Multiscale simulations of environment-induced conformational transitions in peptides: folding, partitioning and aggregation

肽在各个领域发挥着越来越重要的作用，从生物医学研究，食品科学到生物材料应用。例如，两亲性肽可用于药物递送目的以帮助穿透细胞膜。这里，另外的益处是肽可以是刺激响应性的，即可以通过pH或离子强度的变化诱导折叠/解折叠。由于它们倾向于形成定义明确的结构聚集体，肽经常用作生物和仿生材料的模板。为了以良好控制的方式系统地利用肽，理解序列、二级结构形成以及折叠与环境条件和周围介质变化（pH、离子强度、亲水性/疏水性、界面的存在、其他肽的存在等）的偶联之间的关系是非常重要的。分子模拟可以提供对相关相互作用、溶剂化热力学等的微观理解。然而，单独的原子模型是有限的，特别是当涉及聚集现象时，需要粗粒度（CG）模型来访问相关系统的大小和时间尺度。为了提供肽聚集体和肽基材料中结构形成的真实表示，聚集、折叠和分割的相互作用需要在CG模型中反映。这是一个可以被视为一种形式的可转移性挑战的问题。可转移性是指-由于自由度的数量减少- CG模型强烈依赖于状态点（热力学条件，浓度，化学成分等）的问题。它们被参数化了。这是CG模型最严重的局限性之一，这就是为什么理解和解决与可转移性相关的问题是CG模型开发中最重要的问题之一的原因。在本项目中，我们计划开发基于肽的生物材料的可转移CG模型，这些模型能够在环境变化时再现正确的构象行为。我们计划开发一种策略，如何参数化CG模型，捕获肽折叠，环境变化，如pH值或离子强度，聚集，与界面的相互作用，以及水和疏水介质之间的分区之间的耦合和相互影响。为此，选择了三种靶肽，其允许分别关注上述方面中的不同方面：（i）二苯基丙氨酸在聚集或与亲水/疏水界面相互作用时表现出反式/顺式构象变化，而（ii）GALA和（iii）KL肽显示出受周围介质影响的螺旋/卷曲转变。

Peptides play an increasingly important role in various fields, from biomedical research, food science, to biomaterials applications. For example, amphiphilic peptides can be used for drug delivery purposes to help penetrate cell membranes. Here, an additional benefit is that peptides may be stimulus responsive, i.e. one can induce folding/unfolding by a change in pH or ion strength. Thanks to their proneness to form well-defined structured aggregates, peptides frequently serve as templates for biological and biomimetic materials. In order to systematically make use of peptides in a well-controlled manner, it is of immense importance to understand the relationship between sequence, secondary-structure formation, and the coupling of the folding to environmental conditions and to changes in the surrounding medium (pH, ion strength, hydrophilicity/ hydrophobicity, presence of an interface, presence of other peptides, etc.). Here, molecular simulation can provide microscopic understanding of the relevant interactions, solvation thermodynamics, etc. However, atomistic models alone are limited, and - in particular when aggregation phenomena are involved - coarse grained (CG) models are required to give access to the relevant system sizes and timescales.In order to provide a realistic representation of structure formation in peptide aggregates and peptide-based materials, the interplay of aggregation, folding and partitioning needs to be reflected in the CG model. This is a problem that can be viewed as a form of transferability challenge. Transferability refers to the issue that - due to the reduction of the number of degrees of freedom - CG models strongly depend on the state point (thermodynamic conditions, concentrations, chemical compositions etc.) where they were parameterized. This is one of the most severe limitations of CG models, which is the reason why the understanding and solving of transferability-related problems is one of the most important issues in the development of CG models.In this project we plan to develop transferable CG models for peptide-based biomaterials which are capable of reproducing the correct conformational behavior upon environment change. We plan to develop a strategy how to parameterize CG models that capture the coupling between and mutual influence of peptide folding, environment change such as pH or ion strength, aggregation, interaction with interfaces, and partitioning between aqueous and hydrophobic media. To this end, three target peptides were selected which allow to separately focus on different of the above aspects: (i) diphenylalanine exhibits a trans/cis conformational change upon aggregation or interaction with a hydrophilic/hydrophobic interface, while the (ii) GALA and (iii) KL peptides show a helix/coil transitions that are influenced by the surrounding media.