1. Introduction
Chronic gut inflammation, also known as inflammatory bowel disease (IBD), is marked by enduring irritation that harms the digestive tract, encompassing the conditions of Crohn’s disease and ulcerative colitis [
1]. These ailments are typified by a decline in the quantity of goblet cells in the colon and a reduction in mucus synthesis [
2]. Goblet cells positioned at the epithelium are specialized cells that produce mucus, which lubricates and protects the gut [
3]. The mucosal barrier lines the intestinal tract and is a layer of viscous gel consisting of mucin glycoproteins, water, lipids, and ions [
4]. The mucin glycoproteins (primarily MUC2) are the structural components that form the matrix of the mucosal barrier [
5]. The mucosal barrier prevents the adhesion and invasion of harmful microorganisms that can damage the epithelium [
6], while also providing attachment sites and nutrients for interactions with beneficial gut microbiota [
7]. Such microbiota used the mucosal barrier as an ecological niche for growth and promote the digestion and uptake of nutrients, as well as modulating immune responses [
8].
Although various drugs—including antibiotics, corticosteroids, and immunomodulators—are available for clinically managing IBD symptoms, a considerable proportion of patients are unable to mount a suitable response to standard treatments or experience loss of efficacy and off-target adverse effects from long-term drug use [
9]. Therefore, innovative approaches for managing IBD are sought. Fecal microbiota transplantation (FMT)—that is, transferring fecal bacteria and other microbes from a healthy individual into another individual as a treatment—is now being investigated as an approach for managing refractory IBD and has shown considerable potential as a promising alternative to standard drug therapies, especially for patients with chronic IBD that have become irresponsive to standard medications [
10].
Prior research has shown that administering polyphenols orally to rodents can enhance the presence of mucin in their fecal matter, indicating that these compounds somehow induce the microbiome (or host tissues) to modulate the mucosal barrier physiology, including anti-inflammatory impacts [
11]. Grape seed proanthocyanidins are common dietary polyphenols that have been identified as natural agents with excellent anti-inflammatory activities [
12]. Research findings have demonstrated that, while a specific group of B-type procyanidin oligomers can be absorbed by the small intestine, the majority of these oligomers make their way to the colon, where they undergo fermentation by the microbiota present in the colonic environment [
13], [
14]. Previous studies have examined the utility of treating experimental colitis with procyanidins (and their analogs)—including procyanidin [
15], procyanidin A1 [
16], and procyanidin B2 [
17]—and respectively attributed the observed therapeutic benefit to activation of the signal transducer and activator of transcription 3 (STAT3) and nuclear factor-kappa B (NF-κB) pathways, the activated protein kinase (AMPK)/mammalian target of rapamycin (mTOR)/p70S6K signaling pathway, and a combination of the nuclear factor (erythroid-derived 2)-like 2 (Nrf2) and Wnt/
β-catenin signaling pathways. Administration of proanthocyanidins has been shown to inhibit digestive enzymes and alter microbiota composition [
18]. Procyanidin C1 (PCC1) is a B-type proanthocyanidin (comprising three (−)-epicatechin units joined by two successive (4β-8)-linkages) found in grape and other plants, with demonstrated anti-inflammatory activity in mice [
19].
Here, we report that PCC1 confers protective benefits against a classic DSS-induced mouse model of IBD. A PCC1 pretreatment conferred both anti-inflammatory effects and protection against multiple pathological phenotypes in DSS-induced IBD model mice, including reduced weight loss, reduced disease activity index (DAI) levels, and enhanced colon size, as well as preventing degradation of the mucosal barrier. 16S sequencing showed that the PCC1 pretreatment shifts the composition of colonic microbiota; for example, elevating the abundance of Akkermansia muciniphila and Christensenella minuta. Mice that received FMT from PCC1-treated mice phenocopied the protective benefits against IBD, again displaying obvious mucus layer thickening. Gas chromatography-mass spectrometry (GC-MS)-based metabolic profiling revealed that both the PCC1-pretreatment and the PCC1 FMT groups had elevated levels of the microbiota-derived metabolite valeric acid (VA) and that VA supplementation conferred strong protection against IBD. Experiments combining VA supplementation with an inhibitor of FOXO1 showed that FOXO1 signaling in goblet cells of the intestinal epithelium mediates the benefits of VA. Our study establishes that PCC1 (and attendant FMT) treatment and VA supplementation are promising interventions to modulate the mucus layer to treat IBD.
2. Methods
2.1. Animal experimentation
For this investigation, we procured male specific-pathogen-free (SPF) C57BL/6J mice aged 8 weeks from Charles River Laboratories (China). These mice were housed in a controlled SPF barrier system, and all animal experiments were conducted while adhering to the guidelines established by the Animal Care Committee of China Agricultural University. The primary objective of our study was to assess the protective effects of PCC1 (> 98% purity; Sigma Aldrich, USA) treatment in mice afflicted with colitis. We categorized the mice into three groups: ➀ a control group (CON); ➁ a dextran sulfate sodium (DSS-IBD) group, in which the mice were administered sterile water with 2.5% DSS to induce colitis; and ➂ a PCC1-pretreatment IBD model (DSS-PCC1) group. We collected fresh fecal samples from the mice and immediately snap-froze them in 1.5 mL Eppendorf (EP; Thermo Fisher Scientific, USA) tubes using dry ice. The samples were stored at -80 °C for FMT. To remove the primitive microbiota and prepare the mice for the FMT studies, we administered four non-absorbable antibiotic combinations (penicillin, neomycin, vancomycin, and metronidazole) to the SPF mice via gavage. In the FMT experiments, mice were pretreated with antibiotic combinations for 7 days prior to the FMT procedure. On the following day, FMT was conducted using thawed fecal material: each antibiotic combination-pretreated recipient received a 200 μL suspension via oral gavage using animal-feeding needles. In addition, 100 μL of the suspension was applied to the fur of each mouse. For short-chain fatty acid (SCFA) supplementation, 200 mmol·L
−1 each of VA and acetic acid (AA) was administered in the drinking water, following a previously described protocol [
20]. FOXO1 (AS1842856; Selleck Chemicals, USA), phosphate-buffered saline (PBS; Sigma Aldrich), tubulin (Sigma Aldrich).
2.2. DSS-induced colitis model for evaluation using the DAI
To establish a colitis model, SPF C57BL/6J mice were administered sterile water containing 2.5% DSS for a duration of 6 days. The DSS solution was replaced every 48 h to maintain its freshness. We closely monitored various parameters, including food consumption, body weight, mental status, stool consistency, and bleeding, daily during the induction of colitis to track disease progression. In order to assess the progression of colitis in mice, we employed the DAI, which incorporates three objective parameters: ➀ weight loss (a score of 0 for no weight loss, 1 for 1%-5% weight loss, 2 for 5%-10% weight loss, 3 for 10%-15% weight loss, and 4 for weight loss > 15%); ➁ diarrhea (scored as 0 for normal stools, 2 for loose stools, and 4 for watery diarrhea); and ➂ the presence of blood in stools (scored as 0 for absent, 2 for slight bleeding, and 4 for heavy bleeding). The DAI score was determined by the summation of the individual scores for weight loss, diarrhea, and the presence of blood in stools.
2.3. RNA purification and cDNA preparation
Trizol was used to homogenize the mouse colon tissues, and chloroform was added to separate the total RNA from the homogenate. The resulting RNA was utilized to synthesize complementary DNA (cDNA) using the cDNA Synthesis SuperMix kit for quantitative polymerase chain reaction (qPCR) from Servicebio Technology Co., Ltd. (China), which contains reverse transcriptase, random primers, and dNTPs to promote efficient cDNA synthesis. We used equal amounts of total RNA for the reaction to ensure consistent cDNA yields. The resulting cDNA could be used for a variety of downstream applications, including qPCR and gene expression profiling.
2.4. Quantitative real-time PCR
Expression analysis was conducted through qPCR with 2× SYBR Green qPCR Master Mix (Servicebio Technology Co., Ltd.) using a CFX Connect Instrument (Bio-Rad Laboratories, Inc., USA). The reactions were 10 μL in volume. The polymerase chain reaction (PCR) cycling conditions were as follows: a single hold at 95 °C for 30 s, followed by 40 cycles at 95 °C for 15 s, 65 °C at 10 s, and 72 °C at 30 s. The reaction volume was 10 μL (98-well format), with 10 μmol·L−1 of each primer and cDNA. The data were collected using CFX Manager software (Bio-Rad Laboratories, Inc.). Relative gene expression was assessed using the ΔΔCT method, which involves a comparison of the cycle threshold (CT) of the gene of interest with that of a reference gene—in this case, reduced glyceraldehyde-phosphate dehydrogenase. To ensure the specificity of the PCR primers, we evaluated them by performing a melting curve analysis.
2.5. Histopathology
We adhered to an established set of instructions to investigate the tissue-level damage of the DSS-induced colitis mouse colons. First, we preserved tissue integrity by fixing the colon in 4% paraformaldehyde. Next, we gradually dehydrated the tissue by washing it with 30%, 50%, and 70% ethanol for 30 min each. We embedded the colon in 2% agar and stored it in 70% ethanol for future use. To prepare the colon for microscopic analysis, we performed paraffin embedding and used a precision tissue slicer to slice the tissue into 5 μm sections. Subsequently, we performed staining using hematoxylin and eosin (H&E) and blindly examined the resultant slides. The histological scoring of the colon tissue was determined by evaluating four crucial aspects: the magnitude and intensity of inflammation, the extent of crypt damage, and the proportion of tissue affected.
2.6. Alcian blue tissue analysis
To measure the thickness of the mucus layer in the colon, a modified methodology was employed. The colonic samples were subjected to fixation in Carnoy’s buffer for a duration of 3 h, followed by washing with methanol. The fixed samples were then placed in cassettes and stored in fresh methanol for further experiments. To visualize the mucus layer, alcian blue (AB) staining was performed, and the stained colon samples were thoroughly scanned using a panoramic camera. The thickness of the mucus layer was quantified using ImageJ software (National Institutes of Health, USA), which was employed to measure the thickness across two or three representative areas of each sample.
2.7. DNA extraction and gut microbial community profiling via 16S ribosomal RNA sequencing
To evaluate the impact of PCC1 intervention on the gut microbiota in a mouse model of acute colitis induced by DSS, fecal samples were obtained and subjected to bacterial DNA extraction. Amplification of the V3-V4 hypervariable region of the 16S ribosomal RNA (rRNA) gene was performed, followed by sequencing of the resulting amplicons on an Illumina MiSeq platform (Illumina Inc., USA) using paired-end sequencing with 2 × 300 bp read lengths. The initial processing of the raw reads was carried out with the DADA2 (GNU Project, USA) algorithm on the QIIME2 platform (QIIME2; QIIME, USA) to generate high-quality amplicon sequence variants (ASVs) with a minimum frequency threshold of 10, which were retained for downstream analysis.
2.8. Untargeted metabolites
Stool samples were collected from the mice and frozen in liquid nitrogen before being transferred to a -80 °C freezer. For SCFA concentration analysis, the contents, combined with 1 mL of water, were added to 2 mL EP tubes and vortexed for 10 s. The supernatant was further vortexed and exposed to ultrasound (on ice) for 10 min. After another round of centrifugation (15 min, 10 000 r·min-1, 4 °C), the supernatant was moved into a new 2 mL glass tube for analysis using a GC-MS (GC2030-QP2020, Shimadzu Co., Ltd., Japan), with an HP-FFAP capillary column (Agilent Technologies, USA).
2.9. Statistical analysis
The objective of this investigation was to examine the variances in structural dissimilarities within the 16S rRNA gene sequence data, in addition to evaluating the alpha diversity indexes and weighted UniFrac distances. The significance of these variances was assessed using a Kruskal-Wallis rank-sum test. Enzyme activity was assessed using a repeated-measures analysis of variance (ANOVA) with the aid of the “nlme” software package. Statistical analysis was performed using R version 4.2.0, and the data were represented as the mean ± standard error of the mean (SEM). Statistical analysis involved one-way ANOVA followed by a Tukey test. Statistical significance was determined using an unpaired Student’s t-test, with * p < 0.05, ** p < 0.01, *** p < 0.001, and **** p < 0.0001 indicating the level of significance. Any result that fell below these cutoffs was considered to be statistically significant.
3. Results
3.1. PCC1 pretreatment confers protective benefits against IBD
We investigated the potential therapeutic utility of PCC1 for treating IBD using a classic murine model based on 6 days of exposure to 2.5% DSS [
21]. For PCC1, we opted for a pretreatment approach in which mice were given PCC1 for 7 days prior to model induction (50 mg per kilogram daily via intraperitoneal injection;
Fig. 1(a)). After inducing the IBD model, an evaluation of the phenotypes on day 6 showed the anticipated differences in body weight and DAI levels, which included factors such as weight loss, diarrhea, and the presence of blood in the stool [
22]. The evaluation also compared the length of the colon with that of the control animals not used in the experimental model (
Figs. 1(b)-(d)). The IBD model mice exhibited significantly reduced body weights (
Fig. 1(b);
p < 0.0001), higher DAI scores (
Fig. 1(c);
p < 0.0001), and shorter colon lengths (
Fig. 1(d);
p < 0.0001).
Analysis of the post-model induction showed significant weight loss for the IBD model mice in both the DSS-PCC1 and DSS-IBD groups, as compared with the non-model control animals. The DSS-PCC1 group had significantly lower DAI scores than the non-model mice (
Fig. 1(c)). An examination of the dissected/resected colons from animals sacrificed on day 6 revealed a noteworthy elongation of the colon in the DSS-PCC1 group when compared with DSS-IBD and CON group (
Fig. 1(d)). However, there were no discernable disparities in colon length between the DSS-PCC1 and CON group (
Fig. 1(d)). Histopathology at day 6 after model induction revealed that the non-model mice exhibited multiple serious features of colitis that were not evident in the DSS-PCC1 group, including loss of epithelial crypts, inflammation, and ulceration (
Fig. 1(e)).
We also used qPCR to assess the levels of dysregulated genes in the colonic tissue, as reported in previous studies of DSS model mice. For example, infiltrating macrophages are known to produce proinflammatory cytokines during IBD pathogenesis [
23]. While induction of the IBD model led to the expected elevation of interferon-
γ (IFN-
γ), tumor necrosis factor-
α (TNF-
α), and human interleukin-6 (IL-6) expression, PCC1 pretreatment significantly reduced the extent of induction for these three proinflammatory cytokines (
Fig. 1(f)). These findings collectively show that PCC1 pretreatment confers protective effects against DSS-induced IBD in mice.
3.2. PCC1 pretreatment prevents degradation of the mucosal barrier
Both humans and mice have a mucosal barrier in their digestive tract [
24]; this layer is secreted by goblet cells and includes the oligomeric mucus gel-forming protein MUC2, which organizes as a net-like skeleton [
25]. Previous studies [
26], [
27] of IBD patients and studies of DSS mice have reported pathogenic thinning of the mucosal barrier. Moreover, MUC2-deficient mice spontaneously develop colitis and are relatively more sensitive to DSS induction [
28]. We assessed the potential impacts of PCC1 pretreatment on the mucosal barrier by staining with AB, which can detect sulfated and carboxylated acid mucopolysaccharides alongside sulfated and carboxylated sialomucin molecules [
29]. Compared with the non-model mice, the PCC1-pretreatment mice had an obviously thicker mucus layer (
Fig. 2(a)). Moreover, immunoblotting against MUC2 in extracts from the colon tissue samples showed that the PCC1-pretreatment colons had significantly more MUC2 (
Fig. 2(b)). Notably, neither the thickness nor the MUC2 phenotypes of the PCC1-pretreatment mice differed from those of the CON group (
Fig. 2(b)).
Previous study [
30] has enzymatically demonstrated the roles played by multiple enzymes in dynamically regulating the composition of the mucosal barrier, including α-fucosidase and β-
N-acetylglucosaminidase, which DSS- recognize the O-glycan moieties on mature MUC2. Given the thicker mucosal layer we observed in the DSS-PCC1 group, we reasoned that the activities of mucus-degrading enzymes would be highest in the non-model mice. Quantification of the activities of β-
N-acetylglucosaminidasein and α-fucosidase assays of fecal extracts showed that the non-model mice had significantly elevated activities in comparison with the PCC1-pretreatment group (
Fig. 2(c)).
Since FOXO1 has been reported to regulate mucin secretion by colonic goblet cells [
31], we used immunoblotting to assess the FOXO1 amounts present in intestinal epithelial cells. It was found that the PCC1-pretreatment group exhibited markedly higher FOXO1 levels compared with the IBD model and non-model mice (
Fig. 2(d)). We also analyzed the autophagy marker LC3 and found—in line with previous reports on FOXO1’s functions in mucus secretion—that the level of LC3 was significantly elevated in the intestinal epithelial cell samples of the PCC1-pretreatment group in comparison with the non-model mice samples (
Fig. 2(d)). Collectively, these findings support the suggestion that reduced degradation of the mucosal barrier contributes to PCC1’s protective benefits against DSS-induced IBD.
3.3. PCC1 pretreatment shifts the composition of colonic microbiota
Given previous reports [
32] of microbiome-altering impacts from proanthocyanidins, we conducted 16S sequencing for a fecal microbiome analysis. We characterized the impacts of the PCC1 pretreatment on the comparative quantity of bacteria present at the phylum clade, noting a dramatic increase Verrucomicrobia after PCC1 pretreatment as compared with the non-model mice samples (
Fig. 3(a)). A previous clinical study [
33] reported decreased Verrucomicrobia in individuals with IBD compared with healthy individuals. When analyzing the fecal microbiome at the class level, the PCC1-pretreatment mice exhibited a greater relative level of Verrucomicrobiae, Betaproteobacteria, and Erysipelotrichia compared with the IBD model and non-model mice (
Fig. 3(b)). At the family level, the PCC1-pretreatment microbiome had higher Sutterellaceae, Verrucomicrobiaceae, Erysipelotrichaceae, Rikenllaceae, and Prevotellaceae (
Fig. 3(c)). Notable differences in the PCC1-pretreatment samples included a decreased abundance of
Helicobacter typhoons, which has been shown to drive excess TNF-
α production to promote colitis development [
34], and increased
Akkermansia muciniphila abundance (
Figs. 3(d) and
(e)).
3.4. Fecal transplants from the PCC1-pretreatment mice phenocopy the protective benefits against IBD
FMTs—that is, transferring fecal bacteria and other microbes from a healthy individual into another individual—are increasingly routine in both research bioscience and clinical medicine [
35]. We reasoned that, if the protective impacts observed in the PCC1-pretreatment animals are mediated by alteration of the microbiome, then we would expect an from PCC1-exposed healthy donor mice to endow some protection upon induction of the DSS-model in PCC1-naive recipient mice. We designed an experiment in which antibiotics-treated C57BL/6J mice were used as the recipients, while the donors were mice given an oral gavage of either PCC1 (PCC1 donor) or PBS (CON donor) (
Fig. 4(a)). At 7 days after the FMT, we exposed both groups of recipient mice to 2.5% DSS to induce the IBD model.
Compared with the non-model mice, the PCC1-recipient mice had significantly milder IBD symptoms, including a reduced extent of weight loss (
Fig. 4(b)), lower DAI scores (
Fig. 4(c)), and longer colons (
Fig. 4(d)). An analysis of colon tissues stained with H&E indicated that the PCC1-recipient mice had significantly decreased histological scores (
Fig. 4(e)). A qPCR analysis of the colon tissues showed that the levels of transcripts encoding IL-6, IL-1β, and TNF-
α—which are markers of inflammation—were observed to be at substantially lower levels in the PCC1-recipient mice compared with the non-model mice (
Fig. 4(f)). Thus, the protective effects initially observed upon PCC1 pretreatment were indeed mediated by the microbiome.
Analyzing the bacterial community of the recipient mice using 16S rRNA gene sequencing 7 days after the FMT revealed significant modifications in the microbiota compositions of the PCC1-recipient mice, compared with those of the non-model mice. These alterations were observed at the phylum, class, and family levels (
Figs. 5(a)-(c)). More specifically, the PCC1-recipient mice had increased relative abundance of
Akermansia maciniphila, Blautia producta, Christensenella minuta,
Parabacteroides distasonis,
Hydrogenoanaero blacterium,
Enterorhabdus caecimuris,
Faecalicoccus pleomorphus,
Helicobacter hepaticus,
and Clostridium aldenense (
Figs. 5(d) and
(e)). Beyond reinforcing that the microbiome changes induced by PCC1 deliver protective and anti-inflammatory impacts, these findings establish that FMT is a successful means for delivering a protective microbiome into PCC1-naïve animals.
3.5. Fecal microbiota transplantation from PCC1-pretreatment mice resulted in mucus layer thickening
To examine the impact of PCC1 on the mucosal barrier in the colon, we utilized AB staining to evaluate the thickness of the mucosa. Our observations revealed that the mice receiving PCC1 exhibited a significantly thicker mucosal barrier than the non-model mice (
Fig. 6(a)). Consistent with the thicker mucus layer, we observed elevated MUC2 levels (
Fig. 6(b)), as well as dramatically decreased activities of both α-fucosidase and β-
N-acetylglucosaminidase (
Fig. 6(c)) in the PCC1-recipient mice.
Considering that goblet cells play a crucial role in producing various components of the mucus layers, we examined the colon tissues of the FMT-recipient mice to assess the involvement of host cells in the observed mucus layer thickening. In brief, a previous study [
36] reported FOXO1-knockout mice exhibited a thinned mucus layer and decreased mucin secretion from goblet cells; thus, it was notable—and consistent with our observations for the aforementioned PCC1-pretreatment experiments—that the PCC1 FMT recipients had significantly higher FOXO1 levels compared with the non-model mice (
Fig. 6(d)). Moreover, our observations of significantly higher LC3 levels in the PCC1-recipient colon epithelial cells support the suggestion that the induction of autophagy contributes to PCC1’s protective impacts against IBD.
3.6. Increased VA resulting from PCC1 intervention confers protective benefits against IBD via FOXO1 signaling
Untargeted GC-MS method was used to profile the metabolites in fecal samples from PCC1-pretreatment and DSS-control mice, which showed that PCC1 resulted in large-scale changes in metabolites (
Figs. 7(a) and
(b)). Notably, a dramatically increased VA level was detected in the PCC1-pretreatment group, compared with the DSS-control fecal samples, which was subsequently confirmed to be significant in a targeted quantitative analysis (
Fig. 7(c)). Importantly, a similar analysis of fecal samples from the FMT mice groups showed that the PCC1-recipient mice also had dramatically increased VA levels as compared with the non-model mice (
Fig. 7(c)). We therefore conducted VA supplementation experiments with DSS-model mice. To be specific, after 7 days of DSS induction, we supplemented mice with VA or with AA (
Fig. 7(d)). AA treatment group was a control, as the level of this other SCFA did not change in the metabolite profiling experiment. Compared with AA supplementation, VA supplementation resulted in significant improvements in colon length (
Figs. 7(e) and
(f)), body weight (
Fig. 7(h)), DAI (
Fig. 7(i)), and intestinal permeability (
Figs. 7(h) and
(j)). In addition, we noted significant reductions in the levels of proinflammatory cytokines (TNF-α, IFN-γ, and IL-6) following VA supplementation (
Figs. 7(k)-(m)).
Finally, given the known role of FOXO1 in regulating the secretion of mucin from goblet cells [
31], we used a FOXO1 inhibitor to assess the potential involvement of goblet cells and the mucus layer in VA’s protective effects. Administration of FOXO1 at the same time as VA completely abolished the benefit from VA supplementation, in terms of colon length (
Figs. 7(e) and
(f)), body weight (as exemplified in
Fig. 7(h)), disease activity (
Fig. 7(i)), and intestinal permeability (
Fig. 7(j)), demonstrating that FOXO1 signaling directly mediates VA’s protective effects against IBD.
4. Discussion
In this work, we have demonstrated that a grape seed procyanidin compound confers protection against DSS-induced IBD by preventing degradation of the mucus layer. We found that a PCC1 pretreatment enriches
Akkermansia muciniphila, a genus that is recognized for its ability to flourish within the mucus layer and that can provide protection to host intestinal cells [
37]. Our FMT experiments confirmed that a PCC1-induced alteration in the fecal microbiome was responsible for the observed benefits against IBD. We also found that both the PCC1-pretreatment and PCC1-recipient FMT animals had elevated levels of the microbiota-derived metabolite VA, and we showed that supplementation with VA protects against IBD. By coupling supplementation experiments with a small-molecule inhibitor, we verified that the beneficial effects of VA are mediated through FOXO1 signaling in host cells. Earlier investigations have shown a regulatory role for FOXO1 in the secretion of mucus by goblet cells [
31]. These studies have reported that mice lacking FOXO1 exhibit (among other phenotypes) defective goblet cell autophagy and microbiome dysbiosis. Our findings with a FOXO1 inhibitor directly implicate FOXO1’s regulation of the mucus layer in the anti-IBD benefits conferred by PCC1. Interestingly, a study has linked the dysbiosis resulting from the loss of FOXO1 to deficiencies in downstream microbiota-derived metabolites (including SCFAs) [
31]. This is a particularly notable finding in light of ➀ our observations about VA levels (and supplementation) and ➁ a clinical study reporting that IBD patients have lower levels of VA than healthy controls. It also bears mention that a study of radiation injury showed that increased VA promoted intestinal epithelial integrity, thus linking a downstream (microbiota-derived) metabolite to the cells that control mucus secretion [
38]. While obviously quite complex and difficult to isolate experimentally, it will be intriguing to gain further insights into how goblet-cell FOXO1 signaling may alter the mucus layer thickness, structure, and/or composition to reliably favor the production of specific metabolites from particularly taxa of commensal bacteria that in turn promote goblet cell health (
Fig. 8).
Human cell and animal model studies have extensively investigated the anti-inflammatory properties of proanthocyanidins [
15], [
16]. These studies have revealed that proanthocyanidins exert their effects through various pathways, such as NF-κB, mitogen-activated protein kinase (MAPK), and phosphatidylinositol-3-kinase (PI3K)/Akt pathways. Some compounds from grape seeds have been shown to interact with cell surface receptors (e.g., toll-like receptor 4 and purinergic receptor P2X7) to modulate immune responses. We found that both the PCC1-pretreatment and PCC1-recipient FMT mice exhibited diminished levels of the inflammatory markers IL-6, IL-1β, and TNF-α. The observed impacts of PCC1 on colon tissues and the microbiome could be mediated through interactions with host cells and/or through interactions with microbial cells, and the FMT and VA supplementation experiments support the suggestion that multiple aspects of PCC1’s benefit against IBD can be attributed to impacts that are downstream of any direct PCC1 action. It will be interesting to see how the anti-inflammatory effects of this and other proanthocyanidin compounds ultimately shape the host environment to promote mucus layer properties and fecal microbiome composition to favor gut health.
The observation of higher levels of
Akkermansia muciniphila in the PCC1-pretreatment samples make good sense in light of extensive previous research that has linked this species to improved intestinal barrier function and to protecting hosts from intestinal inflammation [
39], [
40]. As its Latin binomial implies, this species engages with mucin and thrives in the mucosal barrier [
41]. Somewhat counterintuitively, although
Akkermansia muciniphila is a mucin-degrading bacterium, it is now understood that an elevated
Akkermansia muciniphila abundance engages in a positive feedback loop that ultimately results in the enhanced protection of host intestinal cells [
42]. Moreover, fecal transplants with
Akkermansia muciniphila-containing microbiota have been reported to increase the mucus layer thickness, thereby supporting the reconstruction of the mucus layer [
43], [
44]. Ultimately, alterations observed in the gut microbiome caused by the PCC1 pretreatment (including
Akkermansia muciniphila) contribute to the observed protection against DSS-induced IBD by preventing the degradation of the mucus layer.
Christensenella minuta has been extensively characterized in the human colon, and its relative abundance has been positively correlated with lean individuals (i.e., low body mass index [
45]). It has been investigated as a probiotic intervention for treating obesity and a variety of metabolic diseases, including IBD. Studies examining the oral administration of
Christensenella minuta in rodents have reported that the interventions reduced colonic inflammation (in one case, reducing the extent of inflammatory lesions by 36%) by inducing an immunomodulatory response [
45].
Alteration of the gut microbiota has been repeatedly implicated in the therapeutic benefits conferred by a diversity of interventions against IBD. Beyond the microbiota-altering impacts of diverse natural products [
15] and drug-like molecules [
17], studies investigating inflammation site-specific targeted nanomedicines have also detected the consistent alteration of gut microbial taxa [
46]. It will be interesting to specifically track the fate of PCC1 molecules upon oral administration to help determine the relative contributions of host versus microbiome physiology on therapeutic impacts. It will also be interesting to examine how PCC1 confers the observed increase in the VA content in the gut, which could reflect the enrichment of VA-producing bacteria (potentially
Clostridium sp.) or maybe the result of interactions between colonies, as VA is known to influence the composition and behavior of microbial communities. Regardless of how PCC1 induces the VA increase, it bears repeated mention that this increase was evident in the PCC1-pretreatment mice and the FMT mice, and that inhibiting FOXO1 signaling blocked the beneficial impacts from VA supplementation.
Extensive studies and trials examining FMT for treating patients with IBD have been encouraging, including for patients in flare and for patients seeking to maintain remission. Our findings from the FMT experiments showed that a PCC1-induced alteration in the fecal microbiome was responsible for the benefit when treating the IBD model mice. Moreover, the FMT results added lines of evidence to our initial observations from the PCC1-pretreatment experiments supporting a thickened mucus layer, decreased inflammation, activation of autophagy, elevated FOXO1 signaling activity, and increased VA content. Thus, beyond tracking the fecal microbiome per se, future FMT studies and studies of PCC1 (or other proanthocyanidins) as interventions against IBD in human patients should monitor markers of inflammation and should consider measuring autophagy markers and/or profiling fecal SCFA levels.
5. Conclusions
The grape seed procyanidin compound PCC1 confers protection against DSS-induced IBD by preventing degradation of the mucus layer. PCC1 enriches Akkermansia muciniphila, a species that resides in the mucus layer, and promotes its turnover. Our FMT experiments showed that a PCC1-induced alteration in the fecal microbiome can account for the observed benefits against IBD. Both the PCC1-pretreatment and PCC1-recipient FMT animals had elevated levels of the microbiota-derived metabolite VA. Finally, we show that supplementation with VA protects against IBD. Supplementation experiments coupled with the inhibition of FOXO1 signaling verified that the beneficial effects of VA against IBD occur through this pathway.
Acknowledgments
We would like to express our gratitude to Professor Benjamin Willing from the University of Alberta for his thoughtful and helpful discussion of our work. This work was supported by the 111 projects of the Education Ministry of China (B18053), the National Natural Science Foundation (32130081), the National Key Research and Development Program of China (2022YFF0710402), and the Pinduoduo-China Agricultural University Research Fund (PC2023B01014).
Compliance with ethics guidelines
Xifan Wang, Pengjie Wang, Yixuan Li, Ran Wang, Siyuan Liu, Ju Qiu, Xiaoyu Wang, Yanling Hao, Yunyi Zhao, Haiping Liao, Zhongju Zou, Josephine Thinwa, and Rong Liu declare that they have no conflict of interest or financial conflicts to disclose.
PDF
(6182KB)
