The mechanisms by which tumor-associated macrophages, key components of the glioblastoma (GBM) microenvironment, impair chemotherapy efficacy remain poorly understood. Resistance to temozolomide (TMZ), the standard chemotherapeutic agent for GBM, is associated with poor prognosis due to efficient DNA repair mechanisms. While low expression of the DNA repair enzyme O6-methylguanine-DNA methyltransferase (MGMT) has been linked to improved TMZ response, our previous findings suggest that N-methylpurine-DNA glycosylase (MPG) may also contribute to chemoresistance in GBM. Here, we report for the first time that conditioned medium from pro-inflammatory macrophages (CM-M1) enhances TMZ cytotoxicity by suppressing STAT3 phosphorylation, resulting in decreased MGMT and MPG expression in GBM cells. Proteomic profiling of CM-M1 revealed a unique, cytokine-rich secretome that may promote STAT1 activation, thereby inhibiting pSTAT3 and reducing DNA repair enzymes levels. Clinically, elevated MGMT and MPG protein levels were associated with increased pSTAT3 in our GBM patient cohort, and analysis of the TCGA database further showed that their combined overexpression correlates with significantly reduced progression-free survival. Gene silencing experiments confirmed the contribution of both enzymes to TMZ resistance, with dual knockdown producing a synergistic sensitizing effect. These findings uncover a novel mechanism of macrophage secretome–mediated chemoresistance and support the development of M1-based strategies to improve TMZ efficacy in GBM.
Pan‐cyclin‐dependent‐kinase (CDK) inhibitors are a new class of targeted therapies that can act on multiple CDKs, with dinaciclib being one of the most promising compounds. Although used as a monotherapy, an interesting approach could be to combine it with radiotherapy. Here, we show that dinaciclib increases radiosensitivity in some experimental models of lung and colon cancer (A549 or HCT 116) but not in others (H1299 or HT‐29). Dinaciclib did not alter serine‐protein kinase ATM signalling or cell cycle profiling after ionising‐radiation exposure, which have been described for other CDK inhibitors. Interestingly, in terms of apoptosis, although the combination renders a clear increase, no potentiation of the ionising‐radiation‐induced apoptosis was observed. Mechanistically, inhibition of CDK12 by dinaciclib diminishes BRCA1 expression, which decreases homologous recombination (HR) and probably promotes the nonhomologous end joining repair process (NHEJ), which ultimately promotes the induction of ionising‐radiation‐associated cellular senescence in a TP53‐dependent manner, explaining the lack of effect observed in some experimental models. In conclusion, our report proposes a molecular mechanism, based on the signalling axis CDK12–BRCA1, involved in this newly identified therapeutic effect of dinaciclib, although other players implicated in HR should not be discarded. In addition, our data provide a rationale for more selective and personalised chemo/radiotherapy treatment according to the genetic background of the tumour.