Molecular Changes in Sub-lesional Muscle Following Acute Phase of Spinal Cord Injury

Abstract

To clarify the molecular changes of sublesional muscle in the acute phase of spinal cord injury (SCI), a moderately severe injury (40 g cm) was induced in the spinal cord (T10 vertebral level) of adult male Sprague–Dawley rats (injury) and compared with sham (laminectomy only). Rats were sacrificed at 48 h (acute) post injury, and gastrocnemius muscles were excised. Morphological examination revealed no significant changes in the muscle fiber diameter between the sham and injury rats. Western blot analyses performed on the visibly red, central portion of the gastrocnemius muscle showed significantly higher expression of muscle specific E3 ubiquitin ligases (muscle ring finger-1 and muscle atrophy f-box) and significantly lower expression of phosphorylated Akt-1/2/3 in the injury group compared to the sham group. Cyclooxygenase 2, tumor necrosis factor alpha (TNF-α), and caspase-1, also had a significantly higher expression in the injury group; although, the mRNA levels of TNF-α and IL-6 did not show any significant difference between the sham and injury groups. These results suggest activation of protein degradation, deactivation of protein synthesis, and development of inflammatory reaction occurring in the sublesional muscles in the acute phase of SCI before overt muscle atrophy is seen.

Introduction

Approximately 12,500 new spinal cord injuries (SCIs) occur each year and currently 270,000 people live with SCI in the US [1], leading to a huge financial burden on the individual’s family and the community. SCIs cause various neurological manifestations as well as universal reductions in sublesional muscle mass [2]. Individuals with complete SCI experience about 50–75 % muscle atrophy [3], and those with incomplete SCI have 20–30 % atrophy [4] within 6 months to a year after SCI. In an animal study, the greatest amount of atrophy was measured at 2-weeks post injury (11–26 %), and spontaneous recovery in muscle size was observed by 4-weeks post-SCI, however, compared to humans, rats show greater neurological recovery [5]. Muscle atrophy in SCI is associated with reductions of maximal force generation and endurance, creating significant negative effects on the rehabilitation [2]. Because muscle is a major organ for metabolism, muscle atrophy in SCI patients also has been associated with increased risk for secondary metabolic co-morbidities, including diabetes mellitus [6, 7], metabolic syndrome [8, 9], and cardiovascular disease [9]. However, the precise pathomechanisms for these have not been fully investigated.

Following muscle atrophy, muscle fiber type switch occurs in individuals with SCI [10]. This fiber-type transformation from type I, slow twitch, to type II, fast twitch fibers, may explain the increased fatigability in patients with SCI [10]. In humans, muscle fiber-type transformation begins 4–7 months post SCI and reaches a new steady state in 20–70 months [10]. Oxidative enzyme activity starts to decline after the first couple of months, possibly corresponding to the transformation from type I to type II fibers [10]. In an important study performed by McCall et al. [11] plasmids of MHC type IIb gene promoters were injected into the soleus muscle of female rats followed by spinal cord isolation (SI) surgery. Aim of this study was to identify the type of promoter region that leads to the de novo transcription of MHC type IIb gene. Seven days post SI the concentration of the pre-mRNA and mRNA was significantly higher in the soleus muscles. Muscles in individuals with chronic SCI show less resistance to fatigue and this fatigue negatively correlates to the time from injury [10].

At the molecular level, muscle atrophy occurs as a result of a network of very complex mechanisms, but mainly occurs due to an imbalance of protein synthesis and degradation. The prime targets in the control of muscle protein synthesis and breakdown are insulin-like growth factor-1(IGF-1) and protein kinase B (PKB)/Akt. An upregulation of IGF-1 and Akt promote muscle protein synthesis through activation of the mammalian target of rapamycin (mTOR) and its downstream effector ribosomal S6 kinase 1 (S6K1), the two key regulators of translation initiation and protein synthesis, and by blocking a major muscle protein breakdown pathway involving forkhead transcription factor (FoxO) 1, 3 and 4. A decrease in the activation of IGF-1 and Akt leads to a decrease in protein synthesis and, subsequently, muscle loss. Furthermore, reduced IGF-1 and Akt activity relieves inhibition of FoxO (hypophosphorylation), thereby allowing FoxO to assume its role as a nuclear transcription factor and promote the activation of several key genes of the ubiquitin–proteasome pathway, especially muscle-specific E3 ligases, such as muscle atrophy F-box (MAFbx)/atrogin-1 and muscle ring finger 1 (MuRF1) [12, 13].

In the sublesional muscles of individuals with chronic SCI, various molecular changes had been identified. In a study of human subjects with chronic SCI (between 3 months and 30 years post injury) [14], a significant reduction in mRNA levels of MuRF1, MAFbx, myostatin, and in protein levels of FoXO1, FoXO3a, MAFbx, IGF-1, phosphorylated GSK-3β, and 4E-BP1 was observed in the left vastus lateralis compared to age-matched healthy subjects. After 10 weeks of spinal cord transection, mRNA levels of PKB/Akt, ribosomal S6 kinase (S6K1), and myogenin were significantly decreased, while expression of MuRF1, FoxO4 protein was increased significantly in the soleus muscles [13]. Likewise, in another 10-week experimental study with complete spinal cord transection, total Akt protein expression in the soleus muscles of male Sprague–Dawley rats were significantly lower in the injury group, compared to the control group [15]. Using a spinal cord isolation model, Kim et al. [16] found significantly higher mRNA levels of MuRF1 and MAFbx as well as higher Akt protein expression in the medial gastrocnemius muscles in SCI rats at 1-week post injury induction compared to controls. An interesting study was conducted by Thomason et al. in 1989 to study the muscle protein synthesis. They suspended the hindlimbs of rats from any movement for a period of 5 h and 7 days and injected [3H] Leu and measured the aminoacyl-tRNA precursor and protein products 5 h and 7 days after the suspension. The results showed that there was not a consistent decrease in protein synthesis but rather a gradual decrease, which they modeled through a half-life equation [17].

Along with studying and reversing the molecular changes occurring after SCI, successful attempts have been made to improve the overall functional recovery. A multitude of studies on SCI individuals and experimental rodents have been performed with electrical stimulations to facilitate voluntary movement [18, 19] Dudley-Javoroski et al. have used electrical stimulation to improve bone strength by stimulating muscles periodically to improve the bone mineral density [20–22]. Some studies such as that of Kim et al. [16] have studied the muscle degradation markers (MuRF1 and MAFbx) on rats whose muscles were electrically stimulated after undergoing spinal cord isolation surgeries. The mRNA levels of these degradative ubiquitin proteasome system markers decreased significantly after receiving the electrical stimulations in their soleus muscles where their contralateral limb served as the control. These electrical stimulations have been helpful in recovering the overall functionality of the limbs due to the existence of the central pattern generation in the spinal cord. A recent review by Roy et al. [23] summarizes the potential the spinal cord has for recovering after an SCI. It reviews multiple studies done in rodents, cats and humans which all imply that giving the spinal cord enough sensory cues without any input from the supraspinal centers can lead to motor recovery due to the existence of the property that spinal cords have known as central pattern generation.

While muscular atrophy has been moderately studied in individuals with chronic SCI, such investigations in acute rodent SCI models are limited. A multitude of data in this field implies that sublesional muscle atrophy does not occur immediately following SCI. To our knowledge, there has been a study done by Urso et al. [24] examining the molecular changes that occur in the acute phase of SCI at subsequent time points (2 and 5 days post SCI). However, this study (done on the biopsy samples of the vastus lateralis muscles of individuals suffering from SCI) did not demonstrate any significant differences in the protein concentrations of MuRF1 and MAFbx, two members of the E3 ligase system but the qRT-PCR analyses of those genes showed a significant increase at both 2 and 5 days post SCI. Such information is important in order to develop treatments to cure or prevent muscle atrophy in SCI, and will be critical in identifying when to initiate the treatment. In order to clarify these issues, we induced a clinically relevant SCI in rats and examined molecular changes in their sublesional muscles in its acute phase.

Materials and Methods

SCI Induction

Male Sprague–Dawley rats (250–300 g) were housed in individual cages and given water and food ad libitum. All animal care, surgery, and induction of SCI was conducted in accordance with the Guide for the Care and Use of Laboratory Animals of the US Department of Health and Human Services (National Institutes of Health, Bethesda, MD, USA) and was approved by the Institutional Animal Care and Use Committee (IACUC) at the Medical University of South Carolina. An experienced technician induced SCI employing the modified weight-drop method of Perot [25, 26]. Laminectomy was performed at T10, the spine was immobilized, and an impounder was gently placed onto the dura. A clinically relevant model of moderately severe SCI was induced by dropping a 5 g weight from a height of 8 cm to produce a 40 g cm force. Shamoperated rats underwent laminectomy alone. Two sets of rats were used in this study. The first cohort consisted of six sham and six injury rats, and the second cohort consisted of four-sham and five-injury rats (n = 10 sham and n = 11 injury). At 48 h post-injury, rats were anesthetized with ketamine 80 mg/kg and xylazine 20 mg/kg and then sacrificed by decapitation.

Muscle Sample Collection

Immediately after sacrificing the rats, gastrocnemius muscles were dissected, separated into visibly white and red portions, and saved in two different ways. Muscles from the right hindlimb were stored in −80 °C for further processing via Western blot analysis. The muscles from the left hindlimbs were placed on embedding blocks (Thomas Scientific, Swedesboro, NJ, USA) mounted with tragacanth gum mix—(5 g tragacanth powder (Sigma-Aldrich) mixed with 1 mL phenol and 50 mL water. After the muscles were mounted, they were swirled in isopentane that had been kept in a container filled with liquid nitrogen. The frozen muscle samples were left in a covered box of dry ice for 30 min before being stored in −80 °C.

Western Blot Analyses

Frozen gastrocnemius muscles were cut into smaller pieces with a scalpel. The finely cut tissue was homogenized with 500 μL of homogenizing buffer [50 mM Tris–HCl, (pH 7.4) containing 5 mM EGTA, and 1 Mm phenylmethyl-sulfonyl fluoride]. Samples were homogenized with intermittent sonication until a completely homogenous suspension of tissue was attained. Protein estimation was performed, and all the samples were boiled and later diluted to 1.5 μg/ μL concentration with sample buffer [62.5 mM Tris–HCl, pH 6.8, 2 % sodium dodecyl sulfate, 5 mM β-mercaptoethanol, 10 % glycerol]. SDS-PAGE was performed with the protein samples using 4–20 % precast gels (Bio-Rad Laboratories, CA, USA) and resolved at 100 V for 1 and 1/2 h. Immunoblotting was performed according to Samantaray et al. [27]. The resolved gel was transferred to the Immobilon^™-polyvinylidene fluoride microporous membranes (Millipore, MA, USA). The membranes were then blocked with non-fat dry milk powder in wash buffer (0.1 % Tween-20 in 20 mM Tris–HCl, pH 7.6). Immunoblotting was then performed with primary antibodies: Anti-fast myosin heavy chain (dilution 1:1000), Anti-slow myosin heavy chain (dilution 1:5000, Abcam, MA, USA); AKT1, Caspase-1, Cox-2, GAPDH, IL-1β, MAFbx, MuRF1, p-AKT1/2/3 (dilution 1:200, Santa Cruz, CA, USA); TNF-α, S6 ribosomal protein and p-S6 ribosomal protein (dilution 1:1000, Cell Signaling, MA, USA). Membranes were immersed in primary antibodies on a shaker overnight at 4 °C. The membranes were then incubated with respective secondary antibodies (dilution 1:2000) tagged with horseradish peroxidase (primary goat anti-mouse or primary goat anti-rabbit antibodies) at room temperature. In between incubations, the membranes were washed with wash buffer 3× for 5 min. Chemiluminescent reagents were used to bind the immunoreactive bands on the membranes, and the images of protein bands were viewed via Alpha Innotech FluorChem FC2 Imager. The immunoreactive bands were then quantified using NIH’s Image J software.

Quantitative Real-Time Polymerase Chain Reaction Analyses

Total RNA was isolated from the right gastrocnemius muscle tissue using the TRIzol reagent solution (Ambion^® Life Technologies, NY, USA). The isolated mRNA was reverse transcribed to cDNA with the iScript Advanced cDNA synthesis kit (Bio-Rad Laboratories) using an Eppendorf-AG thermal cycler. qRT-PCR was performed using the Bio-Rad CFX96 C1000 Thermal Cycler with the forward and reverse primers (Integrated DNA Technologies) as listed in Table 1 along with SsoAdvanced Universal SYBR Green Supermix reagent (Bio-Rad Laboratories) and data was analyzed using the 2^−ΔΔCt method. Beta-tubulin was used as a reference gene for normalizing the sham and injury samples.

Table 1.

Rat oligonucleotide primers used in the study

Gene	Primer sequence 5′–3′		Accession no.
TNF-α	Forward	AAGCTGTCGCTACATCACTG	NM_012675.3
	Reverse	CTCTCAATGACCCGTAGGGC
IL-6	Forward	TTTCTCTCCGCAAGAGACTTCC	NM_012589.2
	Reverse	TGTGGGTGGTATCCTCTGTGA
Tubulin	Forward	CTCTCTGTCGACTACGGAAAG	NM_001044270.2
	Reverse	TGGTGAGGATGGAATTGTAGG

Morphological Analysis

Skeletal muscle blocks from sham (n = 4) and injury (n = 3) rats were sliced to identify morphological changes. Muscle blocks stored at −80 °C were warmed to −20 °C in a Reichart-Jung cryostat (Cryocut 1800; Leica, Wetzlar, Germany). The blocks with the muscle tissue were mounted on a chuck and sliced using the cryostat to a 10 μm thickness. Coronal slices were put on superfrost plus microscope slides, temporarily kept in the crystostat at −20 °C until enough slides were obtained from one block, and later stored in −80 °C. The morphology was visualized using hematoxylin and eosin staining. The muscle slices were viewed under an Olympus BH-2 microscope with Magnafire software and images were captured at 200× magnification.

Fiber Diameter Measurement

Muscle fiber diameter was measured in sham and injury rats according to Park et al. [28]. The shortest fiber diameter of each fiber was measured using the NIH’s image J program (http://rsb.info.nih.gov/ij/). We examined five fields per rat, and counted at least 170 fibers (in μm ± SEM) per group. A microscopic scale was first used to set up the scale size in micrometers. Consequently, a function was defined in the Image J program so that the diameter of the enclosed fiber area could be measured. The boundaries of the fibers were traced with the freehand selection tool, and the value of the shortest Feret’s diameter of the fiber was recorded and saved.

Statistical Analyses

The raw values of immunobands obtained by the Image J analyses were exported to Microsoft Excel. For each blot, the original raw values of the bands were normalized with the reprobed GAPDH bands, and the final normalized values were exported to the Graphpad Prism statistical software. The qRT-PCR values were normalized with Tubulin in Microsoft Excel and T-tests were performed using the Graphpad Prism statistical software. All T-tests were performed between the sham and injury groups at least 95 % CI, p ≤ 0.05, and the bar graphs were also plotted. T test was also performed to compare the fiber sizes of both the sham and injury groups.

Results

Identifying the Differences Between the Visibly Red and White Muscle Tissues

In the visual inspection of the medial portion of the gastrocnemius muscle, the central portion of the muscle is visibly red, while the peripheral portion is visibly white (Fig. 1a). In order to examine the difference in protein expression between them, we obtained samples from each portion and performed Western blot analyses. There was a significantly higher (p < 0.01) MHC type I (slow) immunoreactivity at 200 kDa in the central red muscles compared to the peripheral white muscles (Fig. 1b). The MHC type II (fast) at 223 kDa also revealed a statistically significant (**p < 0.01) increase in the visibly red muscle compared to the visibly white muscle. These results suggest a higher content of myofibril proteins in the medial visibly red portion. Thus, we performed all remaining studies using samples from this portion.

Fig. 1.

Myofibril protein expression in the visibly red and white parts of skeletal muscle. a Representative image of the gastrocnemius muscle in adult male rats. b Western blots to show levels of 200 kDa MHC I (slow myosin skeletal heavy chain) and 223 kDa MHC II (fast myosin skeletal heavy chain) in the red and white parts of the gastrocnemius muscle, and determination of the densitometric difference of each protein expression in red and white parts of the muscle (p < 0.01)**. n = 6 in each

Morphological Examination of the Sub-lesional Muscle Following Acute SCI

To examine the morphological changes of the sublesional skeletal muscle in rats during the acute phase of SCI, we induced a clinically relevant injury, sacrificed rats at 48 h, and obtained muscle samples from medial gastrocnemius muscles. We used sham (laminectomy only) for comparison. H&E staining of axial muscle slices showed no significant morphological changes in injury compared to sham (Fig. 2a). In addition, there was no statistically significant difference in fiber diameter between these two groups (Fig. 2b). The injury group also did not show any infiltration by inflammatory cells.

Fig. 2.

Morphology of sublesional skeletal muscle in the acute phase of SCI. a Representative axial thin slice (10 μm) with H&E staining of gastrocnemius muscle of adult male rat at 48 h after the induction of SCI (Injury) showed no significant morphological difference compared to the laminectomy only (Sham). b Determination of muscle fiber diameter. There was no statistically significant difference in the minimum Feret diameter of muscle fiber between sham and injury at ×200 magnification

Altered Protein Expression in Sub-lesional Muscle Following Acute SCI

We performed Western blot analyses of MuRF1 and MAFbx to examine whether SCI induced any molecular changes in sublesional muscles in the acute phase. We first examined the expression level of muscle specific E3 ligase proteins to examine protein degradation. The injury group had a significantly (**p < 0.01) higher expression of the MuRF1 (44 kDa) and MAFbx (42 kDa) compared to sham (n = 6, each) (Fig. 3a, b).

Fig. 3.

Changes in muscle specific E3 ligase protein expression in sublesional muscle of acute SCI. Western blot and determination of densitometric changes of 44 kDa MuRF1 and 42 kDa MAFbx in gastrocnemius muscle of SCI at 48 h after its induction showed statistically significantly increased expression of both proteins compared to sham (p < 0.01)**. n = 6 in each

We then examined the expression level of the protein synthesis related proteins by Western blot analyses and found p-Akt (active bands at 60/56/60 kDa) and p-S6 ribosomal protein (active band 32 kDa) were significantly (*p < 0.05) downregulated in the SCI group (n = 5) compared to the sham group (n = 4)(Fig. 4a, b). Total Akt1 expression showed a downward trend in the injury rats whereas S6 ribosomal protein showed a slight increase in the injury group (n = 5) compared to the sham group (n = 4) although the difference was not statistically significant in both markers.

Fig. 4.

Changes in protein synthesis related protein expression in sublesional muscle of acute SCI. Western blot and determination of densitometric changes of 62 kDa Akt1, 60/56/60 kDa p-Akt1/2/3 and 32 kDa of both S6 Ribosomal and p-S6 Ribosomal proteins in gastrocnemius muscle of SCI at 48 h after its induction showed statistically significantly decreased expression of p-Akt1/2/3 (injury, n = 5) and p-S6 Ribosomal protein (injury, n = 5) compared to sham (n = 4) (p < 0.05)*

Finally, we examined the expression level of inflammatory proteins and found that Cox-2 (*p < 0.05), TNF-α (*p < 0.05) and Caspase-1 (*p < 0.05) were significantly higher in expression in the injury rats (n = 5) compared to the sham rats (n = 4) (Fig. 5a, b).

Fig. 5.

Changes in inflammation related protein expression in sublesional muscle of acute SCI. Western blot and determination of densitometric changes of 70/72 kDa Cox-2, 28 kDa TNF-α, 20/22 kDa caspase-1, and 17 kDa IL-1β in gastrocnemius muscle of SCI and sham (laminectomy only) at 48 h after their induction are shown. Expression of Cox-2 immunobands at 70/72 kDa, TNF-α at 28 kDa, and caspase-1 at 20/22 kDa in injury animals were significantly increased compared to those in sham animals (p < 0.05)*. The expression of IL-1β showed no statistically significant difference between these two groups, but there was a tendency of higher expression in injury compared to sham. Sham (n = 4); injury (n = 5)

Changes in Gene Expression of Inflammatory Markers in Gastrocnemius Muscles Following Acute SCI

We performed qRT-PCR of inflammatory gene markers TNF-α, and IL-6 with the reference gene tubulin. Neither of the inflammatory markers showed a significant difference between the sham (n = 4) and injury (n = 5) groups (Fig. 6). Contrary to what was expected, the IL-6 expression in the injury group showed a downward trend in the injury group compared to the sham group.

Fig. 6.

mRNA expression of the known inflammatory markers in the sublesional muscle 48 h post SCI. qRT-PCR of Tnf-α and IL-6 show no statistically significant difference between the sham (n = 4) and injury (n = 5) groups as shown. These were first normalized with a reference gene Tubulin and later plotted using the numbers derived from the 2^−ΔΔCt method

Discussion

Molecular changes that may have developed in the sublesional skeletal muscles during the acute phase of SCI were examined. Initial studies where peripheral and central muscles were not distinguished yielded inconsistent immunoband signals (data not shown), which explained why we decided to separate the peripheral (visibly white) and central (visibly red) portions of the gastrocnemius muscles. Higher expression of both the fast and slow fiber proteins was identified in the central (visibly red) portion, suggesting a sample enriched for muscle proteins, and thus, all studies were conducted with those samples.

The morphological analyses of medial gastrocnemius muscle showed no significant morphological changes including muscle fiber diameter between SCI and sham rats at 48 h post injury. In order to identify whether protein degradation leading to muscular atrophy begins in this stage, immunoblotting of proteins involved in the E3 ligase ubiquitin proteosome system was performed. Our studies identified a significantly higher expression of MuRF1 and MAFbx in the gastrocnemius muscles of the SCI rats at 48 h compared to the sham. To examine whether protein synthesis in sublesional muscles is affected in the acute phase of SCI, expression of Akt, p-Akt1/2/3 (p-Akt), S6 and p-S6 ribosomal proteins in the medial gastrocnemius muscles was examined. In our experiments p-Akt and p-S6 ribosomal protein were significantly reduced 48 h after SCI. Akt showed a decreasing trend in SCI rats compared to sham whereas the S6 ribosomal protein showed a slight increase in the injury group compared to sham, however it was not significant. These results clearly indicate that increased muscle protein degradation and decreased muscle protein synthesis begin during the acute phase after induction of SCI prior to the development of muscle atrophy.

The molecular mechanisms that induce protein degradation and/or suppress protein synthesis in sublesional muscles following SCI have not been previously identified. Because the immune/inflammatory process is an acute response occurring from various stimulations—and is known to influence skeletal muscle cells including their anabolic and catabolic processes [29], we hypothesized that a molecular process serves as an inducing factor for the protein degradation and suppression of protein synthesis. Indeed, our study showed up-regulation of inflammatory markers including TNF-α, Cox-2, and Caspase-1 at the protein level in the medial gastrocnemius muscles of SCI rats compared to those of sham rats at 48 h after their induction. TNF-α is known to induce ubiquitin-dependent proteolytic system in skeletal muscles [30]. Our study showed upregulation of muscle specific E3 ligases in UPS (MuRF1 and MAFbx); it is possible that these reactions are induced by TNF-α. The mRNA levels of TNF-α did not show any significant alterations in the gastrocnemius muscles 48 h post SCI. This result suggests that the changes in the protein levels are occurring due to post-translational modifications. Additionally, because Cox-2 blockade in skeletal muscle eliminates the normal increase in muscle protein synthesis following resistance exercise, it is hypothesized that Cox-2 mediates the increase in postexercise muscle protein synthesis [31]. Indeed, upregulation of Cox-2 is associated with Akt activation/phosphorylation [32]. Our study showed upregulation of Cox-2 and downregulation of p-Akt. Hence, it is possible that Cox-2 upregulation as shown in the current study is a compensatory response to p-Akt downregulation. Protein expression of TNF-α, IL-1β, NOS2, and caspase-1 was found in the cytoplasm and sarcolemma of dystrophic skeletal muscle fibers, and thus, they are considered to be associated with muscle fiber damage [33]. TNF-α is also known to activate NF-κB, which is involved in disuse atrophy [34, 35].

Because the morphological studies showed no inflammatory cell infiltration, we hypothesize that these inflammatory markers are secreted from muscle cells. Cytokines or other peptides that are produced, expressed and released by muscle fibers exert autocrine, paracrine, or endocrine effects and are classified as myokines [36]. Clinical significance of cytokines for skeletal muscle has been debated. For example, IL-6 is known to induce skeletal muscle atrophy [37]. Interestingly, IL-6 is also produced and released in response to muscle contractions, resulting in the activation of AMPK and/or PI3 kinase to increase glucose uptake and fat oxidation, creating beneficial effects for the muscle. Our qRT-PCR results showed that there was no significant difference in the mRNA levels of IL-6 in the gastrocnemius muscles, in fact, there was trend toward reduction in IL-6 expression in the injury group compared to the sham group. Whether IL-6 acts as a pro-inflammatory or anti-inflammatory agent is considered to be more dependent on the environment (muscles vs. immune cells) [36]. There are also some recent publications that suggest that the levels of circulating cytokines after a traumatic CNS injury increases as early as 6 h post injury [38, 39] however, the H&E staining did not show any inflammatory cell infiltration in our study.

Current study suggests skeletal muscle protein breakdown (MuRF1 and MAFbx), downregulation of protein synthesis (p-Akt, p-S6 ribosomal protein) and induction of inflammatory reaction (TNF-α, Cox-2, and Caspase-1) in sub-lesional medial gastrocnemius muscles of SCI rats occur in the acute phase (48 h), prior to the morphological alterations. Currently, no effective pharmaceutical treatment is available to cure or prevent sub-lesional muscle atrophy following SCI. To this end, the intracellular changes we have reported earlier in muscle cells in response to extracellular inflammatory stimulation in vitro has been attenuated by calpain inhibition [40, 41]. The involvement of pathological calpain in CNS injury and diseases has been reported and implicated it to be a therapeutic target [42–45]. Our results offer suggestions for future research and development of such a treatment, including possibilities that the molecular changes noted herein can be targeted and the treatment should be initiated in the acute phase before muscle atrophy develops.