Background/Purpose: In a given patient with rheumatoid arthritis (RA), it is difficult to predict disease progression or identify to which treatments they will respond. Macrophages have been strongly implicated in the pathogenesis of RA and reduction in the number of macrophages in the synovial sublining of the joint is a key biomarker for better outcomes. However, without a deeper knowledge of the genes regulated in vivo, such as provided by unbiased high-throughput sequencing, it is difficult to discern the function of these macrophages in RA. Here, we describe our work to profile the genome-wide transcriptome of synovial macrophages in RA patients.

Methods: We used tissue obtained from synovial biopsies of the wrist joints in RA patients for histology and cell sorting via FACS. We gated for CD45⁺Lin^–CD64⁺CD11b⁺CD14⁺HLA-DR⁺ macrophages and distinguished 2 populations based on the expression of CD206. As a comparison, we also collected and sorted macrophages from OA patients via surgical discards. We processed all macrophage populations for RNA-seq and sequenced the libraries on an Illumina NextSeq 500 to an average depth of 5 million reads. The reads were aligned and mapped to genes using bowtie and HTseq, respectively, followed by bioinformatic analysis as described below.

Results: The quality of the RNA-seq data from RA macrophages was comparable to OA, confirming the feasibility of our approach. Next, we assessed the ability of our macrophage-specific RNA-seq to identify differentially expressed genes between RA and OA by fold-change, as compared with RNA-seq of the unprocessed whole synovial tissue. While the approaches generally agreed, we found many genes that were differentially expressed only in macrophages, supporting the value of cell-specific RNA-seq to uncover gene pathways that would be obscured in whole tissue RNA-seq. Next, we considered the heterogeneity of macrophages across RA patients by calculating the pairwise correlation between samples. We found that there were significant differences between RA transcriptional profiles that were best explained by classification of patients into lymphoid, myeloid, or pauci-immune phenotypes based on the dominant cell type. In order to overcome this difficulty in identifying genes associated with RA pathology, we focused on clustering genes into co-regulated modules based on their expression across samples. This allowed us to identify pathways with significant expression patterns, even if the relevant genes were up-regulated in only a portion of RA samples. For example, we found that a subset of RA patients demonstrated over-expression of certain cytokine-mediated pathways (including IL3, IL5, and IL13), while another subset was enriched for type 1 interferon signaling. Similarly, we found that Dock2, a gene involved in actin remodeling that was not significant at the whole-tissue level, was preferentially expressed in RA macrophages that also expressed Ccr1.

Conclusion: Together, these results help us to understand the role of different macrophage populations in RA heterogeneity. Our goal is to use these studies as the basis for predicting clinical outcomes and choosing between therapeutic options for treatment.

« Back to ACR/ARHP Annual Meeting 2017

ACR Meeting Abstracts - https://acrabstracts.org/abstract/transcriptional-profiling-of-synovial-macrophages-from-ra-patients-to-capture-disease-heterogeneity/