Article - Cell Press

Molecular Cell

Article XPG Stabilizes TFIIH, Allowing Transactivation of Nuclear Receptors: Implications for Cockayne Syndrome in XP-G/CS Patients Shinsuke Ito,1 Isao Kuraoka,1,4 Pierre Chymkowitch,3 Emmanuel Compe,3 Arato Takedachi,1 Chie Ishigami,1 Fre´de´ric Coin,3 Jean-Marc Egly,3,* and Kiyoji Tanaka1,2,* 1

Laboratories for Organismal Biosystems, Graduate School of Frontier Biosciences, Osaka University, Osaka 565-0871, Japan Solution-Oriented Research for Science and Technology (SORST), Japan Science and Technology Agency (JST), 1-3 Yamadaoka, Suita, Osaka 565-0871, Japan 3 Institut de Ge´ne´tique et de Biologie Mole´culaire et Cellulaire, CNRS (UMR7104)/INSERM (U596)/ULP, BP 163, 67404 Illkirch Cedex, C.U. Strasbourg, France 4 Present address: Institute for Clinical Research, National Kyushu Cancer Center, 3-1-1 Notame, Fukuoka 811-1395, Japan. *Correspondence: [email protected] (K.T.), [email protected] (J.-M.E.) DOI 10.1016/j.molcel.2007.03.013 2

SUMMARY

Mutations in the human XPG gene give rise to an inherited photosensitive disorder, xeroderma pigmentosum (XP) associated with Cockayne syndrome (XP-G/CS). The clinical features of CS in XP-G/CS patients are difficult to explain on the basis of a defect in nucleotide excision repair (NER). We found that XPG forms a stable complex with TFIIH, which is active in transcription and NER. Mutations in XPG found in XP-G/CS patient cells that prevent the association with TFIIH also resulted in the dissociation of CAK and XPD from the core TFIIH. As a consequence, the phosphorylation and transactivation of nuclear receptors were disturbed in XP-G/CS as well as xpg/ MEF cells and could be restored by expression of wild-type XPG. These results provide an insight into the role of XPG in the stabilization of TFIIH and the regulation of gene expression and provide an explanation of some of the clinical features of XP-G/CS. INTRODUCTION Cockayne syndrome (CS) is a rare autosomal recessive disorder, which is characterized by a broad range of clinical symptoms such as growth failure, mental retardation, microcephaly, cataracts, optic atrophy, sensorineural hearing loss, and dental caries. In addition, CS patients exhibit photosensitivity of the skin but have no predisposition to sunlight-induced skin cancer (Andressoo and Hoeijmakers, 2005). Cells from CS patients show a reduced recovery of RNA synthesis after exposure to UV. The cellular abnormalities in CS have been attributed to a specific defect in transcription-coupled nucleotide excision repair (TC-NER), a subpathway of nucleotide excision

repair (NER), by which transcription-blocking damage is repaired rapidly (Friedberg et al., 2006; Hanawalt, 2002). Complementation analysis has defined two genetic complementation groups, CS-A and CS-B. Defects in NER also lead to other autosomal recessive disorders such as xeroderma pigmentosum (XP) and trichothiodystrophy (TTD) (Lehmann, 2003). XP is characterized by abnormal pigmentation and a predisposition to skin cancers in sun-exposed areas, while TTD is characterized by brittle hair and ichthyosis in addition to the features of CS. About half of TTD patients exhibit photosensitivity and have a NER defect. NER-defective XP is classified into seven complementation groups, XP-A to XP-G. XP-C and XP-E are specifically defective in a global genome nucleotide excision repair (GG-NER) that repairs the damage on the nontranscribed strand or in the genome overall, while the other XP groups involve deficiencies in both TC-NER and GG-NER. Photosensitive TTD is associated with mutations in XPB, XPD, and TTDA that encode three of the subunits of the general transcription factor TFIIH (GigliaMari et al., 2004). On the other hand, mutations of the XPA gene, which is indispensable for both TC-NER and GG-NER, do not cause CS, suggesting that the features of CS may not be exclusively linked to TC-NER. One clue to understanding the pathogenesis of the CStype features comes from the evidence that, in rare cases, mutations in XPB, XPD, and XPG cause features of CS combined with XP (XP-B/CS, XP-D/CS, and XP-G/CS). XPB and XPD are subunits of TFIIH, which is a multifunctional complex involved in basal transcription, transactivation, the cell cycle, and NER (Fisher, 2005; Zurita and Merino, 2003). TFIIH can be divided into two subcomplexes: the core TFIIH (IIH6) composed of XPB, p62, p52, p44, p34, and p8 (TTDA), and a cdk-activating kinase (CAK) subcomplex that contains cdk7, cyclin H, and MAT1. Both subcomplexes are bridged by XPD (Drapkin et al., 1996; Reardon et al., 1996). In NER, TFIIH unwinds the duplex DNA around the lesion to allow the recruitment of the NER factors XPA, RPA, XPG, and XPF-ERCC1 (Evans et al., 1997b). In basal transcription, TFIIH

Molecular Cell 26, 231–243, April 27, 2007 ª2007 Elsevier Inc. 231

Molecular Cell XPG Stabilizes TFIIH

functions in concert with at least four other basal transcription factors (TFIIB, TFIID, TFIIE, and TFIIF). In addition to having an indispensable role in NER, XPB helicase is essential for transcription initiation and promoter escape (Coin et al., 1999), steps regulated by phosphorylation in the carboxy-terminal domain (CTD) of the largest subunit of RNA polymerase II (RNA Pol II) mediated at least partially by the cdk7 kinase of TFIIH (Svejstrup et al., 1996). Although XPD helicase is dispensable for transcription, XPD facilitates optimal transcription by anchoring the CAK subcomplex to the core TFIIH (Tirode et al., 1999). Some mutations found in XP-D patients prevent the interaction of XPD with p44, resulting in the dissociation of CAK from the core TFIIH (IIH6) (Coin et al., 1998; Dubaele et al., 2003). It has been reported that TFIIH phosphorylates numerous nuclear receptors (NRs) and that in XP-D and XP-D/TTD cells, the ligand-dependent phosphorylation of NRs was reduced and the transactivation was not induced (Bastien et al., 2000; Chen et al., 2000; Compe et al., 2005; Drane et al., 2004; Rochette-Egly et al., 1997). These results suggested that the clinical features of XP-D and XP-D/TTD patients can be accounted for by the defect in basal transcription and transactivation as well as NER. The question remains as to how some mutations in the XPG gene, which encodes a structure-specific endonuclease required for making the 30 incision during NER, cause the features of CS in XP-G/CS (Clarkson, 2003). XPG interacts with hNTH1, CSB, and several subunits of TFIIH (Bessho, 1999; Bradsher et al., 2002; Iyer et al., 1996; Klungland et al., 1999; Sarker et al., 2005). Moreover, RAD2, a yeast counterpart of XPG, has been shown to play a role in the galactose-induced transcription of GAL7 and GAL10 (Lee et al., 2002). However, the causative relationship between these functions of XPG and the features of CS in XP-G/CS has not been clarified. Here, we showed that XPG forms a stable complex with TFIIH and functions in maintaining the architecture of TFIIH, underlining the contribution of XPG to transcription. Moreover, we demonstrated that the XPG mutations found in severe XP-G and XP-G/CS patients disturb the interaction of both CAK and XPD with the core TFIIH, resulting in defective transactivation of NRs. RESULTS Purification of the XPG Complex We first established a HEK293 cell line stably expressing a C-terminally FLAG-V5-6xHis-tagged XPG (e-XPG). The HEK293 nuclear extracts were first immunoprecipitated with an antibody directed against the FLAG tag (M2) followed by chromatography with a nickel chelate (Figure 1A). Silver staining indicated that the purified XPG formed a protein complex with several polypeptides (Figure 1B), which were then analyzed by mass spectrometry. As a result, several subunits of TFIIH (XPD, p52, p44, and p34) were identified. Immunoblot analysis revealed that the XPG complex contained the TFIIH subunits XPB,

XPD, p62, p44, cdk7, cyclin H, MAT1, and p8, but the mock-purified control did not (Figure 1C). Converse immunoprecipitation experiments using an antibody against the p44 subunit of TFIIH also revealed that the immunoprecipitated TFIIH (as depicted by XPB) from HeLa nuclear extract specifically contained XPG (Figure 1D). Moreover, this complex seems to be rather stable, since it was resistant to 0.4 M KCl. In addition, the six subunits of the core TFIIH were overexpressed together with XPG using a baculovirus/insect cell system. The immunoprecipitation of the insect cell extracts by the antibody against either p44 or XPG indicated that XPG binds directly with TFIIH (Figure 1E). The XPG complex was then loaded on a Superose 6 gel filtration column (Figure 1F). Silver staining revealed that XPG was eluted either with TFIIH (Figure 1F, fractions 8– 12) or alone (fractions 13–16). The immunoblot analysis showed that XPG overlapped with TFIIH as depicted by XPB (Figure 1F, lower panels). It is likely that fractions 13–16 contain XPG homodimers as previously reported (Constantinou et al., 1999), while fractions 8–12 contain the XPG-TFIIH complex. Taken together, these results indicated that XPG forms a stable complex with TFIIH in the cells. The XPG-TFIIH Complex Can Repair Damaged DNA and Phosphorylate the CTD We next examined whether the purified XPG-TFIIH complex is enzymatically active. It has been demonstrated that XPG makes a 30 incision where DNA is damaged during NER and cleaves specifically the junction of singlestranded/double-stranded DNA in bubble structures (Evans et al., 1997a). Therefore, a DNA substrate, which consisted of a centrally unpaired bubble of 30 nt, flanked by two duplex regions of 30 nt (Figure 2A), was incubated with the XPG-TFIIH complex. This resulted in the generation of 60 nt fragments cleaved at the 30 junction of a single-stranded/double-stranded DNA (Figure 2B). We also observed that such XPG endonuclease activity peaked in fractions 11–13 of the gel filtration column (Figures 1F and 2C). Then, the XPG-TFIIH complex was examined in a NER assay using a single cisplatinated DNA substrate (Figure 2D) and whole-cell extracts from either XP-B, XP-D, or XP-G cells (Araujo et al., 2000). When the DNA substrate was incubated with HeLa extracts, a 29 nt fragment that includes the damaged site was generated, indicating DNA repair synthesis (Figure 2E, lane 1). In contrast, extracts from three patients’ cells were defective in DNA repair synthesis (Figure 2E, lanes 2, 4, and 6). However, the ability to repair the damaged DNA in XP-B, XP-D, and XP-G cell extracts was restored by the addition of the XPG-TFIIH complex (Figure 2E, lanes 3, 5, and 7). These results indicated that the XPG-TFIIH complex was competent in terms of the NER activities of XPG and TFIIH. We then examined if the activity of the cdk7 kinase of TFIIH to phosphorylate the CTD of the largest subunit of RNA Pol II was maintained in the XPG-TFIIH complex.

232 Molecular Cell 26, 231–243, April 27, 2007 ª2007 Elsevier Inc.


Figure 1. XPG Forms a Protein Complex with TFIIH (A) Experimental procedure for purification of the XPG-TFIIH complex. (B) Silver staining of the XPG complex. As a control, a mock purification was performed with nuclear extracts from nontransfected HEK293 cells. Purified polypeptides were resolved by SDS-PAGE and visualized by silver staining. (C) The XPG complex was immunoblotted with antibodies against XPG and eight subunits of TFIIH. (D) Nuclear extracts of HeLa cells were immunoprecipitated with anti-p44 antibody in the presence of 0.4 M KCl. (E) Insect cells were infected with baculoviruses expressing either the core TFIIH (IIH6) (lane 1) or IIH6 plus XPG (lanes 2 and 3). The cell extracts were immunoprecipitated with antibody against p44 (lanes 1 and 2) or XPG (lane 3). The immunoprecipitants were resolved by SDS-PAGE, and XPG and TFIIH subunits were detected by immunoblotting using each antibody. (F) The XPG complex was further separated on a Superose 6 column. After fractionation, each fraction was resolved by SDS-PAGE and visualized by silver staining (upper panel) or immunoblotted with anti-XPG and anti-XPB antibodies (bottom panels). Estimated molecular sizes are depicted in the upper panel.



Figure 2. The XPG-TFIIH Complex Possesses Both 30 Endonuclease and TFIIH Activities In Vitro (A) The structure of the bubble substrate BS-A (90-mer) bearing a 30 nt unpaired region as described (Evans et al., 1997a). The DNA substrate was 50 -labeled with 32P on one strand. (B) Autoradiograph after denatured 12% PAGE of the cleaved products, demonstrating a structure-specific endonuclease activity of the XPG-TFIIH complex. BS-A (100 fmol) was incubated with the XPG-TFIIH complex at 30 C for the period indicated. M, size markers (MspI digest of pBR322 30 -labeled with 32P). (C) BS-A (100 fmol) was incubated at 30 C for 30 min with the XPG-TFIIH complex fractionated by Superose 6 gel filtration (Figure 1F). (D) Structure of the DNA substrate pBSII KS-GTG containing a specifically located 1,3-intrastrand d(GpTpG)-cisplatin crosslink. The expanded region illustrates the DNA sequence flanking the cisplatin crosslink. Seven BstNI restriction sites are indicated. Digestion of the closed circular pBSII KS-GTG DNA containing a cisplatinated adduct with BstNI generates a 29 nt fragment encompassing the DNA adduct and, to a lesser extent, larger fragments surrounding the adduct (140 and 241 nt). (E) Autoradiograph after denatured 14% PAGE, demonstrating DNA repair synthesis in the BstNI fragment containing a cisplatinated adduct (29 nt). DNA was incubated with [a-32P]dCTP and whole-cell extracts from HeLa (lane 1), XP11BE (XP-B) (lanes 2 and 3), XP17BE (XP-D) (lanes 4 and 5), or XP125LO (XP-G) (lanes 6 and 7) in the presence or absence of the XPG-TFIIH complex, and subsequently digested with BstNI before electrophoresis.



Figure 3. Truncated XPG Cannot Form an XPG-TFIIH Complex (A) Schematic representation of the different XPG proteins examined in this study. Hatched N and I boxes represent the conserved regions required for nuclease activity. Box C, conserved C-terminal region; NLS, nuclear localization signal. A792V mimics the mutant XPG protein presumably expressed in a patient with mild XP-G, XP125LO, while D926–1186 mimics the mutant XPG with a C-terminal truncation derived from XP-G/CS, XPCS1RO. (B) Each mutant XPG as well as WT XPG was fused with a FLAG-V5-6xHis tag at the C terminus and expressed in HEK293 cells. Whole-cell extracts were immunoprecipitated with antiFLAG M2 antibody. Purified XPG complexes were then analyzed by immunoblotting with the antibodies indicated.

When incubated with GST-CTD in the presence of [g-32P]ATP, the XPG-TFIIH complex phosphorylated the CTD substrate (Figure 2F).

Truncation Mutation in the C-Terminal Region of XPG Prevents Its Association with TFIIH We then examined if the mutant XPG proteins derived from XP-G and XP-G/CS patients affect the formation of the XPG-TFIIH complex (Figure 3A). The first XPG mutant contains the missense mutation A792V within the conserved I region of XPG found in a patient (XP125LO) with mild XP-G (Nouspikel and Clarkson, 1994). This mutation was reported to weaken the endonuclease activity of XPG (Constantinou et al., 1999). The second one mimicked a mutation in the patient XPCS1RO, who showed severe CS-type features as well as the XP phenotype and was homozygous for a T deletion at nucleotide position 2972, resulting in a frameshift after amino acid 925, followed by 55 amino acids that are unrelated to XPG (Emmert et al., 2002). The corresponding XPG-A792V and XPGD926-1186 cDNAs with the FLAG-V5-6xHis tag were subcloned into the vector pcDNA5/FRT and transfected into HEK293 cells. Lysates prepared from the corresponding stable cell lines were immunoprecipitated with anti-FLAG M2 and subjected to an immunoblot analysis. The wild-type (WT) and XPG-A792V, but not XPGD9261186, were able to immunoprecipitate TFIIH (Figure 3B).

Dissociation of Holo-TFIIH in XP-G Cells It has been reported that XPD interacts with the MAT1 subunit of CAK and with the p44 subunit of the core TFIIH and that the mutations in the C-terminal domain of XPD weakened the interaction with p44, resulting in the dissociation of both CAK and XPD from the core TFIIH (Coin et al., 1998; Keriel et al., 2002). The present finding that XPG forms a stable complex with TFIIH prompted us to investigate whether or not the composition of TFIIH was affected by the XPG mutations in the XP-G (XP3BR) and XP-G/CS (XPCS1RO and XPCS1LV) cell lines as well as the XPD and XPB mutations in the XP-D (XP6BE) and XP-D/CS (XPCS-2 and XP8BR) cell lines and the XP-B/CS (XPCS1BA) cell line, respectively. Mutations of the XPG, XPD, or XPB gene and clinical manifestations in each patient are summarized in Figure 4A and in Table S1 (in the Supplemental Data available with this article online). Whole-cell extracts were prepared from WI38VA13 (WT), BJ1 (WT), and the patients’ cell lines and immunoprecipitated with either anti-XPB antibody directed toward the C terminus of XPB or anti-XPG antibody directed toward the C terminus (Ser947 to Ala1165) of XPG in the presence of 150 mM NaCl. The immunoprecipitation using anti-XPB antibody revealed that, in the XP-G and XP-G/CS cell lines, which are supposed to produce the C-terminally truncated XPG, the XPB and p62 subunits of the core TFIIH were immunoprecipitated normally, while cdk7, a subunit of

(F) Autoradiograph after SDS-PAGE, demonstrating CAK activity. GST-CTD was incubated without (lane 1) or with increasing amounts of the XPGTFIIH complex (lanes 2–4).



Figure 4. XPG and CAK Were Dissociated from the Core TFIIH in the Extracts from XP-D, XP-D/CS, XP-G, and XP-G/ CS Cells (A) Schematic representation of the mutant XPG presumably expressed in XP3BR, XPCS1RO, and XPCS1LV. Hatched N and I boxes and NLS represent the same region as described in Figure 3A. Black boxes indicate non-XPG residues. The numbers on the right side of the boxes indicate the total length of the WT and mutant XPG proteins. (B and C) Immunoprecipitation of whole-cell extracts derived from XP6BE, XPCS-2, XP8BR, XP3BR, XPCS1RO, and XPCS1LV cells as well as WI38VA13 (WT) and BJ1 (WT) cells using anti-XPB (B) and anti-XPG (C) antibodies, respectively.

the CAK subcomplex, was almost absent (Figure 4B, lanes 4, 5, and 7). Such a dissociation of CAK from the core TFIIH was also observed in the XP-D/CS cell lines (Figure 4B, lanes 3 and 10). In addition, XPG was weakly associated with the core TFIIH in the XP-D/CS cell lines (Figure 4B, lanes 3 and 10) when compared with the WT and XP-B/CS cell lines (Figure 4B, lanes 1, 6, 8, and 9), although XPG was produced normally in these cell lines (Figure 4C, lanes 3 and 10). In XP6BE (XP-D) cells, cdk7 and XPG were slightly dissociated from the core TFIIH in the presence of 150 mM NaCl (Figure 4B, lane 2). It has been reported that in HD2 cells, which contain the same XPD mutation as XP6BE cells, cdk7 is completely dissociated from the core TFIIH in the presence of 0.4 M KCl, while it is not dissociated in the presence of 50 mM KCl (Keriel et al., 2002). In WT HeLa cells, cdk7 is not dissociated even in the presence of 0.4 M KCl. These results sug-

gest that XP6BE cells have an unstable TFIIH depending on the salt concentration, leading to less dissociation of cdk7 and XPG from the core TFIIH than in the other XPD/CS cell lines in the presence of 150 mM NaCl (Figures 4B and 4C, lanes 2, 3, and 10). The different levels of dissociation of cdk7 and XPG from the core TFIIH are probably characteristic to the XPD mutation (R683W) in XP6BE. The immunoprecipitation using anti-XPG antibody revealed that, in the WT and XP-B/CS cell lines, normal amounts of TFIIH were coimmunoprecipitated with XPG, as depicted by XPB, p62, and cdk7 subunits (Figure 4C, lanes 1, 6, 8, and 9). However, XPG was not immunoprecipitated, and therefore no TFIIH was coimmunoprecipitated in the XP-G and XP-G/CS cell lines (Figure 4C, lanes 4, 5, and 7). In the XP-D and XP-D/CS cell lines, normal amounts of XPG were immunoprecipitated, but smaller or only trace amounts of the core TFIIH and cdk7 were



coimmunoprecipitated with XPG (Figure 4C, lanes 2, 3, and 10). These results indicated that the mutations in XPG led to the dissociation of CAK from the core TFIIH, while the mutations in XPD that affect the anchoring of CAK to the core TFIIH also disturbed the binding of XPG to the core TFIIH. Therefore, both XPG and XPD are required for the association of the CAK subcomplex with the core TFIIH. Introduction of WT, but Not Truncated XPG, Restores the Stability of TFIIH in XPG-Downregulated Cells To provide further evidence that XPG is required for the maintenance of the TFIIH architecture, we established two XPG-downregulated HeLa cells (HeLa/siXPG-1 and HeLa/siXPG-2) by stably expressing short-hairpin RNA (siXPG-1 and siXPG-2) derived from the vector pSUPERIOR (Figure 5A). As a control, we established cells expressing short-hairpin RNA for the luciferase GL2 sequence (HeLa/siLuc). Real-time quantitative PCR and immunoblot analyses revealed that siXPG-1 and siXPG2 specifically repressed the expression of XPG in HeLa cells, although levels of TFIIH subunits were not affected (Figures 5B and 5D, lanes 1–3). These XPG-downregulated HeLa cells showed about 10-fold hypersensitivity to UV irradiation as compared with HeLa/siLuc cells (Figure 5C). We then investigated the stability of TFIIH in the XPGdownregulated HeLa cells. Again, anti-XPB antibody was unable to immunoprecipitate the entire TFIIH in either of the XPG-downregulated HeLa cells. XPG, XPD, p62, and cdk7 were coimmunoprecipitated in HeLa/siLuc cells, whereas the binding of XPD and cdk7 to the core TFIIH was compromised in XPG-downregulated HeLa cells (Figure 5D, lanes 4–6). Such phenomena were also observed in HEK293 cells in which the expression of XPG was downregulated by siXPG-2 (HEK293/siXPG-2) (Figures 5E and 5F, lanes 2 and 5). Then the HEK293/ siXPG-2 cells were corrected with the V5-tagged WT XPG cDNA (XPG-2M-V5) in which silent mutations were introduced to prevent the siXPG-2-induced downregulation (Figures 5A and 5F, lanes 3 and 6). In the V5-tagged WT XPG (XPG-2M-V5)-corrected HEK293/siXPG-2 cells, using either anti-XPG antibody or anti-XPB antibody, we were able to immunoprecipitate XPG together with TFIIH including XPD and cdk7 (Figure 5F, lanes 3 and 6; arrow indicates XPD bound to the core TFIIH). HEK293/siXPG2 cells were also transfected with either the V5-tagged XPG-A792V or the XPGD926-1186 mutant cDNA (Figure 3A). Immunoprecipitation of the corresponding cell extracts with anti-XPB antibody showed that the overexpression of either XPG-WT or XPG-A792V allowed the binding of XPD and cdk7 to the core TFIIH when compared with that in the parental HEK293/siXPG-2 cells (compare lanes 1 and 3 in Figure 5G). On the other hand, the binding of XPD and cdk7 to the core TFIIH was not restored in the XPGD926-1186-transfected cells (compare lanes 1 and 4 in Figure 5G). The relative amount

of cdk7 bound to the core TFIIH, which was standardized by the expression level of each XPG protein, is shown in the histogram in the lower panel of Figure 5G. Similar results were obtained when the immunoprecipitation was performed with anti-p44 antibody (Figure S1). These observations suggested that the C-terminal portion of XPG is required for maintaining CAK and XPD together with the core TFIIH and that the A792V mutation of XPG found in mild XP-G patient retains the ability to stabilize the XPD and CAK within TFIIH. XP-G Cells Are Deficient in Ligand-Induced Transactivation of Nuclear Receptors The above data showed how the XPG mutation in severely affected XP-G and XP-G/CS patients could be detrimental in maintaining the architecture of TFIIH within the XPGTFIIH complex. On the other hand, it has been reported that NRs are phosphorylated by CAK within TFIIH (Bastien et al., 2000; Chen et al., 2000; Compe et al., 2005; Drane et al., 2004; Rochette-Egly et al., 1997). These findings led us to examine the ligand-induced transactivation of NRs in XP6BE, XPCS-2, XP3BR, XPCS1RO, and xpg/ MEF cells as well as WI38VA13 and xpg+/+ MEF cells. Cells were cotransfected with a reporter vector containing an estrogen-responsive element upstream of the promoter (pERE-Luc), an internal control vector encoding b-galactosidase (pCH110), and either an expression vector for ERa (pcDNA3-ERa) or an empty vector (pcDNA3) (Keriel et al., 2002). In the absence of ERa transfection, the luciferase reporter gene is not activated by treatment with 17b-estradiol (E2) (Figures 6A and 6B). Upon expressing ERa, WI38VA13 and xpg+/+ MEF cells showed ligand-induced transactivation, whereas XP6BE, XPCS-2, XP3BR, XPCS1RO, and xpg/ MEF cells did not (Figures 6A and 6B). Moreover, defective E2-induced transactivation of ERa in the xpg/ MEF and XPCS1RO cells was restored by the introduction of WT XPG cDNA (Figures 6B and 6C), confirming that the defective E2-induced transactivation of ERa in xpg/ MEF and XP-G/CS cells is due to the XPG mutations. We then examined whether the defect in ERa-mediated transactivation observed in XP-G cells could be due to underphosphorylation of ERa, knowing that the Ser118 of ERa is targeted by TFIIH in an E2-dependent manner (Chen et al., 2000). After transfection with pcDNA3-ERa and subsequent treatment with E2 ligand, cell lysates were prepared for immunoblot analysis using an antibody directed toward the phosphorylated Ser118 of ERa. ERa was significantly phosphorylated in WI38VA13 cells in the presence of E2, while very little ERa was phosphorylated in XPCS1RO cells (Figure 6D, lanes 1–6). Interestingly, in the XPG cDNA-corrected XPCS1RO cells (XPCS1RO/XPG-GFP), the E2-induced phosphorylation of ERa was restored (Figure 6D, lanes 7–9). Taken together, these results indicate that the XPG mutation that affected the integrity of TFIIH also decreased the E2induced phosphorylation of ERa and consequently the TFIIH transactivation activity.



Figure 5. Dissociation of CAK and XPD from the Core TFIIH in the XPG-Downregulated Cells (A) Two XPG siRNA sequences are depicted. Cells expressing luciferase (control)- or XPG-siRNA were established. To express siXPG-2-resistant V5 tagged-XPG, silent mutations were introduced into XPG cDNA (XPG-2M). (B) Real-time quantitative PCR analysis of XPG, XPB, and GAPDH was performed with cDNA prepared from control siLuc (black)-, siXPG-1 (gray)-, or siXPG-2 (white)-transfected HeLa cells. The histograms represent the mean ± SEM of the relative expression level normalized with actin in three independent experiments.



DISCUSSION In the present study, we have isolated a transcription/DNA repair complex containing both XPG and TFIIH. Moreover, we have found that mutations in XPG partially disrupted the architecture of TFIIH itself within the XPG-TFIIH complex in XPG-deficient cells. As a consequence, E2induced transactivation of ERa was impaired in these cells due to the disintegration of TFIIH subunits. XPG Forms a Stable Complex with TFIIH Several lines of evidence have indicated that the CS-type features of XP-G/CS patients are related to a defect in some additional functions of XPG besides its endonuclease activity to remove damaged oligonucleotides (Clarkson, 2003). By an epitope tag method, we isolated an XPG protein complex that contains TFIIH (Figure 1). It is worth noting that all of the gene products responsible for XP/CS (XPB, XPD, and XPG) are integrated in the XPGTFIIH complex. Interaction between XPG and TFIIH (Araujo et al., 2001; Dunand-Sauthier et al., 2005; Gervais et al., 2004; Mu et al., 1995; Sarker et al., 2005; Thorel et al., 2004) has been reported, and the focus has been on the NER function. Consistent with these reports, we showed here that the XPG-TFIIH complex possessed a structure-specific 30 endonuclease (Figures 2B and 2C) and was able to complement a defect in DNA repair synthesis in XP-G, XP-B, and XP-D cell extracts (Figure 2E). These results indicated that the XPG-TFIIH complex is competent to repair damaged DNA. However, no other NER factor, such as XPA, RPA, XPF-ERCC1, or XPCHR23B, was included in the XPG-TFIIH complex (Figure S2 and data not shown), indicating that XPG-TFIIH is not preassembled into the overall NER protein complex. We assumed that the XPG-TFIIH complex is required for transcription besides NER. This hypothesis is supported by the fact that some types of mutations in XPG, XPB, and XPD, but not in the other XP genes, cause similar clinical features in several XP/CS patients. Relevant to transcription, the XPG-TFIIH complex could allow the phosphorylation of the CTD of the largest subunit of RNA Pol II in vitro (Figure 2F), indicating that the XPG-TFIIH complex is competent in CAK activity. Moreover, we found that the XPG-TFIIH complex is active in basal transcription in vitro (data not shown).

The XPG-TFIIH Complex Is Altered in XP-G and XP-G/CS Importantly, we showed that the A792V mutant XPG derived from a patient with mild XP-G (XP125LO) could form a complex with TFIIH similar to the WT XPG, while the D926–1186 truncated XPG derived from a XP-G/CS patient (XPCS1RO) could not (Figure 3). These results suggest that the C-terminal region of XPG is required for the formation of the XPG-TFIIH complex. Indeed, a short internal deletion (D225–231 amino acids) of XPG impaired the interaction with TFIIH (Thorel et al., 2004). Considering that broader regions of XPG are required for interactions with subunits of TFIIH (Iyer et al., 1996), a full-length XPG may be required for the formation of a stable XPGTFIIH complex (Figure 7). We also found that the binding of cdk7 and XPD to the core TFIIH was decreased in the XP-G (XP3BR) and XP-G/CS (XPCS1RO and XPCS1LV) cells (Figure 4B) as well as in the XPG-downregulated HeLa and HEK293 cells (Figures 5D and 5G). These results implied that XPG facilitates and/or stabilizes the binding of XPD and the anchoring of CAK to the core TFIIH and suggested that the inadequate interaction between XPG and TFIIH will disturb the architecture of TFIIH and its function in the transcription process. This phenomenon was also observed in XP-D/CS cells. Indeed, coimmunoprecipitation using anti-XPB antibody revealed that the C-terminal mutations in XPD derived from XP-D/CS cells (XPCS-2 and XP8BR) weakened the interaction between CAK and the core TFIIH, while the XPD-R683W mutant derived from a XP-D patient (XP6BE) retained an ability to anchor CAK to the core TFIIH in the context of the XPG-TFIIH complex under 150 mM NaCl (Figure 4B). Moreover, the interaction between XPG and TFIIH was decreased slightly in the XP-D (XP6BE) and significantly in XP-D/CS (XPCS-2 and XP8BR) cells (Figures 4B and 4C). It has been shown that XPD directly interacts with the MAT1 subunit of CAK and p44, another subunit of the core TFIIH (Sandrock and Egly, 2001). Therefore, some mutations in XPD weaken the binding of XPD to MAT1, p44, and/or XPG, thus disturbing TFIIH’s architecture and resulting in the dissociation of XPD, CAK, and/or XPG from the core TFIIH. Taken together, the present results indicated that XPG is required for the stable association of CAK and XPD with the core TFIIH, while XPD is

(C) UV sensitivity of XPG-knockdown cells as determined by the colony assay. siLuc (closed circles)-, siXPG-1 (closed triangles)-, or siXPG-2 (open squares)-transfected HeLa cells were UV irradiated at the indicated dose. Results represent the mean ± SD of three independent experiments. (D) Immunoblot of whole-cell extracts from siLuc-, siXPG-1-, or siXPG-2-transfected HeLa cells (lanes 1–3). Whole-cell extracts were then immunoprecipitated with anti-XPB antibody, and the immunoprecipitants were analyzed by immunoblotting with the antibodies indicated (lanes 4–6). (E) Real-time quantitative PCR analysis was performed as in (B) with cDNA prepared from siLuc (black)- and siXPG-2 (white)-transfected HEK293 cells, respectively. (F) XPG-downregulated cells were transfected with pcDNA5 vectors encoding siRNA-resistant WT XPG. Twenty-four hours later, cells were lysed, and then XPG and TFIIH were immunoprecipitated with anti-XPG (lanes 1–3) or anti-XPB (lanes 4–6) antibodies, followed by immunoblotting with the antibodies indicated. (G) WT and mutant XPG cDNA, which avoid siRNA-induced downregulation, were transfected into XPG-downregulated HEK293 cells, respectively. Twenty-four hours later, cells were lysed, and then XPG and TFIIH were immunoprecipitated with anti-XPB antibody. The histogram in the lower panel presents the relative amount of cdk7 bound to the core TFIIH normalized by the expression level of each XPG protein.



Figure 6. Nuclear Receptor-Mediated Transactivation in WT, XP-D, XP-D/CS, XP-G, XP-G/CS, and xpg/ Cells (A) SV40-transformed WT, XP-D, XP-D/CS, XP-G, and XP-G/CS cells were cotransfected with pERE-Luc, pCH110, and either pcDNA3-ERa or empty pcDNA3 and subsequently treated with 107 M 17b-estradiol (E2) for 16 hr. Luciferase activity was measured and normalized relative to b-galactosidase activity. Histograms show the ratio of the luciferase activity with or without E2 and indicate the values without (open boxes) or with ERa (closed boxes). The results are the mean ± SEM of at least three independent experiments performed in duplicate. (B) ERa-mediated transactivation in the xpg+/+ and xpg/ MEF, and the MEF transfected with pcDNA5-XPG. The transactivation assay was performed as described in (A). Histograms indicate the E2-induced transactivation in the xpg+/+ MEF (open boxes) or the xpg/ MEF (closed boxes). (C) E2-induced transactivation of ERa in WI38VA13, XPCS1RO, and XPG-GFP cDNA-corrected XPCS1RO cells. The transactivation assay was performed as described in (A). (D) E2-induced phosphorylation of ERa in WI38VA13, XPCS1RO, and the XPG-GFP-corrected XPCS1RO cells. After transfection with pcDNA3-ERa and subsequent treatment with E2, cell lysates were prepared for immunoblot analysis using an antibody elicited against ERa specifically phosphorylated at Ser118. b-tubulin was used as an internal control. The quantitative analysis of ERa phosphorylation represents the ratio ERa phosphorylated at Ser118/ERa signals and was set up to 1 for WI38VA13 cells.

required for the anchoring of CAK and XPG to the core TFIIH. These observations suggest that XPG and XPD cooperatively mediate the anchoring of CAK to the core TFIIH. Defective Transactivation of Nuclear Receptors in Severe XP-G and XP-G/CS Cells The clinical features of CS in XP-G/CS may not be exclusively linked to a defect in either GG-NER or TC-NER. This idea is supported by the notion that (1) mutations of the XPA gene, which is indispensable for both TC-NER and

GG-NER, do not cause CS and (2) although xpa/ and xpg/ mice shared defects in both TC-NER and GGNER, only xpg/ mice showed CS-like features such as postnatal growth failure and early death before weaning (Shiomi et al., 2005). Additionally, it was suggested that the hypoplasia of adipose tissues in XP-D/TTD mice is due to a defective transactivation of PPAR NRs, which are essential for lipid metabolism, differentiation, and the survival of adipocytes in vivo (Compe et al., 2005). Furthermore, it is remarkable that CS and TTD show similar phenotypes such as postnatal growth failure, neurological



Figure 7. Schematic Model for the Role of XPG in Maintaining the Integrity of TFIIH and Implications for CS Features in XP-G/CS Patients WT XPG forms a complex with TFIIH and functions in maintaining the integrity of TFIIH (left panel). Full-length XPG with a missense mutation in the nuclease domain that was found in a patient with mild XP-G forms a complex with TFIIH as the WT XPG does (middle panel), while a mutant XPG with a C-terminal deletion that is derived from a XP-G/CS patient caused the dissociation of CAK and XPD from the core TFIIH (right panel).

abnormalities, loss of subcutaneous fat tissue, hypogonadism, and so on. These results suggest that the CS features could be caused by an abnormality in the transcriptional processes as well. Protein modifications such as phosphorylation are crucial to the transcription process. Besides the phosphorylation of the CTD of RNA PoI II, TFIIH is involved in the phosphorylation of activators such as NRs, which is crucial for optimal ligand-dependent transactivation. Some mutations in the C-terminal end of XPD cause a defect in the anchoring of CAK to the core TFIIH, thereby reducing the phosphorylation and transactivation of NRs in response to the ligands (Compe et al., 2005; Drane et al., 2004; Keriel et al., 2002). We reasoned that the dissociation of CAK from the core TFIIH caused by XPG mutations also leads to a transactivation defect. In line with this assumption, our results indicated that severe XP-G and XP-G/CS cells as well as xpg/ MEFs could not allow a transactivation process such as the one mediated by ERa after treatment with E2 ligand. In this case, we indeed showed that the E2-induced phosphorylation of ERa was deficient in XPCS1RO cells (Figure 6D). However, the defective E2-induced phosphorylation and transactivation of ERa in the XPCS1RO cells were restored to normal levels by the introduction of WT XPG, confirming that the defective transactivation in XPCS1RO cells is due to the XPG mutation (Figures 6C and 6D). Moreover, a similar transactivation deficiency was observed in XP3BR and XPCS1RO cells with regards to other NR (PPARg2) (data not shown). Consistent with these findings, E2-induced transactivation of ERa was normal in XP-A (XP12RO) cells, while it was deficient in two XP-G/CS cell lines that have truncation mutations of XPG (XPCS1RO and XPCS1LV) (Figure S3). Taken together, these results suggested that the features of CS in XP-G/CS patients, such as hypogonadism and loss of subcutaneous fat tissue, are at least partly due to a defective transactivation of NRs such as

ERa and PPARg2, which resulted from a loss of TFIIH’s integrity caused by the truncation mutation of XPG. Our results raise the question of whether the CS patients with mutations in CSA and CSB, who show the same features of CS as XP-G/CS patients, have an unstable TFIIH and therefore exhibit the same transactivation deficiency as XP-G/CS patients. We examined the stability of TFIIH in CS3BE (CS-A) and CS1AN (CS-B) cells. As shown in Figure S4, cdk7 was normally integrated in TFIIH in both cell lines. Moreover, the activity of ligand-induced transactivation of ERa was normal in CS1AN cells (our unpublished data), suggesting that the CS features in CS-A and CS-B patients are not caused by a decrease in the stability of TFIIH. We also demonstrated that (1) XP-B/ CS cells have an intact TFIIH without any dissociation of CAK from the core TFIIH (Figures 4B and 4C) and (2) CSB participates in the early step of the transcription process as it is associated with RNA Pol I (Bradsher et al., 2002) and RNA Pol II at the promoter (Proietti-De-Santis et al., 2006). Thus, the same features of CS could be caused by various transcriptional deficiencies. Although CSA and CSB are not involved in the stability of TFIIH and the phosphorylation of NRs, one should emphasize the possibility that they may function in transcription initiation and/or elongation during the transactivation processes of NRs. In conclusion, we showed that XPG forms a stable protein complex with TFIIH and is involved in maintaining the integrity of TFIIH in cooperation with XPD. The mutations of XPG found in the patients with severe XP-G and XPG/CS resulted in an alteration of the architecture of TFIIH, leading to a dysregulation of gene expression as illustrated in the case of transactivation by ERa. These results indicate that XPG would play a crucial role in transcription in helping TFIIH to perform some of its functions, and that the features of CS in XP-G/CS are at least partly due to abnormal transcriptional activity.



EXPERIMENTAL PROCEDURES Purification of XPG Complex HEK293 cells stably expressing C-terminally FLAG-V5-6xHis-tagged XPG (e-XPG) were established by using a pcDNA5/FRT/V5-His TOPO TA Expression Kit (Invitrogen) according to the manufacturer’s instructions. The XPG complex was purified from nuclear extracts of HEK293 cells expressing e-XPG by immunoprecipitation with antiFLAG M2-conjugated agarose (Sigma) (Nakatani and Ogryzko, 2003). The bound polypeptides were eluted with the FLAG peptide and further affinity purified with nickel-NTA agarose (QIAGEN). The bound proteins were eluted with imidazole and dialyzed against buffer containing 20 mM Tris-HCl (pH 8.0), 100 mM KCl, 0.2 mM EDTA, 20% glycerol, and 1 mM dithiothreitol. The obtained XPG complex was further purified using Superose 6 PC 3.2/30 (GE Healthcare) with buffer containing 20 mM Tris-HCl (pH 8.0), 150 mM KCl, 10% glycerol, 0.1% Tween 20, and 10 mM 2-mercaptoethanol. In Vitro Assays The nuclease assay was carried out as described (Evans et al., 1997a). The DNA repair synthesis assay was carried out as described with some modifications (Araujo et al., 2000). Briefly, reaction mixtures (10 ml) containing 50 ng of pBSII KS-GTG plasmid with a cisplatinated adduct and whole-cell extracts were incubated in the presence of [a-32P]dCTP. DNA was recovered from the reaction mixtures and digested with BstNI at 60 C for 1 hr. The incorporation of [a-32P]dCTP into 29 nt of the BstNI fragment was examined for evaluating DNA repair synthesis. Whole-cell extracts were prepared as described previously (Evans et al., 1997b). The CTD kinase assay was carried out as described (Coin et al., 1999). GST-CTD was purchased from ProteinOne. Real-Time Quantitative PCR Total RNA was isolated by using an RNeasy Mini kit (QIAGEN) and reverse transcribed by using TaqMan Reverse Transcription Reagents (Applied Biosystems) and the random hexamer primer. Real-time PCR was performed by using the TaqMan Universal PCR Master Mix and a TaqMan Probe and was detected with an ABI PRISM 7700 Sequence Detection System (Applied Biosystems). Immunoprecipitation Cells were lysed for 30 min on ice in lysis buffer containing 50 mM TrisHCl (pH 7.8), 150 mM NaCl, 1 mM EDTA, 1% Nonidet P-40, 10 mM 2-mercaptoethanol, 0.5 mM phenylmethylsulfonyl fluoride, and a protease inhibitor cocktail (complete, Roche Diagnostics). Lysates were clarified by centrifugation, and the protein concentrations were determined with protein assay (Bio-Rad). Equal amounts of lysate were incubated at 4 C for 3 hr with protein G Sepharose beads (GE Healthcare) precoupled to each antibody and then washed extensively with lysis buffer. Bound materials were detected by immunoblotting.

Transactivation Assays The transactivation assay was performed as described previously with some modifications (Compe et al., 2005). Approximately 60% confluent cells cultured overnight were transfected with plasmid using the transfection reagent Jet-PEI (Polytransfection). Six hours later, the cells were refed with red phenol-free medium containing 5% fetal calf serum (FCS) and 5% charcoal-treated FCS. After 16 hr of incubation, cells were treated with the vehicle or 17b-estradiol for 16 hr. Cells were then washed with PBS and lysed, and assays for luciferase and b-galactosidase were performed as previously described (Compe et al., 2005). Supplemental Data Supplemental Data include Supplemental Experimental Procedures, four figures, one table, and Supplemental References and can be found with this article online at http://www.molecule.org/cgi/content/ full/26/2/231/DC1/. ACKNOWLEDGMENTS We thank Wim Vermeulen for XPCS1RO cells and XPG-GFP cDNAcorrected XPCS1RO cells, Alan R. Lehmann for XP8BR cells, Tadahiro Shiomi for xpg/ and xpg+/+ cells, Richard D. Wood for whole-cell extracts from XP cells, and Colm Ryan for critical reading of the manuscript. K.T. is supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science, and Technology of Japan (MEXT) and Solution-Oriented Research for Science and Technology (SORST) of Japan Science and Technology Agency (JST). J.-M.E. and F.C. are supported by a European contract (CEE Integrated Project LSHG-CT-2005-512113), the Association pour la Recherche sur le Cancer grant (number 3113), and the Agence Nationale de la Recherche Maladies Rares Grant (number ANR-05MRAR-005-02). Received: September 12, 2006 Revised: December 25, 2006 Accepted: March 5, 2007 Published: April 26, 2007 REFERENCES Andressoo, J.O., and Hoeijmakers, J.H. (2005). Transcription-coupled repair and premature ageing. Mutat. Res. 577, 179–194. Araujo, S.J., Tirode, F., Coin, F., Pospiech, H., Syvaoja, J.E., Stucki, M., Hubscher, U., Egly, J.M., and Wood, R.D. (2000). Nucleotide excision repair of DNA with recombinant human proteins: definition of the minimal set of factors, active forms of TFIIH, and modulation by CAK. Genes Dev. 14, 349–359.

Antibodies The mouse monoclonal antibodies against TFIIH subunits and ERa were described previously (Chen et al., 2000; Coin et al., 1998, 2006). Other antibodies were purchased from commercial suppliers as follows: anti-XPG, anti-XPB, anti-XPD, anti-p62, and anti-XPA, Santa Cruz Biotechnology, Inc.; anti-cdk7 and anti-XPC, Abcam; and anti-V5 tag, Invitrogen.

Araujo, S.J., Nigg, E.A., and Wood, R.D. (2001). Strong functional interactions of TFIIH with XPC and XPG in human DNA nucleotide excision repair, without a preassembled repairosome. Mol. Cell. Biol. 21, 2281– 2291.

siRNA Experiments The small interfering RNA (siRNA) for XPG and luciferase GL2 control were constructed in the vector pSUPERIOR (Oligoengine) to include the specific 19 bp target sequence for siXPG-1 (50 -GGAGAAAGA ATTTGAGCTA-30 ) and siXPG-2 (50 -GAAAGAAGATGCTAAACGT-30 ) and for siLuc (50 -CGTACGCGGAATACTTCGA-30 ). All constructs were verified by sequencing. HeLa and HEK293 cells were transduced with each siRNA plasmid, and the transfectants were selected with medium containing 1 mg/ml puromycin.

Bessho, T. (1999). Nucleotide excision repair 30 endonuclease XPG stimulates the activity of base excision repairenzyme thymine glycol DNA glycosylase. Nucleic Acids Res. 27, 979–983.

Bastien, J., Adam-Stitah, S., Riedl, T., Egly, J.M., Chambon, P., and Rochette-Egly, C. (2000). TFIIH interacts with the retinoic acid receptor gamma and phosphorylates its AF-1-activating domain through cdk7. J. Biol. Chem. 275, 21896–21904.

Bradsher, J., Auriol, J., Proietti de Santis, L., Iben, S., Vonesch, J.L., Grummt, I., and Egly, J.M. (2002). CSB is a component of RNA Pol I transcription. Mol. Cell 10, 819–829. Chen, D., Riedl, T., Washbrook, E., Pace, P.E., Coombes, R.C., Egly, J.M., and Ali, S. (2000). Activation of estrogen receptor alpha by



S118 phosphorylation involves a ligand-dependent interaction with TFIIH and participation of CDK7. Mol. Cell 6, 127–137.

Hanawalt, P.C. (2002). Subpathways of nucleotide excision repair and their regulation. Oncogene 21, 8949–8956.

Clarkson, S.G. (2003). The XPG story. Biochimie 85, 1113–1121.

Iyer, N., Reagan, M.S., Wu, K.J., Canagarajah, B., and Friedberg, E.C. (1996). Interactions involving the human RNA polymerase II transcription/nucleotide excision repair complex TFIIH, the nucleotide excision repair protein XPG, and Cockayne syndrome group B (CSB) protein. Biochemistry 35, 2157–2167.

Coin, F., Bergmann, E., Tremeau-Bravard, A., and Egly, J.M. (1999). Mutations in XPB and XPD helicases found in xeroderma pigmentosum patients impair the transcription function of TFIIH. EMBO J. 18, 1357–1366. Coin, F., Marinoni, J.C., Rodolfo, C., Fribourg, S., Pedrini, A.M., and Egly, J.M. (1998). Mutations in the XPD helicase gene result in XP and TTD phenotypes, preventing interaction between XPD and the p44 subunit of TFIIH. Nat. Genet. 20, 184–188. Coin, F., De Santis, L.P., Nardo, T., Zlobinskaya, O., Stefanini, M., and Egly, J.M. (2006). p8/TTD-A as a repair-specific TFIIH subunit. Mol. Cell 21, 215–226. Compe, E., Drane, P., Laurent, C., Diderich, K., Braun, C., Hoeijmakers, J.H., and Egly, J.M. (2005). Dysregulation of the peroxisome proliferator-activated receptor target genes by XPD mutations. Mol. Cell. Biol. 25, 6065–6076. Constantinou, A., Gunz, D., Evans, E., Lalle, P., Bates, P.A., Wood, R.D., and Clarkson, S.G. (1999). Conserved residues of human XPG protein important for nuclease activity and function in nucleotide excision repair. J. Biol. Chem. 274, 5637–5648. Drane, P., Compe, E., Catez, P., Chymkowitch, P., and Egly, J.M. (2004). Selective regulation of vitamin D receptor-responsive genes by TFIIH. Mol. Cell 16, 187–197. Drapkin, R., Le Roy, G., Cho, H., Akoulitchev, S., and Reinberg, D. (1996). Human cyclin-dependent kinase-activating kinase exists in three distinct complexes. Proc. Natl. Acad. Sci. USA 93, 6488–6493. Dubaele, S., Proietti De Santis, L., Bienstock, R.J., Keriel, A., Stefanini, M., Van Houten, B., and Egly, J.M. (2003). Basal transcription defect discriminates between xeroderma pigmentosum and trichothiodystrophy in XPD patients. Mol. Cell 11, 1635–1646. Dunand-Sauthier, I., Hohl, M., Thorel, F., Jaquier-Gubler, P., Clarkson, S.G., and Scharer, O.D. (2005). The spacer region of XPG mediates recruitment to nucleotide excision repair complexes and determines substrate specificity. J. Biol. Chem. 280, 7030–7037. Emmert, S., Slor, H., Busch, D.B., Batko, S., Albert, R.B., Coleman, D., Khan, S.G., Abu-Libdeh, B., DiGiovanna, J.J., Cunningham, B.B., et al. (2002). Relationship of neurologic degeneration to genotype in three xeroderma pigmentosum group G patients. J. Invest. Dermatol. 118, 972–982. Evans, E., Fellows, J., Coffer, A., and Wood, R.D. (1997a). Open complex formation around a lesion during nucleotide excision repair provides a structure for cleavage by human XPG protein. EMBO J. 16, 625–638. Evans, E., Moggs, J.G., Hwang, J.R., Egly, J.M., and Wood, R.D. (1997b). Mechanism of open complex and dual incision formation by human nucleotide excision repair factors. EMBO J. 16, 6559–6573. Fisher, R.P. (2005). Secrets of a double agent: CDK7 in cell-cycle control and transcription. J. Cell Sci. 118, 5171–5180. Friedberg, E.C., Walker, G.C., Siede, W., Wood, R.D., Schultz, R.A., and Ellenberger, T. (2006). DNA Repair and Mutagenesis, Second Edition (Washington, D.C.: ASM Press). Gervais, V., Lamour, V., Jawhari, A., Frindel, F., Wasielewski, E., Dubaele, S., Egly, J.M., Thierry, J.C., Kieffer, B., and Poterszman, A. (2004). TFIIH contains a PH domain involved in DNA nucleotide excision repair. Nat. Struct. Mol. Biol. 11, 616–622. Giglia-Mari, G., Coin, F., Ranish, J.A., Hoogstraten, D., Theil, A., Wijgers, N., Jaspers, N.G., Raams, A., Argentini, M., van der Spek, P.J., et al. (2004). A new, tenth subunit of TFIIH is responsible for the DNA repair syndrome trichothiodystrophy group A. Nat. Genet. 36, 714–719.

Keriel, A., Stary, A., Sarasin, A., Rochette-Egly, C., and Egly, J.M. (2002). XPD mutations prevent TFIIH-dependent transactivation by nuclear receptors and phosphorylation of RARalpha. Cell 109, 125–135. Klungland, A., Hoss, M., Gunz, D., Constantinou, A., Clarkson, S.G., Doetsch, P.W., Bolton, P.H., Wood, R.D., and Lindahl, T. (1999). Base excision repair of oxidative DNA damage activated by XPG protein. Mol. Cell 3, 33–42. Lee, S.K., Yu, S.L., Prakash, L., and Prakash, S. (2002). Requirement of yeast RAD2, a homolog of human XPG gene, for efficient RNA polymerase II transcription. implications for Cockayne syndrome. Cell 109, 823–834. Lehmann, A.R. (2003). DNA repair-deficient diseases, xeroderma pigmentosum, Cockayne syndrome and trichothiodystrophy. Biochimie 85, 1101–1111. Mu, D., Park, C.H., Matsunaga, T., Hsu, D.S., Reardon, J.T., and Sancar, A. (1995). Reconstitution of human DNA repair excision nuclease in a highly defined system. J. Biol. Chem. 270, 2415–2418. Nakatani, Y., and Ogryzko, V. (2003). Immunoaffinity purification of mammalian protein complexes. Methods Enzymol. 370, 430–444. Nouspikel, T., and Clarkson, S.G. (1994). Mutations that disable the DNA repair gene XPG in a xeroderma pigmentosum group G patient. Hum. Mol. Genet. 3, 963–967. Proietti-De-Santis, L., Drane, P., and Egly, J.M. (2006). Cockayne syndrome B protein regulates the transcriptional program after UV irradiation. EMBO J. 25, 1915–1923. Reardon, J.T., Ge, H., Gibbs, E., Sancar, A., Hurwitz, J., and Pan, Z.Q. (1996). Isolation and characterization of two human transcription factor IIH (TFIIH)-related complexes: ERCC2/CAK and TFIIH. Proc. Natl. Acad. Sci. USA 93, 6482–6487. Rochette-Egly, C., Adam, S., Rossignol, M., Egly, J.M., and Chambon, P. (1997). Stimulation of RAR alpha activation function AF-1 through binding to the general transcription factor TFIIH and phosphorylation by CDK7. Cell 90, 97–107. Sandrock, B., and Egly, J.M. (2001). A yeast four-hybrid system identifies Cdk-activating kinase as a regulator of the XPD helicase, a subunit of transcription factor IIH. J. Biol. Chem. 276, 35328–35333. Sarker, A.H., Tsutakawa, S.E., Kostek, S., Ng, C., Shin, D.S., Peris, M., Campeau, E., Tainer, J.A., Nogales, E., and Cooper, P.K. (2005). Recognition of RNA polymerase II and transcription bubbles by XPG, CSB, and TFIIH: insights for transcription-coupled repair and Cockayne syndrome. Mol. Cell 20, 187–198. Shiomi, N., Mori, M., Kito, S., Harada, Y.N., Tanaka, K., and Shiomi, T. (2005). Severe growth retardation and short life span of double-mutant mice lacking Xpa and exon 15 of Xpg. DNA Repair (Amst.) 4, 351–357. Svejstrup, J.Q., Vichi, P., and Egly, J.M. (1996). The multiple roles of transcription/repair factor TFIIH. Trends Biochem. Sci. 21, 346–350. Thorel, F., Constantinou, A., Dunand-Sauthier, I., Nouspikel, T., Lalle, P., Raams, A., Jaspers, N.G., Vermeulen, W., Shivji, M.K., Wood, R.D., and Clarkson, S.G. (2004). Definition of a short region of XPG necessary for TFIIH interaction and stable recruitment to sites of UV damage. Mol. Cell. Biol. 24, 10670–10680. Tirode, F., Busso, D., Coin, F., and Egly, J.M. (1999). Reconstitution of the transcription factor TFIIH: assignment of functions for the three enzymatic subunits, XPB, XPD, and cdk7. Mol. Cell 3, 87–95. Zurita, M., and Merino, C. (2003). The transcriptional complexity of the TFIIH complex. Trends Genet. 19, 578–584.
