Human ribosomal protein L13a is dispensable for canonical ribosome function but indispensable for efficient rRNA methylation

Generation of L13a knocked down monocytic cells

Nucleotide positions 460–478 and 147–169 of the L13a ORF were selected as 19-nt target sequences following published guidelines (Elbashir et al. 2001; Musiyenko et al. 2007). Synthetic sense and antisense oligonucleotides encoding the shRNAs (Fig. 1A) were annealed and cloned into the lentiviral vector pLU6EGFP (Fig. 1B). Recombinant lentivirus harboring the target shRNA sequences (shRNA1 and shRNA2) and control shRNA were made from this vector and used to transduce U937 cells (see Materials and Methods for details). Cells were stably transduced with the recombinant lentivirus and clonally selected by serial dilution and selection in blasticidin. Individual clones were then tested for expression of EGFP and functional L13a shRNAs by immunoblot analysis with anti-L13a antibody. Two clones each (Clones #1 and #2) were isolated for shRNA1 and shRNA2. Immunoblot analysis (Fig. 1C, upper panel) shows that Clone #1 for both shRNA1 and shRNA2 exhibits 50% reduction of L13a compared to control shRNA and that Clone #2 exhibits ∼98% reduction. Reprobing with anti-actin antibody showed no change in actin expression, demonstrating the target specificity of the shRNAs (Fig. 1C, lower panel). Subsequent experiments used the shRNA1 and shRNA2 Clone #2 cell lines, denoted as shRNA1 and shRNA2. The same lentiviral constructs were also transduced into 293T cells, and clonally selected cells that stably express shRNA1 and shRNA2 showing ∼95% reduction of L13a were also generated (data not shown). The effective depletion of this protein from the polyribosomal fraction has also been confirmed and shows immunologically undetected level of L13a (Supplement Fig. 1, available upon sending an e-mail message containing the keyword “supple_Fig. 1_L13a siRNA” to b.mazumder@csuohio.edu).

Abrogation of L13a overcomes IFN-γ mediated silencing of Cp

L13a was previously identified as an essential component of the GAIT complex, capable of silencing Cp translation (Mazumder et al. 2003; Sampath et al. 2004). U937 cells stably expressing the shRNA of L13a and showing abrogation of L13a expression were employed in the current study to test the in vivo role of L13a in translational silencing. Wild-type U937 cells were induced by IFN-γ and de novo synthesis of Cp was measured by pulse labeling with [³⁵S]methionine followed by immunoprecipitation with anti-Cp antibody. Consistent with previous observations (Mazumder and Fox 1999), Cp protein synthesis in U937 cells ceased after ∼16 h of IFN-γ treatment. However, complete restoration of Cp protein synthesis from the IFN-γ-mediated silencing activity was observed when cells expressing shRNA of L13a were used (Fig. 1D). These results demonstrate the in vivo requirement of L13a in the IFN-γ-mediated silencing of Cp translation.

Ribosome biogenesis is unaffected in L13a deficient cells

Studies in yeast have shown that one or more steps of pre-rRNA processing are blocked and cells accumulate pre-rRNPs upon depletion or inactivation of certain ribosomal proteins (e.g., S14, L11) (Moritz et al. 1990; Kressler et al. 1999; Harnpicharnchai et al. 2001; Jakovljevic et al. 2004). However, no comparably detailed study on the role of ribosomal protein(s) in ribosome biogenesis has been performed in higher eukaryotes. The human monocytic cell lines generated in this study enabled us to monitor the effects of L13a deficiency on ribosome biogenesis in metazoan cells.

Assembly and maturation of eukaryotic ribosomes takes place in the nucleolus, where the 35S–47S primary rRNA transcript associates with a number of ribosomal proteins and other trans-acting assembly factors to form the 90S pre-rRNP. This serves as the substrate for a series of endo- and exonucleolytic digestions that ultimately generate 60S and 40S ribosomal subunits (Warner 2001; Fromont-Racine et al. 2003; Tschochner and Hurt 2003). To directly assess the effect of L13a deficiency on the accumulation of 40S, 60S, and 90S pre-ribosomes, nuclear extracts made from the L13a-deficient cells (shRNA1 and shRNA2) and from controls (nontransduced and scrambled shRNA transduced) were fractionated by sucrose gradient centrifugation. No differences were found in the nuclear accumulation of the 90S pre-ribosomes (Fig. 2A). To check the authenticity of the 90S preribosome, RNAs were isolated from 90S and 80S fractions obtained from nuclear and post-nuclear lysates. RT-PCR amplification was carried out using primers corresponding to the different regions of the unprocessed 47S rRNA (Fig. 2B, right panel), the primary constituent of 90S pre-ribosome. The primer pair from the 5′ external transcribed spacer (5′ ETS) region that is only present in unprocessed rRNA was able to amplify the product of the expected size when the RNA from the 90S peak fractions was used. On the other hand, no product was obtained using the same primers and the RNA isolated from the 80S peak. Similarly, the amplification product of the expected size was obtained using the RNA isolated from the 90S peak fractions when one primer from the 5′ ETS region and the other from the 18S region were used (Fig. 2B, left panel). These results confirmed the presence of unprocessed rRNA in the 90S pre-ribosome peak.

It is possible that nuclear accumulation of pre-ribosomes remained unaffected in L13a-deficient cells, but there might be a defect in processing or turnover of 47S pre-rRNA (Sugimoto et al. 2003). It has been reported that inactivation of the S6 protein of small ribosomal subunit causes inhibition of pre-rRNA processing in T lymphocytes (Sulic et al. 2005) and liver (Volarevic et al. 2000). To address the possibility of pre-rRNA processing defects, the L13a-deficient cells were metabolically labeled with a short pulse of [³H]-uridine followed by chase without the radioactive precursor, and samples were collected at specified time points. Processing of the de novo synthesized pre-rRNA was observed following the major pre-rRNA intermediates in the processing pathway, e.g., 47S, 41S, 36S, 32S, and the mature 28S and 18S. No significant difference was observed in their rate of appearance or disappearance of the processing intermediates with different chase time between the controls and L13a-depleted cells (Fig. 2C). Maintenance of the effective depletion of L13a has been confirmed by immunoblot analysis using the anti-L13a antibody (Fig. 2D).

Reduced methylation of 47S pre-rRNA in L13a-deficient cells

Nascent pre-rRNA undergoes a series of modifications including methylation at the 2′-O ribose position (Kiss-Laszlo et al. 1996; Bachellerie and Cavaille 1997). The role of this modification in ribosome assembly and function remains speculative. To examine the requirement of L13a in pre-rRNA methylation, L13a-depleted U937 cells were metabolically labeled with a short pulse of [methyl-³H]-methionine (Sugimoto et al. 2003). Significant reduction in the level of methylation of 47S pre-rRNA was observed in L13a-deficient cells (Fig. 3, upper panel). To rule out the possibility that the methylation defect could be due to a defect in the rate of rRNA synthesis, L13a-deficient cells were metabolically labeled by [³H]-uridine. No differences were observed between the L13a-deficient and control cells in either [³H]-uridine incorporation (Fig. 3, middle panel) or in the steady-state levels of 28S and 18S rRNA (Fig. 3, lower panel). To test whether the dependence of rRNA methylation on L13a is specific for U937 monocytic cells only, we tested rRNA methylation in human embryonic kidney 293T cells similarly depleted of L13a by lentiviral RNAi and observed essentially identical loss of methylation (data not shown). Together, these results demonstrate that depletion of L13a causes significant reduction of pre-rRNA methylation without affecting pre-rRNA synthesis.

Polysome profiles, total protein synthesis, and cell proliferation are normal in L13a-deficient cells

The experiments described above show that pre-rRNA processing, which is a nucleolar event, is not affected by L13a deficiency. However, L13a could potentially have a role in the cytoplasmic compartment, where mature 40S and 60S ribosomes come together to make the elongation competent 80S complex followed by the formation of active polysomes on the translating pool of cellular mRNA. Thus, it is possible that L13a deficiency can affect translation without influencing the processing of rRNA in the nucleolus.

Sucrose gradient analysis of cytoplasmic extracts made from L13a-deficient and wild-type control cells was used to investigate whether L13a deficiency and the subsequent defect in pre-rRNA methylation (Fig. 3) might directly affect ribosomal translational competence. No significant differences in polysome formation or in the pools of free 40S, 60S, or 80S ribosomes were observed between the two sets of extracts (Fig. 4A). If L13a played an essential role in the process of protein synthesis, its depletion would have caused not only a reduction in the number of polysomes, but also a simultaneous increase in the pool of free 40S, 60S, and 80S ribosomes. In a parallel experiment, metabolic labeling of the L13a-depleted U937 cells with [³⁵S]methionine showed no significant differences between wild-type and L13a-depleted cells in the amounts of incorporated radioactivity (Fig. 4B). The maintenance of effective L13a depletion has been verified by immunoblot analysis (Fig. 4C). Together, these data show that depletion of L13a does not significantly affect the global protein synthesis activity of ribosomes.

The ability to form active polysomes and unfettered ribosome synthesis is absolutely required for cell proliferation. Reciprocally, cells that exit the cell cycle and attain a quiescent state exhibit greatly reduced ribosome production and overall protein synthesis (Pardee 1989; Lewis and Tollervey 2000; Ruggero and Pandolfi 2003). There are conflicting reports in the literature about the role of ribosomal protein L13a on cell growth and apoptosis (Chen et al. 1998; Chen and Ioannou 1999; Kobayashi et al. 2006). Rates of cell growth comparing the L13a-depleted versus wild-type controls were performed by measuring the number of viable cells at 24, 48, and 72 h after initial seeding. This assay was performed in order to directly investigate the effect of L13a depletion on cell proliferation. The results show that depletion of L13a did not affect rates of cell proliferation (Fig. 4D). Essentially identical results were obtained in L13a-depleted HEK 293T cells, indicating the universality of this finding, independent of cellular background (data not shown).

L13a depletion has no significant effect on ribosomal frameshifting activity

All of the experiments described above show that L13a is unlikely to have any significant role in the overall function of the ribosome. However, this ribosomal protein may still contribute to the fine-tuning of the ribosome function, e.g., influencing translational fidelity. For example, in bacterial systems, ribosomal proteins L7/L12 (Kirsebom and Isaksson 1985) and S5 (Kirthi et al. 2006) have been found to play important roles in maintaining translation fidelity. In yeast ribosomal proteins L3, L5, and P0 (Peltz et al. 1999; Meskauskas and Dinman 2001; Kinzy et al. 2002) and specific 2′OH methylation defects in 25S rRNA (Baxter-Roshek et al. 2007) have been shown to alter rates of programmed ribosomal frameshifting (PRF). To address this issue, a bicistronic dual luciferase system was used in which the downstream firefly luciferase reporter can only be translated consequent to a programmed −1 ribosomal frameshift (−1 PRF) directed by sequences derived from either the HIV-1 or SARS-coronaviruses (SARS-CoV) (Dinman et al. 2002; Harger and Dinman 2003; Plant et al. 2005). Depletion of L13a did not significantly affect either HIV-1 or SARS-CoV mediated −1 PRF in U937 cells (Fig. 5).

L13a depletion has a significant effect on the cap-independent mechanism of translation initiation

Our data have shown that general protein synthesis (cap dependent) is unaffected in L13a-depleted cells. We next hypothesized that L13a depletion and the subsequent defect in rRNA modification might affect the translation of certain mRNAs competent for cap-independent translation. Global defects in another rRNA modification process, pseudouridylation, have been implicated in X-linked dyskeratosis congenita and in defects of the translation initiation promoted by internal ribosome entry sites (IRES) (Yoon et al. 2006). To test the hypothesis that the rRNA modification defect engendered by L13a depletion might affect translation of mRNAs competent for cap-independent translation, the activities of several previously characterized cellular and viral IRESes were determined using a well established bicistronic expression system (Pelletier and Sonenberg 1989). In this assay the first cistron is translated by a cap-dependent initiation mechanism while the second is under the control of an IRES element (Fig. 6A). IRES-dependent translation of cellular p53 (Ray et al. 2006), p27 (Kullmann et al. 2002; Cho et al. 2005; Yoon et al. 2006), and SNAT2 (Gaccioli et al. 2006) have been documented using this bicistronic assay. The same bicistronic assay was also used to test the cap-independent translation mediated by HCV (Honda et al. 2000) and CrPV (Wilson et al. 2000; Jan and Sarnow 2002; Yoon et al. 2006) IRESes. These reports also rigorously addressed the authenticities of these cellular IRES activities by ruling out the presence of splice sites and cryptic promoters within the IRES elements. Here, the constructs pRF-p53, pRF-p27 (Yoon et al. 2006), and pRF-SNAT2 were used as bicistronic reporter mRNAs to measure the IRES activities of p53, p27, and SNAT2. The bicistronic constructs pRF-HCV (Honda et al. 2000) and pRF-CrPV (Wilson et al. 2000) were used to measure the IRES activities of HCV and CrPV. Construction of pRF-p53 and pRF-SNAT2 are described in Materials and Methods, while pRF-p27 and pRF-CrPV were kindly provided by L. Hengst (Max-Planck-Institute of Biochemistry) and P. Sarnow (Stanford University). To quantitatively assess IRES-dependent translation in L13a-depleted cells, the bicistronic reporter mRNAs were transfected into wild-type U937 cells and cells stably expressing functional shRNAs of L13a (shRNA1 and shRNA2). The results obtained suggest that translation initiation from the cellular p53, p27, and SNAT2 IRES elements were significantly inhibited upon depletion of L13a (Fig. 6B–D). Interestingly, internal initiation driven by the IRES elements of CrPV (Pestova et al. 2004; Wilson et al. 2000) and HCV (Pestova et al. 1998; Honda et al. 2000) remained unaffected by the depletion of L13a (Fig. 6E,F). Similar results were obtained using L13a-depleted HEK 293T cells (data not shown). However, these experiments did not address the possibility that the inhibition of the cellular IRES activities might not result from L13a depletion per se, but can be a consequence of the decreased rRNA methylation (Fig. 3). To address this possibility we have further measured the IRES activities in normal cells in the presence of cycloleucine, a chemical inhibitor of rRNA methylation (Caboche and Bachellerie 1977; Scolnik and Eliceiri 1979; O'Connor and Yutzey 1988; Afman et al. 2005). Treatment of the cells with cycloleucine did not change the level of L13a (Fig. 7A); however, as expected, it resulted in abrogation of rRNA methylation (Fig. 7B). We further found that the activities of three cellular IRESes (p53, p27, and SNAT2), but not the two viral IRESes (HCV and CrPV) (Fig. 7C) were decreased under the conditions of cycloleucine treatment. This observation is similar to what we have observed in L13a-depleted cells (Fig. 6). Together, these results suggest that the inhibition of IRES-mediated translation seen in L13a-depleted cells is not due to the L13a depletion per se but very likely due to the inhibition of rRNA methylation.

Cells and culture conditions

Human monocytic U937 cells were maintained in RPMI1640 and HEK 293T cells were maintained in DMEM supplemented with 10% heat inactivated fetal bovine serum (FBS), 2 mM glutamine, and 100 U/mL of penicillin and streptomycin at 37°C and 5% CO₂.

Construction of lentiviral vector for transduction of shRNA sequences against L13a

Nucleotide positions 460–478 and 147–169 of the ORF of L13a were selected as a 19-nt target sequence following previously published guidelines (Elbashir et al. 2001; Musiyenko et al. 2007). The target shRNA sequences (shRNA1 and shRNA2) were included in sense and antisense oligonucleotides encoding the shRNAs that can serve as a precursor for the siRNAs following published guidelines (Brummelkamp et al. 2002). A scrambled shRNA sequence was used as a control. Synthetic sense and antisense oligonucleotides encoding the shRNAs (Fig. 1A) were annealed and ligated into the SalI and EcoRI sites of lentiviral vector pLU6EGFP (Fig. 1B) downstream of RNA polymerase III-specific U6 promoter. The vector pLU6EGFP (a gift from Dr. Ratan Maitra, Cleveland Biolabs, Inc.) was made by subcloning the U6 promoter into the ClaI site of the vector LRV-EGFP (Malur et al. 2005). A scrambled shRNA sequence was used to generate control shRNA. The resulting plasmid pLU6EGFP-shL13a was used to generate the recombinant lentivirus.

Preparation of the packaged lentivirus particle containing the shRNA sequences against L13a

A previously published method (Lee et al. 2004) was followed to make the packaged lentivirus particle with the shRNA sequences. HEK 293T cells (2×10⁶) were plated on 100 mm tissue culture plates. After 16 h, cells were transfected with 20 μg of transducing vector pLU6EGFP-shL13a and vectors encoding the packaging components HIV gagpol (10 μg), HIV rev (8 μg), and VSVG (6 μg) in the presence of LipofectAMINE 2000 (Invitrogen). The medium was changed the next day and cells were cultured for another 48 h, after which GFP expression was monitored. Conditioned medium was then collected and cleared of debris by low speed centrifugation (2500g for 5 min), passed through 0.45-μm filters, and stored at −70°C. The collection process was repeated one more time after 48 h, and conditioned medium from the two collections was pooled and ultracentrifuged at 100,000g at 4°C for 2.5 h. Pelleted viral particles were suspended in 1× sterile PBS and stored in aliquots at −70°C. The viral stocks were assayed for the p24 core antigen using HIV-1 Antigen ELISA kit according to manufacturer's instructions (BioMerieux, Inc.).

Transduction of cells with recombinant lentivirus and isolation of shRNA expressing cells

U937 and HEK 293T cells (1×10⁶) were seeded into 24 well plates in 0.5 mL of complete medium (RPMI for U937 cells and DMEM for HEK 293T cells) and transduced by recombinant lentivirus at a M.O.I of 10:1. Transduction was carried out in the presence of polybrene (10 μg/mL) and IL-2 and IL-3 (100 ng/mL). After 24 h, transduced cells were transferred to 175 mm flasks in selection medium with 10 μg/mL of blasticidin. After 48 h in selection medium, cells were counted and seeded into 96 well plates in a ratio of 1 cell per well and were monitored for GFP expression. After 15 d in selection media, GFP positive cells were transferred to 25 mm flasks and tested for silencing of L13a expression by immunoblot analysis with anti-L13a.

Metabolic labeling of cells and monitoring of specific and general protein synthesis

To monitor the de novo protein synthesis of Cp, U937 cells expressing shRNA1, shRNA2, or control shRNA (8×10⁶ cells in 4 mL of RPMI 1640 medium) were treated with IFN-γ (500 U/mL) for up to 24 h. Cells were collected by centrifugation and metabolically labeled by incubation for 2 h with [³⁵S]methionine (100 mCi/mL; Trans-Label, ICN) in methionine-free medium. Cells were collected by centrifugation, lysed in 50 mM Tris (pH 7.6), 50 mM NaCl, 1 mM PMSF, 1 mM DTT, and 0.5% NP40, and newly synthesized ³⁵S-labeled Cp was immunoprecipated using anti-Cp antibody (Accurate Chemical). Proteins were resolved by 7% SDS-PAGE, gel was fixed, treated with Amplify (Amersham), dried, and exposed to MR film (Kodak).

To measure total protein synthesis, U937 cells were incubated in methionine-cysteine-free RPMI1640 medium for 30 min in 24 well plates (10⁶ cells per well) before labeling. In vivo metabolic labeling was performed by adding Trans[³⁵S]label (MP Biomedicals) at a final concentration of 100 μCi/mL containing labeled L-methionine and L-cysteine. After 90 min, the cells were washed with ice-cold PBS and incubated with 1:1 1× PBS and 1 M NaOH for 10–15 min at 37°C to stop the labeling. Ice-cold TCA was then added to the aliquots of cell lysates to a final concentration of 10%, mixed well, placed on ice for 1 h, and filtered through GF/C filter discs (Millipore). The GF/C discs were washed three times with 5% TCA and once with 10 mL of EtOH (ice cold). Air-dried filter discs were placed in scintillation vials with scintillation fluid and counted using a liquid scintillation counter. Total protein content was measured using the Bradford assay to determine the counts per minute per microgram of protein.

Metabolic labeling of cells to examine rRNA processing and methylation

To examine the rRNA processing, 1×10⁶ U937 cells expressing either shRNA1, shRNA2, or control shRNA were incubated in 1 mL of 0.5% serum containing RPMI medium in the presence of 50 μCi of [³H]-uridine for 1 h. At the end of incubation the cells were chased with nonradioactive medium for 0, 45, or 90 min. RNAs were extracted at the end of the indicated chase periods, resolved through a 1% agarose gel, and transferred to Hybond N+ membrane (Amersham). Dried membranes were treated with Amplify (Amersham) and exposed to MS film (Kodak). Turnover rates of the 47S precursor rRNA to the processed and matured 28S and 18S rRNA were determined. To examine rRNA methylation, 1×10⁶ U937 cells expressing either shRNA1, shRNA2, or control shRNA were preincubated in 1 mL of methionine-free RPMI medium for 30 min followed by pulse labeling with 50 μCi of [methyl-³H]-methionine (Amersham) for 1 h. At the end of the pulse period, cells were washed and total RNAs were extracted and analyzed as above. To monitor total RNA synthesis in parallel, the same number of cells were pulse labeled with 50 μCi of [³H]-uridine for 1 h followed by extraction of total RNA. The RNAs were processed as above.

Cell proliferation assays

Cell proliferation was measured using the MTT cell proliferation assay kit (ATCC). Cells were seeded at 2×10⁴ cells per well in 96-well plates and incubated at 37°C for 0, 24, 48, and 72 h, respectively. Absorbance was determined at 570 nm with background subtraction at 650 nm. This assay relies on the reduction of yellow tetrazolium MTT [3-(4,5-dimethylthiazolyl-2)-2,5-diphenyltetrazolium bromide] by metabolically active cells to purple colored formazan followed by spectrophotometric quatification. The number of cells was calculated from a standard curve of cell number versus absorbance.

Analysis of polysomes and 90S pre-ribosomes

Polysomes were analyzed from cytoplasmic extracts and 90S pre-ribosomes were analyzed from nuclear extracts. For preparation of cytoplasmic extracts, cells from three 175 cm² tissue culture plates (60% confluence) were treated with 100 μg/mL cycloheximide (MP Biomedicals) for 15 min at 37°C, harvested, and washed with ice-cold phosphate-buffered saline containing cycloheximide (100 μg/mL). Cells were pelleted by centrifugation, lysed in 375 μL of lysis buffer containing Polysome Buffer (PB; 10 mM HEPES at pH 7.5, 100 mM KCl, 2.5 mM MgCl₂, 1 mM dithiothreitol [DTT], 50 U recombinant RNasin [Promega], and 0.1% Igepal-CA630 [Sigma]) followed by two cycles of rapid freeze–thawing. Nuclei were pelleted by centrifugation in a microcentrifuge at 13,000 rpm for 5 min at 40°C. Supernatants (cytoplasmic extracts) were transferred to new 1.5 mL tubes on ice. Five O.D. units of cytoplasmic extracts were carefully layered over 10%–50% linear sucrose gradients (in PB) and centrifuged at 17,000 rpm in a Beckman SW32.1 Ti rotor for 18 h at 4°C. For preparation of nuclear extracts, cells from five 175 cm² tissue culture plates (60% confluency; ∼5×10⁷ cells) were harvested and washed with ice-cold phosphate-buffered saline twice, lysed in 1 mL hypotonic buffer (10 mM HEPES at pH 7.5, 10 mM KCl, 10 mM NaF, 1 mM PMSF, 0.1% Igepal-CA 630) by vortexing at half maximal speed for 15 sec. Nuclei were collected by centrifugation at 1400g for 5 min, lysed in nuclei lysis buffer (50 mM Tris-HCl at pH 8, 150 mM NaCl, 0.5% Igepal-CA 630, 1 mM PMSF), and kept on ice for 30 min with occasional vortexing. Nuclear extracts were collected by centrifugation at 15,000 rpm for 15 min at 4°C, loaded on top of 10%–40% sucrose gradients (in PB), and centrifuged at 20,000 rpm in a Beckman SW32.1 Ti rotor for 18 h at 4°C. Gradients were fractionated using an ISCO Gradient Fractionation System equipped with UA-6 detector. The presence of unprocessed pre-rRNA intermidiate was confirmed by RT-PCR analysis of the total RNA isolated from the 90S using primers specific for the 5′ETS and 18S regions. The sequences of the primers are 5′ETS (forward): 5′-GCGCACGTCCCGTGCTC-3′, 5′ETS (reverse): 5′-GAGGGGGAAGCGGAGGAGG-3′, 18S (forward): 5′-GCGCTGACCCCCTTCGC-3′, and 18S (reverse): 5′-CTCCCCGGGTCGGGAGTG-3′.

Construction of bicistronic IRES reporter plasmids

The IRES sequence of p53 [p53(+39) 5′UTR] was amplified by RT-PCR amplification from U937 cell total RNA using the primers based on published sequence (Ray et al. 2006). The sequence was verified and cloned into pRF vector (Yoon et al. 2006) to generate pRF-p53. To make pRF-SNAT2, the SNAT2 IRES sequence was excised from the construct m/SNAT2-1 (Gaccioli et al. 2006) and subcloned into pRF vector. Plasmids to monitor HIV-1 and SARS-Corona virus directed programmed −1 ribosomal frameshifting were previously described (Dinman et al. 2002; Harger and Dinman 2003; Plant et al. 2005).

Dual luciferase assays

Dual luciferase assays were performed using the DLR assay kit (Promega). Cells were seeded at 5×10⁵ cells per well in 24-well plates and incubated at 37°C. After 24 h HEK293T and U937 cells were transfected with dual luciferase reporter plasmids using the Lipofectamine-2000 (Invitrogen) and Amaxa neucleofection kit (Amaxa) following the manufacturer's suggested procedures. Cell lysates were prepared after an additional 24 h, and reagents for the assay were prepared following the manufacturer's instruction. Luminescence was measured using a 20/20n luminometer (Turner BioSystems). Ratios of firefly and Renilla luciferase derived luminescence were recorded, and efficiencies of either programmed ribosomal frameshifting or IRES-mediated initiation were determined as previously described (Pelletier and Sonenberg 1989; Harger and Dinman 2003).