- Email: [email protected]
Ribonucleotide reductase activity in intact mammalian cells: Stimulation of enzyme activity by MgCl2, dithiothreitol, and several nucleotides
ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 231, No. 1, May 15, pp. 9-16, 1984
Ribonucleotide Reductase Activity in Intact Mammalian Cells: Stimulation of Enzyme Activity by MgC12, Dithiothreitol, and Several Nucleotides ROBERT Departments
of Mkmbio~ Cell Bidosy,
G. HARDS
and Biochemistry, 100 Olivia Street,
Received
June
AND
JIM A. WRIGHT’
University of Manitoba, Winnipeg, Manitoba R3E
30, 1983, and in revised
form
and Manitoba OV9, Canada
January
Institute
of
9, 1934
An intact cell assay system based on Tween-30 permeabilization was used to investigate ribonucleotide reductase activity in Chinese hamster ovary cells. Dithiothreitol, a reducing agent, is required for optimum activity. Analysis of dithiothreitol stimulation of CDP and ADP reductions indicated that in both cases the reducing agent served only to increase the reaction rate without altering the affinity of the enzyme for substrates. Magnesium chloride significantly stimulated the reduction of CDP but not ADP; this elevation in CDP reduction was due to an increase in both the affinity of the enzyme for substrate and the I’,,,,,. In addition to ATP and dGTP, well-known activators of CDP and ADP reductase activities, it was found that dCTP and GTP were also able to activate CDP and ADP reductase activities, respectively. For the dCTP-activated reaction the I’,,,, was 0.158 nmol dCDP formed 5 X lo6 cells-’ h-’ and the K, was 0.033 mM CDP, while for the GTP-activated reduction a V,,, of 0.667 nmol dADP formed 5 X lo6 cells-’ h-’ and a K, of 0.20 IIIM ADP were observed. Kinetic analysis revealed that dCTP, dGTP, and GTP stimulate ribonucleotide reduction solely by increasing the affinity of the enzyme for substrate without affecting the V,, of the respective reactions. ATP behaves in a different manner as it stimulates CDP reduction by altering both the affinity of the enzyme for substrate and the V,,. Cellular concentrations of riboand deoxyribonucleoside di- and triphosphate pools were measured to help evaluate the relative physiological importance of the nucleotide activators. These determinations, along with the reaction kinetic studies, strongly imply that ATP is a much more important regulator of CDP reduction than dCTP, whereas GTP may serve as well or better than dGTP as the in vivo activator of ADP reduction.
A continuous and balanced supply of deoxyribonucleoside triphosphates originating from the reduction of ribonucleotides is required during DNA synthesis. In mammalian cells this reduction occurs with ribonucleoside diphosphates and is carried out by ribonucleotide reductase (EC 1.17.4.1). This enzyme, which is likely a rate-limiting step in DNA synthesis and therefore in cell proliferation, occupies a key position in the biology of the cell (1). Alterations in the reduction of ribonucleotides have wide-ranging consequences. 1 To whom
correspondence
should
For example, the enzyme appears to play an important role in tumor transformation (2, 3), certain immunodeficiency diseases in man (4), alterations in the spontaneous mutation rates of cultured cells (5), the senescence of human diploid fibroblasts (6), and perhaps in certain aspects of cellular differentiation (7). In keeping with the importance of ribonucleotide reductase activity in the physiology of the cell, enzyme specificity and activity is strictly regulated in a very complex manner by nucleoside triphosphate effecters (1,9). Furthermore, structural features of the enzyme are unusual @-lo), and, during DNA synthesis,
be addressed. 9
ooo3-9361/34 Copyright All rights
$3.00
0 1924 by Academic Press. Inc. of reproduction in any form reserved.
10
HARDS
AND
the reductase appears to be physically and functionally associated with other enzymes of DNA synthesis in a multienzyme complex which channels deoxyribonucleotides to the replication forks (11-13). From biological, regulatory, and structural points of view, ribonucleotide reductase is one of the most complex enzymes in the cell (14). The properties of mammalian ribonucleotide reductase have been investigated, for the most part, in cell and tissue extracts. Several models based on these results have been proposed to account for the in vivo regulation of the enzyme activity (8,15). There is increasing evidence, however, that the properties of an enzyme can be affected by the intracellular environment (16). For example, the dilution that accompanies cell breakage can significantly affect enzyme characteristics. This is likely to be especially true for an enzyme as complex as ribonucleotide reductase. Therefore, to study this activity in a situation as close to the normal in vivo environment as possible, this laboratory has developed an intact cell assay system (1,1’7,18) based upon Tween-80 permeabilization. In this investigation, we used the assay system to study the intracellular properties of CDP and ADP reductase activities in the presence of various activators, and to examine the possibility that dCTP and GTP may be physiologically important regulators of enzyme activity. In addition, ribo- and deoxyribonucleotide pool sizes have been determined and used to evaluate kinetic findings. MATERIALS
AND
METHODS
Chinese hamster ovary (CHO)’ cells were cultured at 3’7°C on the surface of plastic tissue culture plates (LUX Scientific) or glass bottles (Brockway Glass Co.) in o-minimal essential medium (a-MEM, Flow Laboratories) supplemented with antibiotics and 10% (v/v) fetal calf serum (FCS, Gibco) as described in previous publications (19, 20). The cells grew with a doubling time of 16 h.
‘Abbreviations used: CHO, Chinese hamster ovary; (u-MEM, a-minimal essential medium; FCS, fetal calf serum; Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid, D’IT, dithiothreitol; PBS, phosphatebuffered saline.
WRIGHT Intracellular ribonucleotide reductase activity was determined in CHO cells made permeable to ribonucleotides as described previously (&l&20). In brief, exponentially growing cells were plated at a density of 5 x 106/150-mm culture plate containing growth medium and incubated at 37°C for 40 h to achieve a density of 2.2 f 0.2 X 10’ cells/plate. The cells were removed with trypsin solution, centrifuged, washed with growth medium, and resuspended at 10’ cells/ ml of permeabilizing buffer consisting of 1% Tween80 (J. T. Baker Co.), 0.25 M sucrose, 50 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (Hepes) buffer, pH 7.2, and 2 mM dithiothreitol (D’IT). The cells were incubated at 22°C for 30 min with occasional agitation and then centrifuged. The cell pellet was resuspended in fresh permeabilizing buffer at a concentration of 2.5 X lo7 cells/ml. Aliquots of 5 X lo6 cells were then added to assay tubes and intracellular CDP and ADP reductase activities were determined. Standard reactions contained either 1.5 mM ATP or 0.5 mM dGTP as activators for CDP or ADP reductions, respectively. A recent evaluation of the intact cell assay system has shown that potential problems involving nucleotide pools and nucleotide metabolizing enzymes do not present serious problems for examining the reductase in the physiologically relevant environment of the whole cell (18). In the present study, determinations of nucleotide concentrations present during the assay period were made as previously described (18) and taken into account to obtain corrected estimations of ribonucleotide reductase activity as outlined earlier (18). To estimate ribo- and deoxyribonucleoside di- and triphosphate pool sizes, the nucleotides were extracted from the CHO cells using the method of Garrett and Santi (21). The cells were grown to a density of 2.2 -C 0.2 X 10’ cells/l50-mm plastic tissue culture plate in a-MEM plus 10% FCS. For each extraction, the cells from five such plates were removed with trypsin solution, washed twice with ice-cold PBS, counted with a Coulter particle counter (Coulter Electronics Ltd.), and resuspended in PBS at a concentration of 0.5 to 3.0 X 10s cells/ml. The PBS into which the cells were resuspended contained [‘Hluridine (130,000 cpm/ml) as a concentration marker and aliquots of PBS were removed for scintillation counting. Seventy percent perchloric acid was then added to the suspension to give a final concentration of 0.5 M. After 30 min on ice, the precipitate was removed by centrifugation in an Eppendorf microcentrifuge, 10 pl of bromocresol blue was added, and the supernatant was neutralized with 4 M KOH containing 0.4 M KHzPO,. The neutralized extract was kept on ice for at least 10 min to allow for the precipitation of potassium perchlorate. The solution was again cleared by centrifugation and an aliquot of the supernatant was analyzed for radioactivity in order to determine dilution factors. As deoxyribonucleotides are often obscurred by the
RIBONUCLEOTIDE
REDUCTASE
ACTIVITY
presence of ribonucleotides, analysis of each required the separation of the two. The method of Garrett and Santi (21), which involves destruction of the ribonucleotides to allow for detection and quantitation of the deoxyribonucleotides, was utilized with slight modifications. Forty microliters of 0.5 M sodium periodate was added to 1 ml of neutralized cell extract. After 5 min, 50 ~1 of 4 M methylamine, which had been slowly brought to pH 7.5 with phosphoric acid, was added. The reactants were mixed and incubated at 37°C for 60 min. Ten microliters of 1 M rhamnose was then added to destroy any remaining periodate and the samples were immediately put on ice. In experiments where there was less than 1 ml extract, proportionally smaller volumes of reagents were used. The solution was cleared by centrifugation and an aliquot was removed for scintillation counting to determine dilution factors. Aliquots of the ribonucleotide plus deoxyribonucleotideand deoxyribonueleotide-containing solutions were analyzed using the HPLC method described previously (18). Separation was achieved using a Partisil10125 SAX (Whatman) strong anion-exchange column (25 cm X 4.6 mm). The initial eluant, 0.007 M KHzPOI(HPLC grade, Fisher Chemical Co.) and 0.007 M KCI(ACS grade, J. T. Baker Co.) was adjusted to pH 4.0 with HCl. The final eluant, 0.25 M KH,PO* and 0.5 M KCI, was adjusted to pH 5.0 with NaOH. An isocratic elution on the low concentration buffer lasting 20 min was followed by a linear gradient from 0% of the final eluant to 100% of the final eluant run in 35 min. Finally, an isocratic elution on the high concentration buffer was run for 15 min. The flow rate was 2.0 ml/min. All buffers were filtered through 0.22-pm Millipore membrane filters (HA, Millipore), and degassed prior to use. Identification was by retention times as compared to prepared standards. For radioactivity determinations of fractions, the fractions collected were transferred to scintillation vials containing 10 ml ACS cocktail and radioactivity was determined with a Beckman Model LS230 liquid scintillation spectrophotometer. Radiochemicals were purchased from Amersham. All other biochemicals were obtained from the Sigma Chemical Company.
IN
Since ribonucleotide reductase catalyzes the reduction of both pyrimidine and purine ribonucleotides (l), the properties of this activity had been examined in the presence of a pyrimidine (CDP) and a purine (ADP) substrate. Enhanced Activity with Nonnucleotide COW&pOUnds While MgClz and the reducing agent, dithiothreitol, are not essential for the re-
MAMMALIAN
CELLS
11
duction of CDP in intact CHO cells (1,20), they do significantly enhance the activity. In order to determine how these compounds stimulate reduction, double-reciprocal plots of velocity against CDP concentration at several fixed levels of MgClz and DTT were constructed (Fig. 1). Magnesium ions increased both the rate of reduction and the affinity of the enzyme for CDP. As expected, DTT, which acts as a hydrogen donor for the reaction, increased only the velocity of the reaction and did not affect the affinity for CDP. Double-reciprocal plots of the variation in velocity of CDP reduction at varying MgCl, or DTT concentrations at several fixed levels of CDP (plots not shown) indicated that in neither case did the substrate level affect the affinity of the enzyme for MgClz or DTT. The K, values determined for the two compounds are 2.4 mM MgClz and 2.8 mM DTT. Although MgCl, does not enhance ADP reductase activity (20), a reducing agent such as DTT is required for optimum activity (1,20). As shown in Fig. 2, DTT stimulates the reduction of ADP by increasing the rate of reduction. As was the case with
-80
RESULTS
INTACT
0
80
160 CDP hM-‘1
FIG. 1. (a) Double-reciprocal plot of the variation in rate of CDP reduction with CDP concentration at several fixed levels of MgCl,. Assays were performed in the absence (0) or presence of 2 mM (A), 2.5 mM (Cl) and 6 mM (0) MgClz. Activity was expressed as nmol dCDP formed 5 X lo6 cells-’ h-l. (b) Doublereciprocal plot of the variation in rate of CDP reduction with CDP concentration at several fixed levels of DTT. Assays were performed in absence (0) or presence of 1.5 mM (A), 3 mM (0) and 6 m&i (0) DTT. Activity was expressed as nmol dCDP formed 5 X lo6 cells-’ h-i.
12
HARDS
AND
r
WRIGHT
2
a
B i2 I: 0.15 8 “Do 2 9 0.10
Y
8B
0
3 0 3 *
P\ w &
0.050
i
0
2
4
GTP (mM)
dCTP hM ADP hd’)
FIG. 2. Double-reciprocal plot of the variation in rate of ADP reduction with ADP concentration at several fixed levels of DTT. Assays were performed in the presence of 1.5 mM (0), 3 mM (A), 4.5 rnrd (0), and 6 mM (0) DTT. Activity was expressed as nmol dADP formed 5 X lo6 cells-’ h-l.
CDP reduction, DTT did not affect the enzyme affinity for ADP. Rearranging the data to give a double-reciprocal plot of velocity against DTT concentration at several fixed levels of ADP (plots not shown) indicated, as expected, that ADP had no effect on the interaction of ribonucleotide reductase with DTT. A K, value of 2.7 mM DTT was determined. With regard to the reducing agent, DTT, CDP, and ADP reductase activities respond in an identical manner.
Nucleotide
FIG. 3. (a) Response of CDP reductase activity to dCTP concentrations. Standard CDP reductase assay conditions were used except for the omission of ATP and the presence of varying concentrations of dCTP. (b) Response of ADP reductase activity to GTP concentrations. Standard ADP reductase assay conditions were used except for the omission of dGTP and the presence of varying concentrations of GTP.
previously characterized. It is apparent from Fig. 3 that maximal enzyme activity occurs in the presence of about 3.6 mM dCTP and 0.2 mM GTP. Figure 4 shows that the reduction of CDP in the presence of 3.6 mM dCTP exhibits a Km of 0.033 mM CDP and a V,, value of 0.158 nmol of dCDP formed 5 X lo6 cells-’ h-l. In comparison, the values observed (18, 20) with 1.5 mM ATP as an activator of the reaction were a Km of 0.021 mM CDP and a Vmax of 0.49 nmol dCDP formed 5 X lo6
Activators
Previous studies with intact mammalian cells and cell-free extracts have shown that ribonucleotide reductase exhibits an absolute requirement for allosteric activators (1, 8, 1’7, 20). It has been well documented that ATP acts as a positive effector of CDP reduction and dGTP activates the reduction of ADP. In preliminary experiments (1) with intact CHO cells, we have also observed that dCTP and GTP stimulate the reductions of CDP and ADP, respectively. The GTP effect, although noted in other investigations (22), has generally not been well characterized or considered in models of ribonucleotide reductase regulation. Deoxy-CTP stimulation has been noted by only one other group (39) but has not been
‘0 ADP (md’)
0
60
120
130
CDP hd)
FIG. 4. (a) Double-reciprocal plot of the variation in rate of ADP reduction with ADP concentrations. Reduction was activated by 0.2 mM GTP. Activity was expressed as nmol dADP formed 5 X 10’ cells-’ h-r. (b) Double-reciprocal plot of the variation in rate of CDP reduction with CDP concentrations. Reduction was activated by 3.6 mre dCTP. Activity was expressed as nmol dCDP formed 5 X lo6 cells-’ h-l.
RIBONUCLEOTIDE
REDUCTASE
ACTIVITY
cells-’ h-‘. In the presence of 0.20 mM GTP, double-reciprocal plots of ADP reduction showed a Km value of 0.20 mM ADP and a V,,,, of 0.67 nmol dATP formed 5 X lo6 cells-’ h-l. In comparison, kinetic studies with optimum dGTP concentrations gave a Km value of 0.095 mM ADP and a V,,, of 1.00 nmol dADP formed 5 X lo6 cells-’ h-’ (18, 20). To learn more about the activation of CDP and ADP reductase in intact cells, the four positive effecters of these reactions were further examined by constructing double-reciprocal plots of velocity against substrate concentrations at various fixed levels of effecters. As shown in Fig. 5a, ATP increases both the rate of CDP reduction and the affinity of the enzyme for CDP. A similar effect for ATP has been reported with the enzyme from cell-free extracts (23). However, it appears that dCTP activates the reduction of CDP solely by increasing the affinity of the enzyme for the substrate (Fig. 5b). The effects of dGTP and GTP on the reduction of ADP are shown in Fig. 6. Similar to the studies described above with dCTP, it appears that both GTP and dGTP activate intracellular ADP reduction entirely by increasing enzyme substrate affinity. The K, values for
30 CDP
0
80
160
hbl-‘1
FIG. 5. (a) Double-reciprocal plot of the variation in rate of CDP reduction with CDP concentration at several fixed levels of ATP. Assays were performed in the presence of 0.6 mM (0), 0.9 mM (A), 1.2 mM (O), and 1.5 mM (0) ATP. Activity was expressed as nmol dCDP formed 5 X 106cells-’ h-l. (b) Double-reciprocal plot of the variation in rate of CDP reduction with CDP concentration at several fixed levels of dCTP. Assays were performed in the presence of 0.9 mbi (O), 1.35 mM (Cl), and 3.6 mM (0) dCTP. Activity was expressed as nmol dCDP formed 5 X lo6 cells-’ h-l.
IN
INTACT
MAMMALIAN
13
CELLS
,40
0
40
80
120
FIG. 6. (a) Double-reciprocal plot of the variation in rate of ADP reduction with ADP concentration at several fixed levels of dGTP. Assays were performed in the presence of 0.046 mM (0), 0.092 mM (A), and 0.134 mM (Cl) dGTP. Activity was expressed as nmol dADP formed 5 X lo6 cells-’ h-i. (b) Double-reciprocal plot of the variation in rate of ADP reduction with ADP concentration at several fixed levels of GTP. Assays were performed in the presence of 0.18 mM (0), 0.36 mM (A), and 0.54 mM (Cl) GTP. Activity was expressed as nmol dADP formed 5 X lo6 cells-’ h-l.
the various activators were 1.37 mM ATP, 1.13 mM dCTP, 0.096 mM GTP, and 0.012 mM dGTP.
Nucleotide
Pools
Cellular concentrations of ribo- and deoxyribonucleoside di- and triphosphate pools were determined to help evaluate the relative significance of the various effecters as regulators in the cell. The pool sizes were determined in logarithmically growing cells cultured in a manner identical to the cells routinely used in enzyme assays. The nucleotides were extracted and analyzed as described under Materials and Methods and the results of these studies are shown in Table I. The deoxyribonucleoside triphosphate levels found for CHO cells were well within the range of values obtained with many types of mammalian cells (e.g. (21, 24-27)). The CHO ribonucleoside triphosphate pools were also within the range of values reported for other animal cells, except that the values were towards the lower end of the range (e.g., (21, 24-27)). Although relatively few studies of intracellular ribonucleoside diphosphate pool sizes have been carried out, those that have been reported agree rea-
14
HARDS TABLE NUCLEOTIDE
AND
I
POOL SIZES OF EXPONENTIALLY GROWING CHO CELLS Intracellular concentration
Nucleotide
nmol/106 f f -+ f
cells
mM*
GTP ATP CTP UTP
0.334 1.589 0.604 0.387
0.038 0.258 0.379 0.030
0.223 1.059 0.403 0.238
GDP ADP CDP UDP
0.157 + 0.025 1.060 0.048 f 0.009 0.121 f 0.011
0.106 0.558 0.026 0.075
dGTP dATP dCTP dTTP
ND” 0.028 f 0.007 0.105 f 0.065 0.053 f 0.035
0.019 0.070 0.035
Note. Results are the averages trials. ‘Not detected. *Based on a CHO cell volume
of two independent
of 1.7 pl (37).
sonably well with the values found for CDP and UDP, whereas the purine ribonucleoside diphosphates, ADP and GDP, were detected in slightly higher levels in CHO cells as compared to many other mammalian sources (24, 25, 27). DISCUSSION
While the two small proteins, thioredoxin through thioredoxin reductase and glutaredoxin via glutathione reductase, function as natural hydrogen carriers in the reduction of ribonucleotides, assays of ribonucleotide reductase utilize dithiols such as dithiothreitol as direct hydrogen donors. In this study, we have shown that dithiothreitol stimulates both CDP and ADP reduction by increasing the velocity of the reactions (Figs. lb and 2). In neither case did the dithiol affect the affinity of the enzyme for the substrates. These findings are consistent with the proposed role for DTT in these reactions. The dithiol apparently reduces a disulfide at the active site, which in turn reduces the ribonucle-
WRIGHT
otide substrate (28). Our results suggest that the reduction of the active site disulfide is required for enzyme activity but it is apparently not needed for substrate binding. Magnesium chloride exerted a dual effect on CDP reduction as it increased both the velocity of the reaction and the affinity of the enzyme for CDP (Fig. la). Magnesium ions are thought to promote the association of nonidentical subunits (Ml and M2) of mammalian ribonucleotide reductase to form a holoenzyme capable of reducing CDP (29,30). This association could explain the observed elevation in velocity caused by MgCla. The increased affinity of the enzyme induced by MgClz could be due to several reasons. For example, it appears that nucleotides bind only to the Ml subunit of the reductase (8, 31, 32), and it is possible that an association of subunits induced by magnesium ions causes a conformational change in Ml, which facilitates the binding of CDP. However, optimal binding of CDP may require interaction between Ml and M2; therefore, it is also possible that a MgClz-induced association of these subunits could lead to increased binding and stimulation of enzyme activity. Although ATP and dGTP are wellknown activators of CDP and ADP reductase activities, respectively, we have observed that dCTP can activate CDP reduction while GTP is a significant activator of ADP reduction in intact hamster cells. It is worth noting that HPLC analysis of the nucleotides present at various times during the reaction have been performed as described in detail in a previous report (18). These studies indicated that the activations observed with dCTP and GTP were due to their direct effects on activity and not to a conversion to another nucleotide or to a stimulation in the reduction of a second substrate. Activation of CDP reduction by ATP and dCTP seemed to involve two different processes. While ATP increased both the affinity of the reductase for CDP and the rate of reduction (Fig. 5a), the sole effect of dCTP was to elevate the affinity of the enzyme for CDP (Fig. 5b). Thus, it would seem that in its native conformation ri-
RIBONUCLEOTIDE
REDUCTASE
ACTIVITY
bonucleotide reductase is capable of reducing CDP but the Km value for CDP is prohibitively high; binding of dCTP to the enzyme would facilitate binding of the substrate without altering the maximum velocity obtainable. It has been suggested (38, 40) that binding of dCTP to an allosteric site on a multienzyme complex involved in DNA synthesis increases the affinity of the complex for nucleoside diphosphates. The kinetic studies with dCTP fit this suggested role, although more direct evidence is required. When ATP is used as an activator it apparently induces a conformational change which not only enhances binding of CDP but also increases the maximum velocity that can be reached. The pattern of ATP activation in intact cells is similar to that seen with cell-free preparations from rat tumor (23). Conformational changes during the binding of substrates in the presence of effecters of the enzyme have been observed in the Escherichia coli system and with highly purified Ml protein from calf thymus (32, 33). With regards to the reduction of ADP, the nucleotides GTP and dGTP both activate the reaction entirely by increasing the affinity of the enzyme for the substrate (Fig. 6); a similar observation for dGTP activation has been obtained with cell-free enzyme preparations as well (23). The intracellular concentrations of nucleotides shown in Table I may provide some insight into the relative physiological importance of the CDP and ADP reductase activators. The intracellular concentration of CDP (0.026 mM) was similar to the K,,, value estimated for an ATP-stimulated reaction (0.021 mM), but lower than the estimate for a dCTP-activated reaction (0.033 mM). Also, the ATP level within the cell is about 70% of the concentration required for maximum activation, whereas the dCTP level represents only about 2% of the amount required for optimal enzyme activation. These observations, coupled with the finding that ATP induced a greater rate of CDP reduction than dCTP, strongly indicate that ATP is a much more physiologically important regulator of CDP reduction than dCTP in intact cells. The situation is not as clear with regard
IN
INTACT
MAMMALIAN
CELLS
15
to the reduction of ADP. The intracellular concentration of ADP (0.558 mM) indicates that the substrate concentration in the cell is several times higher than the Km values for ADP when dGTP (Km for ADP = 0.095 mM) and GTP (Km for ADP = 0.20 mM) are involved as activators; dGTP induced the greatest rate of ADP reduction, with GTP exhibiting 65% of this rate. However, the intracellular pool of dGTP was undetectable while GTP was present at 110% of the level required to induce its maximum effect. Thus dGTP may be the most potent activator of ADP reduction, but its low intracellular level may mean that GTP is the major in vivo activator of ADP reductase. One important consideration must be paid to the interpretation of the intracellular nucleotide pool levels discussed above. Whereas the values presented represent the average concentration throughout a cell, there is a great deal of evidence that intracellular nucleotide pools are compartmentalized (34-36) in mammalian cells. If, as proposed (11,12), ribonucleotide reductase is a component of a large multienzyme complex which channels ribonucleoside diphosphates to DNA, the deoxynucleotide levels within this complex are very likely much higher than the average intracellular levels. It would be these locally concentrated levels which would regulate ribonucleotide reductase activity. Thus, even though dGTP and dCTP levels are low within the entire cell, they may be present within the complex in sufficient concentrations to affect the reduction of ribonucleotides. ACKNOWLEDGMENTS We appreciate research funds provided for this investigation by NSERC and the National Cancer Institute of Canada (J.A.W.), and we thank NSERC and the Faculty of Graduate Studies, University of Manitoba, for studentship support during the course of these studies (R.G.H.). J.A.W. is a recipient of a Terry Fox Cancer Research Scientist Award. REFERENCES 1. WRIGHT, J. A., HARDS, R. G., AND DICK, J. E. (1981) Advan Enzyme Reg. 19,105-127. 2. ELFORD, H. L., FREESE, M., PASSAMANI, E., AND MORRIS, H. P. (1970) .I Biol Ch 245, 522% 5223.
16
HARDS
AND
3. TAKEDA, E., AND WEBER, G. (1981) Life Sti 28, 1007-1014. 4. CHAN, T.-S. (1978) CeU 14, 523-530. 5. MEUTH, M. (1981) Mol. CeU. Bid 1, 652-660. 6. DICK, J. E., AND WRIGHT, J. A. (1982) Me& Ageing Den 20, 103-109. 7. CREASEY, D. C., PARFETT, C. L. J., AND WRIGHT, J. A. (1982) Proc Canad Fed Biol Sot. 25,172. 8. THELANDER, L., AND REICHARD, P. (1979) Annu. Rev. B&hem. 48, 133-158. 9. YOUDALE, T., MACMANUS, J. P., AND WHITFIELD, J. F. (1982) Canad J. Biochem 60,463-470. 10. CORY, J. G., AND FLEISCHER, A. E. (1982) Arch. Biodem. Biopfiys. 217, 546-551. 11. REDDY, G. P. V., AND PARDEE, A. B. (1982) J. Bid Chem. 257, 12526-12531. 12. FORSDYKE, D. R. AND SCOTT, F. W. (1980) in Cell Compartmentation and Metabolic Channeling (Noyer, L., Lynen, F. and Mothes, K., eds.), pp. 177-185, Elsevier/North-Holland, Amsterdam/ New York. 13. ELFORD, H. L. (1974) Arch Biochem Biophys. 163, 537-543. 14. WRIGHT, J. A. (1983) Pharm Theraped. 22, 81102. 15. HUNTING, D., AND HENDERSON, J. F. (1982) CRC Grit. Rev. B&hem. 13, 325-348. 16. ARAGON, J. J., FELIO, J. E., FRENKEL, R. A., AND SOLS, A. (1980) Proc Nat1 Acad Sti USA 77, 6324-6328. 17. LEWIS, W. H., KUZIK, B. A., AND WRIGHT, J. A. (1978) J. CelL Physiol 94, 287-298. 18. HARDS, R. G., AND WRIGHT, J. A. (1983) Arch Biochem Biophys. 220,576-583. 19. WRIGHT, J. A. (1973) J. Cell Bid 56, 666-675. 20. HARDS, R. G., AND WRIGHT, J. A. (1981) J. Cell Physiol 106, 309-319. 21. GARRETT, C., AND SANTI, D. V. (1979) Ad B&hem 99, 268-273.
WRIGHT 22. CHANG, C.-H., ANLI CHENG, Y.-C. (1979) Cancer Res. 39, 5081-5086. 23. MOORE, E. C., AND HURLBERT, R. B. (1966) J. Bid Chem. 241, 4802-4809. 24. WEBER, M. J., AND EDLIN, G. (1971) J. Bid Chem+ 246, 1828-1833. 25. COHEN, M. B., MAYBAUM, J., AND SADEE, W. (1980) J. Chrornutogr. 198, 435-441. 26. KINAHAN, J. J., PAVELIC, Z. P., LEONARD, R. J., BLOCK, A., AND GRINDEY, B. G. (1980) Cancer Rex 40, 598-603. 27. MAYBAUM, J., KLEIN, F. K., AND SADEE, W. (1980) J. Chromatogr. 188, 149-158. 28. THELANDER, L. (1974) J. Biol C~QWL 249, 48584862. 29. HOPPER, S. (1972) J. Bid Chem 247, 3336-3340. 30. MATTALIANO, R. J., SLOAN, A. M., PLUMER, E. R., AND KLIPPENSTEIN, G. L. (1981) Biochem. Bk+ phys. Rex Commun 102. 667-674. 31. ERIKSSON, S., GUDAS, L. J., CLIFT, S. M., CARAS, I. W., ULLMAN, B., AND MARTIN, D. W., JR. (1981) J. Bid Chem 256,10193-10197. 32. THELANDER, L., ERICKSSON, S., AND AKERMAN, M. (1980) J. Biol Chem. 255, 7426-7432. 33. VON DOBELN, U. (1977) Biochemistry 16,4368-4371. 34. GENCHEV, D. D., KERMEKCHIEV, M. B., AND HADJIOLOV, A. A. (1980) Biochem J. 188,85-90. 35. PIPER, P. W. (1981) FEBS Z&t. 131, 373-376. 36. SKOOG, L., AND BJURSELL, G. (1974) J. Biol Chem 249, 6434-6438. 37. Fox, T. O., AND PARDEE, A. B. (1970) Science 167, 80-82. 38. NICANDER, B., AND REICHARD, P. (1981) Biochem Biophys. Res. Cwrnmun 103, 148-155. 39. FUJIOKA, S., AND SILBER, R. (1970) J. Bid C&m. 245, 1688-1693. 40. WIEROWSKI, J. V., LAWTON, K. G., HOCKENSMITH, J. W., AND BAMBARA, R. A. (1983) J. Bid Chem 258. 6250-6254.
Ribonucleotide Reductase Activity in Intact Mammalian Cells: Stimulation of Enzyme Activity by MgC12, Dithiothreitol, and Several Nucleotides ROBERT Departments
of Mkmbio~ Cell Bidosy,
G. HARDS
and Biochemistry, 100 Olivia Street,
Received
June
AND
JIM A. WRIGHT’
University of Manitoba, Winnipeg, Manitoba R3E
30, 1983, and in revised
form
and Manitoba OV9, Canada
January
Institute
of
9, 1934
An intact cell assay system based on Tween-30 permeabilization was used to investigate ribonucleotide reductase activity in Chinese hamster ovary cells. Dithiothreitol, a reducing agent, is required for optimum activity. Analysis of dithiothreitol stimulation of CDP and ADP reductions indicated that in both cases the reducing agent served only to increase the reaction rate without altering the affinity of the enzyme for substrates. Magnesium chloride significantly stimulated the reduction of CDP but not ADP; this elevation in CDP reduction was due to an increase in both the affinity of the enzyme for substrate and the I’,,,,,. In addition to ATP and dGTP, well-known activators of CDP and ADP reductase activities, it was found that dCTP and GTP were also able to activate CDP and ADP reductase activities, respectively. For the dCTP-activated reaction the I’,,,, was 0.158 nmol dCDP formed 5 X lo6 cells-’ h-’ and the K, was 0.033 mM CDP, while for the GTP-activated reduction a V,,, of 0.667 nmol dADP formed 5 X lo6 cells-’ h-’ and a K, of 0.20 IIIM ADP were observed. Kinetic analysis revealed that dCTP, dGTP, and GTP stimulate ribonucleotide reduction solely by increasing the affinity of the enzyme for substrate without affecting the V,, of the respective reactions. ATP behaves in a different manner as it stimulates CDP reduction by altering both the affinity of the enzyme for substrate and the V,,. Cellular concentrations of riboand deoxyribonucleoside di- and triphosphate pools were measured to help evaluate the relative physiological importance of the nucleotide activators. These determinations, along with the reaction kinetic studies, strongly imply that ATP is a much more important regulator of CDP reduction than dCTP, whereas GTP may serve as well or better than dGTP as the in vivo activator of ADP reduction.
A continuous and balanced supply of deoxyribonucleoside triphosphates originating from the reduction of ribonucleotides is required during DNA synthesis. In mammalian cells this reduction occurs with ribonucleoside diphosphates and is carried out by ribonucleotide reductase (EC 1.17.4.1). This enzyme, which is likely a rate-limiting step in DNA synthesis and therefore in cell proliferation, occupies a key position in the biology of the cell (1). Alterations in the reduction of ribonucleotides have wide-ranging consequences. 1 To whom
correspondence
should
For example, the enzyme appears to play an important role in tumor transformation (2, 3), certain immunodeficiency diseases in man (4), alterations in the spontaneous mutation rates of cultured cells (5), the senescence of human diploid fibroblasts (6), and perhaps in certain aspects of cellular differentiation (7). In keeping with the importance of ribonucleotide reductase activity in the physiology of the cell, enzyme specificity and activity is strictly regulated in a very complex manner by nucleoside triphosphate effecters (1,9). Furthermore, structural features of the enzyme are unusual @-lo), and, during DNA synthesis,
be addressed. 9
ooo3-9361/34 Copyright All rights
$3.00
0 1924 by Academic Press. Inc. of reproduction in any form reserved.
10
HARDS
AND
the reductase appears to be physically and functionally associated with other enzymes of DNA synthesis in a multienzyme complex which channels deoxyribonucleotides to the replication forks (11-13). From biological, regulatory, and structural points of view, ribonucleotide reductase is one of the most complex enzymes in the cell (14). The properties of mammalian ribonucleotide reductase have been investigated, for the most part, in cell and tissue extracts. Several models based on these results have been proposed to account for the in vivo regulation of the enzyme activity (8,15). There is increasing evidence, however, that the properties of an enzyme can be affected by the intracellular environment (16). For example, the dilution that accompanies cell breakage can significantly affect enzyme characteristics. This is likely to be especially true for an enzyme as complex as ribonucleotide reductase. Therefore, to study this activity in a situation as close to the normal in vivo environment as possible, this laboratory has developed an intact cell assay system (1,1’7,18) based upon Tween-80 permeabilization. In this investigation, we used the assay system to study the intracellular properties of CDP and ADP reductase activities in the presence of various activators, and to examine the possibility that dCTP and GTP may be physiologically important regulators of enzyme activity. In addition, ribo- and deoxyribonucleotide pool sizes have been determined and used to evaluate kinetic findings. MATERIALS
AND
METHODS
Chinese hamster ovary (CHO)’ cells were cultured at 3’7°C on the surface of plastic tissue culture plates (LUX Scientific) or glass bottles (Brockway Glass Co.) in o-minimal essential medium (a-MEM, Flow Laboratories) supplemented with antibiotics and 10% (v/v) fetal calf serum (FCS, Gibco) as described in previous publications (19, 20). The cells grew with a doubling time of 16 h.
‘Abbreviations used: CHO, Chinese hamster ovary; (u-MEM, a-minimal essential medium; FCS, fetal calf serum; Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid, D’IT, dithiothreitol; PBS, phosphatebuffered saline.
WRIGHT Intracellular ribonucleotide reductase activity was determined in CHO cells made permeable to ribonucleotides as described previously (&l&20). In brief, exponentially growing cells were plated at a density of 5 x 106/150-mm culture plate containing growth medium and incubated at 37°C for 40 h to achieve a density of 2.2 f 0.2 X 10’ cells/plate. The cells were removed with trypsin solution, centrifuged, washed with growth medium, and resuspended at 10’ cells/ ml of permeabilizing buffer consisting of 1% Tween80 (J. T. Baker Co.), 0.25 M sucrose, 50 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (Hepes) buffer, pH 7.2, and 2 mM dithiothreitol (D’IT). The cells were incubated at 22°C for 30 min with occasional agitation and then centrifuged. The cell pellet was resuspended in fresh permeabilizing buffer at a concentration of 2.5 X lo7 cells/ml. Aliquots of 5 X lo6 cells were then added to assay tubes and intracellular CDP and ADP reductase activities were determined. Standard reactions contained either 1.5 mM ATP or 0.5 mM dGTP as activators for CDP or ADP reductions, respectively. A recent evaluation of the intact cell assay system has shown that potential problems involving nucleotide pools and nucleotide metabolizing enzymes do not present serious problems for examining the reductase in the physiologically relevant environment of the whole cell (18). In the present study, determinations of nucleotide concentrations present during the assay period were made as previously described (18) and taken into account to obtain corrected estimations of ribonucleotide reductase activity as outlined earlier (18). To estimate ribo- and deoxyribonucleoside di- and triphosphate pool sizes, the nucleotides were extracted from the CHO cells using the method of Garrett and Santi (21). The cells were grown to a density of 2.2 -C 0.2 X 10’ cells/l50-mm plastic tissue culture plate in a-MEM plus 10% FCS. For each extraction, the cells from five such plates were removed with trypsin solution, washed twice with ice-cold PBS, counted with a Coulter particle counter (Coulter Electronics Ltd.), and resuspended in PBS at a concentration of 0.5 to 3.0 X 10s cells/ml. The PBS into which the cells were resuspended contained [‘Hluridine (130,000 cpm/ml) as a concentration marker and aliquots of PBS were removed for scintillation counting. Seventy percent perchloric acid was then added to the suspension to give a final concentration of 0.5 M. After 30 min on ice, the precipitate was removed by centrifugation in an Eppendorf microcentrifuge, 10 pl of bromocresol blue was added, and the supernatant was neutralized with 4 M KOH containing 0.4 M KHzPO,. The neutralized extract was kept on ice for at least 10 min to allow for the precipitation of potassium perchlorate. The solution was again cleared by centrifugation and an aliquot of the supernatant was analyzed for radioactivity in order to determine dilution factors. As deoxyribonucleotides are often obscurred by the
RIBONUCLEOTIDE
REDUCTASE
ACTIVITY
presence of ribonucleotides, analysis of each required the separation of the two. The method of Garrett and Santi (21), which involves destruction of the ribonucleotides to allow for detection and quantitation of the deoxyribonucleotides, was utilized with slight modifications. Forty microliters of 0.5 M sodium periodate was added to 1 ml of neutralized cell extract. After 5 min, 50 ~1 of 4 M methylamine, which had been slowly brought to pH 7.5 with phosphoric acid, was added. The reactants were mixed and incubated at 37°C for 60 min. Ten microliters of 1 M rhamnose was then added to destroy any remaining periodate and the samples were immediately put on ice. In experiments where there was less than 1 ml extract, proportionally smaller volumes of reagents were used. The solution was cleared by centrifugation and an aliquot was removed for scintillation counting to determine dilution factors. Aliquots of the ribonucleotide plus deoxyribonucleotideand deoxyribonueleotide-containing solutions were analyzed using the HPLC method described previously (18). Separation was achieved using a Partisil10125 SAX (Whatman) strong anion-exchange column (25 cm X 4.6 mm). The initial eluant, 0.007 M KHzPOI(HPLC grade, Fisher Chemical Co.) and 0.007 M KCI(ACS grade, J. T. Baker Co.) was adjusted to pH 4.0 with HCl. The final eluant, 0.25 M KH,PO* and 0.5 M KCI, was adjusted to pH 5.0 with NaOH. An isocratic elution on the low concentration buffer lasting 20 min was followed by a linear gradient from 0% of the final eluant to 100% of the final eluant run in 35 min. Finally, an isocratic elution on the high concentration buffer was run for 15 min. The flow rate was 2.0 ml/min. All buffers were filtered through 0.22-pm Millipore membrane filters (HA, Millipore), and degassed prior to use. Identification was by retention times as compared to prepared standards. For radioactivity determinations of fractions, the fractions collected were transferred to scintillation vials containing 10 ml ACS cocktail and radioactivity was determined with a Beckman Model LS230 liquid scintillation spectrophotometer. Radiochemicals were purchased from Amersham. All other biochemicals were obtained from the Sigma Chemical Company.
IN
Since ribonucleotide reductase catalyzes the reduction of both pyrimidine and purine ribonucleotides (l), the properties of this activity had been examined in the presence of a pyrimidine (CDP) and a purine (ADP) substrate. Enhanced Activity with Nonnucleotide COW&pOUnds While MgClz and the reducing agent, dithiothreitol, are not essential for the re-
MAMMALIAN
CELLS
11
duction of CDP in intact CHO cells (1,20), they do significantly enhance the activity. In order to determine how these compounds stimulate reduction, double-reciprocal plots of velocity against CDP concentration at several fixed levels of MgClz and DTT were constructed (Fig. 1). Magnesium ions increased both the rate of reduction and the affinity of the enzyme for CDP. As expected, DTT, which acts as a hydrogen donor for the reaction, increased only the velocity of the reaction and did not affect the affinity for CDP. Double-reciprocal plots of the variation in velocity of CDP reduction at varying MgCl, or DTT concentrations at several fixed levels of CDP (plots not shown) indicated that in neither case did the substrate level affect the affinity of the enzyme for MgClz or DTT. The K, values determined for the two compounds are 2.4 mM MgClz and 2.8 mM DTT. Although MgCl, does not enhance ADP reductase activity (20), a reducing agent such as DTT is required for optimum activity (1,20). As shown in Fig. 2, DTT stimulates the reduction of ADP by increasing the rate of reduction. As was the case with
-80
RESULTS
INTACT
0
80
160 CDP hM-‘1
FIG. 1. (a) Double-reciprocal plot of the variation in rate of CDP reduction with CDP concentration at several fixed levels of MgCl,. Assays were performed in the absence (0) or presence of 2 mM (A), 2.5 mM (Cl) and 6 mM (0) MgClz. Activity was expressed as nmol dCDP formed 5 X lo6 cells-’ h-l. (b) Doublereciprocal plot of the variation in rate of CDP reduction with CDP concentration at several fixed levels of DTT. Assays were performed in absence (0) or presence of 1.5 mM (A), 3 mM (0) and 6 m&i (0) DTT. Activity was expressed as nmol dCDP formed 5 X lo6 cells-’ h-i.
12
HARDS
AND
r
WRIGHT
2
a
B i2 I: 0.15 8 “Do 2 9 0.10
Y
8B
0
3 0 3 *
P\ w &
0.050
i
0
2
4
GTP (mM)
dCTP hM ADP hd’)
FIG. 2. Double-reciprocal plot of the variation in rate of ADP reduction with ADP concentration at several fixed levels of DTT. Assays were performed in the presence of 1.5 mM (0), 3 mM (A), 4.5 rnrd (0), and 6 mM (0) DTT. Activity was expressed as nmol dADP formed 5 X lo6 cells-’ h-l.
CDP reduction, DTT did not affect the enzyme affinity for ADP. Rearranging the data to give a double-reciprocal plot of velocity against DTT concentration at several fixed levels of ADP (plots not shown) indicated, as expected, that ADP had no effect on the interaction of ribonucleotide reductase with DTT. A K, value of 2.7 mM DTT was determined. With regard to the reducing agent, DTT, CDP, and ADP reductase activities respond in an identical manner.
Nucleotide
FIG. 3. (a) Response of CDP reductase activity to dCTP concentrations. Standard CDP reductase assay conditions were used except for the omission of ATP and the presence of varying concentrations of dCTP. (b) Response of ADP reductase activity to GTP concentrations. Standard ADP reductase assay conditions were used except for the omission of dGTP and the presence of varying concentrations of GTP.
previously characterized. It is apparent from Fig. 3 that maximal enzyme activity occurs in the presence of about 3.6 mM dCTP and 0.2 mM GTP. Figure 4 shows that the reduction of CDP in the presence of 3.6 mM dCTP exhibits a Km of 0.033 mM CDP and a V,, value of 0.158 nmol of dCDP formed 5 X lo6 cells-’ h-l. In comparison, the values observed (18, 20) with 1.5 mM ATP as an activator of the reaction were a Km of 0.021 mM CDP and a Vmax of 0.49 nmol dCDP formed 5 X lo6
Activators
Previous studies with intact mammalian cells and cell-free extracts have shown that ribonucleotide reductase exhibits an absolute requirement for allosteric activators (1, 8, 1’7, 20). It has been well documented that ATP acts as a positive effector of CDP reduction and dGTP activates the reduction of ADP. In preliminary experiments (1) with intact CHO cells, we have also observed that dCTP and GTP stimulate the reductions of CDP and ADP, respectively. The GTP effect, although noted in other investigations (22), has generally not been well characterized or considered in models of ribonucleotide reductase regulation. Deoxy-CTP stimulation has been noted by only one other group (39) but has not been
‘0 ADP (md’)
0
60
120
130
CDP hd)
FIG. 4. (a) Double-reciprocal plot of the variation in rate of ADP reduction with ADP concentrations. Reduction was activated by 0.2 mM GTP. Activity was expressed as nmol dADP formed 5 X 10’ cells-’ h-r. (b) Double-reciprocal plot of the variation in rate of CDP reduction with CDP concentrations. Reduction was activated by 3.6 mre dCTP. Activity was expressed as nmol dCDP formed 5 X lo6 cells-’ h-l.
RIBONUCLEOTIDE
REDUCTASE
ACTIVITY
cells-’ h-‘. In the presence of 0.20 mM GTP, double-reciprocal plots of ADP reduction showed a Km value of 0.20 mM ADP and a V,,,, of 0.67 nmol dATP formed 5 X lo6 cells-’ h-l. In comparison, kinetic studies with optimum dGTP concentrations gave a Km value of 0.095 mM ADP and a V,,, of 1.00 nmol dADP formed 5 X lo6 cells-’ h-’ (18, 20). To learn more about the activation of CDP and ADP reductase in intact cells, the four positive effecters of these reactions were further examined by constructing double-reciprocal plots of velocity against substrate concentrations at various fixed levels of effecters. As shown in Fig. 5a, ATP increases both the rate of CDP reduction and the affinity of the enzyme for CDP. A similar effect for ATP has been reported with the enzyme from cell-free extracts (23). However, it appears that dCTP activates the reduction of CDP solely by increasing the affinity of the enzyme for the substrate (Fig. 5b). The effects of dGTP and GTP on the reduction of ADP are shown in Fig. 6. Similar to the studies described above with dCTP, it appears that both GTP and dGTP activate intracellular ADP reduction entirely by increasing enzyme substrate affinity. The K, values for
30 CDP
0
80
160
hbl-‘1
FIG. 5. (a) Double-reciprocal plot of the variation in rate of CDP reduction with CDP concentration at several fixed levels of ATP. Assays were performed in the presence of 0.6 mM (0), 0.9 mM (A), 1.2 mM (O), and 1.5 mM (0) ATP. Activity was expressed as nmol dCDP formed 5 X 106cells-’ h-l. (b) Double-reciprocal plot of the variation in rate of CDP reduction with CDP concentration at several fixed levels of dCTP. Assays were performed in the presence of 0.9 mbi (O), 1.35 mM (Cl), and 3.6 mM (0) dCTP. Activity was expressed as nmol dCDP formed 5 X lo6 cells-’ h-l.
IN
INTACT
MAMMALIAN
13
CELLS
,40
0
40
80
120
FIG. 6. (a) Double-reciprocal plot of the variation in rate of ADP reduction with ADP concentration at several fixed levels of dGTP. Assays were performed in the presence of 0.046 mM (0), 0.092 mM (A), and 0.134 mM (Cl) dGTP. Activity was expressed as nmol dADP formed 5 X lo6 cells-’ h-i. (b) Double-reciprocal plot of the variation in rate of ADP reduction with ADP concentration at several fixed levels of GTP. Assays were performed in the presence of 0.18 mM (0), 0.36 mM (A), and 0.54 mM (Cl) GTP. Activity was expressed as nmol dADP formed 5 X lo6 cells-’ h-l.
the various activators were 1.37 mM ATP, 1.13 mM dCTP, 0.096 mM GTP, and 0.012 mM dGTP.
Nucleotide
Pools
Cellular concentrations of ribo- and deoxyribonucleoside di- and triphosphate pools were determined to help evaluate the relative significance of the various effecters as regulators in the cell. The pool sizes were determined in logarithmically growing cells cultured in a manner identical to the cells routinely used in enzyme assays. The nucleotides were extracted and analyzed as described under Materials and Methods and the results of these studies are shown in Table I. The deoxyribonucleoside triphosphate levels found for CHO cells were well within the range of values obtained with many types of mammalian cells (e.g. (21, 24-27)). The CHO ribonucleoside triphosphate pools were also within the range of values reported for other animal cells, except that the values were towards the lower end of the range (e.g., (21, 24-27)). Although relatively few studies of intracellular ribonucleoside diphosphate pool sizes have been carried out, those that have been reported agree rea-
14
HARDS TABLE NUCLEOTIDE
AND
I
POOL SIZES OF EXPONENTIALLY GROWING CHO CELLS Intracellular concentration
Nucleotide
nmol/106 f f -+ f
cells
mM*
GTP ATP CTP UTP
0.334 1.589 0.604 0.387
0.038 0.258 0.379 0.030
0.223 1.059 0.403 0.238
GDP ADP CDP UDP
0.157 + 0.025 1.060 0.048 f 0.009 0.121 f 0.011
0.106 0.558 0.026 0.075
dGTP dATP dCTP dTTP
ND” 0.028 f 0.007 0.105 f 0.065 0.053 f 0.035
0.019 0.070 0.035
Note. Results are the averages trials. ‘Not detected. *Based on a CHO cell volume
of two independent
of 1.7 pl (37).
sonably well with the values found for CDP and UDP, whereas the purine ribonucleoside diphosphates, ADP and GDP, were detected in slightly higher levels in CHO cells as compared to many other mammalian sources (24, 25, 27). DISCUSSION
While the two small proteins, thioredoxin through thioredoxin reductase and glutaredoxin via glutathione reductase, function as natural hydrogen carriers in the reduction of ribonucleotides, assays of ribonucleotide reductase utilize dithiols such as dithiothreitol as direct hydrogen donors. In this study, we have shown that dithiothreitol stimulates both CDP and ADP reduction by increasing the velocity of the reactions (Figs. lb and 2). In neither case did the dithiol affect the affinity of the enzyme for the substrates. These findings are consistent with the proposed role for DTT in these reactions. The dithiol apparently reduces a disulfide at the active site, which in turn reduces the ribonucle-
WRIGHT
otide substrate (28). Our results suggest that the reduction of the active site disulfide is required for enzyme activity but it is apparently not needed for substrate binding. Magnesium chloride exerted a dual effect on CDP reduction as it increased both the velocity of the reaction and the affinity of the enzyme for CDP (Fig. la). Magnesium ions are thought to promote the association of nonidentical subunits (Ml and M2) of mammalian ribonucleotide reductase to form a holoenzyme capable of reducing CDP (29,30). This association could explain the observed elevation in velocity caused by MgCla. The increased affinity of the enzyme induced by MgClz could be due to several reasons. For example, it appears that nucleotides bind only to the Ml subunit of the reductase (8, 31, 32), and it is possible that an association of subunits induced by magnesium ions causes a conformational change in Ml, which facilitates the binding of CDP. However, optimal binding of CDP may require interaction between Ml and M2; therefore, it is also possible that a MgClz-induced association of these subunits could lead to increased binding and stimulation of enzyme activity. Although ATP and dGTP are wellknown activators of CDP and ADP reductase activities, respectively, we have observed that dCTP can activate CDP reduction while GTP is a significant activator of ADP reduction in intact hamster cells. It is worth noting that HPLC analysis of the nucleotides present at various times during the reaction have been performed as described in detail in a previous report (18). These studies indicated that the activations observed with dCTP and GTP were due to their direct effects on activity and not to a conversion to another nucleotide or to a stimulation in the reduction of a second substrate. Activation of CDP reduction by ATP and dCTP seemed to involve two different processes. While ATP increased both the affinity of the reductase for CDP and the rate of reduction (Fig. 5a), the sole effect of dCTP was to elevate the affinity of the enzyme for CDP (Fig. 5b). Thus, it would seem that in its native conformation ri-
RIBONUCLEOTIDE
REDUCTASE
ACTIVITY
bonucleotide reductase is capable of reducing CDP but the Km value for CDP is prohibitively high; binding of dCTP to the enzyme would facilitate binding of the substrate without altering the maximum velocity obtainable. It has been suggested (38, 40) that binding of dCTP to an allosteric site on a multienzyme complex involved in DNA synthesis increases the affinity of the complex for nucleoside diphosphates. The kinetic studies with dCTP fit this suggested role, although more direct evidence is required. When ATP is used as an activator it apparently induces a conformational change which not only enhances binding of CDP but also increases the maximum velocity that can be reached. The pattern of ATP activation in intact cells is similar to that seen with cell-free preparations from rat tumor (23). Conformational changes during the binding of substrates in the presence of effecters of the enzyme have been observed in the Escherichia coli system and with highly purified Ml protein from calf thymus (32, 33). With regards to the reduction of ADP, the nucleotides GTP and dGTP both activate the reaction entirely by increasing the affinity of the enzyme for the substrate (Fig. 6); a similar observation for dGTP activation has been obtained with cell-free enzyme preparations as well (23). The intracellular concentrations of nucleotides shown in Table I may provide some insight into the relative physiological importance of the CDP and ADP reductase activators. The intracellular concentration of CDP (0.026 mM) was similar to the K,,, value estimated for an ATP-stimulated reaction (0.021 mM), but lower than the estimate for a dCTP-activated reaction (0.033 mM). Also, the ATP level within the cell is about 70% of the concentration required for maximum activation, whereas the dCTP level represents only about 2% of the amount required for optimal enzyme activation. These observations, coupled with the finding that ATP induced a greater rate of CDP reduction than dCTP, strongly indicate that ATP is a much more physiologically important regulator of CDP reduction than dCTP in intact cells. The situation is not as clear with regard
IN
INTACT
MAMMALIAN
CELLS
15
to the reduction of ADP. The intracellular concentration of ADP (0.558 mM) indicates that the substrate concentration in the cell is several times higher than the Km values for ADP when dGTP (Km for ADP = 0.095 mM) and GTP (Km for ADP = 0.20 mM) are involved as activators; dGTP induced the greatest rate of ADP reduction, with GTP exhibiting 65% of this rate. However, the intracellular pool of dGTP was undetectable while GTP was present at 110% of the level required to induce its maximum effect. Thus dGTP may be the most potent activator of ADP reduction, but its low intracellular level may mean that GTP is the major in vivo activator of ADP reductase. One important consideration must be paid to the interpretation of the intracellular nucleotide pool levels discussed above. Whereas the values presented represent the average concentration throughout a cell, there is a great deal of evidence that intracellular nucleotide pools are compartmentalized (34-36) in mammalian cells. If, as proposed (11,12), ribonucleotide reductase is a component of a large multienzyme complex which channels ribonucleoside diphosphates to DNA, the deoxynucleotide levels within this complex are very likely much higher than the average intracellular levels. It would be these locally concentrated levels which would regulate ribonucleotide reductase activity. Thus, even though dGTP and dCTP levels are low within the entire cell, they may be present within the complex in sufficient concentrations to affect the reduction of ribonucleotides. ACKNOWLEDGMENTS We appreciate research funds provided for this investigation by NSERC and the National Cancer Institute of Canada (J.A.W.), and we thank NSERC and the Faculty of Graduate Studies, University of Manitoba, for studentship support during the course of these studies (R.G.H.). J.A.W. is a recipient of a Terry Fox Cancer Research Scientist Award. REFERENCES 1. WRIGHT, J. A., HARDS, R. G., AND DICK, J. E. (1981) Advan Enzyme Reg. 19,105-127. 2. ELFORD, H. L., FREESE, M., PASSAMANI, E., AND MORRIS, H. P. (1970) .I Biol Ch 245, 522% 5223.
16
HARDS
AND
3. TAKEDA, E., AND WEBER, G. (1981) Life Sti 28, 1007-1014. 4. CHAN, T.-S. (1978) CeU 14, 523-530. 5. MEUTH, M. (1981) Mol. CeU. Bid 1, 652-660. 6. DICK, J. E., AND WRIGHT, J. A. (1982) Me& Ageing Den 20, 103-109. 7. CREASEY, D. C., PARFETT, C. L. J., AND WRIGHT, J. A. (1982) Proc Canad Fed Biol Sot. 25,172. 8. THELANDER, L., AND REICHARD, P. (1979) Annu. Rev. B&hem. 48, 133-158. 9. YOUDALE, T., MACMANUS, J. P., AND WHITFIELD, J. F. (1982) Canad J. Biochem 60,463-470. 10. CORY, J. G., AND FLEISCHER, A. E. (1982) Arch. Biodem. Biopfiys. 217, 546-551. 11. REDDY, G. P. V., AND PARDEE, A. B. (1982) J. Bid Chem. 257, 12526-12531. 12. FORSDYKE, D. R. AND SCOTT, F. W. (1980) in Cell Compartmentation and Metabolic Channeling (Noyer, L., Lynen, F. and Mothes, K., eds.), pp. 177-185, Elsevier/North-Holland, Amsterdam/ New York. 13. ELFORD, H. L. (1974) Arch Biochem Biophys. 163, 537-543. 14. WRIGHT, J. A. (1983) Pharm Theraped. 22, 81102. 15. HUNTING, D., AND HENDERSON, J. F. (1982) CRC Grit. Rev. B&hem. 13, 325-348. 16. ARAGON, J. J., FELIO, J. E., FRENKEL, R. A., AND SOLS, A. (1980) Proc Nat1 Acad Sti USA 77, 6324-6328. 17. LEWIS, W. H., KUZIK, B. A., AND WRIGHT, J. A. (1978) J. CelL Physiol 94, 287-298. 18. HARDS, R. G., AND WRIGHT, J. A. (1983) Arch Biochem Biophys. 220,576-583. 19. WRIGHT, J. A. (1973) J. Cell Bid 56, 666-675. 20. HARDS, R. G., AND WRIGHT, J. A. (1981) J. Cell Physiol 106, 309-319. 21. GARRETT, C., AND SANTI, D. V. (1979) Ad B&hem 99, 268-273.
WRIGHT 22. CHANG, C.-H., ANLI CHENG, Y.-C. (1979) Cancer Res. 39, 5081-5086. 23. MOORE, E. C., AND HURLBERT, R. B. (1966) J. Bid Chem. 241, 4802-4809. 24. WEBER, M. J., AND EDLIN, G. (1971) J. Bid Chem+ 246, 1828-1833. 25. COHEN, M. B., MAYBAUM, J., AND SADEE, W. (1980) J. Chrornutogr. 198, 435-441. 26. KINAHAN, J. J., PAVELIC, Z. P., LEONARD, R. J., BLOCK, A., AND GRINDEY, B. G. (1980) Cancer Rex 40, 598-603. 27. MAYBAUM, J., KLEIN, F. K., AND SADEE, W. (1980) J. Chromatogr. 188, 149-158. 28. THELANDER, L. (1974) J. Biol C~QWL 249, 48584862. 29. HOPPER, S. (1972) J. Bid Chem 247, 3336-3340. 30. MATTALIANO, R. J., SLOAN, A. M., PLUMER, E. R., AND KLIPPENSTEIN, G. L. (1981) Biochem. Bk+ phys. Rex Commun 102. 667-674. 31. ERIKSSON, S., GUDAS, L. J., CLIFT, S. M., CARAS, I. W., ULLMAN, B., AND MARTIN, D. W., JR. (1981) J. Bid Chem 256,10193-10197. 32. THELANDER, L., ERICKSSON, S., AND AKERMAN, M. (1980) J. Biol Chem. 255, 7426-7432. 33. VON DOBELN, U. (1977) Biochemistry 16,4368-4371. 34. GENCHEV, D. D., KERMEKCHIEV, M. B., AND HADJIOLOV, A. A. (1980) Biochem J. 188,85-90. 35. PIPER, P. W. (1981) FEBS Z&t. 131, 373-376. 36. SKOOG, L., AND BJURSELL, G. (1974) J. Biol Chem 249, 6434-6438. 37. Fox, T. O., AND PARDEE, A. B. (1970) Science 167, 80-82. 38. NICANDER, B., AND REICHARD, P. (1981) Biochem Biophys. Res. Cwrnmun 103, 148-155. 39. FUJIOKA, S., AND SILBER, R. (1970) J. Bid C&m. 245, 1688-1693. 40. WIEROWSKI, J. V., LAWTON, K. G., HOCKENSMITH, J. W., AND BAMBARA, R. A. (1983) J. Bid Chem 258. 6250-6254.