Restoration of Sensitivity to Oxazaphosphorines by ...cancerres.aacrjournals.org/content/canres/45/4/1549.full.pdf · Restoration of Sensitivity to Oxazaphosphorines by Inhibitors

ICANCERRESEARCH45,1549-1555,AprÂ«1985]

Restoration of Sensitivity to Oxazaphosphorines by Inhibitors of AldehydeDehydrogenase Activity in Cultured Oxazaphosphorine-resistant L1210and Cross-Linking Agent-resistant P388 Cell Lines1

N. E. Sladek2 and G. J. Landkamer

Departmentof Pharmacology,University of Minnesota Medical School, Minneapolis,Minnesota55455

ABSTRACT

The sensitivity of cultured L1210 and P388 cells sensitive(L1210/0, P388/0) and resistant (L1210/OAP, P388/CLA) tooxazaphosphorines, to 4-hydroperoxycyclophosphamide, ASTAZ-7557, phosphoramide mustard, and acrolein was determinedin the absence and presence of known (disulfiram, diethyldithio-carbamate, cyanamide) or suspected [ethylphenyl(2-formyl-

ethyl)phosphinate] inhibitors of aldehyde dehydrogenase activity.The L1210/OAP cell line is resistant specifically to the oxazaphosphorines; P388/CLA cells are partially cross-resistant toother cross-linking agents. All four inhibitors of aldehyde dehy

drogenase activity potentiated the cytotoxic action of the oxazaphosphorines, 4-hydroperoxycyclophosphamide and ASTA Z-

7557, against L1210/OAP and P388/CLA cells; in the presenceof a sufficient amount of inhibitor, sensitivity was essentially fullyrestored in both cases. The inhibitors did not potentiate thecytotoxic action of the nonoxazaphosphorines, phosphoramidemustard and acrolein, against these cell lines. The cytotoxicaction of the oxazaphosphorines and nonoxazaphosphorinesagainst L1210/0 and P388/0 cells was not potentiated by any ofthe aldehyde dehydrogenase inhibitors. Inhibitors of xanthineoxidase or aldehyde oxidase activities did not potentiate thecytotoxic action of the oxazaphosphorines against L1210/OAPcells. These observations strongly suggest that (a) aldehydedehydrogenase activity is an important determinant with regardto the sensitivity of a cell population to the oxazaphosphorines;(o) L1210/0 and P388/0 cells lack the relevant aldehyde dehydrogenase activity; (c) the phenotypic basis for the resistance tooxazaphosphorines by L1210/OAP cells is aldehyde dehydrogenase activity; and (d) the major reason that P388/CLA cellsare resistant to oxazaphosphorines is aldehyde dehydrogenaseactivity.

INTRODUCTION

The studies reported herein are part of our continuing effortto identify the phenotypic basis for the relatively favorable marginof therapeutic safety exhibited by the 2-chloroethylamido-

oxazaphosphorines, e.g., cyclophosphamide.Tumor cell populations frequently become resistant to a cyto

toxic agent when repeatedly exposed to it. Identification of thephenotypic basis for this resistance often serves to identify oneof the determinants influencing the sensitivity of a given cellpopulation to the drug. Such an approach was taken in the

1Supported by USPHSGrant CA 21737.2To whom requests for reprints should be addressed, at the University of

Minnesota, Department of Pharmacology,3-260 Millard Hall, 435 DelawareStreetS.E., Minneapolis,MN 55455.

Received8/6/84; revised 12/11/84; accepted 12/27/84.

present investigation. Cultured L1210 and P388 cells sensitive(L1210/0, P388/0) and resistant (L1210/OAP,3 P388/CLA) to

oxazaphosphorines (36) were used for this purpose. The L1210/OAP cell line is resistant specifically to the oxazaphosphorines,i.e., it does not exhibit cross-resistance to other cross-linkingagents; indeed, it is collaterally sensitive4 to such compounds.

The P388/CLA cell line exhibits partial cross-resistance to othercross-linking agents.

Cyclophosphamide is a prodrug (Chart 1). It is first hydrox-ylated to 4-hydroxycyclophosphamide/aldophosphamide, whichis also a prodrug. 4-Hydroxycyclophosphamide/aldophospham-

ide can then give rise to phosphoramide mustard; the cytotoxicaction of cyclophosphamide is apparently due to the cross-linkingof DNA by this metabolite. Alternatively, 4-hydroxycyclophos-phamide/aldophosphamide can be oxidized to carboxyphos-

phamide. This metabolite is without cytotoxic activity and doesnot, under physiological conditions, give rise to cytotoxic metabolites. Oxidation to carboxyphosphamide is catalyzed by NAD-

linked aldehyde dehydrogenases (6, 7, 15, 16, 20, 21, 34, 35).Thus, differentially greater aldehyde dehydrogenase activitycould account for the relative insensitivity of some cells tocyclophosphamide, and inhibitors of this activity should confersensitivity.

The ability of Known (disulfiram, diethyldithiocarbamate, cyanamide) and suspected (EPP5) inhibitors of aldehyde dehydro

genase activity (3, 9-12, 30, 33) to potentiate the cytotoxicactivity of oxazaphosphorines (ASTA Z-75576, 4-hydroperoxy

cyclophosphamide) and nonoxazaphosphorines (phosphoramidemustard, acrolein) against cultured L1210/0, L1210/OAP, P388/0, and P388/CLA cells was tested.

MATERIALS AND METHODS

ASTA Z-7557 was supplied by Dr. P. Hilgard, Asta-Werke AG, Bielefeld, Federal Republic of Germany. 4-Hydroperoxycyclophosphamide

was supplied by Dr. A. Takamizawa, Shionogi and Co., Fukushimaku,Osaka, Japan. Phosphoramide mustard-cyclohexylamine was supplied

by Dr. H. B. Wood, Jr., Drug Development Branch, Division of CancerTreatment, National Cancer Institute, Bethesda, MD. Acrolein, cyanamide, and pyridoxal hydrochloride were purchased from the Aldrich Chem-

'L1210/OAP and P388/CLA cells were referred to as L1210/CPA and P388/CPA cells in a previous publication (36).

4The term "collateral sensitivity" is used to describe the phenomenonwhereby

a cell population,while acquiringresistanceto onedrug or group of drugs, becomesmore sensitive to another drug or group of drugs (24).

"The abbreviations used are: EPP, ethylphenyl(2-formylethyl)phosphinate;ASTA Z-7557, 2-[bis-(2-chloroethyl)amino]-4-(2-sulfoethylthio)tetrahydro-2-H-1,3,2-oxazaphosphorine2-oxkJecyclohexylaminesalt.

* ASTA Z-7557 and 4-hydroperoxycyclophosphamideare relatively stable precursors of 4-hydroxycyclophosphamide. Both precursors rapidly and spontaneously (without benefit of enzymatic involvement) give rise to the 4-hydroxyintermediateunder physiologicalconditions (Chart 1).

CANCER RESEARCH VOL. 45 APRIL 1985

1549

on May 23, 2018. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

PARENT

REVERSAL OF ACQUIRED RESISTANCE TO OXAZAPHOSPHORINES

NH-CH2

R-P=0 CH2

0-CH2

CYCLOPHOSPHAMIDE

Cl-CH2-CH2

N-

CI-CH2-CH2

OOH1

NH-CH mixedfunctionR-P

= 0 CH2 H20-i-NADP**'^

\ / ^ H200 -CH2 ^\f~

NADPH + H+02

oxidaseOH

)^4-HYDROPEROXYCYCLOPHOSPHAMIDE>^NH-CH ^/^S-CH2-CH2-SOiMH-CH

^/s\R-P=0CH2

^ 0-CH2ASTA

Z 7557

H,0;2u2 R-P = ( t,H2 HS_CH2_CH2_SO-"O -CH2 MESNA

4-HYDROXYCYCLOPHOSPHAMIDE

TRANSPORT

R-P = 0\

0-CH2-CH2-CH

ALDOPHOSPHAMIDE

PHOSPHORAMIDE MUSTARD

TOXIC

NAD + H20

aldehyaeS^jJehydrogenase

NADH-t-H*

0-CH2-CH2-C-OH

CARBOXYPHOSPHAMI DE

0li

INACTIVE

Chart 1. Metabolism of oxazaphosphorines.

ical Co., Milwaukee, Wl. Disulfiram, diethyldithiocarbamate, pargylinehydrochloride, menadione, and allopurinol were purchased from theSigma Chemical Co., St. Louis, MO. EPP was supplied by Dr. L. A.Cates, College of Pharmacy, University of Houston, Houston, TX.

All drugs, except disulfiram and menadione, were dissolved in sterile,triple-distilled water; disulfiram and menadione were dissolved in 95%

ethanol. All solutions except for those containing acrolein or 95% ethanolwere sterilized by passage through 0.22-//m Millipore filters; all wereused within 1 h of preparation and were kept at approximately 4Â°C prior

to their use. Ethanol, at the concentrations used, was not cytotoxic totumor cells.

Drug-exposure medium contained CaCI2 (100 mg/liter), KCI (200 mg/liter), MgCI2-6H2O (100 mg/liter), KH2PO< (42 mg/liter), Na2HPO4.7H2O

(453 mg/liter), and NaCI (8614 mg/liter). The pH was adjusted to a valueof 7.4. Sterilization was effected by passage through a 0.22-//m Millipore

filter.Cultured mouse L1210 and P388 cells sensitive (L1210/0, P388/0)

and resistant to oxazaphosphorines specifically (L1210/OAP) or to cross-

linking agents in general (P388/CLA) (36) were obtained from the Southern Research Institute, Birmingham, AL through the courtesy of Drs. R.F. Struck and L. J. Wilkoff. The resistant sublines were originally obtainedin vivo by means of serial treatment with cyclophosphamide (13, 27). Inour laboratory, the 4 cell lines were grown in static suspension cultureat 37Â°C. RPMI 1640 culture medium supplemented with 10% horse

serum and a humidified atmosphere of 5% C02 in air was used to growthe cells. Mean population-doubling times were approximately 9 h.

The back-extrapolation method of Alexander and Mikulski (1) was

used, essentially as described previously (36), to determine the sensitivityof cultured tumor cells to drugs. The only departure from the protocolutilized previously was that the total drug exposure period was 90 min.Inhibitors of aldehyde dehydrogenase activity and/or vehicle were present during the entire 90-min period; ASTA Z-7557, 4-hydroperoxycyclo-

phosphamide, phosphoramide mustard, acrolein, and/or vehicle werepresent during only the last 30 min of the drug exposure period.

RESULTS

The sensitivities of cultured L1210/0, L1210/OAP, P388/0,and P388/CLA tumor cells to fixed concentrations of 4-hydro-peroxycyclophosphamide, ASTA Z-7557, phosphoramide mus

tard, and acrolein in the presence of increasing concentrationsof disulfiram, diethyldithiocarbamate, cyanamide, and EPP areshown in Charts 2 to 5. Sensitivities of cultured L1210/OAP andP388/CLA cells to increasing concentrations of 4-hydroperoxy-cyclophosphamide and ASTA Z-7557 in the presence of fixedconcentrations of disulfiram, diethyldithiocarbamate, cyanamide,and EPP are shown in Charts 6 and 7 and in Table 1.


1550



j'/^-j'Â«â€¢""Â«"*.<"


o

S

Oz

o. -4

-5

LI 2 IO/O LI2IO/OAP

IO

P388/0

â€”i 1â€”

P388/CLA

IO 5 IODISULFIRAM.pM

IO

Chart 2. Sensitivity of cultured tumor cells to 4-hydroperoxycyclophosphamide,ASTA Z-7557, phosphoramide mustard, and acrolein in the presence of disulfiram.Tumor cells were incubated with vehicle or disulfiram for 60 min at 37Â°C. 4-Hydroperoxycyclophosphamide (A), ASTA Z-7557 (â€¢).phosphoramide mustard(A), acrolein (O), or vehicle (â€¢)was added, and incubation was continued at 37Â°C

for an additional 30 min. Concentrations of cytotoxic agents were 12 (L1210/0), 35(L1210/OAP), 1.5 (P388/0), and 36 (P388/CLA) MM4-hydroperoxycydophospham-ide; 17 (L1210/0), 43 (L1210/OAP), 4.5 (P388/0), and 20 (P388/CLA) MMASTA Z-7557; 90 (L1210/OAP) and 220 (P388/CLA) Â»HIphosphoramide mustard; and 5(L1210/OAP, P388/CLA) MM acrolein. Cells were harvested at the end of theincubation period and subsequently resuspended and grown in drug-free medium.Surviving fractions were determined by the back-extrapolation method describedin "Materials and Methods." Each point represents the mean of measurements on

duplicate cultures.

100 200 300 48 100 200 300DI ETHY LOI THIOCARBAMATE, J

Chart 3. Sensitivity of cultured tumor cells to 4-hydroperoxycyclophosphamide,ASTA Z-7557, phosphoramide mustard, and acrolein in the presence of diethyldithiocarbamate. Tumor cells were incubated with vehicle or diethyldithiocarbamatefor 60 min at 37Â°C. 4-Hydroperoxycydophosphamide (A), ASTA Z-7557 (â€¢),

phosphoramide mustard (A), acrolein (O), or vehicle (â€¢)was added, and incubationwas continued at 37Â°C for an additional 30 min. Concentrations of cytotoxic

agents were 12 (L1210/0), 35 (L1210/OAP), 1.5 (P388/0), and 36 (P388/CLA) MM4-hydroperoxycyclophosphamide; 17 (L1210/0): 43 (L1210/OAP), 4.5 (P388/0),and 20 (P388/CLA) MMASTA Z-7557; 90 (L1210/OAP) MMphosphoramide mustard;and 5 (L1210/OAP) MMacrolein. Cells were harvested at the end of the incubationperiod and subsequently resuspended and grown in drug-free medium. Survivingfractions were determined by the back-extrapolation method described in "Materialsand Methods." Each point represents the mean of measurements on duplicate

cultures.

Two of the four inhibitors, namely, cyanamide and EPP, werethemselves without cytotoxic action toward any of the 4 celllines (Charts 4 and 5). Disulfiram and diethyldithiocarbamate

100 200 300 100 200 300 100 200 300CYANAMIDE,>|M

100 200 300

Chart 4. Sensitivity of cultured tumor cells to 4-hydroperoxycyclophosphamide,ASTA Z-7557, phosphoramide mustard, and acrolein in the presence of cyanamide.Tumor cells were incubated with vehicle or cyanamide for 60 min at 37Â°C. 4-Hydroperoxycyclophosphamide (A), ASTA Z-7557 (â€¢),phosphoramide mustard(A), acrolein (O), or vehicle (â€¢)was added, and incubation was continued at 37Â°C

for an additional 30 min. Concentrations of cytotoxic agents were 12 (L1210/0), 35(L1210/OAP), 1.5 (P388/0), and 24 (P388/CLA) MM4-hydroperoxycyclophosphamide; 17 (L1210/0), 43 (L1210/OAP), 4.5 (P388/0), and 30 (P388/CLA) MMASTA Z-7557; 90 (L1210/OAP) and 220 (P388/CLA) MM phosphoramide mustard; and 5(L1210/OAP, P388/CLA) MM acrolein. Cells were harvested at the end of theincubation period and subsequently resuspended and grown in drug-free medium.Surviving fractions were determined by the back-extrapolation method describedin "Materials and Methods." Each point represents the mean of measurements on

duplicate cultures.

50 IOO 50 IOO 50 IOO 50 100ETHYLPHENYU2-FORMYLETHYUPHOSPHINATE,j<M

Chart 5. Sensitivity of cultured tumor cells to 4-hydroperoxycyclophosphamide,ASTA Z-7557, phosphoramide mustard, and acrolein in the presence of ethyl-phenyl(2-formylethyl)phosphinate. Tumor cells were incubated with vehicle or ethyl-phenyl(2-formylethyl)phosphinate for 60 min at 37Â°C. 4-Hydroperoxycyclophos-phamide (A), ASTA Z-7557 (â€¢).phosphoramide mustard (A), acrolein (O), or vehicle(â€¢)was added, and incubation was continued at 37Â°C for an additional 30 min.

Concentrations of cytotoxic agents were 12 (L1210/0), 35 (L1210/OAP), 1.5 (P388/0), and 24 (P388/CLA) MM 4-hydroperoxycyclophosphamide; 17 (L1210/0), 43(L1210/OAP), 4.5 (P388/0), and 30 (P388/CLA) MMASTA Z-7557; 90 (L1210/OAP)and 181 (L1210/0, P388/CLA) MM phosphoramide mustard; and 5 (L1210/OAP,P388/CLA) MMacrolein. Cells were harvested at the end of the incubation periodand subsequently resuspended and grown in drug-free medium. Surviving fractionswere determined by the back-extrapolation method described in "Materials andMethods." Each point represents the mean of measurements on duplicate cultures.

were not toxic to L1210/0 or L1210/OAP cells (Charts 2 and 3).However, both were toxic to P388/CLA cells at low concentrations but not at relatively higher concentrations (Charts 2 and 3).A single concentration of each of these agents was used when


1551




25 50 75 1004-HYDROPEROXYCYCLOPHOSPHAMIOE,

50 100 ISO 200ASTA Z 7557, >iM

Chart 6. Sensitivity of cultured L1210/OAP tumor cells to 4-hydroperoxycydc-phosphamide and ASTA Z-7557 in the presence of known or suspected inhibitorsof aldehyde dehydrogenase activity. L1210/OAP cells were incubated with vehicle(x), 3 MMdisulfiram (â€¢),3 MMdiethyldithiocarbamate (A), 30 MMcyanamide (T), or100 MMethylphenyK2-formytethyl)phosphinate (â€¢)for 60 min at 37Â°C. 4-Hydrope-

roxycyclophosphamide or ASTA Z-7557 was added, and incubation was continuedat 37Â°C for an additional 30 min. Cells were harvested at the end of the incubationperiod and subsequently resuspended and grown in drug-free medium. Survivingfractions were determined by the back-extrapolation method described in "Materialsand Methods." Each point represents the mean of measurements on duplicate

cultures. The response of cultured L1210/0 tumor cells to 4-hydroperoxycyclo-phosphamide and ASTA Z-7557 is included for comparative purposes (O).

4-HYDROPEROXYCYCUOPHOSPHAMIDE, >iM ASTA Z 7557.

Chart 7. Sensitivity of cultured P388/CLA tumor cells to 4-hydroperoxycydo-phosphamide and ASTA Z-7557 in the presence of known or suspected inhibitorsof aldehyde dehydrogenase activity. P388/CLA cells were incubated with vehicle(x), 3 MMdisulfiram (â€¢).10 MMdiethyldithiocarbamate (A), 30 Â»Mcyanamide (T), or100 MMethylphenyK2-formytethyl)phosphinate (â€¢)for 60 min at 37Â°C. 4-Hydrope-roxycyclophosphamide or ASTA Z-7557 was added, and incubation was continuedat 37Â°C for an additional 30 min. Cells were harvested at the end of the incubationperiod and subsequently resuspended and grown in drug-free medium. Survivingfractions were determined by the back-extrapolation method described in "Materialsand Methods." Each point represents the mean of measurements on duplicate

cultures. The response of cultured P388/0 tumor cells to 4-hydroperoxycyclophos-phamide and ASTA Z-7557 is included for comparative purposes (O).

Table 1Sensitivity of cultured tumor cells to 4-hydroperoxycyclophosphamide and ASTA

Z-7557 in the presence of known or suspected inhibitors of aldehydedehydrogenase activity

LCn values* were obtained from the data shown in Charts 6 and 7.

LCÂ»(MM)

CellsL1210/OAPInhibitorNone

Disulfiram, 3 MMDiethyldithiocarbamate, 3 MMCyanamide, 30 MMEPP, 100 MMcydophosphamide

Z-755763

16419 2914 3625 36

8 11

13

559

12156

L1210/0 None 12

P388/CLA None 32Disulfiram, 3 MM 9Diethyldithiocarbamate, 10 MM 9Cyanamide, 30 MM 19EPP, 100 MM 7

P388/0 None 4 6" LCÂ«,99% lethal concentration (concentration of drug required to effect a 99%

cell kill).

P388/0 cells were tested but, presumably, a similar biphasicresponse would have been obtained had more concentrationsbeen examined. The biological basis for the biphasic responseis not apparent and was not investigated further. A maximum of10 MMdisulfiram was used in these experiments because greaterconcentrations of this agent were highly toxic to all 4 cell lines.

The 4 inhibitors of aldehyde dehydrogenase activity effectedlittle or no potentiation when the cytotoxic action of 4-hydroperoxycyclophosphamide or ASTA Z-7557 against L1210/0 or

P388/0 cells was determined (Charts 2 to 5). In contrast, all 4inhibitors greatly potentiated the cytotoxic action of these agentsagainst L1210/OAP and P388/CLA cells (Charts 2 to 7; Table1). Serving to illustrate the point are the facts that, when L1210/OAP cells were exposed to 35 UM 4-hydroperoxycyclophos

phamide, the log surviving fraction decreased from approximately -1 to approximately -5 if 3 MMdisulfiram (Chart 2), 3 MM

diethyldithiocarbamate (Chart 3), 100 U.Mcyanamide (Chart 4),or 40 MM EPP (Chart 5) were included in the drug exposuremedium, and that the concentration of ASTA Z-7557 required toeffect a 99% cell-kill of L1210/OAP cells was reduced from 164

MM to 29, 36, 36, and 11 MM when 3 MM disulfiram, 3 MMdiethyldithiocarbamate, 30 MM cyanamide, or 100 MM EPP, respectively, were included in the drug exposure medium (Chart 6and Table 1). Inclusion of a sufficient amount of inhibitor in thedrug exposure medium, i.e., one that gave maximum potentiation(Charts 2 to 5), restored virtually full sensitivity to 4-hydroperoxycyclophosphamide and ASTA Z-7557 in the resistant sublines;

i.e., the sensitivity of the resistant sublines L1210/OAP andP388/CLA to 4-hydroperoxycyclophosphamide and ASTA Z-

7557 in the presence of inhibitors of aldehyde dehydrogenaseactivity was comparable to that exhibited by the respectiveparent lines U 210/0 and P388/0 to these agents in the absence(or presence) of these inhibitors (Charts 6 and 7; Table 1).

None of the inhibitors potentiated the cytotoxic action ofacrolein against L1210/OAP or P388/CLA cells (Charts 2 to 5).Disulfiram, diethyldithiocarbamate, and cyanamide did not potentiate the cytotoxic action of phosphoramide mustard againstthese cell lines (Charts 2 to 4). These results demonstrate the


1552




specificity of the potentiation, i.e., that the cytotoxic action ofonly the oxazaphosphorines is potentiated. This is as expected,given that the current understanding of oxazaphosphorine metabolism and pharmacodynamics is correct and that disulfiram,diethyldithiocarbamate, and cyanamide effect their potentiatingaction by inhibiting aldehyde dehydrogenase activity. A smallamount of potentiation was observed when L1210/0, L1210/OAP, or P388/CLA cells were exposed to phosphoramide mustard in the presence of EPP (Chart 5). This may mean that, inaddition to inhibiting aldehyde dehydrogenase activity, thus potentiating the cytotoxic action of oxazaphosphorines specifically,EPP also exerts an action which leads to an increase in sensitivityto all nitrogen mustards.

Pargyline (100 ^M), yet another inhibitor of NAD-linked aldehyde dehydrogenase activity (32), did not potentiate the antitu-mor activity of 4-hydroperoxycyclophosphamide or ASTA Z-

7557 against L1210/OAP cells (data not presented). Inhibition ofNAD-linked aldehyde dehydrogenase activity by pargyline hasbeen shown to be effected by a metabolite of this agent, namely,propiolaldehyde (32). Apparently, L1210/OAP cells lack the enzyme required for this conversion.

Allopurinol (100 UM), a known inhibitor of xanthine oxidaseactivity (17), and pyridoxal (100 MM),believed to be a physiological substrate for aldehyde oxidase (38) and therefore expectedto inhibit the oxidation of other aldehydes by this enzyme, alsofailed to potentiate the antitumor activity of 4-hydroperoxycyclophosphamide or ASTA Z-7557 against L1210/OAP cells (data

not presented). Menadione (3 >Â¿M),yet another inhibitor of aldehyde oxidase activity (25), slightly potentiated the antitumoractivity of 4-hydroperoxycyclophosphamide and ASTA Z-7557

against L1210/OAP cells, but it effected a potentiation of similarmagnitude when phosphoramide mustard was used as the an-

titumor agent (data not presented), indicating that the basis forthe potentiation was independent of its ability to inhibit aldehydeoxidase activity. Moreover, it effected a slight potentiation ofoxazaphosphorine activity against L1210/0 cells as well (datanot presented).

DISCUSSION

One suspected and 3 known inhibitors of aldehyde dehydrogenase activity potentiated the cytotoxic action of the oxazaphosphorines, 4-hydroperoxycyclophosphamide and ASTA Z-7557, against oxazaphosphorine-resistant L1210/OAP cells andcross-linking agent-resistant P388/CLA cells; they did not poten

tiate the cytotoxic action of the nonoxazaphosphorines, phosphoramide mustard and acrolein, against these cells, nor didthey potentiate the cytotoxic action of the oxazaphosphorinesor nonoxazaphosphorines against the corresponding sensitiveL1210/0 and P388/0 parent cells. Potentiation was substantial;a 4- to 5-log increase in cell kill could be achieved.

Our findings that the inclusion of disulfiram in the drug-expo

sure medium restores the sensitivity of L1210/OAP and P388/CLA cells to the oxazaphosphorines confirm and extend anearlier report by Hilton and Cohen (21) on the same subject.These investigators also found that the inclusion of disulfiram inthe treatment regimen restored the sensitivity of their cyclophos-phamide-resistant L1210 cells to the oxazaphosphorines but hadlittle effect on the sensitivity of cyclophosphamide-sensitive

L1210 cells to these agents.

Potentiation by diethyldithiocarbamate, cyanamide, and EPPhas not been reported previously. That diethyldithiocarbamate,a metabolite of disulfiram, did so is not surprising, since it isknown to inhibit aldehyde dehydrogenase activity in vivo, eventhough it is a very poor inhibitor of these enzymes in cell-free

systems (9, 15). It may be that the actual inhibitor is disulfiramand that diethyldithiocarbamate is oxidized to disulfiram in vivo(9).

Cyanamide is also an inhibitor of aldehyde dehydrogenaseactivity in vivo and without inhibitory action in cell-free systems

(10,15,16). This is because inhibition of aldehyde dehydrogenase activity by cyanamide is dependent on its conversion to anactive form; activation is known to be catalyzed by intact rat livermitochondria (11,12,33). Our findings indicate that mitochondriaof L1210/OAP and P388/CLA cells are also capable of activatingcyanamide. Whether all cells are capable of activating cyanamideis not known. Given that the basis for the relatively favorablemargin of safety exhibited by the oxazaphosphorines is due todifferentially greater aldehyde dehydrogenase activity in criticalnormal cells ordinarily sensitive to nitrogen mustard analogues(see below), and that some neoplastia cells are relatively insensitive to oxazaphosphorines because of a large aldehyde dehydrogenase content, it would, from the viewpoint of therapeuticstrategy, be fortuitous indeed if only the latter were able toactivate cyanamide.

EPP is a phosphorylated aldehyde structurally resemblingaldophosphamide. It is a demonstrated competitive inhibitor ofaldehyde oxidase activity (3). Increases in the life span of L1210-

implanted mice obtained when EPP was concurrently administered with cyclophosphamide were greater than those obtainedwhen cyclophosphamide alone was given (3). EPP had no anti-

tumor activity itself. Preliminary experiments in our laboratoryindicate that, as expected in view of its chemical structure, EPPmay also be a substrate for NAD-linked aldehyde dehydrogen-ases.7 In addition to inhibiting aldehyde dehydrogenase activity

and thus potentiating the cytotoxic action of the oxazaphosphorines specifically, EPP may be acting at another site topotentiate the cytotoxic action of all nitrogen mustards in eachof the sensitive and resistant cell lines used in this study. This issuggested by the observation that EPP slightly potentiated thecytotoxic action of phosphoramide mustard in both sensitive andresistant cell lines and would explain why the combination of 4-hydroperoxycyclophosphamide or ASTA Z-7557 and EPP was

slightly more toxic to L1210/OAP cells than were the oxazaphosphorines alone to L1210/0 cells. It would also explain whyvirtually full sensitivity to the oxazaphosphorines was restoredin P388/CLA cells when EPP was included. P388/CLA cells arenot only resistant to oxazaphosphorines but they are also partially cross-resistant to nonoxazaphosphorine cross-linking

agents (36). We have postulated that at least 2 changes inphenotype are operative, one effecting a decrease in sensitivityto all cross-linking agents and a second additionally effecting adecrease only to the oxazaphosphorines. Given the correctnessof this postulate, full restoration of sensitivity in P388/CLA cellswould not be expected if the sole action of EPP was to effectthe latter by inhibiting aldehyde dehydrogenase activity.

Our observations and those of Hilton and Cohen (21) stronglysuggest that (a) aldehyde dehydrogenase activity is an important

7N. E. Sladek, unpublishedobservations.


1553



REVERSAU OF ACQUIRED RESISTANCE TO OXAZAPHOSPHORINES

determinant with regard to the sensitivity of normal and neoplas-

tic cell populations to the oxazaphosphorines; (b) L1210/0 andP388/0 cells lack the relevant aldehyde dehydrogenase activity;(c) the phenotypic basis for the resistance to oxazaphosphorinesby L1210/OAP cells is aldehyde dehydrogenase activity; and (d)the major reason that P388/CLA cells are resistant to oxazaphosphorines is aldehyde dehydrogenase activity.

Several additional observations can be offered in support ofthese conclusions. Hilton and Cohen (21) found that extractsfrom cyclophosphamide-resistant L1210 cells contained much

more aldehyde dehydrogenase activity than did extracts fromcyclophosphamide-sensitive L1210 cells and that extracts from

cells pretreated with disulfiram were unable to catalyze theformation of carboxyphosphamide from 4-hydroxycyclophos-

phamide. Recently, Hilton and Colvin (22) reported an approximately inverse correlation between aldehyde dehydrogenaseactivity and the sensitivity of several rodent tumor cell lines, 6human leukemia cell lines, and leukemic cells derived from marrow aspirates of several patients, to activated cyclophospha-

mide; disulfiram potentiated the cytotoxic action of activatedcyclophosphamide against those cells with relatively highamounts of aldehyde dehydrogenase activity. Initial results ofongoing experiments in our laboratory, in which aldehyde dehydrogenase activity (substrates were 4-hydroxycyclophospham-

ide, acetaldehyde, butyraldehyde, and propionaldehyde) wasmeasured in extracts of L1210/0, L1210/OAP, P388/0, andP388/CLA cells, indicate that the sensitive cells lack the aldehydedehydrogenase that is present in the 105,000 x g soluble fractionof resistant cells.8

The presence of aldehyde dehydrogenase activity may explainwhy plunpotent hematopoietic stem cells are relatively insensitiveto the oxazaphosphorines (26); differentiation of these cells tocommitted cells appears to be accompanied by a decrease in oreven total loss of aldehyde dehydrogenase activity (22,26). Thisobservation is particularly supportive of the idea, first proposeda decade ago (4-7, 14, 35), that the relatively favorable margin

of safety exhibited by the oxazaphosphorines, particularly cyclophosphamide, when used as immunosuppressive agents and inthe treatment of certain neoplasms, may well have as its basisaldehyde dehydrogenase activity. Indirect evidence is providedby the observation of Cox et al. (7), who reported that disulfiramessentially did not potentiate the therapeutic action of cyclophosphamide against the ADJ/PC6 mouse plasma cell tumor butmarkedly potentiated its lethal action against the host mice; suchresponses would, according to our hypothesis, be expected ifcritical normal cells contained aldehyde dehydrogenase activityand the neoplastic cells did not. This is not to say that otheroxazaphosphorine-specific determinants, e.g., bioactivation (2,

23, 28, 29, 37, 39), cannot also be operative, although suchdeterminants have yet to be conclusively demonstrated as beingso. Nonspecific determinants such as DNA repair and sulfhydrylcontent, i.e., those that would affect the cytotoxic action of allnitrogen mustards, would of course also be operative.

A number of aldehyde dehydrogenase isozymes have beenidentified in various subcellular fractions of normal mammaliancells (31). Which of these catalyze the oxidation of aldophos-phamide to carboxyphosphamide is not known. Also not knownis the genotypic basis of acquired resistance to oxazaphosphor-

*C. L. Manthey and N. E. Sladek, unpublished observations.

ines resulting from increased aldehyde dehydrogenase activity.It could be due to increased steady-state levels of the enzyme,

to the production of an isozyme not normally present, or theselection of a subpopulation of cells with relatively greater aldehyde dehydrogenase activity.

Acquired resistance to cyclophosphamide is often observed inhumans. Whether it is occasionally/inevitably due to increasedaldehyde dehydrogenase activity is not known. Chemosensitivitytesting of the type described by Daniels et al. (8) could answerthe question. In such experiments, the sensitivity of neoplasticbiopsy material, taken prior to the initiation of cyclophosphamidetherapy and again following relapse, to oxazaphosphorines inthe presence and absence of aldehyde dehydrogenase inhibitors,would be quantified. The relevant aldehyde dehydrogenase activity could also be quantified directly.

Even though acquired resistance to the oxazaphosphorinesmight be prevented or reversed, inclusion of an aldehyde dehydrogenase inhibitor in the treatment regimen would not appearto offer any therapeutic advantage if aldehyde dehydrogenaseactivity is the basis for the margin of safety exhibited by theoxazaphosphorines; the same result would be achieved by simply increasing the dose of the oxazaphosphorine. The observations that disulfiram potentiates cyclophosphamide-induced leu-

kopenia (19) and host lethality (7,18) without (7), or only slightly(19) increasing the antitumor action of cyclophosphamide lendscredence to this conclusion.

Along the same vein, a cursory review of the medical literaturereveals that more than 30 drugs in current clinical use produceadisulfiram-likeeffect, i.e., nausea, vasodilation, pulsating head

ache, etc., when taken prior to the ingestion of alcohol. Disulfiramis believed to induce these symptoms by virtue of its ability toinhibit aldehyde dehydrogenases, thus preventing the oxidationof acetaldehyde, generated from ethanol, to acetic acid. Presumably, some or all of these drugs, many of which are used incancer patients, also induce the disulfiram-like effect by inhibiting

aldehyde dehydrogenases. Whether they inhibit the isozyme(s)catalyzing the oxidation of aldophosphamide to carboxyphosphamide is not known, nor is it known whether they inhibit therelevant enzyme activity in all tissues. However, in the light ofour findings and the foregoing discussion, the potential for clinically relevant drug interactions is obvious.

Several drugs, e.g., phÃ©nobarbital, are known to induce hepatic aldehyde dehydrogenase activity in rodents. It is not knownif they induce this activity in humans, if they induce the activityof the relevant isozyme(s), or if they induce aldehyde dehydrogenase activity in all tissues. Again, the potential for clinicallyrelevant drug interactions is obvious.

REFERENCES

1. Alexander, P., and Mikulski, Z. B. Differences in the response of leukaemiacells in tissue culture to nitrogen mustard and to dimethyl myleran. Biochem.Pharmacol.,5. 275-282.1961.

2. Bielicki, L., Voelcker, G., and Hohorst, H-J. Enzymatic toxicogenation of"activated" cyclophosphamide by 3',5'-exonucleases. J. Cancer Res. Clin.Oncol., 705: 27-29,1983.

3. Cates, L. A., Jones, G. S., Jr., Good. D. J., Tsai, H. Y-L. Ã¼,V-S., CarÃ³n,N.,Tu, S-C., and Kimball, A. P. Cyclophosphamide potentiation and aldehydeoxidase inhibition of phosphorylated aldehydes and acetÃ¡is.J. Med. Chem.,23:300-304,1980.

4. Connors, T. A., Cox, P. J., Farmer,P. B., Foster, A. B., and Jarman, M. Somestudies of the active intermediates formed in the microsomal metabolism ofcyclophosphamideand isophosphamide.Biochem. Pharmacol.,23:115-129,1974.


1554




5. Cox, P. J., Farmer,P. B., and Jarman, M. The microsomalmetabolismof someanaloguesof cyclophosphamide:4-methylcyclophosphamideand 6-methylcy-clophosphamide. Biochem. Pharmacol.,24: 599-606, 1975.

6. Cox, P. J., Phillips,B. J., and Thomas, P. The enzymatic basis of the selectiveaction of cyclophosphamide.Cancer Res., 35: 3755-3761,1975.

7. Cox, P. J., Phillips, B. J., and Thomas, P. Studies on the selective action ofcyclophosphamide (NSC-26271): inactivation of the hydroxylated metaboliteby tissue-solubleenzymes. Cancer Treat. Rep., 60: 321-326,1976.

8. Daniels, J. R., Daniels, A. M., Luck, E. E., Whitman, B., Casagrande, J. T.,and Skinner, D. G. Chemosensitivity of human neoplasms with in vitro doneformation. Cancer Chemother. Pharmacol.,6: 245-251,1981.

9. Deitrich, R. A., and Erwin, V. G. Mechanism of the inhibition of aldehydedehydrogenase in vivo by disulfiram and diethyldithiocarbamate. Mol. Pharmacol., 7: 301-307,1971.

10. Deitrich, R. A., Troxell, P. A., Worth, W. S., and Erwin, V. G. Inhibition ofaldehyde dehydrogenase in brain and liver by cyanamide. Biochem. Pharmacol., 25: 2733-2737,1976.

11. DeMaster, E. G., Kaplan, E., Shirota, F. N., and Nagasawa, H. T. Metabolieactivation of cyanamide by liver mitochondria, a requirement for the inhibitionof aldehydedehydrogenaseenzymes. Biochem.Biophys. Res.Commun., 107:1333-1339,1982.

12. DeMaster, E. G., Nagasawa, H. T., and Shirota, F. N. Metabolieactivation ofcyanamide to an inhibitor of aldehyde dehydrogenase in vitro. Pharmacol.Biochem. Behav., 78 (Suppl. iy 273-277,1983.

13. DeWys, W. D. A dose-response study of resistance of leukemia L1210 tocyclophosphamide.J. Nati. Cancer Inst., 50: 783-789. 1973.

14. Domeyer, B. E., and Sladek, N. E. Potential for in vivo inactivation of acyclophosphamide metabolite (aldophosphamide) by tumor enzymes. Fed.Proc., 33. 581,1974.

15. Domeyer,B. E., and Sladek,N. E. Metabolismof 4-hydroxycyclophosphamide/aldophosphamidem vitro. Biochem. Pharmacol.,29: 2903-2912,1980.

16. Domeyer, B. E., and Sladek, N. E. Inhibition by cyanamideof 4-hydroxycyclo-phosphamide/aldophosphamideoxidation to carboxyphosphamide.Biochem.Pharmacol.,30: 2065-2073,1981.

17. Elion, G. B., Kovensky, A., Hitchings, G. H., Metz, E., and Randes, R. W.Metabolie studies of allopurinol, an inhibitor of xanthine oxidase. Biochem.Pharmacol., 75: 863-880, 1966.

18. Habs, H., Habs, M., and SchmÃ¤hl,D. Influenceof pretreatment with disulfiramon the acute toxicity of cyclophosphamide in male Sprague-Dawley rats.Arzneim.-Forsch./Drug Res., 37: 530-531,1981.

19. Hacker, M. P., Ershler, W. B., Newman, R. A., and Camelli, R. L. Effect ofdisulfiram (tetraethylthiuramdisulfide)and diethyldithiocarbamateon the bladder toxicity and antitumor activity of cyclophosphamidein mice. Cancer Res.,42:4490-4494,1982.

20. Hill, D. L., Laster, W. R., Jr., and Struck, R. F. Enzymatic metabolism ofcyclophosphamide and nicotine and production of a toxic cyclophosphamidemetabolite. Cancer Res., 32: 658-665,1972.

21. Hilton, J., and Cohen, D. Role of aldehyde dehydrogenase in the response ofL1210 cells to cyclophosphamideand 4-methyl cyclophosphamide.Proc. Am.Assoc. Cancer Res., 23:169,1982.

22. Hilton, J., and Cotvin, M. The role of aldehyde dehydrogenase activity in

cyclophosphamidesensitivity of hematopoietic and leukemic cell populations.Proc. Am. Assoc. Cancer Res., 25: 339, 1984.

23. Hohorst, H^J.,Bielicki,L., and Voelcker,G. Releaseof phosphoramidemustardfrom activated cyclophosphamideby DNA-polymeraselinked 3',5'-exonucle-

ase as basis for cytotoxic specificity. In: Proceedingsof the 13th InternationalCancerCongress, p. 384,1982.

24. Hutchison, D. J., and Schmid, F. A. Cross-resistanceand collateral sensitivity.In: E. Mihich (ed.), Drug Resistance and Selectivity, pp. 73-126. New York:Academic Press, Inc., 1973.

25. Johns, D. J. Human liver aldehyde oxidase: differential inhibition of oxidationof charged and unchargedsubstrates. J. Clin. Invest., 46: 1492-1505,1967.

26. Kohn, F. R., and Sladek, N. E. Aldehyde dehydrogenaseactivity as the basisfor the differentialsensitivityof murinepluripotentandprogenitorhematopoieticcells to oxazaphosphorinesin vitro. Proc. Am. Assoc. Cancer Res., 25: 689,1984.

27. Lane, M., and Yancey, S. T. Developmentof a leukaemia resistant to cyclophosphamide("Cytoxan"). Nature (Lond.), 788: 756-757,1960.

28. Low, J. E., Borch, R. F., and Sladek, N. E. Conversionof 4-hydroperoxycyclo-phosphamideto phosphoramidemustard andacroleinmediatedby bifunctionalcatalysts. Cancer Res., 42: 830-837,1982.

29. Low, J. E., Borch, R. F., and Sladek, N. E. Further studies on the conversionof 4-hydroxyoxazaphosphorinesto reactivemustards and acroleinin inorganicbuffers. Cancer Res., 43: 5815-5820,1983.

30. Marchner, H., and Tottmar, 0. Studies In vitro on the inactivationof mitochon-drial rat-liver aldehyde dehydrogenaseby the alcohol-sensitizingcompoundscyanamide, 1-aminocyclopropianoland disulfiram. Biochem. Pharmacol.,32:2181-2188,1983.

31. Pietruszko, R. Aldehyde dehydrogenase isozymes. In: M. C. Rattazzi, J. G.Scandalios,and G. S. Whitt (eds.), Isozymes:Current Topics in BiologicalandMedical Research. Cellular Localization, Metabolism, and Physiology, Vol. 8,pp. 195-217. New York: Alan R. Uss, Inc., 1983.

32. Shirota, F. N., DeMaster,E. G., Elberling,J. A., and Nagasawa,H. T. Metabolicdepropargylationand its relationship to aldehyde dehydrogenase inhibition invivo. J. Med. Chem., 23: 669-673, 1980.

33. Shirota, F. N., DeMaster, E. G., and Nagasawa,H. T. Studies on the cyanam-Â¡de-ethanolinteraction. Dimethylcyanamideas an inhibitor of aldehyde dehydrogenasein vivo. Biochem. Pharmacol.,37: 1999-2004, 1982.

34. Sladek, N. E. Evidence for an aldehyde possessing alkylating activity as theprimary metabolite of cyclophosphamide.Cancer Res., 33: 651-658,1973.

35. Sladek, N. E. Bioassay and relative cytotoxic potency of cyclophosphamidemetabolitesgeneratedin vitro and in vivo. CancerRes., 33:1150-1158,1973.

36. Sladek, N. E., Low, J. E., and Landkamer,G. J. Collateralsensitivity to cross-linking agents exhibited by cultured L1210 cells resistant to oxazaphosphorines. Cancer Res., 45: 625-629,1985.

37. Sladek, N. E., Powers, J. F., and Grage, G. M. Half-lifeof oxazaphosphorinesin biologicalfluids. Drug Metab. Dispos., 72: 553-559,1984.

38. Stanulovic, M., and Chaykin, S. Aldehyde oxidase: catalysis of the oxidationof AT-methylnicotinamideand pyridoxal. Arch. Biochem. Biophys., 745: 27-34, 1971.

39. Voelcker, G., Bielicki, L., and Hohorst, H-J. Evidencefor enzymatic toxificationof activated cyclophosphamide(4-hydroxycyclophosphamide).J. Cancer Res.Clin. Oncol., 99: A58-A59,1981.


1555



1985;45:1549-1555. Cancer Res N. E. Sladek and G. J. Landkamer Agent-resistant P388 Cell LinesOxazaphosphorine-resistant L1210 and Cross-LinkingAldehyde Dehydrogenase Activity in Cultured Restoration of Sensitivity to Oxazaphosphorines by Inhibitors of

Updated version

http://cancerres.aacrjournals.org/content/45/4/1549

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/45/4/1549To request permission to re-use all or part of this article, use this link



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

mailto:[email protected]



Documents

Restoration of Sensitivity to Oxazaphosphorines by ...cancerres.aacrjournals.org/content/canres/45/4/1549.full.pdf · Restoration of Sensitivity to Oxazaphosphorines by Inhibitors