Large-scale production of dsRNA and siRNA pools for RNA interference utilizing bacteriophage phi6 RNA-dependent RNA polymerase

Abstract

The discovery of RNA interference (RNAi) has revolutionized biological research and has a huge potential for therapy. Since small double-stranded RNAs (dsRNAs) are required for various RNAi applications, there is a need for cost-effective methods for producing large quantities of high-quality dsRNA. We present two novel, flexible virus-based systems for the efficient production of dsRNA: (1) an in vitro system utilizing the combination of T7 RNA polymerase and RNA-dependent RNA polymerase (RdRP) of bacteriophage 6 to generate dsRNA molecules of practically unlimited length, and (2) an in vivo RNA replication system based on carrier state bacterial cells containing the 6 polymerase complex to produce virtually unlimited amounts of dsRNA of up to 4.0 kb. We show that pools of small interfering RNAs (siRNAs) derived from dsRNA produced by these systems significantly decreased the expression of a transgene (eGFP) in HeLa cells and blocked endogenous pro-apoptotic BAX expression and subsequent cell death in cultured sympathetic neurons.

FIGURE 1.

Production methods for long dsRNA. (A) In vitro dsRNA production systems. ➊ Existing method of generating dsRNA by ssRNA hybridization. cDNA of the desired gene, in this case egfp, was cloned in both orientations into a suitable vector under a T7 promoter. T7 RNA polymerase was used to synthesize both plus- and minus-stranded ssRNAs, which were annealed to yield dsRNA (dotted ends symbolize nonspecific annealing). ➋ New method of generating dsRNA by bacteriophage ϕ6 RdRP. cDNA of the desired gene was cloned into a suitable vector under a T7 promoter and ssRNA was synthesized by T7 RNA polymerase or the viral ssRNA was used directly. The RdRP of ϕ6 efficiently synthesizes an exact complementary minus strand beginning from the very 3′ end of the plus strand. (B) New in vivo dsRNA production system utilizing bacteriophage ϕ6. The diagram depicts the formation of a stable carrier state relationship between ϕ6 and P. syringae Cit7 (pLM1086) host cells. ➊ Plasmids containing cDNA of the ϕ6 L_kan segment (pLM991) and the S_eGFP segment (egfp flanked by ϕ6 s-segment 5′-packaging (*) and 3′-replication (†) signals [pPS9]), placed under a T7 promoter, were electroporated into the host cells and maintained by kanamycin selection. ➋ The cells contain a plasmid (pLM1086) that constitutively expresses T7 RNA polymerase, transiently synthesizing ssRNA from the cDNA plasmids, which are nonreplicative in P. syringae. The ϕ6 L_kan segment contains the viral RdRP and other genes necessary for the formation of empty polymerase complexes (PCs), and a kanamycin resistance (kan) gene. Packaging begins with S_eGFP ssRNA (the s-segment specific ssRNA is packaged first; Gottlieb et al. 1992), followed by l-segment-specific ssRNA. ➌ Upon packaging, an exact complementary strand is synthesized inside the PC particle by the viral RdRP. Packaged capsids contain on average three copies of the S_eGFP segment and one copy of the ϕ6 L_kan segment.

FIGURE 2.

(A, left panel) dsRNA produced by hybridization of complementary ssRNAs or by ϕ6 RdRP. (Lane 1) ssRNA from pCR3.1-eGFP, (lane 2) ssRNA from pCR3.1-eGFPminus, (lane 3) dsRNA generated by hybridization, (lane 4) dsRNA generated by hybridization after LiCl precipitation, (lane 5) dsRNA generated by ϕ6 RdRP from plus-stranded ssRNA, and (lane 6) dsRNA generated by ϕ6 RdRP from plus-stranded ssRNA after LiCl precipitation. (Right panel) Agarose gel analysis of dsRNA produced by the in vivo production system. (Lane i) LiCl precipitated dsRNA from eGFP carrier state cells, (lane ii) LiCl precipitated dsRNA from eGFP carrier state cells after HPLC, (lane iii) LiCl precipitated dsRNA from ϕ6 S-segment carrier state cells, (lane iv) LiCl precipitated dsRNA from ϕ6 S-segment carrier state cells after HPLC, and (lane v) wild-type ϕ6 dsRNA as size markers (L 6374 bp, M 4063 bp, S 2948 bp). (B) The effects of siRNAs on transgene expression. HeLa-eGFP cells were transfected with siRNA pools or siRNA oligonucleotides as indicated and fluorescence levels were analyzed by a VICTOR2 1420 multilabel counter 3 d later. The data (emission at 535 nm) are expressed as a percentage of mock-transfected (Lipofectamine) cells. Background values from medium were similar to those from parental HeLa cells and were subtracted. The averages of five independent experiments are shown ± S.E.M. *** = P < 0.001 as compared to 10 nM respective control values. The exact P values are given in Supplemental Table 2. (C) Microscopic images of typical HeLa-eGFP cultures transfected with indicated siRNAs. Scale bar represents 100 μm.

FIGURE 3.

(A) The effect of siRNAs on endogenous BAX protein (21 kDa) expression in Neuro2A-20 cells 3 d after transfection (upper panel), determined by Western blot analysis. (Lane 1) 3 nM mBax pool, (lane 2) 3 nM pLM659 control pool, (lane 3) 3 nM mBax siRNA oligo, (lane 4) 3 nM control siRNA oligo, (lane 5) Lipofectamine. The lanes contained equal amounts of total protein (β-tubulin, 50 kDa; lower panel). (B) The effect of siRNAs on apoptotic neurons. The neurons were microinjected with indicated siRNAs along with a plasmid for eGFP expression and deprived of NGF 2 d later. Fluorescent neurons living 3 d after NGF deprivation are expressed as the percentage of fluorescent neurons counted immediately after NGF deprivation. Neurons injected with eGFP plasmid alone serve as control for microinjection procedure. Shown are the averages of three independent experiments ± S.E.M. ** = P < 0.01. The exact P values are given in Supplemental Table 2.