Multidimensional analytical and computational approach to determine diastereomer compositions in oligonucleotide drug products

点击翻译按钮获取中文摘要

PROJECT ABSTRACT Synthetic (short) oligonucleotides are an exciting class of therapeutics to treat and prevent a wide variety of diseases, including diseases where traditional drug development efforts have failed. In contrast to native nucleic acids, which are prepared enzymatically, synthetic oligonucleotide drugs are prepared chemically via solid-phase chemical synthesis. Solid-phase chemical synthesis provides an advantageous platform for introducing chemical modifications during drug development to overcome the poor pharmacological properties of native nucleic acids. A common chemical modification for oligonucleotide drugs involves the replacement of one oxygen in the natural phosphodiester (PO) linkage with a sulfur creating a phosphorothioate (PS) linkage. The PS linkage enhances pharmacology properties by increasing nuclease resistance and facilitating protein binding. The PS modification is routinely used in antisense oligonucleotide (ASO) therapeutics that are a prevalent platform for drug discovery with several ASOs gaining United States Food and Drug Administration (FDA) approval. The consequence of chemically replacing the PO linkage with a PS linkage is the creation of a chiral center at the phosphorous atom. This leads to the PS center having two stereochemical configurations. The number of potential PS diastereomers is 2n where n is the number of PS linkages in the oligonucleotide. For a 20-mer oligonucleotide which has 19 PS linkages this equates to 524,288 potential stereoisomers. Despite the increasing popularity of ASO drugs, there is a pressing need to develop high-resolution analytical methods that can comprehensively characterize the diastereomer composition in oligonucleotides for quality control purposes in manufacturing and generic drug development. Consequently, the purpose of this proposal is to develop new and innovative analytical methods for analyzing the diastereomer composition in ASO drugs. We will use TEGSEDI as our model ASO drug product. Our proposal will determine whether and how the manufacturing process may affect the diastereomeric distributions in the drug product. We intend to successfully accomplish these objectives by developing a multidimensional analytical approach that computationally integrates state-of-the-art liquid chromatography, mass spectrometry (MS) and nuclear magnetic resonance (NMR) spectroscopy. Our experimental approach will include in-house chemical synthesis of inotersen (generic name for TEGSEDI) and short chain fragments to investigate the effect chemical synthesis (i.e., manufacturing) has on the stereochemical outcome in PS linkage formation. We will use data science including multivariate analysis and clustering to assess the diastereomer composition and apply machine learning techniques to predict the diastereomer compositions of TEGSEDI and our in-house prepared inotersen. Our developed methods will be directly transferable to other synthetic oligonucleotides and will greatly facilitate the development and assessment of generic oligonucleotide drug products.