Adult Gekko japonicus were used as animal models, as Dong et al. [22]. Geckos were fed with water and mealworms and housed in the room with saturated humidity and controlled temperature (22–25 ${}^{\circ }$C). Amputation was performed at the sixth caudal vertebra, which was described previously [24]. Briefly, it was established by using nylon thread and pulling gently until the tail was detached.

All experimental protocols were approved by the Animal Care and Use Committee of Nantong University and the Jiangsu Province Animal Care Ethics Committee.

To obtain the full length of gecko SOCS1, both a sense primer 5’- CCT GGG AGC GGA AGG TGT GGA AGT G -3’ and an anti-sense primer 5’- CCG CGT CCC CCT GAA TCC GGT TTT A -3’ were designed based on the genome sequences [25]. Both 5’- and 3’-RACE were performed using BD SMART RACE cDNA Amplification Kit (Clontech, Mountain View, CA, USA). The obtained protein was compared against the protein database in GenBank using the Protein-BLAST network server at NCBI [26]. Alignment of multiple protein sequences was performed by using the MegAlign program with the CLUSTAL method [27]. A phylogenetic tree was constructed using the Maximum-Likelihood of PHYML, with the GTR (CDS sequences) and JTT (protein sequences) substitution model [28].

SOCS1-overexpression adenovirus (GV314-SOCS1) was produced by Genechem Co. Ltd. (Shanghai, China). The ORF of SOCS1 was cloned into a GV314 vector via the AgeI and Bam HI sites. The EF-1$\alpha$ promoter drove the expression of SOCS1, and the eGFP expression was driven by CMV promoter. Both sequences were incorporated into an adenovirus produced in 293T cells with 2 $\times$ 10${}^{10}$ PFU/mL of virus titer.

Gsn3, a cell line of gecko oligodendrocytes was cultured in Dulbecco’s Modified Eagles Medium (DMEM, Gibco BRL) supplemented with 10% fetal bovine serum in a 30 ${}^{\circ }$C cell incubator with 5% CO${}_2$.

For IFN-$\gamma$-induced oligodendrocyte apoptosis, Gsn3 was treated with 0–250 ng/mL IFN-$\gamma$ and/or 10-50 ng/mL TNF-$\alpha$ for 24–72 h (R&D Systems). For glucose deprivation-induced oligodendrocyte apoptosis, GV314-vector or GV314-SOCS1 adenovirus was transfected in the Gsn3 cells with 5% fetal bovine serum for 48 h. Cells were then cultured in the glucose-free DMEM for another 24 h.

For the TUNEL assay, cells were detected by the In Situ Cell Death Detection Kit (Roche Molecular Biochemicals, Basel, Switzerland). Briefly, cells were fixed with 4% paraformaldehyde for 1 h at room temperature. Cells were incubated in a TUNEL reaction mixture at 37 ${}^{\circ }$C for 30 min. The nuclei were counterstained with Hoechst 33342. At last, cells were viewed with a Zeiss fluorescence microscope.

In the cell viability assay, cells were determined by the Cell Counting Kit-8 (CCK-8, Dojindo). The absorbance of culture plates was measured at 450 nm with a microplate reader.

Protein samples were lysed from 0.5 cm cord segments (six in each sample) at lesion sites or from cells with RIPA lysis buffer (Beyotime). The concentration of each protein sample was detected by the BCA kit (Beyotime). The extracts were denatured and boiled at 95 ${}^{\circ }$C for 5 min, resolved by 10% SDS-PAGE, and transferred to PVDF membranes. The membranes were blocked with 5% nonfat milk in TBS and then incubated with primary antibodies buffer at 4 ${}^{\circ }$C for 18 h, followed with secondary antibody at room temperature for 2 h. The signal of HRP activity was detected using an enhanced chemiluminescence kit (Beyotime). Antibodies used in Western blot were: $\beta$-actin (1 : 5000, Proteintech, 66009-1-Ig), SOCS1 (1 : 1000, polyclonal rabbit anti-gecko antibody prepared from polypeptides), STAT1 (1 : 1000, Cell Signaling Technology, 14994), pSTAT1 (Tyr701) (1 : 1000, abcam, ab29045), cleaved caspase3 (1 : 1000, Cell Signaling Technology, 9661S), caspase3 (1 : 1000, Cell Signaling Technology, 9665), Bcl-xL (1 : 1000, Cell Signaling Technology, 2764), Akt (1 : 1000, Cell Signaling Technology, 4685S), p-Akt (1 : 1000, Cell Signaling Technology, 4060S), p-ERK (1 : 1000, Cell Signaling Technology, 4370S), ERK (1 : 1000, Cell Signaling Technology, 9102S), and goat anti-rabbit or goat anti-mouse HRP (1 : 1000, Proteintech, SA00001-2, SA00001-1).

Spinal cord segments were sectioned and incubated with SOCS1 antibody (1 : 500 dilution), GALC antibody (1 : 200 dilution, sigma, SAB1402780), p-STAT1 antibody (1 : 200 dilution, abcam), or NeuN antibody (1 : 200 dilution, abcam, ab104224) at 4 ${}^{\circ }$C for 36 h. Then, the sections were further incubated with Cy3-labeled goat anti-rabbit or anti-mouse IgG (1 : 400 dilution, Thermo Scientific, A16101, A16071), or FITC-labeled donkey anti-mouse or anti-rabbit IgG (1 : 400 dilution, Thermo Scientific, A24507, A16030) at 4 ${}^{\circ }$C for 18 h. The images were obtained by a confocal laser scanning microscope (Leica, Heidelberg, Germany).

Differences between groups were analyzed by one-way analysis of variance (ANOVA) with SPSS 23 software (SPSS, Chicago, IL, USA). Normality and homoscedasticity of the data were verified using Levene’s test before statistical analysis. p 0.05 was considered statistically significant.

To understand the structural characteristics of gSOCS1, the cDNA of gSOCS1 (GenBank accession number: XP_015265413) was first cloned by 5’- and 3’-RACE. The full length of the amplified product is 1372 bp, encoding a protein of 200 amino acid residues (Fig. 1). The gSOCS1 protein contains an N-terminal variable region, a KIR domain, a SH2 domain, a 12-aa ESS domain and a SOCS box conserved in other vertebrates homologs (Fig. 1). Alignment with other vertebrates has shown that gSOCS1 shares 37.3–70.1% identity with human, mouse, chicken, lizard, frog and zebrafish SOCS1, respectively (Fig. 1), indicating differential conservation of the protein in the phylogeny.

A phylogenetic tree demonstrated that gecko SOCS1 clustered with lizard and zebrafish, suggesting a closer evolutionary relationship. It also suggests that the gecko SOCS1 is more similar to mice and other mammals than lizards and zebrafish in an evolutionary relationship (Fig. 2).

Fig. 1.

Multiple alignment of amino acid sequences of gSOCS1 with those of representative vertebrates. Gaps introduced into sequences to optimize alignment are represented by dashes. KIR, SH2, ESS and SOCS box domain are indicated. Sequences of SOCS1 obtained from GenBank: gecko (XP_015265413), human (NP_003736), mouse (NP_001258532), rat (NP_665886), chicken (NP_001131120), crocodile (XP_019395658), lizard (XP_028562651), frog (NP_001011327), zebrafish SOCS1a (NP_001003467), and zebrafish SOCS1b (NP_001239559).

Fig. 2.

Phylogenetic analysis of SOCS1 proteins. Unrooted phylogenetic tree of SOCS1 proteins from the gecko and other representative species constructed by the neighbor-joining method within the package PHYLIP 3.5c. Bootstrap majority consensus values on 1000 replicates are indicated at each branch point as a percentage.

To elucidate the potential roles of gSOCS1 in the injured spinal cord, the amputation of the gecko tail was performed and 0.5-cm cord segments were collected at 0 d, 1 d, 3 d, 7 d and 2 w. The protein level of gSOCS1 detected by western blot analysis increased from 1 d onwards, with a peak at 7 d, followed by a return to the control level at 2 w (Fig. 3A,B). As activation of the STAT1 transcription factor is associated with the inducible expression of SOCS1, the phosphorylation of STAT1 was also determined. Results showed that phosphorylated STAT1 was significantly increased following gecko SCI (Fig. 3A,C). The data indicate that spinal cord injury induces the expression of gSOCS1 and its activator STAT1 protein.

To reveal the potential target cells within which gSOCS1 mediates its physiological functions, immunofluorescence staining was performed to detect colocalization of gSOCS1 with neurons and oligodendrocytes. The results showed that gSOCS1 was found to be distributed in NeuN-positive neurons and GALC-positive oligodendrocytes in the severed gecko cord (Fig. 4A,B). Meanwhile, gSOCS1 and its activator p-STAT1 were well colocalized (Fig. 4C), indicating that activation of STAT1 protein is possibly involved in inducing the expression of gSOCS1 in the neurons and oligodendrocytes in the injured spinal cord of gecko.

Fig. 3.

Determination of gSOCS3 and STAT1 protein levels in the injured spinal cord of gecko. Western blot analysis of gSOCS1 and phosphorylated STAT1 (A) in the 0.5 cm segments of the injured cord following gecko tail amputation (n = 6) at 0 d, 1 d, 3 d, 7 d and 2 w, respectively; (B) and (C) are statistical analysis of (A), respectively. The asterisks represent a significant difference between each time point and the control. Experiments were performed in triplicate. Data are expressed as mean $\pm$ SEM; *p 0.05, **p 0.01, ***p 0.001.

Fig. 4.

Tissue distribution of gSOCS1 in the injured spinal cord following gecko tail amputation at 0 d and 1 d (n = 5). (A,B) Immunohistochemistry shows the colocalization of gSOCS1 (red) with NeuN-positive neurons (green), GALC-positive oligodendrocytes (green). (C) Colocalization of gSOCS1 (green) with pSTAT1 (red). Scale bar, 100 $\mu$m. The rectangle indicates the region magnified. Arrowheads indicate the colocalized signals.

Injury of the mammalian spinal cord often leads to apoptosis of oligodendrocytes due to insults of various inflammatory cytokines. SOCS1 has been shown to have the capability to inhibit such deleterious effects [29]. To examine the protective roles of gSOCS1 on the oligodendrocytes in the regenerative spinal cord, an apoptotic model of the Gsn3 gecko oligodendrocyte cell line was established. As IFN-$\gamma$ is shown to efficiently induce apoptosis of mammalian oligodendrocytes [29, 30], Gsn3 cells were stimulated with 0–250 ng/mL recombinant IFN-$\gamma$ for 24, 48 and 72 h, respectively. However, the CCK-8 kit assay demonstrated that different concentrations of recombinant IFN-$\gamma$ were insufficient to promote Gsn3 apoptosis (Fig. 5A). TNF-$\alpha$ can potentiate IFN-$\gamma$ in inducing the cell death of oligodendrocytes [31]. Thus a combined treatment with TNF-$\alpha$ (0–50 ng/mL) and IFN-$\gamma$ (250 ng/mL) was subsequently performed on Gsn3 cells. Consequently, neither TNF-$\alpha$ nor IFN-$\gamma$ was able to induce cell apoptosis (Fig. 5B).

Fig. 5.

Effects of IFN-$\gamma$, TNF-$\alpha$ or glucose deprivation on the apoptosis of gecko Gsn3 cells. CCK-8 assay determined the cell viability following treatment with 0–250 ng/mL IFN-$\gamma$ (A), or combinations of TNF-$\alpha$ (0–50 ng/mL) and IFN-$\gamma$ (250 ng/mL) for 24, 48 and 72 h (B); (C) TUNEL detection of glucose deprivation-induced apoptosis of Gsn3 cells; (D) Assay of cell viability under glucose deprivation stress for 2, 4, 6 and 8 h, respectively; (E) Statistical analysis of (C) in 20 visual fields. Experiments were performed in triplicates. Data are expressed as mean $\pm$ SEM; *p 0.01. Scale bar, 50 $\mu$m.

We then turned to a GD-induced cell apoptosis model, which has been successfully established to potentiate the apoptosis of gecko astrocytes [32]. Following the culture of Gsn3 in a glucose-free medium for 24 h, the apoptotic cell number increased significantly, as evidenced by the TUNEL method (Fig. 5C,E). The CCK8 assay also confirmed cell viability decreased following cell culture in glucose-free medium for 2–8 h (Fig. 5D). The data indicate that glucose deprivation, rather than stimulation with IFN-$\gamma$, effectively induces the apoptosis of Gsn3 cells.

To evaluate the protective role of gSOCS1 in apoptotic oligodendrocytes, gSOCS1 adenovirus expression vectors were constructed that exhibited a transfection rate of over 80% (Fig. 6A,B). Gsn3 cells were transfected with GV314-gSOCS1 or GV314-vector adenovirus for 48 h, followed by incubation in a glucose-free culture for 24 h. Overexpression of gSOCS1 significantly reduced the apoptotic cell number of oligodendrocytes (Fig. 6C–E). The data indicate that gSOCS1 protects Gsn3 from glucose deprivation-induced apoptosis.

Fig. 6.

Determination of gSOCS1 protective roles against glucose deprivation-induced apoptosis on Gsn3. (A) Transfection efficiency of adenovirus on Gsn3 cells; (B) Statistical analysis of (A); (C,D) TUNEL assay of Gsn3 cells transfected with either GV314-gSOCS1 or GV314-vector adenovirus for 48 h, followed by glucose deprivation for 24 h; (E) Statistical analysis of (C, D). Data are expressed as mean $\pm$ SEM; *p 0.01. Scale bars, 25 $\mu$m in (A), 50 $\mu$m in (C,D).

Gekko japonicus has a remarkable regenerative ability to regenerate the injured tail comprised of muscles, cartilage, and spinal cord. The cells of regenerating ependymal tube begin to proliferate and penetrate into the blastema at 8–15 d after tail amputation [24]. Accompanied by this process, the number of either surviving or newly generated neurons increases from 7 d onwards (Supplementary Fig. 1). To clarify the potential effects of gSOCS1 on tail regeneration, a total of 9 $\mu$L GV314-gSOCS1 or GV314-vector adenovirus was injected into the amputated caudal vertebrae in three equal parts 2–3 mm from the plane. The GFP fluorescence intensity was used to detect the transfection efficiency in the sections of the gecko spinal cord (Supplementary Fig. 2A). Histopathology with LFB (Luxol fast blue) staining revealed that the GV314-vector group showed vacuolar changes and microtubule loss from the myelin sheath relative to the GV314-SOCS1 group at 7 d (Supplementary Fig. 2B). The data indicate that the overexpression of SOCS1 reduced the degree of myelin sheath destruction. However, the enforced expression of gSOCS1 did not promote regeneration of the tail following amputation at 7 d and 2 w (Supplementary Fig. 2C).

To unveil the mechanism of gSOCS1-mediated resistance to cell apoptosis, we determined apoptosis-related signaling after Gsn3 transfection with GV314-gSOCS1. Western blot analysis demonstrated that the activity of Caspase3 was significantly attenuated. At the same time, the protein level of Bcl-xL, a death-inhibitory member of the Bcl-2 family, was markedly increased following GV314-gSOCS1 transfection for 48 h before glucose deprivation for 24 h (Fig. 7A–C). The MEK/ERK and PI3K/Akt pathways promote the survival of cells by mediating the upregulation of Bcl-xL [33]. Therefore, the levels of both phosphorylated Erk1/2 and Akt in gSOCS1-mediated Gsn3 were determined. As expected, the phosphorylation levels of both Erk1/2 and Akt were significantly increased in comparison with control (Fig. 7A,D,E). The data indicate that gSOCS1-mediated resistance to oligodendrocyte apoptosis is involved in activating of MEK/ERK and PI3K/Akt pathways.

Fig. 7.

Western blot analysis for the apoptosis-related proteins following GV314-gSOCS1 or GV314-vector adenovirus transfection for 48 h, followed by glucose deprivation for 24 h on Gsn3. (A) Cleaved-caspase3, Bcl-Xl, p-Akt, and p-ERK were detected by western blot. (B–E) Statistical analysis of (A). Experiments were performed in triplicate. Data are expressed as mean $\pm$ SEM; *p 0.01.