DNA dose was chosen to be lower than any dose previously tested by us in guinea pigs. The detection of neutralizing serum antibodies, and responses to both gD and gB by ELISA, indicated that these levels of DNA were effective. Furthermore, the combination at this dose was highly effective in preventing primary genital disease. Following challenge, control animals developed severe disease characterized by high lesion scores and the systemic involvement; in contrast, the DNA-immunized guinea pigs were nearly free of disease. Mean lesion scores for the DNA-immunized group were statistically indistinguishable from background. This background was established by inclusion of a mock-infected control group. The assignment of non-zero scores to some control animals on some observation days (Fig. 3) was likely the result of the slight irritation caused by the infection procedure. There was no evidence that these scores were due to HSV infection transmitted from infected cage mates; all mock-infected control animals were HSV seronegative 4 weeks after the challenge (data not shown). The extent of disease in those DNA-immunized animals which did develop lesions was lower than that seen in the infected control animal. The highest score attained by any DNA-immunized animal was 2.0 on 2 successive days. In contrast, seven of the eight infected controls were scored 4.0 on 2 or more successive days.
It has been postulated that due to competition for DNA uptake or expression, or antigen competition, immunization with DNAs in combination might result in reduced responses to the individual components. The combination of gD-2 and ΔgB-2 DNA did not appear to compromise the response to either component. Moreover, the protection achieved with this low-dosage combination was as good as, or better than, that seen in similar challenge studies using 100-μg doses of gD-2 DNA or ΔgB-2 DNA alone (unpublished observations). Because the combination of gD-2 and ΔgB-2 DNA could induce responses to the broader spectrum of epitopes contained in two separate antigens, it had the potential to be more effective than either component alone. The results are consistent with the combination being more effective than the individual components; however, that cannot be concluded from this study because the comparison was not made directly. Further titrations of gD-2 DNA and ΔgB-2 DNA, both individually and in combination, are in progress to address the question directly, and to establish minimally-effective doses.
Because of the small number of surviving control guinea pigs, latent infection and recurrent disease could not be evaluated. It has been shown with protein subunit vaccines (5, 6) and recently with high-dose DNA immunization (24) that significant reduction in primary genital disease also resulted in reduced latent infection and decreased recurrence. The extent of protection against primary disease found in the study presented here suggests that prophylactic immunization with a low-dose combination of gD and ΔgB-2 DNA would be effective against recurrence; experiments to evaluate this are ongoing.
We have not yet identified which DNA-induced immune responses are protective in our challenge models. Injection of gD-2 or ΔgB-2 DNA, individually or in combination, induced substantial neutralizing serum antibody titers, which may fully account for the protection observed. Others have shown by passive transfer that antibody alone can be protective in some mouse infection models (47, 48). However, DNA immunization has the capacity to elicit cell-mediated, as well as humoral immune responses (26, 35). Recently, Manickan et al. (20) showed that in mice immunized with HSV-1 gB DNA, a CD4+ cytotoxic T-lymphocyte response protected the animals from zosteriform infection with HSV-1. We have detected antigen-specific lymphoproliferative responses in mice and guinea pigs immunized with the gD-2 or ΔgB-2 DNA, but have not yet shown the induction of cytotoxic T lymphocytes (unpublished observations) and cannot rule out a contribution of cell-mediated immunity to the protection observed in these studies. The relative protective roles of humoral and cell-mediated immunity induced by immunization with gD-2 and ΔgB-2 DNA are yet to be defined and may depend on the infection model used.
We have demonstrated that immunization with low doses of DNA was highly effective in generating protective immunity in two animal models of HSV infection and we found, in mice, that a single immunization was protective. (Single immunizations have not been evaluated in the guinea pig model.) These results suggest that simple i.m. injection has the potential to be an efficient form of DNA delivery, and support the feasibility of developing DNA vaccines for human use where low dose and limited numbers of injections are desirable characteristics. We also found that the combination of gD-2 and ΔgB-2 DNA induced immune responses to both proteins and was effective in preventing HSV-2-induced mucosal disease. This result supports the concept that multivalent vaccines can be made by simply combining DNAs, and provides a starting point for the development of such a vaccine for genital herpes.
To date, gD and gB have been the focus of DNA vaccine development (20, 22–24) just as they have been for protein subunit vaccines (reviewed in ref. 3). However, the inherent simplicity of DNA immunization should allow the rapid identification of additional immunogens for inclusion in multivalent DNA vaccines for HSV-induced disease. It is now possible to scan the genomes of complex pathogens for novel immunogens (49) and readily test their capacity to elicit protective immunity. As potentially useful immunogens are identified, they can be easily evaluated in the context of an existing DNA vaccine. Using DNA immunization as both a discovery tool and as a method of delivering combinations of antigens should expedite the development of vaccines with greater potency and breadth of protection. Clearly this approach needs extensive evaluation before clinical efficacy and safety are demonstrated, but these early results with the combination of gD-2 and ΔgB-2 DNA are encouraging.
We wish to thank Mr. Timothy Schofield (Merck Research Laboratories Biometrics Department) for statistical analyses of the data.
1. Roizman, B. (1991) Rev. Infect. Dis. 13, Suppl. 11, S892–S894.
2. Whitley, R.J. & Meignier, B. (1992) in Vaccines: New Approaches to Immunological Problems, ed. Davies, J.E. (Butterworth-Heinemann, Boston), pp. 223–254.
3. Burke, R.-L. (1993) Semin. Virol. 4, 187–197.
4. Burke, R.L. (1991) Rev. Infect. Dis. 13 (Suppl. 11), S906–S911.
5. Stanberry, L.R., Bernstein, D.I., Burke, R.L., Pachl, C. & Myers, M.G. (1987) J. Infect. Dis. 155, 914–920.
6. Stanberry, L.R., Myers, M.G., Stephanopoulos, D.E. & Burke, R.L. (1989) J. Gen. Virol. 70, 3177–3185.
7. Sanchez-Pescador, L., Burke, R.L., Ott, G. & Van Nest, G. (1988) J. Immunol. 141, 1720–1727.
8. Whitley, R.J., Kern, E.R., Chatterjee, S., Chou, J. & Roizman, B. (1993) J. Clin. Invest. 91, 2837–2843.
9. Meignier, B., Longnecker, R. & Roizman, B. (1988) J. Infect. Dis. 158, 602–614.
10. Farrell, H.E., McLean, C.S., Harley, C., Efstathiou, S., Inglis, S. & Minson, A.C. (1994) J. Virol. 68, 927–932.
11. McDermott, M.R., Graham, F.L., Hanke, T. & Johnson, D.C. (1989) Virology 169, 244–247.
12. Gallichan, W.S., Johnson, D.C., Graham, F.L. & Rosenthal, K.L. (1993) J. Infect. Dis. 168, 622–629.
13. Wachsman, M., Aurelian, L., Smith, C.C., Lipinskas, B.R., Perkus, M.E. & Paoletti, E. (1987) J. Infect. Dis. 155, 1188–1197.
14. Aurelian, L., Smith, C.C., Wachsman, M. & Paoletti, E. (1991) Rev. Infect. Dis. 13 (Suppl. 11), S924–S930.
15. Cantin, E.M., Eberle, R., Baldick, J.L., Moss, B., Willey, D.E., Notkins, A.L. & Openshaw, H. (1987) Proc. Natl. Acad. Sci. USA 84, 5908–5912.
16. Heineman, T.C., Connelly, B.L., Bourne, N., Stanberry, L.R. & Cohen, J. (1995) J. Virol. 69, 8109–8113.