Pathological Features of Reinnervated Skeletal Muscles Following Crush Injury of the Sciatic Nerve in Ob/ob Mice

Background: Obesity is a factor for insucient improvement of motor function for peripheral nerve disorders. The aims of this study were to evaluate the skeletal muscles during denervation and reinnervation following nerve crush injury in ob/ob mice. Methods: Experiments were performed on the skeletal muscles of the hindlimbs in 20 male leptin-decient (ob/ob) mice and control mice. Firstly, the characteristics of the gastrocnemius muscles in the mice were evaluated by histological analysis, immunohistological analysis, and Sircol-collagen assay after measurement of body weight and wet weight of the skeletal muscles and by walking tracking analysis. In the histological analysis, nicotinamide adenine dinucleotide tetrazolium reductase (NADH-TR) staining, oil red O staining, and Picro-sirius red staining were performed to assess the type of myobers, lipid accumulation, and collagen deposition, respectively. Then, the models for denervation and re-innervation were made by crushing the sciatic nerves with smooth forceps. The same assessments were performed on the skeletal muscles of nerve crush models. Results: The wet weight of the gastrocnemius muscles was signicantly less in the ob/ob mice than the control mice, whereas body weight was signicantly more. Histological analyses demonstrated a smaller cross-sectional area of type II bers and increase of type I ber grouping of the skeletal muscles in the ob/ob mice. In addition, there was excessive deposition of lipids and collagens between the myobers. Following the nerve injury, the recovery of motor function was equal between both groups, while the cross-sectional area of type II bers was signicantly smaller in the ob/ob mice than the control mice at 4 weeks. Furthermore, the denervated muscles showed an increase in collagen deposition to the area of intermyobers, which were predominant in the ob/ob mice after the nerve injury. Conclusions: The present study showed an increase of collagen deposition, delayed recovery of type II myobers, and type I ber grouping during denervation and re-innervation in the skeletal muscles of ob/ob mice. We suggest through these ndings


Background
The recovery of motor strength and sensory de ciency are the goals of treatment for peripheral nerve disorders including entrapment neuropathies and spinal disorders. Nonetheless, clinicians have occasionally encountered patients with insu cient improvement of motor function, whereas denervated muscles due to disorders of the peripheral nerves could be recovered by proper treatment facilitating axonal regeneration, which results in re-innervation to the skeletal muscles. The insu ciency is a serious problem because the reduction of muscle strength has a potentially devastating outcome on daily activity.
For functional recovery after peripheral nerve injury, the condition of the distal portion, including denervated muscle and the distal stump of the injured nerve, is likely to be more crucial than the proximal portion [1]. Most researchers have paid attention to the pathophysiology of the peripheral nervous system for recovery of peripheral nerve injuries [2,3], whereas we can nd few studies regarding the pathophysiology of skeletal muscles in processes during denervation and re-innervation. The denervation-induced skeletal muscles have biochemical and physiological changes including loss of muscle mass, reduction in the number of satellite cells, and formation of brotic tissue [4,5]. In addition, underlying disease may aggravate the denervation and re-innervation process of the skeletal muscles.
Obesity, one of the metabolic syndromes, is characterized by elevated adipose storage in subcutaneous and visceral tissues and non-adipose organs, a phenomenon called ectopic lipid accumulation [6,7].
Owing to deposition of lipids, in ammation and brosis are caused to various organs including the heart and liver, leading to diastolic dysfunction and non-alcoholic steatohepatitis (NASH), respectively [8,9].
These conditions nally progress to serious disorders including heart failure and liver cirrhosis. The skeletal muscles are also a target organ of ectopic lipid accumulation in obese individuals, by which in ammation and brosis impair skeletal muscle function [10]. Taken together, we suspect there could exist excessive deposition of collagen due to both denervation and lipid toxicity in the denervated muscles of obese individuals with peripheral nerve disorders.
Leptin-de cient (ob/ob) mice become insulin resistant due to de ciency of leptin and develop severe obesity. These animals have been used for various studies in the pathophysiology of the diseases related to obesity including cardiovascular disease, renal disorders, and diabetic mellitus. Previous studies show that the number of non-myelinated bers decreased in ob/ob mice compared with control mice, whereas there existed no change of the number of myelinated nerve bers, such as motor nerve bers [3].
Nonetheless, we could nd no research regarding the skeletal muscles during re-innervation in the animal model of obesity following denervation due to peripheral nerve injuries.
Besides, shifts in ber type could also be the cause of insu ciency of skeletal muscle function during reinnervation. A previous study demonstrated that the process of re-innervation was faster in type I bers than type II bers [11]. As a result, re-innervation to type I bers led to an increase of ber type grouping of type I bers in reinnervated muscles. In addition, several investigators showed an increase of type I bers and a decrease of type II bers with advanced age [11]. Therefore, it should be understood whether there would be differences of ber type in reinnervated muscles under chronic in ammation conditions such as obesity.
The purposes of this study are 1) to evaluate the skeletal muscle in ob/ob mice, 2) to assess the myo bers and the deposition of lipids and collagen on skeletal muscles after denervation in ob/ob mice, 3) to evaluate the ber type of reinnervated skeletal muscles.

Subjects
Experiments were performed on 20 male C57/BL6-ob/ob mice and the same number of their control strain, C57/BL6 mice (10 weeks old; SLC, Hamamatsu, Japan). The animals were housed in a temperature-controlled environment and maintained on a 12-hour light-dark cycle with food and water available ad libitum. The experimental protocol was approved by the committee of animal research at Mie University.

Surgical Procedure
The mice were deeply anesthetized with an intraperitoneal injection of sodium pentobarbital (0.05 mg/g body weight) following the measurement of body weight. For sciatic nerve crush, a longitudinal cutaneous incision was made on the right hind limb under aseptic conditions. Then, the sciatic nerve was exposed by separation of the gluteal muscle, and crushed with smooth forceps for 15 seconds at the level of the sciatic notch, according to previously reported methods [12,13]. The indentation of the sciatic nerve was con rmed; the skin incision was closed with 5 − 0 surgical suture. At 1, 2, and 4 weeks after the surgery (n = 5 at each week), the skeletal muscles of gastrocnemius and tibialis anterior were fully excised from the attachment site. After the measurement of the muscles' wet weight, the gastrocnemius muscles were immersed in isopentane and cooled to freezing point with liquid nitrogen for 10-20 sec, to allow complete freezing to make frozen muscle tissue specimens [14]. The tibialis anterior muscles were snap frozen with liquid nitrogen for the collagen assays. In addition, the other ve mice in each ob/ob and control group served as normal controls, and the skeletal muscles were collected.

Footprint Recording And Sciatic Functional Index (s ) Analysis
A walking track analysis was performed based on previous methods [15][16][17], before and at 1, 2, and 4 weeks after the nerve crush. The total spread (TS), print length (PL), and intermediate toes were measured on the experimental (nerve crushed) side (E) and the contralateral normal side (N), in each mouse. Ten whereas around − 100 SFI represents total dysfunction.

Histological Analysis
The frozen specimens of the gastrocnemius muscle were transversely cut to 10 µm at the center of the muscle using a cryostat. The tissue sections were subjected to nicotinamide adenine dinucleotide tetrazolium reductase (NADH-TR) staining, oil red O staining, and Picro-sirius red staining according to standard protocols, for muscle ber type indication, lipid droplets, and collagen deposition, respectively.
The images of the stained sections were obtained at a magni cation of 200x using an optical microscope (BX50; Olympus, Tokyo, Japan). On the NADH stained sections in 15 random elds, the number of each myo ber type was counted using image analysis software (Lumina Vison, MITANI, Japan). In addition, the improvement rates were calculated based on the number of myo bers in the normal control. The percentage of enclosed type I bers was also determined to evaluate the degree of ber type grouping. The grouping was considered enclosed if a type I ber was completely surrounded by type II bers within the same muscle bundle. Moreover, areas of lipids droplets and collagen deposition between myo bers were measured using the same software. The areas were compared between the sections of the normal controls and the ob/ob mice at 4 weeks after nerve crush. Sircol-collagen Assay (sca)

Immunohistochemistry
The muscles of the tibialis anterior muscle frozen with liquid nitrogen were excised into 40 mg pieces and homogenized; immersed in 400 µl of 0.5 M acetic acid with stirring for 12 hours, centrifuged at 15,000 g x 60 min, and the supernatants were collected. The extracted collagen was measured according to SCA protocol (Sircol soluble collagen assay kit, Biocolor, China). Spectrophotometric readings were taken at 540 nm on a Fluostar Optima microplate reader. Absolute values were attained with a standard graph composed from collagen type I standard supplied with the kit in the range 5-50 µg per 100 µl. The values were expressed as percentage by muscle wet weight.

Statistical analysis
Analyses were performed with SPSS version 22.0 statistical package (SPSS Inc., Chicago, IL, USA). Data between ob/ob and control mice were analyzed using the Mann-Whitney U test. P < 0.05 was considered statistically signi cant.

Results
Skeletal muscles in ob/ob mice had small cross-sectional area of type II bers with excessive deposition of lipid and collagen in the area of the intermyo bers Body weights of ob/ob mice were 41.9 ± 2.3 g (mean ± SD) at 10 weeks after the birth, and were signi cantly heavier than that of control mice 25.4 ± 1.2 g (p < 0.01). In contrast, muscle wet weights of both tibialis anterior and gastrocnemius of ob/ob mice were lower in ob/ob mice (35.3 ± 2.1 mg and 97.5 ± 9.3 mg, respectively) than control mice (47.9 ± 2.7 mg and 127.9 ± 7.9 mg, respectively) (p < 0.01) (Fig. 1). Histological analysis demonstrated that the size of the cross-sectional area in type II ber was signi cantly smaller in ob/ob mice (1459.0 ± 392.2 µm 2 ) than that in control mice (2355.0 ± 610.0 µm 2 ) (p < 0.05). The percentage of enclosed type I bers was also signi cantly higher in ob/ob mice (1.4 ± 1.7%) than that in control mice (7.0 ± 5.1%) (p < 0.01). Meanwhile, the sizes of type I bers were not different between both groups (Fig. 2). Furthermore, the area of lipid droplets and collagen deposition and the number of adipocytes were signi cantly larger in ob/ob mice compared with controls ( Fig. 3, 4) (p < 0.05). Skeletal muscle collagen levels measured by SCA also showed signi cantly higher collagen content in skeletal muscle in ob/ob mice (4.7 ± 0.3%) than in the control group (3.0 ± 0.4%; p < 0.05).
Recovery of motor function was equal between ob/ob and control mice following the nerve injury, even though the cross-sectional area of type II bers was small in ob/ob mice SFI was immediately reduced after crush and then gradually improved until 4 weeks (-108.1 ± 17.0 at 1 day, -26.1 ± 5.4 at 4 weeks after crush). The ob/ob mice showed a slight tendency for low SFI at 1 day, and at 4 weeks (-115.7 ± 11.3, -27.0 ± 16.2, respectively), but no signi cant difference was observed between the two groups, and the ob/ob mice showed equal improvement at 4 weeks (Fig. 5). Furthermore, cross-sectional areas in both type I and type II bers of the gastrocnemius muscles were predominantly decreased in ob/ob mice as well as control mice at 1 week after nerve crush. Thereafter, the crosssectional areas were increased in type I and type II bers at 1 week and 2 weeks later after the nerve injury. In control mice, the cross-sectional areas were recovered up to 86.2 ± 1.4% and 83.1 ± 9.2% of the non-affected side in type I and type II bers at 4 weeks after the nerve injury, respectively. By contrast, the ob/ob mice had an improvement of 79.6 ± 8.9% of cross-sectional area in type I bers, but had insu cient improvement of 71.2 ± 6.9% of the non-affected side in type II bers. In addition, there was a signi cant difference of the cross-sectional areas of type II bers after the nerve injury (p < 0.05) (Fig. 6). Moreover, the percentage of enclosed type I bers was increased in both the ob/ob and control groups (13.7 ± 2.3% and 4.2 ± 4.0%, respectively). It signi cantly increased in ob/ob mice (p < 0.05) (Fig. 7).
Collagen deposition was increased due to nerve injury, especially in ob/ob mice Furthermore, we noticed the increase of collagen deposition after the nerve injury on specimens for histological analysis. Picro-sirius red staining showed that collagen deposition was clearly increased between bers in the skeletal muscles at 4 weeks after the nerve injuries in both ob/ob and control mice. Especially, the increase of the collagen was predominant in ob/ob mice, and the rates of collagen area were signi cantly larger in specimens of ob/ob mice than in the control mice (p < 0.05), regardless of the crush injury to the nerves (Fig. 8). SCA also showed that the amount of collagen was 8.3 ± 2.4% in the reinnervated muscles of ob/ob mice following the crush injury of the sciatic nerve, which was signi cantly higher compared to ob/ob and control mice without the nerve injury (4.7 ± 0.3% and 3.0 ± 0.4%, respectively) (p < 0.05) (Fig. 9).

Discussion
The skeletal muscles in ob/ob mice had a smaller cross-sectional area in type II bers, with excessive deposition of lipid and collagen in the area of the intermyo bers, when compared to the control group. Despite this, the recovery of motor function was equal between ob/ob and control mice following the nerve injury. Collagen deposition was increased both groups due to nerve injury, especially in ob/ob mice.
Obesity is a causative factor for the disorders of various organs including cardiovascular tissues and the liver, which is a target of ectopic lipid accumulation. The excessive lipid accumulation to the non-adipose tissues can cause cell dysfunction or cell death via pro-in ammation, and these processes have been de ned as lipotoxicity [18]. As a result, in ammation and brosis are caused to the heart and the liver, leading to diastolic dysfunction of the heart and non-alcoholic steatohepatitis (NASH), respectively [8,9]. Consequently, these conditions can progress to serious conditions including heart failure and liver cirrhosis. The lipotoxicity is a concern for the skeletal muscles, since skeletal muscles are also a target for ectopic lipid accumulation in individuals with obesity. In fact, we found a number of studies regarding the pathophysiology of the lipotoxicity to skeletal muscle cells based on in vitro study [19,20], while the adverse effect of lipotoxicity in vivo and the functional consequences are unresolved.
In the present study, we used ob/ob mice as the animal model of obesity, in which lipid accumulation and increases of adipocytes were clearly shown in the gastrocnemius muscles. Therefore, the ob/ob mice are an appropriate model of in vivo exploration into the adverse effect of lipotoxicity to the skeletal muscles. Furthermore, the skeletal muscles were observed to be atrophic in the ob/ob mice, in which reduction of cross-sectional areas of type II bers was predominant. In addition, the skeletal muscles had more collagen deposition between the myo bers in the ob/ob mice than in the control mice. These ndings are similar to those in liver and heart patients with obesity [21,22], suggesting that the skeletal muscles in the obese subjects could be atrophic through chronic in ammation owing to lipotoxicity. In fact, a previous study showed that obese adults had a reduction of quadriceps muscle strength relative to body mass compared to non-obese adults [23].
Furthermore, previous studies reported that patients with obesity had inferior function in comparison to non-obese patients after surgery for peripheral nerve disorders, even though the surgical treatments had good clinical results in comparison to the preoperative condition. Roh et al. stated that patients with metabolic syndrome, which is strongly related to obesity, had decreased pinch strength and delay of functional recovery after the surgical treatment for carpal tunnel syndrome [24]. Burgstaller et al. described fewer obese patients with meaningful improvement than non-obese individuals in the surgical treatment for lumbar canal stenosis [25]. These studies suggest obesity is associated with insu cient improvement due to impairment during the denervation and re-innervation process in the surgical treatment of peripheral disorders.
Nonetheless, the improvements in the functional status were equal between ob/ob and control mice after the nerve injury, while histological analysis showed that recovery of the cross-sectional area in type II bers was delayed and poor in the ob/ob mice than in the control mice. In general, type I bers are associated with endurance, while type II bers are characterized as 'fast bers', and associated with strength of skeletal muscles [26]. The results might indicate inferior recovery of muscle strength after the nerve injury in patients with obesity, even though the muscle strength was not measured in this study. By contrast, the recovery of cross-sectional area of type I bers was not different between the two groups, which is likely to explain the reason for no difference in functional status.
Likewise, an increase of type I ber grouping was observed in the ob/ob mice, especially after the nerve injury. Previous studies have demonstrated that re-innervation of type I bers is faster than that of type II bers, in which preferential re-innervation of type I bers in recovery following nerve injury leads to an increase in type I grouping in reinnervated muscles [11,27]. Furthermore, shifts of ber groupings are associated with aging in the denervation and re-innervation processes, suggesting that an increase of type I grouping might in uence the function of reinnervated muscles, such as contraction of the myo bers [11]. This study suggests obesity could not only cause a shift of ber type in the skeletal muscles which result in decreased muscular strength, but also cause insu cient recovery after the nerve injury.
Furthermore, the collagen deposition on the intermyo bers was increased in the skeletal muscles following the nerve injury [28]. The brotic formation is characterized by an anomalous accumulation of the extracellular matrix, such as collagen around in amed tissues, and leads to the dysfunction in various tissues including the heart and liver [29]. In myocardial tissue, this reactive and progressive interstitial brosis contributes to myocardial stiffness with progression to ventricular dysfunction. Even though the pathological signi cance of the brosis is less elucidated in the skeletal muscles than in other organs, brosis of the skeletal muscle is a characteristic feature of muscular dystrophin, myopathies, and traumatic injuries [30]. In addition, previous studies show that brosis caused a decrease in skeletal muscular strength [31][32][33]. Therefore, abnormal accumulation of brotic tissues, which were observed in the denervated and reinnervated muscles of ob/ob mice, also may lead to stiffness of the skeletal muscles and result in a reduction of contractile force. In the future, reduction of body weight, anti-brotic drugs, and inhibition of fatty acid for the recovery of motor function in peripheral nerve disorders need to be evaluated.
In conclusion, this study showed that the skeletal muscles of ob/ob mice were smaller in cross-sectional area of type II bers with excessive deposition of lipid and collagen at the interstitial area at myo bers. Following injury of the sciatic nerve, recovery of motor function was equal between ob/ob and control mice following nerve injury, whereas the skeletal muscles of ob/ob mice not only had a signi cantly smaller sectional area of type II bers than the control mice, but also comprised an increase of type I ber grouping during denervation and re-innervation. In addition, collagen deposition was signi cantly increased in the skeletal muscles of the ob/ob mice after the nerve injury. We suggest that lipotoxicity to type II bers, grouping of type I bers, and interstitial brosis due to collagen deposition could be the causes of insu cient improvement after surgery for peripheral nerve injury in obese individuals. All procedures involving animals were approved by the committee of animal research at Mie University.

Consent for publication
Not applicable.

Availability of data and material
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Figure 1 Body weight and muscle wet weight The muscle wet weight of ob/ob mice is signi cantly less than that of the control mice in both the tibialis anterior and gastrocnemius muscles. *Mann-Whitney U test Page 13/17 p<0.01.

Figure 2
NADH staining a: The size of the type II ber cross-sectional area is signi cantly smaller in ob/ob mice than in the control mice. b: The percentage of enclosed type I bers in ob/ob mice is signi cantly higher. *p<0.05.

Figure 4
Page 15/17 Collagen deposition The areas of collagen deposition are signi cantly larger in ob/ob mice compared with the controls. * p<0.05.

Figure 5
Sciatic functional index SFI was immediately reduced after crush and then gradually improved. A signi cant difference was not observed between the two groups.

Figure 6
Page 16/17 Cross-sectional area of gastrocnemius after sciatic nerve crush The cross-sectional area of the muscle ber decreased after nerve crush, and then improved. The cross-sectional area of the type II muscle ber is signi cantly poorly improved in the ob/ob mice. There is no obvious difference in type I. * p<0.05

Figure 7
Enclosed Type I bers Enclosed Type I is signi cantly increased after nerve crush. This is especially noticeable in the ob/ob mice. * p<0.05. Figure 8