Abstract
There are striking population disparities in prostate cancer (PCa) risk and survival outcome borne out of current health statistics data. This is particularly evident between African Americans (AA) and their Caucasian American (CA) counterparts. Epidemiologic studies have shown that higher mortality and recurrence rates of PCa is still seen in AA men even after adjustment for socioeconomic status, environmental factors and health care access. To test whether differential mRNA splicing may play a role in PCa health disparities, RNA from PCa and patient-matched normal prostate needle biopsies from AAs and CAs was processed and hybridized onto the Affymetrix human Exon 1.0 ST arrays. A 4-way statistical design (10% FDR, >1.5 fold-change) was employed in the following comparisons: AA normal vs. CA normal, AA cancer vs. CA cancer, AA cancer vs. AA normal, and CA cancer vs. CA normal. Our analysis identified >800 genes undergoing differential mRNA splicing between AA and CA cancer specimens. Notably, many of these genes reside within canonical oncogenic signaling pathways. To further delineate the functional consequences of these splice variants, exon-specific and/or exon junction-specific small interfering RNAs (siRNAs) were employed to suppress splice variants for 4 oncogenes (PIK3CD, FGFR3, MET and RASGRP2) in PCa cell lines derived from the AA and CA patient populations. The targeted knockdown of population-specific splice variants appears to have differential effects on cell proliferation and invasive. For example, the splice variant for phosphoinositide-3 kinase delta (PI3KCD) found in AA PCa specimens encodes a more aggressive version of the gene (i.e., leading to greater proliferation and invasion of cancer cells) compared to the variant counterpart found in CA PCa specimens. Taken together, our data suggest that differential mRNA splicing may be critical in PCa health disparities between the AA and CA populations. These findings should advance our knowledge on the molecular mechanisms underlying PCa health disparities, and may facilitate the development of novel strategies for PCa detection, diagnosis, prognosis, and therapy in the AA population. This work was supported by NCI grant 5U01-CA-116937, ACS-IRG-08-091-01 and the Affymetrix Collaborations in Cancer Research Program.
Citation Information: Cancer Epidemiol Biomarkers Prev 2011;20(10 Suppl):PL08-01.