New Results

Recombination proteins differently control the acquisition of homeologous DNA during Bacillus subtilis natural chromosomal transformation

Ester Serrano, Cristina Ramos, Juan C. Alonso, Silvia Ayora
doi: https://doi.org/10.1101/2020.05.13.090589

Abstract

In naturally competent Bacillus subtilis cells the acquisition of closely related genes occurs via homology-directed chromosomal transformation (CT), and its frequency decreases log-linearly with increased sequence divergence (SD) up to 15%. Beyond this and up to 23% SD the interspecies boundary prevails, the CT frequency marginally decreases, and short (<10-nucleotides) segments are integrated via homology-facilitated micro-homologous integration. Both poorly known CT mechanisms are RecA-dependent. Here we identify the recombination proteins required for the acquisition of interspecies DNA. The absence of AddAB, RecF, RecO, RuvAB or RecU, crucial for repair-by-recombination, does not affect CT. However, inactivation of dprA, radA, recJ, recX or recD2 strongly interfered with CT. Interspecies CT was abolished beyond ~8% SD in ΔdprA, ~10% in ΔrecJ, ΔradA, ΔrecX and 14% in ΔrecD2 cells. We propose that DprA, RecX, RadA/Sms, RecJ and RecD2 help RecA to unconstrain speciation and gene flow. These functions are ultimately responsible for generating genetic diversity and facilitate CT and gene acquisition from bacteria of the same genus.

Introduction

The unidirectional nonsexual incorporation of DNA into the genome of a recipient bacterium (horizontal gene transfer) has been the major evolutionary force that has constantly reshaped genomes and persisted through evolution to maximize adaptation to new ecological conditions and genetic diversity 1;Fraser, 2007 #154. The exchange of chromosomal genes among members of the same or related species occurs via RecA-dependent homologous recombination (HR) by Hfr conjugation, viral transduction or natural chromosomal transformation (CT) 2–4. In Hfr conjugation and transduction the recombination machinery processes double-stranded (ds) DNA, whereas during CT the machinery integrates single-stranded (ss) DNA 4,5. Bacteria have developed mechanisms to fight gene transfer by conjugation and transduction, but cells are less efficient to impose barriers to natural transformation 1. Restriction-modification systems rapidly fragmentate internalised non-self dsDNA, but ssDNA, internalised by transformation, is refractory to most restriction systems 6,7 Adaptive immune systems, as CRISPR-Cas, are usually absent in naturally competent bacteria 8.

Genes devoted to natural competence are encoded in the genome of many bacteria, among them the model organism Bacillus subtilis, which occupies a wide range of aquatic and terrestrial niches, and colonizes animal guts 9. Upon stress, a small fraction of B. subtilis stationary phase cells induce competence. During competence development, ongoing DNA replication is halted, a transcriptional programme is activated, a DNA uptake apparatus is assembled at a cell pole, and lysis of kin is induced, releasing DNA for uptake 7,10–12. This DNA uptake apparatus binds any extracellular dsDNA, degrades one strand, and internalises the complementary strand into the cytosol 10–12.

Naturally competent cells can integrate fully homologous DNA (homogamic CT), and with lower efficiency homeologous (similar but not identical) DNA (heterogamic CT) 2. This interspecies CT can replace the recipient by a homeologous DNA, leading to mosaic structures 13–15. The transformation frequency of heterogamic DNA decreases with increased sequence divergence (SD) both in B. subtilis and S.pneumoniae cells 14,15. In B. subtilis, the transformation assays with donor DNAs from different Bacilli species revealed a biphasic curve. Interspecies CT decreased log-linear up to 15% SD, and integration in this part of the curve occurred via one-step homology-directed RecA-dependent gene replacement 16,17 Beyond 15% SD the CT reached a plateau, and the integration of just few nucleotides, at micro-homologous segments (<10-nucleotides [nt]), was observed at a very low efficiency (~1 x 10-5) 17, suggesting another recombination mechanism.

CT also contributes to expand the pan-genome of a species, because it allows the integration of a heterologous DNA sequence, if flanked by two regions identical with recipient, usually >400-nt, but longer homologous regions significantly increase the integration frequency 7,18. Here, two homologous recombination (HR) events in the flanks integrate the heterologous DNA via recipient-deletion /donor-insertion with ~10-fold lower efficiency than homologous gene replacement 7,9. Another mechanism is observed in replicating competent Streptococcus pneumoniae or Acinetobacter baylyi cells. Here, homology at only one flank (anchor region) facilitates illegitimate recombination of short (<10-nt) micro-homologous segments with subsequent deletion of the intervening host DNA and integration of long heterologous DNA segments, albeit with very low efficiency when compared to homogamic CT 20,21. The length of the anchor region affects the efficiency of this homology-facilitated illegitimate recombination (HFIR) 19.

The proteins responsible for the acquisition of natural homeologous DNA remain poorly known. The main player during HR is the RecA recombinase. RecA from a naturally competent bacterium (e.g., B. subtilis) has evolved to catalyse strand exchange in either the 5’→3’ or 3’→5’ direction and to tolerate 1-nt mismatch in an 8-nt region 16,17 In naturally competent cells, the essential SsbA and competence-specific SsbB coated the incoming linear ssDNA as soon it leaves the entry channel. B. subtilis RecA cannot filament onto SsbA- or SsbB-coated ssDNA 22,23. The RecA accessory proteins are divided into four broad classes. First, the mediators that act before RecA-mediated homology search. The DprA mediator facilitates partial disassembly of SsbA and SsbB from the ssDNA, allows RecA binding, and in concert with SsbA assists RecA to catalyse DNA strand exchange 23. In the absence of DprA, RecO assists RecA to filament onto SsbA-coated ssDNA 22. Second, the modulators, which act during DNA identity search and strand exchange, and either promote RecA nucleoprotein filament assembly as RecF (in the DrecX context) or its disassembly from the ssDNA, as RecX or RecU (in the DrecX context) 24–28. With the help of both, mediators and modulators, a dynamic RecA-ssDNA filament is engaged in sequence identity search. E. coli RecA requires ~15-nt of homologous ssDNA to promote strand exchange, defining the in vitro minimal efficient processing segment (MEPS) 29,30. In vivo, E. coli RecA significantly recombines two duplex DNAs with 25- to 30-bp MEPS 31, thus this length was proposed for the B. subtilis protein 32. In vivo, upon finding a MEPS, RecA initiates strand invasion to form a heteroduplex, known as displacement loop (D-loop) 33. Then, branch migration translocases (RecD2, RuvAB, RecG and RadA/Sms) allow D-loop extension, and help to generate a stable heteroduplex 28,34. Finally, after DNA strand exchange a structure-specific nuclease must resolve the D-loop. The RecU resolvase is unable to cleave D-loops 35–37, and the enzyme responsible for such activity remains unknown. The contribution of these RecA accessory proteins to homogamic CT has been studied in deep 7,28,38, but little is known about their role in the integration of heterogamic DNA. In B. subtilis, two exonucleases, the AddAB complex (counterpart of E. coli RecBCD) and RecJ are crucial to process double-strand breaks (DSBs) during DNA repair 33,39, but their putative role in the degradation of the displaced strand during CT remains to be documented.

DNA transfer occurs in ecologically cohesive communities. In this study we aimed to identify the recombination mechanisms and the proteins that contribute to heterogamic CT. We found that interspecies CT frequency is similar to rec+ cells in the addAB, recO, recF, ruvAB and recU context. Our results suggest that for interspecies CT only a subset of recombination proteins is required. RecA, RecX, RecJ, RecD2, DprA and RadA/Sms participate in both, homology-directed HR and homology-facilitated micro-homologous integration mechanisms that are active depending on the degree of SD in B. subtilis competent cells.

Results

Experimental design

The rational to select B. subtilis competent cells to analyse how the recombination machinery contributes to heterogamic CT is summarised in Supplementary Annex 1. Competence development is a stochastic process driven by the expression of ComK, and it occurs only in a small fraction (0.1 - 5%) of cells 11,12. Interspecies CT, at SD above 8%, is at the limit of detection in wild type cells 14. To overcome such technical difficulty, Rok, which directly represses comK expression, was deleted 40. Inactivation of rok increases the subpopulation of cells that develop natural competence 41, and thereby the sensitivity of the assay. The different rec mutations were mobilised into the rok strain (see Table 1). All the strains additionally lack Rok, but for simplicity we only state the rec gene that is mutated.

View this table:
Table 1.

Homogamic CT frequency of rec-deficient strains

The rpoB gene encodes for the essential ß subunit of the RNA polymerase. The 2997-bp donor DNAs have a mutation at codon 482 that confers rifampicin resistance (RifR, rpoB482) (Fig. S1B). The rpoB482 DNA from B. subtilis 168 with 1 mismatch (Bsu 168 rpoB482) was used for homogamic CT (i.e, the recipient’s own DNA, with just the RifR mutation). For heterogamic CT a fixed concentration of purified rpoB 482 donor DNAs derived from the B. subtilis clade (2.47% SD and 8.35% SD), the B. amyloliquefaciens clade with 10.12% SD, the B. licheniformis clade (14.52% SD and 17% SD), the B. thuringiensis clade with 20.83% SD, or a far distant Bacillus with 22.74% SD (B. smithii DSM4512) (Supplemental Annex 2, Fig. S1A). A further description of these DNAs is presented in Supplementary material Annex 2. As revealed in Fig. S1B, the mismatches in these natural homeologous DNAs are almost homogeneously distributed 16,17, although there are some short regions with higher SD (see below). To avoid fitness costs, the promoter of the rpoB gene is provided by the recipient strain, and the sequence identity at the protein level varied from 99% to 89% (Fig. S1C).

DprA, RecX, RadA/Sms, RecJ and RecD2 contribute to homogamic CT, but not RecF, RecO, RecU, RuvAB and AddAB

Except in ΔrecD2 cells, the number of RifR spontaneous mutant colonies, which appeared in the absence of rpoB482 DNA in the different rec− strains was similar to the wt control (see Supplementary material Annex 3). To test how the different rec- mutants affect intraspecies CT, competent cells were transformed with Bsu 168 rpoB482 DNA with direct selection for RifR (Table 1, Fig. 1A). The homogamic CT frequency with this donor DNA was similar to the one obtained with Bsu 168 met+ DNA, also with a single donor-recipient mismatch 17. As previously reported, intraspecies CT was blocked in the DrecA strain (Table 1, Fig. 1A) 16,42.

Figure 1.
Figure 1.

CT frequencies as a function of SD in different rec mutants. Donor DNA was a rpoB482 DNA conferring Rif; derived from B. subtilis 168 (0.04% SD, homologous DNA), B. subtilis W23 (2.47%), B. atrophaeus 1942 (8.35%), B. amyloliquefaciens DSM7 (10.12%), B. licheniformis DSM13 (14.52%), B. gobiensis FJAT-4402 (17%), B. thuringiensis MC28 (20.83%) and B. smithii DSM4216 (22.74% SD). The rpoB482 DNA (0.1 μg DNA/ml) from these different Bacillus species with the selectable Rif mutation was used to transform BG1359 (rec+, ●) competent cells and its isogenic derivatives. The values are plotted dividing the number of transformants/CFUs obtained in each condition by the number of transformants/CFUs obtained when the rec+ cells are transformed with Bsu168 rpoB482 DNA. In (A): BG1611 (recF15, □), BG1641 (ΔrecO,○) and BG1633 (ΔrecA, ▽); in (B): BG1653 (ΔrecU, □), BG1485 (ΔruvAB, ○) and BG1631 (ΔaddAB, ▽); in (C): BG1813 (ΔrecJ, □), BG1397 (ΔrecX, ○) and BG1647 (DradA, ▽); in (D): BG1549 (ΔrecD2, □) and BG1811 (ΔdprA, ○). All data points are mean ± standard error of the mean (SEM) derived from 3 to 5 independent experiments

Natural homogamic CT was barely affected in recF15, ΔrecO, ΔrecU, ΔruvAB, ΔaddAB cells (Table 1, Fig. 1A-B). Their CT frequencies were similar to results previously observed for these recombination mutants in the rok+ background 28,43. However, the frequency of intraspecies CT was significantly reduced in competent ΔrecJ (by ~80-fold), ΔrecX (~500-fold), ΔrecD2 (~950-fold), ΔradA (~1600-fold) and ΔdprA (~24,000-fold) cells (Table 1, Fig. 1C-D). Except in ΔrecJ, intraspecies CT efficiency in these mutants was lower in Δrok than in rok+ 26,34. The deleterious effect of additionally mutate rok in these backgrounds will be reported elsewhere.

Interspecies CT requires DprA, RecX, RadA/Sms, RecD2 and RecJ

To gain insight into the contribution of these recombination proteins to interspecies CT, we used rpoB482 DNAs with different degree of SD (Supplementary Annex 2). The frequency of interspecies CT decreased logarithmically with increased SD up to ~15% in competent recF15, ΔrecO, ΔrecU, ΔruvAB or ΔaddAB cells (Fig. 1A-B). When SD was further increased, beyond 15% and up to ~23% SD, the interspecies CT frequencies varied <3-fold in recF15, ΔrecO, ΔrecU, ruvAB and addAB (Fig. 1A-B), as in the rec+ control (Fig. 1A) 17, suggesting that these functions do not limit genetic recombination in otherwise rec+ cells.

A different outcome was observed in ΔrecJ, ΔrecX, ΔradA, ΔrecD2 and ΔdprA. The CT frequency was similar to the frequency of spontaneous RifR mutations beyond ~8% SD in ΔdprA, ~10% in ΔrecX and ΔradA, and ~15% in ΔrecJ and ΔrecD2 competent cells, (Fig. 1C-D). We believe that this strong defect was not due to an impairment in fitness cost. The colony size observed after overnight growth at 37°C under selective pressure was similar in all cases. Furthermore, the sequence analysis of the chimeric rpoB482 genes revealed a RpoB482 protein only bearing the RifR mutation (see below).

The interspecies barrier is altered in ΔrecJ, ΔrecX, ΔrecD2, DradA and ΔdprA

To simplify the analysis of the strains that are significantly impaired in heterogamic CT, we gave a value of 1 to the transformation rate obtained by the given mutant in the intraspecies CT assay, and plotted the data relative to this value. Hence only the impact of SD is evaluated (Fig. 2A-B). Different outcomes were observed. First, in competent ΔrecX, ΔradA and ΔrecJ cells interspecies CT decreased logarithmically with increased SD, but only up to ~10% (ΔrecJ, ~14%) SD, to reach a plateau at higher divergence (Fig. 2A). Second, competent recD2 reached a plateau already at 8% SD (Fig. 2B). Third, upon inactivation of dprA, the decline in the rate of recombination with SD was only observed at ~2% SD, to reach a plateau at higher divergence (Fig. 2B).

Figure 2.
Figure 2.

Normalised CT frequencies. The CT frequencies were normalized to give a value of 1 to the transformation frequency of the indicated rec-deficient strain when transformed with homogamic DNA, and heterogamic CTs are plotted relative to these values. In (A): ΔrecJ, ΔrecX and ΔradA (A); in (B): ΔrecD2 and ΔdprA RifR cells. For comparison, the values obtained for the rec+ strain are also plotted.

RecO, RecF, RecU, RuvAB and AddAB are dispensable in heterogamic transformation

The above results suggest that RecO, RecF, RecU, RuvAB and AddAB play no apparent roles in interspecies CT, but their contribution to the integration length is unknown. To analyse this, we sequenced the rpoB482 gene from RifR clones and calculated the mean integration length.

The maximal integration length that can be detected with the donor DNA of ~2% SD (Bsu W23 DNA) is 2628-bp, because the first mismatch is located at position 350 and the last at position 2978. Previous analysis showed that when the rec+ strain was transformed with this donor DNA, the mean integration length was close to this maximal integration length, around ~2300-bp 16,17. Similar results were obtained with the donor DNA of ~2% SD in the ΔrecO, ΔrecU, ΔruvAB and ΔaddAB backgrounds (Fig 3A-B).

Figure 3.
Figure 3.

Determination of integration length of donor DNA with increased SD in different rec mutants. The mean length of integration was determined by sequencing the 2997-bp rpoB482 DNA from different transformants obtained in rec+, recF15 and ΔrecO (A); ΔrecU, ΔruvAB and ΔaddAB (B); ΔrecD2, ΔrecJ and ΔrecX (C) and ΔradA, ΔdprA and ΔrecA (D). Integration endpoints were defined as the average between the last single nucleotide polymorphism present and the next SNP absent in the sequence of trasformants (integration endpoint). The empty bars highlight the values obtained with strains that undergo both homology-directed and homology-facilitated CT, and filled bars denote strains impaired in homology-directed and blocked in homology-facilitated CT. ND, not detected.

Up to 15% divergence range, RecA is believed to integrate the DNA by a homology-directed HR mechanism, initiating recombination in a MEPS 16,22. A comparison of the nucleotide sequence of donor with recipient DNA revealed the presence of 22 MEPS at or above 25-nt in donor DNA with ~8% SD, and 21 at ~10% SD. In both donor DNAs there is a long stretch of ~200-nt of sequence identity upstream of the rpoB482 mutation, and several regions with a MEPS longer than 54-nt downstream of the rpoB482 mutation. They could define the left and right recombination endpoints, being the region in-between integrated independently of its SD, as it occurs in the insertion of heterologous DNA by two-step HR at the flanks (see Introduction). However, integrated fragments of ~1600-nt (i.e, recombination endpoints at the 157-nt [at position 1-157] and the 81-nt MEPS [at position 1509-1589]) was not observed in rec+ transformations with ~8% SD donor DNA, and the same was observed with 10% SD (Fig 3A). This result suggests that the two-step deletion/insertion CT may not take place with interspecies DNA, probably because it requires two longer flanking homologous regions (see Introduction), or the sequence in-between plays a relevant role.

The analysis of 10-20 RifR clones obtained in the transformation of competent recF15, ΔrecO, ΔrecU, ΔruvAB and ΔaddAB cells with donor DNA of ~8% SD revealed that the mean integration length was as in rec+, 700-900-nt, except in ΔruvAB, which was ~490-nt (i.e, ~2 times less) (Fig. 3A-B, and Table S1). A sequence analysis of the integrated region revealed eight MEPS (four upstream [50-, 35-, 38- and 36-nt] and four downstream [35-, 81-, 41- and 33-nt]) of the rpoB482 mutation. One recombination endpoint was usually at one of these MEPS, but the other endpoint was usually at other region, with a size below MEPS (Table S1). Several hypotheses can explain these results: i) the MEPS used in vivo are shorter, ii) RecA is insensitive to 1-nt mismatch every ~8-nt, but not to higher mismatches. Indeed, if 1-nt mismatch is allowed at the short endpoints, longer MEPS regions can be predicted in all cases, and iii) integration starts at one MEPS and proceeds uni- or bidirectionally until RecA finds a barrier (e.g., 2-nt mismatch every ~8-nt). We noticed that the DNA inserts are usually followed by higher local SD (~20% to ~28%) in the next 25-nt interval, suggesting that recombination ended there because RecA found a heterologous block.

In recF15, ΔrecO, ΔrecU, ΔruvAB, ΔaddAB and rec+ cells, when SD is ~10%, the mean integration length was ~300-nt (Fig. 3A-B). In these transformants, the nucleotide sequence changes do not alter the RpoB482 protein sequence (see Fig. S1C). Within a 600-nt interval centred at the rpoB482 mutation, there are only three sequences at or above MEPS (see Fig. 4A). Analysis of 10 different RifR clones in rec+ showed that in ~70% of the cases, the 3’-endpoint was at the 54-nt MEPS (at position 1536-1590, Fig. 4A). Recombination in this region did not extend beyond, probably because it is followed by a local ~24% SD in the next 25-nt that could act as a strong heterologous barrier. The other endpoints were located in regions below MEPS which are usually followed by barriers with a higher SD (e.g., endpoint at position 1371 is preceded by 25-nt with ~24% SD, at position 1251 by ~24% SD). No patched sequences were observed, discarding that in some transformants multiple recombination events had occurred at different loci. As in rec+, the recombination endpoints in the recF15, ΔrecO, ΔrecU, ΔruvAB and ΔaddAB transformants did not always coincide with the longest MEPS present (Table S1). In these mutants one endpoint was usually at a region where a MEPS was located, whereas the other endpoint was less specific, and often below MEPS. All these results suggest that a MEPS is necessary to initiate strand invasion, but a SD barrier may halt RecA-mediated uni- or bidirectional DNA strand exchange.

Figure 4.
Figure 4.

Mapping of recombination endpoints during interspecies CT in rec+ cells. The length of integrated DNA and recombination endpoints were determined by sequencing ten different RR clones obtained following transformation with Bam DSM7 rpoB482 donor DNA (10.12% SD, in A) or Bli DSM13 rpoB482 donor DNA (14.52% SD, in B). In the top of the figure, the features of the donor DNA are indicated: The mutation that confers RifR, located at 1443 bp in the rpoB genes, is marked by a pin. In Bam DSM7 rpoB482 DNA (10.12% SD) the regions in the sequence where MEPSs (i.e., fully homologous regions larger than 26-bp) are found are denoted by a vertical dotted line. Their length is indicated also. MEPS regions are not observed in Bli DSM13 rpoB482 DNA (14.52 % SD). The MEPS and integration lengths of ten different RifR clones are shown, in brackets are indicated the location of recombination endpoints.

Nucleotide sequence analysis of 15-30 RifR clones obtained with ~15% SD in ΔruvAB and rec+ cells showed that all were genuine transformants, but these values dropped to ~40-50% in ΔrecU, ΔrecO, ΔaddAB, and recF15. The mean integration length was between 400- to 130-nt in recF15, ΔrecO, ΔrecU, ΔruvAB and ΔaddAB, as well as in the rec+ control (Table S1 and Fig. 4B). This donor DNA has a MEPS of 104-nt (position 54-157) upstream of the rpoB482 mutation, and a 38-nt MEPS (position 2295-2332) downstream, but ~2100-nt inserts were not observed, confirming that the region between the MEPS are relevant. Potential MEPS are not present in the integrated region (Fig. 4B). The sequence analysis of rec+ transformants revealed a great heterogenicity in the recombination endpoints, as well as in the length of the MEPS. In all cases the MEPS used were short: between 5- to 20-nt without mismatches (Fig. 4B and Table S1). However, if 1-mismatch is allowed a longer MEPS is detected (e.g., the endpoint at position 1329 increases from 5- to 26-nt) (Fig. 4B).

Beyond 15% SD the transformation efficiency and the frequency of spontaneous mutation almost overlap, although there is a ~3-fold difference, and it is higher than in a ΔrecA mutant, suggesting that still this is a RecA-dependent integration event (Fig. 1). Four regions at or above MEPS (74-, 56-, 28- and 26-nt) upstream and one downstream (26-nt) of the rpoB482 mutation are present in the 17% SD donor, but they were not used to integrate the region in-between.

Nucleotide sequence analyses of RifR rec+ clones obtained with ~17% SD showed that only ~37% were genuine transformants, i.e, they had incorporated two or more nucleotides of donor DNA. Similarly, 20-30% of the RifR clones obtained in ΔrecO, ΔaddAB, ΔrecU, ΔruvAB and recF15 were genuine transformants (Fig. 3A-B), and the mean integration length was also 4- to 8-nt, or ~5-fold below MEPS (Table S1). The observed low efficiency of micro-homologous integration cannot be attributed to a defect in the resulting RpoB482 protein, because at ~17% SD, except the amino acid change that confers RifR, there is no mutation in the 50 residues intervals up and downstream the RifR change (Fig. S1C). The close inspection in donor DNA with 17% SD of the region surrounding the rpoB482 mutation showed that this mutation is embedded in a region with strong SD with recipient DNA. The 25-nt region upstream of the rpoB482 mutation has 8 mismatches (i.e, 32% SD) and the one downstream 10 mismatches (40% SD) that could act as heterologous barriers.

At ~21% SD there are two sites at or above MEPS (42-nt [at position 90-131] and 27-nt [at position 2781-2807]), one on each side of the rpoB482 mutation, but they were not used. After sequencing of transformants obtained in recF15, ΔrecO, ΔrecU, ΔruvAB, ΔaddAB or rec+, we found that again just few nucleotides from the donor had been integrated, and that the fraction of genuine transformants was ~20% (Fig 3A-B). Here, the analysis showed that the rpoB482 mutation is also surrounded by higher SD: 8 mismatches in 25-nt upstream (i.e, ~32% SD) and the one downstream 6 mismatches (i.e, ~24% SD). Finally, at ~23% SD, only one MEPS exists (at position 801-826). The proportion of genuine RifR transformants accounts to only ~6% of the sequenced clones in the rec+ strain. The mean integration length in rec+ transformants was 4- to 8-nt (Fig. 3A-B). Similar results were obtained in recF15, ΔrecO, ΔrecU, ΔruvAB and ΔaddAB.

RecD2, RecJ, RecX, RadA/Sms and DprA are crucial for interspecies CT

Nucleotide sequence analyses of 10-30 RifR clones obtained in the ΔradA, ΔrecJ, ΔrecX, ΔrecD2 or ΔdprA context revealed that when divergence was low (~2%), the mean integration length of the rpoB482 DNA was similar to the rec+ control (Fig 3C-D). The proportion of genuine transformants was different in these mutants. At ~8% SD, all sequenced clones obtained in DradA were genuine transformants, but this number was reduced to ~50 % in ΔdprA, ΔrecJ and ΔrecX, and even lowered to ~25% in ΔrecD2. Representatives of the genuine RifR clones are documented in Table S2.

The efficiency of interspecies transformation at ~10% SD was strongly reduced in these mutants, to levels similar to the spontaneous mutation rate (Fig. 1). Only ~25% of the RifR clones obtained in ΔrecX were genuine transformants, and these values lowered to ~15% in ΔrecD2, and ~5% in ΔrecJand ΔradA cells. Furthermore, in the ΔdprA strain all 25 sequenced RifR clones were spontaneous mutants (Table S2). The mapping of recombination endpoints in the ΔrecJ, ΔrecD2, ΔrecX and ΔradA transformants showed that, as in the rec+ transformants, the 3’-endpoint of recombination was located at the 54-nt MEPS, except in some ΔrecX transformants. The 5’-endpoint of recombination was usually located at a larger MEPS when compared with the rec+ control (Tables S1 and S2).

At ~15% SD, all sequenced RifR clones obtained were spontaneous mutants in the ΔrecX, ΔradA, ΔrecJ and ΔdprA strains. In ΔrecD2, only one of the sequenced RifR clones (1/19) was a genuine transformant with a mean integration length >120-nt (Table S2). At ~17% SD, all sequenced RifR clones were spontaneous mutants in the ΔrecX, ΔradA, ΔrecJ, ΔdprA, and ΔrecD2 backgrounds. These results show that these functions are all needed for homology-directed integration when SD is larger than 10%, and also for micro-homologous integration.

Discussion

From the data presented in this work, it can be inferred that recombination proteins differently facilitate adaptation and genetic diversity, and that two different recombination mechanisms occur depending on the SD of the interspecies DNA. Up to 15% SD, homology-directed HR accounts for the integration of divergent sequences longer than 130-nt. Beyond this, integration of just few nucleotides at micro-homologous segments is observed.

Inactivation of the main end-resection complex (addAB), a positive mediator (recO), a positive (recF) or a negative (recU) modulator or enzymes essential for Holliday junction processing and cleavage (ruvAB, recU) barely affects heterogamic CT (Fig. 1A-B), although they are essential for DNA DSB repair 39,43,44. The interpretation of such result may not be so simple, because cells lacking both RecF and AddAB are blocked even in homogamic CT 43, suggesting that they are backup functions during CT (e.g., RecF is essential in the recX context) 25. We have observed that a specific subset of HR proteins (RecA, DprA, RecX, RadA/Sms, RecD2 and RecJ) contributes to acquire homeologous DNA. Except DprA which is competence specific and present in all transformable bacteria, and even in non-transformable ones 10, the recombination proteins identified here also participate, together with another subset of HR proteins, in the accurate repair of lethal DSBs and in the restart of stalled replication forks. In some distantly related competent cells (e.g., Proteobacteria phylum) RecD2 is absent, and RadA/Sms is replaced by ComM 45. They differently participate in the recombination mechanisms that may be active during the acquisition of homeologous DNA. In the absence of RecA CT is blocked 16. Interspecies homology-directed CT was blocked with DNA of the same clade (up to ~8% SD) in ΔdprA, beyond ~10% SD in the ΔrecX, ΔrecJ and ΔradA backgrounds, and beyond ~15% SD in ΔrecD2 cells (Fig. 3C-D), showing an essential role for these proteins in the acquisition of interspecies DNA.

We can envision that during homology-directed CT, RecA nucleates in the incoming ssDNA coated by SsbA and SsbB, with the help of the DprA-SsbA two-component mediator. Then, a dynamic RecA nucleoprotein filament, with the contribution of DprA-SsbA and RecX, identifies an identical sequence in the recipient genome. Once a homologous region is found RecA promotes DNA strand invasion to produce a metastable heteroduplex DNA 23,27. RecA at this D-loop interacts with and loads the branch-migration translocase RadA/Sms as documented in Firmicutes 34,46. RadA/Sms, in concert with RecA, facilitates bidirectional branch migration until a region with a SD >20% is found. This is consistent with the in vitro observation that homology-directed RecA-mediated strand-exchange halted at DNA patches >16% SD 16.

The contribution of RecD2 helicase and RecJ exonuclease to homology-directed recombination during interspecies CT is poorly understood. Recently, it was suggested that RecD2 contributes to branch migrate the heteroduplex DNA in a relaxed molecule, perhaps upon cleavage of the displaced strand 28. As RecJ of other naturally competent cells 47, B. subtilis RecJ possesses an extra C-terminal domain, absent in E. coli RecJ, critical for protein-protein interaction (e.g., SsbA). Perhaps in concert with the RecD2 helicase, it might degrade the displaced strand and the non-paired tails. Finally, the ends are sealed, leading to the acquisition of homeologous DNA >130-nt.

It was remarkable the heterogeneity of the MEPS at recombination endpoints in homology-directed HR. In vitro assays with E. coli RecA show that identity search relies on probing tracts of 8-nt homology, based on the transient interactions between the stretched ssDNA within the filament and bases in a locally melted and stretched DNA duplex 48,49. It has been shown that RecA evolved to tolerate 1-nt mismatch every ~8-nt region in vitro, albeit DNA strand exchange with a short region of 16% SD is delayed 16,50. We propose that a sum of delays might compromise DNA strand exchange and determines the length of DNA integrated.

Beyond 15% sequence, integration of <10-nt segments was observed in ΔaddAB, recF, ΔrecO, ΔrecU and ΔruvAB (Fig. 3A-B), but it was not detected in ΔrecA, ΔdprA, ΔrecJ, ΔrecX, ΔradA and ΔrecD2 cells (Fig. 3C-D). These results showed that a similar set of recombination functions are also required for this short integration, which occurs up to 23% SD, although with low efficiency. Here we propose a homology-facilitated micro-homologous integration mechanism: As above, a RecA dynamic filament, with the help of DprA-SsbA and RecX, searches for and identifies a MEPS on the rpoB482 DNA, which is used in this case as an anchor region, to produce a metastable D-loop intermediate, in concert with RadA/Sms. Once the DNA is anchored at MEPS, DprA could mediate the annealing of short stretches of homeologous DNA (3- to 8-nt) around the rpoB482 mutation 23. Then, the donor ssDNA loop between the anchored region and the micro-homologous paired segment has to deleted, perhaps by RecJ in concert with RecD2. Finally, the ends of the integrated segment are sealed and rapidly expressed 45. This mechanism differs from HFIR observed in replicating competent S. pneumoniae and A. baylyi cells (see Introduction) 20,21. In these competent bacteria, inactivation of recBCD or recJ significantly increased HFIR 19,21. In contrast, in non-replicating competent B. subtlis cells integration of thousands of nucleotides of the heterogamic DNA with the subsequent deletion of the recipient DNA was not observed, and inactivation of recJ blocked homology-facilitated micro-homologous integration. Competent A. baylyi cells can also integrate short ssDNA (20-nt) in the recipient genome by another mechanism, which occurs with extremely low efficiency.

This mechanism requires active DNA replication, inactivation of recJ and is independent of RecA 51. It is likely that competent B. subtilis cells may use homology-facilitated micro-homologous integration to restore genes inactivated by mutations and thereby prevent the irreversible deterioration of genomes (Muller’s ratchet) 4. At ~23% SD interspecies CT frequency was similar to spontaneous mutations, suggesting that beyond ~23% SD micro-homologous integration might be inefficient, probably because the unique MEPS present is too short to serve as a stable anchor region 19. Similarly, competent H. pylori cells cannot be transformed by Campylobacter jejuni DNA with ~24% SD 52.

Can the above observations be extrapolated to interspecies Hfr chromosomal conjugation? The frequency of interspecies Hfr chromosomal conjugation also decreases log-linearly with increased SD up to ~16% SD. Inactivation of mutSL alleviates interspecies Hfr conjugation by ~1000-fold, and deletion of recBCD or ruvAB reduced interspecies Hfr conjugation, but inactivation of recJ increases it 2,3,53,54, suggesting that interspecies Hfr chromosomal conjugation uses the repair-by-recombination mechanism. In contrast, inactivation of mutSL marginally prevents interspecies CT with up to 15% SD in Firmicutes 15,16,55,56 and inactivation of recJ, but not addAB or ruvAB, inhibits interspecies CT. All these results suggest that bacteria have evolved different genetic recombination mechanisms devoted to interspecies genetic exchange to generate diversity. Bacterial recA, radA and recX genes, which play crucial roles in interspecies CT, perhaps contributed in their transfer from mitochondria or chloroplasts to the nucleus of land plants, green algae and moss 57, although the evolutionary force and molecular functions that contributed to the transfer of these genes well beyond the species boundaries is poorly understood.

Materials and methods

Bacterial strains and donor DNAs

The parental strain was B. subtilis BG1359. The rec mutations listed in Table 1 were introduced by SPP1 transduction 58.

The rpoB gene from different species, which encodes for the ²-subunit of RNA polymerase, was used as donor DNA, and the rpoB482 mutation, which renders cells RifR, was introduced into all the donor DNAs (Annex 2, supplementary material). Plasmid DNA was prepared by Qiagen extraction and extensive dialysis in Tris-EDTA buffer 17.

Transformation assays

Natural competence was induced as described 42. Competent cells were incubated with 0.1 μg·ml-1 of the indicated rpoB482 donor DNA (30 min, 37°C), and then plated on Rif (8 μg·ml-1) containing LB-agar plates. A control was performed in which competent cells were treated equally, but with no donor DNA to score the appearance of spontaneous RifR mutants. These values (spontaneous RifR mutants) were extracted to the number of transformants. Transformation frequency was calculated as the number of RifR transformants per colony-forming-unit (CFU).

Mapping of integration endpoints

The integration endpoint is defined by the end of the donor sequence followed by the sequence of the recipient. To map integration endpoints, the rpoB gene from the RifR transformants was amplified by PCR, and its nucleotide sequence compared with the one of recipient and donor strains. The presence or the absence of the mismatches between the donor and the recipient DNA were used to determine the MEPS. Endpoints are defined as described 59, and integration length is calculated as the distance between endpoints 59.

Author contributions

S.A. and J.C.A. conceived the project and designed the experiments, E.S., C.R. performed the experiments, E.S., C.R. S.A. and J.C.A. evaluated the results, S.A. and J.C.A. wrote the manuscript.

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data or in the decision to publish the results.

Acknowledgements

ES and CR thank the Ministerio de Ciencia e Innovación [Agencia Estatal de Investigación] (MCIU[AEI]), BES-2013-063433 to E.S., and BES-2017-080504 to C.R. for the fellowships. This work was partially supported by MCIU/AEI/FEDER, EU PGC2018-097054-B-I00 to J.C.A and S.A. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Footnotes

  • ↵✉ Juan C. Alonso jcalonso{at}cnb.csic.es and Silvia Ayora sayora{at}cnb.csic.es

References

  1. ↵
    Gogarten, J. P., Doolittle, W. F. & Lawrence, J. G. Prokaryotic evolution in light of gene transfer. Mol Biol Evol 19, 2226–2238, doi:10.1093/oxfordjournals.molbev.a004046 (2002).
    OpenUrlCrossRefPubMedWeb of Science
  2. ↵
    Fraser, C., Hanage, W. P. & Spratt, B. G. Recombination and the nature of bacterial speciation. Science 315, 476–480, doi:10.1126/science.1127573 (2007).
    OpenUrlAbstract/FREE Full Text
  3. ↵
    Matic, I., Taddei, F. & Radman, M. Genetic barriers among bacteria. Trends Microbiol 4, 69–72, doi:10.1016/0966-842X(96)81514-9 (1996).
    OpenUrlCrossRefPubMedWeb of Science
  4. ↵
    Takeuchi, N., Kaneko, K. & Koonin, E. V. Horizontal gene transfer can rescue prokaryotes from Muller’s ratchet: benefit of DNA from dead cells and population subdivision. G3 (Bethesda) 4, 325–339, doi:10.1534/g3.113.009845 (2014).
    OpenUrlCrossRef
  5. ↵
    Hanage, W. P. Not So Simple After All: Bacteria, Their Population Genetics, and Recombination. Cold Spring Harb Perspect Biol 8, doi:10.1101/cshperspect.a018069 (2016).
    OpenUrlAbstract/FREE Full Text
  6. ↵
    Trautner, T. A. & Spatz, H. C. Transfection in B. subtilis. Curr Top Microbiol Immunol 62, 61–88 (1973).
    OpenUrlPubMed
  7. ↵
    Kidane, D., Ayora, S., Sweasy, J. B., Graumann, P. L. & Alonso, J. C. The cell pole: the site of cross talk between the DNA uptake and genetic recombination machinery. Crit Rev Biochem Mol Biol 47, 531–555, doi:10.3109/10409238.2012.729562 (2012).
    OpenUrlCrossRefPubMedWeb of Science
  8. ↵
    Oliveira, P. H., Touchon, M. & Rocha, E. P. Regulation of genetic flux between bacteria by restriction-modification systems. Proc Natl Acad Sci USA 113, 5658–5663, doi:10.1073/pnas.1603257113 (2016).
    OpenUrlAbstract/FREE Full Text
  9. ↵
    Brito, P. H. et al. Genetic Competence Drives Genome Diversity in Bacillus subtilis. Genome Biol Evol 10, 108–124, doi:10.1093/gbe/evx270 (2018).
    OpenUrlCrossRef
  10. ↵
    Johnston, C., Martin, B., Fichant, G., Polard, P. & Claverys, J. P. Bacterial transformation: distribution, shared mechanisms and divergent control. Nat Rev Microbiol 12, 181–196, doi:10.1038/nrmicro3199 (2014).
    OpenUrlCrossRefPubMed
  11. ↵
    Dubnau, D. & Blokesch, M. Mechanisms of DNA Uptake by Naturally Competent Bacteria. Annu Rev Genet 53, 217–237, doi:10.1146/annurev-genet-112618-043641 (2019).
    OpenUrlCrossRefPubMed
  12. ↵
    Maier, B. Competence and Transformation in Bacillus subtilis. Curr Issues Mol Biol 37, 57–76, doi:10.21775/cimb.037.057 (2020).
    OpenUrlCrossRef
  13. ↵
    Spratt, B. G., Bowler, L. D., Zhang, Q. Y., Zhou, J. & Smith, J. M. Role of interspecies transfer of chromosomal genes in the evolution of penicillin resistance in pathogenic and commensal Neisseria species. J Mol Evol 34, 115–125, doi:10.1007/bf00182388 (1992).
    OpenUrlCrossRefPubMedWeb of Science
  14. ↵
    Zawadzki, P., Roberts, M. S. & Cohan, F. M. The log-linear relationship between sexual isolation and sequence divergence in Bacillus transformation is robust. Genetics 140, 917–932 (1995).
    OpenUrlAbstract/FREE Full Text
  15. ↵
    Humbert, O., Prudhomme, M., Hakenbeck, R., Dowson, C. G. & Claverys, J. P. Homeologous recombination and mismatch repair during transformation in Streptococcus pneumoniae: saturation of the Hex mismatch repair system. Proc Natl Acad Sci U S A 92, 9052–9056 (1995).
    OpenUrlAbstract/FREE Full Text
  16. ↵
    Carrasco, B., Serrano, E., Martín-González, A., Moreno-Herrero, F. & Alonso, J. C. Bacillus subtilis MutS modulates RecA-mediated DNA strand exchange between divergent DNA sequences. Front Microbiol 10, 237, doi:10.3389/fmicb.2019.00237 (2019).
    OpenUrlCrossRef
  17. ↵
    Carrasco, B., Serrano, E., Sanchez, H., Wyman, C. & Alonso, J. C. Chromosomal transformation in Bacillus subtilis is a non-polar recombination reaction. Nucleic Acids Res 44, 2754–2768, doi:10.1093/nar/gkv1546 (2016).
    OpenUrlCrossRefPubMed
  18. ↵
    Simpson, D. J., Dawson, L. F., Fry, J. C., Rogers, H. J. & Day, M. J. Influence of flanking homology and insert size on the transformation frequency of Acinetobacter baylyi BD413. Environ Biosafety Res 6, 55–69, doi:10.1051/ebr:2007027 (2007).
    OpenUrlCrossRefPubMed
  19. ↵
    Brigulla, M. & Wackernagel, W. Molecular aspects of gene transfer and foreign DNA acquisition in prokaryotes with regard to safety issues. Appl Microbiol Biotechnol 86, 1027–1041, doi:10.1007/s00253-010-2489-3 (2010).
    OpenUrlCrossRefPubMed
  20. ↵
    Prudhomme, M., Libante, V. & Claverys, J. P. Homologous recombination at the border: insertion-deletions and the trapping of foreign DNA in Streptococcus pneumoniae. Proc Natl Acad Sci U S A 99, 2100–2105, doi:10.1073/pnas.032262999 (2002).
    OpenUrlAbstract/FREE Full Text
  21. ↵
    de Vries, J. & Wackernagel, W. Integration of foreign DNA during natural transformation of Acinetobacter sp. by homology-facilitated illegitimate recombination. Proc Natl Acad Sci U S A 99, 2094–2099, doi:10.1073/pnas.042263399 (2002).
    OpenUrlAbstract/FREE Full Text
  22. ↵
    Carrasco, B., Yadav, T., Serrano, E. & Alonso, J. C. Bacillus subtilis RecO and SsbA are crucial for RecA-mediated recombinational DNA repair. Nucleic Acids Res 43, 5984–5997, doi:10.1093/nar/gkv545 (2015).
    OpenUrlCrossRefPubMed
  23. ↵
    Yadav, T., Carrasco, B., Serrano, E. & Alonso, J. C. Roles of Bacillus subtilis DprA and SsbA in RecA-mediated genetic recombination. J Biol Chem 289, 27640–27652, doi:10.1074/jbc.M114.577924 (2014).
    OpenUrlAbstract/FREE Full Text
  24. ↵
    Carrasco, B., Ayora, S., Lurz, R. & Alonso, J. C. Bacillus subtilis RecU Holliday-junction resolvase modulates RecA activities. Nucleic Acids Res 33, 3942–3952, doi:10.1093/nar/gki713 (2005).
    OpenUrlCrossRefPubMedWeb of Science
  25. ↵
    Cardenas, P. P. et al. RecX facilitates homologous recombination by modulating RecA activities. PLoS Genet 8, e1003126, doi:10.1371/journal.pgen.1003126 (2012).
    OpenUrlCrossRefPubMed
  26. ↵
    Serrano, E., Carrasco, B., Gilmore, J. L., Takeyasu, K. & Alonso, J. C. RecA Regulation by RecU and DprA During Bacillus subtilis Natural Plasmid Transformation. Front Microbiol 9, 1514, doi:10.3389/fmicb.2018.01514 (2018).
    OpenUrlCrossRef
  27. ↵
    Le, S. et al. Bacillus subtilis RecA with DprA-SsbA antagonizes RecX function during natural transformation. Nucleic Acids Res 45, 8873–8885, doi:10.1093/nar/gkx583 (2017).
    OpenUrlCrossRef
  28. ↵
    Serrano, E., Ramos, C., Ayora, S. & Alonso, J. C. Viral SPP1 DNA is infectious in naturally competent Bacillus subtilis cells: inter-and intramolecular recombination pathways. Environ Microbiol, doi:10.1111/1462-2920.14908 (2020).
    OpenUrlCrossRef
  29. ↵
    Hsieh, P., Camerini-Otero, C. S. & Camerini-Otero, R. D. The synapsis event in the homologous pairing of DNAs: RecA recognizes and pairs less than one helical repeat of DNA. Proc Natl Acad Sci U S A 89, 6492–6496 (1992).
    OpenUrlAbstract/FREE Full Text
  30. ↵
    Yang, D., Boyer, B., Prevost, C., Danilowicz, C. & Prentiss, M. Integrating multi-scale data on homologous recombination into a new recognition mechanism based on simulations of the RecA-ssDNA/dsDNA structure. Nucleic Acids Res 43, 10251–10263, doi:10.1093/nar/gkv883 (2015).
    OpenUrlCrossRefPubMed
  31. ↵
    Watt, V. M., Ingles, C. J., Urdea, M. S. & Rutter, W. J. Homology requirements for recombination in Escherichia coli. Proc Natl Acad Sci U S A 82, 4768–4772 (1985).
    OpenUrlAbstract/FREE Full Text
  32. ↵
    Majewski, J. & Cohan, F. M. DNA sequence similarity requirements for interspecific recombination in Bacillus. Genetics 153, 1525–1533 (1999).
    OpenUrlAbstract/FREE Full Text
  33. ↵
    Kowalczykowski, S. C. An Overview of the Molecular Mechanisms of Recombinational DNA Repair. Cold Spring Harb Perspect Biol 7, doi:10.1101/cshperspect.a016410 (2015).
    OpenUrlAbstract/FREE Full Text
  34. ↵
    Torres, R., Serrano, E. & Alonso, J. C. Bacillus subtilis RecA interacts with and loads RadA/Sms to unwind recombination intermediates during natural chromosomal transformation. Nucleic Acids Res 47, 9198–9215, doi:10.1093/nar/gkz647 (2019).
    OpenUrlCrossRef
  35. ↵
    Ayora, S., Carrasco, B., Doncel-Perez, E., Lurz, R. & Alonso, J. C. Bacillus subtilis RecU protein cleaves Holliday junctions and anneals single-stranded DNA. Proc Natl Acad Sci U S A 101, 452–457, doi:10.1073/pnas.2533829100 (2004).
    OpenUrlAbstract/FREE Full Text
  36. McGregor, N. et al. The structure of Bacillus subtilis RecU Holliday junction resolvase and its role in substrate selection and sequence-specific cleavage. Structure 13, 1341–1351, doi:10.1016/j.str.2005.05.011 (2005).
    OpenUrlCrossRefPubMed
  37. ↵
    Cañas, C. et al. Interaction of branch migration translocases with the Holliday junctionresolving enzyme and their implications in Holliday junction resolution. J Biol Chem 289, 17634–17646, doi:10.1074/jbc.M114.552794 (2014).
    OpenUrlAbstract/FREE Full Text
  38. ↵
    Kidane, D. et al. Evidence for different pathways during horizontal gene transfer in competent Bacillus subtilis cells. PLoS Genet 5, e1000630, doi:10.1371/journal.pgen.1000630 (2009).
    OpenUrlCrossRefPubMed
  39. ↵
    Ayora, S. et al. Double-strand break repair in bacteria: a view from Bacillus subtilis. FEMS Microbiol Rev 35, 1055–1081, doi:10.1111/j.1574-6976.2011.00272.x (2011).
    OpenUrlCrossRefPubMed
  40. ↵
    Hoa, T. T., Tortosa, P., Albano, M. & Dubnau, D. Rok (YkuW) regulates genetic competence in Bacillus subtilis by directly repressing comK. Mol Microbiol 43, 15–26 (2002).
    OpenUrlCrossRefPubMedWeb of Science
  41. ↵
    Maamar, H. & Dubnau, D. Bistability in the Bacillus subtilis K-state (competence) system requires a positive feedback loop. Mol Microbiol 56, 615–624, doi:10.1111/j.1365-2958.2005.04592.x (2005).
    OpenUrlCrossRefPubMedWeb of Science
  42. ↵
    Alonso, J. C., Tailor, R. H. & Luder, G. Characterization of recombination-deficient mutants of Bacillus subtilis. J Bacteriol 170, 3001–3007 (1988).
    OpenUrlAbstract/FREE Full Text
  43. ↵
    Alonso, J. C., Stiege, A. C. & Luder, G. Genetic recombination in Bacillus subtilis 168: effect of recN, recF, recH and addAB mutations on DNA repair and recombination. Mol Gen Genet 239, 129–136 (1993).
    OpenUrlCrossRefPubMedWeb of Science
  44. ↵
    Sanchez, H. et al. The RuvAB branch migration translocase and RecU Holliday junction resolvase are required for double-stranded DNA break repair in Bacillus subtilis. Genetics 171, 873–883, doi:10.1534/genetics.105.045906 (2005).
    OpenUrlAbstract/FREE Full Text
  45. ↵
    Dalia, A. B. & Dalia, T. N. Spatiotemporal Analysis of DNA Integration during Natural Transformation Reveals a Mode of Nongenetic Inheritance in Bacteria. Cell 179, 1499–1511 e1410, doi:10.1016/j.cell.2019.11.021 (2019).
    OpenUrlCrossRef
  46. ↵
    Marie, L. et al. Bacterial RadA is a DnaB-type helicase interacting with RecA to promote bidirectional D-loop extension. Nat Commun 8, 15638, doi:10.1038/ncomms15638 (2017).
    OpenUrlCrossRef
  47. ↵
    Cheng, K. et al. Structural basis for DNA 5 - end resection by RecJ. Elife 5, e14294, doi:10.7554/eLife.14294 (2016).
    OpenUrlCrossRefPubMed
  48. ↵
    Folta-Stogniew, E., O’Malley, S., Gupta, R., Anderson, K. S. & Radding, C. M. Exchange of DNA base pairs that coincides with recognition of homology promoted by E. coli RecA protein. Mol Cell 15, 965–975, doi:10.1016/j.molcel.2004.08.017 (2004).
    OpenUrlCrossRefPubMedWeb of Science
  49. ↵
    Qi, Z. et al. DNA sequence alignment by microhomology sampling during homologous recombination. Cell 160, 856–869, doi:10.1016/j.cell.2015.01.029 (2015).
    OpenUrlCrossRefPubMed
  50. ↵
    Bazemore, L. R., Folta-Stogniew, E., Takahashi, M. & Radding, C. M. RecA tests homology at both pairing and strand exchange. Proc Natl Acad Sci U S A 94, 11863–11868, doi:10.1073/pnas.94.22.11863 (1997).
    OpenUrlAbstract/FREE Full Text
  51. ↵
    Overballe-Petersen, S. et al. Bacterial natural transformation by highly fragmented and damaged DNA. Proc Natl Acad Sci U S A 110, 19860–19865, doi:10.1073/pnas.1315278110 (2013).
    OpenUrlAbstract/FREE Full Text
  52. ↵
    Levine, S. M. et al. Plastic cells and populations: DNA substrate characteristics in Helicobacter pylori transformation define a flexible but conservative system for genomic variation. FASEB J 21, 3458–3467, doi:10.1096/fj.07-8501com (2007).
    OpenUrlCrossRefPubMedWeb of Science
  53. ↵
    Rayssiguier, C., Thaler, D. S. & Radman, M. The barrier to recombination between Escherichia coli and Salmonella typhimurium is disrupted in mismatch-repair mutants. Nature 342, 396–401, doi:10.1038/342396a0 (1989).
    OpenUrlCrossRefPubMedWeb of Science
  54. ↵
    Matic, I., Rayssiguier, C. & Radman, M. Interspecies gene exchange in bacteria: the role of SOS and mismatch repair systems in evolution of species. Cell 80, 507–515 (1995).
    OpenUrlCrossRefPubMedWeb of Science
  55. ↵
    Majewski, J. & Cohan, F. M. The effect of mismatch repair and heteroduplex formation on sexual isolation in Bacillus. Genetics 148, 13–18 (1998).
    OpenUrlAbstract/FREE Full Text
  56. ↵
    Majewski, J., Zawadzki, P., Pickerill, P., Cohan, F. M. & Dowson, C. G. Barriers to genetic exchange between bacterial species: Streptococcus pneumoniae transformation. J Bacteriol 182, 1016–1023 (2000).
    OpenUrlAbstract/FREE Full Text
  57. ↵
    Brieba, L. G. Structure-Function Analysis Reveals the Singularity of Plant Mitochondrial DNA Replication Components: A Mosaic and Redundant System. Plants (Basel) 8, doi:10.3390/plants8120533 (2019).
    OpenUrlCrossRef
  58. ↵
    Valero-Rello, A., Lopez-Sanz, M., Quevedo-Olmos, A., Sorokin, A. & Ayora, S. Molecular Mechanisms That Contribute to Horizontal Transfer of Plasmids by the Bacteriophage SPP1. Front Microbiol 8, 1816, doi:10.3389/fmicb.2017.01816 (2017).
    OpenUrlCrossRef
  59. ↵
    Serrano, E. & Carrasco, B. Measurement of the Length of the Integrated Donor DNA during Bacillus subtilis Natural Chromosomal Transformation. Bio-Protocol 9, e3338, doi:10.21769/BioProtoc.3338 (2019).
    OpenUrlCrossRef
  60. Torres, R., Serrano, E., Tramm, K. & Alonso, J. C. Bacillus subtilis RadA/Sms contributes to chromosomal transformation and DNA repair in concert with RecA and circumvents replicative stress in concert with DisA. DNA Repair (Amst) 77, 45–57, doi:10.1016/j.dnarep.2019.03.002 (2019).
    OpenUrlCrossRef
Back to top
Posted May 13, 2020.
Download PDF

Supplementary Material

Email
Recombination proteins differently control the acquisition of homeologous DNA during Bacillus subtilis natural chromosomal transformation
Ester Serrano, Cristina Ramos, Juan C. Alonso, Silvia Ayora
bioRxiv 2020.05.13.090589; doi: https://doi.org/10.1101/2020.05.13.090589
Reddit logo Twitter logo Facebook logo LinkedIn logo Mendeley logo
Citation Tools

Subject Area