Evaluation of synthetic mRNA with selected UTR sequences and alternative poly(A) tail, in vitro and in vivo

UTRs and the poly(A) tail play a crucial role in regulating mRNA intracellular kinetics.

“Untranslated regions (UTRs) and the poly(A) tail play a crucial role in regulating mRNA intracellular kinetics.”

• cites [4] Kang, 2023, Advancements of in vitro transcribed mRNA (IVT mRNA) to enable translation into the clinics, Adv. Drug Deliv. Rev. sourceSupported

The presence of the poly(A) tail on the plasmid template for mRNA production is not stable and prone to shortening in bacteria during plasmid DNA amplification.

“However, the presence of the poly(A) tail on the plasmid template for mRNA production is not stable and prone to shortening in bacteria during plasmid DNA amplification.17”

• cites [17] Grier, 2016, pEVL: A Linear Plasmid for Generating mRNA IVT Templates With Extended Encoded Poly(A) Sequences, Mol. Ther. Nucleic Acids sourcePartly supported

  The Results section (‘Homopolymeric poly(A) tracts 175 bp or longer are extremely unstable in circular plasmids’; page not provided) reports shortening after bacterial transformation of circular plasmids carrying long poly(A) tracts. The claim omits important limits: 70-bp tracts remained stable, and another Results section reports that poly(A) tracts in the linear pEVL plasmid were highly stable during extended propagation, except that the ~500-bp tract tended to shorten. (LLM, conf=0.88)

  Experimental — accuracy not yet measured

The translation initiation efficiency of mRNA is affected by the sequence and structure of the 5′ UTR region.

“The translation initiation efficiency of mRNA is affected by the sequence and structure of the 5′ UTR region.5”

• cites [5] Hinnebusch, 2016, Translational control by 5′-untranslated regions of eukaryotic mRNAs, Science sourceSupported

The synthetic 5′ UTR UTR4 allows strong expression of mRNA in cells and Zebrafish.

“the synthetic 5′ UTR UTR4 allowing strong expression of mRNA in cells and Zebrafish27”

• cites [27] Linares-Fernández, 2021, Combining an optimized mRNA template with a double purification process allows strong expression of in vitro transcribed mRNA, Mol. Ther. Nucleic Acids sourceSupported

Incorporation of non-A nucleotides, particularly guanosine, can interfere with the deadenylation process, thereby enhancing mRNA stability.

“Previous studies have shown that the incorporation of non-A nucleotides, particularly guanosine, can interfere with the deadenylation process, thereby enhancing mRNA stability.41”

• cites [41] Lee, 2024, Deadenylation kinetics of mixed poly(A) tails at single-nucleotide resolution, Nat. Struct. Mol. Biol. sourceSupported

Shortening of the poly(A) sequence directly affects the translation efficiency of the mRNA.

“This instability, characterized by shortening of the poly(A) sequence, directly affects the translation efficiency of the mRNA.17”

• cites [17] Grier, 2016, pEVL: A Linear Plasmid for Generating mRNA IVT Templates With Extended Encoded Poly(A) Sequences, Mol. Ther. Nucleic Acids sourceSupported

The finding that cap-independent 5′ UTRs have poor translation rate is consistent with previous studies showing IRES are more efficient under stress conditions.

“Our results demonstrated that cap-independent 5′ UTRs exhibited a poor translation rate across three different cell lines compared to cap-dependent 5′ UTRs, which is consistent with previous studies showing that IRES are more efficient under stress conditions.42,43”

• cites [42] Weingarten-Gabbay, 2016, Systematic discovery of cap-independent translation sequences in human and viral genomes, Science sourceNot checked — no full text of the source
• cites [43] Gebauer, 2016, IRES unplugged, Science sourceSupported

The synthetic 5′ UTR neoUTR3 shows enhanced protein production.

“the synthetic 5′ UTR neoUTR3 showing enhanced protein production (UTR3).9”

• cites [9] Cao, 2021, High-throughput 5’ UTR engineering for enhanced protein production in non-viral gene therapies, Nat. Commun. sourceSupported

SRCD analysis showed that UTR3-AES Fluc mRNA had a different profile compared to UTR3-VP6 Fluc mRNA.

“Additionally, SRCD analysis showed that UTR3-AES Fluc mRNA had a different profile compared to UTR3-VP6 Fluc mRNA”

In BALB/c mice two weeks after the first injection, mRNA with AES-mtRNR1 or VP6 3′ UTRs induced higher production of IgG anti-RBD spike protein compared to SOD 3′ UTR mRNA.

“Two weeks after the first injection, mRNA with AES-mtRNR1 or VP6 3′ UTRs induced higher production of IgG anti-RBD spike protein compared to the SOD 3′ UTR mRNA (Figure 7F).”

In HeLa cells, A30L70-mRNA showed progressively higher bioluminescence than A/G-mRNA from 24 to 96 h, reaching up to 2-fold greater at 96 h.

“However, from 24 to 96 h, A30L70-mRNA exhibited a progressively higher bioluminescence signal, reaching up to 2-fold greater intensity at 96 h (Figure 2A).”

In HeLa and AB1079 cells, VP6 and MnSOD 3′ UTRs produced more spike protein than the AES-mtRNR1 3′ UTR.

“The VP6 and MnSOD 3′ UTRs produced more spike protein than the AES-mtRNR1 3′ UTR (Figures 7B and 7C)”

One week after the second injection (boost), no significant differences in IgG production were observed among the three 3′ UTRs.

“Interestingly, 1 week after the second injection (boost), no significant differences in IgG production were observed among the three 3′ UTRs (Figure 7G).”

FISH-Flow showed comparable Nluc mRNA levels at 2 and 24 h after transfection for hAg-AES, hAg-VP6, and hAg-SOD mRNAs.

“FISH-Flow results showed comparable Nluc mRNA levels at 2 and 24 h after transfection for hAg-AES, hAg-VP6, and hAg-SOD mRNAs (Figure S5A).”

Translation of all constructs except CrPV was inhibited by 90% using the eIF4G initiation pathway inhibitor 4EGI-1.

“Moreover, the translation of all constructs except CrPV was inhibited by 90% using the canonical eIF4G initiation pathway inhibitor (4EGI-1).”

A30L70-tailed mRNA had a higher melting temperature (76.6 ± 1.3°C) than A120-tailed (69.4 ± 1.1°C) and A/G-tailed (69.8 ± 1.4°C) mRNA.

“A30L70-tailed mRNA had a higher Tm (76.6 ± 1.3°C) compared to A120 (69.4°C ± 1.1°C) and A/G-tailed mRNA (69.8°C ± 1.4°C).”

In cell-free translation assays, the nine mRNAs were similarly translated.

“Furthermore, in cell-free translation assays, the nine mRNAs were similarly translated (Figure S3A)”

mRNAs with CleanCap AG yielded two times more protein than mRNAs with the ARCA cap analog.

“However, as expected, mRNAs with CleanCap AG yielded two times more protein than mRNAs with the ARCA cap analog (Figure S4).”

In HeLa cells, A/G-mRNA produced 25% less total protein over time (AUC) than A30L70-mRNA.

“The total amount of protein produced was quantified using area under the curve (AUC) analysis, which showed that A/G-mRNA produced 25% less protein overall compared to A30L70-mRNA (Figure 2B).”

In HeLa cells, transfection with Fluc mRNA containing UTR3-AES resulted in lower protein production than the benchmark hAg-AES.

“In line with the in vivo findings, the transfection of HeLa cells with Fluc mRNA containing UTR3-AES resulted in lower protein production than the benchmark mRNA (hAg-AES) (Figure S6).”

In BALB/c mice after intramuscular injection, A/G-tailed and A30L70-tailed Fluc mRNAs exhibited comparable bioluminescence over 72 h.

“As illustrated in Figure 2C, mRNAs with both tails exhibited comparable bioluminescence expression over the 72-h period post-administration.”

mRNAs containing IRES 5′ UTRs (VCIP, CrPV, EMCV) translated less than 20% of the luciferase expression achieved with cap-dependent human alpha-globin 5′ UTR across HeLa, C2C12, and DC2.4 cells.

“The results showed that the translation of mRNAs containing IRES 5′ UTRs derived from mammalian or viral sources (VCIP, CrPV, EMCV) was less than 20% of the luciferase expression achieved with the cap-dependent human alpha-globin 5′ UTR across the three cell lines tested (Figures 3B–3D).”

The melting temperatures of the 3′ UTRs were equivalent: 76.6 ± 1.4°C for AES-mtRNR1, 76.4 ± 0.7°C for VP6, and 69.7 ± 5.7°C for SOD.

“Tm of the 3′ UTRs were equivalent: 76.6°C ± 1.4°C for AES-mtRNR1, 76.4°C ± 0.7°C for VP6, and 69.7°C ± 5.7°C for SOD.”

mRNAs containing hAg, UTR3, or UTR4 5′ UTRs had comparably high Nluc expression across HeLa, C2C12, and DC2.4 cells.

“We also observed comparably high Nluc expression levels with mRNAs containing hAg, UTR3, or UTR4 5′ UTRs across all three cell lines tested.”

Modified (m1Ψ) Spike mRNA yielded 10 times more protein than unmodified Spike mRNA in HeLa cells.

“The results showed that the modified mRNA yielded 10 times more than the unmodified mRNA (Figure S5B).”

In C2C12 cells, hAg-VP6 and hAg-SOD combinations produced significantly higher protein than the benchmark hAg-AES.

“Noteworthy, in C2C12 cells, hAg-VP6 and hAg-SOD combinations produced significantly a higher amount of protein than the benchmark (Figure 4D).”

UTR4 had a drastically lower melting temperature (59.6 ± 1.6°C) compared to hAg (79.2 ± 0.9°C) and UTR3 (76.3 ± 0.9°C).

“However, UTR4 had a drastically lower Tm (59.6°C ± 1.6°C) compared to hAg (79.2°C ± 0.9°C) and UTR3 (76.3°C ± 0.9°C).”

When paired with 3′ UTR SOD, UTR4 yielded 2-fold less protein compared to the reference.

“Despite the 5′ UTR UTR4 being identified in high ribosome load analysis,27 when paired with 3′ UTR SOD, it yielded 2-fold less protein compared to the reference (Figure 4).”

The use of m1Ψ led to decreased dsRNA levels.

“Interestingly, the use of m1Ψ led to decreased dsRNA levels, as did the use of CleanCap AG compared to ARCA cap analog (Table S3).”

• AssumptionRests on an assumption: The parameters of the IVT and the choice of downstream purification are crucial for improved translation efficiency.

CleanCap AG yielded up to 2-fold more expression compared to ARCA in DC2.4 and MoDCs.

“CleanCap AG yielded up to 2-fold more expression compared to ARCA in DC2.4 and MoDCs (Figure S4), likely due to higher capping efficiency (up to 98%) and lower dsRNA levels.”

• AssumptionRests on an assumption: The parameters of the IVT and the choice of downstream purification are crucial for improved translation efficiency.

The A/G tail demonstrated superior stability compared to BioNTech's A30L70 tail in bacterial systems.

“This A/G tail demonstrated superior stability compared to BioNTech’s A30L70 tail in bacterial systems while maintaining comparable expression levels of reporter gene in cellulo and in vivo.”

The use of CleanCap AG compared to ARCA cap analog led to decreased dsRNA levels.

“Interestingly, the use of m1Ψ led to decreased dsRNA levels, as did the use of CleanCap AG compared to ARCA cap analog (Table S3).”

• AssumptionRests on an assumption: The parameters of the IVT and the choice of downstream purification are crucial for improved translation efficiency.

The synthetic 5′ UTR UTR4 yielded different results depending on the paired 3′ UTRs, especially in MoDCs.

“However, the synthetic 5′ UTR UTR4 yielded different results depending on the paired 3′ UTRs (Figure 4), especially in MoDCs.”

• AssumptionRests on an assumption: Another key element in mRNA translation efficiency and stability is the 5′ and 3′ UTRs.

The hypothesis that UTR-Fluc ORF interaction leads to unsuitable mRNA folding was confirmed through in vitro translation assays and SRCD analysis.

“We confirmed this hypothesis through in vitro translation assays and SRCD analysis (Figures S3B and S6).”

The A/G tail maintained comparable expression levels of reporter gene in cellulo and in vivo compared to BioNTech's A30L70 tail.

“This A/G tail demonstrated superior stability compared to BioNTech’s A30L70 tail in bacterial systems while maintaining comparable expression levels of reporter gene in cellulo and in vivo.”

Cap-independent 5′ UTRs exhibited a poor translation rate across three different cell lines compared to cap-dependent 5′ UTRs.

“Our results demonstrated that cap-independent 5′ UTRs exhibited a poor translation rate across three different cell lines compared to cap-dependent 5′ UTRs, which is consistent with previous studies showing that IRES are more efficient under stress conditions.42,43”

Purification using oligo dT magnetic beads reduced but did not completely eliminate double-stranded RNA (dsRNA) contamination.

“We demonstrated that purification using oligo dT magnetic beads reduced but did not completely eliminate double-stranded RNA (dsRNA) contamination (Table S3).”

• AssumptionRests on an assumption: The parameters of the IVT and the choice of downstream purification are crucial for improved translation efficiency.

In vivo immunization tests showed comparable levels of antigen-specific IgG production and neutralizing antibodies across the groups.

“However, in vivo immunization tests showed comparable levels of antigen-specific IgG production and neutralizing antibodies across the groups (Figure 7).”

The A/G tail yielded similar results in vaccination against SARS-CoV-2.

“Moreover, it yielded similar results in vaccination against SARS-CoV-2.”

By selecting the 3′ UTRs (VP6 and SOD) paired with hAg, protein yields were improved by 40% compared to the benchmark in human muscle cells.

“By selecting the 3′ UTRs, we improved protein yields by 40% compared to the benchmark in human muscle cells (Figure 7).”

• AssumptionRests on an assumption: Another key element in mRNA translation efficiency and stability is the 5′ and 3′ UTRs.

Improved protein production from combinations of three cap-dependent 5′ UTRs (hAg, UTR3, UTR4) with three 3′ UTR sequences (AES, VP6, SOD) in various cellular models was due to increased protein translation efficiency rather than enhanced mRNA stability.

“The evaluation of the combinations of three cap-dependent 5′ UTRs (hAg, UTR3, and UTR4) with three 3′ UTR sequences (AES, VP6, and SOD) in various cellular models revealed that improved protein production was due to increased protein translation efficiency rather than enhanced mRNA stability (Figures 4 and 5).”

• AssumptionRests on an assumption: Another key element in mRNA translation efficiency and stability is the 5′ and 3′ UTRs.

The 5′ UTR hAg yielded similar results regardless of the 3′ UTR sequence across different cell models.

“Our result showed that the 5′ UTR hAg yielded similar results regardless the 3′ UTR sequence across different cell models, suggesting that the 5′ UTR is the primary driver for protein translation from synthetic administered mRNA.”

• AssumptionRests on an assumption: Another key element in mRNA translation efficiency and stability is the 5′ and 3′ UTRs.

The 5′ UTR neoUTR3 paired with the 3′ UTR AES-mtRNR1 yielded different results depending on the ORF used.

“The 5′ UTR neoUTR3 paired with the 3′ UTR AES-mtRNR1 yielded different results depending on the ORF used.”

The observed differences in total protein production among UTR combinations can be attributed to ribosome loading rather than mRNA stability or immunogenicity.

“Taken together, these results suggest that the observed differences in total protein production among mRNAs UTR combinations can be attributed to ribosome loading rather than to mRNA stability or immunogenicity.”

Expression of the UTR constructs is cell-type dependent.

“Altogether, these results prove that the expression is cell-type dependent, relieving the difficulty in predicting the efficacy of UTR combinations and reinforcing the need for evaluation in cellular and in vivo models.”

VP6 and SOD 3′ UTRs are viable alternatives to AES-mtRNR1 in the context of mRNA vaccine development.

“Altogether, these results demonstrate the potential of VP6 and SOD 3′ UTRs, as well as the A/G tail, as viable alternatives to AES-mtRNR1 and A30L70, respectively, in the context of mRNA vaccine development.”

The efficiency of mRNA translation depends on the specific combination of 5′ UTR and 3′ UTR sequences, rather than only on the 5′ UTR sequence.

“This polysome profiling analysis revealed that the efficiency of mRNA translation depends on the specific combination of 5′ UTR and 3′ UTR sequences, rather than only on the 5′ UTR sequence.”

Cap-dependent 5′ UTRs exhibit significantly higher translation efficiency than cap-independent 5′ UTRs.

“These results illustrated that cap-dependent 5′ UTRs exhibit significantly higher translation efficiency compared to cap-independent 5′ UTRs.”

The sequence of the coding region may interfere with the folding of the 5′ UTR and 3′ UTR, leading to a conformation of the mRNA less favorable for translation.

“These results indicate that the sequence of the coding region may interfere with the folding of the 5′ UTR and 3′ UTR, leading to a conformation of the mRNA less favorable for translation.”

The selected 3′ UTRs (VP6 and SOD) and heterologous A/G tail have great potential as new elements for therapeutic mRNA design.

“Overall, the selected 3′ UTRs and heterologous A/G tail have great potential as new elements for therapeutic mRNA design.”

• Also stated in: § Evaluation of therapeutic potential · boosted

  “Altogether, these results demonstrate the potential of VP6 and SOD 3′ UTRs, as well as the A/G tail, as viable alternatives to AES-mtRNR1 and A30L70, respectively, in the context of mRNA vaccine development.”

The difference observed in translation potency among the nine mRNAs is not caused by mRNA half-life.

“These results supported similar kinetics decay for the nine mRNA transcripts, indicating that the difference observed in translation potency is not caused by mRNA half-life.”

The heterologous A/G tail is as potent as the Pfizer-BioNTech A30L70 tail for mRNA translation.

“Our results showed that the heterologous tail we introduced is as potent as the Pfizer-BioNTech tail and confirmed the high potency of the human α-globin 5′ UTR.”

• Also stated in: § Alternative poly(A) tail · boosted

  “These results demonstrate that the A/G tail is as potent as A30L70 and represents a viable alternative for mRNA design.”

Guanosine residues are the most frequent non-adenosine nucleotides in poly(A) tails in humans and other species.

“Guanosine residues are the most frequent non-adenosine nucleotides in poly(A) tails in humans and other species.20,21”

• cites [20] Lim, 2018, Mixed tailing by TENT4A and TENT4B shields mRNA from rapid deadenylation, Science sourceNot checked — no full text of the source
• cites [21] Chang, 2014, TAIL-seq: Genome-wide Determination of Poly(A) Tail Length and 3′ End Modifications, Mol. Cell sourceNot checked yet

  Checked before — its earlier verdict is held back until a new check runs.

The A/G tail is a viable alternative for mRNA design.

“These results demonstrate that the A/G tail is as potent as A30L70 and represents a viable alternative for mRNA design.”

The A30L70 and A/G tails have comparable structural properties.

“The structural analysis revealed that the A30L70 and A/G tails exhibited a similar Tm and SRCD profiles, suggesting comparable structural properties.”

Factors regulating translation and mRNA stability are cell-type-specific.

“In contrast to previous studies that optimize mRNA domains separately, our study highlights the importance of considering the entire mRNA sequence, the production parameters of IVT mRNA, and the target cell types of the desired application, as factors regulating translation and mRNA stability are cell-type-specific.”

• AssumptionRests on an assumption: Translation efficiency and mRNA stability of synthetic mRNA are directly linked to its design elements (Cap, UTR, codon usage, and poly(A) tail), IVT parameters, and downstream purification.
• AssumptionRests on an assumption: Another key element in mRNA translation efficiency and stability is the 5′ and 3′ UTRs.
• AssumptionRests on an assumption: The parameters of the IVT and the choice of downstream purification are crucial for improved translation efficiency.

Ribosome load on mRNA is influenced by the 5′ and 3′ UTR base-pairing interactions.

“This observation is supported by the polysome profiling results (Figure 5), suggesting that ribosome load on mRNA is influenced by the 5′ and 3′ UTR base-pairing interactions.”

Open reading frame (ORF) may affect the mRNA secondary structure and, consequently, impact its translation efficiency.

“The in vivo evaluation of these 5′ and 3′ UTR combinations highlighted the fact that open reading frame (ORF) may affect the mRNA secondary structure and, consequently, impact its translation efficiency.45”

• cites [45] Bicknell, 2024, Attenuating ribosome load improves protein output from mRNA by limiting translation-dependent mRNA decay, Cell Rep. sourceNot checked yet

  Checked before — its earlier verdict is held back until a new check runs.

The instability of the traditional poly(A) tail during plasmid amplification in bacterial systems presents a significant challenge in mRNA production.

“The instability of the traditional poly(A) tail during plasmid amplification in bacterial systems presents a significant challenge in mRNA production.”

• cites [17] Grier, 2016, pEVL: A Linear Plasmid for Generating mRNA IVT Templates With Extended Encoded Poly(A) Sequences, Mol. Ther. Nucleic Acids sourceNot checked yet

  Checked before — its earlier verdict is held back until a new check runs.

The 5′ UTR is the primary driver for protein translation from synthetic administered mRNA.

“Our result showed that the 5′ UTR hAg yielded similar results regardless the 3′ UTR sequence across different cell models, suggesting that the 5′ UTR is the primary driver for protein translation from synthetic administered mRNA.”

A possible explanation for the ORF-dependent difference is the interaction between the UTRs and the Fluc ORF, leading to mRNA folding that is not suitable for efficient translation.

“A possible explanation is the interaction between the UTRs and the Fluc ORF, leading to mRNA folding that is not suitable for efficient translation.46”

• cites [46] Lewis, 2025, Quantitative profiling of human translation initiation reveals elements that potently regulate endogenous and therapeutically modified mRNAs, Mol. Cell sourceNot checked yet

  Checked before — its earlier verdict is held back until a new check runs.

The enhanced performance of the A/G tail may be attributed to increased mRNA stability within cells.

“The enhanced performance of the A/G tail may be attributed to increased mRNA stability within cells.”

Transfection with LFM did not induce detectable cytotoxicity.

“Transfection with LFM did not induce detectable cytotoxicity (Figure S9).”