Background/Purpose: Many genes, many of them still unknown, are involved in the etiology and development of OA. Today, different tools are available to try to identify some of the key genes related to the OA process. Therefore, the objective of this work is to perform a genome-wide expression assay in order to identify different expression profiles in the OA disease

Methods: Total RNA from OA cartilage samples was isolated with RNeasy Kit (Qiagen, Madrin, Spain) following manufacturerxs instructions. RNA was checked for integrity and purity with the Agilent Bioanalyzer (Agilent Technologies) and NanoDrop spectrophotometer (Thermo Scientific). 150 nanograms of total RNA were used for cDNA synthesis using the Ambion WT Expression kit (Ambion). The fragmented cDNA was hybridized against the Human Gene 1.1 ST array strip (Affymetrix) and scanned using the GeneTitan system (Affymetrix). Quality controls, normalization, pre-processing, differential gene expression and functional analyses were carried out with Bioconductor packages using R software.

Results: Human Gene 1.1 ST Array, which interrogates more than 28,000 well-annotated genes of 33,297 probes, was used for studying the genome wide expression profile of 23 OA-patients cartilage samples. A non-specific filtering was previously applied for removing those probes with non-annotation information and with low intra-array variation. An unsupervised machine learning approach revealed a group of samples highly different (Figure 1). The differential expression analysis between this cluster, comprising a total of six OA-patients, and the rest of the samples allowed for the identification of 176 differentially expressed probes with an adjusted p value below 0.0001. The analysis of the biological processes related to these differentially expressed genes showed that inflammation and immune processes were the main pathways found to be altered when a gene set enrichment analysis was applied.

Conclusion: The genome-wide expression analysis shows a clearly distinct profile for a group of OA patients. Both inflammation and immune response processes appeared to be significantly altered and revealed as key factors in the development of the OA disease.

Figure 1. Dendrogram showing the two different groups of OA patients. Cluster 2 shows a characteristic expression profile among OA patients.