Acute beetroot juice reduces blood pressure in young Black and White males but not females

Abstract

A complex interplay of social, lifestyle, and physiological factors contribute to Black Americans having the highest blood pressure (BP) in America. One potential contributor to Black adult's higher BP may be reduced nitric oxide (NO) bioavailability. Therefore, we sought to determine whether augmenting NO bioavailability with acute beetroot juice (BRJ) supplementation would reduce resting BP and cardiovascular reactivity in Black and White adults, but to a greater extent in Black adults. A total of 18 Black and 20 White (∼equal split by biological sex) young adults completed this randomized, placebo-controlled (nitrate (NO₃⁻)-depleted BRJ), crossover design study. We measured heart rate, brachial and central BP, and arterial stiffness (via pulse wave velocity) at rest, during handgrip exercise, and during post-exercise circulatory occlusion. Compared with White adults, Black adults exhibited higher pre-supplementation resting brachial and central BP (Ps ≤0.035; e.g., brachial systolic BP: 116(11) vs. 121(7) mmHg, P = 0.023). Compared with placebo, BRJ (∼12.8 mmol NO₃⁻) reduced resting brachial systolic BP similarly in Black (Δ-4±10 mmHg) and White (Δ-4±7 mmHg) adults (P = 0.029). However, BRJ supplementation reduced BP in males (Ps ≤ 0.020) but not females (Ps ≥ 0.299). Irrespective of race or sex, increases in plasma NO₃⁻ were associated with reduced brachial systolic BP (ρ = −0.237, P = 0.042). No other treatment effects were observed for BP or arterial stiffness at rest or during physical stress (i.e., reactivity); Ps ≥ 0.075. Despite young Black adults having higher resting BP, acute BRJ supplementation reduced systolic BP in young Black and White adults by a similar magnitude, an effect that was driven by males.

Graphical abstract

Highlights

• •Racial disparities in blood pressure (BP) are well-established.
• •Reduced NO bioavailability in Black adults may contribute to the BP disparities.
• •Acute beetroot juice increases NO metabolites irrespective of race or sex.
• •Sex, but not race, influences blood pressure responsiveness to beetroot juice.

1. Introduction

Cardiovascular (CV) diseases account for nearly 875,000 deaths per year in the United States [1]. A complex interplay of social, lifestyle, and underlying physiological factors contribute to Black Americans having the highest prevalence of CV diseases of all racial/ethnic groups in America [[2], [3], [4], [5], [6]]. More than half of the racial disparity in CV disease is attributable to the significantly higher prevalence of high blood pressure (BP) in Black adults [7,8]. In Black individuals, high BP develops earlier in life, is more challenging to control, and is associated with an increased likelihood of adverse health outcomes, such as damage to the heart, kidneys, and brain [[9], [10], [11]] compared with other races/ethnicities in America. Additionally, Black adults exhibit altered vascular structure [12,13] and endothelial dysfunction in some [[14], [15], [16], [17], [18], [19]], but not all [12,14,20], studies. Several [[21], [22], [23], [24]], but not all [[25], [26], [27]], studies have also reported elevated CV reactivity (e.g., increases in heart rate [HR] and/or BP during stress) in Black compared with White adults. The heterogeneity in these findings is likely due to differences in the age of cohorts, methodology, social determinants of health, and health behaviors between cohorts [2,3,5]. Nonetheless, these studies collectively provide evidence for racial disparities in CV health.

Addressing social determinants is important in understanding and attenuating racial health disparities [[3], [4], [5], [6],28]. However, understanding the physiological underpinnings of these racial health differences will presumably aid in the development of evidence-based interventions for improving CV health in Black individuals [29]. One possible contributor to racial differences in CV health is reduced nitric oxide (NO) bioavailability in Black adults. Mechanistically, NO influences BP and arterial stiffness through the regulation of vascular tone [30] and arterial elasticity [31], and some [[32], [33], [34]], but not all [[35], [36], [37]], studies demonstrate reduced NO bioavailability in Black compared with White adults. Dietary nitrate (NO₃⁻) supplementation is a cost-effective approach to augment NO availability [38]. Importantly, some studies indicate dietary NO₃⁻ reduces resting BP [39,40], exercising BP [41,42], and arterial stiffness [[43], [44], [45]]. However, there is limited evidence for whether dietary NO₃⁻ supplementation can attenuate racial disparities in CV health [46], particularly in the peripheral vasculature.

Therefore, the purpose of this study was to determine whether an acute NO₃⁻ rich beetroot juice (BRJ) supplement reduces resting CV meaures (HR, BP, and carotid-femoral pulse wave velocity [cf-PWV]) and CV reactivity during stress (isometric hand grip [HG] exercise and post-exercise circulatory occlusion [PECO]) in young Black and White adults. We hypothesized that acute BRJ supplementation would reduce (i.e., improve) resting CV measures and CV reactivity in young Black adults and White adults, but to a greater extent in Black adults. In consideration of existing research demonstrating sex differences in response to dietary NO₃⁻ supplementation [47,48], we also compared CV measures following actute BRJ supplementation in males and females. We hypothesized that acute BRJ supplementation would yield more pronounced reductions in BP in males compared with females.

2. Methods

Findings from the present study were collected as part of a registered clinical trial (NCT05132556). All procedures involving human participants were approved by the Institutional Review Board at Georgia Southern University (H22040). The study was performed in accordance with the ethical standards of the Declaration of Helsinki. All participants provided written consent to participate.

2.1. Participants

All participants were free of any overt CV, metabolic, or neurological disorders. Inclusion criteria included age between 18 and 39 years and body mass index <35 kg/m². Exclusion criteria included current or recent use of any CV-acting medication(s) within the past six months. All participants self-reported their ethnicity and race and reported the ethnicity and race of their parents. Because race is a social construct, we adopted multiple approaches to characterize our participants [49], including assessing skin pigmentation, social determinants of health, and health behaviors (i.e., sleep and dietary patterns).

2.2. Assessment of skin pigmentation

As previously described [50], skin pigmentation was measured as the melanin index (M-index) using reflectance spectrophotometry (DermaSpectrometer; Cortex Technology, Hadsund, Denmark). We measured the M-index on the inner aspect of the participant's upper arm. The M-index was measured in this region because of its ease of access and because it represents constitutive skin pigmentation due to its relatively low sun exposure [50].

2.3. Racial/ethnic discrimination

The Perceived Ethnic Discrimination Questionnaire (PEDQ) is a 22-item self-report survey designed to assess the frequency of various acts of racial and ethnic discrimination [51,52]. Although the PEDQ references ethnic discrimination, the initial development of the questionnaire involved analysis undertaken according to racial groupings [51]. While we acknowledge the limitation, the terminology used in the PEDQ is consistent with the interchangeable use of race and ethnicity in the discrimination-related literature [51,52]. Participants were asked to reflect on the previous three months and indicate how often certain events have occurred, such as being called a name that is an ethnic slur or receiving unfair treatment based on their ethnicity. Examples of the items are “How often have you been subjected to nonverbal harassment because of your ethnicity?“, “How often have you been exposed to offensive comments because of your ethnic group?“, and “How often have you received unfair treatment from service people (e.g., waiters, bank tellers, security guards) because of your ethnicity?” Participants respond using a Likert scale with 1 being never and 7 being very often. The PEDQ is comprised of four subscales—Disvaluation, Avoidance, Verbal Rejection, and Threat Aggression—as well as a full-scale score [51,52]. Scores are derived by summing the responses to the individual items such that higher scores represent more of the experience being measured. The PEDQ was initially developed on an undergraduate population and was developed for use across all racial and/or ethnic groups [51,52].

2.4. Socioeconomic status

As previously described in detail [50], participants responded to a 10-item socioeconomic status (SES) questionnaire measuring constructs associated with increased daily stress, incident type 2 diabetes, and cross-sectional vascular dysfunction [50,53,54]. Briefly, participant responses were coded and added to provide an index of childhood SES, adulthood SES, and lifetime SES (childhood + adulthood SES).

2.5. Self-reported sleep

Self-reported sleep habits over one month were characterized using the Pittsburgh Sleep Quality Index (PSQI) [55]. The PSQI is a self-rated questionnaire intended to comprehensively assess sleep quality by generating seven component scores: sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleeping medication, and daytime dysfunction. By summing component scores in each of these domains, a global PSQI score is generated ranging from 0 to 21, with a lower score indicating healthier sleep.

2.6. Objective sleep

Actigraphy-derived sleep metrics were characterized as previously described [56]. In brief, participants were provided with an ActiGraph GT9X Link activity monitor (ActiGraph Corp, Pensacola FL, USA) and were instructed to wear the monitor on the dorsal side of the non-dominant wrist for 24 h a day for seven consecutive days and nights. Upon return, data were collected and analyzed in 60-s epochs using the Cole-Kripe algorithm and ActiLife software [57]. Sleep data were then visually inspected and manually adjusted with consideration of self-reported sleep logs by a single investigator (BLC) and checked by another investigator (GJG). Participants who wore the device for less than five nights were excluded from analysis. Actigraphy-derived sleep duration was operationalized as total nightly sleep time, not including wake after sleep onset, as time in bed (via sleep logs) and sleep duration (via movement patterns) are conceptually different outcomes. Sleep efficiency was defined as the ratio of total sleep time to time in bed. Sleep duration and efficiency were quantified for each night the accelerometer was worn and an average was calculated for each participant [[58], [59], [60]]. Sleep duration regularity (sleep SD) was operationalized using the standard deviation (SD) of nightly sleep duration [60]. The validity of GT9X-derived sleep parameters has been previously established against gold-standard polysomnography [61].

2.7. Dietary assessment and analysis

After the screening, we provided participants instructions for completing a three-day diet record including at least two weekdays and one weekend day. As recently described [62], to help with estimating portion sizes, participants were provided with a paper printout of well-established two-dimensional food models [63]. The diet record was returned at the laboratory testing visit, and a trained researcher reviewed the record with the participant to obtain additional clarification and detail. The records were analyzed for food group and nutrient intakes using the Nutrient Data System for Research (version 2020, University of Minnesota, Minneapolis, MN). Participants with at least two days of diet record data were included in the current analyses (mean = 3.0 ± 0.3 days). We report the average nutrient intake for all reported days.

2.8. Study design

This randomized, placebo-controlled, blinded, crossover design, acute BRJ supplement study consisted of three laboratory visits. The initial visit included written informed consent followed by the completion of a medical history questionnaire surveys to characterize social determinants of health and health behaviors. Anthropometric measurements were then obtained. Maximal HG strength (left hand) was determined using the highest force from three efforts (Jamar HG dynamometer, Sammons Preston, Bolingbrook, IL) as previously described [64]. We selected the left hand to enable BP and cf-PWV measures to be performed on the right side of the body, consistent with expert consensus [65].

Study visits two and three consisted of identical procedures, except participants received either 140 ml of the active NO₃⁻-rich BRJ supplement (∼12.8 mmol NO₃⁻; Beet It, James White Company) or an indistinguishable placebo (PLA, i.e., matched for texture and flavor) BRJ supplement with NO₃⁻ removed (Beet It Placebo, James White Company). All testing was performed in a temperature-controlled (21–23 °C) and dimly lit room [66]. Participants reported for testing between 0500 and 1000 (matched ± 1 h between visits) following an overnight fast and avoidance of strenuous physical activity, caffeine, tobacco, over-the-counter medication, and alcohol for 24 h, and water for 1 h [67].

An overview of the testing procedures is provided in Fig. 1. Following 10-min of quiet, supine rest, we collected at least duplicate measures of HR, BP, and cf-PWV and report the average herein. If differences in the initial two BP (>5 mmHg) or cf-PWV (>0.5 m/s) measures were observed, a third measure was collected and the closest two measures were averaged for analysis [68].

Fig. 1.

Study design. In the upper panel, we depict the crossover study design. In the bottom panel, we depict the timing of resting cardiovascular (CV) measures (heart rate, blood pressure, and carotid-femoral pulse wave velocity), CV responses to stress (isometric hand grip and an ischemic stimulus), and saliva and blood collection with respect to administration of nitrate-rich (active) or placebo beetroot juice (time = 0) are indicated by arrows.

Next, CV reactivity during physical stress and an ischemic stimulus was evaluated as previously described [64,69]. In brief, participants completed 3 min isometric HG exercise at 30% MVC, which was immediately followed by 3-min PECO via supra-systolic (∼240 mmHg) brachial cuff inflation. Participants’ HR and BP were measured during the second minute and cf-PWV was acquired during the final minute of each stimulus. Venous blood and saliva samples were then obtained, after which participants consumed the allocated supplement. Following 105-min of quiet, seated rest, participants laid supine for 10 min and we repeated experimental procedures to capture resting CV measurements and CV reactivity. Last, post-supplement blood and saliva samples were obtained. Experimental trials were separated by 2–14 days (Black adult average = 7 ± 3 days, White adult average = 6 ± 3 days).

2.9. Cardiovascular measures

As previously described [64,70], we placed a brachial cuff on the participant's right arm to acquire HR, brachial BP, and central BP using a SphygmoCor XCEL device (AtCor Medical, Naperville, IL, USA). Next, central arterial stiffness was assessed via cf-PWV, as previously described [71]. All cf-PWV measures were obtained by one of two trained investigations (BLC or JDV), with only high-quality recordings (defined as intra-device quality index ≥80%) included in the analysis.

2.10. NO metabolites

Nitric oxide metabolites in plasma and salivary samples were measured via chemiluminescence (nitrite; NO₂⁻) and HPLC-Griess assay (NO₃⁻), as described elsewhere [72]. In brief, blood collected in acid-citrate dextrose tubes was centrifuged at 2000×g for 5 min at 4 °C. Plasma and saliva samples were then immediately frozen in liquid nitrogen, and subsequently stored at −80 °C. For analysis, plasma and saliva samples were thawed and NO₂⁻ and NO₃⁻ were extracted using ice-cold methanol. NO₂⁻was detected using ozone-based chemiluminescence with a Sievers Nitric Oxide Analyzer. NO₃⁻ was measured using HPLC-coupled to the Griess reaction using an ENO-30 apparatus (Eicom, Japan). NO₂⁻and NO₃⁻ levels were calculated by comparison to standard curves generated daily and adjusted for extraction efficiency [73].

2.11. Circulating 17β-estradiol and progesterone

We used enzyme-linked immunosorbent assays (Enzo Life Sciences, Inc., Farmingdale, NY) to quantify endogenous concentrations of 17β-estradiol (kit sensitivity: 14 pg/ml) and progesterone (kit sensitivity: 15.6 pg/ml) from female participants. We collected plasma samples during both experimental trials in acid-citrate dextrose vacutainers. For 17β-estradiol samples were assayed in triplicate using a 1:4 dilution. The intraassay coefficient of variation for 17β-estradiol was 5.0%. Progesterone samples were assayed in triplicate using a 1:20 dilution. The intraassay coefficient of variation for progesterone was 5.5%.

2.12. Statistical analyses

Statistical analyses were performed using jamovi version 2.3.0. Descriptive characteristics are presented as means and standard deviation (SD). Normality was assessed through the Shapiro-Wilk test and visual inspection of quantile-quantile plots. Pre-supplement NO metabolites and resting CV measures were averaged between visits and compared between Black and White participants using independent t-tests or the non-parametric Mann-Whitney U test, as appropriate. We also calculated effect size using Cohen's d or rank biserial correlation, as appropriate. Pre-supplementation CV reactivity (Δ; post – pre) to stress (i.e., HG and PECO) was compared between Black and White participants using linear mixed-effects models, both before and after adjusting for resting CV measures. Associations between pre-supplementation NO metabolites and resting CV measures and changes in NO metabolites were assessed using Spearman's correlations. Post-supplementation NO metabolites, resting CV measures, and CV reactivity (Δ; post – pre) to stress (i.e., HG and PECO) were compared between races and treatments using linear mixed-effect models, controlling for pre-supplement values. In all instances, when a mixed-effect model revealed a significant F value, Bonferroni post-hoc tests were performed. Associations between changes (Δ; post-pre) in NO metabolites and changes in CV measures following BRJ supplementation were assessed using Spearman's correlations. Though not a primary focus of our investigation, in consideration of previous literature demonstrating sex differences in NO₃⁻ to NO₂⁻ conversion and BP responsiveness to dietary NO₃⁻ [47,48], we conducted exploratory analyses implementing the same statistical procedures mentioned above to compare CV responses to acute BRJ supplementation in male and female participants. For plasma 17β-estradiol and progesterone, equivalence testing was conducted between visits according to recent recommendations using the confidence interval method [74]. Significance was informed by setting α at or below 0.05 for all analyses.

3. Results

3.1. Participant characteristics

Forty individuals completed the study, but two Black female participants were excluded due to hemodynamic irregularities (e.g., abnormal heart rhythm) for a final sample size of 38 (self-identified: 18 Black adults, 10 males and 8 females; 20 White adults, 10 males and 10 females; Fishers exact test for sex P = 0.76). Participants' descriptive characteristics are summarized in Table 1. Black and White participants’ age and body mass index values were not different (Ps ≥ 0.266). Skin pigmentation (M-index) was greater in Black compared with White participants (P < 0.001).

Table 1.

Participant characteristics and pre-supplementation measures.

	Black (n=18)	White (n=20)	p-value	Effect Size
Descriptives
Age (yrs)	21.4 ± 3.4	21.1 ± 3.7	0.804	0.081
Height (m)	1.7 ± 1.1	1.7 ± 0.9	0.265	0.370
Weight (kg)	73 ± 13	67 ± 11	0.105	0.540
Body mass index (kg/m2)	24 ± 3	23 ± 3	0.266	0.367
M-index	62 ± 13	33 ± 3	< 0.001	3.172
Nitric oxide metabolites
Plasma nitrite (NO2−, μM)	0.3 (0.2)	0.4 (0.3)	0.650	0.089
Plasma nitrate (NO3−, μM)	10.3 ± 2.6	12.3 ± 3.4	0.057	0.639
Salivary nitrite (NO2−, μM)	184 (240)	271 (223)	0.496	0.133
Salivary nitrate (NO3−, μM)	267 (367)	236 (446)	0.874	0.033
Cardiovascular measures
cf-PWV (m/s)	6.1 ± 0.7	5.7 ± 0.7	0.083	0.580
Heart rate (bpm)	64 ± 9	62 ± 6	0.374	0.292
Brachial Systolic BP (mmHg)	121 (7)	116 (11)	0.023	0.433
Brachial Diastolic BP (mmHg)	72 ± 7	66 ± 6	0.007	0.922
Brachial Mean BP (mmHg)	88 ± 7	82 ± 6	0.014	0.841
Central Systolic BP (mmHg)	105 (7)	101 (10)	0.035	0.403
Central Diastolic BP (mmHg)	72 ± 7	67 ± 6	0.012	0.865
Central Mean BP (mmHg)	85 ± 8	80 ± 7	0.025	0.763

Normally distributed variables represented as Means ± SD and non-normally distributed variables as Median (interquartile range). Hypothesis testing performed using t-test and Cohen's d for effect size or non-parametric equivalent (Mann-Whitney U, rank biserial correlation), as appropriate.

3.2. Elevated pre-supplementation resting BP in Black participants vs. White participants

Table 1 provides average pre-supplementation NO metabolites and resting CV measures in Black and White participants. Apart from plasma NO₃⁻ (P = 0.057), no race-related differences in pre-supplementation NO metabolites were observed (Ps ≥ 0.496). Similarly, combined plasma NO₂⁻ and NO₃⁻ was lower in Black compared with White participants (10.7 ± 3.5 vs. 12.7 ± 2.7 μM; P = 0.056), but no racial difference in combined salivary NO₂⁻ and NO₃⁻ was observed (P = 0.407).

Compared with White participants, Black participants exhibited higher brachial systolic (P = 0.023), diastolic (P = 0.007), and mean BP (P = 0.014); and central systolic (P = 0.035), diastolic (P = 0.012), and mean BP (P = 0.025). Table 2 provides average pre-supplementation CV reactivity in Black and White participants averaged between trials, both before (unadjusted) and after (adjusted) controlling for resting values. During handgrip, no race-related differences were observed in unadjusted (Ps ≥ 0.545) or adjusted (Ps ≥ 0.233) models. Similarly, during PECO no race-related differences were observed in unadjusted (Ps ≥ 0.559) or adjusted (Ps ≥ 0.272) models.

Table 2.

Comparison of pre-supplementation cardiovascular stress responses (absolute Δ) between Black and White participants).

	Black (n=18)	White (n=20)	Unadjusted	Adjusted
Handgrip
cf-PWV (m/s)	1.3 ± 0.8	1.4 ± 0.8	0.635	0.927
Heart rate (bpm)	17 ± 11	17 ± 10	0.842	0.755
Brachial Systolic BP (mmHg)	16 ± 9	17 ± 9	0.718	0.513
Brachial Diastolic BP (mmHg)	18 ± 10	19 ± 8	0.545	0.313
Brachial Mean BP (mmHg)	17 ± 9	18 ± 8	0.567	0.233
Central Systolic BP (mmHg)	17 ± 9	17 ± 9	0.837	0.499
Central Diastolic BP (mmHg)	19 ± 11	20 ± 9	0.668	0.473
Central Mean BP (mmHg)	21 ± 10	21 ± 9	0.736	0.428
Post-Exercise Cuff Occlusion
cf-PWV (m/s)	0.9 ± 0.9	0.7 ± 1.2	0.559	0.664
Heart rate (bpm)	3 ± 8	3 ± 5	0.690	0.370
Brachial Systolic BP (mmHg)	14 ± 12	14 ± 9	0.946	0.945
Brachial Diastolic BP (mmHg)	14 ± 11	15 ± 9	0.575	0.919
Brachial Mean BP (mmHg)	14 ± 10	14 ± 9	0.875	0.907
Central Systolic BP (mmHg)	15 ± 12	14 ± 9	0.670	0.674
Central Diastolic BP (mmHg)	14 ± 11	14 ± 9	0.940	0.669
Central Mean BP (mmHg)	15 ± 12	14 ± 13	0.591	0.272

3.3. Black participants experience greater discrimination and exhibit poorer sleep and micronutrient intake

Participant social determinants of health, sleep variables, and nutrient intake data are summarized in Table 3. Regarding social determinants of health, perceived ethnic discrimination (total score), disvaluation, verbal rejection, and avoidance were greater in Black compared with White participants (Ps ≤ 0.023). Black participants exhibited greater sleep duration irregularity (P = 0.037), but PSQI score, sleep duration, and sleep efficiency were not different between races (Ps ≥ 0.089). Black and White participants did not exhibit different caloric or macronutrient intakes. However, Black participants exhibited lower fiber (P = 0.046) and calcium (P = 0.006) intake and Black participants tended to have a higher dietary sodium:potassium ratio (P = 0.055).

Table 3.

Participant social determinants of health, sleep variables, and self-reported dietary intake.

	Black (n = 18)	White (n = 20)	p-value	Effect Size
Social determinants
PEDQ Total	29 (16.5)	19 (3.3)	< 0.001	0.667
Disvaluation	10.3 (8.8)	5.2 (0.0)	< 0.001	0.711
Threat Aggression	4.2 (0.0)	4.2 (0.0)	0.369	0.111
Verbal Rejection	6 (4.4)	2.8 (2.4)	0.008	0.489
Avoidance	3.0 (4.4)	2.3 (0.5)	0.023	0.397
SES Total	17.3 ± 5.0	16.5 ± 4.4	0.568	0.187
Childhood	8.4 ± 3.2	9.1 ± 3.6	0.559	0.191
Adult	8.8 ± 2.7	7.4 ± 2.0	0.068	0.612
Sleep variables
PSQI	3.9 ± 1.6	5.3 ± 3.0	0.089	0.568
Sleep duration (min/night)†	366 ± 48	379 ± 52	0.476	0.257
Sleep duration SD (min)†	79 ± 23	59 ± 26	0.037	0.778
Sleep efficiency (%)†	81.3 (6.4)	84.4 (5.0)	0.311	0.215
Dietary intake
Energy (kcal)	2112 ± 541	2100 ± 658	0.953	0.019
Total fat (g)	86 ± 30	89 ± 32	0.762	0.099
Total carbohydrate (g)	253 ± 73	232 ± 80	0.399	0.277
Total protein (g)	85 (20)	88 (22)	0.740	0.067
Fiber (g)	14 ± 5	18 ± 7	0.046	0.670
Calcium (mg)	671 ± 273	967 ± 344	0.006	0.948
Magnesium (mg)	242 (93)	291 (94)	0.059	0.361
Sodium (mg)	3630 ± 1137	3793 ± 1270	0.679	0.135
Potassium (mg)	2279 ± 717	2813 ± 1165	0.103	0.544
Sodium:Potassium	1.6 (0.32)	1.4 (0.43)	0.055	0.367

Normally distributed variables represented as Means ± SD and non-normally distributed variables as Median (interquartile range). Hypothesis testing performed using t-test and Cohen's d for effect size or non-parametric equivalent (Mann-Whitney U, rank biserial correlation), as appropriate. ^†n = 13 for Black participants 1.4 (0.43).

Associations between social determinants of health and lifestyle variables with resting cardiovascular measures are provided in Supplemental Table 1. In consideration of prior data associating discrimination with BP [75,76] and the race differences in PEDQ in our investigation (Table 3), we performed mediation analysis to determine the extent to which total PEDQ score explained race differences in BP (Supplemental Table 6). In all models, the direct effect of race on BP was attenuated (Ps ≥ 0.05). Moreover, there was an indirect effect of PEDQ total on relations between race and BP (Ps ≤ 0.052) for brachial systolic BP (65.4%); and central systolic (73.8%) and central mean BP (54.1%).

3.4. Salivary, but not plasma, NO metabolites associate with resting BP

Pre-supplementation salivary NO₃⁻ was associated with brachial diastolic (ρ = 0.404, P = 0.012) and mean BP (ρ = 0.347, P = 0.033); and central diastolic (ρ = 0.390, P = 0.016) and mean BP (ρ = 0.388, P = 0.016). However, no associations between resting CV measures with plasma NO₂⁻ (Ps ≥ 0.176) and NO₃⁻ (Ps ≥ 0.600), or salivary NO₂⁻ (Ps ≥ 0.106) were observed.

3.5. Acute BRJ supplementation increases NO metabolites and reduces resting BP in young Black and White participants by a comparable magnitude

Post-supplementation plasma and salivary NO₂⁻ and NO₃⁻ are displayed in Fig. 2. Greater plasma NO₂⁻ (Fig. 2 A, P = 0.020) and NO₃⁻ (Fig. 2 B, P < 0.001) and salivary NO₂⁻ and NO₃⁻ (Fig. 2C & D, Ps < 0.001) were observed following BRJ vs. PLA. No effects of race (Ps ≥ 0.200) or treatment × race interactions (Ps ≥ 0.152) were observed. Likewise, when NO₂⁻ and NO₃⁻ were combined, greater plasma and salivary NO metabolites were observed following BRJ vs. PLA (Ps < 0.001), but no effects of race (Ps ≥ 0.118) or treatment × race interactions (Ps ≥ 0.086) were observed. Baseline plasma NO₂⁻ was inversely associated with changes in plasma NO₂⁻ (ρ = −0.700, P < 0.001). Additionally, baseline plasma NO₃⁻ was inversely associated with changes in plasma NO₃⁻ (ρ = −0.321, P = 0.008) as well as salivary NO₂⁻ and NO₃⁻ (ρs = −0.384 to −0.387, Ps ≤ 0.001).

Fig. 2.

Post-supplementation plasma NO₂⁻ (A) and NO₃⁻ (B) and salivary NO₂⁻ (C) and NO₃⁻ (D) concentration in White and Black participants. Individual (dots) and mean (bars) data depicted as unadjusted values. Hypothesis testing performed and presented adjusting for pre-treatment values as a time-varying covariate. For Black participants, n = 16–18 and for White participants n = 17–20 for plasma measures. For Black participants, n = 15–17 and for White participants n = 16–18 for salivary measures.

Post-supplementation resting CV measures are displayed in Fig. 3. Compared to PLA, HR was higher (Fig. 3 B, P = 0.031) and brachial systolic BP was lower (Fig. 3C, P = 0.029) following BRJ supplementation. Additionally, post-supplementation central diastolic BP was greater (Fig. 3 G, P = 0.031) in Black participants vs. White participants. No other effects of treatment (Ps ≥ 0.075), race (Ps ≥ 0.083), or treatment × race interactions (Ps ≥ 0.316) were observed.

Fig. 3.

Post-supplementation resting cardiovascular measures in White and Black participants. Individual (dots) and mean (bars) data depicted as unadjusted values. Hypothesis testing performed and presented adjusting for pre-treatment values as a time-varying covariate. For Black participants n = 17–18 and for White participants n = 20.

3.6. Changes in plasma NO metabolites inversely associate with changes in brachial BP

Increases (Δ; post – pre) in plasma NO₃⁻ following supplementation were associated with reductions in brachial systolic BP (Supplemental Fig. 1, ρ = −0.237, P = 0.042). However, no associations between changes in CV measures following supplementation with plasma NO₂⁻ (P ≥ 0.544), or salivary NO₂⁻ (P ≥ 0.068) and NO₃⁻ (P ≥ 0.060) were observed.

3.7. No influence of acute BRJ supplementation on CV reactivity

Post-supplementation CV responses to HG exercise are displayed in Fig. 4. No effects of treatment (Ps ≥ 0.415), race (Ps ≥ 0.367), or treatment × race interactions (Ps ≥ 0.381) were observed. Post-supplementation CV responses to PECO are displayed in Fig. 5. Changes in post-PECO HR (Fig. 5 B, P = 0.018) were reduced following BRJ supplementation. No other effects of treatment (Ps ≥ 0.411) or race (Ps ≥ 0.235) were observed. Apart from Δ brachial systolic BP (P = 0.052), no treatment × race interactions (Ps ≥ 0.144) were observed.

Fig. 4.

Post-supplementation cardiovascular responses to handgrip (HG) exercise in White and Black participants. Individual (dots) and mean (bars) data depicted as unadjusted values. Hypothesis testing performed and presented adjusting for pre-treatment values as a time-varying covariate. For Black participants n = 17–18 and for White participants n = 18–20.

Fig. 5.

Post-supplementation cardiovascular responses to post-exercise cuff occlusion (PECO) in White and Black participants. Individual (dots) and mean (bars) data depicted as unadjusted values. Hypothesis testing performed and presented adjusting for pre-treatment values as a time-varying covariate. For Black participants n = 16–18 and for White participants n = 18–20.

3.8. Acute BRJ supplementation reduces resting BP in young male but not female participants

Supplemental Table 2 provides descriptive characteristics and pre-supplementation resting CV measures of male and female participants. Compared with female participants, male participants exhibited higher cf-PWV and brachial systolic BP (Ps ≤ 0.018), and greater increases in cf-PWV during PECO (Supplemental Table 3, P = 0.019). Despite a lack of sex difference in increases in NO metabolites following acute BRJ supplementation (Supplemental Fig. 2, Ps ≥ 0.201), significant treatment × sex interactions for post-supplementation brachial and central systolic, diastolic, and mean BP were observed (Supplemental Fig. 3C–H, Ps ≤ 0.003); acute BRJ supplementation reduced brachial and central BP indices in males (Ps ≤ 0.020), but not females (Ps ≥ 0.299). Post-supplementation CV responses during HG exercise and PECO are displayed in Supplemental Figs. 4 and 5. Changes in post-PECO HR were reduced following BRJ supplementation (Supplemental Fig. 5 B, P = 0.009). No other effects of treatment, sex, or treatment × sex interactions were observed for CV reactivity (Ps ≥ 0.081).

3.9. 17β-estradiol and progesterone

17β-estradiol was not different between BRJ and PLA trials in Black (183 ± 68 vs. 213 ± 90 pg/ml) or White (269 ± 193 vs. 242 ± 135 pg/ml) female participants (Ps ≥ 0.275), and no racial differences were observed (Ps ≥ 0.216). Equivalence testing indicated that 17β-estradiol was not equivalent between trials in either Black (upper bound) or White (lower bound) female participants (Supplemental Table 4). Progesterone was not different between BRJ and PLA trials in Black (3971 ± 4021 vs. 3315 ± 2589 pg/ml) or White (5981 ± 3234 vs. 5626 ± 3079 pg/ml) female participants (Ps ≥ 0.659), and no racial differences were observed (Ps ≥ 0.125). Equivalence testing indicated that progesterone concentration was equivalent between trials in both Black and White female participants (Supplemental Table 4).

4. Discussion

The purpose of this study was to determine whether an acute, NO₃⁻-rich BRJ supplement would improve resting CV health indices and reduce CV reactivity in young Black and White adults, with a greater effect anticipated in Black adults. The main findings from this investigation were that despite young Black adults having higher resting systolic BP, acute BRJ supplementation reduced systolic BP by a similar magnitude in young Black and White adults, an effect that was driven by males. Irrespective of race or sex, increases in circulating NO₃⁻ were associated with reductions in systolic BP.

In partial support of our hypothesis, acute BRJ supplementation increased NO metabolites and reduced brachial systolic BP. A 25% increase in plasma NO₃⁻ is considered to be a critical threshold for physiological impact [77]. Congruent with our observations, past human studies consistently demonstrate that BRJ supplementation raises plasma NO₃⁻ [[78], [79], [80], [81], [82], [83]]. Further, BRJ lowers brachial BP in those with [45,78,84,85] and without hypertension in most [79,80,82,84,86,87], but not all [81] studies. Additionally, there are prior data demonstrating that increases in plasma NO₂⁻, but not NO₃⁻, associate with reductions in brachial BP [88]. However, our finding of an association between increased plasma NO₃⁻ and reduced brachial systolic BP is not unprecedented [89]. Consistent with previous research on sex differences with dietary NO₃⁻supplementation, we report BRJ-mediated reductions in BP that were exclusive to males [47,48], despite a lower relative NO₃⁻ dose (NO₃⁻/kg). This may be explained by the lower baseline systolic BP values in our female participants, which is common in the literature [88]. However, this explanation fails to account for reductions in diastolic and mean BP in our male but not female participants, as baseline values were similar. Sex-specific reductions (i.e., males only) in systolic and diastolic BP in apparently healthy males and females with similar baseline BP values have been previously reported [90]. Though the reason for this apparent sex difference is uncertain, these findings demonstate the utility of dietary NO₃⁻ as an effective BP lowering strategy in males. However, more research is needed in post-menopausal females and also to better understand apparent sex differences in BP responsiveness to dietary NO₃⁻ supplementation in premenopausal females compared with males.

Contrary to our hypothesis, acute BRJ supplementation did not affect central BP or arterial stiffness in the sample as a whole, and resting CV responses were comparable between Black and White participants. A prior investigation also reported no effect of acute BRJ on arterial stiffness in young adults (n = 15, 11 females and four males) with normal BP [91]. However, this investigation did report reductions in central BP 30 min after BRJ ingestion. Additionally, in patients with chronic kidney disease (n = 17, 10 females and 7 males), acute dietary NO₃⁻ reduced brachial and central BP after 4 h [83]. The reason for these conflicting findings is uncertain, as these studies consisted of a greater proportion of female participants in whom we observed no reductions in BP. However, inter-study differences in the time-point at which post-supplementation BP measures were acquired may contribute to this discrepancy. Although, plasma NO metabolite pharmacokinetic data from prior investigations indicate that the BP-lowering effects of BRJ would be maximized at 2 h, when we performed our CV measurements [92].

We observed no associations between plasma NO metabolites and resting CV measures. In contrast, in clinical populations and older adults, NO bioavailability predicts hypertension, and recently an inverse association between urinary/plasma biomarkers of NO synthesis and central BP was observed in young Black adults [93]. However, we did observe an unanticipated positive association between salivary NO₃⁻ with brachial and central diastolic and mean BP. This may be explained by a lower abundance of BP-lowering oral NO₃⁻-reducing bacteria in participants with higher salivary NO₃⁻. Indeed, a higher abundance of oral NO₃⁻-reducing bacteria would favor the conversion of salivary NO₃⁻ to NO₂⁻ [94], and is associated with lower systolic BP in individuals with normal BP [95]. In further support of this idea, exploratory analyses revealed higher NO₃⁻:NO₂⁻ ratio in participants with a higher salivary concentration of NO₃⁻ (ρ = 0.662, P < 0.001), and NO₃⁻:NO₂⁻ ratio was associated with higher brachial and central diastolic BP (Ps ≤ 0.048). Regarding racial differences, we observed some differences in plasma but no differences in salivary NO metabolites between Black and White participants. This finding adds to a small but growing body of literature in support of reduced circulating NO bioavailability in young Black compared to White adults [[32], [33], [34]], though conflicting reports [[35], [36], [37]] highlight the need for more research in this area. In consideration of the elevated baseline plasma NO metabolites in White compared to Black participants, it is somewhat surprising that CV responses to BRJ were similar. Although we acknowledge NO-dependent signaling can be influenced by reactions of NO with oxygen radical species (e.g., superoxide, lipid radicals) [38]. Additionally, skeletal muscle can serve as a reservoir and source of NO₃⁻ but was not considered in the current investigation [96].

Acute BRJ supplementation failed to alter CV reactivity during HG exercise, a finding that is consistent with previous work in young adults [42,[97], [98], [99], [100]]. However, studies investigating whether BRJ supplementation may alter BP responses during exercise in other populations have produced mixed results. For example, BRJ reduced exercising BP in post-menopausal female participants with hypertension [101], but not in patients with type 2 diabetes [102] or older adults [103]. Interestingly, BRJ supplementation attenuated brachial BP responses during knee-extensor, but not handgrip, exercise in adults with hypertension [78]. In contrast to static exercise, BRJ has been demonstrated to attenuate BP responses during upper body [104] and lower-body [82] dynamic exercise in young adults. Regarding our ischemic stimulus, we found that BRJ did not affect BP responses during PECO, whereas previous work demonstrated BRJ attenuates BP responses during PECO in older adults [105]. Collectively, these findings suggest that while acute dietary NO₃⁻ supplementation may reduce CV reactivity in clinical populations and/or older adults, these effects are less pronounced in younger, apparently healthy individuals.

In the present study, Black participants exhibited elevated resting brachial and central BP values compared with White participants. This observation agrees with findings from larger-scale cross-sectional studies demonstrating that racial differences in BP appear to emerge before 30 years of age [1,9,11]. Thus, our findings futher emphasize the significance of multi-level, primordial and primary prevention strategies to control BP in Black Americans [106]. Nonetheless, we did not observe racial differences in arterial stiffness or CV responses during HG exercise and/or an ischemic stimulus. We measured arterial stiffness using cf-PWV, which is the gold standard measure for central arterial stiffness and vascular biological aging. The observed lack of racial differences in central arterial stiffness is in contrast to the greater cf-PWV values previously reported in Black compared with White American children [107] and young adults [108]. However, consistent with our observations several international studies have demonstrated no difference in cf-PWV between Black and White adults [109,110], and one study even reported greater PWV in White compared with Black adults with normal BP [111]. Notably, in the present study, all participants were recruited from a college campus, which likely tempered racial differences in social determinants of health and consequently CV health status compared with a community sample. Indeed, social factors are suspected to account for a significant portion of CV health disparities [3,6].

A particular strength of the present investigation was our thorough characterization of social determinants and health behaviors, which unveiled some noteworthy differences that are worthy of discussion. Despite comparable self-reported SES in our Black and White participants, perceived ethnic discrimination (total score and all domains aside from threat aggression) was demonstrably higher (i.e., worse) in Black compared with White participants. Additionally, discrimination mediated a substantial portion of the higher resting brachial and central systolic BP in Black participants. There is a growing body of evidence indicating discrimination may contribute to the worse CV health observed in racial/ethnic minorities. For instance, in the longitudinal Jackson Heart Study, both medium and high levels of discrimination were associated with a higher incidence of hypertension [112]. Furthermore, in the Brazilian Longitudinal Study of Adult Health, racial discrimination was shown to mediate the greater PWV measured in darker skinned compared with lighter skinned individuals [112]. Additionally, smaller-scale studies in racial/ethnic minorities have linked perceived racism to elevations in ambulatory blood pressure and lack of nocturnal dipping [113], as well as CV reactivity [114]. Collectively, these data provide important context to underlying determinants of racial disparities in CV health, and specifically highlight the role of racial/ethnic discrimination in mediating these disparities.

We also observed key differences in health behaviors between our young Black and White participants, particularly sleep and diet. As we have reviewed previously, differences in social determinants of health can influence health behaviors such as leisure physical activity, sleep, and diet [3]. For example, self-reported childhood neighborhood safety mediates racial differences in sleep in young adults [115], and our group and others have observed that poor sleep contributes to worse CV health, even in young adults [[116], [117], [118]]. More specifically, in a comparable population of young adults we previously demonstrated that higher BP was associated with sleep irregularity (i.e., variability in nightly sleep duration) [56], which was greater in Black vs. White participants in the present study. Disparities in diet quality may also influence racial differences in cardiovascular health [119]. While Black and White participants did not exhibit different caloric or macronutrient intakes, the lower dietary intake of calcium and fiber and the higher sodium:potassium ratio in our young Black participants indicates potential disparities in diet quality. In sum, these findings highlight the importance of considering health behaviors, particularly sleep and diet, as possible contributors to the earlier incidence of high BP in young Black Americans.

Our findings should be interpreted in the context of our studies strengths and limitations. A potential limitation of our study is that we did not control for the menstrual phase by separating trials across cycles. The menstrual cycle can influence CV control in females [120]. However, a recent study found no influence of the menstrual cycle on ischemic pain perception (i.e., the stimulus for subsequent CV responses) [121]. Importantly, our female participants did not exhibit a difference in estrogen or progesterone concentrations between trials and some have argued that not controlling for menstrual cycle increases ecological validity [122]. Additionally, we recruited the majority of our participants from a college campus, which limits the generalizability of our findings. However, our sample is representative of the background of many adults in CV disease trials, such as the Multi-Ethnic Study of Atherosclerosis in which 69% of participants attended college [123]. Another limitation was our relatively modest sample size, which prevented inspection of race × sex × treatment interactions (see Supplemental Table 5). Ideally, follow-up trials will include larger sample sizes and include other racial and ethnic groups (e.g. Asian and Hispanic/LatinX). Notable strengths of this investigation include our use of a randomized, placebo-controlled, blinded, crossover design, thorough characterization of participant sociocultural and lifestyle factors that may influence CV health, and our evaluation of central BP and cf-PWV using validated and highly reliable equipment [124].

In conclusion, we demonstrate that acute BRJ supplementation reduced brachial systolic BP by a similar magnitude in young Black and White adults, an effect that was driven by males, and is consistent with previous research in the field [89]. Irrespective of race or sex, reductions in brachial systolic BP were associated with increases in circulating NO₃⁻. However, acute BRJ supplementation did not influence resting arterial stiffness or central BP, or BP and arterial stiffness responses during handgrip exercise or PECO. Additionally, no race differences in NO metabolite or CV responses to BRJ supplementation were observed. Nonetheless, racial differences in social determinants of health (i.e., racial discrimination) and lifestyle (i.e., sleep irregularity and micronutrients) were observed, and may contribute to the elevated brachial and central BP values in Black compared with White participants. Collectively, these findings highlight the difficulty of untangling the complex interplay of social, lifestyle, and underlying physiological factors that mediate racial CV health disparities, but also demonstrate the utility of dietary nitrate as an effective blood pressure lowering strategy in young Black and White males.

Funding

This project was supported by a Georgia Southern University Internal Seed Grant (GJG) and NHLBI grants K01HL160772 (JCW), K01HL147998, R15HL165325 (ATR), UL1TR003096 (BAL).

Declaration of interest

None.

Appendix A. Supplementary data

The following is the Supplementary data to this article.

Data availability

Data will be made available on request.