QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Data Supplement All Versions of this Article: jpet.107.124602v1 323/3/979    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Liao, M. Articles by Correia, M. A. Search for Related Content PubMed PubMed Citation Articles by Liao, M. Articles by Correia, M. A.

CELLULAR AND MOLECULAR

Impaired Dexamethasone-Mediated Induction of Tryptophan 2,3-Dioxygenase in Heme-Deficient Rat Hepatocytes: Translational Control by a Hepatic eIF2 Kinase, the Heme-Regulated Inhibitor

Departments of Cellular & Molecular Pharmacology (M.L., M.K.P., YQ.W., M.A.C.) and Pharmaceutical Chemistry (D.A.M., K.F.M., S.P.S.-C., M.A.C.), Biopharmaceutical Sciences (M.A.C.), and Medicine (C.H., J.Y., J.J.M., M.A.C.), and the Liver Center, University of California, San Francisco, California

Received for publication April 18, 2007
Accepted August 28, 2007.

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
Tryptophan 2,3-dioxygenase (TDO), a liver-specific cytosolic hemoprotein, is the rate-limiting enzyme in L-tryptophan catabolism and thus a key serotonergic determinant. Glucocorticoids transcriptionally activate the TDO gene with marked enzyme induction. TDO is also regulated by heme, its prosthetic moiety, as its expression and function are significantly reduced after acute hepatic heme depletion. Here we show in primary rat hepatocytes that this impairment is not due to faulty transcriptional activation of the TDO gene but rather due to its posttranscriptional regulation by heme. Accordingly, in acutely heme-depleted hepatocytes, the de novo synthesis of TDO protein is markedly decreased (>90%) along with that of other hepatic proteins. This global suppression of de novo hepatic protein syntheses in these heme-depleted cells is associated with a significantly enhanced phosphorylation of the -subunit of the eukaryotic initiation factor eIF2 (eIF2), as monitored by the phosphorylated eIF2/total eIF2 ratio. Heme supplementation reversed these effects, indicating that heme regulates TDO induction by functional control of an eIF2 kinase. A cDNA was cloned from heme-depleted rat hepatocytes, and DNA sequencing verified its identity to the previously cloned rat brain heme-regulated inhibitor (HRI). Proteomic, biochemical, and/or immunoblotting analyses of the purified recombinant protein and the immunoaffinity-captured hepatic protein confirmed its identity as a rat heme-sensitive eIF2 kinase. These findings not only document that a hepatic HRI exists and is physiologically relevant but also implicate its translational shut-off of key proteins in the pathogenesis and symptomatology of the acute hepatic heme-deficient conditions clinically known as the hepatic porphyrias.

Hepatic TDO is induced via glucocorticoid-mediated transcriptional activation and by substrate stabilization (Danesch et al., 1987; Nakamura et al., 1987). We have reported that heme is required for both basal and dexamethasone (DEX)-inducible TDO expression and function (Ren and Correia, 2000). Accordingly, in rats acutely depleted of hepatic heme by in vivo treatment with DDEP [3,5-dicarbethoxy-2,6-dimethyl-4-ethyl-1,4-dihydropyridine, a cytochrome P450 suicide inactivator that destroys the prosthetic heme by converting it into N-ethylprotoporphyrin(s), highly potent inhibitors of the heme synthetic enzyme ferrochelatase (De Matteis et al., 1980; Ortiz de Montellano et al., 1980, 1981)], both basal and DEX-mediated transcriptional activation of hepatic TDO was significantly (40–50%) impaired as monitored by Northern analyses. This impairment was reversed by in vivo heme administration (Ren and Correia, 2000). Run-on assays with hepatic nuclei isolated from these rats revealed that acute hepatic heme depletion markedly reduced the basal TDO transcriptional rates and that DEX treatment of these rats only partially restored these rates to those of DEX-treated controls. Full restoration was observed after in vivo hemin administration but not after inclusion of hemin in vitro in these assays, thereby indicating the importance of intrahepatic heme for this transcriptional process but excluding a direct effect of hemin. These findings thus suggested that heme may play a critical role in DEX-mediated induction of the TDO gene by interacting with a transcriptional activator or repressor.

As an initial approach to elucidation of the potential mechanism of such heme regulation, we examined the influence of heme on DEX-mediated expression of endogenous TDO mRNA and protein in primary cultured hepatocytes acutely depleted of heme. Our findings reveal that although heme regulates hepatic TDO expression, this regulation is at a posttranscriptional step through translational control by a heme-sensitive hepatic eukaryotic initiation factor 2 -subunit (eIF2) kinase (also known as heme-regulated inhibitor, HRI). Functional unleashing of this eIF2 kinase in heme deficiency results in the phosphorylation of the -subunit of eIF2 (Chen et al., 1989, 1991; Pal et al., 1991; Crosby et al., 1994; Chen and London, 1995; Chen, 2006) and consequent global suppression of de novo syntheses of hepatic proteins, including TDO and CYP2B (Han et al., 2005).

Given its critical importance in controlling hepatic protein synthesis in heme-deficient states and because the heme-regulated eIF2 kinase (HRI) has been claimed to be specific to erythroid cells and their progenitors (Pal et al., 1991; Crosby et al., 1994; Chen and London, 1995), we have successfully cloned the cDNA for this elusive eIF2 kinase from acutely heme-depleted intact rat livers and size-elutriated rat hepatocytes and, through DNA sequencing, found it to be identical to the previously cloned rat brain HRI cDNA (Mellor et al., 1994). Our quantitative real-time PCR (qRT-PCR) analyses of mRNA isolated from acutely heme-depleted hepatocytes coupled with HRI-immunoblotting analyses of their lysate protein indicated that hepatic HRI transcriptional activation and/or content are unaffected by heme deficiency. Thus, the global suppression of protein synthesis observed in these heme-depleted hepatocytes is most probably due to the functional activation of an HRI enzyme and consequent enhancement of eIF2 phosphorylation. We have heterologously expressed the isolated full-length rat hepatic HRI cDNA and purified the recombinant protein to near homogeneity and immunoaffinity-purified the HRI protein from size-elutriated rat hepatocytes. Proteomic, immunoblotting, and/or biochemical analyses of the recombinant and native proteins unequivocally establish their functional identity as a rat hepatic heme-sensitive eIF2 kinase (HRI).

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Materials. Common cell culture medium and supplements such as Williams' Medium E (WME), insulin-transferrin-selenium-G (100x), bovine serum albumin, penicillin/streptomycin, L-glutamine, liver digestion medium, and liver perfusion medium were obtained from Invitrogen Life Technologies (Carlsbad, CA). Methionine/cysteine-free WME was prepared by the University of California San Francisco (UCSF) Cell Culture Facility (San Francisco, CA). Collagen type I was prepared by the UCSF Liver Center Cell and Tissue Biology Core Facility. Matrigel was obtained from BD Biosciences (Bedford, MA). Petri dishes (60 mm, Permanox) were purchased from Nalge Nunc International (Rochester, NY). Phenylmethylsulfonyl fluoride, hemin¹ hydrochloride, and DEX were purchased from Sigma-Aldrich (St. Louis, MO). DDEP was synthesized as described previously (Litman and Correia, 1983, 1985). N-Methylprotoporphyrin IX (NMPP, a mixture of all four isomers) was purchased from Frontier Scientific Inc (Logan, UT). E-64, leupeptin, and aprotinin were purchased from Roche Applied Science (Indianapolis, IN). Dimethyl pimelimidate·2HCl was purchased from Pierce (Rockford, IL). [³⁵S]EXPRESS methionine was purchased from PerkinElmer Life and Analytical Sciences (Boston, MA). Cloning reagents, such as restriction enzymes, ligases, and Vent polymerase, were obtained from New England BioLabs (Beverly, MA). Primers and probes were from IDT Inc. (Coralville, IA). Polyclonal antibodies against eIF2 and phosphospecific eIF2 (pS52) were purchased from BioSource Intel. Inc. (Camarillo, CA). Rabbit polyclonal IgGs were raised commercially against purified recombinant rat hepatic tryptophan 2,3-dioxygenase and eIF2 kinase (HRI) and purified by Hi-Trap Protein A-Sepharose affinity chromatography. Reagents for TaqMan analyses were obtained from ABI Prism (Foster City, CA).

Rat Hepatocyte Isolation. Male Sprague-Dawley rats (225–250 g) were purchased from Charles River (Wilmington, MA) and maintained on a 12-h light/dark cycle in a controlled environment at the UCSF Animal Care Facility. Hepatocytes were isolated from rats by in situ perfusion of the liver with collagenase and purified by centrifugal elutriation as described previously (Han et al., 2005).

Endogenous TDO or HRI mRNA and Protein Expression. These were examined in primary rat hepatocytes cultured in a Matrigel-collagen type I sandwich exactly as described previously (Han et al., 2005). In these experiments, DEX (0.1 µM) is included throughout the cell culture to maintain P450 inducibility and function, required for DDEP-mediated inhibition of ferrochelatase and consequent hepatic heme depletion. Cells were allowed to recover for 72 h with a daily change of medium before initiation of any treatment. At 72 h, cells were treated with DDEP (10 µM) to deplete heme, and at 84 h, DEX (5 µM) was added along with DMSO (vehicle) or DDEP (10 µM). Wherever indicated, hemin (20 µM) was also included at 73 h, and all cells were harvested at 90 h. In most experiments, DDEP was replaced by NMPP (1 µM). Total RNA was analyzed by qRT-PCR (TaqMan) analyses with a TDO or HRI-specific probe and rat -glucuronidase (GUS) gene as the housekeeping control. Cell lysate protein was subjected to TDO- or HRI-immunoblotting analyses (Ren and Correia, 2000).

Quantitative RT-PCR (TaqMan) Analyses. Total RNA was extracted with the QIAGEN RNeasy mini kit (Valencia, CA) and then reverse-transcribed to cDNA by Moloney murine leukemia virus reverse transcriptase (Invitrogen Life Technologies) exactly as described previously (Han et al., 2005). Primers and probes for qRT-PCR analysis were designed with the Primer Express software (Version 2.0; Applied Biosystems, Foster City, CA). The TDO-specific primers were 5'-GGCTATTATTATCTGCGCTCAACTG-3' (forward) and 5'-GAACCAGGTACGATGAGAGGTTAAA-3' (reverse). The TDO-specific TaqMan probe was 5'-AGCGACAGGTACAAGGTGTTCGTGGATT-3'. Amplification reactions were performed in a mixture (50 µl) with final concentrations as follows: 1 x TaqMan buffer, pH 8.3, MgCl₂ (5.5 mM), dATP (200 µM), dGTP (200 µM), dCTP (200 µM) and dTTP (200 µM), AmpliTaq Gold DNA Polymerase (1.25 U), probes (200 nM), primers (900 nM), diethylpyrocarbonate-treated water, and cDNA (10 ng, based on the RNA concentration added to RT reaction). PCR was performed employing an ABI Prism TaqMan 7300 sequence detector system with denaturation at 95°C for 2 min, 40 cycles with denaturation at 95°C for 15 s, and annealing/extension at 60°C for 1 min (Han et al., 2005). The expression level of each gene was normalized to the endogenous expression of the control gene (rat GUS) as described previously (Han et al., 2005). Values are expressed as percentage of the value from nontreated hepatocytes.

qRT-PCR analyses of HRI were similarly conducted on RNA isolated from untreated DEX- and/or NMPP-treated ± heme-treated cultured rat hepatocytes. Primers and probes were designed similarly for the detection of mRNA sequences for rat HRI and GUS. The HRI primers were 5'-CACTGCATGGATAGAGCACGTT-3' (forward) and 5'-GGGCAGTTGAATGGGAACTC-3' (reverse). The HRI-specific TaqMan probe was 5'-CGTGCTTCAGCCACAAG-3'. The 25 µl-reaction mixture for PCR was similar to the one above with the exception of the following: TaqMan Universal PCR buffer, 200 nM probes, 900 nM primers, and 25 ng of cDNA. PCR and data analyses were performed exactly as described above for the TDO gene and expressed relative to that of the untreated control value.

[³⁵S]Methionine/Cysteine Labeling of Cell Protein. Cells were cultured and labeled as described previously (Han et al., 2005) with modifications. In brief, cell cultures were treated with various chemicals for 23 h. The media were replaced with methionine/cysteine-free WME for 1 h and then pulsed with 50 µCi/ml [³⁵S]EXPRESS for 1 h. Each ³⁵S-labeled culture was washed twice with ice-cold PBS containing methionine (0.2 mM) and cysteine (1.4 mM) and then lysed with 0.4 ml of HEPES buffer (20 mM), pH 7.5, containing 1% Triton X-100, NaCl (150 mM), 10% glycerol, EDTA (1 mM), NaF (100 mM), tetrabasic sodium pyrophosphate (10 mM), -glycerophosphate (17.5 mM), phenylmethylsulfonyl fluoride (1 mM), leupeptin (2 µM), aprotinin (0.04 U/ml), and E-64 (50 µM). The lysate was sedimented at 10,000g at 4°C for 10 min, and the supernatant was saved for further study. To insure equivalent [³⁵S] uptake into hepatocytes, the radioactivity of aliquots of the initial lysate was monitored in Ecolume (4 ml; ICN Biomedicals Inc., Costa Mesa, CA) by liquid scintillation spectrometry using a Beckman LS3801 liquid scintillation counter (Beckman Instruments, Fullerton, CA).

[³⁵S]TDO Immunoprecipitation. Rabbit anti-rat TDO IgGs (0.5 mg) were added to ice-cold PBS-equilibrated Protein A-Sepharose slurry (50 µl) and then incubated at 4°C overnight to prepare the antibody-conjugated beads. Lysate protein (450 µg) was adjusted to 2% SDS, boiled for 5 min, and diluted 1:4 (v/v), with Tris-HCl (50 mM, pH 7.4), Triton X-100 (2.5%, v/v), NaCl (190 mM), and EDTA (6 mM), and added to the antibody-conjugated beads. The mixture was then incubated at 4°C for 2 h with end-to-end rotation. The antigen/antibody/Protein A-Sepharose complex was collected by centrifugation and washed five times with PBS containing 0.1% SDS and 0.5% Nonidet P-40. The antigen was eluted by heating the complex for 5 min in the sample loading buffer (80 µl, 62.5 mM Tris-HCl, pH 6.8, 25% glycerol, 10% SDS, 5% -mercaptoethanol, and 0.01% bromphenol blue). The radioactivity of a 10 µl-aliquot was monitored in Ecolume (4 ml) by liquid scintillation spectrometry as described above.

Total Protein [³⁵S] Incorporation. Cold rat liver microsomal protein (6–7 mg) was added as a protein carrier to ³⁵S-labeled cell lysate protein (1 mg), and the protein was precipitated with 5% H₂SO₄ in methanol (v/v) followed by at the least five washes with the same solution. Pellets were then washed sequentially with organic solvents and dissolved exactly as described previously (Han et al., 2005). Aliquots were used for determination of the protein concentration and specific radioactivity. Total ³⁵S protein incorporation was calculated as counts per min/milligram protein/hour.

Immunoblotting Analyses. Cultured hepatocytes were washed three times with ice-cold PBS before harvesting with "lysate" buffer (0.4 ml) consisting of Tris-HCl (20 mM, pH 7.5), NaCl (150 mM), Na₂EDTA (1 mM), EGTA (1 mM), 1% Triton X-100, sodium pyrophosphate (2.5 mM), -glycerophosphate (1 mM), Na₃VO₄ (1 mM), protease inhibitors [leupeptin (4 µM), pepstatin (3 µM), aprotinin (1.3 inhibitory units/ml), antipain (3 µM), 4-(2-aminoethyl)benzenesulfonyl fluoride (1 mM), bestatin (6 µM), and E-64 (50 µM)], 10% glycerol, and NaF (100 mM). The cell lysates were clarified by sedimentation at 10,000g for 10 min at 4°C, and the supernatants were stored at –80°C before immunoblotting analyses. TDO content was determined by Western immunoblotting analysis with rabbit polyclonal anti-rat TDO IgGs [1:7000 (v/v)] and a secondary goat anti-rabbit horseradish peroxidase (HRP)-coupled antibody [1:20,000 (v/v)]. eIF2 and phosphorylated eIF2 (eIF2P) immunoblotting analyses were carried out exactly as described previously (Han et al., 2005). Whereas the anti-human eIF2 IgGs recognize both phosphorylated and nonphosphorylated species of the protein, anti-human eIF2P IgGs recognize only the phosphorylated species. HRI-immunoblotting analysis was performed similarly using rabbit polyclonal IgGs [1:3000 (v/v)] raised against the recombinant intact rat hepatic HRI and purified by HiTrap Protein A affinity chromatography (Amersham Pharmacia Biotech, Little Chalfont, Bucking-hamshire, UK), a secondary goat anti-rabbit HRP-coupled antibody [1:20,000 (v/v)], and Pierce SuperSignal detection system. Parallel immunoblotting analyses of actin in these lysates served as the loading control for normalization.

Cloning of a Rat Liver eIF2 Kinase (HRI) cDNA by RT-PCR. Total RNA was isolated from well perfused DDEP-treated rat livers or NMPP-treated cultured size-elutriated hepatocytes by the QIAGEN RNeasy Mini Kit as detailed previously (Han et al., 2005) and treated with DNase I to get rid of any possible trace DNA contamination. DNA-free RNA was used to synthesize eIF2 kinase cDNA by reverse transcription and PCR techniques using the primers 5'-AAC GAT GCT GGG GGG CGG CTC-3' (forward) and 5'-CCC TCT CAT CTC TTC AGC CCT TTG TCC TGC G-3' (reverse). The primers were based on the known rat brain HRI gene sequence (GenBank accession number NM013223) (Mellor et al.,1994). The RT-PCR fragments were then cloned into pGEM-T Easy TA vectors and confirmed by DNA sequencing. The full-length rat hepatic HRI cDNA was sequenced twice from both directions. Site-directed mutagenesis was used to introduce an NdeI restriction enzyme site upstream of the HRI gene, and the full-length HRI cDNA was then cloned into pET28b vector, resulting in a plasmid encoding an HRI fusion protein with an N-terminal (His)₆ tag.

Expression and Purification of Recombinant HRI in Escherichia coli. This full-length (His)₆HRI was expressed in E. coli BL21 cells. Bacterial culture was grown at 37°C to an OD₆₀₀ of 0.6 and then induced with 1 mM isopropyl -D-thiogalactoside, and the temperature was then reduced to 15°C for 18 h. After induction, cells were harvested at the indicated times, suspended, and lysed by sonication in a 50 mM imidazole-containing buffer (300 mM NaCl, 50 mM NaH₂PO₄, pH 8.0). The 90,000-g supernatant was subjected to fast-performance liquid chromatography separation on a Ni²⁺-NTA-agarose column with a 50 to 500 mM imidazole concentration gradient. The protein eluting around 275 mM imidazole was dialyzed and stored at –80°C in PBS containing 10% glycerol. Protein purity was verified by SDS-PAGE. Fractions containing the purified HRI protein were pooled and concentrated by Amicon ultrafiltration. The identity of the purified protein was confirmed by LC-MS/MS of tryptic digests. A larger-scale purification of this protein with an additional size-exclusion chromatographic step for further purification was conducted for commercial polyclonal antibody generation in rabbits.

LC-MS/MS of Tryptic Digests. Aliquots of the purified recombinant HRI protein were reduced, alkylated with iodoacetamide, and digested with side-chain protected porcine trypsin (http://ms-facility.ucsf.edu/ingel.html). To confirm the identity of the protein, an aliquot from this digest was subjected to LC-MS/MS analysis on a quadrupole-orthogonal acceleration-time-of-flight hybrid tandem mass spectrometer (QSTAR XL; MDS Sciex, Concord, ON, Canada) in an information-dependent fashion; one-second mass measurements were followed by 3-s collision-induced dissociation (CID) experiments on computer-selected multiply charged ions. The collision conditions were adjusted automatically to the mass and the charge state of the selected precursor. A database search (http://prospector2.ucsf.edu) was performed with the acquired CID data against the SwissProt database. A mass accuracy of 200 and 300 ppm was considered for the precursor and fragment ions, respectively. Only tryptic peptides were accepted, and one missed cleavage was permitted. Cys carbamidomethylation was selected as a fixed modification, whereas the acetylation of protein N termini and Met oxidation and pyroglutamic acid formation from N-terminal Gln residues were considered as variable modifications. Peptides with probability scores of p < 0.05 were accepted.

Protein Kinase Assay. The purified recombinant HRI was functionally assayed for its ability to phosphorylate eIF2 as its substrate in vitro (Berlanga et al., 1998). In a final volume of 20 µl, the incubation reaction contained HRI (50 ng), purified recombinant eIF2 (500 ng) in 20 mM Tris buffer (pH 7.4) containing KCl (40 mM), magnesium acetate (3 mM), and DTT (1 mM). The reactions were initiated by the addition of 5 µCi of [-³²P]ATP (3000 Ci/mmol) and nonradioactive ATP (50 µM), incubated at 30°C for 20 min, and terminated by the addition of SDS-sample buffer. Aliquots were subjected to 4 to 15% SDS-PAGE, and the gels quantitated by fluorography with phosphorimaging analyses.

Heme Sensitivity. Purified recombinant HRI was incubated with various concentrations of hemin (0–5 µM) at 30°C for 10 min as described above (Berlanga et al., 1998). The relative sensitivity of eIF2 phosphorylation to each hemin concentration was assayed as described above.

Immunoaffinity Chromatographic Isolation of HRI Protein from Rat Hepatocytes. This was carried out by a modification of previously reported methods (Schneider et al., 1982; Chefalo et al., 1998). Rabbit polyclonal anti-HRI IgG (20 mg), purified by HiTrap Protein A affinity chromatography, was incubated with PBS-equilibrated Protein A-Sepharose (3 mg/ml) by gentle end-to-end mixing for 2 h at room temperature and then washed extensively to remove unbound IgGs. The IgG-bound Protein A-Sepharose matrix was equilibrated with cross-linking buffer (0.1 M triethanolamine, pH 8.2) and then treated with gentle mixing with three sequential (5, 2.5, and 2.5 ml) aliquots of freshly prepared cross-linker dimethyl pimelimidate·2HCl (20 mM) in 0.1 M triethanolamine, pH 8.2. The IgG-cross-linked Protein A-Sepharose beads were washed with 5 column volumes of 0.1 M triethanolamine, pH 8.2, and the cross-linking reaction was terminated with 5 column volumes of 0.1 M ethanolamine, pH 8.0, followed by equilibration of the beads with 20 column volumes of PBS.

Freshly isolated, size-elutriated, and/or Percoll gradient-purified rat hepatocytes (120 x 10⁶) were washed in PBS and then lysed in the lysis buffer (Cell Signaling Technology Inc., Danvers, MA) supplemented with the additional protease inhibitors described above, as well as Triton X-100 and deoxycholate (DOC) to a final concentration of 2 and 1%, respectively. The mixture was allowed to stand on ice for 15 min before centrifugation at 20,000g for 30 min to remove insoluble cell debris. The supernatant was collected, adjusted to 0.3 M NaCl, and then precleared with PBS-equilibrated Protein A-Sepharose beads at 4°C with gentle rolling for 30 min, followed by recentrifugation at 20,000g for 30 min. The precleared cell lysate supernatant was then mixed with the PBS-equilibrated IgG-cross-linked Protein A-Sepharose beads with gentle mixing at 4°C for 4 h. The mixture was poured in a 20-ml glass column and then washed sequentially with i) 0.5 M NaCl, 0.05 M Tris-HCl, pH 8.2, 1 mM EDTA, and 0.5% Nonidet P-40; ii) 0.15 M NaCl, 0.05 M Tris-HCl, pH 8.2, 1 mM EDTA, 0.5% Nonidet P-40, and 0.1% SDS; iii) 0.15 M NaCl and 0.5% DOC; and iv) 0.15 M NaCl and 0.5% DOC. The bound HRI protein was then eluted sequentially with i) diethylamine (10 mM, pH 10.5), ii) triethylamine (10 mM, pH 11.5), and iii) triethylamine (100 mM, pH 11.5) in the presence of 10% glycerol and 0.5% DOC. The last step was repeated once. The eluates were collected and rapidly adjusted to pH 7.4 by the addition of Tris-HCl (1 M, pH 6.8) and concentrated with a Centricon 10 centrifugal device (Millipore Corporation, Billerica, MA), followed by three sequential washing dialyses steps with a Tris-HCl buffer (pH 7.8) containing 50 mM KCl, 0.1 mM EDTA, and 10% glycerol, the last step including additional DTT (2 mM) in the buffer. Each concentrated dialyzed eluate was supplemented with DTT (4 mM final) and stored in aliquots at –80°C until further analyses.

Aliquots of the initial rat hepatocyte lysate (RHL), the flow-through (FT), and each of the four eluates termed E1, E2, E3, and E4 were subjected to SDS-PAGE analyses on 7.5% gels, as well as immunoblotting analyses with rabbit polyclonal anti-rat HRI IgGs [5.5 mg/ml; 1:3300 v/v in 3% nonfat milk-Tris-buffered saline containing 0.5% Tween-20], as the primary antibody and overnight incubation at room temperature with gentle rotation, followed by washes and secondary antibody [goat anti-rabbit HRP-coupled antibody (1:30,000, v/v) in 3% nonfat milk-Tris-buffered saline containing 0.5% Tween-20] for 1 h at room temperature, and detection with Pierce SuperSignal system. The eluate predominantly exhibiting a 76-kDa protein band after SDS-PAGE was subjected to proteomic analysis after slicing of the gel band, reduction, and alkylation with iodoacetamide, in situ tryptic digestion, and LC-MS/MS analyses of the tryptic digests as described above.

Analysis of the SDS-PAGE-Fractionated Proteins. Samples were in-gel-digested with trypsin (http://ms-facility.ucsf.edu/ingel.html) and subjected to LC-MS/MS analysis as follows. Peptide fractionation was carried out on an 1100 nanoHPLC system (reversed phase, C18 column, 75 µm x 150 mm; flow rate 300 nl/min; solvent A: 0.1% formic acid in water; solvent B: 0.1% formic acid in acetonitrile) (Agilent Technologies, Palo Alto, CA). The column was equilibrated with 2% B; 2 µl of tryptic digests were injected and eluted with a gradient: 2 to 50% B over 48 min. Mass spectrometric data acquisition was performed on a hybrid linear ion trap-Fourier transform-ion cyclotron resonance mass spectrometer (Thermo, Bremen, Germany) equipped with a 7T ion cyclotron resonance magnet. Mass measurements and 10-Da zoom-in scans of the three most abundant ions were carried out in the FT; CID analyses of the same ions, if multiply charged, were performed in the ion trap; and dynamic exclusion was enabled. Peaklists were generated using Mascot Distiller version 2.1.0.0 program; ProteinProspector version 4.25.3 was used for the database search with the following parameters: digest (Trypsin), maximal missed cleavages: database, SwissProt.2007.04.19 (264492/264492 entries searched); precursor ion tolerance, 10 ppm; fragment tolerance, 0.8 Da; fixed modifications, carbamidomethyl Cys; variable modifications, acetylation of protein N termini; Met oxidation; and pyroglutamic acid formation from N-terminal Gln. Acceptance criteria: maximal expectation value, 0.05; minimal peptide score, 15.

Statistical Analyses. After statistical consultation, data were analyzed for statistically significant differences between control and treated rat hepatocyte cultures by the Student's t test at the 5% level of significance (Rothman, 1990) rather than adjusted for multiple comparisons. As discussed previously (Han et al., 2005), the latter methods are controversial (Rothman, 1990; Bacchetti, 2002) because they require that the results of each analysis detract from the other. That is, each analysis is considered to be less believable than if it had been the only analysis. Because there are really only two primary comparisons of interest after NMPP-induced heme depletion [DEX versus DEX + NMPP ± heme], the extent of the multiplicity is relatively small.

	Results

Top Abstract Materials and Methods Results Discussion References

 
Effect of Heme on DEX-Inducible Endogenous TDO mRNA and Protein Expression. We examined the role of heme in DEX-inducible endogenous TDO mRNA and protein expression by qRT-PCR (TaqMan) and immunoblotting analyses, respectively, in primary hepatocytes cultured on type I collagen substratum with a Matrigel overlay that preserves many transcriptional functions, such as phenobarbital-induction of CYP2B (Han et al., 2005). Because of DEX inclusion at plating, the basal levels of endogenous TDO mRNA expression were high and further increased 3 to 4-fold by DEX (5 µM) induction. No significant effect on this DEX-inducible TDO mRNA expression was observed after DDEP-elicited heme depletion, nor was this mRNA expression affected by heme resupplementation (Fig. 1A). As in intact rats (Ren and Correia, 2000), immunoblotting analyses revealed the expected DEX-inducible increase of endogenous TDO protein that was significantly reduced by DDEP-elicited heme deficiency and restored to normal in these cells by heme resupplementation (Fig. 1B).

View larger version (27K): [in this window] [in a new window]   Fig. 1. Effects of DDEP-induced heme deficiency on endogenous TDO mRNA and protein expression. Hepatocytes were plated on collagen I with 0.1 µM DEX, overlaid with Matrigel 2 h later, and allowed to recover for 72 h with daily change of medium. At 72 h, cells were treated with DDEP (10 µM) to deplete heme, and at 84 h, DEX (5 µM) was added along with DMSO (vehicle) and DDEP (10 µM). Values for DDEP + hemin were obtained from cells treated with hemin (20 µM) 1 h after DDEP. All cells were harvested at 90 h. A, total RNA was analyzed by qRT-PCR (TaqMan) analyses with a TDO-specific probe and GUS gene as the housekeeping control. Values are mean ± S.D. of three individual cell cultures. Asterisks indicate statistically significant differences between the two indicated values at p < 0.05. NS, nonstatistically significant differences between the two indicated values. B, lysate protein was subjected to TDO-immunoblotting analyses (Ren and Correia, 2000). A prototype immunoblot is shown with actin analyzed in parallel as a loading control. For experimental details see Materials and Methods.

View larger version (16K): [in this window] [in a new window]   Fig. 2. Effects of NMPP-induced heme depletion on endogenous TDO mRNA and protein expression. Hepatocytes were plated on collagen I with 0.1 µM DEX, overlaid with Matrigel 2 h later, and allowed to recover for 72 h with daily change of medium in the absence of DEX. At 72 h, cells were treated with NMPP (1 µM) to inhibit heme synthesis. At 73 h, DMSO (vehicle) or hemin (20 µM) was added. DEX (5 µM) was added 6 h before cell harvesting at 90 h. A, total RNA was analyzed by qRT-PCR (TaqMan) analyses with a TDO gene and GUS gene (as the internal control)-specific primers and probes. TDO mRNA values are expressed relative to control (untreated cell) value. B, lysate protein (10 µg) from each cell culture was subjected to TDO-immunoblotting analyses as detailed and densitometrically quantified (Ren and Correia, 2000). Parallel actin immunoblots are included as loading controls. For experimental details see Materials and Methods. Values are mean ± S.D. of three individual cell cultures. Asterisks indicate statistically significant differences between the two indicated values at p < 0.001. NS, nonstatistically significant differences between the two indicated values.

Role of Heme on de Novo Synthesis of Total Hepatic and TDO-Specific Protein. To determine whether this impairment of TDO protein stemmed from impaired translation, cell cultures were labeled for 1 h with [³⁵S]Met/Cys (50 µCi/ml) in the absence of cold Met/Cys in the medium and then chased with nonradioactive Met/Cys-containing medium before cell harvest (see Materials and Methods). After ascertaining that [³⁵S]Met/Cys uptake was equivalent in these cell cultures, the rate of [³⁵S]Met/Cys incorporation into total cell lysate protein (counts per minute/milligram protein/hour) was determined as detailed under Materials and Methods. Cell lysate protein was also immunoprecipitated with TDO-specific polyclonal IgGs to monitor the specific rate of [³⁵S]Met/Cys incorporation into TDO protein. As illustrated (Fig. 3, solid bars), DEX increased the specific rate of [³⁵S]Met/Cys incorporation into TDO protein >15-fold. This was significantly attenuated after NMPP-induced heme depletion (Fig. 3). Heme resupplementation of DEX/NMPP-treated cultures restored this rate nearly to DEX-inducible levels. In parallel, although DEX per se had no appreciable effect, cotreatment with NMPP significantly reduced the de novo synthesis of hepatic protein (Fig. 3, cross-hatched bars) to 3.6 ± 0.7% of control (untreated) levels, whereas heme resupplementation of these NMPP/DEX-pretreated cultures restored the de novo synthesis of total hepatic protein to 102.8 ± 10% (Fig. 3). These findings thus indicate that hepatic heme depletion impairs both the de novo synthesis of TDO-specific and total hepatic protein. These effects were reversed by heme, thereby revealing a heme-dependent posttranscriptional control of several hepatic proteins, including TDO.

View larger version (24K): [in this window] [in a new window]   Fig. 3. Effect of NMPP-induced heme depletion and heme resupplementation on total protein and DEX-induced TDO protein syntheses in primary rat hepatocyte cultures. For experimental details, see Fig. 2. Cells were pulse-chased with [35S]Met/Cys (50 µCi/ml) 1 h before harvesting at 12 h of DEX-treatment. 35S incorporation (counts per minute/milligram protein/hour) into total cellular protein (cross-hatched bars) or TDO immunoprecipitates (solid bars) was determined as described under Materials and Methods. Values (mean ± S.D., n = 3) are expressed as the percentage of the corresponding DEX value. Statistically significant differences between the two indicated values are as follows: *, p < 0.001; **, p < 0.0001; ***, p < 0.0002. NS, nonstatistically significant differences between the two indicated values.

Effect of NMPP on eIF2 Phosphorylation. Parallel immunoblotting analyses of phosphorylated eIF2 and total eIF2 in corresponding cell lysates revealed that DEX treatment per se enhanced basal eIF2 phosphorylation slightly, albeit not statistically significantly, whereas NMPP treatment enhanced it 5-fold (Fig. 4), whereas the combined NMPP/DEX treatment resulted in an additive effect that was largely attenuated by heme resupplementation to the levels of eIF2 phosphorylation observed with DEX treatment alone (Fig. 4). As expected, these treatments failed to affect total hepatic eIF2 or actin (loading control) content (Fig. 4). These findings thus indicate that the impairment of de novo hepatic TDO and total protein syntheses most probably results from translational control by a hepatic heme-sensitive eIF2 kinase.

View larger version (39K): [in this window] [in a new window]   Fig. 4. Effect of NMPP-induced heme depletion and heme resupplementation on hepatic eIF2 phosphorylation. Aliquots of lysate protein (10 µg) from each cell culture were subjected to immunoblotting analyses (A) as detailed and densitometrically quantified (B) as described by Han et al. (2005). Actin was used as the loading control. For experimental details, see Materials and Methods and Fig. 2. Values are mean ± S.D. of three individual cell cultures. Asterisks indicate statistically significant differences between the two indicated values at p < 0.005. NS, nonstatistically significant differences between the two indicated values.

Heterologous Expression, Purification, Identification, and Functional Characterization of Rat Hepatic HRI. The N-terminally (His)₆-tagged HRI was expressed in E. coli BL21 cells and purified on a Ni²⁺-NTA-agarose column. SDS-PAGE analysis of the purified protein eluting around 275 mM imidazole revealed 95% purity, with one predominant band of molecular mass of 76 kDa (Fig. 5). This purified recombinant HRI was subjected to SDS-PAGE and in situ tryptic digestion followed by LC-MS/MS analyses and database mining. These analyses revealed the presence of 34 of the 44 theoretical tryptic peptides, including the C-terminal 26 residues of rat brain eIF2 kinase (sequence Q642C7/Q63185) (Mellor et al., 1994) in the tryptic digest of this protein, thereby confirming its identity as rat liver HRI.

View larger version (63K): [in this window] [in a new window]   Fig. 5. Expression and purification of recombinant rat liver HRI. (His)6HRI was expressed in E. coli BL21-RIL at 15°C overnight (18 h). Cell lysates were pooled, and 90,000 g supernatants subjected to SDS-PAGE before and after Ni2+-NTA-agarose/fast-performance liquid chromatography purification. M, molecular weight markers; C, control uninduced lysates; I, isopropyl -D-thiogalactoside-induced lysates; Ft, flow-through; Ni, Ni2+-NTA agarose-purified HRI.

The purified recombinant protein was functionally authenticated by its in vitro eIF2 kinase activity and its hemin sensitivity. SDS-PAGE followed by fluorography of the eIF2 kinase reactions in the presence of 0 to 5 µM hemin revealed a concentration-dependent reduction of phosphorylated eIF2 band at 36 kDa (Fig. 6A), with an IC₅₀ of 0.65 µM, consistent with its well recognized hemin sensitivity (Fig. 6B). Furthermore, as expected, a parallel hemin-mediated inhibition of HRI autophosphorylation was observed during this process, as determined by the progressive concentration-dependent reduction of phosphorylated 76-kDa band. Together, these findings document that the purified recombinant protein is a bona fide heme-sensitive eIF2 kinase.

View larger version (30K): [in this window] [in a new window]   Fig. 6. Functional characterization of the recombinant rat liver protein as a hemin-sensitive eIF2 kinase. A, SDS-PAGE and fluorography of aliquots (20 µl) from reaction mixtures incubated with the recombinant protein and eIF2 in the presence (0–5 µM) or absence of hemin as described under Materials and Methods. B, a plot of the concentration-dependent hemin inhibition of these reactions from three individual experiments (values are mean ± S.D.).

View larger version (25K): [in this window] [in a new window]   Fig. 7. Effects of NMPP-induced heme depletion on relative hepatic HRI mRNA and protein content. Hepatocytes were cultured and treated as described in Fig. 2. A, HRI mRNA quantitation by qRT-PCR analyses of total RNA isolated from each treated culture. Values are mean ± S.D. from three separate experiments. B, corresponding immunoblotting analyses of liver cell lysates (30 µg) with rabbit polyclonal anti-HRI antibodies raised against the purified recombinant intact rat liver HRI protein. Positive and negative controls refer to mouse reticulocyte HRI (+/+) and (–/–) lysates (0.5 µl), respectively, kindly provided by Dr. J.-J. Chen. The upper band (92 kDa) in the positive control lane seems to correspond to the hyperphosphorylated eHRI. Values are mean ± S.D. from three separate experiments. C, corresponding immunoblots of mouse reticulocyte HRI (+/+) and (–/–) lysates (0.5 µl) and rat hepatocyte lysates using purified rabbit preimmune IgGs.

Immunoaffinity Capture of HRI Protein from Rat Hepatocytes. To unequivocally establish its presence, rat hepatic HRI was isolated by immunoaffinity chromatography of lysates from size-fractionated rat hepatocytes as described under Materials and Methods). The hepatic lysate along with its flow-through fraction and the E1, E2, E3, and E4 eluates from the anti-HRI IgG-cross-linked Protein A-Sepharose beads were collected and subjected to SDS-PAGE analyses on 7.5% gels as well as immunoblotting analyses with rabbit polyclonal anti-HRI IgGs as described above (Fig. 8). SDS-PAGE analysis of the eluates revealed that E3, eluted with 0.1 M triethylamine (pH 11.5), contained the major fraction of a hepatic protein that electrophoretically migrated around 76 kDa (Fig. 8A). Considerably lesser amounts of this protein were detected in E1 and E2 eluates (Fig. 8A), and none was detected in the E4 (data not shown), thereby suggesting that most of the bound protein was eluted with the first round of 0.1 M triethylamine (pH 11.5). That this 76-kDa hepatic protein was indeed HRI was established not only by parallel immunoblotting analyses (Fig. 8B) but also by proteomic analyses of the 76-kDa protein band. Thus, consistent with the presence of HRI in the rat liver, the initial lysates of rat hepatocytes contained detectable HRI protein that was absent in the corresponding flow-through of the anti-HRI IgG-cross-linked Protein A-Sepharose column after initial passage of the lysate. That the HRI protein had indeed been sequestered from the lysate was verified by immunoblotting analyses of the four eluates that documented that HRI was sequentially eluted with the major fraction present in E3 and none in E4. More importantly, LC-MS/MS analyses of the in situ trypsin-digested 76-kDa protein band after SDS-PAGE of the E3 eluate (Fig. 8A) indicated the unambiguous presence of rat hepatic HRI protein, confirmed from 30 tryptic peptides with 48% sequence coverage (Q63185 [GenBank] /Q642C7; see supplementary data), versus 34 tryptic peptides with a corresponding 54% sequence coverage for the recombinant hepatic HRI analyzed as the control. The detection of IgG-derived peptides as minor contaminants in the tryptic digests of the 76-kDa protein after proteomic analyses indicated that the highly alkaline elution buffers also leached out some cross-linked anti-HRI IgG, clearly detectable as a 55-kDa protein band after SDS-PAGE. Collectively, these findings unambiguously establish the existence of a rat hepatic HRI.

View larger version (57K): [in this window] [in a new window]   Fig. 8. SDS-PAGE and immunoblotting analyses of immunoaffinity-isolated HRI from freshly isolated, size-fractionated rat hepatocytes. A, SDS-PAGE analyses with colloidal Coomassie Brilliant Blue staining of the initial lysates (RHL, 60 µg) from size-fractionated 120 x 106 rat hepatocytes; the corresponding FT fraction (60 µg) and eluates E1, E2, and E3 (15 µg) are shown along with the purified recombinant HRI protein (2 µg) and molecular weight (MW) standards. E3* eluate from a repeat experiment using lysates from size-fractionated 45 x 106 hepatocytes is also included. For experimental details, see Materials and Methods. Proteomic analyses of the E3 lane reveal that the protein band at 76 kDa predominantly contains the HRI protein with minor IgG contaminants, whereas the ones at 150 and 185 kDa contain some HRI as well as rabbit IgG leached from the immunoaffinity matrix. B, immunoblotting analyses of RHL (80 µg), FT (80 µg), and E1, E2, E3, and E4 fractions (6 µg each) along with recombinant HRI (0.4 µg) as described under Materials and Methods.

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
To determine whether acute hepatic heme depletion impairs DEX-mediated TDO induction through defective mRNA transcription (Ren and Correia, 2000), we examined this process in a more defined and controllable system: primary rat hepatocytes cultured in a collagen-Matrigel "sandwich". This system, expected to more competently preserve transcription factors, revealed no specific effects of heme depletion and/or repletion on DEX-mediated transcriptional activation of the endogenous TDO mRNA. These findings in cultured hepatocytes thus differ from those in intact rats² (Ren and Correia, 2000). Nevertheless, in both models, hepatic heme depletion definitely reduced DEX-inducible hepatic TDO protein content, an effect abrogated by heme resupplementation. More importantly, our findings in cultured hepatocytes indicate that heme regulates TDO gene expression not transcriptionally but rather post-transcriptionally. Such posttranscriptional regulation could entail translational control as well as enhanced proteolytic degradation of the heme-denuded TDO apoprotein as discussed previously (Ren and Correia, 2000). Indeed, our findings in heme-depleted hepatocytes that the de novo syntheses of TDO and total protein was blocked and that this blockade was associated with enhanced eIF2 phosphorylation in a heme-reversible manner implicated a heme-sensitive hepatic eIF2 kinase in this translational regulation.

Such eIF2 kinase-mediated translational control is exerted via Ser51 phosphorylation of the -subunit of the eIF2. This in turn results in the sequestration of eIF2B, the exchange factor required for eIF2 recycling between the inactive GDP-bound species (released after GTP hydrolysis before protein chain elongation) and the active GTP-bound species critical for each fresh cycle of translational initiation. To date, four different eIF2 kinases have been identified: i) eHRI (EIF2AK1), the erythroid heme-regulated inhibitor, activated by heme deficiency and oxidative and heat stresses in reticulocytes and fetal liver erythroid progenitors (Chen, 2006); ii) PKR (RNA-dependent protein kinase), an interferon-inducible antiviral defense mechanism (EIF2AK2) (Wek et al., 2006); iii) pancreatic eIF2 kinase or PERK [PKR-like endoplasmic reticulum (ER)-bound eIF2-kinase (EIF2AK3)] (Harding et al., 2003; Kaufman, 2004) induced by ER-stress and unfolded protein response (UPR); and iv) GCN2 (general control nonderepressible-2 or EIF2AK4) induced by amino acid deprivation and non-nutritional stresses such as UV irradiation and proteasome inhibition (Kimball and Jefferson, 2004; Wek et al., 2006). Of these, GCN2 and PKR may be excluded a priori because they neither contain "heme-regulatory motifs" nor are they "heme-sensitive". Their exclusion is also justified because the cultured hepatocytes were adequately supplied with essential amino acids and thus were neither "amino acid-starved" nor exposed to any identifiable interferon-inducing agents. We (Han et al., 2005) have previously excluded PERK, the ER eIF2 kinase activated by various ER-stresses such as UPR, protein misfolding, and hypoxia on grounds of no detectably increased cellular content above basal levels of Bip (Grp78), the intraluminal ER chaperone, and an accepted marker of ER stress and PERK activation. Furthermore, the likelihood of NMPP-induced cellular ER-stress activating PERK was excluded by another UPR functional marker, XBP-1 (X-box binding protein 1) mRNA splicing by IRE-1 (inositol requiring)/ERN1 (ER to nucleus signaling 1) pathway. IRE-1 is a highly conserved, ER-stress-activated integral ER endonuclease, which upon activation, splices out a small 26-nucleotide intron from XBP-1 mRNA (Calfon et al., 2002) verifiable by RT-PCR and PstI restriction digest. Using this index, no differences in XBP-1 mRNA splicing were detected between untreated and NMPP-treated hepatocytes (data not shown). These combined considerations and the specific activation of this hepatic eIF2 kinase on heme depletion and its marked attenuation by heme resupplementation implicate a hepatic HRI-like eIF2 kinase in this process. Indeed, this possibility was greatly strengthened by our cloning of a HRI cDNA from heme-depleted rat liver and cultured size-elutriated hepatocytes identical to the previously cloned rat brain HRI cDNA (Mellor et al., 1994). Furthermore, that this heme-sensitive eIF2 kinase is truly hepatic and unrelated to the fetal liver, eHRI (Lu et al., 2001) was established by size-fractionation (30 µ) of hepatocytes to exclude erythroid cells and/or progenitors. We have heterologously expressed this hepatic HRI and conclusively identified it both structurally via proteomic analyses and functionally via kinase and heme-inhibition assays as a bona fide heme-sensitive rat liver eIF2 kinase.

However, unlike the extensively characterized "erythroid" HRI that controls globin synthesis and thus hemoglobin formation under limited heme availability (Chen and London, 1995; Chen, 2006), relatively little is known about the "nonerythroid" HRIs, fueling the notion that HRI is indeed erythroid-specific (Chen and London, 1995). Northern blot analyses of rabbit liver RNA failed to detect any HRI-specific mRNA, in contrast to the robust expression detected in rabbit bone marrow, peripheral blood, and reticulocytes (Crosby et al., 1994). Furthermore, immunoblotting analyses with anti-rabbit reticulocyte eHRI monoclonal antibodies or polyclonal antibodies failed to reveal detectable cross-immunoreactivity in rabbit and mouse liver or tissues other than reticulocytes and anemic bone marrow (Crosby et al., 1994). A mouse liver ortholog of eHRI (mHRI), with similar hemin-sensitive autokinase and eIF2Ser51 kinase activities, has been isolated (Berlanga et al., 1998).³ Northern and qRT-PCR analyses indicated that mHRI is expressed ubiquitously in mouse, with the highest expression detected in liver, kidney and testis (Berlanga et al., 1998). A recombinant mHRI cloned from a C57BL/6 mouse liver cDNA library has also been recently structurally and functionally characterized in vitro (Miksanova et al., 2006).

Because, to our knowledge, the references cited above are the only reports of mammalian liver heme-sensitive HRI, it has remained somewhat elusive with an as yet undefined physiological role. One apparent reason is that not only is HRI mRNA relatively less abundant in hepatocytes than in erythroid cells but also that the HRI protein is apparently immunochemically undetectable in mouse or rabbit liver (J.-J. Chen, personal communication). Furthermore, our attempts to identify it by immunoprecipitation of rat hepatocyte lysates with polyclonal antibodies against the mouse eHRI N-terminal 138 residues were unsuccessful (A.-P. Han, J.-J. Chen, X.-M. Han, and M. A. Correia, preliminary findings). However, our qRT-PCR analyses and immunoblotting analyses using rabbit polyclonal antibodies against the purified recombinant intact rat hepatic HRI clearly revealed its presence in control and heme-depleted rat hepatocytes (Fig. 7). Furthermore, proteomic and immunoblotting analyses of the immunoaffinity-captured protein from lysates of size-fractionated rat hepatocytes unequivocally confirmed its identity as HRI (Fig. 8). Collectively, our findings argue that this hepatic heme-sensitive eIF2 kinase is physiologically and/or pathologically relevant, as it is functionally activated in heme-depleted rat hepatocytes (Fig. 4) (Han et al., 2005). Such heme depletion stems from a profound (75%) inhibition of hepatic heme synthesis due to ferrochelatase inhibition coupled with DDEP-induced prosthetic heme destruction of several P450s (Sugiyama et al., 1989). We propose that such activation of this heme-sensitive eIF2 kinase (HRI) in heme-depleted rat hepatocytes is responsible for the eIF2 phosphorylation-mediated shutdown of translational initiation and the consequent global suppression of de novo syntheses of total hepatic protein, including TDO (discussed above) and CYP2B1 and 2B2 (Han et al., 2005). Accordingly, heme supplementation blocked its activity and normalized the de novo syntheses of hepatic protein and TDO, thereby revealing its heme reversibility.

Acute hepatic heme deficiency is a hallmark of life-threatening attacks of the genetic diseases clinically known as the hepatic porphyrias. These attacks are signaled by early biochemical events, such as the derepression of the rate-limiting heme synthetic enzyme, -aminolevulinic acid synthase (-ALAS), followed by acute ill-defined abdominal and neurological symptoms (Anderson et al., 2005). In genetically predisposed individuals, such acute attacks of hepatic porphyria are triggered when the extent of their hepatic heme depletion is severe, requiring intravenous hemin administration as the primary and definitive treatment (Anderson et al., 2005). This life-saving measure restores intrahepatic heme levels, thereby dramatically reversing all of the biochemical and clinical symptoms. Indeed, hepatic heme depletion needs to be quite severe as in our model, even for the early biochemical manifestations of acute hepatic porphyria. Thus, mice with a targeted genetic lesion in the heme synthetic enzyme porphobilinogen deaminase that reduces their heme synthetic activity by 55%⁴ (Jover et al., 2000) exhibit both normal hemoprotein (P450) turnover and no hallmark biochemical signs, such as the relief of -ALAS negative feedback inhibition by hepatic heme depletion (Granick et al., 1975). Indeed, -ALAS induction in these porphobilinogen deaminase-deficient mice occurs only after repeated challenge by P450 inducers, such as phenobarbital, that by inducing P450 protein elicits an increased demand for hepatic heme biosynthesis, thereby further depleting the "free heme pool" and thus exacerbating hepatic heme deficiency (De Matteis, 1978; Jover et al., 2000).

The neurological symptoms observed during clinical attacks apparently ensue from disruption of signaling pathways in acutely heme-deficient neuronal cells (Mense and Zhang, 2006). Acute hepatic heme depletion reduces TDO function, impairs L-Trp catabolism, and alters the serotonergic tone in the central nervous system, thereby also possibly contributing to these neurological symptoms (Litman and Correia, 1983, 1985). Our current findings suggest that acute hepatic heme deficiency through activation of a heme-sensitive eIF2 kinase can shut off the synthesis of myriad intracellular proteins, including hemoproteins. Similar translational suppression of other short-lived hepatic proteins could similarly affect various vital homeostatic functions and/or alter the flux of additional neuroactive amino acids, thereby also contributing to the panoply of clinical symptoms observed in these acute heme-deficient states.

	Acknowledgements

 
We thank Dr. J.-J. Chen (Massachusetts Institute of Technology, Boston, MA) for the valuable discussions on HRI/eIF2 kinase methodology, reticulocyte HRI (+/+) and (–/–) lysates, and preliminary studies of HRI immunoprecipitation. We also thank Prof. Peter Bacchetti [University of California, San Francisco (UCSF), San Francisco, CA] for valuable advice on appropriate statistical methodology and Gene Lee (UCSF Liver Center Cell and Tissue Biology Core Facility, San Francisco, CA) for providing isolated hepatocytes used in the preliminary studies. We thank Dr. Ping Kang for assistance with the GraphPad plot and are grateful to Dr. A. L. Burlingame for the use of the UCSF Mass Spectrometry Facility supported by the Biomedical Research Technology Program of the National Center for Research Resources. We also acknowledge the many literature contributions that regretfully could not be credited due to the stipulated limit on the number of citations.

	Footnotes

 
This work was supported by National Institutes of Health (NIH) Grants DK26506 (to M.A.C.) and DK61510 (to J.J.M.). We also acknowledge the UCSF Liver Center Cores on Molecular Analyses (Spectrophotometry and Mass Spectrometry) and on Cell and Tissue Biology supported by NIH NIDDK Center Grant P30DK26743. The UCSF Mass Spectrometry Facility is supported by NIH National Center for Research Resources BRTP 01614, RR019934, and RR015804.

Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.

doi:10.1124/jpet.107.124602.

ABBREVIATIONS: TDO, tryptophan 2,3-dioxygenase; eIF2, -subunit of the eukaryotic initiation factor eIF2; eIF2P, phosphorylated eIF2; DTT, dithiothreitol; CID, collision-induced dissociation; DDEP, 3,5-dicarbethoxy-2,6-dimethyl-4-ethyl-1,4-dihydropyridine; DEX, dexamethasone; ER, endoplasmic reticulum; eHRI, erythroid HRI; GCN2, general control nonderepressible-2; HRI, heme-regulated inhibitor; IRE-1/ERN1, inositol requiring ER to nucleus signaling 1; mHRI, mouse HRI; NMPP, N-methylprotoporphyrin IX; PERK, PKR-like ER-bound eIF2-kinase; WME, Williams' Medium E; PKR, RNA-dependent protein kinase; RHL, rat hepatocyte lysate; qRT-PCR, quantitative reverse transcription-polymerase chain reaction; UPR, unfolded protein response; XBP-1, X-box binding protein 1; E-64, N-(trans-epoxysuccinyl)-L-leucine 4-guanidinobutylamide; DMSO, dimethyl sulfoxide; GUS, -glucuronidase; PBS, phosphate-buffered saline; PAGE, polyacrylamide gel electrophoresis; LC, liquid chromatography; MS, mass spectrometry; FT, flow-through; DOC, deoxycholate; -ALAS, -aminolevulinic acid synthase; HRP, horseradish peroxidase.

The online version of this article (available at http://jpet.aspetjournals.org) contains supplemental material.

¹ The terms heme and hemin for iron-protoporphyrin IX is used interchangeably throughout the text.

² It is conceivable that, although cultured hepatocytes are adequately supplied with nutrients and O₂, rats with acute hepatic heme depletion being neurologically and/or otherwise physiologically impaired are incapable of maintaining normal hepatocellular gene transcription.

³ An eHRI-like inhibitor was isolated from rat liver perfused free of contaminating blood and erythroid cells (Delaunay et al., 1977), but apparently its identity has remained equivocal.

⁴ As monitored by ¹⁴C--ALA-incorporation into heme (Jover et al., 2000).

Address correspondence to: M. A. Correia, Dept. of Cellular and Molecular Pharmacology, Mission Bay Campus, Genentech Hall, 600 16th Street, N572F/Box 2280, University of California, San Francisco, CA 94158. E-mail: almira.correia{at}ucsf.edu

	References

Top Abstract Materials and Methods Results Discussion References

 

Anderson KE, Bloomer JR, Bonkovsky HL, Kushner JP, Pierach CA, Pimstone NR, and Desnick RJ (2005) Recommendations for the diagnosis and treatment of the acute porphyrias. Ann Intern Med 142: 439–450.[Abstract/Free Full Text]
Bacchetti P (2002) Peer review of statistics in medical research: the other problem. Br Med J 324: 1271–1273.[Free Full Text]
Berlanga JJ, Herrero S, and de Haro C (1998) Characterization of the hemin-sensitive eukaryotic initiation factor 2alpha kinase from mouse nonerythroid cells. J Biol Chem 273: 32340–32346.[Abstract/Free Full Text]
Bissell DM and Hammaker LE (1977) Effect of endotoxin on tryptophan pyrrolase and delta-aminolaevulinate synthase: evidence for an endogenous regulatory haem fraction in rat liver. Biochem J 166: 301–304.[Medline]
Calfon M, Zeng H, Urano F, Till JH, Hubbard SR, Harding HP, Clark SG, and Ron D (2002) IRE1 couples endoplasmic reticulum load to secretory capacity by processing the XBP-1 mRNA. Nature 415: 92–96.[CrossRef][Medline]
Chefalo PJ, Oh J, Rafie-Kolpin M, Kan B, and Chen JJ (1998) Heme-regulated eIF-2alpha kinase purifies as a hemoprotein. Eur J Biochem 258: 820–830.[Medline]
Chen JJ, Yang JM, Petryshyn R, Kosower N, and London IM (1989) Disulfide bond formation in the regulation of eIF-2 alpha kinase by heme. J Biol Chem 264: 9559–9564.[Abstract/Free Full Text]
Chen JJ and London IM (1995) Regulation of protein synthesis by heme-regulated eIF-2 alpha kinase. Trends Biochem Sci 20: 105–108.[CrossRef][Medline]
Chen JJ, Throop MS, Gehrke L, Kuo I, Pal JK, Brodsky M, and London IM (1991) Cloning of the cDNA of the heme-regulated eukaryotic initiation factor 2 alpha (eIF-2 alpha) kinase of rabbit reticulocytes: homology to yeast GCN2 protein kinase and human double-stranded-RNA-dependent eIF-2 alpha kinase. Proc Natl Acad SciUSA 88: 7729–7733.[Abstract/Free Full Text]
Chen JJ (2006) Regulation of protein synthesis by the heme-regulated eIF2{alpha} kinase: relevance to anemias. Blood 109: 2693–2699.
Crosby JS, Lee K, London IM, and Chen JJ (1994) Erythroid expression of the heme-regulated eIF-2 alpha kinase. Mol Cell Biol 14: 3906–3914.[Abstract/Free Full Text]
Danesch U, Gloss B, Schmid W, Schutz G, Schule R, and Renkawitz R (1987) Glucocorticoid induction of the rat tryptophan oxygenase gene is mediated by two widely separated glucocorticoid-responsive elements. EMBO J 6: 625–630.[Medline]
Delaunay J, Ranu RS, Levin DH, Ernst V, and London IM (1977) Characterization of a rat liver factor that inhibits initiation of protein synthesis in rabbit reticulocyte lysates. Proc Natl Acad SciUSA 74: 2264–2268.[Abstract/Free Full Text]
De Matteis F (1978) Hepatic porphyrias, in heme and hemoproteins. Handbook Exp Pharmacol 44: 129–155.
De Matteis F, Gibbs AH, and Smith AG (1980) Inhibition of protohaem ferro-lyase by N-substituted porphyrins. Structural requirements for the inhibitory effect. Biochem J 189: 645–648.[Medline]
Granick S, Sinclair P, Sassa S, and Grieninger G (1975) Effects by heme, insulin, and serum albumin on heme and protein synthesis in chick embryo liver cells cultured in a chemically defined medium, and a spectrofluorometric assay for porphyrin composition. J Biol Chem 250: 9215–9225.[Abstract/Free Full Text]
Greengard O and Feigelson P (1961) The activation and induction of rat liver tryptophan pyrrolase in vivo by its substrate. J Biol Chem 236: 158–161.[Free Full Text]
Han XM, Lee G, Hefner C, Maher JJ, and Correia MA (2005) Heme-reversible impairment of CYP2B1/2 induction in heme-depleted rat hepatocytes in primary culture: translational control by a hepatic alpha-subunit of the eukaryotic initiation factor kinase? J Pharmacol Exp Ther 314: 128–138.[Abstract/Free Full Text]
Harding HP, Zhang Y, Zeng H, Novoa I, Lu PD, Calfon M, Sadri N, Yun C, Popko B, Paules R, et al. (2003) An integrated stress response regulates amino acid metabolism and resistance to oxidative stress. Mol Cell 11: 619–633.[CrossRef][Medline]
Jover R, Hoffmann F, Scheffler-Koch V, and Lindberg RL (2000) Limited heme synthesis in porphobilinogen deaminase-deficient mice impairs transcriptional activation of specific cytochrome P450 genes by phenobarbital. Eur J Biochem 267: 7128–7137.[Medline]
Kaufman RJ (2004) Regulation of mRNA translation by protein folding in the endoplasmic reticulum. Trends Biochem Sci 29: 152–158.[CrossRef][Medline]
Kimball SR and Jefferson LS (2004) Amino acids as regulators of gene expression. Nutr Metab (Lond) 1: 3.[CrossRef][Medline]
Knox WE (1966) The regulation of tryptophan pyrrolase activity by tryptophan. Adv Enzyme Regul 4: 287–297.[CrossRef][Medline]
Litman DA and Correia MA (1983) L-Tryptophan: a common denominator of biochemical and neurological events of acute hepatic porphyria? Science 222: 1031–1033.[Abstract/Free Full Text]
Litman DA and Correia MA (1985) Elevated brain tryptophan and enhanced 5-hydroxytryptamine turnover in acute hepatic heme deficiency: clinical implications. J Pharmacol Exp Ther 232: 337–345.[Abstract/Free Full Text]
Lu L, Han AP, and Chen JJ (2001) Translation initiation control by heme-regulated eukaryotic initiation factor 2alpha kinase in erythroid cells under cytoplasmic stresses. Mol Cell Biol 21: 7971–7980.[Abstract/Free Full Text]
Marver HS, Tschudy DP, Perlroth MG, and Collins A (1966) Coordinate synthesis of heme and apoenzyme in the formation of tryptophan pyrrolase. Science 154: 501–503.[Abstract/Free Full Text]
Mellor H, Flowers KM, Kimball SR, and Jefferson LS (1994) Cloning and characterization of cDNA encoding rat hemin-sensitive initiation factor-2 alpha (eIF-2 alpha) kinase. Evidence for multitissue expression. J Biol Chem 269: 10201–10204.[Abstract/Free Full Text]
Mense SM and Zhang L (2006) Heme: a versatile signaling molecule controlling the activities of diverse regulators ranging from transcription factors to MAP kinases. Cell Res 16: 681–692.[CrossRef][Medline]
Miksanova M, Igarashi J, Minami M, Sagami I, Yamauchi S, Kurokawa H, and Shimizu T (2006) Characterization of heme-regulated eIF2alpha kinase: roles of the N-terminal domain in the oligomeric state, heme binding, catalysis, and inhibition. Biochemistry 45: 9894–9905.[CrossRef][Medline]
Nakamura T, Niimi S, Nawa K, Noda C, Ichihara A, Takagi Y, Anai M, and Sakaki Y (1987) Multihormonal regulation of transcription of the tryptophan 2,3-dioxygenase gene in primary cultures of adult rat hepatocytes with special reference to the presence of a transcriptional protein mediating the action of glucocorticoids. J Biol Chem 262: 727–733.[Abstract/Free Full Text]
Ortiz de Montellano PR, Kunze KL, Cole SP, and Marks GS (1980) Inhibition of hepatic ferrochelatase by the four isomers of N-methylprotoporphyrin IX. Biochem Biophys Res Commun 97: 1436–1442.[CrossRef][Medline]
Ortiz de Montellano PR, Kunze KL, Cole SP, and Marks GS (1981) Differential inhibition of hepatic ferrochelatase by the isomers of N-ethylprotoporphyrin IX. Biochem Biophys Res Commun 103: 581–586.[CrossRef][Medline]
Pal JK, Chen JJ, and London IM (1991) Tissue distribution and immunoreactivity of heme-regulated eIF-2 alpha kinase determined by monoclonal antibodies. Biochemistry 30: 2555–2562.[CrossRef][Medline]
Ren S and Correia MA (2000) Heme: a regulator of rat hepatic tryptophan 2,3-dioxygenase? Arch Biochem Biophys 377: 195–203.[CrossRef][Medline]
Rothman KJ (1990) No adjustments are needed for multiple comparisons. Epidemiology 1: 43–46.[Medline]
Schimke RT, Sweeney EW, and Berlin CM (1965) The roles of synthesis and degradation in the control of rat liver tryptophan pyrrolase. J Biol Chem 240: 322–331.[Free Full Text]
Schneider C, Newman RA, Sutherland DR, Asser U, and Greaves MF (1982) A one-step purification of membrane proteins using a high efficiency immunomatrix. J Biol Chem 257: 10766–10769.[Abstract/Free Full Text]
Schutz G and Feigelson P (1972) Purification and properties of rat liver tryptophan oxygenase. J Biol Chem 247: 5327–5332.[Abstract/Free Full Text]
Sugiyama K, Yao K, Rettie AE, and Correia MA (1989) Inactivation of rat hepatic cytochrome P-450 isozymes by 3,5-dicarbethoxy-2,6-dimethyl-4-ethyl-1,4-dihydropyridine. Chem Res Toxicol 2: 400–410.[CrossRef][Medline]
Wek RC, Jiang HY, and Anthony TG (2006) Coping with stress: eIF2 kinases and translational control. Biochem Soc Trans 34: 7–11.[Medline]

This article has been cited by other articles:

				P. Acharya, J. C. Engel, and M. A. Correia Hepatic CYP3A Suppression by High Concentrations of Proteasomal Inhibitors: A Consequence of Endoplasmic Reticulum (ER) Stress Induction, Activation of RNA-Dependent Protein Kinase-Like ER-Bound Eukaryotic Initiation Factor 2{alpha} (eIF2{alpha})-Kinase (PERK) and General Control Nonderepressible-2 eIF2{alpha} Kinase (GCN2), and Global Translational Shutoff Mol. Pharmacol., September 1, 2009; 76(3): 503 - 515. [Abstract] [Full Text] [PDF]

				J. Igarashi, M. Murase, A. Iizuka, F. Pichierri, M. Martinkova, and T. Shimizu Elucidation of the Heme Binding Site of Heme-regulated Eukaryotic Initiation Factor 2{alpha} Kinase and the Role of the Regulatory Motif in Heme Sensing by Spectroscopic and Catalytic Studies of Mutant Proteins J. Biol. Chem., July 4, 2008; 283(27): 18782 - 18791. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Data Supplement All Versions of this Article: jpet.107.124602v1 323/3/979    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Liao, M. Articles by Correia, M. A. Search for Related Content PubMed PubMed Citation Articles by Liao, M. Articles by Correia, M. A.