Detection of genomic changes in Mycobacterium tuberculosis

Resistance development in Mycobacterium tuberculosis

Tuberculosis (TB) remains the leading cause of death due to bacterial infections worldwide. About 8 million new cases of active TB arise each year resulting in 3 million annual deaths. Roughly 1 billion individuals are believed to harbour latent tuberculosis.

Primary infections with M. tuberculosis are generally asymptomatic but will in some cases remain as a latent infection.

Tuberculosis is a secondary disease caused by reactivation of the M. tuberculosis bacteria not fully eliminated after the primary infection.

Sequencing and sequence analysis of bacterial and fungal genomes and proteins/peptides have led to a better general understanding of the pathogenesis of bacterial and fungal infections. Future understanding of the regulatory events at the molecular level will increase and be accelerated by using a variety of new technologies and technology platforms within microarrays, protein chips and sequence analysis tools (CLC bio workbenches). The aim is to develop more specific and effective drugs much faster to target e.g. the expanding multi resistance Mycobacterium tuberculosis strains in human populations all over the world.

Multidrug resistant bacterial strains arise by sequential accumulation of resistance mutations for individual drugs. A diverse array of strategies is available to assist in rapid detection of drug resistance-associated gene mutations.

Bioinformatics (CLC bio workbenches) together with functional genomics and functional proteomics have been used to identify expression pattern (signature) changes in multi resistant M. tuberculosis strains.

Mycobacterium tuberculosis genome

The Mycobacterium tuberculosis genome was sequenced in 1998 [Cole et al., 1998] (see figure 1). Today many strains of M. tuberculosis have been sequenced. In pest control laboratories, sequencing is routinely performed to monitor strain genomic changes.

Presently with the latest re-annotation of the M. tuberculosis genome, it is possible to assign a function to 2058 protein coding sequences (52% of the 3995 proteins predicted). Only 376 putative proteins share no homology with known proteins and thus could be unique to M. tuberculosis.

Drug resistance in Mycobacterium tuberculosis

Today eight agents are used in the treatment of TB. Multi drug resistant M. tuberculosis strains are resistant both to isoniazid (INH) and rifampicin (RFP). These agents have the most effective bactericidal activity towards M. tuberculosis. Nearly 95% of the RFP resistant strains possess a mutation in the rpoB gene encoding a DNA-dependent RNA polymerase. Approximately 90% of INH resistant strains have a mutation in the inhA, katG, and ahpG genes encoding enzymes related to a mycolic acid synthesis of cell wall. Pyrazinamide (PZA) resistant strains have a mutation in the pncA gene encoding a pyrazinamidase which degrades PZA to the bactericidal substance, pyrazinoic acid. Streptomycin resistant strains have a mutation in the rrs and rpsL gene encoding the 16S rRNA and the 12S ribosomal subunit protein, respectively. Kanamycin resistance is due to nucleotide substitutions in the rrs gene encoding 16S rRNA. Ethambutol resistant strains have a mutation in the embB gene encoding a arabinosyl transferase which catalyzes cell wall synthesis. Ethambutol resistance is in approximately 60% of organisms due to amino acid replacements at position 306 of an arabinosyltransferase encoded by the embB gene.

CLC bio provide some important bioinformatics solutions for detection and identification of mutations in selected genes of new strains of M. tuberculosis and other important pathogenic microbes. An example of this is described below.

Detection of mutations in Mycobacterium tuberculosis embB, rrs and pncA genes

The work flow for detecting and identifying mutations using the CLC Combined Workbench is summarized in figure 2.

The multi resistance M. tuberculosis strains are identified and the genomic DNA extracted. The encoding region of the embB, rrs and pncA genes are marked with primers using the graphical primer design. The relevant regions are sequenced automatically and assembled into contigs. BLAST searches in selected databases (both local and NCBI databases) are performed. Subsequently homologous sequences are aligned and annotations are transferred to the contig sequence.

In the next two sections, two of the steps in the work flow are described in further detail.

Zooming in on Primer design

Using the integrated GenBank search function, the annotated M. tuberculosis is downloaded. In a few clicks, the relevant genes (embB, rrs and pncA) are extracted. Guided by the CDS annotations, a set of primers are calculated for each gene (see figure 3). Using the interactive primer design functionality of the CLC Combined Workbench, the primers can be designed to match very specific needs.

Zooming in on Assembly

The sequencing data is imported into the CLC Workbench, and during the assembly, the genes are automatically divided into separate contigs.

The inconsistencies that exist between different reads are inspected using the variance table (see figure 4).

The reads are both forward and reverse, which are automatically detected during the assembly. The orientation is reflected in the color of the reads (in figure 4, the read at the top is reversed which is indicated by the red color of the residues). Quality scores are assigned in order to trim low-quality trace data, so that reads align properly. The quality scores can also be shown graphically below the sequence.

The result of the assembly is a contig sequence which is used in the following steps for BLAST searches and alignments.

The description above is only a brief example of how bioinformatics software, such as the CLC Workbenches, can be used in clinical microbiology. Although remarkable advances have been made, much remains to be learned about the molecular genetic basis of drug resistance development including M. tuberculosis. New therapeutics will be developed based on improved bioinformatic data-mining of results from bacterial drug resistance studies.