Elimination of Snf1 reduces eIF2α-P in histidine-starved cells. (A) Strains H1642 (WT), H1895 (gcn2Δ), HQY343 (snf1Δ) and HQY344 (snf1Δ gcn2Δ) were grown in liquid SC, and serial dilutions were spotted onto synthetic complete (SC) or SC plates with 3AT and incubated at 30°C for 2 or 4 days. (B) Transformants of HQY343 (GCN2 snf1Δ) with the empty vector, p235 (GCN4^c, lacking uORFs 3 and 4), p238 (GCN4^c, no uORFs), pHQ1162 (lc SNF1), p722 (lc GCN2), or p630 (hc GCN2) were replica printed to 3AT medium as for panel A. These strains exhibited equal growth on SC plates without 3AT (not shown). (C) Strains H1895 (gcn2Δ), H1642 (WT), and HQY343 (snf1Δ) were grown in SC-His, diluted in SC-His to an optical density at 600 nm of 0.2, cultured for 6 h (−3AT) or 5 h prior to treatment with 10 mM 3AT for 1 h (+3AT). WCEs were subjected to Western analysis with antibodies specific for eIF2α-P or total eIF2α. (D) Strain H1642 (WT) and transformants of snf1Δ mutant strain HQY343 with the empty vector, p722 (lc GCN2), p630 (hc GCN2), or pHQ1162 (lc SNF1) were analyzed as for panel C. The ratio of eIF2α-P normalized to eIF2α in the presence or absence of 3AT is shown at the bottom.

Reg1 negatively regulates Snf1 to stimulate Gcn2 phosphorylation on T882 and eIF2α phosphorylation. (A and B) Transformants of strains H1642 (WT), HQY343 (snf1Δ), HQY305 (reg1Δ), and HQY347 (snf1Δ reg1Δ) harboring pDH101 (FL-GCN2) were cultured in the absence or presence of 3AT, and the eIF2α-P/eIF2α (A) and T882-P/FL-Gcn2 (B) ratios were determined as for Fig. 2A and B. For analysis of eIF2α phosphorylation (A), aliquots of WCE with 10 and 20 μg of protein were loaded for each strain. For analysis of Gcn2 phosphorylation (B), 4 and 6 mg of protein were immunoprecipitated for each strain. (C) Transformants of H1642 (WT) or HQY305 (reg1Δ) bearing the empty vector or pHQ817 (REG1) were cultured and analyzed as for panel A. Ten and 20 μg of protein were loaded for each strain. (D) Transformants of HQY347 (snf1Δ reg1Δ) with pDH101 (FL-GCN2) and pMO18 (HA-SNF1), pMO19 (HA-SNF-T210A), or the vector pSK134 (V) were analyzed as for panels A and B; gcn2Δ, strain H1895. In the upper panels, 10 and 20 μg of protein were loaded for each strain. For detection of Gcn2 phosphorylation (lower panels), 4 and 6 mg of protein were immunoprecipitated for each strain. For panels A, B, and D, means and standard errors were calculated from the results of three or more independent experiments.

Functional and physical interactions of Snf1 with Gcn2 under conditions of histidine starvation. (A) Transformants of HQY344 (gcn2Δ snf1Δ) harboring pSNF1-316 (SNF1-HA) and either pAH15 (GCN2) (lanes 1 to 4) or the empty vector YEplac181 (lanes 5 to 8) or transformants containing pHQ1162 (SNF1) and pAH15 (lanes 9 to 12) were cultured as described in the legend to Fig. 2 in the presence of 3AT for 90 min, and WCEs were immunoprecipitated (IP) with anti-HA agarose and subjected to Western blot (WB) analysis using antibodies against Gcn2 or Gcd6 (top two panels). Gcd6 is a subunit of eIF2B, which presumably does not interact with Snf1, analyzed as a loading control. (Bottom of panel) WCEs were immunoprecipitated with antibodies against Gcn2 and subjected to Western analysis using anti-HA antibodies. In, P, and S correspond to the input, pellet, and supernatant, respectively. For inputs, 1% and 0.2% of the samples were loaded; for supernatants, 1% of the samples was loaded. (B) Transformants of HQY357 (gcn2Δ snf1Δ) harboring plasmids containing WT GCN2 (pB107), gcn2-K628R (pB115), or gcn2-T882A, T887A (pB11) and either WT HA-SNF1-HA (pMO18) or HA-SNF1-T210A (pMO19) (SNF1-TA) were cultured with 40 mM 3AT and analyzed as described in the legend to Fig. 2B. Four and 6 mg of protein were immunoprecipitated for each strain. Means and standard errors were calculated from three independent experiments. (C) Cells were grown in SD medium and treated with 40 mM 3AT for 90 min. WCEs were prepared, HA-Snf1 was immunoprecipitated, and one-half of the immunocomplexes were subjected to Western analysis with anti-Gcn2 and anti-HA antibodies (bottom two panels); the other half was used for in vitro kinase assays with [γ-³²P]ATP, which were resolved by SDS-PAGE and subjected to autoradiography (top two panels). The strains used were transformants of HQY357 (snf1Δ gcn2Δ) with plasmids pMO18 (HA-SNF1) and p2326 (pB34) (gcn2-K628R) (lanes 1 and 2), pMO19 (HA-SNF1-T210A) and p2326 (pB34) (lanes 3 and 4), pMO18 and p630 (GCN2) (lane 5 and 6), pMO19 and p630 (lanes 7 and 8), and pMO18 and the empty vector YEplac195 (lanes 9 and 10). One-half and 1 mg of protein were immunoprecipitated for each strain.

Evidence that Snf1 antagonizes the eIF2α phosphatase activities of Glc7 and Sit4. (A to D) Strains of the indicated genotype were cultured in galactose for 20 min as described in the legend to Fig. 6A and analyzed for eIF2α-P, T882-P, and S577-P as described in the legend to Fig. 2A and B. (A) Transformants of glc7-1 snf1Δ mutant strain CY1194 harboring plasmids containing SNF1 (pHQ1163) and GLC7 (p28.1) (WT), GLC7 and the empty vector (snf1Δ), SNF1 and the empty vector (glc7-1), or two empty vectors (glc7-1 snf1Δ). (B) The glc7-1 SNF1 and glc7-1 snf1Δ mutant strains are transformants of CY1194 with pDH101 (FL-GCN2) and either plasmid pHQ1163 (SNF1) or the empty vector (snf1Δ); GLC7 SNF1 strains are cognate transformants of CY1323 (GLC7 snf1Δ) containing the same plasmids. (C) The WT and snf1Δ strains are H1642 and HQY343, respectively, transformed with pDH101 (FL-GCN2). The sit4Δ SNF1 and sit4Δ snf1Δ mutant strains are CY1297 and CY1298 transformed with pDH101. (D) The strains employed are transformants of CY1194 harboring plasmids containing SNF1 (pHQ1163) and the empty vector (glc7-1 SNF1), two empty vectors (glc7-1 snf1Δ), GLC7 (p28.1) and pHQ1163 (GLC7 SNF1), or p28.1 and the empty vector (GLC7 snf1Δ). The sit4Δ mutant strains are transformants of CY1299 with two empty vectors (glc7-1 sit4Δ snf1Δ), with SNF1 (pHQ1163) and the empty vector (glc7-1 sit4Δ), with p28.1(GLC7) and pHQ1163 (sit4Δ), or with p28.1 and the empty vector (sit4Δ snf1Δ). For detection of eIF2α phosphorylation (A and D), 10 and 20 μg of protein were loaded for each sample. For detection of Gcn2 phosphorylation (B and C), 4 and 6 mg of protein were immunoprecipitated for each strain. Means and standard errors were calculated from three or more independent experiments. (E and F). Cells were grown in SD medium, and one-half of each culture was used for preparation of WCEs. The other half was washed with SD containing galactose in the place of glucose and cultured in the same galactose medium for 20 min and used for the preparation of WCEs. Sit4-HA (E) or Glc7-MYC (F) was immunoprecipitated from the WCEs with anti-HA or anti-myc agarose (Santa Cruz Biotechnology), respectively, resolved by SDS-PAGE, and subjected to Western analysis using antibodies against eIF2α. Strains: E, sit4Δ (CY1297) transformed with SIT4 (CB1770) or SIT4-HA (pSL2667); F, H1515 transformed with either untagged GLC7 (p29.1) or GLC7-MYC (pHQ1774). In, P, and S correspond to the input, pellet, and supernatant, respectively. For inputs, 1% and 0.2% of the samples were loaded; 1% of the supernatants was loaded.

Elimination of Snf1 reduces T882-P in the Gcn2 activation loop. (A) Lanes 2 to 9, transformants of snf1Δ gcn2Δ mutant strain HQY344 with an lc plasmid containing GCN2-FL (pDH101) or FL-GCN2-S577A (pCB149) and either SNF1 (pHQ1163) or the empty vector; lane 1, H1895. Cultures grown in SD with minimal supplements were divided, and growth continued in the absence (−) or presence of 40 mM 3AT (+) for 90 min. WCEs were subjected to Western analysis with antibodies against total eIF2α or phospho-specific antibodies against eIF2α-P. Western signals were quantified, and the eIF2α-P to eIF2α ratios were plotted. (B) Transformants of H1895 (gcn2Δ) or HQY344 (gcn2Δ snf1Δ) with plasmids containing FL-GCN2 (pDH101) or FL-GCN2-S577A (pCB149) were cultured as for panel A, and FL-Gcn2 was immunopurified with anti-Flag M2 agarose affinity gel and subjected to Western analysis with antibodies against T882-P or Ser577-P in Gcn2 or total Gcn2 (FL-Gcn2). Aliquots with 8 and 6 mg of protein were immunoprecipitated and analyzed in adjacent lanes. Western signals were quantified, and the T882-P to FL-Gcn2 ratios were plotted. Means and standard errors were calculated from three independent experiments.

Snf1 kinase activity is induced in histidine-starved cells and is required to promote eIF2α phosphorylation. (A) Cells were grown in SD with minimal supplements and treated with 40 mM 3AT for 90 min. WCEs were prepared, and eIF2α-P was determined as described in the legend to Fig. 2A. Strains are transformants of HQY357 (snf1Δ gcn2Δ) with plasmids pHQ1162 (SNF1), YCplac33 (V), pHQ1873 (snf1-K84R), and pHQ1874 (snf1-T210A). Ten and 20 μg of protein were loaded for each strain. Means and standard errors were calculated from three or more independent experiments. (B) Transformants of HQY343 (snf1Δ) with pMO19 (HA-SNF-T210A) or pMO18 (HA-SNF1) and HQY347 (snf1Δ reg1Δ) with pMO18 (HA-SNF1 reg1Δ) were cultured in SD medium and left untreated or treated with 40 mM 3AT for 1 h. WCEs were subjected to Western analysis with antibodies specific for mammalian AMPK phosphorylated on T172.

Snf1 stimulates eIF2α phosphorylation in galactose-grown cells without activating Gcn2. (A) Strains H1642 (WT) and HQY343 (snf1Δ) were grown in minimally supplemented SD medium, harvested, washed, and cultured for the indicated times in the same medium containing galactose in the place of glucose. WCEs were analyzed as described in the legend to Fig. 2A. (B and C). Effects of snf1Δ and reg1Δ on eIF2α-P and phosphorylation of T882 and S577 in FL-Gcn2 immunopurified from cells cultured in galactose. (B) Transformants described in Fig. 2B were cultured for 90 min in galactose as for panel A and analyzed as described in the legend to Fig. 2B. (C) Strains described in Fig. 3A were cultured as for panel B and analyzed as described in the legend to Fig. 3A and B. For detection of eIF2α phosphorylation, 10 and 20 μg of protein were loaded for each strain. For detection of Gcn2 phosphorylation, 4 and 6 mg of WCEs were immunoprecipitated for each strain. Means and standard errors were calculated from three or more independent experiments.

Postulated roles of Snf1 in regulating eIF2α-P levels by distinct mechanisms in cells limited for histidine or glucose. (A) In cells replete with glucose but limited for histidine, uncharged tRNA^His accumulates and activates Gcn2, increasing the phosphorylation of T882 in the Gcn2 activation loop and evoking increased eIF2α phosphorylation (bold arrows). Snf1's contribution to Gcn2 activation is indicated by thinner arrows. High-level T882-P and eIF2α-P formation depends on the phosphorylation of T210 in the Snf1 activation loop, which is negatively regulated by the Glc7/Reg1 PP1 complex. Snf1 might also promote eIF2α-P formation by inhibiting Sit4 or the Glc7 complex with eIF2α-PPase activity, whose regulatory/targeting subunit (X) is unknown. (B) In amino acid-replete, glucose-limited cells growing on galactose, uncharged tRNA is at low levels (depicted by gray shading) and Gcn2 activity/T882-P is at the basal level. Snf1 does not stimulate Gcn2 function; in fact, it seems to inhibit Gcn2 by promoting S577-P formation (not depicted here; see text for details). Hence, Snf1 promotes eIF2α-P by inhibiting the eIF2α-P phosphatase activities of Glc7/X and Sit4.