The Rh-blood-group antigens (often described as Rhesus antigens) are associated with erythrocyte membrane proteins of approx. 30 kDa. We have determined the N-terminal 54 amino acid residues of the 30 kDa Rh D polypeptide (D30 polypeptide). We used primers based on these sequence data and the polymerase chain reaction (PCR) on human reticulocyte cDNA and genomic DNA to clone two types of PCR product of identical size. The two PCR products had related translated amino acid sequences between the 3' ends of the primers, one of which was identical with that found for the D30 polypeptide. We designate the two related mRNA species which gave rise to the PCR products as Rh30A and Rh30B, the latter corresponding to the D30 polypeptide. We have isolated cDNA clones for the Rh30A protein which encode a hydrophobic membrane protein of 417 amino acids. The Rh30A protein has the same N-terminal 41 amino acids as the D30 polypeptide, but beyond this point the sequence differs, but is clearly related. The Rh30A protein probably corresponds to the R6A32 polypeptide, another member of the Rh 30 kDa family of proteins, which may carry the C/c and/or E/e antigens. Hydropathy analysis suggests that the Rh30A protein has up to 12 transmembrane domains. Three of these domains are bordered by a novel cysteine-containing motif, which might signal substitutions at these cysteine residues. Information which supplements this paper (amino-acid-sequence-analysis histograms) is reported in Supplementary Publication SUP 50160 (4 pages), which has been deposited at the British Library Document Supply Centre, Boston Spa, Wetherby, West Yorkshire LS23 7BQ, U.K., from whom copies can be obtained on the terms indicated in Biochem. J. (1990) 265, 5.

The erythrocyte chemokine receptor, a receptor for Plasmodium vivax, carries the antigens of the Duffy blood group system. Sequence analysis of reticulocyte RNA from individuals of known Duffy phenotype showed that the Fya antigen differs from the Fyb antigen as a result of a single nucleotide difference (A131 or G) encoding amino acid Gly44 (Fya) or Asp (Fyb) in the N-terminal extracellular domain of the glycoprotein. Evidence is presented for two different genetic backgrounds giving rise to the Fy(a-b-) phenotype. The most likely genetic mechanism in most individuals of the Fy(a-b-) phenotype is down-regulation of Duffy glycoprotein mRNA. However, the Duffy gene from a very rare Caucasian individual (AZ) with the Fy(a-b-) phenotype has a 14 base-pair deletion (nucleotides 287-301) resulting in a frameshift which introduces a stop codon and produces a putative truncated 118 amino acid protein. The occurrence of this mutation in an apparently healthy individual raises questions about the functional importance of the Duffy glycoprotein not only in normal erythrocytes but also in all human cells and tissues.

Membranes from erythrocytes with a new Gerbich (Ge)-negative phenotype (Leach phenotype), as well as those from two other Ge-negative phenotypes, were examined. Whereas cells of the Leach phenotype apparently lack three minor sialoglycoproteins (beta, beta 1 and gamma), the membranes of Ge- Yus- and Ge- Yus+ erythrocytes apparently lack beta- and gamma-sialoglycoproteins but contain additional diffusely migrating components of apparent Mr 30 500-34 500 and 32 500-36 500 respectively. Immunoprecipitation experiments showed that the abnormal components of both Ge- Yus- and Ge- Yus+ erythrocytes reacted with two monoclonal antibodies, BRIC 4 and BRIC 10. These antibodies have been shown to react with sialoglycoproteins beta and beta 1 in normal erythrocytes. Cytoskeletal preparations from Ge- Yus- and Ge- Yus+ erythrocyte membranes contained the abnormal components. In contrast with cells of the Leach phenotype, which are elliptocytic, Ge- Yus- and Ge- Yus+ were of normal shape, despite their apparent lack of beta- and gamma-sialoglycoproteins. It seems likely that the abnormal components in these cells contribute to their normal shape. Ovalocytic erythrocytes were shown to incorporate more radioactivity in the sialoglycoprotein-beta 1 region than normal erythrocytes after labelling by the periodate/NaB3H4 technique. It is suggested that abnormal components in Ge- Yus- and Ge- Yus+ erythrocytes result from chromosomal misalignment with unequal crossing-over at meiosis between the genes giving rise to beta-, beta 1- and gamma-sialoglycoproteins.

The use of prophylactic anti-D to prevent Rh D immunization in Rh D- women and subsequent hemolytic disease in Rh D+ infants is widespread, but has led to shortages of the anti-D Ig. With the aim of substituting monoclonal anti-D for Rh D prophylaxis, we have compared the abilities of monoclonal and polyclonal anti-D to clear Rh D+ red blood cells (RBCs) infused into Rh D- male volunteers and to suppress Rh D immunization. Two human monoclonal antibodies (MoAbs), BRAD-3 (IgG3) and BRAD-5 (IgG1), produced from stable Epstein-Barr virus-transformed B-lymphoblastoid cell lines, were selected because of their proven in vitro activity in promoting RBC lysis in antibody-dependent cell- mediated cytotoxicity assays. RBC clearance was assessed by intravenous injection of 3 mL of 51chromium-labeled D+ RBCs into 27 volunteers 48 hours after intramuscular injection of monoclonal or polyclonal anti-D. Further 3-mL injections of unlabeled D+ cells were administered at 6 and 9 months to induce immunization. Blood samples were taken throughout the 12-month period of study for the serologic detection of anti-D. The mean half-life (t50%) of RBCs in 7 recipients of 300 micrograms BRAD-5 (5.9 hours) was similar to that in 8 recipients of 500 IU polyclonal anti-D (5.0 hours), whereas D+ cells were cleared more slowly in some of the 8 subjects injected with 300 micrograms BRAD- 3 (mean t50% 12.7 hours) and in 1 individual administered 100 micrograms BRAD-3 (t50% 41.0 hours). The rate of RBC clearance in both groups administered 300 micrograms monoclonal anti-D correlated with the amount of antibody bound per cell, determined by flow cytometry. There was no evidence of primary immunization having occurred in any subject after 6 months of follow-up. Five of 24 subjects produced anti- D after one or two further injections of RBCs, confirming that they were responders who had been protected by the monoclonal or polyclonal anti-D administered initially. Four of these responders were recipients of monoclonal anti-D (3 BRAD-3, 1 BRAD-5). One individual who received BRAD-5 produced accelerated clearance of D+ RBCs at the third unprotected RBC challenge but did not seroconvert. This study shows that the human MoAbs BRAD-3 and BRAD-5 can prevent Rh D immunization, and indicates that they may be suitable replacements for the polyclonal anti-D presently used in prophylaxis of Rh D hemolytic disease of the newborn.(ABSTRACT TRUNCATED AT 400 WORDS)

Retroviral-mediated gene transfer using cDNA transcripts of the RHD and RHCE genes resulted in the isolation of K562 clones expressing D and G or c and E antigens, respectively. These results represent the first direct demonstration that the RHD gene encodes the D and G antigens and the RHCE gene encodes the c and E antigens. Both c and E antigens were expressed after transduction of K562 cells with a single cDNA, indicating that the c antigen does not arise by alternative splicing (exon skipping) of the product of the RHCE gene, as has been suggested.

The production and characterization of a new monoclonal antibody, Y2/51, against platelet glycoprotein IIIa is described. A useful feature of this antibody is its ability to recognize platelets and megakaryocytes in formalin-fixed routinely processed material. It could also be used to reveal platelets both in thrombi in large vessels and in microthrombi too small to be readily apparent on conventional microscopic examination. For this purpose it was helpful to use the antibody in conjunction with a new monoclonal reagent (Ret40f) against red cell sialoglycoprotein beta which detects red cells and their precursors in routinely processed tissue. The use of these antibodies should be valuable for the detection of thrombi in a variety of situations such as renal transplant rejection, coronary artery disease and vasculitis.

The human erythrocyte blood group system Cromer consists of high- incidence and low-incidence antigens that reside on decay-accelerating factor (DAF; CD55), a glycosyl-phosphatidylinositol-anchored membrane protein that regulates complement activation on cell surfaces. In the Cromer phenotypes Dr(a-) and Inab there is reduced or absent expression of DAF, respectively. This study investigated the molecular basis of the reduced DAF expression by polymerase chain reaction amplification of genomic DNA and RNA/cDNA obtained from Epstein-Barr virus- transformed lymphoblastoid cell lines. Sequence analysis of the Inab propositus showed a single nucleotide substitution in exon 2 of the DAF gene and at the corresponding position in the cDNA, G314-->A resulting in Trp53-->Stop. This truncation near the amino terminus explains the complete absence of surface DAF in the Inab phenotype. A similar analysis was performed for two Dr(a-) individuals, including KZ, who was previously reported to be Inab phenotype but is now shown by immunochemical and serologic methods to be Dr(a-) phenotype. A single nucleotide change was found in exon 5 of the DAF gene, C649-->T resulting in Ser165-->Leu, which we had previously shown to lead to loss of the Dra epitope. However, two species of cDNA were found, one encoding full-length DAF with the single amino acid change and the more abundant species having a 44-nucleotide deletion. The 44 nucleotide deletion includes the single polymorphic site, which creates a cryptic branch point in the Dr(a-) allele that leads to use of a downstream cryptic acceptor splice site. This shifts the reading frame and leads to a premature stop codon that precludes membrane anchoring. Thus, the single point mutation in the Dr(a-) phenotype results in a novel use of alternative splicing and provides a molecular explanation for both the antigenicity and the reduced DAF expression seen in this phenotype.

The flexibility of the human erythrocyte membrane is mediated by an underlying network of skeletal proteins which interact with the membrane through ankyrin and protein 4.1. The nature of the membrane attachment site(s) for protein 4.1 has yet to be fully elucidated. In this paper we show that purified protein 4.1 binds much less strongly to alkali-stripped membranes from erythrocytes of individuals with total glycophorin C and D deficiency (Leach phenotype) than to alkali-stripped normal membranes. We further show that a synthetic peptide corresponding to amino acid residues 82-98 of the cytoplasmic domain of glycophorin C specifically binds to purified protein 4.1 and inhibits protein 4.1 binding to alkali-stripped normal membranes. The same synthetic peptide binds directly to membranes from individuals with glycophorin C and D deficiency but not to normal membranes. These results indicate that glycophorins C and D provide major membrane attachment sites for protein 4.1 in normal erythrocytes and that this interaction is mediated by protein 4.1 binding to amino acid residues 82-98 of glycophorin C and 61-77 of glycophorin D.