The Sick and the Weak: Neuropathies/Myopathies in the Critically Ill.

Critical illness polyneuropathies (CIP) and myopathies (CIM) are common complications of critical illness. Several weakness syndromes are summarized under the term intensive care unit-acquired weakness (ICUAW). We propose a classification of different ICUAW forms (CIM, CIP, sepsis-induced, steroid-denervation myopathy) and pathophysiological mechanisms from clinical and animal model data. Triggers include sepsis, mechanical ventilation, muscle unloading, steroid treatment, or denervation. Some ICUAW forms require stringent diagnostic features; CIM is marked by membrane hypoexcitability, severe atrophy, preferential myosin loss, ultrastructural alterations, and inadequate autophagy activation while myopathies in pure sepsis do not reproduce marked myosin loss. Reduced membrane excitability results from depolarization and ion channel dysfunction. Mitochondrial dysfunction contributes to energy-dependent processes. Ubiquitin proteasome and calpain activation trigger muscle proteolysis and atrophy while protein synthesis is impaired. Myosin loss is more pronounced than actin loss in CIM. Protein quality control is altered by inadequate autophagy. Ca(2+) dysregulation is present through altered Ca(2+) homeostasis. We highlight clinical hallmarks, trigger factors, and potential mechanisms from human studies and animal models that allow separation of risk factors that may trigger distinct mechanisms contributing to weakness. During critical illness, altered inflammatory (cytokines) and metabolic pathways deteriorate muscle function. ICUAW prevention/treatment is limited, e.g., tight glycemic control, delaying nutrition, and early mobilization. Future challenges include identification of primary/secondary events during the time course of critical illness, the interplay between membrane excitability, bioenergetic failure and differential proteolysis, and finding new therapeutic targets by help of tailored animal models.

N-acetylcysteine inhibits muscle fatigue in humans.

N-acetylcysteine (NAC) is a nonspecific antioxidant that selectively inhibits acute fatigue of rodent skeletal muscle stimulated at low (but not high) tetanic frequencies and that decreases contractile function of unfatigued muscle in a dose-dependent manner. The present experiments test the hypothesis that NAC pretreatment can inhibit acute muscular fatigue in humans. Healthy volunteers were studied on two occasions each. Subjects were pretreated with NAC 150 mg/kg or 5% dextrose in water by intravenous infusion. The subject then sat in a chair with surface electrodes positioned over the motor point of tibialis anterior, an ankle dorsiflexor of mixed-fiber composition. The muscle was stimulated to contract electrically (40-55 mA, 0.2-ms pulses) and force production was measured. Function of the unfatigued muscle was assessed by measuring the forces produced during maximal voluntary contractions (MVC) of ankle dorsiflexor muscle groups and during electrical stimulation of tibialis anterior at 1, 10, 20, 40, 80, and 120 Hz (protocol 1). Fatigue was produced using repetitive tetanic stimulations at 10 Hz (protocol 1) or 40 Hz (protocol 2); intermittent stimulations subsequently were used to monitor recovery from fatigue. The contralateral leg then was studied using the same protocol. Pretreatment with NAC did not alter the function of unfatigued muscle; MVC performance and the force-frequency relationship of tibialis anterior were unchanged. During fatiguing contractions stimulated at 10 Hz, NAC increased force output by approximately 15% (P < 0.0001), an effect that was evident after 3 min of repetitive contraction (P < 0.0125) and persisted throughout the 30-min protocol. NAC had no effect on fatigue induced using 40 Hz stimuli or on recovery from fatigue. N-acetylcysteine pretreatment can improve performance of human limb muscle during fatiguing exercise, suggesting that oxidative stress plays a causal role in the fatigue process and identifying antioxidant therapy as a novel intervention that may be useful clinically.

Cardiac-specific overexpression of tumor necrosis factor-alpha causes oxidative stress and contractile dysfunction in mouse diaphragm.

BACKGROUND: We have developed a transgenic mouse with cardiac-restricted overexpression of tumor necrosis factor-alpha (TNF-alpha). These mice develop a heart failure phenotype characterized by left ventricular dysfunction and remodeling, pulmonary edema, and elevated levels of TNF-alpha in the peripheral circulation from cardiac spillover. Given that TNF-alpha causes atrophy and loss of function in respiratory muscle, we asked whether transgenic mice developed diaphragm dysfunction and whether contractile losses were caused by oxidative stress or tissue remodeling. METHODS AND RESULTS: muscles excised from transgenic mice and littermate controls were studied in vitro with direct electrical stimulation. Cytosolic oxidant levels were measured with 2', 7'-dichlorofluorescin diacetate; emissions of the oxidized product were detected by fluorescence microscopy. Force generation by the diaphragm of transgenic animals was 47% less than control (13.2+/-0. 8 [+/-SEM] versus 25.1+/-0.6 N/cm(2); P:<0.001); this weakness was associated with greater intracellular oxidant levels (P:<0.025) and was partially reversed by 30-minute incubation with the antioxidant N:-acetylcysteine 10 mmol/L (P:<0.01). Exogenous TNF-alpha 500 micromol/L increased oxidant production in diaphragm of wild-type mice and caused weakness that was inhibited by N:-acetylcysteine, suggesting that changes observed in the diaphragm of transgenic animals were mediated by TNF-alpha. There were no differences in body or diaphragm weights between transgenic and control animals, nor was there evidence of muscle injury or apoptosis. CONCLUSIONS: Elevated circulating levels of TNF-alpha provoke contractile dysfunction in the diaphragm through an endocrine mechanism thought to be mediated by oxidative stress.

RyR1 modulation by oxidation and calmodulin.

Alteration of skeletal muscle function by reactive oxygen species and nitric oxide (NO) may involve regulation of the activity of the skeletal muscle Ca2+ release channel (also known as RyR1). We have shown that oxidants can activate RyR1 and produce inter-subunit disulfide bonds. Both effects are prevented by pretreatment with either NO donors or N-ethylmaleimide under conditions that modify less than 5% of the total sulfhydryls on RyR1. Oxidation-induced intersubunit crosslinking can also be prevented by the binding of either Ca2+ calmodulin or apocalmodulin to RyR1. Also, both Ca2+ calmodulin and apocalmodulin binding are blocked by oxidation of RyR1. In contrast, alkylation with N-ethylmaleimide or reaction with NO donors preferentially blocks apocalmodulin binding to RyR1, suggesting the existence of a regulatory cysteine within the apocalmodulin binding site. We have demonstrated that Ca2+ calmodulin and apocalmodulin bind to overlapping, but nonidentical, sites on RyR1 and that cysteine 3635 is close to or within the apocalmodulin-binding site on RyR1. This cysteine is also one of the cysteines that form the intersubunit disulfide bonds, suggesting that calmodulin binds at an intersubunit contact site. Our findings are consistent with a model in which oxidants regulate the activity of RyR1 directly by altering subunit-subunit interactions and indirectly by preventing the binding of either Ca2+-bound calmodulin or apocalmodulin. NO also has both a direct and an indirect effect: it blocks the ability of oxidants to generate intersubunit disulfide bonds and prevents apocalmodulin binding.

Mitochondria mediate tumor necrosis factor-alpha/NF-kappaB signaling in skeletal muscle myotubes.

Tumor necrosis factor-alpha (TNF-alpha) is implicated in muscle atrophy and weakness associated with a variety of chronic diseases. Recently, we reported that TNF-alpha directly induces muscle protein degradation in differentiated skeletal muscle myotubes, where it rapidly activates nuclear factor kappaB (NF-kappaB). We also have found that protein loss induced by TNF-alpha is NF-kappaB dependent. In the present study, we analyzed the signaling pathway by which TNF-alpha activates NF-kappaB in myotubes differentiated from C2C12 and rat primary myoblasts. We found that activation of NF-kappaB by TNF-alpha was blocked by rotenone or amytal, inhibitors of complex I of the mitochondrial respiratory chain. On the other hand, antimycin A, an inhibitor of complex III, enhanced TNF-alpha activation of NK-kappaB. These results suggest a key role of mitochondria-derived reactive oxygen species (ROS) in mediating NF-kappaB activation in muscle. In addition, we found that TNF-alpha stimulated protein kinase C (PKC) activity. However, other signal transduction mediators including ceramide, Ca2+, phospholipase A2 (PLA2), and nitric oxide (NO) do not appear to be involved in the activation of NF-kappaB.

Detection of reactive oxygen and reactive nitrogen species in skeletal muscle.

Reactive oxygen species (ROS) and reactive nitrogen species (RNS) are usually identified with pathological states and mediators of cellular injury. However, over the last decade ROS and RNS have been identified in skeletal muscle under physiological conditions. Detection of ROS and RNS production by skeletal muscle cells is fundamental to the problem of differentiating between physiological and pathological levels. The goal of this paper is to review the techniques that have been used to detect ROS and RNS in skeletal muscle. Electron spin resonance, fluorescent assays, cyotchrome c reduction, chemiluminescence, hydroxylation of salicylate, and nitration of phenylalanine are some of the assay systems that have been used thus far. A large body of evidence now indicates that ROS and RNS are continually produced by many different skeletal muscle types studied in vivo, in situ, and in vitro. Under resting conditions, ROS and RNS are detectable in both intracellular and extracellular compartments. Production increases during both non-fatiguing and fatiguing muscle contractions. In the absence of disease, the individual molecular species detected in skeletal muscle include parent radicals for the ROS and RNS cascades: superoxide anions and nitric oxide. Both are generated at rates estimated to range from pmol-to-nmol/mg muscle/minute. Evidence indicates that hydrogen peroxide, hydroxyl radicals, and peroxynitrite are also present under physiological conditions. However, the molecular species that mediate specific biological effects remains largely undetermined, as do the sources of ROS and RNS within muscle fibers. Eventual delineation of the mechanisms whereby ROS and RNS regulate cellular function will hinge on our understanding of the production and distribution of ROS and RNS within skeletal muscle.

Invited Review: redox modulation of skeletal muscle contraction: what we know and what we don't.

Over the past decade, reactive oxygen species (ROS) and nitric oxide (NO) derivatives have been established as physiological modulators of skeletal muscle function. This mini-review addresses the roles of these molecules as endogenous regulators of muscle contraction. The article is organized in two parts. First, established concepts are briefly outlined. This section provides an overview of ROS production by muscle, antioxidant buffers that oppose ROS effects, enzymatic synthesis of NO in muscle, the effects of endogenous ROS on contractile function, and NO as a contractile modulator. Second, a selected group of unresolved topics are highlighted. These more controversial issues include putative source(s) of regulatory ROS, the relative importance of the two NO synthase isoforms constitutively coexpressed by muscle fibers, molecular mechanisms of ROS and NO action, and the physiological relevance of redox regulation. By discussing current questions, as well as the established paradigm, this article is intended to further debate and stimulate research in this area.

N-acetylcysteine depresses contractile function and inhibits fatigue of diaphragm in vitro.

We have previously shown that antioxidant enzymes (superoxide dismutase and catalase) depress contractility of unfatigued diaphragm fiber bundles and inhibit development of acute fatigue. In the present study, we tested for similar effects of N-acetyl-cysteine (NAC), a nonspecific antioxidant approved for clinical use. Diaphragms were excised from deeply anesthetized rats. Fiber bundles were removed, mounted isometrically at 37 degrees C, and stimulated directly using supramaximal current intensity. Studies of unfatigued muscle showed that 10 mM NAC reduced peak twitch stress (P < 0.0001), shortened time to peak twitch stress (P < 0.002), and shifted the stress-frequency curve down and to the right (P < 0.05). Fiber bundles incubated in 0.1-10 mM NAC exhibited a dose-dependent decrease in relative stresses developed during 30-Hz contraction (P < 0.0001) with no change in maximal tetanic (200 Hz) stress. NAC (10 mM) also inhibited acute fatigue. Throughout 10 min of intermittent contraction at 30-40 Hz, treated bundles developed higher stresses than time-matched control bundles (P < 0.0001). NAC concentrations > or = 30 mM were toxic, causing a prompt irreversible decrease in maximal tetanic stress (P < 0.0001). Because NAC effects mimic the effects of other antioxidant agents with different mechanisms of action, we conclude that exogenous antioxidants exert stereotypical effects on contractile function that differ between unfatigued and fatiguing muscle. Unlike antioxidant enzymes, however, NAC has been approved for clinical use and may be used in future studies of human muscle fatigue.

Dimethyl sulfoxide depresses skeletal muscle contractility.

Dimethyl sulfoxide (DMSO) is commonly used in studies of skeletal muscle as a selective antioxidant (DMSO preferentially scavenges hydroxyl radicals) or as a solvent for drugs. The present experiments tested DMSO for direct effects on diaphragm contractile properties. Fiber bundles were removed from anesthetized rats, mounted in vitro at optimal length (37 degrees C), curarized, and stimulated directly. Protocol 1 tested for contractile depression and dose dependence by comparing bundles treated with DMSO (0.6-640 mM) with time- and stimulus-matched controls. Protocol 2 tested reversibility of 220 mM DMSO effects by using each bundle as its own control. DMSO decreased the relative forces developed during twitch and submaximal tetanic (30- and 60-Hz) contractions, shifting the force-frequency relationship down and to the right. These effects were strongly dose dependent and were reversed by DMSO washout. DMSO had no detectable effect on the forces developed during maximal tetany (200 Hz). DMSO depresses contractile function of diaphragm fibers by reversible dose-dependent inhibition of excitation-contraction coupling.

Effects of metabolic blockade on distribution of blood flow to respiratory muscles.

Sublethal inhibition of citrate metabolism in the tricarboxylic acid (TCA) cycle with monofluoroacetate (MFA) has been shown to cause a fivefold increase in myocardial blood flow without any change in cardiac output, blood pressure, or O2 consumption (C. Liang, J. Clin. Invest. 60: 61-69, 1977); however, blood flow did not increase to any organs examined other than the heart, including resting limb skeletal muscle. Preferential inhibition of glycolysis with iodoacetate (IA) failed to cause similar changes in distribution of blood flow. This unique response of myocardium to TCA cycle inhibition suggested a unique metabolic control of cardiac vasodilation. An alternate explanation is that MFA is preferentially concentrated in active muscle. After MFA, tissue citrate accumulates behind the block and the highest levels are reported in the heart and diaphragm, suggesting enhanced blockade or enhanced compensation in these two continuously active muscles. To test the hypothesis that vasodilation in the heart after MFA is not unique and that similar vasodilation will be evoked in active respiratory muscles, we measured blood flow to the myocardium, kidney, diaphragm, intercostals, transverse abdominals, and triceps brachii in anesthetized dogs using radionuclide-labeled microspheres, before and after MFA, and in another set of dogs before and after IA. Before MFA or IA, inspiratory loading significantly increased blood flow to active muscles of breathing in proportion to the added load. After MFA, blood flow to active muscles of breathing as well as to the heart became abnormally elevated with respect to mechanical work, and loading evoked no further increase in blood flow.(ABSTRACT TRUNCATED AT 250 WORDS)

Theophylline does not increase maximal tetanic force or diaphragm endurance in vitro.

These experiments tested the capacity of theophylline to improve diaphragm strength (maximal force development) and endurance (maintenance of force output during repeated contractions). Rodent diaphragm strips were mounted at optimal length in oxygenated Krebs-Ringer solution (37 degrees C, pH 7.37). Direct stimuli used supramaximal current density, 0.2-ms pulses, and 250-ms tetanic trains. Theophylline (500 mg/ml) increased force development at low stimulation frequencies but did not increase maximal force [25.7 +/- 0.5 for theophylline vs. 26.0 +/- 0.4 (SE) N/cm2 for control (n = 34)]. During repeated submaximal (25-36 Hz) tetanic contractions, theophylline did not affect endurance. During repeated maximal (160 Hz) tetanic contractions theophylline reduced endurance, accelerating the fall of developed force. Theophylline also inhibited recovery of force after endurance trials ended. We conclude that theophylline does not increase maximal tetanic force and can reduce diaphragm endurance in vitro.

Respiratory muscle activity during repeated airflow interruption.

We observed striking differences in respiratory muscle electromyogram activity when active expirations were interrupted in rapid succession, depending on the mode of interruption. When the interruptions were produced at the level of the glottis (utterances, uh-uh-uh-uh, at 5-8 Hz) there were synchronous bursts of activity from expiratory muscles in all three subjects during the periods of no flow and rapid bursts of diaphragmatic activity during the flow phases in one subject. In contrast, when similarly rapid interruptions of active expirations were produced with the tongue on a mouthpiece (utterance, te-te-te-te) or with an external valve, no synchronous bursts were observed. Since all interruptions would have been mechanically similar at expiratory muscular and pulmonary levels, we reasoned that the bursts with glottic interruptions were either programmed centrally or driven reflexly at the laryngeal level.

Relationship between parasternal intercostal length and rib cage displacement in dogs.

The relationship between parasternal intercostal length and rib cage cross-sectional area was examined in nine supine dogs during passive inflation and during quiet breathing before and after phrenicotomy. Parasternal intercostal length (PSL) was measured with a sonomicrometry technique, and rib cage cross-sectional area (Arc) was measured with a Respitrace coil placed around the middle rib cage. During active inspiration as well as during passive inflation, PSL decreased as Arc increased. However, the relationship between PSL and Arc during active inspiration, whether in the intact or phrenicotomized animal, was almost invariably different from that during passive inflation, so that the same increase in Arc was associated with a greater decrease in PSL in the former than in the latter instance. This difference between passive inflation and active inspiration is probably due to the active contraction of the parasternals during inspiration and the consequent caudal displacement of the sternum. In upright humans, the sternum moves cephalad and not caudad during inspiration, so the relationship between PSL and Arc during active breathing might be similar to that during passive inflation.

Respiratory changes in diaphragmatic intramuscular pressure.

We attempted to measure diaphragmatic tension by measuring changes in diaphragmatic intramuscular pressure (Pim) in the costal and crural parts of the diaphragm in 10 supine anesthetized dogs with Gaeltec 12 CT minitransducers. During phrenic nerve stimulation or direct stimulation of the costal and crural parts of the diaphragm in an animal with the chest and abdomen open, Pim invariably increased and a linear relationship between Pim and the force exerted on the central tendon was found (r greater than or equal to 0.93). During quiet inspiration Pim in general decreased in the costal part (-3.9 +/- 3.3 cmH2O), whereas it either increased or slightly decreased in the crural part (+3.3 +/- 9.4 cmH2O, P less than 0.05). Similar differences were obtained during loaded and occluded inspiration. After bilateral phrenicotomy Pim invariably decreased during inspiration in both parts (costal -4.3 +/- 6.4 cmH2O, crural -3.1 +/- 0.6 cmH2O). Contrary to the expected changes in tension in the muscle, but in conformity with the pressure applied to the muscle, Pim invariably increased during passive inflation from functional residual capacity to total lung capacity (costal +30 +/- 23 cmH2O, crural +18 +/- 18 cmH2O). Similarly, during passive deflation from functional residual capacity to residual volume, Pim invariably decreased (costal -12 +/- 19 cmH2O, crural -12 +/- 14 cmH2O). In two experiments similar observations were made with saline-filled catheters. We conclude that although Pim increases during contraction as in other muscles, Pim during respiratory maneuvers is primarily determined by the pleural and abdominal pressures applied to the muscle rather than by the tension developed by it.

Developmental changes in diaphragm contractile properties.

The contractile properties of pre- and early postnatal respiratory muscles are incompletely understood. We examined the effects of development on isometric contractile properties, with an emphasis on properties at 37 degrees C. One-day-old (n = 10), 3-wk-old (n = 10), and adult (n = 10) rabbits were studied. Isometric contractile properties of costal diaphragm strips were measured in vitro by using direct stimulation. Twitch and maximal, i.e., fused, tetanic force production increased with strip dimension and with age. Maximal tetanic force developed per unit cross-sectional area (stress) was significantly decreased in muscle from 1-day olds, whereas it was greatest in muscle from 3-wk olds. Twitch stress was similar in all three groups. Only when the stimulus duration was prolonged did twitch and fused tetanic force achieve maximal values values for the 1-day-old and 3-wk-old strips, suggesting less effective excitation-contraction coupling in those muscles. We conclude that immature rabbit diaphragm has unique isometric contractile properties and stimulus parameter requirements that cannot be deduced from studies using mature diaphragm.

Reactive oxygen in skeletal muscle. III. Contractility of unfatigued muscle.

This study tested the hypothesis that reactive oxygen intermediates present in unfatigued skeletal muscle act to enhance contractile function. Fiber bundles from rat diaphragm were incubated with exogenous catalase (an antioxidant enzyme that dehydrates hydrogen peroxide to molecular oxygen and water) to decrease the tissue concentration of reactive oxygen intermediates. Catalase (10(3) U/ml) significantly decreased twitch characteristics (time to peak tension, half-relaxation time, peak force, and twitch-to-tetanus force ratio), thereby shifting the force-frequency relationship to the right. Catalase effects were dose dependent. Concentrations of 1 to 10(5) U/ml progressively depressed submaximal (30-Hz) tetanic stress, whereas concentrations > 10(5) U/ml were toxic, inhibiting maximal (200-Hz) tetanic stress (P < 0.0001). Exogenous hydrogen peroxide (10(-4) to 10(-2)M) increased peak twitch stress (P < 0.03) and lengthened both time to peak tension (P < 0.02) and half-relaxation time (P < 0.02). Selective removal of superoxide anion radicals with the use of superoxide dismutase produced dose-dependent contractile inhibition similar to that produced by catalase. We conclude that the reactive oxygen intermediates present in unfatigued skeletal muscle have a positive effect on excitation-contraction coupling and are obligatory for optimal contractile function.

Passive mechanics of upright human chest wall during immersion from hips to neck.

We have determined the mechanical effects of immersion to the neck on the passive chest wall of seated upright humans. Repeated measurements were made at relaxed end expiration on four subjects. Changes in relaxed chest wall configuration were measured using magnetometers. Gastric and esophageal pressures were measured with balloon-tipped catheters in three subjects; from these, transdiaphragmatic pressure was calculated. Transabdominal pressure was estimated using a fluid-filled, open-tipped catheter referenced to the abdomen's exterior vertical surface. We found that immersion progressively reduced mean transabdominal pressure to near zero and that the relaxed abdominal wall was moved inward 3-4 cm. The viscera were displaced upward into the thorax, gastric pressure increased by 20 cmH2O, and transdiaphragmatic pressure decreased by 10-15 cmH2O. This lengthened the diaphragm, elevating the diaphragmatic dome 3-4 cm. Esophageal pressure became progressively more positive throughout immersion, increasing by 8 cmH2O. The relaxed rib cage was elevated and expanded by raising water from hips to lower sternum; this passively shortened the inspiratory intercostals and the accessory muscles of inspiration. Deeper immersion distorted the thorax markedly: the upper rib cage was forced inward while lower rib cage shape was not systematically altered and the rib cage remained elevated. Such distortion may have passively lengthened or shortened the inspiratory muscles of the rib cage, depending on their location. We conclude that the nonuniform forcing produced by immersion provides unique insights into the mechanical characteristics of the abdomen and rib cage, that immersion-induced length changes differ among the inspiratory muscles according to their locations and the depth of immersion, and that such length changes may have implications for patients with inspiratory muscle deficits.

Pulmonary afferent activity during high-frequency ventilation at constant mean lung volume.

We recorded the responses of 21 slowly adapting pulmonary stretch receptors (PSRs) and 8 rapidly adapting pulmonary stretch receptors (RARs) from the vagi of anesthetized open-chest dogs to high-frequency ventilation (HFV) at 15 Hz, at constant mean end-expiratory lung volume, and constant end-tidal PCO2. HFV applied in this way has been shown to prolong expiration. The responses of pulmonary afferents during HFV at constant mean volume have not been described. In the present experiments, receptor discharge during HFV was compared with that during the end-expiratory pause of normal-frequency ventilation. Average PSR discharge increased when HFV was applied, although not all PSRs exhibited increases. RARs were generally silent during normal and high-frequency ventilation at functional residual capacity and above. However, at low lung volumes, RAR discharge increased greatly when HFV was applied. We conclude that PSR discharge is increased during HFV in the absence of increased lung volume and that increases in PSR discharge during HFV are sufficient to explain the reflex that prolongs expiration in dogs.

Reactive oxygen in skeletal muscle. II. Extracellular release of free radicals.

We have tested the hypothesis that diaphragm muscle fibers release superoxide anion radicals (O2-.) into the extracellular space. Fiber bundles were isolated from rat diaphragm and incubated in Krebs-Ringer solution containing cytochrome c (10(-5) M), a standard assay for O2-.. Bundles were either passive or active, i.e., directly stimulated to contract rhythmically. After 1 h, absorbance of reduced cytochrome c in the incubation medium was measured at 550 nm. Absorbance was greater in medium exposed to passive muscle than in medium without muscle (P < 0.01), indicating O2-. release by passive muscle. Absorbance was greater in medium exposed to active muscle than in that exposed to passive muscle (P < 0.01), an increase inhibited by superoxide dismutase (10(3) U/ml). Active bundles fatigued; bundles developing the lowest final stresses produced the greatest absorbance increases (P < 0.001), suggesting that the magnitude of fatigue was inversely related to O2-. release. We conclude that O2-. is released by diaphragm myocytes into the interstitium and surrounding medium, a process accelerated by fatiguing muscular contractions.