Versatile applications of transcriptional pulsing to study mRNA turnover in mammalian cells

Simultaneous transfection with plasmid DNA and synthetic siRNA

To facilitate identification of trans-acting factors involved in mRNA decay and elucidate their role in mRNA turnover, we set out to develop protocols to integrate RNAi-mediated gene knockdown with the transcriptional pulsing system. As an effort to simplify the procedure without compromising the transfection efficiency and knockdown effect, we first tried to develop protocols that can simultaneously deliver both the plasmid DNA and the siRNA to the cells effectively. Various transfection reagents were tested because the calcium phosphate-mediated DNA transfection method used in the earlier c-fos serum-inducible promoter and Tet-off promoter protocols is finicky and does not provide sufficient transfection efficiency in many cell lines. Although recent advances in transfection reagents may circumvent this problem, many cannot be used in the presence of antibiotics, a requirement of the Tet-off system. For siRNA-mediated knockdown, we tested various transfection reagents and conditions to deliver nonspecific or UNR-specific siRNA into NIH3T3 mouse fibroblasts. UNR is an RNA-binding protein that has been shown to play a key role in mRNA degradation mediated by the c-fos major protein-coding determinant of instability (mCRD) (Triqueneaux et al. 1999; Grosset et al. 2000; Chang et al. 2004). Among the transfection reagents tested, RNAiFect from QIAGEN worked particularly well with NIH3T3 cells, consistently achieving 85%–95% knockdown efficiency (Fig. 3). We then tested the feasibility of cotransfecting the cells with both plasmid DNA and siRNA. PolyFect reagent was chosen for introducing DNA into NIH3T3 cells not only because of its high transfection efficiency but also because it tolerates antibiotics, a characteristic required for application of the Tet-off system. The plasmid pcDNA V5-PARN, coding for a mammalian nuclear poly(A) nuclease (Dehlin et al. 2000; Gao et al. 2000; Martinez et al. 2000), was used for this test because the V5-tagged PARN protein expresses well and is readily detected by antibody against the tag. The plasmid DNA was transfected either at the same time as the siRNA or 8 h later. Our results showed that siRNAs were capable of mediating efficient (70%–80%) knockdown of UNR in either case (Fig. 3A, lanes 3–5), and that expression levels of V5-PARN protein were higher if the siRNA and the plasmid DNA were transfected simultaneously than when DNA was transfected after siRNA (Fig. 3A, lanes 4,5).

Open in a separate window

FIGURE 3.

DNA and siRNA can be cotransfected effectively without compromising the knockdown efficiency. (A) Western blots of whole-cell lysates showing knockdown of UNR protein expression by the specific (UNR) siRNA and not by the nonspecific (ns) siRNA. DNA transfection was carried out at 8 h after siRNA transfection (lanes 3,4) or at the same time (lane 5) as the siRNA was transfected. (B, Left and middle panels) Northern blots showing decay and deadenylation of BBBspc+cd87 mRNA isolated from cells cotransfected with the DNA encoding pBBBspc+cd87 and either nonspecific siRNA (left) or UNR-specific siRNA (middle). The plasmid pSV-BBB, from which β-globin was expressed constitutively (SV-BBB) and which served as an internal standard, was cotransfected at the same time. (Right panel) Western blots of whole-cell lysate showing knockdown of UNR protein expression by the specific (UNR) siRNA and not by the nonspecific (NS) siRNA.

Applying simultaneous DNA-siRNA transfection to the c-fos promoter transcriptional pulsing system

Combining simultaneous transfection of siRNA and reporter DNA with transcription pulsing systems would allow examination of the effects of decreasing trans-acting factor levels on mRNA decay kinetics. This approach was first tried with the c-fos serum-inducible promoter system to verify the role of UNR in mRNA degradation mediated by the c-fos mCRD. The pBBBspc+cd87 plasmid coding for a hybrid β-globin transcript bearing the c-fos mCRD, and a control plasmid, pSV-BBB, encoding a stable β-globin message that served as an internal standard (Chen and Shyu 1994), were mixed with PolyFect transfection reagent. The siRNA against mouse UNR was prepared with RNAiFect and then added to the cells together with the DNA/PolyFect mixture. Twenty-one hours later, the cells were serum-starved in DMEM/0.5% calf serum for 25 h and then stimulated with DMEM/20% calf serum. Cell extracts and RNA samples were isolated at the indicated times for Western blot and Northern blot analyses. As shown in Figure 3B, UNR protein expression was significantly decreased by UNR-specific siRNA (∼70% knockdown efficiency), accompanied by a decreased rate of BBBspc+cd87 mRNA deadenylation and decay, consistent with previous findings (Chang et al. 2004). These results show that investigation of the mRNA decay mechanism and identification of participating factors can be facilitated by combining the simultaneous transfection protocol with the transcriptional pulsing approach.

Optimizing induction time and reporter levels for the Tet-off promoter system using PolyFect reagent

Applying the approach described above to time-course experiments using the Tet-off promoter driven transcription pulsing system would further facilitate investigation of mRNA decay under a wide variety of physiological conditions in mammalian cells. To optimize conditions for the Tet-off system using the PolyFect reagent, time-course experiments were performed to determine how long a transcriptional pulse is needed to produce a uniformly sized pool of reporter mRNA for study of deadenylation and decay kinetics. NIH3T3 B2A2 cells, which express the tTA transcription factor to support transcription from the Tet-off promoter system in the absence of tetracycline (Xu et al. 1998), were cultured in medium containing tetracycline (25–30 ng/mL) for 24 h before transfection with a plasmid encoding an unstable mRNA, the β-globin mRNA carrying a nonsense codon (BBB-PTC) that targets the transcript for rapid decay via the nonsense-mediated decay pathway (Lim et al. 1992; Thermann et al. 1998; Chen and Shyu 2003). A DNA/PolyFect mixture was prepared and added to the culture dish as described in Materials and Methods. After 48 h, transfected cells were cultured in fresh medium without tetracycline for various time periods to produce a transcription pulse before transcription was shut down by adding tetracycline (500 ng/mL) to the medium. RNA was isolated and analyzed by Northern blotting. As shown in Figure 4A, 100–110 min of induction produced a homogeneous pulse of mRNA synthesis when using PolyFect reagent, compared with the 150-min induction used with calcium phosphate transfection (Xu et al. 1998). By 120 min, some mRNA had already undergone deadenylation, giving a somewhat smeared band.

Open in a separate window

FIGURE 4.

Test for the optimal induction time and reporter plasmid levels for the Tet-off promoter system using PolyFect reagent. (A) NIH3T3 B2A2 cells were transfected with 1 μg of pTetBBB-PTC and 1 μg of pSV α-globin/GAPDH in the presence of Tet (25 ng/mL). Cytoplasmic RNA was isolated at various times after removal of tetracycline as indicated. (B) NIH3T3 B2A2 cells transfected with different amounts of pTetBBB-PTC were transcriptionally pulsed as described in Materials and Methods, and transcription was shut down after 110 min by adding 500 ng/mL tetracycline to culture medium. The number above each lane corresponds to hours (hr) after tetracycline addition. α-Globin/GAPDH mRNA was expressed constitutively and served as an internal standard for transfection efficiency and sample handling. (C) Semi-log plot showing the decay kinetics for BBB-PTC mRNA. Quantification of RNA was obtained by scanning radioactive blots with an imager (Bio-Rad).

Another critical factor to be considered is the optimal amount of the DNA required for transfection, as the transfection efficiency varies when different reagents are used. While using too little DNA will not produce enough signal for accurate data analysis, introducing too much DNA into a cell can result in excess mRNA synthesis, which may saturate the decay machinery and thus interfere with rapid mRNA decay. Therefore, even though the transfection efficiency with the Polyfect reagent is much higher than that with calcium phosphate, it is crucial to empirically identify the optimal amount of plasmid DNA. We performed time course experiments using different amounts of pTet-BBB-PTC DNA with the Tet-off system and the PolyFect reagent in NIH3T3 B2A2 cells. Using an unstable mRNA as a reporter here allowed us to calibrate the system for rapid decay kinetics. Transcription was pulsed for 110 min as described above, and the mRNA was analyzed by Northern blot. The results showed that the reporter mRNA was readily detected by Northern blot analysis when a 10-cm dish of NIH3T3 B2A2 cells was transfected with 0.1–0.5 μg but not with 0.05 μg of reporter plasmid DNA (Fig. 4B). However, rapid decay of BBB-PTC mRNA was retarded when the cells were transfected with 0.2 or 0.5 μg of plasmid DNA (Fig. 4B,C).

Analyzing mRNA deadenylation and decay kinetics using the Tet-off system combined with consecutive siRNA knockdown and simultaneous transfection

The knockdown efficiency of different siRNAs exhibits significant variability. Insufficient knockdown of a target gene may complicate interpretation of experimental results, especially when partial knockdown of the target does not give rise to a phenotype. We therefore developed a “consecutive-knockdown” protocol involving repeated siRNA transfections that significantly increased knockdown efficiency (Fig. 5A). Exploiting the ability to cotransfect plasmid DNA and siRNA, the plasmid DNA was simply included in the second knockdown step. Combining consecutive knockdown with simultaneous transfection of DNA and siRNA provides a new protocol to analyze the impact on mRNA decay of the knockdown of a decay factor. Time-course experiments were performed using the Tet-off transcription pulsing system combined with repeated siRNA knockdown and simultaneous plasmid DNA/siRNA cotransfection (Fig. 5). In this example, we examined the effect of Dcp2 knockdown in the presence of ectopically expressed mutant Pan2 poly(A) nuclease on decay of BBB-PTC mRNA. Mammalian Dcp2 is the catalytic subunit of the Dcp1a-Dcp2 decapping complex that removes the 5′ 7-methyl guanosine cap from mRNA (Lykke-Andersen 2002; Wang et al. 2002; Piccirillo et al. 2003). Coexpression of a dominant-negative mutant of PAN2 to further block the first phase of deadenylation was necessary to observe an intermediate with a full-length poly(A) tail when Dcp2 expression was knocked down (Chang et al. 2004). NIH3T3 B2A2 cells were split into a 10 cm dish and transfected with either nonspecific or Dcp2 siRNA. Eight hours later, the cells were trypsinized, recultured to 30% confluence, and incubated for another 22 h before simultaneous addition of the siRNA/RNAiFect and DNA/PolyFect mixtures. The DNA/PolyFect mixture combined PolyFect reagent with two plasmid DNAs, one encoding BBB-PTC mRNA and the other coding for a dominant-negative mutant of PAN2 poly(A) nuclease (Pan2mt). Following further incubation for 41 h, a 110-min transcription pulse was induced from the pTet BBB-PTC construct as described above. At the indicated times, cytoplasmic RNA was extracted for Northern blot analysis, and cytoplasmic lysates were prepared for Western blot analysis of knockdown efficiency. Conventional siRNA/DNA simultaneous transfection, representing “single-knockdown,” was performed in parallel as a control; these cells were harvested 43–48 h after siRNA-DNA transfection. As shown in Figure 5A, the levels of Dcp2 protein were reduced more by the consecutive siRNA transfection (Fig. 5A, lane 5, ++, ∼85% knockdown efficiency) than the control single siRNA transfection (Fig. 5A, lane 4, ∼50% knockdown efficiency). The results from Northern blot analysis showed that a decay intermediate with full-length poly(A) tail was observed only with the consecutive knockdown of Dcp2, not with the single knockdown (Fig. 5E,F).

Open in a separate window

FIGURE 5.

A decay intermediate with full-length poly(A) tail can be detected using the “consecutive-knockdown” approach. (A) Western blot analysis showing the expression of HA-tagged mutant PAN2 and knockdown of DCP2 protein expression in NIH3T3 B2A2 cells (“+” indicates the DNA or siRNA was transfected once, and “++” indicates the siRNA was transfected twice). The blot was probed with antibodies against HA-tag, DCP2, and GAPDH. (B–F) Northern blots showing deadenylation and decay of nonsense codon containing β-globin (BBB-PTC) mRNA in the (B,C) absence or (D,F) presence of ectopically expressed mutant PAN2, or combined with siRNA-mediated (E) single or (F) consecutive knockdown of DCP2. α-Globin/GAPDH mRNA was expressed constitutively and served as an internal standard for transfection efficiency and sample handling.

Cell synchronization and flow cytometry

K562 III-2 cells were maintained as a suspension in RPMI 1640 with 10% fetal bovine serum and 500 ng/mL tetracycline at 37°C in 8% CO₂. Cells enriched in the desired phases were obtained by culturing them for 24 h in the presence of 2.5 μg/mL aphidicolin for G₁/S phases, 40 ng/mL nocodazole for G₂/M phases, or 1 mM colchicines for M phases. The cells were fixed in methanol followed by RNase treatment and 10 μg/mL propidium iodide staining of DNA. FACS analysis for DNA content was accomplished using a flow cytometer with an excitation wavelength of 420 nm.

Analysis of mRNA deadenylation and decay in K562 cells

Cell culture, DNA transfection, isolation of total cytoplasmic RNA, and Northern blot analysis were conducted as described previously (Shyu et al. 1996; Loflin et al. 1999b). Briefly, K562 III-2 cells were cultured to a density of 10⁶ cells/mL, harvested, washed with PBS, and resuspended in PRMI 1640 at a density of 7.5 × 10⁷ cells/mL. Transfection of cells with 50 μg of plasmid DNA by electroporation was accomplished in a 2-mm gap cuvette (3–5 × 10⁷ cells/0.4 mL). After electroporation, cells were resuspended in medium containing 40 ng/mL tetracycline with or without the appropriate cell cycle inhibitor for 24 h. Aliquots (3 × 10⁶ cells) were washed with PBS and re-suspended in medium without tetracycline but with cell cycle inhibitor for 3 h to induce a transcriptional pulse from the Tet-off promoter-driven construct, pTet-BBB+ARE. Cytoplasmic RNA was extracted for the zero time point. Alternatively, 500 ng/mL tetracycline was immediately added back to the cell suspension for the indicated times, and then cytoplasmic RNA was extracted. Northern blot analysis was performed as described previously (Shyu et al. 1996) using a DNA probe spanning 80 bp of the rabbit β-globin 3′-UTR for detection of the transfected β-globin message containing an ARE (BBB+ARE) without cross-detection of endogenous human α-globin mRNA. Note that human K562 erythroleukemic cells do not express β-globin mRNA. A 300-bp DNA fragment spanning the second exon of GAPDH was labeled for detection of endogenous GAPDH mRNA, as an internal control for variation due to sample handling and transfection efficiency.

Culturing and transfecting NIH3T3 cells

NIH3T3 cells were cultured in DMEM medium containing 10% calf serum, 2 mM glutamine, 100 U/mL penicillin, and 100 mg/mL streptomycin at 37°C and 8% CO₂. For transfection of DNA driven by the c-fos serum-inducible promoter, NIH3T3 cells were split to a density of 1.8 × 10⁶/10-cm dish 24 h before transfection. For siRNA transfection, 30 μg of siRNAs was diluted into 1 mL of serum-free medium, and 90 μL of RNAiFect (QIAGEN) was added and mixed well. For DNA transfection, 7 μg of DNA (1.5 μg of test plasmid, 3.0–4.5 μg of internal control plasmid pSVBBB, and 1.0–2.5 μg of carrier plasmid pT18) was diluted into 1 mL of DMEM medium containing 10% calf serum, and 40 μL of PolyFect (QIAGEN) was added and mixed well. Both mixtures were then added simultaneously to the culture dish. For serum induction, after 21 h, cells were serum-starved in DMEM/0.5% calf serum for 25 h and then stimulated with DMEM/20% calf serum.

For the time-course experiments using the Tet-off promoter-driven transcription pulsing system, 1.8 × 10⁶ NIH3T3 B2A2 cells (30% confluence) were plated per 10-cm plate the day before transfection. Seven micrograms of plasmid DNAs (0.05–2 μg of reporter plasmid pTet-BBB-PTC, 1 μg of internal control plasmid pSVα-globin/GAPDH, and 4–5.95 μg of carrier plasmid pT18) were transfected using 40 μL of PolyFect (QIAGEN). Forty-one hours after transfection, cells were cultured in fresh medium without tetracycline to get a transcription pulse of either 150 min or 110 min. Transcription was terminated by adding tetracycline to a final concentration of 500 ng/mL of culture medium. For transfection involving both siRNA and DNA, 90 μL of RNAiFect (QIAGEN) mixed with 30 μg of siRNA (Dharmacon) and 40 μL of PolyFect (QIAGEN) mixed with 7 μg of DNA (0.11 μg of reporter plasmid pTet-BBB-PTC, 0.22 μg of internal control plasmid pSVα-globin-GAPDH, and 6.67 μg of either carrier plasmid pT18 or the mutant PAN2 expression vector pSR-HA-PAN2-mt) were prepared separately and then added simultaneously to the 10-cm culture dish. For consecutive knockdown, cells were first split to 10-cm dishes and then transfected with 30 μg of siRNA using 90 μL of RNAiFect. Eight hours later, cells were trypsinized, re-cultured in 10-cm dishes to 30% confluence, and incubated for another 22 h before the siRNA/RNAiFect mixture and the DNA/PolyFect mixture (prepared and transfected as described above) were added simultaneously to the culture dish.

Establishment of the BEAS-2B stable cell line expressing the tTA gene

The human bronchial epithelial cell line, BEAS-2B, was purchased from ATCC and was maintained in F12/DMEM (Invitrogen) containing 5% heat-inactivated FCS, 2 mM L-glutamine, penicillin (100 U/mL), and streptomycin (100 mg/mL). Individual, and not pooled, stable BEAS-2B transfectants of pUHD15-1 encoding tTA, Tet-regulated transactivator, were obtained as described previously except that the cells were transfected using FuGENE 6 reagent (Roche) (Xu et al. 1998; Loflin et al. 1999b). The individual transfectants harboring the tTA plasmid, pUHD15-1, were first screened for tTA activity by transiently transfecting them with pUHC13-1 encoding the firefly luciferase gene driven by the Tet-off promoter. Luciferase activity was analyzed according to the manufacturer's instructions (Promega). The clone showing the highest luciferase activity in the absence of tetracycline and also the highest repression of luciferase activity by tetracycline was chosen for further characterization as described in the Results section.

Preparation of RNA samples and Northern blot analysis

Cytoplasmic RNA samples were isolated at various time points as described in each case using RNeasy RNA preparation kit (QIAGEN) and analyzed by Northern blotting. Preparation of probes and Northern blot analyses were conducted as described previously (Shyu et al. 1996). RNase H treatment of cytoplasmic mRNA to generate poly(A)⁻ RNA and analysis of deadenylation and decay curves were carried out as described previously (Shyu et al. 1991).

We thank R. Kulmacz and J. Lever for critical reading of the manuscript, M. Kiledjian for anti-hDCP2 antiserum, and P. Loflin and Y. Zhai for technical assistance. This work was supported by the National Institutes of Health (GM 46454) and in part by the Houston Endowment, Inc. and by the Sandler Program for Asthma Research to A.-B.S.