Lsr2, a pleiotropic regulator at the core of the infectious strategy of Mycobacterium abscessus

Abstract

Mycobacterium abscessus is a non-tuberculous mycobacterium, causing lung infections in cystic fibrosis patients. During pulmonary infection, M. abscessus switches from smooth (Mabs-S) to rough (Mabs-R) morphotypes, the latter being hyper-virulent. Previously, we isolated the lsr2 gene as differentially expressed during S-to-R transition. lsr2 encodes a pleiotropic transcription factor that falls under the superfamily of nucleoid-associated proteins. Here, we used two functional genomic methods, RNA-seq and chromatin immunoprecipitation-sequencing (ChIP-seq), to elucidate the molecular role of Lsr2 in the pathobiology of M. abscessus. Transcriptomic analysis shows that Lsr2 differentially regulates gene expression across both morphotypes, most of which are involved in several key cellular processes of M. abscessus, including host adaptation and antibiotic resistance. These results were confirmed through quantitative real-time PCR, as well as by minimum inhibitory concentration tests and infection tests on macrophages in the presence of antibiotics. ChIP-seq analysis revealed that Lsr2 extensively binds the M. abscessus genome at AT-rich sequences and appears to form long domains that participate in the repression of its target genes. Unexpectedly, the genomic distribution of Lsr2 revealed no distinctions between Mabs-S and Mabs-R, implying more intricate mechanisms at play for achieving target selectivity.IMPORTANCELsr2 is a crucial transcription factor and chromosome organizer involved in intracellular growth and virulence in the smooth and rough morphotypes of Mycobacterium abscessus. Using RNA-seq and chromatin immunoprecipitation-sequencing (ChIP-seq), we investigated the molecular role of Lsr2 in gene expression regulation along with its distribution on M. abscessus genome. Our study demonstrates the pleiotropic regulatory role of Lsr2, regulating the expression of many genes coordinating essential cellular and molecular processes in both morphotypes. In addition, we have elucidated the role of Lsr2 in antibiotic resistance both in vitro and in vivo, where lsr2 mutant strains display heightened sensitivity to antibiotics. Through ChIP-seq, we reported the widespread distribution of Lsr2 on M. abscessus genome, revealing a direct repressive effect due to its extensive binding on promoters or coding sequences of its targets. This study unveils the significant regulatory role of Lsr2, intricately intertwined with its function in shaping the organization of the M. abscessus genome.

Keywords: Lsr2; Mycobacterium abscessus; S/R morphotypes; antibiotic resistance; nucleoid-associated protein; transcription factor.

Fig 1

Analysis of Lsr2 regulon in M. abscessus morphotypes. (A) Volcano plots showing differentially expressed genes (DEG) (red) in Mabs-S-Δlsr2 vs Mabs-S-WT (left) and Mabs-R-Δlsr2 vs Mabs-R-WT (right). Genes exhibiting a greater than two fold changes in expression between two conditions [log₂ fold change (log₂FC) ≤−1 or ≥1] and with a false discovery rate (FDR) of less than 0.05 are considered significantly deregulated. Genes with log₂FC ≥1 correspond to downregulated genes by Lsr2 (repressed genes), and genes with log₂FC ≤−1 correspond to upregulated genes by Lsr2 (activated genes). (B) Plot representation showing all genes regulated by Lsr2 at least in one of the two morphotypes of M. abscessus with a log₂FC ≥1 and log₂FC ≤−1 of gene expression between the Δlsr2-mutant strains and wild-type strains. (C) Gene ontology (GO) enrichment analysis was performed with the topGO R package. Enriched GOs (Biological Process in blue and Molecular Function in red) are sorted according to their enrichment factor, corresponding to the ratio of significant DEGs assigned to the GO over expected assigned DEGs to the GO as defined by topGO. Enriched GOs are represented by circles whose size is proportional to the amount of significant DEGs assigned. Positive statistical tests are given that face each GO.

Fig 2

Lsr2 regulated genes involved in different pathways in M. abscessus. (A and B) Heat maps representing genes regulated by Lsr2 and implicated in various cellular mechanisms (intracellular survival and growth, virulence and pathogenicity, glycopeptidolipids biosynthesis, transport, antimicrobial resistance genes, and central carbon metabolism) in Mabs-S and Mabs-R. The color scale represents differentially expressed genes with a log₂ fold change (log₂FC) ranging from −8 to 8. Genes with log₂FC ≥1 correspond to downregulated genes by Lsr2 (repressed genes), while genes with log₂FC ≤−1 correspond to upregulated genes by Lsr2 (activated genes).

Fig 3

Conservation of Lsr2 regulatory effect on transport genes in both planktonic growth and intracellular growth conditions. Quantification of relative expression of mmpL8 and MAB_2037 measured by RT-qPCR under planktonic growth conditions (left side) and intracellular growth conditions (right side) in Mabs-S-WT (red), Mabs-S-Δlsr2 (orange), Mabs-S-Δlsr2-Clsr2 (purple), Mabs-R-WT (blue), Mabs-R-Δlsr2 (light blue), and Mabs-R-Δlsr2-Clsr2 (dark blue). sigA transcript levels were used for normalization. For each assay, n = 3, and error bars are SEM. ***P < 0.001 and ****P < 0.0001 (Student’s t test).

Fig 4

Enhanced sensitivity to aminoglycosides and macrolides due to diminished transcriptional expression of antimicrobial resistance genes in Mabs-S-Δlsr2 and Mabs-R-Δlsr2 strains. (A) Quantification of relative expression of antimicrobial resistance genes, eis2, MAB_1409c, MAB_2355c, and erm41, by RT-qPCR under planktonic growth conditions (upper side) and intracellular growth conditions (lower side) in Mabs-S-WT (red), Mabs-S-Δlsr2 (orange), Mabs-S-Δlsr2-Clsr2 (purple), Mabs-R-WT (blue), Mabs-R-Δlsr2 (light blue), and Mabs-R-Δlsr2-Clsr2 (dark blue). sigA transcript levels were used for normalization. For each assay, n = 3, and error bars are SEM. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 (Student’s t test). (B) Intracellular growth of M. abscessus strains: Mabs-S-WT (red), Mabs-S-Δlsr2 (orange), Mabs-R-WT (blue), and Mabs-R-Δlsr2 (light blue), following antibiotic treatments. Murine J774.2 macrophages were infected with mycobacteria at an MOI of 10 and subsequently treated with 1×MIC liposomal amikacin or 1×MIC clarithromycin. Intracellular growth post-antibiotic treatment was assessed by counting CFUs at various time points after treatment (days 0, 1, 3, and 6). The data are representative of three independent experiments and are presented as means ± SEM. Differences between means were analyzed by two-way ANOVA and the Tukey post-hoc test, allowing multiple comparisons. ns, non-significant, *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Fig 5

Modalities of Lsr2 binding on the M. abscessus genome. (A) Coverage of Lsr2 ChIP-seq fragments on the genomes of Mabs-S and Mabs-R morphotypes. (B) Distribution of Lsr2 ChIP-seq fragments is strongly correlated with enhanced AT content as exemplified for genomic regions encompassing genes from MAB_0827 to MAB_0842 and from MAB_4463 to MAB_4467. The bottom part corresponds to gene positions. (C). Box plots showing the distribution of the lengths of binding regions of Lsr2 in the Mabs-S and Mabs-R morphotypes (with mean indicated as a « + »). (D) Box plots summarizing the preference of Lsr2 to bind to AT-rich regions of the M. abscessus Mabs-S and Mabs-R genome. (E) Histogram representing the percentage of Lsr2 within genomic features of operons (promoters, CDS, or promoters with CDS).

Fig 6

Integrative analysis RNA-seq/ChIP-seq. (A) Venn diagram showing the number of direct target genes of Lsr2, the number of genes regulated by Lsr2 without binding, and the number of genes bound by Lsr2 without regulatory effect in Mabs-S and Mabs-R. The numbers in the smallest circles show the number of regulated genes in the same operons as the direct targets. (B) Histogram representing the percentage of direct target genes of Lsr2 in the categories of metabolism, transport, genetic information processing, and antimicrobial resistance genes. The number of genes bound by Lsr2 and the total number of genes annotated in the KEGG database are also indicated on each histogram. (C) Violin plots comparing log₂FC expression levels of all genes regulated by Lsr2 with those of direct target genes in Mabs-S and Mabs-R. (D) Box plots comparing the Log2FC expression levels of all genes regulated by Lsr2 with those of direct target genes, categorized by two domain size groups delineated by Lsr2, one with sizes less than 1 Kb and the other exceeding 1 Kb. Differences between means were analyzed by ordinary one-way ANOVA. ns, non-significant and ****P < 0.0001.