Minding your caps and tails

Applications of synthetic mRNA have grown and become considerably diversifed in recent years. Examples include the generation of pluripotent stem cells (1-3), accines and therapeutics (4-5), and CRISPR/Cas9 genome editing applications (6-8). The basic requirements for a functional mRNA – a 7-methylguanylate cap at the 5´ end and a poly(A) tail at the 3´ end – must be added in order to obtain effcient translation in eukaryotic cells. Additional considerations can  include the incorporation of internal modifed bases, modifed cap structures and polyadenylation strategies. Strategies for in vitro synthesis of mRNA vary according to the desired scale of synthesis. This article discusses options for the selection of reagents and the extent to which they influence synthesized mRNA functionality.

by Breton Hornblower, Ph.D., G. Brett Robb, Ph.D. and George Tzertzinis, Ph.D., New England Biolabs, Inc

A nascent mRNA, synthesized in the nucleus, undergoes different modifcations before it can be translated into proteins in the cytoplasm. For a mRNA to be functional, it requires modifed 5´ and 3´ ends and a coding region (i.e., an open reading frame (ORF) encoding for the protein of interest) flanked by the untranslated regions (UTRs). The nascent mRNA (premRNA) undergoes two signifcant modifcations in addition to splicing. During synthesis, a 7-methylguanylate structure, also known as a “cap”, is added to the 5´ end of the pre-mRNA, via 5´ → 5´ triphosphate linkage. This cap protects the mature mRNA from degradation, and also serves a role in nuclear export and efcient translation. The second modifcation occurs posttranscriptionally at the 3´ end of the nascent RNA molecule, and is characterized by addition of approximately 200 adenylate nucleotides (poly(A) tail). The addition of the the poly(A) tail confers stability to the mRNA, aids in the export of the mRNA to the cytosol, and is involved in the formation of a translation-competent ribonucleoprotein (RNP), together with the 5´ cap structure. The mature mRNA forms a circular structure (closed-loop) by bridging the cap to the poly(A) tail via the cap-binding protein eIF4E (eukaryotic initiation factor 4E) and the poly(A)- binding protein, both of which interact with eIF4G (eukaryotic initiation factor 4G), (Figure 1, (9)).

Figure 1. Translation initiation complex.
A mature mRNA, consisting of the 5´ and 3´ untranslated regions (UTRs) and the open reading frame (ORF), forms a “closed-loop” structure via interactions mediated by protein complexes that bind the cap structure and the poly(A) tail.

RNA can be efciently synthesized in vitro (by in vitro transcription, IVT) with prokaryotic phage polymerases, such as T7, T3 and SP6. The cap and poly(A) tail structures characteristic of mature mRNA can be added during or after the synthesis by enzymatic reactions with capping enzymes and Poly(A) Polymerase (NEB #M0276), respectively.

There are several factors to consider when planning for IVT-mRNA synthesis that will influence the ease-of-experimental setup and yield of the fnal mRNA product. These are discussed in the following sections.

DNA template

The DNA template provides the sequence to be transcribed downstream of an RNA polymerase promoter. There are two strategies for generating transcription templates: PCR amplifcation and linearization of plasmid with a restriction enzyme (Figure 2). Which one to choose will depend on the downstream application. In general, if multiple sequences are to be made and transcribed in parallel, PCR amplifcation is recommended as it generates many templates quickly. On the other hand, if large amounts of one or a few templates are required, plasmid DNA is recommended, because of the relative ease of producing large quantities of high quality, fully characterized plasmids. There are different versions of plasmids available that allow for propagation of homopolymeric A-tails of defned length (1).

PCR allows conversion of any DNA fragment to a transcription template by appending the T7 (or SP6) promoter to the forward primer (Figure 2A). Additionally, poly(d)T-tailed reverse primers can be used in PCR to generate transcription templates with A-tails. This obviates the need for a separate polyadenylation step following transcription. Repeated amplifcations should, however, be avoided to prevent PCR-generated point mutations. Amplifcation using PCR enzymes with the highest possible fdelity, such as Q5® High-Fidelity DNA Polymerase (NEB #M0491), reduces the likelihood of introducing such mutations. (2)

Figure 2. Methods for generating transcription templates.
(A) PCR can be used to amplify target DNA prior to transcription. A promoter can be introduced via the upstream primer.

(B) When using plasmid DNA as a template, linearize with an enzyme that produces blunt or 5´-overhanging ends. Using a type IIS restriction enzyme (e.g., BspQI) allows RNA synthesis with no additional 3´-nucleotide sequence from the restriction site.

The quality of the PCR reaction can be assessed by running a small amount on an agarose gel, and DNA should be purifed before in vitro transcription using a spin column or magnetic beads (e.g., AMPure® beads). Multiple PCR reactions can be purifed and combined to generate a DNA stock solution that can be stored at -20°C and used as needed for in vitro transcription. 

Plasmid templates are convenient if the template sequence already exists in a eukaryotic expression vector also containing the T7 promoter (e.g., pcDNA vector series). These templates include 5´- and 3´-untranslated regions (UTR), which are important for the expression characteristics of the mRNA.

Plasmid DNA should be purifed and linearized downstream of the desired sequence, preferably with a restriction enzyme that leaves blunt or 5´ overhangs at the 3´ end of the template. These are favorable for proper run-off transcription by T7 RNA Polymerase (NEB #M0274), while 3´ overhangs may result in unwanted transcription products. To avoid adding extra nucleotides from the restriction site to the RNA sequence, a Type IIS restriction enzyme can be used (e.g., BspQI, NEB #R0712), which positions the recognition sequence outside of the transcribed sequence (Figure 2B, page 2). The plasmid DNA should be completely digested with the restriction enzyme, followed by purifcation using a spin column (e.g., Monarch® PCR & DNA Cleanup Kit (5 μg) NEB #T1030) or phenol extraction/ethanol precipitation. Although linearization of plasmid involves multiple steps, the process is easier to scale for the generation of large amounts of template for multiple transcription reactions.

In vitro transcription

There are two options for the in vitro transcription (IVT) reaction depending on the capping strategy chosen: standard synthesis with enzyme-based capping following the transcription reaction (posttranscriptional capping) or incorporation of a cap analog during transcription (co-transcriptional capping) (Figure 3). Method selection will depend on the scale of mRNA synthesis required and number of templates to be transcribed.

Figure 3. In vitro transcription options based upon capping strategy
Enzyme-based capping (left) is performed after in vitro transcription using 5´-triphosphate RNA, GTP, and S-adenosylmethionine (SAM). Cap 0 mRNA can be converted to cap 1 mRNA using mRNA cap 2´-O-methyltransferase (MTase) and SAM in a subsequent or concurrent reaction. The methyl group transferred by the MTase to the 2´-O of the frst nucleotide of the transcript is indicated in red. Conversion of ~100% of 5´-triphosphorylated transcripts to capped mRNA is routinely achievable using enzyme-based capping.

Co-transcriptional capping (right) uses an mRNA cap analog (e.g., ARCA; anti-reverse cap analog), shown in yellow, in the transcription reaction. The cap analog is incorporated as the frst nucleotide of the transcript. ARCA contains an additional 3´-O-methyl group on the 7-methylguanosine to ensure incorporation in the correct orientation. The 3´-O-methyl modifcation does not occur in natural mRNA caps. Compared to reactions not containing cap analog, transcription yields are lower. ARCA-capped mRNA can be converted to cap 1 mRNA using mRNA cap 2´-O-MTase and SAM in a subsequent reaction.

Transcription for enzyme-based capping (post-transcriptional capping)

Standard RNA synthesis reactions produce the highest yield of RNA transcript (typically ≥100 μg per 20 μl in a 1 hr reaction using the HiScribe Quick T7 High Yield RNA Synthesis Kit, NEB #E2050S). Transcription reactions are highly scalable, and can be performed using an all-inclusive kit (e.g., HiScribe kits), or individual reagents. More information on the HiScribe kits can be found later in the article.

Following transcription, the RNA is treated with DNase I (NEB #M0303) to remove the DNA template, and purifed using an appropriate column, kit or magnetic beads, prior to capping. This method produces high yields of RNA with 5´-triphosphate termini that must be converted to cap structures. In the absence of templateencoded poly(A) tails, transcripts produced using this method bear 3´ termini that also must be polyadenylated in a separate enzymatic step, as described below in “Post-transcriptional capping and Cap-1 methylation”.

Transcription with co-transcriptional capping

In co-transcriptional capping, a cap analog is introduced into the transcription reaction, along with the four standard nucleotide triphosphates, in an optimized ratio of cap analog to GTP 4:1. This allows initiation of the transcript with the cap structure in a large proportion of the synthesized RNA molecules. This approach produces a mixture of transcripts, of which ~80% are capped, and the remainder have 5´-triphosphate ends. Decreased overall yield of RNA products results from the lower concentration of GTP in the reaction (Figure 4, page 3).

Figure 4. RNA yields from transcriptional capping reactions
Reactions were set up according to recommended conditions for two templates: Gaussia luciferase (GLuc) and Cypridina luciferase (CLuc). The RNA was quantifed spectrophotometrically after purifcation with spin columns.

There are several cap analogs used in cotranscriptional RNA capping (3,4). The most common are the standard 7-methyl guanosine (m7G) cap analog and anti-reverse cap analog (ARCA), also known as 3´ O-me 7-meGpppG cap analog (Figure 5). ARCA is methylated at the 3´ position of the m7G, preventing RNA elongation by phosphodiester bond formation at this position. Thus, transcripts synthesized using ARCA contain 5´-m7G cap structures in the correct orientation, with the 7-methylated G as the terminal residue. In contrast, the m7G cap analog can be incorporated in either the correct or the reverse orientation.

Figure 5. Structure of the anti-reverse cap analog, ARCA
The 3′ position of the 7-methylated G is blocked by a methyl group.

HiScribe T7 ARCA mRNA Synthesis Kits (NEB #E2060 and #E2065) contain reagents, including an optimized mix of ARCA and NTPs, for streamlined reaction setup for synthesis of cotranscriptionally capped RNAs

Transcription with complete substitution with modified nucleotides

RNA synthesis can be carried out with a mixture of modifed nucleotides in place of the regular mixture of A, G, C and U triphosphates. For expression applications, the modifed nucleotides of choice are the naturally occurring 5´-methylcytidine and/ or pseudouridine in the place of C and U, respectively. These have been demonstrated to confer desirable properties to the mRNA, such as increased mRNA stability, increased translation, and reduced immune response in the key applications of protein replacement and stem-cell differentiation (1). It is important to note that nucleotide choice can influence the overall yield of mRNA synthesis reactions.

Fully substituted RNA synthesis can be achieved using the HiScribe T7 High-Yield RNA Synthesis Kit (NEB #E2040) or HiScribe SP6 RNA Synthesis Kit (NEB #E2070) in conjunction with NTPs with the desired modifcation. Transcripts made with complete replacement of one or more nucleotides may be post-transcriptionally capped (see next section), or may be co-transcriptionally capped by including ARCA or another cap analog, as described previously

If partial replacement of nucleotides is desired, the HiScribe T7 ARCA mRNA Synthesis Kits (NEB #E2060 and #E2065), may be used with added modifed NTPs, to produce co-transcriptionally capped mRNAs, as described above. Alternatively, the HiScribe T7 Quick RNA Synthesis Kit (NEB #E2050) may be used to prepare transcripts for post-transcriptional capping (see below). 

Post-transcriptional capping and Cap-1 methylation

Post-transcriptional capping is often performed using the mRNA capping system from Vaccinia virus. This enzyme complex converts the 5´-triphosphate ends of in vitro transcripts to the m7G-cap structures. The Vaccinia Capping System (NEB #M2080) comprises three enzymatic activities (RNA triphosphatase, guanylyltransferase, guanine N7-methyltransferase) that are necessary for the formation of the complete Cap-0 structure, m7Gppp5´N, using GTP and the methyl donor S-adenosylmethionine. As an added option, the inclusion of the mRNA Cap 2´ O-Methyltransferase (NEB #M0366) in the same reaction results in formation of the Cap-1 structure, which is a natural modifcation in many eukaryotic mRNAs. This enzymebased capping approach results in the highest proportion of capped message, and it is easily scalable. The resulting capped RNA can be further modifed by poly(A) addition before fnal purifcation 

A-tailing using E. coli Poly(A) Polymerase

The poly(A) tail confers stability to the mRNA and enhances translation efciency. The poly(A) tail can be encoded in the DNA template by using an appropriately tailed PCR primer, or it can be added to the RNA by enzymatic treatment with E. coli Poly(A) Polymerase (NEB #M0276). The length of the added tail can be adjusted by titrating the Poly(A) Polymerase in the reaction (Figure 6).

Figure 6. Analysis of capped and polyadenylated RNA
A. Agilent® Bioanalyzer® analysis of capped and polyadenylated RNA. Longer tails are produced by increasing the enzyme concentration in the reaction. Calculated A-tail lengths are indicated over each lane. Lanes: L: size marker,1: No poly-A tail, 2: 5 units, 3 :15 units, 4 : 25 units of E. coli Poly(A) Polymerase per 10 µg CLuc RNA in a 50 µl reaction.

B. Effect of enzymatic A-tailing on the luciferase reporter activity of CLuc mRNA

The importance of the A-tail is demonstrated by transfection of untailed vs. tailed mRNA. When luciferase activity from cells transfected with equimolar amounts of tailed or untailed mRNAs were compared, a signifcant enhancement of translation efciency was evident (Figure 6). HiScribe T7 ARCA mRNA Synthesis Kit (with tailing) (NEB #E2060) includes E. coli Poly(A) Polymerase, and enables a streamlined workflow for the enzymatic tailing of co-transcriptionally capped RNA.

For mRNA synthesis from templates with encoded poly(A) tails, the HiScribe T7 ARCA mRNA Synthesis Kit (NEB #E2065) provides an optimized formulation for co-transcriptionally capped transcripts.


In summary, when choosing the right workflow for your functional mRNA synthesis needs, you must balance your experimental requirements for the mRNA (e.g., internal modifed nucleotides) with scalability (i.e., ease-of-reaction setup vs. yield of fnal product).

In general, co-transcriptional capping of mRNA with template encoded poly(A) tails or post-transcriptional addition of poly(A) tail is recommended for most applications. This approach, using the HiScribe T7 ARCA mRNA Synthesis Kits (NEB #E2060 and #E2065), enables the quick and streamlined production of one or many transcripts with typical yields of ≥20 μg per reaction, totaling ~400-500 μg per kit. 

Post-transcriptional mRNA capping with Vaccinia Capping System is well suited to larger scale synthesis of one or a few mRNAs, and is readily scalable to produce gram-scale quantities and beyond. Reagents for in vitro synthesis of mRNA are available in kit form or as separate components to enable research and large-scale production.

Products from NEB are available for each step of the RNA Synthesis Product Workflow

View a PDF of this feature article


  1. Warren, L., et al. (2010) Cell Stem Cell, 7, 618-630.
  2. Angel, M. and Yanik, M.F. (2010) PLoS One, 5:e11756.
  3. Yakubov, E., et al. (2010) Biochem. Biophys. Res. Commun. 394, 189.
  4. Geall, A.J., et al. (2012) Proc. Natl. Acad. Sci. USA, 109, 14604-14609.
  5. Ramaswamy, S., et al. (2017) Proc. Natl. Acad. Sci. USA, 114, E1941-E1950.
  6. Ma, Y., et al. (2014) PLoS One, 9:e89413.
  7. Ota, S., et al. (2014) Genes Cells, 19, 555-564.
  8. Bassett, A. R., et al. (2013) Cell Rep. 4, 220–228.
  9. Wells, S.E., et al. (1998) Molecular Cell 2, 135–140.