selective inhibition of pyrimidine biosynthesis and effect...

[CANCER RESEARCH 37, 3088-3087, September 1977)

Selective Inhibition of Pyrimidine Biosynthesis and Effect onProliferative Growth of Colonie Cancer Cells1

Kenneth K. Tsuboi, Henry N. Edmunds,2 and Linda K. Kwong

Department ot Pediatrics, Section of Developmental Medicine, Stanford University, Stanford, California 94305

SUMMARY

A highly selective inhibition of de novo pyrimidine synthesis in the intact cell has been demonstrated by the action ofA/-(phosphonacetyl)-L-aspartate (PALA), a transition-stateanalog inhibitor of the reaction catalyzed by asparate trans-carbamylase. The effect of pyrimidine deprivation inducedby this agent on the viability and survival of human normal(WI-38) and colonie cancer cells (HT-29) was examined. ThePALA-treated, pyrimidine-deprived cells failed to grow butdemonstrated a normal rate of glucose utilization with impaired glycogen synthesis. Pyrimidine deprivation and lackof cell growth were maintained long after PALA removal.Growth inhibition of HT-29 cells by PALA was found toreflect an apparent steady state between newly formed anddying cells induced by limited pyrimidine availability. Thehighly selective nature of PALA action was confirmed by theability of an exogenous source of pyrimidine to restore thenormal growth pattern of the cell. Significant antitumoractivity by PALA was found against a transplantable colonietumor (line 26) carried in mice.

INTRODUCTION

Potent and selective inhibition of mammalian aspartatetranscarbamylase by an analog of the activated complex ofthe reaction catalyzed by this enzyme has been reported (5,15). The analog, PALA,3 has been shown to inhibit prolifera-

tive growth of mammalian cells (15, 19) and to exhibit considerable antitumor activity against certain transplantabletumors in mice (9).

In the present study, inhibition of de novo pyrimidinesynthesis by PALA has been examined in relation to thegrowth and viability of normal (WI-38) and colonie cancer(HT-29) cells of human origin. PALA was found to be ahighly selective and intense inhibitor, resulting in littlemeasurable de novo synthesis of pyrimidine both duringand after treatment of the cells with this agent. Short- andlong-term effects of cell pyrimidine deficiency induced bythis agent are described in this report. Also included in thisreport are the results of in vivo studies demonstrating sig-

1This work was supported by USPHS Grants CA 14917, administeredthrough the National Large Bowel Cancer Project, HD 000391, and HD 02147.Part of this work has been presented at the annual meeting of the Federationof American Societies for Experimental Biology, April 1 to 8, 1977, Chicago,Illinois.

2 USPHS Postdoctoral Fellow, HD 00049.3The abbreviation used is: PALA, N-(phosphonacetyl)-L-aspartate.Received April 7, 1977; accepted June 2, 1977.

nificant antitumor activity by PALA against a transplantablecolonie tumor (line 26) in mice.

MATERIALS AND METHODS

Materials. [14C]Carbamylphosphate (7.4 mCi/mmole),Ba'4CO,, (60 mCi/mmole), and [mefny/-3H]thymidine (86

mCi/mmole) were obtained from New England Nuclear,Boston, Mass. Na214CO,.twas prepared from Ba14CO:,asdescribed previously (7), and [MC]carbamylphosphate waspurified as described by Adair and Jones (1). Purified carba-mylphosphate was stored in small aliquots at -70Â°. Cyto-sine arabinoside, hexokinase, glucose-6-phosphate de-hydrogenase (Sigma Chemical Co., St. Louis, Mo.), 5-fluo-rouracil, and dilithium carbamylphosphate (Calbiochem,LaJolla, Calif.) were obtained from the indicated sources.PALA and [3H]PALA were prepared as described by Swyryd

et al. (15).Cell Lines and Culture. HT-29, a human colonie epithelial

cancer cell line (3), was obtained from Dr. J. Fogh. WI-38, ahuman diploid fibroblast cell line originating from normalfetal lung, was obtained from Dr. L. Hayflick. The cells weresubcultured and grown in Eagle's minimal essential me

dium supplemented with gentamicin at 50 /^g/ml, vitamins,and amino acids at twice the normal concentrations anddialyzed or nondialyzed fetal bovine serum at 5% for HT-29and 10% for WI-38 cells. Cells were subcultured twiceweekly and also were maintained as frozen stocks. Cellswere grown for experimental use in plastic T-75 flasks, 60-and 35-mm Petri dishes, and cluster dishes containing 24wells of 15-mm diameter. Growth medium was changedevery 3 days unless specified otherwise.

Tumor Transplant. Line 26, a transplantable colonie tumor (undifferentiated grade IV carcinoma) induced inBALB/c mice by 1,2-dimethylhydrazine, was obtained fromDr. T. H. Corbett (for tumor details see Ref. 2). The tumorwas maintained by serial passage at 2- to 3-week intervals inBALB/c mice and also as frozen stock. The mice were fed astandard laboratory chow (Berkeley diet), except a syntheticpyrimidine-free diet (63% sucrose, 18% casein, 5% vegetable oil, 5% fiber, vitamins, and balanced salts) was substituted during the period of experimental PALA administration. Mice were inoculated s.c. with 0.1 ml of a 10% finelyminced suspension of the tumor in Hanks' medium.

Incorporation of ["CJBicarbonate into Pyrimidines. Cellsgrown in T-75 culture flasks were incubated in the stoppered flask for specified periods in Hanks' medium modifiedto contain 10 rriM A/-tris(hydroxymethyl)methyl-2-amino-

3080 CANCER RESEARCH VOL. 37

on March 29, 2020. © 1977 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

Inhibition of Pyrimidine Biosynthesis

ethanesulfonic acid buffer, pH 7.4, and 1.0 rnw sodium[14C]carbonate (50 /xCi). After incubation the cells werewashed with cold Hanks' medium, scraped, transferred,

and centrifuged in capped 1.5-ml plastic tubes. The cells

were air dried after extraction with 1.0 ml absolute ethanolcontaining 10 mivi potassium acetate. To the dried cellswere added 200 /Â¿Iof 70% perchloric acid followed by 1 hrheating at 100Â°to effect release of free pyrimidines and

purines from acid-soluble nucleotides and nucleic acids.The procedure was adapted from an original application forquantitative base isolations from nucleic acids (18) in whichcomplete hydrolysis of bases has been shown to beachieved with minimal deamidation of cytosine to uracilresulting. After addition of 200 Â¿Â¿IH2O to the hydrolysate,the total sample (less 10-/U.Ialiquot for protein) was appliedin 50-/nl aliquots to a 9.5- x 22.5-inch sheet of Whatman No.

3 MM paper and chromatographed (placed in Chromatocabimmediately) for 24 hr in isopropyl alcohol/HCI/H2O (65/17/18). The solvent resolves cleanly guanine, adenine, cytosine, uracil, and thymine in ascending order from theirmixture (18). After elution (0.1 N HCI) amounts of each basewere determined by UV absorption and purified further byadsorption and elution from charcoal (17) and rechromato-graphed for 24 hr in H^O-saturated butyl alcohol containing

excess NH4HCO.t (17). Radiochemical purity was establishedby coincidence of radioactivity (determined by radioautog-

raphy) and UV absorption of the final isolated pyrimidine(and purine) deposits (17) as illustrated in Fig. 1.

Cell Glycogen Determinations. Cells were washed free ofmedium and transferred, after scraping in 1.0 ml of 0.5 Mperchloric acid in 70% ethanol, to 1.5-ml capped plastic

centrifuge tubes. After centrifugation, the packed cells wereextracted with 1.0 ml absolute ethanol containing 10 mwpotassium acetate and air dried. To the dried cells wereadded 150 /Â¿Iof 0.6 N HCI, and the capped tubes wereheated 2 hr at 100Â°.After neutralization with NaOH, aliquots

were removed for protein and glucose determination.Miscellaneous Methods. Aspartate transcarbamylase as

says were as described by Swyryd ef al. (15). Glucose wasdetermined by enzyme-coupled stoichiometric reduction of

NADP, which was measured fluorimetrically (11). Cell extracts prepared by 20-min hydrolysis at 75Â°in 0.5 M per

chloric acid were measured for DMA (4) and RNA (14) withmodification of the published procedures to reduce finalcolor volumes to 1 ml. Application of these methods to cellcultures has been described previously (16). Protein wasdetermined by the method of Lowry ef al. (12). Cell growthrate was determined by protein and DNA accumulation andin certain cases by absorbance at 260 nm of a solubilizedsolution of the cells prepared by sonic extraction in 1.0 MNaOH (15).

RESULTS

Inhibition of Pyrimidine Synthesis. In Chart 1 are showninhibition of aspartate transcarbamylase, a component enzyme of the pyrimidine-biosynthetic pathway, of HT-29 (human colonie cancer), and of WI-38 (human diploid fibro-

blasts) cells by PALA, an analog of the activated complex ofthe enzyme-catalyzed reaction (15). Inhibition by PALA was

1.5

i.o

0.5

HT-29 WI-38

IO 20 30 40 50 IO 20 30 40 50I/[C P], mM-l

Chart 1. Inhibition of aspartate transcarbamylase of HT-29 and WI-38 cellsby PALA. Sonic extracts of cells prepared in 20 mMTris-chloride, pH 8.5, plus1 rr>M EDTA were centrifuged for 10 min at 10,000 x g. The supernatantfractions were adjusted to contain 1 to 2 mg protein per ml and were useddirectly as enzyme source. Reaction mixtures contained, in final 0.5-ml volume, 50 n\ enzyme solution, 0.05 M Tris-chloride (pH 8.5), 0.04 M sodiumaspartate, and [14C]carbamylphosphate (1 ^Ci/Mmole) at designated concentrations and PALA at: 0.0 M (Curve J), 2 x 10"' M (Curve 2), and 4 x 10'" M(Curve 3). Incubations were for 10 min at 37Â°.Velocity was computed in

nmoles CP converted per minute per mg protein. Approximate kinetic constants were: HT-29 aspartate transcarbamylase, Km = 1.4 x 10~5 M, Vâ€ž,â€ž=3.7, K, for PALA = 1.1 x 10'' M; WI-38 aspartate transcarbamylase, Km = 1.0x 10-= M, Vmax= 2.3, K, for PALA = 0.85 x 10 â€¢M.

competitive with carbamylphosphate (5) and equally intenseon the aspartate transcarbamylases of both HT-29 and WI-

38 cells. Kinetic constants derived from the plotted datayielded very similar apparent Km and KÂ¡values for bothenzymes, with Km of 1.0 to 1.4 x 10~5 M for carbamylphosphate and KÂ¡of 0.85 to 1.1 x 10~8 M for PALA. Relative

aspartate transcarbamylase activities of HT-29 and WI-38cells were Vmax = 3.7 and 2.3 nmoles carbamylphosphate

conversion per min mg protein, respectively.The effect of PALA, as a selective and intense inhibitor of

aspartate transcarbamylase, on de novo pyrimidine nucleo-tide synthesis in intact HT-29 cells is shown in Table 1. Inthese experiments, de novo synthesis of pyrimidine wasdetermined by measuring incorporation of "CCX into total

cell pyrimidines. Isolation of the free pyrimidine bases cytosine, thymine, and uracil followed their hydrolytic releasefrom total cellular nucleotide and nucleic acids treated as acommon pool. Marked inhibition of pyrimidine synthesiswas induced in HT-29 cells by PALA at 1 x 10 3 M. Inhibitionwas apparent within 2 hr and near maximal after 4 hr. '"CCs

incorporation into cell pyrimidines was 2 to 4% and 60% ofcontrol values after maintaining cells for 24 hr in the presence of 1 x 10~3 and 1 x 10 4 PALA, respectively. Inhibition

of pyrimidine synthesis in HT-29 cells was also demonstra

ble by uridine alone (6) or in the combined presence ofPALA. Inhibition by uridine was evident within 2 hr at 1 x10~5 M and near maximal at 1 x 10 3 M. Labeled CO, was

incorporated only into uracil after 30 min incubation (seeExperiment 2) with further distribution of the label intocytosine and thymine evident only on extending the incubation period with labeled CO, to 2 hr (see Experiment 1).

Sustained Inhibition of Pyrimidine Synthesis. Inhibitionof pyrimidine biosynthesis in HT-29 cells induced in the

presence of PALA was found to be maintained even afterremoval of the inhibiting agent from the culture medium as

SEPTEMBER 1977 3081



K. K. Tsuboi et al.

shown in the experiment summarized in Table 2. HT-29 cellsmaintained in the presence of 1 HIM PALA or 24 hr (andlonger) showed little incorporation of '"CO., into pyrimidine.

Inhibition of pyrimidine synthesis by PALA resulted in anaccompanying inhibition of de novo purine synthesis asevidenced by a lack of '"CO, incorporation into cell purines

Table 1Inhibition of pyrimidine synthesis in HT-29 cells by PALA and

uridineHT-29 cells in T-75 flasks were incubated for the indicated pe

riods with PALA and/or uridine at specified concentrations addedto the growth medium. "CO;, incorporation into pyrimidines wassubsequently determined as described in "Materials and Methods."

['4C]Bicarbonate incorporation(cpm/jumole)"

Additions(M)Experiment1*

ControlPALA, 1 x 10-1 (24)rPALA, 1 x IO"3 (24)PALA, 1 x 10~3 + uridine,

1 x irr3 (24)Uridine, 1 x 10~3 (24)Cytosine5,500

2,000360200Thymine8,300

2,200250970Uracil77,000

47,0003,0001,4003,200

Experiment 2''

ControlPALA, 1 x IO'3 (2)PALA, 1 x 1Q-3 (4)PALA, 1 x 10-3(24)Uridine, 1 x 10~5 (2)Uridine, 1 x 10-4 (2)Uridine, 1 x 10~3 (2)

36,0006,1002,300

80011,5006,9003,200

" Pyrimidines were isolated in each case from 5 to 7 mg cell

protein. Cell pyrimidine contents in /Â¿moles per mg protein were:cytosine, 0.10 Â±0.02; uracil, 0.06 Â±0.02; and thymine, 0.08 Â±0.02.Specific activities were determined in each case by analysis of topand bottom halves of the final isolated deposits with the averagevalues recorded.

* 14CO2incorporation into pyrimidines was determined after 2-hrincubation in modified Hanks' medium containing 1 mw NaH14CO3

(50 /Â¿Ci)and PALA and/or uridine as indicated.r Numbers in parentheses, time (hr).rf 14C02 incorporation into pyrimidines was determined as in

Experiment 1, except incubation time was limited to 30 min.

as shown. Restoration of pyrimidine (and purine) synthesisdid not occur subsequent to removal of the inhibiting agent,with little incorporation of 14CO2 into cell pyrimidines and

purines evident on testing over an extended period (1 to 6days) after the removal of PALA. Total cell purine and pyrimidine levels (estimated per mg protein) were not markedlyreduced over the sustained inhibition period examined.

Lack of incorporation of 14C02 into cell purines was con

sidered to be a secondary consequence of inhibition ofpyrimidine synthesis by PALA. This conclusion was supported by the finding of little inhibition of 14C02 incorpora

tion into cell purines by PALA when administered to cells inthe presence of uridine (not shown).

Inhibition of Pyrimidine Synthesis and Effect on Prolifer-ative Growth of HT-29 and WI-38 Cells. The effect of PALA,

added to growth medium at concentrations inducing inhibition of pyrimidine synthesis (as measured by labeled CO,incorporation) on proliferative growth of HT-29 and WI-38

cells is shown in Chart 2. In these experiments, PALA wasadded to growing cell cultures and maintained over a 6-day

period at 0.1 and 1.0 mM concentrations. The cells weremaintained an additional 6 days in culture in the absence ofPALA. Proliferative growth was measured in these experiments on the basis of protein, DMA, and RNA accumulation,with only the protein values shown (other measurementsshowed similar results). Marked inhibition of growth of bothHT-29 and WI-38 cells was induced with little accumulationof protein resulting during the 6-day culture period in the

presence of PALA. Growth inhibition induced by PALA wasfound to persist even after the removal of PALA from thegrowth medium, which is consistent with the demonstratedinability of PALA-treated cells to resume de novo pyrimidine

synthesis subsequent to removal of the inhibiting agent (seeTable 2). In the presence of either 1 x 10~4 or 1 x 10~3 M

PALA, little increase in cell protein occurred after severaldays in culture, with the existing low levels of protein maintained during the subsequent culture period in the absenceof PALA. Cultures treated with the higher PALA concentration contained only slightly less cell protein as shown.

Metabolic Effects of Pyrimidine Deprivation. The effectof PALA-induced inhibition of pyrimidine synthesis on cell

Table 2Sustained inhibition of pyrimidine (and purine) synthesis in PALA-treated HT-29 cells

HT-29 cells in T-75 flasks were maintained over the indicated periods in the presence of1 mM PALA and after removal of PALA from the culture medium. The ability of the cells toincorporate 14CO2 into pyrimidines (and purines) was subsequently determined as described in "Materials and Methods."

OnPALA(days)01111247Off

PALA(days)0246444Cell

protein(mg)5.68.39.78.97.99.09.79.0[14C]Bicarbonate

incorporationGuanine1

1,000<500<500<500<500<500<500<500Adenine14,800<500<500<500<500<500<500<500Cytosine<500Â»<500<500<500<500<500<500<500(cpm/itmole)aThymine<500<500<500<500<500<500<500<500Uracil34,000<500<500<500<500<500<500<500

" 14CO2 incorporation into pyrimidines and purines was determined after 30-min incubation in modified Hanks' medium containing 1 mw NaH'4CO3 (50 tiCi).

6 Specific activity values of<500 were not significant.





IO 12 24

DAYS IN CULTURE

8 IO 12

Chart 2. Growth inhibition of HT-29 and WI-38 cells by PALA. Retri disheswere seeded with 200,000 HT-29 (35 mm) and WI-38 (60 mm) cells. After 24hr, growth was determined at specified subsequent periods by proteinaccumulation. Curve J, control cells; Curve 2. cells maintained for 6 dayswith PALA at 0.1 mM and then 6 days minus PALA, Curve 3, cells maintained 6 days with PALA at 1.0 mM and then 6 days minus PALA. All pointsrepresent the average of duplicate cultures.

oLUN

jj LJCO I-

8Â§3 O.

^

Lu

O

i

1.5

I.O

0.5

HT-29

Xâ€”X CONTROL

â€¢â€”â€¢+ PALA

WI-38

Xâ€”X CONTROL

â€¢ â€¢+ PALA

30 60 90 30 60 90

INCUBATION TIME (min at 37Â°

Chart 3. Lack of effect of pyrimidine deprivation induced by PALA onglucose utilization by HT-29 and WI-38 cells. Cells were grown in 35-mm (HT-29) and 60-mm (WI-38) Petri dishes. PALA was added at 1 mM for 24 hr to one-half of each set of Petri dishes, before testing for glucose utilization. PALA-treated and control cells were incubated with 1 mM glucose in modifiedHanks' medium, and disappearance of glucose was determined (see "Materials and Methods") after 30, 60, and 90 min incubation. Protein content of thePetri dishes was 0.46 Â±0.08 mg for HT-29 cells and 1.3 Â±0.1 mg for WI-38cells.

viability was examined on the basis of glucose utilizationrates. In Chart 3 are shown the relative rates of glucoseutilization by PALA-treated and control HT-29 and WI-38

cells. Little impairment of glucose utilization was evident ineither of the cells as a consequence of inhibition of pyrimidine (and purine) synthesis by PALA as shown. The HT-29

neoplastic cells were found to exhibit a higher rate of glucose utilization than the normal diploid WI-38 cells.

The metabolic effects of PALA-induced pyrimidine depri

vation were examined further in relation to glycogen synthesis. Participation of uridine nucleotides in glycogen synthesis involves UDP as carrier and UDP-glucose as the

ultimate glucosyl donor to the amylose chain. Since thenormal WI-38 cell and the neoplastic HT-29 cell were found

to engage in an active glycogen synthesis, both were testedto determine the resultant effect of PALA-induced pyrimi

dine deprivation on their ability to synthesize glycogen. InChart 4 are shown relative rates of glycogen restoration inPALA-treated and control cells that had previously been

maintained in the absence of glucose to reduce existingglycogen stores. In contrast to the control cells, little accumulation of glycogen resulted on incubation of the PALA-treated WI-38 and HT-29 cells in the presence of excess

glucose. Impairment of glycogen metabolism as a consequence of cellular pyrimidine deficiency was evident fromthese studies. Glycogen synthesis in the control cells wasfound to be more rapid in the WI-38 as compared to the HT-

29 cells.DMA Synthesis in the Pyrimidine-deprived Cells. PALA-

treated cells were examined in relation to DNA synthesis.During and subsequent to PALA treatment, little net accumulation of cell protein was found to occur (see Chart 2),indicative of complete growth inhibition. On the other hand,the possibility that some proliferative growth was occurringwhich was balanced by a PALA-induced, enhanced rate of

cellular destruction (due to synthesis of defective cells?)could not be excluded. To examine this possibility the DNAof HT-29 and WI-38 cells was labeled with [3H]-

zLUOo

I. CONTROL HT-29

2. PYRIMIDINE DEPRIVED (PALA)

30 60 90 30 60 90INCUBATION TIME (mm at 37")

Chart 4. The effect of pyrimidine deprivation (PALA-induced) on the abilityof WI-38 and HT-29 cells to restore their glycogen pools after glucose starvation. Cells were grown in 60-mm (WI-38) and 35-mm (HT-29) Petri dishes.PALA was added at 1.0 mM for 48 hr to 4 of each set of 8 Petri dishes. PALA-treated and control cultures were maintained for an additional 16 hr inglucose-free Eagle's minimal essential medium containing dialyzed (glucose-

free) fetal calf serum to deplete existing glycogen stores. The cells weresubsequently incubated in Hanks' medium containing excess glucose (5mM), and glycogen accumulation was measured as described in "Materialsand Methods."

thymidine, and the cells were subsequently cultured in thepresence and absence of PALA. Dilution of the label (decrease in specific activity) was followed relative to the DNAcontent of the cells during the subsequent culture periods.The results of a representative experiment obtained withHT-29 cells are shown in Chart 5. Generally similar findingswere obtained with WI-38 cells and are not shown. Duringthe 11-day period examined, the control cultures increasedtheir DNA content 7-fold, whereas the PALA-treated cul

tures showed little accumulation of DNA. The specific activity (cpm/Vg) of the DNA in the control cultures decreasedabout 7-fold (from a high of 37 to 5), correlating well with

the measured accumulation of DNA. The specific activity ofthe DNA in PALA-treated cells decreased 4.5-fold (from a

high of 34 to 7.5), with little associated accumulation ofDNA occurring. These results would imply that considerable

SEPTEMBER 1977 3083



K. K. Tsuboi et al.

8 IO 246

DAYS IN CULTURE

8 IO

Chart 5. Dilution of ['HJthymidine-labeled DMA (A) and correspondingDNA levels (B) in HT-29 cells grown in the presence and absence of PALA.Cells were seeded at 500,000 in 2 sets of 60-mm Petri dishes. After an initial24-hr growth period, all cells were incubated for 2 hr in growth medium(minimal essential medium) containing 1 /UM[3H]thymidine (1 ^Ci/nmole).Following the labeling procedure. PALA was added to 1 mm to 1 set ofcells over a 6-day period and then was removed. The PALA-treated andcontrol sets of cells were analyzed at the indicated times for DNA levelsand radioactivity (see "Materials and Methods" for details).

DNA synthesis, i.e., proliferative growth, was occurring inthe PALA-treated cells (60% of the control rate), balancedby equivalent cellular destruction. On this basis, the PALA-treated cells would appear to represent a steady-state popu

lation in which new cells are being continually formed under conditions of pyrimidine deprivation, leading to a reduced survival ability. New cell production would be supported by a limited source of pyrimidine, available throughsalvage mechanisms from the dying cells with the balanceobtained by leakage through the inhibited de novo pathway.

Growth Inhibition by PALA, Prevention and Reversal byUridine. Growth inhibition of HT-29 cells by PALA could beprevented by supplying the cells with the preformed pyrimi-

dines, cytidine and uridine, in confirmation of selectiveinhibition of de novo pyrimidine synthesis by PALA as thebasis of its antiproliferative action. Cells cultured for 6 daysin the presence of both PALA and uridine (each at 0.1 HIM)and then for 6 days in their absence showed little impairedgrowth compared to control cultures as illustrated in Chart6 (compare Curves 1 and 2). On reducing uridine concentrations, growth inhibition by PALA was increasingly evident(not shown). Cells cultured with PALA alone for 6 days andthen for 6 days in its absence, showed little growth (seeChart 6, Curve 4). The sustained growth inhibition demonstrated by the PALA-treated cells even after removal of the

inhibitor from the culture medium was examined further. Inthis regard, it was shown previously (Table 2) that PALA-

treated cells did not resume de novo pyrimidine synthesison subsequent culture in the absence of the inhibitor. Inhibition of pyrimidine synthesis by PALA, as a competitiveinhibitor of aspartate transcarbamylase (see Chart 1), wouldbe expected to be freely reversible on removal of PALA fromthe culture medium. The inability to restore pyrimidine synthesis after PALA removal was considered, therefore, torepresent a further secondary impairment of the pathwayresulting as a consequence of PALA action. Whether thecontinued impairment of pyrimidine synthesis was the solecause of the inability of the PALA-treated cells to resume

2 4 6 8 IO I2

DAYS IN CULTURE

Chart 6. Prevention and reversal by uridine of growth inhibition of HT-29cells induced by PALA. Cells were plated at 200.000 in 35-mm Petri dishes,and growth was determined by protein accumulation after 6 and 12 days ofculture. Curve Ã•,control cells; Curve 2, cells maintained for 6 days with PALAand uridine at 0.1 mM each and then for 6 days minus PALA and uridine;Curves 3a, 30. and 3c, cells maintained for 6 days with PALA at 0.1 mM andthen for 6 days minus PALA but with uridine added at 0.1 mM (3a). 0.01 mM(3b), and 0.005 mM (3c); Curve 4, cells maintained for 6 days with PALA at 0.1mM and then for 6 days minus PALA. All points represent the average ofduplicate cultures.

growth was tested. In this case, cells cultured initially in thepresence of PALA were supplied with pyrimidine (uridine)subsequent to the removal of PALA (see Chart 6, Curves 3a,b, and c). The sustained growth inhibition induced by PALAwas found to be reversed by providing the cells with preformed pyrimidine. Uridine supplied at 0.1 mM restoredgrowth to nearly control rates (compare slopes in Chart 6,Curves 1 and 3a) in the PALA-treated cells. It was concluded

from these findings that pyrimidine deprivation was the solecause of the growth inhibition of HT-29 cells observed both

during and after PALA treatment.Growth Inhibition by PALA and Dose Schedules. Effec

tive PALA dose schedules, with respect to duration andfrequency, which resulted in complete inhibition of growthof HT-29 cells were as follows. Cells maintained in thepresence of 1 mivi PALA for at least 2 consecutive daysshowed little growth thereafter within a 26-day culture period examined. Single 24-hr dose schedules repeated at

intervals of 3, 5, 7, and 9 days were also effective in preventing growth. The short-term effects of PALA showed little

distinguishable gross morphological effects on cell populations; however, with prolonged culture, the cells appearedto flatten and spread and contained increasing numbers ofvacuoles (both HT-29 and WI-38). Of particular interest was

the appearance of small colonies of apparent mutant cellsin the growth-inhibited cultures which were maintained for

prolonged periods. Selction of these mutant strains andtheir further study are in progress.

Effect of Pyrimidine Analogs on Growth of HT-29 Cells.Comparative growth inhibition patterns of HT-29 cells cul

tured in the presence of selected pyrimidine analogs areshown in Chart 7. Little growth inhibition was induced by 6-

azauridine when tested at concentrations up to 0.1 mM.Cells cultured in the presence of cytosine arabinoside and5-fluorouracil failed to grow at minimal concentrations of

each at 1 and 10 /U.M,respectively. Complete growth inhibition induced in the presence of 1 /U.Mcytosine arabinosidewas not sustained on removal of the inhibiting agent from





the culture medium (Chart 7, Curve 3). On the other hand,growth inhibition induced by 5-fluorouracil at 10 U.Mcontin

ued even after its removal from the culture medium, similarto the inhibition pattern obtained with PALA (Chart 7; compare Curves 4 and 5).

Antitumor Activity of PALA in Mice with Colonie TumorTransplants. The antitumor activity of PALA was tested invivo. In these studies, mice bearing a transplantable colonietumor were treated with PALA to determine its effect as anantitumor agent. The results of 3 preliminary experimentsare summarized in Table 3. The tumor used in these studieswas highly metastatic and resulted in relatively short meansurvival times of the host animals as shown by the 3 sets ofcontrols. In Experiment 3, a longer mean survival time of thecontrol group of animals was found relative to the initial 2

I8 2I

DAYS INCULTURE

Chart 7. Growth inhibition characteristics of HT-29 cells by various pyrimi-dine analogs. Cells were plated at 200,000 in 35-mm Petri dishes. After aninitial 24-hr growth period, cells were cultured for 6 days in the presence andan additional 15 days in the absence of the designated analog: Curve 1,control cells; Curve 2, 6-azauridine at 0.1 ITIM,Curve 3, cytosine arabinosideat 1.0 fiM; Curve 4, 5-fluorouracil at 0.01 mw; Curve 5, PALA at 0.1 mw.Growth was measured by accumulation of cell protein.

experiments, presumably due to the use of older and largeranimals. Within each experiment, however, the administration of PALA resulted in both delay in growth of the tumortransplant and extension of the mean survival time of thehost animals relative to the corresponding controls. PALAwas administered in these experiments at concentrationsranging from 230 to 360 mg/kg body weight per i.p. injection, with the administration repeated 3 to 4 times at theintervals indicated. Extension of mean survival time byPALA ranged from 47 to 56% in the 3 experiments conducted. Simultaneous administration of cytosine arabinoside or 5-fluorouracil in combination with PALA was also

examined in a preliminary manner. Although the combinations did not result in any remarkable extensions of meansurvival times beyond that obtained with PALA alone, further combination studies with cytosine arabinoside wereindicated (mean survival time extended 75% in 1 experiment). The concentrations of PALA administered in thesestudies were below the toxicity levels by a factor of 2 to 3times. Alternative routes of PALA administration, optimaltreatment schedules, and combination chemotherapy haveyet to be examined systematically. Whether simultaneousadministration of uridine with PALA would reverse the anti-

tumor activity of PALA in vivo was not tested.

DISCUSSION

De novo pyrimidine synthesis was determined in thesestudies by measuring 14C02 incorporation into uracil, cyto

sine, and thymine prepared in high radiochemical purityfrom all cell sources (i.e., nucleotides and nucleic acids)treated as a common pool. The method utilized improves on

Table 3Effect of PALA alone and in combination with cytosine arabinoside or 5-fluorouracil on survival

times of BALB/c mice bearing line 26 colonie tumor

No. ofmiceExperiment

1Â°ControlPALAPALA + cytosine arabi

noside4

44Appearance

of palpable tumor (days Â±S.D.)10

2016.3

Â±.0 Â±.8 Â±0

81.5

.05Increase

ofDelay Mean survival mean sur-time time (days Â± vivai time(%) S.D.) p(%)15

95 2363 26.3

Â±.8 Â±.8 Â±2.6

6.2 <0.05"4.8 <0.0155.7 75.4

Experiment 2CControl 4 9.0 Â±2.8 16.3 Â±4.7PALA 3 11.0 Â±3.5 22 26.0 Â±8.9 <0.10 60.0PALA + cytosine arabi- 3 14.0 Â±1.0 56 24.7 Â±2.3 <0.05 51.8

nosidePALA + 5-fluorouracil 3 12.0 Â±2.7 33 22.7 Â±4.7 >0.10 39.5

Experiment 3''Control 6 8.9 Â±1.6 22.8 Â±7.3PALA 6 12.2 Â±0.4 38 33.5 Â±10.6 <0.10 46.7" PALA was given i.p. either alone at 360 mg/kg or combined with cytosine arabinoside at 17.4

mg/kg of Days 1, 4, and 7, after tumor cell implants.6 Significance value on Student's f test. Mean survival times relative to controls.r PALA was given i.p. alone at 230 mg/kg or at 343 mg/kg with cytosine arabinoside at 24 mg/kg

or with 5-fluorouracil at 13 mg/kg on Days 1,4,7, and 14, after tumor cell implants." PALA was given i.p. at 343 mg/kg on Days 1, 4, 7, and 10, after tumor cell implants.

SEPTEMBER 1977 3085



K. K. Tsuboi et al.

that reported by others (e.g., see Ref. 8) in which pyrimidinesynthesis was estimated by measuring '"CO, incorporationonly into the acid-soluble uridine nucleotides, which represents a transient pool of varying size and turnover. Themethod also offers the advantage of allowing simultaneousdetermination of de novo purine synthesis, by measuring14COoincorporation into adenine and guanine which are

also isolated (see Fig. 1).Pyrimidine synthesis was found to be blocked in cells

treated with PALA, which acts as a competitive inhibitor ofaspartate transcarbamylase. Pyrimidine synthesis was notrestored on removal of PALA, indicating either a secondaryderangement of the de novo pathway or an unexplainedprolonged retention of PALA by the cells.

The PALA-treated, pyrimidine-deprived cells could bemaintained as viable cultures for prolonged periods withoutnet change in cell numbers. Such cultures of HT-29 cellseventually developed mutant cell colonies [also found withother cell types (10)]. Metabolic studies showed the PALA-treated cells to be capable of utilizing glucose at normalrates; however, some impairment was found in their abilityto synthesize glycogen, indicative of a deficiency of uridinenucleotide.

The PALA-treated, HT-29 malignant cells, although appearing to have shifted from a normal proliferative state to aquiescent state, continued to synthesize DMA. A steady-state cell population was thus indicated in which new cellswere being produced, of limited survival ability due to pyrimidine deprivation, in equal numbers to cell deaths. Thiswould be consistent with the proposal by Pardee (13) thatmalignant cells demonstrate limited survival ability underadverse conditions (e.g., pyrimidine deficiency) due to lossof restriction point control and stop randomly in their division cycle and die.

That pyrimidine deprivation accounted solely for the sustained growth defect of PALA-treated HT-29 cells was confirmed. By providing an exogenous pyrimidine source (uri-dine), rapid restoration of normal growth patterns was observed with the PALA-treated cells (Chart 6). The utility ofPALA as a highly selective inhibiting agent of cell pyrimidinesynthesis was also confirmed by these studies.

Antitumor activity of PALA was tested in in vivo studiesagainst a transplantable colonie tumor in mice. Formulationof drug schedules was aided by some guidelines providedby the in vitro cell culture studies. Our preliminary testingindicated both delay in tumor growth and extension ofmean survival time of the host animal by PALA, warrantingfurther study of its antitumor action.

ACKNOWLEDGMENTS

The authors wish to acknowledge valuable consultations with Dr. G.Stark, Dr. T. Yoshida. and Dr. N. Kretchmer, and the excellent technicalassistance of M Boswell and E. Chow.

REFERENCES

1. Adair, L. B., and Jones, M. E. Purification and Characteristics of Aspartate Transcarbamylase from Pseudomonas lluorescens. J. Biol. Chem.,247: 2308-2315. 1972.

2. Corbett, T. H., Griswold, D. P., Roberts. B. J., Peckham, J. C., andSchabel, F. M. Tumor Induction Relationships in Development of Transplantable Cancers of the Colon in Mice for Chemotherapy Assays, with aNote of Carcinogenic Structure. Cancer Res., 35. 2434-2439, 1975.

3. Fogh, J., and Trempe, G. New Human Tumor Cell Lines./n: J. Fogh (ed.),Human Tumor Cells in Vitro, pp. 115-159. New York: Plenum Press,1975.

4. Giles, K. W., and Myers, A. An Improved Diphenylamine Method for theEstimation of Deoxyribonucleic Acid. Nature, 206. 93. 1965.

5. Hoogenraad. N.J. Reaction Mechanisms of Aspartate Transcarbamylasefrom Mouse Spleen. Arch. Biochem. Biophys., ÃŽ6ÃŽ:76-82, 1974.

6 Hoogenraad, N. J., and Lee, D. C. Effect of Uridine on de Novo Pyrimidine Biosynthesis in Rat Hepatoma Cells in Culture. J. Biol. Chem., 249.2763-2768. 1974.

7. Hoogenraad, N. J., Levine, R. L., and Kretchmer, N. Copurification ofCarbamylphosphate Synthetase and Aspartate Transcarbamylase fromMouse Spleen. Biochem. Biophys. Res. Commun., 44: 981-988, 1971.

8. Ito. K., and Uchino. H. Control of Pyrimidine Synthesis in Human Lymphocytes. Induction of Glutamine-utilizing Carbamyl Phosphate Synthetase and Operation of Orotic Acid Pathway during Blastogenesis. J. Biol.Chem., 246. 4060-4065. 1971.

9. Johnson, R. K., Inouye, T.. Goloin, A., and Stark, G. R. Antitumor activityof AV-(Phosphonacetyl)-L-aspartic Acid, a Transition-State Inhibitor ofAspartate Transcarbamylase. Cancer Res., 36: 2720-2725, 1976.

10. Kempe, T. D., Swyryd, E. A., Bruist, M.. and Stark, G. R. Stable Mutantsof Mammalian Cells That Overproduce the First Three Enzymes of Pyrimidine Nucleotide Biosynthesis. Cell. 9. 541-551, 1976.

11. Lowry, O. H., Passonneau, J. V. Hasselberger, F. X., and Schultz, D. W.Effect of Ischemia on Known Substrates and Cofactors of the GlycolyticPathway in Brain. J. Biol. Chem., 239: 18-30, 1964.

12. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J. ProteinMeasurement with the Folin Phenol Reagent. J. Biol. Chem., 793: 265-275. 1951.

13. Pardee, A. B. A Restriction Point for Control of Normal Animal CellProliferation. Proc. Nati. Acad. Sei. U. S. A., 71: 1286-1290. 1974.

14. Schneider, W. C. Determination of Nucleic Acids in Tissues by PentoseAnalysis. Methods Enzymol., 3: 448-450, 1957.

15. Swyryd. E. A., Seaver, J. S., and Stark, G. R. 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.

16. Tsuboi, K. K., Mustien. J. B., Seifried, A. S., and Petricciani, J. C.Biochemical Characteristics of Human and Nonhuman Primate DiploidCell Lines. J. Biol. Standardization, 5: 19-30, 1977.

17. Tsuboi, K. K., and Price, T. D. Isolation, Detection and Measure ofMicrogram Quantities of Labeled Tissue Nucleotides. Arch. Biochem.Biophys.. 81: 223-237, 1959.

18. Wyatt, G. R. The Purine and Pyrimidine Composition of DeoxypentoseNucleic Acids. Biochem. J., 48: 584-590, 1951.

19. Yoshida, T., Stark, G. R., and Hoogenraad, N. J. Inhibition by N-(phos-phonacetyl)-L-aspartate of Aspartate Transcarbamylase Activity andDrug-induced Cell Proliferation in Mice. J. Biol. Chem., 249: 6951-6955.1974.





U

CFig. 1. Radioautograph of "C-labeled cell pyrimidines (and purities), after

Chromatographie resolution to high radiochemical purity. Each of the freebases was first isolated from perchloric acid cell digests by chromatographyin isopropyl alcohol/HCI/H2O (65/17/18), purified by adsorption and elutionfrom charcoal, and chromatographed further in water-saturated butyl alcohol (see "Materials and Methods"). The final chromatographed products are

shown in the radioautograph. G, guarirne; -4. adenine; C, cytosine; U, uracil;T, thymine.

SEPTEMBER 1977 3087



1977;37:3080-3087. Cancer Res Kenneth K. Tsuboi, Henry N. Edmunds and Linda K. Kwong Proliferative Growth of Colonic Cancer CellsSelective Inhibition of Pyrimidine Biosynthesis and Effect on

Updated version

http://cancerres.aacrjournals.org/content/37/9/3080

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/37/9/3080To request permission to re-use all or part of this article, use this link



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

mailto:[email protected]



selective inhibition of pyrimidine biosynthesis and effect...

Documents