6.3. LNCaP model and transgenic mice model

The LNCaP cells were originated from a lymph node metastatic lesion of human prostatic adenocarcinoma, and have been widely used in the study of prostate cancer (Vaalara et al. 1998, 2000). Unlike prostatic cell lines PC-3 and DU-145, the growth of LNCaP cells is androgen-dependent, and the cells could be transformed to androgen-independent clones (van Steenbrugge et al. 1991). The secretion of PSA and hPAP, two major prostate epithelium-specific differentiation antigens, serves conventionally as a marker of androgen action on prostate cells (Andrews et al. 1992, Henttu et al. 1992). The androgen-responsive feature of LNCaP also makes it a useful model in studying the transcriptional regulation of prostate-related genes, since the expression of many prostate-specific proteins requires functionally differentiated, androgen-responsive cells. LNCaP is even more valuable in examining the transcriptional regulation of the hPAP gene because it is the only commercially available human prostate cancer cell line which can express endogenous hPAP (Horoszewice et al. 1983, Andrews et al. 1992, Henttu et al. 1992). The androgen-independent prostatic cancer cell line PC-3 does not express detectable levels of AR mRNA (Brolin et al. 1992) or hPAP mRNA in Northern blot analysis (Solin et al. 1990), and PC-3 cells are unable to secrete hPSA or hK2 (Garcia-Arenas et al. 1995). In PC-3 cells, the hPAP gene seems to be normal according to Southern blot analysis (Garcia-Arenas et al. 1995) and partial sequencing (unpublished data). The forced expression of AR in PC-3 cells cannot trigger the transcription of the hPAP gene (Garcia-Arenas et al. 1995). Most likely, some transcription factors essential for hPAP expression in LNCaP cells are missing from PC-3 cells.

Preparing a method for the transfection of LNCaP cells offers a fascinating possibility for studying the promoter activity of the hPAP gene and the transcriptional regulation of the hPAP promoter in a cell culture model. The density of the cells at the moment of plating is important for successful transfection (I). Cell culture density can influence gene expression and protein levels of biologically important molecules, including growth factors and their receptors (Singh et al. 1996, Monget et al. 1998, Pfeiffer et al. 1998), in epithelial cells. In LNCaP, the cell culture density not only affects the level of hPAP mRNA (Lin et al. 1994), but also modulates the androgen regulation of hPAP mRNA (Lin & Garcia-Arenas 1994). The optimal cell density plated for transfection is 1 x 106cells/100mm plate (I). Under this growth condition, androgen suppresses hPAP mRNA and stimulates the secretion of hPAP (Henttu et al. 1992, Lin et al. 1993a, Lin et al. 1993b). LNCaP maintained at such a high density with a slow growth rate could mimic the differentiated state of prostate epithelial cells (Lin et al. 1994).

The puberty effect indicates that the expression of the hPAP gene is under androgen influence. In normal rat prostate, two of the transcripts encoding PAP were up-regulated by androgens, whereas one of the mRNA species was insensitive to the hormone status. The androgen-sensitive transcripts were detectable 4 days after castration (Porvari et al. 1995). These results indicate that PAP gene expression in a normal prostate is a complicated process and is not exclusively androgen-regulated. The mRNA studies using tissue slices from various benign prostatic hyperplastic glands suggested that DHT is necessary to sustain the expression of hPAP in hyperplastic prostates (Dulinska et al. 1997). Results from the transient transfection of LNCaP cells might reflect the case in the cancer status. A number of changes might occur in LNCaP cancer cells compared to normal prostate epithelial cells. A T877A mutation at the LBD of AR in LNCaP makes the receptor responsive to other steroid hormones. AR coactivators could be also differently expressed in LNCaP compared to the normal prostate cells or other prostate cancer cell lines. Normal prostate cells express the receptor protein phosphatase PTP, whereas LNCaP cells do not. PTP regulates the PKC pathway to restore E-cadherin-dependent adhesion of the prostate epithelial cells via its interaction with RACK1 (Hellberg et al. 2002). It is possible that the control of hPAP expression is disturbed in prostate cancer cells and may have little linkage to the function of the AR (Garcia-Arenas et al. 1995).

The first intron fragment +57/+467 of the hPAP gene decreases promoter activity in transient transfections, while the reporter gene led by hPAP -734/+467 has a relatively high expression in the prostate of transgenic mice compared to that by hPAP -734/+50. This might be due to the difference between the normal prostate cells and cancer cells, but it could be also due to the difference between the cell culture and animal models. In the cell culture model, cells lack the interaction with other types of cells. The cells might then lose the in vivo autocrine or paracrine influences. Interleukin-6 (Il-6) is an autocrine factor in human prostate cancer (Giri et al. 2001). Il-6 protein concentrations are increased approximately 18-fold in clinical localized prostate cancers when compared to normal prostate tissue. Normal and neoplastic prostatic epithelial cells in culture, with the exception of LNCaP cells, secrete Il-6. The concentration of the Il-6 receptor is increased eightfold in prostate cancer tissues and increased in prostate cancer cell lines. Addition of exogenous Il-6 to primary epithelial cells in culture or to LNCaP cells leads to phosphorylation of signal transducers and activators of transcription 3 (STAT3) and increases in net cell proliferation. Furthermore, the Il-6 signal transduction pathway could cross-talk with AR signal transduction pathway (Ueda et al. 2002). Immunoprecipitation and transactivation studies showed a direct interaction between the amino acids 234-558 of the AR and STAT3 following Il-6 treatment of LNCaP cells. Inhibitors of MAPK and JAK decreased Il-6-induced phosphorylation of MAPK and activation of the AR N-terminal domain. Although native AR was always co-transfected in the cell culture model, it is quite possible that AR is not fully activated in LNCaP cells.

The transgenic animal model provides the in vivo information of gene regulation, but this model is expensive and time-consuming. The variation in copy number of the transgene and the integration site could sometimes cause problems.