l-Arginine Treatment Alters the Kinetics of Nitric Oxide and Superoxide Release and Reduces Ischemia/Reperfusion Injury in Skeletal Muscle
Background Constitutive nitric oxide synthase (cNOS) may produce species involved in ischemia/reperfusion (I/R) injury: NO in the presence of sufficient l-arginine and superoxide at the diminished local l-arginine concentration accompanying I/R.
Methods and Results During hindlimb I/R (2.5 hours/2 hours), in vivo NO was continuously monitored (porphyrinic sensor), and l-arginine (chromatography), superoxide (chemiluminescence), and I/R injury were measured intermittently. Normal rabbits were compared with those infused with l-arginine 4 mg·kg−1·min−1 for 1 hour. In both groups, ≈6 minutes into ischemia, a rapid increase of NO from its basal level of 50±17 to 115±7 nmol/L, P<.005 (microvessels), was observed. In animals not treated with l-arginine, NO dropped below basal to undetectable levels (<1 nmol/L) during reperfusion. In animals treated with l-arginine, the decrease of NO was slower, such that substantial amounts accumulated during reperfusion (25 nmol/L). Decreased NO during I/R was accompanied by increased superoxide, which during reperfusion reached 50 nmol/L without or 23 nmol/L with l-arginine treatment. Calcium-dependent cNOS was a major source of superoxide release (inhibited 70% by L-NMMA and 25% by L-NAME) during I/R.
Conclusions l-Arginine treatment decreased superoxide generation by cNOS while increasing NO accumulation, leading to protection from constriction (microvessel area, 17.77±0.95 versus 11.66±2.21 μm2 untreated, P<.0005) and reduction of edema after reperfusion (interfiber area, 16.56±2.13% versus 27.68±7.70% untreated, P<.005).
Reestablishing blood flow to ischemic tissue or organs (reperfusion) is a necessary step in many surgical procedures.1 However, reperfusion, especially after prolonged ischemia, leads to change of vasomotility and an increase of microvascular permeability causing tissue reperfusion edema.1 2 These sequelae are constant features of I/R injury.
Under normal conditions in mammalian tissue, cNOS receives and stores enough electrons from reduced NADPH to transform the substrates O2 and l-arginine into the products l-citrulline, NO, and water.3 Any l-citrulline produced is normally resynthesized into l-arginine by two cytosolic enzymes of the Krebs-Henseleit urea cycle.4 But NO production has a plethora of physiological effects mediated by its demonstrated interaction with many biomolecular sites.5 For example, NO generated in the vascular endothelium freely diffuses through the cell membrane into the surrounding smooth muscle target cells, where it initiates a cascade of events resulting in smooth muscle relaxation.6 Concurrently, NO freely diffuses through the cell membrane on the luminal side of the endothelium into the bloodstream to help prevent platelet aggregation and adhesion to the vessel walls.7
We hypothesized that in ischemia, there may be a depletion of local l-arginine concentration around cNOS, initially due to the high production of NO observed after the onset of ischemia.8 l-Arginine depletion may be maintained, under ischemic conditions, by the inability of ATP-dependent argininosuccinate synthetase to resynthesize l-arginine from l-citrulline and l-aspartate and by hindered mass transport of ingested l-arginine to the cells. The model used to design the experiments in this study suggests that these processes of local depletion of l-arginine may lead to disarrangement of the oxidase and reductase domains of the cNOS subunits (conformational changes relative to each other), just as better-characterized enzymes are known to disarrange in low substrate concentration environments.9 Of course, this useful model cannot be proved without tertiary crystallographic data for cNOS, which are unavailable at present. We can only point out that it has been demonstrated that cNOS, when activated in l-arginine–starved environments, can still receive electrons from NADPH and donate them to its other substrate, O2, resulting in a one-electron reduction to form O2−.10
Even though NO production can be high, especially shortly after the onset of ischemic conditions, the concomitant progressively increasing production of O2− may rapidly react with it to produce the stable product OONO−.8 When this becomes protonated (pKa=6.8), the HOONO formed usually undergoes isomerization (t1/2<1 second) to form hydrogen cation and nitrate anion.11 However, as the HOONO concentration increases as maximal O2− accumulations react with freshly synthesized NO during the initial stages of reperfusion, local HOONO concentration may become sufficient to ensure its efficient transport to reactive sites as far as several cell diameters away.11 In the vicinity of certain reactive centers, HOONO may undergo homolytic cleavage to a hydroxyl free radical (OH) and nitrogen dioxide free radical (NO2) or heterolytic cleavage to a nitronium cation (NO2+) and hydroxide anion (OH−).11 Three of these cleavage products (OH, NO2 radicals, and NO2+) are among the most reactive and damaging species in biological systems and may be major contributors to the severe I/R damage signaled by profound microvessel constriction and persistent interstitial edema.11
The near diffusion limited reaction of O2− with NO to form OONO− (k=3.8×109 L·mol−1·s−1) is even faster than the reaction of O2− with SOD to form peroxide and oxygen (k=2×109 L·mol−1·s−1).12 13 This suggests that the 5 to 10 mmol/L SOD in tissue cannot effectively prevent the nonenzymatic reaction of NO with O2− as long as basal concentrations of NO are maintained.12 Since OONO− is a weak vasodilator, NO is a potent vasodilator, and O2− is a vasoconstrictor, it is not surprising that vasodilatation is diminished during the course of I/R.6 14 15
Therefore, at least two major strategies are viable to prevent I/R injury. The first is based on preventing the disarrangement of the cNOS during its extensive work over the course of I/R by treatment with l-arginine or l-arginine analogues, thus preventing the production of O2−. The second is based on scavenging O2− already produced under I/R conditions by disarranged cNOS and other potential sources of O2− by use of SOD and/or other free radical scavengers.16
This article describes the use of the first strategy, which was done by observing the consequences of I/R in the rabbit hindlimb at normal physiological and elevated concentrations of l-arginine (administered before ischemia or before reperfusion). In these experiments, continuous in vivo measurement of NO production was correlated with intermittent measurements of O2− production, physical changes in interfiber space (interstitial edema formation), and microvasculature changes (microvessel diameter alteration) as indicators of I/R injury in the skeletal muscles.
The studies were performed on adult male New Zealand White rabbits (Charles River GmbH, Sulzfeld, Germany) in accordance with institutional guidelines. The rabbits had free access to food (Altromin 2120 standard diet pellets; Marek) and water. The animals were randomly divided into three experimental groups: (1) The AI/R group received l-arginine treatment and a hindlimb operation with I/R; (2) the XI/R group did not receive l-arginine treatment but did receive a hindlimb operation with I/R; and (3) the sham group did not receive l-arginine treatment but did receive a sham operation without I/R.
The animals from all three groups were fasted for 24 hours before surgery. Intravenous anesthesia was attained with a mixture of ketamine hydrochloride (15 mg/kg Ketalar, Parke-Davis GmbH) and xylazine hydrochloride (1.5 mg/kg Rompun, Bayer). The animals were intubated via tracheotomy, and lung ventilation was maintained with N2O/O2 (Fio2=0.35) and isoflurane (Forane, Abbott) (1 to 2 vol%) at a tidal volume of 15 to 20 mL/kg and at a rate of 30 to 35 cycles/min. Arterial blood samples were taken from the left common carotid arterial line. Ventilator parameters were adjusted according to blood gas values (Po2, 148.52±16.41; Pco2, 41.13±3.36; pH, 7.32±0.11) (AVL 995-Hb, AVL LIST GmbH) and oxygen saturation (OSM 2, Radiometer). Blood arterial pressure was monitored with a physiological pressure transducer (Berg GmbH). Ringer solution (Leopold Pharma GmbH) (0.2 mL·kg−1·min−1) was infused via the auricular vein. Body temperature was kept constant.
A catheter was placed via the internal iliac vein into the distal interior caval vein of animals in all three groups. Venous blood samples were obtained from all three groups at the following time points: tc=0 hours, before the onset of ischemia; ti=2.5 hours, at the end of ischemia; and tr=2 hours, 2 hours after reperfusion. An adequate quantity of isotonic saline was injected after each sampling to compensate for blood loss. For both the XI/R and the AI/R groups of animals, bilateral hindlimb ischemia was achieved according to established techniques.17 Common femoral arteries were clamped in the groins, and collateral flow was occluded by a rubber arterial tourniquet (Stripp-Quick, KaWe) wrapped around each thigh at the proximal third of the leg. Blood flow occlusion was confirmed with a blood perfusion monitor, Laserflo BPM2 (Vasamedics). Both limbs were rendered ischemic for 2.5 hours. Reperfusion was achieved by releasing the clamps and tourniquets and was confirmed by restoration of pulsatile blood flow in the femoral arteries as well as by the blood perfusion monitor.
Exclusively in the AI/R group, animals were treated by continuous intravenous infusion of l-arginine (Leopold GmbH) through an auricular vein catheter. l-Arginine (4 mg·kg−1·min−1) was infused for 1 hour before ischemia or 30 minutes before the end of ischemia and during the first 30 minutes of reperfusion (AI/R group only). Exclusively in the sham group, the sham operation consisted of groin incision and femoral blood vessel dissection.
Preparation of NO Sensors for In Vitro and In Vivo Measurements
NO was measured with a porphyrinic microsensor that was free of interference (at physiological conditions) from all reagents used in these experiments and all known readily oxidizable secretory products that may be found in mammalian blood to at least two orders of magnitude greater than their expected concentrations. The porphyrinic sensors for in vitro and in vivo measurements were prepared by methodologies described in detail previously.18 19 20 Briefly, the needle from a intravenous catheter unit (24 gauge, 25 mm long, Angiocath, Becton-Dickinson) was roughened along the shaft, then truncated and polished flat so that it was shorter than its 24-gauge, 25-mm-long catheter. A single carbon fiber (6 μm in diameter, protruding 3 mm, Amoco) was mounted inside the hollow truncated 24-gauge needle with conducting epoxy. After curing, the exterior of the truncated needle was coated with nonconductive epoxy (2-TON, Devcon) and allowed to cure again.
The protruding 3-mm carbon fiber tip was made more sensitive to NO and less sensitive to potential interference by the cyclic voltammetric deposition (−0.20→1.00 V at 100 mV/s for 10 cycles) of a highly conductive polymeric porphyrin from a solution of 0.25 mmol/L nickel (II) tetrakis(3-methoxy- 4-hydroxyphenyl) porphyrin in 0.1 mol/L NaOH under nitrogen. Dip-coating of the dried catalyzed carbon fiber tip (3 times for 5 seconds) in 1% Nafion in alcohol (Aldrich), after drying, produced a thin anionic film that repelled or retarded charged species while allowing small neutral and hydrophobic NO access to the underlying catalytic surface.
A single carbon fiber sensor is flexible and can be bent and placed directly on the surface of a microvessel without catheter protection for in vivo or in vitro measurements. A lower limb of a rabbit was positioned under a dissecting microscope, and surgical microdissection was performed until a microvessel (200±50 μm) was visualized. No differentiation was made between arterial and venous vessels. A porphyrinic sensor was lowered with the help of a stereotactic micromanipulator until the surface of the observed blood vessel was reached. This was indicated by a small (picoampere) and short (ms) piezoelectric signal. Two auxiliary electrodes (platinum and SSCE) were positioned on the surface of adjacent tissue.
To implant the porphyrinic NO sensor in the wall of a large-diameter vein or artery, smooth muscle tissue was pierced with a standard 24-gauge angiocatheter needle (clad with its catheter with 4×50-μm ventilation holes near the tip). The catheter/needle unit was advanced to a desired place in the vessel wall. The position of the catheter was secured, and the placement needle was removed and replaced by a truncated needle with a mounted porphyrinic NO sensor. A platinum wire counterelectrode and SSCE were placed in contact with adjacent tissue.
Two techniques for measuring NO, DPV and CA, were performed with a PAR model 273 voltammetric analyzer interfaced with an IBM 80486 computer with data acquisition and control software. DPV was used to measure the basal NO concentration. Briefly, in the DPV method, current versus potential curves were generated in the potential range between 0.45 and 0.75 V versus SSCE. The DPV peak current at the peak potential characteristic for NO oxidation (0.65 V) was found to be directly proportional to the local NO concentration in the immediate vicinity of the sensor. CA, fixed at the peak potential for the oxidation of NO versus SSCE, was used for fast (resolution time, 0.1 to 1 ms) and continuous measurement of the changes of NO concentration from its basal level with time.
The volume sampled is approximately equal to the volume of the sensor (10−10 to 10−12 L). Therefore, the concentration of NO measured was a local or surface concentration (not a bulk or global concentration). The porphyrinic microsensor had a response time of 0.1 ms at micromolar NO concentrations and 10 ms at the detection limit of 1 nmol/L. Linear calibration curves were constructed for each sensor from 2×10−9 to 2×10−5 mol/L NO before and after in vivo or in vitro measurements with aliquots of saturated NO prepared as described.21
O2− was measured in the adductor magnus muscle tissue samples freshly excised from the right hindlimb before and during I/R. The concentration of O2− was determined by a chemiluminescence method.22 O2− produced chemiluminescence of lucigenin (bis-N-methylacridinium nitrate), which was detected with a scintillation counter (Beckman 6000LS, with a single-photon monitor). Each tissue sample (0.8 to 1.5 mg) was placed in 2 mL of HBSS adjusted to pH 7.4 at 25°C, then enough lucigenin was added to make its concentration 0.25 mmol/L. Basal O2− concentration produced by the tissue was measured after a 2-minute incubation in HBSS. The sum of the O2− produced by disarranged cNOS and other sources (basal) was measured in a similar manner, except that the 2-minute incubation period was followed by injection of 20 mL of 1 mmol/L A23187 calcium ionophore (a receptor-independent cNOS agonist). Photon counts were calibrated as O2− concentration by construction of standard curves based on photons emitted by O2− stoichiometrically generated by treatment of xanthine with xanthine oxidase. The measured concentration of O2− (unlike the sensor-monitored NO) represents an average concentration in the volume of the sample and is reported per 1 mg wet tissue.
Measurement of l-Arginine and l-Citrulline
For measurement of l-arginine and l-citrulline concentration in blood plasma, heparinized venous blood samples were centrifuged at 3000g for 7 minutes. The plasma was separated and deproteinized with sulfosalicylic acid (30%) containing 1 mmol/L β-(2-thienyl)(±)alanine as an internal standard. The samples were stored at 4°C for 30 minutes and centrifuged at 12 000g for 5 minutes, and the supernatant was analyzed for l-arginine and l-citrulline by HPLC.
For measurement of the total concentrations of l-arginine and l-citrulline in the skeletal muscle, tissue samples of the adductor magnus muscle were removed from the right hindlimb at time points tc=0 hours, ti=2.5 hours, and tr=2 hours. Muscle biopsies were immediately wet-weighed and homogenized in a glass container filled with 500 μL of ice-cold sulfosalicylic acid (4%) containing 0.1 mmol/L of β-(2-thienyl)(±)alanine as an internal standard. The homogenate was cooled on ice for 30 minutes, then centrifuged at 12 000g for 5 minutes. The supernatant was analyzed by HPLC, and total concentrations of l-arginine and l-citrulline in the skeletal muscle were expressed as μmol/L of analyzed liquid.
The l-arginine and l-citrulline contents in the skeletal muscle were calculated by total l-arginine and l-citrulline concentrations in the skeletal muscle and fat-free dry mass of skeletal muscle sample (after extraction by petrol ether). The l-arginine and l-citrulline contents were expressed as μmol/100 g fat-free dry mass.
Intracellular concentration of l-arginine and l-citrulline were calculated by subtracting the extracellular amino acid concentration from the total amino acid concentration in the muscle tissue. Extracellular amino acid concentration in muscle tissue for this calculation was assumed to be equal to the concentration in plasma. This assumption is invalid during the ischemic period; therefore, the intracellular concentrations of amino acids were measured before ischemia (tc=0 hours) and 2 hours after reperfusion (tr=2 hours). The quantification of intracellular volume of water needed to calculate concentration was assessed by the chloride method.23
After dilution of plasma or tissue extract in membrane-filtered water (Waters Millipore) and 2-minute derivatizations with o-phthaldialdehyde, a 10-μL sample was injected onto the column by an autosampler (Spark Triathlon). Separation of the derivatized amino acids was obtained by an HPLC (Beckman) equipped with a 3-mm particle size and 125×4.6-mm ODS Hypersil column (Bischoff) with a two-buffer-system gradient elution and a fluorescence detector (Jasco).
At time points tc=0 hours, ti=2.5 hours, and tr=2 hours, adductor magnus muscle tissue samples were removed from the right hindlimb and immediately immersed for quick freezing at −70°C in a 2-methylbutane solution (Uvasol; Merck) for 2 minutes. Then the samples were stored at −80°C until they were sectioned. Transverse cryosections 10 μm thick (Kryostat 1720, Leitz) were stained for actomyosin ATPase activity at pH 4.3 and pH 10.4 for comparison. Sections stained at pH 4.3 were examined by light microscopy (Axiomat, Zeiss) by an unbiased observer.
The measurements and counts of the fibers and microvessels were performed with a pen linked to a personal computer using a semiautomatic image-analyzing system (LUCIA M, Nikon Laboratory Imaging) at ×100 and ×1000 magnification, respectively. In three random fields, the muscle fibers were counted in each section and the %MIFA was determined. Fifty microvessels per random field were measured in each section, and the microvessel area was determined.
To ensure that alterations of %MIFA measured by morphometry could accurately reflect the process of interstitial edema formation, the water content in the skeletal muscle was determined in the animals of the XI/R group. In the muscle biopsies removed at time points tc=0 hours, ti=2.5 hours, and tr=2 hours, the wet mass, dry mass, and fat-free dry mass were measured and the total water content was calculated. Calculation of the intracellular and extracellular water content in the muscle tissue was carried out by the chloride method.23 After completion of the experimental investigation, the animals were killed by intravenous administration of an overdose of potassium chloride.
The mean±SD are given. Comparisons between groups and between different time points were performed by blocked and unblocked ANOVA and paired and unpaired t tests. Correlations between parameters were determined by Pearson’s correlation coefficient. All analyses were made with the statistical software of SAS (SAS 1990, SAS/STAT User’s Guide, Version 6; SAS Institute).
Fig 1a⇓ shows a typical amperometric curve (current calibrated as NO concentration versus time) measured in vivo during I/R with a porphyrinic sensor placed on the wall of a microvessel (200±50 μm) in the lower limb of a rabbit. A short time (360±50 seconds) after the femoral artery was clamped, a rapid increase of NO concentration from its basal concentration of 50±17 nmol/L, P<.005, was observed. The average rate of NO concentration increase was 0.19±0.03 nmol·L−1·s−1. The concentration of NO reached a peak of 115±7 nmol/L after 12 minutes of ischemia, which persisted for 60 seconds and then decayed at a rate of −0.07±0.02 nmol·L−1·s−1. After ≈50 minutes, NO concentration decreased to its basal level and after another 30 minutes dropped significantly below its normal preischemic basal concentration. During reperfusion, a rapid decrease of NO concentration to the detection limit of the porphyrinic sensor (1 nmol/L) was noticed within the first 5 minutes. Then, after 30 minutes of reperfusion, a very slow increase of NO concentration to ≈2 nmol/L was observed.
Infusion of an NOS inhibitor, L-NMMA (20 mg/kg), before femoral artery occlusion caused significant changes of the kinetics of NO release for the duration of the I/R (Fig 1b⇑). NO concentration decreased 2 minutes after administration of L-NMMA from its original level of 50±17 nmol/L to 20±5 nmol/L (data not shown). The occlusion of the femoral artery initially produced a small increase of NO concentration to a plateau of 26±5 nmol/L, then slowly decreased with time. At the end of ischemia, the NO concentration was 12±4 nmol/L. During the first 15 minutes of reperfusion, the NO level decreased slightly to a minimum (8±2 nmol/L). However, this level was 4 to 8 times higher than the minimum concentration of NO observed during reperfusion of a rabbit without L-NMMA treatment.
Preischemic treatment of rabbits with l-arginine (4 mg·kg−1·min−1 for 1 hour) did not change the pattern (kinetics) of NO release during the initial period of ischemia (Fig 2a⇓). The maximum concentration of 123±8 nmol/L was reached 350±20 seconds after ischemia and decayed gradually during ischemia but at a rate (−0.020±0.005 nmol·L−1·s−1) about one third as fast as the decay rate during a comparable time span without l-arginine treatment. However, the kinetics of NO release during reperfusion were different: NO concentration decreased only slightly and stayed above basal levels with preischemic l-arginine treatment, compared with NO concentration extinction without any l-arginine treatment.
Treatment with l-arginine initiated 30 minutes before reperfusion and continued for 30 minutes during reperfusion affected the kinetics of NO release and the level of its accumulation during reperfusion (Fig 2b⇑). The initial kinetics of NO release during ischemia was the same as observed in Fig 1a⇑ without l-arginine treatment. During reperfusion, the NO concentration dropped slightly, to 25±6 nmol/L. This slight drop followed by slow decay of NO concentration during reperfusion is in distinct contrast to NO reperfusion kinetics in normal l-arginine concentration environments (Fig 1a⇑).
We have observed (data not shown) that the peak concentration of NO increased with increasing diameter of veins or arteries. However, the general pattern of NO release shown in Figs 1⇑ and 2⇑ and relative changes of NO concentration during I/R do not depend on the size of veins or arteries. The peak concentration of NO measured in the wall of the femoral artery (diameter, 12 mm) during ischemia was 360 nmol/L, and this concentration changed exactly according to the patterns shown in Figs 1⇑ and 2⇑, decreasing to an undetectable level during reperfusion.
The mode of l-arginine treatment with resultant NO release pattern depicted in Fig 2b⇑ may be of clinical relevance during some surgical procedures. However, the mode of l-arginine treatment and the pattern of NO release (Fig 2b⇑) at the end of ischemia and beginning of reperfusion will be of significance for most of the clinical cases of ischemia; hence, the primary focus of data presented in this work involved studying this mode of l-arginine treatment.
Changes of NO concentration released from freshly excised skeletal muscle after stimulation with the receptor-independent calcium ionophore A23187 (10 μmol/L) were measured in vitro with a single-fiber porphyrinic sensor placed on the wall of a microvessel. The CA curves showing the change of NO concentration with time recorded in the absence and presence of SOD (100 U/mL) are depicted in Fig 3⇓. Since SOD is a rapid scavenger of superoxide, we used this indirect approach to estimate production of O2− at the time of NO release.
The peak release of NO in preischemic tissue (XI/R group) was 123±5 nmol/L and increased slightly to 135±4 nmol/L in the presence of SOD (Fig 3a⇑). In the tissue collected at 30 minutes of ischemia (ti=0.5 hour), the NO concentration was 71±6 nmol/L (42% lower than before ischemia) and increased in the presence of SOD (Fig 2c⇑) but did not reach the level before ischemia. A further decrease of NO concentration to 42±6 nmol/L with concomitant increase in the presence of SOD was observed in tissue collected at the end of ischemia (ti=2.5 hours). During reperfusion (tr=2 hours), a further reduction of NO production (12 nmol/L) was observed. The production of NO recovered to the level of 120±3 nmol/L in the presence of SOD. These data suggest that calcium ionophore not only stimulates NO release but also activates the release of O2−. It may be that the rising calcium-dependent O2− production during ischemia was responsible for the decreasing net production of NO concentration observed in Figs 1⇑ and 2⇑.
This possibility was substantiated by the use of a chemiluminescence method to make direct in vitro measurement of O2− concentration in freshly excised tissue from each experimental group. The average basal concentration of O2− in the tissue before ischemia was 10±3 nmol/L (n=7). This concentration increased to 11±3 and 13±3 nmol/L at ti=0.5 and 2.5 hours, respectively, in the XI/R group (Fig 3⇑). A further increase of basal O2− concentration to 15±2 nmol/L was observed in the tissue during reperfusion. After addition of calcium ionophore A23187, a receptor-independent agonist that turns on cNOS, an increase of O2− production was observed. In tissue freshly excised from the XI/R group, O2− concentration was 16±3 nmol/L before ischemia and increased substantially to 27±4 nmol/L at 0.5 hour of ischemia. At the end of ischemia (ti=2.5 hours) and during reperfusion (tr=2 hours), a further increase of O2− concentration to 36±4 and 50±5 nmol/L, respectively, was observed.
Also, increased production of O2− (25±3 and 31±4 nmol/L after 2.5 hours of ischemia and 2 hours of reperfusion, respectively) was observed in the presence of acetylcholine (1 μmol/L), a receptor-dependent calcium agonist known to activate cNOS (data not shown). As a negative control, 10-μmol/L doses of known inhibitors of cNOS after a 2-minute incubation were found to reduce calcium-dependent O2− release from freshly excised tissue from group XI/R at time ti=2.5 hours after A23187 stimulation as above. L-NMMA (70% inhibition) was more potent than L-NAME (25% inhibition) at inhibiting the production of O2− from ischemic adductor muscle tissue.
In animals treated with l-arginine (AI/R group), the concentration of O2− was similar to that of untreated animals (XI/R group) before ischemia and also at the end of ischemia (ti=2.5 hours) (Fig 3c⇑). However, a significant difference in O2− concentration between these two groups of animals was observed during reperfusion (tr=2 hours). For l-arginine–treated animals (AI/R group), the O2− concentration was 23±3 nmol/L, a 52% decrease from the level of O2− observed for untreated animals (XI/R group).
l-Arginine and l-Citrulline Concentrations
The concentration of l-arginine and l-citrulline in the blood of animals from the XI/R group did not change significantly during the I/R period. In the AI/R group, after infusion of l-arginine (ti=2.5 hours), the l-arginine concentration in blood increased 17 times from basal, 108±11 to 1954±700 μmol/L (P<.005) (Table⇓). The level of l-arginine decreased continuously thereafter but at time point tr=2 hours still remained significantly higher than the physiological concentration before ischemia. l-Citrulline concentration in blood increased after l-arginine treatment in the AI/R group. However, this increase (≈15%) was relatively small compared with the increase of l-arginine concentration in the blood.
Data obtained from the analysis of l-arginine and l-citrulline in tissue as well as their intracellular concentrations were more accurate than blood concentrations in reflecting changes of these two compounds during I/R. It is interesting to note that both muscle tissue content and intracellular concentration of l-arginine increase naturally during reperfusion in animals not treated with l-arginine (XI/R group). In the AI/R group, both l-arginine and l-citrulline concentrations increased during reperfusion. The increase of l-citrulline concentration was less pronounced but significant in l-arginine–treated animals (the AI/R group).
Percent Muscle Interfiber Area
In the sham group, the %MIFA value was 13.70±0.81% at ti=2.5 hours. In the XI/R group, the %MIFA at ti=2.5 hours did not differ significantly from the corresponding value recorded for the animals of the sham group (Fig 4⇓). However, significant interstitial edema was observed in the muscles of the XI/R group after reperfusion (Fig 5B⇓); the %MIFA increased from 16.75±2.40% at ti=2.5 hours to 27.68±7.70% at tr=2.0 hours, at the end of reperfusion. At tr=2.0 hours, %MIFA was 91% higher than that of the sham group, P<.005 (Fig 4⇓). However, in animals treated with l-arginine, only a slight edema was observed during the reperfusion period (Fig 5C⇓). The %MIFA in the AI/R group increased from 13.43±0.84% at ti=2.5 hours to 16.56±2.13% at tr=2.0 hours. At ti=2.5 hours, the %MIFA in the AI/R group was 15% higher than that of the sham group, P<.05 (Fig 4⇓).
The %MIFA alteration was accompanied by a change of water content in the skeletal muscles of the XI/R group animals. An increase of total water content from 349±27 to 386±34 mL/100 g dry mass was observed between ti=2.5 hours and tr=2.0 hours. This increase was primarily a result of an increase of the extracellular water content from 85±34 to 117.5±40 mL/mg at times ti=2.5 hours and tr=2.0 hours, respectively, whereas the intracellular water content remained constant during the I/R period. It should be pointed out that in animals of the XI/R and AI/R groups, morphological alterations were most pronounced in the central part of the skeletal muscle, leaving peripheral subfacial portions of muscle less damaged.
The sham group MVCSA remained the same during I/R experiments (Fig 6⇓). For the XI/R and AI/R groups, no significant changes in the size of microvessels during the ischemia period were observed. However, 2 hours after reperfusion, severe microvessel constriction was found in the muscles of the XI/R group of animals (Fig 7B⇓). The MVCSA decreased from 18.33±2.72 μm2 at the end of ischemia to 11.66±2.21 μm2 at 2.0 hours of reperfusion. At 2.0 hours of reperfusion, the MVCSA was reduced by 34% in the XI/R group compared with the sham group, P<.005. In the AI/R group, no significant change in microvessel size was observed either at the end of ischemia or after 2.0 hours of reperfusion compared with the sham group (Fig 7C⇓). Therefore, at 2.0 hours of reperfusion, the MVCSA was significantly higher in l-arginine–treated than in untreated animals, P<.0005 (Fig 6⇓).
Normally, l-arginine concentration is biosynthesized in adult mammals by a two-enzyme transformation of l-aspartate and l-citrulline into l-arginine and fumarate in the Krebs-Henseleit urea cycle, mainly within the kidney but also in endothelial and muscle cells.4 The rationale underlying our studies is that local depletion of l-arginine concentration during I/R results in the disarrangement of the domains of cNOS and reduction of NO concentration and concomitant increase of O2− production. Therefore, l-arginine treatment before or during I/R should prevent disarrangement of cNOS, decrease O2− production, and minimize I/R by enhanced NO production.
As evident from data obtained by direct and continuous in vivo measurement of NO, a few minutes after the onset of ischemia, a rapid increase of NO concentration in the vasculature of skeletal muscles was observed. This pattern of NO release is similar to that observed in the brain after middle cerebral artery occlusion.8 Endothelial cNOS is anchored to the plasmalemmal caveolae of the vascular endothelium.24 Neuronal cNOS is cytosolic and is found in skeletal muscles, the kidney, and the central and peripheral nervous systems.25 It appears that some phenomenon connected with the first few minutes of ischemia dramatically elevates transient intracellular calcium concentration gradients and turns on both of these cNOS isoforms to start transforming l-arginine and O2 into l-citrulline, NO, and water. During ischemia, and especially in the first few minutes of reperfusion, the release of NO depends not only on the availability of the substrates O2 and l-arginine but also on the concentration of O2− accumulated.
Confirmation that cNOS rather than other sources produced most of the O2− in ischemic tissue came from experiments showing O2− release after treatment with known NO agonists as well as its inhibition after incubation with known cNOS inhibitors. The rapid accumulation of O2− concentration in the presence of calcium ionophore or acetylcholine suggests that production of O2− is calcium dependent like the production of NO by cNOS. Also as in the production of NO, the production of O2− can be inhibited by certain l-arginine derivatives. But it was quite surprising to observe that L-NMMA was more potent than L-NAME at inhibiting the production of O2− from freshly excised XI/R group tissue (after A23187 treatment). A previous report on isolated l-arginine–starved neuronal cNOS found L-NAME effective but L-NMMA ineffective at inhibiting O2− production.26
Of course, there are many other sources of O2− besides disarranged calcium-dependent cNOS. Normally these (calcium-independent) other sources contribute to the basal concentration of O2−, but during I/R, especially during the first few minutes of reperfusion, we observed that the sum of the O2− produced by all these other sources accounts for only 30% of the O2− produced during reperfusion. The O2− produced by some other sources, such as xanthine oxidase, under normal conditions is efficiently scavenged by SOD or basal NO. In addition, it is interesting to note that enhanced NO concentration may actually inhibit other enzymatic sources of O2−, such as NADPH oxidase.27 Conversely, the depressed NO production during the latter stages of I/R may allow increased O2− release from these same (non-cNOS) sources. This effect was observed in this study (Fig 3b⇑), in which the basal O2− (measured in the absence of calcium agonist) increased from 8±3 nmol/L before ischemia to 13±2 and 17±3 nmol/L during ischemia and reperfusion, respectively.
Under normal physiological conditions, even after 24 hours of fasting, the concentration of l-arginine in the endothelial cells is high enough (≈1 to 3 mmol/L) to provide a sufficiently high concentration to facilitate fast transport of this molecule to the active site of the cNOS enzymes.4 In cells, mass transport of l-arginine to the active site of the cNOS is limited by the diffusion process, which depends only on concentration gradient (the difference between l-arginine concentration at the active center of cNOS and its bulk concentration in the cytoplasm). Higher concentration gradients encourage faster mass transport. However, during the extensive NO production during I/R, the mass transport of l-arginine (at its physiological concentration) does not appear to be fast enough to provide its continuous supply to the cNOS active site.
On the basis of our current data, we cannot exclude the possibility that when local l-arginine concentration decreases below some critical level, cNOS (without any conformational changes) simply starts to produce O2− from its other substrate O2. Neither this alternative static enzyme model nor the substrate-directed cNOS conformation model can be proved at this time without tertiary crystallographic structures of the enzyme with and without the substrate l-arginine. Therefore, at this time it is safe to assert, on the basis of direct measurements of O2− in tissue by the chemiluminescence method, that O2− is produced in relatively high concentration during I/R and that this production is inversely related to NO accumulation. The onset of increasing production of O2− (above its basal level) was observed ≈20±6 minutes after the onset of ischemia. This increasing concentration of O2− parallels the decreasing concentration of NO measured by the NO sensor (Fig 1a⇑). A dramatic decrease of NO concentration was observed during reperfusion, when production of O2− was the highest. After several minutes of reperfusion, the net concentration of NO decreased significantly below its preischemic basal level, to the detection limit of the NO sensor (1 nmol/L). Paralleling this low NO concentration during reperfusion, a significant decrease of the diameter of microvessels was observed. Blood flow was hindered, and reperfusion damage to the tissue occurred.
Some l-arginine analogues, such as L-NMMA and L-NAME, inhibit the production of O2−, but by coincidentally inhibiting NO generation, these analogues prevent the local depletion of l-arginine concentration and subsequent disarrangement of cNOS. However, these analogues, by inhibiting NO production, enhance vasoconstriction in nonischemic tissue. Thus, the possibility of a clinical application for l-arginine analogues for the prevention of I/R injury seems rather remote.
In excised skeletal muscle samples, we found an increase of l-citrulline concentration in the AI/R group and elevated concentration of l-arginine in both the XI/R and AI/R groups compared with the sham group. Because of a significant dose of l-arginine administered during I/R, it was expected and confirmed that AI/R group muscle tissue l-arginine content, as well as intracellular l-arginine concentration, would be much higher than in the sham group. It was interesting to note that although it was not treated with extra l-arginine, the XI/R group also showed a slight increase in l-arginine concentration in muscle tissue compared with the sham group. This effect was most likely generated in the early stages of ischemia from cytosolic ATP-dependent argininosuccinate synthetase, the rate-determining enzyme for resynthesis of l-arginine and fumarate from l-citrulline and l-aspartate.4 However, this recycled l-arginine seems to be insufficient for sustained NO synthesis during I/R.
Treatment with l-arginine prevented microvessel constriction in the reperfused muscle despite reduced but still apparent interstitial edema. This suggests that the changes in concentration of substances that can actively affect vasomotility (not merely passive changes of external pressure) should be considered most important in interpreting data on changes in microvessel diameter.
Another beneficial effect of l-arginine treatment is a significant reduction of muscular reperfusion edema. Interstitial reperfusion edema was documented by the increased %MIFA in both groups of animals without or with l-arginine treatment. However, the edema was markedly reduced by l-arginine treatment. There is considerable evidence of NO influence on microvascular permeability mediated by endothelium- and leukocyte-dependent mechanisms. Endothelial regulation of vascular permeability by NO is not fully understood, but the inhibition of NO production was shown to increase the microvascular protein efflux and to potentiate permeability.28 Also, it has been suggested that NO decreases permeability by relaxing endothelial cells, thereby narrowing the width of endothelial junctions.29 The protective effect of NO donors on microvascular permeability may also be related to a leukocyte-dependent mechanism that involves the reduction of leukocyte–endothelial cell interaction.30
In conclusion, the present study demonstrates a protective effect of l-arginine treatment of skeletal muscle against I/R injury. NO deficiency due to its consumption by O2− produced in high concentration during ischemia and reperfusion, which was documented by in vivo measurement, plays an important role in the pathophysiology of I/R injury. l-Arginine treatment during I/R prevents excessive release of O2− and preserves NO concentration at the level needed for vasorelaxation. Therefore, l-arginine treatment during I/R can be a powerful tool to prevent microvessel constriction in reperfused skeletal muscle and also to significantly reduce muscular reperfusion edema.
Selected Abbreviations and Acronyms
|cNOS||=||constitutive NO synthase|
|DPV||=||differential pulse voltammetry|
|HPLC||=||high-performance liquid chromatography|
|L-NAME||=||Nω-nitro-l-arginine methyl ester|
|%MIFA||=||percentage of muscle interfiber area|
|MVCSA||=||microvessel cross-sectional area|
|SSCE||=||silver/silver chloride electrode|
This study was supported in part by MUCOS Pharma GmbH & Co, Geretsried, Germany; Gebro Broschek GmbH, Fieberbrunn, Austria; and Research Excellence Funds, Institute of Biotechnology, Oakland University, Rochester, Mich.
- Received October 16, 1996.
- Revision received December 23, 1996.
- Accepted January 15, 1997.
- Copyright © 1997 by American Heart Association
Korthuis RJ, Granger DN, Townsley MI, Taylor AE. The role of oxygen-derived free radicals in ischemia induced increases in canine skeletal muscle vascular permeability. Circ Res. 1985;57:599-609.
Leone AM, Palmer RM, Knowles RG, Francis PL, Ashton DS, Moncada S. Constitutive and inducible nitric oxide synthases incorporate molecular oxygen into both nitric oxide and citrulline. J Biol Chem. 1991;266:23790-23795.
Reyes AA, Karl IE, Klahr S. Role of arginine in health and disease. Am J Physiol. 1994;267:F331-F346.
Bennet WS, Steitz TA. Glucose-induced conformational change in yeast hexokinase. Proc Natl Acad Sci U S A. 1978;75:4848-4852.
Klatt P, Schmidt K, Uray G, Mayer B. Multiple catalytic functions of brain nitric oxide synthase. J Biol Chem. 1994;268:14781-14787.
Beckman JS, Chen J, Ischiropoulos H, Crow JP. Oxidative chemistry of peroxynitrite. In: Packer L, ed. Methods of Enzymology. San Diego, Calif: Academic Press Inc; 1994;233:229-240.
Kobayashi K, Miki M. A direct demonstration of reaction of nitric oxide with superoxide anion by the use of pulse radiolysis. In: Asada K, Yoshikawa T, eds. Frontiers of Reactive Oxygen Species in Biology and Medicine. Amsterdam, Netherlands: Elsevier Science BV; 1994:223-224.
Fridovich I. The biology of oxygen radicals. Science. 1978;201:875-880.
Liu S, Beckman JS, Ku DD. Peroxynitrite, a product of superoxide and nitric oxide, produces coronary vasorelaxation in dogs. J Pharmacol Exp Ther. 1994;268:1114-1121.
Consentino F, Sill JC, Katusic ZS. Role of superoxide anions in the mediation of endothelium-dependent contractions. Hypertension. 1994;23:229-235.
Huk I, Nanobashvili J, Punz A, Lassner H, Mueller M, Siegl H, Neumayer CH, Mittlboeck M, Roth E. Influence of a short-term acute ischemia and reperfusion on skeletal muscle metabolism and morphology in rabbits. Int J Angiol. 1996;5:59-63.
Malinski T, Czuchajowski L. Nitric oxide measurements by electrochemical methods. In: Feelisch M, Stamler JS, eds. Methods of Nitric Oxide Research. Chichester, UK: John Wiley & Sons Ltd; 1996:319-339.
Jia L, Furchgott RF. Inhibition by sulfhydryl compounds of vascular relaxation induced by nitric oxide and endothelium-derived relaxing factor. J Pharmacol Exp Ther. 1993;267:371-378.
Roth E. Veränderungen im Aminosäuren- und Proteinstoffwechsel bei chirurgischen Patienten. Munich, Germany: Zuckschwerdt Verlag; 1986:24-25.
Shaul PW, Smart EJ, Robinson LJ, German Z, Yuhanna IS, Ying Y, Anderson RGW, Michael T. Acylation targets endothelial nitric-oxide synthase to plasmalemmal caveolae. J Biol Chem. 1996;271:6518-6522.
Pou S, Pou WS, Bredt DS, Snyder SH, Rosen GM. Generation of superoxide by purified brain nitric oxide synthase. J Biol Chem. 1992;267:24173-24176.
Clancy RM, Leszczynska-Piziak J, Abramson SB. Nitric oxide, an endothelial cell relaxation factor, inhibits neutrophil superoxide anion production via a direct action on the NADPH oxidase. J Clin Invest. 1992;90:1116-1121.
Kurose I, Wolf R, Grisham MB, Granger DN. Modulation of ischemia/reperfusion-induced microvascular dysfunction by nitric oxide. Circ Res. 1994;74:376-382.