Mouse study links endothelial FoxO1 to female vascular resilience

Endothelial FoxO1 has opposite effects on vascular and metabolic adaptation in female and male mice, according to a study combining conditional genetic experiments, tissue imaging, RNA sequencing and cultured adipose endothelial cells. Reducing FoxO1 specifically in endothelial cells impaired adipose angiogenesis, enlarged and disorganized capillaries, reduced vascularization in brown fat and skeletal muscle, and worsened glucose regulation in females. In males, the same intervention produced the opposite pattern: greater capillarization and improved metabolic outcomes.

The experiments identify FoxO1 as a sex-dependent regulator of endothelial resilience but do not establish a clinical strategy or a treatment effect in humans. The investigation began with an observation that FoxO1 expression and activity were higher in female endothelial cells. Foxo1 messenger RNA in freshly isolated mouse adipose endothelial cells was approximately twice as high in females as in males.

Cultured female cells also contained more FoxO1 protein and showed greater activity in a forkhead response-element reporter assay, indicating that the difference persisted outside the intact animal. Analysis of a published single-cell atlas containing 68,503 vascular cells from 65 human donors found higher FOXO1 expression in subcutaneous adipose endothelial cells from female donors than from male donors. That human analysis compared expression only; it did not test whether manipulating FOXO1 would affect human vascular function or disease.

To examine the consequences of reduced endothelial FoxO1 in females, researchers used a tamoxifen-inducible, endothelial-selective model that reduced Foxo1 in endothelial cells. Foxo1 messenger RNA in adipose endothelial cells fell by 58% in females, a reduction similar to the 60% measured in males. The animals received either a low-fat diet or a high-fat diet for periods of four, seven or 16 weeks.

Under the high-fat diet, females with reduced endothelial FoxO1 gained weight more rapidly, accumulated more white adipose tissue and developed adipocytes approximately 1.7 times larger than those in controls. Their adipose capillary density fell by about 1.3-fold after seven weeks and 1.5-fold after 16 weeks, while capillary diameters increased about 1.7-fold and 2.7-fold at those respective times. Some vessels had monolayer-like or glomerular-like forms, consistent with abnormal remodeling.

Tissue Vegfa expression did not differ between depleted and control females, suggesting that the observed defect was not explained by a simple change in this angiogenic signal. The vascular abnormalities were not confined to white adipose tissue. In high-fat-fed females with reduced endothelial FoxO1, brown adipose tissue had lower expression of the endothelial markers Pecam1, Cdh5 and Tie1, as well as lower Ucp1 expression, a marker used in the study to assess brown-fat thermogenic function.

The capillary-to-muscle-fiber ratio in skeletal muscle was about 18% lower than in controls. At the systemic level, the females had higher fasting glucose, reduced insulin sensitivity and moderately impaired glucose tolerance. Defects also appeared without high-fat feeding: after 16 weeks on a low-fat diet, females with reduced endothelial FoxO1 showed modest additional weight gain, enlarged adipocytes, approximately 1.6-fold less adipose capillarization, 2.8-fold greater capillary diameter, lower brown-fat endothelial and Ucp1 expression, less skeletal-muscle capillarization and impaired insulin sensitivity.

Glucose tolerance under the low-fat diet was unaffected. The results therefore show that the phenotype was already detectable without diet-induced obesity but became more pronounced with prolonged high-fat feeding. RNA sequencing showed that reduced FoxO1 produced broader transcriptional changes in female endothelial cells than in male cells.

After seven weeks of high-fat feeding, the analysis identified 1,303 differentially expressed genes in females and 1,006 in males at an adjusted significance threshold of less than 0.05. Principal-component analysis separated female control and FoxO1-reduced samples, while the equivalent groups of males were not significantly distinct along the first component. Importantly, female cells depleted of FoxO1 shifted toward the expression profile of control male cells and no longer differed significantly from them.

In females, reduced FoxO1 suppressed gene programs associated with proliferation, cellular respiration, nucleosome organization and DNA repair. It increased programs related to superoxide metabolism and an endothelial senescence-associated secretory phenotype. Programs involving transforming growth factor beta receptor signalling, immune activation, neutrophil chemotaxis and cell-shape regulation changed in both sexes, but with greater magnitude in females.

These are expression patterns rather than direct evidence of every corresponding physiological process. The transcriptional results were reflected in vivo by a selective loss of endothelial proliferation after FoxO1 depletion in females. Labeling of dividing cells in adipose tissue showed fewer proliferating endothelial cells in females with reduced FoxO1 than in female controls.

The corresponding depletion did not significantly reduce endothelial proliferation in males. This result qualifies the view of FoxO1 solely as a brake on cell division: in female adipose endothelium, sustained reduction of FoxO1 was associated with lower proliferation and broader cellular dysfunction. The study did not identify the full sequence of events connecting the transcriptomic changes to impaired tissue angiogenesis, and it did not directly establish that senescence alone caused the vascular phenotype.

Cultured adipose endothelial cells revealed that both duration and sex changed FoxO1’s effects. One day of inhibition with AS1842856 increased proliferation in cells from both sexes, consistent with FoxO1’s short-term role in restraining cell-cycle entry. After six days, proliferation was reduced in female cells but remained above control levels in male cells, although the male increase was smaller than after brief inhibition.

They also increased Tgfb1 messenger RNA and accumulated phosphorylated Smad2, a downstream mediator of transforming growth factor beta signalling. Yet blocking that pathway with RepSox further reduced, rather than restored, proliferation. The anti-proliferative effect therefore appeared to depend on p21 but not on transforming growth factor beta signalling under the tested conditions.

Energy metabolism and mitochondrial organization provided a possible link between prolonged FoxO1 inhibition and the arrest of female endothelial cells. After one day of treatment, glycolytic ATP production increased in both sexes, but mitochondrial ATP production declined specifically in female cells. After six days, total and mitochondrial ATP production were impaired in females, whereas male cells showed no significant differences from controls in these measures.

Female mitochondria became diffusely stained and their measured network length decreased, while mitochondrial networks expanded in male cells. Transcripts encoding several electron-transport-chain components increased in both sexes, and the tested fusion factor Mfn2 and fission factor Fis1 were unchanged. These observations identify a sex-specific effect on mitochondrial structure and function without establishing that altered expression of those particular genes was responsible.

Mitochondrial disruption was accompanied by increased oxidative stress and DNA damage in female endothelial cells. Control female cells had lower levels of 4-hydroxy-2-nonenal, a product of lipid peroxidation, than control male cells. Six days of FoxO1 inhibition significantly increased this marker in females until it reached the level observed in males, while having no significant effect in males.

Antioxidant genes Prdx3 and Gpx8 were upregulated in both sexes, showing that the greater lipid peroxidation in females did not result from a general failure to induce those antioxidant genes. Female cells also accumulated more γH2AX foci, a marker associated with DNA double-strand breaks. Nuclear p53 increased in both sexes, whereas Gadd45a, a gene linked to DNA repair and genomic stability, was induced only in male cells.

The study thus associated female vulnerability with a combination of elevated oxidative injury and a less pronounced induction of this repair-associated response. Prolonged inhibition produced a senescence-like phenotype primarily in female endothelial cells. Their nuclei enlarged, the senescence marker p16INK4a increased and the number of cells expressing that marker rose.

Nuclear enlargement was associated with greater γH2AX accumulation, although unchanged ribosomal protein S6 phosphorylation indicated that the enlargement was not explained by increased protein synthesis. After nine days of FoxO1 inhibition, male endothelial cells recovered proliferative capacity during three days without the inhibitor, whereas female cells remained below control levels. Co-treatment with the antioxidant N-acetylcysteine partially rescued the female phenotype: it improved proliferation, prevented nuclear enlargement and reduced p16INK4a.

This intervention implicated reactive oxygen species as a contributor, although the study did not directly measure endothelial mitochondrial morphology or oxidative stress in female adipose tissue in vivo. Taken together, the findings show that endothelial FoxO1 is a context-dependent regulator whose function cannot be inferred from short-term proliferation experiments alone. It initially restrains cell growth, but under sustained depletion it supports mitochondrial organization, redox balance, DNA repair programs and resistance to senescence-like arrest, particularly in female endothelial cells.

The experiments were conducted in mice and cultured mouse cells, with some outcomes measured under low-fat and high-fat diets and others after pharmacological inhibition. Researchers were not blinded to sex, genotype or diet, experiments were not randomized, and no statistical method was used to predetermine sample sizes. The work therefore identifies a molecular dependency and a set of mechanisms for further study, rather than demonstrating that manipulating FOXO1 would protect people from obesity-related vascular or metabolic disease.

Subject of Research: Sex-dependent endothelial FoxO1 control of vascular and metabolic adaptation to obesity in mice

Article Title: Female-specific reliance on endothelial FoxO1 governs vascular resilience and metabolic adaptation to obesity

Article References: Pislaru, A., Roudier, E., Rezvan, O., Breadmore, C., Fathi, M., Nader, G., AlZaim, I., Nwadozi, E., Lee, G., Yavari, E., Fawzy, A., Rullman, E., Kalucka, J., Gustafsson, T., Rudnicki, M., & Haas, T. L. (2026). Female-specific reliance on endothelial FoxO1 governs vascular resilience and metabolic adaptation to obesity. Angiogenesis, 29(4), Article 79. https://doi.org/10.1007/s10456-026-10093-1

Image Credits: AI Generated

DOI: 10.1007/s10456-026-10093-1

Keywords: FoxO1, endothelial cells, sex differences, vascular resilience, obesity, adipose tissue, angiogenesis, mitochondrial dysfunction, oxidative stress, cellular senescence, glucose metabolism, mouse study