03-01-96 21:25:15 viras AP: Virology 252 SHORT COMMUNICATION

We strongly suspect that the RCNV/FPV recombinant 9. Studdert, M. J., and Peterson, J. E., Arch. Gesamte Virusforsch. 42, is safe and efficacious for cats because no untoward 346–354 (1973). 10. Jongeneel, C. V., Sahli, R., McMaster, G. K., Hirt, B., J. Virol. 59, effects were apparent during regular examinations of the 564–573 (1986). SQ-vaccinated animals described here. The USDA Stan- 11. Tattersall, P., Cawte, P. J., Shatkin, A. J., and Ward, D. C., J. Virol. dard Requirement for licensure of feline panleukopenia 20, 273–289 (1976). virus vaccines requires that the vaccine stimulates a VN 12. Tattersall, P., Shatkin, A. J., and Ward, D. C., J. Mol. Biol. 111, 375– antibody titer of at least 1:8 with at least 20 vaccinated 394 (1977). 13. Tsao, J, Chapman, S. M., Agbandje, M., Keller, W., Smith, K., Wu, cats and 5 nonvaccinated cats and less than 25% drop H., Luo, M., Smith, T. J., Rossmann, R., Compans, R. W., and from the normal white blood cell count after challenge Parrish, C. R., Science 251, 1456–1464 (1991). (USDA Code of Regulations; Ref. 28). Although only 9 14. Kelly, D. C., and Elliot, R. M., J. Virol. 21, 408 (1977). vaccinated cats and 2 nonimmunized cats were used in 15. Siegl, G., Virol. Monogr. 15, 1–109 (1976). this study, the results here suggested to us that further 16. Ngichabe, C. K., Ph.D. Thesis, Cornell University, Ithaca, NY (1992). 17. Wu, W. H., Bloom, M. E., Berry, B. D., McGinley, M. J., Platt, K. B., studies including development of an oral polyvalent vac- J. Vet. Diagn. Invest. 6, 23–29 (1994). cine candidate would be warranted. 18. Herman, J. F., Bacteriol. Proc., 64th Annu. Meeting Amer. Soc. Mi- crobiol., Abstract 117 (1964). ACKNOWLEDGMENTS 19. Scott, F. W., 69th Conf. Res. Workers Anim. Dis., Abstract 60 (1988). 20. Esposito, J. J. Sumner, J. W., Brown, D. R., Ebert, J. W., Shaddock, We thank C. W. Parrish for kindly advising us during these studies J. H., Bai, X. H., Dobbins, J. G., and Fekadu, M., In ‘‘Vaccines 92: and for the infectious clone of FPV. We are grateful to I. Krumina for Modern Approaches to New Vaccines Including Prevention of excellent technical support and to R. Jacobson, J. Casey, and M. Barr AIDS’’ (F. Brown, R. M. Chanock, H. Ginsberg, and R. A. Lerner, for constructive suggestions on this work. Eds.), pp. 321–330. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 1992. REFERENCES 21. Esposito, J. J., and Knight, J. C., Virology 143, 230–251 (1985). 22. Esposito, J. J., Brechling, K., Baer, G., and Moss, B., Virus Genes 1. Greene, C. E., and Scott, F. W., In ‘‘Infectious Diseases of Dogs 1, 7–21 (1987). and Cats’’ (C. E. Greene, Ed.), pp. 291–299. Saunders, Philadel- 23. Falkner, F. G., and Moss, B., J. Virol. 62, 1849–1854 (1988). phia, 1990. 24. Shuman, S., Golder, M., and Moss, B., Virology 170, 302–306 (1989). 2. Scott, F. W., In ‘‘Diseases of The Cat’’ (J. Holzworth, Ed.), pp. 182– 25. Barrett, N., Mitterer, A., Mundt, W., Eibl, J., Eibl, M., Gallo, R. C., 193. Saunders, Philadelphia, 1987. Moss, B., and Dorner, F., AIDS Res. Hum. Retroviruses 5, 159– 3. Scott, F. W., In ‘‘Handbook of Parvoviruses’’ (P. Tijssen, Ed.), pp. 171 (1989). 103–111. CRC Press, Boca Raton, 1990. 26. Mackett, M., Smith, G. L., and Moss, B., J. Virol. 49, 857–864 (1984). 4. Wosu, L. O., Vet. Microbiol. 16, 137–143 (1988). 27. Moss, B., Semin. Immunol. 2, 317–327 (1990). 5. Parrish, C. R., and Carmichael, L. E., Virology 129, 401–414 (1983). 28. Anonymous, Fed. Regist. 38, 8426 (1973). 6. Tratschin, J. D., McMaster, G. K., Kronauer, G., and Siegl, G., J. Gen. 29. Scott, F. W., Csiza, C. K., and Gillespie, J. H., Cornell Vet. 60, 183– Virol. 61, 33–41 (1982). 191 (1970). 7. Parrish, C. R., Virology 183, 195–205 (1991). 30. Rosenfeld, S. J., Young, N. S, Alling, D., Ayub, J., and Saxinger, C., 8. Parrish, C. R., Adv. Virus Res. 38, 403–450 (1990). Arch. Virol. 136, 9–18 (1994).

AID VY 7820 / 6a13$$$321 03-01-96 21:25:15 viras AP: Virology