Graft-versus-host disease (GvHD) remains a significant challenge following allogeneic hematopoietic cell transplantation (HCT). Prophylaxis and treatment strategies for GvHD also need to preserve graft-versus-leukemia (GvL) activity.1 Multiple molecular targets have emerged for GvHD, including the interaction of programmed cell death-ligand 1 (PD-L1), expressed in GvHD target tissues, with programmed cell death protein 1 (PD-1), expressed on donor CD4+ and CD8+ T cells. PD-L1/PD1 interaction generates T-cell exhaustion, anergy, and apoptosis to reduce GvHD activity. Alongside this, blocking interleukin-2 (IL-2) signaling in donor T cells may augment the PD-L1/PD1 interaction to synergistically attenuate GvHD. GvL activity, conversely, requires the proliferation and preservation of donor CD8+ and CD4+ T memory progenitor (Tmp) cells and T effector (Teff) cells in lymphoid tissues.
Song et al. recently published their findings in Blood, investigating the use of an anti-IL-2 monoclonal antibody, JES-6, in a preclinical murine model of GvHD.1 JES-6 interferes with IL-2 and dimeric IL-2 receptor interaction, through beta and gamma subunits, but not alpha. It is of interest to investigate whether the interference of IL-2 interaction with IL-2Rβ and IL-2Rγ receptors, expressed on donor T cells, can attenuate GvHD while also preserving GvL activity in lymphoid tissues. The role of PD-L1 was also investigated to differentiate independent and synergic mechanisms of GvHD attenuation. We summarise key findings of this study below.1
The authors performed a series of experiments with a primary goal to test whether JES-6 treatment attenuates GvHD and preserves GvL activity. The following questions were investigated.
- Does JES-6 attenuate GvHD?
- A murine model was used with BALB/c (H-2d) and C57BL/6 (H-2b) mice to create a histocompatibility mismatch. JES-6 treatment was compared with immunoglobulin G (IgG) and anti-IL-2-S4B6 controls, focusing on symptom reduction and tissue damage in GvHD target tissues.
- Does JES-6 preserve GvL?
- Mice were inoculated with luciferase-transfected BCL1 cells before HCT. The effect of JES-6 treatment on tumor cell clearance was compared with the calcineurin inhibitor tacrolimus (TAC), which inhibits IL-2 production from donor T cells.
- Does JES-6-IL-2 blockage require PD-L1 expression to attenuate GvHD?
- PD-L1−/− mice were given JES-6 treatment following HCT and compared with wildtype mice.
- What mechanisms are PD-L1 dependent?
- The authors investigated the effect of JES-6 treatment on T-regulatory (Treg) cells, T-helper type 1 cells (Th1), and cytotoxic T cell type 1 (Tc1) populations in wildtype and PDL1−/− mice, at Day 6 post-HCT.
- Is inhibition of IL-2-Stat-5 and AKT/mTOR pathways required to attenuate GvHD?
- Wildtype mice treated with JES-6 and IgG were compared for inhibition of IL-2-Stat-5 signaling.
- Wildtype and PD-L1−/− JES-6-treated mice were analyzed for expression of phosphorylated AKT/mTOR.
- How does JES-6 preserve GvL activity?
- Single-cell RNA sequencing (scRNA-seq) was used to generate gene cluster profiles for donor T cells in JES-6 and TAC-treated mice, at Day 7 post-HCT.
- The percentage of Tmp, Teff, and exhausted T (Tex) cells were compared in GvHD target tissues of JES-6 and TAC-treated mice.
Effect of JES-6 treatment on GvHD and GvL
JES-6 attenuates GvHD while preserving GvL activity
- JES-6 administration in mice reduced the severity of acute GvHD, with a reduction in symptoms including weight loss and diarrhea, when compared with both control IgG and anti-IL-2-S4B6. The authors also observed a reduction in damage of GvHD target tissues (liver, colon, and small intestine) in mice treated with JES-6 compared with control antibody groups.
- In terms of GvL, no effect on tumor growth was observed with JES-6, with all recipients dying after 20 days. However, for mice given both splenocytes and TCD-BM cells, JES-6 treatment led to a reduction in tumor growth compared with immunoglobulin controls, with all mice surviving over 30 days without relapse.
- Finally, using TAC as a control, the authors initially observed similar GvHD and GVL activity between the treatment groups, engrafted with spleen (2.5 × 106) and TCD-BM (2.5 × 106) cells, and inoculation with 5 × 106 Luc/BCL1 cells. However, when reducing the number of spleen cells (1.25 × 106) at the same concentration of inoculation, JES-6 treated mice managed to clear tumour cells by Day 12, while only 60% of mice treated with TAC cleared cells by Day 17 and the other 40% died by Day 30.
- At the highest dose (10 × 106 Luc/BCL1 cells), all TAC recipients died by Day 9, indicating that under these conditions, JES-6 showed improved GvL activity.
JES-6 attenuation of GvHD requires PD-L1 expression
- The authors observed a reduction in GvHD activity in wildtype mice treated with JES-6, but not in PD-L1−/− mice.
- The same was observed for the yield of CD4+ T cells but not CD8+ T cells in the liver and colon, which were markedly reduced in JES-6-treated wildtype mice compared with IgG controls; however, no difference was observed in PD-L1−/−-treated mice.
- Compared with IgG treatment, JES-6 treatment also increased the percentage of PD-1+ Eomes+ CD4+ and CD8+ T cells in the liver and colon tissues at Day 6 after HCT in the PD-L1−/− recipients. The percentages of PD-1+ Eomes+ CD4+ and CD8+ T cells in the liver and colon tissues of JES-6-treated PD-L1−/− recipients, were markedly lower than in JES-6-treated wildtype recipients at Day 6 after HCT.
PD-L1 interaction and JES-6 IL-2 interference synergistically reduce the percentage of GM-CSF-producing Th1 and Tc1 cells
- Granulocyte-macrophage colony-stimulating factor (GM-CSF)-producing Th1 and Tc1 cells are involved in the pathogenesis of acute GvHD, with release of this inflammatory cytokine leading to recruitment of monocytes and neutrophils to produce GvHD symptoms.
- The authors observed a reduction in the percentage of GM-CSF IFN-γ+ Th1 cells in the colon and liver of wildtype mice given JES-6, at Day 6 post-HCT compared with anti-IL-2-S4B6 controls. A reduction was also observed in PD-L1−/− mice, albeit far less than that of wildtype-treated mice.
- JES-6 treatment reduced the percentage but not yield of Tc1 cells in the liver and colon of wildtype mice compared with control, however this reduction was again lessened in PDL1−/− mice.
PD-L1-induced proliferation of IL-10+ Tr1 cells is required for attenuation of GvHD
- FoxP3+ Treg and FoxP3−IL-10+ type 1 regulatory (Tr1) cells are required to attenuate GvHD activity. JES-6 treatment increased the percentage of FoxP3+ Treg cells among infiltrating donor-type CD4+ T cells in the liver, but not in the colon of wildtype recipients.
- The authors did not observe any increase for either organ of PD-L1−/− mice with JES-6 treatment.
- Foxp3−IL-10+ Tr1 cells are in abundance in allo-HCT grafts. Expression of the nuclear factors Eomes and Blimp-1, is required for the proliferation of these cells, and the authors observed increased expression of such factors in CD4+ cells of wildtype mice with the JES-6 treatment compared with anti-IL-2-S4B6 and IgG controls; however, not in JES-6-treated PD-L1−/− mice.
JES-6 inhibition of IL-2-Stat-5 signalling and PD-L1 inhibition of AKT-mTOR pathways are required to attenuate GvHD
- In wildtype mice, inhibition of the IL-2-Stat-5 pathway of CD4+ and CD8+ T cells infiltrating the colon and spleen was observed with the JES-6 treatment, at Day 6 post-HCT compared with IgG controls.
- A reduction in phosphorylated AKT and mTOR in CD4+ and CD8+ T-cells was detected in the colon of JES-6-treated mice, unlike anti-IL-2-S4B6 treatment. In contrast, a reduced expression of phosphorylated forms was not observed in JES-6-treated PD-L1−/− mice, demonstrating PD-L1 is essential for this process.
JES-6 and TAC produce distinct populations of donor T cells in the spleen
- As JES-6 treatment produced significantly greater preservation of GvL activity compared with TAC, the authors next compared the populations of donor T cells between the treatment groups.
- scRNA-seq revealed that subsets of infiltrating CD4+ and CD8+ cells had similar expressional profiles; however, clusters of TCF-1+ CD4+ and CD8+ self-renewing memory progenitors were larger in the JES-6-treated group than in the TAC group. Such cells can further differentiate into cytolytic CD8+ T cells under CD4+ T-cell help. Thus, evidence was demonstrated for greater preservation of Tmp cells in lymphoid tissues of JES-6-treated mice.
JES-6 preserves CD8+ Tmp progenitors more effectively than TAC
- JES-6 treatment reduced the percentage of terminally differentiated effector CD8+ T-cells in the spleen compared with TAC treatment; however, there was a greater proportion of CD107a+/Granzyme B+ subset with stronger cytolytic function.
- Additionally, the percentage of CD8+ Tmp cells in the lymph node, liver, spleen, and colon of JES-6-treated mice, at Day 7 post-HCT were greater than the TAC group. The percentage of Teff cells were greater only in the lymph node and spleen of JES-6-treated mice compared with TAC, while conversely, the percentage of Tex cells was fewer in the lymph node and spleen for JES-6-treated mice.
Overall, these data provide preclinical evidence for the use of a monoclonal anti-IL-2 antibody, JES-6, in attenuating GvHD and preserving GvL activity. Song et al. were also able to provide insight into multiple molecular steps involved for this. Firstly, unlike previous IL-2 inhibitors, JES-6 appeared to have minimal effect on proliferation of Foxp3+ Treg cells, instead of acting primarily through inhibition of the IL-2-Stat-5 pathway, and in turn reducing GM-CSF release from donor CD4+ and CD8+ T cells. Secondly, GvL activity was produced through preservation of PD-1+ Tmp CD4+ and CD8+ cells resulting from low PD-L1 expression in lymphoid tissues. Thirdly, in GvHD target tissues, a high expression of PD-1 led to increased PD-L1 interaction in donor CD4+ and CD8+ T cells, reducing AKT-mTOR signalling, increasing Eomes and Blimp-1 expression, resulting in T-cell exhaustion.