BACKGROUND: Regulatory T cells (Tregs) are non-redundant mediators of immunity and tolerance. Transfer of expanded Tregs is a promising therapy for autoimmune disease, organ rejection and graft-versus-host disease (GVHD) following allogeneic hematopoietic stem cell transplantation (aHSCT) (Marek-Trzonkowska N, Diabetes Care 2012; Tang Q, J Mol Cell Biol 2012; Hanash AM, Blood 2005). Cell purity, functional stability and practical issues, including economics, pose challenges to employ ex-vivo Treg expansion. We recently developed a novel approach which markedly expands (>50% of CD4+ T cells) and selectively activates Tregs in vivo by targeting TNFRSF25 (with TL1A-Ig fusion protein) and CD25 (with low dose IL-2) (Wolf D, BBMT 2017). Based on these findings, we asked if there is a phenotypic/functional difference of these two-pathway in vivo expanded Tregs compared with unexpanded Tregs.

METHODS: Mice were administered TL1A-Ig+IL-2 over 6 days. Splenic and lymph node (LN) Treg phenotype was determined by flow cytometry. Sorted Treg functionality was assessed using an in vivo MHC-mismatched aHSCT.

RESULTS: Following TL1A-Ig + low dose IL-2 treatment, there was a strong increase in the overall levels of CD4+FoxP3+ within the CD4+ T cell compartment (>30%). We observed significantly fewer naïve Tregs and increased central memory (CD62-LhiCD44+) and effector/memory (CD62-LloCD44+) Tregs in the spleen (Figure 1) and LN (data not shown) from these mice. Furthermore, expanded Tregs exhibited a significant decrease in Ly6C expression - consistent with greater suppressive function. Notably, the expanded Tregs exhibited higher levels of PD-1 whereas CTLA4 and Tim-3, other immune regulatory molecules, remained unchanged. Additionally, increase in CD103 (activation marker) and Nrp1 expression was observed in the expanded vs unexpanded Tregs. Notably, no differences in Annexin V were observed.

To determine if TL1A-Ig+IL-2 expanded Tregs possess a more potent regulatory activity compared to control i.e. normal unexpanded populations, an MHC-mismatched aHSCT (donor/recipient = C57BL/6-BALB/c) was performed. We employed varying numbers of highly purified, sorted Tregs (>96% purity by CD4+FoxP3+ selection from C57BL/6-FoxP3RFP reporter mice) combined with T conventional cells. As anticipated, using only 3.5x105 control Tregs together with 1.0 x106 T conventional lymphocytes was not enough to prevent GVHD. However, transfer of as few as 1.75-3.5x105 expanded Tregs ameliorated acute GVHD (Figure 2) (*p<0.05 **p<0.01 3.5x105 unexpanded vs expanded Tregs). Lower numbers of expanded Tregs (0.05 to 1.0 x105) were not sufficient to diminish GVHD.

CONCLUSION: Our TL1A-Ig+IL-2 Treg expansion protocol promotes more activated and effector Treg subpopulations. The primary subsets increased were central and effector memory and these are characterized by elevated levels of CD103 and PD-1. Consistent with this phenotype, the expanded Tregs exhibited a markedly greater suppressive activity as shown by amelioration of GVHD following a pre-clinical MHC-mismatched aHSCT. Overall, this approach could provide a new strategy to ameliorate clinical GVHD and could potentially be considered to treat autoimmunity and prevent organ rejection.

Disclosures

Levy: OccuRx: Research Funding; Shire: Research Funding; Allergan: Consultancy; Capricor Therapeutics: Consultancy.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution