Targeted deep sequencing to detect heterogeneity in Mycobacterium tuberculosis populations

Da Camara, Ncite Lima (2017-03)

Thesis (MMed)--Stellenbosch University, 2017.

Thesis

ENGLISH ABSTRACT: Antibiotic resistance in Mycobacterium tuberculosis is a worldwide problem as it drastically affects patient treatment outcome. The development of drug resistance is due to the acquisition of mutations in drug resistance conferring genes. Early detection of drug resistance is vital to improve patient therapy and prevent the transmission of drug resistant strains. It is therefore important to develop a method that is able to accurately detect minority variants conferring drug resistance to prevent treatment failure. The aim of this study was to develop an ultrasensitive method to detect underlying resistance causing variants in specific M. tuberculosis fluoroquinolone resistance causing genes (gyrA and gyrB). Efficient primer sets were used to amplify the quinolone resistance-determining region of gyrA and gyrB. Targeted deep sequencing was done using the Ion Torrent Personal Genome Machine (PGM) and Illumina MiSeq platforms and sequencing data were analysed using the appropriate bioinformatics tools for the respective platforms. The method was validated using synthetic heterogeneous mixtures and was subsequently applied to identify underlying variants in patient isolates showing the acquisition of fluoroquinolone resistance. The Illumina MiSeq platform was shown to be superior to the Ion Torrent PGM platform as it accurately detected the correct proportion of mutant DNA to a minimum frequency of 0.1%. We also showed that targeted deep sequencing is sensitive and able to detect underlying variants emerging and fluctuating during the evolution of fluoroquinolone resistance. These results show great promise for the development of an ultrasensitive diagnostic method for the early detection of fluoroquinolone resistance that could ultimately be used to improve the tuberculosis control program.

AFRIKAANSE OPSOMMING: Antibiotikum weerstandigheid in Mycobacterium tuberculosis is ‘n wêreld-wye probleem omdat dit die uitkoms van die behandeling van tuberkulose pasiënte drasties beinvloed. Die ontwikkeling van middelweerstandigheid is as gevolg van die verkryging van mutasies in middelweerstandige veroorsakende gene. Die vroeë identifisering van middelweerstandigheid is belangrik om die behandeling van pasiënte te bevorder, asook om die transmissie van middelweerstandige stamme te verhoed. Dit is dus belangrik om ‘n metode te ontwikkel wat akkuraat is in die identifisering van onderliggende mutasies wat middelweerstandigheid veroorsaak. Die doel van hierdie studie was om ‘n uiters sensitiewe metode te ontwikkel vir die identifisering van onderliggende mutasies in die fluoroquinolone weerstandigheids gene (gyrA en gyrB). Effektiewe inleiers was gebruik om die quinolone weerstandigheids-bepalende area van gyrA en gyrB te amplifiseer. Geteikende diep deoksiribonukleïensuur (DNS) volgorde bepaling was gedoen deur gebruik te maak van die Ion Torrent Personal Genome Machine (PGM) en die Illumina MiSeq platform. Die volgorde bepaling data was geanaliseer deur toepaslike bioinformatika sagteware vir die onderskeie platforms. Die metode was getoets deur gebruik te maak van sintetiese heterogeniese DNS mengsels en was verder toegepas om onderliggende mutasies in pasiënt isolate wat die verkryging van fluoroquinolone weerstandigheid wys, te identifiseer. Die Illumina MiSeq platform was beter as die Ion Torrent PGM platform, siende dat dit met akkuraatheid die korrekte verhouding van mutant DNS by ‘n minimum frekwensie van 0.1% kon wys. Die studie het ook gevind dat geteikende diep DNS volgorde bepaling sensitief en instaat is om onderliggende mutasies te identifiseer wat ontluik en wisselend voorkom tydens die evolusie van fluoroquinolone weerstandigheid. Hierdie resultate is belowend vir die ontwikkeling van ‘n uiters sensitiewe diagnostiese metode vir die vroeë opsporing van fluoroquinolone weerstandigheid wat teneinde die tuberkulose beheerprogram kan verbeter.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/101365
This item appears in the following collections: