T7 RNA polymerases backed up by covalently trapped proteins catalyze highly error prone transcription

Abstract

RNA polymerases (RNAPs) transcribe genes through the barrier of nucleoproteins and site-specific DNA-binding proteins on their own or with the aid of accessory factors. Proteins are often covalently trapped on DNA by DNA damaging agents, forming DNA-protein cross-links (DPCs). However, little is known about how immobilized proteins affect transcription. To elucidate the effect of DPCs on transcription, we constructed DNA templates containing site-specific DPCs and performed in vitro transcription reactions using phage T7 RNAP. We show here that DPCs constitute strong but not absolute blocks to in vitro transcription catalyzed by T7 RNAP. More importantly, sequence analysis of transcripts shows that RNAPs roadblocked not only by DPCs but also by the stalled leading RNAP become highly error prone and generate mutations in the upstream intact template regions. This contrasts with the transcriptional mutations induced by conventional DNA lesions, which are delivered to the active site or its proximal position in RNAPs and cause direct misincorporation. Our data also indicate that the trailing RNAP stimulates forward translocation of the stalled leading RNAP, promoting the translesion bypass of DPCs. The present results provide new insights into the transcriptional fidelity and mutual interactions of RNAPs that encounter persistent roadblocks.

FIGURE 1.

DPCs in TS constitute strong but not absolute blocks to T7 RNAP. A, PAGE analysis of transcripts produced with TS-DPC templates. Transcription reactions were performed with the indicated templates (15 fmol), T7 RNAP (150 fmol), all four NTPs, [α-³²P]UTP, and RNase inhibitor at 37 °C for up to 20 min, and products were separated by 12% denaturing PAGE. Bands were visualized by [α-³²P]UTP incorporated into transcripts. The positions of RNA size markers run alongside are shown on the left. The weak bands of ³²P-labeled DNA templates that migrated slower than runoff transcripts are not shown. The positions of the T7 promoter, transcription start site (+1), damage (DPCs or FLU, +73), and the runoff site (+103) are shown at the top of the gel. The sequences of the templates are shown in supplemental Fig. S1C. B, same as panel A except that the transcription reactions were performed with NTS-DPC templates. DPCs and FLU are at position +74. C, time courses of the formation of runoff transcripts for TS-DPC and NTS-DPC templates. D, the number of copies of runoff transcripts produced at 20 min. The copy number was calculated by comparison of the radioactivity of a runoff product with that of a known amount of [α-³²P]UTP spotted on the same gel, and is plotted against the damage size on a logarithmic scale. The data in panels C and D are means of two independent experiments.

FIGURE 2.

Back up of T7 RNAPs by DPCs results in highly error prone transcription. A, distribution of transcriptional mutations induced by TS-DPCs. Transcription reactions were performed as described in the legend to Fig. 1A for 60 min using TS-DPCs and an RNAP/template ratio = 10. Runoff transcripts were converted to cDNA and cloned. Forty (except for NEI) or 20 (NEI) sequences between 50 nucleotides upstream (−) and 10 nucleotides downstream (+) of DPCs were analyzed for individual TS-DPC templates (supplemental Fig. S5). The number of transcriptional errors was counted for every 5-nucleotide area, and is plotted against the position relative to DPCs or FLU: deletion (blue), base substitution (green), and insertion (yellow). Consecutive deletions, base substitutions, or insertions were counted as one event. Mutation areas (U1, U2, and D1) and the putative positions of the backed up leading and trailing T7 RNAPs are shown at the top of the graph. B, percentages of mutated transcripts for PLA and END at RNAP/template ratios of 1 and 10. C, changes in mutation distributions with the RNAP/template ratios. Relative mutation distributions in areas U1, U2, and D1 are shown. In panels B and C, the data for RNAP/template = 10 are from Table 1 and panel A, and those for RNAP/template = 1 were obtained by separate experiments (sequence data not shown).

FIGURE 3.

Transcriptional mutations are associated with the backed up leading and trailing T7 RNAPs. A, EMSA analysis of DNA-RNAP complexes formed with the control (G) and TS-DPCs (END or H2A) templates. Transcription reactions were performed with the indicated templates (15 fmol), T7 RNAP (150 fmol), and all four NTPs as described in Fig. 1A for 30 and 60 min. After incubation, complexes were separated by native PAGE. Bands were detected by the ³²P label in template DNA. B, sensitivity of DNA-RNAP complexes to restriction enzymes. Transcription reactions were performed with the G and TS-END templates as in panel A for 60 min. Subsequently the reaction mixture was incubated with the indicated restriction enzyme (HinfI (Hin), HpyCH4III (Hpy), or BsiHKAI (Bsi)) and separated by native PAGE. C, analysis of restriction fragments retained in DNA-RNAP complexes. The individual bands of DNA-RNAP complexes in panel B for END were excised and DNA was electroeluted from the gel. The DNA was treated with proteinase K and separated by denaturing PAGE. Referencing restriction fragments were generated by treatment of the G template with Hin, Hpy, and Bsi (lanes 2–4, see also panel D). D, positions of stalled T7 RNAPs (green) in complexes 1–3 for DPC templates and complex 1 for the G template. Restriction sites, DPCs, mutation areas, and the T7 promoter are indicated by black, red, white, and gray boxes, respectively. The sizes of restriction fragments bearing 5′-³²P labels (asterisks) are also shown to clarify the correspondence to the data in panel C.

FIGURE 4.

Trailing RNAP stimulates forward translocation of stalled RNAPs through DPCs. A, EMSA analyses of the formation of DNA-RNAP complexes 1, 2, and 3 with varying amounts of T7 RNAP. Transcription reactions were performed with the indicated TS-DPC templates (15 fmol), T7 RNAP (15–300 fmol), and all four NTPs for 20 min. After incubation, complexes were separated by native PAGE. Bands were detected by the ³²P label in DNA. The ratios of RNAP/template (RNAP/Temp) are indicated at the top of gels. B, same as panel A except that the transcription reactions were performed with the G and TS-FLU templates. C, variation of the amount of runoff products with the RNAP/template ratios. Transcription reactions were performed with the indicated TS-DPC templates (15 fmol), T7 RNAP (7.5–75 fmol), all four NTPs, and [α-³²P]UTP for 20 min. Products were separated by denaturing PAGE and runoff products were quantified. The amounts of runoff products are plotted against the RNAP/template ratios. The existing regions of complexes 1, 2, and 3 are indicated by arrows at the top of the graph. The data for FLU were replotted from panel D. D, same as panel C except that the transcription reactions were performed with the G and TS-FLU templates. The data in panels C and D are mean ± S.D. for four independent experiments.