Hidden Vascular Flaw in PCOS Emerges Only When Blood Sugar Rises

Women with polycystic ovary syndrome, or PCOS, carry a sharply elevated risk of cardiovascular disease that has long outpaced explanations based on weight, hormones, or cholesterol alone. Now a research team in Taiwan has uncovered what may be a piece of that puzzle hiding inside the blood vessels themselves: endothelial cells derived from women with PCOS harbor a latent mitochondrial vulnerability that stays quiet under normal conditions but collapses under the metabolic stress of high blood sugar. The finding, published in iScience, also points to myeloperoxidase, an enzyme better known as a marker of inflammation, as a potential intracellular accomplice in the damage.

The study tackled a stubborn technical problem. Endothelial cells, the thin lining of blood vessels that governs vascular health, are nearly impossible to extract directly from living patients for study. To get around this, the researchers turned to induced pluripotent stem cells, or iPSCs, generated from blood samples of three women with PCOS and three women without the condition. Because iPSC reprogramming resets most environmental epigenetic marks, any defects that persist into cells derived from these lines likely reflect genetic susceptibility or stable epigenetic memory rather than the current hormonal environment of the donor. The team coaxed these stem cells into endothelial cells with high efficiency, producing a controlled platform in which the innate, cell-autonomous properties of PCOS vasculature could be examined in isolation.

The central experiment was elegantly simple. The researchers exposed both PCOS-derived and control endothelial cells to either normal glucose at 5.5 millimolar or high glucose at 30 millimolar, a concentration routinely used to model the hyperglycemic states seen in diabetes, for 24 hours. Under normal glucose, both cell types formed capillary-like tubes on Matrigel with equal competence. Under high glucose, the picture split dramatically. Tube formation, a standard readout of angiogenic capacity, was markedly and significantly impaired in all three PCOS lines, while the control cells carried on as if nothing had happened. The defect was consistent across every independent PCOS cell line tested, a sign that the vulnerability is not an artifact of one donor.

What the researchers found inside the cells explains the functional collapse. Confocal microscopy of the outer mitochondrial membrane protein TOM20 revealed that PCOS endothelial cells exposed to high glucose underwent striking mitochondrial fragmentation, their normally interconnected tubular networks breaking into short, disconnected pieces. Quantitative analysis with the Mitochondrial Network Analysis toolset confirmed reductions in mitochondrial footprint, mean branch length, and the number of branches per network. Control cells maintained their mitochondrial architecture under identical conditions. The dynamics proteins told a subtler story: control cells responded to glucose stress by upregulating the fusion proteins MFN1, MFN2, and OPA1, a known protective maneuver that dilutes damaged mitochondrial DNA, while the PCOS cells failed to mount this compensatory response and instead increased the fission proteins DRP1 and FIS1. In other words, the pathology involves not just susceptibility to injury but an impaired adaptive response to stress.

Functional measurements drove the point home. Using the MitoSOX Red fluorescent probe, the team detected a significant accumulation of mitochondrial reactive oxygen species in high-glucose-treated PCOS endothelial cells but not in controls. Seahorse extracellular flux analysis then showed that the same cells suffered reductions in basal respiration, maximal respiration, ATP production, and spare respiratory capacity, the metabolic reserve that healthy mitochondria keep for emergencies. Control cells showed none of these deficits. Taken together, the data indicate that a single day of hyperglycemia is enough to unmask a latent bioenergetic defect in PCOS endothelium, one that healthy endothelium simply does not share.

The second half of the story concerns myeloperoxidase, or MPO, a heme peroxidase classically produced by neutrophils and monocytes and long recognized as a cardiovascular risk biomarker, particularly in type 2 diabetes, where elevated MPO independently predicts cardiovascular events beyond traditional risk factors. Prior studies had already reported elevated circulating MPO in women with PCOS. What makes the new work provocative is the finding that endothelial cells themselves express MPO. Reverse transcription quantitative PCR detected MPO mRNA in all PCOS and control lines, and western blotting revealed the characteristic autocatalytic MPO fragments at 43 to 47 kilodaltons. After 24 hours of high glucose, MPO mRNA and protein rose significantly in the PCOS cells. Notably, MPO was undetectable in the culture supernatant, indicating that in these endothelial cells the enzyme acts intracellularly rather than being secreted, a localization consistent with recent reports of nuclear MPO in endothelial colony-forming cells from coronary artery disease patients.

The most striking result came from pharmacological intervention. When the researchers pre-treated the cells with 4-aminobenzoic acid hydrazide, or 4-ABAH, a potent MPO inhibitor, the high-glucose-induced surge in mitochondrial ROS was blunted and the impaired tube formation was substantially restored. A mannitol osmotic control confirmed that the damage was driven by glucose itself rather than by osmotic pressure. This suggests that endogenous, intracellular MPO is not merely a passive bystander but an active contributor to the mitochondrial oxidative stress and angiogenic failure that hyperglycemia provokes in PCOS endothelium. The authors are careful to note that pharmacological inhibition alone cannot establish definitive causality without genetic gain- and loss-of-function models, but the convergent evidence is compelling.

To anchor the cellular findings in clinical reality, the team measured serum MPO in 22 treatment-naïve women with PCOS and 20 age- and body mass index-matched controls at National Taiwan University Hospital. The PCOS group averaged 668.6 nanograms per milliliter of MPO compared with 393.9 in controls, a statistically significant difference. The PCOS cohort also showed the expected burden of metabolic comorbidity: 77 percent met criteria for insulin resistance, 50 percent for pre-diabetes, and 50 percent for hypertension, against 10, 0, and 10 percent respectively in controls. Pearson correlation analysis revealed that in the PCOS group, serum MPO rose in parallel with diastolic blood pressure, fasting glucose, and LDL cholesterol, associations absent in controls. Crucially, multivariable regression showed that the MPO-glucose relationship remained significant after adjusting for BMI and insulin, suggesting that hyperglycemia drives MPO elevation independently of insulin resistance, the hallmark traditionally blamed for PCOS metabolic pathology. The association weakened when blood pressure was added to the model, hinting at interconnections among these vascular parameters that future work will need to disentangle.

The study’s implications reach beyond PCOS. It reframes vascular disease risk in these women as stemming partly from a stable, cell-autonomous mitochondrial defect rather than solely from systemic hormonal and metabolic derangements. Because iPSC reprogramming should erase acquired environmental marks, the persistence of the phenotype raises the possibility of genetic susceptibility or durable epigenetic memory, echoing the concept of hormonal imprinting. It also elevates MPO from biomarker to potential therapeutic target, at least in principle, since inhibiting it functionally rescued the cells in vitro. The authors acknowledge limitations: only three iPSC lines per group, a clinical cohort too small for phenotypic subgroup analysis, an acute rather than chronic glucose challenge, and the absence of direct intracellular MPO measurements in patient endothelial cells. Still, the consistency across independent lines, complementary assays, and the parallel clinical cohort gives the model considerable weight.

For the millions of women living with PCOS, the message is a mechanistic warning: the blood vessel lining may already be primed to fail when blood sugar climbs, even before diabetes develops. The half of the PCOS cohort with pre-diabetes in this study illustrates how common that trigger already is. Whether MPO-targeted strategies or aggressive early glucose control can translate this cellular insight into reduced cardiovascular events remains the essential next question, one that the intracellular MPO-mitochondrial axis now provides a concrete framework for testing.

Subject of Research: Mitochondrial dysfunction and myeloperoxidase dysregulation in hyperglycemia-stressed endothelial cells derived from PCOS patient iPSCs

Article Title: Hyperglycemia induces endothelial mitochondrial dysfunction and myeloperoxidase dysregulation in PCOS iPSCs

Article References: Wu, C.-E., Huang, C.-C., Chen, T.-H., Tsou, C.-C., Yang, I.-J., Chang, Y.-Y., Chen, M.-J., & Ho, H.-N. (2026). Hyperglycemia induces endothelial mitochondrial dysfunction and myeloperoxidase dysregulation in PCOS iPSCs. iScience, 29(11), Article 117852. https://doi.org/10.1016/j.isci.2026.117852

Image Credits: AI Generated

DOI: 10.1016/j.isci.2026.117852

Keywords: PCOS, endothelial dysfunction, mitochondria, myeloperoxidase, hyperglycemia, induced pluripotent stem cells, reactive oxygen species, cardiovascular risk, insulin resistance, angiogenesis, oxidative stress, iScience