J. Pineal Res. 2014; 56:175–188
© 2013 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Molecular, Biological, Physiological and Clinical Aspects of Melatonin
Doi:10.1111/jpi.12110
Journal of Pineal Research
Melatonin treatment combined with treadmill exercise accelerates muscular adaptation through early inhibition of CHOP-mediated autophagy in the gastrocnemius of rats with intra-articular collagenase-induced knee laxity Abstract: The purpose of this study was to determine the effects of melatonin intervention on gastrocnemius remodeling in rats with collagenase-induced knee instability. Type VII collagenase was injected into the right knee to induce joint laxity with cartilage destruction. Melatonin (MT; 10 mg/kg) injection was performed twice daily subcutaneously, and treadmill exercise (Ex; 11 m/min) was conducted for 1 hr/day at a frequency of 5 days/wk for 4 wks. The gastrocnemius mass, which was reduced with collagenase injection only (Veh), was increased with collagenase injection with melatonin treatment with and without exercise in the early phase, and the mass in both limbs was significantly different in the Veh compared with the MT group. However, there was an increase in the relative muscle weight to body weight ratio in the Veh group at the advanced stage. Insulin-like growth factor receptor (IGF-IR) was downregulated in the Veh group, whereas IGF-IR was upregulated in the MT and MT + Ex groups. Joint laxity induced enhancement of autophagic proteolysis (LC3 II) in the muscle, which was recovered to values similar to those in the normal control group (Con) compared with those in the MT and MT + Ex groups. Although intra-articular collagenase increased the total C/ EBP homology protein (CHOP) levels at 1 wk and decreased them at 4 wks in the Veh group, CHOP in the nucleus was upregulated continuously. Prolonged melatonin treatment with and without exercise intervention suppressed nuclear localization of ATF4 and CHOP with less activation of caspase-3, at the advanced phase. Moreover, the interventions promoted the expression of myosin heavy chain (MHC) isoforms under the control of myogenin. Concomitant with a beneficial effect of melatonin with and without exercise, step length of the saline-injected limb and the collagenase-injected supporting side was maintained at values similar to those in control rats. Taken together, the findings demonstrate that melatonin with and without exercise accelerate remodeling of the gastrocnemius through inhibition of nuclear CHOP in rats with collagenase-induced knee instability.
Yunkyung Hong1,2,a, Joo-Heon Kim3,a, Yunho Jin2,4, Seunghoon Lee1,2, Kanghui Park1,5, Youngjeon Lee1,6, Kyu-Tae Chang6 and Yonggeun Hong1,2,4,7 1
Department of Rehabilitation Science, Graduate School of Inje University, Gimhae, Korea; 2Cardiovascular & Metabolic Disease Center, College of Biomedical Science & Engineering, Inje University, Gimhae, Korea; 3 Institute of Animal Medicine, College of Veterinary Medicine, Gyeongsang National University, Jinju, Korea; 4Department of Physical Therapy, Graduate School of Inje University, Gimhae, Korea; 5Division of Rehabilitation Medicine, Dongwei Medical Center, Busan, Korea; 6National Primate Research Center (NPRC), Korea Research Institute of Bioscience and Biotechnology (KRIBB), Ochang, Korea; 7Ubiquitous Healthcare Research Center (UHRC), Inje University, Gimhae, Korea Key words: CHOP, knee joint instability, melatonin, muscle remodeling, treadmill exercise Address reprint request to Yonggeun Hong, Department of Rehabilitation Science, Graduate School of Inje University, 197 Inje-ro, Gimhae, Gyeong-nam 621-749, Korea. E-mail: [email protected] Kyu-Tae Chang, National Primate Research Center (NPRC), Korea Research Institute of Bioscience and Biotechnology (KRIBB), 30 Yeongudanji-ro 30, Ochang, Chung-buk 363-883, Korea. E-mail: [email protected] a These authors contributed equally to this work. Received September 30, 2013; Accepted December 3, 2013.
Introduction Skeletal muscle produces voluntary movement, absorbs loading, and provides dynamic joint stability [1, 2]. Shrier [3] proposed that muscle dysfunction and weakness resulting from injury, overuse, inadequate rehabilitation, and inactivity are the primary drivers of osteoarthritis (OA). The muscles absorb most of the forces that load to the knee joint under normal conditions [4], and the remaining forces are transmitted to articular cartilage, leading to redistribution to bone [5]. Abnormal force distribution causes cartilage damage [3], and chronic mechanical stimulation eventually induces OA [6]. OA affects at least 30%
of people aged older than 60 yr and is a major cause of functional disability. With the population aging, the prevalence of OA in the developed world is expected to increase. Moreover, it is anticipated that OA will become the fourth leading cause of disability in the coming decades [7]. Inflammatory joint pathology induced by intra-articular adjuvant injection causes peri-articular muscle atrophy in the early phase [8]. Muscle atrophy refers to wasting of muscle mass due to disuse, denervation, and systemic disease. The skeletal muscle maintains a balance between the anabolic and catabolic machinery under nonstimulus conditions. However, skeletal muscle reprograms gene 175
Hong et al. expression to alter the characteristics of muscle fibers against environmental stimuli through activation of myogenic transcription factors (MRFs) [9, 10], a process termed muscle adaptation or plasticity [11]. Proteolysis through the ubiquitin–proteasome pathway and autophagy is upregulated under atrophic conditions, whereas protein synthesis is downregulated [12]. In genetic screens, two markers, MAFbx and MuRF1, have been shown to be significantly enhanced prior to the initiation of muscle loss in an animal model [13]. The autophagy– lysosome system is also involved in the regulation of muscle mass [14]. Autophagy removes damaged organelles and unfolded proteins, controls the quality of protein folding in the endoplasmic reticulum (ER), and affects DNA stability [15, 16]. The uncontrolled autophagic machinery accumulates abnormal toxic proteins in the myofibers, leading to muscular degeneration [14], and autophagosomes are observed in the muscle of human with myopathies [17]. Korolchuk and colleagues have suggested that cross-talk may be present between the ubiquitin–proteasome pathway and autophagy–lysosome system [18]. The ER controls the synthesis, folding, and assembly of proteins and provides intracellular Ca2+ storage. ER homeostasis is disrupted by various stimuli such as abnormal nutrients level, Ca2+ imbalance, and increased protein synthesis, leading to accumulation of nonfunctional proteins within the ER lumen [19]. Increased chaperone expression enhances the protein-processing capacity to restore equilibrium when ER stress is applied, a process that is referred as the acute unfolded protein response (UPR). Acute UPR activation is an adaptive response to ER stress, whereas sustained UPR with chronic stress regulates ER stress-related cell fate (e.g., apoptosis or autophagy) [20, 21]. ER stress is present in nonpathological skeletal muscle [22]. UPR in the skeletal muscle is activated to adapt and protect against stress after exercise [23]. C/EBP homology protein (CHOP), a transcription factor, is known to be upregulated by ER stress, which is induced in both apoptotic and differentiating myoblasts through caspase-12 activation [24]. In addition to catabolic pathways, activation of PI-3K/ IGF-I/Akt/mTOR signals can prevent muscle atrophy [25] and is required for a normal hypertrophic response [26]. Mechanical loads induce IGF-I release in skeletal muscle in vitro and in vivo [27]. Additionally, Akt/mTOR signaling regulates postnatal muscle size under conditions of increased mechanical loading [28]. Prolonged melatonin treatment prevents muscle atrophy and activates IGF-Imediated hypertrophic signaling in the gastrocnemius of focal cerebral ischemic rats [29]. The pineal hormone melatonin modulates circadian rhythm [30], oxidative stress [31, 32], and inflammation [33]. Previous studies have reported the beneficial effects of melatonin on muscle damage [34–37]. In the present study, we determined the effect of melatonin with and without exercise on muscle remodeling. Our findings show that melatonin with and without exercise suppresses CHOP-mediated cell death, leading to early adaptation in the gastrocnemius of rats with knee joint laxity. 176
Materials and methods Induction of knee joint instability Eight-wk-old male Sprague–Dawley rats weighing 250– 285 g were used for the present study and were obtained from Daehan Link (Hyochang Science, Daegu, Korea). Rats were housed under light–dark conditions with a controlled temperature. All rats were provided water and food ad libitum. Experimental procedures were approved by the ethics committee for animal care and use at Inje University (Approval no. 2010-72). Experimental rats were randomly divided into three groups after intra-articular injection of collagenase: Veh (collagenase injection only); MT (collagenase injection with melatonin treatment); MT + Ex (collagenase injection with melatonin treatment and treadmill exercise). Rats were anesthetized with an intraperitoneal injection of Zoletil (80 mg/kg of body weight) combined with Xylazine (50 mg/kg of body weight). Alcohol was applied to the shaved skin before collagenase injection. Using a Hamilton microliter syringe (Hamilton Company, Reno, NV, USA), 6 lL of highly purified type VII collagenase (Sigma-Aldrich, St. Louis, MO, USA) was injected into the intra-articular space of the right knee to induce destruction of the ligament and cartilage. For comparison, saline solution was injected at the same volume into the left knee [38]. Intervention Melatonin (Sigma-Aldrich) was dissolved in absolute ethanol and further diluted in 0.9% NaCl solution. Rats were subcutaneously injected with the melatonin solution at doses of 10 mg/kg (twice daily at 07:00 and 19:00). A motor-driven treadmill was used for the exercise intervention. The rats were placed on a moving belt and were trained to run in the direction opposite to the movement of the belt. In 2004, Galois and colleagues [39] proposed that slight (30 cm/s for 15 min, total distance of 7.5 km over 28 days) or moderate exercise (30 cm/s for 30 min, total distance of 15 km over 28 days) exhibits beneficial effects on severity on chondral lesions in the rats with anterior cruciate ligament transection. Based on the previous study, we selected approximate moderate exercise intensity as following. The exercise was conducted for 30 min twice daily (at 17:00 and 22:00) at a frequency of 5 days/wk. The speed and inclination of the treadmill were 11 m/min and 0°, respectively. To allow adaptation to the treadmill, pretraining was carried out for 20 min/day twice daily (at 17:00 and 22:00) during 1 wk before collagenase injection. Collection of animal tissues The gastrocnemius muscle was collected under anesthesia, washed with cold phosphate-buffered saline, and frozen at 70°C. Before freezing, the wet muscle weight was measured using an electronic balance, and the weight was compared among the groups.
Melatonin induces muscular remodeling in osteoarthritis Table 1. Oligonucleotide primers used for PCR Gene
Primer sequence (5′–3′)
Size (bp)
GenBank accession no.
ATF4
F: tcc tga aca gcg aag tgt tg R: cat cca tag cca gcc att ct F: agc aga ggt cac aag cac ct R: ctg ctc ctt ctc ctt cat gc F: ggg gag ttt ggt caa tca ga R: ttt gca tag act ggc tga cg F: gta tga ctc cac tca cgg caa a R: ggt ctc gct cct gga aga tg F: tcc caa gct gtg tgt ctc tg R: gtg cca cgt tat gat gat gc F: aca gag gaa gac agg aag aac cta c R: ggg ctt cac agg cat cct tag F: cct ctt act tcc cag ctg cac ctt ct R: act ttc cct gcg tct ttg ctc tga at F: agc ctg cct cct tct tca tct gg R: cac ggt tgc ttt cac ata gga ctc F: acg gtc gaa gtt gca tcc cta aag R: cac ctt cgg tct tgg ctg tca c
174
NM_024403
166
NM_001109986
170
NM_001106395
100
NM_008084
178
NM_052807.2
288
K01463
239
DQ872905
229
DQ872907
263
DQ872906
CHOP FoxO3 Gapdh IGF-IR MHC Ib MHC IIa MHC IIx MHC IIb
RNA isolation and reverse transcription polymerase chain reaction (RT-PCR) analysis Animal muscle tissue was homogenized with TRI reagent (Sigma-Aldrich) to isolate total RNA. Total RNA was quantified using an Optizen 2120 UV spectrophotometer (Mecasys, Daejeon, Korea) and reverse transcribed using reverse transcriptase. cDNA was amplified using specific primers (Table 1). RT-PCR was performed in a Px2 Thermal cycler HBPX2220 (Thermo Electron Corporation, Beverly, MA, USA). The cycling conditions were as follows: a 10 min initial activation step at 94°C, followed by denaturation at 94°C for 30 s, primer-specific annealing temperature (°C) for 1 min, and extension at 72°C for 1 min. PCR products were confirmed by 1.5% agarose gel electrophoresis. The Image J program ver. 1.6 (NIH, Bethesda, MD, USA) was used to quantify the density. Subcellular fractionation Subcellular components were separated by methods modified from those used in a previous study [40]. After removal of the connective tissues, the gastrocnemius was resuspended and homogenized on ice in STM buffer (250 mM sucrose, 50 mM Tris-HCl (pH 7.4), 5 mM MgCl2) supplemented with protease inhibitor cocktail (Roche, Rotkreuz, Switzerland). The homogenates were maintained on ice for 30 min, vortexed at maximum speed for 15 s, and centrifuged at 800 g for 15 min; then, the supernatants were separated from the pellets. The pellet (P0), which contained nuclei and debris, was resuspended in STM buffer, vortexed at maximum speed for 15 s, and then centrifuged at 500 g for 15 min. Then, the pellet (P1) was washed in STM buffer, vortexed at maximum speed for 15 s, and centrifuged at 1,000 g for 15 min. Thereafter, the washed pellet (P2) was resuspended in NET buffer (20 mM HEPES (pH 7.9), 1.5 mM MgCl2, 0.5 M NaCl, 0.2 mM EDTA, 20% glycerol, 1% Triton X-100) supplemented with protease inhibitor cocktail (Roche) to triturate until homogeneous. The pellet (P2) was vortexed at
maximum speed for 15 s and incubated on ice for 30 min. The nuclei were lysed by 10–20 passages through an 18gauge needle. The lysate was centrifuged at 9,000 g for 30 min, and the supernatant was collected as the ‘nuclear fraction’. The supernatant (S0), which contained not only the mitochondrial but also the cytosolic fraction, was centrifuged at 800 g for 10 min. The pellet was discarded, and the supernatant (S1) was collected and centrifuged at 11,000 g for 10 min. The supernatant (S2), which contains the cytosol and microsomal fractions, was precipitated in 100% cold acetone at 20°C for 2 hr and centrifuged at 12,000 g for 5 min. The pellet was resuspended in STM buffer, which was used to analyze cytosolic protein expression. Protein extraction and Western blotting The medial gastrocnemius muscles were lysed in a buffer (135 mM NaCl, 1 mM MgCl2, 2.7 mM KCl, 20 mM TrisHCl (pH 8.0), 0.5 mM Na3VO4, 10 mM NaF, 1% Triton X-100) supplemented with protease inhibitor cocktail (Roche). The lysate was incubated on ice for 30 min and centrifuged at 14,920 g for 10 min at 4°C; then, the supernatant was collected. The concentration of proteins was measured using an Optizen 2120 UV spectrophotometer (Mecasys) and the Bradford protein assay (Bio-Rad Laboratories, Richmond, CA, USA). The proteins were boiled at 95°C for 10 min and mixed with 29 sample buffer. Thirty-microgram protein samples were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), and the separated protein bands on the gel were transferred to a polyvinylidene fluoride membrane (Merck Millipore, Billerica, MA, USA). The membrane was blocked with 5% nonfat milk in Tris-buffered saline with Tween 20 (TBST) for 1 hr at room temperature. Primary antibodies were incubated with blocked membranes overnight. Membranes were incubated with secondary antibody in TBST for 40 min. Finally, specific bands were visualized using a Western blotting luminal reagent (Santa 177
Hong et al. Cruz Biotechnology, Santa Cruz, CA, USA) and quantified using Image J ver. 1.6.
(A)
Footprint measurements Gait measurements were obtained using a modified footprint method [41]. Both the hind toes and soles of the experimental rats were painted with colored inks (salineinjected limb: blue, collagenase-injected limb: red) before placing them in a run/walk, open-ended acrylic container (1 m in length, 0.08 m in width, and 0.15 m in height). A white paper was laid beneath the acrylic container, and the rats walked toward the open end of the acrylic container. To determine significance, measurements between paw prints were made using at least seven legibly marked strides. Step length and width were measured while the unstable limb was weight bearing and the saline-injected limb was in motion. Footprint images were quantified using the Image J program ver. 1.6.
(B)
Statistical analysis Data were collected from at least three repeated experiments and were presented as the mean standard deviation (S.D.) in graphs. Statistically significant differences between groups were assessed using one-way analysis of variance (ANOVA) with the post hoc Tukey’s test. Differences were deemed statistically significant at a P < 0.05. All analyses were performed using the statistical software SPSS ver. 19.0 (IBM, New York, NY, USA).
(C)
Results Because skeletal muscle has a great impact on whole-body metabolic homeostasis [23], we measured not only the wet weight of the gastrocnemius (Fig. 1B) but also the body weight (Fig. 1A) of rats at 1 wk (early) and 4 wks (advanced). Compared with normal control (Con) and vehicle treatment, melatonin intervention with and without exercise was associated with reduced body weight in rats with knee joint instability at 4 wks (aP < 0.05, bP < 0.05). In the early phase, collagenase injection with no treatment attenuated the gastrocnemius weight compared with the ipsilateral side in the Con group (aP < 0.05) and the contralateral limb in the Veh group (*P < 0.05). Melatonin treatment augmented the mass of the unaffected side only compared with Con (aP < 0.05) and Veh (bP < 0.05), indicating a significant difference between the contralateral and ipsilateral sides (*P < 0.05), although no difference was observed in the collagenase-injected limb of the MT group compared with that of the Con group. However, melatonin with exercise induced a significant increase in the mass in both limbs compared with Con (aP < 0.05), Veh (bP < 0.05), and MT (cP < 0.05) groups. Concomitant with this, the value of the muscle mass relative to body weight showed an increase with melatonin with exercise at 1 wk (Table 2; a,b,cP < 0.05). A significant increase in mass was demonstrated only in the Veh group compared with the Con group at the advanced stage (Fig. 1B; aP < 0.05). Although body weight was reduced in the MT and MT + Ex groups (bP < 0.05), 178
Fig. 1. Intra-articular collagenase-induced changes in body weight and gastrocnemius muscle mass. (A) Body weight of rats was measured at 1, 3, 7, 14, 21, and 28 days after collagenase injection. The values are the means S.D. aP < 0.05 versus Con; b P < 0.05 versus Veh. (B) The wet weight of the gastrocnemius of both hindlimbs (unaffected, saline; affected, collagenase injection) was compared among the groups at 1 and 4 wks. The bars represent the means S.D. aP < 0.05 versus Con; bP < 0.05 versus Veh; cP < 0.05 versus MT at 1 and 4 wks, respectively. Student’s t-test was used to assess the significance (*P < 0.05) of differences between the saline- and collagenase-injected sides. (C) IGF-I receptor (IGF-IR) mRNA was significantly downregulated in the gastrocnemius of the unstable knee of rats, whereas melatonin with and without treadmill exercise induced maintenance or an increase in the IGF-IR level compared with Con. aP < 0.05 versus Con; bP < 0.05 versus Veh; cP < 0.05 versus MT at 1 and 4 wks, respectively.
Melatonin induces muscular remodeling in osteoarthritis Table 2. Relative mass values to body weight Con
1 wk Significance 4 wks Significance
Veh
Saline
Collagenase
Saline
Collagenase
Saline
4.55 0.06 – 4.51 0.01 –
4.55 0.06 – 4.51 0.02 –
4.68 0.24 – 5.41 0.21
4.11 0.12
5.32 0.04
5.38 0.24
5.04 0.07
a
MT + Ex
MT
a,*
a
a,b
a
Collagenase
Saline
Collagenase
4.55 0.33
5.34 0.08
5.19 0.09
5.16 0.07
5.27 0.19
5.48 0.15
*
a
a,b
a
a,b,c
a
The values represent the means S.D. ap < 0.05 versus Con; bp < 0.05 versus Veh; cp < 0.05 versus MT in the each limb at 1 and 4 wks, respectively. Student’s t-test was used to compare the saline- and collagenase-injected sides (*P < 0.05).
the relative muscle mass was not significant among the Veh, MT, and MT + Ex groups at 4 wks. No significant difference was also found between the saline- and collagenase-injected limbs among all the groups (Table 2). Regarding the changes in muscle mass, we analyzed IGF-IR expression (Fig. 1C). IGF-IR expression is important for myoblast proliferation and maintenance of normal muscle mass [42]. In this study, IGF-IR level was the lowest in the Veh group at 1 wk after collagenase injection (aP < 0.05). However, IGF-IR was upregulated in the MT and MT + Ex groups (bP < 0.05), and the level was higher with the combined intervention than with melatonin alone (cP < 0.05). The expression of IGF-IR in the Veh group remained at a lower level than that of the Con group at 4 wks (aP < 0.05). Compared with the Veh group, both melatonin alone and melatonin with exercise intervention enhanced IGF-IR levels (bP < 0.05). Particularly, IGF-IR expression was highest with melatonin alone treatment in the advanced phase. We compared protein synthesis and degradation to identify the mechanisms involved in mass alteration during the progression of joint instability. The IGF-I/Akt/ mTOR pathway is the predominant signal to stimulate protein synthesis, leading to hypertrophy [26]. Total Akt expression was significantly increased in the gastrocnemius of the collagenase-injected limb in the Veh and MT groups compared with that in the Con group (Fig. 2A; a P < 0.05). However, the total Akt expression level in the MT + Ex group was recovered to that of the Con group (b,cP < 0.05). Total mTOR expression in the rats with knee laxity was markedly reduced at 1 wk (aP < 0.05). As joint ligaments were continuously damaged, the mTOR level was upregulated by melatonin treatment only (a,b,cP < 0.05). Furthermore, its expression was lower in the Veh and MT + Ex groups than in both the Con and MT groups (aP < 0.05). Although the muscle mass of the MT + Ex group was augmented in the early phase, Akt and mTOR levels were maintained or downregulated. Therefore, we need to further analyze the expression of p-Akt and p-mTOR for confirmation of their activation. We analyzed the key markers of the ubiquitin–proteasome pathway (e.g., atorgin-1/MAFbx and MuRF1) and autophagy (e.g., LC3 II) signal to determine the proteolytic mechanisms involved in the changes in mass (Fig. 2B). MAFbx was significantly upregulated in the MT and MT + Ex groups at 1 wk compared with the Con (aP < 0.05), whereas no significance was shown
among the groups at 4 wks. MuRF1, which leads to breakdown of the structural muscle proteins such as myosin heavy chains (MHCs), was decreased in the MT + Ex group at 1 wk (a,b,cP < 0.05). However, no significant differences among the groups were found in the advanced phase. Autophagosome-bound LC3 expression (LC3 II) was dramatically increased in the Veh group at 1 wk (aP < 0.05). Prolonged melatonin treatment induced upregulation of LC3 II expression in the advanced phase (a,bP < 0.05), whereas melatonin treatment combined with exercise restored it to a level similar to that of the Con group (cp < 0.05). To confirm the activation of autophagy, we analyzed Foxo3 as a negative regulator of autophagy (Fig. 2C). Melatonin treatment with and without exercise intervention induced upregulation of Foxo3 compared with vehicle treatment (bP < 0.05), which was similar to that of Con group at 1 wk. No reduction in Foxo3 expression was noted in the Veh group at 4 wks, whereas the MT and MT + Ex groups showed enhanced Foxo3 expression compared with the Con group (a,bP < 0.05). These results show not only activation of the autophagic machinery through Foxo3 suppression in the Veh group but also inhibition of MuRF-1-mediated ubiquitin–proteasomal proteolysis in the MT + Ex group in the early phase. Next, we examined whether early loss of muscle mass induced by autophagic activation is associated with ER stress in myofibers. Using data from a previous study [22], we selected ATF4 and CHOP as markers of UPR activated by ER stress (Fig. 3). In the early phase, the mRNA levels of both ATF4 and CHOP decreased with vehicle treatment (aP < 0.05) but were restored by melatonin with and without exercise (Fig. 3A; bP < 0.05;). Particularly, the increase in the mRNA levels of both ATF4 and CHOP in the MT + Ex group was greater than that in the MT-alone group (cP < 0.05). Concomitant with ATF4 mRNA expression, the ATF4 protein level in whole lysates was also decreased compared with the Con group ( Fig. 3B; aP < 0.05). However, total CHOP expression was incredibly increased in the Veh group (aP < 0.05) and was also induced in the MT and MT + Ex groups with caspase-3 activation (a,b,cP < 0.05). To identify the functions of ATF4 and CHOP as transcription factors, we separated the nuclear fraction from the cytosol fraction (Fig. 3C). Compared with the Con group, attenuation of nuclear ATF4 was demonstrated in the rats with knee instability at 1 wk (aP < 0.05). Compared with the Veh group, ATF4 levels in the nuclei were increased by melatonin 179
Hong et al. (A)
(C)
(B)
Fig. 2. Melatonin with and without treadmill exercise regulates muscle mass through both protein synthesis and degradation. The markers involved in (A) protein synthesis (e.g., IGF-I-Akt-mTOR signals) and (B) degradation (e.g., ubiquitin–proteasomal pathway, autophagy), were analyzed to determine causes of the alteration in muscle mass. (A) The expression of Akt and mTOR, which are downstream of IGF-I signals, was analyzed. (B) MAFbx and MuRF1 (ubiquitin–proteasome system), and LC3 II expression (autophagy) were compared. (C) Additionally, Foxo3, which acts as a negative regulator of autophagy, was also analyzed by RT-PCR. The bars represent the means of at least three independent experiments S.D. aP < 0.05 versus Con; bP < 0.05 versus Veh at 1 and 4 wks, respectively.
treatment with and without exercise (bP < 0.05). However, nuclear CHOP was significantly increased in all of the groups (aP < 0.05) and was the greatest in the Veh group. Melatonin with and without exercise remarkably reduced the localization of CHOP within the nuclei (bP < 0.05). Although chaperone proteins are induced by ER stress, ATF4 was downregulated in the Veh group, and nuclear CHOP was increased possibly through compensatory mechanisms in the early phase of knee joint instability. Yu and colleagues [43] suggested a role of CHOP as a new activator of macroautophagy in SBMA mice. Moreover, O’Leary and colleagues [44] reported that reduction in adaptive plasticity in aged animals with not only less activation of autophagy but also marked increase of CHOP was detected in aged muscle following denervation. There was 180
an incredible upregulation of nuclear CHOP in Veh group at early phase, although LC3 II level was elevated in our results. Based on those previous studies [43, 44], these results indicate that nuclear CHOP elevation may be responsible for blockade of muscular adaptation through regulation of autophagy following intra-articular collagenase injection. During the advanced phase, no significant difference in ATF4 mRNA was noted between the Veh and Con groups, whereas ATF4 mRNA was overexpressed by the melatonin alone and melatonin with exercise compared with the Veh and Con groups (Fig. 3A; a,bP < 0.05). Although CHOP mRNA was induced in the MT-alone group at 4 wks (a,bP < 0.05), the MT + Ex group showed significantly CHOP mRNA (a,b,cP < 0.05). No significant
Melatonin induces muscular remodeling in osteoarthritis (A)
(C)
(B)
Fig. 3. Prolonged melatonin treatment with and without exercise inhibits C/EBP homology protein (CHOP)-mediated apoptosis in the gastrocnemius. (A–B) The mRNAs and proteins of ATF4–CHOP–Caspase-3 signals, which are activated by endoplasmic reticulum (ER) stress, were analyzed by RT-PCR and Western blotting, respectively. The bars represent the means of at least three independent experiments S.D. aP < 0.05 versus Con; bP < 0.05 versus Veh; cP < 0.05 versus MT at 1 and 4 wks. (C) The levels of CHOP and ATF4 in the nucleus were measured using subcellular fractionation. The protein fractions (N, nuclear; C, cytosolic fraction) were confirmed by Western blotting using anti-lamin A (nuclear control) and anti-b-tubulin (cytosolic control). The bars represent the means of at least three independent experiments S.D. aP < 0.05 versus Con; bP < 0.05 versus Veh; cP < 0.05 versus MT at 1 and 4 wks.
difference of ATF4 protein expression was shown, whereas the total CHOP level in MT and MT + Ex groups was maintained, being upregulated at 1 wk (Fig. 3B; a,b P < 0.05). However, collagenase injection with no intervention restored the CHOP level in whole lysates. Although nuclear ATF4 was downregulated in the muscle (a,b,cP < 0.05), the nuclear CHOP level was increased in all groups (Fig. 3C; aP < 0.05). Particularly, the nuclear level of CHOP in the Veh group was maintained with a dramatic increase, and both of melatonin with and without exercise reduced its expression (bP < 0.05). Augmentation of nuclear CHOP during the progression of knee instability may indicate chronic UPR induced by sustained stress. Concomitantly, caspase-3, a downstream mediator of CHOP, was activated in all of the groups (aP < 0.05). The level of cleaved caspase-3 was greatest in the Veh group, whereas significantly reduced cleaved caspase-3 was
observed in the MT and MT + Ex groups (Fig. 3B; P < 0.05). Then, we determined a role of cleaved caspase-3 on muscle adaptation during progression of knee instability. A previous study suggested that caspase-12, an upstream of caspase-3, was activated in nonapoptotic but differentiating myoblasts [24]. Therefore, we confirmed the level of MHC isoforms to determine the involvement of activated caspase-3 in differentiation. Myofibers that determine intrinsic contractile properties are classified into type Ib, type IIa, type IIx/d, and type IIb MHCs in adult rodents [45]. Based on the composition by type, mammalian skeletal muscles can generally be classified into slow- and fasttwitch types. Slow-twitch muscles of mammals, including the soleus, predominantly express the slow-type MHC I isoform, with some varied proportion of type IIa, which plays a key role in antigravity function. Fast-twitch
b
181
Hong et al. (A)
(B)
Fig. 4. Melatonin with and without exercise accelerates myosin heavy chain isoform expression through upregulation of nuclear myogenin. (A) Regarding the reduction in caspase-3 activity, we analyzed myosin heavy chain isoforms (MHCs) to identify the effect of longterm interventions on the functional diversity of muscle fibers in the gastrocnemius. The values of MHCs relative to those of glyceraldehyde-3-phosphate dehydrogenase were normalized with Con (% of control). The bars represent the means of at least three independent experiments S.D. One-way ANOVA with the post hoc Tukey’s test was used to compare the Veh, MT, and MT + Ex groups. a P < 0.05 versus Veh; bP < 0.05 versus MT at 1 wk; cP < 0.05 versus Veh; dP < 0.05 versus MT at 4 wks. (B) Myogenin, which is a myogenic regulatory factor (MRF), was analyzed by immunoblotting of subcellular fractions and whole lysates. The bars represent the means of at least three independent experiments S.D. aP < 0.05 versus Con; bP < 0.05 versus Veh at 1 and 4 wks.
muscles of small mammals, which include the gastrocnemius-plantaris complex, predominantly express the two fast isoforms, IIx and IIb, with variable proportions [46]. Regarding the gastrocnemius, the MHC levels in the Veh group, particularly the fast-twitch types, were lower than those in the Con group at the early phase of knee instability, indicating a loss of MHCs in the Veh group (Fig. 4A). Melatonin treatment augmented the MHC Ib level compared with that in the Veh group (aP < 0.05), and the level of MHC IIa was restored to that of the Con 182
group (aP < 0.05). However, the levels of MHC IIx and IIb in the MT group remained lower than those in the Con group. Melatonin with exercise upregulated the expression of both the slow and fast types, resulting in significant differences among the Veh, MT, and MT + Ex groups (a,bP < 0.05). MHC loss was recovered in the Veh group at 4 wks, a finding that was greatest for MHC IIa. Prolonged melatonin administration promoted induction of MHC Ib and IIa compared with the Veh group (cP < 0.05). Changes in the MHC IIx and IIb levels were
Melatonin induces muscular remodeling in osteoarthritis not significantly different between the Veh and MT groups. Melatonin with exercise enhanced the MHC Ib level compared with that in the Veh and MT groups (c, d P < 0.05). These findings indicate that melatonin with and without exercise stimulated MHC expression, particularly the slow types, and that melatonin with exercise accelerated these effects in the early phase. This can be summarized that melatonin with and without exercise enhanced the level of slow-twitch MHCs with caspase-3 activation and induced less localization of CHOP within the nuclei during the early phase. In the advanced phase, there was greater cleavage of caspase-3 than that of interventional groups, with sustained nuclear CHOP expression in the Veh group. Although melatonin and melatonin with exercise induced attenuation of caspase-3 expression at 4 wks, MHCs, particularly slow types, were upregulated than those of Veh group. These results show that melatonin with and without exercise may promote myoblast differentiation for muscular adaptation induced by intra-articular collagenase injection. The skeletal muscle reprograms gene expression through activation of MRFs to alter the characteristics of muscle fibers in response to environmental stimuli [47]. Based on the increase in slow-type MHCs, we analyzed myogenin as a target of regulatory factors (Fig. 4B). Compared with the Con group, total myogenin was significantly increased in the MT and MT + Ex groups at the early and advanced phases (a,bP < 0.05). However, myogenin, which is located in the nuclei, was significantly reduced by collagenase injection at 1 and 4 wks (aP < 0.05). Compared with the Veh group, there was an augmentation of nuclear myogenin in the MT and MT + Ex groups at 1 wk (bP
© 2013 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Molecular, Biological, Physiological and Clinical Aspects of Melatonin
Doi:10.1111/jpi.12110
Journal of Pineal Research
Melatonin treatment combined with treadmill exercise accelerates muscular adaptation through early inhibition of CHOP-mediated autophagy in the gastrocnemius of rats with intra-articular collagenase-induced knee laxity Abstract: The purpose of this study was to determine the effects of melatonin intervention on gastrocnemius remodeling in rats with collagenase-induced knee instability. Type VII collagenase was injected into the right knee to induce joint laxity with cartilage destruction. Melatonin (MT; 10 mg/kg) injection was performed twice daily subcutaneously, and treadmill exercise (Ex; 11 m/min) was conducted for 1 hr/day at a frequency of 5 days/wk for 4 wks. The gastrocnemius mass, which was reduced with collagenase injection only (Veh), was increased with collagenase injection with melatonin treatment with and without exercise in the early phase, and the mass in both limbs was significantly different in the Veh compared with the MT group. However, there was an increase in the relative muscle weight to body weight ratio in the Veh group at the advanced stage. Insulin-like growth factor receptor (IGF-IR) was downregulated in the Veh group, whereas IGF-IR was upregulated in the MT and MT + Ex groups. Joint laxity induced enhancement of autophagic proteolysis (LC3 II) in the muscle, which was recovered to values similar to those in the normal control group (Con) compared with those in the MT and MT + Ex groups. Although intra-articular collagenase increased the total C/ EBP homology protein (CHOP) levels at 1 wk and decreased them at 4 wks in the Veh group, CHOP in the nucleus was upregulated continuously. Prolonged melatonin treatment with and without exercise intervention suppressed nuclear localization of ATF4 and CHOP with less activation of caspase-3, at the advanced phase. Moreover, the interventions promoted the expression of myosin heavy chain (MHC) isoforms under the control of myogenin. Concomitant with a beneficial effect of melatonin with and without exercise, step length of the saline-injected limb and the collagenase-injected supporting side was maintained at values similar to those in control rats. Taken together, the findings demonstrate that melatonin with and without exercise accelerate remodeling of the gastrocnemius through inhibition of nuclear CHOP in rats with collagenase-induced knee instability.
Yunkyung Hong1,2,a, Joo-Heon Kim3,a, Yunho Jin2,4, Seunghoon Lee1,2, Kanghui Park1,5, Youngjeon Lee1,6, Kyu-Tae Chang6 and Yonggeun Hong1,2,4,7 1
Department of Rehabilitation Science, Graduate School of Inje University, Gimhae, Korea; 2Cardiovascular & Metabolic Disease Center, College of Biomedical Science & Engineering, Inje University, Gimhae, Korea; 3 Institute of Animal Medicine, College of Veterinary Medicine, Gyeongsang National University, Jinju, Korea; 4Department of Physical Therapy, Graduate School of Inje University, Gimhae, Korea; 5Division of Rehabilitation Medicine, Dongwei Medical Center, Busan, Korea; 6National Primate Research Center (NPRC), Korea Research Institute of Bioscience and Biotechnology (KRIBB), Ochang, Korea; 7Ubiquitous Healthcare Research Center (UHRC), Inje University, Gimhae, Korea Key words: CHOP, knee joint instability, melatonin, muscle remodeling, treadmill exercise Address reprint request to Yonggeun Hong, Department of Rehabilitation Science, Graduate School of Inje University, 197 Inje-ro, Gimhae, Gyeong-nam 621-749, Korea. E-mail: [email protected] Kyu-Tae Chang, National Primate Research Center (NPRC), Korea Research Institute of Bioscience and Biotechnology (KRIBB), 30 Yeongudanji-ro 30, Ochang, Chung-buk 363-883, Korea. E-mail: [email protected] a These authors contributed equally to this work. Received September 30, 2013; Accepted December 3, 2013.
Introduction Skeletal muscle produces voluntary movement, absorbs loading, and provides dynamic joint stability [1, 2]. Shrier [3] proposed that muscle dysfunction and weakness resulting from injury, overuse, inadequate rehabilitation, and inactivity are the primary drivers of osteoarthritis (OA). The muscles absorb most of the forces that load to the knee joint under normal conditions [4], and the remaining forces are transmitted to articular cartilage, leading to redistribution to bone [5]. Abnormal force distribution causes cartilage damage [3], and chronic mechanical stimulation eventually induces OA [6]. OA affects at least 30%
of people aged older than 60 yr and is a major cause of functional disability. With the population aging, the prevalence of OA in the developed world is expected to increase. Moreover, it is anticipated that OA will become the fourth leading cause of disability in the coming decades [7]. Inflammatory joint pathology induced by intra-articular adjuvant injection causes peri-articular muscle atrophy in the early phase [8]. Muscle atrophy refers to wasting of muscle mass due to disuse, denervation, and systemic disease. The skeletal muscle maintains a balance between the anabolic and catabolic machinery under nonstimulus conditions. However, skeletal muscle reprograms gene 175
Hong et al. expression to alter the characteristics of muscle fibers against environmental stimuli through activation of myogenic transcription factors (MRFs) [9, 10], a process termed muscle adaptation or plasticity [11]. Proteolysis through the ubiquitin–proteasome pathway and autophagy is upregulated under atrophic conditions, whereas protein synthesis is downregulated [12]. In genetic screens, two markers, MAFbx and MuRF1, have been shown to be significantly enhanced prior to the initiation of muscle loss in an animal model [13]. The autophagy– lysosome system is also involved in the regulation of muscle mass [14]. Autophagy removes damaged organelles and unfolded proteins, controls the quality of protein folding in the endoplasmic reticulum (ER), and affects DNA stability [15, 16]. The uncontrolled autophagic machinery accumulates abnormal toxic proteins in the myofibers, leading to muscular degeneration [14], and autophagosomes are observed in the muscle of human with myopathies [17]. Korolchuk and colleagues have suggested that cross-talk may be present between the ubiquitin–proteasome pathway and autophagy–lysosome system [18]. The ER controls the synthesis, folding, and assembly of proteins and provides intracellular Ca2+ storage. ER homeostasis is disrupted by various stimuli such as abnormal nutrients level, Ca2+ imbalance, and increased protein synthesis, leading to accumulation of nonfunctional proteins within the ER lumen [19]. Increased chaperone expression enhances the protein-processing capacity to restore equilibrium when ER stress is applied, a process that is referred as the acute unfolded protein response (UPR). Acute UPR activation is an adaptive response to ER stress, whereas sustained UPR with chronic stress regulates ER stress-related cell fate (e.g., apoptosis or autophagy) [20, 21]. ER stress is present in nonpathological skeletal muscle [22]. UPR in the skeletal muscle is activated to adapt and protect against stress after exercise [23]. C/EBP homology protein (CHOP), a transcription factor, is known to be upregulated by ER stress, which is induced in both apoptotic and differentiating myoblasts through caspase-12 activation [24]. In addition to catabolic pathways, activation of PI-3K/ IGF-I/Akt/mTOR signals can prevent muscle atrophy [25] and is required for a normal hypertrophic response [26]. Mechanical loads induce IGF-I release in skeletal muscle in vitro and in vivo [27]. Additionally, Akt/mTOR signaling regulates postnatal muscle size under conditions of increased mechanical loading [28]. Prolonged melatonin treatment prevents muscle atrophy and activates IGF-Imediated hypertrophic signaling in the gastrocnemius of focal cerebral ischemic rats [29]. The pineal hormone melatonin modulates circadian rhythm [30], oxidative stress [31, 32], and inflammation [33]. Previous studies have reported the beneficial effects of melatonin on muscle damage [34–37]. In the present study, we determined the effect of melatonin with and without exercise on muscle remodeling. Our findings show that melatonin with and without exercise suppresses CHOP-mediated cell death, leading to early adaptation in the gastrocnemius of rats with knee joint laxity. 176
Materials and methods Induction of knee joint instability Eight-wk-old male Sprague–Dawley rats weighing 250– 285 g were used for the present study and were obtained from Daehan Link (Hyochang Science, Daegu, Korea). Rats were housed under light–dark conditions with a controlled temperature. All rats were provided water and food ad libitum. Experimental procedures were approved by the ethics committee for animal care and use at Inje University (Approval no. 2010-72). Experimental rats were randomly divided into three groups after intra-articular injection of collagenase: Veh (collagenase injection only); MT (collagenase injection with melatonin treatment); MT + Ex (collagenase injection with melatonin treatment and treadmill exercise). Rats were anesthetized with an intraperitoneal injection of Zoletil (80 mg/kg of body weight) combined with Xylazine (50 mg/kg of body weight). Alcohol was applied to the shaved skin before collagenase injection. Using a Hamilton microliter syringe (Hamilton Company, Reno, NV, USA), 6 lL of highly purified type VII collagenase (Sigma-Aldrich, St. Louis, MO, USA) was injected into the intra-articular space of the right knee to induce destruction of the ligament and cartilage. For comparison, saline solution was injected at the same volume into the left knee [38]. Intervention Melatonin (Sigma-Aldrich) was dissolved in absolute ethanol and further diluted in 0.9% NaCl solution. Rats were subcutaneously injected with the melatonin solution at doses of 10 mg/kg (twice daily at 07:00 and 19:00). A motor-driven treadmill was used for the exercise intervention. The rats were placed on a moving belt and were trained to run in the direction opposite to the movement of the belt. In 2004, Galois and colleagues [39] proposed that slight (30 cm/s for 15 min, total distance of 7.5 km over 28 days) or moderate exercise (30 cm/s for 30 min, total distance of 15 km over 28 days) exhibits beneficial effects on severity on chondral lesions in the rats with anterior cruciate ligament transection. Based on the previous study, we selected approximate moderate exercise intensity as following. The exercise was conducted for 30 min twice daily (at 17:00 and 22:00) at a frequency of 5 days/wk. The speed and inclination of the treadmill were 11 m/min and 0°, respectively. To allow adaptation to the treadmill, pretraining was carried out for 20 min/day twice daily (at 17:00 and 22:00) during 1 wk before collagenase injection. Collection of animal tissues The gastrocnemius muscle was collected under anesthesia, washed with cold phosphate-buffered saline, and frozen at 70°C. Before freezing, the wet muscle weight was measured using an electronic balance, and the weight was compared among the groups.
Melatonin induces muscular remodeling in osteoarthritis Table 1. Oligonucleotide primers used for PCR Gene
Primer sequence (5′–3′)
Size (bp)
GenBank accession no.
ATF4
F: tcc tga aca gcg aag tgt tg R: cat cca tag cca gcc att ct F: agc aga ggt cac aag cac ct R: ctg ctc ctt ctc ctt cat gc F: ggg gag ttt ggt caa tca ga R: ttt gca tag act ggc tga cg F: gta tga ctc cac tca cgg caa a R: ggt ctc gct cct gga aga tg F: tcc caa gct gtg tgt ctc tg R: gtg cca cgt tat gat gat gc F: aca gag gaa gac agg aag aac cta c R: ggg ctt cac agg cat cct tag F: cct ctt act tcc cag ctg cac ctt ct R: act ttc cct gcg tct ttg ctc tga at F: agc ctg cct cct tct tca tct gg R: cac ggt tgc ttt cac ata gga ctc F: acg gtc gaa gtt gca tcc cta aag R: cac ctt cgg tct tgg ctg tca c
174
NM_024403
166
NM_001109986
170
NM_001106395
100
NM_008084
178
NM_052807.2
288
K01463
239
DQ872905
229
DQ872907
263
DQ872906
CHOP FoxO3 Gapdh IGF-IR MHC Ib MHC IIa MHC IIx MHC IIb
RNA isolation and reverse transcription polymerase chain reaction (RT-PCR) analysis Animal muscle tissue was homogenized with TRI reagent (Sigma-Aldrich) to isolate total RNA. Total RNA was quantified using an Optizen 2120 UV spectrophotometer (Mecasys, Daejeon, Korea) and reverse transcribed using reverse transcriptase. cDNA was amplified using specific primers (Table 1). RT-PCR was performed in a Px2 Thermal cycler HBPX2220 (Thermo Electron Corporation, Beverly, MA, USA). The cycling conditions were as follows: a 10 min initial activation step at 94°C, followed by denaturation at 94°C for 30 s, primer-specific annealing temperature (°C) for 1 min, and extension at 72°C for 1 min. PCR products were confirmed by 1.5% agarose gel electrophoresis. The Image J program ver. 1.6 (NIH, Bethesda, MD, USA) was used to quantify the density. Subcellular fractionation Subcellular components were separated by methods modified from those used in a previous study [40]. After removal of the connective tissues, the gastrocnemius was resuspended and homogenized on ice in STM buffer (250 mM sucrose, 50 mM Tris-HCl (pH 7.4), 5 mM MgCl2) supplemented with protease inhibitor cocktail (Roche, Rotkreuz, Switzerland). The homogenates were maintained on ice for 30 min, vortexed at maximum speed for 15 s, and centrifuged at 800 g for 15 min; then, the supernatants were separated from the pellets. The pellet (P0), which contained nuclei and debris, was resuspended in STM buffer, vortexed at maximum speed for 15 s, and then centrifuged at 500 g for 15 min. Then, the pellet (P1) was washed in STM buffer, vortexed at maximum speed for 15 s, and centrifuged at 1,000 g for 15 min. Thereafter, the washed pellet (P2) was resuspended in NET buffer (20 mM HEPES (pH 7.9), 1.5 mM MgCl2, 0.5 M NaCl, 0.2 mM EDTA, 20% glycerol, 1% Triton X-100) supplemented with protease inhibitor cocktail (Roche) to triturate until homogeneous. The pellet (P2) was vortexed at
maximum speed for 15 s and incubated on ice for 30 min. The nuclei were lysed by 10–20 passages through an 18gauge needle. The lysate was centrifuged at 9,000 g for 30 min, and the supernatant was collected as the ‘nuclear fraction’. The supernatant (S0), which contained not only the mitochondrial but also the cytosolic fraction, was centrifuged at 800 g for 10 min. The pellet was discarded, and the supernatant (S1) was collected and centrifuged at 11,000 g for 10 min. The supernatant (S2), which contains the cytosol and microsomal fractions, was precipitated in 100% cold acetone at 20°C for 2 hr and centrifuged at 12,000 g for 5 min. The pellet was resuspended in STM buffer, which was used to analyze cytosolic protein expression. Protein extraction and Western blotting The medial gastrocnemius muscles were lysed in a buffer (135 mM NaCl, 1 mM MgCl2, 2.7 mM KCl, 20 mM TrisHCl (pH 8.0), 0.5 mM Na3VO4, 10 mM NaF, 1% Triton X-100) supplemented with protease inhibitor cocktail (Roche). The lysate was incubated on ice for 30 min and centrifuged at 14,920 g for 10 min at 4°C; then, the supernatant was collected. The concentration of proteins was measured using an Optizen 2120 UV spectrophotometer (Mecasys) and the Bradford protein assay (Bio-Rad Laboratories, Richmond, CA, USA). The proteins were boiled at 95°C for 10 min and mixed with 29 sample buffer. Thirty-microgram protein samples were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), and the separated protein bands on the gel were transferred to a polyvinylidene fluoride membrane (Merck Millipore, Billerica, MA, USA). The membrane was blocked with 5% nonfat milk in Tris-buffered saline with Tween 20 (TBST) for 1 hr at room temperature. Primary antibodies were incubated with blocked membranes overnight. Membranes were incubated with secondary antibody in TBST for 40 min. Finally, specific bands were visualized using a Western blotting luminal reagent (Santa 177
Hong et al. Cruz Biotechnology, Santa Cruz, CA, USA) and quantified using Image J ver. 1.6.
(A)
Footprint measurements Gait measurements were obtained using a modified footprint method [41]. Both the hind toes and soles of the experimental rats were painted with colored inks (salineinjected limb: blue, collagenase-injected limb: red) before placing them in a run/walk, open-ended acrylic container (1 m in length, 0.08 m in width, and 0.15 m in height). A white paper was laid beneath the acrylic container, and the rats walked toward the open end of the acrylic container. To determine significance, measurements between paw prints were made using at least seven legibly marked strides. Step length and width were measured while the unstable limb was weight bearing and the saline-injected limb was in motion. Footprint images were quantified using the Image J program ver. 1.6.
(B)
Statistical analysis Data were collected from at least three repeated experiments and were presented as the mean standard deviation (S.D.) in graphs. Statistically significant differences between groups were assessed using one-way analysis of variance (ANOVA) with the post hoc Tukey’s test. Differences were deemed statistically significant at a P < 0.05. All analyses were performed using the statistical software SPSS ver. 19.0 (IBM, New York, NY, USA).
(C)
Results Because skeletal muscle has a great impact on whole-body metabolic homeostasis [23], we measured not only the wet weight of the gastrocnemius (Fig. 1B) but also the body weight (Fig. 1A) of rats at 1 wk (early) and 4 wks (advanced). Compared with normal control (Con) and vehicle treatment, melatonin intervention with and without exercise was associated with reduced body weight in rats with knee joint instability at 4 wks (aP < 0.05, bP < 0.05). In the early phase, collagenase injection with no treatment attenuated the gastrocnemius weight compared with the ipsilateral side in the Con group (aP < 0.05) and the contralateral limb in the Veh group (*P < 0.05). Melatonin treatment augmented the mass of the unaffected side only compared with Con (aP < 0.05) and Veh (bP < 0.05), indicating a significant difference between the contralateral and ipsilateral sides (*P < 0.05), although no difference was observed in the collagenase-injected limb of the MT group compared with that of the Con group. However, melatonin with exercise induced a significant increase in the mass in both limbs compared with Con (aP < 0.05), Veh (bP < 0.05), and MT (cP < 0.05) groups. Concomitant with this, the value of the muscle mass relative to body weight showed an increase with melatonin with exercise at 1 wk (Table 2; a,b,cP < 0.05). A significant increase in mass was demonstrated only in the Veh group compared with the Con group at the advanced stage (Fig. 1B; aP < 0.05). Although body weight was reduced in the MT and MT + Ex groups (bP < 0.05), 178
Fig. 1. Intra-articular collagenase-induced changes in body weight and gastrocnemius muscle mass. (A) Body weight of rats was measured at 1, 3, 7, 14, 21, and 28 days after collagenase injection. The values are the means S.D. aP < 0.05 versus Con; b P < 0.05 versus Veh. (B) The wet weight of the gastrocnemius of both hindlimbs (unaffected, saline; affected, collagenase injection) was compared among the groups at 1 and 4 wks. The bars represent the means S.D. aP < 0.05 versus Con; bP < 0.05 versus Veh; cP < 0.05 versus MT at 1 and 4 wks, respectively. Student’s t-test was used to assess the significance (*P < 0.05) of differences between the saline- and collagenase-injected sides. (C) IGF-I receptor (IGF-IR) mRNA was significantly downregulated in the gastrocnemius of the unstable knee of rats, whereas melatonin with and without treadmill exercise induced maintenance or an increase in the IGF-IR level compared with Con. aP < 0.05 versus Con; bP < 0.05 versus Veh; cP < 0.05 versus MT at 1 and 4 wks, respectively.
Melatonin induces muscular remodeling in osteoarthritis Table 2. Relative mass values to body weight Con
1 wk Significance 4 wks Significance
Veh
Saline
Collagenase
Saline
Collagenase
Saline
4.55 0.06 – 4.51 0.01 –
4.55 0.06 – 4.51 0.02 –
4.68 0.24 – 5.41 0.21
4.11 0.12
5.32 0.04
5.38 0.24
5.04 0.07
a
MT + Ex
MT
a,*
a
a,b
a
Collagenase
Saline
Collagenase
4.55 0.33
5.34 0.08
5.19 0.09
5.16 0.07
5.27 0.19
5.48 0.15
*
a
a,b
a
a,b,c
a
The values represent the means S.D. ap < 0.05 versus Con; bp < 0.05 versus Veh; cp < 0.05 versus MT in the each limb at 1 and 4 wks, respectively. Student’s t-test was used to compare the saline- and collagenase-injected sides (*P < 0.05).
the relative muscle mass was not significant among the Veh, MT, and MT + Ex groups at 4 wks. No significant difference was also found between the saline- and collagenase-injected limbs among all the groups (Table 2). Regarding the changes in muscle mass, we analyzed IGF-IR expression (Fig. 1C). IGF-IR expression is important for myoblast proliferation and maintenance of normal muscle mass [42]. In this study, IGF-IR level was the lowest in the Veh group at 1 wk after collagenase injection (aP < 0.05). However, IGF-IR was upregulated in the MT and MT + Ex groups (bP < 0.05), and the level was higher with the combined intervention than with melatonin alone (cP < 0.05). The expression of IGF-IR in the Veh group remained at a lower level than that of the Con group at 4 wks (aP < 0.05). Compared with the Veh group, both melatonin alone and melatonin with exercise intervention enhanced IGF-IR levels (bP < 0.05). Particularly, IGF-IR expression was highest with melatonin alone treatment in the advanced phase. We compared protein synthesis and degradation to identify the mechanisms involved in mass alteration during the progression of joint instability. The IGF-I/Akt/ mTOR pathway is the predominant signal to stimulate protein synthesis, leading to hypertrophy [26]. Total Akt expression was significantly increased in the gastrocnemius of the collagenase-injected limb in the Veh and MT groups compared with that in the Con group (Fig. 2A; a P < 0.05). However, the total Akt expression level in the MT + Ex group was recovered to that of the Con group (b,cP < 0.05). Total mTOR expression in the rats with knee laxity was markedly reduced at 1 wk (aP < 0.05). As joint ligaments were continuously damaged, the mTOR level was upregulated by melatonin treatment only (a,b,cP < 0.05). Furthermore, its expression was lower in the Veh and MT + Ex groups than in both the Con and MT groups (aP < 0.05). Although the muscle mass of the MT + Ex group was augmented in the early phase, Akt and mTOR levels were maintained or downregulated. Therefore, we need to further analyze the expression of p-Akt and p-mTOR for confirmation of their activation. We analyzed the key markers of the ubiquitin–proteasome pathway (e.g., atorgin-1/MAFbx and MuRF1) and autophagy (e.g., LC3 II) signal to determine the proteolytic mechanisms involved in the changes in mass (Fig. 2B). MAFbx was significantly upregulated in the MT and MT + Ex groups at 1 wk compared with the Con (aP < 0.05), whereas no significance was shown
among the groups at 4 wks. MuRF1, which leads to breakdown of the structural muscle proteins such as myosin heavy chains (MHCs), was decreased in the MT + Ex group at 1 wk (a,b,cP < 0.05). However, no significant differences among the groups were found in the advanced phase. Autophagosome-bound LC3 expression (LC3 II) was dramatically increased in the Veh group at 1 wk (aP < 0.05). Prolonged melatonin treatment induced upregulation of LC3 II expression in the advanced phase (a,bP < 0.05), whereas melatonin treatment combined with exercise restored it to a level similar to that of the Con group (cp < 0.05). To confirm the activation of autophagy, we analyzed Foxo3 as a negative regulator of autophagy (Fig. 2C). Melatonin treatment with and without exercise intervention induced upregulation of Foxo3 compared with vehicle treatment (bP < 0.05), which was similar to that of Con group at 1 wk. No reduction in Foxo3 expression was noted in the Veh group at 4 wks, whereas the MT and MT + Ex groups showed enhanced Foxo3 expression compared with the Con group (a,bP < 0.05). These results show not only activation of the autophagic machinery through Foxo3 suppression in the Veh group but also inhibition of MuRF-1-mediated ubiquitin–proteasomal proteolysis in the MT + Ex group in the early phase. Next, we examined whether early loss of muscle mass induced by autophagic activation is associated with ER stress in myofibers. Using data from a previous study [22], we selected ATF4 and CHOP as markers of UPR activated by ER stress (Fig. 3). In the early phase, the mRNA levels of both ATF4 and CHOP decreased with vehicle treatment (aP < 0.05) but were restored by melatonin with and without exercise (Fig. 3A; bP < 0.05;). Particularly, the increase in the mRNA levels of both ATF4 and CHOP in the MT + Ex group was greater than that in the MT-alone group (cP < 0.05). Concomitant with ATF4 mRNA expression, the ATF4 protein level in whole lysates was also decreased compared with the Con group ( Fig. 3B; aP < 0.05). However, total CHOP expression was incredibly increased in the Veh group (aP < 0.05) and was also induced in the MT and MT + Ex groups with caspase-3 activation (a,b,cP < 0.05). To identify the functions of ATF4 and CHOP as transcription factors, we separated the nuclear fraction from the cytosol fraction (Fig. 3C). Compared with the Con group, attenuation of nuclear ATF4 was demonstrated in the rats with knee instability at 1 wk (aP < 0.05). Compared with the Veh group, ATF4 levels in the nuclei were increased by melatonin 179
Hong et al. (A)
(C)
(B)
Fig. 2. Melatonin with and without treadmill exercise regulates muscle mass through both protein synthesis and degradation. The markers involved in (A) protein synthesis (e.g., IGF-I-Akt-mTOR signals) and (B) degradation (e.g., ubiquitin–proteasomal pathway, autophagy), were analyzed to determine causes of the alteration in muscle mass. (A) The expression of Akt and mTOR, which are downstream of IGF-I signals, was analyzed. (B) MAFbx and MuRF1 (ubiquitin–proteasome system), and LC3 II expression (autophagy) were compared. (C) Additionally, Foxo3, which acts as a negative regulator of autophagy, was also analyzed by RT-PCR. The bars represent the means of at least three independent experiments S.D. aP < 0.05 versus Con; bP < 0.05 versus Veh at 1 and 4 wks, respectively.
treatment with and without exercise (bP < 0.05). However, nuclear CHOP was significantly increased in all of the groups (aP < 0.05) and was the greatest in the Veh group. Melatonin with and without exercise remarkably reduced the localization of CHOP within the nuclei (bP < 0.05). Although chaperone proteins are induced by ER stress, ATF4 was downregulated in the Veh group, and nuclear CHOP was increased possibly through compensatory mechanisms in the early phase of knee joint instability. Yu and colleagues [43] suggested a role of CHOP as a new activator of macroautophagy in SBMA mice. Moreover, O’Leary and colleagues [44] reported that reduction in adaptive plasticity in aged animals with not only less activation of autophagy but also marked increase of CHOP was detected in aged muscle following denervation. There was 180
an incredible upregulation of nuclear CHOP in Veh group at early phase, although LC3 II level was elevated in our results. Based on those previous studies [43, 44], these results indicate that nuclear CHOP elevation may be responsible for blockade of muscular adaptation through regulation of autophagy following intra-articular collagenase injection. During the advanced phase, no significant difference in ATF4 mRNA was noted between the Veh and Con groups, whereas ATF4 mRNA was overexpressed by the melatonin alone and melatonin with exercise compared with the Veh and Con groups (Fig. 3A; a,bP < 0.05). Although CHOP mRNA was induced in the MT-alone group at 4 wks (a,bP < 0.05), the MT + Ex group showed significantly CHOP mRNA (a,b,cP < 0.05). No significant
Melatonin induces muscular remodeling in osteoarthritis (A)
(C)
(B)
Fig. 3. Prolonged melatonin treatment with and without exercise inhibits C/EBP homology protein (CHOP)-mediated apoptosis in the gastrocnemius. (A–B) The mRNAs and proteins of ATF4–CHOP–Caspase-3 signals, which are activated by endoplasmic reticulum (ER) stress, were analyzed by RT-PCR and Western blotting, respectively. The bars represent the means of at least three independent experiments S.D. aP < 0.05 versus Con; bP < 0.05 versus Veh; cP < 0.05 versus MT at 1 and 4 wks. (C) The levels of CHOP and ATF4 in the nucleus were measured using subcellular fractionation. The protein fractions (N, nuclear; C, cytosolic fraction) were confirmed by Western blotting using anti-lamin A (nuclear control) and anti-b-tubulin (cytosolic control). The bars represent the means of at least three independent experiments S.D. aP < 0.05 versus Con; bP < 0.05 versus Veh; cP < 0.05 versus MT at 1 and 4 wks.
difference of ATF4 protein expression was shown, whereas the total CHOP level in MT and MT + Ex groups was maintained, being upregulated at 1 wk (Fig. 3B; a,b P < 0.05). However, collagenase injection with no intervention restored the CHOP level in whole lysates. Although nuclear ATF4 was downregulated in the muscle (a,b,cP < 0.05), the nuclear CHOP level was increased in all groups (Fig. 3C; aP < 0.05). Particularly, the nuclear level of CHOP in the Veh group was maintained with a dramatic increase, and both of melatonin with and without exercise reduced its expression (bP < 0.05). Augmentation of nuclear CHOP during the progression of knee instability may indicate chronic UPR induced by sustained stress. Concomitantly, caspase-3, a downstream mediator of CHOP, was activated in all of the groups (aP < 0.05). The level of cleaved caspase-3 was greatest in the Veh group, whereas significantly reduced cleaved caspase-3 was
observed in the MT and MT + Ex groups (Fig. 3B; P < 0.05). Then, we determined a role of cleaved caspase-3 on muscle adaptation during progression of knee instability. A previous study suggested that caspase-12, an upstream of caspase-3, was activated in nonapoptotic but differentiating myoblasts [24]. Therefore, we confirmed the level of MHC isoforms to determine the involvement of activated caspase-3 in differentiation. Myofibers that determine intrinsic contractile properties are classified into type Ib, type IIa, type IIx/d, and type IIb MHCs in adult rodents [45]. Based on the composition by type, mammalian skeletal muscles can generally be classified into slow- and fasttwitch types. Slow-twitch muscles of mammals, including the soleus, predominantly express the slow-type MHC I isoform, with some varied proportion of type IIa, which plays a key role in antigravity function. Fast-twitch
b
181
Hong et al. (A)
(B)
Fig. 4. Melatonin with and without exercise accelerates myosin heavy chain isoform expression through upregulation of nuclear myogenin. (A) Regarding the reduction in caspase-3 activity, we analyzed myosin heavy chain isoforms (MHCs) to identify the effect of longterm interventions on the functional diversity of muscle fibers in the gastrocnemius. The values of MHCs relative to those of glyceraldehyde-3-phosphate dehydrogenase were normalized with Con (% of control). The bars represent the means of at least three independent experiments S.D. One-way ANOVA with the post hoc Tukey’s test was used to compare the Veh, MT, and MT + Ex groups. a P < 0.05 versus Veh; bP < 0.05 versus MT at 1 wk; cP < 0.05 versus Veh; dP < 0.05 versus MT at 4 wks. (B) Myogenin, which is a myogenic regulatory factor (MRF), was analyzed by immunoblotting of subcellular fractions and whole lysates. The bars represent the means of at least three independent experiments S.D. aP < 0.05 versus Con; bP < 0.05 versus Veh at 1 and 4 wks.
muscles of small mammals, which include the gastrocnemius-plantaris complex, predominantly express the two fast isoforms, IIx and IIb, with variable proportions [46]. Regarding the gastrocnemius, the MHC levels in the Veh group, particularly the fast-twitch types, were lower than those in the Con group at the early phase of knee instability, indicating a loss of MHCs in the Veh group (Fig. 4A). Melatonin treatment augmented the MHC Ib level compared with that in the Veh group (aP < 0.05), and the level of MHC IIa was restored to that of the Con 182
group (aP < 0.05). However, the levels of MHC IIx and IIb in the MT group remained lower than those in the Con group. Melatonin with exercise upregulated the expression of both the slow and fast types, resulting in significant differences among the Veh, MT, and MT + Ex groups (a,bP < 0.05). MHC loss was recovered in the Veh group at 4 wks, a finding that was greatest for MHC IIa. Prolonged melatonin administration promoted induction of MHC Ib and IIa compared with the Veh group (cP < 0.05). Changes in the MHC IIx and IIb levels were
Melatonin induces muscular remodeling in osteoarthritis not significantly different between the Veh and MT groups. Melatonin with exercise enhanced the MHC Ib level compared with that in the Veh and MT groups (c, d P < 0.05). These findings indicate that melatonin with and without exercise stimulated MHC expression, particularly the slow types, and that melatonin with exercise accelerated these effects in the early phase. This can be summarized that melatonin with and without exercise enhanced the level of slow-twitch MHCs with caspase-3 activation and induced less localization of CHOP within the nuclei during the early phase. In the advanced phase, there was greater cleavage of caspase-3 than that of interventional groups, with sustained nuclear CHOP expression in the Veh group. Although melatonin and melatonin with exercise induced attenuation of caspase-3 expression at 4 wks, MHCs, particularly slow types, were upregulated than those of Veh group. These results show that melatonin with and without exercise may promote myoblast differentiation for muscular adaptation induced by intra-articular collagenase injection. The skeletal muscle reprograms gene expression through activation of MRFs to alter the characteristics of muscle fibers in response to environmental stimuli [47]. Based on the increase in slow-type MHCs, we analyzed myogenin as a target of regulatory factors (Fig. 4B). Compared with the Con group, total myogenin was significantly increased in the MT and MT + Ex groups at the early and advanced phases (a,bP < 0.05). However, myogenin, which is located in the nuclei, was significantly reduced by collagenase injection at 1 and 4 wks (aP < 0.05). Compared with the Veh group, there was an augmentation of nuclear myogenin in the MT and MT + Ex groups at 1 wk (bP
