in vivo and in vitro evaluation of the alkylating agent...

[CANCER RESEARCH 51. 4581-4587, September I, 1991]

In Vivo and in Vitro Evaluation of the Alkylating Agent CarmethizoleWilliam L. Elliott,1 David W. Fry, Wayne K. Anderson, James M. Nelson, Kenneth E. Hook, Peggy A. Hawkins,

and W. R. Leopold IIIParke-Davis Pharmaceutical Research Division, Ann Arbor, Michigan 48105 Â¡W.L. E., D. W. F., J. M. N., K. E. H., P. A. H., W. K. L.J and College of Pharmacy, StaleUniversity of New York, Buffalo, New York 14260 [W. K. A.J

ABSTRACT

Carmethizole, a novel bis-carbamate alkylating agent, was evaluatedin vitro for potential mechanisms of interaction with DNA and in vivofor spectrum and degree of antitumor activity. In vitro, the concentrationof Carmethizole required to produce a 50% reduction in clonogenic cellsurvival was identical in 06-alkylguanine DNA alkyltransferase-positiveand -negative human cell lines. The CHO cell line UV4, hypersensitiveto mono- and bifunctional alkylating agents, was 37-fold more sensitiveto Carmethizole than normal cells. The UVS cell line, which is nothypersensitive to cross-linkers, was 13-fold more sensitive to carmethi-zole than normal cells. Alkaline elution studies in LI 210 cells exposedto Carmethizole showed the presence of DNA-protein and DNA-DNAcross-links but not DNA strand breaks. These data suggested that theinteraction of Carmethizole with DNA produces monoadducts, DNA-protein, and DNA-DNA interstrand cross-links at several sites.

In vivo, Carmethizole was not cross-resistant with l,3-bis(2-chloroe-thyl)-i-nitrosourea or Cytoxan as determined by testing against P388leukemias resistant to the latter 2 agents. Carmethizole activity wassimilar to that of melphalan across the murine solid tumor panel, whichconsisted of B16 melanoma; colon adenocarcinomas 1la, 26, and 36; andthe KHT sarcoma. Carmethizole, Cytoxan, and melphalan were all activeand had comparable activity against the IK "I-K and MX-ÃŒhuman tumor

xenografts. The in vivo spectrum of activity and efficacy of Carmethizolewas similar to that of melphalan.

INTRODUCTION

Alkylating agents that interact with DNA have an importanttherapeutic role in the treatment of malignant disease. Carmethizole (NSC 602688) (Fig. 1) is an imidazole biscarbamatewith structural homology to the pyrrolizine biscarbamate anti-cancer compounds. However, Carmethizole has enhanced watersolubility due to its basicity, permitting salt preparation andgreater solution stability over the pyrrolizine compounds (1).Carmethizole has demonstrated antitumor activity in severalmurine tumor and human tumor xenograft models (1-3). Thepharmacokinetics of Carmethizole in mice and beagles has beenreported (4).

The goal of the current investigation was to evaluate thetherapeutic potential of Carmethizole relative to currently approved alkylating agents such as BCNU,2 Cytoxan, and mel

phalan. In the model systems examined, the data indicate thatCarmethizole is similar to melphalan in both its degree andspectrum of antitumor activity but is somewhat different in itsmechanism of interaction with DNA.

MATERIALS AND METHODS

Chemicals.Tetrapropylammonium hydroxide was from the EastmanKodak Co., Rochester, NY. [me/A>>/-14C]Thymidineand [methyl-'H]

Received 1/28/91; accepted 6/13/91.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 inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1To whom requests for reprints should be addressed, at Parke-Davis Pharmaceutical Research Division, Warner-Lambert Co., 2800 Plymouth Road, AnnArbor, MI 48105-2430.

2The abbreviations used are: BCNU, l,3-bis(2-chloroethyl)-l-nitrosourea;BrdUrd, 5-bromo-2-deoxyuridine; PI, propidium iodide; FBS, fetal bovine serum;LCÂ«hconcentration of test agent to kill 50% of the cells.

thymidine were purchased from Amersham, Arlington Heights, IL.Carmethizole was obtained from Dr. Wayne Anderson, State Universityof New York, Buffalo, NY. Melphalan and Cytoxan were from SigmaChemical Co., St. Louis, MO. BCNU was from Bristol-Meyers Co.,Syracuse, NY.

Cell Lines. LI210 murine leukemia cells were grown as a suspensionin RPMI 1640 (GIBCO, Grand Island, NY) supplemented 10% FBSand 50 Mg/ml gentamicin (Sigma). P388 leukemia cells (never morethan 4 in vitro passages from in vivo) and CHO cells were grown assuspensions in Ã -minimal essential media (GIBCO) supplemented with10% FBS and 50 Mg/ml gentamicin (Sigma). The MR1 and A1235human astrocytoma cell lines were provided by Dr. D. B. Yarosh(Applied Genetics, Inc.) and were cultured as monolayers in Dulbecco'smodified Eagle's medium (GIBCO) with nonessential amino acids plus

0.1 Mg/ml biotin, 1.36 Mg/ml vitamin B,2, 20% undialyzed FBS, and 50Mg/ml gentamicin (all from Sigma). Cells were grown at 37Â°Cin 5%

COi in air. When treated, cells were in the early to mid-log phase ofgrowth. Viability was greater than 95% as determined by trypan blueexclusion.

DNA Strand Scission and DNA-Protein Cross-Linking Assays. DNAsingle-strand breaks were measured by alkaline elution as described byKohn et a!. (5). LI210 cells in early log growth were grown in 1 MM[methyl-"C]ÃhymiÃ¡Ãne,at a specific activity of 0.02 MCi/nmol or 1 MMlmethyl-3H]thym\aine at 0.1 MCi/nmol for 24 h. The [14C]thymidine-

labeled cells were exposed to drug at a specified concentration and timeinterval, after which 5x10* cells were mixed with an equal number of[3H]thymidine-labeled cells that had been previously X-irradiated with

300 rads as an internal standard.The cells were placed on polycarbonate filters and lysed by passing

5 ml of 2% SDS-25 mM EDTA, pH 9.7, through the filter. Elutingfractions (20 Ml/min with 1-h fractions) were collected in scintillationvials and mixed with Ready-Solv (Beckman) containing 0.7% aceticacid. Counting efficiencies were determined with 14Cand 3H standards

(Amersham), and radioactivity contributed by each isotope determinedby the method of dual channel counting. Data were graphed as a log-log plot of I4C versus 3H retained on the filter, and single strand break

frequency expressed as rad equivalents was calculated as describedpreviously (5).

The frequency of DNA-protein cross-links was assessed by themethod described by Zwelling et al. (6). Drug-treated or untreatedcontrol 14C-labeled cells were X-irradiated at 4Â°Cwith 500 rads andcombined with an equal number of 3H-labeled cells irradiated with 300rads. Elution methods were identical to those for single-stranded breaksexcept for the following. The cells were placed on 0.8-Mm pore sizepoly(vinylchloride) acrylic copolymer filters (Metricell DM-800; Gel-man Sciences, Ann Arbor, MI), and the lysis solution contained 2 MNaCl, 0.2% sodium dodecyl sarcosine, and 50 mM EDTA, pH 10.Treatment with Proteinase K was also omitted. DNA-protein crosslinks were calculated using the bound-to-one-terminus model of Rosset al. (7). DNA-DNA interstrand cross-links were assayed and quanti

tated as described by Kohn et al. (5).Those procedures requiring X-irradiated cells used a 120-kV, 5-mA

X-ray tube (Torr X-ray, Burlington, MA) set in a self-contained lead-lined cabinet (Test Equipment Distributors, Troy, MI). Cells were kepton ice during and after irradiation until they were lysed on the filters.Radiation doses were monitored with a model 500 dual polarity electrometer, equipped with a 0.33-ml probe (Victoreen, Inc., Cleveland,OH).

Growth Inhibition. Dilutions of test compounds in 20 ><\of waterwere placed into wells of 24-well Linbro plates (1.7 x 1.6 cm flatbottomed) followed by the addition of CHO cells (3 x 104/ml) in 2 mlof media. Plates were incubated for 3 days at 37Â°Cin a humidified

4581

on May 23, 2021. © 1991 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

CARMETHIZOLE ACTIVITY IN VITRO AND IN VIVO

CH3HNCO

CH3HNCO

II0

Fig. 1. Chemical structure of carmethizole (l//-imidazole-4,5-dimethanol, 1-methyl-2-(methylthio)-,bis(methylcarbamate)(ester),monohydrochloride).

1001,

DOSE (uM)

Fig. 2. In vitro activity of carmethizole and BCNU against MR1 (mer+) andA1235 (merâ€”)cell lines. Cells were seeded in tissue culture flasks and allowed togrow to midlog phase. Cell cultures were exposed to various concentrations ofeither carmethizole or BCNU for 3 h. The cells were washed with drug-freemedium, and single cell suspensions were plated at various cell densities. After10 to 11 days, colonies were fixed, stained, and counted to determine percentsurvival. O, BCNU versus A1235; â€¢,BCNU versus MR1; *. carmethizole versusA1235; O, carmethizole versus MR1; A, melphalan versus A1235; A, melphalanversus MR1. Data for BCNU and carmethizole are the means of 2 experiments.Data for melphalan are from a single experiment.

Table 1 Growth inhibition ofDNA repair-deficient CHO cells by carmethizole ormelphalan

Cellline"AA8

UV4UV51C,Â»

(MM)*Carmethizole

Melphalan166

Â±22 4.9 Â±0.14.5 Â±0.4 0.02 Â±0.0113 Â±2.2 5.1 Â±0.6Hypersensitivity

tocarmethizole'37

13Â°AA8 cells are the wild-type parent cells with intact excision repair. UV4 cells

exhibit hypersensitivity to far-UV light, chemicals that form large monoadducts,and bifunctional agents. They are deficient in the incision step of excision repairand cross-link removal. UV5 cells are hypersensitive to far-UV and chemicalsthat form large monoadducts and are deficient in the incision step of excisionrepair. See Ref. 11 for additional characteristics of these cell lines.

* Values represent the mean of 3 experiments Â±SE.' The ICjo of carmethizole-treated normal cells divided by the IC!0 of carmeth-

izole-treated repair-deficient cells.

atmosphere containing 5% CO2 in air. Cell growth was determined bycell count with a Coulter model ZM electronic cell counter (CoulterElectronics, Inc., Hialeah, FL).

Clonogenic Survival. The MR1 and the A1235 cell lines were seededinto separate 25-cm2 tissue culture flasks (Corning Laboratories, Corning, NY) at 5 x IO5 cells/flask. The cells were allowed to grow untilthey reached mid-log phase of the growth cycle. This period was 4 and5 days for the MR1 and A1235 cell lines, respectively. The cell cultureswere exposed to various concentrations of either carmethizole, BCNU,or melphalan for 3 h. Cells were then trypsinized, washed once with

drug-free medium, and seeded at various densities in 60-mm Petridishes (Corning). After 10 to 11 days, colonies were fixed with 70%ethanol for 5 min, stained with 0.05% Giemsa stain 620 in water(Diagnostic Systems, Inc., Gibbstown, NJ) for 20 min, and colonies(>75 cells) counted to determine clonogenic survival. The LC50s weredetermined by a best fit log linear regression over the linear portion ofthe curve.

Cell Cycle Studies. For the in vitro cell cycle studies, P388 cells weregrown in a spinner flask overnight at 37Â°Cin 5% CO2 in air. Ten mlof this suspension (2 x 10s cells/ml) were distributed into 25-cm2 tissue

culture flasks (Corning) in which the cells were exposed to carmethizole(0, 10, 50, 100, and 500 UM) for 24 h. The cells were treated with 10MMBrdUrd (Sigma) during the last 30 min of carmethizole exposure.Following centrifugation, cells were stained for BrdUrd and PI (8).Briefly, cells were fixed in 70% ethanol, incubated for 30 min with anti-BrdUrd antibodies (Becton Dickinson, Mountain View, CA), washed,and incubated with FITC-conjugated goat anti-mouse F(ab') antibody

FRACTION OF [3H] RETAINED

Fig. 3. Alkaline elution profile for detection of DNA single strand breaks inLI 210 cells exposed to carmethizole. Cells were exposed to the following concentrations of drug for 3 h. Details are given in "Materials and Methods." O,

untreated control; A, 500 MMcarmethizole; D, l mM carmethizole.


Fig. 4. Alkaline elution profile for detection of DNA interstrand cross-links inL1210 cells exposed to carmethizole. Cells were exposed to the indicated concentrations of drug for 3 h. Details are given in "Materials and Methods." O,Untreated control; D, 500 rads of X-irradiation; O, 500 rads of X-irradiation +500 MMcarmethizole; A, 500 rads of X-irradiation + 1 mM carmethizole.

4582





Fig. 5. Alkaline elution profile for the detection of DNA-protein cross-links inL1210 cells exposed to carmethizole. Cells were exposed to the indicated concentrations of drug for 3 h. Details are given in "Materials and Methods." O,untreated control; A, 500 rads of X-irradiation; D, 500 rads of X-irradiation +500 MMcarmethizole; O, 500 rads of X-irradiation + 1 mM earmethizole.

Table 2 Production of DNA-protein cross-links in L1210 cells exposed tocarmethizole

Cells were exposed to the indicated concentrations of carmethizole for 3 h.Values represent the mean of 2 experiments.

ConcentrationDNA-protein cross-links

(rad equivalents)

100 MM500 MM

1 mM

43151332

(Cappel, Malvern, PA) for 30 min to stain the BrdUrd incorporatedinto newly synthesized DNA. The cells were centrifuged and resus-pended in phosphate-buffered saline containing 20 /in/ml PI (Sigma)to stain total DNA. Samples were analyzed on a fluorescence-activatedcell sorter analyzer (Becton Dickinson). Red (PI) fluorescence wasmeasured with a linear amplifier through a DF 650/60 filter (OmegaOptical), while green (BrdUrd) fluorescence was measured with a logamplifier after passing through a BP 535/40 filter (Becton Dickinson).

P388 leukemia cells were also labeled in vivo by i.p. injection ofBrdUrd. Mice (3/group) were given one injection of 200 mg/kg carmethizole, i.p. Ascites fluid was harvested and pooled 2, 4, 8, 12, and24 h after carmethizole treatment. BrdUrd (100 mg/kg) was injectedi.p. 30 min before cell harvest at all time points. Mononuclear cellswere isolated on Ficoll-Paque (Pharmcia, Piscataway, NJ) and werefurther processed for staining of BrdUrd and total DNA as above.

Two similar studies were done with the KHT sarcoma. Mice (3/group) bearing 300-mg s.c. KHT sarcomas were selected for drugtreatment. They received 50, 100, or 200 mg/kg carmethizole or 0.5ml of diluent. In one experiment, carmethizole-treated mice were giveninjections of BrdUrd 30 min before tumor harvest at 24 h post-carmethizole injection. In the other experiment, carmethizole-treatedmice were given injections of BrdUrd 30 min after carmethizole injection. These mice were held for 24 h before tumor harvest. This wasdone to determine if the cells would continue to cycle after treatmentwith carmethizole. Tumors were then excised, teased apart with pipets,and disaggregated by a 15-min exposure to an enzyme cocktail consisting of Pronase (Calbiochem, San Diego, CA), 5000 units/ml; collagen-ase (Millipore Corp., Bedford, MA), 20.6 units/ml; and DNase (Wor-thington, Freehold, NJ), 280 units/ml. The cells were filtered throughgauze, layered over Ficoll-Paque, centrifuged, and then processed forstaining of BrdUrd and total DNA as above.

In Vivo Chemotherapy. Immunocompetent mice were housed in wire-bottomed cages on automatic flush racks. Immunodeficient mice (NCr-nu) used for human tumor xenograft testing and passage were housed

in microisolator cages in a barrier facility. All mice were provided withfood and water ad libitum, and maintained on a 12-h light/dark cycle.Housing was in accordance with American Association for Accreditation of Laboratory Animal Care guidelines. Tumor passage was carriedout in the inbred strain of tumor origin (9).

Anticancer activity against murine tumors was evaluated in F, hybridmice derived from the inbred strain of tumor origin. For all tests, themice ( 18 to 20 g) were randomized and then inoculated with countedtumor cells or trocar fragments as appropriate on Day 0. The micewere then rerandomized for distribution into treatment cages. The testcompounds Cytoxan, melphalan, and carmethizole were routinely dissolved in 0.9% aqueous NaCl. BCNU was dissolved in ethanol to which0.9% aqueous NaCl was added to obtain a final ethanol concentrationof 2%. Controls were treated with 0.9% aqueous NaCl. For all tests themice were weighed on the first and last days of treatment for dailytreatment schedules or on each treatment day for intermittent treatmentschedules. Additional weighings were made weekly for tumor growthdelay assays. Host body weight change data are the difference betweenthe mean group weights on the first and last days of treatment for lifespan assays, and the maximum drug-related weight loss for growthdelay assays. If animals gained weight during treatment, weight isrecorded as a (+) for the growth delay assays.

Life span and growth delay assays were evaluated on the basis of%T/C (median life span of treated mice/median life span of controlmice x 100%) and T-C (the median difference, in days, for the treatedand control tumors to reach a predetermined evaluation size), respectively (9). Cures were excluded from these calculations. In both assaysan estimation of the net tumor cell kill (logs) was made by standardmethods (9, 10). A positive net log cell kill indicates that the tumorburden at the end of therapy was smaller than that at the beginning oftreatment. Negative net cell kill values indicate that the tumor grew


Fig. 6. Alkaline elution profile for detection of DNA interstrand cross-links inLI210 cells exposed to carmethizole. Cells were exposed to the indicated concentrations of drug for I h. The cells were then washed free of the drug and allowedto incubate a further 3 h in drug-free media before irradiation and analysis byalkaline elution. O, untreated control; A, 500 rads X-irradiation; D, 500 rads X-irradiation + 100 MMcarmethizole; O, 500 rads X-irradiation + 500 MMcarmethizole; â€¢,500 rads of X-irradiation + 1 mM carmethizole.

Table 3 Production of DNA-inters! rand cross-links in LI 210 cells exposed tocarmethizole

Cells were incubated with the indicated drug concentrations for 1 h, washed,resuspended into drug-free media, and incubated for an additional 3 h. Valuesrepresent the mean of 2 experiments.

Carmethizole(MM)

DNA-interstrand cross-links(rad equivalents)

100 MM500 MM

1 HIM

22174578

4583



CARVIETHIZOLE ACTIVITY IN VITRO AND IN VIVO

Table 4 In vivo activity of carmethizole against murine Â¡eukemias"

DoseSchedule (mg/kg/ %T/C Net log

Tumor Drug (day no.) injection) (cures)* cell kill

P388P388/BCNUP388/CytoxanP388/melphalancCarmethizoleCytoxanBCNUCarmethizoleCytoxanBCNUCarmethizoleCytoxanBCNUCarmethizole3-113,7, 113,

7, 113,

7, 113,

7, 113,

7, 113,

7,113,

7,113,

7,113.

7, 111-5389510016016019199510010016019199510010010019195150206

(2/6)213(2/6)196(2/6)(6/6)(6/6)234

(2/6)391(4/6)254

(2/6)310(2/6)(6/6)328

(4/6)110107197154102105229220

(2/6)1111165.26.03.27.06.77.06.77.67.67.67.6-1.1-1.22.80.4-2.1-0.75.38.0NAÂ¿NA

Â°Leukemia cells were implanted i.p. on Day 0. Treatments were administeredi.p. on the days indicated.Leukemia cell burdens on Day 3 ranged from 7x10'to 1 x 10* cells depending on the cell line. Optimal doses from a complete dose

response are reported.Sixty-day tumor-free survivors, these mice are excluded from percent cures/

total animals/group calculations.c Data from Division of Cancer Treatment, National Cancer Institute.d NA, not available.

during the treatment regimen. In each experiment, all compounds weretested over a complete dose response (toxic to nontoxic). For life spanassays the optimal dose was reported, i.e., the dose yielding the maximum %T/C value. For growth delay assays, the maximum tolerateddose (i.e., <10% lethal dose) was reported.

RESULTS

Over a complete dose response range in vitro, both carmethizole and melphalan, a clinically used nitrogen mustard, wereas active against A1235 (mer-) cells as they were against MR1

(mer+) cells (Fig. 2. A1235 cells, in contrast to MR1 cells, aredeficient in C^-alkylguanine DNA alkyltransferase and are unable to repair O-alkylguanine lesions. The LC50for carmethizolewas 20 MMin both cell lines, whereas the LC5o for melphalanwas l UM in both lines. In contrast, BCNU was 8-fold moreactive against A1235 cells, the mer- line. To further character

ize the type of DNA adduct formed by carmethizole, 2 strainsof Chinese hamster ovary cells were examined that are hypersensitive to certain classes of alkylators (Table 1). Both UV4and UV5 cells were hypersensitive to carmethizole relative tothe parent CHO line. Carmethizole was most active againstUV4 cells, which are hypersensitive to both mono- and bifunc-tional alkylators. The UV5 cell line, which is not sensitive toDNA interstrand cross-linkers, was 3-fold less sensitive tocarmethizole than the UV4 cell line but still 13-fold moresensitive than normal cells. The UV4 line was extremely hypersensitive to melphalan and carmethizole, whereas the UV5line was sensitive only to carmethizole and not melphalan.

The possibility that carmethizole may produce DNA strandbreaks, alkaline labile sites, DNA-protein cross-links, or DNAinterstrand cross-links was investigated by filter elution techniques. Fig. 3 shows an alkaline elution profile designed todetect DNA single strand breaks in L1210 cells exposed for 3h to carmethizole. Concentrations as high as 1 HIM producedno DNA scissions. Fig. 4 indicates that under the same condi

tions as in Fig. 3, no interstrand cross-links were produced (i.e.,there was no decrease in the elution rate of drug-treated cellsirradiated with 500 rads compared to irradiated control cells).In contrast, substantial numbers of DNA-protein cross-linksdid form under these conditions (Fig. 5) and are quantitated asrad equivalents in Table 2. Considering these results, DNAinterstrand cross-links were again investigated but in a differentmanner. This time, cells were treated with varying concentrations of drug but for only 1 h. The cells were then washed freeof the drug and allowed to incubate a further 3 h in drug-freemedia before irradiation and analysis by alkaline elution. Fig.6 shows that under these experimental conditions DNA interstrand cross-links did form as indicated by a decreasing elutionrate with increasing drug concentrations. These data are expressed numerically as rad equivalents in Table 3.

To expand upon the National Cancer Institute data thatprompted our interest in this novel agent (12), carmethizolewas assessed for activity against alkylating agent-resistant murine leukemias (Table 4). Carmethizole, Cytoxan, and BCNUall had curative or near curative activity against P388 leukemia.Carmethizole retained full activity against P388/BCNU,whereas its activity was slightly reduced against P388/Cytoxan.Thus, carmethizole lacked significant cross-resistance with Cytoxan and BCNU. National Cancer Institute data indicated thatcarmethizole was completely cross-resistant with melphalanagainst P388/melphalan (Table 4).

Carmethizole was evaluated against 5 murine solid tumors:B16 melanoma, colon Ila, colon 26, colon 36, and KHTsarcoma (Table 5. Carmethizole was inactive against colon Ila,colon 36, KHT sarcoma, and B16 melanoma, and had onlyslight activity against colon 26. Overall, carmethizole was inferior to BCNU and Cytoxan, while having a spectrum ofactivity similar to that of melphalan in the murine solid tumorpanel.

Carmethizole was active against the HCT-8 colon and theMX-1 mammary human xenografts. Against HCT-8 carmethizole gave a 15.2-day growth delay that was superior to thatobtained with the 3 standard agents (Table 5). Against the MX-1 xenograft, carmethizole gave a 35.4-day growth delay,whereas growth delays for melphalan and Cytoxan were 47.8and 44.6 days, respectively.

Carmethizole toxicity appeared similar to that of the otheralkylating agents tested. At doses above the 10% lethal dose,animals generally died in a 5- to 12-day period after the lasttreatment with no signs of diarrhea, suggesting that carmethizole is myelosuppressive.

Studies were carried out in vitro and in vivo to determine thecell cycle effects of carmethizole treatment. A 24-h exposure tocarmethizole in vitro caused an accumulation of P388 cells inthe G2M phase of the cell cycle that was dose dependent (Table6).

In contrast, P388 leukemia-bearing mice treated with carmethizole at various time points and given injections of BrdUrd30 min before harvest appeared to show an accumulation ofcells in the G\ phase of the cell cycle (Table 7). The apparentcontradiction between the in vitro and in vivo results with P388leukemia prompted us to repeat these studies with the KHTsarcoma. An accumulation of cells in G2M was found in agreement with the in vitro P388 data (Table 8; Fig. 7). The KHTexperiments also showed an accumulation of some cells in lateS phase (Fig. 7).

To determine whether cells were completely blocked or onlyretarded in G2M after treatment with carmethizole, mice bearing the KHT sarcoma were given injections of BrdUrd (100

4584



CARMETHIZOLE ACTIVITY IN VITROAND IN VIVO

Table 5 In vivo activity of carmethizole against solid tumors"

TumorColon

1laColon

36Colon

26KHTB16HCT-8

(humanxenograft)MX-1

(human xenograft)DrugCarmethizoleCytoxanBCNUCarmethizoleCytoxanBCNUCarmethizoleCytoxanBCNUMelphalanCarmethizoleCytoxanBCNUMelphalanCarmethizoleCytoxanBCNUMelphalanCarmethizoleCytoxanBCNUMelphalanCarmethizoleCytoxanBCNUMelphalanSchedule

Dose(day no.)(mg/kg/injection)3,

7, 1195160303.

7, 1195100303.7,11

62100128.56,

10. 1410016030131,5,9

10016019812.16.20

160220191513,17,21

1603503115Wt

change(g)-0.4+*-1.9-1.5+-4.6+-1.7+-1.9-1.5-0.6-1.5-0.5-0.2-1.0-0.1-0.9-3.5-1.2+-1.4-2.4-3.6-1.6-2.2T-C(days)2.318.418.97.622.5(9/10)c1.97.914.821.3(3/6)14.61.126.1

(1/10)10.81.82.723.88.25.415.210.67.38.235.4(7/10)(9/10)0.347.8(8/10)Net

logcellkill-0.40.70.701.1-0.500.71.30.6-1.74.50.7-1.6-0.92.80-0.50.60.2-0.401.7-0.52.5

" All tumors were implanted s.c. Drug treatments were i.p. on the indicated schedule.* +, continual weight gain during therapy.' Sixty-day tumor free survivors (cures/total animals/group), cures excluded from T-C and net log kill calculations.

Table 6 Effect ofPD 131607 on P388 cell cycle fractions after a 24-hin vitro exposure

Drugconcn.GlM)%G,0

18.91013.750

5.610010.3%S75.858.653.550.8%G2M5.327.740.938.9Increase

incellno.(%

ofcontrol)100533427

mg/kg) followed 30 min later by carmethizole (200 mg/kg).Tumors were harvested 24 h later, a single cell suspensionprepared, and the cells stained for BrdUrd and total DNA.Contour plots (Fig. 8) showed that the block in late S/G2Mwas incomplete, as evidenced by the presence of BrdUrd-labeledcells in G i phase, which could only occur if cells were delayedor partly blocked in G2M but then continued through the cellcycle.

Table 7 Effect of carmethizole on the ascitic cell cyclefractions of P388-bearingmice in vivo'

Time aftercarmethizoletreatment(h)2

48

1224Control%G,57.6

49.359.754.252.3%S34.2

42.534.839.137.4%G2M8.2

7.85.5

6.810.4%G,64.7

60.670.680.485.9Treated%S27.0

30.718.47.53.7%G2M8.3

11.611.012.210.5

Â°Mice (3/group) were treated with carmethizole 200 mg/kg i.p. and were given

injections of BrdUdR (100 mg/kg) 30 min before harvest. Cells were stained forBrdUdR and total DNA as described in "Materials and Methods."

Table 8 Cell cycle fractions of KHT cells treated with carmethizolein vivofor 24 h"

Dose(mg/kg) %Gi %S %G2M

02001005045.428.233.037.945.548.144.044.89.123.723.017.3

Â°Mice (3/group) bearing the KHT sarcoma (200-500 mg tumors) were treated

i.p. with the indicated doses of carmethizole. After 24 h, mice were given i.p.injections of 100 mg/kg BrdUdR, and tumors were harvested 30 min later. Cellswere stained for BrdUdR and total DNA as described in "Materials and Methods."

DISCUSSION

For carmethizole to warrant development, several importantquestions needed to be addressed:(a) are the DNA adductsformed by carmethizole removed by the enzyme O6-alkylguan-ine DNA alkyltransferase? (b) are carmethizole-DNA adductsdifferent from those produced by other clinically used alkylatingagents? (c) does carmethizole lack cross-resistance with Cytoxan, BCNU, and melphalan? and (d) does carmethizole havesuperior in vivo activity when compared with the latter 3 alkylating agents?

The type of DNA adduct formed by carmethizole was investigated with mer(+)/mer(-) cell lines, repair-deficient CHO

cells, and alkaline elution techniques. The DNA damage induced by carmethizole was not repaired by the enzyme O6-

alkylguanine DNA alkyltransferase (Fig. 2; Ref. 13). Theexpression of this enzyme confers the mer(+) phenotype; cellspossessing it are capable of repairing (X'-alkylguanine residuessuch as those produced by BCNU (14-16). Many human tumors are mer(+), whereas most murine tumors are mer(â€”)(17-19). Hence, murine tumor models might overpredict for theclinical activity of some alkylating agents (20). Since carmethizole DNA damage was not influenced by the mer phenotype,murine tumor models would not be expected to overpredict for

4585




clinical activity of this agent with regard to this particular repairenzyme.

Results with CHO repair-deficient cell lines indicated thatcarmethizole most likely produces cytotoxicity through DNAalkylation forming mono-adducts and cross-links; the UV4 cellline was hypersensitive to both carmethizole and melphalan.whereas the UV5 line was sensitive only to carmethizole (Table1).

The alkaline elution studies showed that whereas DNA-protein cross-links are rapidly formed during a 3-h continuousdrug exposure, DNA-DNA cross-links formed only after extracellular drug was removed (Table 3). The reason for this isunclear unless the high concentration of carmethizole in solution somehow inhibits formation of the second adduci. Carmethizole thus appears to form a different type of DNA adducithan either BCNU or melphalan.

Two-color flow-cytometric analyses were performed on car-methizole-treated cells to determine in which phase of the cellcycle carmethizole-treated cells accumulate. Carmethizole produced an accumulation of P388 cells in G2M in vitro (Table 6).This was in contrast to the in vivo P388 experiment showing aGÃŒblock (Table 7). The in vivo result appeared to be an artifactcaused by contamination of the peritoneal tumor sample withhost cells indistinguishable from P388 cells on the basis ofDNA content. This hypothesis prompted us to repeat the invivo studies with s.c. implanted KHT sarcoma, which is aneu-ploid and can thus be distinguished from host cell populations.The G2M block in the KHT experiments confirmed the in vitroP388 result (Fig. 7; Table 8). When KHT tumors were left insitu for 24 h after BrdUrd and carmethizole treatment, cells

oONCOLU

CO

LU-z.

IOO

m

B

G2M

TOTAL DNA CONTENTFig. 7. In rivo carmethizole treatment causes KHT cells to accumulate inG2M.

Mice (3/group) bearing 200- to 500-mg tumors were given i.p. injections of 200mg/kg carmethizole. Twenty-four h later, the mice received 100 mg/kg BrdUrdi.p. Tumors were harvested 30 min later and cells stained for BrdUrd and totalDNA as described in "Materials and Methods." A, control; B. carmethizole-

treated.

zooBcoLU

Eco

LUz^h-

iooocâ€¢o

m

TOTAL DNA COMENTFig. 8. In vivo carmethizole treatment showing a partial block in Ci;M KHT

cells. Mice (3/group) bearing 200- to 500-nig tumors were given i.p. injections of200 mg/kg carmethizole. Thirty min later the mice received 100 mg/kg BrdUrd.Tumors were harvested 24 h later and cells stained for BrdUrd and total DNA asdescribed in "Materials and Methods." A. control; B, carmethizole-treated.

labeled with BrdUrd were present in all phases of the cell cycle,indicating that carmethizole produces an incomplete G2Mblock. Based on these analyses, carmethizole appeared similarto other alkylating agents that also cause an accumulation ofcells in the G2M phase of the cell cycle (21, 22).

The in vivo studies also indicated that carmethizole is similarto melphalan with respect to both spectrum and degree ofactivity. The observed lack of cross-resistance with Cytoxanand BCNU and complete cross-resistance with melphalan support the conclusion that the therapeutic and mechanistic characteristics of carmethizole are melphalan-like. Our observationthat carmethizole had no activity against murine tumors butshowed significant activity against human tumor xenograftsthat are usually considerably more refractory to therapy wasunexpected. The reason for the difference between the murineand human tumors remains unclear.

In conclusion, carmethizole is a novel anticancer agent withdemonstrated activity against human tumor xenografts. Itcauses an accumulation of cells in the G2M phase of the cellcycle as do other alkylating agents, and interacts with DNAand protein, resulting in mono-adducts and cross-links. Weconclude that this compound is "melphalan-like" due to both

its pattern of activity against drug-resistant cells and the observed in vivo activity.

REFERENCES

1. Anderson. \V. K., Bhattacharjee. P.. and Houston, D. M. Design, synthesis,antincoplastic activity, and chemical properties of bis(carbamate) derivativesof 4:5-bis(hydroxymethyl)imidazole. J. Med. Chem.. 32: 119-127. 1989.

2. Leopold. NV.R., Fry, D. W., Elliott, W. I... Hawkins, P., and Anderson, W.

4586




K. In vivo and biochemical evaluation of the novel alkylator, carmethizole.Proc. Am. Assoc. Cancer Res., 30: 1863, 1989.

3. Waud, W. R., Plowman, J., Anderson, W. K., Dykes, D. J., Harrison, Jr., S.D., and Griswold, D. P., Jr. Treatment site and schedule dependence ofcarmethizole hydrochloride (NSC-602668) in preclinical models. Proc. Am.Assoc. Cancer Res., 30: 2448, 1989.

4. Brodfuehrer, J. I., Wilke, T. J., Kinder, D. H., and Powis, G. Preclinicalpharmacologie studies of the new antitumor agent carmethizole (NSC-602668) in the mouse and beagle dog. Cancer Chemother. Pharmacol., 24:277-283, 1989.

5. Kulm. K. W., Ewig, R. A. G., Erickson, L. C., and Zwelling, L. A. Measurement of strand breaks and cross-links by alkaline elution. In: E. C. Friedbergand P. C. Hanawalt (eds.), DNA Repair. A Laboratory Manual of ResearchProcedures, pp. 379-401. New York: Marcel Dekker. 1981.

6. Zwelling, L. A., Michaels, S., Erickson, L. C., Ungerleider. R. S., Nichols,M., and Kohn, K. W. Protein-associated deoxyribonucleic acid strand breaksin L1210 cells treated with the deoxyribonucleic acid intercalating agents 4'

(9-acridinylamino)methanesulfon-m-anisidide and Adriamycin. Biochemistry, 20:6553-6563, 1981.

7. Ross, W. E., Glaubiger, D., and Kohn, K. W. Qualitative and quantitativeaspects of intercalator-induced DNA strand breaks. Biochim. Biophys. Acta,562:41-50, 1979.

8. Dolbeare, F., Gratzner, H., Pallavicini, M. G., and Gray, J. W. Flowcytometric measurement of total DNA content and incorporated bromode-oxyuridine. Proc. Nati. Acad. Sci. USA, 80: 5573-5577, 1983.

9. Schabel, F. M., Jr., Griswold, D. P., Jr., Laster, W. R., Jr., CorbeÂ«,T. H.,and Lloyd, H. H. Quantitative evaluation of anticancer agent activity inexperimental animals. Pharmacol. Ther.. /: 411-435. 1977.

10. Leopold, W. R., Nelson, J. M., Plowman, J., and Jackson, R. C. Anthrapyr-azoles, a new class of intercalating agents with high level broad spectrumactivity against murine tumors. Cancer Res., 45: 5532-5539, 1985.

11. Hoy, C. A., Salazar. E. P., and Thompson, L. H. Rapid detection of DNA-

damaging agents using repair-deficient CHO cells. MutÃ¢t.Res., 130: 321-332, 1984.

12. Anderson, W. K., Bhattacharjee, D., Houston, D. M., Plowman, J., andHarrison, S. D., Jr. Design, preparation, chemical properties, and antitumoractivity of carmethizole hydrochloride. Proc. Am. Assoc. Cancer Res., 29:1298, 1988.

13. Toshimitsu, A., and Bodell, W. J. Cellular resistance to chloroethylnitrosu-reas, nitrogen mustard, and m-diamminedichlorosoplatinum (II) in humanglial-derived cell lines. Cancer Res., 47: 1361-1366, 1987.

14. Tong, W. P., Kirk, M. C., and Ludlum. D. B. Mechanism of action of thenitrosureas V. Formation of 0-(2-fluoroethyl)guanine and its probable rolein the crosslinking of deoxyribonucleic acid. Biochem. Pharmacol., 32:2011-2015, 1983.

15. Yarosh. D. B. The role of O'-methylguanine-DNA methyltransferase in cellsurvival, mutagenesis, and carcinogenesis. MutÃ¢t.Res., 145: 1-16, 1985.

16. Brent, T. P. Inactivation of purified human O'-alkylguanine-DNA alkyltrans-ferase by alkylating agents or alkylated DNA. Cancer Res., 46: 2320-2323,1986.

17. Myrnes, B., Norstrand, K., Giercksky, K-E., Sjunneskog, C., and Krokan, H.A simplified assay for O-methylguanine-DNA methyltransferase activity andits application to human neoplastic and non-neoplastic tissues. Carcinogenesis (Lond.), 5: 1061-1064, 1984.

18. Futscher, B. W., Micetich, K. C, Barnes, D. M., Fisher, R. I., and Erickson,L. C. Inhibition of a specific DNA repair system and nitrosourea cytotoxicityin resistant human cancer cells. Cancer Communications, /: 65-73, 1989.

19. Yagi, T., Yarosh, D. B., and Day, R. S. III. Comparison of repair of O6-methylguanine produced by A'-methyl-A"-nitro-A'-nitrosoguanidine in mouseand human cells. Carcinogenesis (Lond.), 5: 593-600, 1984.

20. Carter, S. Proceedings of the seventh new drug seminar on the nitrosoureas.Cancer Treat. Rep., 60: 645-811. 1976.

21. Hill, B. T. Cancer chemotherapy. The relevance of certain concepts of cellcycle kinetics. Biochim. Biophys. Acta, 516: 389-417, 1978.

22. Meyn, R. E., and Murray, D. Cell cycle effects of alkylating agents. Pharmacol. Ther., 24(2): 147-163, 1984.

4587



1991;51:4581-4587. Cancer Res William L. Elliott, David W. Fry, Wayne K. Anderson, et al. Carmethizole

Evaluation of the Alkylating Agentin Vitro and In Vivo

Updated version

http://cancerres.aacrjournals.org/content/51/17/4581

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/51/17/4581To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



in vivo and in vitro evaluation of the alkylating agent...

Documents