Development of a vaccine to prevent congenital cytomegalovirus infection is a major public health priority. Live vaccines attenuated through mutations targeting viral mechanisms responsible for evasion of host defense may be both safe and efficacious. Safety and vaccine efficacy were evaluated using a guinea pig cytomegalovirus (GPCMV) model. Recombinant GPCMV with a targeted deletion of gp145 (designated ⌬145), a viral protein kinase R (PKR) inhibitor, was generated. Attenuation was evaluated following inoculation of 10 7 PFU of ⌬145 or parental virus into guinea pigs immunosuppressed with cyclophosphamide. Efficacy was evaluated by immunizing GPCMV-naive guinea pigs twice with either 10 5 or 10 6 PFU of ⌬145, establishing pregnancy, and challenging the guinea pigs with salivary gland-adapted GPCMV. The immune response, maternal viral load, pup mortality, and congenital infection rates in the vaccine and control groups were compared. ⌬145 was substantially attenuated for replication in immunocompromised guinea pigs. Vaccination with ⌬145 induced enzyme-linked immunosorbent assay (ELISA) and neutralizing antibody levels comparable to those achieved in natural infection. In the higher-and lower-dose vaccine groups, pup mortality was reduced to 1/24 (4%) and 4/29 (14%) pups, respectively, whereas it was 26/31 (81%) in unvaccinated control pups (P < 0.0001 for both groups versus the control group). Congenital infection occurred in 20/31 (65%) control pups but only 8/24 (33%) pups in the group vaccinated with 10 6 PFU (P < 0.05). Significant reductions in the magnitude of maternal DNAemia and pup viral load were noted in the vaccine groups compared to those in the controls. Deletion of a GPCMV genome-encoded PKR inhibitor results in a highly attenuated virus that is immunogenic and protective as a vaccine against transplacental infection.
Infection with human cytomegalovirus (HCMV) causes considerable morbidity and occasional mortality in immunocompromised solid organ transplant and hematopoietic stem cell transplant patients, HIV-infected individuals, and newborns that acquire infection in utero (1, 2). Due to the lifelong morbidity associated with congenital CMV infection, a preconception vaccine capable of preventing virus transmission to the fetus would provide a highly cost-effective public health advance (3). Unfortunately, the lack of clear immunological correlates of protective immunity has hampered development of an HCMV vaccine. In spite of this uncertainty, there is evidence that virus-neutralizing antibody responses targeting viral envelope glycoproteins, as well as cellular immune responses (CD4 ϩ and CD8 ϩ ) targeting multiple structural and regulatory proteins, play important roles in protection against acquisition and reactivation of infection (4-7).