Construction of a T m-value prediction model and molecular dynamics study of AmNA-containing gapmer antisense oligonucleotide

Masataka Kuroda; Yuuya Kasahara; Masako Hirose; Harumi Yamaguma; Masayuki Oda; Chioko Nagao; Kenji Mizuguchi

doi:10.1016/j.omtn.2024.102272

Construction of a T _m-value prediction model and molecular dynamics study of AmNA-containing gapmer antisense oligonucleotide

Mol Ther Nucleic Acids. 2024 Jul 16;35(3):102272. doi: 10.1016/j.omtn.2024.102272. eCollection 2024 Sep 10.

Authors

Masataka Kuroda^{1

2}, Yuuya Kasahara^{1

3}, Masako Hirose⁴, Harumi Yamaguma¹, Masayuki Oda⁵, Chioko Nagao⁶, Kenji Mizuguchi^{1

6}

Affiliations

¹ National Institutes of Biomedical Innovation, Health and Nutrition (NIBIOHN), Osaka 566-0002, Japan.
² Mitsubishi Tanabe Pharma Corporation, Yokohama 227-0033, Japan.
³ Graduate School of Pharmaceutical Science, Osaka University, Osaka 565-0871, Japan.
⁴ Malvern Panalytical, Spectris, Tokyo 105-0013, Japan.
⁵ Graduate School of Life and Environmental Sciences, Kyoto Prefectural University, Kyoto 606-8522, Japan.
⁶ Institute for Protein Research, Osaka University, Osaka 565-0871, Japan.

Abstract

RNase H-dependent antisense oligonucleotides (gapmer ASOs) represent a class of nucleic acid therapeutics that bind to target RNA to facilitate RNase H-mediated RNA cleavage, thereby regulating the expression of disease-associated proteins. Integrating artificial nucleic acids into gapmer ASOs enhances their therapeutic efficacy. Among these, amido-bridged nucleic acid (AmNA) stands out for its potential to confer high affinity and stability to ASOs. However, a significant challenge in the design of gapmer ASOs incorporating artificial nucleic acids, such as AmNA, is the accurate prediction of their melting temperature (T _m ) values. The T _m is a critical parameter for designing effective gapmer ASOs to ensure proper functioning. However, predicting accurate T _m values for oligonucleotides containing artificial nucleic acids remains problematic. We developed a T _m prediction model using a library of AmNA-containing ASOs to address this issue. We measured the T _m values of 157 oligonucleotides through differential scanning calorimetry, enabling the construction of an accurate prediction model. Additionally, molecular dynamics simulations were used to elucidate the molecular mechanisms by which AmNA modifications elevate T _m , thereby informing the design strategies of gapmer ASOs.

Keywords: MT: Bioinformatics; RNase H-dependent antisense oligonucleotide, gapmer ASO; Tm value prediction; amido-bridged nucleic acid, AmNA; machine learning; molecular dynamics, MD.