Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Use of adeno-associated viral vector for delivery of small interfering RNA


Post-transcriptional gene silencing by small interfering RNAs (siRNAs) is rapidly becoming a powerful tool for genetic analysis of mammalian cells. Delivery of siRNA into mammalian cells is usually achieved via the transfection of double-stranded oligonucleotides or plasmids encoding RNA polymerase III promoter-driven small hairpin RNA. Recently, retroviral vectors have been used for siRNA delivery, which overcome the problem of poor transfection efficiency seen with the plasmid-based systems. However, retroviral vectors have several limitations, such as the need for active cell division for gene transduction, oncogenic potential, low titers and gene silencing. In this report, we have adapted a commercially available adenoassociated virus (AAV) vector for siRNA delivery into mammalian cells. We demonstrate the ability of this modified vector to deliver efficiently siRNA into HeLa S3 cells and downregulate p53 and caspase 8 expression. Our results suggest that AAV-based vectors are efficient vectors for the delivery of siRNA into mammalian cells. Based on the known ability of these vectors to infect both dividing and nondividing cells, their use as a therapeutic tool for the delivery of siRNA deserves further study.


RNA interference (RNAi) is an evolutionarily conserved mechanism of sequence-specific post-transcriptional gene silencing mediated by double-stranded (ds) RNA molecules that match the sequence of the target gene (Sharp, 1999,2001). RNAi has been successfully used in plants and invertebrates for genetic analysis (Hamilton and Baulcombe, 1999; Hammond et al., 2000). Initial attempts to exploit this phenomenon for gene silencing in mammalian cells failed due to activation by long dsRNA molecules of dsRNA-dependent kinase, which led to nonspecific translational suppression (Manche et al., 1992; Stark et al., 1998). This problem was overcome by the recent discovery that, unlike long dsRNA, transfection of short 21-nucteotide dsRNA duplexes, termed small interfering RNA (siRNA), into mammalian cells effectively inhibits endogenous genes in a sequence-specific manner (Elbashir et al., 2001). Since this initial report, post-transcriptional gene silencing by siRNAs has quickly emerged as a powerful tool for genetic analysis in mammalian cells and has the potential to produce novel therapeutics. Unfortunately, oligonucleotides-mediated siRNA suffers from several limitations, such as high cost, nonspecific toxicity during transfection and transient suppression of gene expression. To overcome these problems, several groups have recently described alternative systems for the expression of siRNA within mammalian cells that rely on DNA vectors (Brummelkamp TR et al., 2002; Lee et al., 2002; Miyagishi and Taira, 2002; Paddison et al., 2002; Paul et al., 2002; Sui et al., 2002). In general, these DNA vectors use an RNA polymerase III (Pol III) promoter to express short dsRNA in the form of an inverted repeat sequence containing a hairpin loop. Such small hairpin RNAs were shown to be efficiently processed into siRNAs inside the cells and achieved sequence-specific gene silencing (Brummelkamp TR et al., 2002; Lee et al., 2002; Miyagishi and Taira, 2002; Paddison et al., 2002; Sui et al., 2002). In order to overcome the problem of poor transfection efficiency with plasmid-based vectors, several groups subsequently developed retroviral-based vectors for the delivery of siRNA into mammalian cells (Barton and Medzhitov, 2002; Brummelkamp T et al., 2002; Devroe and Silver, 2002). Unfortunately, these vectors suffer from the limitations inherent to oncoretroviral vectors, such as inability to infect nondividing cells, silencing or position effects on gene expression and low titers. Thus, there is a need for continued development of improved vectors for the delivery of siRNA into mammalian cells.

Adenoassociated virus (AAV) is currently being tested in several human gene therapy trials because of its several unique features that distinguish it from other gene therapy vectors (Stilwell and Samulski, 2003). These include (i) ability to infect both dividing and nondividing cells; (ii) a broad host range; (iii) wild-type AAV has never been associated with any disease and cannot replicate in infected cells; (iv) lack of cell-mediated immune response against the vector and (v) ability to integrate into a host chromosome or persist episomally, thereby creating potential for long-term expression (Stilwell and Samulski, 2003). Moreover, recent DNA microarray studies have shown that infection with AAV vectors has minimal influence on changing the pattern of cellular gene expression (Stilwell and Samulski, 2003).

AAVs require coinfection with a helper adenovirus or herpesvirus for productive infection. In order to develop an AAV-based vector for siRNA delivery, we took advantage of the AAV Helper free system (Stratagene), which allows the production of infectious recombinant human AAV-2 virions without the need of a helper virus. The wild-type AAV-2 genome consists of two inverted terminal repeats (ITR) that flank the viral rep (replication) and cap (capsid) genes, respectively. In the AAV Helper free system, the rep and cap genes have been removed from the viral vector and are provided in trans on the plasmid pAAV-RC. A second plasmid, pHelper, provides the adenovirus gene products required for the production of infective AAV particles, such as E2A, E4 and VA RNA genes. The remaining adenoviral gene products are provided by the HEK293 cells that stably express adenovirus E1 gene.

We created a modified AAV vector for the delivery of p53 siRNA, designated pH1-si-p53/AAV-GFP. As shown in Figure 1, this vector expresses enhanced green fluorescent protein (EGFP) under the control of CMV immediate-early promoter and a hairpin siRNA against p53 under the control of H1 RNA promoter. A control vector containing an irrelevant siRNA was constructed similarly and designated pH1-si-cont/AAV-GFP. Recombinant AAV vectors encoding siRNAs were generated by transfection of 293 cells with the pH1-si-p53/AAV-GFP or pH1-si-cont/AAV-GFP vectors along with pAAV-RC and pHelper plasmids. Infectious viral particles were collected from cell lysates and growth media 72 h post-transfection and used to infect HeLa S3 cells. Transduction efficiency was analysed by examining the cells under a fluorescent microscope 72 h after infection. As shown in Figure 2a, we observed EGFP expression in nearly 100% of infected cells after only a single round of infection with the unconcentrated virus. Furthermore, we did not observe any detrimental effect of infection on the morphology of transduced cells. Collectively, the above results suggest that recombinant AAV vectors containing the siRNA expression cassette can be packaged efficiently and can be used to infect successfully the target cells at high frequency and with minimal toxicity.

Figure 1

Schematic representation of pH1-si-p53/AAV-GFP vector. The vector expresses the hairpin form of p53 siRNA under the control of human H1 RNA Pol III promoter and an EGFP cDNA under the control of CMV immediate-early promoter (pCMV). This vector was constructed in the following steps. A cDNA encoding the EGFP was inserted into the multicloning site of pCMV-MCS shuttle vector (Stratagene) to generate pCMV-GFP vector. Second, restriction sites for EcoRI and BamHI were created immediately upstream of the CMV promoter in the pCMV-GFP vector. Third, an H1 promoter-driven expression cassette encoding a hairpin siRNA against p53 was inserted in the above vector upstream of the CMV promoter using the EcoRI and BamHI sites. The resulting vector was termed pH1-si-p53/CMV-GFP. Finally, a NotI fragment containing the H1 promoter-driven siRNA and CMV promoter-driven EGFP was cut out from the pH1-si-p53/CMV-GFP vector and inserted between two AAV-2-derived ITRs present in the pAAV-MCS vector (Stratagene). The sequence of the oligonucleotides encoding the p53 hairpin (top strand) was 5′GACTCCAGTGGTA ATCTACTTCAAGAGAGTAGATTACCACTGGAGTCTTTTT3′. The underlined sequence corresponds to nucleotides 775–793 of human p53 open reading frame. Additional details of cloning strategy are available from the authors on request. hGH-pA, human growth hormone polyA signal; ITR, inverted terminal repeats derived from AAV-2 virus; pUC ori, pUC origin of replication

Figure 2

Downregulation of p53 expression by AAV-delivered siRNA. (a) High-efficiency gene transduction by si-p53 encoding AAV vectors. HEK293 cells were cotransfected with pH1-si-p53/AAV-GFP, pAAV-RC and pHelper plasmids and recombinant adenoviral particles collected 3 days post-transfection according to the instructions of the manufacturer. Virus stock was frozen at −80°C prior to use. HeLa S3 cells were plated at a density of 1.5 × 105 cells in each well of a 12-well plate and next day infected with 0.5 ml of virus-containing supernatant. At 72 h postinfection, cells were examined under a fluorescent microscope to visualize the expression of GFP and photographed. (b) Suppression of p53 expression in HeLa S3 cells by AAV-delivered siRNA. HeLa S3 cells were infected with pH1-si-cont/AAV-GFP and pH1-si-p53/AAV-GFP viruses, respectively. Cellular lysates were prepared 48 h after infection. Western blot analysis demonstrates significant downregulation of p53 expression in pH1-si-p53/AAV-GFP-infected cells. Blot was reprobed with an antibody against actin to show equal loading of all lanes and to serve as a specificity control. Primary antibodies against p53 (clone DO-1) and actin were obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA)

We were next interested in analysing the effect of AAV-mediated siRNA delivery on the expression of the target protein. For this purpose, we fractionated an equal amount of total protein by SDS–PAGE and probed the blot with monoclonal antibodies to p53 and actin, respectively. As shown in Figure 2b, Western blot demonstrated near-complete downregulation of p53 expression in cells infected with the pH1-si-p53/AAV-GFP virus as compared to those infected with the control virus. However, there was no significant difference in the expression of actin between the two cell populations, thereby suggesting that the observed downregulation of p53 expression was a gene-specific effect.

We also generated an AAV vector encoding an siRNA against caspase 8. Initial experiments revealed that expression of caspase 8 siRNA using the H1 RNA promoter was ineffective in downregulating its expression (Figure 3a). As such, we constructed a modified AAV vector in which the expression of caspase 8 hairpin siRNA is being driven by human U6 promoter. This construct also contained the first 27 nucleotides of U6 snRNA to direct methylation of the 5′-γ-phosphate and stabilize the transcript (Paul et al., 2002). We generated infectious recombinant AAV virions in 293 cells and used 100, 200 and 500 μl of the unpurified viral supernatant to infect HeLa S3 cells. As shown in Figure 3b, we observed a dose-dependent decrease in caspase 8 expression in the infected cells, whereas no difference in the expression of actin was observed.

Figure 3

Downregulation of caspase 8 expression by AAV-delivered siRNA. (a) AAV vector containing H1 promoter-driven caspase 8 siRNA fails to achieve gene silencing. HeLa S3 cells were infected with pH1-si-cont/AAV-GFP and pH1-si-caspase 8/AAV-GFP viruses, respectively. Cellular lysates were prepared 24 and 48 h after infection and used for the detection of caspase 8 and actin expression by Western blot. Primary antibody against caspase 8 was obtained from Pharmingen. The sequence of oligonucleotides encoding caspase 8 hairpin (top strand) was GTTCCTGAGCCTGGACTACTTCAAGAGAGTAGTCCAG GCTCAGGAACTTTTT. (b) AAV vector containing U6 promoter-driven caspase 8 siRNA achieves downregulation of caspase 8 expression in a dose-dependent fashion. HeLa S3 cells were infected with the indicated volumes of pU6-siCasp8/AAV-GFP viral supernatants. Cellular lysates were prepared 48 h after infection and used for the detection of caspase 8 and actin by Western blot. The U6 promoter used in this vector contained nucleotides −265 to +27 of human U6 snRNA upstream promoter (Paul et al., 2002). The sequence of oligonucleotides encoding the caspase 8 hairpin (top strand) was as described above

Recombinant AAV has shown promise as a vector for the delivery of genes to treat several human diseases, such as hemophilia A and B, cystic fibrosis and Parkinson's disease (Stilwell and Samulski, 2003). Our results suggest that these vectors also have the potential for the therapeutic delivery of siRNA. Although we achieved significant downregulation of p53 expression using the H1 promoter, we failed to downregulate caspase 8 expression using a similar construct. HeLa S3 cells are known to express a relatively high level of caspase 8, which may have contributed to the failure of the H1 promoter-driven caspase 8 siRNA to achieve effective gene silencing. In contrast to the H1 promoter construct, we were able to downregulate successfully caspase 8 expression using a U6 promoter construct. While the exact reason for the difference in the abilities of H1- and U6-promoter-driven caspase 8 siRNA constructs is currently under study, it is unlikely due to the difference in the titers of the two virus preparations, as we observed nearly 100% GFP-positive cells in both cases. It is conceivable that the relatively higher level expression from the U6 promoter might have contributed to this effect.

Alternatively, the presence of the first 27 nucleotides of the U6 snRNA might have stabilized the caspase 8 siRNA transcript produced by the pU6-siCasp8/AAV-GFP construct. Our results further suggest that the degree of gene silencing achieved with Pol III-based siRNA vectors may depend on the titer of the vector and multiplicity of infection. The use of AAV-based vectors is advantageous in this regard, as it is possible to generate preparations of these viruses with very high titer (1012 viral particles/ml after concentration of the primary virus stock).


  1. Barton GM and Medzhitov R . (2002). Proc. Natl. Acad. Sci. USA, 99, 14943–14945.

  2. Brummelkamp T, Bernards R and Agami R . (2002). Cancer Cell., 2, 243.

  3. Brummelkamp TR, Bernards R and Agami R . (2002). Science, 296, 550–553.

  4. Devroe E and Silver PA . (2002). BMC Biotechnol., 2, 15.

  5. Elbashir SM, Harborth J, Lendeckel W, Yalcin A, Weber K and Tuschl T . (2001). Nature, 411, 494–498.

  6. Hamilton AJ and Baulcombe DC . (1999). Science, 286, 950–952.

  7. Hammond SM, Bernstein E, Beach D and Hannon GJ . (2000). Nature, 404, 293–296.

  8. Lee NS, Dohjima T, Bauer G, Li H, Li MJ, Ehsani A, Salvaterra P and Rossi J . (2002). Nat. Biotechnol., 20, 500–505.

  9. Manche L, Green SR, Schmedt C and Mathews MB . (1992). Mol and Cell. Biol., 12, 5238–5248.

  10. Miyagishi M and Taira K . (2002). Nat. Biotechnol., 20, 497–500.

  11. Paddison PJ, Caudy AA and Hannon GJ . (2002). Proc. Natl. Acad. Sci. USA, 99, 1443–1448.

  12. Paul CP, Good PO, Winer I and Engelke DR . (2002). Nat. Biotechnol., 20, 505–508.

  13. Sharp PA . (1999). Genes and Dev., 13, 139–141.

  14. Sharp PA . (2001). Genes and Dev., 15, 485–490.

  15. Stark GR, Kerr IM, Williams BR, Silverman RH and Schreiber RD . (1998). Ann. Rev. Biochem., 67, 227–264.

  16. Stilwell JL and Samulski RJ . (2003). Biotechniques, 34, 148–150, 152, 154 passim.

  17. Sui G, Soohoo C, Affar el B, Gay F, Shi Y and Forrester WC . (2002). Proc. Natl. Acad. Sci. USA, 99, 5515–5520.

Download references


We thank Carolyn Pierce and Alejandra Herrera for preparation of figures. This work was supported in parts by grants from the Nearburg Family Foundation, Children Cancer Fund and a Specialized Program of Research Excellence (SPORE) grant, P50-CA70907 from the NIH.

Author information



Corresponding author

Correspondence to Preet M Chaudhary.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Tomar, R., Matta, H. & Chaudhary, P. Use of adeno-associated viral vector for delivery of small interfering RNA. Oncogene 22, 5712–5715 (2003).

Download citation


  • AAV
  • siRNA
  • RNAi
  • RNA interference

Further reading


Quick links