Alteration of the genes for glycophorin A and B in glycophorin-A-deficient individuals

Glycophorins A and B are homologous glycoproteins of the red cell membrane which carry the blood-group MN and Ss antigens, respectively, and are encoded by two distinct genes closely linked on chromosome 4, which are probably derived from each other by duplication during evolution. The lack of glycophorin A is associated with the rare phenotype En(a-), indicating individuals who are defective for MN antigens, as well as for the Ena antigens, also located on this glycoprotein. The En(a-) condition is heterogenous and includes two categories of variants exemplified by the Finnish and the English types referred to as En(Fin) and En(UK), respectively. By Southern blot and preliminary genomic clone analyzes we have compared the status of the genes for glycophorins A and B, as well as that of the gene encoding glycophorin C, another unrelated red cell membrane glycoprotein, in the En(a-) variants and in the En(a+) control donors. Our data indicate that the En(Fin) variant is homozygous for a complete deletion of the glycophorin A gene without any detectable abnormality of the genes encoding glycophorins B or C. In the genome of the En(UK) variant, with the presumed genotype Mk/En(UK), and where the Mk condition abolishes the expression of MN and Ss antigens, we have identified several abnormalities of the glycophorin A and B genes, but the glycophorin C gene was unaffected. Our results strongly support the view that in Mk chromosome the glycophorin A and B genes are largely deleted, whereas the En(UK) chromosome probably contains a gene fusion product encoding a hybrid glycoprotein AM-B, composed of the N-terminal portion of a blood group M-type glycophorin A and of the C-terminal portion of glycophorin B. The determination of the 5' and 3' limits of the hybrid gene and elucidation of the mechanism involved will require sequencing of the rearranged DNA of the variant and a full knowledge of the organization of the glycophorin A and B genes.     

Rearrangements of the red-cell membrane glycophorin C (sialoglycoprotein beta) gene. A further study of alterations in the glycophorin C gene.

We have cloned portions of the glycophorin C (sialoglycoprotein beta) gene from individuals with red cells of normal, Gerbich and Yus phenotypes. The clones contain up to three exons of the glycophorin C gene (designated exons 2, 3 and 4). Analysis by restriction mapping and DNA sequencing confirmed that the deletions causing the Gerbich and Yus phenotypes are located entirely within the glycophorin C gene. Sequencing of the normal gene showed that not only do exon 2 and exon 3 have related DNA sequences, but also that both the 5' and 3' flanking intronic DNA sequences are almost identical. The two variant genes each lack a different exon: the Yus type gene lacks exon 2, whereas the Gerbich-type gene lacks exon 3. We suggest that the observed deletions are due to recombination between the regions of homologous intronic repeats. We also provide evidence that an unequal cross-over mechanism may be responsible for a number of observed glycophorin C gene rearrangements, including an insertion mutation in Lewis II (Lsa)-type red cells that has not previously been reported.     

The identification of specific Rhesus-polypeptide-blood-group-ABH-active-glycoprotein complexes in the human red-cell membrane.

1. RhD,c and E immune complexes isolated from 3H- and 125I-surface-radiolabelled and unlabelled intact human red cells were analysed by SDS/polyacrylamide-gel electrophoresis. 2. Apparent Mr values of 31,900 for RhD polypeptide and 33,100 for Rhc,E polypeptide were obtained under both reducing and non-reducing conditions. Glycosylation of RhD,c and E polypeptides was not detected. 3. RhD,c and E immune complexes also contain a glycoprotein component. RhD glycoprotein (apparent Mr 45,000-100,000) is distinct from Rhc,E glycoprotein(s) (apparent Mr 35,000-65,000). Rh (Rhesus) glycoprotein carbohydrate moieties are susceptible to endo-beta-galactosidase digestion and carry blood-group-ABH determinants. This suggests the presence of polylactosaminoglycan-type structures. 4. Rh glycoproteins are not present in Rh immune complexes as a result of non-specific adsorption of membrane glycoproteins during the membrane-solubilization phase of immune-complex isolation because RhD immune complexes isolated from a 1:1 (v/v) mixture of Acde/cde and OcDE/cDE red cells do not contain blood-group-A-active glycoprotein. 5. Blood-group-A immune complexes isolated from group-A red cells of the appropriate Rh phenotypes contain the 31,900- and 33,100-apparent-Mr Rh polypeptides. 6. It was concluded from the above evidence that non-covalent Rh-glycoprotein-Rh-polypeptide complexes exist in the native red-cell membrane. 7. The 31,900- and 33,100-apparent-Mr Rh polypeptides are absent from blood-group-A immune complexes isolated from regulator type Rhnull cells (donor A.L.), but are replaced by a 33,800-apparent-Mr Rhnull-specific polypeptide (Rhnull polypeptide). It is suggested that Rhnull polypeptide is an aberrant product of the Rh gene complex.     

Subunit structure of human erythrocyte glycophorin A.

Glycophorin A is a sialoglycoprotein isolated from human erythrocyte membranes which seems to exist as stable dimeric complexes in the presence of sodium dodecyl sulfate. When analyzed by dodecyl sulfate acrylamide electrophoresis this molecule forms two PAS-stainable bands (PAS-U and PAS-2) which are reversibly interconvertible. This change in electrophoretic mobility is dependent on the concentration of dodecyl sulfate, the use of Trisbuffer systems, the protein concentration in the incubation mixture, and the duration and temperature of incubation before electrophoresis. Reducing agents do no influence the results. Chromatography of the sialoglycopeptides on Sepharose columns in dodecyl sulfate before and after heat treatment gave similar results. A small hydrophobic peptide (T-6) derived from glycophorin A was able to prevent reassociation of the monomeric subunits back to the higher molecular weight form. This peptide was able to bind to the subunit of glycophorin A, but not to the high molecular weight complex. These results are consistent with a model of glycophorin A composed of two subunits which can dissociate and reassociate in the presence of detergents. These subunits may interact via the hydrophobic portions of the polypeptide chains.     

A novel gene member of the human glycophorin A and B gene family. Molecular cloning and expression

A new gene closely related to the glycophorin A (GPA) and glycophorin B (GPB) genes has been identified in the normal human genome as well as in that of persons with known alterations of GPA and/or GPB expression. This gene, called glycophorin E (GPE), is transcribed into a 0.6-kb message which encodes a 78-amino-acid protein with a putative leader peptide of 19 residues. The first 26 amino acids of the mature protein are identical to those of M-type glycophorin A (GPA), but the C-terminal domain (residues 27-59) differs significantly from those of glycophorins A and B (GPA and GPB). The GPE gene consists of four exons distributed over 30 kb of DNA, and its nucleotide sequence is homologous to those of the GPA and GPB genes in the 5' region, up to exon 3. Because of branch and splice site mutations, the GPE gene contains a large intron sequence partially used as exons in GPA and GPB genes. Compared to its counterpart in the GPB gene, exon 3 of the GPE gene contains several point mutations, an insertion of 24 bp, and a stop codon which shortens the reading frame. Downstream from exon 3, the GPE and the GPB sequences are virtually identical and include the same Alu repeats. Thus, it is likely that the GPE and GPB genes have evolved by a similar mechanism. From the analysis of the GPA, GPB and GPE genes in glycophorin variants [En(a-), S-s-U- and Mk], it is proposed that the three genes are organized in tandem on chromosome 4. Deletion events within this region may remove one or two structural gene(s) and may generate new hybrid structures in which the promoter region of one gene is positioned upstream from the body of another gene of the same family. This model of gene organization provides a basis with which to explain the diversity of the glycophorin gene family.     

Coisolation of glycophorin A and polyphosphoinositides from human erythrocyte membranes.

The lipid composition of purified erythrocyte membrane glycophorin was measured. Diphosphoinositide, triphosphoinositide, and phosphatidylserine are the major phospholipids in glycophorin preparation. Nearly all of the radioactive diphosphoinositide and triphosphoinositide extracted from erythrocyte membranes by lithium d?odosalicylate are recoverd in purified glycophorin. There appeared to be no significant enrichment of other acidic membrane phospholipids in the protein. The results do not permit a firm conclusion as to whether the polyphosphoinositides are associated specifically with the membrane protein or whether fortuitous binding has occurred during purification.     

Preparation and properties of human red-cell ankyrin.

We describe a procedure for the preparation of ankyrin from human red cells with a yield of 2-3 mg of protein from 30 ml of packed cells. This represents an improvement of an order of magnitude over the usual earlier procedure. Moreover, the product is, in our hands, much more stable against adsorption and proteolysis, and can in general be stored for at least 2 months at 4 degrees C without significant decrease in concentration and binding activity. The preparation depends on the release of the ankyrin-band-3 complex from the membrane cytoskeleton when intact cells are lysed in a medium containing concentrated Triton X-100. The complex is dissociated at high ionic strength, and the final purification is achieved by gel filtration in a medium containing 2 M-Tris or 0.6 M-NaBr. The ankyrin contains all the progression of components present in the intact membrane. All react with affinity-purified polyclonal anti-ankyrin antibodies, and all give widely similar patterns of peptides in partial proteolytic digests. The ankyrin is fully active, as judged by its capacity to bind to band-3-containing membrane vesicles and to Sepharose-coupled spectrin. All components bind to the membrane vesicles. Purified components 2.1 and 2.2, as well as the calmodulin-binding cytoskeletal constituent adducin, can be isolated in pure form by a single anion-exchange column step.     

Evolution of glycophorin A in the hominoid primates studied with monoclonal antibodies, and description of a sialoglycoprotein analogous to human glycophorin B in chimpanzee

Comparison of human and primate erythrocyte membrane sialoglycoproteins showed that common chimpanzee, dwarf chimpanzee, gorilla, orangutan, and gibbon have major periodic acid Schiff-positive proteins resembling human glycophorin A (GPA) monomer and dimer in electrophoretic mobility on sodium dodecyl sulfate-polyacrylamide gels. Immunoperoxidase staining of Western blots with monoclonal antibodies to human GPA showed that these primate bands express some GPA antigenic determinants. A new sialoglycoprotein analogous to human glycophorin B (GPB) was detected in common chimpanzee. Although human MN blood group phenotype results from an amino acid polymorphism of GPA, Western blots showed that in chimpanzee sialoglycoprotein (GPAch) always expresses the M blood group, whereas chimpanzee sialoglycoprotein (GPBch) expresses either the N blood group or a null phenotype. This result explains the detection of M and MN, but not of N, blood group phenotypes in chimpanzee. GPBch has higher apparent m.w. than human GPB, is present in the erythrocyte membrane in greater quantity than human GPB, and contains trypsin cleavage site(s) and the 10F7 determinant (both found on human GPA but not GPB). Expression of human GPA antigenic determinants was consistent with the phylogeny of the hominoid primates; common and dwarf chimpanzee expressed most of the determinants tested, gorilla and orangutan an intermediate number, and gibbon and siamang the least. Of the GPA antigenic determinants examined, the MN blood group determinants were most consistently expressed during evolution of the hominoid primates. The results suggested that variability in expression of GPA antigenic determinants between species was due to both differences in amino acid sequence and glycosylation.     

Modification of glycophorin A during oxidation of erythrocyte membrane

Human erythrocyte ghosts were oxidized with tert-butyl hydroperoxide and subsequently treated with tritiated borohydride to label the membrane proteins modified during the membrane oxidation. From the ghosts, oxidized-and-tritiated glycophorin A was isolated and characterized. No intermolecular cross-links were observed as analyzed by sodium dodecylsulfate gel electrophoresis. But, the number of lysine residues was significantly reduced and susceptibility to proteinases such as trypsin, chymotrypsin and pronase was lower than that of control glycophorin A. Trypsinization of the oxidized-and-tritiated glycophorin A gave insoluble and soluble trypsin fragments. After dansylation, N-terminal amino acids of the trypsin-fragments were determined. Dansyl amino acids from the insoluble trypsin fragments were not identical with those from control insoluble counterparts in the membrane-spanning region of glycophorin A molecule. Fractionation by gel filtration of dansyl-soluble trypsin fragments, and the N-terminal amino acid analysis of the fractionated peptides indicated that the peptides derived from the glycosylated region located in the outside of the membrane matrix were identical with those from control soluble counterparts. The results suggest that the glycosylated outside region of glycophorin A was modified only slightly but the hydrophobic membrane-spanning region was extensively modified during membrane oxidation, most likely by oxidized lipids.     

Studies on chimeric fusion proteins of human aldolase isozymes A and B.

Several kinds of fusion proteins between human aldolases A and B were prepared by recombinant DNA technology and their enzymic properties were examined. AB chimeras, which have aldolase A at the N-terminal region and aldolase B at the C-terminal region, were scarcely obtained, while BA chimeras were abundant (Kitajima et al., (1990), J. Biol. Chem., 265, 17493-17498). All the BAB chimeras, aldolase A fragments inserted in aldolase B, showed activity assignable to aldolase B type, which imply an essential role of Tyr residue at the C-terminus of aldolase A in the binding of fructose-1,6-bisphosphate (Fru-1,6-P2). BAB chimeras also showed reactivity to effectors such as fructose-2,6-bisphosphate (Fru-2,6-P2) and pyridoxal 5-phosphate (PLP), in a similar manner to aldolase B. BAB108 has a similarity to the BA108 chimera, but acts differently from other BAB chimeras, suggesting that its structure around active site looks like that of aldolase A.     

Cooperative action between band 3 and glycophorin A in human erythrocytes: immobilization of band 3 induced by antibodies to glycophorin A.

The ability of transmembrane receptor proteins to change their association with the cytoskeleton in response to ligand binding seems to be a key mechanism of signal transduction across membranes. To investigate the molecular features of this mechanism we have used the red cell membrane as a model system to study signal transduction through the integral protein, glycophorin A. In these studies the lateral mobility of integral proteins was measured in situ by fluorescence recovery after photobleaching, and membrane rigidity was characterized by micropipette aspiration technique. We found that binding either a monoclonal antibody or its monovalent Fab to the exoplasmic domain of glycophorin A in normal red cells immobilized the receptor and rigidified the membrane. Further, immobilization and rigidification did not occur when antibodies were bound to Miltenberger V cells containing a mutant form of glycophorin A lacking the cytoplasmic domain. These results imply that the site of the immobilization/rigidification lies within the membrane skeletal structure, not in exofacial receptor crosslinking, and requires the extended cytoplasmic domain of normal glycophorin A. In addition, we found that glycophorin A immobilization and membrane skeletal rigidification were accompanied by immobilization of band 3 receptors. This unexpected result indicates a cooperative coupling between liganded glycophorin A, band 3, and the membrane skeleton. We speculate that cooperation of this type may represent a general mechanism for cytoskeletal linkage and transformation initiated by receptors with short cytoplasmic sequences, such as integrins.     

Molecular organization of glycophorin A: implications for membrane interactions

Physical studies and conformational analysis of human glycophorin A suggest a revised model for its molecular organization, self-association, and interactions with the erythrocyte membrane. Intrinsic viscosity has been used to study, under more physiological conditions, the monomer–dimer equilibrium demonstrated previously by polyacrylamide–SDS gel electrophoresis. The results show that the equilibrium persists in the absence of detergent and support earlier indications that the dimer is probably the physiologically relevant form and that it is promoted by salt, inhibited by conventional denaturants, and abolished by carboxymethylation. Combined application of CD, fitted to the poly-(L -lysine) model spectra of Greenfield and Fasman, and conformational prediction, by the statistical method of Chou and Fasman and the stereochemical approach of Lim, suggests five helical sequences in glycophorin A: Arg-39 to Tyr-52 (A); Gln-63 to Glu-70 (B); Glu-72 to Leu-89 (C); Ile-95 to Lys-101 (D); and Leu-118 to Asn-125 (E). Sequence A occurs only at low pH and may be stabilized by favorable noncovalent interactions of O-linked tetrasaccharide side chains. The other four helices all occur in the dimeric form of glycophorin A at physiological pH and ionic strength. Sequence D is destroyed by trypsin, and is also lost on conversion to the monomeric form of the glycoprotein at low ionic strength. Sequence E is denatured by 6M guanidine hydrochloride/4M urea. Sequences B and C, which are separated by a single proline residue, are stable under all these conditions. Dimerization of the major, hydrophobic helical sequence, (C) may be promoted and directed by an adjacent short sequence of intermolecular parallel β-sheet (Leu-90 to Tyr-93). It is proposed that these two structures span the lipid bilayer in vivo, and that helices B and D lie, respectively, along the outer and inner surfaces of the membrane. Molecular organization in the N- and C-terminal regions of the molecule is discussed in terms of evidence from the present work and from other recent investigations.     

Organization of membrane lipids and proteins in human En(a-) erythrocytes that lack the major sialoglycoprotein, glycophorin A. A spin-label study.

Membrane fluidity was studied by electron-spin-resonance techniques in human En(a-) erythrocytes that lack the major membrane sialoglycoprotein, glycophorin A. By using stearic acid spin labels with a doxyl group in the C-12 or C-15 positions, we demonstrated that the hydrophobic core in these cells was more fluid than in normal cells. Surface-located regions in isolated En(a-) membranes, when probed with stearic acid labelled in the C-5 position, appeared more stable than in normal membranes. In isolated En(a-) membranes, protein motion was decreased when probed with a nitroxide derivative of maleimide. After incubation with anti-(glycophorin A) antibodies protein motion and membrane fluidity were increased in normal membranes. This effect was observed also after spectrin depletion, which by itself increased protein motion but decreased membrane fluidity in the hydrophobic core of the membrane. The results show that membrane proteins influence the fluidity of membrane lipids.     

Genetic variants of human red-cell membrane sialoglycoprotein beta. Study of the alterations occurring in the sialoglycoprotein-beta gene.

We have studied the DNA of individuals who express an altered sialoglycoprotein beta on their red cells by using Southern blotting with sialoglycoprotein-beta cDNA probes. Individuals of the Leach phenotype do not express any beta (sialoglycoprotein beta) or gamma (sialoglycoprotein gamma) on their red cells, and we show that about 7 kb of DNA, including the 3' end of the beta gene, is deleted in this DNA. Any protein product of this gene is likely to lack the membrane-associating domain of beta. We have also examined the DNA of two types of other individuals (Yus-type and Gerbich-type) who have red cells that lack beta and gamma, but contain abnormal sialoglycoproteins related to beta. These two types of DNA contain different internal deletions of about 6 kb in the beta gene. We suggest that these deletions result from the presence of two different sets of internal homology in the beta gene, and on this basis we propose structures for the abnormal Yus-type and Gerbich-type sialoglycoproteins which are consistent with the other evidence that is available. We provide evidence that beta and gamma are products of the same gene and suggest a possible mechanism for the origin of gamma based on leaky initiation of translation of beta mRNA.     

Structure of the transmembrane dimer interface of glycophorin A in membrane bilayers

The hydrophobic transmembrane domain of glycophorin A contains a sequence motif that mediates dimerization in membrane environments. Long-range interhelical distance measurements using magic angle spinning NMR spectroscopy provide high-resolution structural constraints on the packing of the dimer interface in membrane bilayers. We show that direct packing contacts occur between glycine residues at positions 79 and 83 in the transmembrane sequence. Additional interhelical constraints between Ile76 and Gly79 and between Val80 and Gly83 restrict the rotational orientation and crossing angle of the interacting helices. These results refine our previously proposed structure of the glycophorin A dimer [Smith, S. O., and Bormann, B. J. (1995) Proc. Natl. Acad. Sci. U.S.A. 92, 488-491] which revealed that the methyl groups of Val80 and Val84 are packed against Gly79 and Gly83, respectively.     

The mechanism of production of multiple mRNAs for human glycophorin A.

The major sialoglycoprotein in the human red cell surface membrane, glycophorin A is encoded by a single gene. However, this gene gives rise to three species of glycophorin A mRNA of sizes about 1.0, 1.7 and 2.8 kilobases in reticulocytes, foetal liver cells and erythroleukaemic K562 cells. In an investigation of how the three mRNAs originated, we showed by primer extension analysis that all three mRNAs in K562 cells had identical 5' termini and, by nucleotide sequencing of correlated cDNAs, that they had identical coding regions, except for the well-known glycophorin AM-AN polymorphism. However, we found also by sequencing the cDNAs that the mRNAs apparently differed from each other in the lengths of their 3' untranslated regions. This was confirmed by Northern blot analysis which also provided evidence that the three mRNAs originated by use of different polyadenylation signals of which seven were found in the longest cDNA we analyzed.