- Email: [email protected]
Persistent structural change in replicating hepatic DNA isolated from diethylnitrosamine-treated rats
263
Cancer Letters, 35 (1987) 263269 Elsevier Scientific Publishers Ireland Ltd.
PERSISTENT STRUCTURAL CHANGE IN REPLICATING DNA ISOLATED FROM DIETHYLNITROSAMINE-TREATED
BERNARD
W. STEWART*
and CAROLYN
HEPATIC RATS
HRISTOFORIDIS
children’s Leukaemia and Cancer Research Unit, Prince of Wales Children’s Hospital, Randwick 2031, New South Wales and School of Pathology, University of New South Wales, Kensington 2033, New South Wales (Australia) (Received 6 November 1986) (Revised version received 6 February (Accepted 9 February 1987)
1987)
SUMMARY
Structural analysis, by benzoylated 0-( diethylaminoethyl) (DEAE)-cellulose chromatography, was made of DNA from the livers of rats receiving 100 mg/kg diethylnitrosamine and subsequently subjected to partial hepatectomy. Under these conditions, different DNA labelling procedures permit damage to be associated with pre-existing or newly synthesised DNA. Persistent single stranded regions could be detected in DNA isolated more than 3 days after carcinogen treatment only if the animals were subjected to hepatectomy. This damage was attributable to lesions impeding DNA replication. Induction of proliferative activity up to at least 14 days after nitrosamine treatment made manifest DNA damage, the extent of which was not decreased as the interval between carcinogen treatment and surgery was increased.
INTRODUCTION
Hepatic tumorigenesis occurs in rats after single dose treatment with certain N-nitroso compounds if the animals are subject to partial hepatectomy [2]. Hepatic metabolism of dialkyl nitroso compounds results in the alkylation of purine and pyrimidine bases, as well as of phosphate groups, in DNA [ll]. Persistence of such alkylated bases in vivo has been intensively
*To whom correspondence search Unit.
should be addressed at the Children’s
0304.3835/87/$03.50 c;1987 Elsevier Scientific Published and Printed in Ireland
Publishers
Leukaemia
Ireland Ltd.
and Cancer Re-
264
studied, as have the processes by which they are eliminated [12]. The consequences of DNA replication in carcinogen-exposed cells have been less extensively studied. DNA may be fractionated according to its secondary structure by stepwise elution from benzoylated DEAE -(BD-) cellulose, double stranded DNA being eluted in 1.0 M NaCl whilst DNA containing single stranded regions is recovered by addition of caffeine to the salt solution [16]. Hepatic DNA isolated following administration of simple dialkyl nitrosamines to rats contains transient single stranded regions associated with the excision repair of alkylated bases [13,15]. In addition, the generation and disappearance of single stranded regions in pulse-chase labelled DNA may be associated with migration of the replicating fork [7,14]. This observation encouraged reappraisal of the persistence of nitrosamine-induced structural change in hepatic DNA with particular regard to the effect of cell division. Data included in the present paper indicate that when rats which have received a dose of 100 mg/kg diethylnitrosamine (DEN) are subjected to partial hepatectomy, single stranded regions are generated in hepatic DNA. Such damage is persistent and may be generated even if surgery is undertaken when repair-associated single stranded regions can no longer be detected. MATERIALS
AND METHODS
Female Wistar rats (160 g), maintained as previously described [ 131, were treated with DEN (in saline solution, administered by intraperitoneal injection, 100 mg/kg body wt.) and later subjected to partial hepatectomy. Hepatic DNA was radiolabelled in vivo either during neonatal life, and hence several weeks before administration of the carcinogen (‘parental strand labelling’) or following partial hepatectomy (‘daughter strand labelling’). For neonatal labelling, pups were given a daily intraperitoneal injection of [3H]thymidine (25 PCi) for the first 6 days after birth. Labelling after hepatectomy was accomplished by administration of a single intraperitoneal injection of [3H]thymidine (100 PCi). The kinetic relationships between DEN treatment, partial hepatectomy, radiolabelling and DNA isolation are variously indicated in the figures. Procedures for isolation of DNA and stepwise elution from BD-cellulose have been previously published [ 131. RESULTS
Initial experiments involved animals labelled by administration of [ 3H] thymidine during the first 6 days of life. These animals were treated as adults with DEN, killed up to 6 days later and sheared preparations of isolated hepatic DNA subjected to BD-cellulose chromatography. The amount of DNA containing single stranded regions was markedly dependent on the interval between carcinogen treatment and death. After reaching a maximum within 12 h of injection of DEN, the proportion of DNA containing
265
single stranded regions fell rapidly, approaching control levels within 6 days. However, when rats given DEN were subject to partial hepatectomy 12 h after injection of the nitrosamine, structural analysis of DNA revealed a different pattern. The level of damage detected in the respective DNA preparations fluctuated but did not decrease with time (Fig. 1). The implication that such relatively persistent damage was a consequence of replication was examined by thymidine-labelling newly synthesised DNA. As anticipated [7], in hepatic DNA radiolabelled during regenerative hyperplasia, the amount of radioactive DNA exhibiting single stranded character falls rapidly within 1 h of isotope administration. The same experiment performed using animals treated with DEN 12 h before surgery revealed an effect as a consequence of carcinogen treatment. In both control and carcinogen-treated rats the amount of radioactivity recovered in the caffeine-eluted fraction decreased in similar fashion as the labelled precursor was incorporated into mature DNA. However, DNA isolated 30 min after administration of [3H] thymidine to rats subjected to hepatectomy 3 days
3H
TIME
, I
DEN ,
t +
(DAYS)
Fig. 1. Effect of partial hepatectomy on the amount of DNA containing single stranded regions detected in preparations isolated from the livers of rats injected with 100 mg/kg body wt. DEN. Twelve hours after carcinogen treatment, half of the animals were subject to partial hepatectomy (0) while the remainder were given no other treatment (0). Animals were pre-labelled with [3H] thymidine as described in the Materials and Methods section. DNA was isolated from 12 h to 6 days after carcinogen treatment, this interval being denoted as ‘Time’ on the figure and indicated by ‘t’ on the schematic treatment protocols (PH = partial hepatectomy). In this and other figures, increase in the amount of DNA eluted with caffeine from BD-cellulose is expressed relative to the proportion in adult rat liver (0.067 + 0.006) as previously described [13], each point being the mean (with the range indicated) of at least 2 determinations based on identically treated animals from which individual preparations of DNA were obtained and subjected to separate chromatographic analysis.
266
earlier contained a significantly larger caffeine-eluted fraction if the animals were treated with DEN before surgery (Fig. 2). This suggested that DEN-induced damage might be detected using an assay involving short term-labelling of DNA. Results described thus far indicate that DNA isolated from the livers of DEN-treated rats could be structurally distinguished from control preparations if respective animals were subjected to partial hepatectomy. The slower ‘decay’ in the caffeine-eluted fraction caused by DEN treatment (Fig. 2) was detected when proliferative activity was induced only 12 h after administration of the carcinogen. Nevertheless, DEN-induced effects on single stranded character could be detected, at least under some conditions, using DNA labelled in the parental strand (i.e. before carcinogen treatment, Fig. 1) or the daughter strand (Fig. 2). The extent to which carcinogen-induced structural defects, expressed as a consequence of partial hepatectomy, could be regarded as persistent was examined by varying the period between DEN treatment and surgery. These experiments involved DNA labelling by each of the two protocols employed earlier. Thus, in one series of experiments DNA was labelled prior to carcinogen treatment and in the other DNA was
zF 300 2 E :: 5 iii g 200 iii k 5 z : 2 100 8 + 2 Y !z
Or
I 15 TIME
I 30
1 45
I 60
(MINUTES)
Fig. 2. Inhibition of the decrease in the proportion of caffeine-eluted DNA consequent upon incorporation of [ 3H] thymidine into replicating DNA. Rats were treated either with saline ( 0) or 100 mg/kg DEN (0) and subjected to partial’hepatectomy 12 h later. DNA was labelled 3 days after the saline or carcinogen treatment and animals were killed 1560 min later as shown on the figure. Each point is the mean of at least 2 experiments, expressed as described in the legend to Fig. 1, the range of f 1 S.D. being shown for data involving 3 or more determinations.
267
3
s 8 L
k’ a L
0
7
14
TIME (DAYS)
Fig. 3. The pattern of increase in the amount of hepatic DNA containing single stranded regions in DNA isolated from rats receiving 100mg/kg body wt. DEN and subjected to partial hepatectomy. For the purpose of structural analysis, DNA was labelled either prior to carcinogen treatment (0) or during restorative hyperplasia ( 0), the interval between DEN treatment and surgery being varied as shown by ‘t’ on the figure.
labelled after hepatectomy. A similar pattern of increased binding the BDcellulose was detected regardless of the labelling procedure for periods up to 5 days between carcinogen treatment and hepatectomy. However, incorporation of isotope into newly synthesised DNA provided a slightly more sensitive indication of structural abnormality (Fig. 3). When this latter protocol was used, the same figure indicates that there was virtually no decrease in the proportion of single stranded DNA as the interval between carcinogen treatment and partial hepatectomy was extended to 14 days. DISCUSSION
It would appear that on the basis of BD-cellulose chromatography, 2 types of structural damage can be distinguished in DEN-treated rats. Firstly, single stranded regions may be detected in DNA isolated within 24 h of carcinogen treatment of animals (not subjected to surgery) and these regions are characterized by their rapid loss (Fig. 1). A second class of BD-cellulose binding lesion(s) are generated as a consequence of DNA replication in livers of DEN-treated rats and are not readily lost. Previous work [13, 151 suggests that the former are a consequence of excision repair, being detected when the final stages of DNA repair (polymerization and ligation) are rate limiting and hence repair intermediates accumulate. The latter are most immediately associated with DNA replication (Fig. 2), the relative persistence of single stranded regions being attributable to lesions in template strand
268
consequent upon DEN treatment. Caffeine-gradient elution of the present DNA samples indicated short single stranded regions as distinct from the extensive degradation associated with hepatic necrosis [8]. Carcinogen adducts are known to inhibit DNA elongation [18]. In studies using dimethylnitrosamine, Sarma and his colleagues [lo] have reported the greater susceptibility of 06-methylguanine, in replicated DNA to digestion by Sl nuclease, than 7-methylguanine. Discontinuities in newly synthesised DNA because of alkylated bases in the template strand would appear an even greater likelihood for DEN, since compared to its methyl analogue, the products of DNA ethylation are less rapidly eliminated from rat liver DNA [3]. Thus altered sedimentation in alkaline sucrose of rat liver DNA isolated up to 56 days after DEN treatment [4] is certainly compatible with replication-induced structural damage being generated by hepatectomy 14 days (Fig. 3) after exposure to the nitrosamine. The similar response pattern obtained using parental or daughter strandlabelled DNA is consistent with BD-cellulose binding lesions being present in hybrid molecules such that they are detected by radioactivity contained in either strand. Such single stranded regions would be substrates for ‘postreplication repair’ [5]. The detection of single stranded regions in DNA isolated 6 days after hepatectomy (Fig. l), suggests that, if post-replication repair does occur, it does not result in the elimination of all replicationassociated structural damage. Little is known about how specific DNA lesions are translated into aberrations at the chromosomal level [ 1,9]. However, structural damage may be the genesis of micronuclei described by Den Engelse and his colleagues [17] in rats subjected to partial hepatectomy l-56 days after administration of DEN. This pre-clastogenic damage was associated with 0-alkylation. Although carcinogen adducts may thus be related to the generally accepted critical role of DNA replication in initiation, exploitation of rat liver models involving surgically induced hyperplasia in DEN-treated rats has been principally at the cellular level (for review, see ref. 6). If structural damage to DNA, as described here and evidenced by others [17], could be related to specific sequences, it may be possible to associate such lesions with recognized enzymic markers of pre-malignant cell populations.
ACKNOWLEDGEMENT
This research was made possible Medical Research Council.
by a grant from the National
Health and
REFERENCES 1 Bender, M.A. (1980) Relationship of DNA lesions and their repair to chromosomal tion production. Basic Life Sci., 15, 245-265.
aberra
269 2 Craddock, V.M. (1976) Cell proliferation and experimental liver cancer. In: Liver Cell Cancer, pp. 152-201. Editors: H.M. Cameron, D.A. Linsell and G.P. Warwick. Elsevier, Amsterdam. 3 Den Engelse, L., Floot, B.G.J., De Brij, R.J. and Tates, A.D. (1983) The induction of chromosomal damage in rat hepatocytes and lymphocytes. II. Alkylation damage and repair of rat liver DNA after diethylnitrosamine, dimethylnitrosamine and ethyl methanesulfonate in relation to clastogenic effects. Mutat. Res., 107, 153-156. 4 Floot, B.G.J., Philippus, E.J., Hart, A.A.M. and Den Engelse, L. (1979) Persistence and accumulation of (potential) single strand breaks in liver DNA of rats treated with diethylnitrosamine or dimethylnitrosamine: correlation with hepatocarcinogenicity. Chem.-Biol. Interact., 25, 229242. 5 Fujiwara, Y. (1975) Postreplication repair of alkylation damage to DNA of mammalian cells in culture. Cancer Res., 35, 27892789. 6 Goldsworthy, T.L., Hanigan, M.H. and Pitot, H.C. (1986) Models of hepatocarcinogenesis in the rat - contrasts and comparisons. CRC Crit. Rev. Toxicol., 17, 61-89. 7 Haber, M., Kavallaris, M. and Stewart, B.W. (1986) Two-stage incorporation of thymidine triphosphate into mammalian DNA as indicated by chromatography on benzoylated DEAEcellulose. J. Chromatogr., 382, 127-134. 8 Haber, M. and Stewart, B.W. (1985) Patterns of structural change in DNA during tissue necrosis indicated by benxoylated DEAE-cellulose chromatography. Chem.-Biol. Interact., 53, 247-255. 9 Kihlman, B.A. (1971) Molecular mechanisms of chromosome breakage and rejoining. Adv. Cell Mol. Biol., 1, 59107. 10 Ledda, G.M., Columbano, A., Rao, P.M., Rajalakshmi, S. and Sarma, D.S.R. (1980) In uivo replication of carcinogen-modified rat liver DNA: increased susceptibility of O-methylguanine compared to 7-methylguanine in replicated DNA to Sl nuclease. Biochem. Biophys. Res. Commun., 95, 816821. 11 Margison, G.P. and O’Connor, P.J. (1979) Nucleic acid modification by N-nitrose compounds. In: Chemical Carcinogens and DNA, pp. 111-159. Editor: P.L. Glover. CRC Press, Baltimore. 12 Preussmann, R. and Stewart, B.W. (1984) N-nitroso Carcinogens. In: Chemical Carcinogens, 2nd Edition, pp. 64%828. Editor: C.E. Searle. American Chemical Society, Washington, D.C. 13 Stewart, B.W. (1981) Generation and persistence of carcinogen-induced repair intermediates in rat liver DNA in uiuo. Cancer Res., 41, 32383243. 14 Stewart, B.W. and Brian, M.J. (1979) Evidence of carcinogen-induced replication of partially-repaired DNA in target cells during nitrosamine carcinogenesis. Eur. J. Cancer, 15, 251-256. 15 Stewart, B.W., Hristoforidis, C. and Haber, M. (1985) Dose-dependent persistence of alkylation-induced single stranded regions in rat liver DNA in uiuo. Cancer Letters, 28, 27-33. 16 Stewart, B.W., Huang, P.H.T. and Brian, M.J. (1979) Structural defects in rat liver deoxyribonucleic acid. Endogenous single-stranded regions in comparison with damage induced in vivo by a carcinogen. Biochem. J., 179, 341-352. 17 Tates, A.D., Neuteboom, I., Rotteveel, A.H.M., de Vogel, N., Menkveld, G.J. and Den Engelse, L. (1986) Persistence of preclastogenic damage in hepatocytes of rats exposed to ethylnitrosourea, diethylnitrosamine, dimethylnitrosamine and methyl methanesulphonate. Correlation with DNA 0.alkylation. 18 Zurlo, J., Eustice, D.C., Mignano, J.E., Poirior, M.C. and Yager, J.D. (1986) Effects of carcinogen treatment on rat liver DNA synthesis in viva and on nascent DNA synthesis and elongation in cultured hepatocytes. Mutat. Res., 161, 143154.
Cancer Letters, 35 (1987) 263269 Elsevier Scientific Publishers Ireland Ltd.
PERSISTENT STRUCTURAL CHANGE IN REPLICATING DNA ISOLATED FROM DIETHYLNITROSAMINE-TREATED
BERNARD
W. STEWART*
and CAROLYN
HEPATIC RATS
HRISTOFORIDIS
children’s Leukaemia and Cancer Research Unit, Prince of Wales Children’s Hospital, Randwick 2031, New South Wales and School of Pathology, University of New South Wales, Kensington 2033, New South Wales (Australia) (Received 6 November 1986) (Revised version received 6 February (Accepted 9 February 1987)
1987)
SUMMARY
Structural analysis, by benzoylated 0-( diethylaminoethyl) (DEAE)-cellulose chromatography, was made of DNA from the livers of rats receiving 100 mg/kg diethylnitrosamine and subsequently subjected to partial hepatectomy. Under these conditions, different DNA labelling procedures permit damage to be associated with pre-existing or newly synthesised DNA. Persistent single stranded regions could be detected in DNA isolated more than 3 days after carcinogen treatment only if the animals were subjected to hepatectomy. This damage was attributable to lesions impeding DNA replication. Induction of proliferative activity up to at least 14 days after nitrosamine treatment made manifest DNA damage, the extent of which was not decreased as the interval between carcinogen treatment and surgery was increased.
INTRODUCTION
Hepatic tumorigenesis occurs in rats after single dose treatment with certain N-nitroso compounds if the animals are subject to partial hepatectomy [2]. Hepatic metabolism of dialkyl nitroso compounds results in the alkylation of purine and pyrimidine bases, as well as of phosphate groups, in DNA [ll]. Persistence of such alkylated bases in vivo has been intensively
*To whom correspondence search Unit.
should be addressed at the Children’s
0304.3835/87/$03.50 c;1987 Elsevier Scientific Published and Printed in Ireland
Publishers
Leukaemia
Ireland Ltd.
and Cancer Re-
264
studied, as have the processes by which they are eliminated [12]. The consequences of DNA replication in carcinogen-exposed cells have been less extensively studied. DNA may be fractionated according to its secondary structure by stepwise elution from benzoylated DEAE -(BD-) cellulose, double stranded DNA being eluted in 1.0 M NaCl whilst DNA containing single stranded regions is recovered by addition of caffeine to the salt solution [16]. Hepatic DNA isolated following administration of simple dialkyl nitrosamines to rats contains transient single stranded regions associated with the excision repair of alkylated bases [13,15]. In addition, the generation and disappearance of single stranded regions in pulse-chase labelled DNA may be associated with migration of the replicating fork [7,14]. This observation encouraged reappraisal of the persistence of nitrosamine-induced structural change in hepatic DNA with particular regard to the effect of cell division. Data included in the present paper indicate that when rats which have received a dose of 100 mg/kg diethylnitrosamine (DEN) are subjected to partial hepatectomy, single stranded regions are generated in hepatic DNA. Such damage is persistent and may be generated even if surgery is undertaken when repair-associated single stranded regions can no longer be detected. MATERIALS
AND METHODS
Female Wistar rats (160 g), maintained as previously described [ 131, were treated with DEN (in saline solution, administered by intraperitoneal injection, 100 mg/kg body wt.) and later subjected to partial hepatectomy. Hepatic DNA was radiolabelled in vivo either during neonatal life, and hence several weeks before administration of the carcinogen (‘parental strand labelling’) or following partial hepatectomy (‘daughter strand labelling’). For neonatal labelling, pups were given a daily intraperitoneal injection of [3H]thymidine (25 PCi) for the first 6 days after birth. Labelling after hepatectomy was accomplished by administration of a single intraperitoneal injection of [3H]thymidine (100 PCi). The kinetic relationships between DEN treatment, partial hepatectomy, radiolabelling and DNA isolation are variously indicated in the figures. Procedures for isolation of DNA and stepwise elution from BD-cellulose have been previously published [ 131. RESULTS
Initial experiments involved animals labelled by administration of [ 3H] thymidine during the first 6 days of life. These animals were treated as adults with DEN, killed up to 6 days later and sheared preparations of isolated hepatic DNA subjected to BD-cellulose chromatography. The amount of DNA containing single stranded regions was markedly dependent on the interval between carcinogen treatment and death. After reaching a maximum within 12 h of injection of DEN, the proportion of DNA containing
265
single stranded regions fell rapidly, approaching control levels within 6 days. However, when rats given DEN were subject to partial hepatectomy 12 h after injection of the nitrosamine, structural analysis of DNA revealed a different pattern. The level of damage detected in the respective DNA preparations fluctuated but did not decrease with time (Fig. 1). The implication that such relatively persistent damage was a consequence of replication was examined by thymidine-labelling newly synthesised DNA. As anticipated [7], in hepatic DNA radiolabelled during regenerative hyperplasia, the amount of radioactive DNA exhibiting single stranded character falls rapidly within 1 h of isotope administration. The same experiment performed using animals treated with DEN 12 h before surgery revealed an effect as a consequence of carcinogen treatment. In both control and carcinogen-treated rats the amount of radioactivity recovered in the caffeine-eluted fraction decreased in similar fashion as the labelled precursor was incorporated into mature DNA. However, DNA isolated 30 min after administration of [3H] thymidine to rats subjected to hepatectomy 3 days
3H
TIME
, I
DEN ,
t +
(DAYS)
Fig. 1. Effect of partial hepatectomy on the amount of DNA containing single stranded regions detected in preparations isolated from the livers of rats injected with 100 mg/kg body wt. DEN. Twelve hours after carcinogen treatment, half of the animals were subject to partial hepatectomy (0) while the remainder were given no other treatment (0). Animals were pre-labelled with [3H] thymidine as described in the Materials and Methods section. DNA was isolated from 12 h to 6 days after carcinogen treatment, this interval being denoted as ‘Time’ on the figure and indicated by ‘t’ on the schematic treatment protocols (PH = partial hepatectomy). In this and other figures, increase in the amount of DNA eluted with caffeine from BD-cellulose is expressed relative to the proportion in adult rat liver (0.067 + 0.006) as previously described [13], each point being the mean (with the range indicated) of at least 2 determinations based on identically treated animals from which individual preparations of DNA were obtained and subjected to separate chromatographic analysis.
266
earlier contained a significantly larger caffeine-eluted fraction if the animals were treated with DEN before surgery (Fig. 2). This suggested that DEN-induced damage might be detected using an assay involving short term-labelling of DNA. Results described thus far indicate that DNA isolated from the livers of DEN-treated rats could be structurally distinguished from control preparations if respective animals were subjected to partial hepatectomy. The slower ‘decay’ in the caffeine-eluted fraction caused by DEN treatment (Fig. 2) was detected when proliferative activity was induced only 12 h after administration of the carcinogen. Nevertheless, DEN-induced effects on single stranded character could be detected, at least under some conditions, using DNA labelled in the parental strand (i.e. before carcinogen treatment, Fig. 1) or the daughter strand (Fig. 2). The extent to which carcinogen-induced structural defects, expressed as a consequence of partial hepatectomy, could be regarded as persistent was examined by varying the period between DEN treatment and surgery. These experiments involved DNA labelling by each of the two protocols employed earlier. Thus, in one series of experiments DNA was labelled prior to carcinogen treatment and in the other DNA was
zF 300 2 E :: 5 iii g 200 iii k 5 z : 2 100 8 + 2 Y !z
Or
I 15 TIME
I 30
1 45
I 60
(MINUTES)
Fig. 2. Inhibition of the decrease in the proportion of caffeine-eluted DNA consequent upon incorporation of [ 3H] thymidine into replicating DNA. Rats were treated either with saline ( 0) or 100 mg/kg DEN (0) and subjected to partial’hepatectomy 12 h later. DNA was labelled 3 days after the saline or carcinogen treatment and animals were killed 1560 min later as shown on the figure. Each point is the mean of at least 2 experiments, expressed as described in the legend to Fig. 1, the range of f 1 S.D. being shown for data involving 3 or more determinations.
267
3
s 8 L
k’ a L
0
7
14
TIME (DAYS)
Fig. 3. The pattern of increase in the amount of hepatic DNA containing single stranded regions in DNA isolated from rats receiving 100mg/kg body wt. DEN and subjected to partial hepatectomy. For the purpose of structural analysis, DNA was labelled either prior to carcinogen treatment (0) or during restorative hyperplasia ( 0), the interval between DEN treatment and surgery being varied as shown by ‘t’ on the figure.
labelled after hepatectomy. A similar pattern of increased binding the BDcellulose was detected regardless of the labelling procedure for periods up to 5 days between carcinogen treatment and hepatectomy. However, incorporation of isotope into newly synthesised DNA provided a slightly more sensitive indication of structural abnormality (Fig. 3). When this latter protocol was used, the same figure indicates that there was virtually no decrease in the proportion of single stranded DNA as the interval between carcinogen treatment and partial hepatectomy was extended to 14 days. DISCUSSION
It would appear that on the basis of BD-cellulose chromatography, 2 types of structural damage can be distinguished in DEN-treated rats. Firstly, single stranded regions may be detected in DNA isolated within 24 h of carcinogen treatment of animals (not subjected to surgery) and these regions are characterized by their rapid loss (Fig. 1). A second class of BD-cellulose binding lesion(s) are generated as a consequence of DNA replication in livers of DEN-treated rats and are not readily lost. Previous work [13, 151 suggests that the former are a consequence of excision repair, being detected when the final stages of DNA repair (polymerization and ligation) are rate limiting and hence repair intermediates accumulate. The latter are most immediately associated with DNA replication (Fig. 2), the relative persistence of single stranded regions being attributable to lesions in template strand
268
consequent upon DEN treatment. Caffeine-gradient elution of the present DNA samples indicated short single stranded regions as distinct from the extensive degradation associated with hepatic necrosis [8]. Carcinogen adducts are known to inhibit DNA elongation [18]. In studies using dimethylnitrosamine, Sarma and his colleagues [lo] have reported the greater susceptibility of 06-methylguanine, in replicated DNA to digestion by Sl nuclease, than 7-methylguanine. Discontinuities in newly synthesised DNA because of alkylated bases in the template strand would appear an even greater likelihood for DEN, since compared to its methyl analogue, the products of DNA ethylation are less rapidly eliminated from rat liver DNA [3]. Thus altered sedimentation in alkaline sucrose of rat liver DNA isolated up to 56 days after DEN treatment [4] is certainly compatible with replication-induced structural damage being generated by hepatectomy 14 days (Fig. 3) after exposure to the nitrosamine. The similar response pattern obtained using parental or daughter strandlabelled DNA is consistent with BD-cellulose binding lesions being present in hybrid molecules such that they are detected by radioactivity contained in either strand. Such single stranded regions would be substrates for ‘postreplication repair’ [5]. The detection of single stranded regions in DNA isolated 6 days after hepatectomy (Fig. l), suggests that, if post-replication repair does occur, it does not result in the elimination of all replicationassociated structural damage. Little is known about how specific DNA lesions are translated into aberrations at the chromosomal level [ 1,9]. However, structural damage may be the genesis of micronuclei described by Den Engelse and his colleagues [17] in rats subjected to partial hepatectomy l-56 days after administration of DEN. This pre-clastogenic damage was associated with 0-alkylation. Although carcinogen adducts may thus be related to the generally accepted critical role of DNA replication in initiation, exploitation of rat liver models involving surgically induced hyperplasia in DEN-treated rats has been principally at the cellular level (for review, see ref. 6). If structural damage to DNA, as described here and evidenced by others [17], could be related to specific sequences, it may be possible to associate such lesions with recognized enzymic markers of pre-malignant cell populations.
ACKNOWLEDGEMENT
This research was made possible Medical Research Council.
by a grant from the National
Health and
REFERENCES 1 Bender, M.A. (1980) Relationship of DNA lesions and their repair to chromosomal tion production. Basic Life Sci., 15, 245-265.
aberra
269 2 Craddock, V.M. (1976) Cell proliferation and experimental liver cancer. In: Liver Cell Cancer, pp. 152-201. Editors: H.M. Cameron, D.A. Linsell and G.P. Warwick. Elsevier, Amsterdam. 3 Den Engelse, L., Floot, B.G.J., De Brij, R.J. and Tates, A.D. (1983) The induction of chromosomal damage in rat hepatocytes and lymphocytes. II. Alkylation damage and repair of rat liver DNA after diethylnitrosamine, dimethylnitrosamine and ethyl methanesulfonate in relation to clastogenic effects. Mutat. Res., 107, 153-156. 4 Floot, B.G.J., Philippus, E.J., Hart, A.A.M. and Den Engelse, L. (1979) Persistence and accumulation of (potential) single strand breaks in liver DNA of rats treated with diethylnitrosamine or dimethylnitrosamine: correlation with hepatocarcinogenicity. Chem.-Biol. Interact., 25, 229242. 5 Fujiwara, Y. (1975) Postreplication repair of alkylation damage to DNA of mammalian cells in culture. Cancer Res., 35, 27892789. 6 Goldsworthy, T.L., Hanigan, M.H. and Pitot, H.C. (1986) Models of hepatocarcinogenesis in the rat - contrasts and comparisons. CRC Crit. Rev. Toxicol., 17, 61-89. 7 Haber, M., Kavallaris, M. and Stewart, B.W. (1986) Two-stage incorporation of thymidine triphosphate into mammalian DNA as indicated by chromatography on benzoylated DEAEcellulose. J. Chromatogr., 382, 127-134. 8 Haber, M. and Stewart, B.W. (1985) Patterns of structural change in DNA during tissue necrosis indicated by benxoylated DEAE-cellulose chromatography. Chem.-Biol. Interact., 53, 247-255. 9 Kihlman, B.A. (1971) Molecular mechanisms of chromosome breakage and rejoining. Adv. Cell Mol. Biol., 1, 59107. 10 Ledda, G.M., Columbano, A., Rao, P.M., Rajalakshmi, S. and Sarma, D.S.R. (1980) In uivo replication of carcinogen-modified rat liver DNA: increased susceptibility of O-methylguanine compared to 7-methylguanine in replicated DNA to Sl nuclease. Biochem. Biophys. Res. Commun., 95, 816821. 11 Margison, G.P. and O’Connor, P.J. (1979) Nucleic acid modification by N-nitrose compounds. In: Chemical Carcinogens and DNA, pp. 111-159. Editor: P.L. Glover. CRC Press, Baltimore. 12 Preussmann, R. and Stewart, B.W. (1984) N-nitroso Carcinogens. In: Chemical Carcinogens, 2nd Edition, pp. 64%828. Editor: C.E. Searle. American Chemical Society, Washington, D.C. 13 Stewart, B.W. (1981) Generation and persistence of carcinogen-induced repair intermediates in rat liver DNA in uiuo. Cancer Res., 41, 32383243. 14 Stewart, B.W. and Brian, M.J. (1979) Evidence of carcinogen-induced replication of partially-repaired DNA in target cells during nitrosamine carcinogenesis. Eur. J. Cancer, 15, 251-256. 15 Stewart, B.W., Hristoforidis, C. and Haber, M. (1985) Dose-dependent persistence of alkylation-induced single stranded regions in rat liver DNA in uiuo. Cancer Letters, 28, 27-33. 16 Stewart, B.W., Huang, P.H.T. and Brian, M.J. (1979) Structural defects in rat liver deoxyribonucleic acid. Endogenous single-stranded regions in comparison with damage induced in vivo by a carcinogen. Biochem. J., 179, 341-352. 17 Tates, A.D., Neuteboom, I., Rotteveel, A.H.M., de Vogel, N., Menkveld, G.J. and Den Engelse, L. (1986) Persistence of preclastogenic damage in hepatocytes of rats exposed to ethylnitrosourea, diethylnitrosamine, dimethylnitrosamine and methyl methanesulphonate. Correlation with DNA 0.alkylation. 18 Zurlo, J., Eustice, D.C., Mignano, J.E., Poirior, M.C. and Yager, J.D. (1986) Effects of carcinogen treatment on rat liver DNA synthesis in viva and on nascent DNA synthesis and elongation in cultured hepatocytes. Mutat. Res., 161, 143154.