International Journal of Radiation Biology
ISSN: 0955-3002 (Print) 1362-3095 (Online) Journal homepage: http://www.tandfonline.com/loi/irab20
Effect of Aphidicolin on DNA Synthesis, PLDrecovery and DNA Repair of Human Diploid Fibroblasts C. Baumstark-Khan To cite this article: C. Baumstark-Khan (1992) Effect of Aphidicolin on DNA Synthesis, PLDrecovery and DNA Repair of Human Diploid Fibroblasts, International Journal of Radiation Biology, 61:2, 191-197, DOI: 10.1080/09553009214550811 To link to this article: http://dx.doi.org/10.1080/09553009214550811
Published online: 03 Jul 2009.
Submit your article to this journal
Article views: 15
View related articles
Citing articles: 1 View citing articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=irab20 Download by: [McMaster University]
Date: 04 April 2016, At: 16:33
INT . J . RADIAT . BIOL .,
1992,
VOL .
61,
NO .
2, 191 -197
Effect of aphidicolin on DNA synthesis, PLD-recovery and DNA repair of human diploid fibroblasts C. BAUMSTARK-KHANt
Downloaded by [McMaster University] at 16:33 04 April 2016
(Received 5 April 1991; revision received 28 June 1991 ; accepted 15 July 1991)
Abstract . The effect of aphidicolin, a specific inhibitor of DNA polymerases a and 6, was studied on DNA synthesis, PLD-recovery and DNA double-strand break rejoining in Xirradiated human fibroblasts. In unirradiated, exponentially growing cells, aphidicolin (0 . 5-5,ug ml) inhibited DNA synthesis almost completely. This effect depended not only on aphidicolin concentration but also on the duration of pre-incubation . The action of aphidicolin was found to be reversible . When aphidicolin had been removed, colony forming ability was not affected in aphidicolin pretreated cells . Aphidicolin pretreated and irradiated cells showed a reduction in PLD-recovery, dependent on aphidicolin concentration and duration of pretreatment . The initial number of DNA double-strand breaks (calibrated by 1251 decay) was not affected by aphidicolin . However, after incubation for 90 min in the presence of aphidicolin there was a large reduction in double-strand break rejoining. With long incubation periods in aphidicolin rejoining was almost completely inhibited .
1.
Introduction
The radiosensitivity of normal tissue and tumour cells, although dependent on many secondary factors, is generally considered to result from irradiation-induced DNA damage and DNA repair . A knowledge of repair mechanisms and the possibility of interfering with them are thus of considerable interest in the radiation therapy of tumours . Post-irridation conditions may modify the processes of repair and recovery from potentially lethal irradiation damage (Phillips and Tolmach 1966) . Incubation of cells for some hours in non-growth medium or in the plateau phase of growth (contact inhibition) increased the fraction of surviving cells (Little 1969), whereas a brief anisotonic shock caused a decrease in cell survival (Iliakis et al. 1985) . Another way to influence the level of surviving cells is to use special inhibitors that affect DNA metabolism (Collins 1987) . The ideal inhibitor should be highly specific, so that any effect it may have can clearly be attributed to the involvement of a particular enzyme . j'Institut fur Strahlenbiologie, Sigmund-Freud-Str . D-5300 Bonn 1, Germany .
15,
Studies with JJ-arabinofuranosyladenine (/3-araA), an inhibitor of the mammalian DNA polymerases a, /3 and S, pointed to the importance of DNA polymerization in PLD repair processes . Since all these polymerases appeared to be inhibited by /1-araA to a comparable extent, it was not possible to identify the actions of a specific polymerase in DNA repair synthesis . Aphidicolin, a tetracyclic diterpene tetraol obtained from Cephalosporium aphidicola is another inhibitor of DNA synthesis . It was found to inhibit the growth of eukaryotic cells by inhibiting replicative DNA synthesis . Its effect on DNA synthesis could be attributed specifically to the inhibition of DNA polymerases a and S (Ikegami et al . 1978, Hubermann 1981, Goscin and Byrnes 1982), whereas polymerases /3 and y were found to be resistant to the drug (Hiibscher et al. 1979) . The main function of polymerase a is DNA synthesis (Kornberg 1982) . Additionally, it is likely that the enzyme plays a role in the repair of DNA lesions . Since most radiation-induced lesions are repaired via short repair patches, it has been assumed that polymerase a, which requires a gap of 25-50 nucleotides to initiate DNA synthesis, is not involved in DNA repair synthesis . This assumption, however, has not been proven by experimental data . The function of polymerase S in mammalian cells is still unknown ; a role in DNA repair of u .v.-induced DNA damage has been discussed (Dresler et al . 1989) . Aphidicolin has been reported to increase the frequency of X-ray-induced chromosomal aberrations in human lymphocytes (Kihlman and Anderson 1985), and was also used by Iliakis et al. (1985) to study PLD recovery and DNA repair in EAT-cells . Radford and Broadhurst (1988) postulated deregulation of DNA replication in aphidicolin-synchronized mouse L cells . This paper describes the action of aphidicolin on DNA synthesis, cellular survival and recovery kinetics as well as on DNA double-strand break induction and rejoining in cultured human fibroblasts .
0020-7616/92 $3 .00 © 1992 Taylor & Francis Ltd
Downloaded by [McMaster University] at 16:33 04 April 2016
192
C. Baumstark-Khan
2. Materials and methods
2 .4 .
2 .1 . Cell line and growth conditions
For survival experiments cells were trypsinized immediately after irradiation JP) or after a recovery period of 24h (LP) in non-growth medium (standard medium with 0 .5% FCS) and plated in complete medium at low cell densities in Petri dishes . The number of cells per dish was adjusted to compensate for cloning efficiency of the cell line and any anticipated lethal effect of the treatment, in order to obtain between 20 and 60 colonies per culture dish (9 cm diameter) . Cells were allowed to grow for 1821 days without any change of medium, afterwards colonies were stained with crystal violet (1 mg ml -1 in 3% formaldehyde solution) . Only colonies containing more than 50 cells were scored as survivors . Experiments were performed with six dishes per dose and repeated at least twice . All data from the irradiated samples were fit by least squares linear regression analysis to In (N/No) dose versus dose . The resulting survival curves [S= exp ( - aD - f D 2 ) ] were characterized by the parameters a and fl. For recovery studies confluent cell monolayers were exposed to 4 Gy of X-rays . These cells were plated at low densities in complete medium either directly after irradiation or were kept under nongrowth conditions for different time intervals of 1-72 h before plating . Further processing was done as described above . Recovery enhancement (RE) was calculated according to Baumstark-Khan (1990) for survival levels at a given time of recovery minus survival level at time zero (RE =S,-S,= o) . All these data for time intervals > 1 h were fitted by least squares linear regression analysis to I /RE versus I /time (comparable to Lineweaver-Burk transformation) by a FORTRAN program. Extrapolation to time t. reveals the reciprocal of maxiumum recovery enhancement (I /RE ordinate intercept) and extrapolation to RE . results in the half time of recovery by -1 /R, (abscissa intercept) .
The primary cell line NHF-23 originated from a skin biopsy obtained from a normal healthy female donor (age 42 years) and was established in this laboratory . Cells were sub-cultured under a standard set of conditions in Eagle's minimal essential medium (MEM, Boehringer, 209996) supplemented with 15% foetal calf serum (FCS, Biochrom KG, S-0115) glutamine (0 . 292µg ml -1 ) and pyruvate (0. 11 sg ml -1), buffered with sodium hydrogen carbonate . Cultures were maintained in a gas environment of 95% air, 5% carbon dioxide at 37°C. Experiments were performed with cells between population doublings 10 and 25 . Cells were harvested from plateau phase cultures by trypsinization (0 .5 mg ml -1 ) at 37°C . After counting, cells were inoculated at a density of 1 .0 x 10 4 cells cm -2 in 5 ml of complete medium (MEM+ 15% FCS) in 25 cm 2 flasks or Petri dishes (60 mm) . The medium was changed after 3 days of culture and confluent cell layers were obtained within 7 days.
2 .2. Irradiation conditions For cell survival studies confluent cell layers were irradiated at room temperature with X-rays generated by a X-ray unit (Miiller, RT 200) operated at 200 kV and 20 mA using a 0 . 5 mm copper filter yielding a dose-rate of 2 .3 Gy min -1 (focus-object distance: 300 mm) . To determine DNA double-strand breaks confluent cell layers were irradiated on ice with X-rays generated by a X-ray unit (Miiller, MG 300) operated at 300 kV and 10 mA yielding a dose-rate of 10 Gy min -1 (focus-object distance : 178 mm) .
Survival and PLD-recovery studies
2 .3. DNA synthesis DNA synthesis was measured after incorporation of 3H-thymidine (1 . 85 x 104 Bq/ml -1 [methyl- 3 H] thymidine (Amersham TRK 120, 185 GBq mmol -1 ) into cellular DNA (Painter 1986) . Liquid scintillation counting in cocktail (Quickszint 2000, Zinsser) was performed with a LSC 75000 counter (Beckman Instruments), programmed for automatic quench correction.
2 .5 Detection of DNA double-strand breaks DNA double-strand breaks were measured by a modification of the neutral filter elution method (Bradley and Kohn 1979) . Exponentially growing cells were labelled with 1 . 85 x 10 4 Bq/ml -1 [methyl3H] thymidine (Amersham TRK 120, 185 GBq mmol -1 ) . After 3 days the radioactive medium was discarded, cells were washed twice with phosphate
Downloaded by [McMaster University] at 16:33 04 April 2016
Effect of aphidicolin on human fibroblasts
buffered saline (PBS) and incubated for an additional period of 24 h in non-radioactive medium . Cells were layered onto 25mm (0.22 µm pore size) polyvinyichloride filters (Millipore GVWP 25) either immediately after X-irradiation (IE) or after a 90 min recovery period at 37 °C (LE), washed twice with ice-cold PBS, and lysed on the filters at room temperature for 30 min with 2 .5 ml lysis buffer (25mmol litre -1 EDTA, 50mmol litre -1 Tris, 70 mmol litre` 1 SDS, 50 mmol litre -1 glycine, pH 9.6 containing 0 . 5 mg ml -1 proteinase K) . Elution was performed with continuous pumping (flow rate of 0 .05 ml/min -1 with fractions collected at 90 min intervals) using elution buffer (lysis buffer without proteinase K, pH 9 .6) . Elution was finished after collection of 10 fractions per sample. The radioactivities of 1 ml aliquots of all samples and of the filters were measured after addition of 8 ml scintillation cocktail (Quickszint 2000, Zinsser) in an LS 7500 (Beckman) liquid scintillation counter, programmed for automatic quench correction . It is possible that neutral elution at pH 9 .6 is affected by some types of radiation damage other than double-strand breaks (Tilby et al . 1984) . The higher elution rate at this pH may also be attributed to more effective removal of DNA bound proteins (Okayasu and Iliakis 1989) . In the experiments described here elution was performed at pH 9 .6. The relative fraction of non-eluted DNA (FND) was calculated by FR D /FR D = o , where FR D and FRD=O were the fractions of DNA retained on the filters after 22 .5 ml elution (fraction 5), determined by a least squares minimization procedure from elution profile curves for irradiated cells and nonirradiated cells, respectively . The dose effect curves were calculated using computer fitting of the equation: In FND = - D/D o + 1 n n, where D is absorbed dose in Gy (or the number of 1251 decays), whereas Do (-1 /slope) and n represent constants. The doseresponse curves for double-strand breaks eluted immediately after irradiation (IE) or eluted after recovery (LE) were calculated by linear regression analysis . 2.6 1251-calibration 12'1
decay in DNA allows the numerical calibration of the neutral filter elution technique to measure DNA double-strand breaks (Radford and Hodgson 1985, Peak et al. 1988) . For this purpose exponentially growing cells were labelled with 1 .85 x 10 ° Bq ml-1 5- [125 I]-iodo-2'-deoxyuridine (Amersham IM 355, 74 TBq mmol -1 ) . After 2 days
1 93
of labelling, the radioactive medium was discarded, cells were washed twice with PBS, and incubated in non-radioactive medium for an additional period of 24h . Cells were then harvested by trypsinization, washed in PBS and resuspended in several samples of non-radioactive medium at a concentration of 5 x 105 cells ml -1 . The cell samples were dispensed into freezer vials, frozen slowly and stored in liquid nitrogen. The amount of 1251 incorporation into the DNA was measured by counting aliquots of cells in a LS 7500 (Beckman) liquid scintillation counter (counts per minute-cpm), programmed for automatic quench correction . The absolute activity (disintegrations per minute dpm) of the samples was computed using external standards (5-[ 12 I]-iodo2'-deoxyuridine in Quickszint 2000 cocktail) . The initial decay rate was measured to be 278 disintegrations per cell per day. The number of cumulative decays per cell at any time was calculated, based on the half-life time of 1251 (59 . 6 days) . Cells were thawed and assayed at weekly intervals by neutral elution. Assuming that the number of 125I decays is equal to the number of double-strand breaks, the number of double-strand breaks in irradiated samples (DSB(X)) can be calculated by : DSB(X)=FND (X-ray)/slope ( 125 I) .
2 .7 Aphidicolin treatment
Aphidicolin (Boehringer, Mannheim, Germany) dissolved in ethanol (10 mg ml -1 ) was added to plateau phase cells 1 h or 24 h before X-irradiation. For survival and recovery studies aphidicolin was washed out during trypsinization . For DNA doublestrand break determinations aphidicolin remained within the cells during incubation with nonradioactive medium for I h or 24h before Xirradiation as well as during the 90 min recovery period. 3. Results 3 .1 . Inhibition of DNA synthesis and viability of aphidicolin treated cells
In order to test the inhibition of DNA polymerases a and 6 by aphidicolin in the human fibroblast cell
line NHF-23, exponentially growing cells were exposed to_ various aphidicolin concentrations 1) (0-50 pg ml for 1 h or for 24 h . After this period
C. Baumstark-Khan
Downloaded by [McMaster University] at 16:33 04 April 2016
194
[methyl- 3H] thymidine (1 .85 x 104 Bq ml -1 ) was added for an additional period of 24 h . Incorporated 3H-activity was used as a measure of DNA synthesis (Figure 1) . Concentrations of 0 .05-0 .2Etg ml -1 produced a 50% inhibition . Concentrations higher than 2 µg ml- 1 are assumed to induce complete inhibition because the remaining 3-5% of DNA synthesis could be attributed to mitochondrial DNA synthesis by polymerase y which is resistant to aphidicolin (Hubscher et al . 1979) . The viability of aphidicolin pretreated unirradiated fibroblasts, when aphidicolin was washed out before plating was scarcely affected . For this test plateau-phase cells were incubated with various aphidicolin concentrations (0-50 ltg ml -1 ) for 8, 24 and 36 h . Cells were then trypsinized, washed thoroughly with PBS and plated for survival . In the range of 1-20 pg aphidicolin, surviving fractions are not significantly different from 100% levels . A slight decrease of survival was obvious with 50pg ml - ' (results not shown) . These results show that aphidicolin exposure before plating does not affect colonyforming ability of unirradiated cells . Therefore, aphidicolin inhibition of polymerases a and 6 seems to be reversible .
3 .3 . PLD recovery studies
3 .2 . Survival after X-irradiation
The neutral elution assay for the detection of DNA double-strand breaks is based on the rate at which double-stranded DNA elutes through an inert filter membrane under non-denaturating conditions . Figure 3 shows selected profiles for X-irradiated samples . As described by Bradley and Kohn (1979) the elution kinetics is not first order and the elution
Survival curves of cell line NHF-23 for immediate JP) and delayed plating (LP) are characterized by a and fl-values of 0 .784 and 0 .048 JP) and 0 . 239 and 0 . 047 (LP), respectively . Comparing LP to IP data a constant recovery ratio of 1 . 7 is obtained .
Figure 2 shows the time course of PLD recovery for X-irradiated (4 Gy) NHF-23 cells . The delayed plating kinetic (surviving fraction versus recovery time) displays saturation behaviour . The changes in survival levels are most prominent during the first hours after X-irradiation . With increasing time cell survival reaches a plateau at about 24h . Aphidicolin treatment resulted in a reduced PLD recovery . After 1 h pre-incubation with aphidicolin (1 pg ml -1 ), the increase in cell survival appeared to be unaffected . The higher aphidicolin concentration (20 Mg ml - ') showed stronger inhibition of recovery from potentially lethal lesions . The surviving fraction in this case was comparable to that observed for cells plated immediately after irradiation . The inhibition effect for low aphidicolin concentrations becomes more pronounced with longer preincubation times . After a period of 24 h preincubation there was no increase in cell survival also at the concentration of 1 pg ml -1 . 3 .4 DNA double-strand break induction and rejoining
1 .00N
u a)
0 .50-
t
C
0 .10-
- o----- --------
2
0 .01 0 10
20
30
40
•
0 1
0
20
µg/ml µg/ml µ /ml
• •
o
µg/ml µg/ml 20µ~/ml 0 1
50 0
12 24 36 48 60 72
0
12 24 36 48 60 72
ophidicolin (µg/ml) incubation time (h)
Figure 1 . Inhibition of DNA synthesis as a function of aphidicolin concentrations for NHF-23 fibroblasts treated with aphidicolin for ( •) 1 and (0) 24 h, respectively.
Experiments were repeated twice (6 samples per concentration) ; where no error bars are drawn, the size of the point symbol is larger than the standard deviation .
incubation time (h)
Figure 2 . PLD-recovery as a function of time after 4 Gy of X-rays after pre-incubation with two different aphidicolin concentrations for 1 and 24 h . Experiments were repeated twice (12 samples per concentration) ; where no error bars are drawn, the size of the point symbol is larger than the standard deviation .
Effect of aphidicolin on human fibroblasts
195
1 .000 .50-
0 .10 0 .05-
0 .01 0
0 Gy, IE ~--~ 100 Gy, IE w 100 Gy, LE ~ 100 Gy, LE, aphidicolin
2
4
6
8
10 0
Downloaded by [McMaster University] at 16:33 04 April 2016
fraction number
Figure 3 . Elution of DNA from polyvinylchloride filters at pH 9 .6 . Cells were lysed on the filters and eluted immediately after X-irradiation (IE) or after a 90-min recovery period (LE) in absence or presence of aphidicolin (20,ug ml -1 , 24h preincubation) .
rate decreases with elution time (fraction number) less rapidly than simply exponential . From such elution profiles, dose-response relationships could be established . The dose-response curve (non-eluted DNA versus 1251 decays) for double-strand breaks induced by 1251 decays could be fitted to In FND= -1 .00276 x 10 -4 decays per cell +ln 0 . 999 (r2 =0 . 935) . Assuming one 125 I decay to induce one double-strand break in cellular DNA allows dividing of the FND-values from X-irradiated samples by the sloe of the regression line from In FND (1 5 1) versus 12 I decays . This results in the number of doublestrand breaks for the X-irradiated samples . Figure 4 shows the numbers of double-strand breaks per cell after application of the 1251 calibration . The amount of initial DNA damage, measured by neutral elution immediately after Xirradiation (IE) is in the same range for untreated and for aphidicolin treated cells . The slopes of the dose response curves are not significantly different (Table 1) . When recovery was allowed by incubating the X-irradiated cells for 90 min before elution (LE) the response was different . Cells incubated without aphidicolin were able to rejoin almost all DNA double-strand breaks during the recovery period (90 min) . Aphidicolin-treated cells, however, show a significant reduction in repair capacity (Table 1) . It was also found, that an aphidicolin preincubation period of 24 h resulted in a significant loss of double-strand break rejoining when compared to the 1 h pre-incubation period . Double-strand break rejoining is also more inhibited at the higher aphidicolin concentration (20 µg ml -1 ) .
Figure 4. Induction of double-strand breaks measured with the neutral elution method immediately after Xirradiation (IE) and remaining double-strand breaks after a 90 min recovery period (LE) both with aphidicolin pretreatment for 1 and 24h before irradiation . Experiments were repeated twice (6 samples per concentration) ; where no error bars are drawn, the size of the point symbol is larger than the standard deviation .
4. Discussion
There has been considerable controversy over the effects of aphidicolin-treated cells . This is partly due to the wide range of drug concentrations and to variations in application protocols used by different investigators. Additionally dose-relationships for basic parameters such as DNA synthesis inhibition and cytotoxicity studies for the cell types used are missing . The results described here show that DNA synthesis could be almost completely inhibited by concentrations of 0. 5-1 . 0 yg ml -1 aphidicolin . In spite of this inhibition of DNA synthesis, unirradiated cells, pretreated with aphidicolin show no significant differences in colony formation after removal of aphidicolin. Thus aphidicolin could be used to investigate PLD recovery in X-irradiated cells . Iliakis et al . (1982) found that aphidicolin does not inhibit PLD recovery in plateau-phase EAT-cells
196
C. Baumstark-Khan
Downloaded by [McMaster University] at 16:33 04 April 2016
Table 1 .
Parameters of dose-response curves for DNA double-strand breakage, measured by neutral elution' .
Aphidicolin concentration (tg m1 - ')
Preincubation (h)
dsb cell -1 Gy - '
t-test
IEb
LE`
2p IE
2p LE
0 1 1 5 5 20 20
0 1 24 1 24 1 24
168 . 4 190 . 0 186. 5 153 . 5 156 . 9 141 . 8 156 .9
13. 9 28. 8 17 . 9 40 . 5 80. 6 71 . 2 104.0
>0-2 >0 . 3 >0 . 3 >0 . 1 >0 . 8 >0 . 3
0 .4 < 0 .0001
ISSN: 0955-3002 (Print) 1362-3095 (Online) Journal homepage: http://www.tandfonline.com/loi/irab20
Effect of Aphidicolin on DNA Synthesis, PLDrecovery and DNA Repair of Human Diploid Fibroblasts C. Baumstark-Khan To cite this article: C. Baumstark-Khan (1992) Effect of Aphidicolin on DNA Synthesis, PLDrecovery and DNA Repair of Human Diploid Fibroblasts, International Journal of Radiation Biology, 61:2, 191-197, DOI: 10.1080/09553009214550811 To link to this article: http://dx.doi.org/10.1080/09553009214550811
Published online: 03 Jul 2009.
Submit your article to this journal
Article views: 15
View related articles
Citing articles: 1 View citing articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=irab20 Download by: [McMaster University]
Date: 04 April 2016, At: 16:33
INT . J . RADIAT . BIOL .,
1992,
VOL .
61,
NO .
2, 191 -197
Effect of aphidicolin on DNA synthesis, PLD-recovery and DNA repair of human diploid fibroblasts C. BAUMSTARK-KHANt
Downloaded by [McMaster University] at 16:33 04 April 2016
(Received 5 April 1991; revision received 28 June 1991 ; accepted 15 July 1991)
Abstract . The effect of aphidicolin, a specific inhibitor of DNA polymerases a and 6, was studied on DNA synthesis, PLD-recovery and DNA double-strand break rejoining in Xirradiated human fibroblasts. In unirradiated, exponentially growing cells, aphidicolin (0 . 5-5,ug ml) inhibited DNA synthesis almost completely. This effect depended not only on aphidicolin concentration but also on the duration of pre-incubation . The action of aphidicolin was found to be reversible . When aphidicolin had been removed, colony forming ability was not affected in aphidicolin pretreated cells . Aphidicolin pretreated and irradiated cells showed a reduction in PLD-recovery, dependent on aphidicolin concentration and duration of pretreatment . The initial number of DNA double-strand breaks (calibrated by 1251 decay) was not affected by aphidicolin . However, after incubation for 90 min in the presence of aphidicolin there was a large reduction in double-strand break rejoining. With long incubation periods in aphidicolin rejoining was almost completely inhibited .
1.
Introduction
The radiosensitivity of normal tissue and tumour cells, although dependent on many secondary factors, is generally considered to result from irradiation-induced DNA damage and DNA repair . A knowledge of repair mechanisms and the possibility of interfering with them are thus of considerable interest in the radiation therapy of tumours . Post-irridation conditions may modify the processes of repair and recovery from potentially lethal irradiation damage (Phillips and Tolmach 1966) . Incubation of cells for some hours in non-growth medium or in the plateau phase of growth (contact inhibition) increased the fraction of surviving cells (Little 1969), whereas a brief anisotonic shock caused a decrease in cell survival (Iliakis et al. 1985) . Another way to influence the level of surviving cells is to use special inhibitors that affect DNA metabolism (Collins 1987) . The ideal inhibitor should be highly specific, so that any effect it may have can clearly be attributed to the involvement of a particular enzyme . j'Institut fur Strahlenbiologie, Sigmund-Freud-Str . D-5300 Bonn 1, Germany .
15,
Studies with JJ-arabinofuranosyladenine (/3-araA), an inhibitor of the mammalian DNA polymerases a, /3 and S, pointed to the importance of DNA polymerization in PLD repair processes . Since all these polymerases appeared to be inhibited by /1-araA to a comparable extent, it was not possible to identify the actions of a specific polymerase in DNA repair synthesis . Aphidicolin, a tetracyclic diterpene tetraol obtained from Cephalosporium aphidicola is another inhibitor of DNA synthesis . It was found to inhibit the growth of eukaryotic cells by inhibiting replicative DNA synthesis . Its effect on DNA synthesis could be attributed specifically to the inhibition of DNA polymerases a and S (Ikegami et al . 1978, Hubermann 1981, Goscin and Byrnes 1982), whereas polymerases /3 and y were found to be resistant to the drug (Hiibscher et al. 1979) . The main function of polymerase a is DNA synthesis (Kornberg 1982) . Additionally, it is likely that the enzyme plays a role in the repair of DNA lesions . Since most radiation-induced lesions are repaired via short repair patches, it has been assumed that polymerase a, which requires a gap of 25-50 nucleotides to initiate DNA synthesis, is not involved in DNA repair synthesis . This assumption, however, has not been proven by experimental data . The function of polymerase S in mammalian cells is still unknown ; a role in DNA repair of u .v.-induced DNA damage has been discussed (Dresler et al . 1989) . Aphidicolin has been reported to increase the frequency of X-ray-induced chromosomal aberrations in human lymphocytes (Kihlman and Anderson 1985), and was also used by Iliakis et al. (1985) to study PLD recovery and DNA repair in EAT-cells . Radford and Broadhurst (1988) postulated deregulation of DNA replication in aphidicolin-synchronized mouse L cells . This paper describes the action of aphidicolin on DNA synthesis, cellular survival and recovery kinetics as well as on DNA double-strand break induction and rejoining in cultured human fibroblasts .
0020-7616/92 $3 .00 © 1992 Taylor & Francis Ltd
Downloaded by [McMaster University] at 16:33 04 April 2016
192
C. Baumstark-Khan
2. Materials and methods
2 .4 .
2 .1 . Cell line and growth conditions
For survival experiments cells were trypsinized immediately after irradiation JP) or after a recovery period of 24h (LP) in non-growth medium (standard medium with 0 .5% FCS) and plated in complete medium at low cell densities in Petri dishes . The number of cells per dish was adjusted to compensate for cloning efficiency of the cell line and any anticipated lethal effect of the treatment, in order to obtain between 20 and 60 colonies per culture dish (9 cm diameter) . Cells were allowed to grow for 1821 days without any change of medium, afterwards colonies were stained with crystal violet (1 mg ml -1 in 3% formaldehyde solution) . Only colonies containing more than 50 cells were scored as survivors . Experiments were performed with six dishes per dose and repeated at least twice . All data from the irradiated samples were fit by least squares linear regression analysis to In (N/No) dose versus dose . The resulting survival curves [S= exp ( - aD - f D 2 ) ] were characterized by the parameters a and fl. For recovery studies confluent cell monolayers were exposed to 4 Gy of X-rays . These cells were plated at low densities in complete medium either directly after irradiation or were kept under nongrowth conditions for different time intervals of 1-72 h before plating . Further processing was done as described above . Recovery enhancement (RE) was calculated according to Baumstark-Khan (1990) for survival levels at a given time of recovery minus survival level at time zero (RE =S,-S,= o) . All these data for time intervals > 1 h were fitted by least squares linear regression analysis to I /RE versus I /time (comparable to Lineweaver-Burk transformation) by a FORTRAN program. Extrapolation to time t. reveals the reciprocal of maxiumum recovery enhancement (I /RE ordinate intercept) and extrapolation to RE . results in the half time of recovery by -1 /R, (abscissa intercept) .
The primary cell line NHF-23 originated from a skin biopsy obtained from a normal healthy female donor (age 42 years) and was established in this laboratory . Cells were sub-cultured under a standard set of conditions in Eagle's minimal essential medium (MEM, Boehringer, 209996) supplemented with 15% foetal calf serum (FCS, Biochrom KG, S-0115) glutamine (0 . 292µg ml -1 ) and pyruvate (0. 11 sg ml -1), buffered with sodium hydrogen carbonate . Cultures were maintained in a gas environment of 95% air, 5% carbon dioxide at 37°C. Experiments were performed with cells between population doublings 10 and 25 . Cells were harvested from plateau phase cultures by trypsinization (0 .5 mg ml -1 ) at 37°C . After counting, cells were inoculated at a density of 1 .0 x 10 4 cells cm -2 in 5 ml of complete medium (MEM+ 15% FCS) in 25 cm 2 flasks or Petri dishes (60 mm) . The medium was changed after 3 days of culture and confluent cell layers were obtained within 7 days.
2 .2. Irradiation conditions For cell survival studies confluent cell layers were irradiated at room temperature with X-rays generated by a X-ray unit (Miiller, RT 200) operated at 200 kV and 20 mA using a 0 . 5 mm copper filter yielding a dose-rate of 2 .3 Gy min -1 (focus-object distance: 300 mm) . To determine DNA double-strand breaks confluent cell layers were irradiated on ice with X-rays generated by a X-ray unit (Miiller, MG 300) operated at 300 kV and 10 mA yielding a dose-rate of 10 Gy min -1 (focus-object distance : 178 mm) .
Survival and PLD-recovery studies
2 .3. DNA synthesis DNA synthesis was measured after incorporation of 3H-thymidine (1 . 85 x 104 Bq/ml -1 [methyl- 3 H] thymidine (Amersham TRK 120, 185 GBq mmol -1 ) into cellular DNA (Painter 1986) . Liquid scintillation counting in cocktail (Quickszint 2000, Zinsser) was performed with a LSC 75000 counter (Beckman Instruments), programmed for automatic quench correction.
2 .5 Detection of DNA double-strand breaks DNA double-strand breaks were measured by a modification of the neutral filter elution method (Bradley and Kohn 1979) . Exponentially growing cells were labelled with 1 . 85 x 10 4 Bq/ml -1 [methyl3H] thymidine (Amersham TRK 120, 185 GBq mmol -1 ) . After 3 days the radioactive medium was discarded, cells were washed twice with phosphate
Downloaded by [McMaster University] at 16:33 04 April 2016
Effect of aphidicolin on human fibroblasts
buffered saline (PBS) and incubated for an additional period of 24 h in non-radioactive medium . Cells were layered onto 25mm (0.22 µm pore size) polyvinyichloride filters (Millipore GVWP 25) either immediately after X-irradiation (IE) or after a 90 min recovery period at 37 °C (LE), washed twice with ice-cold PBS, and lysed on the filters at room temperature for 30 min with 2 .5 ml lysis buffer (25mmol litre -1 EDTA, 50mmol litre -1 Tris, 70 mmol litre` 1 SDS, 50 mmol litre -1 glycine, pH 9.6 containing 0 . 5 mg ml -1 proteinase K) . Elution was performed with continuous pumping (flow rate of 0 .05 ml/min -1 with fractions collected at 90 min intervals) using elution buffer (lysis buffer without proteinase K, pH 9 .6) . Elution was finished after collection of 10 fractions per sample. The radioactivities of 1 ml aliquots of all samples and of the filters were measured after addition of 8 ml scintillation cocktail (Quickszint 2000, Zinsser) in an LS 7500 (Beckman) liquid scintillation counter, programmed for automatic quench correction . It is possible that neutral elution at pH 9 .6 is affected by some types of radiation damage other than double-strand breaks (Tilby et al . 1984) . The higher elution rate at this pH may also be attributed to more effective removal of DNA bound proteins (Okayasu and Iliakis 1989) . In the experiments described here elution was performed at pH 9 .6. The relative fraction of non-eluted DNA (FND) was calculated by FR D /FR D = o , where FR D and FRD=O were the fractions of DNA retained on the filters after 22 .5 ml elution (fraction 5), determined by a least squares minimization procedure from elution profile curves for irradiated cells and nonirradiated cells, respectively . The dose effect curves were calculated using computer fitting of the equation: In FND = - D/D o + 1 n n, where D is absorbed dose in Gy (or the number of 1251 decays), whereas Do (-1 /slope) and n represent constants. The doseresponse curves for double-strand breaks eluted immediately after irradiation (IE) or eluted after recovery (LE) were calculated by linear regression analysis . 2.6 1251-calibration 12'1
decay in DNA allows the numerical calibration of the neutral filter elution technique to measure DNA double-strand breaks (Radford and Hodgson 1985, Peak et al. 1988) . For this purpose exponentially growing cells were labelled with 1 .85 x 10 ° Bq ml-1 5- [125 I]-iodo-2'-deoxyuridine (Amersham IM 355, 74 TBq mmol -1 ) . After 2 days
1 93
of labelling, the radioactive medium was discarded, cells were washed twice with PBS, and incubated in non-radioactive medium for an additional period of 24h . Cells were then harvested by trypsinization, washed in PBS and resuspended in several samples of non-radioactive medium at a concentration of 5 x 105 cells ml -1 . The cell samples were dispensed into freezer vials, frozen slowly and stored in liquid nitrogen. The amount of 1251 incorporation into the DNA was measured by counting aliquots of cells in a LS 7500 (Beckman) liquid scintillation counter (counts per minute-cpm), programmed for automatic quench correction . The absolute activity (disintegrations per minute dpm) of the samples was computed using external standards (5-[ 12 I]-iodo2'-deoxyuridine in Quickszint 2000 cocktail) . The initial decay rate was measured to be 278 disintegrations per cell per day. The number of cumulative decays per cell at any time was calculated, based on the half-life time of 1251 (59 . 6 days) . Cells were thawed and assayed at weekly intervals by neutral elution. Assuming that the number of 125I decays is equal to the number of double-strand breaks, the number of double-strand breaks in irradiated samples (DSB(X)) can be calculated by : DSB(X)=FND (X-ray)/slope ( 125 I) .
2 .7 Aphidicolin treatment
Aphidicolin (Boehringer, Mannheim, Germany) dissolved in ethanol (10 mg ml -1 ) was added to plateau phase cells 1 h or 24 h before X-irradiation. For survival and recovery studies aphidicolin was washed out during trypsinization . For DNA doublestrand break determinations aphidicolin remained within the cells during incubation with nonradioactive medium for I h or 24h before Xirradiation as well as during the 90 min recovery period. 3. Results 3 .1 . Inhibition of DNA synthesis and viability of aphidicolin treated cells
In order to test the inhibition of DNA polymerases a and 6 by aphidicolin in the human fibroblast cell
line NHF-23, exponentially growing cells were exposed to_ various aphidicolin concentrations 1) (0-50 pg ml for 1 h or for 24 h . After this period
C. Baumstark-Khan
Downloaded by [McMaster University] at 16:33 04 April 2016
194
[methyl- 3H] thymidine (1 .85 x 104 Bq ml -1 ) was added for an additional period of 24 h . Incorporated 3H-activity was used as a measure of DNA synthesis (Figure 1) . Concentrations of 0 .05-0 .2Etg ml -1 produced a 50% inhibition . Concentrations higher than 2 µg ml- 1 are assumed to induce complete inhibition because the remaining 3-5% of DNA synthesis could be attributed to mitochondrial DNA synthesis by polymerase y which is resistant to aphidicolin (Hubscher et al . 1979) . The viability of aphidicolin pretreated unirradiated fibroblasts, when aphidicolin was washed out before plating was scarcely affected . For this test plateau-phase cells were incubated with various aphidicolin concentrations (0-50 ltg ml -1 ) for 8, 24 and 36 h . Cells were then trypsinized, washed thoroughly with PBS and plated for survival . In the range of 1-20 pg aphidicolin, surviving fractions are not significantly different from 100% levels . A slight decrease of survival was obvious with 50pg ml - ' (results not shown) . These results show that aphidicolin exposure before plating does not affect colonyforming ability of unirradiated cells . Therefore, aphidicolin inhibition of polymerases a and 6 seems to be reversible .
3 .3 . PLD recovery studies
3 .2 . Survival after X-irradiation
The neutral elution assay for the detection of DNA double-strand breaks is based on the rate at which double-stranded DNA elutes through an inert filter membrane under non-denaturating conditions . Figure 3 shows selected profiles for X-irradiated samples . As described by Bradley and Kohn (1979) the elution kinetics is not first order and the elution
Survival curves of cell line NHF-23 for immediate JP) and delayed plating (LP) are characterized by a and fl-values of 0 .784 and 0 .048 JP) and 0 . 239 and 0 . 047 (LP), respectively . Comparing LP to IP data a constant recovery ratio of 1 . 7 is obtained .
Figure 2 shows the time course of PLD recovery for X-irradiated (4 Gy) NHF-23 cells . The delayed plating kinetic (surviving fraction versus recovery time) displays saturation behaviour . The changes in survival levels are most prominent during the first hours after X-irradiation . With increasing time cell survival reaches a plateau at about 24h . Aphidicolin treatment resulted in a reduced PLD recovery . After 1 h pre-incubation with aphidicolin (1 pg ml -1 ), the increase in cell survival appeared to be unaffected . The higher aphidicolin concentration (20 Mg ml - ') showed stronger inhibition of recovery from potentially lethal lesions . The surviving fraction in this case was comparable to that observed for cells plated immediately after irradiation . The inhibition effect for low aphidicolin concentrations becomes more pronounced with longer preincubation times . After a period of 24 h preincubation there was no increase in cell survival also at the concentration of 1 pg ml -1 . 3 .4 DNA double-strand break induction and rejoining
1 .00N
u a)
0 .50-
t
C
0 .10-
- o----- --------
2
0 .01 0 10
20
30
40
•
0 1
0
20
µg/ml µg/ml µ /ml
• •
o
µg/ml µg/ml 20µ~/ml 0 1
50 0
12 24 36 48 60 72
0
12 24 36 48 60 72
ophidicolin (µg/ml) incubation time (h)
Figure 1 . Inhibition of DNA synthesis as a function of aphidicolin concentrations for NHF-23 fibroblasts treated with aphidicolin for ( •) 1 and (0) 24 h, respectively.
Experiments were repeated twice (6 samples per concentration) ; where no error bars are drawn, the size of the point symbol is larger than the standard deviation .
incubation time (h)
Figure 2 . PLD-recovery as a function of time after 4 Gy of X-rays after pre-incubation with two different aphidicolin concentrations for 1 and 24 h . Experiments were repeated twice (12 samples per concentration) ; where no error bars are drawn, the size of the point symbol is larger than the standard deviation .
Effect of aphidicolin on human fibroblasts
195
1 .000 .50-
0 .10 0 .05-
0 .01 0
0 Gy, IE ~--~ 100 Gy, IE w 100 Gy, LE ~ 100 Gy, LE, aphidicolin
2
4
6
8
10 0
Downloaded by [McMaster University] at 16:33 04 April 2016
fraction number
Figure 3 . Elution of DNA from polyvinylchloride filters at pH 9 .6 . Cells were lysed on the filters and eluted immediately after X-irradiation (IE) or after a 90-min recovery period (LE) in absence or presence of aphidicolin (20,ug ml -1 , 24h preincubation) .
rate decreases with elution time (fraction number) less rapidly than simply exponential . From such elution profiles, dose-response relationships could be established . The dose-response curve (non-eluted DNA versus 1251 decays) for double-strand breaks induced by 1251 decays could be fitted to In FND= -1 .00276 x 10 -4 decays per cell +ln 0 . 999 (r2 =0 . 935) . Assuming one 125 I decay to induce one double-strand break in cellular DNA allows dividing of the FND-values from X-irradiated samples by the sloe of the regression line from In FND (1 5 1) versus 12 I decays . This results in the number of doublestrand breaks for the X-irradiated samples . Figure 4 shows the numbers of double-strand breaks per cell after application of the 1251 calibration . The amount of initial DNA damage, measured by neutral elution immediately after Xirradiation (IE) is in the same range for untreated and for aphidicolin treated cells . The slopes of the dose response curves are not significantly different (Table 1) . When recovery was allowed by incubating the X-irradiated cells for 90 min before elution (LE) the response was different . Cells incubated without aphidicolin were able to rejoin almost all DNA double-strand breaks during the recovery period (90 min) . Aphidicolin-treated cells, however, show a significant reduction in repair capacity (Table 1) . It was also found, that an aphidicolin preincubation period of 24 h resulted in a significant loss of double-strand break rejoining when compared to the 1 h pre-incubation period . Double-strand break rejoining is also more inhibited at the higher aphidicolin concentration (20 µg ml -1 ) .
Figure 4. Induction of double-strand breaks measured with the neutral elution method immediately after Xirradiation (IE) and remaining double-strand breaks after a 90 min recovery period (LE) both with aphidicolin pretreatment for 1 and 24h before irradiation . Experiments were repeated twice (6 samples per concentration) ; where no error bars are drawn, the size of the point symbol is larger than the standard deviation .
4. Discussion
There has been considerable controversy over the effects of aphidicolin-treated cells . This is partly due to the wide range of drug concentrations and to variations in application protocols used by different investigators. Additionally dose-relationships for basic parameters such as DNA synthesis inhibition and cytotoxicity studies for the cell types used are missing . The results described here show that DNA synthesis could be almost completely inhibited by concentrations of 0. 5-1 . 0 yg ml -1 aphidicolin . In spite of this inhibition of DNA synthesis, unirradiated cells, pretreated with aphidicolin show no significant differences in colony formation after removal of aphidicolin. Thus aphidicolin could be used to investigate PLD recovery in X-irradiated cells . Iliakis et al . (1982) found that aphidicolin does not inhibit PLD recovery in plateau-phase EAT-cells
196
C. Baumstark-Khan
Downloaded by [McMaster University] at 16:33 04 April 2016
Table 1 .
Parameters of dose-response curves for DNA double-strand breakage, measured by neutral elution' .
Aphidicolin concentration (tg m1 - ')
Preincubation (h)
dsb cell -1 Gy - '
t-test
IEb
LE`
2p IE
2p LE
0 1 1 5 5 20 20
0 1 24 1 24 1 24
168 . 4 190 . 0 186. 5 153 . 5 156 . 9 141 . 8 156 .9
13. 9 28. 8 17 . 9 40 . 5 80. 6 71 . 2 104.0
>0-2 >0 . 3 >0 . 3 >0 . 1 >0 . 8 >0 . 3
0 .4 < 0 .0001
