![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Previous Article | Next Article 
Applied and Environmental Microbiology, July 2005, p. 3589-3598, Vol. 71, No. 7
0099-2240/05/$08.00+0 doi:10.1128/AEM.71.7.3589-3598.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Department of Life Science, Fu Jen University, Taipei, Taiwan, Republic of China
Received 1 October 2004/ Accepted 20 January 2005
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The LexA repressor of X. axonopodis pv. citri is a protein of 213 amino acids which shares a high degree of homology with the LexA proteins of other gram-negative bacteria (31). The amino acid sequence of the domain important for the RecA-induced cleavage of LexA in E. coli is conserved in the X. axonopodis pv. citri LexA protein, suggesting that X. axonopodis pv. citri LexA is proteolytically cleaved, similar to E. coli LexA. The recA gene, which is located adjacent to the 3' end of lexA in X. axonopodis pv. citri, encodes a 37-kDa protein (34). Another gene, located immediately downstream of recA, is called recX, but its function is unknown (32).
Many lexA-like genes of various bacteria have been identified and well characterized (7, 8, 9, 12, 13, 15, 19, 20). In E. coli, LexA suppresses the expression of the SOS network under normal growth conditions by binding to the SOS box (11, 22, 28). In Bacillus subtilis, DinR binds to the DinR box and carries out the same function as E. coli LexA (3, 18, 29). Surprisingly, the sequence of the X. axonopodis pv. citri LexA binding site (TTAGN11CTAA) is different from that of the SOS box (CTGTN8ACAG) or the DinR box (CGAACRNRYGTTYC) (33).
We previously created a lexA mutant (XLE22) of X. axonopodis pv. citri (31). In this mutant, RecA production was still inducible by mitomycin C treatment, suggesting the existence of additional genes that also regulate recA expression in X. axonopodis pv. citri. The entire genome of X. axonopodis pv. citri has been sequenced (4). During an analysis of the genomic sequence of X. axonopodis pv. citri, we found another lexA-like gene at a different location from the region containing the lexA, recA, and recX genes. For this study, we characterized this novel lexA-like gene and found that it has functions similar to those of the previously identified lexA gene of X. axonopodis pv. citri.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(Invitrogen, Carlsbad, Calif.). E. coli strain ER2566 (New England Biolabs, Beverly, Mass.) was used for the overexpression of X. axonopodis pv. citri LexA2. Alignments of deduced sequences of various LexA proteins were performed with CLUSTAL W (24). The aligned sequence files thus generated were imported into GeneDoc (16) to determine sequence identities (Fig. 1; see Table 2). PCRs were performed by the use of Taq DNA polymerase (Invitrogen, Carlsbad, Calif.). The genomic DNA and total RNA of X. axonopodis pv. citri were isolated as described previously (30). Southern and Northern hybridizations were performed by using DNA probes labeled with digoxigenin (DIG) or 32P by random priming. The sensitivity of X. axonopodis pv. citri to mitomycin C (MMC) and methylmethane sulfonate (MMS) was determined by plating bacterial cells on tryptic soy broth (TSB) agar plates containing various concentrations of these agents and incubating the cells overnight as described previously (30). The survival rate was calculated by dividing the CFU/ml value for each treated sample by that for an untreated sample, as described previously (34).
|
|
The plasmid pLexA1 (31), which contains the lexA1 gene of X. axonopodis pv. citri, was digested with XbaI and XhoI, and the resulting 630-bp fragment containing the entire lexA1 gene was ligated between the XbaI and XhoI sites of pBluescript SK(+), generating pBC630. An 850-bp EcoRI DNA fragment containing the gentamicin resistance gene was isolated from pUC4G (Pharmacia, Piscataway, N.J.) and then inserted into the internal EcoRV site (nucleotide 531) of X. axonopodis pv. citri lexA1 in pBC630, generating pLexA1G, which was used to create the lexA1 mutant.
Generation of lexA1, lexA2, and lexA1 lexA2 mutants.
To investigate the role of LexA2 in the response to DNA damage, we constructed lexA1, lexA2, and lexA1 lexA2 mutants by homologous recombination. The plasmids pLexA1G and pLexA2K were separately introduced into the wild-type X. axonopodis pv. citri strain XW47 by electroporation, and gentamicin- or kanamycin-resistant cells were selected. Since pLexA1G and pLexA2K cannot replicate in X. axonopodis pv. citri, cells that become gentamicin or kanamycin resistant must have the gentamicin or kanamycin resistance gene integrated into the chromosome by homologous recombination between the lexA1 gene carried by pLexA1G or the lexA2 gene carried by pLexA2K and that carried on the chromosome, resulting in the inactivation of the genomic lexA1 or lexA2 gene. A single crossover between the lexA1 or lexA2 gene carried by pLexA1G or pLexA2K and that carried on the chromosome would result in an insertional inactivation of the chromosomal gene and the integration of the entire plasmid, including the chloramphenicol resistance gene. In contrast, a double crossover between the chromosomal lexA1 or lexA2 gene and the lexA1 or lexA2 sequences flanking the gentamicin or kanamycin resistance gene in the plasmid would result in the replacement of the chromosomal lexA1 or lexA2 gene with the gentamicin or kanamycin resistance gene; the resulting mutants would thus be gentamicin or kanamycin resistant but chloramphenicol sensitive. Southern blot analyses were performed, and the results confirmed that the gentamicin or kanamycin resistance gene cassette had integrated into the chromosomes of these mutants by the double-crossover event described above. The lexA1 mutant thus generated was named XLN-A1, and the lexA2 mutant was named XLN-A2. To construct the lexA1 lexA2 double mutant, we introduced the pLexA1G plasmid into the lexA2 mutant XLN-A2, and cells that were resistant to both gentamicin and kanamycin were selected and characterized. The lexA1 lexA2 double mutant was named XLN-A1A2. PCRs using primer pairs (UL1-LL1 and ULA2-LLA2) that amplified the internal regions of lexA1 (nucleotides –2 to 646 relative to the initiation codon) and lexA2 (nucleotides –7 to 610 relative to the initiation codon) were then performed, and the results indicated that the lexA1 and/or lexA2 gene was indeed deleted from the lexA mutants.
Expression and purification of X. axonopodis pv. citri LexA2.
To express the LexA2 protein in E. coli, we amplified the entire coding region of lexA2 by PCR, using the primers ULA2 and LLA2 described above. NcoI and XhoI restriction sites were built into the 5' ends of primers ULA2 and LLA2, respectively, to simplify cloning. The PCR products were digested with NcoI and XhoI, and the resulting 600-bp fragment was cloned between the NcoI and XhoI sites in pTYB4 (New England Biolabs, Beverly, Mass.), thus fusing LexA2 to the intein tag at the C-terminal end of LexA2. The recombinant lexA2-intein gene was driven by the T7 promoter and regulated by the lac operator, and the recombinant plasmid was named pLexA2E.
pLexA2E was introduced into the lexA-deficient E. coli strain DM1187 (14). Cells containing pLexA2E were grown in 1 liter of TSB to an optical density at 600 nm of 0.6 and then induced with IPTG (isopropyl-ß-D-thiogalactopyranoside) at a final concentration of 1 mg/ml. After 60 min of IPTG induction, the cells were pelleted, resuspended in 5 ml of column buffer, and then sonicated. The LexA2-intein fusion protein in the cell lysate was then purified by use of the IMPACT (intein-mediated purification with an affinity chitin-binding tag) system (New England Biolabs, Beverly, Mass.). Since the intein tag contained a chitin-binding domain, the cell lysate was loaded onto a column packed with chitin beads. After being washed with column buffer, the LexA2 protein was dissociated from intein by elution from the column with 30 µM dithiothreitol, which caused the intein tag to undergo autocatalytic cleavage (31), releasing the LexA2 protein (23 kDa) from the chitin beads. The LexA2 protein thus purified was electrophoresed in a 10% sodium dodecyl sulfate-polyacrylamide gel, and only one band, of 23 kDa, was seen.
Detection of LexA1, LexA2, and RecA proteins in X. axonopodis pv. citri.
Cell lysates (each containing 30 µg of protein) prepared from X. axonopodis pv. citri were electrophoresed in a 10% sodium dodecyl sulfate-polyacrylamide gel. The separated proteins were transferred to a nitrocellulose membrane and subjected to Western blot analysis using rabbit polyclonal antibodies against E. coli RecA and X. axonopodis pv. citri LexA1 (35) and LexA2. The anti-LexA2 antibody was produced in BALB/c mice by injecting 40 µg each of purified LexA2 and Freund's complete adjuvant three times (3 weeks apart). Ten days after the third injection, the mice were bled and antisera were collected. Immune complexes on Western blots were detected with horseradish peroxidase-conjugated goat anti-mouse secondary antibodies (Jackson ImmunoResearch Laboratories, Inc., West Grove, Pa.) and an enhanced chemiluminescence substrate (EZ-Link kit; Pierce, Rockford, Ill.). The intensities of bands representing immunoreactive proteins were quantified by densitometry as described previously (33).
Preparation of lexA2-luxAB fusion constructs and measurement of bioluminescence.
A 390-bp DNA fragment containing 256 bp upstream and 134 bp downstream of the initiation codon of X. axonopodis pv. citri lexA2 was generated by a PCR using the primers ULA2-1 (5'-GCTCTAGATCTTGAGTTTCATGGCG-3') and LLA1 (5'-CACAAGCTTTTGTGCGGCATTGCG-3'). A shorter (205 bp) DNA fragment containing 71 bp upstream and 134 bp downstream of the initiation codon of X. axonopodis pv. citri lexA2 was also generated by a PCR using the primers ULA2-2 (5'-GCTCTAGAAGCCAGTCTCGCC-3') and LLA1. These two fragments were separately cloned into the XbaI and HindIII sites of the promoterless luciferase reporter plasmid pMY3 (27), generating pLP256 and pLP71, respectively. These two plasmids were introduced into the wild type and the lexA mutants of X. axonopodis pv. citri to investigate the lexA2 promoter activity in these hosts. The cells were grown in TSB for 16 h. One hundred microliters of n-decylaldehyde (0.1% suspension in ethanol) was then added to 500 µl of each culture. The bioluminescence thus generated was measured over three 10-s intervals with a luminometer (LB953 AutoLumat; EG&G Berthold, Bad Wildbad, Germany). Luminescence was expressed in relative light units (RLU).
EMSAs.
The LexA1 and LexA2 proteins were purified from lysates of E. coli (lexA mutant strain DM1187) cells harboring pLexA1E, which was formerly called pLE630 (31), and pLexA2E, respectively. The DNA probes used for binding analyses were generated by PCR and consisted of 195-bp (nucleotides –40 to +155), 205-bp (nucleotides –71 to +134), and 209-bp (–204 to +5 relative to the translation start site) fragments from the promoter regions of X. axonopodis pv. citri lexA1 (33), lexA2, and recA (30), respectively. Electrophoretic mobility shift assays (EMSAs) were performed by use of a DIG gel shift kit (Roche Applied Science, Indianapolis, Ind.). The DNA probes were labeled at their 3' ends with DIG-11-ddUTP by the use of terminal transferase. Each EMSA reaction mixture (10 µl), which contained 0.75 ng of DIG-labeled probe and various amounts of LexA in binding buffer (50 mM NaCl, 20 mM Tris-HCl [pH 7.5], 1 mM dithiothreitol, and bovine serum albumin [100 µg/ml]), was incubated at room temperature for 15 min and then mixed with 2.5 µl of loading buffer (0.25x Tris-borate-EDTA buffer, 40% glycerol, 0.2% [wt/vol] bromophenol blue). The entire mixture was loaded into a native 5% polyacrylamide gel (acrylamide:bisacrylamide ratio, 29:1 [wt/wt]). A Mini-PROTEAN II dual-slab cell electrophoresis apparatus (Bio-Rad, Hercules, Calif.) was used. Electrophoresis was carried out with 0.5x Tris-borate-EDTA buffer (pH 8) at 64 V at room temperature for 2 h, after which the gel was blotted onto a positively charged nylon membrane by use of a Mini Trans-Blot (Bio-Rad, Hercules, Calif.) apparatus at 30 V for 30 min. The membrane was then soaked in 2x SSC (1x SSC is 0.15 M NaCl plus 0.015 M sodium citrate) for 1 min and cross-linked at 120 mJ in a UV cross-linker (Stratalinker; Stratagene, La Jolla, Calif.) for 1 min. The membrane was exposed to X-ray film for 15 min at room temperature. For binding competition experiments, excess unlabeled competitor DNA was included in the reaction mixture.
Statistics.
Determinations of significant differences between two samples were performed by Student's t test. The SigmaStat (Jandel Scientific, San Rafael, Calif.) software package was used to make the calculations.
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1, 2, and 3 [amino acid positions 9 to 54 of LexA2]) with the potential to form a helix-turn-helix structure was localized in the N-terminal region of the deduced amino acid sequence. The autocleavage site, Ala-Gly, which is conserved in all LexA-like proteins, was found at amino acid positions 87 and 88 of this putative LexA protein. A serine residue and a lysine residue that are usually spaced 37 amino acids apart and are also required for the autocleavage of LexA were located at amino acid positions 122 and 159, respectively. These characteristics suggest that this ORF encodes a LexA-like protein.
|
|
Effects of MMC treatment on production of RecA, LexA1, and LexA2 in the wild type and in lexA mutants.
The production of RecA, LexA1, and LexA2 in the wild type and the three lexA mutants was examined next. Cells were grown in 200 ml of TSB overnight and then pelleted. The cell pellet was resuspended and washed three times with 100 ml of fresh TSB. After the final wash, the cells were incubated in 10 ml of TSB containing 0.1 µg/ml of mitomycin C at 28°C for 2 h. An aliquot of the same cells was incubated in TSB without mitomycin C to serve as the control. The cells were then pelleted, resuspended in 0.5 ml of Tris-EDTA buffer, and lysed in sample buffer at 95°C for 10 min. The protein concentration of each cell lysate was determined, and 30 µg of protein from each sample was analyzed by Western blotting using polyclonal antibodies against LexA1, LexA2, and the E. coli RecA protein, which shares significant homology with the RecA protein of X. axonopodis pv. citri (25).
As expected, the 37-kDa RecA protein was detected in the lysates of wild-type XW47 cells (Fig. 3A, lane 1), and the amount of RecA was profoundly increased (10-fold) upon exposure of the cells to mitomycin C (Fig. 3A, lane 2). The same pattern of recA induction by MMC was observed for the lexA1 (Fig. 3A, lanes 3 and 4), lexA2 (Fig. 3A, lanes 5 and 6), and lexA1 lexA2 (Fig. 3A, lanes 7 and 8) mutants. However, the increase in the RecA level in these mutants after MMC induction was not as large as that in the wild type. The RecA concentration increased eightfold in the lexA1 mutant, ninefold in the lexA2 mutant, and fivefold in the lexA1 lexA2 double mutant in response to MMC treatment. No RecA protein was detected in the recA mutant strain XCK75 (Fig. 3A, lane 9).
|
The LexA2 protein was detected only in the wild type (Fig. 3C, lane 1) and the lexA1 mutant (Fig. 3C, lane 3) without MMC treatment (Fig. 3C, lane 1). No LexA2 was detected in these two strains (Fig. 3C, lanes 2 and 4) after MMC treatment, indicating that MMC treatment causes LexA2 to undergo autocleavage. As expected, the lexA2 (Fig. 3C, lanes 5 and 6) and lexA1 lexA2 (Fig. 3C, lanes 7 and 8) mutants produced no LexA2 protein before or after MMC treatment, indicating that the lexA2 gene was indeed mutated in these mutants. The production of LexA2 in the recA mutant strain XCK75 was found to be normal (Fig. 3C, lane 9).
Effects of LexA1 and LexA2 on transcription of the recA gene.
The transcription of the recA gene in the wild type and the three lexA mutants was also examined to determine the effects of LexA1 and LexA2 on recA expression. The cells were grown and treated with MMC or left untreated as described above. The total cellular RNA (5 µg) from each culture was isolated and then analyzed by Northern hybridization with a 32P-labeled recA gene probe which was isolated from pXC560 (34) as a 560-bp EcoRI-HindIII fragment. A recA-deficient mutant (XCK75) was also analyzed as a control. The 1.2-kb recA mRNA was detected in all cultures, with or without MMC treatment, except that of the recA mutant. MMC treatment caused a 19-fold increase in recA transcription in the wild-type strain XW47 and a 20-fold, 15-fold, and 16-fold increase in transcription in the lexA1, lexA2, and lexA1 lexA2 mutants, respectively (Fig. 4). Without MMC induction, recA expression in the lexA1, lexA2, and lexA1 lexA2 mutants was 0.98-, 0.78-, and 0.61-fold, respectively, that in XW47.
|
The basal levels (before mitomycin C treatment) of lexA2 promoter activity in the wild type, the lexA1 mutant, the lexA2 mutant, and the lexA1 lexA2 double mutant were first determined by measuring the luciferase activity in transformed cells. The luciferase activity derived from the lexA2 promoter in the wild type was 29.2 ± 1.5 RLU. A similar level (12.6 ± 1.0 RLU) of lexA2 promoter activity was observed for the lexA1 mutant. A 2.4-fold higher lexA2 promoter activity was observed for the lexA2 mutant than for the wild type (70.8 ± 2.2 versus 29.2 ± 1.5 RLU; P < 0.001), suggesting that LexA2 suppresses its own expression. Interestingly, the lexA2 promoter was much more active in the lexA1 lexA2 double mutant than in the lexA1 mutant. A 6.8-fold (197.2 ± 5.5 versus 29.2 ± 1.5 RLU; P < 0.001) higher lexA2 promoter activity was observed for the double mutant than for the wild type (Table 2), suggesting that LexA1 has some effect on the expression of the lexA2 gene.
After 1 h of mitomycin C (0.1 µg/ml) treatment, the lexA2 promoter activity in the wild type was increased 2.0-fold (from 29.2 ± 1.5 to 58.6 ± 2.3 RLU; P < 0.001). A similar increase (2.7-fold; 33.9 ± 2.1 versus 12.6 ± 1.0 RLU; P < 0.001) in lexA2 promoter activity was observed for the lexA1 mutant upon mitomycin C induction. However, no significant increase in lexA2 promoter activity was observed for both the lexA2 mutant (1.4-fold; 70.8 ± 2.2 versus 100.7 ± 4.3 RLU; P = 0.02) and the lexA1 lexA2 double mutant (1.1-fold; 197.2 ± 5.5 versus 224.8 ± 6.2 RLU; P = 0.04) after mitomycin C treatment (Table 2); this result further supports the notion that LexA2 regulates its own expression and that without LexA2 suppression, lexA2 expression becomes constitutive and is no longer responsive to mitomycin C induction.
Effect of LexA2 on lexA1 gene expression.
To investigate whether LexA2 regulates the expression of lexA1, we transformed pLEP171 (35), which contained the lexA1 promoter fused to the luxAB gene, into the wild type and the lexA1, lexA2, and lexA1 lexA2 mutants and then assayed them for luciferase activity (Table 3). A higher basal level of luciferase activity was observed for the wild type harboring pLEP171 (42.5 ± 3.0 RLU) than for the wild type harboring pLP71 (29.2 ± 1.5 RLU) (Tables 2 and 3), suggesting that the lexA1 promoter is more active than the lexA2 promoter. A dramatic increase (from 42.5 ± 3.0 to 262.6 ± 12.0 RLU; P < 0.001) in lexA1 promoter activity was seen for the lexA1 mutant, indicating that LexA1 suppresses its own expression, as expected. The activity of the lexA1 gene promoter was found to be the same in the wild type (42.5 ± 3.0 RLU) as in the lexA2 mutant (40.7 ± 2.8 RLU), suggesting that LexA2 does not regulate lexA1 gene expression. In addition, the lexA1 promoter was found to be as active in the lexA1 lexA2 double mutant (258.4 ± 12.5 RLU) as in the lexA1 mutant (262.6 ± 12.0 RLU; P = 0.04). This result also supports the notion that LexA2 has no effect on lexA1 gene expression. Treatment with mitomycin C increased the lexA1 promoter activity in the wild type (from 42.5 ± 3.0 to 106.5 ± 4.5 RLU; 2.5-fold; P < 0.001) and the lexA2 mutant (from 40.7 ± 2.8 to 92.7 ± 3.5 RLU; 2.3-fold; P < 0.001), as expected. No increase in lexA1 promoter activity after MMC treatment was observed for the lexA1 mutant (262.6 ± 12.0 RLU before and 261.3 ± 12.1 RLU after treatment; P = 0.01) and the lexA1 lexA2 double mutant (258.4 ± 12.5 RLU before and 268.6 ± 13.2 RLU after treatment; P = 0.04), indicating that a loss of LexA1 production renders the lexA1 gene unable to respond to mitomycin C induction.
|
Incubation of the purified LexA1 protein with the lexA1 promoter resulted in a mobility shift of the DNA fragment in the gel (Fig. 5, lane 2). The presence of an unrelated DNA, pBC, in the binding reaction had no effect on this binding (Fig. 5, lane 3). No mobility shift was observed when LexA1 was incubated with the lexA2 promoter fragment (Fig. 5, lane 8), indicating that LexA1 does not bind to the lexA2 promoter. The LexA2 protein was found to bind the lexA2 promoter as expected (Fig. 5, lane 6), but did not bind to the lexA1 promoter (Fig. 5, lane 4). The binding of LexA2 to the lexA2 promoter was not affected by the presence of the unrelated DNA pBC (Fig. 5, lane 7).
|
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The lexA2 gene of X. axonopodis pv. citri also has functions that are characteristic of a lexA gene. The LexA protein is the major regulator of the SOS system in E. coli and several other bacteria that have been studied. In these bacteria, LexA binds to the SOS box and suppresses the expression of the SOS system under normal conditions. When a DNA damage signal is sensed by the bacterium, the RecA protein is activated and becomes a coprotease, causing LexA to autocleave; therefore, the SOS system is activated to produce enzymes required for DNA repair, such as the UvrABCD endonuclease, the UmuCD protein, and SulA (25). Mitomycin C (MMC) and methylmethane sulfonate (MMS) are two DNA-damaging agents which are commonly used to investigate the SOS response. The treatment of bacterial cells with MMC or MMS causes cells to die due to DNA damage. In response, cells activate the SOS system to repair the damage in order to survive. In E. coli, LexA suppresses the SOS system; inactivation of the lexA gene renders the SOS system constitutively expressed. Therefore, lexA mutants of bacteria such as E. coli are usually more resistant to MMC or MMS. In this study, we found that the inactivation of the lexA2 gene did indeed cause X. axonopodis pv. citri to become more (4 orders of magnitude) resistant to MMC at a concentration of 0.1 µg/ml. The same results were observed when the lexA gene or both the lexA1 and lexA2 genes were mutated (Fig. 3A). However, at higher concentrations (0.3 or 0.5 µg/ml) of MMC, lexA1, lexA2, and lexA1 lexA2 mutants were as sensitive as the wild type, suggesting that the degree of DNA damage caused by such high concentrations of MMC is beyond the capacity of the SOS repair system in X. axonopodis pv. citri.
Treatment of the wild type and the lexA mutants of X. axonopodis pv. citri with MMS revealed more interesting features. Both the lexA1 and lexA2 single mutants responded the same way to MMS treatment and were 1 order of magnitude more resistant to 30 µg/ml of MMS than the wild type, whereas the lexA1 lexA2 double mutant was 2 orders of magnitude more resistant to MMS at the same concentration. At this concentration (30 µg/ml), the lexA1 lexA2 double mutant was completely resistant to MMS (Fig. 3B). A similar pattern of response, with the lexA1 lexA2 double mutant being more resistant to MMS than the lexA1 or lexA2 single mutant, was also observed at a higher concentration of MMS (45 µg/ml). These results suggest that the lexA1 and lexA2 genes play different roles in SOS repair in response to MMS treatment.
MMC causes cross-linking of DNA bases. This type of damage is repaired by the combined action of homologous recombination and nucleotide excision repair pathways (25); both of these repair systems require the RecA protein (26). The observation that all lexA mutants responded to MMC treatment similarly suggests that lexA1 and lexA2 play a similar role in repairing the damage caused by MMC. Therefore, the inactivation of either one or both lexA genes results in the same pattern of response to MMC. MMS causes DNA base methylation, which is repaired by transmethylases as well as base excision and nucleotide excision mechanisms (25). Therefore, both RecA and transmethylases are required to repair the damage. The observation that the lexA1 and lexA2 single mutants were more sensitive to MMS than the lexA1 lexA2 double mutant suggests that lexA1 and lexA2 play different roles in suppressing the activity or production of the transmethylases that are involved in repairing the DNA damage caused by MMS.
The responsiveness of the lexA2 gene to MMC treatment was also confirmed by the use of promoter assays. The lexA2 promoter was found to be located within the region 71 bp upstream of the translation initiation codon (Table 2). This promoter region was able to bind LexA2 (Fig. 5), suggesting that LexA2 autoregulates its own expression. MMC treatment may cause LexA2 to undergo autocleavage, thus increasing the transcription of the lexA2 gene (Table 2). LexA1 appears to have some effect on the expression of the lexA2 gene, as the lexA2 gene promoter was more active in the lexA1 lexA2 double mutant (Table 2). This observation is consistent with the finding of the sequence TTAGTACTAAAGTTATAA at nucleotides –134 to –114 relative to the translation start site of lexA2; this sequence is homologous to the LexA1 binding sequence, which has been determined to be TTAGN11CTAA (33). Whether LexA1 binds to this sequence remains to be determined. Another sequence, GGTGTACAAATGTACACC, located in the 5' untranslated region of the lexA2 gene (nucleotides –43 to –26 relative to the translation start site), also has a perfect inverted repeat and may be bound by a certain transcription regulator. LexA2 appears to have no effect on lexA1 expression, since the lexA1 promoter was equally active in the wild type and the lexA2 mutant. The lexA1 and lexA2 genes do not appear to affect each other in the responsiveness of X. axonopodis pv. citri to MMC induction since the lexA2 promoter did not respond to MMC treatment in the lexA2 mutant, although it produced normal amounts of the LexA1 protein (Table 2). Similarly, the lexA1 promoter did not respond to MMC in the lexA1 mutant, which produced normal amounts of the LexA2 protein (Table 3).
The results from the EMSA experiments indicate that LexA2 binds to the recA gene promoter in addition to its own gene promoter. However, LexA2 does not appear to regulate recA expression, as the production of both the RecA protein (Fig. 3) and recA mRNA (Fig. 4) in the LexA2 mutant was still inducible by MMC treatment to the same degree as in wild-type X. axonopodis pv. citri. This is in stark contrast to the LexA protein in E. coli, in which a lexA mutation would render the expression of the recA gene constitutive (25). Surprisingly, similar results for recA expression were observed for both the lexA1 and lexA1 lexA2 mutants of X. axonopodis pv. citri. RecA expression in these mutants was the same as that in the wild type. These results suggest that both LexA1 and LexA2 do not regulate recA expression under native conditions. It is possible that the amounts of LexA1 and LexA2 in the cell are not sufficient to bind the recA promoter in vivo, as we found that 0.9 µg of LexA1 or 1.0 µg of LexA2 was required to achieve a positive EMSA result for the binding of these proteins to the recA promoter and that <100 ng of LexA1 or 60 ng of LexA2 was estimated to be present before MMC induction in a cell, based on our immunoblot results. This notion is consistent with our previous result indicating that the overexpression of X. axonopodis pv. citri LexA1 in E. coli reduced the abundance of E. coli RecA (35); this observation led us to conclude previously that LexA1 may regulate the expression of X. axonopodis pv. citri recA (35). Another possibility for why LexA1 or LexA2 binds to the recA promoter but does not affect recA expression is that other regulators also bind the recA promoter and either work alone or in conjunction with LexA1 or LexA2 to regulate the recA promoter. Therefore, the inactivation of LexA1 and LexA2 has no significant effect on the expression of recA. One such factor may be the product of the recX gene, since the recX genes of Pseudomonas aeruginosa (21) and Xanthomonas oryzae pv. oryzae (23) have been shown to regulate recA expression. It is possible that the recX gene of X. axonopodis pv. citri has a similar function.
The existence of two lexA genes in an organism has also been found for Geobacter sulfurreducens, which is a delta-proteobacterium that oxidizes organic compounds by using Fe3+ as the terminal electron acceptor (10). Similar to the case for X. axonopodis pv. citri, the two lexA genes of Geobacter sulfurreducens also regulate their own expression and do not regulate RecA production (10). The lack of an effect of LexA on recA expression has also been observed in other bacteria. The expression of the recA gene of Deinococcus radiodurans is induced by DNA damage but is not regulated by LexA (1). Myxococcus xanthus was found to have two recA genes, recA1 and recA2, but only recA2 is repressed by LexA (2). The recA gene of Mycobacterium tuberculosis was found to be driven by two different promoters. One is regulated by RecA and LexA, like that seen with E. coli recA, but the activity of the other promoter is not affected by either RecA or LexA (5, 17). It is possible that these recA genes are regulated by non-LexA regulators that have yet to be identified or that the LexA proteins of these bacteria affect the activity of the RecA protein but not its expression.
|
ACKNOWLEDGMENTS |
|---|
This study was supported by a grant (NSC91-2311-B-030-001) from the National Science Council, Taiwan, Republic of China.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| J. Bacteriol. | Microbiol. Mol. Biol. Rev. | Eukaryot. Cell | All ASM Journals |
|---|
