OUP user menu

Simple sequence repeats (microsatellites): mutational mechanisms and contributions to bacterial pathogenesis. A meeting review

Christopher D. Bayliss, Kevin M. Dixon, E. Richard Moxon
DOI: http://dx.doi.org/10.1016/S0928-8244(03)00325-0 11-19 First published online: 1 January 2004


This review summarises the presentations and discussions that took place during a European Science Foundation-funded workshop whose purpose was to gain current perspectives on the mutational mechanisms of simple sequence repeats and the contribution of localised hypermutation in such repeats to bacterial pathogenesis. In vitro biophysical and biochemical assays of mutational mechanisms were covered as well as genetic studies in various eukaryotic and prokaryotic organisms. Presentations on bacterial pathogenesis elaborated investigations of the use of repeats for typing of strains, epidemiological investigations of mutation rates and functions of loci whose expression is controlled by simple sequence repeats. This review tabulates current perspectives on the cis- and trans-acting factors for mutation of simple sequence repeats and the orientations of mononucleotide repeats in some bacterial species that utilise repeats for adaptation.

  • Microsatellites
  • Simple sequence repeat
  • Contingency genes
  • Phase variation
  • Bacterial pathogenesis
  • Mutation rates

1 Introduction

Microbial fitness and survival is dependent on a balance between genome maintenance and generation of genetic diversity. Bacterial pathogens faced by temporally and spatially dynamic and often hostile host environments have evolved multiple mutational and recombinatorial mechanisms for increasing their rates of generation of genetic diversity and hence the rapidity with which they can adapt to environmental fluctuations. Consequentially, these processes also impinge on pathogenic potential. Simple sequence repeats are hypermutable DNA sequences that act as a major source of functional genetic diversity for many bacterial pathogens [1]. Similar sequences, termed microsatellites, are abundant in the human genome and are highly polymorphic between individuals resulting in their extensive use for genetic mapping [2]. Alterations in the lengths of microsatellites also cause a number of genetic disorders and global instability of microsatellites is associated with several cancers [3]. Consequently, the mutational mechanisms of microsatellites have been intensely studied using Escherichia coli, yeast and human cells as model systems. Simple sequence repeats/microsatellites are, therefore, of major importance to the biology of both bacteria and higher eukaryotes and yet researchers in these fields rarely interact.

In March 2003, under the auspices of and with funding from the European Science Foundation, a workshop took place in Oxford, United Kingdom, that was attended by researchers whose expertise lay in either eukaryotic or prokaryotic microsatellites. The title of this workshop was ‘Hypermutation of simple sequence repeats (microsatellites) and the contribution of hypermutation to bacterial pathogenesis’ and its aims were to explore current knowledge of two fundamental questions. Firstly, what are the mechanisms controlling the mutation rates of microsatellites in various organisms? Secondly, what contribution is made to bacterial pathogenesis/adaptive behaviour by elevated mutation rates in either localised regions of genomes (e.g. loci containing microsatellites) or across the whole genome? The meeting encompassed four sessions:- (i) cis-acting factors for microsatellites; (ii) trans-acting factors for microsatellites; (iii) cis-/trans-acting factors in phase variable gene expression; and (iv) studies of localised or global hypermutation using epidemiological samples or models systems. This review summarises the content of the talks and presents perspectives on the mutational processes of microsatellites and the influence of these processes on bacterial adaptation.

2 Cis-acting factors

DNA structure is a field poorly understood by many non-specialists such that the proportional relationship between melting temperature and (G+C)% seen for ‘bulk’ DNA is often applied to specific sequences. Mikael Leijon described that in pure A-tracts the base pairs are ‘propeller twisted’ optimising stacking interactions between neighbouring bases whereas in G-tracts the base pairs are more planar due to steric interference of the propellar twist by the N2-amino group. Contrastingly, dipolar interactions between GC base pairs make stacking more favourable in CG or GC-steps than GG-steps. Using nuclear magnetic resonance (NMR), the high propellar twist of A-tracts was found to lower base pair disassociation constants and base pair opening times in comparison to G-tracts [4,5]. This higher stability of A-tracts in internal positions may, however, be countered by the higher ‘end-fraying’ of A-tracts. Increases in length for G-tracts elevated base pair disassociation constants, suggesting that long poly-G tracts may be particularly unstable.

Intricate interactions occur between protein and nucleic acid during DNA polymerisation. Using an in vitro assay of DNA replication fidelity and detailed structural models, Tom Kunkel expounded on the impact of protein structure and nucleotide sequence on mutation rates. DNA polymerases make multiple contacts with the DNA substrate over a number of base pairs (10 for T7 DNA polymerase) reconfiguring the DNA structure and permitting accurate selection of nucleotides for polymerisation and detection of base substitution errors/misalignments [6]. Multiple types of DNA polymerases have been identified and these enzymes exhibit variations in structure that alter protein:DNA contacts at the replication fork and hence DNA replication error rates. Similarly variations in nucleotide sequence influence these contacts and hence error rates for DNA polymerisation. For frameshift mutations, processivity of the DNA polymerase was a key factor, possibly as a result of the DNA opening required for entry of one strand into the exonuclease site. Mismatch repair (MMR), however, was highlighted as the major factor for reducing frameshift mutations in repeat tracts. Saturation of MMR — as a result of rapid growth rate, DNA damage or inhibition by environmental carcinogens — is therefore a critical aspect of microsatellite stability [7].

The differing mutagenic effects of leading and lagging strand DNA synthesis have been investigated by Iwona Fijalkowska's group using chromosomally located reporter constructs in E. coli. For missense mutations the leading strand is more mutagenic during normal growth than the lagging strand [8], this phenomenon being reversed during untargeted mutagenesis when Pol V (encoded by umuCD) is induced [9]. This result suggests that the SOS mutator activity operates with differing efficiencies on the two strands and that the lagging strand is most susceptible to the SOS mutator effect. The rate of generation of frameshift mutations also differed between leading and lagging strand. Notably, G-tracts had higher mutation rates than A-tracts [10]. Presented results indicate that DNA strand orientation effects the fidelity of replication in a manner dependent on the type of errors made by the DNA polymerase, the involvement of particular DNA polymerases and on the physiological conditions in the cell.

The discussion, led by Thomas Petes, split the cis-acting factors into those pertaining either to the properties or the context of the repeat tracts (Table 1). Predictive criteria were identified for two properties, repeat number and repeat unit size. In the first case, mutation rate increases as the number of perfect tandem repeats in a repeat tract increases. In the second case, there is an interplay between repeat unit size and MMR. Thus in prokaryotes repeat tracts with unit sizes of less than four are subject to MMR but larger ones are not, whilst in eukaryotes all unit sizes up to about 15 bp are subject to MMR. For other properties and for contextual properties general rules are still not clear. It was noted, however, that polyG tracts are usually more mutagenic than polyA tracts of equivalent repeat number and that imperfect repeats within a tract increase the stability of the tract.

View this table:
Table 1

Cis-acting factors that influence the stability of simple sequence repeats

Cis-acting factors
Repeat numberLocal DNA sequence context
Repeat unit lengthTranscription
Repeat sequence (includes DNA structures)Orientation relative to DNA replication
Purity of repeat tractGlobal effects (positioning relative to nucleosomes, chromosome structure)

3 Trans-acting factors

Organisms have multiple pathways of DNA repair and new activities are still being found. Erling Seeberg described some new alkylation repair enzymes whose identification began with complementation of alkylation-sensitive E. coli mutants [11]. The Ada protein is active on O6-methyl(me)-guanine and O4-me-thymine and is also a transcriptional regulator. AlkA is active on 7-me-guanine and 3-me-adenine and over-expression of this protein increases mutation rate by introducing abasic sites into the genome. AlkB is active on 1-me-adenine and 3-me-cytosine and on methylated nucleotides in ssRNA, i.e. it is an RNA damage repair enzyme [12]. Serendipitously, these studies led to the identification of non-coding RNAs that are up-regulated in response to DNA damage and whose promoters contain LexA binding sites. An exciting prospect is that there are many non-coding RNAs and that these genes may modulate the responses of pathogens to host environments.

The mutagenic effects of mutations in MMR-genes on microsatellites of both yeast and Caenorhabditis elegans were described by Thomas Petes. In yeast, mutations in CAN1 were increased 33-fold by inactivation of msh2 (involved in MMR) with the percentage of mutations due to frameshifts changing from 15% in wild-type cells to 75% in mutant cells. MMR mutations in yeast were also shown to destabilise microsatellites, for example the mutation rate of a 33 bp 5′GT repeat tract was increased 340-, 47- and 2.1-fold by msh2, msh3 and msh6 mutations, respectively [13]. In nematode worms, 14 and 26 5′GT repeat tracts were destabilised 100- and 220-fold, respectively, by an msh2 mutation [14]. Intriguingly, MMR mutations in nematodes lead to extinction of some worm lineages during serial passage, presumably due to the accumulation of mutations and likely through effects on fecundity and senescence. Telomeres are a specialised type of microsatellite consisting of imperfect repeats. Deletion of telomerase in yeast results in loss of 2 repeats per division. Mutations in two protein kinases, TEL1 and MEC1, resulted in very short telomeres and an increase in chromosome loss and telomere–telomere fusions [15].

Translesion DNA synthesis (TLS) is the process whereby DNA replication proceeds through DNA damage with the trade-off of introducing mutations into the genome. Jerome Wagner discussed research using defined substrates and genetic mutants in E. coli that demonstrates roles for all three SOS-induced DNA polymerases (II, IV and V) in TLS. Moreover, depending on both the lesion and the sequence context, some mutation pathways exhibit absolute requirements for specific combinations of DNA polymerases whereas for some other pathways there is functional redundancy [16]. These data illustrate the complexity of TLS processes and imply exchanges of DNA polymerases in the vicinity of the lesion. Within this ‘polymerase switching model’ context, biochemical, structural and genetic studies point to the β-clamp (the processivity subunit of the replicative DNA Pol III holoenzyme) as an essential interaction platform [17].

The contribution of misfolding of simple sequence repeats to instability of microsatellites is a daunting issue. David Leach has tackled this issue for human triplet repeats using assays of phage λ plaque size or of alterations in size of plasmid-borne repeat tracts in E. coli. It was observed that certain triplet repeats, such as CAG.CTG, demonstrate a propensity to form hairpin structures and that the preference for even or odd membered hairpins is sequence dependent, determined by a requirement for a 5′C.3′G base-pair closing the terminal loop of unpaired bases [18]. For a repeat tract containing 24 triplets of the sequence TGG, propagation is strongly orientation dependent, being reduced when the TGG repeat strand forms the template for lagging-strand synthesis [19]. This effect is relieved in mutS, recA or sbcCD (encoding a dsDNA exonuclease known to cleave hairpins) mutants but not mutL, indicating a role for MutS in promoting or stabilising an unusual DNA secondary structure containing mismatches rather than a requirement for MMR. Triplet repeats can also form other structures. A long TGG repeat tract can form a quadraplex and this tract cannot be maintained on the lagging strand of a plasmid in E. coli. Maintenance is restored by inactivation of sbcCD or mutS and recA. The absence of a correlation between repeat array length changes and folding preferences argues that expansion/contraction of these tracts in E. coli involves a replicative mechanism that is insensitive to misfolding preferences.

Multiple human DNA polymerases (∼14) are known to exist. Tom Kunkel described the biochemical activities of these polymerases and the contribution of loss of these activities to human diseases. Thus a reduction in fidelity of Polγ results in progressive external ophthalmoplegia (PEO) whilst inactivation of Polη, which can copy past a thymidine dimer, causes Xerodosum pigmentosum [20]. The role of different DNA polymerases in somatic hypermutation is also being clarified with Polη and Polτ, which inserts incorrect bases in up to 72% of replication events, being strong candidates for generation of the sequence divergence characteristic of this process [20]. An exciting speculation concerned a demonstration of the molecular mechanism of damage caused by an environmental carcinogen and the potential to modify these effects.

Many human disorders are associated with the expansion of simple sequence repeats. Darren Monckton presented analyses of alterations in repeat tracts for human sperm, blood and ex-vivo-cultured cells derived from patients with myotonic dystrophy and other diseases. Very high mutation rates were observed for some tracts with large biases towards expansions (in one example 98% of all mutations were expansions). Cis-factors demonstrated to effect repeat stability included tract length and context [21] with a high G+C% content of the flanking DNA increasing mutation rates [22]. Notably, expansions seem to result from many small changes over time. Observations, using a mouse model of repeat instability [23], of expansions in non-dividing cells and varying effects of MMR mutations indicates that inappropriate MMR of unusual structures, and not replication slippage, has a major impact on repeat instability and as a consequence also on disease progression. Significantly, preliminary data were noted for effects on repeat instability from genotoxic agents.

The metabolism of DNA (replication, repair, transcription, recombination) involves a large number of inter-connected and co-ordinately regulated pathways and hundreds of different proteins. These pathways may be regulated temporally and/or by physiological parameters. The mutation rates for a particular organism, or even for a particular cell, will be dependent on the DNA metabolism pathways/proteins active in that organism and the physiological parameters to which that organism is exposed. The discussion, led by Erling Seeberg, explored our current knowledge of the genetic and physiological trans-acting factors that influence simple sequence repeat stability (see Fig. 1). The genetic factors are summarised in Table 2. One known physiological parameter is oxidative stress which destabilises microsatellites in E. coli[24], however, the mechanism is unknown.

Figure 1

Potential cis- and trans-acting factors for destabilisation of simple sequence repeats. This figure presents a generic view of the processes and factors that are known or potential determinants of the stability of simple sequence repeat tracts. Note that particular simple sequence repeat tracts are not subject to every factor depicted herein. The figure depicts a continuous simple sequence repeat tract that has undergone partial DNA replication. Repeats are represented by filled grey rectangles.

View this table:
Table 2

Trans-acting factors that influence the stability of simple sequence repeats

DNA metabolism pathwayInfluence on repeat instabilityUnit sizes (1–9 bp)Eukaryotic/prokaryotic
DNA replication:
Okazaki fragment processingMedium/highAllBoth
DNA recombinationNone/low?All?Prok
None/low?3, others?Euk
DNA transcriptionLow2, 3, others?Both?
DNA repair:
NERNone/NDFew studiesNR
BERNDNo studies?NR
TLSMedium1, others?Prok
Damage reversalNDNo studies?NR
dsDNA break repairNDNo studies?NR
ssDNA break repairNDNo studies?NR
Post-replication repairNDNo studies?NR
  • ND=No data available, NR=not relevant.

4 Cis- and trans-acting factors in bacteria with simple sequence contingency loci (SSCL)

The generation of genetic diversity is critical for bacterial adaptation. Richard Moxon introduced the mechanisms in pathogenic bacteria that generate genetic diversity and the revolution in our thinking that has come from the availability of complete genomes. The inflexibility of gene regulation as a response to stringent selection was contrasted with the ‘smart’ behaviour offered by horizontal gene transfer, mutation, ‘mutators’ (i.e. bacterial strains with high global mutation rates) and localised hypermutation [1]. The focus, however, was on the stochastic and rapid generation of genetic diversity by localised hypermutation in ‘contingency loci’ encoding products (e.g. surface molecules) that are under stringent selection. Simple sequence repeats provide a major mechanism of hypermutation in such loci. Genome sequencing has led to the identification of multiple contingency loci containing simple sequence repeat tracts in numerous genomes of bacterial pathogens [25,26]. The interplay between localised hypermutation and mutators was highlighted by reference to Igor Stojiljkovic's discovery of the widespread occurrence of mutator strains in Neisseria meningitidis group A strains [27].

Capsule switching rates of N. meningitidis are not altered by non-polar inactivation of dam but are increased by bicyclomicin [28], a Rho-inhibitor, were the initial points of Pietro Alifano. Through linkage analysis of different strains, allelic variations in recB were identified and shown to alter UV sensitivity, transformation rates and pilin phase variation rates [29].

The genome of N. meningitidis contains many repeat tracts. Patricia Martin described a comparative sequencing approach that defined predictive length/sequence criteria for microsatellites being associated with contingency loci in N. meningitidis[30]. Analysis of switching rates in defined N. meningitidis mutants showed that inactivation of mutS destabilised long and short mononucleotide repeat tracts whereas over-expression of Pol IV only effected short tracts. Tetranucleotide repeat tracts were not destabilised in either mutant. Finally, a novel mechanism of tetranucleotide repeat tract-mediated alterations in gene expression was elaborated.

The Haemophilus influenzae genome contains multiple tetranucleotide repeat tracts and Chris Bayliss reviewed the research into the cis-/trans-acting factors controlling the mutation rates of these and other repeats in this bacterial species. Analysis of mutation rates using reporter constructs and defined mutants showed that tract length, Pol I polymerase activity or RnaseH are major factors controlling tetranucleotide repeat tract stability in H. influenzae whilst MMR and some other repair pathways have no effect [31,32]. In contrast, dinucleotide repeat-mediated switching of reporter constructs, but not surprisingly pilus whose switching is mediated by 5′TA repeats, were shown to be elevated by MMR mutations.

Transformation in N. meningitidis and H. influenzae is enhanced by a particular DNA sequence, the uptake sequence. Tone Tonjum presented a bioinformatic analysis of genomes of these species that has led to the observation of an overabundance of uptake sequences in DNA maintenance genes, potentially indicating that these genes are frequently subject to horizontal transfer.

Discussion, led by Richard Moxon, highlighted the paucity of knowledge relating to the cis- and trans-acting factors controlling mutation rates of simple sequence repeats in bacteria containing multiple SSCL. One theme was the absence of an overview of bioinformatic information for these species. Table 3 provides a limited set of this type of information. This table enumerates the numbers of mononucleotide repeats with different sequences (both totals and numbers that mediate switches in gene expression) and the orientation with respect to DNA replication and transcription. Consideration of this type of data may indicate whether there are correlations between, for example, orientation of repeats and DNA replication.

View this table:
Table 3

Orientation of mononucleotide repeats in SSCL

OrganismRepeat typeTotal no.SSCLTranscriptional orientationReplication orientation
N. meningitidis MC58Poly(G)≥5 (≥7)1628 (47)331627
Helicobacter pylori 26695Poly(G)≥5 (≥7)2717 (188)1449
C. jejuni NCTC 11168Poly(G)≥5 (≥7)351 (29)282721
H. influenzae KW20Poly(G)≥5 (≥7)619 (25)0NRNR
  • Numbers of SSCL within a genome containing each mononucleotide repeat type are taken from relevant publications [25,26,33,49]. Repeat tract lengths for SSCL are as defined [30].

  • Number of SSCL in which the Poly(G) tract is on the coding strand.

  • Number of SSCL in which the Poly(G) tract is on the leading strand.

  • NR=not relevant.

5 Epidemiology and functional consequences of elevated mutation rates

Campylobacter jejuni is a commensal of chickens that causes a wide-spectrum of diseases in humans ranging from mild gastoentritis to Guillain Barre syndrome. Brendan Wren highlighted the critical role of genome sequencing in advancing studies of localised hypermutation in C. jejuni. Analysis of the genome sequence enabled the rapid identification of loci containing microsatellites [33]. Subsequent genetic and biochemical analysis demonstrated that some of these microsatellites control expression of loci involved in biosynthesis of capsule, addition of terminal sugar residues to lipopolysaccharide (LPS) and motility [34]. It was speculated that escape of phage infection is major factor driving localised hypermutation in this bacterial species.

Mycoplasmas are wall-less microorganisms with small genomes. David Yogev introduced us to the extensive antigenic variation of lipoproteins that occurs in many mycoplasma species. Sectoring of colonies was an obvious, common and noteworthy phenomenon in these studies. A specific analysis was described of alterations in expression and size of Mycoplasma hyorhinis Vlp [35] and Mycoplasma bovis Vsp proteins [36]. In the former case variation was mediated by alterations in the numbers of repeats in micro- and mini-satellites. In the latter case, variations resulted from juxtaposition of an active promoter to a silent vsp gene by site-specific DNA inversions mediated by a site-specific recombinase, homologous to the E. coli Xer protein, as well as from intragenic recombination between closely related vsp genes.

The occurrence of mutators in clinical populations of bacteria has been investigated by measuring the rate of rifampicin resistance. Fernando Baquero reminded us of this methodology and then described the findings. Mutators of different strengths were observed to occur in both carriage and disease-causing isolates of E. coli, Klebsiella sp., Pseudomonas aeruginosa[37], and Streptococcus pneumoniae[38]. In some cases (e.g. cystic fibrosis patients) mutators persist for many years. Interesting points concern the potential contribution to genetic variation of ‘weak’ mutators, the possibility of an optimum mutation rate for a particular life-style and the implications of mutators for development of antibiotic resistance.

Highly monomorphic bacterial species are difficult to type and this new field of bacterial forensics is the provenance of Gilles Vergnaud. Isolates were separated based on polymorphisms in tandem repeat tracts for Mycobacterium tuberculosis[39], Yersinia pestis, Bacillus anthracis[40] and other species. The validity of this method was tested by construction of phylogenetic trees and comparison to other typing schemes. A database and website (http://minisatellites.u-psud.fr) were developed for analysis of tandem repeats in bacterial genomes and for comparison of results between laboratories.

Clinical isolates of bacteria from similar isolates of a bacterial species are difficult to type reliably and rapidly. Alex van Belkum demonstrated that variations in microsatellites has utility for differentiation of isolates during the course of infections by H. influenzae[41] and Staphylococcus aureus[41]. In some persistent infections alterations in particular loci were observed suggesting adaptation. This analysis has also been applied to Candida albicans isolates from HIV-infected patients and provides an alternative to schemes such as amplified fragment length polymorphism [42].

LPS is the major surface glycolipid of Haemophilus and Neisseria species. Derek Hood presented the molecular analyses that have shown how variations in structural epitopes are key to virulence and commensal behaviour in these bacteria. Genome analysis, construction of defined mutants and biochemical analysis of LPS molecules by both gel electrophoresis and mass spectrometry has identified the functions of many repeat-associated loci of H. influenzae[25,43] and N. meningitidis[44] whose products mediate biosynthesis of LPS epitopes. Use of this knowledge to design biological assays, has resulted in the demonstration that expression of H. influenze loci involved in addition of sialic acid is correlated with survival in in vitro bactericidal assays and with persistence in animal models [45,46]

Helicobacter pylori is responsible for gastritis, ulcers and gastric cancers/lymphomas. Certain H. pylori strains expresses Lewis blood group antigens on their LPS molecules and Ben Appelmelk's presentation discussed the biological roles of these antigens. Phase variation of the Lewis antigens in H. pylori is mediated by C-tracts in genes encoding glycosyltransferases [47]. Expression of these antigens is required for survival of H. pylori in a mouse model of infection and for adherence of H. pylori LPS to stomach tissues and to a dendritic cell marker, DCSIGN [48]. Binding to DCSIGN altered the intracellular cytokine profile, an important modifier of H. pylori infection, thus Lewis antigen phase variation could critically influence host adaptation and persistence of this bacterial species.

This session underlined the wide range of genes and organisms in which localised hypermutation is functional. One focus of the discussions, led by Ben Appelmelk, concerned evolution of this mechanism of bacterial adaptation with genome size, environmental niche and exposure to environmental fluctuation being proposed as selective forces. A general conclusion was that the majority of bacterial microsatellites control expression at a population level of surface determinants resulting in alterations in function and/or escape of immune responses. A unique feature of these discussions concerned the extent to which surface molecules, including LPS and capsules, act as receptors for bacteriophages such that alterations in expression of these molecules may permit escape of a bacteriophage infection. The presence of repeat-associated restriction–modification systems in many genomes is a further indication of the potential importance of bacteriophages in the biology of SSCL.

6 Perspectives on current knowledge and future research avenues

Simple sequence repeats or microsatellites are terms that cover a large range of different DNA sequences that are found at varying frequencies in all DNA based genomes and are associated to differing degrees with functional consequences. These DNA sequences exhibit the unifying feature of being repetitive. As this feature is a major determinant of the mutation rate, comparisons between organisms and repeat sequences can be informative. This workshop provided a forum for exploration of such comparisons.

Misalignment of DNA strands, slipped-strand mispairing, is the simple model that explains the high frequency of mutations in simple sequence repeats. Whilst more complex versions of this model have been proposed to explain particular observations or experimental results, a detailed molecular model of a slippage event is yet to be attained due to uncertainties surrounding many of the molecular determinants of slippage. The roles for example of strand transfer between the polymerisation and proof-reading domains of a DNA polymerase complex or of the structures formed by certain simple sequence repeats in vivo are unclear. DNA replication is usually considered to be the main source of slippage mutations in repeat tracts. An area of interest concerns the effects on slippage of pausing of the replication complex, whether due to particular sequences flanking the repeats or effects of variations in nucleotide concentrations. Recently, evidence has emerged of slippage of simple sequence repeats in non-dividing eukaryotic cells. Complex interactions between MMR, basal excision repair and transcription have been evoked to explain these observations. The importance of a similar phenomenon in nutrient-starved stationary phase bacterial cells has now to be considered. Finally, some evidence of physiological parameters influencing microsatellite mutations rates is emerging, providing further support for the idea that localised hypermutation in bacteria can be regulated by environmental signals.

Localised hypermutation in contingency loci and the occurrence of such loci in multiple copies in diverse species of bacterial pathogens provides a clear indication that mutation rates can have a major influence on bacterial adaptation. This workshop explored this issue in relationship to both SSCL and mutators. Novel observations highlighted in this meeting were of a bimodal distribution of mutation rates in S. pneumoniae isolates [38] and of a high frequency of mutators among epidemic N. meningitidis serotype A isolates [27]. During discussions it was concluded that more epidemiological data is required to address the extent to which mutators are present in natural bacterial populations and are associated with particular infections. Similarly, epidemiological data is required for simple sequence repeats in order to determine if high switching rates in a SSCL or group of loci are correlated with pathogenesis. These studies would be aided by a website that permitted collation and comparison of ‘global’ and localised mutation rates for multiple isolates of different bacterial species. A resource of this type would also eliminate discrepancies in methodologies facilitating comparisons. Whilst the function(s) of many of the genes whose expression is controlled by a microsatellites is known, the effects of variations in the rates of switching of these genes are rarely more than mere speculation. More robust experimental models are required to demonstrate the impact of high mutation rates on bacterial virulence.


The authors thank all the participants of the workshop for their enthusiastic presentations and their help with putting together this review.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
  29. [29].
  30. [30].
  31. [31].
  32. [32].
  33. [33].
  34. [34].
  35. [35].
  36. [36].
  37. [37].
  38. [38].
  39. [39].
  40. [40].
  41. [41].
  42. [42].
  43. [43].
  44. [44].
  45. [45].
  46. [46].
  47. [47].
  48. [48].
  49. [49].
View Abstract