effect of methotrexate with 5-methyltetrahydrofolate ... · of the time the blood supply to the...

[CANCER RESEARCH 43, 5210-5216, November 1983]

Effect of Methotrexate with 5-Methyltetrahydrofolate Rescue and

Dietary Homocystine on Survival of Leukemic Mice and onConcentrations of Liver Adenosylamino Acids1

Mary A. Hilton,2 Jerald L. Hoffman, and Margaret K. Sparks

Department of Biochemistry, School of Medicine, University of Louisville, Louisville, Kentucky 40292

ABSTRACT

We have increased significantly the survival time of DBA/2mice bearing methionine-dependent L1210 or L5178Y leukemia

cells by i.p. administration of lethal doses of methotrexate (fivedaily doses of 25 mg/kg body weight) followed by rescue with5-methyl tetrahydrofolate (five daily doses of 20 mg/kg bodyweight). The mice were maintained on a semipurified choline-and cyst(e)ine-free diet containing 0.32% L-methionine. We further increased significantly the survival time of the treated animals bearing L5178Y cells, but not those bearing L1210 cells,by substitution of 0.86% DL-homocystine for the methionine in

the diet.We have examined the effects of both diets in mice treated

with methotrexate and 5-methyl tetrahydrofolate, singly and incombination, on the concentrations of S-adenosylmethionine andS-adenosylhomocysteine in the liver, a tissue highly active in the

metabolism of these amino acids. The substitution of homocys-tine for methionine in the diet of untreated animals led to asignificant increase in S-adenosylhomocysteine and decrease inS-adenosylmethionine in the liver, with a resultant profounddecrease in the ratio of S-adenosylmethionine to S-adenosylhomocysteine which was not further altered significantly by administration of methotrexate.

INTRODUCTION

Observations that certain normal and malignant cells exhibitdifferences in the metabolism of methionine (17, 31, 40) havegenerated interest in the metabolic pathways involving methionine as possible sites of intervention in the chemotherapy ofcancer (28). Chart 1 outlines the metabolism in mammalian cells(13) of methionine, which plays a key role in cell division throughparticipation in the synthesis of polyamines and the initiation ofprotein synthesis. Methionine also figures prominently in themetabolism of THF,3 a cofactor which is essential for the biosyn

thesis of the purines and thymidylate required for the synthesisof DMA and RNA also required for cell division. The ability ofantifolates, such as MTX, to inhibit the rapid division of neoplasticcells has long been recognized (11) and results from inhibitionby this drug of dihydrofolate reducÃasewith subsequent depletion of cellular THF. The chemotherapeutic efficacy of MTX

1This study was supported by Grant CA 24108 from the National Cancer

Institute, NIH, Department of Health, Education, and Welfare.2To whom requests for reprints should be addressed.3The abbreviations used are: THF, tetrahydrofolate; 5-formyl-THF, 5-formyte-

trahydrololate; 10-formyl-THF, 10-formyltetrahydrofolate; methylene-THF, 5,10-methytenetetrahydrofolate; 5-methyl-THF, 5-methyltetrahydrofolate; MTX, methotrexate; AdoMet, S-adenosylmethionine; AdoHcy, S-adenosylhomocysteine; HPLC,high-pressure liquid chromatography.

Received March 7,1983; accepted August 2,1983.

against murine leukemia has been improved by administration oflethal doses of the drug which are tolerated by the host whenfollowed by "rescue" with 5-formyl-THF (citrovorum factor) (16),

a fully reduced form of folate which bypasses the need for theinhibited enzyme. As shown in Chart 1,5-formyl-THF (citrovorum

factor) can be converted to other forms of THF carrying d unitsin the CHi oxidation state, one of which (10-formyl-THF) participates in purine biosynthesis and, after reduction to methylene-THF, in the biosynthesis of thymidylate, thus supporting celldivision in the face of significant concentrations of MTX. Protocols using high-dose MTX:citrovorum factor rescue have been

adapted successfully to the treatment of cancer in humans (29).Examination of Chart 1 reveals the importance of Reaction A

in normal cells in (a) the recycling of homocysteine to methionineand (u) converting the major circulating form of reduced folate,5-methyl-THF (20), to THF which can then participate in the

biosynthesis of the purines and thymidylate essential for celldivision. In the presence of MTX, there is an accumulation ofdihydrofolate, generated by the synthesis of TMP, with a resultant decrease in the concentration of 5-methyl-THF (33), impeding

Reaction A.In the work reported, we have sought to manipulate the

relationships shown in Chart 1 to produce a selective attack onthe growth of the neoplastic cells, L1210 and L5178Y, injectedi.p. into mice. These cells do not grow in vitro when homocysteineis substituted for methionine in the culture medium and thus aresaid to be methionine dependent (7, 17). We fed mice a semi-purified diet in which DL-homocystine was substituted for L-

methionine, injected the tumor cells, and later administered lethaldoses of MTX in conjunction with 5-methyl THF, which others(2, 37) have found to provide effective "rescue" from the anti

folate. Both the homocystine diet and the form of rescue weredesigned to provide adequate nutrition for host cells, which arecapable of meeting the need for methionine through Reaction A,while selectively impairing the growth of tumor cells, which arenot.

We also examined the effects, singly and in combination, ofthe homocystine diet and of MTX and the rescue protocol on themetabolism of the sulfur amino acids as revealed by measurements of the molar ratio of AdoMet to AdoHcy in the liver, whichparticipates actively in the metabolism of these amino acids. Bythis means, we hoped to determine whether or not MTX and thehomocystine diet might produce additive effects on methioninemetabolism, as monitored in the liver.

MATERIALS AND METHODS

Animals and Diets. Male DBA/2 mice were obtained from The Jackson Laboratory, Bar Harbor, Maine. Animals, weighing 25 Â±3 g at thestart of each experiment, were caged individually in wire-bottomed cages,

5210 CANCER RESEARCH VOL. 43

on March 8, 2021. © 1983 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

MTX, Reduced Folate, Homocystine, and Tumors

FOUCACIOUUtmloHFR*

POLMNIKfS

CAMMEI

CH2THF ^CrSTATHIMME AMp-

CTSTEINE

ÃADOHCT

INOSIUE

â€¢MHIIITEDBTNTXChart 1. Metabolism of methionine in mammalian cells. Standard designations

for amino acids and nucleotides are used. In addition, the following abbreviationsare used: S12, a coenzyme derived from vitamin B12; Ado, adenosine; CH-THF,forms of THF carrying C, units at the CH, oxidation state, e.g., 5,10-methenylte-trahydrofolate or 10-formyl-THF; CH2-THF, methylene-THF; CH3-THF, 5-methyl-THF; CF, citrovorum factor (5-formyl-THF); DHF, dihydrofolate; DHFR, dihydrofolatereducÃase;Hey, homocysteine; Met, methionine; Gly. glycine; Ser, serine.

suspended in racks of 20 cages, in an animal room maintained at 24 Â±1Â°on a 6:30 p.m. to 6:30 a.m. dark cycle. The mice were fed ad libitum

an agar-based semipurified diet containing crystalline amino acids as a

source of nitrogen (22). The published composition was modified toprovide carbohydrate in the form of 85% dextrin-15% sucrose by weight.

The basal diet, containing no methionine, cyst(e)ine, or choline, wasformulated by ICN Nutritional Biochemicals. Cleveland, Ohio. t-Methionine or DL-homocystine was mechanically mixed into the dry diet before

suspension in the agar gel. The finished diet was refrigerated or frozenuntil used. A cube of fresh diet, weighing 8 to 9 g, was placed in eachcage daily in late afternoon. It was possible to assess the food intake ofeach animal by inspection of these cubes.

Cell Lines. Murine leukemia lines L1210 and L5178Y were obtainedin ascitic form in female DBA/2 mice from the Mason Research Institute,Worcester, Mass. Cells were maintained by approximately weekly i.p.transplantation into male DBA/2 mice of 10s cells in 0.1 ml of 0.15 M

NaCI. Cells used in experiments were obtained from host animals, killedby cervical dislocation 5 to 7 days after i.p. inoculation with leukemiccells. The ascitic fluid was rinsed from the abdominal cavity with a filter-

sterilized (MF filter; pore size, 0.025 pm; Millipore Corp., Bedford, Mass.)solution containing 72 mw NaCI, 142 rnw o-glucose, and 27 mw trisodium

citrate, adjusted with HCI to pH 6.1. Cells were sedimented in sterilegraduated 12-ml centrifuge tubes at 500 x g in a clinical centrifuge,

resuspended, and washed in the same medium until contaminating RBCwere no longer apparent. Leukemia cells were then resuspended andrinsed twice in heat-sterilized 0.15 M NaCI. A 0.5-ml aliquot of a stock

solution, diluted with 0.15 M NaCI to a cell density appropriate forcounting, was added to a small tube in which 0.1 ml of a 1% alcoholicsolution of Eosin Y (Harieco, Gibbstown, N. J.) had been dried previously.Cells were placed in a hemacytometer chamber, and those which demonstrated viability by exclusion of the dye were counted (18). The stocksolution was diluted to provide the desired number of cells per 0.1 ml of0.15 M NaCI for use in each experiment.

Chemicals and Drugs. MTX for pilot experiments was generouslyprovided by Lederle Laboratories of American Cyanamid Co. and for thereported experiments by the Drug Synthesis and Chemistry Branch,Division of Cancer Treatment, National Cancer Institute, Bethesda, Md.L-Methtonine, DL-homocystine, and d/-5-methyl-THF, sodium salt, were

obtained from Sigma Chemical Co., St. Louis, Mo. The folate wasanalyzed by reverse-phase HPLC and shown to contain less than 0.2%of 5-formyl-THF (citrovorum factor). The HPLC reverse-phase method

used was one published for separation of the aromatic amino acids (21)

in which 20% methanol in 15 mM H3PO4 Ã©lÃ»tes5-methyl-THF at 11 minand 5-formyl-THF at 16 min. Ammonium sulfate, special enzyme grade,

was obtained from Schwarz/Mann, Inc., division of Mediscience, SpringValley, N. Y. All other chemicals were reagent grade and were obtainedfrom Fisher Scientific Co., Pittsburgh, Pa. Solutions of MTX for injectionwere prepared by dissolving MTX in 0.15 M NaCI by gradually adjustingthe pH to 7 with NaOH. Solutions of 5-methyl-THF were prepared in

0.15 M NaCI containing 10 mM sodium ascorbate, pH 6.5 to 7, andbubbled with nitrogen to protect against oxidation of the reduced folate.Each solution used for i.p. injection was filtered through an Acrodisc filter(Gelman Sciences, Inc., Ann Arbor, Mich.; pore size, 0.45 ^m), dispensedin equal volumes into 5 sterile tubes, and kept frozen at -20Â° until the

day of use. Solutions were injected in volumes equivalent to 0.01 ml perg body weight.

Determination of S-Adenosyl Amino Acids. Mice were anesthetized

with ether, the abdominal cavity was opened with minimal blood loss,and a 0.4- to 0.8-g portion of the liver was quickly removed and dropped

into a preweighed centrifuge tube containing 2.5 ml 5% sulfosalicylicacid. Homogenization, using a Polytron homogenizer(Kinematica GmbH,Lucerne, Switzerland) at a setting of "5," was completed within 5 sec

of the time the blood supply to the liver was severed. Homogenateswere weighed to determine the weight of the liver sample, and the tubeswere centrifuged in the cold at 5500 x g for 10 min. Supernatant s werefrozen at -20Â° for later analysis by HPLC.

HPLC analyses were made at ambient temperature on a 10-^m PartisilSCX column, 250- x 4.6-mm internal diameter (Whatman, Inc., Clifton,N. J.), fitted with a guard column (42- x 3.2-mm internal diameter), and

liver metabolites were determined by a modification of the method ofHoffman (25). Equipment consisted of a Beckman Model 332 liquidChromatograph, equipped with an Altex Model 210 injection valve, aHitachi Model 110-10 spectrophotometer, a Beckman Model 155 variable-wavelength detector equipped with a 20-^1 flow cell with a 10-mmlight path, and a Beckman Model B-5000 Omni-Scribe 28-cm strip-chart

recorder, all obtained from Beckman Instruments, Inc., Fullerton, Calif.Elution, at 2 ml/min, utilized Solvent A, 10 mM ammonium formate,adjusted to pH 3.7 with 88% formic acid, and Solvent B, 0.2 M ammoniumsulfate in Solvent A, pH 3.7. Liver supernatant (20 /il) was injected inrunning Solvent A at 0 time. At 5 to 7.5 min, the flow of Solvent B wasincreased linearly from 0 to 12.5%. From 10 to 12.5 min, the flow ofSolvent B was increased linearly from 12.5 to 100%. At 17 min, the flowof Solvent B was decreased linearly to 0% over a period of 0.1 min. Thecolumn was then reequilibrated with a flow of 30 ml of Solvent A beforeaddition of the next sample. Under these conditions, AdoHcy andAdoMet, detected by absorbance at 254 nm, 0.01 A units full scale, areeluted at 11.2 and 16.6 min, respectively. Quantitation was achieved bycomparison of peak heights with those of standard adenosylamino acidsobtained from Sigma Chemical Co. Solvents were prepared in glass-redistilled water and were filtered through a Millipore type HA 0.45-/im

filter before use. Samples were filtered through Gelman Acrodisc filters,0.45-/jm pore size, before analysis.

Statistical Analysis. Data were analyzed for significant differences byanalysis of variance, with post hoc analysis by Duncan's studentized

range test (10). Further statistical treatments are noted in legends toCharts 2 and 3.

RESULTS

The basal diet, which contained no choline or cyst(e)ine, wasdesigned to provide adequate nutrition but also to place a highdemand on added methionine, both as the sole dietary sourceof preformed methyl groups and of cysteine sulfur (Chart 1). Thedry diet contained 5 mg of pyridoxine-HCI per 100 g, or 10 timesthe usual supplement, so that any complex formation betweenhomocysteine and pyridoxal phosphate would not produce adeficiency of vitamin B6(38).

NOVEMBER 1983 5211



M. A. Hilton et al.

so-,

TREATED

3 UNTREATED80

MET HCYP

Chart 2. Mean survival time of teukemic animals as a function of dietary sulfuramino acid and of treatment with MTX and 5-methyl-THF. In Experiments A to D,animals received 0.1 ml 0.15 M NaCI containing leukemia cells at 5 to 7 p.m. onDay 0 as follows: Experiment/*, 4 x 103 L1210 cells; Experimente, 4 x 103 L1210cells; Experiment C, 103 L1210 cells; and Experiment D, 105 L5178Y cells. Treated

animals were given injections i.p. of MTX (25 mg/kg body weight) on 5 successivedays at 4 to 5 p.m., beginning on Day 1, in a volume of 0.01 ml/g body weight;they were similarly given injections of 5-methyl-THF (20 mg/kg body weight) at 11a.m. to 12 noon on 5 successive days beginning on Day 1 (Experiment A) or Day2 (Experiments B to D). Untreated animals were given injections of equivalentvolumes of 0.15 M NaCI when treated animals received MTX and with 0.15 M NaCIcontaining 10 mm ascorbate, pH 6.5 to 7.0, when treated animals received 5-methyl-THF The basal diet was supplemented with either 0.32% L-methionme(Met) or 0.86% DL-homocystine (Hey). Number at the bottom of each column,number of animals tested. Columns with asterisks, survival times of treated animalswithin a single experiment which differ significantly (p < 0.001), when comparedby f test for groups of unequal size. Bars, S.E.

We showed earlier that the basal diet, supplemented with0.32% L-methionine, provided for adequate growth with no microscopic evidence of fatty liver in animals maintained on thediet for 52 days (22). We then tested diets in which the methio-nine was replaced by levels of DL-homocystinerepresenting 1.5and 3 times the sulfur content of the methionine diet. In animalsinitially weighing 25 to 26 g, diets containing either level of DL-homocystine in place of methionine, fed for 28 days, produced

Chart 3. Groups of 5 to 6 animals on each diet received the indicated treatment:MTX and MTX^-methyl-THF groups received i.p. injections of 25 mg MTX per kgbody weight at 4 p.m. for 5 consecutive days, beginning on Day 1; 0.15 M NaCIand W*-methyl THF groups received i.p. injections of 0.15 M NaCI at the same time./V-Methyl THF and MTXJV-methyl THF groups received i.p. injections of 20 mg5-methyi-THF per kg body weight at 11 a.m. for 5 consecutive days beginning onDay 1; 0.15 M NaCI and MTX groups received i.p. injections of 0.15 M NaCIcontaining 10 mw ascorbate, pH 6.5 to 7, at the same time. Animals were sacrificedon Day 6, and livers were analyzed by HPLC as described in "Materials andMethods." Each injection was given in a volume equivalent to 0.01 ml per g body

weight. Animals were allowed to adjust to the basal gel diet, supplemented with0.32% i -methionine. for 7 days. One-half of the animals were then switched to thebasal gel diet supplemented with 0.43% OL-homocystine for 2 days before treatments were begun. Asterisks above columns, significant differences where p <0.05 (â€¢)or p < 0.01 (") when animals on the 2 diets within one treatment group

are compared. Significant differences between ratios of AdoMet to AdoHcy weredetermined after data were normalized by logarithmic transformation (36). Met,methionine; Hey, homocystine; Sai, 0.15 M NaCI (saline); NMTF, Af-methyt-THF.Bars. S.E.

JMet DM*

Hcij Diet

SdÃ¬ NMTF MTX MTXNMTF





mean gains in body weight of 11.5 Â±7.1% (S.E.) and 11.9 Â±4.9%, respectively, as compared to 21.5 Â±4.7% for the dietcontaining methionine. Older animals, initially weighing 32 to 36g, were shown on the average to maintain these weights whendiets containing either level of homocystine were fed for 90 days;when animals in this weight range were fed the methionine dietfor 3 months, body weights increased an average of 29.8 Â±8.2%. Thus, although the choline-free diet containing homocys

tine did not support continued increases in body weight of olderanimals as did the methionine diet, both levels of homocystinetested did permit long-term maintenance of body weight, withno toxicity evident in 3-month feeding trials.

In experiments testing the survival of animals hosting asciticleukemia cells, the animals were fed the methionine diet for 4 to5 days to permit adjustment to the agar gel. When all animalshad accepted the gel diet, designated animals were fed thehomocystine diet for 1 to 2 days before leukemia cells wereinjected. We have shown previously that, within 2 days, dietaryhomocystine induces near-maximal changes in metabolism of S-

adenosylamino acids in the liver (23).The lethality of doses of 5 to 25 mg of MTX per kg of body

weight was tested on animals fed 0.32% L-methionine or 0.43%or 0.86% DL-homocystine diets. All animals receiving 5 daily i.p.

injections of 7.5 mg of MTX or more per kg of body weight diedafter mean survival times ranging from 6.3 to 8.2 days. Meansurvival time was similar among animals on methionine andhomocystine diets and was little influenced by the dose of thedrug within the range tested. The average loss of body weightfrom the day of the first injection of drug to the time of deathwas 26%, again with little variation induced by the diet or by thedose of the drug. All control animals receiving injections ofequivalent volumes of 0.15 M NaCI survived.

Rescue of 10 mice from the lethality of 5 daily injections of 25mg of MTX per kg of body weight was effected by 5 dailyinjections of 20 mg of d/-5-methyl-THF per kg of body weight;one-half of this dosage of the reduced folate rescued only 9 of10 animals tested. Rescue by the higher dose of 5-methyl-THF

was achieved whether the reduced folate was given 5 hr before(2), or delayed until 19 hr after, each injection of MTX; theaverage maximum weight losses before recovery with these 2modes of rescue were 13 and 17%, respectively. White ef al.(39) have shown that both the natural (/)and unnatural (d) isomersof 5-methyl-THF are substrates for the high-affinity THF cofactor

membrane transport carrier in Ehrlich ascites tumor cells andthat both isomers stimulate a net efflux of MTX from these cellswhen they have been preloaded with the antifolate.

In preliminary experiments, we established that the meansurvival time of mice given injections i.p. of L1210 cells did notchange significantly when the number of cells injected variedfrom 103 to 2 x 105 cells in 0.1 ml of 0.15 M NaCI, or when

standard laboratory chow was fed in place of the semipurifieddiets. The effects of diets supplemented with methionine orhomocystine, and of treatment with MTX and 5-methyl-THF, on

the survival of mice bearing the ascitic form of L1210 or L5178Ycells are shown in Chart 2. In every experiment, where eachanimal received the same inoculum of cells, the mean survivaltime was prolonged significantly by treatment with 5 daily i.p.injections of MTX and of 5-methyl-THF, as compared with thesurvival of tumor-bearing animals receiving 0.15 M NaCI injec

tions. We first followed the procedure of Blair and Searle (2),

who had shown that 5-methyl-THF could protect non-tumor-

bearing FT hybrid mice from the toxicity of 5 daily injections ofMTX when the reduced folate was administered 5 hr before eachinjection of MTX. This procedure was followed in Chart 2,Experiment A. We then found that the mean survival time oftumor-bearing animals could be further increased significantly (p

< 0.01) by delaying the rescue with reduced folate until 19 hrafter each injection of MTX, as can be seen by comparingExperiments A and B of Chart 2. The greatest increase in survivaltime is seen in Experiment C where treated, homocystine-fed

animals lived 108% longer than did their untreated counterparts.Examination of Experiments A through C of Chart 2 reveals thatthe substitution of dietary homocystine for methionine leads tono significant differences in the mean survival time of animalshosting L1210 cells, when the effects of the 2 diets are comparedin either treated or untreated groups of animals. In Chart 2,Experiment D, in which animals were given injections of L5178Ycells, the methionine and homocystine diets are seen to have nodifferential effect on the mean survival time of untreated animals;however, among treated animals, the homocystine diet producedan 80% increase in mean survival time which was significantlygreater (p < 0.001) than the mean survival time of treatedanimals receiving the methionine diet.

We wished to explore whether or not the above combinationof chemotherapy and the homocystine diet could be shown toaffect the metabolism of sulfur amino acids and chose to measurethe concentrations of AdoMet and AdoHcy in the liver, which isthe organ containing the highest combined specific activity ofenzymes catalyzing the formation of the former and hydrolysisof the latter amino acid. Accordingly, we treated groups ofanimals, fed methionine or homocystine diets, for 5 days withMTX and 5-methyl-THF, as outlined in Experiment A of Chart 2,

and sacrificed the animals beginning at 10 a.m. on the sixth day.The livers were removed and analyzed as described in "Materialsand Methods." The results of these experiments, along with

appropriate controls, are depicted in Chart 3. Liver AdoHcy andAdoMet concentrations and the molar ratios of AdoMet toAdoHcy are compared between groups of animals on the methionine and homocystine diets receiving the same treatments, withsignificant differences indicated in Chart 3.

Additionally, it is possible to compare the effects of the 4treatments shown in Chart 3 on the concentrations of liveradenosylamino acids in animals on either diet. In Chart 3/4,among methionine-fed animals, the liver AdoMet concentration

in the group treated with MTX is significantly lower than that ofanimals treated with both MTX and 5-methyl THF (p < 0.01) orof 0.15 M NaCI-treated controls (p < 0.05). The concentration ofliver AdoMet in methionine-fed animals receiving both MTX and

reduced folate was significantly higher (p < 0.01) than that of allgroups on the homocystine diet. The concentration of liverAdoMet in the methionine-fed 0.15 M NaCI control group wassignificantly higher (p < 0.01) than that of all animals on thehomocystine diet except those treated with MTX only, wherethe differences were less significant (p < 0.05). The concentration of liver AdoMet in homocystine-fed animals was not signifi

cantly different when comparisons were made among the 4treatment groups. Treatment of homocystine-fed animals withboth MTX and 5-methyl THF led to liver AdoMet concentrationswhich were not different from those of methionine-fed animals

treated with MTX alone but were significantly lower than those

NOVEMBER 1983 5213



M. A. Hilton et al.

of methionine-fed reduced folate-treated control animals (p <0.05) or of methionine-fed 0.15 M NaCI-treated control animals

or those treated with both MIX and reduced folate (p < 0.01).In Chart 3B, there are no significant differences in liver con

centrations of AdoHcy in homocystine-fed animals when treated

groups are compared with 0.15 M NaCI controls. However, theconcentration of AdoHcy in the livers of homocystine-fed animalsreceiving both MIX and 5-methyl-THF was significantly lower (p

< 0.01) than that in animals receiving either MIX or reducedfolate alone. In animals on the methionine diet, the AdoHcyconcentration in the livers of groups treated with MIX wassignificantly higher (p < 0.05) than that of either the 0.15 M NaCI-or reduced folate-treated control groups; treatment of methionine-fed animals with both MIX and reduced folate did not lower

the liver AdoHcy concentration significantly from that of thosetreated with MIX only. Additionally, Chart 3B reveals that thehomocystine diet fed to the 0.15 M NaCI control group, ortreatment of methionine-fed animals with MIX, elevates liver

AdoHcy to about the same extent (differences not significant).Treatment of homocystine-fed animals with MTX elevates liver

AdoHcy still further so that, although it is not increased significantly over that of homocystine-fed 0.15 M NaCI- or reducedfolate-treated control animals, it is significantly higher (p < 0.01)than that of all methionine-fed animals, including those treatedwith MTX. Reduced folate-treated animals on the homocystine

diet have liver AdoHcy concentrations which are significantlyhigher (p < 0.01) than those of homocystine-fed animals treatedwith both MTX and reduced folate, as well as those of all groupson the methionine diet.

In Chart 3C, the liver AdoMet:AdoHcy ratio is significantlyhigher in methionine-fed animals receiving 0.15 M NaCI, reducedfolate, or both MTX and reduced folate than in all homocystine-fed animals. These differences are all highly significant (p < 0.01)except for that between homocystine- and methionine-fed ani

mals receiving both MTX and the reduced folate, where the pvalue is < 0.05, as indicated in Chart 3. Treatment of methionine-fed animals with MTX alters the liver AdoMet:AdoHcy ratio tosignificantly (p < 0.01) lower than that of methionine-fed animalstreated with 0.15 M NaCI or reduced folate and above that ofreduced folate-treated animals on the homocystine diet (p <0.01). Treatment of methionine-fed animals with both MTX and

reduced folate leaves the liver AdoMet:AdoHcy ratio still significantly below (p < 0.05) that of 0.15 M NaCI controls, although itis significantly higher (p < 0.01) than that of methionine-fedanimals treated with MTX only. In homocystine-fed animals,treatment with both MTX and 5-methyl-THF leads to an increase

in the liver AdoMet:AdoHcy ratio over that seen in groups treatedwith reduced folate or MTX alone (p < 0.01).

In each of the 8 groups of animals receiving the 2 diets andthe 4 treatments, one animal was not sacrificed but was continued on the appropriate diet to monitor the response to the 4treatments. All animals survived except those treated with MTXonly; these died on the seventh day after the first injection of thedrug with losses in body weight of 29 and 32.5% for mice onthe homocystine and methionine diets, respectively. Animalstreated with both MTX and 5-methyl-THF survived after a maxi

mal loss in body weight of 11.0% on Day 5 and 11.5% on Day6 for homocystine- and methionine-fed animals, respectively.Reduced folate- and 0.15 M NaCI treated control animals on

either diet showed only minor fluctuations in body weight duringthe course of the daily injections.

DISCUSSION

Malignant and transformed cells characteristically showrepression (4) or inadequate activity (26) of endogenous metabolic pathways leading to the biosynthesis of nutrients whichcan be supplied by the host. Examples of various approaches tomanipulation of these metabolic changes were discussed in aseries of reports on amino acid imbalance in the treatment ofcancer, summarized by Muggia ef al. (35). Simply removing anamino acid from the diet of the host animal is generally ineffectivein reducing the supply reaching malignant cells in vivo if thenormal tissues of the host are capable of synthesizing the aminoacid. The complement of enzymes present in the liver endowsthis organ with the ability to control plasma concentrations ofmost of the dietary amino acids including the sulfur amino acids(12).

The work reported here describes a combined nutritional andchemotherapeutic approach to depriving methionine-dependent

tumor cells of methionine. Substitution of homocystine for alldietary methionine leaves the normal animal dependent upon 2known routes of methionine regeneration from homocysteine.One of these involves the transfer of a methyl group from betaineto homocysteine (Chart 10), catalyzed by betaine:homocysteinemethyltransferase (EC 2.1.1.5), which has been shown to havesignificant activity in liver only among 11 rat tissues tested (12).We have attempted in these studies to minimize the contributionof this enzyme to the synthesis of methionine from homocysteineby eliminating choline, the metabolic precursor of betaine, fromthe semipurified diet used.

A second route for the conversion of homocysteine to methionine (Chart 1A) involves the vitamin B12-dependent enzyme [5-methyl-THF:L-homocysteine methyltransferase (EC 2.1.1.13)],

which permits endogenous conversion of homocysteine to methionine. This enzyme is widely distributed in mammalian tissues(12), with the small intestine the only tissue found to lack activityof this enzyme. The 2 murine tumor cell lines we have used areincapable of growing in culture medium in which DL-homocy-steine replaces t-methionine (7, 31). This suggests deficient

utilization of both homocysteine in the biosynthesis of methionineand of 5-methyl-THF in meeting cellular needs for THF. The use

of MTX in conjunction with a diet in which homocystine issubstituted for methionine should block the formation in mosttissues of both methionine and THF. Administration of supplements of 5-methyl-THF should rescue only those cells capable

of converting this compound to THF as methionine is formedfrom homocysteine.

The present paper confirms the utility of 5-methyl THF in the

rescue of mice from the toxicity of MTX, while producing highlysignificant increases in the mean survival times of animals receiving i.p. injections of either leukemic cell line over thosereceiving no chemotherapy. This finding extends the pioneeringwork of Goldin er al. (15) in the use of 5-formyl-THF (citrovorum

factor) in overcoming drug toxicity and in extending the survivalof mice inoculated with L1210 cells and treated with the antifolate, aminopterin. Mead ef al. (34) used 5-methyl-THF to reduce

the toxicity of MTX given to leukemic (L1210) hybrid F, mice;these workers reported a 27% increase in the median survivaltime of leukemic animals treated with this form of rescue. Smithef al. (37) found 5-methyl-THF superior to 5-formyl-THF (citro

vorum factor) in overcoming the toxicity of high doses of MTXadministered 7 times, with Cytoxan administered 6 times. Both





drugs were given at 3- to 10-day intervals to C57BL x DBA/2

Fi mice given injections s.c. of L1210 cells. These workersreported that rescue with 5-methyl-THF, administered twice as

frequently as MTX, produced a 328% increase in mean survivaltime over that of untreated tumor-bearing animals. Deaths in the

group receiving chemotherapy were attributed to drug toxicityas well as to tumors.

Borsa ef al. (3) have shown that the persistence in mouseserum of MTX injected i.p. is biphasic, with a half-life of 31 minover the first 4 hr and of 12 hr thereafter. Since 5-methyl-THF is

transported into L1210 cells by an active transport system whichalso transports MTX (19), we have separated by 5 hr the i.p.injections of these compounds on a single day so as to limit theactivity of the reduced folate in competitively inhibiting the penetration of MTX into tumor cells or stimulating efflux of MTXfrom preloaded cells.

The higher (0.86%) of the 2 concentrations of dietary DL-

homocystine treated has been used in the experiments outlinedin Chart 2 in an attempt to exploit the reported effect of homo-

cysteine in retarding the growth of L5178Y cells in culture (7).The higher concentration of dietary homocystine should alsointerfere more effectively with the transport of methionine intotumor cells. Methionine, which normally occurs in low concentrations in the plasma of mice (1), is an excellent substrate for boththe A and L systems for transport of neutral amino acids (9).Homocysteine also interacts with both of these systems and ismore effective than is L-methionine in inhibiting the transport ofL-histidine by both the A and L systems of S37 ascites tumor

cells (32). Homocysteine, with its sulfhydryl group, should alsobe efficiently transported by the ASC system, for which methionine is not a substrate (9).

It is apparent that, in animals hosting L1210 cells, the homocystine diet in conjunction with chemotherapy with MTX and 5-methyl-THF does not increase the mean survival time over thatof methionine-fed animals receiving the same chemotherapy. Inuntreated animals inoculated with 10s L5178Y cells, the mean

survival time of animals on either diet is longer than that inuntreated animals inoculated with 103 L1210 cells (Chart 2).

Although Halpem ef al. (17) have shown L5178Y cells grown inculture to be more sensitive to methionine deprivation than areL1210 cells, we see no change in mean survival time of animalsbearing L5178Y cells when homocystine is substituted for methionine in the diet; presumably, the host tissues provide adequate methionine. However, the combination of the homocystinediet and chemotherapy does produce a significant increase inmean survival time of these animals over those on the methioninediet. This finding provides an interesting puzzle in view of theobservations that L5178Y cells grown in vitro are capable ofutilizing 5-methyl-THF, when it is supplied in the culture medium(7), and that certain methionine-dependent rat cells are capableof converting DL-homocysteine thiolactone to methionine and

grow well in cultures containing homocysteine when low levelsof methionine are also present (26). Chello and Berlino (8) havereported that preincubation of L5178Y cells in methionine-free

medium interferes with MTX cell kill in vitro.We have observed that treated animals hosting either L1210

or L5178Y cells characteristically eat little for 3 to 4 days beforethey die. Thus, any effect of the diet on survival must occurbefore that time. We have found previously that the livers ofanimals on the choline-free homocystine diet accumulate large

amounts of lipid, compared to the normal amount of lipid found

in the livers of animals fed the basal diet containing methionine(23). The data in Chart 2 indicate that these stores of lipid in theliver play no significant role in extending the survival of anorexicanimals bearing either L1210 or L5178Y cells, as the homocystine diet produced no increase in survival of untreated tumor-bearing animals over those on the methionine diet.

Hryniuk (27) has studied the mechanism of action of MTX incultured L5178Y cells and has concluded that interference withpurine synthesis contributes initially to cell kill but that, ultimately,the cells die predominantly a thymineless death. As a third action,MTX should interfere with the generation of methyl groups denovo via the THF-mediated Ci pool, blocking the recycling of

homocysteine to methionine (Chart 1). Because of the uniquerole of methionine in its activated form (AdoMet) as the primarybiological methylating agent, we sought to examine the effect ofMTX on the metabolism of this amino acid in the liver, which ishighly active in methyltransferase reactions. Cantoni (5) haspostulated that the intracellular ratio of AdoMet to AdoHcy playsa critical role in regulating biological methylation reactions, andhe and his Â«workers (6) have studied the relative sensitivity ofa number of methyltransferases to changes in this ratio. Sincemethyltransferases catalyze the methylation of proteins, nucleicacids, lipids, and a variety of small molecules such as neurotrans-

mitters, a disturbance in the ratio of AdoMet to AdoHcy inducedby MTX might be expected to affect many aspects of cellularfunction. The data in Chart 3 reveal that administration of MTXdoes lower the ratio of hepatic AdoMet to AdoHcy significantlyin methionine-fed animals. However, 0.15 M NaCI-treated animalson a diet containing 0.43% OL-homocystine experience the same

decrease in this ratio, and administration of MTX to these animalshas no further effect on the concentrations of liver adenosylaminoacids. Since mice tolerate the diet containing homocystine at0.86% quite well for as long as 3 months, and since the homocystine diet alone has no effect on the mean survival time ofleukemic mice, we suggest that the lowering of the hepatic ratioof AdoMet to AdoHcy produced by MTX is not physiologicallysignificant. These findings also demonstrate that the effects ofMTX and of dietary homocystine on the ratio of AdoMet toAdoHcy in the liver are not additive.

The effects of MTX and dietary homocystine on concentrationsof S-adenosylamino acids in other tissues and in tumor cells In

vivo remain to be examined. Isolation of tumor cells from theascitic fluid will demand special care if the in vivo state is to berevealed. Hoffman et al. (24) have shown that levels of AdoHcyin the liver increase markedly within sec of the death of theanimal, possibly due to rapid accumulation of adenosine.

The significant increase in the AdoMet:AdoHcy ratio in thelivers of homocystine-fed animals treated with both MTX and 5-methyl-THF over those receiving either the drug or reduced folate

alone results from a significant decrease in liver AdoHcy in theformer group of animals. The equilibrium of the reaction for thehydrolysis of AdoHcy favors condensation of homocysteine andadenosine; thus, the hepatic concentrations of both of thesecompounds should affect the amount of AdoHcy in the liver.MTX, by inhibiting purine synthesis, would be expected to decrease liver adenosine; the administration of 5-methyl-THF

should favor the conversion of homocysteine to methionine viaReaction A (Chart 1). The combination of MTX and the reducedfolate could reduce the hepatic concentrations of both adenosineand homocysteine with a resultant decrease in AdoHcy.

The present work extends our understanding of the effects of

NOVEMBER 1983 5215



M. A. Hilton et al.

MIX on hepatic methionine metabolism in vivo and demonstratesthe reversal of MTX-induced changes in the concentrations of S-adenosylamino acids by "rescue" with 5-methyl-THF. We have

also demonstrated the utility of combined nutritional and chem-

otherapeutic regimens, both affecting a metabolic pathway whichfunctions differently in normal and in malignant cells, in increasingthe survival of mice bearing L5178Y cells. Nutritional manipulation may be more successful in larger animals, including humans,where continuous infusion of amino acid mixtures is feasible.This nutritional and chemotherapeutic approach to selectivestarvation of cancer cells may benefit from augmentation withthe catabolic enzyme, methionase (EC 4.4.1.11), which has beenused by Kreis (30) to inhibit the growth of certain malignant cellsin culture. Also, the use of vincristine and probenecid may behelpful in prolonging the retention by tumor cells of MTX inpolyglutamate forms (14).

ACKNOWLEDGMENTS

The excellent technical assistanceof George H. Raque,Jr., Khris J. Basham,J.Meal Sharpe, and Martene C. Steifen is gratefully acknowledged. We also thankDr. J. K. Hilton for assistance in rapid dissection of mouse livers and for drawingCharts 2 and 3.

REFERENCES

1. Battistm. L. Grynbaum. A., and Lajtha, A. The uptake of various amino acidsby the mouse brain in vivo. Brain Res., 29: 85-99, 1971.

2. Blair, J. A., and Scarte, C. E. Reversal of methotrexate toxicity in mice by 5-methyltetrahydrofolic acid. Br. J. Cancer, 24: 603-609,1970.

3. Borsa, J., Whttmore, G. F., Vateriote, F. A., Collins, D., and Bruce, W. R.Studies on the persistence of methotrexate, cytosine arabmoside,and leuco-vorin in serum of mice. J. Nati. Cancer Inst., 42; 235-242,1969.

4. Broome. J. D., and Schwartz, J. H. Differences in the production of L-asparagine in asparaginase-sensitiveand resistant lymphoma cells. Biochim.Btophys. Acta, 138: 637-639,1967.

5. Cantoni, G. L. S-Adenosylamino acids thirty years later: 1951-1981. In: E.Usdin, R. T. Borchardt, and C. R. Creveling (eds.), Biochemistry of S-Adeno-sylmethionine and Related Compounds, pp. 3-10. London: MacmillanPress,Ltd., 1982.

6. Cantoni, G. L., Richards. H. H., and Chiang, P. K. Inhibitorsof S-adenosylhom-ocysteine hydrolase and their role in the regulation of biological methylation.In: E. Usdin,R. T. Borchardt, and C. R. Creveling(eds.),Transmethylation.pp.154-164. Amsterdam: Elsevier/North-HollandBiomÃ©dicalPress, 1979.

7. Cheiio, P. L., and Bertino, J. R. Dependence of 5-methyltetrahydrofolateutilization by L5178Y murine leukemia cells in vitro on the presence ofhydroxycobalammand transcobalammII. Cancer Res., 33:1898-1904,1973.

8. Chello, P. L., and Bertino, J. R. Effect of methionine deprivation on L5178Ymurine leukemia cells in culture. Interferencewith the antineoplasticeffect ofmethotrexate. Biochem. Pharmacol.,25: 889-892,1976.

9. Chnstensen. H. N. Implications of the cellular transport step for amino acidmetabolism. Nutr. Rev., 35: 129-133,1977.

10. Duncan, D. B. Multiple range and multiple F tests. Biometrics, 11:1-42,1955.11. Farber, S. Some observations on the effect of folk: acid antagonists on acute

leukemia and other forms of incurablecancer. Blood, 4:160-167,1949.12. Finkelstem. J. D. Regulation of methionine metabolism in mammals. In. E.

Usdin, R. T. Borchardt, and C. R. Creveling(eds.), Transmethylation,pp. 49-58. Amsterdam: Elsevier/North-HollandBiomÃ©dicalPress, 1979.

13. Finkelstein, J. D., Kyle, W. E., and Harris, B. J. Methionine metabolism inmammals. Regulation of homocysteine methyltransferasesin rat tissue. Arch.Biochem. Btophys., 146: 84-92,1971.

14. Fry, D. W., Yatowich,J. C., and Goldman,I. D. Augmentationof the intracellularlevelsof poiygiutamyiderivativesof methotrexate by vincristineandprobenecidin Ehrlich asertestumor cells. Cancer Res., 42:2532-2536,1982.

15. Goldin, A., Mantel, N., Greenhouse,S. W., VenditÂ«,J. M., and Humphreys, S.R. Effect of delayed administration of citrovorum factor on the antileukerniceffectiveness of aminopterin in mice. Cancer Res., 14: 43-48,1954.

16. Goldin, A., Venduti, J. M., Kline, I., and Mantel, N. Eradication of leukaemiccells (L1210)by methotrexate and methotrexate plus citrovorum factor. Nature(Lend.), 272:1548-1550, 1966.

17. Halpem, B. C., Clark, B. R., Hardy, D. N., Halpem, R. M., and Smith, R. A.The effect of replacement of methionine by homocysteine on survival ofmalignant and normal adult mammaliancells in culture. Proc. Nati. Acad. Sci.U. S. A., 71:1133-1136,1974.

18. Hanks, J. H., and Wallace,J. H. Determinationof cell viability. Proc. See. Exp.Biol. Med., 98: 188-192,1958.

19. Henderson,G. B., Grzelakowska-Sztabert, B., Zevely,E. M., and Huennekens,F. M. Binding properties of the 5-methyltetrahydrofolate/methotrexate transport system in L1210 cells. Arch. Biochem. Biophys., 202: 144-149,1980.

20. Herbert, V., Larrabee,A. R., and Buchanan,J. M. Studies on the identificationof a folate compound of human serum. J. Clin. Invest., 41:1134-1138,1962.

21. Hilton, M. A. Ã¼quid-chromatographicdirect determinationof phenylalanineandtyrosine in serum or plasma,with application to patients with phenylketonuria.Clin. Chem., 28:1215-1218,1982.

22. Hilton, M. A., and Raque, G. H., Jr. An amino acid diet supporting superiorgrowth in mice. J. Nutr., 110:2409-2413,1980.

23. Hilton, M. A., and Sparks, M. K. Effect of dietary OL-homocystineon lipid andS-adenosylaminoacids in mouse liver. Fed. Proc., 41:1168,1982.

24. Hoffman. D. R., Comatzer, W. E., and Guerre. J. A. Relationship betweentissue levels of S-adenosylmethionme.S-adenosylhomocysteine,and trans-methylation reactions. Can. J. Biochem.,57: 56-65,1979.

25. Hoffman, J. L. A rapid liquid Chromatographiedetermination of S-adenosyl-methionmeand S-adenosylhomocysteinein subgram amounts of tissue. Anal.Biochem., 68: 522-530,1975.

26. Hoffman, R. M., and Erbe, R. W. High in vivo rates of methioninebiosynthesisin transformed humanand malignant rat cells auxotrophic for methionine.Proc.Nati. Acad. Sei. U. S. A., 73:1523-1527,1976.

27. Hryniuk, W. M. The mechanismof action of methotrexate in cultured L5178Yleukemiacells. Cancer Res., 35: 1085-1092,1975.

28. Huennekens.F. M., DeGirolamo,P. M., Fujii, K., Jacobsen, D. W., and Vitols,K. S. B,2-dependentmethionine synthetase as a potential target for cancerchemotherapy. Adv. Enzyme Regul., 14:187-205,1976.

29. Jaffe, N.. Paed, D., Frei, E., Ill, Traggis. D., and Watts, H. Weekly high-dosemethotrexate-citrovorum factor in osteogenic sarcoma. Cancer (Phila.), 39:45-50,1977.

30. Kreis, W. Tumor therapy by deprivation of u-methionine:rationale and results.Cancer Treat. Rep.. 63:1069-1072,1979.

31. Kreis. W., and Goodenow, M. Methionine requirement and replacement byhomocysteine in tissue cultures of selected rodent and human malignant andnormal cells. Cancer Res., 38: 2259-2262,1978.

32. Lewis, N. J., inioes. R. L., Hes, J.. Matthews. R. H., and Milo, G. Syntheticsulfur-containing amino acids. Inhibition of transport systems in S37 ascitestumor cells. J. Med. Chem..21: 1070-1072,1978.

33. Matthews, R. G., and Daubner,S. C. Modulationof methylenetetrahydrofolatereducÃaseactivity by S-adenosylmethionmeand by dihydrofolate and its polyglutamate analogues.Adv. Enzyme Regul., 20:123-131,1982.

34. Mead, A. R., Venduti. J. M., Schrecker, A. W., Goldin, A., and Keresztesy, J.C. The effect of reduced derivatives of folie acid on toxicity and antileukerniceffect of methotrexate in mice. Biochem. Pharmacol., 12: 371-383,1963.

35. Muggia, F. M., Catane, R., Bouros, J., and Cooney, D. Amino acid manipulations in cancer treatment: a perspective from the National Cancer Instituteworkshop. Cancer Treat. Rep., 63:1137-1138,1979.

36. Myers, J. L. Fundamentalsof Experimental Design, p. 77. Boston: Allyn andBacon. Inc., 1972.

37. Smith, R. A., Halpem, R. M., and Halpern, B. C. A/6-Formyltetrahydrofolateversus N^-methyltetrahydrofolate rescue in vivo and in vitro. ChemioterapiaOncologica,2 (Suppf.4V 109-116,1978.

38. Tunnidiff, G., and Ngo, T. T. The mode of action of homocysteine on mousebrain glutamic decarboxylase and gamma-aminobutyrate aminotransferase.Can. J. Biochem., 55:1013-1018.1977.

39. White, J. C., Bailey, B. D., and Goldman, I. D. Lack of stereospecificity atcarbon 6 of methyltetrahydrofolate transport in Ehrlich ascites tumor cells. J.Biol. Chem., 253: 242-245,1978.

40. Wilson, M. J., and Poirier, L. A. An increased requirement for methionine bytransformed rat liver epithelial cells in vitro. Exp. Cell Res., 111: 397-400.1978.




1983;43:5210-5216. Cancer Res Mary A. Hilton, Jerald L. Hoffman and Margaret K. Sparks Concentrations of Liver Adenosylamino AcidsDietary Homocystine on Survival of Leukemic Mice and on Effect of Methotrexate with 5-Methyltetrahydrofolate Rescue and

Updated version

http://cancerres.aacrjournals.org/content/43/11/5210

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/43/11/5210To request permission to re-use all or part of this article, use this link



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

mailto:[email protected]



effect of methotrexate with 5-methyltetrahydrofolate ... · of the time the blood supply to the...

Documents