Nitric oxide synthase-independent generation of nitric oxide in rat skeletal muscle ischemia-reperfusion injury.

We have used electron paramagnetic resonance to investigate the time course of nitric oxide (NO) generation and its susceptibility to inhibitors of nitric oxide synthase (NOS) in ischemia-reperfusion (IR) injury to rat skeletal muscle in vivo. Significant levels of muscle nitroso-heme complexes were detected 24 h postreperfusion, but not after at 0.05, 3, and 8 h of reperfusion. The levels of muscle nitroso-heme complexes were not decreased by the NOS inhibitor N-nitro-L-arginine methyl ester as a single dose (30 mg/kg) prior to reperfusion or as multiple doses continued throughout the reperfusion (total administered, 120 mg/kg) or by the potent NOS inhibitor S-methylisothiourea (3 mg/kg). In contrast, nitroso-heme levels were reduced by the glucocorticoid dexamethasone (2.5 mg/kg). Muscle necrosis in vitro did not result in the formation of nitroso-heme complexes. The finding that reperfusion after ischemia is necessary for NO formation suggests that an inflammatory pathway is responsible for NOS-independent NO formation in IR injury to skeletal muscle.     

Endogenous vascular remodeling in ischemic skeletal muscle: a role for nitric oxide.

To test the hypothesis that nitric oxide (NO) production is essential for endogenous vascular remodeling in ischemic skeletal muscle, 22 New Zealand White rabbits were chronically instrumented with transit-time flow probes on the common iliac arteries and underwent femoral ligation to produce unilateral hindlimb ischemia. Iliac blood flow and arterial pressure were recorded at rest and during a graded exercise test. An osmotic pump connected to a femoral arterial catheter continuously delivered N-nitro-l-arginine methyl ester (a NO synthase inhibitor) or a control solution (N-nitro-d-arginine methyl ester or phenylephrine) to the ischemic limb over a 2-wk period. At 1, 3, and 6 wk after femoral ligation, maximal treadmill exercise blood flow in the ischemic limb was reduced compared with baseline in each group. However, maximal exercise blood flow was significantly (P < 0.05) lower in the l-NAME-treated group than in controls for the duration of the study: 48 +/- 4 vs. 60 +/- 5 ml/min at 6 wk. Consistent with the reduction in maximal blood flow response, the duration of voluntary exercise was also substantially (P < 0.05) shorter in the l-NAME-treated group: 539 +/- 67 vs. 889 +/- 87 s. Resting blood flow was unaffected by femoral ligation in either group. The results of this study show that endogenous vascular remodeling, which partially alleviated the initial deficit in blood flow, was interrupted by NO synthase inhibition. Therefore, we conclude that NO is essential for endogenous collateral development and angiogenesis in ischemic skeletal muscle in the rabbit.     

Involvement of nitric oxide synthase in skeletal muscle adaptation to chronic overload.

The purpose of this study was to determine the necessity of nitric oxide (NO) for hypertrophy and fiber-type transition in overloaded (OL) skeletal muscle. Endogenous NO production was blocked by administering N(G)-nitro-L-arginine methyl ester (L-NAME; 0.75 mg/ml; approximately 100 mg x kg-1 x day-1) in drinking water. Thirty-eight female Sprague-Dawley rats (approximately 250 g) were randomly divided into four groups: control-nonoverloaded (Non-OL), control-OL, L-NAME-Non-OL, and L-NAME-OL. Chronic overload of the plantaris was induced bilaterally by surgical removal of the gastrocnemius and soleus. Rats in the Non-OL groups received sham surgeries. L-NAME treatment began 24 h before surgery and continued until the rats were killed 14 days postsurgery. Although OL induced hypertrophy in both control (+76%) and L-NAME (+39%) conditions (P < 0.05), mean plantaris-to-body mass ratio in the L-NAME-OL group was significantly lower (P < 0.05) than that in the control-OL group. Microphotometric analysis of histochemically determined fiber types revealed increases in cross-sectional area (P < 0.05) for all fiber types (types I, IIA, and IIB/X) in the OL plantaris from control rats, whereas L-NAME-OL rats exhibited increases only in type I and IIB/X fibers. SDS-PAGE analysis of myosin heavy chain (MHC) composition in the plantaris indicated a significant (P < 0.05) OL effect in the control rats. Specifically, the mean proportion of type I MHC increased 6% (P < 0.05), whereas the proportion of type IIb MHC decreased approximately 9% (P < 0.05). No significant OL effects on MHC profile were observed in the L-NAME rats. These data support a role of NO in overload-induced skeletal muscle hypertrophy and fiber-type transition.     

An improved histochemical method for measuring nitric oxide synthase activity

The previous quantitative histochemical method for measuring nitric oxide synthase (NOS) activity in tissue sections involved the loss of about 15 per cent of the NOS, presumably from the section into the reaction medium. Two changes are now described. The first is concerned with the preparation in the laboratory of the active reagent, lead ammonium citrate/acetate (LACA). The second change involves an improvement of the procedure for measuring NOS activity. The new method appears to retain all the measurable NOS activity inside the section. Copyright © 1999 John Wiley & Sons, Ltd.     

Evidence that nitric oxide increases glucose transport in skeletal muscle

Balon, Thomas W., and Jerry L. Nadler. Evidence thatnitric oxide increases glucose transport in skeletal muscle.J. Appl. Physiol. 82(1): 359-363, 1997.Nitric oxide synthase (NOS) is expressed in skeletal muscle.However, the role of nitric oxide (NO) in glucose transport in thistissue remains unclear. To determine the role of NO in modulatingglucose transport, 2-deoxyglucose (2-DG) transport was measured in ratextensor digitorum longus (EDL) muscles that were exposed to either amaximally stimulating concentration of insulin or to an electricalstimulation protocol, in the presence ofN^G-monomethyl-L-arginine,a NOS inhibitor. In addition, EDL preparations were exposed to sodiumnitroprusside (SNP), an NO donor, in the presence of submaximal andmaximally stimulating concentrations of insulin. NOS inhibition reducedboth basal and exercise-enhanced 2-DG transport but had no effect oninsulin-stimulated 2-DG transport. Furthermore, SNP increased 2-DGtransport in a dose-responsive manner. The effects of SNP and insulinon 2-DG transport were additive when insulin was present inphysiological but not in pharmacological concentrations. Chronictreadmill training increased protein expression of both type I and typeIII NOS in soleus muscle homogenates. Our results suggest that NO maybe a potential mediator of exercise-induced glucose transport.

     

Nitric oxide in skeletal muscle: inhibition of nitric oxide synthase inhibits walking speed in rats.

Nitric oxide (NO*) is a multifunctional messenger molecule generated by a family of enzymes called the nitric oxide synthases (NOSs). Although NOSs have been identified in skeletal muscle, specifically brain NOS (bNOS) and endothelial NOS (eNOS), their role has not been well clarified. The goals of this investigation were to (1) characterize the immunoreactivity, Ca(2+) dependence, and activity of NOS in human and rat skeletal muscle and (2) using a rat model, investigate the effect of chronic blockade of NOS on skeletal muscle structure and function. Our results showed that both human and rodent skeletal muscle had NOS activity. This NOS activity was similar to that of the endothelial and brain NOS isoforms in that it was calcium-dependent. However, Western blot analysis consistently showed that a polyclonal antibody raised against a peptide sequence of human inducible NOS (iNOS) reacted with a protein with a molecular weight (95 kDa) that was different from that of other NOS isoforms. RT-PCR analysis identified the mRNA expression of not only eNOS and bNOS but also iNOS in human and rat muscle. Inhibition of nitric oxide synthase in rats with N(omega)-nitro-L-arginine methyl ester (L-NAME) resulted in a progressive, severe reduction in walking speed (30-fold reduction in walking velocity at day 22, P < 0.001), muscle fiber cross-sectional area (40% reduction at day 22, P < 0.001), and muscle mass (40% reduction in dry weight at day 22, P < 0.01). Rats fed the same regimen of the enantiomer of L-NAME (d-NAME) had normal motor function, muscle fiber morphology, and muscle mass. Taken together, these results imply that there may be a novel nitric oxide synthase in muscle and that NO. generated from muscle may be important in muscle function.     

A method for measuring sarcoplasmic reticulum calcium uptake in skeletal muscle using Fura-2

We have presented an assay for measuring the rate of sarcoplasmic reticulum (SR) Ca²⁺ uptake and Ca²⁺ release in skeletal muscle homogenates using the fluorescent Ca²⁺ probe Fura-2. Using this assay, we investigated the effects of an elevated temperature (40°C) and lowered pH (6.8), two factors proposed to be involved in skeletal muscle fatigue, on SR Ca²⁺ uptake. The EDL muscle was found to have a higher rate of Ca²⁺ uptake than the soleus (34%). Exposure of the muscles to a raised temperature, but not a reduced pH, resulted in a reduction in the rate of Ca²⁺ uptake in both the EDL and soleus homogenates. This uptake process was blocked by cyclopiazonic acid (CPA) a specific inhibitor of the major transport protein of the sarcoplasmic reticulum, the Ca²⁺-ATPase. Calcium release was induced using AgNO₃ after loading of the vesicles during the uptake process. It was found that AgNO₃ was only effective in producing Ca²⁺ release in the EDL muscles. The soleus muscles did not release Ca²⁺ under varying [Mg²⁺] or with Hg²⁺ substitution for Ag⁺, suggesting that fast- and slow-twitch muscle fibres require different conditions for maximum Ca²⁺-release, or that different isoforms of the Ca²⁺ release channels are present in the different fibres.     

Regulation of nitric oxide production in response to skeletal muscle activation

Nitric oxide(NO) is synthesized in normal muscle fibers by the neuronal (nNOS) andthe endothelial (ecNOS) isoforms of nitric oxide synthase (NOS). NOcontributes to the regulation of several processes such asexcitation-contraction coupling and mitochondrial respiration. Weassessed in this study whether NO production is regulated in responseto an acute increase in muscle activation. Three groups ofanesthetized, tracheostomized, spontaneously breathing rats wereexamined after an experimental period of 3 h. Group 1 served as a control (no loading), whereasgroups 2 and3 were exposed to moderate and severeinspiratory resistive loads, respectively, which elicited trachealpressures of 30 and 70% of maximum, respectively. Ventilatory(diaphragm, intercostal, and transverse abdominis) and limb(gastrocnemius) muscles were excised at the end of the experimentalperiod and examined for NOS activity and NOS protein expression.Neither submaximal nor maximum tracheal pressures were altered after 3 h of resistive loading. Diaphragmatic and intercostal muscle NOSactivities declined significantly in response to moderate and severeloading, whereas those of transverse abdominis and gastrocnemiusmuscles remained unchanged. On the other hand, resistive loading had nosignificant effect on ventilatory and limb muscle NOS isoformexpression. We propose that a contraction-induced decline in muscle NOSactivity represents a compensatory mechanism through which musclecontractility and mitochondrial function are protected from theinhibitory influence of NO.

     

Stretch-induced nitric oxide modulates mechanical properties of skeletal muscle cells

In this study, we examined the hypothesis that stretch-induced (nitric oxide) NO modulates the mechanical properties of skeletal muscles by increasing accumulation of protein levels of talin and vinculin and by inhibiting calpain-induced proteolysis, thereby stabilizing the focal contacts and the cytoskeleton. Differentiating C₂C₁₂ myotubes were subjected to a single 10% step stretch for 0–4 days. The apparent elastic modulus of the cells, E_app, was subsequently determined by atomic force microscopy. Static stretch led to significant increases (P < 0.01) in E_app beginning at 2 days. These increases were correlated with increases in NO activity and neuronal NO synthase (nNOS) protein expression. Expression of talin was upregulated throughout, whereas expression of vinculin was significantly increased only on days 3 and 4. Addition of the NO donor L-arginine onto stretched cells further enhanced E_app, NOS activity, and nNOS expression, whereas the presence of the NO inhibitor N-nitro-L-arginine methyl ester (L-NAME) reversed the effects of mechanical stimulation and of L-arginine. Overall, viscous dissipation, as determined by the value of hysteresis, was not significantly altered. For assessment of the role of vinculin and talin stability, cells treated with L-NAME showed a significant decrease in E_app, whereas addition of a calpain inhibitor abolished the effect. Thus our results show that NO inhibition of calpain-initiated cleavage of cytoskeleton proteins was correlated with the changes in E_app. Together, our data suggest that NO modulates the mechanical behavior of skeletal muscle cells through the combined action of increased talin and vinculin levels and a decrease in calpain-mediated talin proteolysis. mechanical stimulation; apparent elastic modulus; skeletal muscle cells; nitric oxide; stretch     

Agonist-dependent variablity of contributions of nitric oxide and prostaglandins in human skeletal muscle.

The relative contributions of endothelium-dependent dilators [nitric oxide (NO), prostaglandins (PGs), and endothelium-derived hyperpolarizing factor (EDHF)] in human limbs are poorly understood. We tested the hypothesis that relative contributions of NO and PGs differ between endothelial agonists acetylcholine (ACh; 1, 2, and 4 microg.dl(-1).min(-1)) and bradykinin (BK; 6.25, 25, and 50 ng.dl(-1).min(-1)). We measured forearm blood flow (FBF) using venous occlusion plethysmography in 50 healthy volunteers (27 +/- 1 yr) in response to brachial artery infusion of ACh or BK in the absence and presence of inhibitors of NO synthase [NOS; with NG-monomethyl-L-arginine (L-NMMA)] and cyclooxygenase (COX; with ketorolac). Furthermore, we tested the idea that the NOS + COX-independent dilation (in the presence of L-NMMA + ketorolac, presumably EDHF) could be inhibited by exogenous NO administration, as reported in animal studies. FBF increased approximately 10-fold in the ACh control; L-NMMA reduced baseline FBF and ACh dilation, whereas addition of ketorolac had no further effect. Ketorolac alone did not alter ACh dilation, but addition of L-NMMA reduced ACh dilation significantly. For BK infusion, FBF increased approximately 10-fold in the control condition; L-NMMA tended to reduce BK dilation (P < 0.1), and addition of ketorolac significantly reduced BK dilation. Similar to ACh, ketorolac alone did not alter BK dilation, but addition of L-NMMA reduced BK dilation. To test the idea that NO can inhibit the NOS + COX-independent portion of dilation, we infused a dose of sodium nitroprusside (NO-clamp technique) during ACh or BK that restored the reduction in baseline blood flow due to L-NMMA. Regardless of treatment order, the NO clamp restored baseline FBF but did not reduce the NOS + COX-independent dilation to ACh or BK. We conclude that the contribution of NO and PGs differs between ACh and BK, with ACh being more dependent on NO and BK being mostly dependent on a NOS + COX-independent mechanism (EDHF) in healthy young adults. The NOS + COX-independent dilation does not appear sensitive to feedback inhibition from NO in the human forearm.     

Effect of nitric oxide synthase inhibition on mitochondrial biogenesis in rat skeletal muscle.

The purpose of this study was to determine whether nitric oxide synthase (NOS) inhibition decreased basal and exercise-induced skeletal muscle mitochondrial biogenesis. Male Sprague-Dawley rats were assigned to one of four treatment groups: NOS inhibitor N(G)-nitro-l-arginine methyl ester (l-NAME, ingested for 2 days in drinking water, 1 mg/ml) followed by acute exercise, no l-NAME ingestion and acute exercise, rest plus l-NAME, and rest without l-NAME. The exercised rats ran on a treadmill for 53 +/- 2 min and were then killed 4 h later. NOS inhibition significantly (P < 0.05; main effect) decreased basal peroxisome proliferator-activated receptor-gamma coactivator 1beta (PGC-1beta) mRNA levels and tended (P = 0.08) to decrease mtTFA mRNA levels in the soleus, but not the extensor digitorum longus (EDL) muscle. This coincided with significantly reduced basal levels of cytochrome c oxidase (COX) I and COX IV mRNA, COX IV protein and COX enzyme activity following NOS inhibition in the soleus, but not the EDL muscle. NOS inhibition had no effect on citrate synthase or beta-hydroxyacyl CoA dehydrogenase activity, or cytochrome c protein abundance in the soleus or EDL. NOS inhibition did not reduce the exercise-induced increase in peroxisome proliferator-activated receptor-gamma coactivator 1alpha (PGC-1alpha) mRNA in the soleus or EDL. In conclusion, inhibition of NOS appears to decrease some aspects of the mitochondrial respiratory chain in the soleus under basal conditions, but does not attenuate exercise-induced mitochondrial biogenesis in the soleus or in the EDL.     

Upregulation of neuronal nitric oxide synthase in skeletal muscle by swim training

Exercise enhances cardiac output and blood flow to working skeletal muscles but decreases visceral perfusion. The alterations in nitric oxide synthase (NOS) activity and/or expression of the cardiopulmonary, skeletal muscle, and visceral organs induced by swim training are unknown. In sedentary and swim-trained rats (60 min twice/day for 3-4 wk), we studied the alterations in NOS in different tissues along with hindquarter vasoreactivity in vivo during rest and mesenteric vascular bed reactivity in vitro. Hindquarter blood flow and conductance were reduced by norepinephrine in both groups to a similar degree, whereas N(G)-nitro-L-arginine methyl ester reduced both indexes to a greater extent in swim-trained rats. Vasodilator responses to ACh, but not bradykinin or S-nitroso-N-acetyl-penicillamine, were increased in swim-trained rats. Ca(2+)-dependent NOS activity was enhanced in the hindquarter skeletal muscle, lung, aorta, and atria of swim-trained rats together with increased expression of neuronal NOS in the hindquarter skeletal muscle and endothelial NOS in the cardiopulmonary organs. Mesenteric arterial bed vasoreactivity was unaltered by swim training. Physiological adaptations to swim training are characterized by enhanced hindquarter ACh-induced vasodilation with upregulation of neuronal NOS in skeletal muscle and endothelial NOS in the lung, atria, and aorta.     

Reperfusion injury is reduced in skeletal muscle by inhibition of inducible nitric oxide synthase.

This study evaluated the effects of the selective inducible nitric oxide synthase (iNOS) inhibitor N-[3-(aminomethyl)benzyl]acetamidine (1400W) on the microcirculation in reperfused skeletal muscle. The cremaster muscles from 32 rats underwent 5 h of ischemia followed by 90 min of reperfusion. Rats received either 3 mg/kg 1400W or PBS subcutaneously before reperfusion. We found that blood flow in reperfused muscles was <45% of baseline in controls but sharply recovered to near baseline levels in 1400W-treated animals. There was a significant (P < 0.01 to P < 0.001) difference between the two groups at each time point throughout the 90 min of reperfusion. Vessel diameters remained <80% of baseline in controls during reperfusion, but recovered to the baseline level in the 1400W group by 20 min, and reached a maximum of 121 +/- 14% (mean +/- SD) of baseline in 10- to 20-micro m arterioles, 121 +/- 6% in 21- to 40-micro m arterioles, and 115 +/- 8% in 41- to 70-micro m arteries (P < 0.01 to P < 0.001). The muscle weight ratio between ischemia-reperfused (left) and non-ischemia-reperfused (right) cremaster muscles was 193 +/- 42% of normal in controls and 124 +/- 12% in the 1400W group (P < 0.001). Histology showed that neutrophil extravasation and edema were markedly reduced in 1400W-treated muscles compared with controls. We conclude that ischemia-reperfusion leads to increased generation of NO from iNOS in skeletal muscle and that the selective iNOS inhibitor 1400W reduces the negative effects of ischemia-reperfusion on vessel diameter and muscle blood flow. Thus 1400W may have therapeutic potential in treatment of ischemia-reperfusion injury.     

Endotoxin-induced skeletal muscle contractile dysfunction: contribution of nitric oxide synthases

The aims of thisstudy were to assess the role of nitric oxide (NO) and the contributionof different NO synthase (NOS) isoforms in skeletal muscle contractiledysfunction in septic shock. Four groups of conscious rats wereexamined. Group 1 served as control; groups 2, 3, and4 were injected withEscherichia coli endotoxin [lipopolysaccharide (LPS), 20 mg/kg ip] and killed after 6, 12, and 24 h, respectively. Protein expression was assessed byimmunoblotting and immunostaining. LPS injection elicited a transientexpression of the inducible NOS isoform, which peaked 12 h after LPSinjection and disappeared within 24 h. This expression coincided with a significant increase in nitrotyrosine formation (peroxynitrite footprint). Muscle expression of the endothelial and neuronal NOSisoforms, by comparison, rose significantly and remained higher thancontrol levels 24 h after LPS injection. In vitro measurement of musclecontractility 24 h after LPS injection showed that incubation with NOSinhibitor (S-methyliosothiourea)restored the decline in submaximal force generation, whereas maximalmuscle force remained unaffected. We conclude that NO plays asignificant role in muscle contractile dysfunction in septic animalsand that increased NO production is due to induction of the inducibleNOS isoform and upregulation of constitutive NOS isoforms.

     

Reperfusion injury in skeletal muscle is reduced in inducible nitric oxide synthase knockout mice.

Inducible nitric oxide synthase (iNOS) participates in many pathological events, and selective inhibition of iNOS has been shown to reduce ischemia-reperfusion (I/R) injury in different tissues. To further confirm its role in this injury process, I/R injury was observed in denervated cremaster muscles of iNOS-deficient (iNOS-/-) and wild-type mice. After 3-h ischemia and 90-min reperfusion, blood flow in reperfused muscle was 80 +/- 8.5% (mean +/- SE) of baseline at 10-min reperfusion and completely returned to the preischemia baseline after 20 min in iNOS-/- mice. In contrast, blood flow was 32 +/- 7.4% at 10 min and increased to 60 +/- 20% of the baseline level at 90 min in wild-type mice (P < 0.001 vs. iNOS-/- mice at all time points). The increased muscle blood flow in iNOS-/- mice was associated with significantly less vasospasm in all three sizes of arterial vessel size categories. The weight ratio to the contralateral muscle not subjected to I/R was greater in wild-type mice (173 +/- 11%) than in iNOS-/- mice (117 +/- 3%; P < 0.01). Inflammation and neutrophil extravasation were also more severe in wild-type mice. Western blot analysis demonstrated an absence of iNOS protein band in iNOS-/- mice and upregulation of iNOS protein expression in wild-type mice. Our results confirm the importance of iNOS in I/R injury. Upregulated iNOS exacerbates I/R injury and appears to be a therapeutic target in protection of tissues against this type of injury.     

Doxorubicin chemotherapy affects intracellular and interstitial nitric oxide concentrations in skeletal muscle

The role of nitric oxide (NO) in doxorubicin (DOX; cancer chemotherapeutic)-induced cardiotoxicity is well established. In skeletal muscle (SM), NO regulation plays a critical role in health, biogenesis, and function. Despite the increasing evidence that indicates the negative impact of DOX on SM function, the effect of DOX on NO production in SM has yet to be examined. The purpose of the current study was to simultaneously examine intracellular and interstitial NO concentrations in the SM following the administration of DOX. A single dose of 1.5 or 4.5 mg/kg was administered intraperitoneally to male Sprague Dawley rats, and interstitial (IS) and intracellular (IC) NO was quantified every 24 up to 192 h post-injection. There was no significant difference in IC NO following the injection of 1.5 mg/kg DOX when compared to the control; however, the administration of 4.5 mg/kg DOX resulted in lower (P?<?0.05) concentrations of NO in the IC. Interestingly, a consistently higher (P?<?0.05) concentration of NO in the IS was established following the administration of 1.5 mg/kg compared to the control while no significant changes in IS NO resulted from the administration of the 4.5 mg/kg dose. The fluctuation of IS and IC NO was not a result of substrate availability as arginine concentrations remained stable throughout the experiment. By utilizing the microdialysis technique, we have simultaneously quantified for the first time the IS and IC concentrations of NO in SM following DOX administration. These data provide important insight in the possible mechanisms leading to DOX-related SM dysfunction.     

Leucine degradation in cell-free extracts of skeletal muscle.

Since skeletal muscle is the major site in the body for oxidation of leucine, isoleucine and valine, the pathway and control of leucine oxidation were investigated in cell-free preparations of rat muscle. Leucine was found to be transaminated to 4-methyl-2-oxopentanoate, which was then oxidatively decarboxylated. On differential centrifugation 70--80% of the transaminase activity was recovered in the soluble fraction of the cell, and the remaining amount in the mitochondrial fraction. The transaminase, from both fractions had similar pH optima and both were markedly inhibited by Ca2+. Thus changes in cellular Ca2+ concentration may regulate transaminase activity. Both transaminases had a much higher affinity for 2-oxoglutarate than for pyruvate. Therefore the utilization of amino groups from leucine for the biosynthesis of alanine in muscle [Odessey, Khairallah & Goldberg (1974) J. Biol. Chem. 249, 7623--7629] in vivo involves transamination with 2-oxoglutarate to produce glutamate, which is then transaminated with pyruvate to produce alanine. The dehydrogenase activity assayed by the decarboxylation of methyl-2-oxo[1-14C]pentanoate was localized exclusively in the fraction containing mitochondria and required NAD+, CoA and thiamin pyrophosphate for optimal activity. Measurements of competitive inhibition suggested that the oxo acids of leucine, isoleucine and valine are all decarboxylated by the same enzyme. The enzyme activity was decreased by 90% upon freezing or sonication and was stimulated severalfold by Mg2+, K+ and phosphate ions. In addition, it was markedly inhibited by ATP, but not by non-metabolizable analogues. This observation suggests that splitting of ATP is required for inhibition. The oxidative decarboxylation of 4-methyl-2-oxopentanoate by the dehydrogenase appears to be the rate-limiting step for leucine oxidation in muscle homogenates and also in intact tissues. In fact, rat muscles incubated with [1-14C]leucine release 1-14C-labelled oxo acid into the medium at rates comparable with the rate of decarboxylation. Intact muscles also released the oxo acids of [1-14C]valine or [1-14C]isoleucine, but not of other amino acids. These findings suggest that muscle is the primary source of the branched-chain oxo acids found in the blood.