Position-specific ORF nucleoside-ribose modifications enabled by complete chemical synthesis enhance mRNA stability and translation

Iwai, Hiroto; Kimura, Yasuaki; Honma, Masakazu; Nakamoto, Kosuke; Hashimoto, Atsushi; Motosawa, Keiichi; Atago, Takayuki; Asano, Kana; Hashiya, Fumitaka; Abe, Naoko; Kobayashi, Keiko; Ogisu, Ryoko; Yamada, Hiroki; Hiraishi, Keiko; Saito, Seiji; Yamamoto, Junichiro; Abe, Hiroshi

doi:10.1038/s41467-025-65788-8

Download PDF

Article
Open access
Published: 19 November 2025

Position-specific ORF nucleoside-ribose modifications enabled by complete chemical synthesis enhance mRNA stability and translation

Nature Communications volume 16, Article number: 9995 (2025) Cite this article

10k Accesses
2 Citations
38 Altmetric
Metrics details

Subjects

Abstract

Despite the remarkable success of mRNA vaccines, improving the translational efficiency of mRNA therapeutics remains a critical challenge to their widespread clinical application. Here we systematically evaluate chemical modifications to improve the translational activity and stability of uncapped mRNA. We employ a primarily chemistry-based synthetic approach, which is crucial for the position-specific introduction of chemical modifications, enabling detailed structure-activity relationship studies, hitherto unattainable with conventional methods. A pivotal innovation herein is the introduction of 2´-F modification at the first nucleoside of the codon in the open reading frame, which significantly bolsters the stability of mRNA without compromising its translation. Additional modifications at the 5´-UTR and poly(A) tail with other types of nucleoside and phosphate analogs also exemplify the importance of terminal modifications for improved translation. Precise control of these modification patterns achieves higher peptide expression than conventional in vitro-transcribed mRNA. These findings offer a unique framework for designing effective mRNA-based therapeutics.

Expedient production of site specifically nucleobase-labelled or hypermodified RNA with engineered thermophilic DNA polymerases

Article Open access 09 April 2024

Combinatorial optimization of mRNA structure, stability, and translation for RNA-based therapeutics

Article Open access 22 March 2022

RNA modifications in physiology and disease: towards clinical applications

Article 15 September 2023

Introduction

Research on the development of messenger RNA (mRNA) therapeutics using mRNA as drug molecules has been greatly accelerated by the successful application of mRNA vaccines against SARS-CoV-2^1,2. mRNA therapeutics^3,4 have a variety of applications, including vaccines against infectious diseases using viral and bacterial antigens⁵, cancer vaccines based on cancer cell-specific antigens^6,7, and protein replacement therapy for diseases caused by genetic mutations^8,9,10. Compared to protein replacement therapy using proteins as drug molecules, mRNA therapeutics offer advantages in terms of molecular design simplicity, predictability of molecular properties, and inherent eukaryotic post-translational modifications vital for protein function. Unlike conventional DNA-based gene therapy, on the other hand, mRNA therapeutics do not require transport to the nucleus for gene expression, have little risk of genome integration, and allow for transient expression of therapeutic effects, enabling better therapeutic control.

The following issues are generally recognized for the future application of mRNA therapeutics as general-purpose medicines^3,5: mRNA in vivo stability, insufficient translational activity, induction of undesired immune responses, and devising effective delivery methods to target tissues and cells. Among these challenges, introducing non-canonical chemical modifications has shown promise in enhancing mRNA stability and translational activity and reducing immune responses^11,12,13. Improvement of mRNA function by introducing chemical modifications has been studied mainly for the 5´-cap, and various cap derivatives have been developed to improve stability by imparting decapping resistance^{14,15,16,17,18,19}, as well as for tool-oriented applications^20,21,22. Although less common, modifications to the poly(A) tail^23,24,25 and region-specific thiophosphate modifications in the 5´-untranslated region (UTR)²⁶ have been explored to enhance protein production. In addition, the use of non-canonical nucleobases^27,28,29,30 can effectively mitigate undesired immune responses^{31,32,33,34,35} triggered by exogenous mRNA. It is well recognized that the use of these modified nucleobases plays a significant role in the success of mRNA vaccine development against SARS-CoV-2. These studies underscore the potential of chemical modifications and the use of non-canonical structures to address the challenges of mRNA therapeutics. However, conventional enzyme-based preparative methods for mRNA greatly limit the types and patterns of chemical modifications that can be introduced. Modification of the ribose ring would significantly improve biological stability, as exemplified by oligonucleotide therapeutics^36,37,38,39 such as antisense oligonucleotides (ASOs) and siRNAs. However, there are two significant challenges in applying the same concept to mRNAs. One is that ribose modifications in ORFs, as known from examples such as 2´-O-methyl (2´-OMe)^30,40,41, would significantly decrease translational activity. The second issue is the synthetic method used for chemically modified mRNA. mRNA is basically synthesized by enzymatic transcription, but even in transcription using engineered RNA polymerases, the types of sugar-modified nucleosides that can be introduced are very limited^{42,43,44,45,46}. In addition, when preparing mRNA with a single composition, as in pharmaceutical applications, substitutions with modified nucleosides are inevitably made throughout for the specific nucleobase. In other words, by using conventional transcription-based synthetic methods, it is impossible to introduce position-specific chemical modifications to maximize both mRNA stability and translational activity.

On the other hand, our group has also been conducting research to solve issues in mRNA therapeutics from different aspects, including nano-structure design of mRNAs such as circular mRNA^47,48,49,50, introduction of chemical modifications such as specific phosphorothioate modification to 5´-UTR²⁶, development of cap analogs for in vitro transcription (IVT) that enable purification of capped mRNAs with high-purity and translational activity⁵¹, and chemical capping reaction on RNA for complete chemical synthesis of mRNA⁵².

Here, we show the possibility of introducing chemical modifications into the ribose scaffold of nucleosides, especially those in the open reading frame (ORF) regions, as a pioneering approach for developing effective mRNAs (Fig. 1A). Chemically modified mRNAs developed in this study are prepared by first synthesizing the corresponding RNA oligonucleotides using phosphoramidite chemistry and then concatenating these RNA fragments by enzymatic or chemical ligation reactions (Fig. 1B) to obtain full-length mRNAs. This method enables the introduction of many different types of chemical modifications in various patterns in the synthesis of RNA fragments and allows precise evaluation of their effects on the translational activity and stability of the mRNA.

**Fig. 1: Site-specific nucleosugar-modified mRNA to enhance translation.**

Results and discussions

First, to screen for modification patterns, we designed a sequence of 91 nucleotides (nt) uncapped RNA that could be directly synthesized without ligation, and evaluated its translation activity in a cell-free translation system with HeLa lysate (Fig. 2A). The sequence consisted only of 5´-UTR (β-globin UTR)⁵³ and ORF, and was designed to encode Flag-His₆ peptide⁵² to perform a sandwich ELISA for the evaluation of translational activity. As modifications, 2´-fluoro (2´-F), 2´-OMe, 2´-O-methoxyethyl (2´-O-MOE), locked nucleic acid (LNA), and DNA etc., were selected. mRNA modified with 2´-OMe at both ends (NK002) showed 4-fold higher translational activity than unmodified RNA (NK001) (Fig. 2B). Other types of modifications, such as 2´-F, LNA, DNA, 2´-O-MOE, and phosphorothioate, were also examined for possible introduction into the mRNA terminus (Fig. S1A). LNA (NK011) modification resulted in a significant loss of translational activity (Fig. S1B), whereas DNA (NK012) and 2´-O-MOE (NK009) modifications maintained translational activity, including additional phosphorothioate modifications (NK014 and NK015), as well as for 2´-OMe (NK013) (Fig. S1C). Next, the introduction of the 2´-F modification into a specific position in the codon unit was tested; the modification was specifically introduced at the codon’s first, second, or third nucleoside over the ORF (NK003, NK004, NK005) (Fig. 2A). Interestingly, while the translational activity of mRNA with the modification at all of the second or third nucleoside in the codon unit (2nd NC, 3rd NC) was suppressed by nearly 30–50 % level, the modification at the first nucleoside in the codon unit (1st NC) did not have strong deleterious effects on translational activity. This translational acceptability of the 2´-F modification of the 1^st NC was also observed in another sequence (67 nt mRNA coding Flag-His₆). (Fig. S2). In contrast, other types of sugar modifications showed a negative effect on translation (Fig. 2C) when all of the 1st NC were modified, translation was suppressed in the case of the 2´-OMe (NK006) and 2´-O-MOE (NK007) modifications, and a nearly 50% reduction was observed for the deoxyribose modification (NK008).

**Fig. 2: Effect of sugar modification of the ORF of mRNA on translation activity.**

Subsequently, screening for the optimum chemical modification patterns of the terminal region (6 nt at the 5´ terminus and 3 nt at the 3´ terminus) was performed with a longer mRNA (145 nt) encoding three repeats of FLAG peptide and His-Tag (Fig. 3A). The mRNAs were prepared by chemical or enzymatic ligation reactions between the 5´ side 80 nt and the 3´ side 65 nt RNA fragments that were prepared using an automated oligonucleotide synthesizer. The method of ligation (enzymatic or chemical) was determined based on the relative positions of the ligation point and chemically modified sites. For substrates without chemical modifications near the ligation point, enzymatic ligation was selected based on the reported tolerance of sugar-modified nucleoside analogs to ligation by RNA ligase 2⁵⁴. In cases where the region around the ligation point contains 2´-modified nucleosides, the chemical ligation method was selected with the 3´ end of the 5´side -RNA fragment on being 2´-F, 3´-phosphate, and the 5´ end of the 3´ side- RNA fragment being a hydroxyl group. The conditions for the chemical ligation method were slightly modified from those previously reported^55,56 In the case of RNA prepared by chemical ligation, the target ligated RNA strand was confirmed by LC-MS analysis (Fig. S3). 2´-F modification on the terminus increased peptide production (NK024) to some extent, while LNA and BNA-NC (N-Me) modification significantly restrained translation (NK026, NK027) (Fig. 3B). 2´-O-MOE modification showed a positive effect on the translation (NK025) with a similar level as 2´-OMe modification (NK023), and the positive effect was further improved by the additional phosphorothioate modification to the terminal region (NK028).

**Fig. 3: Effect of sugar and backbone modification of the 5’ -UTR and 3’ poly(A) of mRNA on translation activity.**

Based on these findings, terminal modification of mRNA with polyadenine (poly(A)) tails was evaluated. To the 145 nt RNA with 2´-O-MOE modification at the terminus, 20 nt of poly(A) with various modification types was also prepared (Fig. 3C). The types of modifications tested on poly(A) were as follows: 2´-F modification for every 2 nt (NK030), cross modification with 2´-F and 2´-OMe (NK031), complete modification with 2´-OMe (NK032), and 2´-O-MOE (NK033). The terminal 3 nt of the modified poly(A) was set as 2´-O-MOE with a phosphorothioate linkage. Evaluation of the translational activity revealed that the addition of non-modified poly(A) (20 nt) increased the translated peptide amount, and the positive effect was further enhanced by these modifications on poly(A), with 2´-F modification for every 2 nt (NK030) being slightly better than the others (Fig. 3D). In contrast to the ORF cases, various modifications were accepted on poly(A) without loss of translational activity. To evaluate whether the modified poly(A) tail retained the intrinsic biological function, we measured the binding affinity of Poly(A)-Binding Protein (PABP) to chemically-synthesized poly(A) fragments based on the Surface Plasmon Resonance binding assay. The 5´-biotinylated poly(A) fragments were immobilized onto a streptavidin-coated sensor chip, and the response rate for PABP binding was measured (Fig. S13). As expected, PABP exhibited strong binding to unmodified poly(A) RNA (NK065; K_D = 4.4 ± 0.3 × 10⁻¹⁰mol/L), while it did not bind to the randomized sequence (C5A5U5G5; NK067).⁵⁷ Regarding chemically modified fragments, modified poly(A) showed a weaker binding to PABP than unmodified poly(A) (NK066; K_D = 4.8 ± 0.2 × 10⁻⁹mol/L); however, the binding was superior to nonspecific binding observed for C5A5U5G5 sequence with the same modification pattern (NK068; K_D = 4.4 ± 0.1 × 10⁻⁸mol/L). These results supported the binding of NK066 to PABP. A previous study reported that sugar modifications of poly(A) impair PABP binding.⁵⁸ Nevertheless, the combined effect of sugar and PS modification has not been addressed, suggesting the potential for a positive effect from this combination. The biological function of chemically modified poly(A) tail was also evaluated by the subsequent translation competition assays. The translation from synthesized mRNA with modified poly(A) (NK041) was competitively inhibited by the addition of poly(A) RNA fragment, while it remained unchanged by the addition of poly(C) (Fig. S14).

To clarify the effect of chemical modification of mRNA at various positions, 5´-UTR, ORF, and poly(A), mRNA samples with different modification sites and patterns were prepared (Fig. 3E), and their translational activities were compared in HeLa cell extracts. The samples tested were as follows: RNA with no modification (NK034), modified only at the 5´-UTR (NK035), ORF (NK036), poly(A) (NK037), modified both at the 5´-UTR and poly(A) (NK038) and modified over all regions (NK039). As shown in Fig. 3F, each modification to either the 5´-UTR, ORF, or poly(A) increased translational activity, and a stronger increment of the translated peptide was observed by the modification of ORF and poly(A) than that for the 5´-UTR. Interestingly, the combination of the 5´-UTR and poly(A) modification (NK038) counteracted the positive effects of each modification, and the additional modification of the ORF enhanced translation (NK039). Additionally, LC-MS/MS analysis was performed for the extracted peptide after the in vitro translation in rabbit reticulocyte lysate, and both non-modified (NK034) and chemically modified (NK041) mRNA administration predominantly afforded the target peptide MS peak (Fig. S4), suggesting no negative effect on mRNA decoding due to chemical modification in the ORF.

After identifying the candidates for the optimum modification patterns, four mRNA samples were designed and synthesized with various modification patterns for each region (Fig. 4A). All four samples contained ribose modification in the ORF except for the start codon, 2´-F modification at the 1st NC, and the level of chemical modification on both terminals was changed among these four samples, as shown in Fig. 4A. Evaluation of the translational activity of these modified mRNA revealed that, as the level of terminal modification increased, the amount of translated peptide slightly decreased (NK030 vs. NK031, NK040, NK041) (Fig. 4B). The time course of translation was also evaluated over 30 min (Fig. 4C). While mRNA without any modifications (NK034) showed maximum translation at 5 min after administration, the modified RNA (NK030, NK031, NK041) increasingly produced the peptide over 30 min, with moderately modified RNA (NK030) showing the highest translation level at every time point.

**Fig. 4: Effects of sugar and backbone modifications on mRNA stability and translation activity.**

The stability of mRNA in HeLa lysates was evaluated using reverse transcription quantitative polymerase chain reaction (RT-qPCR) for the reaction mixture (Fig. 4D). While the unmodified RNA (NK034) showed fast clearance from the mixture, the modified mRNAs (NK030, NK031, and NK041) stably remained in the reaction mixture without degradation for 30 min at 37 ^oC. While the amount of remaining mRNA was nearly comparable for the three types of modified mRNAs, there was a significant difference in the amount of the translation product (Fig. 4C). These results suggest that moderate and high levels of chemical modification at the terminus improved the stability of the mRNA, but the latter negatively influenced its translational activity.

The effect of introducing chemical modifications to each region of the mRNA on translation activity was also evaluated in a cell-based system (NK034 – NK039, Fig. S6B). Consistent with the results in the cell lysate (Fig. 3F), the introduction of modifications to the ORF led to a strong enhancement in translation activity (NK036 and NK039). Additional modification both in the 5´-UTR and poly(A) cooperatively augmented the translational activity (NK039), which was more obvious than the case with the cell lysate system (Fig. 3F). The 2’-F modification at the 1^st NC of start codon did not affect the translation activity of the mRNA, demonstrating it could be incorporated throughout the ORF (NK041 vs. NK052, Fig. S6D). The time course of translation was evaluated using HeLa cells over an extended incubation period of up to 48 h. In the case of lipofection (Fig. 4E), mRNAs that were highly modified in the poly(A) moiety (NK031 and NK041) afforded the highest amount of translational product over 48 h compared to mRNAs with the other two types of modification (NK030 and NK040). Notably, the translation level of the unmodified mRNA (NK034) was significantly lower than that of the modified mRNAs, clearly reflecting the results of the stability evaluation (Fig. 4D). While the modification level in the 5´-UTR region had little effect on the translation level in the case of mRNAs with a higher level of modification in the poly(A) region (NK031 vs. NK041), a lower level of modification of the 5´-UTR resulted in better translation activity for mRNA with less modification of the poly(A) region (NK030 vs. NK040). In the case of electroporation (Fig. 4F), the trend was somewhat different from that observed in the lipofection cases; mRNA with high modification at both termini (NK041) showed the highest level of translation without decreasing the peptide amount over 24 h. Both these experiments suggested that modification of the poly(A) region was more critical for increasing the translation amount than that on the 5´-UTR region in the cellular system.

The amount of mRNA remaining in HeLa cells after electroporation was quantified by RT-qPCR (Fig. 4G). The result was consistent with the translation level in the electroporation experiment (Fig. 4F); the mRNA was more stable as the level of chemical modification increased. In addition, no cytotoxicity was observed for the transfection of both non-modified and chemically modified mRNA (Fig. S7, NK034, 030, 031, 040, 041) at the concentration range evaluated for translational activity.

Various reasons can be considered for the different optimal modification patterns observed in HeLa cell lysates and cell lines. One possible explanation for this is the varying impact of the mRNA degradation system. Specifically, in the HeLa cell lysate system, where the effect of mRNA degradation was relatively minor, the introduction of extensive chemical modifications at both ends of the mRNA inhibited translation. In contrast, in the cell system, the significant influence of the mRNA degradation system suggests that resistance to degradation plays a major role in determining translational activity.

A similar trend in the positive effect of the chemical modification was also observed in other cell lines, where mRNA was administrated to hAoSMCs (human aortic smooth muscle cells) by electroporation, and the time course of the translation was evaluated (Fig. S5A). In addition, the versatility of the design strategy was confirmed by other mRNA sequences, the ones encoding the same hybrid peptide, three repeats of FLAG and HisTag, but with different 5´-UTR sequences (Fig. S5B, C), and the others encoding a different peptide composed of FLAG and Epidermal Growth Factor (EGF) (Fig. S5D). Notably, the latter has a different ORF sequence, but the 2´-F modification on the 1st NC except for the start codon effectively increased translational activity, demonstrating the broad applicability of the effects of the modification on translational activity. For the FLAG-EGF construct, the time course of translation was also evaluated. By introducing chemical modifications into mRNA (NK051), it was found that the peak translation activity was around 12 h, and the half-life from the maximum translation level was more than 36 h (Fig. S5E). Considering a previous report that the stability of recombinant EGF in several buffer solutions is limited to a few hours,⁵⁹ it could be suggested that mRNA, particularly chemically modified mRNA, has advantages over direct peptide administration. Nonetheless, further investigation is required to test whether chemically modified mRNA administration with established delivery methods such as LNP^60,61,62 afforded more sustained pharmacological effects in vivo than direct administration of recombinant EGF.⁶³

Finally, the stability and translation activity of chemically modified and canonical mRNA prepared by IVT (NK060) with co-transcriptional capping with CleanCap were compared (Fig. 5A). The capping ratio of the co-transcriptionally capped mRNA was 84% based on DNAzyme-mediated 5´-terminus cleavage reaction (Fig. S12A). No damage to a cap structure nor structural change during isolation by dPAGE was confirmed by LC-MS analysis of the isolated mRNA (Fig. S12C). Based on the above studies, chemically modified uncapped RNA (NK041) was designed to have intensive chemical modification at both termini and 2´-F modification at the 1st NC in ORF except for the start codon and was entirely prepared by chemical method; the 5´ side 80 nt RNA fragment with 3´ phosphate terminus was chemically ligated with the 3´-side 85 nt RNA fragment with a 5´ -hydroxy terminus. As additional control samples, m1Ψ-modified co-transcriptionally capped IVT RNAs with 20 nt (NK062) and 80 nt (NK063) poly(A) were also prepared.

**Fig. 5: Improvement of mRNA stability and translation activity by sugar and backbone modifications.**

For stability evaluation, mRNA was incubated in 1.6% diluted mouse serum at 37 ^oC, and the amount of remaining mRNA was evaluated by RT-qPCR (Fig. 5B), where a calibration curve was created for each mRNA sample to offset the effects of modifications on reverse transcription (Fig. S8E). IVT mRNA (NK034) degraded rapidly, leaving less than 5 % of the sample after 90 min. Introducing m1Ψ and extending poly(A) length were both effective, as these modified mRNAs remained at approximately 6.0% (NK060) and 21% (NK063) at the same time point. In contrast, chemically modified RNA without a cap (NK041) showed higher stability; ~48% of RNA remained after 90 min of incubation.

In terms of translational activity, while non-modified mRNAs (NK034 and 060) afforded a very low translation level, regardless of the presence or absence of a 5´ cap (Figs. 5C, D, S8A), m1Ψ-modified capped mRNAs with 20 nt (NK062) and 80 nt (NK063) poly(A) enhanced the translational activity, especially at earlier time points, with longer poly(A) effectively enhancing the activity. Chemically modified uncapped mRNA (NK041) stably expressed much higher levels of the peptide in HeLa cells over 24–48 h compared to other mRNA (Fig. 5C, D). Notably, m1Ψ-modified mRNA with longer poly(A) showed higher translational activity than chemically modified mRNA at the early time point, but its decay was faster than that of NK041. After 24 h, NK041 showed 6–16 fold higher translational activity than NK063, suggesting a strong stabilizing effect by chemical modification (Fig. 5C). The same trend in translational level was observed in the electroporation system (Fig. 5D). These results clearly demonstrate that appropriate chemical modifications enhance peptide production from mRNA in cellular systems.

Immune responses (IL6, TNF, and IL1B) upon the administration of mRNA were also evaluated (Fig. S8B–D). Chemically synthesized mRNAs (NK034 and NK041), as well as m1Ψ-modified mRNAs (NK062 and 063), did not show significant immunogenicity, while non-modified IVT mRNA (NK060) showed a high levels of immune-related gene expression, especially at higher doses. Uncapped IVT mRNA (NK056-059) showed the same trend of immune responses (Fig. S9A, D-I), and these results were consistent with the dsRNA detection results (Fig. S10).

The effect of cap introduction on this short mRNA sequence was also evaluated using several control samples (NK056-NK063, Fig. S9A-C); In terms of translational activity, only capped and m1Ψ-modified mRNA (NK062 and NK063) showed a significant level of peptide expression 5 h after administration. Collectively, no immune response was observed with m1Ψ-modified mRNA, and the introduction of the cap surely enhanced translational activity. In other words, in this construct, if the influence of the immune response is eliminated, enhancement of translational activity by the introduction of the cap can be observed. Furthermore, the effect of cap introduction on chemically modified mRNA was evaluated (Fig. S11). An improvement in translational activity by the introduction of modifications in capped mRNA (NK055 vs. NK053, 054) was confirmed. On the other hand, in comparisons among chemically modified mRNAs, uncapped mRNA (NK052) showed higher translational activity than capped mRNA (NK055), suggesting certain contribution of cap-independent translation mechanisms.^64,65,66 Cap-dependency of translation activity on sugar-modified synthetic mRNA remains a topic for future investigation.

In conclusion, we have systematically demonstrated the effectiveness of chemical modification of the sugar moiety of ribonucleosides in enhancing peptide production from uncapped mRNA. The pivotal discovery in this study is that the position-specific 2´-F modification in the 1st NC (first nucleoside in the codon unit) could be accepted without reducing the translational activity and effectively the biological stability of the mRNA. Augmented by the additional chemical modification on both termini, the chemically modified mRNA, even without a 5´ cap, yielded a greater amount of peptide over a longer duration than the IVT-derived canonical mRNA. Site-specific chemical modification for a high level of translation was enabled by chemical synthesis of the RNA fragments. Ligation reactions with these RNA fragments allowed for the preparation of chemically modified mRNA exceeding 200 nt in length. Building on the SAR insights from this study, we are pursuing the development of highly efficacious mRNA therapeutics. Owing to their high and long-lasting activity, chemically modified mRNA might provide novel therapeutic options that are difficult to achieve with IVT-derived mRNAs in the evolving area of mRNA therapeutics, such as cancer vaccines⁷, protein replacement, and tissue regeneration therapies.

Methods

Synthesis of RNA fragments and splint DNAs

RNA fragments and splint DNAs for enzymatic or chemical ligation reactions were purchased from Gene Design or Hokkaido System Science, where they were synthesized using standard phosphoramidite methods. The oligonucleotides used in this study are listed in Tables S1–S6. The dPAGE analysis of the evaluated RNA samples is presented in Figs. S15–S19. The concentration of the oligonucleotide and ligated mRNA was measured by NanoDrop2000 (Thermo Fisher Scientific) using the extinction coefficient calculated by OligoAnalyzer (Integrated DNA Technologies, Inc.).

Mass spectrometry analysis

The structures of the solid-phase synthesized oligonucleotides were determined by liquid chromatography-mass spectrometry (LC-MS) using an Agilent 6120 series single quadrupole LC-MS system (Agilent Technologies). Liquid chromatography was performed using an ACQUITY BEH C18 column (1.7 μm, 2.1 × 50 mm; Waters) with buffers A (8.6 mmol/L trimethylamine and 100 mmol/L hexafluoroisopropanol in water) and B (methanol). The column temperature was 60 °C. The gradient ran from 10% to 90% B over 9.0 min (for Tables S1–S3) or 18.0 min (for Fig. S3) at 0.60 mL/min. Water injections served as blanks. Mass spectrometry was performed in negative ion mode with a negative capillary voltage of 3000 V, full scan range m/z 500–2000. The acquired MS data were deconvoluted and reviewed using Agilent OpenLab CDS ChemStation Edition (Version C.01.09). All data were obtained from single measurements of each sample (n = 1).

General procedure for enzymatic ligation reactions

A mixture containing RNA fragments No.1 and 2 (each 10 nmol, final conc. of 50 μmol/L) (Table S2) and splint DNA (10 nmol, final conc. of 50 μmol/L) (Table S3) in T4 RNA Ligase 2 Reaction Buffer (New England BioLabs, NEB) was heated at 90 °C for 5 min and then gently cooled to room temperature. Then, T4 RNA Ligase 2 (10 units) and 60% PEG6000 (final conc. of 15%) were added and incubated at 37 °C for 16 h. The ligated RNAs were purified using preparative 5% denaturing PAGE and isolated using the crush-and-soak method to obtain the ligated products. The gel extraction was performed with 1× TE buffer and desalting with Amicon 10 K (Merck Millipore) and ethanol precipitation was performed and the mRNA was stored in 1 mmol/L sodium citrate (pH 6.5) buffer. For the preparation of some lots of NK023 and NK064, the following RP-HPLC purification was performed. After the enzymatic ligation, the crude samples were treated with Recombinant DNase I and then purified by HPLC with XBridge Oligonucleotide BEH C18 OBD Prep Column (130Å, 2.5 μm, 10 mm × 50 mm) and gradient buffer containing 0–50% MeCN, 100 mmol/L hexylammonium acetate in H₂O. The yield of each ligated mRNA is shown in Table S4.

General procedure for chemical ligation reactions

A mixture containing RNA fragments Nos. 56 and 57 (each 10 nmol, final conc. of 50 μmol/L) (Table S2), splint DNA (10 nmol, final conc. of 100 μmol/L) (Table S3) in 100 mmol/L NaCl was heated at 90 °C for 5 min and gently cooled to room temperature. ZnCl₂ solution (final conc. of 5 mmol/L) and N-Cyanoimidazole (final conc. of 5 mmol/L) in DMSO were added and incubated at 30 °C for 20 h. The ligated uncapped RNA was purified using a NAP-10 column and preparative 5% denaturing PAGE. The yield of each ligated mRNA is shown in Table S4. Generally, the ligation yield was moderate (approximately 40 %, Fig. S20); however, the final isolated yield remained low, mainly because of sample loss during the isolation process, which could be overcome by scaling up the reaction.

Chemical synthesis of 5´ capped mRNA

5´capped RNA fragment No. 69 was prepared from starting fragment No. 68 based on a previously described chemical capping reaction.⁵² After lyophilizing the RNA solution, the following solvents and reagents were added at the final concentrations listed below, and the solution was incubated at 55 °C for 3 h: DMSO (10 µmol/L RNA), 7-methylguanosine 5´-diphosphate imidazolide (10 mmol/L), and N-methyl imidazole (1.3 mol/L). After the reaction, the mixture was diluted with MQ and desalted using Amicon 10 K, followed by isopropanol precipitation. The product was purified using reverse-phase HPLC (Column: YMC-Triart Bio C18, Eluent: A: 50 mmol/L TEAA, 5% MeCN, B: MeCN, Linear gradient 0–40 % B over 30 min). Nearly quantitative capping was confirmed by dPAGE analysis (Fig. S12B). LC-MS analysis of the isolated fragment was performed; calculated: 26877.6, observed: 26880.7. The condition was shown below; Column: ACQUITY UPLC BEH C18 Column, 130 Å, 1.7 µm, 2.1 mm × 50 mm; Column temp. of 60 °C.; Linear gradient of 5–30% over 20 min; Solvent A: MeOH; Solvent B: 8.6 mmol/L TEA/100 mmol/L HFIP solution; and Flow rate: 0.3 mL/min.

The obtained fragment No. 69 was determined by LC-MS as described above. RNA fragments Nos. 69 and 70 were chemically ligated and then treated with Recombinant DNase I as described above. The reaction mixture was then purified by HPLC with a COSMOSIL RNA-SEC-1000 Packed Column (7.5 × 300 mm) and 20 mmol/L phosphate buffer (pH 7.4).

Preparation of in vitro-transcribed mRNA and its purification by dPAGE

Template DNA for IVT was prepared by PCR using pUC57amp/G-3×FLAG-6×His-DHFR as the template. The plasmid and PCR primer was purchased from IDT and Sigma-Aldrich, respectively. These sequences are listed in Table S5.

To prepare 5′ capped mRNA by co-transcriptional capping, the following PCR primers were used: FW-AGG and RV-PA20 (for preparation of poly(A)20) or FW-AGG and RV-PA80 (for preparation of poly(A)80). The PCR mixture consisted of 0.25 μmol/L primers, 0.25 ng/μL pUC57amp/G-3×FLAG-6×His-DHFR vector, 1× PCR buffer of PrimeSTAR MAX DNA Polymerase (Toyobo). The mixture was subjected to the thermal cycling reaction as follows: 98 °C, 30 s, (98 °C, 10 s, 55 °C, 5 s, 72 °C, 5 s) × 30 cycles, 72 °C, 5 min. The reaction was treated with 100 U/μL DpnI and then analyzed by agarose gel electrophoresis, and gel-excised dsDNA was purified with the QIAquick PCR Purification Kit (QIAGEN).

Transcription reactions were conducted with MEGAScript T7 Transcription Kit (Invitrogen) based on the manufacturer’s protocol. A reaction mixture containing 1 ng/μL DNA (PCR product) and 8 mmol/L NTPs with or without 6 mmol/L CleanCap AG was incubated at 37 °C for 6 h. To the reaction mixture was added Turbo DNase I, which was further incubated at 37 °C for 15 min, and then purified using the Monarch RNA Cleanup Kit (NEB). The purified capped-RNA was then treated with 5 U/μL Antarctic Phosphatase Enzyme (NEB) and 1×Antarctic Phosphatase Buffer at 37 °C for 1 h. The obtained mRNA was analyzed by dPAGE, and gel-purified mRNA was extracted using buffer. mRNA was recovered by alcohol precipitation.

To prepare 5′ capped mRNA by post-transcriptional enzymatic capping, FW-GGG and PA20 were used as PCR primers. PCR was performed with the KAPA HiFi HotStart ReadyMixPCR Kit (Roche) based on the manufacturer’s protocol. Transcription reactions were conducted using the MEGAScript T7 Transcription Kit, 4 ng/μL DNA, and 9 mmol/L NTPs. After DNase I treatment and purification, the resultant RNA was capped with Vaccinia Capping System (NEB) and ScriptCap 2’-O-Methyltransferase Kit (CELLSCRIPT) based on the manufacturer’s protocol. The reaction mixture containing 0.5 μg/μL RNA, 0.1 mmol/L SAM, 0.5 mmol/L GTP, 0.5 U/μL Vaccinia Capping Enzyme, 2.5 U/μL 2’-O-Methyltransferase, 2 U/μL RNase Inhibitor, and 1× Capping Buffer was incubated at 37 °C for 1 h and then purified with Monarch RNA Cleanup Kit. Purified capped-RNA was then treated with Antarctic Phosphatase and gel-purified. The other procedures were the same as those described above. The sequences of the prepared mRNAs are described in Table S6.

In vitro translation reaction using HeLa cell lysate

Chemically synthesized uncapped mRNAs were preheated at 90 °C for 3 min and cooled to 4 °C before the translation assay and degradation evaluation. The in vitro translation reaction was performed using a 1-Step Human Coupled IVT Kit (Thermo Fisher Scientific) according to the manufacturer’s protocol. The reaction mixture (10 μL) containing each concentration of mRNA and 0.8 U/μL Murine RNase Inhibitor was incubated at 37 °C. The reaction was stopped by incubation at 72 °C for 5 min.

In vitro translation reaction using rabbit reticulocyte lysate (translation competition assay)

The in vitro translation reaction was performed using a rabbit reticulocyte lysate kit (Promega) according to the manufacturer’s protocol. The reaction mixture (10 μL) containing 300 nmol/L mRNA, 10 μmol/L poly(A) fragments and 1.6 U/μL Murine RNase Inhibitor was incubated at 37 °C. After incubation, reaction mixture was diluted in 0.1 mol/L Carbonate buffer (pH 9.4) to stop the reaction.

Surface plasmon resonance analysis for the measurement of binding affinity between recombinant PABP protein and modified poly(A) fragments

The vector for expressing human PABP, pCS-6xHis/TEV/hPABPC1[NM_002568.4], was constructed and packaged by VectorBuilder Inc. with vector IDs VB231222-2539wqp. The protein was expressed in E. coli BL21 (DE3) Star cells (Invitrogen) and cell pellets were lysed with B-PER (Thermo Fisher Scientific) containing 125 units/mL benzonase (Merck). Nucleic acids bound to PABP were digested by benzonase and further removed with poly(ethyleneimine) precipitation. The proteins were purified using cOmplete His-Tag Purification Resin (Roche), followed by a Superdex 200 increase GL 10/300 column (Cytiva).

The binding affinity of human PABP to the poly(A) fragments was determined using Biacore T200 (GE Healthcare). 5’-biotinylated poly(A) fragments were purchased from Gene Design. The poly(A) fragments were captured on streptavidin-coated sensor chip (GE Healthcare) in a running buffer (10 mmol/L HEPES, 600 mmol/L NaCl, 3 mmol/L EDTA, 0.005% v/v surfactant P20 pH 7.4). Recombinant PABP was injected for 2 min and allowed to dissociate for 3 min. The flow-cell surface was regenerated with 0.1% SDS for 2 min to completely remove PABP bound to the poly(A) fragments. Kinetic parameters were determined using 1:1 binding model.

In vitro transfection of mRNAs for measurement of RNA stability and translation activity using cultured HeLa cells

The translational activity of each uncapped mRNA was evaluated in vitro using HeLa cells. In the lipofection experiment, cells were seeded and cultured at 1.0 × 10⁴ cells/well in RPMI1640 medium containing 10% FBS in 96-well cell culture plates at 37 °C and 5% CO₂. After overnight cultivation, 10 μL/well of transfection reagent containing each concentration of mRNA and 0.3% Lipofectamine MessengerMAX Transfection Reagent diluted with Opti-MEM were added to 40 μL/well of the cultured cells. After 5 h of transfection, the culture supernatant was removed, replaced with RPMI1640 medium containing 10% FBS, and the cells were cultivated. Cultured cells were then washed with ice-cold D-PBS(−) and lysed with 20 μL/well iScript RT-qPCR Sample Preparation Reagent (Bio-Rad) containing 2% Protease Inhibitor Cocktail (EDTA free) (Nacalai Tesque). For the FLAG-EGF mRNA experiment, 0.45% Lipofectamine MessengerMAX was used. Electroporation of mRNAs was performed using the Nucleofector 96-well Shuttle System (Lonza) and the SE Cell Line 96-well Nucleofector Kit (Lonza). Twenty microliters of reaction mixture containing each concentration of mRNA, 2.0 × 10⁵ HeLa cells, Nucleofector Solution and Supplement 1 was pulsed with FF-150 condition, and then seeded and cultivated at 5.0 × 10⁴ cells/well. Cell lysis was performed under the same conditions as the lipofection experiment.

In vitro transfection of mRNAs for measurement of translation activity using cultured human primary aortic smooth muscle cells

Electroporation of uncapped human primary aortic smooth muscle cells (hAoSMC) (Lonza) was performed using the Nucleofector 96-well Shuttle System and P1 Primary Cell 96-well Nucleofector Kit (Lonza). hAoSMC were pulsed with FF-130 condition. After electroporation, 2.0 × 10⁴ cells/well were seeded and cultivated with Smooth Muscle Cell Growth Medium 2 (Lonza) and lysed as described above. The translational activity of each mRNA was evaluated as described in the section on assays using HeLa cells.

Sandwich-ELISA for the evaluation of the translation activity

FLAG-His tag peptides in the translation reaction were measured using sandwich ELISA. The following standard peptides were purchased from COSMO-BIO: 1×FLAG-His₆: NH₂-MDYKDDDDKGGHHHHHH-COOH and 3×FLAG-His₆: NH₂-MDYKDDDDKIIDYKDDDDKGGDYKDDDDKHHHHHH-COOH. Maxisorp 96-well plate (Sigma-Aldrich) was coated with 3 μg/mL of Anti-His-Tag Antibody (Proteintech) in 0.1 mol/L Carbonate buffer (pH 9.4) and then blocked with 3% BSA-TBST. The plate was washed with 1×TBST, and then each translation reaction and standard peptides in 3% BSA-TBST were added. After 1 h of incubation, a washed plate was added with 0.01 % Monoclonal ANTI-FLAG M2-Peroxidase (HRP) Ab produced in mice (Thermo Fisher Scientific) in 3% BSA-TBST. After washing, the plate was developed with 1-Step Ultra TMB-ELISA and then stopped with 0.5 mol/L sulfuric acid. Abs. 450 subtracted by abs. 570 was measured using a spectrophotometer (Bio-Rad).

Evaluation of immune stimulation of mRNA in vitro

To evaluate immune stimulation by each mRNA transfected into in vitro cultured cells, cDNA was prepared with the SuperPrep Cell Lysis & RT Kit for qPCR (Toyobo) from cultured cells 5 h after the transfection of mRNAs. The procedure was based on the manufacturer’s protocol. The gene expression of IL6 (Hs00174131_m1), TNF (Hs00174128_m1), and IL1B (Hs01555410_m1) was evaluated by RT-qPCR using TaqMan probes and TaqMan Gene Expression Master mix (Applied Biosystems) and normalized by GAPDH (Hs02786624_g1) expression.

Cell viability assay

The viability of HeLa cells at 24 h after lipofection with 30 nmol/L chemically modified mRNA was evaluated with the CellTiter-Glo Luminescent Cell Viability Assay (Promega) based on the manufacturer’s protocol.

Evaluation of mRNA stability in vitro

The concentration of the remaining mRNA in the translation reaction was measured using RT-qPCR. The sequences and structures of the primers (Sigma-Aldrich) and TaqMan probes (Applied Biosystems) are described below. RT primer: 5’ -TCAGTGGTGGTGGTGGTGGTGTTTG −3’, FW primer: 5’ -GAATACAAGCTACTTGTTCTTTT −3’, RV primer: 5’ -ATCTTGTCGTCGTCGTCCTT −3’, FAM-MGB probe: 5’-FAM-CAGCCACCATG-NFQ-MGB-3’. Reverse transcription of mRNA was performed using the iScript Select cDNA Synthesis Kit (Bio-Rad) according to the manufacturer’s protocol. The reaction mixture contained 0.2 μmol/L RT primer, and the translation reaction was diluted with RNase/DNase-free water containing 0.2 U/μL Recombinant RNase Inhibitor. RT-qPCR of the obtained cDNAs was performed with Taqman qPCR Master Mix (Applied Biosystems) and QuantStudio 12 K Flex Real-Time PCR System (Applied Biosystems), as described below. Twenty microliters of reaction mixture containing 4 μL of cDNA, 1.65 μmol/L FW primer, 1.40 μmol/L RV primer, 0.10 μmol/L FAM-MGB probe and 1×reaction master mix were subjected to the thermal cycling reaction as follows: 50 °C for 2 min, 95 °C for 10 min, (95 °C, 15 s, 60 °C, 1 min) × 50 cycles.

Serum stability assay

The stability of each mRNA in mouse serum was measured as described below. Ten microliters of reaction mixture containing 1.6% mouse serum and 1 μmol/L of each mRNA in THE RNA Storage Solution (Ambion) was incubated at 37 °C for 0, 15, 30, and 60 min, and then 1.2 U/μL of Recombinant RNase inhibitor was added to stop the degradation.

The remaining mRNAs in the reaction mixture were measured using RT-qPCR, as described in the section of “Evaluation of mRNA stability in vitro”.

dsRNA ELISA

dsRNA content in each mRNA sample was detected with a Double-stranded RNA (dsRNA) ELISA kit (K1 based) (Exalpha). Poly(I:C) HMW (InvivoGen) was used as the standard dsRNA. The general procedure was based on the manufacturer’s protocol.

LC-MS analysis of translated peptides

Chemically synthesized mRNAs were translated using a rabbit reticulocyte in vitro translation system (Promega) at 37 °C for 30 min. The translated peptides were purified with an anti-FLAG antibody (Millipore). LC-MS analysis of the peptides and synthetic standards (NH₂-MDYKDDDDKIIDYKDDDDKGGDYKDDDDKHHHHHH-COOH) was performed by Medical ProteoScope (Kanagawa, Japan). Peptides were resuspended in [TFA]:[ACN]:[H2O] = 0.1:2:98 (%vol).

Analysis was conducted using an Ultimate 3000 RSLCnano LC system coupled to a Q Exactive Orbitrap MS (Thermo Fisher Scientific). Chromatography used a Nano HPLC Capillary Column (3.0 µm, 75 µm × 15 cm) with buffers A (0.1% formic acid) and B (0.1% formic acid, 90% acetonitrile). The gradient ran from 5% to 95% B over 50 min at 350 nL/min. Water injections served as blanks.

Mass spectrometry was performed in positive ion mode with a capillary temperature of 250 °C, full scan range m/z 310–1500, and resolution 70,000 (MS) and 17,500 (MS/MS). Lock mass correction used m/z 391.28429 and 445.12003. Data were reviewed with Xcalibur Qual Browser (v4.2.47). Extracted ion chromatograms targeted m/z 859.89–860.89 (pentavalent ion).

MS/MS data were analyzed with Mascot (Matrix Science) using standard settings; however, MS/MS spectra are not shown. Peptide identifications were confirmed by high-resolution LC-MS and comparison with synthetic standards. Technical replicates ensured reproducibility.

Capping ratio analysis

The solution containing RNA (0.5 μmol/L, as final conc.), DNAzyme (1.0 μmol/L), Tris-HCl (pH 8.0, 5 mmol/L), MgCl₂ (5 mmol/L) in MQ was prepared and the mixture was incubated at 37 °C for 1 h. DNAzyme sequence: CTGCAAAAAGAAGGCTAGCTACAACGAAAGTAGCTTG (Eurofins Genomics). After the reaction, iPrOH precipitation was performed and the sample was subjected to dPAGE analysis.

Statistical analysis

No statistical method was used to determine the sample size. No data were excluded from the analysis. The experiments were not randomized. The Investigators were not blinded to the allocation during the experiments and outcome assessment. Statistical analyses were performed using GraphPad Prism9 ver. 9.3.1 (GraphPad Software, USA). The methods used are described in each figure.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

All data generated or analyzed in this study are included in this published article, in the supplementary information file, or in the source data files, and are available from the corresponding author(s) upon request. Source data are provided with this paper.

References

Verbeke, R., Lentacker, I., De Smedt, S. C. & Dewitte, H. The dawn of mRNA vaccines: the COVID-19 case. J. Control. Release 333, 511–520 (2021).
Article CAS PubMed PubMed Central Google Scholar
Fang, E. et al. Advances in COVID-19 mRNA vaccine development. Signal Transduct. Target. Ther. 7, 94 (2022).
Article CAS PubMed PubMed Central Google Scholar
Rohner, E., Yang, R., Foo, K. S., Goedel, A. & Chien, K. R. Unlocking the promise of mRNA therapeutics. Nat. Biotechnol. 40, 1586–1600 (2022).
Article CAS PubMed Google Scholar
Şahin, U., Karikó, K. & Türeci, Ö mRNA-based therapeutics—developing a new class of drugs. Nat. Rev. Drug Discov. 13, 759–780 (2014).
Article PubMed Google Scholar
Chaudhary, N., Weissman, D. & Whitehead, K. A. mRNA vaccines for infectious diseases: principles, delivery and clinical translation. Nat. Rev. Drug Discov. 20, 817–838 (2021).
Article CAS PubMed PubMed Central Google Scholar
Liu, C. et al. mRNA-based cancer therapeutics. Nat. Rev. Cancer 23, 526–543 (2023).
Article PubMed Google Scholar
Imani, S., Tagit, O. & Pichon, C. Neoantigen vaccine nanoformulations based on chemically synthesized minimal mRNA (CmRNA): small molecules, big impact. npj Vaccines 9, 14 (2024).
Article CAS PubMed PubMed Central Google Scholar
Papukashvili, D. et al. Self-amplifying RNA approach for protein replacement therapy. Int. J. Mol. Sci. 23, 12884 (2022).
Article CAS PubMed PubMed Central Google Scholar
Qin, S. et al. mRNA-based therapeutics: powerful and versatile tools to combat diseases. Signal Transduct. Target. Ther. 7, 166 (2022).
Article CAS PubMed PubMed Central Google Scholar
Vavilis, T. et al. mRNA in the context of protein replacement therapy. Pharmaceutics 15, 166 (2023).
Article CAS PubMed PubMed Central Google Scholar
Oh, S. N. & Kessler, J. A. Design, assembly, production, and transfection of synthetic modified mRNA. Methods 133, 29–43 (2017).
Article PubMed PubMed Central Google Scholar
Elkhalifa, D., Rayan, M., Negmeldin, A. T., Elhissi, A. M. A. & Khalil, A. Chemically modified mRNA beyond COVID-19: potential preventive and therapeutic applications for targeting chronic diseases. Biomed. Pharmacother. 145, 112385–112385 (2021).
Article PubMed PubMed Central Google Scholar
De La Vega, R. E. et al. Efficient healing of large osseous segmental defects using optimized chemically modified messenger RNA encoding BMP-2. Sci. Adv. 8, eabl6242 (2022).
Article ADS PubMed Google Scholar
Muttach, F., Muthmann, N. & Rentmeister, A. Synthetic mRNA capping. Beilstein J. Org. Chem. 13, 2819–2832 (2017).
Article CAS PubMed PubMed Central Google Scholar
Shanmugasundaram, M., Senthilvelan, A. & Kore, A. R. Recent advances in modified cap analogs: synthesis, biochemical properties, and mRNA based vaccines. Chem. Rec. 22, e202200005 (2022).
Article CAS PubMed PubMed Central Google Scholar
Kore, A. R., Shanmugasundaram, M., Charles, I., Vlassov, A. V. & Barta, T. J. Locked nucleic acid (LNA)-modified dinucleotide mRNA cap analogue: synthesis, enzymatic incorporation, and utilization. J. Am. Chem. Soc. 131, 6364–6365 (2009).
Article ADS CAS PubMed Google Scholar
Wojtczak, B. A. et al. 5’-phosphorothiolate dinucleotide cap analogues: reagents for messenger RNA modification and potent small-molecular inhibitors of decapping enzymes. J. Am. Chem. Soc. 140, 5987–5999 (2018).
Article ADS CAS PubMed Google Scholar
Warmiński, M. et al. Structural insights into the interaction of clinically relevant phosphorothioate mRNA cap analogs with translation initiation factor 4E reveal stabilization via electrostatic thio-effect. ACS Chem. Biol. 16, 334–343 (2021).
Article PubMed PubMed Central Google Scholar
Kozarski, M. et al. Towards superior mRNA caps accessible by click chemistry: synthesis and translational properties of triazole-bearing oligonucleotide cap analogs. RSC Adv. 13, 12809–12824 (2023).
Article ADS CAS PubMed PubMed Central Google Scholar
Ogasawara, S. Duration control of protein expression in vivo by light-mediated reversible activation of translation. ACS Chem. Biol. 12, 351–356 (2017).
Article CAS PubMed Google Scholar
Bollu, A., Peters, A. & Rentmeister, A. Chemo-enzymatic modification of the 5′ cap to study mRNAs. Acc. Chem. Res. 55, 1249–1261 (2022).
Article CAS PubMed Google Scholar
Klöcker, N. et al. Photocaged 5′ cap analogues for optical control of mRNA translation in cells. Nat. Chem. 14, 905–913 (2022).
Article PubMed PubMed Central Google Scholar
Strzelecka, D. et al. Phosphodiester modifications in mRNA poly(A) tail prevent deadenylation without compromising protein expression. RNA 26, 1815–1837 (2020).
Article CAS PubMed PubMed Central Google Scholar
Aditham, A. et al. Chemically modified mocRNAs for highly efficient protein expression in mammalian cells. ACS Chem. Biol. 17, 3352–3366 (2022).
Article CAS PubMed Google Scholar
Li, C. Y. et al. Cytidine-containing tails robustly enhance and prolong protein production of synthetic mRNA in cell and in vivo. Mol. Ther. Nucleic Acids 30, 300–310 (2022).
Article CAS PubMed PubMed Central Google Scholar
Kawaguchi, D. et al. Phosphorothioate modification of mRNA accelerates the rate of translation initiation to provide more efficient protein synthesis. Angew. Chem. Int. Ed. 59, 17403–17407 (2020).
Article CAS Google Scholar
Liu, A. & Wang, X. The pivotal role of chemical modifications in mRNA therapeutics. Front. Cell Dev. Biol. 10, 901510 (2022).
Article PubMed PubMed Central Google Scholar
Roy, B. Effects of mRNA modifications on translation: an overview. Methods Mol. Biol. 2298, 327–356 (2021).
Article CAS PubMed Google Scholar
Song, J. & Yi, C. Chemical modifications to RNA: a new layer of gene expression regulation. ACS Chem. Biol. 12, 316–325 (2017).
Article CAS PubMed Google Scholar
Li, B., Luo, X. & Dong, Y. Effects of chemically modified messenger RNA on protein expression. Bioconjugate Chem. 27, 849–853 (2016).
Article CAS Google Scholar
Anderson, B. R. et al. Incorporation of pseudouridine into mRNA enhances translation by diminishing PKR activation. Nucleic Acids Res. 38, 5884–5892 (2010).
Article CAS PubMed PubMed Central Google Scholar
Freund, I., Eigenbrod, T., Helm, M. & Dalpke, A. H. RNA modifications modulate activation of innate toll-like receptors. Genes. 10, 92 (2019).
Article CAS PubMed PubMed Central Google Scholar
Karikó, K., Buckstein, M., Ni, H. & Weissman, D. Suppression of RNA recognition by Toll-like receptors: the impact of nucleoside modification and the evolutionary origin of RNA. Immunity 23, 165–175 (2005).
Article PubMed Google Scholar
Karikó, K. et al. Incorporation of pseudouridine into mRNA yields superior nonimmunogenic vector with increased translational capacity and biological stability. Mol. Ther. 16, 1833–1840 (2008).
Article PubMed Google Scholar
Karikó, K. & Weissman, D. Naturally occurring nucleoside modifications suppress the immunostimulatory activity of RNA: implication for therapeutic RNA development. Curr. Opin. Drug Discov. Dev. 10, 523–532 (2007).
Google Scholar
Glazier, D. A. et al. Chemical synthesis and biological application of modified oligonucleotides. Bioconjugate Chem. 31, 1213–1233 (2020).
Article CAS Google Scholar
McKenzie, L. K., El-Khoury, R., Thorpe, J. D., Damha, M. J. & Hollenstein, M. Recent progress in non-native nucleic acid modifications. Chem. Soc. Rev. 50, 5126–5164 (2021).
Article CAS PubMed Google Scholar
Egli, M. & Manoharan, M. Chemistry, structure and function of approved oligonucleotide therapeutics. Nucleic Acids Res. 51, 2529–2573 (2023).
Article CAS PubMed PubMed Central Google Scholar
Wan, W. B. & Seth, P. P. The medicinal chemistry of therapeutic oligonucleotides. J. Med. Chem. 59, 9645–9667 (2016).
Article CAS PubMed Google Scholar
Aurup, H., Siebert, A., Benseler, F., Williams, D. & Eckstein, F. Translation of 2′-modified mRNA in vitro and in vivo. Nucleic Acids Res. 22, 4963–4968 (1994).
Article CAS PubMed PubMed Central Google Scholar
Choi, J. et al. 2′-O-methylation in mRNA disrupts tRNA decoding during translation elongation. Nat. Struct. Mol. Biol. 25, 208–216 (2018).
Article CAS PubMed PubMed Central Google Scholar
Padilla, R. & Sousa, R. Efficient synthesis of nucleic acids heavily modified with non-canonical ribose 2’-groups using a mutantT7 RNA polymerase (RNAP). Nucleic Acids Res. 27, 1561–1563 (1999).
Article CAS PubMed PubMed Central Google Scholar
Meyer, A. J. et al. Transcription yield of fully 2′-modified RNA can be increased by the addition of thermostabilizing mutations to T7 RNA polymerase mutants. Nucleic Acids Res. 43, 7480–7488 (2015).
Article CAS PubMed PubMed Central Google Scholar
Houlihan, G., Arangundy-Franklin, S. & Holliger, P. Engineering and application of polymerases for synthetic genetics. Curr. Opin. Biotechnol. 48, 168–179 (2017).
Article CAS PubMed Google Scholar
Ohashi, S., Hashiya, F. & Abe, H. Variety of nucleotide polymerase mutants aiming to synthesize modified RNA. ChemBioChem 22, 2398–2406 (2021).
Article CAS PubMed Google Scholar
Wang, G. et al. Thermophilic nucleic acid polymerases and their application in xenobiology. Int. J. Mol. Sci. 23, 14969 (2022).
Article CAS PubMed PubMed Central Google Scholar
Abe, N. et al. Rolling circle amplification in a prokaryotic translation system using small circular RNA. Angew. Chem. Int. Ed. 52, 7004–7008 (2013).
Article CAS Google Scholar
Abe, N. et al. Rolling circle translation of circular RNA in living human cells. Sci. Rep. 5, 16435 (2015).
Article ADS CAS PubMed PubMed Central Google Scholar
Abe, N., Kodama, A. & Abe, H. Preparation of circular RNA in vitro. Methods Mol. Biol. 1724, 181–192 (2018).
Article CAS PubMed Google Scholar
Nakamoto, K. & Abe, H. Chemical synthesis of circular RNAs with phosphoramidate linkages for rolling-circle translation. Curr. Protoc. 1, e43 (2021).
Article CAS PubMed Google Scholar
Inagaki, M. et al. Cap analogs with a hydrophobic photocleavable tag enable facile purification of fully capped mRNA with various cap structures. Nat. Commun. 14, 2657 (2023).
Article ADS CAS PubMed PubMed Central Google Scholar
Abe, N. et al. Complete chemical synthesis of minimal messenger RNA by efficient chemical capping reaction. ACS Chem. Biol. 17, 1308–1314 (2022).
Article CAS PubMed Google Scholar
Kozak, M. Features in the 5′ non-coding sequences of rabbit α and β-globin mRNAs that affect translational efficiency. J. Mol. Biol. 235, 95–110 (1994).
Article CAS PubMed Google Scholar
Nandakumar, J. et al. How an RNA Ligase Discriminates RNA versus DNA Damage. Mol. Cell. 16, 211–221 (2004).
Article CAS PubMed Google Scholar
Chen, H. et al. Template-directed chemical ligation to obtain 3′-3′ and 5′-5′ phosphodiester DNA linkages. Sci. Rep. 4, 4595 (2014).
Article ADS PubMed PubMed Central Google Scholar
Bowler, F. R. et al. Prebiotically plausible oligoribonucleotide ligation facilitated by chemoselective acetylation. Nat. Chem. 5, 383–389 (2013).
Article CAS PubMed PubMed Central Google Scholar
Khanam, T., Sondekoppa, M. R., Avak, K., Nahum, S. & Brosius, J. Poly(A)-binding protein binds to a-rich sequences via RNA-binding domains 1+2 and 3+4. RNA Biol. 3, 170–177 (2006).
Article CAS PubMed Google Scholar
Perzanowska, O., Smietanski, M., Jemielity, J. & Kowalska, J. Chemically modified Poly(A) analogs targeting PABP: structure activity relationship and translation inhibitory properties. Chem. A Eur. J. 28, e202201115 (2022).
Article CAS Google Scholar
Yang, C.-H., Huang, Y.-B., Wu, P.-C. & Tsai, Y.-H. The evaluation of stability of recombinant human epidermal growth factor in burn-injured pigs. Process Biochem. 40, 1661–1665 (2005).
Article CAS Google Scholar
Hou, X., Zaks, T., Langer, R. & Dong, Y. Lipid nanoparticles for mRNA delivery. Nat. Rev. Mater. 6, 1078–1094 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Jeong, M., Lee, Y., Park, J., Jung, H. & Lee, H. Lipid nanoparticles (LNPs) for in vivo RNA delivery and their breakthrough technology for future applications. Adv. Drug Deliv. Rev. 200, 114990 (2023).
Article CAS PubMed Google Scholar
Xu, X. & Xia, T. Recent advances in site-specific lipid nanoparticles for mRNA delivery. ACS Nanosci. Au 3, 192–203 (2023).
Article CAS PubMed PubMed Central Google Scholar
Kim, H. et al. Hyaluronate—epidermal growth factor conjugate for skin wound healing and regeneration. Biomacromolecules 17, 3694–3705 (2016).
Article CAS PubMed Google Scholar
Meyer, K. D. et al. 5′ UTR m6A promotes cap-independent translation. Cell 163, 999–1010 (2015).
Article CAS PubMed PubMed Central Google Scholar
Shatsky, I. N., Terenin, I. M., Smirnova, V. V. & Andreev, D. E. Cap-independent translation: What’s in a name?. Trends Biochem. Sci. 43, 882–895 (2018).
Article CAS PubMed Google Scholar
Trainor, B. M., Ghosh, A., Pestov, D. G., Hellen, C. U. T. & Shcherbik, N. A translation enhancer element from black beetle virus engages yeast eIF4G1 to drive cap-independent translation initiation. Sci. Rep. 11, 2461 (2021).
Article CAS PubMed PubMed Central Google Scholar

Download references

Acknowledgements

This work was supported by the Japan Society for the Promotion of Science (JSPS), International Leading Research (JP22K21346 to Y.K.), and Japan Agency for Medical Research and Development (AMED) (JP22gm0010008, JP22fk0310506 to HA, JP23fk0210133 to Y.K.). This work was funded by Kyowa Kirin Co., Ltd.

Author information

These authors contributed equally: Hiroto Iwai, Yasuaki Kimura.

Authors and Affiliations

Research Unit, Research Division, Kyowa Kirin Co., Ltd., Tokyo, Japan
Hiroto Iwai, Masakazu Honma, Atsushi Hashimoto, Keiichi Motosawa, Takayuki Atago, Kana Asano, Keiko Kobayashi, Hiroki Yamada, Keiko Hiraishi, Seiji Saito & Junichiro Yamamoto
Department of Chemistry, Graduate School of Science, Nagoya University, Nagoya, Japan
Yasuaki Kimura, Kosuke Nakamoto, Naoko Abe, Ryoko Ogisu & Hiroshi Abe
Research Center for Materials Science, Nagoya University, Nagoya, Japan
Fumitaka Hashiya & Hiroshi Abe
CREST, Japan Science and Technology Agency, Tokyo, Japan
Hiroshi Abe
Institute for Glyco-core Research (iGCORE), Nagoya University, Nagoya, Aichi, Japan
Hiroshi Abe

Authors

Hiroto Iwai
View author publications
Search author on:PubMed Google Scholar
Yasuaki Kimura
View author publications
Search author on:PubMed Google Scholar
Masakazu Honma
View author publications
Search author on:PubMed Google Scholar
Kosuke Nakamoto
View author publications
Search author on:PubMed Google Scholar
Atsushi Hashimoto
View author publications
Search author on:PubMed Google Scholar
Keiichi Motosawa
View author publications
Search author on:PubMed Google Scholar
Takayuki Atago
View author publications
Search author on:PubMed Google Scholar
Kana Asano
View author publications
Search author on:PubMed Google Scholar
Fumitaka Hashiya
View author publications
Search author on:PubMed Google Scholar
Naoko Abe
View author publications
Search author on:PubMed Google Scholar
Keiko Kobayashi
View author publications
Search author on:PubMed Google Scholar
Ryoko Ogisu
View author publications
Search author on:PubMed Google Scholar
Hiroki Yamada
View author publications
Search author on:PubMed Google Scholar
Keiko Hiraishi
View author publications
Search author on:PubMed Google Scholar
Seiji Saito
View author publications
Search author on:PubMed Google Scholar
Junichiro Yamamoto
View author publications
Search author on:PubMed Google Scholar
Hiroshi Abe
View author publications
Search author on:PubMed Google Scholar

Contributions

H.I., Y.K., M.H., J.Y., and H.A. conceptualized and designed the study. M.H., K.N., K.M., T.A., K.K., R.O., and H.Y. performed chemical synthesis. H.I., K.N., A.H., K.A., N.A., and F.H. designed and performed the biochemical experiments, including cell experiments. A.H., K.H. and S.S. performed LC-MS analysis of the translated peptides, and H.I., Y.K., M.H., A.H., J.Y., and H.A. wrote the manuscript.

Corresponding authors

Correspondence to Yasuaki Kimura, Junichiro Yamamoto or Hiroshi Abe.

Ethics declarations

Competing interests

Competing Interest Statement: H.A. is a cofounder of Crafton Biotechnology. The company focuses on the development of mRNA therapeutics. H.I., M.H., K.M., T.A., K.A., K.K., H.Y., A.H., K.H., S.S., and J.Y. are employees of Kyowa Kirin Co. Ltd. International patents application covering part of this work has been filed by Nagoya University and Kyowa Kirin Co., Ltd. The remaining authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks anonymous reviewers for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information (download PDF )

Reporting Summary (download PDF )

Transparent Peer Review file (download PDF )

Source data

Source data (download XLSX )

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Iwai, H., Kimura, Y., Honma, M. et al. Position-specific ORF nucleoside-ribose modifications enabled by complete chemical synthesis enhance mRNA stability and translation. Nat Commun 16, 9995 (2025). https://doi.org/10.1038/s41467-025-65788-8

Download citation

Received: 01 May 2024
Accepted: 23 October 2025
Published: 19 November 2025
Version of record: 19 November 2025
DOI: https://doi.org/10.1038/s41467-025-65788-8