All experiments were approved by the Temple University Institutional Animal Care and Use Committee (Temple University IACUC) in compliance with NIH guidelines for the humane care and use of laboratory animals. Rats were housed individually in the central animal facility in transparent plastic cages in a 12 hour light: 12 hour dark cycle with free access to water. Studies were conducted on a total of 142, young adult (3.5–4 months of age at onset of experiments), female, Sprague-Dawley rats.

Seventeen rats served as age-matched normal controls (NC) with free access to food, and did not undergo training or task performance; 9 more NC rats received ibuprofen treatment (NC+IBU). All but NC and NC+IBU rats were food restricted (n = 117) to within 5% of their naïve weights as described previously [9], [16]. These food restricted rats went through an initial training period in which they learned to perform either a high repetition negligible force, food retrieval task (TR-NF, n = 51) as described previously [5], or a high repetition high force, handle-pulling task (TR-HF, n = 56) as described previously [7], [18], [21], for a food reward. The rats were trained for 10 min/day, 5 days/week. This training period was either 10 days (TR-NF) or 5–6 weeks (TR-HF) (learning to perform the high force task took longer than the negligible force task).

Forty-one of the TR-NF rats went on to perform the high repetition negligible force, food retrieval task (HRNF). These 41 rats performed the HRNF task for 2 hr/day, 3 days/week for 3 weeks (n = 12), 6 weeks (n = 11) or 9 weeks (n = 18). Thirty-four of the TR-HF rats went on to perform the high repetition high force, handle-pulling task (HRHF). These 34 rats performed the HRHF task for 2 hr/day, 3 days/week, for either 6 weeks (n = 29), or 9 weeks (n = 5). Subcohorts of the 6-week HRHF rats either went untreated (6 HRHF, n = 11), or were treated with anti-rat TNF-α (6HRHF+anti-TNF, n = 6) or ibuprofen (6HRHF+IBU, n = 12) in weeks 5 and 6 of HRHF task performance. The remaining TR-NF (n = 10) or TR-HF rats (n = 32) were euthanized after the training period, and served as trained-only controls. Subcohorts of TR-HF rats, either went untreated (TR-HF, n = 17), or received anti-rat TNF-α (TR-HF+anti-TNF, n = 6) or ibuprofen (TR-HF+IBU, n = 9) before euthanasia, as described further below.

All rats were weighed weekly and food adjusted accordingly. In addition to food pellet rewards, all rats received Purina rat chow daily. Trained-only rats received food pellets and rat chow that matched allotments given to task rats. In accordance with the recommendations of the Panel on Euthanasia of the American Veterinary Medical Association, animals were euthanized using sodium pentobarbital (120 mg/kg body weight) before tissue collection.

The HRNF and HRHF tasks have been described in detail previously [5], [7], [18], [21]. Briefly, the HRNF task was a forelimb food pellet retrieval task in which rats reached and retrieved a food pellet at a target reach rate of 4 reaches/min and <5% maximum pulling force (MPF; the 45 mg food pellet was estimated as <5% MPF) in customized operant behavioral apparati (Med Associates, St Albans, VT), as described previously [5], [6]. Force-dependent changes were examined by having additional rats perform a high repetition high force task (HRHF) at a target rate of 4 reaches/min and 60% maximum pulling force. This task was a forelimb handle-pulling task for a food reward that used customized operant behavioral chambers as described previously [7], [17], [18].

Ibuprofen: At the end of the 4th week of HRHF task performance, 12 of the HRHF rats were administered liquid ibuprofen (Children's Motrin Grape Flavored, Johnson & Johnson, New Brunswick, New Jersey) in drinking water daily (45 mg/kg body weight), as described previously [42]. These animals continued to perform the HRHF task regimen for two more weeks while receiving the ibuprofen treatment (6HRHF+IBU). Nine of the TR-HF rats (TR-HF+IBU) and 9 normal controls (NC+IBU) also received ibuprofen for 2 weeks. The ibuprofen dose used was moderate, yet effective in reducing HRHF-induced increases in tissue cyclooxygenase 2 and pro-inflammatory cytokines in our rat model [42]. The amount of medicated water consumed/day was tracked for each animal by measuring the difference between the initial and final volume of suspended solution daily. Blood was collected, centrifuged, and serum levels of ibuprofen were tested using National Medical Services (NMS, Willow Grove, PA). By these measures, it was determined that all ibuprofen treated rats had between 48.8±6.3 mg/kg body weight of ibuprofen daily.

Anti-TNF-α: Another subcohort of HRHF rats were administered an anti-rat TNF-α (CNTO 1081; generously provided by Janssen R&D), as described previously [12]. Briefly, this drug was injected intraperitoneally (i.p., 15 mg/kg body wt.) into HRHF rats beginning mid-task week 4, at the end of task week 4, and at the end of task week 5 (6HRHF+anti-TNF, n = 6). Six of the TR-HF rats received similar anti-TNFα treatment (TR-HF+anti-TNF).

Daily exposure (reaches/day) was obtained from: (a) 22 randomly chosen HRNF animals at the end of task weeks 1 and 3 (n = 22 each), and weeks 6 and 9 (n = 18 each) of HRNF task performance [the number reduces across weeks due to euthanasia for tissue analysis]; and (b) 14 randomly chosen untreated HRHF animals at the end of task weeks 1 and 3 (n = 14 each), and weeks 6 and 9 (n = 7 each) of HRHF task performance. The mean daily exposure, reaches/day, was determined by multiplying the reach rate by the number of minutes per day that the rats participated in the task.

Average grip strength was tested using previously described methods [6] in: (a) 24 randomly chosen HRNF animals at the naïve time point, at the end of training (week 0; TR), and in weeks 1–5, and weeks 6–9 (n = 18 each) of HRNF task performance; and from (b) 38 randomly chosen untreated HRHF animals at the naïve time point, at the end of training (week 0; TR), and in weeks 1–4, weeks 5 and 6 (n = 35), and weeks 7–9 (n = 9) of HRHF task performance. Average grip strength was also determined in NC+IBU (n = 9), TR-HF+IBU (n = 9), and 6HRHF+IBU rats (n = 12). Grip strength has already been reported for anti-TNF-α treated rats [12].

Fifteen animals were euthanized with an overdose of sodium pentobarbital (120 mg/kg body weight) and forearm flexor muscles (and associated proximal tendon slips) were collected from Region A of the flexor digitorum mass as shown in Figure 1A from NC, TR-NF, and 3-, 6- and 9-week HRNF rats (n = 3/group). Each muscle belly was divided into two longitudinal parts; half was used for RNA extraction and half for protein extraction. The half for RNA extraction was put into RNAlater RNA Stabilization Reagent (QIAGEN, Valencia, CA) for 2 hours at room temperature, and then stored at −80°C. Total RNA was isolated using TRIzol reagent (Invitrogen, Carlsbad, CA). The concentration of each RNA sample was determined using a spectrophotometer; the integrity was monitored on 1% formaldehyde denatured gels. After confirming RNA integrity, cDNA was prepared from mRNA extracts from the above tissue samples (n = 3/group), using a High Capacity cDNA Reverse Transcription kit (Applied Biosystems™, Foster City, CA). PCR primer sets for the following were used: Collagen type I alpha1 (Col1; Cat# PP42922APPR42922A), CTGF (Cat# PPR46426A) and GAPDH (Cat# PPR51520A) (SABiosciences, Frederick, MD). qPCR was then performed in duplicate for 20 µl reactions, using the SYBR Green PCR Master Mix method (Applied Biosystems, Foster City, CA) on an ABI 7500 Fast Real-Time PCR system (Applied Biosystems™). PCR cycles consisted of an initial cycle of 50°C for 2 min and the second cycle of 95°C for 10 min, followed by a two-step program of 95°C for 15 sec and 60°C for 1 min for 40 cycles. Using GAPDH as the internal control, relative gene expression among samples was calculated by comparison of C_t (threshold cycle) values. A dissociation curve was checked for each qPCR run to confirm specific amplification of target RNA.

A CTGF cDNA fragment of approximately 700 bp of the 3^′ end of the rat CTGF cDNA was cloned into PCR-Script vector (Stratagene, LaJolla, CA) flanked by the T3 and T7 promoters to generate CTGF sense and CTGF anti-sense riboprobes. Concentration of the sense and anti-sense riboprobe was determined using an in situ hybridization labeling kit (Roche Diagnostic). For in situ hybridization of CTGF in flexor digitorum muscles, animals were euthanized and flexor digitorum muscles and tendons from NC and 9-week HRNF rats were harvested (n = 3/group), fixed in 4% paraformaldehyde in phosphate buffered saline (PBS) at 4°C, dehydrated, embedded in paraffin, sectioned (5 µm sections), and placed on charged slides (Super Frost Plus, ThermoFisher Scientific, Pittsburgh, PA). In situ hybridization was then carried out as described previously [43].

Animals received an overdose of sodium pentobarbital (120 mg/kg body weight) before being perfused transcardially with 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4): NC (n = 6); TR-NF (n = 5); TR-HF (n = 10), TR-HF+IBU (n = 4), TR-HF+anti-TNF (n = 3); 3-week HRNF (n = 7); 6-week HRNF (n = 5); 9-week HRNF (n = 10); 6-week HRHF (n = 6); 9-week HRHF (n = 4); 6HRHF+IBU (n = 7); and 6HRHF+anti-TNF (n = 3). Tissues were collected and postfixed “en bloc” by immersion overnight. A proximal portion of the muscle mass was then removed with a scalpel for cross-sectional sectioning (as shown with vertical dashed lines in Fig. 1A), while the remaining flexor digitorum muscles and tendons were separated en bloc as a flexor mass from the bones for longitudinal sectioning (included both Regions A and B as shown in Fig. 1A). All tissues were cryoprotected in 30% sucrose in PBS before frozen-sectioning using a cryostat into 15 µm longitudinal or cross sectional slices. Sections were then placed onto charged slides (Fisher, Super Frost Plus) and allowed to dry overnight before storage at −80°C.

Sections on slides were treated with 3% H₂O₂ in methanol to block for endogenous peroxidase for 30 min, washed in PBS, and then blocked with 10% goat serum in PBS for 20 min at room temperature. For localization of OA, sections were first permeabilized with 0.05% pepsin in 0.01N HCL, prior to blocking with 10% goat serum in PBS for 20 min, and prior to primary antibody incubation. Sections were then incubated overnight at room temperature with primary antibody diluted in 10% goat serum in PBS. Primary antibodies were as follows: anti-CTGF (Cat# sc-14939, 1∶400 dilution, Santa Cruz Biotechnology, Santa Cruz, CA), anti-collagen type I (Cat# C2456, 1∶500 dilution, Sigma-Aldrich, St. Louis, MO), and TGFB1 (Cat# MAB240, 1∶300 dilution, R&D Systems). On the 2^nd day, after washing, sections were incubated with the appropriate secondary antibody conjugated to HRP (Jackson Immunoresearch Laboratories, West Grove, PA; each were diluted 1∶100 in PBS, and incubated 2 hours at room temperature before washing in PBS) and visualized using DAB with or without cobalt (black versus brown, respectively; Fast DAB, Sigma), or with appropriate secondary antibody conjugated to Cy2 (green fluorescent tag) or Cy3 (red fluorescent tag) (Jackson Immunoresearch Laboratories). DAB-treated sections were counterstained lightly with eosin. Slides were either dehydrated and coverslipped with DPX mounting medium for bright field microscopy (for HRP-DAB) or washed with PBS and coverslipped with 80% glycerol in PBS for epifluorescence microscopy. Negative control staining was performed by omitting either the primary antibody or the secondary antibody. Specificity of the TGFB1 antibody was determined via the use of a TGFB1 blocking (Santa Cruz Biotechnology Cat# sc-146 P). Western blot analysis, using methods described previously [12], was used to show that proteins of the correct molecular weight were immunodetected with the CTGF and collagen type I antibodies in skeletal muscle (as shown in the results section and figures). Quantification of the immunohistochemical findings was performed using an image analysis program (Bioquant, Nashville, TN) and methods, as described previously [17], [19]. Other subsets underwent Masson's trichrome staining as described [44].

Animals were euthanized with an overdose of sodium pentobarbital (120 mg/kg body weight) before tissue collection from Region A of the flexor digitorum mass (Fig. 1A) from NC (n = 8), NC+IBU (n = 5), TR-NF (n = 5), TR-HF (n = 5), TR-HF+IBU (n = 5), TR-HF+anti-TNF (n = 3), 6-week HRNF (n = 6), 6-week HRHF (n = 5), 6HRHF+IBU (n = 5), and 6HRHF+anti-TNF (n = 3). As described above, the muscles were divided into half longitudinally. The half for protein analysis was snap frozen and stored at −80°C until homogenization, and then prepared for protein analysis, as described previously [6], [12]. Total protein was determined using BCA-200 protein assays (Bicinchoninic Acid, Pierce, Rockford, IL). For ELISA, tissue lysates (50 microliter aliquots) were analyzed using a commercially available ELISA kit for TGFB1 (Cat# PA1-9574, Pierce), according to the manufacturers' protocol. Each sample was run in duplicate. ELISA assay data (pg protein) were normalized to µg total protein.

Repeated measures, mixed model, two-way ANOVAs were used to analyze grip strength, with the factors group (HRNF and HRHF) and week (naïve, 0–9). Two-way ANOVAs, with the factors group and week, were used to analyze daily exposure and CTGF. Univariate ANOVA was used to analyze quantitative PCR and TGFB1 ELISA for differences in task weeks versus controls. Univariate ANOVA was also used to examine for drug effects on CTGF, TGFB1 and grip strength, to examine for any significant changes between controls, task and treatment groups. Statistical analyses were performed using Prism 5 (GraphPad Software, La Jolla, CA). All post hoc analyses were carried out using the Bonferroni test for multiple comparisons as indicated; adjusted p values are reported. An adjusted p value of <0.05 was considered significant for all analyses. Data are presented as the mean ± standard error of the mean (SEM).

Daily exposure showed significant declines towards the target of 480 reaches/day with continued task performance (Fig. 1B; 2-way ANOVA: week, p = 0.003; group, p = 0.01), suggestive of skill acquisition, decreased performance abilities, or a combination of the two. Specifically, in both groups, rats tended to overreach initially. By week 9, reaches/day declined below the target in HRNF rats, compared to HRNF week 1 (p<0.01), suggesting skill acquisition because it declined toward their target reach rate. In HRHF rats, reaches/day declined by week 9, but not significantly, compared to HRHF week 1, suggesting that rats in this higher force group are self-regulating their reach rate due to discomfort (as shown in Fedorczyk et al, 2010). Nevertheless, since reaches/day were similar statistically between the groups, force level rather than reaches/day was a key difference between the two groups (<5% maximum pulling force in HRNF compared to 60% in HRHF tasks rats).

Grip strength showed significant declines with continued task performance in both groups (Fig. 1C; 2-way ANOVA: week, p<0.001; group, p = 0.03). Specifically, TR-NF rats did not have declines in grip strength compared to naïve rats (indicated as 0(TR) in Fig. 1C), but TR-HF rats showed significant declines by the end of their training period, compared to naïve rats (p<0.01). Each group showed declines in grip strength with continued task performance, compared to naïve rats (See Fig. 1C for p values). Compared across groups, 9-week HRHF rats had the greatest grip strength declines, compared to 9-week HRNF rats (p<0.05; Fig. 1C). Thus, both force level and chronicity of exposure contributed to declines in grip strength.

Since we have shown that administration of the anti-rat TNF-α during the early stages of inflammation attenuates HRHF task-induced declines in grip strength [12], we sought to determine here if ibuprofen treatment had a similar effect. Grip strength showed significant declines in each group with continued task performance (Fig. 1D; 1-way ANOVA: p<0.001). Specifically, grip strength declines observed in TR-HF rats, compared to NC (p<0.01), were attenuated in TR-HF+IBU rats. Likewise, grip strength declines observed in 6-week HRHF rats, compared to NC rats (p<0.001), were attenuated in 6-week HRHF+IBU rats (p<0.05 compared to 6-week HRHF). The 6-week HRHF+IBU rats had grip strengths similar to NC and NC+IBU rats (Fig. 1D).

We initially explored if there were changes in expression of CTGF and collagen type I in rats performing the HRNF task, and found that each increased with continued HRNF task performance (Fig. 2A–D). Using qPCR analyses, we observed a significant upregulation of CTGF expression in 6- and 9-week HRNF, compared to TR-NF rats (Fig. 2A, p<0.01 each) and a significant upregulation of Col1 expression in 3-, 6- and 9-week HRNF, compared to TR-NF rats (Fig. 2B; p<0.05, p<0.001, and p<0.01, respectively). In situ hybridization was used to confirm the CTGF gene expression and showed increased CTGF in 9-week HRNF (Fig. 2D; note small dark stained cells on periphery of myofibers and in surrounding connective tissues), compared to NC muscles (Fig. 2C).

Immuohistochemistry was used to determine if CTGF protein levels also increased, as well as the cellular and spatial location of the CTGF protein. HRNF flexor digitorum muscles showed only a small but non-significant increase in CTGF immunostaining in 9-week HRNF, compared to NC rats (Fig. 3B versus Fig. 3A; Fig. 3E). Therefore, we extended our study to include a higher force demand task (i.e. the HRHF task), and found increased CTGF in large mast-like cells near blood vessels (arrowheads in Fig. 3C), and in many small cells at edges of myofibers (Fig. 3D). Quantification of the immunostaining showed increased CTGF immunoreactivity in 9-week HRHF muscles, compared to NC and compared to 9-week HRNF rats (p<0.001 each; Fig. 3E). Figure 3F shows that the anti-CTGF antibody detected a 38 kDa protein, the expected molecular weight of CTGF, in a western blot of flexor digitorum skeletal muscle from a HRHF animal.

qPCR findings for collagen type I were confirmed qualitatively using immunohistochemistry (Fig. 4). There was no collagen type I immunostaining in or around NC myofibers (Fig. 4A). However, collagen type I immunostaining was visible around myofiber slips at sites where muscles attached to tendons in 9-week HRNF rats (Fig. 4B and inset). Collagen type I immunostaining was even more pronounced in flexor digitorum muscle of 9-week HRHF rats in the endomyseum (Fig. 4C), small cells around myofibers, and within some individual myofibers (Fig. 4D). Figure 4E shows that the anti-collagen 1 type antibody detected bands that matched the expected molecular weights of collagen type 1 precursors and mature collagen type I, in western blots of flexor digitorum skeletal muscles from HRHF rats, matching previously published molecular weights for collagen type 1 [45], [46].

Since TGFB1 has been shown to be an upstream modulator of CTGF and collagen production in myofibroblasts [32], [41], and has been implicated in muscle fibrosis [41], we looked for its increase in the flexor digitorum muscles (Fig. 5). Immunohistochemistry showed little to no TGFB1 immunostaining in NC rats (Fig. 5A). In contrast, a clear increase of small cells immunopositive for TGFB1 were observed at the periphery of myofibers and in the endomyseum between myofibers in 6-week HRHF rats (Fig. 5B). ELISA was used for quantification of TGFB1 protein levels and revealed significant increases of TGFB1 in TR-HF and 6-week HRHF flexor digitorum muscles, compared to NC rats (p<0.05 each; Fig. 5E). A two-week treatment of HRHF rats with ibuprofen treatment reduced TGFB1 immunoexpression, compared to untreated 6-week HRHF rats (p<0.05; compare Fig. 5C to 5B; Fig. 5F). A two week treatment of HRHF rats with anti-TNF-α treatment also reduced TGFB1 immunoexpression, although not significantly (compare Fig. 5D to 5B; Fig. 5F). TGFB1 levels in 6HRHF+antiTNF and 6HRHF+IBU were not significantly different from NC levels (Fig. 5F).

Since production of CTGF by fibroblasts is also regulated by TNF-α [37], [38], we examined the effects of both anti-TNF-α and ibuprofen on CTGF and matrix deposition in the flexor digitorum muscles. Both treatments decreased the initial HRHF-training and HRHF-task induced increases in CTGF and collagen matrix (Fig. 6). TR-HF rats had small increases in CTGF immunoreactivity (Fig. 6A), but little staining was seen in TR-HF-anti-TNF rats (Fig. 6B), although neither change was significantly different from NC (Fig. 6G). The amount of CTGF in TR-HF+IBU was also not significantly difference from that in NC (Fig. 6G). In contrast, CTGF immunoreactivity was greater in untreated 6-week HRHF muscles, compared to NC rats (Fig. 6C,F,G; p<0.01). However, CTGF staining was significantly lower in 6-week HRHF rats treated with either anti-TNF-α or ibuprofen (6HRHF+anti-TNF and 6HRHF+IBU rats), compared to untreated 6-week HRHF rats (Fig. 6D–G; p<0.01 each). Lumbrical muscles of the palm also showed decreased collagen matrix staining in 6HRHF+anti-TNF rats compared to 6-week HRHF rats (compare Fig. 6I to 6H). Similar decreases in collagen matrix staining were observed lumbricals of 6HRHF+IBU rats (data not shown).