Nature Standard Company, Shanghai, China, isolated and purified authentic standards of PND ((3R)-9,10-epoxy-1-ene-4,6-diyn-3-ol). Nuclear magnetic resonance (NMR) spectroscopy (${}^1$H-) was used to identify PND, and the full spectral datasets were consistent with previously published values [23, 26, 27]. Purity was determined to be greater than 98% by high-performance liquid chromatography (HPLC). PND was stored at –20 °C in the dark and brought to room temperature before use. The solvent control solution was composed of 12% Kolliphor HS 15 (Sigma-Aldrich Inc., St. Louis, MO, USA) and 88% normal saline. Prior to injection, PND was further diluted in solvent control solution to a final concentration of 10 mg/mL. Kolliphor HS 15 has been widely used as a pharmaceutical excipient, including as a solubilizer, absorption enhancer, emulsifier, plasticizer, and vehicle for controlled drug delivery applications. According to published reports, this concentration has no discernible effect [28].

Twenty adult SD rats (female, 7 weeks of age) were provided by the Animal Care Facility of Shanghai Jiao Tong University, School of Medicine (Shanghai, China). The rats were housed at a constant temperature (25 $\pm$ 2 °C) and were given ad libitum access to food on a 12-h light/dark cycle. All experiments were performed in accordance with the Chinese Council for Animal Care guidelines and were performed in accordance with the Animal Care Committee of Shanghai Jiao Tong University School of Medicine.

Anesthesia was administered by intraperitoneal injection of sodium pentobarbital (40 mg/kg body weight, i.p.), and all animal studies were performed under aseptic conditions. Sciatic nerve transection injury was established as previously described [29]. The right sciatic nerve was exposed 1.0 cm distal to the sciatic notch and transected in the middle with sharp scissors. The lesion site was then repaired with four stitches of 9–0 microsutures (Jin Huan Co., China) through the epineurium under 20–40$\times$ magnification. The contralateral nerve was exposed but was not transected (control).

After surgery, the rats were randomly assigned to two groups: the PND group or the solvent control group. Simultaneously, all groups were given an intraperitoneal injection of the corresponding medication for two consecutive weeks at the same fixed time. Rats in the PND group received PND (10 mg/kg) daily, whereas rats in the control group received the same volume of solvent daily. Then, the rats were randomly divided into two separate experiments, as shown in Fig. 1; in experiment 1, n = 3 per group per timepoint, while in experiment 2, n = 4 per group.

Fig. 1.

Schematic diagram of the experimental procedure. On day 0, sciatic nerve transection was performed in all rats (n = 20). Following injury, the rats were randomly assigned to two groups: the PND group or the solvent control group (n = 10/group). All groups received an intraperitoneal injection of the corresponding medication for two consecutive weeks. Rats in the control group received 0.5 mL solvent daily, whereas rats in the PND group received 0.5 mL PND (10 mg/kg) daily. Then, in experiment 1, the rats were sacrificed at the indicated time points for the sampling of nerve tissues for real-time PCR (n = 3/group) analyses. In experiment 2, the walking track analysis (for evaluating motor function) along with the Hargreaves plantar test (for evaluating sensory function) were performed at the indicated time points (n = 4/group). After functional and behavioral tests were completed, the rats were weighed and sacrificed for the sampling of liver, nerve, muscle, and plantar skin tissues at 16 weeks for H&E staining, IF staining, and transmission electron microscopy (n = 4/group).

The mRNA levels of BDNF, NGF, tropomyosin-related kinase B (TrkB), and p75 neurotrophin receptor (p75NTR) in the injured distal nerve were examined by real-time PCR one and two weeks after surgery. A 1.0-cm-long section of the sciatic nerve on the distal end of the lesion was isolated and removed from anesthetized animals of both groups.

Total RNA was extracted using TRIzol (15596-018, Invitrogen, Carlsbad, CA, USA), and 2 $\mu$g of total RNA was reverse transcribed using the PrimeScript RT Reagent Kit (Perfect Real Time, DRR037A, TaKaRa Biotechnology, Kusatsu, Japan) according to the manufacturer’s instructions. The mRNA levels of BDNF, NGF, TrkB, and p75NTR were quantitatively measured using SYBR Select Master Mix (4472908, Roche Diagnostics, Mannheim, Germany) and the Applied Biosystems Step-One Plus Real-Time PCR System (Life Technologies, Austin, TX, USA). The unregulated control glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used to normalize mRNA expression. The sequences of the forward and reverse primers are presented in Table 1. The threshold cycle (Ct) was calculated using the second-derivative maximum method. The data were analyzed using the delta-delta method. The assays were performed three times using triplicate wells. The final values are reported as the ratio to the control.

All behavioral experiments described below were conducted in a blinded manner in a quiet room (temperature 25 $\pm$ 1 °C) from 9 AM to 6 PM. The surgeon who performed the transection injury and the behavioral observers were different individuals. The rats were habituated to the experimental setting for three consecutive days before the start of the experiment, and the daily acclimatization time was calculated to be the duration of the experiment.

The Hargreaves plantar test (Yuyan, Shanghai, China) was used to assess the injured hindlimb’s heat sensitivities and sensory functional recovery of rats (n = 4/group). The tests were performed at six time points: 1 day before surgery and 1 day, 4 weeks, 8 weeks, 12 weeks, and 16 weeks after surgery. The light intensity of the selected heat source beam was IR = 75%, and the upper limit of the pain threshold latency was set to 20.1 s to avoid scalding the rat’s foot. Briefly, each rat was placed on the glass plate of the pain tester, separated by a special transparent plastic box, and allowed to adapt for more than 30 min. The light source of the tester was switched on while ensuring that the heat source beam emitted by the transmitter aim was located outside of the middle part of the sole of the foot. The time interval between initial activation and paw withdrawal in response was recorded automatically. Heat stimulation was stopped to prevent thermal injury if the animals did not withdraw their paws after 20 s. The test was repeated three times for each rat (15-min intervals were included between tests), and the average value of each measurement was recorded to represent the thermal pain threshold.

The recovery of motor function in rats (n = 4/group) was examined by walking track analysis at six time points: 1 day before surgery and 1 day, 4 weeks, 8 weeks, 12 weeks, and 16 weeks after surgery. The rats were habituated to the experimental setting for three consecutive days before the start of the experiment, and the daily acclimatization time was calculated to be the duration of the experiment.

Then, the sciatic functional index (SFI) was calculated as previously described [30, 31]. Briefly, the rats were trained to walk across a custom-made walking box (100 cm long, 12 cm wide, and 10 cm high) leading to a darkened box. White paper (12 cm $\times$ 100 cm) lined the bottom of the track. The rat’s hind paws were dipped in red dye before it walked the track at each timepoint. Each rat was then allowed to walk the track three or four times until five measurable footprints were collected. Then, the SFI was measured and calculated using the following formula:

SFI = (–38.3 $\times$ (EPL-NPL)/NPL) + (109.5 $\times$ (ETS-NTS)/NTS) + (13.3 $\times$ (EIT-NIT)/NIT) – 8.8

PL: the distance from the heel to the top of the third toe;

IT (intermediary toe spread): the distance from the second to the fourth toe;

TS (toe spread): the distance between the first and the fifth toe.

The normal print length (NPL), normal toe spread (NTS), and normal intermediary toe spread (NIT) represent the PL, IT, and TS recorded from the uninjured, normal foot; experimental print length (EPL), experimental toe spread (ETS), and experimental intermediary toe spread (EIT) represent the PL, IT, and TS recorded from the paired, injured, experimental foot, respectively. An SFI value of approximately 0 indicates full recovery, whereas an SFI value of approximately 100 represents complete dysfunction.

After functional and behavioral tests were completed, the rats (n = 4/group) were weighed and sacrificed carefully 16 weeks after surgery. Organs and tissues including the liver, injured nerve, plantar skin, and gastrocnemius muscles were collected from the injured and uninjured sides. The livers and gastrocnemius muscles from the injured and uninjured sides were then weighed. Blood vessels and the deep fascia covering the surfaces of the liver and muscle were removed and discarded before weighing. The following equation was used to determine the rate of body weight gain, liver hepatosomatic, and recovery rate of the gastrocnemius muscles:

Rate of body weight gain = (final body weight-initial body weight)/initial body weight

Hepatosomatic index = wet weight of the liver/final body weight

The recovery rate of the gastrocnemius muscle = wet weight of the injured muscle/wet weight of the uninjured muscle

After weighing, the liver and mid belly of the gastrocnemius muscle specimens were cut into small pieces, fixed in 4% paraformaldehyde/PBS, and dehydrated in a graded series of ethanol. Then, the samples were paraffin-embedded, cut into 5-$\mu$m-thick sections, and stained with H&E. The H&E-stained fibers were analyzed in a blinded manner using ImageJ software (NIH, Bethesda, MD, USA). The fibers were manually evaluated, and the program was used to calculate the cross-sectional area of each fiber. The quantitative study was conducted using the cross-sectional area of muscle fibers as the unit of measurement.

After functional and behavioral tests were completed, a 0.5-cm-long piece of the sciatic nerve at the 1 cm distal end of the lesion was separated and removed from the rats (n = 4/group). The nerves were sliced into 5-$\mu$m-thick cross-sections and stained with toluidine blue. We assessed the morphology in cross-sections of nerve to measure and calculate the diameter, myelin thickness, and g-ratio of the myelinated nerve fibers of the distal nerve stump.

Myelin staining with toluidine blue: The segments were stained with 1% toluidine blue at room temperature for 30 min, washed gently in water, and consecutively soaked in 95% ethanol and 100% ethanol for 2–3 s. Images were acquired at 400$\times$ magnification using a dissecting microscope (M205FA, Leica, Wetzlar, Germany). The axon and fiber diameters were measured using ImageJ software (NIH, Bethesda, MD, USA). Myelin thickness and g-ratio were calculated as follows: myelin thickness = (fiber diameter – axon diameter)/2; g-ratio = axon diameter/fiber diameter.

Transmission electron microscopy: Sections of regenerated nerves were fixed in precooled 2.5% glutaraldehyde for 3 h followed by postfixation in a 1% osmium tetroxide solution for 1 h, washed, dehydrated, embedded in Epon 812 epoxy resin, cut into ultrathin sections at 60 nm, and stained with lead citrate and uranyl acetate. The stained sections were observed under a transmission electron microscope (JEM-1200, JEOL Ltd., Tokyo, Japan).

Nerve, skin, and muscle were postfixed in 4% paraformaldehyde/PBS overnight after which tissues were incubated in 30% sucrose/PBS overnight for dehydration. Then, the tissues were frozen in Tissue-Tek O.C.T. (4583, Sakura Finetek, CA, USA). Then, 10-$\mu$m-thick specimens were sectioned using a cryostat (Thermo Fisher Scientific, Austin, TX, USA), mounted on glass slides, and dried for future immunostaining. The slides were washed three times in 0.01 M PBS for 5 min, blocked with 10% NGS/0.1% Triton X-100 in 0.1 M PBS for 1 h at room temperature, and incubated overnight at 4 °C with primary antibodies (PGP 9.5, nerve terminals marker, Abcam ab8189, 1:500). After three 10-min washes in PBS, the slides were incubated with CY3-conjugated goat anti-mouse IgG (Abcam, ab97053, 1:100) for 1 h at 37 °C and rinsed three times in PBS for 5 min each time. To stain the end plate, the slides containing muscle tissue were incubated with FITC-conjugated bungarotoxin (Life Technologies, B-13422, 1:1000) for an additional 5 h at 37 °C, followed by three 10-min washes in PBS. Gel/Mount aqueous mounting media (Biomeda Corp., Foster, CA, USA) was used to mount the coverslips, after which slides were viewed under a digital fluorescence microscope (M205FA, Leica, Welzlar, Germany). Five random visual fields of the gastrocnemius muscle sections at 100$\times$ magnification were selected, and the numbers of end plates in 10 sections from each reinnervated muscle specimen were counted. Only end plates double-stained with both bungarotoxin and PGP9.5 were counted.

Statistical analyses were performed using GraphPad Software (GraphPad 9, La Jolla, CA, USA). The data are presented as the means $\pm$ standard deviations. One-way analysis of variance followed by the Student-Newman–Keuls post hoc test was used to compare the PND and solvent control groups or other groups as indicated. A p value 0.05 was considered statistically significant.

The toxicity and safety of traditional Chinese medicine have garnered increased attention in recent decades. The liver is the primary site of drug transformation and metabolism and is also the main target organ for drug toxicity [32, 33, 34]. To evaluate the side effects and toxicity of PND, we investigated body weight gain, hepatosomatic index, and liver histopathologic characteristics of both groups of rats. Rats in both groups survived the experiment without obvious ascites, emaciation, or self-mutilation. Each rat gained approximately 40–52 g of body weight in both groups, and no significant difference was observed in the rate of body weight gain or in the hepatosomatic index between the two groups (Fig. 2C; n = 4, p $>$ 0.05). As shown in Fig. 2, 16 weeks after nerve transection, H&E staining and light microscopy revealed an intact structure of liver lobules in rats of both groups; liver plates were neatly arranged, and no obvious morphological changes, such as swelling, necrosis, or inflammatory cell infiltration, were observed. PND did not cause pathological changes in the liver when administered at a dose of 10 mg/kg/d for two consecutive weeks.

Fig. 2.

H&E staining and histological analysis of the liver in each group. Representative light micrographs of H&E-stained liver sections from the solvent control group (A) and the PND group (B) 16 weeks after surgery. The rate of weight gain in each group is shown in panel (C). The hepatosomatic index in each group is shown in panel (D). All data are presented as the means $\pm$ standard deviations. p $>$ 0.05 for comparisons with the solvent control group. Scale bar: 100 $\mu$m.

After 16 weeks, cross-sections of tissue 1 cm distal to the injury were obtained for the morphologic analysis of the ingrowth of regenerated nerve fibers. SCs surround regenerated nerve processes to reform myelin sheaths or unmyelinated fibers, which provides a structural basis for the subsequent recovery of nerve conduction function [35]. Along with the density of regenerated nerve fibers, the thickness of the myelin sheath, the diameter of nerve fibers, and the g-ratio all serve as important indicators of regenerated nerve fiber maturation. Toluidine blue staining was used to assess distal nerve fiber formation and myelination. The results of toluidine blue staining at the distal end of the injury are shown in Fig. 3A,B. Sixteen weeks after sciatic nerve injury, both groups demonstrated significant regeneration of nerve fibers at the distal end of the injury, with some fibers forming myelin sheaths. Additionally, the average fiber diameter and myelin sheath thickness were significantly greater in the PND group than in the solvent control group (Fig. 3C; n = 4, p 0.05). However, PND treatment did not significantly improve the degree of myelination (g-ratio) after injury compared with the solvent control (Fig. 3C; n = 4, p $>$ 0.05). The ultrastructure of regenerated fibers was revealed by transmission electron microscopy. The morphological appearance and mature structure of the regenerated axons in the PND group were superior to those in the control group (Fig. 3D–G).

Fig. 3.

Morphological appearance and morphometric assessments of regenerated and myelinated nerves in each group. Representative images show toluidine blue staining of regenerated axons (A,B) 1 cm distal to the injury site in the solvent control group (A) and PND group (B) 16 weeks after injury. A, B: scale bar: 50 $\mu$m. Morphometric assessments of regenerated and myelinated nerves (C). All data are presented as the means $\pm$ standard deviations. *p 0.05 compared with the solvent control group. Representative electron micrographs of regenerated axons (D,E) and the myelin sheath (F,G) 1 cm distal to the injury in the solvent control group (D,F) and PND group (E,G) 16 weeks after surgery. D,E: scale bar: 5 $\mu$m; F,G: scale bar: 1 $\mu$m.

By collecting, weighing, and staining the gastrocnemius muscle with H&E 16 weeks after nerve injury, we assessed the healing of target organs following treatment.

As illustrated in Fig. 4A,B, gastrocnemius muscle atrophy was detected in both groups of rats following denervation.

Fig. 4.

Morphological appearance and morphometric assessments of gastrocnemius muscle fiber recovery in each group. Representative light micrographs of H&E-stained transverse sections of the gastrocnemius muscle (A,B) from rats in the solvent control group (A) and the PND group (B) 16 weeks after injury. A, B: scale bar: 100 $\mu$m. The gastrocnemius recovery rate and the area of a single myocyte in the control group were both lower than those in the PND group (C). All data are presented as the means $\pm$ standard deviations. *p 0.05 for comparisons with the solvent control group.

Fig. 4 illustrates the morphometric evaluations (C). The gastrocnemius recovery rate (the ratio of muscle wet weight on the injured side to that on the contralateral side) and the area of a single myocyte in the control group were both lower than those in the PND group (Fig. 4C; n = 4, p 0.05), which implies that muscle atrophy was more severe in the control group.

To evaluate the target organ condition, immunofluorescence (IF) staining was used to detect the reinnervation of muscle and skin by sciatic nerve terminals in the two groups 16 weeks after injury. Neuromuscular junction staining is shown in Fig. 5A,B. Double staining and colocalization of markers of the end plate and nerve terminals was seen in both groups, which indicates effective reinnervation. The number of reinnervated end plates in each visual field in the PND group was not significantly different from that in the control group (Fig. 5C; n = 4, p 0.05).

Fig. 5.

Morphological appearance and morphometric assessments of reinnervated muscle and skin in each group. Representative images of IF staining for PGP 9.5 (red) and FITC-conjugated $\alpha$-bungarotoxin (green) in reinnervated muscle (A,B) and skin (D,E) 1 cm distal to the injury in the solvent control group (A,D) and PND group (B,E) 16 weeks after injury. (A,B,D,E): scale bar: 50 $\mu$m. Partial neuromuscular junctions (NMJs) were colocalized with PGP 9.5-positive fibers. Morphometric assessments of reinnervated muscle (C) and skin (F). All data are presented as the means $\pm$ standard deviations. p $>$ 0.05 for comparisons with solvent control groups.

Both groups displayed regenerated nerve fiber ingrowth in the plantar skin, as illustrated in Fig. 5D,E. Imaging analysis revealed no statistically significant difference in nerve fiber length per unit area between the PND and control groups (Fig. 5F; n = 4, p 0.05).

The Hargreaves plantar test data were collected every four weeks after injury and revealed that sensory functional recovery was complete in both groups. As illustrated in Fig. 6, the plantar skin of rats gradually recovered sensitivity to heat and pain. After approximately four weeks for the PND group and after approximately eight weeks for the control group, the rats recovered to near-normal levels of sensory function (Fig. 6; n = 4, * p 0.05, compared with before injury). The PND group recovered sensory function more quickly than the control group. Additionally, the PND group experienced greater sensory recovery than the control group after four weeks (Fig. 6; n = 4, ${}^{\mathrm{\#}}$p 0.05).

Fig. 6.

Motor and sensory functional recovery in each group. The SFI (a score of –100 and below corresponds to total paralysis, whereas uninjured animals typically have a score of –8.8 or higher) and plantar test reaction time rate tested 1 day before injury and 1 day, 4 weeks, 8 weeks, 12 weeks, and 16 weeks after injury. *p 0.05 for comparisons with values before injury. ${}^{\mathrm{\#}}$p 0.05 for comparisons with the solvent control group at paired time points.

A greater SFI suggests stronger sciatic nerve function. The return to preinjury SFI levels indicates complete recovery of hindlimb motor coordination and function, including that of the most distal paw muscles. Every four weeks, animals in both groups were given an SFI test to measure the recovery of motor function of the sciatic nerve. Statistical analysis of SFI data indicated that motor functional recovery was only partially achieved in both groups. SFI values remained low compared with preinjury levels at all time intervals evaluated up to 16 weeks (Fig. 6; n = 4, * p 0.05, compared with values before injury), which demonstrates poor recovery of motor function. However, at 8, 12, and 16 weeks after injury, the SFI values in the PND group were substantially higher than those in the control group (Fig. 6; n = 4, ${}^{\mathrm{\#}}$p 0.05), which shows that rats in the PND group experienced greater motor functional recovery.

NGF and BDNF mRNA levels in the injured distal nerve were examined using real-time PCR one and two weeks after injury. As shown in Fig. 7, NGF mRNA levels were similar in the PND group and control group at one and two weeks after surgery (n = 3, p $>$ 0.05). At one and two weeks after surgery, the BDNF mRNA levels in the PND group were 1.57-fold and 1.89-fold higher, respectively, than those in the control group (n = 3, p 0.05). The functions of BDNF are mediated by its high-affinity receptor (TrkB) and low-affinity receptor (p75NTR). Additionally, we used real-time PCR to determine the mRNA levels of p75NTR and TrkB in the injured distal nerve one and two weeks after injury. The expression levels of p75NTR and TrkB mRNA in the PND group were 1.99-fold and 2.09-fold greater, respectively, than those in the control group at one week (n = 3, p 0.05). In addition, the expression of p75NTR mRNA remained high in the PND group (1.90-fold higher than that of the control group) at two weeks (n = 3, p 0.05).

Fig. 7.

Expression of neurotrophic factors and their receptors in each group one and two weeks after injury. The mRNA levels of NGF, BDNF, and their receptors TrkB and p75NTR at the injured sites were determined in the solvent control and PND groups at one and two weeks after injury. Each test was repeated three times. All data are presented as the means $\pm$ standard deviations. *p 0.05 for comparisons with the solvent control group at the same timepoint after injury.