Emerging evidences indicate bromodomain-containing proteins 2 and 4 (BRD2 and BRD4) play critical roles in cancers, inflammations, cardiovascular diseases and other pathologies. Multiple short molecular dynamics (MSMD) simulations combined with molecular mechanics generalized Born surface area (MM-GBSA) method were applied to investigate the binding selectivity of three inhibitors 87D, 88M and 89G towards BRD2 over BRD4. The root-mean-square fluctuation (RMSF) analysis indicates that the structural flexibility of BRD4 is stronger than that of BRD2. Moreover the calculated distances between the C_α atoms in the centres of the ZA_loop and BC_loop of BRD4 are also bigger than that of BRD2. The rank of binding free energies calculated using MM-GBSA method agrees well with that determined by experimental data. The results show that 87D can bind more favourably to BRD2 than BRD4, while 88M has better selectivity on BRD4 over BRD2. Residue-based free-energy decomposition method was utilized to estimate the inhibitor-residue interaction spectrum and the results not only identify the hot interaction spots of inhibitors with BRD2 and BRD4, but also demonstrate that several common residues, including (W370, W374), (P371, P375), (V376, V380) and (L381, L385) belonging to (BRD2, BRD4), generate significant binding difference of inhibitors to BRD2 and BRD4.

越来越多的证据表明，含溴结构域的蛋白质2和4（BRD2和BRD4）在癌症，炎症，心血管疾病和其他疾病中起着关键作用。结合分子动力学广义博恩表面积（MM-GBSA）方法，结合多种短分子动力学（MSMD）模拟，研究三种抑制剂87D，88M和89G对BRD2的结合选择性高于BRD4。均方根波动（RMSF）分析表明，BRD4的结构柔韧性强于BRD2。此外，该C之间的计算的距离_αBRD4的ZA_loop和BC_loop中心的原子也大于BRD2的原子。使用MM-GBSA方法计算的结合自由能的等级与实验数据确定的等级非常吻合。结果表明，87D比BRD4更能与BRD2结合，而88M对BRD4的选择性优于BRD2。利用基于残基的自由能分解法估算了抑制剂-残基的相互作用谱，结果不仅确定了抑制剂与BRD2和BRD4的热相互作用点，还证明了几种常见的残基，包括（W370，W374），属于（BRD2，BRD4）的（P371，P375），（V376，V380）和（L381，L385）会产生抑制剂与BRD2和BRD4的显着结合差异。