Back

Mitochondrial Oxygen Consumption Drives Lung Tumor Hypoxia and Resistance to Therapy via Copy Number Alteration in Mitochondrial Electron Transport Subunit NDUFB5

Benej, M.; Benejova, K.; Fergatova, A.; Lisi, R.; Travis, K.; Kreamer, M.; Webb, A.; Dravillas, C.; Hoyd, R.; Bayrali-Ulker, E.; Sai Thoutham, A.; Spakowicz, D.; Denko, N. C.

2026-07-28 cancer biology
10.64898/2026.07.27.741033 bioRxiv
Show abstract

Decades of research have shown that tumor hypoxia is associated with resistance to anti-cancer treatments. Analysis of TCGA gene expression profiles indicates that NSCLC is among the most hypoxic of cancers despite the high levels of oxygen in the surrounding lung tissue. Several groups have shown that extrinsic factors such as poorly formed tumor vascular contributes to tumor hypoxia. Here, we have investigated the possibility that genetic abnormalities within the tumor also contribute to the development of hypoxia. Our analysis of NSCLC patient datasets in the Cancer Genome Atlas (TCGA) PanCancer and ORIEN datasets revealed a strong correlation between tumor hypoxia and amplification of chromosome 3q which is found in up to 40% of NSCLC. Several oncogenic driver genes have been identified in 3q, and we identified a passenger gene encoding mitochondrial complex I subunit NDUFB5 at 3q26.33. To provide experimental evidence that NDUFB5 amplification can drive tumor hypoxia, we have used CRISPR activation technology to generate murine cells overexpressing the endogenous NDUFB5 gene. We found that cells overexpressing NDUFB5 have elevated rates of oxygen consumption, and tumors grown from these cells have increased amounts of hypoxia with associated treatment resistance. Here, we investigate the impact of manipulating NDUFB5 gene expression on mitochondrial complex I activity and experimentally validate the clinical observations that NDUFB5 overexpression leads to increased levels of intratumoral hypoxia and increased resistance to radiation therapy and immunotherapy.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.