Connecting Drug Discovery with Solid State Formulation Design

计算机辅助药物设计包括根据生物活性分子的分子结构对其功能和特性进行预测。这些模型没有考虑到这些分子的固态特性，这些特性可能会控制溶解度和生物利用度，这两个物理化学特性可以控制药物的体内活性。活性药物成分（API）的固体形式以及盐和水合物形成等因素会影响其表面性质和溶解速率，这在考虑固体原料药制剂的生物利用度时至关重要。将早期药物发现中使用的计算方法与可以预测固态特性的计算模型相结合，将对制药和农用化工行业产生强大的影响，并有可能大大降低开发过程中的损耗率。该项目的目的是通过生成定量结构-性质关系（QSPR）计算模型将分子的结构与其溶解度特性联系起来，该模型可以预测离子生物活性分子的固态特性。我们假设固态的作用在目前的溶解度和生物利用度预测模型中没有得到充分的考虑。首先，我们将重点介绍与制药工业高度相关的两种化学结构：喹啉-4- 1和1,8-萘啶。之所以选择这些药物，是因为：(i)它们是在许多临床相关药物中发现的特殊支架；（ii）它们构成了利兹Fishwick小组正在进行的研究项目的一部分，其中典型的生物活性分子在各种水系统（<10 uM）中表现出较差的溶解度。为了实现整个项目目标，我们有几个目标：目标1：我们的目标是开发一套新的描述符，该描述符基于固态独有的各种结构特征，以帮助我们捕捉固态在生物利用度中的作用。这包括“合成相互作用”，3D排列，堆叠和升华焓，以及其他特征。我们认为表面化学起着重要的作用，不同的3D结构产生的表面类型。我们希望使用易于形成不同多晶和盐结构的离子材料来帮助我们量化这些相互作用在固态中的作用。目标2：使用实验数据（如不同多晶型的水溶性测量和原子力显微镜）训练计算模型，以便这些新的描述符可以用于进一步的预测研究，以帮助改进与生物利用度相关的特性的建模。一旦达到合适的终点，这些知识就可以反馈到计算机辅助药物设计的设计阶段。目标3：扩大范围，包括来自工业伙伴的额外特权脚手架。该项目涉及EPSRC的多个研究主题：表面科学（主要对准）、复杂性科学和控制工程。我们的目标是探索离子物质表面化学中的许多未知数，将实验数据与QSPR模型联系起来。

Computer-aided drug design involves making predictions relating to the function and properties of bioactive molecules based on their molecular structure. These models do not take into account the solid-state properties of such molecules, which may control the solubility and bioavailability, two physicochemical properties which can govern the in vivo activity of drugs. The solid form of active pharmaceutical ingredients (APIs), and factors such as salt and hydrate formation, can affect the surface properties and dissolution rate which are critical when considering the bioavailability of solid API formulations. Combining computational approaches used in early stage drug discovery with computational models which can predict solid-state properties, would be a powerful assert to the pharmaceutical and agrochemical industries, with the potential to drastically reduce rates of attrition during development.The aim of the project is to link the structure of a molecule to its solubility properties by generating quantitative structure-property relationship (QSPR) computational models which can predict solid-state properties of ionic bioactive molecules. We postulate that the role of the solid state is not adequately accounted for in the current status-quo models for solubility and bioavailability prediction. To begin we will focus on two chemical structures which are highly relevant to the pharmaceutical industry: quinolin-4-ones and 1,8-naphthyridines. These have been chosen since (i) they are privileged scaffolds, found in a number of clinically relevant drugs; (ii) they form part of active research projects in the Fishwick group at Leeds, where exemplar bioactive molecules display poor solubility in various aqueous systems (<10 uM).To achieve the overall project aim, we have a number of objectives:Objective 1: We aim to develop a new set of descriptors based on various structural features exclusive to the solid state, to help us capture the role of the solid state in bioavailability. This include 'synthonic interactions', the 3D arrangement, stacking and sublimation enthalpies, amongst other features. We believe surface chemistry plays an important role, and the types of surfaces generated from different 3D structures. We wish to use ionic materials which have a propensity to readily form different polymorphs and salt structures to help us quantify the role of these interactions exclusive to the solid state. Objective 2: Train the computational models using experimental data (such as aqueous solubility measurements of different polymorphs and atomic force microscopy), so that these new descriptors may be used in further predictive studies to help improve modelling of properties relevant to bioavailability. Once suitable endpoints have been achieved, the knowledge can feedback into the design-phase of computer-aided drug design.Objective 3: Widen the scope to include additional privileged scaffolds from industrial partners.The project relates to a number of EPSRC research themes: surface science (major alignment), complexity science and control engineering. Our aims will explore many unknowns in surface-based chemistry of ionic substances, connecting experimental data with QSPR modelling.