Heterogeneity among Mycobacterium avium complex species isolated from pulmonary infection in Taiwan

Abstract

Mycobacterium avium complex (MAC) is an emerging pathogen causing nontuberculous pulmonary infections globally. However, clinical treatment guidelines regard MAC as a single entity, recommending a universal anti-mycobacterial combination therapy. Our study aimed to distinguish species among MAC and investigate the antimicrobial susceptibility for the selection of optimal antimicrobial agents in Taiwan. Two hundred ninety-four consecutive sputum samples confirmed as MAC by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry were collected from 1 November 2015 to 31 August 2020 at Linkou Chang Gung Memorial Hospital in Taiwan. These isolates were identified through 16S rRNA gene, 23S rRNA gene, heat-shock protein 65 gene (hsp65), internal transcribed spacer, and beta subunit of RNA polymerase (rpoB) gene sequencing and phylogenetic analyses. Antimicrobial susceptibility testing (AST) was performed with 13 antimicrobial agents. The predominant pathogen identified was Mycobacterium intracellulare clade A (122/294, 41.5%), followed by M. intracellulare subsp. chimaera (87/294, 29.6%), M. intracellulare subsp. intracellulare (39/294, 13.3%), M. avium (35/294, 11.9%), and four other species (11/294, 3.7%). AST showed that clarithromycin and amikacin had high susceptibility rates against M. intracellulare clade A, M. intracellulare subsp. chimaera, M. intracellulare subsp. intracellulare, and M. avium, while linezolid and moxifloxacin exhibited higher resistance. The comparison of minimum inhibitory concentrations among species within the same antimicrobial agent showed variability in susceptibility. A diverse clonality of M. intracellulare might exist in MAC pulmonary infections in Taiwan. Among MAC species, M. avium exhibited multidrug resistance. Although international guidelines recommend macrolide, ethambutol, and rifampin for treating MAC pulmonary disease, our study highlights the importance of considering species identification and regional AST results when selecting anti-mycobacterial agents.IMPORTANCEThere are more than 10 (sub)species within the Mycobacterium avium complex. According to modern biotechnology, such as matrix-assisted laser desorption/ionization time-of-flight mass spectrometry, it is still difficult to differentiate the complex into specific species precisely. We utilize concatenated multi-gene sequencing to classify this complex at the (sub)species level. Indeed, we encountered poorer treatment outcomes when facing Mycobacterium avium pulmonary infections compared to other species causing pulmonary infections. Individualized anti-mycobacterial therapy should focus on each species responsible for pulmonary disease.

Keywords: combination therapy; concatenated gene sequences; genetic diversity; multidrug resistance; phylogenetic analysis.

Fig 1

The phylogenetic tree of full-length 16S rRNA-partial hsp65-partial rpoB of 294 isolates and type strains. The multisequence alignment, including full-length 16S rRNA (1,426 bp), partial hsp65 (354 bp), and partial rpoB (680 bp) genes, was calculated by using the CLUSTAL W program, followed by the neighbor-joining method with Kimura’s two-parameter procedure to construct the phylogenetic tree and displayed as the circular pattern (18).

Fig 2

In vitro susceptibility of 13 antimicrobial agents among the Mycobacterium avium complex (sub)species. GraphPad Prism 9 was used to create statistical graphs for the susceptibility of 13 antimicrobial agents between the species. The MICs of rifampin, rifabutin, and trimethoprim/sulfamethoxazole group comparisons were not included due to the lack of statistical significance based on the Kruskal-Wallis test. A column table was selected to input and organize the MIC data, with each species represented in separate columns. Statistical comparisons between species were performed using the two-sample Mann-Whitney U test. Groups showing statistically significant differences were marked on the graph using asterisks. *P < 0.05; **P < 0.01; and ***P < 0.001.