IgA-mediated control of host-microbial interaction during weaning reaction influences gut inflammation

Figures & data

Figure 1. Weaning induces decreased gut barrier function and an increase of IgA-coated bacteria.

a, Body weight of piglets in SP and WP groups over 7 days (n = 32). b, Comparison of the total histology scores and individual histological criterion scores between the two groups (n = 6). c, Ileum morphology visualization via H&E-stained sections and scanning electron microscopy in SP and WP groups (scale bars, 100 µm). d and e, Assessment of intestinal tight junction protein expression by western blotting (n = 6). f, IF staining for goblet cells (MUC2, pink; scale bars, 100 µm). g, Visualization of CD19⁺ B cells in intestinal tissue via representative IHC images. h, Intestinal tissue IgA, IgG, and IgM levels analysis via ELISA (n = 6). i, Serum and stool IgA analysis via ELISA. j, Flow cytometry analysis of fecal bacterial IgA binding in SP and WP groups, with quantification of IgA-bound fraction (n = 6). k, PCoA emperor plot reveals separation between SP and WP microbiomes (n = 24). Statistical analysis: unpaired t-test. *p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 2. The bacteroides taxon is enriched upon weaning and associated with diarrhea when not highly coated with IgA.

a, Illustration depicting the longitudinal study design encompassing collection of fecal samples, gut microbiota analysis, subject grouping, and correlation analysis. b, Visualization of molecular ecological networks spanning from birth to weaning. Diverse color-coded large modules, consisting of ≥3 nodes, with smaller modules displayed in gray. c, Linear discriminant analysis (LDA) was employed to discern the variation in piglet microbiota composition over time. The length and direction of the arrows depict the normalized scaling for each genus. d, Relative abundance of the Bacteroides in SP (green) and WP (purple) groups. Adjusted R and p values were derived from linear regression analyses. e, The percentage of IgA-coated bacteria was correlated by linear regression with diarrhea score in weaned piglets (n = 48). Linear regression and Pearson’s correlation coefficient were shown. f, The relative abundance of the Bacteroides was positively correlated by linear regression with diarrhea score in group with low levels of IgA-coated bacteria, whereas no correlation was observed in group with high levels of IgA-coated bacteria (n = 24). Linear regression and Pearson’s correlation coefficient were shown. g, WP were subjected to the isolation and characterization of gut bacterial species, employing a combination of qPCR screening, custom anaerobic medium cultivation, and Sanger sequencing techniques. h, Alignment of 16S rRNA sequences of each isolate to the Bacteroides sp. (Feature ID: 5) identified from Silva database. Black lines indicate mismatch. i, Occupancy–abundance curves for the SP and WP communities. Bacteroides sp. features detected as enriched in WPs are highlighted. j, The abundance levels of individual members within the community were ranked over time. The high abundance member of the WP group, Bacteroides sp. (Feature ID: 5), is highlighted.

Figure 3. Weaning stress remodels epithelial and immune cells to emphasize IgA-producing CD138⁺ PCs.

a, UMAP visualization of all ileal cells in SP and WP, with each cell colored based on its corresponding cell-type compartment. b, Boxplots showing CytoTRACE values for progenitor cells, immature enterocytes and mature enterocytes c, Differentiation trajectory of progenitor cells, immature enterocytes and mature enterocytes, assigning distinct colors to each cell type for visualization. d, Heatmap presenting the signaling pathways that display significant variations along the pseudotime. e, 3D differentiation trajectories of mature enterocyte, where each cell was assigned a finite pseudotime value. f, UMAP plots of enterocyte cluster overlay for the AUC scores of given genesets. g, UMAP visualization of all B cells in SP and WP, with each cell colored based on its corresponding cell-type compartment. h, UMAP plots of B cell clusters are colored based on whether they pass the AUC score assignment threshold for selected pathway activates. Cells that surpassed the threshold were colored in red. i, Boxplots showing individual cell AUC score for IgA production activity.

Figure 4. B. uniformis elicits IgA responses in CD138⁺ PCs and alleviates DSS-induced colitis.

a, Timeline of antibiotic treatment, mono-colonization of individual gut bacteria and DSS-induced colitis in SPF mice. b, Weight change of C57BL/6J mice exposed to piglet isolated bacteria pre- and post-DSS colitis. c, Fecal IgA level in C57BL/6 mice colonized with individual gut-isolated bacteria of piglet (n = 6). d, Flow-cytometry analysis of fecal bacterial IgA binding in mice colonized with individual isolates, shown by histogram of IgA-coated bacteria fraction. e, Whole-body imaging and intestines imaging ex-vivo assessing the B. uniformis burden in intact animals. Mice were administrated with FDAA-labeled B. uniformis. f, Mice were gavaged for 2 weeks with medium (control), B. uniformis, or heat-killed B. uniformis and treated with 2.5% DSS for 1 week. g, Weight change of mice exposed to B. uniformis under DSS intervention for 7 days (n = 6). h, Macroscopic image of the colon on day 7 and length of colon. i, Comparison of the total histology scores and individual histological criteria scores across groups (n = 6). j, Representative micrographs of H&E-stained colonic sections, demonstrating that B. uniformis administration results in a reduction in intestinal inflammation compared to both control and heat-killed B. uniformis. Scale bar, 100 µm). k, Flow cytometry analysis of fecal bacterial IgA binding in medium (control), B. uniformis, or heat-killed B. uniformis groups, with quantification of IgA-bound fraction (n = 6). l, PBMCs of all mice were collected for flow cytometry analysis of frequency of CD138⁺ IgA⁺ among live CD45⁺ CD3⁻ CD19⁺ cells, with quantification of CD138⁺ IgA⁺ precent (n = 6). m, LPMCs of all mice were collected for flow cytometry analysis of frequency of CD138⁺ IgA⁺ among live CD45⁺ CD3⁻ CD19⁺ cells, with quantification of CD138⁺ IgA⁺ precents (n = 6). Statistical analysis: (i, k, l, and m) One-way ANOVA, (c) Mann–Whitney U test or (b, g) Two-way ANOVA. *p < 0.05, **p < 0.01.

Figure 5. Low IgA responses in CD138⁺ PCs fail in entrapping B. uniformis and aggravate colitis.

a, Mice were subjected for a 2-week regimen with gel (control), B. uniformis, Bz or B. uniformis + Bz followed by a one-week treatment with 2.5% DSS. b, Weight change of mice under DSS intervention for 7 days (n = 6). c, Macroscopic image of the colon on day 7 and length of colon. d, Representative micrographs of H&E-stained colonic sections. Comparison of the total histology scores and individual histological criteria scores across groups demonstrated that B. uniformis + Bz administration results in an increase in intestinal inflammation compared to B. uniformis group (scale bar, 100 µm; n = 6). e, Flow cytometry analysis of fecal bacterial IgA binding. f, Confocal images of gut luminal bacteria stained with SYTO BC (green) and anti-IgA-PE. Scale bars, 20 µm. g, Representative images of intestinal sections after IF staining with CD138 (green) and IgA (red); scale bar, 50 µm. h, LPMC of all mice were collected for flow cytometry analysis of frequency of CD138⁺ IgA⁺ among live CD45⁺ CD3⁻ CD19⁺ cells, with quantification of CD138⁺ IgA⁺ precents (n = 6). i, 3D-reconstructed imaging of transplanted B. uniformis pre-labeled by 7-BADA in CUBIC-cleared ileum and colon tissue from B. uniformis + Bz treated mice. j, Representative TEM projections of intestine from mice subjected to B. uniformis and B. uniformis + Bz treatments (scale bar, 2 µm). k, Magnetically sorted IgA+ and IgA- bacteria from fecal samples and co-cultured with IPEC-J2 cells. Confocal images of immunofluorescence staining revealed a high level of cell adhesion in IgA-negative bacteria (scale bar, 20 µm). l, Magnetically sorted IgA⁺ and IgA⁻ bacteria from fecal samples and co-cultured with IPEC-J2 cells. Apoptosis was detected by concurrent staining with Annexin V-FITC and PI. Statistical analysis: (b) Two-way ANOVA, (d and h) One-way ANOVA or (l) unpaired t-test. *p < 0.05, **p < 0.01, mean ± SEM.

Figure 6. B. uniformis exacerbates colitis in the absence of IgA or CD138⁺ PCs.

a, Rag1^−/− and IgA^−/− mice were subjected for a two-week regimen with medium (control) or B. uniformis, followed by a one-week treatment with 2.5% DSS. b, LPMCs of Rag1^−/− and IgA^−/− mice were collected for flow cytometry analysis of frequency of CD138⁺ IgA⁺ among live CD45⁺ CD3⁻ CD19⁺ cells, with quantification of CD138⁺ IgA⁺ precent (n = 6). c, Flow-cytometry analysis of fecal bacterial IgA binding across groups, with quantification of the IgA-bound fraction (n = 6). d, Representative TEM projections of intestine from Rag1^−/− and IgA^−/− mice subjected to control or B. uniformis treatments (scale bar, 2 µm). e, Representative micrographs of H&E-stained colonic sections, demonstrating that B. uniformis causes an increase in intestinal inflammation relative to controls (scale bar, 100 µm). f, Comparison of the total histology scores and individual histological criteria scores across groups (n = 6). g, CD138^−/− mice were gavaged for 2 weeks with medium (control), B. uniformis or magnetically sorted IgA-positive B. uniformis and treated with 2.5% DSS for 1 week. h, LPMCs of CD138^−/− mice were collected for flow cytometry analysis of frequency of CD138⁺ IgA⁺ among live CD45⁺ CD3⁻ CD19⁺ cells, with quantification of CD138⁺ IgA⁺ precents (n = 6). i, Flow cytometry analysis of fecal bacterial IgA binding across groups, with quantification of the IgA-bound fraction (n = 6). j, Representative TEM projections of intestine from CD138^−/− mice treated with medium (control), B. uniformis or magnetically sorted IgA positive B. uniformis (scale bar, 2 µm). k, Representative micrographs of H&E-stained colonic sections. Comparison of the total histology scores and individual histological criteria scores across groups demonstrate that B. uniformis causes an increase in intestinal inflammation of CD138^−/− mice relative to other groups (scale bar, 100 µm; n = 6). Statistical analysis: (b, c, h, i and k) One-way ANOVA. *p < 0.05, **p < 0.01. ns, not significant. Mean ± SEM.

Figure 7. CD138⁺ PCs mediate interactions with enterocytes and bacterium in weaning stress.

a, Flow cytometry analysis of frequency of CD138⁺ IgA⁺ PCs in piglets collected at days 14, 21 (per-weaning) and 28 (post-weaning) with quantification of the CD138⁺ IgA⁺ PC subset across groups (n = 8). One way-ANOVA, **p < 0.01, ****p < 0.0001, mean ± SEM. WP/HD: weaning piglets with high diarrheal rate, WP/LD: weaning piglets with low diarrheal rate. b, Circle plot showing differential cell–cell communication network between two groups. The width of edges represents the relative number of interaction strength. Red colored edges represent increased signaling in the WP group. c, Histogram showing significant interactions (ligand-receptor pairs) and related signaling genes for the given signaling pathways. d, Chord diagram visualizing the cell–cell interactions mediated by two selected signaling pathways. e, Representative images of intestine sections from SP and WP group after IF staining with CD45 (pink), CD138 (green) and IgA (red); scale bar, 500 µm. f, Individual cell AUC score for host–microbiota interaction activities were subjected to statistical analysis, followed by liner regression analysis with the ranking of each B cell cluster.

Data availability statement

The raw sequencing data have been deposited in the National Center for Biotechnology Information (NCBI) Sequence Read Archive (SRA) repository with accession number PRJNA970096 and PRJNA974859. Source data are provided with this article. R markdown scripts enabling the analysis for data are uploaded at https://github.com/Tangwenjie34 and https://github.com/yusenWei. Specific codes are accessible from the authors. All database and detailed description of the methods are available in the main text or the supplementary materials.