Population Genomics of Nontuberculous Mycobacteria Recovered from United States Cystic Fibrosis Patients

Abstract

Nontuberculous mycobacteria (NTM) pose a threat to individuals with cystic fibrosis (CF) due to an increased prevalence of pulmonary infections, innate drug resistance of the bacteria, and potential transmission between CF patients. To explore the genetic diversity of NTM isolated from CF patients within the United States (US) and to identify potential transmission events, we sequenced and analyzed the genomes of 341 NTM isolates from 191 CF patients as part of a nationwide surveillance study. The most abundant species in the isolate cohort were Mycobacterium abscessus (59.5%), followed by species in the Mycobacterium avium complex (37.5%). Phylogenomic analyses of the three M. abscessus subspecies revealed that more than half of CF patients had isolates in one of four dominant clones, including two dominant clones of M. abscessus subspecies abscessus and two dominant clones of M. abscessus subspecies massiliense. M. avium isolates from US CF patients, however, do not have dominant clones and are phylogenetically diverse. Longitudinal NTM isolates were compared to determine genome-wide single nucleotide polymorphisms (SNPs) that occur within patients over time. This information was used to compare between and within-patient SNP distributions, to quantitatively define SNP thresholds suggestive of transmission, and calculate a posterior probability of recent transmission given the SNP distance between two isolates from different patients. Out of 114 patients with M. abscessus subspecies, ten clusters of highly similar isolates from 26 patients were identified. Among the 26 patients in the M. abscessus clusters, 12 attended the same CF care centers. No highly similar isolate clusters were observed in M. avium. Our study reveals the contrasting genomic diversity and epidemiology of two major NTM taxa and the potential for between-patient exposure and cross-transmission of these emerging pathogens.

Characterization of integrated prophages within diverse species of clinical nontuberculous mycobacteria

“…The second dataset includes a collection of 318 NTM isolates from 168 individuals diagnosed with cystic fibrosis (CF) [31]. All samples were cultured on solid media, converted to glycerol stock aliquots, and the DNA was extracted for whole genome sequencing using an optimized mycobacterial DNA preparation protocol [31,32]. Bacterial isolates were collected in a longitudinal manner, however, only one isolate genome per patient per species was retained for this prophage analysis.…”

­­­Characterization of integrated prophages within diverse species of clinical nontuberculous mycobacteria

“…The second dataset includes a collection of 318 NTM isolates from 168 individuals diagnosed with cystic brosis (CF) (31). All samples were cultured on solid media, converted to glycerol stock aliquots, and the DNA was extracted for whole genome sequencing using an optimized mycobacterial DNA preparation protocol (31,32). Bacterial isolates were collected in a longitudinal manner, however, only one isolate genome per patient per species was retained for this prophage analysis.…”

“…Species identi cations of draft genomes were determined using a method of average nucleotide identity (ANI) as described previously (31) and sequence reads mapped to active reference genomes to generate phylogenies (34,35). Reads were mapped using Bowtie2 software (36) and single nucleotide polymorphisms (SNPs) were determined using Samtools mpileup (37).…”

­­­Characterization of integrated prophages within diverse species of clinical nontuberculous mycobacteria

“…The second dataset includes a collection of 318 NTM draft genomes from 168 individuals diagnosed with cystic brosis (CF) (24). Isolates were collected in a longitudinal manner, however, only one isolate genome per patient per species was retained for prophage analysis.…”

“…Species identi cations of draft genomes were determined using a method of average nucleotide identity (ANI) as described previously (24) and sequence reads mapped to active reference genomes to generate phylogenies (26,27). Reads were mapped using Bowtie2 software (28) and single nucleotide polymorphisms (SNPs) were determined using Samtools mpileup (29).…”