Logo of narLink to Publisher's site
Nucleic Acids Res. 2002 Jun 15; 30(12): e56.
PMCID: PMC117298

A universal method to produce in vitro transcripts with homogeneous 3′ ends


A method is described that allows a general drawback of in vitro transcription assays to be overcome: RNA polymerases tend to add extra nucleotides to the RNA 3′ end that are not encoded in the linearized DNA template. Furthermore, these polymerases show a considerable rate of premature termination close to the RNA’s 3′ end. These features lead to a decreased yield of full-length transcripts and often make it difficult to determine and isolate the correctly transcribed full-length RNA. The hammerhead ribozyme is frequently used in cis to cleave off these extra nucleotides. However, the upstream sequence requirements of this ribozyme restrict its general usability. In contrast, the hepatitis delta virus ribozyme has no such requirements and can therefore be applied to any RNA sequence in cis. Due to the catalytic activity of the ribozyme, the desired transcript is released as an RNA molecule with a homogeneous 3′ end. The resulting 2′,3′-cyclo-phosphate group of the released RNA can be easily and efficiently removed by T4 polynucleotide kinase treatment. The presented method can be applied for virtually any sequence to be transcribed and is therefore superior to other ribozyme strategies, suggesting possible applications in every field where transcripts with homogeneous 3′ ends are required.


The in vitro synthesis of RNA molecules is a widely used technique in molecular biology laboratories (13). Usually, the gene of interest is inserted into special vector plasmids carrying a phage promoter (T7, T3, SP6) upstream of the multiple cloning site. These vectors are then linearized by restriction enzymes and used for run-off transcription with the appropriate RNA polymerase. Another frequently used type of run-off transcription template consists of PCR products with a promoter sequence introduced by one of the PCR primers.

While this technique is very useful and the quality of the resulting transcripts sufficient for many applications, it does not meet the requirements for experiments where homogeneity of the transcripts is needed. Since RNA as well as DNA polymerases (such as Taq polymerase) show a tendency to fall off the template before they reach the very end of the DNA (premature termination) or add non-encoded nucleotides to the 3′ end of the nascent nucleic acid, the resulting transcripts usually show highly heterogeneous 3′ ends (35). This fact often makes it impossible to isolate the correct transcript band from a preparative polyacrylamide gel and leads to a dramatically decreased yield of correctly sized RNA (see Fig. Fig.2,2, left).

Figure 2
Comparison of the conventional run-off transcription and the ribozyme mediated generation of homogeneous transcripts on a denaturing 10% polyacrylamide gel. While the run-off transcription leads to a set of products of different length, the ribozyme ...

In order to improve the yield of correctly terminated transcripts and to reduce the amount of incomplete or elongated RNAs, previously described methods use hammerhead or hairpin ribozymes as cis-cleaving autocatalytic cassettes (6). However, due to specific upstream sequence requirements, the applicability of these cassettes is restricted to certain RNA sequences. To overcome this drawback, we established a method that uses an optimized version of the hepatitis delta virus (HDV) ribozyme to generate homogeneous RNA 3′ ends (7,8). The HDV ribozyme is the only known catalytic RNA domain that has almost no prerequisites on the sequence composition upstream of the cleavage position and is therefore perfectly suited for such a strategy (9,10). In our model system, we used a chimpanzee mitochondrial tRNAPhe that should be produced as a homogeneous in vitro transcript.


Construction of HDV-tRNA constructs

The deletion variant of the HDV ribozyme (ΔHDV, 5′-GGG UCG GCA UGG CAU CUC CAC CUC CUC GCG GUC CGA CCU GGG CUA CUU CGG UAG GCU AAG GGA GAA G-3′, 67 nt) was constructed by PCR (Fig. (Fig.1A:1A: construction, 20 cycles of 1 min, 94°C; 1 min, 60°C; 30 s, 72°C) using two primers which overlap at their 3′ terminus for 23 bases (underlined): HDVforward (5′-GGG TCG GCA TGG CAT CTC CAC CTC CTC GCG GTC CGA CCT GGG CTA-3′) and HDVreverse (5′-CTT CTC CCT TAG CCT ACC GAA GTA GCC CAG GTC GGA CCG CGA GGA-3′). A second PCR reaction (2 min, 94°C; 40 cycles of 1 min, 94°C; 1 min, 56°C; 30 s, 72°C) was performed to amplify the construct of the first PCR, introducing an elongated downstream primer (Fig. (Fig.1A,1A, amplification): HDVamp/forward (5′-GGG TCG GCA TGG CAT CTC CAC-3′) and HDVext/reverse (5′-AAA CGA CGG CCA GTG CCA AGC TTC TCC CTT AG-3′). This elongation does not influence the reactivity of the ribozyme, but makes it easier to distinguish between the bands of the released tRNA transcript of interest (70 nt) and the HDV ribozyme (87 nt) on an acrylamide gel.

Figure 1
Scheme showing the construction of PCR products consisting of an RNA sequence of interest attached to the HDV ribozyme sequence. (A) Construction of the HDV ribozyme sequence starting with an overlapping primer pair. (B) Amplification of the RNA sequence ...

PCR amplification of the tRNA substrate (2 min, 94°C; 30 cycles: 1 min, 94°C; 1 min, 45°C; 1 min, 72°C) was performed using the following primers (Fig. (Fig.1B;1B; tRNA/HDV crossover intermediate): tRNA Phe-T7 (5′-GGA GAT CTA ATA CGA CTC ACT ATA GTT TAT GTA GCT TAC C-3′) and tRNA Phe HDV reverse (5′-TGG AGA TGC CAT GCC GAC CCT GTT TAT GGG GTG-3′). The resulting PCR product had a length of 115 bp.

The final tRNA/HDV crossover construct (182 bp) was created using an overlap extension PCR assay (11,12) (Fig. 1C; tRNA/HDV crossover end product). PCR products A and B were incubated in a PCR assay without addition of primers (4 min, 94°C; 10 cycles: 1 min, 94°C; 2 min, 40°C; 45 s, 72°C). After addition of the primer pair (tRNA Phe-T7 and HDVext/reverse) a standard PCR was performed to amplify the final construct consisting of T7 promoter, tRNAPhe gene and HDV ribozyme sequence.

For the construction of transcription templates containing A, G or T residues immediately upstream of the HDV sequence, the procedure steps B and C (Fig. (Fig.1)1) were repeated using the appropriate tRNA Phe HDV reverse primers.


Transcription using T7 RNA polymerase (New England Biolabs, Schwalbach, Germany) was carried out in the presence of [a-33P]UTP according to the supplier’s instructions. Transcripts were separated on a 10% polyacrylamide gel containing 8 M urea and visualized by autoradiography. The band corresponding to the tRNA released by the HDV ribozyme self-cleavage reaction was cut out with a sterile blade. tRNAs were eluted by incubation in 500 mM ammonium acetate, pH 5.7, 0.1 mM EDTA, 0.1% SDS at 4°C overnight and ethanol precipitated (13).

Dephosphorylation of RNA 3′ ends

Radioactively labeled transcripts carrying a 2′,3′ cyclic phosphate at the 3′ end (as a consequence of the self-cleavage reaction of the downstream HDV ribozyme) were dephosphorylated using either of the following methods. (i) Up to 50 pmol RNA were incubated with 6 U T4 polynucleotide kinase (New England Biolabs) in 100 mM Tris–HCl pH 6.5, 100 mM magnesium acetate, 5 mM β-mercaptoethanol in a final volume of 50 µl for 6 h at 37°C (14). (ii) Transcripts were incubated with 0.1 mM ATP, 100 mM imidazole–HCl pH 6.0, 10 mM MgCl2, 10 mM β-mercaptoethanol, 20 µg/ml BSA and 1 U T4 polynucleotide kinase (New England Biolabs) per 100 pmol RNA in a final volume of 50 µl for 6 h at 37°C (15). (iii) Up to 300 pmol tRNA were incubated in 100 mM morpholinoethanesulfonate-NaOH pH 5.5, 10 mM MgCl2, 10 mM β-mercaptoethanol, 300 mM NaCl and 10 U T4 polynucleotide kinase (New England Biolabs) in a final volume of 20 µl for 6 h at 37°C (16).

All assays were desalted on microspin G25 columns according to the manufacturer (Amersham Pharmacia Biotech, Piscataway, USA), subjected to phenol/chloroform extraction, ethanol precipitated and dissolved in water.

The efficiency of dephosphorylation was controlled by analyzing 1 pmol RNA on a 10% polyacrylamide gel containing 8 M urea. The removal of the 2′,3′ cyclic phosphate leads to a reduction in the net charge of the RNA and results therefore in a decrease in mobility which can be easily observed (see Fig. Fig.4)4) (16).

Figure 4
Removal of the terminal 2′,3′ cyclic phosphate group of the released tRNA resulting from the HDV ribozyme cleavage reaction. Treatment with T4 polynucleotide kinase (T4 PNK) leads to the removal of the phosphate group and therefore to ...

3′-Sequence analysis

Isolated tRNAs were ligated to an EcoRI linearized pUC18 restriction fragment using T4 RNA ligase. Subsequent cDNA synthesis and PCR amplification were performed as described (17). The PCR products were cloned using a TA cloning kit according to the manufacturer’s instructions (Invitrogen, Groningen, The Netherlands). DNA sequencing was carried out on an ABI Prism 377 automated sequencer.


In order to achieve the highest catalytic activity of the HDV ribozyme for an optimal yield of the desired transcript, a slightly modified version containing a shortened form of stem 4 was used (7,8). To generate the transcription template encoding the desired RNA sequence followed by the HDV ribozyme, an overlap extension PCR technique was applied (Fig. (Fig.1).1). The resulting tRNA/HDV PCR product was used for run-off transcription and the obtained RNA was separated on a 10% polyacrylamide gel and visualized by autoradiography (Fig. (Fig.2,2, right). The figure shows three major bands appearing in the gel, representing the precursor transcript, and the two cleavage products consisting of the released RNA molecule and the free HDV ribozyme. This result indicates that the ribozyme is active during transcription, leading to a highly efficient release of the RNA of interest, a fact that renders additional time-consuming incubations for the cleavage reaction unnecessary. Furthermore, compared with the usual run-off transcription technique which leads to heterogeneous 3′ termini (Fig. (Fig.2,2, left), the ribozyme reaction leads to a released RNA of interest that has a defined length, as indicated by the main product band in the autoradiography, while side products are represented only by very faint bands (Fig. (Fig.2,2, right). Sequence analysis of the RNA 3′ end according to Hetzer and Mueller (17) confirmed that the released RNA had the correct size and carried the expected 3′ terminus, whereas the amount of side products was below the detection level.

Therefore, the use of the HDV ribozyme leads not only to a facilitated isolation of correctly sized transcripts, but also to an increase in yield, since nearly no side products with incorrect 3′ ends appear.

While there is no virtual requirement for a specific sequence upstream of the ribozyme, several authors report on base preferences 5′ of the cleavage position by the HDV domain (9,18). However, when we replaced the base upstream of the cleavage position and compared the cleavage efficiency of all four variants as a ratio between the signal intensities of the released products and the remaining precursor transcript on a denaturing 10% polyacrylamide gel, we observed no dramatic differences (Fig. (Fig.3):3): the naturally occurring C residue led only to a marginally increased cleavage efficiency (79.0%) in comparison with A (71.2%), U (75.3%) or G (75.3%). Furthermore, alternative upstream tRNAs that were tested in our laboratory showed consistently similar cleavage efficiencies as the example presented in this study.

Figure 3
Influence of the nucleotide immediately upstream of the cleavage position. The calculated cleavage efficiencies as the ratio between the signal intensities of the released tRNAPhe and the remaining precursor molecules show that all four nucleotides are ...

Similar to the hammerhead cleavage, the HDV ribozyme reaction leads to a released RNA of interest that carries a 2′,3′ cyclic phosphate group at the 3′ end. Such a 3′ terminus interferes with many functional aspects of RNA molecules. For instance, such a transcript cannot be used for ligation reactions or base incorporation assays. Furthermore, tRNA molecules carrying a 2′,3′ cyclic phosphate group are no longer substrates for aminoacyl tRNA synthetases and can consequently not be used in aminoacylation reactions. Therefore, in order to render these transcripts functional, the interfering cyclic phosphate group has to be removed. In the literature, several techniques are described that use a 3′-phosphatase activity of T4 polynucleotide kinase—an enzyme that is known to add a phosphate group to 5′-hydroxyl ends of both RNA and DNA (1416,19). In order to find a suitable and convenient method to remove the 2′,3′ cyclic phosphate group, three different protocols with the released transcripts of interest were tested. All methods turned out to work well with the used transcripts and led to a complete removal of the terminal 2′,3′ cyclic phosphate group, a reaction that can be monitored by the reduction of the electrophoretic mobility of the RNA on a denaturing 10% polyacrylamide gel due to the loss of negative charges (Fig. (Fig.44).

In order to prove the functionality of this 3′-dephosphorylated tRNA, the isolated transcript was tested for its ability to function as a substrate for CCA addition by ATP(CTP):tRNA nucleotidyltransferase. Almost 100% of the T4 polynucleotide kinase treated tRNA showed CCA incorporation, indicating that the dephosphorylation of the 3′ end led to a functional RNA molecule.

Taken together, these results demonstrate that the utilization of the HDV ribozyme allows the generation of RNAs of defined length carrying functional 3′ termini without the usual side reaction products due to premature termination or incorporation of non-encoded extra nucleotides. Furthermore, the 2′,3′ cyclic phosphate group at the RNA terminus as a result of the ribozyme cleavage reaction does not limit the usefulness of the method, since it can be removed efficiently using the 3′-phosphatase activity of T4 polynucleotide kinase. While the usability of other ribozymes like hammerhead or hairpin is restricted to certain RNA sequences, the HDV ribozyme has no such constraints, but can be applied to virtually any transcription assay where homogeneous RNA 3′ ends are required. Examples for such applications are RNA crystallization and NMR studies. Whereas homogeneous RNA preparations tend to crystallize under several conditions, the use of heterogeneous RNA samples often implies problems due to different conformations in the RNA subpopulations (20). Furthermore, heterogeneous RNA preparations may cause unwanted quenching effects in binding studies using fluorescence labeled RNA (21).

Therefore, this method should be useful for many different biological, biochemical and biophysical applications where homogeneous in vitro transcripts are a prerequisite for successful experiments.


We thank Roland Hartmann, Hanna Mörl and Andreas Reichert for critical discussion. Expert technical assistance by Kathrin Köhler is greatly appreciated. This work was financed by the Max-Planck Society.


1. Chamberlin M.J. and Ryan,T. (1982) Bacteriophage DNA-dependent RNA polymerases. In Boyer,P.D. (ed.), The Enzymes. Academic Press, New York, Vol. 15B, pp. 87–109.
2. Krieg P.A. and Melton,D.A. (1987) In vitro RNA synthesis with SP6 RNA polymerase. Methods Enzymol., 155, 397–415. [PubMed]
3. Milligan J.F. and Uhlenbeck,O.C. (1989) Synthesis of small RNAs using T7 RNA polymerase. Methods Enzymol., 180, 51–62. [PubMed]
4. Draper D.E., White,S.A. and Kean,J.M. (1988) Preparation of specific ribosomal RNA fragments. Methods Enzymol., 164, 221–237. [PubMed]
5. Kholod N., Vassilenko,K., Shlyapnikov,M., Ksenzenko,V. and Kisselev,L. (1998) Preparation of active tRNA gene transcripts devoid of 3′-extended products and dimers. Nucleic Acids Res., 26, 2500–2501. [PMC free article] [PubMed]
6. Shub D.A., Peebles,C.L. and Hampel,A. (1994) Ribozymes. In Higgins,S.J. and Hames,B.D. (eds.), RNA Processing: A Practical Approach. IRL Press, Oxford, Vol. 2, pp. 211–239.
7. Chowrira B.M., Pavco,P.A. and McSwiggen,J.A. (1994) In vitro and in vivo comparison of hammerhead, hairpin and hepatitis delta virus self-processing ribozyme cassettes. J. Biol. Chem., 269, 25856–25864. [PubMed]
8. Been M.D., Perrotta,A.T. and Rosenstein,S.P. (1992) Secondary structure of the self-cleaving RNA of hepatitis delta virus: applications to catalytic RNA design. Biochemistry, 31, 11843–11852. [PubMed]
9. Been M.D. (1994) Cis- and trans-acting ribozymes from a human pathogen, hepatitis delta virus. Trends Biochem. Sci., 19, 251–256. [PubMed]
10. Perrotta A.T. and Been,M.D. (1992) Cleavage of oligoribonucleotides by a ribozyme derived from the hepatitis delta virus RNA sequence. Biochemistry, 31, 16–21. [PubMed]
11. Landt O., Grunert,H.-P. and Hahn,U. (1990) A general method for rapid site-directed mutagenesis using the polymerase chain reaction. Gene, 96, 125–128. [PubMed]
12. Horton R.M., Cai,Z.L., Ho,S.N. and Pease,L.R. (1990) Gene splicing by overlap extension: tailor-made genes using the polymerase chain reaction. Biotechniques, 8, 528–535. [PubMed]
13. Peattie D.A. (1979) Direct chemical method for sequencing RNA. Proc. Natl Acad. Sci. USA, 76, 1760–1764. [PMC free article] [PubMed]
14. Royer-Pokora B., Gordon,L.K. and Haseltine,W.A. (1981) Use of exonuclease III to determine the site of stable lesions in defined sequences of DNA: the cyclobutane pyrimidine dimer and cis and trans dichlorodiammine platinum II examples. Nucleic Acids Res., 9, 4595–4609. [PMC free article] [PubMed]
15. Busch S., Kirsebom,L.A., Notbohm,H. and Hartmann,R.K. (2000) Differential role of the intermolecular base-pairs G292-C(75) and G293-C(74) in the reaction catalyzed by Escherichia coli RNase P RNA. J. Mol. Biol., 299, 941–951. [PubMed]
16. Povirk L.F. and Steighner,R.J. (1990) High ionic strength promotes selective 3′-phosphatase activity of T4 polynucleotide kinase. Biotechniques, 9, 562. [PubMed]
17. Hetzer M. and Mueller,M.W. (1993) PCR mediated analysis of RNA sequences. Nucleic Acids Res., 21, 5526–5527. [PMC free article] [PubMed]
18. Wu H.N. and Huang,Z.S. (1992) Mutagenesis analysis of the self-cleavage domain of hepatitis delta virus antigenomic RNA. Nucleic Acids Res., 20, 5937–5941. [PMC free article] [PubMed]
19. Cameron V. and Uhlenbeck,O.C. (1977) 3′-Phosphatase activity in T4 polynucleotide kinase. Biochemistry, 16, 5120–5126. [PubMed]
20. Holbrook S.R., Holbrook,E.L. and Walukiewicz,H.E. (2001) Crystallization of RNA. Cell Mol. Life Sci., 58, 234–243. [PubMed]
21. Pfeiffer T., Busch,S., Feltens,R., Heide,C., Held,R., Warnecke,J.M. and Hartmann,R.K. (1998) In vitro-Studien an RNA-Molekülen-das Problem homogener und authentischer Enden. Biospektrum, 6, 29–34.

Articles from Nucleic Acids Research are provided here courtesy of Oxford University Press
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...