Genetic Analysis of Gene Expression and the Underlying Polymorphisms in Plasmodium Falciparum

The malaria parasite, Plasmodium falciparum, continues to infect millions of people and kills over a million children annually. Multiple drug resistant (MDR) parasite strains are a primary factor in the prevalence of this disease. In some regions of the world, only a single class of antimalarials, the artemisinin-based compounds, remains effective against MDR malaria. Once parasite sensitivity is lost to these drugs, the people of these regions will be left with no drug to combat the disease. Successful future drug and vaccine policies must incorporate an understanding of how the parasite rapidly adapts to new environments. Ultimately, underlying this adaptation are mutations. These sequence variants range from those that affect single nucleotides (e.g. single nucleotide polymorphisms) to those impacting large regions of the genome (e.g. copy number variants). Mutations may first influence gene expression in the path to downstream phenotypes, and dissection of this dynamic will assist in understanding parasite adaptation. In Chapter 2, expression levels for candidate drug resistance genes are measured in a drug sensitive parasite line, HB3, and a drug resistant parasite line, Dd2, in the presence of chloroquine (CQ). While CQ exposure does not elicit a transcriptional response from the resistant Dd2 line, the sensitive HB3 line stalls in development and gene expression levels remain consistent with its stage of development in the presence of CQ. To further examine whether different parasite strains exhibit variation in gene expression, Chapter 3 describes the mapping of gene expression levels in the progeny of the HB3ÌÄ' Dd2 genetic cross as eQTL. Contrary to previous results suggesting little to no variation in gene expression levels between parasite strains, we observe significant expression level differences between HB3 and Dd2 and their progeny lines. These expression level differences stem from the inheritance of polymorphic regulatory loci across the genome. eQTL map to local regulatory variation (i.e. gene's expression levels mapping back to the gene's location in the genome) and to distant regulatory variation (i.e. gene's mapping distant from their genomic location). As has been reported for other eukaryotic organisms, ~20% of the genome is variably regulated by eQTL, and > 75% of these transcripts are regulated via trans-mechanisms, corresponding to distant eQTL. The chromosome 5 amplification of the pfmdr1 (multi-drug resistance gene) locus is found to associate with expression levels of genes across the genome, highlighting the impact drug selection has on transcriptional regulation in the parasite. To further assess the prevalence of structural polymorphisms in the progeny's genomes, we conducted a comparative genome hybridization (CGH) microarray experiment, presented in Chapter 4. 195 structural polymorphisms larger than 1 kb are detected among the 35 progeny and two parent lines as hybridization signal variants (HSVs). We detect HSVs in one of the parent lines that segregated among the progeny ('segregating HSVs') and non-parental HSVs unique to the progeny lines. The rate at which these HSVs appear in the progeny genomes suggest that the parasite genome is highly plastic. In addition, these HSVs are found to influence transcript levels of those genes residing within the structural polymorphism. Chapter 5 outlines the inheritance of polymorphisms detected on the CGH microarray more closely by using each probe's relative signal intensity as a phenotype for mapping 's'tructural QTL ('s'QTL). sQTL mapped as local, polymorphism maps back to its location in the genome, or distant, polymorphism maps elsewhere in the genome. The local sQTL correspond to a dense set of genetic markers with which to saturate the existing microsatellite linkage map, while the distant sQTL represent loci at which potentially interesting structural polymorphisms may reside. Using direct co-segregation in progeny lines, gene expression data is integrated with the structural data to find that sQTL influence gene expression levels for genes across the genome, as is observed with the Chr 5 amplification. Through these studies, we use novel whole genome approaches to identify genome-wide polymorphisms, specify their influence on transcriptional variation in a segregating population, and provide insights into the biological complexity underlying parasite adaptation.