[CANCER RESEARCH 50, 5269-5274. September 1. 1990]
`
`Tiazofurin Is Phosphorylated
`Cells1
`
`by Three Enzymes from Chinese Hamster Ovary
`
`Priscilla P. Saunders,2 Christy D. Spindler, Mei-Tao Tan, Esperanza Alvarez, and Roland K. Robins
`
`Department of Medical Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas 77030 [P. P. S., C. D. S., M-T. T., E. A.], and Nucleic Acid
`Research Institute, Costa Mesa, California 92626 [R. K. RJ
`
`to these agents.
`kinase deficient cell lines are highly resistant
`Studies have shown, however,
`that a deficiency of adenosine
`kinase does not render CHO cells resistant
`to tiazofurin (9).
`Human lymphoblastoid cells phosphorylate
`tiazofurin via aden
`osine kinase and another enzyme,
`tentatively identified as 5'-
`nucleotidase (15). Recent data from this laboratory have indi
`cated that,
`in CHO cells, another unusual nucleoside, 3-deaza
`guanosine, can be phosphorylated
`by a nicotinamide
`ribonucle
`oside kinase (16) that also demonstrated
`some activity with
`tiazofurin.
`In this communication we describe the phosphoryl
`ation of tiazofurin by this enzyme, as well as by two others.
`Physical
`separation
`of the three cellular enzymes capable of
`catalyzing this reaction, adenosine kinase, nicotinamide
`ribo
`nucleoside kinase, and 5'-nucleotidase,
`is demonstrated.
`
`MATERIALS AND METHODS
`
`ABSTRACT
`
`The growth inhibitory activity of tiazofurin toward adenosine kinase
`deficient Chinese hamster ovary (CHO) cells was partially reversed by
`the presence of nicotinamide riboside. Similarly,
`the formation of tiazo
`furin 5'-monophosphate
`and the active metabolite,
`tiazofurin 5'-adenine
`dinucleotide could be partially inhibited by 100 MMnicotinamide riboside
`in CHO cells and substantially inhibited (80-90%)
`in adenosine kinase
`deficient cells. Tiazofurin phosphorylating activity from CHO cell ex
`tracts was resolved into two peaks by DEAE-cellulose chromatography.
`The
`first
`peak
`of activity was
`identified
`as
`adenosine
`kinase
`(ATP:adenosine S'-phosphotransferase,
`EC 2.7.1.20). The second peak
`of activity correlated with a previously described 3-deazaguanosine phos
`phorylating activity that was identified as a nicotinamide ribonucleoside
`kinase. Contaminating purine nucleoside phosphorylase was removed by
`sedimentation through a sucrose density gradient which also resolved the
`tiazofurin phosphorylating activity into two peaks, one requiring just
`ATP and the other
`requiring both ATP and IMP. Of the substrates
`tested with the lower density peak, nicotinamide riboside was most
`efficient and was the only natural substance that competed well with
`tiazofurin for phosphorylation, substantiating its suggested identity as a
`nicotinamide ribonucleoside kinase. The apparent Ä,„value for nicotina
`mide riboside (2 /IM) was significantly less than that for tiazofurin (13.6
`MM).ATP was the best phosphate donor; CTP and UTP were utilized
`less efficiently and IMP did not support the reaction. The best substrate
`for the higher density peak of tiazofurin phosphorylation was inosine and
`both ATP and IMP were required for the reaction, suggesting its identity
`as a S'-nucleotidase.
`In summary,
`it appears
`that adenosine kinase,
`nicotinamide ribonucleoside kinase, and 5'-nucleotidase may all contrib
`ute to the phosphorylation of tiazofurin in CHO cells.
`
`Chemicals. Selenazofurin (2-/3-D-ribofuranosylthiazole-4-carboxam-
`ide),
`ribavirin (l-;3-D-ribofuranosyl-l,2,4-triazoIe-3-carboxamide),
`3-
`deazaguanosine,
`tiazofurin 5'-monophosphate,
`and tiazofurin 5'-ade-
`nine dinucleotide were prepared as described elsewhere (16-20).
`[3H]
`Tiazofurin (0.5-1.5 Ci/mmol) was obtained from Research Triangle
`Institute, Research Triangle Park, NC, through the Drug Synthesis and
`Chemistry Branch of the National Cancer Institute, Bethesda, MD. [8-
`3H]Deoxyguanosine (7 Ci/mmol),
`[8-3H]deoxyadenosine (11 Ci/mmol),
`[5-3H]deoxycytidine
`(27 Ci/mmol),
`[me/A>>/-3H]thymidine (50 Ci/
`mmol), and [4,5-3H]uridine (42 Ci/mmol) were purchased from ICN
`Pharmaceuticals,
`Inc. (Irvine, CA). [4-3H]Nicotinamide adenine dinu
`cleotide (2.2 Ci/mmol) was purchased from Amersham (Arlington
`Heights,
`IL). [Ã(cid:173)/-nAoÃ(cid:173)>'/-1''C]-3-Deazaguanosinewas prepared as previ
`ously described (21).
`Cells and Media. CHO cells were carried as monolayers in McCoy's
`5a medium as described previously (16). Mutant cell lines were derived
`in this laboratory from CHO cells as indicated previously (4). Cell line
`RbR-l is deficient in adenosine kinase and has been described elsewhere
`(23); line RbM/AAR-10/TGR-6
`(RAT) was derived from RbR-l, se
`lected sequentially with 8-azaadenine and 6-thioguanine,
`and is defi
`cient in adenosine kinase, adenine phosphoribosyltransferase,
`and hy-
`poxanthine-guanine
`phosphoribosyltransferase. Both mutant cell lines
`are stable, were derived from the same parent CHO lines, and demon
`strate sensitivity to tiazofurin and 3-deazaguanosine. Determination of
`minimum inhibitory concentration was as previously described (21).
`Although the nature of the method precludes statistical evaluation the
`data obtained are highly reproducible.
`Metabolic Experiments. Logarithmically growing cultures were de
`tached with trypsin and dispensed into 125-rnl plastic tissue culture
`flasks, 2.5 x IO6cells/flask in 5 ml medium. After incubation for 18 h
`the media were removed and replaced with 5 ml fresh medium contain
`ing dialyzed serum and the desired additions. After
`incubation as
`indicated the media were removed,
`the monolayers were washed with
`solution A, and the cells were extracted directly with 0.4 N perchloric
`acid as described previously (21). Extracts were analyzed by HPLC as
`described below.
`Analyses of Metabolites and Reaction Products by HPLC. Method
`A: tiazofurin and its metabolites in neutralized perchloric acid extracts
`were separated by using Water's Associates equipment described pre
`viously (16) and a column of Partisil-10 SAX aniónexchange resin (25
`x 4.6 mm; Whatman,
`Inc.). Extracts were injected as described and
`5269
`
`INTRODUCTION
`
`Tiazofurin (1), a C-nucleoside, was initially observed to have
`unusual antitumor
`activity (2) and to interfere with guanine
`nucleotide synthesis (3). It has since been shown to be uniquely
`metabolized by mammalian
`cells to tiazofurin 5'-adenine
`di
`nucleotide,
`an analogue of NAD in which the nicotinamide
`moiety is replaced by the thiazole of tiazofurin (4, 5). The latter
`is a very potent
`inhibitor of IMP dehydrogenase
`(4-9), and
`depletes cellular GTP pools (10). Phosphorylation
`of tiazofurin
`to form the 5'-monophosphate
`derivative, assumed to be the
`immediate precursor of TAD,3 appears to occur readily in most
`cells (6, 9, 11, 12); however,
`the identity of enzyme(s)
`involved
`has been unclear. The ability of cells to metabolize tiazofurin
`to TAD appears to vary considerably among different cell lines
`(11). Other compounds having structural
`similarities
`to tiazo
`furin, such as ribavirin and pyrazofurin,
`appear
`to be phos-
`phorylated exclusively by adenosine kinase (13, 14); adenosine
`
`revised 5/30/90.
`Received 3/20/90;
`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
`in
`accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
`1Supported by Grant CA 35788 from the National
`Institutes of Health and
`by Grant CH-283 from the American Cancer Society.
`2To whom requests for reprints should be addressed.
`3The abbreviations used are: TAD, tiazofurin 5'-adenine dinucleotide; HPLC,
`high pressure liquid chromatography; CHO, Chinese hamster ovary; NR, nicotin
`amide ribonucleoside.
`
`
`
`Downloaded from on October 2, 2020. © 1990 American Association for Cancercancerres.aacrjournals.org
`
`
`Research.
`
`THORNE - EXHIBIT 1028
`
`

`

`TIAZOFL'RIN PHOSPHORYLATION
`
`from
`eluted with a concave gradient (curve 7 on the System Controller)
`100% 0.005 M ammonium phosphate, pH 2.8, to 30% 0.75 M ammo
`nium phosphate, pH 3.5, at a flow rate of 2 ml/min. Absorbance was
`monitored at 254 nm. Method B: tiazofurin and tiazofurin S'-mono-
`phosphate could also be separated on a 3.9- x 30-cm reversed phase
`column of Ci8-¿iBondapak(Waters Associates) and isocratic elution
`with 0.5 M ammonium acetate, pH 6.4. This method was used primarily
`to determine [3H]tiazofurin 5'-monophosphate
`formed in enzyme as
`says because the latter could be easily eluted from DE81 disks with the
`ammonium acetate buffer used for the HPLC analysis.
`of [4-3H]
`Preparation of Nicotinamide Riboside. The preparation
`nicotinamide riboside, [car¿>o/i}7-14C]nicotinamideriboside, as well as
`the unlabeled compound was by enzymatic degradation of correspond
`ingly labeled NAD. The procedure was derived from that of Kaplan
`and Stolzenbach (22) and has been described previously (16).
`I nu-ticmation of Cell Extracts by DEAE-Cellulose Chromatograph}-.
`Growth of cells in roller bottles, preparation
`of crude extracts and
`fractionation on DEAE-cellulose are described in a previous commu
`nication (23), with the following modifications: a 1.5- x 20-cm column
`was used with a 250-ml gradient
`(125 ml each of 0 and 0.3 M KC1 in
`buffer A) and approximately
`90 fractions
`containing
`2.7 ml were
`collected and (usually each fourth) analyzed for enzyme activities.
`Buffer A (0.05 M Tris-HCl, pH 7.8, 1 m\i 2-mercaptoethanol,
`or 0.5
`HIM dithiothreitol), with the indicated additions, was used for all
`manipulations of the enzyme preparation.
`Analysis of Enzyme Activities. Assay mixtures for nucleoside phos-
`phorylation contained 50 mM Tris-HCl, pH 7.8, 5 mM MgCI2, 5 mM
`2-mercaptoethanol,
`8 mM ATP (adjusted to pH 7), and 10-100 MM3H-
`or 14C-labeled nucleoside (1-100 ^M) unless indicated otherwise, and
`were incubated 1-3 h at 37°C.At the indicated time(s) aliquots were
`removed and applied to Whatman DE81 filter paper disks which were
`washed 3 times, 10 min each, with 1 mM ammonium formate,
`twice
`with distilled water, and once with 95% ethanol. The disks were allowed
`to partially dry and then were added to scintillation vials containing 1
`ml of l N HCI. After 10 min counting solution was added and the
`radioactivity was determined. For analysis of [I4C]NMN formation,
`reactions were terminated with 0.4 N perchloric
`acid followed by
`neutralization and analysis by HPLC method B. The reaction proceeds
`linearly for at least 3 h, is proportional
`to the amount of enzyme added,
`and is dependent on ATP. Assays for other enzymes are described
`elsewhere (21, 23).
`Sucrose Density Gradient Centrifugation. The Peak II fractions from
`DEAE-cellulose chromatography were combined and concentrated by
`ultrafiltration, using an Amicon device with a PM10 filter. The concen
`trated preparation (0.5-1.0 ml) was then layered onto a 12-ml 5-20%
`sucrose gradient
`in 0.05 M Tris-HCl, pH 7.5, 0.1 M KCI, and 1 mM
`dithiothreitol,
`and centrifuged in a Beckman Model L5-75 ultracentri-
`fuge at 37,000 rpm for 24 h in a SW 40 rotor. Gradients were
`fractionated with the aid of a Buchler Gradient Fractionator.
`
`RESULTS
`
`can give valuable
`reversal experiments
`Growth inhibition
`leads
`in studies of drug metabolism. The observation
`that
`nicotinamide riboside effectively reversed the inhibitory activity
`of 3-deazaguanosine
`( 16) stimulated a similar investigation with
`tiazofurin,
`since the mechanism of phosphorylation
`of both of
`these agents remained unclear. Although the growth inhibitory
`activity of tiazofurin could not be totally reversed by the pres
`ence of nicotinamide
`riboside (Table 1), a very slight effect
`could be seen with CHO cells and a somewhat more significant
`effect, a 10-fold increase in the minimum inhibitory' drug con
`centration, with RAT cells. The latter are deficient
`in adenosine
`kinase,
`adenine
`phosphoribosyltransferase,
`and
`hypoxan-
`and
`thine:guanine
`phosphoribosyltransferase
`demonstrate
`greater
`sensitivity to tiazofurin
`than the parent CHO cells,
`presumably because of the inability to salvage guanine.
`
`the toxicity of
`riboside to affect
`The ability of nicotinamide
`either by competition
`at
`the
`tiazofurin
`could be explained
`mononucleotide
`level (nicotinamide mononucleotide
`formed
`from the riboside would compete with tiazofurin 5'-monophos-
`phate for NAD* pyrophosphorylase)
`or perhaps at the level of
`phosphorylation.
`The observation (Table 1) that growth inhi
`bition by tiazofurin was unaffected by nicotinamide is suggestive
`that
`it is not the production of increased levels of nicotinamide
`mononucleotide
`that
`results in reversal since nicotinamide
`is
`utilized via this intermediate. The metabolic
`experiment
`of
`Table 2 is consistent with this and shows that nicotinamide
`riboside was able to effectively inhibit
`formation of tiazofurin
`5'-monophosphate
`in CHO cells resulting in significantly de
`pressed, but not totally eliminated, TAD formation. The earlier
`observation that
`tiazofurin toxicity was only partially affected
`by 100 nM nicotinamide
`riboside can be explained
`by the
`residual TAD synthesis which, although substantially reduced,
`was probably sufficient
`to inhibit
`IMP dehydrogenase. This
`experiment was carried out
`in Ham's F12 medium which con
`tains only nicotinamide
`at
`low concentration
`(0.3 /¿M)as pri
`mary precursor of NAD. Since it has been shown (15) that
`tiazofurin is phosphorylated
`by adenosine kinase, and also that
`an adenosine kinase deficient mutant
`is not resistant
`to tiazo
`furin, an experiment
`similar to that of Table 2 was carried out
`with the adenosine kinase deficient RAT cell line that
`is also
`unable to salvage guanine. Table 3 shows that TAD formation
`
`Table 1 Reversal of tiazofurin toxicity to CHO and adenosine kinase deficient
`(RA T) cells by nicotinamide riboside
`The minimum inhibitory drug concentration (MIC) was determined by allow
`ing cells to clone for 7-8 days in the presence of drug in medium containing
`dialyzed fetal bovine serum as previously described (9). The MIC is the lowest
`tested concentration of drug in which clones contain less than 50 cells.
`MIC of tiazofurin
`
`AdditionsNoneNicotinamide
`
`Concentration
`
`ribosideNicotinamide(MM)51050100100CHO55557.55RAT15510105
`
`Table 2 Cellularmetabolism of [5-3Hftiazofurin to tiazofurin 5'-monophosphate
`(TRMP) and TAD; effects of nicotinamide and nicotinamide riboside
`CHO cells were dispensed into 25-ml
`tissue culture flasks, 3 x lO'/lask,
`incubated overnight,
`and treated with Ham's F12 medium containing
`5 u\i
`|5-3H]tiazofurin (500 mCi/mmol) and the indicated additions
`for 3 h. Cells were
`then extracted as described in "Materials
`and Methods" and were analyzed by
`HPLC. Values given are the mean ±SD of duplicate determinations.
`Concentration
`[JH]TRMP formed |3H]TAD formed
`(MM)
`(pmol)
`(pmol)
`±0.08
`±0.09
`1.19±0.11
`4.79 ±0.09
`0.32 ±0.044.61
`2.33 ±0.03
`
`None
`Nicotinamide
`Nicotinamide riboside100too1.18
`
`Additions
`
`to TAD in CHO and adenosine
`Table 3 Cellular metabolism of['H]tiazofurin
`kinase deficient (RA T) cells; inhibition by nicotinamide riboside
`treated
`CHO and RAT cells were dispensed into flasks, incubated overnight,
`with medium containing
`5 ><M['H]tiazofurin
`(500 mCi/mmol)
`for 3 h and
`extracted as described. Extracts were analyzed by HPLC; values reflect 4x10*
`cell equivalents. Values are representative of three similar experiments
`carried
`out.
`
`['HITAD formed
`
`(pmol)AdditionsNone
`
`Nicotinamide ribosideConcentration10
`50100CHO4.48
`
`1.09
`2.53
`2.19
`0.54
`1.82RAT3.15
`0.51
`
`5270
`
`
`
`Downloaded from on October 2, 2020. © 1990 American Association for Cancercancerres.aacrjournals.org
`
`
`Research.
`
`

`

`TIAZOFURIN PHOSPHORYLATION
`
`riboside to a much greater extent
`was inhibited by nicotinamide
`in these cells than in the CHO line. Taken together,
`these data
`suggest
`that
`the amount of TAD formed in a cell is a function
`of the amount of tiazofurin phosphorylated
`by more than one
`enzyme. Loss of one of these is apparently
`not sufficient
`to
`render a cell resistant
`to the drug.
`by
`The major nucleoside
`kinases can be easily separated
`conventional DEAE-cellulose
`chromatography
`as described
`elsewhere (23); a typical
`fractionation
`of a CHO cell crude
`extract
`is shown in Fig.
`\A. Under these conditions,
`tiazofurin
`phosphorylating
`activity from CHO cells was resolved into two
`peaks
`(Fig.
`IB), one (Peak
`I)
`that eluted coincident with
`adenosine kinase and another
`(Peak II)
`that eluted more or
`less with purine nucleoside phosphorylase
`and coincident with
`the 3-deazaguanosine
`phosphorylating
`activity (Fig.
`ID) that
`we have identified previously as a nicotinamide
`ribonucleoside
`kinase (16). Although some stimulation of the Peak II activity
`was observed upon addition of IMP to the reaction mixtures
`(not shown), a phosphorylating
`activity dependent
`on IMP
`alone (which would reflect a 5'-nucleotidase) was not observed.
`The identity of Peak I as adenosine kinase was verified by the
`absence of the peak in an extract of the adenosine
`kinase
`deficient cell line, RbR-l (Fig. 1C). To eliminate the possibility
`
`-
`
`15
`ni wihp"
`
`20
`
`55
`
`activity from purine
`Fig. 2. Separation of Peak II tiazofurin phosphorylating
`nucleoside phosphorylase by sucrose density gradient centrifugation. The method
`is described in "Materials and Methods." The top of the gradient
`is on the left,
`the bottom is on the right. (•)Tiazofurin phosphorylation;
`(A) purine nucleoside
`phosphorylase activity. TRMP, tiazofurin 5'-monophosphate.
`
`20r-
`
`••
`
`:
`
`m
`25 O
`
`X
`20 -°
`
`E
`
`»i
`
`S103
`£
`5 E
`
`0123
`
`Hours Incuaatiof.
`Fig. 3. Effect of time and enzyme concentration on tiazofurin phosphorylation
`by the SG-A enzyme; requirement
`for ATP. Reaction mixtures, 100 pi, were as
`described in "Materials and Methods" with (•)26 ^g, (A) 78 fig, or (•)260 /ug
`of SG-enzyme. Other reaction mixtures contained 260 fig of protein: (A) without
`ATP; (O) without ATP. with 8 mM IMP; (d) without enzyme. Points represent
`IO M!of reaction mixture. TRMP, tiazofurin 5'-monophosphate.
`
`l
`
`tiazofurin phosphorylation might actually
`the apparent
`that
`reflect phosphorylation
`of a contaminant,
`the products of the
`reactions of both peaks were characterized by two HPLC meth
`ods.
`In each case,
`tiazofurin
`5'-monophosphate
`was clearly
`formed,
`indicating that
`there are indeed at
`least
`two enzyme
`activities in CHO cell extracts that catalyze the same reaction.
`Initial competition
`experiments with the Peak II enzyme
`raised concerns that the contaminating
`purine nucleoside phos
`phorylase might affect
`the accuracy of the determinations,
`particularly with guanosine and related compounds,
`and give
`misleading results. To eliminate this complication the Peak II
`fractions were combined, concentrated, and subjected to sucrose
`density gradient
`centrifugation which effected a reasonable
`separation of the two enzyme activities (Fig. 2). The tiazofurin
`phosphorylating
`activity remained near the top of the gradient
`in Fractions 4-6, while purine nucleoside phosphorylase was
`entirely in Fractions 8-10. Fractions 4-6 were pooled and are
`referred to here as SG-A enzyme. This fraction is the same as
`that described in an earlier communication
`that was designated
`Peak A and was identified as a nicotinamide
`ribonucleoside
`kinase. Characterization
`of the reaction catalyzed by the SG-A
`enzyme preparation, with [3H]tiazofurin as substrate (Fig. 3),
`indicated that
`it was linear with time, proportional
`to the
`amount of enzyme added, and dependent upon the presence of
`ATP.
`sub
`A variety of compounds were considered as potential
`strates of both the Peak II preparation and the sucrose gradient
`5271
`
`20
`
`80
`
`60
`40
`Fraction Number
`Fig. 1. DEAE-cellulose chromatography of known nucleoside kinase activities
`of CHO cells (A); tiazofurin phosphorylating activities of CHO cells (B); adeno
`sine kinase deficient RbR-l cells (C); 3-deazaguanosine phosphorylating activity
`of CHO cells (D). Crude, dialyzed cell extracts containing 51 mg protein (A) or
`15 mg protein (B and C) were applied to the columns which were developed as
`described in "Materials and Methods." Indicated enzyme activities were measured
`as described. AK, adenosine kinase; PNPase, purine nucleoside phosphorylase.
`
`
`
`Downloaded from on October 2, 2020. © 1990 American Association for Cancercancerres.aacrjournals.org
`
`
`Research.
`
`

`

`TIAZOFURIN PHOSPHORYLATION
`
`Table 4 Comparative substrate activities of SG-A and SG-C in the presence of
`both ATP and IMP
`The activities of SG-A [nicotinamide
`riboside kinase (NRK). 4.6 ng protein/
`reaction mixture] and SG-C (5'-nucleotidase
`(5'-NTase),
`0.8 ^g protein/reaction
`mixture] were determined with the indicated substrates. Enzyme activity was
`determined as described in "Materials
`and Methods" with one exception; ATP
`and IMP were present
`in all
`reaction mixtures at 4 mM concentration
`each.
`Incubation was for 1 h; values reflect 40-(;I aliquots of 100 /il reaction mixtures.
`[MC]Nicotinamide
`mononucleotide
`formation was measured by HPLC as de
`scribed, and is given as mean ±SD of two determinations.
`
`Phosphorylated
`(pmol)
`
`product
`
`Substrate3H]Tiazofurin
`
`CIM)1225
`
`3H]Adenosine
`3H]Guanosine
`l4C]-3-Deazaguanosine
`14C]Nicotinamide riboside
`'"CJInosineConcentration
`
`±0.1
`9 ±0.3
`1.9+ 0.3
`2.0±0.01
`25
`62.5 ±0.5
`20.3 ±0.1
`14.4±0.9
`21.2 ±0.3
`10
`2.0 ±0.9
`78.6 ±3.6
`20
`25SG-A(NRK)11.
`150.3±10.3
`0.0SG-C(S'-NTase)27.4
`
`(SG-A) enzyme and were tested for their abilities to compete
`with (or inhibit)
`tiazofurin phosphorylation
`in the presence of
`ATP as phosphate
`donor. As expected, adenosine
`competed
`well with tiazofurin for phosphorylation
`by Peak I (adenosine
`kinase). but poorly with the reactions catalyzed by Peak II and
`the SG-A preparation. Conversely, nicotinamide riboside inter
`fered with Peak II and SG-A activity but did not affect the Peak
`I catalyzed reaction. The apparent
`affinity of nicotinamide
`riboside for this enzyme activity again suggested the importance
`of
`the enzyme
`in nicotinamide
`ribonucleoside metabolism.
`Other compounds demonstrating
`significant
`interference with
`the Peak II and SG-A activities include selenazofurin,
`an ana
`logue of tiazofurin which is metabolized similarly, and 3-dea-
`zaguanosine. Agents demonstrating
`little or no competition
`with tiazofurin included ribavirin,
`inosine, and all of the deox-
`yribonucleosides
`tested.
`as indicated in
`The best substrate for the SG-A preparation,
`our earlier communication
`( 16), was NR, followed by tiazofurin
`(12% of NR), guanosine (8.9% of NR), and 3-deazaguanosine,
`which could not be directly compared because it was not pos
`sible to test
`it at comparable
`concentrations
`(owing to the
`method of its preparation). Deoxyguanosine,
`deoxycytidine,
`and adenosine demonstrated little, if any, activity. Apparent Km
`values for nicotinamide riboside,
`tiazofurin, and guanosine have
`since been determined to be 2.0, 13.6, and 102 /¿M,respectively.
`ATP was the most efficient phosphate donor
`in the reaction
`catalyzed by the SG-A preparation (not shown). CTP and UTP
`were somewhat
`less efficient (47 and 36% of ATP, respectively)
`and GTP, AMP, and IMP supported little,
`if any, phosphoryl
`ation.
`To clarify the earlier observation that Peak II phosphoryla
`tion was stimulated by addition of IMP, a larger scale prepa
`ration was carried out and the sucrose gradient
`fractions were
`analyzed with a variety of conditions and substrates
`(Fig. 4). In
`the presence of ATP alone (Fig. 4A) there was a single peak
`(Fractions 4-6, SG-A) of tiazofurin phosphorylation; with IMP
`alone there was no activity observed in Fractions 4-6 and a
`trace of apparent phosphorylation
`in Fractions 15-17 (SG-C).
`With both ATP and IMP present
`(Fig. 4B) there were clearly
`two substantial, well separated, peaks of activity. 3-Deazagu-
`anosine was similarly phosphorylated
`by both activities
`(Fig.
`4C). Inosine, on the other hand, was phosphorylated exclusively
`
`therefore levels of activity are not directly comparable from one graph to another.
`TRMP,
`tiazofurin
`5'-monophosphate;
`3-DeazaGTP, 3-deazaguanosine triphos-
`phate;
`jVA/7V. nicotinamide mononucleotide;
`3-Dea:aGuo.
`3-deazaguanosine:
`WO.
`inosine, TR, tiazofurin.
`5272
`
`15
`
`t ATP
`
`l \
`
`t
`
`IMP
`
`poo.
`
`[3H]TH t ATP »IMP
`
`I
`
`,0
`
`Ii 5
`
`0
`
`25
`
`I
`
`* "
`Ia
`
`I
`
`10
`5
`
`15
`
`H s io
`
`1.00
`
`0.75
`
`114C|INO
`t ATPt IMP
`
`r 0.50
`
`0.25
`
`0.00
`
`50
`
`30
`
`l««C|NHt ATP
`
`I L
`
`20
`1 io
`
`1.0
`
`L 0.5
`
`20
`
`0.0
`
`0 T
`
`15
`10
`BOTTOM
`numoer
`traction
`OP
`Fraction Number
`activities detectable after sucrose gradient
`fractiona-
`Fig. 4. Phosphorylating
`tion of Peak II assayed in the presence of ATP and IMP. Gradients were layered
`with 2.5 mg of Peak II protein.
`Incubation mixtures were as described with the
`following variations: A, 6.7 <JM[3H|tiazofurin
`(0.88 Ci/mmol).
`8 HIM ATP (*) or
`8 mivi IMP (O); B, 6.7 MM [3H]tiazofurin.
`8 mM ATP. 3 mM IMP; C, 3 »JM(14C]-
`3-deazaguanosine (23 mCi/mmol).
`6 mM ATP, 3 ITEMIMP; 0. 73 pM |'*C|inosine
`(57 mCi/mmol).
`6 mM ATP. 3 mM IMP; E. 10.5 JIM |"C|nicotinamide
`riboside
`(44 mCi/mmol).
`6 mM ATP; F, 133 /iM [8-MC]IMP (8 mCi/mmol).
`no other
`nucleotides.
`Incubation mixtures
`contained 25 ¡i\of sucrose gradient
`fractions
`and incubation was for 2.5 h at 37'C. Aliquots
`of each (35 ^1) were applied to
`DE81 disks which were treated as described for nucleoside kinase assays. Assays
`were carried out over a period of
`ten days (enzyme was stored at
`-20°C):
`
`
`
`Downloaded from on October 2, 2020. © 1990 American Association for Cancercancerres.aacrjournals.org
`
`
`Research.
`
`

`

`15-17, Fig. 4D), which
`by the second peak, SG-C (Fractions
`corresponded with IMP hydrolytic activity (Fig. 4F), allowing
`the conclusion that
`this enzyme probably reflects a 5'-nucleo-
`tidase similar to that described by Fridland et al. (15). Compar
`ative substrate activities of several relevant compounds with the
`two enzymes,
`in the presence of both ATP (4 HIM)and IMP (4
`HIM) is tabulated in Table 4. It
`is of interest
`to note that 3-
`deazaguanosine
`appears to be more efficiently phosphorylated
`by SG-A (presumably nicotinamide
`ribonucleoside kinase) than
`by SG-C (presumably 5'-nucleotidase), while the reverse seems
`to be true for tiazofurin phosphorylation.
`This is consistent
`with the observation that nicotinamide
`ribonucleoside
`reverses
`the toxicity produced by 3-deazaguanosine more effectively than
`that resulting from exposure to tiazofurin.
`In other experiments
`(not shown) phosphorylation
`of tiazofurin by SG-A was clearly
`inhibited by the presence of nicotinamide
`riboside while the
`reaction catalyzed by SG-C was essentially unaffected. SG-C
`required both IMP and ATP (or dATP)
`for activity while SG-
`A required ATP alone and was unaffected by the addition of
`IMP. All of these properties
`taken together support our conclu
`sion that one of the activities (SG-A) represents
`the nicotina
`mide ribonucleoside kinase that we have previously associated
`with 3-deazaguanosine
`phosphorylation,
`and that
`the SG-C
`enzyme is a cytoplasmic 5'-nucleotidase
`such as that described
`by Fridland et al. (15).
`
`DISCUSSION
`
`TIAZOFURIN PHOSPHORYLATION
`a loss of
`requiring formation of tiazofurin 5'-monophosphate,
`the ability to phosphorylate
`the compound would not result
`in
`resistance. The latter possibility seems improbable since tiazo
`furin is a C-nucleoside and thus cannot enter
`into those reac
`tions whereby nicotinamide
`is initially metabolized to NAD,
`whether via a phosphoribosyltransferase
`to form the 5'-mono-
`phosphate or to NAD directly by an exchange reaction. These
`routes have been suggested by Nelson et al. (25) as possibilities
`in the metabolism of a related agent, 2-amino-l,3,4-thiadiazole.
`Cytoplasmic 5'-nucleotidase
`has recently been implicated in
`the phosphorylation
`of inosine (26), acyclovir
`(27),
`tiazofurin
`(15) and 2',3'-dideoxyinosine
`(28) through an exchange reac
`tion mediated by an enzyme-phosphate
`intermediate. The phos
`phorylation reaction with these nucleosides
`is characterized by
`a requirement
`for IMP as phosphate
`donor,
`stimulation
`by
`ATP, and inhibition by high concentrations
`of inosine. Phos
`phorylation of tiazofurin by the partially purified enzyme (SG-
`A) from CHO cells demonstrates
`a clear requirement
`for ATP
`in the absence of added IMP. Nicotinamide
`riboside is by far
`the best substrate for this enzyme. The SG-C activity, on the
`other hand,
`requires both IMP and ATP, although minimal
`reaction can be detected with IMP alone. Inosine is clearly the
`best substrate
`for SG-C. From these observations we have
`concluded that SG-A represents a nicotinamide
`ribonucleoside
`kinase while SG-C is probably a 5'-nucleotidase
`such as that
`described by Fridland et al. ( 15).
`kinase in cellular
`The role of nicotinamide
`ribonucleoside
`metabolism remains unclear. Although this reaction has been
`observed to occur
`(29, 30),
`it is not well documented and its
`importance in NAD synthesis and cycling is uncertain. We have
`presented evidence elsewhere (16) for the activation by this
`enzyme, of another unusual nucleoside, 3-deazaguanosine.
`In
`view of the possible role of this phosphorylating
`activity in drug
`metabolism and action, additional work directed toward a fun
`damental understanding of its occurrence,
`its substrate specific
`ity, and its significance in NAD metabolism seems warranted.
`
`REFERENCES
`
`of tiazofurin is assumed to be a key
`The phosphorylation
`reaction in the metabolism of this agent
`to its active form.
`While adenosine
`kinase,
`in partially purified form, has the
`capability of phosphorylating
`tiazofurin,
`it is not clear whether
`it does so to a significant extent
`in intact CHO cells; adenosine
`kinase deficient cells demonstrate
`little or no resistance to the
`drug. This
`is in contrast
`to the structurally
`related agents
`ribavirin and pyrazofurin which are readily phosphorylated
`by
`adenosine kinase in whole cells and are inactive toward adeno
`sine kinase deficient
`cells.
`In this communication we have
`demonstrated
`the existence of an enzyme in CHO cells, other
`than adenosine kinase or 5'-nucleotidase,
`that can phosphoryl-
`1. Srivastava, P. D., Pickering, M. U., Allen, L. B., Streeter, D. G., Campbell,
`M. T., Witkowski, J. T., Sidwell, R. W., and Robins, R. K. Synthesis and
`ate tiazofurin. We have physically separated the three phospho
`antiviral activity of certain thiazole C-nucleosides. J. Med. Chem., 20: 256-
`rylating activities and have demonstrated
`conditions
`for the
`262, 1977.
`2. Robins. R. K., Srivastava, P. C, Narayanan, V. L., Plowman, J., and Paull,
`selective measurement of each. Moreover, we present evidence
`K. D. 2-/3-D-Ribofuranosylthiazole-4-carboxamide,
`a novel potential antitu-
`that
`the new phosphorylating
`activity is relevant
`to the utiliza
`mor agent for lung tumors and métastases.J. Med. Chem., 25: 107-108,
`tion of the drug in growing cells, and tentatively identify it as a
`1982.
`3. Streeter, D. G., and Miller, J. P. The in vitro inhibition of purine nucleotide
`nicotinamide
`ribonucleoside kinase,
`the same enzyme to which
`biosynthesis by 2-$-D-ribofuranosylthiazole-4-carboxamide.
`Biochem. Bio-
`we have attributed the phosphorylation
`of 3-deazaguanosine
`in
`phys. Res. Commun., 103: 1409-1412, 1981.
`hypoxanthine-guanine
`phosphoribosyltransferase
`deficient
`Inhibition of inosinate
`4. Kuttan, R., Robins, R. K., and Saunders, P. P.
`dehydrogenase by metabolites of 2-fi-D-ribofuranosylthiazole-4-carboxamide.
`CHO cells (16).
`Biochem. Biophys. Res. Commun., 107: 862-868, 1982.
`involved in
`factors
`have investigated
`Several
`laboratories
`5. Cooney, D. A., Jayaram, H. N., Gebeyehu, G., Betts, C. R., Kelley, J. A.,
`Marquez, J. E., and Johns, D. G. The conversion of 2-/3-D-ribofuranosylthia-
`resistance to tiazofurin, both natural and acquired (6, 9, 11, 12,
`zole-4-carboxamide
`to an analog of NAD with potent
`IMP dehydrogenase-
`24), and have reported
`several conditions
`resulting
`in this
`inhibitory properties. Biochem. Pharmacol., 31: 2133-2136. 1982.
`phenotypic property,
`the most common of which is an inability
`6. Jayaram, H. N., Cooney, D. A., Glazer, R. I., Dion, R. L., and Johns, D. G.
`Mechanism of resistance to the oncolytic C-nucleoside 2-ii-D-ribofuranosyl-
`to form TAD. None, however, have reported a deficiency of
`thiazole-4-carboxamide
`(NSC 286193). Biochem. Pharmacol.,
`31: 2557-
`phosphorylation
`as a mechanism of resistance
`to this agent.
`2560, 1982.
`7. Jayaram, H. N., Dion, R. L., Glazer, R. I., Johns, D. G., Robins, R. K.,
`There are several possible explanations
`for this: a) if there are
`Srivastava, P. C., and Cooney, D. A. Initial studies on the mechanism of
`three enzymes catalyzing the reaction (i.e., adenosine kinase,
`action of a new oncolytic thiazole nucleoside, 2-/3-D-ribofuranosylthiazole-4-
`5'-nucleotidase,
`and nicotinamide
`ribonucleoside
`kinase)
`the
`carboxamide (NSC 286193). Biochem. Pharmacol., 31: 2371-2380,
`1982.
`8. Jayaram, H. N., Smith, A. L., Glazer, R. I., Johns, D. G., and Cooney, D.
`probability of losing all of them through mutation would be
`A. Studies on the mechanism of action of 2-(J-D-ribofuranosylthiazole-4-
`very low; b)
`if the enzyme responsible
`for phosphorylating
`carboxamide (NSC 2