Unveiling the molecular and immune heterogeneity across distinct disease activity states of systemic lupus erythematosus through integrative bulk and single-cell transcriptomics.

Abstract

Objective: Systemic lupus erythematosus (SLE) is a complex autoimmune disease characterised by substantial variability in clinical presentations and disease severity. This study aims to investigate the molecular and cellular mechanisms underlying distinct SLE Disease Activity Index (SLEDAI)-defined disease states of SLE and to identify potential prognostic biomarkers and therapeutic targets associated with heightened inflammatory activity.

Methods: We analysed bulk and single-cell RNA sequencing datasets from SLE patients. Patients were stratified into clinically distinct SLEDAI-defined disease states using established SLEDAI thresholds, which served as operational surrogates of inflammatory disease burden. Differentially expressed genes and gene co-expression networks were analysed to identify severity-associated hub genes. The immune microenvironment and intercellular communication were evaluated using CIBERSORTx and CellChat algorithms, respectively.

Results: Differential expression and network analyses identified five key hub genes (KLRK1, IL2RB, RAG1, SH2D1B and NCR1) strongly correlated with SLE disease activity states, which were used to develop a five-gene SLE severity (SLEsev) score. The SLEsev score was significantly higher in high-activity/severe SLE (sSLE) compared with low-activity/mild SLE (mSLE) (p<0.001), which showed a consistent trend in an independent validation cohort. Immune profiling revealed that mSLE was enriched in regulatory T cells (Tregs) and memory B cells, whereas sSLE exhibited elevated levels of CD8+T cells, natural killer cells and monocytes (p<0.05). Single-cell analysis demonstrated that sSLE possessed a more complex cellular communication network (311 vs 283 inferred interactions) and greater overall interaction strength than mSLE, with prominent pro-inflammatory signalling pathways such as IFN-II, GAS and CCL playing a key role in sSLE.

Conclusions: This study unveils the profound immunological and transcriptomic dysregulation underlying SLE severity. The identified five-gene SLEsev score serves as a potential biomarker for disease stratification, while the hyperactive intercellular signalling networks highlight specific molecular targets for modulating immune responses in severe SLE.