Department of Chemistry, Georgia State University, Atlanta, Georgia 30302-3965, United States.
Center for Diagnostics and Therapeutics, Georgia State University, Atlanta, Georgia 30302-3965, United States.
J Chem Inf Model. 2023 Jun 26;63(12):3892-3902. doi: 10.1021/acs.jcim.3c00535. Epub 2023 Jun 7.
Drug resistance in antiviral treatments is a serious public health problem. Viral proteins mutate very fast, giving them a way to escape drugs by lowering drug binding affinity but with compromised function. Human immunodeficiency virus type I (HIV-1) protease, a critical antiretroviral therapeutic target, represents a model for such viral regulation under inhibition. Drug inhibitors of HIV-1 protease lose effectiveness as the protein evolves through several variants to become more resistant. However, the detailed mechanism of drug resistance in HIV-1 protease is still unclear. Here, we test the hypothesis that mutations throughout the protease alter the protein conformational ensemble to weaken protein-inhibitor binding, resulting in an inefficient protease but still viable virus. Comparing conformational ensembles between variants and the wild type helps detect these function-related dynamical changes. All analyses of over 30 μs simulations converge to the conclusion that conformational dynamics of more drug-resistant variants are more different from that of the wild type. Distinct roles of mutations during viral evolution are discussed, including a mutation predominantly contributing to the increase of drug resistance and a mutation that is responsible (synergistically) for restoring catalytic efficiency. Drug resistance is mainly due to altered flap dynamics that hinder the access to the active site. The mutant variant showing the highest drug resistance has the most ″collapsed″ active-site pocket and hence the largest magnitude of hindrance of drug binding. An enhanced difference contact network community analysis is applied to understand allosteric communications. The method summarizes multiple conformational ensembles in one community network and can be used in future studies to detect function-related dynamics in proteins.
抗病毒治疗中的耐药性是一个严重的公共卫生问题。病毒蛋白突变非常快，通过降低药物结合亲和力但降低功能，使它们能够逃避药物。人类免疫缺陷病毒 1 型（HIV-1）蛋白酶是一种关键的抗逆转录病毒治疗靶点，代表了这种在抑制下的病毒调节的模型。HIV-1 蛋白酶的药物抑制剂随着蛋白质通过几种变体进化而失去效力，从而变得更具耐药性。然而，HIV-1 蛋白酶耐药性的详细机制仍不清楚。在这里，我们检验了这样一个假设，即整个蛋白酶中的突变改变了蛋白质构象集合，从而削弱了蛋白质-抑制剂的结合，导致蛋白酶效率降低，但病毒仍然存活。比较变体和野生型之间的构象集合有助于检测这些与功能相关的动力学变化。超过 30 μs 模拟的所有分析都得出结论，即更多耐药变体的构象动力学与野生型的差异更大。讨论了突变在病毒进化过程中的不同作用，包括主要导致耐药性增加的突变和负责（协同）恢复催化效率的突变。耐药性主要是由于瓣动态的改变，阻碍了进入活性部位。显示出最高耐药性的突变变体具有最“塌陷”的活性部位口袋，因此药物结合的阻碍程度最大。应用增强的差异接触网络社区分析来理解变构通讯。该方法将多个构象集合总结在一个社区网络中，可用于未来的研究中，以检测蛋白质中的功能相关动力学。
