The aim of this study was to examine the effect of hindlimb suspension (HS) on the expressions of COL1A2 (type I collagen α2 chain) mRNA and its regulatory factors, transforming growth factors (TGF)-β1, -β2, and -β3, phosphorylated Smad3, and tumor necrosis factor-α (TNF-α) in rat hindlimb muscles. Forty-eight male Wistar rats (age, 5 wk) were randomly assigned to HS for 1, 3, 7, and 14 days and control (n = 6 for each). During the exposure to HS, COL1A2 mRNA expression decreased in the soleus muscle at day 3 and recovered to control level at day 7. The content of TNF-α, one of the negative regulatory factors for COL1A2, increased from day 3 until day 14. On the other hand, the contents of TGF-β1, TGF-β3, and Smad3, positive regulatory factors for COL1A2, increased at day 7. The in situ hybridization for COL1A2 and the immunohistochemistry of TGF-β1 and TNF-α revealed their expressions around nerve-related tissues, including muscle spindles and connective tissue sheath. The results indicate that the transcriptional activity of COL1A2 in the soleus muscle initially decreases in response to unloading through an increase in TNF-α production; thereafter, it returns toward normal level through the activated TGF-β/Smad pathway.
- muscle atrophy
- extracellular matrix
skeletal muscle shows a rapid adaptation to mechanical environment, such as unloading and overloading. Unloading the hindlimb muscles of the rat by hindlimb suspension (HS) has been shown to cause biochemical and physiological changes, which are similar to those associated with spaceflight (1). Muscle-fiber atrophy takes place in mostly slow-twitch fiber-specific fashion (10, 25, 31, 32). In addition, muscle fiber type is shifted from oxidative to glycolytic subtypes (10, 12, 25).
In muscles, connective tissues, such as endomysium and perimysium, are important for the transmission of force and the structural support (6). Type I collagen, one of the major proteins in these connective tissues, has a triple helical structure composed of two α1(I) chains and one α2(I) chain (8, 27). This characteristic structure of the molecule gives the muscle a high-tensile strength and a limited elasticity, when aggregated into fibrils (19). Thus changes in the synthesis of type I collagen would have a physiological significance in the adaptation of muscle to mechanical environment.
When hindlimb muscles of the rat are cast immobilized, the expression of type I procollagen mRNA has been shown to decrease at day 3 and returns to control level at day 7 (13). Also, it has been reported that the activities of prolyl 4-hydroxylase and galactosylhydroxylysine glucosyltransferase, both of which are enzymes for collagen biosynthesis, and the concentration of hydroxyproline in the soleus muscle decrease after 3 and 42 days of cast immobilization (30). On the other hand, different responses have been shown to occur when the muscles are subjected to denervation in combination with cast immobilization for 1–3 wk. The specific activity of prolyl 4-hydroxylase in the gastrocnemius muscle increases and that of galactosylhydroxylysine glucosyltransferase increases in both gastrocnemius and soleus muscles. The muscle hydroxyproline concentration also increased in both muscles (30). These studies suggest that the biosynthesis of type I collagen is controlled by not only the mechanical stress but also the neuronal factors.
In regard to unloading, it has been reported that 1, 14, and 28 days of HS do not affect the type I collagen mRNA synthesis in the rat soleus muscle (23); however, the type I collagen mRNA expression may respond rapidly to unloading, e.g., within 1 wk of HS. In addition, even though previous studies have focused on the proteins related to the posttranscriptional modification of collagen (13, 30), the role of transcriptional regulators for collagen in atrophy models is not yet well understood.
Among upstream regulators of type I collagen α2 (COL1A2), the transforming growth factor (TGF)-β/Smad pathway is a strong potentiator for the transcription of COL1A2 expression (2, 7, 14). On the other hand, TNF-α is known as an antagonist of TGF-β signaling (15, 16). The cross talk between TGF-β and TNF-α has been thought to be one of the critical mechanisms for the synthesis of collagen and other components of extracellular matrix (17). Here, we hypothesize that the TGF-β/Smad pathway and TNF-α are affected by unloading, in association with changes in the type I collagen mRNA expression. To examine the hypothesis, the present study investigated changes in the COL1A2 mRNA expression and the contents of the TGF-βs (-1, -2 and -3), phosphorylated Smad3, and TNF-α in the rat soleus muscle. Also, the abundance of TGF-β receptor I and activin receptor I, which is also a TGF-β receptor (28), was measured. The distributions of COL1A2 mRNA expression, TGF-β1, and TNF-α in lower limb muscles were detected. From the obtained results, we discuss the possible mechanisms of the regulation of type I collagen mRNA transcription, depending on the HS unloading.
MATERIALS AND METHODS
Male Wistar rats (age, 5 wk; body weight, 110–120 g) were used. They were housed in an animal room with regulated temperature (20°C), humidity (60%), and illumination cycles (12-h light and 12-h dark). They were allowed to feed on commercial rat chow (CE-7 CLEA, Japan) and drink water ad libitum. The study was approved by the Ethical Committee for Animal Experiments at the University of Tokyo.
The rat was subjected to HS model, according to Morey et al. (24), with some modifications. Forty-eight rats were divided into six for each HS and control time course group. After the rat was anesthetized with diethyl ether, a 0.5-mm stainless wire was inserted through the tail. The wire was attached to a rota-table (360°), and then its length was adjusted so that the hindlimbs did not touch the cage bottom. The rats could walk around with their forelimbs and get food and water ad libitum.
The rats were anesthetized with nenbutal and killed. The soleus muscles of hindlimbs from both control and HS groups were excised, weighed, and frozen in liquid nitrogen. The samples were stored at −80°C until biochemical analysis and in situ hybridization and immunohistochemistry.
For total RNA preparation, frozen samples were pulverized in chilled mortars. Muscle powder was then transferred to sterile tubes containing 1 ml of Isogen (Nippon gene Japan) and homogenized with a polytron homogenizer.
After homogenization, 0.2 ml of chloroform were added to each tube and mixed thoroughly. Specimens were then centrifuged at 12,000 g for 15 min at 4°C. The aqueous phase was transferred to a fresh tube, and 0.25 ml of isopropanol and 0.25 ml of 0.45 M NaCl were added. After incubation for 10 min at room temperature, samples were centrifuged at 12,000 g for 10 min at 4°C. The supernatant was discarded, and 80% ethanol was added for washing the precipitate. After centrifugation, precipitate was dried and dissolved in 50 ml of 0.05M Tris·Cl, pH 8.0, containing 50 mM EDTA. Total RNA was quantified by measuring absorbance at 260 nm. For Northern blot analysis, 5 μg total RNA were incubated in a buffer containing formaldehyde for 5 min at 65°C and then cooled on ice. RNA samples were loaded onto a 1% agarose gel, electrophorated, and then transferred overnight to nylon membrane by upward capillary action. The membrane was exposed to UV in the presence of UV cross-linker (ATTO, Japan) and then hybridized with digoxigenin (DIG)-labeled cRNA probes.
For the type I collagen probe, both 5′ and 3′ ends of the clone corresponded to residues of mRNA sequences for rat α2 type (I) collagen. The cRNA probes [300 nucleotides (nts)] were made as follows: cDNA sample of rat soleus muscles was obtained by using superscript II (Stratagene). Primers (forward: 5′-taagggagaaaatggcatcg-3′, reverse: 5′-ccttctttaccagcagcacc-3′) were used for PCR to amplify target regions. A T7 phage RNA polymerase promoter was added to obtained PCR fragments by using the Lig'nScribe kit (Ambion). Obtained fragments were sequenced to confirm that a target sequence was correctly obtained. The cRNA probe labeled with DIG was obtained by using the RNA labeling kit (Roche). The membrane was also hybridized for the 18S ribosomal RNA with DIG-labeled RNA probe (359 nts, forward: 5′-tatggttcctttggtcgctc-3′ reverse: 5′-cttggatgtggtagccgttt-3′). The membranes were prehybridized in hybiridization solution (Ultrahyb, Ambion) for more than 1 h at 65°C. Then DIG-labeled cRNA probe was added and hybridized for more than 18 h at 65°C. After hybridization, the membranes were washed twice in 2× saline sodium citrate (SSC) and 0.1% SSC at room temperature. High-stringency washing was performed twice at 65°C for 15 min in 0.1× SSC and 0.1% SDS. After the membranes were briefly washed with 2× SSC, the membrane was incubated in 0.1 M mareic acid containing 0.15 M NaCl and 1% blocking reagent (Roche Diagnostics) at room temperature for 30 min. The peroxidase-conjugated antibody against DIG (Roche Diagnostics) was added to that solution, and incubation was continued for another 60 min. After the membrane was washed with 0.1 M mareic acid containing 0.15 M NaCl and 0.3% Tween 20 to remove nonspecific binding antibodies, CDP-star (Roche Diagnostics) was added to obtain chemiluminescent signals.
To quantify COL1A2 mRNA expression, its band intensity was measured with Light Capture (ATTO) and expressed relative to that of 18S ribosomal. The ratio between COL1A2 and 18s ribosome in each experimental group was further normalized to that of the control group, as previously reported (18).
The specimens were homogenized in 10 mM Tris·HCl (pH 7.4), 0.1% SDS. The homogenate was incubated at 80°C for 10 min, then centrifuged at 15,000 g (4°C) for 15 min. The supernatant was removed, and its protein content was determined using a protein determination kit (Bio-Rad, Richmond, CA). Then equal amounts of extracted muscle proteins (30 μg total protein) were mixed with sample buffer, boiled, and separated on SDS-polyacrylamide gel (10%) electrophoresis. The proteins were then electrically transferred to pure nitrocellulose membranes (Scheleicher and Schell, Dassel, Germany). The membranes were blocked with blocking solution (Roche). All primary antibodies were incubated with blotted membranes overnight at 4°C. Antibodies used for Western blotting were as follows: rabbit polyclonal anti-TGF-β1, rabbit polyclonal anti-TGF-β2, rabbit polyclonal anti-TGF-β3, rabbit polyclonal anti-TGF-β receptor I, rabbit activin receptor I, rabbit polyclonal anti-phosphorylated Smad3, and rabbit polyclonal anti-TNF-α (all were from Santa Cruz, Biotechnology, Santa Cruz, CA), and diluted to 1:100 (vol/vol). Membranes were washed for 5 min three times at reverse transcriptase. The secondary horseradish peroxidase-conjugated goat anti-rabbit IgG was used at a dilution of 1:1,000 (Santa Cruz Biotechnology). After several washes in PBS, the blots were developed by using Enhanced Chemiluminescence Western Blotting Analysis System (Amersham) and were imaged and quantified with the Light Capture (ATTO). The band densities were expressed relative to those in control.
The stored tissues at −80°C were preincubated for ∼1 h at −20°C before use. Sections (10 μm) were cut with a cryostat at −20°C, fixed with 4% paraformaldehyde in 0.1 M phosphate buffer for 15 min, and then washed with 0.1 M phosphate buffer in saline (PBS; pH 7.4). To block any nonspecific reaction, the sections were incubated with 0.1 M PBS containing 10% normal serum and 1% Triton X-100 for 1 h. The sections were then incubated overnight at 4°C with the primary antibodies. The primary antibodies, a rabbit polyclonal anti-TGF-β1, a rabbit polyclonal anti-TNF-α, a rabbit polyclonal anti-TNF-α receptor I, a rabbit polyclonal anti-TGF-β receptor I, a rabbit polyclonal anti-TGF-β receptor II (Santa Cruz Biotechnology), a mouse monoclonal anti-dystrophin (Sigma, St. Louis, MO), and a rabbit polyclonal laminin (Santa Cruz Biotechnology), were diluted with 0.1 M PBS containing 5% normal serum and 0.3% Triton X-100. After being washed with PBS, the sections were incubated overnight at 4°C with the secondary antibodies (horseradish peroxidase-conjugated anti-rabbit and anti-goat IgG). The secondary antibodies were diluted with 0.1 M PBS containing 5% normal serum and 0.1% Triton X-100. The sections were washed in 0.1 M PBS and mounted in Vectashield mounting medium with 4,6-diamidino-2-phenylindole (Vector Laboratories) to visualize the nuclei.
In situ hybridization.
The sections (10 μm) were fixed with 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) for 15 min and were washed twice with PBS for 5 min. The sections were incubated in 0.2 N HCl for 8 min and then washed with PBS for 5 min twice. The sections were incubated in 20 μg/ml proteinase K at 37°C for 5 min and then washed in PBS for 5 min. The sections were incubated in 0.1 M triethanolamine for 2 min. The sections were then incubated in 0.25% acetic anhydride in triethanolamine for 10 min, were incubated in 1× SSC twice, and then air dried for 10 min. Hybridization buffer (50% formamide, 2× SSC, Denhardt's, 10% dextran sulfate, 0.5 mg/ml yeast tRNA, and 0.5 mg/ml salmon sperm DNA) containing probe was dispersed on the sections. The sections were incubated at 55°C overnight and were washed in 2× SSC at 45°C for 10 min, and then washed with 1× SSC containing 50% formamide at 55°C for 20 min. The sections were washed in 0.2× SSC at 55°C for 10 min, were incubated in 100 mM Tris·HCl (pH 7.5) containing 150 mM NaCl for 1 min, then were blocked with 1.5% blocking reagent (Roche) in 100 mM Tris·HCl (pH 7.5) containing 150 mM NaCl for 30 min. The sections were incubated with anti-DIG-antibody (1:500 with 1.5% blocking reagent in 100 mM Tris·HCl, pH 7.5, containing 150 mM NaCl) for 1 h and were washed in 100 mM Tris·HCl (pH 7.5) containing 150 mM NaCl for 15 min. The slides were washed in 100 mM Tris·HCl (pH 9.5) containing 150 mM NaCl and 50 mM MgCl2 for 3 min and were incubated in nitroblue tetrazolium (450 μg/ml) and 5-bromo-4-chloro-3-indolyl phosphate (175 μg/ml) in 100 mM Tris·HCl (pH 9.5) containing 150 mM NaCl and 50 mM MgCl2 for color development. The sections were washed with 10 mM Tris·HCl (pH 7.5) containing 1 mM EDTA to stop the reaction and then were washed with water for 5 min.
Variables are expressed as means ± SD. Statistical analyses were performed with two-way ANOVA, followed by Tukey's post hoc test. P < 0.05 was regarded as significant.
Changes in the weight of animals and soleus muscle.
Changes in body weight and soleus muscle weight after 1, 3, 7, and 14 days of HS are shown in Table 1. During HS, body weight relative to preexperiment was significantly smaller than that of age-matched control in each point of the time course. After 14-day HS, the soleus muscle weight relative to body weight in the HS group was also smaller than that in the control group. In addition, the relative soleus muscle weight gradually decreased as HS proceeded.
Changes in COL1A2 mRNA.
During HS, the content of COL1A2 mRNA relative to that of 18S mRNA initially decreased for a few days and thereafter returned toward the original level (Fig. 1). The content after a 3-day HS was significantly smaller than that in control and after a 14-day HS.
Changes in TNF-α, TGF-β1, -β2, and -β3, and Smad3.
HS resulted in a significant increase in the TNF-α protein content in the soleus muscle on 3, 7, and 14 days of HS compared with all control groups and 1-day HS (Fig. 2).
On the other hand, the content of TGF-β1 protein, known as a strong potentiator for collagen synthesis, appeared to increase on day 7 and then decrease again to the control level on day 14 (Fig. 3). Other TGF-βs, TGF-β2 and -β3, showed a similar pattern of abundance to that of TGF-β1 (Figs. 4 and 5). The content of TGF-β3 in the 7-day HS was significantly different from that of the 1-day HS. The content of phosphorylated Smad3 appeared to change during HS in a similar manner to that of TGF-β1 (Fig. 6). The value on 7-day HS was significantly different from that of control. Moreover, the value at day 7 was significantly larger than that at day 14, indicating a transient increase in TGF-β during HS.
Changes in TGF-β receptor I and activin receptor I.
In situ hybridization of COL1A2.
In situ hybridization of COL1A2 for transverse sections of the soleus muscle showed that positive stainings were located around nervous tissues, blood vessels, and muscle spindles (MS) (Fig. 9). No considerable differences were seen between staining pattern in the control group and those in the HS groups. This indicates that, regardless of HS, the synthesis and turnover of type I collagen are higher in those tissues and organs than in other loci within the soleus muscle.
Immunohistochemistry for TNF-α, TGF-β1, TNF-α-receptor, TGF-β-receptor I, and TGF-β-receptor II.
Immunohistochemistry for TNF-α, TGF-β1, and receptors showed the positive signals around veins, MS, and neurons (Figs. 10, 11, and 12). The localizations of TNF-α, TGF-β1, and receptors were similar to that of COL1A2 mRNA shown by the in situ hybridization; as for COL1A2 expression, no differences were recognized between staining patterns in control and those after the 7-day HS.
The synthesis of type I collagen within skeletal muscle should play an important role in the maintenance of tissue tensile strength. There has been only one report on the effect of unloading on collagen synthesis in muscle, where the expression of COL1A2 in rat hindlimb muscle was examined after 1, 14, and 28 days of HS. This report showed no change in the expression of COL1A2 during the examined time (23). However, the present study showed a decrease in COL1A2 on the 3 days of HS, suggesting that the expression of COL1A2 responds relatively quickly to the HS. The expression of COL1A2, then, returned to its original level at day 7 and was constant until day 14. Consequently, the expression levels of COL1A2 from day 7 until day 14 did not differ from that in control, and this was consistent with the previous report by Miller et al. (23).
Concerning disuse, another research examined the expression of type I collagen α1 (COL1A1) chain mRNA in the rat hindlimb muscle immobilized at the plantar flexed position. This study showed that the expression of COL1A1 mRNA was reduced at the 3rd day of immobilization; thereafter, it returned to the control level at the 7th day (13). These changes are consistent with the present study on COL1A2, suggesting that the expression of type I collagen mRNA responds in a similar manner to both immobilization and unloading.
The pattern of COL1A2 mRNA expression may be divided into two phases, i.e., decreasing phase (0–3 days) and a recovery phase (3–7 days). Day 3 TNF-α, an inhibitory factor for COL1A2 mRNA expression, increased. On the other hand, the expression of TGF-β1 slightly decreased until day 3 and then increased on day 7. The content of phosphorylated Smad3 showed a similar response pattern to that of TGF-β1, especially after day 3. This increase in TGF-β/Smad signal transduction implies the cue of the rebound of COL1A2 mRNA expression. However, the signal decreased at 14-day HS, even though the expression of COL1A2 mRNA was maintained, which remains to be elucidated. These results suggests that both TNF-α and TGF-β/Smad signal transduction are involved in the present changes in COL1A2 mRNA expression after HS.
TGF-β signals through transmembrane receptors are mediated by phosphorylated cytoplasmic mediators of the Smad family (21, 26). TGF-β1 binds to the TGF-β receptor, and the activated TGF-β receptor interacts with Smad2 and Smad3 to phosphorylate them (9). After phosphorylation, Smad3 forms a heteromeric complex with Smad4. The heterocomplex is then translocated into the nucleus, where they function as transcription factors, binding DNA of the target gene (22).
On the other hand, TNF-α binds and activates two membrane-bound receptors to affect the transcription factor families, activator protein-1 and NF-κB. Activator protein-1 directly interferes at the level of Smad3-DNA interactions (34) to inhibit the TGF/Smad pathway. NF-kB can induce the expression of Smad7, which prevents phosphorylation and nuclear translocation of Smads (5). TNF-α also downregulates the expression of TGF-β receptor II, the TGF-β receptor (36).
Therefore, again, we consider that the increase of TNF-α from the 3rd day to the 14th day of HS in this study suppresses the TGF-β/Smad signaling through the mediators of activator protein-1 and NF-κB. However, on the 7th day of HS, increased TGF-β (and mediators) strengthened the TGF/Smad pathway to activate COL1A2 mRNA production. We will further examine which mediator is critical to the reduction of COL1A2 mRNA in HS to create a preventive way of connective tissues atrophy.
In situ hybridization and immunohistochemistry were performed to elucidate the intramuscular localization of COL1A2 mRNA, TGF-β1, and TNF-α protein. Positive stainings were seen in connective tissues and especially in the nerve-related tissues, such as the nerve endings and the MS in both in situ hybridization for COL1A2 mRNA and immunohistochemistry for TGF-β1 and TNF-α. Localizations of TGF-β in hypertrophied skeletal muscles were observed by Sakuma et al. (29) and Akutsu et al. (3). In those reports, the TGF-β were detected in the connective tissues among muscle fibers. This is consistent with our result in terms of connective tissue. However, in the case of our experiment, TGF-β1 was detected in the connective tissue of MS. This is partially because the condition of our experiment, which is HS, is different from that of previous reports, which is hypertrophied muscle. Anyway, it implies that the synthesis of type I collagen is active around these tissues and organs. Some have shown that physiological and structural changes occur in MS during HS. Tang et al. (33) reported that MS of the rat soleus under 7-day HS caused a marked decrease in spontaneous discharge frequency and the response to ramp-and-hold stretch. Although the effects of HS on the structure of MS has also been reported (35), changes in collagenous tissue in MS have been observed. On the other hand, changes in the structure of collagenous tissue in MS have been studied with another atrophy model. The capsules of MS, in which type I collagen is one of the major components (20), are thickened in the limb muscle of the rabbit after a cast immobilization (11). Together with these previous reports, the present results suggest that changes in type I collagen expression after HS are associated with changes in activities and morphological properties of MS.
In conclusion, the expression of type I collagen mRNA in muscle initially decreased and then returned to its original level at day 7. The first reduction may be mediated by TNF-α, and TGF-β appear to trigger the recovery of type I collagen mRNA. The expression of type I collagen mRNA was localized mainly in nervous tissues and MS, suggesting that changes in type I collagen expression are related to changes in muscle proprioceptive activities after unloading.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society