Expression of novel genes linked to the androgen-induced, proliferative shutoff in prostate cancer cells

 Music & Video

of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Expression of novel genes linked to the androgen-induced, proliferative shutoff in prostate cancer cells
  Pergamon J. Steroid Biochem. Molec. Biol. Vol. 63, No. 4-6, pp. 211-218, 1997 © 1997 Published by Elsevier Science Ltd. All rights reserved Printed in Great Britain PII: S0960-0760(97)00122-2 0960-0760/97 S19.00 + o.00 Expression of Novel Genes Linked to the Androgen-induced, Proliferative Shutoff in Prostate Cancer Ceils Peter Geck, Jozsef Szelei, Jesus Jimenez, Tien-Min Lin, Carlos Sonnenschein and Ana M. Soto* Department of Anatomy and Cell Biology, Tufts University School of Medicine, Boston, MA 0211 I, U.S.A. Androgens control cell numbers in the prostate through three separate pathways: (a) inhibition of cell death, (b) induction of cell proliferation (Step-l) and (c) inhibition of cell proliferation (Step-2, proliferative shutoff). The mechanisms underlying these phenomena are incompletely understood. The human prostate carcinoma LNCaP variants express these pathways as follows: LNCaP-FGC express both steps, LNCaP-LNO expresses Step-2, LNCaP-TAC expresses Step-l, and LNCaP-TJA cells express neither step. These cells facilitated the search for mediators of the androgen-induced proliferative shutoff pathway. Androgen exposure for 24 h or longer induced an irreversible prolif- erative shutoff in LNCaP-FGC cells. The Wang and Brown approach for identifying differentially expressed mRNAs was used to search for mediators of Step-2. Ten unique inserts were identified and from those ten, three genes were further studied. The basal expression of these genes in shut- off-negative variants was not affected by androgen exposure. They were induced by androgens in shutoff-positive LNCaP variants and the androgen receptor-transfected, shutoff-positive, MCF7- AR1 cells. These genes were induced only in the range of androgen concentrations that elicited the shutoff response. Time course analysis showed that their induction precedes the commitment point by 12-I8 h. In addition, they were expressed in the normal prostate during proliferative shutoff. These features suggest that the candidate genes have a role in the regulation cascade for prolifera- tive shutoff. © 1997 Published by Elsevier Science Ltd. All rights reserved J. Steroid Biochem. Molec. Biol., Vol. 63, No. 4-6, pp. 211-218, 1997 INTRODUCTION Cell numbers in the prostate are regulated by andro- gens through three distinct pathways: (a) inhibition of cell death (apoptosis) [1], (b) induction of cell pro- liferation (Step-l), and (c) inhibition of cell prolifer- ation (proliferative shutoff, Step-2) [2,3]. Two of these mechanisms [(a) and (b)] have been used for therapeutic manipulation in prostate cancer by andro- gen ablation [4, 5]; the potential therapeutic use of the proliferative shutoff mechanism (c) remains unex- plored. The complexity of interactions among various cell types in animals hinders the understanding of these androgen-mediated phenomena. Established *Correspondence to A. M. Soto. Tel: 636 6954; Fax: 636 6536; e-mail: Received 12 May 1997; ~ccepted 21 Jul. 1997. 21 hormone target cell lines together with recombinant DNA technology now offer an opportunity for testing hypotheses regarding the control of cell proliferation, without being affected by variables present in intact animals. The human prostate LNCaP-FGC cell line has been extensively used for this purpose. These cells are inhibited from proliferating by sex-steroid stripped human serum (CDHuS); low levels of andro- gens reverse this effect (Step-l) and at higher concen- trations androgens induce an irreversible proliferative shutoff (Step-2) [3,6]. LNCaP variants expressing only one of these two androgen-controlled pathways were isolated. From LNCaP-FGC cells, the LNCaP- TAC variant that expresses Step-1 only was selected [7]. The LNCaP-LNO cells proliferate maxi- mally in the presence of CDHuS, while responding to androgens by undergoing G0/G~ arrest. Finally, the 1  212 Peter Geck et al. LNCaP-TJA variant is impervious to the inhibitory effect of both CD serum and androgens [6]. By using a novel differential subtractive amplifica- tion procedure [8], we are now able to describe candi- date genes responsible for the androgen-mediated proliferative shutoff. MATERIALS AND METHODS Cell lines and tissue culture protocols LNCaP-FGC cells and LNCaP-LNO cells derived from a metastatic lymph node of a patient with pros- tate adenocarcinoma were generously supplied by Dr. Julius Horoszewicz from the State University of New York at Buffalo [9]. The LNCaP-LNO variant was continuously maintained in 5% CD stripped fetal bovine serum (FBS) (HYCLONE, Logan, UT) sup- plemented to Dulbecco's modified Eagle medium (DME) (GIBCO-BRL, Gaithersburg, MD) [10]. The LNCaP-FGC, TAC and TJA variants were routinely grown in 5% FBS [6,7]. The MCF7-AR1 cell line resulted from transfection of the androgen receptor into the MCF7 cell line; it was propagated routinely in 5% FBS[ll]. Rat prostate experiments Sprague-Dawley male rats (200g) were used (Taconic, Germantown, NY). Ventral lobes of intact, castrated and castrated rats treated with testosterone propionate for 3 and 7 days [2] were dissected and frozen on dry ice and used to prepare RNA (see below). Cell proliferation analysis The time-course of androgen exposure required for commitment to express the shutoff phenotype was studied in LNCaP-FGC cells. 40 000 cells/well were seeded in 12-well plates (Costar, Cambridge, MA) in 5% FBS. Medium was changed after 48h to 5% CDHuS in phenol red-free DME. Cells were treated with 30nM testosterone (T) for 0 to 120h. Subsequently, the medium was removed, cells were washed twice with DME and fresh 5% CDHuS con- taining 30 nM estradiol (E2) was supplied to allow proliferation of cells not committed to the androgen- induced shutoff. Cells were lysed, and the nuclei were counted by a Coulter Counter Model ZM (Coulter, Hialeah, FL) [6]. The Wang-Brown differential amplification This procedure first amplifies short fragments of cDNAs. Then, successive subtractions and amplifica- tions between the two cDNAs result in differential sequence pools in both directions [8] (Fig. 1). LNCaP-FGC cells expressing androgen-induced pro- liferative shutoff were harvested after 24 h of 30 nM R1881 treatment. R1881 is a synthetic, non-metab- Shutoff* cDNA R) RsaI AluI digestion adaptor ligation PCR > +cDNA Long hybridization l +I cDNA Short hybridization PCR *2 cDNA repeat procedure cDNA BD shutoff- negative, biotinyl.) CD) cDNA BD CD) Shutoff* > probe subtracted library R) shutoff*) Cloning into pCRII vector Fig. 1. Subtractive amplification of androgen-induced sequences. Shutoff ~ RNA was prepared from LNCaP-FGC cells treated with 30nM R1881 for 24h; the R (for R1881- treated) sequences srcinate from this preparation. 'Driver' cDNA was prepared from LNCaP-FGC cells grown in CDHuS, a condition in which the proliferative shutoff is not expressed; the shutoff-negative cDNA sequences are srci- nated from this pool, called CD (for CDHuS-treated). A similar protocol was performed to identify sequences that were selectively expressed in cells made quiescent by ex- posure to CDHuS; namely, shutoff-negative cDNAs from LNCaP-FGC cells grown in CDHuS were subtracted with cDNAs derived from shutoff +, R1881-treated cells. olized androgen (methyltrienolone, RousselI-UCLAF, Romainville). A shutoff-negative control was obtained by using LNCaP-FGC cells harvested after 3 days of CDHuS treatment. Total RNA was prepared by the acidic guanidinium-thiocyanate method [12] and poly-A + RNA was purified by the FastTrack kit (Invitrogen, San Diego, CA). Double-stranded cDNA pools from both RNA preparations were synthesized using the Copy Kit (Invitrogen), with oligo-dT prim- ing. A 2 #g aliquot of cDNA was digested by AluI and RsaI enzymes; specific adapters with EcoRI and PCR primer sites were ligated to the ends, and the constructs were PCR amplified (GeneAmp Kit, Perkin Elmer, Foster City, CA). 50 /~g aliquots of amplified DNA were digested with EcoRI, photobioti- nylated (driver cDNA) and hybridized to a 2 #g ali- quot of non-biotinylated cDNA from the other pool. Long and short hybridizations were performed to increase subtraction efficiency for low and high abun- dance transcripts [8]. After 3 cycles, the final ampli- fied pools of the androgen-induced shutoff (R) and  Gene Expression in Prostate Cancer Arrest 213 serum arrested (CD) sequences were digested with EcoRI, and ligated into the BlueScript SK vector (Stratagene, La Jolla, CA) for transformation into E. coli (OneShot strain, Invitrogen). Colonies were trans- ferred to nylon filters and differentially screened using radiolabelled subtracted R and CD pools as probes. Multiple cross-hybridizations of dot-blot filters were used to investigate relatedness among cloned inserts. DNA sequence analysis PCR sequencing reactions were performed using the dsDNA Sequencing System (Life Technologies, Gaithersburg, MD). The DNA sequences were searched for homology to known sequences using the FASTA (Genetics Computer Group, Madison, WI) and BLAST (National Center for Biotechnology Information, Bethesda, MD) programs. Northern blot analysis Total RNA was prepared using the RNeasy kit (Qiagen, Chatsworth, CA). RNA samples (20 g) were run in 1% agarose-formaldehyde gels and trans- ferred to nylon membranes using the Turbo-Blot System (Schleicher and Schuell, Keene, NH). The PCR amplified inserts were labelled by the Random Primed DNA Labeling Kit (Boehringer-Mannheim, Indianapolis, IN) and used as probes. Hybridizations were performed overnight at 68°C in ExpressHyb sol- ution (Clontech, Palo Alto, CA). Membranes were analyzed by PhosphorImager (Molecular Dynamics, Stanton, CA) using the ImageQuant program. Northern filters containing poly-A ÷ RNA preparations from an array of normal human tissues were pur- chased from Clontech. Hybridizations of rat Northern blots were at 65°C and the washing temperature was 48°C. RNA loading wets normalized to either glyceral- dehyde 3-phosphate dehydrogenase (GAPDH) or fl- actin (Clontech) signals in the same blot. Logarithmic phase RT-f~CR This procedure was used to measure the expression of low copy number transcripts [13-15]. 2 g aliquots of total RNA were used as templates for cDNA syn- thesis with gene specific primers (cDNA Synthesis Modul, Invitrogen). GAPDH levels were measured first to normalize the: assay for subsequent exper- iments. PCR reactions were performed with the GeneAmp kit. Primers were labelled with 32p T-ATP. From the 50 1 reaction volume, 7 1 aliquots were withdrawn between the 15th and 35th cycles at vari- able points, and resolved by polyacrylamide gel elec- trophoresis. Patterns were analyzed by Phosphorimager. Optical densities of the bands were integrated and plotted against the amplification factor. Only data within the logarithmic phase were used for further analyses. RESULTS Time-course of induction of the proliferative shutoff LNCaP-FGC cells were exposed to T for 0 to 120 h to determine the minimal duration of androgen exposure required for triggering the proliferative shut- off. Figure 2 shows that exposure for 24 h or longer resulted in the irreversible expression of the prolifera- tive shutoff. Consequently, the search for shutoff can- didate genes was done using RNA from cells exposed to androgen for 24 h. Characterization of subtracted sequences 230 and 250 recombinants were selected and plated from the CD and R preparations, respectively. Using the labeled CD- and R-subtracted, PCR-amplified DNA master mixes as probes, double hybridizations revealed 11 and 14 clones present exclusively in the CD and R clone sets. Ten unique inserts were ident- ified by multiple cross-hybridizations and sequenced. Five of the sequences were found to match sequences in GenBank (Table 1). Differential expression analysis Expression patterns of the new genes in androgen- induced shutoff and CD serum-arrested cells are shown in Fig. 3. Panels A and B display genes readily detectable by Northern hybridizations. Panel A shows genes with higher than 3-fold induction, including the positive control for androgen induction, i.e., the mRNA for Prostate Specific Antigen (PSA). Genes 1.3 and 9 were induced 10-fold and 3.9-fold, re- ~00 400 30o '-~ ~00 % ] 100 N I I 0.1 ~ 10 100 200 'I I.~STO,~I ERONE EXPOSURE (hours) Fig. 2. Time-course of androgen exposure required for com- mitment of LNCaP-FGC cells to the proliferative shutoff re- sponse. Cells were exposed to 300 nM testosterone in 10% CDHuS at time = 0 for the intervals indicated by the abscissa. The medium was then removed, wells were rinsed to remove androgens, and fresh medium containing 30 nM estradiol was added to elicit proliferation of cells not com- mltted to the shutoff. Cells were harvested on the 6th day. The ordinate represents final cell yields on the 6th day; the abscissa indicates exposure to 300 nM testosterone. The bot- tom box on the left (stippled) represents the cell yield of con- trois exposed to 10% CDHuS alone for 6 days. The open box at the top indicates maximal cell yield with estradioL  214 Peter Geck et al. Table 1. Sequences isolated by the Wang-Brown method Clone Insert (bp) Homology mRNA (Kb) Up-regulated mRNAs expressed in proliferative shutoff AS1 (9) 190 -- 3.9 AS2 (11) 188 99.2% (X83543) 7.5; 6.5; 5.5; 4.4 AS3 (27) 262 98.0% (U50533) 8.0; 5.5 AS4 (1.3) 109 70% (M25680) 6.0; 3.8; 2.5 AS5 (3) 160 -- 3.0; 1.3; 0.3 (21) 90 96% (H16580) 4.1 (31) 155 N/A (38) 309 -- N/A Up-regulated mRNAs in CDHuS-arrested cells SA1 (7) 119 -- N/A SA3 (2) 293 97.5% (X06704) 2.7 Numbers in parenthesis are the srcinal clone numbers for reference. The Insert column indicates the cloned sequence tag sizes in base pairs. The mRNA column shows the transcript sizes (in kilobases) as determined by Northern analysis. Accession X83543 is the human APXL (formerly ocular albinism-associated) gene. Accession M25680 is the rat catalase. Accession : H16580 is an expressed sequence tag from a human infant brain library. Accession X06704 is the human trk-2h oncogene mRNA. For accession U50533, see text. spectively. Induction of genes in panel B was less manifest; for 21, 31 and 38 the difference was 2.5-, 1.5- and 2.0-fold, respectively. The abundance of genes in panel C was considerably lower; it was about two orders of magnitude less than that of GAPDH. Quantitative analysis of their expression was performed by logarithmic phase RT-PCR. The genes for 3, 11 and 27 were upregulated 9.2-, 4- and 7- fold, respectively. We also found two genes that were turned off during androgen-induced shutoff while being expressed in CDHuS arrested cells. Analysis of these genes will be reported elsewhere. PSA #1.3 #9 A B C SA AS SA AS / #21 #31 #38 SA AS 3 11 ~ #27 GAPD Fig. 3. Analysis of differential expression. Panel (A), Northern analysis of medium abundance genes with higher than 3-fold induction. Panel (B), Northern analysis of reed- Jura abundance genes with lower than 3-fold induction. Panel (C), logarithmic phase quantitative RT-PCR analysis of low abundance genes. SA and AS indicate RNA preparations from CD-serum arrested cells, or from cells expressing androgen-induced shutoff, respectively. Probes are indicated at the left of the blot. PSA, Prostate Specific Antigen. GAPD, glyceraldehyde phosphate dehydrogenase (normaliza- tion control). Time course of the expression of the shutoff candidate sequences Of the 10 differentially expressed sequences, only 9, 11b and 27 were analyzed in detail. The 3.9 kb mRNA for 9 (its gene is tentatively called Androgen Shutoff gene 1, AS1) showed a 2-fold increase at 6 h after exposure to 30 nM R1881, reach- ing its peak (3.9-fold) at 18 h, and was expressed con- tinuously throughout the 72 h-period of study. The 7.5 kb mRNA (and the 6.6 and 4.4 kb splicing var- iants, not shown) of the 11 sequence (AS2) showed a 2-fold induction at 2 to 4 h, gradually increasing to 4-fold after 12h; its expression was maintained throughout the 72-h observation period. The 8.0 and 5.5kb variants of the 27 mRNA (AS3) were increased about 2-fold at the 2 h time point and increased 5 to 7-fold at 18 h with an expression pat- tern similar to that of AS2 (Fig. 4). AS1 AS2 0 2 4 6 12 18 24 72 AS3 Actin Fig. 4. Northern analysis of the time-course of induction of gene expression by 30nM R1881 on LNCaP-FGC cells. Numbers atop the lanes indicate time in hours. Probes are shown on the left. The human p-actin mRNA was used as normalization control.  Gene Expression in Prostate Cancer Arrest 215 AS1 AS2 AS3 FGC LNO AR-I TAC TJA 12 12 12 12 12 iiii i i i : ~ii~i ii iii~iii :i iiii~ ~i~/i~ii~ i iiiii ~ ~i ~//~i/ ~ Androgen concentration-dependent induction of candidate shutoff genes Induction of the proliferative shutoff required R1881 concentrations of 0.3nM and higher [3]. Remarkably, induction of the three candidate shutoff genes occurred also at 0.3nM R1881 and higher. Estradiol increased the proliferation yield of LNCaP- FGC cells, but failed to elicit a shutoff response [3]. Neither of the three genes were induced by estradiol [Fig. 6(A and B), lanes 6]. No gene induction was detected in the shutoff-negative variant LNCaP-TJA [Fig. 6(B)] and LNCaP-TAC (not shown). GAP I/ Fig. 5. Comparison of the expression pattern of AS1, AS2 and AS3 in different cell variants by Northern analysis. The cell variants are indicated atop the lanes. 1, expression pat- tern in CDHuS arrest. 2, expression pattern after exposure for 24 h to 30 nM R1881 (proliferative shutoff). GAP, human glyceraldehyde phosphate dehydrogenase mRNA (normaliza- tion control). Expression of the candidate genes correlated with the ex- pression of the proliferative shutoff in LNCaP variants and MCF7-AR1 cells Figure 5 shows differential expression of the candi- date genes in the three shutoff-positive cell lines, LNCaP-FGC, LNCaP-LNO and MCFT-AR1. No differential expression was observed in the shutoff- negative variants LNCaP-TAC and LNCaP-TJA. Expression of the shutoff candidate genes in human tissues Northern blots of mRNA preparations from 16 nor- mal human tissues were probed (Fig. 7). The three candidate genes were expressed in normal prostate. The 7.5 kb isoform of AS2 was mainly expressed in the prostate and placenta. Testis and ovaries also expressed the same isoform, but to a lesser extent. Minor size variants were also observed in other tis- sues. AS1 was expressed in most tissues; the lung, colon and kidney showed low level expression. The pattern of AS3 expression was similar to that of AS1. The 5.5 and 8.0 kb mRNA isoforms were present in comparable ratio in different tissues. Expression of the shutoff-associated candidate genes in the rat prostate The expression of the rat homologs of AS1, AS2 and AS3 was studied in rat prostate under different proliferative conditions. Human probes of AS1 and AS1 As2 AS3 Actin A I 2 3 4 5 6 B 2 3 4 5 6 Fig. 6. Northern analysis of the expression of shutoff-candldate genes after exposure to R1881 or estradiol. Panel (A), LNCaP-FGC cells. Panel (B), LNCaP-TJA cells. Lane 1: 10% CDHuS; Lanes 2 to 5:R1881 (2, 30 pM; 3, 0.3 nM; 4, 3 nM; 5, 30 nM). Lane 6:30 nM estradiol. Human fl-actin mRNA was used as normaliza- tion control.
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks