BH3-only protein expression determines hepatocellular carcinoma response to sorafenib-based treatment.

Hepatocellular carcinoma (HCC) represents a global health challenge with limited therapeutic options. Anti-angiogenic immune checkpoint inhibitor-based combination therapy has been introduced for progressed HCC, but improves survival only in a subset of HCC patients. Tyrosine-kinase inhibitors (TKI) such as sorafenib represent an alternative treatment option but have only modest efficacy. Using different HCC cell lines and HCC tissues from various patients reflecting HCC heterogeneity, we investigated whether the sorafenib response could be enhanced by combination with pro-apoptotic agents, such as TNF-related apoptosis-inducing ligand (TRAIL) or the BH3-mimetic ABT-737, which target the death receptor and mitochondrial pathway of apoptosis, respectively. We found that both agents could enhance sorafenib-induced cell death which was, however, dependent on specific BH3-only proteins. TRAIL augmented sorafenib-induced cell death only in NOXA-expressing HCC cells, whereas ABT-737 enhanced the sorafenib response also in NOXA-deficient cells. ABT-737, however, failed to augment sorafenib cytotoxicity in the absence of BIM, even when NOXA was strongly expressed. In the presence of NOXA, BIM-deficient HCC cells could be in turn strongly sensitized for cell death induction by the combination of sorafenib with TRAIL. Accordingly, HCC tissues sensitive to apoptosis induction by sorafenib and TRAIL revealed enhanced NOXA expression compared to HCC tissues resistant to this treatment combination. Thus, our results suggest that BH3-only protein expression determines the treatment response of HCC to different sorafenib-based drug combinations. Individual profiling of BH3-only protein expression might therefore assist patient stratification to certain TKI-based HCC therapies.

Direct impact of cisplatin on mitochondria induces ROS production that dictates cell fate of ovarian cancer cells.

Patients with high-grade serous ovarian cancer (HGSC) frequently receive platinum-based chemotherapeutics, such as cisplatin. Cisplatin binds to DNA and induces DNA-damage culminating in mitochondria-mediated apoptosis. Interestingly, mitochondrial DNA is critically affected by cisplatin but its relevance in cell death induction is scarcely investigated. We find that cisplatin sensitive HGSC cell lines contain higher mitochondrial content and higher levels of mitochondrial ROS (mtROS) than cells resistant to cisplatin induced cell death. In clonal sub-lines from OVCAR-3 mitochondrial content and basal oxygen consumption rate correlate with sensitivity to cisplatin induced apoptosis. Mitochondria are in two ways pivotal for cisplatin sensitivity because not only knock-down of BAX and BAK but also the ROS scavenger glutathione diminish cisplatin induced apoptosis. Mitochondrial ROS correlates with mitochondrial content and reduction of mitochondrial biogenesis by knock-down of transcription factors PGC1α or TFAM attenuates both mtROS induction and cisplatin induced apoptosis. Increasing mitochondrial ROS by inhibition or knock-down of the ROS-protective uncoupling protein UCP2 enhances cisplatin induced apoptosis. Similarly, enhancing ROS by high-dose ascorbic acid or H2O2 augments cisplatin induced apoptosis. In summary, mitochondrial content and the resulting mitochondrial capacity to produce ROS critically determine HGSC cell sensitivity to cisplatin induced apoptosis. In line with this observation, data from the human protein atlas (www.proteinatlas.org) indicates that high expression of mitochondrial marker proteins (TFAM and TIMM23) is a favorable prognostic factor in ovarian cancer patients. Thus, we propose mitochondrial content as a biomarker for the response to platinum-based therapies. Functionally, this might be exploited by increasing mitochondrial content or mitochondrial ROS production to enhance sensitivity to cisplatin based anti-cancer therapies.