Alu RNP and Alu RNA regulate translation initiation in vitro

Abstract

Alu elements are the most abundant repetitive elements in the human genome; they emerged from the signal recognition particle RNA gene and are composed of two related but distinct monomers (left and right arms). Alu RNAs transcribed from these elements are present at low levels at normal cell growth but various stress conditions increase their abundance. Alu RNAs are known to bind the cognate proteins SRP9/14. We purified synthetic Alu RNP, composed of Alu RNA in complex with SRP9/14, and investigated the effects of Alu RNPs and naked Alu RNA on protein translation. We found that the dimeric Alu RNP and the monomeric left and right Alu RNPs have a general dose-dependent inhibitory effect on protein translation. In the absence of SRP9/14, Alu RNA has a stimulatory effect on all reporter mRNAs. The unstable structure of sRight RNA suggests that the differential activities of Alu RNP and Alu RNA may be explained by conformational changes in the RNA. We demonstrate that Alu RNPs and Alu RNAs do not stably associate with ribosomes during translation and, based on the analysis of polysome profiles and synchronized translation, we show that Alu RNP and Alu RNA regulate translation at the level of initiation.

Figure 1

Secondary structure homology between Alu RNA and the SRP RNA Alu domain. (A) Secondary structure of the human SRP RNA. The SRP RNA is divided in two functional domains called S and Alu. The S domain of SRP binds nascent chains carrying a signal sequence while they emerge from the ribosome; the Alu domain mediates a transient delay in elongation. Boldface indicates the binding sites of SRP9/14 according to Refs (61) and (46). Three base pairs are formed between two loops and are indicated by dots. (B) Secondary structure of the synthetic Alu RNA used in this study. It was drawn based on a previously determined secondary structure (62) and adapted to the sequence of the Alu element of intron 4 of the α-Fetoprotein gene (Alu Y) (33). Boldface and dots indicate the binding sites of SRP9/14 and the tertiary base pairing between the two loops, respectively, by analogy to SRP RNA. Open arrow indicates the 3′ end of scAlu RNA (116 nt) and closed arrows the 5′ and 3′ ends of sRight RNA (155 nt). scAlu and sRight RNAs represent monomeric left and monomeric right arms, respectively.

Figure 2

Alu RNP purification on Superdex 200. (A) Denaturing acrylamide gel. Synthetic Alu RNA migrates as a single band with the expected size of 305 nt. (B) Native acrylamide gel. Alu RNA migrates in a defined band indicating that it is homogeneously folded. Trace amounts of an RNA dimer are also observed (star). (C) OD₂₅₄ elution profile of a Superdex 200 column. Free Alu RNA, blue; Alu RNA bound to recombinant SRP9/14, red. Fractions 7–10 (grey box) containing Alu RNA in complex with two SRP9/14 proteins were pooled for subsequent experiments. The second RNP peak most likely represents Alu RNA bound to one protein and free RNA. (D) Aliquots of 1 and 2 µl (I and II, respectively) of the purified RNP fraction were subjected to denaturing acrylamide gel electrophoresis after proteinase K digestion. (E) Aliquots of 1 and 2 µl (I and II, respectively) of the purified RNP fraction were subjected to immunoblotting with anti-SRP14 antibodies. (F) Native agarose gel electrophoresis of the purified RNP fraction and free RNA.

Figure 3

Quantification of the effects of purified Alu RNPs and Alu RNAs on protein synthesis. Wheat germ translation reactions programmed with cyclin, preprolactin (pPL), luciferase and PAI-2 mRNAs were supplemented with increasing amounts of (A) Alu RNP, (B) scAlu RNP, (C) sRight RNP, (D) Alu RNA, (E) scAlu RNA, (F) sRight RNA, (G) (−)h14 mRNA, (H) BC200 RNA and (I) SRP9/14. The translation products were analysed by SDS–PAGE (Supplementary Figure S1), quantified and normalized to the buffer control, which was set to 100%. The results represent the average of at least two independent experiments.

Figure 4

Effects of Alu RNP and Alu RNA on the translation of cytoplasmic RNA from HeLa cells. Wheat germ translation reactions were programmed with 5 ng/µl⁻¹ of cytoplasmic RNA and the translation products were analysed by SDS–PAGE (A). Lane 1, buffer control; lane 2, 660 nM Alu RNA; and lane 3, 660 nM Alu RNP. (B) Quantification of the results shown in (A). Total protein synthesis was determined by measuring the intensities of identical elongated squares covering the translation products in the approximate size range of 10–100 kDa in all three lanes. The results represent the average of two independent experiments and were normalized to the buffer control, which was set to 100%.

Figure 5

Alu RNP and Alu RNA act at the level of translation initiation. (A) Polysomes profile of wheat germ translation reactions programmed with cyclin mRNA in presence (black) or absence (grey) of 100 nM Alu RNP. (B) Idem (A) with PAI-2 mRNA. (C) Polysomes profile of wheat germ translation programmed with cyclin mRNA in presence (black) or absence (grey) of 300 nM Alu RNA. (D) Idem (C) with PAI-2 mRNA. Profiles were monitored by the incorporation of [³⁵S]methionine into the nascent chains (cpm).

Figure 6

Alu RNP and Alu RNA migrate in different fractions than ribosomes. (A) Polysome profile of a wheat germ translation reaction programmed with PAI-2 mRNA and supplemented with 300 nM Alu RNA. Northern blot analysis of the gradient fractions with probes against scAlu (B) and sRight RNAs (C) as well as against 28S rRNA (D). (E) Polysome profile of a wheat germ translation reaction programmed with PAI-2 mRNA and supplemented with 100 nM Alu RNP. Northern blot analysis of the gradient fractions with probes against scAlu (F) and sRight RNAs (G) and against 28S rRNA (H).

Figure 7

Alu RNP and Alu RNA do not influence translation elongation. Translation reactions containing [³⁵S]methionine and programmed with cyclin mRNA were allowed to initiate 2 min before the addition of edeine at a final concentration of 5 µM. After two more minutes at 26°C, Alu RNA and Alu RNP were added at final concentrations of 300 and 100 nM, respectively. Aliquots of the reaction were removed at the time points indicated and analysed by SDS–PAGE. Autoradiograms of reactions containing (A) buffer control, (B) Alu RNA and (C) Alu RNP. (D) The signals were quantified and normalized to the average of the time points 12 and 14, which were arbitrarily set to 100%. Circles, buffer control; triangles, Alu RNP; squares, Alu RNA. (E) Negative control reactions for translation initiation. Translation reactions programmed with cyclin mRNA were allowed to initiate 2 min before the addition of edeine at a concentration of 5 µM. After two more minutes of incubation, a second mRNA encoding preprolactin was added to the reaction, which was then incubated 20 min before being subjected to SDS–PAGE.