Abstract B026: Implicating SATB2 loss in a metaplasia-dysplasia sequence in human inflammatory bowel disease

Abstract

Abstract Epithelial metaplasia, or the histologically evident replacement of one differentiated somatic cell type with another, offers insights into mucosal homeostasis as an adaptation to diverse inflammatory pathogenic stimuli. The cellular framework for metaplasia can be created by coordination of tissue-specific transcription factors (TFs) with epigenetic regulators of chromatin remodeling, allowing altered TF binding of inflammation-associated exposed genomic loci. For example, intestinal metaplasia (aberrant goblet cell differentiation and CDX2 expression, a pan-intestinal TF) is a common metaplastic response to chronic injury in the upper gastrointestinal tract, associated with increased risk of epithelial dysplasia and cancer. Inflammatory bowel disease (IBD) is an etiologically complex chronic inflammatory disease characterized by relapsing cycles of intestinal injury and healing. IBD patients carry increased risk of colitis-associated dysplasia and colorectal cancer (CAC) proportional to disease duration, extent, and severity. SATB2, or special AT-rich binding protein 2, is a chromatin organizing protein with homeostatic roles in osteoblastic, neural, and colonic differentiation. Intriguingly, others have recently shown that SATB2 is uniquely lost in human IBD dysplasia and CAC, in contrast to sporadic CRC where its expression is typically retained. (By contrast, CDX2 shows retained expression in both CAC and CRC.) Other recent work has shown that colon-specific Satb2 knockout in mice results in conversion of colon to small intestine, including emergence of villi and Paneth cells. Colonic mucosal biopsies from chronic IBD patients show histologic features reminiscent of small intestine with so-called Paneth cell metaplasia (PCM) and villiform, distorted growth. The molecular mediators of this epithelial remodeling in IBD are unknown. We hypothesized that SATB2 loss could result in a form of “small intestinal metaplasia” in human IBD colon. To investigate this, we selected a cohort of 20 IBD distal colon mucosal biopsies with marked crypt architectural changes and PCM and performed SATB2 immunostaining (n=7 Crohn’s disease, n=13 ulcerative colitis). Indeed, 19 of 20 such samples showed striking evidence of SATB2 loss. The pattern of SATB2 loss was heterogeneous; some tissue fragments showed complete epithelial loss as expected in small intestine, whereas others showed patchy loss in alternating crypts, even some showing heterogeneous expression within the same crypt. As positive and negative controls, we included normal distal colon and terminal ileum samples, observing expected intact and negative SATB2 expression, respectively. Of note, CDX2 was strongly expressed across the entire cohort, even in IBD tissue with loss of SATB2 expression. We propose that SATB2 loss is a pre-neoplastic metaplastic response to chronic injury in human colon. Future work will investigate how SATB2 loss enables a plasticity state permissive for emergence of TP53 mutations, dysplasia, and CAC. Citation Format: Maged Zeineldin, Reem Abu-Shamma, Tatianna Larman. Implicating SATB2 loss in a metaplasia-dysplasia sequence in human inflammatory bowel disease [abstract]. In: Proceedings of the AACR Special Conference on Colorectal Cancer; 2022 Oct 1-4; Portland, OR. Philadelphia (PA): AACR; Cancer Res 2022;82(23 Suppl_1):Abstract nr B026.