Share this article on:

Effect of l-Arginine Infusion on Glucose Disposal during Exercise in Humans


Medicine & Science in Sports & Exercise: September 2011 - Volume 43 - Issue 9 - p 1626-1634
doi: 10.1249/MSS.0b013e318212a317
Basic Sciences

Purpose: We have previously shown that local infusion of a nitric oxide synthase (NOS) inhibitor attenuates increases in leg glucose uptake during exercise in humans. We have also shown that infusion of the NOS substrate, l-arginine (l-Arg), increases glucose clearance, although the mechanisms involved were not determined. A potential mechanism for NO-mediated glucose disposal is via interactions with NOS and the energy sensor AMP-activated protein kinase (AMPK). The aim of this study was to determine the mechanism(s) by which l-Arg infusion increases glucose disposal during exercise in humans by examining total NOS activity and AMPK signaling.

Methods: Seven males cycled for 120 min at 64% ± 1% V˙O2peak, during which the [6,6-2H]glucose tracer was infused. During the final 60 min of exercise, either saline alone (Control, CON), or saline containing l-Arg HCl (l-Arg, 30 g at 0.5 g·min−1) was coinfused in a double-blind, randomized, counterbalanced order.

Results: l-Arg increased the glucose rate of disappearance and glucose clearance rate during exercise; however, this was accompanied by a 150% increase in plasma insulin concentration from 65 to 75 min (P < 0.05) that remained significantly elevated until 90 min of exercise. Skeletal muscle AMPK signaling, nNOSμ phosphorylation by AMPK, and total NOS activity increased to a similar extent in the two trials.

Conclusions: The increase in glucose disposal after l-Arg infusion during exercise is likely due to the significantly higher plasma insulin concentration.

1Exercise Physiology and Metabolism Laboratory, Department of Physiology, The University of Melbourne, Victoria, AUSTRALIA; 2School of Community Health, Charles Sturt University, Albury, New South Wales, AUSTRALIA; 3School of Exercise and Nutrition Sciences, Centre for Physical Activity and Nutrition Research, Deakin University, Burwood Victoria, AUSTRALIA; and 4Institute of Sport, Exercise and Active Living and School of Biomedical and Health Sciences, Victoria University, Victoria, AUSTRALIA

Address for correspondence: Glenn K. McConell, Ph.D., Institute of Sport, Exercise and Active Living and Biomedical and Health Sciences, Victoria University, Victoria 8001, Australia; E-mail:

Submitted for publication July 2010.

Accepted for publication January 2011.

Insulin and contraction both increase skeletal muscle GLUT4 translocation and glucose uptake; however, the mechanisms involved differ (25). We have evidence from rodents (34) and humans (5,22) that nitric oxide (NO) production during contraction may be essential for the regulation of skeletal muscle glucose uptake during exercise. We have previously reported that local femoral artery infusion of the competitive NO synthase (NOS) inhibitor, N G-monomethyl-l-arginine (l-NMMA) into healthy men during exercise attenuates the increase in leg glucose uptake, without influencing leg blood flow or plasma insulin concentration (5). Interestingly, the attenuation in leg glucose uptake seemed to be reversed by coinfusion of the NOS substrate l-arginine (l-Arg) (5). In a follow-up study in people with type 2 diabetes and age-matched controls, l-NMMA infusion attenuated leg glucose uptake to a greater extent in the diabetic than in the control participants (22).

Our laboratory has also shown that infusion of l-Arg increases glucose disposal during exercise in well-trained men (26). We proposed that l-Arg infusion, by increasing the substrate availability for NOS (3), may have increased NO production and therefore mediated the increase in glucose uptake via a NO-dependent mechanism (26). Indeed, systemic infusion of l-Arg at rest in humans increases the content of the NO second messenger cGMP in both plasma and urine and elevates urine levels of the NO breakdown product, nitrate (2). In addition, l-Arg incubation increases NO production of isolated rodent skeletal muscle (1).

Given that local infusion of the NOS inhibitor NG-nitro-l-arginine methyl ester (l-NAME) attenuates the increase in NOS activity (and skeletal muscle glucose uptake) during in situ muscle contractions in rats (34), we have assumed that contraction increased NOS activity and l-NMMA attenuated increases in NOS activity during exercise in our human studies (5,22). We have also assumed that skeletal muscle total NOS activity is increased to a greater extent during exercise when l-Arg is infused during exercise in humans (5,26). Two studies have examined human skeletal muscle NOS activity at rest (9,20); however, no study to date has examined skeletal muscle NOS activity during exercise in humans. Therefore, in the current study, we examined whether skeletal muscle NOS activity was increased during exercise and also whether l-Arg infusion further increased NOS activity during exercise in humans.

It is also possible, however, that the increased glucose disposal observed in our previous study (26) was due, at least in part, to increases in plasma insulin concentration. Although not statistically significant, there was a tendency for l-Arg infusion to increase plasma insulin concentration during exercise in our previous human study (26). Indeed, l-Arg is a known insulin secretagogue, and infusion of l-Arg at rest significantly increases plasma insulin concentration and insulin-stimulated glucose disposal in humans (13). The effect of l-Arg infusion on plasma insulin concentration during exercise in humans has not previously been closely examined.

Therefore, the aim of the current study was to examine the mechanisms by which l-Arg infusion increases glucose disposal during exercise in humans by assessing skeletal muscle total NOS activity and also closely examining the time course of plasma insulin concentration changes during exercise. A potential mechanism for NO-mediated glucose disposal is the ubiquitously expressed energy sensor, AMP-activated protein kinase (AMPK). AMPK activity increases during contraction and is reported to increase skeletal muscle glucose uptake (7,15). There is also evidence for interactions between AMPK and NOS (7,34). Indeed, Higaki et al. have previously shown that high levels (5 mM) of the NO donor, sodium nitroprusside, increase basal AMPK activity and glucose uptake (18). As such, it may be that l-Arg infusion during exercise increases skeletal muscle NO availability, which activates AMPK and increases glucose uptake. Therefore, we also examined AMPK activity and AMPK phosphorylation and the phosphorylation status of the major skeletal muscle isoform of NOS, neuronal NOS (nNOS), by AMPK during exercise. Because the serine-threonine kinase AKT (protein kinase B) is an important signaling protein for insulin-stimulated glucose uptake (21), we also examined skeletal muscle AKT signaling during exercise with and without l-arginine infusion. We hypothesized that l-Arg infusion would increase glucose disposal by causing greater increases in skeletal muscle total NOS activity during exercise, independent of AMPK and insulin signaling in humans.

Back to Top | Article Outline



Seven recreationally active nonsmoker males provided informed written consent to participate in this study, which was approved by the Human Research Ethics Committee of the University of Melbourne and was conducted in accordance with the Declaration of Helsinki. Participants' characteristics were as follows: age = 23 ± 1 yr, weight = 72 ± 4 kg, and peak pulmonary oxygen consumption during cycling (V˙O2peak) = 3.3 ± 0.3 L·min−1 (mean ± SEM).

Back to Top | Article Outline

Preliminary Testing and Diet Control

Participants were required to attend the laboratory on four separate occasions with 1-2 wk between each visit. The first visit involved a cycling V˙O2peak test that was determined during a graded exercise test to voluntary exhaustion on an ergometer (Lode, Groningen, The Netherlands). Participants returned to the laboratory on a separate day for a 30-min familiarization ride at a workload calculated to be ∼65% V˙O2peak. Approximately 1 wk later, participants returned to the laboratory for the first of two exercise trials, which involved cycling for 120 min at ∼65% V˙O2peak (128 ± 8 W).

All participants were instructed to refrain from any formal exercise and to avoid ingesting alcohol and caffeine for 24 h before each exercise trial. To ensure that the energy intake before each trial was consistent, subjects were asked to complete a food diary during the day before their first experimental trial, which was photocopied and returned. They were asked to match the food diary for the second trial and to consume water ad libitum. Participants were instructed to finish eating by 10 p.m. the evening before the experimental trial and attended the laboratory in a fasted state. Between trials, participants were instructed to maintain their regular exercise patterns.

Back to Top | Article Outline

Experimental Trials

On the morning of each exercise trial, a catheter was inserted into an antecubital forearm vein for blood sampling and another into the contralateral arm for infusion of the stable isotope of glucose ([6,6-2H]glucose; Cambridge Isotope Laboratories, Cambridge, MA) and coinfusion of l-arginine or saline control. After a basal blood sample, a bolus of 45.2 ± 0.4 μmol·kg−1 of the glucose tracer was administered intravenously and was immediately followed by a constant (0.72 ± 0.02 μmol·kg−1) 240-min infusion, which was continued until the end of exercise. After 120 min of rest, the exercise protocol consisted of cycling for 120 min. Blood was sampled at −120, −30, and −10 min and immediately before the commencement of exercise and every 15 min during exercise for the measurement of plasma glucose and [6,6-2H]glucose. Plasma glycerol, free fatty acids, and insulin were measured at rest and every 30 min of exercise. In addition, from time 60 to 90 min, blood was sampled every 5 min to closely examine insulin concentration, which has been reported to peak 10 min after the commencement of l-Arg infusion at rest (13). Expired air was collected into Douglas bags for 3 min every 15 min during exercise, and HR (Polar Favor, Oulu, Finland) was recorded every 30 min. Muscle was obtained from the vastus lateralis under local anesthesia, using the percutaneous needle biopsy technique, with suction. Before the commencement of exercise, two separate incisions were made ∼1 cm apart, and muscle was sampled in a distal-to-proximal order at rest and after exercise (120 min). Muscle biopsies were obtained immediately after exercise with the needle containing the muscle frozen in liquid N2 within 8-12 s of the subjects ceasing exercise. One leg (randomly selected) was used for each of the experimental trials. Participants received 8 mL·kg−1 body weight of water at the start of exercise, followed by a further 2 mL·kg−1 body weight every 15 min of exercise and were fan-cooled throughout the trial. In a double-blind randomized crossover design, participants received a coinfusion during the second 60 min of exercise of either 30 g of l-arginine hydrochloride (Ophthalmic Laboratories for Pharmalab, Brookvale, NSW, Australia) mixed with saline (l-Arg; 0.5 g·min−1 intravenously) or a placebo control treatment of 0.9% saline (0.5 g·min−1 intravenously; CON). Infusion of 30 g of l-arginine for 30-60 min in humans raises plasma l-arginine concentration from ∼0.1 to ∼6.2-7.2 mM (3).

Back to Top | Article Outline

Blood Analysis

Plasma glucose, lactate, and glycerol were determined using enzymatic fluorometric procedures; plasma nonesterified fatty acid (NEFA) was determined by an enzymatic colorimetric method (NEFA-C test; Wako, Osaka, Japan); and plasma insulin was determined using a human insulin-specific radioimmunoassay kit (Linco Research, St. Charles, MO). Glucose kinetics were estimated at rest and during exercise using a modified one-pool non-steady-state model as previously described (24,26). Briefly, the rapidly mixing portion of the glucose pool was assumed to be 0.65, and the apparent glucose space was estimated to be 25% of body mass. The technique used in this investigation estimates the rates of plasma glucose appearance (glucose Ra) and disappearance (glucose Rd) from the changes in the percentage enrichment of [6,6-2H]glucose and the plasma glucose concentration. Glucose clearance rate (glucose CR) was calculated by dividing the glucose Rd by the plasma glucose concentration.

Back to Top | Article Outline

Muscle Analysis

Preparation of skeletal muscle lysate.

Frozen muscle was homogenized in ice-cold lysis buffer (10 μL·mg−1 tissue; 50 mM Tris-HCl, pH 7.5 containing 1 mM EDTA, 1 mM EGTA, 10% v/v glycerol, 1% v/v Triton X-100, 50 mM NaF, 5 mM Na4P2O7, 1 mM DTT, 1 mM PMSF, 1 μL·mL−1 trypsin inhibitor, and 5 μL·mL−1 protease inhibitor cocktail (P8340; Sigma, St. Louis, MO)), incubated on ice for 20 min and centrifuged at 16,000g for 20 min at 4°C. The protein concentration of samples was determined using the bicinchoninic acid protein assay (Pierce, Rockford, IL) with bovine serum albumin as the standard.

Back to Top | Article Outline


Polyclonal rabbit antibodies specific for phospho-AKT Thr308, total AKT, and total AMPKα protein were purchased from Cell Signaling Technology (Beverly, MA). The phospho-specific antibodies AMPKα Thr172 and ACCβ Ser221 were purchased from Upstate (Lake Placid, NY), and AMPKα1 and α2 antibodies were purchased from BD Transduction Laboratories (Franklin Lakes, NJ). The predominant skeletal muscle nNOS isoform (23), nNOSμ, was detected with a monoclonal mouse antibody (BD Transduction Laboratories), whereas the phospho-specific nNOSμ Ser1451 phosphorylation site was detected with a polyclonal rabbit antibody (7). ACCβ was detected using IRDye™ 700-labeled streptavidin (Rockland, Gilbertsville, PA).

Back to Top | Article Outline


Skeletal muscle lysates (80 μg) were boiled in Laemmli sample buffer and subjected to SDS-PAGE. Binding of purified proteins was detected by immunoblotting after an overnight incubation with the primary antibody. Membranes were incubated in Odyssey antirabbit IRDye™ 800- or antimouse IRDye™ 700-labeled secondary antibody (Rockland) and were scanned for infrared fluorescence using an Odyssey Infrared Imaging System (LI-COR Biosciences, Lincoln, NE). When both total protein and protein phosphorylation were measured, membranes were probed first for total protein and were stripped of antibodies (2% SDS in 25 mM glycine, pH 2.0); successful stripping was verified by incubating with the secondary antibody followed by infrared detection. Stripped membranes were then reprobed with the anti-phospho antibody. Phosphorylation was expressed relative to the total protein (24,34).

Back to Top | Article Outline

NOS activity.

Total NOS activity (including contributions from both eNOS and nNOS) was determined in tissue extracts by measuring the production of radiolabeled l-citrulline from radiolabeled l-arginine. It should be noted, however, that we have shown previously that almost all NOS expressed in human skeletal muscle is nNOS with very little eNOS expressed (27). Frozen muscle was homogenized in ice-cold homogenizing buffer (10 μL·mg−1 tissue; 250 mM Tris-HCl, pH 7.4, 10 mM EDTA, 10 mM EGTA). NOS activity of the supernatant was then determined using a NOS Activity Assay Kit (Cat 781001; Cayman Chemicals, Ann Arbor, MI) as per the manufacturer's instructions. Ten microliters of the supernatant was combined with 40 μL of reaction mix (in final concentrations, 25 mM Tris-HCl (pH 7.4), 3 μM tetrahydrobiopterin, 1 μM flavin adenine dinucleotide, 1 μM flavin adenine mononucleotide, 1.25 mM NADPH, 0.75 mM CaCl2 and 3 μM l- [14C] arginine monochloride (Amersham Biosciences, Piscataway, NJ)) for 30 min at 37°C in the presence or absence of 1 mM of the NOS inhibitor NG-nitro-l-arginine (l-NNA). The concentration of l-NNA was sufficient to fully block NOS activity (data not shown). NOS activity was calculated as the difference between samples incubated in the presence or absence of l-NNA, and values are expressed as picomoles of l-[14C] citrulline formation per minute per milligram of protein (34).

Back to Top | Article Outline

AMPK activity.

Skeletal muscle lysates (50 μg) were combined with 15 μL of protein A sepharose beads (Pierce), bound to either AMPKα1 or AMPKα2 polyclonal antibodies, and incubated for 2 h at 4°C. Immunocomplexes were washed in lysis buffer containing 0.5 M NaCl and resuspended in 25 μL of 0.05 M Tris buffer (pH 7.5). To commence the assay, 25 μL of reaction mixture containing (in final concentrations) 50 mM Tris-HCl pH 7.5, 0.1 mM EGTA, 0.1% (by volume) 2-mercaptoethanol, 10 mM magnesium acetate, 0.1 mM [32P]ATP (∼200 cpm·pmol−1; Perkin Elmer Life and Analytical Sciences, Boston, MA), 30 μM AMARA peptide (Upstate), and 200 μM AMP was added to each sample at 30°C for 20 min with agitation. Forty microliters of each sample was then transferred onto P81 chromatography paper and washed 3 × 10 min in 75 mM H3PO4, once in 100% ethanol, and air-dried. The P81 paper was then placed in an organic scintillation fluid (Opti-Fluor O; Perkin Elmer), and radioactivity was counted on a β counter (Perkin Elmer). The AMARA peptide has the same AMPK phosphorylation site as ACCβ; therefore, AMPK activities were calculated as units of γ-[32P]ATP incorporated into the AMARA peptide (ACCα (73-87)A77) per minute per milligram of total protein subjected to immunoprecipitation (28).

Back to Top | Article Outline

Statistical Analysis

Data are expressed as mean ± SEM. The two experimental trials were compared using two-factor (trial × time) repeated-measures ANOVA using the SPSS statistical package, and if there was a significant interaction, a post hoc comparison using Fisher LSD test was conducted. Statistical significance was set at P < 0.05. The sample size of six participants was determined by a statistical power analysis (β = 0.2, α = 0.05, two-tailed t-test) of an expected increase of 100% in skeletal muscle AMPKα2 activity obtained from people after 2 h of cycling at ∼65% V˙O2peak.

Back to Top | Article Outline


Pulmonary gas measurements, substrate oxidation, and HR.

During the first 60 min of exercise (before the commencement of the CON or l-Arg infusion), there was no difference between oxygen consumption, RER, ventilation, and HR between the two trials (data not shown). During the second 60 min of exercise, there was no difference in oxygen consumption (2.23 ± 0.14 vs 2.23 ± 0.13 L·min−1), ventilation (53.3 ± 4.4 vs 55.9 ± 5.8 L·min−1), RER (0.91 ± 0.2 vs 0.91 ± 0.01), or HR (154 ± 5 vs 152 ± 5 beats·min−1) between the CON and l-Arg infusion trials (P > 0.05).

Back to Top | Article Outline

Plasma glucose, lactate, insulin, glycerol, and NEFA concentrations.

In both trials, plasma glucose decreased, whereas plasma lactate and glycerol increased during the 120 min of exercise. However, during the period of CON and l-Arg infusion, there was no significant difference in plasma glucose or lactate (P > 0.05; Figs. 1A, B) or plasma glycerol concentration (Fig. 2A) between trials. Plasma NEFA increased in both trials during exercise; however, it was significantly lower in the l-Arg trial at 90 min (P < 0.05; Fig. 2B). In the CON trial, plasma insulin levels decreased during the 120 min of exercise (P < 0.05; Fig. 1C). Infusion of l-Arg resulted in a rapid increase in plasma insulin concentration by ∼150% ± 40% compared with CON values at 65, 70, and 75 min, which remained significantly elevated by ∼50% ± 28% of CON values until 90 min (P < 0.05; Fig. 1C). At 105 min, there was a trend for increased plasma insulin concentration in the l-Arg trial compared with the CON trial (P = 0.06; Fig. 1C).





Back to Top | Article Outline

Glucose kinetics.

Glucose Ra, glucose Rd, and glucose CR increased during exercise in both trials (P < 0.05; Fig. 3). Infusion of l-Arg significantly increased glucose Ra during the final 45 min, glucose Rd over the final 30 min, and glucose CR at 90 and 105 min of exercise above that of the CON trial (P < 0.05).



Back to Top | Article Outline

AMPK signaling and AKT Thr308 phosphorylation.

Exercise significantly increased AMPKα1 activity in both the CON (62% ± 21%, P < 0.05; Fig. 4A) and l-Arg (85% ± 19%, P < 0.05; Fig. 4A) trials with no significant difference between trials. AMPKα2 activity was significantly elevated by 140% ± 51% after exercise in the CON trial (P < 0.05; Fig. 4B), and although it was 73% ± 12% higher after exercise than at rest in the l-Arg trial, this was not significant (P = 0.06). It must be noted that one sample was lost during the analysis of AMPKα2, and as such, the n = 6. Exercise significantly increased AMPKα Thr172 phosphorylation by 180% ± 63% in the CON trial (P < 0.05), and by 48% ± 16% in the l-Arg trial (P < 0.05; Fig. 5A), with no significant difference between trials. ACCβ-Ser221 phosphorylation increased to a similar extent in both trials (P < 0.05; Fig. 5B). Exercise significantly increased AKT Thr308 phosphorylation in both trials with no significant difference between the trials (P < 0.05; Fig. 5C).





Back to Top | Article Outline

Skeletal muscle NOS activity and nNOSμ Ser1451 phosphorylation.

Total NOS activity during exercise was significantly increased from rest by ∼90% ± 33% in both trials with no significant difference between the trials (P < 0.05; Fig. 6A). nNOSμ Ser1451 phosphorylation also increased significantly in both trials with no significant difference between trials (P < 0.05; Fig. 6B).



Back to Top | Article Outline


This is the first study to demonstrate that skeletal muscle total NOS activity is increased with exercise in humans. It seems unlikely, however, that the increased rate of glucose disposal observed with l-Arg infusion is a direct result of increased NO production because total NOS activity increased similarly during exercise in both the CON and l-Arg infusion trials. In addition, l-Arg infusion did not affect AMPK signaling during exercise, so greater AMPK activation could not have accounted for the greater glucose disposal with l-Arg infusion. It is more likely that the large insulin response observed during the l-Arg infusion mediated the greater glucose disposal during exercise in that trial. It is also likely that the insulin surge increased glucose uptake in all insulin sensitive tissues and, as such, elicited a whole-body increase in glucose uptake. However, despite the higher insulin levels during exercise, skeletal muscle AKT phosphorylation after exercise was not higher in the l-Arg trial.

Exercise is characterized by a fall in plasma insulin concentration due to catecholamine inhibition of insulin secretion (12), and indeed, in the current study, plasma insulin concentration decreased during the first 60 min of exercise in both trials (Fig. 1C). l-Arg is a potent insulin secretagogue at rest, acting directly on pancreatic β cells to stimulate insulin secretion (16). Somewhat surprisingly, despite the inhibitory effects of exercise on insulin secretion, l-Arg infusion rapidly caused a very large increase in plasma insulin levels that peaked at approximately 150% above CON levels (Fig. 1C). In our previous study involving endurance-trained individuals, we saw only a trend for an increase in plasma insulin with l-Arg infusion (26). It is possible that there was a greater insulin response in the current study because active, not specifically endurance-trained, individuals were involved, whereas in our first study, well-trained endurance subjects were involved. Indeed, endurance training increases hepatic insulin clearance in rodents (40), and a single bout of exercise has been observed to increase whole body insulin clearance for up to 15 h after the cessation of exercise (29). However, it is also possible that, in our previous study, a higher level of plasma insulin may have been missed within the first 5-10 min of l-Arg infusion because blood samples were only collected every 15 min after the commencement of the l-Arg infusion in that study. Indeed, in the current study, plasma insulin concentration increased within 5-10 min after the commencement of the l-Arg infusion (Fig. 1C). Interestingly, the timing and pattern of the insulin response in the current exercise study were similar to what have been observed with l-Arg infusion in humans at rest (13). For example, a 30-min l-Arg infusion (1 g·min−1) at rest into young, healthy men and women has been shown to elicit a monophasic insulin response, peaking at 10 min and followed by a plateau that remained elevated for the duration of the 30-min infusion (13).

It is known that the addition of insulin has an additive effect on contraction-induced increases in skeletal muscle glucose disposal during exercise in humans (39). It is therefore likely that the substantial increase in the concentration of plasma insulin during the l-Arg trial in the present study would have had an additive effect on glucose uptake during exercise, resulting in the significant increases in glucose Rd and glucose CR (Figs. 3B, C) compared with the CON trial. To verify this potential additive signaling effect, we measured skeletal muscle AKT Thr308 phosphorylation because it closely parallels increases in insulin-stimulated glucose uptake after exercise (11) and is an appropriate marker of skeletal muscle insulin signaling (6). We also found that exercise increased AKT Thr308 phosphorylation in human skeletal muscle (Fig. 5C), which is consistent with some (19), but not all (35), studies. Surprisingly, however, AKT Thr308 phosphorylation at 120 min of exercise was similar in the two trials, despite the higher plasma insulin levels in the l-Arg trial (Fig. 5C). It is possible, however, that the lack of difference in AKT Thr308 phosphorylation was because plasma insulin levels in l-Arg had returned to normal levels by the end of exercise when the muscle biopsy was conducted (Fig. 5C). On the basis of these findings, future exercise studies should examine if there is an additive effect on skeletal muscle AKT phosphorylation at time points closer to the l-Arg-induced plasma insulin surge.

An important novel finding of the present study is that skeletal muscle total NOS activity increases during exercise in humans (Fig. 6). Although NO has been implicated as a potential signaling intermediate in contraction-mediated glucose uptake in humans, previous studies have used indirect measures of NO accumulation/NOS activity (5,22). Previously, human studies had only examined skeletal muscle NOS activity at rest (9,20), and it has been assumed that skeletal muscle NOS activity increases during exercise in humans. Our findings in humans are in line with similar results during exercise (33) and in situ contractions (34) in rats and increased NO release during contraction of isolated rat skeletal muscle (1).

We hypothesized that infusion of the NOS substrate, l-Arg, during exercise would augment the increased total NOS activity during exercise above what was observed in the control exercise trial. Indeed, others have shown that l-Arg incubation increases NO production in noncontracting rat muscle (1). However, we found no greater increase in total NOS activity during exercise when l-Arg was infused than during exercise alone. This may have been because the NOS enzymes in skeletal muscle could have been already saturated with l-Arg substrate before the infusion of l-Arg (32). The Km for nNOS (1.4-2.2 μM) and eNOS (2.9 μM) are much less than the plasma concentrations of l-Arg in healthy humans (∼100 μM), and as such, NOS should have been saturated in the control trial (8). Elevating plasma l-Arg, commonly with 30 g intravenously, such as in the present study, increases the plasma concentration in humans to ∼7000 μM (32).

Despite this, surprisingly, numerous studies (4,5,17) have reported increases in indirect indices of systemic NO production after l-Arg infusion such as increased leg blood flow, cGMP accumulation, and NOx production. This phenomenon, where increased l-Arg availability causes effects despite the NOS enzymes already likely to have been saturated with substrate, is known as the l-arginine paradox. Interestingly, it has also been suggested that many of the vascular effects of l-Arg infusion, such as vasodilation, inhibition of platelet aggregation, and reduced blood viscosity, are mediated by l-Arg-stimulated insulin release and not by l-Arg itself (13). It is not known if l-Arg infusion during exercise exerts its effects solely through increased insulin release. However, results from the present study indicate that, within human skeletal muscle during exercise, systemic l-Arg infusion does not increase total NOS activity above that of control exercise.

Although l-Arg is a precursor for the vasodilator NO, it has previously been shown that l-Arg infusion during exercise has no effect on either leg blood flow (5) or skeletal muscle blood flow (17) in humans. Therefore, it is unlikely that the l-Arg infusion increased glucose uptake during exercise by increasing muscle blood flow and therefore glucose delivery. It is, however, possible that the increase in plasma insulin concentration observed during the l-Arg infusion had a vasodilatory effect of its own (13). Although high levels of plasma insulin are required for an increase in total blood flow, lower, more physiological increases in plasma insulin increase blood flow through muscle capillaries (38). It is possible that the higher l-Arg availability and higher plasma insulin levels in the current study may have combined to increase capillary flow. It is known that increases in plasma insulin phosphorylate AKT, which then phosphorylates and activates eNOS in endothelial cells (30). Furthermore, NOS inhibition can prevent the increases in muscle capillary flow in response to increases in insulin (37).

It is well known that AMPK is phosphorylated and activated during exercise, and there is evidence to suggest that AMPK mediates glucose uptake during exercise (7). However, several studies have demonstrated dissociations between skeletal muscle AMPK activity and glucose uptake during exercise/muscle contraction (10). Interestingly, in the present study, AMPKα2 activity (Fig. 5B), which increased during exercise in the control trial, only tended to increase in the l-Arg trial (P = 0.06). It is not clear why AMPKα2 activation during exercise was blunted in the l-Arg trial. It is possible that the increases in plasma insulin with l-Arg infusion inhibited AMPKα2 activity; however, there is limited evidence to suggest that AMPK activity is attenuated in conditions of high insulin (31). In addition, insulin has no effect on rat skeletal muscle AMPK activity in vitro (15) or in humans with carbohydrate ingestion-mediated hyperinsulinemia during exercise (24).

There was a small, transient, yet significant attenuation of the increase in plasma NEFA levels at 90 min in the l-Arg trial (Fig. 2B). We found more pronounced reductions in plasma NEFA in our previous study during exercise in endurance-trained individuals infused with l-Arg (26). We think it is unlikely that the lower plasma NEFA concentrations in l-Arg accounted for the higher glucose disposal rate in that trial. Some (41), but not all (14), studies have shown that raising plasma NEFA levels has no effect on glucose disposal during exercise in humans. In addition, very large reductions in plasma NEFA concentration during exercise have no (36) or have only modest (41) effects on glucose disposal during exercise in humans.

In summary, we have shown for the first time that exercise increases skeletal muscle total NOS activity in humans. l-Arg infusion during exercise increased glucose uptake, but it seems that this was not via an increase in skeletal muscle NO production because there was no further increase in skeletal muscle NOS activity with l-Arg infusion during exercise. l-Arg infusion also had little effect on skeletal muscle AMPK activation. However, l-Arg infusion resulted in a very large increase in plasma insulin concentration during exercise, and because exercise and insulin are additive on skeletal muscle glucose uptake, it is possible that l-Arg infusion increases glucose uptake during exercise via increases in plasma insulin concentration.

This work was supported by a grant from the National Health and Medical Research Council of Australia (G.K.M.: 237002).

The authors have no conflicts of interest to declare.

The authors thank the participants for taking part in this study and acknowledge the technical assistance of Professor Benedict Canny, Dr. Sean McGee, and Vince Murone.

The results of the present study do not constitute endorsement by the American College of Sports Medicine.

Back to Top | Article Outline


1. Balon TW, Nadler JL. Nitric oxide release is present from incubated skeletal muscle preparations. J Appl Physiol. 1994;77(6):2519-21.
2. Bode-Boger SM, Boger RH, Creutzig A, et al. l-Arginine infusion decreases peripheral arterial resistance and inhibits platelet aggregation in healthy subjects. Clin Sci (Lond). 1994;87(3):303-10.
3. Bode-Boger SM, Boger RH, Galland A, Tsikas D, Frolich JC. l-Arginine-induced vasodilation in healthy humans: pharmacokinetic-pharmacodynamic relationship. Br J Clin Pharmacol. 1998;46(5):489-97.
4. Bode-Boger SM, Boger RH, Schroder EP, Frolich JC. Exercise increases systemic nitric oxide production in men. J Cardiovasc Risk. 1994;1(2):173-8.
5. Bradley SJ, Kingwell BA, McConell GK. Nitric oxide synthase inhibition reduces leg glucose uptake but not blood flow during dynamic exercise in humans. Diabetes. 1999;48(9):1815-21.
6. Casaubon L, Sajan MP, Rivas J, Powe JL, Standaert ML, Farese RV. Contrasting insulin dose-dependent defects in activation of atypical protein kinase C and protein kinase B/Akt in muscles of obese diabetic humans. Diabetologia. 2006;49(12):3000-8.
7. Chen ZP, Stephens TJ, Murthy S, et al. Effect of exercise intensity on skeletal muscle AMPK signaling in humans. Diabetes. 2003;52(9):2205-12.
8. Forstermann U, Closs EI, Pollock JS, et al. Nitric oxide synthase isozymes: characterization, purification, molecular cloning, and functions. Hypertension. 1994;23(6 Pt 2):1121-31.
9. Frandsen U, Hoffner L, Betak A, Saltin B, Bangsbo J, Hellsten Y. Endurance training does not alter the level of neuronal nitric oxide synthase in human skeletal muscle. J Appl Physiol. 2000;89(3):1033-8.
10. Fujii N, Hirshman MF, Kane EM, et al. AMP-activated protein kinase α2 activity is not essential for contraction- and hyperosmolarity-induced glucose transport in skeletal muscle. J Biol Chem. 2005;280(47):39033-41.
11. Funai K, Schweitzer GG, Sharma N, Kanzaki M, Cartee GD. Increased AS160 phosphorylation, but not TBC1D1 phosphorylation, with increased postexercise insulin sensitivity in rat skeletal muscle. Am J Physiol Endocrinol Metab. 2009;297(1):E242-51.
12. Galbo H, Holst JJ, Christensen NJ. The effect of different diets and of insulin on the hormonal response to prolonged exercise. Acta Physiol Scand. 1979;107(1):19-32.
13. Giugliano D, Marfella R, Verrazzo G, et al. The vascular effects of l-arginine in humans. The role of endogenous insulin. J Clin Invest. 1997;99(3):433-8.
14. Hargreaves M, Kiens B, Richter EA. Effect of increased plasma free fatty acid concentrations on muscle metabolism in exercising men. J Appl Physiol. 1991;70(1):194-201.
15. Hayashi T, Hirshman MF, Kurth EJ, Winder WW, Goodyear LJ. Evidence for 5′ AMP-activated protein kinase mediation of the effect of muscle contraction on glucose transport. Diabetes. 1998;47(8):1369-73.
16. Henquin JC, Meissner HP. Effects of amino acids on membrane potential and 86Rb+ fluxes in pancreatic β-cells. Am J Physiol. 1981;240(3):E245-52.
17. Hickner RC, Fisher JS, Ehsani AA, Kohrt WM. Role of nitric oxide in skeletal muscle blood flow at rest and during dynamic exercise in humans. Am J Physiol. 1997;273(1 Pt 2):H405-10.
18. Higaki Y, Hirshman MF, Fujii N, Goodyear LJ. Nitric oxide increases glucose uptake through a mechanism that is distinct from the insulin and contraction pathways in rat skeletal muscle. Diabetes. 2001;50(2):241-7.
19. Howlett KF, Sakamoto K, Yu H, Goodyear LJ, Hargreaves M. Insulin-stimulated insulin receptor substrate-2-associated phosphatidylinositol 3-kinase activity is enhanced in human skeletal muscle after exercise. Metabolism. 2006;55(8):1046-52.
20. Kashyap SR, Roman LJ, Lamont J, et al. Insulin resistance is associated with impaired nitric oxide synthase activity in skeletal muscle of type 2 diabetic subjects. J Clin Endocrinol Metab. 2005;90(2):1100-5.
21. Katome T, Obata T, Matsushima R, et al. Use of RNA interference-mediated gene silencing and adenoviral overexpression to elucidate the roles of AKT/protein kinase B isoforms in insulin actions. J Biol Chem. 2003;278(30):28312-23.
22. Kingwell BA, Formosa M, Muhlmann M, Bradley SJ, McConell GK. Nitric oxide synthase inhibition reduces glucose uptake during exercise in individuals with type 2 diabetes more than in control subjects. Diabetes. 2002;51(8):2572-80.
23. Kobzik L, Reid MB, Bredt DS, Stamler JS. Nitric oxide in skeletal muscle. Nature. 1994;372(6506):546-8.
24. Lee AD, Hansen PA, Holloszy JO. Wortmannin inhibits insulin-stimulated but not contraction-stimulated glucose transport activity in skeletal muscle. FEBS Lett. 1995;361(1):51-4.
25. Lee-Young RS, Palmer MJ, Linden KC, et al. Carbohydrate ingestion does not alter skeletal muscle AMPK signaling during exercise in humans. Am J Physiol Endocrinol Metab. 2006;291(3):E566-73.
26. McConell GK, Huynh NN, Lee-Young RS, Canny BJ, Wadley GD. l-Arginine infusion increases glucose clearance during prolonged exercise in humans. Am J Physiol Endocrinol Metab. 2006;290(1):E60-6.
27. McConell GK, Lee-Young RS, Chen ZP, et al. Short-term exercise training in humans reduces AMPK signalling during prolonged exercise independent of muscle glycogen. J Physiol. 2005;568(Pt 2):665-76.
28. McGee SL, Mustard KJ, Hardie DG, Baar K. Normal hypertrophy accompanied by phosphorylation and activation of AMP-activated protein kinase α1 following overload in LKB1 knockout mice. J Physiol. 2008;586(6):1731-41.
29. Mikines KJ, Sonne B, Farrell PA, Tronier B, Galbo H. Effect of training on the dose-response relationship for insulin action in men. J Appl Physiol. 1989;66(2):695-703.
30. Montagnani M, Chen H, Barr VA, Quon MJ. Insulin-stimulated activation of eNOS is independent of Ca2+ but requires phosphorylation by Akt at Ser(1179). J Biol Chem. 2001;276(32):30392-8.
31. Park SH, Kim SY, Baek WK, et al. Regulation of glucose-dependent insulin secretion by insulin: possible role of AMP-activated protein kinase. Life Sci. 2009;85(3-4):178-83.
32. Rhodes P, Barr CS, Struthers AD. Arginine, lysine and ornithine as vasodilators in the forearm of man. Eur J Clin Invest. 1996;26(4):325-31.
33. Roberts CK, Barnard RJ, Jasman A, Balon TW. Acute exercise increases nitric oxide synthase activity in skeletal muscle. Am J Physiol. 1999;277(2 Pt 1):E390-4.
34. Ross RM, Wadley GD, Clark MG, Rattigan S, McConell GK. Local nitric oxide synthase inhibition reduces skeletal muscle glucose uptake but not capillary blood flow during in situ muscle contraction in rats. Diabetes. 2007;56(12):2885-92.
35. Treebak JT, Birk JB, Rose AJ, Kiens B, Richter EA, Wojtaszewski JF. AS160 phosphorylation is associated with activation of α2β2γ1- but not α2β2γ3-AMPK trimeric complex in skeletal muscle during exercise in humans. Am J Physiol Endocrinol Metab. 2007;292(3):E715-22.
36. van Loon LJ, Thomason-Hughes M, Constantin-Teodosiu D, et al. Inhibition of adipose tissue lipolysis increases intramuscular lipid and glycogen use in vivo in humans. Am J Physiol Endocrinol Metab. 2005;289(3):E482-93.
37. Vincent MA, Barrett EJ, Lindner JR, Clark MG, Rattigan S. Inhibiting NOS blocks microvascular recruitment and blunts muscle glucose uptake in response to insulin. Am J Physiol Endocrinol Metab. 2003;285(1):E123-9.
38. Vincent MA, Clerk LH, Lindner JR, et al. Microvascular recruitment is an early insulin effect that regulates skeletal muscle glucose uptake in vivo. Diabetes. 2004;53(6):1418-23.
39. Wasserman DH, Geer RJ, Rice DE, et al. Interaction of exercise and insulin action in humans. Am J Physiol. 1991;260(1 Pt 1):E37-45.
40. Wirth A, Holm G, Bjorntorp P. Effect of physical training on insulin uptake by the perfused rat liver. Metabolism. 1982;31(5):457-62.
41. Zderic TW, Davidson CJ, Schenk S, Byerley LO, Coyle EF. High-fat diet elevates resting intramuscular triglyceride concentration and whole body lipolysis during exercise. Am J Physiol Endocrinol Metab. 2004;286(2):E217-25.


©2011The American College of Sports Medicine