计算生物学视角下的骨老化与再生机制
本文为综述,围绕骨基质中胶原与羟基磷灰石的结构与相互作用、非胶原蛋白对骨矿化的调控作用,以及以硬化蛋白、RANKL和雌激素受体为靶点的骨再生药物研发三个方向,梳理计算方法在骨生物学与治疗策略研究中的贡献,并讨论机器学习与多尺度建模的应用前景。
要点
- 分子动力学与分子对接可用于筛选和优化作用于关键通路的天然及合成骨再生治疗药物。
- 非胶原蛋白如骨形态发生蛋白、骨钙素、骨桥蛋白和纤连蛋白参与调控骨矿化过程。
- 需更精确的计算技术重建骨再生与退变通路的相互作用网络,并结合实验验证与临床数据。
核心结论
- 计算方法有助于理解骨重塑、再生与稳态的分子机制,并推动个体化骨质疏松治疗策略的开发。
研究设计与样本
- 综述类文章,未报告系统检索范围与纳入文献数量,聚焦骨基质结构、非胶原蛋白调控矿化及骨再生药物研发三个领域。
主要发现
- 分子动力学与对接方法已用于识别和优化作用于硬化蛋白、RANKL和雌激素受体等关键通路的天然与合成治疗药物。
- 生物信息学工具被用于分析骨再生与退变通路,但重建其相互作用组仍需更精确的计算技术。
临床怎么用
- 可将计算预测与实验验证及临床数据结合,用于骨再生靶点筛选与治疗药物优化,服务个体化治疗。
局限性
- 综述未报告检索与纳入标准及文献数量,且现有计算技术重建相互作用组的准确性仍有限。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
Bone degeneration diseases, such as osteoporosis, are skeletal disorders characterized by diminished bone mass and increased susceptibility to fractures and represent a growing global health challenge, particularly in aging populations. The development of effective therapeutic strategies necessitates a deep understanding of the complex biological processes underlying bone remodeling, regeneration, and homeostasis. To address these challenges, computational approaches have played a crucial role in advancing our understanding of bone biology and improving therapeutic strategies. This review explores these contributions across three main areas: (1) elucidating the structural organization and interactions within the bone matrix, particularly between collagen and hydroxyapatite; (2) investigating the regulatory roles of non-collagenous proteins, such as bone morphogenetic proteins, osteocalcin, osteopontin, and fibronectin, in bone mineralization; and (3) facilitating drug discovery and development for bone regeneration by targeting key pathways and molecules, including sclerostin, RANKL, and estrogen receptors. Molecular dynamics and docking have helped identify and optimize natural and synthetic therapeutic agents for these critical pathways. Additionally, we apply bioinformatics tools to analyze bone regeneration and degeneration pathways, emphasizing the need for more accurate computational techniques to reconstruct their interactome. As these techniques continue to evolve, integrating advancements in machine learning, molecular dynamics, and multi-scale modeling, their potential to bridge the gap between experimental research and clinical application is becoming increasingly apparent. A multidisciplinary approach that combines computational predictions with experimental validation and clinical data is poised to drive the development of personalized and effective osteoporosis therapies, ultimately reducing the global burden of this debilitating disease.
引用
- Stojceski F, Zaverdas H, Danani A, et al. Molecular-level understanding of the aging bone and regeneration mechanisms using computational methods. Osteoporos Int. 2026. PMID: 41805841. doi:10.1007/s00198-026-07903-z
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。