Zhang Initiative Research Unit, Advanced Science Institute, RIKEN, 2-1 Hirosawa, Wako, Saitama 351-0198, Japan.
Curr Med Chem. 2012;19(30):5128-47. doi: 10.2174/092986712803530467.
Fragment based drug design has emerged as an effective alternative to high throughput screening for the identification of lead compounds in drug discovery in the past fifteen years. Fragment based screening and optimization methods have achieved credible success in many drug discovery projects with one approved drug and many more compounds in clinical trials. The fragment based drug design starts with the identification of fragments or low molecular weight compounds that generally bind with weak affinity to the target of interest. The fragments that form high quality interactions are then optimized to lead compounds with high affinity and selectivity. The weak affinity of fragments for their target requires the use of biophysical techniques such as nuclear magnetic resonance, X-ray crystallography or surface plasmon resonance to identify hits. These techniques are very sensitive and some of them provide detailed protein fragment interaction information that is important for fragment to lead optimization. Despite the huge advances in technology in the past years, experimental methods of fragment screening suffer several challenges such as low throughput, high cost of instruments and experiments, high protein and fragment concentration requirements. To address challenges posed by experimental screening approaches, computational methods were developed that play an important role in fragment library design, fragment screening and optimization of initial fragment hits. The computational approaches of fragment screening and optimization are most useful when they are used in combination with experimental approaches. The use of virtual fragment based screening in combination with experimental methods has fostered the application of fragment based drug design to important biological targets including protein-protein interactions and membrane proteins such as GPCRs. This review provides an overview of experimental and computational screening approaches used in fragment based drug discovery with an emphasis on recent successes achieved in discovering potent lead molecules using these approaches.
在过去的十五年中，基于片段的药物设计已成为高通量筛选的有效替代方法，可用于药物发现中先导化合物的鉴定。基于片段的筛选和优化方法在许多药物发现项目中取得了令人信服的成功，其中一种药物已获得批准，还有许多化合物正在临床试验中。基于片段的药物设计始于鉴定与目标具有弱亲和力的片段或低分子量化合物。然后，优化与目标形成高质量相互作用的片段，以获得高亲和力和选择性的先导化合物。片段与目标的弱亲和力需要使用生物物理技术，如核磁共振、X 射线晶体学或表面等离子体共振，以鉴定命中物。这些技术非常灵敏，其中一些提供了对片段到先导优化很重要的详细蛋白质片段相互作用信息。尽管过去几年技术取得了巨大进步，但片段筛选的实验方法仍面临一些挑战，例如低通量、仪器和实验成本高、对蛋白质和片段浓度要求高。为了解决实验筛选方法带来的挑战，开发了计算方法，这些方法在片段库设计、片段筛选和初始片段命中物的优化中发挥着重要作用。计算片段筛选和优化方法在与实验方法结合使用时最有用。虚拟片段筛选与实验方法的结合促进了基于片段的药物设计在包括蛋白质-蛋白质相互作用和 GPCR 等膜蛋白在内的重要生物靶标中的应用。本综述概述了基于片段的药物发现中使用的实验和计算筛选方法，重点介绍了使用这些方法发现有效先导分子的最新成功案例。
