Molecular modeling of carborane-containing drug molecules

碳硼烷簇合物及其衍生物近年来成为药物化学和药物设计领域的研究热点。用碳硼烷取代有机环系统被证明是一种成功的策略，用于新型和改进的药物分子。然而，到目前为止，只有少数与靶蛋白结合的含碳硼烷药物分子的晶体结构是已知的。在缺乏实验数据的情况下，本文提出的分子建模研究将提供对含碳硼烷的药物分子与其靶生物分子的结合模式以及动态过程（例如碳硼烷与周围蛋白质之间的氢键形成和解离）的深入了解。经典的分子力学（MM）模拟碳硼烷蛋白质系统是不容易实现的，因为良好的校准和通用力场参数的碳硼烷还没有。目前，只有对接模拟或QM/MM计算可以用来描述含碳硼烷的药物分子和生物分子之间的相互作用。对接模拟可以合理地预测这些配体的结合模式，但未能提供准确的结合自由能值。获得具有低误差容限的自由能值对于研究诸如环氧合酶（考克斯）抑制剂的选择性等问题是必不可少的。此外，由于缺乏足够的碳硼烷力场参数，从对接结果获得的含碳硼烷的候选药物的排名不能在MM自由能计算的帮助下进行细化。QM/MM方法的计算要求的性质限制了分子动力学（MD）模拟的时间尺度，并限制了可实现的计算类型。因此，本项目的主要目标是开发和验证含碳硼烷药物分子（如吲哚美辛衍生物）的MM力场参数。新开发的MM力场将允许访问更长的时间尺度MD模拟，在相当少的计算成本，并将允许访问自由能计算使用MM为基础的方法。此外，新开发的MM力场参数将为类似的含碳硼烷分子的参数化提供一个起点。良好校准的碳硼烷力场将大大有助于发现和研究潜在的生物活性的含碳硼烷化合物。

Carborane clusters and their derivatives have recently become the focus of intense research in the fields of medicinal chemistry and drug design. The substitution of organic ring systems with carboranes proved to be a successful strategy towards novel and improved drug molecules. However, only a few crystal structures of carborane-containing drug molecules bound to target proteins are known so far. In lack of experimental data, the molecular modeling studies proposed herein will provide insight into the binding mode of carborane-containing drug molecules to their target biomolecules and into dynamic processes such as dihydrogen bond formation and dissociation between carboranes and the surrounding proteins. Classical molecular mechanics (MM) simulations on carborane-protein systems are not readily achievable, since well calibrated and general purpose force field parameters for carboranes are not yet available. Currently only docking simulations or QM/MM calculations can be used to describe the interactions between carborane-containing drug molecules and biomolecules. Docking simulations can reasonably predict the binding modes of these ligands, but fail to provide accurate binding free energy values. Obtaining free energy values with a low error margin is essential for investigating problems such as the selectivity of cyclooxygenase (COX) inhibitors. Furthermore, the ranking of carborane-containing drug candidates obtained from docking results cannot be refined with the aid of MM free energy calculations due to the lack of adequate carborane force field parameters. The computationally demanding nature of the QM/MM method limits the time scale of molecular dynamic (MD) simulations and restricts the type of achievable calculations. Thus, the main objective of this project is to develop and validate MM force field parameters for carborane-containing drug molecules, such as indomethacin derivatives. The newly developed MM force fields would grant access to much longer time scale MD simulations, at considerably less computational costs, and would grant access to free energy calculations using MM based methods. Furthermore, the newly developed MM force field parameters will provide a starting point for parameterization of similar carborane-containing molecules. A well calibrated carborane force field will considerably aid the discovery and study of potential biologically active carborane-containing compounds.