Advertisement

  • Loading metrics

Open Access

Peer-reviewed

Research Article

The SigD regulon of Mycobacterium abscessus determines cell envelope composition and antibiotic susceptibility

  • Sean R. Jones,

    Roles Formal analysis, Investigation, Writing – review & editing

    Affiliation School of Public Health, University at Albany, Albany, New York, United States of America

    ⨯
  • Kelly Maune ,

    Contributed equally to this work with: Kelly Maune, Kelley Hurst-Hess

    Roles Formal analysis, Investigation, Writing – review & editing

    Affiliation Division of Genetics, Wadsworth Center, New York State Department of Health, Albany, New York, United States of America

    ⨯
  • Kelley Hurst-Hess ,

    Contributed equally to this work with: Kelly Maune, Kelley Hurst-Hess

    Roles Investigation

    Affiliation Division of Genetics, Wadsworth Center, New York State Department of Health, Albany, New York, United States of America

    ⨯
  • Pallavi Ghosh

    Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Methodology, Supervision, Validation, Writing – original draft, Writing – review & editing

    Pallavi.Ghosh@health.ny.gov

    Affiliations School of Public Health, University at Albany, Albany, New York, United States of America, Division of Genetics, Wadsworth Center, New York State Department of Health, Albany, New York, United States of America

    ⨯

Abstract

A major determinant of the exceptional intrinsic resistance of M. abscessus is the lipid-rich cell envelope, yet the regulatory systems that remodel envelope-associated pathways remain poorly defined. Here, we determine the σD regulon in M. abscessus and establish its role in cell envelope homeostasis and intrinsic resistance to hydrophobic antibiotics. RNA-Seq analysis of a MabΔsigD mutant identified 447 differentially expressed genes, while ChIP-Seq mapped 72 σD binding sites and defined a conserved promoter motif (GTAACA/G-N16-CGAT). Using a combination of σD binding, motif orientation and expression data, we identified a core set of directly regulated genes, distinct from what was previously observed in M. tuberculosis, many of which encode proteins involved in envelope-associated functions. These include loci involved in trehalose polyphleate (TPP) biosynthesis, the antigen 85 complex and peptidoglycan remodeling enzymes. Deletion of sigD resulted in a significant reduction in TPPs in the cell envelope and an increase in ethidium bromide accumulation. Consistent with these changes, loss of σD selectively sensitized M. abscessus to hydrophobic antibiotics, including rifampicin and tigecycline. Deletion of mmpL10, which is required for transport of TPP precursors, recapitulated the drug sensitivity of MabΔsigD, implicating envelope composition as a key effector of the phenotype. Expression of the σD regulon further increased during starvation and in response to SDS, isoniazid, and ethambutol, mediated by degradation of RsdA, consistent with a role in stress-responsive envelope adaptation. Together, these findings demonstrate σD is active during logarithmic growth in rich media where it regulates the expression of envelope-associated genes that influence envelope permeability and basal level susceptibility to hydrophobic antibiotics; its activity further increases in response to cell envelope stress, presumably promoting envelope remodeling to counteract damage.

Author summary

Mycobacterium abscessus is an increasingly important bacterial pathogen that causes difficult-to-treat infections because it is naturally resistant to many antibiotics. A major contributor to this resistance is its complex, lipid-rich cell envelope, which acts as a barrier that limits the entry of antibiotics. How this protective barrier is regulated and remodeled in response to environmental stress is not well understood. Here, we investigated the role of the alternative sigma factor D (σD), a protein that controls the expression of specific groups of genes. We found that σD regulates numerous genes associated with the bacterial cell envelope, including genes involved in the production and transport of specialized surface lipids. Loss of σD altered the abundance of trehalose polyphleates and made bacteria more susceptible to several hydrophobic antibiotics. We further found that σD activity increases during nutrient starvation and when the cell envelope is disrupted. Our findings identify σD as part of a regulatory network that helps maintain the Mycobacterium abscessus cell envelope and influences its intrinsic resistance to antibiotics.

Citation: Jones SR, Maune K, Hurst-Hess K, Ghosh P (2026) The SigD regulon of Mycobacterium abscessus determines cell envelope composition and antibiotic susceptibility. PLoS Genet 22(9): e1012286. https://doi.org/10.1371/journal.pgen.1012286

Editor: Shumin Tan, Tufts University School of Medicine, UNITED STATES OF AMERICA

Received: April 6, 2026; Accepted: August 13, 2026; Published: September 2, 2026

Copyright: © 2026 Jones et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All data underlying the findings of this study are fully available without restriction. Unedited DESQ2 files for RNA-Seq experiments and processed CHIP-Seq data are included within the manuscript and its Supporting Information files. Sequencing data files have been deposited in the NCBI Gene Expression Omnibus (accession # GSE338978).

Funding: PG is supported by National Institutes of Health (NIH) awards AI155473 and AI187284. SJ is supported by an NIH T32 (T32GM132066) award to the RNA Institute, University at Albany.The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Mycobacterium abscessus is a rapidly growing non-tuberculous mycobacterium and a major cause of chronic pulmonary, skin, and soft tissue infections, particularly in individuals with underlying lung disease such as cystic fibrosis or bronchiectasis [1–3]. M. abscessus infections are among the most difficult bacterial diseases to treat, requiring prolonged multidrug regimens that are frequently ineffective and poorly tolerated [4,5]. This exceptional intrinsic resistance to antibiotics is driven by an impermeable lipid-rich cell envelope and a plethora of drug-and target-modifying enzymes that are induced in the presence of antibiotics [6,7].

The mycobacterial cell envelope is comprised of an outer mycomembrane (MM), peptidoglycan (PG) and arabinogalactan (AG) [8]. Long-chain (C60-90) α-alkyl, β-hydroxy fatty acids known as mycolic acids form the inner leaflet of the MM and are covalently linked to AG which in turn is attached to the PG layer. The outer leaflet of the MM is composed of mycolic acids linked to trehalose and other lipids and proteins such as glycopeptidolipids (GPLs), trehalose polyphleates (TPPs), and phosphatidyl-myo-inositol dimannosides (PIMs). [9]. Mycolic acid biosynthesis requires a coordinated action of a number of enzymes. Short-chain fatty acids synthesized by the fatty acid synthase I (FAS-I) system are iteratively elongated by the FAS-II system, consisting of the enzymes FabD, KasA, KasB, HadABC, and InhA [10]. Mycolic acids are then carboxylated by acyl-CoA carboxylase (AccD3), activated by acyl-AMP ligase (FadD2) and transferred to trehalose by Pks13 to form trehalose monomycolate (TMM) [11]. Once synthesized, TMM is transported across the plasma membrane by MmpL3 and incorporated into the cell envelope [12]. Additionally, mycolic acids are transferred from TMM either to AG to form mycolylarabilogalactan (mAG), or to another TMM to form trehalose dimycolate (TDM) by a set of secreted mycolyltransferases (the antigen 85 complex), that are essential for mycomembrane assembly and maintenance [13]. Disruption of mycolic acid synthesis or transport profoundly compromises envelope integrity and viability [14].

Although M. abscessus (Mab) shares a conserved mycobacterial cell envelope architecture with M. tuberculosis (Mtb), their mycomembranes show differences in composition and dynamics that may have functional consequences for antibiotic susceptibility. For example, M. abscessus appears to have a lower abundance of oxygenated mycolic acids, which are key determinants of envelope permeability [15,16]. Moreover, in contrast to the M. tuberculosis mycomembrane which is dominated by long-chain mycolic acids and complex lipids such as phthiocerol dimycocerosates and sulfolipids, the M. abscessus mycomembrane is enriched in surface-exposed glycopeptidolipids (GPLs), which influence colony morphology, biofilm formation and disease progression [17,18]. Additional cell wall associated lipids such as trehalose polyphleates (TPP) - high molecular weight glycolipids structurally related to sulpholipids found in M. tuberculosis, and a glycosyl diacylated nonadecyl diol (GDND) have been reported in M. abscessus [19,20].

Despite the low permeability of the mycobacterial outer membrane, antibiotics reach the cytosol where they elicit a global reprogramming of gene expression mediated by transcription factors and alternate sigma (σ) factors. WhiB7 is the best studied transcription factor that induces genes involved in intrinsic resistance to ribosome targeting antibiotics [7,21,22]. Sigma factors are interchangeable subunits of RNA polymerase that confer promoter specificity and enable global remodeling of gene expression in response to defined cues [23]. In addition to the principal sigma factor, σA, 12 alternate sigma factors belonging to groups II-IV have been identified in M. tuberculosis and 18 are known in M. abscessus [24,25]. In M. tuberculosis, σF induction is observed upon ethambutol (EMB), rifampin (RIF), streptomycin (STR) and cycloserine treatment [26], and σB and σF are induced upon vancomycin (VAN) exposure [27]. Furthermore, a MtbΔsigE mutant was found to display VAN, RIF, STR, EMB, gentamicin (GENT) and isoniazid (INH) sensitivity, whereas MtbΔsigB was INH and EMB sensitive [28].

The σD regulon has also been previously studied in M. tuberculosis [29,30]. MtbsigD is transcribed throughout the logarithmic and stationary phases, but its activity is controlled primarily at the post-translational level by a transmembrane anti-sigma factor RsdA that sequesters σD, keeping it inactive. Extracellular signaling and subsequent intramembrane cleavage of RsdA by a site 2 protease, Rip1, effectively liberates σD into the cytoplasm allowing it to associate with the RNAP core enzyme [31]. Although no growth defects were demonstrated in the MtbΔsigD mutant in rich media, its expression was observed to increase during starvation and under INH exposure [29,32]. Strikingly, ΔsigD mutant strains of M. tuberculosis are significantly attenuated in mouse infection models, exhibiting reduced survival and an altered inflammatory response in the lung compared to the wild-type strain [29,30].

In M. abscessus, with the exception of σH that confers resistance to tigecycline (TIG) and amikacin (AMK) [22], much less is known about the role of the other 17 alternate sigma factors in adaptation to antibiotic stress. Here, we define the σD regulon of M. abscessus and demonstrate that activity of this sigma factor during exponential growth controls genes involved in biosynthesis/transport of cell envelope components which together influence envelope permeability and basal level susceptibility to hydrophobic antibiotics including RIF, CLA and TIG. In particular, we show that genes for TPP biosynthesis are σD regulated and deletion of either sigD or mmpL10 results in increased drug sensitivity. Additionally, we also demonstrate that nutrient starvation and cell wall stress is sensed by membrane bound RsdA (regulator of σD), leading to increased expression of the regulon which presumably counteracts the damage by remodeling the cell envelope.

Results

Determination of the σD regulon in M. abscessus using RNA-Seq and ChIP-Seq

To determine the σD regulon of M. abscessus we generated an unmarked deletion of sigD (MAB_3724c) in a rough variant of the ATCC19977 type strain (CIP 104536R) using recombineering (S1 Fig). A complementing strain expressing sigD from its native promoter (MabΔsigD::psigDsigD) was created by integration into the L5 attB site. Bacteria were grown to mid-exponential phase (A600 = 0.6) and changes in gene expression were determined using RNA sequencing (RNA-Seq) using the criteria of >1.5-fold change in the MabΔsigD strain as compared to wild-type (WT) bacteria (padj < 0.01). This identified 447 genes comprising the σD dependent regulon in M. abscessus (Fig 1a, S1 Data). Of these, 315 genes were downregulated and 132 genes were upregulated in MabΔsigD. Changes in gene expression between the complemented strain were also compared to MabΔsigD and identified 291 σD responsive genes (>1.5- fold change, padj < 0.01) (Fig 1b, S1 Data).

thumbnail
Download:
Fig 1. Determination of the M. abscessus σD regulon using RNA-Seq and ChIP-Seq.

Volcano plots of differentially expressed genes between a) M. abscessus WT and MabΔsigD and b) the MabΔsigD::psigDsigD complemented strain and MabΔsigD, determined by RNA-Seq. The horizontal dashed line represents an adjusted p-value of 0.01 and vertical lines represent a fold change of 1.5. Genes with a positive fold change represent genes underexpressed in the ΔsigD mutant, while a negative fold change indicates genes overexpressed in the mutant. Genes with direct binding sites for σD determined by ChIP-Seq are indicated in magenta. Genes strongly dependent on σD are labeled and MAB_0937c and MAB_0938c are circled as the TPP locus. c) ChIP-Seq was conducted using the MabΔsigD::psigDsigD3’-FLAG strain and peaks were called with a custom script using an FDR = 0.005. The distribution of peaks along the genome are shown. d) Representative examples of σD binding sites at different genomic locations relative to annotated open reading frames are shown. A representative intragenic σD peak within MAB_0938c coinciding with a known TSS suggesting a possible alternate start site is also shown. e) Sequence logo of the enriched motif in σD bound sites identified using MEME Suite 5.5.9 (E-value = 4.6e-030) showing consensus -35 and -10 elements. f) RNA-Seq reads for MAB_0938c in WT and MabΔsigD visualized using IGV shows a large portion of reads starting within the MAB_0938c ORF corresponding to the binding site of σD.

https://doi.org/10.1371/journal.pgen.1012286.g001

To determine the direct binding sites of σD we constructed a strain containing a 3X-FLAG tag at the 3’-end of sigD along with 200nt of its upstream regulatory sequence, integrated at the L5 attachment site of MabΔsigD and confirmed that it was fully functional (S2 Fig). The MabΔsigD::psigDsigD3’-FLAG strain was grown to mid-exponential phase (A600 = 0.6); DNA-nucleoprotein complexes were immunoprecipitated with anti-FLAG monoclonal antibodies followed by library preparation and Illumina sequencing. Using a previously published Python script, Peakcaller [33,34], we identified a total of 72 σD binding sites (FDR = 0.005) constituted by 44 intergenic binding sites (between annotated ORFs), and 28 intragenic sites (within annotated ORFs) (Fig 1c and 1d, Tables 1 and 2). Sequences corresponding to the MabσD ChIP-Seq peaks were analyzed using MEME tools and a conserved motif containing clear consensus -35 and -10 elements, separated by 16 nucleotides, was identified in all 72 sites (Fig 1e). We then used a combination of peak location, motif orientation and gene expression changes between WT and MabΔsigD at all the 72 sites to determine the identity of genes regulated. Of the 44 intergenic peaks, 40 were found to be associated with the M. abscessus genes shown in Table 1, including sigD itself; 4 are likely involved in transcription of antisense RNAs. Of the 28 intragenic binding sites, 5 likely correspond to alternate start sites, and 18 are potentially involved in transcription of antisense RNAs (Table 2). Notably we observed that only MAB_0733, MAB_1097, MAB_1439c, MAB_1616, MAB_2488c, MAB_2960, MAB_3165c, MAB_4114 and MAB_4296c, were downregulated >2-fold in the MabΔsigD strain and were therefore strongly σD dependent. Most genes with σD binding sites in their promoter regions were either modestly downregulated or showed no significant difference in MabΔsigD (S1 Data), suggesting that these genes are not exclusively transcribed by σD and that their expression is likely compensated by additional sigma factors in the MabΔsigD strain. Interestingly, we observed a large intragenic σD binding site within MAB_0938c that coincided with the RNA-Seq profile as well as a previously described internal transcription start site (TSS); the expression of both MAB_0938c (papA3) and the operonic MAB_0937c (mmpL10) were significantly downregulated in MabΔsigD indicating a functional role of σD binding at this site (Fig 1d and 1f, Table 2) [35]. We also investigated if genes within the same transcription unit as the direct targets of σD were differentially expressed in wild-type M. abscessus when compared to MabΔsigD (RNA-Seq) and identified 9 additional genes within the σD regulon that are regulated directly by σD (Table 3). Therefore, binding of σD to 72 chromosomal locations suggests the regulation of 50 genes in M. abscessus (40 intergenic, 1 intragenic with a clear alternate start site, and 9 operonic genes).

thumbnail
Download:
Table 1. Intergenic σD binding sites determined by ChIP-Seq in M. abscessus ATCC/CIPR. Genetic location and coordinates of peaks, gene function and fold downregulation in ΔsigD mutant compared to WT and complemented (ΔsigD + psigDsigD) strains (from RNA-Seq) are shown. Identity of genes regulated by σD binding was determined using a combination of peak location, the orientation of the conserved motif identified by MEME analysis as well as expression in ΔsigD and is also included. σD binding sites associated with a fold change >2 are highlighted in gray. Function of genes annotated as hypothetic proteins were verified using Uniprot(*).

https://doi.org/10.1371/journal.pgen.1012286.t001

thumbnail
Download:
Table 2. σD binding sites within annotated ORFs (Intragenic) determined by ChIP-Seq in M. abscessus ATCC/CIPR. Genetic location and coordinates of peaks, gene function and fold downregulation in ΔsigD mutant compared to WT and complemented (ΔsigD + psigDsigD) strains (from RNA-Seq) are shown. Possible genes functions were assigned by using the orientation of the conserved motif identified by MEME analysis (**).

https://doi.org/10.1371/journal.pgen.1012286.t002

thumbnail
Download:
Table 3. Expression of operonic genes regulated by direct σD binding.

https://doi.org/10.1371/journal.pgen.1012286.t003

Lastly, we mapped previously identified transcription start sites (TSSs) located near σD binding sites (Fig 2) [35]. Approximately 50% of σD regulated promoters were associated with TSSs that were identified in cells grown in rich media (Fig 2). Several of these TSSs were linked to strongly σD dependent genes, including MAB_1439c, MAB_4114, and MAB_0733 suggesting that σD exhibits basal activity under these growth conditions. However, a number of σD bound promoter regions also contained a consensus −10 motif recognized by σA/σB (TANNNT) at an appropriate distance from the annotated TSSs [34] (Fig 2). Many of the corresponding genes showed little or no dependence on σD for expression, suggesting that transcription from these promoters may be mediated by σA or σB associated RNA polymerase during logarithmic growth. These findings provide a potential explanation for the limited effect of sigD deletion on the expression of genes with σD promoters.

thumbnail
Download:
Fig 2. Sequence of σD promoters upstream of regulated genes.

Sequence logo of the enriched motif in σD bound sites is shown. Location of -35 and -10 elements in σD promoters upstream of the regulated genes identified in Tables 1-2 are underlined and in bold. Approximately 50% of σD bound sites are associated with previously identified transcription start sites shown in green. Location of the σA/σB −10 motif, ANNNT, with the conserved T appropriately positioned relative to the annotated transcription start site are in magenta. The fold-change in expression of each gene in WT compared to MabΔsigD is shown in parenthesis.

https://doi.org/10.1371/journal.pgen.1012286.g002

The MabσD regulon contains numerous cell envelope genes and MabΔsigD is deficient in TPPs

Functional classification of genes directly regulated by σD revealed that many were associated with cell envelope-related processes. Approximately 27% of the σD direct targets were involved in cell envelope biosynthesis/transport and peptidoglycan remodeling; another 27% were either secreted or membrane proteins (Fig 3a). Thus, more than half of the direct target set comprises genes with known or predicted connections to the cell surface or envelope. This suggested a role of the σD regulon in cell envelope homeostasis and prompted us to examine envelope-associated phenotypes. Strikingly, σD binding sites were identified upstream of the antigen 85 complex genes involved in TDM and mAGP synthesis, as well as within MAB_0938c, a gene involved in TPP glycolipid biosynthesis (Fig 3b,3d-3e). While no change in expression of the ag85 genes was detected in MabΔsigD compared to WT bacteria, expression levels of both MAB_0937c and MAB_0938c were significantly downregulated in MabΔsigD (Tables 1 and 2). We therefore investigated the differences in cell envelope lipids in wild-type and MabΔsigD strains. Consistent with mRNA levels of ag85 genes, no differences were observed in the total amounts of TDM or methyl esters of mycolic acids (Fig 3c). In contrast, TPP levels were significantly lower in MabΔsigD, and restored in in the complemented strain (Fig 3f-3g). The identity of TPP was based on previous literature as well as by its absence in MabΔmmpL10 (Fig 3f-3g) [19].

thumbnail
Download:
Fig 3. MabσD regulon is enriched in cell envelope genes and a MabΔsigD mutant is deficient in TPPs.

a) Pie chart of the functional categories of genes directly regulated by σD. b) Schematic of the antigen 85 locus encoding proteins responsible for the production of the glycolipid TDM and mAG. Arrows indicate binding sites of σD. Numerical values within each gene indicate the fold-change in expression of the gene in WT/ MabΔsigD. c) Thin Layer Chromatography of apolar lipids and mycolic acid methyl esters (MAMES) purified from WT and ΔsigD strains. Apolar lipids were resolved in CHCl3:MeOH:H2O (90:10:1 v/v) x2 and visualized by spraying with 10% H2SO4 in EtOH and charring at 120°C. MAMEs were resolved in petroleum ether:diethyl ether (95:5 v/v) x6 and visualized by spraying 5% phosphomolybdic acid and charring at 110°C. Data is representative of >3 biological replicates. FAME = Fatty acid methyl ester, αα’ = alpha and alpha prime mycolic acids. d) Schematic of the genes within the TPP locus responsible for TPP synthesis. Arrow indicates location of the σD binding site. The numerical values within each gene indicates the fold-change in expression of the genes in WT/ MabΔsigD; genes with a change in expression in the ΔsigD strain are underlined with a red bar. e) Schematic of the TPP biosynthesis pathway (Created in BioRender. Ghosh, P. 2026, https://BioRender.com/w7kkigq) showing the synthesis of diacyl trehalose (DAT) from trehalose in the cytosol, its transport to the cell envelope by MmpL10, and its subsequent conversion to TPP by the enzyme PE. f) Thin-layer chromatography analysis of total free lipids extracted from WT, ΔsigD, MabΔsigD::psigDsigD, and MabΔmmpL10 strains showing the migration of trehalose polypheates (TPP) and trehalose dimycolate (TDM) and is representative of 5 biological replicates. Total lipids were resolved using CHCl3/CH3OH (90:10 v/v) and visualized by spraying the plates with a 0.2% anthrone solution (w/v) in concentrated H2SO4 and charring. Images are representative of 5 independent experiments. TDM = Trehalose monomycolate, DAT = Diacyltrehalose and TPP = Trehalose polyphleate are annotated as per literature [19]. g) Quantification of TPP abundance from TLC plates by densitometric analysis using ImageJ. Data represent the mean relative spot density from five independent experiments (n = 5), with error bars indicating standard deviation (SD). Statistical significance was determined by one-way ANOVA followed by Dunnet’s multiple comparison test; P values are indicated.

https://doi.org/10.1371/journal.pgen.1012286.g003

The MabσD regulon is induced by nutrient starvation and agents that perturb the cell wall

While our studies thus far showed that σD is functional during growth in rich media, we wondered if perturbation of the cell envelope could further increase the expression the σD regulon. We therefore monitored the expression of sigD as well as MAB_1439c (strongly σD dependent and a proxy for σD activity) under conditions that perturb the cell envelope: SDS that induces cell wall stress, EMB that interferes with arabinogalactan synthesis and INH that interferes with mycolic acid biosynthesis [36,37]. Fig 4a shows that exposure to SDS modestly induced the regulon; EMB and INH caused ~2-fold increase in sigD transcript levels and a 5–10 fold increase in MAB_1439c levels; a similar induction was not observed in RIF treated bacteria. Moreover, the increase in transcript levels of MAB_1439c upon EMB and INH exposure was abrogated in a MabΔsigD mutant confirming that expression of MAB_1439c was σD dependent (Fig 4b). Consistent with this observation, induction of the σD regulon was identified in RNA-Seq of wild-type bacteria exposed to EMB and INH but not in RIF treated cells (Fig 4c, S1 Data) and [38]. Lastly, we compared the gene expression pattern in WT and MabΔsigD upon INH exposure which demonstrated that the induced expression of the σD regulon was abrogated in MabΔsigD (Fig 4d, S1 Data). Notably, the genes exhibiting σD - dependent expression under INH exposure largely overlapped with those identified under uninduced conditions, although the WT/MabΔsigD expression ratios were greater following INH treatment, consistent with induction of the σD regulon. Thus, INH exposure primarily increased the magnitude of σD-dependent expression rather than recruiting a substantial number of additional genes into the regulon (S1 Table and S1 Data).

thumbnail
Download:
Fig 4. Expression of the σD regulon is induced by nutrient starvation and cell envelope stress.

a) Change in relative mRNA expression of sigD and MAB_1439c in WT M. abscessus following exposure to 0.05% SDS, RIF 32μg/mL, EMB 70μg/mL, or INH 64μg/mL for 2h, nutrient starvation in PBS for 6h, or in stationary phase (A600 = 6.0). Transcript levels were measured using qRT-PCR, normalized to the endogenous control sigA, and relative expression was calculated using the 2−ΔΔCt method and expressed as fold change relative to the uninduced control. Data are from 3 independent experiments and are presented as mean ± SD. Statistical analysis shown was performed for MAB_1439c with each treatment condition compared with the uninduced control using two-way ANOVA followed by Dunnet’s multiple comparison test. P values for these comparisons are indicated above the brackets. b) Change in relative mRNA expression of the σD dependent gene MAB_1439c in WT and ΔsigD when induced with EMB 70 μg/mL and INH 64 μg/mL for 2h. Transcript levels were measured using RT-qPCR, normalized to the endogenous control sigA, and relative expression was calculated using the 2−ΔΔCt method and expressed as fold change relative to the wild-type uninduced control. Data represents mean ± SD, n = 3. Statistical analysis was performed within each condition through two-way ANOVA followed by Tukey’s multiple comparison test. P values are indicated. c) Heat map showing the change in expression of the top 10 most σD responsive genes as well as their expression upon treatment with EMB, INH, and RIF compared to uninduced WT using RNA-Seq. d) Volcano plot of differentially expressed genes between M. abscessus WT and the ΔsigD strain after 3h INH treatment, determined by RNA-Seq. The horizontal dashed line represents an adjusted p value of 0.01 and vertical lines represent a fold change of 1.5-fold. Genes with a positive fold change represent genes underexpressed in the ΔsigD mutant, while a negative fold change indicates genes overexpressed in the mutant. Genes with direct binding sites for σD determined by ChIP-Seq are indicated in magenta. Genes strongly dependent on σD are labeled.

https://doi.org/10.1371/journal.pgen.1012286.g004

In M. tuberculosis, nutrient-starvation by transferring exponentially growing cells to phosphate buffered saline (PBS) leads to arrest in growth and respiration, yet bacteria remain viable and become phenotypically drug tolerant [32]. This state is associated with changes in lipid composition of the cell envelope, specifically accumulation of phthiocerol dimycocerosate (PDIM), and a remodeling of peptidoglycan involving L-, D- transpeptidase mediated enrichment in 3 → 3 crosslinks [32,39,40]. Curiously, three L-, D- transpeptidases were found to be direct targets of MabσD: MAB_1530, MAB_3165c and MAB_4537c. Since MtbsigD is also known to be induced during nutrient starvation [32,41], we monitored the induction of MAB_1439c, a proxy for the σD regulon, under PBS starvation and in late stationary phase. As seen in Fig 4a, stationary phase growth and nutrient starvation in PBS resulted in an ~ 6-fold increase in transcription of MabsigD, and an ~ 40–60 fold induction of MAB_1439c. Given the prominent induction of the σD regulon under starvation, we monitored the growth and survival of wild-type and MabΔsigD mutant during late stationary phase and in PBS; however, no significant difference in survival was observed between the strains under these conditions (S3 Fig).

Induction of the σD regulon is controlled by RsdA degradation

Alternate sigma factors are believed to be constitutively transcribed; their activity is however regulated post-translationally by anti-sigma factors that sequester them in an inactive state until activation is required. Anti-sigma factors regulating extracytoplasmic (ECF) sigma factors are often membrane proteins, encoded by genes adjacent to their cognate sigma factors genes on the chromosome. A putative anti-σD/RsdA (MAB_3723c) lies immediately adjacent to MabsigD in the M. abscessus genome (Fig 5a). Transmembrane topology prediction (DeepTMHMM 2.0) suggests a single-pass membrane protein with an N-terminal cytoplasmic sigma factor-binding domain and a C-terminal periplasmic domain, consistent with a role in sensing envelope status and relaying this signal through regulated intramembrane proteolysis, analogous to activation pathways described for M. tuberculosis RsdA, RskA, RslA, and RsmA [14,31] (Fig 5b). To determine if MAB_3723c functions as a regulator of σD, we created a deletion of MAB_3723c in a ΔsigD mutant background (ΔsigD/ΔrsdA) and monitored the induction of MAB_1439c in the presence of EMB and INH. Fig 5c shows a lack of MAB_1439c expression in a ΔsigD/ΔrsdA strain as expected; in a ΔsigD/ΔrsdA::psigDsigD strain background, induced expression levels of MAB_1439c was observed even in the absence of EMB and INH, confirming that RsdA controls the induced expression of the σD regulon.

thumbnail
Download:
Fig 5. Induction of the σD regulon is controlled by RsdA degradation.

a) Genomic organization of sigD and its adjacent anti-sigma factor rsdA. hp = hypothetical protein. b) Model of σD activation by intramembrane proteolysis of RsdA (Created in BioRender. Ghosh, P. 2026 https://BioRender.com/8vxlby1). RsdA is a membrane protein that anchors σD to the cell membrane. Activation of site-1 protease by cell wall stresses and nutrient starvation presumably cleaves RsdA at site1; subsequent cleavage by site-2 protease liberates functional σD. RsdA was tagged at the C-terminal periplasmic domain with a 3X FLAG tag that enables monitoring its degradation under various conditions. c) Change in relative mRNA expression of MAB_1439c in WT, ΔsigD/ΔrsdA and ΔsigD/ΔrsdA::psigDsigD3’-FLAG strain when induced with EMB 70μg/mL or INH 64μg/mL for 4h measured using RT-qPCR and plotted as fold change over wild-type uninduced control. Data represents mean ± SD, n = 3. Statistical analysis was performed through one-way ANOVA followed by Tukey’s multiple comparisons test. d) WT untagged and Mab:: MAB_3723c3’FLAG strains were grown either in PBS for 6h or exposed to EMB 70μg/mL or INH 64μg/mL for 6h followed by detection of RsdAFLAG using anti-FLAG antibodies. Location of RsdA is indicated with an arrow; bands above and below RsdA are non-specific and present in the untagged (WT) strain as well. e) Quantification of RsdA from western blots by densitometric analysis using ImageJ. Data represents mean ± SD, n = 3. Statistical significance was determined by one-way ANOVA followed by Tukey’s multiple comparisons test; P values are indicated.

https://doi.org/10.1371/journal.pgen.1012286.g005

To determine if the observed induction of the σD regulon is associated with degradation of MabRsdA, we introduced a 3X FLAG tag at the 3’-end of MAB_3723c at its native chromosomal location using recombineering. This generated a C-terminal, periplasmically oriented FLAG tagged MabRsdA protein. Although an N-terminal cytoplasmic tag would have been preferable to track the cleavage product of RsdA, this was not technically feasible because the open reading frames of MAB_3723c and MAB_3724c overlap. The resulting Mab::MAB_3723c3’ FLAG strain was exposed to EMB, INH or PBS starvation for 6h, and the cleavage of RsdA was assessed by monitoring the abundance of RsdA C-FLAG by western blotting using α-FLAG antibodies. As seen in Fig 5d-5e, RsdA levels decreased significantly following exposure to INH, EMB and nutrient starvation indicating that induction of the σD regulon is a consequence of RsdA degradation.

The MabΔsigD mutant displays multi-drug sensitivity

Nutrient-starved bacteria accumulate less antibiotics than replicating cells which cannot be reversed by efflux pump inhibitors indicating a reduced cell envelope permeability as a driver of starvation-associated tolerance [42]. Presumably, these changes are associated with cell envelope remodeling following transcriptional reprogramming by core stress regulators and sigma factors. Since the σD regulon constitutes several genes encoding cell wall associated proteins which are upregulated by starvation and envelope stress, we evaluated the sensitivity of the MabΔsigD strain to an array of antibiotics. As seen in Fig 6a, MabΔsigD was hypersensitive to RIF, TIG, clarithromycin (CLA) and erythromycin (ERT), but not to EMB, INH, AMK or spectinomycin (SPC) even though EMB and INH induce the σD regulon. We noticed that the antibiotic sensitivity pattern appeared to be a function of drug hydrophobicity (RIF > CLA/TIG/ERT > SPC > EMB/INH/AMK) rather than that of drug size (RIF > ERT/CLA > AMK/TIG > SPC > EMB > INH) and wondered if the observed hypersensitivity of MabΔsigD was suggestive of a compromised cell envelope that increased passive drug permeability. To test this, we monitored real-time ethidium bromide (EtBr) accumulation in wild-type, MabΔsigD, and complemented strains over one hour. Indeed, MabΔsigD demonstrated a marked increase in EtBr accumulation compared to both the wild-type and complemented strains (Fig 6b).

thumbnail
Download:
Fig 6. MabΔsigD is sensitive to hydrophobic antibiotics and displays increased accumulation of ethidium bromide.

a) Growth of ten-fold serial dilutions of WT, MabΔsigD and MabΔsigD::psigDsigDC-FLAG strains on Middlebrook 7H10 OADC containing indicated concentrations of RIF, TIG, ERT, CLA, AMK, EMB, INH, and SPC. Data is representative of >3 independent experiments. b) WT, MabΔsigD and MabΔsigD::psigDsigDC-FLAG strains were evaluated using the ethidium bromide (EtBr) uptake assay. Cells were incubated with PBST, 0.8% glucose, and 4 μg/mL EtBr and fluorescence was measured over the course of 1h. An increase in fluorescence is indicative of increased EtBr accumulation. Data represents mean ± SD, n = 3.

https://doi.org/10.1371/journal.pgen.1012286.g006

Given the observed depletion of TPP in MabΔsigD we hypothesized that the antibiotic hypersensitivity of MabΔsigD stems, at least in part, from reduced TPP levels in the cell envelope. We therefore constructed an isogenic deletion of mmpL10; the ΔmmpL10 mutant was clearly hypersensitive to RIF, TIG and CLA suggesting that TPPs in the mycobacterial cell envelope may play a protective role against antibiotics (Fig 7).

thumbnail
Download:
Fig 7. MabΔmmpL10 recapitulates the antibiotic sensitive phenotype of MabΔsigD.

Growth of ten-fold serial dilutions of M. abscessus WT, MabΔsigD, MabΔsigD::psigDsigDC-FLAG and MabΔmmpL10 on Middlebrook 7H10 OADC containing indicated concentrations of RIF, TIG, ERT, CLA, AMK, EMB, INH, and SPC. Data is representative of >3 independent experiments.

https://doi.org/10.1371/journal.pgen.1012286.g007

Discussion

σD is one of sixteen group IV extra-cytoplasmic (ECF) sigma factors in M. abscessus and is therefore appropriately poised to transduce signals from the cell surface or the extracellular environment. Here we define the σD regulon of M. abscessus and show that 447 genes are differentially expressed in a MabΔsigD strain relative to wild-type (>1.5-fold change, padj < 0.01) during exponential growth. Only a subset of this regulon (~50 genes) appears to be directly controlled by σD and we identify a strong consensus promoter motif (GTAACA/G-N16-CGAT). Among the direct targets are four transcription factors from the TetR and WhiB families which likely amplify the σD response through secondary regulatory cascades. Interestingly, several genes containing σD promoters in their upstream regions showed little or no reduction in expression in the MabΔsigD mutant under either uninduced or INH-induced conditions. This suggests that these promoters can also be recognized by additional sigma factors, allowing transcription to be maintained when σD is absent. This apparent redundancy is consistent with the complex architecture of transcriptional regulation in mycobacteria, which encode an unusually large repertoire of sigma factors with partially overlapping promoter specificities and stress-responsive functions. In M. abscessus, the presence of approximately 18 alternative sigma factors provides multiple opportunities for compensatory promoter recognition, such that loss of a single sigma factor may have only a limited effect on transcription of genes. Indeed, overlap in promoter recognition among mycobacterial sigma factors has been demonstrated previously, supporting a model in which functional crosstalk between sigma factor regulons buffers gene expression against the loss of individual regulatory pathways [24,25,28,34].

Notably, more than half of the genes directly regulated by σD are cell envelope associated, including those that encode proteins involved in mycomembrane biosynthesis/transport, membrane associated proteins, secreted proteins of hitherto unknown function, and several peptidoglycan hydrolases and L,D- transpeptidases which mediate 3 → 3 crosslinks known to strengthen the cell wall during stress and starvation. In line with this expectation, nutrient starvation and agents that cause envelope stress (SDS) or disrupt envelope biosynthesis (INH, EMB) increased expression of the σD regulon via degradation of RsdA. Together these findings support a role for the σD regulon in cell envelope maintenance during exponential growth, and an induced expression during adaptation to perturbations at the cell surface.

Within the σD regulon, two loci were particularly striking - the TPP locus implicated in trehalose polyphleate (TPP) biosynthesis/transport, and the Ag85 locus involved in TDM and mAGP formation; of these TPPs have not been detected in the envelope of M. tuberculosis. Remarkably, the MabΔsigD strain showed a significant decrease in TPP levels without any detectable change in TDM content (Fig 3f). Furthermore, the amounts of diacyltrehalose precursors (DAT) in the mutant strain were indistinguishable from that of wild-type bacteria. This observation is consistent with the presence of a σD regulated promoter intragenic to MAB_0938c (papA3) and an ~ 4-fold decrease in expression of mmpL10 which is known to be required for translocation of DATs to the cell surface where they are transacylated to phleic acids to form TPPs [43,44]. The reduced expression of mmpL10 in MabΔsigD likely contributes to the observed differences in TPP.

Curiously, despite strong induction of the σD regulon during nutrient starvation and stationary phase, the MabΔsigD mutant exhibited no significant defect in survival under these conditions, suggesting that genes required for persistence in nutrient-limited environments are controlled by redundant regulatory pathways. Similarly, although EMB and INH induced the σD regulon, the susceptibility of MabΔsigD to these antibiotics was indistinguishable from that of wild-type bacteria. Instead, the most pronounced and reproducible phenotypes were observed with RIF, TIG, CLA and ERT, all of which are hydrophobic in nature. In contrast, antibiotics without detectable effects in our assay were generally more hydrophilic (AMK, EMB, INH, SPC). Furthermore, deletion of sigD also increased intracellular accumulation of ethidium bromide. Although net EtBr accumulation is a cumulative measure of both passive influx and active efflux, the deletion of the σD regulon, which heavily regulates envelope-associated genes, suggests that this increased accumulation is driven by changes envelope permeability, though a concurrent contribution of impaired efflux cannot be entirely ruled out.

Collectively, these findings suggest that loss of σD likely increases permeability to hydrophobic compounds, rather than altering susceptibility through a drug-specific target or pathway. This interpretation is supported by previous studies in M. tuberculosis, M. smegmatis and M. marinum which show that reduced mycolates are linked to increased cell envelope fluidity and greater permeability to hydrophobic antibiotics such as RIF [45,46] and that nutrient-starved bacteria accumulate less rifamycins, fluoroquinolones and linezolid than replicating cells and have a reduced cell envelope permeability [42]. However, because the mycolic acid content of the MabΔsigD mycomembrane was indistinguishable from that of wild-type bacteria, altered mycolate abundance is unlikely to account for the increased permeability observed here. Instead, the MabΔsigD envelope contained substantially lower levels of TPP and a deletion of mmpL10, which eliminates envelope-associated TPP, also increases susceptibility to the same hydrophobic antibiotics. Together these findings support a model in which TPP contributes to mycomembrane permeability and limits antibiotic penetration. Although an additional role for MmpL10 in drug efflux cannot be excluded, previous studies argue against this [19].

It is noteworthy that several genes regulated by MabσD (MAB_0175, MAB_0177, MAB_0405c, MAB_1439c, MAB_1616, MAB_2871c and MAB_3355) have been previously associated with macrophage TLR2 activation [47], and an Mtb ΔsigD mutant has been reported to reduce macrophage TNF-α production [30]. Because TLR2 signaling promotes TNF-α production in macrophages, it is plausible that the M. abscessus ΔsigD mutant may similarly attenuate TNF-α responses. This possibility is particularly intriguing given the proposed role of TPPs in survival of M. abscessus in macrophages by inhibiting phagosome acidification [19]. Future work would involve testing if the σD regulon influences macrophage signaling and virulence phenotypes of MabΔsigD in mouse models.

Finally, although the primary amino-acid sequence of MabσD is ~ 74% identical to MtbσD, a comparison of their regulons reveals a limited overlap (15 genes) which includes papA3, rpfC and ag85-C (S2 Data). While all three genes are strongly dependent on σD in M. tuberculosis, only papA3 shows a strong σD dependence in M. abscessus; rpfC and ag85C exhibit σD occupancy without detectable expression changes, again suggesting compensatory or redundant regulation. Additionally, the strong dependence of MtbσD on iniA and iniB transcription is completely absent in M. abscessus; conversely a homologue of MAB_1439c, encoding a putative secreted protein in M. abscessus and reliant on MabσD, is absent in M. tuberculosis. Furthermore, the consensus sequence for MabσD promoters identified here also differs from motifs described previously for MtbσD by two groups: Raman et al reported σD dependent promoters in H37Rv strain to contain a clear -35 consensus of GTAACGct and no -10 consensus motif, whereas Calamita et al reported a consensus of AGAAAG-N16-20-CGTTAA in the CDC1551 strain [29,30], but is more likely a consequence of different methods employed in these studies (ChIP-Seq in M. abscessus v RNA-Seq in M. tuberculosis). Surprisingly, a comparison of the σD regulon of M. abscessus with the closely related actinobacterium, Corynebacterium glutamicum revealed more similarities than with M. tuberculosis. Overexpression of Cgu sigD induces expression of multiple genes involved in corynomycolic acid synthesis, corynomycolyl transferases and L,D-transpeptidases, and results in accumulation of trehalose dicorynomycolate (TDCM) in the envelope [48,49]. Additionally, the C. glutamicum σD consensus promoter sequence GTAACA/G-N17-18-GAT is nearly identical to that identified here for M. abscessus [50]. Despite the similarities in their regulons and promoter sequences, a change in mycolic acid content was not observed upon overexpression of Mab_sigD in MabΔsigD (Fig 4c). This is not surprising given the essential nature of mycolic acids and cell envelope glycolipids in Mycobacterium spp. as compared to corynebacteria which necessitates the existence of redundant regulatory pathways for envelope lipid synthesis. Overall, the differences among mycobacterial σD orthologs in the gene sets they regulate illustrate how related transcriptional regulators can be evolutionarily repurposed across actinobacteria. While σD contributes to growth and virulence in Mycobacterium tuberculosis, our findings establish a role for MabσD in cell-envelope maintenance; its potential contribution to host interactions and virulence remains to be determined experimentally.

Materials and methods

Media and bacterial strains

M. abscessus CIP104536R, an isogenic, glycopeptidolipid (GPL)-negative rough mutant derived from the wild-type ATCC19977 strain [18], and its derivatives were grown in Middlebrook 7H9 (DIFCO) supplemented with 0.05% Tween 80 and 10% OADC. Antibiotics were added at appropriate concentrations when indicated. Isogenic deletion of MAB_3724c (sigD) and MAB_0937c (mmpL10) were generated in the CIP104536R strain using recombineering, followed by the removal of the apramycin cassette by Cre-mediated recombination at loxP sites as previously described [22]. A deletion of MAB_3723c was similarly created in a ΔsigD background. MAB_3723c was endogenously tagged with a C-terminal FLAG tag using recombineering to create Mab::MAB_3723c3’FLAG. Unmarked deletions were confirmed through PCR and sequencing of product (S1 Fig). Complementing strains were created by cloning a gene of interest into pMH94 under the control of either a constitutive promoter (hp60) or its native promoter followed by integration of the plasmid into the L5 attB site of the appropriate deletion background. MabsigD was cloned with a 3’-3X FLAG tag, expressed from its native promoter and integrated into the L5 location of a MabΔsigD strain- MabΔsigD::psigDsigD3’-FLAG.

Growth of bacteria under nutrient starvation

Bacterial strains under study were grown in Middlebrook 7H9 + OADC to A600 = 0.6. The cells were harvested, washed with four times the original volume of the culture of phosphate buffered saline (PBS) with 0.05% (vol/vol) Tyloxapol, as previously described [51]. The bacteria were resuspended in PBS with 0.05% (vol/vol) Tyloxapol and incubated at 37°C for the indicated duration of time (6h, 24h, 48h). Cells were removed and assayed as required either for viability, drug sensitivity or for western blotting.

Antibiotic Sensitivity Assays

Wild-type, mutant, and complementing strains of M. abscessus were grown to A600 = 0.8. Cells were tested for drug susceptibility by spotting a 10-fold dilution series onto Middlebrook 7H10 agar supplemented with 10% OADC and the indicated concentration of antibiotic.

Western Blotting

To follow the degradation of RsdA, the Mab:: MAB_3723c3’FLAG strain was grown either in PBS for 6h or exposed to INH/EMB for 6h. The samples were normalized by A600 and weight, gently sonicated followed by separation of proteins on a 5–15% gradient SDS polyacrylamide gel (Biorad). Gels were either stained with Coommassie Blue or transferred to a PVDF membrane. RsdAFLAG was detected using anti-FLAG antibodies (Invitrogen). 3 biological replicates were performed.

RNA Preparation and RT-qPCR

Wild-type M. abscessus, MabΔsigD and MabΔsigD::psigDsigD3’-FLAG were grown to exponential phase (A600 = 0.6) followed by treatment with 70μg/mL ethambutol (EMB), 64μg/mL isoniazid (INH), 32 μg/mL of rifampicin (RIF), or 0.05% SDS for indicated times. To study the effect of starvation, bacteria were grown to late stationary phase (A600- = 6.0) or washed with phosphate buffered saline (PBS) and incubated for an additional 6h. Total RNA was prepared using the Qiagen RNA preparation kit followed by TURBO DNase treatment. cDNA was generated using random hexamers and Maxima reverse transcriptase (ThermoFisher), and RT-qPCR was performed with the SYBR Green RT-qPCR Master Mix (ThermoFisher) using the following primer pairs: sigD 5’-CCGTTCCTGGCATTTGTGTA and 5’-ACGCGCAAGATCAGGATTTC; MAB_1439c- 5’-ATGGCGACGACGATGAC and 5’-AGTGGACCA

CGCCATTC. Applied Biosystems Quant Studio 5 Real-Time PCR System was used with cycling conditions: 50°C for 2 min, 95°C for 10 min, and 40 cycles of 95°C for 15 s, 60°C for 1 min. Relative transcript levels were determined using the comparative Ct (2−ΔΔCt) method. Ct values for each target gene were normalized to the endogenous control gene sigA (ΔCt), and relative fold changes were calculated by comparison with the indicated uninduced control (ΔΔCt). Data represents mean ± SD, n = 3.

RNA-Seq analysis

Wild-type M. abscessus, MabΔsigD and MabΔsigD::psigDsigD3’-FLAG strains were grown to exponential phase (A600 = 0.6) and exposed to either 70μg/mL EMB, 32μg/mL RIF, or 64μg/mL INH when indicated. RNA extraction, DNase treatment, cDNA synthesis and library prep was performed by SeqCenter (Pittsburgh) and sequenced on an Illumina Novaseq platform. The sequence data was analyzed using the reference-based analysis and default parameters on Rockhopper v2.03. The raw counts generated were then analyzed using DESeq2 that reports differential gene expression and padj values. [52–54]. RNA-Seq experiments were performed using 2 biological replicates. Raw data for all RNA-Seq have been deposited in GEO (accession # GSE338978).

Chromatin immunoprecipitation sequencing (ChIP-Seq) and data analysis

ChIP-Seq was performed as previously described with minor modifications [7,55]. The MabΔsigD::psigDsigD3’-FLAG strain was grown at 37°C in Middlebrook 7H9 supplemented with 10% OADC and 0.05% Tween 80 to an A600 of 0.6 followed by crosslinking with 1% formaldehyde for 30 min and quenching with 250mM glycine. Cells were lysed using the CryoMill (Retsch) followed by sonication for 30 mins using the Bioruptor sonicator (Diagenode). The DNA protein complex was immunoprecipitated with anti-FLAG monoclonal antibody M2 (Sigma) for 18h at 4°C and processed as previously described [7]. Each ChIP-Seq experiment was performed using two biological replicates. Genomic DNA libraries enriched for σD binding were prepared using the NEB NextUltra II Library Prep kit for Illumina followed by sequencing on the Illumina platform. Reads were aligned to the reference genome using Rockhopper v2.03. Regions of enrichment were identified using a custom Python script as described previously [7]. Relative enrichment is reported as fold above threshold (FAT) score. The enriched regions were analyzed using MEME Suite 5.5.9 using default parameters [56]. ChIP-Seq experiments were performed using 2 biological replicates. Raw data for all ChIP-Seq have been deposited in GEO (accession # GSE338978).

Ethidium Bromide (EtBr) uptake assay

Ethidium bromide uptake assays were performed as previously described with minor modifications [57]. Wild-type, MabΔsigD, and MabΔsigD::psigDsigD3’-FLAG were grown to A600 of 0.8 washed in PBS containing 0.05% Tween 80 (PBST) and resuspended to an A600 of 0.8 in PBST. 100μL of each strain was mixed with 100μL of PBST supplemented with 0.8% glucose and 4μg/mL EtBr. A control sample lacking bacteria was also incorporated. Real-time fluorescence measurements were carried out over a time frame of 60 mins using a fluorescence spectrometer (Molecular Devices, SpectraMax M5), with excitation and emission wavelength set at 530 and 595 nm, respectively. Results were normalized by subtracting out fluorescence from the control wells. Experiments were performed using >3 biological replicates.

Lipid extraction and Thin Layer Chromatography (TLC)

Wild-type M. abscessus, ΔsigD, ΔmmpL10, and complemented strains were grown overnight to A600 = 0.8, pelleted and wet cell weight was determined. Total free lipids, apolar lipids and mycolic acid methyl esters (MAMEs) were purified as previously described with minor modifications [19,58]. Total lipids were extracted from ~100 mgs of cell pellets treated successively with CHCl3/CH3OH (1:2)) and CHCl3/CH3OH (2:1), washed with dH2O and dried. Apolar lipids were extracted from ~100mg of cell pellets resuspended in MeOH:0.3% NaCl (100:10) through 2 additions of petroleum ether, and dried. For purification of crude MAMEs, ~ 100mg of cells were resuspended in 15% tetrabutylammonium hydroxide (TBAH) and heated at 105°C for 18h. The mixture was then diluted and mixed with dichloromethane and iodomethane. After mixing for 1h, the organic layer was washed 3 times with dilute HCl followed by 2 washes with dH2O and dried. Dried total lipids, apolar lipids and MAMEs were normalized by weight before resuspension in an appropriate solvent. Total lipids were resuspended in CHCl3; apolar lipids and MAMEs were resuspended in dichloromethane, and equal quantities were loaded on Silica gel 60 F254 TLC plates (Sigma Aldrich). Total lipids were separated using CHCl3/CH3OH (90:10 v/v) and visualized by spraying the plates with a 0.2% anthrone (Sigma, 90-44-8) solution (w/v) in concentrated H2SO4 and charring. Apolar lipids were resolved in a solvent system of CHCl3:MeOH:H2O (90:10:1 v/v) x2 and visualized by spraying with 10% H2SO4 in EtOH and charring at 120°C. MAMEs were resolved in a solvent system containing petroleum ether:diethyl ether (95:5 v/v) x6 and visualized by spraying 5% phosphomolybdic acid and charring at 110°C. Lipid purification and TLCs were performed using >5 biological replicates.

Supporting information

S1 Fig. Validation of mutant strains using PCR.

Deletion of sigD and mmpL10 were checked using the following primer pairs: 5’- GCGGCCGAAGAG. CTGGCG and 5’-TCCACCTGCGCGAGCTCG (sigD) and 5’-CGGAGCCGTCACGTATTACGG and 5’-TTCGGGAACCAGTCCTTGCC (mmpL10).

https://doi.org/10.1371/journal.pgen.1012286.s001

(PDF)

S2 Fig. Functional analysis of untagged and 3XFLAG-tagged SigD.

Growth of ten-fold serial dilutions of WT, MabΔsigD, MabΔsigD::psigDsigD and MabΔsigD::psigDsigDC-FLAG strains on Middlebrook 7H10 OADC containing indicated concentrations of RIF. Data is representative of >3 independent experiments.

https://doi.org/10.1371/journal.pgen.1012286.s002

(PDF)

S3 Fig. Survival of WT M. abscessus WT and ΔsigD during nutrient starvation.

Survival of WT M. abscessus WT and MabΔsigD strains under nutrient starvation was assessed by a) incubation in PBS or b) growth to late stationary phase (A600 = 6.0). CFU/mL for each strain were measured at various timepoints by plating ten-fold serial dilutions on Middlebrook 7H10 OADC.

https://doi.org/10.1371/journal.pgen.1012286.s003

(PDF)

S1Table. Comparison of fold changes in gene expression of the direct targets of σD between WT and MabΔsigD under uninduced and INH induced conditions.

https://doi.org/10.1371/journal.pgen.1012286.s004

(PDF)

S1 Data. Differential gene expression in WT and MabΔsigD, MabΔsigD::psigDsigDC-FLAG when grown in Middlebrook 7H9 + ADC+tween to logarithmic phase and when induced with INH and EMB determined using RNA-Sequencing (RNA-Seq).

https://doi.org/10.1371/journal.pgen.1012286.s005

(XLSX)

S2 Data. Comparison of the σD regulons of M. abscessus and M. tuberculosis.

https://doi.org/10.1371/journal.pgen.1012286.s006

(XLSX)

S1 Raw Image. a-b) Wt and Mab:: MAB_3723c3’FLAG strains were grown either in PBS for 6 h or exposed to EMB 70μg/mL or INH 64μg/mL for 6 h followed by detection of RsdAFLAG using anti-FLAG antibodies.

Location of RsdA is indicated with an arrow; bands above and below RsdA are nonspecific and present in the untagged (WT) strain as well. Samples were normalized by OD, cell weight and coomassie blue staining (c-d).

https://doi.org/10.1371/journal.pgen.1012286.s007

(PDF)

Acknowledgments

We thank The Wadsworth Center’s Applied Genomics Technology Core for sequencing of ChIP-Seq libraries and the Media Core for preparation of media and buffers. We thank Anil Ojha for his expertise and guidance in purification of mycobacterial lipids, and for helpful discussions during manuscript preparation.

References

  1. 1. Olivier KN, Weber DJ, Wallace RJ Jr, Faiz AR, Lee J-H, Zhang Y, et al. Nontuberculous mycobacteria. I: multicenter prevalence study in cystic fibrosis. Am J Respir Crit Care Med. 2003;167(6):828–34. pmid:12433668
  2. 2. Kothavade RJ, Dhurat RS, Mishra SN, Kothavade UR. Clinical and laboratory aspects of the diagnosis and management of cutaneous and subcutaneous infections caused by rapidly growing mycobacteria. Eur J Clin Microbiol Infect Dis. 2013;32(2):161–88. pmid:23139042
  3. 3. Lee M-R, Sheng W-H, Hung C-C, Yu C-J, Lee L-N, Hsueh P-R. Mycobacterium abscessus Complex Infections in Humans. Emerg Infect Dis. 2015;21(9):1638–46. pmid:26295364
  4. 4. Esteban J, García-Pedrazuela M, Muñoz-Egea MC, Alcaide F. Current treatment of nontuberculous mycobacteriosis: an update. Expert Opin Pharmacother. 2012;13(7):967–86. pmid:22519767
  5. 5. Daley CL, Iaccarino JM, Lange C, Cambau E, Wallace RJ, Wallace Jr, et al. Treatment of nontuberculous mycobacterial pulmonary disease: an official ATS/ERS/ESCMID/IDSA clinical practice guideline. Clin Infect Dis. 2020;71(4):e1–36. pmid:32628747
  6. 6. Nessar R, Cambau E, Reyrat JM, Murray A, Gicquel B. Mycobacterium abscessus: a new antibiotic nightmare. J Antimicrob Chemother. 2012;67(4):810–8. pmid:22290346
  7. 7. Hurst-Hess K, McManaman C, Yang Y, Gupta S, Ghosh P. Hierarchy and interconnected networks in the WhiB7 mediated transcriptional response to antibiotic stress in Mycobacterium abscessus. PLoS Genet. 2023;19(12):e1011060. pmid:38055757
  8. 8. Brennan PJ, Nikaido H. The envelope of mycobacteria. Annu Rev Biochem. 1995;64:29–63. pmid:7574484
  9. 9. Marrakchi H, Lanéelle M-A, Daffé M. Mycolic acids: structures, biosynthesis, and beyond. Chem Biol. 2014;21(1):67–85. pmid:24374164
  10. 10. Takayama K, Wang C, Besra GS. Pathway to synthesis and processing of mycolic acids in Mycobacterium tuberculosis. Clin Microbiol Rev. 2005;18(1):81–101. pmid:15653820
  11. 11. Gavalda S, Léger M, van der Rest B, Stella A, Bardou F, Montrozier H, et al. The Pks13/FadD32 crosstalk for the biosynthesis of mycolic acids in Mycobacterium tuberculosis. J Biol Chem. 2009;284(29):19255–64. pmid:19436070
  12. 12. Grzegorzewicz AE, de Sousa-d’Auria C, McNeil MR, Huc-Claustre E, Jones V, Petit C, et al. Assembling of the Mycobacterium tuberculosis Cell Wall Core. J Biol Chem. 2016;291(36):18867–79. pmid:27417139
  13. 13. Belisle JT, Vissa VD, Sievert T, Takayama K, Brennan PJ, Besra GS. Role of the major antigen of Mycobacterium tuberculosis in cell wall biogenesis. Science. 1997;276(5317):1420–2. pmid:9162010
  14. 14. Armitige LY, Jagannath C, Wanger AR, Norris SJ. Disruption of the genes encoding antigen 85A and antigen 85B of Mycobacterium tuberculosis H37Rv: effect on growth in culture and in macrophages. Infect Immun. 2000;68(2):767–78. pmid:10639445
  15. 15. Di Capua CB, Belardinelli JM, Carignano HA, Buchieri MV, Suarez CA, Morbidoni HR. Unveiling the Biosynthetic Pathway for Short Mycolic Acids in Nontuberculous Mycobacteria: Mycobacterium smegmatis MSMEG_4301 and Its Ortholog Mycobacterium abscessus MAB_1915 Are Essential for the Synthesis of α’-Mycolic Acids. Microbiol Spectr. 2022;10(4):e0128822. pmid:35862962
  16. 16. Dubnau E, Chan J, Raynaud C, Mohan VP, Lanéelle MA, Yu K, et al. Oxygenated mycolic acids are necessary for virulence of Mycobacterium tuberculosis in mice. Mol Microbiol. 2000;36(3):630–7. pmid:10844652
  17. 17. Krasowska A, Sigler K. How microorganisms use hydrophobicity and what does this mean for human needs?. Front Cell Infect Microbiol. 2014;4:112. pmid:25191645
  18. 18. Howard ST, Rhoades E, Recht J, Pang X, Alsup A, Kolter R, et al. Spontaneous reversion of Mycobacterium abscessus from a smooth to a rough morphotype is associated with reduced expression of glycopeptidolipid and reacquisition of an invasive phenotype. Microbiology (Reading). 2006;152(Pt 6):1581–90. pmid:16735722
  19. 19. Malmsheimer S, Daher W, Tasrini Y, Hamela C, Aguilera-Correa JJ, Chalut C, et al. Trehalose polyphleates participate in Mycobacterium abscessus fitness and pathogenesis. mBio. 2024;15(12):e0297024. pmid:39475242
  20. 20. Dubois V, Viljoen A, Laencina L, Le Moigne V, Bernut A, Dubar F, et al. MmpL8MAB controls Mycobacterium abscessus virulence and production of a previously unknown glycolipid family. Proc Natl Acad Sci U S A. 2018;115(43):E10147–56. pmid:30301802
  21. 21. Burian J, Thompson CJ. Regulatory genes coordinating antibiotic-induced changes in promoter activity and early transcriptional termination of the mycobacterial intrinsic resistance gene whiB7. Mol Microbiol. 2018;107(3):402–15. pmid:29205551
  22. 22. Hurst-Hess K, Rudra P, Ghosh P. Mycobacterium abscessus WhiB7 Regulates a Species-Specific Repertoire of Genes To Confer Extreme Antibiotic Resistance. Antimicrob Agents Chemother. 2017;61(11):e01347-17. pmid:28874378
  23. 23. Feklístov A, Sharon BD, Darst SA, Gross CA. Bacterial sigma factors: a historical, structural, and genomic perspective. Annu Rev Microbiol. 2014;68:357–76. pmid:25002089
  24. 24. Manganelli R. Sigma Factors: Key Molecules in Mycobacterium tuberculosis Physiology and Virulence. Microbiol Spectr. 2014;2(1):MGM2-0007–2013. pmid:26082107
  25. 25. Sachdeva P, Misra R, Tyagi AK, Singh Y. The sigma factors of Mycobacterium tuberculosis: regulation of the regulators. FEBS J. 2010;277(3):605–26. pmid:19951358
  26. 26. Michele TM, Ko C, Bishai WR. Exposure to antibiotics induces expression of the Mycobacterium tuberculosis sigF gene: implications for chemotherapy against mycobacterial persistors. Antimicrob Agents Chemother. 1999;43(2):218–25. pmid:9925509
  27. 27. Provvedi R, Boldrin F, Falciani F, Palù G, Manganelli R. Global transcriptional response to vancomycin in Mycobacterium tuberculosis. Microbiology (Reading). 2009;155(Pt 4):1093–102. pmid:19332811
  28. 28. Pisu D, Provvedi R, Espinosa DM, Payan JB, Boldrin F, Palù G, et al. The Alternative Sigma Factors SigE and SigB Are Involved in Tolerance and Persistence to Antitubercular Drugs. Antimicrob Agents Chemother. 2017;61(12):e01596-17. pmid:28993339
  29. 29. Raman S, Hazra R, Dascher CC, Husson RN. Transcription regulation by the Mycobacterium tuberculosis alternative sigma factor SigD and its role in virulence. J Bacteriol. 2004;186(19):6605–16. pmid:15375142
  30. 30. Calamita H, Ko C, Tyagi S, Yoshimatsu T, Morrison NE, Bishai WR. The Mycobacterium tuberculosis SigD sigma factor controls the expression of ribosome-associated gene products in stationary phase and is required for full virulence. Cell Microbiol. 2005;7(2):233–44. pmid:15659067
  31. 31. Schneider JS, Sklar JG, Glickman MS. The Rip1 protease of Mycobacterium tuberculosis controls the SigD regulon. J Bacteriol. 2014;196(14):2638–45. pmid:24816608
  32. 32. Betts JC, Lukey PT, Robb LC, McAdam RA, Duncan K. Evaluation of a nutrient starvation model of Mycobacterium tuberculosis persistence by gene and protein expression profiling. Mol Microbiol. 2002;43(3):717–31. pmid:11929527
  33. 33. Fitzgerald DM, Bonocora RP, Wade JT. Comprehensive mapping of the Escherichia coli flagellar regulatory network. PLoS Genet. 2014;10(10):e1004649. pmid:25275371
  34. 34. Hurst-Hess K, Biswas R, Yang Y, Rudra P, Lasek-Nesselquist E, Ghosh P. Mycobacterial SigA and SigB Cotranscribe Essential Housekeeping Genes during Exponential Growth. mBio. 2019;10(3):e00273-19. pmid:31113892
  35. 35. Miranda-CasoLuengo AA, Staunton PM, Dinan AM, Lohan AJ, Loftus BJ. Functional characterization of the Mycobacterium abscessus genome coupled with condition specific transcriptomics reveals conserved molecular strategies for host adaptation and persistence. BMC Genomics. 2016;17:553. pmid:27495169
  36. 36. Takayama K, Kilburn JO. Inhibition of synthesis of arabinogalactan by ethambutol in Mycobacterium smegmatis. Antimicrob Agents Chemother. 1989;33(9):1493–9. pmid:2817850
  37. 37. Vilchèze C, Jacobs WR Jr. The mechanism of isoniazid killing: clarity through the scope of genetics. Annu Rev Microbiol. 2007;61:35–50. pmid:18035606
  38. 38. Hurst-Hess KR, Saxena A, Rudra P, Yang Y, Ghosh P. Mycobacterium abscessus HelR interacts with RNA polymerase to confer intrinsic rifamycin resistance. Mol Cell. 2022;82(17):3166-3177.e5. pmid:35905736
  39. 39. Block AM, Namugenyi SB, Palani NP, Brokaw AM, Zhang L, Beckman KB, et al. Mycobacterium tuberculosis Requires the Outer Membrane Lipid Phthiocerol Dimycocerosate for Starvation-Induced Antibiotic Tolerance. mSystems. 2023;8(1):e0069922. pmid:36598240
  40. 40. Lavollay M, Arthur M, Fourgeaud M, Dubost L, Marie A, Veziris N, et al. The peptidoglycan of stationary-phase Mycobacterium tuberculosis predominantly contains cross-links generated by L,D-transpeptidation. J Bacteriol. 2008;190(12):4360–6. pmid:18408028
  41. 41. Dahl JL, Kraus CN, Boshoff HIM, Doan B, Foley K, Avarbock D, et al. The role of RelMtb-mediated adaptation to stationary phase in long-term persistence of Mycobacterium tuberculosis in mice. Proc Natl Acad Sci U S A. 2003;100(17):10026–31. pmid:12897239
  42. 42. Sarathy J, Dartois V, Dick T, Gengenbacher M. Reduced drug uptake in phenotypically resistant nutrient-starved nonreplicating Mycobacterium tuberculosis. Antimicrob Agents Chemother. 2013;57(4):1648–53. pmid:23335744
  43. 43. Burbaud S, Laval F, Lemassu A, Daffé M, Guilhot C, Chalut C. Trehalose Polyphleates Are Produced by a Glycolipid Biosynthetic Pathway Conserved across Phylogenetically Distant Mycobacteria. Cell Chem Biol. 2016;23(2):278–89. pmid:27028886
  44. 44. Thouvenel L, Prevot G, Chiaradia L, Parra J, Mouton-Barbosa E, Locard-Paulet M, et al. The final assembly of trehalose polyphleates takes place within the outer layer of the mycobacterial cell envelope. J Biol Chem. 2020;295(32):11184–94. pmid:32554804
  45. 45. Liu J, Nikaido H. A mutant of Mycobacterium smegmatis defective in the biosynthesis of mycolic acids accumulates meromycolates. Proc Natl Acad Sci U S A. 1999;96(7):4011–6. pmid:10097154
  46. 46. Gao L-Y, Laval F, Lawson EH, Groger RK, Woodruff A, Morisaki JH, et al. Requirement for kasB in Mycobacterium mycolic acid biosynthesis, cell wall impermeability and intracellular survival: implications for therapy. Mol Microbiol. 2003;49(6):1547–63. pmid:12950920
  47. 47. Le Moigne V, Roux AL, Jobart-Malfait A, Blanc L, Chaoui K, Burlet-Schiltz O. A TLR2-activating fraction from Mycobacterium abscessus rough variant demonstrates vaccine and diagnostic potential. Front Cell Infect Microbiol. 2020;10:432. pmid:32984067
  48. 48. Taniguchi H, Busche T, Patschkowski T, Niehaus K, Pátek M, Kalinowski J, et al. Physiological roles of sigma factor SigD in Corynebacterium glutamicum. BMC Microbiol. 2017;17(1):158. pmid:28701150
  49. 49. Toyoda K, Inui M. Extracytoplasmic function sigma factor σD confers resistance to environmental stress by enhancing mycolate synthesis and modifying peptidoglycan structures in Corynebacterium glutamicum. Mol Microbiol. 2018;107(3):312–29. pmid:29148103
  50. 50. Hart EM, Lyerly E, Bernhardt TG. The conserved σD envelope stress response monitors multiple aspects of envelope integrity in corynebacteria. PLoS Genet. 2024;20(6):e1011127. pmid:38829907
  51. 51. Lee J, Ammerman N, Agarwal A, Naji M, Li S-Y, Nuermberger E. Differential In Vitro Activities of Individual Drugs and Bedaquiline-Rifabutin Combinations against Actively Multiplying and Nutrient-Starved Mycobacterium abscessus. Antimicrob Agents Chemother. 2021;65(2):e02179-20. pmid:33168614
  52. 52. Tjaden B. De novo assembly of bacterial transcriptomes from RNA-seq data. Genome Biol. 2015;16(1):1. pmid:25583448
  53. 53. McClure R, Balasubramanian D, Sun Y, Bobrovskyy M, Sumby P, Genco CA, et al. Computational analysis of bacterial RNA-Seq data. Nucleic Acids Res. 2013;41(14):e140. pmid:23716638
  54. 54. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15(12):550. pmid:25516281
  55. 55. Jaini S, Lyubetskaya A, Gomes A, Peterson M, Tae Park S, Raman S, et al. Transcription Factor Binding Site Mapping Using ChIP-Seq. Microbiol Spectr. 2014;2(2):10.1128/microbiolspec.MGM2-0035–2013. pmid:26105820
  56. 56. Bailey TL, Johnson J, Grant CE, Noble WS. The MEME Suite. Nucleic Acids Research. 2015;43(W1):W39-49. pmid:25953851
  57. 57. Rodrigues L, Aínsa JA, Viveiros M. Measuring Efflux and Permeability in Mycobacteria. Methods Mol Biol. 2021;2314:231–45. pmid:34235655
  58. 58. Ojha AK, Trivelli X, Guerardel Y, Kremer L, Hatfull GF. Enzymatic hydrolysis of trehalose dimycolate releases free mycolic acids during mycobacterial growth in biofilms. J Biol Chem. 2010;285(23):17380–9. pmid:20375425