Elevated expression of 12/15-lipoxygenase and cyclooxygenase-2 in a transgenic mouse model of prostate carcinoma.

Changes in expression of arachidonic acid (AA) metabolizing enzymes are implicated in the development and progression of human prostate carcinoma (Pca). Transgenic mouse models of Pca that progress from high-grade prostatic intraepithelial neoplasia (HGPIN) to invasive and metastatic carcinoma could facilitate study of the regulation and function of these genes in Pca progression. Herein we characterize the AA-metabolizing enzymes in transgenic mice established with a prostate epithelial-specific long probasin promoter and the SV40 large T antigen (LPB-Tag mice) that develop extensive HGPIN and invasive and metastatic carcinoma with neuroendocrine (NE) differentiation. Murine 8-lipoxygenase (8-LOX), homologue of the 15-LOX-2 enzyme that is expressed in benign human prostatic epithelium and reduced in Pca, was not detected in wild-type or LPB-Tag prostates as determined by enzyme assay, reverse transcription-PCR, and immunohistochemistry. The most prominent AA metabolite in mouse prostate was 12-HETE. Wild-type prostate (dorsolateral lobe) converted 1.6 +/- 0.5% [(14)C]AA to 12-HETE (n = 7), and this increased to 8.0 +/- 4.4% conversion in LPB-Tag mice with HGPIN (n = 13). Quantitative real-time reverse transcription-PCR and immunostaining correlated the increased 12-HETE synthesis with increased neoplastic epithelial expression of 12/15-LOX, the leukocyte-type (L) of 12-LOX and the murine homologue of human 15-LOX-1. Immunostaining showed increased L12-LOX in invasive carcinoma and approximately one-half of metastatic foci. COX-2 mRNA was detectable in neoplastic prostates with HGPIN but not in wild-type prostate. By immunostaining, COX-2 was increased in the neoplastic epithelium of HGPIN but was absent in foci of invasion and metastases. We conclude that (a) AA metabolism in wild-type mouse prostate differs from humans in the basal expression of LOXs (15-LOX-2 in human, absence of its 8-LOX homologue in mouse prostate); (b) increased expression of 12/15-LOX in HGPIN and invasive carcinoma of the LPB-Tag model is similar to the increased 15-LOX-1 in high-grade human Pca; and (c) the LPB-Tag model shows increased COX-2 in HGPIN, and therefore, it may allow additional definition of the role of this enzyme in the subset of human HGPINs or other precursor lesions that are COX-2 positive, as well as investigation of its contribution to neoplastic cell proliferation and tumor angiogenesis in Pca.

Pathological parameters of radical prostatectomy for clinical stages T1c versus T2 prostate adenocarcinoma: decreased pathological stage and increased detection of transition zone tumors.

PURPOSE: Studies of radical prostatectomy specimens have suggested that the majority of prostate specific antigen detected (clinical stage T1c) tumors are clinically significant. We compared tumor location and pathological parameters in the radical prostatectomy specimens of stages T1c versus T2 cases in a 3-year period. The percent of stage T1c disease represented a stable majority of patients undergoing treatment for clinically localized prostate cancer. MATERIALS AND METHODS: From January 1, 1998 to December 31, 2000, 417 radical prostatectomies were performed at Vanderbilt University, including 246 for stage T1c and 108 for stage T2 disease. A total of 37 patients were excluded from study because of neoadjuvant antiandrogen treatment. Pathological parameters, including tumor location in the transition and/or peripheral zone, tumor Gleason grade, tumor stage, total tumor volume and surgical margins were compared in stages T1c and T2 cases, and in transition versus peripheral zone stage T1c tumors in completely embedded whole mount specimens. RESULTS: In contrast to stage T2 lesions, stage T1c tumors were of significantly lower Gleason score with a higher percent of Gleason score 5 and lower percent of Gleason score 6, 7 and 8 or greater. They also had a significantly smaller volume and lower pathological stage. Of stage T1c tumors 77% were organ confined versus 62% of stage T2 tumors. There was no statistically significant increase in clinically insignificant neoplasms in stages T1c versus T2 cases (13% versus 7%) when using a volume criterion of less than 0.2 cc but a statistically significant increase in clinically insignificant disease was observed using a volume criterion of less than 0.5 cc (22% versus 9%). Whereas none of the T2 tumors were located in the transition zone and 17% were located in the transition and peripheral zones, 14% of stage T1c lesions were exclusively in the transition zone, with another 17% in the transition and peripheral zones. Compared with peripheral zone tumors transition zone stage T1c tumors had a lower Gleason score with an increase in Gleason score 5 and lower percent of Gleason score 6, 7 and 8 or greater. Although transition zone stage T1c lesions were significantly larger than peripheral zone stage T1c lesions, they had a lower pathological stage with 94% versus 72% organ confined. CONCLUSIONS: Prostate specific antigen detected stage T1c tumors had a lower grade, stage and volume than stage T2 tumors during the same period. Lower tumor grade in stage T1c cases is due at least in part to the increased detection of Gleason pattern 2 containing transition zone tumors. Despite the larger size, T1c transition zone tumors appear to be more favorable with higher rates of organ confined and lower grade tumors. If such transition zone tumors prove to be biologically distinct, improved strategies to identify these lesions preoperatively may result in more conservative treatment recommendations.