Machine learning-powered, high-affinity modification strategies for aptamers

Abstract

The binding affinity of aptamers to targets has a crucial role in the pharmaceutical and biosensing effects. Despite diverse post-systematic evolution of ligands by exponential enrichment (post-SELEX) modifications explored in aptamer optimization, accurate prediction of high-affinity modification strategies remains challenging. Sclerostin, which antagonizes the Wnt signaling pathway, negatively regulates bone formation. Our screened sclerostin aptamer was previously shown to exert bone anabolic potential. In the current study, an interactive methodology involving the exchange of mutual information between experimental endeavors and machine learning was initially proposed to design a high-affinity post-SELEX modification strategy for aptamers. After four rounds of interactive training (a total of 422 modified aptamer-target affinity datasets with diverse modification types and sites), an antifcial intelligence model with high predictive accuracy with a correlation coefficient of 0.82 between the predicted and actual binding affinities was obtained. Notably, the machine learning-powered modified aptamer selected from this work exhibited 10⁵-fold higher affinity (picomole level K_D value) and a 3.2-folds greater Wnt-signal re-activation effect compared to naturally unmodified aptamers. This approach harnessed the power of machine learning to predict the most promising high-affinity modification strategy for aptamers.

Keywords

Machine learning / Aptamer / High affinity / Post-SELEX / Sclerostin

References

[1] Zhao QG, Zhou YJ, Cao DX, Tang AN, Kong DM: DNA-Functionalized Porphyrinic Metal-Organic Framework-Based Drug Delivery System for Targeted Bimodal Cancer Therapy. Journal of Medicinal Chemistry 2023, 66: 15370-15379.

[2] Bashir A, Yang Q, Wang J, Hoyer S, Chou W, McLean C, et al.: Machine Learning Guided Aptamer Refinement and Discovery. Nature Communications 2021, 12: 2366.

[3] Chen Z, Luo H, Gubu A, Yu S, Zhang H, Dai H, et al.: Chemically Modified Aptamers for Improving Binding Affinity to the Target Proteins via Enhanced Non-covalent Bonding. Frontiers in Cell and Developmental Biology 2023, 11: 1091809.

[4] Alkhamis O, Canoura J, Wu Y, Emmons NA, Wang Y, Honeywell KM, et al.: High-Affinity Aptamers for In Vitro and In Vivo Cocaine Sensing. Journal of the American Chemical Society 2024, 146: 3230-3240.

[5] Ma Y, Zhang H, Shen X, Yang X, Deng Y, Tian Y, et al.: Aptamer Functionalized Hypoxia-potentiating Agent and Hypoxia-inducible Factor Inhibitor Combined with Hypoxia-activated Prodrug for Enhanced Tumor Therapy. Cancer Letters 2024, 598: 217102.

[6] Chan KY, Kinghorn AB, Hollenstein M, Tanner JA: Chemical Modifications for a Next Generation of Nucleic Acid Aptamers. ChemBioChem: A European Journal of Chemical Biology 2022, 23: e202200006.

[7] Elskens JP, Elskens JM, Madder A: Chemical Modification of Aptamers for Increased Binding Affinity in Diagnostic Applications: Current Status and Future Prospects. International Journal of Molecular Sciences 2020, 21: 4522.

[8] Hasegawa H, Savory N, Abe K, Ikebukuro K: Methods for Improving Aptamer Binding Affinity. Molecules 2016, 21: 421.

[9] Davies DR, Gelinas AD, Zhang C, Rohloff JC, Carter JD, O’Connell D, et al.: Unique Motifs and Hydrophobic Interactions Shape the Binding of Modified DNA Ligands to Protein Targets. Proceedings of the National Academy of Sciences of the United States of America 2012, 109: 19971-19976.

[10] Gawande BN, Rohloff JC, Carter JD, von Carlowitz I, Zhang C, Schneider DJ, et al.: Selection of DNA Aptamers with Two Modified Bases. Proceedings of the National Academy of Sciences of the United States of America 2017, 114: 2898-2903.

[11] Wu D, Katilius E, Olivas E, Dumont Milutinovic M, Walt DR: Incorporation of Slow Off-Rate Modified Aptamers Reagents in Single Molecule Array Assays for Cytokine Detection with Ultrahigh Sensitivity. Analytical Chemistry 2016, 88: 8385-8389.

[12] Ji D, Feng H, Liew SW, Kwok CK: Modified Nucleic Acid Aptamers: Development, Characterization, and Biological Applications. Trends in Biotechnology 2023, 41: 1360-1384.

[13] Paul D, Sanap G, Shenoy S, Kalyane D, Kalia K, Tekade RK: Artificial Intelligence in Drug Discovery and Development. Drug Discovery Today 2021, 26: 80-93.

[14] Ballarotto M, Willems S, Stiller T, Nawa F, Marschner JA, Grisoni F, et al.: De Novo Design of Nurr1 Agonists via Fragment-Augmented Generative Deep Learning in Low-Data Regime. Journal of Medicinal Chemistry 2023, 66: 8170-8177.

[15] Zhavoronkov A, Ivanenkov YA, Aliper A, Veselov MS, Aladinskiy VA, Aladinskaya AV, et al.: Deep Learning Enables Rapid Identification of Potent DDR1 Kinase Inhibitors. Nature Biotechnology 2019, 37: 1038-1040.

[16] Knight CG, Platt M, Rowe W, Wedge DC, Khan F, Day PJ, et al.: Array-based Evolution of DNA Aptamers Allows Modelling of An Explicit Sequence-fitness Landscape. Nucleic Acids Research 2009, 37: e6.

[17] Iwano N, Adachi T, Aoki K, Nakamura Y, Hamada M: Generative Aptamer Discovery Using RaptGen. Nature Computational Science 2022, 2: 378-386.

[18] Yu Y, Wang L, Ni S, Li D, Liu J, Chu HY, et al.: Targeting Loop3 of Sclerostin Preserves Its Cardiovascular Protective Action and Promotes Bone Formation. Nature Communications 2022, 13: 4241.

[19] Wang L, Yu Y, Ni S, Li D, Liu J, Xie D, et al.: Therapeutic Aptamer Targeting Sclerostin Loop3 for Promoting Bone Formation Without Increasing Cardiovascular Risk in Osteogenesis Imperfecta Mice. Theranostics 2022, 12: 5645-5674.

[20] Zhang H, Yu S, Ni S, Gubu A, Ma Y, Zhang Y, et al.: A Bimolecular Modification Strategy for Developing Long-lasting Bone Anabolic Aptamer. Molecular Therapy. Nucleic Acids 2023, 34: 102073.

[21] Amu G, Zhang G, Jing N, Ma Y: Developing Stapled Aptamers with a Constrained Conformation for Osteogenesis Imperfect Therapeutics. Journal of Medicinal Chemistry 2024, 67: 18883-18894.

[22] Shen WX, Liu Y, Chen Y, Zeng X, Tan Y, Jiang YY, et al.: AggMapNet: Enhanced and Explainable Low-sample Omics Deep Learning with Feature-aggregated Multi-channel Networks. Nucleic Acids Research 2022, 50: e45.

[23] Zhou Z-H: Ensemble Methods: Foundations and Algorithms . New York: CRC Press; 2012.

[24] Gubu A, Ma Y, Yu S, Zhang H, Chen Z, Ni S, et al.: Unique Quinoline Orientations Shape the Modified Aptamer to Sclerostin for Enhanced Binding Affinity and Bone Anabolic Potential. Molecular Therapy. Nucleic Acids 2024, 35: 102146.

[25] Ma Y, Zhao W, Li Y, Pan Y, Wang S, Zhu Y, et al.: Structural Optimization and Additional Targets Identification of Antisense Oligonucleotide G3139 Encapsulated in a Neutral Cytidinyl-lipid Combined with a Cationic Lipid In Vitro and In Vivo. Biomaterials 2019, 197: 182-193.

[26] Shahrezaee M, Salehi M, Keshtkari S, Oryan A, Kamali A, Shekarchi B: In Vitro and In Vivo Investigation of PLA/PCL Scaffold Coated with Metformin-loaded Gelatin Nanocarriers in Regeneration of Critical-sized Bone Defects. Nanomedicine: Nanotechnology, Biology, and Medicine 2018, 14: 2061-2073.

[27] Sugatani T, Agapova O, Malluche HH, Hruska KA: SIRT6 Deficiency Culminates in Low-turnover Osteopenia. Bone 2015, 81: 168-177.

[28] Koblan LW, Erdos MR, Wilson C, Cabral WA, Levy JM, Xiong ZM, et al.: In Vivo Base Editing Rescues Hutchinson-Gilford Progeria Syndrome in Mice. Nature 2021, 589: 608-614.

[29] Kim W, Song ES, Lee SH, Yang SH, Cho J, Kim S-J: A New DNA Aptamer Which Binds to SARS-CoV-2 Spike Protein and Reduces Pro-inflammatory Response. Scientific Reports 2024, 14: 7516.

[30] Poojara LRK, Rawal RM: Computational Approaches Screening DNA Aptamers Against Conserved Outer Membrane Protein W of Vibrio Cholerae O1- An Investigation Expanding the Potential for Point-of-Care Detection with Aptasensors. Journal of Biomolecular Structure & Dynamics 2023, 41: 14438-14449.

[31] Yan Y, Tao H, He J, Huang SY: The HDOCK Server for Integrated Protein-Protein Docking. Nature Protocols 2020, 15: 1829-1852.

[32] Adasme MF, Linnemann KL, Bolz SN, Kaiser F, Salentin S, Haupt VJ, et al.: PLIP 2021: Expanding the Scope of the Protein-ligand Interaction Profiler to DNA and RNA. Nucleic Acids Research 2021, 49: W530-W534.

[33] Aldrees A, Khan M, Taha ATB, Ali M: Evaluation of Water Quality Indexes with Novel Machine Learning and SHapley Additive ExPlanation (SHAP) Approaches. Journal of Water Process Engineering 2024, 58: 104789.

[34] Ma Y, Zhu Y, Wang C, Pan D, Liu S, Yang M, et al.: Annealing Novel Nucleobase-lipids with Oligonucleotides or Plasmid DNA Based on H-bonding or π-π Interaction: Assemblies and Transfections. Biomaterials 2018, 178: 147-157.

[35] Niu B, Wu Y, Zhou M, Lin R, Ge P, Chen X, et al.: Precise Delivery of Celastrol by PEGylated Aptamer Dendrimer Nanoconjugates for Enormous Therapeutic Effect via Superior Intratumor Penetration Over Antibody Counterparts. Cancer Letters 2023, 579: 216461.

[36] Ma Y, Xie D, Chen Z, Shen X, Wu X, Ding F, et al.: Advancing Targeted Combination Chemotherapy in Triple Negative Breast Cancer: Nucleolin Aptamer-mediated Controlled Drug Release. Journal of Translational Medicine 2024, 22: 604.

[37] Tang W, Xu J, Xu C: Noncoding RNAs in the Crosstalk between Multiple Myeloma Cells and Bone Marrow Microenvironment. Cancer Letters 2023, 556: 216081.

[38] Cong W, Shen H, Jiang Y, Li L, Kong X, Chen S, et al.: Design, Synthesis, and Anti-Osteoporotic Characterization of Arginine N-Glycosylated Teriparatide Analogs via the Silver-catalyzed Solid-Phase Glycosylation Strategy. Journal of Medicinal Chemistry 2024, 67: 1360-1369.