Iron Regulates the Intracellular Degradation of Iron Regulatory Protein 2 by the Proteasome (∗)

  • Bing Guo
    Affiliations
    Eccles Program in Human Molecular Biology and Genetics and the Department of Medicine, University of Utah, Salt Lake City, Utah 84112
    Search for articles by this author
  • John D. Phillips
    Affiliations
    Eccles Program in Human Molecular Biology and Genetics and the Department of Medicine, University of Utah, Salt Lake City, Utah 84112
    Search for articles by this author
  • Yang Yu
    Affiliations
    Eccles Program in Human Molecular Biology and Genetics and the Department of Medicine, University of Utah, Salt Lake City, Utah 84112
    Search for articles by this author
  • Elizabeth A. Leibold
    Correspondence
    To whom all correspondence should be addressed: University of Utah, Bldg. 533, Rm. 4220, HMBG, Salt Lake City, UT 84112. Tel.: 801-585-5002; Fax: 801-585-3501
    Affiliations
    Eccles Program in Human Molecular Biology and Genetics and the Department of Medicine, University of Utah, Salt Lake City, Utah 84112
    Search for articles by this author
  • Author Footnotes
    ∗ This work was supported by National Institutes of Health Grant GM45201 (to E. A. L.) and by National Cancer Institute Grant CA4201 (to the Protein Core Facility of the Utah Cancer Center at the University of Utah). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore by hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
      Iron regulatory proteins (IRP1 and IRP2) are RNA-binding proteins that bind to specific structures, termed iron-responsive elements (IREs), that are located in the 5′- or 3′-untranslated regions of mRNAs that encode proteins involved in iron homeostasis. IRP1 and IRP2 RNA binding activities are regulated by iron; IRP1 and IRP2 bind IREs with high affinity in iron-depleted cells and with low affinity in iron-repleted cells. The decrease in IRP1 RNA binding activity occurs by a switch between apoprotein and 4Fe-4S forms, without changes in IRP1 levels, whereas the decrease in IRP2 RNA binding activity reflects a reduction in IRP2 levels. To determine the mechanism by which iron decreases IRP2 levels, we studied IRP2 regulation by iron in rat hepatoma and human HeLa cells. The iron-dependent decrease in IRP2 levels was not due to a decrease in the amount of IRP2 mRNA or to a decrease in the rate of IRP2 synthesis. Pulse-chase experiments demonstrated that iron resulted in a 3-fold increase in the degradation rate of IRP2. IRP2 degradation depends on protein synthesis, but not transcription, suggesting a requirement for a labile protein. IRP2 degradation is not prevented by lysosomal inhibitors or calpain II inhibitors, but is prevented by inhibitors that block proteasome function. These data suggest the involvement of the proteasome in iron-mediated IRP2 proteolysis.

      INTRODUCTION

      Iron regulatory proteins (IRPs)
      The abbreviations used are: IRP1 and IRP2
      iron regulatory proteins 1 and 2
      IRE
      iron-responsive element
      FAC
      ferric ammonium citrate
      PAGE
      polyacrylamide gel electrophoresis
      MG-132
      carbobenzoxyl-leucinyl-leucinyl-leucinal-H.
      are cytosolic RNA-binding proteins that regulate the post-transcriptional expression of genes that are involved in iron homeostasis(
      • Leibold E.A.
      • Guo B.
      ,
      • Klausner R.D.
      • Rouault T.A.
      • Harford J.B.
      ,
      • Theil E.C.
      ,
      • Beinert H.
      • Kennedy M.C.
      ). IRPs were formerly known as the iron-responsive element-binding protein (IRE-BP), the ferritin repressor protein (FRP), and the iron regulatory factor (IRF). IRPs bind with high affinity to RNA stem-loops known as iron-responsive elements (IREs). IREs are located in the 5′- untranslated regions of ferritin and erythroid Δ-aminolevulinic acid synthase mRNAs where binding causes translational repression (
      • Walden W.E.
      • Patino M.M.
      • Gaffield L.
      ,
      • Goossen B.
      • Caughman S.W.
      • Harford J.B.
      • Klausner R.D.
      • Hentze M.W.
      ,
      • Gray N.K.
      • Hentze M.W.
      ). Five IREs are located in the 3′-untranslated region of transferrin receptor mRNA(
      • Casey J.L.
      • Hentze M.W.
      • Koeller D.M.
      • Caughman S.W.
      • Rouault T.A.
      • Klausner R.D.
      • Harford J.B.
      ,
      • Mullner E.W.
      • Neupert B.
      • Kuhn L.C.
      ), and binding of the IRP stabilizes transferrin receptor mRNA(
      • Mullner E.W.
      • Neupert B.
      • Kuhn L.C.
      ,
      • Binder R.
      • Horowitz J.A.
      • Basilion J.P.
      • Koeller D.M.
      • Klausner R.D.
      • Harford J.B.
      ).
      Two distinct IRPs have been cloned and characterized in mammalian cells and have been designated as IRP1 and IRP2. IRP1 has been cloned from a variety of mammalian species(
      • Walden W.E.
      • Patino M.M.
      • Gaffield L.
      ,
      • Rouault T.A.
      • Tang C.K.
      • Kaptain S.
      • Burgess W.H.
      • Haile D.J.
      • Samaniego F.
      • McBride O.W.
      • Harford J.B.
      • Klausner R.D.
      ,
      • Yu Y.
      • Radisky E.
      • Leibold E.A.
      ). IRP1 has a molecular mass of 98,000 Da and shares 30% amino acid identity with the 4Fe-4S enzyme, mitochondrial aconitase(
      • Rouault T.A.
      • Stout C.D.
      • Kaptain S.
      • Harford J.B.
      • Klausner R.D.
      ). The 18 active site residues in mitochondrial aconitase, including the 3 cysteines that serve as ligands for the 4Fe-4S cluster are conserved in IRP1(
      • Rouault T.A.
      • Stout C.D.
      • Kaptain S.
      • Harford J.B.
      • Klausner R.D.
      ). In addition, IRP1 is an active cytosolic aconitase(
      • Kennedy M.C.
      • Mende-Mueller L.
      • Blondin G.A.
      • Beinert H.
      ,
      • Kaptain S.
      • Downey W.E.
      • Tang C.
      • Philpott C.
      • Haile D.
      • Orloff D.G.
      • Harford J.B.
      • Rouault T.A.
      • Klausner R.D.
      ). In iron-repleted cells, IRP1 exhibits aconitase activity and contains iron, but binds the IRE with low affinity. In contrast, in iron-depleted cells, IRP1 lacks aconitase activity and iron, but binds the IRE with high affinity. UV cross-linking studies have shown overlap between RNA binding and the aconitase active sites, indicating that RNA binding and aconitase activities are mutually exclusive(
      • Basilion J.P.
      • Rouault T.A.
      • Massinople C.M.
      • Klausner R.D.
      • Burgess W.H.
      ,
      • Swenson G.R.
      • Walden W.E.
      ). Recent data indicated that aconitase activity is not necessary for iron regulation of IRP1, since substitution of an alanine for an active site serine does not prevent assembly and disassembly of the 4Fe-4S cluster(
      • Philpott C.C.
      • Klausner R.D.
      • Rouault T.A.
      ).
      IRP2 has been characterized in rat tissues by RNA band shift assays (
      • Leibold E.A.
      • Munro H.N.
      ,
      • Guo B.
      • Yu Y.
      • Leibold E.A.
      ,
      • Cairo G.
      • Pietrangelo A.
      ,
      • Cairo G.
      • Tacchini L.
      • Pogliaghi G.
      • Anzon E.
      • Tomasi A.
      • Bernelli-Zazzera A.
      ) and has been purified from rat liver and rat hepatoma cells (
      • Guo B.
      • Yu Y.
      • Leibold E.A.
      ). The partial amino acid sequence of rat IRP2 is similar to the predicted protein sequence encoded by a cDNA isolated from a human T cell library(
      • Rouault T.A.
      • Tang C.K.
      • Kaptain S.
      • Burgess W.H.
      • Haile D.J.
      • Samaniego F.
      • McBride O.W.
      • Harford J.B.
      • Klausner R.D.
      ,
      • Rouault T.A.
      • Haile D.J.
      • Downey W.E.
      • Philpott C.C.
      • Tang C.
      • Samaniego F.
      • Chin J.
      • Paul I.
      • Orloff D.
      • Harford J.B.
      • Klausner R.D.
      ), suggesting that this is the rat version of the human protein. A second IRP has been characterized from mouse tissues by RNA band shift analysis(
      • Henderson B.R.
      • Seiser C.
      • Kuhn L.C.
      ), which is presumed to be homologous to the rat and the human IRP2.
      IRP2 contains similar biochemical properties to IRP1 in that it binds IREs with similar affinity (
      • Guo B.
      • Yu Y.
      • Leibold E.A.
      ,
      • Henderson B.R.
      • Seiser C.
      • Kuhn L.C.
      ) and represses translation of IRE-containing mRNAs in vitro(
      • Guo B.
      • Yu Y.
      • Leibold E.A.
      ,
      • Kim H.
      • Klausner R.D.
      • Rouault T.A.
      ). IRP1 and IRP2 differ in two aspects: first, unlike IRP1, IRP2 does not exhibit aconitase activity, indicating that aconitase activity is not necessary for regulation by iron(
      • Guo B.
      • Yu Y.
      • Leibold E.A.
      ), and second, that although iron results in a decrease in IRP1 and IRP2 RNA-binding activities, the amount of IRP1 remains constant(
      • Yu Y.
      • Radisky E.
      • Leibold E.A.
      ,
      • Tang C.K.
      • Chin J.
      • Harford J.B.
      • Klausner R.D.
      • Rouault T.A.
      ), whereas the amount of IRP2 protein is substantially reduced (
      • Guo B.
      • Yu Y.
      • Leibold E.A.
      ,
      • Samaniego F.
      • Chin J.
      • Iwai K.
      • Rouault T.A.
      • Klausner R.D.
      These novel properties of IRP2 raised questions as to how intracellular iron regulates IRP2 RNA binding activity. To answer this question, we analyzed the regulation of IRP2 by iron in rat hepatoma cells and human HeLa cells. We found that the marked reduction in IRP2 RNA binding activity and protein levels in iron-treated cells is due to increased turnover of IRP2. IRP2 synthesis and IRP2 mRNAs levels are unaffected by iron treatment. The iron-mediated degradation of IRP2 requires protein synthesis, but not transcription, suggesting that the synthesis of a labile protein is required. We also demonstrate that the proteasome complex is required for iron-mediated degradation of IRP2. Regulation of IRP2 protein levels by iron occurs in a variety of cell types, indicating that iron-mediated degradation is a common pathway for regulating IRP2 RNA binding activity.

      EXPERIMENTAL PROCEDURES

       Materials

      The proteasome inhibitor MG-132 (carbobenzoxyl-leucinyl-leucinyl-leucinal-H) was a gift from Myogenics, Inc., Cambridge, MA. Calpain inhibitor II and actinomycin D was purchased from Sigma. Antibodies generated against rat IRP1 and IRP2 were prepared as described(
      • Guo B.
      • Yu Y.
      • Leibold E.A.
      ).

       Cell Culture

      The rat hepatoma cell line FTO2B and human HeLa cells were grown at 37°C in an 8% CO2 atmosphere in Dulbecco's modified Eagles's medium supplemented with 10% heat-inactivated fetal bovine serum. For iron and iron chelation studies, cells were treated with either 50 μg/ml ferric ammonium citrate (FAC) or 200 μM desferrioxamine, respectively. For protease inhibition studies, cells were pretreated for 1 h with either 0.1 mM MG-132 or 0.1 mM calpain II inhibitor. To inhibit lysosomes, cells were pretreated for 1 h with either 20 mM ammonium chloride or 0.15 mM chloroquine prior to the addition of FAC.

       Measurement of IRP2 Degradation Rates

      Cells were preincubated in methionine-free media in the presence of 100 μCi/ml Tran35S-label (ICN Biomedicals) for 4 h, after which fresh media containing an excess of unlabeled methionine with or without 50 μg/ml FAC was added. At the indicated times, cells were lysed in Buffer A (20 mM HEPES, pH 7.5, 2 mM dithiothreitol, 5% glycerol, 40 mM KCl) containing 0.5% Nonidet P-40, 1 mM phenylmethylsulfonyl fluoride), and an aliquot of each sample was used to determine protein concentration using the bicinchoninic acid protein assay (Pierce). Immunoprecipitation of IRP2 was carried out by incubating 30 μg of labeled extracts with 5 μl of rabbit anti-rat IRP2 antisera for 3 h, followed by the addition of 20 μl of protein A-agarose suspension for 2 h. The immunocomplexes were washed with RIPA buffer (150 mM NaCl2, 0.5% deoxycholate, 0.1% SDS, 1% Nonidet P-40, 50 mM Tris-HCl, pH 8.0), boiled in SDS sample buffer (50 mM Tris-HCl, pH 6.8, 2% SDS, 1%β-mercaptoethanol, 10% glycerol), and analyzed by 8% SDS-PAGE. The labeled protein was transferred to polyvinylidene difluoride (Millipore) membrane and subjected to autoradiography. The t of IRP2 was determined by densitometric analysis of the labeled IRP2 bands.

       Measurement of IRP2 Synthesis Rates

      FTO2B cells were preincubated for 2.5 h in medium in the presence or the absence of 50 μg/ml FAC. The medium was replaced with methionine-free medium for 15 min in the absence or presence of FAC. Tran35S-label (100 μCi/ml) was added to the cells and then cells were harvested at 10, 20, 40, or 60 min after labeling. IRP2 was immunoprecipitated from each extract (100 μg) and analyzed by 8% SDS-PAGE as described above.

       RNA Band Shift Assays and Immunoblot Analysis

      Cell lysates for RNA band shift assays and immunoblots were prepared by lysing cells in Buffer A containing 0.5% Nonidet P-40, 1 mM phenylmethylsulfonyl fluoride. The lysates were spun in at 13,000 × g for 10 min, and protein concentration was quantitated using the bicinchoninic acid protein assay (Pierce). For RNA band shift assays using anti-IRP2 antisera to “supershift” the IRP2•IRE complex from the lysates, 5 μl of anti-IRP2 antisera generated against the 73-amino acid insertion in IRP2 (
      • Guo B.
      • Yu Y.
      • Leibold E.A.
      ) or 5 μl of rabbit preimmune antisera was preincubated with extracts for 5 min before the addition of the 32P-labeled RNA. RNA band shift assays were carried out as described previously(
      • Leibold E.A.
      • Munro H.N.
      ).
      For immunoblot analysis, 50 μg of protein from cell lysates was fractionated on 8% SDS-PAGE. The protein was transferred to nitrocellulose membranes, and the membranes were incubated with a chicken anti-IRP1 antibody generated against the entire IRP1 protein (
      • Guo B.
      • Yu Y.
      • Leibold E.A.
      ) or a rabbit anti-IRP2 antibody(
      • Guo B.
      • Yu Y.
      • Leibold E.A.
      ). After 1 h, the membranes were washed and incubated with horseradish peroxidase-conjugated goat anti-chicken or goat anti-rabbit IgG for 1 h. The protein was visualized using the enhanced chemiluminescence Western blotting detection system (Amersham Corp.) according to the manufacturer.

       Northern Blotting

      Total RNA was isolated from FTO2B cells treated with FAC or desferrioxamine using TRIzol (Life Technologies, Inc.). The RNA was fractionated on a 1% formaldehyde-agarose gel and was transferred to nylon membranes. The membranes were hybridized with a random-primed (
      • Feinberg A.P.
      • Vogelstein B.
      ) 32P-labeled IRP2 cDNA using Rapid Hyb (Amersham) according to the manufacturer and washed in 1 × SSC (20 × SSC = 0.3 M sodium citrate, 3 M NaCl2) containing 0.1% SDS for 30 min at 65°C. The membrane was stripped and reprobed with random-primed 32P-labeled glyceraldehyde phosphate dehydrogenase cDNA to control for gel loading. The filters were subjected to autoradiography and the intensity of the bands quantified by densitometry.

      RESULTS

       Iron Reduces IRP2 Levels in Rat Hepatoma and in Human HeLa Cells

      We have previously demonstrated that treatment of FTO2B cells with FAC results in a 5-fold decrease in IRP1 RNA binding activity and undetectable IRP2 RNA binding activity after 4 h(
      • Guo B.
      • Yu Y.
      • Leibold E.A.
      ). The decrease in IRP2 RNA binding activity correlated with a decrease in IRP2 concentration; however, the IRP1 levels were not measured in those experiments. Therefore, FTO2B cells were treated with FAC for 1-24 h, and IRP1 and IRP2 RNA binding activity and their protein levels were measured by RNA band shift gels and immunoblot analysis using anti-IRP1 or anti-IRP2 antisera, respectively (Fig. 1, A and B). Anti-IRP2 antibody was generated against the IRP2 73 amino acid insertion not present in IRP1 and is therefore specific for IRP2. Fig. 1A shows that IRP1 and IRP2 RNA binding activities begins to decrease at 2.5 h after iron treatment and remain low during the 24-h time course. During this time, the amount of IRP1 remained constant (Fig. 1B, lanes 1-8), whereas the amount of IRP2 is reduced about 5-fold and reflected the decreases in RNA binding activity (Fig. 1B, lanes 9-16).
      Figure thumbnail gr1
      Figure 1:Effect of iron treatment on RNA binding activity and levels of IRP1 and IRP2 in rat hepatoma cells. A, FTO2B cells were grown in the presence (lanes 2-7) or the absence (lanes 1 and 8) of 50 μg/ml FAC for 1-24 h after which lysates were prepared as described under “Experimental Procedures.” C24, untreated cells harvested at 24 h. Equal amounts of protein (10 μg) were incubated with a 32P-labeled IRE RNA followed by electrophoresis of the RNA-protein complexes by a 5% native polyacrylamide gel. The positions of IRP1•IRE and IRP2•IRE complexes and free IRE RNA are indicated. B, equal amounts of protein (50 μg) from A were subjected to 8% SDS-PAGE followed by immunoblot analysis using anti-rat IRP1 or anti-IRP2 antisera. Molecular weight standards are indicated.
      To determine if the iron-mediated decrease in IRP2 levels occurs in other cell types, we measured RNA binding activity and protein levels for IRP1 and IRP2 in human HeLa cells treated with FAC for 1-24 h (Fig. 2). Since human IRP1•IRE and IRP2•IRE complexes comigrate on native polyacrylamide gels, we carried out supershift assays using anti-IRP2 antisera in RNA band shift assays. We have previously demonstrated that anti-IRP2 antisera does not interfere with RNA binding and results in a supershifted IRP2•IRE complex(
      • Guo B.
      • Yu Y.
      • Leibold E.A.
      ). Treatment of cells with FAC caused RNA-binding activity of IRP1 and IRP2 to decrease 2- and 5-fold, respectively (Fig. 2, A and C). Immunoblot analysis indicated that the amount of IRP1 remained constant during iron treatment, whereas the amount of IRP2 decreased 5-fold (Fig. 2, B and C). We have observed reductions in the amount of IRP2 in mouse 3T3 fibroblasts and transformed human primary embryonal 293 kidney cells treated with FAC for the same time course (data not shown). These data indicated that the decrease in IRP1 and IRP2 RNA binding activities induced by iron are mediated by different cellular processes and occurs in a variety of cell types.
      Figure thumbnail gr2
      Figure 2:Effect of iron treatment on RNA binding activity and protein levels of IRP1 and IRP2 in human HeLa cells. A, HeLa cells were grown in the presence (lanes 2-7) or absence (lane 1) of 50 μg/ml FAC. Equal amounts of protein (10 μg) were incubated with (lanes 9-15) or without (lanes 1-8) anti-IRP2 antisera for 5 min followed by the addition of 32P-labeled IRE RNA. As a control, rabbit preimmune antisera was added to an extract from untreated cells (C/PI, lane 8). IRP2•IRE complexes are indicated by an asterisk. B, equal amounts of protein (50 μg) from extracts in A were subjected to immunoblot analysis using anti-IRP1 or anti-IRP2 antisera. C, the data in A and B were quantified by densitometry and plotted using untreated control as 100%.

       Iron Does Not Affect Amounts of the IRP2 6.4-kb mRNA

      The decrease in IRP2 protein levels induced by iron could reflect reduced mRNA levels due to changes in transcription or mRNA stability, or a decrease in the rate of translation of its mRNA, or an increased rate of IRP2 degradation. To determine if iron affects IRP2 mRNA levels, we quantified IRP2 mRNA in FTO2B cells treated with FAC for 0-24 h and desferrioxamine, an intracellular iron chelator for 16 h (Fig. 3A). Our previous studies have indicated that desferrioxamine increases IRP2 levels about 2-fold in FTO2B cells(
      • Guo B.
      • Yu Y.
      • Leibold E.A.
      ). As a control for gel loading, the amounts of IRP2 mRNA were normalized to glyceraldehyde phosphate dehydrogenase mRNA (Fig. 3B). Northern blotting using a labeled rat IRP2 cDNA showed that the IRP2 cDNA hybridizes to two major transcripts of 6.4 and 3.7 kb and a minor transcript of 4.2 kb after high stringency washes. All three transcripts are capable of encoding the 104,000-Da IRP2. Treatment of cells with FAC had no significant effect on the amounts of the 6.4- and 4.2-kb transcripts (Fig. 3A). The amount of the 3.7-kb transcript increases 2-fold in desferrioxamine-treated cells. Our previous studies have suggested that changes in the 3.7-kb levels may be due to increased utilization of an alternative polyadenylation site in iron-deprived cells; however, the significance of this result is unclear(
      • Guo B.
      • Brown F.M.
      • Phillips J.D.
      • Yu Y.
      • Leibold E.A.
      ). These data indicate that the 5-fold reduction in IRP2 protein levels and RNA binding activity induced by iron is not due to alterations in the steady-state levels of the IRP2 mRNAs.
      Figure thumbnail gr3
      Figure 3:Effect of iron treatment on IRP2 mRNA levels in rat hepatoma cells. FTO2B cells were grown in the presence of 50 μg/ml FAC (Iron) for 1-24 h or 200 μM desferrioxamine (Df) for 16 h. Total RNA was isolated and analyzed on an 1% formaldehyde-agarose gel. C0 and C24, untreated cells harvested at 0 and 24 h, respectively. The RNA was transferred to a membrane and sequentially hybridized with a 32P-labeled IRP2 cDNA (A) or a 32P-labeled glyceraldehyde phosphate dehydrogenase cDNA (B) to control for gel loading. The size of the IRP2 transcripts are indicated by arrows. RNA molecular weight standards are from Life Technologies, Inc.

       Iron Increases the Rate of IRP2 Degradation, but Has No Effect on the Rate of IRP2 Synthesis

      The reduction in IRP2 levels by iron could be due to either an increase in IRP2 turnover, a decrease in IRP2 synthesis, or a combination of both. To test whether iron increases the degradation rate of IRP2, pulse-chase experiments were carried out in the presence or absence of FAC, labeled IRP2 was immunoprecipitated with anti-IRP2 followed by fractionation of the immunocomplexes by SDS-PAGE (Fig. 4A). IRP2 was identified by comparison with immunoprecipitated protein from a control extract with preimmune rabbit antisera (Fig. 4A, lanes 1 and 2). FAC treatment increased the rate of degradation of IRP2 approximately 3-fold (Fig. 4A, lanes 12-18). The half-life of IRP2 in the presence and in the absence of FAC was reduced from 4.5 h in control cells to 1.5 h in iron-treated cells (Fig. 4B).
      Figure thumbnail gr4
      Figure 4:Effect of iron on the rate of IRP2 degradation. A, FTO2B cells were pulse-labeled for 4 h with Tran35S-label and chased in medium containing an excess of unlabeled methionine in the absence (lanes 1-11) or the presence (lanes 12-18) of 50 μg/ml FAC for 0-8 h. IRP2 was immunoprecipitated using anti-IRP2 antisera (lanes 3-18). As a control, IRP2 was immunoprecipitated from extracts from untreated cells using preimmune rabbit serum (PI) (lanes 1 and 2). Labeled immunoprecipitated protein was analyzed by 8% SDS-PAGE. The positions of the molecular weight standards and IRP2 are indicated. B, the turnover data in A was quantified by densitometry, and the intensity of the IRP2 bands were plotted relative to the percent of radioactivity remaining after 0 h (lanes 3 and 4). These experiment were carried out three times, and one representative experiment is shown. Symbols: ■, no addition; ♦, iron.
      It was possible that in addition to decreasing the half-life of IRP2, iron could also reduce its rate of synthesis. Measurements of synthesis rates in the presence of iron could be misleading, since it would not only measure synthesis, but would also measure degradation of newly synthesized protein. To determine if synthesis of IRP2 was affected by iron, we treated cells with FAC for 2.5 h, then quantified the amount of methionine incorporated into IRP2 during a short 1-h time course. Our turnover data indicated that in 1 h after iron treatment approximately 30% of labeled pre-existing IRP2 is degraded. Labeled IRP2 was immunoprecipitated using anti-IRP2 antibodies at 10, 20, 40, and 60 min followed by fractionation of the immunocomplexes by SDS-PAGE (Fig. 5A) and quantification of the radioactivity in the IRP2 bands by densitometry (Fig. 5B). At time points between 10 and 40 min, the rate of IRP2 synthesis was not significantly different in iron-treated or control cells. After 1 h of labeling, the amount of labeled IRP2 in iron-treated cells decreased slightly compared with the amount of IRP2 in control cells. The decrease in IRP2 label at 1 h is presumably due to the increase in the degradation of newly synthesized protein. We conclude that the iron-mediated reduction in IRP2 levels is due to an increased rate of degradation without changes in the rate of IRP2 synthesis.
      Figure thumbnail gr5
      Figure 5:Effect of iron on the rate of IRP2 synthesis. A, FTO2B cells were grown in the presence (lanes 5-8) or the absence (lanes 1-4) of 50 μg/ml FAC for 2.5 h. The cells were incubated in methionine-free medium for 15 min with or without FAC. After 15 min, the cells were labeled with 100 μCi/ml Tran35S-label and were then harvested after 10, 20, 40, and 60 min. IRP2 was immunoprecipitated using anti-IRP2 antisera and analyzed by 8% SDS-PAGE. Molecular weight standards and IRP2 are indicated. PI, control 60-min lysate immunoprecipitated with preimmune serum. B, the synthesis data in A was quantified by densitometry and the integrated density of labeled IRP2 bands was plotted. These data represent the results from two experiments.

       Protein Synthesis, but Not Transcription Is Required for the Increased Rate of Degradation of IRP2 by Iron

      To determine if protein synthesis is required for the increased degradation rate of IRP2 induced by iron, FTO2B cells were treated with the protein synthesis inhibitor, cycloheximide, in the presence or absence of FAC for 0-4 h, and IRP1 and IRP2 RNA binding activity and IRP2 protein levels were measured (Fig. 6, A and B). FAC caused a decrease in IRP1 and IRP2 RNA binding activity (Fig. 6B) and IRP2 protein levels (Fig. 6A). Cycloheximide alone had no effect on IRP1 or IRP2 RNA binding activity or IRP2 protein levels. When cells were treated with cycloheximide and FAC, IRP1 RNA binding activity decreased similar to cells treated with FAC alone. In contrast, the reduction in IRP2 RNA binding activity and IRP2 protein levels observed with iron treatment did not occur when protein synthesis was inhibited (Fig. 6, A and B). The expected decrease in IRP1 RNA-binding activity in cells treated with cycloheximide and iron showed that cycloheximide did not interfere with iron uptake into cells. Identical results were obtained in FTO2B cells treated with the protein synthesis initiation inhibitor, emetine (data not shown).
      Figure thumbnail gr6
      Figure 6:Effect of cycloheximide on the iron-mediated degradation of IRP2 by iron. FTO2B cells were grown in the presence (lanes 3-5) or absence (lanes 1 and 2) of 50 μg/ml FAC (Iron), 20 μg/ml cycloheximide (Cyx) (lanes 6-8), or FAC plus cycloheximide (Iron + Cyx) (lanes 9-11) for 0-4 h. A, equal amounts of protein (50 μg) were subjected to 8% SDS-PAGE for immunoblot analysis using anti-IRP2 antisera. Molecular weight standards and the positions of IRP2 and a nonspecific immunoreactive band (ns) are indicated. B, equal amount of protein (10 μg) from extracts in A was incubated with 32P-labeled IRE followed by electrophoresis of the RNA-protein by 5% native polyacrylamide gels. The positions of IRP1•IRE and IRP2•IRE complexes are indicated.
      We also determined if transcription is required for the degradation of IRP2 induced by iron. FTO2B cells were treated with the transcription inhibitor, actinomycin D alone, or in the presence or absence of FAC for 0, 1, 2.5, and 4 h, and IRP1 and IRP2 RNA binding activities and IRP2 levels were measured (Fig. 7, A and B). Actinomycin D alone had no effect on IRP1 or IRP2 RNA binding activities (Fig. 7B) or IRP2 protein levels (Fig. 7A). When cells were treated with FAC and actinomycin D, IRP1 and IRP2 RNA binding activities and IRP2 levels decreased, but not to the levels observed with iron alone. These data indicated that the iron-mediated degradation of IRP2 requires protein synthesis, but to a lesser extent transcription, suggesting that the synthesis of a labile protein is required for IRP2 degradation.
      Figure thumbnail gr7
      Figure 7:Effect of actinomycin D on the iron-mediated degradation of IRP2. FTO2B cells were grown in the presence (lanes 5-7) or absence (lanes 1-4) of 50 μg/ml FAC (Iron), 10 μM actinomycin D (lanes 8-10), or FAC plus actinomycin D (Act D + Iron) (lanes 11-13) for 0-4 h. Immunoblot analysis (A) and RNA band shift assays (B) were carried out as described in the legend to .

       A Proteasome Inhibitor Blocks the Iron-mediated Degradation of IRP2 in FTO2B Cells

      To identify the proteolytic system responsible for the iron-mediated degradation of IRP2, we tested whether proteosomal, lysosomal, and cysteine protease inhibitors prevented IRP2 iron-mediated degradation. The multicatalytic 26 S proteasome complex catalyzes the degradation of proteins via either ubiquitin-dependent or ubiquitin-independent pathways(
      • Rechsteiner M.
      • Hoffman L.
      • Dubiel W.
      ,
      • Ciechanover A.
      ). First, to determine whether IRP2 is degraded via proteasomes, we tested the effect of the potent proteasome inhibitor MG-132 on IRP2 degradation. MG-132 is a peptide-aldehyde that inhibits the chymotrypic activities of the proteasome (
      • Palombella V.J.
      • Rando O.J.
      • Goldberg A.L.
      • Maniatis T.
      ,
      • Rock K.L.
      • Gramm C.
      • Rothstein L.
      • Clark K.
      • Stein R.
      • Dick L.
      • Hwang D.
      • Goldberg A.L.
      ) and can inhibit intracellular proteolysis for many hours without cellular toxicity(
      • Rock K.L.
      • Gramm C.
      • Rothstein L.
      • Clark K.
      • Stein R.
      • Dick L.
      • Hwang D.
      • Goldberg A.L.
      ). FTO2B cells were pretreated with MG-132 1 h prior to the addition of FAC for 1-4 h, and cytoplasmic lysates were analyzed for IRP1 and IRP2 RNA binding activity and IRP2 protein levels. Fig. 8 shows that in cells treated with MG-132 in the presence of FAC, the decrease in IRP2 RNA binding activity (bottom panel, lanes 9-11) and IRP2 protein levels (top panel, lanes 9-11) observed with iron alone (lanes 6-8) is blocked. Densitometric analysis indicated that MG-132 inhibited the iron-mediated degradation of IRP2 and the decrease in IRP2 RNA binding activity by 90% at 4 h after treatment. Treatment of cells with MG-132 alone had no effect on IRP2 RNA binding activity (bottom panel, lanes 2-5) or IRP2 protein levels (top panel, lanes 2-5). Surprisingly, IRP1 RNA binding activity decreased in MG-132-treated cells (lanes 2-5) similar to that observed in iron-treated cells (lanes 6-8). Immunoblot analysis showed that IRP1 protein levels did not change during MG-132 treatment (data not shown), indicating that decreased RNA binding activity was not due to decreased IRP1 protein levels. Although the mechanism causing IRP1 RNA binding activity to decrease in MG-132-treated cells is uncertain, we believe that it may be due to changes in intracellular iron levels mediated by inhibition of protein degradation by MG-132.
      Figure thumbnail gr8
      Figure 8:Effect of a proteasome inhibitor on IRP2 iron-mediated degradation. A, FTO2B cells were pretreated with 100 μM MG-132 in 1.0% dimethyl sulfoxide (Me2SO) for 1 h prior to the addition of 50 μg/ml FAC for 1-4 h (lanes 9-11). Cells were also treated with MG-132 for 1-5 h (lanes 2-5) or FAC in Me2SO for 1-4 h (lanes 6-8). CO control, untreated cells harvested at 0 h. The top panel is an immunoblot using anti-IRP2 antisera, and the bottom panel is an RNA band shift assay. The positions of IRP1 and IRP2 are indicated.
      Because MG-132 also inhibits calpains and lysosomal cysteine proteases, such as cathepsin B,
      F. L. Stassen, personal communication.
      we tested whether calpain II (N-acetyl-leucinyl-leucinyl-methional-H), a cysteine protease inhibitor, and the lysosomal inhibitors, ammonium chloride and chloroquine, prevented iron-mediated IRP2 degradation. Previous studies demonstrated that calpain II inhibitors have little effect on proteasome function(
      • Rock K.L.
      • Gramm C.
      • Rothstein L.
      • Clark K.
      • Stein R.
      • Dick L.
      • Hwang D.
      • Goldberg A.L.
      ). Fig. 9A shows that the treatment of cells with calpain II inhibitor in the presence of FAC has no effect on IRP2 iron-mediated degradation. Ammonium chloride also did not inhibit IRP2 degradation by iron (Fig. 9B). We conclude from these studies that the proteasomes, and not the lysosomes, are required for iron-mediated degradation of IRP2.
      Figure thumbnail gr9
      Figure 9:Effect of calpain II and lysosomal inhibitors on IRP2 iron-mediated degradation. A, cells were pretreated with 100 μM calpain II inhibitor in 0.4% dimethylformamide for 1 h prior to the addition of 50 μg/ml FAC (lanes 6-8) or in FAC in DMF (lanes 2-4) for 1-4 h. CO control, untreated cells harvested at 0 h. B, cells were pretreated with 20 mM ammonium chloride (lanes 5-7) or 0.15 mM chloroquine (lanes 8-10) for 1 h prior to the addition of 50 μg/ml FAC for 1-4 h. Cells were also treated with FAC (lanes 2-4) for 1-4 h. The top panels of A and B are immunoblots using anti-IRP2 antisera, and the bottom panels are RNA band shift assays. The positions of IRP1 and IRP2 are indicated.

      DISCUSSION

      In this paper we report the differential regulation of IRP1 and IRP2 by iron in mammalian cells. IRP1 exhibits two functions in cells dependent on iron levels: IRP1 with an 4Fe-4S cluster functions as an cytosolic aconitase converting citrate into isocitrate when iron is abundant and as an RNA binding apoprotein regulating the translation and stabilization of IRE-containing mRNAs when iron is scarce(
      • Philpott C.C.
      • Klausner R.D.
      • Rouault T.A.
      ,
      • Haile D.J.
      • Rouault T.A.
      • Harford J.B.
      • Kennedy M.C.
      • Blondin G.A.
      • Beinert H.
      • Klausner R.D.
      ,
      • Emery-Goodman A.
      • Hirling H.
      • Scarpellino L.
      • Henderson B.
      • Kuhn L.C.
      ,
      • Hirling H.
      • Henderson B.R.
      • Kuhn L.C.
      ). The switch between the 4Fe-4S form and the apoprotein forms occurs without changes in IRP1 levels(
      • Yu Y.
      • Radisky E.
      • Leibold E.A.
      ,
      • Tang C.K.
      • Chin J.
      • Harford J.B.
      • Klausner R.D.
      • Rouault T.A.
      ). By contrast, IRP2 lacks aconitase activity and functions solely as an RNA binding protein(
      • Guo B.
      • Yu Y.
      • Leibold E.A.
      ). Our results indicate that IRP2 is regulated by specific proteolysis induced by iron in a variety of cells types and that the proteasome is responsible for IRP2 degradation.
      Our data suggest a mechanism for the iron-mediated degradation of IRP2. When intracellular iron is scarce, IRP2 binds IREs with high affinity. An increase in intracellular iron results in the induction of a labile protein that is required for IRP2 degradation. Although we do not know the identity and function of this protein, it is possible that it is a targeting protein that binds IRP2 via the 73-amino acid domain, marking it for degradation. Iron could also cause the assembly of an 4Fe-4S cluster in IRP2, similar to the cluster in IRP1. Rat IRP2 contains the 3 conserved cysteines that coordinate the 4Fe-4S cluster in IRP1(
      • Samaniego F.
      • Chin J.
      • Iwai K.
      • Rouault T.A.
      • Klausner R.D.
      ,
      • Guo B.
      • Brown F.M.
      • Phillips J.D.
      • Yu Y.
      • Leibold E.A.
      ). In addition, the presence of 4 cysteines and 1 histidine in the 73-amino acid insertion of IRP2 suggests that this region might also participate in iron binding(
      • Guo B.
      • Brown F.M.
      • Phillips J.D.
      • Yu Y.
      • Leibold E.A.
      ). Preliminary data suggests that in vitro reconstitution of IRP2 with iron results in loss in RNA binding activity.
      J. D. Phillips and E. A. Leibold, unpublished data.
      Thus, according to our model, cluster assembly woud lead to a conformational change in IRP2 and subsequent loss in RNA binding activity. IRP2 would then be recognized by the targeting protein and rapidly degraded by the proteasome. Finally, our data indicate that the decrease IRP2 RNA binding activity mediated by iron is also prevented when IRP2 proteolysis is blocked either by MG-132 or by cycloheximide. One possibility to explain these data is that the putative Fe-S cluster is unstable in IRP2 and is disassembled during extract purification, leading to the generation of an apoprotein containing RNA binding activity.
      The 26 S proteasome contains subunits which are important in the degradation of ubiquitin-conjugated proteins(
      • Rechsteiner M.
      • Hoffman L.
      • Dubiel W.
      ,
      • Ciechanover A.
      ). We have not detected higher molecular weight IRP2 complexes by gel electrophoresis, which might be suggestive of ubiquitination of IRP2. However, since ubiquitin-conjugated proteins are very labile, they are generally difficult to detect. The 26 S proteasome also degrades non-ubiquitinated proteins(
      • Ciechanover A.
      ,
      • Murakami Y.
      • Matsufuji S.
      • Kameji T.
      • Hayashi S.
      • Igarashi K.
      • Tamura T.
      • Tanaka K.
      • Ichihara A.
      ). The signals required for targeting non-ubiquitinated proteins to the proteasome are poorly understood; however, it is possible that the putative targeting protein discussed above could mark IRP2 for degradation by the proteasome.
      Although we cannot eliminate the possibility that MG-132 may affect other unknown proteases and enzymatic activities in cells, the utilization of these inhibitors both in vitro and in vivo have demonstrated the specificity and effectiveness of these compounds against the proteasome(
      • Rock K.L.
      • Gramm C.
      • Rothstein L.
      • Clark K.
      • Stein R.
      • Dick L.
      • Hwang D.
      • Goldberg A.L.
      ). Our data suggested that MG-132 may increase cellular iron levels, perhaps by blocking the degradation of iron transporter proteins. Peptide-aldehyde inhibitors have been used to demonstrate the role of the proteasome in the generation of peptides presented on the major histocompatibility class I molecules (
      • Rock K.L.
      • Gramm C.
      • Rothstein L.
      • Clark K.
      • Stein R.
      • Dick L.
      • Hwang D.
      • Goldberg A.L.
      ) and in the proteolytic processing of the transcription factor NF-κB1(
      • Palombella V.J.
      • Rando O.J.
      • Goldberg A.L.
      • Maniatis T.
      ).
      The structural determinants required for IRP2 iron-mediated degradation are unknown. IRP2 does not contain PEST regions (sequences rich in proline, glutamine, serine, and threonine) which are commonly found in proteins that are rapidly degraded(
      • Rogers S.
      • Wells R.
      • Rechsteiner M.
      ). However, IRP2, the 73-amino acid insertion, contains a site that is susceptible to proteolysis during purification and results in the production of an 83,000-Da proteolytic polypeptide(
      • Guo B.
      • Yu Y.
      • Leibold E.A.
      ). The cleavage site has the sequence SQ- IENTP and is not a known protease cleavage sequence. Whether proteolysis at this site represents a physiological mechanism for iron-mediated degradation or whether the 73-amino acid insertion is a determinant required for degradation remains to be determined.
      The biological relevance of two IRPs in cells is unclear. Both IRP1 and IRP2 bind IREs with high affinity (
      • Guo B.
      • Yu Y.
      • Leibold E.A.
      ,
      • Henderson B.R.
      • Seiser C.
      • Kuhn L.C.
      ,
      • Samaniego F.
      • Chin J.
      • Iwai K.
      • Rouault T.A.
      • Klausner R.D.
      ) and function as translational repressors of IRE-containing RNAs in vitro(
      • Guo B.
      • Yu Y.
      • Leibold E.A.
      ). First, it is possible that IRP2 binds to a subset of IRE-containing mRNAs containing slightly different sequences. A recent study using in vitro synthesized IREs demonstrated that mouse IRP2 has a preference for specific IRE sequences, suggesting that IRP2 may bind to specific IRE-containing mRNAs in vivo(
      • Henderson B.R.
      • Menotti E.
      • Bonnard C.
      • Kuhn L.C.
      ). Second, since IRP2 is present in the highest amounts in skeletal muscle and heart, this suggests that IRP2 may regulate muscle-specific mRNAs(
      • Guo B.
      • Brown F.M.
      • Phillips J.D.
      • Yu Y.
      • Leibold E.A.
      ). Third, IRP2 RNA binding activity is decreased in the livers of rats treated with chemicals to induce oxidative stress (
      • Cairo G.
      • Tacchini L.
      • Pogliaghi G.
      • Anzon E.
      • Tomasi A.
      • Bernelli-Zazzera A.
      ) and increased in regenerating rat livers(
      • Cairo G.
      • Pietrangelo A.
      ), suggesting that IRP2 is regulated under a variety of physiological states. It is unclear whether these effects are due to alterations in intracellular iron levels or to stimuli other than iron.
      A recent study suggested that iron-mediated regulation of IRP2 degradation may be cell-specific(
      • Samaniego F.
      • Chin J.
      • Iwai K.
      • Rouault T.A.
      • Klausner R.D.
      ). A c-myc-tagged recombinant IRP2 expressed in HeLa cells treated with iron or hemin for 16 h resulted in a decrease in RNA binding activity, but no change in the amount of protein. Our experiments analyzing the iron-mediated regulation of endogenous IRP2 in HeLa cells treated with iron for up to 24 h showed a steady decrease in RNA binding activity and protein levels up to 6 h, after which RNA binding activity and protein levels gradually increased. The half-life of recombinant IRP2 expressed in RD-4 cells was greater than 24 h in desferrioxamine-treated cells and 6 h in iron-treated cells(
      • Samaniego F.
      • Chin J.
      • Iwai K.
      • Rouault T.A.
      • Klausner R.D.
      ). By contrast, our data indicated that the half-life of endogenous IRP2 in untreated FTO2B cells was 6 h and 1.5 h in iron-treated cells. The discrepancies between these studies may reflect differences in experimental design due to use of overexpressed protein or to different cell growth conditions.
      The regulation of gene expression by specific proteolysis provides a way by which cells can change the concentration of specific proteins depending on the metabolic state of the cell. The iron-dependent regulation of IRP2 turnover may be similar to the mechanisms regulating the mammalian enzyme ornithine decarboxylase. Ornithine decarboxylase is the first enzyme in the polyamine biosynthesis pathway and is degraded when intracellular polyamine levels increase(
      • McConlogue L.
      • Dana S.L.
      • Coffino P.
      ). Polyamines induce antizyme, a protein which binds with high affinity to ornithine decarboxylase (
      • Li X.
      • Coffino P.
      ,
      • Li X.
      • Coffino P.
      ) and targets ornithine decarboxylase for degradation by the proteasome(
      • Murakami Y.
      • Matsufuji S.
      • Kameji T.
      • Hayashi S.
      • Igarashi K.
      • Tamura T.
      • Tanaka K.
      • Ichihara A.
      ). Thus, it is possible that IRP2, like ornithine decarboxylase, may utilize other proteins that specify its degradation during changes in intracellular iron levels. The characterization of the other components responsible for IRP2 iron-mediated degradation will provide a clearer understanding of the mechanism by which IRP2 is targeted and degraded by the proteasome.

      ACKNOWLEDGEMENTS

      We thank Dennis Winge, Liz Wyckoff, Andy Sewell, and members of the laboratory for insightful comments during the course of this work and for critical readings of the manuscripts.

      REFERENCES

        • Leibold E.A.
        • Guo B.
        Annu. Rev. Nutr. 1992; 12: 345-368
        • Klausner R.D.
        • Rouault T.A.
        • Harford J.B.
        Cell. 1993; 72: 19-28
        • Theil E.C.
        Biofactors. 1993; 4: 87-93
        • Beinert H.
        • Kennedy M.C.
        FASEB J. 1993; 7: 1442-1449
        • Walden W.E.
        • Patino M.M.
        • Gaffield L.
        J. Biol. Chem. 1989; 264: 13765-13769
        • Goossen B.
        • Caughman S.W.
        • Harford J.B.
        • Klausner R.D.
        • Hentze M.W.
        EMBO J. 1990; 9: 4127-4133
        • Gray N.K.
        • Hentze M.W.
        EMBO J. 1994; 13: 3882-3891
        • Casey J.L.
        • Hentze M.W.
        • Koeller D.M.
        • Caughman S.W.
        • Rouault T.A.
        • Klausner R.D.
        • Harford J.B.
        Science. 1988; 240: 924-928
        • Mullner E.W.
        • Neupert B.
        • Kuhn L.C.
        Cell. 1989; 58: 373-382
        • Binder R.
        • Horowitz J.A.
        • Basilion J.P.
        • Koeller D.M.
        • Klausner R.D.
        • Harford J.B.
        EMBO J. 1994; 13: 1969-1980
        • Rouault T.A.
        • Tang C.K.
        • Kaptain S.
        • Burgess W.H.
        • Haile D.J.
        • Samaniego F.
        • McBride O.W.
        • Harford J.B.
        • Klausner R.D.
        Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 7958-7962
        • Yu Y.
        • Radisky E.
        • Leibold E.A.
        J. Biol. Chem. 1992; 267: 19005-19010
        • Rouault T.A.
        • Stout C.D.
        • Kaptain S.
        • Harford J.B.
        • Klausner R.D.
        Cell. 1991; 64: 881-883
        • Kennedy M.C.
        • Mende-Mueller L.
        • Blondin G.A.
        • Beinert H.
        Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 11730-11734
        • Kaptain S.
        • Downey W.E.
        • Tang C.
        • Philpott C.
        • Haile D.
        • Orloff D.G.
        • Harford J.B.
        • Rouault T.A.
        • Klausner R.D.
        Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 10109-10113
        • Basilion J.P.
        • Rouault T.A.
        • Massinople C.M.
        • Klausner R.D.
        • Burgess W.H.
        Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 574-578
        • Swenson G.R.
        • Walden W.E.
        Nucleic Acids Res. 1994; 22: 2627-2633
        • Philpott C.C.
        • Klausner R.D.
        • Rouault T.A.
        Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 7321-7325
        • Leibold E.A.
        • Munro H.N.
        Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 2171-2175
        • Guo B.
        • Yu Y.
        • Leibold E.A.
        J. Biol. Chem. 1994; 269: 24252-24260
        • Cairo G.
        • Pietrangelo A.
        J. Biol. Chem. 1994; 269: 6405-6409
        • Cairo G.
        • Tacchini L.
        • Pogliaghi G.
        • Anzon E.
        • Tomasi A.
        • Bernelli-Zazzera A.
        J. Biol. Chem. 1994; 270: 700-703
        • Rouault T.A.
        • Haile D.J.
        • Downey W.E.
        • Philpott C.C.
        • Tang C.
        • Samaniego F.
        • Chin J.
        • Paul I.
        • Orloff D.
        • Harford J.B.
        • Klausner R.D.
        Biometals. 1992; 5: 131-140
        • Henderson B.R.
        • Seiser C.
        • Kuhn L.C.
        J. Biol. Chem. 1994; 268: 27327-27334
        • Kim H.
        • Klausner R.D.
        • Rouault T.A.
        J. Biol. Chem. 1995; 270: 4983-4986
        • Tang C.K.
        • Chin J.
        • Harford J.B.
        • Klausner R.D.
        • Rouault T.A.
        J. Biol. Chem. 1992; 267: 24466-24470
        • Samaniego F.
        • Chin J.
        • Iwai K.
        • Rouault T.A.
        • Klausner R.D.
        J. Biol. Chem. 1994; 269: 30904-30910
        • Feinberg A.P.
        • Vogelstein B.
        Anal. Biochem. 1983; 132: 6-13
        • Guo B.
        • Brown F.M.
        • Phillips J.D.
        • Yu Y.
        • Leibold E.A.
        J. Biol. Chem. 1995; 270: 16529-16535
        • Rechsteiner M.
        • Hoffman L.
        • Dubiel W.
        J. Biol. Chem. 1993; 268: 6065-6068
        • Ciechanover A.
        Cell. 1994; 79: 13-21
        • Palombella V.J.
        • Rando O.J.
        • Goldberg A.L.
        • Maniatis T.
        Cell. 1994; 78: 773-785
        • Rock K.L.
        • Gramm C.
        • Rothstein L.
        • Clark K.
        • Stein R.
        • Dick L.
        • Hwang D.
        • Goldberg A.L.
        Cell. 1994; 78: 761-771
        • Haile D.J.
        • Rouault T.A.
        • Harford J.B.
        • Kennedy M.C.
        • Blondin G.A.
        • Beinert H.
        • Klausner R.D.
        Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 11735-11739
        • Emery-Goodman A.
        • Hirling H.
        • Scarpellino L.
        • Henderson B.
        • Kuhn L.C.
        Nucleic Acids Res. 1993; 21: 1457-1461
        • Hirling H.
        • Henderson B.R.
        • Kuhn L.C.
        EMBO J. 1994; 13: 453-461
        • Murakami Y.
        • Matsufuji S.
        • Kameji T.
        • Hayashi S.
        • Igarashi K.
        • Tamura T.
        • Tanaka K.
        • Ichihara A.
        Nature. 1992; 360: 597-599
        • Rogers S.
        • Wells R.
        • Rechsteiner M.
        Science. 1986; 234: 364-368
        • Henderson B.R.
        • Menotti E.
        • Bonnard C.
        • Kuhn L.C.
        J. Biol. Chem. 1994; 269: 17481-17489
        • McConlogue L.
        • Dana S.L.
        • Coffino P.
        Mol. Cell. Biol. 1986; 6: 2865-2871
        • Li X.
        • Coffino P.
        Mol. Cell. Biol. 1992; 12: 3556-3562
        • Li X.
        • Coffino P.
        Mol. Cell. Biol. 1994; 14: 87-92