- Email: [email protected]
Dynamcis of the thymidine triphosphate pool during the cell cycle of synchronized 3T3 mouse fibroblasts
Mutation Research, 200 (1988) 37-43 Elsevier
37
MTR 02313
Dynamics of the thymidine triphosphate pool during the cell cycle of synchronized 3T3 mouse fibroblasts Giannis Spyrou and Peter Reichard Department of Biochemistry, Medical Nobel Institute, Karolinska lnstitutet, Box 60400, S-104 01 Stockholm (Sweden)
(Received 28 January 1988) (Accepted 19 February 1988)
Keywords: dTI'P pool; Turnover; Cell synchronization; DNA repair
Summary To investigate whether resting cells of 3T3 mouse fibroblasts carry out de novo synthesis of deoxyribonucleoside triphosphates, we determined the turnover of the thymidine triphosphate pool of G O cells obtained by starvation of cultures for platelet-derived growth factor. These cells were contaminated by less than 1% S-phase cells. In the absence of deoxyribonucleosides in the medium one million G Ocells contained 5 pmole of dTTP with a turnover of 0.09 pmole/min. S-phase cells in comparison contained a 20 times larger dTTP pool with a more than 200-fold faster turnover. Our results suggest that G O cells carry out a slow but finite de novo synthesis of deoxyribonucleoside triphosphates to satisfy the cells' requirement for DNA repair and mitochondrial DNA synthesis.
The synthesis and turnover of deoxynucleoside triphosphates (dNTPs) is tightly coupled to DNA synthesis (for a recent review see Reichard, 1988). The activities of several enzymes involved in dNTP synthesis increase greatly when resting cells are induced by different means to replicate or repair their DNA. Cells in S phase contain much larger dNTP pools than resting cells. In cycling cell populations, S-phase cells contain the largest pools and the highest activities of enzymes participating in dNTP synthesis. In recent reports from this
Correspondence: Dr. P. Reichard, Department of Biochemistry, Medical Nobel Institute, Karolinska Institutet, Box 60400, S-104 01 Stockholm (Sweden). Abbreviations: dNTP, deoxyribonucleoside triphosphate; PPP, platelet-poor plasma.
laboratory the turnover of the dTTP pool was investigated in cycling cells (Nicander and Reichard, 1983, 1985). Exponentially growing cultures with close to 60% cells in S phase contained a single kinetic dTTP pool whose turnover could be accounted for by its consumption for DNA replication. This suggested that dTFP turnover and synthesis occurred mainly, or even exclusively, in S-phase cells and raised the question whether resting cells carry out any de novo synthesis of dTTP (and other dNTPs). The answer to this question is of interest for understanding how cells obtain a balanced supply of dNTPs for DNA repair and also for the interpretation of experiments demonstrating the interference with DNA repair in resting cells by agents that block or upset the normal supply of dNTP by ribonucleotide reductase (Cohen and Thompson, 1986; Seto et al., 1986).
0027-5107/88/$03.50 © 1988 Elsevier Science Publishers B.V. (Biomedical Division)
38 Here we describe experiments concerning the turnover of the dTTP pools of resting and S-phase cell populations of 3T3 mouse fibroblasts. Synchronization was obtained by starvation of the cells in medium lacking platelet-derived growth factor followed by stimulation with complete serum (Pledger et al., 1977). Before stimulation, more than 99% of the cells were in the resting G O state, as judged from the lack of thymidine incorporation into nuclear DNA. Between 15 and 18 h after stimulation more than 95% were in S phase. In both instances the dTTP pools were labeled rapidly from [3H]thymidine and, after change of medium, the decay of radioactivity was used to measure the turnover of the pools. Similar experiments were made with G 1 cells 3-8 h after serum stimulation. The results suggest that both G O and G 1 cells contain small dTTP pools whose half-lives are about 10 times longer than that of S-phase cells. Materials and methods
Chemicals [Methyl-3H]thymidine with a specific activity of 25 C i / m m o l e (20000 c p m / p m o l e ) was obtained from Amersham. Polyd(AT), polyd(IC) and DNA polymerase I from E. coli for pool assays were from Boehringer-Mannheim. All other chemicals were of highest purity available.
Platelet-poor plasma Platelet-poor plasma (PPP) was prepared from blood of human blood donors at the Karolinska Hospital, Stockholm (Sweden) by slight modification of a published method (Pledger et al., 1977). Approximately 450 ml of blood was collected from 1 donor, without using a tourniquet, into a plastic bag containing 63 ml citrate buffer (15 mM citric acid, 96 mM sodium citrate, 20 mM sodium phosphate monobasic, 129 mM glucose). The bag was centrifuged immediately at room temperature for 5 min at 5000 × g in a clinical centrifuge, the plasma was transferred to a second empty bag and recentrifuged for 5 min at 5000 x g. The supernatant solution was centrifuged for 30 min at 28 000 x g with a final yield of 250 ml of PPP. This was heat-inactivated at 5 6 ° C for 30 min, centrifuged at 28 000 x g for 30 min, sterilized by
filtration through 0.2-/~m amicon filters and stored at - 7 0 ° C before use. When indicated PPP was dialyzed for 48 h against several changes of 0.15 M NaC1.
Growth and synchronization of cells Swiss 3T3 mouse fibroblasts were obtained from Dr. Angela Otto, Friedrich Miescher Institut, Basel (Switzerland). The cells were grown in Dulbecco's modified Eagle's medium (DMEM) with 10% inactivated fetal calf serum and maintained for at most 6 passages (never allowed to reach confluency) before use. For experiments with resting cells, 0.15-0.2 million cells were seeded onto a 5-cm petri dish in 5 ml D M E M containing 5% PPP. After 4 days at 37 ° C the cell number was not increased and less than 1% of the cells were in S phase. For experiments with S-phase cells, resting cells were stimulated with 5% inactivated human serum. After 18 h more than 95% of the cells were in S phase as judged from autoradiography. Cells were labeled for autoradiography by a 45-min pulse with labeled 1 /~M thymidine. The percentage of labeled nuclei was determined by counting at least 4000 cells of duplicate autoradiograms.
Incorporation of labeled thymidine into cells Three hours before the addition of isotope, the medium was reduced from 5 to 2.5 ml. The dishes were divided into 2 groups: one received labeled thymidine, while the other received an identical amount of unlabeled thymidine and served as a source of conditioned medium. To measure the decay of the label (started at arrows in Figs. 3, 4 and 6) the labeled medium was sucked off and replaced with conditioned medium. All manipulations were made rapidly inside the incubators. Incubations were at 37 ° C in a humidified 7% CO2 atmosphere. Duplicate samples were used for each time point. Following incubation, the dishes were immediately transferred to an ice bath, the medium was sucked off, the plates were washed 3 times with 3-5 ml ice-cold Tris saline, and then left tilted for 1 min before removing the remaining buffer. Nucleotides were extracted from the cells with 60% methanol and the size and the specific activity of the dTTP pools were determined as described (Lindberg and Skoog, 1970; Hellgren et
39 10 3
al., 1979). The remaining cell pellet was used to determine incorporation of radioactivity into D N A (Nicander and Reichard, 1985). All results are normalized to 106 cells.
i !
I- O.7 I I-0. $ e
to," 10 z .
d N T P pools in relation to DNA synthesis in synchronized 3T3 cells After 4 days incubation of sparse 3T3 cells in growth medium containing PPP less than 1% of the cells were replicating their D N A , as judged from autoradiography with [3H]thymidine. Such cells were stimulated to synthesize D N A by addition of medium containing 5% human serum and at time intervals during the ensuing 21 h their entrance into S phase was determined by pulsed autoradiography with thymidine and the size of their 4 d N T P pools was determined (Fig. 1). D N A synthesis started quite synchronously after 9 h, and between 15 and 18 h more than 95% of the cells were in S phase. As demonstrated earlier, entrance into S phase was accompanied by dramatic pool changes (Waiters et al., 1973; Skoog et al., 1973). In resting cells the size of the d G T P pool was too small to be measured accurately ( < 2 pmole/106). This pool also remained the smallest throughout the whole time. The largest pool was
t0 z
-e-dTTP, dATP,
u
-o-dCTP
/
~dGTP
*
o ~ ~-o~.
t l~
•
I I
~'X
~ e
"%
100
LS0
"7"
,
,
10:
"o
. . . .
/
101
I 1
+,'
~
i"
I
+'
IO Hours
Fig. 1. Deoxyribonucleoside triphosphate pools after serum stimulation of resting 3T3 cells. Resting cells were obtained by incubation of parallel cultures in DMEM containing 5% PPP. At time 0 the cells were shifted to medium containing 5% human serum and incubated at 37 °C for the indicated times before pool analyses. The amount of S-phase cells was determined by autoradiography after labeling for 45 min with [3H]thymidine.
1
O.S .
¢:,
Results
nl
u
a.
I-°'4 T=
o E 101.
I-o.3
x
I
I- o.1 ~--dTTP ,
10 o
0
;o
- i - d C TP ,
~dATP
;o
,
~dGTP
:0
i 0.0 8O
Hours
Fig. 2. Deoxyribonucleoside triphosphate pools in synchronized and cycling 3T3 cells. This experiment was similar to the one described in Fig. 1 except for the use of dialyzed PPP and human serum. Pool determinations were made during a 3-day period. The amount of cells was determined by counting with a hematometer.
dCTP. It had already increased after 9 h, before onset of D N A synthesis, and, after that time, its size mirrored the percentage of cells in S phase. dATP and d T T P followed each other closely, had started to increase at the 12-h time point and continued their increase also when part of the cells had left S phase. These pool fluctuations during the cell cycle are similar to earlier results obtained with other synchronized cell populations (Waiters et al., 1973; Skoog et al., 1973). At first sight the results found with synchronized cells are difficult to reconcile with data from rapidly growing, cycling cell populations, containing close to 60% S-phase cells. The discrepancy can be exemplified with the dCTP pool. In cycling 3T3 cell populations this pool is smaller (40 p m o l e / 1 0 6 ) than expected from the percentage of cells in S phase. In the experiment depicted in Fig. 2 changes in d N T P pools were analyzed during 3 days after serum stimulation to clarify this point. In this experiment we used dialyzed PPP and dialyzed serum to avoid complications that might arise from the possible presence of deoxyribonucleosides in the medium. Again the dCTP pool showed a sharp increase during the first S phase but then returned to more ' n o r m a l ' values by 36 h when the first wave of D N A synthesis had passed. We believe that the initial overshoot of dNTPs was caused by the fact that the synchronization
40 procedure had led to a small increase in ribonucleotide reductase activity already before the onset of DNA synthesis (data not shown). Due to the allosteric regulation of ribonucleotide reductase (Thelander and Reichard, 1979) this led to an overproduction of dCTP and only with time did the cells attain a better balance of the 4 dNTPs.
TABLE 1 RATES OF DNA SYNTHESIS IN G 0- A N D S-PHASE 3T3 CELLS The total radioactivity incorporated into DNA between 30 and 60 min (S phase) or 30 and 90 min (G O phase) was determined in the experiment described in the legend to Fig. 3. The rate of DNA synthesis was calculated from the radioactivity accumulated between the 2 time points in DNA divided by the specific activity of the dTTP pool. All values are for 106 cells.
Turnover of the dTTP pool during the cell cycle To measure the turnover of the dTTP pool cells are first labeled from [3H]thymidine followed by a chase with cold nucleoside. The turnover of the pool can then be calculated from the decay of isotope. In the closely related cycling 3T6 cells the half-life of dTTP was 4-5 min and the turnover closely matched incorporation into DNA, indicating that essentially all dTTP synthesized was used for DNA replication. With synchronized 3T3 ceils it was now possible to measure dTTP turnover in
1
1o'-
I I
o. Il,"0
'3 E
\.
10 3. 0 O.
Go-phase
E
\ X~x
X-,~ x
S-phase 10 2
' 0
I 30
'
! 60
'
I 90
'
I 120
'
I 150
'
I t80
rain Fig. 3. Turnover of the dTTP pool in G o and S-phase cells. Resting and S-phase cells were obtained as described in Fig. 1. The cells were first labeled with 0.3 ~tM [ 3H]thymidine (90 min for G O cells, 60 min for S-phase cells). At the arrows the cells were shifted to conditioned medium containing 0.3 /~M cold thymidine. The specific activity of dTTP was determined on duplicate samples removed during the labeling and chase periods and the values are plotted on a semiiog scale. • = G O cells, × = S-phase cells.
dTI?P pool
S phase G o phase
DNA synthesis
Size (pmole)
Specific Radioactivity activity incorporated (cpm/ (cpm/min) pmole)
dTMP incorporated into D N A (pmole/min)
109 8.5
9700 10 700
20.8 0.24
201700 2 570
cell populations in which essentially all cells were in G 0, G 1 or S phase. In a first experiment we compared G O and S-phase cells. G O cells were obtained by incubation of cultures for 4 days in medium containing 5% PPP, S-phase cells after an additional 18.5 h stimulation of parallel sets of cells with 5% human serum. Both kinds of cell were prelabeled with 0.3 /xM [3H]thymidine and then shifted to conditioned medium with cold thymidine. The specific activity of the dTTP pools was determined with samples removed at short time intervals, with the results shown in Fig. 3. In S-phase cells the chase caused an immediate, rapid and linear decay of the radioactivity when the results were plotted on a semilog scale. The half-life of the pool was 3.7 min corresponding to a pool turnover of 20.5 pmole of d T T P / m i n (pool size = 109 pmole). The incorporation of dTMP into DNA determined from isotope incorporation and the specific activity of dTTP before the chase period was 20.8 p m o l e / m i n (Table 1) demonstrating that in S-phase cells all newly synthesized dTTP was used for D N A replication. In G O cells, the loss of isotope from dTTP was biphasic, with a first short rapid phase followed by a second slow and longer decay period. During the second phase the half-life of the pool was 21 min corresponding to a turnover of 0.27 p m o l e / m i n (pool size = 8.5 pmole). The cells incorporated 0.24 pmole d T M P / m i n into their DNA (Table 1). This incorporation may be due to
41
mitochondrial D N A synthesis a n d / o r D N A replication in a small proportion of contaminating S-phase cells. The reason for the initial short and rapid decay of the specific activity of dTFP is not evident. In part it might be due to contaminating S-phase cells but it may also be caused artifactually by the handling of the cells during the medium change. These results suggest that resting 3T3 cells continuously turn over their dTTP pool and therefore also continuously synthesize this deoxynucleotide, at least in the presence of 0.3 /~M thymidine. In particular in G O cells the presence of the deoxynucleoside in the medium might affect the observed turnover. The following experiment was designed to investigate this question. Cells were starved in medium containing PPP that had been dialyzed extensively to remove all deoxynucleosides. They were then labeled for 90 min with only 0.1 #M thymidine and finally chased in 2 different
1 1 0 4.
°/o'"~*
°--~
I 30
'
/ °
G. t"
0
E 10 3
E
c}.
10 2
'
! 6 0
I 9 0
I 120
'
I 150
'
I 180
vain
Fig. 4. Effect of thymidine in the medium on the turnover of dTTP. This experiment was similar to the Go cell experiment of Fig. 3 except that dialyzed PPP was used to starve the cells, that labeling was done with 0.1 /.LMthymidine and that the case was made with either 0.1/xM cold thymidine (e) or with no thymidine(<3)in the medium.
7J o o
o o
0 U 0
I
0
0
0
I
'
I 90
0
0
E 1"
0
'
0
I
'
30
60 min
Fig. 5. Effect of thymidine in the medium on the size of the dTTP pool. The graph shows the size of the d T r P pools between 10 and 90 min of the chase in the experiment of Fig. 4. N o thymidine = o (mean 4.96, SD 0.53), 0.1/~M thymidine = • (mean 6.15, SD 0.63). Statistical evaluations were performed with Student's t test. The difference was significant at the 1% level.
ways: one chase was with conditioned medium containing 0.1 /~M cold thymidine, the second without deoxynucleoside. In both cases the size and specific activity of the dTTP pools were followed. Fig. 4 shows that in both instances the decay of the specific activities was biphasic, similar to the results for G O cells shown in Fig. 3. The presence of thymidine during the chase shortened the halflife of the pool decay during the second phase (27 min vs. 37 min). Both these values are larger than the 21 min found in the previous experiment with 0.3/~M thymidine. The presence of thymidine also affected the size of the dTTP pool (Fig. 5) which had a mean value of 5.0 pmole (no thymidine) or 6.1 pmole (0.1/tM thymidine) 10-90 min after the start of the chase. Calculations of turnover rates gave values of 0.09 p m o l e / m i n (not thymidine) and 0.16 p m o l e / m i n (0.1/~M thymidine). In a final experiment we determined the turnover of the d T T P pool in G 1 cultures either at 3.5 or 8 h after the switch from PPP to complete human serum. Labeling and chase were made with 0.3/~M thymidine under the same conditions as in the experiment depicted in Fig. 3. From the data shown in Fig. 6 one can calculate half-lives of 23 and 25 min and turnover rates of 0.15 and 0.18
42
1 104
\\
O. I-
a. 103 o,
E o. u
10 2
! 30
'
I 60
'
I 90
I 120
I 150
rain Fig. 6. Turnover of dTTP of G 1 cells. Cultures were set up as described in the legend to Fig. 1 and G 1 cells were labeled with 0.3/~M [3H]thymidine for 60 min starting either at 2.5 or at 7 h after addition of serum. At the arrows (3.5 h after addition of serum = A; 8 h after addition of serum = zx) the cells were shifted to conditioned medium with 0.3 #M cold thymidine. The specific activity of dTTP was determined at the indicated time intervals and is plotted on a semilog scale.
pmole for the 3.5 and 8 h experiments, respectively. Thus the turnover of dTTP in G1 cells did not differ significantly from that in G Ocells. Discussion
Both G O and G 1 cells, i.e., cells not involved in D N A replication, apparently contain small but measurable pools of dNTPs with a slow but finite turnover of at least the dTTP pool. This suggests that resting cells catalyze a small amount of de novo synthesis of dNTPs. Two complications inherent in our experimental approach could invalidate this conclusion: (1) slight contamination
by S-phase cells might be responsible for the observed result; and (2) the turnover might be caused by the addition of thymidine during the labeling and chase periods. The resting cell populations contained 0.5-1% cells that incorporated labeled thymidine into nuclear D N A and thus probably were in S phase. If these contaminants behaved as ordinary S-phase cells they would have contributed at most one pmole of dTTP with a pool half-life of 3.7 min. This amount of dTTP would represent only a minor part of the total dTI'P and the major part of the d T T P was therefore present in resting cells. As to the turnover, it seems possible that the first rapid decay during the chase was in part due to S-phase cells, but the second slow period should be ascribed to resting cells. As to the second complication, the effects of thymidine addition, the results shown in Figs. 4 and 5 demonstrate that the presence of the nucleoside during the chase did influence both the size and the turnover of the pool. As little as 0.1 /~M thymidine increased the pool from 5 to 6 pmole and the turnover from 0.09 to 0.16 pmole/min. We could not, of course, delete thymidine during the labeling period, and one might argue that the presence of thymidine at that stage affected the turnover during the later part of the chase. This seems unlikely, however, and we propose that our data show that 1 million resting 3T3 cells contain a pool of 5 pmole of dTTP with a turnover of 0.09 pmole/min. The latter value should correspond to the rate of de novo synthesis since the turnover was measured without deoxynucleosides in the medium. In comparison, S-phase cells contained a more than 20-fold larger d T T P pool with a 10 times faster turnover. This implies that the de novo rate of dTTP synthesis was more than 200 times faster in S-phase cells. A comparison of the turnover rate of d T T P (20.5 pmole) with the rate of d T M P incorporation into D N A (20.8 pmole) shows that all d T F P synthesized was used for DNA replication and confirms with S-phase cells our previous finding a single kinetic dTTP pool. In resting cells, the turnover of the dTTP pool may be accounted for by incorporation into mitochondrial DNA, D N A repair a n d / o r degradation and excretion of thymidine into the medium.
43
One final point concerns the specific activities of dTTP after labeling from thymidine in resting and S-phase cells. Radioactivity was rapidly introduced into the pools which in both cases during the steady state reached a specific activity of half that of the external thymidine. This result suggests that in both types of cells about half of the dTTP was synthesized de novo from unlabeled precursors and reinforces our earlier conclusion that de novo synthesis of dTTP does occur in resting cells. It is interesting to note that the same balance between de novo synthesis of dTTP and salvage from thymidine is found in G 0, G 1 and S-phase cells even though the activities of the many enzymes involved in the 2 processes undergo large changes after serum stimulation. A similar situation did not apply when synchronized cells were labeled from deoxycytidine (Leeds and Mathews, 1987).
Acknowledgements This work was supported by grants from the Swedish Medical Research Council and the Medical Faculty of the Karolinska Institute. G. Spyrou was the recipient of a fellowship from the Alexander F. Onassis Foundation.
References Cohen, A., and E. Thompson (1986) DNA repair in nondividing human lymphocytes: inhibition by deoxyadenosine, Cancer Res., 46, 1585-1588.
Hellgren, D., S. Nilsson and P. Reichard (1979) Effects of arabinosyl-cytosine on thymidine triphosphate pools and polyoma DNA replication, Biochem. Biophys. Res. Commun., 88, 16-22. Leeds, J.M., and C.K. Mathews (1987) Cell cycle-dependent effects on deoxyribonucleotide and DNA labeling by nucleoside precursors in mammalian cells, Mol. Cell. Biol., 7, 532-534. Lindberg, U., and L. Skoog (1970) A method for the determination of dATP and dTTP in picomole amounts, Anal. Biochem., 34, 152-160. Nicander, B., and P. Reichard (1983) Dynamics of pyrimidine deoxynucleoside triphosphate pools in relationship to DNA synthesis in 3T6 mouse fibroblasts, Proc. Natl. Acad. Sci. (U.S.A.), 80, 1347-1351. Nicander, B., and P. Reichard (1985) Relations between synthesis of deoxyribonucleotides and DNA replication in 3T6 fibroblasts, J. Biol. Chem., 260, 5376-5381. Pledger, W.J., C.D. Stiles, H.N. Antoniades and C.D. Scher (1977) Induction of DNA synthesis in BALB/c 3T3 cells by serum components: reevaluation of the commitment process, Proc. Natl. Acad. Sci. (U.S.A.), 74, 4481-4485. Reichard, P. (1988) Interactions between deoxyribonucleotide and DNA synthesis, Annu. Rev. Biochem., 57, 349-374. Seto, S., C.J. Carrera, D.B. Wasson and D.A. Carson (1986) Inhibition of DNA repair by deoxyadenosine in resting human lymphocytes, J. Immunol., 136, 2839-2843. Skoog, K.L., B.A. Nordenskj~51d and K.G. Bjursell (1973) Deoxyribonucleoside-triphosphate pools and DNA synthesis in synchronized hamster cells, Eur. J. Biochem., 33, 428-432. Thelander, L., and P. Reichard (1979) Reduction of ribonucleotides, Annu. Rev. Biochem., 48, 133-158. Waiters, R.A., R.A. Tobey and R.L. Ratliff (1973) Cell-cycledependent variations of deoxyribonucleoside triphosphate pools in Chinese hamster cells, Biochim. Biophys. Acta, 319, 336-347.
37
MTR 02313
Dynamics of the thymidine triphosphate pool during the cell cycle of synchronized 3T3 mouse fibroblasts Giannis Spyrou and Peter Reichard Department of Biochemistry, Medical Nobel Institute, Karolinska lnstitutet, Box 60400, S-104 01 Stockholm (Sweden)
(Received 28 January 1988) (Accepted 19 February 1988)
Keywords: dTI'P pool; Turnover; Cell synchronization; DNA repair
Summary To investigate whether resting cells of 3T3 mouse fibroblasts carry out de novo synthesis of deoxyribonucleoside triphosphates, we determined the turnover of the thymidine triphosphate pool of G O cells obtained by starvation of cultures for platelet-derived growth factor. These cells were contaminated by less than 1% S-phase cells. In the absence of deoxyribonucleosides in the medium one million G Ocells contained 5 pmole of dTTP with a turnover of 0.09 pmole/min. S-phase cells in comparison contained a 20 times larger dTTP pool with a more than 200-fold faster turnover. Our results suggest that G O cells carry out a slow but finite de novo synthesis of deoxyribonucleoside triphosphates to satisfy the cells' requirement for DNA repair and mitochondrial DNA synthesis.
The synthesis and turnover of deoxynucleoside triphosphates (dNTPs) is tightly coupled to DNA synthesis (for a recent review see Reichard, 1988). The activities of several enzymes involved in dNTP synthesis increase greatly when resting cells are induced by different means to replicate or repair their DNA. Cells in S phase contain much larger dNTP pools than resting cells. In cycling cell populations, S-phase cells contain the largest pools and the highest activities of enzymes participating in dNTP synthesis. In recent reports from this
Correspondence: Dr. P. Reichard, Department of Biochemistry, Medical Nobel Institute, Karolinska Institutet, Box 60400, S-104 01 Stockholm (Sweden). Abbreviations: dNTP, deoxyribonucleoside triphosphate; PPP, platelet-poor plasma.
laboratory the turnover of the dTTP pool was investigated in cycling cells (Nicander and Reichard, 1983, 1985). Exponentially growing cultures with close to 60% cells in S phase contained a single kinetic dTTP pool whose turnover could be accounted for by its consumption for DNA replication. This suggested that dTFP turnover and synthesis occurred mainly, or even exclusively, in S-phase cells and raised the question whether resting cells carry out any de novo synthesis of dTTP (and other dNTPs). The answer to this question is of interest for understanding how cells obtain a balanced supply of dNTPs for DNA repair and also for the interpretation of experiments demonstrating the interference with DNA repair in resting cells by agents that block or upset the normal supply of dNTP by ribonucleotide reductase (Cohen and Thompson, 1986; Seto et al., 1986).
0027-5107/88/$03.50 © 1988 Elsevier Science Publishers B.V. (Biomedical Division)
38 Here we describe experiments concerning the turnover of the dTTP pools of resting and S-phase cell populations of 3T3 mouse fibroblasts. Synchronization was obtained by starvation of the cells in medium lacking platelet-derived growth factor followed by stimulation with complete serum (Pledger et al., 1977). Before stimulation, more than 99% of the cells were in the resting G O state, as judged from the lack of thymidine incorporation into nuclear DNA. Between 15 and 18 h after stimulation more than 95% were in S phase. In both instances the dTTP pools were labeled rapidly from [3H]thymidine and, after change of medium, the decay of radioactivity was used to measure the turnover of the pools. Similar experiments were made with G 1 cells 3-8 h after serum stimulation. The results suggest that both G O and G 1 cells contain small dTTP pools whose half-lives are about 10 times longer than that of S-phase cells. Materials and methods
Chemicals [Methyl-3H]thymidine with a specific activity of 25 C i / m m o l e (20000 c p m / p m o l e ) was obtained from Amersham. Polyd(AT), polyd(IC) and DNA polymerase I from E. coli for pool assays were from Boehringer-Mannheim. All other chemicals were of highest purity available.
Platelet-poor plasma Platelet-poor plasma (PPP) was prepared from blood of human blood donors at the Karolinska Hospital, Stockholm (Sweden) by slight modification of a published method (Pledger et al., 1977). Approximately 450 ml of blood was collected from 1 donor, without using a tourniquet, into a plastic bag containing 63 ml citrate buffer (15 mM citric acid, 96 mM sodium citrate, 20 mM sodium phosphate monobasic, 129 mM glucose). The bag was centrifuged immediately at room temperature for 5 min at 5000 × g in a clinical centrifuge, the plasma was transferred to a second empty bag and recentrifuged for 5 min at 5000 x g. The supernatant solution was centrifuged for 30 min at 28 000 x g with a final yield of 250 ml of PPP. This was heat-inactivated at 5 6 ° C for 30 min, centrifuged at 28 000 x g for 30 min, sterilized by
filtration through 0.2-/~m amicon filters and stored at - 7 0 ° C before use. When indicated PPP was dialyzed for 48 h against several changes of 0.15 M NaC1.
Growth and synchronization of cells Swiss 3T3 mouse fibroblasts were obtained from Dr. Angela Otto, Friedrich Miescher Institut, Basel (Switzerland). The cells were grown in Dulbecco's modified Eagle's medium (DMEM) with 10% inactivated fetal calf serum and maintained for at most 6 passages (never allowed to reach confluency) before use. For experiments with resting cells, 0.15-0.2 million cells were seeded onto a 5-cm petri dish in 5 ml D M E M containing 5% PPP. After 4 days at 37 ° C the cell number was not increased and less than 1% of the cells were in S phase. For experiments with S-phase cells, resting cells were stimulated with 5% inactivated human serum. After 18 h more than 95% of the cells were in S phase as judged from autoradiography. Cells were labeled for autoradiography by a 45-min pulse with labeled 1 /~M thymidine. The percentage of labeled nuclei was determined by counting at least 4000 cells of duplicate autoradiograms.
Incorporation of labeled thymidine into cells Three hours before the addition of isotope, the medium was reduced from 5 to 2.5 ml. The dishes were divided into 2 groups: one received labeled thymidine, while the other received an identical amount of unlabeled thymidine and served as a source of conditioned medium. To measure the decay of the label (started at arrows in Figs. 3, 4 and 6) the labeled medium was sucked off and replaced with conditioned medium. All manipulations were made rapidly inside the incubators. Incubations were at 37 ° C in a humidified 7% CO2 atmosphere. Duplicate samples were used for each time point. Following incubation, the dishes were immediately transferred to an ice bath, the medium was sucked off, the plates were washed 3 times with 3-5 ml ice-cold Tris saline, and then left tilted for 1 min before removing the remaining buffer. Nucleotides were extracted from the cells with 60% methanol and the size and the specific activity of the dTTP pools were determined as described (Lindberg and Skoog, 1970; Hellgren et
39 10 3
al., 1979). The remaining cell pellet was used to determine incorporation of radioactivity into D N A (Nicander and Reichard, 1985). All results are normalized to 106 cells.
i !
I- O.7 I I-0. $ e
to," 10 z .
d N T P pools in relation to DNA synthesis in synchronized 3T3 cells After 4 days incubation of sparse 3T3 cells in growth medium containing PPP less than 1% of the cells were replicating their D N A , as judged from autoradiography with [3H]thymidine. Such cells were stimulated to synthesize D N A by addition of medium containing 5% human serum and at time intervals during the ensuing 21 h their entrance into S phase was determined by pulsed autoradiography with thymidine and the size of their 4 d N T P pools was determined (Fig. 1). D N A synthesis started quite synchronously after 9 h, and between 15 and 18 h more than 95% of the cells were in S phase. As demonstrated earlier, entrance into S phase was accompanied by dramatic pool changes (Waiters et al., 1973; Skoog et al., 1973). In resting cells the size of the d G T P pool was too small to be measured accurately ( < 2 pmole/106). This pool also remained the smallest throughout the whole time. The largest pool was
t0 z
-e-dTTP, dATP,
u
-o-dCTP
/
~dGTP
*
o ~ ~-o~.
t l~
•
I I
~'X
~ e
"%
100
LS0
"7"
,
,
10:
"o
. . . .
/
101
I 1
+,'
~
i"
I
+'
IO Hours
Fig. 1. Deoxyribonucleoside triphosphate pools after serum stimulation of resting 3T3 cells. Resting cells were obtained by incubation of parallel cultures in DMEM containing 5% PPP. At time 0 the cells were shifted to medium containing 5% human serum and incubated at 37 °C for the indicated times before pool analyses. The amount of S-phase cells was determined by autoradiography after labeling for 45 min with [3H]thymidine.
1
O.S .
¢:,
Results
nl
u
a.
I-°'4 T=
o E 101.
I-o.3
x
I
I- o.1 ~--dTTP ,
10 o
0
;o
- i - d C TP ,
~dATP
;o
,
~dGTP
:0
i 0.0 8O
Hours
Fig. 2. Deoxyribonucleoside triphosphate pools in synchronized and cycling 3T3 cells. This experiment was similar to the one described in Fig. 1 except for the use of dialyzed PPP and human serum. Pool determinations were made during a 3-day period. The amount of cells was determined by counting with a hematometer.
dCTP. It had already increased after 9 h, before onset of D N A synthesis, and, after that time, its size mirrored the percentage of cells in S phase. dATP and d T T P followed each other closely, had started to increase at the 12-h time point and continued their increase also when part of the cells had left S phase. These pool fluctuations during the cell cycle are similar to earlier results obtained with other synchronized cell populations (Waiters et al., 1973; Skoog et al., 1973). At first sight the results found with synchronized cells are difficult to reconcile with data from rapidly growing, cycling cell populations, containing close to 60% S-phase cells. The discrepancy can be exemplified with the dCTP pool. In cycling 3T3 cell populations this pool is smaller (40 p m o l e / 1 0 6 ) than expected from the percentage of cells in S phase. In the experiment depicted in Fig. 2 changes in d N T P pools were analyzed during 3 days after serum stimulation to clarify this point. In this experiment we used dialyzed PPP and dialyzed serum to avoid complications that might arise from the possible presence of deoxyribonucleosides in the medium. Again the dCTP pool showed a sharp increase during the first S phase but then returned to more ' n o r m a l ' values by 36 h when the first wave of D N A synthesis had passed. We believe that the initial overshoot of dNTPs was caused by the fact that the synchronization
40 procedure had led to a small increase in ribonucleotide reductase activity already before the onset of DNA synthesis (data not shown). Due to the allosteric regulation of ribonucleotide reductase (Thelander and Reichard, 1979) this led to an overproduction of dCTP and only with time did the cells attain a better balance of the 4 dNTPs.
TABLE 1 RATES OF DNA SYNTHESIS IN G 0- A N D S-PHASE 3T3 CELLS The total radioactivity incorporated into DNA between 30 and 60 min (S phase) or 30 and 90 min (G O phase) was determined in the experiment described in the legend to Fig. 3. The rate of DNA synthesis was calculated from the radioactivity accumulated between the 2 time points in DNA divided by the specific activity of the dTTP pool. All values are for 106 cells.
Turnover of the dTTP pool during the cell cycle To measure the turnover of the dTTP pool cells are first labeled from [3H]thymidine followed by a chase with cold nucleoside. The turnover of the pool can then be calculated from the decay of isotope. In the closely related cycling 3T6 cells the half-life of dTTP was 4-5 min and the turnover closely matched incorporation into DNA, indicating that essentially all dTTP synthesized was used for DNA replication. With synchronized 3T3 ceils it was now possible to measure dTTP turnover in
1
1o'-
I I
o. Il,"0
'3 E
\.
10 3. 0 O.
Go-phase
E
\ X~x
X-,~ x
S-phase 10 2
' 0
I 30
'
! 60
'
I 90
'
I 120
'
I 150
'
I t80
rain Fig. 3. Turnover of the dTTP pool in G o and S-phase cells. Resting and S-phase cells were obtained as described in Fig. 1. The cells were first labeled with 0.3 ~tM [ 3H]thymidine (90 min for G O cells, 60 min for S-phase cells). At the arrows the cells were shifted to conditioned medium containing 0.3 /~M cold thymidine. The specific activity of dTTP was determined on duplicate samples removed during the labeling and chase periods and the values are plotted on a semiiog scale. • = G O cells, × = S-phase cells.
dTI?P pool
S phase G o phase
DNA synthesis
Size (pmole)
Specific Radioactivity activity incorporated (cpm/ (cpm/min) pmole)
dTMP incorporated into D N A (pmole/min)
109 8.5
9700 10 700
20.8 0.24
201700 2 570
cell populations in which essentially all cells were in G 0, G 1 or S phase. In a first experiment we compared G O and S-phase cells. G O cells were obtained by incubation of cultures for 4 days in medium containing 5% PPP, S-phase cells after an additional 18.5 h stimulation of parallel sets of cells with 5% human serum. Both kinds of cell were prelabeled with 0.3 /xM [3H]thymidine and then shifted to conditioned medium with cold thymidine. The specific activity of the dTTP pools was determined with samples removed at short time intervals, with the results shown in Fig. 3. In S-phase cells the chase caused an immediate, rapid and linear decay of the radioactivity when the results were plotted on a semilog scale. The half-life of the pool was 3.7 min corresponding to a pool turnover of 20.5 pmole of d T T P / m i n (pool size = 109 pmole). The incorporation of dTMP into DNA determined from isotope incorporation and the specific activity of dTTP before the chase period was 20.8 p m o l e / m i n (Table 1) demonstrating that in S-phase cells all newly synthesized dTTP was used for D N A replication. In G O cells, the loss of isotope from dTTP was biphasic, with a first short rapid phase followed by a second slow and longer decay period. During the second phase the half-life of the pool was 21 min corresponding to a turnover of 0.27 p m o l e / m i n (pool size = 8.5 pmole). The cells incorporated 0.24 pmole d T M P / m i n into their DNA (Table 1). This incorporation may be due to
41
mitochondrial D N A synthesis a n d / o r D N A replication in a small proportion of contaminating S-phase cells. The reason for the initial short and rapid decay of the specific activity of dTFP is not evident. In part it might be due to contaminating S-phase cells but it may also be caused artifactually by the handling of the cells during the medium change. These results suggest that resting 3T3 cells continuously turn over their dTTP pool and therefore also continuously synthesize this deoxynucleotide, at least in the presence of 0.3 /~M thymidine. In particular in G O cells the presence of the deoxynucleoside in the medium might affect the observed turnover. The following experiment was designed to investigate this question. Cells were starved in medium containing PPP that had been dialyzed extensively to remove all deoxynucleosides. They were then labeled for 90 min with only 0.1 #M thymidine and finally chased in 2 different
1 1 0 4.
°/o'"~*
°--~
I 30
'
/ °
G. t"
0
E 10 3
E
c}.
10 2
'
! 6 0
I 9 0
I 120
'
I 150
'
I 180
vain
Fig. 4. Effect of thymidine in the medium on the turnover of dTTP. This experiment was similar to the Go cell experiment of Fig. 3 except that dialyzed PPP was used to starve the cells, that labeling was done with 0.1 /.LMthymidine and that the case was made with either 0.1/xM cold thymidine (e) or with no thymidine(<3)in the medium.
7J o o
o o
0 U 0
I
0
0
0
I
'
I 90
0
0
E 1"
0
'
0
I
'
30
60 min
Fig. 5. Effect of thymidine in the medium on the size of the dTTP pool. The graph shows the size of the d T r P pools between 10 and 90 min of the chase in the experiment of Fig. 4. N o thymidine = o (mean 4.96, SD 0.53), 0.1/~M thymidine = • (mean 6.15, SD 0.63). Statistical evaluations were performed with Student's t test. The difference was significant at the 1% level.
ways: one chase was with conditioned medium containing 0.1 /~M cold thymidine, the second without deoxynucleoside. In both cases the size and specific activity of the dTTP pools were followed. Fig. 4 shows that in both instances the decay of the specific activities was biphasic, similar to the results for G O cells shown in Fig. 3. The presence of thymidine during the chase shortened the halflife of the pool decay during the second phase (27 min vs. 37 min). Both these values are larger than the 21 min found in the previous experiment with 0.3/~M thymidine. The presence of thymidine also affected the size of the dTTP pool (Fig. 5) which had a mean value of 5.0 pmole (no thymidine) or 6.1 pmole (0.1/tM thymidine) 10-90 min after the start of the chase. Calculations of turnover rates gave values of 0.09 p m o l e / m i n (not thymidine) and 0.16 p m o l e / m i n (0.1/~M thymidine). In a final experiment we determined the turnover of the d T T P pool in G 1 cultures either at 3.5 or 8 h after the switch from PPP to complete human serum. Labeling and chase were made with 0.3/~M thymidine under the same conditions as in the experiment depicted in Fig. 3. From the data shown in Fig. 6 one can calculate half-lives of 23 and 25 min and turnover rates of 0.15 and 0.18
42
1 104
\\
O. I-
a. 103 o,
E o. u
10 2
! 30
'
I 60
'
I 90
I 120
I 150
rain Fig. 6. Turnover of dTTP of G 1 cells. Cultures were set up as described in the legend to Fig. 1 and G 1 cells were labeled with 0.3/~M [3H]thymidine for 60 min starting either at 2.5 or at 7 h after addition of serum. At the arrows (3.5 h after addition of serum = A; 8 h after addition of serum = zx) the cells were shifted to conditioned medium with 0.3 #M cold thymidine. The specific activity of dTTP was determined at the indicated time intervals and is plotted on a semilog scale.
pmole for the 3.5 and 8 h experiments, respectively. Thus the turnover of dTTP in G1 cells did not differ significantly from that in G Ocells. Discussion
Both G O and G 1 cells, i.e., cells not involved in D N A replication, apparently contain small but measurable pools of dNTPs with a slow but finite turnover of at least the dTTP pool. This suggests that resting cells catalyze a small amount of de novo synthesis of dNTPs. Two complications inherent in our experimental approach could invalidate this conclusion: (1) slight contamination
by S-phase cells might be responsible for the observed result; and (2) the turnover might be caused by the addition of thymidine during the labeling and chase periods. The resting cell populations contained 0.5-1% cells that incorporated labeled thymidine into nuclear D N A and thus probably were in S phase. If these contaminants behaved as ordinary S-phase cells they would have contributed at most one pmole of dTTP with a pool half-life of 3.7 min. This amount of dTTP would represent only a minor part of the total dTI'P and the major part of the d T T P was therefore present in resting cells. As to the turnover, it seems possible that the first rapid decay during the chase was in part due to S-phase cells, but the second slow period should be ascribed to resting cells. As to the second complication, the effects of thymidine addition, the results shown in Figs. 4 and 5 demonstrate that the presence of the nucleoside during the chase did influence both the size and the turnover of the pool. As little as 0.1 /~M thymidine increased the pool from 5 to 6 pmole and the turnover from 0.09 to 0.16 pmole/min. We could not, of course, delete thymidine during the labeling period, and one might argue that the presence of thymidine at that stage affected the turnover during the later part of the chase. This seems unlikely, however, and we propose that our data show that 1 million resting 3T3 cells contain a pool of 5 pmole of dTTP with a turnover of 0.09 pmole/min. The latter value should correspond to the rate of de novo synthesis since the turnover was measured without deoxynucleosides in the medium. In comparison, S-phase cells contained a more than 20-fold larger d T T P pool with a 10 times faster turnover. This implies that the de novo rate of dTTP synthesis was more than 200 times faster in S-phase cells. A comparison of the turnover rate of d T T P (20.5 pmole) with the rate of d T M P incorporation into D N A (20.8 pmole) shows that all d T F P synthesized was used for DNA replication and confirms with S-phase cells our previous finding a single kinetic dTTP pool. In resting cells, the turnover of the dTTP pool may be accounted for by incorporation into mitochondrial DNA, D N A repair a n d / o r degradation and excretion of thymidine into the medium.
43
One final point concerns the specific activities of dTTP after labeling from thymidine in resting and S-phase cells. Radioactivity was rapidly introduced into the pools which in both cases during the steady state reached a specific activity of half that of the external thymidine. This result suggests that in both types of cells about half of the dTTP was synthesized de novo from unlabeled precursors and reinforces our earlier conclusion that de novo synthesis of dTTP does occur in resting cells. It is interesting to note that the same balance between de novo synthesis of dTTP and salvage from thymidine is found in G 0, G 1 and S-phase cells even though the activities of the many enzymes involved in the 2 processes undergo large changes after serum stimulation. A similar situation did not apply when synchronized cells were labeled from deoxycytidine (Leeds and Mathews, 1987).
Acknowledgements This work was supported by grants from the Swedish Medical Research Council and the Medical Faculty of the Karolinska Institute. G. Spyrou was the recipient of a fellowship from the Alexander F. Onassis Foundation.
References Cohen, A., and E. Thompson (1986) DNA repair in nondividing human lymphocytes: inhibition by deoxyadenosine, Cancer Res., 46, 1585-1588.
Hellgren, D., S. Nilsson and P. Reichard (1979) Effects of arabinosyl-cytosine on thymidine triphosphate pools and polyoma DNA replication, Biochem. Biophys. Res. Commun., 88, 16-22. Leeds, J.M., and C.K. Mathews (1987) Cell cycle-dependent effects on deoxyribonucleotide and DNA labeling by nucleoside precursors in mammalian cells, Mol. Cell. Biol., 7, 532-534. Lindberg, U., and L. Skoog (1970) A method for the determination of dATP and dTTP in picomole amounts, Anal. Biochem., 34, 152-160. Nicander, B., and P. Reichard (1983) Dynamics of pyrimidine deoxynucleoside triphosphate pools in relationship to DNA synthesis in 3T6 mouse fibroblasts, Proc. Natl. Acad. Sci. (U.S.A.), 80, 1347-1351. Nicander, B., and P. Reichard (1985) Relations between synthesis of deoxyribonucleotides and DNA replication in 3T6 fibroblasts, J. Biol. Chem., 260, 5376-5381. Pledger, W.J., C.D. Stiles, H.N. Antoniades and C.D. Scher (1977) Induction of DNA synthesis in BALB/c 3T3 cells by serum components: reevaluation of the commitment process, Proc. Natl. Acad. Sci. (U.S.A.), 74, 4481-4485. Reichard, P. (1988) Interactions between deoxyribonucleotide and DNA synthesis, Annu. Rev. Biochem., 57, 349-374. Seto, S., C.J. Carrera, D.B. Wasson and D.A. Carson (1986) Inhibition of DNA repair by deoxyadenosine in resting human lymphocytes, J. Immunol., 136, 2839-2843. Skoog, K.L., B.A. Nordenskj~51d and K.G. Bjursell (1973) Deoxyribonucleoside-triphosphate pools and DNA synthesis in synchronized hamster cells, Eur. J. Biochem., 33, 428-432. Thelander, L., and P. Reichard (1979) Reduction of ribonucleotides, Annu. Rev. Biochem., 48, 133-158. Waiters, R.A., R.A. Tobey and R.L. Ratliff (1973) Cell-cycledependent variations of deoxyribonucleoside triphosphate pools in Chinese hamster cells, Biochim. Biophys. Acta, 319, 336-347.