HIF-1 attenuates high-fiber diet-mediated proliferation and stemness of colonic epithelium

Abstract

The complex interplay between diet, microbiota, and host is exemplified by the effects of dietary fiber on the intestine. Inulin ingestion triggers epithelial changes in the colon that depend on microbiota-derived molecules, including enhanced proliferation, increased mucus production, and elevated antimicrobial peptide secretion. Here we employed a multilayered and multi-omics approach, including dietary interventions, intestinal organoids, and both genetic and pharmacological interventions to investigate the impact of inulin on two aspects of diet-microbiota-host interactions: intestinal hypoxia and hypoxia-inducible factor (HIF)-1 signaling in intestinal epithelial cells (IECs). We found that inulin, a soluble fiber, promotes intestinal hypoxia, stabilizing HIF-1 in IECs in a microbiota- and host-dependent manner. Furthermore, we show that HIF-1 stabilization modulates intestinal stem cell (ISC) function through metabolic reprogramming in a microbiota-dependent manner. Our findings reveal an unrecognized role for HIF-1 in orchestrating microbiota-dependent epithelial metabolism and proliferation in the colon, underscoring the complexity of diet-microbiota-host interactions.

Keywords: Hypoxia; dietary fiber; intestinal stem cells; inulin; metabolism; microbiota.

Figure 1.

Inulin ingestion induces distinct transcriptional profiles in enterocytes and proliferating cells in the colon. (A) Venn diagram comparing the upregulated genes among intestinal stem cells (ISC), transit-amplifying cells (TA) and enterocytes (EC) obtained from the scRNA-seq analysis of intestinal epithelial cells (IECs) in inulin-treated animals. (B) Venn diagram comparing the downregulated genes among ISC, TA and EC obtained from the scRNA-seq analysis of IECs in inulin-treated animals. (C) KEGG pathway analysis by enrichR highlighting the top five terms for the up- (red) and downregulated (blue) genes among ISC, cycling TA and EC obtained from the scRNA-seq analysis of IECs in inulin-treated mice. (D) Average expression of selected oxidative phosphorylation (OXPHOS) genes in ISC, TA, and EC populations from control and inulin-fed mice. Genes were selected from significantly enriched pathways identified by gene ontology analysis. Color scale represents Z-score normalized expression values.

Figure 2.

Inulin ingestion increases intestinal hypoxia and HIF-1a stabilization in intestinal epithelial cells. (A) Experimental design illustrating diets treatments and analyses performed in mice. (B) PMDZ staining was quantified by scoring sections of proximal colon from conventional mice fed with a control or a 10% inulin diet (n = 6). (C) Representative images of colonic sections stained with PMDZ from conventional mice treated with control and inulin diet for 3 weeks. Bars = 200 µm, 20x magnification. Zoom insets were taken from different areas or sections. Scale bars in the insets represent 100 µm (D) Representative images captured on the IVIS-SPECTRUM equipment 5 min after luciferin injection in ODD-mice fed with control or inulin diets (n = 3–5). (E) Luminescence quantification over time after luciferin injection in mice fed with control and inulin diet (n = 3–5). (F) Normalized HIF-1 luciferase activity in the small intestine and colon from mice treated with control or inulin diet (n = 3–5). (G) Relative quantification of Hif1a and its target genes expression on colonic epithelial cells by RT-qPCR (n = 8). Data were analyzed using Student’s t-test. In all graphs, each point represents an individual mouse. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 3.

HIF-1α stabilization in response to inulin diet is dependent on colon microbiota and epithelial Pparg. (A) Experimental design illustrating diet and ABX treatments in C57BL/6 mice. The ABX mix consisted of 1 g/L ampicillin, 1 g/L neomycin, 1 g/L metronidazole, and 0.5 g/L vancomycin. (B) Median fluorescence intensity (MFI) for HIF-1α⁺ labeling in CD45⁻/EpCAM⁺ cells, as determined by flow cytometry, in mice fed control or inulin diet and treated with a mix of antibiotics (ABX + inulin) (n = 5). (C) Relative expression of Hif1a and HIF-1α target genes in mice fed with an inulin diet and treated with an antibiotic mix (n = 5). (D) Relative expression of Hif1a and HIF-1α target genes in specific pathogen-free (SPF), germ-free (GF) and conventionalized (CV) mice fed with a normal chow diet (n = 4). (E) In vitro growth curves of BOV and BT in media containing different amounts of inulin (n = 3). (F) Experimental design illustrating gnotobiotic mice colonized with B. ovatus (BOV) or B. thetaiotaomicron (BT), fed a control or inulin diet, followed by pimonidazole injection. (G) Hypoxia score in monocolonized mice fed control or inulin diet (n = 6). (H) Representative images of colonic sections stained with PMDZ from monocolonized mice treated with control and inulin diet for 3 weeks. Bars = 200 µm, 20x magnification. (I) Experimental design illustrating diets treatments and analyses performed in Pparg^fl/fl and Pparg^∆IEC mice. (J) Hypoxia score in Pparg^fl/fl and Pparg^ΔIEC fed with control or inulin diet (n = 4–7). (K) Relative mRNA expression of Hif1a in colonic epithelial cells from Pparg^fl/fl and Pparg^ΔIEC, as determined by RT-qPCR. (n = 4–7). Data were analyzed using one-way ANOVA followed by Tukey’s multiple comparisons test (B and D), Student’s t-test (C) or two-way ANOVA followed by Sidak’s multiple comparisons test (J and K). In all graphs, each point represents an individual mouse. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 4.

Epithelial deletion of HIF-1α increases intestinal epithelial proliferation and modulates metabolism pathways in response to inulin diet. (A) Experimental design illustrating diets treatments and analyses performed in HIF-1α^fl/fl and HIF-1α^∆IEC mice. (B) Colon length normalized by body weight in HIF-1α^fl/fl and HIF-1α^∆IEC mice fed with inulin diet (n = 8–9). (C) Crypts length normalized by body weight in HIF-1α^fl/fl and HIF-1α^∆IEC mice fed with inulin diet (n = 8–9). (D) Representative images of colonic sections stained with H&E from mice fed with inulin diet for 3 weeks. Bars = 100 µm, 20x magnification (n = 8–9). (E) Quantification of EdU-positive cells per crypt (n = 4). (F) Visualization of EdU-positive cells in colonic crypts by fluorescence microscopy following EdU Click-iT reaction . Bars = 100 µm, 20 ×magnification. (G) Biological process enrichment map from differentially regulated proteins (HIF-1α^ΔIEC vs HIF-1α^fl/fl) in mice fed with inulin diet, highlighting the modules associated with cell metabolism (n = 5). (H) Process-protein network showing proteins differentially regulated (blue-red gradient for down- and upregulation, respectively) across metabolic modules. The red edges highlight key proteins and processes linking metabolic effects with epithelial development. Data were analyzed using Student’s t-test. In all graphs, each point represents an individual mouse. *p < 0.05, **p < 0.01.

Figure 5.

HIF-1α plays a critical role in intestinal stem cell (ISC) function by modulating oxidative metabolism. (A) Quantification of the clonogenicity capacity of colon crypts in mice with and without epithelial deletion of HIF-1α (n = 9). (B) Growth kinetics of organoids (n = 9). (C) Representative images of organoids from mice with HIF-1α epithelial deletion. (D) Experimental design for sorting cell experiments with HIF-1α^ΔISC mice (n = 5–6). (E) Clonogenicity capacity of sorted-ISCs from HIF-1α^ΔISC mice and their controls (n = 5–6). (F) Organoid volume average from HIF-1α^ΔISC mice and their controls (n = 5–6). (G) Representative images of organoids from HIF-1α^ΔISC mice and their controls (n = 5–6). (H) Oxygen consumption rates (OCRs) of organoids, as measured by CellMito stress test and a Seahorse XF extracellular flux analyzer (n = 3). (I) Basal respiration of colonic organoids (n = 3). (J) ATP-linked respiration of colonic organoids (n = 3). (K) Maximal respiration of colonic organoids (n = 3). Data were analyzed using Student’s t-test. In all graphs, each point represents an individual mouse. *p < 0.05, **p < 0.01.

Figure 6.

HIF-1 controls colon cell proliferation by regulating oxidative phosphorylation and mitochondrial function in ISCs. (A) Percentage of live EdU-positive cells in colon organoids after treatment with the FAO inhibitor, etomoxir (n = 4). (B) Clonogenicity capacity of colon crypts treated with etomoxir (n = 4). (C) Measurement of organoids volume after 4 days of etomoxir treatment (n = 4). (D) Experimental design of etomoxir in vivo treatment. (E) EdU-positive cells in colon sections from mice treated with etomoxir (20 mg/kg i.p.) (n = 4). (F) Representative images of EdU-positive cells in the colon of mice treated with PBS or etomoxir (n = 4). Bars = 100 µm, 20 ×magnification. (G) Clonogenicity capacity of colon crypts from mice treated with etomoxir (n = 4). (H) Experimental design of organoids treatment with different inhibitors of glycolysis and mitochondrial complexes. (I) Clonogenicity capacity of colon crypts treated with glycolysis inhibitor, 2-deoxyglucose (2-DG), and inhibitors of mitochondrial complexes (n = 3–5). (J) Organoid volume average from organoids treated with oligomycin, antimycin, rotenone or 2-DG (n = 3–5). Data were analyzed using two-way ANOVA followed by Sidak’s multiple comparisons test (A, B, C, I and J) or Student’s t-test (E and G). In all graphs, each point represents an individual mouse. *p < 0.05, **p < 0.01, ***p < 0.001.