Dendrobium nobile was bought from the Chinese planting base in Chishui (Guizhou, China). The polysaccharide was isolated using water extraction and alcohol precipitation, and purified to 98.1% by a DEAE Sepharose Fast Flow (57407-08-6, Bio-Resin, Beijing, China) column.

Specific pathogen Free (SPF)-grade healthy female Sprague Dawley (SD, purchased from Hunan SJA Laboratory Animal Company, Changsha, Hunan, China) rats, aged 12–14 weeks and weighing 220–250 g, were employed in our experiment. The rats were housed by the Animal Center of Gannan Medical University. Animal use and care protocols conformed to the principle of the National Institutes of Health (NIH) of China and were approved by the Ethics Committee of the first affiliated hospital of Gannan Medical University (No. GYYFY2022-10).

The experiment design is shown in Fig. 1. The experimental rats were divided into three groups: the sham operation group, the SCI group, and the DNP group. In the sham group, the lamina was simply cut off, but the spinal cord was not damaged. A modified Allen’s SCI method at T10–T11 was used for the SCI and DNP groups [19]. Following injury, the DNP group was intragastrically administered 100 mg/Kg DNP (purity 98%, Chengdu Alfa Biotechnology Co., Ltd., Chengdu, China) once a day for 14 consecutive days, and the other two groups were intragastrically administered equal amounts of normal saline.

Fig. 1.

Schematic design of the experiment. SCI, spinal cord injury; DNP, dendrobium nobile polysaccharide; GSH, glutathione; Gpx4, glutathione peroxidase 4; GRSF1, G-rich RNA sequence binding factor 1; BBB, Basso-Beattie-Bresnahan; xCT, Xc${}^{-}$ light chain; i. g., oral gavage.

The BBB scale, to observe the walking, trunk movement, and coordination of the hip, knee, and ankle joints by rats crawling, and footprint analysis were used to rate the recovery of hind limb motor function, scoring from 0 to 21. Hind limb motor function was valued at 1, 3, 7, 14, 21, and 28 days after SCI.

In the footprint analysis test, rats’ hind paws were stained with red ink, and their forepaws were stained with black ink. The rats then walked through a narrow tunnel of 60 cm in length and 7.5 cm wide. The opposite end of the tunnel was illuminated to guide the movement of the rats. The bottom of the tunnel was covered with white test paper. The evaluation indicator was clarity of footprint.

The injured segmental spinal cord was sampled at 12, 24 and 48 hours after SCI. Spinal cord tissue was homogenized mechanically to obtain supernatant under ice water bath conditions. Iron content was detected using a tissue iron determination kit (A039-2-1; Nanjing Jiancheng, Nanjing, Jiangsu, China), according to the manufacturer’s instructions, and GSH content was determined using a glutathione assay kit (A006-2-1, Nanjing Jiancheng), with the bicinchoninic acid (BCA) method, according to the manufacturer’s instructions.

At 24 h after SCI, the injured spinal cord tissue was quickly removed and made into 1 mm $\times$ 1 mm $\times$ 1 mm blocks. Tissue was fixed with 2.5% glutaraldehyde in a 0.1 M phosphate buffer (pH 7.4) for 2 h at room temperature, and then treated with 1% samarium tetroxide at 0.1 M phosphate buffer (PB) [20]. Photographs were then taken using transmission electron microscopy (TEM) (7600TEM, Hitachi, Tokyo, Japan).

Spinal cord tissue was dissolved in radio-immunoprecipitation assay (RIPA) lysis buffer. The protein level was measured using a bicinchoninic acid assay (BCA) protein analysis kit (Beyotime, Shanghai, China). Fifty micrograms of protein were separated on a 10% or 12% sodium dodecyl sulfate polyacrylamide gel and transferred to a polyvinylidene fluoride (PVDF) membrane. After being incubated with 5% albumin from bovine serum (BSA) for 1 h, the membrane was incubated overnight on a shaker with primary antibody. The primary antibodies included rabbit anti-xCT (diluted at 1:1000, Boster, Wuhan, Hubei, China), mouse anti-Gpx4 (diluted at 1:2000, Proteintech, Shanghai, China), rabbit anti-GRSF1 (diluted at 1:1000, Abcam, Boston, MA, USA), and mouse anti-GAPDH (diluted at 1:1000, Beyotime, Wuhan, Hubei, China). After washing with tris buffered saline + tween (TBST) three times, the secondary antibody (A0216, Beyotime, Shanghai, China ) was coupled with a 1:10,000 dilution of horseradish peroxidase and incubated at room temperature for 1 h. Finally, a Chemi Dox XRS chemiluminescence imaging system (Bio-Rad, San Francisco, CA, USA) was used to visualize the signal. All tests were replicated three times.

Spinal cord tissue specimens were removed from the freezer at –80 °C. cDNA (2 µL) was quantified according to the manufacturer’s instructions. Data are shown using 2${}^{-\mathrm{\Delta }\mathrm{\Delta }\text{Ct}}$ values [21]. Primer design, synthesis, and sequences of target genes were derived from the database (https://www.ncbi.nlm.nih.gov/gene/?term=), and carried out by GENERAL Biosystems (Anhui) Corporation Limited (Chuzhou, Anhui, China) (Table 1).

Rats were sacrificed by intraperitoneal injection of 10% sodium pentobarbital (3.5 mL/kg, 57-33-0, Damao chemical reagent factory of Tianjin, Tianjin, China). The heart was then infused with saline and 4% formaldehyde. After perfusion, the injured central spinal cord segment (0.5 cm long) was taken and immediately stored in 4% formaldehyde. The spinal cord, soaked in formaldehyde, was dehydrated and embedded; it was then sectioned horizontally into 4 µm slices [22]. For immunofluorescence experiments, the preparation process of tissue specimens was the same as described above. The tissue was then stained according to the hematoxylin–eosin (HE) staining kit instructions (Solarbio, Beijing, China).

Paraffin sections of spinal cord specimens were prepared and 0.1% BSA was added to cover the specimens completely. The slices were placed in a wet box at room temperature for 15 minutes. The primary antibody (NeuN, Proteintech, 1:100) was diluted with phosphate buffered saline (PBS) solution, dropped onto the slices, and then placed in a refrigerator at 4 °C overnight. The slices were removed and washed with PBS. The secondary antibody was dropped onto the sections of the spinal cord specimen, and the slices were left standing at room temperature for 1 hour (in the dark), and then were removed and washed with PBS. Drops of 4${}^{\prime }$,6-Diamidino-2-Phenylindole (DAPI) were added until the specimen was completely covered. Finally, drops of anti-fluorescence quenching agent were added and the specimen was sealed with cover glass. Photos were captured under a fluorescence microscope (DM2500, Leica, Berlin, Germany).

Statistical analyses were carried out using SPSS 20.0 statistical software (SPSS Inc., Chicago, IL, USA). Data are represented as the mean $\pm$ standard error (SE). Repeated measures analysis of variance (ANOVA) was performed for comparisons between two groups and multiple groups. p 0.05 was deemed statistically significant.

The rats were evaluated at 1, 7, 14, 21, and 28 days post-injury. On day 1, the mean BBB scores of the sham group were 21 points, those of the SCI group were 1 point, and those of the DNP group were 1.5 points, but the differences were not statistically significant between the SCI and the DNP groups (p $>$ 0.05). At 7, 14, 21, and 28 days, the scores of the DNP group were markedly increased compared with the SCI group (p 0.05). Although the BBB scores were increased as time progressed, the scores in the DNP and SCI groups were decreased compared with the sham group (p 0.05) (Fig. 2A).

Fig. 2.

The effect of DNP on hind limb function of SCI rats. (A) BBB scores of different groups at 1, 7, 14, 21, 28 d post-SCI (values are expressed as the mean $\pm$ SE. n = 3, ${}^{\text{ns}}$p $>$ 0.05, *p 0.05, **p 0.01, ***p 0.001, ****p 0.0001 vs the DNP group). (B) Footprint of three groups at 28 d post-injury. SE, standard error; ns, not significant.

The footprint test showed that rats in the sham group had clear footprints and no drag on the hind limbs. In the SCI group, the drag of the hind limbs was obvious, and footprints were not distinguishable. The footprints in the DNP group were clearer than those in the SCI group, but there were still drag marks (Fig. 2B).

Iron content was increased significantly after SCI, but gradually decreased over time. After DNP treatment, the iron content decreased at 12 h post-injury, but there were no marked differences compared with the SCI group (p $>$ 0.05). However, the iron content decreased significantly compared with the SCI group at 24 h and 48 h post-injury (p 0.05) (Fig. 3A).

Fig. 3.

Effects of DNP on iron content and mitochondrial morphology in injured spinal cord. (A) Iron content at 12, 24 and 48 hours after SCI (values are expressed as the mean $\pm$ SE, n = 3. ${}^{\text{ns}}$p $>$ 0.05, *p 0.05, **p 0.01, ***p 0.001 vs the SCI group). (B) Mitochondrial form in the spinal cord tissue at 24 h after SCI (Scale bar = 200 nm). red arrow means: Outstanding performance.

We observed ultrastructural changes in mitochondria in injured spinal cord using TEM within 24 h after SCI. The mitochondrial crest and membrane were intact in the sham group. In the SCI group, the mitochondrial membrane was disrupted and the crest almost vanished. Compared with the SCI group, the mitochondrial crest and membrane in the DNP group were improved (Fig. 3B).

We measured the mRNA and protein expression of xCT, Gpx4, GRSF1, and GSH after SCI. Compared with the sham group, the xCT, Gpx4, and GSH levels in the SCI group were markedly reduced at 14 d and 28 d after SCI (p 0.05). After DNP treatment, the mRNA and protein expression of xCT, Gpx4, and GSH were upregulated, suggesting that DNP could prevent ferroptosis by upregulating the expression of GSH, xCT, and Gpx4. In addition, the level of mRNA and protein expression of xCT and Gpx4 in the DNP group at 28 d were significantly increased compared with that at 14 d (p 0.05), suggesting that the inhibition of ferroptosis in SCI by DNP continued with the DNP treatment duration (Figs. 4,5). Meanwhile, the mRNA and protein level trends of GRSF1 and Gpx4 were the same compared with each other. Since GRSF1 and Gpx4 interact with each other and GRSF1 is the upstream protein of Gpx4, we believed that DNP increased the expression of Gpx4 by regulating GRSF1.

Fig. 4.

Effects of DNP on GSH content and level of xCT mRNA, Gpx4 mRNA, and GRSF1 mRNA at 14 d and 28 d after SCI. (A–D) Effect of DNP on GSH content (A), xCT mRNA (B), Gpx4 mRNA (C), and GRSF1 mRNA (D) in injured spinal cord (values are expressed as the mean $\pm$ SE, n = 3. *p 0.05, **p 0.01, ***p 0.001 vs the SCI group. ${}^{\text{ns}}$p $>$ 0.05, ${}^{\mathrm{\#}}$p 0.05 vs the DNP group).

Fig. 5.

Effects of DNP on xCT, Gpx4, and GRSF1 protein level at 14 and 28 days after SCI. (A) Levels of xCT, Gpx4, and GRSF1 at 14 and 28 days after SCI by western blot. (B–D) Levels of xCT, Gpx4, and GRSF1 proteins were quantitatively detected (values are expressed as the mean $\pm$ SE, n = 3. *p 0.05, **p 0.01, ***p 0.001 vs the SCI group. ${}^{\mathrm{\#}}$p 0.05 vs the DNP group).

Compared with the sham group, the damaged spinal cord transverse sections showed obvious destructive cavities. The mean value of cavity area was 5.998 um${}^2$ at 14 days and 5.448 um${}^2$ at 28 days in the SCI group, and 4.997 um${}^2$ at 14 days and 4.180 um${}^2$ at 28 days in the DNP group. Compared with the SCI group, the mean tissue cavity area was markedly improved in the DNP group (14 d SCI vs 14 d DNP: p = 0.0257; 28 d SCI vs 28 d DNP: p = 0.0060). At 28 days, the cavities of the spinal cord tissue in the DNP group were smaller than at 14 days (p = 0.0086), and the tissue damage was relatively mild. These results suggest that DNP treatment can repair injured spinal cord tissue, and the injured spinal cord tissue recovered to a greater extent with the prolongation of DNP treatment duration (Fig. 6).

Fig. 6.

Effect of DNP on histopathological changes of injured spinal cord. (A) HE staining images of the sham, SCI, and DNP groups at 14 d and 28 d of SCI (image magnification: 4$\times$, 200$\times$). (B) Quantification of the cavity areas (values are displayed as the mean $\pm$ SE, n = 3. *p 0.05, **p 0.01 vs the SCI group. ${}^{\mathrm{\#}}$p 0.05 vs the DNP group). HE, hematoxylin–eosin.

Neurons are the most vital cells in the central nervous system, and NeuN is a specific marker of neurons. The amount of NeuN${}^{+}$ positive cells was assessed by immunofluorescence assay. Compared with the sham group, the number of NeuN${}^{+}$ cells was markedly reduced in the SCI group, and DNP treatment increased the amount of NeuN${}^{+}$ cells at 14 and 28 days after SCI. Furthermore, the number of NeuN${}^{+}$ cells at 28 d was significantly greater than that at 14 d in the DNP group (p 0.05). These results demonstrated that the therapeutic effect of DNP continued with the treatment duration of DNP (Fig. 7).

Fig. 7.

Effects of DNP on the number of NeuN${}^{\mathrm{+}}$ cells in injured spinal cord. (A) NeuN immunofluorescence staining of the injured area at 14 and 28 days after SCI, Scale bar = 50 µm. (B) Number of NeuN${}^{+}$ cells (values are displayed as the mean $\pm$ SE, n = 3. **p 0.01.