1. ¹Department of Dermatology, University of Bonn, Bonn, Germany
2. ²Miltenyi Biotec GmbH, MACSmolecular Business Unit, Cologne, Germany
3. ³Department of Dermatology and Allergology, University Hospital RWTH Aachen, Aachen, Germany
4. ⁴Department of Dermatology and Venerology, University of Cologne, Cologne, Germany

Correspondence: Dr Joerg Wenzel, Department of Dermatology, University of Bonn, Sigmund-Freud-Strasse 25, Bonn 53105, Germany. E-mail: joerg.wenzel@ukb.uni-bonn.de

⁵These authors have contributed equally to this work.

Received 22 January 2007; Revised 13 March 2007; Accepted 29 March 2007; Published online 16 August 2007.

Abstract

Here, we present data of a gene expression profiling approach to apply the diagnostic value and pathological significance of this method in different inflammatory skin diseases, using whole skin biopsies. Initially, SAGE™ was performed to identify frequent tags differentially expressed in various skin diseases. On the basis of these results, a new skin pathology-oriented PIQOR™ microarray was designed. Lichen planus (LP) was chosen as a model disease to evaluate this system. Controls included healthy skin, atopic dermatitis (AD), and psoriasis (Pso). Gene expression analyses using the topic-defined microarray followed by unclassified clustering was able to discriminate LP from AD and Pso. Genes significantly expressed in LP included type I IFN inducible genes and a specific chemokine expression pattern. The CXCR3 ligand, CXCL9, was the most significant marker for LP. In situ hybridization and immunohistochemistry confirmed the results and revealed that keratinocytes are type I IFN producers in LP skin lesions. Our results show that gene expression profiling using a skin-specific microarray is a reliable method to identify patients with LP in the chosen context and reflect recent models concerning the pathogenesis of this disease. Gene expression profiling might complement the diagnostic spectrum in dermatology and may provide new pathogenetic insights.