Effects of Androgens on Telomerase Activity inNormal and Malignant Prostate Cells In Vitro

Hiroshi Soda,1,2* Eric Raymond,1 Sunil Sharma,1 Richard Lawrence,1Karen Davidson,1 Mikio Oka,2 Shigeru Kohno,2 Elzbieta Izbicka,1 and

Daniel D. Von Hoff1

1Institute for Drug Development, Cancer Therapy and Research Center, San Antonio, Texas2Second Department of Internal Medicine, Nagasaki University School of Medicine,

Sakamoto, Nagasaki, Japan

BACKGROUND. Recent studies have shown that sex hormones regulate telomerase activityin endometrium and breast tissues. The present study was designed to clarify the effects ofandrogen on telomerase activity in normal and malignant prostate cells.METHODS. Androgen-sensitive (LNCaP) and -independent (TSU-Pr1 and DU145) prostatecancer cell lines and normal prostate cells including basal cells were cultured in the presenceor absence of 5a-dihydrotestosterone (DHT).RESULTS. Prostate cancer cell lines exhibited high telomerase activity, and normal prostatecells showed low activity. Short or prolonged androgen-deprivation reduced telomerase ac-tivity in LNCaP cells, and DHT induced telomerase activity at the G1 phase of the cell cycle.DHT did not modulate telomerase activity in TSU-Pr1, DU145, and normal cells.CONCLUSIONS. LNCaP cells had an androgen-dependent pathway to activate telomerase,whereas TSU-Pr1 and DU145 cells as well as normal prostate cells had an androgen-independent pathway. These findings suggest that the regulatory mechanism of telomerasevaries during the progression of prostate cancers. Prostate 43:161–168, 2000.© 2000 Wiley-Liss, Inc.

KEY WORDS: telomerase regulation; testosterone; LNCaP cells; cell cycle

INTRODUCTION

Human telomeres are a repeated TTAGGG se-quence located at the end of chromosomes. Telomeresshorten with cell division, and the loss of telomeres isthought to trigger cellular senescence. Telomerase isan RNA-dependent DNA polymerase which synthe-sizes the telomere repeats, using an internal RNA tem-plate [1]. Approximately 80–90% of prostate cancershave telomerase activity with shorter telomeres [2,3].Although telomerase activity is not usually observedin normal prostate tissue, telomerase activity is de-tected in proliferating normal human prostate cells inculture [4] and collagenase-dispersed cells from nor-mal rat prostate [5].

Sex steroid hormones regulate telomerase activityin normal endometrium and breast tissues during themenstrual cycle [6,7]. Since normal prostate glands are

maintained by circulating androgen and prostate can-cers show an initial response to androgen-deprivationtherapy, we hypothesized that androgen may regulatethe telomerase activity of normal and cancerous pros-tate cells. This study showed that androgen regulatedtelomerase activity in the androgen-sensitive LNCaPcell line, but not in androgen-independent cancer celllines or proliferating normal cells in culture.

Grant sponsor: National Cancer Institute; Grant number: NationalCooperative Drug Discovery Grant U19CA67760; Grant sponsor:Association pour la Recherche contre le Cancer; Grant sponsor: Na-tional Institutes of Health.*Correspondence to: Hiroshi Soda, M.D., Second Department of In-ternal Medicine, Nagasaki University School of Medicine, 1-7-1Sakamoto, Nagasaki 852-8501, Japan.E-mail: okamikio@net.nagasaki-u.ac.jpReceived 20 August 1999; Accepted 28 December 1999

The Prostate 43:161–168 (2000)

© 2000 Wiley-Liss, Inc.

MATERIALS AND METHODS

Cell Culture

Cells of human prostate cancer cell lines, LNCaP,DU145, and TSU-Pr1, were maintained in RPMI-1640medium with 10% fetal calf serum (FCS) (JRH Biosci-ences, Lenexa, KS), 50 U/ml penicillin and 50 mg/mlstreptomycin. Normal human prostate epithelial cells(Clonetics Co., San Diego, CA) were obtained from awhite male without prostate cancer and benign pros-tatic hypertrophy. Normal cells were positive for cy-tokeratins 5 and 8, suggesting that these cells containbasal cells of the prostate [8]. Normal cells were cul-tured in serum-free PrEBM medium (Clonetics Co.).Normal cells had a limited life span in culture, andcells were analyzed after two passages. Cells were cul-tured at 37°C in a humidified 5% CO2 atmosphere.

Cell Proliferation

Prostate cancer cells at 2 × 104/well and normalcells at 3 × 103/well were plated in 24-well cultureplates in RPMI-1640 with 10% FCS, and allowed toadhere for 24 hr. The next day, the media were re-moved, and the cells were cultured for 4 days inRPMI-1640 with 10% charcoal-stripped calf serum(CCS) (Sigma Chemical Co., St. Louis, MO) in the pres-ence or absence of 10−6–10−10 M 5a-dihydrotestoster-one (DHT) (Sigma Chemical Co.). Normal cells werealso cultured for 4 days in the PrEBM medium with orwithout 10−6–10−10 M DHT. To further evaluate theeffect of prolonged androgen-deprivation, LNCaPcells were cultured in RPMI-1640 with 10% CCS for 1month. Cell numbers were measured in triplicate witha Coulter (Hialeah, FL) counter, and the viable cellswere determined on a hemocytometer with trypanblue.

Telomeric Repeat Amplification Protocol(TRAP) Assay

Telomerase activity was evaluated using a modi-fied TRAP assay kit (Oncor, Inc., Gaithersburg, MD)[9]. Prostate cancer cells were cultured for 7 days inRPMI-1640 containing 10% FCS, or 10% CCS with/without 10−6–10−10 M DHT, and normal cells were cul-tured in PrEBM with/without 10−6–10−10 M DHT.Harvested cells were resuspended in the detergent-based lysis buffer (200 ml/107 cells). After 45 min onice, the lysate was centrifuged for 1 hr at 100,000g. Theprotein concentration of the supernatant was mea-sured using a Bradford assay (Bio-Rad Laboratories,Hercules, CA). The cell extract was diluted to the con-centration of 0.1 mg/ml of protein, and 2 ml of the

extract were incubated at 30°C for 30 min in 50 ml ofreaction solution containing 0.1 mg 32P-end labeled TSprimer (58-AATCCGTCGAGCAGAGTT-38), 50 mMeach of dNTPs, 20 mM Tris-HCl (pH 8.3), 1.5 mMMgCl2, 63 mM KCl, 0.05% Tween-20, 1 mM EGTA,0.01% bovine serum albumin, 0.1 mg reverse primer,and 2 U Taq polymerase. To monitor the efficiency ofthe polymerase chain reaction (PCR), an internal con-trol together with a specific primer was added to thereaction mixture. The solution was subjected to 27PCR cycles (94°C for 30 sec, 60°C for 30 sec). The PCRproducts were electrophoresed on a 12.5% nondena-turing polyacrylamide gel.

The presence of telomerase activity was shown by6-bp ladders starting from 50 bp. A 36-bp internal con-trol band to monitor PCR efficiency was visible in ev-ery lane. Since telomerase is a ribonucleoprotein, theextracts that produced 6-bp ladders were tested forsensitivity to RNase pretreatment. Primer-dimer con-tamination was confirmed by omitting cell extractsfrom the TRAP reaction mixtures. The radioactivity ofeach band of 6-bp ladders was measured using Image-Quant software (Molecular Dynamics, Sunnyvale,CA). The relative telomerase activity was expressed asa percentage of that in LNCaP cells cultured in RPMI-1640 with 10% FCS.

Non-PCR-Based Telomerase Assay

To further confirm the specificity of telomerase andexclude the influence of PCR inhibitors in cell extracts,we used a non-PCR-based telomerase assay withTTAGGG-repeat primer, as previously reported [10–12]. The cell extract was adjusted to the concentrationof 3 mg/ml of protein, and 4 ml of the extract wereincubated at 37°C for 60 min in 20 ml of reaction so-lution containing 1 mM 58-biotinylated (TTAGGG)3

primer (Genosys Biotechnologies, Inc., Woodland,TX), 1.5 mM [a-32P]dGTP, 2 mM dATP, 2 mM dTTP, 50mM potassium acetate, 50 mM Tris acetate (pH 8.5), 1mM MgCl2, 5 mM b-mercaptoethanol, and 1 mM sper-midine. Telomerase reaction was terminated by theaddition of streptavidin-coated magnetic Dynabeads(Dynal A.S., Oslo, Norway). Streptavidin-coated Dy-nabeads bound selectively to the 58-biotinylatedtelomerase product. Using a magnet, the telomerase-product complexed with beads was collected from thereaction solution, and washed several times to elimi-nate the remaining [a-32P]dGTP. The telomerase prod-ucts were isolated with guanidine-HCl, and then pre-cipitated with ethanol. The telomerase product waselectrophoresed on an 8% polyacrylamide denaturinggel.

162 Soda et al.

Cell Cycle Experiments

LNCaP cells were cultured in RPMI-1640 with 10%CCS for 7 days. Afterwards, cells were cultured withthe supplement of 10−6–10−10 M DHT in the presenceor absence of 10 mM hydroxyurea (Sigma ChemicalCo.) for 2 days, because the doubling time of LNCaPcells is approximately 2 days. For cell cycle analysis,cells were stained with a technique described previ-ously [13]. Cells were mixed with 50 mg/ml prop-idium iodide in a hypotonic sodium citrate solutionwith 0.3% NP40 and 1.0 mg/ml RNase at 106 cells/ml.All samples were analyzed with an EPICS ELITE flowcytometer (Coulter Cytometry, Miami, FL).

RESULTS

A TRAP assay detected telomerase activity in LNCaPcells with a minimum concentration of 0.01 mg/ml ofprotein, and showed semiquantitation up to 1 mg/ml:8.0% at 0.01 mg/ml, 100% at 0.1 mg/ml, and 147.4% at1.0 mg/ml of protein (Fig. 1). RNase pretreatment ab-rogated 6-bp ladders of telomerase activity, but not a36-bp internal control band. A primer-dimer contami-nation control (lysis buffer lane in Fig. 1) revealed noproduct except for a 36-bp internal control band. Theintensity of a 36-bp internal control band was some-times decreased when the cell extracts contained ex-cessively high telomerase activity due to semicompeti-tive amplification of the TRAP products and internalcontrol.

Androgen-sensitive LNCaP cells showed arrestedcell growth in RPMI-1640 with 10% CCS, yet the cellsretained their viability (Table I). The addition of DHTrestored cell growth, and peak cell growth was ob-served at 10−8 M DHT. In a TRAP assay, LNCaP cellsgrown in 10% FCS had high telomerase activity, and10% CCS reduced the activity to 2.8% (Fig. 2). Theaddition of DHT increased telomerase activity: 27.8%at 10−10 M, 36.8% at 10−8 M, and 95.7% at 10−6 M DHT.The effect of DHT on the cell cycle was also assessedby flow cytometry (Tables II and III). Ten percent CCSaccumulated LNCaP cells into the G0/G1 phase of thecell cycle, and the addition of DHT increased the pro-portion of S-phase cells. The highest percentage of S-phase cells was observed at 10−8 M DHT (Table II).Hydroxyurea blocked S-phase progression by DHT,but not the induction of telomerase (Table III). In thecell cycle analysis, LNCaP cells with less than 2 NDNA indicating the apoptotic cells was undetectablein the presence or absence of DHT and hydroxyurea(data not shown). After 1 month of androgen-depriva-tion, the appearance of LNCaP cells was changed tosmall-sized cell bodies with long, branched processes.Growth of these cells was still arrested, and the cells

did not show telomerase activity. The addition of DHTinduced LNCaP cells to restore morphology, to slowcell growth, and to reactivate telomerase (data notshown).

Androgen-independent TSU-Pr1 and DU145 cellsshowed reduced cell growth in RPMI-1640 with 10%CCS, and the addition of DHT did not restore growth(Table I). Charcoal-stripped factors besides androgenare considered to be important for the growth of bothcell lines. TSU-Pr1 and DU145 cells showed hightelomerase activity (153.7% and 125.9% of the LNCaPactivity, respectively) (Fig. 3). Ten percent CCS and

Fig. 1. A telomeric repeat amplification protocol (TRAP) assaydetected telomerase activity in LNCaP cells with a minimum con-centration of 0.01 µg/µl of protein, and showed semiquantitationup to 1 µg/µl. An internal control band to monitor the efficiencyof the polymerase chain reaction is visible in every lane. RNasepretreatment abrogated 6-bp ladders of telomerase activity butnot an internal control band.

Androgen and Telomerase in Prostate 163

the addition of DHT did not influence telomerase ac-tivity of both cell lines.

In a TRAP assay (Fig. 4), normal prostate cellsshowed low intensity of 6-bp ladders (8.1% of LNCaPactivity), which was abrogated by RNase pretreat-ment. In a non-PCR-based telomerase assay (Fig. 5),the extract of normal prostate cells added one telo-meric repeat to the (TTAGGG)3 primer, whereasLNCaP,TSU-Pr1, and DU145 showed several exten-sions to the primer. The 6-bp ladders in normal pros-tate cells resulted from telomerase activity, and lowactivity was not due to the presence of PCR inhibitorsin cell extract. The addition of DHT did not modifycell growth (Table I) and telomerase activity in normalprostate cells (Fig. 4).

DISCUSSION

The present study demonstrated that prostate can-cer cell lines had high telomerase activity, whereasnormal prostate cells with a limited life span had lowtelomerase activity. In addition, we showed that an-drogen regulated telomerase activity in androgen-sensitive LNCaP cells, but not in two androgen-independent cancer cell lines or normal cells inculture.

Three prostate cancer cell lines used in the presentstudy had high telomerase activity. A series of ourprior studies showed short telomeres of LNCaP andDU145 cells and long telomeres of TSU-Pr1 cells[11,12]. LNCaP and DU145 cells each have a specifictelomere length, which is stable regardless of cell di-vision [14]. These findings suggest that a regulatorymechanism exists for limiting telomere elongation byhigh telomerase activity. A recent study showed thattelomeric repeat binding factor 1 (TRF1) may be in-volved in this regulation [15]. Telomere length is regu-lated in a complex fashion, and further studies are

TABLE I. Effects of Dihydrotestosterone on Cell Proliferation*

LNCaP TSU-Pr1 DU145 Normal cells

RPMI + FCS 17.0 ± 0.1 58.9 ± 1.0 26.5 ± 2.0 Not doneRPMI + CCS 5.7 ± 0.2 39.0 ± 0.5 16.7 ± 0.2 Not donePrEBM Not done Not done Not done 9.3 ± 0.410−10 M DHT 8.9 ± 0.1 36.8 ± 0.5 15.2 ± 0.4 9.5 ± 0.610−8 M DHT 12.3 ± 0.6 34.9 ± 0.2 15.7 ± 0.2 10.1 ± 0.510−6 M DHT 8.8 ± 0.4 36.4 ± 0.5 15.7 ± 0.2 9.6 ± 0.4

*Prostate cancer LNCaP, TSU-Pr1, and DU145 cells at 2 × 104/well and normal cells at 3 ×103/well are plated in RPMI-1640 with 10% fetal calf serum (FCS) and serum-free conditionedmedia (PrEBM), respectively, on day 0. The media are exchanged on day 1. 5a-dihydrotestosterone (DHT) is added to RPMI-1640 with 10% charcoal-stripped calf serum (CCS)and PrEBM for prostate cancer cells and normal cells, respectively. The numbers of viable cells(× 104 cells, mean ± SD) are measured in triplicate on day 5.

Fig. 2. Telomerase activity was detected with a TRAP assay inLNCaP cells cultured in RPMI-1640 with 10% fetal calf serum(FCS). Ten percent charcoal-stripped calf serum (CCS) reducedtelomerase activity, and the addition of 5a-dihydrotestosterone(DHT) restored telomerase activity.

164 Soda et al.

necessary to clarify the relationship of telomere lengthand telomerase activity.

LNCaP cells showed growth arrest and a loss oftelomerase activity in androgen-depleted serum. Thegrowth-arrested cells mainly accumulated into theG1/G0 phase of the cell cycle without any change intheir viability or the proportion of apoptotic cells.Typical ultrastructural changes in apoptosis and thefragmented DNA ladder are not observed in LNCaPcells treated with CCS [16]. The reduced telomeraseactivity caused by androgen-depletion probably re-sults from the cells becoming quiescent rather thanfrom apoptosis.

DHT induced telomerase activity as well as cellgrowth in LNCaP cells. DHT induced telomerase ac-tivity during the G1 phase of the cell cycle becausehydroxyurea, an inhibitor of S-phase progression, didnot block the induction of telomerase. Several studiesalso showed that telomerase is induced when cells exitthe G0 phase of the cell cycle [17]. Recently, telomeraseactivity was demonstrated to have a close correlationwith the regulatory cell-cycle proteins of G1 phase,regardless of cellular quiescence [18,19]. The prolifera-tion of LNCaP cells is mediated, at least in part,through telomerase activation at the G1 phase byDHT.

LNCaP cells exhibited a bell-shaped growth in re-sponse to DHT, but the highest telomerase activity didnot correspond to the peak of cell growth. The per-centage of S-phase cells was proportional to cellgrowth, but not telomerase activity. These findingswere compatible with telomerase reactivation at G1phase. Several investigators have also reported thatthe expression of proliferative cell nuclear antigen(PCNA) and prostate-specific antigen (PSA) in LNCaPcells is increased with concentrations of DHT [20,21].In contrast, DHT induces the production of transform-

ing growth factor-b (TGF-b), which is known as anegative growth regulator, in a concentration-dependent manner, and the growth inhibition seen athigh concentrations of DHT is abolished after the ac-tion of TGF-b is blocked [22]. High concentrations ofDHT may activate different pathways to stimulate andinhibit cell proliferation, respectively.

Normal prostate cells with a limited life span inculture had low telomerase activity. Although telom-erase activity is not usually detected in unculturednormal prostate tissue, low telomerase activity isfound in proliferating normal prostate cells in culture[4]. Also, telomerase activity is present in collagenase-dispersed cells from normal rat prostate, but thewhole-tissue extract does not express telomerase ac-tivity due to the PCR inhibitors in the prostate secre-tion [5]. Normal prostate glands are composed mainlyof androgen-independent basal cells and androgen-dependent luminal secretory cells. The basal cells arethe proliferative compartment, and the secretory lu-minal cells are the terminally differentiated compart-ment [23]. In a castrated rat, the basal cells becomeprominent and show telomerase activity, and this ac-tivity disappears when the secretory cells regrow fol-lowing the administration of testosterone [24]. Lowtelomerase activity in normal prostate epithelium isconsidered to result from the basal cells.

In the present study, DHT did not influence telom-erase activity in cultured normal prostate cells, unlikethe results in castrated rats treated with testosterone[24]. Recently, similar findings regarding telomeraseactivity and sex steroid hormone were reported [7].Endometrium treated with antiestrogen drugs andpostmenopausal endometrium show decreased telo-merase activity in vivo, whereas estrogen treatmentdoes not modify telomerase activity in cultured endo-metrium cells [7]. It is difficult to observe the differ-entiation into the prostate secretory cells from thebasal cells in conventional culture [25]. The discrep-ancy may result from cell differentiation in in vitroand in vivo systems.

Although prostate cancers respond to initial andro-gen-ablation therapy, most tumors eventually prog-ress to the androgen-independent stage. The presentstudy did not completely reflect the clinical situation,since the period of androgen-deprivation for LNCaPcells was short. A recent study using a mouse modelsuggested that androgen-independent clones outgrowfrom androgen-dependent tumors at least 6 monthsafter androgen-deprivation [26]. Androgen-indepen-dent tumors are known to overexpress genes, such asthe Bcl-2 anti-apoptotic gene, observed normally inprostate basal cells [27]. The present study revealedthat normal prostate cells, including basal cells, hadandrogen-independent telomerase activity. In addi-

TABLE II. Effects of Dihydrotesterone on Cell Cycle inLNCaP Cells*

Medium

Cell cycle (%)

G0/G1 S G2/M

FCS 76.7 10.6 12.7CCS 88.2 3.0 8.8CCS + 10−10 M DHT 85.0 3.9 11.1CCS + 10−8 M DHT 83.9 5.7 10.4CCS + 10−6 M DHT 89.0 2.8 8.2

*LNCaP cells are cultured in RPMI-1640 with 10% charcoal-stripped calf serum (CCS) for 7 days. Then, cells are cultured inthe supplement of 10−6–10−10 M 5a-dihydrotestosterone (DHT)for 2 days. As a control, LNCaP cells are cultured in RPMI-1640with 10% fetal calf serum (FCS).

Androgen and Telomerase in Prostate 165

tion, LNCaP cells retained the ability to reactivatetelomerase in response to DHT after 1 month of an-drogen-deprivation, although androgen-independentsublines were not established. Further studies oftelomerase in prostate basal cells and androgen-

independent sublines derived from androgen-dependent cancers may provide new insights into themechanisms of the progression to androgen-indepen-dent cancers.

In conclusion, androgen-sensitive LNCaP cells had

TABLE III. Effects of Hydroxyurea on Androgen-Induced Telomerase Activity inLNCaP Cells*

Medium

Cell cycle (%)Telomeraseactivity (%)G0/G1 S G2/M

FCS 77.9 13.8 8.4 100.0CCS 91.1 2.1 6.8 2.3CCS + 10−8 M DHT 84.7 6.5 8.8 53.4CCS + 10−8 M DHT + HU 93.4 1.6 5.0 61.2

*LNCaP cells are cultured in RPMI-1640 with 10% charcoal-stripped calf serum (CCS) for 7 days.Afterwards, cells are cultured in the supplement of 10−8 M 5a-dihydrotestosterone (DHT) withor without 10 mM hydroxyurea (HU) for 2 days. As a control, LNCaP cells are cultured inRPMI-1640 with 10% fetal calf serum (FCS).

Fig. 3. Telomerase activity was detected with a TRAP assay inTSU-Pr1 and DU145 cells cultured in RPMI-1640 with 10% fetalcalf serum (FCS). Ten percent charcoal-stripped calf serum (CCS)and the addition of 5a-dihydrotestosterone (DHT) did not modifytelomerase activity in both cells.

Fig. 4. In a TRAP assay, cultured normal cells showed extremelylow telomerase activity, which was abrogated by RNase pretreat-ment. The addition of 5a-dihydrotestosterone (DHT) did notmodify telomerase activity.

166 Soda et al.

high telomerase activity, which DHT induced at theG1 phase of the cell cycle. Short or prolonged andro-gen-deprivation reduced telomerase activity in LNCaPcells. Normal prostate cells, including basal cells,showed low levels of androgen-independent telomer-ase activity, whereas TSU-Pr1 and DU145 cells exhib-ited high levels of androgen-independent telomeraseactivity. These findings suggest that the regulatorymechanism of telomerase varies during the progres-sion of prostate cancers. Our in vitro model shouldencourage further study of the mechanisms of prostatecarcinogenesis and androgen-independent progres-sion.

ACKNOWLEDGMENTS

The authors thank Gene Stephens, Peggy Durack,and Gregory Hannibal for their excellent assistance.E.R. was supported by a grant from Association pourla Recherche contre le Cancer in Villejuif, France. S.S.was supported by a training grant from the NationalInstitutes of Health.

REFERENCES

1. Sharma S, Raymond E, Soda H, Sun D, Hilsenbeck SG, SharmaA, Izbicka E, Windle B, Von Hoff DD. Preclinical and clinicalstrategies for development of telomerase and telomere inhibi-tors. Ann Oncol 1997;8:1063–1074.

2. Sommerfeld H-J, Meeker AK, Piatyszek MA, Bova GS, Shay JW,Coffey DS. Telomerase activity: a prevalent marker of malignanthuman prostate tissue. Cancer Res 1996;56:218–222.

3. Lin Y, Uemura H, Fujinami K, Hosaka M, Harada M, Kubota Y.Telomerase activity in primary prostate cancer. J Urol 1997;157:1161–1165.

4. Belair CD, Yeager TR, Lopez PM, Reznikoff CA. Telomeraseactivity: a biomarker of cell proliferation, not malignant trans-formation. Proc Natl Acad Sci USA 1997;94:13677–13682.

5. Banerjee PP, Banerjee S, Zirkin BR, Brown TR. Lobe-specifictelomerase activity in the intact adult brown Norway rat pros-tate and its regional distribution within the prostatic ducts. En-docrinology 1998;139:513–519.

6. Tsao J-L, Zhao Y, Lukas J, Yang X, Shah A, Press M, Shibata D.Telomerase activity in normal and neoplastic breast. Clin Can-cer Res 1997;3:627–631.

7. Tanaka M, Kyo S, Takakura M, Kanaya T, Sagawa T, YamashitaK, Okada Y, Hiyama E, Inoue M. Expression of telomerase ac-tivity in human endometrium is localized to epithelial glandularcells and regulated in a menstrual phase-dependent mannercorrelated with cell proliferation. Am J Pathol 1998;153:1985–1991.

8. Yang Y, Hao J, Liu X, Dalkin B, Nagle RB. Differential expres-sion of cytokeratin mRNA and protein in normal prostate, pros-tatic intraepithelial neoplasia, and invasive carcinoma. Am JPathol 1997;150:693–704.

9. Holt SE, Norton JC, Wright WE, Shay JW. Comparison of thetelomeric repeat amplification protocol (TRAP) to the new TRA-Peze telomerase detection kit. Methods Cell Sci 1996;18:237–248.

10. Sun D, Lopez-Guajardo CC, Quada J, Hurley LH, Von Hoff DD.Regulation of catalytic activity and processivity of humantelomerase. Biochemistry 1999;38:4037–4044.

11. Izbicka E, Wheelhouse RT, Raymond E, Davidson KK,Lawrence RA, Sun D, Windle BE, Hurley LH, Von Hoff DD.Effects of cationic porphyrins as G-quadruplex interactiveagents in human tumor cells. Cancer Res 1999;59:639–644.

12. Raymond E, Sun D, Izbicka E, Mangold G, Silvas E, Windle B,Sharma S, Soda H, Lawrence R, Davidson K, Von Hoff DD. Ahuman breast cancer model for the study of telomerase inhibi-tors based on a new biotinylated-primer extension assay. Br JCancer 1999;80:1332–1341.

13. Sharma S, Raymond E, Soda H, Izbicka E, Davidson K,Lawrence R, Von Hoff DD. Dimethyl sulfoxide (DMSO) causesa reversible inhibition of telomerase activity in a Burkitt lym-phoma cell line. Leuk Res 1998;22:663–670.

14. Kageyama Y, Kamata S, Yonese J, Oshima H. Telomere length

Fig. 5. In a non-PCR-based telomerase assay, the extract ofproliferating normal prostate cells in culture added one telomericrepeat to (TTAGGG)3 primer, whereas LNCaP, TSU-Pr1, andDU145 cells showed several extensions to the primer.

Androgen and Telomerase in Prostate 167

and telomerase activity in bladder and prostate cancer cell lines.Int J Urol 1997;4:407–410.

15. Van Steensel B, De Lange T. Control of telomere length by thehuman telomeric protein TRF1. Nature 1997;385:740–743.

16. Saeed B, Zhang H, Ng S-C. Apoptotic program is initiated butnot completed in LNCaP cells in response to growth in charcoal-stripped media. Prostate 1997;31:145–152.

17. Holt SE, Aisner DL, Shay JW, Wright WE. Lack of cell cycleregulation of telomerase activity in human cells. Proc Natl AcadSci USA 1997;94:10687–10692.

18. Landberg G, Nielsen NH, Nilsson P, Emdin SO, Cajander J,Roos G. Telomerase activity is associated with cell cycle deregu-lation in human breast cancer. Cancer Res 1997;57:549–554.

19. Balasubramanian S, Kim K-H, Ahmad N, Mukhtar H. Activa-tion of telomerase and its association with G1-phase of the cellcycle during UVB-induced skin tumorigenesis in SKH-1 hairlessmouse. Oncogene 1999;18:1297–1302.

20. Perry JE, Tindall DJ. Androgens regulate the expression of pro-liferating cell nuclear antigen posttranscriptionally in the hu-man prostate cancer cell line, LNCaP. Cancer Res 1996;56:1539–1544.

21. Lee C, Sutkowski DM, Sensibar JA, Zelner D, Kim I, Amsel I,Shaw N, Prins GS, Kozlowski JM. Regulation of proliferationand production of prostate-specific antigen in androgen-

sensitive prostatic cancer cells, LNCaP, by dihydrotestosterone.Endocrinology 1995;136:796–803.

22. Kim IY, Kim J-H, Zelner DJ, Ahn H-J, Sensibar JA, Lee C. Trans-forming growth factor-b1 is a mediator of androgen-regulatedgrowth arrest in an androgen-responsive prostatic cancer cellline, LNCaP. Endocrinology 1996;137:991–999.

23. Bonkhoff H, Remberger K. Differentiation pathways and histo-genetic aspects of normal and abnormal prostate growth: a stemcell model. Prostate 1996;28:98–106.

24. Meeker AK, Sommerfeld HJ, Coffey DS. Telomerase is activatedin the prostate and seminal vesicles of the castrated rat. Endo-crinology 1996;137:5743–5746.

25. Fong C-J, Sherwood ER, Sutkowski DM, Abu-Jawdeh GM, Yo-koo H, Bauer KD, Kozlowski JM, Lee C. Reconstituted basementmembrane promotes morphological and functional differentia-tion of primary human prostatic epithelial cells. Prostate 1991;19:221–235.

26. Craft N, Chhor C, Tran C, Belldegrun A, DeKernion J, Witte ON,Said J, Reiter RE, Sawyers CL. Evidence for clonal outgrowth ofandrogen-independent prostate cancer cells from androgen-dependent tumors through a two-step process. Cancer Res 1999;59:5030–5036.

27. Bui M, Reiter RE. Stem cell genes in androgen-independentprostate cancer. Cancer Metastasis Rev 1999;17:391–399.

168 Soda et al.