Nuclei Density Analysis in A Mouse Autoimmunity Model Using Deep Learning-Based Segmentation
Year: 2026
Type: article
Abstract: Regulatory T cells (Tregs) are a specialized subset of CD4⁺ T cells that maintain immunological unresponsiveness to self-antigens and suppress excessive immune responses. Tregs are essential for maintaining peripheral tolerance, preventing autoimmune diseases, and limiting chronic inflammatory diseases. In humans, paucity or loss of functional Tregs leads to IPEX syndrome, a severe multi-organ autoimmune disorder that can be fatal in infancy [3]. In mice, Foxp3 deficiency produces the scurfy phenotype — fatal lymphoproliferative multi-organ infiltration within 3–4 weeks of birth. To explore the organ-wide effects of the loss of Treg cells, we utilized the Foxp3-DTR mouse model. In this model, diphtheria toxin administration drives selective depletion of Foxp3⁺ regulatory T cells (Tregs). The loss of Treg cells precipitates the proliferation and infiltrates of immune cells across many organs. Tregs are known to interact with various immune and non-immune cell types through cytokine secretion, metabolic control, and cell contact inhibition. We hypothesize that the removal of Tregs will trigger profound changes in cell density as well as cell type composition in many different tissues across the body. Hematoxylin and eosin (H&E) staining is a widely available histological technique that enables the visualization of tissue architecture and cell morphology. Whole-mouse scale H&E in our lab further enables simultaneous assessment of immune infiltration across multiple tissues in a single section, capturing organ-to-organ variation that single-tissue approaches cannot. Here, we deployed Cellpose, a generalist deep learning segmentation tool, to batch quantify nuclei across regions of interest across whole-mouse scans, enabling high-throughput nuclei counting across multiple organs and timepoints.
Authors: Jha, Richa, Mei, Linghan, Chevrier, Nicolas
Institution: University of Chicago
Language: English
Subfield: Biophysics
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W7155220362