海藻糖和氨基酸之间相互作用的分子动力学模拟

虽然海藻糖已经广泛用于蛋白质稳定性研究,但海藻糖稳定蛋白质的作用机理尚不清晰. 本文利用全原子分子动力学模拟研究了20种常见氨基酸和海藻糖之间的分子机理. 结果表明,所有氨基酸,尤其是极性和带电氨基酸,均优先与水分子结合. 相反,仅有疏水性氨基酸与海藻糖发生相互作用,尤其是芳香族和疏水性氨基酸的侧链更易于和海藻糖接触. 所有氨基酸的主链与水分子接触的趋势一致. 虽然氨基酸和海藻糖与水之间均形成氢键,但氨基酸和海藻糖之间的氢键相互作用要弱于氨基酸和水之间的氢键相互作用. 上述分子模拟的结果对于海藻糖稳定蛋白质作用机理的解析及高效蛋白质稳定剂的理性设计具有非常重要的理论指导意义.     

光谱法研究尿素对水溶液中血红蛋白构象的影响

应用荧光猝灭法和动态光散射法测定尿素-水混合溶剂中血红蛋白(Hb)与联苯胺的结合距离和Hb的流体动力学半径. 结合Hb的荧光光谱和吸收光谱, 探讨尿素与蛋白质分子在水溶液中相互作用的机理及其对蛋白质构象的影响. 结果显示, 尿素分子取代水分子在蛋白质周围形成溶剂化层, 并与骨架肽链和亲水侧链形成氢键, 从而积聚在蛋白质分子表面. 尿素分子与蛋白质分子之间的直接相互作用对蛋白质的构象具有复杂的影响, 高浓度的尿素-水混合溶剂破坏蛋白质的构象, 而低浓度的混合溶剂则有利于蛋白质形成更紧密的构象. 在高浓度的尿素-水混合溶剂中, Hb血红素疏水空穴失去原有的三级结构后形成一个与熔球态相类似的结构.     

4-巯乙基吡啶配基与IgG相互作用的分子模拟研究

4-巯乙基吡啶(MEP)是一种新型的疏水性电荷诱导层析(HCIC)配基,具有良好的抗体分离性能,采用分子模拟方法研究MEP配基与IgG间的分子相互作用,以探讨HCIC分离机制.先通过分子对接搜索IgG分子Fc片段A链的蛋白表面,确定12个可能与MEP结合的位点区域,然后用分子动力学模拟考察了其中6个位点的结合性能.结果表明,MEP在Fc-A链表面的结合具有疏水倾向性,pH中性条件下,MEP能稳定地结合在TYR319和LEU309附近的位点,并与两者形成氢键作用,该区域具有疏水性极强的口袋结构;其他位点的结合较不稳定,受MEP取向影响大.在pH4.0酸性条件下,原先稳定结合的MEP快速从Fc-A链表面脱离,主要原因是MEP和Fc间的静电排斥作用,以及疏水作用减弱和氢键结合的消失.通过分子模拟方法,从分子水平验证了HCIC独特的作用机理:疏水相互作用主导吸附,静电排斥作用协助解吸.     

多肽抑制剂抑制淀粉质多肽42构象转换的分子动力学模拟和结合自由能计算

应用分子动力学模拟和结合自由能计算方法研究了多肽抑制剂KLVFF、VVIA和LPFFD抑制淀粉质多肽42 (Aβ42)构象转换的分子机理. 结果表明, 三种多肽抑制剂均能够有效抑制Aβ42的二级结构由α-螺旋向β-折叠的构象转换. 另外, 多肽抑制剂降低了Aβ42分子内的疏水相互作用, 减少了多肽分子内远距离的接触, 有效抑制了Aβ42的疏水塌缩, 从而起到稳定其初始构象的作用. 这些抑制剂与Aβ42之间的疏水和静电相互作用(包括氢键)均有利于它们抑制Aβ42的构象转换. 此外, 抑制剂中的带电氨基酸残基可以增强其和Aβ42之间的静电相互作用(包括氢键), 并降低抑制剂之间的聚集, 从而大大增强对Aβ42构象转换的抑制能力. 但脯氨酸的引入会破坏多肽的线性结构, 从而大大降低其与Aβ42 之间的作用力. 上述分子模拟的结果揭示了多肽抑制剂KLVFF、VVIA和LPFFD抑制Aβ42构象转换的分子机理, 对于进一步合理设计Aβ的高效短肽抑制剂具有非常重要的理论指导意义.     

海藻糖抑制淀粉质多肽42构象转变的分子动力学模拟

应用分子动力学模拟方法研究了海藻糖抑制淀粉质多肽42(Aβ42)构象转变的分子机理.结果表明,海藻糖溶液浓度对Aβ42构象转变具有非常重要的影响.在水和低浓度海藻糖溶液(0.18mol·L-1)中,Aβ42可由初始的α-螺旋结构转变成β-折叠的二级结构;但海藻糖浓度为0.37mol·L-1时即可有效抑制Aβ42的构象转变.这是因为海藻糖利用其优先排阻作用使水分子在多肽周围0.2nm内富集,而其自身却在距离多肽0.4nm的位置附近团聚.另外,海藻糖还可通过降低多肽间的疏水相互作用,减少多肽分子内远距离的接触,有效抑制多肽的疏水塌缩和构象转变.上述分子模拟的结果对于进一步合理设计阿尔茨海默病的高效抑制剂具有非常重要的理论指导意义.     

蛋白质-蛋白质相互作用界面统计分析

以作者开发的从蛋白质结合部位推导出其界面所具有的疏水性质和氢键性质的计算程序PP_SITE为基础,利用蛋白质结构数据库(PDB),对蛋白质-蛋白质相互作用界面进行了统计分析.从PDB中挑出非冗余的链间相互作用对,计算出这个数据集中所有链间界面的疏水和氢键相互作用特征.对得到的界面特征进行统计分析,寻找能够明显聚类的界面特征.结果表明，界面大小、氢键和疏水相互作用在界面所占比例以及疏水相互作用的集中程度可以作为分类的依据.     

确定分子量的聚丙烯酸-β-羟乙酯的合成及其水溶液中水氢键缺损的拉曼光谱学研究

水溶性高分子经化学交联可得水凝胶 ,水凝胶也可由水溶性高分子经物理交联如部分结晶微区 ,疏水相互作用及缠绕交联得到 .线型水溶性高分子在水中以高浓度溶解时 ,高分子链之间相互搭迭缠绕也能形成物理“交联” .Ｋｉｔａｎｏ[1～ 4 ] 等研究了聚环氧乙烷、聚丙烯酸、聚丙烯酰胺和聚乙烯基吡咯烷酮等亲水性高分子水溶液中水分子间的氢键缺损情况 ,在重量浓度相同的情况下 ,随着分子量的增大 ,水分子间的氢键缺损加剧 .他们认为分子量大 ,高分子之间的缠绕就比较严重 ,在水溶液中就会形成许多小的微区 ,水分子在这些小微区中形成分子间氢键…     

碱/脲水溶液体系中纤维素包合物构型及纤维素与溶剂分子间相互作用力的分子动力学模拟

采用分子动力学模拟方法研究了纤维素分子在碱/脲水溶液体系中形成的包合物结构,研究了纤维素包合物的空间构型、氢键网格结构、纤维素分子与溶剂分子的相互作用以及碱金属阳离子对包合物稳定性的影响.在纤维素包合物结构中,碱金属阳离子和OH-主要吸附在纤维素分子链羟基的附近,与纤维素上的羟基氧直接接触形成稳定的吸附构型;尿素分子更倾向于在纤维素糖环面结构上聚集,可以与纤维素上的羟基氧和醚键氧相互作用形成氢键.通过对纤维素与溶剂分子间非键相互作用的研究发现,在纤维素羟基附近,羟基与金属阳离子之间的相互作用能最大,其次为与尿素分子、氢氧根离子的相互作用,最小的为与水分子的相互作用;在纤维素糖环面结构上,Na~+、OH~-、尿素、水与纤维素醚键氧的相互作用远小于与纤维素羟基的相互作用,纤维素上的醚键氧与尿素分子相互作用能最大.比较KOH/尿素和NaOH/尿素2种溶剂体系中碱金属阳离子与纤维素羟基形成的吸附构型的结合能,发现Na~+对纤维素分子内和分子间的氢键具有更强的破坏作用,NaOH/尿素溶剂体系中的分子与纤维素分子形成的包合物构型更稳定.     

核糖核酸酶Sa表面水和尿素分子的分布和动力学行为的分子动力学模拟

利用分子动力学模拟研究了在不同尿素浓度下,核糖核酸酶Sa(RNase Sa)表面水和尿素分子的分布和动力学行为。 结果表明,尿素分子可与RNase Sa酶形成较强的相互作用,并取代其表面的水分子而富集在蛋白质表面。 尿素分子更倾向与RNase Sa酶的疏水残基作用,与RNase Sa酶主链形成氢键的能力更强。 尿素分子的平动和转动远远慢于水分子的平动和转动。 RNase Sa酶表面水分子的平动和转动随着尿素浓度增加而逐渐变慢,但RNase Sa酶表面尿素分子的动力学并不依赖于尿素浓度变化。 本研究中明晰的RNase Sa酶表面水和尿素分子分布和动力学有助于理解水和尿素分子对蛋白质稳定性的影响。     

分子动力学方法研究聚甲基乙烯基醚／水体系中氢键和准氢键的结构与分布

用分子动力学方法模拟室温下不同浓度的聚甲基乙烯基醚／水体系的微观溶剂化结构．得到的径向分布函数和氢键给体和受体距离分布表明,聚合物与水形成的氢键比水之间形成的氢键短约0．005nm．准氢键C—H…O的数目是范德华作用对的7．2％．我们发现,在各浓度下,水分子并不能均匀地分布在聚合物结构单元上,即使在很稀的溶液（3．3％,质量分数）中,仍然有10％左右的醚氧没有和水分子形成氢键．这说明在溶液中,不但高分子链间有紧密的接触,而且高分子链内的链段间也有紧密的接触,导致链上的一些醚氧不能和水分子有效地接触而形成氢键．准氢键随浓度的变化和氢键的变化趋势类似,但形成准氢键的结构单元数目与形成氢键的结构单元数目比值在0．2附近．文献上用动态DSC测量低分子量聚甲基乙烯基醚（PVME）水溶液的相转变焓发现,在浓度为30％左右有一转折,与本模拟所得出的在浓度为27％左右氢键和准氢键比例的转折相关,这给相转变焓的转折点提供了分子尺度的微观解释．另外,浓度小于54％的溶液中存在“自由水”,在86％的浓溶液中每个结构单元大约与1．56个水分子缔合．     

Interaction of urea with amino acids: implications for urea-induced protein denaturation

The molecular mechanism of urea-induced protein denaturation is not yet fully understood. Mainly two opposing mechanisms are controversially discussed, according to which either hydrophobic, or polar interactions are the dominant driving force. To resolve this question, we have investigated the interactions between urea and all 20 amino acids by comprehensive molecular dynamics simulations of 22 tripeptides. Calculation of atomic contact frequencies between the amino acids and solvent molecules revealed a clear profile of solvation preferences by either water or urea. Almost all amino acids showed preference for contacts with urea molecules, whereas charged and polar amino acids were found to have slight preferences for contact with water molecules. Particularly strong preference for contacts to urea were seen for aromatic and apolar side-chains, as well as for the protein backbone of all amino acids. Further, protein-urea hydrogen bonds were found to be significantly weaker than protein-water or water-water hydrogen bonds. Our results suggest that hydrophobic interactions are the dominant driving force, while hydrogen bonds between urea and the protein backbone contribute markedly to the overall energetics by avoiding unfavorable unsatisfied hydrogen bond sites on the backbone. In summary, we suggest a combined mechanism that unifies the two current and seemingly opposing views.     

Molecular dynamics study on the stabilization of dehydrated lipid bilayers with glucose and trehalose

In this study, we use molecular dynamics simulations to investigate and compare the interactions of DPPC bilayers with and without saccharides (glucose or trehalose) under dehydrated conditions. Results from the simulations indicate that unilamellar bilayers lose their structural integrity under dehydrated conditions in the absence of saccharides; however, in the presence of either glucose or trehalose, the bilayers maintain their stability. Hydrogen bond analysis shows that the saccharide molecules displace a significant amount of water surrounding the lipid headgroups. At the same time, the additional hydrogen bonds formed between water and saccharide molecules help to maintain a hydration layer on the lipid bilayer interface. On the basis of the hydrogen bond distributions, trehalose forms more hydrogen bonds with the lipids than glucose, and it is less likely to interact with neighboring saccharide molecules. These results suggest that the interaction between the saccharide and lipid molecules through hydrogen bonds is an essential component of the mechanism for the stabilization of lipid bilayers.     

The effect of aqueous solutions of trimethylamine-N-oxide on pressure induced modifications of hydrophobic interactions

To understand the mechanism of protein protection by the osmolyte trimethylamine-N-oxide (TMAO) at high pressure, using molecular dynamics (MD) simulations, solvation of hydrophobic group is probed in aqueous solutions of TMAO over a wide range of pressures relevant to protein denaturation. The hydrophobic solute considered in this study is neopentane which is a considerably large molecule. The concentrations of TMAO range from 0 to 4 M and for each TMAO concentration, simulations are performed at five different pressures ranging from 1 atm to 8000 atm. Potentials of mean force are calculated and the relative stability of solvent-separated state over the associated state of hydrophobic solute are estimated. Results suggest that high pressure reduces association of hydrophobic solutes. From computations of site-site radial distribution function followed by analysis of coordination number, it is found that water molecules are tightly packed around the nonpolar particle at high pressure and the hydration number increases with increasing pressure. On the other hand, neopentane interacts preferentially with TMAO over water and although hydration of neopentane reduces in presence of this osmolyte, TMAO does not show any tendency to prevent the pressure-induced dispersion of neopentane moieties. It is also observed that TMAO molecules prefer a side-on orientation near the neopentane surface, allowing its oxygen atom to form favorable hydrogen bonds with water while maintaining some hydrophobic contacts with neopentane. Analysis of hydrogen-bond properties and solvation characteristics of TMAO reveals that TMAO can form hydrogen bonds with water and it reduces the identical nearest neighbor water molecules caused by high hydrostatic pressures. Moreover, TMAO enhances life-time of water-water hydrogen bonds and makes these hydrogen bonds more attractive. Implication of these results for counteracting effect of TMAO against protein denaturation at high pressures are discussed.     

beta-Sheet ligands in action: KLVFF recognition by aminopyrazole hybrid receptors in water

Little is known about the precise mechanism of action of beta-sheet ligands, hampered by the notorious solubility problems involved with protein misfolding and amyloid formation. Recently the nucleation site for the pathogenic aggregation of the Alzheimer's peptide was identified as the KLVFF sequence in the central region of Abeta. A combination of two aminopyrazole ligands with di- or tripeptides taken from this key fragment now furnished water-soluble Abeta-specific ligands which allow model investigations in water. A detailed conformational analysis provides experimental evidence for an increased beta-sheet content induced in the peptide. Strong indications were also found for the peptide backbone recognition via hydrogen bonds plus hydrophobic contributions between aminopyrazole nuclei and Phe residues. The affinity of these new ligands toward the KKLVFF fragment is highly dependent on their sequence and composition from natural and artificial amino acids. Thus, for the first time, detailed insight is gained into the complexation of beta-sheet ligands with model peptides taken directly from Abeta.     

From protein denaturant to protectant: comparative molecular dynamics study of alcohol/protein interactions

It is well known that alcohols can have strong effects on protein structures. For example, monohydric methanol and ethanol normally denature, whereas polyhydric glycol and glycerol protect, protein structures. In a recent combined theoretical and NMR experimental study, we showed that molecular dynamics simulations can be effectively used to understand the molecular mechanism of methanol denaturing protein. In this study, we used molecular dynamics simulations to investigate how alcohols with varied hydrophobicity and different numbers of hydrophilic groups (hydroxyl groups) exert effects on the structure of the model polypeptide, BBA5. First, we showed that methanol and trifluoroethanol (TFE) but not glycol or glycerol disrupt hydrophobic interactions. The latter two alcohols instead protect the assembly of the α- and β-domains of the polypeptide. Second, all four alcohols were shown to generally increase the stability of secondary structures, as revealed by the increased number of backbone hydrogen bonds formed in alcohol/water solutions compared to that in pure water, although individual hydrogen bonds can be weakened by certain alcohols, such as TFE. The two monohydric alcohols, methanol and TFE, display apparently different sequence-dependence in affecting the backbone hydrogen bond stability: methanol tends to enhance the stability of backbone hydrogen bonds of which the carbonyl groups are from polar residues, whereas TFE tends to stabilize those involving non-polar residues. These results demonstrated that subtle differences in the solution environment could have distinct consequences on protein structures.     

Unexpectedly strong energy stabilization inside the hydrophobic core of small protein rubredoxin mediated by aromatic residues: correlated ab initio quantum chemical calculations

The formation of a hydrophobic core of globular proteins is believed to be the consequence of exterior hydrophobic forces of entropic nature. This, together with the low occurrence of hydrogen bonds in the protein core, leads to the opinion that the energy contribution of core formation to protein folding and stability is negligible. We show that stabilization inside the hydrophobic core of a small protein, rubredoxin, determined by means of high-level correlated ab initio calculations (complete basis set limit of MP2 stabilization energy + CCSD(T) correction term), amounted to approximately 50 kcal/mol. These results clearly demonstrate strong attraction inside a hydrophobic core. This finding may lead to substantial changes in the current view of protein folding. We also point out the inability of the DFT/B3LYP method to describe a strong attraction between studied amino acids.     

The relative helix and hydrogen bond stability in the B domain of protein A as revealed by integrated tempering sampling molecular dynamics simulation

Molecular dynamics simulations using the integrated tempering sampling method were performed for the folding of wild-type B domain of protein A (BdpA). Starting from random and stretched structures, these simulations allow us to fold this protein into the native-like structure frequently, achieving very small backbone (1.7 A?) and all heavy-atom root-mean-square deviation (2.6 A?). Therefore, the method used here increases the efficiency of configuration sampling and thermodynamics characterization by molecular dynamics simulation. Although inconsistency exists between the calculation and experiments for the absolute stabilities, as a limitation of the force field parameters, the calculated order of helix stability (H3 > H2 > H1) is consistent with that determined by experiments for individual separate helices. The lowest free energy folding pathway of BdpA was found to start with a barrierless and non-cooperative structural collapse from the entirely extended (E) state, which leads to a physiologically unfolded (P) state consisting of multiple stable structures with few native inter-helical hydrophobic interactions formed. In the P state, only H3 is fully structured. The final formation of H1 (and to a lesser extent, H2) in the folded (F) state requires the packing of the inter-helical hydrophobic contacts. In addition, it was found that stabilities of backbone hydrogen bonds are significantly affected by their positions relative to the inter-helical hydrophobic core. As temperature increases, the stability of the hydrogen bonds exposed to the solvent tends to increase while that of the hydrogen bonds buried within the hydrophobic core decreases. Finally, we discuss implications of this study on the general folding mechanism of proteins.     

Halogen bonding in complexes of proteins and non-natural amino acids

In this work, we have analyzed the influence of halogen bonding to the stability of 44 complexes of proteins and non-natural amino acids. Fluorine- and chlorine-containing non-natural amino acids are more prevalent in the dataset, and an even larger number of contacts made by iodine-containing ligands are found. Only few halogen bonds with the hydroxyl oxygens and carboxylate side chains are found in the dataset. Halogen bonds with the nitrogen-containing side chains have higher occurrence than other acceptors. Backbone carbonyl oxygens and nitrogens are to a substantial extent involved in our dataset. We have observed a small percentage of interactions involving water as hydrogen bond donors. Additionally, most of the interacting residues comprising the interfaces also show a great degree of conservation. There is a clear interaction hot spot at distances of 3.5–3.7 Å and Θ₁ angles of 100–120°. There is also a cluster of contacts featuring short distances (2.6–2.9 Å) but only nearly optimal Θ₁ angles (140–160°). 51.3% of stabilizing residues are involved in building halogen bonds with the non-natural amino acids. We discovered three types of structural motifs significantly over-represented: beta-turn-ir, beta-turn-il and niche-4r. The halogen-bonding statistics of the dataset do not show any preference for α-helices (36%), β-sheets (36%), or turns/coils (28%) structures. Most of the amino acid residues that were involved in halogen bonds prefer to be in the solvent excluded environment (buried). Furthermore, we have shown that in amino acid–protein complexes halogen atoms can sometimes be involved in hydrogen bonding interactions with hydrogen bonding-donors. The results from this study might be used for the rational design of halogenated ligands as inhibitors and drugs, and in biomolecular engineering.     

Aqueous urea solutions: structure, energetics, and urea aggregation

Urea is ubiquitously used as a protein denaturant. To study the structure and energetics of aqueous urea solutions, we have carried out molecular dynamics simulations for a wide range of urea concentrations and temperatures. The hydrogen bonds between urea and water were found to be significantly weaker than those between water molecules, which drives urea self-aggregation due to the hydrophobic effect. From the reduction of the water exposed urea surface area, urea was found to exhibit an aggregation degree of ca. 20% at concentrations commonly used for protein denaturation. Structurally, three distinct urea pair conformations were identified and their populations were analyzed by translational and orientational pair distribution functions. Furthermore, urea was found to strengthen water structure in terms of hydrogen bond energies and population of solvation shells. Our findings are consistent with a direct interaction between urea and the protein as the main driving force for protein denaturation. As an additional, more indirect effect, urea was found to enhance water structure, which would suggest a weakening of the hydrophobic effect.