Mutations in Ire1 kinase abolish phosphate transfer but preserve RNase activity. (A) A schematic representation of Ire1 depicting the location of each functional domain. Residues D797 and K799 in the nucleotide-binding pocket of the kinase domain hydrogen bond with the terminal phosphate of ATP to catalyze phosphate transfer to the substrate serine. Mutation of D797 and K799 to noncatalytic asparagines is predicted to block phosphate transfer but allow for ATP binding. (B) The kinase activity of recombinant Ire1KR32 (WT, 474 amino acids; lanes 1–3) and Ire1KR32(D797N,K799N) (lanes 4–6) were measured in an in vitro kinase assay. Recombinant Ire1 was mixed with 0.033 µM [γ³²]P-ATP and incubated at 30°C for the time indicated. Reactions were stopped in 1% SDS loading buffer and separated by SDS-PAGE. A truncated version of WT Ire1, Ire1KR (442 amino acids; lanes 7–9), was mixed with Ire1KR32(D797N,K799N) (lanes 10–12). (C and D) In vitro RNA cleavage assays were performed using purified substrate RNA, HP21 (C) or Xbp1 (D), and either WT Ire1KR32 or Ire1KR32(D797N,K799N). Reactions were performed in the presence and absence of 2 mM ADP. The lower sensitivity of the Xbp1 mRNA cleavage reaction to the ADP cofactor during cleavage suggests that a longer RNA substrate may independently stabilize the Ire1 oligomer, perhaps by bridging between multiple adjacent monomers. Bar values were obtained from single-exponential fitting of time courses. Error bars show standard errors of the single-exponential fitting. The time courses were repeated multiple times and k_obs values reproduced within twofold.

Ire1 kinase activity, uncoupled from HAC1 mRNA splicing, is important for cell survival during the UPR. (A) Cells bearing WT IRE1 (lanes 1 and 2), a deletion of ire1 (lanes 3 and 4), or ire1(D797N,K799N) (lanes 5 and 6) were left uninduced (−) or induced with 2 mM DTT (+). HAC1 mRNA splicing was analyzed by Northern blotting. The positions of the unspliced (u; 1449 nucleotides) and spliced (s; 1197 nucleotides) forms of HAC1 mRNA are indicated with arrows. Splicing intermediate i1 (980 nucleotides) corresponds to the 5′ exon–intron hybrid species, whereas i2 (728 nucleotides) corresponds to the 5′ exon alone. (B) Cells carrying WT IRE1, a deletion of ire1, or ire1(D797N,K799N) were grown in culture, diluted to equal cell number, serially diluted 1:5, and plated onto permissive medium (−Tm) or medium containing 0.25 µg/ml tunicamycin (+Tm). (C) WT IRE1 or ire1(D797N,K799N) cells were grown in culture to OD₆₀₀ 0.2, the UPR was induced by the addition of 2 mM DTT. The value for percent viable cells was determined by measuring the number of colony-forming units over time (see Materials and methods). DTT was refreshed and cells were kept at an OD at or below 0.2 throughout the duration of the experiment. (D) WT or mutant ire1 cells carrying HA-tagged Hac1 were left uninduced or induced with 2 mM DTT, and total protein was isolated. Samples were separated by SDS-PAGE and subjected to Western blotting using an anti-HA antibody or an anti-PGK1 antibody. (E) Hac1 protein was quantified, normalized to PGK1 levels, and plotted. (F) Total protein was isolated from WT- or Ire1(D797N,K799N)- expressing cells, separated by SDS-PAGE, and subjected to Western blotting using an anti-FLAG antibody. Ire1 protein levels are equivalent in WT and ire1(D797N,K799N) cells.

Downstream events in UPR activation are normal in ire1(D797N,K799N) cells. (A) Microarray analysis was performed to assess the total mRNA expression profiles of WT IRE1 or ire1(D797N,K799N) cells over time after induction with 2 mM DTT. Cells were sampled at 0, 30, 60, and 120 min. Canonical target genes, KAR2, ERO1, DER1, PDI1, and LHS1 were up-regulated and YDJ1 was down-regulated equally upon UPR induction in both strains. (B) Total protein was isolated from cells bearing WT IRE1 or ire1(D797N,K799N) after 0, 30, or 60 min in 2 mM DTT and analyzed by Western blot for Kar2 protein. Characteristic increase in Kar2 protein upon UPR induction was observed in both strains. (C and D) ER expansion was measured in WT and ire1(D797N,K799N) cells. The UPR was induced in WT or mutant ire1(D797N,K799N) cells bearing a GFP-tagged version of Sec63 as an ER marker. Images were taken before and after 2 h UPR induction, and cortical ER expansion was quantified as described in Schuck et al. (2009) and expressed as the index of expansion (IE). Error bars indicate SEM. Bar (C), 2 µm.

Activation of Ire1(D797N,K799N) continues after WT activity has plateaued. (A) A schematic of the fluorescent splicing reporter (SR) in which the HAC1 ORF was replaced with GFP such that Ire1-mediated splicing of this reporter produces fluorescent GFP. (B and C) A dilution series of DTT, from 0 to 3.3 mM, was added to cells in culture. WT (B) or ire1(D797N,K799N) (C) cells were grown at 30°C and sampled at 30-min intervals over a 4-h time course. GFP signal was measured by flow cytometry, normalized to baseline and plotted over time. (D) SR fluorescence was plotted as a function of increasing [DTT] in WT and ire1(D797N,K799N) cells at 60 min after DTT addition. (E) The dose–response of SR fluorescence plotted as a function of [DTT] at 240 min after induction reveals that Ire1(D797N,K799N) was significantly more active than WT at all concentrations of DTT. The dotted line indicates the concentration of DTT at which the two curves most deviated. (F) WT or ire1(D797N,K799N) cells bearing GFP-tagged Ire1 were treated with 1 mM DTT and formation of Ire1 foci was imaged by confocal microscopy over time. (G) Foci formation measured in F was quantified as percentage of Ire1 in foci and plotted over time. The maximum value, 100%, is reached when all pixels containing Ire1-GFP signal are in foci (see supplementary methods in Aragón et al., 2009 for a detailed description of quantitation).

The oxidation potential of the ER is restored in ire1(D797N,K799N) cells. (A–D) Re-oxidation of the ero-GFP reporter occurs in the absence of kinase activity. (A) WT, (B) ire1(D797N,K799N) and (C) ire1Δ cells bearing the ER-targeted redox reporter, ero-GFP, were treated with 0, 1, or 2 mM DTT. The ratio of reduced-to-oxidized signal (r/o ratio) was measured by flow cytometry and plotted over time. (D) Percent reoxidation of ero-GFP was calculated for WT, ire1(D797N,K799N), and ire1Δ cells in 0, 1, and 2 mM DTT (see Materials and methods). (E) Cells expressing ero-GFP were analyzed by spinning disk confocal microscopy before and after treatment with 2 mM DTT for 45 min.

Shut-off of Ire1(D797N,K799N) is delayed after removal of ER stress. (A) WT or ire1(D797N, K799N) cells were treated with 5 mM DTT for 60 min before DTT washout. Cell samples were taken after DTT washout and total RNA was analyzed by Northern blot for HAC1 mRNA. Unspliced (u) and spliced (s) forms of HAC1 mRNA are indicated by arrows. (B and C) GFP-tagged variants of Ire1 were visualized by fluorescence microscopy after washout of 5 mM DTT. Quantitation of Ire1 foci is displayed in the bottom panel. (D) Ire1ΔHPL cells retain RNase activity. A dilutions series of DTT, from 0 to 3.3 mM, was added to ire1ΔHPL cells bearing the splicing reporter, SR. Cells were grown at 30°C and sampled at 30-min intervals over a 4-h time course. GFP signal was measured by flow cytometry and plotted over time.