Expression of human prostatic acid phosphatase gene is regulated by upstream negative and positive elements

Biochimica et Biophysica Acta 1491 (2000) 123^132 www.elsevier.com/locate/bba

Expression of human prostatic acid phosphatase gene is regulated by upstream negative and positive elements Stanislav Zelivianski a , Christopher Larson b , James Seberger a , Rodney Taylor Ming-Fong Lin a;b;c;d; * a b

b;c;d

,

Department of Biochemistry and Molecular Biology, College of Medicine, University of Nebraska Medical Center, 984525 Nebraska Medical Center, Omaha, NE 68198-4525, USA Section of Urologic Surgery, College of Medicine, University of Nebraska Medical Center, Omaha, NE 68198, USA c UNMC Eppley Cancer Center, University of Nebraska Medical Center, Omaha, NE 68198, USA d Omaha VA Hospital, Omaha, NE 68198, USA Received 11 November 1999; received in revised form 1 February 2000; accepted 3 February 2000

Abstract Human prostatic acid phosphatase (PAcP) is a prostate epithelium-specific differentiation antigen. To understand the regulation of expression of the PAcP gene, we studied the cis-regulatory elements of its promoter. A DNA fragment from 32899 to +87 base pairs (bp) of PAcP gene was fused to the chloramphenicol acetyltransferase (CAT) reporter gene and introduced into PC-3 and LNCaP human prostate cancer cells. The expression of the CAT gene driven by the PAcP promoter was assessed in transient expression assays. Sequential 5P deletions of the promoter were constructed and analyzed to reveal the positive and the negative regulatory elements that are involved in regulating the transcription of the PAcP gene. Our data showed that the proximal sequence 31305/+87 bp directs a high level of the CAT activity in both cell lines. Deletion of the region from 31305 to 3779 resulted in approximately a 10- and three-fold decrease of the PAcP promoter activity in PC-3 and LNCaP cells, respectively. Interestingly, an inverse correlation of the CAT activity with the cell growth was observed when the reporter gene was driven by the 31305/+87 fragment, but not by the 3779/+87 fragment. Two regions of transcriptional suppression were identified and located in positions from 32899 to 32583, and from 32583 to 31305 bp. Furthermore, the activity of the core promoter region from 3779 to +87 bp can be activated by a SV-40 enhancer. The results, thus, clearly demonstrate the presence of positive and negative cis-elements in the promoter region of the PAcP gene. ß 2000 Elsevier Science B.V. All rights reserved. Keywords: Promoter; Prostatic acid phosphatase; Transcription regulation; Cell growth regulation

1. Introduction The correct temporal and spatial expression of * Corresponding author. Department of Biochemistry and Molecular Biology, University of Nebraska Medical Center, 984525 Nebraska Medical Center, Omaha, NE 68198-4525, USA; Fax: +1-402-559-6650; E-mail: [email protected]

speci¢c genes governs the identity of eukaryotic cells. The core promoter elements (TATA boxes and initiator elements) determine, and reside immediately 5P to and overlap with the mRNA start points. The minimal level of transcription can be supported by the basal transcription machinery, which is assembled on the core promoter region. In addition, promoters contain recognition sites for regulatory

0167-4781 / 00 / $ ^ see front matter ß 2000 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 7 8 1 ( 0 0 ) 0 0 0 3 7 - 3

BBAEXP 93381 27-3-00

124

S. Zelivianski et al. / Biochimica et Biophysica Acta 1491 (2000) 123^132

transcription factors, which can in£uence transcription levels by either enhancing (activator) or antagonizing (repressor) the assembly or activity of the basal transcription machinery [1,2]. The architecture of promoter and enhancer elements re£ects the combinatorial control of gene expression by di¡erent transcriptional factors. Each gene is indeed under the control of a multiplicity of elements, sometimes redundant, that can be dispersed over several kilobases (kb) around or within a gene [3^5]. Furthermore, speci¢c DNA binding transcription factors function against a large excess of genomic sequences that may contain binding sites with variable a¤nities. All these features make it di¤cult to predict the localization of regulatory regions in genomic sequences. Another striking aspect of gene regulation in higher eukaryotes is the tissuerestricted expression of several genes [6,7]. This can be controlled by tissue-speci¢c transcriptional factors which are themselves expressed only in a limited number of cell types, or by a segment of DNA, the tissue-speci¢c cis-element, which will direct gene expression in a tissue-speci¢c manner. Human prostatic acid phosphatase (PAcP) is a prostate epithelium-speci¢c di¡erentiation antigen [8]. PAcP has a long history of serving as a tumor marker of prostate cancer, due to its tissue-speci¢c manner of expression [8,9]. The cellular form of PAcP functions as a negative regulator of the tyrosine phosphorylation signal by dephosphorylating phosphotyrosine on c-ErbB-2 protein that leads to the down-regulation of prostate cell growth [10,11]. However, relative little is known regarding the molecular mechanism of tissue-speci¢c regulation of the expression of PAcP gene in prostate cells. In the present study, we carried out functional analyses of the 5P £anking region of the PAcP gene for elucidating the putative cis-acting elements that mediate the transcription of the PAcP gene in prostate cancer cells. We cloned and characterized a V3 kb DNA fragment from the promoter region of the PAP gene (32899/+87), and determined the regulatory cis-elements. In this communication we reported that this 3 kb DNA fragment exhibits a promoter activity in PC-3 and LNCaP prostate cancer cells, and contains cis-positive as well as cis-negative regulatory elements.

2. Materials and methods 2.1. Materials Cell culture medium, fetal bovine serum (FBS), gentamicin and Lipofectin1 reagent were obtained from Life Technologies. The MasterAmp PCR Optimization kit was from Epicentre Technologies. Zero Blunt1 PCR cloning kit, and pCR-Blunt vector were obtained from Invitrogen. Vectors containing chloramphenicol acetyltransferase (CAT) gene: pCATBasic, pCATEnchancer, and pCATPromoter, the CAT assay kit, and pSV-L-galactosidase vectors were purchased from Promega. DNA manipulations of plasmids were performed by conventional molecular biology techniques [12]. 2.2. Cells culture LNCaP and PC-3 human prostate cancer cells were routinely maintained in RPMI-1640 medium supplemented with 5% FBS, 1% glutamine, and 0.5% gentamicin [13]. 2.3. Isolation of the PAcP promoter region Two clones containing the entire PAcP gene in the pBAC vector were isolated by screening bacterial arti¢cial chromosome (BAC) high-density ¢lters (Genome Systems, MI, USA) using a 32 P-labeled PAcP promoter DNA fragment (between position 3345 and +87 base pairs (bp)). This promoter fragment was obtained utilizing polymerase chain reaction (PCR) with two primers 5P-AGC TGA GAT TGT GCC ACT GTC-3P and 5P-ACT TCG GTC TAG CCA GAA AAA-3P, which were prepared based on a published sequence [14]. In PCR reactions, the genomic DNA from LNCaP human prostate carcinoma cells was used as the template. The isolated genomic clones in the pBAC vector were designated as p16221 and p16222, respectively. 2.4. Cloning of the PAcP promoter The promoter fragment of PAcP gene was obtained by a PCR ampli¢cation using the p16221 clone as the template. The PCR reaction was con-

BBAEXP 93381 27-3-00

S. Zelivianski et al. / Biochimica et Biophysica Acta 1491 (2000) 123^132

ducted in a volume of 100 Wl in the presence of Pfu DNA polymerase (Stratagene), and bu¡er F from the MasterAmp1 PCR Optimization kit utilizing a Perkin-Elmer GeneAmp PCR System 2400 (PerkinElmer). Two oligonucleotide primers were utilized: 5P-AAA GTT CAA GGT GCC ACA G-3P and 5P-ACT TCG GTC TAG CCA GAA AAA-3P. The PCR mixture was ¢rst denatured by heating at 95³C for 5P. The ampli¢cation was performed for 30 cycles using the following conditions: 30Q at 94³C, 1P at 59.6³C, 1P30Q at 72³C. A DNA fragment of V3 kb was obtained and cloned into the pCR-Blunt vector. 2.5. Nucleotide sequence determination The nucleotide sequence of the entire 3 kb was determined in both directions using a cycle sequencing kit fmol0 DNA (Promega, USA) based on the dideoxy-chain-termination method. Sequencing reactions were also conducted with the Dye terminator cycle sequencing system, and analyzed by the DNA Sequencing Core Facility at the University of Nebraska Medical Center. Sequencing analyses were performed three times in two clones using both DNA strands. 2.6. Plasmid constructs To analyze the active promoter region of the human PAcP gene, a series of reporter plasmid constructs was made using the backbone of the pCATBasic vector (Promega). To assess the promoter activity, a HindIII^XbaI fragment of the PAcP promoter from the pCR-Blunt vector was cloned into the pCATBasic plasmid. The constructed plasmid, p2899, contained approximately a 3 kb promoter DNA fragment of the PAcP gene covering the region from 32899 to +87. The plasmid p1305 containing a 1.4 kb fragment of promoter DNA from 31305 to +87 was obtained from p2899 by digestion with A£II following by treatment with Mung Bean nuclease (New England Biolabs) and then religated. The promoter DNA in plasmid p2583 was excised from plasmid p2899 by digestion with KpnI, treated with Mung Bean nuclease, and religated. The reporter construction p2899v containing the promoter DNA with a deletion region 32583 to 31305 was obtained by digestion p2899 with KpnI and A£II restriction

125

enzymes, after overhead removal by Mung Bean nuclease, and religated. The plasmid p779, containing a promoter DNA of 866 bp (3779/+87), was subcloned from p2899 after subsequent HindIII and HincII digestions, Mung Bean nuclease treatment, and religation. Plasmid pCATEPAP containing a PAcP promoter region (31356/+87), a PCR product, and a SV-40 enhancer was obtained as described previously [13]. The plasmid p779E, containing a HincII^XbaI fragment from pCR-Blunt vector, was subcloned into the pCATEnhancer plasmid, after overhead removal of the HindIII site, followed by XbaI digestion and religation. The plasmid p205 (3205/+87) was obtained by PCR using two oligonucleotide primers 5P-TAA GCT TTG GGG AAA ACT GTG ATC TCT CT-3P and 5P-CTT CGC TGC AGC CAG AAA AAA AGC AGC A-3P containing sites for HindIII and PstI restriction enzymes, respectively. The ampli¢cation was performed for 30 cycles using the following conditions: 1P at 94³C, 1P at 59³C, 30Q at 72³C. A DNA fragment of approximately 270 bp was sequenced and cloned into the pCATBasic vector using HindIII and PstI restriction sites. 2.7. Transfection and reporter assays For transient transfection, cells were routinely plated at a density of 2.5U105 cells per well in a 6-well plate in RPMI-1640 medium containing 5% FBS 48 h before transfection. The adherent cells were transiently transfected, using 6 Wl cationic lipid reagent LipofectAMINE PLUS1 reagent (Life Technology), with 1 Wg of plasmid DNA for PAcP promoter/CAT reporter gene constructs in the serum-free Opti-MEM medium (Life Technology). The second plasmid, p-SV-L-galactosidase (Promega), was co-transfected at a ratio of 1:10 to the promoter/reporter gene constructs, serving as an internal control. After 4 h incubation, an equal amount of medium containing 10% FBS was added and incubated for an additional 2 h, which was then replaced with fresh RPMI-1640 medium supplemented with 5% FBS. 48 h after transfection cells were washed twice with phosphate-bu¡ered saline (PBS), scraped, and lysed in 1U reporter lysis bu¡er (Promega) for CAT and L-galactosidase assays. The protein concentration of cell extracts was deter-

BBAEXP 93381 27-3-00

126

S. Zelivianski et al. / Biochimica et Biophysica Acta 1491 (2000) 123^132

mined using the Bio-Rad protein assay kit (Bio-Rad Laboratories) with bovine serum albumin as a standard. Quantitative CAT assays were performed with the same amount of total cell lysate proteins in a reaction volume of 125 Wl in the presence of 14 C-chloramphenicol (Amersham Life Science) as described in the Promega CAT assay manual accompanying with the assay kit. Samples were incubated overnight followed by a single extraction with 300 Wl xylene. An aliquot of 250 Wl organic phase was transferred to a scintillation vial containing 2 ml EcoLume1 scintillation £uid (ICN) and counted with a Beckman LS 1801 scintillation counter. All experiments were repeated at least three times in triplicates. 2.8. L-Galactosidase assay Cultured PC-3 or LNCaP cells were co-transfected with a pSV-L-galactosidase vector containing the L-galactosidase gene driven by a SV-40 promoter as described above. Quantitative L-galactosidase assays were performed with the same amount of total cell lysate proteins in a reaction volume of 200 Wl, as described in the Promega CAT assay manual accompanying with the assay kit. Brie£y, cell lysate proteins were incubated in 100 mM sodium phosphate bu¡er (pH 7.3), containing 50 mM L-mercaptoethanol, 1 mM MgCl2 , and 0.66 mM o-nitrophenyl-L-Dgalactopyranoside (ONPG) at 37³C. The incubation time was 30 min for PC-3 and 16 h for LNCaP cell lysate proteins, respectively. The optical density was measured at 420 nm. 2.9. FBS e¡ect Cells were seeded in 6-well culture plates with 2.5U105 cells/well in RPMI-1640 medium supplemented with 5% FBS and maintained for 2 days before transfection. 24 h after transfection, the medium was replaced with a steroid-reduced medium, i.e. phenol red-free RPMI-1640 medium containing 1% heat-inactivated steroid-reduced FBS (SR-FBS) [10,15,16] for an additional 24 h. The cells were then fed with fresh medium containing 5% FBS, 5% or 1% SR-FBS, respectively, for 3 days prior to the determination of the cell growth and the CAT activity.

3. Results 3.1. Sequence of the 3 kb PAcP promoter DNA fragment The 3 kb promoter DNA fragment of the PAcP gene cloned in the pCR-Blunt vector was sequenced. Compared with the GenBank sequence data [14,17], four di¡erent nucleotides were noted in the region of our sequence from 32823 to 32574 (Fig. 1). The di¡erence at 32798 from G to C which eliminates the XbaI site was the most signi¢cant di¡erence from the published sequence [14,17]. This unique feature was con¢rmed by restriction analyses on 12 independent subclones (data not shown). 3.2. Identi¢cation of multiple cis-acting regulatory elements in the PAcP promoter To understand the regulation of the human PAcP gene expression at the transcriptional level, various portions of the PAcP promoter region were placed in front of a CAT reporter gene. Preliminary studies showed that plasmid p205 containing the shortest promoter region 3205/+86 exhibits the same transcriptional activity as plasmid p779 in PC-3 and LNCaP cells (data not shown). Thus, the transcriptional activity of p779 was utilized as the reference for further experiments. As shown in Fig. 2, the 3 kb DNA fragment of the immediate upstream sequence of the PAcP gene was divided into four clones (p779, p1305, p2583 and p2899). Among these clones, the p1305 exhibited the highest promoter activity, approximately 10- and three-fold higher than the p779 construct in PC-3 cells (Fig. 2A) and LNCaP cells (Fig. 2B), respectively. Inclusion of more upstream sequences resulted in a reduction of the CAT activity (Fig. 2, constructs p2583 and p2899). The region of 32583 to 31305 had a strong negative regulatory e¡ect in both PC-3 and LNCaP cells (Fig. 2). However, the region of 32899 to 32583 only showed a strong negative e¡ect in PC-3 cells (Fig. 2A). These results indicated that a transcriptional active cis-element, which can enhance the activity of basic promoter, is presented between positions 31305 and 3779 bp; while at least one negative regulatory region is located between 32583 and 31305.

BBAEXP 93381 27-3-00

S. Zelivianski et al. / Biochimica et Biophysica Acta 1491 (2000) 123^132

127

Fig. 1. Sequence of the PAcP promoter DNA. The 3 kb DNA sequence of the PAcP promoter region was determined. Four nucleotide di¡erences were detected, and compared with deposited sequences in GenBank [14,17]. These nucleotides are shown at the top line in bold typeface. The XbaI site in the reported sequence is underlined.

3.3. Region 32899/32583 functioning as a negative regulator To determine whether the 32899/32583 region could indeed function as a negative element, we constructed a reporter plasmid, i.e., p2899v, in which the fragment DNA between positions 32583 and 31305 was deleted. The CAT activity of this construct was approximately ¢ve-fold lower than the activity of p1305 in PC-3 cells (Fig. 2A), and 40% lower in

LNCaP cells (Fig. 2B). Thus, the region from 32899 to 32583 could indeed function as a transcriptional suppressor in both human prostate cancer cell lines. 3.4. Transcriptional activity of the cis-active fragment from 31305 to 3779 inversely correlates with cell growth We examined the biological activity of p1305 con-

BBAEXP 93381 27-3-00

128

S. Zelivianski et al. / Biochimica et Biophysica Acta 1491 (2000) 123^132

Fig. 3. Serum e¡ect on the PAcP promoter activity in PC-3 cells. PC-3 cells were transiently transfected with the promoter construct containing 3779/+87 bp (p779) (A) or 31305/+87 bp (p1305) (B), respectively. After 24 h of incubation in the medium containing 1% SR-FBS, cells were grown in di¡erent serum conditions for three days. The relative CAT activity was calculated as the mean of triplicates from three independent experiments. Cell growth was calculated based on total cellular proteins. Bar represents the standard deviation. Fig. 2. Identi¢cation of putative cis-acting DNA elements in the PAcP gene. Di¡erent parts of the 5P upstream region of PAcP gene (from 32899 to +87) were inserted into the pCATBasic reporter plasmid containing the CAT gene. The relative CAT activities were determined with cell extracts from transiently transfected PC-3 (A) and LNCaP (B) cells, normalized by the L-galactosidase activity and the protein amount. The data are presented as relative CAT activities over the value of the p779 expression vector. Data shown are mean of three independent experiments in triplicates. Bar represents a standard deviation. *P 6 0.05, p2583 versus p2899 plasmid for LNCaP cells (n = 9).

struct, in comparison with p779. After transfection, PC-3 cells were grown in a medium containing 5% FBS, 5% or 1% SR-FBS, respectively. Cells were then harvested to determine the cell growth and the CAT activity. As shown in Fig. 3A, in comparison with cells grown in 5% FBS, the cell growth was decreased in 5% SR-FBS, and further diminished in 1% SR-FBS. Concurrently, the promoter activity of p779 construct was decreased (Fig. 3A). Nevertheless, a decreased cell growth correlated with an increased activity of p1305 plasmid (Fig. 3B). A similar

Fig. 4. E¡ect on the 3779/+87 PAcP promoter activity by a SV-40 enhancer. All transfections were normalized with the L-galactosidase activity. Relative CAT activities were further normalized to the activity of basic promoter (3779 to +87). Each value represents the mean of CAT activity in triplicates of three independent experiments. Bar represents the standard deviation. SV-40P: SV-40 promoter; SV-40E: SV-40 enhancer.

BBAEXP 93381 27-3-00

S. Zelivianski et al. / Biochimica et Biophysica Acta 1491 (2000) 123^132

result was obtained in LNCaP cells (data not shown). Thus, the transcriptional activity of the PAcP promoter fragment from 31305 to 3779 bp inversely correlated with the cell growth. 3.5. Activation of the PAcP promoter by a SV-40 enhancer To determine whether a heterologous enhancer can activate the 779 bp fragment of the PAcP promoter DNA, this region was ligated with a SV-40 enhancer in the pCATEnhancer plasmid. In PC-3 cells, this SV-40 enhancer exhibited approximately a nine-fold induction of the CAT activity over the p779; while it

129

was lower than the activation by the `authentic' enhancer in p1305 plasmid (Fig. 4). Interestingly, in the presence of both SV-40 and PAcP 31305 to 3779 enhancer regions, the CAT gene was activated by over 35-fold (Fig. 4). Nevertheless, a 22-fold induction was observed for the plasmid containing the SV40 promoter and its `authentic' enhancer region, and about an eight-fold activity for the plasmid containing the SV-40 promoter alone, compared with p779 plasmid. Thus, the data indicated that a SV-40 enhancer can activate the PAcP promoter, and both SV-40 and PAcP enhancers have an additive e¡ect on the PAcP promoter activity.

Fig. 5. Location of consensus transcriptional factor binding sequences in the cis-active promoter region. The promoter sequence from 31332 to 3782 of the PAcP gene was analyzed by the MatInspector data bank search. Several consensus nuclear factor-binding sites, which are identi¢ed by an underline, were shown.

BBAEXP 93381 27-3-00

130

S. Zelivianski et al. / Biochimica et Biophysica Acta 1491 (2000) 123^132

3.6. Computer analysis of putative recognition sites for transcriptional factors Due to the potential importance of the cis-active element, we performed computer analyses to identify putative nuclear factor-binding sites (Fig. 5). The sequence of the cis-active region between 31332 and 3782 was analyzed by the MatInspector data bank searches [18]. Two recognition sites for the homeo domain factor Nkx-2.5 were identi¢ed to be located at nucleotides 31111 to 31103 and nucleotides 3970 to 3964. Two sequences consistent with the binding site for the hepatocyte nuclear factor HNF occurred at nucleotides 31233 to 31226 and 31214 to 31205. In addition, the data bank search also located one recognition site for the nuclear factor AP1 between 31300 and 31293, and a binding sequence for the cAMP-binding protein CREB at nucleotides 3976 to 3970. Thus, sequence analyses indicated that there are several potential recognition sites for known transcription factors in this cis-active region. 4. Discussion Transcriptional regulation of gene expression is generally achieved by modulation of the promoter function. Binding of RNA polymerase and the formation of a closed transcription complex have been shown to be dependent upon the speci¢c association of transcription factors. These factors recognize cisacting DNA sequences, which can lie either within the promoter or at some distance from it. The identi¢cation and characterization of such sequences, which are involved in the regulation of the PAcP promoter activity, would be a key step in analyzing the regulation of PAcP gene expression at the transcriptional level. This is the ¢rst report, to the best of our knowledge, in studying the structural organization of the 3 kb promoter region of the PAcP gene. Although DNA sequences of PAcP gene have been reported [14,17] the heterogeinity of PAcP cDNA exist [19]. Southern blot analyses with PAcP cDNA probes on the TaqI-digested genomic DNAs from unrelated individuals and members of large families from northern Finland reveal two simultaneous diallelic restriction fragment length polymorphisms [19]. Comparing

our sequence with the published sequence data, only four di¡erent nucleotides are noted in a region of about 250 bp, i.e., between 32823 and 32574 bp in a total of 3 kb sequence. The di¡erence in these sequences could be due to the natural PAcP polymorphism [20] or may re£ect the heterogeneity of the PAcP gene [15,19], a further study is needed. Results from transient transfection using prostate cancer cell lines LNCaP and PC-3 show that the PAcP transcriptional activation requires at least 200 bp of the 5P-£anking sequence. However, sequences further upstream, i.e. 5P up to 3779 bp, show no signi¢cant e¡ect on the transcriptional activity (data not shown and reference [21]). It is still possible that this region of 3779 to 3205 contains some unknown sequences which can be involved in the regulation of PAcP promoter activity. For example, the PAcP promoter fragment from 32899 to +87 exhibits the same level of CAT activity as the fragment 3779/+87 in PC-3 cells (Fig. 2A), and only approximately 50% higher in LNCaP cells (Fig. 2B). Thus, more detailed characterizations of this region are needed. Interestingly, the PAcP promoter activity in LNCaP cells is much lower than in PC-3 cells. It is possible that the low activity is due to the low transfection e¤ciency [13,22]. Another possibility is that LNCaP cells exhibit a low transcriptional activity, which is indicated by a slow growth rate as well as a low transcriptional activity of the SV-40 promoter [13,22]. Alternatively, the low promoter activity may be due to the competition by the endogenous PAcP gene for the limited transcription factors since LNCaP cell express endogenous PAcP but not PC3 cells. Further studies are required to clarify the molecular mechanisms. Deletion analyses of the PAcP promoter (Fig. 2) indicate that the 31305/+87 bp proximal sequence directs the highest level of the reporter gene activity in both human prostate cancer cell lines PC-3 and LNCaP. This activation can be suppressed by two regions including 32583 to 31305 and 32899 to 32583 along the 3 kb PAcP promoter fragment. The second suppressor (32899 to 32583) is very active in PC-3 cells and has a position independent activity (Fig. 2A), although it has much less e¡ect in LNCaP cells (Fig. 2B). The simple presence of the region 32899 to 32583 can suppress the activity of the enhancer element by up to ¢ve-fold in PC-3 (Fig.

BBAEXP 93381 27-3-00

S. Zelivianski et al. / Biochimica et Biophysica Acta 1491 (2000) 123^132

2A). Nevertheless, in the presence of 32583 to 31305 fragment, only approximately a three-fold inhibition was observed (Fig. 2A), indicating a cooperative e¡ect between these two regions. The transcriptional activity of the fragment from 3779 to +87 is decreased when cells are in non-permissive conditions (Fig. 3A). Interestingly, the activity is increased in the presence of the region from 31305 to 3779. The data indicate that the speci¢c transcriptional factors required for PAcP expression are actively functioning despite the suppression of the growth machinery. These factors interact with cis-active enhancer elements between positions 31305 and 3779 and enhance the transcriptional activity (Fig. 3). Currently, the molecular mechanism is not known if the cell growth per se or other factors regulated by FBS are involved in regulating the promoter activity. No matter what is the reason, the results are consistent with our observations that cellular PAcP activity as well as its protein level is inversely correlated with the growth rate [15,16,23,24]. Although, a SV-40 enhancer can activate the transcriptional activity of the PAcP core promoter, this activation is lower than the core promoter plus its authentic enhancer, i.e., p1305 (Fig. 4). Thus, the data indicate that the PAcP core promoter (3779/ +87) requires the presence of its own enhancer element, which is located between positions 31305 and 3779 bp. Results from transfection experiments clearly show that the PAcP promoter is highly active in PC-3 cells and contains positive as well as negative regulatory elements. However, our data can not exclude the possibility that additional distal regulatory elements are existed for an even higher level of expression in vivo. Furthermore, the regulation of PAcP gene expression speci¢cally in human prostate epithelium is complicated and not completely understood [10,21]. If the function of the PAcP enhancer element is to assist the assembly of a unique multiprotein complex in prostate epithelial cells, understanding of such a complicated structure is important [25,26]. Sequence analyses reveal presence several potential recognition sites for known transcription factors in the cis-active region including AP-1 and CREB. The characterization of cis-regulatory elements of the PAcP promoter will provide useful information to identify transcriptional factors that are involved in regulating the

131

PAcP promoter activity. In addition, an understanding of the molecular mechanisms of tissue-speci¢c gene expression may facilitate the development of novel pharmaceutical agents. Acknowledgements We thank Ms. Fen-Fen Lin for preparing the genomic DNA from LNCaP cells, and Drs. Tzu-Ching Meng and Michael Verni for critical comments. This study was supported in part by a NIH Grant CA 72274, Nebraska Research Initiative, Nebraska Department of Health LB 506 (#2000-19), and LB 595, and BMB Prostate Cancer Research Fund.

References [1] J.J. Eloranta, S. Goodbourn, Positive and negative regulation of RNA polymerase II transcription, 12 (1996) 1^33. [2] P.J. Mitchell, R. Tjian, Transcriptional regulation in mammalian cells by sequence-speci¢c DNA binding proteins, Science 245 (1989) 371^378. [3] E.R. Schuur, G.A. Henderson, L.A. Kmetec, J.D. Miller, H.G. Lamparski, D.R. Henderson, Prostate-speci¢c antigen expression is regulated by an upstream enhancer, J. Biol. Chem. 271 (1996) 7043^7051. [4] P. Vorgia, V.I. Zannis, D. Kardassis, A short proximal promoter and the distal hepatic control region-1 (HCR- 1) contribute to the liver speci¢city of the human apolipoprotein C-II gene. Hepatic enhancement by HCR-1 requires two proximal hormone response elements which have di¡erent binding speci¢cities for orphan receptors HNF-4, ARP-1, and EAR-2, J. Biol. Chem. 273 (1998) 4188^4196. [5] D.S. Latchman, Eukaryotic Transcription Factors, Academic Press, London, 1995. [6] J.E. Darnell, Variety in the level of gene control in eukaryotic cells, Nature 297 (1982) 365^371. [7] D.S. Latchman, Gene Regulation: A Eucaryotic Perspective, Chapman and Hall, London, 1995. [8] M.F. Lin, G.M. Clinton, Human prostatic acid phosphatase and its phosphotyrosil protein phosphatase activity, Adv. Protein Phosphatses 4 (1987) 199^208. [9] M.F. Lin, G.M. Clinton, Human prostatic acid phosphatase has phosphotyrosyl protein phosphatase activity, Biochem. J. 235 (1986) 351^357. [10] M.F. Lin, T.C. Meng, P.S. Rao, C. Chang, A.H. Schonthal, F.F. Lin, Expression of human prostatic acid phosphatase correlates with androgen-stimulated cell proliferation in prostate cancer cell lines, J. Biol. Chem. 273 (1998) 5939^ 5947. [11] T.C. Meng, M.F. Lin, Tyrosine phosphorylation of c-ErbB-2

BBAEXP 93381 27-3-00

132

[12]

[13]

[14]

[15]

[16]

[17]

[18]

S. Zelivianski et al. / Biochimica et Biophysica Acta 1491 (2000) 123^132 is regulated by the cellular form of prostatic acid phosphatase in human prostate cancer cells, J. Biol. Chem. 273 (1998) 22096^22104. J. Sambrook, E.F. Fritsch, T. Maniatis, Molecular Cloning a Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1989. S. Zelivianski, D. Comeau, M.F. Lin, Cloning and analysis of the promoter activity of the human prostatic acid phosphatase gene, Biochem. Biophys. Res. Commun. 245 (1998) 108^112. F.S. Sharief, S.S. Li, Nucleotide sequence of human prostatic acid phosphatase ACPP gene, including seven Alu repeats, Biochem. Mol. Biol. Int. 33 (1994) 561^565. M.F. Lin, R. Garcia Arenas, Y.C. Chao, M.M. Lai, P.C. Patel, X.Z. Xia, Regulation of prostatic acid phosphatase expression and secretion by androgen in LNCaP human prostate carcinoma cells, Arch. Biochem. Biophys. 300 (1993) 384^390. M.F. Lin, R. Garcia-Arenas, M. Kawachi, F.F. Lin, Regulation of the expression of prostatic acid phosphatase in LNCaP human prostate carcinoma cells, Cell Mol. Biol. Res. 39 (1993) 739^750. P. Virkkunen, P. Hedberg, J.J. Palvimo, E. Birr, K. Porvari, M. Ruokonen, P. Taavitsainen, O.A. Janne, P. Vihko, Structural comparison of human and rat prostate-speci¢c acid phosphatase genes and their promoters: identi¢cation of putative androgen response elements, Biochem. Biophys. Res. Commun. 202 (1994) 49^57. K. Quandt, K. Frech, H. Karas, E. Wingender, T. Werner, MatInd and MatInspector: new fast and versatile tools for detection of consensus matches in nucleotide sequence data, Nucleic Acids Res. 23 (1995) 4878^4884.

[19] P. Vihko, P. Virkkunen, P. Henttu, K. Roiko, T. Solin, M.L. Huhtala, Molecular cloning and sequence analysis of cDNA encoding human prostatic acid phosphatase, FEBS Lett. 236 (1988) 275^281. [20] R. Winqvist, P. Virkkunen, K.H. Grzeschik, P. Vihko, Chromosomal localization to 3q21- - -qter and two TaqI RFLPs of the human prostate-speci¢c acid phosphatase gene (ACPP), Cytogenet. Cell Genet. 52 (1989) 68^71. [21] J.D. Shan, K. Porvari, M. Ruokonen, A. Karhu, V. Launonen, P. Hedberg, J. Oikarinen, P. Vihko, Steroid-involved transcriptional regulation of human genes encoding prostatic acid phosphatase, prostate-speci¢c antigen, and prostate-speci¢c glandular kallikrein, Endocrinology 138 (1997) 3764^ 3770. [22] M. Ruokonen, J.D. Shan, P. Hedberg, L. Patrikainen, P. Vihko, Transfecting well-di¡erentiated prostatic cancer cell line LNCaP, Biochem. Biophys. Res. Commun. 218 (1996) 794^796. [23] M.F. Lin, J. DaVolio, R. Garcia Arenas, Expression of human prostatic acid phosphatase activity and the growth of prostate carcinoma cells, Cancer Res. 52 (1992) 4600^4607. [24] M.F. Lin, R. Garcia Arenas, X.Z. Xia, B. Biela, F.F. Lin, The cellular level of prostatic acid phosphatase and the growth of human prostate carcinoma cells, Di¡erentiation 57 (1994) 143^149. [25] R. Tjian, T. Maniatis, Transcriptional activation a complex puzzle with few easy pieces, Cell 77 (1994) 5^8. [26] S. Goodbourn, Eukaryotic Gene Transcription, IRL Press at Oxford University Press, Oxford, 1996.

BBAEXP 93381 27-3-00