Blood group antigens and immune responses–detailed knowledge is necessary to prevent immunization and to follow up immunized individuals


  • 3B-S03-01

A. Husebekk, University Hospital North Norway, 9038 Tromsø, Norway


Background  The immune system is educated to detect and react with foreign antigens and to tolerate self-antigen. Transfusion of blood cells and plasma and pregnancies challenge the immune system by the introduction of foreign antigens. The antigens may cause an immune response, but in many instances this is not the case and the individual is not immunised after exposure of blood group antigens.

Aims  The aim of the presentation is to dissect some immune responses to blood group antigens in order to understand the mechanism of immunisation.

Methods  The results of immune responses to blood group antigens can be detected by the presence of antibodies to the antigens. If the antibodies are of IgG class, the activated B cells have received help from antigen specific T cells. Both antibodies, B cells and T cells can be isolated from immunised individuals and studied in the laboratory. Also B-cell receptors and T-cell receptors as well as MHC molecules on antigen presenting cells can be studied and models of the immune synapses can be created in vitro.

Results  The most classic immune responses in transfusion medicine and in incompatible pregnancies are immune responses to the RhD antigen on red cells, HLA class I molecules on white cells and platelets and human platelet antigens. The nature of these antigens are different; RhD antigens are part of a large complex, present on red cells from RhD positive individuals and completely lacking on red cells from RhD negative individuals. It is likely that many peptides derived from this antigen complex may stimulate T cells and B cells. HLA antigens are highly polymorphic and the antigens are known to induce strong alloimmune responses. The HPA antigens are created by one amino acid difference in allotypes based on a single nucleotide polymorphism at the genetic level. HPA 1a induce immune responses in 10% of HPA 1b homozygote pregnant women. The result of these immune responses is destruction of blood cells with clinical consequences connected to the effect of transfusions or the outcome of pregnancies.

Summary/Conclusions  Even though there is emerging knowledge about the immune responses to some of the blood group antigens, more information must be gained in order to understand the complete picture. The action of the innate immune response initiating the adaptive immune response to blood group antigens is not well understood. A detailed understanding of both the innate ad the adaptive part of the immune response is necessary to identify individuals at risk for immunisation and to prevent immunisation to blood group antigens.

All allogeneic blood transfusions challenge the immune system by introducing non-self molecules that frequently are targeted by the immune system. Usually no alloantibodies are generated, but in a few cases, we can detect alloantibodies showing that an immune response has taken place. What does this tell us about immune responses to blood group or platelet antigens?

This question will be discussed in the context of the immune response to the human platelet antigen (HPA) 1a. This immune response has been studied in more detail than immune responses to many other blood group or platelet antigens. It may be regarded as a typical immune response to blood group or platelet antigens, defined by a single nucleotide polymorphism (SNP) resulting in a difference of one amino acid between the two allotypes.

Foetal and neonatal alloimmune thrombocytopenia (FNAIT)

Foetal and neonatal alloimmune thrombocytopenia is caused by the transfer of maternal alloantibodies against platelet antigens from mother to foetus. Foetal platelets are then removed by the foetal reticuloendothelial system as a result of the maternal opsonizing antibodies. The foetus or neonate may become thrombocytopenic and at risk of bleeding. The most severe complication is intracranial haemorrhage (ICH), which may induce foetal or neonatal death, or survival with lifelong disabilities. Although FNAIT can be recognized in 1:1200 pregnancies in Caucasians, ICH occurs more rarely, probably in 1:12 500–25 000 pregnancies [1,2].

The antigen

In most cases of FNAIT, the HPA-1a antigen is responsible for the immunization. HPA-1 is located on β3 integrin in the glycoprotein IIb/IIIa (αIIb/β3) complex on the platelet surface. The glycoprotein (GP) complex carries the receptor for proteins carrying RGD motifs such as fibrinogen. In Caucasians, 98% carry at least one HPA-1a allotype (Leu33), whereas two per cent are homozygous for HPA-1b (Pro33) and are therefore susceptible for immunization against the Leu33 variant. The basis of the amino acid difference is a SNP. The β3 integrin is, in addition to being a part of the GP IIb/IIIa molecule, also a part of the vitronectin receptor on trophoblasts, endothelial cells and sperm. Immunization may take place as a consequence of incompatible pregnancies or incompatible transfusions. In pregnancies, only 10 per cent of HPA-1a negative women are immunized based on detection of anti-HPA-1a antibodies [3].

The innate immune response

Little, if anything, is known about the innate immune response to HPA-1a. It is reasonable to assume that the adaptive immune response associated with antibody production is preceded by activation of innate immune cells, including antigen-presenting cells (APC), which is a requirement for progression of the adaptive response. This would require proinflammatory signals, which normally are inherent in infectious agents such as LPS on gram negative bacteria that bind to Toll-like receptors on APCs, or “danger signals” such as the release of intracellular content like ureic acid from necrotic cells. In the case of FNAIT, the source of proinflammatory signals is unknown. Candidates could include “danger signals” associated with the pregnancy itself or activated foetal platelets that could potentially activate APCs directly.

The adaptive immune response

The adaptive immune system can respond vigorously to alloantigens, in particular, foreign MHC molecules. These molecules can be recognized as such on foreign cells (direct allorecognition) or processed and presented in the context of self-MHC (indirect allorecognition). Minor alloantigens, i.e. HPA-1a in an HPA-1b homozygous individual (HPA-1bb), are recognized in the context of self-MHC, and production of anti-HPA-1a antibodies in an HPA-1bb individual mirrors a complex immune response of which only some elements are known.

T cells and antigen presentation by MHC

As mentioned earlier, HPA-1a represents a foreign antigen in HPA-1bb women. However, the presence of HPA-1a by itself is not sufficient for the initiation of antibody production. There are certainly additional requirements. In that respect, it has long been known that there is a strong association between the presence of the MHC allele HLA-DRB3*0101 and production of antibodies against HPA-1a. This allele is present in more than 90% of HPA-1a-immunized women compared to <30% in the general population. This association suggests that anti-HPA-1a antibody production may be dependent on T-cell responses; T cells recognize foreign antigens as peptide bound to MHC molecules. Binding of the T-cell antigen receptor to specific peptide–MHC complexes on the surface of antigen-presenting cells activates the T cell. Indeed, most antibody responses are T-cell dependent; while B cells become activated by cross-linking of immunoglobulin antigen receptors, differentiation into antibody-producing plasma cells usually requires “help” from activated antigen-specific CD4 “helper” T cells. Therefore, antigen-specific CD4 T cells control the initiation and progression of most antibody responses. In the case of FNAIT, the strongest support for the presence of T-cell responses associated with anti-HPA-1a antibody production is the recent isolation of HPA-1a-specific CD4 T cells clones from HPA-1bb women who have undergone pregnancies with HPA-1a-positive children. Furthermore, these clones are restricted by the heterodimeric MHC class II molecule encoded by the HLA-DRB3*0101 allele (together with the invariant HLA-DRA): the same allele associated with the FNAIT. Studies of the fine specificities of these T-cell clones are providing insight into the biology of anti-HPA-1a responses. Findings from these studies may be translated into therapeutic potential to prevent anti-HPA-1a antibody production and FNAIT [4,5].

B cells and anti-HPA-1a antibodies

In HPA-1a-immunized women, anti-HPA-1a antibodies of IgG class can be detected, most often IgG1 or IgG3 isotypes. Both subclasses can cross the placenta from maternal to foetal circulation. It is shown that there is a correlation between antibody level in maternal plasma and the severity of thrombocytopenia in the newborn [6].

While much effort has been aimed at studies of antibodies that can be isolated from the plasma of HPA-1a-immunized women, the biology of the B cells and plasma cells that produce the antibodies has got less attention. In this regard, we have initiated studies in our laboratory aimed at elucidating the cellular requirements for activation and differentiation of HPA-1a-specific B cells. These studies will also address the nature of HPA-1a-specific T-cell and B-cell interaction and examine in which tissue naive B cells are primed by the HPA-1a antigen. In the course of finding answers to these questions, we hope to identify strategies to block progression of B-cell activation and plasma cell differentiation to block the formation anti-HPA-1a antibodies (Fig. 1).

Figure 1.

 The figure shows the interaction between B and T cells inducing B-cell differentiation into anti-HPA-1a-producing plasma cells in an HPA-1a negative individual.


Studies of the immune reaction to the platelet antigen HPA-1a have elucidated certain details of the immune response to this particular antigen. This insight may shed new light on the immune response to blood group and platelet antigens in general. Important issues that remain to be solved are: (i) we do not know anything about the innate immune response to the antigen. (ii) We cannot explain why only certain women are at high risk of being immunized. (iii) We do not know why thrombocytopenia develops in some pregnancies without detectable antibodies, while in some pregnancies with high levels of antibodies the platelet counts are normal. (iv) We are unable to explain why intracranial haemorrhage occur in some cases of severe thrombocytopenia, while in other cases with equally low platelet count there are no bleeding symptoms.

The knowledge we have about FNAIT serves as the basis of possible immune modulation to prevent immunization or making the antibodies less harmful in already immunized women. Detailed understanding of the immune response towards blood group antigens is necessary for modulation of the immune response.