IgA function – variations on a theme

Authors


Jenny M. Woof, Division of Pathology and Neuroscience, University of Dundee Medical School, Ninewells Hospital, Dundee DD1 9SY, UK.

Immunoglobulin A (IgA), as the major class of antibody present in the mucosal secretions of most mammals, represents a key first line of defence against invasion by inhaled and ingested pathogens at the vulnerable mucosal surfaces. IgA is also found at significant concentrations in the serum of many species, where it functions as a second line of defence mediating elimination of pathogens that have breached the mucosal surface. Receptors specific for the Fc region of IgA, FcαR, are key mediators of IgA effector function and studies described in this issue of Immunology by Rogers and colleagues1 add to a growing appreciation of IgA function in different species.

Although IgA appears to share common protective functions in the different species where it is found, variations in gene number, allotypes and molecular forms have been noted. Humans, chimpanzees, gorillas and gibbons have two IgA heavy constant region (Cα) genes which give rise to the two subclasses, IgA1 and IgA2,2,3 whereas most other species examined (orangutan, rhesus and cynomolgus macaques, cow, horse, pig, dog, mouse, rat, echnida and possum) have just one Cα gene.2,4–11 An interesting exception is provided by the rabbit which has 13 Cα genes,12 of which 11 appear to be expressed.13 Single IgA genes may also assumed to be present in most birds as they have been described in chickens and ducks, considered to be among the most primitive extant birds.14,15

The two IgA subclasses in hominoid primates, having arisen through gene duplication, share considerable sequence similarity. The major difference between IgA1 and IgA2 resides in the hinge region that lies between the two Fab arms and the Fc region. IgA1 features a very extended hinge due to the insertion of a duplicated stretch of amino acids, which is lacking in IgA2. The longer hinge in IgA1 may have evolved to offer advantages in antigen recognition by allowing higher avidity bivalent interactions with distantly spaced antigens.16,17 However, this extended reach is accompanied by an increased vulnerability to proteolytic attack. Indeed, a number of important pathogenic bacteria, such as Streptococcus pneumonaie, Neisseria meningitidis and Haemophilus influenzae, have evolved enzymes independently that cleave specifically in the IgA1 hinge.18 The ability of these IgA1 proteases to compromise IgA function is thought to facilitate the initiation of infection at mucosal surfaces by these microorganisms.

Allelic variation in IgA has been investigated in some species but remains to be investigated in many others. In humans, there are two (or possibly three19) alleles of the IgA2 subclass. Rhesus macaque IgA also displays allelic polymorphism,4,20 while restriction fragment length polymorphism (RFLP) evidence points towards the existence of bovine IgA and equine IgA allotypes.21 Mouse IgA exists in different allelic forms that vary particularly in their hinge regions.22 The two allelic variants of pig IgA differ similarly in the hinge region.21

The serum distribution of the different molecular forms of IgA also varies from species to species. IgA has the capacity to form dimers, in which two monomer units, each comprising two heavy chains and light chains, are arranged in an end-to-end configuration stabilized by disulphide bridges and incorporation of J (joining) chain.23,24 For reasons that remain unclear, serum IgA is chiefly monomeric in humans (and presumably in other primates), but mainly dimeric in other animals.

Dimeric IgA, produced locally at mucosal sites, is transported across the epithelial cell boundary and out into the secretions by interaction with the polymeric immunoglobulin receptor (pIgR). During this process the pIgR is cleaved and the major fragment, termed secretory component (SC), becomes covalently attached to the IgA dimer. Thus SC is an integral part of the IgA molecule released into the secretions (secretory IgA) and is thought to help protect the antibody against degradation. Although IgA is the predominant Ig in the majority of secretions in most mammals, the proportion of IgA in colostrum and milk varies between species, reflecting the development of differing modes of transfer of antibodies from mother to offspring. In primates and rodents, IgG is transferred selectively across the placenta, and the chief Ig in colostrum is IgA. The serum antibody of the newborn is supplemented further in rodents with IgG derived from the colostrum that is actively transported across the neonatal gut by the neonatal Fc receptor (FcRn). In contrast, large mammals such as horses, pigs and cows, lack transplacental transfer of IgG. Instead, IgG is the major immunoglobulin class in colostrum and is absorbed across the neonatal gut in a brief time period prior to ‘closure’ at around 12 hr after birth. In the colostrum of these species IgA content is low, and it remains low in milk in cattle. However, in pigs and horses IgA is the major antibody in milk.21

From the above, it is clear that there are significant variations within the IgA systems of different species. A further important consideration relates to the triggering of IgA effector function through interaction with FcαR. The human FcαR (now termed FcαRI or CD89) was the first to be cloned and characterized (reviewed in Monteiro & van de Winkel25). Ligation of FcαRI by bacterial and viral targets coated in IgA initiates potent responses such as phagocytosis, respiratory burst and release of cytokines, to bring about elimination of the invading microorganism.25 Although FcαRI displays homology to other human FcR specific for IgG and IgE, namely FcγRI, FcγRII, FcγRIII and FcɛRI, it is a more distantly related member of the family. Indeed, it resembles more closely a family of receptors including natural killer cell Ig-like receptors (KIR) and NKp46 receptors, the genes of which flank that of FcαRI in the leukocyte Ig-like receptor cluster (LRC) on chromosome 19. In contrast, the FcγR and FcɛRI genes are clustered on chromosome 1.

While information on human FcαRI has amassed steadily, the FcαR systems in other species are not yet understood fully. The search for FcαR orthologues centred initially on the mouse, but despite exhaustive efforts no mouse equivalent was found. Unexpectedly, an IgG Fc receptor in cattle termed Fcγ2R was found to more closely resemble human FcαRI than human FcγR.26 However, it is only in the last few months that the first true orthologues of human FcαRI have been described in rats27 and cattle.28 The family is now extended further with the description of FcαR in Rhesus macaques.1

The similarity between all the receptors suggests that the FcαR gene may have been generally conserved in mammals. The chromosomal position of the receptor gene adjacent to killer cell receptor genes is a consistent feature in all species investigated. In humans, rats and cattle, the FcαR genes cluster with KIR and NKp46 genes (or their homologues).27–29 Rats are closer to mice in evolutionary terms than to gerbils and hamsters, but in Southern blots a rat FcαR probe showed hybridizing bands in gerbils and hamsters but not in mice.27 This finding, along with the fact that searches of the mouse genome database with known FcαR sequences fail to identify matches, indicates that the mouse genome has lost the FcαR gene. Studies on mouse killer cell receptors have thrown light on how this may have occurred. Mice were thought earlier to lack KIR-like genes but they have been found recently.30,31 Surprisingly, these are located on the X chromosome rather than within the murine LRC on chromosome 7 to which the mouse NKp46 gene localizes. Thus the mouse probably lost the FcαR gene during a translocation event when its KIR-like genes moved from the LRC to the X chromosome. It seems likely that mice have evolved alternative receptors for IgA. A receptor specific for both IgA and IgM Fc (Fcα/µR) has been characterized on B cells and macrophages32 and there is evidence that mouse IgA can trigger mouse neutrophils.33

Interest in the potential of antibody-based therapeutic approaches centred on IgA/FcαR continues to mount,34,35 and results with tumour and infectious targets are very encouraging.36–40 In order to assess readily the efficacy of such reagents, and also to understand better IgA-related diseases such as IgA nephropathy, suitable animal models could prove highly valuable. There are limitations with mouse models in this context, due to the lack of both FcαR orthologue and monomeric serum IgA. An increased knowledge of the IgA effector systems in other mammals will facilitate the development of appropriate models. Rhesus macaques are already used widely as models in numerous infectious diseases, including the simian immunodeficiency virus (SIV) or simian-human immunodeficiency virus (SHIV) model in AIDS research.41 The study of Rogers and colleagues1 will further the possibilities for macaque models relevant to the IgA system.

Acknowledgment

J.M.W. thanks the Leukaemia Research Fund for support.

Ancillary