Ten healthy men with a mean (SD) age of 23 (3) years, body mass of 78.0 (12.5) kg, stature of 180.3 (8.1) cm, and sea-level maximal rate of oxygen uptake (V˙O2max) of 60.9 (10.1) ml·kg⁻¹·min⁻¹ volunteered and provided fully informed written consent to participate in the present study. All participants were non-smokers, normotensive and not currently taking any medication. None of the participants had traveled to an altitude >1,500 m during the preceding 3 months and all were currently residing <500 m. The study received institutional ethical approval and adhered to the principles of the Declaration of Helsinki.

Participants attended the laboratory on five separate occasions within a 7 week period. Visits were separated by a minimum of five and maximum of 12 days, and conducted at the same time each day (±1 h) to minimize the influence of circadian variance. On the first laboratory visit, participants completed an incremental exercise test to volitional exhaustion in normoxia to elucidate V˙O2max. Following a rest period of ~20 min, participants were then familiarized with the experimental procedures. All subsequent visits involved exercise in a normobaric hypoxic chamber (Sporting Edge, Sherfield on Loddon, UK), situated ~113 m above sea-level. The F_IO₂ inside the normobaric hypoxic chamber was adjusted on a daily basis, accounting for fluctuations in barometric pressure and for 47 mmHg water vapor pressure (Conkin, 2011), to simulate either a moderate altitude (F_IO₂ ~14.5%; ~3,000 m) or very-high altitude (F_IO₂: ~11.7%; ~4,300 m). There were, in total, four different experimental conditions, which were conducted in a randomized order. These were: (1) 3,000 m simulated altitude with NO3−-rich beetroot juice (BRJ) supplementation, (2) 3,000 m simulated altitude with NO3−-deplete placebo beetroot juice (PLA) supplementation, (3) 4,300 m simulated altitude with BRJ supplementation; and, (4) 4,300 m simulated altitude with PLA supplementation. Prior to the experimental trials, participants consumed either 140 ml concentrated BRJ (~12.5 mmol NO3−) or PLA (~0.01 mmol NO3−) (Beet It, James White Ltd., Ipswich, UK) administered double blind, 2 h before arriving at the laboratory. The NO3− content of supplements was determined via ozone-based chemiluminescence, as previously described (Shannon et al., 2016). Participants were asked to avoid alcohol and caffeine consumption, and abstain from intense exercise for 24 h prior to each trial. Additionally, participants were asked to avoid antibacterial mouthwash and chewing gum throughout the study duration, given these have previously been demonstrated to destroy the oral nitrate reducing bacteria (Govoni et al., 2008). Participants gave verbal confirmation at each trial that they had fully adhered to the stipulated pre-trial controls.

A two-part incremental test was conducted on a motorized treadmill (Woodway, Cranlea, Birmingham, UK) at sea-level. During the first part of the test, participants completed four submaximal stages of 3 min duration, interspersed with 1 min recovery periods. The treadmill speed was set at 1 km·h⁻¹ for the first stage, and increased by 1 km·h⁻¹ each subsequent stage. The treadmill gradient was set to 10% for all submaximal stages, and participants were required to carry a 10 kg backpack. Gas data obtained during this first part of the test was used to calculate the V˙O2-speed relationship, and allow determination of the appropriate treadmill speed for submaximal exercise in the experimental trials. The exercise modality has previously been utilized to approximate the physiological demands of altitude hiking (Matu et al., 2017). Following four exercise stages, participants rested for ~5 min, after which the second phase of the test commenced. Participants removed the backpack, and the treadmill gradient was lowered to 1%. Participants were then required to run at a fixed speed, which was determined based around perceived fitness and eliciting an initial rating of perceived exertion (RPE) of ~12. The treadmill gradient was increased by 1% every minute until participants reached volitional exhaustion. Expired gas was monitored continuously throughout exercise via an online gas analysis system, which was calibrated prior to testing in accordance with the manufacturer's instructions (MedGraphics Ultima CPX, MGC Diagnostics, MN, USA). Gas data obtained during the second part of the test was used to determine V˙O2max (highest 30-s average in V˙O2). Following a rest period of ~20 min, participants were familiarized with the cognitive testing procedures and the 3 km walk test, as applied in the experimental trials.

Approximately 1 week after preliminary testing, participants completed the first of four experimental trials. Experimental trials commenced with ~45 min of sitting at sea-level, during which time pre-hypoxic exposure measurements were obtained. Participants then entered the normobaric hypoxic chamber, simulating an altitude of either 3,000 or 4,300 m, where they rested for a further 45 min. An exercise period then commenced, 1.5 h after participant arrival at the laboratory and 3.5 h post-supplementation. Exercise comprised 45 min of steady-state walking at 30% V˙O2max, followed immediately by a 3 km TT. An exercise intensity of 30% of sea-level V˙O2max elicited a walking speed of 2.6 (0.6) km·h⁻¹, which replicated the low intensity nature of high altitude hiking (Matu et al., 2017; Mellor et al., in press). During the TT, participants were informed of the distance completed every 500 m, but were blinded to the speed and time. All exercise was conducted at a 10% gradient whilst participants carried a 10 kg backpack. Following exercise participants rested in the normobaric hypoxic chamber for a further 30 min, after which they returned to sea-level. A schematic representation of the experimental trials protocol is presented in Figure 1.

Data were analyzed using IBM SPSS version 24 for Windows. Two-way (supplement × altitude) repeated measures ANOVA was employed to assess differences in plasma [NO2−], BP, HR, SpO₂, RPE, cognitive task performance, pulmonary gas exchange data, muscle [HbO₂], muscle [HHb], muscle [tHb], and TT performance across experimental conditions. Cerebral [HbO₂], cerebral [HHb], and cerebral [tHb] were compared during TT via two-way (supplement × altitude) ANOVA, and during each cognitive testing period via three-way (supplement × altitude × task) repeated measures ANOVA. Significant three-way interaction effects were explored using one-way ANOVA and two-way interactions were explored using paired t-tests. Statistical significance was set at p < 0.05 and Bonferroni corrections were applied to all post-hoc tests. Cohen's d is presented for information on effect sizes. Effect sizes were interpreted as ≤0.2 trivial, >0.2 small, >0.6 moderate, >1.2 large, >2 very large, and >4 extremely large (Hopkins, 2004).

A statistical spreadsheet was also employed to derive qualitative probabilistic inferences for TT performance data (Hopkins, 2007). Verbal descriptors were used to express the chance that the true value of the effect was beneficial, trivial, or harmful, according to the following scale: <0.5%, “almost certainly not”; 0.5–5%, “very unlikely not”; 5–25%, “unlikely”; 25–75%, “possibly”; 75–95%, “likely”; 95–99.5%, “very likely”; >99.5%, “almost certainly.” The effect was deemed unclear when an odds ratio of benefit to harm of <66 was identified, corresponding to a 25% chance of benefit and 0.5% risk of harm. Data are presented as means (SD) unless otherwise stated. Individual responses are presented within figures where appropriate to allow further exploration of the findings.

Power calculations were conducted using G^*Power (Faul et al., 2007) to estimate an appropriate sample size. Based on previously published data from our laboratory (Shannon et al., 2016), power analysis indicated that 10 participants would provide >90% power to detect a 3.2% change in TT performance with an alpha level of 0.05.

Prior to simulated altitude exposure, plasma [NO2−] was 630.9 (253.7), 203.9 (70.6), 555.3 (171.9), and 238.3 (125.3) nM for BRJ-3000m, PLA-3000m, BRJ-4300m, and PLA-4300m, respectively. Plasma [NO2−] was significantly elevated in BRJ compared with PLA (main effect of supplement, p = 0.001, d = 2.22). There was no difference in plasma [NO2−] prior to 3,000 and 4,300 m simulated altitude exposure (main effect of altitude, p = 0.612, d = 0.08). Likewise, there was no supplement × altitude interaction effect for plasma [NO2−] (p = 0.089). Pooled plasma [NO2−] data for the two BRJ trials and two PLA trials is presented in Figure 2.

Prior to simulated altitude exposure, mean arterial pressure (MAP) was 88 (8), 92 (4), 88 (6), and 92 (11) mmHg for BRJ-3000m, PLA-3000m, BRJ-4300m, and PLA-4300m, respectively. MAP did not differ significantly between supplements (p = 0.323, d = 0.41) or altitudes (p = 0.687, d = 0.07). Likewise, there was no supplement × altitude interaction effect for MAP (p = 0.912).

Resting SpO₂ and HR were not significantly different between conditions prior to hypoxic exposure (all effects p ≥ 0.110, d ≤ 0.26, Table 1). There was a significant main effect of supplement on SpO₂ during the pre-exercise rest period at simulated altitude and steady-state exercise period (both p ≤ 0.017, d ≥ 0.28) reflecting an elevated SpO₂ overall in BRJ compared with PLA at these measurement points. During simulated altitude exposure, SpO₂ was lower at all measurement points at 4,300 m compared with 3,000 m simulated altitude (main effect of altitude, all p ≤ 0.002, d ≥ 1.37). There were no supplement × altitude effects at any measurement point (all p ≥ 0.127).

HR was significantly higher at 4,300 m compared with 3,000 m during the pre-exercise rest period at simulated altitude and steady-state exercise (main effect of altitude, both p ≤ 0.026, d ≥ 0.38). HR was not significantly different between BRJ and PLA at any measurement point (main effect of supplement, all p ≥ 0.132, d ≤ 0.32). No supplement × altitude interactions were detected for HR (all p ≥ 0.072).

During steady-state exercise, RPE was 10 (2), 10 (2), 12 (1), and 11 (2) for BRJ-3000m, PLA-3000m, BRJ-4300m, and PLA-4300m, respectively. RPE did not differ significantly between supplements (p = 0.382, d = 0.18), but was significantly greater at 4,300 m compared with 3,000 m (main effect of altitude, p = 0.007, d = 0.93). There was no supplement × altitude interaction effect (p = 1.000). During TT exercise, RPE was 19 (2), 19 (2), 19 (2), and 19 (1) for BRJ-3000m, PLA-3000m, BRJ-4300m, and PLA-4300m, respectively. RPE did not differ significantly between supplements (p = 0.604, d = 0.06) or altitudes (p = 0.678, d = 0.06) during the TT. Likewise, there was no supplement × altitude interaction effect for RPE (p = 0.879).

Data for the cognitive test results is located in the Supplementary data Table 1. There was a tendency toward a reduced number of false alarms in the RVP task in BRJ compared with PLA prior to altitude exposure, although this did not attain statistical significance (main effect of supplement, p = 0.056, d = 0.30). There was no significant difference in cognitive function for other cognitive tests between BRJ and PLA at all measurement points (main effect of supplement, all p ≥ 0.141, d ≤ 0.59). There was a significant decline in cognitive function at 4,300 m compared with 3,000 m during steady-state exercise for AST number of correct responses, AST response time, and RVP response time (main effect of altitude, all p ≤ 0.041, d ≥ 0.32). There was also a significant decline in cognitive function at 4,300 m compared with 3,000 m during the post-exercise measurement period for AST response time, RVP number of correct responses, RVP response time, and RVP false alarms (main effect of altitude, all p ≤ 0.016, d ≥ 0.48). A tendency toward a reduced number of correct responses at 4,300 m compared with 3,000 m was also detected during the post-exercise period for AST, although this did not attain significance (p = 0.053, d = 0.51). Significant supplement × altitude interaction effects were detected for AST number of correct responses during steady-state exercise (p = 0.032), AST response time during the pre-exercise rest period (p = 0.019), and SSP mean number of attempts to pass during the post-exercise rest period (p = 0.040). However, none of these achieved statistical significance during post-hoc analysis once appropriate adjustments were made for multiple comparisons (p ≥ 0.325, d ≤ 0.84).

Muscle TSI did not differ significantly between BRJ and PLA during steady-state exercise (main effect of supplement, p = 0.139, d = 0.15), but was significantly higher in BRJ compared with PLA during TT exercise (main effect of supplement, p = 0.011, d = 0.71, Figure 3A). In the post-exercise rest period, muscle TSI was significantly greater at 3,000 m compared with 4,300 m (p = 0.020, d = 0.83). No supplement × altitude interaction effects were observed (p ≥ 0.656).

Muscle O₂Hb did not differ significantly between supplements (all p ≥ 0.525, d ≤ 0.17, Figure 3B) or altitudes (all p ≥ 0.087, d ≤ 0.53) at any measurement point. Likewise, there were no supplement × altitude interaction effects (all p ≥ 0.205).

Muscle HHb did not differ significantly between BRJ and PLA at any measurement point (all p ≥ 0.251, d ≤ 0.34, Figure 3C). Muscle HHb was significantly greater at 4,300 m compared with 3,000 m during steady-state exercise (p = 0.007, d = 0.52), and tended to be greater at 4,300 m compared with 3,000 m during the post-exercise rest period (p = 0.051, d = 0.38). No significant supplement × altitude interaction effects were observed (all p ≥ 0.307).

Muscle tHb did not differ significantly between supplements (all p ≥ 0.360, d ≤ 0.29, Figure 3D) or altitudes (all p ≥ 0.436, d ≤ 0.20), and no supplement × altitude interaction effects were detected (all p ≥ 0.209) at any measurement point.

Cerebral O₂Hb tended to be greater in PLA compared with BRJ during the TT (main effect of supplement, p = 0.056, d = 0.55, Figure 4A), but did not differ significantly between supplements at other measurement points (all p ≥ 0.487, d ≤ 0.17). Cerebral O₂Hb was significantly greater during steady-state and TT exercise at 3,000 m compared with 4,300 m simulated altitude (main effect of altitude, both p ≤ 0.001, d ≥ 0.88). A significant effect of task was identified during steady-state exercise (p < 0.001), reflecting an increase in cerebral O₂Hb during cognitive tasks compared with values obtained in the 5 min walking period prior to cognitive tasks. A significant supplement × altitude × task interaction effect was also detected during steady-state exercise (p = 0.034), however no significant effects were identified during post-hoc analysis following adjustments for multiple comparisons (all p ≥ 0.909, d ≤ 0.71).

Cerebral HHb was significantly greater in PLA compared with BRJ during TT exercise (main effect of supplement, p = 0.047, d = 0.58, Figure 4B), and tended to be greater during steady-state exercise in PLA compared with BRJ (p = 0.053, d = 0.53). Cerebral HHb was greater at 4,300 m compared with 3,000 m during steady-state exercise and the post-exercise rest period (both p ≤ 0.034, d ≥ 0.41). A significant supplement × altitude interaction effect was observed during the pre-exercise rest period (p = 0.047). Post-hoc analysis revealed greater cerebral HHb in PLA compared with BRJ at 3,000 m during the pre-exercise rest period (p < 0.001, d = 0.66).

Cerebral tHb was greater in PLA compared with BRJ during TT exercise (main effect of supplement, p = 0.043, d = 0.64, Figure 4C). There was no difference in cerebral tHb between altitudes (p ≥ 0.064, d ≤ 0.42). A significant effect of task was identified during steady-state exercise (p = 0.016), reflecting the greater tHb during cognitive tasks relative to the 5 min pre-testing walking period. There was also a significant supplement × altitude interaction during the pre-exercise rest period (p = 0.021). Post-hoc analysis revealed greater tHb in PLA compared with BRJ at 3,000 m simulated altitude during the pre-exercise rest period (p < 0.001, d = 0.63).

Completion times for the 3 km TT were 1,653.9 (261.3), 1,718.7 (213.0), 1,809.8 (262.0), and 1,889.1 (203.9) s for BRJ-3000m, PLA-3000m, BRJ-4300m, and PLA-4300m respectively. There was a significant main effect of supplement (p = 0.019, d = 0.31, Figure 5), reflecting faster TT completion time overall in BRJ compared with PLA. TT completion times were faster at 3,000 m compared with 4,300 m (main effect of altitude, p < 0.001, d = 0.72). No supplement × altitude interaction effect was detected (p = 0.890). Individual threshold values of 42.6 s and 40.8 s were calculated for magnitude-based inferences at 3,000 and 4,300 m, respectively, based around the standardized effect size of 0.2 multiplied by the between subject standard deviation of the placebo trial at each altitude. Magnitude-based inferences indicated a possibly beneficial (74.7%), unlikely trivial (24.9%), most unlikely harmful (0.4%) effect of BRJ on TT performance compared with PLA at 3,000 m. Additionally, the effect of BRJ on TT performance at 4,300 m altitude was deemed likely beneficial (81.2%), unlikely trivial (17.9%), very unlikely harmful (0.9%) compared with PLA.

In the present investigation, plasma [NO2−] was significantly elevated consequent to NO3− supplementation, which substantiates previous research in this area (Webb et al., 2008; Bailey et al., 2009; Vanhatalo et al., 2010, 2011; Wylie et al., 2013; Shannon et al., 2016, 2017). NO generation via L-arginine appears to be blunted in hypoxic conditions (Shaul et al., 1993; Fish et al., 2007). Conversely, the reduction of NO2− into NO is potentiated at low oxygen tensions (Castello et al., 2006). Consequently, the increase plasma [NO2−] observed following NO3− supplementation may be especially effective at maintaining or enhancing NO bioavailability and signaling in hypoxic conditions, including the moderate and very-high altitude environments simulated in this study.

Elevated NO bioavailability following NO3− supplementation might be expected to lower BP, due to NO-dependent relaxation of smooth muscle and subsequent vasodilation (Larsen et al., 2006; Webb et al., 2008; Vanhatalo et al., 2010; Siervo et al., 2013, 2015). However, in the present study there was no apparent difference in resting MAP assessed prior to simulated altitude exposure between BRJ and PLA. This is also similar to the findings of a recent study by our group, in which trained runners/triathletes showed no change in MAP following administration of an identical NO3− dose, despite substantial increases in plasma [NO2−] (Shannon et al., 2017). Recently, Ashworth et al. (2015) reported a significant positive relationship between baseline BP and the change in BP following ingestion of a high-NO3− diet. Thus, it seems plausible that the already low BP in our participants limited the capacity to further reduce BP following NO3− supplementation.

In accordance with multiple previous investigations in both normoxia (Larsen et al., 2007; Bailey et al., 2009; Vanhatalo et al., 2010; Pawlak-Chaouch et al., 2016) and hypoxia (Masschelein et al., 2012; Kelly et al., 2014; Muggeridge et al., 2014; Shannon et al., 2016), NO3− supplementation reduced V˙O2 during steady-state exercise. The novelty in our findings is that we demonstrate this reduction in V˙O2 during simulated altitude hiking, a finding with practical relevance for the thousands of individuals who ascend to altitude for this purpose each year. This was accompanied by an elevated SpO₂ during steady-state exercise, and a greater muscle TSI during TT exercise consequent to NO3− supplementation. Collectively these data might indicate a reduced tissue oxygen extraction during exercise with NO3− supplementation. NO3− ingestion does not appear to “shift” metabolism toward non-oxidative means of ATP resynthesis (Bailey et al., 2010), and did not alter substrate oxidation in this study as indicated by similar RER values in BRJ and PLA. Therefore, it is likely that these effects reflect a genuine improvement in exercise efficiency (i.e., less oxygen is required for a given work rate). Mechanistically, this could be explained by an improved efficiency of muscle contraction, such that the ATP and oxygen requirements for a particular rate of force generation are reduced (Bailey et al., 2010). Alternatively, NO3− supplementation has been reported to improve the efficiency of mitochondrial ATP resynthesis, an effect likely attributable to reduced proton leak through the ATP/ADP translocase protein (ANT) and possibly uncoupling protein 3 (UCP3), thus lowering the oxygen cost of ATP regeneration (Larsen et al., 2011). However, recent data question whether mitochondrial efficiency is improved with beetroot juice, or whether this effect might be restricted to sodium-NO3− supplementation (Whitfield et al., 2015). Contrasting the mechanisms of action of these two NO3− vehicles may be a potential avenue for future research (Kemp, 2016).

Interestingly, in a study by Bourdillon et al. (2015), NO3− supplementation increased pulmonary ventilation and elevated SpO₂ during exercise in hypoxia. A higher ventilation has previously been associated with increased SpO₂ (Benoit et al., 1995). Thus, increased ventilation could contribute toward the elevated SpO₂ with NO3− supplementation in hypoxia. Ventilation-related increases in SpO₂ might also be expected to increase muscle oxygenation, given an elevated SpO₂ would increase the pressure gradient driving oxygen diffusion from the blood to the tissue (Masschelein et al., 2012). However, the increased SpO₂ (during steady-state exercise) and elevated muscle TSI (during TT exercise) occurred during different phases of the exercise test in this study, thus questioning the link between these variables. An acknowledged limitation of this study is that we did not obtain ventilation data, and as such it is unclear whether this parameter was altered following NO3− supplementation.

Some (Gilchrist et al., 2014; Thompson et al., 2015, 2016; Wightman et al., 2015) but not all (Kelly et al., 2013; Bondonno et al., 2014; Thompson et al., 2014) studies conducted at sea-level have reported beneficial effects of NO3− supplementation on cognitive function. Prior to this investigation, to the authors' knowledge only one study had evaluated the effects of NO3− supplementation on cognitive function at simulated altitude (Lefferts et al., 2016). Lefferts et al. (2016) reported a significant decline in short-term memory, information processing efficiency, and emotional recognition during rest at simulated altitude (F_IO₂: 11.6%; 4,600 m) compared with sea-level. However, there was no difference in cognitive performance between NO3− and placebo conditions. Likewise, middle cerebral artery blood flow, an index of neurovascular coupling, was unchanged following NO3− supplementation. In the present study, cognitive function was compromised during exercise and the post-exercise rest period at 4,300 m compared with 3,000 m simulated altitude, as indicated by either slower response times or a decreased number of correct responses in the AST and RVP tasks. Resting cognitive function did not differ between supplements at simulated altitude, which is in accordance with the findings of Lefferts et al. (2016). Extending these findings, the present study demonstrated no effect of NO3− supplementation on cognitive function during or following exercise.

The previously reported beneficial effects of NO3− supplementation on cognitive function may be related to alterations in cerebral blood flow, in particular by ensuring optimal delivery of blood (and thus oxygen) to match neural activity (i.e., enhanced neurovascular coupling) (Wightman et al., 2015). In elderly individuals, NO3− supplementation enhanced perfusion to areas of the brain known to play a role in executive function (Presley et al., 2011). Furthermore, NO3− supplementation has been reported to facilitate a faster, smaller, and more homogenous haemodynamic response to visual stimulation (Aamand et al., 2013), and result in an initial increase in cerebral blood flow during cognitive tasks, followed by a consistent reduction in cerebral blood flow during less demanding cognitive tasks (Wightman et al., 2015). In the latter study, these blood flow changes also coincided with improved cognitive performance (Wightman et al., 2015). We observed no difference in pre-frontal cortex [tHb] changes between NO3− and placebo supplementation during the cognitive testing periods. Interestingly, however, NO3− supplementation decreased the change in cerebral [tHb] during the TT, which might be a consequence of lower blood flow to the pre-frontal cortex during the TT in BRJ compared with PLA. Cerebral blood flow changes at altitude are largely determined by four key reflex mechanisms. These are the hypoxic ventilatory response, hypercapnic ventilatory response, hypoxic cerebral vasodilation, and hypocapnic cerebral vasoconstriction (Ainslie and Subudhi, 2014). It is possible that one or more of these variables may have been influenced by NO3− supplementation, with a concomitant effect on cerebral blood flow.

As mentioned, Bourdillon et al. (2015) observed an increase in the hypoxic ventilatory response following NO3− supplementation, which might result in a lower cerebral blood flow by elevating arterial oxygen tensions and decreasing arterial carbon dioxide tensions (Ainslie and Subudhi, 2014). Although we did not assess ventilation in this study, it seems unlikely that a ventilation-related increase in arterial oxygen content accounts for the lower [tHb] in BRJ compared with PLA, given SpO₂ did not differ between supplements during the TT. A reduction in arterial carbon dioxide tensions could result in cerebral vasoconstriction and hence reduced cerebral blood flow, accounting for the reduced [tHb] during the TT with NO3− supplementation. Arterial carbon dioxide tension was not evaluated in the present study. However, Bourdillon et al. (2015) previously reported no effects of NO3− supplementation on arterial carbon dioxide pressure or end tidal carbon dioxide, which brings into question this mechanism. Alternatively, it is possible that differences in cerebral [tHb] resulted from differences in cerebral activation as a consequence of the higher mean power in BRJ (given time to complete the 3 km TT was lower) compared with PLA (Brümmer et al., 2011).

Cerebral [HHb] was significantly lower during TT exercise, and tended to be lower during steady-state exercise (which involved simultaneous cognitive tasks) following NO3− supplementation. A significant supplement × altitude interaction effect was also observed during the pre-exercise rest period, indicating reduced changes in cerebral [HHb] specifically at 3,000 m simulated altitude consequent to NO3− supplementation. This may indicate a lower cerebral oxygen extraction in the pre-frontal cortex following NO3− supplementation during these measurement periods. Alternatively, as both cerebral [O₂Hb] and [tHb] were also lower during the TT with BRJ, these findings may reflect a smaller volume of blood in the area under investigation, possibly as a consequence of differences in cerebral blood flow as discussed above. Interestingly, and in contrast to our findings, Masschelein et al. (2012) reported no change in cerebral [O₂Hb], [HHb], or [tHb] in conditions of extreme simulated altitude (F_IO₂ 11%; 5,000 m) with NO3− supplementation. This discordance in findings could be related to the different altitudes simulated in this study and by Masschelein et al. (2012). In hypoxia, cerebral blood flow increases to ensure adequate delivery of oxygen to the brain, with greater changes in blood flow observed at higher compared with lower altitudes (Ainslie and Subudhi, 2014). Thus, it is possible that NO3− may be ineffective at influencing cerebral oxygenation/ blood flow at extreme but not moderate or high altitude, given the comparably greater hyperaemic response in the former.

The effects of NO3− supplementation on exercise TTE and/or TT performance at simulated altitude have been evaluated in a number of previous studies, with some (Vanhatalo et al., 2011; Masschelein et al., 2012; Kelly et al., 2014; Muggeridge et al., 2014; Shannon et al., 2016) but not all (Arnold et al., 2015; Bourdillon et al., 2015; MacLeod et al., 2015) reporting beneficial effects. In contrast to previous studies which have employed high-intensity leg-extension (Vanhatalo et al., 2011), cycle ergometry (Masschelein et al., 2012; Kelly et al., 2014; Muggeridge et al., 2014; Bourdillon et al., 2015; MacLeod et al., 2015), or treadmill running (Arnold et al., 2015; Shannon et al., 2016) exercise protocols at simulated altitude, we evaluated the effects of NO3− supplementation using a protocol which more closely replicates the demands of altitude hiking. For this purpose, our participants were required to complete a 3 km uphill (10% gradient) treadmill test whilst carrying a weighted (10 kg) backpack. We observed a 3.8 and 4.2% improvement in performance in BRJ compared with PLA at 3,000 and 4,300 m simulated altitude, respectively. The effects of BRJ on 3 km TT performance were deemed to be “possibly beneficial” and “likely beneficial” at 3,000 and 4,300 m simulated altitude, respectively (i.e., similar beneficial changes at both simulated altitudes). BRJ supplementation was also deemed “most unlikely harmful” and “very unlikely harmful” (i.e., both <1% chance harmful) for performance at 3,000 and 4,300 m simulated altitude respectively, suggesting adverse performance effects are unlikely with BRJ supplementation. Thus, BRJ supplementation may have potential beneficial applications for individuals hiking at altitude, perhaps in situations where they are required to cover a particular distance quickly, such as in response to changing weather conditions or to ensure completion of a hike within daylight hours.

Several mechanisms have been suggested to explain the beneficial effects of NO3− supplementation on exercise performance. In particular, and as mentioned, NO3− has been proposed to enhance the efficiency of muscle contraction (Bailey et al., 2010) and mitochondrial respiration (Larsen et al., 2011). Such effects might account for the lower V˙O2 and elevated SpO₂ observed in the present study following NO3− supplementation, and may also be expected to enhance performance by allowing a higher work rate to be maintained for a given V˙O2.

The effects of NO3− supplementation on muscle and cerebral oxygenation might also be relevant from a performance perspective. The greater muscle TSI observed during TT exercise with NO3−, reflective of greater muscle oxygenation, could be beneficial by enhancing rates of oxidative ATP resynthesis, and limiting phosphocreatine breakdown and rates of anaerobic glycolysis. This might be beneficial by attenuating the accumulation of fatigue associated metabolites, such as H⁺, P_i, and ADP (Vanhatalo et al., 2011). Furthermore, it has been suggested that cortical deoxygenation during exercise might limit performance by compromising executive functions and contributing toward the decision to cease exercise, with this effect particularly apparent in hypoxia (Subudhi et al., 2009). It is therefore possible that the beneficial effects of NO3− supplementation on performance might also partly be accounted for by the reduced cerebral deoxygenation during TT exercise.

The present study makes an important contribution toward our understanding of the applications of NO3− supplementation for individuals exercising at altitude. Notably, our data indicate potential physiological and performance benefits of NO3− supplementation during hiking, which is a popular form of activity amongst individuals ascending to altitude. It is particularly promising that these effects occurred following acute NO3− consumption, such that individuals ascending to altitude may not require prolonged time consuming and costly supplementation strategies to elicit potentially beneficial effects. Nevertheless, certain limitations should be acknowledged.

Firstly, our research was conducted in a normobaric hypoxic chamber which simulated the low PO₂ of altitude by reducing the F_IO₂. This exercise environment allowed greater control over potentially confounding variables vs. testing at terrestrial altitude, and helped minimize participant burden. Recent work has also reported broadly similar responses to normobaric hypoxia and terrestrial altitude (Woods et al., 2017), suggesting that normobaric hypoxia is a reasonable surrogate for the genuine high-altitude environment. Nevertheless, further research is warranted to confirm whether the beneficial effects of NO3− supplementation observed in the present investigation also manifest at terrestrial altitude. It is noteworthy that one recent study reported improvements in flow mediated dilation (FMD) following NO3− ingestion during a trek to 3,700 m terrestrial altitude (Bakker et al., 2015), suggesting the potential capacity to alter some physiological parameters at genuine high altitude with this supplement. The relatively modest sample size of the present study may also be regarded as a limitation. However, our number of participants is similar to several previous investigations that have reported significant beneficial effects of NO3− supplementation (e.g., Bailey et al., 2009; Lansley et al., 2011; Breese et al., 2013; Wylie et al., 2013).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.