COMPUTATIONAL MODELING OF INTERACTIONS BETWEEN HYALURONAN AND LINK MODULES

这个子项目是许多研究子项目中利用 资源由NIH/NCRR资助的中心拨款提供。子项目和 调查员(PI)可能从NIH的另一个来源获得了主要资金， 并因此可以在其他清晰的条目中表示。列出的机构是 该中心不一定是调查人员的机构。 在脊椎动物中，透明质酸(HA)是一种细胞外基质多糖，以游离状态存在于软结缔组织中，在软结缔组织中传递其粘弹性特性，并作为细胞间隙填充物。HA还可以与蛋白质相互作用(通过Link模块结构域)，并可以形成大的大分子复合体，这些复合体对细胞外基质的组装和重塑非常重要，并在软骨和其他组织中提供膨胀。目前，HA的局部溶液结构还存在争议，对大分子HA-蛋白质复合体的三维结构知之甚少，从而限制了我们对细胞外基质和各种疾病，如关节炎、癌症和不孕不育的了解。最近，我们已经展示了如何使用HA片段(例如，8个糖的长度)的原子尺度模拟来帮助解释核磁共振(NMR)实验。模拟结果揭示了透明质酸在溶液中的局部构象状态，它是决定透明质酸粘弹性和流体力学性质的重要因素。同样，从核磁共振数据中获得了来自蛋白质TSG-6(记为Link_TSG6)的HA结合结构域(Link模块)的高分辨结构，该结构既有HA结合的，也有未连接的。通过单独使用核磁共振，还不可能表征HA配体从头开始的取向/构象。然而，使用实验数据和来自分子模拟的结构约束有助于构建二元络合物的初步模型(1)。已经为其他包含蛋白质的Link模块产生了表达构建体，即aggrecan和Verscan的G1结构域，软骨链接蛋白和Link Protein 3，它们都在其HA结合域中包含一对相连的Link模块。这些蛋白质结构域促进蛋白多糖与透明质酸链的结合，从而形成大分子聚集体。这种相互作用通常是通过透明质酸、蛋白多糖和稳定的‘连接蛋白’之间的三元复合体来调节的，从而导致一种基本上不可分离的结构。同源建模，结合从可能的HA结合位点推导出的约束，已经允许为这些蛋白质构建双链模块的模型(2)。研究重点：我们的主要目标是(A)表征HA和各种包含蛋白质的Link模块之间发生的分子相互作用，以及(B)通过使用新的化合物，如特定同位素浓缩的透明质酸和特定的HA结合蛋白表达结构，结合高场核磁共振和X射线结晶学来确定超分子HA-蛋白质复合体的三维结构。然而，HA聚合物是动态的，充分了解HA与其结合的蛋白质之间的分子相互作用将需要进行分子动力学研究。在这方面，我们已经在显式溶剂和离子存在的情况下使用全原子方法对TSG-6的Link模块进行了建模，并表明与实验核磁共振数据有很大的重叠。我们的目标是使用计算建模来帮助理解Link模块和HA之间的动态相互作用及其超分子组织。特别是，我们打算使用全原子分子模型和最终的粗粒度分子模型来模拟HA-蛋白质复合体。我们的目标是模拟大型分子复合体，这需要比我们目前拥有的更多的计算资源。这个PSC项目有三个具体的目标：i)使用新获得的核磁共振数据中的约束条件对蛋白质-碳水化合物复合体进行分子建模。Ii)使用显式溶剂和离子对蛋白质-碳水化合物复合体进行全原子分子动力学模拟，以提供有关界面动力学的信息并帮助解释实验数据。Iii)用粗粒模型模拟多个蛋白质与大分子HA之间的相互作用。计算要求：使用CHARMM程序对Lemieux、Rachel或Jonas进行20-30,000个原子的分子力学最小化/分子动力学计算。参考文献1.Blundell，C.D.；MaHony，D.J.；Almond，A.；DeAngelis，P.L.；Kahmann，J.D.；Teriete，P.；Pickford，A.R.；Campbell，I.D.；Day，A.J.，J Biol Chem 2003，第278,49261-70页。2.布伦德尔，C.D.；阿蒙德，A.；马奥尼，D.J.；DeAngelis，P.L.；坎贝尔，I.D.；戴，A.J.，《生物化学》，2005年，第280页，18189-201年。

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. In vertebrates, hyaluronan (HA) is an extracellular matrix polysaccharide that occurs in a free state in soft connective tissue where it conveys its viscoelastic properties and acts as an intercellular spacefiller. HA also interacts with proteins (via Link module domains) and can form large macromolecular complexes that are important in extracellular matrix assembly and remodelling and provide turgor in cartilage and other tissue. At present, the local solution structure of HA is subject to debate and little is known about the three dimensional organisation of the macromolecular HA-protein complexes thus limiting our understanding of the extracellular matrix and various diseases, e.g., arthritis, cancer, and infertility. Recently, we have shown how atomic-scale simulations of HA fragments (e.g., 8 saccharides long) may be used to help interpret nuclear magnetic resonance (NMR) experiments. Simulations have provided much insight into the local conformational state of hyaluronan in solution which is an important determinant for the viscoelastic and hydrodynamic properties of HA. Similarly, a high-resolution structure for a recombinant HA-binding domain (Link module) from the protein TSG-6 (denoted Link_TSG6) in both its HA-bound and un-liganded forms, has been obtained from NMR data. By exclusive use of NMR it has not been possible to characterize the orientation/conformation of the HA ligand de novo. However, the use of experimental data and structural constrains from molecular modeling have facilitated construction of a preliminary model of the binary complex (1). Expression constructs have been produced for other Link module containing proteins, namely the G1-domains of aggrecan and versican, cartilage link protein and link protein 3 that all contain a contiguous pair of Link modules in their HA-binding domains. These protein domains facilitate the attachment of proteoglycans to HA chains, responsible for the formation of large macromolecular aggregates. The interaction is often mediated through a ternary complex between HA, the proteoglycan and a stabilising 'link protein' leading to a structure that is essentially undissociable. Homology modeling, combined with constraints deduced from the likely HA binding site, has allowed models of the double link modules to be constructed for these proteins (2). RESEARCH FOCUS: Our main goals are (a) characterizing the molecular interactions that occur between HA and the variety of Link module containing proteins and (b) determining the three dimensional organisation of supramolecular HA-protein complexes by using novel compounds such as specifically isotopically enriched hyaluronan and specific HA-binding protein expression constructs together with high-field NMR and X-ray crystallography. However, the HA polymer is dynamic and a full understanding of the molecular interactions between HA and the proteins it binds will necessitate an investigation of molecular dynamics. In this regard, we have already modeled the Link module from TSG-6 using an all-atom approach in the presence of explicit solvent and ions and have shown that there is a significant overlap with experimental NMR data. We aim to use computational modeling to aid the process of understanding dynamic interactions between Link modules and HA and their supramolecular organisation. In particular, we intend to model HA-protein complexes using all-atom molecular modelling and eventually coarse-grained molecular modelling. The large molecular complexes that we are aiming to model necessitate more computational resources than we currently have at our disposal. This PSC project has three Specific Aims: I) Molecular modeling of protein-carbohydrate complexes using restraints from newly obtained NMR data. II) All-atom molecular dynamics simulations of protein-carbohydrate complexes using explicit solvent and ions to provide information about interfacial dynamics and to help interpret experimental data. III) Model interactions between multiple proteins and large molecular mass HA using coarse-grained modeling. Computational Requirements: Molecular mechanics minimization/molecular dynamics of 20-30,000 atoms with CHARMM program on Lemieux, Rachel or Jonas. REFERENCES 1. Blundell, C. D.; Mahoney, D. J.; Almond, A.; DeAngelis, P. L.; Kahmann, J. D.; Teriete, P.; Pickford, A. R.; Campbell, I. D.; Day, A. J., J Biol Chem 2003, 278, 49261-70. 2. Blundell, C. D.; Almond, A.; Mahoney, D. J.; DeAngelis, P. L.; Campbell, I. D.; Day, A. J., J Biol Chem 2005, 280, 18189-201.