Acute Beetroot Juice Supplementation Enhances Intermittent Running Performance but Does Not Reduce Oxygen Cost of Exercise among Recreational Adults

Nitrate (NO3−) supplementation has been reported to enhance intermittent exercise performance; however, its impact on oxygen (O2) cost during intermittent running exercise is unclear. The aim of this study was to assess if acute NO3− supplementation would elicit performance benefits in recreationally active individuals during the Yo–Yo intermittent recovery level 1 (Yo-Yo IR1) test, with its potential benefit on O2 consumption (VO2), in a double-blind, randomized, crossover study, 12 recreational males consumed NO3−-rich (NIT; ~12.8 mmol), and NO3−-depleted (PLA; 0.04 mmol) concentrated beetroot juice 3 h before completing the Yo-Yo IR1 test. VO2 was measured at 160, 280 and 440 m (sub-maximal) and when the test was terminated (peak). Performance in the Yo–Yo IR1 was greater with NIT (990 ± 442.25 m) compared to PLA (870 ± 357.4 m, p = 0.007). The VO2 was not significantly different at 160 m (1.92 ± 0.99 vs. 2.1 ± 0.88 L·min−1), 280 m (2.62 ± 0.94 vs. 2.83 ± 0.94 L·min−1), 440 m (3.26 ± 1.04 vs. 3.46 ± 0.98 L·min−1) and peak (4.71 ± 1.01 vs. 4.92 ± 1.17 L·min−1) between NIT and PLA trials (all p > 0.05). The present study has indicated that acute supplementation of NO3− enhanced intermittent running performance but had no effect on VO2 during the Yo–Yo IR1 test in recreational young adults.


Introduction
The ergogenic effect of dietary nitrate (NO 3 − ) supplementation is attributed to its reduction of NO 3 − to nitrite (NO 2 − ) and, subsequently, nitric oxide (NO) [1]. This ingestion of NO 3 − -rich sources is known to increase plasma NO 2 − and be beneficial for reducing the oxygen (O 2 ) cost for a given workload [2][3][4][5], improving muscle contractile properties [6,7], and supporting fatigue resistance [8][9][10]. Interestingly, it has been shown that a vegetable source is more effective than NO 3 − salts [11], taken as a supplement (e.g., concentrated beetroot juice). Furthermore, existing evidence also supports the notion that the reduction of NO 2 − to NO is enhanced in intra-muscular hypoxic conditions such as that observed within the skeletal muscle during high-intensity activity [12,13].The potential benefits of NO 3 − supplementation were also shown in muscle contractile properties (e.g., evoked contractile force), and these effects seem in preferentially type II compared to type I muscle fibers [14]. This enhancement in muscle contractility after NO 3 − supplementation has been attributed to improved calcium handling and release [6,7] and improved skeletal muscle blood flow and vascular conductance during submaximal efforts [15]. As such, enhanced intermittent running performance in moderately- [16,17] and well-trained individuals [18] following NO 3 − supplementation might be associated with the potential type II fibers' The sample size of this study was based on a priori calculation using G*Power software (version 3.1.9.4, Universität, Düsseldorf, Germany). In determining the minimal estimated sample, we have considered two key outcomes, namely total distance achieved and the difference in overall change of VO 2 . Considering total distance, a standardized mean difference of 1.03 was used based on the work of Nyakayiru et al. [18]. Considering the difference in change of VO 2 , a standardized mean difference of 0.93 was used based on the work of Bailey et al. [8]. In both instances, a t-test family was used with matched pairs, a power of 0.80, a two-tailed approach, and α was set at 0.05. The results from these estimations indicated that sample of 10-12 participants would be sufficient to detect a difference between NO 3 − (NIT) and placebo (PLA) supplementation. Twelve recreationally active males (mean ± SD: age 27 ± 10 years, body mass 78.1 ± 11.8 kg, stature 180.5 ± 5.3 cm) were recruited for this study. All participants were involved in regular moderate-intensity exercise~3 days per week and muscle-strengthening activities~2 days per week. The participants were non-smokers, healthy, and did not use dietary supplements at the time of Nutrients 2022, 14, 2839 3 of 10 data collection. All participants were university students in the Sport Science department and were familiar with the Yo-Yo IR1 test. Ethics approval for this study was given by the Manchester Metropolitan University Research Ethics Committee (Reference no: 33132). All participants were informed of the nature and possible risks of the experimental procedures before providing written informed consent.

Experimental Design
The participants visited the testing facility on two separate occasions. Participants were assigned in a randomized, double-blind, placebo-controlled, crossover design to consume either a NO 3 − -rich (NIT) or a NO 3 − -depleted concentrated beetroot juice (PLA). The experimental trials were all carried out at the similar time of the day (±2 h). Mean and standard deviation of ambient temperature, humidity and pressure during the two trials were 17 ± 2.1 • C, 56.0 ± 4.3% and 1018 ± 2 mbar, respectively. A four-to six-day washout period separated the supplementation periods, as suggested by Wylie et al. [9]. Each participant was asked to record their dietary intake in the 24 h before the first experimental trial and replicate this in the 24 h before the subsequent trial. Participants were instructed to avoid strenuous exercise and the consumption of alcohol and caffeine for at least 24 h before each experimental trial. Participants avoided using antibacterial mouthwash throughout the duration of the study due to its prevention of the reduction of NO 3 − to NO 2 − in the oral cavity [29].

Procedures
Upon arrival at the testing facility (an indoor fourth-generation artificial grass pitch), participants completed a standardized warm-up (10 min), ending with the first two shuttles of the Yo-Yo IR1 test in order to familiarize themselves with the audio and initial speeds. After 10 min of passive recovery, a resting capillary blood lactate (BLa) sample was taken from the pad of the index finger of the left hand using the Lactate Pro-2 (Lactate Pro analyser, Arkay, Kyoto, Japan). Immediately after the Yo-Yo IR1, a second capillary BLa sample was taken.
Participants had an online gas analyzer fitter using a custom-made harness with the device positioned on their back. Pulmonary gas exchange was measured continuously using a Cosmed K5 (Cosmed, K5, Cosmed, Rome, Italy) with the system set for breathby-breath analysis. The calibration of the K5 gas analyzer was performed before each test, according to the manufacturer's instructions including a gas, volume, carbon dioxide (CO 2 ) and breathing frequency. Breath-by-breath VO 2 , CO 2 production (VCO 2 ) and minute ventilation (VE) data from each test were linearly interpolated to provide second-by-second values. Subsequently, mean VO 2 , VCO 2 and VE were assessed during each run and recovery period and averaged to provide the overall mean VO 2 , VCO 2 and VE during the run and recovery periods for each stage of the Yo-Yo IR1 test. The mean sub-maximal values of 160, 280 and 440 m were based on those previously used [30]. Peak values for each variable were considered as the highest value achieved during the test. Previous literature has reported that the COSMED K5 had excellent reliability for VO 2 (CV: 4.4%, CI: 3.2-6.7%, concordance correlation coefficient [CCC]: 0.95), VCO 2 (CV: 6.2%, CI: 4.5-9.7%, CCC: 0.92) and VE (CV: 6.9%, CI: 4.9-10.7%, CCC: 0.89) during more than 2 h of continuous field tests [31]. The Yo-Yo IR1 test has been described elsewhere [20]. Briefly, the test consists of repeated 2 × 20 m runs, interspersed by a 10 s active recovery period, at progressively increasing speeds controlled by audio bleeps from a portable audio system. First four shuttles were at the speed of 10-13 km·h −1 (0-160 m), then three shuttles at 13.5 km·h −1 (200-280 m) and four shuttles at 14.0 km·h −1 (320-440 m); thereafter, the speed increased 0.5 km·h −1 every eight shuttles (i.e., 760, 1080, 1400 m, etc.). The final distance successfully covered was recorded after the second failed attempt to meet the start/finish line in the allocated time.

Statistical Analysis
All data were presented as means ± SD. Differences between NIT and PLA in distance covered during the Yo-Yo IR1 test, VO 2peak , VCO 2peak and VE peak , RER peak and BLa preand post-exercise test were analyzed using a paired samples t-test. Effect sizes (d) were calculated through Cohen's d as: large d > 0.8, moderate d = 0.8 to 0.5, small d = 0.5 to 0.2, and trivial d < 0.2 [32]. Differences in VO 2 , VCO 2 , VE and RER at 160, 280 and 440 Im were determined using two-way repeated-measure ANOVA (supplement × distance). In addition, effect size was calculated as partial eta-squared ( CCC: 0.92) and VE (CV: 6.9%, CI: 4.9-10.7%, CCC: 0.89) during more than 2 h of continuous field tests [31]. The Yo-Yo IR1 test has been described elsewhere [20]. Briefly, the test consists of repeated 2 × 20 m runs, interspersed by a 10 s active recovery period, at progressively increasing speeds controlled by audio bleeps from a portable audio system. First four shuttles were at the speed of 10-13 km·h −1 (0-160 m), then three shuttles at 13.5 km·h −1 (200-280 m) and four shuttles at 14.0 km·h −1 (320-440 m); thereafter, the speed increased 0.5 km·h −1 every eight shuttles (i.e., 760, 1080, 1400 m, etc.). The final distance successfully covered was recorded after the second failed attempt to meet the start/finish line in the allocated time.

Statistical Analysis
All data were presented as means ± SD. Differences between NIT and PLA in distance covered during the Yo-Yo IR1 test, VO2peak, VCO2peak and VEpeak, RERpeak and BLa pre-and post-exercise test were analyzed using a paired samples t-test. Effect sizes (d) were calculated through Cohen's d as: large d > 0.8, moderate d = 0.8 to 0.5, small d = 0.5 to 0.2, and trivial d < 0.2 [32]. Differences in VO2, VCO2, VE and RER at 160, 280 and 440 Im were determined using two-way repeated-measure ANOVA (supplement × distance). In addition, effect size was calculated as partial eta-squared (ŋp 2 ) varying small (<0.25), medium (0.26-0.63) and large (>0.63) [33]. All data were analyzed using SPSS 27.0 (IBM Corp., Armonk, NY, USA). Significance was determined at p < 0.05.

Results
The distance covered in the Yo-Yo IR1 test was significantly greater in NIT (990 ± 442.25 m) compared to PLA (870 ± 357.4 m, p = 0.007, d = 0.30, Figure 1). The group mean relative and absolute VO2 and absolute VCO2 responses at 160m, 280m, 440m and peak during the Yo-Yo IR1 following both NIT and PLA supplementation were shown in Figure 2, and values (including VE and RER values) were reported in Table 1. The mean relative VO2 responses at submaximal distances were similar for NIT p 2 ) varying small (<0.25), medium (0.26-0.63) and large (>0.63) [33]. All data were analyzed using SPSS 27.0 (IBM Corp., Armonk, NY, USA). Significance was determined at p < 0.05.

Results
The distance covered in the Yo-Yo IR1 test was significantly greater in NIT (990 ± 442.25 m) compared to PLA (870 ± 357.4 m, p = 0.007, d = 0.30, Figure 1). CCC: 0.92) and VE (CV: 6.9%, CI: 4.9-10.7%, CCC: 0.89) during more than 2 h of continuous field tests [31]. The Yo-Yo IR1 test has been described elsewhere [20]. Briefly, the test consists of repeated 2 × 20 m runs, interspersed by a 10 s active recovery period, at progressively increasing speeds controlled by audio bleeps from a portable audio system. First four shuttles were at the speed of 10-13 km·h −1 (0-160 m), then three shuttles at 13.5 km·h −1 (200-280 m) and four shuttles at 14.0 km·h −1 (320-440 m); thereafter, the speed increased 0.5 km·h −1 every eight shuttles (i.e., 760, 1080, 1400 m, etc.). The final distance successfully covered was recorded after the second failed attempt to meet the start/finish line in the allocated time.

Statistical Analysis
All data were presented as means ± SD. Differences between NIT and PLA in distance covered during the Yo-Yo IR1 test, VO2peak, VCO2peak and VEpeak, RERpeak and BLa pre-and post-exercise test were analyzed using a paired samples t-test. Effect sizes (d) were calculated through Cohen's d as: large d > 0.8, moderate d = 0.8 to 0.5, small d = 0.5 to 0.2, and trivial d < 0.2 [32]. Differences in VO2, VCO2, VE and RER at 160, 280 and 440 Im were determined using two-way repeated-measure ANOVA (supplement × distance). In addition, effect size was calculated as partial eta-squared (ŋp 2 ) varying small (<0.25), medium (0.26-0.63) and large (>0.63) [33]. All data were analyzed using SPSS 27.0 (IBM Corp., Armonk, NY, USA). Significance was determined at p < 0.05.

Results
The distance covered in the Yo-Yo IR1 test was significantly greater in NIT (990 ± 442.25 m) compared to PLA (870 ± 357.4 m, p = 0.007, d = 0.30, Figure 1). The group mean relative and absolute VO2 and absolute VCO2 responses at 160m, 280m, 440m and peak during the Yo-Yo IR1 following both NIT and PLA supplementation were shown in Figure 2, and values (including VE and RER values) were reported in Table 1. The mean relative VO2 responses at submaximal distances were similar for NIT The group mean relative and absolute VO 2 and absolute VCO 2 responses at 160 m, 280 m, 440 m and peak during the Yo-Yo IR1 following both NIT and PLA supplementation were shown in Figure 2, and values (including VE and RER values) were reported in CCC: 0.92) and VE (CV: 6.9%, CI: 4.9-10.7%, CCC: 0.89) during more than 2 h of continuous field tests [31]. The Yo-Yo IR1 test has been described elsewhere [20]. Briefly, the test consists of repeated 2 × 20 m runs, interspersed by a 10 s active recovery period, at progressively increasing speeds controlled by audio bleeps from a portable audio system. First four shuttles were at the speed of 10-13 km·h −1 (0-160 m), then three shuttles at 13.5 km·h −1 (200-280 m) and four shuttles at 14.0 km·h −1 (320-440 m); thereafter, the speed increased 0.5 km·h −1 every eight shuttles (i.e., 760, 1080, 1400 m, etc.). The final distance successfully covered was recorded after the second failed attempt to meet the start/finish line in the allocated time.

Statistical Analysis
All data were presented as means ± SD. Differences between NIT and PLA in distance covered during the Yo-Yo IR1 test, VO2peak, VCO2peak and VEpeak, RERpeak and BLa pre-and post-exercise test were analyzed using a paired samples t-test. Effect sizes (d) were calculated through Cohen's d as: large d > 0.8, moderate d = 0.8 to 0.5, small d = 0.5 to 0.2, and trivial d < 0.2 [32]. Differences in VO2, VCO2, VE and RER at 160, 280 and 440 Im were determined using two-way repeated-measure ANOVA (supplement × distance). In addition, effect size was calculated as partial eta-squared (ŋp 2 ) varying small (<0.25), medium (0.26-0.63) and large (>0.63) [33]. All data were analyzed using SPSS 27.0 (IBM Corp., Armonk, NY, USA). Significance was determined at p < 0.05.

Results
The distance covered in the Yo-Yo IR1 test was significantly greater in NIT (990 ± 442.25 m) compared to PLA (870 ± 357.4 m, p = 0.007, d = 0.30, Figure 1). CCC: 0.92) and VE (CV: 6.9%, CI: 4.9-10.7%, CCC: 0.89) during more than 2 h of continuous field tests [31]. The Yo-Yo IR1 test has been described elsewhere [20]. Briefly, the test consists of repeated 2 × 20 m runs, interspersed by a 10 s active recovery period, at progressively increasing speeds controlled by audio bleeps from a portable audio system. First four shuttles were at the speed of 10-13 km·h −1 (0-160 m), then three shuttles at 13.5 km·h −1 (200-280 m) and four shuttles at 14.0 km·h −1 (320-440 m); thereafter, the speed increased 0.5 km·h −1 every eight shuttles (i.e., 760, 1080, 1400 m, etc.). The final distance successfully covered was recorded after the second failed attempt to meet the start/finish line in the allocated time.

Statistical Analysis
All data were presented as means ± SD. Differences between NIT and PLA in distance covered during the Yo-Yo IR1 test, VO2peak, VCO2peak and VEpeak, RERpeak and BLa pre-and post-exercise test were analyzed using a paired samples t-test. Effect sizes (d) were calculated through Cohen's d as: large d > 0.8, moderate d = 0.8 to 0.5, small d = 0.5 to 0.2, and trivial d < 0.2 [32]. Differences in VO2, VCO2, VE and RER at 160, 280 and 440 Im were determined using two-way repeated-measure ANOVA (supplement × distance). In addition, effect size was calculated as partial eta-squared (ŋp 2 ) varying small (<0.25), medium (0.26-0.63) and large (>0.63) [33]. All data were analyzed using SPSS 27.0 (IBM Corp., Armonk, NY, USA). Significance was determined at p < 0.05.

Results
The distance covered in the Yo-Yo IR1 test was significantly greater in NIT (990 ± p 2 = 0.169), while these increased with distance (ANOVA: distance, F = 63.56; p < 0.001; FOR PEER REVIEW 4 of 11 CCC: 0.92) and VE (CV: 6.9%, CI: 4.9-10.7%, CCC: 0.89) during more than 2 h of continuous field tests [31]. The Yo-Yo IR1 test has been described elsewhere [20]. Briefly, the test consists of repeated 2 × 20 m runs, interspersed by a 10 s active recovery period, at progressively increasing speeds controlled by audio bleeps from a portable audio system. First four shuttles were at the speed of 10-13 km·h −1 (0-160 m), then three shuttles at 13.5 km·h −1 (200-280 m) and four shuttles at 14.0 km·h −1 (320-440 m); thereafter, the speed increased 0.5 km·h −1 every eight shuttles (i.e., 760, 1080, 1400 m, etc.). The final distance successfully covered was recorded after the second failed attempt to meet the start/finish line in the allocated time.

Statistical Analysis
All data were presented as means ± SD. Differences between NIT and PLA in distance covered during the Yo-Yo IR1 test, VO2peak, VCO2peak and VEpeak, RERpeak and BLa pre-and post-exercise test were analyzed using a paired samples t-test. Effect sizes (d) were calculated through Cohen's d as: large d > 0.8, moderate d = 0.8 to 0.5, small d = 0.5 to 0.2, and trivial d < 0.2 [32]. Differences in VO2, VCO2, VE and RER at 160, 280 and 440 Im were determined using two-way repeated-measure ANOVA (supplement × distance). In addition, effect size was calculated as partial eta-squared (ŋp 2 ) varying small (<0.25), medium (0.26-0.63) and large (>0.63) [33]. All data were analyzed using SPSS 27.0 (IBM Corp., Armonk, NY, USA). Significance was determined at p < 0.05.

Results
The distance covered in the Yo-Yo IR1 test was significantly greater in NIT (990 ± 442.25 m) compared to PLA (870 ± 357.4 m, p = 0.007, d = 0.30, Figure 1). The group mean relative and absolute VO2 and absolute VCO2 responses at 160m, 280m, 440m and peak during the Yo-Yo IR1 following both NIT and PLA supplementation were shown in Figure 2, and values (including VE and RER values) were reported in Table 1. The mean relative VO2 responses at submaximal distances were similar for NIT p 2 = 0.094; supplementation × distance, F = 0.35, p = 0.966,

Statistical Analysis
All data were presented as means ± SD. Differences between NIT and P covered during the Yo-Yo IR1 test, VO2peak, VCO2peak and VEpeak, RERpeak an post-exercise test were analyzed using a paired samples t-test. Effect sizes lated through Cohen's d as: large d > 0.8, moderate d = 0.8 to 0.5, small d = trivial d < 0.2 [32]. Differences in VO2, VCO2, VE and RER at 160, 280 an determined using two-way repeated-measure ANOVA (supplement × dist tion, effect size was calculated as partial eta-squared (ŋp 2 ) varying small (< (0.26-0.63) and large (>0.63) [33]. All data were analyzed using SPSS 27.0 (I monk, NY, USA). Significance was determined at p < 0.05.

Results
The distance covered in the Yo-Yo IR1 test was significantly greater 442.25 m) compared to PLA (870 ± 357.4 m, p = 0.007, d = 0.30, Figure 1). The group mean relative and absolute VO2 and absolute VCO2 resp 280m, 440m and peak during the Yo-Yo IR1 following both NIT and PLA tion were shown in Figure 2, and values (including VE and RER values) w Table 1. The mean relative VO2 responses at submaximal distances were s p 2 = 0.003), while these increased with distance (ANOVA: distance, F = 60.51; p < 0.001; allocated time.

Results
The distance covered in the Yo-Yo IR1 test was significantly greater in 442.25 m) compared to PLA (870 ± 357.4 m, p = 0.007, d = 0.30, Figure 1). The group mean relative and absolute VO2 and absolute VCO2 response 280m, 440m and peak during the Yo-Yo IR1 following both NIT and PLA sup tion were shown in Figure 2, and values (including VE and RER values) were Table 1. The mean relative VO2 responses at submaximal distances were simi p 2 = 0.85). The mean VCO 2 response were also very similar at submaximal distances between NIT and PLA (ANOVA: supplementation, F = 1.85; p = 0.201; 0.5 km·h −1 every eight shuttles (i.e., 760, 1080, 1400 m, etc.). The final distance successfully covered was recorded after the second failed attempt to meet the start/finish line in the allocated time.

Statistical Analysis
All data were presented as means ± SD. Differences between NIT and PLA in distance covered during the Yo-Yo IR1 test, VO2peak, VCO2peak and VEpeak, RERpeak and BLa pre-and post-exercise test were analyzed using a paired samples t-test. Effect sizes (d) were calculated through Cohen's d as: large d > 0.8, moderate d = 0.8 to 0.5, small d = 0.5 to 0.2, and trivial d < 0.2 [32]. Differences in VO2, VCO2, VE and RER at 160, 280 and 440 Im were determined using two-way repeated-measure ANOVA (supplement × distance). In addition, effect size was calculated as partial eta-squared (ŋp 2 ) varying small (<0.25), medium (0.26-0.63) and large (>0.63) [33]. All data were analyzed using SPSS 27.0 (IBM Corp., Armonk, NY, USA). Significance was determined at p < 0.05.

Results
The distance covered in the Yo-Yo IR1 test was significantly greater in NIT (990 ± 442.25 m) compared to PLA (870 ± 357.4 m, p = 0.007, d = 0.30, Figure 1). The group mean relative and absolute VO2 and absolute VCO2 responses at 160m, 280m, 440m and peak during the Yo-Yo IR1 following both NIT and PLA supplementation were shown in Figure 2, and values (including VE and RER values) were reported in (200-280 m) and four shuttles at 14.0 km·h −1 (320-440 m); thereafter, the speed increased 0.5 km·h −1 every eight shuttles (i.e., 760, 1080, 1400 m, etc.). The final distance successfully covered was recorded after the second failed attempt to meet the start/finish line in the allocated time.

Statistical Analysis
All data were presented as means ± SD. Differences between NIT and PLA in distance covered during the Yo-Yo IR1 test, VO2peak, VCO2peak and VEpeak, RERpeak and BLa pre-and post-exercise test were analyzed using a paired samples t-test. Effect sizes (d) were calculated through Cohen's d as: large d > 0.8, moderate d = 0.8 to 0.5, small d = 0.5 to 0.2, and trivial d < 0.2 [32]. Differences in VO2, VCO2, VE and RER at 160, 280 and 440 Im were determined using two-way repeated-measure ANOVA (supplement × distance). In addition, effect size was calculated as partial eta-squared (ŋp 2 ) varying small (<0.25), medium (0.26-0.63) and large (>0.63) [33]. All data were analyzed using SPSS 27.0 (IBM Corp., Armonk, NY, USA). Significance was determined at p < 0.05.

Results
The distance covered in the Yo-Yo IR1 test was significantly greater in NIT (990 ± 442.25 m) compared to PLA (870 ± 357.4 m, p = 0.007, d = 0.30, Figure 1). The group mean relative and absolute VO2 and absolute VCO2 responses at 160m, 280m, 440m and peak during the Yo-Yo IR1 following both NIT and PLA supplementation were shown in Figure 2, and values (including VE and RER values) were reported in Table 1. The mean relative VO2 responses at submaximal distances were similar for NIT p 2 = 0.143), while these increased with distance (ANOVA: distance, F = 40.26; p < 0.001; shuttles were at the speed of 10-13 km·h −1 (0-160 m), then three shuttles at 13.5 km·h −1 (200-280 m) and four shuttles at 14.0 km·h −1 (320-440 m); thereafter, the speed increased 0.5 km·h −1 every eight shuttles (i.e., 760, 1080, 1400 m, etc.). The final distance successfully covered was recorded after the second failed attempt to meet the start/finish line in the allocated time.

Statistical Analysis
All data were presented as means ± SD. Differences between NIT and PLA in distance covered during the Yo-Yo IR1 test, VO2peak, VCO2peak and VEpeak, RERpeak and BLa pre-and post-exercise test were analyzed using a paired samples t-test. Effect sizes (d) were calculated through Cohen's d as: large d > 0.8, moderate d = 0.8 to 0.5, small d = 0.5 to 0.2, and trivial d < 0.2 [32]. Differences in VO2, VCO2, VE and RER at 160, 280 and 440 Im were determined using two-way repeated-measure ANOVA (supplement × distance). In addition, effect size was calculated as partial eta-squared (ŋp 2 ) varying small (<0.25), medium (0.26-0.63) and large (>0.63) [33]. All data were analyzed using SPSS 27.0 (IBM Corp., Armonk, NY, USA). Significance was determined at p < 0.05.

Results
The distance covered in the Yo-Yo IR1 test was significantly greater in NIT (990 ± 442.25 m) compared to PLA (870 ± 357.4 m, p = 0.007, d = 0.30, Figure 1). The group mean relative and absolute VO2 and absolute VCO2 responses at 160m, 280m, 440m and peak during the Yo-Yo IR1 following both NIT and PLA supplementation were shown in Figure 2, and values (including VE and RER values) were reported in Table 1. The mean relative VO2 responses at submaximal distances were similar for NIT p 2 = 0.785). There were also no significant differences between trials for the mean VE response (ANOVA: supplementation, F = 2.94; p = 0.114; increasing speeds controlled by audio bleeps from a portable audio sys shuttles were at the speed of 10-13 km·h −1 (0-160 m), then three shuttles (200-280 m) and four shuttles at 14.0 km·h −1 (320-440 m); thereafter, the s 0.5 km·h −1 every eight shuttles (i.e., 760, 1080, 1400 m, etc.). The final distan covered was recorded after the second failed attempt to meet the start/fi allocated time.

Results
The distance covered in the Yo-Yo IR1 test was significantly greate 442.25 m) compared to PLA (870 ± 357.4 m, p = 0.007, d = 0.30, Figure 1). The group mean relative and absolute VO2 and absolute VCO2 resp 280m, 440m and peak during the Yo-Yo IR1 following both NIT and PLA tion were shown in Figure 2, and values (including VE and RER values) w repeated 2 × 20 m runs, interspersed by a 10 s active recovery period, at progressively increasing speeds controlled by audio bleeps from a portable audio system. First four shuttles were at the speed of 10-13 km·h −1 (0-160 m), then three shuttles at 13.5 km·h −1 (200-280 m) and four shuttles at 14.0 km·h −1 (320-440 m); thereafter, the speed increased 0.5 km·h −1 every eight shuttles (i.e., 760, 1080, 1400 m, etc.). The final distance successfully covered was recorded after the second failed attempt to meet the start/finish line in the allocated time.

Statistical Analysis
All data were presented as means ± SD. Differences between NIT and PLA in distance covered during the Yo-Yo IR1 test, VO2peak, VCO2peak and VEpeak, RERpeak and BLa pre-and post-exercise test were analyzed using a paired samples t-test. Effect sizes (d) were calculated through Cohen's d as: large d > 0.8, moderate d = 0.8 to 0.5, small d = 0.5 to 0.2, and trivial d < 0.2 [32]. Differences in VO2, VCO2, VE and RER at 160, 280 and 440 Im were determined using two-way repeated-measure ANOVA (supplement × distance). In addition, effect size was calculated as partial eta-squared (ŋp 2 ) varying small (<0.25), medium (0.26-0.63) and large (>0.63) [33]. All data were analyzed using SPSS 27.0 (IBM Corp., Armonk, NY, USA). Significance was determined at p < 0.05.

Results
The distance covered in the Yo-Yo IR1 test was significantly greater in NIT (990 ± 442.25 m) compared to PLA (870 ± 357.4 m, p = 0.007, d = 0.30, Figure 1). The group mean relative and absolute VO2 and absolute VCO2 responses at 160m, 280m, 440m and peak during the Yo-Yo IR1 following both NIT and PLA supplementation were shown in Figure 2, and values (including VE and RER values) were reported in Table 1. The mean relative VO2 responses at submaximal distances were similar for NIT The Yo-Yo IR1 test has been described elsewhere [20]. Briefly, the test consists of repeated 2 × 20 m runs, interspersed by a 10 s active recovery period, at progressively increasing speeds controlled by audio bleeps from a portable audio system. First four shuttles were at the speed of 10-13 km·h −1 (0-160 m), then three shuttles at 13.5 km·h −1 (200-280 m) and four shuttles at 14.0 km·h −1 (320-440 m); thereafter, the speed increased 0.5 km·h −1 every eight shuttles (i.e., 760, 1080, 1400 m, etc.). The final distance successfully covered was recorded after the second failed attempt to meet the start/finish line in the allocated time.

Statistical Analysis
All data were presented as means ± SD. Differences between NIT and PLA in distance covered during the Yo-Yo IR1 test, VO2peak, VCO2peak and VEpeak, RERpeak and BLa pre-and post-exercise test were analyzed using a paired samples t-test. Effect sizes (d) were calculated through Cohen's d as: large d > 0.8, moderate d = 0.8 to 0.5, small d = 0.5 to 0.2, and trivial d < 0.2 [32]. Differences in VO2, VCO2, VE and RER at 160, 280 and 440 Im were determined using two-way repeated-measure ANOVA (supplement × distance). In addition, effect size was calculated as partial eta-squared (ŋp 2 ) varying small (<0.25), medium (0.26-0.63) and large (>0.63) [33]. All data were analyzed using SPSS 27.0 (IBM Corp., Armonk, NY, USA). Significance was determined at p < 0.05.

Results
The distance covered in the Yo-Yo IR1 test was significantly greater in NIT (990 ± 442.25 m) compared to PLA (870 ± 357.4 m, p = 0.007, d = 0.30, Figure 1). The group mean relative and absolute VO2 and absolute VCO2 responses at 160m, 280m, 440m and peak during the Yo-Yo IR1 following both NIT and PLA supplementation were shown in Figure 2, and values (including VE and RER values) were reported in Table 1. The mean relative VO2 responses at submaximal distances were similar for NIT p 2 = 0.796). There were also no significant differences between conditions for RER (ANOVA: supplement, F = 0.46; p = 0.513; .92) and VE (CV: 6.9%, CI: 4.9-10.7%, CCC: 0.89) during more than 2 h of continuld tests [31]. e Yo-Yo IR1 test has been described elsewhere [20]. Briefly, the test consists of d 2 × 20 m runs, interspersed by a 10 s active recovery period, at progressively ing speeds controlled by audio bleeps from a portable audio system. First four s were at the speed of 10-13 km·h −1 (0-160 m), then three shuttles at 13.5 km·h −1 0 m) and four shuttles at 14.0 km·h −1 (320-440 m); thereafter, the speed increased ·h −1 every eight shuttles (i.e., 760, 1080, 1400 m, etc.). The final distance successfully d was recorded after the second failed attempt to meet the start/finish line in the ed time.
. The distance covered in the Yo-Yo IR1 test was 14% greater with NIT compared to PLA. hed lines indicate the responses of individual participants. The solid line indicates the ean (±SD). * p < 0.05. e group mean relative and absolute VO2 and absolute VCO2 responses at 160m, 440m and peak during the Yo-Yo IR1 following both NIT and PLA supplementare shown in Figure 2, and values (including VE and RER values) were reported in . The mean relative VO2 responses at submaximal distances were similar for NIT CCC: 0.92) and VE (CV: 6.9%, CI: 4.9-10.7%, CCC: 0.89) during more than 2 h of continuous field tests [31]. The Yo-Yo IR1 test has been described elsewhere [20]. Briefly, the test consists of repeated 2 × 20 m runs, interspersed by a 10 s active recovery period, at progressively increasing speeds controlled by audio bleeps from a portable audio system. First four shuttles were at the speed of 10-13 km·h −1 (0-160 m), then three shuttles at 13.5 km·h −1 (200-280 m) and four shuttles at 14.0 km·h −1 (320-440 m); thereafter, the speed increased 0.5 km·h −1 every eight shuttles (i.e., 760, 1080, 1400 m, etc.). The final distance successfully covered was recorded after the second failed attempt to meet the start/finish line in the allocated time.

Statistical Analysis
All data were presented as means ± SD. Differences between NIT and PLA in distance covered during the Yo-Yo IR1 test, VO2peak, VCO2peak and VEpeak, RERpeak and BLa pre-and post-exercise test were analyzed using a paired samples t-test. Effect sizes (d) were calculated through Cohen's d as: large d > 0.8, moderate d = 0.8 to 0.5, small d = 0.5 to 0.2, and trivial d < 0.2 [32]. Differences in VO2, VCO2, VE and RER at 160, 280 and 440 Im were determined using two-way repeated-measure ANOVA (supplement × distance). In addition, effect size was calculated as partial eta-squared (ŋp 2 ) varying small (<0.25), medium (0.26-0.63) and large (>0.63) [33]. All data were analyzed using SPSS 27.0 (IBM Corp., Armonk, NY, USA). Significance was determined at p < 0.05.

Results
The distance covered in the Yo-Yo IR1 test was significantly greater in NIT (990 ± 442.25 m) compared to PLA (870 ± 357.4 m, p = 0.007, d = 0.30, Figure 1). The group mean relative and absolute VO2 and absolute VCO2 responses at 160m, 280m, 440m and peak during the Yo-Yo IR1 following both NIT and PLA supplementation were shown in Figure 2, and values (including VE and RER values) were reported in Table 1. The mean relative VO2 responses at submaximal distances were similar for NIT p 2 = 0.030; supplementation × distance, F = 0.02; p = 0.946; CCC: 0.92) and VE (CV: 6.9%, CI: 4.9-10.7%, CCC: 0.89) during more than 2 h of continuous field tests [31].
The Yo-Yo IR1 test has been described elsewhere [20]. Briefly, the test consists of repeated 2 × 20 m runs, interspersed by a 10 s active recovery period, at progressively increasing speeds controlled by audio bleeps from a portable audio system. First four shuttles were at the speed of 10-13 km·h −1 (0-160 m), then three shuttles at 13.5 km·h −1 (200-280 m) and four shuttles at 14.0 km·h −1 (320-440 m); thereafter, the speed increased 0.5 km·h −1 every eight shuttles (i.e., 760, 1080, 1400 m, etc.). The final distance successfully covered was recorded after the second failed attempt to meet the start/finish line in the allocated time.

Statistical Analysis
All data were presented as means ± SD. Differences between NIT and PLA in distance covered during the Yo-Yo IR1 test, VO2peak, VCO2peak and VEpeak, RERpeak and BLa pre-and post-exercise test were analyzed using a paired samples t-test. Effect sizes (d) were calculated through Cohen's d as: large d > 0.8, moderate d = 0.8 to 0.5, small d = 0.5 to 0.2, and trivial d < 0.2 [32]. Differences in VO2, VCO2, VE and RER at 160, 280 and 440 Im were determined using two-way repeated-measure ANOVA (supplement × distance). In addition, effect size was calculated as partial eta-squared (ŋp 2 ) varying small (<0.25), medium (0.26-0.63) and large (>0.63) [33]. All data were analyzed using SPSS 27.0 (IBM Corp., Armonk, NY, USA). Significance was determined at p < 0.05.

Results
The distance covered in the Yo-Yo IR1 test was significantly greater in NIT (990 ± 442.25 m) compared to PLA (870 ± 357.4 m, p = 0.007, d = 0.30, Figure 1). The group mean relative and absolute VO2 and absolute VCO2 responses at 160m, 280m, 440m and peak during the Yo-Yo IR1 following both NIT and PLA supplementation were shown in Figure 2, and values (including VE and RER values) were reported in Table 1. The mean relative VO2 responses at submaximal distances were similar for NIT p 2 = 0.002).   There were no significant differences in the VO 2peak p = 0.114, d = 0.19), VCO 2peak (p = 0.085, d = 0.22), VE peak (p = 0.295, d = 0.18), and RER peak (p = 0.079, d = 0.80) between conditions. There were no significant differences in pre-BLa (p = 0.45, d = 0.36) and post-BLa (p = 0.24, d = 0.48) between NIT and PLA.

Discussion
This study set out to determine whether the acute supplementation of NO 3 − would enhance the metabolic responses to intermittent running performance as measured via the Yo-Yo IR1 test among recreationally active adults. The original finding of the present study was that the acute supplementation of NO 3 − via beetroot juice significantly enhanced intermittent running performance in the Yo-Yo IR1 test (by 14%, p = 0.007). However, the metabolic responses (e.g., VO 2 , VCO 2 and VE) did not alter at submaximal distances (at 160, 280 and 440 m) or peak during the Yo-Yo IR1 test after acute NO 3 − supplementation compared to placebo. These findings indicate that acute NO 3 − supplementation can enhance performance without altering exercise efficiency during intermittent running.
The Yo-Yo IR1 test in the present study compared to previous observations, this also supports the existence of potential responders and non-responders to NO 3 − supplementation [1,34]. Marked differences exist between individuals in the erogenicity of NO 3 − supplementation. Assuming the potential type II preference of NO 3 − is on those factors, it might be speculated that because some of the participants in the present study had a high proportion of type II muscle fibers, this might have theoretically increased the ergogenic potential of NO 3 − [34]. The greater improvement in the present study compared to the previous observations would be also due to a different population being used (recreational individuals vs. moderately/highly trained team-sport players) given that existing evidence has shown that potential ergogenic effect of NO 3 − supplementation is more evident in individuals who have low levels of aerobic fitness [1,10]. Highly endurance-trained athletes could be less responsive to NO 3 − supplementation mediated by a lower fraction of type II muscle fibers or other factors such as greater NO synthase activity, mitochondrial efficiency or better muscle oxygenation compared to moderately trained subjects [35]. Therefore, further research is required to determine whether acute (2-3 h pre-exercise) high-dose supplementation of NO 3 − (>6 mmol) can benefit in highly trained team sports athletes. Given the considerable differences in technical requirements when executing specific skills, the metabolic cost of the activity and the participants' characteristics, extrapolating these findings to every team sport activity and/or players is fraught. Future studies are therefore warranted to explore whether the findings of the present study can be reproduced in different team sports activities and/or athletes.
To best of our knowledge, this is the first study that assessed pulmonary VO 2 , VCO 2 and VE responses following NO 3 − supplementation during the Yo-Yo IR1 test. This study shows that NO 3 − supplementation had no effect onVO 2 response during the Yo-Yo IR1 test. These findings are consistent with observations that have reported an enhancement in intermittent exercise performance, consisting of repeated sprints, but no alteration in VO 2 [16,36]. However, more recent studies reported inconsistent findings regarding the impact of NO 3 − supplementation to improve performance or/and metabolic responses during different high-intensity intermittent exercise [37,38]. Whilst Kent at al. [37] found that supplementation of NO 3 − did not enhance repeated-sprints performance in hypoxia, but may reduce VO 2 , Sousa et al. [38] showed no impact of NO 3 − supplementation either on VO 2 or repeated-sprint training. Besides inter-study differences (e.g., participant training status, supplementation regimen and environmental conditions), previous studies applied considerably different exercise modalities and exercise protocols regarding work-to-rest ratio (e.g., intensities, durations, and numbers of work and/or rest) [17,[36][37][38][39]. In earlier studies, reduced VO 2 following NO 3 − supplementation during submaximal exercise [2,8,9,25] was attributed to a reduced adenosine-tri-phosphate (ATP) cost of muscle force production. The lack of effect in VO 2 in this study and in previous studies conducted intermittent-exercise protocols might be, at least partly, due to the regular fluctuations in exercise intensity (e.g., non-steady-state conditions). Therefore, the findings of the present study suggest that the observed ergogenic effect of NO 3 − in intermittent exercise may work through divergent mechanisms of action independent of alterations in the efficiency of oxidative metabolism.
Some other mechanistic underpinnings, such as improved muscle contractility, for the effect of NO 3 − supplementation have been also identified [1]. Animal-based studies reported that the effect of NO 3 − supplementation is more apparent in type II compared with type I fibers regarding physiological response and performance, such as improving force output selectively in type II fibers via increasing sarcoplasmic reticulum calcium handling and/or release [6,7]. Despite remaining to be elucidated in human, this mechanism may better explain the ergogenic basis for the enhanced performance in the Yo-Yo IR1 test where a greater recruitment of type II fibers is expected [19]. Improved performance in the Yo-Yo IR1 might be related to delaying fatigue as a result of preserving the reduction in muscle excitability [40,41]. Wylie et al. [9] have previously attributed enhanced performance in the Yo-Yo IR1 test to attenuated muscle excitability due to a net loss of potassium. Further, it has been reported that NO 3 − supplementation attenuated the increase in motor unit action