Inter-strain recombinants of human cytomegalovirus as an in vitro model of in vivo genetic diversity.

Pathogen genetic diversity presents an obstacle to effective interventions against infectious diseases. The development of drugs and vaccines often rely on in vitro models that poorly reflect the real-world genetic diversity of the pathogen. This is the case for human cytomegalovirus (HCMV), which despite vast public and private research investments over 50 years, continues to harm immunocompromised adults and congenitally infected children. Nineteen of the canonical 170 genes of HCMV have 2-14 fixed alleles with inter-allelic variation up to 30% of predicted amino acids. The allelic genes are distributed throughout an otherwise conserved genome with low linkage disequilibrium. Mixed infections in people are common and pervasive recombination can shuffle the allelic genes into a vast number of recombinant variant haplotypes. Research into the molecular biology of HCMV, the basis of intervention development, typically utilizes one or another of a handful of reference strains. We have published extensive comparative analyses of 3 genetically distinct reference strains of HCMV and documented dramatic phenotypic variation in glycoprotein display, specific infectivity, the preferences for cell- free vs direct cell-to-cell modes of spread and sensitivity to neutralizing antibodies and anti-HCMV drugs. To investigate the genetic correlates of these phenotypic differences, we have generated a recombinant library by coinfecting cells with 2 reference strains that differ at 17 of the 19 allelic loci and allowing recombination. Over 1100 isolates were recovered and low resolution RFLP analyses of 15 confirmed complex recombination crossover patterns. The observed phenotypes among these recombinant isolates span the gamut between the 2 parental phenotypes, suggesting polygenic phenotypes that are influenced by genome-wide epistasis. The goal of this R03 project is to capture recombinants on bacterial artificial chromosomes (BAC) to facilitate comprehensive genomic characterization of the recombination patterns, de novo point mutations or other genetic alterations in relation to the parental strains. Future projects will use this set of BAC clones as an in vitro model of the vast recombinatorial diversity of HCMV in vivo for studies of how complex genome-wide epistasis influences viral phenotypes and selection through bottlenecks imposed by antiviral drugs and neutralizing antibodies.