Whole exome sequencing uncovers genetic predisposition to pediatric HIV disease progression and hyperhemolysis syndrome

Abstract

Background: Populations of African ancestry hold a disproportionate burden of infectious and non-infectious diseases. Genetic variants, with population allele frequencies ranging from rare to common variants, are believed to underlie several disease phenotypes. Although genomic studies have gone a long way in understanding the etiologies across this disease spectrum, the high cost of high-throughput sequencing and the general lack of expertise has historically excluded these populations from genomics studies. Whole-exome sequencing (WES) is an approach to increasing the cost-effectiveness of Next Generation Sequencing (NGS) by focusing primarily on the protein-coding genes (the exome) to understand the genetic underpinnings of disease phenotypes. Additionally, WES generates off-target reads that are typically left out of analysis pipelines; however, these contain information that could further broaden its applicability by investigating additional genetic factors such as mitochondrial genetics and viral infections. Method: In this thesis, two datasets were used: a primary CAfGEN dataset, which consisted of a cohort of pediatric HIV disease progression cases, as well as a secondary dataset consisting of patients with hyperhemolysis syndrome. WES was thus used to investigate two disease phenotypes; the first was pediatric HIV Long Term Non- Progression (LTNP), a phenotype seen in about 10% of the population. Here, two African populations totaling 813 children from Uganda and Botswana were studied using WES in a non-traditional fashion by analyzing off-target reads from mitochondrial DNA and viral DNA. The second disease phenotype was hyperhemolysis syndrome (HHS) in sickle cell disease (SCD), a rare disease phenotype seen in less than 5% of the population. For this, WES data from a cohort of 12 HHS patients from the South-West United States was utilized in the traditional approach to investigate the protein-coding variants. Results: Anelloviridae co-infection was found to be associated with an extended AIDS-free period in the African pediatric population. Additionally, lower mitochondrial DNA copy numbers were associated with increased participant age, and with increased duration of highly active antiretroviral therapy (HAART). This suggests a possible role of both HAART and chronic HIV infection in mitochondrial toxicity, building onto previous studies purporting the same. For HHS in SCD, three rare, protein-truncating variants implicating dysfunction in the innate immune system were identified. Conclusions: Rare, putative loss-of-function variants associated with HHS were identified; additionally, analyzing off-target reads from WES also provided more information on the role of the host virome in HIV disease progression. The results illustrate the broad application of NGS to understanding both the host and viral genetic factors influencing two disease phenotypes in populations of African ancestry. As NGS technologies improve, this may also serve as a prototype for future cost-effective sequencing strategies. Such approaches would incorporate host and pathogen genetics in a single sequencing assay to study disease etiology.