Quantitative assessment of AKAP12 promoter methylation in human prostate cancer using methylation-sensitive high-resolution melting: correlation with Gleason score.

OBJECTIVES: To quantitatively investigate the A kinase anchoring protein 12 (AKAP12) gene promoter methylation and its association with clinicopathologic variables in human prostate cancer (PCa). The AKAP12 gene has shown reduced expression and marked hypermethylation in a variety of cancers.

METHODS: The percentage levels of DNA methylation were measured in 78 PCa, 22 benign prostatic hyperplasia, and 22 normal adjacent tissue samples using an AKAP12 methylation-sensitive high-resolution melting assay. AKAP12 gene expression was also examined in 4 human prostate carcinoma cell lines, PC-3, DU145, LNCaP, and 22RV1, using quantitative reverse transcriptase-polymerase chain reaction and methylation-sensitive high-resolution melting analysis and after DNA methyltransferase inhibition with 5-aza-2'-deoxycytidine.

RESULTS: Methylation (>1%) of the AKAP12 promoter region was present in 47 (60.2%) of the 78 PCa, 5 (22.7%) of the 22 benign prostatic hyperplasia, and 2 (9.1%) of the 22 adjacent normal tissue samples. AKAP12 methylation was significantly greater in the PCa than in the benign prostatic hyperplasia or adjacent tissue samples (P < .01). AKAP12 methylation was significantly greater in the PCa samples with higher Gleason scores (P = .03); however, no correlation was found with age, pT category, or serum prostate-specific antigen level. Reverse transcriptase-polymerase chain reaction demonstrated that PC-3 and DU-145 cells expressed AKAP12 RNA and LNCaP and 22RV1 did not. The AKAP12 locus was methylated in the LNCaP and 22RV1 cells. Treatment of LNCaP cells with 5-aza-2'-deoxycytidine markedly decreased the methylation levels and increased the expression of AKAP12.

CONCLUSIONS: The results of the present study have demonstrated that AKAP12 promoter methylation is a frequent event in human PCa. AKAP12 methylation represents a potential molecular biomarker for predicting the malignancy of PCa.