Objective. We attempted to elucidate possible pathogenetic mechanisms in scleroderma by analysis of gene expression patterns of purified monocytes and lymphocytes, as well as protein profiles of cytokines and growth factors.Methods. Expression analysis was performed on messenger RNA (mRNA) from cells that had been purified with magnetic beads. Plasma samples from the same patients were used for multiplex cytokine analysis. Potential sources of proteins were also examined by in situ hybridization of skin specimens.Results. A total of 1,800 genes from monocytes and 863 genes from CD4؉ T cells were differentially expressed in scleroderma patients. As observed by other investigators using unfractionated peripheral blood cells from patients with autoimmune connective tissue diseases, the cell type-specific analyses of our scleroderma samples showed expression of genes suggesting the presence of interferon-␣ (IFN␣), despite the apparent absence of this cytokine in plasma. IFN␣ RNA was, however, expressed at enhanced levels in vascular and perivascular cells in scleroderma skin samples. While levels of interleukin-1␣ (IL-1␣) and IL-16 were among 10 proteins found to be significantly elevated in scleroderma patients, none of the large panel of plasma cytokines we analyzed correlated with the expression levels of putative IFN response genes.Conclusion. The pattern of up-regulation of mRNA in both the monocytes and CD4 lymphocytes of scleroderma patients, together with the detection of IFN␣ RNA in the microvasculature, suggests that leukocytes respond to this cytokine locally in the vessels. Detection of high levels of IL-1␣ and IL-16 in plasma and the independence of these protein levels from the IFN signature, implicates an independent contribution of other cytokines to immune activation and/or inflammation in scleroderma.

Material

Inflammatory markers have consistently been associated with vascular disease. Evidence of genetic polymorphisms in inflammatory loci that predict severe carotid artery disease (CAAD) would suggest that this relationship is not secondary to other correlated factors, but related to inflammation itself. We examined the full common genetic variation in 42 inflammatory loci for prediction of severe CAAD versus ultrasound proven controls using a tagSNP approach. For selected loci, monocyte RNA levels were contrasted in subjects with and without CAAD. We confirm the association of IL6(-174), FGB (-455), and ALOX5 with CAAD and show that multiple ALOX5 SNPs independently predict CAAD. We provide evidence for previously unreported associations of SNPs in IL4R, NFKBIA, and PLG with CAAD, and weaker evidence for associations with CSF3, IL10RA, and VCAM1. The NFKBIA and IL10RA expression levels significantly differed between subjects with CAAD and controls. These results support a role for genetic variation related to inflammation in CAAD and a causal role for specific gene products.