Dissociation between androgen responsiveness for malignant growth vs. expression of prostate specific differentiation markers PSA, hK2, and PSMA in human prostate cancer models

Abstract Background A detailed understanding is evolving as to how androgen receptor (AR) functions as a transcriptional regulator via its binding to androgen response elements (ARE) within promoter and enhancer regions of prostate‐specific differentiation markers such as PSA, hK2, and PSMA. It has been assumed that an understanding of regulation of expression of these marker proteins would also provide an understanding of the mechanisms whereby AR interactions regulate proliferation and survival of malignant prostate cells. In order to validate this hypothesis, we used a series of human prostate cancer models [i.e., LAPC‐4, CWR22Rv1, MDA PCA‐2b, LNCaP, and C4‐2B (derived from LNCaP)] to test whether there is a consistent concordance between androgen responsive regulation for malignant growth vs. regulation of expression of prostate differentiation specific markers PSA, hK2, and PSMA. Methods In order to define androgen growth responsiveness in vivo, human prostate cancer cell lines were inoculated as xenografts into intact vs. surgically castrated adult male nude mice and the subsequent tumor growth response monitored. To assess androgen regulation of PSA and hK2 expression in these cell lines, the concentration of PSA and hK2 in the conditioned standard media and charcoal stripped media ± androgen from each cell line was determined using an immunoassay system. PSMA enzymatic activity was determined using the PSMA substrate 3 H N ‐acetylaspartylglutamate ( 3 H NAAG). Results Wild‐type AR expressing LAPC‐4 cells are androgen responsive for their in vivo growth. This cell line is also androgen sensitive for the expression of both PSA and hK2 in vitro and express PSMA. CWR22Rv1 cells have a mutated AR and are androgen responsive for growth in vivo and androgen sensitive for hk2 but not PSA expression. CWR22Rv1 produce ∼1.4‐fold more PSA, ∼18‐fold more hK2, and have 21‐fold higher PSMA activity than LAPC‐4 cells. MDA PCA‐2b cells are androgen responsive for growth in vivo and androgen sensitive for PSA expression. MDA PCA‐2b cells produce ∼250‐fold more PSA but almost equivalent amounts of hK2 compared to LAPC‐4 and have ∼19‐fold higher PSMA activity. Both late passage LNCaP and C4‐2B are androgen independent for growth in vivo but remain androgen sensitive for both PSA and hK2 expression. LNCaP cells produce ∼50‐fold more PSA, ∼35‐fold more hK2, and have 28‐fold higher PSMA activity compared to LAPC‐4. C4‐2B cells produce ∼80‐fold higher levels of PSA, ∼250‐fold higher levels of hK2. C4‐2B also the highest PSMA activity of the cell lines with 105‐fold higher PSMA activity than LAPC‐4 and ∼4‐fold higher activity than late passage LNCaP cells. Conclusions Androgen can coordinately regulate both the tumor growth and expression of prostate specific marker genes as observed for the LAPC‐4 human prostate cancer cells. Such coordinated regulation, however, is not universal. In all of the other cell lines, there is a dissociation between androgen responsive regulation of malignant growth vs. regulation of expression of prostate specific markers PSA and hK2. In addition, PSMA activity in these cell lines increases as cells become more androgen independent for growth in vivo. These results emphasize that tumor growth and the expression of the specific secretory genes are independently regulated molecular events even if they share a requirement for androgen and/or AR function. Additional independent mechanisms occur in prostate cancer cells for regulation of expression for even the highly related PSA and hK2 genes. Further studies are needed to clarify the mechanisms for androgen ligand‐independent, AR‐dependent regulation of the genes that directly effect the growth of androgen (i.e., ligand) independent prostate cancer cells. Unfortunately, the data in this present report do not validate the use of the PSA or hK2 gene as surrogates for a model system for such critically important mechanistic studies. Prostate 54: 249–257, 2003. © 2003 Wiley‐Liss, Inc.