Expression and regulation of prostate androgen regulated transcript-1 (PART-1) and identification of differential expression in prostatic cancer

Prostate androgen regulated transcript 1 (PART-1), is a gene predominantly expressed in the prostate gland and is regulated by androgens in human prostate cancer cell lines. Here, we report additional characteristics of PART-1 tissue expression and hormonal regulation and study its expression profile in human normal and matched prostate cancer tissues. Since PART-1 shows similarity to prostate-specific antigen (PSA) in prostate specificity and regulation, we hypothesized that it may be implicated in prostate carcinogenesis or may be a potential new biomarker. We used reverse transcriptase polymerase chain reaction (RT-PCR) to further characterize PART-1 tissue expression and hormonal regulation in the LNCaP prostate cancer cell line. RT-PCR analysis revealed that PART-1 is expressed not only in the prostate and salivary gland, but also in other tissues, including the thymus and placenta. In addition to androgen stimulation, PART-1 is also up-regulated by progestins, oestrogens and glucocorticoids. We further studied the expression of PART-1 in 27 paired (from the same patient) cancerous and non-cancerous prostatic tissues, with qualitative and quantitative RT-PCR (LightCycler®technology), in order to examine whether PART-1 is overexpressed or underexpressed in cancer. Our results indicated that PART-1 is more frequently overexpressed in the cancerous prostatic tissue. We conclude that this gene is overexpressed in prostate cancer and may represent a novel prostate cancer tumour marker. © 2001 Cancer Research Campaign http://www.bjcancer.com


Prostate cancer cell line and hormonal stimulation procedures
The prostate cancer cell line LNCaP was purchased from the American Type Culture Collection (ATCC), Rockville, MD. Cells were cultured in RPMI media (Life Technologies, Inc.) and supplemented with glutamine (200 mmol l -1 ), bovine insulin (10 mg l -1 ), fetal bovine serum (10%), antibiotics and antimycotics in plastic flasks, to near confluency. The cells were then aliquoted into 24-well tissue culture plates and cultured to 50% confluency. 24 h prior to each procedure, the culture media were changed into phenol red-free media containing 10% charcoal-stripped fetal bovine serum. For stimulation, various steroid hormones dissolved in 100% ethanol were added into the culture media, at a final concentration of 10 -8 M. Cells stimulated with 100% ethanol were included as controls. The cells were grown for 24 h, after which they were harvested and total RNA extracted (see below).

Reverse transcriptase polymerase chain reaction
Using Trizol™ reagent (Life Technologies, Inc) and following the manufacturer's instructions, we extracted total RNA from the cell line LNCaP. Through spectrophotometry, RNA concentrations were determined. Using the Superscript™ preamplification system (Life Technologies, Inc), 2 µg of total RNA was reversetranscribed into first strand DNA. The gene-specific primers (Table 1) were designed through information on the genomic and cDNA structure of PART-1, obtained from NCBI. PCR reactions were carried out in a reaction mixture, containing 1 µl of cDNA, 1.5 mM MgCl 2 , 2.5 µl of 25 mM PCR buffer, 0.5 µl of 10 mM dNTPs (deoxynucleoside triphosphates), 0.5 µl (150 ng) of primers, 20 µl of H 2 O and 0.25 µl (1 unit) of HotStar Taq polymerase (Qiagen, Valencia, CA) on an Eppendorf Mastercycler gradient system (Eppendorf, Westbury, NY). The cycling conditions were 95°C for 15 minutes to activate the HotStar Taq polymerase, followed by 35 cycles of 94°C for 30 s, 62°C for 1 minute and a final extension at 72°C for 10 minutes.
Quantitative PCR was performed on a Roche LightCycler ® system (Roche Molecular Biochemicals, Mannheim, Germany). PCR reactions were carried out in a reaction mixture consisting of 15.2 µl of H 2 O, 2 mM MgCl 2 , 0.5 µl (150 ng) of primers, 1 µl of cDNA and 2 µl of LightCycler DNA Master SYBR ® Green I (Roche). Protocol conditions consisted of denaturation at 95°C for 10 minutes, followed by 40 cycles of 63°C for 55 s and 72°C for 40 s of extension and data acquisition, with a subsequent melting at 95°C for 1 s. Quantitative data analysis was made possible, through the use of PART-1 RNA from serially diluted prostate cDNA, using 1, 10, 100 and 1000-fold dilutions. These 4 samples provided the template on which a line of best fit was plotted and used as a standard curve for data interpretation after each run.
Construction of standard curves and the tabulation of second derivative peaks displayed the beginning and end of the log-linear phase of PCR products.
The amplified PCR products were also separated using 1.5% agarose gels. The isolated bands were then purified using the Qiagen gel purification kit. The PCR products were cloned into the pCR 2.1-TOPO vector (Invitrogen, Carlsbad, CA, USA) in order to verify their identity by sequencing. The inserts were sequenced in both directions using vector-specific primers with an automated DNA sequencer.

Screening for PART-1 transcripts in prostatic tissues
Included in this study were tissue samples from 27 patients with prostatic cancer, that had been surgically removed by radical retropubic prostatectomy. Patient age ranged from 50-68 years, with a median of 64. Cancerous and non-cancerous pieces of the whole prostates were carefully excised and verified by histopathological examination. All patients had a histologically confirmed diagnosis of primary cancer and received no treatment before surgery (except patient 1080, who was treated with antihormonal therapy 4 weeks prior to surgery).
All tissue samples were minced with a scalpel, on ice and immediately transferred into a 2 ml polypropylene tube. Tissue samples were then homogenized and RNA extracted using Trizol™ reagent (Life Technologies, Inc) following the manufacturer's recommendations. The RNA concentration was determined spectrophotometrically and 2 µg of total RNA was reverse-transcribed into first strand cDNA as described above.

Tissue expression of the PART-1 gene
As shown in Figure 1, the PART-1 gene is highly expressed in the prostate, placenta, thymus, and salivary gland, and at lower levels in trachea, kidney and brain. The PCR-products were subsequently sequenced to verify the RT-PCR specificity.

Hormonal regulation of the PART-1 gene
To verify whether the PART-1 gene is under steroid hormone regulation, the prostate carcinoma cell line LNCaP was used. PSA, which is up-regulated by androgens and progestins, was used as a control gene. Our results show that PART-1 is up-regulated by dihydrotestosterone (DHT) and to a lower extent by oestradiol, norgestrel (a synthetic progestin) aldosterone and dexamethasone ( Figure 2).

Malignant versus non-malignant prostatic tissues
To determine the differential expression levels of PART-1 in normal (benign) and malignant (cancer) tissues, we analysed 27 pairs of prostatic tissue extracts (normal/cancer) through qualitative and quantitative PCR. We determined that 18 out of 27 patients had elevated PART-1 in the cancerous tissue, 7 had decreased PART-1 levels in the cancerous tissue compared to the non-cancerous tissue, while two patients had similar levels in both ( Table 2). The differential expression of PART-1 between normal and tumour tissues was 93%, with 67% (18 out of 27) of the matched prostate samples displaying tumour-associated over expression of PART-1. The distribution of quantitative data in the cancerous and non-cancerous tissues is displayed in Figure 3. The ratio of PART-1 versus actin was used to normalize the data. The difference between PART-1/actin levels in cancerous versus non-cancerous tissues ranged from approximately 0.03 to 0.24. Our quantitative results were further correlated with qualitative data which were generated by running the amplified samples on gels. For 20 out of the 27 pairs (74%) the data were concordant by both methods (Figure 4). The discordant samples included those with relatively small differences in PART-1 transcripts between cancer and normal tissues.

DISCUSSION
Through the use of a cDNA microarray-based approach (Lin et al, 2000), were able to identify PART-1 and characterize it as an androgen-induced gene in prostate adenocarcinoma cells, with an expression restricted to the prostate and salivary glands. They further demonstrated that the cDNA sequence of PART-1 encodes for a 60-amino acid protein (Lin et al, 2000).
Our study extends the data related to PART-1 expression and regulation. By using a more sensitive detection system, RT-PCR, rather than the Northern blot analysis used by Lin et al, we were able to detect and verify that PART-1 is also highly expressed in the placenta and thymus and to a lower extent in other tissues. PART-1 is regulated, in prostate carcinoma cells, by a variety of hormones, although primarily by androgens. It is possible that the up-regulation of PART-1 by hormones other than androgens may by due to cross talk between these hormones and the mutant androgen receptor of LNCaP cells (Veldscholte et al, 1992;McDonald et al, 2000). This mutant androgen receptor loses its binding specificity to androgens.
The finding that PART-1 is frequently present at higher levels in malignant than in benign tissue, suggests that PART-1 expression is  altered during cancer and that its regulation follows different mechanisms than PSA, which is usually found in higher levels in benign tissue (Magklara et al, 2000). The data with PSA, which were derived by using tissue extracts from the same patients, further confirm that PART-1 overexpression is not due to higher levels of luminal epithelium in the cancerous tissues, since PSA is also produced by prostatic luminal epithelium. PART-1 may be a secreted protein and its overexpression in cancer lead us to hypothesize that PART-1 may have value as an additional circulating biomarker for prostate cancer. This possibility merits investigation. PART-1 is localized to human chromosome 5q12 and encodes for a 60-amino acid protein (Lin et al, 2000). The small size of this  polypeptide, along with its hormonal regulation and its restricted tissue expression pattern, indicates that PART-1 has characteristics reminiscent of a hormone or a growth factor. Isolation of the protein and detailed functional and homology analysis should further illuminate PART-1's function.
In conclusion, our finding of overexpression of PART-1 in a significant percentage of prostate cancers, provides a foundation for future studies examining the potential of PART-1 as a prostate cancer marker and its biological function in the prostate and other tissues.