Background/Purpose:

Localized scleroderma (LS) has both inflammatory and fibrotic components affecting skin and underlying tissue.  Extent and duration of inflammation during active LS is thought to be the major contributor to long-term disease damage and disability.  Identifying the cellular phenotype expressed during active LS would be instrumental to improve outcomes. As a preliminary evaluation, pediatric LS and healthy pediatric control samples were analyzed by mass cytometry by time-of-flight spectrometry (CyToF). CyToF uses heavy metal ion tags to mark cells and perform high-dimensional phenotypic and functional analysis on single cells. Our panels focused on monocytes and T cells, respectively, to determine key phenotypic populations in LS samples.

Methods:

Paired pediatric LS PBMC samples (n=9 subjects, 18 samples) from individuals with initial active and later inactive disease states and healthy pediatric controls (n=8) were collected (IRB #PRO11060222) and analyzed using CyToF. Both monocyte and T cell panels of 34 markers each were performed at the Stanford Human Immune Monitoring Center (HIMC) using singlet and viability based analysis. Data was analyzed using SPADE and CITRUS (Cytobank) and Cytosplore software.  CITRUS clusters were made from all samples, and cell types were then identified from all levels of the cluster hierarchy that were significantly associated with active LS, inactive LS, and healthy groups. SPADE trees were built using 11 marker calculated cell clusters to define key populations that differed from the subject’s active PBMC sample acting as the baseline. Cytosplore software then allowed for sub-analysis of SPADE branches through t-Distributed Stochastic Neighbor Embedding (t-SNE) plots.

Results:

CyTOF revealed a dramatic decrease in frequency of monocyte subsets, DC cells (both CD16+/-), and NK cells (CD56+CD16+) in the inactive state.  The CD16+CD86+ monocyte subset (M1 phenotype) showed a 3-fold decrease in the inactive state, consistent with our prior Luminex™ findings of the peripheral LS blood signature.  Further study of NK subsets shows NK^CD56bright cells having a 4-fold increase in the active state and 50% increase in IFN-γ production after lipopolysaccharide (LPS) stimulation.Further CyToF analysis showed an overall increase in T cell (CD3+) populations in active LS, with decreased T_H1 (CD3+CD4+IFNγ) and Tc1 (CD3+CD8+IFNγ+) populations when transitioning from active to inactive disease state. Granzyme positive cytotoxic T cells (CD8+) were elevated in active LS and when comparing overall LS to healthy controls.

Conclusion:  

These findings support increased levels of type-1 specific M1 and NK cells during active LS. These cell types are thought to contribute to activation and stimulation of T cells and DCs through IFNγ expression, which would support the elevation of T_H1-like (IFNγ+ T_H and Tc) cells and DCs observed in our cohort. Previous work showed that the IFNγ inducible chemokines (CXCL9, CXCL10 and CXCL11) are present during active LS, possibly further polarizing these phenotypes.  Further study is ongoing to evaluate these phenotypes in active LS skin, and the association to the PBMC signature.