Profiling genome-wide recombination in Epstein Barr virus reveals type-specific patterns and associations with endemic-Burkitt lymphoma

Abstract

Background: Endemic Burkitt lymphoma (eBL) is potentiated through the interplay of Epstein Barr virus (EBV) and holoendemic Plasmodium falciparum malaria. To better understand EBV's biology and role in eBL, we characterized genome-wide recombination sites and patterns as a source of genetic diversity in EBV genomes in our well-defined population of eBL cases and controls from Western Kenya.

Methods: EBV genomes representing 54 eBL cases and 32 healthy children from the same geographic region in Western Kenya that we previously sequenced were analyzed. Whole-genome multiple sequence alignment, recombination analyses, and phylogenetic inference were made using multiple alignment with fast Fourier transform, recombination detection program 4, and molecular evolutionary genetics analysis.

Results: We identified 28 different recombination events and 71 (82.6%) of the 86 EBV genomes analyzed contained evidence of one or more recombinant segments. Associated recombination breakpoints were found to occur in a total of 42 different genes, with only 7 (16.67%) being latent genes. Recombination events were major drivers of clustering within genome-wide phylogenetic trees. The occurrence of recombination segments was comparable between genomes from male and female participants and across age groups. More recombinant segments were found in EBV type 1 genomes (p = 6.4e - 06) and the genomes from the eBLs (p = 0.037). Two recombination events were enriched in the eBLs; event 47 (OR = 4.07, p = 0.038) and event 50 (OR = 14.24, p = 0.012).

Conclusions: EBV genomes have extensive evidence of recombination likely acquired progressively and cumulatively over time. Recombination patterns display a heterogeneous occurrence rate across the genome with enrichment in lytic genes. Overall, recombination appears to be a major evolutionary force impacting EBV diversity and genome structure with evidence of the association of specific recombinants with eBL.

Keywords: Endemic-Burkitt lymphoma; Epstein–Barr virus; Genome-wide recombination.

Fig. 1

Phylogenetic tree of EBV isolates showing diversity related to genomic recombination events. The analysis involved 86 EBV genome sequences. The evolutionary history was inferred using the NJ method. Evolutionary distances were computed using the Jukes–Cantor model. Ambiguous nucleotides were removed using pairwise deletion. Bootstrap analysis of 5000 replicates was performed. A circular heatmap was used to annotate the tree using the 28 recombination events detected

Fig. 2

Recombination events per genome stratified by A Viral Type, B BL Status, and C Viral Type Relative to BL Status. Center Lines represent medians, with lower and upper boundaries of the boxes representing first and third quartiles respectively. Wilcoxon (A, B and C) tests were performed and p value < 0.05 was considered significant

Fig. 3

EBV genome map with positions of recombination breakpoints. From outer to inner, circles display genomic positions for (i) gene positions, (ii) breakpoints, (iii) aligned covered regions, (iv) repetitive regions, and (v) scale. Genes are color-coded based on the gene exons. Genes on the outside are transcribed clockwise and the inner are counterclockwise. This figure was drawn by GenomeVx

Fig. 4

Distribution of recombination breakpoints in EBV coding sequences. Abbreviations. Kbp, Kilobase pair; CDS, Coding Sequence. Each colored bar represents an EBV gene. The total number of CDS = 42. Of the 42 CDS, 6 (14.3%) are latent genes and 36 (85.7%) are lytic genes. The bars are colored according to the classification of the genes: EBV lytic cycle,Blue; Immediate Early genes, Red; Early lytic genes, Green; Late Lytic genes, Purple; Latent genes, Brown; and Uncharacterized genes, Maroon. The black dotted strip denotes the 3rd interquartile for the number of recombination breakpoints per Kbp for all the 42 genes (2.0)