Search

Band 3 is the predominant membrane-spanning polypeptide and the mediator of anion transport in the human erythrocyte. In addition, it provides the sites of association for fructose 1,6-bisphosphate aldolase and other cytoplasmic proteins with the membrane. The aldolase-binding activity of water-soluble fragments of band 3 was measured by their inhibition of aldolase catalytic activity and by their displacement of aldolase from ghosts. At saturation, the binding of one band 3 or certain of its fragments per aldolase molecule partially inhibited the catalytic activity and band 3 binding of the unliganded subunits of the tetramer through an apparently cooperative mechanism. An NH2-terminal 23,000-dalton fragment generated by S-cyanylation of the cytoplasmic pole of band 3 was approximately 20% as avid in binding aldolase as was native band 3. Several fragments cleaved from the NH2-terminal portion of the 23,000-dalton peptide by trypsin, mild acid hydrolysis, and cyanogen bromide digestion all bound aldolase, while fragments from the rest of the polypeptide were essentially inactive. The first 31 residues of band 3 contained 16 Asp plus Glu, no basic residues, and a blocked alpha-amino terminus. The highly acidic composition of this region is consistent with the strongly electrostatic character of the interaction between band 3 and aldolase, presumably at the strongly basic catalytic center of the enzyme. We conclude that the NH2-terminal region of band 3 bears the membrane-binding site for aldolase.

In order to examine the widely held hypothesis that the reticulum of proteins which covers the cytoplamsic surface of the human erythrocyte membrane controls cell stability and shape, we have assessed some of its properties. The reticulum, freed of the bilayer by extraction with Triton X-100, was found to be mechanically stable at physiological ionic strength but physically unstable at low ionic strength. The reticulum broke down after a characteristic lag period which decreased 500-fold between 0 degrees and 37 degrees C. The release of polypeptide band 4.1 from the reticulum preceded that of spectrin and actin, suggesting that band 4.1 might stabilize the ensemble but is not essential to its integrity. The time-course of breakdown was similar for ghosts, the reticulum inside of ghosts, and the isolated reticulum. However, at very low ionic strength, the reticulum was less stable within the ghost than when free; at higher ionic strength, the reverse was true. Over a wide range of conditions the membrane broke down to vesicles just as the reticulum disintegrated, presumably because the bilayer was mechanically stabilized by this network. The volume of both ghosts and naked reticula varied inversely and reversibly with ionic strength. The volume of the naked reticulum varied far more widely than the ghost, suggesting that its deformation was normally limited by the less extensible bilayer. The contour of the isolated reticulum was discoid and often dimpled or indented, as visualized in the fluorescence microscope after labeling of the ghosts with fluoroscein isothiocyanate. Reticula derived from ghosts which had lost the ability to crenate in isotonic saline were shriveled, even though the bilayer was smooth and expanded. Conversly, ghosts crenated by dinitrophenol yielded smooth, expanded reticula. We conclude that the reticulum is a durable, flexible, and elastic network which assumes and stabilizes the contour of the membrane but is not responsible for its crenation.

We have isolated 5 families of proteins from human red blood cell membranes and characterized their secondary structure by ultraviolet circular dichroism measurements. The protein families were prepared by selective solubilization from ghosts under nondenaturing conditions. We find that the intact ghost has a mean alpha-helix fraction of 0.37, whereas a low-ionic-strength extract (bands 1, 2, 5, "spectrin") has a substantially higher helix fraction, 0.55. Further extraction of the ghosts with para-chloromercuribenzoate yields bands 2.1, 4.1, 4.2, and 6; their helix content is only 0.17. Finally, the major intrinsic protein, band 3, was solubilized by a non-ionic detergent. Its helix fraction is 0.38.

Band 3, the anion transport protein of the human erythrocyte, provides the site of association of certain glycolytic enzymes with the membrane. We have now demonstrated that glyceraldehyde-3-P dehydrogenase is inhibited, reversibly and completely, when membrane bound. The inhibition was competitive with respect to NAD+ and arsenate, but was noncompetitive with glyceraldehyde-3-P. Peptide fragments containing the NH2-terminal 23 residues of band 3 also inhibited the enzyme and displaced it from ghosts. Thus, the red cell membrane binding site for glyceraldehyde-3-P dehydrogenase is the same as that for aldolase, the polyanionic NH2-terminal region of the band 3 polypeptide.

We have determined the amino acid sequence of the N alpha-terminal portion of band 3, the anion transport protein of the human erythrocyte membrane. The material analyzed was a 201-residue, 23,053-Da fragment cleaved from the cytoplasmic end of band 3 by S-cyanylation. The sequence had these notable features. 1) The N alpha-terminal region was extraordinarily acidic, second only to a segment of similar size from the sigma factor of Escherichia coli RNA polymerase. The first 33 residues contained 6 aspartic acid and 12 glutamic acid residues, no basic residue, and a blocked N alpha-amino group. 2) The first 11 residues of the protein had a striking resemblance to the following 11 residues. 3) In contrast to the acidic N alpha-terminal third, the COOH-terminal two-thirds of the 23,053-Da fragment had a predominantly basic character. The highly acidic character of the N alpha-terminal portion of band 3 accounts for the capacity of this part of the protein to bind glycolytic enzymes in a highly electrostatic fashion, presumably through interaction with their cationic substrate-binding sites.

Aldolase is a trace protein in isolated human red cell membrane preparations. Following total elution of the endogenous enzyme by a saline wash, the interaction of this membrane with rabbit muscle aldolase was studied. At saturation, exogenous aldolase constituted over 40% of the repleted membrane protein. Scatchard analysis revealed two classes of sites, each numbering approximately 7 X 10(5) per ghost. Specificity was suggested by the exclusive binding of the enzyme to the membrane's inner (cytoplasmic) surface. Furthermore, milimolar levels of fructose 1,6-bisphosphate eluted the enzyme from ghosts, while fructose 6-phosphate and NADH (a metabolite which elutes human erythrocyte glyceraldehyde-3-phosphate dehydrogenase (G3PD) from its binding site) were ineffectuve. Removing peripheral membrane proteins with EDTA and lithium 3,5-diiodosalicylate did not diminish the binding capacity of the membranes. An aldolase-band 3 complex, dissociable by high ionic strength or fructose 1,6-bisphosphate treatment, was demonstrated in Triton X-100 extracts of repleted membranes by rate zonal sedimentation analysis on sucrose gradients. We conclude that the association of rabbit muscle aldolase with isolated human erythrocyte membranes reflects its specific binding to band 3 at the cytoplasmic surface, as is also true of G3PD.