In mammals, the number of fibers in skeletal muscles is determined at birth and changes little through life; however, muscle fiber cross-sectional area could be modified as result of a reduction or increase in the number of myofibrils (Tamaki et al., 2002). There are many factors that can modify the cross-sectional area such as immobilization, starvation, aging, gonadal hormones levels, and denervation (Dedkov et al., 2001; Jackman and Kandarian, 2004). In rats, skeletal muscles are drastically affected by denervation which reduces weight and fiber size (Nnodim, 1999; Dedkov et al., 2001), commonly accompanied by muscle fiber nuclei loss as well as muscle fiber death (Borisov and Carlson, 1995). Additionally, denervation leads to an increase in packed bundles of collagen fibers, an early increase followed by a decline in satellite cell number (Lu et al., 1997), and a fast mass loss resulting in a lack of contractile force (Araki et al., 1991). It has also been shown that gonadal hormones can influence skeletal muscle morphology, since low levels of androgens following castration reduce weight and fiber size of bulbocavernosus/levator ani muscle complex, which are recovered after testosterone treatment (Foster and Sengelaub, 2004).
The morphology of striated muscle fiber depends on both neural activity and gonadal hormones. Gonadal hormone effects are commonly described to act through a direct action on the muscle fiber (Herbst and Bhasin, 2004); however, recently evidence has been presented for an indirect action on the neuromuscular complex which includes the neuromuscular junction dimensions and activity (Balice-Gordon et al., 1990; Foster and Sengelaub, 2004), motoneuron soma size modification (Breedlove and Arnold, 1981; Kurz et al., 1986), increase of acetylcholine receptor number (Bleisch and Harrelson, 1989), and production of brain-derived neurotrophic factor by the muscle (Yang et al., 2004). Meanwhile, the neural effect on the muscle fiber is only through the motoneuronal innervation. In the current literature, most of studies exploring the effects produced by denervation on striated musculature, usually did not take into account the remaining circulating gonadal hormones. Likewise, when exploring gonadectomy effects it is often forget the remaining action of the muscle innervation. Therefore and in order to avoid the above mentioned neglects, in the present study we analyzed the unilateral nerve transection from the pubococcygeus muscle (Pcm) of gonadally intact and castrated male rats to explore both neural and hormonal effects under the same environmental factors, that is, within the same animal.
It has been claimed that the Pcm in humans belongs to the levator ani complex (Jundt et al., 2005) and participates in fecal and urinary continence (Fucini et al., 2001). In male rats, this muscle is not part of the levator ani complex (Yuan et al., 2003); however, its participation in micturition and ejaculation is highlighted through behavioral studies (Manzo et al., 2000) and by electrical stimulation of its nerve (Manzo et al., 1997). Furthermore, by analyzing the cross-sectional area of its fibers, we have shown that this muscle is sensitive to gonadal hormones (Alvarado et al., 2008), but we still do not know whether this effect is due to a direct action of these hormones on its fibers and/or through the neuromuscular complex, in which hormonal changes affect soma size, synaptic efficiency, and neuromuscular junction size of muscles such as the bulbocavernosus and levator ani (Breedlove and Arnold, 1981; Balice-Gordon et al., 1990; Tanaka and Arnold, 1993).
The effect of denervation on the cross-sectional area of Pcm fibers in male rats is currently unknown. Therefore, the aim of the present study was to explore the effect of denervation as well as the gonadal hormone influence on denervation effects upon Pcm fibers. Hence, considering the bilaterality of this muscle we analyzed the effect of denervation versus non-denervation throughout unilateral nerve transection of the corresponding muscle in the same animal. Moreover, we have also analyzed the effect of denervation versus non-denervation in castrated male rats, in order to explore the magnitude of the action on fiber cross-sectional area of the Pcm by both surgeries.
MATERIALS AND METHODS
Throughout the study, animals were treated and maintained according to the Policies on the Use of Animals in Neuroscience Research (The Society for Neuroscience), the Policy on Humane Care and Use of Laboratory Animals (National Institutes of Health), and the guidelines of the Instituto de Investigaciones Biomédicas (Universidad Nacional Autónoma de México) of which the internal Bioethical Committee carefully reviewed and subsequently approved this research protocol.
The left and right Pcm of 12 adult male Wistar rats (approximately 100 days old) initially weighing 277 ± 5.6 g were randomly allocated into two equal groups. After surgery (see below), the animals were individually housed in acrylic cages of 32 × 27 × 15 cm3. Animals were housed in a temperature-controlled room with a 12:12 hr reverse light/dark cycle with ad libitum access to food (Harlan, Mexico) and water.
The rationale of this experiment was to study the effect of six weeks of unilateral denervation on the cross-sectional area of Pcm fibers in six male rats, and compare the results with those of the non-denervated muscle.
Assignment of Pcm groups: Denervated (Den) and non-denervated (N-Den).
The rationale of this experiment was to study the effects of six weeks of simultaneous castration and unilateral denervation on the cross-sectional area of Pcm fibers in six male rats, and compare the results with those of the non-denervated muscle.
Assignment of Pcm groups: Castrated plus denervated (C+Den) and castrated non-denervated (C+N-Den).
Through an abdominal midline incision made under intraperitoneal anesthesia with sodium pentobarbital (45 mg/kg b.w. Anestesal, Smith Kline, Mexico), either a unilateral right Pcm denervation or castration plus unilateral right Pcm denervation were performed. Denervation consisted of a 5 mm removal of the somatomotor branch from the right pelvic nerve close to the muscle (see Fig. 1). For castration, the vas deferens and the corresponding blood vessels were ligated and the testes removed.
Pcm Extraction and Histological Procedure
Six weeks after surgery, animals were deeply anesthetized using intraperitoneal sodium pentobarbital. Both Pcm were carefully extracted following partial removal of the pubic bone. Immediately after extraction, the animals were euthanized with an overdose of anesthesia. Muscles were then immersed in Helly's fixer for 24 hr, washed with distilled water for 2 hr, and dehydrated in alcohols. Using a similar procedure previously described (Alvarado et al., 2008), 20 transverse serial sections (7 μm each) were obtained from a piece of muscle (4 mm in length) close to the origin, and stained with hematoxylin and eosin.
One of the 20 transversal sections was randomly selected (simple random sampling) from each of the denervated and non-denervated muscles. The section at ×10 magnification was then divided using a grid with 12 × 8 compartments (100 μm × 100 μm each compartment). From this grid, five sample fields were randomly selected, and from each one of these sample fields, five fibers (similar to those represented in Fig.1 from Alvarado et al., 2008) were randomly selected and a blinded second observer carried out the measurements. A total of 25 fibers per muscle, that is, 150 fibers per group were analyzed using a light microscope (Olympus BH-2) and software for morphology measurement (Sigma ScanPro 4; Aspire Software International, USA). The mean cross-sectional area of the 25 Pcm fibers was calculated for each muscle within each group of muscles and a grand mean was determined for comparisons among groups. Furthermore, the percentage distribution of the cross-sectional area values from Pcm fibers grouped in bins of 500 μm2 was included only for qualitative descriptive purposes.
Cross-sectional area average values of Pcm fibers were analyzed within each experiment using paired Student's t-tests. Comparisons among groups of cross-sectional area values were analyzed with a linear mixed-effects model in which subject effect was used as random factor and treatment effect was used as fixed factor and a Bonferroni test was used as post-hoc. Frequency histograms were included only for qualitative purposes. The data in the text are presented as the mean ± SEM. The significance level for all tests was considered as α = 0.05. Data analyses were done using SPSS 20 (IBM, USA).
Fiber cross-sectional area mean value of the Den group was significantly lower than that of the N-Den group [t(5) = 12.78, P < 0.05]. According to percentage distribution of fibers (Fig. 2), N-Den group of muscles presented a cross-sectional area range distribution from 500 to 3999 μm2 with a high percentage of fibers at the 1,000–1,499 μm2 bin. The Den group of muscles presented a cross-sectional area range distribution from 500 to 2499 μm2 with a high percentage of fibers at the 1,000–1,499 μm2 bin. Moreover, in comparison with N-Den, the Den group showed an increase in the percentage of fibers from 500 to 1499 μm2 and a lack of fibers from 2,500 to 3,999 μm2.
As compared to cross-sectional area average values from fibers of the C+N-Den group, values of the C+Den group were significantly lower [t(5) = 6.99, P < 0.05]. The percentage distribution of fibers (Fig. 2) in C+N-Den group of muscles presented a cross-sectional area range distribution from 500 to 1,999 μm2, with a high percentage of fibers at the 1,000–1,499 μm2 bin. The C+Den group of muscles had a cross-sectional area range distribution from 500 to 1,999 μm2 with a high percentage of fibers at the 500–999 μm2 bin. Furthermore, in comparison with the C+N-Den, C+Den group presented a higher percentage within the 500–999 μm2 bin but a lower percentage of fibers between 1,000 and 1,999 μm2.
Comparisons Between Experiments
When the cross-sectional area mean values from groups of Experiments 1 and 2 were compared (Fig. 3), we found that all groups were significantly different from each other [F (3, 20) = 1.66, P < 0.001]. The cross-sectional area values of the N-Den group were the largest, while the C+Den group had the lowest cross-sectional area value. Moreover, we found that the mean value of C+N-Den group was significantly lower than that of Den group.
The percentage distribution of fibers from groups in Experiments 1 and 2 showed that compared to N-Den group, the C+N-Den group showed a lack of fibers from 2,000 to 3,999 μm2 whilst a higher percentage of fibers in both 500–999 μm2 and 1,000–1,499 μm2 bins. However, when compared to Den group, the C+Den group presented a lack of fibers from 2,000 to 2,499 μm2, and a lower percentage of fibers in both 1,000–1,499 μm2 and 1,500–1,999 μm2 bins, while presenting an important increase in the percentage of fibers from 500 to 999 μm2 (Fig. 2).
In skeletal muscles, neural activity regulates its contractility, maintaining the muscle fiber size. Thus, a lack of muscle activity reduces the cross-sectional area of the fibers (Brown et al., 2001), while denervation decreases number of capillaries promoting a low blood circulation (Borisov et al., 2000). Studies have shown that during the first few weeks after denervation, there is a tendency for the number of satellite cells to double or triple, inducing myogenesis (Viguie et al., 1997; Borisov et al., 2001). In soleus muscle, denervation increases fiber size four weeks after surgery suggesting an important regeneration process (Wanek and Snow, 2000). This phenomenon can explain why electrical properties of contraction and relaxation are gradually affected after denervation, since original fibers are replaced by regenerated ones (Lewis and Schmalbruch, 1994).
Gonadal hormones also influence skeletal muscle morphology through a direct and indirect action on muscle fibers (Rand and Breedlove, 1992; Herbst and Bhasin, 2004; Verhovshek et al., 2010). Interestingly, denervation affects all striated muscles while castration does not, for example, plantaris is not sensitive to changes in gonadal hormone levels (Antonio et al., 1999). However, it has been reported that the bulbocavernosus complex responds equally to both denervation and castration (Godinho et al., 1987; Araki et al., 1991). Moreover, studies involving differences between castration and denervation in limb musculature of male rats have found that denervation had a greater effect in reducing the fiber cross-sectional area (Lu et al., 1999), similar to that described in the levator ani complex (Nnodim, 1999). Through our model of study, it was also possible to compare the effect of denervation versus castration in Pcm fibers within the same male rat. We found that castration had a greater effect than denervation, that is, castration presented a lack of ranges in the cross-sectional area distribution from 2,000 to 3,999 μm2, while in denervation it was from 2,500 to 3,999 μm2.
In relation to our findings, it seems that other than the reduction from a direct hormonal effect on the muscle fibers, castration reduces the indirect effect on the neuromuscular complex (Monks et al., 2004; Verhovshek et al., 2010) resulting in a reduction of the Pcm reflex activity, similar to that described in the bulbocavernosus muscle (Hart, 1967; Davidson et al., 1978). In contrast, denervation produced a lesser effect because denervation itself does not prevent the direct and indirect action of gonadal hormones on the Pcm fibers. This differential effect of denervation versus castration here described was corroborated by the simultaneous combined surgeries, through which an additional effect was observed where the fiber cross-sectional area ranges were mostly located from 500 to 1,499 μm2 and where as shown in Fig. 3, the mean cross-sectional area value was significantly the lowest.
The different effects observed in Pcm fibers versus those reported for the limb musculature (Lu et al., 1999) could be explained by the type and function of the muscle in study. Pcm location close to the midline of the body, its insertion into the sacral vertebrae, and its participation in micturition (Manzo et al., 1997), ejaculation (Manzo et al., 2000), and tail posture (Martinez-Gomez et al., 1992) are characteristics that allows us to consider the Pcm as an epaxial muscle; thus, implying that when compared to castration, denervation effect in epaxial muscles is not as strong as it is in striated muscles of limbs. Another possible explanation is that most studies of limb muscles were carried out in a separate way, that is through bilateral denervation in which besides the evident lack of muscle activity, environmental factors such as hormonal status, feeding behavior, mobilization, and a different muscular blood flow supply could be involved. In contrast, our model of study avoided these variables and additionally, it allowed us to explore a probable compensatory effect on the contralateral non-denervated muscle as it has been described in other musculature (Lu et al., 1999). In this regard, the Pcm fibers in the contralateral non-denervated muscle of gonadally intact rats showed a tendency toward an increase (data not shown) that was noticed when compared to values from intact males (Alvarado et al., 2008); however, further studies are necessary to address this issue properly.
In summary, this is the first study demonstrating that castration is more effective than denervation in reducing the cross-sectional area of striated fibers. According to present results, the Pcm that is involved in micturition and ejaculatory processes should be considered as an epaxial muscle, which cross-sectional area of its fibers depends on neural as well as direct and indirect gonadal hormone effects. Our study pointed out that during illness in which striated musculature is involved, it is highly important to consider in the therapeutic treatment the type of muscle, likewise the balance of neural and gonadal hormone dependence.
Authors would like to express their great appreciation to Dr. Leanne Fraser and Carolina Rojas for their constructive suggestions to this study.