ASSOCIATION OF PEPTIDES IN MEMBRANES

膜中肽的缔合

• 批准号: 7367745
• 负责人: ANDREW POHORILLE
• 金额: $ 0.77万
• 依托单位: UNIVERSITY OF CALIFORNIA, SAN FRANCISCO
• 依托单位国家: 美国
• 财政年份: 2006
• 资助国家: 美国
• 起止时间: 2006-07-01 至 2007-06-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/7367745
• 关键词: ASSOCIATION; PEPTIDES; MEMBRANES

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. Single transmembrane helices are rarely capable of performing biological functions. Instead, they form functional units after self-assembling into higher order structures. However, not all helices self-assemble. Consequently, it is necessary to understand sequence-specific interhelical recognition before we can predict the kinds of structures that can form in membranes. The simplest models for peptide association are helical dimers. Simple experimental models have recently been developed for helix associations in membranes. The best-studied system, both structurally and thermodynamically, is the 24-residue transmembrane region of glycophorin A (GpA). GpA forms non-covalent dimers through the reversible association of its membrane-spanning domain, which adopts an alpha-helical conformation. The structure has recently been determined by NMR spectroscopy of 40-residue peptides that contain the transmembrane segment, solvated in detergent micelles. This study also showed that the alpha-helices formed a right-handed, coil-coiled structure. We simulated the wild-type dimer of GpA in a lamella of dodecane, placed between two lamellae of water. The width of the dodecane layer was approximately the same as that of the hydrophobic core of a palmitoyloleylphosphatidylcholine (POPC) membrane. The starting structure was based on the NMR model of the dimer. The comparison between the calculated, time-averaged structure of the dimer after 30.0 ns of molecular dynamics trajectory and the nuclear Overhauser effect data of MacKenzie, et al., provided an assessment of the accuracy of our model and the potential energy functions utilized. The distance root mean square deviation was less than 1.5 Angstroms for the backbone atoms and 2.75 Angstroms for all atoms (Hydrogens excluded) and the monomers remained alpha-helical. Engelman, et. al, have observed interhelical close contacts between several Hydrogens bound to alpha Carbons and carbonyl Oxygens and argue that these can form weak hydrogen bonds that stablize the dimer. We observe that between 4 and 5 of these contacts are maintained over the course of the simulations and could contribute significantly to the stability of the dimer. The dissociation free energy of the GpA dimer in a detergent pentaoxyethylene (C8E5) has been estimated to be about 9.0 kcal/mol. It has been also demonstrated that single-residue mutations can markedly influence the free energy of association of the helices. Mutants in which either one of two leucine residues were substituted with alanine or glycine was substituted with isoleucine were found to be less stable than the wild-type dimer by 1-3 kcal/mol. All these residues are involved in interhelical interactions in the NMR determined model of GpA. We separated the helices in a series of molecular dynamics simulations using the distance separating the centers of mass of the two helices as the reaction coordinate. The free energy of dissociation was calculated using the Adaptive Biasing Force (ABF) method of Darve and Pohorille. The complete free energy pathway joining the dimer from 5.5 Angstroms at contact to dissociated monomers at 20 Angstroms was divided into a series of intermediate states corresponding to different values of the reaction coordinate. The free energy of dissociation is given as the sum of free energy differences between consecutive intermediate states. The computed free energy is approximately 10.0 kcal/mol. By comparison, the experimentally determined free energy of dissociation in the detergent pentaoxyethylene (C8E5) is equal to 9.0 kcal/mol. Since the molecular environments of GpA in the computational and experimental studies are different - ie. dodecane versus C8E5 - the free energies are not expected to be identical, but should be similar. Given that the volume accessible to the alpha-helical dimer is much smaller in C8E5 than in dodecane, which corresponds to a smaller entropic contribution, we expect that the free energy of dissociation should be higher in dodecane than in C8E5. Based on similar considerations, it is expected that the influence of the surroundings on the computed point mutations would, in principle, be limited due to compensation of entropic effects in the free energy differences. However, this may not be necessarily the case. The I76L point mutation was carried out computationally by decoupling the annihilation of the electrostatic and the van der Waals and internal parameter contributions. The free energy difference for the electrostatic term was 0 kcal/mol to within the statistical errors. The contribution due to the modification of the van der Waals parameters and the participating chemical bonds and valence angles, and, therefore, the total free energy difference, is equal to 0.4 kcal/mol, which is somewhat less than the experimental value of 1.7 kcal/mol. It should be emphasized, however, that, as the L-isoleucine is transformed into L-alanine, the disruptive effect of the mutation found experimentally in C8E5 should be less pronounced in dodecane, because of the greater volume accessible to the alpha-helices in that environment. The free energy change upon single-point mutation is fairly small. In contrast, the free energy of dissociation is large and positive. This might suggest that the dimer is strongly favored for both the wild-type and the mutants. This would indeed be the case if we only considered the equilibrium between the associated state and separated, transmembrane helices. However, this comparison is not appropriate because the transmembrane orientation of helices is not necessarily stable. The free energy of insertion of a helical peptide into the membrane is positive (unfavorable) and may be substantial. Thus, the equilibrium that needs to be considered is between the transmembrane dimer and the individual helices at the water-membrane interface. This equilibrium is governed by the balance between the unfavorable free energy of insertion into the membrane and favorable free energy of interhelical association. This balance could be subtle, and modest changes in either term could shift the equilibrium, possibly disrupting dimerization.

本子项目是利用由NIH/NCRR资助的中心赠款提供的资源的众多研究子项目之一。子项目和研究者（PI）可能已经从另一个NIH来源获得了主要资金，因此可以在其他CRISP条目中表示。列出的机构是中心的，不一定是研究者的机构。单个跨膜螺旋很少能够执行生物学功能。相反，它们在自组装成更高阶结构后形成功能单元。然而，并不是所有的螺旋都能自我组装。因此，在我们能够预测可以在膜中形成的各种结构之前，有必要了解序列特异性螺旋间识别。最简单的肽结合模型是螺旋二聚体。简单的实验模型，最近已开发的螺旋结合膜。在结构和热力学上研究得最好的系统是糖蛋白A （GpA）的24个残基跨膜区。GpA通过其跨膜结构域的可逆结合形成非共价二聚体，该结构域采用α -螺旋构象。该结构最近已通过核磁共振光谱确定了40个残基肽，其中包含跨膜段，溶剂化在洗涤剂胶束中。这项研究还表明，α -螺旋形成了一个右旋的、盘绕的结构。我们模拟了野生型GpA二聚体在十二烷片层中，置于两个水片层之间。十二烷层的宽度与棕榈酰油基磷脂酰胆碱（POPC）膜的疏水核的宽度大致相同。初始结构是基于二聚体的核磁共振模型。将计算得到的经过30.0 ns分子动力学轨迹后二聚体的时间平均结构与MacKenzie等人的核Overhauser效应数据进行比较，可以评估我们的模型和所使用的势能函数的准确性。主链原子的距离均方根偏差小于1.5埃，所有原子（氢除外）的距离均方根偏差小于2.75埃，单体保持α -螺旋结构。Engelman等人观察到与α碳和羰基氧结合的几个氢之间的螺旋间紧密接触，并认为这些可以形成弱氢键，稳定二聚体。我们观察到，在模拟过程中，这些接触中有4到5个保持着，这对二聚体的稳定性有很大的贡献。估计GpA二聚体在清洁剂五氧乙烯（C8E5）中的解离自由能约为9.0 kcal/mol。研究还表明，单残基突变可以显著影响螺旋的自由结合能。当两个亮氨酸残基中的一个被丙氨酸取代或甘氨酸被异亮氨酸取代时，突变体的稳定性比野生型二聚体低1-3千卡/摩尔。在核磁共振确定的GpA模型中，所有这些残基都参与了螺旋间相互作用。在一系列分子动力学模拟中，我们用两个螺旋质心之间的距离作为反应坐标来分离螺旋。利用Darve和Pohorille的自适应偏压力（ABF）法计算解离自由能。二聚体从接触时的5.5埃到20埃解离单体的完整自由能途径被划分为一系列对应于不同反应坐标值的中间态。解离的自由能用连续中间态之间的自由能差的和来表示。计算得到的自由能约为10.0 kcal/mol。通过比较，实验测定的洗涤剂五氧乙烯（C8E5）的解离自由能为9.0 kcal/mol。由于GpA的分子环境在计算和实验研究中是不同的。十二烷和C8E5的自由能不会相同，但应该相似。考虑到螺旋二聚体在C8E5中的可接近体积比在十二烷中的小得多，这对应于更小的熵贡献，我们预计十二烷的解离自由能应该比C8E5高。基于类似的考虑，由于补偿了自由能差中的熵效应，预计环境对计算的点突变的影响原则上是有限的。然而，情况未必如此。通过解耦静电湮灭与范德华和内部参数贡献，计算了I76L点突变。静电项的自由能差为0千卡/摩尔，在统计误差范围内。由于范德华参数和参与的化学键和价角的修改，因此，总自由能差的贡献等于0.4 kcal/mol，略低于1.7 kcal/mol的实验值。然而，应该强调的是，当l -异亮氨酸转化为l -丙氨酸时，在C8E5实验中发现的突变的破坏性影响在十二烷中应该不那么明显，因为在该环境中α -螺旋可接近的体积更大。单点突变时的自由能变化很小。相反，离解的自由能很大，而且是正的。这可能表明二聚体对野生型和突变型都非常有利。如果我们只考虑结合态和分离的跨膜螺旋之间的平衡，情况确实是这样的。然而，这种比较是不合适的，因为螺旋的跨膜取向不一定稳定。螺旋肽插入膜的自由能是正的（不利的），可能是巨大的。因此，需要考虑的平衡是跨膜二聚体和水膜界面上的单个螺旋之间的平衡。这种平衡是由插入膜的不利自由能和螺旋间结合的有利自由能之间的平衡所控制的。这种平衡可能是微妙的，任何一项的适度变化都可能改变平衡，可能破坏二聚化。