Radiation-induced changes in the structure of erythrocyte membrane proteins

Structural changes in proteins of erythrocyte membranes induced by gamma-radiation at doses of 10-10(3) Gy were studied using the method of tryptophan fluorescence quenching by acrylamide. It was found that the exposure to ionizing radiation leads to a decrease in intramolecular dynamics of membrane proteins.     

Spin-label detection of hemoglobin-membrane interaction at physiological pH

The interaction between hemoglobin and the cytoplasmic surface of human erythrocyte membranes at physiological pH was studied by monitoring the electron paramagnetic resonance (EPR) signal of spin-labeled membrane ghosts in hemoglobin solutions of various concentrations. The EPR spectra indicate the existence of a significant hemoglobin-membrane interaction which exhibits a substantial hemoglobin concentration dependence over the concentration range 0-12 mg/mL. An equilibrium binding model yields a hemoglobin-membrane dissociation constant, Kd, on the order of 10(-4) M, at and above physiological pH; the interaction is classified as very low-affinity binding. The interaction increases significantly when the pH is decreased. Half-saturation of the binding sites occurs at a ratio of about 10(8) hemoglobins per cell.     

Interaction of hemoglobin with the red blood cell membrane. A saturation transfer electron paramagnetic resonance study

Human hemoglobin has been labeled on cysteine 93(beta) with the maleimide spin label, 3-maleimido-2,2,5,5-tetramethyl-1-pyrrolidinyloxyl and reassociated with erythrocyte membrane previously stripped of hemoglobin and glyceraldehyde-3-phosphate dehydrogenase. The affinity of hemoglobin for the membrane is not affected by the presence of the label. Saturation transfer electron paramagnetic resonance measurements show that the diffusion rotational movements of hemoglobin are considerably slowed down when it is bound to the erythrocyte membrane. The correlation time of rotation, tau c, is found to be 8 . 10(-6) s as compared with 2 . 10(-8) s when the hemoglobin molecule is in solution. The same values are obtained whether the protein is associated with its high- or low-affinity binding sites. They depend on the viscosity of the solution. The high-affinity sites are presumably located on the segment of the band 3 protein which extends into the cytoplasm and which links through ankyrin, the spectrin-actin cytoskeleton to the membrane. When band 3 is cross-linked into a dimer after reaction with the copper-ortho-phenanthroline chelate, the correlation time of rotation of spin-labelled hemoglobin is unchanged. It is also independent of the presence of the spectrin-actin network and ankyrin. These results show tha the movements of hemoglobin bound by ionic linkage to different part (protein or phospholipid) of the cytoplasmic surface of the membrane are similarly highly restricted by some potential or energetic barrier. They give also evidence for independent movements and flexibility in the assembly of the macromolecules which link the spectrin-actin cytoskeleton to the erythrocyte membrane.     

Susceptibility of erythrocytes from several animal species to Vibrio vulnificus hemolysin

The hemolytic activity of Vibrio vulnificus hemolysin (VVH) against erythrocytes from several animal species (sheep, horse, cow, rabbit, chicken) was investigated. VVH was active against erythrocytes from all species, but the amount of VVH causing 50% hemolysis under identical conditions (hemolytic susceptibility to VVH) differed. The degree of 125I-labeled VVH (125I-VVH) binding to each erythrocyte species correlated with the susceptibility of the cells to hemolysis. However, marked differences in the binding ability of 125I-VVH were not observed against liposomes constructed with lipids from each erythrocyte membrane. On the other hand, release of hemoglobin (Hb) differed for each of the erythrocyte species despite administration of approximately the same hemolytic VVH concentration to each species. Furthermore, under hypotonic conditions, the stability of each erythrocyte species varied markedly; the more susceptible the erythrocyte to VVH, the more unstable it was under such conditions. These results, therefore, suggest that the susceptibility of erythrocytes to VVH may be closely associated with the binding ability of VVH and erythrocyte membrane stability.     

The use of cis-parinaric acid to determine lipid peroxidation in human erythrocyte membranes. Comparison of normal and sickle erythrocyte membranes

The recently developed parinaric acid assay is shown to offer possibilities for studying peroxidation processes in biological membrane systems. Taking the human erythrocyte membrane as a model, several initiating systems were investigated, as well as the effect of residual hemoglobin in ghost membrane preparations. The effectivity of a radical generating system appeared to be strongly dependent upon whether radicals are generated at the membrane level or in the water phase. Thus, cumene hydroperoxide at concentrations of 1.0-1.5 mM was found to be a very efficient initiator of peroxidation in combination with submicromolar levels of hemin-Fe3+ as membrane-bound cofactor. In combination with cumene hydroperoxide, membrane-bound hemoglobin appeared to be about 6-times more effective in promoting peroxidation than hemoglobin in the water phase. Results comparing the behaviour of normal and sickle erythrocyte ghost suspensions in the peroxidation assay suggest that the increased oxidative stress on sickle erythrocyte membranes could be due to enhanced membrane binding of sickle hemoglobin, but also partly to a characteristically higher capability of sickle hemoglobin to promote peroxidation. The order of peroxidation-promoting capabilities that could be derived from the experiments was hemin greater than sickle hemoglobin greater than normal hemoglobin.     

Binding of prostaglandin E1 to human erythrocyte membrane

Prostaglandin E1 is known to alter the structural and functional characteristics of red blood cells, yet, little is understood about the membrane receptors mediating this process. We therefore studied the binding of tritium-labeled prostaglandin E1 to the intact human erythrocyte membrane and demonstrated that the interaction is highly specific, rapid, saturable and reversible. Scatchard analysis of prostaglandin E1 binding to the membrane preparations showed the presence of two independent classes of prostaglandin E1 binding sites which differed in their affinity for the autacoid. The high-affinity class had Kd = 3.6 X 10(-9) M and the low-affinity class had Kd = 5.6 X 10(-5) M. The optimum pH for the binding of [3H]prostaglandin E1 to the erythrocyte membrane was found to be around 7.5 and maximum specific binding occurred at a concentration of 5 mM Mg2+ in the incubation mixture. [3H]Prostaglandin E1 bound to the membrane preparation could not be displaced by GTP or by its stable derivative Gpp[NH]p. However, prostaglandin E1 bound to the erythrocyte membrane preparation could be rapidly displaced by cyclic AMP. The IC50 (concentration of the nucleotide displacing 50% bound [3H]prostaglandin E1 from the membrane) was 75 nM. Other adenine nucleotides or cyclic GMP could not substitute for cyclic AMP. Unlike the right-side-out erythrocyte membrane, the inside-out membrane preparations do not bind [3H]prostaglandin E1. Treatment of right-side-out erythrocyte membrane preparation with neuraminidase markedly decreases the binding of prostaglandin E1. Incubation of the erythrocyte membrane preparation with trypsin resulted in total loss of the binding activity. These results indicate that the prostaglandin E1 binding sites located on the cell surface and sialic acid residues are required for prostaglandin E1 binding to the human erythrocytes. These results also indicated that the binding sites are glycoprotein in nature.     

Distribution of insulin receptors in human erythrocyte membranes. Insulin binding to sealed right-side-out and inside-out human erythrocyte vesicles

Analyses of insulin binding to human erythrocytes and to resealed right-side-out and inside-out erythrocyte membrane vesicles have revealed that high affinity insulin binding receptors are present on both sides of the erythrocyte membranes. Insulin binding to human erythrocytes was examined with the use of a binding assay designed to minimize the potential errors arising from the low binding capacity of this cell type and from non-specific binding in the assay. Scatchard analysis of equilibrium binding to the cells revealed a class of high affinity sites with a dissociation constant (Kd) of (1.5 +/- 0.5) X 10(-8) M and a maximum binding capacity of 50 +/- 5 sites per cell. Interestingly, both resealed right-side-out and inside-out membrane vesicles exhibited nearly identical specific sites for insulin binding. At the high affinity binding sites, for both right-side-out and inside-out vesicles, the dissociation constant (Kd) was (1.5 +/- 0.5) X 10(-8) M, and the maximum binding capacity was 17 +/- 3 sites per cell equivalent. These findings suggest that insulin receptors are present on both sides of the plasma membrane and are consistent with the participation of the erythrocyte insulin receptors in an endocytic/recycling pathway which mediates receptor-ligand internalization/externalization.     

Identification of the hemoglobin binding sites on the inner surface of the erythrocyte membrane

The hemoglobin binding sites on the inner surface of the erythrocyte membrane were identified by measuring the fraction of hemoglobin released following selective proteolytic or lipolytic enzyme digestion. In addition, binding stoichiometry to and fractional hemoglobin release from inside-out vesicle preparations of human and rabbit membranes were compared since rabbit membranes differ significantly from human membranes only in that they lack glycophorin. Our results show that rabbit inside-out vesicles bind about 65% less human or rabbit hemoglobin under conditions of optimal and stoichiometric binding, despite being otherwise similar in composition. We suggest that this difference is either directly or indirectly due to the absence of glycophorin in rabbit membranes. Further supportive evidence includes demonstrating (a) that neuraminidase treatment of human membranes did not affect hemoglobin binding and (b) that reconstitution of isolated glycophorin into phospholipid vesicles increased the hemoglobin binding capacity in a manner proportional to the fraction of glycophorin molecules oriented with their cytoplasmic sides exposed to the exterior of the vesicle. Proteolysis of human inside-out vesicles either before or after addition of hemoglobin reduced the binding capacity by about 25%. This is consistent with the known proportion of total hemoglobin binding sites involving band 3 protein and the selective lability of the cytoplasmic aspect of band 3 protein to proteolysis. Phospholipid involvement in hemoglobin binding was determined using various phospholipase C preparations which differ in their reactivity profiles. Approximately 38% of the bound hemoglobin was released upon cleavage of phospholipid headgroups. These results suggest that the predominant sites of binding for hemoglobin on the inner surface of the red cell membrane are the two major integral membrane glycoproteins.     

Hemoglobin degradation, lipid peroxidation, and inhibition of Na+/K(+)-ATPase in rat erythrocytes exposed to acrylonitrile

The effect of acrylonitrile (VCN) on erythrocyte lipid metabolism was investigated in vitro in metabolically active red cells from male Sprague-Dawley rats containing three types of hemoglobins: oxyhemoglobin, methemoglobin, and carbon monoxyhemoglobin. VCN at the concentration of 10 mM rapidly depleted erythrocyte glutathione (GSH) (75% of control) and induced lipid peroxidation (274% of control). Degradation of oxy- and methemoglobin was directly proportional to the extent of lipid peroxidation (r = 0.89). Addition of glucose to the incubation medium decreased hemoglobin degradation while it slightly increased VCN-induced lipid peroxidation. The highest amount of lipid peroxidation occurred in erythrocytes containing carbon monoxyhemoglobin and glucose. In the isolated red cell membranes incubated with 10 mM VCN, the lipid peroxidation was 400% of controls. VCN (25 mM) noncompetitively inhibited erythrocyte membrane Na+/K(+)-ATPase activity and the degree of inhibition was inversely proportional to the reaction temperature (r = -0.88). These findings indicate that the VCN induced hemoglobin degradation and lipid peroxidation are two extremes of a spectrum of oxidative damage in red cells leading to a change in physical state of membrane structure causing inhibition of adenosine triphosphate (ATPase) activity.     

Interaction of hemoglobin with phospholipid vesicular membranes of different composition

Incorporation of hemoglobin into vesicles and its oxidation were studied as a function of composition of the vesicular membrane containing erythrocyte membrane lipids as main components. The addition of negatively charged lipids such as phosphatidylserine and phosphatidic acid, was shown to considerably increase the extent of hemoglobin binding, while sphingomyelin did not produce such an effect. Cholesterol and dipalmitoylphosphatidylcholine stabilized oxyHb.     

The influence of ferrylhemoglobin and methemoglobin on the human erythrocyte membrane

The aim of the study was to examine and compare the effects of methemoglobin (metHb) and ferrylhemoglobin (ferrylHb) on the erythrocyte membrane. Kinetic studies of the decay of ferrylhemoglobin (*HbFe(IV)=O denotes ferryl derivative of hemoglobin present 5 min after initiation of the reaction of metHb with H(2)O(2); ferrylHb) showed that autoredecay of this derivative is slower than its decay in the presence of whole erythrocytes and erythrocyte membranes. It provides evidence for interactions between ferrylHb and the erythrocyte membrane. Both hemoglobin derivatives induced small changes in the structure and function of the erythrocyte membrane which were more pronounced for ferrylHb. The amount of ferrylHb bound to erythrocyte membranes increased with incubation time and, after 2 h, was twice that of membrane-bound metHb. The incubation of erythrocytes with metHb or ferrylHb did not influence osmotic fragility and did not initiate peroxidation of membrane lipids in whole erythrocytes as well as in isolated erythrocyte membranes. Membrane acetylcholinesterase activity increased by about 10% after treatment of whole erythrocytes with both metHb and ferrylHb. ESR spectra of membrane-bound maleimide spin label demonstrated minor changes in the conformation of label-binding proteins in ferrylHb-treated erythrocyte membranes. The fluidity of the membrane surface layer decreased slightly after incubation of erythrocytes and isolated erythrocyte membranes with ferrylHb and metHb. In whole erythrocytes, these changes were not stable and disappeared during longer incubation.     

Erythrocyte membrane antigen expression during friend cell differentiation: Analysis of two non-inducible variants

Erythrocyte membrane antigens have been detected on induced Friend erythroleukemic cells with a rabbit antiserum raised against mouse erythrocyte membranes. The antibody specificities of this antiserum have been quantitatively analyzed using a cellular radioimmunoassay. After absorption with thymocytes, the rabbit anti-erythrocyte membrane serum bound to dimethylsulfoxide (DMSO)-induced Friend erythroleukemic cells and to mouse erythrocytes but not to uninduced Friend cells or thymocytes. Reciprocal inhibition studies demonstrated that, following complete thymocyte absorption, the antiserum detected similar antigenic specificities, termed erythrocyte membrane antigens (EMA), on both mature erythrocytes and induced Friend cells. The expression of these erythrocyte membrane antigens was also induced on Friend cells by other agents, such as ouabain and dimethylacetamide (DMA). In contrast, exogenous hematin, which did not induce hemoglobin synthesis in the Friend cell clones used in this study, also did not induce erythrocyte membrane antigen expression. Two independently derived variant clones which do not produce hemoglobin in reponse to DMSO were analyzed for their ability to produce erythrocyte membrane antigens in response to various inducers of Friend cell differentiation. Clone TG-13 is not inducible by DMSO or hematin but is weakly induced by DMA for both hemoglobin production and erythrocyte membrane antigen expression. Another variant clone, M18, was also analyzed. This clone does not synthesize detectable hemoglobin when grown in either DMSO or hematin alone, but undergoes extensive hemoglobin synthesis when grown in medium containing both DMSO and hematin. M18 does, however, express erythrocyte membrane antigens when grown in DMSO alone: the presence of hematin and DMSO together in the growth medium does not enhance expression of these antigens. Thus M18 appears to be defective for hemoglobin inducibility, and this defect can be overcome by exogenous hematin; however, the expression of erythrocyte membrane antigens is not affected by this block in hemoglobin synthesis. The results with the variant clones are discussed in terms of a program for Friend cell differentiation in which the induction of hemoglobin synthesis and erythrocyte membrane antigen expression are under both co-ordinate and separate controls.     

Classification and localization of hemoglobin binding sites on the red blood cell membrane.

The binding of hemoglobin to the red cell membrane was characterized over a wide range of free hemoglobin concentrations by measurement of membrane bound and supernatant hemoglobin. Scatchard analysis of the binding data revealed two classes of sites: high affinity sites with a binding constant of 1 X 10(8) M-1 and 1.2 X 10(6) sites per cell, and a second, low affinity class of sites with a binding constant of 6 X 10(6)M-1 and 6 X 10(6) sites per cell. The low affinity sites are shown to be nonspecific and appear to be a result of the ghost preparation. The high affinity sites are shown to be specific to the inner surface of the red cell membrane. The competition of hemoglobin and glyceraldehyde-3-phosphate dehydrogenase suggests band III proteins as a potential binding site for hemoglobin.     

Hemoglobin degradation lipid peroxidation,and inhibition of na+/k+-atpase in rat erythrocytes exposed to acrylonitrile

The effect of acrylonitrile (VCN) on erythrocyte lipid metabolism was investigated in vitro in metabolically active red cells from male Sprague-Dawley rats containing three types of hemoglobins: oxyhemoglobin, methemoglobin, and carbon monoxyhemoglobin. VCN at the concentration of 10 mM rapidly depleted erythrocyte glutathione (GSH) (75% of control) and induced lipid peroxidation (274% of control). Degradation of oxy- and methemoglobin was directly proportional to the extent of lipid peroxidation (r = 0.89). Addition of glucose to the incubation medium decreased hemoglobin degradation while it slightly increased VCN-induced lipid peroxidation. The highest amount of lipid peroxidation occurred in erythrocytes containing carbon monoxyhemoglobin and glucose. In the isolated red cell membranes incubated with 10 mM VCN, the lipid peroxidation was 400% of controls. VCN (25 mM) noncompetitively inhibited erythrocyte membrane Na⁺/K⁺-ATPase activity and the degree of inhibition was inversely proportional to the reaction temperature (r = ?0.88). These findings indicate that the VCN induced hemoglobin degradation and lipid peroxidation are two extremes of a spectrum of oxidative damage in red cells leading to a change in physical state of membrane structure causing inhibition of adenosine triphosphatase (ATPase) activity.     

Binding of hemoglobin to red cell membranes with eosin-5-maleimide-labeled band 3: analysis of centrifugation and fluorescence data

We have studied the binding of hemoglobin to the red cell membrane by centrifugation and fluorescence methods. The intact red cell was labeled with eosin-5-maleimide (EM), which specifically reacts with lysine 430 of band 3. Even though this residue is not part of the cytoplasmic domain of band 3 (cdb3) associated with hemoglobin binding, fluorescence quenching was observed when hemoglobin bound to inside-out vesicles (IOVs). The use of fluorescence quenching to measure band 3 binding was quantitatively compared with the binding determined by centrifugation, which measures binding to band 3 and non-band 3 sites. For the centrifugation it was necessary to include the non-band 3 association constants determined from chymotrypsin-treated IOVs. The binding of hemoglobin to band 3 was interpreted in terms of the binding of two hemoglobin tetramers to each band 3 dimer. An anticooperative interaction associated with the conformational change produced when hemoglobin binds results in a 2.8-fold decrease in the intrinsic constant of (1.54 +/- 0.25) x 10(7) M(-1) for the binding of the second hemoglobin molecule. From the changes in lifetime produced by binding the first and second hemoglobin molecules, it was possible to show that the conformational change associated with binding the second hemoglobin molecule results in a decrease of the heme-eosin distance from 47.90 to 44.78 A. Reaction of cyanate with the alpha-amino group of hemoglobin (HbOCN) is shown to produce a very dramatic decrease in the binding of hemoglobin to both the band 3 and non-band 3 sites. The intrinsic constant for binding the first hemoglobin molecule to band 3 decreases by a factor of 29 to (5.34 +/- 0.15) x 10(5) M(-1). The anticooperative interaction is greater with the intrinsic constant decreasing by a factor of 3.8 for the binding of the second hemoglobin tetramer to band 3. In addition, the nature of the conformational change produced by binding hemoglobin is very different with the second HbOCN increasing the heme-eosin distance to 55.99 A. The utilization of eosin-5-maleimide-reacted red cell membrane to study hemoglobin binding makes it possible to directly study the binding to band 3. At the same time a sensitive probe of the conformational changes, which occur when hemoglobin binds to band 3, is provided.     

Adenosine 3',5' monophosphate binds only to the inner surface of human erythrocyte membranes

The binding of adenosine 3′,5′ monophosphate (cyclic AMP) to each surface of the isolated human erythrocyte membrane was measured. Unsealed ghosts, in which both membrane faces are accessible, and sealed inside-out vesicles, which expose only the cytoplasmic side of the membrane, both bound approximately 6,000 cyclic AMP molecules per cell membrane equivalent with a dissociation constant, K ? 2.5 × 10^?9. The binding of this nucleotide by preparations rich in sealed ghosts and right-side-out vesicles, which sequester the inner surface, was limited and could be correlated precisely with small amounts of exposed cytoplasmic surface. We conclude that these binding sites for cyclic AMP are confined to the cytoplasmic side of the erythrocyte membrane.     

The interaction of phenyldichloroarsine with erythrocytes

The purpose of the study was to identify binding sites of organic arsenic in the erythrocyte and to explain species differences in binding. Washed erythrocytes were exposed to graded concentrations of [U-14C]phenyldichloroarsine (PDA) in phosphate-buffered saline containing 0.1% glucose and 0.1% bovine serum albumin. At low PDA concentrations, all cells bound the arsenical rapidly (within 10 min) and quantitatively. Human, pig, hamster, guinea pig, and mouse erythrocytes approached saturation at 0.02-0.3 mumol PDA/10(9) cells, depending on the species. Saturation points correlated well with each respective species' erythrocyte glutathione content. In contrast, rat erythrocytes showed no sign of saturation at PDA loads as high as 3.0 mumol/10(9) cells. Hemolysates of PDA-treated erythrocytes were subjected to Sephadex G-75 gel filtration chromatography. 14C from rat hemolysate was distributed between the hemoglobin and small molecular weight (glutathione-containing) fractions. In all other species, the 14C eluted almost exclusively with the glutathione-containing fractions. In equilibrium dialysis experiments, human hemoglobin did not bind PDA, whereas rat hemoglobin bound 2 PDA/mol with Kd approximately 5 microM. In conclusion, glutathione is the principal binding site of phenyldichloroarsine in erythrocytes. In most species, the arsenical does not bind to hemoglobin, even though it has free (titratable) sulfhydryls considerably in excess of the glutathione concentration. In rat erythrocytes, phenlydichloroarsine binds both to glutathione and to hemoglobin. Arsenical binding by rat hemoglobin is presumably due to the unique location of the extra titratable cysteine in that protein.     

Effect of hemoglobin A and S on human erythrocyte ghosts

The interaction of erythrocyte ghosts and vesicles with chromatographed hemoglobin (Hb) A and Hb S was studied under various conditions. Although no binding of either Hb A or Hb S to inside-out vesicles was detected, under conditions of physiological ionic strength and pH, several properties of white membrane ghosts were effected by the presence of Hb. Addition of Hb A and Hb S (2 g/dl) to membrane ghosts in 6 mM MgATP, 150 mM NaCl, 10 mM Tris-HCl buffer, pH 7.4, was found to effect the echinocyte-discocyte transition, the extent of endocytosis, the volume, and the sealing of ghosts. Our observations suggest that the structure of membrane ghosts is influenced by cytosol proteins and that the environment of the red cell membrane plays an important role in the definition and the control of the membrane structure and function.     

The influence of ferrylhemoglobin and methemoglobin on the human erythrocyte membrane

Abstract

The aim of the study was to examine and compare the effects of methemoglobin (metHb) and ferrylhemoglobin (ferrylHb) on the erythrocyte membrane. Kinetic studies of the decay of ferrylhemoglobin (^*HbFe(IV)=O denotes ferryl derivative of hemoglobin present 5 min after initiation of the reaction of metHb with H₂O₂; ferrylHb) showed that autoredecay of this derivative is slower than its decay in the presence of whole erythrocytes and erythrocyte membranes. It provides evidence for interactions between ferrylHb and the erythrocyte membrane. Both hemoglobin derivatives induced small changes in the structure and function of the erythrocyte membrane which were more pronounced for ferrylHb. The amount of ferrylHb bound to erythrocyte membranes increased with incubation time and, after 2 h, was twice that of membrane-bound metHb. The incubation of erythrocytes with metHb or ferrylHb did not influence osmotic fragility and did not initiate peroxidation of membrane lipids in whole erythrocytes as well as in isolated erythrocyte membranes. Membrane acetylcholinesterase activity increased by about 10% after treatment of whole erythrocytes with both metHb and ferrylHb. ESR spectra of membrane-bound maleimide spin label demonstrated minor changes in the conformation of label-binding proteins in ferrylHb-treated erythrocyte membranes. The fluidity of the membrane surface layer decreased slightly after incubation of erythrocytes and isolated erythrocyte membranes with ferrylHb and metHb. In whole erythrocytes, these changes were not stable and disappeared during longer incubation.     

Affinity of hemoglobin for the cytoplasmic fragment of human erythrocyte membrane band 3. Equilibrium measurements at physiological pH using matrix-bound proteins: the effects of ionic strength, deoxygenation and of 2,3-diphosphoglycerate

The cytoplasmic fragment of band 3 protein isolated from the human erythrocyte membrane was linked to a CNBr-activated Sepharose matrix in an attempt to measure, in batch experiments, its equilibrium binding constant with oxy- and deoxyhemoglobin at physiological pH and ionic strength values and in the presence or the absence of 2,3-diphosphoglycerate. All the experiments were done at pH 7.2, and equilibrium constants were computed on the basis of one hemoglobin tetramer bound per monomer of fragment. In 10 mM-phosphate buffer, a dissociation constant KD = 2 X 10(-4)M was measured for oxyhemoglobin and was shown to increase to 8 X 10(-4)M in the presence of 50 mM-NaCl. Association could not be demonstrated at higher salt concentrations. Diphosphoglycerate-stripped deoxyhemoglobin was shown to associate more strongly with the cytoplasmic fragment of band 3. In 10 mM-bis-Tris (pH 7.2) and in the presence of 120 mM-NaCl, a dissociation constant KD = 4 X 10(-4)M was measured. Upon addition of increasing amounts of 2,3-diphosphoglycerate, the complex formed between deoxyhemoglobin and the cytoplasmic fragment of band 3 was dissociated. On the reasonable assumption that the hemoglobin binding site present on band 3 fragment was not modified upon linking the protein to the Sepharose matrix, the results indicated that diphosphoglycerate-stripped deoxyhemoglobin or partially liganded hemoglobin tetramers in the T state could bind band 3 inside the intact human red blood cell.