The Location of CD4⁺FoxP3⁺ Cells with Regard to CD25 and TNFRSF25 Receptor Signals Matters: Different Requirements for GI and Peripheral Tregs

Regulatory T cells (Tregs) are critical to maintaining immune homeostasis and generating tolerance in the gastrointestinal (GI) tract. GI complications play a prominent role following allogeneic HSCT (aHSCT) and we are interested in manipulating GI Tregs to regulate GVHD. Previously our lab has shown that using a two-pathway strategy of stimulating the TNFRSF25 and CD25 receptors with a TL1A-Ig fusion protein (FP) and IL-2_{low dose}, Treg frequency and numbers in the lymphohematopoietic (LHC) compartment can be markedly increased (Wolf, BBMT 2017; Copsel, BBMT 2018). As a consequence of microbes and food antigens, Treg populations in the GI have a relative highly diverse composition compared to the lymphohematopoietic Treg compartment. This includes, but is not limited to, the presence of a stable and suppressive FoxP3⁺RORyt⁺ Treg population. Additionally, the GI tract contains various populations of innate lymphoid cells (ILCs) that interact with Tregs. ILCs express CD25 as well as TNFRSF25, and respond individually to IL-2 and TL1A administration respectively (Danny, JCI 2017; Verneris, Blood 2013). Based on the extensive differences between the lymphohematopoietic and GI Treg compartments, we evaluated how our two-pathway strategy may be differentially affecting the levels and activation status of Tregs in the GI tract.

To address this question, we initially utilized B6-Nur77^GFP mice, where GFP expression occurs with activation of the Nur77 promoter. However, since Nur77 is only produced following TcR engagement and not inflammatory signals, the strength of TcR stimulation in all T cell populations can be monitored by GFP levels. We first examined Tregs from 2-pathway treated B6-Nur77^GFP mice and subsequently generated B6-Nur77^GFPFoxP3^RFP mice to readily assess the TcR signaling status of Tregs in different compartments. Mice were systemically administered FP, IL-2, or FP + IL-2 over a 1 wk period.

The frequency of Tregs (FoxP3⁺CD4⁺) / Tcon (CD4⁺FoxP3^-) in the lamina propria (LP) of the large intestine (LI) reached levels >60%. (1A). This elevation of Treg / Tcon frequency included FoxP3⁺RORyt^- Tregs as well as FoxP3⁺RORyt⁺ double positive Treg populations. In contrast to our previous findings in the LHC, treatment with TL1A-Ig FP alone elevated levels to the same extent as the combination of FP + IL-2 (1A). Importantly, IL-2 treatment alone - as reported in the LHC - again had only a modest effect on elevating the frequency of Tregs in the LI/LP (1A). These observations suggest that the activation status of Tregs may differ based on the compartmental location. To asses activation status, we evaluated 2-pathway treated B6-Nur77^GFP mice. Tregs in the LN/spleen had elevated frequency of GFP⁺ Tregs and higher GFP MFI than untreated mice. This elevated TcR stimulation was present in peripheral Tregs - but not CD8 - T cells (1B). Without exogenous stimulation, Tregs exhibited higher baseline TcR activation levels vs. Tcon cells (1B). The frequency of GFP⁺ Tregs and the GFP MFI was clearly highest (>45%) in the SI and LI LP vs. LN/spleen (<35%) (1C). In an independent experiment, we verified these findings and learned that only conjunctival Tregs demonstrated similarly increased frequency of GFP⁺ Tregs with elevated GFP MFI (1D). We also found that TL1A-Ig + IL-2 treatment in vivo increases the relative frequency of both innate ILC2 and ILC3 lymphoid populations (2A). The combination also increased the frequency of ILC2 cells greater than either reagent alone in the GI tract (2B).

In total, our results objectively validate selective Treg vs. Tcon targeting via TNFRSF25. These findings also demonstrate that basal Treg activation status differs depending on the compartment. Notably, Tregs in mucosal vs. LHC tissue expressed higher TcR activation levels. Such Tregs have the potential for co-stimulation via TNFRSF25. Moreover, since IL-2 is required to maintain Tregs under both basal and activated conditions, our findings suggest a local source of IL-2 is present to maintain GI Tregs. ILC3 cells are recently reported to generate IL-2 (Zhou L, Nature 2019). We hypothesize that TNFRSF25 stimulation in ILC3 cells results in a local source of IL-2 which can account for our finding that TL1A-Ig alone vs. TL1A-Ig + IL-2 stimulation results in equivalent levels of GI Tregs. Ongoing experiments are examining how GI GVHD in aHSCT recipients are affected by TNFRSF25 +/- CD25 Treg/ILC stimulation.

View largeDownload slide

View largeDownload slide

Close modal