Transcription, replication and homologous recombination are intrinsically connected and it is well established that an increase of transcription is associated with an increase in homologous recombination. Here, we have studied how homologous recombination is affected during transcription inhibition by 5,6-dichloro-1-beta-D-ribofuranosylbenzimidazole (DRB), a compound that prevents activating phosphorylations of the RNA Pol II C-terminal domain. We identify that DRB triggers an increase in homologous recombination within the hprt gene as well as increasing RAD51 foci formation in mammalian cells. Furthermore, we find that DRB-induced transcriptional stress is associated with formation of the nuclear foci of the phosphorylated form of H2AX (γH2AX). We accounted that about 72% of RAD51 foci co-localized with the observed γH2AX foci. Interestingly, we find that XRCC3 mutated, homologous recombination defective cells are hypersensitive to the toxic effect of DRB and fail to form RAD51 foci. In conclusion, we show that DRB-induced transcription inhibition is associated with the formation of a lesion that triggers RAD51-dependent homologous recombination repair, required for survival under transcriptional stress.

Familial breast and ovarian cancers are often defective in homologous recombination (HR) due to mutations in the BRCA1 or BRCA2 genes. Cisplatin chemotherapy or poly(ADP-ribose) polymerase (PARP) inhibitors were tested for these tumors in clinical trials. In a screen for novel drugs that selectively kill BRCA2-defective cells, we identified 6-thioguanine (6TG), which induces DNA double-strand breaks (DSB) that are repaired by HR. Furthermore, we show that 6TG is as efficient as a PARP inhibitor in selectively killing BRCA2-defective tumors in a xenograft model. Spontaneous BRCA1-defective mammary tumors gain resistance to PARP inhibitors through increased P-glycoprotein expression. Here, we show that 6TG efficiently kills such BRCA1-defective PARP inhibitor-resistant tumors. We also show that 6TG could kill cells and tumors that have gained resistance to PARP inhibitors or cisplatin through genetic reversion of the BRCA2 gene. Although HR is reactivated in PARP inhibitor-resistant BRCA2-defective cells, it is not fully restored for the repair of 6TG-induced lesions. This is likely to be due to several recombinogenic lesions being formed after 6TG. We show that BRCA2 is also required for survival from mismatch repair-independent lesions formed by 6TG, which do not include DSBs. This suggests that HR is involved in the repair of 6TG-induced DSBs as well as mismatch repair-independent 6TG-induced DNA lesion. Altogether, our data show that 6TG efficiently kills BRCA2-defective tumors and suggest that 6TG may be effective in the treatment of advanced tumors that have developed resistance to PARP inhibitors or platinum-based chemotherapy. Cancer Res; 70(15); 6268-76. (C) 2010 AACR.

Severe photosensitivity of the skin and predisposition to cancer development are two important features whichcharacterising a genetic syndrome known as Xeroderma pigmentosum. An interesting class of patients has beendescribed, characterised by a proficiency in nucleotide excision repair and a defect in the DNA damage avoidancepathways. This class is termed Xeroderma pigmentosum variant (XP-V) and is known to be caused by a deficiencyin the function of the specific DNA Polymerase η. Cells derived from XP-V patients are sensitivie not only to UVlight but also to crosslinkers such as cisplatin, and Polη overexpression is potentially relevant to development ofcisplatin resistance. In this paper we investigate the importance of the Polη status in the response to treatment withthe chemotherapeutic agent 6-thioguanine (6TG). Our results show that Polη deficient cells are more resistant totreatment with 6TG in comparison to Polη complemented cells. This is in contrast to the typical UV sensitivity ofPolη deficient cells, which is confirmed in the same cells. We also show that 6TG has a growth retardation effect,regardless of the Polη status. There were no DNA double-strand breaks detected in a short period after theexposure to 6TG in physiologically relevant doses, although a DNA damage response was observed in high doses.It was previously demonstrated that 6TG can be used to kill cisplatin resistant cells and these data may indicateone potential explanation.