Logo: to the web site of Uppsala University

Glioblastoma is the most aggressive malignant primary brain tumor and remains associated with poor clinical outcome despite multimodal treatment. Radiotherapy is a central component of glioblastoma management; however, its efficacy is limited by efficient DNA damage repair and intratumoral radioresistance. Because CDK12 and CDK13 regulate transcriptional elongation of DNA damage response genes, including genes involved in homologous recombination, pharmacological CDK12/13 inhibition represents a rational strategy to compromise DNA repair capacity and enhance the response to ionizing radiation.

This thesis investigated the effects of the CDK12/13 inhibitor SR-4835 in U87MG and U343MG glioblastoma cells, both as a single agent and in combination with irradiation. XTT assays were first used to define short-term drug sensitivity and guide the selection of sub-IC₅₀ concentrations for subsequent combination experiments. SR-4835 reduced metabolic activity in both cell lines within the nanomolar range, with greater sensitivity observed in U87MG cells (IC₅₀ = 74.51 nM) than in U343MG cells (IC₅₀ = 203.6–248.1 nM).

The interaction between SR-4835 and irradiation was then evaluated using spheroid growth, DNA damage-associated signalling, apoptosis, and clonogenic survival, with wound healing included as a supplementary functional readout. In spheroid models, pretreatment duration was identified as an important determinant of the combination response. A 1 h SR-4835 pretreatment produced predominantly additive effects under the Loewe and Bliss reference models, with only limited localized interaction. In contrast, 24 h pretreatment generated localized positive interactions at sub-IC₅₀ concentrations in both cell lines, with hotspots centered around 25 nM SR-4835 in U87MG cells and 100 nM SR-4835 in U343MG cells. These concentration ranges were also associated with increased apoptosis-associated cell populations after combined SR-4835 + 4 Gy treatment compared with the corresponding single treatments, with the strongest response observed in U343MG cells.

In U343MG cells, western blot time-course analysis at 1 h and 24 h after irradiation showed that SR-4835 increased γH2AX independently of irradiation, amplified the early γH2AX response after irradiation, and maintained elevated γH2AX at 24 h, when the irradiation-only signal had largely declined. Clonogenic survival analysis further demonstrated enhanced loss of reproductive capacity after combined treatment. The most interpretable radiosensitizing condition was 25 nM SR-4835, where the surviving fraction after 2 Gy decreased from 0.317 to 0.136, corresponding to a sensitization ratio of 2.34. At higher SR-4835 concentrations, interpretation was limited by stronger drug-only suppression. Wound-healing analysis also showed reduced wound closure in U343MG cells, most prominently after 75 nM SR-4835 + 4 Gy, whereas rapid wound closure limited interpretation in U87MG cells.

Overall, these findings provide functional evidence that SR-4835 can enhance radiation -associated effects in glioblastoma cells at sub-IC₅₀ concentrations and that this response is dependent on pretreatment duration. However, the underlying mechanism was not directly established. Key DNA damage response and homologous recombination factors, including BRCA1, RAD51, ATM, ATR, and RNA polymerase II CTD Ser2 phosphorylation, were not assessed, limiting conclusions regarding homologous recombination impairment or direct CDK12/13 target engagement. Further studies are required to validate these findings and determine whether SR-4835-mediated radiation enhancement is driven by impaired homologous recombination, broader disruption of DNA damage response signalling, or additional mechanisms.