C57BL/6 mice (16–20 g, 6–8 weeks old, equal numbers of males and females) were purchased from Chengdu Dossy Experimental Animals Co., Ltd. (Chengdu, China) and housed under standard laboratory conditions (12-h light/dark cycle, 22 ± 2°C, free access to food and water). The experimental groups were structured as follows: Experiment 1 sought to identify the target miRNA, its downstream bacterial species, and metabolic pathways involved in EA treatment of FC using multi-omics data. Mice were divided into four groups: normal control (NC) group, FC group, FC + EA group, and FC + EA + Abx group (n = 8 per group). Experiment 2 investigated whether EA’s regulation of gut motility and microbiota in FC mice is mediated by host-derived miR-205-5p. Adeno-associated virus serotype 9 with knockdown miR-205 (AAV9-mir205) was administered via tail vein injection to determine its potential to antagonize the effects of EA. Mice were randomly allocated into the AAV9-ctrl+FC + EA group (n = 8) and AAV9-mir205 + FC + EA groups (n = 8). One mouse in the AAV9-mir205 + FC + EA group died within 24 h of receiving the tail vein injection of the AAV9-mir-205. The potential causes were hypothesized to include stress from the injection procedure, tail vein injury, or individual variability in dose sensitivity to AAV9. To mitigate similar occurrences, tail vein injection techniques were optimized, and postoperative health monitoring of mice was intensified in subsequent experiments. Experiment 3 examined the effects of AAV9-mir-205 on gut motility, microbiota, and metabolism in normal mice. Mice were randomly divided into the AAV9-ctrl group and AAV9-mir205 group (n = 8 per group). All animals were assigned to their respective groups through a computer-generated randomization sequence. Researchers conducting the experiments and analyzing data were blinded to group assignments. All experimental procedures were approved by the Animal Ethics Committee of Chengdu University of Traditional Chinese Medicine (2021-59) and conformed to the guidelines of the National Institutes of Health (NIH) for the care and use of laboratory animals. The detailed experimental process is illustrated in Figures 1A, 2A, 3A, respectively.

A suspension of loperamide hydrochloride (0.98 mg/mL) was administered via gavage to establish the FC model at a dose of 9.8 mg/kg body weight, twice daily for 14 days (Li et al., 2023). During the EA intervention period, the same dose was administered once daily in the morning, 30 min prior to the intervention, to maintain the model for 12 days. Preliminary experiments confirmed that once-daily administration during the 12-day intervention period was sufficient to maintain the constipation model. Mice in the FC + EA + Abx group received an antibiotic cocktail to establish a pseudo-germ-free model. Neomycin sulfate, ampicillin, metronidazole and vancomycin hydrochloride were mixed at concentrations of 1, 1, 1, and 0.5 g/L, respectively, with the solution replaced twice a week. On the 15th day, fecal parameters were measured and compared to those of the NC group to determine whether the FC model was successfully established.

EA treatment was conducted as previously described (Wang L. et al., 2023; Wang M. et al., 2023). Acupoints Tianshu (ST 25) and Shangjuxu (ST 37) were selected for this study. The mice were anesthetized with isoflurane gas before treatment. Stainless steel acupuncture needles (0.25 mm in diameter, 13 mm in length, Hwato, Suzhou, China) were inserted unilaterally into specific acupoints at a depth of 4–5 mm. The needles were then connected to an EA device (SDZ-V, Hwato, Suzhou, China) using alligator clips. The EA stimulation parameters were set as following: 1 mA, 3/15 Hz alternating frequency, 30 min duration. Mice in the EA treatment groups received stimulation once daily for 2 weeks, with a 2-day rest period after every 5 treatments. EA treatment was conducted at a consistent time each morning (7 a.m. to 9 a.m.), while the other two groups underwent anesthesia only during the same time frame.

Recombinant AAV9 vectors encoding miR-205-5p or the control were sourced from Saiye Biotechnology Co., Ltd. The viral titer was specified as 2E+13 GC/mL.

The viral suspension was diluted in sterile saline to a final concentration of 5E+12 GC/mL prior to injection. After 7 days of acclimatization feeding, mice were administered approximately 0.2 mL of AAV9 suspension via the tail vein injection under isoflurane gas anesthesia. Following injection, the mice were monitored until full recovery from anesthesia and were then returned to their home cages.

Each mouse was administered 200 μL of the Lactobacillus reuteri DSM 17938 drops (#22DB336, BioGaia, Sweden), equivalent to 1 × 10⁸ CFU of Lactobacillus reuteri, once daily. A sterile, flexible gavage needle (20-gage, 1.5-inch length) was used for gavage, carefully inserted into the esophagus to deliver the bacterial suspension directly into the stomach.

Following a 12-h fasting period, each mouse was administered 0.1 mL of activated charcoal suspension via gavage. The mice were subsequently placed in individual cages, and the time of first black stool defecation was recorded (Wang L. et al., 2023; Wang M. et al., 2023).

Genomic DNA was extracted from fecal samples, with its concentration and purity assessed using electrophoresis and a NanoDrop 2000 spectrophotometer. Library construction and sequencing were outsourced to Majorbio Bio-Pharm Technology Co., Ltd. (Shanghai, China). Full-length PCR amplification of the 16S rRNA gene was conducted using 338F (5′-ACTCCTACGGGAGGCAGCAG-3′) and 806R (5′-GGACTACHVGGGTWTCTAAT-3′) primers, followed by library construction and sequencing. All sequencing data were uploaded to the Majorbio Cloud Platform for further analysis (Ren et al., 2022). Sequencing results were clustered into operational taxonomic units (OTUs) at 97% sequence similarity using UPARSE 7.1, with chimeras removed during the process. Taxonomic annotation of OTUs was performed using the RDP classifier with the Silva 16S rRNA gene database at a confidence threshold of 70%. Partial Least Squares Discriminant Analysis (PLS-DA) based on OTUs was employed to evaluate microbial community structures similarity between samples. Community composition differences between groups were analyzed at the genus level based on species abundance means within each group. Differential species analysis was performed using Linear Discriminant Analysis Effect Size (LEfSe) analysis to identify significant differences in abundance across phylum to species level, with thresholds of LDA > 3 and p < 0.05. The Kruskal-Wallis rank sum test, with FDR correction, was applied to evaluate differences between multiple groups.

Post hoc multiple comparisons were conducted using “Tukeykramer,” with significance set at p < 0.05.

Total RNA extraction, quantitative determination and quality control, library construction and sequencing were performed by Kangcheng Biological Co., Ltd. (Shanghai, China). Subsequently, miRDeep2 software was utilized to align sequencing results to the miRBase v22 database and calculate read counts. Differentially expressed miRNAs between groups were analyzed using DESeq2 package in the R software, with thresholds set at adjust p-value ≤ 0.05 and |Log2FC| ≥ 0.5.

Total RNA was extracted from fecal and colonic samples for miRNA analysis using RNAiso Plus (#9109, Takara). RNA was reverse-transcribed into miRNA cDNA synthesis using All-in-One miRNA First-Strand cDNA Synthesis Kit 2.0 (#QP114, GeneCopoeia, China). Quantification of miRNA cDNAs was conducted via real-time PCR using Hieff^® qPCR SYBR Green Master Mix (Low Rox Plus) (#11202ES08, GeneCopoeia, China) on ViiA 7 Real-Time PCR System (ABI, United States), following the manufacturer’s protocol. Bacterial DNA for gut bacteria analysis was extracted from feces using the E.Z.N.A.^® Soil DNA Kit (Omega Bio-Tek, United States) with quantity and quality assessed using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific). Quantitative PCR (qPCR) was conducted on an ABI7300 fluorescence quantitative PCR system (Applied Biosystems, United States) using ChamQ SYBR Color qPCR Master Mix (2X) (Novozymes, China). The specific Primer sequences were listed in Supplementary Table S1.

Metabolites extraction from feces, untargeted liquid chromatography–tandem mass spectrometry (LC–MS/MS) analysis, and data preprocessing and annotation were outsourced to Majorbio Bio-Pharm Technology Co., Ltd. (Shanghai, China). All data were uploaded to the Majorbio Cloud Platform for subsequent analysis (Ren et al., 2022). Metabolic profiling was conducted using orthogonal partial least squares discriminant analysis (OPLS-DA) for anions and cations, with Paletto conversion applied for data transformation. The model was validated using a 200-cycle permutation test. Differential metabolite screening identified significant metabolites between groups (NC group vs. FC group, FC group vs. FC + EA group, FC + EA group vs. FC + EA + Abx group), based on variable importance projection (VIP) >1 from OPLS-DA analysis, p-value < 0.05, and fold change (FC) >1 from Student’s t-test. Pathway enrichment analysis was conducted by mapping common differential metabolites to the KEGG database (Kyoto Encyclopedia of Genes and Genomes, http://www.genome.jp/kegg/). Multiple testing correction was performed using the “Bonferroni” method, and “Relative Betweenness Centrality” was applied as the topological method. Likewise, targeted metabolomics analysis of fecal samples was outsourced to Shanghai Majorbio Bio-pharm Technology Co., Ltd. (Shanghai, China). Fecal metabolites in tryptophan pathway were characterized using UHPLC–MS/MS, as detailed in previous publications (Cheng et al., 2022; Liu et al., 2024a).

The colon tissue samples were dehydrated and fixed prior to embedding in OCT compound. Tissue sections were prepared using a cryostat (Thermo Fisher, United States), and stained with hematoxylin and eosin (H&E) to observe morphological changes.

Sections were washed with PBS (pH 7.4) and incubated with DAPI. GFP expression in the mucosa and submucosa was visualized using a panoramic pathology scanner (PanoVIEW VS200, Olympus, Japan) (Ma et al., 2022).

Unless otherwise specified in the methods sections, all statistical analyzes were conducted using Prism 8 (GraphPad Software, San Diego, CA). Multiple group comparisons were conducted using least significant difference (LSD) and Kruskal-Wallis rank-sum tests based on the results of normality and homogeneity of variance tests. Similarly, two-group comparisons were analyzed using unpaired t-tests, Welch’s corrected unpaired t-tests, or Mann–Whitney non-parametric tests, except for the comparison between the AAV9-mir205 group and the AAV9-mir205 + LR group, which was performed using a paired t-test. A p-value of < 0.05 was considered statistically significant.

To evaluate the therapeutic effect of EA treatment on loperamide-induced constipation in mice, gut motility was assessed using the defecation time for the first black stool, 6-h fecal pellet count, wet weight. After a 14-day loperamide intervention, the FC group exhibited a significantly longer defecation time for the first black stool compared with the NC group. Similarly, the fecal pellet count and wet weight over 6 h significantly reduced in the FC group compared to the NC group (Figure 1B; Supplementary Figure S1). Thus, loperamide successfully induced gastrointestinal transit disorder in FC model mice. After 2 weeks of EA treatment, all gut motility indicators significantly improved in the FC + EA group, compared to the FC group (Figure 1B; Supplementary Figure S1). In addition, the FC + EA + Abx group mice, treated with a broad-spectrum antibiotic in drinking water, exhibited a significantly longer defecation time for the first black stool and reduced 6-h fecal pellet count and wet weight compared to the FC + EA group (Figure 1B; Supplementary Figure S1). These findings suggest that gut microbiota mediates the positive effects of EA on gut motility, consistent with previous reports (Xu et al., 2023).

Intestinal microbial alterations were investigated using 16S rDNA sequencing of feces collected from each group of mice. Alpha diversity analysis revealed significant differences between the FC + EA + Abx group and all three other groups (Supplementary Figure S2A). Similarly, microbial beta-diversity analysis showed that the microbial structure in the FC + EA + Abx group was distinct from the other groups, indicating that antibiotics dramatically affected gut microbiota composition (Figure 1C). PLS-DA also demonstrated that the NC group was clearly distinguishable from the FC and FC + EA groups, clustering into different taxa (Figure 1C). Despite some overlap in microbial structure between the FC and FC + EA groups, significant differences were observed in the relative abundance of specific intestinal microbes. To identify the specific intestinal microbes that were influenced by EA treatment, we compared the relative abundance of gut microbial species identified in the feces of each group of mice. At the genus level, the FC + EA + Abx group was dominated by Klebsiella spp., norank_f__Muribaculaceae spp., and Bacteroides spp., whereas the other groups were dominated by norank_f__Muribaculaceae spp., Lactobacillus spp., and Dubosiella spp. (Figure 1D). Notably, loperamide modeling reduced Lactobacillus abundance in mouse feces, which was restored following EA treatment (Figure 1D). Additionally, broad-spectrum antibiotics negated the positive effects of EA on Lactobacillus. LEfSe was employed to identify differential species between the NC group and FC group, FC group and FC + EA group, and FC + EA group and FC + EA + Abx group (Figure 1E; Supplementary Figure S3). The intersection of all species-level differences from the three comparisons was analyzed. As depicted in the Venn diagrams (Figure 1F), three differential species were identified: s_uncultured_bacterium_g_Allobaculum, s_Lactobacillus_reuteri, s_Lactobacillus_johnsonii. The relative abundance of these three differential species were significantly reduced in the FC group compared to NC group. However, only the abundance of Lactobacillus reuteri was restored following EA treatment (Figure 1G). Compared to the FC + EA group, the relative abundance of Lactobacillus reuteri was significantly lower in the FC + EA + Abx group, suggesting that EA’s regulation of Lactobacillus reuteri was inhibited by antibiotics. Previous studies have shown that administrating Lactobacillus reuteri DSM 17938 alleviates clinical symptoms in patients with constipation (Riezzo et al., 2019).

Previous studies have demonstrated altered fecal miRNA profiles in FC patients compared to healthy individuals (Yao et al., 2023). Host intestinal epithelial cells and homeodomain-only protein homeobox (HOPX)-positive cells are primary sources of miRNAs that influence the growth of intestinal luminal microbiota (Zhang C. et al., 2023; Zhang X. et al., 2023). To identify host-derived miRNAs regulated by EA in alleviating loperamide-induced constipation, small RNA sequencing was performed on colonic tissues from mice in the NC, FC, and FC + EA groups. PCA revealed that the FC group formed distinct clusters from the FC + EA and NC groups, while the NC and FC + EA groups overlapped (Figure 4A). The analysis showed that 11 miRNAs were upregulated and 4 were downregulated in the NC group compared to the FC group, while 14 miRNAs were upregulated and 9 were downregulated in the FC + EA group compared to the FC group (Figure 4B; Supplementary Tables S2, S3). Among these changes, 9 miRNAs (mmu-miR-615-3p, mmu-miR-10a-3p, mmu-miR-10b-3p, mmu-miR-10b-5p, mmu-miR-10a-5p, mmu-miR-672-5p, mmu-miR-205-5p, mmu-miR-224-5p, and mmu-miR-708-5p) were upregulated, while 3 miRNAs (mmu-miR-490-3p, mmu-miR-490-5p, and mmu-miR-148a-5p) were downregulated in the FC group compared to the NC or FC + EA groups (Figure 4C). Combined with prior sequencing of fecal miRNAs from FC patients, miR-205-5p appears to be a key miRNA in EA treatment (Yao et al., 2023). Increased miR-205-5p levels in colon tissue and feces of the FC + EA group compared to the FC group were confirmed by qPCR (Figure 4D). Antibiotic treatment significantly elevated miR-205-5p expression levels in colon tissues and feces of the FC + EA + Abx group compared to the FC group (Figure 4D). This suggests that removal of the intestinal flora does not alter EA’s effect on miR-205-5p expression in the intestinal lumen.

To investigate the relationship between gut microbiota and colonic miRNAs, we correlated the 12 differentially expressed miRNAs with the top 20 gut microbial species ranked by total abundance. As shown in Figure 4E, the three downregulated miRNAs were significantly negatively correlated with Lactobacillus reuteri abundance, whereas the nine upregulated miRNAs, including miR-205-5p were significantly positively correlated with Lactobacillus reuteri abundance. Thus, host-derived miR-205-5p may influence the growth of Lactobacillus reuteri, contributing to the regulation of intestinal function.

To determine whether miR-205-5p influences the therapeutic effects of EA in loperamide-induced constipation, AAV9-mir205 was synthesized and administered via tail vein injection (Figure 2A). The results showed a significant prolongation of the first black stool defecation time in the AAV9-mir205 + EA + FC group (Figure 2B). Histopathological examination revealed no significant pathological damage to colonic tissues in either group, suggesting that AAV9 exhibited minimal toxicity (Figure 2C). Immunofluorescence analysis indicated that GFP⁺ cells were present in the colonic tissues of both groups, primarily in the mucosal and submucosal layers, with sparse distribution in the muscularis propria (Figure 2D). qPCR analysis revealed significantly reduced miR-205-5p expression in colon tissue and fecal samples of AAV9-mir205 + FC + EA group compared to the AAV9-ctrl+FC + EA group (Figure 2E). Tail vein injection of AAV9 successfully targeted the colonic tissues of mice and was functionally effective. 16S rDNA sequencing was performed to analyze overall microbial composition. While alpha diversity showed no significant differences (Supplementary Figure S4), PLS-DA analysis of beta diversity revealed that the fecal microbial structure in the AAV9-mir205 + FC + EA group was distinct from that in the AAV9-ctrl+FC + EA group (Figure 2F). At the genus level, intestinal flora in both groups was dominated by Lactobacillus spp., norank_f__Muribaculaceae spp., and Ileibacterium spp. The relative abundance of Lactobacillus spp. was significantly lower in the AAV9-mir205 + FC + EA group (20.7%) compared to the AAV9-ctrl+FC + EA group (26.2%) (Figure 2G). LEfSe analysis showed significant enrichment of several strains, including Lactobacillus reuteri, in fecal samples from the AAV9-mir205 + FC + EA group compared to the AAV9-ctrl+FC + EA group (Figure 2H). qPCR results indicated a significant reduction in the absolute expression of Lactobacillus reuteri in feces from AAV9-mir205 + FC + EA group (Figure 2I). Thus, the AAV9-mir205 inhibited the effects of EA on gut motility and Lactobacillus reuteri in mice with loperamide-induced constipation.

No prior studies have investigated the role of miR-205-5p in the regulation of intestinal motility and Lactobacillus reuteri. AAV9-mir205 was first administered to untreated mice by tail vein injection, showing minimal effects on their body weight (Figure 3B). Since the onset of AAV9 typically occurs at 7 days post-injection, intestinal motility was assessed from day 7 onwards. The defecation time for the first black stool was significantly prolonged on days 7 and 10 post-injection in AAV9-mir205 mice compared to controls (Figure 3C). Results of HE staining and immunofluorescence staining were consistent with previous findings, indicating that AAV9 successfully targeted the mucosal epithelium of colon tissue without disrupting the tissue structure (Figures 3D,E). On day 10 post-administration, miR-205-5p levels in feces were lower in AAV9-mir205 mice than in controls, confirming its role in intestinal motility regulation (Figure 3F). To investigate the role of miR-205-5p in shaping the intestinal microbiota, alpha diversity showed no significant differences, whereas beta diversity revealed some separation between the groups (Supplementary Figure S5; Figure 3G). At the genus level, both AAV9-ctrl and AAV9-mir205 groups were dominated by Lactobacillus spp. and norank_f__Muribaculaceae spp., with relative abundances of 77.8% vs. 30.2% and 9.5% vs. 34.6%, respectively (Figure 3H). LEfSe analysis identified significant differences at multiple species levels, with the abundance of species, such as Lactobacillus reuteri, Lactobacillus intestinalis, and Lactobacillus_johnsonii, significantly reduced in the miRNA-205 group compared to controls (Supplementary Figure S6). qPCR results confirmed a significant reduction in the absolute expression of Lactobacillus reuteri in feces from AAV9-mir205 mice (Figure 3I). Diminished gut motility in miR-205 knockdown mice was associated with reduced Lactobacillus reuteri expression in the intestine.

To confirm the role of Lactobacillus reuteri in miR-205-5p-mediated regulation of intestinal motility, AAV9-mir205 group mice were gavaged with Lactobacillus reuteri DSM17938 (BioGaia), forming the AAV9-mir205 + LR group. As shown in the Figure 3J, pairwise comparisons indicated that the defecation time for the first black stool was significantly reduced on day 12 and day 15 of consecutive Lactobacillus reuteri gavage compared to pre-gavage. Immunofluorescence results confirmed persistent colonization of colonic tissues by AAV9-mir205 (Figure 3E). Pairwise comparisons with the AAV9-mir205 group revealed no statistically significant difference in fecal miR-205-5p expression levels in the AAV9-mir205 + LR group (p > 0.05, Figure 3K). These results suggest that the miR-205 knockdown effect of AAV9 was sustained. Gavage of Lactobacillus reuteri significantly altered the microbial structure, as indicated by 16S rDNA sequencing of feces (Supplementary Figure S5). Species difference analysis showed increased abundance of Lactobacillus reuteri, Akkermansia muciniphila, and Bifidobacterium animalis in the feces of the AAV9-mir205 + LR group compared to the AAV9-mir205 group (Supplementary Figure S7). As seen in the Figure 3L, qPCR further confirmed increased Lactobacillus reuteri abundance in feces. These findings demonstrate that Lactobacillus reuteri successfully colonized the intestine and counteracted the inhibitory effect of AAV9-mir205 on gut motility.

Previous studies suggest that gut bacteria regulate intestinal motility through their metabolites (Waclawikova et al., 2022). To identify flora-regulation-related fecal metabolites affected by EA in relieving constipation, we conducted LC–MS untargeted metabolomics analysis on feces from loperamide-induced constipated mice with or without broad-spectrum antibiotic treatment post-EA. After preprocessing, differences in fecal metabolic profiles across the four groups were analyzed using two methods, PLS-DA and OPLS-DA, respectively. As shown in the Figure 5A, the PLS-DA model for anions and cations revealed overlapping fecal metabolic profiles between the NC group and FC + EA group, which were distinct from FC group. In contrast, the FC + EA + Abx group displayed clearly distinct fecal metabolic profiles compared to the other three groups. OPLS-DA analysis identified significant differences in fecal metabolic profiles between groups (NC group vs. FC group, FC group vs. FC + EA group and FC + EA group vs. FC + EA + Abx group, Supplementary Figure S8). Further screening identified 271 shared differential fecal metabolites between groups, as shown in Figure 5B. These metabolites were primarily associated with flavone and flavonol biosynthesis, tyrosine metabolism, tryptophan metabolism, histidine metabolism (Figure 5C). Notably, both in vivo and in vitro experiments have demonstrated that Lactobacillus reuteri metabolizes and activates tryptophan (Cervantes-Barragan et al., 2017; Montgomery et al., 2022). Association analysis from this study showed that Lactobacillus reuteri, one of the top 20 species by total abundance, was significantly linked to several tryptophan metabolites (Figure 5D).

One-way ANOVA followed by Tukey–Kramer tests identified differential tryptophan metabolites (p < 0.05) across the four groups (Figure 5E). Compared to the NC group, the FC group exhibited reduced levels of 4 tryptophan metabolites (Formyl-5-hydroxykynurenamine, 5-Hydroxy-N-formylkynurenine, 5-Hydroxyindoleacetate, Indole-3-Carboxaldehyde) in the feces, while 2-Aminophenol levels increased. Further comparison revealed that EA treatment reduced 2-Aminophenol levels while increasing Formyl-5-hydroxykynurenamine, 5-Hydroxyindoleacetate, and Indole-3-Carboxaldehyde levels in feces of FC model mice. Additionally, antibiotic treatment inhibited EA’s regulatory effects on these 4 tryptophan metabolites in the feces of FC mice. These findings suggest that EA’s regulation of tryptophan metabolites in the feces of constipated mice is dependent on gut microbiota.

We investigated the impact of miR-205 knockdown on the EA-regulated tryptophan metabolites in the feces of loperamide-induced constipated mice. Compared to the AAV9-ctrl+FC + EA group, AAV9-mir205 + FC + EA group showed significantly reduced levels of picolinic acid (PCL 016), N-(3-indolylacetyl)-L-alanine, rac-kynurenine, kynurenic Acid, indole-3-carboxaldehyde, indole-3-acetic acid, and indole-3-acetamide (all p < 0.05), while tryptamine levels were significantly elevated (p < 0.01; Figure 6; Supplementary Table S4). Further analysis explored whether the regulation of these metabolites was associated with Lactobacillus reuteri. Transfection of normal mice with AAV9-mir205 resulted in significantly reduced levels of indole-3-acetamide, indole-3-carboxylic acid, indole-3-carboxaldehyde, indole-3-acetic acid, and tryptophol in the feces (all p < 0.05), whereas indole levels significantly increased (p < 0.05; Figure 6; Supplementary Table S5). After Lactobacillus reuteri gavage in AAV9-mir205 mice, fecal levels of 2-aminophenol, rac-kynurenine, xanthurenic acid, kynurenic acid, and 5-hydroxyindole-3-acetic acid were significantly reduced (all p < 0.05), while L-tryptophan, indole-3-acetamide, indole-3-acetic acid, and indole-3-carboxaldehyde levels significantly increased (all p < 0.05; Figure 6; Supplementary Table S6). Notably, Lactobacillus reuteri restored the levels of indole-3-acetamide, indole-3-acetic acid, and indole-3-carboxaldehyde reduced by AAV9-mir205 infection. Thus, the improvement of gut motility in loperamide-induced constipated mice by EA treatment, through increased levels of these 3 tryptophan metabolites, may involve host-derived miR-205-5p promoting the growth of Lactobacillus reuteri.