Genome sequencing of Mycobacterium abscessus strains that infect the lungs suggests a possible shift in the bacterium's mode of infection from environmental acquisition to human transmission. This finding has clinical implications.
Next-generation DNA sequencing has revolutionized infectious-disease research and surveillance, and allows the evolutionary history of pathogens to be observed at a resolution unimaginable just a decade ago. The technique offers a particularly useful way to investigate outbreaks of any emerging infectious pathogens, because high-throughput testing of bacterial or viral genomes can determine the genetic relatedness of strains. Mycobacterium abscessus causes human disease in several tissues and organs, including the lungs1, and people with cystic fibrosis have a weakened lung defence that makes them particularly vulnerable to M. abscessus infection2. This bacterial infection has increased in prevalence3, which is of concern because of the microbe's virulence and drug resistance. Bryant et al.4 set out to investigate the rise of this infection in individuals with cystic fibrosis, and, writing in Science, report a sequencing analysis of more than 1,000 clinical strains of M. abscessus from around the world.
The Mycobacterium genus is often divided into two groups — one containing the species that cause tuberculosis (Mycobacterium tuberculosis) and leprosy (M. leprae), and the other known as nontuberculous mycobacteria (NTM), encompassing all other species, including M. abscessus. Infections of M. tuberculosis are spread by person-to-person transmission5. By contrast, NTM infections have long been thought to be acquired almost exclusively from the environment6, where the bacteria are omnipresent, occupying diverse natural6 and artificial7 niches. In some cases, these niches serve as reservoirs for the bacteria to spread and infect humans6, although more research is needed to identify the environmental habitats that are the most probable sources of the NTM bacteria that infect humans.
The steady increase in NTM infections in many regions of the world8,9 has resulted in greater awareness among doctors of the global impact of such infections, particularly because some individuals incorrectly diagnosed with tuberculosis have been shown to have an NTM infection instead10. To study NTM infection, Bryant et al. used next-generation sequencing to analyse the genomes of 1,080 strains of M. abscessus isolated from 517 patients in the United Kingdom, continental Europe, the United States, Ireland and Australia. The authors combined the sequencing information with publicly available genomic data to perform phylogenomic comparisons, and to create relational-alignment trees to examine the genetic relatedness of the clinical bacterial isolates that affect individuals with cystic fibrosis.
This analysis confirmed the presence of three M. abscessus subspecies, termed M. a. abscessus, M. a. massiliense and M. a. bolletii. Bryant and colleagues found three large and seemingly clonal clusters of bacterial samples that have near-identical sequences in the analysed genomic regions. Two of these clusters were in M. a. abscessus and one was in M. a. massiliense. Because the strains within each cluster were so similar at the genomic level, the authors inferred that, rather than representing the environmental acquisition of bacteria presumed to be genetically diverse, these strains might represent widespread human transmission of three bacterial clones, which they termed dominant circulating clones (Fig. 1).
The authors propose that the local transmission of such clones could be mediated through asymptomatic human carriers, long-lived infectious cough aerosols or an inanimate intermediate, such as an infected surface. Cystic-fibrosis clinics follow strict infection-control guidelines developed specifically to reduce the potential for patient-to-patient transmission of other types of bacterium that infect the airways in cystic fibrosis. The dominant clones were observed on different continents, and a mechanism for such long-distance transmission has not been identified.
To test whether a strain is truly clonal, the authors compared whole-genome-sequence information over time for strains derived from the same patients. This revealed the extent that a single clone might be expected to evolve within an infected individual, and was used to define a baseline metric for the expected rate of sequence change, when comparing strains from different people. The authors conclude that, if in the analysed regions there are fewer than 20 single DNA-base differences — known as single nucleotide polymorphisms — between strains from different individuals, this suggests a probable human-transmission event, rather than independent environmental acquisition of the strains. However, epidemiological evidence of cross-patient exposures was not available to corroborate this.
Bryant and colleagues conducted infection tests using cellular assays and mouse models to determine whether the dominant circulating clones are more virulent than other M. abscessus strains. Compared with the non-clustered bacterial strains tested, the dominant clones exhibited greater uptake into cells and increased intracellular survival, and yielded higher numbers of bacteria in mice, suggesting that they have adapted to become more robust and successful pathogens. The dominant clones also had more mutations associated with drug resistance and correlated with poorer clinical outcomes. Whether the dominant clones are better at being transmitted than are non-clustered strains was not directly tested.
Although Bryant and colleagues' impressive feat of genomic analysis illuminates the global population structure of a potentially deadly pathogen, further investigation is needed. The genomic diversity of environmental M. abscessus is unknown, and this raises the question of whether environmental bacterial clades might show some genetic overlap with the observed dominant clinical clades. If the genotypes of the dominant transmissible clones are also dominant in the environment, how might this affect the inference of widespread human transmission? To definitively define the occurrence, frequency and existence of a human-to-human transmission mechanism, epidemiological studies must also be conducted, together with genomic pathogen surveillance, as Bryant and colleagues did in a previous study focused on a local M. abscessus outbreak11. Hypotheses that challenge paradigms are essential for scientific progress, but they also warrant careful follow-up and should be accompanied by rigorous testing of the previously held dogma.
Bryant and co-workers' application of high-throughput genomic sequencing of pathogens underscores the importance of continuing to push the boundaries of such technology in biological and medical fields. This approach offers researchers a tool with which to potentially capture snapshots of evolution, identify previously unobserved pathogen characteristics, and monitor pathogens in a state of transition as their infection capabilities or modes of transmission evolve. When Charles Darwin and Alfred Russel Wallace developed the theory of evolution by natural selection, they could only have imagined how scientists' ability to investigate evolution would improve over time to reach the fine-scale analysis that is now possible.