<<

(CANCER RESEARCH 51. 2346-2352, May I. 1991) Mechanism of Cell Death following Thymidylate Synthase Inhibition: 2'-Deoxyuridine-5'-triphosphate Accumulation, DNA Damage, and Growth Inhibition following Exposure to CB3717 and Dipyridamole Nicola J. Curtin,1 Adrian L. Harris, and G. Wynne Aherne Cancer Research L'nil, Medical School, University of Newcastle upon Tyne, Newcastle upon Tyne [N. J. C.J; Imperial Cancer Research Fund Clinical Oncology Unit, Churchill Hospital, Headington, Oxon ¡A.L. H.]; Department of Biochemistry, University of Surrey, Guildford, Surrey [G. W. A.], England

ABSTRACT but one hypothesis, based on the study of bacterial mutants (3- The thymidylate synthasc inhibitor /V'°-propargyl-5,8-dideazafolic 5), is that TS inhibition leads not only to a reduction in dTTP levels but also, as dUMP accumulates behind the block, to the acid (CB3717) inhibits the growth of human lung carcinoma A549 cells. formation of dUTP. The levels of dUTP eventually overwhelm The cytotoxicity of CB3717 is potentiated by the transport inhibitor dipyridamole (DP), which not only inhibits the uptake and dUTPase (the enzyme which breaks down dUTP to dUMP) therefore salvage of but also inhibits the efflux of deoxyuridine, and the levels of dUTP increase. DNA polymerase can utilize thereby enhancing the intracellular accumulation of deoxyuridine nucleo- dUTP and dTTP with equal efficiency (6), such that is tides. Measurement of intracellular deoxyuridine triphosphate (dUTP) misincorporated into DNA. Uracil in DNA is excised rapidly pools, by sensitive radioimmunoassay, demonstrated a large increase in by uracil glycosylase, leaving an apyrimidinic site. During repair response to CB3717, in a dose- and time-related manner, and this of apyrimidinic sites, in the presence of unbalanced dUTP/ accumulation was enhanced by coincubation with DP. In untreated cells dTTP ratios, uracil is likely to be reinserted, causing a futile and those treated with DP alone, dUTP was close to or below the limit cycle of excision, repair, and reinsertion, leading to DNA strand of detection of the assay. In cells treated for 24 h with 3 MMCB3717 breakage and ultimately cell death. Whether or not thymineless (concentration producing 50% growth inhibition) the intracellular dUTP was 46.1 ±9.6 (SEM) pumi III" cells and after 24 h exposure to 30 n\i death occurs by this mechanism in mammalian cells has not CB3717, 337.5 ±37.9 pmol dUTP/106 cells was detected. There was been fully determined; some investigators have presented evi significant enhancement by DP of the accumulation of ill 11*in cells dence supporting this theory (7-9), whereas others have pre treated with CB3717; coincubation of cells with 1 MMDP + 3 MMCB3717 sented evidence to the contrary (10, 11). for 24 h resulted in intracellular dLTP levels of 174.7 ±57.7 pmol/10" In order to investigate the role of dUTP in thymineless death cells. Accumulation of DNA strand breaks, measured by alkaline elution, in mammalian cells we have studied the combined effect of also increased in response to CB3717 concentration and exposure period. CB3717 and the nucleoside transport inhibitor, DP on dUTP Newly synthesized (nascent) DNA was more sensitive to damage by pools in relation to DNA damage and cytotoxicity. Salvage of CB3717 than was mature DNA. As with the accumulation of dUTP, dThd is a mechanism by which cells may bypass TS inhibition. coincubation with DP also enhanced the accumulation of strand breaks, By inhibiting dThd uptake, DP potentiated CB3717 toxicity; whereas DP alone had little or no effect on DNA fragmentation. however, this potentiation was greater than that achieved by When data for cells treated with CB3717 alone and CB3717 in limiting dThd availability with dialyzed serum (12). This dis combination with DP were combined, there was a significant correlation crepancy could be explained by the observation that DP also of intracellular dUTP levels with the level of DNA strand breaks. This inhibited deoxyuridine efflux and we postulated that DP might strongly suggests that growth inhibition following thymidylate synthase inhibition is mediated through an increase in intracellular dill', leading therefore help to maintain high deoxyuridine pools to uracil misincorporation into DNA, its subsequent excision, and result and that this was responsible for its greater potentiating effect ant strand breakage. than dialyzed serum (12). Indeed, by inhibiting deoxyuridine efflux, DP has been shown to increase intracellular dUMP levels in fluorouracil-treated cells (13). INTRODUCTION To confirm our hypothesis that DP was potentiating CB3717 TS2 is a key enzyme of DNA synthesis and therefore a prime cytotoxicity in part by promoting dUTP accumulation, it was target for cancer chemotherapy. Methotrexate and fluorodeox- necessary to measure dUTP levels in cells treated with CB3717 yuridine can both give rise to TS inhibition but attempts to alone and in combination with DP. The standard methods for determine their mechanism of cytotoxicity are complicated by quantification of deoxynucleotide triphosphates are high per the fact that they also have other sites of action; methotrexate formance liquid chromatography and the indirect DNA polym also inhibits biosynthesis and fluorodeoxyuridine may erase assay. Both methods are unsuitable for dUTP determi be incorporated into DNA. In contrast the antifolate drug nation: high performance liquid chromatography because the method lacks sensitivity and hence large numbers of cells are CB3717 specifically inhibits TS (1, 2) and this makes it an required, while the DNA polymerase assay is very sensitive but excellent tool for the study of thymineless death in mammalian does not distinguish between dUTP and dTTP (6). In view of cells. these problems, we have developed a highly sensitive radio The mechanism of thymineless death has not been defined immunoassay method capable of detecting 3.78 fmol dUTP Received 6/11/90: accepted 2/13/91. (14). This has allowed the detection of small changes in dUTP The costs of publication of this article were defrayed in part by the payment levels using low numbers of cells (IO6) treated with biologically of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. relevant doses of CB3717 and DP. ' This work was supported by the North of England Cancer Research Cam We report here a study of the effects of CB3717 and DP on paign (N. J. C. and A. L. H.) and the Cancer Research Campaign of Great Britain dUTP pools in parallel with determinations of DNA damage (G. W. A.). A preliminary report of this work was presented at the Annual General Meeting of the British Association for Cancer Research, March 1990 (Br. J. by alkaline elution and cytotoxicity. The results show that Cancer, 62: 508. 1990). 2The abbreviations used are: TS. thymidylate synthase: DP, dipyridamole; intracellular dUTP accumulation is directly proportional to dThd. thymidine; CB3717. A'lo-propargyl-5,8-dideazafolic acid; IC50, concentra DNA damage and related to growth inhibition of human lung tion producing 50% growth inhibition; dNTP, deoxynucleoside triphosphate. carcinoma A549 cells treated with CB3717 and DP. 2346

Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 1991 American Association for Cancer Research. CELL DEATH AFTER TS INHIBITION

MATERIALS AND METHODS cells were exposed to fresh control medium or that containing CB3717 and/or DP. After the desired incubation period, the medium was Cell Culture. Human lung carcinoma A549 cells (15, 16) were grown removed by aspiration, and the dish was placed on ice and flooded with in 6-well cluster dishes (Falcon; Becton Dickinson, Oxford, England) 1 ml ice-cold 0.4 M perchloric acid. The cells were removed from each in RPMI 1640 (Northumbria Biologicals, Cramlington, England) sup well with a cell scraper and the suspensions were transferred to chilled plemented with 10% fetal bovine serum (Biological Industries, Glasgow, centrifuge tubes, vigorously vortex mixed, and placed on ice for 30 min. Scotland), 500 ID/ml penicillin, and 500 Mg/ml streptomycin (Gibco, They were centrifuged for 20 min at 1500 x g at 4°Candthe supernatant Paisley, Scotland) at 37°Cin an atmosphere of 5% CO? in air. For was carefully decanted. The volume was measured and neutralized with growth inhibition assays, cells were seeded at 2 x IO4cells/well in 2 ml 0.5 volumes ice-cold 0.72 M KOH in 0.16 M KHCO3. Alter vortex medium. The following day, the medium was replaced with fresh control mixing, the extracts were kept on ice for 1 h and centrifuged as before. or experimental medium containing CB3717 and/or DP. After 72 h The supernatant fluid was transferred to duplicate vials and stored in cells were trypsinized and counted on a Coulter Counter (Coulter liquid nitrogen until assayed for dUTP levels. Replicate wells were used Electronics, Luton, England). for cell counts. Alkaline Elution. The method for the detection of DNA strand breaks Intracellular dUTP pools were measured using a previously described by alkaline elution has been reviewed in detail (15). The technique radioimmunoassay procedure with a sensitivity of 3.78 fmol (14). utilizes filters which mechanically impede the passage of long DNA Briefly, cell extracts were treated with 0.5 Msodium periodate to remove strands. Short DNA strands will be eluted rapidly from the filters and interfering and an aliquot (50-500 //I. depending on larger DNA strands will be retained on the filter for longer periods. the degree of exposure of the cells to CB3717) was subjected to Experimental cells exposed to varying drug concentrations are coeluted chromatography on QAE A25 Sephadex (Pharmacia, Milton Keynes, with internal standard cells, which have been irradiated immediately England). Following elution of dUMP and dUDP with 0.3 M KH2PO4, prior to elution, in alkaline buffer. Measurements of the fraction of dUTP was eluted with 0.7 M KH2PO4 and the eluate collected in 0.8- DNA from the experimental cells remaining on the filter, compared ml fractions. The radioimmunoassay was carried out on aliquots of the with the fraction of DNA from the internal standards remaining on the fractions (typically 100-400 /tl, depending on the dose and exposure filter, with time give the standardized elution rate of the experimentally time of CB3717) using an affinity-purified rabbit antiserum to dUTP- treated cells. Experimental cells were grown in 6-well plates and internal ovalbumin and [ 'H]dUTP (Amersham International). The recovery of standards in 10-cm Petri dishes. Cells were seeded such that at har dUTP added to extracts of control cells was 60.34 ±5.48% (n = 33) vesting there would be approximately 0.5-1 x 10' cells/well (6-well for a concentration range of 4-400 pmol dUTP/ml. plates) and 3x10' cells/10-cm Petri dish. The alkaline elution filter assay was used to measure DNA strand breaks in both nascent and mature DNA. RESULTS For studies with mature DNA, experimental cells were incubated with 0.4 nCi [14C]dThd/well (51.4 mCi/mmol, Amersham Interna As we have previously reported, DP potentiates the growth tional, Aylesbury, England) and internal standards were incubated with inhibitory effect of CB3717. At 10 ^M, but not at 1 MM,DP 1 /iCi [m«A>'/-3H]dThd/dishfor 24 h. Following a 4- to 24-h postlabel- alone is growth inhibitory. Fig. 1 shows the inhibition of cell ing chase period with fresh medium the medium was replaced with growth by CB3717 alone and with two concentrations of DP. medium containing CB3717 and/or DP as indicated for 4-24 h. Cells When normalized to percentage of appropriate control (no were then harvested for alkaline elution and the internal standards were CB3717) the IC50 values for CB3717 were as follows: CB3717 given 300 cGy X-irradiation prior to coelution with experimental cells. For studies with nascent DNA, [3H]dThd or |'4C]dThd were present alone, 2.5 /

Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 1991 American Association for Cancer Research. CELL DEATH AFTER TS INHIBITION affect dUTP levels significantly but, in the presence of 3 UM caused a 2- to 4-fold expansion of the pool compared to 3 or CB3717, DP further expanded the pool significantly (P < 0.05), 30 MMCB3717 alone and 10 MMDP caused an approximately 1 MMDP caused a 3-fold expansion, and 10 MMDP caused a 10-fold increase over that of 3 MMCB3717 alone. 10-fold expansion of the pool. The effect of DP on dUTP levels Alkaline elution studies to determine the time course of DNA in 30 /¿MCB3717-treated cells was not significant; however, damage by 30 MMCB3717 with and without 1 MMDP are shown this may simply reflect the fact that maximal dUTP accumu in Figs. 4 and 5. In nascent DNA (Fig. 4), strand breaks were lation was already achieved by CB3717 alone. just detectable after only 4 h with CB3717 alone (Fig. 4A) and In alkaline elution studies, following a 24-h exposure to they continued to accumulate with increasing time. At all time CB3717, DNA strand breakage was, as expected, greater with points, 1 MMDP increased the DNA damage (Fig. 4B) and this 30 MMCB3717 than with 3 MMCB3717 and the effect was more was particularly obvious at early time points with a large marked in nascent DNA than in mature DNA. Nascent DNA amount of breakage being apparent after only 4 h with 30 MM was more labile generally, the slopes for control cells being CB3717 + 1 MMDP. In mature DNA (Fig. 5) elution profiles steeper than with mature DNA (Figs. 2 and 3). DP alone at 4 and 8 h with CB3717 alone (Fig. 5/4) were virtually induced some DNA strand breakage in nascent DNA (Fig. 2, indistinguishable from controls. Strand breaks did, however, B and C) but the elution profiles for mature DNA treated with accumulate with time after the first 8 h. In the presence of 1 DP alone were virtually indistinguishable from controls (Fig. MMDP (Fig. 5B), strand breakage was detectable at 4 h and the 3, B and C). However, DP, both at 1 MMand to a greater extent damage continued to accumulate as the incubation period was at 10 MM,augmented the DNA strand breakage caused by 3 extended. and 30 MMCB3717. This effect was more marked in nascent To investigate possible relationships between DNA damage DNA and with 3 MMCB3717. and dUTP accumulation the data for all experiments were Time-course studies showed that there was a significant ele combined. DNA damage was quantitated as the relative elution vation in dUTP after only 4 h with 3 MMCB3717 and a greater rate (16): elevation with 30 MMCB3717 (Table 2). Levels were signifi log retention CB3717 (±DP)-treated sample cantly higher after 16 h exposure. Coincubation with 1 MMDP - log retention control (±DP) Table I Inlracellular dl'TP following 24 h exposure W CB3717 ±DP (taken when the internal standard retention was 0.4). cells)1.5 CB37I7(„M)00Ãœ333IO303030DP<„M)0110011000110dUTP(pitiol/10' Analysis showed a significant correlation of dUTP levels and ±0.3(26)°2.9 relative elution rate for both mature (r = 0.9481, P < 0.001; ±0.7(9)1.4 Fig 6A) and nascent (r = 0.8152, P < 0.005; Fig. 65) DNA. ±0.5(7)46.1 (21)174.7±±9.6* 57.7*'c(7)482.0 DISCUSSION (7)134.0±±80.3'"

(8)337.5 11.9* The aim of this study was to investigate the role of dUTP in the mechanism of thymineless death in mammalian cells using (18)370.1±37.9* CB3717 alone and in combination with DP. The rapid and (8)660.3±68.0* ±351.2*(3) substantial increase in dUTP pools following a dose- and time- a Mean ±SEM; number in parentheses, number of observations. related response to CB3717 confirms our earlier observations * Significantly different from control. (14). Increases in intracellular dUTP have been reported with ' Significantly different from CB3717 alone. P < 0.05 (Student's / test). CB3717 (2) and other TS-acting drugs (9, 17, 18). However,

1.0«

QL y 0.1 0.1 <

0.01 0.01 0.01 4. 01 0.01 1.0 0.1 0.01 1.0 0.1 0.01

Fraction 3H Retained Fraction 3H Retained Fraction 3H Retained Fig. 2. Alkaline elution of nascent DNA. A, no DP: A, +1 ¿IMDP;C, + 10 JIMDP. Incubation with drug(s) were for 24 h prior to elution: •,noCB37I7;O, 3 IÕM CB3717; A. 30nMCB3717 2348

Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 1991 American Association for Cancer Research. CELL DEATH AFTER TS INHIBITION

1.0. 1.0 \ Vs. a 0.1 0.1 . \

001 0.01 0.01 0.01 1.0 0.1 0.01 0.1 1.0 0.1 0.01 Fraction 3H Retained Fraction 3H Retained Fraction 3H Retained

Fig. 3. Alkaline elution of mature DNA. A, no DP; B, +1 MMDP; C +10 MMDP. Incubations with drug(s) were for 24 h prior to elution: •,no CB3717; O, 3 nM CB3717;A, 3ÛMMCB37I7.

Table 2 Intracellular dUTP following 4, 16, and 24 h incubation with that they overcame the inhibition of efflux by DP, preventing CB3717 + DP dUTP(pmol/10' cells) the accumulation of deoxyuridine beyond a certain level. h°46.1 (MM)3 (MM)h0 4 b±4.7(10) The mechanism of DP enhancement of dUTP accumulation ±9.6(21)I74.7±57.7f(7)is probably 2-fold: (a) by inhibiting deoxyuridine efflux (12), ±8.4 ±68.6'(4) 3 1 ±3.9 (5) 128.5 higher levels of intracellular deoxyuridine nucleotides are main 3 10 30.0 ±14.5C(4) 386.3 ±76.6f(3) 482.0 ±80.3r(7) 30 0 9.6±27.2 3.7 (7) 125.9 ±25.6(7) 337.5±37.9(18) tained, since it has been calculated in methotrexate-treated cells 30DP 13.3 ±1.3(10)*3.7C(5)32.2312.216±81.9'(5)24370.1 ±68.0(8) that for every molecule of dUMP that accumulates 15 molecules * Data from Table 1 for comparison. * Mean ±SEM; number in parentheses, number of observations. of deoxyuridine are released from the cell (19); (b) by inhibiting "Significantly different from CB3717 alone; all results were significantly the uptake of thymidine, dTTP pools will be further depressed, different from 24-h controls (P < 0.05, Student's t test). resulting in a greater derepression of dCMP deaminase, the major source of dUMP (19) and hence dUTP. In parallel to the increase in dUTP, there was a dose- and exposure of A549 cells to CB3717 results in higher dUTP pools time-related increase in DNA strand breakage in CB3717- than have been reported previously. For example, following a treated cells. In studies of mature DNA there was only a very 24-h incubation with 30 /ÕMCB3717 alone, dUTP levels were slight increase in elution rate following 8 h exposure to 30 /UM >300 pmol/106 cells, more than twice that found following CB3717, but after 12 h strand breakage was easily detected. incubation with 10 UM methotrexate + 10 RIM deoxyuridine Similarly, DNA strand breakage has been reported in NIH3T3 (9). cells treated with 200 MM CB3717 after 16 h but not 8 h When the time course of dUTP accumulation was studied, exposure (20). There was a more profound effect on the integrity significant increases in dUTP were seen after only 4 h with the of nascent DNA. This would be expected, because the fragmen IC50concentration of CB3717; thus an increase in dUTP is an tation of mature DNA is the result of the inability to repair early event. Furthermore it was just possible to detect an spontaneously occurring damage (the cytotoxic effect) and the increase in DNA fragmentation in nascent but not mature DNA degradation of newly synthesized DNA (the cytostatic effect) at 4 h with 30 /*MCB3717; thus the effect on DNA synthesis reflects errors due to deoxynucleotide imbalance introduced is also an early event. The intracellular dUTP content was not during semiconservative replication. DNA polymerase a has an significantly altered by DP alone. However, DP significantly estimated error rate of 1/30,000 (21) but this will be much increased dUTP accumulation in CB3717-treated cells. The greater in the presence of increased dUTP/dTTP ratios, since enhancing effect of DP on dUTP accumulation was more DNA polymerase utilizes both dUTP and dTTP with equal apparent at the low concentration of CB3717 and following efficiency (6). In an in vitro system using isolated cell nuclei, shorter exposure periods, i.e., when dUTP levels were <200 replacement of as little as 2.5% of dTTP by dUTP resulted in pmol/106 cells, 1 ¿ÕMDPresulted in a 3-fold increase in dUTP the formation of short Okazaki fragments and an increase in and 10 UM a 10-fold increase. It is possible that the reduced the proportion of small fragments (22). This increase in short effect of DP on the dUTP level in cells treated with 30 ¿IM fragments could be prevented by the inhibition of uracil glyco- CB3717 for 24 h was because near-maximum achievable levels sylase by 6 ITIMuracil and was therefore due to excision of of dUTP had been attained. The nucleotide pool may have misincorporated uracil. Furthermore, studies using an antibody become so imbalanced and DNA damage so great that the raised against partially purified human uracil glycosylase indi synthesis of dUTP was significantly slowed or halted after a cate a physical attachment of this enzyme to DNA polymerase certain level had been attained. Alternatively, intracellular deox a (23); therefore the excision will be immediately after insertion. yuridine concentrations may have increased to such an extent In studies with fluorouracil, DP, and deoxyuridine, Grem et 2349

Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 1991 American Association for Cancer Research. CELL DEATH AFTER TS INHIBITION

Control

Fig. 4. Alkaline elution of nascent DNA: time course of 30 MMCB3717 exposure. /4, no DP; B, +1 MMDP. Before elution, cells were treated with no CB.V717(•)or30 MMCB3717 for: O, 4 h; A, 8 h; A, 12 h; • 16 h; or D, 24 h.

0.01 0.01 1.0 0.1 0.01 1.0 0.1 0.01 Fraction 3H Retained Fraction 3H Retained al. (24) observed biphasic elution profiles of nascent DNA from hundred residues are lost each generation in human treated cells but nearly linear control elution profiles. However, cells (26). The error rate of DNA polymerase ßis1/3000-1/ in our study, nascent control DNA had a biphasic elution profile 7000 (21) but will again be increased in the presence of increased reflecting the proportion of different sized newly synthesized dUTP/dTTP ratios. The progressive accumulation of DNA DNA replication intermediates from Okazaki fragments to strand breakage in mature DNA, therefore, is the product of larger fragments. With increasing CB3717 concentration and two factors: (a) the accumulation of spontaneous damage with exposure time, the proportion of the rapidly eluting shorter time and (b) the increased perturbation of the dUTP/dTTP fragments increased and the profiles became more linear, sug ratio, resulting in an increasing inability to repair this sponta gesting a greater heterogeneity of fragment size. This is similar neous damage. to the observation that a 5-min pulse of ['Hjthymidine in In nascent, but not mature, DNA DP alone induced a small control cells resulted in two labeled DNA populations of 10 amount of DNA fragmentation; possibly DP reduces the overall kilobases and Okazaki fragments, whereas in fluorouracil- dNTP pool (without affecting the balance) and the competition treated cells there was a heterogeneous population of variable for dNTPs for initiation of new fragments with those for chain fragment sizes, mostly of smaller sized fragments (25). elongation results in shorter fragment length. We have not, The integrity of mature DNA is dependent on repair of however, tested this hypothesis experimentally. Strand break spontaneous damage. Spontaneous depurination is estimated age caused by CB3717 was increased by DP in both mature and to occur at a rate of 10,000 residues/generation and several nascent DNA. As with dUTP pools, DP had the most marked

24hr Fig. 5. Alkaline elution of mature DNA: 0.1 time course of 30 MMCB3717 exposure. .4, no DP; B, +1 MMDP. Before elution. cells were treated with no CB3717 (•)or with 30 MM CB3717 for: O, 4 h; A, 8 h; A, 12 h; •16 h; or D, 24 h.

0.01 0.01 1.0 0.1 0.01 0.1 0.01

Fraction 3H Retained Fraction 3H Retained

2350

Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 1991 American Association for Cancer Research. CELL DEATH AFTER TS INHIBITION

0.400 It remains to be determined whether perturbations in dNTP

0.350 • pools other than those described for dUTP also contribute to the DNA damage and growth inhibition caused by CB3717. 0.300- An additional factor in the potentiation of CB3717 cytotox- o: c 0250- icity by DP may be the effect of DP on drug accumulation. The accumulation of etoposide, Adriamycin, vinblastine, vindesine, 0.200 • and vincristine (36-38) is increased by DP by inhibition of drug 0.150- efflux, and accumulation of Adriamycin (39) and cisplatinum (40) is increased by DP by stimulation of drug uptake. More 0.100- directly relevant is the observation that DP has been shown to 0.050- increase intracellular concentrations of methotrexate, its poly- glutamation, and its retention following incubation in drug-free 0.000 100 200 300 400 500 medium (41, 42). Since methotrexate and CB3717 are structur ally related, it is possible that DP will increase the retention dUTP pmol/1CT6 cells and polyglutamation of CB3717. However, radiolabeled CB3717 will be required to test this possibility. 0.800 i In conclusion, our findings show that there is a positive 0.700 - correlation between elevated dUTP level and DNA strand breakage, both with CB3717 alone and in combination with 0600- DP. Since DP also potentiates the growth inhibition of CB3717 0.500- these are potent arguments in favor of the hypothesis that

0.4OO- thymineless death in mammalian cells is mediated largely through uracil misincorporation and its rapid excision. 0.300-

0.200- ACKNOWLEDGMENTS 0.100- We are grateful for discussions with Dr. D. R. Newell regarding this 0.000 manuscript and K. Mortimer and J. Wake for typing this manuscript.

dUTP pmol/10-6 cells B REFERENCES Fig. 6. Correlation of relative elution with intracellular dUTP. Relative elution was calculated as described in "Results." Data were taken from Tables I and 2 1. Jones, T. R., Calvert, A. H., Jackman, A. L., Brown, S. L., Jones, M., and and Figs. 2-5. A, mature DNA; B, nascent DNA; •,CB3717 alone; A, CB3717 Harrap, K. R. A potent antitumour quinazoline inhibitor of thymidylate + 1 nM DP; •.CB3717 + 10 »MDP. Regression lines are shown for mature synthctase synthesis, biological properties and therapeutic results in mice. DNA [r = 0.9481 (P< 0.001); intercept, 0.02; slope, 6.61] and for nascent DNA Eur.J. Cancer, / 7: 11-19, 1981. [r = 0.8152 (P < 0.005); intercept, 0.227: slope, 9.595]. 2. Jackson, R. C., Jackman, A. L., and Calvert, A. H. Biochemical effects of quinazoline inhibitor of thymidylate synthetase CB3717 on human lympho- blastoid cells. Biochem. Pharmacol., 33: 3783-3790, 1983. 3. Tamanoi. F., and Okazaki, T. Uracil incorporation into nascent DNA of requiring mutant Bacillus sublilis 168. Proc. Nati. Acad. Sci. USA, effect on lower concentrations of CB3717 and shorter exposure 75:2195-2199. 1978. times, i.e., when damage was slight to start with. Again, this 4. Tye. B-K., Chien, J., Lehman, I. R., Duncan, B. K., and Warner, H. R. Uracil may be because maximum achievable DNA fragmentation was incorporation: a source of pulse-labelled DNA fragments in the replication attained with the higher CB3717 concentrations at 24 h. The of Escherichia coli chromosome. Proc. Nati. Acad. Sci. USA. 75: 233-237, 1978. cells may reach a point of maximum damage beyond which 5. Makino, F., and Munakata. N. Deoxyuridine residues in DNA of Bacillus cellular processes, including DNA synthesis and repair, are subtilis strains with defective JV-glycosidase activity for uracil containing DNA. J. Bacteriol., 134: 24-29. 1978. halted completely and the cell is destined to die. It is possible 6. Richardson. C. C., Schildkraut, C. L., and Kornberg, A. Studies on the that the greatest fragmentation achieved in mature DNA rep replication of DNA by DNA polymerases. Cold Spring Harbor Symp. Quant. resents lack of effective repair of all spontaneously occurring Biol., 2Ä.-9-19, 1963. 7. Goulian, M., Bleile, B., and Tsang, B. Y. Methotrexate-induced misincor base loss. poration of uracil in DNA. Proc. Nati. Acad. Sci. USA, 77: 1956-1960, The progressive accumulation of DNA strand breaks has 1980. been observed with other agents which induce dNTP imbalance, 8. Sedwick, W. D., Kutler, M., and Brown, O. E. Antifolate-induced misincor poration of deoxyuridine monophosphate into DNA: inhibition of high e.g., hydroxyurea (decreased dATP and dGTP but increased molecular weight DNA synthesis in human lymphoblastoid cells. Proc. Nati. dCTP and dTTP) (27), fluorodeoxyuridine (decreased dTTP Acad. Sci. USA, 78: 917-921, 1981. and dGTP but increased dATP) (28), a combination of deoxy- 9. Ingraham, H. A., Dickey, L., and Goulian, M. DNA fragmentation and cytotoxicity from increased cellular deoxyuridylate. Biochemistry, 25: 3225- and deoxycofomycin (decreased dGTP but increased 3230. 1986. dATP) (29) and methotrexate (decreased dTTP, increased 10. Fraser. D. C., and Pearson, C. K. Is uracil misincorporation into DNA of mammalian cells a consequence of methotrexate treatment. Biochem. Bio- dUTP and dCTP) (9). Because of activation or repression of phys. Res. Commun., 135: 886-889. 1982. synthetic pathways, alterations in one dNTP often lead to 11. Ayusawa, D., Shimizu, K., Koyama, H., Takeishi, K., and Seno, T. Accu mulation of DNA strand breaks during thymineless death in thymidylate changes in other dNTPs. Furthermore, nucleotide pool imbal synthase negative mutants of mouse FM3A cells. J. Biol. Chem., 25S 12448- ance, particularly altered dCTP/dTTP ratios, can induce nu 12454, 1988. merous genetic changes, including altered sensitivity to muta- 12. Curtin, N. J., and Harris, A. L. Potentiation of quinazoline antifolate gens, point mutations, recombination events, activation of on- (CB3717) toxicity by dipyridamole in human lung carcinoma. A549, cells. Biochem. Pharmacol.. 37: 2113-2120. 1988. cogenes, chromosome and chromatid breaks, and chromosome 13. Grem, J. L.. and Fischer. P. H. Alteration of fluorouracil metabolism in shattering (see Refs. 30-35 for reviews and references therein). human colon cancer by dipyridamole with a selective increase in fluorodeox yuridine monophosphate levels. Cancer Res.. 46: 6191-6199. 1986. Thus, maintaining the correct nucleotide pool balance is of 14. Piali, E. M., Curtin, N. J.. Aherne, G. W., Harris, A. L., and Marks. V. The critical importance in all aspects of DNA replication and repair. quantitation by radioimmunoassay of 2'-deoxyuridine-5'-triphosphate in 2351

Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 1991 American Association for Cancer Research. CELL DEATH AFTER TS INHIBITION extracts of thyniidylulc synthase-inhibited cells. Anal. Biochem., 177: 347- imbalance: 5-fluorodeoxyuridine induced DNA double strand breaks in 352. 1989. mouse FM3A cells and the mechanism of cell death. J. Biol. Chem., 262: 15. Kohn, K. W.. Ewig, R. A.. Erickson. L. C.. and /.welling. L. A. Measurement 8235-8241. 1987. of strand breaks and cross-links by alkaline elution. In: E. C. Friedbcrg and 29. Yoshioka. A., Tanaka. S.. Hiraoka, O., Koyama, Y'., Hirota, Y., and Wataya, P. C. 11.ni.i».ili(eds.). DNA Repair: A Laboratory Manual of Research Y. triphosphate imbalance death: -in- Procedures. Vol. 1. Part B. pp. 379-401. New York: Marcel Dekker Inc.. duced dNTP imbalance and DNA double strand breaks in mouse FM3A 1981. cells and the mechanism of cell death. Biochem. Biophys. Res. Commun., 16. Fornace. A. J.. and Little. J. B. DNA crosslinking induced by X-rays and 146: 258-264. 1987. chemical agents. Biochim. Biophys. Acta. 477: 343-355. 1977. 30. Kun/, B. A. Genetic effects of deoxyribonucleotide pool imbalances. Environ. 17. Goulian. M., Bleile. B.. and Tseng. B. Y. The effect of methotrexate on levels Mutagenesis, 4: 695-725, 1982. of dUTP in animal cells. J. Biol. Chem.. 255: 10630-10637. 1980. 31. Hayncs. R. H., and Kunz, B. A. The influence of thymidine nucleotide 18. Ingraham. H. A.. Tseng. B. Y.. and Goulian. M. Nucleotide levels and depletion on genetic stability and change in eukaryotic cells. Curr. Sci., 55: incorporation of 5-fluorouracil and uracil into DNA of cells treated with 5- 1-11, 1986. fluorodeoxyuridine. Mol. Pharmacol.. 21: 211-216. 1982. 32. Kun/., B. A. Mutagenesis and deoxyuribonucleotide pool imbalance. Mutât. Res.. 200: 133-147. 1988. 19. Jackson, R. C. The regulation of thymidylate biosynthesis in Novikoff hepatoma cells and the effect of amethopterin. 5-fluorodeoxyuridine and 3- 33. Reichard, P. Interactions between deoxyribonucleotide and DNA synthesis. deazauridinc. J. Biol. Chem.. 253: 7440-7446. 1978. Annu. Rev. Biochem., 57: 349-374, 1988. 20. Lorico. A.. Toffoli. G.. Boiocchi. M.. Erba, E.. Broggini. M.. Rappa. G., and 34. Meuth, M. The genetic consequences of nucleotide precursor pool imbalance D'Incaici, M. Accumulation of DNA strand breaks in cells exposed to in mammalian cells. Mutât.Res.. 126: 107-112. 1984. methotrexate or Nlo-propargyl-5.8-didea/afolic acid. Cancer Res.. 48: 2036- 35. Meuth. M. The molecular basis of mutations induced by triphosphate pool imbalances in mammalian cells. Exp. Cell Res., 181: SOS- 2041. 1988. SI 6, 1989. 21. Kunkel. I. A., and L.oeb. L. A. Fidelity of mammalian DNA polymerases. Science (Washington DC), 213: 765-767. 1981. 36. Howell. S. B.. Horn. D. K.. Sanga. R., Vick, J. S.. and Chan, T. C. K. Dipyridamole enhancement of etoposide sensitivity. Cancer Res., 49: 4147- 22. Brynolf. K.. Eliasson. R.. and Reichard. P. Formation of Okazaki fragments in polyoma DNA synthesis caused by incorporation of uracil. Cell. 13: 573- 4153. 1989. 37. Hirose. M.. Takeda. E.. Ninomiya, T.. Kuroda. Y., and Miyao, M. Synergistic 580. 1978. inhibitory effects of dipyridamole and vincristine on the growth of human 23. Seal. G.. and Sirover. M. A. Physical association of the human base-excision leukaemia and lymphoma cell lines. Br. J. Cancer, 56: 413-417, 1987. repair enzyme uracil DNA glycosylase with the 70.000 dalton catalytic 38. Howell. S. B.. Horn. D.. Sanga, R., Vick. J. S.. and Abramson, I. S. subunit of DNA polymerase «.Proc. Nati. Acad. Sci. USA. S3: 7608-7612. Comparison of the synergistic potential of etoposide, doxorubicin and vin- 1986. blastine cytotoxicity by dipyridamole. Cancer Res., 49: 3178-3183. 1989. 24. Grem. J. L.. Mulcahy. R. T.. Miller. E. M.. Allegra. C. J.. and Fischer. P. 39. Kusumoto. H.. Maehara. Y.. Anai. H., Kusumoto. T.. and Sugimachi. K. II. Interaction of deoxyuridine with fluorouracil and dipyridamoleina human Polentiation of Adriamycin cytotoxicity by dipyridamole against HeLa cells colon cancer cell line. Biochem. Pharmacol.. 38: 51-59. 1989. in rim. Cancer Res.. 48: 1208-1212. 1988. 25. Lonn. I'., and Lonn. S. Interaction between 5-fluorouracil and DNA of 40. Howell, S. B.. Vick. J.. and Andrews, P. A. Biochemical modulation of human colon adenocarcinoma. Cancer Res.. 44: 3414-3418. 1984. cisplatin by dipyridamole. Proc. Am. Assoc. Cancer Res., 28: 313, 1987. 26. I mil.ihl. T. DNA glycosylase. endonucleases for upurinic/apyrimidinic sites 41. Kennedy. D. G., Van Den Berg, H. W.. Clarke, R., and Murphy, R. F. and base excision repair. Prog. Res. Mol. Biol., 22: 135-142. Enhancement of methotrexate cytotoxicity towards the MDA MB436 human 1979. breast cancer cell line by dipyridamole. The role of methotrexate polygluta- 27. Li. J. C.. and Kaminskas. E. Progressive formation of DNA lesions in mates. Biochem. Pharmacol.. 33: 3053-3056, 1986. cultured Ehrlich ascites tumor cells treated with hydroxyurea. Cancer Res.. 42. Cabrai. S.. Leis. S.. Bover, L., Nembrot. M., and Mordoh. J. Dipyridamole 47: 2755-2758, 1987. inhibits reversion by thymidine of methotrexate effect and increases drug 28. Y'oshioka. A.. Tanaka. S.. Hiraoka. O.. Koyama. Y.. Hirota. Y.. Ayusawa. uptake in sarcoma 180 cells. Proc. Nati. Acad. Sci. USA, 81: 3200-3203, D.. Seno. T.. Garetl. C., and VVataya. Y. Deoxynucleotide triphosphate 1984.

2352

Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 1991 American Association for Cancer Research. Mechanism of Cell Death following Thymidylate Synthase Inhibition: 2 ′-Deoxyuridine-5′-triphosphate Accumulation, DNA Damage, and Growth Inhibition following Exposure to CB3717 and Dipyridamole

Nicola J. Curtin, Adrian L. Harris and G. Wynne Aherne

Cancer Res 1991;51:2346-2352.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/51/9/2346

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/51/9/2346. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.

Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 1991 American Association for Cancer Research.