Deciphering Cellular Networks in Creeping Fat Free

Crohn’s disease (CD) not only is frequently characterized by luminal inflammation of the gut, but also can feature hyperplasia of mesenteric adipose tissue and concomitant immune cell infiltration of fat tissue.^1,2 These morphologic changes of mesenteric adipose tissue adjacent to inflamed areas of the intestine in CD are referred to as creeping fat (CF) and were first described by Burill Crohn.³ However, the underlying pathophysiology of the development of CF and its contribution to disease course and severity of CD are still not well elucidated.

The current understanding of CF is thereby limited by the fact that CF represents a platform of highly complex interactions between various cell types such as adipocytes, immune cells, endothelial cells, and fibroblasts, as well as bacteria and other gut antigens.⁴

Thus, bacterial translocation from the intestinal lumen to the adjacent adipose tissue can be observed upon intestinal epithelial barrier breaches in CD leading to the infiltration of adipose tissue, with a variety of different immune cells such as B cells, T cells, and macrophages ultimately causing the production of various cytokines and adipokines, including leptin and adiponektin.^5,6 How these factors affect systemic and intestinal immune cell functions in human disease is currently not well deciphered, and it remains elusive if CF serves as an alternative barrier preventing the bacterial spread from leaky areas of inflamed gut or whether CF might contribute to intestinal inflammation and fibrogenesis by the secretion of proinflammatory factors and by providing an antigen-rich niche for autoreactive T cells, sustaining systemic and intestinal inflammation.

On one hand, observations from CD patients undergoing radical resection of the mesentery have suggested that CF might directly contribute to disease severity, as the resection of CF prolongs the recurrence-free survival of CD in a retrospective analyses of CD patients.⁷ Likewise, several groups have reported that CF might act as a source of elevated leptin-production in CD patients,^6,8 which can induce the production of tumor necrosis factor α in macrophages and T cells and which can deteriorate intestinal inflammation both in men and mice.^9,10 In contrast, T cells that do not express the leptin receptor are incapable of inducing intestinal inflammation in murine transfer models of colitis,^11,12 and leptin-deficient ob/ob mice are protected from chemically induced intestinal inflammation.¹⁰ Moreover, obesity and increased body mass index have been linked to a significantly higher risk of developing CD, but not to a significantly higher risk of developing ulcerative colitis, in adults in a large meta-analysis including 601 009 individuals, highlighting that fat-derived signals might specifically contribute to the pathogenesis of CD.¹³

On the other hand, increased abundances of interleukin-10–producing M2-like macrophages have been detected in the CF of CD patients, suggesting that CF might possess immune modulatory functions in the pathogenesis of CD by suppressing overwhelming immune responses.¹⁴ Accordingly, a recent publication has found that CF is characterized by high infiltration with viable bacteria including the bacterium Clostridium innocuum, which possess the potential to shift the differentiation of macrophages toward an immune modulatory M2-like phenotype in vitro.² Using single-cell RNA sequencing, the translocation of C. innocuum was associated with a higher infiltration of CF, with macrophages producing (among other cytokines) high levels of transforming growth factor β, suggesting that these cells might also contribute to the fibrogenesis and tissue repair frequently observed in CD patients.² In line with a potentially protective function of CF a study led by Ewe et al¹⁵ previously found that the radical resection of the mesentery of CD patients with ileal CD was associated with a higher recurrence rate of disease than patients undergoing conventional surgery, thereby opposing the recent findings of Coffey et al.⁷

Despite these advances in the understanding of the interplay between adipose tissue, microbiota, and fat-residing immune cells, the complex cellular networks underlying the development and function of CF in CD remain elusive, and other cell types such as endothelial cells and fat-residing fibroblasts have so far not been well characterized in CF, and their contribution to the modulation of CD is currently unclear.

With the help of single-cell RNA sequencing and immune-histochemical analyses, in this issue of Inflammatory Bowel Diseases Shu et al¹⁶ describe the cellular landscape of the stromal vascular cell fraction (SVF) in CF of patients with active ileal CD and provide subanalyses of fat-residing endothelial, fibroblast, and macrophage subsets. Especially the focus on endothelial cells and fibroblasts provides additional insight to the previously published single-cell RNA sequencing results of Ha et al² and helps to better delineate the complex interactions between the single-cell types of the SVF fraction in CF.

Thereby, Shu et al report that CF is composed of a heterogeneous population of immune cells, vascular and lymphatic endothelial cells, and fibroblasts. In intraindividual comparisons with mesenteric fat tissue from noninflamed areas of the gut of the same patients, CF is characterized by high abundances of plasma cells as well as by a significant enrichment of vascular endothelial cells expressing high levels of the transcription factor PPARγ, which is subsequently correlated to a higher expression of PPARγ-regulated targets such as the lipoprotein lipase, which is likely not only to functionally impact lipid metabolism in vascular endothelial cells, but also to facilitate the conversion of vascular endothelial cells into adipocytes.¹⁶

Furthermore, the authors delineate several subclusters of fibroblasts in CF, including a subset of cells predominantly showing features of inflammation-associated fibroblasts, which are characterized by a differential regulation of pathways that have been associated with fibrosis and inflammation, suggesting that these cells might contribute to fibrogenesis and tissue repair in CD.¹⁶

By a subanalysis of CF-infiltrating macrophages, Shu et al furthermore confirm previous observations of high macrophage abundances in CF, and their analyses now discriminate the presence of 6 different subpopulations of specific macrophages in CF, including interleukin-10–producing macrophages as well as inflammatory macrophages,¹⁶ underlying the importance to further discriminate the contribution of the different macrophage subclusters to CF development and CD.

By performing computational analyses of predicted ligand-receptor interactions, Shu et al additionally describe the complex interactions between single cell types arguing in favor of a distinguished network of intercellular cross-communication between fibroblasts, endothelial cells, and immune cells controlling the tissue remodeling from conventional mesenteric adipose tissue into CF.¹⁶

Despite the rather small sample size of the study and the descriptive nature of the presented data, Shu et al unambiguously demonstrate that further efforts should be undertaken to investigate the contribution of fat-residing vascular and lymphatic endothelial cells as well as fibroblasts to better understand their contribution to intestinal inflammation and fibrosis.

It will be, in our eyes, interesting to see if these changes in the SVF homeostasis observed in CF can be also reiterated in mouse models of CD such as in TNF^ΔARE mice,¹⁷ SAMPyit/FC mice,¹⁸ or Caspase8^fl/fl-Villin-Cre mice,¹⁹ which are characterized by ileal inflammation, which will be crucial to understand which of the described cell types might be specifically targeted in novel therapeutic approaches.

Moreover, it will be interesting to study in the future how the cellular composition of CF differs between patients undergoing specific immune-suppressive therapies and to better delineate the impact of the SVF on adipocyte function and differentiation in CF. This will be especially important to finally understand how CF contributes to the clinical disease course of CD in patients and whether the surgical removal of CF might be beneficial for achieving clinical remission in CD patients. Therefore, the results of the ongoing clinical phase 2 SPICY (Mesenteric SParIng Versus Central mesenterectomY in Ileocolic Resection for Terminal Ileitis in Crohn’s Disease) trial²⁰ comparing conventional surgery with the radical resection of the mesentery in CD (NCT04538638) are highly anticipated and will help to further decipher the role of CF in the pathogenesis of CD.

Conflict of Interest

C.W. declares no competing interests. B.S. has served as consultant for Abbvie, Arena, BMS, Boehringer, Celgene, Falk, Galapagos, Janssen, Lilly, Pfizer, Prometheus and Takeda and received speaker’s fees from Abbvie, CED Service GmbH, Falk, Ferring, Galapagos, Janssen, Novartis, Pfizer, Takeda (served as representative of the Charité).

References