J. Cell Sci. 72, 213-226 (1984) 213Printed in Great Britain © The Company of Biologists Limited 1984

DNA REPLICATION AND REPAIR IN TILAPIA CELLSI. THE EFFECT OF ULTRAVIOLET RADIATION

F. H. YEW AND L. M. CHANGDepartment of Zoology, National Taiwan University, Taipei,Taiwan, R.O.C.

SUMMARY

The effect of ultraviolet radiation on a cell line established from the warm water fish Tilapia hasbeen assessed by measuring the rate of DNA synthesis, excision repair, post-replication repair andcell survival. The cells tolerate ultraviolet radiation better than mammalian cells with respect toDNA synthesis, post-replication repair and cell survival. They are also efficient in excision repair,which in other fish cell lines has been found to be at a low level or absent. Their response to theinhibitors hydroxyurea and 1-jS-D-arabinofuranosylcytosine is less sensitive than that of other celllines, yet the cells seem to have very small pools of DNA precursor.

INTRODUCTION

Cell lines derived from fish have attracted considerable interest in recent years.Regan, Carrier, Samet & Olla (1972) studied the cells from two closely related marinefish with very different life spans, which, they found, corresponded to their photo-reactivation activities. Mitani, Etoh & Egami (1982) found that cells from goldfishwere more resistant to y-irradiation than mammalian cells; this was consistent withearlier studies on the individual goldfish (Hyodo-Taguchi & Egami, 1969; Purdom& Woodhead, 1973). However, the same cells were found to be more sensitive toultraviolet (u.v.) radiation than mammalian cells (Mano, Mitani, Etoh & Egami,1980), although the amount of DNA damage was the same as that found in mam-malian cells (Mitani & Egami, 1982; Mano, Kator & Egami, 1982).

Photoreactivation of pyrimidine dimers on DNA has long been a feature of fish cellrepair. The work of Hart, Setlow & Woodhead (1977) related the induction ofpyrimidine dimers to tumour formation and their photoreactivation to the absence oftumorigenicity. More recent investigations into the lethal and mutagenic effects ofradiation and chemicals (Mitani, 1983) have all demonstrated that photoreactivationplays a vital role in survival and transformation of fish cells from the cell lines studied;it was concluded that excision repair was minimal and probably of minor importance(Mano etal. 1980, 1982; Regan et al. 1972).

In this paper we present the results of studies of the effects of u.v. radiation on cellsderived from the warm water fish Tilapia. We have investigated the effects of radia-tion on excision repair, DNA replication, post-replication repair and cell prolifera-tion.

Key words: Tilapia, DNA replication, DNA repair.

214 F.H. Yew and L. M. Chang

MATERIALS AND METHODS

Cell culture

Tilapia kidney (TK) cells were originally derived from the hybrid species of Tilapia mossainbicaand T. nilotica (CheneZ al. 1983). They were routinely cultured in 25-ml plastic flasks in Lebovitz-15 medium supplemented with 10% (v/v) foetal calf serum and antibiotics at 31 °C. For mostexperiments, cells were transferred to 24-well plate at a 1-5 split ratio, from confluent culture.

Chinese hamster ovary (CHO) cells were cultured in McCoy's medium supplemented with 10 %foetal calf serum.

Cell killingCells were trypsinized, washed twice and resuspended in cold phosphate-buffered saline (PBS),

u.v. radiation was carried out in a 30 mm Petridish in 1 ml of cold PBS at a dose rate of l-0Jm~2s~'.for various lengths of time. When a 24-well plate was used, because of the shielding effect of the wallof the well, the dose rate was reduced to 0-9Jm~2s~'. After irradiation the cells were centrifugedand transferred to normal growth medium in either 25-ml flasks or 24-well plates, the number of cellsinoculated was recorded. To prevent photoreactivation the cells were kept in the dark duringincubation. For proliferation studies the cells in the flasks were trypsinized and counted onsubsequent days. The percentage of cells engaged in DNA synthesis was also measured. The cellsin the 24-well plates were pulse-labelled with [3H]thymidine (1 /lCi/ml, 31 Ci/mmol) for 1 h, thenlysed in 0-5 ml of alkaline solution (5% (w/v) sucrose, 0-3M-NaOH, 0-5 M-NaCl) for 5 min andDNA was precipitated by adding a drop of 50% cold trichloroacetic acid and standing at 4°Covernight. The precipitate was collected by filtering through Whatman 2-4cm GF/C glass-fibrefilters, washed with 5% cold trichloroacetic acid and 95% (v/v) ethanol, dried and counted in ascintillation counter.

Nucleoid sedimentationThe sedimentation behaviour of nucleoids derived from TK cells was studied using the method

of Cook & Brazell (1975). Cells were labelled with [3H]thymidine (OS^Ci/ml, 31 Ci/mmol) for48 h at 31 °C. They were trypsinized, washed twice and suspended in PBS; 50/il of cell suspensioncontaining 5 X 104 cells was gently layered on 150^d of a lysis buffer (2M-NaCl, 0-01 M - E D T A ,0 1 % Triton X-100, pH 8-0) on top of a linear sucrose gradient (15 % to 30 % sucrose, 2-0 M-NaCl,0-01 M - E D T A , pH 8 0 ; total volume 4-8ml). For the titration of DNA supercoils, ethidiumbromide (0—10 ^g/ml) was present in the lysis buffer as well as in the gradient. The gradients weretransferred to a RPS 50-2 rotor and centrifuged at 30000 rev./min on a Hitachi SCP 85H ultra-centnfuge at 20°C for 40 min. After centrifugation, 20 fractions were collected directly into count-ing vials mixed with 5 ml scintillation fluid (66-67 % toluene, 33-33 % Triton X-100, 5 g PPO, 0-1 gPOPOP in 1 litre) and counted on a LKB 1217 Rackbeta liquid scintillation counter.

Rate of DNA synthesisThe method was based on the work of Painter (1977). Cells were generally labelled with [14C]-

thymidine (0-02jiCi/ml, 56mCi/mmol) for 48 h. The radioactive medium was replaced with freshmedium and the cells were further incubated for 4 h. After u.v. radiation at various doses, cells werepulse-labelled with [3H]thymidine (20 /iCi/ml, 31 Ci/mmol) for 10 min at various times (0, 30 min,1 h, 2 h), they were then lysed with alkali, precipitated with trichloroacetic acid, filtered and coun-ted.

Alkaline sucrose gradient sedimentation

Alkaline sucrose gradient sedimentation was used in assessing the presence of single-strand breaksin the parental DNA and post-replication (daughter strand) repair. The gradient contained 0-3 M-NaOH, 0-5 M-NaCl, 0-01 M - E D T A , 5% to 20% sucrose (pH 12-3) and was always kept at 4°C.After centrifugation, 20 fractions were collected. DNA was precipitated by adding a drop of 50 %ice-cold trichloroacetic acid. The fractions were kept in the cold overnight, then filtered through

DNA replication and repair in Tilapia cells 215

Whatman 2-4 cm GF/C glass-fibre filters. They were washed with 5 % cold trichloroacetic acid and95 % ethanol and counted in a liquid scintillation counter phage A DNA was used as a molecularweight marker.

Assay of DNA single-strand breaksThe method was based on those of Collins, Schor & Johnson (1977) and Johnson & Collins

(1978). The cells were generally labelled with [3H]thymidine (0-5/iCi/ml, 31 Ci/mmol) for 48hbefore u.v. radiation. The cells were incubated with or without inhibitors for 1 h before radiation.After u.v. radiation followed by incubation with or without inhibitors, the cells were scraped off thesurface, washed with PBS and suspended in 0-2 ml PBS. The sample were layered on top of analkaline sucrose gradient and centrifuged immediately at 30000 rev./min for 30 min at 4°C.

Post-replication repairThe DNA was uniformly labelled with [14C]thymidine (0-02/iCi/ml, 56mCi/mmol) for 48 h.

After u.v. radiation, the cells were pulse-labelled with [3H]thymidine (20/iCi/ml, 31 Ci/mmol) for30 min. The medium was replaced by fresh medium containing 10 mM-thymidine and chased from0 to 1 h. After the chase period, the cells were scraped off the surface and lysed on top of an alkalinesucrose gradient for 12 h in the cold. The gradients were then centrifuged for 3 h at 30 000 rev./minat4°C.

RESULTS

The effect of u.v. radiation on cell proliferation

TK cells do not form well-defined colonies. It is therefore not possible to constructa dose-response curve for single cell survival. Instead, by counting cell number eachday after irradiation, the effect of u.v. on proliferation can be established (Fig. 1). Wefound that on the first day after u.v. radiation, there was a dose-dependent decreasein the number of attached cells in all the flasks except the control. TK cells tolerate4-5 Jm~2 of u.v. well enough to resume the control proliferation rate after 2 days. Cellirradiated with 9 Jm~2 recovered to the starting cell number after 3 days, while cul-tures exposed to higher u.v. doses showed no recovery. This was further confirmedby [3H]thymidine pulse-labelling the cultures 1 day and 2 days post-irradiation toassess replicative ability (Fig. 2).

Inhibition of replicative DNA synthesis by u.v. light and the effect metabolic inhibitors

The inhibition of semi-conservative DNA synthesis by u.v. radiation has been welldocumented in both bacteria (Swenson & Setlow, 1966) and mammalian cells(Cleaver, Kaufmann, Kapp & Park, 1983). In general, DNA synthesis decreasedimmediately after u.v. radiation. Fig. 3 shows how the rate of DNA synthesis duringa 10-min pulse label was affected by u.v. This was the method used by Painter (1977)to assess the damaging effect of mutagens on DNA. Our results on CHO cells weresimilar to those reported by Meyn, Hewitt, Thomson & Humphrey (1976) andDoniger (1978). On the other hand, TK cells showed a drop in their rate of DNAsynthesis immediately after 9Jm~2 irradiation (see Oh in Fig. 3), reached theminimum at 30 min, but recovered within 2h, which is comparable to CHO after4-5 Jm~"2, but considerably faster than CHO after the same dose of u.v. The ac-cumulation of [3H]thymidine counts during the first 30 min post-irradiation is shown

216 F. H. Yew and L. M. Chang

Time (days)

Fig. 1. The proliferation of TK cells after u.v. radiation. The cells were trypsinized andsuspended in PBS, then irradiated with 0(A A), 4-5(# • ) , 9(D • ) ,18(^ • ) , 36(A • ) and 45Jm~2(B • ) of u.v. They were transferred intogrowth medium in several flasks and the cell numbers were recorded. After 1, 2 and 3 days,one flask of each dose was trypsinized and the cell numbers were counted. The cell numberplated in the control flask was taken as 100%.

in Fig. 4. In both types of cells u.v. induced a dose-related decline in the incorporationof [3H]thymidine, TK cells being slightly more sensitive (Fig. 4A, B).

The metabolic inhibitor hydroxyrea (HU) inhibits DNA synthesis by blocking theenzyme ribonucleotide reductase (Sjoeberg, Reichard, Graeslund & Ehrenberg,

Fig. 2. The incorporation of [3H]thymidine into TK cells 1 (A A) and 2 ( • 9)days after u.v. radiation. The cells after irradiation in PBS were plated in a 24-well plate.One and two days after irradiation the cells were pulse-labelled for 30 min with [3H]thymi-dine. The number of counts registered on unirradiated control is taken as 1. The error barindicates the standard deviation of four experiments.Fig. 3. The rate of DNA synthesis for TK and CHO cells. The cells were generallylabelled with [HC]thymidine, then replated into 24-well plate in subconfluence for 12 hwithout 14C. u.v. radiation was carried out in PBS followed by a 10-min pulse label with[3H]thymidine in complete medium at 0, 30 min, lh and 2h after u.v. The ratio ofH/14C for control cells was taken as 100%. The standard deviation of four experiments

was ±10%forTK ( ) cells and ±13% for CHO ( ) cells, u.v. dose: 4-5(A),9(A), 18(B), 36«» and 45 Jm"2(«).

% Of controlSurvival (%)

omrid

218 F.H. Yew and L. M. Chang

1977). It is widely used in cell synchronization (Adams & Lindsay, 1967; Skoog &Nordenskjoeld, 1974) and may, with caution, be used to measure unscheduled DNAsynthesis (Smith & Hanawalt, 1976). Our results show that in CHO cells (Fig. 4A),as in many other cell lines, hydroxyurea inhibited over 90% of semi-conservativeDNA replication (Timson, 1975). In TK cells (Fig. 4B), however, without u.v.hydroxyurea (10 mM) suppressed only 80% of total DNA synthesis. The inhibitiondid not increase with increasing dose of u.v.

l-/3-D-Arabinofuranosylcytosine (araC) as araCTP inhibits DNA synthesis eitherby affecting the DNA polymerase or as a chain terminator (Chu & Fisher, 1962;Waquar, Burgoyne & Atkinson, 1971; Monparler, Rossi & Labatan, 1973; Skoog &Nordenskjoeld, 1974); it suppressed DNA replication activities by 95% in CHO(Fig. 4A), but in TK cells (Fig. 4B) the effect of araC is not quite as pronounced.However, treatment with HU and araC combined resulted in more than 90% in-hibition of incorporation in both cell types.

We also tried to determine the inhibitory effects of HU and araC at various con-centrations on CHO and TK cells. The results showed that HU (Fig. 5A) was moreeffective with CHO than TK cells at concentrations from 10"' to 10~4M, the dif-ference was approximately 15% throughout. araC (Fig. 5B) at 10~2M inhibited DNAsynthesis in both cell lines by more than 90%; however, the inhibitory effect was

100

10

68

Fig. 4. The effect of u.v. dose and inhibitors on [3H]thymidine incorporation. C H O ( A )and T K ( B ) cells were preincubated with inhibitors for 1 h, u.v.-irradiated at various dosesin PBS, then incubated in complete medium for 30 min. The cells were then labelled with[3H]thymidine for 30 min, and the acid-precipitable material was counted, u.v. only(V V) ; u.v. + 1 0 " 2 M - H U ( • • ) ; u.v. + 10-4M-araC ( • • ) ; u . v . + 10~2M-

H U + 10-4M-araC ( A — A ) .

DNA replication and repair in Tilapia cells 219

diluted away almost completely at 10~6M, while at this concentration there was stilla suppression of 60% of DNA synthesis in CHO cells. The inhibition of DNAsynthesis by HU and araC is reversible (Collins & Johnson, 1981; Downes & Collins,1982). The effectiveness of the drugs probably depends on several factors, includingthe sizes of internal deoxyribonucleotide pools (Skoog & Nordenskjoeld, 1974), andthe efficiency by which HU and araC are transported into the cell-a pathway affectedgreatly by the medium conditions in certain cells, especially CHO cells (Downes &Collins, 1982; Downes, Johnson & Yew, 1983). We compared the effects of fresh andconditioned medium on TK cells pre- and post-u.v. irradiation and found no dif-ference, which suggests that TK cells do not condition the medium so that it affectsnucleotide transport. In order to eliminate the complication caused by the CHO cell-conditioning effect, fresh medium was always used during experiments. Therefore,the discrepancies in the drug inhibition of DNA synthesis are probably due to thedifferences in the endogenous pool sizes of the two cell lines. A simple method ofmeasuring the relative pool sizes (Strauss, 1981) is to use various specific activities of[3H]thymidine to label the cells; the incorporation rate monitors the dilution of theisotope in the endogenous pool. Fig. 6 shows the incorporation of [3H]thymidine(20^Ci/ml) in the presence of various amounts of cold exogenous thymidine intoDNA of CHO and TK cells. The results demonstrate that in TK cells the isotope waseffectively diluted away at an exogenous thymidine concentration of 10 fiM, while inCHO cells the dilution was much less effective. TK cells should have a smaller

100

1 2 3-log[HU](M)

3 4 5-log[araC] (M)

Fig. 5. The effect of concentration of inhibitors on the rate of DNA synthesis. CHO andTK cells were generally labelled with [14C]thymidine, then replated into 24-well plate insubconfluence for 4—6 h without 14C. Inhibitors were added to the media for 1 h, the cellswere then labelled with pHJthymidine for lOmin, and the acid-precipitable material wascounted, A, HU; B, araC. TK cells (A A) ; CHO cells (A A).

220 F. H. Yew and L. M. Chang

8 7 6 5 4 3-log[thymidine] (g/ml)

Fig. 6. The effects of exogenous thymidine concentration on the rate of [3H]thymidineincorporation. CHO(A • ) and T K ( # • ) cells were labelled with ['4C] thymidinefor 40 h, then transferred and replated in 24-well plate for 24 h in complete mediumwithout radioactive label; the cells were in a subconfluent state when the experiment wasperformed. [3H]thymidine (20/iCi/ml, 30Ci/mmol) and various amounts of coldthymidine were added to the wells, followed by 30-min incubation. The acid-precipitablematerial was counted for radioactivity, and the undiluted 3 H/ H C counts were taken as100%. Each point was the average of eight experiments, the error bars indicate thestandard deviation.

endogenous deoxyribonucleotide triphosphate pool than that of CHO cells if thetransport mechanism and other factors are similar in both cell lines.

The capacity for excision repair and the effect of inhibitors

The capacity of TK cells to perform excision repair was assessed by nucleoidsedimentation. Fig. 7 showed the titration of nucleoid sedimentation against ethidiumbromide. The biphasic curve indicated that the DNA in TK nucleoid is supercoiled(Cook&Brazell, 1975). When irradiated with 10 Jm~2 followed by 30 min incubation,there was a reduction in the rate of nucleoid sedimentation probably associated withthe relaxation of a number of supercoil loops by single-strand breakage. Two hoursafter irradiation, the nucleoid sedimentation is restored to control value. This bi-phasic change of sedimentation in relation to the incubation period after u.v. reflectsthe balance between incision near damaged sites and ligation of completed repair sitesduring excision repair (Cook & Brazell, 1976).

The production of single-strand breaks during excision repair is best resolved usinginhibitors of DNA synthesis (HU and araC) as demonstrated by alkaline sucrosesedimentation (Fig. 8). Following alkaline lysis, DNA from cells u.v. irradiated andincubated with these inhibitors is found to contain single-strand breaks. The presence

DNA replication and repair in Tilapia cells 221

20

Fraction no. Bottom

Fig. 7. The effect of u.v. radiation on the sedimentation of nucleoids. Nucleoids wereprepared by lysing 5 x 104 cells in 0-2ml buffer (2-OM-NaCl, 0 - O I M - E D T A , 0-1%Triton X-100, pH 8-0) on top of a 4-6 ml sucrose gradient (15% to 30% sucrose, 2-0 M-NaCl, 0 - O I M - E D T A , 0-002M- Tris, pH 80) , and centrifuged at 30000 rev./min for40 min at 20 °C. The cells were generally labelled with [3H]thymidine, then irradiated with9Jm"2 in PBS. After irradiation the cells were incubated in complete medium for 30min(D • ) and 90 min ( • • ) . Sedimentation was performed with an unirradiatedcontrol (A A) in the same run. Acid-precipitable material was counted for radio-activity. Inset: relative sedimentation rate of TK nucleoids in the presence of variousamounts of ethidium bromide (EthBr) in the lysis buffer and sucrose gradient. Theconditions for centrifugation were the same as above.

of breaks reduces sedimentation in alkaline sucrose gradients (Burg, Collins & John-son, 1977; Collins e* al. 1977; Collins, 1977). When TK cells were gently layered onan alkaline sucrose gradient and sedimented immediately, the velocity of the DNApeak migration was inversely correlated with the number of single-strand breaks. Theresults shown in Fig. 8 indicate that after 30 min of post-irradiation incubation, theamount of single-strand break accumulation depends on the type and concentrationof the inhibitors. With u.v. alone, there was little change in DNA sedimentation,indicating few single-strand breaks, even when the dose was increased to HOJrrT2.Incubation with HU increased the amount of radioactivity in the top half of thegradient. araC by itself caused a more drastic reduction in sedimentation rate, while

222

10

F. H. Yew and L. M. Chang

^ - 100Ioxc

c2?b

50

10

;' hJ[ E

• • I • i • • i i i

F

0 2T

10

Fraction no.

20B

10

Fraction no.

20 2BT

10

Fraction no.

20B

Fig. 8. Rate of DNA unwinding during sedimentation in alkaline sucrose gradients. TKcells were preincubated with or without inhibitors for 1 h before u.v. radiation. After u.v.challenge the cells were incubated for 30min in the presence or absence of inhibitors,approximately 5 X 104 cells were then lysed on top of an alkaline sucrose gradient andcentrifuged immediately at 30000 rev./min for 30min at 4°C. Fractions were collectedfrom the top and acid-precipitable material was counted for radioactivity, A. Unirradiatedcontrol; B, U.V. l l jm~ 2 ; c, u.v. llOJm"2; D, U.V. l l j m " 2 with 0 - 0 1 M - H U ; E, U.V.11 JrrT2 with l(T4M-araC; F, U.V. 11 Jm"2 with 001 M-HU and 10"4M-araC.

the combined effect of hydroxyurea and araC retained the DNA peak near the top ofthe gradient.

Post-replication (daughter strand) repair

A pulse-chase experiment was performed to assess the ability of TK cells to carryout post-replication repair (Lehmann, 1981). After irradiation 14C-labelled cells werepulsed for 30min with [3H]thymidine and chased for 1 h.

The results show that the behaviour of nascent DNA molecules in control and u.v.-irradiated cells was very similar at 0 h and after 1 h (Fig. 9), indicating that 13-5 Jm~2

of u.v. did not delay the ligation of the nascent species into bulk DNA; but 0-3 mg/ml caffeine delayed the chase rate slightly. The overall rate of incorporation of nascentDNA into the parent was faster than that in CHO cells at the same dose (Stamato,Hinkle, Collins & Waldren, 1981).

DISCUSSION

The DNA repair mechanism in fish cells has not been extensively studied. Of the

DNA replication and repair in Tilapia cells 223

Fraction no. Fraction no. Fraction no.

Fig. 9. Post-replication repair of u.v. irradiated TK cells. Cells were generally labelledwith [14C]thymidine for 48 h. u.v. radiation (13-5 j /m 2 ) was carried out 4—6 h after wash-ing off the [14C]thymidine. The cells were pulse-labelled with [3H]thymidine for 30minimmediately after u.v., then chased for 1 h with/without caffeine (02mg/ml). About5 X 104 cells were lysed on top of an alkaline sucrose gradient at 4°C for 12h. The tubeswere spun at 30000 rev./min for 3 h at 4°C. Fractions were collected from the top, andacid-precipitable material was counted for radioactivity, A. Control, unirradiated, 30-minpulse, no chase, B.- Control, unirradiated, 30-min pulse, 1-h chase without caffeine, c.Control, unirradiated, 30-min pulse, 1-h chase with caffeine, p. u.v. (13-5J/m2)irradiated, pulse, no chase, E. U.V. (13-5 J/m2) irradiated, pulse, 1-h chase without caf-feine, F. u.v. (13-5J/m2) irradiated, pulse, 1-h chase with caffeine. 14C (O O); 3H( • • ) . The arrow indicates A DNA.

repair systems investigated, photoreactivation of the thymine dimer appears to be themost prominent, with a limited amount of excision repair activity (Mano et al. 1980,1982; Regan et al. 1972). Here we have demonstrated that TK cells can undertakeas much excision and post-replication repair as CHO cells.

Recent studies by Elliott & Johnson (1983) showed that excision repair activities weredifferent in cells of different origin; they found that adult mouse kidney epithelial cellsaccumulated more single-strand breaks in the presence of HU and araC than kidney orlungfibroblasts. Gibson-D'Ambrosio,Leong&D'Arnbrosio(1983)alsodemonstratedthat cells cultured from various human foetal organs had different capacities for ex-cision repair. All of the fish cell lines studied in different laboratories were fibroblastsoriginating from connective tissue (ReganeZ al. 1972), erythrophoroma (Mitani, 1983)

224 F. H. Yew and L. M. Chang

or fins (Cook & McGrath, 1967; Regan & Cook, 1967; Mano et al. 1980, 1982).Whether the excision repair activity found in TK cells is related to their epithelialnature and kidney origin remains to be investigated.

The cell survival and proliferation studies showed that the u.v. sensitivity of TKcells is comparable to that of HeLa and CHO cells (Downes et al. 1979; Collins,Downes & Johnson, 1980). The recovery of DNA synthesis after u.v. was, however,faster in TK than in CHO cells. Single-strand scission due to u.v. endonucleasecleavage was not apparent after u.v. irradiation except in the presence of inhibitors.As in other studies (e.g. see Collins & Johnson, 1981), HU and araC were found todiffer in their capacity to accumulate single-strand breaks, while their combinationresulted in the greatest accumulation of breaks.

We found that HU and araC were less effective in suppressing replicative DNAsynthesis in TK cells than in HeLa and CHO cells. Cell survival depends, to a certainextent, on the supply of DNA precursors (Johnson & Collins, 1978; Yew & Johnson,1979). The efficiency of the TK repair system in the presence of inhibitors might beexplained if the internal pool of deoxynucleotides was considerably larger in TK thanin CHO cells. However, the isotope dilution experiment shown in Fig. 6 revealed thatthe extent of [3H]thymidine incorporation was much more readily influenced bychanges in specific activity in TK cells than in CHO cells. Thus we cannot claim thatTK cells have larger DNA precursor pools than CHO cells. One possible explanationis that TK cells might have a very fast turnover rate of DNA precursors. This is atpresent under investigation.

The work was supported by the National Science Council of the Republic of China. The authorsthank Dr S. N. Chen for supplying TK cell lines and other experimental essentials, Miss S. L.Shyng and Mr P. Y. Yeh for technical assistance. Special acknowledgement is made to Dr R. T.Johnson, who has read the manuscript critically and made many valuable suggestions.

REFERENCES

ADAMS, R. L. P. & LINDSAY, J. G. (1967). Hydroxyurea, reversal of inhibition and use as a cellsynchronizing agent. J. biol. Chent. 242, 1314-1317.

BURG, K., COLLINS, A. R. S. & JOHNSON, R. T. (1977). Effects of ultraviolet light on synchronizedChinese hamster ovary cells: potentiation by hydroxyurea. jf. Cell Sci. 28, 29—48.

CHEN, S. N., UENO, Y., WEN, S. C. & K O U , G . H. (1983). Establishment of a cell line from kidneyof Tilapia. Bull. Eur. Ass. Fish Path. 3, 1-4.

CHU, M. Y. & FISHER, G. A. (1962). A proposed mechanism of action of l-/3-D-arabinofurano-sylcytosine as an inhibitor of the growth of leukemic cells. Biochem. Pharmac. 11, 423-430.

CLEAVER, J. E. (1965). Investigation into the effect of ultraviolet light on the rate of deoxyribo-nucleic acid synthesis in mammalian cells. Biochim. biophys. Ada 108, 42-52.

CLEAVER, J. E., KAUFMANN, W. K., KAPP, L. N. & PARK, S. D. (1983). Replicon size andexcision repair as factors in the inhibition and recovery of DNA synthesis from ultravioletdamage. Biochim. biophys. acta 739, 207-215.

COLLINS, A. R. S. (1977). DNA damage in ultraviolet-irradiated HeLa and CHO-K1 cellsexamined by alkaline lysis and hydroxyapatite chromatography. Biochim. biophys. Acta 478,461-473.

COLLINS, A. R. S., DOWNES, C. S. & JOHNSON, R. T. (1980). Cell cycle-related variations inUV damage and repair capacity in Chinese hamster (CHO-K1) cells. J. cell. Phvsiol. 103,179-197.

DNA replication and repair in Tilapia cells 225

COLLINS, A. R. S. &JOHNSON, R. T. (1981). Use of metabolic inhibitors in repair studies. inDNARepair: A Laboratory Manual ofResearch Procedures, vol. 1, partB (ed. E. C. Friedberg&P. C.Hanawalt), pp. 341-360. New York, Basel: Marcer Dekker, Inc.

COLLINS, A. R. S., SCHOR, S. L. & JOHNSON, R. T. (1977). The inhibition of repair in UVirradiated human cells. Mutat. Res. 42, 413-432.

COOK, J. S. & MCGRATH, J. R. (1967). Photoreactivating-enzyme activity in metazoa. Proc. natn.Acad. Sci. U.SA. 18, 2274-2279.

COOK, P. R. & BRAZELL, I. A. (1975). Supercoils in human DNA. J . Cell Sci. 19, 261-279.COOK, P. R. & BRAZELL, I. A. (1976). Detection and repair of single-strand breaks in nuclear DNA.

Nature, Land. 263, 679-682.DONIGER, J. (1978). DNA replication in ultraviolet light irradiated Chinese hamster cells: The

nature of replicon initiation and postreplication repair. J. molec. Biol. 120, 433-446.DOWNES, C. S. & COLLINS, A. R. S. (1982). Effects of DNA replication inhibitors on UV excision

repair in synchonized human cells. Nucl. Acids Res. 10, 5357-5368.DOWNES, C. S., COLLINS, A. R. S. & JOHNSON, R. T. (1979). DNA damage in synchronized HeLa

cells irradiated with ultraviolet. Biophys. jf. 25, 129-150.DOWNES, C. S., JOHNSON, R. T. & YEW, F. F. (1983). Effects of conditioned medium on nucleo-

side uptake, cell cycle progression and apparent DNA repair. J. Cell Sci. 59, 145-158.ELLIOTT, G. C. & JOHNSON, R. T. (1983). DNA repair in mouse embryo fibroblasts. I. Decline

in ultraviolet-induced incision rate accompanies cell transformation, jf. Cell Sci. 60, 267-288.GIBSON-D'AMBROSIO, R. E., LEONG, Y. & D'AMBROSIO, S. M. (1983). DNA repair following

ultraviolet and yV-ethyl-JV-nitrosourea treatment of cells cultured from human fetal brain, intes-tine, kidney, liver and skin. Cancer Res. 43, 5846-5850.

HART, R. W., SETLOW, R. B. & WOODHEAD, A. D. (1977). Evidence that pyrimidine dimers inDNA can give rise to tumors. Proc. natn. Acad. Sci. U.SA. 74, 5574-5578.

HYODO-TAGUCHI, Y. & EGAMI, N. (1969). Development of intestinal injury and recovery atdifferent temperatures in fish. In Comparative Cellular and Species Radiosensitivity (ed. V. P.Bond & T. Sugahara), pp. 244-254. Tokyo: Igaku Shoin.

JOHNSON, R. T. & COLLINS, A. R. S. (1978). Reversal of the changes in DNA and chromosomestructure which follow the inhibition of UV-induced repair in human cells. Biochem. biophys. Res.Cornmun. 80, 361-369.

LEHMANN, A. R. (1981). Measurement of postreplication repair in mammalian cells. In DNARepair: A Laboratory Manual of Research Procedures, vol. 1, part B (ed. E. C. Friedberg & P. C.Hanawalt), pp. 471-485, New York, Basel: Marcer Dekker, Inc.

MANO, Y., KATOR, K. & EGAMI, N. (1982). Photoreactivation and excision repair of thyminedimers in ultraviolet-irradiated cultured fish cells. Radiat. Res. 90, 501-508.

MANO, Y., MITANI, H., ETOH, H. & EGAMI, N. (1980). Survival and photoreactivability ofultraviolet-irradiated cultured fish cells (CAF-MM1). Radiat. Res. 84, 514-522.

MEYN, R., HEWITT, R. R., THOMSON, L. & HUMPHREY, R. M. (1976). Effects of ultravioletirradiation on the rate and sequence of DNA replication in synchronized Chinese hamster cells.Biophys. J. 16,517-524.

MITANI, H. (1983). Lethal and mutagenic effects of radiation and chemicals on cultured fish cellsderived from the erythrophoroma of goldfish (Carassius auratus). Mutat. Res. 107, 279-288.

MITANI, H. & EGAMI, N. (1982). Rejoining of DNA strand breaks after gamma-irradiation incultured fish cells, CAF-MM1. Int. J. Radiat. Res 89, 85-90.

MITANI, H., ETOH, H. & EGAMI, N. (1982). Resistance of a cultured fish cell line (CAF-MM1) tor-irradiation. Radiat. Res. 44, 748-757.

MONTPARLER, R. L., Rossi, M. & LABITAN, A. (1973). Partial purification and properties of twoforms of deoxyribonucleic acid polymerase from calf thymus. jf. biol. Chem. 248, 285—293.

PAINTER, R. B. (1977). Rapid test to detect agents that damage human DNA. Nature, Land. 265,650-651.

PURDOM, C. E. & WOODHEAD, D. S. (1973). Radiation damages in fish. In Genetics andMutagenesis of Fish (ed. J. H. Schroender), pp. 67-73. Berlin: Springer.

REGAN, J. D., CARRIER, VV. L., SAMET, C. & OLLA, B. L. (1972). Photoreactivation in two closelyrelated marine fishes having different longevities. Mech. Aging Dev. 18, 59-66.

REGAN, J. D. & COOK, ?.(1967). Photoreactivation in an established vertebrate cell line. Proc. natn.Acad. Sci. U.SA. 58, 2274-2279.

226 F. H. Yew and L. M. Chang

SJOEBERG, B. M., REICHARD, P., GRAESLUND, A. & EHRENBERG, A. (1977). Nature of the freeradical in ribonucleotide reductase from Escherichia coli.jf. biol. Chem. 252, 536—541.

SKOOG, L. & NORDENSKJOELD, B. (1974). Effect of hydroxyurea and l-/3-D-arabinofuranosylcyto-sine on deoxyribonucleotide pools in mouse embryo cells. Eur.J. Biochem. 19, 81-89.

SMITH, C. A. & HANAWALT, P. C. (1976). Repair replication in human cells. Simplified determina-tion utilizing hydroxyurea. Biochim. biophys. Ada 432, 336-346.

STAMATO, T. D., HINKLE, L., COLLINS, A. R. S. & WALDREN, C. A. (1981). Chinese hamsterovary mutant UV-1 is hypomutable and defective in a postreplication recovery process. Somat.Cell Genet. 7, 307-320.

STRAUSS, B. S. (1981). Use of benzoylated naphthoylated DEAE cellulose. In DNA Repair: ALaboratory Manual of Research Procedures, vol. 1, part B (ed. E. C. Friedberg & P. C.Hanawalt), pp. 319-339. New York, Basel: Marcer Dekker, Inc.

SWENSON, P. A. & SETLOW, R. B. (1966). Effects of ultraviolet radiation on macromolecularsynthesis in Escherichia coli.J. molec. Biol. 15, 201-219.

TIMSON, J. (1975). Hydroxyurea. Mutat. Res. 32, 115-132.WAQUAR, M. A., BURGOYNE, L. A. & ATKINSON, M. R. (1971). Deoxyribonucleic acid synthesis

in mammalian nuclei. Incorporation of deoxyribonucleotides and chain-terminating nucleotideanalogues. Biochem. jf. 121, 803-809.

YEW, F. F. H. & JOHNSON, R. T. (1979). Ultraviolet-induced DNA excision repair in human Band T lymphocytes. II. Effect of inhibitors and DNA precusors. Biochim. biophys. Ada 562,240-251.

(Received 9 April 1984 -Accepted, in revised form, 3 July 1984)