Wang Gaowei, Chiou Joshua, Zeng Chun, Miller Michael, Matta Ileana, Han Jee Yun, Kadakia Nikita, Okino Mei-Lin, Beebe Elisha, Mallick Medhavi, Camunas-Soler Joan, Dos Santos Theodore, Dai Xiao-Qing, Ellis Cara, Hang Yan, Kim Seung K, MacDonald Patrick E, Kandeel Fouad R, Preissl Sebastian, Gaulton Kyle J, Sander Maike
Department of Pediatrics, University of California San Diego, La Jolla CA, USA.
Pediatric Diabetes Research Center, University of California San Diego, La Jolla CA, USA.
bioRxiv. 2023 Jan 2:2022.12.31.522386. doi: 10.1101/2022.12.31.522386.
Altered function and gene regulation of pancreatic islet beta cells is a hallmark of type 2 diabetes (T2D), but a comprehensive understanding of mechanisms driving T2D is still missing. Here we integrate information from measurements of chromatin activity, gene expression and function in single beta cells with genetic association data to identify disease-causal gene regulatory changes in T2D. Using machine learning on chromatin accessibility data from 34 non-diabetic, pre-T2D and T2D donors, we robustly identify two transcriptionally and functionally distinct beta cell subtypes that undergo an abundance shift in T2D. Subtype-defining active chromatin is enriched for T2D risk variants, suggesting a causal contribution of subtype identity to T2D. Both subtypes exhibit activation of a stress-response transcriptional program and functional impairment in T2D, which is likely induced by the T2D-associated metabolic environment. Our findings demonstrate the power of multimodal single-cell measurements combined with machine learning for identifying mechanisms of complex diseases.
胰岛β细胞的功能改变和基因调控是2型糖尿病（T2D）的一个标志，但对驱动T2D的机制仍缺乏全面的了解。在这里，我们将单个β细胞中染色质活性、基因表达和功能测量的信息与遗传关联数据相结合，以确定T2D中疾病因果性的基因调控变化。通过对来自34名非糖尿病、糖尿病前期和T2D供体的染色质可及性数据进行机器学习，我们稳健地识别出两种在转录和功能上不同的β细胞亚型，它们在T2D中发生了丰度变化。定义亚型的活性染色质富含T2D风险变异，表明亚型身份对T2D有因果贡献。两种亚型在T2D中均表现出应激反应转录程序的激活和功能损伤，这可能是由T2D相关的代谢环境诱导的。我们的研究结果证明了多模态单细胞测量结合机器学习在识别复杂疾病机制方面的强大作用。
