Chronically ischemic mouse skeletal muscle exhibits myopathy in association with mitochondrial dysfunction and oxidative damage

Abstract

A myopathy characterized by mitochondrial pathology and oxidative stress is present in patients with peripheral arterial disease (PAD). Patients with PAD differ in disease severity, mode of presentation, and presence of comorbid conditions. In this study, we used a mouse model of hindlimb ischemia to isolate and directly investigate the effects of chronic inflow arterial occlusion on skeletal muscle microanatomy, mitochondrial function and expression, and oxidative stress. Hindlimb ischemia was induced by staged ligation/division of the common femoral and iliac arteries in C57BL/6 mice, and muscles were harvested 12 wk later. Muscle microanatomy was examined by bright-field microscopy, and mitochondrial content was determined as citrate synthase activity in muscle homogenates and ATP synthase expression by fluorescence microscopy. Electron transport chain (ETC) complexes I through IV were analyzed individually by respirometry. Oxidative stress was assessed as total protein carbonyls and 4-hydroxy-2-nonenal (HNE) adducts and altered expression and activity of manganese superoxide dismutase (MnSOD). Ischemic muscle exhibited histological features of myopathy and increased mitochondrial content compared with control muscle. Complex-dependent respiration was significantly reduced for ETC complexes I, III, and IV in ischemic muscle. Protein carbonyls, HNE adducts, and MnSOD expression were significantly increased in ischemic muscle. MnSOD activity was not significantly changed, suggesting MnSOD inactivation. Using a mouse model, we have demonstrated for the first time that inflow arterial occlusion alone, i.e., in the absence of other comorbid conditions, causes myopathy with mitochondrial dysfunction and increased oxidative stress, recapitulating the muscle pathology of PAD patients.

peripheral arterial disease (PAD) is a manifestation of systemic atherosclerosis, producing stenoses and occlusions in the arterial tree supplying the lower extremities. PAD affects ∼10 million people in the United States, most of them elderly (18). In the early stages of PAD, patients present with a significant impairment of their ability to walk (claudication), whereas, in more advanced stages, claudication worsens, and patients also may develop nonhealing foot ulcers and gangrene. Procedures to treat these patients constitute some of the most common operations performed by vascular specialists nationwide (18). With the expected increases in the elderly population, PAD is fast becoming one of the most significant causes of morbidity and mortality in the United States (18).

Claudication is a consequence of compromised blood supply to the ischemic limb and a metabolic myopathy affecting the ischemic muscles (4, 5, 22, 23). Defective mitochondria are central to this myopathy through compromised performance as primary energy producers and regulators of oxygen radical species, contributing to a progressive deterioration in muscle function and microanatomy (4, 5, 17, 22, 23). PAD myopathy is characterized by an increased content of dysfunctional mitochondria having significant defects in electron transport chain (ETC) complexes I, III, and IV (17, 24). These defects are associated with a bioenergetic decline, characterized by inadequate oxidative phosphorylation, decreased ATP energy production, and increased oxidative stress (4, 5, 17, 23–25). All start with inflow arterial occlusions, which produce low blood supply and myopathy (due to ischemic injury), and the two, in conjunction, produce the manifestations of PAD claudication (5, 14, 22, 23).

The long-term effects of arterial occlusive disease on the physiology of skeletal muscle, particularly in relation to its microanatomy and mitochondrial function, have been evaluated only in atherosclerotic PAD patients (5, 14, 22, 23). Patients with PAD differ in disease severity, mode of presentation, and presence of comorbid conditions, and such differences may complicate the study of skeletal muscle physiology in these patients. Rodents subjected to inflow arterial ligation exhibit characteristics resembling moderate-severe PAD in humans and have been used to evaluate the effects of occlusive arterial disease on angiogenesis and muscle physiology and bioenergetics (8, 10). In the present study, we evaluated skeletal muscle microanatomy, mitochondrial respiration, and oxidative stress in an established mouse model of chronic inflow arterial occlusion. The hypothesis driving the present study is that inflow arterial occlusion alone, in the absence of comorbidities, causes myopathy with mitochondrial dysfunction, increased mitochondrial content, and increased oxidative stress.

METHODS

Animals

Female C57Bl/6 mice (N = 48) (Charles River Laboratories), weighing 30–35 g, were placed on study at 6 mo of age, when ligation of the left femoral artery (N = 24) and sham operation (N = 24) were implemented. The experimental protocol was approved by the Institutional Animal Care and Use Committee of the University of Nebraska Medical Center.

Hindlimb Ischemia Model

Arterial ligation and division and tissue collection were carried out on mice anesthetized with an intraperitoneal injection of tribromoethanol (Avertin; 0.25 mg/g body wt). Arterial occlusions were induced in two stages. First, the left femoral artery was ligated proximal to the superficial epigastric artery. Two weeks later, the left common iliac artery was ligated distal to the aortic bifurcation. These animals exhibited no gross changes in hindlimb appearance. Control mice received similarly staged incisions and anatomic dissections without arterial ligation. Twelve weeks after the second operation, both ligated and control mice were anesthetized. Soleus muscle was harvested and used fresh for measurement of mitochondrial respiration. Tibialis anterior muscle was harvested and fixed in methacarn (methanol-chloroform-acetic acid 60:30:10) for microscopy. The remainder of the lower (below knee) hindlimb muscles (including gastrocnemius, plantaris, tibialis posterior, flexor hallucis longus, flexor digitorum longus, extensor digitorum longus, and peroneus muscles) was removed and immediately frozen for later biochemical measurements, including analysis of manganese superoxide dismutase (MnSOD) and citrate synthase (CS) activities and quantification of MnSOD, 4-hydroxy-2-nonenal (HNE) adducts, and carbonyl groups by reverse-phase protein array analysis (RPPA).

Bright-Field and Fluorescence Microscopy

Tibialis anterior muscles were taken from chronically ligated and control mice. Gastrocnemius muscle samples from a PAD patient and a control subject were harvested, in accordance with a protocol approved by the Internal Review Board of the University of Nebraska Medical Center. The samples were obtained from the anteromedial aspect of the muscle belly 10 cm distal to the tibial tuberosity. Muscle specimens were fixed in methacarn, embedded in paraffin, sectioned at 4 μm, and mounted to glass slides. For both bright-field and fluorescence microscopy, slide specimens were deparaffinized in xylene and rehydrated to water. Images of specimens stained with hematoxylin and eosin were captured under bright-field illumination with a Leica DMRXA2 microscope (North Central Instruments), configured with a Leica DFC color camera (North Central Instruments) and ImagePro Plus software (Media Cybernetics). For fluorescence imaging, rehydrated slide specimens were heated at 95°C in 0.01 M citrate buffer (with 0.05% Tween 20; pH 6.0) for 30 min and then soaked in SuperSensitive Wash buffer (BioGenex Laboratories) for 30 min at room temperature. Labeling barriers were applied to dry areas around the tissue specimens, which were kept moist. Specimens were loaded into a programmable autostainer (i6000; BioGenex Laboratories) and washed three times (10 min per wash) with wash buffer and then blocked with a 20-min exposure to 10% goat serum (Zymed Laboratories). Subsequently, the goat serum was manually tapped off of the slides, and the tissue specimens were covered with primary antibody (Ab) or mouse IgG (eBiosciences Laboratories) as a negative control. Primary Abs used were monoclonal anti-MnSOD Ab (R&D Systems), anti-ATP synthase Ab (Mitosciences), and anti-HNE Ab (R&D Systems). After an overnight incubation at 5°C, slides were loaded into the autostainer and labeled for 1 h with a secondary goat anti-mouse IgG Ab, coupled with Alexa Fluor 488 or Alexa Fluor 568 (Molecular Probes), and then washed with wash buffer. Labeled specimens were mounted with ProLong Gold antifade reagent (Molecular Probes), the coverslips were sealed with clear lacquer, and the slides were stored at 5°C. Fluorescence images were captured with the Leica DMRXA2 microscope configured with a charge-coupled device B/W camera (Hamamatsu) and ImagePro Plus software.

Evaluation of Mitochondrial ETC Complexes

Respiration of skinned (saponin-permeabilized) soleus muscle fibers (26) was measured with a Clark electrode (YSI), in four separate assays corresponding to the following sets of conditions: 1) 5 mM glutamate, 5 mM malate, and 1 mM ADP (complex I-dependent respiration); 2) 3 μM rotenone (to inhibit complex I), 1 mM ADP, and 10 mM succinate (complex II-dependent respiration); 3) 3 μM rotenone, 1 mM ADP, and 1 mM duroquinol (complex III-dependent respiration); 4) 3 μM rotenone, 1 mM ADP, 10 mM ascorbate, and 0.2 mM N,N,N′,N′-tetramethyl-p-phenylenediamine (complex IV-dependent respiration). Respiratory rates (nanoatoms of oxygen per minute) were normalized to CS activity and total protein, immediately after completion of the respiratory measurements.

CS activity was measured as the increase in absorbance (412 nm) from the reduction of 5,5′-dithiobis-2-nitrobenzoic acid by newly formed CoA-SH (30). Total protein concentration of each sample was determined with the BCA Protein Assay Kit (Pierce).

Analysis of MnSOD Activities in Calf Muscle Homogenates

MnSOD activity was quantified by a modification of the nitrite method (3, 11, 20). Briefly, a SOD activity reference set (0, 6.25, 12.5, 25, 50, 100, 200, and 400 ng MnSOD/20 μl) was prepared with standard MnSOD (Sigma-Aldrich, St. Louis, MO; catalogue no. S-5639), and each sample set was prepared by adding muscle homogenate at a dilution of 1/20 to each solution of a duplicate reference set. For reference and sample sets, signal reduction was plotted as a function of the quantity of MnSOD in the activity reference set. The shift of the X-intercept at the 50% end point of the sample curve relative to the 50% end point of the reference curve was a linear function (R² = 0.997) of MnSOD added to the reference set. Standard SOD activity units in each sample were computed as nanogram equivalents of standard MnSOD and normalized to homogenate protein.

Quantification of MnSOD, HNE Adducts, and Carbonyl Groups in Whole Muscle Homogenates by RPPA

Preparation of muscle homogenates for RPPA.

Calf muscle homogenates were analyzed by the methods of Calvert et al. (7). Homogenates were diluted in extraction buffer to 100 μg protein/ml (final volume 250–500 μl) and then heated at 70°C for 2 h (Eppendorf Thermomixer 5436). Samples were sonicated (Branson 2110 Branson Ultrasonics, Danbury, CT) for 10 min, heated at 95°C for 10 min, and then centrifuged (Fisher Micro-Centrifuge model 235V Fisher Scientific, Pittsburgh, PA) at 5,000 g and room temperature for 10 min.

General procedures for RPPA.

Samples were spotted (50 nl/spot) in triplicate on duplicate nitrocellulose slides (LI-COR Biosciences Laboratories), with an eight-pin arrayer (VP478; V&P Scientific), according to manufacturer instructions. Slides were treated with primary Ab or isotype control for 2 h at room temperature and then with IRDye 800-conjugated secondary Ab (1:2,500 dilution; LI-COR Biosciences Laboratories). Washed slides were air dried and then analyzed with the Odyssey IR imaging system (LI-COR Biosciences Laboratories). Integrated fluorescence intensities were determined for each spot.

Analysis of MnSOD.

One slide was treated with rabbit anti-MnSOD Ab (Stressgen) diluted to 2 μg/ml blocking buffer, and the duplicate with rabbit IgG (2 μg/ml) (Vector Laboratories). Monospecificity of the Ab was determined by Western analysis of muscle homogenate (data not shown). MnSOD concentration (μg/ml) of each homogenate was determined with a standard curve expressing MnSOD concentration (μg/ml) as a linear function (R² = 0.999) of integrated fluorescence intensity. Expression was normalized to total protein in each homogenate.

Analysis of HNE adducts.

Adducts were labeled with monoclonal Ab (2 μg/ml) (kind gift of Dr. Koji Uchida, Institute for Advanced Research, Nagoya University, Japan), specific for the tetrahydrofuran (ring) form of the Michael adduct (32), and the isotype control slide was treated with mouse IgG (2 μg/ml) (eBiosciences). Labeling was inhibited >90% with 4 × 10⁻⁵ M HNE-histidine (gift of Dr. Koji Uchida) in the primary Ab solution. Additionally, treatment of extracts and HNE standards with 25 mM NaBH₄ prevented labeling (9, 32). Integrated intensity of the HNE-BSA standards (Alpha Diagnostic) was a linear function of concentration of the standards (R² = 0.996), and HNE adducts in sample homogenates were expressed as HNE-BSA standard units per microgram protein.

Analysis of carbonyl groups.

A carbonyl standard was prepared by oxidizing BSA (OxBSA) with ferrous ammonium sulfate (1), and a negative control was prepared by reducing OxBSA with sodium borohydride (1). Borohydride-reduced and nonreduced homogenates and standards were arrayed on duplicate slides. Reduced samples served as background controls. Protein carbonyls were labeled with mouse monoclonal anti-dinitrophenol Ab (Sigma-Aldrich, St. Louis, MO; 2 μg/ml), and the isotype control slide was treated with mouse IgG (eBiosciences Laboratories, San Diego, CA; 2 μg/ml). Protein carbonyls were determined from a standard curve relating integrated intensity of OxBSA standards as a linear function (R² = 0.999) of OxBSA concentration (μg/ml) and expressed as micrograms of OxBSA per milligram homogenate protein.

Data Analysis

Results were expressed as means ± SE. Sigmastat (Jandel) and PlotIT (Scientific Programming Enterprises) software packages were used for computation of means and SE and for statistical analysis. The statistical significance of differences between measurements of ischemic and control muscle was evaluated by a two-tailed Student's t-test for unequal variance. P values of ≤0.05 were considered indicative of significance. Curves were fitted to linear and nonlinear data by regression analyses with Sigmastat and PlotIT software.

RESULTS

Chronically Ischemic Muscle of Murine and Human Origin Exhibits Myopathic Features, As Determined by Bright-Field Microscopy

Histological changes in tibialis anterior muscle from chronically ligated mice were determined in reference to the control mice and compared with ischemic and control muscle from PAD and non-PAD patients, respectively. In the mice, ischemic myofibers (Fig. 1B) exhibited a wider range of size, a more rounded shape, centrally located nuclei, and a shift to a smaller cross-sectional area compared with control fibers (Fig. 1A). Muscle of the PAD patient (Fig. 1D), in contrast to control muscle (Fig. 1C), exhibited characteristics similar to those of the ischemic muscle of the mouse (increased size variability, rounded shape, centrally located nuclei, and smaller size).

Fig. 1.

Comparable histopathological changes in ischemic skeletal muscle of a chronically ligated mouse and a patient with arterial disease. Control and ischemic muscle from a sham-operated (A) and a chronically ligated (B) mouse, and a control patient (C) and an age- and sex-matched patient with arterial disease (D), were stained with hematoxylin and eosin and captured under bright-field conditions with a ×10 objective. Myofibers of control muscles (A and C) exhibit peripherally located nuclei, have a regular polygonal shape, and are fairly uniform in size. Myofibers of ischemic muscles (B and D) display more centrally located nuclei and decreased cross-sectional area. These features are consistent with myopathic alterations in the ischemic mouse muscle. The calibration bar at the bottom right of D represents 50 μm.

Mitochondrial Content Is Increased in Skeletal Muscle of Chronically Ligated Compared With Control Mice

ATP synthase and MnSOD expression in slide specimens, by fluorescence microscopy.

Extending our microscopic studies to the mitochondria, we applied fluorescence microscopy to evaluate distribution and abundance of mitochondria within and among myofibers of the tibialis anterior of chronically ligated and control mice. Adjacent serial sections were labeled for expression of ATP synthase and MnSOD. High magnification (×670) of specimens labeled with a monoclonal anti-ATP synthase Ab (Fig. 2, A and B) revealed the tubular microanatomy of the mitochondrial system (19) and subsarcolemmal condensations. At a lower magnification (×260), subsarcolemmal aggregates and a sarcoplasmic dotlike pattern, typical of mitochondrial labeling in skeletal muscle fibers (19, 27), were seen in specimens labeled with anti-ATP synthase Ab (Fig. 2, C and D). However, more intense labeling was seen in more myofibers of ischemic muscle. Centrally located nuclei were apparent as unstained areas within the ischemic fibers. The same pattern of labeling (Fig. 2, E and F) was seen in specimens labeled with a monoclonal Ab specific for MnSOD. In the isotype control slides, myofibers exhibited no fluorescence labeling (data not shown).

Fig. 2.

Representative images illustrating similar labeling patterns and increased expression of ATP synthase and manganese superoxide dismutase (MnSOD) in ischemic skeletal muscle of a chronically ligated mouse. Tibialis anterior muscle from sham-operated (A, C, and E) and chronically ligated (B, D, and F) mice was fixed in methacarn, embedded in paraffin, and sectioned at 4 μm. Control (A and C) and ischemic (B and D) muscles were labeled with specific monoclonal antibodies against ATP synthase. High magnification (×670) with a ×20 objective (A and B) revealed the classic tubular microanatomy of the mitochondrial system and intensely fluorescent nodules associated with mitochondrial labeling. At a lower magnification (×260) with a ×10 objective (C and D), labeling exhibits subsarcolemmal aggregates and a sarcoplasmic dotlike pattern typical of mitochondrial labeling of skeletal myofibers. ATP synthase expression is variable among myofibers of ischemic and control muscle, reflecting differences in mitochondrial content. Overall, ischemic muscle exhibits increased ATP synthase expression compared with control muscle. The same pattern of labeling was seen in adjacent serial sections of control (E) and ischemic (F) muscle labeled with a monoclonal antibody specific for MnSOD. Myofibers treated with isotype control exhibited no fluorescence labeling (not shown). Calibration bars at the bottom right of B and F represent 25 μm. Magnification was calculated by dividing the length of each bar in the image by 25 μm.

Analysis of CS activity in muscle homogenates.

Total CS activity was measured in homogenates of whole calf muscle. CS activity was significantly increased (P < 0.05) in ischemic (N = 10; 752.42 ± 35.03 μmol·ml⁻¹·min⁻¹·mg protein⁻¹) compared with control muscle (N = 12; 466.97 ± 9.38 μmol·ml⁻¹·min⁻¹·mg protein⁻¹). This increased specific activity was not due to the small but significant (P < 0.05) reduction of total protein in ischemic (11.45 ± 0.44 mg) compared with control (12.63 ± 0.35 mg) calf muscle. Total wet weights of ischemic and control muscles were 0.218 ± 0.015 and 0.243 ± 0.007 g, respectively, and were not significantly different.

Mitochondrial Respiration Is Impaired in Muscle Fibers From Ischemic Mice

Compared with control, the respiratory activities of ETC complexes I, III, and IV were significantly reduced in fibers of ischemic muscle, when normalized to CS activity (Table 1). The activities of complexes I, III, and IV were reduced 34, 45, and 42%, respectively. Normalized to myofiber protein, respiratory activity of complex III was significantly reduced, whereas the activities of complexes I and IV exhibited a downward trend (Table 1). These results are consistent with the higher SEs of complex activities normalized to total protein (Table 1). In agreement with our findings from whole calf muscle homogenates, CS in the soleus muscles used for respirometry was significantly (P > 0.05) higher in ischemic (896.28 ± 29.437 μmol·min⁻¹·mg⁻¹) vs. control specimens (757.61 ± 54.460 μmol·min⁻¹·mg⁻¹).

Table 1.

Analysis of mitochondrial complex I- through IV-dependent respiration in ischemic and control soleus muscle of chronically ligated and sham-operated mice

	Control Muscle	Ischemic Muscle	P Value
Complex-dependent respiration normalized to CS activity, nanoatom O2·min−1·unit CS activity−1
Complex I	148.0±16.3	98.1±5.5	<0.05
Complex II	124.4±7.8	100.4±9.6	NS
Complex III	364.9±43.3	202.2±16.2	<0.05
Complex IV	337.8±23.7	196.4±15.0	<0.05
Complex-dependent respiration normalized to total protein, nanoatom O2·min−1·mg protein−1
Complex I	141.1±33.3	95.1±8.5	NS
Complex II	93.2±12.0	97.3±10.7	NS
Complex III	324.2±53.2	192.2±17.4	<0.05
Complex IV	266.2±45.2	205.9±14.6	NS

Values are means ± SE; N = 12 mice. Respiration determined by polarography as nanoatom O₂/min was normalized to both citrate synthase (CS) activity (determined as μmol·ml^{−1·min−1}) and total protein (mg). NS, not significant.

Michael Adducts of HNE and Carbonyl Groups, Indexes of Oxidative Stress, Are Quantitatively Increased in the Ischemic Muscle of Chronically Ligated Mice

Myofibers of ischemic muscle (Fig. 3B) exhibited brighter fluorescence labeling for HNE-protein adducts compared with control muscle (Fig. 3A), indicating greater oxidative stress (28). Isotype controls did not exhibit fluorescence signal. In addition, labeling specificity was demonstrated by 90% inhibition of fluorescence signal (data not shown) when primary Ab was pretreated with 2 × 10⁻⁵ M HNE-histidine conjugate (kind gift of Dr. K. Uchida).

Fig. 3.

Increased 4-hydroxy-2-nonenal (HNE) content in ischemic calf muscle of chronically ligated mice. Myofibers in control (A) and ischemic (B) muscle were labeled with a mouse monoclonal antibody (gift of Dr. Koji Uchida) specific for HNE-protein adducts and Alexa Fluor-conjugated secondary antibody. Images were captured with a Leica fluorescence microscope (×10 objective). HNE signal is markedly increased in myofibers of the ischemic muscle. The bar at the bottom right of B represents 25 μm. C: homogenates prepared with ischemic (1–3) and control (5–7) muscle were analyzed by reverse-phase protein array analysis (RPPA). Homogenate preparations and dilutions (1/20, 1/40, 1/80, and 1/160) of stock HNE-BSA were arrayed in triplicate on nitrocellulose slides and labeled with the monoclonal anti-HNE antibody and IRDye 800 CW conjugated secondary antibody. Integrated intensity of the HNE signal is higher in ischemic muscle of the chronically ligated mice. For both RPPA and fluorescence microscopy, the isotype control produced no fluorescence signal (not shown).

HNE adducts in homogenates of whole calf muscle from control (N = 9) and ischemic (N = 9) mice were measured by RPPA (7). HNE adducts expressed as HNE-BSA units/μg protein were two times more abundant in homogenates of ischemic muscle (5.66 ± 0.43) compared with control muscle (2.77 ± 0.33). This is in agreement with findings from a previous study evaluating muscle from PAD patients (24). Results obtained with homogenates from three ligated and three control mice are presented in Fig. 3C. Arrays treated with nonimmune IgG did not exhibit fluorescence signal. In addition, fluorescence labeling was reduced by 90% (data not shown) when primary Ab was pretreated with the 2 × 10⁻⁵ M HNE-histidine conjugate (kind gift of Dr. K. Uchida).

To confirm the increase in oxidative stress, we quantified carbonyl groups in the same muscle as derivatives of dinitrophenyl hydrazine, by RPPA. In agreement with our measurements of HNE adducts and similar to findings with muscle from PAD patients (24), carbonyls were significantly increased (P < 0.05) in ischemic muscle (14.7 ± 1.10 μg OxBSA/mg protein) compared with control (9.70 ± 0.44 μg OxBSA/mg protein). Arrays treated with nonimmune IgG did not exhibit fluorescence signal. In addition, reduction of homogenates and OxBSA standard with NaBH₄ eliminated the carbonyl signal.

Alteration of Activity and Expression of MnSOD in Ischemic Muscle of the Mouse Hindlimb

Expression of MnSOD increased significantly (P < 0.05) in ischemic compared with control muscle, but activity displayed no significant change (Table 2). The ratio of MnSOD activity to expression was reduced in ischemic muscle to ∼80% of the ratio in control muscle. The divergence of MnSOD activity and expression supports inactivation of the enzyme.

Table 2.

Quantitative analysis of MnSOD in homogenates of ischemic and control calf muscle of chronically ligated and sham-operated mice

	Control Muscle	Ischemic Muscle	P Value
MnSOD activity (standard SOD activity) normalized per mg total protein	1.75+0.14	1.90+0.05	NS
MnSOD expression (ng) normalized per mg total protein	310.3+14.8	412.1+16.0	<0.05

Values are means ± SE; N = 9 mice. Manganese superoxide dismutase (MnSOD) concentration (μg/ml) and MnSOD activity (SOD activity units/ml) were normalized to total homogenate protein (mg).

DISCUSSION

In this study, we have evaluated mice with hindlimb ischemia secondary to inflow arterial occlusion. The skeletal muscle of these animals exhibited typical histological and biochemical features found in human patients with arterial occlusive disease. Development in ischemic mice, of a myopathy associated with proliferation of dysfunctional mitochondria, oxidative stress, and alterations in MnSOD expression and activity provides the first direct demonstration that arterial occlusive disease produces muscle and mitochondrial pathology.

Myopathic Histological Features in Chronically Ischemic Muscle of the Mouse Hindlimb

Our work demonstrates that mouse skeletal muscle responds to chronic inflow arterial occlusion with the development of appreciable myopathic histological changes, a drop in total protein content, and a trend toward decreased wet weight. These findings are similar to the significant myopathic and atrophic changes that have been well described in the ischemic muscle of PAD patients (22, 23). These changes in human PAD muscle are thought to be related to and responsible for the sarcopenia, loss of strength, and functional impairment that is so characteristic of this patient population (5, 22, 23).

Increased Mitochondrial Content

Our work further demonstrates that arterial occlusive disease generates a significant increase in the mitochondrial content of skeletal muscle (documented both by elevated CS activity and increased ATP synthase expression). Increased mitochondrial expression reflected by CS activity and mitochondrial DNA levels has been documented previously for claudicating PAD patients (4, 5). Furthermore, mitochondrial proliferation is characteristic of mitochondrial diseases and aging (33) and has been described for mice with genetically induced (adenine nucleotide translocator 1, creatine kinase, cytochrome-c oxidase and transcription factor A knockouts, and Twinkle and lipoprotein lipase transgenics) mitochondriopathy (15, 21, 34). Increased mitochondrial content in our mice and in PAD patients suggests that bioenergetic deficiency (as seen by defective mitochondrial respiration) and compensatory proliferation of defective mitochondria are central processes in the pathophysiology of skeletal muscle ischemia (34). In skeletal muscle, an upregulation of mitochondrial biogenesis may be associated with, or expressed in the form of, an alteration of muscle fiber type toward the more oxidative type I and IIa fibers. Interestingly, such a change toward slow myofiber type in PAD patients has been described by several groups (31). The mechanism by which nuclear and mitochondrial genomes are coordinately induced to promote mitochondrial proliferation and the extent to which this proliferation is part of a myofiber type shift was not explored in this paper, but deserves consideration in the future.

Impaired Respiration in Ischemic Muscle Mitochondria

Although increased, the mitochondrial population of ischemic skeletal muscle appears to have significantly compromised respiration (oxidative phosphorylation). Our results demonstrate significant defects in integrated mitochondrial function. We found that complex I-, III-, and IV-dependent oxidative phosphorylation is significantly decreased in ischemic muscle compared with control. These findings are in agreement with previous work from our group and others demonstrating interrelated defective ETC complexes, compromised mitochondrial respiration, and impaired oxidative energy production in the skeletal muscle of PAD patients (24, 25). Our mouse data are unique because they demonstrate that arterial occlusive disease alone, in the absence of other comorbid conditions frequently seen in PAD patients (such as diabetes, smoking, heart/lung disease, etc.), is able to produce significant mitochondrial pathology.

Increased Oxidative Stress and Compromised Mitochondrial Antioxidant Defense

As shown in the present work, chronically ischemic mouse skeletal muscle sustains substantial oxidative injuries indicated by a consistent increase in protein carbonylation and lipid peroxidation adducts. The observed increase in carbonyl and HNE levels can be related to our findings of concurrent mitochondrial dysfunction and possible inactivation of the MnSOD antioxidant defense system. Under resting conditions, a large proportion of cellular reactive oxygen species (ROS) is produced in the mitochondria primarily by ETC complexes I and III (33). Damage to the mitochondria, especially at these complexes, exponentially enhances the production of ROS (33). Our data show potential defects in both complexes I and III, suggesting ischemic muscle may generate significant amounts of ROS, which, in the presence of insufficient MnSOD, are amplified. Our data demonstrate a discrepancy between MnSOD expression and activity, which is consistent with inactivation of MnSOD (the initial line of ROS defense in mitochondria). A similar discrepancy has been described for chronic kidney transplant rejection, a condition characterized by increased ROS. In that system, MnSOD expression was increased, but enzymatic activity was decreased secondary to nitration of tyrosine residues in the active site (16). Dysfunctional ETC complexes, along with diminished MnSOD activity, may be key contributors to a destructive cycle, leading to progressively deteriorating ETC complexes and oxidative stress.

Inflow Arterial Occlusion in the Mouse Model PAD Pathophysiology

Exercise-induced ischemia-reperfusion is the central problem in PAD patients and in animals with inflow arterial occlusion (2, 6, 29). The ischemia-reperfusion of every bout of claudication increases oxidative stress, triggers inflammation and oxidative damage to the tissues (4, 5, 12, 13, 22, 23), and initiates mitochondrial injury and dysfunction (4, 5, 22, 23). Mitochondrial dysfunction can then be perpetuated by repeated destructive cycles of ischemia-reperfusion, causing amplification of respiratory chain defects, compromised bioenergetics, increased ROS production, diminished MnSOD antioxidant activity, and oxidative stress (5, 17). In this process, mitochondrial biogenesis may be induced as the ischemic muscle attempts to compensate for its compromised bioenergetic state. The combination of compromised bioenergetics and worsening oxidative stress may then lead to progressive oxidative damage of structures in the myocytes, eventually producing the myopathic histological changes of chronically ischemic muscle.

Perspectives and Significance

Our data have significant implications for modeling mitochondrial dysfunction and oxidative stress in the pathophysiology of PAD. We have evaluated mice with hindlimb ischemia secondary to inflow arterial occlusion. These mice reproduce important histopathological and biochemical features found in patients with arterial occlusive disease. These features include myopathic histological alterations, increased mitochondrial content, compromised mitochondrial respiration with dysfunctional ETC complexes I, III, and IV, increased oxidative stress, and alterations in MnSOD expression and activity. Our findings indicate that chronic arterial occlusion is the underlying mechanism for the myopathy described in patients with PAD and that the key factors in this myopathy are mitochondrial dysfunction and oxidative stress. Future animal model work will allow further delineation of the mechanisms responsible for the production of this myopathy. Such research is of central importance in delineating the pathobiology of claudication and will support the development of novel therapies for patients with arterial disease.

GRANTS

This work was funded by grants from the National Heart, Lung, and Blood Institute (1 K08 HL079967–01) and the Lifeline Programs of the American Vascular Association (William J. von Liebig Award) to I. I. Pipinos.