Eosinophil-Targeted Therapy: Not the Panacea for Antibody-Mediated Rejection?



The study of Cravedi et al (page 2696) challenges the hypothesis that eosinophils are an essential component of the long-lived plasma cells niche in the bone marrow, and it suggests that the role of eosinophils in long-lived plasma cells survival might be mice strain-dependent.

Historically, eosinophils have been considered end-stage cells involved in immunity against parasitic infections and immunopathology in hypersensitivity diseases. Nowadays, eosinophils are considered multifunctional leukocytes implicated in several physiological processes, including tissue homeostasis, modulation of adaptative immune responses and innate immunity to certain microbes [1]. In 2011, an additional role for eosinophils was uncovered: eosinophils are a crucial component for the survival of long-lived plasma cells (PCs) in their bone-marrow niche [2]. From this publication, the hope rose that eosinophil-targeted specific agents might be introduced to treat antibody-mediated rejection (ABMR). These agents have recently entered clinical testing for several diseases. ABMR is one of the principal mechanisms affecting late graft survival in the context of solid organ transplantation and no specific treatment is available yet. The primary source for long-lasting high-affinity donor-specific antibodies (DSAs) is the long-lived PC. The majority of long-lived PCs reside in the bone marrow and are associated with CXCL12-expressing reticular stromal cells. CXCL12 exerts its chemotactic action on cells that express its main receptor CXCR4, which is expressed on long-lived PCs, megakaryocytes, basophils and eosinophils. These innate myeloid cells are also present in the bone marrow niches, and importantly, these cells are producers of the proliferation-inducing ligand APRIL and IL-6. Both cytokines are crucial for the survival of long-lived PC in vitro and in vivo. In the current issue of The American Journal of Transplantation, Cravedi et al [3] challenge the role for eosinophils in the generation and maintenance of DSAs.

The investigators used ΔdblGATA1 mice on a BALB/c (TH2-phenotype, this strain was also used in the study of Chu et al [2]) and C57BL/6 (TH1-phenotype) that have a deletion in the high-affinity GATA-binding site of the promotor of the GATA1 gene, which results in the complete absence of mature eosinophils. In line with the study of Chu et al, the authors show that after alloantigen priming, the number of bone marrow PC was reduced in the ΔdblGATA1 BALB/c recipients. However, at the same time the spleen of these mice contained a higher frequency of PC. When the same experiment was performed with WT and ΔdblGATA1 on a C57BL/6 background, bone marrow PC numbers were comparable and spleen PC frequencies were even higher in the latter mice. This strain specificity was also observed in DSA titers. Comparable to the results described in the initial report by Chu et al, high-affinity antibody titers were lower in the ΔdblGATA1 on a BALB/c background. However, DSA titers were remarkably similar when the experiment was repeated on a C57BL/6 background. When they performed immunization with allogeneic splenocytes followed by heterotopic cardiac transplantation, the kinetics and strength of DSA were similar in the WT and ΔdblGATA1 mice on both genetic backgrounds.

Both studies pointed toward a role for eosinophils in the differential postgerminal center plasmablast chemotaxis that seemed to be strain-specific to a certain extent but did not result in the loss of DSA production per se. A comparable phenomenon in chemotaxis was observed in lupus-prone NZM mice that are known to produce large amounts of autoantibodies. In these mice, long-lived PCs were shown to be unresponsive to CXCL12, which resulted in absence of PCs in the bone marrow and subsequently migration toward the spleens instead, indicating that the spleen might be an alternative site of long-lived PC survival when bone marrow homing fails [4]. In lupus-prone NZB/W mice, long-lived PCs could even be found in the chronically inflamed kidneys [5]. The latter finding is actually not that surprising because in humans, functional germinal centers are also formed within the parenchyma of renal allografts that underwent chronic rejection. Besides eosinophils, other hematopoietic cells that find their progeny in a common myeloid cell have been related to PC longevity in the bone marrow. Previous work has shown that basophils play an important role in the maintenance of long-lived PCs after sepsis in C57BL/6 mice whereas a role for megakaryocytes was brought forward on both a BALB/c and the C57BL/6 background.

The work by Cravedi et al contributes to the view on the complexity of micro-environmental niches that contain resident mesenchymal cells, common myeloid progenitor-derived cells and long-lived PCs in a “ménage à trois” responsible for humoral memory. The current study questions the possibility of eosinophil-targeted agents as a therapeutical intervention to lower DSAs produced by long-lived PCs in the setting of ABMR. Further research should address which components, including route of immunization, antigen load and diversity, cell type responsible for the PC survival factors and interindividual differences, are applicable to humoral memory in the context of ABMR of solid organ transplants.


The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.