Genistein Restores Nitric Oxide Bioavailability and Attenuates Oxidative Stress in Angiotensin II-Induced Hypertensive Mice

Abstract

Genistein is a dietary isoflavone that is abundant in soy products and has been suggested to exert cardiovascular protective effects; however, its role in hypertension remains incompletely defined. Here, we investigated whether genistein ameliorates angiotensin II (Ang II)-induced hypertension and endothelial dysfunction using complementary in vivo, ex vivo, and in vitro models. The oral administration of genistein substantially attenuated Ang II-induced elevations in systolic and diastolic blood pressure in mice without affecting body weight. Histological analyses revealed that genistein alleviated aortic wall thickening and smooth muscle hypertrophy, whereas vascular reactivity assays demonstrated improved endothelium-dependent relaxation while preserving the endothelium-independent responses. In isolated murine aortic rings, genistein reversed Ang II-induced hypercontractility and restored acetylcholine-mediated vasodilation. Furthermore, in human umbilical vein endothelial cells, genistein suppressed the Ang II-induced overproduction of reactive oxygen species (ROS) and restored nitric oxide (NO) bioavailability. Mechanistically, genistein improved endothelial redox-NO coupling, accompanied by transcriptional remodeling of the redox/NO axis [e.g., suppression of nicotinamide adenine dinucleotide phosphate oxidase 4 (NOX4) and nitric oxide synthase 2 (NOS2) and restoration of nitric oxide synthase 3 (NOS3)], which is consistent with reduced ROS and enhanced NO bioavailability. Collectively, our findings identify genistein as a potent endothelial-protective agent that ameliorates Ang II-induced hypertension, highlighting its therapeutic potential in cardiovascular disease driven by vascular dysfunction.

INTRODUCTION

Hypertension is one of the most consequential and poorly controlled drivers of global cardiovascular morbidity and mortality. Despite the widespread availability of guideline-directed therapy, the detection, treatment, and control rates remain suboptimal across regions, resulting in a large burden of preventable events and target-organ damage (Mills et al., 2020; Mensah et al., 2023). Endothelial dysfunction is a proximate mediator of this risk. By losing the capacity to regulate vascular tone, thrombosis, inflammation, and growth, the diseased endothelium accelerates arterial remodeling and stiffness, thereby amplifying the neurohumoral pressor signals that sustain hypertension (Godo and Shimokawa, 2017; Gallo et al., 2022).

Endothelial dysfunction is characterized by the disturbed redox-nitric oxide (NO) axis. Under physiological conditions, endothelial NO synthase (eNOS)-derived NO regulates leukocyte/platelet activation and maintains vasodilator tone, whereas reactive oxygen species (ROS) are tightly ameliorated (Förstermann and Sessa, 2012; Tejero et al., 2019). In hypertension, upregulated nicotinamide adenine dinucleotide phosphate (NADPH) oxidases generate superoxide that rapidly quenches NO to peroxynitrite, oxidizes tetrahydrobiopterin, and promotes the uncoupling of eNOS, thereby depleting the bioavailable NO and establishing a self-reinforcing cycle of oxidative stress and impaired vasodilation (Kuzkaya et al., 2003; Landmesser et al., 2003). Translational and interventional studies show that restoring NO bioavailability or reducing vascular ROS levels improves endothelial function and can lower blood pressure in relevant models (Drummond et al., 2011; Vita, 2011; Moreau et al., 2012; Ras et al., 2013; Griendling et al., 2021). Nevertheless, even with effective blood pressure and lipid-lowering interventions, considerable “residual risk” remains, thus motivating the identification of adjunct strategies that directly protect the endothelium (Sampson et al., 2012; Reith and Armitage, 2016; Dhindsa et al., 2020; Manta et al., 2024).

Diet-derived bioactives with antioxidant and endothelial activity have attracted interest as mechanistic adjuncts. Meta-analyses of flavonoid-rich interventions report modest but consistent improvements in flow-mediated dilation (FMD), a noninvasive index that predicts future cardiovascular events, and small blood pressure reductions, particularly in high-risk subgroups (Zhang et al., 2015; Ebaditabar et al., 2020; Jenkins et al., 2021; Zhao et al., 2022; Luo et al., 2023). Genistein is a soy-derived isoflavone that is notable for its phytoestrogenic signaling, peroxisome proliferator-activated receptor (PPAR) engagement, and tyrosine kinase inhibition (Akiyama et al., 1987; Markovits et al., 1989; Fukutake et al., 1996; Morito et al., 2001; Dang et al., 2003; Kim et al., 2004). In vascular systems, genistein preserves NO signaling, blunts oxidative/inflammatory pathways, and favorably modulates endothelial biology in cell and animal models (Kapiotis et al., 1997; Dang et al., 2003; Kim et al., 2004; Yang et al., 2010; Lin et al., 2011; Lu et al., 2019; Rahman et al., 2020; Li et al., 2023). In vivo, studies in classic hypertension models (e.g., spontaneously hypertensive rats or NO-deficiency-induced states) suggest that genistein may enhance eNOS activity, reduce NADPH-oxidase-dependent superoxide production, and improve vasodilator responsiveness (Vera et al., 2007; Si and Liu, 2008; Lin et al., 2011; Poasakate et al., 2021). Early clinical data in postmenopausal women further demonstrate improved endothelial function and small blood pressure effects with purified genistein or isoflavone-rich matrix supplementation, albeit with heterogeneity according to dose, formulation, and vascular phenotype (Squadrito et al., 2003, 2013; Taku et al., 2010; Liu et al., 2012; Irace et al., 2013; De Gregorio et al., 2017; Lei et al., 2024; Feng et al., 2025).

Despite these advances, important gaps remain. First, the efficacy of genistein against angiotensin II (Ang II)-driven endothelial injury, which is the canonical hypertension-associated pressor/oxidative stimulus, has not been rigorously tested across in vivo, ex vivo, and cellular models to assess the localized effects on the endothelium (i.e., endothelium-dependent vs. endothelium-independent responses). Second, the relevance of exposure is debated because circulating genistein is predominantly conjugated; whether aglycone exerts direct endothelial actions at physiologically achievable concentrations remains unclear. Third, mechanistic anchoring to the redox-NO axis in Ang II hypertension, including effects on ROS generation and NO bioavailability, has rarely been integrated into functional vascular phenotypes.

In this study, we hypothesized that genistein mitigates Ang II-induced hypertension by restoring the endothelial redox-NO balance. We combined in vivo Ang II infusion with serial blood pressure measurements, ex vivo conduit-artery myography to separate endothelium-dependent vs. endothelium-independent responses, and complementary cellular assays of ROS/NO signaling at exposure and physiologically relevant concentrations. This multiscale design allowed us to define where genistein acts along the vascular wall, establish mechanistic plausibility, and evaluate its translational potential as an adjunct strategy for hypertension treatment.

MATERIALS AND METHODS

Animals

All animal experimental procedures were reviewed and approved by the Animal Ethics Committee of the School of Public Health (Shenzhen), Sun Yat-sen University, and the Institutional Animal Care and Use Committee of Sun Yat-sen University (approval no.: 2024002932). Male C57BL/6 mice (10 weeks old) were purchased from Zhuhai BesTest Bio-Tech Co., Ltd., and housed in the Sun Yat-sen University Laboratory Animal Center. Mice were maintained under specific pathogen-free conditions with controlled temperature, a 12-h light/dark cycle, and ad libitum access to standard specific pathogen-free rodent chow and water. The mice were randomized into four groups: control, Ang II, Genistein, and Ang II+Genistein (n=5).

Genistein (Yeasen) was dissolved in corn oil (Aladdin Scientific) and administered via oral gavage at a dose of 50 mg/kg/d for 28 days. Mice in the control and Ang II (MedChemExpress) groups received an equal volume of corn oil as a vehicle (Servicebio). Beginning on Day 14, the Ang II and Ang II+Genistein groups received continuous subcutaneous infusion of Ang II (800 ng/kg/min), which was freshly prepared in sterile normal saline, via osmotic minipumps (Model 1002, Alzet) for 14 days to induce hypertension (Yang et al., 2025). The control and Genistein groups were implanted with minipumps containing sterile normal saline as vehicles. Mice were anesthetized with inhaled isoflurane (MedChemExpress) and placed on a temperature-controlled operating table to maintain body temperature at 37.0±1.0°C, as previously described (Jiang et al., 2021). A schematic of the timeline is provided in Fig. 1A. Mice were euthanized by carbon dioxide (CO₂) inhalation at the indicated time points before tissue collection (Hennen et al., 2025).

Fig. 1.

Effects of genistein (Gen) on blood pressure and changes of aortic pathology in angiotensin II (Ang II)-induced hypertensive mice. (A) Schematic of the in vivo experimental design. Mice received subcutaneous Ang II via an osmotic minipump to induce hypertension, and Gen or vehicle was administered according to the indicated dosing regimen. Blood pressure was measured serially (baseline and follow-up time points), followed by tissue collection for histology and biochemical assays. (B) Systolic blood pressure (SBP) and diastolic blood pressure (DBP) of mice from the different treatment groups. Statistical analysis for SBP/DBP was performed using two-way analysis of variance with Tukey’s multiple comparisons test. Adjusted P-values were indicated as follows: SBP: Ang II vs. control, P<0.001; Ang II+Gen vs. Ang II, P=0.002; and Gen vs. Ang II+Gen, P=0.016; and for DBP: Ang II vs. control, P<0.001; and Ang II+Gen vs. Ang II, P=0.002. (C) Representative hematoxylin and eosin (H&E) staining of the aorta. (D) Quantification of the aortic wall thickness. Data are presented as the mean±standard error of the mean, n=5 per group. Groups not sharing a common letter are significantly different (P<0.05), whereas groups sharing the same letter are not significantly different. Letters are assigned in descending order of mean values (i.e., a indicates the highest mean).

Blood pressure measurement

Systolic and diastolic blood pressures (DBPs) were measured using noninvasive tail-cuff plethysmography after acclimation (Zhang et al., 2014). For each session, ≥10 valid cycles per mouse were averaged. Measurements were obtained on Day 0, 7, 14, 21, and 28. The tail-cuff methods correlate reasonably with telemetry at the group level but can diverge for individual readings; therefore, we standardized the factors of acclimation, restraint, and warming to minimize variability (Wilde et al., 2017; Harrison et al., 2024).

Tissue collection and processing

Thoracic aortae were harvested immediately after the mice were euthanized and carefully cleaned of surrounding adipose and connective tissues under a dissecting microscope. Each thoracic aorta was divided into four segments and allocated for vascular tension measurement, en face fluorescence staining, cryosection preparation, and paraffin sectioning.

One freshly isolated thoracic aortic ring (2－3 mm in length) was immediately placed in Krebs buffer and used in isometric wire myography experiments. A second freshly isolated aortic segment was incubated with dihydroethidium (DHE, Beyotime) without prior fixation and immediately processed for en face imaging of endothelial ROS. A third freshly isolated aortic ring was cultured in appropriate medium and exposed to Ang II in the presence or absence of genistein for 24 h, after which the vascular function and ROS generation were assessed. The remaining thoracic aortic segment was fixed in 4% paraformaldehyde (Servicebio), processed for paraffin embedding, sectioned, and stained with hematoxylin and eosin for histological analysis.

Isometric wire myography

The aortic rings were mounted on a wire myograph at 37°C in Krebs buffer (95% O₂/5% CO₂) and set to a resting tension of 3 mN. Following equilibration, viability was confirmed with high-K⁺ Krebs (60 mM). The Krebs buffer was prepared using sodium chloride (Sangon Biotech), potassium chloride (Sigma-Aldrich), calcium chloride (Sigma-Aldrich), magnesium sulfate heptahydrate (Yeasen), potassium dihydrogen phosphate (Macklin), sodium bicarbonate (MedChemExpress), D-(+)-glucose (Sangon Biotech), and ethylenediaminetetraacetic acid (MedChemExpress). The cumulative concentration-response curves were generated for 1 nM－10 µM phenylephrine (Phe, Aladdin Scientific). Endothelium-dependent relaxation (EDR) was assessed with 1 nM－10 µM acetylcholine (ACh, MedChemExpress) after Phe preconstriction (∼10 µM). Endothelium-independent relaxation was assessed with 1 nM－10 µM sodium nitroprusside (SNP, MedChemExpress). Tension was continuously recorded and analyzed as previously described (Zhang et al., 2015).

Hematoxylin and eosin staining

Aortae were fixed in 4% paraformaldehyde (48 h), dehydrated through graded ethanol concentrations (XiHua Chemical), cleared with xylene (XiHua Chemical), embedded in paraffin, and sectioned at 5－6 µm. Hematoxylin and eosin staining was performed according to the manufacturer’s instructions. Morphometric analysis was conducted on predefined regions of interest with identical imaging settings across the groups.

En face endothelial ROS imaging

Fresh aortae were incubated in the dark with DHE (5 µM) according to the manufacturer’s instructions. They were opened longitudinally, gently flattened lumen-side down, and imaged with confocal microscopy to detect red fluorescence (Ex/Em 535/570 nm). Elastin autofluorescence (Ex/Em 494/520 nm) was obtained for orientation. Acquisition parameters were kept constant across the groups and samples. Fluorescence intensity was quantified as a relative index of endothelial ROS.

Serum collection

At the study endpoint, blood was collected from the mice by cardiac puncture. Whole blood was allowed to clot at 22－25°C for 2 h and then centrifuged at 3,000 g for 15 min to obtain serum. Aliquots of serum were immediately frozen at −80°C until analysis.

Serum total NO metabolite (NOx) measurement

Because free NO is short-lived in vivo, circulating nitrite and nitrate (total NOx) were quantified as an integrated index of systemic NO availability. Serum total NOx levels were measured using a commercial colorimetric kit based on enzymatic nitrate reduction followed by the Griess reaction (CheKine^TM micro NO assay kit, Abbkine) according to the manufacturer’s instructions. Absorbance was measured at 540 nm using a microplate reader, and the concentrations were calculated from a sodium nitrite (NaNO₂, Sigma-Aldrich) standard curve. Serum samples were diluted as needed to ensure that the values fell within the linear range of the assay.

Untargeted serum metabolomic profiling

Serum untargeted metabolomic profiling was performed by BGI Genomics using liquid chromatography-mass spectrometry (LC-MS). Metabolite features were annotated by comparing the accurate mass, retention time, and/or MS/MS spectra against in-house and public metabolite databases where available. The relative abundance of genistein and genistein-derived metabolites was compared across the experimental groups, and the results are presented as the normalized signal intensity (relative abundance).

Ex vivo Ang II and genistein exposure in aortic rings

To isolate the direct vascular effects, thoracic aortic rings were cultured ex vivo and exposed to Ang II (1 µM) with or without genistein (1 or 10 µM) for 24 h. The contractile responses to Phe, EDR to ACh, and relaxation responses to SNP in the presence or absence of N^ω-nitro-L-arginine methyl ester (L-NAME, MedChemExpress) were recorded as above in the section of isometric wire myography. ROS in the vascular wall was assessed by DHE staining.

Cell culture and treatment

Human umbilical vein endothelial cells (HUVECs, National Collection of Authenticated Cell Cultures) were maintained in high-glucose Dulbecco’s modified Eagle medium (Thermo Fisher Scientific) with Endothelial Cell Growth Supplement (Sigma-Aldrich), 10% fetal bovine serum (Thermo Fisher Scientific), and penicillin/streptomycin (Thermo Fisher Scientific) at 37°C under 5% CO₂ conditions. At ∼70% confluence, the cells were treated with the vehicle, Ang II (1 µM), Ang II+Genistein (1 µM), or Ang II+Genistein (10 µM) for 24 h.

Intracellular ROS and NO detection in HUVECs

After treatment, the cells were incubated with 10 µM 2’,7’-dichlorodihydrofluorescein diacetate (DCFH-DA, Thermo Fisher Scientific) in the dark for 30 min at 37°C, according to the manufacturer’s instructions. They were then washed and imaged using epifluorescence microscopy (Ex/Em 488/525 nm). Representative fields of each group were captured with identical exposure settings. For NO measurement, the cells were incubated with 5 µM 4-amino-5-methylamino-2’,7’-difluorofluorescein diacetate (DAF-FM diacetate, Thermo Fisher Scientific) in the dark for 30 min at 37°C, according to the manufacturer’s instructions. After washing, calcimycin (A-23187, MedChemExpress, 1 µM in Hanks’ Balanced Salt Solution) was added immediately before imaging (Ex/Em ∼495/515 nm). The exposure settings were identical across the group conditions, and representative fields were captured by a blinded operator.

Real-time reverse transcription polymerase chain reaction (RT-qPCR) analysis

Total RNA was extracted from cultured cells using the MolPure^Ⓡ TRIeasy plus total RNA kit (Yeasen), according to the manufacturer’s instructions, and 2 µg total RNA was reverse transcribed into cDNA using Hifair^Ⓡ AdvanceFast One-step RT-gDNA Digestion SuperMix for qPCR (Yeasen). For RT-qPCR analysis, a reaction mixture consisting of 10 µL of 2× SYBR Master Mix (Hieff^Ⓡ qPCR SYBR Green Master Mix, Yeasen), 1 µL of each 10 µM primer, 2 µL of cDNA template, and nuclease-free water adjusted to a final volume of 20 µL was tested on the Roche LightCycler 480 Real-Time Thermocycler 384-Well Block. Gapdh was used as the reference gene, and the relative gene expression levels were calculated using the 2^−ΔΔCt method. Sangon Biotech synthesized the primers, and the primer sequences are listed in Supplementary Table 1.

Statistical analyses

Data are presented as the mean±standard error of the mean. Two-group comparisons used unpaired, two-tailed Student’s t-tests. Multiple-group comparisons were performed using one-way or two-way analysis of variance with planned post hoc tests (Tukey). Statistical significance was set at P<0.05. When letter annotations are present, groups with different letters differ significantly (P<0.05), whereas groups sharing at least one letter do not differ significantly. Analyses and plots were generated using GraphPad Prism 9.0. The experimental unit (n) and the exact tests applied to each dataset are specified in the figure legends.

RESULTS

Genistein lowers blood pressure and ameliorates aortic pathology in Ang II-infused mice

We first evaluated whether oral genistein mitigates the hemodynamic and structural consequences of Ang II-induced hypertension. Compared with the controls, continuous Ang II infusion robustly increased the systolic blood pressure (SBP) and DBP, and genistein administration significantly attenuated both elevations (Fig. 1B). Histopathological analysis via hematoxylin and eosin staining revealed pronounced aortic wall thickening and abnormal smooth muscle cell hypertrophy in the Ang II-induced hypertensive mice, both of which were markedly ameliorated by genistein administration (Fig. 1C and 1D).

Genistein improves vascular function and reduces vascular and endothelial ROS in vivo

To localize the vascular compartment responsible for this hemodynamic benefit, we performed isometric myography on the thoracic aortae after in vivo exposure (Fig. 2A). Ang II infusion shifted the Phe-induced vasocontractility in the aorta. In contrast, genistein reduced the hypercontractility toward that of the control group levels (Fig. 2A and 2B). Ang II markedly impaired EDR to ACh and was restored by genistein treatment (Fig. 2A and 2C). In contrast, endothelium-independent relaxation to SNP in the presence of L-NAME did not differ significantly among the groups (Fig. 2D), thus indicating that genistein primarily exerts its beneficial effects via endothelial modulation. To verify systemic exposure under our dosing regimen, untargeted LC-MS serum metabolomic profiling at the study endpoint confirmed the presence of genistein and several genistein-derived metabolites in the genistein-treated mice, including genistein, genistein 4’-O-glucuronide, genistein sulfate, and 8-hydroxygenistein (Fig. 2E). Among these, 8-hydroxygenistein was significantly elevated in the genistein-treated group compared with the control and Ang II groups, thereby supporting the in vivo exposure and metabolism of genistein in the animals studied.

Fig. 2.

Effects of in vivo genistein (Gen) treatment in vascular function in angiotensin II (Ang II)-induced hypertensive mice. (A) Representative tracings illustrating the effects of Gen on vasoconstriction and vasodilation. (B) Phenylephrine (Phe)-induced contraction of the aorta. Statistical analysis for the contraction curve was performed using two-way analysis of variance with Tukey’s multiple comparisons test. Adjusted P-values were indicated as follows: Ang II vs. control, P=0.012; Ang II+Gen vs. Ang II, P=0.012; Ang II+Gen vs. Gen, P=0.023; and Gen vs. control, P=0.120. (C) Acetylcholine (ACh)-induced endothelium-dependent relaxation of the aorta. Statistical analysis for the relaxation curve was performed using two-way analysis of variance with Tukey’s multiple comparisons test. Adjusted P-values were indicated as follows: Ang II vs. control, P<0.001; Ang II+Gen vs. Ang II, P<0.001; Ang II+Gen vs. Gen, P=0.001; Gen vs. control, P=0.002. (D) Sodium nitroprusside (SNP)-induced endothelium-independent relaxation of mouse aortae in the presence of N^ω-nitro-L-arginine methyl ester (L-NAME). (E) Untargeted serum metabolomic analysis showing the relative abundance of Gen and its major metabolites, including Gen 4’-O-glucuronide, Gen sulfate, and 8-hydroxygenistein, in the control, Ang II, Gen, and Ang II+Gen mice. All data are presented as the mean±standard error of the mean, n=5 per group. Groups not sharing a common letter are significantly different (P<0.05), whereas groups sharing the same letter are not significantly different. Letters are assigned in descending order of the mean values (i.e., a indicates the highest mean).

Given the central role of redox stress in endothelial dysfunction, we quantified ROS using DHE fluorescence. We observed that Ang II infusion significantly elevated the ROS levels in the aortic wall (Fig. 3A and 3B) and en face endothelium (Fig. 3C and 3D) relative to the normotensive controls. Genistein significantly lowered ROS in both compartments in hypertensive mice (Fig. 3A－3D). The ROS levels in normotensive mice were similar with or without genistein administration.

Fig. 3.

Effects of genistein (Gen) on reactive oxygen species (ROS) overproduction in the vascular wall and en face endothelium in angiotensin II (Ang II)-induced hypertensive mice. (A) Representative dihydroethidium (DHE) fluorescence images and (B) summarized relative fluorescence value indicating ROS in the aortic vascular wall. (C) Representative en face endothelial DHE fluorescence images and (D) summarized relative fluorescence value of endothelial ROS detected by DHE staining. (E) Serum total nitric oxide (NO) metabolites (NOx; nitrite+nitrate) as an integrated index of systemic NO bioavailability. All data are presented as the mean±standard error of the mean, n=5 per group. Statistical analysis was performed using one-way analysis of variance with Tukey’s multiple comparisons test.

Because NO is short-lived in vivo, we quantified circulating NO metabolites (total NOx; nitrite+nitrate) in the serum as an integrated index of systemic NO availability. Serum NOx was markedly reduced by Ang II infusion (84.23 nmol/mL) compared with that of the control group (152.03 nmol/mL). Genistein cotreatment restored serum NOx levels toward the control group levels (Ang II+Genistein: 147.70 nmol/mL), whereas the Genistein group exhibited intermediate values (122.72 nmol/mL). Overall, the Ang II group differed significantly from all other groups, whereas the control, Genistein, and Ang II+Genistein groups did not differ significantly from each other (Fig. 3E).

Genistein directly counteracts Ang II-induced vascular dysfunction and oxidative stress in isolated aortic rings

To distinguish the direct vascular actions from systemic effects, we exposed aortic rings to Ang II ex vivo (24 h) with or without genistein (1 or 10 µM). Genistein co-exposure attenuated Ang II-induced hypercontraction and restored ACh-mediated EDR across the tested concentrations (1 and 10 µM), with no notable interdose differences (Fig. 4A and 4B). SNP-mediated relaxation in the presence of L-NAME was unchanged (Fig. 4C). Consistent with in vivo results, cotreatment with 1 or 10 µM genistein reduced Ang II-induced ROS overproduction in the aortic wall (Fig. 4D and 4E). The higher concentration demonstrated marginally enhanced ROS suppression efficacy.

Fig. 4.

Ex vivo effects of genistein (Gen) against angiotensin II (Ang II)-induced vascular dysfunction in isolated mouse aorta. (A) Phenylephrine (Phe)-induced vascular contraction of cultured mouse aorta exposed to Gen w/o Ang II for 24 h. Data are presented as the mean±standard error of the mean (SEM), n=5 per group. Statistical analysis for the contraction curve was performed using two-way analysis of variance with Tukey’s multiple comparisons test. Adjusted P-values were indicated as follows: Ang II vs. control, P<0.001; Ang II+Gen 1 µM vs. Ang II, P=0.012; and Ang II+Gen 10 µM vs. Ang II, P=0.007. (B) Acetylcholine (ACh)-induced endothelium-dependent relaxation of cultured mouse aorta exposed to Gen w/o Ang II for 24 h. Data are presented as the mean±SEM, n=5 per group. Statistical analysis for the relaxation curve was performed using two-way analysis of variance with Tukey’s multiple comparisons test. Adjusted P-values were indicated: Ang II vs. control, P<0.001; Ang II+Gen 1 µM vs. Ang II, P<0.001; Ang II+Gen 1 µM vs. control, P=0.020; Ang II+Gen 10 µM vs. Ang II, P<0.001; and Ang II+Gen 10 µM vs. control, P<0.001. (C) Sodium nitroprusside (SNP)-induced endothelium-independent relaxation of the mouse aorta in the presence of N^ω-nitro-L-arginine methyl ester. (D) Representative dihydroethidium fluorescence images and (E) summarized relative fluorescence value indicating the reactive oxygen species level in the aortic vascular wall. Data are presented as the mean±SEM, n=3 per group. Statistical analysis was performed using one-way analysis of variance with Tukey’s multiple comparisons test.

Genistein restores NO bioavailability and lowers intracellular ROS in Ang II-challenged endothelial cells

HUVECs were exposed to Ang II to model endothelial dysfunction and then treated with genistein (1 or 10 µM). Endothelial NO bioavailability was quantified using DAF-FM diacetate after stimulation with the calcium ionophore A23187 (1 µM), and then the intracellular ROS levels were measured using the DCFH-DA assay. Ang II significantly reduced A23187-induced NO fluorescence relative to the control, whereas genistein restored NO toward control levels at both concentrations (Fig. 5A and 5B). Conversely, Ang II elevated DCFH-DA fluorescence, indicating enhanced ROS production, and genistein treatment lowered the ROS levels compared with Ang II at both concentrations (Fig. 5C and 5D). ROS production was lower in the 10 µM dose than in the 1 µM dose, although this difference was not significant. RT-qPCR analysis further indicated transcriptional remodeling of the endothelial redox-NO axis. Ang II increased the mRNA expression levels of NADPH oxidase 4 (NOX4) and nitric oxide synthase 2 (NOS2) and suppressed those of nitric oxide synthase 3 (NOS3). Genistein displayed concentration-dependent effects on NOX4 and NOS3. Specifically, 1 µM modestly attenuated NOX4 induction but did not fully normalize it and did not rescue NOS3, whereas 10 µM suppressed NOX4 and restored NOS3 toward control levels. In contrast, Ang II-induced NOS2 upregulation was normalized by genistein at 1 and 10 µM (Fig. 5E－5H). These transcriptional changes are consistent with the observed reduction in ROS and restoration of NO bioavailability. To provide a broader view of the transcriptional responses, we profiled the antioxidant defense genes [superoxide dismutase 1 (SOD1), superoxide dismutase 2 (SOD2), and glyoxalase 1 (GLO1)]; inflammatory/redox regulatory genes [nuclear factor erythroid 2-related factor 2 (NRF2) and nuclear factor kappa B (NF-κB)]; and selected cytokines/chemokines [interleukin 6 (IL-6), C-C motif chemokine ligand 5 (CCL5), C-C motif chemokine ligand 2 (CCL2), and transforming growth factor beta 1 (TGF-β1)]. None of these genes exhibited significant differences among the groups (Supplementary Fig. 1). The expression of intercellular adhesion molecule 1 (ICAM1) was modestly increased in the Ang II+Genistein 10 µM group compared with the control. In contrast, the expression of interleukin 1 beta (IL-1β) was elevated in the Ang II-treated groups, with no further modulation by genistein administration (Supplementary Fig. 1).

Fig. 5.

Effects of genistein (Gen) on nitric oxide (NO) production and reactive oxygen species (ROS) generation in angiotensin II (Ang II)-exposed endothelial cells in vitro. (A) Representative 4-amino-5-methylamino-2’,7’-difluorofluorescein diacetate (DAF-FM diacetate) fluorescent images and (B) summarized relative fluorescence value of NO production evoked by calcimycin (A-23187, 1 µM) in human umbilical vein endothelial cells (HUVECs). (C) Representative 2’,7’-dichlorodihydrofluorescein diacetate (DCFH-DA) images and (D) summarized data of intracellular ROS production in HUVECs. (E) Heatmap showing the relative mRNA expression of redox-NO-related genes in HUVECs under the indicated treatments. (F) Relative NOX4 mRNA expression in HUVECs under the indicated treatments [control, Ang II, Ang II+Gen 1 µM, and Ang II+Gen 10 µM]. (G) Relative NOS3 mRNA expression in HUVECs under the indicated treatments. (H) Relative NOS2 mRNA expression in HUVECs under the indicated treatments. mRNA levels were normalized to an internal reference gene and expressed relative to those of the control group. Data are presented as the mean±standard error of the mean, n=4 per group. Statistical analysis was performed using one-way analysis of variance with Tukey’s multiple comparisons test.

DISCUSSION

Using in vivo, ex vivo, and cellular models, we demonstrate that genistein ameliorates Ang II-induced hypertension predominantly via an endothelial mechanism. In vivo, genistein lowered SBP/DBP and reduced aortic wall thickening while restoring EDR without affecting SNP endothelium-independent relaxation. Consistent with impaired NO signaling in Ang II hypertension, Ang II markedly reduced circulating NO metabolites (total NOx), whereas cotreatment with genistein restored serum Nox levels toward control levels, thereby providing in vivo biochemical support for improved systemic NO availability. Ex vivo, genistein directly corrected the Ang II-induced hypercontractility and EDR impairment in aortic rings. In HUVECs, genistein reduced intracellular ROS generation and restored A23187-induced NO production. Consistently, endothelial RT-qPCR provided additional molecular validation for this redox-NO phenotype observed in our functional assays. Ang II enhanced the mRNA expression levels of NOX4 and NOS2 while suppressing those of NOS3, whereas genistein normalized NOX4/NOS2 and restored NOS3 expression in a concentration-dependent manner. This parallels the reduced intracellular ROS and recovering agonist-induced NO bioavailability observed in HUVECs. Together, these data triangulate the improved redox-NO coupling in the endothelium as the proximate driver of genistein benefit. Each model adds a distinct mechanistic layer: the in vivo model establishes physiological relevance and localizes ROS suppression to the intact endothelium; the ex vivo ring model demonstrates vessel-intrinsic protection independent of systemic effects; and the endothelial cell model links reduced ROS production to the restoration of NO signaling, accompanied by redox-NO transcriptional remodeling. These results support a predominantly endothelial mechanism in which genistein restores redox-NO homeostasis. Specifically, it improves endothelium-dependent relaxation while preserving endothelium-independent responses to SNP. Concurrently, it reduces ROS signals while restoring NO bioavailability. A schematic summary of this endothelial redox-NO mechanism is provided in Fig. 6.

Fig. 6.

Schematic summary of genistein-mediated protection against angiotensin II (Ang II)-induced endothelial dysfunction and hypertension. Ang II disrupts endothelial redox-nitric oxide (NO) coupling, which is characterized by increased reactive oxygen species (ROS) production and reduced NO signaling, accompanied by the transcriptional remodeling of key redox-NO pathway genes in endothelial cells [nicotinamide adenine dinucleotide phosphate oxidase 4 (NOX4)↑, nitric oxide synthase 2 (NOS2)↑, and nitric oxide synthase 3 (NOS3)↓]. This redox imbalance impairs NO-dependent endothelium-dependent relaxation [reduced acetylcholine-mediated endothelium-dependent relaxation with preserved sodium nitroprusside (SNP) responses] and contributes to vascular dysfunction and hypertension. Genistein counteracts Ang II-evoked injury by limiting ROS, restoring NO availability [supported by endothelial NO readouts and circulating nitric oxide metabolites (NOx)], and normalizing the redox-NO transcript pattern (NOX4↓, NOS2↓, and NOS3↑/toward control), thereby improving endothelial function and vascular protection. The schematic summarizes the mechanisms supported by the present study and does not imply the direct identification of an upstream molecular target.

However, because we did not directly quantify the protein abundance or post-translational regulation (e.g., eNOS phosphorylation at Ser1177/1179), the BH4/BH2 ratio, eNOS uncoupling, or NADPH oxidase enzymatic activity, we interpret these data as functional rescue accompanied by transcriptional modulation rather than direct evidence of altered enzyme activity. Future work incorporating protein-level and activity assays will be important for defining the precise biochemical steps targeted by genistein and for establishing causality across the NOS/NOX redox network.

The mechanistic framework that integrates our findings with previous studies regarding hypertension is based on the disturbed redox-NO axis. In Ang II hypertension, vascular NADPH oxidases generate superoxide that rapidly reacts with the eNOS-derived NO, thereby forming peroxynitrite, which oxidizes tetrahydrobiopterin and promotes eNOS uncoupling, thus depleting NO bioavailability and reinforcing oxidative endothelial injury (Schulz et al., 2008; Förstermann and Li, 2011; Tejero et al., 2019; Griendling et al., 2021). Our observation that genistein reduces DHE-detected superoxide signals in the vascular wall and en face endothelium, restores ACh-EDR, and leaves the SNP responses unchanged is consistent with a primary action that modulates endothelial NO bioavailability and reduces ROS generation. Genetic studies further support the causal role of the oxidative stress-NO axis in hypertensive endothelial dysfunction; for example, the endothelial-specific deletion of SMAD4 ameliorates vascular dysfunction in hypertensive mice, which is accompanied by reduced oxidative stress and improved NO-dependent endothelial signaling (Yang et al., 2025). Previous reports on classic hypertensive models have indicated that genistein increases eNOS expression/activity, modulates caveolin/calmodulin dynamics, and suppresses NADPH-oxidase-derived superoxide (Vera et al., 2007; Si and Liu, 2008; Lin et al., 2011). Our study extends those observations to an Ang II-driven injury model using a cross-scale design that ties endothelial biochemistry to conduit-artery function and systemic hemodynamics.

Pharmacokinetic studies support the relevance of the low µM range used in our ex vivo and in vitro experiments. In humans, circulating genistein after soy food intake or isoflavone-containing supplements is typically reported in the sub- to low µM range, with most genistein present as glucuronide/sulfate conjugates (Gooderham et al., 1996; Anupongsanugool et al., 2005). In a controlled feeding study using a genistein-rich soy protein isolate, plasma genistein reached ∼0.9 µM (907 nmol/L). In mice, oral dosing models yield total circulating genistein concentrations in the low µM range (e.g., >3 µM after 50 mg/kg), thereby providing contextual support that low µM exposure is biologically relevant in vivo (Cimafranca et al., 2010). Accordingly, we selected 1 µM for ex vivo/in vitro experiments to approximate the exposure-aligned concentrations. We included 10 µM as a commonly used upper-bound concentration in the endothelial mechanistic studies to assess concentration dependence. Notably, 10 µM tended to produce more complete normalization of the redox-NO transcriptional responses (e.g., NOX4 suppression and NOS3 restoration), which is consistent with a concentration-response component in our endothelial results. Because circulating genistein in vivo is largely conjugated, our cell-based experiments used aglycone as a control of endothelial responsiveness; therefore, these concentrations should be interpreted as exposure-informed mechanistic tools rather than direct equivalents of free plasma genistein. Although circulating genistein is largely present as glucuronide/sulfate conjugates in the serum, the direct reversal of dysfunction under aglycone exposure and previous evidence that conjugates may be locally deconjugated at the inflamed endothelium argue that exposure-relevant free compounds can function at the vascular interface. We deliberately avoided over-interpretation and note that metabolite-resolved pharmacokinetic studies, with measurements of circulating conjugates, free aglycone, and vascular tissue levels, will be required to more precisely define the exposure-response relationships.

To address whether the observed vascular protection reflects systemic exposure vs. endothelial delivery, we performed untargeted LC-MS serum metabolomic profiling at the study endpoint. This analysis confirmed the presence of genistein and several genistein-derived metabolites in the serum of genistein-treated mice, including genistein, genistein 4’-O-glucuronide, genistein sulfate, and 8-hydroxygenistein (Fig. 2E). Notably, 8-hydroxygenistein was significantly increased in the genistein-treated mice compared with the control and Ang II groups, thereby providing model-matched biochemical evidence of systemic genistein exposure under our experimental conditions. Because these data are semiquantitative (relative abundance) and do not resolve free vs. conjugated fractions or vascular tissue levels, we do not claim direct endothelial delivery from metabolomics alone. Nevertheless, several functional observations argue for an endothelial site of action. Specifically, genistein selectively restored EDRs, directly rescued Ang II-induced dysfunction in isolated aortic rings ex vivo (independent of systemic factors), and restored agonist-induced NO signals while reducing ROS generation in cultured endothelial cells at exposure-relevant concentrations. Together, these findings support a model in which systemic exposure to genistein, predominantly present as metabolized/conjugated forms in circulation, translates into endothelial protection. We acknowledge that targeted LC-MS/MS pharmacokinetic profiling (including free aglycone and conjugates) and vascular tissue quantification are required to more precisely define endothelial exposure and exposure-response relationships.

The present results also intersect with receptor-proximal signaling that governs Ang II pressor and oxidative responses. Ang II signals promote vasoconstriction, oxidative stress, and remodeling primarily via the angiotensin II type 1 receptor (AT1R), whereas the angiotensin II type 2 receptor exerts counter-regulatory effects (Czepiel et al., 2022; Colin et al., 2023; Okuno et al., 2023). Crosstalk between AT1R and cannabinoid receptor type 1 (CB1R) has been implicated in the modulation of pressor responses and endothelial oxidative stress, and the inhibition of CB1R has been reported to downregulate vascular AT1R and improve endothelial function (Tiyerili et al., 2010; Miklós et al., 2021; Mińczuk et al., 2022; Bo et al., 2024). Recent reports have further verified genistein as a neutral CB1R antagonist with vascular protective effects and limited central nervous system infiltration (Wei et al., 2022). In parallel, genistein is a pleiotropic phytoestrogen that interacts with estrogen receptors (ERα/ESR1 and ERβ/ESR2), and endothelial estrogen receptor signaling is generally considered vasculoprotective, in part by promoting eNOS/NOS3-related signaling and modulating oxidative stress (Kuiper et al., 1998; Chambliss et al., 2010). In addition to classical estrogen receptors, G protein-coupled estrogen receptor 1 (GPER1) has been linked to rapid endothelial signaling that enhances NO bioavailability and vasodilatory function (Meyer et al., 2014). Moreover, genistein has been reported to engage with metabolic/nuclear signaling nodes such as PPARγ, which can exert anti-inflammatory and antioxidant effects on the vascular endothelium and support redox-NO homeostasis (Dang et al., 2003; Xu et al., 2004; Si and Liu, 2007). Conversely, the transactivation of epidermal growth factor receptor is frequently implicated in Ang II-driven ROS generation and endothelial dysfunction, which provides another plausible upstream interface through which genistein could mitigate oxidative injury (Gekle et al., 2023). Importantly, the present study was not designed to establish receptor-level target engagement or necessity (e.g., pharmacological antagonism, genetic perturbation, or binding assays); therefore, these upstream pathways are discussed as literature-supported, testable hypotheses that may be involved in the redox-NO axis observed in our models.

As a phytoestrogen, genistein can engage estrogen receptor-dependent signaling, thus providing an additional route of eNOS activation and endothelial protection (Vera et al., 2007; Si and Liu, 2008; Al-Nakkash et al., 2012). Reports of sex-dimorphic vascular effects-benefits in male hypertensive models under high-salt load and selective blood pressure reductions in female mice at higher dietary doses highlight the importance of explicitly incorporating sex as a biological variable and investigating estrogen receptor pathways in future studies (Montenegro et al., 2009; Al-Nakkash et al., 2012). Our experiments were performed in male mice; therefore, extending these findings to females and to estrogen-deficient states (e.g., ovariectomy or menopause) is a translational priority, particularly considering the clinical reports that purified genistein may improve endothelial function in postmenopausal women (Squadrito et al., 2003; Irace et al., 2013; Squadrito et al., 2013; De Gregorio et al., 2017; Abshirini et al., 2020).

From a translational perspective, endothelial function is a clinically meaningful intermediate phenotype. Brachial FMD can predict future cardiovascular events and provides a noninvasive indication of endothelial health (Green et al., 2011; Vita, 2011; Ras et al., 2013). Small, randomized trials have suggested that genistein or isoflavone-rich matrices improve FMD and may modestly lower blood pressure, although the results exhibit heterogeneity by dose, formulation, and baseline vascular status (Squadrito et al., 2002, 2003; Irace et al., 2013). Meta-analyses of soy isoflavones, including genistein, have demonstrated modest-to-average reductions in SBP (typically ∼2 mmHg) with heterogeneity by population and product matrix, thus aligning with an endothelial-targeted, small-effect model (Taku et al., 2010; Liu et al., 2012; Lei et al., 2024; Feng et al., 2025). Our multiscale data offer a mechanistic link from endothelial redox-NO biology to conduit-artery function and systemic hemodynamics in a clinically relevant Ang II model, thereby supporting genistein as a candidate adjunct to guideline-directed therapy for residual endothelial dysfunction in hypertension. Carefully designed human studies should prespecify the endothelial endpoints (e.g., FMD or reactive hyperemia), stratify by sex and menopausal status, consider background antihypertensive regimens, and incorporate biomarker panels that index the eNOS coupled state and redox status.

Because genistein is a dietary isoflavone that is typically consumed through habitual, long-term intake or supplementation, we initiated genistein administration before Ang II infusion to reflect a preventive supplementation model rather than an acute pharmacological rescue strategy. This timing was mechanistically motivated, as endothelial nitro-oxidative stress and dysfunction can occur early during Ang II exposure and may contribute to subsequent vascular pathology (Wattanapitayakul et al., 2000). Therefore, pretreatment allowed us to evaluate whether genistein preserves endothelial redox-NO coupling at the initiation phase of Ang II-induced injury. Although our study emphasizes prevention, whether genistein can reverse established hypertension when initiated after blood pressure elevation has been established remains an important question for future work (Ma et al., 2020; Feng et al., 2025).

In conclusion, this study provides strong evidence that genistein, as a legume-derived bioactive compound, offers protective effects against cardiovascular diseases through mechanisms that enhance NO bioavailability and suppress oxidative stress. These findings identify genistein as a dietary-derived small molecule with therapeutic potential against vascular dysfunction and hypertension.

SUPPLEMENTARY MATERIALS

Supplementary materials can be found via https://doi.org/10.3746/pnf.2025.290