Sixty male Sprague Dawley (SD) sedentary rats, housed in plexiglass cages since birth (∼8 weeks old), were randomly divided into two groups (30 rats/group): the resistance exercise and the control animals. They were weighed weekly. To minimize the number of animals used in this study, each rat was also subjected to a sham injury on the right leg and a sciatic nerve injury on the left leg (Oñate et al., 2016). Sciatic nerve injury was either a mild sciatic nerve compression (SNC) or a moderate sciatic crush injury (SCI). The rats were kept at an ambient temperature of ∼23 °C under a 12-h light-dark cycle (lights on from 8:00 p.m. to 8:00 a.m.) with free access to pelleted chow and water. Ethical approval was obtained from the Health Sciences Center Ethics Committee of Kuwait University, and the rats were treated in accordance with guidelines on the humane handling of experimental animals.

A newly developed resistance exercise system was used (Al-Sarraf and Mouihate, 2022). Briefly, using a tunnel and a pulley system, the rats were induced to pull a pre-determined weight tied to their tail. Rats were trained three times per week for 10 weeks by increasing the pulling load progressively and the number of runs at each training session, as well as weekly. The resistance exercise protocols were performed during the animals’ dark cycle in a dark room with a small, dim light. The control animals walked in the tunnel in a similar manner to the resistance exercise group without applying any load.

A standard aneurysm clip (Yasergil Phynox, FE-753K) was used to induce SNC. According to the manufacturer, this clip produces a constant 70 g/mm² force. To find a proper compression duration, the sciatic nerve compressions at ∼1 cm away from the greater trochanter were performed for either 3, 5, or 10 min (6 rats/group). The Functional recovery was assessed using the toe spread reflex (TSR) test on days 2, 5, 7, 9, and 12 post-SNC. TSR scores showed no significant difference between 3- and 5-min compressions, with animals appearing fully recovered by day 9 (Figure 1). In contrast, 10-min compression produced a more distinct recovery pattern, but full recovery was delayed until day 12. Accordingly, 10 min of compression and day 5 post-SNC were selected for further analysis.

A 12 cm stainless-steel micro-mosquito haemostatic forceps with a serrated straight tip (1.3 mm in width) at three clicks was used to crush the sciatic nerves for 60 s, as this crush duration was used previously (Altun and Kurutaş, 2016; Renno et al., 2015). The site of injury was at ∼1 cm from the greater trochanter to the midshaft and ∼0.5 cm caudally to avoid the trifurcation of the sciatic nerve (Savastano et al., 2014). We used the TSR test daily during the first week to confirm that sciatic nerve crush produced consistent motor deficits suitable for monitoring recovery. All animals showed clear TSR impairment.

Twenty-four hours after the last resistance exercise or walking with no load through the tunnel, the animals were anesthetized intraperitoneally with ketamine 50 mg/kg (Rotexmedica, Germany) and xylazine 5 mg/kg (Interchemie Company, Holland), and subcutaneously injected with antibiotic Enrofloxacin 5% (T Special LTD., United Kingdom). Ointment Tobrex (Novartis, Spain) was applied to the eyes to prevent dehydration, and both thighs were shaved using an electric shaver. The rat was placed on its side with a rectal probe on a heated pad to maintain the body temperature during the surgery. All surgical tools were sterilized in an autoclave before surgery. The skin was cleaned under the femur bone with betadine and 70% ethanol. Using a surgical blade, the muscles below the femur bone were dissected, and the sciatic nerve was exposed and cleaned of the surrounding tissues. On the left leg, the compression groups received SNC using the aneurysm clip, while the crush groups received SCI using micro-mosquito haemostatic forceps. A few drops of sterilized saline were applied, and the muscles were sutured. The skin was wiped with sterilized H₂O₂ (6%) and closed with 9 mm autoclips or sutured. On the right leg, sham surgery was done following the same procedure but without applying any compression or crush injury to the nerve.

To assess motor functions, toe spread reflex (TSR), foot positioning, extensor postural thrust (EPT), and CatWalk tests were performed on all rats (Renno et al., 2017; Dayawansa et al., 2014; Parvathy and Masocha, 2013). Briefly, the TSR test was performed by holding the rat from its tail and lowering it to a surface to check for toe separation while hanging and on the surface (Renno et al., 2017). This test was evaluated using a three-point scale where score 1 corresponds to no separation of toes, which is the least functional, score 3 corresponds to full separation of the toes, indicating a fully functional toe reflex, and score 2 corresponds to anything in between.

For both SNC and SCI, the motor function tests were performed 1 week pre-surgery. For the SNC, the motor functions were tested on the second and fifth days post-injury. In contrast, the SCI resulted in larger damage for which the recovery took a longer time. In a pilot study we observed a partial recovery by the eighth and the 12 days post-SCI. For this reason, we tested motor functions for a longer period of time lapsing from the 2nd until the 12th days post-SCI to account for the longer recovery period in the SCI model.

This test was performed at 1 week pre-surgery, second, and fifth days post-SNC model, and for the SCI, it was performed at 1 week pre-surgery, 2nd, 5th, 8th, and 12th days post-injury.

A foot positioning test was also used to assess the position and posture of the feet of the rat while standing and walking (Renno et al., 2017). This test was also evaluated using a three-point scale, where score 1 means full eversion, score 2 is mild eversion, and score 3 is when the foot is fully supported on the surface, indicating fully functional foot positioning. This test was done on the same selected time points that were used for the TSR test.

To assess the power of the hind limbs, the EPT test was performed. This test was done by holding the rat with a piece of cloth upright and placing their foot on an open digital scale (Dayawansa et al., 2014). The measurements were obtained in grams corresponding to the force of the lower limb. This test was also done on the same selected time points that were used for the TSR and foot positioning tests. For these three functional tests, an average of three measurements was taken for each leg. Rats’ gait was monitored using the CatWalk gait analysis system (Noldus Information Technology, Netherlands) (Parvathy and Masocha, 2013). The captured videos were analyzed using CatWalk software version 7.1. Parameters such as the maximum contact area, intensity of contact, print area, standing phase, and swinging phase of the hind paws were measured (Parvathy and Masocha, 2013). This test was done 1 week pre-surgery, 2 and 5 days after SNC, and at 1 week pre-surgery, 5 and 12 days after SCI.

Rats were anesthetized with urethane 1.5 mg/kg (Sigma Life Science, China) intraperitoneally and perfused with ice-cold PBS. The sciatic nerves were collected on the fifth day post-SNC (Figure 1) and the 14th day post-SCI, as shown in previous studies (Renno et al., 2017; Ramli et al., 2017; Wang et al., 2018). The intact and injured sciatic nerves with a length of ∼1 cm at the site of injury were extracted, snap-frozen in liquid nitrogen then stored at −80 °C for the Western blot experiment. The site and distal regions (∼1–2 mm) of the injured sciatic nerve were collected and fixed by immersion in 3% glutaraldehyde and Millonig’s phosphate buffer for the transmission electron microscopy (TEM) study.

Nerve samples were added to a solution of lysis base buffer with a protease inhibitor tablet (Roche Applied Sciences, Germany) and homogenized using Next Advance Bullet Blender Storm 24 (USA), then centrifuged at 12,000 rpm for 15 min at 4 °C. The supernatant was collected, and the protein levels were assayed using the bicinchoninic acid (BCA) assay. The proteins were separated using the SDS-PAGE technique using a 15% acrylamide gel. Semi-dry transfer was used to transfer the proteins from the gel to a nitrocellulose membrane (Bio-Rad company, Germany). After the transfer and blocking in 5% fat-free dry milk, the membrane was washed and incubated with primary antibodies overnight at 4 °C. The membranes were incubated in a primary monoclonal mouse anti-myelin basic protein antibody (anti-MBP, Calbiochem, 1:5000) or monoclonal mouse anti-phospho-MBP antibody clone P12 (anti-pMBP, Merk KGaA Darmstadt, Germany, 1:3000). After washing, the membranes were incubated with horseradish peroxidase-conjugated (HRP) donkey anti-mouse IgG (H + L) secondary antibody (Invitrogen, 1:2000) for 2 h at room temperature. Beta actin polyclonal rabbit primary antibody was used as a house-keeping protein (Invitrogen, 1:5000), followed by donkey anti-rabbit IgG secondary antibody (Invitrogen, 1:5000). The membranes were then exposed to an enhanced chemiluminescent solution (clarity western ECL substrate; Bio-Rad, USA), and the immunoreactive bands (IR) were digitally detected using Azure Biosystems scanner (USA). Optical density of the protein bands was measured using Azure Analysis Software version 2.0.028. The ratios of the OD of each protein to that of actin were calculated and used as a semi-quantitative measure of protein levels.

Sciatic nerves were dehydrated using increased concentrations of ethanol and embedded in epoxy resin. Ultra-thin cross sections (100 nm) were cut from the resin blocks and mounted on copper grids. The sections were then stained with uranyl acetate and lead citrate. Images were acquired using a 1500X objective using JEOL’s JEM-1200 EXII, Transmission Electron Microscope (Tokyo, Japan). The morphometric parameters studied were focused on the myelinated axons, including the number of myelinated axons, myelin thickness, and g-ratio. TEM Images were taken from the sciatic nerves of SNC (3 per rat, 3 rats per group) and SCI rats (5 per rat, three rats per group). For each rat, the images were taken from the sham sciatic nerve, and at two sites of the injured sciatic nerve: at the site of injury, and ∼1–2 mm distal to the site of injury. Therefore, a total of 9 (SNC model) to 15 (SCI model) images were taken. The number of myelinated axons was counted per image using Fiji ImageJ software. The mean diameter of axons of the myelinated fibers (d) in µm and the mean diameter of the whole myelinated nerve fibers (D) in µm were calculated automatically using Image Pro Plus version 10 analysis software (Media Cybernetics, Rockville, MD, USA), which is commonly used to analyze TEM nerve cross-sectional images (Renno et al., 2017; Ronchi et al., 2023). From the mean diameters, we calculated the myelin thickness as (D-d/2) and the g-ratio as (d/D). A smaller value of the g-ratio indicates better myelination of the nerve fiber. The number of myelinated axons and the average of each parameter per image were averaged by rat. Comparison was done between each injured group (site of injury or distally) with its corresponding sham group, and between exercised and control rats at each studied area. For the SNC model, a total of 2079 axons were analyzed for myelin parameters with a range of around 210–482 axons/group. For the SCI model, a total of 3278 axons were analyzed for myelin parameters with a range of around 280–703 axons/group.

Statistical analysis was performed using SPSS (version 25.0). All behavioral data and body weight were analyzed using repeated measures ANOVA followed by an adjusted Bonferroni post hoc test. The data from Western blot and TEM were analyzed using a two-way ANOVA followed by a Bonferroni multiple comparison test. Statistical significance was considered when (p-value <0.05). Data were presented as mean ± SEM.

To assess the impact of resistance exercise on peripheral nerve recovery, we applied two types of nerve injuries to the sciatic nerve: the sciatic crush injury (SCI) and the sciatic nerve compression (SNC). As shown in Figure 2A, the SNC resulted in significantly lower scores for TSR (Figure 2Aa), foot positioning (Figure 2Ab), and EPT (Figure 2Ac) in both the control and exercised rats on the second and fifth day post-SNC when compared to the scores in their respective sham-treated legs. A slight, but significant improvement was observed in these three scores on the fifth day post-SNC injury that was not dependent on the resistance exercise but was probably due to the naturally occurring recovery. Similar effects were noticed when the sciatic nerve was subjected to the SCI, as seen in Figure 2B. While there was normal improvement of these scores over time, ranging from day 2 to day 12 post-SCI injury, we observed that resistance exercise enhanced these recoveries, particularly in TSR (Figure 2Ba) and EPT (Figure 2Bc) scores on day 8, and in TSR scores on day 12.

The gait steps were analysed using the CatWalk system (Figure 3). Following SNC, both control and exercised rats showed significantly lower maximum contact areas (Figures 3Aa, Ab), step intensity (Figure 3Ac), and print areas (Figure 3Ad) of the hind paws compared to their corresponding sham-operated leg on the second and the fifth day. However, the standing phase was not significantly affected by the compression in either control or exercised rats on either the second or the fifth day after SNC (Figure 3Ae). The swinging phase was longer in both the control and exercised rats when compared to their corresponding shams on the second day post-SNC (Figure 3Af), while on the fifth day after the compression, only the exercised rats showed significantly longer swinging durations compared to their sham. Resistance exercise did not significantly improve any SNC-induced alteration of these gait parameters.

As shown in Figure 3B., SCI significantly lowered the maximum contact area (Figures 3Ba, Bb), the intensity of the contact of the hind paws (Figure 3Bc), and print area (Figure 3Bd), and increased the swinging phase (Figure 3Bf) in both exercised and control rats when compared to their corresponding shams on the 5th and 12th days. Compared to their corresponding shams, the standing phase was significantly reduced on the fifth day after SCI in the control rats, and on the 12th day in both control and exercised rats (Figure 3Be). As was seen with SNC injury, resistance exercise did not significantly improve any SCI-induced alteration of these gait parameters (p > 0.05).

Levels of MBP and pMBP at the site of the SNC on the fifth day post-compression injury are shown in Figure 4A. The compression had no significant effect on the levels of either MBP or pMBP isoforms at the injured nerves in both the control and exercised rats when compared to their corresponding shams. Also, there was no significant difference in these myelin levels between the injured exercised and control rats.

In contrast, the SCI (see Figure 4B) resulted in a significant reduction in the levels of an isoform of MBP with an apparent MW of 20–22 kDa in both the control and exercised rats when compared to their corresponding sham-operated legs (Figure 4Bc). However, the injury had no significant effect on the levels of any pMBP isoform in the crush-injured nerves of either control or exercised rats when compared to their corresponding shams (Figures 4Be, Bf, Bg). Resistance exercise did not significantly affect the expression levels of either MBP or pMBP in the crush-injured sciatic nerve.

The transmission electron microscope images of sciatic nerves 5 days after SNC show that both the control and exercised shams display normal thick myelin sheath around their axons with occasional scattered unmyelinated axons (Figure 5Aa). Resistance exercise did not appear to affect the morphology, density, and distribution of the myelinated axons of uninjured sham sciatic nerves. Compared with their corresponding shams, the nerve fibers subjected to SNC in both control and exercised rats showed slightly thinner and irregular myelin sheaths with intra-axonal debris, which is more apparent in the areas distal to the site of compression (Figure 5Aa). Resistance exercise did not significantly alter the number of myelinated axons (Figure 5Ab), myelin thickness (Figure 5Ac), and myelin g-ratio (Figure 5Ad) in the sciatic nerves subjected to the compression injury.

Fourteen days after SCI, both exercised and control rats showed damage features at the site of injury and distally when compared to the regular myelinated axons with thick, compact myelin sheaths seen in the shams (micrographs in Figure 5Ba). Injured nerves exhibited irregular, loosely compacted myelin sheaths with intra-axonal debris. Compared to the control, resistance-exercised crush-injured sciatic nerves showed a similar pattern in the morphology of damage at the site of injury, but showed a better morphology with greater density of myelinated fibers and thicker myelin sheaths at the distal areas (micrographs in Figure 5Ba). Morphometric analysis of the micrographs showed no significant difference in the number of myelinated axons (Figure 5Bb) and myelin thickness (Figure 5Bc) of injured nerves at the site of injury for both the control and exercised rats when compared to their corresponding sham-operated legs. Although statistical comparison of the number of myelinated fibers distally did not reach significance (p = 0.08), visual inspection of electron micrographs revealed a greater myelinated fiber density in the resistance-exercised group compared to the non-exercised one (Figure 5Bb and micrographs in Figure 5Ba).