Ovarian hormone status and skeletal muscle inflammation during recovery from disuse in rats
Abstract
Resumption of normal muscle loading after a period of disuse initiates cellular processes related to mass accretion. The renewed loading also induces a significant amount of muscle damage and subsequent inflammation. Ovarian hormone depletion delays atrophied myofibre mass recovery. Ovarian hormones are also global regulators of immune system function. The purpose of this study was to determine whether ovarian hormone depletion-induced deficits in myofibre regrowth after disuse atrophy are related to the induction of muscle damage and the associated inflammatory response. We hypothesized that soleus muscle immune cell infiltration and inflammatory gene expression would be both accentuated and prolonged in ovarian hormone-depleted rats during the first week of recovery from disuse atrophy. Intact and ovariectomized (OVX) female rats were subjected to hindlimb suspension for 10 days and then returned to normal ambulation for a recovery period, the rats were killed and the soleus muscle removed for analysis. Although reloading increased both circulating creatine kinase and myofibre membrane disruption, there was no effect of ovarian hormones on these processes during recovery. Muscle neutrophil concentration was increased above baseline regardless of hormone status at days 1 and 3 of recovery; however, this increase was 43% greater at day 3 in the OVX group. Muscle ED1+ and ED2+ macrophage concentrations were increased during recovery in both groups. However, macropage concentrations remained elevated at day 7 of recovery in the OVX group, whereas they returned to control levels in the intact group. Cyclo-oxygenase-2, interleukin-6 and interleukin-1β muscle mRNA expression increased similarly during recovery, regardless of ovarian hormone status. These results demonstrate that the initial myofibre damage and inflammatory gene expression induced during muscle recovery from disuse atrophy are independent of ovarian hormone status.
Muscle mass recovery from disuse atrophy appears to be more complicated than simple activation of muscle growth programs, since atrophic processes must also be repressed or reversed (Mitchell & Pavlath, 2001). Additionally, resumption of normal activity after a period of disuse induces rapid and extensive damage that is characterized by creatine kinase (CK) release into the serum, membrane constituent protein loss, and the appearance of extracellular matrix-localized proteins in the interior of the myofibre (Tidball, 2002). During the complicated process of muscle mass recovery, a variety of stimuli must be integrated to provide a cellular response that leads to successful regeneration and growth. These stimuli include circulating growth factor abundance, circulating hormone abundance, and inflammatory processes related to damage and injury.
The contribution of cells of the immune system to muscle recovery is believed to be linked to their roles in removal of necrotic tissue and localized production of cytokines and chemokines critical for regeneration and growth (Tidball, 2005). Leukocyte infiltration occurs in response to muscle injury and fibre degeneration and is generally characterized by increased phagocytic (neutrophil and ED1+ macrophage) and non-phagocytic cell types (ED2+ macrophage; St Pierre & Tidball, 1994a,b; Pizza et al. 1998; Frenette et al. 2000, 2002). Neutrophils rapidly infiltrate skeletal muscle during reloading, at about the same time as myofibre structural disruption (Fielding et al. 1993). High concentrations of neutrophils are critical for removal of cellular debris and proteolysis associated with myofibre damage and degeneration, as well as for the production of myofibre-damaging molecules (Tidball, 2005). Proteolytic ED1+ macrophages are capable of phagocytosing apoptic neutrophils and function in the removal of damaged tissue debris (Savill et al. 1989; Tidball, 2005). Mast cells are a population of bone marrow-derived inflammatory cells that are important in wound recovery and serve as an intermediate infiltrator of damaged skeletal muscle between the time of peak neutrophil and macrophage infiltration (Gorospe et al. 1996). ED2+ macrophages function in muscle repair and regeneration by stimulating myoblast proliferation (St Pierre & Tidball, 1994a; Massimino et al. 1997; Tidball, 2005).
Although macrophages are an important source of inflammatory cytokines, both immune and non-immune cells are capable of producing cytokines involved in inflammation during skeletal muscle recovery (St Pierre Schneider et al. 1999). Muscle inflammatory cytokines include interleukin-1 (IL-1β), IL-2, IL-6 and tumour necrosis factor-α (TNF-α; Imura et al. 1996). Interleukin-1β expression increases at the onset of the immune response phase of skeletal muscle repair. Interleukin-1β is responsible for stimulating proteolytic enzyme synthesis, leucocyte adhesion and extravasion, macrophage activation, and the expression of other muscle related cytokines such as IL-2, IL-3, IL-6 and TNF-α (Tidball, 1995). Interleukin-6 is a pro-inflammatory cytokine reported to provide an important early stimulus that modulates satellite cell activation and impacts protein synthesis during skeletal muscle recovery (Cantini et al. 1995). Cytokines are also capable of inducing prostaglandin synthesis, which both mediate inflammatory processes and induce satellite cell activation (Bondesen et al. 2004). Cyclo-oxygenase-2 (COX-2) is a key prostaglandin-synthesizing enzyme critical for successful and timely skeletal muscle regeneration (Bondesen et al. 2004). Temporally, the induction and suppression of cytokines, prostaglandins and growth factors appears to be critical for the timely recovery of skeletal muscle mass from disuse atrophy.
It has been clearly demonstrated that ovarian hormone loss delays rat skeletal muscle mass recovery from disuse-induced muscle atrophy (Brown et al. 2005; Sitnick et al. 2006; McClung et al. 2006). Specifically, oestrogen appears to be necessary for timely recovery of muscle mass following disuse atrophy in female rats (McClung et al. 2006). Ovarian hormone effects on muscle damage and recovery have historically focused on myofibre membrane stabilization and/or altered inflammatory response, and have often been examined using gender comparisons (Stupka et al. 2000; Tiidus, 2000). The ovarian hormone oestrogen possesses a lipophilic structure and fat-soluble nature that give it membrane stabilization properties (Kendall & Eston, 2002) through binding to the phospholipid bilayer and potentially altering membrane fluidity. In addition, neutrophil and macrophage concentrations and inflammatory cytokine expression are ovarian hormone-sensitive components of inflammation (Zuckerman et al. 1996; Angstwurm et al. 1997; Komulainen et al. 1999; Tiidus et al. 2001, 2005). The regulation of immune cell infiltration and inflammatory cytokine expression by ovarian hormones during recovery from disuse-induced atrophy has not yet been determined. The purpose of this study was to determine whether damage and inflammatory processes activated during the initial period of loading after hindlimb muscle disuse are regulated by ovarian hormone loss. Our working hypothesis was that ovarian hormone depletion would prolong and accentuate immune cell infiltration and inflammatory gene expression in female rat skeletal muscle recovering from disuse atrophy.
Methods
Animals and housing
Intact and ovariectomized (OVX) female Sprague–Dawley rats (body weight, ∼200 g) were acquired from Harlan rodent colony (Indianapolis, IN, USA). Previously, we determined that the 14 days of reloading after disuse was the critical ovarian hormone-sensitive period of muscle recovery (McClung et al. 2006). To determine whether ovarian hormone depletion alters the initial myofibre disruption and subsequent inflammatory response of skeletal muscle reloaded after disuse atrophy, intact and OVX females were randomly divided into five separate treatment groups (n= 6 per group): (1) ground control (Con); (2) 10 days of hindlimb suspension (Sus); (3) 10 days of hindlimb suspension with 1 day of reloading (Sus+1); (4) 10 days of hindlimb suspension with 3 days of reloading (Sus+3); and (5) 10 days of hindlimb suspension with 7 days of reloading (Sus+7; Fig. 1). Animals were approximately 8 weeks old at the time of ovariectomy and were approximately 12 weeks old at the onset of treatments. Animals were housed individually and kept on a 12 h–12 h light–dark cycle and given ad libitum access to normal rodent chow (intact female treatment groups) or to phyto-oestrogen-free rodent chow (Purina Test Diet, Richmond, IN, USA; OVX treatment groups) and water for the duration of the study at the fully accredited animal care facilities at the University of South Carolina, Columbia. Intact and OVX ground control and the 10 day hindlimb suspension treatment group muscle weights have been previously described (McClung et al. 2006). At the conclusion of the treatment period, animals were anesthetized with an intramuscular injection of a cocktail containing ketamine hydrochloride (75 mg/kg b.w.), xylazine (3 mg/kg b.w.) and acepromazine (5 mg/kg b.w.), and the postural soleus muscle was frozen in liquid nitrogen for analyses. Daily vaginal smears (over a period of 5 days) were obtained from all animals at the onset of the study to verify oestrous cycling in intact females and the lack of oestrous cycling in OVX animals. Circulating levels of the ovarian hormone oestrogen were examined in plasma obtained via venous blood draw at the time of killing. All procedures were approved by the University of South Carolina Animal Care and Use Committee.
Study design Intact and ovariectomized (OVX) females were subjected to hindlimb suspension and to 1, 3 or 7 days of reloading recovery (see Methods).
Hindlimb suspension-induced disuse and reloading recovery
Hindlimb suspension-induced disuse and recovery was performed as previously described (McClung et al. 2006), after the initial week of animal monitoring. Briefly, unanesthetized animals' tails were cleaned with rubbing alcohol and air dried, covered with a light coat of benzoin tincture, and dried with a hair dryer until tacky. Strips of elastoplast (Biersdorf, Norwalk, CT) adhesive bandage were applied to the proximal two-thirds of all sides of the tail and looped through a swivel attachment mounted above the cage, and designed to allow the animal to move rotationaly 360° with only the forelimbs able to come into contact with the cage floor.
Plasma oestradiol analysis
Plasma oestradiol was extracted from 500 μl of plasma using 4 ml of 100% ethanol, and measured using a commercially available double-antibody radio-immunoassay kit (Diagnostic Products, Los Angeles, CA, USA) as previously described (McClung et al. 2006). The intra- and interassay coefficients of variation were 6.2 and 14.3%, respectively.
Circulating creatine kinase
Serum was assayed spectrophotometrically (wavelength, 340 nm) for creatine kinase activity using a commercially available kit (DG147-UV, Sigma Diagnostics, Columbus, OH, USA) according to the manufacturer's instructions.
Muscle dry weights and water content
The soleus muscle was analysed for dry muscle mass and water content as previously described (McClung et al. 2006). Briefly, ∼30 mg of total soleus muscle wet mass was weighed, subjected to a drying protocol of 70°C, and weighed every 30 min until a stable dry weight was obtained.
Immunohistochemical analysis
Soleus muscle sections (10 μm) were processed for immunohistochemical analysis of inflammatory cell infiltration as previously described (Frenette et al. 2002). Briefly, muscle sections were air-dried, fixed in acetone for 10 min, and air-dried for an additional 30 min. Endogenous peroxidase activity was quenched by incubation in 0.6% peroxide in methanol for 60 min. Sections were then washed in phosphate buffered saline (1 × PBS) and non-specific binding blocked with 4% horse serum for 1.5 h. The following primary antibodies were then incubated on the sections overnight at 4°C: (1) single chain cell surface glycoprotein antigen, 110 kD (anti-ED1+) macrophage subpopulation (1:50; MCA341R, Serotec); (2) cell surface glycoprotein antigen, 175D (anti-ED2+) macrophage subpopulation (1:50; MCA342R, Serotec, Raleigh, NC); and (3) anti-His 48 granulocytes (1:50; MCA54R, Serotec). After a brief wash in 1 × PBS, antimouse secondary antibody (1:200; Vector Laboratories, Burlingame, CA, USA) was then incubated with all sections (control and treatment) for 90 min at room temperature. Sections were washed in 1 × PBS and incubated with the Vectastain ABC reagent (Vector Laboratories) according to the manufacturer's instructions. Sections were then exposed to 3,3′-diaminobenzidine diaminobenzidine tetrahydrochloride (DAB) with nickel (Vector Laboratories) for 10 min. This reaction was stopped by immersion in tap water, and sections were mounted in Permount (Fisher, Pittsburgh, PA, USA). Four different digital images were taken from sections labelled for each individual inflammatory cell and analysed for cell concentration by a blinded investigator. To reduce experimental bias, all immunostained nuclei present on digital images were quantified. The number of labelled cells in each section was counted, and the total area of the section was determined and multiplied by its thickness to express the number of each cell type per cubic millimetre as previously described (Frenette et al. 2002).
Myofibre membrane disruption and myofibre necrosis
Soleus muscle sections (10 μm thick) from the midbelly of the muscle were processed for analysis of myofibre membrane disruption and myofibre necrosis by double immunohistochemical staining for the membrane protein dystrophin (NCL-DYS2; Novacastra, Norwell, MA, USA) and ED1+ macrophage subpopulation. Sections were incubated overnight at 4°C in primary dystrophin antibody. The slides were then incubated in the biotinylated secondary antibody for 90 min at room temperature and in avidin–biotinylated peroxidase for 60 min at room termperature. The final reaction involved incubating the sections with 0.01% hydrogen peroxide and 0.05% DAB. Sections were then incubated overnight at 4°C in primary ED1+ macrophage antibody and subsequently incubated in biotinylated secondary antibody and in avidin–biotinylated peroxidase at room temperature. The final reaction was performed by incubating the sections in DAB. Four different digital images from sections labelled for both dystrophin and ED1+ macrophage cells were analysed by a blinded investigator as previously described (Komulainen et al. 1999; Frenette et al. 2000). Briefly, approximately 300 individual fibres were quantified for both dystrophin disruption and necrosis, which by examination of no additional change in standard deviation determined an appropriate sample size. Myofibre membrane disruption and invasion by ED1+ macrophages were expressed as the percentage of fibres with disrupted dystrophin staining (>50% of fibre membrane area) or macrophage infiltration (percentage of fibres with ED1+ cells inside membrane). Myofibre membrane disruption and invasion by ED1+ macrophages were quantified by light microscopy by a blinded investigator as previously described (Komulainen et al. 1999; Frenette et al. 2000).
Morphological analysis of mast cells
The infiltration of mast cells into skeletal muscle during reloading was analysed as previously described (Nahirney et al. 1997), with the following modifications. Briefly, cryosections (10 μm) were taken from the midbelly of the soleus muscle and stained with 1.0% aqueous Toluidine Blue for 15 s at 60°C. Sections were then counterstained with alcohol–Eosin. Four different digital images at ×20 magnification were obtained from sections labelled for mast cells and analysed for cell concentration by a blinded investigator. To reduce experimental bias, all stained nuclei present on digital images were quantified. The number of labelled cells in each section was counted, and the total area of the section was determined and multiplied by its thickness to express the number of mast cells per cubic millimetre as previously described (Frenette et al. 2002). Digital images at ×20 magnification were taken and analysed for cell concentration by a blinded investigator. The number of mast cells was expressed per cubic millimetre.
Total RNA isolation and cDNA synthesis
Total RNA was isolated using TRIzol reagent (Life Technologies, Grand Island, NY, USA) according to the manufacturer's instructions. Complementary DNA (cDNA) was reverse transcribed from 4 g of total RNA using 1 μl of random hexamers and 50 U of Superscript II RT (Invitrogen, Carlsbad, CA, USA) in a final volume of 20 μl at 25°C for 10 min, followed by 42°C for 50 min and 70°C for 15 min.
Real-time reverse transcriptase-polymerase chain reaction (RT-PCR)
Real-time RT-PCR was performed in a 25 μl reaction containing 1.0 μl of cDNA, 75 nm of primer, 12.5 μl of Taq-Man master mix buffer, and sterile water. Amplification was performed with a thermal cycler (Sequence Detection Systems, model 7300; Applied Biosystems, Foster City, CA, USA) with an initial cycle of 50°C for 10 min, followed by 40 cycles of 95°C for 15 s and 60°C for 45 s. Fluorescence labelled probes for COX-2 (FAM dye) and the ribosomal RNA 18s (VIC dye) were purchased from Applied Biosystems. Interleukin-1β and IL-6 primer sequences were synthesized by Integrated DNA Technologies (IDT) (Caralville, IA, USA), and SYBR Green (Applied Biosystems, Foster City, CA, USA)was used for quantification. The primer sequences for COX-2 were: forward, 5′-ACACTCTATCACTGGCATCC-3′; reverse, 5′-GAAGGGACACCCTTTCACAT-3′. The primer sequences for IL-6 were: forward, 5′-CTTGGGACTGATGTTGTTGAC-3′; reverse, 5′-TCTGAATGACTCTGGCTTTGT-3′. The primer sequences for IL-1β were: forward, 5′-CCTTCTTTTCCTTCATCTTTG-3′; reverse, 5′-ACCGCTTTTCCATCTTCTTCT-3′. Cyclo-oxygenase-2, IL-6 and IL-1β cycle threshold (Ct) was determined, and the ΔCt value calculated as the difference between their Ct value and the 18s Ct value. Final quantification of cytokine and COX-2 gene expression was calculated using the ΔΔCT method {ΔΔCt =[ΔCt(calibrator) −ΔCt(sample)]. Relative quantification was then calculated as 2−ΔΔCt.
Data analysis
Variables for the analysis of differences within treatment control groups due to recovery day (Sus, Sus+1, Sus+3 or Sus+7) were analysed by one-way ANOVA. Where no significant effects occurred in treatment controls due to day, respective groups were pooled. All other variables were analysed using two-way ANOVA for main effects (ovariectomy or recovery day) or interactions (ovariectomy × recovery day). Where significant interactions existed, Bonferroni post hoc analyses were used between groups. Values are presented as means ±s.e.m. Significance was set at P < 0.05.
Results
Circulating oestradiol
Circulating oestradiol measurements were performed to verify the success of the ovariectomy procedure. Ovariectomy decreased circulating oestradiol by 86% (4 ± 2 pg ml−1) from intact control values (29 ± 12 pg ml−1).
Analysis of day of controls
There were no observed effects of within treatment day control groups due to recovery day on any morphological or biochemical measurement analysed in the present study; therefore, data within treatments were combined for all further analysis and discussion.
Muscle wet weights, wet weight/tibia length and water content
Owing to increased water content in soleus muscle from OVX rats (McClung et al. 2006), muscles in the present study were analysed for wet weight (mg), wet weight corrected for tibia length (mg mm−1), water content (mg) and muscle dry weight (mg) to allow accurate description of alterations in weight. There was an interaction (P= 0.004) of ovariectomy and recovery day on muscle wet weight (Table 1), muscle wet weight corrected for tibia length (P= 0.004) and muscle water content (P= 0.003). Muscle wet weight, weight corrected for tibia length and muscle water content were increased by ovariectomy. Suspension decreased wet weight by 27–44%, weight corrected for tibia length by 22–43% and muscle water content by 21–44%, regardless of ovarian hormone status. Wet weight was restored by day 7, and weight corrected for tibia length and muscle water content were restored by day 3 of recovery in soleus muscle from intact, but not OVX rats. There was no interaction of ovariectomy and recovery day on muscle dry weight (Table 1).
| Treatment | Muscle wet wt (mg) | Muscle wet wt/Tibia length (mg mm−1) | Muscle dry wt (mg) | Muscle water content (mg) |
|---|---|---|---|---|
| Intact | ||||
| Con | 102 ± 5 | 2.7 ± 0.1 | 25 ± 4 | 75 ± 6 |
| Sus | 75 ± 3* | 2.1 ± 0.1* | 15 ± 1 | 59 ± 2* |
| Sus+1 | 72 ± 2* | 2.1 ± 0.2* | 16 ± 3 | 58 ± 4* |
| Sus+3 | 87 ± 4*†‡ | 2.5 ± 0.1†‡ | 15 ± 4 | 71 ± 3†‡ |
| Sus+7 | 96 ± 6†‡ | 2.8 ± 0.2†‡ | 18 ± 4 | 75 ± 4†‡ |
| OVX | ||||
| Con | 131 ± 2# | 3.7 ± 0.1# | 27 ± 1 | 104 ± 1# |
| Sus | 74 ± 3* | 2.1 ± 0.1* | 16 ± 2 | 58 ± 2* |
| Sus+1 | 79 ± 4* | 2.3 ± 0.1* | 15 ± 5 | 62 ± 3* |
| Sus+3 | 86 ± 7* | 2.5 ± 0.2*† | 16 ± 3 | 65 ± 4* |
| Sus+7 | 89 ± 4*†‡ | 2.5 ± 0.1*† | 16 ± 4 | 70 ± 3* |
- Muscle wet weight (mg), wet weight corrected for tibia length (mg mm−1), dry weight (mg) and muscle water content (mg) in soleus muscle from intact and ovariectomized (OVX) female rats in ground control (Con), hindlimb suspended (Sus), 1 day recovery (Sus+1), 3 day recovery (Sus+3) and 7 day recovery (Sus+7) groups. Values are presented as means ±s.e.m.* Significantly different (P < 0.05) from treatment-matched Con. † Significantly different (P < 0.05) from treatment-matched Sus. ‡ Significantly different (P < 0.05) from treatment-matched Sus+3. # Significantly different (P < 0.05) from intact Con. There was a main effect of recovery day on muscle dry weight.
Myofibre membrane disruption
To determine whether ovariectomy alters membrane disruption during recovery from disuse, we analysed dystrophin protein and circulating creatine kinase (CK). Dystrophin protein disruption at the plasma membrane is a direct marker of disrupted myofibre membranes in skeletal muscle (Tidball, 2002; Lovering & De Deyne, 2004). Lesions in myofibres result in the release of CK into the serum (Tidball, 2002). Dystrophin disruption (as % of fibres) at the skeletal muscle membrane was quantified by immunohistochemistry (Fig. 2A). There were significant effects (P < 0.0001) of recovery day on dystrophin disruption (Fig. 2B) and circulating CK (Fig. 2C). Both dystrophin disruption and CK were increased at day 1 of reloading and returned to control values at day 7 of recovery. There was no interaction of ovariectomy and recovery day on dystrophin disruption or circulating CK. There was an effect of ovariectomy (P= 0.01) on circulating CK (in units l−1). Compared with intact females, CK increased by 42% with following ovariectomy (Fig. 2C).
Myofibre membrane disruption in skeletal muscle reloaded after disuse atrophy A, representative images of OVX control (Con) and hindlimb suspension with 3 days of reloading (Sus+3). Arrow indicates representative myofibre displaying dystrophin disruption. B, there were main effects of both recovery day (ɛ) and ovariectomy (γ) on circulating creatine kinase concentration (in U L−1). Ground control (Con), hindlimb suspended (Sus), suspension with 1 day of recovery (Sus+1), suspension with 3 days of recovery (Sus+3), and suspension with 7 days of recovery (Sus+7) circulating creatine kinase in soleus muscle from intact (black bars) or OVX rats (grey bars). Compared with intact females, CK was increased following ovariectomy across all groups (γ). Circulating creatine kinase was also increased at day 1 of reloading and returned to control values at day 3 of recovery across both treatments (ɛ). C, there was a main effect of recovery day (ɛ) on the percentage of fibres displaying dystrophin disruption. Dystrophin disruption was increased at days 1 and 3 of reloading and returned to control values at day 7 of recovery across both treatments (ɛ). All values are expressed as means ±s.e.m.
Leukocyte concentrations
Neutrophil and macrophage concentrations increase in skeletal muscle during reloading after disuse atrophy (Frenette et al. 2002). There were interactions of ovariectomy and recovery day on neutrophil (P < 0.0001; neutrophils (mm muscle)−3), phagocytic ED1+ macrophages (P < 0.0001; ED1+ (mm muscle)−3) and non-phagocytic ED2+ macrophages (P < 0.0001; ED2+ (mm muscle)−3; Fig. 3). Neutrophil concentration increased similarly at day 1 of recovery, regardless of ovarian hormone status. At day of recovery, however, the muscle neutrophil concentration was increased by 43% from intact values by ovariectomy (Fig. 4A). The concentrations of phagocytic ED1+ macrophages increased similarly by 511 and 593% at days 1 and 3 of recovery, respectively, regardless of ovarian hormone status (Fig. 4B). However, in OVX muscle, ED1+ macrophages remained elevated by 558% at day 7 of recovery. In intact muscle, the concentration of non-phagocytic ED2+ macrophages increased at day 3 of recovery (Fig. 4C). In OVX muscle, however, ED2+ macrophage concentration increased by 111% at day 1 and remained elevated by 64 and 59% at days 3 and 7 of recovery, respectively.
Representative immunohistochemical images of ovariectomized ground control (Con) and hindlimb suspension with 3 days of reloading (Sus+3) ED1+ and ED2+ macrophages, as well as His 48 (neutrophil) staining
Infiltration of leucocytes into soleus skeletal muscle during reloading after disuse atrophy There were significant interactions of ovariectomy and recovery day on neutrophil, ED1+ and ED2+ macrophage concentrations in soleus muscle. A, neutrophil concentration in soleus muscle of intact (black bars) and OVX female rats (grey bars). B, ED1+ macrophage infiltration during reloading. C, ED2+ macrophage infiltration during reloading. Values are expressed as means ±s.e.m.* Significantly different (P < 0.05) from treatmentmatched Con. † Significantly different (P < 0.05) from treatment-matched Sus. ‡ Significantly different (P < 0.05) from treatment-matched Sus+1. ɛ Significantly different (P < 0.05) from treatmentmatched Sus +7.
ED1+ macrophage localization inside the myofibre membrane in skeletal muscle is an indicator of myofibre necrosis (Frenette et al. 2000). Ovariectomy (P= 0.006) and recovery day (P < 0.0001) independently altered myofibre necrosis (Fig. 5). Ovariectomy increased myofibre necrosis by 51% (P= 0.008). In addition, myofibre necrosis was increased at days 1, 3 and 7 of recovery.
Percentage of myofibres with macrophage infiltration inside the sarcolemma There were main effects of both recovery day (ɛ) and OVX (γ) on the percentage myofibres with macrophage infiltration. Compared with intact females, ovariectomy increased myofibre necrosis across all groups (γ). Myofibre necrosis was also increased at days 1, 3, and 7 of recovery across both treatments (ɛ). All values are expressed as means ±s.e.m.
Mast cell infiltration
Mast cells are present in skeletal muscle connective tissue interstitium and are believed to be critical in triggering and sustaining inflammation and immune responses (Nahirney et al. 1997). Ovariectomy (P= 0.05) and recovery day (P < 0.0001) independently altered mast cell concentrations (mast cells (mm muscle)−3) in the soleus muscle (Fig. 6A). Mast cell concentration increased by 20% (P < 0.008) following ovariectomy. Also, mast cell concentration was increased at days 1 and 3 of recovery but returned to control values by day 7 (Fig. 6B).
Infiltration of mast cells into reloaded skeletal muscle A, representative image of Toluidine Blue staining for mature granulated skeletal muscle mast cells in soleus muscle from OVX ground control (Con) and OVX hindlimb suspension with 3 days of reloading (Sus+3). B, there were main effects of both recovery day (ɛ) and OVX (γ) on mast cell concentration. Compared with intact females, mast cell concentration increased with OVX across all groups (γ). Mast cell concentration was also increased at days 1 and 3 of reloading and returned to control values at day 7 of recovery across both treatments (ɛ). All values are expressed as means ±s.e.m.
Cyclo-oxygenase-2 mRNA abundance
Cyclo-oxygenase-2 (COX-2) synthesis of prostaglandins is critical for early processes associated with skeletal muscle regeneration after injury (Bondesen et al. 2004). Recovery day alone significantly (P < 0.0001) altered COX-2 mRNA expression (Fig. 7). Cyclo-oxygenase-2 mRNA increased by 478% at day 1 and by 1812% at day 3 of recovery in the soleus, returning to control values by day 7 of recovery.
Cyclo-oxegenase-2 (COX-2) mRNA abundance in recovering soleus skeletal muscle Real-Time RT-PCR was performed to determine COX-2 mRNA abundance. There was a main effect of recovery day (ɛ) on COX-2 mRNA abundance in both intact and OVX rats. Cyclo-oxegenase-2 mRNA was increased at days 1 and 3 of recovery in the soleus across both treatments (ɛ). All values are expressed as means ±s.e.m.
Abundance of inflammatory cytokine mRNA
Inflammatory cytokines are critical mediators of muscle regeneration and growth after damaging stimuli (Tidball, 2005). Recovery day alone significantly altered IL-6 (P= 0.004) and IL-1β (P < 0.0001) mRNA abundances. Interleukin-6 mRNA abundance increased by 138% at day 1 of recovery and by 291% at day 3 of recovery, but returned to control values by day 7 of recovery in the soleus (Fig. 8A). Interleukin-1β mRNA abundance increased by 1629% at day 1 of recovery, by 770% at day 3 of recovery and by 455% at day 7 of recovery in soleus muscle (Fig. 8B).
Inflammatory cytokine mRNA abundance in reloaded soleus skeletal muscle Real-Time RT-PCR was performed to determine IL-6 and IL-1β mRNA abundance. A, there was a main effect of recovery day (ɛ) on IL-6 mRNA abundance in both intact and OVX rats. Interleukin-6 mRNA abundance increased at days 1 and 3 of recovery, but returned to control values by day 7 of recovery in both intact and OVX rats (ɛ). B, there was a main effect of recovery day (ɛ) on IL-1β mRNA abundance. Interleukin-1β mRNA abundance increased at days 1, 3 and 7 of recovery in both intact and OVX rats (ɛ). All values are expressed as means ±s.e.m.
Discussion
Ovarian hormone depletion prevents the recovery of soleus muscle fibre cross-sectional area from disuse atrophy during 2 weeks of normal cage ambulation, while oestrogen replacement allows for the restoration of myofibre size (McClung et al. 2006). The present study expands on these findings by demonstrating that ovarian hormones exert critical regulation related to recovery of myofibre size during the initial period of recovery from disuse atrophy. However, we report that the processes related to ovarian hormone loss and muscle recovery are independent of muscle damage. A more promising mechanism appears to be related to the biological response of skeletal muscle to damage. Ovariectomized animals demonstrated elevated soleus muscle neutrophil and macrophage concentrations during the initial week of recovery. However, the induction of inflammatory gene expression was not sensitive to ovarian hormone status during the first week of recovery.
The temporal co-ordination of processes occurring during the initial days of reloading after disuse is critical for the successful recovery of muscle mass and myofibre size. Skeletal muscle subjected to loading after an extended period of disuse undergoes rapid and extensive damage (Tidball, 2002). Ovarian hormones have been hypothesized to function in membrane stabilization and defense against free radicals; effects which could potentially attenuate myofibre damage (Tiidus et al. 1993; Kendall & Eston, 2002). In the present study, both direct (dystrophin disruption) and indirect (circulating creatine kinase) markers of myofibre membrane disruption increased early during the recovery process and dissipated over the latter portion of the first week. The peak in CK efflux from skeletal muscle did not coincide with dystrophin disruption. Histological markers of muscle damage often display a different time course of induction from circulating variables (Van der Meulen et al. 1991). Ovariectomy did not alter myofibre dystrophin disruption or circulating CK during recovery. Ovariectomy did alter baseline circulating creatine kinase, similar to previously reported findings (Tiidus et al. 2001). These facts suggest that during reloading after disuse, the status of ovarian hormones is not sufficient to alter the pattern of myofibre damage that occurs.
Ovarian hormones may therefore function in other capacities to mediate recovery processes in response to myofibre damage, including secondary damage or the subsequent immune response. Cyclo-oxygenase-2 gene expression temporally increases in damaged skeletal muscle (Bondesen et al. 2004). Pro-inflammatory cytokine and COX-2 mRNA abundances increase and subsequently decrease by day 7 of recovery in reloaded female rats. On the whole, our data support the concept that immune and inflammatory processes are stimulated and decline within the initial week of recovery. Overall soleus weight recovery occurs by the 7th day of reloading, but is primarily characterized by sustained oedema. Myofibre cross-sectional area and myofibrillar protein content, indicators of myofibre size, are not restored until the 14th day of recovery (McClung et al. 2006). Disruption of inflammation during the first 7 days of reloading after disuse may delay overall restoration of myofibre size. In other models of muscle damage, the manipulation of inflammation after damaging stimuli results in a delay in recovery of myofibre size (Bondesen et al. 2004). The resolution of oedema and inflammation may be critical for signalling myofibre growth after injury. Potential ovarian hormone-sensitive processes include immune cell infiltration and inflammatory cytokine expression (Kendall & Eston, 2002).
Inflammatory cells are rapid infiltrators of damaged skeletal muscle, and alterations in their temporal appearance are responsible for disruption of muscle healing processes (Bondesen et al. 2004). The immune response is an ovarian hormone-sensitive process of muscle recovery (Stupka et al. 2000; Tiidus et al. 2005). The rapid sequence of inflammatory cell invasion observed after damage (neutrophils, ED1+ macrophages and then ED2+ macrophages) follows a temporal pattern previously published for this model (Frenette et al. 2002). Neutrophils are considered to be one of the first immune cell types to enter skeletal muscle in response to damage and are partly responsible for the stimulation of inflammation through phagocytosis and oxygen free radical release (Tidball, 2005). ED1+ macrophage infiltration increases and decreases within a 2–3 day window during recovery from disuse and coincides with decreasing concentrations of neutrophils. Mast cell infiltration in reloaded skeletal muscle, however, has previously not been characterized. We demonstrate for the first time that mast cells infiltrate reloaded skeletal muscle after disuse in a pattern similar to that observed for neutrophils. Mast cells may function to stimulate inflammation and aid in the release of chemo-attractants and pro-inflammatory cytokines during the initial response to reloading-induced damage (Gorospe et al. 1996).
Ovariectomy did not alter the initial pattern of neutrophil, ED1+ macrophage or mast cell accumulation during the first day of recovery. The present study demonstrates an ovarian hormone-sensitive maintenance of peak concentrations of neutrophils at day 3 of recovery. Both neutrophil infiltration and the expression of neutrophil chemo-attractants are attenuated by the ovarian hormone oestrogen in vivo (Tiidus et al. 2001). The maintenance of peak neutrophil concentrations at day 3 of recovery in OVX rats could contribute to alterations in the temporal infiltration of other immune cells, including macrophages. With ovariectomy, macrophage concentrations remain increased at day 7 of recovery. However, macrophage chemotaxis may be cyclic throughout the initial week of reloading in soleus muscles from ovariectomized females. ED2+ macrophage infiltration appears to be an ovarian hormone-sensitive process during the initial 3 days of recovery. Disruption of the temporal infiltration of ED2+ macrophages at the onset of recovery could be related to alterations in macrophage activity or might occur as a result of disrupted governing of factors regulating their chemotaxis. Macrophage function in damaged muscle includes the phagocytosis of necrotic tissue and the stimulation of inflammation and regeneration by release of cytokines and growth factors. The population of phagocytic ED1+ macrophages in soleus muscle from OVX females was sustained at high levels at day 7 of recovery. ED1+ macrophage concentrations decline in conjunction with the completion of tissue necrosis occurring during reloading recovery (St Pierre & Tidball, 1994a). Prolonged localization of ED1+ macrophages in the present study could indicate delays in the removal of elevated apoptotic neutrophil populations or continued phagocytosis of necrotic myofibre debris. Alveolar macrophage phagocytosis is higher in female rats than in males, indicating that ovarian hormones potentially contribute to macrophage function (Spitzer, 1999). In addition, macrophages from female rats exhibit an ovarian hormone-sensitive increase in superoxide dismutase (SOD) and catalase (CAT) activities (Azevedo et al. 2001). Delays in the activity of ED1+ macrophages may also contribute to alterations in non-phagocytic macrophage infiltration after ovariectomy. Non-phagocytic ED2+ macrophages are implicated in muscle repair processes (Massimino et al. 1997). ED2+ macrophage infiltration occurs during the latter sequence of ED1+ macrophage accumulation in intact muscle, and ovariectomy alters this process. Changes in phagocytosis of tissue debris and stimulation of muscle regeneration by specific macrophage subpopulations may be related to altered inflammatory signalling.
In muscle recovering from damaging stimuli, the immune system provides critical cellular machinery and molecular signals that contribute to muscle repair (Tidball, 2005). The present study demonstrates that macrophages are a critical component of the timely recovery of muscle from disuse atrophy during reloading. The specific services provided by macrophages that are required for timely recovery of muscle have not been pinpointed, but are believed to involve phagocytosis of necrotic debris and neutrophis, cytokine and chemokine release, and myogenic precursor cell activation (Tidball, 2005). Although the gene expressions of IL-1β and IL-6 are not altered by ovariectomy during recovery in the present model, the possibility remains that gene targets of these cytokines are altered. Intracellular adhesion molecule-1 (ICAM-1), and endothelial selectin (E-selectin), and vascular cell adhesion molecules are ovarian hormone-sensitive targets of cytokine signalling involved in immune cell chemotaxis (Oger et al. 2000; Stork et al. 2002). Other candidates for ovarian hormone-sensitive cytokine-mediated inflammation include macrophage migration inhibitory factor (MIF). This is a pro-inflammatory cytokine expressed by activated macrophages during wound healing and results in excessive inflammation and delayed healing when altered (Ashcroft et al. 2003). Myofibres themselves are a rich source of cytokines in recovering muscle, and this provides evidence of a potential mechanism for the dissociation of cytokine expression and inflammatory cell accumulation in recovering muscle (Hamada et al. 2005). Macrophage stimulation of skeletal muscle satellite cell activity is another potential alternative mechanism not explored in the present study. Loss of ovarian hormone influence during the initial week of recovery may create a local regenerative environment non-conducive to macrophage activity and satellite cell activation, resulting in the retardation of non-contractile tissue resolution and recovery. We have already determined that loss of the specific ovarian hormone oestrogen results in prolonged expansion of non-contractile tissue and collagen 1a mRNA abundance during recovery. Future work is necessary to determine whether ovarian hormones such as oestrogen regulate gene expression related to satellite cell activation during recovery from disuse, and whether macrophage behaviour after muscle injury is a mechanism for these potential alterations.
The maintenance of lean body mass is a critical component for health in both chronic wasting diseases and ageing, and this suggests the importance of muscle regeneration and growth in the prevention of disuse-induced muscle atrophy in cachexic states (Anker & Coats, 1999). One of the consequences of the ageing process is an altered hormonal environment, resulting in decreasing oestrogen levels during menopause that result in loss of lean muscle mass and strength (Phillips et al. 1993; Poehlman et al. 1995). Decreases in function and mobility with age and disuse injuries are serious threats to older women owing to their lower functional capacity and longer lifespan compared with men (Rice et al. 1989). Successful recovery during the initial week of muscle reloading after disuse atrophy requires the integration of hormonal and inflammatory pathways that influence the response of the muscle. The present study demonstrates that ovarian hormones can have a significant impact on muscle immune cell infiltration during recovery after disuse. Delineation of ovarian hormone-sensitive mechanisms related to muscle recovery will aid in the development of future pharmaceutical therapies for the rapid and efficient recovery of skeletal muscle from periods of disuse.
Appendix
Acknowledgements
The authors would like to thank Marlene E. Wilson and Edie C. Goldsmith for their contributions to this manuscript and Kristen A. Mehl and Tyrone A. Washington for technical assistance. This work was supported by National Science Foundation/EPSCoR grant no. EPS-0132573 and NIH/Biomedical Research Infrastructure Network (BRIN) grant no. 8-PORR13461A, as well as American College of Sports Medicine/NASA Student grants awarded to Joseph M. McClung for the 2002–2004 funding periods.
