DNA methylation biomarkers of prostate cancer: confirmation of candidates and evidence urine is the most sensitive body fluid for non-invasive detection

BACKGROUND

A prostate cancer (PCa) biomarker with improved specificity relative to PSA is a public health priority. Hypermethylated DNA can be detected in body fluids from PCa patients and may be a useful biomarker, although clinical performance varies between studies. We investigated the performance of candidate PCa DNA methylation biomarkers identified through a genome-wide search.

METHODS

Real-time PCR was used to measure four DNA methylation biomarkers: GSTP1 and three previously unreported candidates associated with the genes RASSF2, HIST1H4K, and TFAP2E in sodium bisulfite-modified DNA. Matched plasma and urine collected prospectively from 142 patients referred for prostate biopsy and 50 young asymptomatic males were analyzed.

RESULTS

Analysis of all biomarkers in urine DNA significantly discriminated PCa from biopsy negative patients. The biomarkers discriminated PCa from biopsy negative patients with AUCs ranging from 0.64 for HIST1H4K (95% CI 0.55-0.72, P < 0.00001) to 0.69 for GSTP1 (95% CI 0.60-0.77, P < 0.00001). All biomarkers showed minimal correlation with PSA. Multivariate analysis did not yield a panel that significantly improved performance over that of single biomarkers. All biomarkers showed greater sensitivity for PCa in urine than in plasma DNA.

CONCLUSIONS

Analysis of the biomarkers in urine DNA significantly discriminated PCa from biopsy negative patients. The biomarkers provided information independent of PSA and may warrant inclusion in nomograms for predicting prostate biopsy outcome. The biomarkers' PCa sensitivity was greater for urine than plasma DNA. The biomarker performances in urine DNA should next be validated in formal training and test studies.

The 17 kDa band identified by multiple anti-aquaporin 2 antisera in rat kidney medulla is a histone

The osmotic water permeability of epithelial cells of the inner medullary collecting duct of the kidney is regulated by antidiuretic hormone (ADH). ADH causes the insertion and removal of cytoplasmic vesicles containing the aquaporin (AQP-2) water channel protein which is recognized by multiple rabbit antipeptide antisera raised against amino acid sequences comprising its cytoplasmic carboxyl terminal. Immunoblots of rat kidney membrane fractions as well as human urine have all shown that AQP-2 is expressed exclusively by collecting duct cells and have identified a 29 kDa band (corresponding to the nonglycosylated AQP-2 protein), a broad 35-45 kDa band (corresponding to the mature glycosylated form of AQP-2 protein) and an additional immunoreactive 17 kDa band of unknown origin. We now report that the 17 kDa band identified by these anti-AQP-2 antisera is not an AQP-2 component but rather a denatured histone protein type H2A1. This binding of anti-AQP-2 antisera to denatured H2A1 present in protein samples derived from both kidney inner medulla and human urine is blocked specifically by preincubation of immunoblots with solutions containing the acidic protein gelatin.