tiazofurin metabolism in human lymphoblastoid cells ... · 286193), an analogue of the antiviral...

[CANCER RESEARCH 46, 532-537, February 1986]

Tiazofurin Metabolism in Human Lymphoblastoid Cells: Evidence forPhosphorylation by Adenosine Kinase and 5'-Nucleotidase1

Arnold Fridland,2 Michele C. Connelly, and Terry J. Robbins

Department of Biochemical and Clinical Pharmacology, St. Jude Children 's Research Hospital, Memphis, Tennessee 38101

ABSTRACT

The exact route of metabolism of tiazofurin, a novel nucleosidewith antitumor activity, is controversial. Using human cell linesseverely deficient in salvage nucleotide enzymes, we were ableto identify the route of activation in tiazofurin metabolism. Withloss of adenosine kinase activity by mutation in two lymphoblas-toid cell lines, CCRF-CEM and WI-L2, the growth sensitivity totiazofurin decreased by 6- and 3-fold, respectively. In contrast,the mutant lines were about 3000- to 1500- and 16- to 4-fold

more resistant to the structurally similar tiazofurin analoguespyrazofurin and ribavirin, respectively. Other mutants with defective deoxycytidine or uridine kinase activity showed normal sensitivity to all three analogues. Both cell lines with defectiveadenosine kinase activity accumulated about 50% wild-typelevels of tiazofurin-5'-monophosphate and thiazole-4-carbox-

amide adenine dinucleotide analogue of tiazofurin at cytotoxicconcentrations of the drug. Extracts of wild-type lymphoblasts

catalyzed the phosphorylation of tiazofurin in the presence ofadenosine 5'-triphosphate and Mg2*. Loss of adenosine kinase

activity in the mutant extract eliminated this phosphorylatingactivity for tiazofurin consistent with the notion that adenosinekinase catalyzes phosphorylation of tiazofurin. However, an enzyme activity that catalyzed the phosphorylation of tiazofurin inthe presence of inosine-5'-monophosphate as donor and Mg2+

was detected in the extracts of both wild-type cells and adenosine kinase-deficient mutants. The monophosphate donor specificity, divalent metal, high salt requirement, and nucleoside acceptor specificity of this enzyme activity paralleled that of a 5'-

nucleotidase (EC 3.1.3.5) which catalyzes inosine phosphorylation. In addition, tiazofurin phosphorylation was competitivelyinhibited by inosine and the apparent KÂ¡value was similar to theapparent Km value for inosine phosphorylation. These resultsindicate that two enzymes, adenosine kinase and a cytoplasmic5'-nucleotidase, are functionally important anabolizing enzymes

for tiazofurin in human cells.

INTRODUCTION

Tiazofurin or 2-D-ribofuranosylthiazole-4-carboxamide (NSC

286193), an analogue of the antiviral and antitumor agentsribavirin and pyrazofurin, respectively, has displayed potent activity against murine leukemias, Lewis lung adenocarcinoma, andis undergoing Phase I clinical trials in humans (1,2). Cells culturedin tiazofurin show inhibited growth due to accumulation of two

Received7/19/85; revised 10/15/85; accepted 10/17/85.The costs of publication of this article were defrayed in part by the payment of

page charges. This article must therefore be hereby marked advertisement inaccordancewith 18 U.S.C. Section 1734 solely to indicate this fact.

1This work was supported by a grant from the National Cancer Institute, CA33017, and by the American LebaneseSyrian Associated Charities.

2To whom requests for reprints should be addressed.

major metabolites, tiazofurin-5'-monophosphate and a dinucleotide analogue of NAD, termed TAD,3 in which the nicotinamide

moiety is replaced by tiazofurin. In mammalian cells, the effectsof tiazofurin have been attributed to an inhibition of guanosinenucleotide synthesis by the accumulation of a sufficient concentration of TAD to inhibit the rate limiting enzyme inosine mono-phosphate dehydrogenase (IMP:NAD+ oxidoreductase, EC

1.1.1.205) (3-5).

The molecular details of how tiazofurin is activated in cellshave not been reported. Some investigators have shown thattiazofurin is taken up and activated to its respective nucleotidesin different cell types and tissues, including human erythrocytes(6), mammalian skeletal muscle, kidney (4,7), and various mouseand human tumors (8, 9). In theory, tiazofurin activation couldbe mediated by one or more cellular kinases that phosphorylatestructurally similar analogues. This could presumably occurthrough the action of adenosine kinase (ATP:adenosine-5'-phos-

photransferase, EC 2.7.1.20) which is primarily responsible forphosphorylation of ribavirin and pyrazofurin (10, 11). However,Saunders ef al. (11) in studies with Chinese hamster ovary cellscould find no evidence for phosphorylation by this pathway orby other anabolic pathways for deoxycytidine or purine deoxy-

nucleosides.In the present study, we have examined the cytotoxicity and

metabolism of tiazofurin in two different human lymphoblastoidcell lines and their sublines with specific deficiencies in salvagenucleotide pathways and compared the cytotoxicity of this drugto that of ribavirin and pyrazofurin. A preliminary account ofthese results has appeared elsewhere (12).

MATERIALS AND METHODS

Materials. Tiazofurin and TRMP were provided by Dr. Robert Engleof the Developmental Therapeutics Program of the Division of CancerTreatment, National Cancer Institute. TAD was kindly supplied by Dr.Robert Jackson (Wamer-Lambert/Parke-Davis, Detroit, Ml). Dr. Roland

K. Robins, University of Utah, generously provided ribavirin.Nucleosides and nucleotides were purchased from Sigma Chemical

Co. (St. Louis, MO). [5-3H]Thiazole riboside (635 mCi/mmol) was ob

tained from Research Triangle Institute, Research Triangle Park, NC,and was more than 98% pure. [2,8,5-3H]Adenosine (51 Ci/mmol) was

purchased from New England Nuclear Corp., Boston, MA.Cell Lines. The human splenic B-lymphoblastoid cell line WI-L2, and

its two variants WI-L2/dCK" and WI-L2/dCK"AK" were obtained from

Dr. M. S. Hershfield (Duke University, Durham, NC) and were initiallydescribed by Ullman e? al. (13). The human lymphoblastoid cell linesCCRF-CEM and its sublines Py9 and CAR-1 were previously described(14) and RPMI 6410 were from Dr. Arnold Welch (St. Jude Children's

Research Hospital, Memphis, TN).In general, all cell lines were cultured at 37Â°Cin 75-cm2 tissue culture

3The abbreviationsused are: TAD, thiazole-4-carboxamidedinucleotide;TRMP,tiazofurin 5'-monophosphate; HPLC, high pressure liquid chromatography; DTT,dithiothreitol; Ado, adenosine; ICÂ»,concentration causing 50% growth inhibition.

CANCER RESEARCH VOL. 46 FEBRUARY 1986

532

Research. on October 18, 2020. © 1986 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

ACTIVATION OF TIAZOFURIN IN HUMAN LYMPHOBLASTS

flasks (Costar 3275; BÃ©licoGlass, Inc., Vineland, NJ) in high humidity,and 5% CO2 in air. The culture medium was composed of RPMI 1640medium supplemented with 2 miuiglutamine (Grand Island Biological Co.,Grand Island, NY), sodium bicarbonate (2.2 g/liter), penicillin (60 ÃŸQ/m\),streptomycin (100 Â¿ig/ml),and 10% heat-inactivated newborn calf serum(Flow Laboratories, Rockville, MD). For WI-L2 cultures, 10% heat-inac

tivated horse serum (Flow Laboratories) was used instead of newborncalf serum. Culture conditions for CEM cells have been described (14).

Cell cultures and serum were tested and found to be free of myco-

plasma and adenosine phosphorylase activity by methods previouslydescribed (15). The average cell volume of exponentially growing cellswas determined by a Coulter Counter (Model Z) calibrated with mono-

sized latex particles.Growth Inhibition Studies. Drug sensitivity studies were conducted

in 25-cm2 tissue culture flasks (Costar 3050). Small volumes (10-200 ^l)

of growth inhibitory agent were pipetted into each flask after which 5 mlof cells (0.1-0.2 x 106 cells/ml) in complete medium was added. After

48 h the number of untreated cells typically increased from 4- to 8-fold.

The initial cell densities were subtracted from the final cell densities, andthe number of cells in the flasks containing the growth inhibitory nucleo-

side was calculated as a percentage of the number of untreated cells incontrol flasks. The IC60was determined from semilog graphs of percentage of cell growth versus drug concentration.

Measurement of Uptake of Radiolabeled Substrates by Cells. Cellswere collected from exponential phase cultures, suspended to a densityof 1.0 x 106 cells/ml in growth medium, and incubated with tritiatedtiazofurin. The suspension was incubated at 37Â°Cand at specified times.Approximately 107 cells were separated from the culture medium by

centrifugation and the cell pellet was resuspended in 0.25 ml of cold 0.5N perchloric acid and transferred to 1.5-ml Microfuge tubes. Ten minlater the acid-insoluble precipitate was removed by centrifugation andthe supernatant, containing the acid-soluble nucleotides, was neutralized

by addition of 12.5 n\ 1 M potassium phosphate, pH 7.5. The resultingpotassium perchlorate was removed by centrifugation and the supernatant was stored at -20Â°C until HPLC analysis. Tritiated tiazofurin nu

cleotides were quantitated by HPLC on a Partisil-10 SAX column (4.6 x

250 mm; Whatman, Inc.). A mobile solvent consisting of 5 rriM potassiumphosphate, pH 4.0, was used to elute the monophosphates at a flowrate of 1.5 ml/min in 18 min followed by a linear gradient over 40 min to0.5 M potassium phosphate, pH 4.O. Fractions were collected every 0.4min and analyzed by liquid scintillation counting. Each tritiated peak wasidentified by comparing its retention time with a known standard; TRMPand TAD had respective retention times of 17 and 51 min.

DEAE-Cellulose Chromatography. Ado kinase was purified from 2'-

deoxycitidine kinase and 2'-deoxyadenosine-2'-deoxyguanosine kinase

activities by DEAE-cellulose Chromatography (DE23; Whatman, Inc.). Asonically disrupted extract of about 6 x 1010CEM cells was centrifugea

at 100,000 x g for 1 h; the supernatant was dialyzed for 36 h againstthree changes of 10 vol of column buffer (50 mw Tris-HCI, pH 7.52-2 mwDTT); and the lysate was applied at 4Â°C to a preequilibrated DEAE-

cellulose column (2.9 x 45 cm) and washed through the column withbuffer to elute unbound protein. Ado kinase activity was eluted with KCIat a concentration of about 0.1 M. The fraction containing enzyme activitywas pooled, concentrated by pressure dialysis on Amicon YM 10 membranes, and stored in 50 mw Tris-HCI, pH 7.5, 2 mM DTT, and 10%glycerol at -20Â°C until assayed. By this procedure, Ado kinase was

purified 20-fold with 85-90% recovery.

Enzyme Assays. The standard assay for nucleoside phosphorylationwas essentially that described previously (14). For tiazofurin phosphorylation, dialyzed extracts (400 ^g protein) or purified Ado kinase (200ng) was incubated with 1 mw [3H]tiazofurin (0.5-1.0 nC\), 50 mM MgCI2,

100 mw Tris-HCI (pH 7.0), 3 mM DTT, 10 mw ATP, 15 mw pyruvate

phospho(enol). and 8 units pyruvate kinase. In the assay containing IMPas phosphate donor the incubations were carried out with the same Tris-

HCI buffer, pH 7.0, 50 mw MgCI2, 500 mw KCI, and 1 mm IMP and[3H]tiazofurin at the concentration above. The reactions (100 pi) were

initiated by the addition of enzyme and after various times at 37Â°C,they

were terminated by adding 20 Â¡Aof the reaction mixture to 80 /il of ice-cold water. A portion (50 ^l) of this mixture was applied to a DE-81 disk

(Whatman), washed with water, and dried. The filter was then extractedwith 1 ml 0.1 M HCI and 0.2 M KCI. After 15 min the eluate was collectedin a scintillation vial, 7 ml of scintillant was added, and radioactivity wasdetermined as described earlier. Recovery of nucleotides by this procedure was >90%.

RESULTS

Cross-Resistance. We first compared the growth inhibitoryeffects of tiazofurin, pyrazofurin, and ribavirin in human lympho-

blasts and their sublines deficient in specific enzymes of thenucleotide salvage pathways. Table 1 presents a summary ofthe IC50values of the 3 drugs for wild type and 5 different cloneswith various kinase deficiencies. Two of these, Py9 (Ado kinasedeficient) and WI-L2/dCK"AK" (Ado kinase and 2'-deoxycytidine

kinase deficient), respectively, required 6- and 3-fold more tia

zofurin for 50% growth inhibition than did their wild type counterparts. By contrast, the 3 lines with deficiency in 2'-deoxyciti-dine kinase (CAR-1, WI-L2/dCÂ«-) or uridine kinase (6410/UK~)

showed normal sensitivity to tiazofurin. Both mutants, Py9 andWI-L2/AK~dCK" had IC50 values for pyrazofurin and ribavirin

1500- to 3000-fold and 4- to 16-fold, respectively, higher thanthat for wild-type cells. All these mutants showed a normalsensitivity to various other nucleosides, such as 5-fluorodeox-yuridine or thymidine, suggesting that these cells had not undergone a generalized transport defect for nucleosides.

Phosphorylation of Tiazofurin. To further examine the role ofAdo kinase in the activation of tiazofurin, its metabolism to theputative active nucleotides was studied. Cells (wild type andmutants) were incubated with radiolabeled tiazofurin and atvarious times the resulting nucleotides were analyzed by HPLC.In agreement with previous studies (16-18), tiazofurin was in

corporated into the nucleotides with accumulation mainly inTRMP and TAD (data not shown). The amount of the nucleotidesformed was found to be proportional to the drug concentrationand time of incubation (up to 6 h). Both TRMP and TAD accumulation did not seem to be saturable at least up to 500 UMexogenous tiazofurin. In the 2 mutants Py9 and WI-L2/AK"-dCK" the levels of both TRMP and TAD was significantly de

creased from that in their respective wild type. Thus, at 40 ^M

Table 1Growth inhibition of analoguesin human lymphoblastoid cells

Experiments to determine the inhibitory effects of the analogueson the growthof lymphoblastoidcells are described in "Materials and Methods."

Concentrationof inhibitor (^M)required for 50% growth inhibition

CelllineCCRF-CEM

Py9CAR-1WI-L2WI-L2/dCK-WI-L2/dCK-,AK-

64106410/UK-Tiazofurin34

Â±3.1a(1)"

200 Â±16(6)24 Â±2.9 (0.7)4.8 Â±0.8(1)4.5 Â±1.0(1)

14.7 Â±2.6(3)13 Â±2.7(1)

17.5 Â±3.8(1.3)Pyrazofurin0.05

Â±0.02(1)150 Â±44(3000)

0.07 Â±0.01 (1.4)0.09 Â±0.02(1)0.08 + 0.03(1)135 Â±29(1500)

NDNDRibavirin180

Â±22(1)700 Â±51 (4)

NDC

95 Â±11(1)100 Â±17(1)

1500 Â±182(16)NDND

a Mean Â±SD." Numbers in parentheses, relative resistance, which is expressed as the ratio

of the ICÂ»value for the mutant to that of wild-type cells.0 ND, not determined.


533




tiazofurin the level of TRMP and TAD was reduced to approximately 50% of wild type in both Py9 and WI-L2/dCK-AK~ (Fig.

1). At 4 UM drug no accumulation of nucleotide analogue couldbe detected in the kinase-deficient mutants. With 500 /Â¿Mexog

enous tiazofurin the level of TRMP and TAD accumulation wasapproximately 70-80% of that of wild type levels (data not

shown). In parallel experiments no appreciable differences wereobserved in tiazofurin anabolism between wild type and either2'-deoxycitidine kinase or uridine kinase-defective mutants (data

not shown). These results indicate that tiazofurin-induced growthinhibition in these human cells is dependent on its phosphoryla-

tion via adenosine kinase.To demonstrate further that tiazofurin is a substrate for aden

osine kinase, the enzyme was partially purified from CEM lym-phoblasts by DEAE-cellulose column chromatography. Followingapproximately 20-fold purification of the enzyme activity andseparation from the kinase activity for 2'-deoxycitidine, 2'-deox-yadenosine, 2'-deoxyguanosine, and thymidine, tiazofurin phos-

phorylation remained associated with Ado kinase activity (datanot shown). As shown in Fig. 2, with fixed ATP and Mg2+ ions,

tiazofurin gives a competitive type inhibition of Ado phospho-

rylation. The kinetic constants for the enzyme of the variousnucleosides are listed in Table 2. These values were determinedfrom double reciprocal plots of initial velocity versus variousnucleoside concentrations at a constant excess of ATP. Theratio of VmaxtoKmwas used to assess overall substrate efficiency.Adenosine was the most efficient substrate for Ado kinasefollowed by pyrazofurin, tiazofurin, and ribavirin in order of decreasing efficiency.

The aforementioned results showed that the elimination of Adokinase activity from both human lymphoblastoid mutant linesonly reduced partially the cytotoxicity and anabolism of tiazofurin,suggesting that this enzyme is not solely responsible for thecytotoxic activation of the drug. Assays of nucleoside phospho-rylating activities in the kinase-deficient subline with ATP as the

phosphate donor disclosed that the phosphorylating activity fortiazofurin was decreased by at least 90% from wild type (Fig.3A). By contrast, both wild-type and kinase-deficient extracts

phosphorylated tiazofurin in the presence of a monophosphate

â€¢5e.

g 20

10 -

TR-5'-P4

â€¢WTaâ€¢AK-C"WT.

O'AK'-aCK'C1*1uTADCEM

WI-L2A"*CT1BnTiDÃ±CEMWI-L2

Fig. 1. Accumulation of total tiazofurin nucleotide derivatives by wild-type andmutant lymphoblasts. Cultures (approximately 2 x 107 cells) were incubated for 6

h with 40 fM tiazofurin (20 //Ci) and the accumulation of nucleoides was analyzedas described in "Materials and Methods." The data are the average of threeseparate experiments. WT, wild type; TR-5'-P, tiazofurin 5'-monophosphate; bars,

SD.

f 6

?. 2

0"0 0.5 IO 1.5 2.0

I/Concentration of Ado (jjM~')

Fig. 2. Lineweaver-Burk plots showing competitive inhibition of adenosine phosphorylation by tiazofurin. The rate of phosphorylation of Ado was measured underthe assay condition specified under "Materials and Methods," except that the

substrate concentration was varied as indicated. At each concentration of substrate, incubations were performed for 15 and 30 min in duplicate. The amount ofenzyme (approximately 200 ng of partially purified protein) was adjusted so thatless than 20% of the substrate was used. In all experiments the amount ofphosphorylation was proportional to time over the assay period.

Table 2Kinetics of adenosine kinase binding with nucleosides

The Km value for Ado kinase was measured by the standard assay at variousconcentrations of the substrate with fixed 10 ITIMATP and 50 mw MgCI2 present,and about 200 ng of partially purified enzyme protein per reaction mixture. Otherconditions were as those described under "Materials and Methods."

RelativevelocitySubstrateAdo

TiazofurinPyrazofurinRibavirinKm(M)2.1

x1fr46.1x irr3

2.5x irr310.7x irr3pmol/min9240

18.3165

2.1%100

0.0020.0170.0002

30 60 0 30

Incubation Time (min}

Fig. 3. Effect of ATP and IMP on tiazofurin phosphorylation in extracts of wild-type and mutant lymphoblasts. Tiazofurin phosphorylation was measured as described in 'Materials and Methods" with either 10 mM ATP and 50 mM MgCI2 (A)

or 1 mw IMP and 50 mw MgCt (B) in the reaction mixtures. The velocity isexpressed as nmol tiazofurin phosphorylated/mg protein. Each reaction containedapproximately 400 Â¿igof protein extract/assay, and standard assay conditionswere used. WT, wild type.

such as IMP as donor (Fig. 3B). In addition to the monophosphatedonor, Mg2+ ions were essential for this phosphorylating activity

for tiazofurin and KCI stimulated tiazofurin phosphorylation atleast 2- to 3-fold (data not shown). The products of the reaction

were analyzed by HPLC and identified as the monophosphateby comparison with a standard; no dinucleotides or higher phosphates were detected.

To characterize further the nature of the latter phosphorylatingactivity, we examined its phosphate donor specificity. As shownby the results in Table 3, the nucleoside monophosphate thatacted as phosphate donor for tiazofurin phosphorylation included


534




TablesDonor specificity for tiazofurin phosphorylation in cell-free extracts of Py9

The reactions were performed as described in "Materials and Methods," with

use of 400 Â»gprotein extract and [3H]tiazofurin as substrate in the assay mixture

supplemented with the indicated compounds. Specific activity was calculated asnmol of reaction product/min/mg of protein and expressed as the percentages ofactivity when 1 /iw IMP is used as the donor.

Donor (1mw)GMPAMPUMPCMPTRMPRibavirin

5'-monophosphateP,PP,2-Glycerophosphatep-NitrophenylpnosphateActivity

(% ofIMP)73.75.016.44.038.03.00000

Table 4Effect of nucleosides on IMF-dependent tiazofurin phosphorylation in

CEM cell extractsCell-free extracts of CEM cells served as the source of enzyme. IMP (1 mM)

served as the donor and other conditions for the assay were as described in"Materials and Methods." Specific activity of the products formed was calculated

as nmol/min/mg of protein and is expressed as a percentage of the control activitywithout nucleoside inhibitor.

Inhibitor (5 mM)

Activity(% control

activity)

InosineAdenosine + e/yi/iro-9-<2-hydroxy-

3-nonyl)adenine

UndineCytidineThymidinePyrazofurinRibavirinSelenazofurin

26.971.0

60.566.091.669.858.032.0

GMP, UMP, and TRMP. Compounds such as AMP, dTMP, andribavirin 5'-monophosphate were not active in this reaction.

Moreover, as shown in Table 3, p-nitrophenylphosphate, 2-

glycerophosphate, PÂ¡,or PP, could not substitute as the phosphate donor.

We also assessed the effect of various nucleosides on tiazofurin phosphorylation in the extracts of Py9 with IMP as thedonor. As shown by the results in Table 4, inosine caused asubstantial reduction in the phosphorylation of the analogue. Theother compounds including Ado [with e/ytf?ro-9-(2-hydroxy-3-

nonyl)adenine], uridine, cytidine, or thymidine inhibited muchmore weakly. Similarly, the analogues pyrazofurin and ribavirincompeted very poorly with tiazofurin for phosphorylation whilethe seleno derivative selenazofurin inhibited about the same asinosine.

The effect of inosine on tiazofurin phosphorylation was examined further to determine the type of interaction involved. Adouble reciprocal plot of tiazofurin phosphorylation was linearover the substrate concentration range used and the intercepton the ordinate indicated competitive inhibition (Fig. 4). A K, valueof 7.7 mM for inosine (Fig. 4, inset) and a Km of 16.5 mw fortiazofurin were calculated.

DISCUSSION

Tiazofurin is a novel C-nucleoside with the ability to induce

cellular differentiation (18) and an antiproliferative agent with

02468

Concentration oftnosine (mM)

0.25 0.50 075 100I/Concentrationof TR (mM"')

Fig. 4. Lineweaver-Burk plot showing inhibition of tiazofurin (TR) phosphorylation by inosine. Tiazofurin phosphorylation was measured as described in 'Materialsand Methods" with 1 mr.i IMP and 50 rnw MgCI2 in the reaction mixture (A); slope

replot of the data shown in the inset (B). Each reaction contained approximately400 ng protein from cell extract.

potent antitumor activity both in vitro and in animal tumor modelsystems. This broad spectrum of biological activities suggeststhat the compound would be effective in patients with cancer.To data, reported clinial studies of tiazofurin have been limitedto Phase I pharmacokinetic evaluations (1, 2). Since phosphorylation of tiazofurin appears essential for its antineoplastic andcytotoxic activities, it was of interest to identify the enzymepathway for activation of the compound.

Previous reports (7, 11) have raised the question of whethertiazofurin is a substrate for nucleoside kinases involved in purinemetabolism. We have attempted to resolve this issue by examining mutants of cultured human lymphoblastoid cells deficientin various enzymes of nucleoside salvage. The mutants werestudied by comparing their growth in the presence of exogenoustiazofurin and measuring their drug-phosphorylating activity in

intact cells and cell extracts. The results of these experimentsshow that in both the CCRF-CEM and WI-L2 lymphoblastoid

lines Ado kinase is required in tiazofurin metabolism. Furtherevidence for a role of Ado kinase in tiazofurin phosphorylationwas provided by investigation with the analogue with the use ofpartially purified enzyme from human lymphoblasts. To beginwith, the two activities for tiazofurin and Ado copurified on ion-

exchange column chromatography. In addition, Ado phosphorylation by the purified enzyme was competitively inhibited bytiazofurin and the apparent inhibition constant (apparent K) beingvery similar to the apparent Kni value for tiazofurin indicate thatthe two compounds share the same binding site.

However, the individual blockade of metabolism via Ado kinasereduced but did not abolish the cytotoxic activation of tiazofurinin the human lymphoblastoid cell lines. This suggested that thekinase is not the sole enzyme involved in the phosphorylation oftiazofurin. Other mutant lymphoblastoid cell lines with functionally defective deoxycytidine kinase or uridine kinase activitymetabolized tiazofurin to its monophosphate form and TAD atnormal rates, and tiazofurin did not compete in kinase reactionswith thymidine or purine deoxyribonucleosides (data not shown).The present experiments have also revealed that extracts of thehuman lymphoblastoid cells catalyze the phosphorylation of tiazofurin in the presence of either high-energy phosphoryl ATP orlow-energy nucleoside monophosphate donor. These two activ-


535




Â¡tiesare distinct because the mutants lacking the Ado kinaseactivity retain only the latter phosphorylation reaction for tiazc-

furin (Fig. 4).Low-energy phosphotransferase has been found in microor

ganisms (19, 20), plants (21), birds (22, 23), and some mammalian tissues (24), but the activity described here appears to bedifferent in several of its catalytic properties. It reacted preferentially with purine nucleoside monophosphates IMP and GMPas phosphate donor but not with CMP, dTMP, or AMP, whichare effective donors for the previously described pnosphotrans-ferases. Compounds such as p-nitrophenylphosphate, glycero-

phosphate, or ribose phosphate are totally inactive as donors asare nucleoside diphosphates and triphosphates. The efficiencytoward the sugar moiety of the nucleotide is less strict. 2'-

Deoxyribose, ribose, and arabinose nucleosides of guanine areactive as donors for tiazofurin phosphorylation. The presentenzyme activity absolutely requires a divalent metal and unlikethe phosphotransferase proteins described in other systems isstrongly stimulated by various salts including KCI and NaCI.

The properties of the tiazofurin phosphorylation activity of thehuman lymphoblastoid cells most clearly resemble those of acytoplasmic 5'-nucleotidase (5'-ribonucleotide phosphohydro

lase, EC 3.1.3.5) previously purified from rat liver (25-27). The

latter enzyme also reacts preferentially with IMP and GMP, isinactive with compounds such as p-nitrophenylphosphate or

glycerophosphate, and requires bivalent metal and high salt foroptimal activity (26). In other recent studies Worku and Newby(27) and Keller ef al. (28) have found that this enzyme will alsoact as a phosphotransferase to phosphorylate inosine and guan-osine and its analogues 9-/8-D-arabinofuranosylguanine and acy-

clovir in the presence of a suitable monophosphate donor. In thisrespect it should be noted that inosine is the most effectivecompetitive inhibitor of the phosphorylation of tiazofurin in theextracts of the lymphoblastoid cells with a KÂ¡of 7.7 ITIM(Fig. 4).This value can be compared with a Km or 6.8 ITIMreported forinosine phosphorylation for the purified nucleotidase of rat liver(27). It seems reasonable therefore to conclude that tiazofurinphosphorylation observed in the Ado kinase-deficient mutant

lymphoblasts is mediated through this or a similar enzyme activity.

Tiazofurin phosphorylation by either of the two enzyme activities reported herein was inefficient; nevertheless, the level ofactivity determined in the extracts appears sufficient to accountfor the amount of tiazofurin nucleotides detected in the lymphoblastoid cells. Thus, CEM cells treated with 40 fiu tiazofurin for6 h anabolized about 11 pmol/106 cells of tiazofurin (Fig. 1). The

total amount of tiazofurin phosphorylating activity in extractsfrom the same cells in this study was about 1.6 pmol/min/106

cells assayed with 1 HIMtiazofurin. Extrapolation to the conditionof 40 nu drug results in an activity (65 nmol/min/106 cells) that

could potentially phosphorylate about 23 pmol of tiazofurin in 6h, enough to account for that accumulated in these cells.

The present experiments have raised a number of questions,and obviously further experiments are needed to characterizethe specificity of cellular nucleotidase for tiazofurin phosphorylation. Both murine and human tumor cells have shown a widedifference in their capacities to accumulate the putative activemetabolite TAD (9, 29). So far, the basis for this difference hasnot been elucidated, although it has been proposed that greaterrates of catabolism of TAD in resistant versus sensitive cells

could be a basis for the difference in sensitivities observed (29).However, differences found in the catabolic activity for TAD wasonly about 4-fold, which appears to be insufficient to account forthe 20- to 50-fold difference in rate of TAD accumulation ob

served in these cells. Another possibility would be that thereexists in sensitive and resistant cells a difference in the efficiencyof either of the two enzyme pathways reported here to converttiazofurin to the nucleotide derivatives. Both Ado kinase and 5'-

nucleotidase are regulated by nucleotide effectors, and differentintracellular nucleotide levels could strongly affect their activityin cells. Characterization of both Ado kinase and nucleotidasefrom both sensitive and resistant cells should be valuable toevaluate this potential mechanism.

ACKNOWLEDGMENTS

We thank Dr. Jim Fyfe of Wellcome Research Laboratories for helpful discussions and communicationof unpublishedresults on S'-nudeotidase.

REFERENCES

1. Trump, D. L, Tutsch, K. D., Koeller, J-M., and Tormey, D. Phase I clinicalstudy with pharmacokinetic analysis of 2-Â¡;-D-ribofuranosylthiazo!e-4-carbox-amide (NSC 286193) administered as a five-day infusion. Cancer Res., 45.2853-2858,1985.

2. Melink, T. J., Van Hoff, D. D., KÃ¼hn,J. G., Hersh, M. R., Stemson, L. A.,Patton, T. F., Siegler, R., BotÃ³t,A. H., and Glauck, G. M. Phase I evaluationand pharmacokinetics of tiazofurin (2-f)-D-ribofuranosylthiazole-4-carboxam-ide, NSC 286193). Cancer Res., 45: 2859-2865,1985.

3. Carney, D. N., Ahluwalia,G. S., Jayaram, H. N., Cooney, D. A., Minna, T. D.,and Johns, D. G. Biochemical correlates of donai growth inhibition humantumor cell lines to tiazofurin. Clin. Res., 31: 507A, 1983.

4. Jayaram, H. N., Dion, R. L., Glazer, R. I., Johns, D. G., Robbins, R. K.,Serivastura, P. C., and Cooney, D. A. Initial studies on the mechanism ofaction of a new oncolytic thiazote nucleoside, 2-/3-D-ribofuranosylthiazole-4-carboxamide(NSC286193). Biochem. Pharmacol.,37:2371-2380,1982.

5. Lui, M. S., Faderan,M. A., Liepuieks,J. J., Natsumeda,Y., Olah, E.. Jayaram,H. N., and Weber, G. Modulationof IMPdehydrogenaseactivity and guanylatemetabolism by tiazofurin (2-,>'-o-nbofuranosylthiazole-4-carboxamide).J. Bio!.Chem..259. 5078-5082,1984.

6. Kovach, J. S., McGovem, R., Rubin, J., Richardson, R. L., and Schut, A. J.Rasma and red blood cell (RBC) pharmacokinetics of tiazofurin (T) andalterations in RBC nucleotides after intravenous T in man. Proc. Am. Assoc.Cancer Res., 25:344A, 1984.

7. Arnold, S. T., Jayaram, H. N., Harper, G. R., Litterst, C. L., Malspeis, L.,DeSouza,J. J. V., Staubus, A. E., AttJuwalia,G. S., Wilson, Y. A., Cooney, D.A., and Johns, D. G. The disposition and metabolismof tiazofurin in rodents,rabbits, and dogs. Drug Metab. Dispos., 12:165-173,1984.

8. Boritzki, T. J., Berry, D. A., Besserer, J. A., Cook, B. D., Fry, D. W., Leopold,W. R., and Jackson, R. C. Biochemicaland antitumor activity of tiazofurin andits seleniumanalog (2-/i-D-ribofuranosyl-4-selenazolecarboxamide).Biochem.Pharmacol.,34:1109-1114,1985.

9. Carney, D. N., Ahluwalia,G. S., Jayaram, H. N., Cooney, D. A., and Johns, D.G. Relationshipsbetween the cylotoxicity of tiazofurin and its metabolism bycultured human lung cancer cells. J. Clin. Invest., 75:175-182,1985.

10. Cadman, E. C., Dix, D. E., and Handschumaker,R. E. Clinical, biological,andbiochemicaleffects of pyrazofurin. Cancer Res., 38: 682-688,1978.

11. Saunders, P. P., Kuttan, R., Lai, L. M., and Robins, R. K. Action of 2-ÃŸ-o-ribofuranosylthiazole-4-carboxamide(tiazofurin)in ChineseHamsterovary andvariant cell lines. Mol. Pharmacol.,23: 534-539,1983.

12. Fridland,A., and Connelly, M. Adenosmekinase deficiency in cultured humanlymphoblasts resistant to tiazofurin. Proc. Am. Assoc. Cancer Res., 26:19A,1985.

13. miman, B., Levinson. B. B., Hershtield, M. S., and Martin, D. W., Jr. Abiochemicalgenetic study of the role of specific nucleosidekinases in deoxy-adenosinephosphorylationby cultured humancells. J. Biol. Chem.,256: 848-852,1981.

14. Verhoef, V., Sarup, J., and Fridland, A. Identification of the mechanism ofactivation of 9-fi-D-arabinofuranosyladeninein human lymphoid cells usingmutants deficient in nucleosidekinases.Cancer Res., 41:4478-4483,1981.

15. Verhoef, V., Germain, G., and Fridland, A. Adenosmephosphorylaseactivityin Mycoplasma-ireegrowth media for mammaliancells. Exp. Cell Res., 749:37-44,1983.

16. Earie, M. F., and Glazer, R. F. Activity and metabolismof 2-0-o-ribofuranosyl-


536




thiazole-4-carboxamide in human lymphoid tumor cells In culture. Cancer Res.,43: 133-137,1983.

17. Cooney, D. A., Jayaram, H. N., Gebeyehu. G., Betts, G. R., Kelly, J. A.,Marquez, V. E., and Johns, D. G. The conversion of 2-/3-o-ribofuranosylthia-zole-4-carboxamide to an analogue of NAD with potent IMP dehydrogenase-inhibitory properties. Biochem. Pharmacol., 37; 2133-2136,1982.

18. Lucas, D. L, Webster, H. K., and Wright, D. G. Purine metabolism in myetoidprecursor cells during maturation. J. Clin. Invest., 72:1889-1900,1983.

19. Brunngraber, E. F., and Chargaff, E. A nucleotide phosphotransferase fromEscherichia coli. Purification and properties. Biochemistry, 12: 3005-3012,

1973.20. Brunngraber, E. F., and Chargaff, E. Transferase from Escherichia coli effective

low-energy phosphate transfer to nucleosides and nucleotides. Proc. Nati.Acad. Sci. USA, 67: 107-112,1970.

21. Tunis, M., and Chargaff, E. Studies on the nucleoside phosphotransferase ofcarrot. I. Partial purification of the nucleoside phosphotransferase. Biochim.Biophys. Acta, 37: 257-266,1960.

22. Kit, S.. Leung, W. C., Trkula, D., and Dubbs, D. R. Characterization ofnucleoside phosphotransferase and thymidine kinase activities of chick embryo

cells and of chick-mouse somatic cell hybrids. Arch. Biochem. Biophys., 769:66-76,1975.

23. Tesoriere, G., Vento, R., Calvaruso, G., and Gueli, M. C. Kinetic properties ofa nucleoside phosphotransferase of chick embryo. Biochim. Biophys. Acta,660:348-358,1981.

24. Brawerman, G., and Chargaff, E. On the distribution and biological significanceof the nucleoside phosphotransferase. Biochim. Biophys. Acta, 76: 524-532,1955.

25. Itoh, R., Mitsui, A., and Tsushima, K. Properties of 5'-nucleotidase from hepatic

tissue of higher animals. J. Biochem., 63:165-169,1968.26. Itoh, R. Purification and some properties of cytoso! 5'-nudeotidase from rat

liver. Biochim. Biophys. Acta, 657: 402-410,1981.27. Worku, Y., and Newby, A. C. Nucleoside exchange catalyzed by the cyto-

plasmic 5'-nucleotidase. Biochem. J., 205: 503-510,1982.28. Keller, P. M., McKee. S. A., and Fyle, J. A. Cytoplasmic 5'-nucleotidase

catalyzes acydovir phosphorylation. J. Biol. Chem., 260: 8664-8667,1985.29. Ahluwalid, G. S., Jayaram, H. N., Plowman, J. P., Cooney, D. A., and Johns,

D. G. Studies on the mechanism of action of 2-/3-D-ribofuranosylthiazole-4-carboxamide. Biochem. Pharmacol., 33:1195-1203,1984.


537



1986;46:532-537. Cancer Res Arnold Fridland, Michele C. Connelly and Terry J. Robbins -Nucleotidase

′Evidence for Phosphorylation by Adenosine Kinase and 5Tiazofurin Metabolism in Human Lymphoblastoid Cells:

Updated version

http://cancerres.aacrjournals.org/content/46/2/532

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/46/2/532To request permission to re-use all or part of this article, use this link



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

mailto:[email protected]



tiazofurin metabolism in human lymphoblastoid cells ... · 286193), an analogue of the antiviral...

Documents