The RNase Rny1p cleaves tRNAs and promotes cell death during oxidative stress in Saccharomyces cerevisiae

Abstract

The cellular response to stress conditions involves a decision between survival or cell death when damage is severe. A conserved stress response in eukaryotes involves endonucleolytic cleavage of transfer RNAs (tRNAs). The mechanism and significance of such tRNA cleavage is unknown. We show that in yeast, tRNAs are cleaved by the RNase T2 family member Rny1p, which is released from the vacuole into the cytosol during oxidative stress. Rny1p modulates yeast cell survival during oxidative stress independently of its catalytic ability. This suggests that upon release to the cytosol, Rny1p promotes cell death by direct interactions with downstream components. Thus, detection of Rny1p, and possibly its orthologues, in the cytosol may be a conserved mechanism for assessing cellular damage and determining cell survival, analogous to the role of cytochrome c as a marker for mitochondrial damage.

Figure 1.

Rny1p is the endonuclease responsible for yeast tRNA cleavage. (A) 5′ tRNA-His(GTG) probe. tRNA cleavage in WT yeast during stationary phase entry. WT yeast were grown for 6 d. (B) 5′ tRNA-His(GTG) probe. tRNA cleavage does not occur in yeast lacking RNY1. rny1Δ yeast were grown as in A. The contrast in panels A and B were adjusted to similar levels to facilitate comparison. (C) 3′ tRNA-Met(CAT) probe. Expression of RNY1 from a plasmid restores tRNA cleavage. The pGAL-RNY1 vector (+) or the empty vector pRS426 (−) were transformed into WT and rny1Δ cells. Yeast were grown to mid-log phase. (D) 3′ tRNA-Met(CAT) probe. Overexpression of rny1-ci in rny1Δ cells does not rescue tRNA cleavage; compare this with the overexpression of RNY1. Yeast were grown as in C. (E) 3′ 25s rRNA probe (arrow). rRNA cleavage in WT yeast during stationary phase entry. This is a reprobing of the blot shown in A. (F) 3′ 25s rRNA probe (arrow). rRNA cleavage in rny1Δ yeast during stationary phase entry. This is a reprobing of the blot shown in B. (G) 3′ tRNA-GluD probe. tRNA cleavage does not occur during postharvest cell lysis. WT (yRP840) and rny1Δ strains carrying a D. discoideum tRNA (tRNA-GluD) were cocultured or grown separately for 3 d in selective medium. (H) 5′ tRNA-His(GTG) probe. Overexpression of human RNASET2 in an rny1Δ strain rescues tRNA cleavage. Yeast were grown as in C. tRNA illustrations indicate full-length and fragment species. L = φX174/Hinf 1 ladder (sizes are indicated in nucleotides). Experiments were repeated at least three times; representative blots are shown.

Figure 2.

Rny1p localization before and during oxidative stress. (A) Localization of Rny1p-GFP during log phase and after a 20-min exposure to 3 mM H₂O₂. (B) Western blot analysis of Rny1p-GFP or Pep4p-GFP levels in cells before and after 1 h exposure to 3 mM H₂O₂. Approximate molecular masses are indicated in kD. (C) Localization of Pep4p-GFP during log phase and after a 20-min exposure to 3 mM H₂O₂. The vacuolar luminal dye CMAC was used to mark vacuoles. Experiments were repeated at least three times; representative examples are shown.

Figure 3.

RNY1 promotes cell death during stationary phase and oxidative stress. (A) Overexpression of RNY1 affects viability. Yeast carrying the pGAL-RNY1 (OE RNY1) or an empty vector were grown for the indicated number of days. 10-fold serial dilutions of cells were plated. (B) Overexpression of RNY1 sensitizes cells to oxidative stress. Indicated strains were grown to mid-log phase, then exposed to 0 or 3 mM H₂O₂ for 1 h. 10-fold serial dilutions were plated. (C) Overexpression of human RNASET2 in an rny1Δ strain results in a growth defect. Yeast carrying the pGAL-RNY1 (OE RNY1), the pGAL-RNASET2 (OE RNASET2), or vector were grown as in A. 10-fold serial dilutions were plated. (D) Deletion of RNY1 rescues the bir1Δ growth defect. Indicated strains were grown for 3 d, then 10-fold serial dilutions were plated. (E) Overexpression of RNY1 exacerbates the bir1Δ growth defect. bir1Δ cells carrying either the pGAL-RNY1 plasmid (OE RNY1) or vector were grown for 3 d, then 10-fold serial dilutions were plated. (F) Overexpression of RNY1 exacerbates the yap1Δ growth defect in response to oxidative stress. yap1Δ cells carrying either the pGAL-RNY1 plasmid (OE RNY1) or vector were grown as in E. Experiments were repeated at least three times; representative experiments are shown. pRS426 was the empty vector control.

Figure 4.

Catalytically inactive and WT RNY1 both promote cell death during oxidative stress. (A) Overexpression of RNY1 or rny1-ci results in a growth defect. Indicated strains were grown for 3 d, then 10-fold serial dilutions of cells were plated to assess viability. (B) Overexpression of RNY1 or rny1-ci sensitizes cells to oxidative stress. Indicated strains were grown to mid-log phase and exposed to 3 mM H₂O₂ for 1 h, then 10-fold serial dilutions were plated. (C) Overexpression of RNY1 or rny1-ci exacerbates the yap1Δ growth defect in response to oxidative stress. Strains were grown as in A. (D) Expression of RNY1 or rny1-ci at endogenous levels in a bir1Δ rny1Δ strain results in reversion to the bir1Δ phenotype. Cells were grown as in A. Experiments were repeated at least three times; representative experiments are shown.

Figure 5.

Organellar release of proteins involved in stress responses. This model compares the cytochrome c–apoptosis pathway to how release of Rny1p from the vacuole might affect cell death after stress or damage.