A novel spatial transcriptomics atlas developed by Northwestern Medicine scientists may improve the understanding of niche cellular interactions in the gastrointestinal tract that promote the development of inflammatory bowel diseases, as detailed in a recent study published in Nature Communications.
Parambir Dulai, MD, associate professor of Medicine in the Division of Gastroenterology and Hepatology, and Yuan Luo, PhD, professor of Preventive Medicine in the Division of Health and Biomedical Informatics, director of the Center for Collaborative AI in Healthcare, were co-corresponding authors of the study.
Inflammatory bowel diseases (IBD) cause chronic inflammation in the digestive tract and mainly consist of Crohn’s disease and ulcerative colitis. IBD impacts nearly three million individuals in the U.S. alone, according to the Centers for Disease Control and Prevention, and while there are no curative treatments, medications can help reduce symptoms and consequences from chronic inflammation.
Cellular signaling and cell-to-cell interactions in the intestinal microenvironment have been known to drive or reduce inflammation in IBD, and previous work has shown that characterizing single-cell transcriptomic signature data can help in identifying associations with IBD development, progression and treatment responsiveness.
Unfortunately, these methods currently lack the spatial context to properly analyze the organization and interactions of cells within gastrointestinal tissue, which can help reveal new biomarkers and inform future drug development, according to Dulai.
“With the growth in spatial omics, there was a tremendous need in the space for us to understand what is the best way to use the technology, which platform would be ideal to give us the most information possible, and is it feasible to start implementing this across large clinical trial programs or large consortiums so that we can start working towards a unified approach for reverse translational science in the field of IBD,” Dulai said.
To address this challenge, Dulai and his collaborators generated a large-scale spatial atlas containing more than three million cells using more than 100 intestinal tissue sections from patients with ulcerative colitis or Crohn’s disease, and from healthy controls.
The scientists then used their atlas to evaluate the interpretability, data quality, biological accuracy and variability of two imaging-based spatial transcriptomics platforms: the CosMX and Xenium multi-tissue panels.
“There’s a tremendous amount of natural variability in clinical research that we had to essentially mimic in a controlled experimental design to see how the platforms would perform and to see if we would lose meaningful biological insights across platforms due to this variability,” Dulai said.
From this comparison, the scientists found that CoxMx achieved higher detection efficiency than Xenium in both ulcerative colitis or Crohn’s disease tissue, while Xenium showed reduced performance associated with tissue type, quality and panel size. Additionally, they found that CosMx identified regulatory T-cell biology in both diseases, which was validated in laboratory experiments and using multi-plex spatial multi-omics techniques.
Going forward, Dulai said he and his collaborators will continue to update their dataset and improve the accessibility of their dataset for the scientific community. Dulai said they are also working towards implementing AI to make data analysis more efficient.
“All diseases are defined by their pathological features on histology and so our hope is that by creating the best ground truth information we can about what these cells are, what they’re doing and how they’re communicating to drive disease processes, that we can use that to better train digital pathology and AI models that can then be deployed more routinely and across clinical practice or clinical research,” Dulai said.
Co-authors include Wenjing Yang, PhD, research assistant professor of Medicine in the Division of Gastroenterology and Hepatology; Tianming Yu, PhD, research assistant professor of Medicine in the Division of Gastroenterology and Hepatology; Jenny Yanyi Ding, a student in the Health Sciences Integrated PhD Program; Stephen Hanauer, MD, the Clifford Joseph Barborka Professor of Medicine in the Division of Gastroenterology and Hepatology; Emanuelle Bellaguarda, MD, associate professor of Medicine in the Division of Gastroenterology and Hepatology; Laura Yun, MD, assistant professor of Medicine in the Division of Gastroenterology and Hepatology; Ronen Sumagin, PhD, associate professor of Pathology; Deyu Fang, PhD, the Hosmer Allen Johnson Professor of Pathology; and Yingzi Cong, PhD, the Stanley Gradowski Professor of Gastroenterology.
James Lewis, MD, professor of Medicine in the Divison of Gastroenterology at the University of Pennsylvania Perelman School of Medicine, was also a co-author of the study.
This study was funded with support from the National Institute of Diabetes, and Digestive and Kidney Diseases grant DK134321 PSD, DK135620 PSD, DK144103 PSD and DK138901. The Northwestern Medicine GI Biorepository was funded through a donation from the Digestive Health Foundation.
