Ubiquitin aldehyde increases adenosine triphosphate-dependent proteolysis of hemoglobin alpha-subunits in beta-thalassemic hemolysates.

Two major causes of the anemia in beta-thalassemia are a deficiency in hemoglobin (Hb) beta-subunit (and consequently HbA) synthesis and, due to the resulting excess of Hb alpha-subunits, erythroid cell hemolysis. The hemolytic component might be ameliorated by increasing the intracellular proteolysis of the excess alpha-subunits. Isolated 3H-labeled alpha-chains are known to be degraded primarily by the adenosine triphosphate (ATP)- and ubiquitin (Ub)-dependent proteolysis pathway in unfractionated beta-thalassemic hemolysates. Our objective was to increase this degradation by targeted intervention. Ub aldehyde (Ubal), a synthetic inhibitor of isopeptidases (proteases that hydrolyze the bond between the Ub polypeptide and its protein adduct), was added to reaction mixtures containing a hemolysate from the blood cells of one of four beta-thalassemic donors and 3H-alpha-chains or 3H-alpha-globin as a substrate. Optimum enhancement of ATP-dependent degradation occurred at 0.4 to 1.5 micromol/L Ubal and ranged from 29% to 115% for 3H-alpha-chains and 47% to 96% for 3H-alpha-globin among the four hemolysates. We suggest that Ubal stimulates 3H-alpha-subunit proteolysis by inhibition of an isopeptidase(s) that deubiquitinates, or "edits," Ub-3H-alpha-subunit conjugates, intermediates in the degradative pathway. In control studies, similarly low Ubal concentrations did not enhance the degradation of 3H-alpha2beta2 (HbA) tetramers or inhibit the activities of methemoglobin reductase and four selected glycolysis pathway enzymes. These and other results may be the basis for a therapeutic approach to beta-thalassemia.

Differential effects of ubiquitin aldehyde on ubiquitin and ATP-dependent protein degradation.

ATP-dependent proteolysis of 125I-labeled human alpha-globin, bovine alpha-lactalbumin, bovine serum albumin, or chicken lysozyme was assessed in a rabbit reticulocyte extract supplemented with ATP, excess ubiquitin, and variable amounts of ubiquitin aldehyde (Ubal), an inhibitor of many ubiquitin-protein isopeptidases. Low concentrations (0.8 microM) of Ubal increased the ATP-dependent degradation of 125I-alpha-globin by approximately 30% after 2 h at 37 degrees C, had little effect on 125I-lysozyme turnover, and decreased 125I-alpha-lactalbumin or 125I-albumin degradation by approximately 20%. The ATP-dependent degradation of all substrates was inhibited by high concentrations (> 3 microM) of Ubal throughout the incubation (15 min to 2 h); after 2 h, this inhibition ranged from 15% for 125I-alpha-globin to approximately 85% for 125I-alpha-lactalbumin and 125I-albumin. Levels of ubiquitin-125I-protein conjugates were increased significantly with Ubal; with > or = 8.0 microM Ubal, high molecular mass multiubiquitinated conjugates were particularly evident for 125I-alpha-globin and 125I-alpha-lactalbumin. These mixtures also accumulated ubiquitin conjugates with sizes expected for di- through pentaubiquitin oligomers. The results are consistent with the following proposed events: The ATP-dependent degradation of 125I-alpha-lactalbumin or 125I-albumin is probably mediated almost exclusively through polyubiquitinated intermediates. High Ubal concentrations inhibit an isopeptidase(s) which normally disassembles "unanchored" polyubiquitin chains that remain after substrate degradation by the 26S proteasome; these chains accumulate to inhibit further conjugate degradation. Much of the ATP-dependent degradation of 125I-alpha-globin and, to a lesser degree, 125I-lysozyme may occur through alternative structures where ubiquitin monomers or short oligomers are ligated to one or more substrate lysines. For 125I-alpha-globin, even low concentrations of Ubal effectively inhibit deubiquitination of these conjugates to enhance alpha-globin degradation.

Degradation of monoubiquitinated alpha-globin by 26S proteasomes.

Ubiquitin-125I-alpha-globin conjugate fractions containing either one (Ub1-alpha), or two (Ub2-alpha), or a mixture of three and four (Ub3,4-alpha) molecules of ubiquitin (Ub), covalently linked to one 125I-alpha-globin molecule were isolated after incubation of a proteolysis reaction mixture containing ATP, ubiquitin aldehyde-treated reticulocyte lysate, and human 125I-alpha-globin. Each of the purified conjugate fractions or an identically-purified control sample of unconjugated 125I-alpha-globin was incubated as a substrate in companion proteolysis reaction mixtures containing either purified 26S or 20S rabbit reticulocyte proteasomes. The initial rate of ATP-dependent degradation of the Ub1-alpha conjugate by the 26S proteasomes was approximately 0.44% (1.1 fmol)/min while that of the free 125I-alpha-globin was undetectable. The initial rates of ATP-dependent degradation by the 26S proteasomes of the Ub2-alpha and Ub3,4-alpha conjugates were 2- to-3-fold that of the Ub1-alpha species. Conversely, the degradation of free 125I-alpha-globin and its ubiquitinated conjugates by the 20S proteasomes was not dependent on ATP, nor did it increase with the size of the Ub adduct. Analysis of the products of a reaction mixture with 26S proteasomes by sodium dodecyl sulfate-polyacrylamide gel electrophoresis showed no conversion of the Ub1-alpha conjugate substrate to higher-molecular-mass conjugates. These results suggest that monobiquitinated alpha-globin can be degraded significantly and specifically by interaction directly with the 26S proteasomes.(ABSTRACT TRUNCATED AT 250 WORDS)

Heterogeneity in the structure of the ubiquitin conjugates of human alpha globin.

Only monoubiquitinated (and to a much lesser extent diubiquitinated) 125I-alpha globin is observed as a conjugated intermediate in the ATP- and ubiquitin (Ub)-dependent proteolysis of 125I-alpha globin catalyzed by an unfractionated reticulocyte lysate (Shaeffer, J.R. (1994) J. Biol. Chem. 269, 22205-22210). A monoubiquitinated 125I-alpha globin (Ub1-alpha) fraction was isolated and incubated with a dilute acid reagent to selectively cleave the 125I-alpha globin moiety between residues 94 (Asp) and 95 (Pro). Analysis by sodium dodecyl sulfate-polyacrylamide gel electrophoresis of the resulting polypeptides showed that the Ub1-alpha conjugate consisted of a mixture of molecules in which 57% had Ub attached to the amino-terminal two-thirds and 43% had Ub attached to the carboxyl-terminal one-third of the 125I-alpha globin monomer. This amino- to carboxyl-terminal region ubiquitination ratio was essentially unchanged when a Ub1-alpha conjugate intermediate, pulse-labeled for 15 min in the presence of ubiquitin aldehyde to inhibit disassembly, was analyzed after (a) a 2-h chase incubation with excess nonradioactive alpha globin or (b) isolation and incubation separately as a proteolysis substrate. Similar analysis of the diubiquitinated 125I-alpha globin (Ub2-alpha) conjugate molecules showed that, after pulse labeling, 58% were diubiquitinated within the amino-terminal two-thirds and 42% were monoubiquitinated within both amino- and carboxyl-terminal regions of the 125I-alpha globin moiety (a small amount of molecules diubiquitinated within the carboxyl-terminal region may also be present) and that the relative amounts of these molecular types changed little, if any, during the chase incubation. This invariance in the amino- to carboxyl-terminal region ubiquitination ratio during degradation of the Ub1-alpha and Ub2-alpha conjugates suggests that 125I-alpha globin molecules ubiquitinated within either the amino- or carboxyl-terminal regions were both intermediates in the proteolysis of the unconjugated substrate. This heterogeneous pattern of Ub conjugation may also occur during the proteolysis of other long-lived intracellular proteins.

Monoubiquitinated alpha globin is an intermediate in the ATP-dependent proteolysis of alpha globin.

A dialyzed whole human reticulocyte (beta-thalassemic) lysate was incubated with human 125I-alpha globin at 37 degrees C for 15 min in a medium supporting proteolysis with ATP and added ubiquitin (Ub). Analysis of this reaction mixture by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) showed that approximately 0.5% of the initial 125I-alpha globin substrate radioactivity (125I cpm) was in the conjugate (Ub1-alpha), of one Ub monomer with one 125I-alpha globin monomer and substantially less in the Ub2-alpha conjugate. When the reaction mixture contained 0.5 microM ubiquitin aldehyde (Ubal) to inhibit endogenous isopeptidases, there was a 7-9-fold increase in the Ub1-alpha and Ub2-alpha conjugates and an appearance of higher molecular weight species including polyubiquitinated 125I-alpha globin (Ub > 10-alpha). Purified fractions of Ub1-alpha, Ub2-alpha, Ub3,4-alpha, and Ub > 10-alpha conjugates were isolated by preparative SDS-PAGE, dialyzed and treated with triethylamine to remove SDS, and dissolved in 10 mM HCOO-(Na+),pH 4.0. Incubation of each of the conjugate fractions with Ubal-treated lysate in a proteolysis reaction mixture showed a progressive increase in the initial rate of degradation (conversion to acid-soluble 125I cpm) with increase in molar amount of Ub from approximately 3 times that of the unconjugated 125I-alpha globin precursor for Ub1-alpha to approximately 6 times for Ub > 10-alpha. Analytical SDS-PAGE of the Ub1-alpha mixture also showed a rapid conversion to higher M(r) conjugates. These results suggest that monoubiquitinated alpha globin, in addition to higher M(r) conjugates, is an intermediate in the ATP-dependent proteolysis of alpha globin. The finding that addition of Ubal enhances the amount of these conjugates suggests that isopeptidase activity contributes to the poor turnover of hemoglobin alpha chains observed in intact beta-thalassemic reticulocytes.

ATP-dependent proteolysis of hemoglobin alpha chains in beta-thalassemic hemolysates is ubiquitin-dependent.

beta-Thalassemia is an inherited human disorder which is characterized by a deficient production of hemoglobin beta chains and an attendant accumulation of structurally normal alpha chains in the erythropoietic cells. The objective of this work is to understand the mechanism of intracellular proteolysis of these excess alpha chains. Dialyzed stroma-free hemolysates (32 mg/ml hemoglobin) of blood reticulocytes from four individuals with beta-thalassemia intermedia were incubated with human hemoglobin 3H-alpha chains (0.13 mg/ml) at 37 degrees C in a reaction mixture supporting protein degradation. In the presence of ATP and an ATP-generating system, the fraction of alpha chain 3H radioactivity made acid-soluble after 4 h ranged from 4 to 12% among the different hemolysates; in the absence of ATP or when hemolysates of normal human erythrocytes were used, only 1 to 2% of the 3H-alpha chains were degraded. It is likely that the ATP-dependent proteolysis of 3H-alpha chains in the beta-thalassemic hemolysates corresponds to the ATP-dependent turnover of newly synthesized soluble alpha chains in intact beta-thalassemic reticulocytes observed previously (Shaeffer, J. (1983) J. Biol. Chem. 258, 13172-13177) because of the following similarities between the two systems: (a) free 3H-alpha chains, but not 3H-labeled tetrameric hemoglobins, were readily degraded; (b) the rate of 3H-alpha chain proteolysis in the cell-free system was at least one-half of that observed for the turnover of newly synthesized alpha chains (t1/2 approximately 6 h) in intact cells; and (c) the ATP-dependent proteolytic activity of both systems was inhibited substantially by certain chemical agents (orthovanadate, N-ethylmaleimide, o-phenanthroline, and phenylmethylsulfonyl fluoride) but only slightly, if at all, by others (epsilon-aminocaproic acid and leupeptin). When excess human erythrocyte ubiquitin was added to the beta-thalassemic cell-free systems, a stimulation in ATP-dependent proteolysis of 3H-alpha chains ranging from 30 to 58% was observed. Conversely, addition of from 1.25 to 2.50 mg/ml affinity-purified rabbit antiubiquitin inhibited almost all (greater than 90%) of the ATP-dependent 3H-alpha chain proteolysis; in control experiments, antiubiquitin neutralized with excess ubiquitin inhibited only 13 to 30% of the total (including ubiquitin-stimulated) ATP-dependent proteolysis.(ABSTRACT TRUNCATED AT 400 WORDS)

Dissociation of dimers of human hemoglobins A and F into monomers.

Dissociation of alpha beta and alpha gamma dimers of human hemoglobins (Hb) A and F into monomers was studied by alpha chain exchange (Shaeffer, J. R., McDonald, M. J., Turci, S. M., Dinda, D. M., and Bunn, H. F. (1984) J. Biol. Chem. 259, 14544-14547). Unlabeled carbonmonoxy-Hb A was incubated with trace amounts of preparatively purified, native, 3H-alpha subunits in 10 mM sodium phosphate, pH 7.0, at 25 degrees C. At appropriate times, free alpha monomers were separated from Hb A tetramers by anion exchange high performance liquid chromatography. Transfer of radioactivity from the alpha chain pool into Hb A was measured, yielding a first order dimer dissociation rate constant, k2 = (3.2 +/- 0.3) X 10(-3) h-1. The Arrhenius plot of k2 was linear between 7 and 37 degrees C, yielding an enthalpy of activation of 23 kcal/alpha beta dimer. As the chloride concentration was raised from 0 to 0.2 M, the dissociation rate increased 3-fold; with higher salt concentrations, however, the rate gradually returned to baseline. This rate was not altered by raising the pH from 6.5 to 7.2, but as pH was further raised to 8.4, kappa 2 increased about 3-fold. Hb F, which has an increased stability at alkaline pH, dissociated into alpha and gamma monomers 3 times more slowly than Hb A. Moreover, the dimer-monomer dissociation of Hb F was characterized by a significantly reduced pH dependence. These results demonstrate that both alpha beta and alpha gamma dimers of Hb A and Hb F dissociate reversibly into monomers under physiologic conditions. The differential pH dependence for dimer dissociation between Hb A and Hb F suggests that specific amino acid replacement at the alpha 1 gamma 1 interface confers increased resistance to alkaline denaturation.

Dimer-monomer dissociation of human hemoglobin A.

Isolated 3H-labeled human hemoglobin alpha chains were incubated with unlabeled carbonmonoxyhemoglobin A for 72 h in 0.01 M potassium phosphate, pH 7.0, at 25 degrees C. Following separation of the free alpha chain monomers and Hb A (alpha 2 beta 2) tetramers by electrophoresis on cellulose acetate strips or by chromatography on DEAE-cellulose columns, analysis of radioactivity showed a slow transfer of 3H-labeled alpha chains into Hb A. Separation of the globin chains of the isolated [3H] Hb A on a CM-cellulose column in 8 M urea showed that the radioactivity was in structurally intact alpha chains. Gel filtration of a reaction mixture on Sephadex G-75 in the presence of 0.15 M NaCl showed that the 3H-labeled alpha-chains in Hb A were an integral part of, and not nonspecifically adsorbed to, the tetramer. These results indicate that the free 3H-labeled alpha chain monomers were incorporated into Hb A by exchange with pre-existing unlabeled alpha chain subunits according to the scheme, alpha 2 beta 2 in equilibrium 2 alpha beta in equilibrium 2 alpha + 2 beta. The rate constant for the dissociation of alpha beta dimers to alpha and beta monomers, the rate-limiting step in this exchange reaction, was 4.0 (+/- 0.7) X 10(-3) h-1.

Acetylation of human fetal hemoglobin occurs throughout erythroid cell maturation.

The biosynthesis of human acetylated fetal hemoglobin (Hb F1) has been examined by incubating the following cell types with [3H]leucine: (a) burst-forming unit erythroid cells cultured from umbilical cord mononuclear cells, (b) infant bone marrow, (c) umbilical cord blood, and (d) peripheral blood cells from adults with elevated fetal hemoglobin. Newly synthesized Hb F1 was 18-20% that of Hb F0 in burst-forming unit erythroid cells which were immature, mature, or in an intermediate state of development. In infant marrow and in infant and adult peripheral blood the extant Hb F1 comprised 10.8 +/- 1.8% of the total Hb F. In marrow cells the specific radioactivity (cpm/mg) of Hb F1 was 1.4-2.0-times greater than that of Hb F0. In peripheral blood cells these ratios were slightly greater. [3H]Leucine-labeled infant bone marrow, umbilical cord blood, and adult peripheral blood cells were subjected to density gradient ultracentrifugation. The ratios of specific radioactivity for Hb F1/Hb F0 increased from 1.0-1.8 in the lightest cell zone to 5.2-9.0 in the more dense cells. Thus the biosynthesis of Hb F1 is enhanced in cells which are more mature than those responsible for the bulk of hemoglobin synthesis, and the acetylation of Hb F provides a marker of erythroid cell maturation.

Turnover of excess hemoglobin alpha chains in beta-thalassemic cells is ATP-dependent.

Blood erythroid cells from five beta-thalassemic donors were incubated with [3H]leucine at 37 degrees C to label the pool of excess, free hemoglobin alpha chains. The 3H-labeled cells were split and reincubated in nonradioactive media that either supported ATP production (with 5 mM glucose) or inhibited ATP production (without glucose but with 20 mM 2-deoxyglucose and 0.20 mM 2,4-dinitrophenol). During the 6-h incubation in the glucose medium, the total cellular protein 3H radioactivity decreased about 30%, while the ATP levels remained constant. Chromatographic separation of the alpha- and non-alpha-globin chains of the crude (stroma included) lysates and electrophoretic separation of the free alpha chains and tetrameric hemoglobins of the stroma-free soluble phases both showed that degradation of the alpha chains was responsible for the decrease in protein 3H radioactivity. Conversely, in the energy-deprived cells, the ATP levels dropped to less than 10% of that of the energy-supported cells, and the turnover of alpha-globin 3H radioactivity of the crude lysates was only 5-10%. These results indicate that proteolysis of excess, newly synthesized alpha chains in beta-thalassemic cells is ATP-dependent. The accumulation, mostly in the stromal fraction, of intact 3H-alpha chains in the ATP-deprived cells suggests that an ATP-dependent step occurs early in the biochemical pathway of alpha chain proteolysis. Denaturation resulting in insolubility of the free alpha chains may be a recognition signal for activation of this proteolysis.

Subunit assembly of hemoglobins A and S.

It has been demonstrated that a difference in affinities of beta A and beta S chains for alpha chains exists. Furthermore, differential rates of dissociation of beta A and beta S chain tetramers have also been measured. It appears that the overall assembly of Hb A and Hb S, in vitro, can be governed by either the dissociation or combination reactions of its non-alpha-subunit.

Evidence for a difference in affinities of human hemoglobin beta A and beta S chains for alpha chains.

A limiting amount of isolated human hemoglobin alpha chains was incubated with a mixture containing excess, but equal, amounts of beta A and beta S chains to form hemoglobin S = alpha 2 beta 2S and hemoglobin = alpha 2 beta 2A, presumably by the reaction 2 alpha + 2 beta leads to 2 alpha beta leads to alpha 2 beta 2. The initial concentration of each type of beta chain was varied from levels (greater than microM) at which the tetrameric form = beta 4 was predominant to levels (less than 1 microM) at which the tetramers had dissociated to monomers. The initial relative concentration of alpha chains remained constant at one-twentieth (0.05) that of total beta chains. About 50% as much hemoglobin S as hemoglobin A was formed in each reaction despite the 500-fold range in the beta chain concentrations among the assays. These results suggest that the difference in amounts of hemoglobins S and A formed was caused by a difference in affinities of individual beta S and beta A chain monomers for alpha chains and not by a difference in the concentrations of beta S and beta A monomers resulting from putative unequal rates of dissociation of beta 4S and beta 4S tetramers.

Sickle cell--betao thalassemia variant with high hemoglobin F and mild clinical course.

A 70 year old Black woman had chronic hemolytic anemia without recurrent painful crises. Hemoglobin pattern by electrophoresis was hemoglobin S (69 to 71 per cent), hemoglobin A2 (4.6 per cent) and hemoglobin F (24 to 27 per cent). No hemoglobin A was detected, and the hemoglobin F was distributed heterogeneously in the red cells. Reticulocyte alpha/nonalpha globin chain synthetic ratios were 1.44 to 1.62. Thus, the patient had a high hemoglobin F variant of S-beta zero (betao) thalassemia which has not been described previously. Her clinical course has been mild in comparison with S-betao thalassemia patients who do not have extremely elevated hemoglobin F levels.

betaS Chain turnover in reticulocytes of sickle trait individuals with high or low concentrations of haemoglobin S.

Reticulocytes, isolated from the blood of sickle cell trait donors with either low (25-30%) or high (40-42%) haemoglobin S(Hb S) concentrations, were incubated with [3H]leucine for various times from 1.25 to 60 min. Samples of the total soluble fractions of the cells were denatured with urea and mercaptoethanol. The mixtures were analysed by electrophoresis on cellulose acetate strips. The specific radioactivities (dpm/mg) of the separated betaS and betaA globin chains were determined. The betaS/betaA ratios of globin chain specific radio activities in the reticulocytes of the 'low Hb S' donors decreased gradually from initial values higher than 1.30 to values near unity. These data suggested that faster turnover of some of the soluble, newly synthesized betaS chains compared to the newly synthesized betaA chains could explain part, but not all, of the disparity in concentrations of Hbs S and A in these people. When reticulocytes from 'high Hb S' donors were 3H-labelled for times longer than 5 min, the betaS/betaA specific radioactivity ratios remained at or near unity. This result suggested that newly synthesized betaS chains were not turning over selectively in these cells. Instead, there was a relative decrease in betaS chain synthesis proportional to the difference in blood concentrations of Hb S and Hb A. Additional calculations suggested that the more rapid turnover of newly synthesized betaS chains in the 'low Hb S' reticulocytes could explain the difference in Hb S concentrations between 'high and low Hb S' people. These results are consistent with previous reports that an alpha-thalassaemia gene, present in 'low Hb S' but absent in 'high Hb S' donors, may be responsible for the selective turnover of betaS chains.

Phosphorylation of ribosome-associated proteins during the mammalian cell cycle. Unique phosphorylation of a specific protein during mitosis.

Chinese hamster ovary cells in monolayer culture were incubated with [32P] phosphate. Ribosome-associated proteins, including both structural proteins and those tightly bound to washed, centrifuged ribosomes, were isolated and separated by electrophoresis on sodium dodecyl sulfate-polyacrylamide gels. The electrophoretic pattern showed five major regions or peaks of 32P radioactivity which represented phosphorylated ribosome-associated proteins with molecular weights of 17 500, 23 000, 30 000, 38 000, and 57 000. When asynchronous cells were pulse-labeled with [32P] phosphate, the predominant peak of 32P radioactivity was associated with the protein of 38 000 daltons. Similar results were obtained with cells synchronized in the G1, S, or G2 phase of the mammalian cell cycle. Conversely, proteins isolated from the ribosomes of mitotic cells, collected and labeled with [32P] phosphate in the presence of colcemid, showed a new and predominant peak of 32P radioactivity migrating with a protein of 45 000 daltons. When cells labeled in mitosis were allowed to progress into G1 phase, this peak of 32P radioactivity rapidly disappeared from the electrophoretic pattern. These results suggest that a specific protein associated with the ribosomes was phosphorylated uniquely during the mitotic phase of the cell cycle.

Hemoglobin synthesis studies of a family with alpha-thalassemia trait and sickle cell trait.

The ratio of total globin alpha to beta chain synthesis was determined in reticulocytes isolated from the blood of the members of a black family, some of whom had sickle cell trait with low blood HbS concentrations (25-30%). The results support the hypothesis that sickle cell trait individuals with low HbS concentrations also carry a gene for alpha-thalassemia.