Neuroprotective Effects of Oxymatrine via Triggering Autophagy and Inhibiting Apoptosis Following Spinal Cord Injury in Rats

Spinal cord injury (SCI) is a devastating neurological disorder characterized by high morbidity and disability. However, there is still a lack of effective treatments for it. The identification of drugs that promote autophagy and inhibit apoptosis in neurons is critical for improving patient outcomes following SCI. Previous studies have shown that increasing the activity of silent information regulator 1 (SIRT1) and downstream protein AMP-activated protein kinase (AMPK) in rat models of SCI is highly neuroprotective. Oxymatrine (OMT), a quinolizidine alkaloid, has exhibited neuroprotective effects in various central nervous system (CNS) diseases. However, its explicit effect and molecular mechanism in SCI are still unclear. Herein, we aimed to investigate the therapeutic effects of OMT and explore the potential role of autophagy regulation following SCI in rats. A modified compressive device (weight 35 g, time 5 min) was applied to induce moderate SCI in all groups except the sham group. After treatment with drugs or vehicle (saline), our results indicated that OMT treatment significantly reduced the lesion size, promoted survival of motor neurons, and subsequently attenuated motor dysfunction following SCI in rats. OMT significantly enhanced autophagy activity, inhibited apoptosis in neurons, and increased SIRT1 and p-AMPK expression levels. Interestingly, these effects of OMT on SCI were partially prevented by co-treatment with SIRT1 inhibitor EX527. Furthermore, combining OMT with the potent autophagy inhibitor chloroquine (CQ) could effectively abolish its promotion of autophagic flux. Taken together, these data revealed that OMT exerts a neuroprotective role in functional recovery against SCI in rats, and these effects are potentially associated with OMT-induced activation of autophagy via the SIRT1/AMPK signaling pathway.


Introduction
Spinal cord injury (SCI) is a catastrophic central nervous system (CNS) disorder that affects patients' physical, psychological, and social well-being and places a substantial financial burden on the healthcare system. SCI may be divided mainly into two pathological phases, primary injury 1 3 [1] and secondary injury [2]. The former is attributed to direct mechanical damage to the spinal cord and is irreversible, while the latter is elicited by a complicated cascade of biochemical changes following the primary injury involving edema [3,4], autophagy [5,6], apoptosis [7,8], inflammation [9,10], oxidative stress [11,12], and scar formation [4,13]. Importantly, secondary injury is considered to be reversible and is thought to have an even more pronounced impact on the pathogenesis of SCI than the primary injury [11,14]. To date, there are several potential treatments of SCI involving gene, molecule, and drugs therapy; however, only methylprednisolone (MP) has demonstrated efficacy in clinical trials and thus has been used commonly in the treatment of patients with SCI [15]. In recent years, a series of complications induced by the high-dose protocol of MP has been gradually emerging such as pneumonia, sepsis, wound infections, gastrointestinal bleeding, as well as thromboembolism, indicating MP administration after SCI carries some substantial risks [15][16][17]. Therefore, identifying more effective clinical drugs preventing or alleviating secondary injury is of vital importance.
Neuronal apoptosis is one of the most prominent pathological features of secondary SCI, and the resultant neuronal loss can seriously impede the recovery of locomotor functions following SCI in rats [1,5,18]. Autophagy is the main catabolic mechanism for cells to degrade intracellular proteins and organelles in the cytoplasm by the autophagosomal-lysosomal pathway and plays a crucial role in the maintenance of cellular homeostasis and regulation of both physiological and pathological processes in the CNS [19][20][21]. The autophagy process is believed to be closely related to apoptosis [22,23]. Emerging data have shown that autophagy can exhibit a dual role, both beneficial and detrimental, following CNS injuries, possibly depending on the location and severity of the injury [24,25]. Nonetheless, it appears clearly that enhancing autophagic flux may significantly accelerate neurological functional recovery by inhibiting neuronal apoptosis following moderate SCI models of rats [5,8], which suggests that modulation of the autophagy pathway could represent a promising therapeutic strategy for SCI.
Silent information regulators (SIRTs or Sirtuins) are a family of conserved nicotinamide adenine dinucleotide (NAD+)-dependent histone deacetylase, which consists of 7 mammalian members (SIRT1-7) [26]. The functions of SIRTs vary depending on their subcellular distribution. SIRT1, localized mainly in the nucleus, has been indicated to play a critical role in the regulation of several physiological and pathological processes, including metabolism, autophagy, apoptosis, inflammation, oxidative stress, and aging [27][28][29][30]. Compelling in vivo and in vitro evidence that the levels of autophagy and apoptosis are interlinked with the SIRT1/AMPK (AMP-activated protein kinase) signaling pathway [31][32][33]. After SCI in rats, the SIRT1 activator resveratrol has been shown to promote activation of the autophagy pathway by modulating AMPK activity [31]. Moreover, melatonin has been shown to significantly enhance autophagy and inhibit apoptosis in neurons, thereby promoting neurological functional recovery via activation of the SIRT1/AMPK signaling pathway in an SCI rat model [32]. These previous studies suggested that the SIRT1/ AMPK signaling pathway plays an essential role in the regulation of neuronal autophagy and apoptosis following SCI in rats.
Oxymatrine (OMT) (Fig. 1A), classified as a quinolizidine alkaloid, is a phytochemical product extracted from the traditional Chinese medicinal herb -Sophora flavescens. OMT is primarily a chemotherapeutic agent used in the treatment of a wide variety of cancers [34,35]. Due to its wide range of pharmacological properties, including anti-viral [36], anti-inflammatory [37], anti-apoptotic [38], anti-oxidative [39], and anti-immune [40] effects, OMT has successfully been used for the treatment of various diseases, such as chronic hepatitis [41], bronchial asthma [36], and several disorders involving ischemia/reperfusion injuries in the heart [42], lung [43], liver [38], kidney [39], intestine [44], and brain [45,37]. After hippocampal ischemia-reperfusion injury in rats, OMT can mediate neuroprotection by enhancing autophagy via upregulation of SIRT1 expression and thereby attenuating injury-induced cognitive deficits [46]. Recently, Guan et al. [10] have demonstrated that OMT was able to elicit neuroprotective effects following SCI by reducing inflammation, oxidative stress, and apoptosis. However, the effects and exact molecular mechanisms of OMT on the autophagic activity of neuronal cells in the spinal cord following SCI have not been elucidated. Here, we hypothesized that OMT could promote neuronal autophagy, reduce neural tissue damages, and ultimately alleviate locomotor impairments via activation of the SIRT1/AMPK signaling pathway following SCI in rats.

Animals
Eighty female Sprague-Dawley rats (weighing 180-220 g, aging 9-10 weeks old) were purchased and bred in the central animal house (12-h light/dark cycle with free access to food and water ad libitum) of Jinzhou Medical University (Jinzhou, China). All animal experiments were performed in compliance with the guidelines for the Institutional Animal Care and Use at Jinzhou Medical University. The procedures were approved by the Animal Ethics Committee of Jinzhou Medical University. Great efforts were made to reduce the number and suffering of the animals used in our study.

Experimental Groups
All rats were randomly and blindly allocated into the Sham group (only laminectomy was performed without SCI induction), SCI group (underwent moderate compressive SCI after laminectomy), OMT group (40 mg/kg, intraperitoneal injection immediately after SCI), OMT+EX527 group (a SIRT1 specific inhibitor, 10 μg/kg, intraperitoneal injection immediately after SCI), chloroquine group (abbreviated to CQ group, a classical autophagy-lysosome inhibitor, 50 mg/ kg, intraperitoneal injection immediately after SCI), and OMT+CQ group. The vehicle (saline) was administrated intraperitoneally directly into the abdominal cavity at the same time point after SCI.

Establishment of Rat SCI Models
The rat models of SCI were established as per our previous description [4,9]. Briefly, rats were anesthetized intraperitoneally (i.p.) with pentobarbital sodium (40 mg/kg). Following anesthesia, a 3-cm midline incision was conducted to expose the T9-T12 spinal cord, followed by compressing the spinal cord vertically with a sterile metal impounder (weight 35 g, diameter 2 mm) for 5 min to induce a moderate SCI model [4,9,47]. Signs of a successful model were symbolized with visible injured spinal cord surface blood stasis, quick tail flick reflex, and rapid tremor of both lower limbs of the rats. Rats that had not been successfully modeled were excluded from experiments. After the surgery, the incision was sutured, and all rats were allowed to recover on a 30 °C heating pad. Next, experimental rats received a continuous injection of ceftriaxone sodium (50 mg/kg, i.p.) daily for 3 days, and their bladders were massaged manually twice a day until the spontaneous urination was restored.

Drugs Administration
Oxymatrine (N1835) with a purity of more than 98% was purchased from APExBIO (USA) and prepared at a final concentration of 40 mg/mL in normal saline. EX527 (S1541) with a purity of more than 99.78% was purchased from Selleck (USA) and dissolved in normal saline to 0.04 M/L. Chloroquine (C6628) with a purity of more than 98.5% was purchased from Sigma-Aldrich (USA) and diluted with normal saline to 25 mg/ml. All experimental reagents were standard commercially available. OMT (40 mg/kg) [10], EX527 (10 μg/kg) [48,32], and CQ (50 mg/kg) [49,50]  Obviously, compared with the SCI group, the BBB scores were significantly higher in the OMT-treated group at day 3 and day 7 and in the OMT-EX527 co-treatment group at day 7. Data are presented as mean ± SEM; significant differences among groups are indicated by # P < 0.05, ** P < 0.01, and *** P < 0.001. (D) Footprint analysis in animals of the different treatment groups. Apparently, rats treated with OMT showed clear hindlimb coordination and little toe dragging compared with the SCI rats. Red indicates forepaw footprints; black indicates hindpaw footprints were administered intraperitoneally daily for consecutive 7 days until the animals were euthanized. An equal volume of saline was injected intraperitoneally at the same time points.

Assessment of Functional Locomotor Recovery
The Basso, Beattie, and Bresnahan (BBB) locomotor rating scale [51] and footprint analysis [52] were conducted in our present study to assess the locomotor function recovery of rats. Both of them were performed in a noise-free environment. Scores of all behavioral measures were done by observers blinded to the groups.

For BBB Scores
All animals were tested at days 0, 1, 3, and 7; the highest score (21 scores) indicates normal motor capacity, and the lowest score (0 score) indicates completely paralyzed. All rats were scored by three experimenters who were blind to the experimental protocols. The mean of three measurements was recorded and analyzed finally.

For Footprint Analysis
All animals were tested at day 7; the modified method was referred to in several studies [52]. Briefly, prepare a manual dark box (50 × 4 × 4 cm) paved with white paper (60 cm long, 5 cm wide) at the bottom to make a straight track. Subsequently, the animal's fore-and hindpaws were inked with red and black dyes, respectively. Induce the animals to walk forward straight by placing food at the far end of the box. The footprints were preserved, and the digitized images were analyzed finally.

Hematoxylin-Eosin and Nissl Staining
Hematoxylin-Eosin (HE) and Nissl staining kits were purchased from Wanleibio (WLA051a, China) and Beyotime (C0117, China), respectively. All protocols were carried out as described previously [4,9]. Prepared sections were taken out of the −20 °C and air-dried for 3-4 h at room temperature (RT), followed by conducting the following experiments. All finishing staining sections were finally observed and imaged using an optical microscope (BX43, Olympus, Japan) at a magnification of ×40 or ×100.

For HE Staining
All the procedures were adjusted as follows: Washing with deionized water for 5 min, staining with hematoxylin for 2 min and eosin for 15 s, respectively, dehydrating with 95% ethanol and clearing with xylene for 2 min twice, respectively, and mounting finally with neutral gum.

For Nissl Staining
Sections were soaked in 100% ethanol and chloroform (1:1, v/v) overnight in the dark at RT. The day late, the sections were passed through 100%, 95% ethanol, and deionized water for 1 min, respectively, followed by being incubated in 37-50 °C pre-warmed crystal violet solution for 5 min. Subsequently, the sections were dehydrated in 100% ethanol and cleared with xylene for 5 min twice, respectively, and mounted finally using neutral gum. The number of Nisslpositive neurons was analyzed in four randomly selected areas of the ventral horn of spinal cords.

Quantitative Real-Time PCR Analysis
Total RNA was isolated from rat spinal cord tissues using TRIzol TM reagent (15596018, Invitrogen; ThermoFisher Scientific, Inc.) according to the manufacturer's protocols; the amount and purity of the mRNA extracts were determined by spectrophotometry (DPI-1, Qiagen) based on the ratio of the optical density value measured under 260 nm and 280 nm. Subsequently, reverse transcription was performed to synthesize cDNA from 1 μg of total mRNA using the Pri-meScript™ RT reagent Kit with gDNA Eraser (RR047A, TaKaRa, Otsu, Japan), and the quantitative Real-Time PCR (qRT-PCR) with TB Green® Premix Ex Taq™ II (RR820A, TaKaRa, Otsu, Japan) was conducted to quantify all the gene transcripts on the ABI StepOnePlus Real-Time PCR System (Thermo Fisher Scientific, CA, USA). The thermocycling conditions were adjusted as follows: initial denaturation (95 °C, 30 s), 40 cycles for amplification reaction including denaturation (95 °C, 5 s), annealing (60 °C, 20 s), and extension (65 °C, 15 s). The specified primers utilized in our present study were designed using the software Primer 6.0 (Applied Biosystems) based on the sequences obtained from Oligo 7 (Applied Biosystems) and displayed in Table 1. We evaluated the ratio of gene expression profiling by the threshold cycle (Ct) and used Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) whose ΔΔCt value was set at 1 as the housekeeping gene. Finally, the relative mRNA expression levels of target genes were normalized to GAPDH and were calculated using the 2 −ΔΔCt method.

Terminal Dexynucleotidyl Transferase-mediated dUTP Nick End Labeling Staining
Sections samples were stained using terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) apoptosis detection kit (KGA7071, Jiangsu KeyGEN Bio-TECH Corp, Jiangsu, China). All the procedures were performed according to the manufacturer's instructions. Briefly, sections were rinsed in 0.3% Triton X-100 for 5 min and subsequently incubated with 50 μL-TUNEL reaction mixture in the dark at RT. Afterward, anti-Neuron (MAB377X, 1:50, Millipore) was added to the sections and incubated overnight at 4 °C. The next day, sections were incubated with the fluorescent secondary antibody Alexa Fluor 594 (A-11005, 1:250, Thermo Fisher Scientific, USA) after being washed with 0.1% PBS (3 × 5 min) for 2 h at RT. Finally, sections were stained with DAPI for 10 min and observed using a fluorescence microscope (IX51, Olympus, Japan).

Transmission Electron Microscope
Rats were sacrificed at 7 days post-surgery, and then a 0.5-cm-long spinal cord sample centered on the injured site was obtained after transcardial perfusion with 0.1M PBS, followed by fixation immediately with 2.5% glutaraldehyde and 1% osmium tetroxide overnight. Subsequently, dissected spinal cords were subjected to gradient alcohol dehydration and then embedded in araldite overnight. Afterward, GGT GGA AGA ATG GGA GTT GCT CTG GAG AAA CCT GCC AAG TATG ultrathin sections (50 nm) were obtained using an ultramicrotome (Leica ultracut UCT) and counterstained with uranyl acetate and lead citrate. Finally, images were captured with a transmission electron microscope (TEM) (HT7800, Hitachi, Japan) to evaluate the intracellular structure in neurons.

Statistical Analysis
Statistical analysis was performed using GraphPad Prism 8.0 Program (Graph Pad Software, San Diego, CA, USA).
All the results were presented as mean ± standard error of the mean (SEM) and are representative of at least three independent experiments. Comparisons among groups using one-way analysis of variance (ANOVA) followed by Tukey's post-hoc Test under the premise of Parametric data (normality and equal variance passed), otherwise using Kruskal-Wallis ANOVA. The BBB rating scores were analyzed using two-factor ANOVA followed by Bonferroni post-hoc for repeated measures. P < 0.05 was considered statistically significant.

OMT Benefited Locomotor Function Recovery after SCI
The protective effect of OMT on locomotor recovery of SCI rats was evaluated using both the average BBB scales at days 0, 1, 3, and 7 and footprint analysis at day 7 following SCI. As shown in Fig. 1C, the BBB scores of rats in the sham group were 21 at all four time points, but declined dramatically at day 1 and thereafter showed evidence of gradual improvements over time in the other treatment groups. Interestingly, OMT-treated rats exhibited markedly higher BBB scores compared with rats in the SCI group at day 3 (P < 0.01). These differences became even more pronounced at day 7 (P < 0.001), suggesting a better functional recovery of injured rats after treatment with OMT. However, this beneficial effect of OMT after SCI was partly abrogated after administration of SIRT1 inhibitor EX527 (P < 0.05) at day 7. In addition, this improvement in functional recovery following OMT treatment was observed in the footprint analysis (Fig. 1D). Rats in the sham group showed fairly coordinated movements of the fore-and hindlimbs without hind limb dragging, while injured animals at day 7 exhibited extensive dragging of hind limbs and inconsistent coordination in hind paw stepping. Conversely, those rats treated with OMT displayed a marked recovery of gait, characterized by clear hindlimb coordination and little toe dragging. Again, the therapeutic effect of OMT on gait was significantly inhibited by co-treatment with EX527. All the above results indicated that OMT contributed to the functional recovery of rats with SCI.

OMT Ameliorated Tissue Damage and Reduced the Loss of Locomotor Neurons after SCI
We next performed HE staining to observe the size of damaged spinal cords at day 7 post-SCI ( Fig. 2A). SCI caused structural damage to the spinal cord, with large and irregular cavities compared with the sham group. In the OMT group, the lesion area was significantly smaller, indicating that OMT could ameliorate tissue damage following SCI.
To determine the effect of OMT on neuronal survival, we utilized Nissl staining to count ventral horn motor neurons around the injured epicenter. Rats in the SCI group had fewer neurons with irregular neuronal structures in spinal cords compared with the sham group (P < 0.001, Fig. 2B, C). After treatment with OMT or co-treatment with EX527, rats exhibited a significantly increased number of surviving neurons (P < 0.001, P < 0.05, Fig. 2B, C) and showed improved neuronal morphology, while the number of surviving neurons was significantly decreased in the OMT-EX527 co-treatment group compared with the OMT-only group (P < 0.05, Fig. 2B, C). This observation suggested that OMT promoted neuronal survival after SCI in rats.

OMT Activates the SIRT1/AMPK Signaling Pathway after SCI
AMPK is a direct downstream target protein of SIRT1 against SCI [31,32]. To investigate the effects of OMT treatment on the SIRT1/AMPK signaling pathway following SCI in rats, we performed Western blotting and qRT-PCR to examine the expression of SIRT1 and phosphorylated AMPK (p-AMPK) at day 7 after SCI. In the protein analysis of band density, there was no significant alteration in the total AMPK expression among groups; however, levels of SIRT1 (P < 0.01, Fig. 3A, B), p-AMPK (P < 0.01, Fig. 3A, C), and p-AMPK/AMPK (P < 0.01, Fig. 3A, D) expression were significantly increased after injury compared with the sham group, suggesting SCI-induced activation of the SIRT1/AMPK signaling pathway. Compared with the SCI group, the protein expression of SIRT1 and p-AMPK, as well as the ratio of p-AMPK/AMPK, was higher in the OMT group (P < 0.05, P < 0.01, P < 0.05, Fig. 3A-D). Importantly, EXT527 treatment resulted in a significantly lower expression of these proteins (all P < 0.01, Fig. 3A-D) compared with the OMT group. qRT-PCR revealed that the mRNA expression levels of SIRT1 and AMPK in the injured spinal cord were increased in the SCI group in comparison to the Sham group (P < 0.05, P < 0.05, Fig. 3E, F). Levels of SIRT1 and p-AMPK were significantly higher in the OMT-treated group compared to the SCI group (P < 0.01, 1 3 P < 0.01, Fig. 3E, F), indicating that OMT treatment could significantly promote the expression of SIRT1 and p-AMPK, while co-treatment with SIRT1 inhibitor EX527 effectively inhibited SIRT1 and p-AMPK expression (P < 0.05, P < 0.001, Fig. 3E, F).

OMT Promotes the Neuronal Autophagy after SCI
To determine the effect of OMT treatment on neuronal autophagy after SCI in rats, we performed Western blotting to assess expression levels of the proteins p62, Beclin-1, and microtubule-associated protein 1 light chain 3B (MAP1LC3B, hereafter referred to LC3B), which are well-established indicators of cellular autophagy [53][54][55] in the injured spinal cords. At day 7 after SCI, levels of p62 (P < 0.001, Fig. 4A, B), Beclin-1 (P < 0.01, Fig. 4A, C), and LC3B (P < 0.01, Fig. 4A, E) expression and the ratio of LC3B-II to LC3B-I (an essential for autophagosome formation) (P < 0.05, Fig. 4A, D) were significantly increased compared with the sham group, indicating induction of autophagy after SCI. It is worthwhile to note that the level of p62 expression was significantly lower, whereas those of Beclin-1, LC3B, and LC3B-II/I ratio were remarkably higher in OMT-treated rats compared with the SCI group (all P < 0.01, Fig. 4A-E), suggesting that OMT treatment elicited enhanced autophagic activity following SCI in rats.
To further confirm these findings, we employed immunofluorescence double staining (Beclin-1/Neuron and LC3B/ Neuron) to evaluate the activity of autophagy in neurons in the ventral horn of the spinal cords around the injured area in rats. Consistent with the results of Western blotting, the number of both Beclin-1-positive neurons (P < 0.01, Fig. 4F, G) and LC3B-positive neurons (P < 0.05, Fig. 4H, I) was significantly higher at day 7 after the injury compared with the sham group. In the OMT-treated group rats, Beclin-1 and LC3B-positive neurons were significantly elevated compared with rats in the SCI group (P < 0.001, P < 0.001) and EX527 group (P < 0.01, P < 0.01) (Fig. 4F-I). Autophagosomes were combined with lysosomes to form autolysosomes for degradation. To examine the fusion of autophagosomes with lysosomes, next, we performed immunofluorescence double staining for lysosomal-associated membrane protein 1 (LAMP1), a lysosome marker [56], with Beclin-1 (Fig. 4J, K) or LC3B (Fig. 4L, M) in neurons at 7 days following SCI. Our results revealed that the co-localization of LAMP1 showed significantly increased numbers of surviving neurons compared to untreated animals with SCI, but the number of surviving neurons was significantly decreased in the OMT-EX527 co-treatment group compared with the OMT-only group. Scale bars = 50 μm. Data are presented mean ± SEM; significant differences among groups are indicated by * P<0.05 and *** P<0.001 with Beclin-1 or LC3B was significantly lower in the sham group, but markedly increased in the SCI group (P < 0.05, P < 0.05). After treatment with OMT or OMT and EX527, the colocalizations of LAMP1/Beclin-1 (P < 0.001, P < 0.001) and LAMP1/LC3B (P < 0.001, P < 0.01) were significantly elevated. These data suggested that neuronal autophagy was significantly increased after SCI, and OMT administration greatly promotes neuronal autophagy following injury via the SIRT1/AMPK signaling pathway.
Based on the fact that the abundance of autophagyrelated proteins was attributed to either an increase in autophagosome for mation or a decrease in Fig. 3 Effect of OMT treatment on SIRT1 and AMPK protein expression after SCI in rats at day 7. (A-D) Representative Western blotting and quantitative results of SIRT1 and p-AMPK after SCI (n = 6). Rats treated with OMT or OMT and EX527 exhibited a significantly increased expression of SIRT1, p-AMPK, and p-AMPK/AMPK ratio after injury, while the levels of SIRT1 and p-AMPK expression, as well as the ratio of p-AMPK/AMPK, were significantly decreased in the OMT-EX527 co-treatment group compared with the OMT-only group. (E, F) qRT-PCR and quantitative analysis of SIRT1 and AMPK mRNAs expression after SCI (n = 8). Rats treated with OMT or OMT and EX527 had a significantly higher expression of SIRT1 and AMPK mRNAs following SCI compared to untreated animals, but the levels of both genes were significantly reduced in the OMT-EX527 co-treated group compared with the OMT-only group. Data are presented mean ± SEM; significant differences among groups are indicated by * P<0.05, ** P<0.01, and *** P<0.001 autophagosome degradation. To further evaluate the effect of OMT on autophagic flux, we intraperitoneally treated injured rats with CQ [49], a well-known autophagy-lysosome inhibitor to block the autophagic flux. At day 7 after SCI, levels of p62 were obviously decreased in the OMT group but significantly increased in the CQ-only group compared with the SCI group (P < 0.05, P < 0.01, Fig. 4N, O). Moreover, we also detected the level of LC3-II and the ratio of LC3B-II to LC3B-I in each group and found both LC3-II (Fig. 4N, Q) expression and the LC3B-II/I ratio (Fig. 4N, P) were significantly increased in the OMT group and were further increased in the CQ group and OMT-CQ co-treatment group compared with the SCI group (all P < 0.001), indicating CQ could suppress downstream activities of the autophagic pathway via inhibition of autophagosome degradation. In addition, compared with the OMT group, those of P62, LC3-II, and LC3B-II/I ratio were significantly increased in OMT-CQ co-treatment group (P < 0.01, P < 0.01, P < 0.05), and the level of p62 expression was significantly decreased with concomitant upregulation of LC3B-II/I ratio in the OMT-CQ co-treatment group compared with the CQ-only group (P < 0.05, P < 0.05) (Fig. 4N-Q). Together, the above data demonstrated OMT could effectively enhance the autophagic flux following SCI in rats.
To locate apoptotic neurons, we performed TUNEL staining with neuronal co-staining to quantify the number of apoptotic neurons in the injured area. After SCI at day 7, the number of TUNEL-positive (apoptotic) neurons was significantly higher than in the sham group (P < 0.001, Fig. 5I, J), indicating the impairment of neurons following injury, but this increase in TUNEL-positive cells could be rescued by OMT treatment (P < 0.001, Fig. 5I, J). Interestingly, co-treatment of OMT with the EX527 group (P < 0.01, Fig. 5I, J) also resulted in fewer TUNEL-positive than untreated SCI. However, the number of TUNEL-positive cells was significantly lower in the OMT-only-treated group compared with the OMT-EX527co-treated group (P < 0.05, Fig. 5I, J). Taken together, the above results demonstrated that OMT substantially suppressed neuronal apoptosis in a rat model of SCI, and the protective effects are partly mediated via activation of the SIRT1/AMPK signaling pathway.

OMT Improves the Neuronal Ultrastructures after SCI
To further confirm the improving effect of OMT on spinal cord neurons at day 7 after SCI, we performed TEM to observe the ultrastructural changes of neurons among groups (Fig. 6). Notably, no detected autophagosomes in the sham group whose rats exhibited intact neuronal intracellular structure as shown by regular nuclei and normal mitochondria with distinct crista. In contrast, in the SCI-group rats, neurons displayed obvious cellular damages characterized by karyotin pyknosis and margination, cytoplasm vacuolization, as well as mitochondrial swelling. In addition, increased autophagosome numbers were observed in the neurons. However, the morphological impairments in neurons of SCI models were significantly alleviated to some extent, and along with remarkably increased autophagosomes when single treatment of OMT or co-treatment of OMT with EXT527. Collectively, these data indicated that the OMT treatment could effectively enhance autophagy and attenuate the structural impairments of neurons following SCI in rats.

Discussion
SCI is well known as a devastating central nervous system trauma characterized by high morbidity and long-term disability, placing a heavy burden on families, healthcare systems, and society overall [62]. OMT is a major quinolizidine alkaloid extracted from the root of Sophora flavescens Ait and has previously been reported to exert anti-inflammatory, anti-apoptotic, and anti-antioxidant functions, with a potential protective effect against ischemia or ischemia/ reperfusion injuries of the brain in rats [63,45,64]. However, the protective effects and molecular mechanisms underlying OMT action following SCI remain unknown. In the present study, we demonstrated that OMT significantly reduced neural tissue damage, inhibited neuronal apoptosis and promoted survival of neurons, and ultimately improved functional recovery in a rat model of compressive SCI. Furthermore, our data also provide evidence that the therapeutic efficiency of OMT involves the enhancement of neuronal autophagy activity mediated by the SIRT1/AMPK signaling pathway, revealing a novel molecular basis for the neuroprotective role of OMT. Taken together, this is the first study to date in that we investigated the link between OMT-mediated neuroprotection and neuronal autophagy and explored the specific molecular mechanisms of the beneficial effects of OMT following SCI in rats. Our data imply that OMT treatment may represent a potential and effective therapeutic strategy against SCI in the future. Locomotor dysfunction is the characteristic hallmark of SCI and is attributed to the damage of locomotor neurons [7]. Apoptosis, a cascade-cleaving process of proteinase, has been recognized to be one main forms of neuronal loss following SCI. Mounting studies have reported that inhibition of apoptosis in neurons resulted in positive effects toward the functional recovery of SCI in rats [65,6,66].
Apoptosis-induced cell death is governed by a large number of genes, including Caspase-3, a key mediator of cell death, which is reported to be involved in the final execution of apoptosis phase and Bax, Bak, Bcl-2, and Bcl-xL (four main members of Bcl-2 family) [57,60]. Caspase-3 is activated following the translocation of the pro-apoptotic protein Bax and Bak from the cytoplasm to the mitochondria, Fig. 5 Effect of OMT treatment on neuronal apoptosis after SCI in rats at day 7. (A-H) Representative Western blotting and quantitative results of Bcl-2, Bcl-xL, Bax, C-caspase-3, and Bak after SCI (n = 6). Rats treated with OMT or OMT and EX527 exhibited a significantly decreased expression of Bax, C-caspase-3, and Bak and significantly increased expression of Bcl-2 and Bcl-xL proteins, as well as the ratio of Bcl-2/Bax after injury compared to the untreated animals. After co-treatment with OMT and EX527, the level of Bax expression was significantly increased, and Bcl-2, Bcl-xL, and Bcl-2/ Bax levels were significantly reduced compared with the OMT-only group. (I, J) Representative immunofluorescence double staining images of TUNEL(green)/Neuron(red), as well as the quantitative analysis of TUNEL-positive neurons after SCI (n = 5). White arrows point to the representative TUNEL-positive neurons. Rats treated with OMT or OMT and EX527 showed a significantly decreased number of TUNEL-positive neurons. The number of TUNEL-positive neurons was significantly increased in the OMT-EX527 cotreated group compared with the OMT-only group. Scale bar is 20 μm. Data are presented mean ± SEM; significant differences among groups are indicated by * P<0.05, ** P<0.01, and *** P<0.001 ultimately leading to nuclear damage and DNA fragmentation [59]. Contrarily, Bcl-2 and Bcl-xL, considered as a key anti-apoptotic factor, can effectively counteract the proapoptotic effects of Bax and Bak, further maintaining cellular homeostasis. Upregulation of the expression of Bcl-2 and Bcl-xL is thought to indicate an inhibition of apoptosis [58,67]. Autophagy, a main intracellular catabolic process for the degradation of cytoplasmic proteins and organelles plays a major role in the occurrence and progression of secondary SCI [20,21]. Accumulating evidence has revealed the importance of autophagy as the same as apoptosis and demonstrated that autophagy occurs prior to apoptosis in the process of neuronal cell death following secondary SCI [5].
Previous studies have reported that autophagy can act as a "double-edged sword," exhibiting both beneficial or detrimental roles for neuronal survival depending on the location and severity of the neuronal injury. In some injury models of the brain [68,69] and spinal cord [70], the enhancement of autophagy results in cell death and contributes to functional outcome deficits. Conversely, autophagy can also effectively suppress apoptosis by clearing damaged organelles, further alleviating neuronal death in response to CNS disorders. For example, after brain injury, Ding et al. [71] and Zhang et al. [72] detected the changes of autophagy-labeled proteins Beclin-1 associated with autophagosome activity [54] and LC3B involved in autophagosome formation [55], whose upregulated expression implied the activation of autophagy, and revealed a protective role of autophagy for promoting neuronal survival and inhibiting neuronal degeneration. Additionally, Li et al. [1,65] also reported that SCI triggered autophagy and enhanced autophagy activity induced by exendin-4 and curcumin could protect neurons from damage and benefit the recovery of neurological functions by inhibiting neuronal apoptosis in a rat model of acute SCI. In the sham-group rats, neurons demonstrated barely autophagosomes, but displayed clear nuclear structure and intact mitochondria with clearly visible crista. On the contrary, in the SCI-group rats, neurons were seriously damaged and exhibited abnormal intracellular structures, including chromatin condensation and karyotin mar-gination, mitochondria swelling and cytoplasm vacuolation, as well as numerous double-membrane structure autophagosomes. However, rats treated with OMT or OMT and EX527 significantly alleviated these morphological impairments of ventral neurons of the spinal cord and enhanced the number of autophagosomes. Scale bar = 1 μm. (N, nucleus; the dovetail arrows indicate karyotin margination; the red arrow points to mitochondria; the black arrow identifies an autophagosome)

3
Consistently, autophagy inhibition resulted in neuronal damage and neurodegeneration in mice [73]. These findings are strongly in agreement with our present results. In this study, we found that the protein expression levels of Beclin-1 and LC3B and the ratio of LC3B-II/I were significantly elevated after SCI and were moreover significantly higher after OMT treatment. Similarly, the number of Beclin-1 and LC3B positive neurons was found to be significantly increased in immunofluorescence staining, which was further dramatically increased in OMT-treated rats after 7 days post-injury. Lysosomes integrate with autophagosomes to form autolysosomes and play a key role in the autophagic process following SCI. Here, we detected the fusion of autophagosomes and lysosomes by immunofluorescence labeled with Beclin-1 or LC3B with LAMP1. Our results indicated that the co-expression of both Beclin-1 and LC3B with LAMP1 was significantly increased after SCI and was further markedly elevated after OMT administration, suggesting OMT could effectively activate autophagy-lysosome pathway. p62, known as autophagic degradation substrate, is widely used as a marker of autophagic flux [53]. Studies have demonstrated that p62 protein interacts with LC3B and involves in the formation and degradation of autophagosomes. Accumulation of p62 protein indicates the blockage of the autophagy degradation, and downregulation of p62 expression represents the enhancement of autophagic flux [53,74]. Here, our study revealed that p62 protein expression was significantly increased at day 7 after SCI and dramatically decreased after treatment with OMT. To further determine the effect of OMT on autophagic flux, CQ, a classical autophagy inhibitor, was coadministered with OMT after injury. Our results found CQ could effectively abolish the favorable effect of OMT on autophagic flux stimulation, suggesting that OMT accelerates activation of a whole autophagic flux following SCI in rats. Similar results were evidenced by increased autophagosomes observed in the OMT-treated group compared to the SCI group by TEM. Together, these data suggested that OMT could effectively enhance SCI-induced neuronal autophagy. Importantly, we also found that OMT significantly reduced the expression of C-caspase-3, Bax, and Bak proteins while increasing the levels of Bcl-2 and Bcl-xL expression, as well as the ratio of Bcl-2/Bax following SCI in rats. TUNEL and Nissl staining confirmed that OMT treatment notably increased the number of surviving neurons after SCI while inhibiting the proportion of apoptotic neurons. In addition, TEM results showed that SCI surgery led to obvious neuronal subcellular structural damage manifested by abnormal nuclei and a proportion of mitochondrial vacuolization, indicating induction of neuronal apoptosis following compression-induced SCI in rats, but treatment with OMT or co-treatment with EX527 could significantly improve the intracellular morphological impairments of neurons and reverse these phenomena to some extent. These data implied that, at least in part, OMT could significantly inhibit SCI-induced neuronal apoptosis which may be associated with its promotion of autophagy activity in rats.
To further clarify the protective role of OMT following SCI in rats, we utilized the BBB scores and HE staining to investigate the locomotor recovery of SCI rats and the morphological changes of spinal cord tissue. Obviously, OMT treatment significantly improved locomotor functions compared with untreated rats subjected to SCI. Moreover, our histopathological results revealed that OMT markedly alleviated tissue damage in the injured spinal cord. The results clearly indicate a positive effect of OMT after SCI in rats.
SIRT1 is known as a conserved nicotinamide adenine dinucleotide (NAD+)-dependent histone deacetylase. Numerous previous studies have demonstrated an active involvement of SIRT1 in cellular metabolism, autophagy, apoptosis, inflammation, oxidative stress, and aging [27][28][29][30]. AMP-activated protein kinase (AMPK) is generally considered as a stress response enzyme, and its activation can induce the promotion of autophagy in various CNS diseases [75 ,6]. Previous studies have revealed a role for AMPK as a direct downstream target protein of SIRT1 after SCI. Zhao et al. [31] have demonstrated that SIRT1 activator resveratrol could target activated AMPK to promote neuronal autophagy, thereby improving the neurological functional recovery following SCI in rats. Moreover, Gao et al. [32] found that melatonin could trigger neuronal autophagy and inhibit apoptosis via activation of the SIRT1/AMPK signaling pathway in rat models of SCI. These findings identified the neuroprotective role of the SIRT1/AMPK signaling pathway in regulating neuronal autophagy and apoptosis following SCI in rats. Interestingly, OMT has been reported to attenuate hippocampal ischemia/reperfusion injury via upregulation of SIRT1 and thereby influencing the processes of autophagy and apoptosis [46]. All of the above evidence prompted us to investigate whether OMT targets the SIRT1/ AMPK signaling pathway to promote autophagy and inhibit apoptosis following SCI in rats. To confirm our hypothesis, we utilized EX527, a specific inhibitor of SIRT1, to investigate the relationship between SCI-induced-enhanced autophagy and the SIRT1/AMPK signaling pathway. Following SCI and subsequently using Western blotting and qRT-PCR to detect protein and gene expression of SIRT1 and p-AMPK, we found that both SIRT1 and p-AMPK were significantly upregulated at 7 days. Interestingly, their expression was further increased after OMT administration, suggesting that OMT further promoted the SIRT1/AMPK signal pathway activation following SCI in rats. In addition, Western blotting or immunofluorescence results revealed changes in representative autophagy biomarkers Beclin-1 and LC3B, as well as apoptosis-related factors Bax, C-caspase-3, and Bcl-2 in neurons. EX527 treatment resulted in significant suppression of autophagy activity following OMT treatment and abrogated the beneficial effects of OMT treatment on suppression of neuronal apoptosis, indicating that OMT promoted autophagy via activation of the SIRT1/ AMPK signaling pathway. However, the neuroprotective effects of OMT were not entirely blocked by EX527, which indicates that there could be other underlying mechanisms involved in the neuroprotection of OMT after SCI.
In conclusion, our present study demonstrates for the first time that OMT significantly inhibits SCI-induced neuronal apoptosis and promotes functional recovery of rat hindlimbs. The protective effects of OMT are related to the enhancement of autophagy in neuronal cells via activation of the SIRT1/AMPK signaling pathway (Fig. 7). Therefore, our results indicate a new potential mechanism which represents another property of OMT in neuroprotection following SCI, and OMT may be considered as a promising therapeutic strategy against SCI in the future. Next, we will conduct a further clinical study to determine the protective roles of OMT against SCI in humans. However, there are several limitations that should be considered in the interpretation of our results. Firstly, while a single dose of OMT (40 mg/kg) was reported to exert protective effects against SCI, it is necessary to clarify the optimal dose and therapeutic window. Secondly, based on the time limitations, we harvested animals at day 7 following SCI, which did not enable us to explore the long-term effects of OMT on SCI. Lastly, as inhibition of autophagy by EX527 only partly prevented the effects of OMT on neuronal apoptosis, the other possible pathophysiological mechanisms will need to be further investigated in vitro in the future.
Author Contribution JL and ZF conceived and designed the study; YC, YW, and GL provided the technical support; JL and LL analyzed the data; XC, JC, and SM assisted with animal care; JL wrote and revised the primary manuscript; JL, YW, ZF, and GL afforded the funding sources. All authors read and approved the final manuscript. Data Availability All data are real and guarantee the validity of results and are available from the authors upon request.

Declarations
Ethics Approval and Consent to Participate Not applicable. Fig. 7 Schematics of a proposed mechanism underlying the neuroprotective effect of OMT against SCI. OMT promotes SCI-induced neuronal autophagy and inhibits neuronal apoptosis via activation of the SIRT1/AMPK signaling pathway in an SCI rat model