Effects of Combined Inorganic Nitrate and Nitrite Supplementation on Cardiorespiratory Fitness and Skeletal Muscle Oxidative Capacity in Type 2 Diabetes: A Pilot Randomized Controlled Trial

Nitric oxide (NO) stimulates mitochondrial biogenesis in skeletal muscle. However, NO metabolism is disrupted in individuals with type 2 diabetes mellitus (T2DM) potentially contributing to their decreased cardiorespiratory fitness (i.e., VO2max) and skeletal muscle oxidative capacity. We used a randomized, double-blind, placebo-controlled, 8-week trial with beetroot juice containing nitrate (NO3−) and nitrite (NO2−) (250 mg and 20 mg/day) to test potential benefits on VO2max and skeletal muscle oxidative capacity in T2DM. T2DM (N = 36, Age = 59 ± 9 years; BMI = 31.9 ± 5.0 kg/m2) and age- and BMI-matched non-diabetic controls (N = 15, Age = 60 ± 9 years; BMI = 29.5 ± 4.6 kg/m2) were studied. Mitochondrial respiratory capacity was assessed in muscle biopsies from a subgroup of T2DM and controls (N = 19 and N = 10, respectively). At baseline, T2DM had higher plasma NO3− (100%; p < 0.001) and lower plasma NO2− levels (−46.8%; p < 0.0001) than controls. VO2max was lower in T2DM (−26.4%; p < 0.001), as was maximal carbohydrate- and fatty acid-supported oxygen consumption in permeabilized muscle fibers (−26.1% and −25.5%, respectively; p < 0.05). NO3−/NO2− supplementation increased VO2max (5.3%; p < 0.01). Further, circulating NO2−, but not NO3−, positively correlated with VO2max after supplementation (R2= 0.40; p < 0.05). Within the NO3−/NO2− group, 42% of subjects presented improvements in both carbohydrate- and fatty acid-supported oxygen consumption in skeletal muscle (vs. 0% in placebo; p < 0.05). VO2max improvements in these individuals tended to be larger than in the rest of the NO3−/NO2− group (1.21 ± 0.51 mL/(kg*min) vs. 0.31 ± 0.10 mL/(kg*min); p = 0.09). NO3−/NO2− supplementation increases VO2max in T2DM individuals and improvements in skeletal muscle oxidative capacity appear to occur in those with more pronounced increases in VO2max.


Introduction
Type 2 Diabetes Mellitus (T2DM) affects over 460 million people worldwide [1] and is associated with a significant increased risk of cardiovascular disease and all-cause mortality [2,3]. Cardiorespiratory fitness (i.e., VO 2 max), which is inversely associated with cardiovascular disease and all-cause mortality risks [4], is decreased by ≥15% in T2DM patients [5,6]. Therefore, strategies that increase VO 2 max may mitigate morbidity and mortality in T2DM patients. Regular exercise is an effective intervention that can increase VO 2 max in T2DM patients along with improvements in insulin sensitivity and glucose metabolism [7][8][9]. However, substantial reductions in VO 2 max invariably cause simple daily living activities to become more physically demanding, which make it more challenging for individuals to engage in exercise programs. In fact, 39% of adults with diabetes report being physically active compared to 58% of those without diabetes [10]. Collectively, these observations indicate that additional strategies to improve VO 2 max should have significant clinical implications in T2DM. VO 2 max is dependent on three main factors, namely respiratory mechanics and alveolar gas exchange, delivery of oxygen to the working muscle via blood flow and utilization of oxygen by muscle mitochondria [11]. Among these, limitations in muscle blood flow and oxygen utilization have been extensively documented in T2DM patients [12][13][14][15]. Of note, however, the bioavailability of the signaling molecule nitric oxide (NO) is decreased in animal models of diabetes as well as in diabetic humans [16,17]. This is relevant because NO has been shown not only to regulate blood flow to contracting muscles, but also to stimulate muscle mitochondrial biogenesis [18][19][20][21]. The L-arginine-NO synthase (NOS) pathway is a major contributor to NO formation in mammals. Once synthesized, NO is rapidly oxidized to form nitrite (NO 2 − ) and nitrate (NO 3 − ). However, the conversion of NO 3 − and NO 2 − back to NO also occurs in blood and various tissues [22][23][24][25] and NO 3 − reduction (i.e., nitrate-nitrite-NO pathway) represents an alternative pathway for generation of NO [25]. In a recent related study, we demonstrated that the combined supplementation of inorganic NO 3 − and NO 2 − increases NO bioavailability in T2DM [26]. Here, we hypothesized that combined supplementation of inorganic NO 3 − and NO 2 − would improve VO 2 max and skeletal muscle mitochondria respiratory function in T2DM participants. We report findings in T2DM participants and age-and BMI-matched non-diabetic controls and address our central hypothesis using a double-blinded, placebo-controlled trial. We have also explored associations between circulating NO 3 − and/or NO 2 − with VO 2 max in T2DM to determine the extent to which these NO-related metabolites or precursors may predict outcomes of the supplementation.

Ethical Approval
Patients provided written, informed consent prior to participating in experimental protocols approved by the University of Iowa's Institutional Review Board. Data reported in this manuscript were collected in accordance with the principles of the Declaration of Helsinki during a registered clinical trial (ClinicalTrials.gov ID: NCT02804932) with some related data previously published. Specifically, baseline data from 30/36 patients in the present study were compared to 15 controls without T2DM to identify if peripheral α-mediated vasoconstriction is exaggerated in patients with T2DM [15]. Demographical, along with pre-intervention endothelium-dependent vasodilation data, were published to examine the relationship between microvascular endothelial function and glycemic management [27]. Impact of the intervention on forearm skeletal muscle perfusion to handgrip exercise and blood pressure in T2DM patients was examined elsewhere [26,28].

Study Participants
Age-and BMI-matched T2DM and non-diabetic control participants were enrolled in the study from 2016 to 2019. Exclusion criteria for all subjects included age of ≤40 and ≥77 years, BMI of ≥42 kg/m 2 , cardiovascular events in the prior year (e.g., heart attack, stroke), heart failure, symptomatic coronary artery disease, hypotension (i.e., resting systolic BP < 90 mmHg), renal impairment with creatinine clearance (estimated glomerular filtration rate of <50 mL/min), current tobacco use or recent (<1 yr.) cessation, use of anticoagulant drugs, use of medication containing nitrates or participation in research studies in which medications or interventions that could alter subject responses in the current study. T2DM patients were included if their diagnosis (along with their medications) was corroborated via review of electronic medical records. Control subjects were examined at baseline and T2DM patients were examined at baseline and after an 8-week, double-blinded, placebo-controlled intervention with beetroot juice. The total number of T2DM patients participating in the intervention was based on a priori sample size calculation related to a different outcome measure of the larger clinical trial (i.e., muscle blood flow) [26].

Maximal Exercise Testing
A 12-lead electrocardiogram (ECG), symptom limited cardiopulmonary exercise testing with gas exchange measurements (Parvo Medics, Sandy, UT, USA) was performed on a cycle ergometer (Lode Corival Bike Ergometer, Groningen, The Netherlands) using a ramp protocol to determine cardiorespiratory fitness (VO 2 max). The rate of perceived exertion (RPE), heart rate and brachial blood pressure were measured throughout the test and all tests were monitored by a physician. The exercise protocol consisted of a two-minute warmup with no resistance. The first minute was performed at 20 W and then increased by 10 W every minute thereafter until volitional fatigue. Maximal effort was defined as achieving 90% age-predicted maximal heart rate (i.e., 220-age), a respiratory exchange ratio (RER) ≥ 1.1 and a RPE ≥ 17 with VO 2 max defined as the average of the two or three highest measurements (each reflecting 15-sec average data points) during the final minute of exercise. Maximal work rate (WRmax) was defined as the highest work rate (in Watts) achieved during exercise test.

Nitrate (NO
As described previously [26], patients were randomly assigned in a double-blinded parallel fashion to consume either beetroot juice containing 250 mg NO 3 − (4.0 mmol) and 20 mg NO 2 − (0.3 mmol), which is comparable to 100 g of red beetroot or fresh spinach [29], or~20 mg NO 3 − (~0.08 mmol) without any NO 2 − (placebo) daily for eight weeks. Beetroot beverages consisted of beetroot powder (Superbeets, HumanN, Inc., Austin, TX, USA) dissolved in 4-6 ounces of water. The NO 3 − and NO 2 − content of both supplements was verified via high-performance liquid chromatography prior to the start of data collection. Participants were instructed to consume those once per day, before the first meal in the morning. Two subjects reported gastrointestinal discomfort with the beetroot juice consumption and were asked to consume the beverage two hours after their first meal. All subjects were instructed to adhere to a low-nitrate diet (e.g., no leafy vegetables or processed meats) 48 h prior to study visits. Given the anti-bacterial properties of mouthwash [30], subjects were also instructed not to use these products before beetroot juice consumption and for two hours afterwards. Post study visits were conducted 18-24 h following consumption of the supplements.

Blood Sampling and Muscle Biopsies
The blood collection and muscle biopsies were performed 48-72 h apart from maximal exercise testing. Blood was collected in tubes containing EDTA. Samples were centrifuged, aliquoted and frozen at −80 • C until further analysis, as described [26]. Plasma NO 3 − and NO 2 − were determined by chemiluminescence (NOA 280i, Sievers Instruments, Boulder, CO, USA). In addition, quantification of fasting blood glucose, insulin and HbA1c were performed at the University of Iowa Hospitals and Clinics using standard laboratory protocols [26].
Vastus lateralis muscle biopsies (~100-150 mg) were obtained with a percutaneous needle using the Bergstrom technique under local anesthesia. All study participants who received a muscle biopsy and/or had blood drawn did so after an overnight (minimum of eight hours) fasting period, refraining from caffeine and alcohol. While physical activity was not monitored throughout the study, all subjects were instructed to refrain from exercise for at least 24 h prior to the study visit. In addition, all subjects withheld their medication(s) the morning of the biopsy. The final number of muscle biopsies provided and analysis performed for each sample, was determined by: (a) number of subjects that consented to the biopsy; and (b) amount of fat contamination from intramuscular stores which at times limited the amount of muscle obtained and the variables that could be reliably assessed from those.

Histology and Fluorescence Microscopy
Approximately 20 mg of tissue obtained from biopsies were immediately embedded in tissue-freezing medium for subsequent immunofluorescence analysis as described [31]. Serial sections of muscle samples (10 µm) were obtained at a temperature of −24-25 • C. For fiber type distribution, sections were stained with specific antibodies for Laminin (Sigma L9393, 1:500), myosin heavy chain (MyHC) type 1 (MYH7 (type 1)-BA-F8), type 2A (MYH2 (Type 2A)-SC-71) and type 2X (MYH1 (type 2X)-6H1). These were from Developmental Studies Hybridoma Bank and were used at 1:250 dilution, as previously described [32]. Samples were mounted using appropriate medium (Prolonged Diamond or Glass; Invitrogen, # P36970 or P36980) and images were obtained via confocal microscopy (Zeiss LSM710; Jena, Germany). An average of 384 fibers was analyzed per section and all histological analyses described were completed in a single-blinded fashion using ImageJ (NIH).

Skeletal Muscle Oxidative Capacity
Skeletal muscle oxidative capacity, also referred to as mitochondrial respiratory capacity, was assessed as previously described [33][34][35][36]. Approximately 50 mg of skeletal muscle tissue obtained from the biopsy sample was immediately placed in ice cold preservation buffer (7.23 mM K 2 EGTA, 2.77 mM CaK 2 EGTA, 20 mM imidazole, 0.5 mM DTT, 20 mM taurine, 5.7 mM ATP, 14.3 mM phosphocreatine, 6.56 mM MgCl 2 ·6H 2 O and 50 mM MES; (pH 7.1, 295 mosmol/kgH 2 O)). Muscle fibers were gently separated under a light microscope and fiber bundles were then permeabilized in preservation buffer with 50 µg/mL of saponin for 30 min. Permeabilized fiber bundles were transferred to ice cold respiration buffer (105 mM K-MES, 30 mM KCl, 1 mM EGTA, 10 mM K 2 HPO4, 5 mM MgCl 2 ·6H 2 O, 20 µm blebbistatin and 2.5 mg/mL bovine serum albumin (BSA, pH 7.4, 290 mosmol/kgH 2 O)) and remained on ice until high-resolution respirometry (OROBOROS Instruments, Innsbruck, Austria) was performed. For carbohydrate (CHO)-supported respiration, fiber bundles (3-4 mg) were used to assess respiration supported with pyruvate (5 mM) and malate (2 mM) with subsequent addition of ADP (5 mM) and glutamate (5 mM). Maximal CHO-supported respiratory capacity was determined in the presence of these substrates with subsequent addition of succinate (5 mM). Subsequently, respiration was assessed under Complex I and Complex V (i.e., ATP synthase) inhibition with the use of Rotenone (0.01 mM) and Oligomycin (10 µg/mL), respectively. Fatty acid (FA)-supported respiration was assessed in separate fiber bundles (also 3-4 mg) exposed to palmitoylcarnitine (0.075 mM) and malate (1 mM). Maximal FA-supported respiratory capacity was determined in the presence of these substrates with subsequent addition of ADP (5 mM). Assessments of mitochondrial respiratory capacity were performed in a blinded fashion throughout the NO 3 − /NO 2 − intervention.

Statistical Analyses
Data is reported as means ± standard deviation for the primary group (i.e., all individuals participating in the study) and sub-group (i.e., individuals from whom muscle biopsies were obtained). Two-sample t-tests were used for continuous variable comparisons, whereas Chi-square tests were used for categorical variable comparisons between non-diabetic controls and T2DM participants before the intervention. For examination of potential benefits of combined NO 3 − /NO 2 − supplementation, individual delta changes for each dependent variable were calculated for each group (i.e., placebo and NO 3 − /NO 2 − supplementation). Then, an Analysis of Covariance (ANCOVA) was used to test differences for each dependent variable with pre-intervention values for that same variable as a covariate. This approach allowed the comparison of the main effects of the intervention while controlling for potential baseline differences between placebo and NO 3 − /NO 2 − supplemented subjects that might have originated from random group assignments. Due to the lower number of subjects providing muscle biopsies, ANCOVA was generally underpowered in this subgroup and additional analyses (e.g., within group paired t test) were conducted to identify potential trends originating from the intervention. Chi-square tests were also used to compare the proportion of participants improving mitochondrial respiration or not under CHO-and FA-supported conditions among those in the NO 3 − /NO 2 − supplementation and placebo groups. Pearson's partial correlation coefficients and related coefficients of determination (R 2 ) were established between plasma NO 3 − or NO 2 − levels and VO 2 max after supplementation. Grubb's test identified one subject of the placebo sub-group as an outlier due to atypical changes in mitochondrial respiration during the intervention. Their data was excluded from such analyses. Statistical significance was determined a priori at p < 0.05. ANCOVA analyses were performed using SPSS (version 27.0; IBM Corp., Armonk, NY, USA), whereas all other analyses were conducted using Prism (version 9.0.0 (121); GraphPad Software, San Diego, CA, USA).

Baseline Characteristics
T2DM patients (N = 65) and non-diabetic controls (N = 50) were originally screened. Thirty-six (36) patients diagnosed with T2DM (primary group of T2DM patients) and 15 control subjects (primary group of controls) signed consents and completed the study ( Figure 1). Muscle biopsies were obtained in 10 of the 15 control subjects (sub-group of controls) and 19 of the 36 T2DM participants (sub-group of T2DM). All clinical baseline characteristics of primary groups and sub-groups are shown in Table 1.
Diagram illustrating enrollment of patients with type 2 diabetes mellitus (T2DM) and non-diabetic control participants for the entire study (Primary group). The number of participants that provided muscle biopsies (Sub-group) is also shown. (*) Non-diabetic control subjects that did not participate in the study included 11 subjects who did not qualify based on screening questionnaire (either because of too much regular exercise, specific medications, or other studies they were enrolled in), 2 subjects because bifurcation of their brachial artery was identified via Doppler examination and 22 subjects that were either out of contact or did not match to our diabetic population. (**) T2DM patients that did not participate in the study included 5 subjects because of medications used, 1 subject because of chronic kidney disease, 1 subject because of current smoking status, 1 subject misdiagnosed with T2DM, 2 subjects because their body mass index was too high, 14 subjects out of contact or declining to participate prior to consenting, 4 subjects who did not complete the entire study and 1 subject who did not reach VO 2 max in one of the exercise capacity evaluations. The 36 T2DM patients enrolled in the study were assigned via random number generator to receive either placebo or nitrate/nitrite supplementation with a one-to-one allocation. Note: the final number of participants studied with muscle biopsies was influenced by: (i) subjects not consenting to the procedure; or (ii) technical problems occurring in at least one of the testing visits (ex: not enough muscle tissue and/or extensive proportion of intramuscular fat obtained). Note: BRJ = beetroot juice.

Clinical Characteristics and Medications
As expected, only T2DM subjects were on medications for glycemic control and had elevated blood glucose and HbA1c levels. In addition, a higher proportion of T2DM subjects vs. non-diabetic controls were taking statins and angiotensin receptor blockers (72 vs. 33% and 25 vs. 0%, respectively). Medications used by participants in the placebo and NO 3 − /NO 2 − supplementation groups were comparable. Clinical characteristics and medications used by non-diabetic controls and T2DM subjects in the sub-group that provided muscle biopsies were consistent with those observed in the original primary sample, except that the reported duration of T2DM was higher among those in the NO 3 − /NO 2 − group vs. placebo.

Plasma NO Metabolites, VO 2 max and Work Rate Capacity in T2DM and Non-Diabetic Controls
As shown in Table 2, T2DM patients had higher plasma NO 3 − both in the primary group and sub-group when compared to non-diabetic controls (100% and 117.5%, respectively; p < 0.001). Conversely, plasma NO 2 − levels were lower in the T2DM patients in both the primary group (−46.8%; p < 0.0001) and sub-group (−42.0%; p < 0.001). VO 2 max was decreased in the primary group of T2DM patients in comparison to non-diabetic controls and this deficit was preserved among participants that provided muscle biopsies (−26.4% and −29.3%, respectively; p < 0.01). Similarly, the maximal individual work rates of T2DM patients were lower in the primary group and sub-group in relation to non-diabetic controls (−23.6% and −30.1%, respectively; p < 0.01). Data are presented as mean ± standard deviation. ** p < 0.05, *** p < 0.001 and **** p < 0.0001 vs. respective non-diabetic control group.

Effects of Combined NO3 − /NO2 − Supplementation
Combined NO3 − /NO2 − supplementation did not alter BMI and variables associated to glycemic control in T2DM in the primary group [26] or sub-groups of subjects (Supplemental Table S1). In addition, no changes in medication occurred during the 8-week intervention in either the placebo or NO3 − /NO2 − groups. . Data are presented as mean ± SD. * p < 0.05; ** p < 0.01.
Changes in plasma NO3 − and/or NO2 − resulting from supplementation were n related with changes in VO2max. However, plasma NO2 − , but not NO3 − , was signif correlated with VO2max after supplementation in both the primary group (r = 0.

Skeletal Muscle Oxidative Capacity
Mean group values of CHO-and FA-supported respiration did not significa change with NO3 − /NO2 − supplementation or placebo ( Figure 5A-D). We did not also serve significant differences with NO3 − /NO2 − supplementation in relation to placebo ANCOVA analyses (p = 0.16 and p = 0.39, respectively, for maximal CHO-and FA-s ported respiration; Supplementary Figure S4A,B). ANCOVA analysis also did not re changes in citrate synthase protein levels between NO3 − /NO2 − supplementation and cebo, but pre-vs. post-supplementation t test analyses showed trends for increases in protein with NO3 − /NO2 − supplementation (Supplementary Figure S4C-E). We then formed additional qualitative analyses comparing the proportion of individuals in cebo and NO3 − /NO2 − groups that presented increases in mitochondrial respiration ab 15.3%, which has been reported as the coefficient of variation for high resolu

Skeletal Muscle Oxidative Capacity
Mean group values of CHO-and FA-supported respiration did not significantly change with NO 3 − /NO 2 − supplementation or placebo ( Figure 5A-D). We did not also observe significant differences with NO 3 − /NO 2 − supplementation in relation to placebo via ANCOVA analyses (p = 0.16 and p = 0.39, respectively, for maximal CHO-and FAsupported respiration; Supplementary Figure S4A,B). ANCOVA analysis also did not reveal changes in citrate synthase protein levels between NO 3 − /NO 2 − supplementation and placebo, but pre-vs. post-supplementation t test analyses showed trends for increases in CS protein with NO 3 − /NO 2 − supplementation (Supplementary Figure S4C-E). We then performed additional qualitative analyses comparing the proportion of individuals in placebo and NO 3 − /NO 2 − groups that presented increases in mitochondrial respiration above 15.3%, which has been reported as the coefficient of variation for high resolution respirometry of permeabilized human muscle fibers [37]. Interestingly, the proportion of subjects that presented improvements in both CHO-and FA-supported respiration was 42% (5 of 12) after NO 3 − /NO 2 − supplementation vs. 0% receiving placebo (p < 0.05; Figure 5E-G). We considered these subjects as highly responsive to the intervention for skeletal muscle oxidative capacity. Despite the small N, this highly responsive sub-sample tended to present higher VO 2 max increases than the rest of the NO 3 − /NO 2 − group that provided biopsies (p = 0.09; Supplementary Table S2).
Nutrients 2022, 14, 4479 11 respirometry of permeabilized human muscle fibers [37]. Interestingly, the proportio subjects that presented improvements in both CHO-and FA-supported respiration 42% (5 of 12) after NO3 − /NO2 − supplementation vs. 0% receiving placebo (p < 0.05; Fi 5E-G). We considered these subjects as highly responsive to the intervention for ske muscle oxidative capacity. Despite the small N, this highly responsive sub-sample ten to present higher VO2max increases than the rest of the NO3 − /NO2 − group that prov biopsies (p = 0.09; Supplementary Table S2).  Figure 5E,F (i.e., highly responsive to the supplementation in terms of chondrial respiration) are denoted by "#". * p < 0.05. (D) Fatty Acid (FA)-supported mitochondrial respiration in the same conditions as described in C. (E) Waterfall plot denoting individual % changes in CHO-supported mitochondrial respiration after intervention in both placebo and NO 3 − /NO 2 − groups. Shaded area represents CV of high resolution respirometry in permeabilized human muscle fibers [37]. (F) Waterfall plot denoting individual % changes in FA-supported mitochondrial respiration as described in E. (G) Percentage of individuals that presented increases in both CHO-supported and FA-supported respiration (black section of columns) vs. others (grey section of columns). Data are presented as mean ± SD for A, B, C and D. Common subjects in Figure 5E,F (i.e., highly responsive to the supplementation in terms of mitochondrial respiration) are denoted by "#". * p < 0.05.

Discussion
Here, we extend previous findings [5,6] and demonstrate that T2DM patients had decreased VO 2 max (approx. −26%) in comparison to age-and BMI-matched non-diabetic controls. Contrary to select previous reports [38,39], these differences were not associated with a decreased percentage of slow twitch muscle fibers (MyHC I) (Figure 2A) or mitochondrial content (as indicated by comparable skeletal muscle CS expression) between T2DM and controls (Supplementary Figure S1). Although other mitochondrial proteins could have been decreased in T2DM muscles, collectively our results suggest that most of the deficit in skeletal muscle oxidative capacity observed in T2DM patients was due to dysregulation/dysfunction of mitochondria rather than its decreased content. It is also well established that T2DM patients present decreased NO bioavailability [17]. In fact, to our knowledge this is the first study reporting beneficial effects of combined NO 3 − /NO 2 − supplementation on VO 2 max and skeletal muscle oxidative capacity in T2DM patients. The observed 5.3% increase in VO 2 max resulting from this nutraceutical or dietetic intervention lasting only eight weeks has very important clinical and therapeutical implications. First, this represents a mean improvement of~1.1 mL/(kg*min) in the group of 18 T2DM participants supplemented with NO 3 − /NO 2 − , which may lead to a 9-10% decrease of all-cause mortality risk [4]. Second, cost of this simple therapeutic approach is low, which should lead to high and broad adherence in this population. Several other interesting observations arise from the present investigation.
Our findings demonstrate that the reduced NO bioavailability reported in T2DM patients, which was primarily associated with decreases in NOS-dependent NO production [17], is also associated with decreased circulating levels of NO 2 − , but not NO 3 − , when compared to age and BMI-matched controls. Individuals with several cardiovascular risk factors have been shown to present low plasma NO 2 − when compared to healthier controls [40] and our results now expand these findings to type 2 diabetic patients. The mechanisms responsible for the low plasma NO 2 − in T2DM remain to be established and may involve multiple processes including alterations of the oral microbiome and reduced NOS activity [41][42][43][44][45]. Nevertheless, our supplementation regimen with a daily dose of 250 mg of NO 3 − and 20 mg of NO 2 − for eight weeks was successful in increasing circulating levels of these NO metabolites in T2DM subjects, as previously reported [26,28]. Because the amount of NO 2 − provided orally was 4-6 times lower than in previous studies targeting its direct supplementation [46,47], the dose used here can be considered relatively low and the~40% increase of NO 2 − in the circulation is likely to have resulted from both its direct supplementation and its production from supplemented NO 3 − . This is relevant as it suggests that larger improvements in VO 2 max may be feasible with supplementation regimens that lead to larger increases in circulating NO 2 − . Indeed, mean NO 2 − levels in T2DM after supplementation were still at 74.6% of those in our group of non-diabetic controls. Considering the safety and feasibility of chronic NO 3 − supplementation [26,28,48] and the fact that higher doses of NO 2 − have been administered for longer periods such as 12 weeks [46,47], future studies examining the impact of larger doses of NO 3 − and/or NO 2 − (combined with a longer period of supplementation) on VO 2 max of T2DM patients are needed.
Although this novel pilot work has clinically significant indications for the management or treatment of T2DM, some aspects related to the supplementation need to be further investigated in future studies. Exercise itself is an attractive therapy for patients with T2DM, given that muscle contraction occurring with exercise can lower blood glucose independent of insulin [49] and the cardiorespiratory benefits of regular exercise in T2DM are well documented [7,50,51]. On initial screening, most subjects reported not being involved in any form of structured exercise and all participants were asked not to alter their physical activity habits during the intervention. Therefore, the potential interaction of NO 3 − /NO 2 − supplementation and exercise needs to be addressed in more detail in future investigations. Another important point to consider in relation to study design is how other aspects of combination therapy may influence the outcomes observed. For example, supplementation of vitamin C and other antioxidants may improve the proportion of bioavailable nitrate/nitrite by inhibiting the production of potentially harmful nitrosative compounds during NO 3 − /NO 2 − supplementation [52]. As such, follow-up studies are needed to understand if vitamin C supplementation (in combination with NO 3 − /NO 2 − ) can provide additional benefits in older adults, particularly those with T2DM. Another important consideration is the potential effects of body composition changes throughout the intervention that may not be detected through assessing BMI alone. We found no differences in BMI for the primary group of T2DM subjects in the study, nor sub-group of T2DM subjects who provided muscle biopsies (Supplemental Table S1). Similar studies with nitrate or antioxidant supplementation for longer durations (10-11 weeks) in younger adults have also not found changes in body composition [53,54]. However, a potential change in body composition throughout this nutrient intervention should be investigated in future randomized controlled trials in older adults and/or T2DM patients. The potential influence of medications on the outcomes of NO 3 − /NO 2 − supplementation also deserves attention. First, it is important to note that in the present study subjects were instructed to refrain from any medications in the morning of tests in order to avoid potential acute effects on the variables assessed. In addition, there were no differences in the proportion of subjects in the placebo and NO 3 − /NO 2 − supplemented groups taking each medication (Table 1) and no changes were self-reported during the intervention. However, future studies examining potential interactions between NO 3 − /NO 2 − and specific medications in T2DM are still needed. Of particular relevance to the present study, more diabetics were taking angiotensin receptor blockers (ARBs) and statin medications than non-diabetic controls. While ARBs do not seem to have deleterious effects in skeletal muscle, the effects of statins on reducing strength and integrity of skeletal muscle fibers and disrupting mitochondrial function have been documented and reviewed [55][56][57]. Still, the number of T2DM patients taking statins (or ARBs) was not different between the placebo and NO 3 − /NO 2 − groups. Furthermore, the use of statins (or ARBs) was also not different between the highly responsive individuals in the NO 3 − /NO 2 − group when compared to the rest of subjects in that same group. Collectively, these observations indicate that the outcomes reported from the supplementation in T2DM were not confounded by ARB or statin use. Nevertheless, whether NO 3 − /NO 2 − supplementation can be particularly beneficial for T2DM taking statins still need to be addressed in future investigations.
Despite the relatively limited number of subjects providing muscle biopsies and the considerations outlined above, our observations on these still provide important insights into the potential mechanisms involved in the improved VO 2 max seen in T2DM patients. Decreases in VO 2 max among T2DM patients that provided muscle biopsies were accompanied by similar decreases in skeletal muscle oxidative capacity (i.e., −29.3% for VO 2 max and −26.1% and −25.5% for maximal CHO-and FA-supported respiration, respectively; Table 2 and Figure 2B,C). However, despite >90% of T2DM subjects (both in the large primary group or among those that provided muscle biopsies) presenting increases in VO 2 max after NO 3 − /NO 2 − supplementation, only 42% of subjects that provided muscle biopsies presented improvements in maximal CHO-and FA-supported respiration. These highly responsive individuals did not present increases in CS protein when compared with the rest of the group suggesting that their superior benefits relied more on improving mitochondrial function than content (Supplemental Table S2). Additionally, the magnitudes of increase in mitochondrial respiratory capacity observed in these individuals (i.e., average of 81% and 82% under maximal CHO-supported and FA-supported respiration, respectively) were disproportionately higher than the increase in VO 2 max in this sub-group (i.e., 5.4%). Collectively, these observations indicate that those improvements in skeletal muscle oxidative capacity are not the main mechanism by which NO 3 − /NO 2 − supplementation leads to smaller and yet consistent increases in VO 2 max of T2DM patients. Regarding this matter, supplemental oxygen has been shown to improve in vivo muscle oxidative phosphorylation in type 2 diabetics [58]. Additionally, we have recently demonstrated that forearm skeletal muscle perfusion to handgrip exercise improves~16% after NO 3 − /NO 2 − supplementation in the same T2DM patients studied here [26]. Given that leg muscles are actively engaged during cycle ergometry (when VO 2 max testing was performed), similar improvements in blood flow and oxygen delivery were likely to be occurring in lower limbs. Therefore, in vivo maximal skeletal muscle oxygen consumption could have increased with supplementation, thereby contributing to the improved VO 2 max observed, independently of changes in maximal mitochondrial respiratory capacity.
In summary, we report beneficial effects of combined NO 3 − /NO 2 − supplementation for eight weeks on VO 2 max of T2DM patients, which should have important clinical implications as this may decrease all-cause mortality risks in this population. Maximal CHO-and FA-supported skeletal muscle oxidative capacity was also robustly improved in some individuals. Future studies should examine if increased doses and duration of the supplementation and, further, whether combined supplementation with an exercise regimen can lead to larger increases in VO 2 max of T2DM patients and to a larger proportion of subjects becoming responsive in terms of skeletal muscle oxidative capacity. Additionally, poor vascular function and mitochondrial dysfunction in skeletal muscle likely contribute to the high prevalence of sarcopenia reported in T2DM patients [59,60]. It will be important to examine if NO 3 − and/or NO 2 − supplementation can help preserve muscle mass and contractile function in this clinical population, logically extending our current findings in skeletal muscle.

Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/nu14214479/s1, Figure S1: Citrate Synthase (CS) expression in T2DM patients and non-diabetic controls; Figure S2: Plasma nitrate (NO 3 − ) and nitrite (NO 2 − ) levels after combined NO 3 − /NO 2 − supplementation vs. Placebo in T2DM participants that provided muscle biopsies (Sub-group); Figure S3: Impact of combined nitrate/nitrite (NO 3 − /NO 2 − ) supplementation on maximal oxygen uptake capacity (VO 2 max) in T2DM participants that provided muscle biopsies (Sub-group); Figure S4: Changes in skeletal muscle mitochondria respiratory capacity and citrate synthase expression resulting from combined NO 3 − /NO 2 − supplementation in T2DM participants; Table S1: Clinical post-supplementation characteristics of T2DM subjects that provided muscle biopsies (Sub-group); Table S2: Baseline characteristics and delta changes with supplementation of highly responsive (in terms of skeletal muscle oxidative capacity) vs. little or not responsive T2DM subjects in the NO 3 − /NO 2 − group that provided muscle biopsies (Sub-group).

Institutional Review Board Statement:
The study was conducted in accordance with the Declaration of Helsinki, and approved by the Institutional Review of the University of Iowa (protocol code 201511802, date of initial approval 5 December 2016).

Informed Consent Statement:
Informed consent was obtained from all subjects involved in the·study.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.