Introduction

The ability to knock down expression levels of an endogenous gene product and to monitor effects on individual cells or whole animals has emerged as a powerful strategy for rapid analysis of specific gene function^{1, 2}. RNAi is widely being adopted as a method to achieve gene silencing of an endogenous protein by means of sequence-specific mRNA degradation, triggered by small interfering RNAs (siRNAs)^{3, 4}. The antisense strand of the siRNA guides its cognate mRNA into an RNA-induced silencing complex, which rapidly cleaves the target mRNA, leading to directed disruption of gene function^{1, 2, 4, 5}.

Chemically synthesized siRNAs⁵, or endogenous expression of shRNA with a fold-back stem-loop structure^{6, 7, 8}, are most commonly used to induce RNAi in mammalian cells. Vector-based constitutive expression of shRNAs by RNA polymerase III promoters can result in stable and efficient knockdown of target genes^{6, 7, 9}. However, constitutive expression of shRNAs imposes major limitations when analyzing the function of genes whose expression is vital for survival of an organism¹⁰.

Inducible RNAi systems circumvent this limitation by enabling the inhibition of expression of an essential gene only when the inducing agent is present. A second potential advantage of such a system is that the level of knockdown of the essential gene can be controlled by the concentration of inducing agent.

The most commonly employed inducible RNAi system uses tetracycline-based regulatory elements, and it has been successfully used to induce gene knockdown both in vitro^{11, 12} and in transgenic mice^{13, 14}. The availability of multiple expression cassettes containing Pol III (U6 or H1)¹² or Pol II (Ubc9)¹⁴ promoters offers this system more flexibility than other systems. However, the chief drawbacks of the tetracycline-inducible system are as follows: relatively high levels of 'leaky' expression in the uninduced state^{15, 16, 17, 18}; and the slow rate of induction of shRNA expression from the induced state^{18, 19}. These limitations often do not permit accurate and quick analysis of loss-of-function phenotypes.

To overcome these limitations, we utilized and modified the inducible vector of a commercially available ecdysone system to enable synthesis of shRNA in mammalian cells^{20, 21}. The efficacy of this system has been previously evaluated against the most abundant essential histone variant, H2A.Z, in both mouse and human cell lines^{20, 22}; the procedures described in this protocol are based on the methods used in these previously published papers^{20, 22}.

An overview of the ecdysone-inducible RNAi system

The ecdysone-inducible shRNA expression system comprises the following: the inducer, ponasterone-A (PonA), which is an ecdysone analog; the ecdysone receptor (EcR) expression vector, pERV3 or pVgRXR; and the shRNA expression vector, pIND-miR30.

The inducer, PonA. PonA is an analog of ecdysone and exhibits a potency similar to that of the insect hormone ecdysone; it is an economical substitute for ecdysone as an inducer.

The EcR expression vector (pERV3 or pVgRXR). The EcR is a member of the Drosophila retinoid-X-receptor (RXR) family of nuclear receptors and is responsible for insect metamorphosis. It does not have a role in mammalian development²³. The functional EcR is composed of two interacting subunits, EcR and RXR, that specifically bind to a promoter and tightly repress transcription^{24, 25}. When the ecdysone inducer binds to the EcR/RXR receptor, the corepressor complexes are released from the promoter and transcription is rapidly activated.

To increase the inducibility and specificity of this system in mammalian cells, a truncated EcR was fused to the transcriptional activation domain of VP16 and a mutated DNA binding domain of the glucocorticoid receptor^{23, 25}, resulting in a modified form of the ecdysone subunit known as VgEcR. The modified EcR, VgEcR/RXR, therefore specifically binds to unique synthetic response elements E/GRE (a hybrid DNA sequence that contains half of the ecdysone response element and half of the glucocorticoid response element²³).

The commercially available receptor vector pERV3 (pVgRXR can be alternatively substituted) contains an expression cassette from which a dicistronic message encoding ecdysone subunits, VgEcR and RXR, is ubiquitously expressed from the cytomegalovirus (CMV) promoter. It is also possible to replace the CMV promoter with a tissue-specific (or other) promoter of interest using the PstI and FseI restriction sites. This would be useful for in vivo studies where cell- or tissue-specific knockdown of a gene is required²⁶; only tissues that express the receptors will express the shRNA in the presence of the inducer. Individual promoters and their expression patterns can be found in the human promoter database (http://tiprod.cbi.pku.edu.cn:8080/index.html) or the mouse tissue-specific database (http://rulai.cshl.edu/cgi-bin/gbrowse?source=MMPD5).

The shRNA expression vector, pIND-miR30. The shRNA to be expressed is cloned into the modified version of the pIND-miR30 plasmid, which places its expression under the control of five tandem repeats of the synthetic response elements (E/GRE). In the absence of the ecdysone inducer, the receptors bind as a heterodimer to the E/GRE repeats (which are located upstream of the minimal heat shock pol II promoter) along with corepressors and tightly repress transcription of the shRNA (Fig. 1). Upon induction, ecdysone (or analogs of ecdysone) tightly bind and change the conformation of the receptor proteins, the corepressors are released and the transcriptional coactivators are recruited to the promoter, resulting in highly induced expression of shRNA.

Figure 1: Ecdysone-inducible RNAi system.

(a) The functional receptors, VgEcR and RXR, are expressed from pERV3 in mammalian cells. In the absence of ecdysone ligand, the receptor proteins exist in an inactive conformation and synthesis of shRNA is repressed. In the presence of ligand (PonA), the receptor forms a heterodimer, binds to the response elements, E/GRE, and synthesizes shRNAs in the form of preprocessed structures, which are subsequently cleaved by Drosha and Dicer complexes. A sample structure of the miR30-based shRNA with predicted cleavage sites of Drosha and Dicer is shown. The sense and antisense sequences of shRNA are depicted in green and red, respectively. (b) Microscopic analysis of a sample cell line tested for range of basal versus induced levels of GFP expression. ER cells were transiently transfected with reporter plasmid expressing GFP (reporter) and assayed for GFP expression in the absence of PonA (uninduced) and presence of 5 M PonA (induced) at 6 h after induction. Note the induced expression of GFP⁺ activity in cells treated only with PonA (but not in uninduced cells).

Full size image (58 KB)

Selection of shRNA target sequences and cloning

The appropriate design of shRNAs is necessary for avoiding off-target or nonspecific target effects. An important factor in designing an effective shRNA is to ensure that it does not contain any stretches of perfect dsRNA of 11 bp (ref. 42). Additionally, the thermodynamic stability of the shRNA precursor should be taken into consideration; the antisense strand with the less stable 5' end (owing to weaker base pairing) is favorably entered into RNA-induced silencing complex⁴³. A practical guide for selecting siRNAs is given in detail elsewhere in the literature^{42, 44}.

A number of web-based algorithms are available at no cost to predict effective shRNA sequences either by simply browsing the accession number of gene(s) or by blast-searching raw sequences (http://codex.cshl.edu/scripts/newmain.pl, http://katahdin.cshl.org/homepage/siRNA/RNAi.cgi?type=shRNA). The shRNAs predicted by these Web tools include miR30-flanking sequences for direct cloning³⁹. Additionally, a substantial number of premade, sequence-verified shRNA libraries covering the mouse and human genomes are publicly available (http://www.openbiosystems.com/RNAi/shRNALibraries/, http://cgap.nci.nih.gov/RNAi/) and are provided in the retroviral vector backbone^{38, 40}; single-step cloning is sufficient to transfer the shRNAs cassette into the pIND-miR30 vector.

Typically, the target sequence should be 21–23 nt long. In general, 3–4 shRNAs should be individually tested for efficient knockdown of a single target gene. A database search is recommended to filter out candidate targets that are present in other genes to avoid unwanted knockdown of these loci. Generally, shRNA consists of short 5' and 3' miR-30 sequences flanking a 22-nt sense and 22-nt antisense target sequence that is separated by a 15-nt miR-30 loop (Figs. 1a and 2). The total length of the shRNA insert is 114 bp. To quickly and efficiently generate shRNA constructs, we employ a PCR-based strategy to clone shRNA sequences into the pIND-miR30 vector using the oligonucleotide itself as a template (see 'Primer design' in REAGENT SETUP). The amplicon encoding each shRNA is inserted between XhoI and EcoRI restriction sites that are flanked by miR-30 sequences.

Figure 2: Design and shRNA cloning strategy.

(a) Schematic of PCR-based shRNA synthesis. The entire miR-30-based shRNA is synthesized by PCR amplification of primers that contain complementary 3' ends, shown in blue, corresponding to part of the miR-30 loop structure. The sense and antisense nucleotides are shown in green and red, respectively, and those in black represent the flanking miR-30 sequences. The XhoI and EcoRI sites are positioned such that the enzyme cuts precisely at the ends of the 5' and 3' miR-30 flanking sequence of shRNA. (b) Schematic of the pIND-miR-30 vector showing the position of 5' and 3' native miR-30 sequences and the cloning sites of XhoI and EcoRI for shRNA insertion.

Full size image (34 KB)

Materials

Reagents

• shRNA expression vector, pIND-miR30 (modified version of a formerly commercially available plasmid, pIND) (see Table 1)

  Table 1: Expression vectors.

  

  
  Full table
  
  
• Receptor expression vector, pERV3 (commercially available from Stratagene, cat. no.217460). Alternatively, pVgRXR can be used but is no longer commercially available
• Custom-designed oligonucleotides (P1 and P2, order at 0.5 M scale from, e.g., Sigma-Proligos; see Table 2 and 'Primer design' in REAGENT SETUP)

  Table 2: Oligonucleotides required.

  

  
  Full table
  
  
• Primer set for sequencing shRNAs (Sigma-Proligos; see Table 2)
• Restriction enzymes: EcoRI, XhoI and MluI
• Transfection reagent: lipofectamine-2000 (Invitrogen, cat. no. 11668-027)
• MinElute reaction cleanup kit (Qiagen, cat. no. 28204)
• Spectinomycin (Sigma, cat. no. S4014; see REAGENT SETUP)
• Zeocin, blasticidin and geneticin (Invitrogen; see REAGENT SETUP)
• Vent DNA polymerase with 10 ThermoPol buffer (New England Biolabs, cat. no. M0254)
• Rapid DNA T4 ligase system (Promega, cat. no. M8221)
• PonA (Invitrogen, cat. no. H101-01)
• QIAprep spin miniprep kit (Qiagen, cat. no. 27106)
• Cloning disc, sterile, -irradiated (Sigma-Aldrich, cat. no. Z374458-100EA)
• Bacterial strains: competent Escherichia coli STBL4 (Invitrogen, cat. no. C7373-03) or E. coli GT115 cells (InvivoGen, cat. no. Lyo-115-21)

Equipment

• Fluorescence microscope (Olympus or equivalent)
• Inverted light microscope (Olympus CK2 or equivalent)
• PCR machine (Bio-Rad or equivalent)
• Tissue culture biohazard hood
• Tissue culture incubator with 5% CO₂

Reagent setup

• 1 mM PonA solution Resuspend 250 g of PonA in 500 l of 100% ethanol. Store at - 20 °C in the dark.
• Low-salt LB medium 10 g of tryptone, 5 g of NaCl and 5 g of yeast extract in 1 liter of H₂O. Adjust pH to 7.5 with 1 N NaOH. Autoclave and allow to cool to at least 55 °C before adding the appropriate antibiotics. Store the plates at +4 °C in the dark.
• 100 g l^{- 1} of zeocin solution Resuspend 100 mg of zeocin (Invitrogen, cat. no. R250-05) in 1 ml of autoclaved deionized water. Zeocin is light sensitive; store at - 20 °C in the dark. It is stable for up to 6 months. Alternatively, premade zeocin is available from Invitrogen (cat. no. R250-01).
  Caution Zeocin is toxic. Do not inhale or ingest solutions containing zeocin. When handling, wear gloves and safety glasses and use a safety biohazard hood.
• 25 g l^{- 1} of spectinomycin solution Dissolve 25 mg of spectinomycin per 1 ml of water.
• 50 g l^{- 1} of blasticidin solution Resuspend 50 mg of blasticidin (Invitrogen, cat. no. R210-01) in 1 ml of autoclaved deionized water.
  Caution Blasticidin is toxic and readily absorbed through the skin. Wear appropriate gloves and safety goggles and use a safety biohazard hood.
• 100 g l^{- 1} of geneticin solution Dissolve 100 mg of geneticin (Invitrogen, cat. no. 11811-031) in 1 ml of 100 mM HEPES buffer, pH 7.2.
• Primer design For knockdown of any target gene(s) in mammalian cells, predesigned miR-30-based shRNAs (140,000 human and mouse genes) can be purchased from Open Biosystems (http://www.openbiosystems.com/RNAi/shRNALibraries/) and cloned directly into the pIND-miR30 vector at XhoI and EcoRI sites ('cut and paste' approach). A free Web tool 'RNAi oligo retriever' is also available (at http://www.cshl.org/public/SCIENCE/hannon.html) to facilitate designing a new miR-30-based-shRNA. Alternatively, select a 22-nt 'sense' sequence from the gene of interest and design primers (P1 and P2; see Table 2) to amplify the shRNA insert. The forward primer (P1) must contain a common 5' flanking sequence of miR-30 with an XhoI site, 22-nt of sense sequence and part of the miR-30 hairpin loop sequences. The reverse primer (P2) includes a 3' flanking sequence of miR-30 with an EcoRI site, 22-nt of antisense sequence and part of the miR-30 loop sequences. The 3' portion of P1 and P2 primer sequences (ideally, 12 bp) must overlap and anneal to each other in a PCR (see Fig. 2), whereby the entire shRNA is amplified to produce a PCR product that can be cloned into pIND-miR30.

Procedure

1. Points from here (point 1) up to and including point 7 are related to Preparation of shRNA for ligationDesign and synthesize two oligonucleotides (P1 and P2) as described in REAGENT SETUP (see Table 2). Resuspend the oligonucleotides in sterile nuclease-free water to stock concentrations of 10 M. We find it is not necessary to phosphorylate the ends of oligonucleotides or to purify them by polyacrylamide gel electrophoresis.
2. Set up a 100-l PCR amplification reaction using P1 and P2 primers as described below. Amplification of PCR products is based on the use of primers that contain complementary 3' ends; the primers use each other as a PCR template (see Fig. 2a).

   Component	Amount	Final concentration
   10 ThermoPol buffer	10 l	1
   DMSO	5 l	5% (vol/vol)
   dNTPs (2 mM stock)	10 l	200 nM
   P1 primer (10 M stock)	5 l	0.5 nM
   P2 primer (10 M stock)	5 l	0.5 nM
   Vent DNA polymerase (2Ul- 1)	1 l	2 U
   Water	64 l

   
   Critical step It is essential to add 5% (vol/vol) DMSO or other similar agents (such as GC-melt PCR reagent from Clontech) to weaken hydrogen bonding and prevent formation of hairpin structures; undesirable hairpins can often interfere with PCR amplification. Note that the Vent Polymerase and 10 ThermoPol buffer (with added 2 mM MgCl₂) are available from NEB. To maintain the thermostability and proofreading activity (i.e., minimizing the error rate) of the enzyme, it is not advisable to add extra MgCl₂ to the supplied reaction buffer.Troubleshooting
3. Amplify the PCR according to the following setup:

   Cycle number	Denaturation	Annealing	Polymerization	Hold
   1	2 min at 94 °C	–	–
   2–29	30 s at 94 °C	30 s at 50 °C	1 min at 72 °C
   30	30 s at 94 °C	30 s at 50 °C	5 min at 72 °C	4°C

4. Analyze an aliquot of the PCR products by 4% (wt/vol) agarose gel electrophoresis, and estimate the concentration and yield of the amplified target. The reaction should produce a single band of 114 bp in size.
5. Purify and elute the double-stranded PCR product using a MinElute PCR purification kit, according to the manufacturer's instructions. We find that this column-based PCR cleanup kit is sufficient to remove the primers and salts with minimal loss of double-stranded PCR products.
6. Double-digest the amplification product with XhoI and EcoRI for 1–2 h at 37 °C.
7. To remove impurities that might interfere with ligation, purify the digested fragments by extraction with phenol/chloroform or using a column-based reaction cleanup kit; make sure that the column is designed to capture products as small as 100 bp.Pause Point Digested DNA can be stored at 4 °C for a few days or at - 20 °C for several months.
8. Points from here (point 8) up to and including point 9 are related to Preparation of pIND-miR30 vector for ligationPrepare pIND-miR30 vector DNA by digesting 2 g with XhoI and EcoRI, following the supplier's instructions.
9. Purify the linearized vector using the MinElute reaction cleanup kit and resuspend the DNA in 20 l of TE buffer.Pause Point Digested DNA can be stored at 4 °C for a few days or at - 20 °C for several months.
10. Points from here (point 10) up to and including point 16 are related to Cloning shRNA into pIND-miR30 vectorSet up a 10-l ligation reaction as indicated in the table below. Incubate the reaction for 15–20 min at 20–25 °C.

    Component	Amount	Concentration
    Vector, pIND-miR30	1 l	100 ng
    PCR insert	1 l	8 ng
    2 ligation buffer	5 l	1
    Rapid DNA ligase (1Ul- 1)	1 l	1 unit
    Water	2 l

    
    Critical step Also set up a negative control reaction with the digested pIND-miR30 alone (i.e., no shRNA insert) to estimate the number of background transformants caused by the presence of undigested vector.Pause Point Ligation reactions can be stored at 4 °C overnight or at - 20 °C for a few days.
11. Use 2 l of each ligation product (including control reactions) to transform individual aliquots of E. coli GT115 or STBL4-competent bacteria and select for growth in the presence of either 25 g ml^{- 1} of spectinomycin or 100 g ml^{- 1} of blasticidin–LB plates; antibiotics are interchangeable for bacterial selection.
    Caution Blasticidin is toxic; avoid inhalation and skin contact.
    Critical step It is essential to add blasticidin to the low-salt LB media and store plates at 4 °C in the dark. Failure to lower the salt content in LB medium will result in the failure of selection due to inactivation of the drug. Plates containing spectinomycin and blasticidin are stable for up to 2 weeks.Troubleshooting
12. Select individual colonies from the experimental plate and grow them overnight in LB medium containing 25 g ml^{- 1} of spectinomycin.
13. Isolate plasmid DNA using the Qiagen miniprep kit, following the supplier's instructions.
14. Determine whether the recombinant clones contain the shRNA insert by digestion with EcoRI and XhoI restriction enzymes, according to the supplier's instructions.
15. Verify the correct insertion by sequencing at least 5–6 bacterial clones for each individual shRNA using appropriate sequencing primers (see Table 2).
    Critical step It is highly recommended to sequence the recombinant clones to confirm the sequence of the shRNA; we find that up to 30% of the clones may contain mutated inserts (generally 1 or 2 bp deletions within the insert). The reason for this is not known but may be due to activation of the DNA repair machinery within E. coli, as a result of the inverted repeat sequence within the shRNA insert.Troubleshooting
16. Transform the verified clones into the cell lines of interest to test their ability to knock down gene expression (as outlined in Box 1) before proceeding with stable transfection.Troubleshooting
17. Points from here (point 17) up to and including point 29 are related to Establishing stable Ecdysone-responsive (ER) cell lines (i.e., transfected with receptor plasmid pERV3 or pVgRXR)Digest 5 g of the receptor plasmid (pERV3 or pVgRXR) with MluI at 37 °C for 3–4 h and then purify the linearized plasmid by using the MinElute reaction cleanup kit. Alternative to Steps 14–26, premade ER cell lines (e.g., HEK, CHO and NIH3T3) can be purchased from Stratagene.
    Critical step It is important to purify the linearized plasmid; DNA must be free from salts and proteins, which can often interfere with transfection reagents.
18. Seed 1 10⁵ cells in a 6-well plate the day before transfection. Incubate the cells at 37 °C in a 5% CO₂ incubator. Cells should be 60–70% confluent at the time of transfection.
19. Just before transfection, remove the medium and replace with fresh serum-free DMEM. Incubate the cells at 37 °C in a 5% CO₂ incubator until the cells are ready for transfection.
20. Transfect the cells with 4 g of DNA using lipofectamine-2000 or an equivalent transfection reagent, according to the supplier's instructions.
21. After transfection, aspirate off the medium and replace with fresh DMEM supplemented with 10% FCS. Allow the cells to recover at 37 °C in a 5% CO₂ incubator for 24 h.
22. The next day, split the cells into a 10-cm plate containing fresh medium supplemented with either geneticin (to select for pERV3) or zeocin (to select for pVgRXR).
    Critical step Split the cells so that they are no more than 25% confluent. If the cells are too dense, the antibiotic will not kill the nontransformed cells. It is recommended to add a predetermined concentration of antibiotics required for the specific cell line; consult the available references or the supplier of cell line for the optimum concentration (i.e., the lowest concentration of antibiotics that kills all of the cells). Alternatively, a simple technique is to grow cells in a 24-well plate with a range of antibiotic concentrations (from 100 to 1,000 g ml^{- 1}) in the individual wells and to determine the concentration of antibiotics that kills the cells within 10 d.
23. Isolate antibiotic-resistant cells by the cloning disc method. Mark at least 20 foci (i.e., individual antibiotic-resistant colonies) to be isolated with a marker, on the bottom of the plate, and then remove the medium from the plate.
24. Dip cloning discs (size 3–5 mm) in trypsin solution and place one disc on each marked focus or colony for 5–10 min to ensure complete trypsinization.
25. Remove cloning disc from the plate (cells adhere to the disc after trypsin treatment) and transfer each into a separate well of a 24-well plate with fresh DMEM supplemented with 10% FCS and 2 mM L-glutamine. Incubate the cells at 37 °C in a CO₂ incubator overnight.
26. The next day, substitute culture medium with medium containing predetermined concentrations of antibiotic. Grow the cells at 37 °C in a CO₂ incubator. Replenish the selective media every 3–4 d until the cells approach 80% confluence.
27. Trypsinize and replate cells into three replicates of the original 24-well plate. Incubate the cells at 37 °C in a CO₂ incubator until ready for Step 28.
    Critical step Use one plate for testing inducibility of cells (Step 28); use second plate as uninduced control (Step 28); and use third plate for stable transfection with shRNA plasmids, as described in Steps 30–33.
28. Use one of the plates to test the inducibility of the cells by transiently transfecting with the reporter plasmid expressing green fluorescent protein (GFP) as described in Box 1 (Steps 3–8). Monitor the expression of GFP using an inverted fluorescence microscope; compare expression levels of the induced cells to that of the uninduced cells (see Fig. 1b).
    Critical step It is highly recommended to test the inducibility and the expression of the regulator plasmid in these stable cells (also called 'receptor-expressing' (ER) cells) using the reporter plasmid expressing GFP. The induced cells must exhibit GFP activity within 6–8 h of induction. An example of sample data for ER cells tested for high induced and low basal expression of GFP is shown in Figure 1b.
29. Pick the cells that express the highest levels of expression in the induced cells. From these highly-expressing cells, choose those that exhibit complete repression of basal transcription in uninduced cells to create a second cell line that will express shRNA from the pIND-miR30-shRNA (Steps 30–33).Pause Point It is advisable to store aliquots of ER cells in liquid nitrogen for future use.Troubleshooting
30. Points from here (point 30) up to and including point 33 are related to Stable transfection of ER cells with inducible shRNA plasmidsTransfect ER cells with pIND-miR30-shRNA as described in Steps 17–21 and select them by their ability to grow in medium-supplemented with blasticidin (to select for pIND-miR30-shRNA) and geneticin (to select for pERV3) or zeocin (to select for pVgRXR).
31. Feed the cells with fresh selective medium containing blasticidin and geneticin or zeocin every 3–4 d until antibiotic-resistant colonies can be identified (generally 8–10 d after selection).
32. When the cells approach 60–70% confluence, induce expression of the shRNA by adding 5 M PonA for 24–48 h, as described in Box 1 (Steps 6 and 7).
33. Measure the level of target gene knockdown by western or northern blotting⁴⁵ or by quantitative RT-PCR⁴⁶ and compare to levels in the uninduced cells.

Timing

Cloning shRNA into the pIND-miR30 vector requires 2 d. Time required to sequence the individual shRNA clones may vary. Validation and test induction of cloned shRNA takes about 5 d. Western blot analysis to determine which shRNA sequences produce the most knockdown of the targeted transcript requires 2 d. The generation of stable ER cell lines and testing the inducibility of the stable cell lines requires about 2 weeks. Establishing a second cell line that stably expresses shRNA requires about 2 weeks.

Troubleshooting

Troubleshooting advice can be found in Table 3.

Table 3: Troubleshooting table.

Full table

Anticipated results

The protocol presented here should result in very robust, reversible, tightly controlled expression of shRNAs with >80% of knockdown in a wide range of mammalian cells. It is, however, essential to validate the extent of the knockdown to interpret the outcome of experiments. This is usually achieved by quantitative western blot analysis^{6, 45} to compare the levels of the targeted protein in induced cells, uninduced cells or control-transfected cells (transfected with the empty vector containing no shRNA insert). Alternatively, the levels of mRNA transcript of the target gene can be measured by quantitative RT-PCR analysis⁴⁶. Stable transfection of this inducible shRNA system should confer a dose-dependent knockdown of the target mRNA ranges from 70% to 90% of the endogenous gene product.

Figure 3 illustrates an example of a western blot and RT-PCR analysis of inducible shRNA system showing the anticipated level of suppression of protein and mRNA transcript of target gene in uninduced and induced cells. Notice the decreased levels of target protein and mRNA in cells treated with PonA (an analogue of ecdysone) compared to untreated controls. Furthermore, the ability to activate shRNA specifically in the induced cells is demonstrated in northern blot analysis. Note the synthesis of 22-nt siRNAs in induced cells (but not in uninduced cells) indicating that shRNA precursor is efficiently processed in the induced cells to produce a functional siRNA. Finally, typical results of the immunofluorescence assay for detection of loss-of-function phenotypes in mammalian cells are shown in Figure 3d. Note the chromosome segregation defects in cells treated with only PonA, resulting in functional inactivation of the targeted gene.

Figure 3: An example of ecdysone-inducible synthesis of shRNA and experimental validation in mammalian cells.

Inducible stable expression of H2A.Z-specific shRNA was developed, as described in this protocol, to investigate the functional consequences of silencing an essential histone variant, H2A.Z, in mammalian cells. The cells were either untreated (control) or treated with 5 M PonA for 48 h. The expression levels of H2A.Z were analyzed by (a) western blot using histone H4 as a loading control, (b) transcript levels of H2A.Z mRNA quantified by quantitative RT-PCR and normalized against GAPDH and (c) northern blot analysis of processed siRNA in the cells grown in the absence or presence of 5-M PonA for 48 h. Blots were probed with a ³²P-labeled sense 22-nt probe corresponding to the targeting sequence. The control 5S-rRNA band was detected with ethidium bromide staining. (d) Immunofluorescence assay for detection of loss-of-function in mammalian cells. Knockdown of H2A.Z by inducible shRNA produces defects in chromosome segregation of both mouse and monkey cells (note the formation of micronuclei when H2A.Z is silenced). Cells were stained with antibodies against H2A.Z (green) and against nuclei with propidium iodide (red) (scale bar, 10 m).

Full size image (59 KB)