The deoxyhypusine synthase mutant dys1-1 reveals the association of eIF5A and Asc1 with cell wall integrity

Abstract

The putative eukaryotic translation initiation factor 5A (eIF5A) is a highly conserved protein among archaea and eukaryotes that has recently been implicated in the elongation step of translation. eIF5A undergoes an essential and conserved posttranslational modification at a specific lysine to generate the residue hypusine. The enzymes deoxyhypusine synthase (Dys1) and deoxyhypusine hydroxylase (Lia1) catalyze this two-step modification process. Although several Saccharomyces cerevisiae eIF5A mutants have importantly contributed to the study of eIF5A function, no conditional mutant of Dys1 has been described so far. In this study, we generated and characterized the dys1-1 mutant, which showed a strong depletion of mutated Dys1 protein, resulting in more than 2-fold decrease in hypusine levels relative to the wild type. The dys1-1 mutant demonstrated a defect in total protein synthesis, a defect in polysome profile indicative of a translation elongation defect and a reduced association of eIF5A with polysomes. The growth phenotype of dys1-1 mutant is severe, growing only in the presence of 1 M sorbitol, an osmotic stabilizer. Although this phenotype is characteristic of Pkc1 cell wall integrity mutants, the sorbitol requirement from dys1-1 is not associated with cell lysis. We observed that the dys1-1 genetically interacts with the sole yeast protein kinase C (Pkc1) and Asc1, a component of the 40S ribosomal subunit. The dys1-1 mutant was synthetically lethal in combination with asc1Δ and overexpression of TIF51A (eIF5A) or DYS1 is toxic for an asc1Δ strain. Moreover, eIF5A is more associated with translating ribosomes in the absence of Asc1 in the cell. Finally, analysis of the sensitivity to cell wall-perturbing compounds revealed a more similar behavior of the dys1-1 and asc1Δ mutants in comparison with the pkc1Δ mutant. These data suggest a correlated role for eIF5A and Asc1 in coordinating the translational control of a subset of mRNAs associated with cell integrity.

Figure 1. The dys1-1 mutant shows a severe growth defect that is not associated with cell lysis.

(A) Serial dilutions of the wild type and dys1-1 mutant strains were plated onto YPD medium in the presence or absence of 1 M sorbitol at the indicated temperatures. (B) Growth curves of the wild type and dys1-1 mutant strains. The strains were grown at 25°C in YPD medium containing 1 M sorbitol, and the cell numbers were counted to monitor the 16- h growth rate. (C) The cells were grown as in Figure 1B, to mid-log phase, treated with methylene blue and counted to analyze the cell lysis. The quantification is shown relative to the wild type (100%). (D) Sensitivity to zymolyase. The cells were grown as in Figure 1C and treated with zymolyase. The turbidity was measured at the indicated time points after the cells were treated with SDS.

Figure 2. The dys1-1 mutant reveals a drastic reduction in Dys1 protein levels resulting in a reduction in hypusine-containing eIF5A.

(A) Determination of mutant Dys1 protein levels. Wild type and dys1-1 mutant strains were grown to mid-log phase at the permissive temperature in YPD medium containing 1 M sorbitol. The cells were lysed and 10 µg of total protein were blotted with the indicated antibodies. Samples were probed for eEF2 as a loading control. (B) Detection of hypusine-containing eIF5A of wild type and dys1-1 mutant strains. Total eIF5A was immunoprecipitated and subjected to SDS-PAGE. Hypusine-containing eIF5A was revealed by autoradiography. (C) Quantification of relative hypusination levels after analysis of hypusine-containing versus total eIF5A, comparing the dys1-1 mutant with the wild type (DYS1) and expressed the quantification as percent of wild type.

Figure 3. The reduction in hypusine formation in dys1-1 mutant results in a reduction in total protein synthesis, and the polysome profile is characteristic of translation elongation defects.

(A) The indicated strains were grown to mid-log phase, as in Figure 1B and radiolabeled [³H]leucine was added to the medium. The incorporation of [³H]leucine into total proteins was measured as described in the Materials and Methods. (B) Whole cell extracts (WCE) of the indicated strains were fractionated through centrifugation in a sucrose density gradient. Optical scans (OD_254nm) of the gradients are shown. The areas of the 80S and polysome peaks were compared to calculate the P/M ratio. The polysome profile fractions and the WCE were collected and blotted against the indicated antibodies. (C) Quantification of the ribosome-bound eIF5A relative to the amount of ribosomes (normalized by ribosomal protein L5) in the polysome profile fractions. The values obtained with the wild type strain were considered as 100% and those obtained with mutant strains were expressed as percentages of the wild type in the bar graphs.

Figure 4. Dys1 and eIF5A genetically interact with Pkc1 and Asc1.

(A) The wild type and mutant strains harboring wild type, inactive or constitutively active forms of Pkc1 protein under a galactose-inducible promoter were plated onto SC-ura medium containing 1 M sorbitol plus glucose (growth control) and plus galactose (inducible condition) and grown at 25°C for 3 days. (B) The indicated strains were plated onto medium not containing or containing 5-FOA and grown at 25°C for 3 days for plasmid shuffle. (C) The indicated strains harboring the empty vector, TIF51A or DYS1 high-copy plasmids (2μ) were grown at the permissive and restrictive conditions for 3 days.

Figure 5. Absence of Asc1 increases association of eIF5A to polysome profile fractions.

(A) Whole cell extracts (WCE) of the indicated strains were fractionated through centrifugation in a sucrose density gradient. Optical scans (OD_254nm) of the gradients are shown. The polysome profile fractions and the WCE were collected and blotted against the indicated antibodies. (B) Quantification of the ribosome-bound eIF5A relative to the amount of ribosomes (normalized by ribosomal protein L5) in the polysome profile fractions. The values obtained with the wild type strain were considered as 100% and those obtained with mutant strains were expressed as percentages of the wild type in the bar graphs.

Figure 6. The Dys1, Asc1 and Pkc1 mutants showed a distinguished sensitivity to compounds affecting cytoplasmic membrane and cell wall integrity.

(A) The strains were plated onto medium supplemented with the indicated drugs and grown at 25°C for 3 days. (B) The growth was measured relative to each respective isogenic wild type strain (100%).

Figure 7. Genetic interaction between Dys1 and Asc1 is related to Asc1 binding to the 40S ribosome subunit.

The indicated strains were plated onto medium not containing or containing 5-FOA and grown at 25°C for 3 days for plasmid shuffle.