Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The formation of 3-nitrotyrosine (NO₂Tyr) is the most commonly studied covalent modification of proteins attributed to nitric oxide (NO). Elevated NO₂Tyr formation has been documented in acute lung injury, sepsis, rheumatoid arthritis, amytrophic lateral sclerosis, Alzheimer, and liver transplantation (1, 2). Several mechanisms are involved in tyrosine nitration, including peroxynitrite (formed from the near diffusion-limited reaction between NO and O₂ anions), the reaction of NO with protein tyrosyl radicals (3), the reaction of nitrite and peroxidases (4), and finally nitrous acid (formed in acidic environment such as the stomach) (5).

Little information is available regarding protein tyrosine nitration in skeletal muscles. Recent studies have documented significant NO₂Tyr formation in the limb and ventilatory muscles of animals with severe sepsis (6). Despite this recent progress, many aspects of NO₂Tyr formation in skeletal muscles remain unexplored. For instance, it is unclear whether NO₂Tyr formation occurs in normal skeletal muscle fibers and whether this formation is dependent on the fiber-type composition of various muscles. Constitutive NO synthesis and nitric oxide synthase (NOS) expression inside skeletal muscles is highly dependent on fiber-type composition especially in rats and mice (9, 10). Another important question that remained unanswered is the contribution of the neuronal (nNOS), endothelial (eNOS), and the inducible (iNOS) isoforms of NOSs to NO₂Tyr formation in muscle fibers. We should emphasize that all of these isoforms contribute to elevated NO synthesis in the muscles of septic animals (6). Finally, the nature of tyrosine-nitrated proteins inside skeletal muscle fibers remains unexplored. Numerous proteins have been shown to be tyrosine nitrated in nonmuscle cells and organs including cytosolic and contractile proteins such as actin, catalase, glutathione-S-transferase, tubulin, carbonic anhydrase, and several mitochondrial enzymes including Mn-superoxide dismutase (Mn-SOD), aconitase, ATP synthase, and glutamate dehydrogenase (11- 13). Whether these proteins, particularly Mn-SOD, actin, and tubulin, are also tyrosine nitrated inside skeletal muscle fibers remain unclear.

We hypothesized in this study that protein tyrosine nitration occurs inside normal skeletal muscle fibers and is dependent on NO synthesis. This hypothesis is based on the fact that skeletal muscles are among few organs in which the two components required for peroxynitrite formation (O₂ and NO) are constitutively synthesized (9, 14). We also hypothesized, on the basis of an excellent correlation between the time course of NO₂Tyr formation and iNOS expression in the ventilatory muscles of septic rats (6), that protein tyrosine nitration in these muscles is dependent mainly on the activity of iNOS. To test these hypotheses, we assessed the presence and localization of protein tyrosine nitration in normal ventilatory and limb muscles of various fiber-type compositions. We also evaluated, using selective pharmacologic inhibitors of NOS isoforms and genetically altered mice, which are deficient in a specific NOS isoforms, the contribution of the iNOS, eNOS, and nNOS isoforms to protein tyrosine nitration in the ventilatory and limb muscles both under normal conditions and in response to severe sepsis.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Reagents for protein measurement were purchased from Bio-Rad Inc. (Hercules, CA). Gels and loading buffers for immunoblotting were obtained from Novex Inc. (San Diego, CA). Aprotinin, leupeptin, trypsin inhibitor, pepstatin A, and phenylmethylsulphonyfluoride (PMSF) were purchased from Sigma Chemicals (St. Louis, MO). Monoclonal, polyclonal, and horseradish peroxidase-conjugated polyclonal antibodies for NO₂Tyr were obtained from Cayman Chemical Inc. (Ann Arbor, MI), Upstate Biotechnology (Lake Placid, NY), and Academy Biomedical Co. (Houston, TX), respectively. Secondary antibodies for immunoblotting and immunohistochemistry were obtained from Transduction Laboratories Inc. (Lexington, KY) and Jackson ImmunoResearch Inc. (West Grove, PA), respectively. Reagents for enhanced chemiluminescence detection were obtained from Chemicon Inc. (Temecula, CA).

Animal Preparation

Rat Experiments. The Animal Research Committee of McGill University and University of Virginia approved all procedures. Pathogen-free male Sprague-Dawley rats (250-275 g) were used. The animals were housed in the animal facility of the hospital, fed food and water ad libitum, and studied 1 wk after arrival.

Nitrotyrosine formation in normal rat muscles. Male rats were anesthetized with sodium pentobarbital (30 mg/kg), and the diaphragm, intercostals, gastrocnemius, and soleus muscles were quickly excised and frozen in liquid nitrogen. For immunostaining, the tissues were first flash frozen in cold isopentane (20 s), then immersed in liquid nitrogen and stored at 80°C.

Acute NOS inhibition. To evaluate the acute influence of NOS inhibition on muscle nitrotyrosine formation, three groups of male rats were studied. Group 1 served as control, whereas Groups 2 and 3 were injected intraperitoneally with either a selective iNOS inhibitor (1400W, 20 mg/kg) or a nonselective NOS inhibitor (L-NAME, 30 mg/kg). Both inhibitors were dissolved in 0.3 ml of phosphate-buffered saline (PBS). All animals were killed 1 h later and the diaphragm (mixed fiber composition) and soleus (rich in type I fibers) muscles were quickly excised and frozen in liquid nitrogen.

Sepsis. Five groups of pathogen-free male Sprague-Dawley rats (250-300 g, n = 5 in each group) were studied 1 wk after arrival. Group 1 was injected with normal saline (control group). Groups 2, 3, and 4 were injected intraperitoneally with Escherichia coli lipopolysaccharides (LPS) (serotype 055:B5; Sigma Inc., 20 mg/kg dissolved in 0.3 ml of PBS) and killed by an overdose of sodium pentobarbital 6, 12, and 24 h after the injection, respectively. Group 5 animals were injected with a selective iNOS inhibitor (1400W, 20 mg/kg) 30 min before LPS administration and every 8 h thereafter and were killed 24 h after LPS administration. The diaphragm was quickly dissected, frozen in liquid nitrogen, and stored at 80°C.

Mice Experiments. To evaluate the separate effects of NOS isoforms on nitrotyrosine formation in the diaphragm, we studied three separate groups of adult (8- to 12-week-old) mice genetically deficient (knockout) in iNOS, eNOS, or nNOS isoforms. For the iNOS isoforms, B6/129 hybrid iNOS knockout mice (iNOS^/) were generated as mentioned previously (15) and a full colony of these mice was maintained at McGill University. Wild-type B6/129 hybrid mice (iNOS^+/+) were purchased from Jackson Laboratories (Bar Harbor, ME) and bred to serve as experimental controls. NO₂Tyr formation in normal mice muscles was studied in two groups (n = 6 in each group) of wild-type (control-iNOS^+/+) and iNOS knockout (control-iNOS^/) mice. Animals were anesthetized with sodium pentobarbital (30 mg/kg), the chest was then opened, and the diaphragm was quickly excised and frozen in liquid nitrogen. To evaluate the role of iNOS in sepsis-induced NO₂Tyr formation, two groups (n = 6 in each group) of iNOS^+/+ (LPS-iNOS^+/+) and iNOS^/ (LPS-iNOS^/) mice were injected with intraperitoneal E. coli LPS (20 mg/kg) and killed 24 h later. The diaphragm was excised and frozen in liquid nitrogen as mentioned above. For the nNOS and eNOS isoforms, C57BL/6 nNOS^/ and eNOS^/ mice were generated as mentioned previously (16, 17). The nNOS^/ mouse colony was maintained at McGill University, whereas the eNOS^/ mouse colony was placed at the University of Virginia. Wild-type (WT) C57BL/6 mice (the background strain for both eNOS^/ and nNOS^/ mice) were purchased from Charles River Inc. The roles of eNOS and nNOS isoforms in NO₂Tyr formation in normal muscles were assessed by comparing diaphragm NO₂Tyr levels among control nNOS^/ and eNOS^/ mice and their corresponding wild-type animals. These groups were designated as control-nNOS^/, control-nNOS^/, control-eNOS^+/+ and control-eNOS^/ mice. Sampling of the diaphragms from these animals was performed as mentioned above. The involvement of the nNOS isoform in sepsis-induced NO₂Tyr formation was evaluated by injecting two groups (n = 6 in each group) of nNOS^+/+ (LPS-nNOS^+/+) and nNOS^/ (LPS-nNOS^/) mice with E. coli LPS (20 mg/kg i.p.). Using an identical approach, we compared diaphragmatic NO₂Tyr formation after 24 h of E. coli LPS (20 mg/kg) injection in eNOS^+/+ (LPS-eNOS^+/+) and eNOS^/ (LPS- eNOS^/) mice. All animals were killed after 24 h of LPS injection and the diaphragm was excised and frozen as mentioned above.

Muscle Sample Preparation

Fractionation of muscle samples. Separation of mitochondrial, membrane, and cytosolic muscle fractions was achieved using the protocol of Rock and colleagues (18). The entire procedure was done at 4°C. In brief, frozen muscle samples were homogenized in 6 vol/wt ice-cooled homogenization buffer A (tris-maleate 10 mM, EGTA 3 mM, sucrose 275 mM, DTT 0.1 mM; leupeptin 2 µg/ml; PMSF 100 µg/ml; aprotinin 2 µg/ml, and pepstatin A 1 mg/ 100 ml, pH 7.2). Samples were then centrifuged at 1000 × g for 10 min. The pellet (P1) was discarded, whereas the supernatant (S1) was designated as crude homogenates. These homogenates were then centrifuged at 12,000 × g for 20 min to yield supernatant (S2) and pellet (P2). Pellet (P2) was then resuspended in buffer B (tris-maleate 10 mM, EDTA 0.1 mM, and KCl 135 mM) and then centrifuged at 12,000 × g for 20 min to yield S3 and P3. The resulting pellet (P3) was resuspended in buffer A and designated as the mitochondrial fraction. Both S2 and S3 fractions were pooled and were used to separate the membrane and cytosolic fractions by centrifugation for 1 h at 100,000 × g. The resulting supernatant (S4) was designated as the cytosolic fraction, whereas the pellet (P4) was resuspended in buffer C (HEPES 10 mM and sucrose 300 mM, pH 7.2), treated for 1 h with 600 mM KCl, and then centrifuged again at 100,000 × g for 1 h. Pellet was resuspended in buffer A and designated as the membrane fraction.

Fractionation of muscle samples into myofibrillar, membrane, and cytosolic fractions was performed according to the protocol of Fagan and colleagues (19). In brief, muscle samples were homogenized in ice-cooled pyrophosphate buffer (tris-maleate 0.01M, 0.1M KCl, 2 mM MgCl₂, 2 mM EGTA, 2 mM Na₄P₂O₇, 0.1M Na₂PO₃, and 1 mM DTT, pH 6.8). Samples were then centrifuged at 1000 × g for 10 min. The pellet (P1) was then washed four times with 10-vol low-salt buffer followed by one wash with low-salt buffer containing Triton X-100 (0.02%) and one wash with sodium deoxycholate (0.02%). The pellet was then washed two additional times in low-salt buffer and was finally suspended in pyrophosphate buffer and designated as the myofibrillar fraction. The supernatant (S1) was then centrifuged at 100,000 × g for 1 h to yield supernatant (S2, cytosolic fraction). The resulting pellet (P2) was resuspended in buffer A (see above) and designated as membrane fraction. Protein concentrations of various muscle fractions were measured according to the Bradford technique (BioRad Inc.).

ImmunoblottingCrude homogenates and various muscle fractions of rat diaphragmatic samples (80 µg per sample) were mixed with sample buffer, boiled for 5 min at 95°C, and were then loaded onto 8 or 10% tris-glycine sodium dodecylsulfate (SDS) polyacrylamide gels and separated by electrophoresis (150 V, 30 mA for 1.5 h). Lysates of human and rat neutrophils were used as positive controls (including control for species differences). Proteins were transferred electrophoretically (25 V, 375 mA for 2 h) to methanol presoaked polyvinylidene difluoride (PVDF) membranes, and then blocked with 1% bovine serum albumin (BSA) for 1 h at room temperature. The PVDF membranes were subsequently incubated with primary monoclonal or polyclonal antibodies raised against NO₂Tyr and dissolved in 1% BSA. In a few samples, we also probed membrane with monoclonal antibodies specific to iNOS and nNOS proteins (Transduction Laboratories Inc.). In addition, the efficiency of separating muscle samples into different fractions was verified by probing various muscle fractions with selective antibodies to cytochrome oxidase (mitochondrial marker), caveolin-3 (marker of sarcolemma), and troponin I (myofibril protein marker). After three 10-min washes with wash buffer on rotating shaker, the PVDF membranes were further incubated with horseradish peroxidase (HRP)-conjugated anti-mouse or anti-rabbit secondary antibodies. Specific proteins were detected with a chemiluminescence kit (Chemicon Inc.) The blots were scanned with an imaging densitometer and optical densities (OD) of positive nitrotyrosine protein bands were quantified with SigmaGel software (Jandel Scientific Inc.) Total NO₂Tyr OD was calculated for each sample by adding OD of individual positive protein band. Specificity of anti-NO₂Tyr antibodies was evaluated by preincubation of each primary antibody with either 10-mM nitrotyrosine or 10-fold excess peroxynitrite-tyrosine-nitrated bovine serum albumin (generously provided by Dr. Ischiropoulos, University of Pennsylvania).

ImmunohistochemistryFrozen tissues sections (5-10 µm thickness) were adsorbed to microscope slides and dried. The sections were fixed with acetone at 20°C, rehydrated with PBS containing 1% BSA (pH 7.4), and were then blocked with solutions of avidin, biotin (15 min each at room temperature), and then 3% BSA for 30 min. The sections were incubated for 1 h at room temperature with primary monoclonal or polyclonal antibodies raised against NO₂Tyr. For negative control, the primary antibody was replaced with nonspecific mouse or rabbit IgG. After three rinses with PBS, sections were incubated with biotin-conjugated anti-mouse or anti-rabbit secondary antibodies at room temperature for 1 h followed by exposure to Cy3-labeled strepavidin for 1 h. Sections were then washed, mounted with coverslips, and examined with a Nikon fluorescence microscope and photographed with a 35 mm camera (Nikon Inc.).

Statistical analysisValues are presented as means ± SEM. Differences in OD of individual NO₂Tyr protein band or total sample NO₂Tyr OD between various conditions were compared with one-way analysis of variance (ANOVA) followed by Tukey test for multiple comparisons. P values of less than 5% were considered significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Protein Tyrosine Nitration in Normal Muscles

Figure 1 illustrates NO₂Tyr immunoreactivity in normal rat diaphragms detected with three different antibodies. Monoclonal anti-NO₂Tyr antibody (Cayman Chemical Inc.) detected five nitrated protein bands of 52, 48, 40, 30, and 10 kD. (Figure 1A). The ODs of each of these protein bands expressed as a percentage of total muscle NO₂Tyr OD are listed in Table 1. Polyclonal anti-NO₂Tyr antibody (Upstate Biotechnology) also detected the above-mentioned nitrated proteins in addition to an 18-kD protein band (Figure 1 B). The intensity of nitrated protein bands detected with the polyclonal antibody differed from those delineated by the monoclonal antibody (Table 1). Similarly, HRP-conjugated anti-NO₂Tyr polyclonal antibody (Academy Biomedical) detected positively nitrated protein bands of 52, 40, 30, 18, and 10 kD; however, an additional positive band of 66 kD was also detected and that of 48 kD was not detected in normal rat diaphragms (Figure 1C). Unlike the monoclonal antibody, which preferentially detects 30- and 52-kD protein bands, the intensity of the 10-kD protein band was the highest proportion of total diaphragmatic NO₂Tyr signal detected with the HRP-conjugated anti-NO₂Tyr antibody (Figure 1C and Table 1). When various rat muscles were compared with respect to nitrotyrosine immunoreactivity, intercostals muscles showed similar pattern of nitrotyrosine proteins to that of the diaphragm, whereas gastrocnemius and soleus muscles have weaker protein nitration at 10 kD compared with the diaphragm and intercostals muscles (Figure 1D). Figure 1E illustrates the specificity of monoclonal anti-NO₂Tyr in detecting NO₂Tyr formation in rat diaphragm. Preincubation of monoclonal anti-NO₂Tyr antibody with 10-fold excess of peroxynitrite-tyrosine-nitrated bovine serum albumin completely eliminated NO₂Tyr immunoreactivity (two right lanes, underlined). Similar results were obtained when this antibody or polyclonal antibodies were preincubated with 10 mM nitrotyrosine (data not shown).

View larger version (38K): [in this window] [in a new window]   Figure 1.   Representative immunoblots of two different rat diaphragmatic (D) homogenates probed with monoclonal (Cayman Chemical Inc. [A]), polyclonal (Upstate Biotechnology Inc. [B]) and HRP-conjugated polyclonal (Academy Biomedical Co. [C]) anti-NO2Tyr antibodies. Note the differences in the intensity of individual protein bands among the three immunoblots. (D) A representative immunoblot of rat diaphragm (D), intercostals (I), gastrocnemius (G), and soleus (S) muscles probed with monoclonal anti-NO2Tyr antibody. (E) Selectivity of monoclonal anti-NO2Tyr antibody. Rat diaphragmatic samples (20 and 40 µg total protein per lane) were probed with monoclonal anti-NO2Tyr antibody (two left lanes). The two right lanes (underlined) show diaphragmatic muscle samples probed with the same antibody, which was preincubated with 10-fold excess of nitrated bovine serum albumin. Note that this preincubation resulted in disappearance of positive nitrotyrosine protein bands shown in the left two lanes.

Figure 2 illustrates the presence of NO₂Tyr immunoreactivity in various muscle fractions. Monoclonal anti-NO₂Tyr antibody detected positive NO₂Tyr protein bands mainly in the cytosolic fraction, whereas only weak bands were detected in the myofibrillar, membrane, and mitochondrial fractions (Figures 2A and 2B).

View larger version (49K): [in this window] [in a new window]   Figure 2.   (A) Representative immunoblot of myofibrillar, membrane, and cytosolic fractions of normal rat diaphragm samples probed with monoclonal anti-NO2Tyr antibody. (B) Membrane, mitochondrial and cytosolic fractions of a rat diaphragm probed with monoclonal anti-NO2Tyr antibody. Note that the majority of tyrosine nitrated protein bands are localized in the cytosolic fraction.

Localization of nitrotyrosine immunoreactivity in normal rat muscles is shown in Figure 3. Monoclonal anti-NO₂Tyr antibody detected positive staining in gastrocnemius (Figures 3A and 3B) and soleus (Figure 3C) in close proximity to the sarcolemma. Positive NO₂Tyr staining was also detected in nerve fibers (arrows in Figure 3D), but not in blood vessels traversing skeletal muscles (Figure 3D). Polyclonal anti-NO₂Tyr also detected positive NO₂Tyr in the diaphragm of normal rats in close proximity to the sarcolemma (Figure 3E). We should emphasize that we detected in this muscle punctate cytosolic-positive NO₂Tyr staining suggesting that a proportion of tyrosine-nitrated proteins is not close to the sarcolemma. Replacement of primary antibodies with nonspecific mouse IgG or rabbit IgG completely eliminated positive NO₂Tyr staining (Figure 3F).

View larger version (128K): [in this window] [in a new window]   Figure 3.   Immunohistochemical localization of tyrosine-nitrated proteins in normal rat muscles. Monoclonal anti-NO2Tyr antibody detected positive staining in gastrocnemius (A and B) and soleus (C) muscle sections in close proximity to the sarcolemma. Positive NO2Tyr staining was also detected in nerve fibers (arrows in D) but not in blood vessels traversing gastrocnemius muscle. Detection of protein tyrosine nitration in the diaphragm of rats with polyclonal anti-NO2Tyr antibody is shown in (E), whereas (F) shows negative control staining in which the primary anti-NO2Tyr antibody was replaced by nonspecific mouse IgG. Magnification of all panels except (B) is ×200. (B) is magnified at ×400.

Role of NOS in Protein Tyrosine Nitration in Normal Muscles

Examination of crude homogenates (Figure 4A) and various muscle fractions (Figure 4B) after 1 h of in vivo administration of either a selective iNOS inhibitor (1400W) or nonselective NOS inhibitor (L-NAME) in normal rats revealed significant reductions in total diaphragmatic NO₂Tyr levels (Figure 4C). Administration of 1400W reduced NO₂Tyr OD at 52 and 48 kD to 72 and 64% of control diaphragms, respectively (P < 0.05). Those of 30 and 10 kD remained similar to control muscles. The inhibitory effect of 1400W on tyrosine nitration of 52- and 48-kD proteins was much more pronounced when muscle homogenates were fractioned into mitochondrial, cytosolic, and membrane fractions (Figure 4B). L-NAME administration produced relatively larger reduction in NO₂Tyr levels of crude diaphragm homogenates compared with that elicited by 1400W, and lowered NO₂Tyr ODs of 52-, 48-, and 30-kD protein bands to 20, 48, and 81% of control muscles, respectively (P < 0.01).

View larger version (32K): [in this window] [in a new window]   Figure 4.   Effects of acute (within 1 h) inhibition of NO synthesis on the intensity of protein tyrosine nitration in the crude lysates (A) and in the mitochondrial, cytosolic, and membrane fractions (B) of the diaphragm. Rats were injected with either 1400W (selective iNOS inhibitor) or L-NAME (nonselective NOS inhibitor). (C) Total muscle NO2Tyr OD of control, 1400W-, and L-NAME- treated animals. *P < 0.05 compared with control.

In addition to pharmacologic inhibition of NOS activity in rats, we compared diaphragmatic NO₂Tyr OD among wild-type mice and mice deficient in either iNOS, eNOS, or nNOS (knockout) isoforms (Figure 5). Anti-NO₂Tyr antibodies detected in the diaphragms of control-iNOS^+/+ (B129/C57Bl6 wild-type), control-nNOS^+/+ and control- eNOS^+/+ (Bl6 wild-type) mice several nitrated protein bands, which are similar to those detected in the diaphragms of normal rats (Figures 5A-5C). In control- iNOS^/ mice, the intensities of these bands, particularly those of 52-, 48-, 40-, and 30-kD protein bands, were significantly lower than in iNOS^+/+ mice (Figure 5A). By comparison, absence of nNOS or eNOS proteins has no effects on the intensity of NO₂Tyr protein bands compared with wild-type mice (Figures 5B and 5C). Figure 5D shows total diaphragmatic NO₂Tyr ODs in the six groups of control animals. Total NO₂Tyr OD was significantly lower in iNOS^/ mice compared with that of iNOS^+/+ mice (P < 0.01), whereas no significant differences were observed in total diaphragmatic NO₂Tyr levels between eNOS^/ and nNOS^/ animals and their corresponding wild-type animals.

View larger version (26K): [in this window] [in a new window]   Figure 5.   (A-C) Representative NO2Tyr immunoblots of diaphragmatic samples obtained from control iNOS/, nNOS/, and eNOS/ mice and their corresponding wild-type (+/+) animals. Note that the absence of iNOS but not eNOS or nNOS, had a major effect on tyrosine nitration of diaphragmatic proteins. (D) Mean values of total diaphragmatic NO2Tyr OD in control wild-type and NOS knockout mice. **P < 0.01 compared with wild-type mice. Note that the total diaphragmatic NO2Tyr OD is significantly lower in iNOS/ mice compared with iNOS+/+ mice.

Protein Tyrosine Nitration in Septic Muscles

The influence of sepsis on NO₂Tyr formation in the ventilatory and limb muscles was assessed both in rats and mice. Injection of E. coli LPS in rats elicited a significant induction of the iNOS isoform in the diaphragm and intercostals muscles, which peaked after 12 h of LPS administration (Figure 6). In addition to iNOS induction, LPS injection produced a significant rise in nNOS protein expression, which peaked after 24 h of LPS administration (Figure 6). Injection of LPS in rats produced a significant rise in NO₂Tyr OD of crude diaphragmatic homogenates with total NO₂Tyr OD after 6, 12, and 24 h of LPS injection reaching 120 ± 3, 134 ± 7, and 175 ± 6% of control samples, respectively (P < 0.05, Figure 7A). Fractionation of muscle homogenates revealed that the increase in NO₂Tyr formation in response to LPS injection was localized mainly in the mitochondrial and membrane fractions rather than in the cytosolic fraction (Figures 7B and 7C). The rise in mitochondrial and membrane NO₂Tyr levels in response to LPS injection was evident mainly at 52-, 48-, and 30 kD-protein bands rather than at the 10-kD band. Selective inhibition of iNOS activity in septic rats by the administration of 1400W before and after LPS injection resulted in significant attenuation of LPS-induced rise in diaphragmatic NO₂Tyr formation. Indeed, NO₂Tyr OD of 52-, 48-, and 30-kD protein bands in septic rats that received 1400W reached 63, 76, and 77% of corresponding values in rats receiving only LPS, respectively (P < 0.05).

View larger version (26K): [in this window] [in a new window]   Figure 6.   (A) Expression of iNOS (top) and nNOS (bottom) proteins in the diaphragm and intercostals muscles of rats in response to LPS injection. 0 refers to control animals, whereas +ve refers to positive controls. Note the transient nature of iNOS expression and the upregulation of nNOS protein after 24 h of LPS administration. (B) Optical densities of nNOS and iNOS proteins (mean values of five animals) in the diaphragm and intercostals muscles of control (0 h) and LPS-injected animals. *, ** P < 0.05 and 0.01, respectively, compared with control values (0 h).

View larger version (29K): [in this window] [in a new window]   Figure 7.   (A) Influence of LPS injection on diaphragmatic NO2Tyr formation. Notice that NO2Tyr levels rose substantially 24 h after LPS compared with control samples. (B) Localization of NO2Tyr formation in control and septic diaphragmatic muscle samples. Notice the appearance of tyrosine-nitrated proteins in the mitochondrial and membrane fractions but not the cytocolic fractions of rat diaphragms in response to LPS injection. (C) Mean values of total NO2Tyr OD in the three fractions of diaphragmatic samples obtained from control (open bars) and septic (24 h after LPS injection; filled bars) rats. *P < 0.05 compared with control animals.

The involvement of NOS isoforms in LPS-induced rise in muscle NO₂Tyr formation was also assessed in wild-type mice, iNOS^/, nNOS^/, and eNOS^/ mice. LPS injection in iNOS^+/+ mice increased total diaphragmatic NO₂Tyr OD by a mean value of 56% of control-iNOS^+/+ mice (P < 0.05, Figure 8A). Only the 10-kD protein band did not show any rise in tyrosine nitration in response to LPS injection in iNOS^+/+ mice (Figure 8A). By comparison, LPS injection in iNOS^/ mice did not elevate diaphragmatic tyrosine nitration (mean values of 91% of control-iNOS^/ mice, Figure 8A). Injection of LPS in nNOS^+/+ and nNOS^/ mice produced a similar degree of increased tyrosine nitration in diaphragmatic lysates (mean values of 161 and 168% of corresponding control animals, respectively). Moreover, similar augmentation in diaphragmatic tyrosine nitration was noticed among eNOS^+/+ and eNOS^/ mice in response to LPS injection (154 and 149% of corresponding control animals, respectively). Figure 8B summarizes total diaphragm NO₂Tyr OD measured in response to LPS injection in nNOS^/, iNOS^/, and eNOS^/ mice (results are expressed as % of their corresponding septic wild-type animals). Whereas total tyrosine nitration in LPS-nNOS^/ and LPS-eNOS^/ mice was similar to their corresponding wild-type animals, total tyrosine nitration in LPS-iNOS^/ mice was significantly lower than LPS-iNOS^+/+ mice. These results suggest that the absence of iNOS, but not nNOS or eNOS, significantly alters the response of muscle tyrosine nitration to LPS injection.

View larger version (33K): [in this window] [in a new window]   Figure 8.   (A) Comparison of diaphragmatic protein tyrosine nitration in wild-type (iNOS+/+) and iNOS knockout (iNOS/) mice under control condition and after 24 h of LPS injection. (B) Mean values of total NO2Tyr OD of diaphragmatic muscle samples obtained after 24 h of LPS injection in iNOS/, nNOS/, and eNOS/ mice. Values are expressed as percentage of corresponding septic wild-type animals. *P < 0.05 compared with wild-type animals.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The main findings of this study are that (i) abundant NO₂Tyr formation is detected in the cytosolic fraction of normal rat and mouse ventilatory and limb muscles; (ii) NO₂Tyr immunoreactivity in the ventilatory and limb muscles is limited to specific protein bands ranging in apparent molecular mass between 66 and 10 kD; (iii) pharmacologic inhibitors and knockout mice experiments revealed that in normal ventilatory muscles, the iNOS isoform is a major contributor to NO₂Tyr formation; and finally, (iv) injection of LPS resulted in augmentation of muscle NO₂Tyr formation, particularly in the mitochondrial and membrane fractions, and was dependent in large part on muscle iNOS activity.

Little information is available regarding NO₂Tyr formation in the ventilatory and limb muscles. Supinski and colleagues (20) used dot-blotting technique to demonstrate a significant increase in NO₂Tyr formation in the diaphragm of septic rats. Our group reported recently that two protein bands (50 and 42 kD) are tyrosine nitrated in normal rat diaphragms, and two additional protein bands (196 and 86) are heavily nitrated in the diaphragm of septic rats (6). In a more recent study, Boczkowski and colleagues (8) described tyrosine nitration of a single 105-kD protein band in diaphragmatic mitochondria of septic rats and in response to exposure of isolated muscle mitochondria to SIN-1 (peroxynitrite donor). We detected in the current study abundant protein tyrosine nitration in the ventilatory and limb muscles of both normal rats and mice. This tyrosine nitration was noticeable with three different antibodies, though various intensities of specific proteins bands were apparent depending on the type of the antibody used. The fact that more tyrosine nitrated protein bands were detected with polyclonal antibodies is not surprising since these antibodies are capable of detecting more diverse epitopes than monoclonal antibodies. We attribute the failure to detect abundant protein nitration in normal skeletal muscles in previous studies to lack of protein separation (as in Supinski and colleagues [20]), incomplete separation of various proteins (as in our previous study [6]), or measurement of NO₂Tyr formation in one muscle fraction (mitochondria, in Boczkowski and colleagues [8]).

Our finding of a significant reduction in intensity of tyrosine-nitrated proteins in the diaphragm of rats injected with NOS inhibitors and in the diaphragms of iNOS^/ mice confirms that NO₂Tyr formation in skeletal muscle fibers is dependent on NO production. Moreover, we found that the level of specific tyrosine nitrated proteins (52 and 48 kD) declined significantly within 1 h of NOS inhibition, whereas other nitrated proteins (30 and 10 kD) were less sensitive to rapid NOS inhibition (Figure 4). We speculate that increased turnover rate of tyrosine-nitrated proteins can explain this finding. Our speculation in this respect is supported by the observation of Souza and colleagues (21), namely, that protein nitration enhances susceptibility to degradation by the proteasome. Another possible mechanism for rapid decline in tyrosine nitration after 1 h of NOS inhibition is that an enzymatic mechanism exists to denitrate nitrated tyrosine residues. The existence of such mechanism (nitrotyrosine denitrase) has been proposed by Kamisaki and colleagues (22), who found that homogenates of rat lung and spleen are capable of modifying tyrosine-containing proteins in a time and protein concentration-dependent fashion. The exact nature and molecular structure of "nitrotyrosine denitrase" remain to be elucidated.

Although the exact contribution of each of the NOS isoforms expressed in skeletal muscle fibers to protein tyrosine nitration in normal muscle remains speculative, our experiments in which NOS isoforms were pharmacologically inhibited or genetically altered in mice clearly suggest that the iNOS isoform is the main contributor to NO₂Tyr formation in normal muscles. These results are rather surprising since relatively low levels of iNOS protein have been detected in normal skeletal muscles of various species compared with the other two isoforms (nNOS and eNOS) (6, 23, 24). It is possible that even low levels of iNOS protein are sufficient to elicit protein tyrosine nitration in skeletal muscles because of the much higher rate of NO synthesis by the iNOS isoform compared with the nNOS and eNOS isoforms. Although previous studies in neurons suggest that the nNOS isoform is involved in protein tyrosine nitration (25), our results using nNOS^/ and eNOS^/ mice suggest that both of these isoforms do not play a major role in protein tyrosine nitration in normal skeletal muscles.

Previous studies have documented an increase in protein tyrosine nitration in the ventilatory muscles of septic humans and rats (6). It has also been shown that systemic inhibition of NOS in septic animals reduces mitochondrial tyrosine nitration (8). The contribution, however, of various NOS isoforms to the rise in protein tyrosine nitration in septic animals or the presence of tyrosine nitration in various muscle compartments were not assessed in the above-mentioned studies. We found that muscle NO₂Tyr levels increase significantly in septic animals and that tyrosine nitration was mainly localized in the mitochondrial and membrane fractions but not in the cytosolic fraction. Our results also indicate that LPS-induced protein nitration is mediated primarily by the iNOS isoform but not by the nNOS and eNOS isoforms despite the fact that the levels of these isoforms in the ventilatory muscles are elevated in response to LPS injection (Figure 6) (6). The mechanisms responsible for in vivo nitration of tyrosine residues remain the focus of intense investigation and debate over the past several years. The most widely accepted mechanism of in vivo tyrosine nitration is peroxynitrite, which is formed from the near diffusion-limited reaction between NO and superoxide anions (1). The identity of peroxynitrite as the reaction product of NO and superoxide anions, as well as its ability of peroxynitrite to nitrate tyrosine residues at physiologic pH has recently been confirmed by Reiter and colleagues (26). We propose that protein tyrosine nitration in skeletal muscles is mediated primarily by peroxynitrite and that iNOS is the primary source of NO required for peroxynitrite formation. Other proposed pathways mediating protein tyrosine nitration include the reaction of NO with tyrosyl radical generated by prostaglandin H synthase-2 (3), and oxidation of NO₂ by H₂O₂ at physiologic pH, which can result in the formation of peroxynitrous acid and consequently lead to nitration of tyrosine residues (2). We speculate that these two pathways are not likely to contribute to tyrosine nitration in skeletal muscles simply because the first pathway necessitates the presence of abundant prostaglandin H synthase-2 expression, which is not usually present in skeletal muscle fibers. The second pathway requires relatively high concentrations of H₂O₂ (> 1 mM), although normal skeletal muscle H₂O₂ levels are quite small (27). Finally, it has been proposed recently that myeloperoxidase (MPO), utilizing both NO₂ and H₂O₂, is capable of nitrating tyrosine residues (28). We believe that this pathway does not play a major role in tyrosine nitration in normal skeletal muscle because MPO is localized mainly in polymorphonuclear leukocytes, whereas protein tyrosine nitration is detected inside muscle fibers. Moreover, the MPO pathway requires relatively high levels of NO₂ and H₂O₂ and more than 1 h to produce tyrosine nitration (29).

A major finding in our study is that muscle NO₂Tyr is limited to specific protein bands ranging in apparent mass between 66 and 10 kD, though the intensity of these bands in normal muscles and the changes in their intensity in response to inhibition of NOS differed significantly depending on the antibody used. Tyrosine nitration is a selective process, which is dependent on several factors such as the nature of the nitrating agent, the exposure of the aromatic ring to the surface of protein, the location of tyrosine on a loop structure, and the presence of glutamate in the local environment of the tyrosine residue (2). Interestingly, tyrosine nitration is not influenced by protein abundance or the abundance of tyrosine residues in a given protein (2). Major progress has recently been made in identifying tyrosine-nitrated proteins. The majority of these proteins are cytosolic, including catalase, glutathione-S-transferase, carbonic anhydrase III, tyrosine hydroxylase, cAMP-dependent protein kinase, lactate dehydrogenase, glycogen synthase, and transketolase (12, 13, 30). In addition, cytoskeletal proteins such as actin, neurofilament L, tubulin, and myofibrillar creatine kinase have recently been shown to be tyrosine nitrated (11, 12, 31). Tyrosine nitration of several mitochondrial proteins such as aconitase, Mn-SOD, ATP synthase, glutamate dehydrogenase, and glutamate oxaloacetate transaminase-2 has also been described (12, 33). Finally, recent studies suggest that nuclear proteins such as histones II-S and VIII-S can be tyrosine nitrated (13). The influence of tyrosine nitration on protein function remains in most cases unclear. There is, however, evidence that tyrosine nitration may result in loss of function. For instance, nitration of specific tyrosine residues of Mn-SOD or 1-antitrypsin causes a significant inhibition of activity of these proteins (33, 34). Nitration of C-terminal tyrosine residue in -tubulin compromises microtubule organization and binding of microtubule-associated proteins (32).

Our findings that the majority of tyrosine-nitrated proteins in normal skeletal muscles are residing in the cytosol is in agreement with the observations Aulak and colleagues (12) that described NO₂Tyr formation in cultured A459 cells and in the liver and lungs of LPS-injected rats. We did observe, however, that protein tyrosine nitration rose significantly in the mitochondrial and membrane fractions in response to LPS injection. Although our study does not provide a comprehensive list of tyrosine-nitrated proteins in skeletal muscles, we can exclude several proteins from the list of tyrosine-nitrated proteins (12, 13, 32, 33). The fact that we did not observe abundant tyrosine nitration in the myofibrillar fraction excludes the possibility of sarcomeric actin and myofibrillar creatine kinase. Moreover, we can exclude Mn-SOD from the list of possible tyrosine-nitrated proteins in our study because the apparent molecular masses of tyrosine-nitrated proteins in various muscle fractions do not correspond to that of Mn-SOD and that this protein is localized in the mitochondria where little tyrosine nitration was detected in normal muscles. Other previously described tyrosine-nitrated proteins such as neurofilament L and lung surfactant protein A can also be excluded because they are not usually abundantly expressed inside normal skeletal muscle fibers. One possible candidate for tyrosine nitration in the cystosol of normal muscles is tubulin, which has a similar molecular mass (49 kD) to protein bands detected by various anti-NO₂Tyr antibodies in normal muscle samples (Figure 1). Both the  and  isoforms of tubulin are known to undergo tyrosine nitration under specific conditions (12, 32). We have conducted several experiments in which tyrosine-nitrated proteins in normal diaphragm muscle lysates were immunoprecipitated with monoclonal and polyclonal anti-NO₂Tyr antibodies. The immunocomplexes were then separated on SDS-PAGE, transferred to PVDF membranes and probed with anti-tubulin antibody (Sigma Chemical Inc.). We were unable to detect positive tubulin protein bands in the NO₂Tyr immunocomplexes suggesting that tubulin is not tyrosine nitrated in normal rat diaphragm. Clearly, more research is needed to elucidate the identity of tyrosine-nitrated proteins in skeletal muscle.

In summary, our study indicates that there is abundant protein tyrosine nitration in the cytosol of both ventilatory and limb muscles and that tyrosine nitration is limited to specific proteins ranging in molecular mass between 66 and 10 kD. We also found that sepsis elicits a significant augmentation of NO₂Tyr formation in the ventilatory muscles that is localized in the mitochondrial and membrane fractions. Finally, NO₂Tyr formation both in normal and septic muscles appears to be dependent on the activity of the inducible NOS.