Diversity and Evolution of Human Cytomegalovirus

Human cytomegalovirus (HCMV) is a ubiquitous virus that infects the majority of the worldwide population through contact of bodily secretions. Like other herpesviruses, HCMV cannot be cleared and remains latent for the lifetime of its host, providing a source of persistent or sporadic virus secretion and resulting transmission. While the virus can occasionally cause febrile illness or infectious mononucleosis in immunocompetent adults and children, infections mostly pass unnoticed. HCMV is, however, the leading infectious cause of congenital disease and intrauterine transmission to the fetus can result in sensorineural hearing loss and other neurodevelopmental damage. Moreover, it causes clinically important opportunistic infections in individuals with a compromised immune system, particularly related to defects in cell-mediated immunity. Development of a vaccine is prioritized, but no candidate is close to market release. HCMV has the largest genome of all viruses that infect humans (235 kb). From neutralization, hybridization and restriction assays, the genetic and antigenic heterogeneity of HCMV isolates became apparent long before the first complete genome sequence was obtained. While this variation could have important implicationsnbsp;the understanding of virus biology, viral determinants of pathogenesis and the development of therapies and prophylaxis, only a handful of complete genome sequences from clinical isolates were publicly available at the start of this PhD project. Since the number of full-genome sequences was limited, knowledge about strain diversity was highly fragmented and biased towards a small set of loci. Characterizing the complete genome of a cytomegalovirus strain with the previous generation of sequencing technologies is a laborious task because of the relatively long length of its genome. The first aim of this PhD project therefore was the implementation of the new generation of sequencing technologies to analyze completenbsp;genomes in high-throughput. To achieve this, a sample preparation and sequence assembly workflow had to be developed to generate viral DNA from clinical isolates suitable for next-generation sequencing analysis and reconstruct the viral consensus sequence from small sequencing reads (Chapter 2). The presented method relies on limited passaging of clinical isolates on human fibroblasts, followed by digestion of cellular DNA and purification of viral DNA. After multiple displacement amplification, highly pure viral DNA is generated. These extracts are suitable for high-throughput next-generation sequencing and assembly of consensus sequences. Throughout a series of validation experiments, we have shown that the workflow reproducibly generated consensus sequences representative for the virus population present in the original clinical material. Additionally, the performance of 454 GS FLX and/or Illumina Genome Analyzer datasets in consensus sequence deduction was evaluated. Based on assembly performance data, the Illumina Genome Analyzer was the platform of choice in the presentednbsp;the established workflow was employed to characterize genetic diversity in a collection of 100 clinical isolates, quadrupling the available information for full-genome analysis (Chapter 3). These data provided the most highly resolved map of HCMV interhost diversity and evolution to date. We showed that cytomegalovirus is significantly more divergent than all other human herpesviruses and highlighted hotspots of diversity in the genome. Importantly, 75% of strains were not genetically intact, but contained disruptive mutations in a diverse set of 26 genes, including immunomodulative genes UL40 and UL111A. These mutants were independent from culture passaging artifacts and circulated in natural populations. Pervasivenbsp;linked to the widespread occurrence of multiple infections, was found throughout the genome. Recombination density was significantly higher than in other human herpesviruses and correlated with strain diversity. While the overall effects of strong purifying selection on virus evolution were apparent, evidence of diversifying selection was found in several genes encoding proteins that interact with the host immune system, including UL18, UL40, UL142 and UL147. These residues may present phylogenetic signatures of past and ongoing virus-host interactions. The presented data are of major value to the development of novel antivirals and a vaccine and to identify potential targets for genotype-phenotype experiments. Furthermore, they have enabled a thorough study of the evolutionary processes that have shaped cytomegalovirus diversity. Like other herpesviruses, HCMV has co-evolved with its human host and it encodes a DNA polymerase enzyme with proofreading capacity. It was therefore assumed that herpesviruses evolve slowly. This was confirmed by the sequence stability of several cytomegalovirus genes inside patients over timescales of one or a few years. Recent studies have, however, identified variability in intrahost viral populations of congenitally infected infants that was comparable to quickly evolving RNA viruses and have suggested this could have implications for HCMV intrahost evolution on short timescales. To further evaluate these conflicting data, we have used our HCMV genome sequencing pipeline to characterize full-genome sequences in serial patient isolates sampled for periods of up to 18.5 months (Chapter 4). Consensus sequences were extremely stable with a maximum of 1 nucleotide substitution per genome. Furthermore, we applied an alternative method to infer population diversity in a pulmonary transplant recipient and found only 208 intrahost variants, of which 111 were length polymorphisms in homopolymer regions. Similar analyses of intraculture populations showed that this method can identify low-frequency sequence changes associated with culture passaging before they are present in the genome consensus. Finally, we have discussed some caveats based on our analyses that should be considered when characterizing HCMV intrahost diversity. While infections with multiple strains are common and can create considerable intrahost diversity and the capacity for evolution through genetic recombination, we believe that intrahost diversity of HCMV populations from a single strain is limited and not comparable to that of quasispecies RNA viruses. While diversity of isolates from European patients had been well documented through the analysis of over 100 complete genome sequences, sequence knowledge for other continents, Africa in particular, was severely limited and mostly restricted to a small set of loci. To provide a more complete picture of cytomegalovirus diversity in Africa, we have analyzed a broad set of 16 genes in a population of Gambian infants (Chapter 5). This selection included both the top 11 of variable loci and the most frequently disrupted open reading frames. These data were analyzed in conjunction with the available sequences from other continents to establish a global picture of HCMV diversity. African isolates contained gene-disruptingnbsp;in genes RL5A, RL6, UL1, UL9, US7 and US9. Several mutations were identical to those identified in European patients, suggesting their widespread circulation and ancient origin. For all variable genes under study, Gambian sequences clustered together with isolates from other continents. For genes RL6, UL9, UL73, UL139, UL144 and UL146, genotype frequencies did fluctuate significantly between different continents. Our analyses further illustrated the impact of recombination on the generation of cytomegalovirus diversity. Lastly, we analyzed serial isolates from Gambian infants and highlighted the stability of genotype sequences. In contrast to other herpesviruses like herpes simplex virus 1 and varicella zoster virus, there seems to be no clear geographical discrepancy in the genetic diversity of HCMV. This result supports the idea that our high-throughput analysis in European patients provides a good overview of worldwide genetic diversity. The commercial introduction of next-generation sequencing technology has changed the scope and pace of genomic research, including that on HCMV. Sequencing the complete genome of a clinical HCMV isolate is now possible in a considerably higher throughput and pace than a few years ago. This has produced a novel resolution on genome-wide interhost diversity. Epidemiological studies that search for viral determinants of pathogenicity on a genome-wide scale are now achievable.nbsp;RNA-Seq and ribosome profiling applications have altered our understanding of the transcriptional and translational complexity during HCMV infection. Deep sequencing applications are providing novel insights on HCMV intrahost diversity, but results are contradictory and methodological issues need to be sorted out. In the meantime, anbsp;generation of single-molecule sequencing technologies are finding their way to the market, giving novel research opportunities that will undoubtedly assist in understanding and controlling this ubiquitous but complex herpesvirus.status: publishe