Crosstalk between Microbiota-Derived Short-Chain Fatty Acids and Intestinal Epithelial HIF Augments Tissue Barrier Function

Abstract

Interactions between the microbiota and distal gut are fundamental determinants of human health. Such interactions are concentrated at the colonic mucosa and provide energy for the host epithelium through the production of the short-chain fatty acid butyrate. We sought to determine the role of epithelial butyrate metabolism in establishing the austere oxygenation profile of the distal gut. Bacteria-derived butyrate affects epithelial O2 consumption and results in stabilization of hypoxia-inducible factor (HIF), a transcription factor coordinating barrier protection. Antibiotic-mediated depletion of the microbiota reduces colonic butyrate and HIF expression, both of which are restored by butyrate supplementation. Additionally, germ-free mice exhibit diminished retention of O2-sensitive dyes and decreased stabilized HIF. Furthermore, the influences of butyrate are lost in cells lacking HIF, thus linking butyrate metabolism to stabilized HIF and barrier function. This work highlights a mechanism where host-microbe interactions augment barrier function in the distal gut.

Figure 1. Microbial-Derived Short-Chain Fatty Acids Promote Localized O₂ Depletion

(A) Mice (n = 12) were administered PMDZ 30 min prior to sacrifice. Colon samples were paraffin-embedded and stained according to the manufacturer’s instructions and counterstained with DAPI. (B) For analysis of “physiologic hypoxia” along the length of the lumen-to-serosa axis, samples (n = 3 per animal) were photographed and imported into ImageJ, where 100 μm sections were divided into equivalent 10 μm and quantified for PMDZ retention. Results are pooled and plotted as fluorescence intensity per μm². (C–E) Real-time O₂ content in responses to 10 mM NaB (C), acetate (D), and propionate (E) (blue) were compared to vehicle control (black). Inhibition of oxidative phosphorylation by the addition of oligomycin to each SCFA treatment increased O₂ content in media (red). Error bars represent SEM. (F and G) The rate of O₂ consumption 30 min after treatment with NaB (F) and acetate (G) was increased (p < 0.05). (H) Sodium propionate did not significantly influence the rate of O₂ consumption. (I) O₂ content of media was inversely related to concentration of NaB treatment with the greatest decrease at the physiologic concentration of 10 mM (p < 0.01). (J) Rate of O₂ consumption 30 min after treatment was also greatest with 10 mM NaB (p < 0.01). (K) Validation of the method for O₂ consumption using the XF Analyzer. Shown here is the influence of buffer (closed circles) and 10mM NaB (closed squares) on Caco2 cell O₂ consumption over a period of 100 min (p < 0.01). (L and M) Mitochondria-depleted Caco2 cells (L) consumed less O₂ compared with control after exposure to NaB (M). Data represent mean ± SEM from three ([I]–[M]) or four ([A]–[H]) independent experiments. Significance was determined using one-way ANOVA ([F]–[K]) and Student’s t test ([L] and [M]).

Figure 2. Buytrate Stabilizes HIF-1α and Transactivates HIF Target Genes

(A) The influence of acetate, propionate or NaB at indicated concentrations on the HIF HRE response (p < 0.01 for propionate and NaB). The response was abolished with oligomycin co-treatment in Caco2 cells transfected with a HIF reporter plasmid. DMOG is shown as a positive control. (B) NaB treatment for 8 hr increased nuclear HIF-1α in Caco2 cells (p < 0.05). This treatment increased retention of the O₂-sensitive probe in Caco2 cells. (C and D) Exposure of Caco2 cells to vehicle (gray bar), NaB (10 mM, 8 hr, black bars), or positive control DMOG (1 mM, 8 hr, white bars) induced indicated HIF target gene expression (D). (E) Influence of NaB on the expression of the HIF target gene CKM in T84 cells expressing lentiviral shRNA against Hif1b (Hif1b KD) relative to non-targeting control lentivirus (Sh Ctl). Data represent average ± SEM of three independent experiments. Significance determined using one-way ANOVA. *p < 0.05; **p < 0.01.

Figure 3. Antibiotics Reduce SCFA Concentration and HIF in the Colon

(A) Denaturing gradient gel electrophoresis, where each lane represents an individual animal and each band represents a distinct bacterial population, revealed depletion of most bacterial populations in the cecum of ODD-Luc mice with Abx treatment. (A and B) Despite appearance of resistant organisms (A), cecal contents from Abx-treated mice lost their ability to ferment inulin to SCFA. Each plate contained cecal contents from an individual animal, and yellow color indicates pH < 5.2 (B). (C) Analysis of culturable stool microbiota from vehicle and Abx-treated mice (n = 7 per group, where ** indicates p < 0.025). (D) Accordingly, cecal acetate, propionate, and NaB were reduced with Abx treatment (p < 0.001). (A–C) Data represent average ± SEM from three independent experiments ([A] and [B]). Each lane (A), dish (B), and symbol (C) represent an individual animal. (E) Significance was determined using Student’s t test. Abx treatment reduced retention of PMDZ in the colon indicating loss of “physiologic hypoxia.” (F and G) This influence reflected reduced luciferase activity ([F], p < 0.001) and HIF-1α protein ([G], p < 0.001) in the colon of ODD-Luc animals. (H) PMDZ localization in control and germ-free mouse colon (n = 6 per group). PMDZ was quantified (H) where *** indicates p < 0.001. (J) HIF-1α protein in control and germ-free mouse colon (n = 6 per group). Data represent average ± SEM from three independent experiments where *** indicates p < 0.001.

Figure 4. Butyrate Rescues “Physiologic Hypoxia” and Protects Barrier via HIF

(A) 4 hr following oral gavage of butyrate, tissue luciferase activity increased in the distal gut of Abx-treated ODD-Luc mice. (B–D) Oral gavage of TBN increased cecal acetate and butyrate (B), luciferase activity (C), and HIF-1α protein (D) in Abx-treated ODD-Luc mice. Daily TBN gavage for 3 days increased retention of O₂-sensitive probe in the colon of Abx-treated mice (n = 6 per group). (H) Validation of lentiviral knockdown of HIF-1β in high-resistance T84 monolayers is shown by western blot. (I) Role of HIF-1β in mediating influences of NaB treatment is shown by the lower flux ratio in NaB/control-treated control cells compared with HIF-1 β knockdown cells (p < 0.05). Data represent average ± SEM from three independent experiments. (C, F, and G) Significance was determined using Kruskal-Wallis (C) and Student’s t test ([F] and [G]).