to BioTechniques free email alert service to receive content updates.
Generation of genetic constructs that simultaneously express several shRNAs
Olga V. Kretova, Ildar R. Alembekov, and Nickolai A. Tchurikov
Department of Genome Organization, Engelhardt Institute of Molecular Biology, Russian Academy of Sciences, Moscow, Russia
BioTechniques, Vol. , No. , May 2012, pp. 1–3
Full Text (PDF)
Method Summary

Stable intramolecular hairpins often hamper the generation of genetic constructs containing short hairpin RNAs (shRNAs). Here we present a method that overcomes this problem through the use of specifically designed oligonucleotides and PCR amplification.


RNAi has potential as an antiviral gene therapy strategy. Cassette constructs simultaneously expressing several siRNAs could prove to be the most efficient technique in developing gene therapy approaches for highly mutable viruses such as HIV-1. Here we describe a rapid and cost-saving protocol to generate cassettes that simultaneously express three siRNAs for repression of HIV-1 and CCR5 transcripts. siRNA biological activity was tested in a non-viral system, and exhibited both efficiency and specificity. Our results suggest this protocol can be used to rapidly generate cassette constructs for antiviral gene therapy applications.

Genetic constructs that express several siRNAs simultaneously are more effective in RNAi-mediated gene therapy compared with those that express only one siRNA (1). This so-called combinatorial RNA interference (co-RNAi) is especially important for treating highly mutable viruses such as HIV-1 because they tend to escape from RNAi (2). There are three main strategies for using co-RNAi in antiviral gene therapy: (1) multiple promoter/shRNA cassettes; (2) long hairpin RNAs; and (3) microRNA-embedded shRNAs (3-5). In this paper we describe a protocol for generation of shRNA cassettes that express several shRNAs under a single promoter in human cultured cells. The cassettes were cloned into the pGeneClip-U1 vector and used to cleave HIV-1 transcripts in cultured human cells. In contrast to other methods (6), we did not use step-by-step cloning, but instead generated the whole cassette construct at once, resulting in a considerable time savings.

Preparation of perfect double-stranded DNA from a mixture of partially overlapping DNA oligos containing several palindromes is hampered by the formation of rather stable intramolecular hairpins that serve as strong stops for the Klenow fragment of Escherichia coli DNA polymerase I or filling by other enzymes. This drastically inhibits the cloning efficiency of full-size double-stranded DNA. Using two partially overlapping oligos of more than 100 nucleotides in length allows researchers to avoid this problem, but such long oligos often contain mistakes and their synthesis is expensive. The protocol described herein, which uses specifically designed shorter oligos, is less costly. It uses pairs of DNA oligos that overlap only by different sequences of loops between palindromes, thus preventing the formation of intramolecular hairpins. Another feature of this protocol is different restriction sites at the ends of each element of the cassette, which allows for predetermined assembly of the cassette.

We generated a genetic construct that expresses one anti-CCR5 and two anti-HIV-1 shRNAs for processing into three different siRNAs in cells (Figure 1). Figure 1A shows the design of the oligos and the main stages of the protocol for generating the expression cassette for the three shRNAs. Figure 1B shows the sequences of the DNA oligos that correspond to each shRNA. The single-stranded regions contain the same 5'–3' sequences and cannot form duplexes. Thus, the three pairs of oligos that overlap only by sequences in their loops may be separately annealed and then treated with the Klenow fragment of E. coli DNA polymerase I.

After digestion with the indicated restriction enzymes, the DNA fragments possessing the 5' protruding end can be ligated into a full-length linear DNA. This DNA may be enriched by PCR amplification using primers corresponding to the termini of the full-length DNA, fractionated in a 2% agarose gel, eluted, and subjected to a second round of amplification using the same pair of primers. The final amplified DNA is then cloned into the pGEM-T Easy Vector for sequencing and finally subcloned into the pGeneClip-U1 expression vector (Figure 1A).

Specifically, our protocol consists of the following steps: 1. oligonucleotide (oligo) design and synthesis; 2. annealing of the oligonucleotides in pairs; 3. filling-in pairs of annealed oligos using the Klenow fragment of E. coli DNA polymerase; 4. ethanol precipitation of the separate fragments of the whole cassette followed by digestion with appropriate restriction endonucleases in order to create cohesive ends; 5. ligation of the separate DNA fragments into one linear molecule; 6. two consecutive PCR amplifications, each followed by fractionation into 2% agarose gels and elution of full-length DNA; 7. cloning into pGEM-T Easy Vector using positive clone selection by colony hybridization; 8. sequencing; 9. subcloning into the pGeneClip-U1 expression vector.

Stages 2–4 of this protocol take 2–4 days; stages 5 and 6 take 1–2 days; and stages 7–8 take 3–4 days. Thus it is possible to create several constructs in the intermediate pGEM-T Easy Vector within 5–7 days. Subcloning into the expression vector normally takes 3–4 days. Overall, several cassette constructs in an expression vector can be generated within a week. We sequenced five random pGEM-T Easy clones selected by the colony hybridization procedure and found a single mutation inside the loop in just one of these clones, indicating that mutations arising during the two rounds of PCR in the suggested procedure are rare.

Several cassettes expressing the three shRNAs were constructed following this protocol, and tested in a non-viral system (4). The results showed effective expression of each shRNA, as illustrated by effective inhibition of the separate targets for all siRNAs (Figure 2A). For this experiment, the artificial targets (domains), which were cloned into the 3' region of the Renilla luciferase gene inside the psiCHECK-2 Vector (Promega, Madison, WI, USA), were used (7-9). The inhibition of separate targets by the siRNAs that were expressed by cassette 1 was rather high; we observed about 75% repression by anti-HIV-1 siRNA, 77% by anti-HIV-2 siRNA, and 53% by anti-CCR5 siRNA. The same inhibition was observed for these targets using constructs that expressed single siRNAs under the U6-promoter (7). t-test p-values shown in Figure 2A indicate the statistical significance of the knockdown seen from each siRNA. As expected, we did not observe differences in similar experiments using the constructs expressing unrelated siRNAs (Figure 2B). The formation of siRNAs in cells transfected by the cassette generated by this protocol was confirmed by an RNase protection assay (Figure 2C), which was performed as described before (10).

These results illustrate that working cassette constructs can be generated using our protocol. The cassettes effectively expressed three different shRNAs under a single promoter and the siRNAs exhibited specificity and biological activity. Our protocol allows the use of synthetic DNA oligos of less than 50 nucleotides in length and generates the whole cassette at one time (in contrast to step-by-step cassette creation). Thus this protocol offers both speed and economy.


This work was supported by the Russian Ministry of Science and Education (contracts 17.740.11.0252 and 16.512.11.2058), CRDF grant RUB2-2960-MO-09, and RFBR grant 11-04-00091-a. We thank Dr. Evgenia Moiseeva for technical support with cassette generation and Maria Gorbacheva for technical assistance.

Competing interests

The authors declare no competing interests.

Address correspondence to Olga V. Kretova, Department of Genome Organization, Engelhardt Institute of Molecular Biology, Russian Academy of Sciences, Moscow, Russia. Email: [email protected]">[email protected];[email protected]">[email protected]

1.) Castanotto, D., and J.J. Rossi. 2009. The promises and pitfalls of RNA-interference-based therapeutics. Nature 457:426-433.

2.) Boden, D., O. Pusch, and B. Ramratnam. 2007. Overcoming HIV-1 resistance to RNA interference. Front. Biosci. 12:3104-3116.

3.) Lambeth, L.S., N.J. Van Hateren, S.A. Wilson, and V. Nair. 2010. A direct comparison of strategies for combinatorial RNA interference. BMC Mol Biol. 11:11-77.

4.) Liu, Y.P., K.J. von Eije, N.C. Schopman, J.T. Westerink, O. ter Brake, J. Haasnoot, and B. Berkhout. 2009. Combinatorial RNAi against HIV-1 using extended short hairpin RNAs. Mol. Ther. 17:1712-1723.

5.) Liu, Y.P., J. Haasnoot, O. ter Brake, B. Berkhout, and P. Konstantinova. 2008. Inhibition of HIV-1 by multiple siRNAs expressed from a single microRNA polycistron. Nucleic Acids Res Published online 2008 Mar 16 36:2811-2824.

6.) Xu, X.-M., M.-H. Yoo, B.A. Carlson, V.N. Gladyshev, and D.L. Hatfield. 2009. Simultaneous knockdown of the expression of two genes using multiple shRNAs and subsequent knock-in of their expression. Nat Protoc. 4:1338-1348.

7.) Alembekov, I.R., O.V. Kretova, and N.A. Tchurikov. 2011. Analysis of genetic constructs expressing antiviral siRNAs in the non-viral test system. Vopr. Virusol. Russian 56:32-35.

8.) Tchurikov, N.A., O.V. Kretova, N.M. Gashnikova, and A.G. Pokrovskii. Patent RF No 2385939 “Genetic constructions, which attack six new targets of RNAi in human HIV-1 transcripts and repressing the virus reproduction in human cells”.

9.) Tchurikov, N.A., O.V. Kretova, N.M. Gashnikova, and A.G. Pokrovskii. Patent RF No 2425150 “ The cassette genetic construction expressing three biologically active siRNAs, which effectively attack human HIV-1 and CCR5 gene transcripts with the use of RNAi”.

10.) Tchurikov, N.A., and O.V. Kretova. 2011. Both piRNA and siRNA pathways are silencing transcripts of the Suffix element in the Drosophila melanogaster germline and somatic cells. PLoS One Published online 2011 Jul 14 6:e21882.