Initiation of DNA repair mediated by a stalled RNA polymerase IIO
2006; Springer Nature; Volume: 25; Issue: 2 Linguagem: Inglês
10.1038/sj.emboj.7600933
ISSN1460-2075
AutoresJean‐Philippe Lainé, Jean‐Marc Egly,
Tópico(s)RNA modifications and cancer
ResumoArticle12 January 2006free access Initiation of DNA repair mediated by a stalled RNA polymerase IIO Jean-Philippe Lainé Jean-Philippe Lainé Search for more papers by this author Jean-Marc Egly Corresponding Author Jean-Marc Egly Institut de Génétique et de Biologie Moléculaire et Cellulaire, CNRS/INSERM/ULP, Illkirch Cedex, CU Strasbourg, France Search for more papers by this author Jean-Philippe Lainé Jean-Philippe Lainé Search for more papers by this author Jean-Marc Egly Corresponding Author Jean-Marc Egly Institut de Génétique et de Biologie Moléculaire et Cellulaire, CNRS/INSERM/ULP, Illkirch Cedex, CU Strasbourg, France Search for more papers by this author Author Information Jean-Philippe Lainé and Jean-Marc Egly 1 1Institut de Génétique et de Biologie Moléculaire et Cellulaire, CNRS/INSERM/ULP, Illkirch Cedex, CU Strasbourg, France *Corresponding author. Institut de Génétique et de Biologie Moléculaire et Cellulaire, CNRS/INSERM/ULP, BP 163, 67404 Illkirch Cedex, CU Strasbourg, France. Tel.: +33 388 65 34 47; Fax: +33 388 65 32 01; E-mail: [email protected] The EMBO Journal (2006)25:387-397https://doi.org/10.1038/sj.emboj.7600933 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info The transcription-coupled repair (TCR) pathway preferentially repairs DNA damage located in the transcribed strand of an active gene. To gain insight into the coupling mechanism between transcription and repair, we have set up an in vitro system in which we isolate an elongating RNA pol IIO, which is stalled in front of a cisplatin adduct. This immobilized RNA pol IIO is used as 'bait' to sequentially recruit TFIIH, XPA, RPA, XPG and XPF repair factors in an ATP-dependent manner. This RNA pol IIO/repair complex allows the ATP-dependent removal of the lesion only in the presence of CSB, while the latter does not promote dual incision in an XPC-dependent nucleotide excision repair reaction. In parallel to the dual incision, the repair factors also allow the partial release of RNA pol IIO. In this 'minimal TCR system', the RNA pol IIO can effectively act as a loading point for all the repair factors required to eliminate a transcription-blocking lesion. Introduction The different mechanisms involved in deciphering the genetic information leading from a gene to a protein are highly regulated and any circumstances that impede them can have severe consequences. Unfortunately, the integrity of cellular DNA is constantly jeopardized by UV radiation, antitumoral drugs (cisplatin) or environmental agents that distort the DNA helix (Hoeijmakers, 2001). Those lesions pose a serious threat to cellular processes such as transcription, as they can obstruct an elongating RNA polymerase II (RNA pol II) and lead to an apoptotic response if left unrepaired (Yamaizumi and Sugano, 1994; Ljungman and Zhang, 1996). Moreover, persistent lesions can also be bound by proteins such as TBP or HMGA/B, which could indirectly lead to an inhibition of transcription and repair (Vichi et al, 1997; Reeves and Adair, 2005). Keeping the gene sequence intact to protect the genetic information then is one of the cell's priorities and is carried out by a transcription-dependent mechanism called transcription-coupled repair (TCR). TCR and global-genome nucleotide excision repair (GG-NER) are subpathways of NER. While GG-NER deals with the removal of lesions present in the overall genome, TCR is specific for removing lesions present on the transcribed strand of active genes (Bohr et al, 1985; Mellon et al, 1987). At the molecular level, TCR and GG-NER can be distinguished by their damage recognition factor. In GG-NER, the lesion is recognized by the complex XPC–HR23B in an ATP-independent manner, and will further allow for the binding of TFIIH and the subsequent NER factors (Sugasawa et al, 1998; Riedl et al, 2003). In TCR, an elongating RNA pol IIO, likely helped by additional specific factors, is thought to be the switch for engaging the repair mechanism (Christians and Hanawalt, 1992; Sweder and Hanawalt, 1992; Svejstrup, 2003). In Escherichia coli, the stalled RNA polymerase triggers the recruitment of the mutation frequency decline (mfd) protein repair, allowing for the release of RNA polymerase and further recruitment of the repair factors (Selby and Sancar, 1993). In mammals, CSA and CSB (rad28 and rad26 in yeast, respectively) are thought to play this function. Indeed, the trademark characteristic of TCR-deficient Cockayne syndrome (CS) cells is a defect in the mRNA recovery synthesis after UV irradiation (Wade and Chu, 1979; Troelstra et al, 1992; van der Horst et al, 1997). CSB was found to be connected to both the transcription and the repair machineries: (1) CSB, as a member of the SWI/SNF family of proteins, behaves as a chromatin remodeling factor most likely through its DNA-dependent ATPase activity (Citterio et al, 2000); (2) CSB/rad26 stimulates RNA synthesis both in vivo and in vitro (Balajee et al, 1997; Selby and Sancar, 1997b; Lee et al, 2001); (3) CSB interacts with transcriptional complexes containing TFIIH, TFIIE, p53, RNA pol II and I, as well as with repair factors such as XPG, XPA and XAB2 (Iyer et al, 1996; Tantin et al, 1997; Selby and Sancar, 1997b; van Gool et al, 1997; Nakatsu et al, 2000; Bradsher et al, 2002). Therefore, it is not surprising that inherited mutations in proteins specifically involved in the TCR pathway give rise to the rare autosomal recessive disorder CS (Venema et al, 1990; van Hoffen et al, 1993). While the majority of CS cases are caused by mutations in the CSA and CSB genes (Troelstra et al, 1992; Henning et al, 1995), CS phenotypes can also be associated with the xeroderma pigmentosum (XP) arising from mutations in XPB/XPD (TFIIH) and XPG genes. CS patients are characterized by progressive neurodegeneration and developmental defects (Nance and Berry, 1992). They also display sun sensitivity manifested as a severe rash. We hypothesized that a stalled RNA pol IIO would trigger the TCR reaction. We then set up an in vitro system in which a DNA fragment containing a single site-specific lesion located downstream from a promoter is immobilized on magnetic beads, allowing for both transcription and repair. We showed that an isolated elongating RNA pol IIO stalled at the lesion is able to sequentially recruit the repair factors in the absence of XPC–HR23B factor. Furthermore, we demonstrate that this RNA pol IIO-associated complex initiates and/or mediates an ATP-dependent incision of the damaged DNA in the presence of CSB. Results A template for transcription and DNA repair To investigate the connection between transcription and DNA repair, we have set up an assay in which the DNA template can be used for both reactions. A plasmid, containing a promoter and cisplatin DNA adduct (at position 105 on the transcribed strand), was cut by two restriction enzymes and biotinylated at the extremity, leading to a damage-containing DNA fragment of 709 base pairs (bp), further immobilized on streptavidin magnetic beads (Cax-Pt, Figure 1A). Figure 1.A template for both transcription and DNA repair reactions. (A) The transcription/repair template (Cax-Pt) contains a single cisplatin adduct (GTG, Pt) at position +105 nt in the transcribed strand downstream from the adenovirus major late promoter. The TATA box is represented by a triangle and the start site +1 by a bent arrow. The positions of the restriction enzymes sites are indicated. (B) Coomassie staining of highly purified transcription and repair factors. (C) Transcription on Cax (lanes 1–5) and Cax-Pt (lanes 6–10) was performed using either RTS or WCE/XPC as indicated at the top of the panel. Full length (328 nt) or prematurated stopped transcripts (105 nt) were resolved on 8% urea/PAGE. The Addition of α-amanitin is indicated. (D) Dual incision on Cax-Pt was carried out with RIS in either the presence of (lane 1) or the absence of XPC (RISΔXPC; lane 2), WCE/Hela (lane 3) or WCE/XPC (lane 4) supplemented with XPC (lane 5). Dual incision is indicated by the occurrence of a 26–34-nucleotides excision products. Download figure Download PowerPoint The ability of the Cax-Pt and the undamaged Cax templates to be transcribed was analyzed by using either a reconstituted transcription system (RTS), containing, in addition to RNA pol II, the basal transcription factors TBP, TFIIB, IIE, IIF and TFIIH (Figure 1B, upper panel), or an XPC-deficient cell extract (WCE/XPC). Run-off transcription on Cax-Pt resulted in a 105-nucleotide (nt)-long RNA transcript, suggesting that the cisplatin lesion presents a strong impediment to the progression of RNA pol II (Figure 1C, lanes 5 and 7). On the contrary, transcription on the undamaged Cax template allowed the synthesis of a full 328 nt length of RNA (lanes 1 and 3). The addition of α-amanitin, a specific RNA pol II transcription inhibitor, prevented RNA synthesis (lanes 2, 4, 6 and 8). We also monitored a dual incision assay on Cax-Pt using either a reconstituted incision system (RIS), containing recombinant XPC–HR23B, TFIIH, XPA, RPA, XPG and XPF/ERCC1 (Figure 1B, lower panel), a HeLa whole-cell extract (WCE/HeLa) or WCE/XPC (Figure 1D). Both the RIS and WCE/HeLa (lanes 1 and 3) were able to release the damaged 26–34 nt dual incision products contrary to WCE/XPC and RIS lacking XPC (RISΔXPC) (lanes 2 and 4). The lack of repair activity of WCE/XPC could be overcome by adding recombinant XPC (lane 5). Recruitment of NER factors onto the stalled RNA pol II Having established a system in which both transcription and repair could be carried out on the same substrate, we next wished to isolate a single RNA pol II stalled at the lesion. The Cax-Pt template was first preincubated with RTS for 15 min at 24°C, and then incubated either in the absence of (Figure 2A, lane 1) or in the presence of nucleotide triphosphates for 45 min (lanes 2 and 3). After being washed at different salt concentration, the supernatant was discarded and the remaining proteins bound to the immobilized DNA were submitted to Western blotting. We observed that the 50 mM washed preinitiation complex (PIC-50, lane 1) contained most of the basal transcription factors in addition to the hypophosphorylated RNA pol IIA. After adding the nucleotide triphosphates and further washes at 50 mM KCl, all the transcription factors, in addition to both the RNA pol IIA associated with the transcription initiation complex and the hyperphosphorylated elongating RNA pol IIO (EC-50, lane 2), were present on the immobilized DNA. Following additional washes of EC-50 at 400 mM KCl plus 0.1% sarkosyl, we observed that only RNA pol IIO remained tightly bound to Cax-Pt (EC-400, lane 3), ridding the promoter of the basal transcription factors and RNA pol IIA as well. However, only a small percentage (5–10%) of PIC efficiently promotes transcription; therefore, most of the damaged region of the substrate is not protected by RNA pol IIO (data not shown). To eliminate the unprotected lesion, EC-400 was then incubated with ClaI restriction enzyme, which cuts at position +86 (Figure 1A). When the ClaI-cut EC-400 immobilized DNA was incubated with RIS, no dual incision repair signal could be observed as compared to an uncut Cax-Pt (Figure 2B, lanes 3 and 1, respectively). As a control, untranscribed Cax-Pt cut by ClaI did not exhibit a dual incision signal when incubated with RIS (lane 2). ClaI-cut EC-400 represents therefore a single lesion-stalled RNA pol IIO on the damaged DNA fragment, whose unprotected remaining cisplatin lesions have been removed. Figure 2.Recruitment of NER factors onto the stalled RNA pol IIO. (A) Western blots of transcription complexes on immobilized DNA before (PIC, lane 1) and after (EC, lanes 2 and 3) the addition of nucleotide triphosphates. Immobilized protein complexes were washed either at 50 mM KCl (PIC-50 and EC-50) or at 400 mM KCl (EC-400). (B) Dual incision reaction on the immobilized Cax-Pt transcribed or not, and cut by ClaI. (C) Transcription reactions on Cax-Pt (lane 1) were further incubated with TFIIS or CSB (lanes 2 and 3). Arrows indicate the different lengths of RNA transcripts. (D) Immobilized ClaI-cut Cax-Pt (lane 1) as well as ClaI-cut EC-400 complex (lanes 2 and 3) were incubated with XPC. Following a second wash, these templates were tested in a dual incision reaction in which XPC is omitted. Fgt-Pt competitor damaged DNA (lane 2) is added where indicated. Complete NER (lane 4) is used as a control. (E) Immobilized Cax-Pt (lanes 1 and 4) or ClaI-cut EC-400 containing the stalled RNA pol IIO (lanes 2 and 3, 5–8) were pretreated with CIP (as indicated), and further incubated with either WCE/XPC or WCE/XPCΔCSB in the presence of ATP and CSB as indicated. Proteins remaining bound to the immobilized Cax-Pt template were next analyzed by Western blots. (F) Immobilized Cax-Pt (lanes 1) or ClaI-cut EC-400 (lanes 2–6) were treated as indicated at the top of the panel similarly to (E); the RNA pol IIO and CSB remaining on the immobilized Cax-Pt template were next analyzed by Western blots. (G) ClaI-cut Cax-Pt (lane 1) or ClaI-cut EC-400 complex (lane 2) were incubated with RISΔXPC, the dual incision system in which XPC is lacking. Following washes at 50 mM KCl, the remaining bound proteins were analyzed by Western blot. Download figure Download PowerPoint The dynamics of the stalled RNA pol IIO was studied by incubating the transcription elongation factor TFIIS, which stimulates the nascent RNA cleavage activity intrinsic of RNA pol II (Reines et al, 1993); its addition leads to a 97 nt transcript (Figure 2C, lane 2). Moreover, the addition of CSB helps RNA pol IIO to move forward (Selby and Sancar, 1997a) and to resume transcription from position 97 nt to position 105 nt (lane 3). We also checked whether XPC was able to target the DNA structure induced by the cisplatin damage when RNA pol IIO was stalled in front. Recombinant XPC was incubated with the ClaI-cut EC-400 complex (see Figure 2A, lane 3), for 30 min at 30°C, and the immobilized template was subsequently rinsed with a buffer containing 50 mM KCl and then further incubated with RISΔXPC. In these conditions, we did not observe the removal of damaged oligonucleotides, suggesting that XPC did not displace RNA pol IIO (Figure 2D, lane 3). Nor did we observe a repair signal when we added a challenge-damaged DNA fragment (Fgt-Pt) to RISΔXPC and the immobilized EC-400 complex (lane 2). However, when a nontranscribed damaged DNA was first incubated with XPC before addition of RISΔXPC, we observed dual incision activity (lane 4); this was not the case when a ClaI-cut DNA fragment was used instead (lane 1). Altogether, our data suggest that XPC neither binds to nor displaces a RNA pol II stalled in front of a DNA damage. We next investigated whether the stalled RNA pol IIO could be specifically recognized by NER factors. ClaI-cut EC-400 was then incubated with WCE/XPC in the presence of or in the absence of ATP. Most of the DNA repair factors including XPG, TFIIH, XPA, XPF/ERCC1 and RPA were recruited onto RNA pol IIO in the presence of ATP (Figure 2E, compare lanes 1 and 2). Interestingly, when ATP was omitted from the incubation, we observed a decrease in the recruitment of XPG, TFIIH, XPA and XPF/ERCC1 (lane 3), while RNA pol IIO was still bound to the template (Figure 2F, lanes 2 and 3). We also tested the influence of CSB in recruiting NER factors on RNA pol IIO. CSB has been suggested to be part of a complex containing RNA pol II, TFIIH and XPG (Iyer et al, 1996; Bradsher et al, 2002), and to allow for TFIIH recruitment onto the RNA pol II (Tantin et al, 1998). However, adding recombinant CSB to WCE/XPCΔCSB (WCE/XPC immunodepleted for CSB; Figure 2F, lane 5 and 6), as well as ATP on ClaI-cut EC-400, did not modify the recruitment of the NER factors (Figure 2E, lanes 7 and 8). In parallel, the amount of RNA pol IIO on the immobilized DNA remained constant (Figure 2F, lanes 5 and 6). To verify that the recruitment was dependent on the presence of the RNA pol IIO on the DNA, we treated RNA pol IIO with a phosphatase as dephosphorylation of a stalled RNA pol IIO CTD destabilizes the RNA pol II from the template when incubated next with a cellular extract (Tremeau-Bravard et al, 2004) (Figure 2F, compare lane 4 with lanes 2, 5 and 6). In these circumstances, we observed a drop in the presence of NER factors (Figure 2E, lanes 5 and 6) and CSB (Figure 2F, lane 4) associated with the loss of RNA pol IIO. Similarly, we showed that when the incubation was carried out in the presence of the five recombinant repair proteins and ATP, an increased recruitment of each protein onto RNA pol IIO could be detected compared to the nontranscribed template (Figure 2G). This further suggests that their recruitment does not necessarily require intermediate proteins. Altogether, our results show the ability of the lesion-stalled RNA pol IIO to specifically recruit DNA repair factors whose recruitment is stimulated by ATP. Furthermore, while CSB appears to be part of the complex, it does not influence the recruitment of the NER factors. CSB together with the stalled RNA pol II promote dual incision Having shown that RNA pol IIO can recruit repair factors, we next wondered whether this complex could mediate the removal of the lesion. Knowing that the lengths of the repair patches were similar for TC-NER and for GG-NER (Bowman et al, 1997), we used our in vitro DNA repair system to check for the release of a damaged single-strand fragment (Aboussekhra et al, 1995). When EC-400 was first incubated with RISΔXPC, no incision signal was detected. However, when increasing amounts of CSB were also added (Figure 3A), we detected a low but significant increased amount of released damaged oligonucleotides (lanes 3–5) characteristic of a repair signal. Nevertheless, to check whether such a dual incision pattern would have been nonspecifically generated by XPG and XPF, we incubated EC-400 and CSB with varying combinations of four of the five NER factors. The omission of TFIIH, XPA, RPA, XPG, or XPF completely abolished the repair reaction, suggesting that the removal of the damaged oligonucleotide is highly specific and not promoted by unspecific endonuclease cuts (Figure 3B, lanes 3–7). Figure 3.RNA pol IIO/CSB-mediated incision. (A) EC-400 was incubated with RISΔXPC and increasing amounts of CSB, and subjected to a dual incision assay. Quantification was made as described in Materials and methods. A graphic depicts the relative intensity of each signal. (B) Dual incision assays were performed on EC-400, which was incubated with different combinations of NER factors and CSB as indicated. (C) EC-400 transcription complex (lanes 1–12) and Cax-Pt (lanes 13–15) pretreated or not with CIP (lanes 5 and 6) were incubated with RISΔXPC (lanes 2–15) and with either wild-type CSB (lanes 3–6, 9 and 15), mutated CSBΔ440, CSBΔ378, CSBR670W (lanes 10–12) or XPC (lane 13), and subjected to a 3′-incision primer extension assay. The position and scans of sensitive bands relative to cisplatin lesion are denoted by asterisks and indicated at the right of the gel. (D) Dual incisions were performed on the untranscribed Cax-Pt with RISΔXPC in the presence (lanes 1, 3–11) or the absence of XPC (lanes 2, 12 and 13), with either a limiting amount of TFIIH (lanes 3–5) or XPG (lanes 9–11). CSB was added in the reaction where indicated. The relative amounts of TFIIH and XPG are indicated by thick boxes (saturating amounts) and thin boxes (limiting amounts). Download figure Download PowerPoint The RNA pol IIO-mediated dual incision is rather weak compared to an XPC-mediated dual incision and might reflect either the low efficiency of the reaction (and the absence of putative additional stimulatory repair factors) and/or a different incision pattern, which cannot be fully detected by our repair system due to the design of our probe. Therefore, instead of using a probe, the DNA fragments generated from the 3′-incision were amplified by primer extension using a radiolabeled oligonucleotide annealed 100 bp upstream from the site of the lesion. Increasing amounts of CSB together with the recruited factors on RNA pol IIO stimulated the 3′-incision activity (Figure 3C, compare lanes 3 and 4 to lane 2). We observed a particular increase in the intensity of the hypersensitive sites at positions +9, +12, +14, +16, +17 and +20 (Figure 3C, lanes 4 and 9), also depicted in the histogram in which the intensity of the bands were quantified and normalized to the background bands located at positions +8 and +24 found in every samples. Dephosphorylation of RNA pol IIO, which leads to its partial destabilization, resulted in a decreased recruitment of the NER factors (Figure 2E) and in a decreased of the 3′-incision pattern (Figure 3C, lanes 5 and 6). The 3′-incision positions mediated on the one hand by RNA pol IIO and CSB and on the other hand by XPC are rather similar (lanes 13 and 9). However, we reproducibly observed in the RNA pol IIO-mediated repair a more sensitive site at position C+11 and less-sensitive sites at positions T+17 and A+20 compared to GG-NER (Figure 3C, right scans). Furthermore, to evaluate the specificity of the CSB-dependent incision reaction, we tested several mutated CSB (provided by A Lehmann): (CSBΔ440, whose 440–446 glycine residues were deleted; CSBΔ378, which lacks seven glutamine residues (from position 378 to position 384); and CSBR670W, found within a CS patient, in which the arginine located in the conserved motif III of the ATPase domain was changed into a tryptophane). We observed a slight decrease incision activity with CSBΔ440, and an inhibition with CSBΔ378 and CSBR670W (Figure 3C, lanes 10–12). To further investigate the specificity of CSB for the RNA pol IIO-mediated incision, we tested the influence of CSB on an XPC-mediated repair reaction, and more specifically its effect on TFIIH and XPG as CSB interacts with these two factors (Iyer et al, 1996) (Figure 3D). Cax-Pt and increasing amounts of CSB were then incubated with RIS (lanes 6–8), RISΔXPC (lanes 12 and 13) or RIS containing limited amounts of either TFIIH (lanes 3–5) or XPG (lanes 9–11). CSB did not affect the efficiency of the repair reaction by stimulating TFIIH or XPG, nor did it stimulate the overall rate of the reaction. Moreover, when CSB alone was incubated with RISΔXPC (lanes 12 and 13), we did not observe any repair signal, suggesting that CSB did not promote the dual incision and that the CSB fraction was not contaminated by XPC. Altogether, these results strongly suggest that the RNA pol IIO-mediated incision reaction is CSB and NER factor specific, as well as XPC-independent. The coming of the TCR components We further investigated whether the recruitment of each of the repair proteins was interconnected. The EC-400 complex (Figure 4B, upper panel) was incubated with different combinations of the five NER factors (in which the omitted factor is referred to by Δ as indicated at the top of each lane), and the remaining RNA pol IIO-associated proteins were next detected by Western blot. Incubation of TFIIH, XPA, RPA, XPG and XPF resulted in their recruitment on the RNA pol IIO (Figure 4B, lane 1; see also Figure 2). We next found that TFIIH, revealed by the p62 subunit (Figure 4B, panel 2), could still be detected in the absence of RPA (lane 4), XPG (lane 5), and XPF (lane 6). In the latter case, we observed a slight decrease of TFIIH. Similarly, XPA (panel 3) could also be detected in ΔRPA (lane 4), ΔXPG (lane 5) and ΔXPF (lane 6). Interestingly, we observed an interdependence between TFIIH and XPA in which, in the absence of one, the other is less present on the damaged complex. We found that if TFIIH was almost completely absent in ΔXPA (lane 2), there was a significant drop of XPA recruitment in ΔTFIIH (lane 2). Since the omission of RPA, XPG or XPF did not prevent them from binding to the RNA pol IIO (panels 2 and 3), we conclude that TFIIH and XPA might be recruited first onto the RNA pol IIO. The level of RPA (panel 4) remained the same in the absence of ΔTFIIH (lane 2), ΔXPA (lane 3) or ΔXPG (lane 5), and only decreased in ΔXPF (lane 6), suggesting that the RPA is able to load independently from the rest of the NER factors onto the complex. As previously observed in GG-NER (Riedl et al, 2003), XPG (panel 5) is no longer recruited onto the RNA pol IIO in ΔTFIIH (lane 2) or ΔXPA (lane 3), and strongly decreases in ΔRPA (lane 4). However, XPG is still present when XPF is omitted (lane 6), suggesting that XPG is recruited late on the RNA pol IIO and likely independently of, or before, XPF. Regarding the recruitment of XPF (panel 6), its presence was strongly diminished when either TFIIH or XPA was omitted (lanes 2 and 3, respectively). Neither the omission of RPA nor the absence of XPG affected the presence of XPF onto the RNA pol IIO (lane 4). Therefore, XPF might be recruited late onto the RNA pol IIO and positioned in relation with XPG. Figure 4.Sequential assembly of NER factors onto RNA pol IIO. (A) Scheme depicting the reaction. Cax-Pt was transcribed and then cut by ClaI to form the ClaI-cut EC-400, which was further incubated with different combinations of four repair factors (the omitted factor is referred to by Δ in each lane), or with different combinations of indicated NER factors. After soft washes (50 mM KCl), the recruited proteins were analyzed by Western blots (B) and (F) or by a functional dual incision complementation assay (see text for details). (C–E) Either RNA pol IIO (B) or RNA synthesis levels (C), (D) and (E) are used as a loading control. Download figure Download PowerPoint The presence of an NER factor onto the RNA pol IIO following the first incubation can be further investigated for its requirement in the dual incision reaction in which the factor of interest is omitted (Riedl et al, 2003). In our experimental conditions, RNA synthesis, and consequently the amount of the elongating RNA pol IIO, was similar in every sample (Figure 4C–E). Similar to what we observed by Western blot, omission of either TFIIH or XPA in the first incubation strongly prevented the recruitment of XPG and XPF (Figure 4D and E, lanes 4 and 5, respectively), underlying the requirement of TFIIH and XPA for the further recruitment of NER factors on RNA pol IIO. Furthermore, we also observed a slight decrease of XPF when XPG was omitted (Figure 4E, lane 7). The recruitment of XPF slightly modified the concentration of TFIIH on the damaged DNA (Figure 4C) as it was observed in GG-NER, where the release of TFIIH from the damage DNA is concomitant with the arrival of XPF (Riedl et al, 2003). We also noticed that the presence of XPF did enhance the recruitment of RPA (Figure 4F, lane 5), a factor that would be later involved in DNA resynthesis. NER factor-mediated partial release of RNA pol II Next, questions arose regarding what becomes of RNA pol IIO during the reaction. EC-400 was incubated with the NER factors for 30 min at 30°C. The supernatant was then separated from the beads and analyzed by Western blot for the presence of RNA pol IIO. We also checked for the presence of remaining RNA pol IIO on the immobilized DNA. However, because of a very low percentage of RNA pol IIO released from the immobilized DNA (5–10%), the remaining RNA pol IIO on the immobilized DNA was in saturating amounts. The addition of each NER factor separately did not significantly promote the release of RNA pol IIO, even in the presence of ATP (Figure 5A, compare lane 1 to lanes 2–6). On the contrary, the sequential addition and recruitment of NER factors led to an increased release of RNA pol IIO from the template (Figure 5A, lanes 7–10). A maximum was obtained when all the NER factors were incubated together (lane 10). We also noticed that when incubated either alone or together with the NER factors, CSB does not help release RNA pol IIO from the DNA (Selby and Sancar, 1997b) (Figure 5B). Furthermore, while RNA pol IIO is 'destabilized' by the five NER factors in the absence of ATP, its removal from the template is stimulated by the addition of increasing amounts of ATP (Figure 5C, lanes 3–6). Both TFIIH and CSB exhibit a DNA-dependent ATPase activity; however, because CSB is not implicated in the release of RNA pol IIO, TFIIH is the most likely to use ATP to unwind the DNA. This hypothesis was confirmed by incubating RNA pol IIO with the repair factors either in the absence of TFIIH or in the presence of TFIIH/XPB-f99s, whose mutation is detrimental for TFIIH DNA-opening activity (Coin et al, 1999). In both cases the RNA pol IIO release is much weaker than in the presence of wild-type TFIIH (Figure 5D and E), likely due to the absence of the 'TFIIH helicase functioning'. Figure 5.NER factors-mediated release of RNA pol IIO. (A–E) EC-400 were incubated for 30 min at 30°C, either alone or together with NER factors and CSB as indicated at the top of each panel. The removal of RNA pol IIO from the DNA template in the supernatant was further analyzed by Western blots. (B) EC-400 was incubated with indicated factors in the absence or the presence of increasing amounts of ATP. IIH/XPB-f99s was also used (E). Quantification was made as described in Materials and methods. A graphic at the bottom of each panel depicts the relative intensity of each signal. Download figure Download PowerPoint Discussion Depending on the type of DNA damage, the elongating RNA pol IIO can stall and therefore face a situation in which it might trigger factors to allow for either the by
Referência(s)