Regulation and Rate Enhancement during Transcription-Coupled DNA Repair

Summary Transcription-coupled DNA repair (TCR) is a subpathway of nucleotide excision repair (NER) that is triggered when RNA polymerase is stalled by DNA damage. Lesions targeted by TCR are repaired more quickly than lesions repaired by the transcription-independent “global” NER pathway, but the mechanism underlying this rate enhancement is not understood. Damage recognition during bacterial NER depends upon UvrA, which binds to the damage and loads UvrB onto the DNA. Bacterial TCR additionally requires the Mfd protein, a DNA translocase that removes the stalled transcription complexes. We have determined the properties of Mfd, UvrA, and UvrB that are required for the elevated rate of repair observed during TCR. We show that TCR and global NER differ in their requirements for damage recognition by UvrA, indicating that Mfd acts at the very earliest stage of the repair process and extending the functional similarities between TCR in bacteria and eukaryotes.

TCR was analyzed using a patch synthesis assay essentially as described in Figure 3.
Reactions contained 1.5 nM UV-irradiated plasmid pHWL1-T7A1-2 (which contains a reporter cassette downstream of the T7A1 promoter), and also 1.5 nM UV-irradiated plasmid pHWL1-2 (which contains a shorter reporter cassette that is not transcribed).
The experiment allowed repair patch synthesis in the T7A1 transcription unit during TCR to be directly compared with repair patch synthesis by global NER in nontranscribed pHWL1-2 DNA. Cleavage of the plasmid mixture by BsrGI and SphI produced nontemplate strand (NT) and template strand (T) fragments from pHWL1-T7A1-2, and two shorter strands from pHWL1-2 (labeled strands 1 and 2). TCR reactions contained RNAP, UvrB, C, D, DNA ligase, DNA polymerase, Mfd and 8 nM UvrA. Global NER reactions (GGR) were performed as for TCR reactions, except that RNAP and Mfd were omitted. The control reaction in lane C was performed as for the GGR reaction except that UvrA, B, C and D were omitted. The gel image shows the incorporation of [α-32 P]dATP into repair patches in the four separate strands under the conditions specified. The chart shows the intensities of strand 1 (pHWL1-2) and strands T and NT (pHWL1-T7A1-2) normalized to the intensity of strand 2 (pHWL1-2). The intensity of strand 1 relative to strand 2 is the same in TCR and GGR reactions as both strands of this nontranscribed reporter are repaired by global NER regardless of the conditions. The intensity of strand T relative to strand 2 increases under TCR conditions, because strand T is a substrate for the TCR pathway and its repair is enhanced by RNAP and Mfd. The intensity of strand NT relative to strand 2 does not differ significantly between GGR and TCR reactions. This result confirms that, under TCR conditions, patch synthesis in the nontemplate strand of pHWL1-T7A1-2 represents global NER. Data are the mean of a 3 independent experiments and are shown with SD.
(B) Effect of UvrA concentration on the strand-bias of TCR in vitro. TCR was analyzed using a patch synthesis assay as described in Figure 3. The gel image shows the incorporation of [α-32 P]dATP into repair patches in the template (T) and nontemplate (NT) strand of a reporter cassette located downstream of the T7A1 promoter. TCR reactions contained RNAP,UvrB,C,D,DNA ligase,DNA polymerase and Mfd,and 0.5,1,2,4,8,16,32,64 and 128 nM UvrA. GGR reactions contained 4 nM UvrA, and were performed as for TCR reactions except that RNAP and Mfd were omitted. The control reaction in lane C was performed as for the GGR reaction except that UvrA, B, C and D were omitted. The chart shows the ratio of incorporation of radioactive label into the T and NT strands. The intercept of the x axis on the y axis denotes the ratio of repair obtained in the GGR reactions. Data are the mean of 3 independent experiments, and are shown with SD.   concentrations from 1-800 nM were incubated with 1 nM radiolabeled 50 bp duplex containing a single FldT adduct for 20 min at 37°C before analysis. ("+" DNA binding was detected at < 300 nM UvrA; "-" No DNA binding detected at 800 nM UvrA.; "ND" not determined). As shown previously, the substitutions within either the proximal ATPase site or the distal ATPase site decreased the NTPase activity of the protein (Myles et al., 1991;Thiagalingam and Grossman, 1991). UvrB did not affect the residual activity of either mutant. UvrA R712A R722A R724A R730A showed a reduced GTPase activity that was not affected by UvrB, and was unable to bind to DNA even at high protein concentration. The GTPase activities of the remaining mutants were all close to WT, but differed in their response to UvrB: UvrB repressed the activity of UvrA Δ290-400 and UvrA G502D, but did not repress the activity of UvrA Δ131-248 (which lacks the UvrB binding domain) or UvrA R216A E222A (in which the UvrB interaction surface is disrupted). promoter: the mfd gene was amplified by PCR using a primer that introduced an NdeI site and a start codon upstream of codon 354, and the product was digested with NdeI and ligated between the NdeI sites of pETMfd-T7. pMfd19 R165A R181A F185A encodes an untagged Mfd derivative under the control of the mfd promoter, and was created from pMfd19 (Selby and Sancar, 1993) by "rolling circle PCR" mutagenesis (Kirsch and Joly, 1998). pETMfd2 R165A R181A F185A also encodes an untagged Mfd derivative under the control of the mfd promoter, and was created by transferring NdeI-NdeI fragments from pMfd19 R165A R181A F185A into NdeI-digested pETMfd2 (Deaconescu et al., 2006). pETMfd2 ΔBHM encodes untagged Mfd ΔBHM under the control of the mfd promoter: a region of the mfd gene spanning the two natural NdeI sites (located 41 bp upstream of the start codon and at codon 910) was amplified by PCR using an upstream primer that deleted codons 2-353, and the product was digested with NdeI and ligated between the NdeI sites of pETMfd2. pETMfd2ΔD7 has been described previously (Deaconescu et al., 2006). pETMfd-T7 R165A R181A F185A

Plasmid and Strain Construction
ΔD7 encodes untagged Mfd ΔD7 carrying R165A R181A F185A substitutions, under the control of a T7 promoter: the mutated mfd gene was amplified from pETMfd2 R165A R181A F185A using a primer that inserted an NdeI site upstream of the start codon, and an NdeI-NdeI fragment was transferred into NdeI-digested pETMfd 1-997 -T7 (Smith et al., 2007).
pETDuetUvrA encodes UvrA under the control of a T7 promoter, and was created in two stages: (i) the uvrA gene was amplified from MG1655 genomic DNA using primers that introduced a BamHI site immediately upstream of the start codon, an NcoI site overlapping the start codon, and a HindIII site immediately downstream of the stop codon; the PCR fragment was ligated as a BamHI -HindIII fragment into the corresponding sites in the polylinker of pBluescript II KS(-) (Stratagene) to create pBSIIUvrA; (ii) the uvrA gene was excised from pBSIIUvrA as an NcoI -HindIII fragment and ligated into the corresponding sites in pETDuet-1 (Novagen). Expression vectors encoding mutant N-terminally His-tagged UvrA or UvrB proteins carrying substitutions or deletions, were derived from pQE30UvrA and pQE30UvrB respectively (Manelyte et al., 2009), using "rolling circle PCR" mutagenesis (Kirsch and Joly, 1998).
Mfd ΔD7 was purified as described in (Smith et al., 2007). E. coli DNA ligase, DNA polymerase I and restriction enzymes were purchased from New England Biolabs.
WT Mfd was purified from BL21(DE3) cells transformed with pETMfd-T7: cells were grown in LB containing appropriate antibiotics at 37°C to mid-log phase (A 600 ~0.5), expression was induced by the addition of 1 mM IPTG and the culture was incubated for 3 hr at 37°C before harvesting. Mfd ΔBHM was purified from BB834(DE3) cells transformed with pETMfdΔBHM-T7: cells were grown in LB containing appropriate antibiotics at 37°C to mid-log phase (A 600 ~0.5), expression was induced by the addition of 1 mM IPTG and the culture was incubated for 1 hr 40 min at 30°C before harvesting.
Cells were grown in LB containing appropriate antibiotics at 37°C to mid-log phase (A 600 ~0.5), expression was induced by the addition of 1 mM IPTG and the culture was incubated for 3 hr at 30°C before harvesting. Mfd R165A R181A F185A (Mfd D2AAA) was purified from FB20256 (a derivative of MG1655 carrying a Tn5(Kan R ) insertion in the chromosomal mfd gene (Kang et al., 2004)) transformed with pMfd19 R165A R181A F185A: cells were grown in LB containing appropriate antibiotics at 37°C until the A 600 was 1.5, and were then harvested. Mfd R165A R181A F185A ΔD7 (Mfd D2AAA ΔD7) was purified from BL21(DE3) Δmfd cells transformed with pETMfd-T7 R165A R181A F185A ΔD7: cells were grown in LB containing appropriate antibiotics at 37°C to mid-log phase (A 600 ~0.5), expression was induced by the addition of 1 mM IPTG and the culture was incubated for 1 hr at 27°C before harvesting. In all cases the untagged Mfd proteins were purified from the harvested cell pellet as described in Chambers et al., 2003.
Untagged WT UvrA (used for experiments presented in Figures 3, S1D, S2, S3A and S4) was purified from BL21(DE3) pLysS transformed with pETDuetUvrA: cells were grown in LB containing appropriate antibiotics at 37°C to mid-log phase (A 600 ~0.5), expression was induced by the addition of 1 mM IPTG and the culture was incubated for sepharose, heparin and MonoQ columns (all GE Healthcare) using the procedures and buffers described for Mfd purification in (Chambers et al., 2003). His-tagged WT and mutant UvrA proteins (used for experiments presented in Figures 5A, 6, 7, S3B, S4 and Table S1) were purified from FB21635 [F´ proAB lacI q Z∆M15 Tn10 (Tet R )] transformed with the appropriate pQE30UvrA derivatives, following the procedure described in (Manelyte et al., 2009). His-tagged mutant UvrB proteins were purified from JW0762-2 [F´ proAB lacI q Z∆M15 Tn10 (Tet R )] transformed with the appropriate pQE30UvrB derivatives, following the procedure described in (Manelyte et al., 2009), except that cultures were grown at 30°C.

In Vitro Patch Synthesis Assay for TCR
Strand-specific repair was analyzed using a modification of the repair synthesis assay reported in (Selby and Sancar, 1993). Supercoiled DNA templates containing randomly were incubated at 37°C for 20 min, then stopped by phenol/chloroform extraction. The extracted DNA was ethanol-precipitated, and each DNA pellet was resuspended in 50 μl NEB buffer 2 + 100 μg/ml BSA, and digested with 5 U BsrGI and 5 U SphI. The digested DNA was purified by phenol/chloroform extraction and ethanol precipitation.

Assay for Mutation Frequency Decline
Mutation frequency decline was monitored using an assay based on that described in (Fabisiewicz and Janion, 1998). The strains used were AB1157 (relevant genotype mfd + argE3(Oc)), and UNCNOMFD (mfdderivative of AB1157) transformed with pETMfd2 derivatives or pET21a. In experiments with UNCNOMFD transformants the liquid LB and agar plates used were supplemented with 100 μg/ml ampicillin. Cultures were grown in LB until the A 600 was approximately 0.3. The cultures were then harvested by centrifugation and the pellets were resuspended at an A 600 of approximately 0.3 in M9 medium that lacked amino acids and glucose. Each resuspended culture was warmed to 37°C, and then a 5 ml aliquot was placed in a 90 mm diameter Petri dish and irradiated with 20 J/m 2 254 nm light. The irradiated culture was immediately supplemented with glucose (0.4% w/v final concentration), and a 1 ml sample was withdrawn and added to 5 ml LB that had been prewarmed to 37°C (zero time sample).
The remaining irradiated culture was incubated at 37°C for 30 min, and then a second 1 ml sample was withdrawn and added to 5 ml LB that had been prewarmed to 37°C (30 min. sample). The two samples were incubated overnight at 37°C. The overnight cultures were serially diluted in M9 medium lacking amino acids and glucose, and appropriate dilutions were plated on LB agar (to determine cfu per ml) and M9 agar supplemented with 0.4% w/v glucose, and 25 μg/ml histidine, leucine, proline and threonine (to identify Arg + mutants). Plates were incubated at 37°C before counting and determining the frequency of reversion to Arg + . All UV-induced mutation frequencies measured were between 1 x 10 -6 and 3 x 10 -5 .

Assays for DNA Binding and Incision Activities of NER Proteins
DNA binding by UvrA was studied by EMSA. The substrates were blunt-ended 32 Plabeled 50 bp duplexes with or without a single FldT adduct 29 bp from the 5′ end of one strand (sequence as described for incision assays

RNAP-Displacement Assays
The RNAP-displacement assays presented in Figure S2 used a slightly modified version of the protocol described in (Smith and Savery, 2005;Smith et al., 2007). The DNA substrate used was a 570 bp PCR fragment generated from plasmid pAR1707 (Levin et al., 1987). This fragment contains the T7A1 promoter and is essentially the same as the RsaI/SmaI fragment of pAR1707 described in (Smith and Savery, 2005). Transcription initiation complexes were formed on this fragment by incubating 0.4 nM end-labeled