Human cytomegalovirus (HCMV) is a common cause of death in immunocompromised hosts including organ allograft recipients, patients with AIDS, and the developmentally immature fetus. HCMV is the most common cause of congenital viral infection in humans with an estimated annual incidence of 1 percent in the US. Approximately 10 percent of infected infants will suffer long term CNS damage, making HCMV the most common viral etiology of brain disease in infancy. Natural history studies have suggested that congenital HCMV infection is a vaccine modifiable disease, and the Institute of Medicine has targeted HCMV for vaccine development. Previous studies have suggested that primary maternal infection in pregnancy was the most significant risk factor for damaging congenital infections. More recent studies of congenital HCMV infection have indicated that non-primary maternal infection is equally important. We have recently demonstrated that non-primary infection of women can occur following reinfection with a new serotype of HCMV and the reinfection can cause damaging congenital infection. Reinfection could result from a lack of strain dependent virus neutralizing antibodies. Consistent with this hypothesis has been the demonstration of significant amino acid variation in the primary sequence of gN, a component of the abundant gM/gN virion envelope glycoprotein complex. Preliminary findings have indicated that the 78 amino acid ectodomain of gN can exhibit striking amino acid heterogeneity with some viruses having over 25 amino acid substitutions in this region compared to a reference virus strain. This level of sequence divergence appears sufficient to limit virus recognition by neutralizing antibody activity in seropositive hosts. In this proposal, we will investigate the hypothesis that the antigenic divergence in the gN/gM complex can lead to reinfection of seropositive women. We will address this hypothesis in four specific aims: (1) characterize the structural requirements for gM/gN complex formation, (2) using a genetic approach define the function of this essential virus glycoprotein complex in virus infectivity and assign functions to specific domains of gM and gN; (3) define and characterize the antigenic domains on gM, gN, and the gM/gN complex and relate these to virus strain dependent antibody responses; and (4) determine the relationship between antigenic variation in gN and reinfection of seropositive women. These studies are essential for the development of a vaccine which will protect offspring born to women at risk for infection, especially women at risk for reinfection.
