Gauto Diego F, Di Lella Santiago, Guardia Carlos M A, Estrin Darío A, Martí Marcelo A
Departamento de Quimica Inorganica, Analitica, y Quimica Fisica, INQUIMAE-CONICET, Universidad de Buenos Aires, Argentina.
J Phys Chem B. 2009 Jun 25;113(25):8717-24. doi: 10.1021/jp901196n.
Formation of protein ligand complexes is a fundamental phenomenon in biochemistry. During the process, significant solvent reorganization is produced along the contact surface and many water molecules strongly bound to the protein's ligand binding site must be displaced. Both the thermodynamics and kinetics of this process are complex and a clear understanding at the microscopic level has been not achieved so far. Special attention has been paid to the structure of water molecules on carbohydrate recognition sites of various proteins, and many studies support the idea that displacement of these water molecules should have a crucial effect on the binding free energy. Molecular dynamics (MD) simulations in explicit water solvent is a very promising approach for this type of studies. Using MD simulations combined with statistical mechanics analysis, thermodynamic properties of these water molecules can be computed and analyzed in a comparative view. Using this idea, we developed a set of analysis tools to link solvation with ligand binding in a key carbohydrate binding protein, human galectin-1 (hGal-1). Specifically, we defined water sites (WS) in terms of the thermodynamic properties of water molecules strongly bound to protein surfaces. In the present work, we selected a group of proteins whose ligand bound complexes have been already structurally characterized in order to extend the analysis of the role of the surface associated water molecules in the ligand binding and recognition process. The selected proteins are concanavalin-A (Con-A), galectin-3 (Gal-3), cyclophilin-A (Cyp-A), and two modules CBM40 and CBM32 of the multimodular bacterial sialidase. Our results show that the probability of finding water molecules inside the WS, p(v), with respect to the bulk density is directly correlated to the likeliness of finding an hydroxyl group of the ligand in the protein-ligand complex. This information can be used to analyze in detail the solvation structure of the carbohydrate recognition domain (CRD) and its relation to the possible protein ligand complexes and suggests addition of OH-containing functional groups to displace water from high p(v) WS to enhance drugs, specially glycomimetic-drugs, protein affinity, and/or specificity.
蛋白质配体复合物的形成是生物化学中的一种基本现象。在此过程中，沿着接触表面会产生显著的溶剂重组，并且许多与蛋白质配体结合位点紧密结合的水分子必须被取代。该过程的热力学和动力学都很复杂，到目前为止尚未在微观层面上实现清晰的理解。人们特别关注各种蛋白质碳水化合物识别位点上水分子的结构，许多研究支持这样的观点，即这些水分子的取代对结合自由能应该有至关重要的影响。在明确的水溶剂中进行分子动力学（MD）模拟是进行这类研究非常有前景的方法。通过将MD模拟与统计力学分析相结合，可以从比较的角度计算和分析这些水分子的热力学性质。基于这一想法，我们开发了一套分析工具，以将溶剂化与关键的碳水化合物结合蛋白人半乳糖凝集素-1（hGal-1）中的配体结合联系起来。具体而言，我们根据与蛋白质表面紧密结合的水分子的热力学性质定义了水位点（WS）。在本工作中，我们选择了一组其配体结合复合物已在结构上得到表征的蛋白质，以便扩展对表面相关水分子在配体结合和识别过程中作用的分析。所选蛋白质为伴刀豆球蛋白A（Con-A）、半乳糖凝集素-3（Gal-3）、亲环蛋白A（Cyp-A）以及多模块细菌唾液酸酶的两个模块CBM40和CBM32。我们的结果表明，相对于本体密度，在WS内发现水分子的概率p(v)与在蛋白质-配体复合物中发现配体羟基的可能性直接相关。该信息可用于详细分析碳水化合物识别结构域（CRD）的溶剂化结构及其与可能的蛋白质配体复合物的关系，并建议添加含羟基的官能团以将水从高p(v)的WS中取代出来，从而增强药物（特别是糖模拟药物）与蛋白质的亲和力和/或特异性。
