Transcription analysis of virulent strains of Mycobacterium tuberculosis

Abstract

Background: Despite the development of new drugs and success of social programs, tuberculosis

remains a leading cause of mortality. This burden falls disproportionately on

developing countries where the high burden of HIV has a potentiating e↵ect, but may soon

return to areas where it was previously brought under control as resistant strains continue

to emerge. In the Western Cape, two closely related strains of the Beijing family have been

isolated that provide an opportunity to study virulence in a system with relatively little noise.

The aim of this project was to identify the cause of the altered virulence displayed between

the two strains, and describe how the di↵erences between the two genomes contributed to

the phenotypic di↵erences.

Results: GenGraph allows for the creation of graph genomes, and facilitated the creation

of a pan-transcriptome that allowed for the mapping of gene annotations between

isolates. This allowed for the mapping of reads to a more suitable Beijing family reference

while interpreting the results with annotations from the H37Rv reference. We generated

expression and target profiles for the known sRNA, and identified a large number of novel

sRNA. Transcriptomic data from 4 di↵erent growth conditions was integrated with this

sRNA data as well as variant data using the Cell pipeline. From this data we identified

multiple sets of genes linked to copper sensing in MTB, including the di↵erentially expressed

MoCo operon. Increasing evidence that macrophages use copper to poison bacteria trapped

in their phagosomes provides the link to virulence and pathogenicity.

Conclusions: Through the integration of data from multiple data types we were able

to elucidate the most probable cause of the altered virulence found between the two isolates

in this study. We developed reusable tools and pipelines, and noted a large number of

undescribed sRNA expressed in these isolates. The identification of the copper response as

a chief contributor to the phenotype increases both our understanding of the isolates, and

the role of the element in infection. These results will be key in guiding further investigation

of the variant linked genes to identify those linked to copper homeostasis or response.