[CANCER RESEARCH 41, 894-904, March 1981 ]0008-5472/81 /0041-OOOOS02.00

Mechanism of Resistance of Variants of the Lewis Lung Carcinoma to/V-(Phosphonacetyl)-L-aspartic Acid1

Thomas W. Kensler, George Mutter, James G. Hankerson, Linda J. Reck, Christine Harley, Nyun Han, BachArdalan, Richard L. Cysyk, Randall K. Johnson, Hiremagalur N. Jayaram, and David A. Cooney2

Laboratories of Toxicology ¡T.W. K.. G. M.. J. G H.. L. J. R.. C. H.. N. H., B. A., H. N. J., D. A. C.J and Chemical Pharmacology [R. L. C.¡,DevelopmentalTherapeutics Program, Division of Cancer Treatment, National Cancer Institute, Bethesda, Maryland 20205, and Arthur D. Little, Inc., Cambridge. Massachusetts02140IR. K J.I

ABSTRACT

Variants of the Lewis lung carcinoma were selected forresistance to N-(phosphonacetyl)-L-aspartic acid (PALA) bytreatment of tumor-bearing mice with repetitive subcurativedoses of PALA. The specific activity of the target enzyme, L-

aspartic acid transcarbamylase (ATCase), was measured in thefour variants developed. Three had markedly elevated ATCaseactivities; however, the fourth line, LL/PALA-C, had an ATCaseactivity identical to that of the parent, PALA-sensitive line (LL/0). One high-ATCase variant, LL/PALA-J, and LL/PALA-C

were compared with LL/0 in subsequent biochemical studieson the mechanism of resistance to PALA. Enzyme activities inthe salvage pathways which phosphorylate pyrimidine nucleo-

sides and deoxynucleosides were found to be similar in allthree lines. ATCase in these lines exhibits closely comparablekinetics with its natural substrates as well as with PALA. Thetime courses of restitution of ATCase after a single therapeuticdose of PALA show that both resistant variants recover fullactivity more rapidly than the parent. Additionally, inhibition ofATCase 24 hr following graded doses of PALA is lower in theresistant lines. The uptake of [14C]PALA in vitro into cell lines

derived from the three Lewis lung carcinomas apparently occurs by passive diffusion and at comparable rates in bothsensitive and resistant cells. Analysis of the nucleotide contentof tumors reveals comparable spectrums of purine and pyrimidine nucleotide levels in the LL/0 and LL/PALA-C lines,whereas the LL/PALA-J line has augmented nucleotide pools.

In all three lines, 24 hr after treatment with PALA (400 mg/kg),uridine and cytidine nucleotide levels were substantially diminished (70 to 80%) while adenosine 5'-triphosphate and gua-nosine 5'-triphosphate levels were elevated (50 to 100%).

Estimations of precursor flux through the de novo pyrimidinepathway by measuring orotate and orotidine levels in tumors ofmice treated with pyrazofurin (an inhibitor of orotidine-5'-mono-

phosphate decarboxylase) and either 0.9% NaCI solution orPALA shows that PALA treatment eliminates orotate and orotidine accumulation in LL/0 but reduces it by only 75 and 50%in LL/PALA-C and LL/PALA-J, respectively. Similarly, PALA

treatment (20 /IM) of tumor lines in culture provokes a dramaticdecrease in the incorporation of NaH14CO3 into pyrimidine

intermediates and nucleotides in the LL/0 cell line only. Determinations of specific activities of the other enzymes in thispathway reveal that the activity of carbamyl phosphate synthe-

1A preliminary account of this work was presented at the 71 st Annual Meeting

of the American Association for Cancer Research in San Diego, Calif. (18).2 To whom requests for reprints should be addressed, at Laboratory of

Medicinal Chemistry and Biology. Bldg. 37/6D-18. National Cancer Institute,Bethesda, Md. 20205.

Received August 5, 1980: accepted November 21, 1980.

tase II, the rate-limiting step, is elevated 2- to 3-fold in both

resistant lines. Since carbamyl phosphate synthetase II existsas a complex with ATCase, the suggestion is made that levelsof carbamyl phosphate synthetase II are collaterally importantdeterminants of PALA activity. An augmented pool of carbamylphosphate in the resistant variants may serve to competitivelydisplace PALA from ATCase, diminish enzyme inhibition, andallow pyrimidine biosynthesis to proceed despite therapy.

INTRODUCTION

PALA3 (NSC 224131) a transition-state analog inhibitor of

ATCase (5), the second enzyme in the de novo pyrimidinebiosynthetic pathway, exhibits oncolytic activity against a number of transplantable solid murine tumors; most notable is thefact that PALA is curative of the Lewis lung carcinoma, whereasit is ineffective against several murine leukemias (13, 14). Thecytotoxic and cytostatic effects of PALA on the growth of cellsin culture can be reversed by uridine; moreover, the toxicityand antitumor activity of the drug can be reversed by uridine or/V-carbamyl-DL-aspartate in mice, showing, in both cases, that

the effects of PALA are due specifically to blockade of de novopyrimidine nucleotide biosynthesis (12, 27, 30). In culturedcells continuously exposed to PALA, the emergence of resistant variants is associated with elevation of ATCase activity(17). A comparison of cultured cell lines from PALA-sensitivesolid tumors and PALA-refractory leukemias suggests that

naturally occurring resistance to the drug is associated withhigh ATCase activity (14). A recent evaluation of ATCase intransplantable murine tumors growing in vivo also shows thatATCase activity is significantly higher in PALA-refractory asopposed to PALA-sensitive tumors (10, 11). However, amongtumors sensitive to PALA, there is no clear-cut relationship

between ATCase activity and degree of sensitivity to PALA.To best assess factors in addition to ATCase activity that

might determine the responsiveness of tumors to PALA, linesof the Lewis lung carcinoma were selected for resistance bytreatment with repeated subcurative doses of PALA. Two suchlines were chosen for evaluation of enzymological and pharmacological parameters which could influence sensitivity toPALA: one line with a markedly elevated ATCase level; and asecond line with an ATCase activity indistinguishable from theparent PALA-sensitive line. The roles of affinity of ATCase for

its substrates and PALA, restitution of ATCase following PALAtreatment, PALA uptake, activities of the other pyrimidine bio-synthetic enzymes, the effects of PALA on the flux of interme-

3 The abbreviations used are: PALA, N-(phosphonacetyl)-L-aspartic acid; ATC

ase, L-aspartic acid transcarbamylase; PRPP, phosphoribosyi 1-pyrophosphate;CPS II, carbamyl phosphate synthetase II; HEPES, 4-(2-hydroxyethyl)-1-pipera-zineethanesulfonic acid; DHOase, L-dihydroorotase.

894 CANCER RESEARCH VOL. 41

Research. on December 6, 2018. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

Resistance to PALA in Lewis Lung Carcinomas

diates through this de novo pathway and nucleotide pool sizes,as well as the anabolism of pyrimidine nucleosides via thesalvage pathways in these 3 lines, have been determined andare discussed in this report.

MATERIALS AND METHODS

Animals. Resistant variants of the Lewis lung carcinomawere established as follows. C57BL/6J x DBA/2J F, (hereafter called B6D2F,) mice bearing s.c. implants of 106 Lewis

lung carcinoma cells were treated on Days 1 to 4 with 200 mgof PALA per kg, a regimen which will generally cure a majorproportion of the animals thus treated (14). On Day 21, the 3largest tumors were excised and transplanted independentlyinto 3 groups of mice. At each transplant generation of eachsubline, the single tumor which grew most rapidly was transplanted to the subsequent generation. The 3 PALA-resistantlines thus derived all had elevated ATCase activities (comparable to the activity of the line designated LL/PALA-J in Table3); this last-mentioned line, LL/PALA-J, was selected for further study. The LL/PALA-C line was developed in an analogous

manner except that tumors from the first generation survivorswere pooled before passage. After 3 such passages, thosetumors that exhibited growth rates closest to that of the untreated parent line were selected for passage to the subsequentgeneration.

Establishment and Maintenance of Cell Lines. Ten daysafter transplantation, LL/0, LL/PALA-C, and LL/PALA/J tumors were excised and placed in cold Hanks' balanced salt

solution. Necrotic and connective tissues were dissected away,and the tumors were finely minced with a scissor. Cells weredispersed by serial extrusion through 19-, 22-, and 25-gaugeneedles followed by incubation at 37°for 25 min in the pres

ence of 0.25% pancreatin (Grand Island Biological Co., GrandIsland, N. Y.). Enzymatic digestion was quenched by additionof an equal volume of cold Eagle's Medium No. 2 (NIH Media

Unit) containing 2x L-glutamine, penicillin, streptomycin and10% fetal calf serum. Cells were washed twice, planted in 150-sq cm culture flasks, and incubated at 37°under a 5% CO2:

95% air atmosphere. The cells, which grow in monolayer, werepassaged approximately twice a week. PALA-resistant lines

were maintained in the absence of drug.Chemicals. L-{4-14C]Aspartic acid (50 mCi/mmol), L-[L/-

14C]aspartic acid (216 mCi/mmol), sodium [14C]bicarbonate(57.5 mCi/mmol), deoxy-[2-14C]uridine (39 mCi/mmol), and[2-"'C]thymidine (60 mCi/mmol) were obtained from Amer-sham/SearleCorp. (Arlington Heights, III.). [carboxy-14C]Oroticacid (41.3 mCi/mmol), [carboxy-14C]orotidine 5'-monophos-phate (30.7 mCi/mmol), [2-'4C]uridine (52.4 mCi/mmol), anddeoxy-[2-14C]cytidine (32 mCi/mmol) were purchased fromNew England Nuclear (Boston, Mass.). [acefy/-14C]PALA (18.6

mCi/mmol) was obtained through the Drug Development Program of the National Cancer Institute and was purified togreater than 99% radiochemical purity on a 1- x 10-cm columnof Aminex A-14 (Bio-Rad Laboratories, Richmond, Calif.) by

elution with a 0 to 1 M gradient of ammonium bicarbonate. Thedilithium salt of carbamyl phosphate and L-glutamate oxaloac-etate transaminase (EC 2.6.1.1), 10 mg/ml (180 ID/mg protein), was obtained from Boehringer Mannheim Corp. (Elms-ford, N. Y.). Pyruvate decarboxylase (EC 4.1.1.1), 12 mg/ml(18 lU/mg protein), orotidine-5'-phosphate pyrophosphorylase

and orotidine-5'-phosphate decarboxylase (0.32 lU/mg pro

tein), PRPP, and L-aspartic acid were products of Sigma Chem

ical Co. (St. Louis, Mo.). Pyrazofurin was supplied by Eli Lillyand Co. (Indianapolis, Ind.) and PALA by the DevelopmentalTherapeutics Program of the National Cancer Institute (Be-thesda, Md.). Versilube-F50 was obtained from Harwick Chem

ical Corp., Cambridge, Mass.W-Carbamyl-L-[L/-"'C]aspartate (216 juCi//imol) was synthe

sized by mixing 250 juCi of L-[L/-14C]aspartic acid (50 /iCi/ml),

50 jitl of 0.1 25 M carbamyl phosphate, and 5 ml of 0.05 M Tris-HCI, pH 8.4; the reactants were then incubated for 9 hr at 37°

with 25 jul of purified Escherichia coli ATCase (365 lU/mg;19.5 mg/ml; New England Enzyme Center, Boston, Mass.).The reaction mixture was lyophilized overnight and redissolvedin 1 ml of 0.01 M LÃŒGI,pH 2.72. A/-Carbamyl-L-[L/-'4C]aspartatewas separated from unreacted L-[L/-'4C]aspartate on a Jeol

JLC-5AH amino acid analyzer equipped with an 8- x 700-mmcolumn packed with Hamilton AN-90 cation-exchange resin

and developed with 0.01 M LiCI, pH 2.72, at a flow rate of 0.67ml/min. A/-Carbamyl-L-aspartate and L-aspartic acid elute at35 and 365 min under these conditions. Following lyophiliza-

tion, extraction of LiCI was accomplished by the procedure ofChristopherson era/. (2). Electrophoresis at 3000 V for 30 minof an aliquot of the resultant A/-carbamyl-L-[L/-14C]aspartate on

Whatman No. 3M paper moistened with 0.1 M sodium phosphate buffer, pH 7.2, revealed radiochemical purity of >97%;a yield of 36% was obtained. To dilute the specific activity ofthe A/-carbamyl-L-{L/-'iC]aspartate to 11.3 /iCi//imol, cold N-

carbamyl-L-aspartate was prepared according to the modifiedmethod of Toninello ef al. (29). L-Aspartic acid (1 g) and KCNO

(0.61 g) were added to 10 ml 0.01 M KOH and incubated at37°for 48 hr. The reaction mixture was then chromatographed

on Dowex 50W-X8 (200 to 400 mesh), and the percolate waslyophilized. L-[carboxy-14C]Dihydroorotic acid was synthesizedenzymatically from [carboxy-14C]orotic acid (41.3 mCi/mmol)

essentially as described by Smithers et al. (26).Enzyme Assays. Mice were killed by cervical dislocation;

tumors were promptly excised and homogenized for 10 secwith a Polytron homogenizer in 1 volume of a buffer consistingof 30% dimethyl sulfoxide, 5% glycerin, 0.1 M Tris-HCI (pH8.4), 0.1 M KCI, 0.2 mW EDTA, and 0.2 mw dithiothreitol. Aftercentrifugation for 3 min at 12,000 x g, supernatants wereimmediately used for assay. All dispensations were made at4°. CPS II activities were measured by determination of theincorporation of NaH14CO3 into N-['4C]carbamyl-L-aspartate.

To 1500-fil Eppendorf reaction vessels were added 10 /il of 50

HIM HEPES buffer, pH 7.2, containing 62.5 mw ATP, 75 mwMgCI2, 37.5 mw KCI, 37.5 mw L-aspartic acid, 8.2 mw L-

glutamine, and 12.5% glycerin; 5 ¿ilof either HEPES buffer or0.05 M 5-chloro-4-oxo-L-norvaline, and 10 n\ of tumor supernatant. After addition of 5 fi\ of NaH14CO3 (1 mCi/ml) to the

underside of the lids, vessels were centrifuged at 12,000 x gfor ~1 sec to mix reactants and incubated at 37°for 15 min.The reaction was terminated by heating at 95°for 2 min, after

which 50 /il of 2 N HCI was added to dissipate unreacted[14C]bicarbonate. Endogenous ATCase quantitatively converted the newly formed [14C]carbamyl phosphate to A/-['4C]-

carbamyl-L-aspartate which was quantitated by immersing the

vessels in vials containing 15 ml of Aquasol and counted at85% efficiency in a Beckman LS-230 scintillation spectrome

ter. CPS II activity was taken as the difference in counts

MARCH 1981 895

Research. on December 6, 2018. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

f. W. Kensler et al.

obtained in reaction vessels lacking and containing the L-glu-tamine antagonist 5-chloro-4-oxo-L-norvaline. Confirmationthat the differential radioactivity was solely A/-[14C]carbamyl-L-

aspartate was accomplished by chromatography on a columnof high-resolution anion-exchange resin (31).

ATCase activity was assayed using L-[4-14C]aspartic acid as

cosubstrate as described elsewhere (11 ). DHOase activityin the forward direction was assayed by measuring the L-[14C]dihydroorotate formed from A/-carbamyl-L-['4C]aspartate.

The reaction mixture containing 10 /il of /V-carbamyl-L-fC]-aspartate (11.3 /iCi/mol, 44.8 /¿Ci/ml)in 0.05 M Tris-HCI, pH8.4, and 5 /il of the tumor supernatant were incubated at 37°

for 15 min. After termination of the reaction by heating for 2min at 95°, 10 /il of the reaction mixture were spotted on

Whatman No. 3M paper, overspotted with L-aspartic acid, N-carbamyl-L-aspartate, and L-dihydroorotate, and electropho-

resed for 30 min at 3000 V in 0.1 M sodium phosphate buffer,pH 7.2. L-Dihydroorotate was located by the method of Fink ef

al. (8), and radioactivity was counted as described earlier.Dihydroorotate dehydrogenase was assayed in intact mito

chondria isolated from individual tumors by the method described by Schnaitman and Greenawalt (25). The assay conditions were 0.25 M sucrose, 10 mw HEPES buffer (pH 7.4),0.1 HIM [MC]dihydroorotic acid, and mitochondrial enzyme in

a final volume of 40 /J. Reaction vessels were incubated for 10min at 37°and then 2 min at 95°.After centrifugation for 1 minat 12,000 x g, the newly formed [carboxy-'"C]orotic acid was

quantitatively converted to UMP by the conjoint actions oforotate phosphoribosyl transferase and orotidine-5'-mono-

phosphate decarboxylase from yeast. In practice, 10 /il of 0.05M Tris-HCI, pH 8.4, containing 0.01 M PRPP-MgCb and 100

/ig of yeast enzyme were added to each vessel, and theliberated 14CO2 was trapped as described below. Reaction

blanks were constituted with heat-denatured mitochondrial en

zyme.For the measurement of orotate phosphoribosyl transferase

activity, the following reagents were admixed in Eppendorfcentrifuge tubes: 5 /il of [canboxy-14C]orotate (50 /tCi/ml) and

either 5 /il of 0.01 M MgCI2 in 0.05 M Tris-HCI, pH 8.4, or 5 /ilof 0.01 M PRPP-MgCI2 in the same buffer. The reaction was

initiated by the addition of 20 /il of tumor supernatant at thesame time as a 5-/J droplet of 40% KOH was deposited on the

underside of the vessel lid. The resultant assembly was closedrapidly and securely and then incubated at 37°for 15 min, at95°for 2 min, and finally overnight at room temperature. At the

end of these incubations, the lids were removed, and theradioactivity trapped in the alkali thereon was measured byscintillation spectrometry. Activity was expressed as PRPP-dependent decarboxylation of [carboxy-14C]orotate. Orotidine-5'-monophosphate decarboxylase activity was assayed simi

larly but with 0.05 M Tris, pH 8.4, in lieu of PRPP-MgCI2, using[MC]orotidine 5'-monophosphate (20 /iCi/ml) as the radiola-

beled substrate.Kinetic studies on ATCase were carried out on partially

purified tumor extracts. Immediately prior to assay, 12,000 xg supernatants were passed through a Sephadex G-25 column

(1 x 20 cm) equilibrated with 2.5% glycerol, 15% dimethylsulfoxide, 0.05 M Tris-HCI (pH 8.4), 0.05 M KCI, 0.1 mw EDTA,

and 2 mw dithiothreitol. The reaction mixtures, in a total volumeof 20 /il, contained 5 /il of i_-[4-14C]aspartic acid, 5 /il of

carbamyl phosphate in 0.05 M Tris-HCI (pH 8.4), 5 /il of PALA

or 5 /il of buffer, and 5 /il of the appropriate tumor extract. Thereactants were driven together by brief centrifugation in anEppendorf centrifuge at 12,000 x g, incubated at 37°for 10

min, and processed as described elsewhere (11). When L-aspartic acid was the variable substrate, the concentration ofcarbamyl phosphate was held fixed at 2.0 mw, and the concentration of L-aspartic acid varied at 8 points from 0.078 to

5.0 mW. When carbamyl phosphate was the variable substrate,the concentration of L-aspartic acid was held constant at 1.25

ITIM,and the concentration of carbamyl phosphate varied at 8points from 1.25 to 10 /ÃŒM.PALA concentrations used were 0,0.0025, 0.01, and 0.1 /IM.

Measurement of pyrimidine nucleoside kinase activities wascarried out by incubating at 37°for 15 min a reaction mixtureconsisting of 5 /il of '4C-labeled precursor (0.25 /iCi; 5 nmol),

5 /il of 0.01 M ATP-MgCI2 in 0.05 M Tris (pH 8.4) or buffer

alone, and 5 /il of tumor supernatant. The reaction was terminated by adding 10 /il of 2 N HCI followed by centrifugation at12,000 x g for 3 min. Samples (5 jul) were subjected toascending paper chromatography on Whatman No. 3M paperfor 18 hr using isopropyl alcohol:EDTA (saturated solution):toluene: 14 M NH4OH (320:44:40:4) as solvent (32). The phos-

phorylated products are retained at the origin.PALA Assay. PALA was assayed in tumor homogenates by

radiometrie enzyme inhibition assay (6).Measurement of PRPP. Tumors were excised and homoge

nized in one volume of cold 3% HCICv Following a 15-seccentrifugation at 12,000 x g, supernatants were rapidly neutralized by the addition of 100 /il 1 M KH2PO„and approximately10 /il 40% KOH. PRPP measurements were conducted byadding 5 /tl of neutralized tumor supernatant and 5 /tl [carboxy-14C]orotate (50 /iCi/ml) to the bottoms of Eppendorf vessels.

The reaction was initiated by the addition of 20 /il of 0.05 MTris-HCI, pH 8.4, containing 0.01 M MgCI2 and yeast orotatephosphoribosyl transferase plus orotidine 5'-monophosphate

decarboxylase (2.5 units/ml) at the same time as a 5-jul dropletof 40% KOH was deposited on the underside of the vessel lid.The resultant assembly was closed rapidly and incubated overnight at room temperature. At the end of this incubation, thePRPP-dependent liberation of 14CO2 was quantified as de

scribed above. PRPP standards, 0.01 to 10 mM, were constituted in 0.05 M Tris-HCI, pH 8.4, and delivered as 5 /tl in lieu

of tumor supernatant.Uptake of [14C]PALA. Lewis lung carcinoma cells were

suspended at a concentration of 3 x 106 cells/ml of Dul-becco's phosphate-buffered saline plus 5.6 mw glucose. Uptake studies were performed at either 37°or 4°in siliconized25-ml spinner flasks. [14C]PALA was added to cell suspensions

at a final concentration of 200 /IM unless otherwise indicated,and at appropriate times, 900-/il aliquots of cells were transferred to microcentrifuge tubes containing 600 /il of Versilube-F50 silicone oil. Incubations were terminated by centrifugationof the cells through the oil at 12,000 x g for 1 min in anEppendorf centrifuge. The cell pellets were solubilized overnight in 1.5 ml of 0.2 N NaOH at 37°. Samples were then

neutralized and suspended in 15 ml of Aquasol prior to scintillation counting. Data were corrected for entrapped extracellular drug by subtracting the radioactivity measured in cellsfollowing a 1-min incubation in the presence of the appropriate[14C]PALA concentration.

[14C]Bicarbonate Incorporation into Cells. Flasks (150 sq

896 CANCER RESEARCH VOL. 41

Research. on December 6, 2018. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

Resistance to PALA in Lewis Lung Carcinomas

cm) containing 5 x 106 logarithmically growing cells were rinsedtwice with HCO3-free Eagle's Medium No. 2 containing 2x L-

glutamine, 25 mw HEPES, and 10% dialyzed fetal calf serum.To begin an experiment, 10 ml of the aforementioned medium,with or without 2 x 10~5 M PALA, were added to flasks. Aftera 3-hr incubation, 200 /iCi of NaH14CO3 was added to each

flask, and the cells were incubated for an additional 3 hr.Following this incubation, the medium was decanted (the unincorporated radiolabel was trapped in alkali for disposal following acidification of the media), and the cells were trypsin-ized, washed, pelleted, and lysed by the addition of 1 ml ice-cold 3% HCIO4. Acid-soluble supernatants were chromato-graphed on an anion-exchange column (31); 2.5-min fractions

were collected and counted.Measurement of Orotate, Orotidine, and Nucleotides. Tu

mor orotate and orotidine levels were measured in a mannersimilar to that described by Moyer and Handschumacher (24).Tumors were homogenized in 2 volumes of ice-cold 1 M HCIO4.Following centrifugation at 12,000 X g for 3 min, the acid-soluble supernatant was heated at 95°for 15 min and subse

quently adjusted to pH 3 to 4 with KOH. Chromatography of

Table 1Activity of PALA against Lewis lung carcinomas

Lewis lung carcinomas (106 viable cells) were implanted s.c. in male B6D2Fimice 24 hr before the initiation of therapy.

TumorlineLL/0LL/PALA-CLL/PALA-JPALA

dose" (mg/kg/day)20010050Untreated

control200100Untreated

control300180108Untreated

controlMedian

survivaltime(days)>60>60>603031342831383828Increaseinlifespan

(%)>100>100>1000112101136360Long-termcures10/1010/106/100/100/100/100/100/100/100/100/10

Days 1 to 4, i.p.

the supernatant was done on a Waters Associates, Inc. (Mil-ford, Mass.) high-performance liquid Chromatographie system

consisting of 2 Model 6000 A pumps, a Model 660 solventprogrammer, a Cis-iiBondapak column, a Model 440 absorb-ance detector monitoring 254 and 280 nm, and a linear dual-

channel recorder (Houston Instruments, Austin, Texas). Orotate had a retention time of 4.2 min when eluted with a 0.05M sodium formate buffer, pH 4.1, at a flow rate of 1 ml/minand was separable from orotidine (elution time, 3.2 min).

Nucleotides were extracted from tumor homogenates asdescribed above and chromatographed on a Partisil 10 SAXcolumn (Whatman Inc., Clifton, N. J.) in a manner similar to thatdescribed by Lui et al. (21). Initial conditions were 0.005 Mpotassium phosphate, pH 2.8, and final conditions were 0.5M potassium phosphate, pH 4.8. After elution for 10 min withthe starting buffer at a flow rate of 1 ml/min, the flow rate wasraised to 2 ml/min, and a linear gradient was started with thefinal buffer; 100% final buffer conditions were reached 20 minlater. Under these gradient conditions, we observed the following retention times (min): CMP, 9.5; AMP, 12.7; UMP, 15.0;GMP, 16.5; UDP /V-acetylglucosamine, 18.2; UDP glucose and

UDP galactose, 18.6; CDP, 20.8; UDP, 22.0; ADP, 23.2; UDPglucuronate, 23.8; GDP, 26.1; UTP, 29.0; CTP, 30.6; ATP,32.8; and GTP, 36.0.

RESULTS

Sensitivity and Resistance to PALA in Vivo and in Vitro.The Lewis lung carcinoma is one of several transplantablemurine tumors that is sensitive to the therapeutic effects ofPALA (14). Repetitive treatment of mice bearing this carcinomawith subcurative doses of PALA yielded a number of variantsthat were totally refractory to the drug. Table 1 summarizes theantitumor activity of PALA against the 3 Lewis lung carcinomalines that were chosen for subsequent biochemical evaluationin this study. Against the parent line (LL/0), daily treatment for4 consecutive days with PALA at a dose of 100 mg/kg produced 100% long-term survivors. However, against the other2 lines, LL/PALA-C and LL/PALA-J, no long-term survivorswere seen at similar or higher doses of PALA. Utilizing previ-

•—•Saline°—°0.05x 10"4M PALA

a—a 0.2 x 1<T4M PALA1.0 x 10~4M PALA

Chart 1. Effect of PALA on growth of Lewis lungcarcinoma cells in culture. Cells were subcultured 24hr prior to the addition of control media or mediacontaining PALA at the indicated concentrations. Atthe appropriate times after treatment, cells were disaggregated with 0.05% trypsin:0.02% EDTA for enumeration with a Coulter counter. Points, mean of 3cultures.

234DAYS

MARCH 1981 897

Research. on December 6, 2018. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

T. W. Kensler et al.

ously defined criteria for PALA sensitivity and resistance (11 ),the less than 40% inhibition of tumor growth (data not shown)and the percentage of increase in life span clearly delineatethese latter 2 lines as resistant to PALA.

Cell lines established from the LL/0, LL/PALA-C, and LL/PALA-J tumors were also evaluated for sensitivity to PALA.Chart 1 shows the effects of continuous exposure to gradedconcentrations of PALA on the growth of these 3 Lewis lungcarcinoma lines in culture. In the absence of PALA, all 3 linesgrew with a doubling time of approximately 24 hr. Exposure ofcultures for 5 days to 5 fiM PALA reduced the growth rate inthe LL/0 line over 50% without appreciable effect on thegrowth of the 2 PALA-resistant lines. Incubation with 20 ¡J.M

PALA eliminated growth of the LL/0 cells while inhibitinggrowth rates by 50% in the 2 resistant lines, denoting a 4- to5-fold differential in the sensitivity of the LL/0 versus LL/PALA-C and PALA-J lines to PALA. The LL/PALA-C and LL/PALA-J lines also grew in the presence of 100 fiM PALA,although at a greatly diminished rate; this concentration ofPALA was cytotoxic to the LL/0 line.

Resistance to PALA in the LL/PALA-C and LL/PALA-J celllines is stable, at least for the short term. Cells (1 x 106) from

passage 12 of each line were evaluated for tumorigenicity andresistance to PALA in vivo (Table 2). All 3 lines retained theirtumorigenicity, and the LL/PALA-C- and LL/PALA-J-derived

tumors were refractory to treatment with 200 mg PALA per kgon Days 1 to 7 postinoculum; the LL/0 cell line maintained itssensitivity to PALA. Subsequent experiments with these celllines utilized cells from the 12th to 15th passage.

Effects of PALA on ATCase. The initial basis for development of PALA-resistant variants was to determine whether the

Table 2

Tumorigenicity of Lewis lung carcinoma cell lines

Cellline"LL/0LL/PALA-CLL/PALA-JTreatmentNonePALA0NonePALACNonePALACTumorgrowth

inhibition6(%)01000000Mediansurvivaltime(days)30>6033342929Increaseinlifespan

(%)0>1000000Long-termsurvivors0/1010/100/100/100/100/10

correlation between low and high ATCase activity and tumorsensitivity and resistance to PALA seen in a broad spectrum oftumors of varying sensitivities (11) could also be observed invivo within a single tumor type. Of the 4 resistant Lewis lungcarcinoma lines developed, 3 had markedly elevated ATCaseactivities (e.g., LL/PALA-J), whereas the fourth, which wasdeveloped in a slightly different manner (LL/PALA-C), had anATCase activity identical to the activity in the parent, PALA-sensitive line (see Table 3, Column 3). The obvious suggestionfrom the appearance of such a line that ATCase activity wasnot the sole determinant of resistance to PALA prompted a fullbiochemical evaluation of the potential factors underlying themechanism of resistance in the LL/PALA-C line.

Although the specific activity of ATCase per se in the LL/PALA-C line was deemed not to be of primary importance, it

was still felt that ATCase, the unquestioned site of PALA action,was likely in some way to be an important determinant of PALAresistance. To examine whether ATCase from resistant tumorsmight have altered affinities for PALA compared to the enzymefrom the sensitive tumor, determinations of the kinetic parameters of ATCase from the 3 lines were made, and the resultsare shown in Table 4. The values for the K¡of PALA for eitherthe competitive substrate carbamyl phosphate (~20 nM) or thenoncompetitive substrate L-aspartic acid (~350 /¿M)were notdissimilar for the sensitive and resistant lines. Additionally, thesubstrate affinity constants were closely comparable in the 3lines.

An assessment of the uptake of PALA and subsequent inhibition of the target enzyme, ATCase, was made by treatingmice bearing established carcinomas with PALA at doses between 10 and 400 mg/kg. Tumors were excised 24 hr laterand assayed for ATCase activity as well as for residual PALA.

Table 4ATCase kinetics in PALA-sensitive and -resistant Lewis lung carcinomas

Kinetic studies on the inhibition of ATCase by PALA were carried out asdescribed in "Materials and Methods." Values are the means of 2 separateexperiments. The Km's of these partially purified enzymes for carbamyl phosphateare lower than the value reported in the literature: 1.8 X 10~6 M (9). This

discrepancy may reflect differences in the enzymes, differences in the analyticaltechniques used, or a biphasic (high-Km as well as Iow-Km) affinity of the enzyme.

1 x 10e cells from passage 12 implanted s.c.6 On Day 15.c 200 mg/kg, i.p.. Days 1 to 7.

CarbamylphosphateTumor

lineLL/0

LL/PALA-CLL/PALA-JK.

W0.5

0.50.5K,

(PALA)(/IM)0.02

0.030.03L-Aspartic

acidK.O.M)800

10001100K,

(PALA)(UM)330

350360

Table 3Activity of pyrimidine biosynthetic enzymes in PALA-sensitive and resistant Lewis lung carcinomas

Enzyme assays were performed on freshly prepared 12.000 x g supernatants of tumor homogenates or on intact mitochondria as described in "Materials andMethods." For each tumor line, assays were done in triplicate on individual tumors from either 4 or 5 mice.

SpecificactivityTumor

lineLL/0

LL/PALA-CLL/PALA-JCPS

IIa0.9

±0.2a1.8 ±0.3'3.0 ±04'ATCase0114+13

116 ±15204±i e'Dihydroorotasec6.1

±0.78.2 ±0.7'11.5

±O.g'Dihydroorotate

dehydrogenased20.6

±2.819.8 ±1.620.5 ±1.4Orotate

phosphoribo-syltransferase13.4

±0.512.3 ±0.812.7 ±2.0Orotidine-5'-

monophosphatedecarboxylase17.7

±3.718.2 ±2.017.9 ±1.7

nmol ['4C]CO2 consumed per mg protein per hr.nmol A/-carbamyl-L-{'4C]aspartate formed per mg protein per hr.

" nmol ['"CJdihydroorotate formed per mg protein per hr.^ nmol [MC]CO2 evolved per mg protein per hr.3 Mean ±S.D.' Differs from LL/0, p < 0.01.

898 CANCER RESEARCH VOL. 41

Research. on December 6, 2018. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

Resistance to PALA in Lewis Lung Carcinomas

As shown in Chart 2/4, there was a log-linear relationship

between the dose of PALA and the degree of inhibition ofATCase at 24 hr in all lines. However, at doses greater than 10mg/kg, the 2 resistant lines showed small but significantly (p< 0.01 ) lower percentages of ATCase inhibition. This disparitywas most pronounced in the LL/PALA-J line, where inhibition

lagged 10 to 20% behind that seen in the LL/0 line for anygiven dose. In parallel with the inhibition of ATCase, the concentrations of PALA in both resistant tumors were lower thanin the sensitive tumor, particularly at high doses of PALA (Chart26). A combination of the data presented in A and B (Chart2C) shows that inhibition of tumor ATCase varies with the logof intratumoral PALA concentration. This log-linear inhibition of

ATCase by PALA has been described previously in mousespleen (6). Comparisons among the 3 lines show that, atcomparable inhibition of tumor ATCase, intratumoral PALAconcentrations are roughly equivalent.

It also could be argued that resistance to PALA might stemfrom a more rapid restitution of ATCase in the refractory linesas contrasted to the drug-sensitive tumor following exposure

to PALA in vivo. Enhanced restoration of enzyme activity couldresult from more rapid synthesis of enzyme molecules in newor existent cells or from more extensive displacement of PALAfrom existent ATCase molecules as a consequence of competitive binding by an augmented pool of carbamyl phosphate.Although it is not, as yet, possible to test these 2 alternativesdirectly, measurements were made of the rate of reappearanceof ATCase activity following single administration of PALA at awell-tolerated dose of 400 mg/kg. As shown in Chart 3/4, PALA

has a dramatic and prolonged inhibitory effect on ATCaseactivity in all 3 tumors. ATCase inhibition recovers slowest inthe LL/0 line; full recovery is still not evident by Day 13posttreatment. By contrast, the LL/PALA-J and LL/PALA-C

lines achieved full restitution of ATCase activity by Days 9 and13, respectively. From Chart 36, it can be seen that thediffering restitutions of ATCase activity correlate with the retention of inhibitory concentrations of PALA. Similarly, at thosetimes when enzyme activity is fully restored, intratumoral PALAconcentrations fall below the limit of detectability (0.1 ¡J.M)forthe enzyme inhibition assay used.

[14C]PALA Uptake into Cells. Inasmuch as the direct assay

of intratumoral PALA concentrations in the 3 Lewis lung carcinoma lines showed that there was less drug in the PALA-

resistant tumors than in LL/0 per given dose (Charts 2 and 3),the possibility was raised that there exist differences betweenthe sensitive and resistant lines in their abilities to accumulateor transport PALA. The kinetics of [14C]PALA uptake were

established in the LL/0 cell line. Drug uptake was linear withtime for at least 3 hr following exposure of cells to [14C]PALA

(200 ¡IM)and apparently occurs by passive diffusion. The timecourses of intracellular accumulation of PALA in cells exposedto [14C]PALA at either 37°or 4°were comparable. Additionally,

uptake was nonsaturable; accumulation of drug at 1 hr waslinear with concentration between 0.1 and 2 HIM [14C]PALA.

This observation is in contrast to the situation in human leukocytes in which [14C]PALA uptake is saturable with an appar

ent Kmof 0.4 rtiM (19). All 3 Lewis lung cell lines accumulated[14C]PALA at a rate of 25 ±2 (S.D.; n = 5) pmol/106 cells/hr.

Thus, differential limitation of PALA accumulation does notappear to play an important role in the resistance of the LL/PALA-C and LL/PALA-J cell lines.

Pyrimidine Biosynthetic Enzymes. Prompted by the conjecture that the biochemical lesion in certain of the resistantvariants was other than at the level of the carbamylation of L-aspartic acid, attention was directed to the other enzymesinvolved in the de novo biosynthesis of pyrimidines. Table 3presents the specific activities of these enzymes in the PALA-sensitive and -resistant Lewis lung carcinomas. Proceeding

seriatim through the enzymes on this pathway, it can be appreciated that the specific activity of CPS II, the soluble L-glutamine

amidotransferase, is the lowest of the enzymes assayed. Additionally, activity of this enzyme is significantly elevated inboth resistant lines: 2- and 3-fold in the LL/PALA-C and LL/PALA-J lines, respectively. This finding assumes pharmacolog

ical relevance when it is recalled that PALA is a competitiveinhibitor of the utilization of carbamyl phosphate by ATCase.By contrast, those enzymes that are not a part of the CPS II-ATCase-DHOase complex exhibit activities that are compara

ble in the sensitive and resistant lines.Because of the marked elevation of CPS II activity in the

•B

n< o

p? u

0.1100

DOSE OF PALA (mg/kgl

100

CWSE OF PALA (mg/kgl

-C

59 30I- 20if

10

INTRATUMORAL PALA CONCENTRATION IpMI

Chart 2. Dose response of inhibition of ATCase activity by PALA and intratumoral concentrations of PALA in Lewis lung carcinomas sensitive and resistant toPALA. In A, 0.9% NaCI solution or indicated doses of PALA were injected i.p. into mice bearing s.c. Lewis lung carcinoma. •,LL/0; O, LL/PALA-C; D, LL/PALA-J.Animals were sacrificed 24 hr later, and the tumors were removed, homogenized, and assayed for ATCase activity as detailed in "Materials and Methods." In B,

intratumoral PALA concentration determinations were made on the homogenates assayed for ATCase activity. Points, means of triplicate measurements done onindividual tumors from groups of 5 mice. In C, for each dose of PALA administered, the percentage of inhibition of ATCase activity (A) is plotted versus the log of theassayed intratumoral PALA concentration (B).

MARCH 1981 899

Research. on December 6, 2018. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

T. W. Kensler et al.

10090

80

70

80

50

40

30

20

10

10

11i fE

II 1

0.12 4 6 8 10 12

DAYS AFTER PALA TREATMENT

2 4 6 8 10 12

DAYS AFTER PALA TREATMENT

Fig. 3. Time course of restitution of ATCase activity and intratumoral concentrations of PALA in Lewis lung carcinomas sensitive and resistant to PALA followinga single dose of PALA. In A, tumor-bearing mice were given injections of 0.9% NaCI solution or 400 mg of PALA per kg 12 days after s.c. implantation with Lewis lungcarcinomas. Animals were sacrificed at the indicated times, and the tumors were removed and homogenized. Analysis of ATCase activity is as described in "Materialsand Methods." •.LL/0; O, LL/PALA-C; D, LL/PALA-J. In B, intratumoral PALA concentration determinations were made on the homogenates assayed for ATCase

activity. The lower limit of detectability for the assay used is 0.1 fiM. Points, means of triplicate measurements done on individual tumors from groups of 5 mice.

resistant variants, it was felt to be of interest to determine theintratumoral concentrations of those intermediary metabolitesknown to exert the strongest allosteric controls over the activityof this enzyme. As shown in Table 5, the levels of PRPP, apositive effector of CPS II at the concentrations measured (23,28), were elevated 2-fold in both resistant variants when com

pared to the parent line. DTP, a negative effector of CPS II,was also elevated in the LL/PALA-J resistant line but to a

concentration that is probably without differential effect on CPSII activity (28).

Effect of PALA on Pyrazofurin-provoked Accumulation ofOrotate and Orotidine. To further define the basis for the stateof resistance in the Lewis lung carcinoma variants, a moredynamic approach was taken. Precursor flux through the denovo pyrimidine pathway was evaluated by measuring theaccumulations of orotic acid and orotidine in tumors of micetreated with pyrazofurin (a fraudulent nucleoside that, afterphosphorylation, inhibits orotidine-5'-monophosphate decar-

boxylase) and with either 0.9% NaCI solution or PALA. Tumor-bearing mice were given injections simultaneously of 100 mgpyrazofurin per kg and either 0.9% NaCI solution or PALA (400mg/kg) and sacrificed 3 hr later, and the tumors were assayedfor orotate and orotidine by high-performance liquid chroma-tography. As already described by Moyer and Handschu-macher (24), such a treatment protocol in the parent Lewislung carcinoma leads to an accumulation of orotate and orotidine (-0.5 /imol/g) in the absence of PALA, but concurrent

PALA treatment abrogates their accumulation. When this experiment was performed on the PALA-resistant variants, appre

ciable orotate and orotidine accumulation persisted as shownin Chart 4; inhibition of orotate and orotidine accumulation bysingle-dose PALA treatment was only 75 and 50% in the LL/PALA-C and LL/PALA-J lines, respectively. This finding is

consistent with the data presented in Charts 1/4 and 2A showing less inhibition of ATCase by PALA in the 2 resistant variants.Clearly then, in the face of PALA insult, these 2 resistantvariants maintain to some degree a functionally operative denovo pyrimidine biosynthetic pathway.

Table 5

Levels of CPS II effectors in Lewis lung carcinomas

Assays for PRPP and DTP were carried out on freshly excised tumors asdescribed in "Materials and Methods."

Concentration(UM)Tumor

lineLL/0

LL/PALA-CLL/PALA-JPRPP(+)29

±6a (12)668 ±9C55 ±8°UTP(-)102

±10(5)118 ±10217 ±13C

Mean ±S.E.b Numbers in parentheses, number of determinations.c Differs from LL/0, p < 0.01.

Incorporation of NaH14CO3 into Pyrimidines. Histograms of

the radiolabeled eluent from an automated anion-exchangecolumn used to separate the intermediates of de novo pyrimidine biosynthesis and pyrimidine nucleotides of the acid-soluble fraction of Lewis lung carcinoma cells labeled withNaH14CO3 are shown in Chart 5. After a 3-hr incubation withNaH14CO3, the majority of the incorporated label was found in

the uridine nucleotides, particularly in UTP, UDP glucose, UDPglucuronate, and UDP /V-acetylglucosamine. Very little labelwas observed in the pyrimidine intermediates, although N-[14C]carbamyl-L-aspartate was readily discernible. When LL/0

cells were treated with 20 JUMPALA, a concentration thatretards cell growth completely in LL/0 and about 50% in LL/PALA-C and LL/PALA-J (Chart 1), a marked reduction in theincorporation of NaH14CO3 into the cytidine and uridine nucleo

tides as well as into /V-carbamyl-L-aspartate is observed. Bycontrast, little perturbation of incorporation is observed in eitherthe LL/PALA-C or LL/PALA-J lines. These findings serve toreiterate the observation made in tumors that the PALA-resistant variants maintain a functional de novo pyrimidine biosynthetic pathway following PALA treatment.

Nucleotide Pools. The effect of PALA on the nucleotidecontents of the 3 Lewis lung carcinoma lines were also determined. Shown in Table 6, the pool sizes for the adenosine,guanosine, cytidine, and uridine nucleotides were comparablein the LL/0 and LL/PALA-C lines. However, the LL/PALA-J

900 CANCER RESEARCH VOL. 41

Research. on December 6, 2018. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

Resistance to PALA in Lewis Lung Carcinomas

or>

0.5

0.4

gO 0.3DCOo< 0.2

I§

01

Pyrazofurin (100 mg/kgl

Pyrazofurin + PALA (400 mg/kg)

1

LUO LUPALA-C LUPALAJ

TUMORChart 4. PALA inhibition of pyrazofurin-induced orotic acid and orotidine

accumulation in PALA-sensitive and -resistant Lewis lung carcinomas. Groups of5 mice each bearing the indicated Lewis lung carcinoma line were treated withPyrazofurin (100 mg/kg) or Pyrazofurin plus PALA (400 mg/kg). Animals weresacrificed 3 hr after drug administration, and the tumors were removed andanalyzed for orotate and orotidine as described in "Materials and Methods."

Columns, means; bars, S.D.; *, differs from Pyrazofurin plus PALA treated LL/0,

p < 0.01.

line, which has elevated activities of the first 3 pyrimidinebiosynthetic enzymes, has pyrimidine and purine nucl-otidelevels that are augmented 70 and 25%, respectively. Measurement of nucleotide pool sizes 24 hr after treatment with 400mg PALA per kg reveals a 70 to 80% reduction in the cytidineand uridine nucleotides in all 3 lines. As reported previously(24), ATP and GTP levels rise 50 to 100% following PALAtreatment. Whether this increase in purine nucleotides levelsreflects an underutilization or an overproduction remains undefined.

Salvage Pathway Enzymology. Although evidence suggested an important role of CPS II in the differential susceptibility of these Lewis lung carcinomas to PALA, the knowledgethat resistance to antimetabolites is frequently pluricausalprompted an examination of the enzymes of pyrimidine salvagein the sensitive and resistant lines. A greater capacity to ana-bolize exogenous pyrimidines or pyrimidine nucleosides couldcontribute to resistance to PALA by virtue of decreased dependence on the de novo pathway for nucleic acid synthesis.The ability of homogenates of the 3 tumor lines to phosphoryl-ate pyrimidine nucleosides and deoxynucleosides is shown inTable 7. The kinases responsible for phosphorylating uridine,cytidine, thymidine, deoxyuridine, and deoxycytidine exhibitedsimilar activities in the sensitive and the resistant lines. However, it is important to note that PALA treatment apparentlystimulates utilization of the salvage pathway. Specific activitiesof these kinases were increased 1.5- to 5-fold in all lines whentumor homogenates were prepared and assayed 24 hr follow-

* 3

l 2

Il 1

° 0

LL/PALA-C

50 100 150 200MINUTES

250 300 350 400

•'••y,

JLJl'i|§g|jL050 100 150 200MINUTES

LL/PALA-J

300 400

Chart 5. Effect of PALA on the incorporation ofNaH14CO3 into pyrimidine precursors and nucleotides in

the Lewis lung carcinoma cell lines. Logarithmically growing cultures were incubated for 3 hr with either 0.9% NaCIsolution or 20 ¡M PALA after which time 200 /iCi ofNaH 'CO. were added to cultures. After an additional 3

hr of incubation, the cells were harvested, and the incorporation of NaH14CO3 into pyrimidines was measured asdescribed in "Materials and Methods." Open columns,

distribution of radioactivity in 0.9% NaCI solution-treatedcells; stippled columns, those of the respective PALA-treated cells; CAA, N-carbamyl-L-aspartic acid; OHO, L-dihydrootoric acid; UOPGNH2, UDP glucosamine;UDPAG, UDP W-acetylglucosamine; UDPG, UDP glucose;UDPGal, UDP galactose; UDPGA, UDP glucuronate.

MARCH 1981 901

Research. on December 6, 2018. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

T. W. Kensler et al.

Table 6Effect of PALA on nucteotide content of sensitive and resistant Lewis lung carcinomas

Groups of 5 mice each bearing the indicated Lewis lung carcinoma line were treated with 0.9% NaCI solution or PALA (400 mg/kg). Animals were sacrificed 24hr after drug administration, and the tumors were removed and analyzed for nucleotide content as described in "Materials and Methods."

Nucleotide content (nmol/g)

LL/0CMPCDPCTPSCytidineUMPU

DPUDPGdUDPAGUDPGAUTPZUridineAMPADPATPSAdenosineGMPGDPGTPEGuanosine0.9%

NaCIsolution19±1*28

±733±48030

±539±146±252±418±2102±10287136

±18205±30364±4570531

±349±5106±10186PALA6

±3b6±2b9±2b219

±3b7±2b23±3b21±3b8±2b21±3b89136

±16262±54819±73b121724

±458±9157±16b239LL/PALA-C0.9%

NaCIsolution19±

226±

534±27937±

549±757±447±217±

4118±10325123±

28192±26384

±4069939±

550±3107±6196PALA6±

3°5

±3b12±5b239

±2b11±1b26±6b16±4b7±

3b34

±10103149

±31334

±58730±80b121338

±864±11170±21b272LL/PALA-J0.9%

NaCIsolution20

±343±6C68±11C13139

±874±12C77±4C73±9C21

±2217±13C50196

±23241±36532±40C86930

±674±10139±5C243PALA8

±4"7±3b15±2b3012

±2b17±1b17±3b20±3Ò11±2b24±3b1011

56±44212±29797

±61"116542

±585±12175±19302

Mean ±S.E.6 Differs from 0.9% NaCI solution-treated controls, p < 0.05.c Differs from LL/0, p < 0.05." UDPG, UDPglucose; UDPAG,UDPN-acetylglucosamine; UDPGA,UDP glucuronate.

Table 7Effect of PALA on the in vitro phosphory/ation of pyrimidine nucleosides and deoxynucleosides by tumor homogenates

Tumor-bearing mice were treated with either 0.9% NaCIsolution or 400 mg PALA per kg 24 hr prior to sacrifice. Homogenates from freshly excised tumors werespun at 12.000 x g for 3 min. and the supernatants were incubated with the labeled precursors for 15 min at 37°and processed as described in "Materials andMethods." Prior to assay, homogenates from 2 tumors were pooled; 5 such pooled homogenates were assayed for each treatment group.

In vitro phosphorylation (nmol nucleotide formed/mg protein/hr)

TumorlineLL/0LL/PALA-CLL/PALA-JTreatment0.9%

NaCI solutionPALA0.9%

NaCI solutionPALA0.9%

NaCIsolutionPALAUridine13.4

±1.2a

18.5 ±1.411.0

±3.021.6 ±2.5b'c15.6

±3.524.2 ±2.1b'cCytidine13.3

±0.819.0 ±1.6b10.7

±2.823.9 ±4.6b'c16.3

±4.123.5 ±2.2b'cThymidine1.1

±0.54.5 ±1.5b0.8

±0.25.8±1.5b2.5

±1.86.2 ±1.Ob'cDeoxyu

ridine1.1

±0.63.0 ±0.8b0.7

±0.23.4 ±1.7"1.6

±1.04.4 ±26'cDeoxycytidine3.0

±0.48.5 ±1.84.4

±1.116.4 ±3.7b'c2.2

±0.912.2 ±1.56-0

Mean ±S.D.' Differs from 0.9% NaCI solution-treated control, p < 0.05.: Greater than PALA-treated LL/0. p < 0.05.

¡ngadministration of a single 400-mg/kg dose of PALA. Additionally, PALA-stimulated kinase activity was elevated to asignificantly (approximately 25%) greater extent in the resistantlines than in LL/0. At 3 days after PALA treatment, kinaseactivities remain elevated over their respective control levels,although the elevation was not in all cases significant (data notshown).

DISCUSSION

PALA is a stable analog of the transition-state intermediate

of ATCase and combines most of the structural features of the2 natural substrates, carbamyl phosphate and L-aspartic acid.As such, PALA exerts its inhibitory effect on ATCase with bothhigh affinity and specificity; as best is known, no other enzy

matic reactions are directly affected by PALA. PALA producescompetitive inhibition with carbamyl phosphate as the variablesubstrate but is noncompetitive with respect to L-aspartic acid;the apparent KÃŒversus carbamyl phosphate is reported at 1CT8to 10~10 M (9, 11, 17). Similar ATCase kinetics are observed

in the 3 Lewis lung carcinoma lines used in the present study.However, the findings that one of the PALA-resistant variants,LL/PALA-C, is closely comparable to the PALA-sensitive par

ent line in terms of (a) ATCase activity, (b) rate of restitution ofATCase activity following a single therapeutic dose of PALAand (c) ATCase kinetics strongly suggests that factors otherthan ATCase contribute in a profound manner to PALA resistance. The 2 most likely ways in which PALA-induced pyrimidine

starvation could be overcome are enhanced salvage of pyrimidine bases and increased synthesis of competitive substrate,

902 CANCER RESEARCH VOL. 41

Research. on December 6, 2018. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

Resistance to PALA in Lewis Lung Carcinomas

carbamyl phosphate. It would appear that the latter possibilityis a major reason for resistance to PALA in the LL/PALA-C

line.The activity of the salvage pathway could influence the

response of tumors to PALA by reducing the dependence onthe de novo pyrimidine pathway for nucleic acid synthesis.Measurements of the anabolism of pyrimidine nucleosides anddeoxynucleosides by tumor homogenates show no real differences in basal kinase activities between the sensitive andresistant lines. However, following PALA treatment, tumor kinase activities were significantly elevated over the respectivenontreated activities in all 3 lines. Furthermore, this stimulationof pyrimidine salvage by PALA was 25 to 50% greater in the 2resistant variants than in LL/0 and could contribute to insen-sitivity to PALA. It should be recalled, though, that mice bearingthe parent Lewis lung carcinoma, LL/0, can be readily rescuedfrom the effects of PALA by administering uridine (12). Thus,even the PALA-sensitive tumor has a functional salvage path

way. The mechanism through which PALA, as well as pyrimidine analogs such as 6-azauridine (3, 16), induce these kinase

activities is not known. However, this induction of uridine kinaseactivity may contribute to the synergistic inhibition of cellgrowth observed between PALA and 5-fluorouracil (1 ) by enhancing the anabolism of the ribosylated derivative of 5-fluorouracil.

Dynamic assessment of PALA perturbation of pyrimidinebiosynthesis, both in vivo and in vitro, demonstrates that thePALA-resistant variants of the Lewis lung carcinoma maintain,to an important degree, a functional de novo pathway followingPALA treatment. Pyrazofurin-provoked accumulation of orotateand orotidine and NaH'4CO3 incorporation into pyrimidine nu-

cleotides can be blocked by PALA in the parent line, LL/0,while substantial flux through the de novo pathway is stilldemonstrable in the 2 resistant variants. Clearly then, theoperative mechanism(s) of PALA resistance in these Lewis lungcarcinomas serve to diminish inhibition of de novo pyrimidinebiosynthesis by the drug.

The first 3 enzymes of the de novo pyrimidine biosyntheticpathway, CPS II, ATCase, and DHOase, have been copurifiedas a single oligomeric protein in mouse ascites hepatoma cells(23, 24) and in a mutant SV-40-transformed hamster lineselected for resistance to PALA (4). In postmitochondrial su-pernatants of this latter cell line, the ratio of the specificactivities of the 3 enzymes is 1:47:7 (4). As derived from Table3, similar ratios are observed in comparable extracts of the 2PALA-resistant Lewis lung carcinomas (LL/PALA-C, 1:65:5;and LL/PALA-J, 1:68:4). By contrast, less CPS II activity is

found in the LL/0 line, both in absolute terms and relative toATCase activity (1:127:7). The appearance of the LL/PALA-C

line showing elevated CPS II and DHOase activities but unaltered ATCase activity is somewhat of an anomaly in that the 3activities are usually elevated together in PALA-resistant cells(17), as is the case for the LL-PALA-J line. However, in mutant

Chinese hamster ovary cells requiring exogenous pyrimidinesfor growth, several reversants have been isolated that also haveincreased amounts of CPS II and DHOase activities but onlyslightly elevated ATCase activity (7). Davidson and Patterson(7) suggest that these low ATCase revertants produce a defective multienzyme complex that is more liable to proteolyticcleavage into CPS ll:DHOase and ATCase fragments with con

sequent loss of ATCase activity. It is possible that a similarstructural defect exists in the CPS ll-ATCase-DHOase complexof the LL/PALA-C line.

Because PALA is a competitive inhibitor of the utilization ofcarbamyl phosphate by ATCase, an increased production ofcarbamyl phosphate by the elevated CPS II in the resistantvariants could serve to displace PALA from ATCase, reducethe inhibition of this enzyme, and allow pyrimidine biosynthesisto proceed despite therapy. It is pertinent to note that, in accordwith this point, examination of the dose responsiveness ofATCase inhibition by PALA reveals less inhibition per givendose of PALA in both resistant variants. The concordant observation that intratumoral concentrations of PALA are similarlydepressed in the LL/PALA-C and LL/PALA-J lines may reflect

the displacement of PALA from its site of sequestration (ATCase) and subsequent efflux of the drug from tumor cells. Druguptake studies suggest that the alternative explanation, thatthere exist differing rates of PALA uptake in the resistant versussensitive lines, is not operative. It would be appropriate to makean assessment of the pool sizes of carbamyl phosphate in thedifferent Lewis lung carcinoma lines; however, attempts toquantitatively extract this extremely labile intermediary metabolite from tumor tissue have been without success. It might wellbe, though, that cellular carbamyl phosphate levels are notrepresentative of the carbamyl phosphate concentration at theactive site of ATCase due to channeling of this substrate fromthe adjacent CPS II (15).

Mammalian CPS II is subject to allosteric regulation by avariety of nucleotides. UTP is the most potent inhibitory effectorof this enzyme; concentrations between 0.5 and 2 mw UTPcause significant increases in the apparent Km of MgATP forCPS II (28). Measurement of UTP levels in Lewis lung carcinomas shows concentrations of 100 /IM in LL/0 and LL/PALA-C and 200 JIM in LL/PALA-J. The in vitro studies of Tatibanaand Shigesada (28) would suggest that these concentrationsof UTP would have little impact on CPS II reaction velocity.PRPP, by contrast, is a positive effector of CPS II activity. Morief al. (22) have demonstrated that concentrations as low as 50/IM PRPP can maximally enhance enzyme activity by loweringthe apparent Kmof MgATP for CPS II. Similar concentrations ofPRPP are observed in the 2 PALA-resistant tumor lines

whereas a lower PRPP concentration is seen in the sensitiveline. Enhanced CPS II activity in the resistant variants thenmay, in part, be due to the augmented concentrations of thiseffector.

The role of elevated CPS II activities as modulators of PALAresistance is reinforced by studies on the chemotherapeuticpotentiation of PALA by the L-glutamine antagonist Acivicin(NSC 163501) in LL/PALA-C tumors. In the following article

(20), we find that Acivicin, a potent inhibitor of CPS II activity,markedly inhibits orotate and orotidine accumulation, as wellas tumor growth when given in combination with PALA.

In general, ATCase activity will vary with CPS II. Thus,naturally, PALA-resistant tumor lines should have elevated CPSII activities as well as high ATCase activities. Conversely, PALA-sensitive tumors would be expected to have lower CPS II levelsin accord with their lower ATCase levels. The possibility thatthose marginally sensitive tumor lines with very low ATCaseactivities [e.g., ovarian teratocarcinoma (11)] have relativelyhigher CPS II levels is presently under investigation.

MARCH 1981 903

Research. on December 6, 2018. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

r. W. Kensler et al.

REFERENCES

1. Ardalan, B.. Glazer, R. !.. Kensler, T. W., Jayaram, H. N., and MacDonald.J. S. Biochemical mechanism for the synergism of 5-fluorouracil and phos-phonacetyl-L-aspartate in human mammary carcinoma cells. Bull. Cancer

(Paris), in press, 1980.2. Christopherson, R. E., Matsurra, T., and Jones, M. E. Radioassay of dihy-

droorotase utilizing ion-exchange chromatography. Anal. Biochem., 89.225-234, 1978.

3. Cihak, A., Vesety, J.. and Harrap, K. R. Enhancement of rat liver uridinekinase activity by various metabolic inhibitors. Biochem. Pharm., 23:1087-1094, 1974.

4. Golemen, P. F., Suttle. D. P., and Stark, G. R. Purification from hamstercells of the multifunctional protein that initiates de novo synthesis of pyrim-idine nucleotides. J. Biol. Chem. 252. 6379-6385, 1977.

5. Collins. K. D., and Stark, G. R. Aspartate transcarbamylase: interaction withthe transition state analogue f\Mphosphonacetyl)-L-aspartate. J. Biol. Chem.,246. 6599-6605, 1971.

6. Cooney, D. A., Karlowicz, M. G.. Cubillan, J., Roettger, M., and Jayaram, H.N. An enzymatic technique for measuring N-phosphonacetyl-L-aspartic acidin tissues. Cancer Treat. Rep., 62: 1503-1507, 1979.

7. Davidson, J. N., and Patterson, D. Alteration in structure of multifunctionalprotein from Chinese hamster ovary cells defective in pyrimidine biosynthesis. Proc. Nati. Acad. Sei. U. S. A., 76. 1731-1735, 1979.

8. Fink, R. M., Cline, R. E., McGaughey, C., and Fink, K. Chromatography ofpyrimidine reduction products. Anal. Chem., 28: 4-6, 1956.

9. Hoogenraad. N. J. Reaction mechanism of aspartate transcarbamylase frommouse spleen. Arch. Biochem. Biophys., 161: 76-82, 1974.

10. Jayaram, H. N., and Cooney, D. A. Analogs of L-aspartic acid in chemotherapy for cancer. Cancer Treat. Rep., 63. 1095-1108, 1979.

11. Jayaram, H. N., Cooney, D. A., Vistica, D. T., Kariya, S., and Johnson, R. K.Mechanisms of sensitivity or resistance of murine tumors to AMphosphona-cetyl>-L-aspartate (PALA). Cancer Treat. Rep., 63. 1291-1302, 1979.

12. Johnson, R. K. Reversal of toxicity and antitumor activity of AMphosphona-cetyl)-L-aspartate by uridine or carbamyl-DL-aspartate in vivo. Biochem.Pharmacol.. 26. 81-84, 1977.

13. Johnson, R. K., Inouye. T., Goldin, A., and Stark, G. R. Antitumor activity ofN-(phosphonacetyl)-L-aspartic acid, a transition-state inhibitor of aspartatetranscarbamylase. Cancer Res.. 36. 2720-2725, 1976.

14. Johnson, R. K.. Swyryd, E. A., and Stark, G. R. Effects of W-(phosphona-cetyl)-L-aspartate on murine tumors and normal tissues in vivo and in vitro

and the relationship of sensitivity to rate of proliferation and level of aspartatetranscarbamylase. Cancer Res., 38. 371-378, 1978.

15. Jones. M. E. Pyrimidine nucleotide biosynthesis in animals: genes, enzymes,and regulation of UMP biosynthesis. Annu. Rev. Biochem., 49: 253-279.1980.

16. Keefer, R. C., Morris, H. P., and Webb, T. E. 5-Azacytidine-modified patternsof uridine kinase activities in normal and neoplastic tissues. Cancer Res..34: 2260-2265, 1974.

17. Kempe. T. D.. Swyryd, E. A., Bruist, M., and Stark, G. R. Stable mutants of

mammalian cells that overproduce the first three enzymes of pyrimidinenucleotide biosynthesis. Cell, 9: 541-550, 1976.

18. Kensler, T. W., Cooney, D. A., Cysyk, R. L., and Johnson, R. K. Studies onthe mechanism of resistance of Lewis lung carcinoma to AHphosphonacetyl)-L-aspartic acid (PALA). Proc. Am. Assoc. Cancer Res., 21: 258, 1980.

19. Kensler, T. W., Erlichman, C., Jayaram, H. N., Tyagi, A. K., Ardalan, B., andCooney, D. A. Peripheral leucocytes as indicators of the enzymic effects of/vXphosphonacetyl)-L-aspartic acid on human L-aspartate transcarbamylaseactivity. Cancer Treat. Rep., in press, 1980.

20. Kensler, T. W., Reck, L. J., and Cooney, D. A. Therapeutic effects of Acivicinand N-(phosphonacetyl)-L-aspartic acid in a biochemically designed trialagainst a A/-(phosphonacetyl)-L-aspartic acid-resistant variant of the Lewislung carcinoma. Cancer Res., 41: 905-909, 1981.

21. Lui, M. S., Jackson, R. C., and Weber, G. Enzyme pattern-directed chemotherapy: effects of antipyrimidine combinations on the ribonucleotide contentof hepatomas. Biochem. Pharmacol., 28. 1189-1195, 1979.

22. Mori, M., Ishida. H., and Tatibana, M. Aggregation states and catalyticproperties of the multienzyme complex catalyzing the initial steps of pyrimidine biosynthesis in rat liver. Biochemistry, 14: 2622-2630, 1975.

23. Mori, M., and Tatibana, M. A multienzyme complex of carbamoylphosphatesynthetase (glutamine):aspartate carbamoyltransferaserdihydroorotase (ratascites hepatoma cells and rat liver). Methods Enzymol., 51: 111-121,1978.

24. Moyer, J. D., and Handschumacher, R. E. Selective inhibition of pyrimidinesynthesis and depletion of nucleotide pools by N-(phosphonacetyl)-L-aspar-tate. Cancer Res., 39. 3089-3094, 1979.

25. Schnaitmann. C. S., and Greenawalt, J. W. Enzymatic properties of the innerand outer membranes of rat liver mitochondria. J. Cell Biol., 38: 158-175,1968.

26. Smithers. G. W.. Gero, A. M., and O'Sullivan, W. J. A simple radioassay for

dihydroòrotate dehydrogenase. Anal. Biochem., 88. 93-103, 1978.27. Swyryd, E. A., Seaver, S. S., and Stark, G. R. W-(Phosphonacetyl)-L-aspar-

tate, a potent transition state analog inhibitor of aspartate transcarbamylase,blocks proliferation of mammalian cells in culture. J. Biol. Chem., 249:6945-6950, 1974.

28. Tatibana, M., and Shigesada, K. Control of pyrimidine biosynthesis inmammalian tissues. J. Biochem. (Tokyo), 72. 549-560, 1972.

29. Toninello, A., and Almirante, L. Incorporation of the precursors of purineand pyrimidine nucleotides in the nucleic acids of rat liver. Farm. Ed. Sci.,28:888-895. 1973.

30. Tsuboi, K. K., Edmunds, H. N., and Kwong, L. K. Selective inhibition ofpyrimidine biosynthesis and effect on proliferative growth of colonie cancercells. Cancer Res., 37: 3080-3087, 1977.

31. Tyagi, A. K., Jayaram, H. N., Anandaraj, S., Taylor, B., and Cooney, D. A.Use of automated chromatography on the amino acid analyzer with lithiumcitrate buffers to separate nucleic acid bases, nucleosides, nucleotides andtheir precursors. J. Biochem. Biophys. Methods, 1: 221-226, 1979.

32. Yamamoto, C. M.. Lauzon, G. J., and Patterson, R. P. Toxicity of combinations of arabinosylcytosine and 3-deazauridine toward neoplastic cells inculture. Biochem. Pharmacol., 27. 181-186, 1976.

904 CANCER RESEARCH VOL. 41

Research. on December 6, 2018. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

1981;41:894-904. Cancer Res   Thomas W. Kensler, George Mutter, James G. Hankerson, et al.  

-(Phosphonacetyl)-l-aspartic AcidNCarcinoma to Mechanism of Resistance of Variants of the Lewis Lung

  Updated version

  http://cancerres.aacrjournals.org/content/41/3/894

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/41/3/894To request permission to re-use all or part of this article, use this link

Research. on December 6, 2018. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from