NEW YORK (GenomeWeb News) – A bevy of studies appearing this week in Nature Genetics traced the evolution of the tuberculosis-causing bacterium from its pre-Neolithic roots to today's modern, and increasingly drug-resistant, strains.
As tuberculosis appears to have traveled with humanity as it spread all over the world, researchers speculated that a better understanding of how the disease has evolved resistance to a number of drugs — a process that appears to be more complex than initially anticipated — may eventually help diagnose resistant disease or even prevent such resistance from cropping up.
Currently, tuberculosis, if left untreated, kills half of the people it infects, and while there has been a decline in the number of TB cases, it causes between 1 million and 2 million deaths each year, mostly in developing nations. However, a number of strains are developing drug, and even multi-drug, resistance. In China, for example, nearly 6 percent of new TB cases are multi-drug resistance and 8 percent are extensively drug resistant.
"We don't really understand why resistance develops so consistently," said Megan Murray, a professor of global health and social medicine at Harvard Medical School, in a statement. "[Our study examining selection in resistant strains] may provide a lens that we can use to see a way to develop better diagnostics for impending resistance, or even ways to prevent it from happening."
Mycobacterium tuberculosis has been stalking and co-existing with people for thousands of years, and researchers led by Sebastien Gagneux, an assistant professor at the Swiss Tropical and Public Health Institute in Basel, reported in Nature Genetics this week that they've traced the evolutionary history of TB back 70,000 years.
By sequencing 186 different strains of tuberculosis from the world over, Gagnaeux and his colleagues identified some 34,000 SNPs that they used to construct a phylogenetic tree for TB. Two Bayesian analyses and one parsimonious analysis indicated that TB likely emerged in Africa.
They also compared their phylogenetic tree for TB to one constructed for humans based on nearly 5,000 mitochondrial genomes, finding them to be very much alike.
"The evolutionary path of humans and the TB bacteria shows striking similarities," Gagneux said in a statement.
Further, as people expanded out of Africa and later into denser population centers, TB appeared to come along for the ride. "We see that the diversity of tuberculosis bacteria has increased markedly when human populations expanded," he added.
Tuberculosis is also evolving regionally, Gagneux and his colleagues noted. They focused part of their analyses on the Beijing family of strains by sequencing the genomes of an additional 39 lineage 2 strains isolated in China — Beijing strains are typically marked by increased drug resistance. The lineages leading to these strains, the researchers found, arose around the Neolithic period as people began to populate and expand into East Asia.
TB, they noted, "has been a constant companion of anatomically modern humans during our evolution and global dissemination over the last 70,000 years."
A separate study appearing in the same journal took a global look at genes and intergenic regions linked to drug resistance in TB. The researchers, led by Lijun Bi from the Chinese Academy of Sciences, sequenced some 160 isolates of M. tuberculosis with a variety of drug-resistance profiles.
From this, they identified a number of known drug-resistance genes like gyrA, rpoB, embB, rrs, and others, as well as novel ones, including 72 new genes, 28 intergenic regions, 11 nonsynonymous SNPs, and 10 intergenic region SNPs associated with drug resistance. While many of the newly uncovered genes are not well characterized, some were linked to pyrimidine metabolism, DNA, topisomerase, or chemical response.
They found that extreme drug resistance appeared to emerge due to an accumulation of nonsynonymous SNPs, rather than the development of a mutation or mutations that conferred pan-resistance.
By comparing the ratio of nonsynonymous SNPs to synonymous SNPs, the researchers found that drugs to treat the disease have exerted a positive selective force on the genome, with the effect more pronounced on drug-resistance genes. This suggests to Bi and his colleagues that "the genetic basis of drug resistance is more complex than previously anticipated."
Indeed, certain regions of the M. tuberculosis genome appeared to be under such positive selection, and Murray and her team from Harvard and elsewhere reported in Nature Genetics that they used those targets of selection to develop biomarkers of drug resistance in TB.
They sequenced nearly 120 M. tuberculosis isolates, some of which were sensitive to drug treatments as well as others that showed signs of emergent drug resistance. By combining that with deep-sequencing data from an additional 35 resistant isolates and public data, the researchers constructed a phylogenetic tree to identify mutations that arose independently.
Using a phylogenetic convergence test, they identified mutations, genes, and intergenic regions linked to resistance to a number of first-line TB drugs, including 11 known and 39 novel targets. Many of the novel targets were linked to DNA repair or cell wall biogenesis.
One family in particular, the PE/PPE gene family appeared to host a number of targets of independent mutation, many within the PGRS subfamily, Murray and her colleagues noted. Members of this family encode surface-exposed cell wall proteins, affect cell wall structure and permeability, or are antigens.
"[T]he preponderance of regions were associated with cell wall permeability phenotypes," Murray and her team noted. "This finding suggests that stable drug resistance phenotypes may evolve through a complex stepwise process involving cell wall remodeling."
A separate team led by David Alland at Rutgers University also found that drug resistance in TB seemed to come about in a stepwise fashion. As they reported in Nature Genetics this week, Alland and his group focused on TB that is resistant to ethambutol, a first-line treatment for the disease.
They selected a number of M. tuberculosis strains grown in vitro for ethambutol resistance. By examining both high- and low-level resistance, they uncovered mutations that appeared to act as a sort of gateway for drug resistance.
For example, they found that mutations in certain codons of embB lead to ethambutol resistance, but only at low levels of the drug, suggesting to the researchers that this canonical resistance mutation is a first-step mutation in the development of drug resistance.
Other mutations, such as ones in embC or Rv3806c, they added, may combine to bring about high levels of ethambutol resistance.
"Our study contradicts commonly held beliefs that drug resistance in M. tuberculosis is caused by single-step mutations," Alland and his group said. "These conclusions are clinically relevant, especially if they are found to be applicable to other tuberculosis drugs."
With more resistance-related genes in hand, researchers may be able to work out their roles and better detect drug resistance in the future. "We have found that more genes might be implicated in resistance than previously thought, and this means that we can start to unravel the role of these genes," added Murray, referring to her work. "This is significant because it implicates new mechanisms in the evolution of resistance that can be further studied now and raises the possibility of more specific targets for the detection of resistance through molecular methods."