Technical Review: Synthesis and Translational Assessment of Trinucleotide 5′-Cap Analogs for Messenger Ribonucleic Acid-Based Therapeutics
- 11 hours ago
- 4 min read
Authors: Jeeyeon Kim, Kyeongwon Moon, Jihun Kim, Sung-Jun Park, Nari Kim, Yonggyu Jung, Yoonsuk Lee, Kyoungmin Lee, Wokchul Yoo, Jong Hoon Kim, Byeong-Won Kim, Daseul Kim, Hyun-Ju Park, Jaeheon Lee, Pargat Singh, and In Su Kim
Affiliations: Sungkyunkwan University, Suwon, South Korea
Background Introduction
The 5′-cap is an essential structural element at the terminus of eukaryotic mRNA that drives translation initiation by recruiting the eukaryotic initiation factor 4F (eIF4F) complex. In therapeutic mRNA, the identity and chemistry of this cap structure directly govern capping efficiency, translational output, protein yield, and mRNA stability against cellular decapping enzymes such as DcpS and hDcp2. Conventional approaches using dinucleotide analogs have been superseded by the trinucleotide Cap 1 paradigm, which offers higher capping fidelity and eliminates the need for 3′-O-methylation of the m7G moiety to prevent reverse incorporation. The commercial reagent CleanCap AG 3′OMe is one example of this technology, and it has been deployed in the production of the Comirnaty mRNA vaccine.
Building on this foundation, the authors sought to explore whether alternative ribose modifications at the 2′- and 3′-positions of the m7G unit could further improve binding affinity to eIF4E, boost translational efficiency, and confer greater resistance to decapping enzymes — all properties critical to next-generation mRNA therapeutics.
Materials and Methodology
The researchers began with in silico molecular docking and molecular dynamics simulations to predict the binding modes of various 2′- and 3′-modified cap analogs to the murine eIF4E receptor. Guided by these computational models, five novel trinucleotide m7GpppAmpG cap analogs were chemically synthesized through a multi-step process. In vitro, these compounds were assessed for IVT yield using T7 RNA polymerase, and capping efficiency was quantified via liquid chromatography-mass spectrometry. Translational output was measured in human embryonic kidney (HEK293T) cells using a dual-luciferase reporter assay, alongside binding affinity validations via fluorescence quenching titration and enzymatic resistance assays using human decapping enzymes. For the subsequent in vivo studies, firefly luciferase mRNAs capped with the most promising analogs were formulated into lipid nanoparticles (LNPs) and injected intravenously into CD-1 mice to track whole-body bioluminescence over 96 hours.
Results
Computational Design and Synthesis
Computational docking of a series of dinucleotide and trinucleotide analogs against eIF4E revealed that the 3′-O-mesylated scaffold retained a binding pose nearly identical to the unmodified reference (compound 1), with a docking score of –9.666 kcal/mol. Importantly, the steric bulk introduced by 3′-O-pivaloyl and mesyl groups did not disrupt key triphosphate and nucleobase interactions with Arg157, Lys159, and Lys162 of eIF4E, providing structural rationale for selecting these modifications as lead candidates. This in silico strategy efficiently narrowed the candidate pool prior to synthesis, streamlining resource allocation in the medicinal chemistry campaign.
In Vitro Transcription and Translation
The 3′-O-mesylated analog (compound 43) yielded excellent IVT mRNA production and achieved a remarkable capping efficiency of 97.1%, slightly outperforming the commercial benchmark. It also demonstrated a 1.8-fold increase in luciferase protein expression in HEK293T cells compared to the standard reference, while also recording the highest equilibrium association constant for the eIF4E receptor. This established compound 43 as a good replacement for existing commercial reagents, and therefore a solid template for further optimization of eIF4E-targeting cap analogs.
Decapping Resistance
mRNAs capped with 43 exhibited the strongest resistance to both hDcp2- and yeast Dcp-mediated decapping, with full-length mRNA integrity preserved under conditions that caused partial degradation of mRNAs capped with 9 and 42. Similarly, in DcpS hydrolysis assays at elevated enzyme concentrations (500 nM), compound 9 showed partial degradation at 60 minutes, while 42 and 43 remained fully intact.
In Vivo Translational Activity
LNP formulations 42a and 43a were administered intravenously to CD-1 mice, and bioluminescence imaging confirmed dose-dependent protein expression in the hepatic region. The 43a group exhibited consistently higher bioluminescence intensities compared to 9a and 42a across the 96-hour observation window. No mortality or adverse clinical signs were observed during the 14-day monitoring period, supporting the preliminary safety profile of this novel capping agent.
Conclusion
Kim and colleagues established the 3′-O-mesylated trinucleotide cap analog as a robust and translationally superior 5′-capping agent for mRNA therapies. The compound consistently outperformed CleanCap AG 3′OMe across capping efficiency, enzymatic stability, and in vivo protein expression.
This paper utilized the PreciGenome NanoGenerator Flex-M for LNP synthesis. Its microfluidic mixing enabled uniform synthesis conditions essential for uniform RNA-LNPs, and therefore efficient targeted delivery of mRNA payloads.
For further details, check out the link to the article here:
Comments