Preferred binding regions for highly selective and orthogonal multimodal protein separations

基于蛋白质的药物的发展正在对许多疾病的治疗产生越来越大的影响，如癌症、糖尿病和多发性硬化症。这些疾病的蛋白质疗法的例子分别包括Rituxan、Insulin和Avonex，还有更多处于不同发现和开发阶段的药物。然而，生产这些药物的成本以及获得副作用最小的高纯度药物的挑战是一个重大挑战。通常，负面副作用是由于少量与所需药物极其相似的杂质造成的，因此很难去除。这个由美国国家科学基金会资助的项目将开发新的方法，有效地从一系列生物制药中去除这些具有挑战性的药物杂质，从而降低生产成本，更安全的药物和更多的患者访问。这个实验和计算项目将专注于开发用于色谱树脂的多模式配体，将用于实现复杂生物产品的高度选择性分离。新的多模式配体将通过靶向蛋白质表面的特定区域来提高选择性，从而能够有效地去除与生物产品相关的杂质，这是生物制造中的一个关键挑战。实验表征技术将验证分子水平的结合信息。这项计算工作将预测关键结合面和相对结合亲和力，并将为生物过程开发产生强大的预测工具。在特定蛋白质表面区域具有独特和有针对性的选择性的多模式系统的开发将使在分子水平的理解和预测工具的指导下开发高效的正交下游过程成为可能。这将导致降低成本，提高产品质量和安全。从事这一项目的研究生和本科生将直接接触到生物物理学、层析和分子模拟方面的最新技术。更广泛地说，这些概念将被纳入伦斯勒理工学院纳米技术中心的旗舰推广和教育项目The Moleculum项目的本科课程和在线动画电影中。

The development of protein-based medicines is having an increasing impact on treatment of many diseases such as cancer, diabetes, and multiple sclerosis. Example protein-based therapeutics for these diseases include rituxan, insulin, and avonex, respectively, with many more in various stages of discovery and development. However, the costs of producing these drugs and the challenge of obtaining highly pure drugs with minimal side effects is a major challenge. Often, negative side effects are due to small amounts of impurities that are extremely similar to the desired medicine and thus very difficult to remove. This NSF funded project will develop new methods for efficiently removing these challenging drug impurities from a range of biopharmaceuticals, resulting in reduced production costs, safer medicines and more patient access.This experimental and computational project will focus on the development of multimodal ligands for chromatographic resins, which will be used to achieve highly selective separations of complex biological products. The new multimodal ligands will increase selectivity by targeting specific regions on protein surfaces, enabling the effective removal of bioproduct related impurities, a critical challenge in biomanufacturing. Experimental characterization techniques will validate molecular-level binding information. The computational work will predict key binding faces and relative binding affinities and will produce powerful predictive tools for bioprocess development. The development of multimodal systems with unique and targeted selectivities on specific protein surface regions will enable the development of highly effective orthogonal downstream processes guided by molecular level understanding and predictive tools. This will result in reduced costs and improved product quality and safety. Graduate and undergraduate students working on this project will be directly exposed to the state of the art in biophysics, chromatography and molecular simulations. More broadly, the concepts will be incorporated into both undergraduate coursework and on-line animated movies for the the Molecularium project, the flagship outreach and education effort of Rensselaer Polytechnic Institute's Nanotechnology Center.