History Features TRAMP and the Neuroendocrine Phenotype Uses TRAMP HistoPathology- self guided slide show Video Instruction Manual and Guide Despite the many parallels between TRAMP and clinical prostate cancer, there has been a lot of discussion and misconception concerning neuroendocrine cancer and the TRAMP model [Abate-Shen and Shen, 2002; Ellwood-Yen et al., 2003]. While the significance of the neuroendocrine compartment in prostate cancer remains to be clarified [see Abrahamsson et al., 1998], the neuroendocrine phenotype is associated with a poor prognosis and emergence of androgen indepepndent disease, and prostate cancer patients with advanced hormone-dependent and hormone-refractory disease often have increased levels of neuroendocrine markers chromogranin A (ChgA) and neuron-specific enolase (NSE) in their sera and tissue specimens [Angelsen et al., 1997; Kadmon et al., 1991; Tarle and Rados, 1991]. Clearly there is a neuroendocrine component to prostate cancer and this should be considered a positive attribute in prostate cancer models. In fact, there is growing evidence in the literature that prostate cancer cells posses an intrinsic plasticity that allows them to either transdifferentiate or de-differentiate and re-differentiate into cells with neuroendocrine-like properties. For example, human LNCaP epithelial cells can display neuronal-like morphology or differentiation when grown in steroid-reduced media or following treatment with IL-6 (50 ng/ml) or dibutyryl cAMP (0.1 mM) [Cox et al., 1999; Qiu et al., 1998; Zelivianski et al., 2001]. Some transgenic models, including the CR2-SV40 [Hu et al., 2002], Fetal-G(gamma)-globin-SV40 [Perez-Stable et al., 1997], 12T-5 and 12T-10 "LADY" [Kasper et al., 1998] models appear to develop cancers originating within the neuroendocrine compartment in the prostate. On the other hand we have recently demonstrated that emergence of the neuroendocrine phenotype in TRAMP as measured by expression of synaptophysin(SynP) is a stochastic event correlated with progression and the loss of differentiation, glandular architecture and hormonal response - features remarkably similar to those observed in clinical disease (see Figure 1 below). Moreover, cells of true neuroendocrine origin should not express AR, and as recently published in The Prostate, we demonstrate that in fact 16 out of 29 (55%) TRAMP tumors express both SynP and AR and that 9 out of 13 (70%) poorly differentiated tumors in castrated TRAMP mice expressed both SynP and AR. These findings further support our position that most TRAMP tumors are not of neuroendocrine origin and also demonstrate that expression of synaptophysin alone can not "prove" neuroendocrine origin. To further distinguish the emergence of the neuroendocrine phenotype in TRAMP from true neuroendocrine carcinoma, we have recently used in silico analysis to establish gene expression profiles in samples obtained from the CR2-Tag [Hu et al., 2002] mice representing neuroendocrine carcinoma and samples of progressive prostate cancer obtained from the TRAMP model . As shown in Figure 2, we note clear differences between the expression profiles for neuroendocrine markers between the TRAMP and CR2-Tag samples. Whereas the markers were uniformly and highly expressed in the primary CR2-Tag lesions, consistent with tumors of neuroendocrine origin, expression of neuroendocrine markers increased as a function of disease progression in TRAMP underscoring the stochastic nature of the TRAMP model and supporting the hypothesis that these adenocarcinomas, like LNCaP cells, display a certain intrinsic plasticity that allows them to phenocopy neuroendocrine cells and display neuroendocrine features. It is therefore our conclusion from these studies that the neuroendocrine phenotype in TRAMP emerges as a consequence of an "epithelial to neuroendocrine" transition (or switch) as a function of cancer progression. Hence, lesions displaying neuroendocrine histology in the TRAMP model should best be identified as "carcinomas with neuroendocrine features" rather than as neuroendocrine carcinomas. Figure 1: Expression of Synaptophysin in Primary and Metastatic Prostate Carcinomas from TRAMP We used immunohistochemistry with an anti-synaptophysin antibody (The Binding Site PH510; 1/200 dilution) to analyze tissue sections procured at necropsy from TRAMP mice. All sections were visualized with ABC detection kit (Vector Labs). Panels A-D (Animal 1228) Panel A: Well differentiated primary tumor characterized by well formed glands and desmoplastic stroma. No evidence of moderately or poorly differentiated carcinoma was identified and staining for synaptophysin was negative. Panel B: Large moderately differentiated lung metastasis that does not express synaptophysin is present on the left. There is also a small, poorly differentiated metastasis that expresses synaptophysin (right). Panel C: Moderately to poorly differentiated liver metastasis with focal expression of synaptophysin (center). Panel D: Moderately differentiated liver metastasis without synaptophysin expression. Panels E-F (Animal 885) Panel E: Well differentiated primary tumor. No evidence of moderately or poorly differentiated carcinoma was identified and stains for synaptophysin were negative. Note the positively staining ganglion on the left that serve as an internal positive control. Panel F, Moderately differentiated liver metastasis. No synaptophysin expression was identified. Panels G-H (Animal 1113) Panel G: Well differentiated primary tumor. No evidence of moderately or poorly differentiated carcinoma was identified and stains for synaptophysin were negative. Panel H, Lung metastasis expressing synaptophysin. A glandular structure is present in the metastasis. Panels I-J (Animal 1057) Panel I: Well differentiated primary tumor. No evidence of moderately or poorly differentiated carcinoma was identified and stains for synaptophysin were negative. Note the positively staining ganglion along the upper portion of tissue. Panel J: Lung metastasis and stains for synaptophysin were negative. All original magnifications: 200X Download High-res images Figure 2: In Silico Analysis of Neuroendocrine Progression in TRAMP and CR2-TAG Mouse Models. Tumor samples from TRAMP and CR2-Tag mice were harvested and RNA isolated and subjected to Expression Array Profiling. A progressive increase in neuroendocrine marker expression is observed to correlate with stage/invasiveness of TRAMP tumors, while expression is uniformly high in the CR2-Tag model. Elevated expression levels (greater than 2-fold) were detected in early TRAMP primary tumors (dorsal prostate), in seminal vesicle invasive extensions, and in lymph node metastasis. SVI, seminal vesicles invasive; MET, metastasis. In collaboration with Dr. Robert Bok, UCSF References Abate-Shen C, Shen MM (2002): Mouse models of prostate carcinogenesis. Trends Genet 18:S1-5. Abrahamsson, P. A., A. T. Cockett, et al. (1998). "Prognostic significance of neuroendocrine differentiation in clinically localized prostatic carcinoma." Prostate Suppl 8: 37-42. Angelsen, A., U. Syversen, et al. (1997). "Use of neuroendocrine serum markers in the follow-up of patients with cancer of the prostate." Prostate 31(2): 110-7. Cox ME, Deeble PD, Lakhani S, Parsons SJ (1999): Acquisition of neuroendocrine characteristics by prostate tumor cells is reversible: implications for prostate cancer progression. Cancer Res 59:3821-30. Ellwood-Yen K, Graeber TG, Wongvipat J, Iruela-Arispe ML, Zhang J, Matusik R, Thomas GV, Sawyers CL (2003): Myc-driven murine prostate cancer shares molecular features with human prostate tumors. Cancer Cell 4:223-38. Hu Y, Ippolito JE, Garabedian EM, Humphrey PA, Gordon JI (2002): Molecular characterization of a metastatic neuroendocrine cell cancer arising in the prostates of transgenic mice. J Biol Chem 277:44462-74 Kadmon, D., T. C. Thompson, et al. (1991). "Elevated plasma chromogranin-A concentrations in prostatic carcinoma." J Urol 146(2): 358-61. Kasper, S., P. C. Sheppard, et al. (1998). "Development, progression, and androgen-dependence of prostate tumors in probasin-large T antigen transgenic mice: a model for prostate cancer [corrected and republished article originally printed in Lab Invest 1998 Mar;78(3):319-33]." Lab Invest 78(6): i-xv. Perez-Stable, C., N. H. Altman, et al. (1997). "Prostate cancer progression, metastasis, and gene expression in transgenic mice." Cancer Res 57(5): 900-6. Qiu Y, Robinson D, Pretlow TG, Kung HJ (1998): Etk/Bmx, a tyrosine kinase with a pleckstrin-homology domain, is an effector of phosphatidylinositol 3'-kinase and is involved in interleukin 6-induced neuroendocrine differentiation of prostate cancer cells. Proc.Natl.Acad.Sci.U.S.A. 95:3644-3649. Tarle, M. and N. Rados (1991). "Investigation on serum neurone-specific enolase in prostate cancer diagnosis and monitoring: comparative study of a multiple tumor marker assay." Prostate 19(1): 23-33. Zelivianski S, Verni M, Moore C, Kondrikov D, Taylor R, Lin MF (2001): Multipathways for transdifferentiation of human prostate cancer cells into neuroendocrine-like phenotype. Biochim.Biophys.Acta 1539:28-43