GATA6-AS1 Regulates Intestinal Epithelial Mitochondrial Functions, and its Reduced Expression is Linked to Intestinal Inflammation and Less Favourable Disease Course in Ulcerative Colitis.

BACKGROUND AND AIMS: Widespread dysregulation of long non-coding RNAs [lncRNAs] including a reduction in GATA6-AS1 was noted in inflammatory bowel disease [IBD]. We previously reported a prominent inhibition of epithelial mitochondrial functions in ulcerative colitis [UC]. However, the connection between reduction of GATA6-AS1 expression and attenuated epithelial mitochondrial functions was not defined. METHODS: Mucosal transcriptomics was used to conform GATA6-AS1 reduction in several treatment-naïve independent human cohorts [n=673]. RNA pull-down followed by mass spectrometry was used to determine the GATA6-AS1 interactome. Metabolomics and mitochondrial respiration following GATA6-AS1 silencing in Caco-2 cells were used to elaborate on GATA6-AS1 functions. RESULTS: GATA6-AS1 showed predominant expression in gut epithelia using single cell datasets. GATA6-AS1 levels were reduced in Crohn's disease [CD] ileum and UC rectum in independent cohorts. Reduced GATA6-AS1 lncRNA was further linked to a more severe UC form, and to a less favourable UC course. The GATA6-AS1 interactome showed robust enrichment for mitochondrial proteins, and included TGM2, an autoantigen in coeliac disease that is induced in UC, CD and coeliac disease, in contrast to GATA6-AS1 reduction in these cohorts. GATA6-AS1 silencing resulted in induction of TGM2, and this was coupled with a reduction in mitochondrial membrane potential and mitochondrial respiration, as well as in a reduction of metabolites linked to aerobic respiration relevant to mucosal inflammation. TGM2 knockdown in GATA6-AS1-deficient cells rescued mitochondrial respiration. CONCLUSIONS: GATA6-AS1 levels are reduced in UC, CD and coeliac disease, and in more severe UC forms. We highlight GATA6-AS1 as a target regulating epithelial mitochondrial functions, potentially through controlling TGM2 levels.

Artificial Intelligence-based Analytics for Diagnosis of Small Bowel Enteropathies and Black Box Feature Detection.

OBJECTIVES: Striking histopathological overlap between distinct but related conditions poses a disease diagnostic challenge. There is a major clinical need to develop computational methods enabling clinicians to translate heterogeneous biomedical images into accurate and quantitative diagnostics. This need is particularly salient with small bowel enteropathies; environmental enteropathy (EE) and celiac disease (CD). We built upon our preliminary analysis by developing an artificial intelligence (AI)-based image analysis platform utilizing deep learning convolutional neural networks (CNNs) for these enteropathies. METHODS: Data for the secondary analysis was obtained from three primary studies at different sites. The image analysis platform for EE and CD was developed using CNNs including one with multizoom architecture. Gradient-weighted class activation mappings (Grad-CAMs) were used to visualize the models' decision-making process for classifying each disease. A team of medical experts simultaneously reviewed the stain color normalized images done for bias reduction and Grad-CAMs to confirm structural preservation and biomedical relevance, respectively. RESULTS: Four hundred and sixty-one high-resolution biopsy images from 150 children were acquired. Median age (interquartile range) was 37.5 (19.0-121.5) months with a roughly equal sex distribution; 77 males (51.3%). ResNet50 and shallow CNN demonstrated 98% and 96% case-detection accuracy, respectively, which increased to 98.3% with an ensemble. Grad-CAMs demonstrated models' ability to learn different microscopic morphological features for EE, CD, and controls. CONCLUSIONS: Our AI-based image analysis platform demonstrated high classification accuracy for small bowel enteropathies which was capable of identifying biologically relevant microscopic features and emulating human pathologist decision-making process. Grad-CAMs illuminated the otherwise "black box" of deep learning in medicine, allowing for increased physician confidence in adopting these new technologies in clinical practice.