(519) Interactions Between Mycobacterium abscessus Dominant Circulating Clones Recovered from People with Cystic Fibrosis and Alveolar Macrophages from Healthy Donors

Background: Mycobacterium abscessus subsp. abscessus (MABS) is a pulmonary pathogen in people with cystic fibrosis (pwCF). MABS infections can be indolent without apparent clinical significance, or may develop into active pulmonary disease requiring intensive antibiotic treatment. Prior genomic studies of MABS isolates from pwCF in the United States (US) identified a highly prevalent dominant circulating clone (DCC1) and genetically diverse strains (non-DCC1). We hypothesized that one or more in vitro phenotypes may explain the fitness of DCC1 isolates in the US CF population and predict clinical outcomes (indolent infection vs. NTM disease).

Methods: We identified a panel of MABS dominant circulating clones (DCC1) and genetically diverse strains (non-DCC1) from pwCF who either met did not meet ATS criteria for NTM pulmonary disease (indolent infection; n=22) or who did meet ATS criteria (n=33). Isolate DCC genotypes were not mutually exclusive with diagnosis of NTM pulmonary disease. This panel of isolates was evaluated for a range of binary in vitro phenotypes including morphology (smooth vs. rough), pellicle formation, presence of glycopeptidolipids (GPL), and inactivation of the cathelicidin antimicrobial peptide (LL-37). We also infected primary human alveolar macrophages (AM) recovered from three healthy donors, and assessed MABS uptake and survival (CFU), and cytokine production (TNFα, MCP-1) at 1h, 24h, 48h, and 96h post infection.

Results: Smooth isolates were more likely to be GPL-positive (p< 0.001; OR = 17.3), and less likely to form pellicles (p< 0.001; OR = 0.05) compared to rough isolates. Rough isolates were more likely to inactivate LL-37 in vitro (p< 0.05) compared to smooth isolates. However, inactivation of LL-37 was not dependent of isolate morphology and presence of GPL. None of the binary in vitro phenotypes was significantly associated with DCC1 genotypes or clinical outcomes. In contrast, MABS uptake and survival in macrophages was significantly higher for non-DCC1 isolates compared to DCC1 isolates (p< 0.001), for GPL-positive isolates compared to GPL-negative isolates (p≤0.001), and for isolates that did not inactivate LL-37 compared to those that inactivated LL-37 (p< 0.001). Furthermore, non-DCC1 isolates from pwCF and NTM pulmonary disease showed higher CFU in macrophages over time compared to non-DCC1 isolates from pwCF and indolent infections (p< 0.01). Increases in CFU over time were positively correlated with increased production of TNFα by infected cells (p< 0.001).

Conclusions: Our data suggest that MABS genotypes and intrinsic phenotypes such as cell wall lipid composition and the ability to inactivate a host-derived antimicrobial peptide, contribute to MABS infectivity and normal immune responses by healthy host cells to infection, and may predict aspects of pulmonary disease in at-risk patient cohorts.

Acknowledgements: This work was supported by the Cystic Fibrosis Foundation (HONDA20A0).

References: