Changes in Musculoskeletal System and Metabolism in Osteoporotic Rats Treated With Urocortin

Objective: In aging population, postmenopausal osteoporosis and decline of musculoskeletal function, referred to as “frailty syndrome” lead to loss of bone and muscle, causing falls, and fall-related injuries. To limit the impact of this portentous duo, simultaneous treatment of both is needed. Urocortin (UCN) has been reported to improve osteoporotic bone properties while its effect on muscle has not been addressed yet. Design and Methods: We aimed to investigate the effect of urocortin in vivo on skeletal muscle structure in osteopenic rats. Sixty Sprague-Dawley rats were divided into five groups: four were ovariectomized (OVX) and one underwent sham operation (SHAM). One ovariectomized group was left untreated (OVX), while one was treated with urocortin s.c. in 3 μg/kg body weight (bw) (OVX+UCN low), one with 30 μg/kg (OVX+UCN high), while one group was treated with estradiol orally (OVX+E: 0.2 mg/kg bw), each for 35 days. Mm. gastrocnemius, longissimus, and soleus were isolated and capillary density as well as diameters of type I and II fibers were measured. In addition, we examined the effect of UCN on tibia using biomechanical, micro-CT and ashing analysis and investigated the blood serum. Results: We demonstrated a positive effect of UCN on M. soleus, in which fiber diameter was positively influenced. The biomechanical and structural parameters of bone were not changed in UCN treated rats. The higher cholesterol, glucose and triglyceride levels in the “UCN high” group raise concern about this treatment. Conclusions: Our results portray urocortin as a substance that can be assessed for future therapeutic treatments of estrogen deficiency. New and Noteworthy: Urocortin has a positive effect on M. soleus (diameter). Urocortin raises serum cholesterol and triglyceride levels. Bone tissue was not affected by UCN.


INTRODUCTION
Due to the high prevalence of postmenopausal osteoporosis (1,2), a strong muscular function is essential for the protection of the incorporated bone and physical coordination to prevent falls. In the postmenopausal women, estrogen deficit does not only cause osteoporosis, but was also related to muscular dysfunction (3,4). As "muscle bone interactions" are focused on more and more and the incidence of osteoporosis and sarcopenia is closely linked, we aimed to assess whether in a rat model for postmenopausal osteoporosis skeletal muscle could be addressed (5)(6)(7). In recent decades and with a distinct diagnostic definition, multiple therapeutic options and breakthroughs have been developed for osteoporosis (8,9), whereas muscle-related research has not proceeded as far (10,11). Post-menopausal estrogen replacement therapy (ERT) was often recommended for prevention and treatment of osteoporosis (12) as well as for maintaining muscle strength and mass (13,14). However, ERT is associated with potential side effects on reproductive tissues (12). Current potential therapeutic options against muscle loss include human recombinant bone morphogenetic protein (BMP)-2/7 and antimyostatin antibodies (15)(16)(17), though certain of the latter (activin decoy receptor IIB: ActRIIB-Fc) seem to elevate bone mass (18)(19)(20). The effect of vitamin D supplementation remains inconclusive, although beneficial effects on muscle strength and prevention of falls in sarcopenic post-menopausal women have been reported (21). Recently, creatine supplementation has been proposed to positively affect sarcopenic state in aging postmenopausal women, while effect on bone remains inconclusive (22).
Hence, innovative therapeutic strategies against muscle loss need to address both, bone and muscular systems.
Urocortin I (UCN), a peptide consisting of 40 amino acids (23), and its fellow members of the corticotropin-releasing factor peptide family positively affect the cardiovascular system in vitro (24)(25)(26)(27)(28) and in vivo (29,30) and modulate stress responses as well as ingestive behavior (31). Urocortin I exerts its effects via the corticotropin-releasing factor 1 and 2 receptor (CRF1R and CRF2R) and has a 40-fold higher affinity than corticotropinreleasing factor (CRF) itself at CRF2Rs (32,33). Since UCN has received more and more attention as a cardio-protective agent (34,35) and seems to be effective at cutting excessive food consumption (36)(37)(38), further effects of these promising peptides need to be elucidated.
To mimic the clinical situation, in which muscle changes often accompany bone loss (43)(44)(45), we chose the ovariectomized rat as an established model of postmenopausal osteoporosis (46), in which UCN could simultaneously affect impaired bone tissue along with the muscle (47). In the present study, the tibia was analyzed to confirm previously reported positive effects (39,40) in bone. M. gastrocnemius (both slow and fast fiber types) and M. soleus (mainly slow fiber types) as lower leg agonists were chosen as surrogates for different fiber types, while M. longissimus (with mainly fast fiber types) represents a holding muscle. We sought to investigate the changes in muscle fibers and capillarization as parameters of nutrient supply (48,49) of several skeletal muscles in osteoporotic rats after subcutaneous administration of urocortin in different doses. Since fiber diameter alterations alone do not sufficiently represent muscle structure and function, quotients between capillaries and fibers were calculated and metabolic muscle enzymes were measured (50). Moreover, the metabolic status of the rats was evaluated.

Animals and Treatment
In this study, 60 female Sprague-Dawley rats (Winkelmann Company, Borchen, Germany) were kept at 22 • C with 55% relative humidity in Makrolon IV R -Cages. One acclimatization week was included prior to the beginning of the experiment, which took place in conformance with the ethical standards of animal care. The experimental design was approved by the local institutional animal care and use committee (district authorities of Oldenburg, Germany, registration numbers: 33.9-42502-04-10/0246).
At an age of 13 weeks, the rats underwent bilateral ovariectomy (OVX) or sham surgery (SHAM) as described elsewhere (51). Operations were carried out under ketamine (90 mg/kg body weight (bw), Hostaket R , Hoechst, Bad Soden, Germany) and xylazine anesthesia (7.5 mg/kg bw, Rompun R , Bayer, Leverkusen, Germany) applied 0.1 ml/100 g bw intraperitoneally. After shaving and disinfection, skin was incised below the ribs on both sides, and the abdomen was opened. Adnexa were prepped and the tubae uterinae tied up. Ovaries were removed with a scalpel before wound closure.
The OVX+Estradiol group was supplied via food (10 mg/kg food) with 0.2 mg/kg bw estradiol-17-benzoate (Sigma-Aldrich Chemie GmbH, Taufkirchen, Germany) per day for 35 days. All rats received a soy-free diet (Ssniff special diet GmbH, Soest, Germany) during the experiment. Food intake and body weight were recorded once weekly and are already published (39,40). In the end, rats were CO 2 -anesthetized and decapitated.

Tissue Isolation and Processing
Rats were sacrificed after 35-day treatments. Uterus, heart, liver, kidneys, spleen and visceral adipose tissue were extracted and weighed. Mm. gastrocnemius and soleus were weighed and transversely cut in half across the whole muscle. A piece of M. longissimus (1 × 3 cm) was extracted. After that, the muscles were snap-frozen in liquid nitrogen and stored at −80 • C. Afterwards, 12 µm frozen sections were cut orthogonally with a cryotome (Leica 2800E Frigocut Microtome Cryostat, Leica Biosystems, Nussloch, Germany).

Staining
To distinguish type I from type II muscle fibers, adenosine triphosphatase (ATPase) staining was carried out as described elsewhere (54). Three different image areas were chosen within the histological section of each muscle. Therein, 30 type I and 30 type IIb fibers were outlined, and the mean diameter was measured with the help of Lucia G image analysis system (Version 4.82, Laboratory Imaging, Prague, Czech Republic). Because slow oxidative fibers (type I) were rarely represented and cannot always be clearly distinguished from fast-twitch oxidative glycolytic fibers (IIa), they were analyzed combined in M. gastrocnemius and M-longissimus (type I) ( Figure 1A) (52). In M. soleus, most of the fiber types are type I (55). Therefore, no further differentiation was conducted, and solely type I fibers were analyzed.
For the relationship of fiber to capillaries, periodic acid-Schiff (PAS) staining was performed, following the method of Andersen (56,57). Two areas of 500 × 500 µm were chosen within the muscle section, wherein fibers and capillaries were counted and their ratio calculated ( Figure 1B).
The selection of image areas as well as their analyses have been performed by one individual who was blinded with respect to the treatment groups.

Analysis of Muscle Enzymes
Muscle specimens were first assimilated in Chappell-Perry medium (0.1 M KCl, 0.05 M Tris, 0.01 M MgCl 2 × 6H 2 O, 1 mM EGTA, pH 7.5). Afterwards, they were assessed with a spectrophotometer (Spectronic Genesys 2PC, Pittsford, USA). The assay was performed within 2 h following the tissue homogenization. The lactate dehydrogenase (LDH) activity was indirectly measured by recording the reduction in absorption at 340 nm, which demonstrates the oxidation of NADH to NAD + (58).
Citrate synthase (CS) activity was measured at 412 nm with Triton X-100 (0.1%) for the liberation of all CS proteins according to Faloona and Srere (59), whereas the complex I activity was assessed at 340 nm before the complex I inhibitor rotenone was added to the incubation solution for measuring the NADH reduction, as reported by Hatefi and Stiggall (60). Protein content was measured with a multilabel reader (Perkin Elmer Precisely Victor X4, Turku, Finland) and software version 4.0 (Perkin Elmer Life and Analytical Science). The activity of enzymes was calculated relative to the protein content. Detailed protocols have been previously described by our group (61)(62)(63).

Biomechanical Analysis
Tibia biomechanics was assessed using a Zwick device (145 660 Z020/TND; Ulm, Germany) as previously described (64). Four screws inside the metallic cylinders helped to fix the tibia on the metal plate as explained in Komrakova et al. (65). The bending test was stopped before plastic deformation ended; measurements were performed with a 5 mm/min motion rate and stopped at a curve-decline more than 2 N. Recording was conducted with testXpert software (Zwick GmbH & Co. KG, Ulm, Germany). The highest force that the tibia could resist was the maximum load (Fmax). The turnaround point from elastic to plastic deformation was referred to as the yield load (yL), while elasticity of the tibia (stiffness (S)) was measured as the slope of the linear rise of the curve during elastic deformation. All these values were quantified using Microsoft Excel (MS Office 2016) (66,67).

Micro-CT
The metaphyseal tibia was scanned with a micro-CT device (eXplore Locus SPScanner, GE Healthcare, Ontario, Canada) with 72 kVp, 90 µA, and 1,600 ms exposure time and 360 • rotation, resulting in 0.0029 mm pixel size. Five hydroxyapatite blocks with defined mineral densities were scanned to transform gray scale values into bone mineral density (mg/cm 3 ). 3D OsteoAnalyze, which was developed for our laboratory, was used to determine bone parameters according to American Society for Bone and Mineral Research (ASBMR) criteria (40,(68)(69)(70). Total, trabecular and cortical BMD (g/cm²) as well as bone volume fraction (BV/TV) were assessed at the metaphysis. Therefore, two measurements next to the growth plate were performed (3 and 5 mm), and the area in between was cut out. The volume and density of cortical, trabecular and total bone were measured in this field. After turning 90 • , the crosssection was used to measure the distal cortical bone area (Ct.Ar), endosteal area (E. Ar), and total area (T. Ar.) of the bone. Thereafter, the cortical bone was cut off, and trabecular structure was analyzed. Trabecular thickness (Tb.Th), number of trabecular nodes (N.Nd), trabecular number (Tb.N), mean trabecular junctions at one node (Tb.N/Nd), and trabecular separation (Tb.Sp) were measured (69,71).

Ashing
The tibia was ashed at 750 • C for 1 h. Afterwards, calcium content was quantified with an atomic absorption spectrometer (4100; PerkinElmer, Baesweiler, Germany). Orthophosphate content was assessed by the colorimetric method (Spectral Photometer DM4; Zeiss, Jena, Germany) (67,72). The inorganic weight was measured after ashing as a percentage of the wet weight before ashing (39).

Serum Analyses
Serum analyses were conducted at the Department of Clinical Chemistry, University of Goettingen according to the manufacturers' instructions (Abbott, Wiesbaden, Germany).
After collecting 5 ml of whole blood from the rats in the end of the experiment, it was centrifuged at 3,000 × g for 10 min. Serum was stored at −20 • C. Creatine kinase (CK), Alanine transaminase (ALT), triglycerides, uric acid, Aspartate aminotransferase (AST/GOT), Cholesterol, glucose, and HDL were measured with the ARCHITECT c System and AEROSET System (Abbott, Wiesbaden, Germany).

Statistics
Statistical analysis was performed with GraphPad Prism (5.04, GraphPad Software, Inc. San Diego, CA). One-way ANOVA (F test, α = 0.05) was conducted to reveal the impact of the treatments on variables. For variables that showed an overall statistically significant difference in group means (p < 0.05, F test), the Tukey-Kramer post hoc test (α = 0.05) was performed to estimate the differences between individual means (Supplemental Table 1

Rat Characteristics
In the beginning of the study, body weight of rats did not differ among the groups, whereas at the end of the treatments, the mean weight of the SHAM (324.8 ± 23.5 g) and OVX+Estradiol (324.3 ± 16.5 g) groups was significantly lower compared to the OVX (396.8 ± 18.9 g), OVX+UCN high (388.3 ± 43.7 g) and OVX+UCN low (375.2 ± 16.6 g) groups [p < 0.05; (39,40)]. Body weight in the OVX+UCN high and OVX+UCN low groups did not differ from that in the OVX group. As expected, the uterus was significantly lighter in all ovariectomized animals, i.e., OVX, OVX+UCN high, and OVX+UCN low groups, compared to those in the SHAM and OVX+Estradiol groups, as shown previously by our group [p < 0.05; (39,40)]. Though the weight of the uterus increased after estradiol treatment, it was still lower than in the SHAM group.
Visceral fat mass was most abundant in the OVX and OVX+UCN high groups (Figure 2A). Regarding heart weight, OVX+Estradiol group showed the lowest weight compared to all other groups ( Figure 2B). In OVX+UCN high group, heart weight was higher than that in SHAM and OVX+UCN low groups. OVX+UCN high led to the highest results in the liver weight, which could also be seen in the kidney weight (Figures 2C,D). OVX+UCN high had the lowest spleen weight compared to all other groups ( Figure 2E). In M. soleus, OVX+UCN high caused significantly higher values compared to all other groups ( Figure 2H). OVX and OVX+UCN low showed higher total M. gastrocnemius weight compared to OVX+Estradiol (Figure 2F). Regarding the ratio of M. gastrocnemius weight to body weight, OVX+UCN high caused lowest ratio, while SHAM led to a significantly higher ratio compared to all (except OVX+Estradiol) treatments ( Figure 2G). In weight/bw-ratio of M. soleus, OVX+UCN high caused significantly higher ratio compared to OVX+UCN low and OVX (Figure 2I). Taken together, there was an effect of UCN treatment on inner organs and fat.
Superficial observational analysis of rat activity did not reveal differences regarding daily levels of exercise. Movements among groups were comparable and treatment with UCN did not cause any gain or decrease of action.

Fiber Diameter in Muscles
Next, to evaluate the effect of urocortin treatment on muscle, we decided to measure muscle fiber parameters in Mm. gastrocnemius, longissimus, and soleus. In M. gastrocnemius, fiber diameter was homogeneous between the all groups. With a mean diameter of 53.3-59.0 µm, no differences in type I fiber diameter were detected (Figure 3A). In fiber type IIb, mean diameter ranged from 75.9 to 79.6 µm, also without any significant differences ( Figure 3B).
In M. soleus, the mean diameter of fibers was significantly higher in OVX+UCN high (81.6 µm) compared to SHAM (67.0 µm), OVX (59.9 µm), and OVX+Estradiol (68.2 µm) ( Figure 3E). Effect of UCN on fiber diameter was most evident in M. soleus, whereas in Mm. gastrocnemius and longissimus, the UCN effect was not significant.

Ratio of Capillaries to Muscle Fibers
To explain the differences in muscle characteristics, we investigated the capillary supply. No statistically significant differences were detected between groups in the ratio of capillaries to M. gastrocnemius (Figure 3F) or M. longissimus fibers (Figure 3G). In M. soleus, no differences could be detected too ( Figure 3H).

Muscle Enzyme Activity
For metabolic insights, we measured enzyme activity in the abovementioned muscles. Lactate dehydrogenase (LDH) activity in M. gastrocnemius, M. longissimus, and M. soleus was not different between the groups (Figures 4A-C).
Regarding citrate synthase (CS) activity, no differences could be obtained in M. gastrocnemius ( Figure 4D) as in M. longissimus ( Figure 4E). In M. soleus, the OVX+UCN high and OVX+Estradiol groups showed significantly higher CS activity compared to the SHAM and OVX groups ( Figure 4F).
Complex I (CI) activity showed the highest levels in OVX compared to all other groups in M. gastrocnemius (Figure 4G), whereas in Mm. longissimus and soleus, no differences could be detected (Figures 4H,I).
Enzyme activity in terms of citrate synthase was significantly enhanced by UCN high treatment, whereas UCN therapy did not lead to differences in LDH or CI activity.

Biomechanical Assessment of Tibia and Ashing
After detection of metabolic muscle alterations, we proceeded to assess bone parameters to learn whether the former lead to changes in the subjacent bone. While stiffness was not altered, yield load and maximum load were significantly diminished in the OVX+UCN low group compared to SHAM and OVX+Estradiol. In the OVX group, yield load and maximum load were lower than in the SHAM group ( Table 1).
In the ashing analysis, the calcium content was significantly lower in OVX+UCN low group and OVX+UCN high than in SHAM (Table 1), corresponding to the inorganic mass of the tibia, for which the OVX+UCN low group showed lower values compared to those of SHAM, OVX+UCN high and OVX+Estradiol ( Table 1). The inorganic mass was significantly lower in all OVX groups than in the SHAM group. The phosphate content did not change significantly between the groups ( Table 1).
UCN treatment did not cause biomechanical changes or mineral differences in bone of OVX rats.

Micro-CT
As a next step we investigated the microstructure of tibiae. Notably, micro-CT analysis showed that all OVX groups, compared to SHAM, had reduced total bone mineral density ( Table 1). Lower trabecular BMD appeared in the OVX+UCN low and OVX groups compared to SHAM ( Table 1). Cortical BMD revealed that OVX+UCN low-values were significantly lower compared to SHAM and OVX+Estradiol ( Table 1). Bone volume fraction was reduced in all OVX groups, similar to total BMD ( Table 1). Regarding the analysis of trabecular structure, the trabecular thickness, nodes, number, and trabecular junctions at single node were significantly decreased in all OVX groups ( Table 1). Cortical bone area, endosteal and total bone area were not altered ( Table 1). Micro-CT measurements confirmed osteoporotic phenotype after OVX, while UCN treatment did not alter trabecular or cortical parameters of OVX rats.

Serum Analyses
Finally, to assess the potential systemic impact of urocortin, investigation of serum parameters followed. Neither CK nor AST/GOT was significantly different among the groups (Figures 5A,B), though in the OVX+UCN high group the values were higher than in the other groups. Alanine transaminase was significantly increased in the OVX+UCN high group compared to all other groups (Figure 5C), as was serum cholesterol (Figure 5D). Glucose was significantly higher in both UCNtreated groups compared to SHAM, OVX and OVX+Estradiol (Figures 5E,F). UCN low showed the highest uric acid values, significantly higher than in SHAM, OVX, and OVX+Estradiol ( Figure 5F). HDL-cholesterol was significantly higher in the OVX+UCN high group than all other groups (Figure 5G), and a similar result was seen in triglycerides ( Figure 5H). Serum analyses demonstrated profound alterations, especially in ALT, cholesterol, glucose, and triglyceride levels, by UCN treatment.

DISCUSSION
To detect the impact of urocortin on musculoskeletal metabolism, we used an established osteoporotic rat model (65,66). Five groups of rats were generated, two of which received different concentrations of urocortin (referred to as "UCN low" and "UCN high"), one was treated with Estradiol (OVX-Estradiol), one was just ovariectomized (OVX) and one was not (SHAM). The treatment regimens were chosen according to preliminary tests after literature review (41,73).
Successful ovariectomy was demonstrated by the significantly raised body weight and reduced uterus weight in all OVX groups, while estrogen treatment lead to similar body weight compared to SHAM rats and an increase in uterus weight; UCN treatments did not alter body weight or uterus weight (40).
Muscle analysis was started with the largest muscle of the lower leg, M. gastrocnemius, which functions as a dynamic muscle because it contains more fast-twitch type IIb fibers in rats than in humans. In the slow-twitch type I fibers and type IIb fibers as well, no differences in diameter after treatments with urocortin were detected. The M. soleus, as a holding muscle, mainly consists of slow and constantly contracting type I fibers (46). Here, type I fibers were significantly thicker after higher urocortin treatment compared to the OVX group. These observations raise the question of cause and effect, which is not clear at this point.
Hinkle et al. (41) reported that urocortin could prevent loss of skeletal muscle mass. In their rat model, they analyzed M.  (41). In our OVX model, we also observed enhanced muscle weight and muscle fiber size in M. soleus at the high dose and M. longissimus at the low dose. Therefore, we were able to demonstrate that urocortin has a positive effect on muscle mass even already at lower doses applied in our study. In M. longissimus, which consists of approximately 20% type I fibers (46), we observed a significant decrease in type I fiber diameter in OVX+Estradiol group compared to the OVX+UCN low, but not between any UCN treatment and OVX. These results suggest a superior effect of the higher urocortin concentration on M. soleus compared to M. longissimus. The UCN high treatment exerted its greatest effects compared to SHAM and OVX in M. soleus, a possible hint that type I fibers benefit more from urocortin treatment than type IIb fibers do. Capillary density did not differ significantly between the groups among all muscle studied.
The analysis of creatine kinase and aspartate aminotransferase in serum, which can serve as an unspecific marker of muscle damage (74,75), did not reveal significant differences between the groups, though the higher concentration of urocortin led to higher levels compared to all other groups. The impact of this substance on muscle damage may be questioned since a common marker of muscle loss, LDH, was not affected by UCN treatment in any muscle studied. Citrate synthase activity was higher in the OVX+UCN high group than OVX, which supports a favorable effect of UCN on muscle tissue, as is known for estradiol (76). However, the decreased activity of complex I in M. gastrocnemius after both UCN doses reflects the reduced mitochondrial activity and impaired possibility for an oxidative stress response (77) and should be evaluated in further studies.
The moderate effect of the urocortin treatment in this study can at least partially be attributed to the low dose administered and different animal model applied. An anabolic effect after urocortin I treatment on triceps surae muscles in dystrophic mdx (5Cv) mice has been seen at a dose of 300 µg/kg (bw) after 2 weeks (compared to our 35 days), which was 10-fold higher than the highest dose that we injected (78). Hinkle et al. (41) reported significant inhibition of muscle loss after 9-day UCN treatments in different atrophied muscle mouse models, which, however responded to the treatment differently. Whereas, glucocorticoidinduced muscle atrophy was inhibited at doses from 10 to 300 µg/kg (bw), the casting model was affected by doses higher than 100 µg/kg (bw). Furthermore, various muscles responded to the UCN treatment differently (41).
The adverse effects of urocortin on the cardiovascular system, with raised values of cholesterol, HDL and triglycerides, cause solicitude. Particularly, the impact of our highest UCN concentration on cholesterol has not been described in the literature. However, the cardio-protective effect of UCN is wellknown (79)(80)(81)(82). Our findings could nonetheless point out to some harming long-term effects of UCN. The results reached statistical significance in alanine transaminase (ALT), for which the high urocortin treatment group showed the highest values, indicating liver damage through this specific treatment. In a rat model with intracerebroventricular administration of UCN (0.01-1 nmol/day), these effects have also been reported after a 13-day treatment (83). The effects on glucose metabolism as the raised uric acid in the lowest-urocortin-concentration have also been described elsewhere (84). Heart, liver and kidney weight were also raised, which could indicate another side effect of urocortin treatment. The reduced spleen weight after the higher urocortin dose reveals once more the broad effect of urocortin, which we do not completely understand to this point.
Since, after entering blood circulation, the potential site of action for UCN are osteoblasts and osteoclasts as well as myocytes and immune cells (85,86), we performed an analysis of bone parameters.
Regarding bone parameters, biomechanical evaluation revealed reduced yield load and maximum load in the UCN low group compared to SHAM and OVX+Estradiol group, similar to that in the OVX group. In the UCN high group, the biomechanical properties of bone were at the level of SHAM and Estradiol groups. Similar results were demonstrated in the inorganic mass in the ashing analysis, confirming the results of our group on rat femora and vertebrae, where UCN showed overall positive effects on bone parameters (39,40). One possible mechanism could be the inhibition of osteoclast differentiation and function described by Combs et al. (87).
These results are, however, contradicted by the calcium levels measured in our study, where both UCN groups had the lowest levels. More precise evaluation of tibia with micro-CT did not confirm the harmful effects of low-dose urocortin treatment on trabecular or cortical bone, but it also did not confirm the positive effects that have been reported for high-dose urocortin treatment in spine and femur (39,40). A possible explanation could be that effects of UCN are site specific, which has never been described before, but bone loss and microarchitecture changes in OVX rat models were described to be site specific (88). An explanation for the differential muscle changes could be that due to the less bone response underneath, the lower leg muscles did not show the same effect as M. quadriceps femoris would above the femur, which was not examined in our study. Contradictory, M. longissimus did not show significant effects despite spine did.
We conclude that the effects of urocortin I seem to be fiber-type specific and that type I fibers benefit more from the treatment. Though postmenopausal muscle loss is a phenomenon known to affect fast-twitch muscle fibers more than slow-twitch fibers, Urocortin seems to be a promising substance for muscle tissue in estrogen-deficient organism.
The diameter of oxidative muscle fibers increased after UCN treatment in M. soleus compared to the SHAM, OVX, and OVX+Estradiol groups. The ratio of capillaries to muscle fibers was higher in M. soleus after UCN high treatment, whereas parameters of muscle damage in serum were not altered due to urocortin-treatment. Bone parameters were not improved significantly, whereas parameters of metabolism during urocortin treatment, including increased cholesterol and triglycerides, should be examined more deeply in future studies. Urocortin is a promising substance for further studies regarding skeletal muscle, but its optimal dose of application and potential harmful effects on healthy and diseased tissues need to be further characterized.

LIMITATIONS
In our model, functional muscle tests like strength test, wire hang test or different activity tests were not implemented. Therefore, the assertions regarding functional muscle parameters, although indirectly obtained by muscle enzyme activity and structural parameters, are missing.
There was no formal, blinded analysis of behavior: future studies should formally assess the effects of UCN treatment on home cage locomotor activity since effects on locomotor activity could affect muscle physiology.
Since this study does not include sarcopenic rats, the described effects just show the influence of UCN on skeletal muscle in an estrogen deficient osteoporotic rat model. Furthermore, the weight of M. longissimus was not measured which is also a limiting factor in this study. The effect of UCN on skeletal muscle in the OVX rat is limited compared to denervated or casted mice since the latter leads to a well-defined muscle atrophy, while the former mimics the estrogen-deficit in postmenopausal women.
Measured bone parameters are just static and do not imply analysis of osteoblast and osteoclast relation or bone turnover. Ultimate statements regarding the safety of UCN cannot be made until all possible side effects would be investigated in detail.

ETHICS STATEMENT
The experimental design was approved by the local institutional animal care and use committee (district authorities of Oldenburg, Germany, registration numbers: 33.9-42502-04-10/0246).