Ultrastructural mapping of methyldopa and anti-D IgG erythrocyte antigen receptors.

The ultrastructural distribution pattern and site density of alpha-methyldopa immunoglobin G (alpha-MD IgG) on the red cell membrane was observed and compared with that of anti-D IgG, with ferritin-conjugated rabbit anti-human IgG and [125I]anti-D. alpha-MD IgG binds to all common types of human red cells, both Rho (D) positive and negative, to give a random, aperiodic distribution pattern grossly indistinguishable from the red cell D receptor site pattern. alpha-MD IgG inhibits the binding of [125I]anti-D to D-positive red cells when the reaction is controlled with respect to total reaction volume, ionic strength, and the appropriate concentrations of the two IgG reactants. To determine if a alpha-MD IgG binds to the D-antigen receptor, D-positive red cells were sensitized with alpha-MD and [125I]anti-D IgG spearately and with both IgG preparations. The cell-bound radioactivity served to identify what proportion of the total ferritin-labeled IgG sites were due to anti-D. With nonsaturating concentrations of anti-D the number of IgG sites observed was equal to the sum of the sites found when the red cell was sensitized separately with alpha-MD and anti-D IgG. With saturating concentrations of anti-D there was a reduction in the expected number of IgG sites, indicating that alpha-MD IgG was excluded from binding. There was no comparable interaction of alpha-MD IgG and anti-D IgG when D-negative red cells were used. The results obtained indicate that alpha-MD IgG does not bind to the D antigen. The interaction between alpha-MD IgG and anti-D IgG for binding sites on the red cell membrane may be due to the close physical proximity of the two receptors, so as to produce steric hindrance in binding of the two IgG preparations when both are present. The alpha-MD IgG receptor appears to be a part of the Rh antigen complex that occurs in both D-positive and D-negative red cells and probably contains receptors for other types of warm-antibody immune hemolytic anemias.

Relationship between Rh-o(D) zygosity and red cell Rh-o(D) antigen content in family members.

The red cells of 63 members of 11 families were tested with (125)I-labeled anti-Rh(0)(D). Families with a history of hemolytic disease of the newborn due to fetomaternal Rh incompatibility were selected for study. In such families it was possible to determine the antibody binding to the Rh(0)(D) heterozygous red cells of the children and to compare within each family this value with the antibody bound to the father's Rh(0)(D)-positive red cells and the mother's Rh(0)(D)-negative red cells. The fathers in all the families studied could be assigned to two classes on the basis of the quantity of antibody bound to their red cells. One group bound about the same quantity of antibody to their cells as did their children, indicating that they were heterozygous for the Rh(0)(D) antigen. The other bound about twice as much antibody to their cells as did their children, indicating that they were homozygous for the antigen. The Rh genotype of the father in all 11 families could be ascertained by using the children in each family as a reference point. The members of two families showed a poor correspondence between antibody binding and zygosity. In one family an Rh heterozygous child (R(1)r) took up 85% of the antibody bound to the father's homozygous cells (R(1)R(1)), and in the other family an Rh heterozygous child (R(1)r) took up 20% more antibody than did the cells of her father, which were of the same Rh phenotype (Rh(1)) and zygosity.The quantity of antibody bound to the red cells of unrelated Rh(0)(D) homozygous individuals of the same Rh phenotype (Rh(1)) showed an almost sixfold variation. A consequence of this observation was that the cells of Rh(0)(D) heterozygous children of high antibody uptake fathers took up more antibody than did the cells of low antibody uptake Rh(0)(D) homozygous fathers. The gene dosage effect for the Rh(0)(D) antigen demonstrable within a family does not appear to apply when unrelated individuals are tested, even though they may be of the same Rh phenotype.

The role of antigen mobility in anti-Rh0(D)-induced agglutination.

Intact human erythrocytes were cross-linked with glutaraldehyde (GA) or dimethyladipimidate (DMA) and tested for their ability to bind [125I]-IgG anti-Rh0(D) and to undergo antibody-mediated hemagglutination. There was no decrease in antibody binding after treatment with GA concentrations up to 1.25% and DMA concentrations up to 1%. Red cells treated with these concentrations of GA and DMA did not agglutinate. The techniques employed to induce agglutination of the cross-linked red cells involved "incomplete" IgG anti-Rho (D) in albumin, "complete" IgM anti-D Rho (D) in saline, and the antiglobulin (Coombs) reaction. The agglutinability of the chemically modified red cells was inversely correlated with the extent of fixation. The dissociation of antibody binding from agglutinability in cross-linked erythrocytes suggests that Rho (D) antigen mobility is required for red cell agglutination. Antigen mobility was manifested by the transition from a relatively monodisperse distribution pattern of Rho (D) antigen sites to one of large aggregates or clusters when agglutination was induced by IgM anti-Rho (D), IgG anti-Rho (D) agglutination of protease modified red cells, and by anti-IgG agglutination of IgG anti-Rho (D)-coated red cells. Antigen clustering was not as prominent in red cells agglutinated by IgG anti-Rho (D) in the presence of albumin. Even though antigen mobility is a prerequisite for antibody-mediated hemagglutination, clustering does not appear to be an absolute requirement. The degree of antigen clustering differs with varying types of agglutination.

Rh antigen immunoreactivity after histidine modification.

125I-anti-D IgG and unlabeled blood group allo-antisera in combination with 125I-protein A were employed in assessing antibody binding to red cells (RBC) treated with histidine reagents. The acylating reagent diethylpyrocarbonate (DEP), the alkylating reagent p-bromophenacyl bromide (pBPB) and the photosensitizer dye Rose Bengal (RB) were used under conditions that usually result in the selective modification of histidine in isolated proteins. Progressive apparent inactivation of the D antigen in ghost membranes occurred with increasing DEP concns, which was not demonstrably reversible by hydroxylamine since this reagent itself inactivated the D antigen. Exposure of red cells to 5 mM p BPB resulted in a 50% decrease in binding of 125I-anti-D IgG. Photo-oxidation of RBC in the presence of Rose Bengal apparently inactivated all the major Rh antigens as detected either by labeled anti-D IgG binding, IgG agglutinating serological reagents, or the binding of 125I-labeled protein A following the sensitization of cells with unlabeled antisera. Under conditions of RB treatment, where hemolysis was absent or minimal, 125I anti-D IgG binding decreased to 38-49% of the level seen in controls. Rose Bengal treatment of R1r RBC revealed varying inactivation of all the Rh antigens, i.e. D 15%, C 89%, c 73%, e 54% inactivated, whereas antibody binding activity of the Fya and Fyb antigens present in the same cell was unaffected. Previous reports as well as the pH profile of anti-D binding have implicated the participation of histidine in Rh antigen expression. Our results are consistent with histidine involvement in Rh activity. Whether Rh antigens have essential histidine(s) involved directly in epitope structure, or instead depend on a critical histidine(s) at the lipid-protein interface that modulates antigen expression remains to be determined.

Characterization of the D, c, E and G antigens of the Rh blood group system with human monoclonal antibodies.

The human MAbs, anti-D, -c, -E and -G of the Rh blood group system, produced by Epstein-Barr virus transformed B-cell lines, were purified by protein A-Sepharose chromatography and used to characterize the Rh antigens of human red cells. Scatchard plot analyses performed with the radiolabelled MABs indicated that each R2R2 red cell carries 0.43, 0.32 and 0.38 x 10(5) D, c and E binding sites, respectively. About half this number of antigen sites are present on erythrocytes from heterozygote individuals using the appropriate antibody. We found, however, that only 0.18 x 10(5)G antigenic sites were present on each R1R1 red cell. The affinity constants of the anti-D, -E and -G were similar varying from 0.6 to 1.5 x 10(8) M-1 whereas that of the anti-c was much lower (0.035 x 10(8) M-1). The blood group specificity and binding properties indicate that the MAbs behave like the polyclonal anti-Rh reagents. Immunoprecipitation experiments carried out with membranes from R2R2 red cells show that a 30-32 kDa component can be identified whatever the antibody used. The immune complexes involving anti-c, -E or -G antibodies could be formed with the detergent lysates from red cell membranes. In contrast, membrane integrity was a prerequisite for the binding of the anti-D antibodies. Finally, from extraction studies of immunocomplexes with non-ionic detergents it was concluded that all the Rh-active components are bound to the membrane skeleton, suggesting that these molecules may have important function for maintaining red cell shape and viability.

Labeling of Rh antibodies on solid-phase protein A.

Rh0(D) antibodies which retain immune specificity after radiolabeling were prepared by a procedure which does not require IgG isolation from serum, requires 10-fold less isotope than conventional techniques and yields antibody solutions of defined composition. The method involves radioiodination of IgG on immobilized protein A, depends on employing human red cells reduced in surface cytophilic IgG, and exploits the inability of goat IgG to interact with Staphylococcus aureus protein A. The technique concentrates IgG by affinity adsorption and should prove useful in preparing radiolabeled alloantibodies from dilute human antisera and for red cell autoantibodies.

The antiglobulin reaction with native and enzyme-modified red cells.

The quantity of anti-IgG bound at equilibrium to cell-bound anti-D failed to increase with enzymatic modification of intact red blood cells. This contrasts with results obtained by immunoelectronmicroscopy with unsealed red blood cell ghosts. Ghosts derived from enzyme-modified, anti-D-sensitized intact red blood cells show an increased ratio of anti-IgG to anti-D. The ratio declined with protease modification of intact cells, possibly because of steric hinderance to anti-IgG binding within the cellular agglutinates. At comparable levels of cell-bound anti-D and anti-IgG, enzyme-modified cells were more agglutinable in the antiglobulin reaction than unmodified cells. As in saline hemagglutination, enzymatic enhancement of agglutinability appears to be unrelated to the amount of antibody bound, but rather may be related to other factors such as alterations of the biophysical properties of the red blood cell membrane.

Blood group D antigen content of nucleated red cell precursors.

The D antigen content of nucleated red cell precursors in human bone marrow was estimated using autoradiography and 125I-anti-D. D antigen first appeared in the pronormoblast, and the quantity of antigen progressively increased during red cell maturation. Maximal anti-D binding occurred on mature red blood cells. Pronormoblasts, basophilic normoblasts, polychromatophilic normoblasts, and orthochromatic normoblasts, respectively, had approximately 1/4, 1/2, 2/3, and 3/4 the quantity of antigen found on mature red cells. None of the other cell types were found in bone marrow labeled with anti-D.

Interaction 125I-protein A with erythrocyte-bound IgG.

The molar combining ratio of 125I-PrA to RBC-bound 125I-labeled IgG anti-D was 0.72 +/- 0.044. There was a significant decrease in the PrA-to-IgG combining ratio when anti-D was bound to protease-modified RBCs or to unmodified RBCs sensitized at low ionic strength, 0.49 +/- 0.034 and 0.56 +/- 0.006, respectively. These findings indicate that the interaction of RBC-bound IgG with PrA may be influenced by alterations in membrane structure, surface density, and distribution of the IgG receptor and possibly other steric factors. The quantity of RBC-bound IgG on RBCs sensitized with unlabeled serum anti-D and anti-Kell could be quantitatively assessed and correlated with antiglobulin agglutinability. Unlabeled alloantibodies were detected with the 125I-PrA at IgG densities lower than those detectable with the standard antiglobulin test. 125I-PrA, in contrast to the antiglobulin reaction, has the potential of providing increased sensitivity as well as quantitative data in assessing IgG alloantibody- or autoantibody-sensitized RBCs. (J Lab Clin Med 99:399, 1982.)

Protease modification enhances anti-D binding to nucleated red blood cell precursors.

Protease modification of nucleated red blood cell precursors resulted in significant enhancement of anti-D binding, indicating the presence of cryptic Rh determinants within the plasma membrane of the developing erythroid cell. Protease-modified nucleated precursor cells bound less anti-D than protease-modified mature red blood cells, however, suggesting that the total content of D antigen in the developing erythropoietic plasma membrane is reduced. These findings support the hypothesis that the D antigen, an integral membrane protein, is progressively synthesized during erythropoiesis.

Quantitative two-dimensional ultrastructural distribution of Rh o (D) antigenic sites on human erythrocyte membranes.

A method is described for determining the two-dimensional distribution of specific antigens on cell surfaces, and is applied to the D antigen of the Rh antigenic system. Rh-positive human erythrocytes are allowed to react with purified (125)I-labeled human anti-Rh(o)(D) gamma-globulin antibodies, and the sensitized cells are then lysed at an air-water interface. The residual cell membranes are spread flat by surface forces, and are picked up on a carbon-strengthened collodion-coated electron microscope grid. The membranes are then stained with ferritin-conjugated goat antibodies directed against human gamma-globulins. Only Rh-positive cells sensitized with anti-Rh(o)(D) antibodies bind the ferritin-conjugated antihuman gamma-globulins. The ferritin particles are found in small clusters on the membrane surface, and the number of such clusters per unit area agrees with the number of (125)I-labeled anti-Rh(o)(D) antibodies bound per unit area. The Rh(o)(D) antigenic sites appear to be molecularly dispersed on the membrane surface, but in a random two dimensional array.

Immunoreactivity of the Rho(D) antigen in cytoskeleton-free vesicles.

Skeleton-free microvesicles derived from D-positive red cells immunospecifically bind anti-D. Anti-D saturation binding yielded an equilibrium constant value (K = 1-2 X 10(8) I mole-1) similar to intact red cells. The vesicles contained band 3, the major sialoglycoproteins and, at high protein loads, low-molecular-weight polypeptides. Anti-D binding was referenced to total protein, phospholipid, and band 3 content to determine whether there was nonrandom segregation of the D antigen into vesicles. Selective segregation of the D antigen into vesicles could not be demonstrated unequivocally based on the methods and the assumptions of this study. The results, however, indicate that D reactivity does not require a membrane skeletal association.

Antiidiotypic IgG crossreactive with Rh alloantibodies in red cell autoimmunity.

IgG autoantibodies eluted from RBCs of antiglobulin positive normal blood donors contained at least two antibody populations, an IgG autoantibody (Ab 1), and an IgG population (Ab 2) that agglutinated RBCs coated with some Rh(D) alloantibodies. Eight of 24 autoantibody eluates tested agglutinated 3 of 10 anti-Rh(D) sensitized RBCs. The agglutinating activity was inhibited specifically by preincubation of the autoantibody eluate with the reactive anti-D. The reaction did not require the Fc domain of the anti-Rh(D), since autoantibody eluates agglutinated RBCs coated with F(ab')2 prepared from the reactive anti-D sera. These findings indicate that the RBCs of some antiglobulin-positive blood donors contain an immunoglobulin auto-antiidiotype (Ab 2) against the RBC autoantibody (Ab 1) which is demonstrable through its cross-reactivity with selected Rh(D) alloantibodies. Identification of auto-antiidiotypes in RBC autoimmunity lends support to the idiotype-antiidiotype network hypothesis of immune regulation and is consistent with the bizarre and complex serology of autoimmune hemolytic anemia. The absence of clinical hemolysis in antiglobulin-positive normal blood donors suggests that immunoglobulin idiotype-antiidiotype interactions may play a role in modulating the effects of RBC autoimmunity.

Red cell autoantibodies characterized by competitive inhibition of iodine 125 Rh alloantibody binding and by immunoprecipitation of membrane proteins.

The relationship between determinants recognized by warm-type immunoglobulin G red cell autoantibodies and the Rh antigens was characterized by autoantibody competitive inhibition of iodine 125 Rh alloantibody binding and autoantibody immunoprecipitation of iodine 125 red blood cell membrane proteins. The majority of blood donor autoantibody recognized epitopes that are closely related to Rh antigens as determined by competitive inhibition studies. Eighteen of 20 (90%) autoantibodies inhibited anti-Rh(c) binding, 15 inhibited anti-Rh(E), 5 inhibited anti-Rh(D), and only 2 failed to inhibit any of the three Rh alloantibodies tested. Autoantibodies that inhibited anti-Rh(D) also inhibited anti-Rh(c) and anti-Rh(E) and all those that inhibited anti-Rh(E) also inhibited anti-Rh(c). Autoantibodies that inhibited all three Rh alloantibodies immunoprecipitated 30 kd membrane polypeptides, as did two of the three autoantibodies that inhibited only anti-Rh(c) and anti-Rh(E). One autoantibody in this group and two autoantibodies that inhibited only anti-Rh(c), as well as an autoantibody that did not inhibit any of the Rh alloantibodies, immunoprecipitated only a single membrane polypeptide identified as band 3. The majority of normal donor red blood cell autoantibodies inhibited the binding of Rh alloantibodies, which indicates that they either bound to the Rh polypeptides or to epitopes on band 3 that were closely associated with the Rh complex.

Differential binding of IgG anti-D and IgG autoantibodies to reticulocytes and red blood cells.

125I IgG anti-D binding to reticulocytes obtained by density fractionation is reduced relative to that bound to all other red cell (RBC) fractions. Maximum D antigen reactivity occurs following reticulocyte maturation with no detectable change in D reactivity of mature RBC throughout their life span. Reticulocytes have in the range of about 60% of the content of mature RBC. Previously reported increased anti-D agglutinability and binding to old RBC it is not due to an intrinsic increase in D antigen with age, but results from an 'apparent' decrease in anti-D binding to young RBC fractions due to reticulocyte enrichment. IgG RBC autoantibodies obtained by elution from the RBC of eight Coombs-positive blood donors, probably associated with alpha-methyldopa (alpha-MD) administration, showed decreased binding to reticulocytes as determined by 125I protein A (PA). Reticulocytes bound about 70% of the IgG bound to mature RBC, indicating that the membrane antigenic determinant defined by these autoantibodies was incompletely expressed in the reticulocyte. This difference in IgG autoantibody binding between reticulocytes and mature RBC is similar to the decreased D antigen content of reticulocytes and consistent with an autoantibody determinant associated with the Rh complex. Direct testing of density fractionated Coombs-positive RBC in four out of five patients with autoimmune haemolytic anaemia (AIHA) showed reduced quantities of IgG on reticulocytes. The distribution of IgG between reticulocytes and mature RBC may be useful in serologically characterizing patients with AIHA and in identifying subpopulations of patients with this disorder.

Exposure of the Rh0(D) antigen on the surface and cytoplasmic domains of the red cell membrane.

Inside-out (IO) and right-side-out (RO) vesicles derived form human red blood cells were tested for their ability to bind 125I-labelled IgG anti-RHO(D). The binding of anti-RHO(D) to RO vesicles from RHO(D)-positive cells was quantitatively similar to that exhibited by intact cells when compared on a membrane surface area basis. There was no significant binding of labelled antibody to IO vesicles from RhO(D)-positive cells or to either RO or IO vesicles derived from RhO(D)-negative cells. The RhO(D) antigen was immunologically accessible on only the plasma side of the membrane in RhO(D)-positive red cells, as has been shown for blood group antigens defined by carbohydrate determinants. No immunologically reactive RhO(D) antigen was present on either RO or IO vesicles derived from RHO(D)-negative red cells.

The IgG binding function of the normal red cell plasma membrane: identification of integral polypeptides that bind IgG.

Non-immune IgG binds to red cell integral membrane proteins obtained by mild alkaline extraction of ghosts. Detergent gel chromatography and electrophoretic analyses of IgG-membrane protein complexes obtained by nonionic detergent solubilization and affinity binding to protein A indicate that three polypeptides participate in the binding of IgG. These have apparent molecular weights of 90 000, 44 000 and 22 000 and are present in a 1:3:0.9 stoichiometry. Evidence obtained indicates that the major sialoglycoproteins are not involved in this type of binding.

Anti-Rho(D) IgG binds to band 3 glycoprotein of the human erythrocyte membrane.

Alkali-extracted erythrocyte ghost membranes from Rho(D)-positive and Rho(D)-negative donors were incubated with human immune anti-Rho(D) IgG and nonimmune IgG. After sensitization with IgG, the integral membrane proteins were solubilized in Brij 36T nonionic detergent and chromatographed by gel filtration. There was a distinct resolution of IgG into free and membrane-complexed forms. The IgG-complexed membrane proteins were isolated by the use of a staphylococcal protein A affinity support. The protein A-bound complexes were examined for polypeptide composition by gel electrophoresis after elution. Only Rho(D)-positive membrane proteins incubated with immune anti-Rho(D) IgG revealed intact band 3. Control Rh-negative membrane proteins that had reacted with immune anti-Rho(D) IgG and the Rh-positive membranes that had reacted with nonimmune IgG showed only low molecular weight fragments of band 3 that bound nonspecifically to IgG. Arguments are presented supporting a band 3 localization for the Rh antigen.

Proteolysis of band 3 is enhanced by anti-Rho(D) binding to the red cell membrane.

Surface radioiodinated human red cells were incubated with IgG fractions and the radioelectrophoretic profile of the ghost membranes determined. The patterns of RhO(D)-negative membranes exposed to anti-RhO(D) IgG and RhO(D)-positive membranes exposed to non-immune IgG fractions remained intact. Membranes of RhO(D)-positive membranes following incubation with anti-RhO(D) IgG showed a sharp reduction in the quantity of intact band 3, the main glycoprotein of the red cell membrane. This process was significantly abrogated in the presence of protease inhibitors. The results suggest a possible role for IgG binding in promoting the generation of band 3-derived fragments described by others as normal constituents of isolated ghosts.

Quantitative immunoferritin microscopy of Fya, Fyb, Jka, U, and Dib antigen site numbers on human red cells.

The Fya, Fyb, Jka, U, and Dib antigen site numbers and ultrastructural distribution patterns on the human erythrocyte membrane were determined using quantitative immunoferritin microscopy. For homozygous antigen-positive red cells, the average number of determinants per red cell was about 14,000 for Jka, 17,000 for Fya and Fyb, 19,000 for Dib, and 23,000 for the U antigen, assuming that the equilibrium binding observed represented 80% saturation of the accessible antigen sites. The site numbers for this group of antigens were less than that for the Rh antigens, but considerably more than the Kell and Cellano antigens. The technique used was capable of demonstrating a twofold difference in antigen density between heterozygous and homozygous Fy (a+) red cells. More than 85% of the Fya and Fyb antigen sites were lost following pretreatment of the red cells with papain, consistent with the serologic lability of the Fy antigens following proteolysis. The ferritin distribution observed following conjugate staining of antibody-sensitized ghost membranes was similar for all five antigens studied and showed a random, clustered ferritin pattern. Although the quantitative estimates are valid, the remarkable similarity in antigen distribution pattern for this diverse group of antigens, as well as other considerations, suggest that the findings with ghose membranes probably do not reflect faithfully the antigen arrangement on the intact red cell membrane.