Antibody-dependent Enhancement of Tick-borne Encephalitis Virus Infectivity

Transcription

1 J. gen. Virol. (1985), 66, Printed in Great Britain 1831 Key words: antibody-mediated enhancement/tbev/flavivirus Antibody-dependent Enhancement of Tick-borne Encephalitis Virus Infectivity By ROBERT J. PHILLPOTTS,* JOHN R. STEPHENSON AND JAMES S. PORTERFIELD t Vaccine Research and Production Laboratory, Centre for Applied Microbiology and Research, Porton Down, Salisbury, Wiltshire SP4 0JG and 1Sir William Dunn School of Pathology, University of OxJbrd, South Parks Road, Oxford OX1 3RE, U.K. (Accepted 14 May 1985) SUMMARY Fourteen mouse monoclonal antibodies raised against tick-borne encephalitis virus (TBEV) and polyclonal antisera raised against six other flaviviruses, Edge Hill (EHV), Japanese encephalitis (JEV), Langat (LGTV), louping ill (LIV), West Nile (WNV) and yellow fever (YFV), were tested for their ability to enhance the replication of TBEV in cells of the mouse macrophage-like line P388 D1, and for their reactivity in ELISA and haemagglutination inhibition (HI) tests. Irrespective of their specificity for either the 51K or 58K polypeptide present in TBEV-infected cells, 13 of the 14 monoclonal antibodies enhanced the replication of TBEV but not of WNV. The remaining monoclonal antibody, which immunoprecipitated the 58K polypeptide of TBEV enhanced WNV but not TBEV, although it reacted strongly with both viruses in ELISA and HI tests. Only polyclonal antisera against viruses within the tick-borne encephalitis virus complex (TBEV, LGTV and LIV) enhanced TBEV replication, although all the polyclonal antisera reacted with TBEV by ELISA; two (against JEV and WNV) also reacted by HI test and all enhanced the replication of WNV. These findings suggest that with TBEV, enhancement may be TBEV complex-specific rather than flavivirus-specific. Data derived from testing both polyclonal and monoclonal antibodies suggest further that not all antibodies that bind to the envelope glycoprotein of TBEV are able to enhance the replication of TBEV, and that enhancement is epitope-specific. Tick-borne encephalitis is the most important arthropod-borne virus disease in Europe, and clinically apparent infections with tick-borne encephalitis virus (TBEV, flavivirus genus of the new family Flaviviridae) have been reported from almost every country on the European mainland. TBEV infection does not occur in the U.K., but infection of animals with a related flavivirus, louping ill (LIV), is prevalent here (Reid et al., 1978; Bannatyne et al., 1980). When assayed in certain cells the infectivity of many viruses is increased by the presence of specific antiviral antibody, for example, members of the Togaviridae (Peiris & Porterfield, 1981 ; Cafruny & Plagemann, 1982), Bunyaviridae (Peiris & Porterfield, 1981), Rhabdoviridae (King et al., 1984; Clerx et al., 1978) and Reoviridae (Burstin et al., 1983). This phenomenon has been extensively studied in the mouse macrophage-like cell line P388 D 1 (Koren et al., 1975) where, in the absence of complement, enhancement by specific antiviral IgG requires the presence of a functional Fc receptor on the cells (Daughaday et al., 1981 ; Peiris et al., 1981). Here we describe the Fc receptor-mediated, antibody-dependent enhancement of TBEV infectivity in P388 D1 cells. To examine the antibody specificity of enhancement we have tested six hyperimmune rabbit antisera raised against other flaviviruses for reactivity with TBEV and West Nile virus (WNV) antigens in the ELISA and haemagglutination inhibition (HI) tests, and for their ability to neutralize and enhance the infectivity of both TBEV and WNV. In addition, 14 mouse SGM

2 1832 Short communication i I I I I I ~2 I I vc log,0 Serum dilution Fig. 1. Enhancement of TBEV by hyperimmune rabbit antisera. Effect on TBEV infectivity of incubation with pre-inoculation rabbit serum (O), or hyperimmune rabbit antiserum to TBEV (O), prior to assay in P388 DI cells, is shown. L15 medium alone was substituted for serum dilution in the virus controls (vc). -qp-, < 105 p.f.u./ml. monoclonal antibodies raised against TBEV, each of which immunoprecipitates one of two infected cell polypeptides (51K or 58K), were tested by ELISA and HI using TBEV and WNV antigens, and for enhancement of both TBEV and WNV infectivity. The central European strain, Neud6rfl, of TBEV, and the Egypt 101 strain of WNV were used throughout. Preparation of the rabbit hyperimmune antisera to the flaviviruses Langat (LGTV), louping ill (LIV), Japanese encephalitis (JEV) and Edge Hill (EHV) has been described previously (Madrid & Porterfield, 1974); hyperimmune rabbit antiserum to yellow fever virus (YFV) was a gift from E. A. Gould, London School of Hygiene and Tropical Medicine, U.K. The preparation of rabbit hyperimmune antiserum to TBEV is described by Hambleton et al. (1983). All sera were inactivated at 56 C for 30 min. The antibody-dependent plaque enhancement (ADPE) assay using P388 D1 cells (Peiris & Porterfield, 1982) was used, but modified for TBEV by the addition of 30 mm-mgc12 and 10 ~tg/ml DEAE-dextran (Pharmacia) to the overlay. The results of a single experiment in which hyperimmune rabbit antiserum to TBEV, and preimmune rabbit serum were tested for their ability to enhance the infectivity of TBEV are shown in Fig. 1. Incubation with pre-immune rabbit serum had no effect on the titre of TBEV, whereas hyperimmune rabbit antiserum neutralized the virus up to a dilution of and enhanced infectivity at dilutions from to The maximum enhancement, which occurred at serum dilutions of and 10-4, was ninefold above the virus control. To demonstrate the requirement for a functional Fc receptor, P388 D1 cells were treated before exposure to virus-antibody complexes with a monoclonal IgG, 2.4 G2, to the trypsinresistant Fc receptor (Peiris et al., 1981). Subconfluent monolayers of P388 D1 cells were incubated at 4 C with either 2.4 G2 hybridoma cell culture supernatant diluted 1 : 2 in growth medium, or growth medium alone. After 2 h the cell sheets were washed three times in phosphate-buffered saline, then infected (for 2 h at 4 C) with TBEV mixed with an equal volume of either rabbit hyperimmune antiserum to TBEV diluted 1 : 104 or diluted pre-immune rabbit serum. In P388 D1 cells pretreated with growth medium alone, the titre of TBEV was p.f.u./ml in the presence of pre-inoculation rabbit serum, compared with p.f.u./ml in the presence of hyperimmune rabbit antiserum (i.e. fivefold enhancement). However, when P388 D I cells were pretreated with 2.4 G2 monoclonal antibody, there was no enhancement of infectivity (titre of TBEV in the presence of pre-inoculation and hyperimmune rabbit serum 10 TM and I07"05

3 Titre with WNV (-log10) Titre with TBEV (-log~ 0) f- Rabbit serum Enhancement* Neutralizationi" HI ELISA Enhancement* Neutralizationt HI ELISA Anti-TBEV 3.5 < 1.3 < , Anti-LIV 2.5 < Anti-LGTV 2.5 < Anti-YFV 3.0 < < 1.0 < 1.3 < Anti-WNV < 1-0 < Anti-JEV < 1.0 < Anti-EHV 4.0 < 1-3 < 1.0 2,9 < 1.0 < 1.3 < Pre-immune < 1.0 < 1.3 < < 1.0 < 1.3 < * Dilution at which the increase in plaque count was >~threefold. Titres represent the mean of two or more assays performed on separate occasions. t Dilution at and below which a > 50% reduction in plaque number was observed. Table 1. Reaction of different ant&era with TBEV and WNV

4 1834 Short communication I I I [ i I I. I f I [ I f I I (a) (b) I I I I I I logj Antibody dilution I I I I I ] I I Fig. 2. Regression lines drawn through the linear portion of the titration curve for two anti-tbev monoclonal antibodies (20/.tg/ml) T7 (a) and T11 (b). A, WNV antigen; O, antigen prepared from uninfected mouse brain. The titration endpoint (0.15) was arbitrarily set at twice the mean A~50 value of wells from which monoclonal antibodies, but not conjugate, had been omitted p. f. u./ml respectively). Using the formula of Detre & White (1970) the reduction in plaque count produced by 2.4 G2 treatment of P388 D1 cells was statistically significant (z = 2.86, P < 0.01). To examine the specificity of enhancement, polyclonal rabbit antisera to TBEV and six other flaviviruses were tested for their ability to react with TBEV and WNV by ELISA, HI, and in the enhancement and neutralization assays (Table 1). The ELISA was performed with mouse brain antigen according to the method of Stephenson et al. (1984) and the HI tests as described by Clarke & Casals (1958). The neutralization test in PS cells is described by Madrid & Porterfield (1969). Only those antisera which neutralized TBEV (TBEV, LGTV and LIV) were also able to enhance its infectivity in P388 D1 cells. This contrasted with WNV, where all the flavivirus antisera tested enhanced infectivity, and there was no obvious relationship between the neutralization and enhancement titres. No correlation was found between infectivity enhancement and ELISA or HI titres for either virus. However, antisera that failed to enhance TBEV infectivity (YFV, WNV, EHV and JEV) did react with TBEV in the ELISA and two of them (WNV and JEV) also inhibited haemagglutination by TBEV. Of 14 monoclonal antibodies tested, all reacted with either one of two TBEV-infected cell polypeptides, 51K or 58K, in immunoprecipitation tests (Stephenson et al., 1984). Thirteen of these enhanced the infectivity of TBEV. The single exception was T7 which failed to enhance TBEV even when tested at high concentration (490 ~tg/ml). However, only T7 enhanced the infectivity of WNV (Table 2). Haemagglutination of both viruses was inhibited by T7, and one antibody, T 17, inhibited haemagglutination by TBEV but not W NV. There was no correlation between ELISA titres and ability to enhance the infectivity of either virus, but there was a relationship between antibody avidity (as shown by the gradient of the linear portion of the titration curve) and ability to enhance WNV. T7 (the only antibody to enhance WNV) bound strongly to WNV antigen in the ELISA (gradient 0.66) whereas the remaining monoclonal antibodies, e.g. T 11, although they bound specifically to W NV antigen, were less avid (gradients 0.04 to 0.08; Fig. 2). All the monoclonal antibodies including T7 bound strongly to TBEV antigen (gradients 0.1 to 0.47). None of the monoclonal antibodies reacted with an ELISA antigen prepared from uninfected mouse brain (Fig. 2, data not shown). The ability to enhance TBEV or WNV infectivity did not appear to be restricted by IgG isotype, or by the infected cell polypeptide (51K or 58K) with which each monoclonal antibody reacts in immunoprecipitation tests (Table 2). The failure of WNV, YFV, EHV and JEV antisera to enhance TBEV infectivity was not due to their inability to react with TBEV antigens. Positive ELISA reactions were obtained with all

5 Table 2. Reactions of mouse monoclonal antibodies to TBEV with TBEV and WNV Infected Titre* with TBEV (-log10) Titre* with WNV (-log10) Monoclonal IgG cell polypeptide c '~ ~ r ~ antibody isotypet precipitatedt Enhancement ELISA + HI Enhancement ELISA HI T6 2a 58K (0.12) 3.3 (0.06) - T7 2a 58K - (490 ~tg/ml) 3.9 (0.41) (0.67) 1.1 T9 2a 58K (0.47) (0.06) - TI 1 2a 58K (0.14) (0.05) TI2 3 51K (0.27) 1.2 (0.04) - T13 2a 58K (0.33) (0.06) - TI5 2b 51K (0.10) (0.04) - TI6 9 51K (0.11) 2.3 (0.05) - TI7 2a 51K (0.25) (0.05) TI8 2a 58K (0.21) (0.06) T33/1 2b 51K (0.12) (60 ~g/ml) (60 ~tg/ml) 3,3 (0.05) - (60 ~tg/ml) T33/2 2b 58K (0.22) - (40 ~tg/ml) - (40 ~tg/ml) 1.6 (0.08) (40 ~tg/ml) T33/3 2b 51K (0.10) (70 ~tg/ml) (40 ~tg/ml) 3.1 (0.05) - (70 ktg/ml) T35/3 2a 58K (0.15) (0.06) * All the titres given are standardized to 100 ~tg/ml of antibody protein (Lowry et al., 1951) and where higher or lower concentrations were tested, these are shown in parentheses. The enhancement titres represent the mean of two or more assays performed on separate occasions. t Data from Stephenson et al. (1984). :~ Figure in parentheses is the gradient of the linear portion of the titration curve in A450 units/dilution step. Negative, ta~

6 1836 Short communication four antisera. Two of them (WNV and JEV) also inhibited haemagglutination, and therefore reacted with the principal surface glycoprotein (E) of the virion (Heinz et al., 1981). These results suggest that with TBEV, enhancement may be TBE complex-specific and not flavivirus group-reactive, as has been argued for dengue virus enhancement (Brandt et al., 1982). Thirteen of the 14 mouse monoclonal antibodies raised against TBEV enhanced TBEV infectivity, but failed to enhance WNV. This may have been because of their low avidity for WNV antigens. The remaining monoclonal antibody, T7, did not enhance TBEV infectivity, even when tested at high concentration (490 ~tg/ml), but did enhance WNV. T7 bound with high avidity to both TBEV and WNV in the ELISA, and inhibited haemagglutination by both viruses. This suggests that antibody that binds specifically to the E glycoprotein of TBEV need not necessarily enhance its infectivity, and that, as has been found with dengue virus, enhancement may be epitope-specific (Halstead et al., 1984). However, monoclonal antibodies that inhibit haemagglutination may also enhance infectivity, for example T17. Either one of two infected cell polypeptides is immunoprecipitated by each of the monoclonal antibodies examined here (Stephenson et al., 1984). The larger (58K) polypeptide is closely related to the E glycoprotein of TBEV, but the origin of the smaller (S1K) polypeptide is uncertain. Enhancement was not restricted to those antibodies which immunoprecipitated the 58K polypeptide, suggesting that epitopes present on the 51K polypeptide are also expressed on the E glycoprotein. Further evidence for this is provided by the finding that T17, which immunoprecipitates the 51K polypeptide, is able to inhibit haemagglutination. The relationship between these two polypeptides is at present unclear. Although the role of antibody-mediated enhancement of virus infectivity in vitro is uncertain, it has been implicated in the pathogenesis of haemorrhagic dengue and dengue shock syndrome (Halstead, 1982; Porterfield, 1982). Circulating antibody may also increase the severity of feline infectious peritonitis virus infection (Weiss & Scott, 1981) and lead to 'early death' in rabies virus-infected animals (Prabhakar & Nathanson, 1981). Exposure to viral antigens may occur either by vaccination or by natural infection. However, we were unable to show enhancement of TBEV infectivity with antisera to J EV or YFV (two commonly used flavivirus vaccines), nor is there any clinical or epidemiological evidence to suggest that TBEV infection is more severe after natural flavivirus infections. When considering the design of future TBEV vaccines (produced for example by recombinant DNA technology) it would nevertheless by desirable to exclude epitopes that elicit enhancing antibody, but which are not involved in the protective immune response. We would like to thank Janet Peat for preparation of the manuscript. This work was supported by a grant from the Medical Research Council of Great Britain. REFERENCES BANNATYNE, C. H., WILSON, R. L., REID, H. W., BURTON, O. & POW, I. (1980). Louping I11 virus infection of pigs. Veterinary Record 106, 13. BRANDT, W. E., McCOWN, I. M., GENTRY, M. K. & RUSSELL, P. K. (1982). Infection enhancement of dengue type 2 virus in the U-937 human monocyte cell line by antibodies to flavivirus cross-reactive determinants. Injection and Immunity 36, BURST1N, S. J., BRANDRISS, M. W. & SCHLESINGER, J. J. (1983). Infection of a macrophage cell line, P388 D1 with reovirus: effects of immune ascitic fluids and monoclonal antibodies on neutralisation and on enhancement of viral growth. Journal of Immunology 130, CAFRUNY, W. A. & PLAGEMANN, P. G. W. (1982). Immune response to lactate dehydrogenase-elevating virus: serologically specific rabbit neutralising antibody to the virus. Injection and Immunity 37, CLARKE, D. H. & CASALS, J. (1958). Techniques for hemagglutination and hemagglutination inhibition with arthropod borne viruses. American Journal of Tropical Medicine and Hygiene 7, CLERX, J. P. M., HORZINEK, M. C. & OSTERHAUS, A. D. M. E. (1978). Neutralization and enhancement of infectivity of non-salmonid fish rhabdoviruses by rabbit and pike immune sera. Journal of General Virology 40, DAUGHADAY, C. C., BRANDT, W. E., Mc'COWN, J. M. & RUSSELL, P. K. (1981). Evidence for two mechanisms of dengue virus infection of adherent human monocytes: trypsin-sensitive virus receptors and trypsin-resistant immune complex receptors. Infection and Immunity 32, DETRE, K. & WHITE, C. (1970). The comparison of two Poisson-distributed observations. Biometrics 26, HALSTEAD, S. B. (1982). Immune enhancement of viral infection. Progress in Allergy 31,

7 Short communication 1837 HALSTEAD, S. B., VENKATESHAN, C. N., GENTRY, M. K. & LARSEN, L. K. (1984). Heterogeneity of infection enhancement of dengue 2 strains by monoclonal antibodies. Journal ofhnmunology 132, HAMBLETON, P., STEPHENSON, J. R., BASKERVILLE, A. & WIBLIN, C. N. (1983). Tick-borne encephalitis in primates, pathology, immune response and efficacy of vaccination. Injection and Immunity 40, HEINZ, F. X., TVMA, W. & KUNZ, C. H. (1981). Antigenic and immunogenic properties of defined physical forms of tick-borne encephalitis virus structural proteins. Injection and Immunity 33, KING, A. A., SANDS, J. J. & PORTERFIELD, J. S. (1984). Antibody-mediated enhancement of rabies virus infection in a mouse macrophage cell line (P388 D1). Journal of General Virology 65, KOREN, J. H., HANDWERGER, B. S. & WUNDERLICH, J. R. (1975). Identification ofmacrophage-like characteristics in a cultured murine tumour line. Journal oflmmunology 114, LOWRY, O. r~., ROSEBROUGn, N. J., FARR, g. t. & RANDALL, R. J. (1951). Protein measurement with the Folin phenol reagent Journal of Biological Chemistry 193, MADRID, DE A. T. & PORTERFIELD, J. S. (1969). A simple micro-cellular method for the study of group B arboviruses. Bulletin of the Worm Health Organization 40, MADRID, DE A. T. & PORTERFIELD, J. S. (1974). The flaviviruses (group B arboviruses): a cross-neutralization study. Journal of General Virology 23, PEIRlS, J. S. M. & PORTERFIELD, J. S. (1981). Antibody-dependent enhancement of plaque formation on cell lines of macrophage origin - a sensitive assay for antiviral antibody. Journal of General Virology 57, PE1RIS, J. S. M. & PORTERFIELD, I. S. (1982). Antibody-dependent plaque enhancement: its antigenic specificity in relation to Togaviridae. Journal of General Virology 58, PEIRIS, J. S. M., GORDON, S., UNKELESS, J. C. & PORTERFIELD, J. S. (1981). Monoclonal anti-fc receptor IgG blocks antibody enhancement of viral replication in macrophages. Nature, London 289, r'orteafield, J. S. (1982). Immunological enhancement and the pathogenesis of dengue haemorrhagic fever. Journal of Hygiene 89, PRABI-IAKAR, B. S. & NATHANSON, N. (1981). Acute rabies death mediated by antibody. Nature, London 290, REID, H. W., DUNCAN, J. S., PHILLIPS, J. D. P., MOSS, R. & WATSON, A. (1978). Studies on Louping I11 virus (flavivirus group) in wild red grouse (Lagopus logapus seoticus). Journal of Hygiene 81, STEPHENSON, J. R., LEE, J. M. & WILTON-SMITH, P. D. (1984). Antigenic variation among members of the tick-borne encephalitis complex. Journal of General Virology 65, WEISS, g. W. & SCOTT, r. w. (1981). Antibody-mediated enhancement of disease in feline infectious peritonitis: comparisons with dengue haemorrhagic fever. Comparative Immunology, Microbiology and Infectious Diseases 4, (Received 18 January 1985)