Inhibition of translation initiation by volatile anesthetics involves nutrient-sensitive GCN-independent and -dependent processes in yeast

Abstract

Volatile anesthetics including isoflurane affect all cells examined, but their mechanisms of action remain unknown. To investigate the cellular basis of anesthetic action, we are studying Saccharomyces cerevisiae mutants altered in their response to anesthetics. The zzz3-1 mutation renders yeast isoflurane resistant and is an allele of GCN3. Gcn3p functions in the evolutionarily conserved general amino acid control (GCN) pathway that regulates protein synthesis and gene expression in response to nutrient availability through phosphorylation of the alpha subunit of eukaryotic initiation factor 2 (eIF2alpha). Hyperphosphorylation of eIF2alpha inhibits translation initiation during amino acid starvation. Isoflurane rapidly (in <15 min) inhibits yeast cell division and amino acid uptake. Unexpectedly, phosphorylation of eIF2alpha decreased dramatically upon initial exposure although hyperphosphorylation occurred later. Translation initiation was inhibited by isoflurane even when eIF2alpha phosphorylation decreased and this inhibition was GCN-independent. Maintenance of inhibition required GCN-dependent hyperphosphorylation of eIF2alpha. Thus, two nutrient-sensitive stages displaying unique features promote isoflurane-induced inhibition of translation initiation. The rapid phase is GCN-independent and apparently has not been recognized previously. The maintenance phase is GCN-dependent and requires inhibition of general translation imparted by enhanced eIF2alpha phosphorylation. Surprisingly, as shown here, the transcription activator Gcn4p does not affect anesthetic response.

Figure 1.

Anesthetic and high-temperature response of zzz2 and zzz3(gcn3) mutants. Approximately 5 × 10³ cells of the indicated strains were spotted on SC medium and incubated for 3 d at 30°C in the presence of various concentrations of isoflurane (Iso) or in the absence of isoflurane at 37.5°C. ZZZ2 ZZZ3 and GCN3 are alternative names for the RLK88-3C wild-type strain. (A) Growth of the zzz2-1 zzz3-1 double mutant is temperature sensitive and isoflurane resistant. (B) Deletion of GCN3 (gcn3Δ) renders cells resistant to isoflurane.

Figure 2.

Mutations affecting eIF2α phosphorylation increase resistance to isoflurane. Cells were spotted and incubated as described in the legend for Figure 1. Mutants in the left-hand panel contain precise deletions of the protein-encoding sequences of the indicated genes. Single-copy plasmids containing wild-type (YCpSUI2) or mutant (YCpSUI2-S51A) SUI2 genes were tested for anesthetic response in a sui2Δ strain background (right-hand panel). The SUI2 gene encodes the eIF2α protein. The eIF2α produced by the SUI2-S51A mutant cannot be phosphorylated on Ser-51 because of mutation of this residue to alanine.

Figure 3.

Isoflurane exposure affects eIF2α phosphorylation. (A) A logarithmically growing culture of P1417 was split and grown in the absence or presence of isoflurane (Iso) or 100 mM 3-amino triazole (3-AT) for various lengths of time. At the indicated times, cells were harvested. Extracts were prepared and fractionated by SDS-PAGE. Blots were probed with antibodies recognizing either total eIF2α or eIF2α phosphorylated at serine-51 (eIF2α∼P). An extract derived from a mutant (P1983) that cannot be phosphorylated at serine-51 because of mutation of the serine to alanine (S51A) is shown as a control. (B) Levels of phosphorylated eIF2α decrease rapidly during isoflurane exposure. Cells from a logarithmically growing culture of RLK88-3C were exposed to isoflurane as described for 5–15 min. Extracts were prepared and fractionated by SDS-PAGE as described. (C) Halothane exposure does not induce the rapid reduction in levels of phosphorylated eIF2α. A logarithmically growing culture of RLK88-3C was split and grown in the presence or absence of halothane for various lengths of time. At the indicated times, cells were harvested and extracts were treated as described above. (D) Exposure of a prototroph to isoflurane produces a rapid reduction of levels of phosphorylated eIF2α that do not recover. A logarithmically growing culture of P1480 was split and grown in the presence or absence of isoflurane. At appropriate times, cells were harvested and extracts were treated as described above. A similar extract prepared from the gcn2Δ mutant P1837 that cannot phosphorylate eIF2α is included as a control.

Figure 4.

Overexpression of a dominant negative GLC7 mutation (YEpglc7Δ209-312) does not affect the isoflurane-induced decrease of phosphorylated eIF2α. (A) Logarithmically growing cultures of RLK88-3C containing the indicated plasmids were split and grown in the presence or absence of isoflurane (Iso) for various lengths of time. At the indicated times, cells were harvested. Extracts were prepared and fractionated by SDS-PAGE. The blot was probed with antibodies recognizing total eIF2α or eIF2α phosphorylated at serine-51 (eIF2α∼P). (B) Overexpression of the dominant negative GLC7 truncation does not alter anesthetic response. Approximately 5 × 10³ cells from cultures of RLK88-3C transformed with the indicated plasmids were spotted on SC-ura medium and tested for response to isoflurane (Iso).

Figure 5.

Deletion of GCN4 (gcn4Δ) does not affect the anesthetic response of RLK88-3C. Cells were spotted and incubated as described in the legend for Figure 1. (A) Although gcn4Δ strains grow more slowly in the absence or presence of isoflurane, this mutation does not affect the isoflurane MIC. (B) As expected, gcn4Δ strains are hypersensitive to sulfometuron methyl (SM).

Figure 6.

A GCN2-dependent and a GCN2-independent pathway are involved in inhibition of translation initiation by isoflurane. Logarithmically growing cultures of various strains were split and incubated in the presence or absence of isofluarane (Iso) for 0, 15, 60, or 120 min before harvesting the cells. Extracts were prepared and polysomes were separated on sucrose gradients. The gradients were fractionated and scanned for UV absorbance at 254 nm. (A) Isoflurane rapidly inhibits translation initiation in RLK88-3C. This inhibition is maintained throughout the length of exposure to the anesthetic. (B) Translation initiation is rapidly inhibited in the gcn2Δ strain P1837 but the inhibition is not maintained. (C) Isoflurane does not inhibit translation in the prototrophic strain P1480.

Figure 7.

Response to isoflurane is GCN2-independent in prototrophic strains. Cells were spotted and incubated as described in the legend for Figure 1.

Figure 8.

Models for cellular responses to volatile anesthetics. (A) Inhibition of translation initiation by isoflurane involves both GCN and a GCN-independent pathway. The dashed parts of the lines indicate the uncertainty as to the times at which each pathway is required to produce normal anesthetic response. Growth inhibition occurs within 15 min of exposure (Keil et al., 1996; Wolfe et al., 1998). (B) Transport of amino acids and neurotransmitters in neuronal cells. Volatile anesthetics may induce anesthesia by affecting transport of amino acids and/or neurotransmitters in neuronal cells. When starved for these nutrients, cells may induce nutrient-sensing/deprivation response pathways such as GCN that could produce anesthesia. Three transport processes that may be affected by volatile anesthetics are indicated and described more completely in the text (indicated by the boxed numbers). Filled rectangle, transporter; unfilled rectangle, neurotransmitter receptor; circles, vesicles; aa, amino acid; aa^*, metabolic derivative of aa; nt, neurotransmitter.