Young (2 month old) and old (22 month old) female C57BL/6N mice were obtained from the Center of Experimental Animals, Capital Medical University (Beijing, China). All mice were housed in an air-conditioned room where the temperature was 22 ± 2 °C and relative humidity was 55% ± 10%, and followed a standard 12:12 light/dark cycle. Food and water were available ad libitum. We euthanized mice by cervical dislocation after they were deeply anesthetized with 5% isoflurane inhalation, confirmed by the absence of a pedal withdrawal reflex upon firm toe pinch. The animal protocols were approved by The Animal Care and Use Committee of Capital Medical University. (Approval No.: AEEI-2022-157).

We used 2% isoflurane for mouse anesthesia. Thereafter, laminectomy was performed to expose the T10 spinal cord, and 70-kilodyne contusion was performed using an Infinite Horizons Impactor (Precision Systems & Instrumentation, Lexington, KY, USA) before suturing of muscles and incision. Throughout the surgery and post-anesthesia recovery, the animals were put into the warming chamber until complete consciousness was reached. The mice were hydrated using Ringer’s solution (0.5 ml, SC) for 5 days postoperatively. The urinary bladder was manually emptied twice or more daily throughout our study period. Surgical interventions and postoperative animal care were conducted following guidelines and relevant policies for rodent survival surgery released by the Experimental Animal Committee of the Capital Medical University. The schematic diagrams of the spinal cord injury modeling process and the experimental groups were shown in Supplementary Figure 1. Mice were separated into 6 groups with 8 mice in each group: sham (Y-sham and O-sham); acute phase following injury (3 days post injury, Y-SCI-A, and O-SCI-A), chronic phase following injury (28 days post injury, Y-SCI-C and O-SCI-C). Y: young mice (2 months); O: old mice (22 months); A:acute phase following injury; C: chronic phase following injury.

The total RNA was extracted from tissue samples using the TRIzol^® Reagent and then subjected to a quality assessment using the 5300 Bioanalyzer (Agilent). The extracted RNA was quantified using the ND-2000 (NanoDrop Technologies). The high-quality RNA samples (OD260/280 = 1.8-2.2, OD260/230 ≥ 2.0, RIN ≥ 6.5, 28S:18S ≥ 1.0, > 1 µg) were then used for library construction. RNA purification, reverse transcription, library preparation, and sequencing were performed at Shanghai MajorBio Bio-pharm Biotechnology Co. Ltd. by following the Illumina protocols. The RNA-seq library was prepared using 1 µg of the total RNA, and the process comprised mRNA isolation using poly-A selection, cDNA synthesis using random hexamer primers, and end-repair using ‘A’ base addition. Size-selected libraries (300 bp) were then PCR amplified and sequenced on a NovaSeq 6000 sequencer (2 × 150 bp).

Raw paired-end reads were subjected to quality trimming and control using fastp with default parameters. The clean reads were then aligned to the reference genome using HISAT2 in orientation mode. The mapped reads were assembled using String Tie in a reference-based manner. The threshold for selection was |log2FC| ≥ 1 and FDR ≤ 0.05. Differentially expressed genes (DEGs) were identified using the DESeq2 package in R. KEGG enrichment analysis was performed using the SciPy library in Python. The above data analysis was conducted using the free online Majorbio Cloud Platform (http://www.majorbio.com). RNA sequencing data have been deposited in Sequence Read Archive (SRA) with the accession number PRJNA1241889.

Fecal samples were collected in 2.0 mL sterile tubes, snap frozen in liquid nitrogen, and stored at –80 °C for further analysis. The fecal samples were then placed in dry ice and sent to Shanghai Majorbio Bio-pharm Biotechnology Co., Ltd. (Shanghai, China) for 16S rRNA Gene Amplicon Sequencing and Bioinformatics and Statistical Analysis. This gene contains conserved and variable regions, where the conserved regions facilitate amplification with universal primers, while the sequence differences in the variable regions serve as molecular markers for identifying bacterial species and genera, making it a standard method for studying the structure of complex microbial communities. Briefly, microbial genomic DNA was extracted using the OMEGA Stool DNA Kit (D4015-02, Omega Bio-Tek, Norcross, GA, USA) according to the manufacturer’s instructions. The quality of extracted DNA was examined by agarose gel electrophoresis, and quantified using QuantiFluor™ -ST (Promega, USA). PCR amplification of the bacterial 16S rRNA genes V3–V4 region was performed using the forward primer 338F (5′-ACT CCT ACG GGA GGC AGC A-3′) and the reverse primer 806R (5′-GGA CTA CHV GGG TWT CTA AT- 3′). After the individual quantification step, amplicons were pooled in equal amounts, and paired-end sequenced (2 × 300 bp) on an Illumina MiSeq platform according to the standard protocols. The raw reads were deposited in Sequence Read Archive (SRA) with the accession number PRJNA1242401.

Alpha/beta diversity analyses, RDA, heatmap generation, and statistical tests based on community data were performed in the QIIME2 and R environment. Data analysis was carried out on the free online Majorbio Cloud Platform (http://www.majorbio.com).

The data are presented as the means and standard errors of the means and were analyzed via SPSS 17.0 (SPSS Inc., Chicago, IL, USA). Two-way ANOVA was applied to evaluate the main effects of mice age (young vs. old) and injury phase (acute phase vs. Chronic phase), as well as their interaction. And two-tailed unpaired t tests were used for comparisons between two groups. Spearman correlation analysis was used to assess correlations. All the statistical tests were performed via GraphPad Prism 9.0 software (San Diego, CA). A significance level of p < 0.05 was considered statistically significant.

Previous investigations have demonstrated that young mice exhibit superior recovery compared to aged mice after mid-thoracic contusion injury (19, 20). To explore the underlying molecular mechanisms, we employed transcriptome sequencing to analyze the differential gene expression profiles between young and aged mice during both acute and chronic phases of SCI. A series of spinal cord samples in the acute stage (3 days post injury, Y-SCI-A, and O-SCI-A), chronic stage (28 days post injury, Y-SCI-C and O-SCI-C) and sham (Y-sham and O-sham) were collected, sequenced, and subjected to bioinformatics analysis. KEGG pathway analysis revealed that in the young group, the differentially expressed genes (DEGs) between Y-sham and Y-SCI-A were significantly enriched in pathways such as neuroactive ligand-receptor interaction, TNF signaling pathway, and MAPK signaling pathway (Figure 1A). The DEGs between Y-sham and Y-SCI-C were primarily enriched in cytokine-cytokine receptor interaction, neuroactive ligand-receptor interaction, and NF-kappa B signaling pathway (Figure 1B). Meanwhile, the DEGs between Y-SCI-A and Y-SCI-C showed significant enrichment in pathways including cytokine-cytokine receptor interaction and PI3K-Akt signaling pathway (Figure 1C). In the aged group, the DEGs between O-sham and O-SCI-A were significantly enriched in neuroactive ligand-receptor interaction, TNF signaling pathway, MAPK signaling pathway, PI3K-Akt signaling pathway, and cytokine-cytokine receptor interaction (Figure 1D). The DEGs between O-sham and O-SCI-C were mainly enriched in cytokine-cytokine receptor interaction, neuroactive ligand-receptor interaction, NF-kappa B signaling pathway, and chemokine signaling pathway (Figure 1E). Additionally, the DEGs between O-SCI-A and O-SCI-C were significantly enriched in pathways such as cytokine-cytokine receptor interaction, cell adhesion molecules, and chemokine signaling pathway (Figure 1F). Overall, the analysis demonstrated that the enriched pathways of DEGs in both young and old groups were highly similar, with a predominant focus on immune response-related pathways.

To further elucidate the differences in spinal cord gene expression profiles between young and old mice at the same time period after SCI, we performed differential gene expression analysis on the sham, acute phase, and chronic phase groups of young and old mice. KEGG annotation data revealed that, DEGs were primarily enriched in pathways related to the immune system, signaling molecules and interaction, and signal transduction in Y-sham vs O-sham, O-SCI-A vs Y-SCI-A, and O-SCI-C vs Y-SCI-C (Figures 2A–C). KEGG pathway enrichment analysis showed that the DEGs between Y-sham and O-sham were significantly enrichment in pathways such as cytokine-cytokine receptor interaction and chemokine signaling pathway (Figure 2D). In the acute phase of SCI, the DEGs between O-SCI-A and Y-SCI-A were mainly enriched in pathways including neutrophil extracellular trap formation and cell adhesion molecules (Figure 2E). In the chronic phase of SCI, the DEGs between O-SCI-C and Y-SCI-C were significantly enriched in pathways such as cytokine-cytokine receptor interaction and Th17 cell differentiation (Figure 2F). Notably, immune responses, particularly DEGs enriched in cytokine-cytokine receptor interaction, showed significant differences in an age-dependent manner following SCI. A heatmap illustrates the expression changes of detected cytokines among groups (Figure 3). The heatmap demonstrates that some cytokines were highly expressed in the acute phase (e.g., ccl2, ccl7, ccl12), while others showed relatively higher expression in the chronic phase (e.g., ccl8, ccl4, CXCR4, cxcl13, cxcl16) (Figure 3). On the whole, the age-dependent upregulation of cytokines was more pronounced under both normal and pathological conditions.

Previous studies showed that the abundance and composition of the gut microbiota are altered, which was observed not only after SCI (16, 21), but also following aging (22, 23). The mechanisms by which injury and aging collaboratively influence gut microbiota changes are yet to be fully elucidated. We explore how gut microbiota is altered in young and old mice after spinal cord injury using 16S rRNA sequencing. As shown in Figures 4A, B, significant differences in the Chao and Shannon indices were observed among the groups. The Chao index increased obviously at 3 days after SCI and declined slowly at 28 days after injury in both young and old mice. However, there was a significant difference in Chao index between groups (Y-sham vs Y-SCI-A and Y-SCI-A vs Y-SCI-C) in young mice, whereas there was no statistically significant difference in Chao index between groups of old mice (O-sham vs O-SCI-A and O-SCI-A vs O-SCI-C). A similar trend was observed in the Shannon index. These data suggest that spinal cord injury significantly affects the diversity and the richness of gut microbiota and that the extent of change is more pronounced in young mice after injury compared to old mice. NMDA revealed distinct clustering patterns in the acute phase when compared to the sham group, with the microbiome structures progressively recovering during the chronic phase in young mice. In old mice, the gut microbial structure showed little variation at different time points after SCI compared to the sham group; whereas the clustering pattern of gut microbiota in young mice at 28 days post-injury closely resembled that of old mice (Figure 4C). At both the phylum and genus levels, the SCI groups at different stages of injury presented distinct microbial profiles compared with those of the sham group in both young and old mice (Figures 4D–G).

After SCI, the microbiota in the young and old mice exhibited distinct responses ((Figure 5). Centered log-ratio transformation was applied to normalize the microbial data, followed by two-way ANOVA. This analysis identified 24 taxa with significantly different responses to SCI between the young and old (Figure 5). The adjusted p values from the two-way ANOVA were shown in Supplementary Table S1. The abundance of Lactobacillus, Dubosiella showed an immediate reduction during the acute phase postinjury, followed by a progressive increase throughout the chronic phase; both young and old mice exhibited similar patterns of gut microbiota alterations following SCI. The abundance of g:Prevotellaceae_UCG-001 swiftly increased at 3 days postinjury, followed by a decrease but remained elevated above sham levels until 28 days postinjury, and young and old mice demonstrated comparable changes in their gut microbiota profiles after SCI. Additionally, certain bacterial genera showed distinct variation pattern after SCI in young and old mice. As the condition progresses from the acute to the chronic phase, the microbial composition gradually shifts toward that of the sham group. While, the fluctuation of gut microbiota is significantly greater in young mice when compared to old mice, suggesting that the self-repair capacity of the gut microbiota is diminished in aged individuals.

To further elucidate the gut microbial differences between young and old mice during the same phase after SCI, we conducted differential analysis of gut microbial between young and old mice in sham group, acute and chronic phase. The results showed that compared with Y-sham group, the abundance of Lactobacillus, Dubosiella, norank_f:norank_o:_Clostridia_UCG-014, Candidatus_Saccharimonas, Faecalibaculum were significantly decreased, while the abundance of Clostridium_ sensu_stricto_1 and Bacteroides were significantly increased in O-sham group (Figure 6A). Compared to Y-SCI-A group, the abundance of norank:f:norank:o:_Clostridia_UCG-014, Prevotellaceae_UCG-001, and Prevotellaceae_NK3B31_group was significantly decreased, while the abundance of norank:f: Lachnospiraceae abundance was significantly increased in O-SCI-A group (Figure 6B). The abundance of Dubosiella, Alloprevotella, Enterococcus were significantly decreased, while the abundance of Lachnospiraceae_UCG-001 was significantly increased in O-SCI-C group compared to Y-SCI-C group (Figure 6C). The results indicate that there are significant differences in the changes of gut microbiota between young and old mice at different injury phases after SCI. Combined with the above analysis of intergroup differences, we noted that g:Dubosiella, which was significantly reduced in the acute phase of injury, continued to show an age-dependent decline in the chronic phase as well.

To explore the relationship between the host and microbiota, we selected the differential inflammatory cytokines as the environmental factor for RDA. RDA/CCA revealed that most of cytokines significantly affected the bacterial composition at the genus level (Figure 7A). Also, the 22 inflammatory cytokines was analyzed by two-way ANOVA, showing 6 inflammatory cytokines (Tnfsf8, Ccl3, Ccl4, Il1rn, Il2rg, Csf3r) with significantly different responses to SCI between the young and old. The adjusted p values from the two-way ANOVA were shown in Supplementary Table S2. A correlation heatmap analysis of the relationships between cytokines and the community compositions of the different groups is shown in Figure 7B. The abundance of Lactobacillus, Dubosiella, Clostridium_sensu_stricto_1 and Bacillus showed a significant negative correlation with the majority of cytokines, while the abundance of Lachnospiraceae_NK4A136_group and Allobaculum exhibited a significant positive correlation with most of the cytokines. Correlation analysis indicated that alterations in the abundance of the gut microbiota were closely associated with changes in the cytokine levels of spinal cord.