Current prostate cancer (PCa) biomarkers such as PSA are not optimal in distinguishing cancer from benign prostate diseases and predicting disease outcome. To discover additional biomarkers, we investigated PCa-specific expression of novel unannotated transcripts. Using Affymetrix Human Exon Arrays and RNA sequencing data, we identified 334 candidates referred to as EMC PCa-associated transcripts (EPCATs). These transcripts are uniquely expressed in subsets of PCa and not or barely expressed in normal prostate and other tissues. To validate their unique expression pattern, 15 EPCATs were validated by RT-PCR in cell lines and patient samples. Combined into a diagnostic panel, 11 EPCATs classified 80% of PCa samples correctly, while maintaining 100% specificity. High specificity was confirmed by in situ hybridization for EPCAT4R966 and EPCAT2F176 (SChLAP1) on extensive tissue microarrays. Besides being diagnostic, EPCAT2F176 and EPCAT4R966 showed significant association with pT-stage and were present in cancer precursor PIN lesions. We also found EPCAT2F176 and EPCAT2R709 to be associated with development of metastases and PCa-related death, and EPCAT2F176 to be enriched in lymph node metastases. Functional significance of expression of nine EPCATs was investigated by siRNA transfection, revealing that knockdown of five different EPCATs impaired growth of LNCaP and 22RV1 PCa cells. Two EPCATs inhibited the migration of PC3 cells in a Boyden chamber assay.
The EPCATs investigated so far do not exhibit a protein coding potential and are classified as long noncoding RNAs (lncRNAs). Ours and many other studies have now shown that our transcriptome consists of a limited number of coding RNAs (~22 000) and a huge variety of small and long noncoding transcripts. Surprisingly, many of these transcripts are uniquely expressed in one type of tissue or cancer. The observation that some of these lncRNAs are functionally relevant indicates that they do not just reflect random disease-related changes, but rather another layer of cellular regulatory complexity. Although, the underlying transcriptional regulation is not fully understood, the novel PCa-associated transcripts are new diagnostic and prognostic markers with functional relevance to PCa growth.