Increased fat deposition in injured skeletal muscle is regulated by sex-specific hormones

Abstract

Sex differences in skeletal muscle regeneration are controversial; comparisons of regenerative events between sexes have not been rigorously defined in severe injury models. We comprehensively quantified inflammation and muscle regeneration between sexes and manipulated sex-specific hormones to determine effects on regeneration. Cardiotoxin injury was induced in intact, castrated and ovariectomized female and male mice; ovariectomized mice were replaced with low- or high-dose 17-β estradiol (E₂) or progesterone (P4). Extent of injury was comparable between intact mice, but females were more efficient in removal of necrotic debris, despite similar tissue levels of inflammatory cells and chemokines. Myofiber size during regeneration was equivalent between intact mice and after castration or ovariectomy (OVX) but was decreased (P < 0.001) in ovariectomized mice with high-dose E₂ replacement. Intermuscular adipocytes were absent in uninjured muscle, whereas adipocyte area was increased among regenerated myofibers in all groups. Interestingly, intermuscular fat was greater (P = 0.03) in intact females at day 14 compared with intact males. Furthermore, castration increased (P = 0.01) and OVX decreased adipocyte accumulation. After OVX, E₂, but not P4, replacement decreased (P ≤ 0.03) fat accumulation. In conclusion, sex-dependent differences in regeneration consisted of more efficient removal of necrosis and increased fat deposition in females with similar injury, inflammation, and regenerated myofiber size; high-dose E₂ decreased myofiber size and fat deposition. Adipocyte accumulation in regenerating muscle was influenced by sex-specific hormones. Recovery following muscle injury was different between males and females, and sex-specific hormones contributed to these differences, suggesting that sex-specific treatments could be beneficial after injury.

skeletal muscle regeneration is an important component of limb salvage following severe injury induced by traumatic and/or ischemic extremity injury. Multiple cellular and molecular components involved in skeletal muscle regeneration have been studied, and controversies exist regarding effects of sex in muscle injury, inflammation, and regeneration (28, 41, 62, 63). A better understanding of the role that sex plays in muscle regeneration could lead to sex-specific treatment strategies for limb salvage.

The male and female response to injury varies by organ system and mechanism of injury. For example, female mice exhibit less ischemia and reperfusion injury than males after hepatic surgery (25); however, alcoholic liver injury is more severe in female rats (35). Also, sex-specific protection leads to decreased acute lung injury in female mice (48), but female rats exhibit increased pulmonary fibrosis after injury (23). Although not entirely understood, these differences have been primarily attributed to sex-specific hormones, specifically estrogen.

Estrogen positively affects immune responses by acting as a modulator of proinflammatory responses (9). Normalization and maintenance of the immune response has been attributed to estrogen leading to decreased organ injury and sepsis in females after trauma (13, 16, 30, 69). Estrogen also is an antioxidant, thereby exerting a protective influence (7, 36, 40, 49, 58).

Estrogen and sex influences on postexercise injury have been debated, with rodent models suggesting decreased injury and inflammation in females; however, similar studies in humans are variable or show no sex differences (28, 41, 62, 63). Exercise-induced muscle injury is a common model used in rodents and humans. In skeletal muscle, sex-specific differences in injury have generally revealed decreased injury in female rodents (1, 8, 18, 20, 60). Most of these studies have used enzyme release as a marker of injury, most commonly creatine kinase, with decreased levels released after injury in females (2, 3, 10). Examination of the morphological changes in skeletal muscle after injury yields mixed results of either similar (4, 53) or decreased muscle damage in females (65). Furthermore, neutrophil recruitment was attenuated in female compared with male rats after exercise-induced injury and estrogen supplementation in males decreased neutrophil recruitment to similar levels observed in females (59).

Sex influences on muscle regeneration would also appear to favor females in regard to satellite cells. Satellite cells, or myogenic progenitor cells, are the primary progenitor cells responsible for successful skeletal muscle regeneration (39, 45, 70); stimulation by estrogen leads to increased activation and proliferation (17, 18, 61). Additionally, muscle-derived stem cells from females, compared with males, were more efficient at regenerating injured skeletal muscle in both sexes (15).

The current study used cardiotoxin (CTX) injury, an established model of extensive muscle injury and regeneration (26, 34, 37, 54). CTX contains lytic factors that degrade the muscle plasma membrane (26), resulting in muscle necrosis followed by an inflammatory response with angiogenesis and skeletal muscle regeneration (34, 37, 54). The larger extent of injury obtainable using CTX, compared with the mild to moderate injury with exercise-induced muscle injury, maybe more consistent with clinically treated traumatic injuries.

Given the reported female advantages in injury and regeneration in rodents in exercise-induced muscle injury (19, 28, 41, 62, 63), we hypothesized that female mice would exhibit less injury and larger regenerated cross-sectional area (CSA) at earlier time points after extensive injury compared with male mice. The purpose of this study was to determine inflammatory cell recruitment, injury, necrosis, inflammatory chemokines, and muscle regeneration in intact male and female mice in a more extensive model of muscle injury. Furthermore, we aimed to determine the influence of sex-specific hormones on muscle regeneration.

MATERIALS AND METHODS

Experimental animals.

C57BL/6J mice were purchased from Jackson Laboratories. Male and female mice, 4–6 mo old, were used in this study. All procedures complied with the National Institutes of Health Animal Care and Use Guidelines and were approved by the Institutional Animal Care and Use Committee of the University of Texas Health Science Center at San Antonio and the South Texas Veterans Health Care System, San Antonio, TX.

Mouse CTX model.

Myonecrosis was induced by the intramuscular injection of CTX (Calbiochem, San Diego, CA), as previously described (34, 37, 54). In brief, two 50-μl CTX [2.5 μM in normal saline (NS)] injections were delivered uniformly into the muscles of the right hind limb anterior compartment. Similarly, the right hind limb posterior compartment received four 50-μl CTX injections. The left hind limb was injected with identical volumes of NS. Baseline mice did not receive intramuscular injections and were used as controls.

Mouse castration and ovariectomy model.

Twelve-week-old male mice underwent castration. The castrations were performed by making a small incision on the scrotum in the midline, and the testicles were elevated bilaterally, suture ligated, and transected. The scrotal skin was closed, and the mouse recovered. Twelve-week-old female mice underwent bilateral ovariectomy (OVX). Two small dorsal incisions were made just below the inferior aspect of the rib cage bilaterally, the peritoneum was incised, the ovaries were identified and removed, the ovarian vessels were suture ligated, and the skin incisions were closed. At death, the uterus and plasma were collected and used to determine adequacy of OVX; plasma samples were stored at −80°C. Following castration and OVX, acetaminophen and codeine were administered via drinking water, and mice were monitored for 10 days to ensure complete recovery. The mice were allowed to recover for at least 4 wk before using the animals at 4–6 mo of age for baseline or CTX injection studies.

17β estradiol and progesterone implants.

Placebo, 17β estradiol (E₂), and progesterone (P4) pellets (Innovative Research of America, Sarasota, FL) were implanted in female mice at 12 wk of age during the OVX procedure. The pellets were implanted per the manufacturer's instructions between the skin and muscle through an incision on the lateral side of the neck. Mice were divided into five different groups: 1) OVX with placebo pellet, 2) OVX with 0.25 mg E₂ pellet, 3) OVX with 2.5 mg E₂ pellet, 4) OVX with 25 mg P4 pellet, and 5) OVX with 100 mg P4 pellet. The pellets were 60-day continuous release. The mice were injected with CTX 4–6 wk after OVX.

Measurement of protein, MCP-1, and MCP-5.

Muscles of the anterior compartment were removed, weighed, and immediately used in preparation of tissue lysates, as previously described (34, 37, 46). Briefly, on ice, muscles were minced, mixed, and a weighed portion was homogenized (Tissumizer, Tekmar, Cincinnati, OH) for 15 s in 1.5 ml of lysate buffer, i.e., 50 mM Tris buffer (pH 7.4) containing 250 mM NaCl, 1% Nonidet P-40 (Roche Applied Science, Indianapolis, IN), 50 mM NaF, 2 mM Na₃VO₄, 5.5 mM EDTA, 1 mM 4-(2-aminoethyl) benzenesulfonyl fluoride hydrochloride, 65 mM bestatin hydrochloride, 7 μM transepoxysuccinyl-l-leucylamido-(4-guanidino) butane, 11 mM leupeptin, 0.15 mM aprotinin and 1 mM phenylmethylsulfonylfluoride (all from Sigma-Aldrich, Saint Louis, MO, unless otherwise specified). The tissue homogenate was immediately centrifuged (4,400 g, 5 min, 4°C), the supernatant was removed and centrifuged (2,300 g, 5 min, 4°C). Aliquots of the final supernatant were stored at −80°C.

Tissue lysates were thawed on ice for 15 min and immediately centrifuged (3,300–16,100 g, 4°C, 10 min). Protein was determined by the Pierce BCA protein assay (Pierce Biotechnology, Rockford, IL) using a microtiter plate format; BSA from ICN Biomedicals (Costa Mesa, CA) in lysate buffer was used as the standard (34, 37, 46). Absorption in microtiter plates was monitored in a SpectraMax Plus plate reader (Molecular Devices, Sunnyvale, CA), and results were analyzed with SOFTmax PRO software (Molecular Devices).

MCP-1 and MCP-5 levels in tissue lysates were assessed by ELISA (R&D Systems, Minneapolis, MN), according to manufacturer's protocol with slight modifications, as previously described (34, 46). Standards and unknowns were diluted in lysate buffer, and results were expressed as picogram per milligram protein.

Measurement of plasma E₂ and P4 levels.

Levels of E₂ and P4 in mouse plasma were assessed in ELISA, according to the manufacturer's protocol (Cayman Chemical Co, Ann Arbor, MI). The lower limit of sensitivity for these assays was 32.8 pg/ml and 31.25 pg/ml for E₂ and P4, respectively.

Tissue monocyte/macrophage and neutrophil quantification.

Cells from muscle were isolated as previously described (54). Briefly, all muscles from the anterior and posterior compartments were removed at 3 or 7 days after injury. The total cell count was manually determined on a hemocytometer and divided by the weight of the tissue. The percentage of cells from each population was determined by flow cytometry and the absolute number of each cell population was calculated and expressed as cells/g tissue.

Monocytes/macrophages and neutrophils were analyzed by a modified procedure for muscle-associated cells as previously described (32, 54). Neutrophils were defined as CD11b+/Gr-1+, and monocytes/macrophages were defined as CD11b+/Gr-1- cells. The fluorescently conjugated monoclonal antibodies (BD Biosciences, San Jose, CA) were as follows: APC Gr-1 and PE CD11b, as well as the corresponding isotype control antibodies. Isotype controls were used to titrate each antibody to minimize nonspecfic binding.

Histology and histomorphometry.

Myofiber CSA and percent fat in tibialis anterior (TA) muscle were analyzed as previously described in detail (34, 37, 54). In brief, mice were killed at baseline or at various time points following CTX and NS injections. The right and left TA muscles were removed, fixed in 10% neutral buffered formalin, and embedded in paraffin prior to sectioning and staining with hematoxylin and eosin (H&E) by routine procedures. The average CSA (μm²) of myofibers in the specimen was determined after outlining individual myofibers in representative, digitized images of a given TA muscle. Only regenerated fibers with centrally located nuclei (11) were measured in the post-CTX specimen to minimize the effect of differences in injury; whereas, mature myofibers with peripherally located nuclei were measured in the baseline specimen. Fat area (%) in a given image was calculated after manual outline of intermuscular adipocyte area and division by the total area of the image. Data for male mice were previously published (34).

The area of muscle injury and residual necrosis in the TA muscle of individual animals obtained at days 5 and 7 post-CTX was analyzed as previously described (34). H&E-stained cross sections were measured after digital scanning of microscope slides using a model CS ScanScope system (Aperio Technologies, Vista, CA). The digitally captured slide was analyzed via ImageScope software (version v10.0.36.1805; Aperio Technologies) to measure the total area of the TA, area of TA injury, and area of residual TA necrosis. The entire area of injury for a given TA muscle in cross section was defined as the area of regenerated muscle cells with centrally located nuclei in combination with the area of residual necrotic myofibers. Percent necrosis was calculated as the area of necrosis relative to the entire area of injury. Percent injury was calculated as the entire area of injury relative to the entire cross-sectional area of the TA. Necrosis and injury data for male mice at the day 7 time point were previously published (34).

Oil red O staining of unfixed frozen sections of tibialis anterior muscle was performed using a kit from Poly Scientific (Bay Shore, NY), as previously described (14).

Determination of percent total body fat.

Body fat (%) in baseline mice only was determined in anesthetized animals (60 mg/kg ip pentobarbital; Abbott Laboratories, Chicago, IL) using a PIXImus Mouse Densitometer (General Electric, Waukesha, WI). Data for male mice were previously published (37, 54).

Data analysis.

Percent fat, MCP-1, MCP-5, neutrophils, and monocytes/macrophages were analyzed with an Exact Wilcoxon test with Hochberg correction for multiple testing to determine whether significant differences existed between intact male, intact female, castrated male, and ovariectomized female at individual time points. Fiber CSA was tested using ANOVA with Hochberg for multiple testing. ANOVA with Dunnett's-corrected P values were used to determine significant differences between intact male, intact female, castrated male, and ovariectomized female at different time points post-CTX injection compared with baseline values for CSA. As described previously, for lysates samples with values below the level of detection in the ELISA for MCP-5 (<15.625 pg/ml) and MCP-1 (<78 pg/ml), a value of the lowest detectable level/√2 pg/ml was assigned to these samples (27), and this value was corrected for the protein in each specimen (34). Percent injury and percent necrosis in intact male, intact female, and ovariectomized female were analyzed using ANOVA with Tukey correction. Percent total body fat was analyzed by an unpaired Student's t-test.

SAS Version 9.2 for Windows (SAS Institute, Cary, North Carolina) was used for statistical analyses, and all statistical testing was two-sided with an experiment-wide significance level of 5%. Data were presented as the means ± SE.

RESULTS

Histologic evaluation of skeletal muscle injury and regeneration.

Acute myonecrosis, edema, and inflammatory cell infiltration developed in both male and female mice following CTX injection. At day 1 postinjury, the predominant inflammatory cell was the polymorphonuclear neutrophilic leukocyte, whereas, by day 3 in both groups, the principal cell was mononuclear. Small regenerated myofibers with centrally located nuclei were prevalent within 5 days postinjury in both male and female mice (Figs. 1, A and B). At this time point, necrotic myofibers persisted in the injured muscle of male mice; however, this residual necrosis was uncommon in the regenerated muscle of female mice. Within 7 days, regenerated muscle fibers prevailed in both groups and were progressively increased in size through 3 wk postinjury (Fig. 1, D and E).

Fig. 1.

Inflammation, adipocyte accumulation, myofiber necrosis, and tissue regeneration in tibialis anterior muscle following cardiotoxin-induced injury in male and female mice. A, B, D, E: representative images obtained from hematoxylin and eosin (H&E)-stained paraffin sections (3–4 μm) of formalin-fixed tibialis anterior muscle derived from male and female mice. Asterisks identify necrotic muscle fibers. Arrows indicate adipocytes among regenerated myofibers. C, F: representative images obtained from oil red O-stained frozen sections (4–6 μm) of unfixed tibialis anterior muscle. Hematoxylin counterstain.

The intramuscular accumulation of adipocytes was evident among regenerated myofibers in both female and male mice as early as 5 days postinjury (Fig. 1, B and C). This fat accretion was most abundant within the regenerated muscle of female mice at 2 wk following injury (Fig. 1, E and F).

Inflammatory cell accumulation and MCP levels in injured skeletal muscle.

Isolated inflammatory cells derived from injured hind limb muscles reached comparable levels in male and female mice and were transiently increased (Fig. 2, A and B). For both sexes, neutrophils and monocytes/macrophages peaked within 3 days following CTX-induced injury and decreased thereafter.

Fig. 2.

Neutrophil and monocyte/macrophage accumulation in cardiotoxin-injured muscles of male and female mice. Tissue neutrophils (CD11b+/Gr-1+) (A) and monocytes/macrophages (CD11b+/Gr-1-) (B) were determined by flow cytometric analysis of single-cell suspensions prepared from anterior and posterior compartments of below the knee hind limb muscles of male and female mice at 3 or 7 days postinjury. Data are presented as means ± SE; n = 4 mice/sex/time point.

Consistent with the above, muscle tissue levels of MCP-1 and MCP-5 were increased (P ≤ 0.03) above baseline within 1 day following muscle injury in both male and female mice (Figs. 3, A and B). MCP-1 and MCP-5 levels in muscle were similar in both sexes at all time points. Interestingly, in injured muscle of male mice, the tissue levels of MCP-5 were approximately twice that of MCP-1 at days 1 and 3 postinjury; in female mice, MCP-1 and MCP-5 levels were comparable at both of these time points.

Fig. 3.

MCP-1 and MCP-5 levels in male and female mice. MCP-1 (A) or MCP-5 (B) in tissue lysates prepared from the anterior compartment muscles of male and female mice at baseline and after cardiotoxin-induced injury. Data are presented as the means ± SE; n = 4–7 mice/sex/time point. For MCP-1, *P ≤ 0.008, and for MCP-5, *P = 0.03 compared with corresponding baseline for each sex.

Sex-dependent histomorphometric differences in muscle regeneration.

While the wet weight of the anterior compartment muscles was increased in males compared with females, similar weights were obtained between the CTX and NS-injected anterior compartments. Given the known differences in body weight between male and female mice, it is not surprising that the baseline (noninjured) myofiber CSA of male mice was larger (P < 0.001) than that of female mice (Fig. 4A). In contrast, the CSA of regenerated myofibers was similar between male and female mice through 21 days postinjury. For both sexes, regenerated myofibers progressively increased in size with time following injury and, at days 7 and 14, were smaller (P ≤ 0.001) than baseline myofibers (Fig. 4A). At 21 days post-CTX, regenerated myofibers in male mice remained smaller (P = 0.01) than baseline muscle, whereas, for female mice, regenerated myofibers were comparable in size to mature, baseline myofibers.

Fig. 4.

Sex-dependent differences in myofiber size and intermuscular fat accumulation. A: cross-sectional area (CSA) of mature myofibers at baseline (noninjured) and after injury (regenerated myofibers) in male and female mice. B: intermuscular adipocyte (fat) area (%) within injured/regenerated muscle; at baseline, intermuscular adipocytes were not detectable (ND) in either sex. Data derived from the tibialis anterior muscles are presented as means ± SE; n = 8–13 mice/sex/time point. *P ≤ 0.01 compared with corresponding baseline for each sex. #P ≤ 0.03 male vs. female mice at corresponding time points.

The extent of skeletal muscle injury within the TA muscle was extensive; >80% of the entire muscle, in both male and female mice (Table 1). At day 5, muscle injury was greater (P = 0.02) in females compared with males. In part, this may reflect the fact that necrotic myofibers were virtually eliminated and replaced by small regenerated myofibers in female mice. In contrast, necrotic myofibers were persistent and increased (P ≤ 0.05) in the injured muscle of male compared with female mice at both days 5 and 7 (Table 1).

Table 1.

Effect of sex on tibialis anterior muscle injury and resolution of necrosis

	Day 5		Day 7
	Injury, %	Necrosis, %	Injury, %	Necrosis, %
Male	81.5 ± 3.8	3.2 ± 1.3	85.2 ± 3.3	0.5 ± 0.2
Female	93.8 ± 1.6†	0.7 ± 0.4†	86.4 ± 3.9	0.03 ± 0.02#

Histomorphometric measurements of the tibialis anterior muscle in cross section were used to determine the extent of injury and residual necrosis at days 5 and 7 post-CTX injection; n = 12–15 mice/sex/time point. Data are presented as means ± SE.

^†Significant difference (P ≤ 0.02) between male and female mice at day 5.

^#Significant difference (P ≤ 0.03) between male and female mice at day 7.

Intermuscular fat was not detectable in baseline (noninjured) TA muscle of male and female mice, and the percentage of total body fat was similar in baseline mice between males (15.7% ± 0.57) and females (16.8% ± 0.52). Following injury, intermuscular adipocytes were increased (P ≤ 0.008) within the regenerated muscle in both sexes through day 21 (Fig. 4B). Somewhat surprising given the effectiveness of removal of necrotic myofibers and the robust growth of regenerated fibers in female mice, the extent of adipocyte accumulation within regenerated muscle was greater in female compared with male mice at day 14 post-CTX (P = 0.03).

Effect of castration or OVX on regenerated myofiber size and intermuscular fat accumulation.

The size of mature, baseline (noninjured) myofibers was similar following castration or OVX compared with intact mice; however, castrated males, similar to intact males, did exhibit larger (P ≤ 0.05) baseline CSA compared with both intact and ovariectomized females (Fig. 5A), and all groups after injury were smaller than baseline (P < 0.001). At day 14 postinjury, regenerated myofiber size in castrated or ovariectomized mice was comparable to that of intact male or female mice, respectively (Fig. 5A). In addition, percent injury (86 ± 4 vs. 94 ± 1) and necrosis (0.03 ± 0.02 vs. 0.02 ± 0.01) was similar between intact and ovariectomized females at day 7 after CTX injury.

Fig. 5.

Myofiber cross-sectional area and fat area in intact male and female mice and castrated male and ovariectomized female mice. Analysis performed on tibialis anterior muscle with evaluation at baseline (no injury) and after cardiotoxin-induced injury, n = 7–15 mice/treatment group/time point. A: cross-sectional area (μm²) of mature myofibers at baseline and after injury (regenerated myofibers) in intact male and female mice and castrated male and ovariectomized female mice. B: fat area (%) at baseline was not detectable (ND) in any group but increased after injury. Data are presented as means ± SE; *P ≤ 0.008 compared with corresponding baseline group. #P ≤ 0.04 between intact male mice vs. the three other groups at day 14. ¥P ≤ 0.05 compared with intact and ovariectomized female mice.

In contrast to regenerated myofiber size, the intermuscular accumulation of fat within regenerated muscle was significantly changed following castration (Fig. 5B). Thus, at 14 days postinjury, the percentage of intermuscular fat within regenerated muscle was increased (P = 0.01) following castration compared with intact males and was decreased, although not significantly, following OVX compared with intact females (Fig. 5B). These observations suggested the involvement of sex hormones in the expansion of adipocytes within regenerating muscle.

Effect of hormone replacement on regenerated myofiber size and intermuscular fat accumulation in injured skeletal muscle of ovariectomized mice.

Given the differences observed in intermuscular fat accumulation in mice, the effects of hormone replacement via implantable pellets were examined. Ovariectomized mice received implantable pellets designed to release either low doses or high doses of E₂ or P4.

Uterine weights are an accurate indicator of the completeness of ovariectomized and hormone replacement (6, 67). Uterine weights at day 14 postinjury were decreased (P < 0.001) following OVX compared with intact female mice (Fig. 6A). After hormone replacement of ovariectomized mice, uterine weights were increased (P < 0.001) compared with intact female mice with both the low doses and high doses of E₂. With P4 replacement following OVX, uterine weights were decreased (P < 0.001) in the low-dose group compared with intact animals (Fig. 6A). Compared with the OVX with placebo group, the high-dose P4 and both E₂ groups demonstrated increased (P ≤ 0.003) uterine weights. E₂ and P4 plasma levels were also determined using ELISA. E₂ levels were below the level of detection (32.8 pg/ml) except for the high-dose E₂ replacement group, 741 ± 66 pg/ml. The P4 levels were detectable for all replacement groups (Fig. 6B). At baseline, uninjured, intact female mice had higher levels than the baseline, uninjured ovariectomized mice, and 14-day placebo ovariectomized mice. This increased level correlates with the cyclicity of the estrous female. The E₂-replaced mice had dose-dependent decreased levels of P4. High-dose P4 replacement mice had higher levels (P < 0.001) than all other groups. Both the low-dose and high-dose P4 replacement mice demonstrated significantly increased levels of P4 (P ≤ 0.008) compared with baseline, uninjured ovariectomized mice.

Fig. 6.

Effect of hormone replacement on uterine weight, regenerated myofiber size, and intermuscular fat accumulation in ovariectomized mice. Animals were killed at 14 days postcardiotoxin injection. ovariectomized animals received implantable pellets containing either placebo, 17β estradiol (E₂; 0.25 or 2.5 mg), or progesterone (P4; 25 or 100 mg) at the time of OVX. Intact animals were not subjected to OVX. A: uterine weight (g) for each treatment group. B: plasma P4 levels (pg/ml) for each treatment group, as well as intact 0 day and OVX 0 day female mice. C: cross-sectional area of regenerated myofibers in the tibialis anterior muscle. D: intermuscular adipocyte (fat) area (%) within injured/regenerated tibialis anterior muscle. Data are presented as means ± SE; n = 8–14 mice/treatment group; *P ≤ 0.03 compared with intact female mice. #P ≤ 0.003 compared with OVX-placebo group. ¥P < 0.001 low dose E₂ vs. all other groups. §P < 0.001 compared with other replacement groups.

Regenerated myofiber size was comparable among all ovariectomized and hormone-treated groups with the exception of the high dose E₂. In the latter, regenerated myofiber size was decreased (P < 0.001) compared with all of the other groups (Fig. 6C). Similarly, the intermuscular accumulation of fat within regenerated muscle was lower, but not significantly different between the P4 groups and intact females. However, both doses of E₂ replacement resulted in decreased (P ≤ 0.03) percent fat compared with intact female mice (Fig. 6D).

DISCUSSION

The present study examined the influence of sex and sex-specific hormones on skeletal muscle regeneration in a model of extensive muscle injury. Despite previous literature reporting an overall female rodent advantage in exercise-induced muscle damage (19, 28, 41, 62, 63), we demonstrated that the extent of injury and regeneration, as measured by CSA, was remarkably similar between the sexes. Female mice removed necrotic myofibers more efficiently with equivalent numbers of inflammatory cells, but exhibited increased intermuscular fat accumulation. Male mice subjected to castration demonstrated increased intermuscular fat, and after OVX, female mice had decreased fat with further decreases in fat in ovariectomized mice replaced by E₂. Taken together, these observations suggested a hormonal influence on the level of fat accumulation after extensive muscle injury.

Female mice exhibited comparable (day 7) or increased (day 5) injury after similar exposure to hind limb injections of CTX compared with male mice. This is contrary to reports that the female sex provides protection against injury, which has been reported in multiple organ systems, such as lung (9, 48), heart (40), brain/nervous system (49), skeletal system (40), and skeletal muscle after exercise-induced muscle damage (17, 57, 62, 63). These studies attributed the differences in extent of injury to sex with potential mediation by estrogen. One possible explanation for the mildly increased level of injury observed in our laboratory is the variability of the CTX injection. However, our model typically achieves high levels of injury with each injection. More likely is that female sex does not confer decreased injury after extensive muscle injury. Skeletal muscle injury after exercise has been the primary model used with the majority of studies showing decreased injury in females (1, 8, 18, 20, 60). Most of these exercise-induced injury studies have used enzyme release as a marker of injury, most commonly creatine kinase, with decreased levels released after injury in females (2, 3, 10). Enzyme levels, however, may be inadequate markers of muscle injury, inaccurately reflecting the degree of histological damage (31, 65). Furthermore, an increase in creatine kinase in males compared with females may not be a reflection of increased injury, but rather, the increased muscle mass and body weight present in male mice. In addition, the membrane-stabilizing effect of estrogen that may protect muscle in exercise-induced injury model (1–3, 8) could be overcome by the more extensive injury induced by CTX.

Female mice were more efficient at removing necrotic muscle fibers after injury, but at the expense of increased fat deposition. This efficiency occurred despite similar levels of inflammatory cells and proinflammatory chemokines (MCP-1 and MCP-3). Neutrophil accumulation was similar between males and females after CTX injury (Fig. 2A). However, neutrophil accumulation, measured using myeloperoxidase levels after exercise-induced muscle damage, was decreased in female compared with male rats (59). Differences in time points, muscles tested, mechanism, and extent of injury may account for these variances. Monocyte/macrophage recruitment was also similar between male and female mice (Fig. 2B), confirming a previous study of muscle injury that demonstrated similar numbers of CD11b+ cells between the sexes; however, differences in macrophage subsets were noted (50). Macrophage subset studies were not performed herein, but they could potentially account for the increased efficiency in removing necrotic fibers in females vs. males despite similar overall numbers of inflammatory cells. While inflammatory cell recruitment and tissue chemokines were not measured after OVX, removal of necrotic fibers and CSA was similar in ovariectomized and intact female mice.

Surprisingly, the CSA of both sexes was remarkably similar throughout regeneration; castration and OVX did not influence regenerated fiber CSA. At baseline before injury, male mice have a known increase in CSA and muscle mass compared with female mice (24, 44). Interestingly, after injury, the baseline difference was not reflected in the similar CSA at days 7–14. By day 21, female CSA was similar to baseline while male mice were still decreased. Muscle-derived stem cells have been used to regenerate muscle in dystrophic mice (42). Female muscle-derived stem cells more efficiently regenerated skeletal muscle in both male and female mice (15). Furthermore, estrogen increases activation and proliferation of satellite cells (17, 18, 61). While satellite cell activation and muscle function were not measured in the current study, the regeneration advantage of estrogen was not observed in the female mice as measured using CSA; in contrast, increased intermuscular fat was present in females, suggesting impaired/altered regeneration.

Uninjured, mature male and female mice had similar percent total body fat in our study and a previous study (47). This finding is different than human males and females with ranges of 10–15% and 20–25%, respectively (43). In spite of similar percent total body fat at baseline, female mice had increased intermuscular fat accumulation after injury. The finding of increased fat accumulation occurring in regenerating skeletal muscle has been reported by our group (14, 34, 37) and others (66, 68) but not specifically related to sex. Interestingly, increased fat accumulation in injured muscle was associated with decreased monocyte/macrophage recruitment observed in MCP-1-/- and CC chemokine receptor 2-/- mice (34). Fat deposition in skeletal muscle has been reported in the elderly (21) and in muscle wasting diseases, such as muscular dystrophy (29), and may represent altered/impaired muscle regeneration.

Studies that show a sex-specific difference in response to injury primarily attribute this result to sex-specific-hormones, in particular, testosterone and estrogen, (9, 30, 38, 40, 48). We performed castrations and OVX to alter the hormone levels and examined the resultant changes in injury and regeneration. Castration resulted in similar CSA during regeneration, and OVX had similar injury and CSA compared with intact mice despite these procedures removing a significant portion of the hormonal influence. Interestingly, intermuscular fat accumulation did increase in castrated males and decrease in ovariectomized females. Age-related decreases in testosterone resulted in an increase in total body fat (12, 64) but were not specific to skeletal muscle deposition. Our findings suggest a hormonal influence on the amount of intermuscular fat after injury.

P4 replacement in ovariectomized mice does not appear to confer any protective benefit after extensive muscle injury. The levels of P4 measured in the E₂ replacement mice correlated with the negative feedback cycle of the mouse. E₂ and P4 both exert a respective negative feedback control on leutinizing hormone. Leutinizing hormone increases the level of E₂ and P4; however, increased levels of either hormone negatively control the level of leutinizing hormone, resulting in decreased levels of the other hormone (22), potentially leading to the decreased levels of P4 measured herein.

However, the OVX with hormone replacement mice revealed a dose-dependent effect of E₂ replacement. Mice receiving both doses of E₂ had less intermuscular fat compared with intact females, while only mice that received high-dose E₂ had significantly smaller CSA compared with intact and all the other ovariectomized groups. A limitation of the current study is the inability to determine whether the effects of estrogen replacement after OVX were primary or secondary. OVX leads to insulin resistance in skeletal muscle, and an increase in adipose tissue and E₂ administration resulted in a decrease in visceral but not subcutaneous adiposity (5). Further studies are needed to determine the mechanism of these effects.

Studies demonstrating a protective effect of estrogen all used low-dose replacement (25, 30, 40). The tantalizing possibility of using high-dose estrogen to improve muscle regeneration was not supported by our study. A potential limitation to our study is the reported variance in E₂ levels delivered via the pellet method, resulting in uneven and unpredictable plasma E₂ concentrations (52). Another study reported opposing effects on neural tissue after stroke relative to the method of administration of E₂ (51). Nevertheless, given these findings, administration of low-dose E₂ in postmenopausal women may have a positive effect on regeneration after injury and is a potential future area of exploration. However, studies in neural and vascular tissue have shown that hormone replacement therapy provides the most benefit when initiated in close proximity to menopause (33, 55, 56); the effect of timing of hormone replacement therapy for skeletal muscle will require further research.

Perspectives and Significance

In conclusion, there are sex-related differences in recovery from skeletal muscle injury; female mice demonstrated similar injury and regenerated myofiber size compared with males while high-dose E₂ resulted in decreased regenerated CSA. Adipocyte accumulation in regenerating muscle was influenced by sex-specific hormones; females healed with increased intermuscular fat compared with males, and removal of male hormones resulted in increased intermuscular fat. While high-dose E₂ decreased intermuscular fat, CSA was also decreased in ovariectomized mice. Given the considerable literature on exercise-induced muscle injury, our study may assist in defining sex-specific differences present after more extensive injury seen in trauma patients. Injury recovery is different between males and females, and sex-specific hormones contribute to these differences, suggesting that sex-specific treatments could be beneficial after injury.

GRANTS

These studies were supported, in part, by the United States Air Force and by grants from the National Institutes of Health R01-HL074236 and Veterans Administration Merit Review. Data were generated in the Core Flow Cytometry Facility, which is supported by the UTHSCSA and NIH-NCI P30 CA054174 (Cancer Therapy & Research Center).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.