Zheng Yi, Li Jincheng, Li Yucan, Wang Jiacheng, Suo Chen, Jiang Yanfeng, Jin Li, Xu Kelin, Chen Xingdong
State Key Laboratory of Genetic Engineering, Zhangjiang Fudan International Innovation Center, School of Life Sciences, Human Phenome Institute, Fudan University, Shanghai, China.
Department of Epidemiology, School of Public Health, and the Key Laboratory of Public Health Safety of Ministry of Education, Fudan University, Shanghai, China.
J Adv Res. 2024 Oct 25. doi: 10.1016/j.jare.2024.10.019.
Exploration of plasma proteins associated with osteoporosis can offer insights into its pathological development, identify novel biomarkers for screening high-risk populations, and facilitate the discovery of effective therapeutic targets.
The present study aimed to identify potential proteins associated with osteoporosis and to explore the underlying mechanisms from a proteomic perspective.
The study included 42,325 participants without osteoporosis in the UK Biobank (UKB), of whom 1,477 developed osteoporosis during the follow-up. We used Cox regression and Mendelian randomization analysis to examine the association between plasma proteins and osteoporosis. Machine learning was utilized to explore proteins with strong predictive power for osteoporosis risk.
Of 2,919 plasma proteins, we identified 134 significantly associated with osteoporosis, with sclerostin (SOST), adiponectin (ADIPOQ), and creatine kinase B-type (CKB) exhibiting strong associations. Twelve of these proteins showed significant associations with bone mineral density (BMD) T-score at the femoral neck, lumbar spine, and total body. Mendelian randomization further supported causal relationships between 17 plasma proteins and osteoporosis. Moreover, follitropin subunit beta (FSHB), SOST, and ADIPOQ demonstrated high importance in predictive modeling. Utilizing a predictive model built with 10 proteins, we achieved relatively accurate prediction of osteoporosis onset up to 5 years in advance (AUC = 0.803). Finally, we identified three osteoporosis-related protein modules associated with immunity, lipid metabolism, and follicle-stimulating hormone (FSH) regulation from a network perspective, elucidating their mediating roles between various risk factors (smoking, sleep, physical activity, polygenic risk score (PRS), and menopause) and osteoporosis.
We identified several proteins associated with osteoporosis and highlighted the role of plasma proteins in influencing its progression through three primary pathways: immunity, lipid metabolism, and FSH regulation. This provides further insights into the distinct molecular patterns and pathogenesis of bone loss and may contribute to strengthening early diagnosis and long-term monitoring of the condition.
探索与骨质疏松症相关的血浆蛋白有助于深入了解其病理发展过程，识别用于筛查高危人群的新型生物标志物，并推动有效治疗靶点的发现。
本研究旨在从蛋白质组学角度识别与骨质疏松症相关的潜在蛋白质，并探究其潜在机制。
该研究纳入了英国生物银行（UKB）中42325名无骨质疏松症的参与者，其中1477人在随访期间患上了骨质疏松症。我们使用Cox回归和孟德尔随机化分析来研究血浆蛋白与骨质疏松症之间的关联。利用机器学习探索对骨质疏松症风险具有强大预测能力的蛋白质。
在2919种血浆蛋白中，我们鉴定出134种与骨质疏松症显著相关，其中硬化蛋白（SOST）、脂联素（ADIPOQ）和肌酸激酶B型（CKB）表现出强烈关联。这些蛋白中有12种与股骨颈、腰椎和全身的骨密度（BMD）T值显著相关。孟德尔随机化进一步支持了17种血浆蛋白与骨质疏松症之间的因果关系。此外，促卵泡激素β亚基（FSHB）、SOST和ADIPOQ在预测模型中显示出高度重要性。利用由10种蛋白质构建的预测模型，我们能够提前5年相对准确地预测骨质疏松症的发病（曲线下面积 = 0.803）。最后，我们从网络角度识别出三个与免疫、脂质代谢和促卵泡激素（FSH）调节相关的骨质疏松症相关蛋白模块，阐明了它们在各种风险因素（吸烟、睡眠、身体活动、多基因风险评分（PRS）和绝经）与骨质疏松症之间的介导作用。
我们鉴定出了几种与骨质疏松症相关的蛋白质，并强调了血浆蛋白通过免疫、脂质代谢和FSH调节这三条主要途径影响其进展的作用。这为深入了解骨质流失的独特分子模式和发病机制提供了进一步的见解，并可能有助于加强对该疾病的早期诊断和长期监测。
