In this issue of Blood Advances, Zhao et al1 improved our understanding of donor T-cell migration after hematopoietic stem cell transplantation (HSCT). Specifically, they demonstrate that the broad inhibition of chemokine receptors on donor effector T cells diminishes their migration into the gastrointestinal (GI) tract, a critical target organ of graft-versus-host disease (GVHD). In contrast, their capacity to migrate into the bone marrow (BM), where graft-versus-leukemia (GVL) effects primarily occur, was unaffected during GVHD. These effects were not alloantigen specific.
Allogeneic HSCT is a potential curative therapy for severe disorders affecting the hematological and immunological systems. Although donor T cells in the graft promote hematopoietic stem cell engraftment and eradicate hematopoietic malignant cells in the BM and lymphoid organs (ie, GVL effects), they also attack normal recipient tissue (ie, GVHD). Given that severe acute and chronic GVHD, particularly steroid-refractory GVHD, cause high mortality and morbidity after HSCT, strategies to minimize GVHD while maximizing GVL effects remain a major unmet need in the field.2 Shlomchik et al recently demonstrated that following the entry of a small number of progenitor-like T cells into GVHD targets early after transplantation, donor alloreactive T cells predominantly expand and are maintained locally thereafter.3 This suggests that the inhibition of initial donor T-cell entry into the various organs early after transplantation may determine subsequent GVHD. In the current study, the authors investigated the effects of broad inhibition of chemokine receptors on effector T-cell infiltration into GVHD target organs (eg, small intestine, colon, skin, and liver) and organs wherein GVL effects occur (eg, BM, spleen, and blood). This was done using 2-photon intravital microscopy and flow cytometry, which included methods to separate extravascular infiltrating cells from locally residing intravascular cells.
Given that donor T-cell migration into target tissue results in acute GVHD, chemokine gradients within target organs and the expression of cognate chemokine receptors by donor T cells should theoretically control this process. For example, CCR7-deficient naïve donor T cells have diminished capacity to invoke GVHD.4 The absence of CCR2 expression by donor CD8+ T cells or CCR6 expression by CD4+ T cells attenuates GVHD.5,6 The role of CCR5 expression on donor T cells is dependent on pretransplant conditioning: CCR5-deficient donor T cells mitigate GVHD following nonirradiation conditioning,7 but enhance GVHD following myeloablative radiation–based conditioning.8 This difference has been attributed to the rapid induction of multiple chemokines within the GVHD target organs following irradiation.8 In addition, CCR5 expression on activated regulatory T cells (Treg) is required for Treg infiltration into target organs during GVHD, contributing to CCR5-dependent mitigation of GVHD following irradiation-based conditioning.9 CXCR3 expression on donor T cells has been shown to increase GVHD lethality in major histocompatibility complex (MHC)-matched recipients but not in MHC-mismatched recipients,10 suggesting a potential difference in chemokine induction between the 2 models. Thus, the migration of effector T cells may be altered by both GVHD-dependent and -independent variables.
Zhao et al compared effector T-cell trafficking between the BM and spleen in GVHD and non-GVHD recipient mice, with or without pertussis toxin (PTX) treatment of T cells, which enzymatically disrupts chemokine receptor signaling. They noted that in the presence of a graft-versus-host (GVH) response, chemokines are no longer required to draw T cells into BM and spleen but remain critical for their recruitment to GVHD target tissues such as the GI tract. Thus, PTX treatment of effector CD8+ T cells did not impair their migration to the BM, nor their capacity for antigen-specific leukemia killing.
The authors focused on the early trafficking of nonalloreactive effector T cells that were isolated from immunized spleen. This transfer of effector T cells mimics adoptive cell therapy with transgenic T-cell receptors or chimeric antigen receptor T cells, which are usually devoid of alloreactivity. However, these insights may also be useful to modulate alloreactive T cells after HSCT. Although donor CD8+ T cells are primarily activated by recipient hematopoiesis-derived antigen-presenting cells,11,12 they are detected in the lymphoid organs 1 day after transplantation,13 suggesting very rapid trafficking and redistribution.3 The contextual nature of antigen presentation potentially influences T-cell recruitment and differentiation.14 To avoid potential confounding factors and evaluate solely chemokine-dependent T-cell behavior, the authors expanded and isolated ovalbumin (OVA)–specific effector CD8+ T cells in vivo from the spleen of OVA-immunized mice, before transferring them into (OVA) antigen-free mice that had been transplanted with T-cell–replete (GVHD) or T-cell–depleted BM (non-GVHD) 14 days earlier. They then assessed their antigen-independent tissue immigration 3 hours later.
The following questions remain. How does the GVH reaction abrogate chemokine-dependent BM migration of effector T cells? How do alloantigen presentation and chemokine gradients interact to guide effector T-cell recruitment into tissue? What chemokine receptors, other than CXCR3 tested in the current study, are PTX sensitive and responsible for the observed effects? Most importantly, does chemokine receptor inhibition preserve a long-term GVL effect and how can chemokine-dependent T-cell migration be interfered with in a more permanent fashion (given that the PTX effect is transient), and is this even necessary? Nevertheless, the current study has demonstrated the crucial and differential roles of chemokine receptors in organ-specific effector T-cell trafficking and the role of GVHD in this process. Given the often redundant role of individual chemokines in T-cell trafficking in highly inflammatory situations that are characteristic of adoptive T-cell therapies, this study provides a rationale and context for broader approaches. These approaches may minimize immune toxicity while allowing for effective antitumor responses in sites relevant to hematological malignancies.
Conflict-of-interest disclosure: M.K. declares no competing financial interests.
