Study of the involvement of eIF5A in the endoplasmic reticulum stress response in Saccharomyces cerevisiae

The eukaryotic translation factor 5A (eIF5A) is highly conserved in archaea and eukaryotes and undergoes a unique and essential post-translational modification called hypusination. Although a function for eIF5A was initially suggested at the initiation of translation, it was in the elongation step that its function was best demonstrated. Previous studies have shown that eIF5A has a role in protein translocation by the co-translational pathway and that mutants of this factor have increased levels of proteins involved with endoplasmic reticulum (ER) stress. Therefore, in this work, the correlation of eIF5A with the stress response pathway of ER (the Unfolded Protein Response - UPR) was studied. Considering that the Hac1 transcription factor is an essential element of the UPR pathway in Saccharomyces cerevisiae, it was first verified whether eIF5A affects the cytoplasmic splicing of the HAC1 mRNA, which is dependent on the activation of Ire1, a sensor of unfolded proteins in the RE. To test it, we did the quantification of the mature and immature mRNA levels of HAC1 by means of qPCR and semi-quantitative RT-PCR assays, which did not reveal any difference in the behavior of the eIF5A mutant hyp2-3 in comparison to the wild-type. On the other hand, different mutants of eIF5A show sensitivity to DTT and resistance to tunicamycin, a behavior not yet described in the literature and that differs from mutants of the SRP pathway (co-translocation pathway for RE) and also this behavior of eIF5A mutants was not modified by overexpression of SRP. In addition, the eIF5A mutant phenotype in DTT and tunicamycin differs from that of knockout strains of ribosomal proteins, knockout of a ribosome biogenesis factor or a dominant-negative mutant of eEF2 (translation elongation factor 2). Thus, these results suggest that the sensitivity to DTT and resistance to tunicamycin upon decreased eIF5A function is not an effect related to the SRP pathway or a general translation defect. Furthermore, the tunicamycin resistance phenotype of the eIF5A mutants has been shown to be Ire1 or Hac1 independent. To better understand these UPR-related eIF5A phenotypes, we determined the proteome profile (using the Synthetic Genetic Array strategy and the S. cerevisiae GFP colection) of the hyp2-3 mutant on treatment with DTT, tunicamycin or in the untreated control condition. Analysis of these results revealed proteins that are decreased in hyp2-3 when compared to the wild-type and that could contribute to explain the phenotypes observed for this mutant. Finally, analysis of this proteomic profile and other data from this study do not suggest a pre-activation of UPR with the decrease of eIF5A function, as observed in a recent work using HeLa cells. (AU)