The acute effects of exercise intensity and inorganic nitrate supplementation on vascular health in females after menopause

Abstract

Menopause is associated with reduced nitric oxide bioavailability and vascular function. Although exercise is known to improve vascular function, this is blunted in estrogen-deficient females post-menopause (PM). Here, we examined the effects of acute exercise at differing intensities with and without inorganic nitrate (NO₃⁻) supplementation on vascular function in females PM. Participants were tested in a double-blinded, block-randomized design, consuming ∼13 mmol NO₃⁻ in the form of beetroot juice (BRJ; n = 12) or placebo (PL; n = 12) for 2 days before experimental visits and 2 h before testing. Visits consisted of vascular health measures before (time point 0) and every 30 min after (time points 60, 90, 120, 150, and 180) calorically matched high-intensity exercise (HIE), moderate-intensity exercise (MIE), and a nonexercise control (CON). Blood was sampled at rest and 5-min postexercise for NO₃⁻, NO₂⁻, and ET-1. BRJ increased N-oxides and decreased ET-1 compared with PL, findings which were unchanged after experimental conditions (P < 0.05). BRJ improved peak Δflow-mediated dilation (FMD) compared with PL (P < 0.05), defined as the largest ΔFMD for each individual participant across all time points. FMD across time revealed an improvement (P = 0.05) in FMD between BRJ + HIE versus BRJ + CON, while BRJ + MIE had medium effects compared with BRJ + CON. In conclusion, NO₃⁻ supplementation combined with HIE improved FMD in postmenopausal females. NO₃⁻ supplementation combined with MIE may offer an alternative to those unwilling to perform HIE. Future studies should test whether long-term exercise training at high intensities with NO₃⁻ supplementation can enhance vascular health in females PM.

NEW & NOTEWORTHY This study compared exercise-induced changes in flow-mediated dilation after acute moderate- and high-intensity exercise in females postmenopause supplementing either inorganic nitrate (beetroot juice) or placebo. BRJ improved peak ΔFMD postexercise, and BRJ + HIE increased FMD measured as FMD over time. Neither PL + MIE nor PL + HIE improved FMD. These findings suggest that inorganic nitrate supplementation combined with high-intensity exercise may benefit vascular health in females PM.

INTRODUCTION

Approximately 1.2 billion females will be postmenopausal by 2030 (1). Menopause involves the loss of endogenous estradiol, which greatly increases metabolic and cardiovascular disease (CVD) risk (2). Specifically, the postmenopausal state (PM) is associated with reduced bioavailable nitric oxide (NO), which can impact vascular function (3) and CVD risk via its impact on vasodilation, blood flow regulation, glucose uptake, and platelet function (4, 5).

Estrogen increases bioavailable NO by promoting endothelial NO synthesis via both genomic and nongenomic pathways and may also reduce NO consumption via increasing antioxidant status and reducing inflammation (5). Estradiol treatment has been shown to improve NO bioavailability and vascular health in females PM, however not to the levels observed in the premenopausal state (6–8). In addition, estradiol treatment when combined with moderate-intensity exercise (MIE) training appears to lower CVD risk in females PM (9). However, the use of estradiol treatment within females PM is not necessarily prescribed following data from the Women’s Health Initiative showing a potentially increased risk of blood clots, stroke, and breast cancer, without a reduction in the risk of CVD (10–13). Consequently, novel interventional approaches that increase NO bioavailability and promote vascular health in females PM are paramount.

Exercise training increases NO bioavailability and improves vascular health in a variety of populations (14, 15). However, the cardiometabolic benefits of exercise training in estrogen-deficient females PM are equivocal when compared with age-matched males and premenopausal females (12, 16, 17). This is partly due to confounding study differences such as length of PM status before training, levels of visceral fat, metabolic/vascular impairment at baseline, and training intensity/duration (7, 12, 17–28). A viable approach to exercise training in PM may be to use exercise intensities above the lactate threshold, as this high-intensity exercise (HIE) is known to induce greater vascular adaptations (15, 29, 30). In addition, because the acute exercise-induced alterations to vascular function appear to predict the chronic exercise-induced adaptations to vascular function (31), exploring factors that acutely impact vascular health in females PM may provide insight into viable factors to aid long-term training adaptations. Importantly, our laboratory has previously shown that acute HIE improves FMD more than MIE (32).

Another novel and innovative approach is to increase bioavailable NO through the conversion of inorganic nitrate (NO₃⁻) and nitrite (NO₂⁻) anions. This method is attractive as it is biologically distinct from endothelial-NO synthase (eNOS) and can be achieved easily via oral beetroot juice (BRJ) administration (33). Briefly, NO₃⁻ is swallowed, absorbed into the circulation, concentrated in the salivary glands, and then resecreted into the oral cavity, where commensal bacteria reduce NO₃⁻ to NO₂⁻ (34). This NO₂⁻ is then swallowed and absorbed into the circulation where it can be reduced to NO. BRJ supplementation improves vascular health (35, 36) and exercise tolerance (37) in some clinical populations. Combining exercise with exogenous NO supplementation may optimize the beneficial effects of each treatment individually but remains to be tested in females PM.

Thus, the purpose of the present study was to determine whether acute calorically matched HIE improves vascular health more than acute MIE in females PM and whether NO₃⁻ supplementation impacts these responses. The primary hypothesis was that HIE would improve brachial artery flow-mediated dilation (FMD) more than MIE and that NO₃⁻ supplementation would further enhance these improvements. A secondary hypothesis was that plasma NO₂⁻ concentration, serum ET-1 concentration, time since menopause (TSM), baseline FMD, fitness, and body composition [e.g., % body fat, abdominal visceral fat (AVF)] will affect exercise-induced changes in FMD.

METHODS

Experimental Design and Protocol

This study was a randomized, double-blind, placebo-controlled trial (NCT05221905). Following a screening visit, participants were randomized to one of two treatment arms: 1) ∼13 mmol of NO₃⁻ in the form of 140-mL beetroot juice (BRJ) and 2) or identical placebo (PL) with the nitrate extracted (<0.1 mmol nitrate), for 2 days before each testing visit, as well as 2-h before each study visits (see Supplementation Protocol). Study visits were performed at least 48 h apart each, but no more than 2 wk apart. Each arm of the trial involved three randomized experimental visits that consisted of HIE, MIE, and a nonexercise control (CON) visit (Fig. 1). All study visits were performed by the same investigator, at the same time of day (∼8:00 AM arrival ± 2 h for between subjects, with time kept consistent within subject) and under the same sensory conditions (i.e., light, temperature, sound). Participants were encouraged to maintain physical activity and diet habits throughout the study period except for the avoidance of high nitrate foods during supplementation. Participants were recruited from the University of Virginia and surrounding Charlottesville, Virginia area. All procedures were approved by the Institutional Review Board at the University of Virginia, and the study was conducted in accordance with the Declaration of Helsinki. All participants provided written informed consent.

Figure 1.

Study schematic. BRJ, beetroot juice; n = 12. PL, placebo; n = 12. Image created with a licensed version of BioRender.com.

Participants

A total of 24 (12 per treatment arm) estrogen-deficient females PM were included in this parallel-arm trial. PM was defined as not having had a menstrual cycle for at least 1 yr. Of note, n = 12 per treatment arm for all conditions except for BRJ CON (n = 11) due to participant dropout. Participants were nonsmokers, sedentary or recreationally active (<3 days/wk of exercise), with no use of hormone replacement therapy in the past year, and participants could not have had a hysterectomy or oophorectomy. Participants were not taking any medications that might interfere with NO₃⁻ supplementation (e.g., nitrates, proton pump inhibitors, H₂ blockers).

Screening Procedures

Participants were phone screened for eligibility. Before all testing, participants abstained from all food or drinks other than water for at least 6 h, caffeine for at least 12 h, and exercise or alcohol for at least 24 h. Participants met at the Clinical Research Unit (CRU) of the University of Virginia School of Medicine and were consented, baseline measures taken (blood pressure, height, weight), and they were screened by a study physician who determined TSM. After enrollment, a venous catheter was placed in the antecubital fossa for serial blood lactate sampling during rest and exercise. Blood was taken at rest for baseline measures of plasma NO₃⁻, NO₂⁻, and serum ET-1. Participants performed an incremental symptom-limited cycle ergometer exercise test at the Exercise Physiology Core Laboratory. Open-circuit spirometry was used to measure respiratory gases (Viasys Vmax Encore, Yorba Linda, CA). The exercise test consisted of 3-min, 20-W stages until participants reached volitional fatigue. The criteria for achieving V̇o_2peak specific for older females included maximum HR within 10 beats/min of age-predicted HR-max, respiratory exchange ratio (RER) > 1.10, and a rating of perceived exertion (RPE) > 17 (38). All included participants met criteria for V̇o_2peak. Blood lactate was sampled in the last 30 s of each stage, analyzed using a YSI 2300 STAT Plus (Yellow Springs, OH), and the resultant data were plotted to determine the lactate threshold (LT) (39).

On a separate day (at least 48 h later), participants completed a DEXA scan (Hologic Horizon) for the assessment of percent body fat and AVF and a baseline vascular testing visit that consisted of measures of blood pressure (BP), pulse wave velocity (PWV), and brachial artery flow-mediated dilation (FMD) before beetroot juice supplementation. This visit started with participants resting in a supine position in a dark, quiet thermoneutral room for 10 min. BP and PWV testing were completed with a SphygmoCor Xcel model (AtCor Medical, Itaska, IL) following manufacturer instructions. FMD was performed according to established guidelines (40, 41). Briefly, participants laid in a supine position with their arm extended and with their hand supinated, while an automatic inflation blood pressure cuff (AI6 Arterial Inflow System, Hokanson Inc., Bellevue, WA) was placed on the forearm directly below the elbow. Brachial artery diameter and blood velocity were recorded at rest with a high-resolution 7.5-MHz linear array transducer using Doppler ultrasound (uSmart 3300 Ultrasound System, Terason, Burlington, MA). A probe holder was used to stabilize the transducer, and both anatomical notes and images of the artery were taken to ensure accurate repeat testing placement. An ECG trigger was used to capture images during end diastole of the cardiac cycle. After resting measures, the cuff was inflated to 200 mmHg for 5 min while the participant was encouraged to remain as still and relaxed as possible. Image capture was resumed during the final 10 s of the occlusion period, and measures continued for 2 min after cuff deflation. Automated edge detection software (Medical Imaging Applications, Coralville, IA) was used to evaluate diameter measurements. All ultrasound images were coded to ensure blinding for analysis.

Supplementation Protocol

Participants were block-randomized (matched for baseline characteristics of FMD, V̇o_2peak, and age) to one of two treatment arms. Participants were assigned to either consume ∼13 mmol of NO₃⁻ daily via two 70-mL bottles (morning and night) of nitrate-rich beetroot juice (BRJ), or a nitrate-depleted beetroot juice (PL) (<0.1 mmol) with identical taste and appearance (Beet It, James White Drinks Ltd., Ipswich, UK). The participants and study personnel were blinded to the beverage allocation. Participants were given instructions to consume the beverages for 2 days before study visits and to consume two bottles 2 h before the beginning of each study visit to account for the time course of nitrate and nitrite bioavailability (42, 43). Participants were provided with a form to log supplementation times and were instructed to bring back the empty bottles at each study visit. Participants were instructed to avoid high nitrate foods (i.e., spinach, arugula, celery, etc.) and grapefruit throughout the study period (44), as well as to avoid factors that could impact oral and microbial environment, such as mouthwash or antibiotics (45–47).

Experimental Visits

Each treatment arm of the study (BRJ or PL) consisted of three experimental visits; HIE, MIE, and a CON visit, assigned in a randomized order. Upon arrival to the laboratory, participants lay supine for 5 min before undergoing a resting blood draw. Participants then underwent a repeat of the vascular tests administered during the screening visit (BP, PWV, and FMD – time point 0 min). Following vascular testing, participants were unblinded to the experimental visit allocated to that day (HIE, MIE, or CON).

Exercise sessions consisted of calorically matched cycle ergometer (Lode Excalibur Sport, Groningen, The Netherlands) exercise at either HIE or MIE, with the goal of reaching 200 kcal of energy expenditure. Gas exchange and heart rate were recorded continuously during the exercise sessions (Cosmed Quark, Cosmed USA Inc., Concord, CA). Time to expend 200 kcal was estimated based on the V̇o_2peak test, with real-time adjustments made if energy during the exercise sessions was expended at a rate other than expected.

• 1) HIE occurred at a power output associated with the power output that was 75% of the difference between LT and V̇o_2peak (75% Δ). If participants could not maintain this power output until 200 kcal was expended, power output was decreased to 50% Δ and then to 25% Δ if necessary.
• 2) MIE occurred at the power output associated with LT.
• 3) CON visits consisted of passive rest.

After each experimental condition and 5 min of supine rest, participants underwent a postcondition blood draw to determine changes in blood markers from pre- to postexperimental condition. Within each experimental condition (HIE, MIE, or CON), FMD for each time point was quantified (i.e., 0, 60, 90, etc.). Peak ΔFMD within each condition was determined as the largest ΔFMD for each individual participant across all time points.

Blood Measures

Blood on experimental visits was obtained upon study arrival (i.e., 2 h after final supplementation) and ∼5-min postexercise via antecubital venipuncture. Blood was drawn into a NO₃⁻-free syringe (BD Luer-Lok) (∼5 mL) for subsequent (end of study) N-oxide analysis. Plasma NO₃⁻ and NO₂⁻ were assessed via ozone-based chemiluminescence using a Sievers NOA model 280i (GE Analytical Instruments, Boulder, CO) as previously described (48). Briefly, plasma samples for NO₃⁻ analysis were deproteinized using cold ethanol precipitation in a 1:3 dilution (plasma:ethanol) followed by a 30-min incubation before being centrifuged at 14,000 g for 10 min. The supernatant was removed for the subsequent NO₃⁻ analysis in the presence of vanadium chloride in hydrochloric acid at 95°C. The NO₂⁻ of the undiluted (nondeproteinized) plasma samples was determined by its reduction to NO in the presence of glacial acetic acid and potassium iodide as previously explained (49). Serum was also obtained at all time points for determination of endothelin-1, according to manufacturer’s instruction (CV: 7%; DET100, Endothelin‐1 Quantikine ELISA Kit, R&D Systems, Minneapolis, MN). Blood analyses were not possible in all markers and time points due to one participant dropout (n = 11 for BRJ CON), failed blood draw attempt, or sample contamination. Final counts for each blood marker are as follows: plasma nitrate (PL: n = 32; BRJ: n = 28), plasma nitrite (PL: n = 35; BRJ: n = 33), serum ET-1 (PL: n = 29; BRJ: n = 30).

Statistics

Unpaired t tests were used to determine differences between participant characteristics at baseline. The highest change in FMD from pre- to postexperimental condition (Peak ΔFMD) was determined for each participant. Two-way ANOVA was used to determine any differences between time and experimental conditions (i.e., exercise intensity and treatment combinations) for peak ΔFMD and for FMD changes over time (condition × time). Two-way ANCOVA with baseline artery diameter was also used to determine differences between time and experimental conditions. One-way ANOVA was used to determine any differences in FMD variables (i.e., baseline diameter, peak diameter, absolute FMD, and FMD %) over time within each condition. Simple linear regression models were used to examine the relationship between supplemented plasma nitrite, serum ET-1, percent body fat, V̇o_2peak, AVF, TSM, and baseline FMD with peak ΔFMD. All statistical analysis was conducted using GraphPad Prism Version 9.3 (GraphPad Software, La Jolla, CA). Statistical significance was determined a priori when P < 0.05, and Hedges g was used to account for lower sample sizes and was interpreted as having trivial effects (0.0–0.19), small effects (0.20–0.49), medium effects (0.50 – 0.79), and large effects (>0.80) (50). Data are presented as mean ± SD unless otherwise noted.

RESULTS

Participant Characteristics

A total of 24 postmenopausal females completed the study (12 per treatment arm). Treatment groups were matched for age, fitness (Table 1), and vascular health (Table 2). FMD, resting brachial artery diameter, PWV, MAP, or DBP were not different from screening compared with after 3 days of supplementation (i.e., visit 2 time point 0), however, SBP was significantly lower after both PL (P = 0.04) and BRJ (P = 0.01) supplementation (Table 2).

Table 1.

Baseline characteristics

Variable	PL	BRJ	P Value
n	12	12
Age, yr	61 ± 5	59 ± 5	0.42
Time since menopause, yr	11 ± 6	9 ± 5	0.30
Body mass index, kg/m2	27 ± 5	28 ± 7	0.68
Body fat, %	36 ± 7	38 ± 6	0.43
Abdominal visceral fat, g	380 ± 180	447 ± 205	0.21
V̇o2peak, mL/kg/min	24 ± 4	23 ± 5	0.55
V̇o2 at LT, mL/kg/min	14 ± 3	14 ± 3	0.40
Medication
β-Blocker	1/12	1/12
Statin	1/12	0/12
ACE inhibitor	0/12	1/12
Thyroid	3/12	4/12

Values are means ± SD. BRJ, beetroot juice; LT, lactate threshold; PL, placebo.

Table 2.

The effects of acute inorganic nitrate (BRJ) or placebo (PL) on resting vascular health

	Placebo (PL)			Beetroot Juice (BRJ)			PL vs. BRJ
Variable	Pre	Post	P Value	Pre	Post	P Value	Pre P Value
Flow-mediated dilation, %	6.1 ± 2.3	6.4 ± 2.0	0.74	5.8 ± 2.2	6.0 ± 1.8	0.84	0.70
Resting brachial artery diameter, mm	3.5 ± 0.5	3.5 ± 0.5	0.84	3.6 ± 0.3	3.5 ± 0.4	0.24	0.55
Pulse wave velocity, m/s	7.4 ± 1.1	7.0 ± 1.1	0.06	7.1 ± 0.7	6.8 ± 0.9	0.12	0.81
Mean arterial pressure, mmHg	93 ± 8	89 ± 5	0.08	91 ± 7	89 ± 8	0.27	0.64
Systolic blood pressure, mmHg	124 ± 12	119 ± 8	0.04	127 ± 12	120 ± 13	0.01	0.51
Diastolic blood pressure, mmHg	77 ± 8	75 ± 5	0.49	74 ± 7	74 ± 7	0.80	0.32

Means ± SD. Pre, baseline resting values before any supplementation; Post, baseline resting values on visit 2 (i.e., after 3 days of supplementation). Bold denotes significance at P ≤ 0.05.

Blood Markers

There were no differences between groups for baseline plasma NO₃⁻ (BRJ: 45 ± 23 µM; PL: 42 ± 16 µM; P = 0.75), plasma NO₂⁻ (BRJ: 123 ± 35 nM; PL: 114 ± 50 nM; P = 0.66), or serum ET-1 (BRJ: 1.61 ± 0.47 pg/mL; PL: 1.69 ± 0.83 pg/mL; P = 0.81). Following BRJ treatment preexperimental condition, plasma nitrate (NO₃⁻) and plasma nitrite (NO₂⁻) were significantly elevated, and serum ET-1 was significantly decreased compared with PL (Fig. 2, A–C). No blood markers changed from pre- to postexperimental condition (Fig. 2, A–C).

Figure 2.

The effects of inorganic nitrate (BRJ) or placebo (PL) on plasma nitrate (A), plasma nitrite (B), and serum endothelin-1 (ET-1; C) pre- and postexperimental condition (i.e., HIE, MIE, or CON). Data are means ± SD. ***P < 0.001. ****P < 0.0001. BRJ, beetroot juice; CON, control; HIE, high-intensity exercise; MIE, moderate-intensity exercise; PL, placebo.

Exercise Parameters

There were no differences between treatment groups for any of the exercise parameters during the prerandomization screening or V̇o_2peak testing (e.g., V̇o_2peak, peak power, heart rate, RER, RPE, blood lactate concentration, or % of V̇o_2peak where lactate threshold occurred; all P > 0.05). Data from the experimental exercise bouts are displayed in Table 3 showing similar energy expenditure between exercise conditions and treatment groups, whereas HIE resulted in higher V̇o₂, heart rate, RPE, and a shorter exercise duration than MIE for both treatment conditions (BRJ vs. PL) (P < 0.05).

Table 3.

Experimental exercise parameters

	Moderate Intensity		High Intensity
Variable	PL	BRJ	PL	BRJ
Power, W (% Powerpeak)*	53 ± 18 (39%)	66 ± 22 (48%)	113 ± 18 (85%)	120 ± 32 (87%)
V̇o2, mL/kg/min (% V̇o2peak)*	16.2 ± 3.8 (66%)	15.6 ± 2.6 (68%)	20.3 ± 4.5 (83%)	19.6 ± 4.5 (85%)
HR, beats/min (% HRpeak)*	107 ± 22 (66%)	115 ± 15 (71%)	147 ± 10 (90%)	142 ± 8 (89%)
Borg-RPE*	12 ± 1	12 ± 1	17 ± 1	17 ± 1
Duration, min*	37.2 ± 4.6	33.9 ± 7.8	29.2 ± 2.1	26.0 ± 4.6
Energy expenditure, kcal	198 ± 19	200 ± 14	200 ± 21	199 ± 17

Values are means ± SD; n = 12 per treatment arm. min, minutes; kcal, kilocalories; RPE, rating of perceived exertion; W, watts. *Significant difference between exercise intensities.

Peak ΔFMD

Analysis of peak ΔFMD revealed a significant treatment (BRJ vs. PL) (P = 0.02) and exercise intensity effect (P < 0.001), with a nonsignificant interaction (P = 0.11; Fig. 3). Plasma NO₂⁻ concentrations following supplementation of all participants (i.e., BRJ and PL) showed a positive correlation with peak ΔFMD (R² = 0.18; P = 0.04) and participants with a larger resting brachial artery diameter had a lower peak ΔFMD (R² = 0.21; P = 0.03). PWV had a negative correlation with peak ΔFMD (R² = 0.18; P = 0.04). No other factors were significant predictors of peak ΔFMD including body fat % (R² = 0.13; P = 0.08), abdominal visceral fat (R² = 0.003; P = 0.78), V̇o_2peak (R² = 0.05; P = 0.28), baseline FMD% (R² = 0.03; P = 0.39), or TSM (R² < 0.01; P = 0.77). In an exploratory analysis to determine whether NO₃⁻ dose relative to body weight may predict changes in FMD, a significant negative correlation was found between body weight and peak ΔFMD when looking at the BRJ group alone (n = 12; R² = 0.52; P = 0.008). Serum ET-1 was lower in the BRJ compared with PL group postexercise (BRJ: 1.00 pg/mL; PL: 1.34 pg/mL; P < 0.001, Fig. 2C), but neither ET-1 (R² = 0.05; P = 0.31) nor ΔET-1 predicted peak ΔFMD (R² = 0.05; P = 0.46).

Figure 3.

The effects of inorganic nitrate (BRJ) or placebo (PL) and exercise intensity on Peak Δ FMD. BRJ, beetroot juice; CON, control; HIE, high-intensity exercise; MIE, moderate-intensity exercise; PL, placebo. Data are means ± SD.

FMD within Each Exercise Condition

PL CON compared with PL MIE showed no effect of time (P = 0.82), condition (P = 0.95), or an interaction (P = 0.61) (Fig. 4A). BRJ CON compared with BRJ MIE showed no effect of time (P = 0.58), condition (P = 0.15), or an interaction (P = 0.46), however, there were medium effect sizes (Hedge’s g) for BRJ MIE improving FMD from 60 to 150 min postexercise (Fig. 4B). When comparing PL MIE to BRJ MIE, there was no significant effect of time (P = 0.49), condition (P = 0.45), or an interaction (P = 0.55) (Fig. 4C). PL CON compared with PL HIE showed no effect of time (P = 0.51), condition (P = 0.65), or an interaction (P = 0.45) (Fig. 4D). BRJ CON compared with BRJ HIE showed no effect of time (P = 0.27), or interaction (P = 0.10), but a significant treatment effect of condition (P = 0.05), with large Hedge’s g effect sizes with BRJ HIE improving FMD (Fig. 4E). Two-way ANCOVA using baseline diameter as a covariate revealed a statistically significant difference between BRJ + HIE versus BRJ + CON (P = 0.005). PL HIE compared with BRJ HIE showed a significant effect of time (P = 0.04), with no condition (P = 0.55) or interaction (P = 0.46) effects (Fig. 4F). PL CON compared with BRJ CON showed no effect of time (P = 0.74), treatment (P = 0.47), or an interaction (P = 0.96) (data not shown). Baseline diameter, peak diameter, absolute change in diameter (mm), and %FMD are listed in detail in Table 4. There were no statistical differences between any time points within any condition (P > 0.05) with BRJ + HIE approaching significance for absolute change in diameter over time (mm; P = 0.06). Contrary to BRJ + HIE versus BRJ + CON, two-way ANCOVA using baseline diameter as a covariate revealed no differences between any other conditions.

Figure 4.

The effects of inorganic nitrate (BRJ - black) or placebo (PL - white) and different experimental visits (Control - CON - circles; Moderate - MIE - squares; High - HIE - triangles) on flow-mediated dilation (FMD). A: placebo control (PL CON) versus placebo moderate intensity (PL MIE). B: PL CON versus PL high intensity (PL HIE). C: beetroot juice (BRJ) CON versus BRJ MIE. D: BRJ CON versus BRJ HIE. E: PL MIE versus BRJ MIE. F: PL HIE versus BRJ HIE. Significant effects of treatment between BRJ CON and BRJ HIE (E; *P = 0.05) and significant effects of time with HIE (F; #P = 0.04). Data are means ± SE. BRJ, beetroot juice; CON, control; HIE, high-intensity exercise; MIE, moderate-intensity exercise.

Table 4.

The effects of acute inorganic nitrate (BRJ) or placebo (PL) on flow-mediated dilation across experimental conditions and time points

	Time point (min)
Placebo	0	60	90	120	150	180	P Value
Control
Baseline, mm	3.51 ± 0.44	3.49 ± 0.49	3.68 ± 0.55	3.48 ± 0.42	3.63 ± 0.46	3.44 ± 0.53	0.13
Peak, mm	3.73 ± 0.44	3.70 ± 0.50	3.83 ± 0.55	3.70 ± 0.43	3.75 ± 0.48	3.66 ± 0.53	0.19
FMD, mm	0.22 ± 0.05	0.21 ± 0.09	0.25 ± 0.06	0.22 ± 0.06	0.23 ± 0.07	0.22 ± 0.05	0.46
FMD, %	6.31 ± 2.01	6.10 ± 2.68	6.87 ± 1.93	6.33 ± 2.01	6.42 ± 2.39	6.49 ± 1.98	0.41
Moderate intensity
Baseline, mm	3.54 ± 0.54	3.62 ± 0.63	3.60 ± 0.64	3.59 ± 0.67	3.51 ± 0.59	3.54 ± 0.60	0.33
Peak, mm	3.77 ± 0.55	3.86 ± 0.67	3.81 ± 0.59	3.80 ± 0.68	3.72 ± 0.57	3.74 ± 0.60	0.36
FMD, mm	0.23 ± 0.05	0.24 ± 0.07	0.23 ± 0.07	0.21 ± 0.05	0.22 ± 0.07	0.20 ± 0.06	0.53
FMD, %	6.54 ± 1.82	6.59 ± 1.55	6.65 ± 2.84	6.12 ± 2.01	6.48 ± 2.28	5.77 ± 2.36	0.69
High intensity
Baseline, mm	3.53 ± 0.60	3.65 ± 0.52	3.63 ± 0.51	3.57 ± 0.52	3.46 ± 0.49	3.54 ± 0.56	0.38
Peak, mm	3.75 ± 0.61	3.88 ± 0.57	3.85 ± 0.52	3.83 ± 0.52	3.69 ± 0.48	3.76 ± 0.56	0.39
FMD, mm	0.22 ± 0.05	0.23 ± 0.09	0.22 ± 0.05	0.25 ± 0.07	0.23 ± 0.04	0.23 ± 0.04	0.47
FMD, %	6.49 ± 2.01	6.41 ± 2.24	6.12 ± 1.59	7.27 ± 2.67	6.91 ± 2.13	6.71 ± 2.23	0.36

Beetroot Juice	0	60	90	120	150	180	P value
Control
Baseline, mm	3.67 ± 0.42	3.70 ± 0.46	3.65 ± 0.57	3.59 ± 0.51	3.76 ± 0.41	3.58 ± 0.47	0.22
Peak, mm	3.88 ± 0.43	3.90 ± 0.44	3.89 ± 0.51	3.78 ± 0.51	3.86 ± 0.41	3.78 ± 0.46	0.13
FMD, mm	0.21 ± 0.05	0.21 ± 0.08	0.21 ± 0.06	0.20 ± 0.05	0.21 ± 0.05	0.20 ± 0.08	0.66
FMD, %	5.83 ± 1.48	5.75 ± 2.28	5.95 ± 2.00	5.61 ± 1.58	5.57 ± 1.62	5.67 ± 2.22	0.69
Moderate intensity
Baseline, mm	3.60 ± 0.43	3.66 ± 0.48	3.61 ± 0.48	3.67 ± 0.46	3.68 ± 0.39	3.63 ± 0.45	0.82
Peak, mm	3.82 ± 0.43	3.92 ± 0.46	3.87 ± 0.48	3.91 ± 0.44	3.94 ± 0.39	3.86 ± 0.42	0.61
FMD, mm	0.21 ± 0.05	0.26 ± 0.05	0.25 ± 0.06	0.25 ± 0.07	0.26 ± 0.08	0.23 ± 0.09	0.16
FMD, %	5.91 ± 1.62	7.23 ± 1.87	7.12 ± 1.88	6.84 ± 2.41	7.23 ± 2.29	6.46 ± 2.85	0.23
High intensity
Baseline, mm	3.52 ± 0.45	3.60 ± 0.45	3.58 ± 0.53	3.61 ± 0.50	3.65 ± 0.34	3.60 ± 0.49	0.61
Peak, mm	3.74 ± 0.46	3.80 ± 0.47	3.87 ± 0.55	3.89 ± 0.48	3.93 ± 0.33	3.86 ± 0.48	0.26
FMD, mm	0.22 ± 0.07	0.20 ± 0.09	0.26 ± 0.06	0.28 ± 0.05	0.28 ± 0.08	0.27 ± 0.10	0.06
FMD, %	6.43 ± 2.12	5.56 ± 2.45	6.64 ± 2.80	8.03 ± 2.02	7.82 ± 2.43	7.72 ± 3.23	0.09

Values are means ± SD. FMD, flow-mediated dilation; mm, millimeters in diameter.

Blood Pressure and Pulse Wave Velocity

MAP (Fig. 5, A–C) significantly increased across time (P < 0.01) for all three experimental conditions (CON, MIE, and HIE), with no treatment effect (BRJ vs. PL), or interaction effects (P > 0.05). PWV (Fig. 5, D–F) significantly increased over time during the CON visit (P < 0.001), with no treatment (P = 0.22) or interaction effects (P = 0.43; Fig. 5D). During MIE and HIE (Fig. 5, E and F), PWV was not significantly altered by time, or treatment, and there was no interaction (all P > 0.05).

Figure 5.

The effects of inorganic nitrate (BRJ) or placebo (PL) on mean arterial pressure (MAP; A–C) and pulse wave velocity (PWV; D–F) during control (A and D), moderate-intensity exercise (B and E), and high-intensity exercise (C and F). A significant main effect of time was observed for MAP in control, moderate-intensity exercise, and high-intensity exercise (denoted by #). Only a significant main effect of time for PWV was found during control. Data are means ± SD. BRJ, beetroot juice.

DISCUSSION

To our knowledge, this is the first study to examine the effects of NO₃⁻ supplementation and different intensities of exercise on vascular health in females PM. The primary hypothesis was that HIE would improve FMD compared with MIE and CON and that NO₃⁻ supplementation would augment these responses. Our findings indicate that FMD increases postexercise, with NO₃⁻ supplementation augmenting acute postexercise peak ΔFMD in estrogen-deficient females PM. This augmentation appears greatest after high-intensity exercise (Fig. 3). When examining the effects of time within each experimental condition, NO₃⁻ supplementation had medium effects (Hedge’s g) on FMD after MIE (Fig. 4B) and statistically significant effects after HIE (Fig. 4E; P = 0.05; P = 0.005 with baseline diameter used as a covariate). HIE regardless of supplementation also significantly changed FMD across time (Fig. 4F). In agreement with our hypothesis, plasma NO₂⁻ levels correlated positively with peak ΔFMD. Finally, BRJ supplementation did not improve measures of vascular health beyond PL in any measure at rest (Table 2).

As changes in FMD postacute exercise have shown in a population of young, healthy males to predict the exercise training-induced changes in FMD (31), the present findings suggest that long-term exercise training at high intensities may provide a greater stimulus for vascular adaptation than moderate-intensity exercise in females PM and that the combination of BRJ with HIE may offer the greatest improvements in FMD. This is notable as a 1% improvement in FMD is considered clinically meaningful and is associated with a relative risk reduction of CVD events of ∼10% (51). Consuming an exogenous source of NO may counteract the decrease of endogenous NO that occurs secondary to the loss of estrogen in menopause.

Effects of Exercise Intensity on FMD

We hypothesized that HIE would produce greater changes to acute postexercise FMD responses because of the greater elevation of shear stress and overall larger stimulus that this exercise induces when compared with MIE. Supporting this, previous data from our laboratory in younger lean adults showed an improvement in FMD 2 h post-HIE (Δ3.2 ± 0.5%), which was largely sustained up to 4 h postexercise (32). However, others have shown in females PM, acute postexercise FMD was improved only ∼1% immediately after high-intensity interval training (HIIT) and returning to baseline 1 h later, with no changes in FMD following MIE (52). These findings are equivocal, as other studies have failed to show a significant increase in FMD after either moderate- or high-intensity exercise in females PM (52). While studies have reported that FMD after acute MIE increases FMD as high as ∼4.6% in females PM (which was not further enhanced by estradiol (53, 54)), long-term exercise training at moderate intensities has failed to improve FMD (12) unless baseline FMD is impaired (17, 21) or unless coadministration of estrogen is administered (7).

Similar to previous data (9), the present study showed no significant differences between PL CON and PL MIE when examining either peak ΔFMD (Fig. 3) or FMD across time (Fig. 4A). When examining whether higher intensities of exercise alone can induce greater improvements in FMD in females PM, PL HIE showed no added benefits to PL MIE when examining peak ΔFMD (Fig. 3), and only a small effect on FMD across time (Hedge’s g = 0.38 at 120 min) compared with PL CON (Fig. 4D). Thus, it appears that neither moderate-intensity nor high-intensity exercise alone provide a potent enough stimulus to induce acute vascular benefits in females PM. Females PM have a loss of eNOS associated with the loss of estrogen and estrogen receptors (ERs) (55). While exercise has been shown to improve eNOS (56), this improvement is known to be blunted in aging (57) and potentially reliant on the presence of ERs in females PM (58). Thus, additional NO mimetics appear needed to replace the NO responsiveness lost in menopause.

Effects of Inorganic Nitrate and Exercise Intensity on FMD

Reintroducing NO via a pathway independent of eNOS may reproduce the exercise-induced benefits seen after estrogen replacement in females PM. Similar to recent work by Somani et al. (59), NO₃⁻ supplementation alone did not improve FMD at rest (Table 2). However, the present study showed that NO₃⁻ supplementation enhanced FMD when combined with acute exercise. This was observed for both peak ΔFMD (Fig. 3) as well as changes in FMD across time after HIE (Fig. 4E). NO₃⁻ supplementation paired with moderate-intensity exercise had medium effects on FMD over time (∼1%; Fig. 4B). Thus, it appears that consuming an exogenous source of NO (which acts via an eNOS-independent pathway) may restore some of the benefits of exercise on vascular health in estrogen-deficient females PM, although this needs to be replicated with larger sample sizes. If these findings are reproducible, chronic moderate-intensity exercise training alone may not improve FMD in females PM, but the addition of exogenous inorganic nitrate may augment the exercise-induced improvements in FMD to levels similar to what is seen with MIE combined with estradiol use. The combination of MIE + BRJ may provide an alternative for females PM who are reluctant to undertake HIE training.

Combining exogenous NO with greater exercise-induced shear stress via higher exercise intensity offers a potentially two-pronged approach to improving FMD in females PM. Indeed, NO₃⁻ supplementation when combined with high-intensity exercise provided the greatest improvements in FMD when observed as both peak ΔFMD (Fig. 3, ∼4%) as well as FMD over time (Fig. 4E, ∼2%). HIE induced a biphasic response to FMD with an initial decrease or lack of change in FMD, followed by a delayed improvement in FMD (Fig. 4, D–F). This resulted in a significant effect of time when examining PL HIE versus BRJ HIE (Fig. 4F). The initial lack of response or worsening of FMD after HIE is commonly reported in the literature (31, 60) and has been attributed to the elevated exercise-induced increased shear rate resulting in a larger baseline diameter of the artery (60). NO₃⁻ supplementation has been reported to provide the greatest improvements in fast-twitch muscle fibers (61, 62), and reduction in NO₂⁻ to NO is known to occur in lower partial pressures of oxygen (63). Thus, the combined benefits of NO₃⁻ and HIE on FMD may reside to some extent from the increased recruitment of fast-twitch muscle fibers during HIE and greater conversion of NO₂⁻ to bioavailable NO.

Predictors of Peak ΔFMD Response

In the present study, the greatest predictor of peak ΔFMD was the supplemented plasma NO₂⁻ level such that those with greater supplemented plasma NO₂⁻ upon experimental visit arrival had greater changes in FMD. Similar to other studies (59), baseline-supplemented FMD did not differ between treatments (time points 0), suggesting that elevated plasma NO₂⁻ from BRJ alone does not improve resting FMD in females PM and that exercise is required to gain the added benefits from BRJ.

Contrary to our secondary hypothesis, neither baseline FMD, V̇o_2peak, TSM, nor AVF correlated with peak ΔFMD. Baseline PWV was a significant predictor of peak ΔFMD, suggesting that those with worsened arterial stiffness experience blunted enhancement to vascular function following exercise. Similarly, body fat percentage trended toward a negative relation with peak ΔFMD. This is in line with previous work from our laboratory, suggesting that improvements in FMD after HIE were only seen in lean individuals, but not in individuals with obesity (32). Thus, it appears that arterial stiffness and obesity may blunt the exercise-mediated improvements in FMD and that this does not change after BRJ supplementation. Because the same absolute NO₃⁻ dose was provided to each person (∼13 mmol/day), and because peak ΔFMD was negatively correlated with body weight in the BRJ group (R² = 0.52; P = 0.008), it appears that participants who have a higher body fat percentage may experience improvements in FMD postexercise if given a larger dose of NO₃⁻.

NO₃⁻ supplementation improved NO bioavailability (NO₃⁻ and NO₂⁻; Fig. 2, A and B) and decreased the vasoconstrictor ET-1 (Fig. 2C) more so than PL. These differences in N-oxides and ET-1 remained significantly different between groups (BRJ vs. PL) after experimental conditions (CON, HIE, MIE; Fig. 2, A–C). Females PM have altered ET-1 activity, and exercise training has recently shown to improve ET-1-mediated vasoconstriction in females PM (64, 65). Thus, NO₃⁻ supplementation may improve vascular responsiveness in females PM by improving the vasodilatory and vasoconstrictor balance of the vasculature. Despite this, ET-1 values did not correlate with peak ΔFMD. A ratio of NO:ET-1 has been suggested as a possible way to examine the balance between vasodilation and vasoconstriction (66), however, this ratio did not predict FMD outcomes. Furthermore, changes in plasma nitrite and ET-1 from pre- to postexperimental condition did not predict peak ΔFMD. Thus, it appears that supplemented levels of plasma NO₂⁻ were the only blood markers measured here that were correlated with the peak ΔFMD. Further enhancing plasma levels of NO₂⁻ by increasing the dose of NO₃⁻ supplementation or tailoring it to an individual’s body weight may provide even greater enhancements in vascular responsiveness, but this remains to be tested. Interestingly, our group recently showed that inorganic nitrate may worsen exercise capacity in premenopausal females (67), potentially implicating a relationship between endogenous estradiol and exogenous inorganic nitrate.

Effects of Exercise and Inorganic Nitrate on Blood Pressure and Pulse Wave Velocity

Acute exercise often induces elevations in blood pressure during exercise, followed by mild rebound hypotension after exercise caused by peripheral vasodilation (68). Postexercise hypotension has recently been shown not to be altered by NO₃⁻ supplementation in females PM (69), findings that agree with ours, as we only observed a time effect in each condition (Fig. 5, A–C). Similarly, acute postexercise PWV has been shown to be either decreased or normalized compared with control conditions (70), results that are similar to our findings (Fig. 5, D–F). Although FMD responses to exercise training are blunted in females PM, exercise-mediated improvements in blood pressure and PWV are still possible (71). Overall, these data suggest the beneficial impacts of NO₃⁻ and acute exercise on FMD occur via mechanisms not related to changes in blood pressure and arterial stiffness. Furthermore, NO₃⁻ supplementation alone did not improve PWV, MAP, or DBP, whereas SBP was improved similarly by both PL and BRJ (Table 2). It is unclear whether improvements in SBP in the PL group are from the presence of vasoactive compounds found in beetroot juice independent of inorganic nitrate (i.e., antioxidants, polyphenols, etc.). Other groups have shown a lack of benefit from NO₃⁻ supplementation in females PM on PWV (72), with mixed results on resting blood pressure (69, 72). Further research is needed to determine what factors may predict responders to NO₃⁻ supplementation.

Limitations

There are several limitations in this study. First, the sample of participants (n = 12 per treatment arm) was powered to examine the primary outcome of peak ΔFMD (regardless of time), but not for a three-way ANOVA for treatment × intensity × time. Consequently, separate two-way ANOVAs within each experimental condition and effect sizes were used to examine these differences.

Another limitation is that the sample of females PM were in relatively good vascular health. Only 9 of the 24 participants had a baseline FMD that is considered “impaired” for females PM (< 4.5% FMD). This may have masked the potential FMD improvements when extended to a less healthy population, as previous data suggest that individuals with impaired baseline FMD improve to a greater extent (73). Most of the participants in this study were Caucasian (n = 23) with only one participant from a Hispanic background. Although resting and postprandial FMD has previously been shown by our group to be unaffected by race in females PM (20), we cannot address whether postexercise FMD may be impacted differently across races with and without BRJ. In addition, we do not report sex hormones to assess menopause status, although at least 1 year after menses is commonly used to define menopause. In addition, participants in this study were on average 10 years beyond their final menstrual cycle. Finally, the follow-up period for this study was ∼2 h postexercise, and thus it is unknown for how long these effects on FMD may persist postexercise.

Conclusions

These findings reveal that inorganic nitrate and HIE combined may be a potential precision approach to improve vascular health in females PM, a population that is growing and is experiencing unmitigated insult to their vascular function. Further research should examine whether chronic exercise training interventions performed at high versus moderate intensities and paired with NO₃⁻ supplementation may rescue the impairments in endothelial function that are seen in postmenopausal females.

DATA AVAILABILITY

Data will be made available upon reasonable request.

GRANTS

This project was funded by the University of Virginia School of Education and Human Development Inclusion Diversity Equity Access (IDEA) Grant and by the University of Virginia Department of Kinesiology. K.M.L. was supported by National Institutes of Health Grant K23DK131327.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

A.C.H., K.M.L., A.W., and J.D.A. conceived and designed research; A.C.H., J.O.d.Z., N.W., J.R.C., V.M., L.C., and D.M. performed experiments; A.C.H., J.R.C., V.M., L.C., and D.M. analyzed data; A.C.H., K.M.L., A.W., and J.D.A. interpreted results of experiments; A.C.H. prepared figures; A.C.H. drafted manuscript; A.C.H., K.M.L., A.W., and J.D.A. edited and revised manuscript; A.C.H., J.O.d.Z., N.W., J.R.C., V.M., L.C., D.M., K.M.L., A.W., and J.D.A. approved final version of manuscript.