JBC PeproTech; Our Business is Cytokines!

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M412882200 on January 31, 2005

J. Biol. Chem., Vol. 280, Issue 17, 16925-16933, April 29, 2005
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
280/17/16925    most recent
M412882200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by McEwen, E.
Right arrow Articles by Kaufman, R. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by McEwen, E.
Right arrow Articles by Kaufman, R. J.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Heme-regulated Inhibitor Kinase-mediated Phosphorylation of Eukaryotic Translation Initiation Factor 2 Inhibits Translation, Induces Stress Granule Formation, and Mediates Survival upon Arsenite Exposure*

Edward McEwen{ddagger}, Nancy Kedersha§, Benbo Song{ddagger}, Donalyn Scheuner{ddagger}, Natalie Gilks§, Anping Han¶, Jane-Jane Chen¶, Paul Anderson§, and Randal J. Kaufman{ddagger}||**

From the {ddagger}Howard Hughes Medical Institute and the ||Department of Biological Chemistry, University of Michigan Medical Center, Ann Arbor, Michigan 48109, the §Division of Rheumatology, Immunology, and Allergy, Brigham and Women's Hospital, Boston, Massachusetts 02115, and the Harvard-MIT Division of Health Sciences and Technology, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139

Received for publication, November 15, 2004 , and in revised form, January 28, 2005.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Exposure to arsenite inhibits protein synthesis and activates multiple stress signaling pathways. Although arsenite has diverse effects on cell metabolism, we demonstrated that phosphorylation of eukaryotic translation initiation factor 2 at Ser-51 on the {alpha} subunit was necessary to inhibit protein synthesis initiation in arsenite-treated cells and was essential for stress granule formation. Of the four protein kinases known to phosphorylate eukaryotic translation initiation factor 2{alpha}, only the heme-regulated inhibitor kinase (HRI) was required for the translational inhibition in response to arsenite treatment in mouse embryonic fibroblasts. In addition, HRI expression was required for stress granule formation and cellular survival after arsenite treatment. In vivo studies elucidated a fundamental requirement for HRI in murine survival upon acute arsenite exposure. The results demonstrated an essential role for HRI in mediating arsenite stress-induced phosphorylation of eukaryotic translation initiation factor 2{alpha}, inhibition of protein synthesis, stress granule formation, and survival.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
The aqueous form of the trivalent arsenical is a potent environmental toxin that significantly contributes to human pathogenesis. Exposure to the arsenite anion (AsO–3) results in skin disease, cancer, peripheral neuropathy, and cardiovascular disease (1). Arsenite exposure is also causative of diabetes (2). Unfortunately chronic exposure to arsenite through contaminated water occurs frequently and contributes to carcinogenicity as well as neurological and renal pathology (3). Based upon incidence, toxicity, and potential for human exposure, this toxin is ranked as the number one hazardous substance (4). Higher doses of arsenite (>5 µM) damage different tissues by induction of apoptosis and necrosis. The dose-dependent and tissue-dependent apoptosis induced by arsenite has led to the use of this compound for the treatment of certain types of cancer including acute promyelocytic leukemia and multiple myeloma (5, 6). As an issue of major public health concern and its potential application to treat cancer, it is essential to obtain an understanding of the mechanisms mediating arsenite toxicity as well as those affording protection against arsenite exposure.

The physiologically important biological target(s) of arsenite exposure is unknown. Arsenite-induced stress is thought to be the chemical paradigm of heat stress. It is believed that acute arsenite toxicity is due to oxidation of cysteine residues in target proteins that directly alters their conformation and/or activity (7). Therefore, many cellular responses to arsenite exposure may be a consequence of protein misfolding due to arsenite modification. Compounding the production of unfolded protein, the ubiquitin-proteosome pathway responsible for degrading misfolded and damaged proteins is impaired upon arsenite exposure (8). Alternatively arsenite may interact with cysteine-rich, redox-sensitive molecules, such as thioredoxin and glutathione, to generate oxidative stress and interfere with specific biological processes. Exposure to arsenite influences numerous signal transduction pathways that include activation of the stress-activated, mitogen-activated protein kinases, extracellular signal-regulated kinase, c-Jun amino-terminal kinase, and p38 to induce activator protein-1-dependent transcription (9–11). In addition, arsenite inactivates the proinflammatory NF{kappa}B signaling pathway through inhibition of I{kappa}B kinase {beta} (12), inhibits the signal transducers and activators of transcription family of transcription factors (13), and activates heat shock transcription factor (14). As a consequence, arsenite suppresses expression of genes encoding antiapoptotic functions and induces transcription of genes encoding proapoptotic functions, cytosolic protein chaperones, and other stress response genes, such as CHOP/GADD153 and metallothionein (15–18). One of the most immediate cellular responses to arsenite exposure is inhibition of protein synthesis that correlates with phosphorylation of the eukaryotic translation initiation factor 2 at Ser-51 on the {alpha} subunit (eIF2{alpha})1 (19–21). Phosphorylation of eIF2{alpha} is a fundamental regulatory mechanism that controls global rates of protein synthesis in all eukaryotic cells (20). Eukaryotic translation initiation factor 2 is a heteromeric GTP-binding protein that binds initiator tRNAMeti to form a ternary complex required for 40 S ribosomal subunit recognition of the AUG initiation codon within mRNAs. Prior to 60 S ribosomal subunit joining, GTP is hydrolyzed to GDP. To catalyze another round of initiation, it is necessary to exchange GDP for GTP. As this spontaneous exchange rate is slow, this reaction requires catalysis by the guanine nucleotide exchange factor eIF2B.

Since the cellular amount of eIF2B is limiting over the amount of eIF2 (22), continued protein translation initiation requires catalytic recycling of eIF2B. The recycling of eIF2B is regulated by phosphorylation of eIF2{alpha} at residue Ser-51. Phosphorylation at this site increases the affinity of eIF2·GDP for eIF2B by 150-fold (23). As a consequence, eIF2B is sequestered in a nonproductive complex with eIF2·GDP and prevents further initiation events. Therefore, small increases in the level of phosphorylated eIF2{alpha} profoundly inhibit the cellular protein synthetic rate. Such a drastic reduction in the rate of translation upon arsenite exposure may be an adaptive mechanism to conserve cellular energy and resources and to prevent the further accumulation of unfolded protein.

It was previously demonstrated that arsenite treatment induces phosphorylation of eIF2{alpha} (21). There are four protein kinases known to phosphorylate eIF2{alpha} at Ser-51 in mammalian cells: the double-stranded RNA activated protein kinase PKR (24), the mammalian homologue of yeast Gcn2p (25), the endoplasmic reticulum (ER)-localized eIF2{alpha} kinase PERK/PEK (26, 27), and the heme-regulated inhibitor kinase HRI (28). Although these protein kinases are homologous in their kinase catalytic domains, they each have different regulatory domains that respond to different stress stimuli. PKR gene expression is induced by class 1 interferons, and the protein is translated in a latent form that requires activation by binding to highly structured or double-stranded RNA. GCN2 is activated by amino acid starvation and mediates translational and transcriptional adaptation to amino acid limitation. PERK is activated by the accumulation of unfolded proteins in the ER. HRI is expressed most abundantly in erythroid cells. HRI is the most important eIF2{alpha} kinase in erythroid cells where it is activated by deficiency in heme, the prosthetic group of hemoglobin, through two heme-binding domains. In this manner, HRI coordinates globin mRNA translation with available iron (29). HRI is required to prevent accumulation of misfolded globin chains in the absence of heme. Previous studies have demonstrated that arsenite mediates phosphorylation of eIF2{alpha} through HRI in erythroid cells (30). However, it is not known which, if any, of these eIF2{alpha} kinases mediates the translational inhibition by arsenite in non-erythroid cells.

Environmental stress initiates the appearance of phase-dense particles termed stress granules (SGs) that were first identified upon heat treatment of tomato cells in culture (31). SG assembly occurs when translation initiation is aborted. When translation is initiated in the absence of functional eIF2·GTP·tRNAMeti, the 40 S ribosome stalls on the mRNA to form a 48 S preinitiation complex. These complexes of 40 S ribosomes, initiation factors, and their associated mRNA transcripts accumulate to form SGs. SG formation requires T-cell internal antigen 1 (TIA-1) and TIA-1-related proteins (32, 33). The mRNAs in SGs are in dynamic equilibrium with the polysomes (34). Stress granule formation in response to ER stress requires the ER stress kinase PERK (35). Although depletion of the eIF2·GTP·tRNAMeti ternary complex correlates with SG formation, it is not known whether eIF2 inactivation through eIF2{alpha} phosphorylation is required for SG formation (36).

In this study we demonstrated that the inhibitory effect of arsenite on protein synthesis was mediated through phosphorylation at Ser-51 on eIF2{alpha}. Phosphorylation of eIF2{alpha} was necessary for arsenite-mediated inhibition of protein synthesis and SG formation. We demonstrated that HRI was expressed and was required for eIF2{alpha} phosphorylation to inhibit protein synthesis in response to arsenite-induced stress in murine embryonic fibroblasts (MEFs). The significance of arsenite-induced protein synthesis inhibition was corroborated in vivo as HRI was required for survival when cells or mice were exposed acutely to arsenite.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Generation of Kinase-null Mice and MEFs—Homozygous HRI–/– mutant mice (37) and mice with the S51A allele of eIF2{alpha} (38) were generated as described previously. PKR–/– mutant mice were kindly provided by Bryan Williams (39). MEFs were generated using standard isolation procedures (40). PERK–/– mutant, GCN2–/– mutant, and control MEFs were kindly provided by David Ron (27, 53). MEFs were maintained in Dulbecco's modified essential medium with high glucose (Invitrogen) with the addition of 10% fetal bovine serum (Hyclone, Logan, UT), glutamine, penicillin/streptomycin, and essential and nonessential amino acids (Invitrogen).

HRI mRNA Expression Analysis—RNA was extracted from wild-type and HRI–/– mutant MEFs using the RNAqueous kit (Ambion, Austin, TX) according to the manufacturer's protocol. The reverse transcriptase reaction mixture was assembled as follows: 2 pmol of oligo(dT)15 (Promega, Madison, WI), 4 µl of 5x avian myeloblastosis virus reverse transcriptase buffer, 10 units of avian myeloblastosis virus reverse transcriptase (Promega), 40 units of rRNAsin recombinant RNase inhibitor (Promega), 2 µg of RNA, 1.25 mM dNTP mixture (Qbiogene, Montreal, Quebec, Canada), and purified water to a final reaction volume of 20 µl. The reaction was heated to 50 °C for 1 h and then to 75 °C for 10 min. The resulting cDNA was purified using a PCR purification column (Qiagen, Valencia, CA) and eluted in 30 µl of water. cDNA eluate (5 µl) or HRI cDNA (5 ng) was used as template for the PCR. The PCR was performed using the Advantage cDNA PCR kit (Clontech) according to the manufacturer's protocol plus 5% dimethyl sulfoxide. The specific primers for HRI were 5'-ATGCTGGGGGGCAGCTCC-'3 and 5'-TCATCTCTTCATCCCTCTG-3. The reaction was cycled at 95 °C for 1 min, then 50 °C for 1 min, and then 72 °C for 2 min for 32 cycles followed by a final extension of 72 °C for 2 min. The expected product size for the HRI+/+ MEFs was 1549 base pairs, and for the HRI–/– MEFs it was 1145 base pairs. The PCR products were electrophoresed on a 0.9% agarose gel for analysis. For Northern blot analysis, total RNA was extracted using the RNAqueous kit (Ambion). RNA was electrophoresed on a formaldehyde gel and then transferred onto a nylon membrane for hybridization to a digoxigenin-labeled DNA probe specific to HRI. Detection was performed using anti-digoxigenin alkaline phosphatase-complexed antibody.

Protein Synthesis Analysis—MEF cultures were grown overnight after passage and incubated for 30 min at 37 °C in media containing 0–200 µM arsenite or 1 µM thapsigargin. Cells were rinsed with phosphate-buffered saline and then radiolabeled with 200 µCi/ml [35S]methionine and [35S]cysteine (specific activity =>1000 Ci/mmol, Amersham Biosciences) in methionine- and cysteine-free medium (Invitrogen) for 10 min. After washing with ice-cold phosphate-buffered saline, cell extracts were prepared. Radioactivity incorporated was measured by precipitation with 10% trichloroacetic acid and scintillation counting. Some cell extracts were analyzed by SDS-PAGE and prepared for autoradiography using EnHance (PerkinElmer Life Sciences).

Western Blot Analysis—MEFs were cultured and treated as indicated. Cell extracts were prepared with lysis buffer (50 mM Tris, pH = 7.4, 150 mM NaCl, 10 mM {beta}-glycerol phosphate, 50 mM NaF, 1 mM orthovanadate, 0.1 mM EDTA, 10% glycerol, 1% Triton X-100, and Complete miniprotease inhibitors (Roche Applied Science)). Proteins were analyzed by SDS-PAGE on reducing gels, and the phosphorylated and total forms of eIF2{alpha} were detected by Western blot analysis as described previously (41). Antibody for phosphorylated eIF2{alpha} was obtained from BIOSOURCE, and mouse monoclonal antibody was used for detection of total eIF2{alpha} (41). Anti-phospho-p38 antibody was obtained from Cell Signaling Technology, Inc. (Beverly, MA). Densitometry scanning was performed and quantified with Image J software obtained from the National Institutes of Health website.

Polysome Profile Analysis—Cells in log phase growth were treated with cycloheximide (20 µg/ml for 10 min) prior to being harvested by scraping and processed for polysome analysis essentially as described previously (42). The fractionated polysomes were eluted using an ISCO fractionator with continual monitoring at A254.

Immunofluorescence of Stress Granules—Cells were treated with different stress treatments including graded concentrations of sodium arsenite for 45 min, heat shock (44 °C for 1 h), puromycin (20 µg/ml for 2 h), and glucose starvation. After treatment, the cells were fixed and stained for goat polyclonal antibodies against TIA-1 (sc-1751, Santa Cruz Biotechnology, Inc.), eIF3 (sc-16378, Santa Cruz Biotechnology, Inc.), and other SG markers as described previously (33). MEFs were transfected using Qiagen reagent as described previously (32).

Cell Viability Assay—MEF cells were cultured in 96-well plates overnight before exposure to medium containing sodium arsenite. Cell viability was measured with a ProCheck cell viability assay kit (Serologicals Corp., Norcross, GA). The assay is based on the conversion of an oxidized tetrazole to a reduced form by mitochondria in metabolically active cells. The absorbance at 480 nm is proportional to the number of viable cells.

In Vivo Response to Arsenite Injection—The survival of wild-type, HRI–/–, PKR–/–, and HRI–/–PKR–/– mutant mice upon acute arsenite exposure was assessed 24 h after subcutaneous injection of freshly prepared, sterile sodium arsenite (0–20 mg/kg of body weight as indicated) or phenylhydrazine (45 mg/kg of body weight). The University of Michigan Committee for the Care and Use of Animals (UCUCA) approved all animal procedures. Blood samples were collected in EDTA by tail bleed before and at 1 h and 4.5–5 h after injection. Reticulocyte counts were analyzed by Coulter counting based on size. Liver samples were taken at 4 h after injection, and extracts were prepared by homogenization in Nonidet P-40 lysis buffer (1% Nonidet P-40, 50 mM Tris-HCl, pH 8.0, 150 mM NaCl) containing protease inhibitors (Roche Applied Science), 150 mM NaF, 0.5 mM orthovanadate, 50 mM {beta}-glycerophosphate, and 100 µg/ml phenylmethylsulfonyl fluoride.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
Arsenite-mediated Inhibition of Protein Synthesis Requires Phosphorylation of eIF2{alpha}—To evaluate the role of eIF2{alpha} phosphorylation in the cellular stress response to arsenite, we studied MEFs that harbor a S51A knock-in mutation into the eIF2{alpha} gene (38). Cells were treated with increasing concentrations of sodium arsenite and then pulse-labeled with [35S]methionine and [35S]cysteine to measure the rates of protein synthesis. Whereas increasing concentrations of sodium arsenite inhibited protein synthesis greater than 90% in wild-type MEFs, the inhibition was significantly less in the homozygous eIF2{alpha} A/A mutant MEFs (Fig. 1, A and B). Although the spectrum of abundant polypeptides synthesized was similar under control and arsenite treatment conditions, the homozygous eIF2{alpha} A/A mutant cells were more resistant to the inhibition mediated by arsenite (Fig. 1A). In three independent experiments, 50 µM sodium arsenite reproducibly inhibited protein synthesis in wild-type MEFs with little change in the homozygous eIF2{alpha} A/A mutant MEFs (Fig. 1C). Immunoblot analysis using antibodies that specifically recognize phosphorylated eIF2{alpha} and total eIF2{alpha} demonstrated a 3-fold increase in eIF2{alpha} phosphorylation in wild-type (S/S) MEFs, whereas phosphorylation was not detected in the homozygous eIF2 A/A mutant MEFs (Fig. 1D).



View larger version (46K):
[in this window]
[in a new window]
  		FIG. 1.
Arsenite inhibits protein synthesis through eIF2{alpha} phosphorylation. Wild-type eIF2{alpha} (S/S) and homozygous S51A mutant eIF2{alpha} (A/A) MEFs were incubated in 0–200 µM arsenite (As) for 30 min and then pulse-labeled with [35S]methionine and [35S]cysteine. A, SDS-PAGE analysis of metabolically labeled proteins in cell extracts. B, quantitation of protein synthesis after treatment with sodium arsenite. Analysis was performed in triplicate as described under "Experimental Procedures." S.E. bars were small and are not apparent on the graph. C, protein synthesis after sodium arsenite treatment. MEFs were treated with 100 µM sodium arsenite for 30 min and then pulse-labeled with [35S]methionine and [35S]cysteine. Cells were harvested, and incorporation was measured as described under "Experimental Procedures." Results are presented as mean ± S.E. from three independent experiments. D, phosphorylation of eIF2{alpha} and p38 upon sodium arsenite treatment. Western blot analysis was performed after exposure of MEFs to 100 µM arsenite for 30 min. eIF2{alpha} (eIF2{alpha}-P) and p38 (p38-P) phosphorylation was quantified and normalized to total eIF2{alpha}. The ratio of the normalized phosphorylation signal of arsenite-treated versus untreated is indicated.

 
Treatment of cells with sodium arsenite activates p38 mitogen-activated protein kinase (7). To determine whether arsenite actually induces a stress response in the homozygous eIF2{alpha} A/A mutant MEFs, we studied the activation of the p38 mitogen-activated protein kinase cascade by Western blot analysis using an antibody that is specific to phosphorylated p38. The results demonstrated that sodium arsenite activated p38 to a similar level in the wild-type (S/S) and homozygous eIF2{alpha} A/A mutant cell lines by comparison to the total eIF2{alpha} loading control (Fig. 1D). Therefore, although arsenite induced a stress response, protein synthesis was not inhibited in the eIF2{alpha} A/A mutant cells. In addition, p38 activation did not require eIF2{alpha} phosphorylation or translational attenuation.

Arsenite-mediated eIF2{alpha} Phosphorylation and Inhibition of Protein Synthesis Requires HRI—To evaluate the requirement for the four different eIF2{alpha} kinases in arsenite-mediated eIF2{alpha} phosphorylation and translational inhibition, we studied arsenite responses in MEFs isolated from mice that harbor gene deletions in the different eIF2{alpha} kinases. An initial experiment compared eIF2{alpha} phosphorylation in wild-type HRI+/+, HRI+/–, HRI–/–, and PKR–/– MEFs. While the eIF2{alpha} phosphorylation in wild type, heterozygous HRI+/–, and PKR–/– MEFs increased 3–4-fold, there was no comparable increase in the HRI–/– MEFs (Fig. 2A). Analysis of the same Western blot using an anti-phospho-p38 antibody indicated similar amounts of activated p38 upon arsenite treatment for all MEFs. Analysis of protein synthesis by [35S]methionine and [35S]cysteine pulse labeling and SDS-PAGE indicated substantial translational inhibition in all cell lines tested except the HRI–/– MEFs (Fig. 2B and data not shown).



View larger version (73K):
[in this window]
[in a new window]
  		FIG. 2.
Arsenite-mediated eIF2{alpha} phosphorylation and inhibition of protein synthesis requires HRI. Wild-type or the indicated eIF2{alpha} kinase-null MEFs were incubated in 100 µM arsenite (As) or 1 µM thapsigargin for 30 min and then pulse-labeled with [35S]methionine and [35S]cysteine. A, Western blot analysis of MEFs treated with 0 and 100 µM arsenite using anti-phospho-eIF2{alpha}, anti-phospho-p38, and anti-total eIF2{alpha} antibody. eIF2{alpha} (eIF2{alpha}-P) and p38 (p38-P) phosphorylation was quantified and normalized to total eIF2{alpha}. The ratio of the normalized phosphorylation signal of arsenite-treated versus untreated is indicated. B, MEFs were treated with increasing concentrations of arsenite (0, 10, 50, and 200 mM) for 30 min, then pulse-labeled with [35S]methionine and [35S]cysteine, and then harvested as described under "Experimental Procedures." Equal amounts of protein extract were analyzed by SDS-PAGE and autoradiography. Quantitation of protein synthesis after treatment with sodium arsenite was performed in triplicate as described under "Experimental Procedures." Data are mean ± S.E.

 
We then analyzed MEFs harboring deletions in each of the eIF2{alpha} kinases: HRI, PKR, GCN2, and PERK. Homozygous deletion of PKR, GCN2, or PERK had no effect on the arsenite-induced phosphorylation of eIF2{alpha} and inhibition of protein synthesis (Fig. 3, A and B). In contrast, deletion of HRI alone prevented both arsenite-mediated eIF2{alpha} phosphorylation and translational inhibition. Although there was a background level of eIF2{alpha} phosphorylation in the HRI–/– mutant MEFs, this phosphorylation was not increased upon sodium arsenite treatment (Fig. 3A, lanes 7 and 8). Treatment of the MEFs with thapsigargin, an agent that induces ER stress, significantly inhibited protein synthesis in all cell lines except the PERK–/– mutant MEFs (Fig. 3B), consistent with the role of PERK in translational attenuation on ER stress (26, 27).



View larger version (29K):
[in this window]
[in a new window]
  		FIG. 3.
Arsenite-mediated eIF2{alpha} phosphorylation and inhibition of protein synthesis requires HRI. Wild-type or the indicated eIF2{alpha} kinase-null MEFs were incubated in 100 µM arsenite (As) for 30 min and then pulse-labeled with [35S]methionine and [35S]cysteine. A, MEFs were treated with 0 and 100 µM for 30 min and then analyzed by Western blot using anti-phospho-eIF2{alpha} and anti-total eIF2{alpha} antibody. eIF2{alpha} phosphorylation (eIF2{alpha}-P) was quantified and normalized to total eIF2{alpha}, and the ratio of arsenite-treated versus untreated is indicated for each MEF. B, MEFs were treated with 100 µM sodium arsenite for 30 and then pulse-labeled with [35S]methionine and [35S]cysteine. Cells were harvested, and incorporation was measured as described under "Experimental Procedures." Results are presented from three determinations. Tg, thapsigargin.

 
Phosphorylation of eIF2{alpha} inhibits translation at the level of initiation and causes dissociation of polyribosomes and accumulation of monoribosomes and ribosomal subunits. Therefore, we analyzed polysome profiles for control and arsenite-treated MEFs (Fig. 4). These results were quantified by measuring the ratio of polysomes/80 S species observed for treated cells relative to untreated cells (values presented in Fig. 4). Comparison of these ratios between wild-type and mutant MEFs shows the area of polysomes/80 S upon arsenite treatment was ~2-fold greater in the HRI–/– and eIF2{alpha} A/A mutant MEFs compared with their wild-type controls. In addition, this ratio between the PKR–/– MEFs and wild-type MEFs did not significantly differ (0.9) (Fig. 4). Thus, less polysome dissociation occurred upon arsenite treatment of homozygous eIF2{alpha} A/A mutant MEFs and HRI–/– mutant MEFs (Fig. 4). These differences in polysome profiles are consistent with the effects of arsenite on protein synthesis in these cells.



View larger version (24K):
[in this window]
[in a new window]
  		FIG. 4.
HRI kinase and phosphorylation of eIF2{alpha} inhibit translation initiation. Polysome profiles were obtained from cell extracts prepared after incubation of wild-type S/S MEFs, homozygous mutant eIF2{alpha} A/A MEFs, PKR–/– mutant MEFs, PKR+/+ (C57Bl/6J MEFs for control of PKR–/– genetic background), HRI+/+ wild-type MEFs, and HRI–/– mutant MEFs in 0 (–As) or 250 µM (+As) arsenite for 30 min. The values presented are the ratio (P/M) of polysomes(P)/80 S (monosomes (M)) species for arsenite-treated relative to the untreated cells. Samples treated with arsenite are shown in bold tracing. Polys, polysomes.

 
Our data support the conclusion that HRI is the dedicated eIF2{alpha} kinase that mediates translational control in response to sodium arsenite in MEFs. Since previous studies questioned the significance of HRI detected in non-erythroid cells (43, 44), we characterized HRI expression. Unfortunately immunoblot analysis with several anti-HRI antibodies did not detect HRI in extracts of wild-type MEFs. Therefore, we performed reverse transcription-PCR as a more sensitive means to detect whether HRI mRNA is expressed in MEFs. This analysis demonstrated the presence of a 1.5-kb product in wild-type MEFs that was absent in the HRI–/– mutant MEFs (Fig. 5A). HRI–/– mutant MEFs displayed a 1.1-kb product, consistent with deletion of 404 bases in the mRNA from the HRI-targeted allele (37). In addition, Northern blot analysis detected a minor RNA species of the expected molecular size in both the wild-type and HRI–/– cells (Fig. 5B). Therefore, we conclude that HRI is expressed at a low but physiologically significant level to mediate arsenite-induced translational attenuation in MEFs.



View larger version (45K):
[in this window]
[in a new window]
  		FIG. 5.
HRI kinase mRNA is expressed in wild-type MEFs. A, RNA was extracted from wild-type and HRI–/– mutant MEF cultures, and reverse transcription (RT)-PCR was performed. The amplification primers produce a 1549-bp product from the wild-type HRI mRNA, amplifying from bp 311 of the coding sequence to the stop codon of the mRNA. In the mRNA of the HRI–/– MEFs, the primers amplify a truncated 1145-bp product. B, Northern blot analysis was performed as described under "Experimental Procedures."

 
Phosphorylation of eIF2{alpha} Is Necessary for Stress Granule Formation—The phosphorylation of eIF2{alpha} correlates with the assembly of SGs (32), discrete cytoplasmic foci at which un-translated mRNAs and stalled preinitiation complexes accumulate in cells subjected to environmental stress (33, 35). To form SGs, it is necessary to inhibit protein synthesis by greater than 90% (34). Transient overexpression of a non-phosphorylatable mutant of eIF2{alpha} (S51A) inhibited the assembly of SGs in cells subjected to arsenite-induced oxidative stress (32), suggesting that phosphorylated eIF2{alpha} is required for SG assembly. Consistent with this hypothesis, we found that wild-type S/S MEFs assembled SGs in response to arsenite, whereas homozygous eIF2{alpha} mutant A/A MEFs did not (Fig. 6, A and C). Similar results were observed in cells subjected to heat shock or carbonyl cyanide p-trifluoromethoxyphenylhydrazone-induced energy deprivation (data not shown) indicating that phosphorylation of eIF2{alpha} is absolutely required for SG assembly. In addition, the HRI–/– mutant MEFs exhibited a severely impaired ability to form SGs in response to sodium arsenite treatment (Fig. 6, B and C), although their ability to form SGs (and to phosphorylate eIF2{alpha}) in response to heat shock was indistinguishable from that of wild-type MEFs (data not shown). We conclude that arsenite treatment activates HRI, which in turn phosphorylates eIF2{alpha} and thus drives SG assembly. In contrast, heat shock likely activates other eIF2{alpha} kinases in addition to HRI.



View larger version (47K):
[in this window]
[in a new window]
  		FIG. 6.
HRI-mediated phosphorylation of eIF2{alpha} is required for stress granule formation. Wild-type eIF2{alpha} S/S, homozygous eIF2{alpha} mutant A/A, HRI wild-type, and HRI–/– mutant MEFs were treated with increasing concentrations of sodium arsenite (As) for 45 min, then processed for immunofluorescence, and visually scored for SGs using TIA-1 and eIF3 as SG markers. More than 100 cells were scored per treatment. Upper panels, graphic data; lower panels, representative images of cells treated with 1000 µM arsenite and stained for TIA-1 or eIF3 to reveal the presence or absence of SGs. No granules were detected in the mutant eIF2{alpha} A/A MEFs at any concentration of arsenite, thus there are no visible data on the bar graph (A). A, wild-type eIF2{alpha} S/S and homozygous eIF2{alpha} A/A MEFs. B, wild-type HRI+/+ and homozygous HRI–/– MEFs. C, morphology of wild-type eIF2{alpha} S/S, homozygous eIF2{alpha} A/A, wild-type HRI+/+, and HRI–/– mutant MEFs treated with 1000 µM arsenite and stained for either TIA-1 or eIF3 as indicated. D, homozygous mutant eIF2{alpha} A/A MEFs were co-transfected with a {beta}-galactosidase ({beta}gal) expression vector in the presence of S51D mutant eIF2{alpha} expression vector. At 48 h post-transfection, cells were analyzed for SGs as described under "Experimental Procedures." The arrowheads indicate a transfected cell expressing {beta}-galactosidase.

 
To further evaluate the requirement for eIF2{alpha} phosphorylation to promote stress granule formation in the S51A eIF2{alpha} mutant MEFs, we transfected these MEFs with expression vectors encoding a S51D mutant eIF2{alpha} to mimic a phosphorylated serine that was previously shown to act as a dominant inhibitor of translation (45). Cells were co-transfected with a {beta}-galactosidase expression vector to identify the transfected cells. Analysis of either eIF3 or TIA-1 stress granule markers demonstrated that the homozygous A/A mutant MEFs formed stress granules in the presence of the S51D mutant eIF2{alpha} without any exogenous stress (Fig. 6D). In contrast, transfection with S51A mutant eIF2{alpha} did not induce the formation of stress granules in these cells (data not shown). Therefore, the homozygous A/A mutant MEFs are able to form stress granules in the presence of a phosphomimetic form of eIF2{alpha}. We conclude that phosphorylated eIF2{alpha} is both necessary and sufficient for stress granule formation.

The HRI Response to Arsenite Promotes Cellular and Whole Animal Survival—To determine whether HRI serves a protective role in response to arsenite, we measured cell survival after sodium arsenite treatment. After 2 h of acute exposure to 100 µM arsenite, the survival of HRI–/– mutant MEF cells was significantly lower than that of wild-type or heterozygous HRI+/– mutant MEFs (Fig. 7A). However, all cell lines were sensitive to arsenite-induced cell death upon longer periods of treatment. Similar analysis of the homozygous eIF2{alpha} A/A MEFs did not yield consistent results possibly due to their slow growth rate and/or the inability to grow at low density,2 which may reflect an increased sensitivity to oxidative stress, as described for PERK–/– MEFs (50).



View larger version (31K):
[in this window]
[in a new window]
  		FIG. 7.
HRI is required for survival to arsenite. A, MEF viability. Wild-type and HRI–/– MEFs were exposed to arsenite (As, 100 µM) for 0–3 h, and cell viability was measured after 48 h. B, dose response for survival after arsenite injection into wild-type mice. Arsenite was injected subcutaneously at dosages of 5–20 mg/kg, and survival was determined after 24 h. n = 2–12 mice per dosage. C, arsenite survival in wild-type, HRI–/–, PKR–/–, and PKR–/–HRI–/– double mutant mice. Mice were injected with 15 mg/kg arsenite, and survival was determined after 24 h. n = number of mice. D, red blood cell counts in phenylhydrazine (PH)- and arsenite-treated mice. Mice were treated with phenylhydrazine or arsenite (15 mg/kg), and red blood cell counts were determined as described under "Experimental Procedures." E, eIF2{alpha} phosphorylation in liver extracts. Wild-type and HRI–/– mice (three each) were treated with sodium arsenite (15 mg/kg) or vehicle, and liver extracts were prepared after 4 h for Western blot analysis using antibodies specific to phosphorylated eIF2{alpha} as described under "Experimental Procedures." Blots were stripped and reprobed with antibodies specific for total eIF2{alpha}. Each lane represents a liver extract sample from different mice. The ratio (P/T) of phosphorylated (P) to total (T) eIF2{alpha} was determined by averaging the densities for each set of three individual mice.

 

To further investigate the role of HRI in survival upon arsenite exposure, we evaluated the toxicity of sodium arsenite in vivo in adult wild-type and HRI–/– mutant mice. In vivo, arsenite is toxic to target tissues such as liver, kidney, and red blood cells (46–48). Wild-type HRI+/+ mice were injected with a single dose of arsenite at concentrations of 5–20 mg/kg of body weight (Fig. 7B). The 24-h survival was 58% (7 of 12 mice) after a dosage of 15 mg/kg, consistent with the reported LD50 for acute arsenite exposure (49). The sublethal dosage of 15 mg/kg was used for comparison of survival for wild-type and HRI–/– mutant mice derived from littermates (Fig. 7C). Whereas greater than 55% of the wild-type mice survived arsenite injection, only 3% of the HRI–/– mutant mice survived (Fig. 7C). The body mass and arsenite dosages were comparable between both groups of mice. There was no significant difference in survival after arsenite exposure observed between male and female mice (not shown).

Previously PKR was suggested to be the eIF2{alpha} kinase that is activated in response to arsenite treatment (19). Therefore, to evaluate the relative roles of PKR and HRI in arsenite-mediated toxicity, we crossed the PKR-null mice with the HRI-null mice. The HRI–/–PKR–/– double mutant mice displayed no obvious significant phenotype. MEFs prepared from the HRI–/–PKR–/– double knock-out mice were resistant to translational inhibition in response to arsenite, similar to the HRI–/– single knock-out MEFs (data now shown). Upon injection of arsenite, the survival of PKR–/– mice was similar to the wild-type mice, whereas the survival of HRI–/–PKR–/– double knock-out mice was reduced to a level comparable to the survival of the HRI–/– single knock-out mice (Fig. 7C). Therefore, we conclude that PKR, in contrast to HRI, does not provide significant protection to arsenite exposure under these conditions in mice.

Analysis of red blood cells from HRI–/– mice indicated they are more sensitive to oxidative stress, such as that induced by the administration of phenylhydrazine (37). The survival of phenylhydrazine-treated HRI–/– mutant mice was dramatically reduced, and this correlated with red blood cell hemolysis. Therefore, we examined whether the reduced survival of HRI–/– mutant mice in response to arsenite treatment correlated with increased red blood cell hemolysis. Whereas phenylhydrazine significantly reduced the number of red blood cells in the wild-type mice, acute arsenite treatment actually slightly increased the red blood cell count in the HRI–/– mice (Fig. 7D). Therefore, we conclude that arsenite toxicity of the HRI–/– mice is likely due to defects in other organs that require HRI to protect against arsenite toxicity. Analysis of liver samples from arsenite-treated mice demonstrated a significant increase in phosphorylated eIF2{alpha} in wild-type mice, whereas there was a 2-fold lesser increase in HRI–/– mice (Fig. 7E). Therefore, upon arsenite treatment, HRI contributes significantly to eIF2{alpha} phosphorylation in the liver; however, there are additional kinase(s) that also mediate eIF2{alpha} phosphorylation under these conditions. It is possible that another eIF2{alpha} kinase may be activated indirectly as a consequence of arsenite treatment, e.g. amino acid imbalance or deficiency may occur and cause GCN2 activation. Thus, these studies establish that the HRI kinase not only mediates control of translation in response to arsenite stress but also identify a mechanism necessary for cellular and whole animal survival upon acute arsenite exposure.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
Treatment of cells with agents that are known to modify protein structure, such as heat shock or sodium arsenite, produce high levels of phosphorylated eIF2{alpha} and inhibit protein translation. However, there is no direct data to support that eIF2{alpha} phosphorylation is necessary and/or sufficient to inhibit protein synthesis in arsenite-treated cells nor has the responsible kinase(s) been identified. Recent studies demonstrate that PERK can mediate eIF2{alpha} phosphorylation in response to oxidative stress (50). Although arsenite induces formation of reactive oxygen species, it is not known whether arsenite treatment can activate PERK. Previous studies also suggested that PKR is the eIF2{alpha} kinase activated by arsenite (19, 20). However, since these studies were performed by expression of a trans-dominant negative mutant PKR, it is not known whether the dominant negative acts through inhibiting PKR activity or interfering with some additional regulator of eIF2{alpha} phosphorylation. Another study demonstrated that arsenite induces activation of the PKR activator PACT that was suggested to mediate eIF2{alpha} phosphorylation (51). However, the studies that implicate PKR in the arsenite translational response did not evaluate the potential importance of HRI. We compared the requirement for PERK, GCN2, PKR, and HRI in the arsenite translational response. Our studies support the idea that arsenite treatment activates HRI to phosphorylate eIF2{alpha} and that eIF2{alpha} phosphorylation is essential to inhibit protein synthesis. In addition, our studies support that PERK, GCN2, and PKR do not significantly mediate phosphorylation of eIF2{alpha} in response to arsenite. It was recently established that HRI is activated by arsenite in erythroid cells (30). However, it was unknown whether eIF2{alpha} phosphorylation is required and whether HRI is expressed at physiologically significant levels to mediate eIF2{alpha} phosphorylation and inhibition of translation upon arsenite exposure to non-erythroid cells.

Our findings that support that HRI is the kinase that mediates eIF2{alpha} phosphorylation upon arsenite treatment include the following. 1) Arsenite treatment did not induce eIF2{alpha} phosphorylation or significantly inhibit protein synthesis in HRI–/– MEFs, whereas protein synthesis was inhibited and eIF2{alpha} was phosphorylated in wild-type, GCN2–/–, PKR–/–, and PERK–/– MEFs. 2) Arsenite treatment disaggregated polysomes to a lesser degree in eIF2{alpha} A/A mutant or HRI–/– MEFs, whereas polysomes were dissociated in wild-type and PKR–/– MEFs. 3) eIF2{alpha} A/A mutant and HRI–/– MEFs did not form stress granules upon arsenite exposure compared with control cells that did form stress granules. 4) HRI–/– MEFs were more sensitive to arsenite treatment compared with control MEFs. 5) HRI–/– mice were more sensitive to arsenite toxicity compared with wild-type or PKR–/– mice. In addition, we demonstrated that HRI mRNA was expressed in MEFs at a low level. Therefore, this low level of expression appeared to be physiologically sufficient to mediate the translational inhibition in response to arsenite. Analysis of the PKR–/–, PERK–/–, GCN2–/–, and HRI–/– mutant MEFs strongly supports the hypothesis that HRI is the only eIF2{alpha} kinase that signals the arsenite stress response under these conditions.

These results also establish an essential requirement for phosphorylation of eIF2{alpha} in SG assembly. Homozygous A/A eIF2{alpha} mutant MEFs did not assemble SGs in response to arsenite, heat shock, or carbonyl cyanide p-trifluoromethoxyphenylhydrazone. In addition, transfection of a S51D eIF2{alpha} mutant, to mimic the phosphoserine, was sufficient to induce SGs in these cells in the absence of exogenous stress. These findings indicate that phosphorylation of eIF2{alpha} is essential and sufficient for SG assembly. Previous studies have demonstrated that acute energy starvation mediated by uncoupling oxidative phosphorylation through carbonyl cyanide p-trifluoromethoxyphenylhydrazone treatment induces the assembly of SGs without detectably increasing the phosphorylation of eIF2{alpha} (35). These findings led to the proposal that reduced levels of eIF2·GTP·tRNAMeti ternary complex is the key trigger for SG assembly rather than phosphorylation of eIF2{alpha} per se (33, 35). It is likely that basal levels of constitutively phosphorylated eIF2{alpha} are sufficient to induce SGs in wild-type cells treated with carbonyl cyanide p-trifluoromethoxyphenylhydrazone. Regardless the essential requirement for phosphorylated eIF2{alpha} demonstrated in this study requires a reconsideration of current models for SG assembly (36) that currently predict that SG assembly is a result of ternary complex insufficiency rather than due to eIF2{alpha} phosphorylation per se. The data presented here suggest a more direct role for phospho-eIF2{alpha} in SG assembly.

Phosphorylation of eIF2{alpha} is implicated in both apoptotic and antiapoptotic responses. It is proposed that eIF2{alpha} phosphorylation promotes apoptosis by preventing translation of inhibitors of apoptosis (41, 52). Alternatively eIF2{alpha} phosphorylation is implicated in survival by inhibiting translation when protein folding conditions are suboptimal and by mediating the preferential translation of protective factors, such as the activating transcription factor ATF4 (27, 38, 53). Therefore, it is possible that eIF2{alpha} phosphorylation may contribute to both survival and death responses upon arsenite exposure. However, our findings in MEFs and in mice support that HRI-mediated phosphorylation of eIF2{alpha} is a protective response to arsenite exposure. In contrast, the utility of arsenite as an anticancer agent may be attributed to its effects on another signaling pathway such as activation of c-Jun amino-terminal kinase or inhibition of NF{kappa}B activation (5, 6). Since HRI-mediated phosphorylation of eIF2{alpha} protects cells from arsenite exposure, there should be consideration to developing agents to inhibit HRI that may be used in conjunction with arsenite for therapeutic intervention in cancer.

Chronic arsenite exposure is a major public health concern affecting a large percentage of the world population. We have shown that in vivo survival upon acute arsenite exposure requires HRI activity. eIF2{alpha} phosphorylation in liver tissue was increased in an HRI-dependent manner, consistent with reports of liver toxicity and arsenite accumulation in the liver (48, 49). The identification of HRI as a protective factor for arsenite toxicity may also provide an avenue toward rational development of HRI agonists for the treatment and/or prevention of arsenite-induced disease and death.


   FOOTNOTES
 
* Portions of this work were supported by National Institutes of Health RO1 Grants DK42394 (to R. J. K.) and DK 16272 (to J.-J. C.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

** To whom correspondence should be addressed: Howard Hughes Medical Inst. and the Dept. of Biological Chemistry, University of Michigan Medical Center, Ann Arbor, MI 48109-0650. Tel.: 734-763-9037; Fax: 734-763-9323; E-mail: kaufmanr{at}umich.edu.

1 The abbreviations used are: eIF, eukaryotic translation initiation factor; HRI, heme-regulated inhibitor; ER, endoplasmic reticulum; SG, stress granule; TIA-1, T-cell internal antigen 1; MEF, murine embryonic fibroblast; PKR, double-stranded RNA-activated protein kinase; GCN, general control of nitrogen metabolism kinase; PERK/PEK, PKR-related ER kinase/pancreatic EIF2{alpha} kinase. Back

2 D. Scheuner and R. J. Kaufman, unpublished observations. Back


   ACKNOWLEDGMENTS
 
We gratefully thank Dr. David Ron and Heather Harding for providing PERK–/– and GCN2–/– MEFs for these studies and Dr. Bryan Williams for providing PKR–/– mice. We thank Jan Mitchell for extensive manuscript and figure preparation.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Abernathy, C. O., Liu, Y. P., Longfellow, D., Aposhian, H. V., Beck, B., Fowler, B., Goyer, R., Menzer, R., Rossman, T., Thompson, C., and Waalkes, M. (1999) Environ. Health Perspect. 107, 593–597[Medline] [Order article via Infotrieve]
  2. Tseng, C. H. (2004) Toxicol. Appl. Pharmacol. 197, 67–83[CrossRef][Medline] [Order article via Infotrieve]
  3. Snow, E. T. (1992) Pharmacol. Ther. 53, 31–65[CrossRef][Medline] [Order article via Infotrieve]
  4. Agency for Toxic Substances and Disease Registry (2003) Toxic Substance and Disease Registry, www.atsdr.cdc.gov, Centers for Disease Control and Prevention, Atlanta, GA
  5. Shen, Z. X., Chen, G. Q., Ni, J. H., Li, X. S., Xiong, S. M., Qiu, Q. Y., Zhu, J., Tang, W., Sun, G. L., Yang, K. Q., Chen, Y., Zhou, L., Fang, Z. W., Wang, Y. T., Ma, J., Zhang, P., Zhang, T. D., Chen, S. J., Chen, Z., and Wang, Z. Y. (1997) Blood 89, 3354–3360[Abstract/Free Full Text]
  6. Hayashi, T., Hideshima, T., and Anderson, K. C. (2003) Br. J. Haematol. 120, 10–17[CrossRef][Medline] [Order article via Infotrieve]
  7. Thompson, D. J. (1993) Chem.-Biol. Interact. 88, 89–114[CrossRef][Medline] [Order article via Infotrieve]
  8. Bredfeldt, T. G., Kopplin, M. J., and Gandolfi, A. J. (2004) Toxicol. Appl. Pharmacol. 198, 412–418[CrossRef][Medline] [Order article via Infotrieve]
  9. Bernstam, L., and Nriagu, J. (2000) J. Toxicol. Environ. Health B Crit. Rev. 3, 293–322[CrossRef][Medline] [Order article via Infotrieve]
  10. Cavigelli, M., Li, W. W., Lin, A., Su, B., Yoshioka, K., and Karin, M. (1996) EMBO J. 15, 6269–6279[Medline] [Order article via Infotrieve]
  11. Bode, A. M., and Dong, Z. (2002) Crit. Rev. Oncol. Hematol. 42, 5–24[Medline] [Order article via Infotrieve]
  12. Kapahi, P., Takahashi, T., Natoli, G., Adams, S. R., Chen, Y., Tsien, R. Y., and Karin, M. (2000) J. Biol. Chem. 275, 36062–36066[Abstract/Free Full Text]
  13. Cheng, H. Y., Li, P., David, M., Smithgall, T. E., Feng, L., and Lieberman, M. W. (2004) Oncogene 23, 3603–3612[CrossRef][Medline] [Order article via Infotrieve]
  14. Mosser, D. D., Duchaine, J., and Massie, B. (1993) Mol. Cell. Biol. 13, 5427–5438[Abstract/Free Full Text]
  15. Palmiter, R. D. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 8428–8430[Abstract/Free Full Text]
  16. Kreppel, H., Bauman, J. W., Liu, J., McKim, J. M., Jr., and Klaassen, C. D. (1993). Fundam. Appl. Toxicol. 20, 184–189[CrossRef][Medline] [Order article via Infotrieve]
  17. Keyse, S. M., and Tyrrell, R. M. (1989) Proc. Natl. Acad. Sci. U. S. A. 86, 99–103[Abstract/Free Full Text]
  18. Bauman, J. W., Liu, J., and Klaassen, C. D. (1993) Fundam. Appl. Toxicol. 21, 15–22[CrossRef][Medline] [Order article via Infotrieve]
  19. Brostrom, C. O., Prostko, C. R., Kaufman, R. J., and Brostrom, M. A. (1996) J. Biol. Chem. 271, 24995–25002[Abstract/Free Full Text]
  20. Hershey, J., and Merrick, W. (2000) in Translational Control of Gene Expression (Sonenberg, N., Hershey, J. W., and Matthews, M. B, eds) pp. 33–88, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY
  21. Duncan, R. F., and Hershey, J. W. (1987) Arch. Biochem. Biophys. 256, 651–661[CrossRef][Medline] [Order article via Infotrieve]
  22. Oldfield, S., Jones, B. L., Tanton, D., and Proud, C. G. (1994) Eur. J. Biochem. 221, 399–410[Medline] [Order article via Infotrieve]
  23. Rowlands, A. G., Panniers, R., and Henshaw, E. C. (1988) J. Biol. Chem. 263, 5526–5533[Abstract/Free Full Text]
  24. Kaufman, R. J. (2000) in Translational Control of Gene Expression (Sonenberg, N., Hershey, J. W., and Matthews, M. B, eds) pp. 503–527, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY
  25. Hinnebusch, A. (2000) in Translational Control of Gene Expression (Sonenberg, N., Hershey, J. W., and Matthews, M. B, eds) pp. 185–243, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY
  26. Ron, D., and Harding, H. (2000) in Translational Control of Gene Expression (Sonenberg, N., Hershey, J. W., and Matthews, M. B, eds) pp. 547–560, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY
  27. Harding, H. P., Zhang, Y., Bertolotti, A., Zeng, H., and Ron, D. (2000) Mol. Cell 5, 897–904[CrossRef][Medline] [Order article via Infotrieve]
  28. Chen, J. J. (2000) in Translational Control of Gene Expression (Sonenberg, N., Hershey, J. W., and Matthews, M. B, eds) pp. 529–426, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY
  29. Chen, J. J., and London, I. M. (1995) Trends Biochem. Sci. 20, 105–108[CrossRef][Medline] [Order article via Infotrieve]
  30. Lu, L., Han, A. P., and Chen, J. J. (2001) Mol. Cell. Biol. 21, 7971–7980[Abstract/Free Full Text]
  31. Nover, L., Scharf, K. D., and Neumann, D. (1983) Mol. Cell. Biol. 3, 1648–1655[Abstract/Free Full Text]
  32. Kedersha, N. L., Gupta, M., Li, W., Miller, I., and Anderson, P. (1999) J. Cell Biol. 147, 1431–1442[Abstract/Free Full Text]
  33. Kedersha, N., Chen, S., Gilks, N., Li, W., Miller, I. J., Stahl, J., and Anderson, P. (2002) Mol. Biol. Cell 13, 195–210[Abstract/Free Full Text]
  34. Kedersha, N., Cho, M. R., Li, W., Yacono, P. W., Chen, S., Gilks, N., Golan, D. E., and Anderson, P. (2000) J. Cell Biol. 151, 1257–1268[Abstract/Free Full Text]
  35. Kimball, S. R., Horetsky, R. L., Ron, D., Jefferson, L. S., and Harding, H. P. (2003) Am. J. Physiol. 284, C273–C284
  36. Anderson, P., and Kedersha, N. (2002) J. Cell Sci. 115, 3227–3234[Abstract/Free Full Text]
  37. Han, A. P., Yu, C., Lu, L., Fujiwara, Y., Browne, C., Chin, G., Fleming, M., Leboulch, P., Orkin, S. H., and Chen, J. J. (2001) EMBO J. 20, 6909–6918[CrossRef][Medline] [Order article via Infotrieve]
  38. Scheuner, D., Song, B., McEwen, E., Lui, C., Laybutt, R., Gillespie, P., Saunders, T., Bonner-Weir, S., and Kaufman, R. J. (2001). Mol. Cell 7, 1165–1176[CrossRef][Medline] [Order article via Infotrieve]
  39. Yang, Y.-L., Reis, L. F., Pavlovic, J., Aguzzi, A., Schäfer, R., Kumar, A., Williams, B. R., Aguet, M., and Weissmann, C. (1995). EMBO J. 14, 6095–6106[Medline] [Order article via Infotrieve]
  40. Hogan, B., Beddington, R., Costantini, F., and Lacy, E. (1994) Manipulating the Mouse Embryo, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY
  41. Srivastava, S. P., Kumar, K. U., and Kaufman, R. J. (1998) J. Biol. Chem. 273, 2416–2423[Abstract/Free Full Text]
  42. Johannes, G., and Sarnow, P. (1998) RNA 4, 1500–1513[Abstract]
  43. Mellor, H., Flowers, K. M., Kimball, S. R., and Jefferson, L. S. (1994) Biochim. Biophys. Acta 1219, 693–696[Medline] [Order article via Infotrieve]
  44. Crosby, J. S., Lee, K., London, I. M., and Chen, J. J. (1994) Mol. Cell. Biol. 14, 3906–3914[Abstract/Free Full Text]
  45. Kaufman, R. J., Davies, M. V., Pathak, V. K., and Hershey, J. W. (1989) Mol. Cell. Biol. 9, 946–958[Abstract/Free Full Text]
  46. Westhoff, D. D., Samaha, R. J., and Barnes, A., Jr. (1975) Blood 45, 241–246[Abstract/Free Full Text]
  47. Feussner, J. R., Shelburne, J. D., Bredehoeft, S., and Cohen, H. J. (1979) Blood 53, 820–827[Abstract/Free Full Text]
  48. Morse, B. S., Conlan, M., Giuliani, D. G., and Nussbaum, M. (1980) Am. J. Hematol. 8, 273–280[Medline] [Order article via Infotrieve]
  49. Liu, J., Kadiiska, M. B., Liu, Y., Lu, T., Qu, W., and Waalkes, M. P. (2001) Toxicol. Sci. 61, 314–320[Abstract/Free Full Text]
  50. Harding, H. P., Zhang, Y., Zeng, H., Novoa, I., Lu, P. D., Calfon, M., Sadri, N., Yun, C., Popko, B., Paules, R., Stojdl, D. F., Bell, J. C., Hettmann, T., Leiden, J. M., and Ron, D. (2003) Mol. Cell 11, 619–633[CrossRef][Medline] [Order article via Infotrieve]
  51. Patel, C. V., Handy, I., Goldsmith, T., and Patel, R. C. (2000) J. Biol. Chem. 275, 37993–37998[Abstract/Free Full Text]
  52. Donze, O., Deng, J., Curran, J., Sladek, R., Picard, D., and Sonenberg, N. (2004) EMBO J. 23, 564–571[CrossRef][Medline] [Order article via Infotrieve]
  53. Harding, H. P., Novoa, I., Zhang, Y., Zeng, H., Wek, R., Schapira, M., and Ron, D. (2000) Mol. Cell 6, 1099–1108[CrossRef][Medline]