Membrane attachment sites for the membrane cytoskeletal protein 4.1 of the red blood cell [see comments]

The identity of the membrane binding sites for the membrane cytoskeletal protein 4.1 of the human red blood cell has been investigated. Exhaustive proteolysis of the membrane with a range of proteases led to the elimination of only some 60% of all binding sites. The predominant integral membrane protein, band 3, as well as glycophorin A, was totally digested at levels of proteolysis that were essentially without effect on the number of 4.1 binding sites. Proteolysis caused scission of the polypeptide chain of glycophorin C (together with the minor product, glycophorin D, of the same gene), but left a fragment from the region of the C-terminus still attached to the membrane. We have found a low-molecular weight protein, possessing an epitope (recognized by an antibody directed against the cytoplasmic domain of glycophorin C) in common with this proteolytic fragment, in cells of a Leach phenotype, which are characterized by lack of extracellular epitopes of glycophorin C. When these membranes were extracted at low ionic strength to dissociate the membrane cytoskeleton, approximately half the content of 4.1 was liberated, compared with only some 25% from normal membranes. Cells of a different variant of the Leach phenotype, which are totally devoid of glycophorin C, lost close to 70% of their 4.1 under these circumstances. The Rh(D) transmembrane protein, which interacts with the membrane cytoskeleton, is also resistant to proteolysis of the cytoplasmic membrane surface, but Rhnull cells, devoid of this protein, showed no decreased retention of 4.1. The results suggest that glycophorin C (with D) may contain two types of binding site for 4.1, which would be sufficient in number to account for all the strong binding of 4.1 on normal membranes; modulation of binding at one of the sites by another protein or by lipid is not excluded. A possible site for reinitiation of translation overlapping the premature stop codon in the mutant expressing the truncated glycophorin C can be discerned.     

Polymerase chain reaction-based detection of MN blood group-specific sequences in the human genome

The MN blood group antigens have traditionally been detected by serotyping; however, development of a DNA-based method offers flexibility in the determination of this highly polymorphic system. Genotyping the MN blood group antigens was performed by polymerase chain reaction amplification of the specific alleles (PASA) in the human genome. In separate paired reactions, M or N allele-specific oligonucleotide primers were amplified with a common distal primer. Only in the presence of the homologous template was a 781-base pair polymerase chain reaction amplification product visible after agarose gel electrophoresis and ethidium bromide staining. This method of genotyping could be performed using either 1 microgram of extracted DNA or 0.5 microL of whole blood, and the results showed 100-percent correlation with those obtained by serotyping. PASA-based genotyping of MN blood group antigens, which requires a small amount of starting material, has application in linkage and population studies and in forensic medicine.     

Purification and physical characteristics of a hemoglobin solution modified by coupling to 2-nor-2-formylpyridoxal 5''-phosphate (NFPLP)

Human stroma-free hemoglobin (Hb) was crosslinked with 2-nor-2-formylpyridoxal 5'-phosphate (NFPLP), purified over crosslinked dextran, and eluted with a linear salt gradient. The oxygen dissociation curve of this crosslinked hemoglobin appeared to be shifted to the right with a standard P50 of 49 torr (PO2 for 50% saturation with oxygen at a pH of 7.40, a PCO2 of 40 torr, and a temperature of 37 degrees C) compared with a P50 of 12 to 15 torr for the unmodified Hb. The Hill coefficient n of HbNFPLP was 2.1, versus 2.8 for Hb. The proton Bohr factor of HbNFPLP, calculated from P50 values in the pH range of 7.1 to 7.7, was found to be -0.19, versus -0.29 for unmodified Hb. The oxygen capacity of HbNFPLP was not affected by the crosslinking and was found to be 1.410 ml of O2 per g of HbNFPLP, versus 1.407 ml of O2 per g of Hb for unmodified Hb. Four derivatives of HbNFPLP, i.e., deoxyhemoglobin, oxyhemoglobin, carboxyhemoglobin, and methemoglobin, were prepared, and the light adsorption spectra were recorded in the region of 480 to 680 nm. No differences were detected in comparison with the spectra of unmodified Hb. The alpha and beta chains of the tetramer were separated by reverse-phase chromatography. Comparison of the elution patterns of the chains of Hb and HbNFPLP revealed a retardation of the beta chains due to crosslinking with NFPLP. This indicates that the binding of NFPLP to Hb occurred only between the beta chains.(ABSTRACT TRUNCATED AT 250 WORDS)