Lycium barbarum glycopeptide alleviates neuroinflammation in spinal cord injury via modulating docosahexaenoic acid to inhibiting MAPKs/NF-kB and pyroptosis pathways

In total, 70 (8-week-old) male Sprague‒Dawley rats (300–350 g) were purchased from the Laboratory Animal Center of Ningxia Medical University. After surgery, all the rats were housed under continuous temperature (21 ± 3 °C) and humidity (50% ± 5%) conditions with a 12 h light/dark cycle. The experimental protocol was approved by the Laboratory Animal Ethical and Welfare Committee of the Laboratory Animal Center of Ningxia Medical University.

The twelfth thoracic vertebral body (T12) hemisection model was established as previously described [16]. To establish the SCI model, rats were anesthetized with isoflurane (2–4% induction, 1.5% maintenance), and their body temperature was maintained at 35 ± 1 °C using a mini heating pad. The skin of the upper thoracic area was exposed after shaving and cleaning with betadine solution. To expose the T12, the fascia and muscle were bluntly dissected to avoid spinal dural injury during the incision of the dorsal lamina. A needle (26G) was bent 90° to establish a lateral hemisection, and the spinal cord was punctured dorsoventrally at the midline by inserting the needle 5 mm while avoiding damage to the dorsal spinal artery. The left half of the spinal cord was pulled and cut, and this step was repeated three times to ensure completeness of the spinal cord hemisection (Additional file 1: Fig. S2A, B). For pain management, carprofen (4 mg/kg; Rimadyl®, Zoetis, Florham Park, NJ, USA) and buprenorphine (0.2 mg/mL; Temgesic®, Intervet International, Boxmeer, the Netherlands) were administered continuously before surgery and every 8 h for three successive days after SCI.

Forty-five adult male rats were randomly assigned to three groups (15 rats per group): sham group, SCI group and post-SCI LbGp treatment group. LbGp was dissolved in 0.01 M phosphate-buffered saline at room temperature, and the solution was administered to the rats in the post-SCI LbGp treatment group via a nasogastric tube (50 mg/kg) once daily until day 28 post-SCI. The rats in the sham and SCI groups were administered 0.01 M phosphate-buffered saline 24 h after SCI.

We next evaluated the hind limb motor function of rats after SCI. Three rats were selected from each group, and the Basso-Beattie-Bresnahan (BBB) test was used to score the motor function of the hind limbs. The BBB test is divided into 22 grades, with 21 points indicating normal paralysis and 0 points indicating paralysis. The scoring was performed at 8 p.m. in a double-blind manner. Each rat was measured three times, and the average value was recorded. The more detail of BBB Score as previously describe: the rats were placed on a circular platform with a diameter of 2 m. The walking and limb activity scores of the hindlimbs were observed and recorded. In the first stage (0–7 points), the joint activity of the hindlimbs of the animals was scored. In the second stage (8–13 points), the gait and the coordination of the hindlimbs were scored. In the third stage (14–21 points), the fine movements of the claws were judged. The scores for the three stages together totalled 21 points. Each group was scored 1 day before surgery and 1, 3, 7, 14, and 28 days post-injury (dpi) [17].

The rat microglia (RM) cell line (BNCC 360237) was purchased from the BeNa culture collection (Henan, China). First, RM cells were seeded in six-well plates (1 × 10^–5 cells/well), activated with LPS (1 µg/ml L3129 Sigma) and ATP (5 mM; Sigma) for 4 h, and then treated with LbGp or DHA for 24 h to establish the LbGp treatment group or DHA treatment group, respectively. In addition, microglia activated with ATP + LPS were incubated with the FADS2 enzyme inhibitor SC26196 for 12 h and then incubated with LbGp for 24 h to construct the ATP + LPS + SC26196 + LbGp treatment group. Spinal cord tissue was harvested on day 7 post-SCI. Tissue or cell lysates were prepared using the Keygen protein extraction kit (KGP250 Keygen BioTECH China), and the protein concentration was measured with the BCA protein assay kit (KGP902 Keygen BioTECH China). Equal amounts of protein were separated with 10 or 15% sodium dodecyl sulfate‒polyacrylamide gels, transferred to polyvinylidene fluoride membranes, and blocked with 5% skim milk. Subsequently, the membranes were probed with the following primary antibodies at 4 °C overnight: anti-brain-derived neurotrophic factor (BDNF) (1:10,000) (Cat. Number: ab108319; Abcam UK), anti-glia-derived neurotrophic factor (GDNF) (1:2000) (Cat. Number: A14639; ABclonal China), anti-P-p-38 (1:1000) (Cat. Number: 28796–1-AP; Proteintech, China), anti-p38 (1:1000) (Cat. Number: 14064–1-AP; Proteintech, China), anti-P-JNK (1:1000) (Cat. Number: 80024–1-RR; Proteintech, China), anti-JNK (1:1000) (Cat. Number: 24164–1-AP; Proteintech, China), anti-P-p65 (1:1000) (Cat. Number: ab76302; Abcam, Cambridge, UK), anti-p65 (1:20,000) (Cat. Number: 80979–1-RR; Proteintech, China), anti-NLRP3 (1:2000) (Cat. Number: 27458–1-AP; Proteintech, China), anti-ASC (1:3000) (Cat. Number: 10500–1-AP; Proteintech, China), anti-caspase-1 p45 (1:1000) (Cat. Number: ab179515; Abcam UK), anti-caspase-1 p20 (1:2000) (Cat. Number: bs-10442R; Bioss, China), anti-GSDMD and anti-GSDMD-N (1:1000) (Cat. Number: ab219800; Abcam, UK), anti-pro-IL-18 and anti-IL-18 (1:10,000) (Cat. Number: 10663–1-AP; Proteintech, China), anti-IL-1β (1:1000) (Cat. Number: ab254360; Abcam, UK), and anti-β-actin (1:40,000) (Cat. Number: 81115–1-RR; Proteintech China). The membrane was washed with TBST, incubated with secondary antibody (1:20,000) (Cat. Number: ab6721 Abcam UK) for 1 h at room temperature, washed again, visualized with ECL reagent (SW134-01 Seven Biotech, China), and quantitatively analyzed using ImageJ software (National Institutes of Health, Bethesda, MD, USA).

The Cell Counting Kit-8 (CCK-8) assay (Dojindo Laboratories Lot. CK04 Japan) was used to assess the toxicity of LbGp toward microglia. Microglia were plated in 96-well plates at 7 × 10^–3 cells/well, and different concentrations of LbGp were added (100, 200, and 400 µg/ml). The microglia were then incubated at 37 °C in a 5% CO₂ incubator for 24 h. The supernatant was discarded, and 100 µl of medium containing 10% CCK-8 reagent was added to each well. The plate was then incubated at 37 °C in a 5% CO₂ incubator for 2 h in the dark. The absorbance of each well was detected with microplate reader at a wavelength of 450 nm.

The level of DHA in each group was detected by COIBO BIO (Shanghai, China). RM cells were seeded in six-well plates (1 × 10^–5 cells/well), activated with LPS (1 µg/ml; L3129, Sigma) and ATP (5 mM; Sigma) for 4 h, and treated with LbGp for 24 h. The cells were centrifuged at 4 °C and 800g for 10 min, and 200 µl of supernatant was then collected for DHA detection. DHA assays were performed according to the protocol provided by the COIBO BIO (CB14901-Ra, Shanghai, China).

Microglia were counted and plated into a six-well plate at a density of 1 × 10^–5 cells/well. Upon reaching 80% confluency, the cells were incubated with 1 µg/ml LPS for 12 h and then treated with different concentrations of LbGp (100, 200, and 400 µg/ml) for 24 h. After three washes with cold phosphate-buffered saline (PBS), 1 ml of TRIzol was added, and the mixture was pipetted into a 1.5-ml EP tube. After the addition of 200 µl of chloroform, the prepared samples were centrifuged at 9000g at 4 °C for 15 min. The supernatant was aspirated, and an equal volume of isopropanol was added. The mixture was then centrifuged at 9000g at 4 °C for 15 min. The supernatant was removed, and 75% ethanol was added. After two rounds of centrifugation at 6000g for 5 min at 4 °C, double-distilled water (ddH2O) was added, and the RNA concentration was measured. The reverse transcription kit and real-time quantitative PCR kit were purchased from Vazyme (lot. R302-01, Q411-02; Nan Jing, China), and the assays were performed according to the manufacturer’s instructions. The gene-specific sequences of the primers used are shown in Additional file 2: Table S1. Relative gene expression was measured using the 2^{− △△ct} method.

Rat spinal cord tissue was collected on day 7 post-SCI and subjected to fixation, paraffinization, dewaxing, dehydration, antigen extraction, and blocking according to the standard protocol. Subsequently, tissue sections were incubated with the following primary antibodies for 24 h: anti-NLRP3, anti-ASC, anti-caspase-1, and anti-NeuN (1:2000) (Cat. Number: 6975-1-AP; Proteintech, China). Next, tissue sections were incubated with fluorescent-labeled secondary antibody colored green (1:500) (ab150077; Abcam UK), red (1:200 SA00013-4; Proteintech, Wuhan, China), rose red (1:200 GB21303 Servicebio China), pink (1:200 GB1232 Servicebio China) for 3 h, and the nuclei were stained with DAPI. Images were scanned using an ortho-fluorescence microscope (Nikon Eclipse C1) and a fluorescence microscope (Olympus, Japan). Quantitative analysis was performed using ImageJ.

One week post-SCI, the rats in the three groups (A: sham group, B: SCI group, C: LbGp treatment groups; n = six rats/group) were anesthetized under excessive isoflurane, and the dorsal skin surface was incised. The vertebrae were cut, and the injured area of the spinal cord tissue was harvested (length of approximately 50 mm in length, weight of approximately 30 mg). The harvested spinal tissue was rapidly placed in liquid nitrogen and transferred to a − 80 °C refrigerator for metabolomic sequencing analysis.

Metabolomics was performed by Shanghai Luming Biotech Co., Ltd. All reagents were of high-performance liquid chromatography (HPLC) grade. L-2-chlorophenylalanine was purchased from Shanghai Heng Chuang Biotechnology (Shanghai, China), and methoxyamine hydrochloride (97%), pyridine, n-hexane, and BSTFA with 1% TMCS were purchased from CNW Technologies GmbH (Düsseldorf, Germany). Chloroform was obtained from Titan Chemical Reagent (Shanghai, China), and water and methanol were obtained from Thermo Fisher Scientific (Waltham, MA, USA).

GC‒MS was performed using a Db-5MS capillary column (30 × 0.25 mm × 0.25 µm; Agilent J&W Folsom, CA, USA), high-purity helium (purity ≥ 99.999%) as the carrier gas, a flow rate of 1.0 ml/min, and an injection port temperature of 260 °C. The injection volume was 1 µl, and the injection was split with a solvent delay of 5 min. The program temperatures were as follows: the temperature of the column oven was initially set to 60 °C and maintained at 60 °C for 0.5 min; the program temperature was then increased to 125 °C at 8 °C/min, to 210 °C at 8 °C/min, to 270 °C at 15 °C/min and to 305 °C at 20 °C/min and maintained at 305 °C for 5 min.

The electron bombardment ion source had the following parameters: ion source temperature, 230 °C; quadrupole temperature, 150 °C; and electron energy, 70 eV. The scanning mode was full scan mode (SCAN) with a mass scanning range of 50–500 m/z.

The collected GS/MS raw data were converted into abf format using Analysis Base File Convert for rapid data retrieval. The data were then imported into MS-DIAL software for characterization, MS2Dec deconvolution, peak alignment, peak identification, peak detection, wave filtering, and missing value interpolation. Metabolites were characterized based on the LUG database. The data matrix included the signal intensity, retention index, retention time, mass-to-charge ratio, sample information, and peak name for each species. After screening, all peak signal intensities were segmented and normalized for each sample according to an internal criterion of RSD ˃ 0.3. After data normalization, redundancy removal and peak merging were performed to obtain a data matrix.

The matrix was imported into R for principal component analysis to observe the overall distribution between samples and the stability of the entire analysis process. The differential metabolites between groups were identified by using orthogonal partial least squares discriminant analysis (PLS-DA). The model quality was assessed by sevenfold cross-validation and a 200-response permutation test.

The projected importance values of the variables in the OPLS-DA model were used to rank the overall contribution of each variable to the group discrimination. Two-tailed Student’s t test was used to verify whether the difference in metabolites between groups was significant. Differential metabolites that met the following criteria were selected: variable importance of projection value ˃ 1.0 and p value < 0.05.

LbGp was purified from LBP and contains 368 metabolites (Additional file 1: Fig. S1A). LBP is known to induce nerve regeneration and promote motor function recovery [18]. To test our conjecture, we established a rat spinal cord hemisection model, orally administered LbGp after the operation, and assessed the motor function of the different treatment groups up to 28 days after surgery using the BBB motor function score. Based on the BBB test results, the score of the SCI group was significantly lower than that of the sham group, and the score of the LbGp group did not significantly differ from that of the SCI group during the first week post-SCI but was higher than that of the SCI group during the second week (Fig. 1A). These results revealed that LbGp treatment could improve the motor function of rats. On day 14, the BBB motor function score of the LbGp intervention group was higher than that of the SCI group.

Previous studies have shown that BDNF and GDNF play an important role in promoting nerve regeneration [19]. To further elucidate the potential role of LbGp in nerve regeneration, we detected the expression levels of BDNF and GDNF in SCI by western blotting and that of NeuN protein by immunofluorescence. Because the inflammatory microenvironment was altered prior to behavioral changes in the animals, we further investigated the mechanism of motor function recovery in rats by examining the levels of brain-derived neurotrophic factor (BDNF) and glia-derived neurotrophic factor (GDNF) in the spinal cord tissues of rats on day 7. On day 7, the protein expression levels of BDNF and GDNF were significantly higher in LbGp-treated rat spinal cord tissues than in the tissues of the SCI group, as determined by western blotting (Fig. 1B–D; Additional file 1: Fig. S3A, B). On day 7, the expression of NeuN protein in spinal cord tissue was consistent with the above-described results, as determined by tissue immunofluorescence experiments (Fig. 1E, F). Therefore, we hypothesize that LbGp could promote neuronal repair and improve the motor function of rats after SCI by modulating BDNF and GDNF expression.

The role of LbGp in inhibiting neuroinflammation has been demonstrated [20]. To further confirm the role of LbGp in neuroinflammation, we administered different concentrations of LbGp after SCI by gavage and identified the optimal concentration based on the therapeutic effect. The inhibitory effect of different concentrations of LbGp on neuroinflammation was assessed by western blotting. Rats were treated with different concentrations (10, 50, and 100 mg/kg) of LbGp after SCI, and the inflammatory factor pro-IL-18 in the different treatment groups was detected. We found that LbGp induced superior inhibition of the inflammatory factory Pro-IL-18 at 50 mg/kg (Fig. 2A) (Additional file 1: Fig. S3C). The accumulated body of evidence shows that LBP can inhibit the expression of the NLRP3 inflammasome and ROS production to suppress inflammation [21]. To further elucidate the mechanism of the role of LbGp in SCI-induced neuroinflammation, we detected the levels of NLRP3, ASC, and caspase-1 by immunofluorescence and western blotting. The expression levels of NLRP3, ASC, and caspase-1 proteins determined by tissue immunofluorescence were significantly lower in the LbGp group than in the SCI group (Fig. 2B–D). Based on the western blot analysis, the rats in the SCI group exhibited elevated expression of P-p-65, NLRP3, ASC, caspase-1p45, GSDMD, GSDMD-N, IL-1β and IL-18 proteins in the spinal cord on day 7 post-SCI, and these expression levels were significantly reduced in the LbGp group (Fig. 2E–M) (Additional file 1: Fig. S3D, E). Taken together, these data indicate that the oral administration of LbGp after SCI can suppress neuroinflammation.

As mentioned above, LbGp can inhibit the NF-kB and pyroptosis pathways to suppress neuroinflammation in vivo, but the target cells on which it acts are unclear. Microglia are intrinsic immune cells in the CNS that are involved in the immune response in vivo and play an essential role in neuroinflammation [22]. To determine the effect of LbGp on microglia, cell viability was examined by a CCK-8 assay. The results indicated no significant changes in cell viability after incubation with 100–400 µg/ml LbGp (Additional file 1: Fig. S4A). To further assess the optimal drug concentration for the LbGp-mediated inhibition of inflammation triggered by microglia, we incubated activated microglia with LbGp at different concentrations (100, 200 and 400 µg/ml) for 24 h and assayed the mRNA expression levels of the inflammatory factor IL-18. The RNA level of IL-18 was reduced in the LbGp groups (100, 200 and 400 µg/ml), and no significant difference was found among the LbGp groups (Fig. 3A). Based on the abovementioned results, we administered 100 µg/ml LbGp after microglial activation. As previously described, the administration of LbGp treatment after the activation of microglia resulted in lower expression of P-p38, P-JNK, P-p65, NLRP3, ASC, caspase-1p20, GSDMD, GSDMD-N, IL-18, and IL-1β proteins in the LbGp-treated group compared with the ATP + LPS group (Fig. 3B–K). Similarly, in vitro experiments showed that microglial activation followed by LbGp intervention can suppress neuroinflammation.

As described previously, LbGp inhibited neuroinflammation both in vivo and in vitro, and to further clarify the mechanism of LbGp treatment for SCI by metabolomic analysis, 66 and 38 metabolites were found to be elevated in cell supernatants and tissues, respectively, in the LbGp-treated group (Additional file 1: Fig. S5A–E). Pairwise comparisons revealed that both ethanolamine and DHA were increased in tissue and cell supernatants after LbGp intervention (Fig. 4C). At present, the role of ethanolamine in neuroinflammation is unclear [23‐25]. The view that DHA suppresses inflammation by inhibiting the NIRP3 protein in activated macrophages has been well confirmed [26‐28]. DHA intervention after SCI can inhibit neuroinflammation in rats [29]. Therefore, we hypothesized that LbGp may suppress neuroinflammation by regulating DHA production by microglia. To test this hypothesis, we administered SC-26196, an inhibitor of FADS2, a key enzyme for DHA production, before LbGp intervention, and the results showed that the inhibitory effect of LbGp on inflammation was significantly weakened after reduced DHA production (Fig. 3B–L). Collectively, our data indicate that LbGp regulates the production of DHA by microglia and thereby suppresses neuroinflammation.

To determine whether LbGp can induce DHA secretion from microglia, we analyzed cell supernatants by ELISA and found that the level of DHA was significantly elevated in the LbGp intervention group (Fig. 4D). Metabolomic analysis showed that microglia are rich in DHA, providing the possibility of DHA secretion [30]. FADS1 and FADS2 play an essential role in the process of DHA synthesis [31]. Q-PCR results showed that LbGp intervention significantly contributed to high mRNA expression of FADS1 and FADS2, key enzymes for DHA production (Fig. 4E, F). As mentioned previously, our data demonstrate that LbGp promotes high expression of the key enzymes FADS1 and FADS2 in microglia and thereby stimulates DHA secretion from microglia.

A previous study suggested that DHA can inhibit microglia-induced neuroinflammation [32]. Therefore, we further verified its role in the MAPK/NF-kB and pyroptosis pathways by Western blotting. As previously described, the administration of DHA after the activation of microglia resulted in lower expression of P-p38, P-JNK, P-p65, NLRP3, ASC, caspase-1p20, GSDMD, GSDMD-N, IL-18, and IL-1β proteins in the DHA-treated group (Fig. 5A–K). Our in vitro results validate that DHA has an inhibitory function on neuroinflammation.