Abstract
BACKGROUND & AIMS: Stress granules (SGs) represent membrane-free cytoplasmic structures rapidly aggregating during cellular stress responses arguably useful as markers of molecular inflammation. To provide an automated, reproducible, and unbiased analytic workflow, we used the open-source software CellProfiler to quantify SGs in distinct cell types in inflammatory bowel disease. METHODS: The EpiCellProfiler (ECP) and PropiCellProfiler (PCP) pipelines enable segmentation within intestinal epithelial cells and lamina propria cells, respectively. The SG marker Ras GTPase-activating protein-binding protein 1 (G3BP1) was quantified for fluorescence intensity, granule size, and morphology on tissue sections of patients with ulcerative colitis (UC) and Crohn's disease (CD) in deep remission. RESULTS: Both pipelines detected elevated G3BP1 fluorescence intensities in inactive UC and CD. Additionally, SGs spot counts and spot sizes were increased in CD and UC compared with controls. The distribution of G3BP1 was homogenous in intestinal epithelial cells, without SG typical aggregations. In UC, PCP analysis revealed nuclear morphology alterations in terms of size, regularity, and compactness. CONCLUSIONS: Herein, we provide a powerful, reproducible, versatile and open-source software tool to quantify remnant molecular inflammation in patients with CD and UC, enabling research to openly share, reproduce and compare results within the field of quantitative image analysis. Our pipeline separates and distinguishes between epithelial and lamina propria events and provides insights into the spatial distribution and dynamics of SGs, revealing their homogeneous distribution and persistent accumulation in patients with CD and UC, notably in such without clinical, endoscopic, biochemical and histological disease activity. The sensitivity of the pipelines allows detection of subtle morphologic alterations that warrant further investigation, as does the usage of G3BP1 as an inflammatory bowel disease stress marker.