Abstract
Introduction
Over 50% of all breast cancers are associated with a BRCA1 mutation. In addition to its roles in DNA repair, cell cycle regulation, ubiquitination, and transcriptional activation, it modulates ERα activity by directly binding to the AF-2 (coactivator-binding) region of ERα. Additionally, BRCA1 was determined to prevent epithelial-to-mesenchymal transition (EMT) and stemness. The cancer stem cell population of the tumor microenvironment is responsible for the development of chemoresistance, radio-resistance, metastasis, invasion, and tumor recurrence. Our lab established that lead BRCA1-mimetic small molecules, A7 and A9, can inhibit ERα, and identified a novel A9 analog, ABJ-1-195, inhibits tamoxifen-resistant stem cells. Directing our novel BRCA1-mimetic, ABJ-1-195, to a breast cancer stem cell-targeting peptide is crucial for specific drug delivery to prevent or halt metastasis and reduce off-target toxicity. The present study focuses on the synthesis of a linker to connect ABJ-1-195 to our fluorescent, stem cell targeting peptide construct.
Previously, EDCI-mediated coupling and conventional heating via the Schotten-Baumann reaction proved unsuccessful in overcoming the poor nucleophilicity of the exocyclic amine in the 4-aminopyrimidines A9 and ABJ-1-195 for linker synthesis. Unlike conduction heating, microwave reactors use dielectric heating, or oscillating high-frequency electric fields, to promote collisions between particles, allowing the swift generation of high heat and low pressure within a sealed reaction tube. Dipolar rotation and ionic conduction are crucial for improving reaction rates, yields, and safety while reducing side reactions and chemical waste in microwave-assisted synthesis.
Methods
In this study, our lab explored the use of predictive modeling to identify relevant parameters to increase the yield of the Schotten-Baumann reaction of A9 and ABJ-1-195 under microwave irradiation. A dataset was compiled from 22 Schotten-Baumann reactions using A9 or ABJ-1-195 performed under microwave conditions with variations in molar equivalent of starting materials, the molecular weights of starting material, the use of a catalyst, the molar equivalent of catalyst used, the dielectric constant of solvent, the dielectric loss of the solvent, the pKa of the used base, pressure applied to the reaction vial, reaction temperature, and reaction time. Additionally, approximately 120 published Schotten-Baumann reactions were also added to this dataset to evaluate the effects of these parameters on predicting reaction yield.
Results
Our model predictions indicate that reaction temperature, the inclusion of a catalyst, and the amount of the selected catalyst are relevant parameters for improving the reaction yield. We successfully incorporated the linker into A9 under the highest-yield microwave conditions, affording 48%.
Conclusion
Predictive modeling is a useful method for identifying relevant parameters to improve the yield of microwave-assisted Schotten-Baumann reactions with poor nucleophiles, A9 and ABJ-1-195. This work supports the synthesis of a novel peptide drug conjugate to target the breast cancer stem cell population. Ultimately, this work connects targeted drug delivery methods and machine learning to advance therapeutic strategies against drug-resistant cancers.