Review: Human trophoblast fusion and differentiation: lessons from trisomy 21 placenta

The syncytiotrophoblast layer plays a major role throughout pregnancy, since it is the site of numerous placental functions, including ion and nutrient exchange and the synthesis of steroid and peptide hormones required for fetal growth and development. Inadequate formation and regeneration of this tissue contributes to several pathologies of pregnancy such as intrauterine growth restriction and preeclampsia, which may lead to iatrogenic preterm delivery in order to prevent fetal death and maternal complications. Syncytiotrophoblast formation can be reproduced in vitro using different models. For the last ten years we have routinely purified villous cytotrophoblastic cells (CT) from normal first, second and third trimester placentas and from gestational age-matched Trisomy 21 placentas. We cultured villous CT on plastic dishes to follow the molecular and biochemical aspects of their morphological and functional differentiation. Taking advantage of this unique collection of samples, we here discuss the concept that trophoblast fusion and functional differentiation may be two differentially regulated processes, which are linked but quite distinct. We highlight the major role of mesenchymal-trophoblast cross talk in regulating trophoblast cell fusion. We suggest that the oxidative status of the trophoblast may regulate glycosylation of proteins, including hCG, and thereby modulate major trophoblast cell functions.

Elevated angiogenin levels in synovial fluid from patients with inflammatory arthritis and secretion of angiogenin by cultured synovial fibroblasts

Angiogenesis is a key process in the pathogenesis of inflammatory arthritis. Angiogenin is one of the most potent inducers of neovascularization in experimental models in vivo. To look for evidence that angiogenin is involved in inflammatory joint disease, we examined plasma and synovial fluid (SF) samples from rheumatology patients and synovial fibroblast cell culture supernatants. Angiogenin levels were determined by radioimmunoassay and ELISA. Plasma angiogenin concentrations ranged from 96 to 478 ng/ml, with no significant difference between patients and normal controls. In SF, angiogenin concentrations were significantly higher in patients with acute or chronic synovitis (rheumatoid arthritis (RA): median, 104 ng/ml; range 13-748, n = 14; crystal-induced arthritis (CIA): median, 149 ng/ml; range, 37-616, n = 14, and other chronic inflammatory arthritis: median, 42 ng/ml; range, 15-205; n = 9) than in the 18 patients with osteoarthritis (OA) (median, 20 ng/ml; range 8-116) (P < 0.0001, anova). Angiogenin levels in SF from RA patients in remission with secondary OA were similar to those achieved in primary OA, and decreased in parallel with the resolution of acute gout. Angiogenin protein was released by cultured synovial fibroblasts from OA and RA patients, and reached 1.18 ng/106 cells/day. These data suggest that angiogenin may mediate local inflammation in arthritis via effects on angiogenesis and leucocyte regulation.

[Angiogenin: involvement in angiogenesis and tumour growth]

Angiogenin is one of the most potent inducers of neovascularization in experimental models in vivo. Angiogenin is normally present in plasma but overexpressed in cancer patients. The possible involvement of angiogenin in the development of cancer is suggested by its overexpression in patients with a variety of tumours and the observation that angiogenin antagonists prevent the growth of human tumour xenografts in athymic mice. This 14.1-kDa protein has 35% amino acid sequence identity with human pancreatic ribonuclease and displays ribonucleolytic activity. As only angiogenin is able to induce angiogenesis, its biological activities are thought to result from structural characteristics. Although the structural characteristics of angiogenin have been extensively studied, the understanding of its physiological role and of how its properties are expressed is still to be deciphered. This article reviews some of the biological, biochemical and structural properties of angiogenin.

Internalization and processing of human angiogenin by cultured aortic smooth muscle cells

Human angiogenin is a 14-kDa plasma protein with angiogenic and ribonucleolytic activities. Angiogenin binds specifically to aortic smooth muscle cells, activates second messenger pathways, and inhibits their proliferation. Human and bovine aortic smooth muscle cells were used to study the internalization and intracellular fate of human angiogenin at 37 degrees C. Using a specific antibody against angiogenin, we found that the internalized native protein was localized in the perinuclear region at 30 min and then dispersed throughout the cytoplasm. In conditions favoring receptor-mediated endocytosis, internalization of iodinated angiogenin showed a first peak at 5 min and then further increased for up to 24 h. The half-life of the molecule, calculated as 12 h in chase experiments, could contribute to its intracellular accumulation. In cell extracts, in addition to the 14-kDa protein, a 8.7-kDa fragment was observed at 24 h, and three fragments with molecular mass of 10.5, 8.7, and 6. 1 kDa were detected at 48 h. Our data point to a specific internalization and processing of human angiogenin by aortic smooth muscle cells.

Angiogenin, a potent mediator of angiogenesis. Biological, biochemical and structural properties

Angiogenin is a heparin-binding 14 kDa plasma protein with angiogenic and ribonucleolytic activity. Although its structural features have been extensively studied, an understanding of its physiological role and of how its properties are expressed continues to elude researchers. This article discussed some of the biological, biochemical, and structural properties of angiogenin.

Expression of receptors for human angiogenin in vascular smooth muscle cells

Human angiogenin is a plasma protein with angiogenic and ribonucleolytic activities. Angiogenin inhibited both DNA replication and proliferation of aortic smooth muscle cells. Binding of 125I-angiogenin to bovine aortic smooth muscle cells at 4 degrees C was specific, saturable, reversible and involved two families of interactions. High-affinity binding sites with an apparent dissociation constant of 0.2 nm bound 1 x 104 molecules per cell grown at a density of 3 x 104.cm-2. Low-affinity binding sites with an apparent dissociation constant of 0.1 micrometer bound 4 x 106 molecules.cell-1. High-affinity binding sites decreased as cell density increased and were not detected at confluence. 125I-angiogenin bound specifically to cells routinely grown in serum-free conditions, indicating that the angiogenin-binding components were cell-derived. Affinity labelling of sparse bovine smooth muscle cells yielded seven major specific complexes of 45, 52, 70, 87, 98, 210 and 250-260 kDa. The same pattern was obtained with human cells. Potential modulators of angiogenesis such as protamine, heparin and the placental ribonuclease inhibitor competed for angiogenin binding to the cells. Together these data suggest that cultured bovine and human aortic smooth muscle cells express specific receptors for human angiogenin.

Basic homopolyamino acids, histones and protamines are potent antagonists of angiogenin binding to ribonuclease inhibitor

A radio-ribonuclease inhibitor assay based on the interaction of 125I-angiogenin with ribonuclease inhibitor (RI) was used to detect pancreatic-type ribonucleases and potential modulators of their action. We show that highly basic proteins including the homopolypeptides poly-arginine, poly-lysine and poly-ornithine, core histones, spermatid-specific S1 protein and the protamines HP3 and Z3 were strong inhibitors of angiogenin binding to RI. A minimum size of poly-arginine and poly-lysine was required for efficient inhibition. The inhibition likely resulted from direct association of the basic proteins with the acidic inhibitor, as RI bound to poly-lysine and protamines while 125I-angiogenin did not. Antagonists of the angiogenin-RI interaction are potential regulators of either angiogenin-triggered angiogenesis and/or intracellular RI function, depending on their preferential target.

Interaction of human angiogenin with copper modulates angiogenin binding to endothelial cells

Angiogenin is a potent inducer of blood-vessel formation with ribonucleolytic activity. Angiogenin binds to high affinity endothelial cell receptors and with lower affinity to extracellular matrix components. Here we report the effect of copper and zinc on these interactions. There was a 4.3-fold increase in angiogenin binding to calf pulmonary artery endothelial cells in the presence of Cu2+ in vitro. A 3.8-fold increase was observed with Zn2+, whereas Ni2+, Co2+, or Li+ had no effect. Specific angiogenin binding to the lower affinity matrix sites was increased by 2.7- and 1.9-fold in the presence of Cu2+ and Zn2+ respectively. Metal ion affinity chromatography and atomic absorption spectrometry were used to show the direct interaction of angiogenin with copper and zinc ions. Angiogenin bound 2.4 mol of copper per mole of protein. We suggest that copper, a modulator of angiogenesis in vivo, may be involved in the regulation of the biological activity of angiogenin.

Secretion of ribonucleases by normal and immortalized cells grown in serum-free culture conditions

The requirement of serum in cell culture is a major limitation for studies on secreted ribonucleases (RNases) because serum contains a high amount of ribonucleolytic activity. Defined culture condition is thus of interest to improve our knowledge of the RNase biology. We report here that cells from three different types and origins, Chinese hamster lung fibroblasts, bovine smooth muscle cells, and human endothelium-derived EA.hy926 cells, proliferate consistently in the presence of a basal medium supplemented with bovine serum albumin, high-density lipoproteins, basic fibroblast growth factor, insulin, and transferrin. Using a new quantitative radio-RNase inhibitor assay, two distinct ribonucleolytic assays, and a radioimmunoassay against angiogenin, it is shown that RNases became apparent in media conditioned by cell monolayers. Both the hamster lung fibroblast and the EA.hy926 cell lines secreted larger amounts of RNase inhibitor-interacting factors and RNase activity than normal smooth muscle cells. The serum-free medium represents an alternative way to grow these cells and allows investigation of biosynthesis and functions of RNases in culture. It should be useful to identify and quantitate unambiguously specific members of the RNase family secreted by normal versus tumor cells in culture.

Visualization of several binding sites for basic fibroblast growth factor (FGF-2) on fibroblasts by photoaffinity labeling: evidence for intracellular complexes

The internalization of basic fibroblast growth factor (FGF-2) was studied in Chinese hamster lung fibroblasts (CCL39). Recombinant FGF-2 was derivatized with a photoactivable agent, N-hydroxysuccinimidyl-4-azidobenzoate (HSAB), iodinated, and used to visualize intracellular FGF-2-affinity-labeled molecules after internalization at 37 degrees C. Iodinated HSAB-FGF-2 maintained the properties of natural FGF-2 such as affinity for heparin, binding to Bek and Fig receptors, interaction with high- and low-affinity binding sites, and reinitiating of DNA synthesis in CCL39 cells. Affinity-labeling experiments at 4 degrees C with 125I-HSAB-FGF-2 led to the detection of several FGF-cell surface complexes with apparent molecular mass of 80, 100, 125, 150, 170-180, 220, 260, and about 320 kDa by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), whereas two specific bands at 80 and 130-160 kDa were obtained using the homobifunctional cross-linking reagent, disuccinimidyl suberate. When the cells, preincubated with 125I-HSAB-FGF-2 at 4 degrees C and then washed, were shifted to 37 degrees C, irradiation of the internalized labeled FGF-2 led to detection of a similar but fainted profile with one major specific band at 80 kDa. Heparitinase II treatment of the cells reduced binding of 125I-HSAB-FGF-2 to its cell surface sites by 80% and internalization by 55%, indicating the involvement of heparan sulfate proteoglycans in these processes. Among the heparitinase-sensitive bands was the 80-kDa complex.

The widespread expression of angiogenin in different human cells suggests a biological function not only related to angiogenesis

Angiogenin is a secreted polypeptide that induces neovascularization in vivo. The expression of angiogenin by human cells in culture was investigated by using a specific radioimmunoassay and by cDNA hybridization. Angiogenin immunoreactivity was widely but differentially produced by anchorage-dependent growing cells including vascular endothelial cells from saphenous and umbilical veins, aortic smooth muscle cells, fibroblasts (from embryos, new-borns and adults), and tumour cells. Endothelial cells from saphenous veins and the endothelium-derived EA.hy926 cell line released immunoreactivity whatever the stage of the culture, including release at the lag phase, during exponential growth and at the confluent phase. However, the rate of accumulation of angiogenin varied as a function of EA.hy926 cell density. As compared to anchored cells, normal peripheral blood cells and tumour cells of myelomonocytic and megakaryocytic origin did not noticeably secrete angiogenin except at low levels. A myeloma cell line supernatant contained as much angiogenin cross-reactivity as did anchored cells, while four tumour T-cell lines expressed the cross-reactivity at different levels, i.e. from undetectable levels to a high level. A 0.9-kb angiogenin messenger RNA was detected by Northern-blot analyses in a variety of representative cells correlating with the presence of immunoreactivity in the cell-culture media. The widespread expression pattern of angiogenin suggests a physiological function that is not restricted to the neovascularization process.

[Angiogenic role of growth factors]

Several growth factors are potential inducers of neovascularization. This property established from in vitro or in vivo studies suggest that these polypeptides might be involved in physiological angiogenesis. Recent advances in the structure of the growth factors, identification of specific cell surface receptors and in their mechanisms of action bring a better understanding of the biochemical events involved in angiogenesis.

Interaction of basic fibroblast growth factor (FGF-2) with nonresponsive HeLa cells

Human HeLa adenocarcinoma cells did not respond to basic fibroblast growth factor (bFGF or FGF-2) but did bind the same amount of bFGF as responsive Chinese hamster lung fibroblasts (CCL 39). Heparinase II treatment of HeLa and CCL 39 cells resulted in a decrease of bFGF binding by 96 and 57%, respectively, indicating that heparan sulfate molecules were involved in bFGF binding. On HeLa cells, bFGF bound to a single family of low-affinity sites. Cross-linking experiments of 125I-bFGF to HeLa cells yielded several labeled complexes. Cell-associated 125I-bFGF was internalized in both cell types either by high-affinity receptors and heparitinase-sensitive sites in CCL 39 cells or by heparitinase-sensitive binding sites only in HeLa cells. The binding of bFGF to nonresponsive HeLa cells and its internalization via a family of heparitinase-sensitive binding sites might illustrate other functions of bFGF unrelated to cell proliferation.

Internalization of basic fibroblast growth factor by Chinese hamster lung fibroblast cells: involvement of several pathways

Subconfluent Chinese hamster lung fibroblast cells (CCL39) which express high- and low-affinity binding sites for basic fibroblast growth factor (bFGF) were used to study bFGF internalization. Kinetics at 37 degrees C indicated that this process was complex and involved various pathways with regard to the ligand concentration used. Internalization with 6 to 45 pM of 125I-r-bFGF led to a steady state that lasted up to 3 h without any appearance of 125I-labeled degradation products in the cell-culture medium, suggesting that the endocytosis reached equilibrium. Furthermore, binding data at steady state, at 37 degrees C, revealed a two-phase Scatchard curve suggesting the involvement of two families of interaction sites in the process of internalization. Apparent dissociation constants were estimated to be 20 pM and 58 nM, respectively, and the number of bFGF molecules involved per cell, 4300 and 1.3 x 10(6), respectively. These data were in good agreement with those obtained from binding experiments at equilibrium at 4 degrees C. Besides, higher concentrations of 125I-r-bFGF (greater than 47 pM) induced an internalization process which did not reach steady state and was not saturable. These results suggest that CCL39 cells could internalize bFGF by various pathways involving high- and low-affinity binding sites.

MCF7 mammary cancer cells respond to bFGF and internalize it following its release from extracellular matrix: a permissive role of cathepsin D

High and low affinity receptors for basic fibroblast growth factor (bFGF) were detected by binding experiments on MCF7 breast cancer cells. These cells were stimulated for growth by physiological concentrations of bFGF. However, in contrast to endothelial cells, these MCF7 cells did not produce detectable amounts of biologically active bFGF or related heparin-binding growth factor(s) of the FGF family. In vitro, the cathepsin D (cath-D) secreted by MCF7 cells was able to digest extracellular matrix (ECM) and to release ECM-bound 125I-bFGF. When MCF7 cells were cultured on ECM containing bound bFGF, they internalized bFGF, which was slowly and partially proteolyzed in the cells. Processing occurred in acidic compartments and was inhibited by leupeptin. Pepstatin A, an inhibitor of aspartyl proteases, had no effect on the processing but reduced internalization of matrix-bound bFGF by MCF7 cells. Taken together, these results suggest a cooperation between cath-D and bFGF, by which the protease could facilitate the release of bFGF from ECM and its subsequent use by breast cancer cells and/or adjacent cells involved in angiogenesis.

In vivo and in vitro studies of angiogenin--a potent angiogenic factor

Angiogenin is a potent blood-vessel-inducing polypeptide with a molecular weight of 14,000 that has a unique ribonucleolytic activity. First isolated from the conditioned medium of tumour cells, angiogenin has since been purified from normal plasma, which suggested that its propensity to induce neovascularization should be strictly controlled. Modulation of that activity might involve interaction of angiogenin with cell-surface receptors and extracellular matrix of endothelial cells, tight-binding inhibition of both its ribonucleolytic activity and cell binding property by ribonuclease inhibitor, as well as the overall influence of divalent copper, a modulator of angiogenesis.

Affinity chromatography of fibroblast growth factors on substituted polystyrene

The heparin-binding growth factors aFGF and bFGF (acidic and basic fibroblast growth factor) from crude bovine brain extract were co-eluted with purified [125I]aFGF and/or [125I]bFGF as tracers from heparin-Sepharose and from several insoluble substituted polystyrenes used as stationary phases in low-pressure affinity chromatography. The ability of the resins to isolate FGFs was determined by measuring the eluted radioactivity. It was demonstrated that the various substituted polystyrene resins retain [125I]aFGF and [125I]bFGF with different specificities according to the chemical nature of the substituted groups bound to the polystyrene support. Bifunctional resins substituted with sulphonate and phenylalanine sulphamide groups adsorbed both [125I]aFGF and [125I]bFGF whereas bifunctional resins substituted with sulphonate and sulphamide serine adsorbed only [125I]bFGF. These stationary phases could be adapted to high-performance affinity chromatography and used to isolate growth factors of the FGF family.

Acidic fibroblast growth factor is a modulator of epithelial plasticity in a rat bladder carcinoma cell line

During normal embryogenesis and neoplastic transformation epithelia change their state of differentiation and degree of cohesiveness. It is thus essential to identify the signals modulating these transitions. We report here that acidic fibroblast growth factor (FGF) induces cells derived from a rat bladder carcinoma to lose their epithelial character and to acquire some properties typical of mesenchymal cells. The structurally related basic FGF did not have such an effect; both factors, however, had a mitogenic activity for these cells. Two distinct populations of receptors for acidic FGF and basic FGF were distinguished by their ligand-binding characteristics. The observations that both acidic and basic FGFs had a mitogenic effect on NBT-II cells and that only acidic FGF caused cell dissociation and dispersion strongly suggest that these two biological activities could be medicated through distinct signaling pathways.

Specific binding of angiogenin to calf pulmonary artery endothelial cells

Specific binding of angiogenin (ANG) to calf pulmonary artery endothelial cells was demonstrated. Cellular binding at 4 degrees C of 125I-labeled human recombinant ANG was time and concentration dependent, reversible, and saturable in the presence of increasing amounts of the unlabeled molecules. The interaction was shown to be specific since a large excess of unlabeled ANG reduced labeled ANG binding by 80%, whereas similar doses of RNase A, a structurally related protein, had no effect. Scatchard analyses of binding data revealed two apparent components. High-affinity sites with an apparent dissociation constant of 5 x 10(-9) M were shown to represent cell-specific interactions. The second component, comprising low-affinity/high-capacity sites with an apparent dissociation constant of 0.2 x 10(-6) M, was essentially associated with pericellular components. High-affinity ANG binding sites varied with cell density and were found on other endothelial cells from bovine aorta, cornea, and adrenal cortex capillary but not on Chinese hamster lung fibroblasts. Divalent copper, a modulator of angiogenesis, was found to induce a severalfold increase in specific cell-bound radioactivity. Placental ribonuclease inhibitor, a tight-binding inhibitor of both ribonucleolytic and angiogenic activities of ANG, abolished 125I-labeled human recombinant ANG binding only in the absence of copper.

Internalization and limited processing of basic fibroblast growth factor on Chinese hamster lung fibroblasts

Using either acidic (pH 2.5) or trypsic treatments, we demonstrated that 125I-labeled basic Fibroblast Growth Factor (125I-bFGF) was submitted to an internalization process on responsive Chinese hamster lung fibroblasts (CCL39) at 37 degrees C. Various experiments based on the measurement of cell-associated radioactivity, as well as on research of degradated products of 125I-bFGF in cellular supernatants, showed that most of the internalized radioactivity remained intracellularly located after up to 5 hr of incubation. Analyses of this radioactivity by NaDodSO4-PAGE revealed the presence of labeled peptides issued from the limited processing of the native 125I-bFGF form (17 kD) and whose molecular weights were estimated to be 9 and 6 kD. Kinetic experiments indicated that proteolysis of the 125I-bFGF began early on incubation (less than 30 min) and led to a prolonged preservation of the 9- and 6-kD peptides which were still detectable after 13 hr of incubation. Preincubation of the cells with different lysosomotropic agents completely inhibited the proteolysis, indicating that this event occurred probably in an intracellular acidic compartment. Two enzyme inhibitors, leupeptin and N-alpha-tosyl-L-lysine chloromethyl ketone (TLCK), were also shown to interfere with the formation of both 9- and 6-kD peptides, thus suggesting a way to control the appearance of these fragments, and hence to determine their potential intracellular role.

Presence of basic fibroblast growth factor receptors in bovine brain membranes

We described a protocol for purification of bovine brain membranes suitable to study the binding of iodinated basic fibroblast growth factor (FGF) to bovine brain membrane preparation. The binding of 125I basic FGF to brain membranes reached equilibrium within 30 min at 20 degrees C, was reversible, and displaced by an excess of unlabeled basic FGF. Scatchard analysis of the data revealed that two classes of binding sites could be detected with an apparent Kd of 30 pM and a capacity of 0.24 pmol/mg of membrane proteins for the high affinity binding site and Kd of 3 nM with a capacity of 51 pmol/mg of membrane proteins for the low affinity binding site. Cross-linking experiments of labeled basic FGF to brain membrane receptor yield the formation of a single major complex with an apparent molecular mass of 170 kDa which is similar to the value obtained for the high affinity binding site for basic FGF on target cells in tissue culture. Hence these data present the first biochemical evidence suggesting that membrane purified from bovine brain contain two classes of specific binding sites for basic FGF and confirm results described with cells grown in vitro.

Isolation and partial purification of lipid-linked oligosaccharides from calf pancreas

Lipid-linked oligosaccharides containing at least five mannose residues (0.7 nmol/g of tissue) were isolated from calf pancreas by chloroform-methanol-water extraction and purification on DEAE-cellulose and Sephadex LH-20, and affinity chromatography on concanavalin A-Sepharose in the presence of nonionic detergent. The addition, prior to the chromatographic steps, of 14C-labeled lipid-linked oligosaccharides (synthesized in vitro by calf pancreas microsomes in the presence of GDP-D-[14C]mannose and UDP-D-[14C]glucose, respectively) as internal standards, indicated a final yield ranging from 38 to 50%. Analysis of the oligosaccharide residues by liquid chromatography of the lipid-free preparation, monitored by u.v. absorbance and radioactivity measurement of the tritiated compounds, indicated a heterogeneous mixture of oligosaccharides. Its components, ranging from Man5(GlcNAc)2 to Glc3Man9(GlcNAc)2, cochromatographed with the 14C-labeled derivatives from in vitro synthesis. Calf pancreas contains lipid intermediates bearing at least six mannose residues, such as Man9(GlcNAc)2-P-P-lipid, in almost equal or even higher amounts than Glc3Man9(GlcNAc)2-P-P-dolichol.

Evidence and characterization of the receptor to eye-derived growth factor I, the retinal form of basic fibroblast growth factor, on bovine epithelial lens cells

Eye-derived growth factor I (EDGF-I), the retinal form of the basic fibroblast growth factor, has been purified to homogeneity from bovine retina by heparin-Sepharose chromatography. The radioiodinated EDGF-I retained full mitogenic activity and was used to study the interaction of the growth factor with bovine epithelial lens cells. We showed that 125I-labeled EDGF-I bound in a saturable and reversible manner to a specific cellular receptor. Scatchard analysis of the equilibrium binding gave a Kd of 53 X 10(-12) M with approximately equal to 20,000 binding sites per cell. Crosslinking experiments using two homobifunctional reagents induced the formation of a specific major complex with a Mr of approximately equal to 145,000, as determined by NaDodSO4/PAGE, and independent of reducing conditions. These data establish the existence of a receptor for the basic growth factor derived from neural tissues and give an estimation of the size of this receptor at Mr 130,000.

Level of the A, B and H blood group glycosyltransferases in red cell membranes from patients with malignant hemopathies

Eight patients (4 suffering from acute myeloid leukemia) exhibiting a loss of ABO red cell antigens, as seen by a mixed-field reaction pattern in agglutination tests, were selected and examined for the level of the A, -B, -H blood group glycosyltransferases within membranes prepared from erythrocyte subpopulations (A or B positive and A or B negative red cells). A or B enzyme activities were largely decreased in membranes which had lost A or B antigens (A or B negative subpopulations) but were within normal level in membrane from cells which had not lost A or B antigens (A or B positive subpopulations). The H enzyme level which was frequently low in the serum was within normal limits in the membrane preparations examined. Since A or B negative subpopulations were normally glycosylated in vitro into A or B reactive structures, the results demonstrate that loss of A or B antigens is related to some alteration of the blood group gene products rather than to significant abnormalities of the membrane precursors.

Atypical transmission of ABO blood groups in a French family

A group AB mother (Mrs P.D.) gave birth to a group O female baby (C.D.). Extensive study of the blood group genetic markers in both the parents and the child, carried out on several occasions, showed nothing unusual outside the ABO system. Mrs P.D. then, gave birth to a second female baby who was also group O. Mrs P.D. had normal amounts of A, B, H and Lewis antigens in her saliva. The H, A and B agglutinability of her red cells was in the range of normal A2B group. This A2B blood group was characterized by very low A gene-specified glycosyltransferase activity in serum. Moreover this activity was undetectable in red blood cell membranes. These results are discussed in the light of various hypotheses in order to explain this unusual transmission of ABO blood group.

alpha-N-acetyl-D-galactosaminyl- and alpha-D-galactosyltransferase activities in sera of cis AB blood group individuals

Thirteen cis AB persons from five families were examined for serum glycosyltransferase activities associated with the biosynthesis of A and B blood group characters. Their transferases were generally homogeneous within one family, except for A2/cis AB genotypes, whose A enzyme level was similar to the A2 normal sera, but they varied from one family to another. These activities differed quantitatively and qualitatively from A, B and AB normal sera. Studies of A transferase showed variations in the pH-dependent curve, the effect of cofactors and the capacity of conversion of O red cells into A-active cells. Moreover, A and B transferases behaved differently with respect to their relative levels than did AB heterozygous normal sera. The results were discussed and it was suggested that a mutation of a single enzyme transferring both galactose and N-acetyl-galactosamine could explain these properties.

Normal blood group B antigen and B transferase activity in a patient with IgM autoanti-B agglutinin

Studies carried out on the red cells of a patient with autoanti-B agglutinin gave further evidence that it is probably not modified red cell antigens which cause autoantibody formation.

Inheritance of cis-AB in three generations (family Lam.)

In a case of disputed paternity genetic incompatibility was observed in the ABO blood group system between mother (O) and child (AB). From biostatistical evaluation of 21 genetic markers, including HLA phenotypes, a high value of probability for paternity, maternity and parentage was found between the child, the child's mother, the accused man and his mother. Substitution of the newborn was thereby excluded. The serostatistical evaluation of maternity and paternity could be supported by anthropological opinion. In serological investigations of the ABH system the A and B antigens of the 'A2B' cells were found to be abnormal in the child, the child's father, and the father's mother: the A was weaker than normal A1 but stronger than normal A2; the B was found to be abnormal which in addition was documented through occurrence of irregular anti-B antibodies in the child. In comparison to normal controls (A1, A2, B, A1B, A2B) diminished activity of alpha-D-N-acetylgalactosamine and alpha-D-galactosyltransferase were observed in the three 'A2B' propositi. These facts confirmed the existence of a cis-AB gene in the Lam. family which the child inherited from her father and the child's father from his mother. Assuming a population frequency of 1.1 X 10(-5) for the cis-AB gene, the probability for paternity was calculated from all genetic markers to be W = 99.9999985%.

[Serum glycosyltransferase activity associated with antigen biosynthesis in blood groups A and B. Study of normal B group and cis AB group subjects]

Two groups of alpha-D-galactosyltransferase activity were described in the sera of 132 caucasian B donors of the National Centre of Blood Transfusion, Paris. The group having the lowest activity represented 84% of the caucasian population, the second one: 16%. They were related neither to the secretory status nor to the genotype of the individuals studied. In some of the 20 african sera studied a third group of activity could be defined. In the caucasian population, these groups were clearly correlated to the agglutinability of the red cells by anti B serum. But no clearcut agglutinability groups could be defined. In the african heterogeneous population, there was no relationship between the two parameters. We did not find the third group of Gibbs. The agglutinability of the B red cells by anti-H (Ulex europaeus) varied inversely as transferase activity and two relationships were distinguished according to the ethnical origin. Only in the caucasian population, this phenomenon was in close relationship with the B agglutinability. Glycosyltransferases activities associated to the biosynthesis of A and B blood group antigens were looked for in 12 cis AB samples from 5 families. The alpha-N-acetyl-D-galactosamyltransferase activity was always weaker than that of normal A and AB sera except for A/cis AB genotype individuals. The presence of alpha-D-galactosyltransferase activity was demonstrated using 2' fucosyllactose as acceptor and by transformation of O red cells into weak B active ones.

Groups of alpha-D- galactosyltransferase activity in sera of individuals with normal B phenotype. II. Relationship between transferase activity and red cell agglutinability

In the Paris population of blood donors with normal B phenotype, two groups can be formed owing to their respective serum alpha-D-galactosyltransferase activity and red cell agglutinability with an anti-B antibody. Both parameters are closely correlated. The agglutinability groups partially overlap. In an African population from various ethnical origins, this correlation was observed only in some individuals. 11 among 20 subjects belonged to a third group defined by a high transferase activity. The third group with the strongest agglutinability previously described by GIBBS et al. [6] were not encountered. On the other hand, serum transferase activity varied inversely as agglutination scores with anti-H (Ulex). Both parameters are closely correlated but not in the same way in Caucasian as in African individuals. In the latter, this relation does not depend on the agglutinability group. The H antigen strength variability, according to ethnical origins, may explain these results.

Assay of alpha-N-acetylgalactosaminyltransferases in human sera. Further evidence for several types of Am individuals

The study of the alpha-N-acetylgalactosaminyltransferase in the sera of 19 individuals belonging to the rare Am blood group makes it possible to confirm the heterogeneity of this phenotype established on genetical and immunological criteria. Two groups of subjects, Am and Ay, can be distinguished. For the individuals of the first group, named Am, 15 samples (7 families) have been studied, the phenotype is inherited as an allele at the ABO locus. 14 of these subjects, have an alpha-N-acetylgalactosaminyltransferase whose kinetic properties were similar to those of A1 subjects. In one family, however, the A transferase detected is of the A1 type. On a quantitative level, the enzyme activities of these sera only reached 30-50 percent of the average value observed for A1 or A2 subjects, respectively. These facts suggest the existence of a genetic inhibitor, possibly linked to the ABO locus, preventing either an A1 or A2 gene from acting at the level of some cellular lines and leading therefore to the recognition of phenotypes named A-m-A1 and A-m-A2. On the contrary, under the experimental conditions used, no alpha-N-acetylgalactosaminyl-transferase activity was detected among the four individuals of the second group, named A-y by Weiner et al. (37), and whose appeareance in siblings results from the action of a recessive modifying y-A gene.