Oral Presentation Annual Meetings of the Endocrine Society of Australia and Society for Reproductive Biology and Australia and New Zealand Bone and Mineral Society 2016

Exploring the function of prostate cancer-associated germline variants in the PSA gene (#190)

Judith Clements 1 2 , Srilakshmi Srinivasan 1 , Carson Stephens 1 , Amanda Spurdle 3 , Hannu Koistinen 4 , The PRACTICAL Consortium , Australian Prostate Cancer BioResource , Jyotsna Batra 1
  1. Australian Prostate Cancer Research Centre-Queensland and Institute of Health and Biomedical Innovation, Queensland University of Technology, Translational Research Insitute-QUT, Brisbane, QLD, Australia
  2. QUT, Woolloongabba, QLD, Australia
  3. QIMR Berghofer Medical Research Institute, Herston, QLD
  4. Department of Clinical Chemistry, , University of Helsinki,, Helsinki, Finland.

Prostate-specific antigen (PSA) is the current clinical biomarker for prostate cancer diagnosis and monitoring disease progression. Our genetic fine-mapping studies identified two non-synonymous SNPs, rs61752561:G>A (Asp102 to Asn102 codon change) and rs17632542:T>C (Ile179 to Thr179 codon change), within the PSA or Kallikrein-3 (KLK3) gene to be significantly associated with prostate cancer risk. Further, these two SNPs are associated with low PSA levels at diagnosis and the rs17632542 SNP with reduced tumour volume (p=0.002). In this study, we sought to understand the potential molecular effect of these two non-synonymous SNPs on PSA mRNA expression and protein function. Recombinant forms of these 2 SNP generated PSA isoforms were used to biochemically assess the stability and activity compared to wild type PSA. PC3 prostate cancer cells stably expressing active wild-type PSA and the protein isoforms harbouring the SNP alleles were used to analyse the functional effect on proliferation, migration and invasion of these cells. Differential allele-based expression analysis was also employed to elucidate any changes in expression levels. Stable expression of wild-type PSA in PC3 cells increased proliferation and migration compared to the two SNP isoforms. Differences in protein stability, protease activity and glycosylation levels for the two SNP-generated protein isoforms were observed. One SNP also affects t PSA mRNA expression levels and, interestingly, an additional splice variant induced by this SNP was observed. Our results provide evidence that these two nonsynonymous SNPs within the PSA/KLK3 gene affect the levels of PSA/KLK3 mRNA expression and splicing, and function of the PSA protease, and suggest that they may be a contributor to the functional role of PSA in prostate cancer pathogenesis. Understanding the biological effect of these potentially functional variants will help to unravel the importance of this region observed by genome-wide association studies and may impact on current interpretation of the PSA test.