Abstract

Background:

Peripheral blood (PB) regulatory T cells (Tregs) in post-transplant alloimmune hepatitis (DAIH) & autoimmune hepatitis (AIH) have poor regulatory function, however little is known about intrahepatic (IH) Tregs in both diseases. As the human Treg compartment encompasses multiple subsets that delineate different developmental stages & their associated regulatory function, we used mass cytometry & unsupervised clustering algorithm Flow self-organizing map (FlowSOM) to interrogate the distribution of subsets within ex-vivo CD25^hiCD127^lo/^negFoxP3⁺ Treg in PB & liver (IH) of children with DAIH & non-transplanted children with AIH.

Methods:

Enriched CD4⁺ T cells from peripheral blood mononuclear cells (PBMC) and intrahepatic lymphocytes (IHL) of children with DAIH (n=5), AIH (n=5), biopsy proven acute rejection (AR) (n=3), liver transplanted children with graft dysfunction (n=6), & liver transplanted children with normal graft function (LTC) (n=14) were expanded in culture with Dyna beads CD3/CD28, rIL-2 & TGF-b for 5-days prior to FACS sorting of Tregs. For FACS sorting, Tregs were identified by high expression of CD25 & low expression of CD127. Ex-vivo sorted PB & IH Tregs were then stained for mass cytometry with metal-conjugated monoclonal antibodies (CD49D, CD4, CCR4, CD45RA, CD3, CD39, FoxP3, CD95, CD45RO, CD25, CD152, HLA-DR, CD127, CD73, NRP1, Helios, LAP, CD45). Cells were acquired using a CyTOF Helios mass cytometer. CyTOF data was time-based bead normalized using standard procedures. The data were then uploaded to OMIQ (Omiq.ai), manually gated using Gaussian parameters to resolve live, intact, single cells for further analysis. Expression levels were arcsinh transformed with a cofactor of 5 and compared across disease groups within the Treg population. Clustering was performed using FlowSOM & visualized using dimensionality reduction using UMAP. All statistical analyses were performed in GraphPad Prism v9 (Dotmatics).

Results:

Demographics reported in Table 1 below.

10 phenotypically distinct clusters were identified based on15 analyzed parameters within PB & IH Tregs of DAIH, AIH patients, as well as LTC subjects. IH Tregs of AIH patients were characterized by a higher CD45RA, & lower CCR4, Helios, FoxP3, CD25, CD73, CD39, CD45RO, & CD95 expression when compared with PB Tregs (p<0.001). IH Tregs of DAIH patients were characterized by a higher expression of CD45RA, CCR4, CD25, CD73, CD95, & lower expression of FoxP3, Helios, CD39 compared to PB Tregs (p<0.001). An important suppressive mechanism mediated by Treg involves the CD39/CD73 adenosine pathway. In humans, CD39⁺ Tregs are implicated in the suppression of Th17 responses & the control of autoimmunity.

Conclusion:

The lower expression of FoxP3 & Helios in IH Tregs of patients with DAIH & AIH could suggest Treg destabilization. Further work is needed to determine if these IH Tregs have the potential to transdifferentiate into effector T cells.