Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2011 Feb;31(3):542-56.
doi: 10.1128/MCB.01162-10. Epub 2010 Nov 22.

La-related Protein 4 Binds poly(A), Interacts With the poly(A)-binding Protein MLLE Domain via a Variant PAM2w Motif, and Can Promote mRNA Stability

Affiliations
Free PMC article

La-related Protein 4 Binds poly(A), Interacts With the poly(A)-binding Protein MLLE Domain via a Variant PAM2w Motif, and Can Promote mRNA Stability

Ruiqing Yang et al. Mol Cell Biol. .
Free PMC article

Abstract

The conserved RNA binding protein La recognizes UUU-3'OH on its small nuclear RNA ligands and stabilizes them against 3'-end-mediated decay. We report that newly described La-related protein 4 (LARP4) is a factor that can bind poly(A) RNA and interact with poly(A) binding protein (PABP). Yeast two-hybrid analysis and reciprocal immunoprecipitations (IPs) from HeLa cells revealed that LARP4 interacts with RACK1, a 40S ribosome- and mRNA-associated protein. LARP4 cosediments with 40S ribosome subunits and polyribosomes, and its knockdown decreases translation. Mutagenesis of the RNA binding or PABP interaction motifs decrease LARP4 association with polysomes. Several translation and mRNA metabolism-related proteins use a PAM2 sequence containing a critical invariant phenylalanine to make direct contact with the MLLE domain of PABP, and their competition for the MLLE is thought to regulate mRNA homeostasis. Unlike all ∼150 previously analyzed PAM2 sequences, LARP4 contains a variant PAM2 (PAM2w) with tryptophan in place of the phenylalanine. Binding and nuclear magnetic resonance (NMR) studies have shown that a peptide representing LARP4 PAM2w interacts with the MLLE of PABP within the affinity range measured for other PAM2 motif peptides. A cocrystal of PABC bound to LARP4 PAM2w shows tryptophan in the pocket in PABC-MLLE otherwise occupied by phenylalanine. We present evidence that LARP4 expression stimulates luciferase reporter activity by promoting mRNA stability, as shown by mRNA decay analysis of luciferase and cellular mRNAs. We propose that LARP4 activity is integrated with other PAM2 protein activities by PABP as part of mRNA homeostasis.

Figures

FIG. 1.
FIG. 1.
Human LARP4(1-286) preferentially binds poly(A) with a length requirement longer than 10 nucleotides. (A to D) EMSA of homopolymeric 20-mer RNAs for binding to the RNA binding domain of LARP4 protein, LARP(1-286), also referred to as LARP4-NTD. Twofold serial dilutions of LARP4(1-286) were used, with a 750-μM final concentration in lane 8, as indicated. (E) EMSA of A(20) and U(20) for La(1-235) and LARP4(1-286). (F to G) Comparison of LARP4(1-286) binding to otherwise identical 36-mer RNAs (see sequences in Materials and Methods) that end with a 3′ OH (F) or a 3′ phosphate (G) (see Materials and Methods). (H) LARP4(1-286) does not bind 10-mer RNAs. R1 and R2 are the same 36-nt RNAs used in panels F and G. (I) La(1-235) binds to U10. (J to L) Isothermal titration calorimetric analysis of LARP4(111-303) interactions with A(15) (J), U(15) (K), and A(10) (L). The Kd and other thermodynamic parameters derived from this analysis are reported in Table 1.
FIG. 2.
FIG. 2.
Human LARP4 cosediments with 40S subunits and polyribosomes and associates with RACK1. (A and B) Polysome profiles in the presence (A) and absence (B) of puromycin. Numbered fractions were collected, fractionated by SDS-PAGE, and immunoblotted onto a membrane that was probed, exposed, stripped, and reprobed to detect the proteins indicated between the panels. (C) Co-IP of RACK1 with FLAG-LARP4 (F-LARP4) but not the control protein, F-La, using anti-FLAG Ab. Cells were transfected with F-LARP4 or F-La and processed for IP. The upper gel shows input extracts (lysate) prior to IP using anti-FLAG for immunoblot detection. The lower gel shows the IP material using anti-RACK1 for immunoblot detection. (D) Co-IP of F-LARP4, but not the control protein F-LARP6 or F-La, with RACK1 using anti-RACK1 Ab for the IP. Cells were transfected with F-LARP4, F-LARP6, or F-La and processed for IP. The upper gel shows input extracts (lysate; lanes 1 to 3), the IPed material (lanes 4 to 6), and the supernatant (lanes 7 to 9), using anti-FLAG for immunoblot detection. The lower gel shows a region of the same blot in the upper gel after stripping and reprobing using anti-RACK1 for immunoblot detection.
FIG. 3.
FIG. 3.
LARP4 knockdown decreases cellular protein synthesis. (A and B) Polysome profiles of extracts made from cells treated with control siRNA (A) or siRNA directed to LARP4 (B). Quantitative areas under the 80S and polysome OD254 tracings are shown as numerical fractions above the panels. Immunoblots of the collected fractions for LARP4 and rpS6 are shown below the OD254 tracings. Ethidium-stained RNA gels showing 28S and 18S rRNAs and tRNA are shown below. (C) Coomassie blue-stained gel containing extracts from cells treated with control siRNA or siRNA directed to LARP4 or PABP (lanes C, L, and P, respectively) and pulsed for 30 min with [35S]methionine. Fifty, 100, and 150 μg of each extract was loaded in lanes 1 to 3, 4 to 6, and 7 to 9, respectively. (D) The gel in panel C was dried and exposed for autoradiography on a Fuji phosphorimager. (E) Quantitation of the total 35S was performed and expressed as [35S]Met incorporation per unit of protein on the y axis for each siRNA-treated cell extract on the x axis; the error bars reflect triplicate data. (F to H) Immunoblot showing relative levels of LARP4, PABP, and GAPDH from the three extracts used for panels C and D; a single membrane was incubated sequentially with the three Abs for panels H to J. LARP localizes to stress granules after exposure to arsenite. (I to P) After transfection with F-LARP4, cells were mock treated (I to L) or treated with arsenite (M to P), which is widely used to induce stress granules, and examined with anti-FLAG (green) and anti-TIA-1 (red) antibodies. LARP4 distribution was homogeneously cytoplasmic in the mock-treated cells (J), whereas a significant fraction became localized in punctate foci after arsenite treatment (N). The stress granule marker TIA-1 also organized into distinct foci after arsenite treatment (K versus O). Image merging revealed superimposed LARP4 foci and TIA-1 foci (P). Endogenous LARP4 also localized to stress granules (see Fig. S7Q to X in the supplemental material).
FIG. 4.
FIG. 4.
LARP4 contains a putative variant PAM2 (PAM2w) motif and interacts with PABP. (A) Sequence alignment of the PAM2 motifs of six Homo sapiens (Hs) proteins, eRF3 (elongation release factor 3), Paip1, Paip2, ataxin 2, and USP10, with the homologous sequences from LARP5, and LARP4. (B) Immunoblot of input extracts from cells transfected with vector only (vec), F-La, and F-LARP4 that were mock treated (−) or treated with RNase A (+) as visualized with anti-FLAG Ab. (C) The extracts shown in panel B were subjected to IP, and the products were examined by immunoblotting using anti-PABP Ab. (D) The blot in panel C was stripped and then probed using anti-FLAG Ab. ns, nonspecific. (E) Immunoblot, using anti-PABP Ab, of input extract (lane 1) and after IP with control IgG (lane 2) and anti-LARP4 Ab raised against a C-terminal peptide of LARP4 (lane 3). PABP remaining in the supernatants is shown in lanes 4 and 5. (F) Immunoblot of input extracts from cells after transfection with empty vector, F-LARP4, and F-LARP4-L15A-W22A (lanes 1 to 3) and after IP (lanes 4 to 6 and 7 to 9) visualized using anti-PABP Ab. (G) The membrane was stripped and reprobed using anti-Flag Ab.
FIG. 5.
FIG. 5.
Binding of the PAM2w peptide of LARP4 to the MLLE domain of PABP. (A) Isothermal titration calorimetry. Shown are baseline corrected thermograms (top) and integrated areas (bottom) of the heat released for binding of PAM2w-long (left) and PAM2w (right). The dissociation constant (Kd), enthalpy (ΔH), entropy (ΔS), and stoichiometry (N) are indicated. (B) X-ray crystal structure of LARP4 PAM2w bound to the MLLE domain of PABPC1. The PAM2w peptide (green) wraps around the surface of the MLLE domain, colored according to the electrostatic potential (negative in red, positive in blue) (PDB ID, 3PKN). (C) Electron density omit map (2Fo − Fc [where Fo is observed structure factor and Fc is calculated structure factor]; 1-standard deviation contour) of the bound PAM2w peptide showing the tryptophan residue. (D) Side-by-side comparison of PAM2w (green) from LARP4 and PAM2 peptide (gray) (PDB entry 3KUS) from Paip2 bound to MLLE. The peptide N termini show strikingly similar confirmations and interactions with the MLLE domain. (E) Comparison of C termini. In LARP4, tryptophan (W22) replaces the phenylalanine found in other PAM2 peptides (38-41). The larger indole ring is accommodated by a displacement of Gln560 of MLLE (not shown) to generate a shallow hydrophobic pocket.
FIG. 6.
FIG. 6.
Two LARP4 regions are required for association with PABP and polyribosomes. (A) Immunoblot of extracts after transient transfection with a subset of Flag-tagged constructs depicted in panel J. The membrane was sequentially probed with the Abs indicated on the right. (B) Immunoblot of IPed material probed as for panel A. (C) Further mapping of the PABP interaction region of LARP4 C terminal to the RRM; probed with anti-PABP. (D) The blot in panel C was probed with anti-FLAG Ab. (E to I) Five polysome profiles run in parallel after transfection with the indicated constructs. Fractions were analyzed by immunoblotting as indicated. (J) Schematic showing the F-LARP4 constructs used here and in Fig. 7 (N-terminl Flag is not shown) with summarized results on the right (+, positive correlation; −, negative correlation; ND, not determined). The PAM2w, LaM, and RRM motifs are shown. PBM refers to a putative PABP binding motif mapped here. PAM-S contains alanines at LARP4 residues 15 to 22.
FIG. 7.
FIG. 7.
The LARP4 RNA binding domain contributes to PABP and polysome association. (A and B) Two polysome profiles run in parallel after transfection with the indicated constructs, F-LARP4-WT and F-LARP4-M3, which contain mutations in the RNA binding domain (see the text); the 40S peaks are indicated. The fractions were analyzed by immunoblotting, as shown under the profiles. (C and D) PABP-LARP4 interaction is resistant to RNase A, which does not cleave poly(A) but is partially sensitive to RNase I. Shown are immunoblots of IPed material after mock treatment (no RNase) or incubation with RNase A or RNase I, as indicated, probed first for PABP (C) and then for F-LARP4 (D). (E) RNAs remaining in supernatants of samples 1 to 6 in panels C and D after IP.
FIG. 8.
FIG. 8.
The effects of LARP4 on transfected luciferase reporter activity reflect luciferase mRNA levels. (A to F) Cells transfected with siRNA (lanes 1 to 4) or F-LARP constructs (lanes 5 to 9) and corresponding controls were secondarily transfected with luciferase reporter plasmid, after which extracts were prepared for luciferase activity (A), Northern blotting (B to D), and immunoblotting (E to F). The blots in panels B to D reflect a single membrane sequentially probed for the RNAs indicated on the left. The immunoblots in panel E were processed using anti-LARP (left) or anti-FLAG (right). (G) The experiment represented in panels A to F was repeated in triplicate using a VA1 plasmid together with the luciferase reporter plasmid to normalize for transfection efficiency. A Northern blot was generated from each of the three experiments and probed for luciferase mRNA, GADPH mRNA for normalization for loading, and VA1 RNA for normalization for transfection. The graph shows luciferase mRNA levels after normalization for loading and transfection. The error bars reflect triplicate data.
FIG. 9.
FIG. 9.
LARP4 can promote mRNA stability. Cells were transfected with a mixture of F-LARP4 and luciferase plasmid or empty vector and luciferase plasmid. Forty hours later, the cells were treated with actinomycin D, and RNA was isolated at time zero and intervals thereafter as indicated. Three independent experiments were carried out, and a Northern blot was made from each (one of the blots sequentially probed is shown in Fig. S5 in the supplemental material). The blots were sequentially probed for the RNAs indicated above the x axes. The blots were also probed for 18S rRNA, GAPDH mRNA, and 7SK snRNA, which were used to normalize for loading. Each graph shows the decay curves for LARP4-transfected and empty-vector-transfected cells; each line reflects the time course for the RNA as corrected for by either 18S rRNA, 7SK snRNA, or GAPDH mRNA as indicated; the error bars reflect data from the three data sets.

Similar articles

See all similar articles

Cited by 32 articles

See all "Cited by" articles

Publication types

MeSH terms

LinkOut - more resources

Feedback