Biomimetic superabsorbent hydrogel acts as a gut protective dynamic exoskeleton improving metabolic parameters and expanding A. muciniphila

Abstract

The rising prevalence of obesity and metabolic disorders worldwide highlights the urgent need to find new long-term and clinically meaningful weight-loss therapies. Here, we evaluate the therapeutic potential and the mechanism of action of a biomimetic cellulose-based oral superabsorbent hydrogel (OSH). Treatment with OSH exerts effects on intestinal tissue and gut microbiota composition, functioning like a protective dynamic exoskeleton. It protects from gut barrier permeability disruption and induces rapid and consistent changes in the gut microbiota composition, specifically fostering Akkermansia muciniphila expansion. The mechanobiological, physical, and chemical structures of the gel are required for A. muciniphila growth. OSH treatment induces weight loss and reduces fat accumulation, in both preventative and therapeutic settings. OSH usage also prevents liver steatosis, immune infiltration, and fibrosis, limiting the progression of non-alcoholic fatty liver disease. Our work shows the potential of using OSH as a non-systemic mechanobiological approach to treat metabolic syndrome and its comorbidities.

Keywords: Akkermansia muciniphila; NAFLD; gut health; gut-liver axis; intestinal barrier; intestinal permeability; metabolic syndrome; microbiota; obesity; superabsorbent hydrogel.

Graphical abstract

Figure 1

OSH administration prevents obesity and metabolic syndrome in mice C57BL/6J male mice were fed for 18 weeks with HFD or HFD supplemented with 2% or 4% OSH or control diet (CD). (A) Body weight variation as percentage of basal; area under the curve (AUC) of body weight variation. (B) Epididymal adipose tissue (EAT) weight. (C) H&E staining of EAT tissue sections, scale bar, 100 μm; epididymal adipocyte area distribution, line at median; 75–100 cells per field per mouse; five mice per group were analyzed. (D) Serum total cholesterol levels. (E–G) High-density lipoprotein (HDL) and low-density lipoprotein (LDL) serum cholesterol levels and total circulating triglyceride levels, respectively. (H) Relative gene expression levels of the lipid and fatty acid transporters Slc27a4, Ffar2, Cd36, and Fabp6 in the ileum at 18 weeks, expressed as fold change of CD-fed group (n = 5 CD; n = 9 HFD and 2%–4% OSH). (I and J) Fasting blood glucose and insulin levels (n = 5 CD, n = 10 HFD, n = 8 2%–4% OSH). (K) HOMA-IR values. (L) Circulating GLP-1 levels. (M) Intraperitoneal GTT expressed as a percentage of basal performed after 17 weeks of feeding (n = 5 mice per group); AUC of GTT. (N) Intraperitoneal ITT as a percentage of basal performed after 17 weeks of feeding (n = 5 mice per group); inverted AUC of ITT (∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001 one-way ANOVA Dunnett post-test, line at mean with SEM).

Figure 2

OSH administration protects intestinal barrier from HFD-induced alterations (A–C) Small intestine, colon, and total intestine length, respectively, after 18 weeks of feeding. (D) Colon tissue sections of mice fed for 18 weeks, stained for Muc-2 (in green), E-cadherin (in red), and DAPI, marker for nuclei (in blue). Scale bar, 50 μm. (E) Plasma FITC-dextran (4 kDa) levels after 2 and 4 weeks of feeding, expressed as fold change of CD (n = 15 for 2 weeks testing, n = 10 for 4 weeks testing). (F) Circulating LPS levels after 18 weeks of feeding. (G) Ileum tissue sections stained for ZO-1, marker of tight junctions (in green); CD34, marker of vessels (in gray); and DAPI, marker for nuclei (in blue). Scale bar, 50 μm. (H) GLP-2 serum levels. Fluorescent signals are expressed as integrated density/μm² and analyzed using Fiji image software (∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001; one-way ANOVA Dunnett’s post-test, line at mean with SEM; n = 5 [CD], n = 10 mice per group).

Figure 3

OSH therapeutic administration over 12 weeks reverses obesity and metabolic alterations induced by HFD feeding (A) Experimental design for therapeutic administration of OSH. (B) Body weight variation expressed in percentage of basal; area under the curve. (C) EAT weight. (D) H&E staining of EAT formalin-fixed paraffin-embedded tissue sections, scale bar, 100 μm. Epididymal adipocyte area distribution, line at median; 75–100 cells per field per mouse, five mice per group were analyzed. (E–H) HDL and LDL serum levels, total circulating triglycerides, and total cholesterol. (I and J) Fasting blood glucose and insulin levels (n = 5 mice per group). (K) HOMA-IR values (n = 5 mice per group. (L) Intraperitoneal GTT as a percentage of basal performed after 23 weeks of feeding (n = 5 mice per group); AUC of GTT. (M) Intraperitoneal ITT (n = 5 per group); inverted AUC of ITT. (N) Circulating GLP-1 levels. (O–Q) Small intestine, colon, and total intestine length. (R) Plasma FITC-dextran (4 kDa) levels (n = 8 CD, n = 8 HFD, n = 7 2% OSH, n = 9 4% OSH; outliers were calculated using GraphPad outlier calculator and excluded from analysis). (S) Circulating LPS levels after 12 weeks of OSH treatment (n = 7 mice per group). (T) Quantification of Muc-2 fluorescent signal expressed as integrated density/μm², performed using Fiji image software. (U) Quantification of ZO-1 fluorescent signal expressed as integrated density/μm², performed using Fiji image software (n = 5 CD and HFD, n = 7 for 2% and 4% OSH). (V) Colon (left) sections stained for Muc-2 (in green), E-cadherin (in red), and DAPI (in blue). Ileum tissue (right) sections stained for ZO-1 (in green); CD34, a marker of vessels (in gray); and DAPI (in blue). Representative images of a single mouse of five for CD and HFD and of seven for 2% and 4% OSH groups. Scale bar, 50 μm (∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001; one-way ANOVA Dunnett’s post-test, line at mean with SEM; number of mice per group n = 10 HFD in all graphs if not specified differently).

Figure 4

OSH administration prevents HFD-induced liver steatosis and limits NAFLD progression (A) Liver tissue sections of C57BL/6J male mice fed for 2 (top), 4 (middle), and 18 (bottom) weeks with an HFD diet or an HFD supplemented with 2% and 4% OSH or CD, stained with oil red O (ORO); representative images of 1 of 5 (2 and 4 weeks) and 1 of 5 (CD) or 9 or 10 mice per group (18 weeks); scale bar, 50 µm. (B) Steatosis grade histologically quantified after 18 weeks of preventative OSH treatment. (C) Steatosis grade prior to and after 12 weeks of OSH therapeutic administration. (D) Liver tissue sections stained with ORO. Top row illustrates one representative image each of liver sections from CD- and HFD-fed mice at the end of disease induction phase (12 weeks HFD, n = 5); bottom row shows liver section images after 12 weeks of OSH therapeutic treatment. Scale bar, 50 μm. (E) Relative gene expression levels of lipid and fatty acid transporter Cd36 and regulators of lipid metabolism and β oxidation Ppara, Cpt1a, Pparg, Mlxipl, and Fasn in the liver after 12 weeks of OSH treatment, expressed as fold change of CD (n = 10). (F) Liver tissue sections of C57BL/6J male mice therapeutically fed for 6 weeks with OSH-supplemented HFHCC diet, stained with H&E (arrows point to inflammatory foci, the star indicates an example of hepatocyte degeneration, and α indicates microvesicular steatosis surrounding zone 3), ORO for liver triglycerides, and Sirius red for liver fibrosis (arrows point to pericellular and perivenular collagen accumulation). Scale bars, 50 and 100 μm (ORO-H&E and Sirius red, respectively). (G–I) Liver triglyceride quantification expressed in milligrams per gram of liver tissue, histological quantification of steatosis grade, and circulating LDL cholesterol levels, respectively. (J–M) NAFLD activity score, ballooning degeneration score, inflammation grade, and percentage of Sirius red-positive area, respectively. (N) IHC staining on liver tissue sections for CD3, F4/80, Ly6G, and B220. Scale bar, 100 μm, arrowheads indicate DAB-positive cells; relative quantification on the right side. (O–Q) Fluorescence-activated cell sorting (FACS) analysis of liver immune populations. (O) Gating strategy for cDC1 (CD11c⁺, MHCII⁺, and CD103⁺) and cDC2 (CD11c⁺, CD11b⁺, CD3⁻, and B220⁻), (P) frequency of cDC1 in CD45⁺ cells, and (Q) frequency of cDC2 in CD45⁺ cells (n = 8 mice per group) (∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001, one-way ANOVA Dunnett’s post-test, line at mean with SEM; n = 10 mice per group in all graphs if not specified differently).

Figure 5

OSH administration shapes gut microbiota composition, preventing and reversing HFD-induced alterations (A) PCoA of fecal microbiota from mice fed for 18 weeks HFD (green) or OSH-supplemented HFD (2% OSH, orange, and 4% OSH, light blue), compared with basal (CD, black), based on the Jaccard index; single dots represent individual mice (PERMANOVA, p < 0.001). (B) Shannon’s diversity index of fecal microbiota from mice fed for 18 weeks with OSH-supplemented HFD (∗p < 0.05; one-way ANOVA, Dunnett’s post hoc test). (C) Relative abundance of detected fecal bacteria phyla (∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001; one-way ANOVA, Dunnett’s post hoc test). (D) Firmicutes/Bacteroidetes ratio. (E) Akkermansia muciniphila relative abundance. (F) Correlation between bacterial taxa with metabolic variables measured after 18 weeks of preventative OSH feeding (∗p < 0.05; ∗∗p < 0.01; Spearman correlation). (G) Cladogram including all bacterial species identified in fecal metagenome. Colored bar plot shows differences between OSH supplemented groups and HFD expressed as log fold change of bacterial abundance (preventative study in orange and therapeutic study in light blue; commonly decreased taxon in red and commonly increased taxon in green). (H) Functional analysis. Venn diagram and plots of differentially abundant pathways showing differences between OSH-supplemented groups and HFD commonly found in preventative and therapeutic mouse studies with a gFC > |0.3|; the bar plot represents the change in pathway abundance (gFC) and the error bars refer to the variability between the preventative and the therapeutic experiments. (I) Heatmap of differentially abundant carbohydrate-active enzymes (CAZymes). (J) Heatmap of Akkermansia muciniphila-associated pathways.

Figure 6

OSH abets rapid Akkermansia muciniphila expansion both in vivo and in vitro (A) Experimental scheme for short-term A. muciniphila abundance quantification upon HFHCC and OSH administration. Relative A. muciniphila abundance was normalized to the total 16S rRNA levels (n = 5 mice/group). (B) Fecal A. muciniphila relative abundance (normalized to total 16S rRNA) 2 weeks after microbiota depletion with a broad-spectrum antibiotic cocktail. (C) Scheme for in vitro culture of A. muciniphila. (D) A. muciniphila growth curve expressed as optical density at 600 nm wavelength (OD_600nm) in the presence or absence of OSH; AUC (n = 6 repetitions), unpaired t test, ∗∗p < 0.01. (E) AUC of A. muciniphila growth curves with decreasing concentrations of type III porcine mucin (0.05%–0%). Multiple unpaired t test, ∗∗p < 0.01; line at mean with SEM. (F) A. muciniphila growth in the presence of OSH vs. fermentable fiber inulin and gel-forming fiber psyllium. (G) A. muciniphila growth curve in presence of OSH vs. OSH individual components: carboxymethyl cellulose (CMC), citric acid (CA), and the two components together un-cross-linked. (H) A. muciniphila growth curve in presence of OSH vs. a gel-forming compound of synthetic origin. (I) A. muciniphila growth curve expressed as OD_600nm in the presence of OSH vs. a gel-forming compound of synthetic origin and combination of the last with OSH individual components (CMC and CA and CMC + CA). (F–I) ^#p < 0.0001, one- or two-way ANOVA, Dunnett’s post-test relative to OSH vs. control, inulin, and psyllium, mean of two independent tests, line at mean with SEM.

Figure 7

FMT of OSH-shaped gut microbiota reduces body weight and restores glucose and insulin metabolism in HFHCC-fed mice (A) Experimental design of the FMT. (B) Recipients’ body weight variation expressed as percentage of basal, where basal is the weight before the first FMT oral gavage; AUC of body-weight variation curve. (C and D) Intraperitoneal GTT and ITT, respectively. (E) Liver weight. (F) Alanine aminotransferase (ALT) serum levels in recipient mice after 4 weeks of FMT. (G) Liver histology. H&E and ORO staining for liver triglycerides. Scale bars, 50 and 100 μm (ORO and H&E). (H and I) Histological score of steatosis grade and NAFLD activity score, respectively, after 4 weeks of FMT. (J) Recipients’ serum LDL levels (∗p < 0.05 and ∗∗p < 0.01, one-way ANOVA, Dunnett’s post-test, line at mean with SEM; number of mice per group n = 4 autologous FMT [HFHCC in HFHCC-fed recipients], n = 4 FMT 2% OSH in HFHCC recipients, n = 5 FMT 4% OSH in HFHCC recipients in A–J).