Intermittent fasting alleviates hyperalgesia in ovariectomized mice via gut microbiota remodeling

Abstract

Pain is a common symptom of menopause, yet effective therapeutic options are limited. Intermittent fasting (IF) has emerged as a promising dietary intervention; however, its effects on menopausal pain are still unclear. In this study, we established a hyperalgesia model in mice through ovariectomy (OVX) and subjected them to an alternate-day fasting regimen. IF significantly elevated the pain thresholds for mechanical, hot and cold stimuli in OVX mice and reduced the expression of pain-related molecules, including transient receptor potential vanilloid 1 (TRPV1) and calcitonin gene-related peptide (CGRP) in dorsal root ganglion (DRG). Moreover, IF remodeled the gut microbiota and metabolite profile, marked by a substantial increase in the abundance of Akkermansia muciniphila and its key metabolite, indole-3-propionic acid (IPA). Depletion of the gut microbiota via antibiotic treatment abolished the analgesic effects of IF on OVX-induced hyperalgesia. Conversely, fecal microbiota transplantation from IF-treated donors restored microbial composition and alleviated hyperalgesia in OVX recipients. Administration of A. muciniphila increased IPA levels and alleviated hyperalgesia in OVX mice. Importantly, exogenous IPA supplementation not only alleviated hyperalgesia but also reduced the excitability of DRG neurons. Together, these findings demonstrate that IF mitigates estrogen deficiency-related hyperalgesia through remodeling gut microbiota and metabolite profile, and identify IPA as a potential therapeutic target, offering new perspectives for the clinical management of menopausal pain.

Introduction

Menopausal pain, a highly prevalent and debilitating condition, is primarily driven by estrogen decline following decreased ovarian function^1–3. It is characterized by heightened pain sensitivity (hyperalgesia) and chronic pain, encompassing a spectrum of symptoms such as musculoskeletal pain, headaches or migraines, and urogenital pain^1,2. Epidemiological studies indicate that nearly 54% of middle-aged women are affected, with significant impairment to quality of life^4,5. Despite this prevalence, there are currently no therapies specifically targeting menopausal pain⁶. While conventional pharmacologic analgesics provide temporary relief, their long-term utility is limited by adverse effects^6,7. Thus, lifestyle interventions, particularly dietary modifications, have emerged as promising strategies for sustainable symptom management and health optimization.

Intermittent fasting (IF), a widely practiced dietary strategy, typically involves a 24h fasting period followed by 24 h of unrestricted eating⁸. Emerging evidence indicates that IF exerts multiple health benefits, including a reduction in the inflammatory burden, regulation of energy metabolism and enhancement of cellular stress resistance^9,10. In addition to potentially extending longevity, IF shows therapeutic promise for various conditions like obesity, diabetes, cancer, asthma, hypertension, cardiovascular diseases, ischemic stroke, and neurodegenerative disorders¹¹. Furthermore, recent studies indicate its potential in pain management, demonstrating antinociceptive effects in a formalin-induced pain mouse model¹² and pain improvement in osteoarthritic patients^13,14. However, the effect of IF on menopausal pain remains unclear.

The gut microbiota, a vast and diverse community of microorganisms in the gastrointestinal tract, plays a critical role in the maintenance of host health¹⁵. Dysbiosis of gut microbiota increases susceptibility to a variety of diseases, such as metabolic diseases, cardiovascular diseases, and neurological diseases¹⁶. Recent evidence indicates the involvement of gut microbiota in various pain conditions, including visceral pain, inflammatory pain, neuropathic pain, headache, and opioid tolerance^17–22. Notably, alterations in the gut microbiota and its metabolites have been documented in both menopausal women and ovariectomized (OVX) mice²³. Previous studies have shown that OVX mice develop pain behaviors such as mechanical, thermal, and cold hyperalgesia^24,25. Fecal microbiota transplantation (FMT) from OVX mice induced hyperalgesia in normal mice, whereas remodeling of gut microbiota via FMT from SHAM-operated mice attenuated pain sensitivity in OVX animals²⁶. Therefore, the gut microbiota might play a crucial role in the regulation of menopausal pain.

IF is well known to dynamically modulate the gut microbiota and its metabolic profile²⁷. Based on this premise, our study was designed to elucidate whether IF alleviates menopausal pain by remodeling the gut microbiota. Using an OVX mouse model of hyperalgesia, we implemented a 24-h alternate-day fasting protocol to test this hypothesis.

Results

IF alleviated hyperalgesia in OVX mice

To investigate the effects of IF on menopausal pain, we utilized an OVX mouse model of hyperalgesia and implemented an alternate-day fasting protocol (Fig. 1a). IF treatment eliminated the slightly higher body weight in OVX mice relative to SHAM controls (Fig. S2a). Behavioral tests revealed that OVX mice developed significant hypersensitivity to mechanical and thermal (cold and hot) stimuli compared to SHAM mice. In contrast, IF treatment markedly increased the paw withdrawal thresholds in response to mechanical stimuli and prolonged withdrawal latencies to thermal stimuli in OVX mice. However, the pain sensitivity in OVX + IF mice remained higher than in SHAM groups (Fig. 1b-d). Consistent with these behavioral findings, immunofluorescence staining demonstrated that IF reduced the number of TRPV1-positive neurons and downregulated CGRP expression in the DRGs of OVX mice (Fig. 1e–h). Collectively, our results indicate that IF effectively alleviates OVX-induced hyperalgesia, but does not fully normalize OVX-induced hyperalgesia, suggesting it can partially reverse the negative effects of ovariectomy.

Fig. 1. IF alleviated hyperalgesia in OVX mice.

a Schematic diagram showing the strategy of the IF intervention protocol in OVX mice, n = 7 for each group. b Change in paw withdrawal threshold to mechanical stimuli. c, d Change in paw withdrawal latency to cold (4 °C) and hot (50 °C) stimuli. e-h Representative immunofluorescence images and quantification of relative fluorescence intensity showing (e, f) TRPV1 and (g, h) CGRP co-localization with NeuN in DRGs. Data were represented as mean ± SEM. SHAM vs. OVX, *p < 0.05, **p < 0.01, ***p < 0.001,****p < 0.001; OVX vs. OVX + IF, #p < 0.05, ##p < 0.01, ###p < 0.001, ####p < 0.0001.

IF remodeled gut microbiota and elevated serum IPA levels in OVX mice

To assess the effects of IF on gut microbiota, we conducted 16S rRNA gene amplicon sequencing of fecal samples. Principal component analysis (PCA) revealed that the microbial community of the OVX group was distinct from the SHAM group. Notably, IF treatment not only shifted the microbial profile away from the OVX cluster but also positioned it separately from both the OVX and SHAM groups (Fig. 2a). A Venn diagram showed an increase in operational taxonomic units (OTUs) in the OVX + IF groups compared to the OVX group (Fig. 2b). Although the Shannon index showed no significant differences in alpha diversity among groups (Fig. 2c), the ACE index was significantly higher in the OVX + IF group than in the OVX group (Fig. 2d). Linear discriminant analysis effect size (LEfSe) revealed that IF treatment induced significant shifts in bacterial communities from the phylum to the genus level (Fig. 2e, f). Specifically, compared to the OVX group, the OVX + IF group displayed a decreased Firmicutes/Bacteroidetes ratio and a lower abundance of Verrucomicrobiota at the phylum level (Fig. 2g), which was accompanied by an increase in Bacteroidia and a decrease in Clostridia at the class level (Fig. 2h), as well as decreased Lachnospirales and increased Bacteroidales and Oscillospirales at the order level (Fig. 2i).

Fig. 2. IF remodeled the gut microbiota in OVX mice.

a PCA for beta diversity of the gut microbiota in the SHAM, OVX and OVX + IF groups, n = 5 for each group. b Venn diagram of operational taxonomic units (OTUs) among the three groups. c, d Shannon index and ACE index in alpha diversity. e Taxonomic cladogram generated from LEfSe analysis, f Linear discriminant analysis (LDA >4.0). g–i Relative abundance of microbiota at the phylum, class and order levels among three groups. *p < 0.05; **p < 0.01, ***p < 0.001 for the comparison.

To further explore the impact of IF on gut microbiota-derived metabolites, serum samples were collected and subjected to metabolomic analysis. The results showed significant differences in serum metabolites among the three groups, with notable differences remaining between OVX mice after IF treatment and the SHAM group (Fig. S3a–e). Furthermore, we focused on the effects of IF on OVX mice. PCA revealed significant overall metabolic differences between the OVX and OVX + IF groups (Fig. 3a). Subsequent OPLS-DA analysis demonstrated distinct metabolic profiles, with minor intra-group variations but pronounced inter-group differences (Fig. 3b). Annotation of the identified metabolites using the KEGG database revealed that tryptophan metabolites accounted for 4.41% of the total (Fig. 3c). Gut microbiota are known to metabolize tryptophan into various indole derivatives, such as indole-3-propionic acid (IPA), indole-3-lactic acid (ILA), and indole acetic acid (IAA)^28,29. In OVX mice, IF treatment selectively elevated serum IPA levels, but not ILA or IAA, according to our ELISA results (Fig. 3d–f). IPA production is mediated by specific gut bacteria, including A. muciniphila and Escherichia coli^30–32. In contrast to the stable levels of E. coli, IF treatment markedly increased the abundance of A. muciniphila in OVX mice (Fig. 3g, h).

Fig. 3. IF elevated serum IPA levels and abundance of Akkermansia muciniphila in OVX mice.

a PCA for beta diversity of serum samples in the OVX and OVX + IF groups, n = 5 for each group. b OPLS-DA score chart between the OVX and OVX + IF groups and correlation coefficient. c Annotation of the identified metabolites using the KEGG database. d–f ELISAs for serum IPA, ILA, and IAA levels, n = 6 for each group. g Abundance of E. coli. h Abundance of A. muciniphila. *p < 0.05; **p < 0.01, ***p < 0.001 for the comparison.

Gut microbiota mediated the protective effects of IF against hyperalgesia in OVX mice

To investigate whether gut microbiota remodeling mediates the alleviation of hyperalgesia, we first administered antibiotics (ABX) to deplete the gut microbiota in IF-treated OVX mice (Figs. 4a and S4). Compared to the IF group, ABX treatment in the IF + ABX group resulted in a significant decrease in mechanical withdrawal thresholds and shortened withdrawal latencies to both thermal and cold stimuli (Fig. 4b–d). Consistent with these behavioral findings, immunofluorescence staining demonstrated that gut microbiota depletion reversed the IF-induced reduction in the number of TRPV1-positive neurons and the downregulation of CGRP expression in the DRG of OVX mice (Fig. 4e–h).

Fig. 4. Depletion of gut microbiota by antibiotics abolished the protective effects of IF against hyperalgesia in OVX mice.

a Schematic diagram illustrating the strategy of the gut microbiota depletion by antibiotic (ABX) in IF-treated OVX mice, n = 6 for each group. b–d The alterations (b) in the mechanical withdrawal threshold after ABX treatment; Paw withdrawal latency of (c) cold plate (4 °C) and (d) hot plate (50 °C) among groups after ABX treatment. e-h Representative immunofluorescence images and quantification of relative fluorescence intensity showing (e, f) TRPV1 and (g, h) CGRP co-localization with NeuN in L3-L5 DRGs of the AL, IF, and IF + ABX groups. Data were represented as mean ± SEM. AL vs. IF, *p < 0.05, **p < 0.01, ***p < 0.001; IF vs. IF + ABX, #p < 0.05, ##p < 0.01, ###p < 0.001.

In addition, we transplanted fecal microbiota from OVX mice and IF-treated OVX mice donors into OVX recipient mice (Fig. 5a). This procedure effectively restored the gut microbial composition in the recipients (Fig. S5). Compared to the FMTOVX group, the FMTOVX + IF group showed increased pain thresholds, with higher mechanical withdrawal thresholds and longer withdrawal latencies to both thermal and cold stimuli (Fig. 5b–d). Meanwhile, the number of TRPV1-positive neurons and the expression level of CGRP in the DRG were lower in the FMTOVX + IF group than in the FMTOVX group (Fig. 5e–h).

Fig. 5. FMT from IF-treated donors alleviated hyperalgesia in OVX recipient mice.

a Schematic diagram showing the strategy of FMT treatment, n = 6 for each group. b The alterations in the mechanical withdrawal threshold over 3 weeks. c, d Paw withdrawal latency of (c) cold plate(4 °C) and (d) hot plate(50 °C) among groups over 3 weeks. e-h Representative immunofluorescence images and quantification of relative fluorescence intensity showing (e, f) TRPV1 and (g, h) CGRP co-localization with NeuN in DRGs. *p < 0.05; **p < 0.01, ***p < 0.001 for the comparison.

Administration of A. muciniphila increased IPA production and alleviated hyperalgesia in OVX mice

To investigate the causal role of the IF-induced increase in A. muciniphila (AKK) abundance on IPA production, we directly supplemented OVX mice with exogenous AKK via oral gavage (Fig. 6a). ELISA results revealed that AKK supplementation elevated the serum IPA levels in AKK -treated OVX mice compared to the OVX group (Fig. 6b). In behavioral tests, AKK treatment significantly raised the withdrawal thresholds in response to mechanical stimuli and extended withdrawal latencies to both cold and hot stimuli in OVX mice (Fig. 6c–e). Pearson correlation analysis revealed positive correlations between the elevated IPA levels and the changes in pain thresholds (Fig. 6f–h). Additionally, immunofluorescence staining showed that AKK administration significantly reduced the number of TRPV1-positive neurons and CGRP expression in the DRGs (Fig. 6i–l).

Fig. 6. Administration of A. muciniphila increased IPA production and alleviated hyperalgesia in OVX mice.

a Schematic diagram illustrating the strategy of A. muciniphila (AAK) administration, n = 7 per group. b Serum concentration of indole-3-propionic acid. c Change in paw withdrawal thresholds to mechanical stimuli. d, e Change in paw withdrawal latencies to cold (4 °C) and hot (50 °C) plates. f–h Pearson correlation analysis. i–l Representative immunofluorescence images and quantification of relative fluorescence intensity showing (i, j) TRPV1 and (k, l) CGRP co-localization with NeuN in DRGs. The data are presented as the mean ± SEM. *p < 0.05; **p < 0.01, ***p < 0.001 for the comparison. In (b–d), SHAM v.s. OVX, *p < 0.05, **p < 0.01, ***p < 0.001; OVX v.s. OVX + AKK, #p < 0.05, ##p < 0.01, ###p < 0.001, for the comparison.

Exogenous IPA supplementation alleviated hyperalgesia in OVX mice and reduced the excitability of DRG neurons

To determine whether elevated IPA levels contribute to the alleviation of hyperalgesia induced by IF, we administered exogenous IPA to OVX mice via oral gavage (Fig. 7a). Behavioral assays revealed that IPA administration elevated mechanical and thermal pain thresholds in OVX mice, as evidenced by increased paw withdrawal thresholds to mechanical stimuli and longer withdrawal latencies in response to both hot and cold stimuli (Fig. 7b–d). Immunofluorescence staining demonstrated that IPA treatment reduced the number of TRPV1-positive neurons and CGRP expression in the DRGs of OVX mice (Fig. 7e–h). The effect of IPA on neuronal excitability was evaluated by whole-cell patch-clamp recordings in primary DRG neurons from OVX mice (Fig. 7i). Under current-clamp conditions, IPA treatment altered the electrophysiological parameters of neurons (Fig. S6). Specifically, IPA perfusion hyperpolarized the resting membrane potential (Fig. 7j). Additionally, IPA decreased the firing frequency of action potentials (APs) (Fig. 7k, l) and lowered their threshold intensity (Fig. 7m). Collectively, IPA treatment reduced the spontaneous excitability of neurons.

Fig. 7. Exogenous IPA supplementation alleviates hyperalgesia in OVX mice and reduces the excitability of DRG neurons.

a Schematic diagram showing the strategy of IPA administration. b Change in paw withdrawal threshold to mechanical stimuli, n = 7 for each group. c, d Change in paw withdrawal latency to cold (4 °C) and hot (50 °C) stimuli. e-h Representative immunofluorescence images and quantification of relative fluorescence intensity showing (e, f) TRPV1 and (g, h) CGRP co-localization with NeuN in DRGs. i Schematic diagram showing the strategy of whole-cell patch-clamp recordings. j Resting membrane potential of DRG neurons from OVX mice (n = 12, p = 0.0038, paired t-test). k Representative traces of AP trains evoked by IPA (20 μM) infusion in DRG neurons of OVX mice under current-clamp stimulation. l The firing frequency of APs at a current intensity of 30 pA. m The rheobase of OVX mice neurons perfused with IPA (20 μM) (n = 12, p = 0.0367, paired t-test).*p < 0.05; **p < 0.01, ***p < 0.001 for the comparison. In (b–d), data were represented as mean ± SEM. SHAM v.s. OVX, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; OVX v.s. OVX + IPA, #p < 0.05, ##p < 0.01, ###p < 0.001, ####p < 0.0001 for the comparison.

Discussion

Menopausal pain is a common symptom that poses a significant clinical challenge due to limited effective therapies³³. Current pharmacologic analgesia offers only transient relief and carries risks of adverse effects, highlighting the need for safe alternatives^6,7. To explore the role of fasting regimens in menopausal pain management, we established an OVX-induced hyperalgesia model and treated mice with an IF protocol (24-h alternating fasting/feeding). Previous studies have shown that as little as 10 days of IF can improve gut microbiota distribution and enhance nerve repair in mice, whereas approximately 4 weeks of IF improves overall metabolism and cognitive function^28,34,35. In humans, 8 weeks of IF is also sufficient to alter gut microbiota composition and exert positive effects³⁶. In this study, behavioral differences between the OVX and OVX + IF groups began to emerge 3–4 weeks after IF initiation, indicating the onset of IF effects. Accordingly, for ABX experiments, we selected mice that had undergone 3 weeks of IF post-OVX as the baseline IF group and initiated ABX treatment at this time point. Behavioral performance in IF-treated OVX mice plateaued between weeks 6 and 7, converging to a stable range. Therefore, week 7 was designated as the observation endpoint. Daily food intake was recorded for each group throughout the experimental period. Food consumption in the OVX + IF group progressively decreased after fasting periods (Fig. S2b), eventually reaching levels comparable to those in the SHAM and OVX groups. IF sustainably reduced mechanical, hot, and cold hyperalgesia. Additionally, IF remodeled the gut microbiota and metabolite profile, with a marked increase in A. muciniphila and its metabolite IPA. Microbiota depletion and exogenous IPA supplementation experiments further confirmed the critical role of the Akkermansia–IPA axis in mediating the analgesic effects of IF.

Dietary strategies have a profound impact on pain and health^10,37–39. Inflammation and oxidative stress are the main pathophysiological pathways of chronic pain^37,38. Accumulating evidence suggests that IF can attenuate central and peripheral inflammation and oxidative damage, thereby modulating nociceptive sensitivity and reducing pain perception^37,40. Both animal and human studies have shown that IF can alleviate musculoskeletal pain and visceral pain^{12,14,41–44}. A previous study has also reported that IF significantly enhanced morphine-induced anti-nociception in the thermal tail-flick and postoperative paw incision model in mice⁴⁵. In our study, OVX mice subjected to IF exhibited antinociceptive actions in response to mechanical, thermal, and cold stimuli. Moreover, evidence shows that estrogen deficiency-induced hyperalgesia is linked to alterations in the expression of pain-related molecules (e.g., TRPV1, CGRP, and substance P) in DRG neurons^24,25,46. We found that IF reduced the number of TRPV1-positive neurons and CGRP expression in the DRGs of OVX mice. Taken together, these findings suggest the potential utility of IF in the comprehensive management of hyperalgesia and pain induced by estrogen deficiency.

Estrogen deficiency is associated with alterations in the composition and abundance of the gut microbiota in both menopausal women and OVX animals²³. Transplantation of an intact microbiota from sham-operated donors markedly alleviates hyperalgesia in OVX mice, indicating the contribution of the microbiota to menopausal pain²⁶. Disturbance of the gut microbiota is also implicated in the pathogenesis of visceral hypersensitivity¹⁸ and other types of chronic pain, including inflammatory pain¹⁹, neuropathic pain²⁰, headache²¹, stress-induced hyperalgesia⁴⁷ and opioid tolerance²². Evidence shows that IF can boost the diversity of the gut microbiota and enrich beneficial bacterial species⁴⁸. Dietary interventions such as IF or caloric restriction may mitigate chronic pain and chronic stress through modulation of gut microbiota⁴⁹. In this study, we found that IF not only alleviated hyperalgesia but also reshaped gut microbiota in OVX mice. Antibiotic-induced microbiota depletion largely abolished the analgesic effects of fasting. Conversely, FMT from fasted donors restored microbial composition and alleviated hyperalgesia in OVX recipients. Taken together, these results suggest that IF alleviates hyperalgesia in OVX mice by remodeling the gut microbiota.

Metabolites produced by gut microbes act as crucial mediators of host-microbial crosstalk²⁷. Evidence shows that microbiota-derived metabolites, such as short-chain fatty acids and amino acids, can regulate neural signaling molecules and pathways, thereby influencing the progression of chronic inflammatory pain and visceral pain^50,51. Notably, IPA, a tryptophan metabolite, is produced by gut bacteria such as A. muciniphila, Clostridium sp, E. coli, Peptostreptococcus and Fournierella^30–32,52. Our results showed that IF induced a marked enrichment of A. muciniphila and an elevated level of serum IPA in OVX mice. Administration of A. muciniphila not only increased IPA production but also alleviated hyperalgesia in OVX mice. It has been demonstrated that IPA is effective in ameliorating bone loss induced by estrogen deficiency⁵³ and also enhances axonal regeneration following sciatic nerve crush²⁸. Recent research has shown that IPA exhibits neuroprotective activity and alleviates pain-like behaviors in diabetic rats⁵⁴ and the complete Freund’s adjuvant-treated mice⁵⁵. Our study demonstrated that exogenous IPA supplementation markedly alleviated hyperalgesia in OVX mice. This analgesic effect was accompanied by the reduction of the number of TRPV1-positive neurons and CGRP expression in the DRGs of OVX mice. Hyperexcitability of DRG neurons contributes to the development of pain²⁵. Importantly, we found that IPA significantly reduced the excitability of DRG neurons from OVX mice. Collectively, these findings suggest that gut metabolite IPA mediates the protective effect of IF against estrogen deficiency-induced hyperalgesia.

However, this study has several limitations. First, while alternate-day fasting was effective, the efficacy of other intermittent fasting regimens (e.g., modified alternate-day fasting or time-restricted feeding) in alleviating menopausal hyperalgesia remains unexplored. Second, although the observed analgesic effect appeared independent of sustained caloric reduction, the impact of dietary composition under an IF protocol warrants further investigation. Third, whether IF can ameliorate other menopausal symptoms, such as vasomotor disturbances or mood changes, is still unknown. Fourth, electrophysiological measurements were taken from the same OVX neurons before and after IPA application, rather than from neurons of mice subjected to chronic IF or IPA treatment. This limits the strength of evidence regarding the sustained effects of these interventions. Additionally, the observed decreases in TRPV1-positive neurons and CGRP expression in DRGs, though consistent with behavioral improvement, are correlative findings. The study did not include functional assays (e.g., with capsaicin) to directly establish a causal link between these molecular changes and the alleviation of hyperalgesia. Finally, the translational potential of an alternate-day fasting regimen from mice to humans may be limited due to species-specific physiological and behavioral differences.

In summary, our findings show that the IF regimen alleviates hyperalgesia in OVX mice by remodeling the gut microbiota and its metabolites. This work identifies IPA as a potential therapeutic target, paving the way for novel clinical strategies against menopausal pain.

Methods

Animal experiments

Female C57BL/6J mice (7–8 weeks old) were obtained from the Animal Experimental Center of Nanfang Hospital, Southern Medical University. Following a one-week acclimation period under specific pathogen-free conditions (22 ± 2 °C, 12-h light/dark cycle) with free access to food and water, the mice underwent OVX surgery as previously described²⁴. After a one-week recovery period, the mice were randomly assigned to either an ad libitum-fed group (OVX group) or an intermittent fasting group (OVX + IF group). Mice in the OVX + IF group were subjected to alternate-day fasting over a 24-h cycle for a total duration of 7 weeks. Under this protocol, food was withheld on fasting days, while on feeding days, standard laboratory chow was provided daily at 10:00 AM. Water was available ad libitum throughout the entire experimental period for all animals. Furthermore, a separate sham-operated group (SHAM group), which underwent a comparable surgical procedure without ovariectomy and received ad libitum feeding, was included as a control. To minimize confounding variables related to social hierarchy and feeding behavior, mice from the same experimental group were co-housed together, while different experimental groups were housed in separate cages. Body weight was recorded, and behavioral tests were conducted weekly. At the end of the intervention, fecal samples, serum samples, and dorsal root ganglia (DRGs) were collected for subsequent analysis. All surgical procedures were performed under isoflurane inhalation anesthesia. For tissue collection, experienced technicians euthanized the mice via cervical dislocation under isoflurane-induced anesthesia to ensure rapid and complete separation of the cervical vertebrae from the skull. All experimental procedures were approved by the Animal Ethics Committee of Nanfang Hospital (approval No. IACUC-LAC-20241108-001) and were conducted in accordance with the Guide for the Care and Use of Laboratory Animals, as well as relevant institutional and national regulations. Animal experiments adhered to the ARRIVE guidelines 2.0, with blinding implemented throughout the study: data collection and data analysis were conducted independently by different personnel, all of whom were blinded to group allocation.

Antibiotics (ABX) treatment

To deplete the gut microbiota, some OVX mice received a broad-spectrum antibiotic cocktail via daily oral gavage (100 μl per mouse at 10:00 a.m.). The cocktail, prepared in ddH₂O, contained ampicillin (200 mg/kg; Macklin, China), metronidazole (200 mg/kg; Macklin, China), neomycin sulfate (200 mg/kg; Macklin, China), and vancomycin (100 mg/kg; OKA, China)⁵⁶. OVX mice were randomly divided into three groups: ad libitum (AL), IF, and IF combined with antibiotics (IF + ABX). Prior to ABX treatment, mice in the IF group had already undergone 3 weeks of IF and continued with the IF regimen in the subsequent protocol. Starting from week 3, the IF + ABX group received daily antibiotic administration for 1 week, while the other two groups maintained their original regimens. Behavioral tests were conducted both before and after antibiotic treatment. Finally, all mice were sacrificed for sample collection.

FMT

Prior to FMT, recipient OVX mice underwent ABX treatment as described above. Fresh fecal samples were collected daily from donor OVX and OVX + IF mice (OVX and OVX + IF groups) at 10 a.m., homogenized in PBS at a concentration of 0.125 g/ml⁵⁷, and promptly administered to recipients via oral gavage (200 µL/mouse). Each FMT procedure was completed within 30 min after sample preparation. FMT was administered daily for 7 days. Throughout this period, all mice had access to a standard diet and water. Behavioral tests were conducted weekly. Mice were euthanized for tissue collection and subsequent analysis.

Indole-3-propionic acid (IPA) treatment

The IPA (Macklin, China) solution was prepared by dissolving the compound in sterile PBS at a concentration of 2.5 mg/ml as previously described²⁸. Mice in the OVX + IPA group received this solution via oral gavage at a dose of 20 mg/kg per day.

Administration of A. muciniphila

The bacterium Akkermansia muciniphila (ATCC BAA-835) was obtained from the Guangdong Microbial Culture Collection Center. They were cultured anaerobically at 37 °C in brain heart infusion (BHI) medium supplemented with 0.5% porcine mucin (Sigma-Aldrich, USA) according to the manufacturer’s instructions. Species identification results were shown in Fig. S1. The bacterial concentration was measured at 600 nm absorbance⁵⁸. OVX mice were orally administered 0.5 mL of PBS with or without 8 × 10⁸ colony-forming units (CFU) of A. muciniphila.

Assessment of pain behaviors

Mechanical and thermal (hot and cold) sensitivity were assessed as previously described^24,25. First, the mechanical sensitivity was measured using an Electronic Von Frey (e-VF, electronic Von Frey, Ugo Basile, Italy). Mice were placed in a transparent plastic enclosure on an elevated metal frame for a 15-min acclimation period. A polypropylene filament was then applied perpendicularly to the center of the plantar surface of the hind paw, with gradually increasing force until a withdrawal response was observed. A positive response was defined as abrupt paw retraction, licking, or shaking. This procedure was repeated once every 15 min for a total of five trials, and the average value was recorded as the paw withdrawal threshold for that individual mouse. Second, the thermal sensitivity was evaluated using a hot/cold plate (ZH-6C Plate) to measure the latency of hind paw withdrawal in response to thermal (50 °C) or cold (4 °C) stimulation. Mice were placed on the temperature-controlled plate, enclosed within a circular Plexiglas wall to prevent escape, and allowed to calm down before testing. The latency to a pain-like behavior (such as paw licking, stomping, jumping, or escape attempts) was recorded. To prevent tissue injury, a 20-s cut-off was enforced. The test was repeated once every 15 min for a total of five trials, and the mean latency across trials was calculated as the paw withdrawal latency. Behavioral testing and data collection were performed by an investigator blinded to group assignment, while unblinding and statistical analysis were conducted by a separate researcher who was also blinded to group allocation.

CFU counting of feces

The CFU counting of mouse feces was determined as previously described⁵⁹. Briefly, 200 mg of fecal samples from each group were collected into sterile tubes containing 1 mL of sterile PBS. After homogenization, the mixture was serially diluted 1000-fold, and 40 μL of the dilution was inoculated onto TSA plates. The plates were incubated overnight at 37 °C, and the colonies were counted.

Immunofluorescence staining

Following 4–6 h of fixation, the DRGs were dehydrated in 30% sucrose for 16–18 h, embedded in OCT compound, and flash-frozen at −80 °C. Serial 10-μm sections were prepared from the OCT-embedded tissues using a cryostat (Leica CM1950). The sections were first blocked with 10% goat serum for 30 minutes at room temperature, followed by an overnight incubation at 4 °C with the following primary antibodies: transient receptor potential vanilloid 1 (TRPV1) (1:100, Absin, abs134462), calcitonin gene-related peptide (CGRP) (1:100, CST, 14959 T), and NeuN (1:250, CST, 94403S). After washing, the sections were incubated for 1 h at room temperature with secondary antibodies: goat anti-rabbit Alexa Fluor 488 (1:200, Abcam, ab150077) and goat anti-rat Alexa Fluor 594 (1:200, Abcam, ab150160). We harvested L3-L5 DRGs from each mouse. Fluorescence images captured by Zeiss Imager D2 microscope were analyzed using ZEN 3.8 software, and the arithmetic mean intensity of each image was recorded. The arithmetic mean intensity values from three fields of view per mouse were averaged to represent the data for that individual mouse (i.e., at least three sections), and inter-group comparisons were performed using the averaged values from each mouse.

LC-MS/MS analysis

The analysis of gut-derived metabolites was performed using a liquid chromatography-mass spectrometry (LC/MS) system consisting of a Waters Acquity I-Class PLUS UHPLC coupled with a Waters Xevo G2-XS QTof mass spectrometer. Separation was achieved on a Waters Acquity UPLC HSS T3 column (1.8 µm, 2.1 × 100 mm) with an injection volume of 1 µL. For both positive and negative ionization modes, the mobile phases consisted of (A) 0.1% formic acid in water and (B) 0.1% formic acid in acetonitrile. Data acquisition was conducted in MSe mode via MassLynx V4.2 software, with a scan rate of 0.2 s per spectrum. The collision energy was set to 2 V for low-energy scans and ramped from 10 to 40 V for high-energy scans. The electrospray ionization (ESI) source parameters were configured as follows: capillary voltage at 2000 V (positive mode) and −1500 V (negative mode), cone voltage at 30 V, source temperature at 150 °C, desolvation gas temperature at 500 °C, cone gas flow at 50 L/h, and desolvation gas flow at 800 L/h. Data processing and related analyses were carried out using BMKCloud (www.biocloud.net).

High-throughput 16S ribosomal RNA gene sequencing

Fecal samples were collected prior to sacrifice and immediately frozen at −80 °C. Total genomic DNA was extracted using the TGuide S96 Magnetic Soil/Stool DNA Kit (Tiangen Biotech, Beijing, China). The V3–V4 hypervariable region of the bacterial 16S rRNA gene was amplified with the primers 338F (5′-ACTCCTACGGGAGGCAGCA-3′) and 806R (5′-GGACTACHVGGGTWTCTAAT-3′). High-quality sequences were clustered into operational taxonomic units (OTUs) at a 97% similarity threshold using USEARCH (v10.0). Taxonomic classification of OTUs/ASVs was conducted with the Naive Bayes classifier in QIIME2 based on the SILVA database (release 138.1) under a 70% confidence threshold. Alpha diversity was assessed to evaluate within-sample species diversity, while beta diversity was analyzed via principal coordinate analysis (PCoA) to compare microbial communities between samples. Differential abundance of bacterial taxa was examined using linear discriminant analysis effect size (LEfSe). All sequencing data were processed with the BMKCloud platform (https://www.biocloud.net).

Enzyme-linked immunosorbent assay (ELISA)

The quantification of serum indole metabolites, including indole propionic acid (IPA), indole-3-acetic acid (IAA) and indole-3-lactic acid (ILA), was performed with commercial ELISA kits specific for each analyte (Bioroyee Biotech, Inc., Beijing, China). According to the manufacturer’s protocol, six hippocampal samples from each group were assessed in triplicate.

Primary neuron cells culture

DRG neurons were isolated and cultured as described previously²⁵. Briefly, DRGs were dissected from OVX mice (n = 6) and transferred into ice-cold DMEM/F12 medium. The ganglia were then subjected to enzymatic digestion at 37 °C for 30 min with 0.25% trypsin and type I collagenase (2 mg/ml; Bio-sharp, China). After enzymatic treatment, the tissue was gently triturated using a Bartholin tube to generate a single-cell suspension. The dissociated cells were plated onto culture dishes pre-coated with poly-D-lysine (100 ng/ml) and incubated for 24 h. Subsequently, the culture medium was replaced with Neurobasal medium (Pricella, China, PM151223) containing NGF (100 ng/ml; Merck, Germany) and B27 supplement (Gibco, USA, 17504044).

Whole-cell patch-clamp recordings

The excitability of DRG neurons was evaluated as described previously²⁵. Whole-cell current-clamp recordings were carried out at room temperature using a MultiClamp 700B amplifier and a DigiData 1440A digitizer (Molecular Devices). Borosilicate glass patch pipettes were pulled to a tip resistance of 3 to 6 MΩ when filled with an internal solution composed of (in mM): 120 K-gluconate, 20 KCl, 1 CaCl₂, 2 MgCl₂, 11 EGTA, 10 HEPES-K, and 2 MgATP, with pH adjusted to 7.2 using Tris-base and osmolarity maintained between 290 and 300 mOsm with sucrose. The extracellular solution contained (in mM): 145 NaCl, 3 KCl, 2 MgCl₂, 2 CaCl₂, 10 glucose, and 10 HEPES, with pH adjusted to 7.4 using NaOH. After a stabilization period (within 3 min), the resting membrane potential was recorded under current-clamp mode. Action potentials (APs) were evoked by injecting depolarizing current steps (500 ms duration), incremented in 5 pA increments via the recording electrode. The number of APs evoked by a suprathreshold stimulus was evaluated by applying a depolarizing current at rheobase intensity. Throughout the experiments, currents were low-pass filtered at 3 kHz, and signals were digitized at 10 kHz using pCLAMP 10 software. Only cells with a stable series resistance (Rs) of <25 MΩ were included for analysis. When Rs was between 20 and 25 MΩ, series resistance compensation (50–70%) was applied to minimize voltage errors. Recordings were excluded if Rs changed by >20% during the experiment. The series resistance was continuously monitored for each neuron, and recordings were excluded if the series resistance varied by more than 20%. Data were recorded again after treatment with neuronal culture medium containing IPA (20 μM). All data analysis were performed using Mini Analysis software (Synaptosoft).

Statistical analysis

All data were presented as mean ± standard error of the mean (SEM). Two groups were compared with the unpaired Welch’s t-test. Whole-cell data were compared with paired t-tests. Behavioral data and three-group comparisons were analyzed by one-way ANOVA followed by Bonferroni post-tests. A P value <0.05 was considered statistically significant. All analyses were conducted using GraphPad Prism (version 10.4).

Author contributions

R.-F.A.: Conceptualization, methodology, data curation, visualization, and writing original draft. Y.-S.H.: Investigation, validation, and data curation. Y.-T.H.: Investigation, validation, and formal analysis. H.-R.Y.: Data curation and writing —review & editing. S.-C.W.: Data curation. H-Z.L.: Formal analysis and writing—review & editing. J.-W.G.: Methodology, writing—review & editing, and funding acquisition. C.T.: Writing—review and editing. J.-S.Z.: Formal analysis, writing—review & editing, and funding acquisition. Z.-M.Z.: Conceptualization, methodology, supervision, writing—review and editing, project administration, and funding acquisition.

Data availability

Data will be made available on request. The raw sequence data reported in this paper have been deposited in the Genome Sequence Archive (Genomics, Proteomics & Bioinformatics 2025) in the National Genomics Data Center (Nucleic Acids Res 2025), China National Center for Bioinformation/Beijing Institute of Genomics, Chinese Academy of Sciences (GSA: CRA032772), which are publicly accessible at https://ngdc.cncb.ac.cn/gsa.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41522-026-00990-0.