AUTHOR=Nicholls Jemma , Cao Benjamin , Le Texier Laetitia , Xiong Laura Yan , Hunter Christopher R. , Llanes Genesis , Aguliar Ethan G. , Schroder Wayne A. , Phipps Simon , Lynch Jason P. , Cao Huimin , Heazlewood Shen Y. , Williams Brenda , Clouston Andrew D. , Nefzger Christian M. , Polo Jose M. , Nilsson Susan K. , Blazar Bruce R. , MacDonald Kelli P. A. TITLE=Bone Marrow Regulatory T Cells Are a Unique Population, Supported by Niche-Specific Cytokines and Plasmacytoid Dendritic Cells, and Required for Chronic Graft-Versus-Host Disease Control JOURNAL=Frontiers in Cell and Developmental Biology VOLUME=9 YEAR=2021 URL=https://www.frontiersin.org/journals/cell-and-developmental-biology/articles/10.3389/fcell.2021.737880 DOI=10.3389/fcell.2021.737880 ISSN=2296-634X ABSTRACT=

Regulatory T cell (Treg) reconstitution is essential for reestablishing tolerance and maintaining homeostasis following stem-cell transplantation. We previously reported that bone marrow (BM) is highly enriched in autophagy-dependent Treg and autophagy disruption leads to a significant Treg loss, particularly BM-Treg. To correct the known Treg deficiency observed in chronic graft-versus-host disease (cGVHD) patients, low dose IL-2 infusion has been administered, substantially increasing peripheral Treg (pTreg) numbers. However, as clinical responses were only seen in ∼50% of patients, we postulated that pTreg augmentation was more robust than for BM-Treg. We show that BM-Treg and pTreg have distinct characteristics, indicated by differential transcriptome expression for chemokine receptors, transcription factors, cell cycle control of replication and genes linked to Treg function. Further, BM-Treg were more quiescent, expressed lower FoxP3, were highly enriched for co-inhibitory markers and more profoundly depleted than splenic Treg in cGVHD mice. In vivo our data are consistent with the BM and not splenic microenvironment is, at least in part, driving this BM-Treg signature, as adoptively transferred splenic Treg that entered the BM niche acquired a BM-Treg phenotype. Analyses identified upregulated expression of IL-9R, IL-33R, and IL-7R in BM-Treg. Administration of the T cell produced cytokine IL-2 was required by splenic Treg expansion but had no impact on BM-Treg, whereas the converse was true for IL-9 administration. Plasmacytoid dendritic cells (pDCs) within the BM also may contribute to BM-Treg maintenance. Using pDC-specific BDCA2-DTR mice in which diptheria toxin administration results in global pDC depletion, we demonstrate that pDC depletion hampers BM, but not splenic, Treg homeostasis. Together, these data provide evidence that BM-Treg and splenic Treg are phenotypically and functionally distinct and influenced by niche-specific mediators that selectively support their respective Treg populations. The unique properties of BM-Treg should be considered for new therapies to reconstitute Treg and reestablish tolerance following SCT.