Gp130 Signaling in NOD2-Driven Crohn’s Disease: A Key Player in Fibrosis and a Novel Target for Refractory Patients

Nayar S, Morrison JK, Giri M, et al. A myeloid–stromal niche and gp130 rescue in NOD2-driven Crohn’s disease. Nature 2021;593:275–281. NOD2 (loss-of-function) mutations are the major genetic risk factor for Crohn’s disease. NOD2-driven Crohn’s disease is characterized by ileal location and a higher frequency of fibrostenotic complications. Although it is well-established that NOD2 recognizes bacterial components, the reasons why mutations in NOD2 cause increased risk for Crohn’s disease are not completely understood. In this study, the authors aimed to elucidate how NOD2 drives pathogenicity in Crohn’s disease using human samples (intestinal tissue and peripheral blood mononuclear cells) and a convenient animal model of intestinal injury (zebrafish treated with dextran sodium sulfate [DSS]). By performing single-cell RNA sequencing from inflamed tissue of human ileum, they found a gene expression signature associated with activated macrophage and fibroblast cells. Moreover, they identified a cell population that expresses myeloid and fibroblast markers (CD14+PDGFRA+ cells). This population increases in inflamed ileum and exhibits enriched expression of profibrotic markers. These findings suggest that a population of inflammatory macrophages differentiates to become activated fibroblasts during the course of disease. Importantly, DSS-treated zebrafish showed conservation of myeloid, stromal, and epithelial populations that shared distinctive signatures between species, thus highlighting the value of the model to study human ileal Crohn’s disease. The authors hypothesized that NOD2 deficiency results in aberrant fibroblast–macrophage homeostasis and differentiation over time. In agreement with that finding , they found that tissue samples from patients carrying NOD2 mutations exhibited a pathogenic activated fibroblast and macrophage signature with significant increases in the expression of CD14, PDGFRA, activated fibroblast markers (PDPN, CHI3L1), matrix metalloproteinases (MMP3, MMP9), and gp130 family genes (IL11, IL6, and OSM). This activity resulted in downstream activation of STAT3, among other chronic inflammatory and fibrotic pathways. In vitro studies revealed a higher number of collagen-high expressing activated fibroblasts after stimulating peripheral blood monocytes from healthy volunteers carrying NOD2 mutations with muramyl dipeptide (MDP), a NOD2 ligand. They also tested the effect of MDP exposure in different zebrafish lines (nod2 wild type, nod2 mutant, and nod2 knockout) and found an up-regulation of profibrotic transcripts in nod2-deficient lines. After DSS injury, lines with nod2 deficiency did not efficiently ameliorate fibrotic and inflammatory signals 24 hours after DSS removal. Interestingly, patients refractory to anti-tumor necrosis factor (TNF) treatment also showed increased expression of gp130 genes and activated fibroblast signatures after treatment. Finally, the author tested the potential value of bazedoxifene, a small molecule gp130 inhibitor, in NOD2-driven Crohn’s disease. The drug was able to significantly reverse the aberrant differentiation of newly recruited CD14+ peripheral blood mononuclear cells from NOD2-mutated carriers and also decreased secretion of IL-6 and profibrotic factors. Moreover, treatment with bazedoxifene reduced intestinal injury (ie, decreased infiltrating leukocytes and alleviated intestinal length shortening) in nod2 mutant zebrafish treated with DSS. In 2001, 2 groups independently reported the gene NOD2 as the first susceptibility gene for Crohn’s disease (Nature 2001;411:599–603; Nature 2001;411:603–606). Different studies have demonstrated an approximately 2-fold risk for Crohn’s disease in individuals carrying heterozygote NOD2 mutations and 20-fold risk in homozygotes or complex heterozygotes (Lancet 2001;357:1925–1928; Gastroenterology 2002;122:867–874) Genome-wide association studies have established >200 loci associated with Crohn's disease risk, but NOD2 remains the strongest genetic risk factor (Nature 2017;547:173–178). NOD2 is an intracellular receptor that senses MDP, a component of bacterial cell walls, and stimulates host innate immune response. Stimulation of NOD2 activates pro-inflammatory pathways such as nuclear factor-κB, mitogen-activated protein kinases, and caspase-1. In contrast, NOD2 deficiency can result in deregulation of host-pathogen interactions and aberrant inflammatory response (PLoS Pathog 2017;13:e1006177). NOD2-driven Crohn’s disease is associated with ileal location, stenosis, fistula, and the need for surgery (Gut 2013;62:1556–1565). However, the mechanisms of NOD2 pathogenesis are not fully elucidated. To investigate them, Nayar et al used human samples and a zebrafish model of intestinal injury. Animal models are powerful research tools for studying the pathogenesis and potential therapeutic approaches for many different diseases. Currently, the most established research models for the majority of human diseases, including inflammatory bowel disease (IBD), are murine models. However, they have some limitations: (i) murine models breed a generation in about 2 months, (ii) housing and maintenance are expensive, (iii) pathologic changes cannot be determined by observation, and (iv) genetic manipulation is complex. These limitations have prompted researchers to find more efficient animal models. Zebrafish possess an immune system quite similar to ours, including orthologs of mammalian Toll-like receptors (TLRs), innate immune receptors that recognize specific microbial molecules, and functionally conserved orthologs of susceptibility genes for IBD such as NOD1 and NOD2 (Front Immunol 2017;8:501). Zebrafish models offer some appealing advantages versus murine models: (i) high fecundity. Females lay hundreds of eggs that can be fertilized in vitro, thus enabling more efficient generation of research models. (ii) Housing and maintenance are easier and cheaper. Several thousand zebrafish can be housed in a small laboratory. (iii) Availability of larger quantity of material to work with (iv) optical transparency and rapid development of the embryos facilitate observation, manipulation, and drug screening; and (vi) genetic manipulation is easier. Despite their advantages, zebrafish have some limitations for modeling IBD. Duplicated genes are common in the zebrafish genome, and therefore, homologs of mammal immune factors might not have conserved functions in zebrafish. Lipopolysaccharide, the TLR4 ligand of mammals, does not bind to zebrafish TLR4. Zebrafish gastrointestinal system lacks lymph glands, Peyer’s patches, and intestinal crypts. The micro-organisms living in mammals do not survive in the zebrafish gut. In any case, it is worth noticing that the authors demonstrated that a DSS-treated zebrafish model recapitulates well cell populations and distinctive signatures of human ileal Crohn’s disease, thus supporting the use of this model in this setting. The present study convincingly demonstrated that NOD2 deficiency drives a pathogenic macrophage-fibroblast program. NOD2 mutations confer risk through altered differentiation of newly recruited blood monocytes over time. Specifically, the loss of NOD2 establishes a pathogenic activated fibroblast and macrophage signature characterized by an enrichment of gp130 ligands (ie, IL-6, IL-11, and OMS) and downstream activation of STAT3, resulting in chronic inflammation and fibrosis. Moreover, they found that patients refractory to anti-TNF treatment also show a post-treatment signature with enrichment of gp130 ligands and activated fibroblast transcripts. Anti-TNF drugs revolutionized the treatment of IBD. Novel biological alternatives for patients non-responsive to anti-TNF drugs are a pressing clinical need. Strategies modulating gp130 signaling are of great interest in this setting because gp130 ligands are known players in the pathogenesis of IBD, and many have been suggested as potential therapeutic targets. Treatments blocking IL-6 and IL-11 have been tested in IBD, but not adopted in clinical practice due to conflicting results and/or concerning side effects. In a pilot study, intravenous tocilizumab, a drug blocking the IL-6 receptor, showed a significantly higher clinical response rate in Crohn’s disease than placebo, although endoscopic and histologic healing was not different between groups (Gastroenterology 2004;126:989–996). Subcutaneous PF-04236921, an anti–IL-6 drug, obtained greater clinical response rates than placebo in Crohn’s disease. Unfortunately, safety concerns, including abscesses and perforations (Gut 2019;68:40–48) were communicated. Specific neutralization of the soluble form of the IL-6 receptor using olamkicept might represent a safer and more effective strategy. A recent trial evaluating olamkicept in patients with active IBD suggested effectiveness during short-term exposure. Clinical remission was associated with changes in mucosal pSTAT3 levels. Importantly, the drug was well-tolerated and neither signs of significant immunosuppression nor intestinal perforations were observed (Gastroenterology 2021;160(7):2354–2366). Initial studies demonstrated recombinant human IL-11 to be safe and significantly superior to placebo in inducing Crohn's disease remission after 6 weeks (Aliment Pharmacol Ther 2002;16:399–406). However, in a later multicenter study, it was less efficient than prednisolone in short-term remission induction and poor in maintaining remission (Am J Gastroenterol 2006;101:793–797). In this article, the author tested bazedoxifene, a small-molecule inhibitor of gp130, in NOD2-driven Crohn’s disease. This drug has shown promising preclinical results in colon and pancreatic cancer, malignancies characterized by excess production of IL-6 and dysfunction of IL-6 signaling, particularly the IL-6–JAK–STAT3 pathway. Bazedoxifene significantly reversed the aberrant differentiation of newly recruited blood monocytes from NOD2 mutated carriers and reduced intestinal injury in DSS-treated nod2 mutant zebrafish. These positive results prompted the authors to file a patent application on repurposing bazedoxifene for clinical use in a subset of patients with Crohn’s disease. Although their preclinical results are promising, it will be essential to evaluate the potential adverse effects of the drug in patients with Crohn’s disease. Moreover, it is an open question if inhibiting gp130 signaling more upstream (specific gp130 ligands) or downstream (ie, JAK and STAT inhibitors) could be more effective in these patients. In this sense, JAK inhibitors are currently in late-stage clinical development for Crohn’s disease. In summary, the authors shed light on the mechanisms of NOD2 pathogenesis in Crohn’s disease. They convincingly demonstrate that NOD2 deficiency drives a pathogenic macrophage-fibroblast program characterized by increased expression of gp130 genes and STAT3 hyperactivation, which leads to chronic inflammation and fibrosis. Blocking gp130 signaling with bazedoxifene might ameliorate pathogenic activation of the myeloid-stromal niche and provide a new therapeutic strategy in these patients. Further research is required to evaluate the therapeutic value of gp130 blockage in Crohn’s disease.