Making Ends Meet: Repairing Breaks in Bacterial DNA by Non-Homologous End-Joining

  • Richard Bowater,


  • Aidan J Doherty mail

    To whom correspondence should be addressed. E-mail:


  • Published: February 24, 2006
  • DOI: 10.1371/journal.pgen.0020008

Reader Comments (1)

Post a new comment on this article

Non-Homologous End Joining and Mismatch Repair: Never the Twain Shall Meet?

Posted by PLoS_Genetics on 20 Feb 2008 at 10:51 GMT

Originally submitted as a Reader Response by Nicole J. Moreland ( on 5 June 2006:

In their recent review [1], Bowater and Doherty explain that many prokaryotes that retain Non-homologous End Joining (NHEJ) spend much of their life cycle in stationary phase and that NHEJ may have evolved as a pathway to repair DNA strand breaks (DSBs) that arise during this cell cycle phase. They describe the sporadic distribution of NHEJ genes amongst prokaryotes and suggest this may be driven by environmental factors rather than by phylogeny.

Building on this speculation, this letter proposes that a reliance on NHEJ is linked to the striking absence of mismatch repair (MMR) in mycobacteria. Mycobacteria have a robust NHEJ pathway that requires Ku and an ATP-dependent ligase, ligD [2]. Indeed, it was in Mycobacterium tuberculosis (Mtb) that bacterial NHEJ was first discovered, but genome analysis and experimental studies have found no evidence for the existence of a MMR pathway in mycobacteria [3,4]. The absence of homologues for the otherwise extremely conserved MMR genes mutS and mutL has puzzled researchers for some time, but with the discovery of NHEJ perhaps an explanation now exists? MMR repair plays a critical role in rejecting recombination between moderately divergent DNA sequences [5] and has been shown to directly modulate NHEJ in eukaryotes [6,7]. Notably, eukaryote homologues of bacterial MutS and MutL (Msh2 and Mlh1) reduce NHEJ events between DNA ends with no microhomology. Could the absence of MMR repair in mycobacteria highlight a unique reliance on NHEJ for survival?

Vital to the pathogenic success of Mtb is its ability to survive in a quiescent state within host macrophages. During this latent phase, the bacterium is exposed to genotoxic agents produced by the host cell and resulting DSBs must be repaired in the absence of a daughter chromatid. Fortunately, NHEJ is a process that allows direct rejoining of DNA ends formed in a DSB. NHEJ has been shown to be extremely mutagenic in mycobacteria with an error rate of ~50% [2], but this is likely balanced by an increased ability to survive in a quiescent state. The high error rate may be linked to the absence of MMR, with MutS and MutL not present to inhibit NHEJ between DNA ends lacking microhomology. This concept is supported by studies in eukaryotes in which an increase in promiscuous NHEJ is seen when MMR machinery is lost [6,7].

But what of other prokaryotes that possess NHEJ pathways? To date, mycobacteria are the only known prokaryotes displaying a complete absence of MMR, so how do other species regulate the two systems? It appears that the cell cycle plays a role via a down-regulation of MMR during stationary phase. Most prokaryotic research on MMR regulation has been performed in Escherichia coli K12 [8,9] which does not harbor NHEJ components [1]. In E. coli, key components of MMR, particularly MutS, are down-regulated in stationary-phase cells and this has been postulated to enhance recombination between homologous DNA [9]. It is not difficult to imagine that in a species with functional NHEJ, a down-regulation of MMR in stationary phase would enhance the NHEJ pathway. Indeed in B. subtillis, which is functional for NHEJ [1], a strain deficient in MutSL showed a significant increase in stationary-phase mutants indicating an increase in error-prone repair [10].

As Bowater and Doherty explain, open reading frames encoding NHEJ proteins exist in a diverse range of bacteria [1], but reliance on this pathway for survival may vary greatly. A complete elimination of MMR in the successful pathogen Mtb suggests this species is primed to perform NHEJ. In other species where MMR is functional, utilization of NHEJ may be less absolute and a balancing act between the two systems in response to the environment is likely to exist.

I am grateful to Professor Ted Baker and Dr. Shaun Lott for their helpful suggestions in preparing this response.

1. Bowater R and Doherty AJ (2006) PLoS Genet 2 (2) e8
2. Gong C, Bongiorno P, Martins A, Stephanou NC, Zhu H, et al. (2005) Nat Struct Mol Biol 12: 304-312
3. Mizrahi V and Anderson SJ (1998) Mol Microbiol 29(6):1331-9
4. Springer B, Sander P, Sedlacek L, Hardt WD, Mizrahi V et al (2004) Mol Microbiol 53(6):1601-9
5. Vulic M, Dionisio F, Taddei F, Radman M (1997) Proc Natl Acad Sci USA 94 (18):9763-7
6. Bannister LA, Waldman BC and Waldman AS (2004) DNA repair (Amst) 3(5):465-74.
7. Smith JA, Waldman BC, Waldman AS (2005) Genetics 170(1):355-63
8. Harris RS, Feng G, Ross KJ, Sidhu R, Thulin C et al (1999) Mutat Res 437(1):51-60
9. Foster PL (2005) Mutat Res 569: 3-11
10. Pedraza-Reyes M, Yasbin RE (2004) J Bacteriol. 186(19):6485-91