Unravelling the significance of genotype-specific immunity against cytomegalovirus

CMV infections cause significant morbidity and mortality in pregnant women and their children, immunocompromised or suppressed patients, HIV-infected patients with advanced disease, as well as patients treated with monoclonal antibodies (Abs) for leukemia. A safe, protective, and cost-effective vaccine against Cytomegalovirus (CMV) infection is urgently needed. This need could not be met during the past four decades. Protection rates of 50% were attained in healthy adults recently in a trial of a vaccine based on the glycoprotein B antigen (Ag) a large step forward but still too short to make a significant difference in clinical terms. We strongly believe that the high genetic variability in Ag-coding regions of the CMV genome contributes at least in part to efficacy of CMV vaccines: (1) CMV is considered a genetically stable virus but sequencing of DNA from clinical CMV strains revealed highly complex dynamics of CMV genotypes over time and in different patient populations. In a previous study, we found evidence for intragenic recombination between two different CMV genotypes which may have been responsible for the emergence of novel CMV genotypes. (2) Particularly in regions coding for antigenic envelope proteins, which are also commonly used as vaccine Ags, considerable genetic variability was observed. (3) Pre-existing immunity to one CMV genotype does not protect from re-infection with a different CMV genotype. In a pilot project, we developed a versatile Ag microarray that is based on the in vitro expression of recombinant proteins including all mammalian posttranslational modifications. This Ag microarray allows the screening of large numbers of human immune sera against multiple different viral Ags derived from multiple different microbial subtypes. We propose to characterize comprehensively the cross-reactivity of polyclonal immune sera against the major targets of neutralizing CMV immunity (envelope glycoproteins gB, gM, gN, gH and gL) with use of this novel microarray. For this purpose, glycoprotein-coding sequences derived from multiple clinical CMV strains will be used for in vitro protein expression, analysis of genetic distances, and Ab cross-reactivity. Antigen microarray results will be validated with use of standard immunological methods (immunoblots, indirect EIA, ELISPOT) and mouse studies. The proposed study will provide important insights into heterosubtypic immunity against infectious pathogens as well as the immunobiology of CMV - knowledge that may greatly facilitate the development of effective vaccines against antigenically diverse infectious pathogens.

Development of an effective vaccine against human cytomegalovirus (HCMV) infection is a high priority due to significant morbidity and mortality in newborns and immunocompromised patients. The glycoprotein B (gB) of HCMV is a dominant target of immunity and therefore used in multiple vaccines candidates. Nevertheless, vaccine trials have been unsuccessful so far. We hypothesized that the highly divergent genome of CMV, with gB as one of the most divergent proteins (Renzette et al, 2015), may allow immune escape from vaccine induced, single-strain immunity. To study cross-reactive, heterosubtypic immunity against the different clinical strains of HCMV, we developed a versatile antigen (Ag) microarray that is based on the in vitro expression of recombinant proteins including all mammalian posttranslational modifications. This Ag microarray allows the screening of large numbers of human immune sera against multiple different viral Ags derived from multiple different microbial subtypes. We could show that the protein in situ array (PISA) protocol we used, is a highly sensitive method and can be used to differentiate between genetic variants. We found that there is extensive cross-reactivity between CMV gB antigenic domain (AD) proteins. Recognition of infecting CMV genotype was not the dominant response in several cases. Primary infected patient sera showed significantly higher reactivity to AD5 proteins than to any other gB AD. Our results show that there is extensive cross-reactivity of CMV-specific antibodies to different subtypes of gB proteins. Thereof we suggest that a highly, strain-specific immunity is not a viable explanation for the failure of CMV gB vaccines.