“Allo” from the (marginal) zone

In this issue of Blood, Zerra et al¹  have identified a unique pathway of induction of antibodies to transfused red blood cell (RBC) antigens involving marginal B cells and CD4⁺ T cells.

RBC transfusions remain a life-savingtreatment for millions worldwide. However, patients can develop antibodies to transfused RBC minor antigens that are not routinely matched, causing life-threatening transfusion reactions and delays in transfusion once alloimmunized. Although progress has been made in our understanding of factors that influence RBC alloimmunization, the immune activation pathways involved in the initiation of RBC alloantibody formation remain elusive.^2-5  Delineation of these pathways can help in the development of existing or new targeted therapies for prevention or inhibition of alloimmunization. Studies in mice have identified mature B cells that reside in the marginal zone (MZ) of the spleen at the boundary between red and white pulp, hence the name MZ B cells, as playing a crucial role in the primary immune response to protein antigens.⁶  These B cells are considered a first line of defense against blood-borne pathogens, rapidly producing antibody responses to T-cell–dependent and T-cell–independent antigens. This contrasts with T-cell–dependent antibody responses by conventional follicular (FO) B cells, located in clusters or follicles in lymphoid organs adjacent to T-cell zones, which take longer to develop and give rise to high-affinity antibodies and memory B cells. Unlike well-characterized immune responses to soluble antigens or infectious agents, our understanding of the antibody response to RBC antigens has lagged simply as a result of the lack of molecular tools. However, with the recent development of several animal models expressing model RBC blood group antigens,²  it is now possible to examine immune activation pathways and regulation of antibody responses to transfused RBCs, including the role of MZ or FO B cells in the induction of humoral response in vivo. Using RBCs from transgenic mice expressing the Kell blood group antigen, it was recently shown that immunoglobulin G (IgG) anti-KEL antibody formation develops through a MZ B-cell–dependent pathway, although the antibody response was T-cell independent.⁷ 

To determine whether other RBC antigens can induce antibodies using a similar pathway, Zerra et al have now used the HOD target antigen (hen egg lysozyme [HEL] and ovalbumin [OVA] fused with the human RBC antigen Duffy) as an alternative model system. Similar to the KEL RBC alloimmunization murine model, MZ B cells were required for an antibody response to HOD RBC antigen, because removal of MZ B cells through floxed Notch2 deletion in mouse B cells or by using antibodies (anti-CD11a and anti-CD49d) inhibited alloimmunization following transfusion with HOD RBCs. Also similar to the anti-KEL antibody response, alloimmunization to HOD RBCs was not affected after depletion of FO B cells. However, IgG anti-HOD antibody formation required CD4⁺ T-cell help, which differs from the T-cell–independent immune response to transfused KEL RBCs. This difference strongly suggests that RBC blood group antigens can elicit an IgG response through different immune-activation pathways. Although, in retrospect, this may not be surprising, given that protein antigens can activate B cells with or without assistance from T cells, it was important to experimentally demonstrate that this also applied to blood group antigens on RBCs. It is known that the degree of CD21/35 engagement on antigen-specific B cells with fixed complement components can regulate the overall B-cell response and dependence on CD4 T cells.⁸  The investigators postulate that, in addition to intrinsic differences in the antigens themselves, which may also allow differential engagement of B-cell receptors, differences in the ability of IgM antibodies to fix complement following engagement of HOD or KEL during the early immune response may be one reason why distinct RBC antigens induce different immune responses. Understanding what influences these differences will be important to explore in future studies examining these important pathways.

The study also highlighted several important features of the B-cell response to an RBC antigen, in part as a result of the availability of a wealth of molecular tools against OVA and HEL portions of the HOD antigen. For example, the overall anti-HOD alloantibody levels persisted for ≥3 months, underscoring the emerging role of CD4 T-cell–driven extrafollicular B-cell responses in producing long-lived antibody secreting cells⁹  that have classically been associated with germinal center responses. Transfused HOD RBCs colocalized with MZ B cells for up to 14 days without any movement of HOD RBCs into the B-cell follicle. Furthermore, MZ B-cell depletion did not affect the localization of transfused RBCs within the marginal sinus or CD4 activation, suggesting that other MZ constituents may be responsible for the localization of HOD RBCs. Based on these data, the investigators propose a working model whereby HOD alloimmunization involves the engagement of HOD-specific CD4 T cells with antigen-specific MZ B cells, without the requirement for MZ B-cell trafficking to the B-cell follicle.

In summary, the present study builds on the premise that antibody responses against RBC antigens can involve the participation of MZ B cells rather than FO B cells which are associated with high affinity antibody responses. In terms of translational applicability of the present finding, the investigators note that even a transient depletion of MZ B cells can prevent IgM or IgG anti-HOD antibody formation, supporting the exciting possibility of using short-term depletion of MZ B cells for suppressing alloimmunization.