Evidence for involvement of UvrB in elicitation of ‘SIR’ phenotype by rpoB87-gyrA87 mutations in lexA3 mutant of Escherichia coli

DNA Repair 11 (2012) 915–925

Contents lists available at SciVerse ScienceDirect

DNA Repair journal homepage: www.elsevier.com/locate/dnarepair

Evidence for involvement of UvrB in elicitation of ‘SIR’ phenotype by rpoB87-gyrA87 mutations in lexA3 mutant of Escherichia coli V. Shanmughapriya 1 , M. Hussain Munavar ∗ Department of Molecular Biology, School of Biological Sciences, Centre for Excellence in Genomic Sciences, Madurai Kamaraj University (University with Potential for Excellence), Madurai 625021, Tamil Nadu, India

a r t i c l e

i n f o

Article history: Received 27 February 2012 Received in revised form 16 July 2012 Accepted 11 September 2012 Available online 9 October 2012 Keywords: Escherichia coli rpoB87 gyrA87 uvrB SIR Mitomycin C SOS

a b s t r a c t An unconventional DNA repair termed SIR (SOS Independent Repair), specific to mitomycin C (MMC) damage elicited by a combination of specific RifR (rpoB87) and NalR (gyrA87) mutations in SOS un-inducible strains of Escherichia coli was reported by Kumaresan and Jayaraman (1988). We report here that the rpoB87 mutation defines a C1565 →T1565 transition changing S522 →F522 and gyrA87 defines a G244 →A244 transition changing D82 →N82 . The reconstructed lexA3 rpoB87 gyrA87 strain (DM49RN) exhibited resistance to MMC but not to UV as expected. When mutations in several genes implicated in SOS/NER were introduced into DM49RN strain, uvrB mutation alone decreased the MMC resistance and suppressed SIR phenotype. This was alleviated about two fold by a plasmid clone bearing the uvrB+ allele. Neither SulA activity as measured based on filamentation and sulA::gfp fluorescence analyses nor the transcript levels of sulA as seen based on RT-PCR analyses indicate a change in sulA expression in DM49RN strain. However, uvrB transcript levels are increased with or without MMC treatment in the same strain. While the presence of lexA3 allele in a plasmid clone was found to markedly decrease the MMC resistance of the DM49RN strain, the additional presence of uvrB+ allele in the same clone alleviated the suppression of MMC resistance by lexA3 allele to a considerable extent. These results indicate the increased expression of uvrB in the DM49RN strain is probably from the LexA dependent promoter of uvrB. The sequence analyses of various uvrB mutants including those isolated in this study using localized mutagenesis indicate the involvement of the nucleotide phosphate binding domain (ATPase domain) and the ATP binding domain and/or the DNA binding domain of the UvrB protein in the MMC repair in DM49RN. The possible involvement of UvrB protein in the MMC damage repair in DM49RN strain in relation to DNA repair is discussed. © 2012 Elsevier B.V. All rights reserved.

1. Introduction The DNA repair mechanisms in Escherichia coli that have been well understood and documented can be classified as the RecA independent, such as the ada response [1] or the RecA dependent [2] SOS response. It is well known and has been established over the years that the SOS mediated inducible repair system is under the control of the recA and the lexA genes [2]. On contact with single stranded DNA the RecA protein gets activated and mediates self-cleavage by LexA auto-protease [3,4]. This cleavage of LexA repressor results in activation of a cascade of genes. During initial stage of DNA damage, a set of genes including uvrA, uvrB, etc.,

Abbreviations: MMC, mitomycin C; Rif, rifampicin; Nal, nalidixic acid; SIR, SOS Independent Repair; tet, tetracycline; kan, kanamycin; cfu, colony forming units. ∗ Corresponding author. Tel.: +91 452 2458210; fax: +91 452 2458210. E-mail address: [email protected] (M.H. Munavar). 1 Tel.: +91 452 2458210; fax: +91 452 2458210. 1568-7864/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.dnarep.2012.09.005

involved in nucleotide excision repair (NER) are expressed [4,5]. When this does not suffice, a second set of genes under tight LexA control including the umuD, umuC, dinB, sulA, etc., are expressed which give rise to mutagenic repair [6,7]. Mutations in recA or lexA have been known to render the cell hypersensitive to DNA damaging agents. While the SOS mediated repair is general to many kinds of lesions [5], the other mechanisms identified are specific to certain specific types of lesions. The ada response is specific for methylated bases of purines or pyrimidines [8]. The repair of methylation at oxygen residues, O6 -MeG and O4 -MeT, caused by SN 1 agents such as N-methyl-N -nitro-N-nitrosoguanidine (MNNG) and N-Methylnitrosourea (MNU) is mediated by the ada gene protein, an O6 -meG-DNA methyltransferase [9–11]. Another lesion specific SOS independent response reported is the UVM (UV Modulation of Mutagenesis) [12]. The UVM response is specific to 3,N4 -ethenocytosine (␧C) [13] or ␧A [14] lesions caused by extrinsic carcinogens such as vinyl chloride and ethyl carbamate [15]. Mitomycin C (MMC) is a highly potent DNA damaging agent which causes DNA crosslinks and it has been shown that a single

916

V. Shanmughapriya, M.H. Munavar / DNA Repair 11 (2012) 915–925

crosslink per genome is sufficient to cause the death of a bacterial cell [16,17]. MMC, after reduction into vinylogous quinone methide, causes two N-alkylations in specific 5 -CpG-3 regions in DNA which causes crosslinks [18]. Mitomycin C has been shown to be a potent inducer of the SOS response and the damage is repaired by the UvrABC dependent nucleotide excision repair [16]. The UvrB protein of the NER has been shown to play a major role in this as it interacts with all the components of the excision repair like the UvrA, UvrC, UvrD (helicase), DNA polymerase I and damaged DNA [19,20]. During the process of excision repair the UvrA and UvrB form a complex together which detects damaged DNA and using helicase activity inserts the ␤ hairpin loop into the DNA helix followed by the release of UvrA to give the preincision complex [21]. UvrC then binds to the UvrB–DNA complex and causes incisions first at the 4th to 5th base in the 3 side followed by a 5 incision at 8th base upstream [22,23]. In related studies, mutant forms of UvrB have been shown to recognize damaged DNA on its own without UvrA [24] and homologue for the UvrC protein has also been identified named Cho [25]. Kumaresan and Jayaraman in 1988 [26] reported an unconventional DNA repair specific to MMC damage in SOS un-inducible strains and this repair was found to be elicited due to a specific combination of RifR (rpoB87) and NalR (gyrA87) mutations. The fact that this repair is seen in SOS uninducible strains led them to postulate that this repair could potentially be an unconventional SOS Independent Repair and hence the phenotype was termed as ‘SIR’. MMC is believed to inhibit DNA synthesis and cause degradation of available DNA. Thus DNA degradation assays with rpoB87 gyrA87 derivatives of lexA3 Ind− strains which are SOS un-inducible but SIR proficient were carried out to analyse the levels of DNA damage recovery. This revealed that the DNA degradation was decreased to a considerable extent in this SIR+ strain as compared to the parent lexA3 rpoB+ gyrA+ strain, DM49. Therefore, it can be said that the rpoB87 and gyrA87 mutations are able to overcome the MMC induced damage caused to DNA to significant levels. It was also shown that the SIR phenotype does not result due to deficiency in uptake of or reduction of mitomycin C because the prophage induction of mitomycin C in recA+ ␭ lysogens was independent of rpoB87 and gyrA87 mutations. Therefore, it was proposed that the elicitation of the SIR+ phenotype could perhaps be due to induction of a partially efficient DNA damage repair system/pathway by the combined effect of rpoB87 and gyrA87 mutations under SOS un-inducible conditions [26]. The elicitation of the proposed SIR phenotype is dependent on the mutant forms of ␤ subunit of RNAP and GyrA subunit of DNA Gyrase apart from the RecA protein and the product encoded by the locus named sir which was found to map between 57 and 61 min [27] and possibly other unidentified gene(s). In this investigation, we have taken efforts to testify the view of SIR phenotype and functions involved in the same. We have shown here that rpoB87 defines a C→T transition in the 522nd codon of rpoB and gyrA87 defines a G→A transition in the 82nd codon of gyrA. Attempts taken to see the effect of dinB::kan, umuDC::cat, uvrD::Tn5, uvrA::Tn10 and uvrB::Tn10kan mutations in the elicitation of SIR phenotype clearly reveal that functional UvrB is mandatory for this unconventional (SIR) DNA repair. The data reported herein further validate that sulA transcription is not at all altered in SIR proficient lexA3 rpoB87 gyrA87 strain but transcription of uvrB is increased irrespective of damage induction and might possibly be from the LexA dependent promoter of uvrB. We also believe our results indicate that the ATPase domain (domain 1a) and the ATP binding domain and/or the DNA binding domain (␤ hairpin) might play a role in the MMC resistance of the DM49RN strain.

2. Materials and methods 2.1. Bacterial strains used and construction Given in Table 1 is the list of bacterial strains used in this study. Genetic nomenclature is according to Demerec et al. [28]. All P1 mediated transductions were performed as described in Miller (1972, 1992) [29,30]. The relevant recipient and donor strains are also mentioned in Table 1. The rpoB87 mutation was transduced from JK10AB using linked argE+ marker and introduced into DM49 and the Arg+ transductants obtained on selective minimal plates lacking arginine were screened for Rif resistant colonies. One such Rif resistant Arg+ transductants was named DM49R and used for further experiments. The gyrA87 mutation from JK10AB was linked with Tetracycline resistant marker zfa723::Tn10 and the TetR transductants were screened for those transductants which retained gyrA87. This strain was named JK10ABT (lexA3 rpoB87 gyrA87 zfa723::Tn10). Using the P1 made on this strain, the recipient strains DM49 and DM49R were transduced for TetR colonies and the obtained TetR transductants were checked for acquisition of Nal resistance. One such TetR NalR transductants from DM49 and DM49R were named DM49N and DM49RN, respectively. In order to enable mobilization of uvrB mutation to desired strains, the uvrB point mutations were first linked with kanamycin resistance marker zbh3108::Tn10kan by transducing the KanR marker into relevant uvrB mutants and looked for UVS ones among the obtained KanR transductants. The uvrB45 mutation in the strain AB2421 could not be directly linked to zbh3108::Tn10kan by screening for UV sensitivity because of the presence of uvrA6 mutation which also affects UV sensitivity. Thus after transduction of zbh3108::Tn10kan marker into AB2421, P1 lysate made from few random transductants were transduced into the strain AB1157 and from those KanR transductants UV sensitive colonies were selected for further use. The respective Tn10kan linked uvrB mutants were then used as donor and the alleles were mobilized into the desired strains using P1 transduction. 2.2. Media and chemicals LB and minimal media [29] with appropriate supplements were used. Cells were routinely grown in LB at 37 ◦ C unless specified otherwise. Whenever required the following chemicals/antibiotics were added to the media in the final concentrations indicated. MMC (0.5 ␮g/ml), Rif (20 ␮g/ml), Nal (20 ␮g/ml), Tet (10 ␮g/ml), Kan (45 ␮g/ml), X-gal (30 ␮g/ml), and amino acids (30 ␮g/ml). The chemicals used were purchased from Himedia, India, Sigma, USA and Sisco Laboratories, India. MMC was purchased from Biobasic Inc., India. The primers used for the study were obtained from Chromous Biotech, Bangalore, India. The enzymes used were obtained from Fermentas, India. 2.3. Mitomycin C and UV survival assays Overnight cultures were sub-cultured into fresh LB broth and grown till mid-log phase (∼0.3 OD). Untreated samples were withdrawn at 0 min and were diluted. Appropriate dilutions were plated on LB plates for the cell titre and to calculate cfu/ml. MMC was added to the cultures to the final concentration of 0.5 ␮g/ml, the samples were withdrawn after 30 min and 60 min time intervals. The samples were then diluted and plated on appropriate LB plates. The cell titre (cfu/ml) value in each case was calculated based on the colony counts after 36 h incubation at 30 ◦ C. The percentage survival of each of these strains at different time intervals after MMC treatment was calculated by taking the cfu/ml at 0 min sample (MMC untreated) as 100%. Survivors at 30 min and 60 min after

V. Shanmughapriya, M.H. Munavar / DNA Repair 11 (2012) 915–925

917

Table 1 List of E. coli K12 strains used in this study, their relevant genotype and source. Strain

Relevant genotype

Source/reference/construction

AB1157

F− thr-1, araC14, leuB6(Am), (gpt-proA)62, lacY1, supE44, hisG4(Oc), rpoS396(Am), rpsL31(strR ), argE3(Oc), thi-1 Same as AB1157 but lexA3 Ind− Same as DM49 but argE+ rpoB87 hisG+ gyrA87 F− zfa-723::Tn10 rph-1 Same as JK10AB but zfa723::Tn10 Same as DM49 but zfa723::Tn10 gyrA87 Same as DM49 but argE+ rpoB87 Same as DM49N but argE+ rpoB87 F− , (araD-araB)567, dinB749::kan, F− , uvrA277::Tn10 F− , (araD-araB)567, lacZ4787(::rrnB-3), uvrB751::kan F− , glnV44(AS), hisG4(Oc), rpsL281(strR ), uvrD282::Tn5, metE46, argH1 F− , thr-1, araC14, leuB6(Am), glnV44(AS), galK2(Oc) (umuD-umuC)595::cat, hisG4(Oc), rpsL31(strR ), uvrA6 F− , thr-1, araC14, (gpt-proA)62, lacY1, − , uvrB29, rpsL31(strR) F− , − , zbh-3108::Tn10kan, rph-1 Same as DM49RN but dinB749::kan Same as DM49RN but uvrA277::Tn10 Same as DM49RN but uvrB751::kan Same as DM49RN but uvrD282::Tn5 Same as DM49RN but umuDC::cat Same as AB2402 but zbh-3108::Tn10kan Same as DM49RN but zbh-3108::Tn10kan, uvrB29 F− tsx-3100::Tn10kan rph-1 F− lon510 cpsB10::lac Same as SG20780 but tsx-3100::Tn10kan Same as AB1157 but tsx-3100::Tn10kan lon Same as DM49 but tsx-3100::Tn10kan lon Same as DM49RN but tsx-3100::Tn10kan lon F− att␭::sulApgfp-mut F− , zbh-3108::Tn10kan F− att␭::sulApgfp-mut,zbh-3108::Tn10kan Same as AB1157 but att␭::sulApgfp-mut,zbh-3108::Tn10kan Same as DM49 but att␭::sulApgfp-mut,zbh-3108::Tn10kan Same as DM49RN but att␭::sulApgfp-mut,zbh-3108::Tn10kan F− , thr-1, araC14, leuB6(Am), (gpt-proA)62, lacY1, tsx-33, − , uvrB45, hisG4(Oc), rpsL31(strR), xylA5, mtlA2, argE3(Oc), thi-1, uvrA6 F− , uvrB153, sulA22 F , thr-1, leuB6(Am), fhuA21, codA1, lacY1, glnV44(AS), − , uvrB501, pyrF32, thyA6, malA19, thi-1, deoC1, R „ F42 Same as AB2421 but zbh-3108::Tn10kan Same as BS-8 but zbh-3108::Tn10kan Same as GY854 but zbh-3108::Tn10kan Same as AB1157 but zbh-3108::Tn10kan uvrB45 Same as AB1157 but zbh-3108::Tn10kan uvrB153 Same as AB1157 but zbh-3108::Tn10kan uvrB501

Laboratory collection

DM49 JK10AB CAG12178 gyrA+ JK10ABT DM49N DM49R DM49RN JW0221-1 N3055 JW0762-2 SK3451 RW82 AB2402 CAG18531 49RNDB 49RNUA 49RNUB 49RNUD 49RNDC MS29K 49RN29 CAG18413 SG20780 MS780LK MS57LK MS49LK MS49RNLK SS996 CAG18531 SS996K AB1157sgK DM49sgK DM49RNsgK AB2421

BS-8 GY854 MS45K MS153K MS501K MSV45 MSV153 MSV501

M.K. Berlyn, CGSC, USAa Kumaresan and Jayaraman (1998) M.K. Berlyn, CGSC, USAa This study, JK10AB X P1/(CAG12178) This study, DM49 X P1/(JK10ABT) This study, DM49 X P1/(JK10AB) This study, DM49N X P1(JK10AB) M.K. Berlyn, CGSC, USAa M.K. Berlyn, CGSC, USAa M.K. Berlyn, CGSC, USAa M.K. Berlyn, CGSC, USAa Laboratory collection M.K. Berlyn, CGSC, USAa Laboratory collection This study, DM49RN X P1/(JW0221-1) This study, DM49RN X P1/(N3055) This study, DM49RN X P1/(JW0762-2) This study, DM49RN X P1/(SK3451) This study, DM49RN X P1/(RW82) This study, AB2402 X P1/(CAG18531) This study, DM49RN X P1/(MS2402K) M.K. Berlyn, CGSC, USAa S. Gottesman, NIH, USAb This study, SG20780 X P1/(CAG18413) This study, AB1157 X P1/(MS780LK) This study, DM49 X P1/(MS780LK) This study, DM49RN X P1/(MS780LK) S.J. Sandler, USA M.K. Berlyn, CGSC, USAa This study, SS996 X P1/(CAG18531) This study, AB1157 X P1/(SS996K) This study, DM49 X P1/(SS996K) This study, DM49RN X P1/(SS996K)

This study, AB2421 X P1/(CAG18531) This study, BS-8 X P1/(CAG18531) This study, GY854 X P1/(CAG18531) This study, AB1157 X P1/(MS45K) This study, AB1157 X P1/(MS153K) This study, AB1157 X P1/(MS501K)

MSV554, MSV593, MSV673, MSV1157, MSV1229, MSV1569 MSV1810, MSV1849 and MSV1995

Same as AB1157 but zbh-3108::Tn10kan uvrB554/uvrB593/uvrB673/uvrB1157/uvrB1229/ uvrB1569/uvrB1810/uvrB1849/uvrB1995 alleles, respectively (see also text)

This study, AB1157 X P1/(NTG treated CAG18531)

49RN45, 49RN153, 49RN501, 49RN554, 49RN593, 49RN673, 49RN1157, 49RN1229, 49RN1569, 49RN1810, 49RN1849 and 49RN1995

Same as DM49RN but zbh-3108::Tn10kan uvrB45/uvrB153/uvrB501/uvrB554/uvrB593/ uvrB673/uvrB1157/uvrB1229/uvrB1569/uvrB1810/ uvrB1849/uvrB1995 alleles, respectively (see also text)

This study, DM49RN x P1(Respective strains carrying zbh-3108::Tn10kan uvrBxxxx alleles)

a b

CGSC: Coli Genetic Stock Center, USA. NIH: National Institutes of Health, USA.

MMC treatment were normalized with respect to it and a logarithmic plot of the survival was made. UV resistance was checked by plating appropriate dilutions of mid-log phase cultures and exposing them to different time intervals of UV light as described in Miller (1972) [29,30]. 2.4. PCR and cloning The whole genomic DNA from appropriate strains were isolated from relevant strains using the protocol described by Chen and Kuo

[31] and treated with RNase at 37 ◦ C for 30 min, heat inactivated at 65 ◦ C for 15 min and used as template for the PCR. The 777 bp region spanning the Rif resistance cassette in rpoB87 and the 546 bp region spanning the Nal resistance region in gyrA87 were amplified using the genomic DNA isolated from DM49RN with the forward primer rifF: 5 CGTCGTATCCGTTCCGTTGG 3 and the reverse primer rifR: 5 GGCAACAGCACGTTCCATACC 3 for rpoB and the forward primer nalF: ATGAGCGACCTTGCGAGAGA and the reverse primer nalR: CGGGATGTTGGTTGCCATAC for the gyrA genes. The PCR was done with thirty amplification cycles with the following

918

V. Shanmughapriya, M.H. Munavar / DNA Repair 11 (2012) 915–925

conditions: 5 min at 94 ◦ C, 30 s at 94 ◦ C, 45 s at 55 ◦ C, 30 s at 72 ◦ C, 7 min at 72 ◦ C and maintained at 4 ◦ C. The 2.1 kb uvrB+ allele was amplified with the primer set F: TCAAGCTTTGCTCATGATTGACAGC and R: GAGGATCCAGTCTTCTTCGCTATCCT. The PCR was done with thirty amplification cycles with the following conditions: 5 min at 94 ◦ C, 30 s at 94 ◦ C, 45 s at 56 ◦ C, 2 min 30 s at 72 ◦ C, 7 min at 72 ◦ C and maintained at 4 ◦ C. The ∼800 bp with lexA3 allele was amplified using the primer set F: TACCATGGGCATTCTGTTATGGTCG and R: TCAAGCTTTCGCGGTCTCAGAGATA. The PCR was done with thirty amplification cycles with the following conditions: 5 min at 94 ◦ C, 30 s at 94 ◦ C, 45 s at 55 ◦ C, 1 min at 72 ◦ C, 7 min at 72 ◦ C and maintained at 4 ◦ C. The obtained products were checked on 1% agarose gels along with a 1 kb DNA ladder. The puvrB+ clone was constructed by ligating the PCR product of uvrB+ allele and the pBR322 vector both digested with HindIII and BamHI restriction enzymes. The plexA3 clone was constructed by ligating lexA3 PCR product and the pBR322 vector both digested with EcoRI and HindIII restriction enzymes. The puvrB+ lexA3 clone was constructed by ligating the uvrB+ PCR product and plexA3 clone both digested with HindIII and BamHI. The obtained clones were checked by releasing the insert by restriction digestion. 2.5. Sequencing and analyses of relevant mutations The obtained PCR products were purified using DNA extraction Kit obtained from Fermentas, India. The purified PCR products were then sequenced using relevant primers by Chromous Biotech, Bangalore, India. The RifR region was sequenced using the primers rifF: 5 CGTCGTATCCGTTCCGTTGG 3 and rifR: 5 GGCAACAGCACGTTCCATACC 3 , the NalR region was sequenced using the primers nalF: ATGAGCGACCTTGCGAGAGA and nalR: CGGGATGTTGGTTGCCATAC, the entire uvrB gene were sequenced using the primers F1: TACCATGGGCATTCTGTTATGGTC, R1: TCAAGCTTTCGCGGTCTCAGAGATA, F2: GATGATGCTCCATCTCACGG and R2: CTACCAGCACGTCGAACTCA. The sequence results obtained were then analysed with BLAST tool in NCBI nucleotide database and mutations/mismatches in DNA sequence of E. coli K12 substrain MG1655 obtained manually analysed. All the sequence reactions were repeated twice with relevant forward primer and reverse primer for the purpose of confirmation of the mutation. 2.6. Construction of lon strains and observation of filamentation The strain SG20780 is lon and carries cps::lac transcriptional fusion [32]. Into this strain the KanR marker was transduced using P1 lysate made on CAG18413 bearing tsx3100::Tn10kan and Kan resistant transductants were screened for Cps-Lac+ phenotype (denoting the retention of lon) and Cps-Lac− phenotype (denoting loss of lon). From one such SG20780 tsx3100::Tn10kan lon strain (named MS780LK), P1 lysate was made and the lysate was independently used to transduce tsx3100::Tn10kan into AB1157, DM49, DM49RN strains. In all the three crosses the KanR transductants obtained on suitable selective plates containing LB + Kan + citrate were checked for their mucoid phenotype (denoting co-transduction of lon with tsx3100::Tn10kan) and non-mucoid phenotype (denoting the transduction of only tsx3100::Tn10kan excluding lon) at 30 ◦ C. Since mucoid phenotype is characteristic of lon mutants due to hyper expression of cps genes only at 30 ◦ C, enough care was taken to see the mucoid phenotype/selection of transductants at only 30 ◦ C. The obtained strains were named as given in Table 1. These set of strains were then observed for filamentation under mitomycin C induced and un-induced conditions by observing the stained preparations of the cells under light microscope with 40× resolution. The un-induced cultures were directly heat fixed on glass slides, stained with

saffranin red and viewed. For inductions, the mid-log phase cells were treated with 0.5 ␮g/ml of MMC for 2 h before processing. The slides were then observed under the Nikon Eclipse Ti light microscope at 40× resolution and images were viewed and photographed using the software NIS elementsD version 3.0.

2.7. Transductional transfer of chromosomal sulA::gfp fusion The strain SS996 with sulA::gfp chromosomal construct at the att site was obtained from Prof. S.J. Sandler, USA as a kind gift [33]. The zbh3108::Tn10kan from CAG18531 at the 17.67 min, which is located near the gal-bio region was transduced into the SS996 through P1 transduction in order to link the KanR marker to sulA::gfp fusion. Transductants resistant to kanamycin were selected on appropriate selective plates. Randomly chosen colonies were screened for retention of sulA::gfp on chromosome using polymerase chain reaction (PCR) with gfp primers: Forward: 5 GGAGAGGGTGAAGGTGATGC 3 and Rev: 5 -GGTCACGCTTTTCGTTGGG 3 . P1 lysate from one isolate (named sgK996) that gave positive result (sulA::gfp linked with KanR marker) in the PCR, was used to transduce into AB1157, DM49 and DM49RN and once again random transductants were screened for presence of sulA::gfp (co-transduction of sulA::gfp with KanR marker) by PCR. Such colonies were named AB1157sgk, DM49sgk and DM49RNsgk and were used for fluorescence studies.

2.8. Observation of fluorescence For studying the fluorescence levels, glass cover slips were taken and dipped in poly-l-lysine solution for 2–5 min and dried in laminar hood. The relevant strains were grown in appropriate medium to an OD600 of ∼0.5. 20 ␮l of these cells were spread on to the polyl-lysine coated cover slip and air dried in laminar hood. These cover slips were then fixed with 4% formaldehyde for 15 min and air dried. For permeabilization of bacterial cells, 0.2% Triton X was added and left for 15 min and washed. This cover slip was then placed upside down on a glass slide, sealed and stored for future purposes. The slides were viewed through a Carl Zeiss fluorescence microscope equipped with water immersion objectives and connected to axio cam Hrm digital camera with an excitation maximum at 488 nm. Images were captured and analysed with Axio-vision-4.8 software. The images were processed using standard image processing techniques.

2.9. RT-PCR analyses Reverse Transcription PCR was performed by isolating RNA from late-log phase cultures of AB1157, DM49 and DM49RN strains before and after treatment with 0.5 ␮g/ml MMC (MMC treatment was done for 2 h). The RNA obtained was then normalized with corresponding OD at 260 nm and used as template for Reverse Transcription. First step cDNA was obtained from the RNA samples using the first strand synthesis kit from Chromous Biotech, Bangalore, India. The PCR with first strand cDNA for sulA was performed with primers sulARTFor – 5 -GCTATGCACATCGTTCTT C-3 and sulARTRev – 5 -GGAAAGTTGTCTCGTGGC and uvrB was performed with uvrBRTFor 5 -CGACGCTGTTTGATTACCTG-3 and uvrBRTRev 5 -CTACCAGCACGTCGAACTCA-3 . A control reaction for RT-PCR was performed with primers for rpoB as follows: rpoB1698F 5 -AACATCGGTTTGATCAAC-3 and rpoB2041R 5 -CGTTGCATGTTGGTACCCAT-3 . The PCR was done with thirty amplification cycles with the following conditions: 15 min at 94 ◦ C, 30 s at 94 ◦ C, 45 s at 55 ◦ C, 30 s at 72 ◦ C, 7 min at 72 ◦ C and

V. Shanmughapriya, M.H. Munavar / DNA Repair 11 (2012) 915–925

919

damage to some extent, only a combination of both gave rise to very high percentage of survival as expected of the phenotype termed SIR+ reported earlier [26,27]. 3.2. Sequence changes due to rpoB87 and gyrA87 mutations and their implications

Fig. 1. The log % survival of 0.5 ␮g/ml MMC treated cells of indicated strains. Relevant genotypes are mentioned wherever appropriate. The data plotted on the graph are the means of four independent sets of experiments performed in duplicates with standard error ranges.

maintained at 4 ◦ C. The amplicon sizes were verified by agarose gel electrophoresis. 2.10. Localized mutagenesis using NTG and selection of UV sensitive mutants The NTG mutagenesis experiment was performed as described in Miller 1992 [30] on the strain CAG18531 carrying a Tn10kan marker close to uvrB. P1 lysate was made from the NTG treated O/N enriched culture of CAG18531 and transduced into AB1157. 8 different transduction experiments were performed and transductants from each case were screened for loss of UV resistance. From 2208 colonies screened, 9 UV sensitive mutants were isolated. The obtained UV sensitive mutations were then used as donors and P1 lysate made from each of the strain was first transduced into AB1157 strain and obtained KanR transductants were screened for loss of UV resistance to ensure co-transduction of the putative uvrB allele with linked zbh3108::Tn10kan. After ensuring the co-transduction of putative UV sensitive uvrB allele with zbh3108::Tn10kan marker in each case, the relevant P1 lysates were used to transduce the KanR marker into DM49RN and the KanR transductants were screened for loss of MMC resistance. 3. Results 3.1. Reconstruction of DM49 rpoB87 gyrA87 strain and mitomycin C survival analyses Although the earlier work of Kumaresan and Jayaraman [26] implicated the involvement of rpoB87 and gyrA87 mutations in elicitation of SIR phenotype in lexA3 or recA defective strains in which the conventional SOS pathway cannot be induced, the genes/functions involved in the proposed SIR pathway remained unidentified till this investigation. One Mud-lac insertion identified by them in a locus tentatively designated as sir was also not studied in detail. Therefore, to get more insights about this novel phenotype/pathway, we initially introduced the rpoB87 and gyrA87 mutations into DM49 (lexA3 Ind− ) strain as described in Section 2.3. UV survival tests performed on relevant strains showed that all the DM49, DM49N, DM49R and DM49RN strains were UV sensitive as expected while AB1157 was UV resistant (data not shown). We then used these set of strains to perform MMC survival analyses as described in Section 2.3 and the results obtained are illustrated in Fig. 1. As could be seen from Fig. 1, in lexA3 Ind− strain, although either rpoB87 or gyrA87 alone increased the survival to MMC

The genetic mapping studies of the Rif resistant (rpoB87) and Nal resistant (gyrA87) mutations indicated the presence of the mutations in the rpoB and gyrA genes, respectively. However, this was not verified earlier by sequencing. Mutations conferring rifampicin resistance have been shown to occur majorly in the rpoB gene [34]. The regions of rifampicin resistance in rpoB gene of E. coli has been well analysed and was believed to have reached near saturation [35,36]. The rifampicin resistant region is concentrated mainly in the central region of the RNAP ␤ subunit spanning amino acid 500–700. This region of rifampicin resistance has been divided into three clusters. The cluster I comprising of amino acids 507–533, cluster II located between amino acids 563 and 572 while the cluster III comprises only one allele located at amino acid 687 [36]. Apart from these clusters, one single Rif resistant mutation has been reported that alters the amino acid 146 [37]. This shows the involvement of the central region of RNAP ␤ subunit in Rif binding and more than 90% of the Rif alleles were located in this region. Primers were designed to amplify a 777 bp region covering the major Rif resistant clusters and PCR amplifications were done as said in Section 2.4. The analysis of the sequencing data from this amplicon indicated a single C→T transition at the 1565th bp of rpoB87. This mutation changes the serine (TCT) to phenyl alanine (TTT) in the 522nd amino acid of RNAP ␤ subunit which falls under cluster 1 of rifampicin resistance and conserved region D. Resistance to nalidixic acid has been found to occur in the gyrA and gyrB genes equally [38]. Mutations in the gyrA subunit of DNA Gyrase which give rise to nalidixic acid resistance arise close to the N-terminus of the gene near amino acid 122, which is the binding site of transiently cleaved DNA [39,40]. The quinolone resistance determining region (QRDR) of the gyrA gene was located within the relatively small region spanning amino acids 67 through 106 [39]. Also, around 80% of the mutations that arose seemed to be concentrated in the amino acids 81–87 [39,41]. Specifically, the region around amino acid 83 is vital for GyrA activity. Thus we designed primers to amplify the initial 546 bp of the gyrA gene which spans all the hotspots reported as of date for Nal resistance and PCR amplification of these regions were done as said in Section 2.4. The sequence analysis of this amplicon indicated that this region harbours a single G→A transition in the Nucleotide 244 of gyrA87 gene, which converts the amino acid 82 of GyrA from aspartate (GAC) to asparagine (AAA). Although the region from 81 to 87 has been considered the mutational hotspot of the Nal resistance region, till date, as far as we know, there have been no reports of a Nal resistant mutation occurring in the amino acid 82 of this region. Admittedly, we are compelled to conclude that gyrA87 defines a novel NalR allele. 3.3. Suppression of SIR phenotype by uvrB mutations Although the rpoB87 and gyrA87 mutations elicit ‘SIR’ phenotype in the lexA3 mutant, the involvement of conventional SOS functions in the so-called SIR pathway remained equivocal since the previous report [26]. Considering the fact that the rpoB87 is same as rpoB3595 reported by Jin and Gross [36] and owing to the fact that this rpoB87/rpoB3595 has been shown to be fast moving RNA polymerase [42], one can imagine somehow that some SOS genes coming under the class – weakly repressed/rapidly derepressed (<1 min) [4] by LexA might get turned on by this mutant form of RNAP bearing rpoB87/rpoB3595 mutation. The mutated form of

920

V. Shanmughapriya, M.H. Munavar / DNA Repair 11 (2012) 915–925

3.4. Study of the SulA activity in SIR proficient/deficient strains

Fig. 2. The log % survival of 0.5 ␮g/ml MMC treated cells of indicated strains. All strains carry lexA3, rpoB87 and gyrA87 mutations and other relevant genotypes are mentioned wherever appropriate. The data plotted on the graph are the means of three independent sets of experiments performed in duplicates with standard error ranges.

GyrA due to gyrA87 lesion can further add to this effect. With this in mind we transduced dinB::kan, umuDC::cat, uvrD::Tn5, uvrA::Tn10 and uvrB::Tn10kan insertions into DM49RN (lexA3 rpoB87 gyrA87) strain and studied the level of MMC resistance. Quite unexpectedly uvrB::Tn10kan led to marked decrease in MMC resistance of the otherwise SIR proficient strain (Fig. 2). These results clearly imply that functional UvrB protein is mandatory for the acquisition of MMCR phenotype of lexA3 rpoB87 gyrA87 mutant and SIR may not be SOS Independent Repair in the true sense. The introduction of uvrB point mutation, uvrB29, from the strain AB2402 also rendered the DM49RN strain (49RN29) MMC sensitive. When the 49RN29 strain was transformed with the clone bearing the uvrB+ allele with its own promoter and other regulating sequences, the MMC resistance of the strain was found to increase by two fold (Fig. 3).

Fig. 3. The log % survival of 0.5 ␮g/ml MMC treated cells of indicated strains. All strains carry lexA3, rpoB87 and gyrA87 mutations and other relevant genotypes are mentioned wherever appropriate. The data plotted on the graph are the means of three independent sets of experiments performed in duplicates with standard error ranges.

As our genetic studies clearly indicate the role of uvrB, a LexA repressed gene, in the MMC repair of SIR proficient strain, our next intention was to study the activity of SulA, another major protein involved in the SOS response. The sulA gene is under tight regulation by the LexA repressor and is induced at high levels during SOS response [43]. The function of SulA in SOS dependent DNA repair is binding to the FtsZ protein involved in cell division and thereby stalling the process of cell division till DNA damage is repaired [44]. The Lon protease has been shown to be the major protease involved in cleavage of SulA after SOS mediated repair thereby making FtsZ free for continuing the stalled cell division [45]. The lack of Lon protease in the cell makes the cell hyper-sensitive to DNA damaging agents due to inability of the cells to divide and thus giving rise to long (therefore, the locus was originally named as lon) filaments [46]. Thus, in the case of induction of SOS in lon mutants the cells would extensively filament due to lack of SulA degradation [46]. Hence, we first studied the extent of filamentation in lon derivatives of DM49RN (lexA3 rpoB87 gyrA87), DM49 (lexA3) and AB1157. The strains were then observed for the formation of filamentation under mitomycin C induced and non-induced conditions as described in Section 2.6. No apparent filamentation was seen in the lon derivative of DM49RN, MS49RNLK, indicating that SulA activity is not increased due to rpoB87-gyrA87 to any considerable level to be detected based on filamentation (Fig. 4). In order to further validate our results, sulA::gfp fusion was transduced into relevant strains as described in Section 2.7 and the strains obtained thus, AB1157sgk (sulA::gfp), DM49sgk (lexA3 sulA::gfp) and DM49RNsgk (lexA3 rpoB87 gyrA87 sulA::gfp), were examined for fluorescence (as described in Section 2.8). The fields observed in each slide showed that AB1157sgk showed increased fluorescence upon MMC induction which was not the case with either DM49sgk or DM49RNsgk (Fig. 5). From these studies it can be said that the rpoB87 gyrA87 mutations do not lead to increased activity or expression of the SulA protein in the lexA3 strain, DM49.

3.5. Increased level of uvrB but not sulA transcription in lexA3 rpoB87 gyrA87 mutant The results presented above indicate that somehow rpoB87 and gyrA87 mutations increase the UvrB function against MMC damage in lexA3 mutant but this is not the case with SulA function. Although one can imagine that increased UvrB function is due to increased expression of uvrB+ gene in rpoB87 gyrA87 strains, it is imperative to verify the same by Reverse Transcription PCR. RTPCR was performed as described in Section 2.9. Analysis of the obtained results for sulA once again confirmed that sulA expression is not increased at the transcriptional level in the DM49RN (lexA3 rpoB87 gyrA87) strain upon MMC treatment (Fig. 6 – sulA DM49RN untreated, DM49RN MMC treated) as expected but increased in the AB1157 strain upon MMC treatment (Fig. 6 – sulA AB1157 untreated, AB1157 MMC treated). The Reverse Transcription PCR for uvrB showed increased levels of the uvrB amplicon even without MMC treatment in DM49RN (Fig. 6 – uvrB DM49RN untreated, DM49RN MMC treated) while the AB1157 strain showed higher amplification of uvrB amplicon only upon MMC treatment (Fig. 6 – uvrB AB1157 untreated, AB1157 MMC treated). The DM49 (lexA3) strain showed no significant increase in amplification with or without MMC treatment either with sulA or with uvrB (Fig. 6 – sulA, uvrB DM49 untreated, DM49 MMC treated). The amplification observed in housekeeping control, rpoB, was found to be similar in the all the cDNA samples analysed (Fig. 6 – rpoB). From the observed result, it is tenable to conclude that the transcription of uvrB is increased to

V. Shanmughapriya, M.H. Munavar / DNA Repair 11 (2012) 915–925

921

Fig. 4. Extent of filamentation in indicated strains as observed under light microscope. Relevant genotypes of the strains are mentioned wherever appropriate.

a certain level in the DM49RN strain by rpoB87-gyrA87 mutations despite lexA3 mutation.

by the puvrB+ lexA3 clone (data not shown). These results reveal that it is indeed the rpoB87 and gyrA87 mutations that give rise to this unconventional ‘SIR’ phenotype which demands increased expression of uvrB.

3.6. Suppression of MMC resistance by lexA3 clone and its alleviation by uvrB+ allele The uvrB gene has been shown to have 3 different promoters which include LexA dependent and LexA independent [47,48]. While promoters P2 and P3 are inhibited or repressed by LexA, the promoter P1 of uvrB has been shown to function independent of LexA repression. But this is not the case with the only promoter of sulA which overlaps with the LexA binding box [49]. Thus, it is quite possible that rpoB87 and gyrA87 mutations could be responsible for the increased transcription from promoter P1 of uvrB which in normal strains accounts for most of the uvrB transcription in uninduced cells [50]. To test this, we constructed a lexA3 multicopy clone in the pBR322 vector as described in Section 2.4 and transformed the same into the DM49RN strain. We found that the DM49RN/plexA3 strains showed marked decrease in the MMC resistance as shown in Fig. 7. This decreased resistance to MMC by presence of lexA3 in multicopy was reverted by the presence of uvrB+ allele in the same clone (Fig. 7). From these results, it is tenable to conclude that the constitutive expression of uvrB found in the DM49RN strain is possibly from the LexA dependent promoter as the presence of increased LexA3 molecules in the cell decreases the resistance. But the presence of the clone bearing uvrB+ allele alone does not give any significant increase in resistance in the DM49RN strain or the DM49 strain. Also, a decrease in MMC resistance is observed in an AB1157 strain due to the plexA3 clone which is not reverted

3.7. Effect of various uvrB point mutations on the MMC resistance of DM49RN As our results indicate that UvrB protein alone (of the entire NER proteins) is needed for the increased survival of DM49RN strain after MMC damage, the question of which domain of UvrB plays a role in ‘SIR’ arises. To answer this question we wanted to introduce many different mutant forms of uvrB gene into the DM49RN strain to study its effect on MMC resistance. We have already shown that the uvrB::kan allele and the uvrB29 allele from AB2402 markedly decreases the MMC resistance of the DM49RN strain (Section 3.3). We further proceeded to introduce other reported UV sensitive uvrB mutants into DM49RN as described in Section 2.1. The uvrB mutant alleles uvrB29, uvrB45, uvrB153 and uvrB501 from the strains AB2402, AB2421, BS-8 and GY854 strains, respectively, were first linked to the zbh3108::Tn10kan marker as described in Section 2.1. Also nine mutations capable of rendering AB1157 strain UVS and perhaps mapping in uvrB were isolated in this study using NTG mediated localized mutagenesis as described in Section 2.10. These nine mutations were named uvrB554, uvrB593, uvrB673, uvrB1157, uvrB1229, uvrB1569, uvrB1810, uvrB1849 and uvrB1995 (Table 1). The linkage of these UV sensitive mutations to zbh3108::Tn10Kan was confirmed by co-transduction of the UV sensitive phenotype (∼80–90%) with zbh3108::Tn10kan with all the 9 mutations

922

V. Shanmughapriya, M.H. Munavar / DNA Repair 11 (2012) 915–925

Fig. 5. Extent of fluorescence in indicated strains as observed under fluorescence microscope. Relevant genotypes of the strains are mentioned wherever appropriate.

isolated. The zbh3108::Tn10kan linked uvrB mutants were then transduced into the DM49RN strain (Section 2.1) and the transductants were then screened for loss of MMC resistance in each case. Each of the case tested was found to give rise to MMC sensitive colonies indicating that the lesion(s) present in each case should be in a nearby gene, in all likelihood the uvrB gene coding for the UvrB protein, that is involved in MMC resistance of the SIR strains. Thus, all the tested uvrB mutants were sequenced to identify the lesion(s) present in each case as described in Section 2.5. We found that both the uvrB29 allele from AB2402 and the uvrB501 allele from GY854 harbour a single base deletion at G224 and T323, respectively, thereby creating a frameshift for the rest of the coding sequence. The uvrB45 and uvrB153 alleles define the same mutation, a single transversion from C137 →A137 changing the codon 46 of the UvrB protein from threonine to asparagine. All the uvrB alleles isolated in this study through the NTG mutagenesis experiment were

Fig. 6. RT-PCR based analysis of expression of sulA, uvrB and rpoB in indicated strains with and without MMC treatment.

found to harbour two mutations: A C244 →T244 transition changing codon 82 from proline (CCG) to serine (TCG) and another C293 →T293 transition changing the codon 98 from proline (CCG) to Leu (CTG). From the results observed, we conclude that the T46 of the UvrB protein plays an essential role in the MMC resistance of DM49RN as does the P82 and/or P98, all of which are located in highly conserved region of UvrB. The results obtained are summarized in Table 2. 4. Discussion In this investigation, we have analysed the possible cause for elicitation of SIR+ phenotype in lexA3 mutant by combination of rpoB87-gyrA87 mutations. What we found as the base change

Fig. 7. The values plotted in the graph indicate the relative % survival of the relevant strains on 0.5 ␮g/ml MMC containing plates taking the cfu/ml on LB plates lacking MMC as 100%. All the values plotted are average of three independent set of experiments. The SE values are plotted as error bars.

V. Shanmughapriya, M.H. Munavar / DNA Repair 11 (2012) 915–925

923

Table 2 Table indicating the base change(s) defined by different uvrB alleles and their consequence on the strain. All the mutations listed were sequenced in this study and except uvrB29, uvrB45, uvrB153 and uvrB501 all other mutations were isolated in this study. S. no.

Designation of allele(s)

Lesion defined by the alleles in uvrB

Amino Acid change in UvrB

UVS/R phenotype when present in AB1157

MMC S/R phenotype when present in DM49RN

Domain involved

Possible activity affected

1 2 3 4

uvrB29 uvrB501 uvrB45, uvrB153 uvrB554, uvrB593, uvrB673, uvrB1157, uvrB1229, uvrB1569, uvrB1810, uvrB1849 uvrB1995

G224 T323 C137 →A137 C244 →T244 and C293 →T293

Frameshift from G75 Frameshift from F108 T46N P82S and P98L

UVS UVS UVS UVS

MMCS MMCS MMCS MMCS

N.A. N.A. 1a 1a and ␤ hairpin

N.A. N.A. ATPase ATPase and SS DNA binding

N.A.: not applicable.

defined by rpoB87, namely the C→T transition at the 1565th bp of rpoB has already been reported by Jin and Gross [36] as rpoB3595 in 1988. This mutation is in one of the highly conserved regions of RNAP falling under cluster I of Rif resistance. Kogoma (1994) reported that this rpoB3595 (rpoB87) mutation enhances the constitutively expressed SOS response in the absence of RNaseHI activity up to 5 fold [51]. The enhancement of SOS expression was analysed by ␤-galactosidase levels from a sfi::lacZ fusion and it was reported that this enhanced expression by rpoB3595 (rpoB87) is under the LexA control and the enhancement does not occur in recA302 and lexA3 Ind− strains. It was shown in another previous report that the RpoB3595 RNA polymerase has an accelerated elongation rate during transcription in vitro and in vivo [42]. It has also been observed that if more than one RNAP is made to initiate from the same promoter, then the RNAP can transcribe through roadblocks caused by either Lac repressor at the relevant operator or by DNA binding proteins at the relevant binding sites [52]. This mechanism is believed to be caused by ‘pushing’ the blocked RNAP of the trailing RNAP. This also acts as a cooperative effect and as more numbers of RNAP are present more is the de-repression. Forward translocation of RNAP during repressional roadblocks has been shown to be favoured by increased intracellular NTP pools [53]. There are evidences that suggest that the rifampicin binding site on the RNA polymerase may interact with DNA [54]. To add to this notion, one Rif resistant mutation located in the conserved region D destabilized RNA polymerase binding to the promoter [55]. Another major finding that adds to these aspects is that a residue in between amino acids 516 and 540 is close to the 5 end of the nascent RNA transcript [55,56]. All these together suggest that RNAP is capable of turning on expression of repressed genes to some degree in altered

cellular conditions like increased concentration of RNAP, mutated RNAP or higher NTP levels. Thus, we believe that this modified form of RNAP can in some way elicit some level of higher expression of certain genes that are repressed or less expressed under normal conditions which may be involved in the MMC damage repair of the strains that carry the rpoB87 mutation. Previous reports suggest that nalidixic acid resistant mutations have effect on the efficiencies of recombination, linkage, and DNA repair and mutation frequency [57]. There have been many reports that the DNA Gyrase is involved in DNA repair mechanisms. Other studies imply the interaction between RNAP and DNA Gyrase which may alter supercoiling of the DNA. Certain Rif resistant RpoB together with DNA Gyrase have been shown to be involved in the enhancement of negative supercoiling of DNA [58,59]. We, thus, believe that the alteration of the acidic amino acid aspartate to a basic amino acid asparagine (due to gyrA87) might induce some altered negative supercoiling activity of DNA Gyrase and that coupled with the altered structural conformation of RNAP ␤ subunit (due to rpoB87) with the replacement of phenyl-alanine with a bulky side chain instead of Serine may induce expression from certain genes that are under certain conditions repressed or silent. The LexA repressor has been known to be auto-catalytic and selfcleavage of LexA occurs under alkaline pH in a RecA independent manner [60]. Also internal pH of the cell can change the DNA binding affinity of LexA repressor [61]. The binding of a protein to its target DNA is said to require energetically favourable interactions with the base pairs of its binding site and conformational changes in both the protein and the DNA [62,63]. Thus, as discussed above the mutant RNAP and GyrA could possibly overcome the blocks during transcription by operator bound repressor complexes. All

Fig. 8. Simulated structure of Escherichia coli UvrB protein from residues 5 to 596 using Swissmodel and the structure of a UvrB variant from Bacillus caldotenax (PDB number: 1T5Lb, UniProtKB: P56981) was used as the template. The quality of the structure predicted was validated using the SAVES server. The structure obtained was processed using Pymol pdb viewer. (a) Structure of Wildtype UvrB protein of E. coli showing the nucleotide phosphate binding region in green and the SS DNA binding ␤ hairpin structure in magenta. (b) Structure of protein from uvrB29 allele showing T46N mutation in red in the nucleotide phosphate binding region denoted in green. (c) Structure of UvrB protein from uvrB554, uvrB593, uvrB673, uvrB1157, uvrB1229, uvrB1569, uvrB1810, uvrB1849 and uvrB1995 alleles showing P82S lesion in blue and P98L lesion in cyan within the SS DNA binding ␤ hairpin structure shown in magenta.

924

V. Shanmughapriya, M.H. Munavar / DNA Repair 11 (2012) 915–925

these information adds support to our finding that the SIR pathway could be the partial activation of the orthodox SOS pathway even in presence of an un-inducible LexA3 to some degree so that the strain could tolerate MMC damage, but not UV damage. The UvrABC mediated nucleotide excision repair is the major repair system acting on MMC mediated DNA damage. It has been already shown by Kumaresan and Jayaraman [26] that an uvrC::Tn10 mutation does not alter the MMC resistance of the rpoB87-gyrA87 strain. Vidal et al. [64] and Lage et al. [65] have shown, in the UvrABC pathway, the inactivation of only uvrB caused sensitivity to MMC or other types of DNA damaging agents like PUVA but not the inactivation of uvrA or uvrC or even the uvrC homologue cho. Vidal et al. have also shown that upon MMC treatment the number of double strand breaks per genome generated was higher in an uvrB mutant strain with increasing incubation times and recovery of high molecular weight DNA was reduced as compared to the wildtype and even uvrA and uvrC mutants [64]. Thus a NER independent role for UvrB in repair of DNA damage and double strand breaks caused by MMC was clearly implicated which was also reflected in our results and favours the proposal that SIR is in some way interrelated with SOS. Although, our studies on the filamentation, fluorescence studies and RT PCR analyses did indicate the non-expression of sulA in the DM49RN strain, it clearly indicates that the uvrB gene is constitutively expressed in the same strain. This indicates that subsets of SOS genes might be expressed by a combination of rpoB87 gyrA87 mutations despite uninducible LexA3 that includes uvrB. The studies on the multicopy effect of lexA3 and lexA3 along with uvrB+ alleles indicate that the increased expression of uvrB is majorly from the LexA dependent promoter of the uvrB gene. Thus, it may be possible that the LexA3 protein might probably have decreased binding affinity to certain sites, which already have weaker LexA boxes like the uvrB, due to possible altered negative super-coiling of DNA by mutant DNA Gyrase. This may make the LexA3 mutant repressor less tightly bound to the DNA, thereby making it easier for the mutant RNAP (RpoB87) to have higher transcription rates from the promoter. The presence of higher number of LexA3 molecules might therefore pose increased hindrance to the transcribing RNA polymerase and as an indirect consequence decrease the MMC resistance. But whether other SOS genes that also have weak LexA boxes show increased expression as well by rpoB87 gyrA87 mutations is to be studied. The sequence analyses of the uvrB mutations that affect MMC resistance of the DM49RN strain indicate the involvement of the T46 of UvrB in the same. The T46 of UvrB falls within the ATPase like domain identified in the UvrB protein (Fig. 8a). It forms an integral part of the sequence motif that is common to many ATPases in UvrB that is rich in glycine and other amino acids such as serine, threonine, alanine and proline [66]. It lies within the nucleotide phosphate binding region · · ·GVTGSGKT· · · present in the domain 1a of UvrB which spans residues 1–89, 116–150, 324–346 and 379–411 which is known to interact with ATP [66,67]. The T46N mutation changes the second T of the conserved nucleotide phosphate binding region and mutations in this domain have been shown to be defective in forming the 3 and 5 incisions and ATPase activity thereby inactivating UvrB functionally [67] (see Fig. 8b). Thus, we can conclude that the ATPase activity of the UvrB protein is essential in exhibiting the MMC resistance in the SIR pathway. Also from the other sequenced mutations the P98 and/or P82 seems to play important role in the MMC resistance of SIR (see Fig. 8c). As said earlier the P82 form a part of the domain 1a involved in ATP binding by UvrB. The P98 forms a part of the ␤ hairpin structure of spanning residues 90–115 of UvrB involved in SS DNA binding and helicase activities of UvrB [67,68] (Fig. 8a). Mutations in this ␤-hairpin region of UvrB have been shown to be defective in binding of UvrB to damaged SS DNA and forming 5 incision but not 3 incision [67,69]. So it is possible that along with the ATPase activity

of UvrB the DNA binding and damage processing activity of UvrB might also be required for the repair of MMC damage in the SIR phenotype. Taken together, the results presented herein reveal that the mutational changes in RNA polymerase and DNA Gyrase causes the increased expression of a supposedly repressed gene uvrB which is responsible for the MMC resistance exhibited in lexA3 rpoB87 gyrA87 strain and the UvrB protein is essential in elicitation of this SIR phenotype. The ATPase activity and also possibly the DNA binding activity of UvrB could be involved in eliciting this ‘SIR’ phenotype. We propose that the increased MMC survival of the DM49RN involves an alternate mechanism which manages to bypass the repression of uvrB, a SOS gene that is under weaker LexA repression by higher transcription from its promoter(s). Conflicts of interest The authors declare that there are no conflicts of interest. Acknowledgements We thank Prof. R. Jayaraman for his expert advice, Prof. G. Mukhopadhyay and his group, Special Centre for Molecular Medicine, Jawaharlal Nehru University, New Delhi, India for their kind support, Prof. S.J. Sandler, University of Massachusetts, USA for providing us with the sulA::gfp chromosomal construct, M.K. Berlyn, CGSC, USA for the E. coli strains, Prof. P. Gunasekaran and Prof. Sripathi Kandula for permitting us to use the instrumentation facilities of Centre for Excellence in Genomic Sciences and Networking Resource Centre in Biological Sciences Instrumentation facilities, Prof. S. Shanmugasundaram and Prof. G. Marimuthu for their support and help. Mr. J. Kumaresan, Dr. B. Singaravelan, T. Ponmani, N. Arul Muthukumaran, S. Vinodha, M. Karthik, S. Meenakshi, S. Vivek Raj and G. Sutharsan for their help and assistance, Shankar Manoharan for his help. S. Poovalingam and P. Jagadish for Laboratory chores. We thank the anonymous referee for his comments that helped to revise this manuscript in its present form. This work was financially supported by Centre for Excellence in Genomic Sciences and Centre for Advanced Studies Programs of School of Biological Sciences, Madurai Kamaraj University originally funded by University Grants Commission, Government of India. V. Shanmughapriya thanks, Council for Scientific and Industrial Research, India for providing Junior and Senior Research Fellowships. References [1] P. Jeggo, Isolation and characterization of Escherichia coli K-12 mutants unable to induce the adaptive response to simple alkylating agents, J. Bacteriol. 139 (1979) 783–791. [2] M. Radman, SOS repair hypothesis: phenomenology of an inducible DNA repair which is accompanied by mutagenesis, Basic Life Sci. 5A (1975) 355–367. [3] J.W. Little, D.W. Mount, The SOS regulatory system of Escherichia coli, Cell 29 (1982) 11–22. [4] C. Janion, Inducible SOS response system of DNA repair and mutageneis in Escherichia coli, Int. J. Biol. Sci. 4 (2008) 338–344. [5] G.C. Walker, Mutagenesis and inducible responses to deoxyribonucleic acid damage in Escherichia coli, Microbiol. Rev. 48 (1984) 60–93. [6] H. Shingawa, H. Iwasaki, T. Kato, A. Nakata, RecA protein-dependent cleavage of UmuD protein and SOS mutagenesis, Proc. Natl. Acad. Sci. U.S.A. 85 (1988) 1806–1810. [7] M. Tang, X. Sjem, E.G. Frank, M. O’Donnell, R. Woodgate, M.F. Goodman, UmuD’(2)C is an error-prone DNA polymerase, Escherichia coli Pol V, Proc. Natl. Acad. Sci. U.S.A. 96 (1999) 8919–8924. [8] P. Jeggot, Isolation and characterization of Escherichia coli K-12 mutants unable to induce the adaptive response to simple alkylating agents, J. Bacteriol. 139 (1979) 783–791. [9] B. Demple, B. Sedgwick, P. Robins, N. Totty, M.D. Waterfeild, T. Lindahl, Active site and complete sequence of the suicidal methyltransferase that counters alkylation mutagenesis, Proc. Natl. Acad. Sci. U.S.A. 82 (1985) 2688–2692. [10] B. Sedgwick, T. Lindahl, Recent progress on the Ada response for the inducible repair of DNA alkylation damage, Oncogene 21 (2002) 8886–8894.

V. Shanmughapriya, M.H. Munavar / DNA Repair 11 (2012) 915–925 [11] S. Dinglay, S.C. Trewick, T. Lindahl, B. Sedgwick, Defective processing of methylated single-stranded DNA by E. coli AlkB mutants, Genes Dev. 14 (2000) 2097–2105. [12] M.Z. Humayun, SOS and Mayday: multiple inducible mutagenic pathways in Escherichia coli, Mol. Micobiol. 30 (1998) 905–910. [13] V.A. Palejwala, R.W. Rzepka, M.Z. Humayun, UV irradiation of Escherichia coli modulates mutagenesis at a site-specific ethenocytosine residue on M13 DNA: evidence for an inducible recA-independent effect, Biochemistry 32 (1993) 4112–4120. [14] M.S. Rahman, P.M. Dunman, G. Wang, H.S. Murphy, M.Z. Humayun, Effect of UVM induction on mutation fixation at non-pairing and mispairing DNA lesions, Mol. Microbiol. 22 (1996) 747–755. [15] J. Nair, C.E. Vaca, I. Velic, M. Mutanen, L.M. Valsta, H. Bartsch, High dietary omega-6 polyunsaturated fatty acids drastically increase the formation of etheno-DNA base adducts in white blood cells of female subjects, Cancer Epidemiol. Biomarkers Prev. 6 (1997) 597–601. [16] W. Szybalski, V.N. Iyer, Crosslinking of DNA by enzymatice or activated mitomycins and porfiromycins, bifunctionally ‘alkylation’ antibiotics, Fed. Proc. 23 (1964) 946–957. [17] M. Tomasz, Mitomycin C: small, fast and deadly (but very selective), Chem. Biol. 2 (1995) 575–579. [18] A.C. Sartorelli, W.F. Hodnick, M.F. Belcourt, M. Tomasz, B. Haffty, J.J. Fischer, S. Rockwell, Mitomycin C: a prototype bioreductive agent, Oncol. Res. 6 (1994) 501–508. [19] A. Sancar, G.B. Sancar, DNA repair enzymes, Annu. Rev. Biochem. 57 (1988) 29–67. [20] D.K. Orren, C.P. Selby, J.E. Hearst, A. Sancar, Post-incision steps of nucleotide excision repair in Escherichia coli. Disassembly of the UvrBC–DNA complex by helicase II and DNA polymerase I, J. Biol. Chem. 267 (1992) 780–788. [21] E.E.A. Verhoeven, C. Wyman, G.F. Moolenaar, J.H. Hoeijmakers, N. Goosen, Architecture of nucleotide excision repair complexes: DNA is wrapped by UvrB before and after damage recognition, EMBO J. 20 (2001) 601–611. [22] A. Sancar, W.D. Rupp, A novel repair enzyme: UvrABC excision nuclease of Escherichia coli cuts a DNA strand on both sides of the damaged region, Cell 33 (1983) 249–260. [23] J.J. Lin, A.M. Phillips, J.E. Hearst, A. Sancar, Active site of (A)BC excinuclease. II. Binding, bending and catalysis mutants of UvrB reveal a direct role in 3 and an indirect role in 5 incision, J. Biol. Chem. 267 (1992) 17693–17700. [24] Y. Zou, H. Ma, I.G. Minko, S.M. Shell, Z. Yang, Y. Qu, Y. Xu, N.E. Geacintov, R.S. Lloyd, DNA damage recognition of mutated forms of UvrB proteins in nucleotide excision repair, Biochemistry 43 (2004) 4196–4205. [25] G.F. Moolenaar, S. van Rossum-Fikkert, M. van Kesteren, N. Goosen, Cho, a second endonuclease involved in Escherichia coli nucleotide excision repair, Proc. Natl. Acad. Sci. U.S.A. 99 (2002) 1467–1472. [26] K.R. Kumaresan, R. Jayaraman, SOS independent survival against mitomycin C induced lethatlity in a rifampicin-nalidixic acid resistant mutant of Escherichia coli, Mutat. Res. 194 (1988) 109–120. [27] K.R. Kumaresan, R. Jayaraman, The sir locus of Escherichia coli: a gene involved in SOS-independent repair of mitomycin C-induced DNA damage, Mutat. Res. 235 (1990) 85–92. [28] M. Demerec, E.A. Adelberg, A.J. Clark, P.E. Hartman, A proposal for a uniform nomenclature in bacterial genetics, Genetics 54 (1966) 61–66. [29] J.H. Miller, Experiments in Molecular Genetics, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 1972. [30] J.H. Miller, A Short Course in Bacterial Genetics: A Laboratory Manual and Hand Book for Escherichia coli and Related Bacteria, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 1992. [31] W.P. Chen, T.T. Kuo, A simple and rapid method for the preparation of Gramnegative bacterial genomic DNA, Nucleic Acids Res. 21 (1993) 2260. [32] J.C. Brill, C. Quinlan-Walshe, S. Gottesman, Fine structure mapping and indentification of two regulators of capsule synthesis in Escherichia coli K-12, J. Bacteriol. 170 (1988) 2599–2611. [33] J.D. McCool, E. Long, J.F. Petrosino, H.A. Sandler, S.M. Rosenberg, S.J. Sandler, Measurement of SOS expression in individual Escherichia coli K-12 cells using fluorescence microscopy, Mol. Microbiol. 3 (5) (2004) 1343–1357. [34] W. Zillig, K. Zechel, D. Rabussay, M. Schachner, V.S. Sethi, P. Palm, A. Heil, W. Seifert, On the rule of different subunits of DNA-dependent RNA polymerase from E. coli in the transcription process, Cold Spring Harb. Symp. Quant. Biol. 35 (1970) 47–58. [35] K. Severinvo, M. Soushko, A. Goldfarb, V. Nikiforov, Rifampicin regions revisited: new rifampicin resistant and streptolidigin resistant mutants in the ␤ subunit of Escherichia coli RNA polymerase, J. Biol. Chem. 268 (1993) 14820–14825. [36] D.J. Jin, C.A. Gross, Mapping and sequencing of mutations in the Escherichia coli rpoB gene that lead to rifampicin resistance, J. Bacteriol. 171 (1989) 5229–5231. [37] N.A. Lisitsyn, E.D. Sverdlov, E.P. Moiseyeva, O.N. Danilevskaya, V.G. Nikiforov, Mutation to rifampicin resistance at the beginning of the RNA polymerase ␤ subunit gene in Escherichia coli, Mol. Gen. Genet. 196 (1984) 173–174. [38] S. Nakamura, M. Nakamura, K. Kojima, H. Yoshida, gyrA and gyrB mutations in quinolone-resistant strains of Escherichia coli, Antimicrob. Agents Chemother. 33 (1989) 254–255. [39] H. Yoshida, M. Bogaki, M. Nakamura, S. Nakamura, Quinolone resistance determining region in the DNA Gyrase gyrA gene of Escherichia coli, Antimicrob. Agents Chemother. 34 (1990) 1271–1272.

925

[40] D.S. Horowitz, J.C. Wang, Mapping the active site tyrosine of Escherichia coli DNA Gyrase, J. Biol. Chem. 262 (1987) 5339–5344. [41] Y. Saenz, M. Zarazaga, L. Brinas, F. Ruiz-Larrea, C. Torres, Mutations in gyrA and parC genes in nalidixic acid-resistant Escherichia coli strains food products, humans and animals, J. Antimicrob. Chemother. 51 (2003) 1001–1005. [42] D.J. Jin, R.R. Burgess, J.P. Richardson, C.A. Gross, Termination efficiency at rhodependent terminators depends on kinetic coupling between RNA polymerase and rho, Proc. Natl. Acad. Sci. U.S.A. 89 (1992) 1453–1457. [43] E.C. Friedberg, G.C. Walker, W. Siede, DNA Repair and Mutagenesis, American Society for Microbiology Press, Washington, DC, 1995. [44] E. Bi, J. Lutkenhaus, Analysis of ftsZ mutations that confer resistance to the cell division inhibitor SulA (SfiA), J. Bacteriol. 172 (1990) 5602–5609. [45] S. Mizusawa, S. Gottesman, Protein degradation in Escherichia coli: the Lon gene controls the stability of SulA protein, Proc. Natl. Acad. Sci. U.S.A. 80 (1983) 358–362. [46] S. Gottesman, E. Halpern, P. Trisler, Role of sulA and sulB in filamentation by Lon mutants of Escherichia coli K-12, J. Bacteriol. 148 (1981) 265–273. [47] P.F. Schendel, M. Fogliano, L.D. Strausbaugh, Regulation of the Escherichia coli K-12 uvrB Operon, J. Bacteriol. 150 (2) (1982) 676–685. [48] G.B. Sancar, A. Sancar, J.W. Little, W.D. Rupp, The uvrB gene of Escherichia coli has both LexA-repressed and LexA-independent promoters, Cell 28 (1982) 523–530. [49] S.T. Cole, Characterisation of the promoter for the LexA regulated sulA gene of Escherichia coli, Mol. Gen. Genet. 89 (3) (1983) 400–404. [50] E.A. van den Berg, R.H. Geerse, H. Pannekoek, P. van de Putte, In vivo transcription of the E. coli uvrB gene: both promoters are inducible by UV, Nucleic Acids Res. 11 (1983) 4355–4363. [51] T. Kogoma, Escherichia coli RNA polymerase mutants that enhance or diminish the SOS response constitutively expressed in the absence of RNaseHI activity, J. Bacteriol. 176 (1994) 1521–1523. [52] V. Epshtein, F. Toulome, A.R. Rahmouni, S. Borukhov, E. Nudler, Transcription through the road blocks: the role of RNA polymerase cooperation, EMBO J. 22 (2003) 4719–4727. [53] C. Morsin-Huaman, C.L. Turnbough, A.R. Rahmouni, Translocation of Escherichia coli RNA polymerase against a protein roadblock in vivo highlights a passive sliding mechanism for transcript elongation, Mol. Microbiol. 51 (2004) 1471–1481. [54] A. Mustaev, E. Zaychikov, K. Severinov, M. Kashlev, A. Polyakov, V. Nikiforov, A. Goldfarb, Topology of the RNA polymerase active center probed by chimeric rifampicin-nucleotide compounds, Proc. Natl. Acad. Sci. U.S.A. 91 (1994) 12036–12040. [55] C.A. Gross, C.L. Chan, M.A. Michael, A structure/function analysis of Escherichia coli RNA polymerase, Phil. Trans. R. Soc. Lond. B 351 (1996) 475–482. [56] K. Severinov, A. Mustaev, E. Severinova, M. Kozlov, S.A. Darst, A. Goldfarb, The ␤subunit Rif cluster I is only angstrom away from the active centre of Escherichia coli RNA polymerase, J. Biol. Chem. 270 (1995) 29428–29432. [57] S. Inan, A. Kalaycioglu, Effect of nalidixic acid on recombination and DNA repair of Escherichia coli K-12 strains, Indian J. Exp. Biol. 34 (1996) 949–953. [58] K. Drlica, R.J. Franco, T.R. Steck, Rifampicin and rpoB mutations can alter DNA supercoiling in Escherichia coli, J. Bacteriol. 170 (1988) 4983–4985. [59] B. Cheng, C. Zhu, C. Ji, A. Ahumada, Y. Tse-Dinh, Direct interaction between Escherichia coli RNA polymerase and the zinc ribbon domains of DNA topoisomerase I, J. Biol. Chem. 278 (2003) 30705–30710. [60] J.W. Little, Autodigestion of lexA and phage lambda repressors, Proc. Natl. Acad. Sci. U.S.A. 81 (1984) 1375–1379. [61] N.K. Relan, E.S. Jenuwine, O.H. Gumbs, S.L. Shaner, Preferential interactions of the Escherichia coli LexA repressor with anions and protons are coupled to binding the recA operator, Biochemistry 36 (1997) 1077–1084. [62] B. Tiebel, N. Radzwill, L.M. Aung-Hilbrich, V. Helbl, H.J. Steinhoff, W. Hillen, Domain motions accompanying Tet repressor induction defined by changes of interspin distances at selectively labeled sites, J. Mol. Biol. 290 (1999) 229–240. [63] G. Wisedchaisri, R.K. Holmes, W.G. Hol, Crystal structure of an IdeR–DNA complex reveals a conformational change in activated IdeR for base-specific interactions, J. Mol. Biol. 342 (2004) 1155–1169. [64] L.S. Vidal, L.B. Santos, C. Lage, A.C. Leitão, Enhanced sensitivity of Escherichia coli uvrB mutants to mitomycin C points to a UV-C distinct repair for DNA adducts, Chem. Res. Toxicol. 19 (10) (2006) 1351–1356. [65] C. Lage, S.R. Gonc¸alves, L.L. Souza, M. de Pádula, A.C. Leitão, Differential survival of Escherichia coli uvrA, uvrB, and uvrC mutants to psoralen plus UV-A (PUVA): evidence for uncoupled action of nucleotide excision repair to process DNA adducts, J. Photochem. Photobiol. B: Biol. 98 (2010) 40–47. [66] T.W. Seeley, L. Grossman, Mutations in the Escherichia coli UvrAB ATPase motif compromise excision repair capacity, Proc. Natl. Acad. Sci. U.S.A. 86 (1989) 6577–6581. [67] K. Theis, M. Skorvaga, M. Machius, N. Nakagawa, B. Van Houten, C. Kisker, The nucleotide excision repair protein UvrB, a helicase like enzyme with a catch, Mutat. Res. 460 (2000) 277–300. [68] G.F. Moolenar, L. Hoglund, N. Goosen, Clue to damage recognition by UvrB: residues in the ␤-hairpin structure prevent binding to non-damaged DNA, EMBO J. 20 (2001) 6140–6149. [69] M. Skorvaga, M.J. Dellavecchia, D.L. Croteau, K. Theis, J.J. Trugio, B.S. Mandavilli, C. Kisker, B. Van Houten, Identification of residues within UvrB that are important for efficient DNA binding and damage processing, J. Biol. Chem. 279 (2004) 51574–51580.