[CANCER RESEARCH 45, 2177-2183, May 1985]

Inhibition of DNA Ligase from Human Thymocytes and Normal or LeukemicLymphocytes by Antileukemic Drugs1

Jean-Claude David,2 ThérèseBassez, Micheline Bonhommet, and RégineRusquet

Laboratoire de Biochimie du Développement,La Centre National de la Recherche Scientifique n 256, Universitéde Rennes I, Campus de Beaulieu, 35042, Rennes Cedex,France

ABSTRACT

Human DNA ligase was purified from both normal and leukemicperipheral lymphocytes and normal thymocytes. The activity ofthe purified enzymes was assayed in the presence of severalwidely used antileukemic drugs. Melphalan and prednisone at 5nriMhad no effect. Carmustine, chlorambucil, and cyclophospha-

mide were more effective at inhibiting the enzyme from leukemiccells, whereas Adriamycin and vinblastine and their derivativeswere stronger inhibitors of the enzyme from normal cells. Vin-

cristine and etoposide inhibited DNA ligase from thymocytes andnormal lymphocytes with a low K, but were totally ineffective onthe enzyme from leukemic cells.

The three classes of intercalating anthracyclines, Vicia alkaloids, and podophyllotoxin derivatives, were the only drugs foundto markedly inhibit DNA ligases from normal cells. Less substituted molecules of the Vicia alkaloids and podophyllotoxinclasses were the more active inhibitors, whereas in the intercalating anthracycline group, it was the more substituted compounds. The clinical consequences of these observations arediscussed with respect to the role of DNA ligase in DNA replication and repair.

INTRODUCTION

Antitumoral chemotherapy essentially began about 40 yearsago with the discovery of the effect of nitrogen mustard (16,17).The first temporary remissions were obtained with the antime-

tabolite, aminopterine in 1948 (11) and complete remissions withthe closely related compound methotrexate (22). In 1956, a thirdclass of antitumoral agents was described by Farber ef al. (11)with actinomycin D and more recently, a fourth class was foundwith Vinca antimitotic (18). At present, all the antitumoral substances are classified into 5 groups: (a) electrophilic agents; (b)inhibitors of DNA synthesis; (c) DNA interacting substances; (d)poisons of the mitotic spindle; and (e) hormones (21). With thepossible exception of the last group, all these substances may,in fact, interfere with DNA metabolism. Several of these compounds are known to induce DNA breaks (35, 41, 42) therebyrendering DNA enzymatic studies of great interest. Among theenzymes involved in DNA replication or repair, DNA ligases havereceived little attention (8).

The existence of enzymes capable of joining DNA fragmentswas proposed more than 20 years ago for prokaryotic cells (19,30). Polynucleotide ligases were first described in 1967 (14,15).Their physiological importance has been assessed in prokaryotes

1This work was supported by a Grant from Institut National de la Santéet de

la Recherche MédicaleNo. 833003 and Federation Nationale des Centres de Luttecontra le Cancer.

1To whom requests for reprints should be addressed.

Received7/2/84; revised 12/11/84;accepted1/9/85.

by the inability of Escherichia coli mutants defective in theseenzymes to repair their DNA (9, 40). However, in eukaryotes,the association of DNA ligase with replication or repair has beenknown for just a few years (44). The presence of active DNAligase in lymphoid and myeloid tissues (26) as well as thestimulation of its activity during rat liver regeneration (44) orlymphocyte division (43) suggest that it may have an importantfunction in proliferative states such as leukemia. However, at thepresent time, DNA ligase activity has not been studied in thedifferent types of leukemias (8). In this report, we describeinvestigations on the effects of antileukemic drugs on DNA ligaseactivity purified from normal thymocytes and lymphocytes andfrom acute myeloblastic leukemia lymphoblasts.

MATERIALS AND METHODS

Drugs

Adriamycin was provided by Roger Bellori Laboratories, Paris, France.AD32 was a gift of Dr. Israel, University of Tennessee Medical School,Knoxville, TN. m-AMSA3 and o-AMSA were kind gifts of Professor B.

Baguley, University of Auckland, New Zealand; Ara-C and podophyllo

toxin, from Sigma. Carmustine and etoposide, from Bristol Laboratories,Paris; chlorambucil, from Techni-Pharma Laboratories, Monaco; cyclo-

phosphamide, from Laboratories Lucien, Paris. DMCOOH, DMCOOK,DMCONHî(1-p-carboxyamidophenyl-3,3-dimethyltriazene), MMCONH2,

generous gifts from Dr. Lassiani, University of Trieste; daunorubicin andmethotrexate, from Specia, Paris; melphalan, Welcome Laboratories,Paris; prednisone, from Roussel Uclaf, Paris; vinblastine, vincristine, andvindesine, from Eli Lilly, Saint Cloud, France. Kinetoplast DNA andpBR322 were gifts from Dr. M. Duguet, Paris, France.

Solutions

Fresh 10 rnw solutions were made in distilled water for the followingchemicals: Adriamycin, m-AMSA, ara-C, cyclophosphamide, vinblastine,

vincristine, vindesine, DMCOOK and MMCONH2. They were made in 1%dimethyl sulfoxide for carmustine, etoposide, podophyllotoxin, prednisone, DMCONH2, and DMCOOH; 2% dimethyl sulfoxide plus 0.01 MNaOH chlorambucil and melphalan, 0.02% Nonidet P-40 for AD32, and

0.01 N HCI for methotrexate. Controls performed with solvents aloneshowed no effect.

Preparation of Cells

Thymuses were obtained from 3 different normal cases of thoracicsurgery (Case 1, female age 15 months; Case 2, male age 30 months;

3The abbreviations used are: m-AMSA, 4'-(9-acridinyl)aminomethanesulfone-m-anisidine; o-AMSA, methoxy-m-4'-(9-acridinyl)aminomethanesulfone-m-anisidine;

ara-C, 1-/3-D-arabinofuranosylcytosine; DMCOOH, p-(3,3-dimethyl-1-triazino)-benzoic acid; DMCOOK, p-3,3-dimethyl-1-triazeno)benzoic acid potassium salt;MMCONH2, 1-p-carboxyamidophenyl-3-methyltriazene; TEM, 50 mw Tris-HCI, pH7.5, 1 mm EDTA, 1.4 rriM 2-mercaptoethanol; poly[d(A-T)]„,poly(deoxyadenylate-

deoxythymidylate) of varying chain lengths; oligo(dT),2-„,oligodeoxythymidylatewith a chain length of 12 to 18 residues.

CANCER RESEARCH VOL. 45 MAY 1985

2177

on May 31, 2020. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

EFFECT OF ANTILEUKEMIC DRUGS ON DMA LIGASE

l -f\ _

+

+

B

Case 3, male 10 years) from the Department of Thoracic Surgery, HôpitalPontchaillou, Rennes, France. Thymocytes were separated after gentlehand homogenizing in RPMI (Eurobio, Paris), separated from RBC onFicoll gradients and washed 3 times for 15 min at 1200 rpm in a Jouanrefrigerated centrifuge at 4 °C.The thymuses were kept no longer than

1 h in ice before processing).Leukemic WBC were obtained from the Centre Anticancéreux, Hôpital

Pontchaillou, Rennes, France. Three cases were selected: Case 1, acutemyeloblastic leukemia M2, female, 54 years (77% blasts); Case 2, chronicmyelogenous leukemia, male, 62 years (85% blasts); and Case 3, chroniclymphocytic leukemia, female, 40 years (90% blasts). WBC were separated from heparinized blood by centrifugation on Ficoll gradients (Flobio,Paris) for 15 min at 1400 rpm and washed 3 times in RPMI for 10 min at1200 rpm at 4 °C. Normal WBC were obtained from the Centre de

Transfusion de Rennes and purified in the same manner.

DNA Ligase Purification

DMA ligase was purified separately from the 3 types of thymocytesand the 3 types of leukemic lymphocytes. DNA ligase purification wasbasically as described previously (6) with the following modifications.Thymocytes (20 g) were suspended in 200 ml of TEM buffer and leukemiclymphocytes (4.5 x 109 cells) in 100 ml of TEM. The suspension was

homogenized for 1 min in a Servali Omni Mixer, centrifugea for 60 minat 140,000 x g, and the supernatant dialyzed for 16 h against 2 changesof TEM buffer containing 20% glycerol (TEMG buffer). This extract wasadsorbed onto a PEn phosphocellulose column equilibrated at pH 7.2.DNA ligase was eluted by 0.3 M KCI and further purified using, consecutively, Sepnadex G 150, Sephadex CM 50, and DNA cellulose chro-

matography as described (6). The sedimentation coefficients of theenzymes from thymocytes, normal lymphocytes, and lymphoblasts were,respectively, 5 ±0.5s, 5 ±0.6s, and 5.2 ±0.6s (SD) and were notsignificantly different. The homogeneity of the enzymes was checked bypolyacrylamide gel electrophoresis and found to be greater than 95% forall the purified enzymes (Fig. 1). No significant differences in the migrationmobilities were observed between the enzymes from the differentsources. In each case, detectable activity was found at the band level.

Enzyme Assays

DNA ligase activity was assayed routinely using a modification of themethod described by Modrich and Lehman (34).

Preparation of [3H]Poly[d(A-T)]„

The incubation mixture contained in 1 ml: 50 HIM Tris HCI pH 8.6, 20HIM dithiothreitol, 18 MM dTTP (Sigma), 18 MMdATP (Sigma), 4 absor-bance units of poly[d(A-T)]n (Boehringer), 5 units of DNa polymerase I,large fragment (New England Nuclear), 1.5 MM [3H]dTTP (78 Ci/mmol;

New England Nuclear), and 0.1 mg of bovine serum albumin (Sigma).After a plateau of incorporation was reached, the reaction was stoppedby 250 M! of 5 M NaCI followed by a 20-min incubation at 70 °C.The

solution was dialyzed against 50 volumes of a solution containing 100mM NaCI and 0.1 rtiM EDTA; 200-Ml aliquots were stored at -80 °Cuntil

further use.

2-

DNA Ligase Assay

Each assay (0.1 ml) contained 25 mM Tris-HCI (pH 8), 4 mw MgCI2, 1HIM ATP, 0.3 mM [3H]d(A-T)„,20 ¡ÕQof bovine serum albumin, 5 mM

dithiothreitol, and 0 to 40 M!of DNA ligase solution.After 30 min incubation at 37 °C,the assays were heated in a boiling

Fig. 1. Polyacrylamidegel electrophoresis of purified DNA ligase. DNA ligase(25 ftg) was submitted to electrophoresis from 150 min under 2.5 mA in a 5%polyacrylamide gel. Top lane, enzyme purified from thymocytes Case 1, median

lane from leukemia. Case 1, bottom lane from normal lymphocytes. A, Line 1,enzyme purified from normal thymocytes Case 1; Line 2, enzyme purified fromlymphoblasts Case 1. B, Line 1, enzyme purified from normal thymocytes Case 1;Line 2, enzyme purified from normal lymphocytes.

CANCER RESEARCH VOL. 45 MAY 1985

2178

on May 31, 2020. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

EFFECT OF ANTILEUKEMIC DRUGS ON DMA LIGASE

water bath for 3 min, and 10 units of exonuclease III (New EnglandNuclear) were added. After incubation for 30 min at 37 °C,the samples

were precipitated with trichloracetic acid (5% final concentration) filteredthrough Whatman GFC filters, and counted. One unit of DNA ligaseactivity is defined as the amount converting 1 nmol of d(A-T)n to anexonuclease Ill-resistant form in 30 min under the assay conditions.

All the kinetics studies were repeated using a second independentenzyme test according to a modification (5) of the method described byOlivera (37). The substrate was prepared as follows. One mwi oligo-

(dT),2-ie(P. L. Biochemicals) in 50 HIM Tris, pH 7.6, was incubated for 30min at 37 °Cwith E. coli alkaline phosphatase (10 units/ml). The reaction

was stopped by the addition of 0.1 volume of 20 rriM KH2PO4. Thismixture was adjusted to pH 10.5 with 1 N NaOH and kept for 15 min ina boiling waterbath. After adjusting the pH to 7.6 with 1 N HCI, theextract was subjected to centrifugation at 10,000 rpm for 10 min. Thesupernatant containing 200 MMdephosphorylated oligotdT^-iswas usedfor 5'-32P labeling in the presence of 60 ITIM Tris-HCI, pH 7.6, 6 HIMMgCI2, 6 FTIM2-mercaptoethanol, 300 UM [7-32P]ATP (20 Ci/mmol), and15 units per ml of T4 injected E. coli 5'-polynucleotide kinase (P. L.

Biochemicals) at 37 °Cuntil the plateau of incorporation was reached.The resulting [S'-^PJoligcKdT)^.^ was stored at -20 °Cuntil use. For

the determination of ligase activity, each assay (0.3 ml) contained 6 UM["PjoligoldT^-is, 6 tíMpolydeoxyadenylate (P. L. Biochemicals), 25 mw

Tris-HCI (pH 7.6), 6 JIM MgCI2, 1 mw ATP, 2.5 mM dithiothreitol, bovine

serum albumin (31 ng/m\), and 100 /¿Iof enzyme solution. The sampleswere incubated at 37 °Cfor 30 min. After the addition of 1 unit of alkaline

phosphatase and dilution with 300 n\ of ice-cold water, they wereincubated for 30 min at 80 °C. The samples were precipitated with

trichloroacetic acid, 5% final concentration, filtered through Milliporefilters, and counted. One unit of DNA ligase is defined as the activitywhich renders 1 nmol of [5'-32P]oligo(dT) resistant to alkaline phospha

tase in 30 min under standard conditions. Specific activities for theenzyme extracted from thymocytes, normal lymphocytes, and lympho-

blasts were, respectively, 75, 64, and 112 units/mg of protein whentested by the poly[d(A-T)] test. They were 82, 70, and 152 units/mg of

protein when tested by the oligo(dT) test.

Controls for Other Enzyme Activities

DNA Polymerase a and ßActivities. The polymerases activities wereassayed on purified ligases using activated calf thymus DNA in thepresence of dATP, dCTP, dGTP, and [3H]dTTP (7.1 Ci/mmol; New

England Nuclear). The «and ßactivities were separately assayed according to the method of Carre ef al. (5).

DNA Topoisomerases. ATP-independent relaxation of supercoiled

DNA was assayed on the purified ligase preparations according to themethod of Duguet ef al. (10). ATP-dependent catenation of pBR322 DNAform I was assayed as described by Liu (23) and ATP-dependent

decatenation was assayed in a similar manner with the exception thatthe DNA substrate was kinetoplast DNA. Controls made with the enzymes from the 3 sources reveal that there was neither DNA polymerasesa or ßnor topoisomerases I or II activities copurified with the ligasepreparation.

RESULTS

Inhibition of DNA Ligase Activity by Antileukemic Drugs.Thirteen widely used and 8 experimental antileukemic drugswere tested on the activity of purified DNA ligases from normalthymocytes and normal and leukemic peripheral lymphoblasts.Cyclophosphamide, melphalan, prednisone, ara-C, and metho-

trexate (0.5 mw) were without effect on ligase activity from eithertype of cell (Table 1). Carmustine was slightly inhibitory, reducingthe ligase activity of leukemic cells to 80% of the control.Chlorambucil and the experimental drugs DMCOOH and

Table 1

Effects of antileukemic drugs on the activity of DNA ligase purified fromthymocytes, normal lymphocytes, and blasts

Purified enzymes (about 0.05 unit in 20 j»l;0.4 to 0.8 jig of protein) were mixedwith 0.5 row concentrations of the different inhibitors. The enzyme assays wereperformed as described under "Materials and Methods."

DNA ligase residual activity (% of control) invarious cells

NormalDrugsIntercalating

anthracyclinesDaunorubicinAdriamycinAD32Vicia

alkaloidsVinblastineVincristineVindesinePodophyllotoxin

andderivativesPodophyllotoxinEtoposideAlkylating

agentsChlorambucilMelphalanNitrosoureasCarmustineCyclophosphamideMiscellaneousm-AMSAo-AMSAPrednisoneMethotrexateara-CDMCOOHDMCOOKDMCONHs8MMCONHj,Thymocytes25105401550402570100100100100401001001001008030100lymphocytes452518601080451010010010010010042100100100100100100100Blasts7250568010010010010040100809010045100100100508010028

" DMCONHj, l-p-carboxyamidophenyl-S.S-dimethyltriazene.

MMCONH2 were more potent inhibitors of the blast enzyme,with residual activities representing 40, 50, and 28% of thecontrol of the 2 other experimental drugs. m-AMSA was found

to be totally ineffective in inhibiting the different enzymes. However, o-AMSA inhibited the ligase activity from thymocytes,

lymphocytes, and blasts to 40,42, and 45% of the control values,respectively. Unexpectedly, daunorubicin, Adriamycin, and AD32were more potent inhibitors of DNA ligase from the normal celltypes.

Thymocyte ligase was almost totally inhibited (25,10, and 5%of control values), whereas the enzyme from normal lymphocytesdisplayed activities of 45, 25, and 18% of control values for the3 drugs, respectively. The enzyme derived from blast cells wasmuch less inhibited with residual activities of 72, 50, and 56%.Podophyllotoxin, etoposide, and the Vicia alkaloids were alsomore effective in inhibiting ligase from normal cells, althoughvinblastine and vindesine were less effective.

For these reasons, kinetic studies were performed for eacheffective drug on DNA ligase activity. As depicted in Chart 1,Adriamycin, etoposide, and vincristine were effective inhibitorsof ligase at a dose of 0.1 ITIMin thymocytes and normal lymphocytes while having little (Adriamycin) or no apparent effect on theenzyme extracted from blasts. Only chloroambucil was moreeffective at inhibiting ligase from leukemic cells.

Effects of Antileukemic Drugs on Mixed DNA Ligases.When the thymic and lymphocytic ligases were mixed in equal

CANCER RESEARCH VOL. 45 MAY 1985

2179

on May 31, 2020. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

EFFECT OF ANTILEUKEMIC DRUGS ON DMA LIGASE

100 100

75

50

I 25

O<

100

75

oÕÕ50

25

6 \

\\\

\o\

\\\\v

0.01 0.1 0.20.250.3 0.4

INHIBITOR CmM)

0.5 0.01 0.1 0.2 0.250.3 0.4 0.5

INHIBITOR CmM)

Chart 1. Dependenceof DNA ligase activity on inhibitor concentration. Purified enzymes (about 0.05 unit in 20 ¿il;0.4 to 0.8 ng of protein) were mixed with differentconcentrations of the drugs. The enzyme assays were performed as described under "Materials and Methods." Ordinate, residual activity in percentage of control;

abscissa, final drug concentration. Activity of the enzyme from leukemic cells ( ), normal lymphocytes ( ) and normal thymocytes (—). A to D, the effects ofadriamycin, chlorambucil,etoposide, and vincristine, respectively.

Table 2Effect of antileukemic drugs on the activity of ONA ligase either from specific

origin or mixed together

About 0.05 unit of DMA ligase representing, respectively, 0.6, 0.8, and 0.4 /¡gof protein in 10 n\ were allowed to react with the drug in the test. When DMAligases from different origins were used together, they were in a volume of 10 M!each at the appropriate dilution to about 0.05 unit.

DMAligase residual activity (% of control) from variousenzyme origins

DrugsAdriamycin

ChlorambucilEtoposideVincristineMMCONHjThymo

cytes10

7025

15100Lympho

cytes25

10010

10100Blasts50

4010010028T"

+L0

8010

5100T

+B37

50627055L

+B41

51606562T

+ L+B25

45404557

8 T, thymocytes; L, lymphocytes; B, blasts.

amounts, inhibition by Adriamycin, etoposide, and vincristine wassimilar to that found for the individual enzymes with completeinhibition in the case of Adriamycin (Table 2).

The mixing of enzyme from normal cells with that of blastsresulted in no further inhibition by the 3 drugs. Chlorambucil andthe experimental antileukemic drug MMCONH2 were more potent inhibitors of blast rather than normal cell-derived ligase with

no further inhibition when assayed on a mixture of enzyme from

normal and blast cells.Kinetics of DNA Inhibition. Kinetic studies were performed

using drugs effective in inhibiting DNA ligase at doses below 0.5rriM. The results for chlorambucil, etoposide, and vincristine arepresented in Chart 2. With respect to poly[d(A-T)]n substrate,the Km were, respectively, 4 x 10~7, 2 x 10~7, and 10~7 M for

ligase purified from thymocytes, normal, and lymphoblasts. Although all the observed inhibition mechanisms were competitive,the K, were different. All the studied mechanisms of inhibition arecompetitive (Table 3). Although etoposide, podophyllotoxin, andvincristine failed to inhibit ligase from blasts, they did reduce thatof normal cells. The thymocyte-extracted ligase was inhibitedwith a K, of 10~5 and 5 x 10~6 M, respectively, for etoposide and

vincristine. The 3 drugs, daunorubicin, Adriamycin, and AD32,inhibited with low K, (10~5,7 x 1(T6, and 5 x 10~6M, respectively),the enzyme from thymocytes, medium K, (2 x 10~5, 2x10~5,and 10~5 M, respectively), the enzyme extracted from lymphocyte, and high K¡(5 x 10~4, 10~4, and 2 x 10~4 M, respectively),

the enzyme extracted from blasts. K¡were very high for the othercompounds studied (chlorambucil, vinblastine, vindesine, o-AMSA, and DMCOOK).

DISCUSSION

For several years, DNA, rather than RNA or proteins, has beenconsidered an important target for carcinogenicity (4) and the

CANCER RESEARCH VOL. 45 MAY 1985

2180

on May 31, 2020. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

EFFECT OF ANTILEUKEMIC DRUGS ON DNA LIGASE

CHLORAMBUCIL

5 10 16

1/ Poly d(A-T)n x 10"7 M"1

ETOPOSIDE

OEo.

O

05mM

0.01mMOmM

05mM

5 10 16

1/ Poly d(A-T)n x 10"7 M"1

VINCRISTINE

O.SmM

o£

e

Table 3

Inhibition constants and type of inhibition of DNA ligases by antileukemic drugs

For legend, see Table 1

Origin ofenzymeThymocytesNormal

lympho-BlastsDrugsDaunorubicinAdriamycinAD32VinblastineVincristineVindesinePodophyllotoxinEtoposideChlorambucilo-AMSADMCOOKMM)10-57x

10-"5x10-«10-5x

10-«3x10-"3

xIO'51x10-51.6x

10-2x10-15X10-1TypeCCCCCCCCCCCMM)2

xIO"52x10-sio-»2

xIO"4io-55

x10-4x1Q-55x10-«2x

10-"TypeCCCCCCCCCMM)5

x10-510-2

x10"45x10-2

x10"*3x10-5x10-TypeCCCCCCC

I/ Poly d(A-T)n x IO'7 M'1

criticai cellular event which leads to mutagenicity seems to be asemiconservative DNA synthesis on a damaged template (29).DNA synthesis, either replication or repair, requires several enzymes such as polymerases (36), DNA glycosylases, endonu-cleases (25), and others (24). Among DNA-metabolizing en

zymes, DNA ligases (46) are of special interest. Although somefunctions of these enzymes are known in prokaryotic systems(20), there is a lack of information concerning eukaryotic DNAligases. Increased activity of DNA ligase has been found inmonkey kidney cells after irradiation (31) or pretreatment withmitogenic drugs (33) and after treatment of human fibroblasts bycarcinogenic drugs (32).

To our knowledge, this is the first time that DNA ligase hasbeen purified to such homogeneity from different human lymph-

oid sources. Ligase activity was found in thymocytes, normallymphocytes, acute myeloblastic, chronic myelogenous, andchronic lymphocytic leukemia peripheral blasts. Whatever theorigin of the enzyme, the sedimentation coefficient was alwaysthe same (5S). This is in good agreement with the existence ofa unique DNA ligase in normal thymus and blood after birth inbirds (7).

Relationships between Structure and Effects of the Drugs.The action of antileukemic drugs depended on the source ofligase being tested. Melphalan and prednisone had no inhibitoryeffect at up to 5 mw. The other alkylating agent chlorambucilwas a potent inhibitor of ligase from thymocytes and blasts.

Among the miscellaneous drugs studied, the inhibitors ofnucleic acid synthesis, ara-C and methotrexate, have virtually noeffect on the enzyme activity. m-AMSA and o-AMSA deservespecial mention, since only o-AMSA appeared to be equallyactive on the enzyme from the 3 different sources. The nitrosou-

reas, carmustine and cyclophosphamide, were almost inactive.Podophyllotoxin and to a greater extent its derivative etoposidewere more effective inhibitors of the ligase from thymocytes andlymphocytes. This indicates an important effect of the sugarmoiety of etoposide. Other poisons of the mitotic spindle, such

Chart 2. Kinetics of inhibition of DNA ligases by chlorambucil, etoposide, andvincristine. About 0.05 unit is incubated under conditions of initial velocity withincreasing amounts of poly[d(A-T)]„in the absence or in the presence of constantamounts of inhibitors. 1/V is expressed in 1/nmol [3H]poly[d(A-T)]n resistant toexonuclease III in 30 min and the 1/S in 1/poly[d(A-T)]„in 10~7 M~V Chlorambucil:

A, thymic enzyme; B, leukemic enzyme. Etoposide: A, thymic enzyme; 6, lymphocyte enzyme. Vincristine: A, thymic enzyme; B, lymphocyte enzyme.

CANCER RESEARCH VOL. 45 MAY 1985

2181

on May 31, 2020. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

EFFECT OF ANTILEUKEMIC DRUGS ON DNA LIGASE

001,0 OH H O 001,0 OH M O 001,0 OM M O

CH.CO'

| HO COOCII,

VINBL AST INE R = CH,

VINCRISTINt H - CMO

POnoPHYLLOTOXtN AtJD DERIVATIVE

H,CO •y^'OCH,

OCH,

POOOPHYLLOTOKIN ETOPOSIDC VPW

ALKYLATINQ AGENTS

HOOC-CH-CH,-^ y-NICH.-CM.-Cllj

UELPHALAN

HOOC-CH,-CH,-CM,-/' ^- H ICH,- CH,-CII,

\ /

CHIORAMBUCIL

H.NOC-

NITROSOURE AS

,NH-CM,-CH,-CI

VN- CHt- CM.-CI

CH¿>

MISCELLANEOUS

CH,-CH,-CI

N-CH.-CM.-CI

CtClOPHOSPMAMIDE

CH.

aiofi*

MCI

PHEONISONE

I

COOHI

VMM> ••TVM,- CM,- LMNMC -<, ^-NCM, N.

O CH,MI i MÕ,i HE i *TE

Chart 3. Molecular structures oìthe antileukemic drugs.

as vincristine, vinblastine, and vindesine have a comparableeffect, possibly enhanced by the presence of a -CH3 group.

Intercalating anthracyclines are also of great interest.The inhibition of the activity of ligase from normal cells by

intercalating anthracyclines appear to be enhanced by the presence of a lateral chain on C-13 of the drug molecule. Although

based upon just a few representatives of each family of drugs,these observations may stimulate further structure-function stud

ies (Chart 3).When enzymes purified from different sources were mixed

together, drug inhibition in the presence of blast-extracted en

zyme was not increased. This raised a major question about thestructural and catalytic similarity of the 3 types of enzymes. Atpresent, it appears that several biochemical properties are identical. All 3 enzymes have a sedimentation coefficient close to 5S,their cofactors and temperature óptimas are identical. However,these observations do not exclude structural differences. Thepossibility that closely related enzymes like DNA polymerasesand topoisomerases copurify with ligases seems unlikely, sinceno such activities have been detected in the purified ligasepreparations. This observation is important since the activity oftopoisomerases is affected by some antileukemic drugs, m-AMSA but not o-AMSA has recently been shown to induce (28)topoisomerase II. In the present study, the effect of o-AMSA andthe absence of an effect of m-AMSA on DNA ligase is in agree

ment with the absence of interference between ligase and topoisomerase. The concomitant involvement of several DNA enzymatic systems points out the study of DNA breaks underdifferent conditions. Using relaxation methods, it was found thatDNA is not damaged in blasts from acute myeloblastic leukemiain resting conditions."

All except the experimental drugs used in this study arecurrently used alone or in combination (47) for the treatment ofleukemias. These agents have been found to affect DNA metabolism. Carmustine is supposed to carbamoylate cellular macro-molecules (3), and this process could be responsible for theinhibition of ligation of X-ray-induced DNA strand breaks (1,13).

Chlorambucil determines sister chromatid exchange in humanlymphocytes (41), and cyclophosphamide induces DNA repair inhuman leukocytes (35). Adriamycin has formerly been reportedto be a noninducer of DNA repair synthesis (38), but more recentwork shows that it causes DNA-protein cross-links and DNAsingle- and double-strand breaks (42). Vincristine determines

sister chromatid exchange in human lymphocytes (42) and hasbeen reported not to induce (2) or to induce poorly, DNA repairsynthesis (40). Etoposide has been shown to cause single-strand

breaks rapidly repaired after drug removal (27, 48).The major antileukemic present therapeutic includes the drugs

studied in this report either alone or in combination (i.e., ara C +anthracycline in acute myeloblastic leukemia, prednisone plusVicia, or ara-C + etoposide alkaloids in acute lymphoblastic

leukemia). The fact that some antileukemic agents like vincristineand etoposide are inhibitors of DNA ligase from normal lymphoidcells without affecting the enzyme purified from leukemic source,and taking into account that this enzyme is required for bothreplication (36) and repair (9, 39), might be a crucial factor forthe chemotherapy of leukemia.

4J. G. David, B. Fedecka Bruner, Z. Mishal, D. Vinson, and C. Rosenfeld,unpublishedobservations.

CANCER RESEARCH VOL. 45 MAY 1985

2182

on May 31, 2020. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

EFFECT OF ANTILEUKEMIC DRUGS ON DMA LIGASE

REFERENCES

1. Bedford, P., and Eisenbrand,G. DNA damage and repair in the bone marrowof rats treatedwith fourchloroethylnitrosoureas.CancerRes., 44: 514-518,1984.

2. Benigni, R., Calcagnile,A., Oogliotti, E., Fanone,E., and Giuliani,A. DNArepairby cytostatic drugs in proliferating and quiescent MRC-5 cells. Teratog.Carcinog. Mutagen., 3. 481-490,1983.

3. Bowden, B. J., and Wheeler, G. P. Reaction of 1,3-bis(2-chlorotesyl)-1-nitro-sourea (BCNU)with protein. Proc. Am. Assoc. Cancer Res., 12: 67.1971.

4. Brookes, P.,andLawley, P. D. Evidencefor the bindingof polynucleararomatichydrocarbons to the nucleic acids of the mouse skin: relation between carcinogenic power of hydrocarbons and their binding to DNA. Nature (Lond.),202:781-784,1964.

5. Carre, D., Signoret, J., Lefresne,J., and David,J. C. Enzymesinvolved in DNAreplication in the Axolotl. I. Analysis of the forms and activities of DNApolymerase and DNA ligase during development. Dev. Biol., 87: 114-125,1981.

6. David, J. C. Purificationand properties of a soluble polynucleotideligase fromchick embryo. Biochimie(Paris),59: 723-728,1977.

7. David, J. C., Fedecka Bruner, B., Mishal, Z.. Vinson, D., and Rosenfeld, C.Differentiationof thymocytes during chicken ontogeny: occurrenceof a specificDNA ligase in relationship to cell size and surface antigens. Eur. J. Immunol.,77:593-596, 1981.

8. David, J. C., Zittoun, R., Bassez, T., Maniey, D., Rusquet, R., Bonhommet,M., Thevenin, D., Suberville, A. M., and Marie, J. P. DNA ligases in humanleukemias. Leuk. Res., in press, 1985.

9. Dean, C., and Pauling,C. Properties of a deoxyribonucleic acid ligase mutantof E coli: X ray sensitivity. J. Bacterio!., 702: 588-589, 1970.

10. Duguet, M., Lavenot, C., Harper, F., Mirambeau,G., and De Recondo, A. M.DNA topoisomerases from rat liver: physiological variations. Nucleic AcidsRes., 77: 1059-1075,1983.

11. Farber, S„Diamond, L. K„Mercer, R. D., Sylvester, R. F., and Wolff, J. A.Temporary remissions in acute leukemia in children produced by folie antagonist, 4-amino pteroylglutamic acid (Aminopterin).N. Engl. J. Med., 238: 787,1948.

12. Farber, S., Toch, R., Sears, E. M., and Pinkel, D. Advances in chemotherapyof Cancer in man. Adv. Cancer Res., 4:1-71,1956.

13. Fornace, A. J., Jr., Kohn, K. W., and Kann, H. E. Inhibition of the ligase stepof excision repair by 2-chloro-ethylisocyanate:a decomposition product of 1,3-bis(2-chloroethyl)-1-nitrosourea.Cancer Res., 38:1064-1069,1978.

14. Getter, M. L, Becker, A., and Hurwitz, J. The enzymatic repair of DNA. I.Formation of circular DNA. Proc. Nati. Acad. Sci. USA,58: 240-247,1967.

15. Geliert, M. Formation of covalent circles of lambda DNA by E. coli extracts.Proc. Nati. Acad. Sci. USA, 57: 148-155, 1967.

16. Oilman,A. The initial clinical trial of nitrogen mustard. Am. J. Surg., 705: 574,1946.

17. Oilman,A., and Philips,F. S. The biologicalactionsand therapeuticapplicationsof #-chloroethylaminesand sulfites. Science(Wash.DC), 703:409-410,1946.

18. Johnson, I. S., Armstrong, J. G., Gorman, M., and Burnett, J. P., Jr. The Vincaalkaloids:a new class of oncolytic agents. CancerRes., 23:1390-1427,1963.

19. Kellenberger,C. M., Zichichi, M., and Land Weigle.J. J. Exchangeof DNA inrecombination of bacteriophage. Proc. Nati. Acad. Sci. USA, 58: 869-875,1961.

20. Lehman, I. R. DNA ligase:structure, mechanismand function. Science(Wash.DC), 186: 790-797, 1974.

21. LePecq, J. B. Spécificitéd'actions des substances antitumorales. J. Pharma-col. (Paris), 73:53-75, 1982.

22. Li, M. C., Hertz, R., and Spencer, D. B. Effect of methotrexate upon chorio-carcinoma and chorioadenoma.Adv. Exp. Med. Biol., 84: 247-264,1956.

23. Liu, L. F. In: B. M. Alberts and C. F. Fox (eds.), Mechanistic Studies of DNAReplication and Genetic Recombination, pp. 817-831. New York: AcademicPress, 1980.

24. Lindahl, T. New class of enzymes acting on damaged DNA. Nature (Lond.),

259: 64-66, 1976.25. Lindahl,T. DNA glycosylases,endonucleasesfor apurinic/apyrimidicsites and

base excision repair. Progr. NucleicAcid. Res. Mol. Biol., 22:135-193,1979.26. Lindahl, T., and Edelman, G. M. Polynucleotide ligase from myeloid and

lymphoid tissues. Proc. Nati. Acad. Sci. USA, 67: 680-687, 1968.27. Loike, J. D., and Horwitz, S. B. Effect of VP16-253 on the ¡ntracellular

degradationof DNA in Hela cells. Biochemistry, 75: 5443-5448,1976.28. Marshall, B.. Darkin, S., and Ralph, R. K. Evidence that m-AMSA induces

topoisomerasesaction. FEBS Lett., 767: 75-78, 1983.29. McCormick, J. J., and Mäher,V. M. Role of DNA lesions and DNA repair in

mutagenesis and transformation of human cells. In: Human Carcinogenesis,pp. 415-429. New York: Academic Press, 1983.

30. Meselson,M., and Weigle,J. J. Chromosomebreakageaccompanyinggeneticrecombination in bacteriophage. Proc. Nati. Acad. Sci. USA, 47: 857-858,1961.

31. Mezzina, M., and Nocentini, S. DNA ligase activity in UV irradiated monkeykidney cells. NucleicAcids Res., 5: 4317-4327,1978.

32. Mezzina, M., Nocentini,S., and Sarasin,A. DNA ligase activity in carcinogen-treated human fibroblasts. Biochimie(Paris),64. 743-748,1982.

33. Mezzina, M., Suarez, H. G., Cassigema,R., and Sarasin,A. Increasedactivityof polynucleotideligasein 5'-iodo-2'-deoxymidine and mytomycinC pretreated

simian virus 40 (SV40) infected monkey kidney cells. Nucleic Acids Res., 70:5073-5084.

34. Modrich, P., and Lehman, I. R. Enzymatic joining of polynucleotides. IX. Asimple and rapid assay of polynucleotide joining ligase by measurement ofcircle formation from linear deoxyadenylate-deoxythymidylatecopolymer. J.Biol. Chem., 245: 3626-3634,1970.

35. Nordenstjöld,M., Molde, U. S. P., and Lambert, B. Effects of ultraviolet lightand cyclophosphamideon replication and repair synthesis of DNA in isolatedrat liver cells and humanleukocytesco-incubatedwith microsomes.Hereditas,89: 1-8, 1978.

36. Okazaki, R., Okazaki,T., Sakabe,K., Sugimoto. K., andSugino,A. Mechanismof DNAchaingrowth. I. Possiblediscontinuityand unusualsecondarystructureof newly synthesized chains. Proc. Nati. Acad. Sci. USA,59:598-605,1968.

37. Olivera. B. M. The DNA joining enzyme from E. coli. Methods Enzymol., 27:311-319,1971.

38. Painter,R. B. DNAsynthesis inhibitionin HeLacells as a simpletest for agentsthat damage human DNA. J. Environ. Pathol. Toxicol., 2: 65-78, 1978.

39. Pauling,C., and Hamm, L. Properties of temperature sensitive radiation sensitive mutant of £coli. Proc. Nati. Acad. Sci. USA,60: 1495-1502,1968.

40. Perocco, P., and Prodi, G. DNA repair in human lymphocytes after treatmentwith vincristine, chlorambucil and cyclophosphamidein vitro. Hematologica,67:522-529,1982.

41. Raposa, T. Sister chromatid exchange studies for monitoring DNA damageand repair capacity after cytostatics in vitro and in lymphocytes of leukemicpatients under cytostatic therapy. Mutât.Res., 57: 241-247,1978.

42. Ross, W. E., and Bradley,M. O. DNAdouble strand breaks in mammaliancellsafter exposure to intercalating agents. Biochem. Biophys. Acta, 654: 129-134, 1981.

43. Saucier,J. M., and Laval.F. DNAligaseactivity in crude extracts of fibroblastsand lymphocytes. Biochem. Biophys. Res. Commun., 776: 657-662,1983.

44. Soderhall. S. DNA ligase during rat liver regeneration. Nature (Lond.), 260:640-643,1976.

45. Soderhall, S., and Lindahl, T. DNA ligases of eukaryotes. FEBS Lett., 67: 1-8,1976.

46. Teraoka, H., and Tsukada, K. Eucaryotic DNA ligases. J. Bid. Chem., 257:4758-4763,1982.

47. Tirelli, U., Crivellar!,D., Carbone,A., Veronesi, A., Galligioni,E., Trovo, M. G.,Turnólo,S., andGrigoletto,E. Combinationchemotherapyfor multiplemyelomawith melpnalan, prednisone, cyclophosphamide, vincristine and carmustine(BCNU)(M2 PROTOCOL).CancerTreat. Rep., 66: 1971-1973, 1982.

48. Wozniak, A. J., and Ross, W. E. DNA damage as a basis for 4'-demethylepi-podophyllotoxin-9-(4,6,O-ethylidene-#-D-glucopyramoside)etoposide toxicity.Cancer Res., 43: 120-124, 1983.

CANCER RESEARCH VOL. 45 MAY 1985

2183

on May 31, 2020. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1985;45:2177-2183. Cancer Res   Jean-Claude David, Thérèse Bassez, Micheline Bonhommet, et al.   or Leukemic Lymphocytes by Antileukemic DrugsInhibition of DNA Ligase from Human Thymocytes and Normal

  Updated version

  http://cancerres.aacrjournals.org/content/45/5/2177

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/45/5/2177To request permission to re-use all or part of this article, use this link

on May 31, 2020. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from