Disease activity drives divergent epigenetic and transcriptomic reprogramming of monocyte subpopulations in systemic lupus erythematosus

Objectives Systemic lupus erythematosus (SLE) is characterised by systemic inflammation involving various immune cell types. Monocytes, pivotal in promoting and regulating inflammation in SLE, differentiate from classic monocytes into intermediate and non-classic monocytes, assuming diverse roles and changing their proportions in inflammation. In this study, we investigated the epigenetic and transcriptomic profiles of these and novel monocyte subsets in SLE in relation to activity and progression. Methods We obtained the DNA methylomes and transcriptomes of classic, intermediate, non-classic monocytes in patients with SLE (at first and follow-up visits) and healthy donors. We integrated these data with single-cell transcriptomics of SLE and healthy donors and interrogated their relationships with activity and progression. Results In addition to shared DNA methylation and transcriptomic alterations associated with a strong interferon signature, we identified monocyte subset-specific alterations, especially in DNA methylation, which reflect an impact of SLE on monocyte differentiation. SLE classic monocytes exhibited a proinflammatory profile and were primed for macrophage differentiation. SLE non-classic monocytes displayed a T cell differentiation-related phenotype, with Th17-regulating features. Changes in monocyte proportions, DNA methylation and expression occurred in relation to disease activity and involved the STAT pathway. Integration of bulk with single-cell RNA sequencing datasets revealed disease activity-dependent expansion of SLE-specific monocyte subsets, further supported the interferon signature for classic monocytes, and associated intermediate and non-classic populations with exacerbated complement activation. Conclusions Disease activity in SLE drives a subversion of the epigenome and transcriptome programme in monocyte differentiation, impacting the function of different subsets and allowing to generate predictive methods for activity and progression.