•Anti-A, reagents can be: a lectin prepared from Dolichos biflorus seeds; sera of group B subjects absorbed with group A2 red cells; or mouse monoclonal antibodies.

"The plasma of 1-8% of group Az and 22-35% of group AjB individuals have weak naturally occurring anti-A,.

•Anti-A, reagents can be: a lectin prepared from Dolichos biflorus seeds; sera of group B subjects absorbed with group A2 red cells; or mouse monoclonal antibodies.

"The plasma of 1-8% of group Az and 22-35% of group AjB individuals have weak naturally occurring anti-A,.

on both homologous chromosomes that code for functionally inert proteins (see below), while A and B code for an N-acetylgalactosaminyl- and galacto-syltransferase respectively. The appropriate alleles that encode the A1, A2 and B-transferases have been cloned and sequenced. The A and B genes code for 41 kDa proteins composed of 353 amino acids, which contain a 21 amino acid long leader peptide that is cleaved off to form the mature transferase. The two genes differ at seven nucleotide positions, which generate only four amino acid differences, found at positions 176, 235, 266 and 268 of the A-or B-transferase polypeptide. These correspond to Arg, Gly, Leu and Gly in the A-transferase, and to Gly, Ser, Met and Ala, respectively, in the B-transferase. The most critical differences are at positions 266 and 268 but the four are collectively responsible for the immunochemical differences between A and B and their associated transferases; substitution of one or more of these four key amino acids can critically affect ABO expression (see O2 below).

Three types of O alleles have been described to date; the commonest has a nucleotide sequence similar to that of the A allele, but with a single base deletion that generates a change in reading frame (hence scrambling the amino acid sequence) at amino acid position 87. This deletion also produces a new in-frame stop codon that generates a truncated O allele polypeptide that is only 116 amino acids long, and which is enzymically inactive. An analogous O allele, with an incidence one-quarter of that of the former and representing an equivalent single base deletion in a B gene, has also been reported. The third type of O (sometimes called O2) arises from Arg—»Gly and Gly—»Arg substitutions at amino acids 176 and 268, respectively, of an A allele which abolishes the enzymatic activity of the resulting protein, thereby emphasizing the metabolic importance of these amino acids at the transferase active site.

The differences between the A1 and A2 alleles arise from a Pro—>Leu mutation at amino acid position 156 coupled to a single base deletion near the C-terminus of an equivalent Abállele. This in turn results in a frame shift so that the original stop codon is no longer recognized, and a new stop codon is generated downstream, giving rise to a polypeptide for the A2 enzyme that is 21 amino acids longer than the A1 enzyme. The extra sequence thus generated, together with the Pro-Leu substitution in the A2-transferase, should account for the differences in kinetics and specificity between the two enzymes (see below). In addition to A1 and A2, the alleles encoding several examples of various other ABO subgroups have also been cloned: A3, A„ B3, B(A) and c/s-AB. All have so far proven to be heterogeneous at the nucleotide level despite having been considered to characterize a single phenotypic subtype.

The A- and B-transferases synthesize A and B carbohydrate antigens by transfer of N-acetylgalac-tosamine (GalNAc) or galactose (Gal) moieties from UDP-GalNAc or UDP-Gal respectively to the C-3 carbon atom of a fucosylated terminal galactose residue present in a variety of oligosaccharide precursors with blood group H activity (see Figure 1). These precursors are the sugar component of glycolipids and glycoproteins. The transferases of the A, and A2 alleles are known to have different pH optima, isoelectric points, thermal stabilities and metal ion cofactor requirements, which result in differences in their substrate transfer kinetics.

It is rarely possible to deduce a genotype from an ABO phenotype unless the groups of the family are available. For example, a group B individual could be B/B or B/O; an A, subject could be A'/A1, A'/A2 or even AVO. In contrast, group AB and group O individuals are A/B and O/O respectively, unless they are of the extremely rare ds-AB (with A and B antigens both derived from a rare allele at a single locus) or O 'Bombay' (see below) phenotypes.

Knowledge of the nucleotide sequences that distinguish the ABO alleles has enabled primers to be designed that can be used to determine ABO genotypes through the application of PCR. These include sequence specific primers that uniquely amplify A, B or O, as well as generic primers that amplify all ABO alleles; the products of the latter are distinguished by their sensitivity to restriction endonucleases that recognize the DNA sequences of specific ABO alleles.

ABO epitopes

ABO antigens are synthesized by glycosylation of oligosaccharides with H antigen activity. The H antigen is synthesized by a fucosyltransferase that is the product of the H gene on the long arm of chromosome 19. Carbohydrate chains carrying the A, B, and H antigens are present on 1) the short-chain oligosaccharides of simple glycolipids in plasma; 2) the heavily branched polysaccharides that form the poly-glycosyl moieties of either soluble glycoproteins present in secretions or of polyglycosyl ceramides in the red cell membrane, and 3) the short O-linked and highly branched N-linked polysaccharides of integral membrane proteins. The immunodominant sugars of the A and B antigens are at the terminal (non-reducing end) of the various polysaccharide chains expressing A or B, and are invariably attached by an al-3 linkage to a fucosylated galactose residue with H antigen activity (see Figure 1) such that the simplest A epitope is a trisaccharide with the structure:

AType 1


Figure 1 Biosynthetic pathways for the synthesis of A, B and H structures on type 1 precursor chains.


|al—2 Fuc where R represents the rest of the polysaccharide chain. The B epitopes have Gal instead of GalNAc as the immunodominant sugar. The presence of the fucosyl residue, i.e. the H antigen, is essential for A and B expression: its absence, as in the rare 'Bombay' phenotype, leads to an inability of the A- and B-transferases to add their respective sugars to the Gal part of Gal—R, and hence A and B antigens are not expressed, even though the relevant transferases can be detected in cells and plasma and are functionally active in in vitro assays using fucosylated substrates from normal subjects as acceptors. Such 'Bombay' individuals lack an H gene, being homozygous for the silent allele h.

The terminal trisaccharides can be attached to R in at least six different ways: e.g. by either a 1 -3 or 1-4 linkage to (BGlcNAc (type 1 and type 2 A or B structures respectively), by a 1-3 link to a GalNAc (type 3), ^GalNAc (type 4) or pGal (type 5) or even a 1-4 link to (3Glc (type 6). Of these, types 1 and 2 are the more abundant in the red cell membrane and the most important in red cell serology: integral red cell membrane proteins and glycolipids have almost exclusively type 2 linked sugars. Red cells may also contain glycolipids, passively adsorbed from plasma, that have exclusively type 1 chains. Secretions have a mixture of type 1 and type 2 epitopes. The existence of these various epitopes probably explains the heterogeneity in reactivity of different ABO antibodies with group A or B variants; seven reaction patterns for monoclonal anti-A and anti-B have been described.

In secretory tissues and other epithelia, ABH antigen expression is modulated by the genes of the 'secretor' (Se) locus which is not linked to the ABO locus; nonsecretor (se/se) individuals fail to produce type 1 and type 2 H in their secretions whereas red cell (type 2) H expression is unaffected.

Expression of ABO on red cells may be deliberately modified by treatment with glycosidases: an a-galactosidase extractable from green coffee beans can remove galactose from group B red cells and hence enzymically convert them to group O. Such 'ECO' (enzymically converted group O) red cells may survive normally when transfused to group O subjects. However, the clinical potential of such techniques is limited.

Biosynthesis and ontogeny

The A- and B-transferases and their respective antigens in adults are most abundant in intestinal and gastric mucosa, lungs and salivary glands. Significant levels are found in kidneys, bladder, urothelial cells, bone marrow and hematopoietic cells. The enzymes exist in the cytoplasm and bound to the membranes of the cells of the above tissues and in the membranes of red cells and platelets. The transferases are in free solution in the transtubular network and their relevant secretions, e.g. mucin droplets, plasma, ovarian cyst fluid, milk, saliva. Molecules glycosylated by the transferases include membrane enzymes, membrane structural proteins and receptors, as well as secreted proteins, e.g. IgA. The ABO transferases, but not others, can be lost in various cancers, e.g. carcinoma of the bladder and of the colon. Both N-Iinked and O-linked oligosaccharide moieties of glycoproteins with A or B activity are synthesized in the Golgi.

During ontogeny, ABH (and Lewis) activity is at its highest in the early embryo from the 5th week postfertilization; ABH antigens are found in large amounts on endothelial cells and most epithelial pri-mordia, and in practically all early organs e.g. blood islands of the yolk sac, erythropoietic foci of the liver, digestive tube epithelia, pharyngeal pouches, the thymus, the pituitary, thyroid glands, trachea and bronchi, hepatic and pancreatic diverticula, the cloaca, urachos and allantois, mesonephros and the ducts of the metanephros. The CNS, liver, adrenal glands and secretory tubules show no ABH activity at this stage.

From the end of the 12-14th week of gestation, there is regression of ABH expression from epithelial cell walls, thyroid pituitary, and other glands and organs, to the adult vestigial pattern described above. The biological significance of this regression is unknown. The number of A and B sites on the red cell is increased approximately fourfold in adults as compared with newborn infants, such that there are 0.25-0.37 x 106 A sites per red cell in the newborn and 0.81-1.2 x 106 in the adult, as compared with 0.2-0.32 x 106 B sites per red cell in the newborn and 0.75 x 106 for adults.

A- and B-like antigens have been detected in a variety of microorganisms and animal tissues. During phylogeny, ABH antigens are confined to the endo-dermal tissues of amphibia and reptiles, and are present on the epidermis of all mammals studied to date. While ABH is found on baboon vascular endothelium, only humans and the great apes have red cells expressing ABH. ABO are 'histocompatibility' antigens; it is simply the peculiarity of their original description that has led to ABH antigens being thought of as 'blood groups'.

ABO antibodies

The clinical importance of the ABO blood group system derives from the universality of its antibodies and their in vivo potency. The 'naturally occurring' antibodies of the majority of group A or B individuals are mainly IgM and produced in response to environmental ABO antigens, e.g. from microbes in the gut and respiratory tract. Such IgM antibodies, although displaying optimal activity in the cold, are reactive at 37°C and can activate the complement cascade up to the C9 stage, leading to the immediate intravascular lysis of transfused incompatible red cells in vivo. In the UK, roughly one in every three randomly selected, ungrouped blood donations would be incompatible with a given recipient; such incompatible transfusions can lead to renal failure, disseminated intravascular coagulation and even death. The majority of the signs and symptoms of severe ABO hemolytic transfusion reactions can be attributed to the generation of C3a and C5a fragments as a result of complement fixation, with the consequent release of vasoactive amines from mast cells and of cytokines such as interleukin-1 (IL-1), IL-6, IL-8 and tumor necrosis factor a (TNFa) from mononuclear cells.

Most, if not all, group O adults, and a small proportion of group A and B subjects, have naturally occurring, usually weak, IgG in addition to stronger IgM ABO antibodies. The IgG component can cross the placenta and bind to fetal red cells; however, lysis of fetal red cells is generally minimal and hemolytic disease of the newborn (HDN) caused by ABO antibodies is usually mild or inapparent in Western Europe and North America. The occurrence of HDN due to ABO antibodies cannot be predicted, but it only affects the offspring of group O mothers. The lack of severity of most cases of ABO HDN is thought to be due to: 1) IgG ABO antibodies being predominantly IgG2, which is incapable of initiating complement-mediated hemolysis or destruction of antibody-coated red cells by the mononuclear phagocytic system (many sera have IgGl ABO antibodies as well as an IgG2 component; a few sera have trace amounts of lgG3 and IgG4 antibodies); 2) substantial amounts of maternal IgG ABO antibodies binding to ABO sites on tissues other than red cells in the fetus; 3) soluble ABH antigens in plasma and body fluids of the fetus which neutralize IgG anti-A and anti-B and inhibit their binding; 4) the ABO epitopes of fetal red cells being present on unbranched oligosaccharides, which are thought to be unable to support the divalent IgG binding needed for complement activation by IgGl or IgG3 antibodies; 5) fetal complement levels being too low to support efficient lysis of cord (or even adult) target red cells. However, in some parts of the world, ABO HDN is often more severe and this is attributed to environmental factors such as the stimulation of ABO antibodies by microbes and parasites.

Some individuals possess plasma IgA ABO antibodies irrespective of immunization; ABO antibodies of colostrum arc often wholly IgA, although sometimes IgM antibodies can also be found.

Cord blood usually does not contain ABO antibodies although maternally derived IgG anti-A or -B can sometimes be detected. Newborn infants do not produce ABO antibodies until the 3rd-6th month of age (median titer = 4), reaching a maximal titer (approx. = 128) between the ages of 5 and 10 years. Anti-A seems to attain higher titers more rapidly than anti-B. Adult titers of anti-A range from 32 to 2048 (median = 256) and anti-B from 8 to 512 (median = 64). The vast majority of healthy adults have readily detected ABO antibodies. Weakening of ABO antibodies can occur naturally in individuals aged over 50; a third of patients over 65 have ABO antibody titers of 4 or less. Occasional subjects may lack the appropriate ABO agglutinins, especially if hypogammaglobulinemia or if their plasma IgM levels are low. Antibodies can be lost by exhaustive plasma exchange (used therapeutically in ABO incompatible bone marrow and organ transplantation) or by immunosuppression caused by therapy or by disease. IgM anti-A or -B are completely absent in individuals with the very rare Wiskott-Aldrich syndrome.

See also: Alloantigens; Blood transfusion reactions; Cold agglutinins; Complement, classical pathway; Embryonic antigens; Erythrocytes; Forssman antigen; Hemolytic disease of the newborn; Maternal antibodies; Natural antibodies; Rh antigens.

Further reading

Clausen H, Bennett EP and Grunnet N (1994) Molecular genetics of ABO histo-blood groups. Transfusion Clinique Biologique 2: 79-89. Daniels G (1995) Human Blood Groups, Chapter 1, pp 8-

120. Oxford: Blackwell. Mollison PL, Engelfriet CP and Contreras M ( 199.5) Blood Transfusion in Clinical Medicine, 9th edn, pp 14S-20i. Oxford: Blackwell. Oriol R, Samuelsson BE and Messetov L i 1990) ABO antibodies - serological behaviour and immunochemical characterisation. Journal of Immunogenetks 17: I"74-299.

Schenkel-Brunner H (1995) Human Blood (.roups, Chemical and Biochemical Basis of Antigen Specificity, pp 47-147. New York: Springer-Verlag. Yamamoto F-I (1995) Molecular genetics of the ABO histo-blood group system. Vox Sanguinis 69: I-7.

Was this article helpful?

0 0
How To Bolster Your Immune System

How To Bolster Your Immune System

All Natural Immune Boosters Proven To Fight Infection, Disease And More. Discover A Natural, Safe Effective Way To Boost Your Immune System Using Ingredients From Your Kitchen Cupboard. The only common sense, no holds barred guide to hit the market today no gimmicks, no pills, just old fashioned common sense remedies to cure colds, influenza, viral infections and more.

Get My Free Audio Book

Post a comment