Advantage of ostarine over raloxifene and their combined treatments for muscle of estrogen-deficient rats

The animal study protocol was approved by the local regional government (14/1396, Oldenburg, Germany) prior to the study. Seventy-five three-month-old Sprague Dawley rats (Fa. Janvier Labs, Saint-Berthevin, France) were used in the experiment. The experiment was conducted as depicted in Fig. 1. All rats were anesthetized with isoflurane, microchips (1,25 × 7 mm, ISO11784/11785, Med Associates, Inc. Fairfax, Virginia, USA) were injected s.c. for further identification of the rats, and the rats were either bilaterally ovariectomized (OVX) or left non-ovariectomized to serve as intact controls (Non-OVX). Thereafter, rats were divided into five groups, each of 15 rats: Group 1, Non-OVX; Group 2, OVX, and Groups 3 to 5, OVX rats treated with OST, RAL, or a combination of both (RAL + OST), with a dosage of 0.4 mg/kg BW for OST and 7 mg/kg BW for RAL for up to 13 weeks. The dosages were taken from the previous studies [11, 18]. Three to four rats were housed in one cage (Type Makrolon® IV, Techniplast Deutschland GmbH, Hohenpreißenberg, Germany). The rats had free access to demineralized water and soy-free pelleted food (ssniff Spezial Diät GmbH, Soest, Germany). In the latter, OST and RAL were supplied. OST was obtained from Shanghai Biochempartner Co., Ltd. (Shanghai, China), and RAL was obtained from Eli Lilly and Company (Evista®, Indianapolis, USA). Food intake and BW were weekly recorded. The average daily food intake of a rat was calculated by dividing the food consumed by the number of rats in a cage and it served for the calculation of the drug intake (OST, RAL, OST + RAL) [19]. The average received dose was 0.6 ± 0.1 mg/kg/day BW for OST and 11.1 ± 1.2 mg/kg/day for RAL.

At the end of the experiment, 13 weeks after OVX, an open field activity test was performed. Four animals at a time were placed in a cage with the bottom divided into 6 squares using a black marker and filmed with a camera (Nikon D5600, Tokyo, Japan) for 5 min. The number of completely crossed lines (horizontal movement activity, transitions), the number of uprights (vertical movement activity, rearing), and the cleaning activity (grooming) were then counted on a computer [20]. Blood samples were then collected by heart puncture under deep isoflurane anesthesia and stored at − 20 °C for further analysis. The uterus and three muscles were removed bilaterally: gastrocnemius (GM), longissimus (LM), and soleus (SM). GM and SM were weighed. All muscles were snap frozen in liquid nitrogen and stored at − 80 °C for either histological or mRNA expression analysis.

A cryotome was used to cut cross-sections of 12-µm thickness serially from the middle part of each frozen muscle (CM 1900; Leica Microsystems, Wetzler, Germany). Until staining, specimens were air dried and stored at − 20 °C. Unless otherwise indicated, all chemicals were obtained from Merck KGaA (Darmstadt, Germany).

Staining of muscle capillaries was performed using a periodic acid–Schiff (PAS) method [21]. Briefly, section was fixed in ethanol/chloroform/glacial acid solution (16:3:1), then incubated in 0.3% α-amylase from porcine pancreas, (Sigma–Aldrich Laborchemikalien GmbH, Seelze, Germany), stained using Schiff’s reagent solution (Roth, Karlsruhe, Germany) and finally treated with a 10% potassium sulfite solution. To avoid overstaining, Schiff’s reagent solution was applied under visual control (2–25 min).

For staining of muscle fibers, a modified staining method with adenosine-triphophatase (ATPase) was applied as described by Horák [22]: Sections were fixed in a solution of 1% paraformaldehyde solution (pH 6.6), 1% CaCl₂ and 6% sucrose and then stained by an incubation in a reduced nicotinamide adenine dinucleotide diaphorase solution (pH 7.4). At last, an acidic incubation (pH 4.2) and incubation in adenosine-5´-triphosphate solution (pH 9.4) were performed [22].

Using hematoxylin eosin (HE) staining, nuclei were analyzed [23]. After muscle sections have been fixed with acetone, Mayer’s hematoxylin solution was used to stain the nuclei, and eosin G solution to stain the muscle fibers.

Aquatex® (Merck) was used to mount sections for fiber type and capillary stainings, whereas Eukitt® (Kindler GmbH, Freiburg, Germany) was taken for sections with nucleus staining. Muscle sections were analzed using a microscope (Eclipse E 600 microscope; Nikon, Tokyo, Japan), a digital camera (DS-Fi2 Digital Camera; Nikon Instruments Europe, Amsterdam, Netherlands) and a software (NIS-Elements AR 4.0 imaging software; Nikon Instruments Europe) at tenfold magnification. For the evaluation of muscle fibers, we used three randomly chosen fields of 1 mm² within each ATPase-stained section. In these fields, 90 slow-twitch oxidative and fast-twitch oxidative (fiber types I and IIa, respectively) STO + FTO and 90 fast-twitch glycolytic (FTG; fiber type IIb) fibers were skirted [24]. In the ML, the fiber distribution was determined, as in this muscle, fibers show a homogenous distribution pattern [25]. In contrast, the GM showed a relative heterogenic distribution of fiber types, whereas the SM mainly consists mostly of STO fibers [26, 27]. Hence, in the latter only STO fibers were measured. The percentage of STO + FTO and FTG fibers was determined within 1 mm² field. The ratio of capillaries to fibers (capillary density) as well as the ratio of nuclei to fibers (nucleus density) were calculated in two randomly chosen fields of 0.5 mm² each withih the cross section [28].

For analyses of enzyme activities and electrolyte concentrations an automated chemistry analyzer Architect c16000 (Abbott, Wiesbaden, Germany) and commercially available kits (Abbott) were used at the Department of Clinical Chemistry, University of Goettingen according to the manufacturer’s instructions (Abbott). Activity of creatine kinase (CK), and concentration of calcium (Ca), magnesium (Mg), and phosphorus (P) in serum were determined. The following methods were applied: Ca and Mg: quantification by arsenazi III dye was used (7D61-20 and 7D70-30, Abbott); P: ammonium molybdate method (7D71-30, Abbott); CK: reactivator method with N-acetyl-L-cysteine (7D63-30, Abbott) [11]. Enzyme Immunoassay kit for rats (Cloud-Clone Corp., Katy, Texas, USA) was used to determine follicle stimulating hormone (FSH) and luteinizing hormone (LH) levels.

GM samples (100 mg; n = 5/group) were homogenized in 750 µl TRIzol (Thermo Fischer Scientific, WA, USA) using 4 mm tungsten carbide beads (Cat. No. 69997 Qiagen, Germany) with the aid of the Tissuelyzer LT system (Qiagen, Germany). Thereafter, the samples were incubated for 5 min at room temperature and further RNA extraction was processed according to the manufacturer’s protocol (Trizol, Thermo Fischer Scientific) using chloroform and isopropanol treatments and ethanol washings. Finally, the RNA pellet was dissolved in 20 µL H₂O, measured by DeNovix DS-11 FX + System (DeNovix, NC, USA) and stored at − 80 °C for further analysis.

Reverse transcription was performed with 1000 ng of total RNA using an iScript cDNA synthesis kit (Biorad, CA, USA). Quantitative real-time Polymerase chain reaction (PCR) was performed on the CFX96 Real-time PCR Detection System (Biorad, CA, USA) using a SYBR Green (Biorad, CA, USA) detection marker. Relative expressions of beta-2-microglobulin (B2M), androgen receptor (Ar), estrogen receptor alpha (Er alpha), Myostatin, insulin-like growth factor 1 (Igf-1), and vascular endothelial growth factor B (Vegf-B, [29]) were measured in triplicate and effects were calculated using the 2^−ΔΔCT method [30]. Ready-to-use primers for B2M, Ar, Er-alpha, Igf-1 and Myostatin were obtained from Qiagen (QuantiTect Primer Assays, Qiagen, Hilden, Germany). Primers for Vegf-B were used with the following sequences: Forward GCCAGACAGGGTTGCCATAC, Reverse GGAGTGGGATGGATGATGTCAG. B2M was taken as a reference gene. We failed to measure the mRNA expression of Er beta, confirming its low and nearly undetectable expression in rodent muscle [31].

Mean food intake was significantly different between the treatment groups (Table 1). The OVX and OST groups had significantly higher food intakes than the Non-OVX and RAL groups. Weekly food intake analysis showed higher food intake in the OVX and OST groups compared to all other groups for up to 8 weeks (Fig. 2). At week 10, higher food intake was particularly observed in the OST group compared to the other groups and in the OST + RAL group compared to the RAL group. Thereafter, it did not differ from that of the OST + RAL group, and at the end of the study, it differed only from that of the RAL group (Fig. 2).

The mean BW of the rats was highest in the OST group and higher in the OVX group than in the Non-OVX, RAL, and OST + RAL groups (Table 1). Similarly, BW was higher in the OVX and OST groups compared to all other groups throughout the experiment (Fig. 2). In addition, the BW of the RAL groups was significantly lower than that of the OST + RAL group at weeks 5, 8, 9, 10, and 11.

The mean OST intake was 0.52 ± 0.03 mg/kg BW for the OST group and 0.58 ± 0.07 mg/kg BW for the OST + RAL group (Fig. 2). The mean RAL intake was 10.62 ± 1.06 mg/kg BW for the OST group and 11.52 ± 1.41 mg/kg BW for the OST + RAL group. RAL intake was significantly higher in the OST + RAL group than in the RAL group during weeks 10 and 11 (Fig. 2).

None of the parameters recorded during the activity test (transitions, rearing and grooming) differed between groups (Table 1).

Uterine weight was significantly lower in the OVX and RAL groups compared with the Non-OVX, OST, and OST + RAL groups (Table 1).

Regarding muscle weight, GM and SM weights were significantly higher in the OST group compared to all other groups. The OVX group had a significantly higher SM weight compared to the RAL group and a higher GM weight compared to the Non-OVX, RAL, and OST + RAL groups (Table 1). Correlation analysis of muscle weight and BW showed a significant relationship between these variables in GM and SM (Table 3).

In the GM, the diameter of STO + FTO fibers of the RAL group was significantly larger than that of the Non-OVX group. The diameter of FTG fibers was significantly larger in the OVX group and the OST group than in the OST + RAL group, and the OST group had a significantly larger diameter than the RAL group. The areas of all fibers did not differ between the treatment groups (Table 2). The size of FTG fibers correlated significantly with BW in the GM (Table 3).

In the LM, the areas and diameters of STO + FTO fibers were significantly larger in the OVX group than in the RAL and OST + RAL groups, whereas those of FTG fibers did not differ between groups (Table 2). All fiber types examined correlated positively with the BW of the rats (Table 3). The ratio of STO + FTO fibers to FTG did not differ significantly between the groups in the LM (Table 2).

In the SM, the area of STO fibers was significantly larger in the OVX group than in the RAL group (Table 2). There was no correlation between fiber size and BW of the rats (Table 3).

Regarding the capillary ratio, the Non-OVX group had a significantly lower ratio than all other groups in the GM (Fig. 3). In the LM, the OVX group and the OST group had a significantly higher ratio compared to the OST + RAL group. In addition, the OST group had a significantly higher ratio compared to the Non-OVX and RAL groups (Fig. 3). In the SM, a significantly higher ratio was observed in the OST group compared to the Non-OVX, OVX, and OST + RAL groups (Fig. 3). In the GM and LM, the capillary ratio was significantly correlated with BW (Table 3).

The analysis of the nucleus ratio (nuclei per muscle fiber) was significantly higher in the OST + RAL group than in the Non-OVX group in the GM. There were no significant differenes in the LM or SM (Table 1).

Correlations between nucleus ratio and BW were not significant for GM and SM, but reached a significant level for LM (Table 3).

Ca and Mg levels were significantly higher in the Non-OVX group compared to all other groups. In addition, the OVX group had a significantly higher Ca level than the OST + RAL group (Table 1). The P level of the OST + RAL group was the highest among the other treatment groups. It was significantly lower in the Non-OVX group than in the OST and RAL groups, and lower in the OVX group than in the OST group (Table 1). There were no significant differences in CK activity (Table 1). In hormonal analysis, significantly higher FSH levels were observed in the RAL and OST + RAL groups compared to the Non-OVX group. LH levels were also significantly higher in the OST + RAL group than in the Non-OVX group (Table 1).

Ar gene expression was significantly higher in the RAL group compared to the Non-OVX, OVX and OST + RAL groups (Fig. 4). No significant differences were found in the Er alpha expression. Vegf-B gene expression was significantly higher in the Non-OVX, and OVX groups than in the OST + RAL group. Igf-1 expression was higher in the OST, RAL, and OST + RAL groups than in the Non-OVX group. Igf-1 gene was expressed expression was significantly higher in the OST and RAL groups than in the OVX group. Myostatin expression was significantly higher in the RAL group than in the Non-OVX, OVX, and OST groups (Fig. 4).