Regulatory T Cell Amelioration of Graft-versus-Host Disease following Allogeneic/Xenogeneic Hematopoietic Stem Cell Transplantation Using Mobilized Mouse and Human Peripheral Blood Donors

Abstract

The present studies examined experimental transplant outcomes using mobilized peripheral blood from mice and humans together with FoxP3+Treg cells. Donor mice were treated with filgrastim and / or plerixafor and their peripheral blood (PB) displayed significant elevations in hematopoietic stem and progenitor populations. Some of these PB donors were concurrently administered a Treg expansion strategy consisting of a TL1A-Ig fusion protein low dose rIL-2. A significant increase (4-5x) in the frequency Tregs occurred during mobilization. C3H.SW PB was collected from mobilized and Treg unexpanded ("TrUM") or mobilized and Treg expanded ("TrEM") donors and transplanted into MHC-matched B6 (H2b) recipients. Recipients of TrEM, exhibited significantly reduced weight loss and clinical GVHD scores compared to recipients of TrUM. Notably, recipients of TrEM exhibited comparable GVL activity to TrUM recipients against leukemia levels. Next, huTregs (CD4+CD25+CD127lo) from a healthy human PB mobilized donor were expanded ex-vivo prior to transplant into NSG/ NOD-scid IL2Rgammanull mice. We found that treatment with ex-vivo expanded huTregs resulted in significant reduction of lethality and clinical xGVHD scores. Notably, post-transplant, PB huTregs levels remained elevated and the frequency of huCD4+Tconv and CD8+ cells was diminished supporting the improved xGVHD outcomes. These findings demonstrated that the use of mPB containing elevated Treg levels significantly reduced GVHD following "MUD" and MHC-mismatched mouse HSCT without loss of GVL activity. Moreover, utilizing ex-vivo expanded huTregs from a mobilized PB donor and added back to donor PB ameliorated xGVHD. In total, these studies support the notion that in vivo or ex-vivo manipulation of donor Tregs together with mobilized peripheral blood could provide therapeutic approaches to improve aHSCT outcomes.

Keywords: Allogeneic Hematopoietic Stem Cell Transplantation; GVT, Xenotransplantation; Graft-versus-Host Disease; Mobilization; TL1A; Tregs, IL-2.

Fig. 1. Tregs can be expanded concomitantly with HSC and progenitor cell increases in peripheral blood following mobilization with Filgrastim and plerixafor infusion.

(A-E) B6-Fir,H2^b mice were administered rGCSF (2.5ugs / injection) on Days 1–4 daily and plerixafor (5.0mgs/kg) on Days 4 and 5. (Fig. S1). Peripheral blood was collected in the morning of Day 5, 1 hr. following plerixafor injection and PBMC isolated (Methods). Following staining with selected mAbs and analysis via flow cytometry, (A-E) mobilization treatment resulted in 24-, 29-, and 60-fold increases of HSC (A) MEP (B) and LSK (C) populations respectively. CLP (3 fold) (D) and GMP (34 fold) (E) were also elevated compared to control (peripheral blood from non-injected normal mice). Data represents results of pooled peripheral blood from 2 mobilized B6-Fir male mice. (F-I) Mice (LP/J, H2)^b were mobilized as above, and administered TL1A-Ig fusion protein and rhIL-2LD. Increased levels of c-kit⁺ and WBC fractions (F-H populations calculated within the non-lymphoid fraction) together with elevated Treg frequency following mobilization and treatment with TL1a-Ig and rhIL-2LD. Data represents pooled peripheral blood from 5 individual mice in each group (n=5 mice / group). (I) FoxP3⁺ Tregs within the CD4⁺ population. Numbers of Tregs were calculated for each group indicated (1.0ml peripheral was collected from each mouse, PBMC collected and pooled (n=5/group). Total PBMC were counted and Treg numbers calculated based on the frequency of CD4⁺FoxP3⁺/CD4⁺ cells. Untreated: 7,502; Mobilized: 102,538; Mobilized plus Treg expansion: 914,514. (J-M) Targeting TNFRS25 with a second agonist (mAb 4C12) also expands Tregs in mobilized peripheral blood. B10.D2 (H2d) mice were mobilized and Treg expanded with either TL1A-Ig (50ug) or mAb 4C12 (100ug) plus rhIL-2LD. (J-L) Heightened levels of c-kit⁺, myeloid cells and monocytes were detected in mobilized animals receiving eitherTNFRSF25 agonistic reagent. (M) Levels of Tregs were increased in all mobilized B10.D2 animals treated with either agonist (n=2 for mAb 4C12+IL-2LD), n=2 for TL1A-Ig+IL-2LD) vs non-mobilized B10.D2 animals (n=4): Cells per 200ul of peripheral blood = Non-mobilized, 818–984; Mobilized via 4C12, 1640, 2476; mobilized via TL1A-Ig, 1485,3320. (J-M) Data were collected from individual mice and are expressed as mean ± SD and were analyzed by one-way ANOVA with Bonferroni correction for multiple comparisons. *P < .05; **P < .01; ***P < .001; ****P < .0001.

Fig 2.. Tregs Expanded with TL1a-Ig and IL-2(LD) in mobilized donor peripheral blood (PB) exhibited suppressive activity and ameliorated GVHD.

Mice were injected i.p. with TL1A-Ig (50ug) (days 1–4) and IL-2 LD (days 4 to 6). Mice administered TL1A-Ig+IL-2_LD showed an increase in overall Treg (CD4⁺ FoxP3⁺) frequency in mobilized (rGCSF + Plerixafor) plus Treg expanded PB versus mobilized or untreated mice (A,F). (B) Diminished frequencies of central Tregs and significantly elevated levels of effector Tregs were present in mobilized plus Treg expansion compared to mobilized or untreated mice. (C) One hour after the last IL-2 injection (combined TL1A-Ig+IL-2_LD), pSTAT5 staining showed heightened activation of PB Tregs compared to Tcon CD4 T cells. (D) Decreased T cell proliferation in mobilized peripheral blood in animals concomitantly Treg expanded. PBMC were activated with anti-CD3 mab and assessed for proliferation after 72 and 96 h (E) GVHD was diminished in animals receiving mobilized and Treg expanded donor cells. MHC-mismatched HSCT (B6→BALB/c) using T cells from donor B6-Fir mice PB either mobilized or mobilized plus Treg expanded (TL1A-Ig+IL-2LD) adjusted to contain 1.0×10⁶ total T cells. Clinical scoring is presented. (F-H) GVHD was diminished in animals after an MHC-matched minor histocompatibility antigen mismatched HSCT (LP/J -> B6) using T cells from donor PB mobilized (rGCSF + Plerixafor) plus Treg expanded (TL1A-Ig fusion protein + IL-2_LD) versus mobilized only. (G) and mPTX mab+IL-2LD (H) adjusted to contain1.0 ×10⁶ total T cells. (F) Treg and Ly6G⁺ granulocyte levels in PB donors for the transplant results in panel (G) are shown. Values are means ± SEM and were analyzed by multiple variable analysis using ANOVA. A P-value < 0.05 was considered significant. (E,G) a 2-tailed unpaired t-test. was used for comparisons between 2 experimental groups (H) *P < .05; **P < .01; ***P < .001.Significance indicated by * p < 0.05, ** p < 0.01, *** p < 0.001, ns=not significant.

Fig 3.. Maintaining mouse GVL and suppressing xenogeneic GVHD : Transplants with mobilized mouse and mobilized human peripheral blood.

(A-C) An MHC-matched transplant C3H.SW⟶B6 HSCT was performed. Mice received PB T cells from either mobilized only (rGCSF) or mobilized (rGCSF) plus expanded Treg donor C3H.SW mice. MLL-AF9 B6 tumor cells were administered to all mice at the time of transplant. (A) GVHD was reduced in recipients of MLL-AF9+mobilized + Treg expanded donors compared to MLL-AF9 + mobilized only recipients. (B) Representative flow contour plots of spleen and bone marrow cells 28–30 days post-HSCT from individual recipients of BM only (syngeneic, C3H.SW), mobilization only (allogeneic, B6), and mobilization (allogeneic, B6) + Treg expansion (TL1A-Ig + IL-2 LD). (C) GVL is maintained in animals with reduced GVHD. Allo-HSCT recipient groups were examined for MLL-AF9 presence post-HSCT. Tumor cell frequency was always greater in the recipients of syngeneic mobilized PB donors compared to levels in recipients who received allogeneic mobilized PB without or with expanded Tregs. (D-H) Addition of ex-vivo expanded human Tregs to mobilized human peripheral blood suppresses xeno GVHD. Sorted Tregs isolated from huPB were cultured and expanded for 7 days using aCD3/aCD28 beads (Fig. S3C). (D) Treg numbers and FoxP3 expression at Day 7 of culture are shown prior to use in transplant. NSG mice were irradiated on day −1 and injected with mobilized huPBMCs with or without these expanded huTregs on day 0 (n=8 mice/group). (E,F) Mice treated with huTregs exhibited significantly less lethality and better clinical GVHD scores. (G) Mice receiving huTregs with huPB showed persistence of elevated Treg levels in PB, less huCD8⁺ and conventional CD4⁺ T cell levels in the blood 13 days post-transplant compared to recipients of huPB without Tregs. (E) Representative histogram and graph of individual mice illustrating huCD4⁺ Tconv proliferation in blood 13 days post-transplant. *** p<0.001; **** p<0.0001.