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

Abstract

Previously, we demonstrated that treatment of monocytic cells with IFN-gamma causes release of ribosomal protein L13a from the 60S ribosome and subsequent translational silencing of Ceruloplasmin (Cp) mRNA. Here, evidence using cultured cells demonstrates that Cp mRNA silencing is dependent on L13a and that L13a-deficient ribosomes are competent for global translational activity. Human monocytic U937 cells were stably transfected with two different shRNA sequences for L13a and clonally selected for more than 98% abrogation of total L13a expression. Metabolic labeling of these cells showed rescue of Cp translation from the IFN-gamma mediated translational silencing activity. Depletion of L13a caused significant reduction of methylation of ribosomal RNA and of cap-independent translation mediated by Internal Ribosome Entry Site (IRES) elements derived from p27, p53, and SNAT2 mRNAs. However, no significant differences in the ribosomal RNA processing, polysome formation, global translational activity, translational fidelity, and cell proliferation were observed between L13a-deficient and wild-type control cells. These results support the notion that ribosome can serve as a depot for releasable translation-regulatory factors unrelated to its basal polypeptide synthetic function. Unlike mammalian cells, the L13a homolog in yeast is indispensable for growth. Thus, L13a may have evolved from an essential ribosomal protein in lower eukaryotes to having a role as a dispensable extra-ribosomal function in higher eukaryotes.

FIGURE 1.

Generation of L13a-depleted U937 monocytic cells by stable expression of shRNA. (A) Designs for the shRNA1 and shRNA2 for targeted depletion of endogenous L13a by RNA silencing. (B) Design of the vector for making the recombinant lentiviruses harboring shRNA sequence against L13a. (C) Identification of single U937 cell clones with abrogated expression of endogenous L13a by immunoblot analysis. U937 cells expressing L13a shRNA1, shRNA2, or control shRNA were made by lentivirus-mediated transduction. Transduced cells were clonally purified by serial dilution and selected in blasticidin. Abrogation of L13a expression was tested by immunoblot analysis of 100 μg of the total lysate with anti-L13a (upper panel) and reprobed with anti-actin antibody (lower panel) as a control. (D) Abrogation of L13a rescues Cp translation from IFN-γ mediated translational silencing. U937 cells expressing shRNA1, shRNA2, or control shRNA were treated with IFN-γ for 0, 8, or 24 h. At the end of each interval cells were metabolically labeled by incubation with [³⁵S]-methionine in methionine-free medium for 2 h. Lysates were subject to immunoprecipitation (IP) with rabbit anti-human Cp IgG and resolved by SDS-PAGE. The radiolabeled bands were detected by autoradiography on X-ray film. The whole image comes from a single gel run and a single experiment, but a duplicate and identical shRNA2 lane was deleted.

FIGURE 2.

Depletion of L13a does not alter ribosome biogenesis. (A) Abrogation of L13a by RNA silencing does not alter 90S pre-ribosome formation. Nuclear extracts were prepared from U937 cells expressing shRNA1, shRNA2, control shRNA, or untransduced cells. The extracts were layered over 10%–40% sucrose gradients and resolved by centrifugation. After centrifugation the fractions were unloaded by upward displacement using an ISCO gradient fractionation system. UA-6 UV detector recorded continuous OD readout at A254. The presence of 90S pre-ribosomes was detected by the indicated peak. (B) Authenticity of the 90S pre-ribosome fraction. Total RNA was isolated from the 90S peak fractions obtained from nuclear lysate and 80S peak fractions obtained from post-nuclear lysate of U937 cells. The RNA was subjected to RT-PCR amplification using primers made against the 5′ external transcribed spacer (ETS) and 18S sequence present in the unprocessed 47S rRNA (left panel). The amplification products of the expected size were observed (right panel). (C) Abrogation of L13a by RNA silencing does not alter ribosomal RNA processing. U937 cells expressing shRNA1, shRNA2, or control shRNA were pulse labeled with [³H]-uridine for 1 h followed by chase in nonradioactive medium for 0, 45, or 90 min. RNAs were extracted at the end of indicated chase period, resolved, and transferred to Hybond N+ membrane (Amersham). Dried membranes were subjected to autoradiography. Turnover of the 47S precursor rRNA to the processed and matured 28S and 18S rRNA was observed. (D) Maintenance of the effective L13a depletion in the cells used in the above experiments. The depletion of L13a was verified by immunoblot analysis of 100 μg of the total cell lysates made from the wild-type U937 cells or cells expressing shRNA1, shRNA2, or control shRNA with anti-L13a antibody (upper panel). For loading control the same blot was reprobed with anti-actin antibody (lower panel).

FIGURE 3.

Depletion of L13a significantly inhibits the methylation of rRNA. U937 cells expressing shRNA1, shRNA2, or control shRNA were pulse labeled with [methyl-³H]-methionine for 1 h. At the end of the pulse period RNA was extracted, resolved, and transferred to Hybond N+ membrane (Amersham). Dried membranes were subjected to autoradiography. Methylation of newly synthesized rRNA is shown by incorporation of radioactive methyl methionine (upper panel). Total rRNA synthesis was determined as a control by pulse labeling the same cells in parallel by [³H]-uridine (middle panel). Steady-state levels of 28S and 18S rRNA were determined by ethidium bromide staining of the membrane from the methylation experiment (lower panel).

FIGURE 4.

Depletion of L13a has no effect on cytoplasmic polysome formation, total protein synthesis, or cell proliferation. (A) Abrogation of L13a by RNA silencing does not alter the ability to form translationally active polysomes. Exponentially growing U937 cells expressing shRNA1, shRNA2, or control shRNA were lysed, and polysomes were resolved by centrifugation through 10%–40% sucrose gradients and 40S, 60S, 80S, and active polysomes were identified by collecting the fractions through continuous monitoring of UV absorption. (B) Total protein synthesis remains unaffected upon L13a depletion by RNA silencing. U937 cells expressing shRNA1, shRNA2, or control shRNA were metabolically labeled with Trans [³⁵S] label in methionine-cysteine free medium for 90 min. After the labeling, cells were hydrolyzed by NaOH followed by tricholoroacetic acid (TCA) precipitation on filter. Filters were washed and counted using a liquid scintillation counter. Protein concentrations of lysates were measured and incorporation of radioactive amino acids into the TCA precipitates were expressed as CPM/μg of protein. The results of each experiment are the average of triplicate and the error bar was drawn on the basis of three independent experiments. (C) Maintenance of effective L13a depletion. The depletion of L13a was verified by immunoblot analysis with anti-L13a antibody of the 100 μg of total cell lysates made from wild-type U937 cells or cells expressing shRNA1, shRNA2, or control shRNA (upper panel). For loading control the same blot was reprobed with anti-actin antibody (lower panel). (D) L13a depletion by RNA silencing has no effect on cell proliferation. Cell proliferation was measured by counting the number of viable cells after 0, 24, 48, and 72 h in culture using an MTT cell proliferation assay kit (ATCC) following the manufacturer's suggested procedure. The results of each experiment are the average of triplicate and the error bar was drawn on the basis of three independent experiments.

FIGURE 5.

Depletion of L13a by RNA silencing does not affect programmed −1 ribosomal frameshifting. (Top panel) Schematic of control and test bicistronic reporter plasmids. (Bottom panel) U937 cells expressing shRNA1, shRNA2, or control shRNA were transfected with bicistronic vectors harboring −1 ribosomal frameshift signals from either SARS-CoV or HIV-1 inserted between the Renilla and Firefly luciferase genes (R Luc and F Luc). A bicistronic vector lacking frameshift signals was used as the readthrough control. Rates of programmed −1 ribosomal frameshifting was expressed by dividing the ratio of F Luc and R Luc in the test plasmid by that of the control plasmid as previously described (Harger and Dinman 2003). The results of each experiment are the average of triplicate and the error bar was drawn on the basis of three independent experiments.

FIGURE 6.

Depletion of L13a inhibits IRES activity. (A) Design of the bicistronic constructs harboring the IRES elements from p53, p27, and SNAT2 between R Luc and F Luc. (B–F) Bicistronic reporters were transfected into the U937 cells expressing shRNA1, shRNA2, or control shRNA by using an Amaxa nucleofection kit (Amaxa). The IRES activities of p53 (B), p27 (C), SNAT2 (D), HCV (E), and CrPV (F) were measured by measuring the ratio of F Luc and R Luc by using a Dual Luciferase Assay (DLR) kit from Promega following the manufacturer's suggested procedure. The results of each experiment are the average of triplicate and the error bar was drawn on the basis of three independent experiments.

FIGURE 7.

Inhibition of rRNA methylation inhibits IRES activity. (A) The methylation inhibitor cycloleucine does not inhibit the cellular level of L13a. U937 cells were treated or not treated with cycloleucine (2 mg/mL) for 8 h. At the end of cycloleucine treatment total cell lysate was made and was subjected to immunoblot analysis with anti-L13a antibody (upper panel). To check equal loading the same blot was reprobed with anti-actin antibody (lower panel). (B) Cycloleucine inhibits de novo methylation of 47S rRNA. U937 cells were pretreated with cycloleucine (2 mg/mL) for 7 h followed by pulse labeling with [methyl-³H]-methionine for 1 h. At the end of labeling the total RNA was extracted, resolved in 1% denaturing agarose gel, transferred to nylon membrane, and subjected to autoradiography. The band of de novo methylated RNA was observed in control cells but not in cycloleucine-treated cells (upper panel). To check the equal loading of RNA, the 28S and 18S bands were shown on the membrane after transfer (lower panel). (C) Cycloleucine inhibits p53, p27, and SNAT2 IRESes but not the HCV or CrPV IRES. U937 cells were treated with cycloleucine (2 mg/mL) for 7 h followed by transfection with the bicistronic reporter by using an Amaxa nucleofection kit (Amaxa). The IRES activities of p53, p27, SNAT2, HCV, and CrPV were measured by measuring the ratio of Firefly luciferase and Renilla luciferase activities using a Dual Luciferase Assay (DLR) kit from Promega following the manufacturer's suggested procedure. The results of each experiment are the average of triplicate and the error bar was drawn on the basis of three independent experiments.