Comprehensive functional analysis of Mycobacterium tuberculosis toxin-antitoxin systems: implications for pathogenesis, stress responses, and evolution - PubMed (original) (raw)
Comprehensive functional analysis of Mycobacterium tuberculosis toxin-antitoxin systems: implications for pathogenesis, stress responses, and evolution
Holly R Ramage et al. PLoS Genet. 2009 Dec.
Abstract
Toxin-antitoxin (TA) systems, stress-responsive genetic elements ubiquitous in microbial genomes, are unusually abundant in the major human pathogen Mycobacterium tuberculosis. Why M. tuberculosis has so many TA systems and what role they play in the unique biology of the pathogen is unknown. To address these questions, we have taken a comprehensive approach to identify and functionally characterize all the TA systems encoded in the M. tuberculosis genome. Here we show that 88 putative TA system candidates are present in M. tuberculosis, considerably more than previously thought. Comparative genomic analysis revealed that the vast majority of these systems are conserved in the M. tuberculosis complex (MTBC), but largely absent from other mycobacteria, including close relatives of M. tuberculosis. We found that many of the M. tuberculosis TA systems are located within discernable genomic islands and were thus likely acquired recently via horizontal gene transfer. We discovered a novel TA system located in the core genome that is conserved across the genus, suggesting that it may fulfill a role common to all mycobacteria. By expressing each of the putative TA systems in M. smegmatis, we demonstrate that 30 encode a functional toxin and its cognate antitoxin. We show that the toxins of the largest family of TA systems, VapBC, act by inhibiting translation via mRNA cleavage. Expression profiling demonstrated that four systems are specifically activated during stresses likely encountered in vivo, including hypoxia and phagocytosis by macrophages. The expansion and maintenance of TA genes in the MTBC, coupled with the finding that a subset is transcriptionally activated by stress, suggests that TA systems are important for M. tuberculosis pathogenesis.
Conflict of interest statement
The authors have declared that no competing interests exist.
Figures
Figure 1. Identification and testing of putative TA systems.
(A) Three approaches were used to identify putative TA systems. These are indicated as BLAST (Non-PIN; homologs found through BLAST analysis that do not contain PIN domains), PIN Domain (PIN domain-containing proteins), and Genome Org. Only (novel and not homologous to known TA systems). (B) M. smegmatis cultures with putative toxins or putative toxin-antitoxin pairs under the control of the inducible acetamidase promoter were serially diluted and plated on solid media with (right panel) or without (left panel) 0.2% acetamide. (C) Summary of toxin and antitoxin testing results: not tested (unable to PCR or clone gene products), not toxic (no toxin activity was detected), toxic only (toxicity was not relieved by the putative antitoxin), or a TA system (toxic activity relieved by antitoxin). (D) Functional TA systems were identified as novel, PIN domain-containing proteins, or homologs found by BLAST (non-PIN domain-containing proteins). Genes found by BLAST were further subdivided by the TA system family to which they belong.
Figure 2. TA system conservation across the genus Mycobacterium.
Phylogenetic tree based on 16S rDNA sequences showing conservation of TA systems. The tree was constructed using Neighbor-joining inference method and nodes supported by bootstrap values>70% (1,000 replicates) are shown. Nocardia farcinica (Nfa) was used as the outgroup. TA systems are arranged according to family (vapBC, mazEF, relBE, parDE, higBA, and novel); for details see Table S6. Orange represents orthologs (BLAST best reciprocal hits displaying synteny), yellow: BLAST best reciprocal hits residing in different genomic contexts (homologs), blue: pseudogenes residing in similar genomic context, green: pseudogenes residing in different genomic contexts, and black indicates no hits were detected by BLAST. Abbreviations: Nfa Nocardia farcinica; Mab Mycobacterium abscessus; Mgi Mycobacterium gilvum; Msm Mycobacterium smegmatis; Mva Mycobacterium vanbaalenii; Mjl Mycobacterium sp. JLS; Mmc Mycobacterium sp. MCS; Mkm Mycobacterium sp. KMS; Mle Mycobacterium leprae; Mka Mycobacterium kansasii; Mpa Mycobacterium avium str. k10; Mav Mycobacterium avium 104; Mul Mycobacterium ulcerans; Mma Mycobacterium marinum; Mtu Mycobacterium tuberculosis; Mmi Mycobacterium microti; Mbo Mycobacterium bovis; Maf M. africanum; Mca Mycobacterium canetti.
Figure 3. Conservation of a novel TA system.
(A) Genomic regions of Rv0909-Rv0910 TA module orthologs (orange) across diverse mycobacteria showing conservation of toxin and cognate antitoxin genes as well as surrounding genomic context, as indicated by color coding of orthologs across species. (B) M. smegmatis Rv0909-0910 orthologs encode a functional TA system. M. smegmatis cells expressing MSEMG_5634 alone or MSMEG_5634-5635 under the control of the inducible acetamidase promoter were serially diluted and plated on solid media with (right panel) or without (left panel) 0.2% acetamide. (A) adapted from the tree-browser function in MicrobesOnline .
Figure 4. VapB Antitoxins are specific for their cognate VapC toxins.
M. smegmatis cultures carrying VapC toxins under control of the inducible acetamidase promoter and VapB antitoxins under the control of a tetracycline-inducible promoter were assessed for toxin activity. The VapC toxin being tested is indicated on the left side and the VapB antitoxins are indicated across the top of each set of panels. Each set of panels includes strains tested on solid media without (top) and with (bottom) inducers. The cognate toxin-antitoxin pair for each set is indicated in blue.
Figure 5. VapC homologs have RNase activity and inhibit translation but a novel toxin does not.
(A) Cultures of M. smegmatis harboring empty vector (pHR100) or acetamide-inducible toxin constructs were treated with 0.2% acetamide. At the indicated times, cells were labeled with of 35S-methionine at 37°C for 1 min and incorporation of radioactivity was measured. Incorporation at t = 0 was set as 100% translation. As controls, cells were treated with 0.5 µg/ml ciprofloxacin (cip), or 25 µg/ml hygromycin (hyg). The average of three experiments is shown and error bars represent the standard deviation. (B) Cultures of M. smegmatis were grown and induced as described above. The OD600 of each culture was measured at the indicated times. Results are plotted as fold-increase of OD600 at each timepoint as compared to OD600 at t = 0. The average of three experiments is shown error bars represent the standard deviation. (C) Purified toxins were incubated with MS2 RNA for 3 h at 37°C. The RNA was then purified and electrophoresed in a 1% denaturing agarose gel. Included as controls were RNA alone (RNA), His-MBP (MBP), and E. coli MazF (MazF). (D) Purified toxins MazF and Rv0301 were incubated with their respective GST-tagged antitoxins MazE (10 µg) and Rv0300 (5 µg) and 0.8 µg MS2 RNA for 3 h at 37°C. Reactions were electrophoresed in a 2% agarose gel.
Figure 6. Subsets of TA systems are activated during stress.
(A) Cultures were grown in 1 liter roller bottles with a headspace ratio of 0.5 (830 ml culture) at 37°C, with slow stirring to induce NRP. At the indicated days, cells were collected and RNA was isolated and amplified. Gene expression was measured using qPCR. One of three similar experiments is shown and error bars represent the standard deviation within this experiment. (B) IFN-γ stimulated bone marrow-derived macrophages were infected with M. tuberculosis at an MOI of 10. Bacterial RNA from intracellular bacilli was isolated and amplified. Gene expression was measured by qPCR and normalized to 16S rRNA as an internal control. Gene expression from in vitro grown log-phase cultures (log) is included for comparison. One of three similar experiments is shown and error bars represent the standard deviation within this experiment.
Figure 7. Model of stress-induced TA system activation and proteome remodeling.
(A) Under normal growth conditions, the antitoxin is bound to its cognate toxin, this complex in turn binds its own promoter, inhibiting transcription. (B) In response to stress, the antitoxin is specifically degraded, releasing the toxin to cleaves existing transcripts. Additionally, degradation of the antitoxin results in increased transcription of the TA system. (C) Upon stabilization of the antitoxin, the TA system is inactivated and toxin-antitoxin complex, resulting in transcriptional inhibition of its operon. This pulse of toxin activity functionally erases the previous transcriptional profile allowing the newly stress induced messages to be preferentially translated.
References
- Gerdes K, Christensen SK, Lobner-Olesen A. Prokaryotic toxin-antitoxin stress response loci. Nat Rev Microbiol. 2005;3:371–382. - PubMed
- Kolodkin-Gal I, Hazan R, Gaathon A, Carmeli S, Engelberg-Kulka H. A linear pentapeptide is a quorum-sensing factor required for mazEF-mediated cell death in Escherichia coli. Science. 2007;318:652–655. - PubMed
Publication types
MeSH terms
Substances
LinkOut - more resources
Full Text Sources
Molecular Biology Databases