A novel method for the determination of basal gene ... - Nature

D 2001 Nature Publishing Group 0929-1903/01/$17.00/+0 www.nature.com/cgt

A novel method for the determination of basal gene expression of tissue-specific promoters: An analysis of prostate-specific promoters Henk G. van der Poel,1 John McCadden,1 Gerald W. Verhaegh,2 Mark Kruszewski,1 Fernando Ferrer,1 Jack A. Schalken,2 Michael Carducci,1 and Ronald Rodriguez1 1

Brady Urologic Institute, Johns Hopkins Medical School, Baltimore, Maryland 21287; and 2Urological Research Laboratory, Nijmegen, Netherlands.

Because the toxicity of suicide gene therapeutics is directly related to basal promoter activity, we developed an assay to test for promoter ‘‘leakiness’’ using a diphtheria toxin mutant. Sequences of 15 prostate - specific gene promoter constructs were cloned in an expression plasmid ( pBK; Stratagene, La Jolla, CA ) backbone driving expression of an attenuated mutant of diphtheria toxin A ( tox176 ). Low expression levels of the DT - tox176 result in significant protein synthesis inhibition reflected by a decreased expression of the luciferase activity of a simultaneously transfected CMV luciferase construct. ID50 ( dose of plasmid with 50% luciferase inhibition ) was calculated for each promoter construct in different cell lines. Highest transactivational activity ( ID50 < 75 ng ) was found for the CMV promoter in all cell lines, which is in agreement with the dual luciferase assay findings. Unlike the dual luciferase findings, however, the DT - tox176 assay showed protein inhibition of CN65 ( PSA promoter / enhancer ) and PSE – hK2 ( PSA enhancer and basal human kallikrein 2 promoter ) in HEK293 and DLD cells indicating ‘‘leakiness’’ of these promoter constructs. Low basal promoter activity in nonprostate cell lines was found for the minimal PSA promoter, hK2, DD3, and OC promoters. The DT - tox176 assay can better predict basal promoter activity compared to less sensitive dual luciferase assay. Cancer Gene Therapy ( 2001 ) 8, 927 – 935 Key words: Prostate cancer; gene promoters; luciferase assay; diphtheria toxin; gene therapy; protein synthesis inhibition.

P

rostate cancer is the most common malignancy in American men. Therapeutic modalities for advanced hormone refractory disease are limited to chemotherapeutic regimens with considerable toxicity. Considering the low proliferation index in these cancers, many cytotoxic agents are ineffective.1 As a result, gene therapeutic modalities have recently been developed. Initially, intraprostatic local injection of a herpes simplex thymidine kinase gene was used to convert the prodrug ganciclovir into a phosphorylated nucleoside poison with limited success.2,3 Subsequent efforts have been made to design adenoviral vectors, which replicate selectively in prostate cells and hence lyse the tumor via its normal lytic cycle (i.e., oncolytic viruses ).4,5 Though these early attempts at utilizing tissue -specific gene expression have been encouraging, optimization of subsequent generation vectors will be greatly facilitated by improving the fidelity of the tissue -specific transgene expression. Promoter activity is commonly assessed by measuring the expression of a reporter gene (e.g., chloramphenicol acetyltransferase, - galactosidase, firefly luciferase, alkaline phosphatase, and - glucoronidase ). Most recently, the use of a ‘‘dual’’ luciferase assay has allowed internal control of Received July 27, 2001. Address correspondence and reprint requests to Ronald Rodriguez, MD, PhD, Marburg Room 145, Johns Hopkins Hospital, 600 North Wolfe Street, Baltimore, MD 21287. E-mail address: [email protected]

Cancer Gene Therapy, Vol 8, No 12, 2001: pp 927 – 935

the variability of transfection efficiencies from day to day and from cell line to cell line.6 Although useful to study the high end of the transgene expression spectrum, low expression levels may be missed due to the detection method used. Because low expression levels of a potent suicide gene may still result in unwanted side effects in a gene therapeutic setting, we evaluated the application of an extremely toxic agent ( diphtheria toxin) for determining the basal promoter activity of prostate -specific gene promoters. These promoters were derived from the prostate - specific antigen ( PSA ) gene,7 the prostate - specific membrane antigen ( PSMA ) gene,8 the human glandular kallikrein gene ( hK2 ),9 the rat probasin gene ( PBN ),10 the mouse osteocalcin gene ( OC ),3 and the human DD3 gene.11 In addition, chimeric promoter constructs were made by fusing the prostate - specific antigen enhancer ( PSE ) to each of the different promoter constructs. Diphtheria toxin is secreted from the Corynebacterium diphtheriae as a single polypeptide. Through partial proteolysis, two disulfide -linked fragments are formed: fragments A and B. DT binds to specific cell surface receptors through domains in the B fragment and is subsequently internalized by endocytosis.12 Translocation of the catalytic A fragment in the cytosol finally results in protein synthesis inhibition by ADP ribosylation of elongation factor 2.13 Although some cells may not express the cell surface receptor, all mammalian cells are sensitive to DT when present in the cytosol. The extreme

927

928

VAN DER POEL, MCCADDEN, VERHAEGH, ET AL: BASAL EXPRESSION OF PROSTATE PROMOTERS

cytotoxicity of the wild type DT, as little as one molecule, is lethal to the cell,14 making this toxin less applicable for gene therapeutic purposes15 because many promoters driving gene expression will have some basal activity, i.e., ‘‘leakiness’’ of promoter activity in presumably nontarget cells. Attenuation of the A fragment of the DT protein showed decreased toxicity compared to the wild type DT.16 The DT-tox176 mutant contains a Gly -to -Asp mutation at codon 638, resulting in an attenuation of 15 – 30 times less toxicity than DTwt. Thus, this mutant is less likely to result in unwanted site effects through ‘‘leakiness’’ of the tissue -specific promoter, i.e., expression of the gene product in transfected cells normally not known to activate the promoter.17 In the present study, basal promoter activity was assessed using DT- tox176 in an analysis of protein synthesis inhibition. Basal promoter activity was higher than estimated compared to a dual luciferase assay for a variety of constructs. This method allows the rapid discrimination of basal promoter activity in a variety of cells lines. The method is particularly well suited to detecting low - level transgene expression and ideally used as a precursor to suicide gene therapeutics, where the inherent toxicity of the approach depends on the basal gene expression in target and nontarget cells. It may also be useful in the generation of transgenic tissue -specific expression studies, by allowing better identification of tightly regulated promoters. Finally, conditionally replication - competent adenoviral constructs rely on the fidelity of the promoter used to drive E1a expression and the use of our basal expression method may allow better discrimination of potentially useful promoter constructs prior to the tedious and expensive process of generating a recombinant viral vector.

MATERIALS AND METHODS Cell culture

Cell lines were cultured in T75 flasks containing 10 mL of the appropriate medium and 10% fetal bovine serum at 378C and 5% CO2. Six prostate cancer cell lines (LNCaP, LA PC4, CWR22R, DU145, TSU -pr1, and PC3 ), one colon cancer ( DLD ), and the human embryonal kidney line 293 ( HEK293 ) were plated in 96- well plates at a density of 104 cells/well. Cells were grown for 2 days in antibiotic - free 10% fetal bovine serum medium. At day 3, transfections were performed. Reporter constructs

Reporter constructs were all cloned in the pBK – CMV backbone (Stratagene ), with promoter substitutions for the CMV promoter as described below. pBK – CN33 (basal PSA promoter ) was constructed by using the CN33 promoter plasmid obtained as a generous gift from D. R. Henderson ( Calydon, Menlo Park, CA ). The PSA promoter ( 541 to + 12) was amplified from CN33 using the following primers: 50 GGGCCACTAGTCGATATC AAGCTTGGGGC 30 and 50 GGCCATATGGAATTCCTGCAG-

CCCGG 30, which contained SpeI and NdeI restriction sites. pBK – CMV was digested with VspI and NheI to remove the CMV promoter. Following digestion of the CN33 polymerase chain reaction ( PCR ) product with SpeI and NheI, it was cloned into the pBK vector. The pBK – CN65 (PSA enhancer and promoter fusion ) was made using the following primers, 50 GGGCCACT AGTCGATATCAAGCTTGGGGC 30 and 50 1GGGCAGATCTAAAGCTGGGT ACCTCTAGAAAATCT 30 to amplify the PSE – PSA complex from CN65 (PSE –PSA construct ) (generous gift from D. R. Henderson; Calydon ). Restriction sites BglII and SpeI were incorporated for cloning purposes. A separate PCR using primers, 50 GGGCAGAT CTTATGTAACGCGGAACTC 30 and 50 TCCGCCCCATTGACGCA 30, was performed to obtain the pBK backbone with BglII and NheI 50 to 30 ends. Following cleavage with the aforementioned enzymes, ligation of the products was performed. The pBK – PBN construct was made in a similar fashion to pBK – PSE – PSA. The same vector product from pBK – CMV was used. The probasin promoter was amplified from MatLyLu genomic DNA using the following primers, 50 GCATAGATCTCCACA AGTGCATTTAGCCTC 30 and 50 GCATGCTAGCGATCGTGTGTG AGCTCTGTAGGT 30, which contained BglII and NheI restriction sites. The PCR product from above was then cut with these restriction enzymes and the cleavage products were ligated. pBK –hK2 was made in a fashion similar to pBK –PBN with the pBK –CMV vector product. The hK2 promoter was cloned by PCR of LNCaP genomic DNA using the following primers: 50 GCATAGATCTGTGCT CACGCCTGTAATCTC 30 and 50 GCATGCTAGCT GCTGACACAGGTGTCCA 30. The pBK – PSMA construct was made using genomic DNA and the following primers, 50 GAAGCAGATCTAATTCCCT TCCTTCCTTGCCCT 30 and 50 GAGTTGCTAGCAATGCCTCGCTTATCAGCCCTG 30. Incorporated within were BglII and NheI restriction sites. Subsequently, the PBN promoter of pBK – PBN was excised and the products of the aforementioned PCR inserted in its place. The luciferase reporter plasmids were created by excising the firefly luciferase reporter gene from pSP -Luc + (Promega, Madison, WI ) using a HindIII and XbaI digest. This was then inserted in pBK – CMV into the same sites. Subsequently, pBK – PSA, pBK – PSE – PSA, pBK – PBN, and pBK –hK2 constructs were made by excising Luc reporter from pBK –CMV –Luc using an SpeI and NotI digest and inserting the restriction fragment into the respective plasmid with the same enzymes. Insertion of the prostatic- specific enhancer ( PSE ) into plasmids was done by isolating the 1592 - bp fragment from CN65 using a BglII and BamHI digest and inserting it into a BglII site of pBK – PBN, pBK – PSMA, and pBK – hK2 to make constructs incorporating the PSE. DD3 was a the generous gift of J. Shalken ( Uro -oncology Research Group, Neimegen, the Netherlands ). pBK – PSE – DD3 – Luc was made by excising PSA from pBK – PSE – PSA – Luc with a BamHI digest and inserting the DD3 sequence into the same excision sites. the pBK –DD3 –Luc construct was similarly made by excising PSE – PSA from pBK – PSE – PSA – Luc and inserting DD3 into this site. Diphteria toxin A chain reporter gene was generated from

Cancer Gene Therapy, Vol 8, No 12, 2001


PCR of pDT201 (ATCC, Rockville, MD ) using the PCR primers: 50 GGATCCATGGGCGCTGATGATGTTG 30 and 50 CCGCGGCCGCCTGACACGATTTCCTGCAC 30, incorporating a BamHI site at the 50 end of the gene and a NotI site at the 30 end of the gene, which were used for directional cloning into the corresponding BamHI and NotI sites of the pBK backbone. DT-tox176 was made by sitedirected mutagenesis utilizing a commercial kit (Stratagene ) by converting a G!A transition at codon 638, resulting in a Gly- to -Asp point mutation. The DD3 promoter was a generous gift from Dr. Verhaegh.25 The fragment no. 1.12 ( 152 to +62 bp) from their publication was used. The AP -1 DD3 clone was obtained by DNA synthesis of two AP -1 response elements (actGACTcagtacatGACTcagt ). To this sequence, XhoI and HincII sites were added to allow cloning it into the DD3(1.12 ) basal promoter fragment. The murine osteocalcin promoter (OC )

a. -1500

-1250

-750

-1000

-500

bp 0

-250

was a generous gift from Dr. T. Gardner. To obtain all basal promoters in combination with one copy of the prostate specific enhancer (PSE ), a 1592 -bp fragment from CN65 was digested using BglII and BamHI ( Fig 1b ). Subsequently, this fragment was cloned into BglII site of the PBN, PSMA, hK2, OC, and DD3 plasmids. All promoter fragments were similarly cloned in the pBK – luc backbone to obtain the reporter plasmid for the dual luciferase assay. For co- reporter, we used the pBK – CMV – luc plasmid (Fig 1c ). The corresponding positions of all of the promoter sequences are shown relative to the transcriptional start site. All PCR -based clones were sequenced and confirmed by comparison to the known GenBank entries before proceeding with subsequent experimentation. Constructs generated by restriction endonuclease digestion were confirmed by at least three different restriction digests to confirm the predicted pattern.

250

500

Not1

BamH1 CMV promoter -154

DT-tox176 +62

DD3 -754

DT-tox176 +62

hK2 -304

DT-tox176 +62

hK2 enhancer

hK2

DT-tox176

-7435

-8605

OC -435

DT-tox176 +33

PBN -541

DT-tox176 +12

PSA -1275

DT-tox176 +62

PSMA

b.

929

BglII

DT-tox176

BamH1 PSE

-5322

DT-tox176 -3738

c.

CMV promoter

DT-tox176

DT-tox176 mRNA

DT-tox176 protein

eEF2

Luciferase protein

Luciferase mRNA

luciferase

CMV promoter

-1500

-1250

-1000

-750


19 -500

-250

bp 0

250

1750

Figure 1. Constructs for the DT - tox176 assay for promoter activity all in a pBK backbone. a: Reporter constructs for CMV ( positive control ), DD3, hK2, hK2 enhancer / promoter ( CP390 ), osteocalcin ( OC ), probasin ( PBN ), PSA, and PSMA. b: All promoters, except for CP390, were also combined with the prostate - specific enhancer from the PSA gene ( PSE ). c: Mechanism of action of the DT - tox176 assay: increased expression of DT - tox176 protein will inhibit luciferase mRNA translation at the level of eukaryotic elongation factor 2 ( eEF2 ).

930


point 0 hour ) for reporter and co -reporter plasmid was applied. For each dose –response curve, the inhibitory dose of 50% (ID50 ) was calculated by assuming linear correlation between the relative protein inhibition (RPI ) and the logarithmic value of the plasmid dose. RPI was calculated by dividing the luciferase expression at a certain dose by the luciferase expression at dose level 0 ng. ID50 values over 110100 ng (logID50 = 100) could not reliably be predicted and were scored as log( ID50 ) >( larger than )100. Basal expression or ‘‘leakiness’’ of the promoter was defined as the presence of protein inhibition in the DTtox176 assay in the absence of detectable promoter activity in the dual luciferase assay. Dual luciferase assay Figure 2. Time interval analysis for pBK – CMV – luc transfection ( 100 ng ) in DLD cells. All transfections were performed using 100 ng of the DT - tox176 containing plasmid at time point 0 hour. pBK – CMV – luc transfections were done either 1 hour before, at time of DT - tox176 transfection, or 1, 2, 4, or 20 hours after that time point ( arrow ) either with R1881 ( R + ) or without ( R ). Protein inhibition was normalized for pBK – CMV – luc without tox176 ( no tox176 ). DLD.

The dual luciferase assays were performed essentially as previously published.6 Briefly, the firefly luciferase buffer ( BufferLuc ) consisted of 25 mM glycylglycine, 15 mM KxPO4 (pH 7.8), 4 mM EGTA, 2 mM ATP, 1 mM DTT, 0.1 mM CoA, 75 M Luciferin with the final pH adjusted

a. DT - tox176 transfections relative protein inhibition

1.0

0.8

0.6

TSU 0.4

0.2

DLD

0.0 0

100

200

300

400

plasmid dose (ng/104 cells)

b.

3.00 2.75 2.50

log(ID50)

2.25 2.00

cell line

log(ID50)

La PC4

1.26

PC 3

1.26

LNCaP

1.29

HEK293

1.29

DU 145

1.44

1.75

DLD

1.73

1.50

TSU

2.66

1.25

U TS

D LD

aP H EK 29 3 D U 14 5

LN C

PC

3

1.00

La PC 4

Reporter transfections were performed at 0, 100, 200, and 400 ng tox176 plasmid on 104 cells per well. To correct for differences in overall plasmid concentration, empty pBK backbone plasmid was added to obtain a final plasmid concentration of 400 ng for each concentration. The pBK – CMV –luc plasmid (Photinus pyralis luciferase) was added at a concentration of 100 ng /well at the same time as transfection with the DT- tox176 plasmid. Several promoters contained androgen response elements ( ARE ) ( PSE enhancer, PSA basal promoter, CP390, and PBN ). An absent or dysfunctional androgen receptor is found in PC3, TSU, and DU145 cells. To test for androgen responsiveness of the promoters in different cell lines, R1881 (methyltrienolone; NEN, Boston, MA ) at final concentration of 5 nM was added 4 hours after transfection. After 48 hours, the medium was removed and cells were lysed with 30 L of passive lysis buffer in 30 minutes at room temperature. Subsequently, plates were read for luciferase activity in an illuminometer ( 1450 microbeta; Perkin Elmer Wallac, Gaithersburg, MD ) using a buffer luciferin containing buffer ( BufferLuc =25 mM glycylglycin, 15 mM KPO4, 4 mM EGTA, 2 mM ATP, 1 mM DTT, 15 mM MgSO4, 0.06 mM CoA, 75 M luciferin ). One hundred microliters of this buffer containing the lysed cells was dispensed per well; the delay time was set at 3 seconds and counting time at 8 seconds. To determine the influence of interval variation on the transfection of the CMV –luc reporter and tox176 containing plasmid, addition of 100 ng of pBK – CMV – luc co -reporter was done at different time points relative to the tox176 plasmid transfection (Fig 2 ). These data showed that transfection of the co -reporter longer than 1 hour after reporter plasmid transfection greatly reduces discriminative power of the assay. Hence, simultaneous transfection ( time

HEK293 DLD DU145 LaPC4 LNCaP PC3 TSU

1.2

Figure 3. Dose - dependent protein synthesis inhibition for pBK – CMV – DT - tox176 plasmid in different cell lines without R1881. a: Dose – response curves. b: Corresponding ID50.


931


RESULTS Protein inhibition

The pBK – CMV – DT- tox176 plasmids showed a dose dependent inhibition of protein synthesis as assessed by double transfection with a pBK – CMV – luc plasmid in all cell lines (Fig 3a). Cell lines used in our analysis included the hormone -responsive prostate lines LnCaP, LA -PC4, CWR22R; the hormone- insensitive prostate lines DU145, TSU -PR1, and PC3; and the nonprostate cell lines DLD ( human colon ) and HEK293 ( human embryonic kidney ). The hormone- sensitive lines have been found to have functional androgen receptors that can be stimulated by the synthetic androgen R1881. Strongest inhibition of protein synthesis was found for HEK293, LaPC4, LNCaP, and PC3 cells with more than 75% inhibition at 100 ng of DT- tox176 plasmid. RPI in TSU, DLD, and DU145 cells was markedly lower, ranging from 65% in DU145 to 36% in the TSU cell line corresponding with the calculated ID50 values (Fig 3b ). These findings confirm earlier data on the variable sensitivity and cell kill mechanism of different cell lines to DT-A. In particular, TSU has been found to undergo DT-mediated cell death by necrosis, while most other cells die by apoptosis. This difference appears to be related to the p53 axis.18 The expression levels of the dual luciferase assay were compared to the log values of the ID50 (Fig 4). If CMV promoter activity would differ between cell lines, this would


a. 1.2

relative protein inhibition

to 8.0. The stock BufferRen assay buffer was made as follows: 1.1 M NaCl, 2.2 mM Na2EDTA, 0.22 M KxPO4 ( pH 7.8 ), 0.44 mg /mL BSA, 1.3 mM NaN3, 1.43 M Coelenterazine, with the final pH adjusted to 5 for analysis of Renilla reniformis luciferase activity. A delay time and counting time, each of 5 seconds, were used for analysis of both luciferase activities.

1.0 0.8 0.6 0.4 0.2 0.0

CWR22R DLD DU145 HEK292 LaPC4 LNCaP PC3

0

100

200

300

400

500

400

500

dose of DT-tox176 plasmid (ng)

b. 1.2

relative protein inhibition

Figure 4. Comparison of CMV promoter activity using the dual luciferase ( relative expression ) and pBK – CMV – DT - tox176 assay ( ID50 ) by cell line without R1881 ( r = 0.701, P = .024, Pearson correlation ).

theoretically, in case of the CMV –tox176 construct, affect both reporter and co -reporter to the same extent. A clear negative correlation existed between the log( ID50 ) and the expression levels of the CMV promoter in the dual luciferase assay for the different cell lines (r = 0.701, P= .024, Pearson correlation ). This indicated that although sensitivity to DT-A among cell lines differed, CMV promoter activity highly correlated with protein inhibition as assessed in the DT- tox176 assay. Log( ID50 ) values ranged from 1.26 ( 18 ng ) for CMV – tox176 in LaPC4 to > 100 in 8 of 15 plasmids in CWR22R, 6 in DLD, 5 in PC3, 4 in HEK293, 3 in DU145, 3 in LNCaP, 3 in TSU, and 0 in LaPC4 without androgen supplementation. After androgen supplementation, these values were 1.26 (18 ng ) for CMV –tox176 in LaPC4 and > 100 in 7 of 15 plasmids in CWR22R, 4 in DLD, 3 in DU145, 6 in HEK293, 3 in LNCaP, 1 in LaPC4, 2 in PC3, and 2 in TSU. For the CN65 promoter (prostate - specific antigen gene enhancer fused to the core prostate - specific antigen promoter ), the dose –response curves are shown in Figure 5. Strongest protein inhibition (log( ID50 ) =3.25 ) was seen in LNCaP cells, whereas no inhibition was detectable in CWR22R and PC3 cells. When compared to the dual luciferase analysis, for PC3, TSU, CWR22R, and

1.0 0.8 0.6 0.4 0.2

CWR DLD DU145 HEK293 LaPC4 LNCaP PC3

0.0 0

100

200

300

dose of DT-tox176 plasmid (ng)

Figure 5. Dose – response curves for pBK – CN65 – tox176 in different cell lines without ( a ) and with R1881 ( b ).

932


LNCaP (androgen -supplemented ), a correlation between dual luciferase expression and log( ID50 ) was found for CN65 (Fig 6 ). However, no such correlation of CN65 promoter activity was found for HEK293, DU145, and LNCaP ( androgendeprived ), suggesting basal promoter ( ‘‘leakiness’’) activity of the CN65 promoter in these cell lines that could not be detected by the dual luciferase assay but was clear in the DT-tox176 assay. Several promoter constructs showed activity in the DTtox176 assay where no expression was found in the dual luciferase assay. In 66 cell line /promoter combinations, the dual luciferase assay did not detect any promoter activity. In 30 of these combinations (45% ), the log( ID50 ) was lower than 5 ng, indicating significant basal expression. Table 1 shows the dual luciferase and log( ID50 ) values for each cell type / promoter combination. To determine relative ‘‘leakiness’’ of each promoter in a specific cell line, the log( ID50 ) was multiplied by the relative expression determined in the dual luciferase assay. Lower values of this multiplier correspond with an increased promoter ‘‘leakiness’’ or, in other words, underestimation of promoter activity in the dual luciferase assay ( Table 2). Promoters with high basal expression levels are the PSE – hK2 ( 4 /8) and the PSMA (3 /8 ) promoter. Cell lines with high ‘‘leakiness’’ of several promoters are DU145 ( 3 of 15 promoters ) and PC3 (3 of 15 promoters ), both of which are

a. 100

TSU PC3 LNCaP LaPC4 HEK293 DU145 DLD CWR22R

log(ID50) (ng)

95

90 9 6 3 0 0

1

2

3 20

30

40

50

dual luciferase (fold induction over pBK-luc)

b.

100

TSU PC3 LNCaP LaPC4 HEK293 DU145 DLD CWR22R

log(ID50) (ng)

95

90 9 6 3 0 0

1

2

3 20

30

40

50

dual luciferase (fold induction over pBK-luc)

Figure 6. Comparison of dual luciferase assay and DT - tox176 assay ( log( ID50 ) ) for CN65 promoter activity in eight cell lines. a: Without androgens. b: After androgen suppletion.

androgen -independent prostate cancer cell lines. Minimal ‘‘leakiness’’ in most cell lines was found for the CN33 (basal prostate - specific antigen promoter ), hK2 ( human glandular kallikrein 2 promoter ), and the OC (osteocalcin ) promoter. Interestingly, basal expression levels for the PSE – DD3 promoter construct were high in five of six prostate cancer lines, whereas no activity was detected in nonprostatederived cell lines ( DLD and HEK293 ), suggesting that this promoter, although relatively weak in the dual luciferase assay, has a highly prostate - specific transactivation pattern. DISCUSSION

The novel DT- tox176 assay of basal promoter activity enhances the detection of promoter specificity compared to the conventional dual luciferase assay. Promoter activity is defined as the number of mRNA copies transcribed in a fixed time period. Although mRNA quantification through Rnase protection or S1 mapping is feasible and can provide information on the activity of cis - acting genetic elements,19,20 reporter gene assays are in general more sensitive and are most widely applied for screening of promoter activity. Gene products such as luciferase, chloramphenicol acetyltransferase ( CAT ), - galactosidase ( - gal ), -glucuronidase ( -glu ), alkaline phosphatase (AP ), growth hormone (GH ), and green - fluorescent protein (GFP ) are generally used in chemiluminescence, bioluminescence, or ELISA assay. Depending on the detection method used, detection limits vary from 103 molecules for the luciferase assay to over 108 for the CAT and GH assays.21 In the case of gene therapy with replicating viruses or highly toxic suicide genes such as diphtheria toxin, lowlevel gene expression will result in toxicity to nontarget cells.14,17 Toxicity will increase when, due to the basal promoter activity, effector genes are expressed to cytotoxic levels in nontarget cells. This becomes exceedingly important when highly toxic effector genes are applied, such as diphtheria toxin, or in conditionally replicating adenoviruses. Even the low levels of promoter activity (basal activity ) can result in unacceptable toxicity. We used the high toxicity of the diphtheria toxin to evaluate the basal activity of several prostate - specific gene promoters in a panel of human prostate - and nonprostate - derived cell lines. Earlier data22 ( including unpublished observations of our own ) showed low levels of luciferase activity for the majority of prostate -specific gene promoters in at least one prostate cancer cell line. None of the prostate - specific promoters is active in all tested prostate cancer lines, limiting their universal application in promoter- driven therapy for all prostate cancers. Since some prostate cancer cell lines are androgen -dependent ( e.g., LNCaP and LAPC - 4), while others are androgen- independent (e.g., TSU and DU145 ), the difference in expression patterns is likely related to the presence or absence of a functional androgen receptor. Similarly, low luciferase levels were found in nonprostate lines. Considering the detection limit of the luciferase assay, it may very well be that low levels of enzyme remain undetected. Inasmuch as we are actively involved in the development of both conditionally replication- competent


933


Table 1. Comparison of log( ID50 ) ( DT - tox176 Assay ) and Dual Luciferase Assay in Eight Cell Lines Without Androgen Suppletion

AP1 ( ) CMV ( ) CN33 ( ) CN65 ( ) CP390 ( ) DD3 ( ) hK2 ( ) OC ( ) PBN ( ) PSEDD3 ( ) PSEhK2 ( ) PSEOC ( ) PSEPBN ( ) PSEPSMA ( ) PSMA ( )

log( ID50 ) dual log( ID50 ) dual log( ID50 ) dual log( ID50 ) dual log( ID50 ) dual log( ID50 ) dual log( ID50 ) dual log( ID50 ) dual log( ID50 ) dual log( ID50 ) dual log( ID50 ) dual log( ID50 ) dual log( ID50 ) dual log( ID50 ) dual log( ID50 ) dual

DLD

DU145

HEK293

LaPC4

LNCaP

PC3

TSU

CWR22R

1.9 10.3 1.9 1901.2 90.0 4.0 7.9 2.0 3.2 27.0 100.0 1.0 20.0 2.0 100.0 1.0 10.2 10.0 100.0 1.0 7.0 1.0 4.8 1.0 18.9 4.0 100.0 1.0 4.4 23.0

2.5 8.4 1.6 2098.3 100.0 1.0 3.9 1.0 2.1 1.0 100.0 1.0 100.0 1.0 100.0 1.0 100.0 3.0 3.7 1.0 2.4 1.0 2.3 1.0 42.5 3.1 3.1 3.0 100.0 2.0

3.7 9.3 1.7 1101.5 100.0 1.0 3.3 1.0 8.1 35.4 19.9 1.0 100.0 1.0 100.0 1.0 100.0 2.7 100.0 1.0 2.9 1.0 4.1 1.0 15.2 1.6 7.8 1.0 2.3 1.0

6.9 20.0 1.3 3680.1 3.7 1.0 5.4 1.0 3.8 2.0 100.0 1.0 2.9 3.0 100.0 1.0 7.1 1.0 8.0 1.0 1.6 1.0 3.7 1.0 3.8 1.0 5.0 1.8 4.6 1.0

2.1 83.9 1.6 3384.2 100.0 1.0 3.3 1.0 2.4 17.6 3.0 1.0 100.0 1.0 5.7 1.0 4.1 1.0 7.0 1.0 2.2 1.0 3.4 1.0 2.4 1.0 2.6 1.0 3.0 1.0

4.7 66.1 1.7 588.3 100.0 1.0 100.0 1.1 2.9 9.8 100.0 1.0 100.0 3.1 10.0 4.2 5.1 1.0 2.9 1.0 3.2 1.0 2.7 4.0 2.9 1.1 2.8 1.0 1.8 1.0

3.1 41.6 2.7 455.7 100.0 1.0 9.0 2.0 15.0 3.0 2.2 1.0 100.0 1.0 100.0 1.0 11.2 5.9 5.0 2.0 100.0 1.0 8.0 1.0 50.0 1.8 50.0 4.0 3.4 4.0

2.0 9.9 1.3 17.2 100.0 1.4 100.0 1.2 100.0 1.0 4.6 1.0 100.0 1.0 9.4 1.5 100.0 1.0 69.9 1.0 100.0 1.0 100.0 1.1 100.0 1.2 4.8 1.0 5.5 1.0

adenoviruses as well as suicide gene – based adenoviral vectors, we developed an assay based on the DT-tox176 to better assess basal promoter activity. The inhibitory effects of DT-tox176 expression on translation resulted in a reduced expression of luciferase from the co- transfected plasmid even in cases where the dual luciferase assay did not show any transactivation. Several

promoters showed such tendency to ‘‘leakiness’’ in multiple cell lines. In particular, PSE – OC, PSE –hK2, and PSMA under androgendeprived conditions showed considerable basal promoter activity in at least four cell lines. Earlier studies showed increased PSMA levels in androgendeprived prostate cancer.23 Low ID50 values for the PSMA promoter under androgen -deprived conditions compared to

Table 2. Multiplication of the log( ID50 ) and Relative Expression from the Dual Luciferase Assay

AP1 ( ) CMV ( ) CN33 ( ) CN65 ( ) CP390 ( ) DD3 ( ) hK2 ( ) OC ( ) PBN ( ) PSEDD3 ( ) PSEhK2 ( ) PSEOC ( ) PSEPBN ( ) PSEPSMA ( ) PSMA ( )

DLD

DU145

HEK293

LaPC4

LNCaP

PC3

TSU

CWR22R

20.0 3560.3 360.0 15.8 86.4 100.0 40.0 100.0 102.1 100.0 7.0 4.8 75.5 100.0 101.8

21.0 3449.0 100.0 3.9 2.1 100.0 100.0 100.0 300.0 3.7 2.4 2.3 131.8 9.3 200.0

34.4 1872.6 100.0 3.3 287.4 19.9 100.0 100.0 266.0 100.0 2.9 4.1 23.7 7.8 2.3

137.1 4632.8 3.7 5.4 7.6 100.0 8.7 100.0 7.1 8.0 1.6 3.7 3.8 9.0 4.6

176.2 5285.0 100.0 3.3 42.2 3.0 100.0 5.7 4.2 7.0 2.2 3.4 2.4 2.6 3.0

311.1 991.2 100.0 114.9 28.7 100.0 308.1 41.9 5.1 2.9 3.2 10.8 3.2 2.8 1.8

129.0 1211.4 100.0 18.0 45.0 2.2 100.0 100.0 65.7 10.0 100.0 8.0 90.0 200.0 13.6

19.7 22.9 137.9 119.2 100.0 4.6 100.0 13.6 100.0 69.9 100.0 107.3 123.0 4.9 5.5

A low value indicates basal expression. Values lower than 3 are marked as they indicate high promoter ‘‘leakiness’’ as defined by a low log( ID50 ) in the DT - tox176 assay and low expression levels in the dual luciferase assay.


934


5 nM R1881 were found only in the DU145 cell line, whereas for all other cell lines, ID50 values were comparable for both conditions. For the dual luciferase assay, no difference in PSMA promoter activity was found with or without R1881, consistent with earlier findings.8 These findings suggest that PSMA promoter-driven gene therapy may not be particularly specific for hormone- resistant prostate cancer as previously indicated by others.24 High basal promoter activity, as assessed by the DT-tox176 assay, may hamper modalities using highly toxic suicide genes. Interestingly, addition of the PSE (PSA enhancer ) to the PSMA basal promoter reduced basal expression in both nonprostate cell lines, while not so in the prostate cell lines, hence increasing specificity.22 The hK2 promoter showed limited to no activity under hormone- deprived conditions. Addition of the PSE fragment increased the transactivation in six of eight cell lines from both prostate and nonprostate tissue. Thus, compared to the PSMA basal promoter, no gain in specificity was obtained by adding PSE. Addition of the hK2 enhancer to its basal promoter (CP390 ) increased expression in seven of eight cell lines again without gain of specificity. Similarly, addition of the PSE to the basal promoters of PBN and OC did not enhance specificity. Based on basal expression levels, the PSE –DD3 promoter construct showed highest specificity for prostate cancer cells. DD3 is a recently discovered prostate cancer –specific gene.11 Its promoter has been analyzed and highest activity found for a 216 -bp promoter fragment.25 Although active in some cell lines, both the activity and specificity of the DD3 minimal promoter were greatly enhanced by addition of the PSE. No promoter showed sole expression in prostate cancer cells. Promoters with frequent high basal expression levels are the PSE –hK2, PSE – OC, and the PSMA promoter. Cell lines with high basal promoter activation are DU145 and PC3. The PSE –DD3 basal promoter showed lowest basal promoter activity in nonprostate lines. The novel DT-tox176 assay clearly shows that many promoters express low levels of transcription in cells where the native protein (e.g., PSA ) is typically not detected. This ‘‘promiscuous’’ low -level gene expression becomes exceedingly important when more toxic gene therapy strategies are applied. The fact that most promoters are ‘‘leaky’’ in nontarget cells is not surprising; however, precise evaluation of the actual basal gene expression patterns has not been readily possible. This assay will be able to identify those promoters with minimal basal activity for highly specific promoter -targeted gene therapy and may also prove useful as a screening method prior to the development of tissue -specific transgene expression in transgenic animal studies. ACKNOWLEDGMENTS

The authors acknowledge the helpful advice of Donald Coffey, William Nelson, Jonathan Simons, Theodore DeWeese, Dale vander Putten, and Thomas Gardner during the course of these experiments. We also thank Donna

Klinedinst for technical assistance, and Helen Kelley for secretarial assistance. Funding came from CaPCure ( Santa Monica, CA ), The Robert Wood Johnson Foundation ( Princeton, NJ ), and the Dutch Cancer Foundation (Amsterdam, the Netherlands ).

REFERENCES 1. Scher HI, Fossa S. Prostate cancer in the era of prostate specific antigen. Curr Opin Oncol. 1995;7:281 – 291. 2. Herman JR, Adler HL, Aguilar - Cordova E, et al. In situ gene therapy for adenocarcinoma of the prostate: a phase I clinical trial. Hum Gene Ther. 1999;10:1239 – 1249. 3. Koeneman KS, Kao C, Ko SC, et al. Osteocalcin - directed gene therapy for prostate cancer bone metastasis. World J Urol. 2000;18:102 – 110. 4. Rodriguez R, Schuur ER, Lim HY, et al. Prostate - attenuated replication competent adenovirus ( ARCA ) CN706: a selective cytotoxic for prostate - specific antigen - positive prostate cancer cells. Cancer Res. 1997;57:2559 – 2563. 5. Yu DC, Chen Y, Seng M, et al. The addition of adenovirus type 5 region E3 enables calydon virus 787 to eliminate distant prostate tumor xenografts. Cancer Res. 1999;59:4200 – 4203. 6. Dyer BW, Ferrer FA, Klinedinst DK, et al. A noncommercial dual luciferase enzyme assay system for reporter gene analysis. Anal Biochem. 2000;282:158 – 161. 7. Schuur ER, Henderson GA, Kmetec LA, et al. Prostate specific antigen expression is regulated by an upstream enhancer. J Biol Chem. 1996;271:7043 – 7051. 8. Good D, Schwarzenberger P, Eastham JA, et al. Cloning and characterization of the prostate - specific membrane antigen promoter. J Cell Biochem. 1999;74:395 – 405. 9. Yu DC, Sakamoto GT, Henderson DR. Identification of the transcriptional regulatory sequences of human kallikrein 2 and their use in the construction of calydon virus 764, an attenuated replication competent adenovirus for prostate cancer therapy. Cancer Res. 1999;59:1498 – 1504. 10. Greenberg NM, DeMayo FJ, Sheppard PC, et al. The rat probasin gene promoter directs hormonally and developmentally regulated expression of a heterologous gene specifically to the prostate in transgenic mice. Mol Endocrinol. 1994;8:230 – 239. 11. Bussemakers MJ, van Bokhoven A, Verhaegh GW, et al. DD3: a new prostate - specific gene, highly overexpressed in prostate cancer. Cancer Res. 1999;59:5975 – 5979. 12. Choe S, Bennett MJ, Fujii G, et al. The crystal structure of diphtheria toxin. Nature. 1992;357:216 – 222. 13. Collier RJ. Diphtheria toxin: mode of action and structure. Bacteriol Rev. 1975;39:54 – 85. 14. Yamaizumi M, Mekada E, Uchida T, et al. One molecule of diphtheria toxin fragment A introduced into a cell can kill the cell. Cell. 1978;15:245 – 250. 15. Maxwell IH, Maxwell F, Glode LM. Regulated expression of a diphtheria toxin A - chain gene transfected into human cells: possible strategy for inducing cancer cell suicide. Cancer Res. 1986;46:4660 – 4664. 16. Maxwell F, Maxwell IH, Glode LM. Cloning, sequence determination, and expression in transfected cells of the coding sequence for the tox 176 attenuated diphtheria toxin A chain. Mol Cell Biol. 1987;7:1576 – 1579. 17. Burrows HL, Birkmeier TS, Seasholtz AF, et al. Targeted ablation of cells in the pituitary primordia of transgenic mice. Mol Endocrinol. 1996;10:1467 – 1477. 18. Rodriguez R, Lim HY, Bartkowski LM, et al. Identification of diphtheria toxin via screening as a potent cell cycle and p53 -



independent cytotoxin for human prostate cancer therapeutics. Prostate. 1998;34:259 – 269. 19. Sperisen P, Wang SM, Reichenbach P, et al. A PCR - based assay for reporter gene expression. PCR Methods Appl. 1992;1:164 – 170. 20. Kastenbauer S, Wedel A, Frankenberger M, et al. Analysis of promoter activity by polymerase chain reaction amplification of reporter gene mRNA. Anal Biochem. 1996;233:137 – 139. 21. Wood KV. Marker proteins for gene expression. Curr Opin Biotechnol. 1995;6:50 – 58. 22. Latham JP, Searle PF, Mautner V, et al. Prostate - specific antigen promoter / enhancer driven gene therapy for prostate


935

cancer: construction and testing of a tissue - specific adenovirus vector. Cancer Res. 2000;60:334 – 341. 23. Wright GL, Grob BM, Haley C, et al. Upregulation of prostate - specific membrane antigen after androgen deprivation therapy. Urology. 1996;48:326 – 334. 24. O’Keefe DS, Uchida A, Bacich DJ, et al. Prostate - specific suicide gene therapy using the prostate - specific membrane antigen promoter and enhancer. Prostate. 2000;45:149 – 157. 25. Verhaegh GW, van Bokhoven A, Smit F, et al. Isolation and characterization of the promoter of the human prostate cancer - specific DD3 gene. J Biol Chem. 2000;275:37496 – 37503.