Ribosome-associated complex binds to ribosomes in close proximity of Rpl31 at the exit of the polypeptide tunnel in yeast

Abstract

Ribosome-associated complex (RAC) consists of the Hsp40 homolog Zuo1 and the Hsp70 homolog Ssz1. The chaperone participates in the biogenesis of newly synthesized polypeptides. Here we have identified yeast Rpl31, a component of the large ribosomal subunit, as a contact point of RAC at the polypeptide tunnel exit. Rpl31 is encoded by RPL31a and RPL31b, two closely related genes. Delta rpl31a Delta rpl31b displayed slow growth and sensitivity to low as well as high temperatures. In addition, Delta rpl31a Delta rpl31b was highly sensitive toward aminoglycoside antibiotics and suffered from defects in translational fidelity. With the exception of sensitivity at elevated temperature, the phenotype resembled yeast strains lacking one of the RAC subunits or Rpl39, another protein localized at the tunnel exit. Defects of Delta rpl31a Delta rpl31b Delta zuo1 did not exceed that of Delta rpl31a Delta rpl31b or Delta zuo1. However, the combined deletion of RPL31a, RPL31b, and RPL39 was lethal. Moreover, RPL39 was a multicopy suppressor, whereas overexpression of RAC failed to rescue growth defects of Delta rpl31a Delta rpl31b. The findings are consistent with a model in that Rpl31 and Rpl39 independently affect a common ribosome function, whereas Rpl31 and RAC are functionally interdependent. Rpl31, while not essential for binding of RAC to the ribosome, might be involved in proper function of the chaperone complex.

Figure 1.

The zuotin subunit of RAC binds to ribosomes in close proximity of Rpl31. (A) Cross-linking of high-salt washed ribosomes after rebinding of purified RAC using the homobifunctional, amino-reactive cross-linker BS³ (+). As a control an aliquot was incubated without addition of BS³ (−). Aliquots with (+) or without (−) BS³-treatment were applied to immunoprecipitation reactions under denaturating conditions using protein A Sepharose beads coated with α-Zuo1. Zuo1-X, cross-link detected with α-Zuo1; U, material unbound; B, material bound to beads coated with α-Zuo1 after immunoprecipitation. Immunoblots were decorated with α-Zuo1. (B) Scale-up of the experiment shown in A. Coomassie-stained gel of the material bound to beads coated with α-Zuo1. The band labeled as Zuo1-Rpl31 was detected only in the sample treated with BS³. (C) Cross-linking of low-salt washed ribosomes with the zero-length cross-linker EDC (+). A mock reaction without addition of EDC is shown as a control (−). Zuo1-Rpl31 indicates the cross-link between Zuo1 and Rpl31. Immunoblots were developed with α-Zuo1 or α-Rpl31, as indicated. For details compare Materials and Methods. (D) Cross-linking of cell lysate with (+) or without (−) the cross-linker EDC. Zuo1-Ssz1 indicates the cross-link between Zuo1 and its partner subunit Ssz1 detected with α-Zuo1 and α-Ssz1. (E) Crystal structure of the large ribosomal subunit of the archaea Haloarcula marismortui (PDB: 1S72 and 1JJ2; Ban et al., 2000; Klein et al., 2001). rRNA in yellow, ribosomal proteins in magenta, ribosomal proteins of the platform around the tunnel exit (E) in violet. The nomenclature of the ribosomal proteins is according to Lecompte (Lecompte et al., 2002). The appendix “e” indicates ribosomal proteins confined to archaea and eukaryotes. H. marismortui proteins in close proximity to the tunnel exit correspond to yeast proteins: L24 is the homolog of Rpl26, L29 of Rpl35, L39e of Rpl39, L23 of Rpl25, L19e of Rpl19, L31e of Rpl31, and L22 of Rpl17. The structural representation was prepared with Pymol (http://pymol.sourceforge.net/).

Figure 2.

Yeast strains lacking Rpl31 display slow growth, sensitivity to high- and low-growth temperatures and suffer from defects in translational fidelity. (A) Phenotypic comparison of wild type, Δrpl31a, Δrpl31b, Δrpl31aΔrpl31b, and Δrpl31aΔrpl31b expressing RPL31a from a low copy number plasmid. Haploid yeast strains were grown to early log phase at 30°C on minimal glucose medium. Serial 10-fold dilutions containing the same number of cells were spotted onto YPD plates and were incubated as indicated. (B) Phenotypic comparison of wild type, Δrpl31aΔrpl31b, Δzuo1, Δrpl31aΔrpl31bΔzuo1, and Δrpl31aΔrpl31b expressing ZUO1 and SSZ1 from high copy number plasmids. Strains were grown as in A. Paro corresponds to 50 μg/ml paromomycin. Total cell extracts of strains shown in A and B were analyzed by immunoblotting using antibodies specifically recognizing Rpl31, and the RAC subunits, Ssz1 and Zuo1, as indicated. (C) Wild-type, Δrpl31aΔrpl31b, and Δzuo1Δssz1 strains were inoculated to the same OD₆₀₀ on YPD or into the same medium containing paromomycin as indicated. Cultures were grown until the control without paromomycin had reached OD₆₀₀ = 1.0, which was set to 100%. (D) Reporter construct for the determination of the frequency of stop codon read-through in vivo. An in-frame fusion between β-galactosidase and luciferase (lacZ-luc) served to determine the relative enzymatic activities upon equimolar expression of the two enzymes. Read-through efficiency was determined using the lacZ-STOP-luc construct in which the two enzymes were separated by a stop codon (for details compare Results and Materials and Methods). (E) Efficiency of stop codon read-through in wild-type and Δrpl31aΔrpl31b strains in the presence of increasing concentrations of paromomycin. Enzymatic activities were determined in extracts derived from cells grown to an OD₆₀₀ of 0.4–1 on YPD or YPD supplemented with paromomycin as indicated. Read-through frequency is expressed as the ratio of luciferase/β-galactosidase activity ([RLU]/[A₄₂₀]) obtained with the lacZ-STOP-luc construct normalized to the ratio obtained with lacZ-luc as described in Rakwalska and Rospert (2004). Experiments were performed at least in triplicate; error bars, SD.

Figure 3.

Phenotypic defects of yeast strains lacking Rpl31 can be partly suppressed by high levels of Rpl39. (A) Ribosome profiles of logarithmically growing wild-type and Δrpl31aΔrpl31b strains. Fractionation of the gradients was monitored at 254 nm. Asterisks indicate the position of halfmer ribosomes. (B) Analysis of large subunit concentration in wild type and Δrpl31aΔrpl31b. Total cell extract corresponding to 0.6 × 10⁷ cells of logarithmically growing wild type and Δrpl31aΔrpl31b was separated on 10% TRIS-Tricine gels. Purified Rpl17 was applied to the same gel and was analyzed by immunoblotting using α-Rpl17 as a large ribosomal subunit marker. The presence of untagged Rpl17a in the standard is due to the addition of total extract as a carrier (Raue et al., 2007). Note that the purified, His₆-tagged standard protein has a slightly higher molecular mass. (C) Quantification of ribosomes in wild type (white) and Δrpl31aΔrpl31b (gray). Densitometric analysis was performed to determine the range of linearity and to quantify protein concentrations in the total cell extracts. Error bars, SD. (D) Wild type, Δrpl31aΔrpl31b, Δrpl31aΔrpl31b, harboring the empty vector, or expressing either Rpl17a, Rpl24a, or Rpl39, Δrpl39, and Δrpl39 expressing Rpl31 were grown and analyzed as in Figure 2A. Paro corresponds to 50 μg/ml paromomycin. Total cell extracts were analyzed by immunoblotting using antibodies specifically recognizing Sse1 (as a loading control), Rpl17, Rpl31, Rps9, Rpl24, and Rpl39. (E) Ribosome profiles of logarithmically growing Δrpl31aΔrpl31b, Δrpl39, and Δrpl31aΔrpl31b expressing Rpl39 from a low copy plasmid were performed as described in A.

Figure 4.

Rpl31 is not essential for binding of RAC to ribosomes. (A) Ribosome profiles were run as described in Figure 3A. After fractionation aliquots were analyzed by immunoblotting using α-Ssz1/α-Zuo1 (RAC), α-Rps9, and α-Rpl24 as indicated. One-twentieth of the material loaded onto the sucrose gradient was analyzed as a total (T). (B) Quantification of Zuo1 in wild type and Δrpl31aΔrpl31b. Ribosomes were isolated under low-salt conditions. Analysis of isolated ribosomes was then performed as described in Figure 3, B and C using purified RAC and Rps9a as a standard for the quantifications. The experiments were performed in triplicate. The occupation of ribosomes with Zuo1 is given in percent (Raue et al., 2007). (C) Zuo1 does not form an alternative cross-link to a ribosomal protein in the absence of Rpl31. Cross-linking with BS³ was performed on low-salt washed ribosomes as described in Materials and Methods. Immunoblots were analyzed with α-Zuo1 and with α-Rpl17 as a control.

Figure 5.

A highly charged segment within Zuo1 affects the affinity for ribosomes. (A) Wild type, Δzuo1, Δzuo1 expressing Zuo1-15A, or Δzuo1 expressing Zuo1Δ282-331 were grown as in Figure 2A. paro, 200 μg/ml paromomycin. (B) Scan of a Zuo1 peptide library for segments interacting with ribosomes. Zuo1–15mer peptides on the cellulose membrane are shifted by one amino acid per spot. The cellulose membrane was incubated with high-salt washed ribosomes and subsequently ribosomes were transferred to nitrocellulose and analyzed by immunoblotting using α-Rpl16 as a ribosomal marker. Box1 corresponds to amino acids 119–130, box2 to amino acids 243–253, box3 to amino acids 257–269, box4 to amino acids 296–305. (C) Amino acid sequence of Zuo1. Interacting segments identified in B are indicated. Amino acids 282–331 are shaded in gray. Mutations introduced in box4, and adjacent amino acids are shown in the bottom panel. (D) Ribosome profiles of Δzuo1 expressing Zuo1-15A and of Δzuo1Δrpl31aΔrpl31b expressing Zuo1-15A were performed as described in Figure 3A. After fractionation aliquots were analyzed by immunoblotting using α-Ssz1/α-Zuo1 (RAC), α-Rps9, and α-Rpl24. (E) Zuo1-15A binding to ribosomes is destabilized in the presence of high concentrations of salt. Salt-dependent release of RAC from ribosomes derived from wild type, Δrpl31aΔrpl31b, Δzuo1 + Zuo1-15A, and Δzuo1Δrpl31aΔrpl31b + Zuo1-15A. Separation of extracts (tot) into a postribosomal supernatant (S) and a ribosomal pellet (P) was performed at increasing concentrations of KAcetate as indicated (mM KAc). Aliquots were separated on 10% TRIS-Tricine gels and subsequently analyzed by immunoblotting using α-Ssz1/α-Zuo1 (RAC), and α-Rpl17 as a ribosomal marker. Boxed is the concentration of KAcetate that resulted in ∼50% of RAC release.