15N‐labeled dietary nitrate supplementation increases human skeletal muscle nitrate concentration and improves muscle torque production

Dietary nitrate (NO3−) supplementation increases nitric oxide bioavailability and can enhance exercise performance. We investigated the distribution and metabolic fate of ingested NO3− at rest and during exercise with a focus on skeletal muscle.


| INTRODUCTION
The signaling molecule, nitric oxide (NO), is essential for the maintenance of normal physiological function. 1,2 The highly reactive nature and relatively short half-life of NO means that sustained provision of this molecule may be compromised if it is not continuously synthesized. 1 Following its production from L-arginine in a reaction catalyzed by the nitric oxide synthase (NOS) enzymes, NO may be oxidized to form the more stable metabolites, nitrite (NO 2 − ) and nitrate (NO 3 − ). NO 2 − and NO 3 − are now considered to be storage forms of NO since they can be also reduced under appropriate physiological conditions (i.e., low PO 2 ) to form NO. 3  ] in biological tissues such as saliva, 4-6 plasma, 7-9 urine 5,10,11 and, most recently, skeletal muscle. [12][13][14][15][16] The augmentation of body NO bioavailability following dietary NO 3 − ingestion may have important physiological and therapeutic effects. 17,18 At any given time, skeletal muscle  ] reflects the balance between metabolism of NO 3 − into other nitrogencontaining species, oxidation of NO produced via NOS into NO 3 − , and NO 3 − and NO 2 − exchange between muscle and blood with the latter facilitated by sialin 15,19 and chloride channels. 20 While muscle  ] has been shown to be elevated following dietary NO 3 − supplementation, 12,[14][15][16] the extent to which this results from accretion of the exogenously-supplied NO 3 − , via uptake from the circulation, is uncertain. Determining the proportional contribution of exogenous NO 3 − and endogenously-generated  12,13 ) and humans (vastus lateralis 14,15 ) has led to speculation that this relatively high muscle  ] may have functional significance. 3 It has been suggested that skeletal muscle serves as a NO 3 − "reservoir" that might be drawn upon, via the circulation, to enhance NO bioavailability in other tissues when access to dietary NO 3 − is restricted. 12,18,21 Moreover, skeletal muscle possesses the enzymatic machinery required for the reduction of NO 3 −  ] was not different from baseline at 1-h (from 28 ± 6 to 28 ± 5 nmol/g) but was lower than baseline at 3-h (27 ± 6 nmol/g; p < 0.05). Total plasma [NO 2 − ] was greater than baseline at 1-h (from 0.13 ± 0.02 to 0.29 ± 0.07 nmol/g; p < 0.001) and 3-h (0.47 ± 0.14 nmol/g; p < 0.001) and the value at 3-h was greater than that at 1-h (p < 0.01) ( Figure 1D). 15  ] was greater than baseline at 1-h (0.08 ± 0.04 nmol/g, p < 0.001) and 3-h (0.21 ± 0.10 nmol/g; p < 0.001) and the value at 3-h was greater than that at 1-h (p < 0.05). Unlabeled plasma [NO 2 − ] was also elevated above baseline at 1-h (from 0.12 ± 0.02 to 0.20 ± 0.03 nmol/g; p < 0.001) and 3-h (0.26 ± 0.05 nmol/g; p < 0.001), and the value at 3-h was greater than that at 1-h (p < 0.05). ] was greater than baseline at 1-h (from 739 ± 225 to 2567 ± 1360 nmol/g, p < 0.05) and 3-h (2466 ± 1070 nmol/g; p < 0.001) following NO 3 − ingestion, and there was no difference between 1-and 3-h ( Figure 1G). 15  ] was not different from baseline at either 1-or 3-h ( Figure 1H). 15  ] ratio was greater than baseline at both 1-and 3-h (p < 0.01) ( Figure 2B). ] did not change from pre-exercise (105 ± 41 nmol/g) to post-exercise either in the control leg (93 ± 35 nmol/g) or the exercised leg (71 ± 40 nmol/g; p = 0.09) in the NIT condition ( Figure 3A). Moreover, exercise did not alter total muscle [NO 3 − ] in the PLA condition with total response in black font. a Significant difference compared to 0-h (p < 0.05). b Significant difference between 1-and 3-h (p < 0.05). Black, blue, and red letters refer to comparisons between total, unlabeled and 15 N labeled data, respectively.

| Exercise performance
The peak MVC torque achieved prior to the 5-min allout test was 266 ± 95 N/m and 260 ± 60 N/m, and the mean torque sustained during this contraction was 221 ± 83 N/m and 220 ± 49 N/m for PL and NIT, respectively. Voluntary activation of the knee extensors achieved during the MVCs performed prior to the 5-min all-out test was 91 ± 8% and 91 ± 5% for PLA and NIT, respectively. Baseline peak MVC torque, mean torque and voluntary activation were not different between conditions. To facilitate comparisons between PLA and NIT conditions, all torque profiles were subsequently normalized to the MVC values achieved during the PLA condition (MVCPL).

| 5-min test torque profile
The profile for peak torque and mean torque across all participants during each contraction for the 5-min test is shown in Figure 4. During the PLA trial, normalized peak torque declined from 90 ± 9% MVC during the first contraction to 40 ± 10% MVC during the final contraction (p < 0.001). The overall decline in normalized peak torque was not significantly different between the NIT (−52 ± 15%) and PLA (−55 ± 20%) conditions. There were no significant differences between the PLA and NIT conditions for the overall decline in mean torque or torque impulse across the 60 MVCs, and no differences were observed for CT or impulse above CT (Table 1).
The potentiated doublet responses and voluntary activation across all participants at each assessment timepoint during the 5-min all-out test are provided in Figure 5. During the PLA trial, potentiated doublet responses decreased from 88 ± 11 N m during the first contraction to 40 ± 12 N m during the final contraction (p < 0.001). This overall decline was not significantly different between PL (−54 ± 15%) and NIT (−55 ± 17%) (p > 0.05). Voluntary activation changed across time (p < 0.05) during the 5-min all-out test ( Figure 1B), but there were no differences between PL and NIT.   ] during exercise was significantly correlated with a greater peak mean torque %MVCPL (r = 0.71, p < 0.001) and a greater mean mean torque %MVCPL (r = 0.62, p < 0.01) during the first 90 s of the 5min test. In contrast, there were no significant correlations between changes in total, or 15  -supplemented (black circles) conditions during the 60 maximal isometric contractions that comprised the 5-min all-out test. All contractions were normalized to a baseline maximal voluntary contraction performed during the placebo condition. # Trend for a significant difference from placebo in mean peak torque for the 90-s time bin (p = 0.05). a Trend for peak mean torque to be higher compared to placebo (p = 0.06). *Mean mean torque significantly different from placebo for the 90-s time bin (p < 0.05). ] remaining elevated at 3-h. These findings are consistent with earlier reports. 5,6,8,11,16,24 Interestingly, at 1-h, 15  ]. This suggests that recently ingested NO 3 − , following its arrival at the muscle, might be more "accessible" than the baseline NO 3 − store, and raises the intriguing possibility that NO 3 − which is generated endogenously and NO 3 − which is supplied exogenously might be stored and processed differently within the muscle, at least in the short-term. One possibility which remains to be explored is the extent to which s-nitrosothiols may be formed in muscle and contribute as NO precursors. 26

| Exercise performance
Study participants completed a 5-min all-out test involving 60 intermittent (3-s contraction, 2-s rest) isometric MVCs of the knee extensors. 29 The muscle was electrically stimulated on the first, 15th, 30th, 45th, and 60th contraction. This protocol enables measurement of dynamic changes in torque development as well as assessment of the contribution of central and peripheral factors to fatigue development. 30 We found no differences between NIT and PLA for voluntary activation or the potentiated doublet response, suggesting that central and peripheral fatigue development was similar between the conditions. Moreover, there was no difference in the end-test torque or impulse above endtest torque. However, the mean torque produced during muscular contractions over the first 90 s of the test (i.e. first 18 contractions) was significantly greater for NIT compared to PLA. These results contrast with one study, which reported no significant effect of acute NO 3 − ingestion on indices of central or peripheral fatigue or on time-to-task-failure during intermittent isometric knee extension exercise, 31 but are consistent with another study, which found that 5 days of NO 3 − supplementation reduced the rate of muscle fatigue development and extended the time-to-task-failure during dynamic knee extension exercise. 32 Recent meta-analyses have highlighted the potential for dietary NO 3 − supplementation to enhance skeletal muscle force or power production during high-intensity exercise. 33,34 Acute NO 3 − ingestion has been reported to result in a ~ 5% increase in peak power during isokinetic dynamometry 35 and sprint cycling 36 whereas chronic (5-6 days) NO 3 − supplementation improved performance during a 30-s cycle sprint, 37 and 5-20-m running sprints. 38 Dietary NO 3 − supplementation also appears to enhance the intrinsic contractile properties of human skeletal muscle, as evidenced by increased force production at low frequencies of electrical stimulation, an increased rate of force development or a lower metabolic cost of force production. [39][40][41] A possible explanation for the improvement in muscle torque production over the first 90 s of the 5-min allout exercise test is differences in calcium (Ca 2+ ) handling or sensitivity between the NIT and PLA conditions. 42,43 Hernandez et al. 42 reported that, in mouse fast-twitch (but not slow-twitch) muscle, 7 days of NO 3 − treatment elevated myoplasmic free [Ca 2+ ] and increased contractile force at ≤50 Hz of electrical stimulation. The present study required participants to make a series of maximum isometric contractions of their knee extensors for 5 min.
In this type of all-out exercise, recruitment of muscle fibers will be near maximal at the outset with a high proportional contribution from type II (fast-twitch) fibers to torque generation. 44 is associated with greater muscle torque production may have implications for improving functional outcomes in a wide range of human populations, from those that may be compromised by senescence or disease through to elite athletes.
It was not the purpose of this study to attempt to quantify the proportional distribution of the ingested NO 3 − in muscle, plasma, saliva, urine and, by subtraction, other tissues. Doing so would require assessment of the participants' body composition and skeletal muscle mass, hematocrit and plasma volume, and total urinary output, as well as assumptions regarding uniformity of distribution of NO 3 − in all skeletal muscle based on measurements made in the vastus lateralis. It should be noted, however, that some portion of the ingested NO 3 − would have entered organs such as the liver, heart, kidney and brain. 21 Given the relatively high concentration of NO 3 − in skeletal muscle and the fact that skeletal muscle mass may represent as much as 50% of total body mass, it is clear that skeletal muscle may represent an important storage site for NO 3 − . 3 In this study, we used a K 15 NO 3 − tracer to address our experimental hypotheses. However, dietary NO 3 − supplementation most often occurs in the form of beetroot juice, and it is unclear if or how our results might have differed if NO 3 -rich beetroot juice had been consumed. We employed acute NO 3 − ingestion in the present study, and it is also unclear how longer-term NO 3 − supplementation (over several days or weeks) might have influenced our findings. Finally, our study was conducted in a young male population and further studies are required to establish whether females and older people respond similarly.

| MATERIALS AND METHODS
This study was approved by the Sport and Health Sciences Ethics Committee (University of Exeter) in line with the principles of the Declaration of Helsinki. Once the associated risks and benefits of the investigation were thoroughly explained, all participants that enrolled in the study provided written consent before taking part in any experimental procedures.

| Participants
Inclusion criteria were ostensibly healthy males and females, free of cardiovascular, respiratory, metabolic and musculoskeletal disorders, or having any contraindication to maximal exercise. Exclusion criteria included use of antibacterial mouthwash or tongue scrapers, dietary supplements, blood pressure medication, and tobacco smoking. Although females were eligible and welcome to participate in this study, only males volunteered (n = 10; age: 23 ± 4 years, height: 1.80 ± 0.07 m, body mass: 87.7 ± 8.5 kg, BMI: 26.4 ± 1.0 kg/m 2 ).

| Experimental design
In a randomized, crossover study, participants were allocated to one of two conditions, which involved the consumption of a K 15 NO 3 tracer (NIT group: 12.8 mmol, ~1300 mg NO 3 − ; 1 g/L, 99% 15 N, CK Isotopes, Desford, UK) or a equimolar potassium chloride (KCl) placebo (PLA: negligible nitrate) ( Figure 6). To minimize the number of muscle biopsies taken from participants and because we have previously shown that no changes occur in skeletal muscle [NO 3 − ] following the ingestion of PLA, 15,16 there were five biopsies during the NIT condition and two biopsies (pre and post-exercise) during the PLA condition. Participants were informed that they would provide a different number of muscle tissue samples between the two conditions, but this would be balanced across the group such that a larger number of biopsies did not indicate which condition was NIT and which condition was PLA. To control for potential variations in the participants' habitual diets, a 3-day dietary control period preceded the experimental visits. These consisted of an initial 2 day period in which participants were provided with a list of foodstuffs containing high NO 3 − and NO 2 − and asked F I G U R E 6 Schematic of experimental protocol including timings of sample collection.
to abstain from consuming them, and a final day during which they were provided with a controlled diet containing ~25-30 mg NO 3 − . The experimental visits were separated by a minimum of 7 days and a maximum of 10 days. This duration has been shown to be sufficient for muscle  ] to return to baseline values following NO 3 − supplementation. 16 On each experimental visit, participants arrived at the laboratory in a rested and fasted state at 07:30 a.m. Upon arrival, participants were asked a series of questions regarding their adherence to the prescribed diet. An initial urine sample was collected after which participants were seated on a bed and were requested to refrain from excessive movement for the remainder of the sample collection period. A saliva sample was then collected, and an intravenous cannula was inserted into the antecubital fossa and a blood sample was collected. Preparations for the muscle biopsies were completed after the initial blood samples had been processed and the initial muscle tissue sample was then collected. Following the biopsy, a low NO 3 − breakfast (two slices of toast with 10 g butter) was provided at ~08:50 a.m., and at 09:00 a.m. the K 15 NO 3 tracer was ingested in the form of a 140 ml drink. The drink was created on the morning of the visit by dissolving 1.31 g of either the K 15 NO 3 tracer (NIT, 12.8 mmol, ~1300 mg NO 3 − ) or KCl (PLA, negligible nitrate), weighed using analytical grade scales, in 140 ml deionized water. The container was vigorously shaken to ensure that the powder had fully dissolved before the participant ingested the drink. The NIT and PLA drinks were indistinguishable in appearance, smell and taste. All subsequent biological samples were collected in relation to the supplement ingestion time. The collection time of the muscle tissue extraction was at 1-and 3-h postsupplement ingestion, with saliva and blood collected before, and urine collected after, the muscle sampling was completed. In the NIT condition, the 3-h biopsy also served as the "pre-exercise" biopsy and, following the exercise protocol, two biopsies were taken (one from each leg) within 20 s of the cessation of exercise ( Figure 6). In the PLA condition, only two biopsies were taken: one at 3 h post-supplement ingestion (i.e., pre-exercise) and a second following the completion of the exercise protocol (i.e., post-exercise). A maximum of two biopsies were taken from each incision site.

| Exercise protocol
The exercise protocol involved the participants performing a series of 60 intermittent isometric maximal voluntary contractions (MVCs) of the knee extensors with muscle stimulation (i.e., the "5-min all-out test" first described by Burnley 29 ). The protocol was unilateral such that one leg served as the exercise leg and the other served as a control.
Participants were seated in the chair of a Biodex System 3 isokinetic dynamometer (Biodex Medical Systems, Shirley, NY), which was calibrated according to the manufacturer's instructions. The dominant leg was attached to the lever arm of the dynamometer with the seating position adjusted to ensure that the lateral epicondyle of the right femur was centered with the axis of rotation of the lever arm. The subjects sat with relative hip and knee angles of 85° and 90°, respectively (full extension being 0°), determined using an inclinometer. The lower leg was firmly attached to the lever arm above the ankle using a padded Velcro strap, and straps secured firmly across the waist and shoulders prevented extraneous movement during the isometric contractions. Carbon rubber electrodes (12 × 10 cm, EMS Physio, Oxfordshire, UK) coated in conductive gel were placed on the anterior thigh and secured using micropore tape. The cathode was placed on the midline of the thigh at ∼30% of thigh length measured in the seated position from the anterior superior iliac spine to the superior border of the patella, while the anode was positioned over the femoral artery. A constant-current, variable voltage stimulator (Digitimer DS7AH, Welwyn Garden City, UK) was used to deliver a series of stimuli between which the anode was adjusted until an optimal electrical stimulus or "saturation point" was achieved. Once the site had been confirmed, the anode was secured using micropore tape and a 0.5 kg sandbag was placed on top. The voltage was set at 400 V and the current was then incrementally increased until there was a plateau in the evoked twitch. The current was then increased to 130% of the plateau current and this was subsequently used to deliver a doublet stimuli (100-μs pulses, 10-ms interval). Participants were familiarized with the experimental setup and exercise protocol prior to the commencement of the main experiment.
Following a warm-up involving a series of submaximal contractions (three at 50%, two at 75% and one at 90% of estimated MVC), participants completed three MVCs, each lasting 3 s and separated by 60 s rest. The second of these contractions was performed with electrical stimulation of the muscle. Subjects were given a countdown followed by very strong verbal encouragement to maximize torque. At 1.5 s into the contraction a doublet was delivered, subjects were instructed to stop the contraction after 3 s, and another doublet was delivered 1 s after the end of the contraction. Following the third MVC, subjects rested for 10 min.
Participants then completed a 5-min all-out test involving 60 intermittent isometric MVCs (3-s contraction and 2-s rest). During the test the subjects were strongly encouraged to maximize torque during each contraction but were not informed of the elapsed time or the number of contractions remaining. The muscle electrical stimulator was triggered to deliver a doublet on the first, 15th, 30th, 45th, and 60th contraction. Stimuli were delivered 1.5 s into each of these contractions, and 1 s after each contraction.
The torque data were analyzed using Spike 2 software (Cambridge Electronic Design Ltd., Cambridge, UK). Briefly, a horizontal cursor was set on the torque axis at 15 N.m to exclude any potential false triggers and the start and end of each contraction was defined as the intersection between two additional vertical cursors with the horizontal cursor. The Spike 2 software then determined peak torque and mean torque for each 3-s contraction for both tests (i.e., NIT and PLA). The peak mean torque was defined as the 3-s contraction that yielded the highest mean torque. For any given time bin (such as over the whole 5 min test or across the first 90-s), the peak torque and the mean torque produced during each contraction was determined and used to calculate the mean peak torque and the mean mean torque, respectively. The torque impulse was calculated as the area under the torque-time curve. The potentiated doublet torque was calculated as the peak torque achieved following the doublet stimuli between contractions, and superimposed doublet torque was calculated as the increment in torque immediately following the stimuli during contraction. The end-test torque, which provides an estimate of the critical torque (CT), was operationally defined as the mean of the last six contractions in the 5-min all-out test. Voluntary activation was determined using the twitch interpolation technique. 51

| Sample collection
Muscle tissue samples were collected from the vastus lateralis muscle using a modified percutaneous Bergström needle procedure adapted for manual vacuum. 52  ]. These samples were collected in lithium heparin vacutainers (Becton Dickinson, NJ) and centrifuged at 3300 g for 7 min at 4°C. The extracted plasma was then placed in liquid nitrogen before being stored in a −80°C freezer. Two-min saliva collection periods were employed to enable participants to generate sufficient saliva before expelling it into a 30-ml universal tube (Thermo Scientific™ Sterilin™; Massachusetts, USA). The saliva was then aliquoted and placed in liquid nitrogen before being stored at −80°C. Urine samples were collected in separate containers (Kartell™; Milan, Italy) and aliquoted into micro-centrifuge tubes for storage. ] in the biological samples collected during the study using helium as the carrier gas (Sievers 280i Nitric Oxide Analyzer, GE Analytical Instruments, Boulder, CO, USA). The initial step for sample processing prior to injection into the NO analyzer was to add methanol to the plasma, urine and saliva samples (1:2 ratio by volume). These were thoroughly vortexed, left to incubate at room temperature for 30 min and subsequently centrifuged at 4°C and 11 000 g for 5 min. The supernatant was then collected and injected into the analyzer configuration. Vanadium chloride or tri-iodide solution was used for nitrate or nitrite analysis, respectively. Muscle samples were weighed and processed to ensure consistency between the sample sizes (~40-60 mg), a NO 2 − preservation solution was added (K 3 Fe(CN) 6  ] was then determined using the methods described by Park et al. 27 For all tissues, part of the supernatant was processed for UPLC-MS/MS analysis (see below) and the rest was used for NOA. 15  were directly subjected to DAN derivatization. High-performance liquid chromatography (HPLC) grade solvents and LC-MS modifiers were purchased from Sigma-Aldrich (St. Louis, MO, USA). Detection and quantification were achieved by UPLC-MS/MS utilizing a Thermo Scientific Vanquish UPLC with a Thermo Scientific Altis triple quadrupole mass spectrometer, heated electrospray ionization (HESI-II) in positive ion mode (3500 V). 50 μl of sample was mixed with 200 μl of acetonitrile (ACN), vortexed for 5 min and then centrifuged at 4°C and 17 000 g for 15 min. The supernatant was transferred to an LC-MS vial for analysis. Injection volume was 1 μl. A Waters Cortecs T3 column, 2.1 × 100 mm, 1.6 μm column was maintained at 35°C. Solvent A: H 2 O with 0.1% formic acid (FA) and Solvent B: ACN with 0.1% FA. The flow rate was 250 μl/min, the gradient was 25% B at 0 min for 0.25 min, increasing to 65% B at 5 min, further increased to 90% B at 5.5 min, remained at 90% B until 7.5 min, and then decreased to 25% B at 8 min. The total running time was 10 min. Samples were analyzed in triplicate. Quantitation of 14

| Statistical analysis
The Statistical Package for Social Scientists (SPSS Version 28, SPSS Inc., Chicago, IL, USA) was used for statistical analysis of the data. Full sets of data were available for muscle, saliva and urine (n = 10) whereas, due to technical error, data were available for plasma for nine participants. Two-way repeated measures ANOVAs were used to determine differences in [NO 3 − ] and [NO 2 − ] in muscle, plasma, saliva and urine across time (pre-exercise and post-exercise) and condition (PLA and NIT). A separate one-way repeated measures ANOVA was also run for the NIT condition for the exercise component of the study, which included a control and experimental leg. Where appropriate, significant main and interaction effects were analyzed further using least significant difference (LSD) post hoc tests. Relationships between variables were evaluated using Pearson's product moment correlation coefficients. The alpha level to denote statistical significance was p < 0.05. All results are expressed as mean ± standard deviation (SD).

| CONCLUSION
We used a stable isotope tracer (K 15  ] and the magnitude of its decrease during exercise are correlated with muscle torque production during maximal voluntary contractions of the knee extensors. These results provide new insight into the regulation of muscle contractile function by mechanisms related to NO generation and suggest that dietary NO 3 − supplementation may provide a means of enhancing human muscular performance.