[CANCER RESEARCH 41, 3621-3628, September 1981 ]0008-5472/81 /0041-OOOOS02.00

Isolation of Tumor Cells from Patients with Osteosarcoma and Analysisof Their Sensitivity to Methotrexate1

JoséMordoh,2 Reinaldo D. Chacón, and Jorge Filmus '

Department of Cancero/ogy, Fundación Centro de Investigaciones Médicas "Albert Einstein, Luis Viale 2831. 1416 Buenos Aires ¡J.M.. J F.], and Department of

Clinical Oncology, PrívateHospital L. Quemes. Av. Córdoba 3933, 1188 Buenos Aires [R. D. C.¡.Argentina

ABSTRACT

Many clinical studies have been conducted on the role ofhigh-dose methotrexate (MTX) in human osteosarcoma, but

information about the in vitro effect of MTX on human osteosarcoma cells is lacking. In this paper, the effect of MTX ontumor cells derived from seven patients with osteosarcoma hasbeen studied in an attempt to correlate clinical and in vitrosensitivity to the drug. Isolation of the cells from the primarytumors (four patients) or metastasis (three patients) was carriedout with a collagenase treatment followed by purificationthrough a density gradient. The osteosarcoma cells were identified by electron microscopy and histochemical reactions. Thecellular sensitivity to MTX was measured by the inhibitory effectof MTX on [3H]deoxyuridine incorporation into DMA. This in

corporation was 50% inhibited in primary tumors at concentrations from 3 x 10~7 to 3 x 10~6 M. The metastatic cells

isolated from patients that were clinically resistant to high-doseMTX had a 50% inhibition ranging from 1.5 x 10~7 to 4 x10~5 M. Human stimulated lymphocytes, Sarcoma 180 cells,

and Ehrlich ascitic mouse cells had a 50% inhibition of about1.5 x 10~7 M. When [3H]thymidine incorporation into DMA of

human osterosarcoma cells was studied, it was observed thatMTX increased its incorporation up to 4-fold. This increasewas stable for at least 6 hr and was only slightly enhanced bythe addition of hypoxanthine. The stimulation by MTX of[3H]thymidine incorporation into DMA in stimulated lymphocytes

and Ehrlich cells is much smaller, between 40 and 60%. Ahypothesis to explain these results is that osteosarcoma cellsbuild their deoxythymidine monophosphate pool largelythrough the de novo pathway, the salvage pathway being lessimportant. It is suggested that the importance of the de novopathway for deoxythymidine monophosphate synthesis is abiochemical characteristic of the osteosarcoma cells whichcould be related to the initial sensitivity of this tumor to MTXand that an activation of the salvage pathway could constitutean additional mechanism of resistance to this drug.

INTRODUCTION

High-dose MTX4 followed by leucovorin rescue has been

used extensively in the treatment of osteosarcoma (4, 13, 19).' This work has been partially supported by Grant 8040/80 from the Consejo

Nacional de Investigaciones Científicas y Técnicas, Argentina, and by Grant70025500-001 of the Secretaría de Estado de Ciencia y Tecnología, Argentina.

1 Research Career Investigator from the former Institution. To whom requests

for reprints should be addressed.1 Fellow from the Comisión de Investigaciones Científicas y Técnicas de la

Provincia de Buenos Aires. Argentina.4 The abbreviations used are: MTX, methotrexate; DHFR, dihydrofolate reduc

íase(EC 1.5.1.3); MEM, minimal essential Eagle's medium; PCS, fetal calf serum;

PHA. phytohemagglutinin; dThd, deoxythymidine; dUrd, deoxyuridine; I50, 50%inhibition.

Received September 12, 1979; accepted June 2, 1981.

This type of treatment appears to be especially effective as acoadjuvant to surgery and only when subclinical métastasesare present. When overt metastatic disease develops, high-dose MTX appears to be less effective since the developmentof clinical resistance is frequent (18). The mechanism of MTXaction has been reviewed recently (1). It is an analog of folieacid which acts essentially through the binding to DHFR, anenzyme which catalyzes the reduction of dihydrofolic acid totetrahydrofolic acid. Goldman (8) has also demonstrated thata certain amount of free intracellular MTX is necessary in orderto achieve a complete inhibition of DNA synthesis. The maineffect of MTX is a cellular depletion of reduced folates, essential for the de novo synthesis of dTMP and purines. Theresulting "thymineless state" or "purineless state" would lead

to cell death, it being at present uncertain which is the predominant mechanism in a given cell type (2, 10, 12). Some authorshave also described a direct inhibitory effect of MTX on thy-midylate synthetase, which catalyzes the synthesis of dTMPfrom dUMP (22).

Several mechanisms have been pointed out to explain thedevelopment of cellular resistance to MTX. The most frequentare: (a) an increase in the levels of DHFR (11, 15, 21); (b) adiminished transport of MTX into the cell (7, 9, 14); (c) analtered DHFR with less affinity for MTX (1 ); (d) the presence ofa large amount of tumor cells in a nonproliferative state (G0)which would make them refractory to the cytocidal effect ofMTX (2).

The rational basis for the utilization of therapeutic regimensincluding high-dose MTX was that the higher drug concentra

tions in blood could overcome cellular resistance. However, toour knowledge, no study has been conducted to analyze MTXresistance at a cellular level in patients with osteosarcoma.

The work reported in this paper was intended to verify if themetastatic osteosarcoma cells isolated from patients clinicallyresistant to MTX were also resistant in vitro to this drug, ascompared with cells isolated from primary tumors. A methoddeveloped to dissociate the osteosarcoma cells is also reported. Some aspects of the origin of the dTMP pool were alsoanalyzed since its changes appear to play a fundamental rolein the lethal effect of MTX.

MATERIALS AND METHODS

Patients. Seven patients were included in this study, and someclinical data relevant to this study are detailed in Chart 1.

Several therapeutic regimens were administered to the patients: (a)LOMA: MTX, 3, 5, and 7 g/sq m on Days 1, 8, and 15 with previousadministration of vincristine (1, 4 mg/sq m) and followed by leucovorinrescue (10 mg/sq m ¡.m.starting 2 hr after MTX infusion; every 6 hrduring 72 hr). Adriamycin (60 mg/sq m) was given on Day 18.

This treatment was administered every 6 weeks. (£>)VIMELECIDA:vincristine (1.4 mg/sq m) on Days 1,8, 15, 22, 29, and 36; MTX, 1.5

SEPTEMBER 1981 3621

Research. on October 31, 2018. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

J. Mordo/1 ef al.

g/sq m on Day 2 with leucovorin rescue; cyclophosphamide, 600 mg/sq m on Day 15; 5-(3,3-dimethyltriazeno)imidazole-4-carboxamide,

200 mg/sq m from Days 15 to 19. Adriamycin, 50 mg/sq m on Days30 and 45. (c) High-dose MIX, 5 and 25 g/sq m with leucovorin

rescue, (d) Adriamycin, 75 mg/sq m.Tumor Cell Separation. The tumor samples were obtained during

surgery performed on primary or metastatic osteosarcoma. The material was collected under sterile conditions in MEM with the followingadditions: 20 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid,

pH 7.2; 1 mM glutamine; penicillin (100 units/ml); streptomycin (100¿ig/ml);and mycostatin (50 units/ml). In the laboratory, the tissue wasminced with a scalpel and resuspended in the above-described medium

at 0.2 to 0.3 g/ml. Different concentrations of collagenase were added,and the suspension was incubated at 37°under magnetic stirring (see"Results"). The turbid supernatant was afterwards filtered through

gauze, and the tumor cells were either used at this step or furthersubmitted to a density gradient centrifugation. The gradient was performed by gently layering 10 ml of the filtered cell suspension on topof a cushion obtained by mixing 7.2 ml of 8% Ficoll and 3 ml of 33%Hypaque, and the suspension was centrifugea in an HN-S International

centrifuge at 2500 rpm for 30 min. The fractions obtained are describedin the text. The interphase containing the tumor cells was aspiratedwith a Pasteur pipet, diluted with MEM, and centrifuged at 2000 rpmfor 10 min. The pellet was resuspended at a concentration of 1 x 106

cells/ml in the culture medium described above with the addition of10% FCS.

The cell viability was measured with the trypan blue exclusion testand ranged from 60 to 95% in different experiments. Alternatively, thecells were resuspended at a concentration of 10 x 106 cells/ml in

MEM-10% FCS-10% dimethyl sulfoxide and frozen in liquid nitrogen.

After thawing, the viability of the cells ranged from 40 to 60%.Lymphocyte Purification and PHA Stimulation. The method used

for lymphocyte purification from normal peripheral blood has beendescribed (6). Purified lymphocytes were resuspended at a concentration of 1 x 106 cells/ml in MEM containing 10% FCS, 20 mM 4-(2-

hydroxyethyl)-1-piperazineethanesulfonic acid (pH 7.2), 1 mM glutamine, penicillin (100 units/ml), and streptomycin (100 fig/ml). Mito-genie stimulation was achieved by adding 0.1 ml of PHA (PHA-P

resuspended in 5 ml of bidistilled water) per 100 ml of culture. Thecultures were incubated at 37°for 72 hr.

Ehrlich Ascitic Mouse Cells. These were obtained from the peritoneal cavity of Swiss mice, which had been inoculated 5 days beforewith 0.2 ml of a suspension containing 3 x 108 cells. The cells were

washed with phosphate-buffered saline (NaCI, 8 g/liter; KCI, 0.2 g/liter; Na2HPO„,1.15 g/liter; MgCI2.6H2O, 0.1 g/liter; phenol red, 10mg/ml) and were resuspended at 1 x 106 cells/ml in the same medium

used for lymphocyte culture.Mouse Sarcoma 180 Cells. This tumor was grown in BALB/c mice,

inside a small glass rod inserted s.c. After 10 days of growth, the tumorwas extracted, and the cells were dissociated with collagenase asdescribed for osteosarcoma. They were then resuspended at 1 x 106

cells/ml in the medium described above.[ 'HJdThd and [3H]dUrd Incorporation into DNA. After the cells were

resuspended in the culture medium, they were divided in aliquots of 2ml (1 x 106 viable cells/ml) and incubated for 1 hr at 37°. Unlessotherwise stated, the cells were pulsed with 5 /¿Ciof [3H]dThd (specificactivity, 40 to 60 Ci/mmol) or with 2 fiCi of [3H]dUrd (specific activity,15 to 30 Ci/mmol) for 60 min at 37°.

The reaction was stopped by centrifugation, and the cell pellet wasresuspended in 1.2 ml of water. After the preparation was mixed andallowed to stand in ice for 5 min, 0.1 ml of 0.1 M sodium pyrophosphate,0.2 ml of heat-denatured DNA (2.5 mg/ml), and 0.25 ml of 12%

perchloric acid were added. The mixture was centrifuged at 5000 rpmfor 10 min, and the precipitate was dissolved in 0.5 ml of 0.2 N NaOH.After 1 ml of water was added, the solution was precipitated with 0.25ml of 12% perchloric acid. This procedure was repeated once more,and the precipitate was finally dissolved in 0.2 ml NCS, added to 10 ml

of Toluene-Omnifluor scintillation mixture, and counted in a Nuclear

Chicago liquid scintillation counter.Materials. Tissue culture media were obtained from Grand Island

Biological Co. (Grand Island, N. Y.), MTX was from Lederle Laboratories (Pearl River, N. Y.), collagenase was from Worthington BiochemicalCo. (Freehold, N. J.), and PHA-P was from Difco Laboratories (Detroit,

Mich.). Radioactive materials and Omnifluor were from New EnglandNuclear (Boston, Mass.), and NCS was from Amersham, The Radi-

ochemical Centre, (Buckinghamshire, England).Electron Microscopy. The cells recovered from the gradient inter-

phase were centrifuged at 1000 rpm for 5 min, fixed with 4% glutar-aldehyde in 0.1 M sodium phosphate buffer (pH 7.4) at 4°for 2 hr, and

rinsed overnight in the same buffer containing 0.1 M sucrose. The cellswere then postfixed in 1% OsCu in the same buffer at 4°for 2 hr. The

specimens were then dehydrated in graded ethanol and embedded inEpon. Ultrathin sections were cut with a Porter-Blum MT-2B ultrami-

crotome, mounted on copper grids, doubly stained with uranyl acetatefollowed by lead citrate, and observed with a Siemens Elmiskop 101electron microscope.

RESULTS

Cellular Disaggregation of Osteosarcoma. Several methodswere assayed to release the osteosarcoma cells from thesurgical sample. Since mechanical dissociation with needlesreleased only negligible amounts of cells, experiments wereperformed in which the tissue was incubated with severalenzymes. Pronase, trypsin, and collagenase were tested, andit was found that the last enzyme was the most efficient inreleasing tumor cells. In order to determine the optimal conditions, several collagenase concentrations (0.1, 0.5, and 1.0mg/ml) and incubation times were assayed. A sample from aprimary osteosarcoma (Patient 3) was resuspended at 0.2 g/ml of culture medium as described previously (see "Materialsand Methods"). After addition of collagenase, the number of

released viable cells was determined at different times. It wasobserved that without collagenase there was practically norelease of cells and that 1 mg/ml was the most efficientconcentration tested, a plateau being reached after 3 hr ofincubation at 37°.These conditions were afterwards used for

the experiments described below.Characterization of Dissociated Tumor Cells. After the

collagenase treatment, phase-contrast microscopy revealedthe presence of large cells with irregular nuclei, often binu-

cleated, and lymphocytes infiltrating the tumor at a variabledegree according to the patient. RBC were also found invariable amounts, mainly depending on the histological varietyof osteosarcoma. The cellular suspension obtained was filteredthrough gauze, and the cells were used at this step in someexperiments described below. However, in order to verify ifsome of the observed biochemical characteristics could betruly ascribed to the osteosarcoma cells, a further purificationstep was attempted. For this purpose, the cell suspension waslayered on the top of a gradient and centrifuged as describedin "Materials and Methods." The interphase contained virtually

pure large cells (Fig. 1), whereas the pellet contained celldebris, lymphocytes, and RBC.

The recovery of the tumor cells after the gradient centrifugation varied between 40 and 67% in different experiments.

Electron microscopy performed on the cells recovered in theinterphase revealed cytological features which have been described as characteristic of osteosarcoma cells (26), i.e., nu-

3622 CANCER RESEARCH VOL. 41

Research. on October 31, 2018. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

Action of MTX on Human Osteosarcoma

¡p-«»V«'

SÎAAutfiî

^^m^i1.«ïte«.-JOJ£ì»:s

Ai^r-«#>,

Fig. 1. Phase-contrast microscopy of disaggregated osteosarcoma cells. A.X 628; B, x 1,570.

clear irregularity and prominent and dilated rough endoplasmicreticulum (Fig. 2). Two histochemical tests were also performedon the original tumor and the isolated cells: (a) the enzymatictest for alkaline phosphatase; and (t>) the periodic acid-Schiff

reaction to detect cellular glycogen. The results (not shown)revealed that both the tumor and the interphase-recoveredcells were strongly positive for these cell components. On thebasis of this evidence, we concluded that the cells recoveredin the gradient interphase were highly purified osteosarcomacells.

The cells recovered from the pellet were inactive with respectto DNA synthesis, as determined by [3H]dThd incorporation

into DNA.Distribution of DNA-synthesizing Cells in a Lung Metas

tasis. In order to investigate if DNA-synthesizing osteosarcomacells were uniformly distributed in a lung metastasis, cells from3 different zones were obtained. A 4-mm-thick slice was cut

along the major axis of the metastasis, and 3 zones (5x5x4mm) were separated: peripheral (the position situated underthe capsula); intermediate; and internal. The cells from eachzone were disaggregated as described in "Materials and Methods." The viabilities of the cells from the 3 zones were very

similar (50, 60, and 52% for the peripheral, intermediate, and

Table 1

DNA synthesis by cells isolated from different zones of an osteosarcomametastasis

A lung metastasis (3x2 cm) from Patient 5 was placed in the mediumdescribed in "Materials and Methods." The cells from 3 different zones weredissociated with collagenase (see "Materials and Methods ") and divided intoaliquots of 3 ml, and 10 fiCi of [3H]dThd were added. After 2 hr at 37°, theradioactivity incorporated into DNA was measured as described in "Materialsand Methods."

[3H]dThd incorporated into DNA

Cells obtainedfromPeripheral

zoneIntermediate zoneCentral zonecpm/1

x 10ecells6.150

±653a

4.560 ±6022.610 ±262%100

74.142.4

" Mean ±S.D.

internal zones, respectively). The cells were resuspended at aconcentration of 1 x 106 viable cells/ml of culture medium,and the [3H]dThd incorporation into DNA was studied. The

results are shown in Table 1. It may be observed that the cellsfrom the peripheral part of the metastasis are more activelysynthesizing DNA.

Correlation between m Vivo and in Vitro Sensitivity to MTXin Patients with Osteosarcoma. As a first approach to theanalysis of the possible mechanisms responsible for clinicalresistance to MTX, it was decided to study in vitro the cellularsensitivity to MTX in patients in which metastatic diseasecontinued to develop despite repeated courses of MTX. As acontrol, cells from primary tumors were also analyzed. Theclinical evolution of the patients, the administered treatments,and the moment at which the surgical samples were obtainedare indicated in Chart 1. The sensitivity to MTX was measuredthrough the inhibition of [3H]dUrd incorporation into DNA. The

tumor cells used for these experiments were obtained afterpurification through a density gradient (see "Materials andMethods"). In some cases, the results were compared with

those obtained using the tumor cells without purification andwere found to be closely similar (data not shown). In Chart 2,A to C, the results obtained with 3 patients are presented. Twoof them displayed similar sensitivity to MTX (Chart 2, A and B)whereas the inhibition results obtained with Patient 6 (Chart2C) revealed the presence of cells more resistant to MTX. Theshoulder observed in the inhibition curve could suggest theexistence of 2 different cell populations. The same experimentwas also performed with different cell types, neither of whichhad been previously exposed to MTX (Chart 2D). In order toquantify the results, the sensitivity to MTX is expressed as I50,i.e., the concentration of MTX which inhibits the incorporationof [3H]dUrd into DNA to 50% of the controls without MTX. The

sensitivity of the cells isolated from all the primary tumors andmétastases is shown in Table 2. Four patients with primarytumors were analyzed, and it may be observed that the I50ranges between 3 x 10~7 and 3 X 10~6 M. With respect to the

sensitivity of the tumor cells isolated from métastases, theresults were variable. In the case of Patient 5, the lung and s.c.métastaseswere obtained during different surgical proceduresseparated by a period of 19 months (see Chart 1). In themeantime, the patient received 20 courses of high-dose MTXwith leucovorin rescue. It may be seen that the sensitivity ofthe cells isolated from both métastasesis similar to that of theprimary tumors, with I50 of 5 x 10~7 and 1.5 x 10~7 M,

respectively. The cells from Patient 6 were isolated from a lung

SEPTEMBER 1981 3623

Research. on October 31, 2018. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

J. Mordati et al.

PATIENT N't G P. MALE, 18 y

OSTEOSARCOMA FROM TIBIA

[SURGERY]MTXTA(4g)— v <*0

5MFfi10 15 2025PATIENT

N'2,J.CL,MALE,30

35 4045ByOSTEOSARCOMA

FROMTIBIA1SURGERY]-^CljafD0

51

SURGERY!10

15 2025PATIENT

N'3, OGP.MALEOSTEOSARCOMA

FROM30

35 4045,19yTIBIA

15 20 25 30

PATIENT N't. 0 J.MALE, 12 y

OSTEOSARCOMA FROM FEMUR

35 W 45

NED

20 30 35 40 45PATIENT N"5, OF. MALE. 22y

OSTEOSARCOMA FROM FEMUR

35 40X) 15 20 25 30PATIENT N'6, Ch G MALE.ISy

TELANGIECTATIC OSTEOSARCOMA FROM FEMUR

t DEATH

15 20 25 30 35PATIENT N°7.CS FEMALE, 14y

OSTEOSARCOMA FROM HUMERUS

45

t DEATH

15 20 25 30

TIME (MONTHS)

35 40 45

Chart 1. Patients' evolution and treatments administered. Outline of main

features of the course of the disease. The surgery of the primary tumor is placedat Time 0. Heavy arrows, moment at which tumor samples were obtained; WED,nonevident disease. Details of treatments administered are in Materials andMethods. ' •local radiotherapy; A. LOMA; •.VIMELECIDA ((1), 3 courses; (2).

7 courses]; • high-dose MTX {( 1), 23 courses; (2). 6 courses]; T, Adriamycin;y , years.

metastasis which developed despite the fact that the patientreceived 7 courses of a combined treatment including MTX(1.5 g/sq m), and 6 courses of high-dose MTX with leucovorinrescue. Patient 7 had also received several courses of high-dose MTX, although the treatment had been interrupted 8months before the resection of the pulmonary metastasis. Thecells from these patients were considerably more resistant toMTX, with I50's of 1.3 x 10~5 and 4 x 10~5 M, respectively.

The sensitivities to MTX of human lymphocytes, Ehrlich cells,and Sarcoma 180 cells are very similar, with an I50 rangingfrom 1.4 x 10~7to2.1 x 1CT7M.

Estimation of the Salvage Pathway for dTMP Synthesisthrough the Effect of MTX on [3H]dThd Incorporation into

DMA. As shown previously, the in vitro inhibition by MTX

120

80

40

120

40

0 Sx»'' Sx»'7 5x«)5 0 5«10' 5xB' 5x»s

| MTX | Molar

Chart 2. Effect of MTX on [3H]dUrd incorporation. The osteosarcoma cellswere obtained as described in "Materials and Methods,' and 30 min afteraddition of MTX at the above-indicated concentrations a pulse of [3H]dUrd was

given. The radioactivity is expressed as percentage of the control without theaddition of MTX. A, Patient 1. B, Patient 5: •,lung metastasis; O, s.c. metastasis.C, Patient 6. In D, the following cells were used: 72-hr PHA-stimulated lymphocytes (•};Ehrlich ascitic tumor cells (O); and mouse Sarcoma 180 cells (A).[3H]dUrd pulses were performed as described above.

Table 2Effect of MTX on [3H]dUrd and ¡3HIdThd incorporation into DNA

The inhibition by MTX of [3H]dUrd incorporation into DNA has been measured

for each patient as described in Chart 2. The data reported for the MTX effect on[3H]dThd incorporation into DNA represent the maximal stimulation obtained. In

every case, a concentration curve similar to those reported in Chart 4 has beenperformed. The effect of each MTX concentration was tested in duplicate, andwhen enough cells were available several experiments were performed. Statisticalanalyses were performed when at least 3 different experiments were done.

Patient12345567Source ofcellsPrimary

tumor,tibiaPrimarytumor,tibiaPrimarytumor,tibiaPrimarytumor,femurLung

metastasiss.c.metastasisLungmetastasisLung

metastasis[3H]dUrd

inhibitionbyMTX

(I»)(M)8.7X10'72X10~63X 1073X10~65X10~71.5XIO"71.3X10~5(2)b4x 10~5Maximal

stimulation of[JH]dThdincorporated

intoDMA(%)NDaND150205310ND249

(2)27

Human peripheral lym- 1.4 ± 0.6 x 10 ' (3) 52.3 ± 8.3 (9)

phocytes, PHA stim

ulatedEhrlich mouse ascitic 2.1 ± 0.1 x 10~7(4) 54.7 ± 2.7 (3)

cellsSarcoma 180 cells 1.5 ±0.2 x 10~7 (3) ND

" ND. not determined.6 Numbers in parentheses, number of experiments.c Mean ±S.D.

of [3H]dUrd incorporation into DNA of metastatic cells was

variably correlated with clinical resistance to the drug. It wastherefore decided to explore the magnitude of the salvagepathway for dTMP synthesis in these cells, which through thesynthesis of dTMP by thymidine kinase could prevent the"thymineless state," which has been pointed out as the crucial

event leading to cell death (12). The rationale of the performedexperiments is the following. If the de novo synthesis of dTMP

3624 CANCER RESEARCH VOL. 41

Research. on October 31, 2018. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

Action of MTX on Human Osteosarcoma

is completely inhibited by MTX at high concentrations (Chart2), the dTMP pool should contract, being solely constituted viathe salvage pathway. If it is assumed that neither the enzymesinvolved in the salvage pathway of dTMP synthesis nor thosedirectly involved in DNA replication are affected by MTX in ourexperimental conditions, when tracer amounts of [3H]dThd aregiven to the cells, the increase in the [3H]dThd incorporated

into DNA would reflect directly the higher specific activity ofthe dTMP pool due to the suppression of the de novo synthesis.

The results obtained with osteosarcoma cells from 2 patients(Patients 5 and 6) in typical experiments are shown in Chart3A. It may be observed that in the presence of MTX [3H]dThd

incorporation into DNA is highly increased, 310 and 249%,respectively. The inverse relationship existing between the[3H]dUrd and [3H]dThd incorporation into DNA strongly sug

gests that the contraction of the de novo dTMP pool is responsible for the increase of [3H]dThd incorporation into DNA [i.e.,

Patient 5 (Charts 2A and 3A)]. In the case of Patient 6, whosepattern of inhibition by MTX of [3H]dUrd incorporation demon

strates the presence of MTX-resistant cells (Chart 2C), the[3H]dThd incorporation into DNA in the presence of MTX ap

pears to increase more gradually and at an MTX concentrationof 5 x 10~5 M has not yet reached a plateau.

This effect was also tested in nonrelated cells, i.e., humanPHA-stimulated lymphocytes and Ehrlich ascitic cells. In Chart36, it is shown that the increase in [3H]dThd incorporation in

the presence of MTX is much smaller, 52.3 and 54.7%, respectively.

Table 2 shows the data of maximal stimulation by MTX of[3H]dThd incorporation into DNA in the osteosarcoma cells

isolated from primary tumors and métastasesand in nonrelatedcells. The smaller increase in osteosarcoma was obtained inPatient 7, whose cells were isolated from a lung metastasis.

In order to test whether the collagenase treatment to whichosteosarcoma cells were submitted could in some way alter theeffect of MTX on [3H]dThd incorporation into DNA, lymphocytes

and Ehrlich cells were submitted to an identical collagenasetreatment without any change being observed in the degree ofstimulation by MTX (data not shown). From the data shown in

®

300

200

É 100

0 5xW' 5xK)7 5x105 0 5x10^

(MTX) Molar

SxW 5xK>5

Chart 3. Effect of MTX on [3HJdThd incorporation. The effect of differentconcentrations of MTX on [3H]dThd incorporation was assayed on different cells.

A, human osteosarcoma: •.Patient 5; O. Patient 6. B, human PHA-stimulatedlymphocytes (•)and Ehrlich ascitic tumor cells (O). The cells were obtained asdescribed in "Materials and Methods." and 30 min after addition of MTX at theabove-indicated concentrations a pulse of [3H]dThd was given. The radioactivity

is expressed as percentage of the control without the addition of MTX.

Chart 3 and Table 2, it may be concluded that in osteosarcomacells (with the exception of Patient 7) the dTMP pool is synthesized mainly through the de novo pathway, whereas in Ehrlichcells and in human stimulated lymphocytes its origin is mainlythrough the salvage pathway.

Effect of Time and Hypoxanthine on MTX Stimulation of[3H]dThd Incorporation into DNA. The high-dose MTX infusions performed on patients maintain high blood levels (>10~5

M) for about 24 hr. It was therefore important to verify if thedTMP pool contraction due to DHFR inhibition obtained with30 min of MTX treatment was maintained for longer periods.Also, since it has been reported that the dTTP pool is largerthan the dCTP, dATP, and dGTP pools in many different typesof cells (5, 24), it was important to verify if purine nucleotides,the synthesis of which is also blocked by MTX, were ratelimiting for [3H]dThd incorporation into DNA. If this were the

case, addition of a purine precursor like hypoxanthine shouldincrease [3H]dThd incorporation into DNA. The result of this

experiment performed with osteosarcoma cells is shown inChart 4. It may be observed that the stimulatory effect of MTXon [3H]dThd incorporation remains quite stable during at least6 hr at 37°and that hypoxanthine produces a small stimulatory

effect.When the experiment was performed with stimulated lym

phocytes, it was observed that the stimulatory effect of MTXwas maximal after 30 min of MTX action and then became verysmall. The stimulatory effect of hypoxanthine was again notvery important. Control experiments demonstrated that hypoxanthine alone did not alter [3H]dThd incorporation into DNA.

These experiments strongly suggest that the dATP and dGTP

400

200

Time (hr)Chart 4. Effect of time and hypoxanthine on MTX stimulation of [3H|dThd

incorporation into DNA in osteosarcoma cells. The cells were obtained from alung metastasis as described in "Materials and Methods and diluted to aconcentration of 1 x 106/ml in the usual culture medium. After 1 hr at 37°,the

culture was divided into aliquots of 2 ml, and MTX was added to a finalconcentration of 5 x 10 6 M with (O) or without (•)addition of 1 x 10"* M

hypoxanthine. Controls without additions were also assayed. At the times indicated above, 5 jiCi of [3H)dThd were added to each aliquot, and incubation wascontinued for another 60 min at 37°.The reaction was terminated as describedin "Materials and Methods." Inset, absolute values of [3H]dThd incorporation of

the control tubes at different times. The results are expressed as percentage ofpHJdThd incorporation of the controls.

SEPTEMBER 1981 3625

Research. on October 31, 2018. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

J. Mordo/7 et al.

pools are not seriously rate limiting for [3H]dThd incorporation

into DNA, at least during the first 6 hr of MTX action.

DISCUSSION

Further advances in the treatment of human osteosarcomawill probably be obtained if some characteristics of the osteosarcoma cells, such as cell growth kinetics, growth requirementfactors, and sensitivity to Chemotherapie agents, may be estimated. To perform many of these experiments, it is desirableto obtain isolated cells from this tumor, and in this paper amethod is described which makes it possible to obtain a largenumber of highly purified osteosarcoma cells. This methodinvolves an enzymatic treatment with collagenase, and similartreatments have been reported to preserve the cloning capabilities of human melanoma and sarcoma (16).

The question of whether the clinical resistance to MTX wasaccompanied by in vitro cellular resistance to the drug wasalso analyzed in this paper. Two methods are commonly usedto measure drug sensitivity of tumor cells: long-term cultures

where the ability of drugs to diminish cellular reproduction andcolony formation is measured (20); or short-term cultures

where the effect of drugs on the incorporation of metabolicprecursors into macromolecules is determined. The results ofboth methods have been reported to be correlated with theclinical response (20, 25). Since attempts performed in ourlaboratory to develop colonies in vitro from osteosarcoma usingthe semisolid agar technique (20) have until now been unsuccessful, the inhibition by MTX of [3H]dUrd incorporation into

DNA was therefore analyzed. For the cells isolated from primarytumors, I50 ranged between 3 x 10~7 and 3 x 10~6 M. The

results obtained with cells isolated from clinically resistantpatients were variable. In one patient (Patient 5), clinical resistance to MTX was not accompanied by in vitro cellularresistance to the drug, even in tumor cells obtained from ametastasis 2 months before the patient's death. In other pa

tients, however (Patients 6 and 7), the I50for MTX was considerably higher. It seems clear, therefore, that clinical resistanceto MTX may or may not be accompanied by increased in vitrocellular resistance to the drug. In the case of tumor cells withan increased I50, it could be assumed that they developedresistance mechanisms such as higher amounts of DHFR, analtered DHFR, or a lesser permeability to MTX, which could bebypassed by higher MTX concentrations. In the second case,where the cells have normal sensitivity to MTX in our assaysystem, clinical resistance could be due to other mechanisms,such as a high number of resting cells or diminished drugaccess to the tumor. Another possible mechanism to circumvent the inhibition by MTX is suggested by the observations inthis paper on the different contribution of the salvage and thede novo pathways to form the dTMP pool in cells of differentorigins. Since the "thymineless state" appears to be a main

factor in the cell death caused by MTX (12), it could beassumed a priori that cells with a relatively important de novopathway for dTMP synthesis would be more sensitive to MTXthan were those with a predominant salvage pathway. In thecourse of studies on the intrinsic sensitivity to MTX of differentcell lines, Tattersall ef al. (24) observed that: (a) a highersensitivity to MTX appeared to be associated with smaller dTTPpools; (£>)after 24 hr of MTX treatment, the only deoxyribo-

nucleoside triphosphate pool consistently diminished was the

dTTP pool, the reduction again being higher in the most sensitive cells; and (c) after the addition of dThd to the extracellularmedium, there was a greater increase in the dTTP pool in themost resistant cells. It is reported here that in human osteosarcoma cells the de novo pathway for dTMP synthesis is predominant and this could explain the reported efficacy of MTX as acoadjuvant chemotherapeutic agent in this tumor. These observations could lead to the speculation that an activation ofthe salvage pathway would make it possible to acquire a certaindegree of resistance to MTX. Partial support for this hypothesishas been obtained in only one patient (Patient 7) whose denovo pathway for dTMP synthesis was considerably smallerthan usual (Table 2). It is interesting to note that in this patientthe increase in the salvage pathway coexisted with an increased I50for [3H]dUrd incorporation into DNA. This observa

tion would suggest that 2 resistance mechanisms could combine in a particular patient. In the case of Patient 6, in whomosteosarcoma cells resistant to MTX were detected (Chart 2),the stimulation of [3H]dThd incorporation by MTX had peculiar

characteristics. Although the maximal increase obtained washigh (249%), the activation curve had several inflection pointsand was still rising at the higher MTX concentration tested.This pattern could be compatible with a heterogeneous cellularpopulation as regards the MTX ability to inhibit the de novo-

pathway.The activation of the salvage pathway has been reported in

other cells. Thus, Cooper ef al. (3) found in leukemic cells thatthe proportion of dThd from the extracellular medium incorporated into DNA varies between 13 and 87% when dThd concentration in the extracellular medium increased from 0.03 to300 ftM. This finding would be particularly important in patientsbeing treated with chemotherapy, since due to cellular destruction large amounts of degradation products are probably released to the extracellular medium. These nucleotides couldsupply the lack of thymidylate produced by the inhibition of thede novo pathway. It is interesting to note that Pinedo ef al. (17)observed that it was not possible to obtain consistent blockageof colony-forming units with MTX when nondialyzed FCS andL-cell supernatant were used. This effect was probably due tothe presence of nucleic acid precursors which after dialysiswere reduced to undetectable levels. These authors have alsofound that the amount of nucleic acid precursors varied between different batches of FCS. This source of variability wasconsidered in the experiments described in this paper; therefore, the same batch of FCS was used throughout.

It seems therefore that the cellular proportion of the de novoand salvage pathways for dTMP synthesis is variable, depending on several factors. One of them is the nature of the cellitself, since it has been shown in this paper that under identicalculture conditions osteosarcoma cells have a de novo pathwaymore important than that of lymphocytes or Ehrlich cells. Another factor appears to be the dThd concentration in theextracellular medium (3).

Once the de novo pathway for dTMP synthesis in a cell hasbeen interrupted by MTX, the ability of the cell to survive willprobably depend on 2 factors, the size of the remaining dTMPpool and the ability of the cell to replenish it through a higheractivity of the salvage pathway. It has been shown in this paperthat the remaining dTMP pool after exposure to MTX is differentin osteosarcoma cells from that in lymphocytes. The capacityof these different cells to activate their salvage pathways is

3626 CANCER RESEARCH VOL. 41

Research. on October 31, 2018. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

Action of MTX on Human Osteosarcoma

unknown, although it has been shown in this paper that thecontraction of the dTMP pool by MTX in osteosarcoma is stablefor at least 6 hr under our culture conditions (Chart 4). It hasalready been mentioned that some cells increase their salvagepathway when dThd extracellular concentration increases. Apparently, this effect can be therapeutically exploited, since ithas been reported that in mice bearing L1210 leukemia administration of dThd could prevent MTX toxicity without affectingits antitumor action (23). In the case of osteosarcoma, it isreasonable to assume that these cells would be less effectivelyrescued by dThd than would normal tissues with more important salvage pathways.

The methodology and data reported in this paper could beuseful for further studies on the biological behavior of osteosarcoma cells including the biochemical basis of the responseand resistance to MTX and other agents.

ACKNOWLEDGMENTS

The authors would like to thank Dr. Adelina Riarte de Tamaroff (ILAFIR-Institute, Universidad del Salvador. Buenos Aires. Argentina), for performing theelectron microscopy observations reported in this paper, and Dr. Christiane D.de Pasqualini, Academia Nacional de Medicina. Buenos Aires, Argentina, for thegift of Sarcoma 180.

REFERENCES

1. Berlino, J. R. Toward improved selectivity in cancer chemotherapy: TheRichard and Hinda Rosenthal Foundation Award Lecture. Cancer Res.. 39.293-304, 1979.

2. Borsa, J., and Whitmore, G. F. Cell killing studies on the mode of action ofmethotrexate on L-cells in vitro. Cancer Res.. 29: 737-744, 1969.

3. Cooper, R. A., Perry. S.. and Breitman, T. R. Pyrimidine metabolism inhuman leukocytes. Cancer Res , 26 2267-2275, 1966.

4. Djerassi. I. High-dose methotrexate (NSC-740) and citrovorum factor (NSC-3590) rescue: background and rationale. Cancer Chemother. Rep., 6: 3-6,1975.

5. Fridland, A. Effect of methotrexate on deoxyribonucleotide pools and DMAsynthesis in human lymphocytic cells. Cancer Res., 34: 1883-1888. 1974.

6. Fridlender, B., Medrano. E., and Mordoh. J. Synthesis of DMA in humanlymphocytes: possible control mechanism. Proc. Nati. Acad. Sei. U. S. A.,71: 1128-1132, 1974.

7. Galivan, J. Transport and metabolism of methotrexate in normal and resistantcultured rat hepatoma cells. Cancer Res.. 39: 735-743, 1979.

8. Goldman, I. D. Analysis of the cytotoxic determinants for methotrexate (NSC-740): a role for "free" intracellular drug. Cancer Chemother. Rep., 6: 51-

61, 1975.9. Hill, B. T.. Bailey, B. D., and Goldman, I. D. Membrane transport of folates

and antifolates in lines of 5178Y cells. One sensitive to methotrexate andone resistant to methotrexate due to a permeability defect. Proc. Am. Assoc.Cancer Res., 79. 49, 1978.

10. Hryniuk, W. M. Purineless death as a link between growth rate and cytotox-icity by methotrexate. Cancer Res.. 32. 1506-1511, 1972.

11. Hryniuk, W. M. The mechanism of action of methotrexate in cultured L5178Yleukemia cells. Cancer Res., 35: 1085-1092, 1975.

12. Hryniuk. W. M., and Berlino, J. R. Treatment of leukemia with large dosesof methotrexate and folinic acid: clinical-biochemical correlates. J. Clin.Invest.. 48: 2140-2155, 1969.

13. Jaffe, N., Frei, E., Ill, Watts, H., and Traggis, D. High-dose methotrexate inosteogenic sarcoma: a 5-year experience. Cancer Treat. Rep., 62: 259-264. 1978.

14. Lindquist, C. A., Moroson, B. A., and Berlino, J. R. A methotrexate sublineof L1210 characterized by impaired transport of methotrexate. but not 5-methyltetrahydrofolate. Proc. Am. Assoc. Cancer Res.. 19: 165, 1978.

15. Nithammer. D., and Jackson. R. C. Changes of molecular properties associated with development of resistance against methotrexate in human lym-phoblastoid cells. Eur. J. Cancer, 11: 845-854, 1975.

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

17. Pinedo, H. M., Zaharko, D. S., Bull, J. M., and Chabner. B. A. The reversalof methotrexate cytotoxicity to mouse bone marrow cells by leucovorin andnucleosides. Cancer Res., 36: 4418-4424. 1976.

18. Pratt, C. B., Howarth, C., Ransom, J. L., Bowles, D., Green, A. A., MaeshKumar, A. P., Rivera. G.. and Evans, W. E. High dose methotrexate usedalone and in combination for measurable primary or metastatic osteosarcoma. Cancer Treat. Rep.. 64. 11-20. 1980.

19. Rosemberg, S. A., Chabner, B. A., Young, R. C., Serpp, C. A., Levine, A. S.,Costa, J., Hanson, T. A., Head, G. C., and Simon, R. M. Treatment ofosteosarcoma. I. Effect of adjuvant high-dose methotrexate after amputation.Cancer Treat. Rep., 63: 739-751, 1979.

20. Salmon. S. E., Hamburger. A. W.. Soehnlen, B., Durie. B. G. M., Alberts, D.S.. and Moon. T. E. Quantitation of differential sensitivity of human tumorstem cells to anticancer drugs. N. Engl. J. Med.. 298: 1321-1327, 1978.

21. Schrecker, A. W., Mead, J A. R., Greenberg, N. H., and Goldin, A. Dihydro-folate reducíase activity of leukemia L1210 during development of methotrexate resistance. Biochem. Pharmacol., 20: 716-718. 1971

22. Sirotnak, F. M., and Donsbach, R. C. The intracellular concentration dependence of antifolate inhibition of DNA synthesis in L1210 leukemia cells.Cancer Res., 34: 3332-3340, 1974.

23. Tattersall, M. H. N., Brown, B., and Frei, E.. III. The reversal of methotrexatetoxicity by thymidine with maintenance of antitumor effects. Nature (Lond.).253: 198-200, 1975.

24. Tattersall, M. H. N., Jackson, R. C.. Jackson. S. T. M.. and Harrap, K. R.Factors determining cell sensitivity to methotrexate: studies of folate anddeoxyribonucleoside triphosphate pools in five mammalian cell lines. Eur. J.Cancer, 10: 819-826, 1974.

25. Volm, M., Wayss, K.. Kaufmann. M., and Mattern, Y. Pretherapeutic detection of tumor resistance. Eur. J. Cancer, /5: 983-993, 1979.

26. Williams, A. H., Schwinn, C. P., and Parker, J. W. The ultrastructure ofosteosarcoma. A review of twenty cases. Cancer (Phila.). 37 1293-1301,1976.

SEPTEMBER 1981 3627

Research. on October 31, 2018. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

J. Mordoh et al.

C »V

ICI •^

- m* -TjKÌaaaSW f >'--À;f^;S-:è£'-'••-^i'ÌSc^

' - r"1 *"

rL' v^ ¿V. « f

2A

i1^

mi.

rer, •.

2 B 2CFig. 2. Electron microscopy of disaggregated osteosarcoma cells (Patient 4). A, osteosarcoma cell containing an irregular nucleus, mitochondria (mi), and

abundant and dilated rough endoplasmic reticulum (rer). x 10.500. 6. high magnification of A showing filamentous structures (0 irregularly arranged or in bundlesand atypical configuration of mitochondria (mi), x 40,500. C, detail of portion of malignant cell with mitochondria (mi), Golgi complex (g), and dilated roughendoplasmic reticulum (rer). x 19,500.

3628 CANCER RESEARCH VOL. 41

Research. on October 31, 2018. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

1981;41:3621-3628. Cancer Res   José Mordoh, Reinaldo D. Chacón and Jorge Filmus  Analysis of Their Sensitivity to MethotrexateIsolation of Tumor Cells from Patients with Osteosarcoma and

  Updated version

  http://cancerres.aacrjournals.org/content/41/9_Part_1/3621

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/41/9_Part_1/3621To request permission to re-use all or part of this article, use this link

Research. on October 31, 2018. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from