The C-terminal Region and SUMOylation of Cockayne Syndrome Group B Protein Play Critical Roles in Transcription-coupled Nucleotide Excision Repair*

Cockayne syndrome (CS) is a recessive disorder that results in deficiencies in transcription-coupled nucleotide excision repair (TC-NER), a subpathway of nucleotide excision repair, and cells from CS patients exhibit hypersensitivity to UV light. CS group B protein (CSB), which is the gene product of one of the genes responsible for CS, belongs to the SWI2/SNF2 DNA-dependent ATPase family and has an ATPase domain and an ubiquitin-binding domain (UBD) in the central region and the C-terminal region, respectively. The C-terminal region containing the UBD is essential for the functions of CSB. In this study, we generated several CSB deletion mutants and analyzed the functions of the C-terminal region of CSB in TC-NER. Not only the UBD but also the C-terminal 30-amino acid residues were required for UV light resistance and TC-NER. This region was needed for the interaction of CSB with RNA polymerase II, the translocation of CS group A protein to the nuclear matrix, and the association of CSB with chromatin after UV irradiation. CSB was modified by small ubiquitin-like modifier 2/3 in a UV light-dependent manner. This modification was abolished in a CSB mutant lacking the C-terminal 30 amino acid residues. However, the substitution of lysine residues in this region with arginine did not affect SUMOylation or TC-NER. By contrast, substitution of a lysine residue in the N-terminal region with arginine decreased SUMOylation and resulted in cells with defects in TC-NER. These results indicate that both the most C-terminal region and SUMOylation are important for the functions of CSB in TC-NER.

Posttranslational modifications play important roles in the functions of CSB (26). It has been reported that CSB is ubiquitinated and degraded in a UV light-and CSA-dependent manner (27). However, another study has reported that CSB is ubiquitinated by BRCA1 even in the absence of CSA (28). Therefore, it is unclear which ubiquitin ligase is responsible for the ubiquitination of CSB. Not only ubiquitination but also deubiquitination of CSB by UVSSA-USP7 might be essential for the progression of TC-NER (22,23). There are at least eight candidate phosphorylation sites in CSB (26). CSB is phosphorylated by c-Abl kinase, and this modification is relevant to the subcellular localization of CSB (29).
Here we showed that the most C-terminal region of CSB affects its function in TC-NER and the UV light sensitivity of cells. In addition, we revealed that CSB is modified by small ubiquitin-like modifier (SUMO)-2/3 in a UV light-dependent manner and that the most C-terminal region is related to this modification. Moreover, the amino acid substitution of Lys-205 with Arg in CSB suppressed SUMOylation and resulted in a lack of TC-NER in cells. These results indicate the importance of the C-terminal region and SUMOylation of CSB in TC-NER.

Experimental Procedures
Expression Constructs and Stable Cell Lines-To generate epitope-tagged CSB expression constructs, WT and mutant CSB cDNA fragments were amplified by PCR and cut with XhoI at the 5Ј end and with XbaI at the 3Ј end. The fragments were cloned in-frame and downstream of the sequence encoding the FLAG epitope, followed by the HA epitope in pcDNA3.1 (Invitrogen). The CSB 2K3 R , CSB 3K3 R , and CSB K205R cDNAs were generated using the QuikChange II-E site-directed mutagenesis kit (Agilent Technologies) according to the instructions of the manufacturer. DNA sequencing of the plasmids ruled out the presence of PCR-derived mistakes. An expression construct for C-terminal FLAG-and HA-tagged WT CSB was also generated. CSB-FLAG-HA cDNA was cut out of pCAGGS-CSB-FLAG-HA (30) using XhoI and NotI and inserted into pcDNA3.1.
To isolate stable transfectants, CS1ANSV cells were transfected with the CSB expression constructs using Effectene transfection regent (Qiagen) according to the procedure of the manufacturer. Stable transfectants were selected in the presence of 500 g/ml G418 (Nacalai Tesque).
UV Irradiation-Cells were washed once with PBS and irradiated with the indicated dose of UV-C light. Culture medium was added immediately after UV irradiation. For the UV light survival assay, 1.0 ϫ 10 3 cells were seeded into 10-cm dishes and incubated overnight. After UV irradiation, cells were cultured for 10 days, fixed with 3% formaldehyde prepared in PBS, and stained with 2.5 mM crystal violet solution. Colonies were counted using a stereo microscope.
Recovery of RNA Synthesis after UV Irradiation-Cells were inoculated into 35-mm dishes at a density of 1.0 ϫ 10 5 cells/ dish and incubated overnight. Cells were washed once with PBS, irradiated with 10 J/m 2 UV-C light or not irradiated, and incubated in 1 ml of culture medium. To measure RNA synthesis, [ 3 H]uridine (PerkinElmer Life Sciences) was added to each dish to a concentration of 370 kBq/ml at each time point. After incubation for 1 h, labeling was terminated by adding NaN 3 to a final concentration of 200 g/ml. Cells were solubilized with 0.8% SDS. Then an equal volume of 10% TCA containing 0.1 M sodium diphosphate was added to the lysates and incubated on ice. Acid-insoluble material was collected on glass microfiber filters (Whatman GF/C). Radioactivity was measured in Insta-Fluor Plus mixture (PerkinElmer Life Sciences) with a liquid scintillation counter (PerkinElmer Life Sciences Tri-Carb 2810TR). The ratio of the radioactivity of UV-irradiated cells to that of non-irradiated cells was considered to reflect the recovery of RNA synthesis after UV irradiation.
Interaction of CSB with Pol II-After irradiation with 20 J/m 2 UV light, cells were incubated for 30 min and harvested. In total, 1.0 ϫ 10 6 cells were resuspended in micrococcal nuclease buffer (20 mM Tris-HCl (pH 7.5), 100 mM KCl, 2 mM MgCl 2 , 1 mM CaCl 2 , 300 mM sucrose, 0.1% Triton X-100, 1 mM DTT, and protease inhibitors) and incubated on ice for 10 min. Lysates were fractionated by centrifugation (5000 rpm, 5 min, 4°C). Pellets were resuspended in micrococcal nuclease buffer containing 15 units/ml micrococcal nuclease and incubated at 25°C for 15 min. The reaction was terminated by adding EDTA to a final concentration of 5 mM and centrifugation at 5000 rpm for 5 min at 4°C. The supernatants were recovered. The pellets were washed with micrococcal nuclease buffer, and the supernatants were combined with the aforementioned supernatants and used as the solubilized chromatin fractions. The solubilized chromatin fractions were incubated with 20 l of anti-FLAG M2-agarose beads (Sigma) overnight at 4°C. The beads were washed five times with NETN buffer (50 mM Tris-HCl (pH 7.8), 150 mM NaCl, 1% Nonidet P-40, 1 mM EDTA, and 0.5 mM PMSF) and then boiled in an equal volume of 2ϫ SDS-PAGE sample buffer (125 mM Tris-HCl (pH 6.8), 4% SDS, 20% glycerol, 0.02% bromphenol blue, and 5% mercaptoethanol). SDS-PAGE was performed, and separated proteins were transferred to a PVDF membrane. Immunoblot detection was performed with anti-Pol II and anti-CSB antibodies using Pierce Western blotting Substrate Plus (Thermo Scientific).
Subcellular Fractionation-After irradiation with 20 J/m 2 UV light, cells were harvested at each time point. In total, 1.0 ϫ 10 6 (Fig. 2) or 4.0 ϫ 10 6 ( Fig. 4) cells were resuspended in CSK-Triton buffer (10 mM PIPES (pH 6.8), 100 mM NaCl, 300 mM sucrose, 3 mM MgCl 2 , 0.5% Triton X-100, 1 mM EGTA, 1 mM DTT, and protease inhibitors) with gentle vortexing and incubated on ice for 10 min. The lysates were fractionated by cen-trifugation (6000 rpm, 3 min, 4°C). The supernatants were mixed with an equal volume of 2ϫ SDS-PAGE sample buffer, boiled for 5 min, and used as the soluble fractions. The pellets were washed once with CSK-Triton buffer, boiled in 1ϫ SDS-PAGE sample buffer, and used as the insoluble fractions.
Toranslocation of CSA to the Nuclear Matrix after UV Irradiation-UV light-induced translocation of CSA in a cellfree system was examined as described previously (21). CS1ANSV cells expressing CSB were irradiated with 20 J/m 2 UV light, incubated for 1 h, and then treated with CSK-Triton buffer to prepare the insoluble fractions (CSK-ppt fraction). The soluble fractions (CSK-sup fraction) were prepared from CS3BESV cells stably expressing CSA-FLAG-HA by treatment with CSK-Triton buffer. The CSK-sup fraction containing CSA-FLAG-HA was incubated with the CSK-ppt fraction on ice for 1 h. After centrifugation (6000 rpm, 3 min, 4°C), the pellet was washed twice with CSK-Triton buffer, incubated with 0.1 units/l DNase I (Takara) prepared in CSK-Triton buffer at 30°C for 10 min, and centrifuged (6000 rpm, 3 min, 4°C). After washing three times with CSK-Triton buffer, 1ϫ SDS-PAGE sample buffer was added to the pellet and boiled for 5 min. CSA retained in the DNase I-insoluble fractions was detected by immunoblotting with an anti-HA antibody.
Detection of SUMOylated Proteins-SUMOylated proteins were detected as described previously (31) with some modifications. After irradiation with 20 J/m 2 UV light, cells were incubated for 1 h and harvested. In total, 2.0 ϫ 10 6 cells were solubilized by adding SDS lysis buffer (62.5 mM Tris-HCl (pH 6.8), 10% glycerol, 2% SDS, and protease inhibitors) and boiling for 10 min. The lysates were passed through a 25-gauge needle to shear DNA and diluted 10-fold with radioimmune precipitation assay buffer (50 mM Tris-HCl (pH 7.4), 1% Nonidet P-40, 0.25% sodium deoxycholate, 150 mM NaCl, and 1 mM EDTA) supplemented with 20 mM N-ethylmaleimide, 20 mM PR-619 (Life Sensors), 1 mM sodium fluoride, 1 mM sodium orthovanadate (V), and 2 mM ␤-glycerophosphate immediately before use. The diluted lysates were incubated with 20 l of protein G-Sepharose 4 Fast Flow (GE Healthcare) and 2 g of an anti-CSB antibody overnight at 4°C. The beads were washed five times with radioimmune precipitation assay buffer and then boiled in an equal volume of 2ϫ SDS-PAGE sample buffer. SDS-PAGE was performed, and separated proteins were transferred to a PVDF membrane. Immunoblot detection was performed with anti-ubiquitin, anti-SUMO 1, 2, and 3, and anti-CSB antibodies using Pierce Western blotting Substrate Plus.
RNA Interference-Ubc9 silencing was performed by transfection of siRNA (Santa Cruz Biotechnology, catalog no. sc-36773) using Lipofectamine RNAiMAX transfection reagent (Invitrogen) according to the procedure of the manufacturer. Reverse transfection was performed first, and forward transfection was performed 24 h later. Cells were irradiated with UV light 24 -48 h after forward transfection, incubated for the indicated amount of time, and harvested.

Results
UV Sensitivity and Recovery of RNA Synthesis after UV Irradiation of Cells Expressing Mutant CSB Proteins with Deletions of the C-terminal Region-To analyze the functions of the C-terminal region of CSB in TC-NER, we generated several C-terminally truncated CSB mutants, including CSB 1-1220 (Fig.  1A). CSB 1-1400 had a deletion from just the N terminus of the UBD to the C terminus. CSB 1-1445 and CSB 1-1463 contained the UBD, and CSB 1-1463 lacked only the C-terminal 30 amino acid residues. There is no report of the functional domains and amino acid residues in the region corresponding to amino acid residues 1221-1400, 1446 -1463, and 1464 -1493 of CSB, except for target sites for phosphorylation (Ser-1348 (32) and Ser-1461 (33)) and ubiquitination (Lys-1457 (34)). CSB ⌬UBD lacked 49 amino acids residues (1385-1433) containing the UBD. All CSB mutants have two nuclear localization signals (amino acid residues 466 -481 and 1038 -1055 (10)). In addition to the expression constructs for the CSB mutants, an expression construct for WT CSB (CSB WT ) was also generated. All CSB proteins had FLAG and HA epitope tags at their N termini. These epitope-tagged CSBs were stably expressed in CS1ANSV cells. CS1ANSV cells are derived from a CS-B patient, theoretically only express short CSB (35) and do not have functional CSB. Immunoblot analysis using anti-CSB and anti-HA antibodies was performed to examine the expression of CSB in each transfectant (Fig. 1B). Expression levels of CSB in the transfectants were similar to those in normal (WI38VA13) cells. Transfectants expressing each CSB protein were designated CSB WT cells, CSB 1-1220 cells, CSB 1-1400 cells, CSB 1-1445 cells, CSB 1-1463 cells, and CSB ⌬UBD cells for convenience.
First, the colony-forming ability of UV light-irradiated transfectants was measured (Fig. 1C). CSB WT cells exhibited almost the same level of UV light sensitivity as normal (WI38VA13) cells. CSB 1-1220 cells showed UV light hypersensitivity comparable with that of CS1ANSV cells, as described previously (18). Cells expressing other deletion mutants were more sensitive to UV irradiation than CSB WT cells but were less sensitive to UV irradiation than CS1ANSV cells regardless of whether the CSB mutant protein contained the UBD. Therefore, all cell lines expressing CSB deletion mutants exhibited hypersensitivity to UV irradiation.
Next, the recovery of RNA synthesis after UV irradiation, which is an index of TC-NER, was measured. RNA synthesis was decreased 2 h after UV irradiation, but, in WI38VA13 and CSB WT cells, recovered to the level of non-irradiated cells 24 h after UV irradiation (Fig. 1D). No such recovery was observed in CS1ANSV cells. All cell lines expressing CSB deletion mutants had little ability to recover RNA synthesis after UV irradiation. CSB 1-1220 cells showed the most severe reduction in RNA synthesis. On the other hand, the reduction in RNA synthesis in CSB ⌬UBD cells was slightly less than that in cells expressing other deletion mutants but was almost the same as that in CS1ANSV cells. Regardless, all cell lines expressing CSB deletion mutants were deficient in TC-NER. These results indicate that not only the UBD but also the C-terminal region downstream of the UBD (1446 -1493) affected TC-NER. Even deletion of only the C-terminal 30 amino acid residues had the same effect. Therefore, we compared the C-terminal amino acid sequences of CSB homologs (Fig. 1E). Approximately 75% of amino acid residues in the C-terminal region (1446 -1493) are identical among six mammals (Fig. 1E, top). At non-conserved amino acid residues, most residues are substituted for similar residues. Amino acid residues in this region of other animals are relatively conserved (Fig. 1E, bottom; ϳ40% identical). In the frog (Xenopus tropicalis) CSB homolog and Rad26 (Saccharomyces cerevisiae CSB homolog), there are no amino acid residues corresponding to this region.
CSB is also needed for the translocation of CSA to the nuclear matrix after UV irradiation (20,21). This translocation is relevant to TC-NER. Therefore, we next investigated the ability of each CSB mutant to translocate CSA to the nuclear matrix using a cell-free system (21). Mutant CSB cells were irradiated with UV light or not irradiated and incubated for 1 h. CSK-ppt fractions were prepared from the cells. CSK-sup fractions were prepared from CS-A (CS3BESV) cells expressing FLAG and HA epitope-tagged CSA. This fraction contained FLAG-HAtagged CSA. Each CSK-ppt fraction was incubated with the CSK-sup fraction. After DNase I treatment, the CSK-ppt fractions were analyzed by Western blotting with an anti-HA antibody (Fig. 2B). When CSA translocation occurs, HA-tagged CSA is detected in a UV light-dependent manner. As reported previously, CSA was detected when the CSK-ppt fraction was prepared from CSB WT cells after UV irradiation (Fig. 2B, lanes  4 and 5) but was not detected when CS1ANSV cells were used (Fig. 2B, lanes 2 and 3). When the CSK-ppt fractions were prepared from mutant CSB cells, CSA was not detected (Fig. 2B,  lanes 6 -15). These results indicate that the UBD and C-terminal region of CSB are required for the translocation of CSA to the nuclear matrix.
Association of the Mutant CSB Proteins with Chromatin after UV Irradiation-TC-NER factors are thought to associate with chromatin for progression of TC-NER. Therefore, we examined the association of mutant CSB proteins with chromatin. Insoluble fractions containing chromatin were prepared from mutant CSB cells after UV irradiation and analyzed by Western blotting using an anti-CSB antibody (Fig. 2C). More CSB WT was detected in the insoluble fractions after UV irradiation. The intensities of the bands were quantified and plotted (Fig. 2D).
The amount of CSB WT in the insoluble fraction increased until 2 h after UV irradiation and decreased to the level of non-irradiated cells by 12 h after UV irradiation. The level of CSB  in the insoluble fraction also increased in a UV light-dependent manner. The levels of CSB 1-1220 and CSB 1-1463 in the insoluble fraction increased slightly until 2 h after UV irradiation and then decreased. The level of CSB ⌬UBD in the insoluble fraction did not increase after UV irradiation. The amount of CSB  in the insoluble fraction was decreased by UV irradiation. These data show that the association of CSB with chromatin after UV irradiation was affected by deletion of the UBD or C-terminal region. Fig. 2C we show that some high molecular weight bands were observed after UV irradiation in the CSB WT panel (lanes 2-4, asterisk). In the panel of CSB ⌬UBD , the shifted bands were extremely faint but visible (Fig. 2C, lanes 27-29). In the panels of other CSB mutants, only a single band of CSB (i.e. no shifted bands) was detected. It has been reported that CSB is modified by ubiquitin and phosphate and that these posttranslational modifications affect the function of CSB in TC-NER (26 -29). In addition, the SUMOplot analysis program (Abgent) showed that Lys-1489 in the C-terminal region has a high potential to be modified by SUMO. The shifted bands were presumed to reflect modifications of CSB, and we hypothesized that there is a relationship between the C-terminal deletion of CSB and specific posttranslational modifications. Therefore, we first examined which modifications of CSB were detected after UV irradiation. Cells were treated with SDS lysis buffer, which disrupts protein interactions, and then CSB was immunoprecipitated from the lysate. Modifications by ubiquitin and SUMO were examined by Western blotting using specific antibodies (Fig. 3). We could not detect ubiquitinated CSB regardless of UV irradiation (Fig. 3, lanes 1-8). No bands were detected using an anti-SUMO-1 antibody (Fig. 3, lanes 9 -12), whereas, remarkably, three bands higher than 250 kDa and some even higher faint bands were detected with anti-SUMO-2 (Fig. 3, lanes 13-16) and anti-SUMO-3 (Fig. 3, lanes 17-20) antibodies specifically after UV irradiation. The amino acid sequences of SUMO-2 and SUMO-3 are 97% identical, and differences in their function have not been found. Therefore, they are usually denoted as SUMO-2/3. Almost the same band patterns were detected with anti-SUMO-2 and anti-SUMO-3 antibodies in Fig. 3 (lanes 16 and 20). Therefore, we subsequently only used an anti-SUMO-3 antibody to detect the shifted bands.

UV-dependent Posttranslational Modification of CSB by SUMO-2/3-In
Analysis of CSB SUMOylation-Ubc9 is the only SUMOconjugating enzyme (E2) in eukaryotes. Ubc9 was knocked down by transfection of siRNA (Fig. 4A). The amount of Ubc9 was extremely low after the transfection but that of CSB was not (Fig. 4A, lanes 1 and 3). Then CSB was immunoprecipitated from the cell lysate. The bands detected with an anti-SUMO-3 antibody became faint (Fig. 4A, lane 8). These data indicate that the bands are due to SUMO modification.
Next, soluble and insoluble fractions were prepared with CSK-Triton buffer, and SUMOylation of CSB was analyzed. Bands were detected in the insoluble fraction but not in the soluble fraction (Fig. 4C). The insoluble fraction contained chromatin, suggesting that SUMOylated CSB is associated with chromatin.
The ATPase activity of CSB is essential for its function in TC-NER. One amino acid substitution in the ATPase domain (K538R) disrupts the ATPase activity and causes cells to become hypersensitive to UV irradiation (36). In cells expressing CSB K538R , this dysfunctional CSB was also SUMOylated after UV irradiation (Fig. 4E). We next examined whether the CSB deletion mutants described above are SUMOylated (Fig.  5). CSB ⌬UBD was SUMOylated after UV irradiation, similar to CSB WT (Fig. 5, lanes 1, 2, 11, and 12), but no other CSB mutants were SUMOylated.
Taken together, UV light-dependent SUMO modification of CSB occurred independently of UVSSA, CSA, and the ATPase activity of CSB. In addition, the C-terminal region of CSB was involved in the SUMOylation of CSB but the UBD was not.
Effects of Substitution of Lysine to Arginine in the C-terminal Region of CSB-As shown in Fig. 5, the C-terminal truncated CSB mutants (CSB 1-1220 , CSB 1-1400 , CSB 1-1445 , and CSB  were not SUMOylated after UV irradiation. CSB  , which has the smallest deletion, lacks only 30 amino acid residues. Therefore, it was assumed that lysine residues of a SUMO acceptor site are present among the most C-terminal 30 amino acid residues of CSB. In this region, there are two lysine residues, Lys-1487 and Lys-1489 (Fig. 6A). To investigate whether these lysine residues were responsible for SUMOylation of CSB, we generated two CSB mutants with FLAG-HA epitope tags at their N termini: CSB K1487R, K1489R (2K3 R) and CSB K1457R, K1487R, K1489R (3K3 R). Lys-1457 is located outside of the 30 amino acid residues but just precedes this region. Transition of a SUMOmodified site to a nearby lysine residue can occur when an authentic lysine for SUMOylation is substituted. Therefore, Lys-1457 was substituted with arginine in conjunction with the substitutions at Lys-1487 and Lys-1489. Lys-1457 is also reportedly modified by ubiquitin independently of UV irradiation (34). In immunoblot analysis, the expression levels of both mutant CSB proteins in the transfectants were similar to those in normal (WI38VA13) and CSA WT cells (Fig. 6B). SUMOylation of the substitution mutants was examined as described above. Substitutions of Lys-1457, Lys-1487, and Lys-1489 with arginine did not abolish SUMOylation of CSB after UV irradiation (Fig. 6C). We noted that the band pattern of CSB 3K3 R in non-irradiated cells was different from that of the other CSB mutants. A single band was detected in CSB 3K3 R cells (Fig. 6C,  lane 3), but no bands were detected in the other mutant cells.
In addition, the colony-forming ability and recovery of RNA synthesis after UV irradiation of the substitution mutants were  examined. The substitutions did not affect UV light sensitivity (Fig. 6D) or RNA synthesis recovery following UV irradiation (Fig. 6E). These results indicate that the three lysine residues in the C-terminal region are not SUMO acceptor sites after UV irradiation and that the substitution of these residues with arginine has no effect on the function of CSB in TC-NER. Substitution of Lys-205 with Arginine Represses SUMOylation of CSB-There is no lysine residue for SUMOylation in the C-terminal region of CSB. The SUMOplot analysis program (Abgent) indicated that Lys-205 has a high probability of being SUMOylated. In addition, according to PhosphoSitePlus (Cell Signaling Technology), many amino acid residues near Lys-205 are modified by posttranslational modifications. Therefore, CSB with Lys-205 substituted (CSB K205R ) and tagged with the FLAG-HA epitope at its N terminus was generated and stably expressed in CS1ANSV cells (Fig. 7A). SUMOylation of CSB K205R was not detected with an anti-SUMO3 antibody even after UV irradiation (Fig. 7B, lanes 3 and 4). When cells over-expressing CSB K205R were used, faint bands were detected, dependent on UV light, but the unique shifted bands were not (Fig. 7B, lanes 5 and 6).
CSB K205R cells were more sensitive to UV irradiation than CSB WT cells but less sensitive to UV irradiation than CS1ANSV cells (Fig. 7C), similar to CSB 1-1400 , CSB 1-1445 , CSB 1-1463 , and CSB ⌬UBD cells. Moreover, the recovery of RNA synthesis after UV irradiation was not detected in CSB K205R cells, similar to all cell lines expressing CSB deletion mutants (Fig. 7D).
To investigate the effects of SUMOylation on TC-NER, Ubc9 knockdown cells were used (Fig. 4B). RNA synthesis in knockdown cells not irradiated with UV was almost equal to that in control cells. Next, the recovery of RNA synthesis after UV irradiation was examined. RNA synthesis was decreased in both cell types 2 h after UV irradiation. RNA synthesis was recovered in control cells 24 h after UV irradiation, whereas RNA synthesis in Ubc9 knockdown cells was only 70% of that in non-irradiated cells. These results indicate that TC-NER is hampered by Ubc9 knockdown. There are no reports showing that any TC-NER-specific factors are SUMOylated in eukaryotes. Regardless of UV irradiation, neither CSA nor UVSSA was modified by any SUMO isoforms (data not shown). These data suggest that SUMOylation of CSB is relevant to TC-NER.

Discussion
When UV light-induced lesions occur at the transcribed strands of actively transcribed genes, Pol II is arrested at the damaged sites (37,38), and TC-NER is initiated. CSB interacts with Pol II, and this interaction is increased after UV irradiation (15). CSB is a key factor to recruit repair factors to stalled Pol II at damaged sites (19,22,39). In this study, we characterized CSB mutants with deletions of various C-terminal regions. All cells expressing deletion mutants showed hypersensitivity to UV light (Fig. 1C) and did not recover RNA synthesis after UV irradiation (Fig. 1D), indicating that both the UBD and C-terminal region are necessary for TC-NER. Although the recovery of RNA synthesis in all mutant cells was as defective as that in CS1ANSV cells, the mutant cells, except CSB 1-1220 cells, were less sensitive to UV irradiation than CS1ANSV cells. These results suggest that the region corresponding to amino acid residues 1221-1400 has functions for cell viability after UV irradiation; for example, suppression of apoptosis. It is noteworthy that Ser-1348 is phosphorylated (32). It is likely that phosphorylation at this site is important for the functions of CSB.
The interaction between CSB with deletion of either the UBD or C-terminal region and Pol II was increased slightly after UV irradiation in contrast to the large increase in the interaction between WT CSB and Pol II ( Fig. 2A), even when CSB was associated with chromatin ( Fig. 2, C and D). These results indicate that both regions of CSB are required for its complete interaction with Pol II after UV irradiation. The interaction between Pol II and CSB is thought to be critical for TC-NER. If the interaction is incomplete after UV irradiation, then it is expected that the TC-NER reaction is not initiated.
Translocation of CSA to the nuclear matrix after UV irradiation is related to TC-NER and is dependent on CSB (20,21). Because the translocation did not occur with the deletion mutants (Fig. 2B), both the UBD and C-terminal regions are necessary for the translocation. It is considered that complete