Potential role of mitochondrial reactive oxygen species in the manifestation of 5-fluorouracil-induced microvascular dysfunction

INTRODUCTION: 5-fluorouracil (5FU) is a common chemotherapy used to treat gastrointestinal cancers. While effective, its use is associated with severe angina in up to 20% of treated patients. This can result in removal of 5FU from the treatment regimen and negatively impact cancer treatment outcomes. As such, understanding how 5FU causes cardiotoxicity is critical for ensuring patients can receive 5FU without increased risk to cardiovascular wellbeing. Microvascular dysfunction is known to contribute to angina, and we have previously demonstrated that cancer patients undergoing 5FU treatment have impaired microvascular function in vivo compared to matched controls. However, the mechanisms for this impairment remain unknown. It is well established that mitochondrial reactive oxygen species (ROS) can lead to endothelial dysfunction, and that this dysfunction can be mitigated by use of mitochondrial targeted antioxidants. Given 5FU has been associated with mitochondrial damage and elevated ROS production, the goal of this study was to determine whether a mitochondrial targeted antioxidant could prevent 5FU-induced microvascular endothelial dysfunction in human arterioles. Additionally, we sought to determine whether 5FU reduces gene expression of the mitochondrial redox regulator superoxide dismutase 2 (SOD2). We hypothesized that 5FU would reduce SOD2 gene expression in endothelial cells (HUVECs) and that pretreating human arterioles with the mitochondrial antioxidant mitoTEMPO would prevent impairments in microvascular endothelial function. METHODS: Human arterioles (max internal diameter 80-200 µM) from healthy donors (≤ 1 CV risk factor) were isolated from surgically discarded adipose tissue and cannulated on glass micropipettes in a temperature-controlled chamber filled with Krebs buffer. Arterioles were treated with 5FU (n=6, 500 µM) or volume matched control (CON, n=8) for 30 minutes. A subset of arterioles was treated with mitoTEMPO (MT+5FU, n=4, 1 µM) 30 min prior to 5FU exposure. Endothelium dependent (flow-mediated dilation, FMD, 5-100 cmH2O) and independent (100 µM Papaverine) dilation was assessed using pressure myography. Data are presented as peak FMD as % max diameter ± SD and compared using 2-way repeated measures mixed effects analysis and Tukey’s multiple comparison test. qPCR to assess SOD2 gene expression was completed using cDNA from HUVECs ± 5FU (500µM) for 30 min. Experiments were completed in triplicate and compared using the ΔΔCT method with an independent samples t-test. Statistical significance was set at 0.05 for all comparisons. RESULTS: FMD at 100 cmH2O was significantly reduced in 5FU compared to CON arterioles (5FU: 66±15%, CON: 90±5%, p =0.006). However, in 5FU+MT, 5FU-induced endothelial dysfunction was prevented (92±6%; p = 0.01 vs 5FU; p = 0.76 vs CON). There were no group differences in papaverine dilation (all p > 0.05). SOD2 gene expression was 55% lower in HUVECs treated with 5FU vs CON, but did not reach statistical significance with current number of biological replicates (p = 0.16). CONCLUSION: Our preliminary findings indicate that 5FU-induced microvascular endothelial dysfunction is prevented by a mitochondrial antioxidant, suggesting a potential role of mitochondrial ROS in the etiology of 5FU-cardiotoxicity. Future studies will determine whether 5FU increases mitochondrial ROS and reduces nitric oxide bioavailability, with the long-term goal of developing modalities to alleviate 5FU cardiotoxicity in the clinical setting. FUNDING: This work is supported by NIH T32 HL134643 and the MCW CVRC A.O. Smith Fellowship Program (STH) and NIH R01HL173549 and AHA SFRN 847970 (AMB). This abstract was presented at the American Physiology Summit 2026 and is only available in HTML format. There is no downloadable file or PDF version. The Physiology editorial board was not involved in the peer review process.