Glen Report 36-14: Application Note — RNA Synthesis

Long RNA synthesis has become increasingly popular for various applications. Many researchers find themselves needing very long (100mer+) RNA oligonucleotides for real-time PCR, sgRNA for CRISPR technology1-3, or short mRNA sequences.4 Long RNAs are also important for studying genomic pathways and drug discovery. Finally, the regulatory role of long RNA in biology requires the synthesis and analysis of long RNAs.5 

Traditionally, RNA synthesis is less efficient than DNA synthesis because of the 2′-O-protecting group required for RNA. The most common protecting group at this position is a silyl group that tends to be bulky and can impede coupling through steric hindrance. TBDMS is the standard 2′-silyl group while TOM is popular as well. TOM is often recommended for long RNA synthesis because the oxymethyl spacer extends the bulky silyl group further away from the active phosphoramidite center (Figure 1). In addition, the impact of secondary structure in regard to RNA complicates the isolation of long, single-stranded RNAs.

Figure 1. DNA and standard RNA phosphoramidites

Figure 1. DNA and standard RNA phosphoramidites 

To better support this research, we sought out to provide a quantitative assessment of various factors impacting the quality of RNA synthesis, such as RNA monomers and universal supports.

To do this, we synthesized 20mer oligonucleotides under various conditions and extrapolated the data to determine efficiency for longer syntheses. DCA deblock was chosen for these experiments to mimic recommended conditions for long oligonucleotide synthesis.6 A 6 min coupling time was used with 0.25M ETT as the activator for all RNA monomers.7 We compared DNA, TBDMS, and TOM monomers as well as the impact of our two universal supports: Universal Support (US) III PS and Glen UnySupport™ CPG.

Our methods are as follows:

  • Synthesis. On an ABI 394 synthesizer, a 20mer oligonucleotide was synthesized on a 1.0 umol support, with the following sequence using DNA, TBDMS, and TOM phosphoramidites:

*For DNA control, T was used in place of U.

  • Cleavage & Deprotection. Each oligonucleotide was deprotected according to support recommendations.
    • For oligonucleotides synthesized using US III PS, oligonucleotides were incubated in 2M ammonia in methanol for 60 min at RT. Without drying down, an equal volume of AMA was added to the solution, and deprotection was continued for 10 min in a 65 °C water bath. The solution was filtered and dried in a sterile microcentrifuge
      tube using a steady flow of argon to evaporate the liquid.
    • For oligonucleotides synthesized using Glen UnySupport CPG, oligonucleotides were
      incubated in AMA for 60 min in a 65 °C water bath. The solution was filtered and dried in a sterile microcentrifuge tube using a steady flow of argon to evaporate the liquid.
  • 2′-Desilylation. The oligonucleotides were redissolved in DMSO (115 μL) and warmed in a 65 °C water bath until fully dissolved. TEA (60 μL) and TEA•3HF (75 μL) were added to the reaction. The mixture was heated in a 65 °C water bath for 2.5 h. The reaction was quenched by adding 750 μL RNA quenching buffer.
  • Desalting. To remove reaction conditions from the previous step, the oligonucleotide was desalted using a Glen Gel-PakTM 1.0 desalting column. The recommended conditions were followed, and crude oligonucleotides were eluted in 0.1M RNase-free TEAA and analyzed by reverse-phase high-performance liquid chromatography (RP-HPLC).

Figure 2. Crude RP-HPLC chromatograms

Figure 2. Crude RP-HPLC chromatograms of 20mer oligonucleotides synthesized on US III PS with DNA (blue), TOM (orange), and TBDMS (green) monomers. 

Coupling efficiency was calculated based on crude oligonucleotide purity. As expected, DNA phosphoramidite coupling performance was the best of the three followed by TOM phosphoramidites and then TBDMS phosphoramidites (Table 1). Crude purity of TOM- and TBDMS-prepared oligonucleotides synthesized on US III PS were 80.1 and 77.6%, respectively (Figure 2). Based on MS and RP-HPLC analysis of our crude RNA oligonucleotides, we detected a small degree of DMT loss. The full-length DMT-OFF oligonucleotide co-eluted with failure sequences. The loss of the trityl group likely occurred during the drying step necessary for cleavage and deprotection and the 2′-desilylation reaction. This was unavoidable and has implications when it comes to choosing a purification method for your long RNA oligonucleotides. This was a contributing factor to the coupling efficiency differences between DNA and RNA
in Table 1.


Universal Support 


Coupling Efficiency 








UnySupport CPG 




Table 1. Summary of coupling efficiencies

Although only 20mer oligonucleotides were synthesized in this exercise, we can use these values to extrapolate the expected overall purity of a 100mer. We recognize a few caveats with this, namely the fact that failure sequences absorb less at 254nm than the full-length oligonucleotide. It is worth remembering that this extrapolation will predict what the chromatogram looks like, rather than the ultimate yield. Using the following equation, a 100mer made of TOM affords a 33% crude oligonucleotide purity while TBDMS provides a 27% crude oligonucleotide purity.

where CE = coupling efficiency (written as a decimal) and L = length

Unsurprisingly, US III PS offered better results than UnySupport CPG (Table 1, Figure 3). UnySupport requires very harsh conditions to cleave the oligonucleotide from the support. For example, standard nucleobase deprotection using AMA requires 65 °C for 10 min but UnySupport requires AMA at 65 °C for 1 hr for complete cleavage. Using this method, we observed some cleavage that yielded shorter DMT-ON fragments in our crude HPLC (highlighted with the red box). This is particularly concerning considering reverse phase cartridge purification would not be able to distinguish between
the full-length DMT-ON sequence and the DMT-ON cleavage fragments.

Figure 3. Crude RP-HPLC chromatograms of 20mer RNA oligonucleotide synthesized with TBDMS monomers on UnySupport (purple) and US III PS (green). 

Considering we synthesized oligonucleotides that were 20 nucleotides long, one thing we did not evaluate in this study is the impact of pore sizes on the length of oligonucleotides that can be prepared on certain solid supports. Our general rule of thumb for synthesis length is shown below, and the Glen UnySupport CPG used in this experiment was 1000 Å.

  • 500 Å CPG is up to 50mer
  • 1000 Å CPG is compatible with 75-100mer
  • 2000 Å CPG is for >100mer
  • PS is comparable to 1000 Å CPG

Based on our data, we can conclude some key takeaways when it comes to long RNA synthesis and RNA synthesis in general:

TOM offers a slightly higher coupling efficiency than TBDMS, which makes a bigger impact the longer the oligonucleotide length is.

Some loss of the DMT group is unavoidable under all conditions evaluated. This most likely occurs during drying the oligonucleotide down and the 2′-desilylation reaction. If pursuing reverse phase purification techniques, this will negatively impact the final yield of recovery as the DMT-OFF full length is washed away with any failure sequences. Another consideration to take is the finite hydrophobicity of the trityl group becomes less effective at DMT-ON purification as the very polar phosphate backbone increases in length. To circumvent these concerns, ion exchange chromatography or PAGE purification may be better options.

US III PS provides cleaner crude results than that of UnySupport. Some strand cleavage occurs during the harsh conditions required for UnySupport cleavage, which yields DMT-ON fragment sequences. The presence of these fragments may interfere with purification using reverse phase techniques. 


  1. P. Zhao, Z. Zhang, H. Ke, Y. Yue, and D. Xue, Cell Res, 2014, 24, 247-50.
  2. A.E. Briner, et al., Mol Cell, 2014, 56, 333-339.
  3. S. Scaringe, F. Wincott, M. Caruthers, J. Am. Chem. Soc, 1998, 120, 11820-11821.
  4. N. Abe, et al., ACS Chem Biol, 2022, 17, 1308-1314.
  5. L. Statello, C.J. Guo, L.L. Chen, and M. Huarte,
    Nat Rev Mol Cell Biol, 2021, 22, 96-118.
  6. The Glen Report, 2009, 21.2, 14-16.
  7. The Glen Report, 2003, 16.2, 4.

Product Information

Universal Supports
Deblock (40-4040)
Activator (30-3140)
RNA Quenching Buffer (60-4120)
Desalting Columns (61-5010)