The nucleotide excision repair (NER) pathway operates through two sub-pathways: global genome repair (ggNER) and transcription-coupled repair (TCR) or gene- and strand-specific DNA repair [1, 2, 4]. The ggNER is a repair mechanism which has the ability to repair DNA damage to the overall genome with equivalent efficiency. In contrast, TCR is a kind of heterogeneous DNA repair, where repair to the damaged DNA in the status of transcription activity is superior to the silenced genes and the repair of the transcribed strand is superior to the untranscribed strand. Some DNA repair proteins and transcription factors have been identified to be involved in TCR such as CSA, CSB, XPG, XAB2, RNA polymerase II, and TFIIH [1, 7, 8, 24]. Blockage of RNA polymerase □ at the DNA damage site is believed to create a conducive environment for DNA repair [7, 9]. In this report, we provide evidence to demonstrate that DNA-PKcs, a known critical component in the NHEJ pathway of DNA double-strand breaks, is also necessary for the TCR pathway of UV radiation-induced DNA damage. We firstly demonstrated that deficient of DNA-PKcs or CSB does not affect the efficiency of the global genome nucleotide excision repair (ggNER). In addition, the ggNER repair kinetics of UV radiation-induced CPDs in HeLa cells (Figure 2A & 2B) was shown faster than that in HL60 cells (Figure 2C). A previous report indicated that the level of UVB-induced γH2AX foci (a molecule marker of DNA double-strand break, DSB) in the nonreplicating G1, G2 and M was much higher in HeLa than in HL60 cells. It is well known that the cyclobutane pyrimidine dimers (CPDs) and pyrimidine (6-4) pyrimidone photoproducts (6-4PPs) are the predominant lesions caused by UV radiation. The formation of DSB in the UV-irradiated cells has been explained as resulting from [26, 27]: 1) the attempted replication of DNA (S phase cells) at the sites containing the stalling replication forks with these sites attracted an endonuclease which subsequently cleaved DNA generating DSB; 2) the processing of the nucleotide excision repair (NER) of CPDs, during which the endonucleases, e.g. ERCC1-XPF, incise the adjacent CPDs in inter-strand of DNA to a DSB. Therefore, the formation of DSB, especially in G1, G2 and M phases, can be considered as the intermediate product of NER, and the increased level of the transient γH2AX foci may reflect a more active NER in HeLa cells than in HL60 cells. At present, we could not give a precise mechanistic explanation on this different kinetics of CPDs repair between HeLa and HL60 cells.
In order to determine the TCR of UV-induced cyclobutane pyrimidine dimers (CPDs), we firstly established a novel method of assaying TCR based on the technology of strand-specific polymerase chain reaction. The classical method of the gene- and strand-specific DNA repair or TCR assay has a tedious procedure including: DNA is firstly cut with appropriate restriction enzymes, and then cut at UV-damaged CPDs with T4-endonuclease V; The digested DNA is electrophoresed in the alkaline agarose gels, blotted to Nylon membranes; Northern hybridization is performed with radioactively labeled strand-specific DNA probes. Obviously, the time consuming and radioactive operation largely limit its application. Up to now, there is still no an ideal alternative method for this purpose. This novel method has solved the major technical obstacles of time consuming and radioactive operation. In principal, it partially evolved from the idea of the quantitative PCR detection of transcription-coupled repair based on the chromatin co-immunoprecipitated DNA fragments  and the technique of the tagged RT-PCR for the strand-specific detection of hepatitis C virus (HCV) RNA , as well as the primers design strategy of the suppression subtractive hybridization PCR method . This novel method only has the procedures of DNA digestion with T4 Endo V, first-strand synthesis of the single strand and the strand-specific PCR (Figure 3A) with a set of smart specific primers as shown in table 1. In order to insert an exogenous primer into the synthesized first-strand, chimera primer (UP-TSP1 or UP-NTSP1) was used for the first-strand synthesis of transcribed- or untranscribed-strand. The chimera primer is composed of the upstream adaptor universal primer (UP), which has no homology with the genomic DNA of eukaryotic cells, and the downstream annealing primer (TSP1 or NTSP1). As the TT sequence in genes easily can be damaged into TT-dimers by UV irradiation, we selected a TT-rich sequence of DHFR as a target to detect the activity of the TCR repair pathway. After exposure to UV radiation, the genomic DNA was digested with T4 Endo V to excise the residual TT dimers, resulting in nicks in the DNA strand, which led to the decrease of template DNA molecules. The first-strands of the transcribed and non-transcribed strands were synthesized separately using the specially designed primers described above. Finally, this synthesized first-strand was used as the template to amplify separately the transcribed and untranscribed strands using the primer pair of an adaptor universal primer (UP) and a primer specifically targeting the DHFR gene. This novel method was verified through the detection of the TCR efficiency of the DHFR gene in a CSB-depleted cell line (HeLasiRNA-107). Our data showed that depletion of CSB led to a decreased efficiency of TCR to repair UV-induced DNA damage in HeLa cells. Using our novel TCR assay, we also confirmed that the TCR efficiency of the c-myc gene was depressed when the c-myc gene transcription was silenced by DMSO treatment in HL60 cells. We found that deficiency of DNA-PKcs does not significantly affect the efficiency of the ggNER activity of UV-damaged genomic DNA. However, we observed that the TCR efficiency of the transcribed strand of the DHFR gene in the DNA-PKcs-depleted HeLa-H1 cells was significantly lower than that of control HeLa-NC cells.
The interaction between DNA-PKcs and Cyclin T2 was identified using a co-immunoprecipitation test and a peptide fingerprint assay in this study. Cyclin T2 is a member of the cyclin T family, which is not directly involved in cell cycle control, but involved in the regulation of the gene transcription process through forming the human transcription elongation factor (P-TEFb) with CDK9. The CyclinT1/2 and CDK9 complex (P-TEFb) can phosphorylate the serine 2 of the carboxyl-terminal domain (CTD) of RNA polymerase II and activate this enzyme, and promote transcription elongation [18, 19, 29, 30]. It is noteworthy that Ku80, another component of the DNA-PK complex, can also bind to the RNA polymerase □ elongation complex (not initiation complex) . Although the authors demonstrated that the binding of Ku80 with Pol □ was not dependant on DNA-PKcs , it was evident that there is a close relationship between DNA-PK and the gene transcription complex. Therefore, interaction of DNA-PKcs with Cyclin T2 (or P-TEFb) suggests that DNA-PK is closely associated with the transcriptional "machinery" both physically and physiologically. The binding of these two molecules might have two biological functions: regulation of both gene transcription and the transcription-coupled repair (TCR) mechanism. Since RNA Pol II has already been verified to play an important role in the initialization of TCR [18, 19], we postulate that the involvement of DNA-PKcs in TCR might be mediated through its interaction with Cyclin T2/CDK9/RNA Pol II.