Sun Zhuolun, Kemter Elisabeth, Pang Yingxian, Bidlingmaier Martin, Wolf Eckhard, Reincke Martin, Williams Tracy Ann
Medizinische Klinik und Poliklinik IV, Klinikum der Universität München, Ludwig-Maximilians-Universität München, Germany (Z.S., Y.P., M.B., M.R., T.A.W.).
Chair for Molecular Animal Breeding and Biotechnology, Gene Center and Department of Veterinary Sciences, Ludwig-Maximilians-Universität München, Germany (E.K., E.W.).
Hypertension. 2025 Feb;82(2):319-332. doi: 10.1161/HYPERTENSIONAHA.124.23817. Epub 2024 Dec 2.
Aldosterone-producing adenomas (APAs) are a common cause of primary aldosteronism that can lead to cardiovascular complications if left untreated. Machine learning-based bioinformatics approaches have emerged as powerful tools for identifying potential disease markers, gaining widespread recognition in biomedical research. We aimed to use machine learning to discover novel biomarkers of APAs to identify new pathophysiological mechanisms.
We applied 2 machine learning algorithms to published RNA sequencing data to identify APA feature genes. Validation was performed using APA tissue samples, spatial transcriptomics, pig adrenal glands, and in vitro assays in a human adrenocortical cell line.
Machine learning identified as a key feature gene in APA, and its upregulation in APAs compared with the adjacent cortex was confirmed by spatial transcriptomics. In human adrenocortical cells, angiotensin II treatment increased gene expression 9.15-fold. Silencing decreased basal expression and aldosterone secretion by 3.51-fold and 1.46-fold, respectively, and by 1.77-fold and 1.94-fold under angiotensin II stimulation. Dietary sodium restriction in pigs significantly increased mRNA and protein levels. Spatial transcriptomics showed that APA cells exhibited higher gene expression compared with all other adrenal cell types. The suppressive effect of silencing on expression was further enhanced by Ca inhibitors.
The gene is highly expressed in APA and is a key regulator of expression and aldosterone production.
醛固酮瘤（APAs）是原发性醛固酮增多症的常见病因，若不治疗可导致心血管并发症。基于机器学习的生物信息学方法已成为识别潜在疾病标志物的强大工具，在生物医学研究中得到广泛认可。我们旨在利用机器学习发现醛固酮瘤的新型生物标志物，以确定新的病理生理机制。
我们将两种机器学习算法应用于已发表的RNA测序数据，以识别醛固酮瘤特征基因。使用醛固酮瘤组织样本、空间转录组学、猪肾上腺以及在人肾上腺皮质细胞系中进行体外试验进行验证。
机器学习确定[具体基因名称未给出]为醛固酮瘤中的关键特征基因，空间转录组学证实其在醛固酮瘤中相对于相邻皮质上调。在人肾上腺皮质细胞中，血管紧张素II处理使[具体基因名称未给出]基因表达增加9.15倍。沉默[具体基因名称未给出]分别使基础[具体基因名称未给出]表达和醛固酮分泌降低3.51倍和1.46倍，在血管紧张素II刺激下分别降低1.77倍和1.94倍。猪的饮食钠限制显著增加[具体基因名称未给出]mRNA和蛋白水平。空间转录组学显示，与所有其他肾上腺细胞类型相比，醛固酮瘤细胞表现出更高的[具体基因名称未给出]基因表达。钙抑制剂进一步增强了沉默[具体基因名称未给出]对[具体基因名称未给出]表达的抑制作用。
[具体基因名称未给出]基因在醛固酮瘤中高度表达，是[具体基因名称未给出]表达和醛固酮产生的关键调节因子。
