Three Distinct Transcriptional Profiles of Monocytes Correlate with Disease Activity in SSc Patients

Abstract

Background/PurposePatients with systemic sclerosis (SSc) display a complex clinical phenotype. There are numerous studies that relate transcriptional signatures from PBMC or whole skin of SSc patients to disease activity. However, analyses of whole tissue RNA-sequencing studies are subjected to changes in cellular composition that can drive gene expression signatures and a loss of the ability to detect biologically important transcriptional changes within minority cell populations. Here, we focused on circulating monocytes, which have been shown to exist as two central populations classical (CM) and non-classical (NCM). MethodsSSc patients were recruited from four different sites that form PRESS: Northwestern University, University of Texas, University of Michigan and University of Utah. Comprehensive clinical data was collected for all patients. We isolated CM and NCM from these patients and age, sex, and race-matched healthy volunteers were used as controls. RNA-seq was performed on CM and NCM populations as well as on isolated bulk macrophages from skin. ResultsWe first performed RNA-seq on CM, which are the predominant population in circulation. In order to capture the variability across the SSc cohort, we defined 1790 differentially expressed genes in each patient. We then used these genes to cluster patients into 3 subgroups: Groups A-C. Group A exhibited the strongest interferon signature and innate immune pathways. Group B patients expressed genes in the same pathways but was also enriched for response to cAMP and corticosteroids. Both Group B and Group C exhibited upregulation of genes associated with vasculature development and blood vessel formation. Group C uniquely upregulated TGFB pathways. Next, we performed RNA-seq on NCM isolated from the same patients. When NCM were clustered based on the same 1790 genes as CM, we found that Groups A and C were recapitulated, while Group B was less cohesive. Our analysis stratified SSc patients based on their transcriptional profiles in monocytes but was agnostic to their clinical presentation. We found that Group B and C patients exhibited significantly worsened lung function at the time of monocyte isolation than Group A patients. However, there were no significant differences in skin disease. We then isolated macrophages from skin biopsies of SSc patients and showed that the transcriptional profile of Group A and C in SSc patients was conserved. We also used gene expression data from another study on monocytes which stratified patients based on disease presentation. We found that Group A accurately distinguished dcSSc and ncSSc patients from controls, but not lcSSc. ConclusionWe are the first to show that transcriptomic analysis of classical and non-classical circulating monocytes can unbiasedly stratify SSc patients and correlate with disease activity outcome measures.