SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Transcriptional activation of the recA gene of Streptococcus pneumoniae was previously shown to occur at competence. A 5.7 kb recA-specific transcript that contained at least two additional genes, cinA and dinF, was identified. We now report the complete characterization of the recA operon and investigation of the role of the competence-specific induction of recA. The 5.7 kb competence-specific recA transcript is shown to include lytA, which encodes the pneumococcal autolysin, a protein previously shown to contribute to virulence of S. pneumoniae. Uncoupling (denoted Ind) of recA and/or the downstream genes was achieved through the placement of transcription terminators within the operon, either upstream or downstream of recA. Prevention of the competence-specific induction of recA severely affected spontaneous transformation. Transformation efficiencies of recA+ (Ind) and of wild-type cells were compared under various conditions and with different donor DNA. Chromosomal transformation was reduced 17- (chromosomal donor) to 45-fold (recombinant plasmid donor), depending on the donor DNA, and plasmid establishment was reduced 129-fold. Measurement of uptake of radioactively labelled donor DNA in transformed cells in parallel with scoring for transformants (chromosomal donor) revealed normal uptake, but a 21-fold reduction in recombination in a recA+ (Ind) strain, indicating that the transformation defect was primarily in recombination. Strikingly enough, a much larger (460-fold) reduction in recombination was observed for the shortest homologous donor fragment used (878 nucleotides long). Possible interpretations of the observation that basal RecA appears unable to promote efficient recombination whatever the number and the length of donor fragments taken up are proposed. The role of recA induction is discussed in view of the potential contribution of transformation to genome plasticity in this pathogen.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Genetic transformation was first discovered in Streptococcus pneumoniae (Griffith, 1928), a human pathogen responsible for various diseases (for review, see Paton et al., 1993). Transformation is one of the simplest modes of genetic exchange and involves uptake of exogenous native donor DNA and subsequent homology-dependent integration into the recipient chromosome. Homologous recombination is strictly dependent upon the presence of the RecA protein (Martin et al., 1995a). In S. pneumoniae, competence for genetic transformation, defined as the ability of bacteria to take up high-molecular-weight exogenous DNA, arises in exponentially growing cultures at a critical cell density. The process is co-ordinated by release of an unmodified heptadecapeptide pheromone, called CSP (for competence-stimulating peptide) (Håvarstein et al., 1995). Previous studies demonstrated that the recA gene was induced at competence (Martin et al., 1995a; Pearce et al., 1995). Western blotting experiments revealed that this induction resulted in a threefold (Martin et al., 1995a) to fivefold (Pearce et al., 1995) increase in the concentration of RecA. Competence-specific induction of the recA gene was also demonstrated in Bacillus subtilis (for review, see Yasbin et al., 1992). This induction was recently shown to be dependent upon ComK, the competence-specific transcription factor of B. subtilis (Haijema et al., 1996). The role of the induction of recA in competent cells has not been investigated in either organism.

recA is the second gene in a competence-induced (cin) operon identified in S. pneumoniae (Martin et al., 1995a). It is preceded by cinA, which has been suggested to encode a recombination accessory protein (Pearce et al., 1995), and followed by dinF (Martin et al., 1995a). The cinA–recA gene arrangement has been conserved in several Gram-negative and Gram-positive bacteria (Martin et al., 1995b), including Escherichia coli (our unpublished observations) and B. subtilis (Albertini et al., 1996), but no competence-specific transcript that would encompass both genes has been reported in the latter. The dinF gene of S. pneumoniae has been only partly sequenced. The 239 residues identified exhibit homology to the N-terminal part of the 459 residue DinF protein of E. coli (Martin et al., 1995a). The E. coli dinF gene is located immediately downstream of the lexA gene and its expression is stimulated by DNA-damaging agents (Krueger et al., 1983).

In this article, we first describe the complete characterization of the cinA–recA operon. We have demonstrated that this operon contained, in addition to dinF, the lytA gene, which encodes the pneumococcal autolysin (Garcia et al., 1986). We have also investigated the importance of the competence-specific induction of recA for transformation. Uncoupling of recA was achieved by placement of transcription terminators in front of the gene and resulted in a strong reduction in recombination. These and previous results give some support to the hypothesis that proposes evolution of transformational recombination as a means of increasing the fitness of a bacterial population.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

The recA competence-specific transcript includes lytA, the gene encoding the pneumococcal amidase

The recA gene of S. pneumoniae was previously shown to be part of a cin operon that included at least two additional genes, cinA and dinF (Martin et al., 1995a). Two recA competence-specific transcripts, a and b (respectively 5.7 kb and 2.55 kb in length), were detected in mRNA extracted from competent cells (Fig. 1 and Fig. 2B, lane 4). As the location of the 3′ extremity of transcript a suggested that it could include at least one additional open reading frame (ORF) downstream of dinF, we decided to characterize this region.

image

Figure 1. . Gene organization of the S. pneumoniae recA chromosomal region. Genes identified in the region are indicated by shaded arrows (each vertical subdivision represents 1 kb). Structures shown at the top correspond to strains harbouring polar insertions (plasmid integration) either upstream (1, left: strains R214, R317 and R322) or downstream (2, right: strains R290 and R318) of the recA gene. Note that plasmid integration generates a partial duplication but does not lead to gene inactivation. Competence- specific transcripts, a and b, identified previously (Martin et al., 1995a) are shown as black arrows. The cat gene confers CmR. The cat-flanking terminators are indicated by ter.

Download figure to PowerPoint

image

Figure 2. . lytA is present on the basal recA transcript and on the competence-specific cinA transcript. A. Hybridization probes (see Experimental procedures) used in Northern blotting are indicated by heavy black bars. Gene organization around the cinA–recA operon is shown for strains R6 (WT) and SBC3. The latter strain, which harbours a polar insertion in front of the lytA gene, was obtained by integration of plasmid pBC into strain R6 (see Experimental procedures). The loop structures represent the lytA terminator (Diaz and Garcia, 1990) and the T1T2 terminators from pBC. The latter terminator is located ≈550 bp downstream of the end of the polylinker in pBC (see Experimental procedures). Transcripts shown as black arrows are labelled by lettering from a to n; e′ indicates the transcript ending at the T1T2 terminator in strain SBC3, whereas l′, which terminates at the same position as e′ and exhibits the same size as l, is not indicated. The competence-inducible promoter, the recA promoter, and the lytA promoter are indicated by Pc, Pb and Pl respectively. B. Northern blot analysis of lytA. About 10 μg of RNA were applied to each lane. Lanes 1, 3, 5 and 7 correspond to RNA extracted from exponentially growing R6 cells; lanes 2 and 4, RNA were prepared from R6 cells made competent by addition of CSP; lanes 6 and 8 correspond to RNA prepared from strain SBC3. Probes used in hybridization experiments are indicated above each panel. The pseudo band corresponding to the size of transcript b in lanes 1–2 and 5–7 does not represent a discrete lytA-specific transcript, but accumulation of breakdown products nearby the 23S RNA band. This is a common observation which has been tentatively attributed to trapping of mRNA species by the abundant ribosomal species (Simpson et al., 1990). As accumulation occurs at the front of the 23S RNA species (data not shown; de Saizieu et al., 1997), we suggest gel occlusion as an alternative explanation.

Download figure to PowerPoint

Two lines of evidence suggested that lytA, which encodes the pneumococcal amidase (Garcia et al., 1986), could be located in the vicinity of dinF. First, recA was localized on the same 72 kb-long SacII generated fragment as lytA (Mortier-Barrière et al., 1997), no. 10 (not no. 9 as erroneously mentioned previously for lytA; Gasc et al., 1991). Second, a tight genetic linkage was detected between two markers, recA::cat (chloramphenicol resistance) and lytA::ermAM (erythromycin resistance) (our unpublished observations). The hypothesis that lytA was located downstream of dinF was checked by a PCR reaction using a primer specific for the 5′-moiety of dinF and a second primer complementary to sequences upstream from the lytA coding region (see Experimental procedures). A 570 bp fragment was obtained, which strongly suggested that dinF and lytA were adjacent. DNA sequencing of this fragment (EMBL/GenBank accession no. Z34303) revealed that the two coding regions were separated by 376 nucleotides (Fig. 1) and that the dinF gene (positions 2799 … 4169 in Z34303) potentially encoded a 456 residue protein (Z34303; data not shown). Twelve potential membrane-spanning α-helices were predicted in S. pneumoniae DinF by the method of Eisenberg et al. (1984). DinF exhibited 19% identity with E. coli DinF (Krueger et al., 1983). TBLASTN and BLASTP searches identified proteins with significant degrees of similarity (25–19% identity) to S. pneumoniae DinF in B. subtilis (YpnP; SWISS-PROT P54181), Methanococcus jannaschii (PIR E64388G), Pyrococcus sp. (Bouyoub et al., 1995) and Helicobacter pylori (PID G2314344). S. pneumoniae DinF and B. subtilis YpnP exhibited 25% identity (data not shown), and transmembrane segments have been predicted to be present in the latter at similar positions.

A transcription termination site was previously mapped 72 nucleotides downstream from the translational stop codon of lytA, near the 3′ end of a potential hairpin in the mRNA with a free energy (ΔG) of −20.8 kcal mol−1 (Diaz and Garcia, 1990). A transcript originating in front of cinA and ending at this terminator would be 5.57 kb long, which fits very well with the size estimated previously for the cinA–recA competence-specific transcript (Martin et al., 1995a). This suggested that lytA could be induced at competence. To check this hypothesis, transcriptional analysis of the region was conducted using two lytA probes, L1 and L2, which cover the 5′ end and the 3′ end of the gene, respectively, and a recA probe, R (Fig. 2A; see Experimental procedures). With RNA prepared from exponentially growing wild-type cells, four transcripts (e, l, m and n) were detected with probe L1 (Fig. 2B, lanes 1 and 5) and L2 (Fig. 2B, lane 7). Transcript e (4.3 kb) was also detected with the recA-specific probe, which suggested that this transcript could originate from the previously identified recA promoter, Pb (Martin et al., 1995a). Analysis of mRNA extracted from strain SBC3 in which a transcription terminator was placed in the middle of the lytA gene (Fig. 2A) confirmed this interpretation because, as expected, transcript e was no longer detected with probe L2 (Fig. 2B, lane 8). Transcript l (1.3 kb), which was detected with both L1 and L2 probes (Fig. 2B, lanes 1 and 5–8), corresponded to the 1267-nucleotide-long lytA-specific transcript identified previously (Diaz and Garcia, 1990). Two additional transcripts (m and n) of ≈1.1 and 0.8 kb were obtained with these two probes. They could correspond to stable degradation products of the 1.3 kb lytA transcript l.

With RNA extracted from competent cells, an additional transcript, a (5.7 kb), was detected with the lytA specific probe, and this appeared strongly induced, whereas the level of transcripts e and l remained unchanged (Fig. 2B, lane 2). Transcript a was previously shown to start immediately upstream of cinA (Martin et al., 1995a). The second competence-specific transcript, b (Fig. 2B, lane 4), encoded only cinA and recA (Martin et al., 1995a).

Altogether, these results demonstrated that the recA operon includes lytA and that transcription of the autolysin gene is strongly induced in competent cells from Pc, the competence-inducible promoter located in front of cinA (transcript a). In non-competent cells, lytA is expressed from Pb, the recA promoter (transcript e) and from Pl, a promoter located immediately upstream of lytA (transcript l ).

Prevention of the competence-specific induction of recA affects spontaneous transformation

Strain R214 was constructed in a previous work to map more precisely the recA competence-specific transcripts (Martin et al., 1995a). This strain was generated by chromosomal integration of plasmid pR322, which carried a DNA fragment overlapping the cinA–recA intergenic region, and harboured the cat cassette immediately downstream of the cinA gene (Fig. 1). Plasmid integration interrupted neither the cinA nor the recA genes, but transcription termination occurred at ter-down and ter-up, the cat-flanking terminators, which prevented the competence-specific induction of recA and of the downstream genes (Martin et al., 1995a; data not shown). This genotype will be designated recA+ (Ind) thereafter. Western blot analysis demonstrated that the RecA protein concentration remained unchanged in recA+ (Ind) competent cells, whereas a large increase in CinA concentration was observed (Fig. 3). In the wild-type parent, increased transcription of the cin operon at competence correlated with increased CinA and RecA concentrations. Densitometric analyses of signal intensities carried out using CinA as an internal standard revealed a fourfold difference in RecA concentration between wild-type and recA+ (Ind) competent cells. Thus, recA+ (Ind) strains offered the unique opportunity to investigate whether induction of recA was essential for full transformation proficiency.

image

Figure 3. . Western blot analysis of RecA in competent and non-competent cells of wild-type S. pneumoniae (R310) and of strains harbouring polar insertions upstream (R317) or downstream (R318) of the recA gene. Precompetent cells were treated (or not) with CSP. Samples corresponding to 3 × 108 cells were subjected to SDS–PAGE (left), followed by immunoblot transfer (right) with a mixture of E. coli RecA and of S. pneumoniae CinA (our unpublished results) antibodies as described previously (Martin et al., 1995a). The RecA and CinA proteins are indicated by filled and open arrows respectively. Protein molecular weight standards (centre: 97.4, 68, 43, 29 and 18.4 kDa) were from Gibco BRL.

Download figure to PowerPoint

Comparison of the spontaneous competence profile of a recA+ (Ind) strain, R317, with that of its wild-type parent, R310, showed that the former was severely affected for transformation (Fig. 4). Maximum transformation frequency in R317 was only 6% of wt (i.e. 16-fold reduction). Addition of CSP did not change the competence profile of the recA+ (Ind) strain (Fig. 4, inset), which suggested that regulation of competence was not affected in this strain.

image

Figure 4. . Competence profiles of wild-type S. pneumoniae (R310) and of strains harbouring polar insertions upstream (R317) or downstream (R318) of the recA gene. The percentage of SmR transformants was determined at 15 min intervals during logarithmic growth (usually from OD550 = 0.1–1) in C+Y medium for the three strains. CSP (25 ng ml−1) was added after 95 min incubation in the indicated R317 culture. OD550: ♦, R310; ▴, R317; ▪, R317 treated with CSP; •, R318. Transformants: ×, R310; +, R317; □, R317 treated with CSP; ○, R318. The inset shows R317 transformants using a 15-fold expanded scale. Data correspond to a single experiment (four cultures run in parallel) representative of several independent experiments. Strains R214 and R290 behaved similarly to strains R317 and R318 respectively (data not shown).

Download figure to PowerPoint

The transformation deficiency of recA+ (Ind) strains could be attributed to the uncoupling of any of the three genes located downstream of the cat-flanking terminators, i.e. recA, dinF and/or lytA. To determine which gene was specifically required for full competence development, we constructed strain R290 by transformation of strain R800 with pR361, a plasmid carrying a DNA fragment that overlapped the recAdinF intergenic region (see Experimental procedures). This construct was identical to that in strains R214-R317, except for the location of the cat cassette, which was placed immediately downstream of the recA gene (Fig. 1). Integration of plasmid pR361 interrupted neither recA nor dinF, but transcription termination occurring at ter-down and ter-up, the cat-flanking terminators, rendered the corresponding strains R290 or R318, dinF+ (Ind) and lytA+ (Ind). As expected, a fourfold increase in RecA protein concentration was observed in R318-competent cells (Fig. 3). The competence profile of strain R318 appeared indistinguishable from that of R310 (Fig. 4). We concluded that the uncoupling of recA from the competence-inducible promoter Pc was responsible for the reduction in transformation observed with strains R214–R317.

Plasmid establishment defect associated with uncoupling of recA expression

As plasmid establishment was previously shown to be blocked in a recA mutant, which suggested the involvement of RecA in the process (Martin et al., 1995a), we investigated the kinetics of dose–response for establishment of plasmid pLS1 in recA+ (Ind) (R214) and recA+ (Ind+) (R800) strains. Plasmid pLS1 is a pMV158 derivative that is autonomously maintained in S. pneumoniae at about 25 copies per cell and replicates by a rolling-circle mechanism (Puyet et al., 1988). At the lowest DNA concentration tested, plasmid establishment in recA+ (Ind) cells was reduced 129-fold relative to the wild-type value (Fig. 5A).

image

Figure 5. . Yield of transformants as a function of donor DNA concentration in the recA+ (Ind) strain R214 and in its wild-type parent R800. Competent cells at 1.5 × 108 cfc ml−1 were obtained by CSP treatment as indicated in Experimental procedures. The concentration of donor DNA added to the culture was varied as indicated. Cells were exposed to DNA for 1 min (R304 chromosomal DNA) or 20 min (pLS1 and pR161 donor DNA). A. pLS1 plasmid DNA in the form of CC monomers. B. R304 chromosomal DNA. C. pR161 plasmid DNA. Scoring was for TcR (pLS1 DNA), SmR (R304 chromosomal DNA), and MTXR (pR161 DNA) transformants. ♦, R800; ▪, R214.

Download figure to PowerPoint

Reduction of the number and length of donor DNA fragments taken up does not restore full recombination proficiency under low RecA conditions

It is well known that under saturating conditions individual recipient cells take up several donor fragments that are likely to compete for RecA. To investigate whether a reduction in the amount of DNA taken up per cell could improve transformation efficiency in the recA+ (Ind) background, dose–response curves were established in parallel for a recA+ (Ind) strain (R214) and its wild-type parent (R800) using chromosomal DNA as donor (Fig. 5B). Even at the lowest DNA concentration tested (1 ng ml−1), R214 yielded 17-fold fewer transformants than wild type. As the lowest concentration corresponded to about 0.1 molecule per cell in the transformed culture (assuming that donor chromosomal DNA fragments are about 40 kb in length, as routinely obtained by normal isolation in the laboratory), these data suggested that the amount of RecA in R214 cells was not high enough to process about 40 kb of donor DNA.

To further reduce the length of recombining donor fragments, we used as donor pR161 plasmid DNA. This recombinant plasmid harbours a 1894 bp pneumococcal DNA fragment from the ami locus with the ami142 mutation that confers resistance to methotrexate (Alloing et al., 1989). With this donor DNA, R214 cells also appeared transformation deficient (45-fold reduction), even at the lowest DNA concentration tested, which corresponded to about 0.3 molecules per cell (Fig. 5C). This indicated that R214 cells could not process 7.2 kb chimeric DNA (the length of pR161) to recombine a 1.9 kb segment homologous with the recipient chromosome as efficiently as wild-type cells.

Recombination defect in recA+ (Ind ) cells

Altogether, our observations suggested that RecA protein concentration could be limiting, and that the reduction in chromosomal transformation or in plasmid establishment in the recA+ (Ind) genetic background could result from a recombination defect. However, none of the experiments described above really assessed the recombination rate as DNA uptake was not measured. We therefore compared the recombination rate of recA+ (Ind) cells (R322) to that of wild-type cells (R246) by measuring uptake of 32P uniformly labelled donor DNA in parallel with scoring for transformants with chromosomal donor DNA (Fig. 6). Although similar amounts of DNA were taken up in both strains (Fig. 6 and data not shown), recombination was reduced 21-fold in the recA+ (Ind) strain, a value in good agreement with that measured in 5Fig. 5B. We concluded that the lack of induction of recA gene expression at competence in the latter strain resulted in chromosomal recombination defect.

image

Figure 6. . Measurement of recombination rate in the recA+ (Ind) strain (R322) and in its wild-type parent (R246). Precompetent cells (1.2 × 108 cfc ml−1 for R322 and 1.7 × 108 cfc ml−1 for R246) were activated with CSP for 10 min at 37°C, followed by trypsin treatment (2 μg ml−1) for 8 min at 37°C to remove CSP. For uptake measurement (tube A), activated cells were incubated at 30°C with a radioactively labelled PCR fragment carrying the str41 marker (3650 cpm ng−1; 230 ng ml−1) (see Experimental procedures). In parallel (tube B), activated cells were incubated at 30°C with R304 chromosomal DNA (0.9 μg ml−1). At time intervals (0.5, 1.5, 3, 6, 12 and 18 min), 100 μl aliquots from tube A were treated with DNase I (2 μg ml−1) for 10 min at 30°C, centrifuged, washed and counted after cell lysis, and 100 μl aliquots from tube B were similarly treated with DNase I and used for scoring of SmR transformants. SmR transformants were also scored from aliquots of tube A taken after 12 and 18 min incubation. Transformants, left axis scale: ▴, R246 recipient; and •, R322 recipient. Right axis scale: ○, R322 recipient. Note the 20-fold difference between both scales that results in near superimposing of R246 and R322 curves. Data correspond to a single experiment (both strains run in parallel) representative of several independent experiments.

Download figure to PowerPoint

In the same experiment, we also measured transformation yields using as donor a PCR-generated short fragment (878 bp), completely homologous to the recipient chromosome (Fig. 6 and Experimental procedures). Recombination rates in the recA+ (Ind) cells were 0.19–0.25% of wt, i.e. c. 460-fold reduction (12 and 18 min uptake points in Fig. 6; data not shown). These results indicated that processing of short homologous fragments was strongly affected in cells containing only a basal (uninduced) level of RecA protein. A similar recombination defect was also observed under conditions allowing uptake of only a single short fragment per cell (data not shown).

Each individual cell in recA+ (Ind ) cultures can process two independent markers

All the experiments described above indicated a decreased recombination rate in recA+ (Ind) cultures. However, RecA concentration could be nearly identical in all cells or could vary from cell to cell. In the first case, a decreased recombination rate would indicate reduced recombination proficiency of all individuals in a culture. In the latter case, recA+ (Ind) cultures would be mixed and a fraction recombining at near wild-type rate may account for most of the recombinants formed.

To distinguish between the two possibilities, single and double transformants for a pair of independent markers were scored in wild-type (R246) and recA+ (Ind) (R322) cultures. The two markers used, str41 and rif23 (rifampicin resistance), were genetically independent (map locations: 11–12.5′ and 21–24′ respectively; Gasc et al., 1991). Transformant yields for each marker were about 20-fold lower in the recA+ (Ind) strain (0.4% transformants) than in the wild-type strain (7% transformants) (data not shown). The yield of double transformants observed for both strains was in good agreement with the expected frequency calculated for each strain as the product of individual marker frequencies measured in the corresponding genetic background, i.e. on the implicit assumption that all cells in a culture were transformed with the same efficiency (data not shown). We concluded that each individual competent cell in a recA+ (Ind) culture recombined at a similar albeit reduced rate. The data also indicated that, although recA+ (Ind) cells processed donor DNA with a reduced efficiency, they were able to process a second independent fragment at the same reduced rate.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

The cinA–recA operon includes a putative SOS repair gene and the lytA gene

The first and to date only cin operon identified in S. pneumoniae contains four genes, cinA, recA, dinF and lytA. Homologues of DinF have been detected in both Gram-negative and Gram-positive bacteria, as well as in archaebacteria (Bouyoub et al., 1995). Although such a conservation during evolution suggests that DinF proteins fulfil an important physiological role, this role remains unknown. The existence of a tight linkage between dinF and either the lexA or the recA gene in phylogenetically distant organisms suggests that co-expression of these genes is functionally important. The co-induction of dinF with recA at competence also suggests that regulation of competence and of the SOS DNA repair system could overlap in S. pneumoniae. This overlap is further suggested by the observation of a RecA-dependent prophage induction after DNA uptake in transformation (Martin et al., 1995a). The existence of a regulatory link between the competence network and the SOS regulon has been firmly established in B. subtilis (Yasbin et al., 1992; Haijema et al., 1996).

The presence of lytA, which encodes the pneumococcal autolysin within a cin operon, was not totally unexpected as a correlation between autolytic activity and competence has been established in S. pneumoniae (Seto and Tomasz, 1975), as well as in two other transformable bacteria, B. subtilis and Streptococcus gordonii (for review, see Dijkstra and Keck, 1996). In Neisseria gonorrhoea, the tpc mutant was found to affect competence and cell division leading to the hypothesis that Tpc is a peptidoglycan hydrolase (Fussenegger et al., 1996). Controlled autolysis has been suggested to be responsible for the release of DNA, thus favouring horizontal exchange (Fussenegger et al., 1996). It has also been proposed that the autolytic activity might be involved in transient remodelling of the peptidoglycan layer to allow DNA uptake (Dijkstra and Keck, 1996; Fussenegger et al., 1996). However, no evidence exists in favour of either hypotheses. Our results show that a lytA+ (Ind) strain behaves normally for transformation, which suggests that the competence-specific induction of lytA is not essential for this process. In addition, S. pneumoniae strains harbouring different lytA mutations (Lopez et al., 1986; Tomasz et al., 1988) and the lyt mutants of B. subtilis retain full transformability (Fein and Rogers, 1976). Therefore, a full understanding of the role of lytA induction in competent S. pneumoniae awaits further investigation. On the other hand, LytA has been demonstrated to contribute to virulence of S. pneumoniae (Berry et al., 1989; Canvin et al., 1995). It is tempting to speculate about the possible regulation of virulence and the existence of a regulatory link between virulence and competence. Further studies will reveal whether competence and virulence represent alternative responses of this organism to changes in environmental conditions.

Based on the location of the cinA gene (called exp10) in the same operon as recA, a gene essential for recombination, and on the observation that cinA mutants transformed 2 log units less efficiently than the parent strain, Pearce et al. (1995) surmised a role for CinA in recombination. However, the presence of dinF and of lytA within the same operon weakens the first argument. Moreover, the use by Pearce et al. (1995) of plasmid insertion to generate cinA mutations precludes conclusions regarding the effect of the mutations as plasmid insertions are likely to be polar on downstream gene expression (including recA), and we have shown here that uncoupling of recA leads to a 21-fold decrease in recombination. Experiments using non-polar cinA mutations are in progress to investigate the function of CinA in transformation.

Induction of recA is required for full recombination proficiency

The importance of the competence-specific induction of recA for transformation is assessed by the 16-fold reduction in spontaneous transformation observed in recA+ (Ind) strains. CSP treatment had no effect on transformation efficiency, which ruled out the hypothesis that uncoupling affected some early steps in the regulation of competence. Measurement of uptake of radioactively labelled DNA demonstrated that uncoupling of recA resulted in a 21-fold defect in chromosomal recombination. Interestingly, it has been shown very recently that the number of conjugational recombinants in E. coli depended on RecA concentration: a fourfold reduction in RecA resulted in a c. 20-fold defect in recombination (Fig. 1 in Boudsocq et al., 1997). The reduced concentration of RecA in recA+ (Ind)-competent cells had even more drastic effects on plasmid establishment than on chromosomal transformation. The fact that plasmid establishment may require interaction of RecA with two independently imported single-stranded molecules for assembly of a double-stranded replicon (Martin et al., 1995a), whereas chromosomal recombination involves a unique donor fragment may account for the difference. Most striking was the observation that reduction of the length of the donor down to 878 nucleotides resulted in a c. 460-fold recombination defect in recA+ (Ind) cells, even under conditions allowing uptake of only one donor fragment per cell.

The last observation deserves further discussion. A calculation based on the assumption that a nucleoprotein complex contains one RecA monomer for every three nucleotides (Kowalczykowski et al., 1994) indicates that about 290 molecules of RecA would be required at the initial presynaptic step to cover entirely the 878-nucleotide-long donor fragment. The number of RecA molecules in recA+ (Ind) cells should therefore be lower than this value. Moreover, as RecA concentration increased only fourfold at competence (Fig. 3), the induced level of RecA would be lower than 1160 molecules per cell. These figures appear very low compared with 9000 molecules and 80 000 molecules, the values reported for E. coli RecA basal and induced levels respectively (Boudsocq et al., 1997 and references therein). In B. subtilis, non-competent and competent cells have been reported to contain, respectively, 4300 molecules and 66 000 molecules (Lovett et al., 1989). Although values in the latter are subject to uncertainty and could be off by a factor of two (Lovett, et al., 1989), they are very similar to those in E. coli. It is, therefore, rather unlikely that S. pneumoniae cells contain so few RecA molecules. Moreover, it would be difficult to explain how such a limited number of RecA molecules permits the efficient processing of up to 100–150 kb DNA per cell (an estimate based on transformation frequencies for single markers — a few percent — and on the size of the pneumococcal chromosome −2.27 Mb). Alternatively, the RecA content of S. pneumoniae could be similar to that in other bacteria, but RecA molecules present before induction of competence could be inefficient for recombination, either because they are sequestered or because the protein is modified and catalyses recombination only poorly. Synthesis of new RecA molecules would then be absolutely required for full recombination proficiency. However, this hypothesis could not by itself account for the drastic reduction of recombination observed with the 878 bp donor fragment.

An alternative explanation to the reduced recombination rate observed in recA+ (Ind) cells would be that competition occurs between RecA and some other protein(s). Depending on the amount of competitor protein, recombination of a short donor fragment could be more drastically affected than recombination of large chromosomal segments. A good candidate as RecA competitor would be an SSB-like protein as it is known that the binding of E. coli SSB and RecA to ssDNA in vitro is competitive (Kowalczykowski et al., 1994), and it has been shown that an SSB-like protein accumulates in S. pneumoniae competent cells (Morrison et al., 1979). In recA+ (Ind) cells, uncoupling of recA expression would change the SSB/RecA balance that could in turn inhibit recombination. It would be interesting to check the effect of the simultaneous uncoupling of recA and ssb on recombination of a short homologous fragment. Alternatively, or in addition, competition occurring between recombination and nucleolytic degradation of incoming DNA (Lataste et al., 1981) could be biased toward degradation in recA+ (Ind) cells. Short donor fragments would obviously be more sensitive to such a bias than larger pieces of chromosomal donor.

Although induction of recA in competent cells has been observed in both B. subtilis (Yasbin et al., 1992) and S. pneumoniae (Martin et al., 1995a), this is the first demonstration that induction of recA is required for full recombination proficiency in a transformable bacteria. Induction of recA has also been reported in E. coli during intraspecies conjugation (Matic et al., 1995). However, induction was weak and subsequent to mating, whereas in S. pneumoniae induction of recA does not depend on the presence of incoming single strands but clearly precedes DNA uptake (Martin et al., 1995a).

Biological significance of transformation: genome plasticity?

Several hypotheses have been advanced to explain the evolutionary origin of transformation, including a role in nutrient acquisition, in repair of chromosome damage and in genome evolution (Dubnau, 1991 and references therein). It is difficult to understand why the recA gene of S. pneumoniae and of B. subtilis should be induced at competence for nutrient acquisition. On the other hand, induction of rec genes would be consistent with both the repair and the genome evolution hypotheses. However, the observation of a co-ordination of competence development, particularly in S. pneumoniae cultures, is difficult to fit in with the repair hypothesis. Co-ordination would be most consistent with the hypothesis that transformational recombination has evolved as a means of increasing fitness of a bacterial population by facilitating the acquisition of new genetic traits. As uptake is not sequence-specific, any DNA present in the environment could be taken up. Therefore, uptake of DNA from related species present in the same ecological niche (for example oral streptococci) followed by recombination into the chromosome can provide the S. pneumoniae genome with enhanced plasticity.

Whereas mutations are frequently neutral or deleterious, transformation is expected to replace genes or gene fragments by sequences that were functional in the original species. In this view, transformation would be an extremely powerful mechanism for rapid evolution as each individual cell in a culture can experiment with new combinations of genes; each cell may randomly take up several independent fragments (up to 1–5% of the genome) and the probability of integrating these fragments varies from close to 1 for homologous DNA to 0.1–0.5 for 2–10% diverged DNA (Humbert et al., 1995). Possible examples of replacement of pneumococcal genes by sequences from related species have been discussed (for example penicillin resistance, optochin resistance, capsular type; Mortier-Barrière et al., 1997). As only a few new combinations of genes are expected to be beneficial, co-ordination of competence development through CSP is essential to ensure that a sufficient number of cells of the species is present to experience the process.

Elements favouring genome evolution by transformation in S. pneumoniae

By increasing the recombination rate, the induction of recA during competence development in S. pneumoniae potentially contributes to genome plasticity. On the other hand, two efficient means of preventing interspecies recombination have proven inefficient in this organism. First, restriction of foreign DNA does not occur in transformation, despite the presence of the powerful DpnI–DpnII restriction system, because DNA is taken up as single strands that are refractory to digestion (Cerritelli et al., 1989). Second, the hypothesis that genomic sequence divergence constitutes a major limitation to gene transfer between bacterial species (Matic et al., 1996) does not apply to S. pneumoniae. Whereas recombination between E. coli and Salmonella typhimurium is reduced as much as 1000-fold in mismatch repair proficient cells (Rayssiguier et al., 1989; Matic et al., 1995), in S. pneumoniae the Hex mismatch repair system, which reduces transformation frequencies for point mutations in homologous DNA (Claverys and Lacks, 1986), is unable to prevent interspecies transformation (Humbert et al., 1995). Hex is easily saturated by excess mismatches (Guild and Shoemaker, 1974; Humbert et al., 1995). Moreover, the finding that two hex genes are not induced at competence, whereas an increase in cellular concentration of either HexA or HexB would increase repair ability and prevent saturation by excess mismatches (Humbert et al., 1995), is also consistent with the genome evolution hypothesis. It strongly suggests that the Hex system has not been tuned to cope with excess mismatches and, therefore, to abort interspecies recombination. Finally, S. pneumoniae exhibits a wide variety of recombination mechanisms favouring integration of heterologous DNA (for review, see Mortier-Barrière et al., 1997).

Altogether, these observations suggest that in S. pneumoniae, and possibly in other transformable bacteria as well, conditions have been evolved to maximize intra- and intergeneric genetic exchanges during transformation. In the absence of SOS mutagenesis in S. pneumoniae (Gasc et al., 1980; Martin et al., 1995a), transformation could provide most of the genetic variability for adaptation of this pathogen to changes in environmental conditions. S. pneumoniae seems, therefore, to have evolved strategies that differ completely from that of E. coli (Matic et al., 1995) with respect to the role of recombination, SOS and mismatch repair systems in the evolution of the species.

Experimental procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Bacterial strains, growth conditions and competence

The bacterial strains and plasmids used are listed in Table 1. Cultures of S. pneumoniae were as described previously (Alloing et al., 1996), with the exception of strains R6 and SBC3 that were routinely grown in Todd–Hewitt broth. Competence profiles, which assess spontaneous transformability (Fig. 4), were generated from cultures grown in C+Y medium, as previously described (Alloing et al., 1996). For CSP-induced competence, precompetent cells prepared in CTM 1, as previously described (Martin et al., 1995a), were incubated with synthetic CSP (25 ng ml−1) in CTM 2 (Martin et al., 1995a) at 37°C for 10–15 min, before addition of DNA. Cells were incubated at 30°C during DNA uptake. Transformants were determined by plating in 10 ml of D medium (Alloing et al., 1996), followed by challenge with a 10 ml of overlay containing the appropriate antibiotic after phenotypic expression for 120 min at 37°C. E. coli strains were grown in LB medium. Antibiotic concentrations used for selection of transformants were: chloramphenicol (Cm), 4.5 μg ml−1; erythromycin (Ery), 2 μg ml−1; methotrexate (MTX), 3 × 10−6 M; rifampicin (Rif), 2 μg ml−1; spectinomycin (Spc), 200 μg ml−1; and streptomycin (Sm), 200 μg ml−1 for S. pneumoniae; ampicillin (Ap), 50 μg ml−1; and Cm, 10 μg ml−1 for E. coli.

Table 1. . Bacterial strains, plasmids and oligonucleotides primers used in this study. pR plasmids replicate autonomously in E. coli only.R/S, resistant/sensitive.a. Small letters in oligonucleotide sequences indicate changes introduced for creation of convenient restriction sites (XhoI and BamHI underlined in lytA-a and lytA-d respectively).Thumbnail image of

Northern blot analysis

S. pneumoniae cells were collected by centrifugation and stored frozen in liquid nitrogen. The cell pellet was resuspended into 3 ml of prewarmed (60°C) acid–phenol. After a 5 min. incubation, at 60°C, 3 ml of NAES buffer (NAES: 50 mM sodium acetate pH 5.1; 10 mM EDTA; 1% SDS) was added and the suspension incubated for an additional 5 min. at 60°C and mixed by repeated inversions. After cooling on ice the phases were separated by centrifugation and an additional two acid–phenol extractions were performed before ethanol precipitation. The resulting RNA were then treated for 20 min with RNase-free DNase and precipitated again. We routinely obtained about 500 μg of total RNA from a 50 ml culture grown to an OD600 of 0.5. Total cellular RNA (10 μg per lane) was electrophoresed on a formaldehyde gel and transferred to a nylon filter (Gene-screen). The RNA size standard was purchased from Gibco. [α-33P]-UTP antisense probes were produced by in vitro transcription from fragments cloned in pBluescript SK+. Probe L1 covers the 5′ end of lytA, from position 781–1020, whereas probe L2 covers the 3′ end of lytA from position 1250–1450. These numberings refer to the EMBL sequence M55415. Probe R covers the 5′ end of recA, from position 1800–2325 according to numbering of the EMBL sequence Z34303.

Pictures of the Northern blot experiments shown in Fig. 2 were generated using a Molecular Dynamics PhosphoImager, and processed using the NIH IMAGE program.

Plasmid and strain constructions

To generate the lytA plasmid pBC, pJDC9 (Chen and Morrison, 1987) was first digested with NdeI and religated to itself after a filling-in reaction with Klenow, leading to pAS0. pAS0 was then digested with HindIII and EcoRI flanking the polylinker and religated with the new polylinker sequence 5′-agctCCTCGAGGTCTAGAATTCAGGTAGTGTACCCATATGCCCGGGAAGCTTGGATCCTGCAGAGCTCaatt-3′ (small letters in the nucleotide sequence indicate the presence of HindIII and EcoRI adapters on one or the other strand) to generate plasmid pAS1. The PCR fragment generated by primers lytA-a and lytA-d (Table 1) containing the lytA promoter region and the first 540 bp of lytA was ligated into XhoI–BamHI-digested plasmid pAS1 to create plasmid pBC. Plasmid pBC was then transformed into S. pneumoniae strain R6 to generate strain SBC3 and correct integration was confirmed by appropriate PCR.

Plasmid pR361 was constructed by cloning the PCR fragment generated with primers IM30 (internal to recA) and IM39 (recA–dinF intergenic region) (Table 1) into the EcoRV site of plasmid pR315. Plasmid pR315 was then transformed into S. pneumoniae strain R800 to generate strain R290 and correct integration was confirmed by Southern hybridization analysis.

DNA manipulations and sequencing

Methods for preparing S. pneumoniae chromosomal DNA and for obtaining plasmid DNA from E. coli or from S. pneumoniae were those described previously (Martin et al., 1995a). PCR products were amplified using Pfu DNA polymerase (Stratagene) with 25 cycles and a primer annealing temperature depending on the primer pair. The PCR product amplified from chromosomal DNA using primers BM31 (internal to dinF ) and IM53 (lytA upstream region) (Table 1) was directly sequenced on both strands using a CircumVent Thermal Cycle dideoxy DNA sequencing kit (New England Biolabs) and BM31 or IM53 as primers. The PCR product (878 bp) containing the str41 SmR marker was amplified from R304 chromosomal DNA using primers IM51 and IM52 (Table 1). For measurement of DNA uptake, DNA was radioactively labelled by amplification in the presence of 50 μCi of [α-32P]-dATP (3000 Ci mmol−1) and the four dNTPs (0.2 mM) in a total volume of 100 μl.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

We thank Patrick Caspers, Martin Stieger and Dietrich Stüber for helpful discussions. We acknowledge the excellent technical assistance of Sabine Demmak and of Marie-Chantal Granadel.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  • 1
    Albertini, A., Caldwell, B., Caramori, T., Errington, J., Foulger, D., Thomaides, H., and Williams, A. (1996) The genetic organization of the recA chromosomal region in Bacillus subtilis. Published DNA sequence with accession number U56238.
  • 2
    Alloing, G., Trombe, M.C., and Claverys, J.P. (1990) The ami locus of the Gram-positive bacterium Streptococcus pneumoniae is similar to binding protein-dependent transport operons of Gram-negative bacteria. Mol Microbiol 27: 633644.
  • 3
    Alloing, G., Granadel, C., Morrison, D.A., and Claverys, J.P. (1996) Competence pheromone, oligopeptide permease and induction of competence in Streptococcus pneumoniae. Mol Microbiol 27: 471478.
  • 4
    Avery, O.T., MacLeod, C.M., and MacCarty, M. (1944) Studies on the chemical nature of the substance inducing transformation of pneumococcal types. I. Induction of transformation by a deoxyribonucleic acid fraction isolated from pneumococcus type III. J Exp Med 27: 137158.
  • 5
    Berry, A.M., Lock, R.A., Hansman, D., and Paton, J.C. (1989) Contribution of autolysin to virulence of Streptococcus pneumoniae. Infect Immun 27: 23242330.
  • 6
    Boudsocq, F., Campbell, M., Devoret, R., and Bailone, A. (1997) Quantitation of the inhibition of Hfr × F recombination by the mutagenesis complex UmuD′C. J Mol Biol 27: 201211.
  • 7
    Bouyoub, A., Barbier, G., Querellou, J., and Forterre, P. (1995) A putative SOS repair gene (dinF-like) in a hyperthermophilic archaeon. Gene 27: 147149.
  • 8
    Canvin, J.R., Marvin, A.P., Sivakumaran, M., Paton, J.C., Boulnois, G.J., Andrew, P.W., and Mitchell, T.J. (1995) The role of pneumolysin and autolysin in the pathology of pneumonia and septicemia in mice infected with a type 2 pneumococcus. J Infect Dis 27: 119123.
  • 9
    Cerritelli, S., Springhorn, S.S., and Lacks, S.A. (1989) DpnA, a methylase for single-strand DNA in the Dpn II restriction system, and its biological function. Proc Natl Acad Sci USA 27: 92239227.
  • 10
    Chen, J.-D., and Morrison, D.A. (1987) Cloning of Streptococcus pneumoniae DNA fragments in Escherichia coli requires vector protected by strong transcriptional terminators. Gene 27: 179187.
  • 11
    Chen, J.-D., and Morrison, D.A. (1988) Construction and properties of a new insertion vector, pJDC9, that is protected by transcriptional terminators and useful for cloning of DNA from Streptococcus pneumoniae. Gene 27: 155164.
  • 12
    Claverys, J.P., and Lacks, S.A. (1986) Heteroduplex deoxyribonucleic acid base mismatch repair in bacteria. Microbiol Rev 27: 133165.
  • 13
    De Saizieu, A., Vankan, P., Vockler, C., and Van Loon, A.P.G.M. (1997) The trp RNA-binding attenuation protein (TRAP) regulates the steady-state levels of transcripts of the Bacillus subtilis folate operon. Microbiology 27: 979989.
  • 14
    Diaz, E., and Garcia, J.L. (1990) Characterization of the transcription unit encoding the major pneumococcal autolysin. Gene 27: 157162.
  • 15
    Dijkstra, A.J., and Keck, W. (1996) Peptidoglycan as a barrier to transenvelope transport. J Bacteriol 27: 55555562.
  • 16
    Dintilhac, A., Alloing, G., Granadel, C., and Claverys, J.P. (1997) Competence and virulence of Streptococcus pneumoniae: adc and psaA mutants exhibit a requirement for Zn and Mn resulting from inactivation of putative ABC metal permeases. Mol Microbiol 27: 727739.
  • 17
    Dubnau, D. (1991) Genetic competence in Bacillus subtilis. Microbiol Rev 27: 395424.
  • 18
    Eisenberg, D., Schwarz, E., Komaromy, M., and Wall, R. (1984) Analysis of membrane and surface protein sequences with the hydrophobic moment plot. J Mol Biol 27: 125142.
  • 19
    Fein, J.E., and Rogers, H.J. (1976) Autolytic enzyme-deficient mutants of Bacillus subtilis. J Bacteriol 27: 14271442.
  • 20
    Fussenegger, M., Kahrs, A.F., Facius, D., and Meyer, T.F. (1996) Tetrapac (tpc), a novel genotype of Neisseria gonorrhoeae affecting epithelial cell invasion, natural transformation competence and cell separation. Mol Microbiol 27: 13571372.
  • 21
    Garcia, P., Garcia, J.L., Garcia, E., and Lopez, R. (1986) Nucleotide sequence and expression of the pneumococcal autolysin gene from its own promoter in Escherichia coli. Gene 27: 265272.
  • 22
    Gasc, A.M., Sicard, N., Claverys, J.P., and Sicard, A.M. (1980) Lack of SOS repair in Streptococcus pneumoniae. Mutat Res 27: 157165.
  • 23
    Gasc, A.M., Kauc, L., Baraillé, P., Sicard, M., and Goodgal, S. (1991) Gene localization, size, and physical map of the chromosome of Streptococcus pneumoniae. J Bacteriol 27: 73617367.
  • 24
    Griffith, F. (1928) The significance of pneumococcal types. J Hyg 27: 113159.
  • 25
    Guild, W.R., and Shoemaker, N.B. (1974) Intracellular competition for a mismatch recognition system and marker specific rescue of transforming DNA from inactivation by ultraviolet irradiation. Mol Gen Genet 27: 291300.
  • 26
    Haijema, B.J., Van Sinderen, D., Winterling, K., Kooistra, J., Venema, G., and Hamoen, L.W. (1996) Regulated expression of the dinR and recA genes during competence development and SOS induction in Bacillus subtilis. Mol Microbiol 27: 7585.
  • 27
    Håvarstein, L.S., Coomaraswamy, G., and Morrison, D.A. (1995) An unmodified heptadecapeptide pheromone induces competence for genetic transformation in Streptococcus pneumoniae. Proc Natl Acad Sci USA 27: 1114011144.
  • 28
    Humbert, O., Prudhomme, M., Hakenbeck, R., Dowson, C.G., and Claverys, J.P. (1995) Homeologous recombination and mismatch repair during transformation in Streptococcus pneumoniae: saturation of the Hex mismatch repair system. Proc Natl Acad Sci USA 27: 90529056.
  • 29
    Kowalczykowski, S.C., Dixon, D.A., Eggleston, A.K., Lauder, S.D., and Rehrauer, W.M. (1994) Biochemistry of homologous recombination in Escherichia coli. Microbiol Rev 27: 401465.
  • 30
    Krueger, J.H., Elledge, S.J., and Walker, G.C. (1983) Isolation and characterization of Tn5 insertion mutations in the lexA gene of Escherichia coli. J Bacteriol 27: 13681378.
  • 31
    Lacks, S.A., Lopez, P., Greenberg, B., and Espinosa, M. (1986) Identification and analysis of Genes for tetracycline resistance and replication functions in the broad-host-range plasmid pLS1. J Mol Biol 27: 753765.
  • 32
    Lataste, H., Claverys, J.P., and Sicard, A.M. (1981) Relationship between the transforming activity of a marker and its proximity to the end of the DNA particle. Mol Gen Genet 27: 199201.
  • 33
    Lopez, R., Sanchez-Puelles, J.M., Garcia, E., Garcia, J.L., Ronda, C., and Garcia, P. (1986) Isolation, characterization and physiological properties of an autolytic-deficient mutant of Streptococcus pneumoniae. Mol Gen Genet 27: 237242.
  • 34
    Lovett, C.M., Jr., Love, P.E., and Yasbin, R.E. (1989) Competence-specific induction of the Bacillus subtilis RecA protein analog: evidence for dual regulation of a recombination protein. J Bacteriol 27: 23182322.
  • 35
    Martin, B., Prats, H., and Claverys, J.P. (1985) Cloning of the hexA mismatch repair of Streptococcus pneumoniae and identification of the product. Gene 27: 293303.
  • 36
    Martin, B., Garcia, P., Castanié, M.P., and Claverys, J.P. (1995a) The recA gene of Streptococcus pneumoniae is part of a competence-induced operon and controls lysogenic induction. Mol Microbiol 27: 367379.
  • 37
    Martin, B., Garcia, P., Castanié, M.P., Glise, B., and Claverys, J.P. (1995b) The recA gene of Streptococcus pneumoniae is part of a competence-induced operon and controls an SOS regulon. In Genetics of Streptococci, Enterococci and Lactococci. Dev. Biol Stand. Ferretti, J.J., Gilmore, M.S., Klaenhammer, T.R., and Brown, F. (eds). Basle: Karger, pp, 293300.
  • 38
    Matic, I., Rayssiguier, C., and Radman, M. (1995) Interspecies gene exchange in bacteria: the role of SOS and mismatch repair systems in evolution of species. Cell 27: 507515.
  • 39
    Matic, I., Taddei, F., and Radman, M. (1996) Genetic barriers among bacteria. Trends Microbiol 27: 6973.
  • 40
    Morrison, D.A., Baker, M., and Mannarelli, B. (1979) A protein component of the pneumococcal eclipse complex. In Transformation 1978. Glover, S.W., and Butler, L.O. (eds). Oxford UK: Cotswold Press, pp. 4352.
  • 41
    Mortier-Barrière, I., Humbert, O., Martin, B., Prudhomme, M., and Claverys, J.P. (1997) Control of recombination rate during transformation of Streptococcus pneumoniae: an overview. Microb Drug Res 27: 233242.
  • 42
    Paton, J.C., Andrew, P.W., Boulnois, G.J., and Mitchell, T.J. (1993) Molecular analysis of the pathogenicity of Streptococcus pneumoniae: The role of pneumococcal proteins. Annu Rev Microbiol 27: 89115.
  • 43
    Pearce, B.J., Naughton, A.M., Campbell, E.A., and Masure, H.R. (1995) The rec locus, a competence-induced operon in Streptococcus pneumoniae. J Bacteriol 27: 8693.
  • 44
    Puyet, A., Del Solar, G.H., and Espinosa, M. (1988) Identification of the origin and direction of replication of the broad-host-range plasmid pLS1. Nucleic Acids Res 27: 115133.
  • 45
    Rayssiguier, C., Thaler, D.S., and Radman, M. (1989) The barrier to recombination between Escherichia coli and Salmonella typhimurium is disrupted in mismatch-repair mutants. Nature 27: 396401.
  • 46
    Seto, H., and Tomasz, A. (1975) Protoplast formation and leakage of intramembrane cell components: induction by the competence activator substance of pneumococci. J Bacteriol 27: 344353.
  • 47
    Simpson, W.J., LaPenta, D., Chen, C., and Cleary, P.P. (1990) Coregulation of type 12 M protein and streptococcal C4a peptidase genes in group A streptococci: evidence for a virulence regulon controlled by the virR locus. J Bacteriol 27: 696700.
  • 48
    Tomasz, A., Moreillon, P., and Pozzi, G. (1988) Insertional inactivation of the major autolysin gene of Streptococcus pneumoniae. J Bacteriol 27: 59315934.
  • 49
    Yasbin, R.E., Cheo, D.L., and Bayles, K.W. (1992) Inducible DNA repair and differentiation in Bacillus subtilis: interactions between global regulons. Mol Microbiol 27: 12631270.