Gluten-free diet reduces severity of dextran sulfate sodium-induced colitis in BALB/c mice

Abstract

BACKGROUND

Inflammatory bowel disease is a chronic inflammatory condition in which gut microbiota, intestinal permeability, and nutrition play critical roles. Dietary gluten and its major component gliadin exhibit pro-inflammatory, lipopolysaccharide-like properties that may contribute to intestinal inflammation.

AIM

To investigate whether gluten withdrawal, i.e., a gluten-free diet (GFD), attenuates inflammation in experimental colitis.

METHODS

We investigated the effect of a GFD in dextran sulfate sodium (DSS)-induced colitis using specific-pathogen-free BALB/c mice fed standard or gluten-free Altromin diets for 4 weeks prior to a 1-week exposure to 3% or 2.5% DSS. Both Altromin open-formula diets were comparable in total protein and energy content. Disease severity was assessed by colon length, intestinal bleeding, weight loss, histology, and flow cytometry analysis of colonic leukocytes and selected T-cell subsets in mesenteric lymph nodes and spleen.

RESULTS

Four-week pre-exposure to a GFD led to statistically significant preservation of colon length (P < 0.0001), decreased intestinal bleeding, and improved histology (crypt length and histology score; P < 0.0001). GFD-fed mice displayed a significant reduction in infiltrating colonic lymphocyte antigen 6C⁺ monocytes (P < 0.0001) and increased proportions of forkhead box P3⁺ regulatory T cells, interleukin-10⁺ T helper cells (P < 0.01), and γδ T cells (P < 0.05) in mesenteric lymph nodes, and to a lesser extent in the spleen, indicating an enhanced anti-inflammatory immune profile.

CONCLUSION

Long-term gluten withdrawal substantially reduced DSS-induced colitis in specific-pathogen-free BALB/c mice. These findings suggest that a GFD may represent a safe therapeutic strategy warranting further evaluation in clinical studies.

Key Words: Gluten-free diet; Inflammatory bowel disease; Colitis; Intestinal inflammation; Nutrition

Core Tip: A gluten-free diet (GFD) markedly attenuated dextran sulfate sodium-induced colitis in BALB/c mice, reducing clinical and histological disease severity. Gluten withdrawal decreased colonic lymphocyte antigen 6C⁺ monocyte infiltration and promoted anti-inflammatory immune responses, including increased proportions of forkhead box P3⁺ regulatory T cells and interleukin-10⁺ T helper cells, as well as γδ T cells. These findings highlight a previously underappreciated role of dietary gluten in intestinal inflammation and support a GFD as a potential non-pharmacological strategy for modulating immune responses in inflammatory bowel disease.

INTRODUCTION

Inflammatory bowel disease (IBD) refers to a group of disorders that affect the gastrointestinal tract and cause chronic, relapsing inflammation. The most prevalent forms of IBD are Crohn’s disease and ulcerative colitis (UC), both of which have shown a steadily increasing global incidence in recent decades[1,2]. The chronic and relapsing inflammatory condition of the gastrointestinal tract is thought to result from a complex interplay of genetic, immune, and environmental factors. The gut microbiota, its metabolites, changes in intestinal permeability, and dietary factors play a critical role in disease pathogenesis[3]. While substantial research interest in IBD has been devoted to probiotic bacteria or beneficial metabolites, considerably less attention has been given to dietary interventions[4], which have recently been more intensively studied[5].

Pharmacologic interventions involving immunosuppressive drugs such as steroids, immunomodulators, biologics, and small molecule inhibitors[6] carry significant risks, including infection, malignancy, and other toxic side effects[7,8]. However, these treatments primarily manage disease symptoms rather than address the underlying causes. Therefore, there is an increasing need to complement conventional treatments with non-pharmacological strategies. Because IBD affects the digestive tract, diet is considered one of the most suitable adjunctive solutions[9], as it significantly influences not only IBD onset and progression[10], but may also be associated with disease relapses[11].

In IBD, chronic intestinal inflammation is often referred to as a pathogenetic factor in genetically susceptible individuals[12,13]. Many years ago, including in our own studies, intestinal inflammation and factors contributing to an enhanced intestinal inflammatory environment were proposed as underlying features in a wide range of immune-mediated diseases[14,15]. This concept has recently attracted more attention in IBD, but also in diseases such as autoimmune hepatitis, gluten ataxia, autoimmune thyreoiditis, type 1 diabetes (T1D), multiple sclerosis (MS), Parkinson’s disease, and rheumatoid arthritis, based on several gut-organ bidirectional axes[16,17].

There is accumulating evidence for the beneficial effects of anti-inflammatory dietary components in IBD[5,9]. Gluten is a complex protein composed of gliadins and glutenins, proteins rich in glutamines and prolines that are resistant to protease degradation, allowing them to escape complete breakdown in the gastrointestinal tract[18]. Gluten peptides, particularly α-gliadin, have been found to be immunogenic and capable of triggering T-cell-mediated responses[19].

Gliadin has been shown to exhibit chronic pro-inflammatory properties similar to those of lipopolysaccharide (LPS)[20,21]. A gluten-free diet (GFD) is used not only in treating celiac disease or gluten allergies, but is also one of the strongest environmental factors in the prevention of T1D in a spontaneous non-obese diabetes (NOD) mouse model[22-24]. Dietary gluten is associated with pro-inflammatory immune mechanisms, which induce pro-inflammatory cytokines in patients with celiac disease[25,26], but also in the standard strain of immunocompetent BALB/c mice[27,28]. Interestingly, one animal study reported worsening signs of dextran sulfate sodium (DSS)-induced colitis by the addition of gluten to a standard mouse diet[29].

In IBD patients, gluten-free products were associated with symptom attenuation[30,31]. However, Zallot et al[32] and Schreiner et al[33] reported insufficient symptom improvement. A recent study by Gregersen et al[34] found no significant association between gluten intake and response to biological treatment in patients with chronic inflammatory diseases, including Crohn’s disease, UC, and rheumatoid arthritis. However, patients with high gluten intake reported a lower health-related quality of life compared with those with a low-to-moderate intake. These findings from human trials are not fully conclusive and highlight the need for further investigation into the relevance of a GFD in IBD management.

The aim of this study was to assess the effect of a GFD on the development of DSS-induced colitis in specific-pathogen-free BALB/c mice and to investigate morphological changes and selected immune parameters in this mouse model of UC.

MATERIALS AND METHODS

Animals, diets and experimental design

Inbred BALB/c mice were obtained from the animal facility of the Institute of Microbiology of the Czech Academy of Sciences (Prague, Czech Republic). Mice were maintained under specific-pathogen-free conditions in accordance with FELASA guidelines, under standard light- and climate-controlled conditions, with ad libitum access to food and water.

Experiments were conducted in compliance with European Directive 2010/63 on the protection of animals used for scientific purposes and were approved by the Institutional Animal Care and Use Committee of the Institute of Microbiology, CAS (approval No. 23-2022-P). Humane endpoint criteria were established to minimize suffering and ensure animal welfare.

Four-week-old female BALB/c mice were fed a standard diet (STD) (Altromin 5c) or a GFD (Altromin 1434 modified 5d) diet (Altromin, Lage, Germany), with compositions described previously[22,24]. In brief, these open-formula Altromin diets were designed to include the same content of milk and soybean proteins, whereas the gluten-containing ingredients were replaced with meat protein, while keeping (with respect to the open-formula character of the diets) the content of total protein, fat, fiber and minerals as similar as possible. The total protein content of the GFD and STD was 22.7% and 22.9%, respectively. Both diets were equally supplemented with vitamins.

Experimental groups of 4-week-old BALB/c mice (n = 8-10) were kept on the GFD and STD for a period of 4 weeks prior to colitis induction by DSS (36-50 kDa; MP Biomedicals, CA, United States) for 7 days. In the initial pilot experiment, an autoclaved 3% DSS water solution was used. In subsequent experiments, the concentration was reduced to 2.5% DSS. Mice were sacrificed 3 days after DSS exposure by isoflurane (IsoFlo, Zoetis CZ, Prague, Czech Republic) anesthesia followed by cervical dislocation, and samples were collected for follow-up analysis. Untreated controls maintained on the STD and GFD were used in parallel. The number of experiments, replicates and animals per group is indicated in the figure legends. Animals were randomly allocated to experimental groups prior to DSS treatment.

The severity of DSS-induced colitis was assessed by changes in body weight, occult intestinal bleeding (Hemoccult Test, Beckman Coulter Life Sciences, Indianapolis, IN, United States), and stool consistency, which were used to calculate the disease activity index (DAI) as previously described[35]. Colon length was measured and colon and lymphoid organ samples were collected for histology and flow cytometry.

Histological analyses

Distal colon samples were fixed for 30 minutes in Carnoy’s solution, transferred to 96% ethanol, and embedded in paraffin. Colonic sections (5-7 μm) were stained with hematoxylin and eosin (HE) or Alcian blue for mucin, followed by nuclear fast red counterstaining (Vector Laboratories, Newark, CA, United States). Subsequent microscopic evaluation of colon morphology and mucin production/goblet cell depletion was performed in a blinded manner by an independent investigator. The mucosal damage was first evaluated on HE-stained sections using the scoring system described by MacPherson and Pfeiffer[36], ranging from stage 0 (no colitis) to stage 3 (severe colitis with ulceration), and applied to colon sections from three distinct regions. Changes in colonic crypt length were assessed on well-positioned transverse sections from 3 colon locations > 400 μm apart, with > 10 individual values per animal (n = 10 per group) and used for calculating crypt length differences between the GFD- and STD-fed groups. Samples were examined using a Leica DM6000 Light microscope, and images were processed by the Leica Application Suite X software (Wetzlar, Germany).

Single cell preparation and flow cytometry

Single cell suspensions from mesenteric lymph nodes (MLN) were prepared by placing collected organs into Petri dishes with 3 mL of complete RPMI medium. The organs were then mechanically gently dissociated by pressing with fine tweezers against a rough part of a microscope slide. The microscope slide was washed with 400 μL of fluorescence activated cell sorter (FACS) buffer (1 × Dulbecco’s phosphate-buffered saline, 2% bovine serum albumin, 0.5 M ethylenediaminetetraacetic acid, sodium azide 0.05%) and samples filtered through a 70 μm cell strainer (BD Biosciences, San Diego, CA, United States) into test tubes. The absolute number of viable cells per sample was counted using trypan blue staining, a Burker chamber, and light microscope.

For the preparation of colonic infiltrating leukocytes, washed colons were cut into approximately 1-2 cm long pieces, and single cell suspensions were gently prepared according to the protocol by Couter and Surana[37]. Additional mice treated in parallel and randomized with those used in other assays were used to isolate colonic leukocytes.

For surface staining, cells were incubated with diluted fluorochrome-conjugated monoclonal antibodies (mAbs) for 30 minutes on ice. Fc block [cluster of differentiation (CD) 16/CD32; clone 2.4G2; immunoglobulin G-2b, κ] was purchased from BD Biosciences and added to reduce the Fc receptor-mediated binding. Live/dead discrimination was performed using eBioscience Fixable Viability Dye eFluor 506 (Thermo Fisher Scientific, Waltham, MA, United States) or by Hoechst 33342 staining immediately before acquisition. Golgi Stop (Cytofix/Cytopem kit, BD Biosciences) was added in all incubation and washing steps prior to fixation. Cells were then fixed and permeabilized using the eBioscience forkhead box P3 (Foxp3)/transcription factor staining buffer set (Thermo Fisher Scientific) according to the manufacturer’s instructions and finally re-suspended in FACS buffer for analysis by flow cytometry. Cells were labeled using fluorochrome-conjugated mAbs specific for the following markers: CD3 (145-2C11), CD4 (RM4-5), CD8a (53-6.7), CD11b (M1/70), CD11c (N418), CD45 (30-F11), CD49b (DX5), Foxp3 (FJK-16s), interleukin (IL)-10 (JES5-16E3), T cell receptor γδ (GL3), major histocompatibility complex class II (I-A/I-E; MS/114.15.2), Siglec-F (1RNM44N), lymphocyte antigen 6C (Ly6C) (HK1.4), Ly6G (1A8-Ly6 g). All mAbs were purchased from Thermo Fisher Scientific. Flow cytometric data were acquired using an LSR II flow cytometer (BD Bioscience) and analyzed with FlowJo software (Tree Star, Ashland, OR, United States).

Statistical analysis

Sample size was determined based on previous experiments with the DSS-induced colitis model in the laboratory, including previous and preliminary experiments showing expected differences in body weight loss and histological scores between groups, as well as previous flow cytometry analyses performed in animal model experiments. Statistical analyses were performed using GraphPad Prism 9 (GraphPad Software, La Jolla, CA, United States). Data are expressed as the mean ± SD. Statistical differences between two groups were assessed by unpaired t tests. One-way analysis of variance followed by Bonferroni’s or Tukey’s multiple comparison post-tests was applied for multiple-group analyses. Differences were considered statistically significant at P < 0.05.

RESULTS

GFD preserved colon length in DSS-induced acute colitis

In an initial pilot experiment, we tested the hypothesis that an open-formula, gluten-free Altromin diet (GFD), which has been shown to strongly prevent T1D in NOD mice[22], might also exert a beneficial effect on the development of DSS-induced colitis in mice, possibly due to its lower pro-inflammatory properties. Four-week-old BALB/c mice (females) were placed on the GF and ST Altromin open-formula diets for a period of 4 weeks and were then treated with 3% DSS in drinking water for a period of 7 days. BALB/c mice were used because their lower sensitivity to DSS-induced colitis may facilitate detection of subtle dietary effects. In this preliminary experiment, we observed a statistically significant (P = 0.017, n = 10) preservation of colon length in BALB/c females on a GFD (n = 10) compared with controls on the STD (n = 10), Figure 1A. Intestinal bleeding was reduced in the GFD group (P < 0.05), while body weight remained comparable between the two dietary groups after 7 days of acute colitis induction (Figure 1B and C). Although the diets differed slightly in total energy content, this did not translate into marked weight changes. By day 10 after DSS treatment, body weights were not significantly different; nevertheless, animals on the STD showed a slight trend toward lower values (19.0 ± 0.7 vs 19.5 ± 0.9) compared with their GFD-fed counterparts.

Figure 1 Gluten-free diet preserves colon length in dextran sulfate sodium-induced acute colitis. In the first pilot experiment, colitis was induced with 3% dextran sulfate sodium (DSS) for 7 days in female BALB/c mice that had been maintained for 5 weeks on a gluten-free diet (GFD) (n = 10) or a standard diet (STD) (n = 10). A: Preservation of colon length in BALB/c mice on the GF or ST diet after DSS treatment (P < 0.05); B: Reduced intestinal bleeding scores (P < 0.05); C: No substantial changes in body weight during DSS-induced colitis. Data are expressed as mean ± SD. ^aP < 0.05. DSS: Dextran sulfate sodium; STD: Standard diet; GFD: Gluten-free diet.

GFD reduces the severity of acute DSS-colitis in BALB/mice

To increase the sensitivity of the DSS-colitis model to subtle environmental influences, for example, by modifying only one protein source in the diet (replacing potential gluten-containing ingredients, i.e., grain, with meat protein), we reduced the DSS concentration to 2.5% in the next series of acute colitis experiments. This experimental design led to much more pronounced changes between the DSS-treated groups on different diets (Figure 2). In particular, a macroscopically visible preservation of colon length was noted. As shown in Figure 2A and D, highly significant preservation of colon length was observed in mice on the GFD compared with those on the STD (P < 0.0001, n = 10). Intestinal bleeding was also significantly decreased in the GFD group (P < 0.001, n = 10) (Figure 2C). There were only minor decreases in body weight during the 7 day course of 2.5% DSS, but the DAI index was statistically significantly lower in the GFD vs STD group (P < 0.01, Figure 2B). There were no substantial differences in colon length and intestinal bleeding between GFD- and STD-fed controls without colitis induction (no DSS) (n = 10; Figure 2A, C and E).

Figure 2 Gluten-free diet reduces severity of dextran sulfate sodium (2.5%)-induced acute colitis in BALB/c mice. Eight-week-old female BALB/c mice fed the gluten-free diet (GFD) (n = 10) and standard diet (STD) (n = 10) from 4 weeks of age received 2.5% dextran sulfate sodium (DSS) in drinking water for 7 days and were sacrificed 3 days later. BALB/c mice on the STD (n = 10) and GFD (n = 10) without DSS exposure were used as controls. A: The GFD led to statistically significant preservation of colon length (P < 0.0001); B: Decreased disease activity index (P < 0.01); and C: Reduced intestinal bleeding (P < 0.001, P < 0.0001) in DSS-treated animals; D: Macroscopic images of representative colons from GFD and STD groups treated with 2.5% DSS; E: Controls. Data are expressed as mean ± SD. ^bP < 0.01. ^cP < 0.001. ^dP < 0.0001. P values were assessed by Student’s t-test or by one-way analysis of variance followed by Bonferroni’s post-test. DAI: Disease activity index; DSS: Dextran sulfate sodium; STD: Standard diet; GFD: Gluten-free diet.

A repeated experiment (n = 10 for each dietary group) confirmed our findings: Colon length was again markedly shortened in the STD-fed mice compared with GFD-fed mice (P < 0.001, n = 10), intestinal bleeding increased on the STD (P < 0.01), and the DAI index was significant higher (P < 0.01) after 7 days of DSS treatment (Figure 3).

Figure 3 Effects of gluten-free diet on severity of dextran sulfate sodium (2.5%)-induced acute colitis in BALB/c mice. Repeated experiment as in Figure 2, groups of 10 BALB/c females were exposed to 2.5% dextran sulfate sodium (DSS) in drinking water for 7 days at 8 weeks of age. A: The gluten-free diet (GFD) led to statistically significant preservation of colon length (P < 0.001); B: Decreased disease activity index (P < 0.01); and C: Reduced intestinal bleeding (P < 0.01) compared with DSS-treated animals on the standard diet (STD); D: Macroscopic images of representative colons from GFD and STD groups. Data are expressed as mean ± SD. ^bP < 0.01. ^cP < 0.001. DAI: Disease activity index; DSS: Dextran sulfate sodium; STD: Standard diet; GFD: Gluten-free diet.

Colon histology

Histological evaluation of colon morphology on HE-stained, paraffin-embedded sections revealed inflammatory cell infiltration, swollen and thickened submucosa, epithelial damage, and ulcerations. Inflammatory infiltrates and occasional ulcerations were observed in both GFD- and STD-fed groups treated with 2.5% DSS (Figure 4A). Although the degree of inflammatory infiltration varied, possibly reflecting the specific site of colon sampled (e.g., the presence of ulceration), histological scoring of HE-stained specimens from three colon locations in each animal revealed a highly significant increase in inflammatory signs in the DSS-treated STD group compared with the DSS-treated GFD group (P < 0.0001, n = 10), demonstrating a protective effect of the GFD against DSS-induced intestinal inflammation (Figure 4B). Conversely, only discrete minor inflammatory infiltrates were observed in the control groups of BALB/c mice fed either the GFD or STD (n = 10; Figure 4B).

Figure 4 Colon morphology and histology of the dextran sulfate sodium-treated BALB/c mice fed gluten-free diet or standard diet. A: Representative hematoxylin and eosin-stained colon sections from BALB/c mice treated with 2.5% dextran sulfate sodium (DSS) for 7 days and fed the gluten-free diet (GFD) or standard diet (STD) (n = 10), magnification 5 × or 10 ×; B: Histology score evaluating colonic mucosa in GFD- and STD-fed BALB/c mice (P < 0.0001, n = 10) after 7 days of 2.5% DSS exposure, including untreated controls; C: Representative histological specimens documenting mucin in colon and goblet cells within colonic epithelium (stained by Alcian blue and nuclear fast red) of BALB/c mice treated for 7 days with 2.5% DSS and fed GFD or STD (n = 10), magnification 5 × or 10 ×; D: Statistically significant (P < 0.0001) preservation of crypt length in GFD vs STD BALB/c mice (n = 10) exposed to 2.5% DSS. Example of two independent experiments. Data are expressed as mean ± SD. ^dP < 0.0001. DSS: Dextran sulfate sodium; STD: Standard diet; GFD: Gluten-free diet.

Mucin production and the number of mucin-producing goblet cells, visualized by Alcian blue and nuclear fast red staining, were better preserved in the GFD group (Figure 4C). Finally, detailed assessment of crypt length at three different locations from each colon revealed a highly significant (P < 0.0001, n = 10) preserved crypt length in mice fed the GFD compared with those fed the STD (Figure 4D). The average crypt length increased by 16.2% in DSS-treated mice on the GFD.

Flow cytometry analysis of colon infiltrating Ly6C^high monocytes

Cells isolated from the colons and MLN of GFD- and STD-fed mice exposed to 2.5% DSS for 7 days were subjected to flow cytometric analysis. After exclusion of debris and doublets and the selection of viable CD45⁺ leukocytes, cells were further stratified according to side scatter properties and surface marker expression, including CD3, CD49b, Siglec-F, Ly6G, CD11c, CD11b, and Ly6C. The gating strategy for the CD11c^low CD11b⁺ Ly6C^high cells is shown in Supplementary Figure 1.

Within these leukocyte subsets, a marked expansion of colonic CD11c^low CD11b⁺ Ly6C^high monocytes was detected in DSS-treated mice maintained on the STD Altromin diet, whereas this population was significantly reduced in animals on the GFD (P < 0.0001; n = 8; Figure 5A). No statistically significant difference in the proportion of CD11c^low CD11b⁺ Ly6C^high monocytes was found in MLN (n = 8; Figure 5A). This finding indicates that dietary gluten contributes to enhanced recruitment of inflammatory monocytes into the colon during DSS-induced colitis. The observed effect is consistent with other measures of disease activity, such as preserved colon length and lower DAI scores in the GFD group (Figures 2 and 3). Representative flow cytometry dot plots illustrating the differential distribution of CD11c^low CD11b⁺ Ly6C^high cells in the colon between GFD- and STD-fed mice are shown in Figure 5B.

Figure 5 Monocyte infiltration in colon and mesenteric lymph nodes of dextran sulfate sodium-treated mice fed gluten-free diet vs standard diet. Flow cytometry analysis of colon-infiltrating monocytes prepared according to Couter and Surana[37] and monocytes in mesenteric lymph nodes. After exclusion of debris and doublets, cells were further gated according to the side scatter, cluster of differentiation (CD) 45, and other markers and are displayed as a percentage of live CD11c^low CD11b⁺ lymphocyte antigen 6C (Ly6C)^high cells. A: Gluten-free diet (GFD)-fed mice displayed a statistically significant reduction in the proportion of colonic CD11c^low CD11b⁺ Ly6C^high cells (n = 8, P < 0.0001); B: Example dot plots of colonic monocytes from the dextran sulfate sodium-treated mice on the GFD and standard diet. Data are expressed as mean ± SD. ^dP < 0.0001. MLN: Mesenteric lymph nodes; CD: Cluster of differentiation; Ly6C: Lymphocyte antigen 6C; SSC-A: Side scatter-area; STD: Standard diet; GFD: Gluten-free diet.

Effect of GFD on Foxp3⁺ regulatory T cells, IL-10⁺ T helper cells, and γδ T cells in MLN and spleen in DSS-induced colitis

Cell suspensions were prepared from gut-draining MLN and spleens. Three days after a 7-day exposure to 2.5% DSS, BALB/c mice on the GFD showed a significant increase in the proportion of CD4⁺ Foxp3⁺ T cells in the MLN (P < 0.01, n = 8), but not in the spleen (Figure 6A; representative data from two independent experiments). This finding is in contrast to the original description of an unchanged proportion of Foxp3⁺ regulatory T cells (Tregs) in BALB/c mice on a GFD[27]. In accordance with the increased proportion of CD4⁺ Foxp3⁺ T cells, there was also a significant increase in unstimulated IL-10⁺ T helper cells (CD3⁺ CD4⁺ IL-10⁺) in the MLN (P < 0.01, n = 8), and to a lesser extent in the spleen (P < 0.05, n = 8), of animals on the GFD. Illustrative flow cytometry dot plots from GFD- and STD-fed groups are shown in Figure 6B-D. The gating strategy for Foxp3⁺ Tregs and IL-10⁺ T helper cells is shown in Supplementary Figure 2.

Figure 6 Effect of gluten-free diet on forkhead box P3⁺ regulatory T cells and interleukin-10⁺ T helper cells in mesenteric lymph nodes and spleen of dextran sulfate sodium-treated mice. A: Proportion of cluster of differentiation (CD) 4⁺ forkhead box P3 (Foxp3)⁺ cells within T cells (CD3⁺-gated); B: Example dot plots of CD4⁺ Foxp3⁺ regulatory T cells in mesenteric lymph nodes (MLN); C: Proportion of interleukin (IL)-10⁺ T helper cells (CD3⁺ CD4⁺-gated) in MLN and spleen of female BALB/c mice fed gluten-free diet and standard diet (n = 8) that were treated for 7 days with 2.5% dextran sulfate sodium; D: IL-10⁺ (CD3⁺ CD4⁺) T helper cells. Data are expressed as mean ± SD. ^aP < 0.05. ^bP < 0.01. P values represent an example of two independent experiments. MLN: Mesenteric lymph nodes; STD: Standard diet; GFD: Gluten-free diet; CD: Cluster of differentiation; TCR: T cell receptor; IL: Interleukin; Foxp3: Forkhead box P3; SSC-A: Side scatter-area; Treg: Regulatory T cells; PE-Cy7: Phycoerythrin-cyanine 7; Th: T helper cell.

We next assessed possible changes in γδ T cells, which have previously been reported to be influenced by a GFD or by mucosally administered gliadin[27,38]. Figure 7A shows a slight increase (P < 0.05; group size n = 8) in γδ T cells in both MLN and the spleen (Figure 7; representative data from two independent experiments). Illustrative flow cytometry dot plots of CD3⁺ γδ⁺ cells in the spleen are shown in Figure 7B. There were no substantial shifts in the proportions of CD8⁺ or CD8^- subsets within the γδ T-cell population in either MLN or the spleen. The prevailing population consisted of CD8^- γδ T cells, representing the majority (~90%) of all γδ T cells (Figure 7C and D). Representative flow cytometry dot plots from MLN are shown in Figure 7D, and the gating strategy for the CD8⁺ or CD8^- γδ T cells are shown in Supplementary Figure 3.

Figure 7 Changes of γδ T cells in mesenteric lymph nodes and spleen of dextran sulfate sodium-treated mice fed gluten-free diet and standard diet. A: Increased proportion of cluster of differentiation (CD) 3⁺ γδ⁺ T cells in mesenteric lymph nodes (MLN) and spleen of female BALB/c mice fed gluten-free diet vs standard diet (n = 8, P < 0.05) after 7 days of 2.5% dextran sulfate sodium; B: Example dot plots of γδ T cells in spleen; C: There were no changes in the proportion of CD8^- vs CD8⁺ γδ T-cell subsets; D: CD8^-/CD8⁺ γδ⁺ T cells in MLN. Data are expressed as mean ± SD. ^aP < 0.05. The results represent an example from two independent experiments. MLN: Mesenteric lymph nodes; STD: Standard diet; GFD: Gluten-free diet; CD: Cluster of differentiation; TCR: T cell receptor; SSC-A: Side scatter-area; PE-Cy7: Phycoerythrin-cyanine 7.

These results may indicate a beneficial effect of the GFD on regulatory or protective mucosal immune cell populations under DSS-induced pro-inflammatory conditions.

DISCUSSION

In this study, we examined the impact of a GFD on DSS-induced colitis in BALB/c mice. Compared with the STD, the GFD resulted in preservation of colon length, reduced intestinal bleeding, improved histological scores, and diminished inflammatory infiltration, particularly of Ly6C⁺ monocytes. In addition, GFD administration was associated with increased frequencies of Foxp3⁺ Tregs, IL-10⁺ T helper cells, and γδ T cells in MLN, highlighting both morphological and immunological benefits in this mouse model of UC.

The protective effect of the GFD on colon length preservation represents the most striking findings of our study. While environmental changes are easy to study in spontaneous models of an autoimmune disease with incomplete disease penetrance, such as the NOD mouse model of T1D, where it is relatively easy to alter the natural disease course, the chemically induced DSS model of acute UC is much more challenging in this respect. Colon shortening is a well-established marker of colitis severity in the DSS model, reflecting the degree of tissue damage and inflammatory response[39]. Our results showing highly significant preservation of colon length (P < 0.0001) in GFD-fed mice are consistent with reduced inflammatory damage to the colonic tissue. This macroscopic improvement was accompanied by significant histological improvements, such as preservation of crypt architecture, which is important for maintaining intestinal barrier function and epithelial homeostasis[40]. The better preservation of mucin-producing goblet cells in GFD-treated mice is also relevant, as these cells form the first line of defense against luminal antigens and pathogens[41]. Loss of goblet cells and mucin production is characteristic of IBD and contributes to increased intestinal permeability[42]. Thus, the GFD may harness the protective barrier function of the gut.

The mechanism underlying the pro-inflammatory properties of gluten has been well documented. Gliadin stimulates monocytes to produce IL-8 and tumor necrosis factor (TNF)-α through nuclear factor kappa-B (NF-κB) signaling pathways, exhibiting LPS-like inflammatory properties[20]. Furthermore, pepsin digestion of wheat gliadin increases IL-1β production via Toll-like receptor 4 (TLR4)/MyD88/Toll/interleukin-1 receptor domain-containing adaptor protein inducing interferon beta/mitogen-activated protein kinase/NF-κB signaling and NOD-like receptor family pyrin domain containing 3 inflammasome activation and was also documented to induce inflammatory cytokine production via the MyD88 pathway[21,43]. Interestingly, several innate immune/pattern recognition receptors e.g., CD14, TLR4 are expressed not only by lamina propria macrophages and dendritic cells but also by gut epithelial cells[44,45]. Furthermore, gliadin increases intestinal permeability through zonulin release and CXCR3 binding, potentially facilitating bacterial translocation and sustaining inflammatory responses[25].

Ly6C⁺ inflammatory monocytes are rapidly recruited to inflamed tissues, where they differentiate into pro-inflammatory macrophages and contribute to tissue damage through production of inflammatory mediators such as TNF-α, IL-1β, and reactive oxygen species[46]. Interestingly, COMMD10 has been shown to negatively regulate inflammasome activity in Ly6C^high monocytes in DSS-induced colitis and IBD[47]. Our findings of decreased Ly6C^high monocyte infiltration in GFD-fed mice support the concept that gluten withdrawal reduces inflammatory cell recruitment to the colon.

The immunological benefits of the GFD in our study were reflected by increased proportions of Foxp3⁺ Tregs, IL-10⁺ T helper cells and γδ T cells in MLN and to a lesser extent in the spleen. Tregs, including specialized intestinal Foxp3⁺ retinoic acid receptor-related orphan receptor γt⁺ Tregs, play important roles in maintaining intestinal immune homeostasis and in preventing excessive inflammatory responses[48-50]. Previous studies have shown that dietary gluten affects immune cells, especially within the mucosal compartment, in immunocompetent BALB/c mice. We have demonstrated that dietary gluten alters the balance between pro-inflammatory and anti-inflammatory cytokines in T cells of BALB/c mice, with the STD promoting inflammatory responses compared with the GFD[28]. Additionally, gluten impacts Tregs and Th17 cells in BALB/c mice[27]. Ejsing-Duun et al[51] reported that dietary gluten significantly decreased the abundance of Tregs in mice compared with those on a STD.

The γδ T cells, particularly gut-resident IL-17A-producing γδ T cells or butyrophilin-like-selected γδ T-cell subsets, contribute to epithelial barrier maintenance, tissue repair, and protection from intestinal inflammation[52,53]. γδ T cells, which are abundant in intestinal tissues, contribute not only to epithelial barrier maintenance but are also essential for mechanisms of mucosal (oral) tolerance[54]. Using single-cell transcriptomics, we recently studied the effect of the gluten-free Altromin diet on lymphocyte subsets in the NOD mouse model of T1D. The GFD did not lead to substantial alterations in the abundance or phenotype of the major lymphocyte subset; however, our data revealed subtle alterations in the differentiation of unconventional, potentially protective subsets, such as Tregs, activated γδ T cells, and natural killer T cells. The GFD was also associated with increased gene expression regulated by IL-2, IL-7 and IL-15, whereas the STD gluten-containing Altromin promoted type I interferon-responsive gene expression[55]. Thus, the enhancement of Foxp3⁺ Tregs and IL-10⁺ T helper cells in MLN, together with a modest increase in γδ T cells following GFD, suggests that gluten withdrawal might modulate both adaptive and innate-like immune regulatory pathways, thereby contributing to the attenuation of intestinal inflammation.

Our findings must be considered within the broader context of how different dietary protein sources influence intestinal inflammation. It has been demonstrated that diets rich in simple sugars promote pro-inflammatory responses through gut microbiota alterations and TLR4 signaling[56]. This suggests that the beneficial effects we observed with the GFD may result not only from the gluten removal but also from the overall dietary composition, which is difficult to specify in open-formula diets. Our results contrast with findings by Kostovcikova et al[57], who reported that animal protein-rich diets promote pro-inflammatory macrophage responses and exacerbate colitis in mice. In our study, the GFD contained increased meat protein content to replace gluten-containing grains while maintaining similar total protein levels. The discrepancy may be explained by differences in specific protein sources used. On the other hand, the protective effect of the GFD is consistent with previous work by Menta et al[29], who demonstrated that wheat gluten supplementation exacerbated DSS-induced colitis by weakening junctional complex proteins and increasing bacterial translocation. Our results extend these findings by showing that complete gluten withdrawal provides significant protection against colitis development. The preservation of colon length (P < 0.0001) and improved DAI scores in GFD-fed mice suggest that gluten removal creates a less inflammatory intestinal environment conducive to maintaining mucosal integrity during chemical challenge.

The clinical relevance of our findings is supported by observational studies in IBD patients. Triggs et al[30] reported that gluten-free products were associated with symptom improvement in Crohn’s disease patients while Herfarth et al[31] found clinical improvement in IBD patients following GFDs. However, other studies have reported mixed results, with Zallot et al[32] and Schreiner et al[33] finding insufficient improvements. A recent randomized placebo-controlled trial in patients with mild-to-moderate UC (n = 26) failed to demonstrate significant benefits of a short-term (6-week) GFD on inflammatory markers or disease severity; however, the authors emphasized that larger clinical trials with longer follow-up periods are needed[58].

The heterogeneity in clinical responses may reflect differences in patient populations, disease severity, concurrent medications, or adherence to dietary interventions[59]. It should also be taken into consideration that because effective therapies for IBD relapse are available, patients cannot be left without appropriate treatment. Thus, prospective clinical trials in IBD with dietary interventions face several challenges[60,61] in addition to issues, such as short duration, small sample size, variable adherence, and inappropriate endpoints. Therefore, randomized clinical trials in IBD with dietary interventions should be designed with stricter parameters[62], such as longer duration of dietary interventions, increased number of participants, and enrollment of subjects at diagnosis or with only short disease duration. Our controlled experimental model eliminates many of these confounding factors and provides more clearly defined and mechanistic insights into how a GFD might benefit IBD patients. Nevertheless, additional mechanistic studies will be required before the therapeutic application of gluten withdrawal in established human IBD can be properly evaluated.

CONCLUSION

In conclusion, our data provide evidence that a GFD significantly ameliorates experimental colitis possibly through multiple mechanisms involving both innate and adaptive immune responses. The preservation of the colon architecture, reduced inflammatory cell infiltration, and expansion of regulatory immune cell populations support the concept that gluten withdrawal represents a safe and potentially effective intervention for IBD management. These findings warrant further investigation in clinical trials to evaluate the therapeutic potential of a GFD as an adjunctive treatment for IBD.

ACKNOWLEDGEMENTS

The authors thank Jan Svoboda from the Centre Cytometry and Microscopy core (Institute of Microbiology of the Czech Academy of Sciences, Prague, Czech Republic) and Jan Valecka from the Service Laboratory of Light Microscopy (Institute of Molecular Genetics of the Czech Academy of Sciences, Prague, Czech Republic) for excellent technical assistance.