Hsp90 and cochaperones have two genetically distinct roles in regulating eEF2 function

Abstract

Protein homeostasis relies on the accurate translation and folding of newly synthesized proteins. Eukaryotic elongation factor 2 (eEF2) promotes GTP-dependent translocation of the ribosome during translation. eEF2 folding was recently shown to be dependent on Hsp90 as well as the cochaperones Hgh1, Cns1, and Cpr7. We examined the requirement for Hsp90 and cochaperones more closely and found that Hsp90 and cochaperones have two distinct roles in regulating eEF2 function. Yeast expressing one group of Hsp90 mutations or one group of cochaperone mutations had reduced steady-state levels of eEF2. The growth of Hsp90 mutants that affected eEF2 accumulation was also negatively affected by deletion of the gene encoding Hgh1. Further, mutations in yeast eEF2 that mimic disease-associated mutations in human eEF2 were negatively impacted by loss of Hgh1 and growth of one mutant was partially rescued by overexpression of Hgh1. In contrast, yeast expressing different groups of Hsp90 mutations or a different cochaperone mutation had altered sensitivity to diphtheria toxin, which is dictated by a unique posttranslational modification on eEF2. Our results provide further evidence that Hsp90 contributes to proteostasis not just by assisting protein folding, but also by enabling accurate translation of newly synthesized proteins. In addition, these results provide further evidence that yeast Hsp90 mutants have distinct in vivo effects that correlate with defects in subsets of cochaperones.

Fig 1. Effect of cochaperone deletion or mutation 6x-His eEF2 levels.

His-eEF2 was isolated from the indicated strain (S1 Table) using nickel resin. Proteins bound to the nickel resin were separated using SDS-PAGE and visualized using stained gels (top). Levels of eEF2 were also detected using immunoblot analysis of whole cell lysates (below). Lane 1: empty vector (YcPlac111). eEF2 levels in the immunoblots in A and B are the combined signal from endogenous eEF2 + plasmid His-eEF2. Anti-Pgk1was used as a loading control. A. Effect of cochaperones previously shown to affect eEF2 folding, as well as a protein (Dph2) required for eEF2 modification. B. Effect of additional Hsp90 cochaperones. C. Quantification of the changes in His-eEF2 levels in panels A and B. The level of His-eEF2 bound to resin in each strain was quantified as described in Materials and Methods. The mean values and standard deviations of three biological replicates, along with a representative of each, are shown. Cochaperones previously linked to eEF2 function are shaded dark gray. Statistical significance was evaluated with GraphPad Prism using Mixed-effects analysis (* P ≤ 0.05; ** P ≤ 0.01; *** P ≤ 0.001). Non-significant values not shown.

Fig 2. Effect of Hsp90 mutation on 6x-His eEF2 levels.

A. Schematic of the different Hsc82 mutants and the approximate location of the mutants. The loading mutants disrupt interaction of Hsp70 and Hsp90 during client loading and the closing mutants disrupt formation of the closed state. The reopening mutants cluster in regions required for ATP binding and/or hydrolysis. The G424D mutation is in a distinct region. Additional details about the Hsc82 mutants are listed in S2 Table. B. Wild-type or mutant Hsp90 (the Hsc82 isoform) was expressed in strain JJ816 (hsc82hsp82). His-eEF2 was isolated using nickel resin. Proteins bound to the nickel resin were separated using SDS-PAGE and visualized using stained gels (top). Levels of eEF2 were also detected using immunoblot analysis of whole cell lysates (below). eEF2 levels in the immunoblot are the combined signal from endogenous eEF2 plus plasmid His-eEF2. Anti-TIM44 was used as a loading control. Lane 1: empty vector (YCplacIII). B. The level of His-eEF2 bound to resin in each strain was quantified as described in Materials and Methods. The mean values and standard deviations of biological replicates, along with a representative of each, are shown. Hsc82 mutants in the reopening category are shaded in dark gray. Statistical significance was evaluated with GraphPad Prism using Mixed-effects analysis (* P ≤ 0.05; ** P ≤ 0.01). Non-significant values (P ≥ 0.05 not shown).

Fig 3. Deletion of HGH1 enhances the growth defect of some hsc82 mutant strains.

Plasmids expressing either wild-type Hsc82 or the indicated mutant were transformed into isogenic hsc82hsp82/URA3-HSP82 (JJ816) or hgh1hsc82hsp82/URA3-HSP82 (JJ1471) strains. A and B. Strains were streaked onto media containing 5-FOA, which counterselects for the HSP82 plasmid. Pictures were taken after three days at 30°C. At least three independent assays were conducted, with a representative shown. C. Growth of strains in B expressing WT HSC82, hsc82-S25P or -E377A after 5 days on 5-FOA at 30°C. D. Strain hgh1hsc82hsp82/URA3-HSP82 (JJ1471) expressing WT HSC82 or hsc82-Q380K was transformed with a plasmid expressing HGH1 or empty vector (pRS424). Transformants were patched onto selective media in triplicate, then replicated onto either selective media or plates containing 5-FOA and grown for two days at 30°C.

Fig 4. Effect of eEF2 mutation on growth and steady state level.

A. Growth of wild-type or mutant His-EF2 in eft1eft2 strain (JJ1472). Strains expressing indicated His-eEF2 plasmid were grown overnight, then serially diluted 10-fold and grown for two days at the indicated temperature. Growth of eft1eft2 strains expressing each eEF2 mutant was normalized to the growth of the cells expressing WT eEF2 strain. B. His-eEF2 was isolated using nickel resin and analyzed by SDS-PAGE followed by staining with Coomassie Blue and immunoblot analysis. Lane 1 is a control that lacks His-tagged protein (JJ1472). The level of His-eEF2 bound to resin in each strain was quantified as described in Materials and Methods. The mean values and standard deviations of three biological replicates, along with a representative of each, are shown. C. Growth of WT or mutant eEF2 in eft1eft2 strain (JJ1472) or eft1eft2hgh1 strain (JJ1481) after 3 d at 23°C. Growth was normalized to growth of WT or mutant eEF2. D. Growth of WT or mutant eEF2 in eft1eft2 strain in the presence of empty vector (pRS424) or HGH1 (prs424-HGH1) after 2 d on selective media at 30°C. The growth defect on selective media is stronger than when grown on rich media (YPD, in A). Growth was normalized to growth of WT eEF2 in the presence of empty vector (EV). Three biological replicates were obtained, with representative pictures of each shown. Statistical significance was evaluated with GraphPad Prism using one way ANOVA. (* P ≤ 0.05; ** P ≤ 0.01; *** P ≤ 0.001; **** P ≤ 0.0001).

Fig 5. Effect of eEF2 mutation sensitivity to diphtheria toxin.

A. WT cells (762) or dph2, or jjj3 strains were transformed with empty vector (-, pB656) or pLMY101, which expresses the catalytic domain of DT under the GAL1 promoter [41]. Cells were struck out onto selective (- uracil) plates containing 2% glucose (as a control) or 2% galactose and grown for four days at 30°C. B. Strain 1472 (eft1eft2) expressing wild-type or mutant His-EF2 was transformed with plasmid pLMY101. A dph2 strain harboring pLMY101 was used as a control. Cells were struck out onto selective (- uracil) plates containing 2% glucose (as a control) plates containing 2% galactose and grown for four days at 30°C. Three biological replicates were obtained, with representative pictures of each shown.

Fig 6. Effect of cochaperone deletion or alteration on sensitivity to diphtheria toxin.

A. WT cells (JJ762), cells lacking HGH1 (JJ1465), CPR7 (JJ1115) or DPH2 (JJ1449) or a cns1 disruption strain (JJ21) expressing WT CNS1 or cns1-G90D were transformed with plasmid pLMY101 and struck out onto selective (- uracil) plates containing 2% glucose (as a control), 1% raffinose and 1% galactose, or 2% galactose and grown for four days at 30°C. B. WT cells (JJ762), cells lacking CPR6 (JJ1138), STI1 (JJ623), AHA1 (JJ73), SBA1 (JJ543), or DPH2 (JJ1449). Three biological replicates were obtained, with representative pictures of each shown.

Fig 7. Effect of Hsc82 cochaperone deletion on sensitivity to diphtheria toxin.

Strain JJ117 expressing WT or mutant Hsc82 was transformed with plasmid pLMY101 and struck out onto selective (- uracil) plates containing 2% glucose (as a control), 1% raffinose and 1% galactose, or 2% galactose and grown for four days at 30°C. A dph2 strain harboring pLMY101 was used as a control. Three biological replicates were obtained, with representative pictures of each shown.

Fig 8. Model of Hsp90 and cochaperone interaction with eEF2.

A. Folding of eEF2 is assisted by Hgh1, Cns1 and Cpr7. Hsp90 binds the eEF2 cochaperone complex, further assisting folding. Reduced function of Hsp90 or these cochaperones results in inefficient folding and/or aggregation of eEF2. B. Hsp90 and Sti1 promote maturation of proteins required for efficient processing of the diphthamide modification, such as Dph6 and Dph7. Reduced function of Hsp90 and/or Sti1 may result in inefficient modification of eEF2, resulting in resistance to DT.