Background/Purpose:

Skin global gene expression profiling indicates presence of distinct gene expression signatures in patients with systemic sclerosis (SSc). We examine the correlation of the skin gene expression profiles with clinical subsets of disease in a large sample of SSc patients.

Methods:

In this study, 61 SSc patients (61% diffuse disease, mean disease duration: 7.7 years) and 36 gender, ethnicity, and age matched controls were examined.  All skin biopsies were obtained from subjects’ arm and processed according to the same procedures.  In 40 SSc patients, the skin was affected at the biopsy site.  Global gene expression profiling was performed by Illumina HumanHT-12 arrays in one batch. Only 1 biopsy per subject was analyzed. Differentially expressed (DE) genes were detected in multivariable analysis with False Discovery Rate of 10% at the confidence level of 80%. Involved Canonical Pathways were identified by Ingenuity Pathway Analysis.  Unsupervised hierarchical clustering was performed with all expressed genes and after filtering the gene list with the previously published TGF-β, IL-13, and diffuse proliferative (Milano et al. 2008) signatures.  We also generated an interferon-α (IFN- α) inducible gene signature using control fibroblast treated with IFNa.

Results:

 The transcriptome of the majority of SSc patients differed from controls. Specifically, 2525 DE genes were identified, belonging to fibrotic, leukocyte extravasation, and oxidative phosphorylation pathways. However, there was significant heterogeneity among the SSc samples. We examined whether the observed heterogeneity correlated with known clinical subtypes of SSc. However, we could not identify any significant differential expression when patients were subgrouped by disease type (limited versus diffuse), antibody subtypes, treatment status (immunosuppression versus no immunosuppression), or disease duration. The presence of anti-RNP antibodies was the exception.  Comparison of anti-RNP positive to the anti-RNP negative patients resulted in 188 DE genes, belonging mainly to IFN pathways.  However, we detected prominent gene expression differences among SSc patients who have affected vs. unaffected skin at the biopsy site. This comparison resulted in 378 DE expressed genes, belonging to fibrotic and leukocyte extravasation pathways. This was further underscored by the fact that comparison of SSc affected skin to controls resulted in 3458 genes while comparing patients wit unaffected skin to controls resulted in only 38 DE genes. We next filtered our gene list by TGF-β, IL-13, diffuse proliferative, and IFN-α gene signatures in separate unsupervised clustering analyses. While these gene signatures were helpful in separating the majority of patients from controls, they did not correspond to known clinical subtypes of SSc including disease type and treatment status.    

Conclusion:

The majority of SSc patients have distinct (but heterogeneous) global gene expression profile when compared to controls. Within SSc group, the status of skin at the biopsy site is a major source of variability (affected versus unaffected skin). The observed heterogeneity in SSc transcriptomes does not correspond to the known clinical subtypes of disease.

---

« Back to 2012 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/correlates-of-skin-gene-expression-profile-in-systemic-sclerosis/