| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Experimental Therapeutics |
Department of Molecular Pharmacology, St. Jude Children’s Research Hospital, Memphis, Tennessee 38105
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-fetoprotein (5)
, c-erbB-2 (6)
, glial fibrillary acidic protein (7)
, and L-plastin (8)
have been evaluated. In each of these studies, some degree of tumor-specific expression of the cDNA of interest was achieved with each promoter. The level of prodrug-activating enzyme required and the precise degree of sensitization needed for clinical applications will likely depend on many factors and differ for individual prodrugs. In this regard, encouraging results have been achieved in translational studies by Teoh et al. (2) . These investigators demonstrated similar efficiencies by DF3 and CMV promoters in regulating Hsvtk activity and in sensitizing multiple myeloma cells to ganciclovir. Also, Pang et al. (9 , 10) compared the efficiency of chimeric PSA/CMV promoter constructs with PSA promoter/enhancer constructs and found the chimeric constructs to be the less efficient in producing high levels of reporter enzyme activity in androgen-responsive human prostate cells. The strength of the PSA promoter/enhancer construct was approximately equivalent to CMV promoter. The studies by Pang et al. suggest that efficient tumor cell-specific expression is better achieved using regulatory sequences that are not chimeras of specific and nonspecific transcription factors. Important to the study presented here, both MYC-responsive promoters (11, 12, 13, 14) and enhancer sequences (15 , 16) have been identified. Initial studies with the unmodified ODC promoter (11) as well as with chimeric MYC-responsive enhancer/SV40 or Hsvtk promoter constructs (15 , 16) suggest that the ODC promoter, when combined with MYC-responsive enhancer sequences, might provide the specificity and efficiency needed to regulate transgene expression in the numerous MYC-expressing tumors (17, 18, 19, 20, 21, 22) .
MYC transcription factors include N-MYC, c-MYC and L-MYC (23)
. Advanced stage neuroblastomas often overexpress the N-MYC protein, with or without amplification of this gene (17)
. Several types of tumors including colon (18)
, breast (19)
, cervical (20)
, and brain (glioblastoma; Ref. 21
) frequently overexpress c-MYC. L-MYC was first identified in lung tumors (22)
. The MYC family of transcription factors forms heterodimers with MAX, another basic helix-loop-helix protein; and MYC-MAX dimers bind with high affinity to a CACGTG E-box sequence present within gene promoters (24)
. Several genes including the ODC gene (11, 12, 13, 14)
contain these E-box motifs in their regulatory sequences. Two isoforms of c-MYC proteins have been identified and characterized (25)
. Decreased expression or loss of the growth-inhibitory 67-kDa form of c-MYC and concomitant overexpression the 64-kDa growth-stimulating isoform has been observed in many tumor cells (26)
. Enhanced expression of MYC-regulated genes, such as ODC, -prothymosin, and Cdc25A, have been reported in primary tumors with elevated levels of MYC (27
, 28)
. Because levels of MYC in normal cells are very low, overexpression of this protein in tumors represents a potentially exploitable difference between tumor and normal tissue.
We demonstrated recently that N-MYC-regulated expression of a CE activates the prodrug CPT-11 (29)
. Those data showed that N-MYC activated the MYC-responsive ODC promoter, and IMR32 neuroblastoma transfectants were sensitized 
10-fold to CPT-11. Results from that study indicated that the ODC promoter/CE/CPT-11 combination might be useful in tumor cells that overexpress N-MYC, and also that it may be beneficial to modify the ODC promoter to retain its specificity to MYC family members but to increase the strength of the promoter to achieve higher CE levels and greater sensitization to CPT-11 or other enzyme/prodrug combinations. The study presented here investigates the possibility that modification of the endogenous ODC promoter with additional MYC-responsive elements might increase the efficiency of MYC/ODC promoter interaction and retain specificity for cells that overexpress N-MYC or c-MYC. Such a construct might be useful for VDEPT applications.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cell Lines
SJ-G2 and SJ-G3 pediatric glioblastoma cell lines were established at SJCRH. These cell lines grow as monolayers in DMEM (BioWhitaker, Walkersville, MD) supplemented with 15% fetal bovine serum (Hyclone, Logan, UT) and 2 mM glutamine in a humidified atmosphere of 10% CO2/90% air. The medium for the SJ-G3 cell line was supplemented with 50 µg of endothelial cell growth supplement (Upstate Biotechnology, Inc., Lake Placid, NY)/ml of medium. SW480 and HT-29 colon carcinoma cells were obtained from the American Type Culture Collection and were grown in RPMI 1640 and 10% fetal bovine serum in a humidified atmosphere of 5% CO2/95% air. The SJNB-4 neuroblastoma cell line was established from the tumor of a patient at St. Jude, in accordance with the policies of the SJCRH Internal Review Board. NB-1691 neuroblastoma cells were obtained from the Pediatric Oncology Group. JR1 rhabdomyosarcoma cells were obtained from Peter J. Houghton (SJCRH). N-MYC (57-kDa isoform) and c-MYC (64-kDa form) cDNAs were ligated into the EcoRI site of the plasmid pIRESneo (Clontech, Palo Alto, CA) and transfected into JR1 cells to create JR1 Nmyc6 and JR1 cmyc5 cell lines, respectively. MYC expression in these clones is regulated by the CMV promoter. JR1 cells transfected with the pIRESneo plasmid are designated JR1neo20 (29)
. Transfected JR1 cell lines were grown in RPMI 1640 containing 10% fetal bovine serum and 200 µg G418/ml in a humidified atmosphere of 5% CO2/95% air.
Plasmids
MYC-responsive Elements.
Sense and antisense oligonucleotides were synthesized by the Hartwell Center for Biotechnology (SJCRH). Each core sequence (CCCACCACGTGGTGCCT) contained a MYC-responsive E-box CACGTG motif. The 17-nucleotide oligomer above is that reported by Kumagi et al. (15)
. 68-mer or 102-mer oligonucleotides containing four or six repeats of the above oligonucleotide were annealed by heating to 90°C in 10 mM Tris-HCl (pH 8.0), 1 mM EDTA, and 50 mM NaCl and allowed to cool to room temperature. Duplex oligonucleotides having the same core sequence as above except that the CACGTG E-box sequence was replaced with CACCTG (CCCACCACCTGGTGCCT) to which MYC-MAX heterodimers bind poorly (28)
were also constructed and used as negative controls.
Plasmids containing the endogenous murine ODC promoter or mutated ODC promoter were generously provided by John L. Cleveland (SJCRH) and Philip Coffino (University of California, San Francisco, CA).
Construction of Vectors Containing Normal or Mutant ODC Promoter/MYC-responsive Enhancer Sequences for Reporter Activity Assays.
The pCAT3 Basic vector (Promega Corp., Madison, WI) encodes the CAT gene but lacks a promoter. ODC and 
ODC promoter sequences (above) were ligated into the XmaI site of the pCAT3 Basic plasmid to create pC3B.ODC.CAT and pC3B.ODC.CAT, respectively. The R4 or R6 MYC-response element sequences (above) and pC3B.ODC.CAT or pC3BODC.CAT were then digested with MluI or BglII or both, and R4 or R6 was inserted into pCAT3 plasmids either 5' (MluI site) or 3' (BglII site) or both 5' and 3' to the ODC promoter. Nomenclature for each construct is defined and schematic representations of each construct are shown in Fig. 1
. R4ODC, for example, contains the endogenous 414-bp ODC promoter with four additional CACGTG MYC binding sites 5' to the endogenous promoter (pC3B.R4ODC.CAT). Mutant sequences are designated as ODC or R6. An additional control plasmid containing the CMV promoter ligated into the MluI and NheI sites of pCAT3 Basic was also constructed.
|
For transient transfection studies, the CMV promoter of the plasmid pCIneo (Promega) was excised by digestion with IppoI and BglII. The ODC or R6ODC MYC-responsive elements were removed from the previously constructed pIRES plasmids using SstI and XhoI. Staggered ends of the restriction sites were filled in the T4 polymerase, and the promoter/enhancer elements were ligated to the linear promoterless pClneo plasmid by blunt end ligation with T4 ligase (Promega) to produce pCI.ODC.CE and pCl.R6ODC.CE constructs. Descriptions of the plasmids used in this study are shown in Table 1
.
|
CAT Reporter Activity Assays
The CAT assay was carried out using the Quan T kit from Amersham (Piscataway, NJ) according to the directions of the manufacturer. ß-gal activity was determined simultaneously with CAT activity to provide an internal control for transfection efficiency. Enzymatic activity of ß-gal was quantitated using the ß-gal enzyme assay system (Promega) as published previously (29)
. Data are expressed as A420 nm o-nitrophenol produced/mg protein/h. CAT activity was quantitated using [3H]acetyl-CoA as a substrate; results are expressed as dpm of [3H]acetylated chloramphenicol produced/h/mg protein. All data points recorded were in the linear range of each assay. Results are expressed as ratios of CAT activity:ß-gal activity.
Immunoblotting
This method has also been reported previously (29)
and follows in brief.
c-MYC Protein.
Degradation of MYC proteins was minimized by preparing samples rapidly at 4°C. Cells (106) were resuspended in 600 µl of sample buffer containing protease inhibitors, and the lysate was mixed immediately with 200 µl of 4x SDS-PAGE sample buffer. DNA was sheared by passing lysates through a 21-gauge needle 10 times, and the lysate was boiled for 5 min. Samples were loaded onto SDS-PAGE gels within 10 min after adding sample buffer to cells. Proteins were separated by standard SDS-PAGE techniques and transferred by semidry blotting to Immobilon-P membranes (Millipore, Bedford, MA). c-MYC protein was detected using clone 9E10 c-MYC antibody (Santa Cruz Biotechnology, Inc., Santa Cruz, CA). Immunoreactive proteins were visualized using ECL chemiluminescence detection reagents (Amersham) and X-OMAT AR film (Eastman Kodak, Rochester, NY). Films were scanned with an HP ScanJet ADF, and bands were quantitated using Imagequant software (Molecular Dynamics). Blots were stripped and reprobed with an anti-tubulin antibody (ICN Biomedical, Inc., Aurora, OH) as a loading control.
N-MYC and L-MYC Proteins.
The same method as above was used to detect N-MYC and L-MYC proteins. The primary antibody for detection of N-MYC was sc-142 and for L-MYC was sc-790 (Santa Cruz Biotechnology).
CE Activity Assay
Whole-cell sonicates were prepared using the method reported previously (30)
. The conversion of o-nitrophenyl acetate to nitrophenol was monitored spectrophotometrically at 420 nm. Protein concentrations were determined using Bio-Rad Protein Reagent (Bio-Rad, Hercules, CA) with BSA as the standard. Results are reported as µmol of nitrophenol produced/mg protein/min.
Growth Inhibition and Clonogenic Assays
Cells in log-phase growth were plated in 35-mm culture dishes and allowed to adhere overnight. Cells were then exposed to various concentrations of CPT-11 for 4 h and resuspended in drug-free medium for a time equivalent to five cell doubling times or exposed to G418 for five cell doubling times. Cell number was determined with a Coulter counter; the number of colonies was determined by an AlphaImager 2000, as reported previously (29)
. Results are reported as a percentage of cells or colonies compared with untreated controls.
Statistical Analysis
Data shown in Figs. 3
, 4
, and 6
and in Table 2
were analyzed using the Student’s t test. P < 0.05 was considered statistically significant.
|
|
|
|
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and described in "Materials and Methods." The endogenous ODC promoter contains two CACGTG E-box sequences. R4ODC, ODCR4, R4ODCR4, and R6ODC constructs thus contain a total of 6, 6, 10, and 8 of these E-box sequences, respectively. The construct encoding the mutated ODC promoter (ODC) and six additional mutated CACCTG binding sequences (R6) was used as a negative control. The R6ODC construct contains eight modified E-box sequences, to which MYC/MAX dimers have little or no binding affinity (28)
. All constructs contain the CAT gene.
Immunoblots for c-MYC and CAT Reporter Activity in SJ-G2 and SJ-G3 Glioblastoma Cell Lines
The relative transcriptional efficiency of the promoter/enhancer constructs described above were first evaluated in the c-MYC-expressing glioblastoma cell line SJ-G2. The cell line used as the negative control in these experiments was the SJ-G3 cell line. Immunoblots in Fig. 2
show that the 64-kDa form of c-MYC protein was readily detectable in SJ-G2 cells but not in SJ-G3 cells. Neither cell line expressed the 67-kDa form of c-MYC.
|
, each construct was ligated into pCAT3Basic, and SJ-G2 cells were transiently transfected with each of the plasmids. The cells were cotransfected with a plasmid encoding ß-gal to correct for transfection efficiency. Results are reported as CAT activity normalized to ß-gal activity, with activity regulated by the unmodified ODC promoter sequence arbitrarily set equal to 1.0.
Data in Fig. 3
show relative levels of CAT activity in lysates of cells transfected with plasmids containing the various MYC-responsive regulatory elements. The endogenous ODC promoter increased CAT activity 3-fold relative to promoterless controls. Four additional CACGTG sequences 5' to the endogenous ODC promoter (R4ODC) increased promoter activity 7.2-fold relative to the unmodified promoter (P < 0.005). Four additional MYC binding sites 3' to the ODC promoter or both 5' and 3' to the endogenous promoter produced CAT activity similar to the unmodified promoter. The highest level of CAT activity, 14-fold greater (P < 0.005) than the ODC promoter, was produced by constructs containing six additional MYC binding sites 5' to the ODC promoter (R6ODC). The negative control R6ODC sequence gave results equivalent to promoterless controls. SJ-G3 cells, which have no immunodetectable c-MYC, expressed only background levels of CAT activity when transfected with plasmids that contained either the R6ODC or the R6ODC sequence. We conclude that the R6ODC enhancer/promoter is the most efficient of the constructs tested in regulating expression of a reporter enzyme in the SJ-G2 glioblastoma cell line that overexpresses c-MYC.
Specificity of the R6ODC MYC-responsive Element to Cells That Overexpress MYC Family Proteins
Immunoblots for MYC Family Members in Colon and Rhabdomyosarcoma Cell Lines
To demonstrate that the results in Fig. 2
were not limited to a single cell line and to compare activation of the R6ODC sequence by N-MYC as well as by c-MYC, CAT assays were also done with several other cell lines that overexpress MYC. Immunoblots for c-MYC, N-MYC, and L-MYC expression in a variety of human tumor cell lines and in JR1 rhabdomyosarcoma cells had been transfected to stably express either N-MYC (Ref. 29
and Fig. 4
) or c-MYC (Fig. 4)
. The three JR1 cell lines were used to compare levels of CAT activity among isogenic cell lines that differed only in their level of 57-kDa N-MYC or 64-kDa c-MYC protein expression.
c-MYC.
SW480 and HT-29 colon carcinoma cells (Fig. 4)
as well as SJ-G2 cells (Fig. 2)
express the 64-kDa form of c-MYC. Also, JR1cmyc5 cells clearly express elevated levels of c-MYC, compared with JR1neo20 cells.
N-MYC.
We have published previously that SJNB-1 neuroblastoma cells express high levels of both the 54- and 57-kDa N-MYC isoforms of N-MYC, that JR1neo20 rhabdomyosarcoma cells express only the 54-kDa form of N-MYC, and that JR1Nmyc6 transfectants express both the endogenous 54-kDa form of N-MYC and also the transfected 57-kDa isoform (Ref. 29
and Fig. 4
).
L-MYC.
Blots for L-MYC were negative for all cell lines with the exception of SW480 cells.
On the basis of results in Figs. 2
and 4
and on the report by Pawlik et al. (29)
, cell lines were designated as MYC- (SJ-G3) and MYC+ (SJ-G2, SW480, HT29, JRcmyc5, SJNB-4, and JR1Nmyc6). Additionally, although JR1neo 20 cells clearly expressed the 54-kDa form of N-MYC and the 64-kDa form of c-MYC, they were designated here as MYC+/- to emphasize the relatively low levels of MYC proteins in this vector-transfected cell line compared with the JR1cmyc5 and JR1Nmyc6 cell lines.
CAT Reporter Activity in N-MYC- and c-MYC-positiveCell Lines
We next assessed the ability of the most efficient of the modified ODC constructs, R6ODC, to induce CAT activity specifically in cell lines that express c-MYC or N-MYC. CAT activity in cells transfected with the promoterless pCAT3Basic plasmid was normalized to a value of 1.0, and results in Fig. 5
are presented as fold-increase in CAT activity compared with promoterless controls. R6ODC controls for all five cell lines produced levels of CAT activity equal to mock transfectants and pCAT3Basic (data not shown).
|
2-fold increase in CAT activity compared with promoterless controls. In contrast, with all cell lines expressing readily detectable levels of c-MYC (64 kDa) or N-MYC (57 kDa), the R6ODC promoter/enhancer mediated a 50–300-fold increase in CAT activity relative to promoterless pCATBasic-transfected control (Fig. 5)
. All of the increases in CAT activity in MYC-positive cell lines were significant at the level of P < 0.05 (HT-29 and SJNB-4); P < 0.025 (JR1Nmyc6); and P < 0.005 (SJ-G2, SW480, and JR1cmyc5). Most informative were the three isogenic JR1 transfectants that differed only in the level of 57-kDa N-MYC or 64-kDa c-MYC protein expression. Results with these cell lines show that the R6ODC regulatory element was specifically activated by N-MYC and by c-MYC. Results with these cell lines also show that the 57-kDa isoform of N-MYC was very efficient in activating the R6ODC promoter/enhancer sequence compared with the 54-kDa form.
CAT Reporter Activity in L-MYC- and MyoD-expressing Cell Lines
As additional controls to examine the specificity of activation of R6ODC by c-MYC and N-MYC, we also assessed the ability of a rhabdomyosarcoma (Rh30) and a lung carcinoma cell line (H460) that overexpress MyoD or L-MYC, respectively, to activate the R6ODC promoter. MyoD and L-MYC are transcription factors of the same bHLH family as c-MYC and N-MYC. After transfection with pC3B.R6ODC.CAT, no CAT expression above background was detected in Rh30 cells or in H460 cells (data not shown).
Comparison of ODC- and R6ODC-mediated Transgene Expression
We next determined that the modified sequence produced higher levels of transgene expression than the endogenous promoter in all of the cell lines in Fig. 5
. The level of CAT activity in cells transfected pC3B.ODC.CAT was compared with that in cells transfected with pC3B.R6ODC.CAT. Data in Table 2
show that in all six cell lines having elevated levels of c-MYC or N-MYC, all cell lines transfected with the plasmid containing the R6ODC sequence had 10–50-fold higher levels of CAT activity than cells transfected with the plasmid containing the unmodified ODC promoter, with the single exception of JR1cmyc5 cells. JR1cmyc5 cells activated both the ODC and the R6ODC promoters relatively efficiently, resulting in a significant increase in CAT activity for both promoters but only a 5-fold difference between the two (Fig. 5
and Table 2
) in this cell line.
Comparison of ODC- and R6ODC-mediated Expression of Rabbit CE
The above reporter assays are informative in that the level of CAT activity is regulated directly by N-MYC or c-MYC and are a direct comparison of the strength of MYC-responsive elements. However, the long-range goal of our studies is to induce expression of a prodrug-activating enzyme to produce tumor cell-specific cytotoxicity (33)
. Therefore, to demonstrate the utility of the R6ODC sequence in inducing expression of a CE that activates CPT-11, we used the bicistronic pIRES.R6ODC.CE.neo vector and attempted to establish stable transfectants of SW480, SJ-G2, HT-29, IMR32, NB-1691, and SJNB-4 cells in which expression of rabbit CE and the neomycin resistance gene were under the control of the R6ODC regulatory element. With a single exception, repeated attempts to select stable transfectants with G418 failed. The exception was a stable SW480 cell line transfected with pIRES.R6ODC.CE.neo. The transfected cells expressed the neomycin resistance gene, because 200 µg of G418/ml killed all untransfected cells, but SW480.R6ODC.CE.neo cells survived exposure to >1000 µg G418/ml. Interestingly, however, rabbit CE was not expressed in these cells. PCR experiments showed that the cDNA encoding the rabbit CE was retained by the stable transfectants (data not shown), but CE activity was equal to untransfected cells. The data suggested that overexpression of CE may be toxic. We then transfected several MYC-expressing cell lines (U373, SW480, JR1cmyc5, and JRNmyc6) with the pIRES.R6ODC.CE plasmid and quantitated CE activity 24–72 h after transfection. CE activity increased transiently by 4–20-fold, but no stable transfectants could be established, again suggesting that overexpression of CE is cytotoxic. Experiments in which adenoviral vectors were used to deliver the cDNA encoding CE confirmed that overexpression of CE decreased the clonogenic survival of six of six human tumor cell lines (data not shown).
Because our long-range goal is tumor cell-specific cytotoxicity, the toxicity of CE overexpression could contribute positively toward tumor cell eradication. However, stable CE-expressing transfectants could not be used to compare the ability of the CMV and R6ODC promoters to regulate protein expression. Therefore, we performed reporter assays with pC3B.CMV.CAT and pC3B.R6ODC.CAT constructs and established stable transfectants with pIRES.CMV.neo and pIRES.R6ODC.neo plasmids to compare promoter efficiencies in specific human tumor cell lines.
Comparison of the CMV and R6ODC Promoters inc-MYC-expressing Cells
Comparison of CAT Activity Regulated by CMV andR6ODC Promoters
We inserted the CMV or the R6ODC promoter into the multiple cloning region of the pCAT3Basic vector and compared CAT activity of HT-29 and SW480 cells transfected with pC3B.CMV.CAT or pC3B.R6ODC.CAT. Data in Fig. 6
show that in HT-29 cells 1.4-fold more CAT activity was seen with the R6ODC compared with CMV promoter, whereas in SW480 cells the CMV promoter generated a 1.7-fold higher level of CAT activity than the R6ODC promoter. Therefore, although the levels of CAT activity mediated by the two promoters differed significantly (P < 0.01) in these two cell lines, the difference was <2-fold.
In Situ Comparison of the Ability of the CMV Promoter and R6ODC Regulatory Sequences to Induce Expression of the Neomycin Resistance Gene
We then replaced the CMV promoter in the parent pIRESneo vector with the R6ODC promoter to compare pIRES.CMV.neo and pIRES.R6ODC.neo constructs and assessed the ability of the CMV and the R6ODC promoters to up-regulate expression of the neomycin resistance gene in the absence of an inserted transgene cDNA. SW480, HT-29, and NB-1691 cells were transfected with either pIRES.CMV.neo or pIRES.R6ODC.neo and transfectants selected with a concentration of G418 that killed 100% of nontransfected cells. Transfected cells were plated for clonogenic assays, and dose-response curves were done to determine levels of the neomycin resistance gene expressed by transfected cell lines. Although a direct comparison of promoter strength in situ is difficult because cells differ in levels of transcription factors, transcription factor dimerization partners, and others that activate each promoter, results in Fig. 7
show that the CMV promoter and the R6ODC promoter/enhancer induced similar levels of expression of the neomycin resistance protein, as indicated by the overlapping cell survival curves over a range of G418 concentrations. We anticipated that the survival curves of each pair of cell lines would be similar at concentrations of G418 equal to or less than the concentration used to select transfectants (SW480, 300 µg/ml; HT29, 50 µg/ml; NB-1691, 750 µg/ml; Fig. 7
, arrows), but the data show further that the R6ODC sequence was comparable with the CMV promoter in regulating expression of the neomycin resistance gene in the three human tumor cell lines shown. We are investigating the utility of the R6ODC sequence in producing tumor cell-specific transgene expression in preclinical mouse models.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
10–50-fold more efficient than the endogenous ODC promoter in regulating reporter gene activity in five of six cell lines (Table 2
and Ref. 29
) and 50–250-fold more effective than promoterless controls in six of six cell lines. We propose that the modified ODC promoter may be of use in VDEPT approaches for the treatment of tumors that overexpress MYC transcription factors.
Other promoters have also been examined for their potential use in VDEPT therapy for specific tumors. These include glial fibrillary acidic protein for glial tumor cells (7)
, -fetoprotein for hepatocellular carcinoma (5)
, tyrosinase for melanoma (34)
, carcinoembryonic antigen for lung cancer (4)
, DF3 for multiple myeloma (Ref. 2
; reviewed in Ref. 35
), and midkine for Wilms’ tumor (36)
. From the viewpoints of tumor specificity and promoter strength, use of the DF3 promoter to regulate expression of a prodrug-activating enzyme for use in purging bone marrow prior to autologous transplant of multiple myeloma patients seems promising (2)
.
There are also several studies in the literature demonstrating that MYC-responsive elements can enhance the activity of the CMV, SV40, or Hsvtk viral promoters (15
, 16
, 27
, 28)
. However, because these promoters are activated in normal cells as well as tumor cells, chimeric MYC-responsive enhancer/viral promoter sequences may not have sufficient tumor cell specificity for VDEPT applications. Additionally, viral promoters can be silenced in mammalian cells, thereby limiting their efficacy in situ (37)
. Of potential relevance to the construction of efficient, tumor-specific regulatory elements, Pang et al. (9)
determined that the efficiency of the PSA promoter or of a chimeric CMV/PSA promoter in regulating expression of a luciferase reporter gene was 10% that seen with the CMV promoter. In contrast, a PSA enhancer/PSA promoter construct was 90% as efficient as the CMV promoter (10)
, suggesting the possibility that chimeric promoter/enhancer elements may be inferior to regulatory elements in which both promoter and enhancer sequences respond to the same stimulatory factors.
Also of relevance to the usefulness of the reported results to VDEPT applications are observations by us and by others that human tumor cells that overexpress c-MYC frequently express only the smaller 64-kDa isoform (26)
. [We note that there is little agreement in the literature as to the exact molecular mass of c-MYC and N-MYC proteins. Because the bands on our immunoblots migrated at 57 and 54 kDa for N-MYC and at 67 and 64 kDa for c-MYC, we have used the nomenclature of Bovini et al. (38)
for N-MYC and of Eisenman and Hann (39)
for c-MYC.] For eventual application to c-MYC-expressing tumors, therefore, it was necessary to determine whether the 64-kDa isoform activated the promoter/enhancer constructs of interest. In contrast, immunoblots for N-MYC in 10 human neuroblastoma cell lines (29)
4
showed that both the 57-kDa and 54-kDa forms of N-MYC are expressed at approximately equal levels in all 10 cell lines. Because results published previously showed that the 57-kDa isoform of N-MYC activated the endogenous ODC promoter more efficiently than the 54-kDa isoform (29)
, the 57-kDa isoform was transfected into the JR1 cell line for reporter activity assays (Fig. 3)
. We also did preliminary assessments of the ability of lung tumor cells that express L-MYC and rhabdomyosarcoma cells that express MyoD to activate the R6ODC promoter (data not shown). Although L-MYC and MyoD are bHLH transcription factors and are structurally and functionally similar to N-MYC and c-MYC, neither L-MYC nor MyoD activated the R6ODC promoter. This observation is similar to that of Tobias et al. (40)
, who reported that another closely related bHLH transcription factor, USF, bound to but failed to activate the endogenous ODC promoter. Therefore, our conclusions regarding promoter activation are restricted to tumor cells that overexpress c-MYC or N-MYC.
We also noted that levels of transgene expression did not necessarily correlate with the level of MYC protein. As a potential explanation for the varied levels of transgene expression, we did immunoblots for MAX and MAD1 proteins for all of the cell lines used in the study, reasoning that low levels of MAX or overexpression of MAD would limit transcriptional activation by MYC family members. However, the immunoblots showed that levels of MAX and MAD1 varied only by a factor of 2 among all of the cell lines (data not shown), and no correlations regarding ratios of MYC:MAX, MAD1:MAX, or MYC:MAD1 with reporter gene expression were evident.
In summary, data presented here suggest that a modified ODC promoter may be useful in achieving tumor-specific expression of transgenes of interest in tumor cells that overexpress c-MYC or N-MYC. Studies to explore the potential for use of the combination of R6ODC.CE delivered by a adenoviral vector and systemic administration of CPT-11 in a VDEPT approach are ongoing.
|
FOOTNOTES |
|---|
1 Supported in part by NIH Grants CA-23099, CA-63512, CA-76202, CA-79763, CA-66124, and Cancer Center Core Grant P30-CA-21765 and the American Lebanese and Syrian Associated Charities. 
2 To whom requests for reprints should be addressed, at Department of Molecular Pharmacology, St. Jude Children’s Research Hospital, 332 North Lauderdale, Memphis, TN 38105. Phone: (901) 495-3440; Fax: (901) 521-1668; E-mail: mary.danks{at}stjude.org
3 The abbreviations used are: VDEPT, virus-directed enzyme prodrug therapy; PSA, prostate specific antigen; CAT, chloramphenicol acetyltransferase; CE, carboxylesterase; CMV, cytomegalovirus; CPT-11, irinotecan, 7-ethyl-10-[4–1-piperidino)-1-piperidino]carbonylcamptothecin; Hsvtk, Herpes simplex virus thymidine kinase; ODC, ornithine decarboxylase; SJCRH, St. Jude Children’s Research Hospital; R4, four repeats of a 17-bp sequence containing the MYC CAGGTG binding sequence; R6, six repeats of a 17-bp sequence containing the MYC CACGTG binding sequence; SN-38, 7-ethyl-10-hydroxycamptothecin; ß-gal, ß-galactosidase; bHLH, basic helix-loop-helix.
Received 9/19/00. Accepted 1/31/01.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B, and Myc/Max. Oncogene, 18: 2728-2738, 1999.[Medline]
This article has been cited by other articles:
|
|
 |
|
  D. Stearns, A. Chaudhry, T. W. Abel, P. C. Burger, C. V. Dang, and C. G. Eberhart c-Myc Overexpression Causes Anaplasia in Medulloblastoma Cancer Res., January 15, 2006; 66(2): 673 - 681. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
