Combined molecular simulation and experimental study to discover, predict and control enzyme immobilization in polymeric nanoparticles

酶是以蛋白质为基础的生物催化剂，具有巨大的潜力，可以彻底改变我们将化学品转化为有用产品、治疗疾病或为环境污染物解毒的方式。然而，酶在实践中的使用受到其通过各种机制失去活性的趋势的限制。提高酶稳定性的一种常见策略是将其固定并包裹在聚合物中，聚合物是由许多重复亚单位组成的大分子。这一策略前景看好，但聚合物的性质、包埋技术和酶/聚合物的相互作用都在决定特定包埋策略的成功与否中发挥着重要作用。本研究项目旨在将计算机模拟和实验结果进行比较，以帮助开发一种有效的方法来预测有益的封装策略。通过与华盛顿大学工程数学学院的学生合作，研究人员提供了许多机会来丰富研究生的培训，改善教育，并吸引当地社区学院(贝尔维尤学院)的本科生参与研究。到目前为止，成功的酶胶囊主要是通过广泛的试验和错误实验和偶然发现的。该项目结合了分子尺度模拟和各种基质中酶活性和释放的实验测量，旨在建立一个合理的设计框架，以发现、预测和控制酶/聚合物界面上的基本驱动力。具体地说，该项目正在展示一种全面的策略，使用快速分子动力学模拟与统计机器学习相结合，来识别包裹在聚合物纳米颗粒中的酶的强结合和弱结合的序列水平描述符。正在进行补充性实验，研究各种条件下的酶负载、释放和活性。合理设计这种相互作用的能力将是一个变革性的进步，可以应用于水凝胶、无机表面或其他类型的酶/聚合物系统。实验和模拟之间的强大相互作用将确保结果是准确的，并可能有助于确定分子动力学力场和高通量实验设计的改进领域。

Enzymes, protein-based biological catalysts, have enormous potential to revolutionize the way we transform chemicals to useful products, treat disease, or detoxify environmental contaminants. However, the use of enzymes in practice is limited by their tendency to lose their activity through a variety of mechanisms. One common strategy to improve enzyme robustness is to immobilize and encapsulate it in a polymer, a large molecule composed of many repeated subunits. This strategy holds great promise, but the polymer properties, encapsulation technique, and enzyme/polymer interactions all play important roles in determining the success of a particular encapsulation strategy. This research project aims to compare and contrast computer simulations with experimental results to help develop an efficient means of predicting beneficial encapsulation strategies. Working with students through the UW College of Engineering Math Academy, the researchers are providing many opportunities for enriching the training of graduate students, improving education, and engaging undergraduates from a local community college (Bellevue College) in research. To date, successful enzyme encapsulations have been discovered largely via extensive trial and error experimentation and serendipity. This project, a combined study of molecular scale simulations and experimental measurement of enzyme activity and release in various matrices, seeks to build a rational design framework to discover, predict, and control the essential governing driving forces at the enzyme/polymer interface. Specifically, this project is demonstrating a comprehensive strategy, using fast molecular dynamics simulations paired with statistical machine learning, to identify sequence level descriptors of strong and weak binding of enzymes encapsulated in polymer nanoparticles. Complementary experiments are being performed that study enzyme loading, release and activity under a wide range of conditions. The ability to rationally design such interactions would be a transformative advance that could be applied to hydrogels, inorganic surfaces, or other types of enzyme/polymer systems. The strong interplay between the experiments and simulations will ensure that the results are accurate and potentially help to identify areas of improvement for molecular dynamics force fields and high throughput experimental design.