Muscle-protective effects of Schisandrae Fructus extracts in old mice after chronic forced exercise

Author’s Accepted Manuscript Muscle-protective effects of Schisandrae Fructus extracts in old mice after chronic forced exercise Ki-Young Kim, Sae-Kwang Ku, Ki-Won Lee, Chang-Hyun Song, Won G An www.elsevier.com/locate/jep

PII: DOI: Reference:

S0378-8741(17)31949-9 https://doi.org/10.1016/j.jep.2017.10.022 JEP11076

To appear in: Journal of Ethnopharmacology Received date: 17 May 2017 Revised date: 19 September 2017 Accepted date: 20 October 2017 Cite this article as: Ki-Young Kim, Sae-Kwang Ku, Ki-Won Lee, Chang-Hyun Song and Won G An, Muscle-protective effects of Schisandrae Fructus extracts in old mice after chronic forced exercise, Journal of Ethnopharmacology, https://doi.org/10.1016/j.jep.2017.10.022 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Original Research Article

Muscle-protective effects of Schisandrae Fructus extracts in old mice after chronic forced exercise Ki-Young Kima,b,1, Sae-Kwang Kuc,d,1, Ki-Won Leeb, Chang-Hyun Songc,d*, and Won G Ana* a

Department of Pharmacology, School of Korean Medicine, Pusan National University, Yangsan 626-870,

Republic of Korea. b

Research Institute, Bio Port Korea, Busan 619-912, Republic of Korea

c

Department of Anatomy and Histology, College of Korean Medicine, Daegu Haany University,

Gyeongsan 712-715, Republic of Korea d

MRC-GHF, College of Korean Medicine, Daegu Haany University, Gyeongsan 712-715, Republic of

Korea [email protected]

[email protected] 1

These authors contributed equally to this work.

*

Corresponding

author

at:

Chang-Hyun

Song,

DVM,

PhD,

1

Haanydaero,

Gyeongsan,

Gyeongsangbuk-do 712-715, Republic of Korea. Tel./fax: +82-53-819-1822. *

Corresponding author. Tel.: +82 51 510 8455/fax: +82-51-510-8447.

Abstract Ethnopharmacological Relevance Schisandrae Fructus (SF), the dried fruit of Schisandra chinensis (Turcz.) Baill., is a well-known traditional herb used in Asia for enhancing physical work capacity as well as providing anti-stress and anti-inflammatory effects. Extracts of SF (SFe) have also been reported to increase skeletal muscle mass

1

and inhibit muscle atrophy. Aim Of The Study We examined whether SFe had muscle-protective effects in old mice after chronic forced exercises, and, if so, relevant mechanisms. Materials And Methods Ten-month-old aged male mice were divided into six groups. One group received no forced swimming after oral administration of distilled water (Intact); the other groups received forced swimming after administration of distilled water (SW), oxymetholone (OXY), or SFe at 500, 250 and 125 mg/kg (SFe500, SFe250, and SFe125, respectively). Forced swimming was conducted for 2 min at 30 min after oral administration; the treatment was repeated for 28 days. Muscle thickness, weight, lean proportion, and strength were examined. The sampled muscles were subjected to histopathological and biochemical analyses. Plasma was examined by biochemical analyses. Results The thicknesses of the calf muscle and the sampled gastrocnemius and soleus, protein proportion and muscle strength increased significantly in the SW group versus Intact, and they were further increased in the SFe and OXY groups versus SW. The forced swimming in the SW group upregulated mRNA expression related to protein synthesis (Akt1, PI3K) and muscle growth (A1R, TRPV4), while it downregulated mRNAs related to protein degradation (atrogin-1, MuRF1) and muscle growth inhibitor (myostatin, SIRT1). The detected upregulation and downregulation were enhanced in the SFe groups. In addition, the SFe administration inhibited lipid peroxidation and reactive oxygen species, and accelerated activities of endogenous anti-oxidants and anti-oxidant enzymes. Plasma biochemistry showed decreases in creatine, creatine kinase and LDH in the SFe groups versus SW, suggesting muscle-protective effects of SFe. In the SFe groups versus SW, histopathological analyses revealed an increase in myofibre diameter,

2

and immunohistochemistry showed increases in myofibres immunoreactive for ATPase and decreases in myobibres for apoptosis markers (caspase-3, PARP) and oxidative stress markers (NT, 4HNE, iNOS). Conclusions Oral administration of SFe, especially SFe500, enhanced exercise-induced adaptive muscle strengthening in aged mice after forced swimming through anti-apoptotic and anti-oxidant effects, mediated via modulation of gene expression related to muscle synthesis or degradation. These results suggest that SFe may be helpful in improvement various muscle disorders as an adjuvant therapy to exercise-based remedies.

Abbreviation ANOVA = analysis of variance; A1R = adenosine A1 receptor; CAT = Catalase; DEXA = dual energy X-ray absorptiometry; DW = distilled water; GSH = Glutathione; HE = haematoxylin and eosin; HNE = hydroxynonenal; iNOS = inducible nitric oxide synthase; LDH = lactate dehydrogenase; MDA = malondialdehyde; MuRF1 = muscle RING-finger protein-1; N = Newtons;

NT

=

nitrotyrosine;

PARP

=

poly

(ADP-ribose)

polymerase;

PI3K

=

phosphatidylinositol 3-kinase; RFU = relative fluorescence unit; ROS = reactive oxygen species; RT-PCR = reverse transcription-polymerase chain reaction; SD = standard deviation; SF = Schisandrae Fructus; SFe = extracts of SF; SIRT = sirtulin; SOD = Superoxide dismutase; TRPV = transient receptor potential cation cannel subfamily V member

Keywords Sarcopenia; Exercise; Schisandrae Fructus; Anti-oxidant; Anti-apoptosis

3

Chemical compounds studies in this article citrate (PubChem CID: 6224); dichlorofluorescein diacetate (PubChem CID: 101615877); eosin (PubChem CID: 11048); haematoxylin (PubChem CID: 442514); H2O2 (PubChem CID: 784); methanol (PubChem CID: 887); nitrotetrazolium blue (PubChem CID: 9281); Sirius red (PubChem CID: 5464587); thiobarbituric acid (PubChem CID: 2723628); Tris–HCl (PubChem CID: 93573); 2-nitrobenzoic acid (PubChem CID: 6254)

1. Introduction Sarcopenia is a condition characterized by a progressive loss of skeletal muscle mass and strength associated with aging, which has profound effects on quality of life (Brooks and Faulkner, 1994). Although the exact mechanisms remain unknown, many factors are responsible for decreased muscle mass and strength in the elderly, such as increased insulin resistance, inflammation, hormonal alterations, perturbations in muscle metabolism, and decreased muscle proliferation (Montero-Fernandez and Serra-Rexach, 2013). The resulting reduction in physical activity induces further muscle atrophy, leading to a vicious cycle of atrophic processes (Glass, 2003). Interventions such as exercise training, nutrition, and mechanical stimulation are currently recommended to prevent such atrophic changes, but it is often hard to obtain improvements in sarcopenia, especially in the elderly with the reduced anabolic potential of their skeletal muscle. It is known that appropriate physical aerobic exercise facilitates muscle thickening and remodelling of muscle fibre compositions. Because exercise demands large amounts of energy, followed by increased oxygen consumption, the ATPase in myofibres adapts to the oxygen demand, leading to increased striated muscle strength (Barcelos et al., 2014; Oe et al., 2011). 4

However, excessive exercise induces several metabolic changes, sufficient to disrupt mitochondrial functioning, and induce reactive oxygen species (ROS) production (Barcelos et al., 2014). The muscle inflammation and oxidative stress induced by long-duration or high-intensity exercises have reported to impair the function of the skeletal muscle parenchyma (Davis et al., 2007). Indeed, use of anti-inflammatory agents is encouraged to avoid this, and other available treatments include androgens, growth hormones and ghrelin (Ali and Garcia, 2014). Oxymetholone is oral anabolic androgen steroid commonly used for improving muscle mass and strength, despite its hepatotoxicity (Pavlatos et al., 2001; Wood and Yin, 1994). However, there is no effective treatment approved for sarcopenia. Schisandrae Fructus (SF), the dried fruit of Schisandra chinensis (Turcz.) Baill. (Magnoliaceae), is a well-known traditional herb in Russia and Asia including China, Japan and Korea for increasing physical work capacity and providing anti-stress, anti-inflammation and anti-heavy metal intoxication effects (Lu and Chen, 2009; Panossian and Wikman, 2008). The SF in Korea is known to have bioactive components, mainly containing lignans of schizandrin, gomisin A and tigloylgomisin H (Lu and Chen, 2009). Recent reports have shown that SF has favourable effects on smooth muscle relaxation (Yang et al., 2011; Young Park et al., 2012), diabetes, and related complications in both in vitro and in vivo studies (Kwon et al., 2011; Zhang et al., 2010). We have also previously reported the muscle-protective effects of extracts of SF (SFe) in muscle atrophy models induced by dexamethasone (Kim et al., 2015a) or denervation, through anti-inflammatory and anti-oxidant properties (Kim et al., 2015b). However, there has been no report involved in the beneficial effects of SFe on normal muscular physiology after exercise training as a remedy of sarcopenia in old-ages. The forced swimming model has been used for chronic exercise-induced muscle damage by inflammatory processes and oxidative stress (Chen 5

et al., 2014; Li et al., 2012); the muscle damage is more severe with aging (Bogdanova et al., 2013). Thus, we examined the muscle-protective effects of SFe in normal old mice after forced swimming for 28 days by comprehensive analyses including gross aspects of muscle thickness, weights and strength, lean body mass, plasma and muscle biochemistry and histopathology.

2. Materials and methods 2.1. Reagents SFe in powdered form were provided by Bioport Korea Inc. (Yangsan, Korea); the lot number (BPK-SC-001) was the same as that used in our previous studies (Kim et al., 2015a; Kim et al., 2015b). Briefly, the collected SF (100 kg) was extracted in 20% ethanol at 90ºC for 4 h, and filtered with a 80-mesh filter. The extract was lyophilised and yielded 23.2% reddish brown powders. The SFe contained 4.92 ± 0.06 mg/g of schizandrin as a specific ingredient by high-performance liquid chromatography analysis (Figure A.1). Oxymetholone (oxymetholone 50 mg tablet, Celltrion Inc., Jincheon, Korea) was used as a reference drug. The powders of SFe and ground oxymetholone were dissolved in distilled water (DW) as a vehicle.

Figure A.1. HPLC profile of extracts of Schisandrae Fructus.

6

2.2. Animals. In total, 48 SPF/VAF 10-month-old-aged male Hsd:ICR (CD-1), weighing 32−54 g, were purchased from Envigo (Indianapolis, IN, USA). The mice were housed individually in a polycarbonate cage in a temperature-controlled (20−25°C) and humidity-controlled (40−45%) room. The light/dark cycle was 12 h/12 h, and normal rodent pellet diet (Purinafeed, Seungnam, Korea) and water were available ad libitum. All experiments were conducted in accordance with international regulations of the use and welfare of laboratory animals, and approved by the Institutional Animal Care and Use Committee of Daegu Haany University (Gyeongsan, Korea) (Approval No. DHU2016-064). 2.3. Experimental design. After a week-acclimatisation, the mice were divided into six groups (n = 8 per group), based on body weight (46.5 ± 4.2 g) and calf thickness (4.99 ± 0.08 mm). One group received oral administration of DW (Intact). The other five groups received administration with DW as an exercise control (SW), oxymetholone at 50 mg/kg (OXY), or SFe at 500, 250 and 125 mg/kg (SFe 500, SFe 250, and SFe 125, respectively), followed by swimming exercise as a chronic forced performance 30 min after administration. The administration was performed in a volume of 10 mL/kg once a day for 28 days. The doses of oxymetholone and SFe were based on previous studies (Kim et al., 2015a; Kim et al., 2015b; Pavlatos et al., 2001). Forced swimming was conducted in a circular pool (diameter, 100 cm; height, 30 cm) containing warm water (temperature, 28 ± 1°C) to a 15-cm depth for 2 min a day, as described elsewhere (Hong et al., 2015). The mice fasted overnight before the initial treatment and euthanasia; they were euthanised with CO2 gas after the 28-day treatment. 7

2.4. Measurement of muscle mass. For physical examinations, body weights of mice and thicknesses of their left calf muscle were measured once a week. On day 28 post-treatment, the amount of muscle mass was measured as the lean muscle, subtracted from contents of bone mineral and fat, using a dual energy X-ray absorptiometry (DEXA) digital radiography system (InAlyzer, Medikors, Seungnam, Korea). The strength of the calf muscle, expressed in force units of Newtons (N), was measured using a computerised testing machine (SV-H1000, Japan Instrumentation System Co., Tokyo, Japan) (Kim et al., 2015a; Kim et al., 2015b). Briefly, animals were restrained in the machines using two separate 1-0 silk suture ties on the left ankle and chest, and the peak tensile loads were recorded when knee angles reached 0° (10−20-mm distances). After the left gastrocnemius and soleus were exposed, their thickness was measured directly using a caliper. The muscle samples were weighed and expressed as an amount (g) or a percentage of the body weight. 2.5. Plasma biochemistry. Blood samples were centrifuged at 1,200 × g for 10 min, and the resulting plasma was collected. The plasma samples were stored at −150°C until analysis. Samples were examined for levels of plasma creatine, CK and lactate dehydrogenase (LDH) using an autoanalyser (Dri-Chem NX500i, Fuji Medical System Co., Ltd., Tokyo, Japan). 2.6. Anti-oxidant defense systems. Muscle samples of the gastrocnemius and soleus were homogenised in ice-cold 0.01 M Tris-HCl (pH 7.4), and then centrifuged at 12,000 × g for 15 min. Total proteins were measured using bovine albumin as the standard (Invitrogen, Carlsbad, CA, USA). Levels of malondialdehyde (MDA), indicating lipid peroxidation, were determined using the thiobarbituric acid test. The 8

level was assessed by absorbance at 525 nm with an ultraviolet/visible spectrometer (Optizen Pop, Mecasys, Daejeon, Korea), and expressed as nM of MDA/g tissue protein (Jamall and Smith, 1985). Levels of ROS were measured at 490/520 nm using 2′,7′-dichlorofluorescein diacetate (DCFDA)-Cellular Reactive Oxygen Species Detection Assay Kit according to the manufacturer’s protocol (ab113851, Abcam, Cambridge, MA, USA). The signal was expressed as relative fluorescence unit (RFU) per μg tissue protein. Glutathione (GSH) content was measured by the absorbance at 412 nm using 2-nitrobenzoic acid (Sigma-Aldrich, St. Louis, MO, USA), and expressed as mg/g tissue (Sedlak and Lindsay, 1968). Catalase (CAT) activity was defined as the amount of enzyme required to decompose 1 nM of H2O2 per min, expressed as U/mg tissue protein. Superoxide dismutase (SOD) activity was measured at 560 nm by the degree of inhibition of the reaction with nitrotetrazolium blue, expressed as U/mg tissue protein (Sun et al., 1988). One unit of SOD enzymatic activity is equal to the amount of enzyme that diminishes the initial absorbance of nitrotetrazolium blue by 50% in 1 min. 2.7. Real-time reverse transcription-polymerase chain reaction (RT-PCR). RNA was extracted from muscle homogenates using the Trizol reagent (Invitrogen, Carlsbad, CA, USA). RNA concentrations and quality were determined with a CFX96TM Real-Time System (Bio-Rad, Hercules, CA, USA). To remove contaminating DNA, samples were treated with recombinant DNase I (DNA-free; Ambion, Austin, TX, USA). RNA was reverse-transcribed using the reagent High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA) according to the manufacturer’s instructions. The oligonucleotide primers for eight targets of atrogin-1, muscle RING-finger protein-1 (MuRF1), phosphatidylinositol 3-kinase (PI3K) p85α, Akt1, adenosine A1 receptor (A1R), transient receptor potential cation cannel subfamily V member (TRPV) 4, myostatin, and sirtulin (SIRT) 1 are listed in Table A.1. The 18s 9

ribosomal RNA was used as an internal control.

Table A.1. Oligonucleotides used for real-time RT PCR. Target

Direction

Primers (5’-3’)

F

CAGCTTCGTGAGCGACCTC

R

GGCAGTCGAGAAGTCCAGTC

F

GACAGTCGCATTTCAAAGCA

R

GCCTAGCACTGACCTGGAAG

F

GCCAGTGGTCATTTGTGTTG

R

ACACAACCAGGGAAGTCCAG

F

ATGAACGACGTAGCCATTGTG

R

TTGTAGCCAATAAAGGTGCCAT

F

TGTTCCCAGGGCCTTTCAC

R

TAATGGACTGAGACTAGCTTGACTGGTA

F

CAGGACCTCTGGAAGAGTGC

R

AAGAGCTAGCCTGGACACCA

F

CCTCCACTCCGGGAACTGA

R

AAGAGCCATCACTGCTGTCATC

F

TTCACATTGCATGTGTGTGG

R

TGAGGCCCAGTGCTCTAACT

18s Ribosomal

F

AGCCTGAGAAACGGCTACC

RNA

R

TCCCAAGATCCAACTACGAG

Atrogin-1 MuRF 1 PI3K p85α Akt 1 Adenosine A1R TRPV4 Myostatin SIRT1

Size

Gene ID

244 bp

67731

194 bp

433766

236 bp

18708

116 bp

11651

155 bp

11539

165 bp

63873

185 bp

17700

175 bp

93759

252 bp

19791

RT-PCR = reverse transcription polymerase chain reaction, F = forward, R = reverse, MuRF1 = muscle RING-finger protein-1, PI3k = phosphatidylinositol 3-kinase, A1R = A1 receptor, TRPV4 = transient receptor potential cation cannel subfamily V member 4, SIRT1 = Sirtulin 1

2.8. Histopathological analysis. Part of the gastrocnemius and soleus were fixed and serially paraffin-sectioned at a thickness of 3−4 μm. The sections were stained with haematoxylin and eosin (HE) for general histopathology or Sirius red for collagen fibres, and they were observed by a histopathologist blinded to the 10

groups under a light microscope (Eclipse 80i, Nikon, Tokyo, Japan). The mean diameters of muscle fibres (μm/fibre) and regions of collagen fibre (%/mm2 of muscle bundles) in muscle bundles were assessed using an automated image analyser (iSolution FL ver 9.1, IMT i-solution Inc., Quebec, Canada) (Kim et al., 2015a; Kim et al., 2015b). 2.9. Immunohistochemical analysis. Serial sections were pretreated with 10 mM citrate buffer (pH 6.0) at 95−100°C for 20 min for antigen retrieval. Endogenous peroxidase activity was blocked by incubation with 0.3% H2O2 in methanol for 30 min, and non-specific binding was blocked by incubation with normal horse serum (Vector Lab., Burlingame, CA, USA. dilution 1:100) for 1 h. Sections were immunostained for ATPase, myostatin, caspase-3, poly (ADP-ribose) polymerase (PARP), nitrotyrosine (NT), 4- hydroxynonenal (4-HNE), and inducible nitric oxide synthase (iNOS) using Vectastain Elite ABC Kits (Vector Lab.). The antibodies used are listed in Table A.2. Cells or muscle fibres occupying more than 20% of the immunoreactivity were regarded as positive, and measured using an automated image analysis process. They were expressed as cell number or fibre area in mm2 of muscle bundle. The histopathologist was blinded to the groups. Table A.2. Primary antibodies used for immunohistochemistry

Antibodies Anti-SERCA2 ATPase antibody Anti- caspase-3 antibody

Product company (Ca. No.)

Dilution

Abcam, Cambridge, UK (ab2861)

1:100

Cell Signaling Technology Inc., Danvers, MA, USA (9661)

1:400

Anti-PARP antibody

Cell Signaling Technology Inc. (9661)

1:100

Anti-4HNEantibody

Abcam (Ab46545)

1:100

Anti-nitrotyrosine antibody

Millipore

Co.,

(06-284)

11

Billerica,

CA,

USA

1:200

Anti-iNOS antibody

Santa Cruz Biotechnology, Burlingame, CA, USA (sc-651)

1:100

Anti-GDF8/Myostatin antibody

Abcam (Ab71808)

1:50

All antibodies were diluted by 0.01 M phosphate buffered saline. SERCA = sarco-endoplasmic reticulum calcium ATPase, PARP = poly (ADP-ribose) polymerase, HNE = hydroxynonenal, iNOS = inducible nitric oxide synthase

2.10. Statistical analyses. All data are expressed as means ± standard deviation. First, data were analysed by the Levene test for homogeneity of variance. If the test indicated no significance, the values were examined by one way-analysis of variance (ANOVA), followed by the least significant differences post hoc test. However, if there was significance, the Kruskal−Wallis H test was conducted for non-parametric comparisons, followed by the Mann−Whitney U post hoc. The kinetics for the body weight and the calf thickness were analysed by two-way ANOVA with a main factor of group; the day measured was treated as a repeated measurement. The analyses focused on the differences among treatment groups. A p-value < 0.05 was considered statistically significant.

12

3. Results 3.1. Effects of SFe on muscle composition 3.1.1. Changes in body weight and weights of gastrocnemius and soleus Two-way ANOVA for the kinetics of body weight showed no significant effects for group and no interactions between group and day (Figure A.2).

However, there were significant main effects

for group in the absolute weights of the gastrocnemius (F=33.9; p<0.01) and soleus (F=17.0; p<0.01) (Table 1). The absolute weights were increased significantly in both muscles of the SW group versus Intact, and they were increased further in the SFe and OXY groups (p<0.05). The relative weight to the body weight in the SW group versus Intact was increased by 10.7% and 32.7% in the gastrocnemius and soleus, respectively. The relative weights of the SFe (SFe500, SFe250, and SFe125) and OXY groups versus SW showed an even higher increase by 24.4%, 17.0%, 14.6%, and 19.9%, respectively, in the gastrocnemius, and 37.5%, 29.0%, 21.7%, and 38.3%, respectively, in the soleus (p<0.05).

Figure A.2. Body weights. Mice body weight was measured once a week and expressed as means ± SD in 8 mice

13

Table 1. Changes in muscle weight

Muscle weights, g (%) Gastrocnemius Intact SW

Soleus

0.1740.007 (0.4340.030) *

0.1980.011

0.0100.002

(0.0260.006) **

0.0140.001 (0.0350.006**)

(0.4810.037)

OXY

0.2380.021*,† (0.5760.042*,†)

0.0190.003**,‡ (0.0480.012*,†)

SFe500

0.2380.009*,† (0.5980.076*,†)

0.0190.003**,‡ (0.0480.007*,†)

SFe250

0.2280.009*,† (0.5620.052*,†)

0.0180.003**,‡ (0.0450.004*,†)

SFe125

0.2220.012*,† (0.5510.057*,†)

0.0170.002**, ‡ (0.0420.006*,‡)

Muscle weight (g) and a percentage to body weight are expressed as mean  standard deviation (SD) in 8 mice. * and **: p<0.01 and p<0.05, respectively, vs. Intact and † and ‡: p<0.01 and p<0.05, respectively, vs. SW.

14

3.1.2. Lean mass in total body and calf muscle DEXA digital radiography showed an increase in lean mass in the SW group compared with that of the Intact, with further increases noted in the SFe and OXY groups compared with SW (Fig. 1A). One-way ANOVA showed significant main effects of group for the lean amounts of total body (F=14.0; p<0.01) and calf muscle (F=36.7; p<0.01) (Fig. 1B). The post hoc tests versus

Figure 1. DEXA densitometric analyses for lean body mass.

Lean body mass was examined by dual energy X-ray absorptiometry (DEXA) digital radiography on day 28 post-treatment, and representative images for bone and tissue are shown in A. Densitometric analyses for amount of lean body mass (B) and its proportion to total body mass (C) were performed in the total body (black bars) and hind limbs between the knee and ankle (white bars). Values are expressed as means ± standard deviation (SD) in 8 mice. *: p<0.01 versus Intact and †: p<0.01 versus SW.

15

Intact showed significant increases in the amounts of both total body and calf muscle in the SW group (p<0.05); multi-comparisons versus SW showed further increases in the amount in the SFe and OXY groups (p<0.05). The proportion of lean mass to total amounts including bone mineral and fat also showed significant increases in the total body and calf muscle of the SW group versus Intact, and further increases in the SFe and OXY groups versus SW (p<0.05) (Fig. 1C). 3.2. Effects of SFe on muscle thickness 3.2.1. Thickness of the calf muscle Two-way ANOVA for the kinetics of calf muscle thickness showed significant effects for group (F=10.6; p<0.01) and significant interactions between group and day (F=153.9; p<0.01) (Fig. 2A). Figure 2. Thickness of the calf muscle.

The calf thickness was measured once a week post-treatment, and the kinetic thickness (A) and the changes on days 0−28 post-treatment are expressed as means ± SD in 8 mice. *: p<0.01 versus Intact and †: p<0.01 versus SW.

16

The post hoc tests versus Intact showed significant increases in the SW group on days 21−28 (p<0.01). However, the calf thickness increased more in the SFe groups on days 21−28 and in the OXY group on days 14−28, compared with the SW group (p<0.05). The increased thickness for 28 days versus Intact was increased by 8.9-fold in the SW group, and the increased thickness versus SW was increased more by 2.0-, 1.8-, 1.6- and 2.4-fold in the SFe500, SFe250, SFe125, and OXY groups, respectively (p<0.05) (Fig. 2B).

17

3.2.2. Thicknesses of the gastrocnemius and soleus The thicknesses of the gastrocnemius and soleus consisting of the calf muscle mass were measured directly in the post-mortem samples after treatments for 28 days. The post-mortem values exhibited increased muscle mass in the SW group compared with Intact, and the muscle Figure 3. Muscle thickness of the gastrocnemius and soleus.

The calf muscle was exposed on day 28 post-treatment, and the component muscles of the gastrocnemius (GC) and soleus (Sol) was sampled (A). Scale bars = 5.5 mm. The thicknesses of the gastrocnemius (B) and soleus (C) were assessed and expressed as means ± SD in 8 mice. *: p<0.01 versus Intact and † and ‡: p<0.01 and p<0.05, respectively, versus SW.

mass was further increased in the SFe and OXY groups compared with SW (Fig. 3A). Non-parametric multiple analyses for muscle thickness showed significant main effects for group 18

in the gastrocnemius and soleus (p<0.01) (Fig. 3B and C). The gastrocnemius thickness versus Intact was non-significantly increased by 1.2-fold in the SW group, however, the soleus thickness increased significantly by 1.1-fold (p<0.05). The thicknesses of both muscles versus SW showed small but significant increases in the SFe and OXY groups (p<0.05). 3.3. Effects of SFe on muscle strength Muscle strength was assessed by contraction intensity of the calf muscle. There was significant main effect of group (F=16.7; p<0.01) (Fig. 4). The post hoc tests versus Intact showed a significant increase by 1.15-fold in the SW group, and the multi-comparisons versus SW showed Figure 4. Muscle strength.

Contractive strength of the hind limbs was assessed before euthanasia on day 28 post-treatment. Values are expressed as means ± SD in 8 mice. * and **: p<0.01 and p<0.05, respectively, versus Intact and †: p<0.01 versus SW.

further increases by 1.30-, 1.20- and 1.32-fold, in the SFe500, SFe250 and OXY groups, respectively (p<0.05). The SFe125 group also showed an increase but it was not different from that of the SW. 3.4. Effects of SFe on plasma biochemistry

19

Table 2. Changes in serum biochemistry

Creatine (mg/dl) Intact

Creatine kinase (IU/l)

0.380.05

98.639.35 *

LDH (IU/l) 174.2534.05

*

326.8847.45*

SW

0.650.09

OXY

0.480.08*,†

118.6313.32*,†

215.8831.85**,†

SFe500

0.490.06*,†

127.5017.96*,†

219.2515.08**,†

SFe250

0.510.05*,†

133.2512.80*,†

238.5034.94*,†

SFe125

0.540.07*,†

136.639.21*,†

265.7532.20*,†

164.6320.78

Values are expressed as mean  SD in 8 mice. * and **: p<0.01 and p<0.05, respectively, vs. Intact and †: p<0.01 vs. SW. LDH = lactate dehydrogenase.

There were significant effects of group on levels of creatine (F=13.9; p<0.01), creatine kinase (F=18.0; p<0.01), and LDH (F=18.8; p<0.01) (Table 2). The three values versus Intact were increased in the SW. However, the contents versus SW were decreased significantly in the SFe and OXY groups (p<0.05). 3.5. Effects of SFe on anti-oxidant defense systems

Table 3. Changes in muscular antioxidant defense system

MDA (nM/mg)

ROS (RFU/μg)

GSH (nM/mg)

Gastrocnemius Intact 3.290.50

26.784.20

0.690.12

SOD (U/mg)

29.187.40

6.591.41

SW

6.941.31*

72.6916.73*

0.260.05*

OXY

4.620.97*,†

38.1110.10†

0.480.11*,†

20.342.39**,†

3.330.62*,†

SFe500

4.420.90**,†

36.7611.07†

0.460.08*,†

20.052.63**,†

3.490.68*,†

SFe250

5.110.76*,†

45.0810.99*,†

0.430.13*,†

18.033.06*,†

3.320.68*,†

SFe125

4.420.90**,†

36.7611.07†

0.460.08*,†

20.052.63**,†

3.490.68*,†

Soleus

20

9.981.84*

CAT (U/mg)

2.100.51*

Intact

2.850.63

31.136.50 *

0.660.11 *

SW

5.931.14

OXY

3.730.67**,†

SFe500

3.680.76**,†

SFe250

4.120.64*,†

44.3611.57**,† 51.875.77*,†

SFe125

4.540.47*,†

56.3411.13*,†

78.4010.82

44.6612.29**,†

0.320.08

27.568.21 *

10.271.75

*

5.650.75 2.350.46*

0.550.13**,†

19.171.92*,†

3.860.37*,†

0.540.10**,†

18.702.41*,†

3.960.58*,†

0.480.09*,†

16.221.75*,†

3.160.38*,†

0.440.06*,‡

14.792.25*,†

2.950.15*,†

Values are expressed as mean  SD in 8 mice. * and **: p<0.01 and p<0.05, respectively, vs. Intact and † and ‡: p<0.01 and p<0.05, respectively, vs. SW. MDA = malondialdehyde, ROS = reactive oxygen species, GSH = glutathione, SOD = superoxide dismutase, CAT = catalase

Levels of MDA and ROS were assessed for oxidative stress. There were significant effects of group for levels of MDA in the gastrocnemius (F=15.7; p<0.01) and soleus (F=15.6; p<0.01) (Table 3). There were also significant effects for group in the ROS levels in the gastrocnemius (F=15.5; p<0.01) and soleus (F=20.1; p<0.01). The levels of MDA and ROS in both muscles were increased in the SW group compared with the Intact group. However, they were reduced in the SFe and OXY groups compared with SW (p<0.05). In particular, the SFe500 and OXY groups showed similar levels of MDA and ROS in the gastrocnemius with those in the Intact group. For muscular anti-oxidant responses, contents of the endogenous anti-oxidant, GSH, and activities of the anti-oxidant enzymes, SOD and CAT, were assessed. One-way ANOVA showed significant effects of group for the contents of GSH in the gastrocnemius (F=14.3; p<0.01) and soleus (F=11.3; p<0.01). Non-parametric analyses also showed significant effects of group for the activities of SOD and CAT in both muscles (p<0.01). The post hoc tests versus Intact revealed significant decreases in the contents of GSH and activities of SOD and CAT in both muscles in the SW group (p<0.05). However, the anti-oxidant activities were increased in the SFe and OXY groups compared with those of the SW group. 21

3.6. Effects of SFe on mRNA expression relevant to muscle growth Expression levels of mRNAs involved in protein synthesis (Akt1, PI3K) or degradation (Atrogin-1, MuRF1) and muscle growth activation (A1R, TRPV4) or inhibition (myostatin, SIRT1) were assessed in the gastrocnemius and soleus (Table 4). Multiple-comparisons showed significant differences in the mRNA levels in both muscles among groups (p<0.01), and the SFe groups showed tendencies for dose-dependent differences in the expression levels compared with those of the SW group. The SW group showed significant increases in the levels of mRNAs for Akt1 and PI3K p85α, whereas it showed decreases in the levels of atrogin-1 and MuRF1, in both muscles compared with the Intact group (p<0.05). However, when compared with the SW group, the levels of atrogin-1 and MuRF1 were further increased and the levels of Akt1 and PI3K p85α were lower in both muscles in the SFe and OXY groups (p<0.05). Similarly, the SW group versus Intact showed significant increases in the levels of A1R and TRPV4 and decreases in the levels of myostatin and SIRT1, in both muscles (p<0.05). The expressions levels versus SW were higher for A1R and TRP4 and lower for myostatin and SIRT1, in both muscles of the SFe and OXY groups (p<0.05).

22

Table 4. Expression levels of mRNAs involved in muscle growth and protein synthesis

Akt1 Gastrocnemius Intac 1.00±0.07 t SW

OX Y SFe 500 SFe 250 SFe 125 Soleus Intac t SW

OX Y SFe 500

PI3K p85α

Atrogin-1

MuRF1

0.99±0.11

1.01±0.11 1.02±0.09

1.55±0.20

1.78±0.21

0.85±0.0

*

*

6

2.20±0.27

2.51±0.40

*,†

0.85±0.07

1.01±0.1 7 1.48±0.1

1.00±0.12

1.53±0.17

Myostati n

SIRT1

1.01±0.0

0.97±0.1

7

3

0.84±0.0

0.82±0.0

8

0.57±0.0

0.53±0.18

*,†

9*,†

2.21±0.50

2.54±0.46

0.56±0.0

*,†

*,†

2.10±0.36

2.35±0.37

0.61±0.0

0.64±0.10

1.87±0.2

1.96±0.20

0.60±0.0

0.60±0.1

*,†

*,†

9*,†

*,†

2*,†

*,†

7*,†

0*,†

1.86±0.14

2.11±0.16

0.67±0.1

0.71±0.10

1.75±0.1

1.87±0.16

0.66±0.0

0.66±0.0

*,†

*,†

0.95±0.13

0.99±0.10

1.37±0.10

1.38±0.09

*

*

7

1.80±0.19

1.96±0.17

0.62±0.0

*,†

*,†

1.71±0.10

1.84±0.20

*,†

*,†

2

*,†

*,†

1.01±0.0 8 0.82±0.0

9

*

*,†

0.63±0.0 9

*,†

*

6

2.10±0.1

2.18±0.40

0.50±0.1

0.56±0.1

*,†

4*,†

*,†

0*,†

3*,†

0.54±0.08

2.09±0.2

2.15±0.21

0.53±0.1

0.56±0.0

*,†

*,†

0.98±0.09

0.76±0.10

4

8

*

TRPV4

*

9

*

A1R

*,†

*,†

1.02±0.0 9 1.39±0.1

*

3

0.55±0.09

1.93±0.1

*,†

0.56±0.08 *,†

23

6

*

*,†

1.86±0.0 9

*,†

*,†

*,†

1.04±0.14

1.36±0.33 **

2

6

*

*,†

*,†

9*

7*,†

9*,†

1.02±0.0

1.04±0.1

8

5

0.82±0.0

0.85±0.0

6

*

8*

2.22±0.45 0.54±0.11 0.60±0.11 *,†

*,†

*,†

2.12±0.32

0.52±0.1

0.58±0.0

*,†

7

*,†

9*,†

SFe 250 SFe 125

1.60±0.18

1.73±0.16 0.66±0.11 0.60±0.08

1.77±0.1

1.88±0.11 *,†

*,†

*,†

1.55±0.10

1.54±0.11

0.70±0.0

0.65±0.06

1.68±0.1

1.77±0.16

0.64±0.1

0.69±0.0

*,**

*,d

9*,d

*,†

2*,†

*,†

0*,†

7*,†

8

*,†

0.65±0.11

*,†

4

*,†

0.61±0.0

*,†

*,†

Relative expressions to 18s ribosomal RNA is expressed as mean  SD in 8 mice. * and **: p<0.01 and p<0.05, respectively, vs. Intact and † and ‡: p<0.01 and p<0.05, respectively, vs. SW. MuRF1 = muscle RING-finger protein-1, PI3k = phosphatidylinositol 3-kinase, A1R = adenosine A1 receptor, TRPV4 = transient receptor potential cation cannel subfamily V member 4 and SIRT1 = sirtulin 1

3.7. Histopathological and immunohistochemical analyses 3.7.1. Changes in myofibres and collagen deposition In the gastrocnemius and soleus of the SW group versus Intact, HE staining exhibited hypertrophic changes in the muscle fibres, and Sirius red staining showed decreases in the collagen-occupied region (Fig. 5A). However, the tendencies for hypertrophic changes and a reduced collagen-occupied region were observed much more in the SFe and OXY groups than in the SW group. One-way ANOVA for the histomorphometric analyses showed significant main effects of group for diameters of muscle fibres in the gastrocnemius (F=32.2; p<0.01) and soleus (F=82.2; p<0.01) (Fig. 5B). The post hoc tests revealed significant increases in the SW group versus Intact, and in the SFe and OXY groups versus SW (p<0.01). Non-parametric analyses also showed significant differences by group for the collagen-occupied fibre regions (p<0.05) (Fig. 5C). The region was significantly reduced in the SW group versus Intact (p<0.01), and reduced further in the SFe and OXY groups versus the SW group (p<0.05).

24

Figure 5. Histopathological analyses.

Samples of gastrocnemius and soleus were serially paraffin-sectioned, and stained with haematoxylin and eosin (HE) or Sirius red (A). Scale bars = 40 μm. Representative HE- and Sirius red-stained sections were analysed for fibre diameters (B) and red-coloured collagen (C), respectively. Values are expressed as means ± SD in 8 mice. *: p<0.01 versus Intact and † and ‡: p<0.01 and p<0.05, respectively, versus SW.

25

3.7.2. Changes in expression levels of ATPase and myostatin in muscle fibres Muscle fibres immunoreactive for ATPase and myostatin were assessed as makers for fast-twitched white muscle fibres and a potent negative regulator, respectively (Fig. 6A). The gastrocnemius and soleus of the SW group versus Intact exhibited increases in the expression of ATPase and decreases in the expression of myostatin. However, treatments with SFe and OXY showed

increases

in

the

ATPase-immunostained

myofibres

but

decreases

in

the

myostatin-immunostained myofibres, compared with the SW group. Multi-comparisons among Figure 6. Immunohistochemistry for ATPase and myostatin.

Serial sections of the gastrocnemius and soleus were immunostained for ATPase and myostatin (A). Scale bars = 40 μm. The muscle fibres immunostained for ATPase (B) and myostatin (C) are expressed as means ± SD in 8 mice. *: p<0.01 versus Intact and †: p<0.01 versus SW.

groups revealed significant increases in the muscle fibres immunoreactive for ATPase and myostatin in both muscles of the SW group versus Intact (p<0.01) (Figs. 6B and C). However, 26

the SFe and OXY groups showed increases in ATPase but decreases in myostatin compared with the SW group (p<0.01). Figure 7. Immunohistochemistry for caspase-3 and PARP.

Serial sections of the gastrocnemius and soleus were immunostained for caspase-3 and PARP (A). Scale bars = 40 μm. The muscle fibres immunostained for caspase-3 (B) and poly (ADP-ribose) polymerase (PARP, C) are expressed as means ± SD in 8 mice. * and **: p<0.01 and p<0.05, respectively, versus Intact and † and ‡: p<0.01 and p<0.05, respectively, versus SW.

3.7.3. Changes in the expression of caspase-3 and PARP in muscle fibres The muscle fibres immunostained for caspase-3 and PARP as makers for apoptosis, were increased in the gastrocnemius and soleus of the SW group compared with those of the Intact group (Fig. 7A). Conversely, the expressions were decreased in both muscles in the SFe and OXY groups. Histomorphometric analyses revealed significant increases in the caspase-3 and PARP specific muscle fibres of the gastrocnemius and soleus in the SW group versus Intact (p<0.01) (Figs. 7B and C). However, the immunostained muscle fibres were decreased 27

significantly in both muscles in the SFe and OXY groups versus the SW group (p<0.05). Interestingly, the expressions were not different in the gastrocnemius of the SFe500 and OXY groups versus Intact. Figure 8. Immunohistochemistry for NT, 4HNE, and iNOS.

Serial sections of the gastrocnemius and soleus were immunostained for nitrotyrosine (NT), 4- hydroxynonenal (HNE), and inducible nitric oxide synthase (iNOS) (A). Scale bars = 40 μm. The muscle fibres immunostained for NT (B), 4HNE (C), and iNOS (D) are expressed as means ± SD in 8 mice. * and **: p<0.01 and p<0.05, respectively, versus Intact and † and ‡: p<0.01 and p<0.05, respectively, versus SW.

3.7.4. Changes in the expression of NT, 4HNE, and iNOS in muscle fibres Similar to the apoptosis data, the expression levels of NT, 4HNE, and iNOS as markers for 28

oxidative stress, were increased in the gastrocnemius and soleus of the SW group versus Intact, however, they decreased in the SFe and OXY groups versus SW (Fig. 8A). Histomorphometric analyses showed significant increases in NT-, 4HNE-, and iNOS-specific muscle fibres in the gastrocnemius and soleus in the SW group versus Intact (p<0.01) (Fig. 8A-D). However, the immunostained muscle fibres were significantly decreased in the SFe and OXY groups versus SW group (p<0.5). In particular, the NT specific muscle fibres were not different in the gastrocnemius of the SFe500 and OXY groups versus the Intact group.

4. Discussion Loss of skeletal muscle associated with aging reduces physical activity, leading to muscle atrophy, characterized by reduced myofibrillar protein, organelles, and cytoplasm (Bassel-Duby and Olson, 2006; Glass, 2003). Many factors are involved in muscle atrophy, such as denervation, musculoskeletal injuries, ligament and joint injuries, glucocorticoid treatment, sepsis and cancer (Glass, 2005). Although previous studies have shown that SFe ameliorates the muscular atrophy induced by dexamethasone or denervation (Kim et al., 2015a; Kim et al., 2015b), how SFe influences in the normal muscular changes after exercise training in old-ages was enigmatic. Since aerobic physical exercise is currently believed to increase the cross-sectional area of muscle fibres and improve muscle function (Montero-Fernandez and Serra-Rexach, 2013), any substances enhancing the exercise effects could be also applied to various cardiometabolic disease and chronic inflammation as an adjuvant therapy. We used forced-swimming aged mice model for the aerobic exercise. The model mice indeed exhibited increases in lean body mass, muscle thickness and strength and weights in the gastrocnemius and soleus. The exercise effects seemed to involve increased gene expression for muscle growth activation and protein synthesis 29

but decreased expression for muscle growth inhibition and protein degradation. Interestingly, the oral administration of SFe enhanced the exercise effects further. However, while contents of plasma creatine, creatine kinase, and LDH and oxidative stress of muscular MDA and ROS, were increased in the SW group, they were reduced in the SFe groups and muscle GSH, SOD, and CAT levels were increased. In addition, histopathological and immunohistochemical analyses revealed the muscle-protective effects of SFe in the forced-swimming aged model, most likely mediated through anti-oxidant and anti-apoptotic pathways. The contractive strength of muscles is determined by the activity of myosin ATPase (Trapani et al., 2011). Skeletal muscle fibres can be divided into slow-twitched red muscle fibres (e.g., the soleus) and fast-twitched with muscle fibres (e.g., the gastrocnemius) according to myosin ATPase subtypes. Generally, it is known that slow-twitched red muscles have relatively stronger aerobic capacity and higher resistance to fatigue whereas fast-twitched white muscles have relatively faster contraction (Schiaffino and Reggiani, 1996). Here, increases in muscle thickness and weight, along with anti-oxidant activities, were slightly higher in the soleus than the gastrocnemius. However, the changes were still significant in both muscles of the SW group versus Intact or the SFe groups versus the SW group. Similarly, the muscle fibres immunostained for ATPase were increased in both muscles of the SW group, especially the soleus, and they were increased further in the SFe groups. In addition, the muscle fibres immunostained for myostatin, a member of the TGF-β protein family that inhibits myogenesis (Gilson et al., 2007), were increased in both the gastrocnemius and soleus of the SW group, however, they were decreased in the SFe groups. By the way, there were inconsistency in the expression levels of myostatin of the SW group between protein and mRNA. This difference was probably due to the time gap between gene transcription and protein translation for muscle remodeling; protein expression of 30

myostain might be increased to inhibit myogenesis after the forced swimming, and then its gene expression might be reduced in response to the increased levels. For any reason, along with the mRNA and protein expression for myostatin, other mRNAs related to muscle growth inhibition and protein degradation decreased more in the SFe groups than the SW group, while expressions related to muscle growth activation and protein synthesis were increased more. The data from the immunostains and mRNA expression provide evidences supporting the increased muscle mass and strength in the SFe groups compared with the SW group. Since an estimated 98% of total body creatine is found in skeletal muscle and plasma activities of creatine kinase and LDH are used as makers of muscle damages (Wyss and Kaddurah-Daouk, 2000; Zhang et al., 2012), their increased plasma levels in the SW group indicated the skeletal muscle damage. In this context, the significant decreases in their levels in the SFe groups versus SW group suggest preventative effects on skeletal muscle fatigue and subsequent damage. This may be related to downregulation of mRNAs involved in protein degradation and muscle growth inhibition in the SFe groups. Furthermore, the significant decreases in the muscle fibres immunostained for early apoptosis markers (caspase, PARP) of the SFe groups support the reduced muscle damages from the biochemical analyses in serum. In particular, the tendency of apoptosis was similar in the gastrocnemius and soleus of the SFe500 and OXY groups compared with the Intact group which received no swimming exercise. Thus, consistent with other studies (Arakawa et al., 2010; Dam et al., 2012; Yen et al., 2012), the administration of SFe may contribute to inhibiting atrophic changes in muscles in the early phase and the resulting muscle damages through anti-apoptosis. It is well known that excessive and chronic exercise induces inflammation as well as several metabolic changes that disrupt mitochondrial functioning, increasing ROS production (Barcelos 31

et al., 2014; Davis et al., 2007). Because the oxidative stress and various toxic substances resulting from lipid peroxidation are involved in impairing function of the skeletal muscle parenchyma, they are considered to be important inducers of muscle atrophy (Powers et al., 2007). In the present study, the SW group showed overproduction of muscular ROS and lipid peroxidation (MDA) and increased expression of NT, 4HNE, and iNOS, markers of oxidative stress, in muscle fibres. However, physiological defence activities, including endogenous anti-oxidants (GSH) and anti-oxidant enzymes (SOD and CAT) were lower. Conversely, the administration of SFe lowered ROS and MDA, while it promoted anti-oxidant activities regardless of muscle fibre types, in agreement with previous studies (Kim et al., 2015a; Kim et al., 2015b). Indeed, SFe has been reported to contain bioactive components including lignans such as schizandrins and gomisins, some of them have anti-inflammatory properties (Oh et al., 2010) and improve anti-oxidant capacity (Chang et al., 2009). Although the exact mechanisms of individual constituents of SFe remain unclear here, the anti-inflammatory and anti-oxidant effects of SFe may contribute to anti-apoptotic effects, leading to protective effects against muscle damage. In conclusion, SFe enhanced adaptive muscle strengthening induced by chronic forced swimming exercise through potent anti-oxidant and anti-apoptotic activities. In particular, the beneficial effects of SFe500 were similar in the mouse model to administration of oxymetholone, an anabolic steroids (Delgado et al., 2010). To date, SFe has long been used in traditional medicine as a non-toxic compound. Its bioactive properties tend to enhance physical working capacity, anti-stress response, and cardiovascular function, involved in anti-inflammatory and anti-oxidant mechanisms (Panossian and Wikman, 2008). These results suggest that SFe may be helpful for improving various muscle disorders as an adjuvant therapy to exercise-based remedies.

32

Conflict of interest The authors disclosed no conflict of interest.

Appendices Appendices of two tables (Table A.1 and Table A.2) and one figure (Figure A.1) were attached for supplementary data.

References Ali, S., Garcia, J.M., 2014. Sarcopenia, cachexia and aging: diagnosis, mechanisms and therapeutic options - a mini-review. Gerontology 60, 294-305. Arakawa, T., Katada, A., Shigyo, H., Kishibe, K., Adachi, M., Nonaka, S., Harabuchi, Y., 2010. Electrical stimulation prevents apoptosis in denervated skeletal muscle. NeuroRehabilitation 27, 147-154. Barcelos, R.P., Souza, M.A., Amaral, G.P., Stefanello, S.T., Bresciani, G., Fighera, M.R., Soares, F.A.A., de Vargas Barbosa, N., 2014. Caffeine intake may modulate inflammation markers in trained rats. Nutrients 6, 1678-1690. Bassel-Duby, R., Olson, E.N., 2006. Signaling pathways in skeletal muscle remodeling. Annu Rev Biochem 75, 19-37. Bogdanova, O.V., Kanekar, S., D'Anci, K.E., Renshaw, P.F., 2013. Factors influencing behavior in the forced swim test. Physiol Behav 118, 227-239. Brooks, S.V., Faulkner, J.A., 1994. Skeletal muscle weakness in old age: underlying mechanisms. Med Sci Sports Exerc 26, 432-439. 33

Chang, C.Y., Chen, Y.L., Yang, S.C., Huang, G.C., Tsi, D., Huang, C.C., Chen, J.R., Li, J.S., 2009. Effect of schisandrin B and sesamin mixture on CCl(4)-induced hepatic oxidative stress in rats. Phytother Res 23, 251-256. Chen, J.-C., Hsiang, C.-Y., Lin, Y.-C., Ho, T.-Y., 2014. Deer antler extract improves fatigue effect through altering the expression of genes related to muscle strength in skeletal muscle of mice. J Evid Based Complementary Altern Med 2014. Dam, A.D., Mitchell, A.S., Rush, J.W., Quadrilatero, J., 2012. Elevated skeletal muscle apoptotic signaling following glutathione depletion. Apoptosis 17, 48-60. Davis, J.M., Murphy, E.A., Carmichael, M.D., Zielinski, M.R., Groschwitz, C.M., Brown, A.S., Gangemi, J.D., Ghaffar, A., Mayer, E.P., 2007. Curcumin effects on inflammation and performance recovery following eccentric exercise-induced muscle damage. Am J Physiol Regul Integr Comp Physiol 292, R2168-2173. Delgado, J., Saborido, A., Megias, A., 2010. Prolonged treatment with the anabolic-androgenic steroid stanozolol increases antioxidant defences in rat skeletal muscle. J Physiol Biochem 66, 63-71. Gilson, H., Schakman, O., Combaret, L., Lause, P., Grobet, L., Attaix, D., Ketelslegers, J.M., Thissen, J.P., 2007. Myostatin gene deletion prevents glucocorticoid-induced muscle atrophy. Endocrinology 148, 452-460. Glass, D.J., 2003. Molecular mechanisms modulating muscle mass. Trends Mol Med 9, 344-350. Glass, D.J., 2005. Skeletal muscle hypertrophy and atrophy signaling pathways. Int J Biochem Cell Biol 37, 1974-1984. Hong, S.S., Lee, J.Y., Lee, J.S., Lee, H.W., Kim, H.G., Lee, S.K., Park, B.K., Son, C.G., 2015. The traditional drug Gongjin-Dan ameliorates chronic fatigue in a forced-stress mouse 34

exercise model. J Ethnopharmacol 168, 268-278. Jamall, I.S., Smith, J.C., 1985. Effects of cadmium on glutathione peroxidase, superoxide dismutase, and lipid peroxidation in the rat heart: a possible mechanism of cadmium cardiotoxicity. Toxicol Appl Pharmacol 80, 33-42. Kim, J.W., Ku, S.K., Han, M.H., Kim, K.Y., Kim, S.G., Kim, G.Y., Hwang, H.J., Kim, B.W., Kim, C.M., Choi, Y.H., 2015a. The administration of Fructus Schisandrae attenuates dexamethasone-induced muscle atrophy in mice. Int J Mol Med 36, 29-42. Kim, J.W., Ku, S.K., Kim, K.Y., Kim, S.G., Han, M.H., Kim, G.Y., Hwang, H.J., Kim, B.W., Kim, C.M., Choi, Y.H., 2015b. Schisandrae Fructus Supplementation Ameliorates Sciatic Neurectomy-Induced Muscle Atrophy in Mice. Oxid Med Cell Longev 2015, 872428. Kwon, D.Y., Kim, D.S., Yang, H.J., Park, S., 2011. The lignan-rich fractions of Fructus Schisandrae improve insulin sensitivity via the PPAR-γ pathways in in vitro and in vivo studies. J Ethnopharmacol 135, 455-462. Li, S.S., Chen, Z.C., Zhang, C.H., 2012. Effect of tao-hong-si-wu-tang, a traditional Chinese herbal medicine formula, on physical fatigue in mice. Afr J Tradit Complement Altern Med 10, 60-65. Lu, Y., Chen, D.F., 2009. Analysis of Schisandra chinensis and Schisandra sphenanthera. J Chromatogr A 1216, 1980-1990. Montero-Fernandez, N., Serra-Rexach, J.A., 2013. Role of exercise on sarcopenia in the elderly. Eur J Phys Rehabil Med 49, 131-143. Oe, K., Ueha, T., Sakai, Y., Niikura, T., Lee, S.Y., Koh, A., Hasegawa, T., Tanaka, M., Miwa, M., Kurosaka, M., 2011. The effect of transcutaneous application of carbon dioxide (CO(2)) on skeletal muscle. Biochem Biophys Res Commun 407, 148-152. 35

Oh, S.Y., Kim, Y.H., Bae, D.S., Um, B.H., Pan, C.H., Kim, C.Y., Lee, H.J., Lee, J.K., 2010. Anti-inflammatory effects of gomisin N, gomisin J, and schisandrin C isolated from the fruit of Schisandra chinensis. Biosci Biotechnol Biochem 74, 285-291. Panossian, A., Wikman, G., 2008. Pharmacology of Schisandra chinensis Bail.: an overview of Russian research and uses in medicine. J Ethnopharmacol 118, 183-212. Pavlatos, A.M., Fultz, O., Monberg, M.J., Vootkur, A., Pharmd, 2001. Review of oxymetholone: a 17alpha-alkylated anabolic-androgenic steroid. Clin Ther 23, 789-801; discussion 771. Powers, S.K., Kavazis, A.N., McClung, J.M., 2007. Oxidative stress and disuse muscle atrophy. J Appl Physiol (1985) 102, 2389-2397. Schiaffino, S., Reggiani, C., 1996. Molecular diversity of myofibrillar proteins: gene regulation and functional significance. Physiol Rev 76, 371-423. Sedlak, J., Lindsay, R.H., 1968. Estimation of total, protein-bound, and nonprotein sulfhydryl groups in tissue with Ellman's reagent. Anal Biochem 25, 192-205. Sun, Y., Oberley, L.W., Li, Y., 1988. A simple method for clinical assay of superoxide dismutase. Clin Chem 34, 497-500. Trapani, L., Melli, L., Segatto, M., Trezza, V., Campolongo, P., Jozwiak, A., Swiezewska, E., Pucillo, L.P., Moreno, S., Fanelli, F., Linari, M., Pallottini, V., 2011. Effects of myosin heavy chain (MHC) plasticity induced by HMGCoA-reductase inhibition on skeletal muscle functions. FASEB J 25, 4037-4047. Wood, P., Yin, J.L., 1994. Oxymetholone hepatotoxicity enhanced by concomitant use of cyclosporin A in a bone marrow transplant patient. Clin Lab Haematol 16, 201-204. Wyss, M., Kaddurah-Daouk, R., 2000. Creatine and creatinine metabolism. Physiol Rev 80, 1107-1213. 36

Yang, J.M., Ip, P.S., Che, C.T., Yeung, J.H., 2011. Relaxant effects of Schisandra chinensis and its major lignans on agonists-induced contraction in guinea pig ileum. Phytomedicine 18, 1153-1160. Yen, Y.-P., Tsai, K.-S., Chen, Y.-W., Huang, C.-F., Yang, R.-S., Liu, S.-H., 2012. Arsenic induces apoptosis in myoblasts through a reactive oxygen species-induced endoplasmic reticulum stress and mitochondrial dysfunction pathway. Arch Toxicol 86, 923-933. Young Park, J., Wook Yun, J., Whan Choi, Y., Ung Bae, J., Won Seo, K., Jin Lee, S., Youn Park, S., Whan Hong, K., Kim, C.D., 2012. Antihypertensive effect of gomisin A from Schisandra chinensis on angiotensin II-induced hypertension via preservation of nitric oxide bioavailability. Hypertens Res 35, 928-934. Zhang, J., Shi, L.L., Zheng, Y.N., 2010. Dibenzocyclooctadiene lignans from Fructus Schisandrae Chinensis improve glucose uptake in vitro. Nat Prod Commun 5, 231-234. Zhang, Y., Huang, J.-J., Wang, Z.-Q., Wang, N., Wu, Z.-Y., 2012. Value of muscle enzyme measurement in evaluating different neuromuscular diseases. Clin Chim Acta 413, 520-524.

Figure legends Figure 1. DEXA densitometric analyses for lean body mass. Lean body mass was examined by dual energy X-ray absorptiometry (DEXA) digital radiography on day 28 post-treatment, and representative images for bone and tissue are shown in A. Densitometric analyses for amount of lean body mass (B) and its proportion to total body mass (C) were performed in the total body (black bars) and hind limbs between the knee and ankle (white bars). Values are expressed as means ± standard deviation (SD) in 8 mice. *: p<0.01 versus Intact and †: p<0.01 versus SW. 37

Figure 2. Thickness of the calf muscle. The calf thickness was measured once a week post-treatment, and the kinetic thickness (A) and the changes on days 0−28 post-treatment are expressed as means ± SD in 8 mice. *: p<0.01 versus Intact and †: p<0.01 versus SW.

Figure 3. Muscle thickness of the gastrocnemius and soleus. The calf muscle was exposed on day 28 post-treatment, and the component muscles of the gastrocnemius (GC) and soleus (Sol) was sampled (A). Scale bars = 5.5 mm. The thicknesses of the gastrocnemius (B) and soleus (C) were assessed and expressed as means ± SD in 8 mice. *: p<0.01 versus Intact and † and ‡: p<0.01 and p<0.05, respectively, versus SW.

Figure 4. Muscle strength. Contractive strength of the hind limbs was assessed before euthanasia on day 28 post-treatment. Values are expressed as means ± SD in 8 mice. * and **: p<0.01 and p<0.05, respectively, versus Intact and †: p<0.01 versus SW.

Figure 5. Histopathological analyses. Samples of gastrocnemius and soleus were serially paraffin-sectioned, and stained with haematoxylin and eosin (HE) or Sirius red (A). Scale bars = 40 μm. Representative HE- and Sirius red-stained sections were analysed for fibre diameters (B)

and red-coloured collagen (C), respectively. Values are expressed as means ± SD in 8 mice. *: p<0.01 versus Intact and † and ‡: p<0.01 and p<0.05, respectively, versus SW.

Figure 6. Immunohistochemistry for ATPase and myostatin. Serial sections of the gastrocnemius and soleus were immunostained for ATPase and myostatin (A). Scale bars = 40 μm. The muscle fibres immunostained for ATPase (B) and myostatin (C) are expressed as means ± SD in 8 mice. *: p<0.01 versus Intact and †: p<0.01 versus SW.

Figure 7. Immunohistochemistry for caspase-3 and PARP. Serial sections of the gastrocnemius and soleus were immunostained for caspase-3 and PARP (A). Scale bars = 40 μm. The muscle fibres immunostained for caspase-3 (B) and poly (ADP-ribose) polymerase (PARP, 38

C) are expressed as means ± SD in 8 mice. * and **: p<0.01 and p<0.05, respectively, versus Intact and † and ‡: p<0.01 and p<0.05, respectively, versus SW.

Figure 8. Immunohistochemistry for NT, 4HNE, and iNOS. Serial sections of the gastrocnemius and soleus were immunostained for nitrotyrosine (NT), 4- hydroxynonenal (HNE), and inducible nitric oxide synthase (iNOS) (A). Scale bars = 40 μm. The muscle fibres immunostained for NT (B), 4HNE (C), and iNOS (D) are expressed as means ± SD in 8 mice. * and **: p<0.01 and p<0.05, respectively, versus Intact and † and ‡: p<0.01 and p<0.05, respectively, versus SW.

Graphical abstract

39