Background/Purpose:

A role for interferon (IFN) in lupus pathogenesis has been inferred from the prominent IFN gene signature (IGS) found in lupus peripheral blood. However, the identification of the major type of IFN up-regulating gene expression in SLE has not been determined. The current experiments were undertaken to differentiate the subtype of IFN driving gene expression perturbations in lupus.

Methods: Raw data from lupus cells and tissues were derived from the GEO repository and collaborators. Determination of differentially expressed (DE) genes was done for each dataset. Gene Set Variation Analysis (GSVA) was used as a non-parametric, unsupervised method for estimating the variation of pre-defined gene sets in gene expression samples. Upstream regulator analysis was employed to identify possible upstream regulators of lupus DE genes. For each potential regulator, an activation z-score was calculated from the experimentally observed information provided for the downstream targets. The first reference dataset used was from Chiche et al, 2014, the second was the DE analysis of the in vitro treatment of PBMC with individual IFNs (Waddell et al, 2010) and the third was the IFNB signature in the whole blood of Multiple Sclerosis patients treated with IFNB (Nickles et al, 2013).

Results:

The IGS was detected in most lupus samples. Comparison of lupus DE genes with previously reported IGS modules or with gene expression data sets obtained by stimulating PBMC with specific IFNs indicated better separation of lupus patient samples from controls using the latter and, additionally, identified patients with TNF and IL12 signatures lacking IFN signatures. Z score calculations to determine the most likely upstream regulator predicted IFNB1 and IFNW as the major IFNs inducing the IGS for SLE skin and kidney. Confirmation of the strong IFNB1 signal was carried out using published microarray data of the IFNB1 signature in Multiple Sclerosis (MS) patients (MS IFNB1). The MS IFNB1 had superior overlap to SLE datasets compared to both IFN-M and the in vitro PBMC IFN Signature. Z score calculations using the MS IFNB1 signature showed high Z scores for all active lupus cells and tissues. The data indicate that IFNB and IFNW are likely IFN family member upstream regulators accounting for the IGS in SLE cells and tissues.

Conclusion:

The type I and type II IGS overlap significantly and are not distinguishable by previously defined interferon modules. In contrast, GSVA using individual cytokine / interferon signatures from in vitro stimulation of PBMC allows interrogation and stratification of individual lupus patient IGS and separates lupus patient from control samples. Z score calculations demonstrate the likely role of IFNB1 and IFNW1 in SLE pathogenesis and suggest that targeting these interferons may have therapeutic value for SLE.

References:

Chiche et al., 2014. Arthritis Rheumatol. 2014 Jun;66(6):1583-95.

Waddell et al., 2010. PLoS One. 5(3): e9753.

Nickles et al., 2013. Hum Mol Genet. 22(20): 4194–4205.

« Back to 2017 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/gene-expression-analysis-demonstrates-that-multiple-type-1-interferons-are-involved-in-lupus-pathogenesis/