Intragraft B Cells, Just Not Like the Others.

Abstract

B-cell infiltrates are found in most human transplants during rejection. Despite years of research, however, their role and exact contribution to the rejection process are still enigmatic. As a first step toward cracking this vexing puzzle, investigators have begun characterizing the phenotype and specificity of these infiltrating cells. In recent years, a few characteristics have emerged from studies conducted primarily with human kidney and heart transplant specimens.1,2 In a study published in this issue, Zheng and colleagues used a fully allogeneic Balb/c into B6 kidney transplant model to address some of these questions (see Zhang et al3 in this issue). Aside from its technical prowess—try transplanting a mouse kidney in your spare time—the experimental model appears clinically relevant because it recapitulates most features of antibody-mediated rejection and T cell–mediated rejection such as donor-specific antibodies, T- and B-cell infiltrates in the graft interstitium, glomerulitis, and C4d deposition in structures resembling peritubular capillaries. Allografts are lost within 1 mo with a median survival time of 21 d. Using μMT mice as recipients decreased the severity of rejection, indicating that B cells were involved in the rejection mechanism. Remarkably, CD4+ T-cell infiltrates were unchanged in these mice, suggesting that B-cell deficiency did not overtly impact the development of T cell–mediated rejection. The authors took advantage of the model to further characterize the graft-infiltrating B cells. Some features had already been observed in human B cells; others are novel and point to an unrecognized function of these cells. Among the common characteristics between mice and men is the fact that infiltrates include a higher proportion of memory cells. This had previously been reported for human kidney and heart transplants.1,2,4,5 Another striking similarity noted between intragraft B cells found in this model and in humans is their peculiar transcriptomic signature, including the expression of genes attributed to innate-like B cells such as Ahnak.1 In contrast, a number of differences were also noticeable. In particular, the fact that graft B cells predominantly express germline-encoded immunoglobulin genes and show only a low level of clonal expansion is perplexing and in disagreement with the past analysis of human kidney and heart transplant infiltrates.1,6,7 This discrepancy could result from short time, only 20 d, between transplant and rejection compared to months or years in human allograft recipients. Another apparent difference comes from the reactivity of these cells. Mouse-switched intragraft B cells appear enriched for allospecific cells, although this is more questionable in humans.1,2,8-10 The authors then investigated the functional properties of intragraft B cells and especially their capacity to promote the proliferation of CD4+ T cells. This part of the study is particularly interesting because it represents one of the first attempts to determine the function of intragraft B cells presumably associated with their innate-like profile. Results revealed a decreased ability to expand CD4 T cells when compared to lymph node (LN) B cells. Moreover, using Foxp3IRES−GFPmice, a model familiar to this research group, Zheng et al showed that intragraft B cells stimulate Tfh cells in a different manner than LN B cells, resulting in reduced proliferation and diminished cytokine secretion with the exception of granulocyte-macrophage colony-stimulating factors. Notwithstanding its inherent limitations, this experimental model offers invaluable insights into kidney graft-infiltrating B cells at the time of active rejection. For one, their contribution to the inflammatory reaction and tissue destruction is now substantiated. Two, and likely the most important observation corroborating previous findings in humans, intragraft B cells appear to be different from LN B cells in their molecular phenotype, especially through the expression of genes associated with an innate-like profile. Moreover, we learn from the present study that intragraft B cells also differ from LN B cells in their ability to interact with Tfh cells. These characteristics are definitely worth noting and raise additional questions. Are intragraft innate-like B cells related to conventional CD5+ B1 B cells? Are they enriched in autoreactive or polyreactive clones as suggested by their innate-like phenotype as documented for human intragraft B cells?1,2 More relevant to patient care, would removing these cells therapeutically correct ongoing rejection? These important questions can now be addressed using a relevant model like the one described by Zheng et al.