Spondin-2 (SPON2), a more prostate-cancer-specific diagnostic biomarker

Abstract

Background: Prostate-specific antigen (PSA) screening, although common, has recently been called into question. To find prostate cancer (PCa) diagnostic biomarkers that can make up for the defects of PSA, we compared the secretomes of several benign and PCa cell lines, selected candidate molecules, and then confirmed their clinical value.

Methodology/principal findings: We first identified extracellular proteins by two-dimensional gel electrophoresis (2-DE) coupled with liquid chromatography-tandem mass spectrometry (LC-MS/MS) identification. We then validated the secreted proteins on a cellular level, and finally determined whether they could be used as PCa diagnostic biomarkers using prostate tissue and serum specimens of Chinese volunteers by immunohistostaining and sandwich ELISA. We obtained credible extracellular protein 2-DE graphs of prostate cell lines. The 5 spots that showed superior repeatability were selected for LC-MS/MS analysis, which identified seven candidate molecules. One of the candidate molecules, spondin-2 (SPON2), was only expressed in the conditioned media (CM) of androgen receptor (AR) positive PCa cell lines. Using tissue microarray by immunohistostaining, we found SPON2 to be over-expressed in PCa. SPON2 staining was more intense in Gleason score sum 7-8 and in PCa patients with metastasis. By receiver operator characteristic (ROC) curve analysis, we found that the serum SPON2 level was elevated in PCa patients, showing sensitivity and specificity suitable for diagnostic use. We also found that SPON2 could be used to identify PCa patients with serum PSA levels no higher than 10 ng/ml from healthy elderly men.

Conclusion/significance: SPON2 is a new serum and histological diagnostic biomarker for PCa. It can avoid some of the problems of PSA testing and was here found to offer relatively high sensitivity and specificity relative to PSA.

Figure 1. Two-dimensional gel electrophoresis profiles of condition media from BPH-1, LNCaP, and C4-2 cells.

Two-dimensional electrophoregram of condition media from (A) BPH-1, (B) LNCaP, and (C) C4-2 cells are shown. (D) Enlarged corresponding areas of selected protein spots from these cell lines are also shown. Proteins in condition media of all cells were resolved by two-dimensional gel electrophoresis and silver-stained. White arrows indicate the differential spots.

Figure 2. Expression of spondin-2 (SPON2) in the condition media of androgen-receptor-positive prostate cancer cell lines.

(A) Semi-quantitative reverse transcription PCR analysis of five candidate molecules. (B) Western blot analysis of cell lysates and condition media protein with anti- SPON2 and anti- prostate-specific antigen (PSA) antibody. (C) Standard curve for SPON2 sandwich ELISA. (D) SPON2 quantitation in 10 µg extracellular protein in condition media of different prostate cell lines.

Figure 3. Spondin-2 (SPON2) immunohistostaining analysis of prostate tissue microarray.

SPON2 integral optical density (IOD) sums of all prostate cancer patients (A), group Gleason score sum 7–8 (B) and metastasis (C) were all higher than those of both normal and benign prostatic hyperplasia (SNK* represents SNK grouping test results and groups with the same letter are not significantly different; GS** represents Gleason score sum; M*** represents metastasis status.). Same regions of same samples staining with hematoxylin and eosin (HE) and SPON2 separately were shown in (D).

Figure 4. Serum concentration testing of spondin-2 (SPON2).

(A) SPON2 in 70 prostate cancer patients and 13 healthy elderly men were compared. Willcoxon 2 sample testing was used to determine the significance of the differences between two groups. (B) A SPON2 receiver operator characteristic (ROC) curve for above cases is shown. (C) SPON2 and prostate-specific antigen (PSA) ROC curves for prostate cancer patients with serum PSA≤10 ng/mL and healthy elderly men are shown.