Distinct colitis-associated macrophages drive NOD2-dependent bacterial sensing and gut homeostasis
Gajanan D. Katkar, Mahitha Shree Anandachar, Stella-Rita C. Ibeawuchi, Ella G. McLaren, Megan L. Estanol, Kennith Carpio-Perkins, Shu-Ting Hsu, Celia R. Espinoza, Jane E. Coates, Yashaswat S. Malhotra, Madhubanti Mullick, Vanessa Castillo, Daniella Vo, Saptarshi Sinha, Pradipta Ghosh
Gajanan D. Katkar, Mahitha Shree Anandachar, Stella-Rita C. Ibeawuchi, Ella G. McLaren, Megan L. Estanol, Kennith Carpio-Perkins, Shu-Ting Hsu, Celia R. Espinoza, Jane E. Coates, Yashaswat S. Malhotra, Madhubanti Mullick, Vanessa Castillo, Daniella Vo, Saptarshi Sinha, Pradipta Ghosh
View: Text | PDF
Research Article Gastroenterology Immunology Microbiology

Distinct colitis-associated macrophages drive NOD2-dependent bacterial sensing and gut homeostasis

Abstract

Single-cell studies have revealed that intestinal macrophages maintain gut homeostasis through the balanced actions of reactive (inflammatory) and tolerant (noninflammatory) subpopulations. How such balance is impaired in inflammatory bowel diseases (IBDs), including Crohn’s disease (CD) and ulcerative colitis (UC), remains unresolved. Here, we define colon-specific macrophage states and reveal the critical role of noninflammatory colon-associated macrophages (niColAMs) in IBD recovery. Through trans-scale analyses—integrating computational transcriptomics, proteomics, and in vivo interventional studies—we identified GIV (CCDC88A) as a key regulator of niColAMs. GIV emerged as the top-ranked gene in niColAMs that physically and functionally interacts with NOD2, an innate immune sensor implicated in CD and UC. Myeloid-specific GIV depletion exacerbates infectious colitis, prolongs disease, and abolishes the protective effects of the NOD2 ligand muramyl dipeptide in colitis and sepsis models. Mechanistically, GIV’s C-terminus binds the terminal leucine-rich repeat 10 (LRR 10) of NOD2 and is required for NOD2 to dampen inflammation and clear microbes. The CD-associated 1007fs NOD2 variant, which lacks LRR 10, cannot bind GIV, which provides critical insights into how this clinically relevant variant impairs microbial sensing and clearance. These findings illuminate a critical GIV•NOD2 axis essential for gut homeostasis and highlight its disruption as a driver of dysbiosis and inflammation in IBD.

Authors

Gajanan D. Katkar, Mahitha Shree Anandachar, Stella-Rita C. Ibeawuchi, Ella G. McLaren, Megan L. Estanol, Kennith Carpio-Perkins, Shu-Ting Hsu, Celia R. Espinoza, Jane E. Coates, Yashaswat S. Malhotra, Madhubanti Mullick, Vanessa Castillo, Daniella Vo, Saptarshi Sinha, Pradipta Ghosh

×

Figure 4

A mouse model of dysbiosis, impaired microbial clearance, patchy chronic transmural ileocolitis, and fibrosis.

Options: View larger image (or click on image) Download as PowerPoint
A mouse model of dysbiosis, impaired microbial clearance, patchy chronic...
(A and B) Schematic (A) and bar graph (B) display the process and outcome of a 16S fecal microbiome analysis (using the QIIME 2 multi-omics data science framework) at baseline in 10-week-old, myeloid-specific (LysMCre) GIV-KO mice and their control littermates (WT). n = 5 mice in each group. (C–I) Panel describe the experimental design (C) and findings (D–I) in an infectious colitis model of GIV-KO and control littermates induced using Citrobacter rodentium (initially named C. freundii biotype 4280 (88); strain name DBS100); 5 × 108 CFU/200 μL/mouse. GIV-KO, n = 8; WT, n = 6. Findings are representative of 2 independent repeats. (D) Line graphs display the bacterial burden in fecal pellets over 7 weeks after the initial oral gavage. (E) Bar graph displays the differences in colon length. H&E-stained (F) or trichrome-stained (H) images representative of Swiss rolls of the entire intestinal tract are shown. Scale bar: 2.5 mm. Magnified fields of the rectum (R), colon (C), and small intestine (SI) of the corresponding boxed regions are shown. Scale bar: 250 μm. Arrows show regions of transmural inflammation or crypt distortion; immune infiltrates (F) correspond also to transmural fibrosis (H). Segments in between these patches appear normal. Bar graphs show the histology index (89) (G) based on submucosal inflammation, percent area involved, inflammatory infiltrates in LP and crypt hyperplasia, and the degree of fibrosis (I), as assessed by H&E and trichrome staining of 5 WT and 5 GIV-KO mice. All results are displayed as mean ± SEM. Significance was tested using 2-tailed Student’s t test. Only significant P values (≤ 0.05) are shown.