Short-chain fatty acids modulate regulatory T cells via NF-κB signaling pathway in a rat model of spinal cord injury

Abstract

Objective

To investigate the effects and mechanisms of short-chain fatty acids on regulatory T cells (Treg cells) in spinal cord injury (SCI) using cell and animal experiments.

Methods

A mixture of SCFAs (acetate, propionate, and butyrate) was used. In vitro, the optimal SCFAs concentration was determined by CCK-8 assay. The expression of IL-4, IL-10, and IL-13 was assessed by RT-qPCR and ELISA. IKKβ, P65, and p-P65 levels were analyzed by Western blot. In vivo, the modified Allen’s weight-drop method was used to establish a moderate contusion SCI model in SD rats. Animals were randomly divided into Sham, SCI, and SCFAs groups. Functional recovery was evaluated using BBB scoring, inclined plane test, and CatWalk analysis. Spinal cord pathology was examined by H&E and Nissl staining. Tissue cytokine levels and Treg cell numbers were measured by ELISA and flow cytometry. Statistical comparisons were performed using one-way or two-way ANOVA with appropriate post-hoc tests.

Results

In vitro, SCFAs at the working concentration (0.5 mM) did not compromise Treg cell viability (P > 0.05), while promoting IL-4, IL-10, and IL-13 expression in SCI-derived Treg cells (P < 0.05). SCFAs significantly decreased IKKβ and p-P65 levels in SCI-derived Treg cells (P < 0.01). In vivo, SCFAs supplementation improved inclined plane test angles and BBB scores from day 7 onwards (P < 0.01). CatWalk analysis revealed decreased hindlimb base of support (P < 0.05) and increased regularity index (P < 0.01). Histological examination showed reduced cavitation and improved neuronal morphology. SCFAs treatment elevated IL-4, IL-10, and IL-13 levels (P < 0.05) and increased Treg cell numbers (P < 0.01).

Conclusions

SCFAs enhanced anti-inflammatory cytokine production by Treg cells and promoted motor recovery after SCI in association with reduced NF-κB signaling activation, highlighting SCFAs as a potential therapeutic strategy for SCI.

Introduction

Spinal cord injury (SCI) represents one of the most devastating disorders of the central nervous system. According to the National Spinal Cord Injury Statistical Center, the annual incidence of SCI is approximately 54 cases per million population in the USA. ¹ Males represent approximately 80% of affected individuals, and the substantial lifetime healthcare costs impose a significant socioeconomic burden on patients, families, and healthcare systems. ¹ Primary mechanical injury leads to vascular disruption, blood-spinal cord barrier damage, and neuronal destruction. ¹ Following injury, neurons release damage-associated molecular patterns such as heat shock proteins and purine metabolites, triggering excessive immune cell activation and inflammatory mediator secretion. This results in dramatic alterations of the immune microenvironment, leading to secondary damage that expands the lesion over days to weeks. ² Therefore, modulating post-injury neuroinflammation has emerged as a crucial therapeutic strategy for SCI.

Regulatory T cells (Treg cells) constitute a distinct T cell subpopulation that modulates autoimmune responses and maintains immune homeostasis. ³ Following SCI, IL-17+γδT cells are recruited to the injury site, amplifying the inflammatory cascade. ⁴ Treg cells can suppress IL-17+γδT cell differentiation through secretion of anti-inflammatory factors such as IL-10 and TGF-β, thereby attenuating neuroinflammation. ⁵ However, the functional status of Treg cells after SCI and factors regulating their anti-inflammatory capacity remain incompletely understood.

The NF-κB signaling pathway plays a central role in inflammatory responses following tissue injury. Upon activation, IKKβ phosphorylates IκBα, leading to its degradation and subsequent release of P65, which then translocates to the nucleus to drive inflammatory gene transcription. ⁶ While NF-κB activation has been extensively studied in macrophages and microglia after SCI, its role in regulating Treg cell function in this context has not been examined. This represents a critical knowledge gap, as understanding how NF-κB signaling regulates Treg cell function after SCI may reveal new therapeutic targets for modulating neuroinflammation.

Short-chain fatty acids (SCFAs), including acetate, propionate, and butyrate, are major metabolites produced by gut microbiota through dietary fiber fermentation. ⁷ Recent studies have established the existence of a gut-spinal cord axis, with SCI causing dysbiosis and notably decreased abundance of SCFA-producing bacteria. ⁸ SCFAs have been shown to alleviate neuroinflammation in Alzheimer’s disease models ⁹ and to reduce ocular inflammation by modulating intestinal Treg cells. ¹⁰ The SCFA mixture ratio (acetate:propionate:butyrate = 12:5:3) used in this study approximates physiological proportions found in the mammalian gut and has been validated in previous metabolic studies. ¹¹

Based on the known anti-inflammatory properties of both Treg cells and SCFAs, we hypothesized that SCFAs might enhance Treg cell anti-inflammatory function after SCI, potentially through modulation of NF-κB signaling. This study aimed to investigate whether exogenous SCFA supplementation could promote Treg cells activity and motor function recovery after SCI, and to examine the potential involvement of the NF-κB pathway. To test this hypothesis, we employed a combination of in vitro cellular assays using SCI-derived Treg cells and in vivo experiments using a rat SCI model to evaluate functional recovery, histological outcomes, and immunological parameters.

Materials and methods

Reporting guidelines

All animal experiments were conducted in accordance with institutional guidelines, and the reporting of this study conforms to the ARRIVE 2.0 guidelines. ¹²

Experimental animals

Fifteen female Sprague-Dawley (SD) rats, approximately 8 weeks old and weighing 200±20 g, were obtained from the Laboratory Animal Center of Guangxi Medical University (SCXK Gui 2020-0003); females were chosen due to their lower incidence of post-operative urinary tract infection and easier manual bladder expression. The sample size was determined by a priori power analysis based on prior SCI studies with comparable endpoints. The animals were housed in a specific pathogen-free (SPF) facility under controlled environmental conditions: temperature maintained at 22-26°C, relative humidity at 40-60%, and a 12-h light/dark cycle. Animals were housed in groups of 2-3 per cage with standard bedding material and environmental enrichment to improve welfare. Throughout the experimental period, animals had ad libitum access to standard rodent chow and water. Following a one-week acclimatization period, the experiments were initiated. All animal experiments were conducted in accordance with the Guidelines for the Care and Use of Laboratory Animals and were approved by The Animal Care & Welfare Committee of Guangxi Medical University (Approval No. 202305541, May 2023). Efforts were made to minimize the number of animals used and their suffering.

Construction of SCI model

The modified Allen’s weight-drop method was used to establish the SCI model. ¹³ SD rats were anesthetized with isoflurane inhalation (3% for induction, 1.5-2% for maintenance) delivered via a nose cone and fixed in a prone position on the operating table. Body temperature was maintained at 37°C using a heating pad. Using T10 as the central point, a 3.0-4.0 cm longitudinal skin incision was made along the midline, followed by layer-by-layer separation of muscles and other soft tissues until the vertebral plate was opened and the spinal cord was exposed. A stainless steel rod weighing 10 g was dropped vertically from a height of 4 cm onto the exposed dura mater. This impact force (40 g·cm) produces a moderate contusion injury characterized by transient hindlimb paralysis with gradual partial recovery. The model was considered successful when the rats exhibited irregular tail movements, spastic convulsions of hindlimbs, and hematoma formation at the dura mater. The surgical wound was sutured in layers.

Following surgery, rats were placed on a heating pad until recovery from anesthesia. Post-operative care included manual bladder expression twice daily until spontaneous voiding recovered. Rats received subcutaneous saline (2 mL) immediately after surgery for fluid support. Animals were monitored daily for signs of distress, infection, autophagia, or weight loss exceeding 20% of pre-surgical body weight. Humane endpoints included complete loss of mobility, severe infection unresponsive to treatment, or loss of more than 20% body weight. No animals reached humane endpoints or died during the study. At 21 days post-surgery, rats were euthanized by CO₂ inhalation in a closed chamber, with 100% CO₂ introduced at a fill rate of approximately 30% of the chamber volume per minute. CO₂ flow was maintained for at least 1 minute after respiratory arrest, and death was further confirmed by cessation of heartbeat and absence of corneal reflex before tissue collection.

Animal groups and experimental design

Fifteen SD rats were randomly allocated using a random number table to three groups (n=5 per group): Sham group (laminectomy only without spinal cord impact), SCI group (spinal cord contusion followed by daily intragastric administration of 2 mL normal saline for 21 days beginning on postoperative day 1), and SCFAs group (spinal cord contusion followed by daily intragastric administration of 2 mL SCFAs solution at 150 mM for 21 days beginning on postoperative day 1). The SCFAs dosage was selected based on previous studies demonstrating efficacy in rodent models of neurological conditions. ¹⁴ Group allocation was concealed from personnel responsible for behavioral testing, data collection, and histological analysis throughout the study. Behavioral assessments were performed at 1, 3, 7, 14, and 21 days post-surgery by two independent observers blinded to group assignment. All rats were euthanized for tissue collection at 21 days post-surgery.

Isolation of Treg cells from SD rats

For in vitro experiments, Treg cells were isolated from either normal uninjured rats or from rats at 72 hours post-SCI. Lymphocytes were isolated from spinal cord tissue using Rat Peripheral Blood Lymphocyte Separation Kit (Solarbio, Beijing, China). The isolated lymphocytes (2×10^8 cells) were resuspended in 100 μL Auto MACS Running Buffer (Miltenyi Biotec, Germany), followed by addition of 10 μL anti-rat CD4 and anti-rat CD25 antibodies (eBioscience, USA). After incubation at 4°C for 15 minutes, cells were washed and resuspended in 80 μL buffer with 20 μL magnetic beads (Miltenyi Biotec, Germany), incubated at 4°C for 20 minutes.

An MS separation column (Miltenyi Biotec, Germany) was pre-wetted and loaded with cell suspension. After elution of CD4+ T cells, 20 μL magnetic beads were added and incubated at 4°C for 15 minutes. The suspension was passed through a new MS column, and collected Treg cells (CD4+CD25+ T cells) were cultured in RPMI 1640 medium (Sigma, USA) supplemented with 10% Fetal Bovine Serum (FBS; Thermo Fisher Scientific, USA) at 37°C with 5% CO₂ in a CO₂ incubator (Thermo Fisher Scientific, USA). The purity and phenotype of isolated Treg cells were confirmed by flow cytometry using anti-CD4 and anti-CD25 antibodies, with the proportion of CD4⁺CD25⁺ cells consistently >90% across all isolations.

In vitro experimental groups

Four experimental groups were established for in vitro studies: Group a consisted of Treg cells isolated from normal SD rats cultured in standard medium; Group b contained Treg cells isolated from SD rats with SCI cultured in standard medium; Group c comprised Treg cells isolated from normal SD rats treated with 0.5 mM SCFAs (concentration selected based on CCK-8 results), which also falls within the physiologically relevant range of circulating SCFA concentrations (0.1–1 mM) ¹⁵ ; and Group d included Treg cells isolated from SD rats with SCI treated with 0.5 mM SCFAs. For clarity regarding the use of different control groups in vitro versus in vivo: in vitro experiments used cells from “normal” uninjured rats as controls because no surgical procedure was performed, whereas in vivo experiments used “Sham” rats that underwent laminectomy without spinal cord impact to control for surgical stress. Both represent appropriate controls for their respective experimental contexts.

Reagents

Fetal bovine serum (FBS) was obtained from Thermo Fisher Scientific (USA). RPMI 1640 medium was purchased from Sigma (USA). Anti-rat CD4 and anti-rat CD25 antibodies were purchased from eBioscience (USA); Auto MACS Running Buffer, magnetic beads, and MS separation columns were obtained from Miltenyi Biotec (Germany). ELISA kits for IL-4, IL-10, IL-13, IL-17A, IL-1β, TNF-α, and IFN-γ were purchased from Shanghai Enzyme-linked Biotechnology (Shanghai, China). The reverse transcription kit was obtained from Qihengxing Biotechnology (China). RNAiso Plus reagent was purchased from Takara (Japan). 4% Paraformaldehyde, PBS, and antigen retrieval solution were obtained from Servicebio (Wuhan, China) or Solarbio (Beijing, China). Trypsin and Percoll cell separation solution were purchased from Solarbio (China), and Collagenase Type IV was obtained from Gibco (USA).

CCK-8 assay

Cell viability of Treg cells under different concentrations of SCFAs was assessed using the CCK-8 assay. Treg cells were seeded in 96-well plates at a density of 1×10^4 cells/well with 100 μL culture medium per well. After overnight culture, cells were treated with varying concentrations of SCFAs (0, 0.1, 0.25, 0.5, 1, and 2 mM) for 24 hours. Each concentration was tested in six replicate wells, and the experiment was repeated three times independently. Subsequently, 10 μL of CCK-8 solution was added to each well, and the plates were incubated at 37°C for 1 hour in the dark. The absorbance was measured at 450 nm using a BioTek ELx800 microplate reader (BioTek, USA). Cell viability was calculated as a percentage relative to untreated control cells.

RT-qPCR

The effects of SCFAs on inflammatory cytokine mRNA expression in Treg cells were analyzed using RT-qPCR. Total RNA was extracted from Treg cells using RNAiso Plus (Takara, Japan) and quantified using a Nanodrop 2000 spectrophotometer (Thermo Fisher Scientific, USA). RNA purity was confirmed by A260/A280 ratios between 1.8 and 2.0. One microgram of total RNA was reverse-transcribed to cDNA using the reverse transcription kit (Qihengxing Biotechnology, China) according to manufacturer’s instructions. RT-qPCR was performed using SYBR Green on a LightCycler 480 (LC480) real-time PCR system (Roche, Germany). The cycling conditions were: initial denaturation at 95°C for 3 minutes, followed by 40 cycles of denaturation at 95°C for 10 seconds and annealing at 60°C for 30 seconds, with a final extension at 72°C for 90 seconds. Melting curve analysis was performed to verify amplification specificity. The relative mRNA expression levels were calculated using the 2^(-ΔΔCT) method with GAPDH as the reference gene. All samples were run in triplicate from three independent experiments. The primer sequences are shown in Table 1.

Table 1.

Primer sequences for RT-qPCR.

Gene	Forward (5′–3′)	Reverse (5′–3′)
IL-4	TGTACCGGGAACGGTATCCA	ACATCTCGGTGCATGGAGTC
IL-10	AATTGAACCACCCGGCATCT	TTTCCAAGGAGTTGCTCCCG
IL-13	TGAGCAACATCACACAAGAC	CATCCGAGGCCTTTTGGTTA
GAPDH	GCATCTTCTTGTGCAGTGCC	GATGGTGATGGGTTTCCCGT

ELISA

The expression levels of inflammatory proteins in Treg cells were measured by ELISA. Supernatants from Treg cells under different treatment conditions were collected and centrifuged at 3000 rpm for 10 minutes using a low-temperature high-speed centrifuge (Eppendorf, Germany) to remove particles and aggregates. After centrifugation, the supernatants were collected. Standard wells and sample wells were set up on the ELISA plate (Shanghai Enzyme-linked Biotechnology, China). Different concentrations of standards were added to the standard wells, and samples from each experimental group were added to the sample wells. The plate was sealed with plate sealer and incubated at 37°C for 1 hour. Subsequently, 50 μL each of chromogenic reagents A and B were added, and the plate was sealed and incubated at 37°C in the dark for 15 minutes. Finally, 50 μL of stop solution was added to each well to terminate the reaction. The absorbance was measured at 450 nm using a BioTek ELx800 microplate reader (BioTek, USA). All samples were measured in duplicate, and the experiment was repeated three times independently.

Western blot

Total protein was extracted from Treg cells using RIPA lysis buffer. Protein concentrations were determined using the BCA Protein Assay Kit (Beyotime, China), and equal amounts of protein (20 μg per lane) were loaded for each sample. Equal amounts of protein were separated by 10% SDS-PAGE and transferred to 0.22 μm PVDF membranes at 200 mA for 120 minutes. Membranes were blocked with 5% non-fat milk at room temperature for 1.5 hours, then incubated overnight at 4°C with the following primary antibodies: rabbit polyclonal anti-phospho-NF-κB P65 (Proteintech, Wuhan, China; Cat. No. 80379-2-RR, RRID:AB_2890123, 1:1000), rabbit polyclonal anti-NF-κB P65 (Proteintech; Cat. No. 10745-1-AP, RRID:AB_2178878, 1:1000), rabbit polyclonal anti-IKKβ (Proteintech; Cat. No. 15649-1-AP, RRID:AB_2283775, 1:1000), and rabbit polyclonal anti-GAPDH (Proteintech; Cat. No. 10494-1-AP, RRID:AB_2263076, 1:2500). After TBST washing, membranes were incubated with HRP-conjugated goat anti-rabbit IgG secondary antibody (Proteintech; Cat. No. SA00001-2, RRID:AB_2722564, 1:5000) and developed using ECL reagents. Band intensities were quantified using ImageJ software (version 1.53, NIH, USA) and normalized to GAPDH. Relative protein expression levels were calculated as the ratio of target protein band intensity to GAPDH band intensity for each sample.

Inclined plane test and BBB score

Rat SCI recovery was assessed using the inclined plane test and BBB (Basso-Beattie-Bresnahan) scoring at 1, 3, 7, 14, and 21 days post-surgery. For the inclined plane test, rats were placed on an inclined platform, and the angle between the platform and the horizontal plane was gradually adjusted. The maximum angle at which rats could maintain their position for 5 seconds without sliding was recorded. BBB scoring was conducted using a double-blind method by two independent observers who were blinded to group allocation and had no involvement in surgical procedures or drug administration. The scores from both observers were averaged for each animal at each time point. The observers evaluated the locomotor activity of rats from each group according to the scoring criteria.

CatWalk gait analysis

At 21 days post-surgery, the VisuGait gait analysis system (Shanghai Xinruan, Shanghai, China) was used. Prior to surgery, all rats received training to familiarize them with the runway, completing 3-5 runway crossings per session. Testing required rats to complete their walk within 10 seconds with speed variation not exceeding 60%. Hindlimb base of support (BOS) and regularity index were analyzed.

H&E staining and nissl staining

After anesthetizing the rats, they were perfused with 4% paraformaldehyde (Servicebio, Wuhan, China). A 2 cm length of spinal cord tissue centered on the injury site was removed and fixed in 4% paraformaldehyde. Paraffin sections were prepared using a pathological slicer (Leica, Germany) and embedding machine (Wuhan Junjie, China). After deparaffinization and dehydration, H&E and Nissl staining (Servicebio, China) were performed. Tissue morphology was observed under a fluorescence microscope (Axio Imager. M2, ZEISS, Germany).

Flow cytometry

Spinal cord tissue fragments extending 5 mm above and below the injury center were collected and cut into 1 mm² pieces, then placed in digestive solution containing 0.25% trypsin (Solarbio, China) and 0.1% collagenase IV (Gibco, USA). After 20 minutes of digestion at 37°C with gentle agitation, the tissue was filtered through a 40 µm nylon mesh. Spinal cord lymphocytes were separated using Percoll density gradient centrifugation (Solarbio, China). 2×10⁶ lymphocytes were resuspended in 100 μL FACS buffer (PBS containing 2% FBS) to prepare a cell suspension. The cell suspension was mixed with anti-rat CD4 (eBioscience, USA) and anti-rat CD25 (eBioscience, USA) antibodies, then incubated at 4°C in the dark for 30 minutes. Subsequently, 200 μL buffer was added and mixed, followed by centrifugation at 500 g for 5 minutes. After removing the supernatant, 100 μL IC fixation buffer was added, and the mixture was incubated at 4°C in the dark for 60 minutes. After another centrifugation at 500 g for 5 minutes and removal of supernatant, cells were finally resuspended in 300 μL buffer for flow cytometric analysis using a CytoFLEX flow cytometer (Beckman Coulter, USA). Data were analyzed using FlowJo software. Cells were first gated on lymphocytes based on FSC/SSC characteristics, followed by singlet discrimination. CD4+ cells were then gated, and CD25+ cells within the CD4+ population were identified as Treg cells. Fluorescence-minus-one (FMO) controls were used to set gating boundaries.

Statistical analysis

Data analysis was performed using GraphPad Prism 8.0. All results are presented as mean ± SD. Normality and homogeneity of variance were verified using Shapiro-Wilk and Brown-Forsythe tests, respectively. When assumptions of normality or homogeneity of variance were not met, non-parametric alternatives (Kruskal-Wallis test) were considered; however, all datasets in this study satisfied both assumptions. The sample size (n=5 per group) was determined to be adequate for detecting statistical significance in preliminary animal studies, consistent with recent guidelines. For the CCK-8 assay, differences were analyzed using one-way ANOVA followed by Dunnett’s post-hoc test. For in vitro experiments involving four groups, comparisons were performed using two-way ANOVA followed by Bonferroni’s test. For in vivo endpoint comparisons among three groups, one-way ANOVA followed by Tukey’s test was used. Behavioral data collected at multiple time points were analyzed using two-way repeated measures ANOVA followed by Bonferroni’s test. A P-value < 0.05 was considered statistically significant.

Results

Effect of different concentrations of SCFAs on Treg cell viability assessed by CCK-8 assay

SCFAs at concentrations at or below 0.5 mM had no significant effect on Treg cell viability (P > 0.05), while concentrations of 1 mM and 2 mM significantly inhibited cell viability (P < 0.01 and P < 0.0001, respectively; Figure 1). Based on these results, 0.5 mM SCFAs was selected for subsequent in vitro experiments.

Figure 1.

Effect of different concentrations of SCFAs on Treg cell viability assessed by CCK-8 assay. Data are mean ± SD; error bars represent SD (n = 3 independent experiments with six replicates per concentration). **P < 0.01, ****P < 0.0001 vs. 0 mM control (one-way ANOVA with Dunnett’s test).

Effect of SCFAs on the mRNA expression levels of inflammatory cytokines in Treg cells

Using this optimized concentration, compared to group a, mRNA levels of IL-4, IL-10, and IL-13 were significantly decreased in group b (all P < 0.0001), indicating that SCI impaired anti-inflammatory cytokine expression in Treg cells. SCFAs treatment significantly increased mRNA levels of all three cytokines in group d compared to group b (all P < 0.0001; Figure 2a–c).

Figure 2.

Effect of SCFAs on mRNA expression of anti-inflammatory cytokines in Treg cells detected by RT-qPCR. (a) IL-4, (b) IL-10, (c) IL-13. Data are mean ± SD; error bars represent SD (n = 3 independent experiments). Asterisks above brackets indicate significant differences between the bracketed group pairs. ****P < 0.0001 (two-way ANOVA with Bonferroni’s test). Group a: Treg cells from normal SD rats; Group b: Treg cells from SCI rats; Group c: Treg cells from normal rats + 0.5 mM SCFAs; Group d: Treg cells from SCI rats + 0.5 mM SCFAs.

Effect of SCFAs on protein expression levels of inflammatory cytokines in Treg cells

Consistent with the mRNA findings, protein levels of IL-4, IL-10, and IL-13 were significantly decreased in group b compared to group a (P < 0.001). SCFAs treatment restored cytokine expression, with group d showing significantly higher levels than group b (P < 0.05; Figure 3a–c).

Figure 3.

Effect of SCFAs on protein expression of anti-inflammatory cytokines in Treg cells measured by ELISA. (a) IL-4, (b) IL-10, (c) IL-13. Data are mean ± SD; error bars represent SD (n = 3 independent experiments). Asterisks above brackets indicate significant differences between the bracketed group pairs. *P < 0.05, ***P < 0.001, ****P < 0.0001 (two-way ANOVA with Bonferroni’s test). Group designations as in Figure 2.

Effect of SCFAs on NF-κB signaling pathway in Treg cells

Having confirmed that SCFAs restored anti-inflammatory cytokine expression at both the mRNA and protein levels, we next investigated the underlying signaling mechanism. SCI activated the NF-κB signaling pathway in Treg cells. Protein levels of p-P65 and IKKβ were significantly elevated in group b compared to group a (both P < 0.0001), while total P65 levels remained unchanged. SCFAs treatment significantly reduced both p-P65 and IKKβ levels in group d compared to group b (both P < 0.01; Figure 4a–c).

Figure 4.

Effect of SCFAs on NF-κB signaling pathway in Treg cells. (a) Representative Western blot images of p-P65, P65, IKKβ, and GAPDH. (b) Quantification of p-P65. (c) Quantification of IKKβ. Protein levels were normalized to GAPDH. Data are mean ± SD; error bars represent SD (n = 4 independent experiments). Asterisks above brackets indicate significant differences between the bracketed group pairs. **P < 0.01, ****P < 0.0001 (two-way ANOVA with Bonferroni’s test). Group designations as in Figure 2.

Effect of SCFAs on motor function recovery after SCI in rats

To determine whether the in vitro immunomodulatory effects of SCFAs translated into functional benefits in vivo, we performed a series of behavioral and histological assessments. Note that in the in vitro experiments above, “normal” refers to Treg cells derived from non-operated rats, whereas in the following in vivo experiments, “Sham” refers to rats that underwent laminectomy without SCI, serving as surgical controls.

All rats had normal BBB scores (21 points) before surgery. On day 1 post-surgery, both SCI and SCFAs groups scored 0, confirming complete hindlimb paralysis. From day 3, gradual recovery was observed in all injured animals, but no significant difference between groups was detected (P > 0.05). From day 7 onwards, rats in the SCFAs group showed significantly higher BBB scores than those in the SCI group, and this difference persisted through day 21 (P < 0.01 at day 7; P < 0.0001 at days 14 and 21; Figure 5(b)). Inclined plane test results showed a similar pattern. From day 7 post-surgery, the SCFAs group maintained stability at significantly steeper angles compared to the SCI group (P < 0.01; Figure 5(a)).

Figure 5.

Effect of SCFAs on motor function recovery after SCI. (a) Inclined plane test scores. (b) BBB locomotor rating scale scores. Data are mean ± SD; error bars represent SD (n = 5 per group). **P < 0.01, ****P < 0.0001 vs. SCI group at corresponding time points (two-way repeated measures ANOVA with Bonferroni’s correction).

Effects of SCFAs on rat gait parameters after SCI

To further characterize locomotor recovery, CatWalk gait analysis was performed at 21 days post-surgery. Sham rats displayed regular gait patterns with clear footprints, while SCI rats showed disordered footprints with obvious dragging and increased print density. The SCFAs group showed notably improved footprint distribution compared to the SCI group (Figure 6(a)).

Figure 6.

Effects of SCFAs on gait parameters after SCI. (a) Representative CatWalk images showing footprint patterns. (b) Hindlimb base of support. (c) Regularity index. Data are mean ± SD; error bars represent SD (n = 5 per group). Asterisks above brackets indicate significant differences between the bracketed group pairs. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 (one-way ANOVA with Tukey’s test).

Hindlimb base of support was significantly increased in the SCI group compared to Sham (P < 0.001). SCFAs treatment partially reversed this deficit (P < 0.05 vs. SCI), though values remained elevated compared to Sham (Figure 6(b)). The regularity index was dramatically decreased in the SCI group compared to Sham (P < 0.0001). SCFAs treatment significantly improved this parameter compared to the SCI group (P < 0.01; Figure 6(c)).

Effects of SCFAs on spinal cord tissue morphology in rats

H&E and Nissl staining were performed at 21 days post-surgery. In the Sham group, spinal cord structure was intact with clear gray-white matter boundaries, no inflammatory cell infiltration or tissue cavitation, and neurons displayed normal morphology with darkly stained, well-organized Nissl bodies (Figure 7a–b).

Figure 7.

Effects of SCFAs on spinal cord histopathology at 21 days post-surgery. (a) Representative H&E staining. (b) Representative Nissl staining. Representative images from n = 5 rats per group are shown.

In the SCI group, the spinal cord structure was severely disrupted with obvious tissue cavitation, extensive inflammatory cell infiltration, significantly reduced neuron numbers, and remaining neurons showed lightly stained, disorganized Nissl bodies (Figure 7a–b). Compared to the SCI group, the SCFAs group displayed more preserved spinal cord architecture with notably reduced cavitation and inflammatory cell infiltration, increased numbers of surviving neurons, and better-preserved Nissl body staining (Figure 7a–b).

Effects of SCFAs on Treg cells following SCI in rats

To evaluate whether SCFAs modulated the immune response in vivo, flow cytometry analysis at 21 days post-surgery showed that compared to the Sham group, the percentage of Treg cells (CD4+CD25+) was significantly increased in the SCI group (P < 0.05). SCFAs treatment further increased Treg cell numbers, with the SCFAs group showing significantly higher percentages than the SCI group (P < 0.01; Supplementary Figure 1a–d).

Effects of SCFAs on anti-inflammatory cytokines following SCI in rats

Consistent with the Treg cells findings, ELISA analysis at 21 days post-surgery showed that compared to the Sham group, protein levels of IL-4, IL-10, and IL-13 were significantly decreased in the SCI group (P < 0.05). SCFAs treatment significantly increased the levels of all three cytokines compared to the SCI group (P < 0.05; Supplementary Figure 2a–c).

Discussion

SCI triggers a cascade of inflammatory responses that extend beyond the initial mechanical insult, with the dysregulated immune microenvironment constituting a major obstacle to neural recovery.^16,17 While accumulating evidence supports the existence of a gut-spinal cord-immune axis and the immunomodulatory potential of microbial metabolites, ¹⁸ the specific cellular targets and molecular mechanisms through which short-chain fatty acids influence post-injury neuroinflammation have remained incompletely characterized. This study provides evidence that SCFAs modulate Treg cell function potentially through modulation of NF-κB signaling, thereby enhancing anti-inflammatory cytokine production and promoting functional recovery after SCI.

Our central finding is that SCFA treatment significantly enhanced the anti-inflammatory capacity of Treg cells, as evidenced by increased expression of IL-4, IL-10, and IL-13 at both transcriptional and protein levels. Concurrently, SCFA-treated Treg cells exhibited reduced IKKβ expression and diminished P65 phosphorylation, indicating suppression of the NF-κB signaling cascade. These parallel changes were not merely correlative; when Treg cells were exposed to SCI-conditioned medium, which mimics the inflammatory microenvironment, the opposite pattern emerged—elevated NF-κB activation with attenuated anti-inflammatory cytokine output. SCFA pretreatment reversed this injury-induced phenotype, restoring both signaling homeostasis and cytokine production. In vivo, SCFA supplementation increased Treg cell infiltration at the lesion site, reduced tissue IL-17 levels while elevating IL-4, IL-10, and IL-13, diminished pathological cavitation, and improved functional outcomes including BBB locomotor scores and CatWalk gait parameters. These convergent observations across experimental systems suggest that SCFAs exert immunomodulatory effects through Treg cells in the context of SCI.

The mechanistic interpretation of these findings centers on the role of IKKβ as a regulatory node within the NF-κB pathway. In canonical NF-κB signaling, the IKK complex—comprising IKKα, IKKβ, and NEMO—phosphorylates IκBα, marking it for proteasomal degradation and liberating P65/P50 dimers for nuclear translocation and transcriptional activation. ¹⁹ IKKβ serves as the principal catalytic subunit for inflammatory signal transduction, and its activity level directly determines the magnitude of downstream NF-κB activation. ²⁰ Our data showing reduced IKKβ expression following SCFA treatment suggest that SCFAs act upstream in this cascade, limiting the initial activation step rather than merely blocking nuclear translocation or DNA binding. This upstream intervention would preserve the cytoplasmic IκBα-P65 complex in its inhibited state, thereby preventing the entire transcriptional program associated with NF-κB activation. ²¹ The reduction in p-P65 levels observed in our study is consistent with this interpretation, as diminished IKKβ would result in less IκBα phosphorylation and consequently less P65 release and activation. Notably, SCFAs exhibited dose-dependent biphasic effects on Treg cells. Physiologically relevant concentrations (≤0.5 mM) preserved cell viability and enhanced anti-inflammatory cytokine production, whereas higher concentrations (≥1 mM) impaired viability, likely due to excessive HDAC inhibition or intracellular acidification. Accordingly, 0.5 mM was selected to balance viability and functional efficacy.

The precise molecular mechanism by which SCFAs reduce IKKβ expression remains to be fully elucidated. SCFAs may act through multiple non-mutually exclusive pathways, including HDAC inhibition that alters chromatin accessibility and suppresses inflammatory gene transcription. ²² Specifically, butyrate-mediated HDAC inhibition may reduce transcriptional activator binding at the IKBKB locus and disrupt the positive autoregulatory feedback on IKBKB transcription through acetylation of RelA/P65, which diminishes its DNA-binding affinity.^22,23 SCFAs also signal through G-protein-coupled receptors such as GPR41, GPR43, and GPR109A expressed on T lymphocytes, ²⁴ where β-arrestin-2 recruitment can directly interfere with IKK complex assembly and promote IKKβ proteasomal degradation. ²⁵ Additionally, intracellular metabolism to acetyl-CoA may activate AMPK, which phosphorylates and inhibits IKKβ kinase activity.^26,27 Our use of a mixed SCFA preparation was designed to approximate physiological gut-derived SCFA composition and capture potential synergistic effects among components, ²⁸ but precludes determination of whether acetate, propionate, or butyrate—each with distinct receptor affinities and HDAC inhibitory potencies—contributes differentially to IKKβ suppression. Systematic comparison of individual SCFAs at equimolar concentrations represents an important avenue for future mechanistic dissection.

A notable aspect of our findings concerns the specific cytokine profile enhanced by SCFA treatment. The coordinated upregulation of IL-4, IL-10, and IL-13 suggests polarization toward a Th2-associated anti-inflammatory phenotype. ²⁹ IL-10 is the canonical immunosuppressive cytokine produced by Treg cells and is critical for limiting inflammatory damage in numerous contexts. ³⁰ IL-4 and IL-13, while classically associated with Th2 helper cells, can also be produced by certain Treg cell subsets and contribute to alternative macrophage activation and tissue repair. ³¹ The enhancement of all three cytokines indicates that SCFAs may promote a tissue-protective Treg cell subphenotype rather than simply amplifying existing functions. This interpretation is supported by the observation that SCFA-treated animals showed not only reduced inflammatory markers but also improved tissue preservation, suggesting active promotion of a reparative microenvironment.

The relationship between our findings and the existing literature on SCFA immunomodulation reveals both consistencies and intriguing divergences. Consistent with our results, Nakamura and colleagues demonstrated that SCFAs ameliorate experimental autoimmune uveitis partly through expansion of Treg cell populations in gut-associated lymphoid tissue, ³² and Smith et al. showed that microbiota-derived SCFAs promote colonic Treg cells differentiation through HDAC inhibition. ³³ More recently, mounting evidence indicates that SCI itself induces significant gut microbiome dysbiosis, characterized by reduced SCFA-producing bacterial genera, which in turn exacerbates systemic inflammation and impairs neurological recovery. ³⁴ This bidirectional gut–spinal cord interaction suggests that SCFA depletion may be both a consequence of and a contributing factor to post-injury immune dysregulation. Beyond SCFAs, other microbiome-derived metabolites such as tryptophan derivatives and secondary bile acids have also been implicated in modulating CNS neuroinflammation through aryl hydrocarbon receptor and farnesoid X receptor signaling, respectively, ³⁵ highlighting a broader role for microbial metabolites in neuroimmune crosstalk. These studies support a generalizable role for SCFAs in enhancing regulatory immunity across different inflammatory contexts. Our data extend this paradigm to SCI, demonstrating that SCFA-Treg cell interactions are relevant not only in mucosal immunity but also in CNS trauma.

Previous studies have primarily attributed SCFA-mediated immunomodulation to enhanced Treg cells differentiation and Foxp3 induction, ³⁶ whereas our data emphasize functional enhancement of existing Treg cells through NF-κB suppression. This distinction likely reflects context-dependent mechanisms: in the gut mucosa, continuous antigen exposure favors de novo Treg cells differentiation, while the acute inflammatory milieu following SCI may preferentially benefit from augmented effector function of already-differentiated Treg cell populations. The route of administration and resultant tissue concentrations also influence which mechanisms predominate. ³⁷ Beyond Treg cells, SCFAs exert broad immunomodulatory effects on macrophages, dendritic cells, and effector T cell populations, ³⁸ and the reduction in tissue IL-17 levels observed in our study may reflect direct effects on Th17 or γδT cells in addition to Treg cell-mediated suppression. ³⁹ The relative contributions of these parallel pathways cannot be fully distinguished in the current study. Moreover, oral SCFA supplementation may exert additional systemic effects—including improved gut barrier function and altered circulating metabolite profiles ⁴⁰ —that could independently influence neuroinflammatory outcomes, and thus the observed benefits likely reflect multiple SCFA-mediated mechanisms rather than Treg cells modulation alone.

The temporal dynamics of functional recovery in our study merit consideration in relation to the known time course of immune cell infiltration following SCI. The divergence in BBB scores between SCFA-treated and control groups became apparent from day 7 post-injury, coinciding with the peak of T cell infiltration and the transition from acute to subacute inflammatory phases. ⁴¹ This timing suggests that SCFA effects are indeed mediated through immune modulation rather than acute neuroprotection. Earlier interventions targeting the initial microglial and neutrophil responses might conceivably yield even greater benefits, ⁴² though whether oral SCFA supplementation could achieve sufficient tissue concentrations rapidly enough for acute effects remains uncertain given the pharmacokinetics of absorption and distribution. ⁴³

The functional improvements observed in our study, while statistically significant, were modest in absolute magnitude, consistent with the broader literature on immunomodulatory interventions in SCI that typically demonstrate partial rather than complete recovery. ⁴⁴ However, it should be noted that in the BBB scoring system, incremental gains within certain score ranges correspond to qualitatively distinct locomotor milestones—for example, the transition from occasional weight-supported stepping to consistent coordinated limb movement—which carry meaningful implications for potential functional independence. ⁴⁵ Moreover, the magnitude of improvement observed in our study is comparable to or exceeds that reported for other single-modality immunomodulatory strategies in preclinical SCI models, ⁴⁶ supporting the translational potential of SCFA supplementation. From a translational perspective, oral SCFA administration or dietary fiber supplementation to promote endogenous SCFA production represents a readily implementable, low-cost intervention that could complement existing rehabilitation protocols, although optimal dosing, timing, long-term safety profile, and the relative efficacy of individual SCFAs versus mixtures remain to be determined in clinical settings. This limitation reflects the complex pathophysiology of SCI, in which inflammation represents only one contributing factor alongside axonal degeneration, demyelination, and glial scarring that may be less amenable to immune modulation.^47,48 Additionally, our 21-day observation period captures meaningful subacute changes but may not fully reflect the ultimate functional outcomes achievable with extended recovery times, and the timing and dosing of SCFA administration may not have been optimized for maximal therapeutic effect. Furthermore, only female rats were used in this study, which may limit the generalizability of findings given known sex differences in immune responses and recovery after SCI. ⁴⁹ Additionally, the relatively modest sample size and the absence of genetic or pharmacological validation of the IKKβ-dependent mechanism limit the causal conclusions that can be drawn from the current data.

These considerations inform directions for future investigation. Studies employing individual SCFAs rather than mixtures would clarify structure-activity relationships and potentially identify the most therapeutically relevant metabolite. Genetic approaches including IKKβ conditional knockout in Treg cells would provide more definitive evidence for the proposed mechanistic pathway. Extended observation periods and validation in chronic injury models would address whether early functional improvements translate to sustained long-term benefits. Inclusion of male animals in future studies is also warranted to address sex-dependent differences in immune responses and improve translational generalizability. Finally, combination approaches pairing SCFA supplementation with other regenerative strategies—such as rehabilitation, growth factor delivery, or cell transplantation—may achieve greater functional restoration than single-modality treatments. ⁵⁰

Conclusion

In conclusion, this study demonstrates that SCFAs enhance Treg cell anti-inflammatory function by suppressing IKKβ expression and P65 phosphorylation, thereby increasing IL-4, IL-10, and IL-13 production and promoting functional recovery after SCI. These findings identify the SCFA-Treg cell–NF-κB axis as a potential therapeutic target for neurotrauma.

Supplemental material

Supplemental material - Short-chain fatty acids modulate regulatory T cells via NF-κB signaling pathway in a rat model of spinal cord injury

Supplemental material for Short-chain fatty acids modulate regulatory T cells via NF-κB signaling pathway in a rat model of spinal cord injury by Hao Deng, Junhong Zhou, Yilin Teng, Deshuang Xi, Shaohui Zong in Science Progress