Technical articles
Experimental Judgment in mRNA In Vitro Transcription: Template Design, 5′ Capping, and dsRNA Control
Experimental Judgment in mRNA In Vitro Transcription: Template Design, 5′ Capping, and dsRNA Control
Introduction
The outcome of mRNA in vitro transcription does not usually depend only on whether an RNA of the target length has been obtained. Whether the template elements are arranged appropriately, whether the 5′ cap structure matches the intended downstream use, and whether key impurities such as double-stranded RNA (double-stranded RNA, dsRNA) have been reduced in time will all directly affect translation output, reproducibility, and immune responses. In its regulatory considerations for mRNA vaccines, the WHO points out that mRNA structural integrity is a critical quality attribute, and that actual control generally needs to pay attention to indicators such as 5′ capping efficiency, the presence or length of the 3′ poly(A) tail, the proportion of full-length mRNA, the proportion of fragments, and the percentage of dsRNA. Summaries of the work of Karikó and Weissman have also clearly shown that modified nucleosides and the control of immunostimulatory impurities can significantly affect the expression and immunological performance of in vitro transcribed RNA.
1. Three Key Points That Need to Be Distinguished First in In Vitro Transcription
1.1 | Overview of the Three Key Judgment Points in In Vitro Transcription
Critical aspect | Main impact | What should be confirmed first before the experiment | Common misjudgment |
Template design | Whether transcription can initiate smoothly, whether product length is uniform, and whether the UTRs and poly(A) on which subsequent translation depends are in an appropriate combination | Whether the promoter, 5′UTR, coding region, 3′UTR, poly(A), and linearization site are matched to one another | Assuming that as long as RNA can be transcribed, the template has no problem |
5′ capping | Translation initiation efficiency, 5′-end maturity, and the risk of cellular recognition of exogenous RNA and translation inhibition | Whether to choose co-transcriptional capping or enzymatic capping, and whether the cap structure matches the experimental objective | Looking only at capping rate, without considering cap structure type or the residual uncapped RNA |
dsRNA control | Whether translation is suppressed, whether innate immune readouts are amplified, and whether the purification burden and the difficulty of process scale-up increase | Whether dsRNA is more likely to be generated upstream or to accumulate downstream, and at which stage it is planned to be reduced | Treating dsRNA as a minor issue to be handled incidentally during final product purification |
2. Key Experimental Judgments at the Template Design Stage
The template is the starting point of in vitro transcription (in vitro transcription, IVT), and many subsequent abnormalities may also have their roots at the template stage. The promoter determines how the polymerase initiates transcription; the 5′ untranslated region (5′ untranslated region, 5′UTR) affects translation initiation; the coding region affects local RNA structure and the translation process; the 3′ untranslated region (3′ untranslated region, 3′UTR) and the poly(A) tail are related to stability and expression performance; and the linearization site determines whether the transcription endpoint is clearly defined. WHO-related documents indicate that these design features should be described because they affect critical quality attributes, manufacturing methods, and control testing. Regarding the 5′UTR, recent studies also support that its length, sequence, and secondary structure can significantly affect translation, and that overly strong structure and upstream start codons can impose additional burdens.
2.1 | Several Judgment Points That Should Be Clarified First in Template Design
Template element | Direct impact | What should be emphasized | Common experimental manifestation |
Promoter | Whether the polymerase can initiate efficiently and whether the sequence boundaries near the initiation site are appropriate | Whether the selected polymerase matches the promoter, and whether the sequence downstream of the promoter imposes additional constraints on initiation | Low total yield and obvious batch-to-batch fluctuation |
5′ untranslated region (5′UTR) | Ribosome assembly and translation initiation efficiency | Whether the secondary structure is too strong, and whether it contains upstream start codons or inappropriate regulatory elements | Normal RNA band, but low protein expression |
Coding region | Local RNA structure, the translation process, and sequence composition | Whether regions of high local structure, repetitive segments, codon usage, and local folding impose burdens | Transcription can proceed, but expression is unstable or relatively weak |
3′ untranslated region (3′UTR) | RNA stability and translation performance | Whether the selected 3′UTR matches the target cell type or intended experimental use | RNA can be detected, but duration or output is unsatisfactory |
poly(A) tail | Stability and translation performance | Whether the tail is template-encoded or added afterward, and whether its length is coordinated with the overall construct | Insufficient expression persistence and poor batch-to-batch consistency |
Linearization site | Whether the transcription endpoint is clearly defined and whether the terminus is clean | Whether linearization is complete, and whether non-target sequences are introduced at the terminus | Band tailing, uneven length, and heavy downstream purification burden |
There is another class of problems on the template side that may appear on the surface to be caused by “unstable reaction conditions,” but in fact have already been introduced earlier. Incomplete linearization, poor template purity, and residual non-target fragments at the template terminus can all amplify subsequent formation of double-stranded RNA (double-stranded RNA, dsRNA) and abnormal length distribution. This point is supported by a 2023 study on linearized template purification, as well as 2018 and 2024 studies on T7 RNA polymerase (T7 RNA polymerase) by-products.
2.2 | Which Abnormal Manifestations Should Be Investigated First from the Template Side
Experimental phenomenon | What to check first | Common cause | Adjustments that should not be rushed into |
Low total yield | Template integrity, linearization quality, promoter, and initiation region | Template not fully linearized, poor template purity, or inappropriate initiation-region design | Directly and sharply increasing polymerase dosage or extending reaction time |
Band tailing or uneven length | Linearization site and template terminal state | Untidy termini or the presence of additional template sources | Immediately attributing the problem to the purification step |
Large batch-to-batch differences | Template batch consistency and template preparation process | Different batches vary in template purity and degree of linearization | Changing only the IVT buffer formulation |
RNA can be produced but expression is weak | 5′ untranslated region, local structure in the coding region, and the combination of 3′ untranslated region and poly(A) tail | The combination of upstream elements is unfavorable for translation output | Attributing everything from the outset to the capping route |
3. Key Experimental Judgments in Choosing a Capping Route
There are two core questions in 5′ capping. One is how to make the RNA more closely resemble the 5′-end state of mature messenger RNA (messenger RNA, mRNA) after it enters the cell. The other is how to consider this together with the current template design, purification route, and experimental objective. WHO-related documents clearly state that the cap structure can be introduced either during in vitro transcription (IVT) or enzymatically after transcription.
On the other hand, the cap structure does not affect only whether the RNA is “capped.” It also affects subsequent translation and how the cell recognizes the RNA. Studies have shown that the identity of the first transcribed nucleotide and its methylation state can alter the biological behavior of the 5′ end. Interferon-induced protein with tetratricopeptide repeats 1 (interferon-induced protein with tetratricopeptide repeats 1, IFIT1) more readily recognizes cap0 structures lacking 2′-O methylation and RNA bearing a 5′ triphosphate terminus, while its inhibitory effect on cap1 structures is weaker. Therefore, when judging whether the 5′ end is appropriate, it is not enough to look only at whether capping has been completed; the cap structure type and the modification state of the first nucleotide must also be considered.
3.1 | Key Points in Choosing Between Co-Transcriptional Capping and Enzymatic Capping
Capping route | Applicable scenario | Main advantage | What should be emphasized | Common downstream situation |
Co-transcriptional capping | When transcription and capping are intended to be completed in a single step, reducing the number of process stages | A more integrated workflow, making it easier to connect IVT directly with 5′-end introduction | Whether the initiation region is compatible with the cap analog, and whether uncapped RNA is prone to accumulate | Transcription yield is already present, but 5′-end maturity may not be ideal |
Enzymatic capping | When transcription products are to be obtained first and the 5′-end maturation process optimized separately | 5′-end processing can be optimized independently of transcription | An additional processing step is introduced, placing higher demands on downstream purification and analysis | The overall process connection is longer and there are more control points |
If one looks only at capping efficiency, two more critical issues are easily overlooked. First, cap0 and cap1 differ in their effects on downstream translation and immune recognition. Second, the first transcribed nucleotide and the overall maturity of the 5′ end also affect the result. Therefore, when encountering situations such as “RNA yield is already sufficient, but expression is low” or “immune readouts are high,” one should first recheck 5′-end maturity, cap structure type, and residual uncapped RNA before deciding whether the route needs to be changed.
3.2 | In Which Abnormal Situations the Capping Route Should Be Checked First
Experimental phenomenon | What to check first | Possible step involved | Adjustments that should not be made first |
RNA amount and length are basically normal, but protein expression is low | Cap structure type, capping completeness, and the state of the first transcribed nucleotide | Insufficient 5′-end maturity, or a cap structure that does not match the experimental objective | Immediately going back and extensively redesigning the template |
Immune response readouts are high, but the template profile is normal | cap0/cap1 status, uncapped RNA, and residual 5′ triphosphate | The 5′ end is more readily recognized as exogenous RNA | Attributing the problem only to the delivery system |
Clear differences are observed between different capping routes | The degree of matching between the route and the template initiation region | Route selection and template design are not coordinated | Switching the capping route directly without first checking whether the template initiation region matches |
4. Formation, Effects, and Control Routes of Double-Stranded RNA
Double-stranded RNA (double-stranded RNA, dsRNA) is one of the product-related impurities in in vitro transcribed (IVT) messenger RNA (messenger RNA, mRNA) that requires focused control. WHO-related documents indicate that such dsRNA can be formed as a by-product during IVT and can induce innate immune responses in cells; it should therefore be removed during manufacturing, or at least measured and controlled. Studies on the formation mechanism indicate that dsRNA produced by T7 RNA polymerase is mainly associated with self-templated extension, promoter-independent transcription, and non-target initiation transcription related to the state of the DNA template terminus. The terminal sequence and structure of the template can significantly affect the formation of these by-products.
4.1 | Sources, Effects, and Control Stages of dsRNA
dsRNA-related stage | Common source or situation | Possible consequence | Control approach that can be prioritized | Applicable stage |
Upstream generation | Self-templated extension, promoter-independent antisense transcription, and non-target transcription triggered by the template terminus | Translation is suppressed, immune response readouts rise, and downstream purification burden increases | Improve the purity of the linearized template, optimize the template terminus and initiation region, and adjust reaction conditions | Template design and IVT small-scale testing stage |
Variables related to immune recognition | Uridine sites are unmodified, making the RNA more likely to induce immune responses | Elevated innate immune readouts and suppressed translation performance | According to the experimental objective, evaluate replacement of uridine sites with pseudouridine or N1-methylpseudouridine to reduce immune stimulation and improve translation performance | Upstream design stage |
Downstream residuals | dsRNA is not effectively reduced, and aberrant-length RNA coexists with other impurities | Insufficient final product purity and poor reproducibility | Use an appropriate purification route, such as high-resolution purification or a cellulose-based method | Purification and process-integration stage |
Accumulation during scale-up | Small-scale experiments are controllable, but the proportion of by-products rises after scale-up | Batch-to-batch differences are amplified and purification losses increase | Fix template quality and key process windows before scale-up | Pre-scale-up evaluation stage |
When choosing modified nucleosides, it is first necessary to determine which type of site is being replaced. Pseudouridine and N1-methylpseudouridine correspond to replacement at uridine sites and are mainly used to reduce immune stimulation and improve translation performance; they do not correspond to replacement at guanosine sites. If this point is not clearly distinguished, subsequent judgments on substrate design, immune readouts, and expression changes can easily deviate.
For dsRNA that has already formed, downstream purification remains a key control step. A 2019 study showed that the cellulose-based method can effectively, reliably, and safely remove dsRNA impurities from IVT mRNA; WHO also regards dsRNA as an impurity that requires focused attention and control. Therefore, dsRNA is better managed through a combined strategy of “reducing generation upstream and lowering residuals downstream,” rather than waiting until the final product stage to attempt centralized remediation.
5. Priority Troubleshooting Sequence After Abnormal Results Appear: Template, 5′ End, and dsRNA
After abnormal results appear, it is usually more effective to first separate the problem into three lines of analysis, namely the template, the 5′ end, and dsRNA, than to change multiple conditions at the same time. The template is at the very front end, the state of the 5′ end directly affects translation initiation, and dsRNA is influenced both by upstream generation and downstream purification. Once the troubleshooting sequence is clarified, many seemingly complex phenomena can be traced more easily to specific variables.
5.1 | Common Abnormal Phenomena and Priority Troubleshooting Points
Experimental phenomenon | Priority check | Common cause | Adjustments that should not be made first |
Low total RNA yield | Template integrity, linearization quality, promoter, and initiation region | The template problem already exists at the upstream stage | Repeatedly increasing enzyme amount first or blindly extending reaction time |
Abnormal bands and uneven length | Linearization site and template terminal state | The template endpoint is unclear, and there are additional transcription sources | Attributing the problem first to capping or downstream processing |
RNA yield is sufficient, but protein expression is low | 5′-end maturity; secondarily, the 5′ untranslated region and local structure in the coding region | The cap structure is inappropriate, or the combination of upstream elements is unfavorable for translation | Making major changes to both the template and the purification workflow at the same time |
High immune readouts | dsRNA, 5′-end state, and modified nucleoside design | The dsRNA burden is high, or the 5′ end is more readily recognized | Attributing the problem only to cell status or the delivery system |
Small-scale tests work, but performance becomes unstable after scale-up | Template batch consistency, key process windows, and purification conditions | Differences in template quality and by-product control problems are amplified after scale-up | Simply copying small-scale parameters without rechecking key points |
6. Key Information That Should Be Defined Before Scale-Up
Before moving from small-scale testing to scale-up, it is helpful to first fix several categories of key information, so that it becomes easier afterward to determine whether a problem originates from the template, the 5′ end, or impurity control. In the sections on process changes and comparability, WHO emphasizes that when the process or scale of an mRNA product changes, critical quality attributes and comparability need to be reassessed. Therefore, the more complete the early-stage records are, the easier it is afterward to distinguish whether the problem lies at the template end, the 5′ end, or in purification and dsRNA control.
6.1 | Key Information Recommended to Be Defined Before Scale-Up
Information category | Content recommended to be fixed | Main purpose |
Template information | Template profile, linearization method, and arrangement of key elements | To facilitate judgment of whether the upstream end is stable |
5′-end information | Capping route, cap structure type, and relevant analytical results | To facilitate interpretation of translation-related and immune-related readouts |
Purification information | Purification steps, dsRNA test results, and the status of aberrant-length RNA | To facilitate judgment of final product purity and by-product burden |
Result information | Total RNA yield, expression readouts, and immune readouts | To facilitate linking the observed phenomena to specific variables |
7. Product Navigation Table for Template Design, Capping Route, and dsRNA Control in mRNA In Vitro Transcription (Choose Table 1 to Table 3 According to Research or Experimental Objective)
Research or experimental objective | Recommended table to consult first | Why this table should be consulted first | Recommended table(s) to consult in combination | Reason for combined consultation |
To first set up the in vitro transcription reaction system, establish basic buffer conditions, and compare yield, reproducibility, and band patterns | Table 1 | Table 1 focuses on buffer systems, reductive protection, divalent metal ions, and polyamine-regulating components, making it suitable for first clarifying whether the reaction can initiate stably and whether the conditions are properly established | Table 2 | After the reaction conditions have been determined, Table 2 can then be used to compare basic nucleotide substrates or modified nucleosides within the same system, so as to judge whether changes in yield arise from the conditions or from substrate composition |
To compare the effects of modified nucleoside routes on expression, immune response, or system stability when a basic transcription system is already available | Table 2 | Table 2 separates basic nucleoside triphosphate substrates from modified nucleosides, making it suitable for first clarifying whether the current study is comparing conventional transcription with modified uridine or modified cytidine routes | Table 1, Table 3 | Consulting Table 1 together can help determine whether reaction conditions need to be readjusted after the introduction of modified nucleosides. Consulting Table 3 together can further compare the combined performance of modified substrates with capping methods and before-and-after dsRNA removal |
To compare different capping routes and observe the effect of 5′-end maturity on translation output or immune readouts | Table 3 | Table 3 focuses on dinucleotide cap analogs, trinucleotide cap analogs, methyl donors, and post-transcriptional purification control components, making it suitable for first distinguishing whether the work is comparing co-transcriptional capping or post-transcriptional processing and methylation arrangements | Table 1, Table 2 | Consulting Table 1 together can help determine whether the current buffer conditions support stable transcription and subsequent processing. Consulting Table 2 together can compare how different substrate compositions, combined with different capping routes, affect expression and immune readouts |
To focus on reducing double-stranded RNA impurities and compare changes in expression and immune response before and after removal | Table 3 | Table 3 includes post-transcriptional precipitation, cellulose-method-related materials, and capping components, making it suitable for first examining both downstream processing and 5′-end control together, so as to judge whether the current problem is more related to impurity removal or insufficient 5′-end maturity | Table 1 | Consulting Table 1 together can help further review whether divalent metal ions, polyamines, and the buffer system have amplified the by-product burden at an earlier stage, avoiding leaving all problems to be handled only during purification |
To design a relatively complete in vitro transcription study, proceeding in sequence from reaction conditions and substrate composition to capping and dsRNA control | Table 1 | Table 1 is suitable as the starting point, first establishing reaction conditions, buffer composition, and transcription feasibility, and then advancing step by step in a clearer order | Table 2, Table 3 | Table 2 can first be used to distinguish and compare basic substrates from modified nucleoside routes. Table 3 can then incorporate 5′ capping, precipitation-based purification, and dsRNA removal into one experimental design, forming a continuous chain of judgment |
To quickly identify the priority troubleshooting direction when encountering problems such as “RNA product is present but expression is weak” or “immune readouts are high” | Table 3 | Such phenomena often simultaneously involve 5′-end maturity, uncapped RNA, double-stranded RNA impurities, and post-transcriptional purification arrangements. Looking at Table 3 first makes it easier to narrow down the problem range | Table 2, Table 1 | If Table 3 suggests that the main problem does not lie in capping or downstream processing, Table 2 can then be consulted to judge whether the issue is related to the modified nucleoside route. Finally, Table 1 can be consulted together to check whether the reaction conditions had already introduced yield and by-product burden problems from the outset |
Table 1 | In Vitro Transcription Reaction Buffers and Condition-Regulating Components
Category | CAS No. | Aladdin Cat. No. | Name | Specification or Purity | Product Features and Applications |
Reaction buffer component | 1185-53-1 | Tris-HCl | Injection grade | Used to prepare in vitro transcription reaction buffers, maintain stable acid-base conditions, and reduce fluctuations in the reaction environment between different batches. | |
Reductive protection component | 3483-12-3 | DL-Dithiothreitol | For electrophoresis, ≥99% | Used to maintain a reducing environment, protect oxidation-sensitive components in the reaction system, and improve the stability of transcription or enzymatic capping steps. | |
Divalent metal ion component | 7786-30-3 | Magnesium chloride | Anhydrous grade, ≥99.9% metals basis, powder | Provides divalent magnesium ions and directly affects polymerase catalytic activity, transcription efficiency, and by-product levels, making it a variable that requires focused adjustment during condition optimization. | |
Polyamine-regulating component | 124-20-9 | Spermidine | UltraBio™, molecular biology grade, ≥99.5% (GC) | Functions as a polyamine component in regulating template and reaction-system stability, and is suitable for establishing a comparison window for transcription initiation and high-yield conditions. |
Table 2 | Examples of Partial Transcription Substrates and Modified Nucleosides
Category | CAS No. | Aladdin Cat. No. | Name | Specification or Purity | Product Features and Applications |
Basic nucleoside triphosphate substrate | 65-47-4 | Cytidine-5'-triphosphate | Moligand™, 10 mM in Water | One of the basic nucleoside triphosphate substrates used in the main in vitro transcription reaction, suitable for establishing substrate ratios together with other nucleoside triphosphates. | |
Basic nucleoside triphosphate substrate | 56-65-5 | Adenosine triphosphate (ATP) | Moligand™, ≥95% | One of the basic nucleoside triphosphate substrates involved in synthesis of the target RNA backbone, and a core component for establishing a balance between yield and substrate consumption. | |
Modified uridine substrate | 1175-34-4 | Pseudouridine-5′-triphosphate Sodium Salt | ≥97%, 100 mM in water | Used to replace uridine sites and is suitable for reducing immune stimulation and improving translation performance during the in vitro transcription stage. | |
Modified cytidine substrate | 327174-86-7 | 5-Methylcytidine 5′-triphosphate | _ | Used as a modified cytidine substrate to adjust the chemical composition of mRNA, and can be combined with modified uridine routes to compare expression and immune response readouts. |
Table 3 | 5′ Capping and Post-Transcriptional Purification Control Components
Category | CAS No. | Aladdin Cat. No. | Name | Specification or Purity | Product Features and Applications |
Post-transcriptional precipitation/desalting component | 7447-41-8 | Lithium chloride | Anhydrous grade, 99.99% metals basis | Commonly used for RNA precipitation after in vitro transcription and for removing most unincorporated nucleotides and enzymes, providing a relatively clean sample for subsequent capping analysis or further purification. Its main role is precipitation and desalting, and it should not be regarded as a dsRNA-specific removal step. | |
dsRNA removal material | 9004-34-6 | Cellulose | Microcrystalline powder | Suitable for reducing the burden of double-stranded RNA impurities by the cellulose method, and can be used to compare changes in expression and immune response before and after removal. | |
Methyl donor for enzymatic capping | 29908-03-0 | S-Adenosyl-L-methionine (SAM) | Moligand™, ≥98% | Serves as a methyl donor in enzymatic capping or subsequent methylation steps, helping to form a more mature 5′-end structure. | |
Dinucleotide cap analog | 75252-10-7 | N7-Methyl-guanosine-5'-triphosphate-5'-adenosine diammonium | ≥98% | A classical dinucleotide cap analog used to establish a co-transcriptional capping control route, suitable for comparison with trinucleotide cap analog routes in terms of 5′-end maturity and residual uncapped RNA. | |
Trinucleotide cap analog (adenosine-initiated type) | 62858-30-4 | m7GpppAmpG | _ | Used to introduce a trinucleotide cap structure by co-transcriptional capping and is suitable for systems in which the first transcribed nucleotide is adenosine, allowing comparison of the effects of different 5′-end structures on translation and immune readouts. | |
Trinucleotide cap analog (guanosine-initiated type) | 1258049-00-1 | m7GpppGmpG | _ | Used to introduce a trinucleotide cap structure by co-transcriptional capping and is suitable for systems in which the first transcribed nucleotide is guanosine, facilitating comparison of 5′-end maturity under different initiating-nucleotide conditions. |
Note: The above are representative Aladdin products. For more product specifications, please search on the Aladdin official website using the product name/CAS/catalog number.
References
[1] World Health Organization. Evaluation of the quality, safety and efficacy of messenger RNA vaccines for the prevention of infectious diseases: regulatory considerations. Geneva: World Health Organization; 2021.
[2] 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. 2005;23(2):165-175. doi:10.1016/j.immuni.2005.06.008.
[3] Karikó K, Muramatsu H, Welsh FA, Ludwig J, Kato H, Akira S, Weissman D. Incorporation of pseudouridine into mRNA yields superior nonimmunogenic vector with increased translational capacity and biological stability. Molecular Therapy. 2008;16(11):1833-1840. doi:10.1038/mt.2008.200.
[4] Andries O, Mc Cafferty S, De Smedt S, Weiss R, Sanders N, Kitada T. N-1-methylpseudouridine-incorporated mRNA outperforms pseudouridine-incorporated mRNA by providing enhanced protein expression and reduced immunogenicity in mammalian cell lines and mice. Journal of Controlled Release. 2015;217:337-344. doi:10.1016/j.jconrel.2015.08.051.
[5] Ma Q, Zhang X, Yang J, Li H, Hao Y, Feng X. Optimization of the 5' untranslated region of mRNA vaccines. Scientific Reports. 2024;14(1):19845. doi:10.1038/s41598-024-70792-x.
[6] Ramanathan A, Robb GB, Chan SH. mRNA capping: biological functions and applications. Nucleic Acids Research. 2016;44(16):7511-7526. doi:10.1093/nar/gkw551.
[7] Diamond MS. IFIT1: a dual sensor and effector molecule that detects non-2'-O methylated viral RNA and inhibits its translation. Cytokine & Growth Factor Reviews. 2014;25(5):543-550. doi:10.1016/j.cytogfr.2014.05.002.
[8] Mu X, Greenwald E, Ahmad S, Hur S. An origin of the immunogenicity of in vitro transcribed RNA. Nucleic Acids Research. 2018;46(10):5239-5249. doi:10.1093/nar/gky177.
[9] Yu B, Chen Y, Yan Y, Lu X, Zhu B. DNA-terminus-dependent transcription by T7 RNA polymerase and its C-helix mutants. Nucleic Acids Research. 2024;52(14):8443-8453. doi:10.1093/nar/gkae593.
[10] Baiersdörfer M, Boros G, Muramatsu H, Mahiny A, Vlatkovic I, Sahin U, Karikó K. A facile method for the removal of dsRNA contaminant from in vitro-transcribed mRNA. Molecular Therapy: Nucleic Acids. 2019;15:26-35. doi:10.1016/j.omtn.2019.02.018.
For more related articles, see below:
