[CANCER RESEARCH 44. 3856-3861, September 1984]

Enhancement of 1,3-Bis(2-chloroethyl)-1-nitrosourea-induced

Cytotoxicity and DMA Damage by a-Difluoromethylornithinein L1210 Leukemia Cells1

Paul F. Cavanaugh, Jr.,2 Zlatko P. Pavelic, and Carl W. Porter3

Department ol Experimental Therapeutics. Grace Cancer Drug Center, Roswell Park Memorial Institute, New York State Department of Health. Buffalo, New York 14263

ABSTRACT

Polyamine depletion by pretreatment with a-difluoromethylor-

nithine (DFMO), a specific and irreversible inhibitor of omithinedecarboxylase, potentiates the cytotoxicity of 1,3-bis(2-chloro-ethyl)-1 -nitrosourea (BCNU) in L1210 leukemia cells grown in a

modified soft agar system. The dose enhancement ratio was1.97 at a control colony formation level of 5%. The basis for thisenhancement was investigated at the level of DMA damage usinga modified fluorometric assay to quantitate the production ofalkaline-labile strand breaks per relative DMA molecular mass.

Pretreatment of cultured L1210 cells for 48 hr with 5 mw DFMOdepleted intracellular putrescine and spermidine (but not sperm-ine) pools and resulted in a 2.3-fold increase in BCNU-induced

(10 fig/ml, 2 hr) DNA strand breaks per relative DNA molecularmass. The inclusion of 10 MM spermidine during the DFMOpretreatment fully prevented growth inhibition and enhancementof BCNU-induced DNA damage while maintaining cellular sper

midine pools at control levels. The inclusion of 2 ^M putrescineor spermidine also prevented growth inhibition and enhancementof DNA damage while maintaining spermidine pools at only 25to 35% of control. Thus, the portion of spermidine essential forcell growth appears to be associated with DNA. BCNU itself wasfound to reduce cellular polyamine levels by causing their leakagefrom cells. In addition, BCNU was found to react directly withspermidine in a cell-free system, resulting in a major reactionproduct detectable by high-performance liquid chromatography.

While decreased interaction of BCNU with polyamines couldaccount, in part, for enhancement effects of DFMO, it is moreprobable that alterations in DNA structure secondary to polyamine depletion are responsible for these effects.

INTRODUCTION

The anticancer agent BCNU4 is thought to exert its cytotoxic

properties through alkylation of DNA followed by the formationof interstrand cross-links (10). Marton ef al. (7, 14) found that

BCNU cytotoxicity can be potentiated in 9L rat brain gliosarcomacells in vitro and in vivo by polyamine depletion secondary topretreatment with DFMO, an irreversible inhibitor of omithinedecarboxylase (9). In additional studies (13), the cytotoxicity of1-(chloroethyl)-3-irans-4-methylcyclohexyl-1 -nitrosourea and

'This investigation was supported by Research Grants CA-22153 and CA-33321 and Core Grant CA-24538 from the National Cancer Institute, Departmentof Health. Education, and Welfare.

2 Recipient of American Cancer Society Postdoctoral Fellowship Grant PF-2425.3To whom requests for reprints should be addressed.'The abbreviations used are: BCNU, 1,3-bis(2-chloroethyl)-1 -nitrosourea (NSC

409962); DFMO. a-difluoromethytomithine; HPLC, high-performance liquid chromatography; PUT, putrescine; SPD, spermidine; SPM, spermine; PBS, phosphate-buffered saline.

Received August 29, 1983; accepted June 12,1984.

chlorozotocin was also found to be potentiated by DFMO. Bycontrast, the effects of c/s-diamminedichloroplatinum(ll), whichis of a different class of DNA-reactive agents (6), were found to

be decreased by DFMO in 9L cells (26, 27).Since all of the above effects of DFMO are rapidly reversed

by subsequent treatment with 1 mw PUT, they are generallyattributed to PUT and SPD depletion, although considerationmight also be given to the possible involvement of drug-inducedalterations in the levels of S-adenosylmethionine and its metab

olites (21). Polyamines are thought to be closely associated withDNA (1, 11), and it has been suggested (7) that their depletionby DFMO might destabilize DNA, rendering it more susceptibleto alkylation. In particular, polyamine deficiency may alter thespatial configuration of DNA nucleophilic groups, favoring interaction with certain cross-linking nitrosoureas but not with cis-

diamminedichloroplatinum(ll) (17). Evidence that the basis for theeffects of DFMO on the cytotoxicity of nitrosoureas does indeedoccur at the level of DNA has been provided by studies in whichDFMO pretreatment was found to approximately double thenumber of sister chromatid exchanges induced by BCNU in 9Lcells (27) and to increase the number of BCNU-induced DNAcross-links by about 2-fold (26).

In the present study, we have examined the effect of DFMOon the cytotoxicity of BCNU in L1210 leukemia cells using amodified soft agar cloning system and on BCNU-induced DNAdamage. We have focused our study on the latter effect, examining the correlation of polyamine pool sizes with DNA damageas detected by a modification of the bisbenzamide fluorescenceenhancement analysis (9). This technique relies on the time-

dependent alkaline denaturation of cellular DNA followed bydetermination of duplex DNA to total DNA ratios using bisbenzamide, which exhibits a differential molar fluorescence whenbound to single- versus double-stranded DNA. The data indicate

that SPD depletion by DFMO leads to enhanced cytotoxicity andDNA damage by BCNU. Interestingly, the same fraction of cellularSPD essential for cell growth is also involved in effecting theenhancement of BCNU-induced DNA damage.

MATERIALS AND METHODS

Cell Culture. Murine Li210 leukemia cells were maintained ¡nlogarithmic growth as a suspension culture in RPM11640 medium containing16 mw 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid, 8 ÕTIM4-mor-pholine:propanesulfonic acid, and 10% Nu-Serum (Collaborative Re

search, Inc., Lexington, MA). Cells were grown under a humidified 5%CO2 atmosphere at 37°. Cultures were treated while in logarithmicgrowth (0.5 to 1 x 105cells/ml) with 5 ITIMDFMO ±2 or 10 /<Mpolyamine.

After 48 hr, an aliquot of cells was removed for counting and viabilitydeterminations. Cell numbers were determined by electronic particlecounting (Model ZF Coulter Counter; Coulter Electronics, Hialeah, FL)and confirmed periodically with hemocytometer measurements. Cell

3856 CANCER RESEARCH VOL. 44

on July 15, 2018. © 1984 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

viability was assessed by trypan blue dye exclusion (0.5% in unbuffered0.9% NaCI solution). The percentage of control growth was determinedas follows.

% of control growth =final treated cell no. - initial inoculum

final untreated cell no. - initial inoculum x 100

Only viable cells were considered in this calculation. Pretreated cellswere washed twice in RPMI 1640 medium, suspended in completemedium, and incubated for 2 hr in the presence or absence of varyingconcentrations of BCNU (Bristol Laboratories, Syracuse, NY). The drugwas freshly dissolved in ethanol and diluted in distilled water just priorto its addition to the cell cultures. All handling of BCNU was performedunder dimmed lighting conditions.

Cytotoxicjty Assays. L1210 cells were incubated with or without 5rnw DFMO for 48 hr. The cells were harvested, washed in completeRPM11640 medium, and then exposed to varying concentrations (0 to1 ng/m\) of BCNU for 2 hr at 37°.They were then washed twice and

suspended in complete medium.The soft agar method of Pluznik and Sachs (23) and Bradley and

Metcalf (2) was slightly modified and used to culture treated L1210 cells.Briefly, a 2.5-ml lower layer of 0.5% agar in double-strength Eagle's

minimal essential medium containing 10% fetal calf serum was placed ina 35- x 10-mm plastic Retri dish and permitted to solidify at room

temperature. L1210 cells from the various treatment groups were suspended in a plating layer (0.85 ml) of 0.3% agar in double-strengthEagle's minimal essential medium with 10% fetal calf serum. Cells

pretreated with DFMO were seeded with 5 ¿IMSPD in the plating layer.After solidifying at room temperature, the agar was inspected to ensurethere were no cell clumps. Cultures were prepared in triplicate andincubated at 37°in a humidified incubator in an atmosphere of 5% COz

and 95% air. Cultures were examined with an inverted-phase microscope

at x 100 and x200. The final number of colonies was quantitated between1 and 2 weeks after plating. Groups of 30 or more cells were identified

as colonies.The effect of DFMO on BCNU was also examined in suspension

cultures of L1210 cells. Cells pretreated in the presence or absence of5 mw DFMO for 48 hr were exposed to varying concentrations (0.1 to40 /zg/ml) of BCNU for 2 hr and then cultured in RPMI 1640 in thepresence or absence of 2 ¿IMSPD for 48 hr. Growth was determined asabove at 3 time intervals during the incubation.

Polyamine Determinations. Polyamine pools were determined onextracts from cells treated in the presence or absence of DFMO ±polyamines or from cells treated for 2 hr with varying concentrations ofBCNU. Cell samples were washed twice in cold PBS, and an aliquot of107 cells was removed for polyamine determination. The cells were

pelleted, and the PBS supernatant was carefully removed with a cottonswab. The remaining cell pellet was extracted with 0.5 ml of 0.6 Mperchloric acid for 30 min at 4°and then centrifuged for 3 min at 12,000x g using a microcentrifuge. The supernatant was frozen at -20° until

analysis by HPLC.Polyamines in a 50-^1 sample of perchloric acid extract were separated

on an HPLC system using a 2.8-mm-diameter glass microbore columnwith a 2-cm column height packed with DC-4A cationic exchange resin

(Durrum Chemical Corp., Palo Alto, CA). The column temperature wasmaintained at 65°with water from a circulating bath. The chromatogramwas developed at 500 Ib/sq in at a flow rate of 16 ml/hr. using a 3-step

sodium borate buffer system. Buffer 1 containing 0.2 M boric acid, 0.5 MNaCI, 0.03% Brij 35 (Pierce Chemical Co., Rockford, IL), and octanoicacid (pH adjusted to 6.0 with saturated KOH) was run through thecolumn for 4 min. Buffer 2 containing 0.2 M boric acid, 2.15 M NaCI,0.03% Brij 35, and 0.0001% octanoic acid (pH 6.0 as above) was runfor 6 min. Buffer 3 containing 0.2 M boric acid, 2.9 M NaCI, 0.0001%octanoic acid (pH 6.0 as above) was run for 6 min. The column wasreequilibrated for 10 min with Buffer 1 prior to loading the next sample.The column eluate was derivatized with 0.05% o-phthalaldehyde (DurrumChemical Corp.) in 0.4 M borate buffer (pH 10.4) containing 1 mw 2-

DFM0 Enhancement of BCNU Effects

mercaptoethanol and 0.09% Brij 35. The o-phthalaldehyde reagent flow

rate was 8 ml/hr. The derivatized eluate was analyzed for polyaminecontent by passing it through the flow cell of a Fluoro-Monitor (American

Instrument Co., Silver Spring, MD). Fluorescence was measured byactivation with UV passing through a primary filter (Corning No. 7-51)

and recording the emitted light through a secondary filter (Wratten No.8) in a Hewlett-Packard Model 3385A automation system for data

analysis. The system variance for a standard containing known quantitiesof PUT, SPD, and SPM hydrochloride is less than 5%. The sensitivity ofthe HPLC system is in the range of 50 pmol/50-/J sample (106 cells).

Run time was less than 20 min/sample plus 10 min for column equilibration.

Bisbenzamide Fluorescence Enhancement Assay for DNA Damage. DNA damage was quantitated using a slight modification of thefluorometric procedure of Kanter and Schwartz (9), developed to detectlow levels of radiation or drug-induced DNA damage in mammalian cells.

The conditions of alkaline unwinding used here were originally developed by Rydberg (25) to detect low levels of radiation-induced DNAdamage in the 0- to 400-rad dose range. Treated L1210 cells werewashed twice in ice-cold PBS and suspended at a final density of 1 x106 cells/ml in PBS. They were pipetted (500 jil) into 9 disposable

borosilicate (10 x 75 mm) glass culture tubes (Fisher Scientific Co.,Pittsburgh, PA) for each treatment group. The tubes were then dividedinto 3 sets: one for no DNA denaturation; one for timed (30 min) alkalineDNA unwinding; and one for complete alkaline DNA denaturation. SinceDNA, when in alkali, can be degraded by visible light (23), alkalineunwinding was performed under controlled lighting conditions in a photographic dark room using a safe light (Kodak type OC filter) in a relativelyvibration- and disturbance-free area. Stock solutions (200 ng/ml in H2O)

of bisbenzamide (Hoechst 33258; Sigma Chemical Co., St. Louis, MO)were made fresh daily. To Set A tubes, where no DNA denaturationoccurs, 500 ii\ of 0.06 N HCI were added, followed by 500 n\ of base[0.06 M NaOH:0.02 M NazHPO^LS M NaCI (pH 12.2)] and 500 n\ ofbuffer [0.16% sodium lauroyl sarcosinate:0.2 M K2HPO4:0.04 M disodiumEDTArbisbenzamide (1 ^g/ml), pH 7.4]. The DNA was then sheared bysonication for 15 sec using a Branson Sonifier No. 185, Setting 4,equipped with a microsonicator tip (Branson Sonic Power Co., Danbury,CT). Set B tubes underwent a timed unwinding period of 30 min andreceived 500 n\ of base. Alkaline denaturation was terminated by theaddition of 500 ß\of 0.06 N HCI, followed by 500 ¡Aof buffer andsonication for 15 sec. Total DNA denaturation in Set C tubes wasaccomplished by the addition of 500 ¡i\of base and sonication for 10sec, followed by the addition of 500 ¿ilof 0.06 N HCI and 500 //I of buffer.The cell lysates in Set C tubes were then resonicated for 5 sec. Thetubes were then covered with Parafilm and stored overnight at ambienttemperature in the dark. The relative fluorescence of each sample wasobtained using borosilicate glass tubes (10 x 75 mm) in an SPF 500fluorometer (American Instrument Co., Silver Spring, MD) operating inthe ratio mode (excitation, 353 nm; emission, 451 nm; band pass, 5 nm).

Calculations of DNA Damage. The duplex:total DNA ratio (F) inL1210 cells after 30 min of alkaline denaturation was calculated usingEquation A.

B - C = duplex DNAA - C ~ total DNA (A)

where A, B, and C are the relative fluorescences observed in the samplesfrom the 3 sets of tubes.

As described by Rydberg (25)

(B)

where M„is the number average molecular weight between 2 breaks, fis the time of unwinding, and ßis the unwinding constant which is lessthan one. Since different pretreatments did, in fact, alter the ratio ofduplex DNA to total DNA, DNA alkaline unwinding after BCNU treatment

SEPTEMBER 1984 3857

on July 15, 2018. © 1984 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

P. F. Cavanaugh, Jr., et al.

was only compared with control cells from the same pretreatmentpopulation.

The number of unwinding points (p) per alkaline unwinding unit ofDNA was determined from the ratio of double-stranded to total DNA (F)as shown in Equation C.

P = In F«,InFo

(C)

where F0 is the ratio obtained for control cells, and F«,is the ratioobtained for BCNU-treated cells. The number of breaks per alkalineunwinding unit (n) is equal to p - 1. Results were then converted todrug-induced strand breaks per relative DNA molecular mass (n/109

dallons).Radiolabelingof L1210 Cells. In orderto detectpossibleleakageof

polyaminesfrom cells as a consequenceof BCNUtreatment, cells werelabeled with [14C]PUTprior to exposure to BCNU. L1210 cells ¡nthe

logarithmic phase of growth were incubated in complete medium containing 0.5 nCi of [14C]PUT(New England Nuclear)per ml for 24 hr. The

cells were then washed twice with PBS and incubated with 0, 20, or 40ng of BCNU per ml for 4 hr. The cells were washed twice in completemedium, solubilized in 1 N NaoH, and neutralized with HCI. Sampleswere mixed with ACS II (Amersham, Arlington Heights, IL) scintillantsolution, and radioactivity was quantitated using a Packard scintillationcounter.

RESULTS

DFMO Effects on Cell Growth and Polyamine Pools. Treatment of cells with 5 mw DFMO for 48 hr prior to exposure toBCNU reduced PUT and SPD pools to undetectable levels, whileincreasing SPM pools slightly (Table 1). Under these conditions,growth was routinely inhibited by 40 to 50% at 48 hr. Treatmentof cells with 5 mM DFMO and 10 ^M PUT or SPD fully supportedcell growth and maintained intracellular SPD pools at controllevels. When PUT or SPD was added at 2 MMinstead of 10 UM,growth was again supported at near control levels (>90%), butSPD pools were only 25 to 35% of control. This SPD level canbe maintained even after extended incubations (>96 hr) and isregarded (data not shown) as the proportion of the normal SPDpool that is essential for cell growth. The extent to which SPDpools were maintained under the conditions of this study wassomewhat greater (25 to 35%) than the 15% reported previously(24). We attribute this difference to the use of L1210 cellsadapted to grow in medium supplemented with NU-Serum rather

than fetal calf serum. Treatment with 2 UM SPM also partiallymaintained SPD pools at 25 to 35% and, as a consequence,prevented DFMO-induced cytostasis. Whether this is a consequence of back-conversion of SPM to SPD or of feedback

inhibition of SPM synthase by SPM has not yet been determined(24). SPM was not used at 10 UM because of an inherentcytotoxicity encountered at this concentration.

TabtelEffects of pretreatment with DFMO ±polyamines on cell growth and polyamine

content of cultured L1210 leukemiacells

Pretreatment(48hr)None

5 RIMDFMO5 mw DFMO + 2 MMPUT5 mMDFMO + 10 MMPUT5 mMDFMO + 2 MMSPD5rriMDFMO + 10 MMSPD5 mMDFMO + 2 MMSPMPolyamine

content(pmol/10"cells)PUT220

<50120220<50220<50SPD2980

<5013303560106028001050SPM980

113015501320131010202190%

of controlgrowthtoo57

979396

10196

Cellular Cytotoxicity Assays. L1210 cells which were treatedfor 48 hr in the presence of 5 mw DFMO did not form coloniesin the soft agar cloning assay when plated in the absence ofadded SPD in the plating layer. By adding 5 UM SPD to theplating layer, colonies were obtained, although at a lower platingefficiency than with cells not pretreated with DFMO. DFMO-

treated cells in the presence of SPD exhibited a plating efficiencyof up to 46% of control values. The data were normalized byexpressing cytotoxicity data as a percentage of control (noBCNU) colony formation. DFMO pretreatment significantly increased the cytotoxicity of BCNU in the population of L1210cells able to grow in soft agar, giving a dose enhancement ratio(7) of 1.97 at the 5% control survival level (Chart 1). This issimilar to what has been observed in 9L cells (7).

The ability of DFMO pretreatment to enhance BCNU cytotoxicity was also examined under suspension culture conditions(Chart 2). As with cells grown in soft agar, DFMO-pretreated

cells recovered poorly, unless exogenous PUT or SPD waspresent in the medium. Exogenous SPM did not significantlyimprove recovery. At the concentrations used (2 /tw), the polyamines did not affect control cell growth. As indicated in Chart 2,pretreatment with DFMO did not increase the cytotoxicity ofBCNU at the concentrations tested.

BCNU-induced DNA Damage. A modification of the assay

described previously (9) for DNA damage using an alkali buffersystem described by Rydberg (25) gave consistent and reproducible data. Variability was minimized by performing the assayin a vibration- and disturbance-free setting with controlled incan

descent lighting conditions.Depletion of PUT and SPD by DFMO consistently resulted in

duplex DNA:total DNA ratios of 0.549 ±0.010 (S.E.) and 0.757

BCNU CONCENTRATION, /Chart 1. Effect of DFMO pretreatment on BCNU cytotoxicity ¡nL1210 cells

grown in soft agar. Cells were incubated in the presence (•)or absence (O) of 5RIMDFMO for 48 hr prior to treatment for 2 hr with varying concentrations ofBCNU. Points, mean of the percentage of control (no BCNU) colonies from 2separate experiments run in triplicate; bars, S.E.

3858 CANCER RESEARCH VOL. 44

on July 15, 2018. © 1984 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

DFMO Enhancement of BCNU Effects

o

X

E

oQ

8.0

6.0

4.0 -

2.0 -

1.0

O.8

0.6

0.4 -

0.2 - DFMO , 48hr+ BCNU , 2hr+ 2 ,iM Spd,48hr

Table 2Effect of DFMO pretreatment on duplex:total DNA ratios and BCNU-induced DNA

strand breaks per relative molecular mass in L1210 leukemia cells

+ 5mM OFMO, 48hr•»•BCNU,2hr•f2 fi M Spd,48hr

IO 24 48 IO 24 48

Time öfter BCNU Treotment (hr)

Chart2. Effects of DFMO pretreatment on BCNU cytotoxidty in L1210 cellsgrown in suspensioncultures. Cellswere incubated in the absence(left)or presence(right) of 5 mMDFMO for 48 hr prior to treatment with 0, 0.1,1,10, 20, and 40 figof BCNU per ml for 2 hr. The cells were then washed and reseeded at a density of5x10* cells/ml in complete medium containing 2 /IM SPD. Cell densities were

determined at 10, 24, and 48 hr after seeding. The inclusion of 2 MMSPD in theculture mediumhad no effect on control cell growth. Points, meanof the percentageof control growth from 4 experiments run in triplicate.

±0.038 (Table 2) for control and DFMO-treated cells, respec

tively, after 30 min in alkali. Further studies (not shown) haveindicated that the rate of unwinding for DNA in DFMO-treated

cells is almost identical to that of control cells; however, theduplex DNA:total DNA ratio in DFMO-treated cells is always

greater than in control cells at the end of the unwinding period.Addition of an amount of exogenous SPD to the assay cellsuspension in excess of the amount found in the control cellpopulation prior to adding base had no effect on the final du-plexrtotal DNA ratio. This indicates that only SPD bound intra-

cellularly has an effect on final duplex:total DNA ratios. Also, anexcess amount of SPD added to the final cell lysate had no effecton the relative fluorescence of the bisbenzamide:DNA complex.Alkaline sucrose gradient centrifugation analysis revealed thatthe molecular weight of L1210 cell DNA did not change withDFMO pretreatment (data not shown).

In the absence of DFMO pretreatment, a 2-hr treatment withvarying concentrations of BCNU (Table 3) gave a dose-depend

ent increase in strand breaks/alkaline unwinding unit of DNA(approximately 2 x 10" daltons/alkaline unwinding unit).5 For the

purpose of this study, a BCNU concentration of 10 /ig/ml wasselected. With a 2-hr treatment, it resulted in 1.50 strand breaks/109 daltons with additional DNA damage occurring at higher

BCNU concentrations. Depletion of PUT and SPD by pretreatment with 5 mM DFMO increased this frequency by about 2.3times. The effect was completely abrogated by the inclusion of10 UM PUT or 10 /IM SPD during the DFMO pretreatment, both

Pretreatment(48hr)None5

mMDFMO5

mMDFMO + 2 UMPUT5

mMDFMO + 2 ¡MSPD5mMDFMO+

10.UMSPD5

mu DFMO + 2 MMSPMBCNU,

2 hr Duplex:total DNA(lOfig/ml) (Ff0.549

±0.0106-1- 0.458±0.0110.757

±0.036+ 0.623 ±0.0100.584

±0.032-I- 0.514 ±0.0400.634

±0.013+ 0.547 ±0.0350.569

±0.001+ 0.491 ±0.0060.575

±0.009+ 0.517 ±0.005Strandbreaks/109

daltons1.503.501.201.601.500.94

" Results were calculated as described in the text. Each experiment was run at

least 3 times in triplicate." Mean ±S.E. of all assays performed.

Table 3Dose dependence of BCNU-induced DNA damage and polyamine depletion in

L1210 leukemia cells

BCNU, 2 hr(Mg/ml)(f

0.115

102040Strand

breaks/10'

daltonsND"ND

0.451.504.104.75Polyamine

content(pmol/10"cells)PUT160

130140ND160<50<50SPD2810

21502100ND

183011701290SPM1220

990920ND860690

670"Solvent control containing 0.25% ethanol.

'P.M. Kanter, personal communication.

of which maintained polyamine pools at near-control levels (Table

2). The enhancement effect was also prevented with 2 ¿ÕMPUTor SPD, even though SPD pools were maintained at 25 to 35%of control levels. It should be noted that SPM pools wereincreased to levels above control values for all of these treatments and could also play a role in the prevention effect. Additionof 2 fiM SPM not only prevented the enhancement of DNAdamage but even reduced the amount of DNA damage to belowcontrol levels (Table 2).

BCNU Interaction with Polyamines. As is apparent in Table3, cellular polyamine levels decreased in a dose-dependent fash

ion, following exposure to concentrations of BCNU up to 20 ng/ml. At higher drug concentrations, they tended to remain relatively unchanged which, for SPD and SPM, was about 40 to60% of control values. This loss was determined to be due to aleakage out of the cell as determined by examination of 14C-

polyamine pools.When cells were prelabeled with [14C]PUT for 24 hr, the total

polyamine-associated radioactivity fell by 31% when exposed toBCNU at 20 ¿ig/mlfor 4 hr and by 44% at 40 /¿g/ml.The valuesare roughly comparable to the decreases seen in specific poly-

amines following identical treatments (Table 3).It is also possible that BCNU reacts directly with polyamines

to yield a product that is not removed during extraction for HPLCanalysis. This possibility was tested by mixing BCNU directlywith radiolabeled SPD. Incubation of an acellular mixture COP-

SEPTEMBER 1984 3859

on July 15, 2018. © 1984 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

P. F. Cavanaugh, Jr., étal.

taining an 11:1 molar ratio of BCNU to [3H]SPD in aqueous

solution led to the formation of at least one major unidentifiedproduct detectable by HPLC (Chart 3), indicating that BCNU canreact with SPD under the appropriate conditions. The chemicalnature of the polyamines leaking out of the cell and whetherBCNU is capable of reacting directly with cellular polyamineswere not determined.

DISCUSSION

It was determined that, as with 9L gliosarcoma cells (see"Introduction"), polyamine depletion achieved by DFMO pretreat

ment enhances the cytotoxicity of BCNU in L1210 leukemia cells.This effect was clearly demonstrable in our soft agar system(Chart 1) but not in routine cell culture (Chart 2). The basis forthis difference is uncertain but may be related to the length ofincubation time, 1 to 2 weeks in soft agar versus 48 hr in culture.In order for cells treated with DFMO alone to grow in eithersystem, SPD had to be added to the medium. The findingconfirms the true cytostatic effect of SPD depletion on cells andreinforces the likelihood that the greatest potential of DFMO incancer chemotherapy may be in combination with cytotoxicantineoplastics.

Further investigations using a fluorometric assay for cellularDNA damage revealed that DFMO pretreatment also potentiatesthe production of BCNU-induced alkaline-labile lesions in the

DNA. Quantitatively, DFMO enhancement of both BCNU effectswas very similar. The dose enhancement factor for cytotoxicityin soft agar (1.97) compares very favorably with the increase inDNA strand breaks (2.3). That these effects were obtained at 2different concentration ranges of BCNU may reflect the relativesensitivities of the 2 assays. Although DFMO enhancement ofDNA damage might be expected from the known mechanism ofaction of alkylating agents, such as BCNU (10), polaymines areubiquitously distributed among cellular organelles, and their involvements in biochemical reactions are many and varied (8).Conceivably, therefore, a large number of possible mechanismscould be responsible for DFMO potentiation effects. The findings

(0

FRACTION NUMBERCharts. HPLC elution profile of a reaction mixture containing an 11-1 molar

ratio of BCNUtpHJSPD. The BCNU and [3H]SPD were mixed in an aqueous solutionof 0.8% ethanol (v/v) and incubated for 6 hr at 37°.The mixture was made 0 6 Min HCIC-4, placed on ice for 30 min, and analyzed for polyamines by HPLC asdescnbed in "Materials and Methods.' Fractions (2 drops/fraction) were collected

and counted for radioactivity.

described here substantiate earlier indications by sister chro-matid exchange studies (27) and alkaline elution assay (26) thatthe basis for the DFMO-mediated potentiation of BCNU cytotoxicity (7) most probably resides at the level of DNA.

DFMO enhancement of all other BCNU-induced effects hasthus far been observed in 9L rat brain gliosarcoma cells. Whileserving as a useful brain tumor model, 9L cells exhibit unusualcharacteristics related to polyamines when compared to othercell lines. For example, they are relatively impermeable to polyamines (27) and methylglyoxalbis(guanylhydrazone) (18), andthey exhibit ultrastructural changes in mitochondria followingtreatment with DFMO (18). It is important, therefore, that DFMOenhancement of BCNU-mediated effects has been confirmed ina totally different cell type such as L1210 cells. Although differentfrom L1210 cells in many respects, 9L cells appear to be comparable in sensitivity with regard to the combined effects ofDFMO and BCNU. Both cell lines are sensitive in the 0- to 1-/ig/ml range of BCNU in the soft agar cloning assay, while BCNU-induced DNA damage occurred in the 10- to 20-/tg/ml range.This similarity of DFMO effects on BCNU cytotoxicity and DNAdamage in 2 widely diverse cell lines suggests a generality in thephenomena among other cell lines.

One interesting observation that has not yet been made in the9L cell system involves the apparent relationship of the polyamines that modulate BCNU-induced DNA damage to those whichare critical for cell growth. Although DFMO treatment depletesboth PUT and SPD, interest was focused on SPD since fromprevious studies (18,22) it was demonstrated that it, rather thanPUT, is critical for cell growth. DFMO-induced cytostasis isprevented to the same extent by either 2 or 10 ^M SPD, eventhough the former maintains intracellular SPD pools at only 30%of the latter. Similarly, both concentrations fully prevent theDFMO enhancement effect of BCNU-induced DNA damage.Thus, the fraction of cellular SPD that is essential for cell growthcontains the same fraction of SPD that modulates BCNU-inducedDNA damage. It may therefore be associated with DNA.

The surplus SPD pool [i.e., that not required for cell growth orfor modulating BCNU effect (-70% of control)] may representthe unbound fraction. It was demonstrated that BCNU treatmentcauses a leakage of labeled polyamines from cells and thatincreasing concentrations of BCNU can reduce cellular pools to-40% but not beyond this value (Table 3). At the same time,

DNA damage by BCNU also fails to increase. Thus, it wouldseem that the remaining SPD, which corresponds roughly to thatrequired for cell growth (-30%), is firmly bound and unable tofreely diffuse out of the cell.

Unexpectedly, increasing intracellular SPM pools in the presence of DFMO consistently resulted in a decrease in BCNU-induced DNA damage (Table 2). We are currently investigatingthe effect of modulating intracellular SPM on BCNU-inducedcytotoxicity and DNA damage, using newly identified inhibitorsof polyamine biosynthesis (4) which deplete SPM as well as SPD.In addition, we have devised recently a means for achievingselective SPM depletion by treating cells with DFMO plus a SPDanalogue (3), which will also be used to modulate BCNU effects.We have already demonstrated (3) that 60% of the SPM pool isessential for L1210 cell growth as compared to 30% for SPD asdetermined here and elsewhere (24). Since SPM is more commonly associated with maintenance of DNA structure than isSPD (1,11), its cellular removal may have a greater impact onBCNU action.

3860 CANCER RESEARCH VOL. 44

on July 15, 2018. © 1984 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

The finding that BCNU is capable of reacting directly withpolyamines raises the possibility that this interaction could initself modulate BCNU activity. Theoretically, BCNU would beexpected to interact chemically with the primary amino groupsof polyamines (5, 16), mainly via carbamoylation. Based on ourfindings in an acellular system, an intracellular chemical interaction between the metabolite of BCNU and polyamines cannot beprecluded at this time. Although none was detected spectrofluo-rometrically in BCNU-treated cell extracts, it is possible that theproduct is cross-linked to cell macromolecules and not extract-

able for HPLC detection.From the present study and those with 9L cells (7, 14), it is

apparent that DFMO has clinical potential as a biological response modifier in combination with certain antineoplasticagents. This investigation has determined that the SPD poolrequired for cell growth must be depleted by more than 70%before a modification in the response to an agent such as BCNUis observed. Therefore, it is imperative that, in experimentalsystems for testing the in vivo potential of DFMO plus BCNU,tumor polyamine concentrations be carefully monitored followingDFMO treatment.

ACKNOWLEDGMENTS

The authors wish to thank Dr. Peter Kanter and Dr. Herbert Schwartz for theirinvaluable advice regarding the assay for DNA damage; Edwin Kelly for hisassistance in polyamine determinations; and Barbara Ganis, Nancy Reska, andSarra Polonskaya for their expert technical assistance. The authors also appreciatethe helpful comments of Dr. Laurence Marlon.

REFERENCES

1. Alten, J. C. Biochemistry of the polyamines. Cell Brachem. Funct., 1:131-140,1983.

2. Bradley, T. K., and Metcalf, D. The growth of mouse bone marrow in vitro.Aust. J. Exp. Biol. Med. Set., 44: 287-299,1966.

3. Casero, R. A., Jr., Bergeron, R. J., and Porter, C. W. Growth inhibition byspermine depletion in U 210 cells. Proc. Am. Assoc. Cancer Res., 25: 356,1984.

4. Cavanaugh, P. F., Jr., Porter, C. W., Stolowich, N. J., and Bergeron, R. J.Antiproliferative activity and biological effects of spermidine derivatives in11210 leukemia cells. Proc. Am. Assoc. Cancer Res., 25:359,1984.

5. Cheng, C. J., Fujimara, S., Grunberger, D., and Weinstein, I. B. Interaction of1-(2-chloroethyl)-3-cyclohexyl-1-nitrosourea. Cancer Res., 32. 22-27,1972.

6. Harder, H. C., and Rosenberg, B. Inhibitory effects of anti-tumor platinumcompounds on ONA, RNA, and protein synthesis in mammalian cells in vitro.Int. J. Cancer, 6: 207-216, 1970.

7. Hung, D. T., Deen, D. F., Seidenfeld, J., and Marton. L. J. Sensitizaron of 9Lrat brain gliosarcoma cells to 1,3-bis(2-chloroethyl)-1-nitrosourea by a-difluc-romethylornithine. an ornithme decarboxylase inhibitor. Cancer Res., 41:2783-

DFM0 Enhancement of BCNU Effects

2785,1981.8. Janne, J., Poso, H., and Raina, A. Polyamines in rapid growth and cancer.

Biochim. Btophys. Acta, 473: 241-293,1978.

9. Kanter, P. M., and Schwartz, H. S. A fluorescence enhancement assay forcellular DNA damage. Mol. Pharmacol.. 22:145-151,1982.

10. Kohn, K. W. Interstrand cross-linking of DNA by 1,3-bis(chloroethyl)-1 -nitro-sourea and other 1-(2-haloethyl)-1-nitrosoureas. Cancer Res., 37:1450-1454,1977.

11. Liquori, A. M., Constantion. L., Crescenzi, V., Clia, V., Giglio, E., Puliti, R.,DeSantis Davino, M., and Vitagliane, V. Complexes between DNA and polyamines: a molecular model. J. Mol. Biol., 24:113-122, 1968.

12. Mamont, P. S., Duchesne. M-C., Grove, J., and Bey, P. Antiproliferativeproperties of DL-a-difluoromethytomitfiine in cultured cells. A consequence ofirreversible inhibition of omithine decarboxylase. Biochem. Biophys. Res. Commun., 81: 58-66,1978.

13. Marton, L. J., Levin, V. A., Hervatin, S. J., Koch-Weser, J., McCann, P. P.,and Sjoerdsma, A. Potentiation of the antitumor therapeutic effects of 1,3-bis(2-chloroethyl)-1-nitrosourea by a-difluoromethylornithine, an omithine decarboxylase inhibitor. Cancer Res., 41: 4426-4431,1981.

14. Marton, L. J., Oredsson, S. M., Hung, D. T., and Deen, D. F. Effects ofpolyamine depletion on the cytotoxicity of cancer chemotherapeutic agents.Adv. Polyamine Res., 4: 33-43,1983.

15. Miktes-Robertson, F., Feuerstein, B. F., Dave, C., and Porter, C. W. Thegenerality of methylgloxal-bis(guanylhydrazone)-induced mitochondrial damage and the dependence of this effect on cell proliferation. Cancer Res., 39:1919-1926,1979.

16. Montgomery, J. A., James, R., McCateb, G. S., and Johnston, T. P. The modesof decomposition of 1.3-bis(2-chlorethyl)-1-nitrosourea (BCNU) and relatedcompounds. J. Med. Chem., 70: 668-674,1967.

17. Oredsson, S. M., Deen, D. F., and Marton. L. J. Decreased cytotoxicity of c/s-diaminedichloroplatinunXII) by a-difluoromethylomithine depletion of polyamines in 9L rat brain tumor cells in vitro. Cancer Res., 42:1296-1299,1982.

18. Oredsson, S. M., Friend, D. S., and Marton, L. J. Changes in mitochondrialstructure and function in 9L rat brain tumor cells treated In vitro with «-

difluoromethykxnithine, a polyamine biosynthesis inhibitor. Proc. Nati. Acad.Sci. USA, 80: 780-784,1983.

19. Favelle. Z. P., Slocum, H. K., Rustum, Y. M., Creaven, P. J., Karakousis, C.,and Takita, H. Colony growth in soft agar of human malanoma, sarcoma, andlung carcinoma cells disaggregated by mechanical and enzymatic methods.Cancer Res., 40: 2160-2164,1980.

20. Pegg, A. G., and McCann, P. P. Polyamine metabolism and function. Am. J.Phystol., 243: C212-C221,1982.

21. Pegg, A. E., Poso, H., Shuttleworth, K., and Bennett, R. A. Effect of inhibitionof polyamine biosynthesis on the content of decarboxylated S-adenosylme-thionine. Biochem. J., 202: 519-526, 1982.

22. Pteshkewych, A., Kramer, D. L., Kelly, E., and Porter, C. W. Independence ofdrug action on mitochondria and polyamines in Li210 leukemia cells treatedwith methylglyoxal-bis(guanylhydrazone). Cancer Res.. 40: 4533-4540,1980.

23. Pluznik, D., and Sachs, L. The cloning of normal "mast" cells in tissue culture.

J. Cell. Comp. Phystol.. 66. 319-324,1966.24. Porter, C. W, and Bergeron, R. J. Spermidine requirement for cell proliferation

in eukaryotic cells: structural specificity and quantitation. Science (Wash. DC),219:1083-1085, 1983.

25. Rydberg, B. The rate of strand separation in alkali of DNA of irradiatedmammalian cells. Radiât.Res., 6Õ:274-287,1975.

26. Tofiton, P. J., Deen, D. F., and Marton, L. J. a-Difluoromethylomithine inducedpolyamine depletion of 9L tumor cells modifies drug-induced DNA cross-linkformation. Science (Wash. DC), 222: 1132-1135,1983.

27. Tofiton, P. J., Oredsson, S. M., Deen, D. F., and Marton, L. J. Polyaminedepletion influences drug-induced chromosomal damage. Science (Wash, DC),2Õ7:1044-1046,1982.

SEPTEMBER 1984 3861

on July 15, 2018. © 1984 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1984;44:3856-3861. Cancer Res   Paul F. Cavanaugh, Jr., Zlatko P. Pavelic and Carl W. Porter  L1210 Leukemia Cells

-Difluoromethylornithine inαCytotoxicity and DNA Damage by Enhancement of 1,3-Bis(2-chloroethyl)-1-nitrosourea-induced

  Updated version

  http://cancerres.aacrjournals.org/content/44/9/3856

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

  .pubs@aacr.orgDepartment 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/44/9/3856To request permission to re-use all or part of this article, use this link

on July 15, 2018. © 1984 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from