Monoclonal antibody-mediated inhibition of the human HLA alloimmune response to platelet transfusion is antigen specific and independent of Fcgamma receptor-mediated immune suppression

Trogocytosis drives red blood cell antigen loss in association with antibody-mediated immune suppression

ABSTRACT

Red blood cell (RBC) alloimmunization to paternal antigens during pregnancy can cause hemolytic disease of the fetus and newborn (HDFN). This severe and potentially fatal neonatal disorder can be prevented by the administration of polyclonal anti-D through a mechanism referred to as antibody-mediated immune suppression (AMIS). Although anti-D prophylaxis effectively prevents HDFN, a lack of mechanistic clarity has hampered its replacement with recombinant agents. The major theories behind AMIS induction in the hematologic literature have classically centered around RBC clearance; however, antigen modulation/loss has recently been proposed as a potential mechanism of AMIS. To explore the primary mechanisms of AMIS, we studied the ability of 11 different antibodies to induce AMIS, RBC clearance, antigen loss, and RBC membrane loss in the HOD (hen egg lysozyme-ovalbumin-human Duffy) murine model. Antibodies targeting different portions of the HOD molecule could induce AMIS independent of their ability to clear RBCs; however, all antibodies capable of inducing a strong AMIS effect also caused significant in vivo loss of the HOD antigen in conjunction with RBC membrane loss. In vitro studies of AMIS-inducing antibodies demonstrated simultaneous RBC antigen and membrane loss, which was mediated by macrophages. Confocal live-cell microscopy revealed that AMIS-inducing antibodies triggered RBC membrane transfer to macrophages, consistent with trogocytosis. Furthermore, anti-D itself can induce trogocytosis even at low concentrations, when phagocytosis is minimal or absent. In view of these findings, we propose trogocytosis as a mechanism of AMIS induction.

ABO Incompatibility between the Mother and Fetus Does Not Protect against Anti-Human Platelet Antigen-1a Immunization by Pregnancy

(1) Background: ABO blood group incompatibility between the mother and fetus protects against anti-D immunization by pregnancy. The possible role of ABO incompatibility in protecting against anti-human platelet antigen-1a immunization is unclear. (2) Methods: This study retrospectively screened 817 families (mother-father-neonate trios) of suspected fetal and neonatal alloimmune thrombocytopenia for inclusion. ABO genotypes were determined in 118 mother-child pairs with confirmed alloimmune thrombocytopenia due to anti-HPA-1a antibodies, and 522 mother-child pairs served as the control group. The expression of blood group antigen A on platelets was determined in 199 consecutive newborns by flow cytometry and compared with adult controls. (3) Results: ABO incompatibility between mother and fetus did not protect against anti-human platelet antigen-1a immunization by pregnancy. ABO blood groups of mothers and/or fetuses were not associated with the severity of fetal and neonatal alloimmune thrombocytopenia. The expression pattern of blood group A antigens on the platelets of newborns mirrored that of adults, albeit on a lower level. Blood group A antigen was detected on a subpopulation of neonatal platelets, and some newborns revealed high platelet expression of A determinants on all platelets (type II high-expressers). (4) Conclusion: The lack of a protective effect of ABO incompatibility between mother and fetus against anti-human platelet antigen-1a immunization by pregnancy may indicate that fetal platelets are not the cellular source by which the mother is immunized.

Induction of non-specific suppression in chicks by specific combination of maternal antibody and related antigen

Specific immune suppression in newly hatched chicks induced by specific maternal antibodies has been reported. Laying hens were immunized with dinitrophenyl-keyhole limpet hemocyanin (DNP-KLH). Purified maternal anti-DNP and non-specific immunoglobulin (Ig) Y antibodies were transferred by yolk sac inoculation to newly hatched chicks, and then, they were immunized with an optimum immunogenic dose of DNP-KLH at 1 and 4 weeks of age. Concentrations of anti-DNP antibodies in serum samples of these chicks were measured by using Enzyme-linked immunosorbent assay (ELISA). Proportions of T-cell subsets in peripheral blood of these chicks were also measured by flow cytometric analysis at 5 weeks of age (one week after the second immunization). Suppression of anti-DNP antibody response and down-regulation of CD3⁺CD4⁺ cells were observed in the chicks received high dose of maternal anti-DNP antibodies and immunized with DNP-KLH. On the other hand, normal anti-DNP antibody response and normal proportion of CD3⁺CD4⁺ cells were observed in the chicks received high dose of non-specific IgY antibodies and immunized with DNP-KLH. Furthermore, when chicks received high dose of maternal anti-DNP antibodies and immunized with DNP-KLH at 1 and 4 weeks of age and then with rabbit serum albumin (RSA) at 5 and 8 weeks of age, their primary anti-RSA response was also significantly suppressed. We indicate here that specific maternal antibodies can affect both B and T cell responses and induce non-specific suppression against different antigens. However, this non-specific suppression does not continue for a long time.

Antibody-mediated immune suppression of erythrocyte alloimmunization can occur independently from red cell clearance or epitope masking in a murine model

Anti-D can prevent immunization to the RhD Ag on RBCs, a phenomenon commonly termed Ab-mediated immune suppression (AMIS). The most accepted theory to explain this effect has been the rapid clearance of RBCs. In mouse models using SRBC, these xenogeneic cells are always rapidly cleared even without Ab, and involvement of epitope masking of the SRBC Ags by the AMIS-inducing Ab (anti-SRBC) has been suggested. To address these hypotheses, we immunized mice with murine transgenic RBCs expressing the HOD Ag (hen egg lysozyme [HEL], in sequence with ovalbumin, and the human Duffy transmembrane protein) in the presence of polyclonal Abs or mAbs to the HOD molecule. The isotype, specificity, and ability to induce AMIS of these Abs were compared with accelerated clearance as well as steric hindrance of the HOD Ag. Mice made IgM and IgG reactive with the HEL portion of the molecule only. All six of the mAbs could inhibit the response. The HEL-specific Abs (4B7, IgG1; GD7, IgG2b; 2F4, IgG1) did not accelerate clearance of the HOD-RBCs and displayed partial epitope masking. The Duffy-specific Abs (MIMA 29, IgG2a; CBC-512, IgG1; K6, IgG1) all caused rapid clearance of HOD RBCs without steric hindrance. To our knowledge, this is the first demonstration of AMIS to erythrocytes in an all-murine model and shows that AMIS can occur in the absence of RBC clearance or epitope masking. The AMIS effect was also independent of IgG isotype and epitope specificity of the AMIS-inducing Ab.

Does Rh immune globulin suppress HLA sensitization in pregnancy?

BACKGROUND

How Rh immune globulin (RhIG) prevents sensitization to D antigen is unclear. If RhIG Fc delivers a nonspecific immunosuppressive signal, then RhIG may inhibit sensitization to antigens other than D. HLA antibody prevalence was compared in previously pregnant D- versus D+ women to investigate whether RhIG suppresses HLA sensitization.

STUDY DESIGN AND METHODS

In the Leukocyte Antibody Prevalence Study (LAPS), 7920 volunteer blood donors were screened for anti-HLA and surveyed about prior pregnancies and transfusions. A secondary analysis of the LAPS database was performed.

RESULTS

D- women not more than 40 years old (presumed to have received antenatal with or without postpartum RhIG in all pregnancies) had a significantly lower HLA sensitization rate than D+ women (relative risk, 0.58; 95% confidence interval [CI], 0.40-0.83). When stratified by deliveries (one, two, three, or four or more), D- women not older than 40 were HLA sensitized less often than D+ women in every case. In contrast, a clear relationship between D type and HLA sensitization was not seen in older previously pregnant women whose childbearing years are presumed to have preceded the use of routine RhIG prophylaxis. In a multivariable logistic regression model, D- women not more than 40 years old remained significantly less likely to be HLA sensitized compared with D+ women after adjusting for parity, time from last pregnancy, lost pregnancies, and transfusions (odds ratio [OR], 0.55; 95% CI, 0.34-0.88).

CONCLUSION

Consistent with a nonspecific immunosuppressive effect of RhIG, younger previously pregnant D- women were less likely than previously pregnant D+ women to be HLA sensitized.

Mechanisms of anti-D action in the prevention of hemolytic disease of the fetus and newborn: what can we learn from rodent models?

PURPOSE OF REVIEW

Hemolytic disease of the fetus and newborn can be effectively prevented by administration of anti-D to the mother. In this setting, the IgG purified from the plasma of D-alloimmunized donors prevents the maternal immune response to D-positive red blood cells (RBC). Several monoclonal anti-D antibodies have recently been developed for potential use in the setting of hemolytic disease of the fetus and newborn; the functional assays used to assess the potential success of these antibodies have often assumed antigen clearance as the predominant mechanism of anti-D. Unfortunately, the in-vivo success of these monoclonal antibodies has thus far been limited. A similar inhibitory effect of IgG has been observed in animal models with a vast array of different antigens, referred to as antibody-mediated immune suppression (AMIS). Here, studies of AMIS are reviewed and the relevance of these findings for anti-D-mediated immunoprophylaxis is discussed.

RECENT FINDINGS

In animal models of AMIS, IgG-mediated antigen clearance was not sufficient for prevention of the antibody response to RBC. Furthermore, anti-RBC IgG inhibited B-cell priming to foreign RBC, but failed to prevent a T-cell response and immunological memory.

SUMMARY

The applicability of AMIS models for determining the true mechanism of anti-D, though uncertain, may nevertheless provide knowledge as to potential mechanisms of action of anti-RBC antibodies.

Mechanisms of anti-D action in the prevention of hemolytic disease of the fetus and newborn

Anti-D is routinely and effectively used to prevent hemolytic disease of the fetus and newborn (HDFN) caused by the antibody response to the D antigen on fetal RBCs. Anti-D is a polyclonal IgG product purified from the plasma of D-alloimmunized individuals. The mechanism of anti-D has not been fully elucidated. Antigenic epitopes are not fully masked by anti-D and are available for immune system recognition. However, a correlation has frequently been observed between anti-D-mediated RBC clearance and prevention of the antibody response, suggesting that anti-D may be able to destroy RBCs without triggering the adaptive immune response. Anti-D-opsonized RBCs may also elicit inhibitory FcgammaRIIB signaling in B cells and prevent B cell activation. The ability of antigen-specific IgG to inhibit antibody responses has also been observed in a variety of animal models immunized with a vast array of different antigens, such as sheep RBCs (SRBC). This effect has been referred to as antibody-mediated immune suppression (AMIS). In animal models, IgG inhibits the antibody response, but the T-cell response and memory may still be intact. IgG does not mask all epitopes, and IgG-mediated RBC clearance or FcgammaRIIB-mediated B-cell inhibition do not appear to mediate the AMIS effect. Instead, IgG appears to selectively disrupt B cell priming, although the exact mechanism remains obscure. While the applicability of animal models of AMIS to understanding the true mechanism of anti-D remains uncertain, the models have nevertheless provided us with insights into the possible IgG effects on the immune response.

Immunoglobulin G-mediated regulation of the murine immune response to transfused red blood cells occurs in the absence of active immune suppression: implications for the mechanism of action of anti-D in the prevention of haemolytic disease of the fetus and newborn?

Anti-D has been widely and effectively used in Rhesus blood group D negative mothers for the prevention of haemolytic disease of the fetus and newborn; its mechanism of action however, often referred to as antibody-mediated immune suppression (AMIS), remains largely unresolved. We investigated, in a murine model, whether active immune suppression or clonal deletion mediated by anti-red blood cell (RBC) immunoglobulin G (IgG) could explain the phenomenon of AMIS. Transfusion of IgG-opsonized foreign RBCs (i.e. AMIS) strongly attenuated antibody responses compared to transfusion of untreated foreign RBCs. When the AMIS-mice were subsequently transfused with untreated RBCs, no immune suppression was observed at 5 and 35 days after AMIS induction; in fact, the mice responded to retransfusion with untreated RBCs in a manner that was characteristic of a secondary immune response. When IgG-opsonized RBCs were transfused concurrently with untreated RBCs, a dose-dependent reduction of the antibody response was observed. This work suggests that the attenuation of the antibody responsiveness by anti-RBC IgG is not associated with active immune suppression or clonal deletion at either the T-cell or B-cell level; rather, the effect appears more characteristic of B-cell unresponsiveness to IgG-opsonized RBCs. These results may have implications for the understanding of the mechanism of action of anti-D in haemolytic disease of the fetus and newborn.

IgG-mediated immunosuppression is not dependent on erythrocyte clearance or immunological evasion: implications for the mechanism of action of anti-D in the prevention of haemolytic disease of the newborn?

Haemolytic disease of the newborn (HDN) can be prevented by the passive administration of anti-D to the mother. The most accepted theory to describe this activity of anti-D is based upon its ability to clear opsonized erythrocytes before their recognition by the maternal immune system. We examined this hypothesis using a murine model of immunity to foreign erythrocytes. Whereas transfusion of foreign erythrocytes into mice induced immunoglobulin (Ig)M and IgG antibodies specific for the erythrocytes, these humoral immune responses were inhibited when the erythrocytes were opsonized with IgG. To specifically determine if immunological evasion occurs with these opsonized erythrocytes, we examined T-cell responses from these mice. An erythrocyte-specific T-cell response was clearly detected. We then tested whether phagocytosis of opsonized erythrocytes is sufficient to prevent the antibody response. We exposed mononuclear phagocytic cells to sheep red blood cells (SRBC) in vitro and then adoptively transferred the phagocytic cells to recipient mice; opsonized SRBC unexpectedly increased, rather than decreased, the antibody response. These data indicate that removal of opsonized erythrocytes by phagocytic cells does not prevent their immunological recognition and suggest that antigen clearance may not be the predominant mechanism of anti-erythrocyte action in downregulating the humoral immune response.

Fc-signalling in the modulation of immune responses by passive antibody

Passive antibody can both suppress and augment immune responses. Until recently, there was virtual unanimity on the importance of the interaction of the Fc portion of modulating antibody with Fc-receptors (Fc-signalling), especially in experiments involving the suppression by antibody. Experiments reported in the last few years, that do not demonstrate the range of Fc-portion/Fc-receptor influences on the suppression of immune responses by passive antibody, have introduced new uncertainty into this field. The purpose of this paper is to review how the initial controversy on the influence of Fc-signalling in inhibition by passive antibody was resolved. Old and new approaches are suggested that may help in resolving the current uncertainty engendered by recent experimental results that were interpreted to mean that passive suppressive antibody does not utilize the inhibitory FcgammaRIIB receptor. An understanding of the factors that influence negative Fc-signalling is needed in order to optimize clinical therapies whose action depends on the suppressive property of antibody.