Expression of biomarkers modulating prostate cancer angiogenesis: differential expression of annexin II in prostate carcinomas from India and USA

Abstract

Background: Prostate cancer (PCa) incidences vary with genetic, geographical and ethnic dietary background of patients while angiogenesis is modulated through exquisite interplay of tumor-stromal interactions of biological macromolecules. We hypothesized that comprehensive analysis of four biomarkers modulating angiogenesis in PCa progression in two diverse populations might explain the variance in the incidence rates.

Results: Immunohistochemical analysis of 42 PCa biopsies reveals that though Anx-II expression is lost in both the Indian and American population with Gleason scores (GS) ranging between 6 and 10, up to 25 % of cells in the entire high grade (GS > 8) PD PCa samples from US show intense focal membrane staining for Anx-II unlike similarly graded specimens from India. Consistent with this observation, the prostate cancer cell lines PC-3, DU-145 and MDA PCa 2A, but not LNCaP-R, LNCAP-UR or MDA PCa 2B cell lines, express Anx-II. Transcriptional reactivation of Anx-II gene with Aza-dC could not entirely account for loss of Anx-II protein in primary PCa. Cyclooxygenase-2 (COX-2) was moderately expressed in most of high grade PIN and some MD PCa and surrounding stroma. COX-2 was not expressed in PD PCa (GS approximately 7-10), while adjacent smooth muscles cells stained weakly positive. Decorin expression was observed only in high grade PIN but not in any of the prostate cancers, atrophy or BPH while stromal areas of BPH stained intensively for DCN and decreased with advancing stages of PCa. Versican expression was weak in most of the MD PCa, moderate in all of BPH, moderately focal in PD PC, weak and focal in PIN, atrophy and adjacent stroma.

Conclusions: Expression of pro- and anti-angiogenic modulators changes with stage of PCa but correlates with angiogenic status. Focal membrane staining of Anx-II reappears in high grade PCa specimens only from US indicating differential expression of Anx-II. COX-2 stained stronger in American specimens compared to Indian specimens. The sequential expression of DCN and VCN in progressive stages was similar in specimens from India and USA indicating no population-based differences. The mechanistic and regulatory role of Anx-II in PCa progression warrants further investigation.

Figure 1

The photomicrograph of representative prostate archival specimens stained with H&E. (A-D) shows areas of BPH, high grade PIN, moderately-differentiated and poorly-differentiated prostate adenocarcinoma (PCa) and residual normal glands (indicated as open arrows). Immunostaining for endothelial cell specific marker (anti-CD34, E-H) at magnification × 200 shows the extent of anti-CD34 stained organizing endothelial cell in areas of PIN as well as intraductal venule cross-sections in PCa (indicated as stealth arrows). No CD34 staining was seen in BPH area.

Figure 2

The photomicrograph of representative prostate archival specimens immunostained with anti-Anx-II (A-D) and anti-COX-2 (E-H) antibodies at magnification × 200 shows positively stained Anx-II regions of normal epithelia (open arrows) and BPH as opposed to loss of Anx-II in PCa. High grade PIN is negative. The COX-2 immunostaining staining is present in high grade PIN but decreases in intensity from moderately-differentiated carcinoma to poorly-differentiated carcinoma. BPH and normal epithelia are negative. The stromal cells in the areas of poorly differentiated PCa however stain strongly.

Figure 3

Representative magnified images of high grade PCa specimens from American (panel A) and Indian (panel B) shows distinct pattern of Anx-II immunostaining. Intense focal staining of cell membranes in American specimens contrasts to very weak staining in Indian specimens.

Figure 4

Western Blot profile of Anx-II and DCN expressed in prostate cancer cell lines. Each lane was loaded with 20 μg of total protein from lysates of indicated cell lines and detected with appropriate antibodies as mentioned in methods section. The samples are HPV-18C-1 (lane 1), LNCaP-R (lane 2), LNCaP-UR (lane 3), PC-3 (lane 4), DU-145 (lane 5), MDA PCa 2A (lane 6) and MDA PCa 2B (lane 7). PGK was included as internal control.

Figure 5

Anx-II mRNA expression, but not protein, is induced by 5-aza-2'-deoxycytidine, A: LNCaP and DU-145 cells were treated with or without 3 μM 5-aza-2'-deoxycytidine for 48 hours, and the annexin II expression was measured by RT-PCR. M: marker; lane 1: untreated LNCaP; lane 2: Aza-dC treated LNCaP; lane 3 untreated DU-145; lane 4 Aza-dC treated DU-145. B: LNCaP cells were treated with or without 3 μM 5-aza-2'-deoxycytidine for 48 hours. 10 μg of total protein was subjected to Western blot as described in Materials and Methods. 10 μg of whole cell extract of K562 cells was used as a positive control. Lane 1: treated LNCaP; Lane 2: untreated LNCaP; Lane 3: mock-treated LNCaP (with Lipofectamine); Lane 4: K562 cell extract. The bottom panel shows expression of an internal control in the same samples. PGK was used as an internal control.

Figure 6

The photomicrograph of representative prostate archival specimens stained with anti-DCN (A-D) and anti-VCN (E-H) at magnification × 200 shows only PIN regions as DCN positive while BPH, moderately-differentiated and poorly-differentiated carcinomas are negative. VCN staining is strongly positive in BPH and moderately-differentiated carcinoma while poorly-differentiated carcinoma shows moderate staining. PIN is negative for VCN expression.