Mol Cancer Ther. 2005;4:1026-1030
© 2005 American Association for Cancer Research
Identification of compounds that inhibit growth of 2-amino-1-methyl-6-phenylimidazo(4,5-b)pyridine–resistant cancer cells
Kurtis E. Bachman1,
Jason Sager1,
Ian Cheong1,
Marco Catto2,
Alberto Bardelli1,
Ben Ho Park1,
Bert Vogelstein1,
Angelo Carotti2,
Kenneth W. Kinzler1 and
Christoph Lengauer1
1 Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University School of Medicine, and Howard Hughes Medical Institute and 2 Dipartimento Farmacochimico, Università di Bari, Bari, Italy
Requests for reprints: Kurtis E. Bachman, University of Maryland Greenebaum Cancer Center, University of Maryland School of Medicine, 9-009 Bressier Research Building, Baltimore, MD 21201. Phone: 410-328-8076; Fax: 410-328-6559. E-mail: kbachman{at}som.umaryland.edu
 |
Abstract
|
|---|
The dietary carcinogen 2-amino-1-methyl-6-phenylimidazo(4,5-b)pyridine (PhIP) is a heterocyclic amine and is a common byproduct of cooked meat and fish. Although most cells undergo apoptosis when exposed to this mutagen, subsets develop resistance. Rather than die, these resistant cells persist and accumulate mutations, thereby driving tumorigenesis of exposed organs within the gastrointestinal tract. By applying a high-throughput cell-based screen of 32,000 small molecules, we have identified a family of compounds that specifically inhibit the growth of PhIP-resistant cancer cells. These compounds may prove useful for the treatment or prevention of gastrointestinal tumors arising after exposure to PhIP and related carcinogens.
|
Introduction
|
|---|
Elucidation of the role of environmental carcinogens in the development of human cancers has long been an important and active area of research. Many dietary mutagens have been identified, and experimental studies in vivo and in vitro suggest that they may play an integral part in the tumorigenic process (1–3). Heterocyclic amines have in particular attracted much attention because they are found in relatively high quantities whenever meat or fish is thoroughly cooked. Several epidemiologic studies have shown that individuals who prefer well-cooked meat have an increased risk of developing cancers of the gastrointestinal tract and other organs (4).
The most prevalent heterocyclic amine is 2-amino-1-methyl-6-phenylimidazo(4,5-b)pyridine (PhIP), which is known to be a potent DNA-damaging agent and a mutagen (1, 3–5). When exposed to this bulky adduct-forming molecule, most cells are either able to repair the DNA damage that results or, if the damage is too high to be efficiently repaired, undergo programmed cell death. It would be expected, however, that a small subset of cells would acquire genetic alterations that provide them with the ability to resist PhIP-induced toxicity. These cells may be particularly prone to neoplasia, as they can persist even in the presence of multiple mutations caused by PhIP exposure. Accordingly, several studies in mice and rats have shown that prolonged exposure to PhIP can induce the formation of colon, prostate, and mammary tumors (6–10).
Despite the above findings, little is known about the precise molecular mechanisms underlying PhIP's carcinogenesis and no therapeutic approaches that may be applicable to PhIP-initiated cancers have been established. To this end, we reasoned that it might be possible to discover small molecules that selectively target PhIP-resistant tumor cells, as described below.
|
Materials and Methods
|
|---|
Cell Lines
The parental cell line used for screening was a derivative of the human colorectal cancer cell line, HCT116, that harbors an additional copy of chromosome 3 to correct for mutL homologue 1 deficiency and microsatellite instability (11). This cell line was used previously to study carcinogen-specific genetic instability, which provided us with several PhIP-resistant clones as well as N-methyl-N'-nitro-N-nitrosoguanidine (MNNG)–resistant controls that differed only in their resistance to certain carcinogens and acquired forms of genetic instability (12). Cells were cultured in McCoy's 5A medium (Invitrogen/Gibco, Rockville, MD) supplemented with 10% (vol/vol) fetal bovine serum (HyClone, Logan, UT), 100 units/mL penicillin, and 0.1 mg/mL streptomycin.
Screening and Compound Libraries
All screening steps were done using a Biomek FX Laboratory Automation Workstation (Beckman Coulter, Inc., Fullerton, CA). Approximately 32,000 small molecules were obtained from ChemBridge Corporation (DiverSet Libraries, San Diego, CA). All compound stocks were maintained in 100% DMSO (Sigma, St. Louis, MO) at –20°C. For primary screening, each compound was at a final concentration of 2.5 µg/mL and 0.5% DMSO. On day 0, 20 µL of complete medium containing the appropriate number of cells (150–300) was plated per well of a 384-well tissue culture plate (Fisher Scientific, Hampton, NH) and incubated overnight at 37°C with 5% CO2 and 90% humidity. The appropriate number of cells to plate per well was determined for each cell line prior to screening. Cells were seeded to yield 70% to 80% confluency after 5 days of growth in medium containing 0.5% DMSO. To minimize screening variability between cell lines, each 384-well plate was divided into four 96-well quadrants, with each quadrant harboring a specific cell line. Therefore, a total of 80 compounds plus 16 DMSO (no-drug) controls were tested against four cell lines in each plate. Twenty to 24 hours post-seeding (day 1), compounds were prepared by serial dilution with complete medium to yield a concentration of 5 µg/mL. Twenty microliters of the 5 µg/mL dilution was added to each well containing 20 µL of medium and cells, yielding the final screening concentration and a volume of 40 µL. Plates were incubated for an additional 3 days (days 2–5) until control cells (0.5% DMSO only) reached 
70% to 80% confluency. On day 6, 40 µL of lysis/detection solution containing 1.2% Igepal CA-630 (Sigma) and a 1:1,000 dilution of SYBR Green I nucleic acid stain (Molecular Probes, Eugene, OR) was added to each well. Following an overnight incubation at 37°C, total fluorescence was measured using a Fluorostar Galaxy plate reader with a 485 nm excitation filter and 520 nm emission filter set (BMG Labtech, Inc., Durham, NC). This data was exported to a custom program that determined growth inhibition by dividing each individual fluorescence read by the average of 16 control fluorescence reads. Compounds that showed 50% growth inhibition compared with the DMSO-only controls average were scored as "hits". Compounds scored as hits were those that showed cell type–specific growth inhibition and those that inhibited the growth of all cell lines at the screening concentration. Secondary screens were done as above except that each hit compound was tested using a concentration range extending from 7.5 down to 0.03 µg/mL in 3-fold increments. Compounds found to specifically inhibit the growth of the PhIP-resistant cell line when compared with parental and MNNG resistant control cell lines at two or more concentrations were repeated for confirmation. Using the same concentration range, the best secondary hit and several of its analogues were then tested against multiple PhIP-resistant, MNNG-resistant, and parental clones as well as other colon cancer cell lines. The PhIP-2 clone was used to derive the IC50 concentrations for Table 1. Cells were plated at desired densities in 384-well plates and exposed to varying concentrations of drug (similar to secondary screen). After 4 days of incubation, cell growth was assessed by fluorescence detection. Growth was normalized against DMSO-treated cells.
Analogue Synthesis
Chemicals and solvents for the preparation of condensed heterocyclic derivatives were from Sigma. Polycyclic derivatives, KLB compounds 1 through 8, were prepared via a one-pot procedure, starting from ninhydrin, a suitable ketone and hydrazine in acetic acid, as previously described (13–16). Dehydrogenation of compounds KLB-1 and KLB-3 with 2,3-dichloro-5,6-dicyano-1,4-benzoquinone in refluxing dioxane afforded aromatic derivatives KLB-2 and KLB-4, respectively. Purity and identity of all the tested compounds were checked by microanalysis (Carlo Erba 1106 elemental analyzer), 1H-nuclear magnetic resonance (Bruker 300, 300 MHz spectrometer) and HPLC (Waters 600, UV detector 486, C-18 Waters reversed phase column). The compound purity was always >99%. Physicochemical, spectroscopic, chromatographic, and analytic data were in full agreement with the proposed structures and agreed well with previously published data (13–16).
Colony Formation Assay
Cell lines were plated on day 0 at low densities (1,200–2,400 cells per well) into 48-well tissue culture plates. Twenty to 24 hours after seeding on day 1, medium containing drug at a concentration of 0.03 µg/mL was added to each well. Control wells received medium with DMSO only. Plates were then incubated for 4 days. Following incubation, cells were washed once with 1x PBS and stained with crystal violet solution (1 g of crystal violet in 500 mL buffered formalin) for 20 minutes. Cells were then rinsed with 1x PBS and photographed.
|
Results and Discussion
|
|---|
The cells used for the screen were derived by exposing the colorectal cancer cell line HCT116 to PhIP at doses which killed >99% of the cells. The clones that were able to survive this exposure were shown to be resistant to subsequent exposure to PhIP (12). We initially selected one such clone for high-throughput screening, comparing it to parental cells as well as to another clone of HCT116 that had been selected for resistance to MNNG, a completely different carcinogen (12).
A high-throughput screen was carried out by plating an equal number of cells of each of the three lines into separate wells of a 384-well tissue culture plate. In the absence of drug, the three cell lines grew at identical rates, with exponential growth occurring for 5 days, at which time 70% to 80% confluency was reached. Cell growth was assessed through a fluorescence-based assay employing SYBR Green I. A commercially available small molecule library totaling 32,000 diverse compounds was initially screened against the three cell line derivatives. This primary screen, with each compound at a concentration of 2.5 µg/mL, yielded 710 compounds (2.4% hit rate) that either selectively inhibited the growth of the PhIP-resistant clone or inhibited the growth of all three cell lines in a nonselective manner.
Secondary screens were done to confirm and quantify the selectivity of these 710 compounds. Through testing concentrations ranging from 0.03 to 7.5 µg/mL (3-fold increments), we identified 10 compounds that showed selectivity at two consecutive concentrations. On repeated assays, KLB-1 was found to be the compound that specifically targeted PhIP-resistant cells at the lowest drug concentrations (Fig. 1). KLB-1 is an indenobenzocinnoline derivative prepared and tested previously, along with a large series of closely related analogues, as a central benzodiazepine receptor ligand and monoamine oxidase inhibitor (13, 15). In both biological assays, KLB-1 displayed low activity.
View larger version (15K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1. Identification of a compound that inhibits the growth of PhIP-resistant cells. A, structure and molecular weight of compound KLB-1. B, growth inhibition of PhIP-resistant cells by compound KLB-1 in primary screen (single read). Black columns, growth of cells exposed to 0.5% DMSO only. Gray columns, growth of cells exposed to KLB-1 at a screening concentration of 2.5 µg/mL (8.8 µmol/L). C, growth inhibition of PhIP-resistant cells from secondary screen using KLB-1 concentrations of 100 nmol/L. Black columns, growth (± SD) of cells exposed to 0.5% DMSO only. Gray columns, growth (± SD) of cells exposed to compound KLB-1. Secondary assays were done in triplicate.
|
|
To ensure high purity for subsequent experiments, KLB-1 and seven of its analogues (Table 1) were resynthesized according to the synthetic pathway previously described (13). All of the newly synthesized compounds were then tested in tertiary screens incorporating three independent PhIP-resistant as well as two independent MNNG-resistant clones of HCT116. Each screen was done in triplicate using duplicate plates to ensure reproducible plating efficiencies. KLB-1 was found to specifically inhibit the growth of all three PhIP-resistant clones, with an IC50 of 32.5 nmol/L (Fig. 2A; Table 1). This yielded a 20-fold difference in selectivity when compared with the parental cell line or to MNNG-resistant control clones. A closely related analogue, KLB-2, was also found to specifically target the PhIP-resistant clones, with an IC50 of 40 nmol/L and a 17-fold difference in selectivity (Fig. 2A; Table 1). Other analogues tested showed a range of growth inhibition and higher IC50 concentrations. However, none of the other six compounds proved to be selective inhibitors of the PhIP-resistant clones (Fig. 2A; Table 1). As far as the structure-activity relationships are concerned, one may notice that pentacyclic derivatives with a "Z-shaped" topology (KLB-1 and KLB-2) showed a higher activity than the corresponding "J-shaped" pentacyclic (KLB-3, KLB-4, and KLB-6) and tetracyclic (KLB-5) congeners, irrespective of the planarity (aromaticity) of the cyclic system condensed to the indanone moiety. Substituted tricyclic compounds (KLB-7 and KLB-8), endowed with a higher conformational mobility and a less extended aromatic heterocyclic skeleton, were consistently less active than their tetracyclic and pentacyclic homologues.
View larger version (33K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2. Compound-specific growth inhibition of multiple PhIP-resistant clones. A, exposure to compound KLB-1 and its analogue KLB-2 at various concentrations (tertiary screen) causes growth inhibition of three PhIP-resistant clones when compared with MNNG-resistant clones and parental control cells (left). Other close analogues (KLB-3 and KLB-4) exhibit no specific growth inhibition of PhIP-resistant cells (right). Cells were plated at desired densities in 384-well plates, and cell growth was assessed by fluorescence detection. Growth was normalized against cells growing in 0.5% DMSO. B, confirmation of compound-specific growth inhibition of PhIP-resistant cells using a colony formation assay. Cells were plated at low densities, exposed to drugs at 0.03 µg/mL and stained with crystal violet solution confirming results obtained from tertiary screen. KLB-1R, resynthesized compound.
|
|
To verify these results with an independent growth assay, we tested the colony-forming abilities of the various clones following exposure to KLB-1 and its analogues (Fig. 2B). As expected, KLB-1 and KLB-2 showed selective growth inhibition of the PhIP-resistant clones. In contrast, analogues KLB-3 and KLB-4 exhibited no selectivity (Fig. 2B).
The mechanisms that underlie PhIP resistance as well as the capacity of KLB-1 and KLB-2 to selectively inhibit the growth of PhIP-resistant cells remains unknown. The selective growth inhibition of multiple unique clones, however, suggests that either a specific genetic lesion gives rise to this resistance, or that varying mutations occur in different genes all involved in the same targeted pathway leading to PhIP resistance. The KLB compounds will therefore be useful resources to further explore the mechanisms that underlie PhIP resistance. Our approach enables us to identify compounds that target a defined phenotype rather than a known genetic lesion. This has powerful implications as it allows for the design of drug library screening that could identify compounds specifically targeting clinically relevant drug resistance. It will be of interest to further test the KLB compounds in in vivo animal models and against multiple tumor types that have acquired resistance to other carcinogens or chemotherapeutic agents.
The impact of dietary carcinogens on the incidence of human cancers remains an area of intense research. Cellular repair systems control the damage caused by these mutagens; however, increased exposure leads to tumor-specific genetic changes, prevention of apoptosis and acquired resistance. Although there is increasing evidence that dietary PhIP exposure is associated with the development of human cancers, it is difficult to determine if this exposure would result in the same genetic changes that presumably took place in the cell lines used in our experiments. An attempt to address this concern was made during the derivation of the PhIP-resistant clones by selection in the presence of activated PhIP with hopes of modeling the metabolites and subsequent genetic alterations produced in vivo (12). Our results do, however, provide proof that small molecules can selectively target cells that have acquired a resistance phenotype. The cell-based screening approach described here should be similarly useful for the identification of agents that target cells that have developed resistance to clinically important chemotherapeutic agents in the absence of known genetic mechanisms.
|
Acknowledgments
|
|---|
We thank Carlo Rago and Victor Velculescu for technical assistance and helpful discussion.
|
Footnotes
|
|---|
Grant support: Supported by the Maryland Cigarette Restitution Fund, the American Cancer Society (#IRG-58-005-41), and the Flight Attendant Medical Research Institute.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Note: K.E. Bachman is currently at University of Maryland Greenebaum Cancer Center, University of Maryland School of Medicine, Baltimore, MD 21201.
A. Bardelli is currently at The Oncogenomics Center, Institute for Cancer Research and Treatment, University of Torino-Medical School, 10143 Candiolo, Italy.
Received 2/ 7/05;
revised 3/14/05;
accepted 3/28/05.
|
References
|
|---|
Schut HA, Snyderwine EG. DNA adducts of heterocyclic amine food mutagens: implications for mutagenesis and carcinogenesis. Carcinogenesis 1999;20:353–68.[Abstract/Free Full Text]
Thorgeirsson UP, Snyderwine EG, Gomez DE, Adamson RH. Dietary heterocyclic amines as potential human carcinogens: experimental data from nonhuman primates. In Vivo 1996;10:145–52.[Medline]
Gooderham NJ, Zhu H, Lauber S, Boyce A, Creton S. Molecular and genetic toxicology of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP). Mutat Res 2002;506–7:91–9.
Felton JS, Knize MG, Bennett LM, Malfatti MA, Colvin ME, Kulp KS. Impact of environmental exposures on the mutagenicity/carcinogenicity of heterocyclic amines. Toxicology 2004;198:135–45.[CrossRef][Medline]
Pfau W, Martin FL, Cole KJ, et al. Heterocyclic aromatic amines induce DNA strand breaks and cell transformation. Carcinogenesis 1999;20:545–51.[Abstract/Free Full Text]
el-Bayoumy K, Chae YH, Upadhyaya P, et al. Comparative tumorigenicity of benzo[a]pyrene, 1-nitropyrene and 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine administered by gavage to female CD rats. Carcinogenesis 1995;16:431–4.[Abstract/Free Full Text]
Ito N, Hasegawa R, Sano M, et al. A new colon and mammary carcinogen in cooked food, 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP). Carcinogenesis 1991;12:1503–6.[Abstract/Free Full Text]
Shirai T, Sano M, Tamano S, et al. The prostate: a target for carcinogenicity of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) derived from cooked foods. Cancer Res 1997;57:195–8.[Abstract/Free Full Text]
Hasegawa R, Sano M, Tamano S, et al. Dose-dependence of 2-amino-1-methyl-6-phenylimidazo[4,5-b]-pyridine (PhIP) carcinogenicity in rats. Carcinogenesis 1993;14:2553–7.[Abstract/Free Full Text]
Ghoshal A, Preisegger KH, Takayama S, Thorgeirsson SS, Snyderwine EG. Induction of mammary tumors in female Sprague-Dawley rats by the food-derived carcinogen 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine and effect of dietary fat. Carcinogenesis 1994;15:2429–33.[Abstract/Free Full Text]
Koi M, Umar A, Chauhan DP, et al. Human chromosome 3 corrects mismatch repair deficiency and microsatellite instability and reduces N-methyl-N'-nitro-N-nitrosoguanidine tolerance in colon tumor cells with homozygous hMLH1 mutation. Cancer Res 1994;54:4308–12.[Abstract/Free Full Text]
Bardelli A, Cahill DP, Lederer G, et al. Carcinogen-specific induction of genetic instability. Proc Natl Acad Sci U S A 2001;98:5770–5.[Abstract/Free Full Text]
Altomare C, Campagna F, Carta V, et al. Synthesis, benzodiazepine receptor affinity and anticonvulsant activity of 5-H-indeno[1,2-c]pyridazine derivatives. Farmaco 1994;49:313–23.[Medline]
Kneubuhler S, Carta V, Altomare C, Carotti A, Testa B. Synthesis and monoamine oxidase inhibitory activity of 3-substituted 5H-indeno[1,2-c]pyridazines. Helv Chim Acta 1993;76:1956–63.[CrossRef]
Kneubuhler S, Thull U, Altomare C, et al. Inhibition of monoamine oxidase-B by 5H-indeno[1,2-c]pyridazines: biological activities, quantitative structure-activity relationships (QSARs) and 3D-QSARs. J Med Chem 1995;38:3874–83.[CrossRef][Medline]
Carotti A, Carta V, Campagna F, Altomare C, Casini G. An efficient route to biologically active 5H-indeno[1,2-c]pyridazine-5-ones. Farmaco 1993;48:137–41.
This article has been cited by other articles:
|
|
|
|
 
E. Gallmeier, T. Hucl, J. R. Brody, D. A. Dezentje, K. Tahir, J. Kasparkova, V. Brabec, K. E. Bachman, and S. E. Kern
High-Throughput Screening Identifies Novel Agents Eliciting Hypersensitivity in Fanconi Pathway-Deficient Cancer Cells
Cancer Res.,
March 1, 2007;
67(5):
2169 - 2177.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
