Dietary n-3 polyunsaturated fatty acids (PUFA) decrease obesity-associated Th17 cell-mediated inflammation during colitis

Abstract

Clinical and experimental evidence suggests that obesity-associated inflammation increases disease activity during colitis, attributed in part to the effects of Th17 cells. Using a model of concurrent obesity and colitis, we monitored changes in critical immune cell subsets and inflammatory biomarker expression in three key tissues: visceral adipose tissue, colon (local inflammatory site) and spleen (systemic inflammatory site), and we hypothesized that n-3 PUFA would reduce the percentage of inflammatory immune cell subsets and suppress inflammatory gene expression, thereby improving the disease phenotype. Obesity was induced in C57BL/6 mice by feeding a high fat (HF) diet (59.2% kcal) alone or an isocaloric HF diet supplemented with fish oil (HF-FO) for 12 weeks. Colitis was induced via a 2.5% trinitrobenzene sulfonic acid (TNBS) enema. The HF-FO diet improved the obese phenotype by reducing i) serum hormone concentrations (leptin and resistin), ii) adipose tissue mRNA expression of inflammatory cytokines (MCP-1, IFNγ, IL-6, IL17F and IL-21) and iii) total (F4/80⁺ CD11b⁺) and inflammatory adipose tissue M1 (F4/80⁺ CD11c⁺) macrophage content compared to HF (P<0.05). In addition, the HF-FO diet reduced both colitis-associated disease severity and colonic mRNA expression of the Th17 cell master transcription factor (RORγτ) and critical cytokines (IL-6, IL-17A, IL-17F, IL-21, IL-23 and IFNγ) versus HF (P<0.05). Compared to HF, the percentage of both splenic Th17 and Th1 cells were reduced by the HF-FO group (P<0.05). Under ex vivo polarizing conditions, the percentage of HF-FO derived CD4⁺ T cells that reached Th17 cell effector status was suppressed (P = 0.05). Collectively, these results indicate that n-3 PUFA suppress Th1/Th17 cells and inflammatory macrophage subsets and reconfigure the inflammatory gene expression profile in diverse tissue sites in obese mice following the induction of colitis.

Figure 1. Characterization of the diet-induced obese phenotype.

C57BL/6 mice were fed a high diet (HF), high fat diet supplemented with FO (HF-FO) or a low fat (LF) control diet for 12 weeks (n = 12−17 TNBS-treated and 4–6 vehicle controls/diet). Mice were presensitized with 1% TNBS or vehicle control (week 11), followed by a 2.5% TNBS enema or vehicle control (week 12) and sacrificed 3 d post-TNBS. A) Changes in body weight over time. B) Visceral adipose tissue weight from individual visceral depots (perinephric, mesenteric and epididymal) or combined (total visceral adipose). Serum concentrations of C) insulin, D) leptin, E) resistin and F) adiponectin. All data were analyzed by two-way ANOVA (main effects: diet and treatment) and P-values are shown. Bars represent means ± SEM and statistical significance was (P≤0.05). Panel A) asterisk (*) indicates statistically significant time points where the HF and HF-FO groups differed from LF (P≤0.05). Panels B-F) bars not sharing a common letter differ (P≤0.05).

Figure 2. Visceral adipose tissue macrophage infiltration.

Stromal vascular cells (SVC) were isolated and quantified from total visceral adipose tissue (HF and HF-FO groups, n = 3 vehicle controls and 6–8 TNBS-treated mice, LF n = 4 pooled samples comprised of 3–4 mice/treatment). A) percentage of F4/80⁺ CD11b⁺ cells (total macrophages), B) percentage of F4/80⁺ CD11c⁺ cells (M1 macrophages), C) percentage of F4/80⁺ CD206⁺ cells (M2 macrophages). Data were analyzed by two-way ANOVA (main effects: diet and treatment) and bars represent mean values ± SEM. Bars not sharing a common letter are significantly different (P≤0.05).

Figure 3. Colon histological disease scores for TNBS-treated mice.

Colonic mucosal injury (0–3) and inflammation (0–3) scores were assessed in a blinded manner by a board-certified pathologist (B. Weeks) and combined for a total score (0–6). Representative images (100 × magnification) are shown for the HF, HF-FO and LF TNBS-treated groups, respectively (panels A-C) and a representative image of a HF vehicle control (panel D) is shown. E) Combined injury/inflammation histological score within the distal colon (n = 10−14 TNBS treated mice/diet). Data were analyzed using the Kruskal-Wallis test followed by Wilcoxon two-sample testing, and bars represent median values. Bars not sharing a common letter are significantly different (P≤0.05).

Figure 4. Effect of diet and colitis on splenic T cell subsets.

A) Tregs (CD4⁺ FOXP3⁺), B) Th17 (CD4⁺ IL17A⁺), and C) Th1 (CD4⁺ IFNγ⁺) cell populations (n = 3–6 vehicle controls and n = 6−12 TNBS treated mice/dietary group). Bars represent mean values ± SEM. Bars not sharing a common letter are significantly different (P≤0.05).

Figure 5. Effect of diet on splenic CD4⁺ T cell polarization.

Splenic CD4⁺ cells were purified by positive selection and cultured for 3 d under A) Th17 or B) Treg polarizing conditions (see Materials and Methods, n = 3−4 TNBS treated mice/dietary group). Bars represent mean values ± SEM. Bars not sharing a common letter are significantly different (P≤0.05).