Molecular Cancer Therapeutics CTRC-AACR San Antonio Breast Cancer Symposium Targeting the PI3-Kinase Pathway in Cancer
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
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hillier, S. M.
Right arrow Articles by Croy, R. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hillier, S. M.
Right arrow Articles by Croy, R. G.
Mol Cancer Ther. 2006;5:977-984
© 2006 American Association for Cancer Research

Research Articles: Therapeutics

DNA adducts formed by a novel antitumor agent 11ß-dichloro in vitro and in vivo

Shawn M. Hillier1, John C. Marquis1, Beatriz Zayas2, John S. Wishnok1, Rosa G. Liberman1, Paul L. Skipper1, Steven R. Tannenbaum1, John M. Essigmann1 and Robert G. Croy1

1 Department of Chemistry and Biological Engineering Division, Massachusetts Institute of Technology, Cambridge, Massachusetts and 2 School of Environmental Affairs, Universidad Metropolitana, San Juan, Puerto Rico

Requests for reprints: Robert G. Croy, Department of Chemistry and Biological Engineering Division, Massachusetts Institute of Technology, Cambridge, MA 02139. Phone: 617-253-6729; Fax: 617-253-5445. E-mail: rgcroy{at}mit.edu

Abstract

The multifunctional molecule 11ß-dichloro consists of a ligand for the androgen receptor linked to a bifunctional alkylating group, permitting it to create DNA adducts that bind the androgen receptor. We propose that binding of the androgen receptor to 11ß-DNA adducts acts to both shield damaged sites from repair and disrupt the expression of genes essential for growth and survival. We investigated the formation 11ß-DNA adducts in tumor xenograft and nontumor tissues in mice. Using [14C]-11ß-dichloro, we show that the molecule remains intact in blood and is widely distributed in mouse tissues after i.p. injection. Covalent 11ß-guanine adducts identified in DNA that had been allowed to react with 11ß-dichloro in vitro were also found in DNA isolated from cells in culture treated with 11ß-dichloro as well as in DNA isolated from liver and tumor tissues of mice treated with the compound. We used accelerator mass spectrometry to determine the levels of [14C]-11ß-DNA adducts in LNCaP cells treated in culture as well as in liver tissue and LNCaP xenograft tumors in treated mice. The level of DNA adducts in tumor tissue was found to be similar to that found in LNCaP cells in culture treated with 2.5 µmol/L 11ß-dichloro. Our results indicate that 11ß-dichloro has sufficient stability to enter the circulation, penetrate tissues, and form DNA adducts that are capable of binding the androgen receptor in target tissues in vivo. These data suggest the involvement of our novel mechanisms in the antitumor effects of 11ß-dichloro. [Mol Cancer Ther 2006;5(4):977–84]

Introduction

Prostate cancer is the second most commonly diagnosed cancer and the fourth leading cause of cancer death among men in developed countries (1). Most prostate cancers depend on androgens for their growth. Therefore, chemical or surgical castration and/or treatment with hormonal antagonists is often given as adjuvant therapy following the surgical removal of the tumor. Androgen ablation results in the apoptotic death of androgen-sensitive cells producing an initial therapeutic response (2–5). Unfortunately, patients who undergo these adjuvant treatments often develop aggressive and metastatic androgen-independent forms of hormone-refractory prostate cancer. New and more effective drugs are needed to treat advanced stages of this disease, as well as to treat more effectively its earlier and less aggressive forms.

The androgen receptor is expressed throughout the development of prostate cancer and is present in most patients with hormone-refractory prostate cancer (6–9). Clinical evidence suggests that disruption of androgen receptor signaling through dysregulation of androgen receptor coregulators, androgen receptor gene amplification, or mutations in the androgen receptor enables it to remain transciptionally active in the presence of androgen receptor antagonists and other therapeutics (10, 11). The continued role of the androgen receptor in hormone-refractory prostate cancer provides an attractive target for therapeutic development.

We have recently reported the development of a novel cytotoxic agent that was designed to take advantage of the dysregulation of the androgen receptor in hormone-refractory prostate cancer (12). 11ß-Dichloro is a bifunctional compound in which an androgen receptor ligand is linked to a p-N,N-bis-(2-chloroethyl)aminophenyl moiety that can produce covalent damage to DNA. A consequence of the stable connection between the groups at either end of the 11ß-dichloro molecule is that the androgen receptor can bind to covalent adducts that are formed in DNA. It is hypothesized that these androgen receptor-DNA adduct complexes both shield the DNA adduct from repair enzymes as well as prevent the androgen receptor from acting (by its normal function) to promote cell growth and survival (Fig. 1 ). This latter mechanism, inactivation of the androgen receptor through its physical association with DNA adducts, is different from the mechanism of currently used antagonists.


Figure 1
View larger version (28K):
[in this window]
[in a new window]
  		Figure 1. A, scheme illustrating the proposed effects of 11ß-DNA adducts on androgen receptor function and DNA repair in target cells. The steroid ligand portion of 11ß adducts could form a complex with the androgen receptor that prevents recognition and removal by the DNA repair complex (left). 11ß-DNA adducts compete with the natural ligand for the androgen receptor (DHT) and antagonize its transcriptional function, leading to diminished gene expression (right). The combination of these mechanisms results in apoptosis and cell death. B, structure of 11ß-dichloro; *, position of 14C atom.

 
The 11ß-dichloro compound rapidly induces apoptosis in LNCaP cells in culture at concentrations >5 µmol/L and is highly effective in preventing the growth of LNCaP tumors as xenografts in mice (90% growth inhibition; ref. 12). The studies described here were designed to examine the tissue distribution of 11ß-dichloro in mice and to determine whether the intact molecule is capable of forming DNA adducts in vivo that are required for our proposed mechanisms of action (i.e., adducts capable of interacting with the androgen receptor). We have also used the highly sensitive technique of accelerator mass spectrometry to investigate the formation of 11ß-DNA adducts in a human prostate tumor growing as s.c. xenografts in mice.

Materials and Methods

Chemicals
[14C]-11ß-Dichloro, (3-{4-[bis-(2-chloro-ethyl)-amino]-phenyl}-3-[14C]propyl)-carbamic acid 2-[6-(17-hydroxy-13-methyl-3-oxo-2,3,6,7,8,11,12,13,14,15,16,17-dodecahydro-1H-cyclopenta[a]phenanthren-11-yl)-hexylamino]-ethyl ester, was synthesized with one 14C atom in the propyl group connecting the carbamate and the p-N,N-bis-(2-chloroethyl)aminophenyl moieties using previously described procedures (13). The position of the radiolabel is indicated in Fig. 1B. Initial radiochemical purity was >98% and specific activity was 50 mCi/mmol. Unlabeled 11ß-dichloro was used to dilute the radiolabeled material where necessary.

Animals
Four- to six-week-old NIH Swiss Webster mice were purchased from Charles River Laboratories (Wilmington, MA). Female athymic NIH Swiss (nu/nu) mice (25 g) were obtained from the National Cancer Institute, Frederick Cancer Center (Frederick, MD). All experiments were done under the guidelines of the Massachusetts Institute of Technology Animal Care Committee.

Dose Formulation
[14C]-11ß-Dichloro or unlabeled 11ß-dichloro dissolved in a small volume of ethanol was added to cremophor-EL/saline (final proportions, cremophor-EL 37%/ethanol 27%/saline 30%) for administration by i.p. injection.

Biodistribution Analysis
NIH Swiss Webster mice received a single dose of 45 mg/kg of [14C]-11ß-dichloro (specific activity, 1.91 mCi/mmol) in 50 µL of vehicle administered via i.p. injection. Animals were sacrificed by carbon dioxide asphyxiation at 0.25, 1, 2, 4, 6, and 24 hours postinjection; blood samples obtained by cardiac puncture were collected in heparinized syringes; and the following tissues were excised from each animal: lung, liver, spleen, kidney, intestine, and in some instances adipose, heart, and skeletal muscle.

Intestinal contents were also collected. Tissue samples were flash-frozen and stored at –20°C before processing. Approximately 100 mg of each tissue (10 µL of whole blood) were placed in preweighed vials and solubilized in 1 mL of Solvable (Packard Biosciences, Meriden, CT) heated to 65°C for 3 hours or until the tissue was completely dissolved. The solutions were decolorized by treatment with 200 µL of 30% hydrogen peroxide. After addition of 15 mL Hionic Fluor scintillation fluid (Packard Biosciences), radioactivity was determined using a Beckman LS1801 Liquid Scintillation Counter.

Analysis of 11ß-Dichloro in Blood
Two volumes of acetonitrile were added to whole blood and the precipitate was isolated by centrifugation (5 minutes; 13,000 x g). The amount of 11ß-dichloro covalently bound to proteins was assessed by determining the 14C activity in aliquots of the supernatant (organic soluble phase) and precipitate.

The precipitated material was solubilized with 1 mL of Solvable (Packard Biosciences), heated overnight at 70°C, and decolorized by addition of 200 µL 30% hydrogen peroxide. The fraction of noncovalently bound 11ß-dichloro equals the amount of 14C in the organic soluble phase divided by the total amount of 14C in blood.

The amount of intact 11ß-dichloro that was present in blood was determined by high-performance liquid chromatography analysis. An aliquot of the organic soluble phase (supernatant) was dried in a Savant SC-110 SpeedVac, dissolved in 100 µL acetonitrile, and injected onto a Beckman octadecyl silane 4.6 x 250 mm Ultrasphere column at a flow rate of 1 mL/min with 10% CH3CN, 50% methanol containing 0.1 mol/L ammonium acetate. Compounds were eluted from the column with a 20-minute linear gradient that increased the concentration of methanol to 100%. Compounds were detected by a tandem configuration of UV (Rainin UV-1 UV Detector) and radiochemical (Packard Flow Scintillation Analyzer Model 150TR) detectors. The radiochemical detector was calibrated with a known amount of [14C]-11ß-dichloro that was used to correlate peak area with the amount of [14C]-11ß-dichloro in the sample.

Reaction of 11ß-Dichloro with DNA In vitro and Identification of Covalent Products
Salmon testes DNA was dissolved in 5 mmol/L sodium cacodylate and N,N-dimethylformamide was added to a final concentration of 25%. 11ß-Dicholoro was added in DMSO (50 µmol/L) and the solution incubated for 16 hours at 37°C. After phenol/chloroform extraction, the DNA was isolated by ethanol precipitation. Covalent products were released by acid hydrolysis in 0.1 N HCl (30 minutes, 70°C); after which, the solution was neutralized with NaOH and adjusted to 20 mmol/L Tris-HCl (pH 7.4) and 10% methanol. The solution was then loaded directly onto a C18 SepPak column (Waters Co., Milford, MA) that was sequentially eluted with 10%, 50% aqueous methanol solutions, and finally 100% methanol. After reducing the volume of the 100% methanol fraction, an aliquot was analyzed by high-performance liquid chromatography using conditions described above. Compounds in fractions collected from the high-performance liquid chromatography were analyzed by electrospray ionization mass spectrometry on an Agilent 1100 Series LC/MSD Trap operated in the positive ion mode. Samples were introduced by flow injection (0.2 mL/min) in methanol/10 mmol/L NH4Ac in H2O/acetonitrile (50:45:5).

Identification of 11ß-Dichloro DNA Adducts In vivo
NIH Swiss Webster mice were treated with 11ß-dichloro (25 mg/kg, i.p.) and sacrificed 4 hours later by asphyxiation with 95% CO2. Livers were removed surgically, snap frozen on dry ice, and stored at –80°C. Thawed tissue was minced and homogenized (Dounce) in 3 volumes of cold 0.01 mol/L Tris-HCl (pH 6.9), 0.25 mol/L sucrose, 2 mmol/L calcium chloride (lysis buffer). After filtering through nylon mesh, Triton X-100 was added to a final concentration of 5% and a crude nuclear fraction collected by centrifugation at 1,000 x g for 20 minutes at 4°C. The pellet was resuspended in 2 volumes of lysis buffer, to which SDS and sodium chloride were added to final concentrations of 1% and 1 mol/L, respectively. The viscous solution was extracted twice with chloroform/isoamyl alcohol (24:1) and nucleic acids collected by ethanol precipitation.

RNA was removed by digestion with RNase A. The isolated nucleic acids were dissolved in 2 mL of 0.05 mol/L Tris-HCl (pH 7.5), 0.1 mol/L NaCl and treated with 0.5 mg RNase A (10 minutes; 37°C). Following extraction with chloroform/isoamyl alcohol (24:1), DNA was isolated by ethanol precipitation. The DNA was then subjected to acid hydrolysis and released products were analyzed by high-performance liquid chromatography and electrospray ionization mass spectrometry as described above.

Cell Culture
LNCaP cells were obtained from the American Type Culture Collection (Rockville, MD) and maintained in RPMI 1640 (Life Technologies, Inc., Carlsbad, CA) supplemented with 2.5 mg/mL glucose, 1 mmol/L sodium pyruvate, 100 mmol/L HEPES, 2 mmol/L glutamax, and 10% fetal bovine serum in a humidified 5% CO2/air atmosphere at 37°C. LNCaP cells were seeded in six-well plates and allowed to attach to the surface for 24 hours. Cells were then exposed to the indicated dose of [14C]-11ß-dichloro dissolved in DMSO for the indicated period of time. At the end of the incubation, the cells were harvested by trypsinization, pelleted, and washed with PBS. The DNA from the cellular pellet was isolated according to the procedure described above.

For growth inhibition experiments, LNCaP cells were seeded in six-well dishes at 105 per well. Forty-eight hours later, the test compound was added in DMSO solution. Cells were detached by trypsin/EDTA after 36 hours and the number of cells in control and 11ß-dichloro-treated wells determined using a Coulter Counter. The percent growth inhibition is the ratio of cell number in treated and control wells multiplied by 100.

14C Accelerator Mass Spectrometry Analysis of DNA Adducts
Accelerator mass spectrometry analyses were conducted by the Biological Engineering Accelerator Mass Spectrometry Lab at Massachusetts Institute of Technology as described in detail elsewhere (14). DNA concentrations were determined by UV absorption at 260 nm using a Beckman DU-65 UV-Vis Spectrophotometer. Solutions of DNA dissolved in water were applied directly to a CuO matrix used for sample combustion. A standard consisting of a solution of [14C-methyl]-bovine serum albumin was used to calibrate the instrument. The amount of 14C in each sample was calculated from the peak area ratio of the sample to the standard. All samples were analyzed at least twice.

Tumor Implantation and Formation of 11ß-Dichloro DNA Adducts in Xenografts
Tumor xenografts were established on the flank of NIH Swiss nu/nu mice by s.c. injection of 2 x 106 LNCaP cells suspended in 0.25 mL of serum-free medium/Matrigel (1:1; Becton Dickinson, Franklin Lakes, NJ). When the tumors reached ~300 mm3 (5-6 weeks), animals received an i.p. injection of 50 mg/kg of [14C]-11ß-dichloro (specific activity, 0.5 mCi/mmol). Four hours later, DNA was isolated as described above from liver and tumor tissues of individual mice and subjected to accelerator mass spectrometry analysis to determine the amount of covalently bound radioactivity.

Analysis of Liver Toxicity in Mice
NIH Swiss nu/nu mice injected with either single or repeated dose(s) of 11ß-dichloro were sacrificed by carbon dioxide asphyxiation. Blood was collected by cardiac puncture and placed in a Becton Dickinson Microtainer Serum Separator tube for blood chemistry analysis, which was done by IDEXX Laboratories (North Grafton, MA).

Results

Distribution of Radioactivity in Blood and Selected Tissues
[14C]-11ß-Dichloro was rapidly absorbed into the circulation following i.p. injection. The peak concentration of radiolabeled 11ß-dichloro in blood of 128 µmol/L was found at 15 minutes and declined rapidly as shown in Fig. 2 (inset) with a half life (t1/2) of 1.3 hours. [14C]-11ß-Dichloro was well distributed into the tissues (Fig. 2). At 15 minutes after injection, the highest levels of radioactivity were found in the liver, intestine, spleen, and lung. High levels of radioactivity were also found in fat (not shown). Tissue concentrations reached maximum levels at 4 hours with the liver and kidney experiencing the highest concentrations. After 4 hours, there was rapid accumulation of radioactivity in feces, which is consistent with biliary excretion as a major route of 11ß-dichloro elimination.


Figure 2
View larger version (22K):
[in this window]
[in a new window]
  		Figure 2. Distribution of [14C]-11ß-dichloro in blood and selected tissues. NIH Swiss Webster mice were administered 45 mg/kg [14C]-11ß-dichloro (specific activity, 1.91 mCi/mmol) via i.p. injection and sacrificed at the indicated time points. Tissues were solubilized and radioactivity measured by scintillation counting (n = 3; bars, SD). Figure 2, blood; Figure 2, liver; Figure 2, lung; Figure 2, kidney; bullet, intestine; {square}, feces. Inset, concentration of 11ß-dichloro in blood.

 
Following organic extraction, <10% of radioactivity in blood remained associated with precipitated proteins, consistent with a low level of covalent modification of blood proteins. High-performance liquid chromatography analysis found that >90% of the organic-soluble radioactivity in blood after 1 hour corresponded to the intact 11ß-dichloro compound (Fig. 3 ). At 4 hours, several earlier eluting compounds were observed, which are probably 11ß-dichloro metabolites (Fig. 3). The majority of the radioactivity (>65%), however, still corresponded to the parent compound.


Figure 3
View larger version (12K):
[in this window]
[in a new window]
  		Figure 3. Reversed phase high-performance liquid chromatography analysis of radiolabeled compounds isolated from blood of mice at 0.25, 1, and 4 h after administration of [14C]-11ß-dichloro. The peak labeled 11ß-dichloro represents the intact compound. For chromatographic conditions, see Materials and Methods.

 
Identification of 11ß-DNA Adducts In vitro
The presence of the parent 11ß-dichloro compound in blood indicated that the intact molecule could be distributed to tissues and was available to react with cellular DNA forming covalent adducts. A key question about the proposed mechanism of action of the new compound is whether the intact molecule is capable of forming DNA adducts in vivo. To answer this question, we first characterized the DNA adducts formed by the reaction of 11ß-dichloro with DNA in vitro. Salmon testis DNA that had been incubated with 11ß-dichloro was subjected to acid hydrolysis and the hydrolyzed material was analyzed by reversed phase high-performance liquid chromatography. A single peak was observed that eluted before the unreacted 11ß-dichloro compound. Analysis of the eluted material by electrospray ionization mass spectrometry identified an ion with 813.5 m/z (Fig. 4A ) corresponding to [M + H]+. This ion is consistent with a chemical structure in which one ethylene group at the aniline nitrogen of the 11ß molecule is attached to a guanine base whereas the other ethylene has a hydroxyl group substituted for chlorine (Fig. 4A, bottom). We did not investigate whether hydrolysis of the chlorine occurred during reaction with DNA or at a subsequent step in our analytic procedure. Further analysis of the 813.4 m/z molecular ion by collision-induced dissociation yielded prominent fragment ions at m/z 662.4 and m/z 372.1 (Fig. 4A, inset). The proposed structure of the guanine adduct formed by 11ß-dichloro has a parent ion and collision-induced dissociation fragments analogous to those produced by a DNA adduct formed by a structurally related compound E2-7{alpha} in which estradiol is linked to the reactive p-N,N-bis-(2-chloroethyl)aminophenyl group (13). It is likely that the 11ß molecule forms a covalent bond at the N7 atom of guanine but we did not have sufficient material to allow complete structural characterization.


Figure 4
View larger version (16K):
[in this window]
[in a new window]
  		Figure 4. Electrospray ionization mass spectrometry analysis of 11ß-dichloro-DNA adducts formed in vitro and in vivo. A, salmon testes DNA reacted with 11ß-dichloro in vitro; B, DNA obtained from LNCaP cells in culture treated with 10 µmol/L 11ß-dichloro for 6 h; C, DNA isolated from liver tissue of NIH Swiss nu/nu mice treated with 25 mg/kg (i.p.) 11ß-dichloro. Proposed structures of the most abundant ion (813.5 m/z) and collision-induced dissociation (CID) fragments (insets) are shown in A. DNA was subjected to hydrolysis with 0.1 N HCl to release covalent adducts (see Materials and Methods).

 
Analysis of 11ß-DNA Adducts in LNCaP Cells in Culture
Following preliminary characterization of the 11ß-DNA adduct formed in vitro, we investigated whether the same adduct was formed in LNCaP cells in culture. The 11ß-dichloro compound rapidly induces apoptosis in this androgen receptor–expressing human prostate cancer cell line (12). We isolated DNA from LNCaP cells after treatment with 10 µmol/L 11ß-dichloro for 6 hours, which was then hydrolyzed using the same conditions used for in vitro modified DNA. The hydrolyzed products were then analyzed by electrospray ionization mass spectrometry. Among the hydrolysis products was a prominent ion at m/z 813.5, which was identical to the product identified in DNA that was directly reacted with 11ß-dichloro in vitro (Fig. 4B). Thus, the 11ß-dichloro molecule remains intact under cell culture conditions and reacts with guanine bases in cellular DNA. There were no marked differences between the DNA adduct that was identified in vitro and the one formed in LNCaP cells in culture.

We next investigated the formation of covalent 11ß-DNA adducts in LNCaP cells by treating them with [14C]-11ß-dichloro. DNA isolated from treated cells was analyzed by accelerator mass spectrometry to determine the amount of covalently bound [14C]-11ß-dichloro. The high sensitivity of the accelerator mass spectrometry technique permitted us to detect and quantify the levels of covalently bound 14C that were present in cellular DNA. We first established a dose-response relationship for adduct formation by treating LNCaP cells with 2.5, 5, or 10 µmol/L [14C]-11ß-dichloro for 4 hours. The amount of 14C per microgram of DNA increased in direct proportion with 11ß-dichloro concentration in the growth media (Fig. 5A ). Based on specific activity, the level of 11ß-DNA adducts rose from 0.3 to 1.0 adducts per 106 bases over the dose range of 2.5 to 10 µmol/L 11ß.


Figure 5
View larger version (6K):
[in this window]
[in a new window]
  		Figure 5. Formation of 11ß-DNA adducts and inhibition of LNCaP cell growth by 11ß-dichloro. A, relationship between 11ß-dichloro concentration in growth media and level of 11ß-DNA adducts formed after 4-h exposure. B, rate of formation of 11ß-DNA adducts in LNCaP cells exposed to 10 µmol/L 11ß-dichloro. C, growth inhibition of LNCaP cells by 11ß-dichloro (bars, SD).

 
We also investigated the rate of formation of 11ß-DNA adducts over a 15-hour period in LNCaP cells treated with a single dose of 10 µmol/L [14C]-11ß-dichloro. Figure 5B shows that the concentration of 11ß-DNA adducts in these cells increased at a constant rate during the 15-hour period. The slope of the line in Fig. 5B implies a rate of adduct formation of 0.25 adducts/106 bases per hour. This rate of adduct formation correlates well with the observed number of adducts present in LNCaP cells after 4-hour exposure to 10 µmol/L 11ß-dichloro (Fig. 5A). These results are also consistent with the stability of the 11ß-dichloro compound in cell culture media because rapid destruction of the compound by hydrolysis or metabolism would be expected to decrease the rate of 11ß adduct formation.

Figure 5C shows the dose-response relationship for growth inhibition of LNCaP cells by 11ß-dichloro. The ED50 for growth inhibition calculated from these data is 5.3 µmol/L.

Identification of 11ß-DNA Adducts in Liver and Xenograft Tumor Tissue
The effectiveness of 11ß-dichloro in inhibiting the growth of LNCaP prostate tumor cells growing as s.c. xenografts in mice (12) led us to examine the formation of 11ß-DNA adducts in xenograft and normal tissues. First, to establish the identity of adducts that were formed in tissues of mice treated with 11ß-dichloro, we investigated the presence of 11ß-DNA adducts in liver because this tissue is exposed to the highest concentration of 11ß-dichloro (Fig. 2). DNA was isolated from liver tissue of mice 4 hours after administration of 11ß-dichloro (45 mg/kg, i.p.), hydrolyzed, and analyzed directly by electrospray ionization mass spectrometry. As in the case of studies in LNCaP cells in culture, we searched for molecular ions of ≥600 m/z and found one abundant ion of 813.5 m/z (Fig. 4C). Further analysis of this ion by collision-induced dissociation also produced fragment ions at m/z 662.5 and m/z 372.1. Thus, the DNA adduct formed by direct reaction of 11ß-dichloro with DNA in vitro is identical by mass spectrometry to the adduct formed in DNA of cells in culture and in tissues in vivo. The fact that the intact 11ß molecule is covalently linked to liver DNA indicates that the molecule has sufficient stability in vivo for distribution and penetration into tissues and, hence, is available to do its intended biological functions.

Having evidenced formation of DNA adducts by intact 11ß-dichloro in vivo, we proceeded to investigate tumor concentrations of the 11ß adducts in an animal xenograft model. Mice bearing xenograft LNCaP prostate tumors were administered a single dose of 50 mg/kg [14C]-11ß-dichloro (specific activity, 0.5 mCi/mmol). After 4 hours, samples of tumor and liver tissues were obtained and DNA was isolated and subjected to 14C accelerator mass spectrometry analysis as described above. The results of these analyses are presented in Table 1 . Based on the number of amol 14C/µg DNA, tumor tissue had ~12% of the concentration of 11ß-DNA adducts as were found in the liver (Table 1). Because accelerator mass spectrometry analysis cannot reveal the identity of the radiolabeled species, it will require further investigation to confirm that the results of our analyses represent the presence of the 11ß-guanine adduct in tumor tissues. Nonetheless, these data ascertain the ability of 11ß-dichloro to react with one of its intended molecular targets (DNA) in tumor tissue.


View this table:
[in this window]
[in a new window]
  		Table 1. [14C] accelerator mass spectrometry analysis of 11ß-DNA adducts in liver tissue and LNCaP tumor xenografts

 
Investigation of Hepatotoxic Effects of 11ß-Dichloro
An animal xenograft tumor study conducted to investigate the efficacy of 11ß-dichloro against prostate cancer found little overt evidence of toxicity throughout a 45-day seven-course regimen using a dose of 30 mg/kg (12). Because 11ß-dichloro concentrations were highest in the liver and this tissue was also found to experience higher levels of DNA damage than tumor tissue, we investigated whether four serum markers of hepatotoxicity were elevated after single or multiple doses of 11ß-dichloro. In the case of acute exposures, blood was collected from NIH Swiss nu/nu mice 24 hours after they had received a single i.p. dose of 10, 30, 50, or 75 mg/kg 11ß-dichloro. Blood was also collected from NIH Swiss nu/nu mice that had been treated repeatedly with 30 mg/kg 11ß-dichloro on the same regimen used in the xenograft tumor study (12). Serum levels of four enzymes that are frequently elevated in drug toxicity were measured in treated animals and compared with levels in animals administered vehicle only as well as those of untreated historical controls.

Increased levels of alanine aminotransferase and aspartate aminotransferase were found in animals treated with ≥50 mg/kg 11ß-dichloro. After a single administration, we observed no changes in either {gamma}-glutamyl transferase or alkaline phosphatase at doses up to 75 mg/kg 11ß-dichloro (Table 2 ) whereas alanine aminotransferase and aspartate aminotransferase levels were elevated from 35- to 40-fold at this dose. No significant changes in the serum markers of liver toxicity were found in mice treated with 10 or 30 mg/kg. Furthermore, normal levels of all four serum enzymes were found in animals that had received repeated doses during seven courses of five daily doses of 30 mg/kg (Table 2).


View this table:
[in this window]
[in a new window]
  		Table 2. Serum enzyme levels in mice after single or repeated administration of 11ß-dichloro

 
We also examined blood reticulocytes as a measure of bone marrow toxicity in mice treated for an extended period of time with the 30 mg/kg dose. The average level of reticulocytes in treated mice (6.9 x 106/mm3; SD, 2.3) was not statistically different from vehicle-treated mice (7.5 x 106/mm3; SD, 1.3). Taken together, these data show that 11ß-dichloro has acceptable toxicity at the level shown to effectively inhibit tumor growth.

Discussion

In this study, we have identified a covalent DNA adduct formed in cells in culture and tissues of mice treated with a novel bifunctional compound, 11ß-dichloro. A ligand for the androgen receptor linked to a reactive N,N-bis-(2-chloroethyl)aniline mustard composes the 11ß-dichloro molecule. These chemical features enable 11ß-dichloro to form covalent DNA adducts capable of binding the androgen receptor. It is proposed that androgen receptor-DNA adduct complexes interfere with both DNA repair and androgen receptor function (see Fig. 1). We have shown that 11ß-dichloro rapidly induces apoptosis in LNCaP prostate cancer cells and that the new compound can prevent tumor growth in vivo (12). This earlier study found that it is essential to have the DNA-damaging and androgen receptor ligand moieties in the same molecule to observe potent activity against cancer cells.

The results of this study show that, following administration by i.p. injection, the intact 11ß-dichloro molecule forms DNA adducts in cells in culture as well as in mouse tissues after absorption and distribution. Levels of radioactivity were widely distributed in tissues after administration of [14C]-11ß-dichloro. Chromatographic analysis of radiolabeled compounds extracted from blood revealed that intact 11ß-dichloro was the predominant molecular species present for up to 4 hours. Several unidentified metabolites represented approximately one third of the radioactivity present in blood at that time.

These findings answer a key question about the relevance of our proposed mechanisms of action of 11ß-dichloro to the antitumor effects of this compound in vivo. The fact that the compound remains intact and capable of forming DNA adducts in cells and tissues is consistent with the view that its combination of biochemical functions may underlie its ability to prevent tumor growth.

The 11ß-guanine adduct we identified is consistent with the major adduct formed by reaction of bifunctional aniline mustard drugs with DNA (15–18) and is the product of the reaction of one arm of the N,N-bis-(2-chloroethyl) group with a guanine base. Reaction of the second arm of this group with another DNA base can produce intrastrand or interstrand cross-links (17–19). Evidence indicates that these bifunctional adducts are primarily responsible for the lethal effects of this class of compounds (20, 21). The efficiency of cross-link formation by nitrogen mustards is not great. Thus, any change in the balance between the removal of 11ß-monoadducts from DNA and their conversion to cross-links will likely have a significant effect on toxicity. Whether the ability of 11ß-DNA adducts to bind the androgen receptor affects this balance requires further investigation.

We found that the levels of 11ß-DNA adducts formed in LNCaP xenograft tumors after a single administration of 11ß-dichloro were in the range of the adduct levels associated with concentrations that inhibited the growth of LNCaP cells in culture. The ED50 for growth inhibition of LNCaP cells in culture by 11ß-dichloro is 5.2 µmol/L. Approximately 0.25 adduct per million DNA bases was found in LNCaP cells after a 4-hour exposure to 2.5 µmol/L 11ß-dichloro in culture. A single dose of 50 mg/kg 11ß-dichloro resulted in a similar level of DNA damage in LNCaP xenografts (average, 0.2 adducts/106 bases).

The measured adduct levels in LNCaP cells within xenograft tumors, however, are likely to vary because of the heterogeneity of the tumor microenvironment. Whereas cells in culture experience uniform exposures to 11ß-dichloro in media, vascularization and rates of perfusion vary within tumors resulting in nonuniform exposures (22, 23). Thus, the levels of DNA adducts we identified should be considered an average with some cells likely having greater amounts of DNA damage and cytotoxicity from 11ß-dichloro whereas others likely experience less.

The levels of 11ß-DNA adducts in liver were 8-fold greater than in tumor tissue. This finding led us to investigate whether single or repeated treatments with 11ß-dichloro resulted in hepatotoxicity. In our examination of four serum enzymes that are diagnostic for liver disease, we found that significant increases in both alanine aminotransferase and aspartate aminotransferase occurred at the 50 mg/kg 11ß-dichloro dose. No increases were found in either {gamma}-glutamyl transferase or alkaline phosphatase. This pattern is consistent with drug-induced injury to hepatocytes after the single dose. At a lower repeated dose of 30 mg/kg over a 7-week period with five consecutive daily doses, none of the four serum enzymes levels were elevated. The fact that inhibition of tumor growth occurred when animals were treated repeatedly with the 30 mg/kg dose in the absence of significant toxicity is encouraging (12).

We propose that in addition to inhibiting DNA repair, association of the androgen receptor with DNA adducts may antagonize androgen receptor transcriptional activity. This unique mechanism of disabling androgen receptor–mediated gene transcription in hormone-refractory prostate cancer may prove effective against cancers in which overexpression or mutation of the androgen receptor or the changes in androgen receptor coregulators underlie escape from androgen blockade (24–26).

Targeting androgen receptor function in hormone-refractory prostate cancer is often ineffective because of the variety of mechanisms that enable tumor cells to defeat current antihormonal therapies. Nonetheless, the androgen receptor remains an attractive therapeutic target for this disease because of its continued role in the growth and survival of advanced prostate cancers. Elimination of androgen receptor function can leave cells vulnerable to other chemotherapeutic agents such as those that act by damaging DNA (7). We propose that 11ß-DNA adducts that capture the androgen receptor could provide a novel way of antagonizing its biological functions in hormone-refractory prostate cancer. The combined effects of persistent DNA damage and the unique mechanism of receptor antagonism by 11ß-DNA adducts may result in the disruption of biochemical pathways and provoke prostate cancer cells into apoptosis, leading to more effective therapy.

Acknowledgments

We thank Jeff Bajko and Ellen Buckley from the Department of Comparative Medicine at Massachusetts Institute of Technology for their assistance with the handling and processing of blood samples; Agilent Technologies for access to mass spectrometers; and Dr. Marr of Agilent for helpful discussions.

Footnotes

Grant support: Department of Defense grant DAMD 17-03-1-0085 and NIH grants R01 CA 77743-06 (J.M. Essigmann) and P30-ES02109 (Massachusetts Institute of Technology Center for Environmental Health Sciences).

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.

Received 11/ 8/05; revised 1/ 2/06; accepted 2/15/06.

References

  1. Weir HK, Thun MJ, Hankey BF, et al. Annual report to the nation on the status of cancer, 1975-2000, featuring the uses of surveillance data for cancer prevention and control. J Natl Cancer Inst 2003;95:1276–99.[Abstract/Free Full Text]
  2. Bruckheimer EM, Gjertsen BT, McDonnell TJ. Implications of cell death regulation in the pathogenesis and treatment of prostate cancer. Semin Oncol 1999;26:382–98.[Medline]
  3. Kozlowski JM, Ellis WJ, Grayhack JT. Advanced prostatic carcinoma. Early versus late endocrine therapy. Urol Clin North Am 1991;18:15–24.[Medline]
  4. Kreis W. Current chemotherapy and future directions in research for the treatment of advanced hormone-refractory prostate cancer. Cancer Invest 1995;13:296–312.[Medline]
  5. Santen RJ. Clinical review 37: Endocrine treatment of prostate cancer. J Clin Endocrinol Metab 1992;75:685–9.[CrossRef][Medline]
  6. Gioeli D. Signal transduction in prostrate cancer progression. Clin Sci (Lond) 2004;108:293–308.
  7. Heinlein CA, Chang C. Androgen receptor in prostate cancer. Endocr Rev 2004;25:276–308.[Abstract/Free Full Text]
  8. Ruizeveld de Winter JA, Janssen PJ, Sleddens HM, et al. Androgen receptor status in localized and locally progressive hormone refractory human prostate cancer. Am J Pathol 1994;144:735–46.[Abstract]
  9. Wang LG, Ossowski L, Ferrari AC. Androgen receptor level controlled by a suppressor complex lost in an androgen-independent prostate cancer cell line. Oncogene 2004;23:5175–84.[CrossRef][Medline]
  10. Buchanan G, Irvine RA, Coetzee GA, Tilley WD. Contribution of the androgen receptor to prostate cancer predisposition and progression. Cancer Metastasis Rev 2001;20:207–23.[CrossRef][Medline]
  11. Heinlein CA, Chang C. Androgen receptor (AR) coregulators: an overview. Endocr Rev 2002;23:175–200.[Abstract/Free Full Text]
  12. Marquis JC, Hillier SM, Dinaut AN, et al. Disruption of gene expression and induction of apoptosis in prostate cancer cells by a DNA-damaging agent tethered to an androgen receptor ligand. Chem Biol 2005;12:779–87.[Medline]
  13. Mitra K, Marquis JC, Hillier SM, et al. A rationally designed genotoxin that selectively destroys estrogen receptor-positive breast cancer cells. J Am Chem Soc 2002;124:1862–3.[Medline]
  14. Liberman RG, Tannenbaum SR, Hughey BJ, et al. An interface for direct analysis of 14c in nonvolatile samples by accelerator mass spectrometry. Anal Chem 2004;76:328–34.[Medline]
  15. Bank BB. Studies of chlorambucil-DNA adducts. Biochem Pharmacol 1992;44:571–5.[Medline]
  16. Kohn KW, Hartley JA, Mattes WB. Mechanisms of DNA sequence selective alkylation of guanine-N7 positions by nitrogen mustards. Nucleic Acids Res 1987;15:10531–49.[Abstract/Free Full Text]
  17. Osborne MR, Lawley PD. Alkylation of DNA by melphalan with special reference to adenine derivatives and adenine-guanine cross-linking. Chem Biol Interact 1993;89:49–60.[CrossRef][Medline]
  18. Povirk LF, Shuker DE. DNA damage and mutagenesis induced by nitrogen mustards. Mutat Res 1994;318:205–26.[Medline]
  19. Haapala E, Hakala K, Jokipelto E, Vilpo J, Hovinen J. Reactions of N,N-bis(2-chloroethyl)-p-aminophenylbutyric acid (chlorambucil) with 2'-deoxyguanosine. Chem Res Toxicol 2001;14:988–95.[Medline]
  20. Ross WE, Ewig RA, Kohn KW. Differences between melphalan and nitrogen mustard in the formation and removal of DNA cross-links. Cancer Res 1978;38:1502–6.[Abstract/Free Full Text]
  21. Sunters A, Springer CJ, Bagshawe KD, Souhami RL, Hartley JA. The cytotoxicity, DNA cross-linking ability and DNA sequence selectivity of the aniline mustards melphalan, chlorambucil and 4-[bis(2-chloroethyl)amino] benzoic acid. Biochem Pharmacol 1992;44:59–64.[CrossRef][Medline]
  22. Jain RK, Munn LL, Fukumura D. Dissecting tumour pathophysiology using intravital microscopy. Nat Rev Cancer 2002;2:266–76.[CrossRef][Medline]
  23. Stohrer M, Boucher Y, Stangassinger M, Jain RK. Oncotic pressure in solid tumors is elevated. Cancer Res 2000;60:4251–5.[Abstract/Free Full Text]
  24. Berrevoets CA, Umar A, Brinkmann AO. Antiandrogens: selective androgen receptor modulators. Mol Cell Endocrinol 2002;198:97–103.[CrossRef][Medline]
  25. Gaddipati JP, McLeod DG, Heidenberg HB, et al. Frequent detection of codon 877 mutation in the androgen receptor gene in advanced prostate cancers. Cancer Res 1994;54:2861–4.[Abstract/Free Full Text]
  26. Montgomery JS, Price DK, Figg WD. The androgen receptor gene and its influence on the development and progression of prostate cancer. J Pathol 2001;195:138–46.[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hillier, S. M.
Right arrow Articles by Croy, R. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hillier, S. M.
Right arrow Articles by Croy, R. G.

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