The CT26 colon adenocarcinoma and C2C12 cell lines were obtained from the American Type Culture Collection (Manassas, VA, USA). These cells were confirmed to be without mycoplasma contamination by using a color one-step mycoplasma detection kit [Yeasen Biotechnology (Shanghai) Co., Ltd., Shanghai, China]. The cells were maintained in Dulbecco's modified Eagle's medium (DMEM; Corning Life Sciences, Corning, NY, USA) supplemented with 10% heat-inactivated fetal bovine serum (Zhejiang Tianhang Biotechnology Co., Ltd., Hangzhou, China), 100 U/ml penicillin and 100 µg/ml streptomycin. The cells were cultured in an incubator at 37°C in a humidified atmosphere of 5% CO₂. Myotubes were induced by culturing the C2C12 cells in DMEM with 2% horse serum (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) for 3–5 days, replacing the medium every 24 h.

Three C2C12 myotube atrophy models were established, according to previous studies (31–33). Briefly, C2C12 myoblasts were differentiated to myotubes by culturing in 2% horse serum at 37°C. Dex (National Institute for the Control of Pharmaceutical and Biological Products, Beijing, China) or TNFα (Novus Biologicals, LLC, Littleton, CO, USA) were then added to the media for 48 h at 100 µM and 50 ng/ml, respectively. For the third model, the myotubes were incubated for 48 h in CM consisting of 33% ‘cachexia liquid’ and 66% fresh DMEM with 2% horse serum; this CM was replaced every 24 h. The ‘cachexia liquid’ was acquired as follows; when CT26 cells reached 90% confluence, their culture medium was replaced with 2% horse serum (differentiation medium) and the supernatant was collected as ‘cachexia liquid’ after 48 h. For the investigation of myoblast differentiation and myotube atrophy, matrine (Shanghai EFE Biological Technology Co., Ltd., Shanghai, China) was added to the culture medium at 0.1 and 0.2 mM for 48 h at 37°C. For the signalling pathway investigation, 0.1 mM matrine was added for 48 h at 37°C and 10 nM wortmannin (MedChemExpress, Monmouth Junction, NJ, USA) was added to culture medium for 48 h at 37°C.

The CCK-8 (Dojindo Molecular Technologies, Inc., Kumamoto, Japan) assay was performed according to the manufacturer's protocol. Briefly, CT26 or C2C12 cells were seeded in 96-well plates at 3,000-5,000 cells/well. Matrine (97%) was added at different concentrations (0.1, 0.3, 0.89, 2.68, 8.05 and 24.16 mM for CT26 cells, and 0.001, 0.004, 0.012, 0.037, 0.111, 0.333 and 1 mM for C2C12) for 48 h. Subsequently, 10 µg/ml CCK-8 regent was added and incubated at 37°C for 1 h. The absorbance was measured at 490 nm with a Synergy™ HT Multi-Mode Microplate Reader (BioTek Instruments, Inc., Winooski, VT, USA).

After being treated with or without matrine (100 µM) and Dex (100 µM) for 48 h at 37°C, the myotubes were washed with cold PBS three times and were fixed with 4% paraformaldehyde at room temperature. The cells were permeabilized by treatment with 0.5% Triton X-100 for 20 min. The myotubes were washed with PBS and blocked in 5% bovine serum albumin (Beijing Solarbio Science and Technology Co., Ltd., Beijing, China) for 1 h at room temperature (25±2°C). The myotubes were incubated with primary antibodies against myosin heavy chain (MHC; cat. no. sc-376157; 1:200), MyoD (cat. no. sc-71629; 1:150), MuRF1 (cat. no. sc-398608; 1:150) and MAFbx (cat. no. sc-166806; 1:150) overnight at 4°C, which were all purchased from Santa Cruz Biotechnology, Inc. (Dallas, TX, USA). The myotubes were then incubated with Alexa Fluor 488-conjugated (cat. no. 4408S; 1:1,000) and Alexa Fluor 594-conjugated secondary antibodies (cat. no. 8889S; 1:1,000) at 4°C overnight, which were both from Cell Signalling Technology, Inc. (Danvers, MA, USA). DAPI (5 µg/ml; Beijing Solarbio Science and Technology Co., Ltd.) was used to stain the nuclei for 5 min at room temperature. Images of the C2C12 myotubes were obtained under a fluorescence microscope equipped with a digital camera (Olympus Corporation, Tokyo, Japan; magnification, ×100 and ×200). A minimum of six representative images were selected and 150 sets of myotube data were measured per group by CellSens standard software (version 1.16; Olympus Corporation).

All procedures involving animals and their care in this study were approved by the Animal Care Committee of Shanghai Sixth People's Hospital (Shanghai, China) in accordance with institutional requirements and Chinese government guidelines for animal experiments. Six-week-old male mice (weight, 20±1.5 g) were housed in a controlled environment (25±2°C; 12 h light/12 h dark cycle; relative humidity of ~50–60%) and provided standard laboratory chow and water ad libitum. The mouse cancer cachexia model bearing CT26 tumor was established as previously described (34,35). Briefly, 40 mice were divided randomly into four groups with ten mice per group. The normal control (NC) group included mice without CT26 inoculation that received normal saline treatment. The normal mice treated with matrine (NC + Mat) group consisted of mice without CT26 inoculation that received matrine-treatment. The cancer cachexia (CC) group included mice with CT26 inoculation that received normal saline treatment. The cancer cachexia mice treated with matrine (CC + Mat) group consisted mice with CT26 inoculation that received matrine-treatment. The CT26 tumor suspension (~2×10⁶ cells) was injected subcutaneously into the right flanks of the mice. Ten days after CT26 inoculation, 50 mg/kg matrine was given via intraperitoneal injection. The food intake, body weights, and longest and shortest tumor diameters were recorded every 2 days. The tumor weights were calculated using the following formula: Weight (mg)=0.52 × (L × S²), where L is the longest diameter and S is the shortest diameter. The longest diameter of the CT26 tumors did not exceed 1.5 cm during the experiment. A total of 21 days after tumor inoculation, the mice were euthanized. The heart, liver, spleen, lungs, kidneys, epididymal fat, gastrocnemius muscle and tibialis anterior muscle were dissected, weighed and snap frozen in liquid nitrogen.

A total of three fresh gastrocnemius muscles were dissected randomly from each group, fixed with GD fixative (v/v=1:19) (Servicebio, Wuhan, China; cat. no. G1111) for 12 h at room temperature, embedded with paraffin, sectioned into 4 µm-thick slices and stained with hematoxylin for 8 min (Servicebio; cat. no. G1004) and eosin-phloxine (Servicebio; cat. no. G1001) for 1 min at room temperature. The muscle sections were observed under a light microscope (Olympus Corporation; magnification, ×200). Myofiber CSAs were calculated using CellSens software (version 1.16; Olympus Corporation), and a minimum of six images and 180 sets of data were acquired per group.

Total RNA from mice gastrocnemius, tibialis anterior and C2C12 myotubes were extracted using RNAiso reagent (Takara Bio Inc., Otsu, Japan). cDNA was synthesized using HiScript II Q RT SuperMix for qPCR (cat. no. R223-01; Vazyme Biotech Co. Ltd., Nanjing, Jiangsu, China) with the following procedure: 37°C for 15 min and 85°C for 5 sec. qPCR was performed with ChamQ™ Universal SYBR qRT-PCR Master mix (cat. no. Q711-02; Vazyme Biotech Co. Ltd.). qPCR and data analysis were conducted on a StepOnePlus™ Real-Time PCR system (Thermo Fisher Scientific, Inc.), according to the manufacturer's protocol. The qPCR amplifications consisted of 40 cycles of 95°C for 30 sec, 95°C for 15 sec and 62°C for 30 sec. The 2^−ΔΔCq method (36) was used for quantification. All data were from three independent experiments. GAPDH was used as the internal control gene. The primers were designed, synthesized and ultra-purified by Sangon Biotech Co. Ltd., Shanghai, China (Table I).

Myotubes and gastrocnemius muscle tissue were lysed using RIPA buffer (Thermo Fisher Scientific, Inc.). Total protein was quantified with a BCA kit (Beyotime Institute of Biotechnology, Haimen China). Subsequently, 50 µg denatured protein per lane was subjected to SDS-PAGE on 10% polyacrylamide gels and then transferred to polyvinylidene difluoride membranes. Following blocking with 5% skim milk in TBS containing 0.05% Tween-20 for 1 h at room temperature, the membranes were incubated with primary antibodies against MuRF1 (cat. no. sc-398608; 1:500), MAFbx (cat. no. sc-166806; 1:500), phosphorylated (p)-Akt (cat. no. sc-81433; 1:500), Akt (cat. no. sc-81434; 1:500), p-FoxO3α (cat. no. sc-101683; 1:500), FoxO3α (cat. no. sc-48348; 1:500), p-mTOR (cat. no. sc-293133; 1:250), mTOR (cat. no. sc-517464; 1:250; all from Santa Cruz Biotechnology, Inc.) and GAPDH (cat. no. 5174S; 1:1,000; Cell Signaling Technology, Inc.) at 4°C overnight. After washing with PBS, the corresponding fluorescently labelled secondary antibodies (catalog nos. 926-68020 and 926-68021, 1:3,000; LI-COR Biosciences, Lincoln, NE, USA) were incubated with the membranes at room temperature for 1.5 h. The membranes were imaged on an Odyssey^® CLx Infrared Imaging system (LI-COR Biosciences). Image Studio 5.0 software (LI-COR Biosciences) was used for measuring integrated optical densities.

Statistical analyses were performed using SPSS version 19.0 (IBM Corp., Armonk, NY, USA). Significance was determined using one-way ANOVA. For comparisons among multiple groups, post-hoc pairwise comparisons were performed using Tukey's multiple-comparison tests. All values are expressed as the mean ± standard deviation (n≥3). P<0.05 was considered to indicate a statistically significant difference.

To evaluate the anti-cachexia effect of matrine (Fig. 1A), a classic CT26 colon adenocarcinoma-bearing murine model was established. The tumors were palpable by the seventh day post-inoculation. Mice bearing CT26 tumors presented with early cancer cachexia symptoms, including decreased food intake from day 12 (Fig. 1B). Total body weights were not significantly different until day 18 in the CT26-bearing mice compared with the negative control (data not shown). However, the free-tumor body weights (total body weight minus tumor weight) were significantly different from day 15 (P<0.05; Fig. 1C). Based on a previous study (28), the mice were intraperitoneally injected with matrine at 50 mg/kg. Matrine-treatment demonstrated a positive effect on the body weights but no influence on food intake.

The organ and tissue weights also suggested less wasting in the cancer cachexia matrine-treatment group compared with in the cancer cachexia group (Table II). The mass loss in the gastrocnemius and tibialis anterior muscles was also reversed with matrine-treatment. In addition, matrine significantly decreased the serum concentrations of inflammatory cytokines, including TNFα (Fig. 1D), IL-6 (Fig. 1E), and IL-1β (Fig. 1F), which are known cachectins (10).

CT26 tumor growth was not significantly inhibited following matrine-treatment in mice (Fig. 1G). To determine the anticancer activity of matrine in CT26 cells, a CCK-8 assay was performed to determine the viability of CT26 cells with or without matrine-treatment. The data demonstrated that when the cells were treated with ≥8.05 mM matrine for 48 h, the cell viability significantly decreased in a concentration-dependent manner (Fig. 1H). The calculated IC₅₀ of matrine was 20.51 mM in CT26 cells (Fig. 1I). All of these results indicate a limited anticancer effect of matrine. Thus, the anti-cachexia effect of matrine appears to occur predominantly through the alleviation of body weight and muscle wasting, rather than anticancer effects.

To determine whether the beneficial effect of matrine on muscle wasting resulted from alleviation of skeletal muscle atrophy, histological analysis of gastrocnemius muscle was performed. Compared with that of the NC group, the CC group exhibited myofiber atrophy, presenting as random loose fiber arrangements and large inter-fiber gaps (Fig. 2A). The gastrocnemius myofibers of the cachexia mice also demonstrated a sharp decrease in CSA, with most myofibers having a value of 200–300 µm². Following matrine-treatment, the shape and arrangement of the myofibers normalized, and there was a marked CSA increase, with most myofibers exhibiting a value of 400–500 µm² (Fig. 2B). Compared with that in the CC group, there was a 70% increase in CSA in the CC + Mat group (Fig. 2C), which supported the muscle weight increases. No negative influence of matrine on the gastrocnemius muscle of normal mice was observed.

Subsequently, the mRNA levels of two E3 ubiquitin ligases, MAFbx and MuRF1, in both gastrocnemius and tibia anterior muscle were quantified by RT-qPCR. The mRNA expression levels of both genes were significantly elevated more than ten-fold in the cachexia mice compared with the NC group. Notably, matrine-treatment significantly downregulated the mRNA levels of both E3 ubiquitin ligases (Fig. 2D). Consistent with the mRNA expression data, western blots demonstrated that the protein expression levels of MuRF1 and MAFbx in gastrocnemius muscle were significantly inhibited by matrine-treatment compared with the CC group (Fig. 2E). Although there were differences between mice, matrine significantly decreased the levels of the two E3 ubiquitin ligases in the cachexia skeletal muscle.

In order to determine the concentrations of matrine that are non-toxic to C2C12 myocytes and myotubes, a CCK-8 assay was performed. The data demonstrated that the IC₁₀ of matrine toward C2C12 myocytes is 0.203 mM (Fig. 3A and B), indicating no significant toxicity at <0.203 mM. In addition, RT-qPCR analysis indicated that 0.2 mM matrine exhibited no significant overall influence on the mRNA expression of MHC, a major structural protein of myotubes. However, matrine treatment at 0.1 mM significantly upregulated the expression of MHC Ib (Fig. 3C). For MHC IIa (Fig. 3D) and IIb (Fig. 3E), there were no significant changes with matrine-treatment at any concentration. In addition, matrine (0.4 mM) significantly inhibited the expression of MHC Ib and MHC IIx (Fig. 3C and F).

Double immunofluorescence staining of MHC and MyoD was used to investigate the effect of matrine on normally differentiated or Dex-stimulated C2C12 myoblasts. As presented in Fig. 4A, the myotubes, but not the undifferentiated myoblasts, were stained for MHC and MyoD. Furthermore, matrine increased myotube diameters at 0.1–0.4 mM (Fig. 4B). Notably, the myoblasts fused into giant and atypical myotubes when matrine was used at 0.4 and 0.8 mM, with higher matrine concentrations demonstrating increased fusion (Fig. 4C). However, compared with that in the other matrine-treatment groups, MyoD expression was lower in the 0.8 mM matrine-treatment group.

It was identified that Dex can significantly inhibit differentiation of C2C12 myoblasts, since there was less myotube formation in the Dex-treated groups. Furthermore, the formed myotubes in the Dex-treated groups were shorter and exhibited fewer nuclei (Fig. 5A). With matrine-treatment at 0.1–0.4 mM, the myotube diameters and fusion indexes increased (Fig. 5B and C). These results indicate that matrine can increase the differentiation of C2C12 myoblasts, even in the presence of Dex, which impairs their differentiation. However, there was no significant difference in fusion index between the Dex + Mat 0.8 mM and Dex groups. In addition, compared with that of the other matrine-treatment groups, the Dex + Mat 0.8 mM group demonstrated lower MyoD expression. This suggests matrine can increase myoblast differentiation when used up to 0.4 mM.

Despite having confirmed that matrine promotes C2C12 myoblast differentiation, whether it can stabilize myotubes in the presence of impairment factors remains unknown. After 3–5 days of differentiation and myotube formation, 100 µM Dex, 50 ng/ml TNFα or CM were added for 48 h to induce myotube apoptosis and atrophy. Dex-treatment resulted in the apoptosis and atrophy of formed myotubes, since discontinuous MHC staining was visible. TNFα- and CM- treatment also resulted in extensive myotube damage. Matrine demonstrated protective effects at both 0.1 mM (Mat-L) and 0.2 mM (Mat-H; Fig. 6A). The diameters of most normal myotubes ranged between 9–15 µm, while they ranged between 5–11 µm after being treated with Dex, TNFα or CM (Fig. 6B and C). With matrine co-treatment, the myotube diameters significantly improved, with values of 7–15 µm. There was no effect of matrine on the morphology of normal myotubes, indicating the absence of toxicity on formed myotubes.

Western blotting and immunofluorescence for MAFbx and MuRF1 (Fig. 7) also supported the ability of matrine to protect myotubes under Dex-induced myotube conditions. Dex-treatment significantly upregulated MAFbx and MuRF1 (Fig. 7A). Matrine (0.1 mM) decreased expression of MAFbx and MuRF1 by 27 and 43%, respectively. These results are consistent with the in vivo experimental findings where matrine downregulated the expression of MAFbx and MuRF1 in cachexia mice skeletal muscle. In addition, matrine exhibited no significant influence on the expression of these two E3 ubiquitin ligases in normal myotubes. Furthermore, it was identified that matrine significantly decreased the mRNA level of myostatin (Fig. S1), an upstream regulator of E3 ubiquitin ligases, which was significantly upregulated by following treatment with Dex.

The western blotting data was then corroborated using immunofluorescence. The images revealed that MAFbx and MuRF1 could not be detected in normal C2C12 myotubes. However, particularly with the Dex-treatment group, both E3 ubiquitin ligases demonstrated higher fluorescence. Co-treatment with matrine substantially decreased the expression of both E3 ubiquitin ligases, with little observable difference between the two concentrations of matrine.

When myotubes were treated with 100 µM Dex for 48 h, accompanied by the upregulated expression of the two E3 ubiquitin ligases, a significant decrease was also identified in the relative phosphorylation of Akt, mTOR and FoxO3α. Treatment with matrine at 0.1 mM for 48 h reversed this effect on Akt, FoxO3α, and mTOR phosphorylation (P<0.05; Fig. 8). However, compared with the treatment with Dex and matrine, the additional treatment with 10 nM wortmannin (a PI3K inhibitor) for 48 h diminished the effect of matrine by decreasing the phosphorylation of Akt, mTOR and FoxO3α, accompanied by the upregulation of MAFbx and MuRF1. This indicates that the anti-atrophic effect of matrine was attenuated by inhibition of the Akt/mTOR/FoxO3α signaling pathway. Therefore, these results suggest that matrine-treatment upregulates anabolism and downregulates catabolism of muscle-specific proteins.