Division of Applied Pharmacology Research, Office of Testing and Research, Office of Pharmaceutical Sciences, Center for Drug Evaluation and Research, Food and Drug Administration, Laurel, Maryland 20708
Received December 20, 2002; accepted April 9, 2003
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key Words: Tg.AC; gadd153; c-fos; -globin; HepG2; K562.
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
For increased understanding of and appreciation for the predictive value of the Tg.AC model, it is important to provide assurance of assay specificity (the appropriate lack of response to properly dosed noncarcinogens) through the careful selection and testing of noncarcinogenic drugs in this assay. A consortial evaluation of the Tg.AC assay and other alternative in vivo model systems, organized by the International Life Sciences Institute (ILSI), was initiated to provide mechanistic information on the response of these assays to several different modes of carcinogenesis, including species-specific mechanisms (Robinson and MacDonald, 2001). One rodent noncarcinogen (sulfisoxazole) has been tested in the Tg.AC mouse by the ILSI consortium. The study described in this paper was designed to complement the ILSI effort through the additional testing of noncarcinogens that might potentially elicit "false positive" activity. A potential false positive response in the Tg.AC assay could theoretically be generated by noncarcinogenic drugs that, through receptor or enzyme modulation, stimulate skin proliferation or hair growth, for example, since it is known that full-thickness wounding or repeated depilation induces papillomas in Tg.AC mice (Cannon et al., 1998
; Hansen and Tennant, 1994
).
Based on the structure of the Tg.AC transgene, the reporter phenotype of the Tg.AC mouse, and the restricted expression of the transgene in untreated Tg.AC mice to tissues involved in erythropoiesis, it could be deduced that the generation of skin papillomas results from activation of the -globin promoter and the resultant induction of transgene expression. Therefore, noncarcinogenic drugs that can induce signal transduction pathways that lead to stimulation of the transcriptional activity of
-globin promoter directly or through upstream enhancer elements could potentially generate positive responses in the Tg.AC assay. Based on this reasoning, in vitro promoter/enhancer element induction assays were viewed as potentially useful rapid throughput screens for identifying potential false positive drugs in the Tg.AC assay. Activation of the
-globin promoter can be measured by placing this sequence in front of a reporter gene in a permissive cell line. Reporter gene assays can also measure the effect of drug treatment on enhancer sequences such as AP-1 and p53 response elements that could reside upstream of the transgene integration site and be activated in response to tumorigen treatment.
Four rapid throughput in vitro reporter gene assays were selected for their potential mechanistic links to papillomagenesis in Tg.AC mice and analyzed using 24 Tg.AC tested compounds (Thompson et al., 2000). One assay measured induction of the
-globin gene promoter in K562 cells. The other three assays examined were Chloramphenicol acetyltransferase- (CAT)Tox (L) assays that measured induction of the gadd153 promoter, c-fos promoter, and p53 response elements in HepG2 cells. Of the four assays, the gadd153-CAT assay showed the strongest overall concordance (81%) with activity in the Tg.AC assay, correctly classifying 78% of Tg.AC positive and 83% of Tg.AC negative compounds. The correlation was further improved to 86% by adding a discriminator of a strong response (eightfold or greater) in a
-globin promoter reporter assay as a second screen. Gadd153 can produce a strong transcriptional response to mechanistically diverse inducers of cellular injury (Luethy and Holbrook, 1992
) and has been used in screening assays for cellular toxicants (Lin et al., 2001
; Todd et al., 1995
). The fos-CAT, Zeta-Luc, and p53-CAT assays correctly classified 67%, 64%, and 36%, respectively, of Tg.AC positive compounds.
In this report, the three in vitro assays shown to have the highest correlation with Tg.AC activity were used to screen 99 noncarcinogenic pharmaceuticals in order to select and prioritize drugs that might challenge the specificity of the Tg.AC assay. For high throughput screening purposes, cell lines were established by stable transfection of plasmid constructs containing the gadd153, c-fos, or -globin promoters linked to luciferase or green fluorescent protein (GFP) reporter genes. The drugs screened had tested negative for carcinogenic activity in male and female rats and mice and were available from commercial or internal sources. Three candidate drugs for subsequent in vivo testing in a Tg.AC assay were selected by evaluation of the resultant data.
![]() |
MATERIALS AND METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Reporter gene constructs.
CAT reporter plasmids containing hamster GADD153 promoter sequences from 778 to +21 bp (p9000) or minimal gadd153 promoter sequences from 36 to +21 bp alone (p9025) were obtained through a material transfer agreement from Dr. Nikki Holbrook, National Institute of Aging, Baltimore, MD. The 799 bp gadd153 promoter sequence was excised from p9000 with EcoRI and HindIII and subcloned into the promoterless GFP vector pGFPemd-b (Packard Instrument Company, Meriden, CT) to make the plasmid gadd153-GFP. The same gadd153 promoter sequence was also inserted into a luciferase plasmid to make the plasmid gadd153-Luc as follows. After p9000 was digested with EcoRI, the EcoRI site was filled in using Klenow polymerase, followed by digestion with HindIII. The resulting promoter fragment was ligated into a SmaI/HindIII-digested pGL3-Basic vector (Promega, Madison, WI). A luciferase plasmid containing three copies of the gadd153 AP-1 site linked in tandem to its minimal promoter (3xAP-1-Luc) was constructed from the plasmid 3xAP-1-CAT (from Dr. Holbrook) by digestion with BglI, blunting the overhanging ends with T4 DNA polymerase, and redigesting with Hind III. The purified promoter sequence was ligated into a SmaI/HindIII double-digested pGL3-Basic plasmid.
The 0.95 kb mouse -globin promoter was excised from the vector Zeta-Luc (Thompson et al., 2000
) with KpnI/HindIII and subcloned into the vector pGFPemd-b to create the plasmid Zeta-GFP.
A pBR322-based plasmid containing the complete coding sequence of the human fos proto-oncogene (pc-fos(human)-1) was purchased from American Type Culture Collection (Rockville, MD). After the plasmid was linearized with EcoRI, the c-fos promoter sequence from 451 to +40 bp was amplified by polymerase chain reaction (PCR) using primers with 5' restriction enzyme sites for EcoRI and BamHI with the sequences: CGGAATTCCGAATGTTCTCTCTCATTCTGC and CGGGATCCCGAGTCTTGGCTTCTCAGTTG. The resultant PCR product was digested with EcoRI and BamHI and ligated into an EcoRI/BamHI-digested pGFPemd-b vector to make the plasmid fos-GFP. The plasmid fos-Luc was constructed by subcloning the promoter from KpnI/HindIII-digested Fos-GFP into a KpnI/HindIII-digested pGL3-Basic vector.
The correct identity of each plasmid construct was confirmed by fluorescence-based DNA sequencing.
Cell lines.
HepG2 cell lines were grown at 37°C in 95% air:5% CO2 and Minimum Essential Medium (MEM) with Earles salt and nonessential amino acids, supplemented with 10% heat-inactivated fetal calf serum (FCS), 1 mM sodium pyruvate, 2 mM L-glutamate, and penicillin-streptomycin. HepG2 cells were transfected with a 2:1 ratio of reporter plasmid (gadd153-Luc, gadd153-GFP, fos-Luc, or 3xAP-1-Luc) and a Rous sarcoma virus promoter-neomycin resistance gene construct (RSV-neo) using FuGENETM 6 transfection reagent (Roche Molecular Biochemicals, Indianapolis, IN) and subsequently placed under selection with 0.3 µg/ml Geneticin (Life Technologies) for one month. To obtain clonal cultures, the cells were plated at a dilution of 1 per well in 96-well plates along with 104 nontransfected HepG2 cells. After growth for 1 week, selection with 0.5 mg/ml Geneticin was begun. The resulting clonal cultures were assayed for basal levels of reporter gene activity and levels induced by TPA. The selected clonal cultures were designated HepG2(gadd153-Luc), HepG2(gadd153-GFP), HepG2(fos-Luc), or HepG2(3xAP-1-Luc). We were unable to isolate a stable line of HepG2 cells transfected with a GFP vector construct carrying the c-fos promoter, presumably because the expressed levels of GFP were toxic to the cells.
K562 cells lines were grown in Roswell Park Memorial Institute 1640 media (RPMI) supplemented with 10% FCS and penicillin-streptomycin. K562 cells (5 x 105/ml) were washed with PBS and transfected with Zeta-GFP and RSV-neo at a 2:1 ratio using DOTAP reagent (Roche Molecular Biochemicals, Indianapolis, IN) in RPMI buffered with N-(2-hydroxylethyl)piperazine-N'-(2-ethanesulfonic acid) (HEPES), pH 7.0, for 4 hours and then replated in RPMI with 10% FCS. After 5 days incubation, the cells were replated in media containing 0.5 mg/ml Geneticin. After 9 days incubation, the cells were plated out singly in the wells of 96-well plates. After two weeks, wells containing single colonies were replated, expanded, and tested for basal and TPA-inducible green fluorescence. A clonal culture with low basal and high TPA-inducible GFP expression was selected and labeled K562(Zeta-GFP).
Measurement of fos-Luc or gadd153-Luc activity in HepG2 cells.
HepG2(fos-Luc) or HepG2(gadd153-Luc) cells were plated in 24-well plates at 35 x 104 cells/well. On day 2, stock solutions of drugs were prepared in ethanol, dimethylsulfoxide (DMSO), or MEM without FCS. Doses were selected using the viability curve as a guide, with the aim of reaching a 50% reduction in viability after 2 days with the higher doses. The maximum percent of ethanol or DMSO allowed was 5%. Stock solutions were diluted into MEM with 10% FCS to make dosing solutions, which replaced the plating media on the HepG2 cells, and the plates were returned to the incubator for 48 h. For a positive control, 1 µg/ml mezerein in MEM/10% FCS was added to duplicate wells. Mezerein typically induced fos-Luc activity by 8.3-fold ± 4 (n = 10) and gadd153-Luc activity by 3.4-fold ± 0.6 (n = 9).
On day 4, the media was removed from the wells, which were washed twice with phosphate-buffered saline (PBS). Each well received 200 µl of 1x Passive Lysis Buffer (Promega) After incubating for 15 min on a rocking platform, the lysate was transferred into microfuge tubes and snap frozen. The lysates were thawed and centrifuged at 4°C for 12 min at 13,000 x g.
Luciferase activity was measured using the Promega Luciferase Assay System (Madison, WI) in a Monolight 2010 luminometer (Analytical Luminescence Laboratories, San Diego, CA). Protein concentrations of the lysates were measured using Coomassie Plus Protein Assay Reagent (Pierce, Rockford, IL). Luciferase activity normalized for protein concentration was calculated as the average of two replicate wells for each drug dose.
Measurement of gadd153-GFP activity in HepG2 cells.
HepG2(gadd153-GFP) cells were plated at 3 x 104 cells /well in 24-well dishes. Triplicate wells were plated for each data point. For the standard curve, four wells were plated with 1, 2, 3, or 4 x 104 cells/well. One well per plate was left blank for normalizing for plate fluorescence.
Cells were dosed as described above. For a positive control, 1 µg/ml mezerein in MEM/10% FCS was added to duplicate wells. Mezerein typically induced gadd153-GFP expression by 4.2-fold ± 1.1 (n = 9). After 48 hours, the media was removed, wells were washed once with PBS, and 0.5 ml phenol-red free MEM plus 1% bovine serum albumin (BSA) was added to each well (including blank wells). Fluorescence was measured in a CytoFluor 4000 plate reader (PE Biosystems, Foster City, CA) using a 485/20 excitation filter and a 530/25 emission filter at a gain setting of 98. The media was subsequently replaced with phenol-red free MEM with 1% BSA and 10 µg/ml Hoechst 432. The plate was covered with foil and placed in a CO2 incubator for 20 min. The dye solution was removed, and the wells were washed with PBS. The PBS was replaced with phenol-red free MEM containing 1% BSA (0.5 ml per well including the blank well). Fluorescence per well was measured using a 360/40 excitation filter and a 460/40 emission filter at a gain setting of 80. Hoechst 432 fluorescence was converted to cell number from the standard curve.
Measurement of Zeta-GFP activity in K562 cells.
K562(Zeta-GFP) cells were plated at 3 x 105/ml in phenol red-free RPMI with 10% FCS at a volume of 0.5 ml/well in 24-well dishes using duplicate wells per data point. Drug solutions were prepared as above at 2x concentrations. Wells were dosed with 0.5 ml of drug solution per well. For the positive control, 0.5 ml of 2 nM mezerein was added to triplicate wells. Mezerein typically induced Zeta-GFP activity by 12.1-fold ± 2.7 (n = 8). Plates were replaced in a CO2 incubator for 48 h. Subsequently, cells were gently resuspended, and 0.4 ml of the sample was transferred to a counting vial. After dilution to 20 ml with Isoton II (Coulter Corporation, Miami, FL), cell counts were measured in a Coulter Counter.
For each well, a second 0.4-ml aliquot of cells was removed to a microfuge tube and centrifuged at 4°C for 2 min at 1900 x g. The cell pellet was resuspended in 200 µl phenol-red free RPMI containing 1% BSA and replated in a 96-well plate. Blank wells containing media alone were also plated for normalization. Plate fluorescence was read in a CytoFluor 4000 plate-reader using a 485/20 excitation filter and a 530/25 emission filter. The results were calculated as fluorescence per 106 cells. For drugs with autofluorescence (e.g., dipyridamole), control wells plated with the parental K562 cell line received the same doses as K562(Zeta-GFP) cells. Autofluorescence levels were normalized to cell number and subtracted from the corresponding K562(Zeta-GFP) cell samples.
Cellular toxicity assay.
K562(Zeta-GFP) cells were plated at 3 x 105/ml in 96-well dishes at 100 µl per well. Drug solutions were prepared in RPMI + 10% FCS, and 100 µl aliquots were added to cells. The plates were incubated for 48 h at 37°C in 95% air:5% CO2. Cell survival was measured using the CellTiter 96 Aqueous nonradioactive cell proliferation assay (Promega).
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Prior to testing in the reporter gene assays, an appropriate dose range for each drug was determined based on K562 cell viability after two days incubation, as measured with a MTT colorimetric assay. Each drug was tested in the three reporter gene assays over a dose range that reached induction of toxicity, or if no toxicity was observed, the drug was tested up to maximal doses of 510 mM. A positive response in each of the three reporter gene assays for each of the 100 drugs tested was judged using the same criteria as in the original evaluation of in vitro correlates of Tg.AC activity (Thompson et al., 2000). In that study, the discriminant ratio of response in the gadd153-CAT assay was calculated to be twofold or greater. Therefore, for this data set, a twofold or greater maximum induction of reporter gene activity at a drug concentration that did not decrease cell number or protein concentration in cell lysates by more than 40% below control values was called a positive response. Most of the drugs were first screened in the gadd153-GFP assay, due to the fact that it is less labor-intensive than the gadd153-luciferase assay because reporter expression can be measured without cell lysis. All strong positives in the GFP-based assay were retested in the luciferase-based assay to screen out false positives due to effects on cellular pH. The variant of GFP used in these studies (S65T) has a pKa of 6.0 (Elsliger et al., 1999
), so intracellular alkalinization could produce an increase in fluorescence in the absence of a change in the rate of transcription of the GFP gene. Drugs that induced GFP fluorescence in the gadd153-GFP assay but did not increase luciferase activity in the gadd153-Luc assay were deprioritized for in vivo testing.
About 20% of the drugs tested were positive in the gadd153 promoter induction assays (Table 1). Higher positive response rates were observed with the set of 99 noncarcinogenic drugs in the two in vitro assays that had lower correlation with Tg.AC activity than in the gadd153 promoter induction assay. Overall, 40% of the drugs tested were positive in the Fos-Luc assay and 50% were positive in the Zeta-GFP assay. Induction of the
-globin gene promoter in K562 cells correlated in some cases with the pharmacologic class of the drug. For example, all eight antihistamines and six of seven beta-blockers tested were positive in the Zeta-GFP assay, while eight of 12 nonsteroidal anti-inflammatory agents were negative. Induction of the gadd153 or c-fos promoters did not show a similar correlation with pharmacology.
For in vivo testing candidates, we focused on the drugs with the strongest response (fourfold or greater) in the gadd153 promoter assay. The gadd153 promoter has a large dynamic range, demonstrating a greater than 100-fold induction ratio with some DNA damaging agents when assayed as a fusion construct with a CAT reporter gene in HeLa cells (Luethy and Holbrook, 1992). Compounds that are inhibitors of growth, DNA synthesis, or cell division typically induce gadd153-CAT activity in HeLa cells by fourfold or less (Luethy and Holbrook, 1992
). The seven Tg.AC positive compounds we had previously found to be active in the gadd153-CAT assay had an average maximum fold induction of 3.9 ± 1.5 (mean ± S.D.) (Thompson et al., 2000
). Of the 99 drugs tested in this study, nine drugs induced gadd1533 promoter activity by fourfold or greater. These drugs are listed in Table 2
along with their maximal fold inductions in a gadd153-GFP or gadd153-Luc assay, the dose at which the maximal induction was observed, and the percent viability of the treated cells at time of harvest, as assayed by total protein or DNA measurements. The observed maximal fold inductions in the fos-Luc and Zeta-GFP assays are also reported. All of the drugs tested that gave the strongest induction of the gadd1533 promoter also induced the c-fos promoter by twofold or greater, suggesting that there may be common elements in the signal transduction pathways that induce the gadd153 promoter and c-fos promoter in HepG2 cells. However, only half of the gadd153-inducing drugs also induced
-globin gene promoter activity, which is the same proportion of Zeta-GFP positives observed among the entire group of 99 noncarcinogenic drugs. In an earlier study, increased correlation with Tg.AC activity had been demonstrated among compounds positive in a gadd153 promoter induction assay that also produced a high (eightfold or greater) induction of Zeta-Luc activity in a second-stage assay (Thompson et al., 2000
). However, of the five drugs that tested positive in both gadd153 and
-globin gene promoter induction assays (Table 2
), none induced the
-globin gene promoter by eightfold or more, so no such added weight was given to a positive response in both assays.
|
Several important factors were considered in the selection of the three best candidates for testing the specificity of the Tg.AC assay amongst the nine noncarcinogenic drugs that demonstrated the strongest in vitro induction of the gadd153 promoter. A target concentration for daily skin paint application was calculated for each drug based on maximally tolerated doses (MTD) or maximally feasible doses in a 2-year carcinogenesis bioassay in mice calculated in mg/kg/day. It was first determined whether the drug was sufficiently soluble in a solvent that was considered suitable for a skin paint study to achieve this target dose. Mebendazole, for example, was excluded from consideration for in vivo testing based on poor solubility in acetone and in ethanol. Second, the cost of administering each drug for 26 weeks at the target concentration and at a lower submaximal dose could not be prohibitive. Third, a higher weight was given to drugs with the highest potency in inducing the gadd153 promoter. Fourth, a lower priority was given to drugs of similar activity or class to those that had been previously tested or were undergoing testing in Tg.AC mice as part of the ILSI initiative.
The three candidate drugs that meet all of the selection criteria for in vivo testing in Tg.AC mice were amiloride, dipyridamole, and pyrimethamine. Amiloride is an antikaliuretic-diuretic agent, dipyridamole is approved for clinical use as a platelet adhesion inhibitor in combination therapies, and pyrimethamine is a folic acid antagonist that is indicated for treatment of toxoplasmosis and malaria in combination with sulfonamides (Physicians Desk Reference, 2002). None of the three drugs showed carcinogenic activity in chronic feeding studies in rodents. No evidence of tumorigenic activity was observed in mice dosed for 111 weeks with dipyridamole or with amiloride for 92 weeks (Physicians Desk Reference, 2002). The duration of dosing with pyrimethamine in a National Toxicology Programsponsored bioassay was 78 weeks followed by 2627 weeks of additional observation (National Toxicology Program, 1978). No statistically significant incidence of tumors was observed in female B6C3F1 mice or in male and female Fisher rats. The rats were dosed to toxicity, as evidenced by the significant levels of bone marrow atrophy. A reduced life span in both dosed and matched-control male mice groups prevented potential observation of late-appearing tumors; therefore, the carcinogenic potential of pyrimethamine to male mice could not be assessed in this study.
Representative induction profiles for each of the three drugs in the gadd153-GFP (or gadd153-Luc) assays, the fos-Luc assays, and the Zeta-GFP assays are shown in Figures 13. Superimposed on each induction profile plot are the dose response curves of drug toxicity, measured by cell number or protein concentration from a portion of the same sample used for the induction measurements. Amiloride was a very strong inducer of the gadd153 and c-fos promoters, with no effect on
-globin gene promoter activity (Fig. 1
). Dipyridamole increased the expression of the gadd153-GFP and fos-Luc reporters to similar extents, inducing maximal protein or enzyme activity levels about tenfold above control levels (Fig. 2
). The effect of dipyridamole on Zeta-GFP expression was moderate, with a 3.5-fold maximal induction observed. Pyrimethamine significantly inhibited cell growth at low doses (510 µM) in both HepG2 and K562 cell lines (Figs. 13
). With this compound, the promoter induction and cell viability curves reached a plateau between doses of 25 and 100 µM. Within this dose range, pyrimethamine induced expression by approximately fivefold in all three of the promoter induction assays.
|
|
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In our study, gadd153 induction was not a general response to cytotoxicity but was selective for certain pathways of cellular injury, because, although >95% of the drugs induced toxicity or growth arrest at the highest doses tested, only 9% induced gadd153 promoter activity by fourfold or more. Induction of gadd153 promoter-driven luciferase activity by > twofold was highly correlative with in vitro activation of the c-fos promoter. Ninety-four percent (16/17) of drugs positive in the gadd153-Luc assay were also positive in the fos-Luc assay (Table 1). The AP-1 binding site in the gadd153 promoter has been shown to be involved in the response of this promoter to oxidative stress (Guyton et al., 1996
). These results suggest that an earlier step in the signal transduction pathway that leads to gadd153 promoter activation by the drugs in this study involves activation of the c-fos promoter, which results in increased production of fos protein, one of the components of AP-1, and is followed by activation of the gadd153 promoter through its AP-1 binding site. We have observed that amiloride, dipyridamole, and pyrimethamine induce luciferase activity directed by a promoter consisting of three copies of the gadd153 AP-1 site linked to a minimal gadd153 promoter with a similar dose response range as observed for induction from the fuller length (799 bp) gadd153 promoter (data not shown). There is substantial evidence that AP-1 is an important component in mouse skin tumor development and in the response to wounding (Angel et al., 2001
). Mice that are double transgenics for v-fos and v-Ha-ras expression targeted to the epidermis develop benign papillomas at an increased rate and earlier onset (Greenhalgh et al., 1993
). The induction of papillomas in Tg.AC mice by full thickness wounding or repeated depilation is thought to be due to the resulting proliferative response of follicular epithelial cells (Sistare et al., 2002
), which may promote sustained dual expression of AP-1 and v-Ha-ras in these cells. Therefore, noncarcinogenic drugs that are able to induce AP-1 activity in follicular cells upon dermal application may elicit a false positive response in the Tg.AC assay.
No correlation was observed between activation of the -globin gene promoter and activation of the gadd153 or c-fos promoters. Only 53% (9/17) of drugs positive in the gadd153-Luc assay were also positive in the Zeta-Luc assay (Table 1
). Although the Tg.AC transgene contains
-globin gene promoter sequence, the nonerythroid site of papilloma origin (i.e., hair follicles) and the apparent requirement for palindromically oriented promoter sequence (Honchel et al., 2001
; Thompson et al., 1998
) suggest that direct activation of the
-globin gene promoter may be necessary but not sufficient for formation of transgene-expressing skin papillomas. Therefore, our previous finding of a low correlation between Tg.AC papilloma induction and
-globin promoter activation in vitro was not surprising (Thompson et al., 2000
).
This study was designed to identify drugs that, while not positive in chronic dosing studies in rodents, might induce papillomas in Tg.AC mice based on their ability to induce certain stress responses in vitro. This approach may conceivably fail because of greater sensitivity of cultured cells to the toxic effects of pharmaceuticals compared to follicular skin cells of the Tg.AC mouse or, alternately, if organ-specific toxicities limit dose levels in the Tg.AC assay prior to achievement of application site tumorigenic activity. Additionally, differences in metabolic competence, in expression of receptors and signal transduction components, and in the extracellular environment between homogenous cultured HepG2 cells and complex mouse skin could factor into the degree of correlation observed between in vitro and in vivo outcomes.
Three drugs (amiloride, dipyridamole, and pyrimethamine) were selected based on their in vitro induction of gadd153 promoter activity and solubility in ethanolic solutions at concentrations that, with daily skin paint application, would approach MTDs determined in 2-year carcinogenicity bioassays in mice. Although the three drugs have different therapeutic classifications and molecular mechanisms of pharmacologic action, there is nonetheless a plausible basis for the observed induction of the gadd153 promoter in HepG2 cells by each of these drugs.
Amiloride is a potassium-sparing diuretic that acts directly on distal renal tubules to inhibit sodium/potassium exchange (McEvoy, 1998). It also inhibits hydrogen ion exchange. In HepG2 cells, which require Na+/H+ exchange activity for growth (Strazzabosco et al., 1995
), amiloride decreases intracellular pH and inhibits DNA synthesis (Garcia-Canero et al., 1999
; Strazzabosco et al., 1995
). The mediator of growth inhibition by amiloride may be ribonucleotide reductase, an enzyme critical for DNA synthesis, which is selectively dependent on Na+/H+ exchange activity (Vairo et al., 1992
). These effects on cell cycle blockage may account for the induction of gadd153 expression by amiloride.
Dipyridamole has pharmacologic activity as a coronary vasodilator and as an inhibitor of platelet adhesion (McEvoy, 1998; Physicians Desk Reference, 2002). The mechanism of vasodilation by dipyridamole is thought to be through inhibition of phosphodiesterase. Dipyridamole also inhibits the uptake of adenosine into platelets, endothelial cells, and erythrocytes in vitro and in vivo. An increase in local concentrations of adenosine may stimulate adenylate cyclase through the platelet A2-receptor, and through elevation of cAMP levels, may cause inhibition of platelet aggregation in response to various stimuli. Dipyridamole can also reverse multidrug resistance in cells overexpressing P-glycoprotein or multidrug resistance-associated protein (MRP), drug efflux pumps for cytotoxic proteins (Asoh et al., 1989
; Curtin and Turner, 1999
). Dipyridamole does not appear to work as a direct blocker of MRP because reversal of resistance of MRP-overexpressing cells to chemotherapeutic drugs can be induced by dipyridamole in the absence of any significant effect on drug accumulation or efflux (Curtin and Turner, 1999
). Instead, intracellular glutathione levels, which can regulate drug efflux by MRP (Versantwoort et al., 1995
), are depleted in MRP-overexpressing cells when chemosensitivity is restored by dipyridamole treatment. The toxicity of dipyridamole (and the induction of the gadd153 promoter) in HepG2 cells, which have high levels of MRP2 (Morrow et al., 2000
), could therefore be a result of depletion of intracellular glutathione.
Pyrimethamine is an antimalarial drug that, as a folic acid antagonist, inhibits plasmodial dihydrofolate reductase. The depletion of nucleotide pools that results from impairment of purine and pyrimidine biosynthesis can lead to errors in DNA synthesis and an increase in DNA and chromosomal lesions. Pyrimethamine is not mutagenic in the Ames assay, Rec assay, or Escherichia coli WP2 assay (Physicians Desk Reference, 2002), but there is evidence of clastogenic activity in vivo and in vitro (Egeli, 1998; Ono et al., 1997
). The induction of gadd153 by pyrimethamine in HepG2 cells is most likely a response to such DNA damage induced by high concentrations of the drug.
This study constitutes a novel approach to select compounds to challenge the specificity of an alternate short-term assay like the Tg.AC assay for pharmaceuticals. Similar to preinitiated mouse skin, the Tg.AC transgenic model contains an oncogene (v-Ha-ras) that requires the cooperation of additional signals for tumor development (Compere et al., 1988). However, the extent and nature of the involvement of the
-globin promoter sequence in the transgene in Tg.AC skin papilloma formation remains unknown (Sistare et al., 2002
). This aspect of the model has raised questions about appropriate specificity, especially towards chemicals with pharmacologic activity. To increase our knowledge base of the response of noncarcinogenic pharmaceuticals in the Tg.AC assay, we first identified a high throughput in vitro reporter gene assay that demonstrated high correlation with in vivo activity in the Tg.AC assay (Thompson et al., 2000
). Next, 99 noncarcinogenic pharmaceuticals were screened for activity in the in vitro assay, and candidates for in vivo testing were selected that could meet the dosing limitations of a skin paint assay. Finally, three of the selected compounds were tested in a 26-week skin paint assay on Tg.AC mice (see accompanying manuscript). These studies were conducted to provide increased understanding of assay specificity, which is an important step in modulating regulatory and industry confidence in alternate short-term carcinogenicity assays.
![]() |
ACKNOWLEDGMENTS |
---|
![]() |
NOTES |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Asoh, K-I., Saburi, Y., Sato, I., Nogae, I., Kohno, K., and Kuwano, M. (1989). Potentiation of some anticancer agents by dipyridamole against drug-sensitive and drug-resistant cancer cell lines. Jpn. J. Cancer Res. 80, 475481.[ISI][Medline]
Cannon, R. E., Spalding, J. W., Virgil, K. M., Faircloth, R. S., Humble, M. C., Lacks, G. D., and Tennant, R. W. (1998). Induction of transgene expression in Tg.AC (v-Ha-ras) transgenic mice concomitant with DNA hypomethylation. Mol. Carcinog. 21, 244250.[CrossRef][ISI][Medline]
Compere, S. J., Baldacci, P., and Jaenisch, R. (1988). Oncogenes in transgenic mice. Biochim. Biophys. Acta 948, 129149.[CrossRef][ISI][Medline]
Contrera, J. F., Jacobs, A. C., and DeGeorge, J. J. (1997). Carcinogenicity testing and the evaluation of regulatory requirements for pharmaceuticals. Regul. Toxicol. Pharmacol. 25, 130145.[CrossRef][ISI][Medline]
Curtin, N. J., and Turner, D. P. (1999). Dipyridamole-mediated reversal of multidrug resistance in MRP over-expressing human lung carcinoma cells in vitro. Eur. J. Cancer 35, 10201026.[CrossRef][ISI][Medline]
Egeli, U. (1998). Induction of sister chromatid exchanges by pyrimethamine in human lymphocyte cultures. Teratog. Carcinog. Mutagen 18, 163169.[CrossRef][ISI][Medline]
Egeli, U., and Erdogen, G. (1991). The clastogenic effect of pyrimethamine (Daraprim) on human chromosomes in lymphocyte cultures. Cell Biol. Toxicol. 7, 347356.[CrossRef][ISI][Medline]
Elsliger, M-A., Wachter, R. M., Hanson, G. T., Kallio, K., and Remington, S. J. (1999). Structural and spectral response of green fluorescent protein variants to changes in pH. Biochemistry 38, 52965301.[CrossRef][ISI][Medline]
Garcia-Canero, R., Trilla, C., Perez de Diego, J., Diaz-Gil, J. J., and Cobo, J. M. (1999). Na+:H+ exchange inhibition induces intracellular acidosis and differentially impairs cell growth and viability of human and hepatocarcinoma cells. Toxicol. Lett. 106, 215228.[CrossRef][ISI][Medline]
Greenhalgh, D. A., Quintanilla, M. I., Orengo, C. C., Barber, J. L., Eckhardt, J. N., Rothnagel, J. A., and Roop, D. R. (1993). Cooperation between v-fos and v-rasHA induces autonomous papillomas in transgenic epidermis but not malignant conversion. Cancer Res. 53, 50715075.[Abstract]
Guyton, K. Z., Xu, Q., and Holbrook, N. J. (1996). Induction of the mammalian stress response gene GADD153 by oxidative stress: Role of AP-1 element. Biochem. J. 314, 547554.[ISI][Medline]
Hansen, L. A., and Tennant, R. W. (1994). Follicular origin of epidermal papillomas in v-Ha-ras-transgenic TG.AC mouse skin. Proc. Natl. Acad. Sci. U.S.A. 91, 78227826.[Abstract]
Honchel, R., Rosenzweig, B. A., Thompson, K. L., Blanchard, K. T., Furst, S. M., Stoll, R. E., and Sistare, F. D. (2001). Loss of palindromic symmetry in Tg.AC mice with a nonresponder phenotype. Mol. Carcinog 30, 99110.[CrossRef][ISI][Medline]
Leder, A., Kuo, A., Cardiff, R. D., Sinn, E., and Leder, P. (1990). v-Ha-ras transgene abrogates the initiation step in mouse skin tumorigenesis: Effects of phorbol esters and retinoic acid. Proc. Natl. Acad. Sci. U.S.A. 87, 91789182.[Abstract]
Lin, X., Gately, D. P., Hom, D., Mishima, M., Los, G., and Howell, S. B. (2001). Quantification of tumor cell injury in vitro and in vivo using expression of green fluorescent protein under the control of the GADD153 promoter. Int. J. Cancer 91, 555562.[CrossRef][ISI][Medline]
Luethy, J. D., and Holbrook, N. J. (1992). Activation of the GADD153 promoter by genotoxic agents: A rapid and specific response to DNA damage. Cancer Res. 52, 510.[Abstract]
Martelli, A., Mattioli, F., Ghia, M., Mereto, E., and Brambilla, G. (1996). Comparative study of DNA repair induced by cyproterone acetate, chlormadinone acetate and megestrol acetate in primary cultures of human and rat hepatocyes. Carcinogenesis 17, 11531156.[Abstract]
McEvoy, G. K., Ed. (1998). American Hospital Formulary Service Drug Information, American Society of Health-System Pharmacists, Bethesda, MD.
Morrow, C. S., Smitherman, P. K., and Townsend, A. J. (2000). Role of multidrug-resistance protein 2 in glutathione S-transferase P11-mediated resistance to 4-nitroquinoline 1-oxide toxicities in HepG2 cells. Mol. Carcinog. 29, 170178.[CrossRef][ISI][Medline]
National Toxicology Program. (1978). Bioassay of pyrimethamine for possible carcinogenicity (CAS No. 58-14-0). Technical Report Series No. 77, U.S. Department of Health and Human Services, Public Health Service, Bethesda, MD.
Ono, T., Sekiya, T., Takahashi, Y., Sasaki, Y. F., and Ohta, T. (1997). Species-specificity of pyrimethamine in the rodent bone marrow micronucleus test. Mutat. Res. 390, 167170.[ISI][Medline]
Physicians Desk Reference Electronic Library (2002). Medical Economics Company, Montvale, NJ.
Price, B. D., and Calderwood, S. K. (1992). Gadd45 and Gadd153 messenger RNA levels are increased during hypoxia and after exposure of cells to agents which elevate the levels of the glucose-regulated proteins. Cancer Res. 52, 38143817.[Abstract]
Robinson, D., and MacDonald, J. S. (2001). Background and framework for ILSIs collaborative evaluation program on alternative models for carcinogenicity assessment. Toxicol. Pathol. 29, 1319.[CrossRef][ISI][Medline]
Sistare, F. D., Thompson, K. L., Honchel, R., and DeGeorge, J. (2002). Evaluation of the Tg.AC transgenic mouse assay for testing the human carcinogenic potential of pharmaceuticals-practical pointers, mechanistic clues, and new questions. Int. J. Toxicol. 21, 6579.[CrossRef][ISI][Medline]
Spalding, J. W., French, J. E., Tice, R. R., Furedi-Machacek, M., Haseman, J. K., and Tennant, R. W. (1999). Development of a transgenic mouse model for carcinogenesis bioassays: Evaluation of chemically induced skin tumors in Tg.AC mice. Toxicol. Sci. 49, 241254.[Abstract]
Strazzabosco, M., Poci, C., Spirli, C., Zsembery, A., Granato, A., Massimino, M.L., and Crepaldi, G. (1995). Intracellular pH regulation in Hep G2 cells: Effects of epidermal growth factor, transforming growth factor-, and insulinlike growth factor-II on Na+/H+ exchange activity. Hepatology 22, 588597.[CrossRef][ISI][Medline]
Thompson, K. L., Rosenzweig, B. A., and Sistare, F. D. (1998). An evaluation of the hemizygous transgenic Tg.AC mouse for carcinogenicity testing of pharmaceuticals. II. A genotypic marker that predicts tumorigenic responsiveness. Toxicol. Pathol. 26, 548555.[ISI][Medline]
Thompson, K. L., Rosenzweig, B. A., Tsong, Y., and Sistare, F. D. (2000). Evaluation of in vitro reporter gene induction assays for use in a rapid prescreen for compound selection to test specificity in the Tg.AC mouse short-term carcinogenicity assay. Toxicol. Sci. 57, 4353.
Todd, M. D., Lee, M. J., Williams, J. L., Nalezny, J. M., Gee, P., Benjamin, M. B., and Farr, S. B. (1995). The CAT-Tox (L) Assay: A sensitive and specific measure of stress-induced transcription in transformed human liver cells. Fundam. Appl. Toxicol. 28, 118128.[CrossRef][ISI][Medline]
Trempus, C. S., Ward, S., Farris, G., Malarkey, D., Faircloth, R. S., Cannon, R. E., and Mahler, J. F. (1998). Association of v-Ha-ras transgene expression with development of erythroleukemia in Tg.AC transgenic mice. Am. J. Pathol. 153, 247254.
Vairo, G., Cocks, B. G., Cragoe, E. J., and Hamilton, J. A. (1992). Selective suppression of growth factor-induced cell cycle gene expression by Na+/H+ antiport inhibitors. J. Biol. Chem. 267, 1904319046.
Van Hummeln, P., Elhajouji, A., and Kirsch-Volders, M. (1995). Clastogenic and aneugenic effects of three benzimidazole derivatives in the in vitro micronucleus test using human lymphocytes. Mutagenesis 10, 2329.[Abstract]
Versantwoort, C. H. M., Broxterman, H. J., Bagrij, T., Scheper, R. J., and Twentyman, P. R. (1995). Regulation by glutathione of drug transport in multidrug-resistant human lung tumour cell lines overexpressing multidrug resistance-associated protein. Br. J. Cancer 72, 8289.[ISI][Medline]
Vijayalaxmi, K. K., and Vishalakshi, M. (2000). Evaluation of the genotoxic effects of pyrimethamine, an antimalarial drug, in the in vivo mouse. Teratog. Carcinog. Mutagen. 20, 6571.[CrossRef][ISI][Medline]
Wang, X-Z., Lawson, B., Brewer, J., Zinzser, H., Sanjay, A., Boorstein, R., Kreibich, G., Hendershot, L., and Ron, D. (1996). Signals from the stressed endoplasmic reticulum induce C/EBP homologous protein (CHOP/GADD153). Mol. Cell. Biol. 16, 42734280.[Abstract]
Werner, S., Kunz, S., Beckurts, T., Heidecke, C.-D., Wolff, T., and Schwarz, L. R. (1997). Formation of DNA adducts by cyproterone acetate and some structural analogues in primary cultures of human hepatocytes. Mutat. Res. 395, 179187.[ISI][Medline]
|