A Novel Strategy to Generate a Recombinant, Live-Virus, Replication Incompetent Congenital Cytomegalovirus Vaccine Candidate

Abstract

Abstract Background Preventing congenital cytomegalovirus (cCMV) infection, the most common infectious cause of disability in pediatrics, is a high public health priority. Subunit vaccine studies based on single viral proteins (usually envelope glycoproteins) have yielded disappointing results in clinical trials. In contrast, Disabled infectious single cycle (DISC) recombinant viruses are promising vaccine candidates, offering the potential for a robust immune response against multiple viral immunogens while maintaining an attenuated phenotype. These studies sought to examine the safety and immunogenicity of this strategy in an animal model of vaccine-preventable cCMV infection, the guinea pig cytomegalovirus (GPCMV) model. Design/Methods Vaccine design consisted of introducing a ligand-regulatable FKBP protein destabilization domain, ddFKBP, into an essential viral gene. This domain directs degradation of the tagged protein in cells unless the synthetic compound Shield-1 is present, allowing generation of viral particles capable of single-cycle replication. Two such essential genes, the GPCMV homologs of CMV genes UL51 and UL52 (GP51 and GP52), were targeted for vaccine design. In this study, we sought to test: 1) whether the DISC viruses GP51 and GP52 were attenuated for replication in a guinea pig model; and 2) whether the replication-incompetent viruses retained immunogenicity comparable to a wild-type virus. GPCMV-seronegative guinea pigs were challenged with DISC vaccines GP51 or GP52. Control animals were challenged with wild-type GPCMV (SG24). Animals (3 per group) were infected by subcutaneous injection and boosted at day 21. Blood collected at days 0, 7, 14, 21, 28, 35, 42, and 84 was used for PCR (to gauge attenuation) and serology (to examine immunogenicity). Results Real-time PCR showed a mean viral load of 8.5 x 105 genome copies/mL of blood at day 7 and 9.3 x 104 genome copies/mL at day 14 in guinea pigs that received SG24 compared to 2.9 x 104 and 1.9 x 103 genome copies/mL at days 7 and 14 respectively in recipients of GP51. No viral DNA was detected in recipients of GP52 at either day 7 or 14. These data demonstrate significantly lower viremia in the GP51 and GP52 groups compared to SG24 by two-way ANOVA (p<0.01; Fig. 1A). In spite of marked attenuation, no statistically significant difference in immune response was noted, with equivalent ELISA responses detected upon comparison of animals that received GP51, GP52, or SG24 CMV strains at days 14, 35, or 84 (Fig. 1B). Western blot analysis of sera used to probe fractions containing wild-type viral particles likewise demonstrated induction of a robust antibody response by GP51 and GP52. Conclusion(s) These data suggest that GP51 and GP52 are attenuated for replication but induce an immune response comparable to wild-type infection. Thus, they are promising options in the development of a vaccine to prevent cCMV. Examination of the extent of attenuation against vertical transmission of these DISC viruses, as well as their efficacy as vaccine candidates against cCMV, is warranted. These questions have important implications for potential future human clinical trials. Funding: Supported by 5 T35 AI118620-07, Medical Student Summer Research Program in Infection and Immunity