"Tip-toeing” to an assay for transplantation tolerance?

In 2000, Orosz and colleagues published a seminal paper using an in vivo delayed-type hypersensitivity (DTH) assay to investigate the immunological basis of human allograft acceptance (1). Donor-specific DTH responses, elicited by the indirect pathway of alloantigen presentation, were measured following injection of recipient peripheral blood mononuclear cells (PBMCs) and donor-cell lysate into the footpad of severe combined immunodeficient (SCID) mice. PBMCs from rejecting individuals triggered strong swelling reactions associated with human cell retention and mouse neutrophil recruitment while PBMCs from clinically tolerant recipients induced reduced or no DTH responses. PBMCs from tolerant recipients retained normal DTH responses to third-party recall antigens such as tetanus/diptheria toxoid or Epstein-Barr virus, but these responses were suppressed in a TGFß- and IL-10-dependent manner when donor and third-party antigens were co-injected. The demonstration that allograft acceptance was associated with reduced donor-specific DTH responses and also increased linked-suppression provided evidence that immune regulation was an important mechanism for renal allograft acceptance in humans.

In this issue of the American Journal of Transplantation, Haynes et al. (2) used this trans-vivo DTH assay to compare the strength of donor-specific and third-party responses across five kidney transplant patient groups: identical twin acceptance (TWIN, n=2), clinically tolerant (TOL, n=11), steroid monotherapy (MONO, n=7), standard immunosuppression (SI, n=18) and chronic rejection (CR, n=7). Indirect anti-donor DTH responses that were IFN-γ and IL-17-dependent were observed in decreasing strengths in the CR>SI>MONO>TOL>TWIN groups. In contrast, linked-suppression of third-party responses was strongest with TOL PBMC, and progressively reduced in the MONO/IS and CR groups. Thus, the degree of linked-suppression inversely correlated with immunosuppression requirements.

Two recent studies sponsored by the Immune Tolerance Network (ITN) and Reprogramming the Immune System for the Establishment of Tolerance (RISET) consortia reported an enriched B cell gene expression signature in tolerant patients and a cross-platform biomarker- and microarray-based index of tolerance (3, 4). These findings, coincident with emerging literature on IL-10-producing regulatory B cells (5), have fueled a hypothesis of regulatory B cells playing a key role in transplantation tolerance. Haynes et al. (2) observed that the TOL and TWIN groups had similar numbers of naïve B cells, which were significantly higher than all other groups receiving immunosuppression. Furthermore the index of tolerance was reached for all patients in the TOL group. However, no gradation in B cell numbers was observed in MONO, versus IS versus CR groups, and the index of tolerance was only achieved in two patients in the SI group and none in the MONO group. Thus, in a head-to-head comparison of these three tolerance assays, only the trans-vivo DTH assay revealed increasing strength of regulatory tolerance with decreased requirement for immunosuppression. Could this outcome be analogous to the biblical story of David and Goliath, and suggest that the trans-vivo DTH assay is more sensitive for diagnosing tolerance than the more technologically advanced techniques utilized by ITN and RISET? If so, the challenge will be to modify the trans-vivo DTH assay into a more user-friendly format that does not require mice tip-toeing on footpads injected with PBMCs and antigens. The alternative interpretation is that the reasoning is incorrect, and that the strength of regulation in the MONO group need not be greater than in the SI and CR groups. But then, how could IS drugs be eliminated in the MONO recipients without triggering acute rejection?

Haynes et al. also used the trans-vivo DTH assay to test whether regulatory B cells are essential for linked-tolerance. Consistent with the lack of a role for IL-10, removal of B cells did not eliminate linked-suppression (2). Moreover, although 2 of the 3 TOL patients with the lowest regulation also had the lowest numbers of B cells, potentially correlating B cells with linked-suppression, the only TOL patient with no regulation had among the highest numbers of naïve B cells. Thus there is limited evidence overall for regulatory B cells driving linked-suppression in this cohort of tolerant patients. It however remains possible that a role of B cells can be uncovered with a different tolerance assay or in other patient cohorts.

Our understanding of how transplantation tolerance is maintained in both experimental rodent models and in humans remains incomplete. Orosz, Sykes and colleagues were prescient in their demonstration that the tolerant state can evolve over time, with regulation playing a critical role early but becoming subsumed by other mechanisms including deletion, T cell anergy and/or exhaustion. Their observations along with those of Haynes et al. (2) raise questions of whether different mechanisms of tolerance are employed in different patients and towards different organs, whether a signature of tolerance established early post-transplantation evolves to a completely different one over time, and what is the cause of this evolution? Finally, what is the best approach to developing assays that take into account the potential diversity and evolution of the tolerance signature?