- Email: [email protected]
Significant existence of deleted mitochondrial DNA in cirrhotic liver surrounding hepatic tumor
Vol. 182, No. 2, 1992 January
BIOCHEMICAL
AND BIOPHYSICAL
31, 1992
RESEARCH COMMUNICATIONS Pages 913-920
Significant Existence of Deleted Mitochondrial DNA in Cirrhotic Liver Surrounding Hepatic Tumor Hideo Yamamoto, Masashi Tanaka*, Makoto Katayama, Toshihiro Obayashi*, Yuji Nimura, and Takayuki Ozawa* First Department of Surgery and *Department of Biomedical Chemistry, Faculty of Medicine, University of Nagoya, Nagoya 466, Japan Received
December
18,
1991
Summary: To understand the role of mitochondria in carcinogenesis, we compared the amount of deleted mtDNAs between human hepatic tumors and surrounding cirrhotic portion of the liver of ten patients by using polymerase chain reaction (PCR). Multiple mtDNA deletions were detected in cirrhotic portion, but no deletions were detected in the tumor portion. Direct sequencing of the fragments revealed a 7,079-bp deletion (nucleotide position 8,992-16,072) involving no direct repeated sequences and a 7,436-bp deletion (position 8,649-16,084) involving a 12-bp directly repeated sequence of SCATCAACAACCG-3’ exists in both the ATP6 gene and the D-loop region. These mtDNA mutations could be one of the endogenous factors that induce somatic mutations in nuclear genome and etiologically contribute to human carcinogenesis. 0 1992 Academic Press, hc.
Most hepatocellular carcinomas in Japan are associated with cirrhosis or fibrosis in the noncancerous portion of the liver (1). Liver cirrhosis is a chronic liver lesion which involves both diffuse destruction and regeneration of hepatic parenchymal cells, and is one of the risk factors for liver cancer (2). In experimental liver cirrhosis models, reduced function of mitochondria has been reported (2, 3,4, 5). It seems possible that the impaired function of hepatocytes in liver cirrhosis is related to dysfunction of mitochondria. Recently, it was reported (6) that somatic mutations in the p53 tumor suppressor gene are common in diverse types of human cancer including the liver tumor, and that both exogenous and endogenous factors may contribute to the somatic mutations from the differences of the p53 mutational spectrum among the tumors. In this paper, it is intended to clarify whether mutations in mitochondrial DNA (mtDNA), as one of the endogenous factors to carcinogenesis, exists in cirrhotic liver cells or not. Mitochondria are unique organellas containing their own genome, which encodes 13 subunits of the ATP producing system (7). For the maintenance of this system, the entire expression of the whole genome is needed. Moreover, mtDNA shows a high mutation rate and does not possess an adequate DNA repair system. Therefore, any type of mutation in the 0006-291X/92 Copyright All rights
$1.50
0 1992 by Academic Press, of reproduction in any form
Inc. reserved.
913
Vol.
182, No. 2, 1992
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
mtDNA is likely to exert profound effects on mitochondrial function.
It is proposed that
deleted mtDNAs are a major contributor to both the ageing process and degenerative diseases (8). In fact, age-dependent increase in deleted mtDNA has been demonstrated among human hearts (9), diaphragms (lo), liver (11) and several organs (12). In addition, mtDNA is considered as a preferential target for certain exogenous mutagens and carcinogens (13). Several mutations of mtDNA have been demonstrated in rat tumor cell lines (14). However, mtDNA mutations have not been analyzed in human liver carcinomas. In order to investigate whether there are any mutations of mtDNA in cirrhotic liver or whether there is any difference in the amount of mutant mtDNA between cirrhotic liver and liver cancer, we analyzed the amount of deleted mtDNA in both hepatic tumors and the surrounding cirrhotic portion by using a sensitive gene amplification method which enables us to detect a small population of deleted mtDNA (15, 16). The results presented here show that few deletions were detected in the tumor portion, but significant amount of multiple mtDNA deletions exists in cirrhotic portion,
PATIENTS
AND METHODS
Patients: Hepatic tumors and surrounding cirrhotic portion of liver tissues were obtained from the resected specimens of ten patients with the liver cancer bearing cirrhosis who underwent surgery between March 1989 and October 1990. A clinical summary of the patients is presented in Table 1. The patients (eight males and two females) ranged in age from 41 to 67 years, with a mean of 56.7 years. Hepatitis B surface antigen was detected in three of ten patients. Histological examination revealed hepatocellular carcinoma in nine patients and cholangiocellular carcinoma in one patient, all of which were associated with liver cirrhosis. Preparation of DNA: Tissues (30 mg) of hepatic tumor and cirrhotic liver were homogenized using a Physcotron Handy Micro Homogenizer (N&on, Tokyo) for 30 set, and then digested in 1 ml of 10 mM Tris-HCl, 0.1 M EDTA (pH 7.4) containing 0.1 mg/ml proteinase K and 0.5% sodium dodecyl sulfate. DNA was extracted twice with an equal volume of phenol/chloroform/isoamyl alcohol (25:25:1), then once with chloroform/isoamyl
Table 1. The list of patients with the hepatic tumors associated with liver cirrhosis No.
Patient
Sex
1 2 3 4 5 6 7 8 9 10
N.S. J.U. T.H. T.M. S.I. N.H. T.S. S.H. S.H. S.M.
F M M M F M M M M M
Age (yead 41 54 54 55 57 58 59 59 63 67
HBs Ag*
Type of hepatic tumor
Hepatocellular Hepatocellular Hepatocellular Hepatocellular Cholangiocellular Hepatocellular Hepatocellular Hepatocellular Hepatccellular Hepatocellular
-
F, female; M, male; *HBs Ag, hepatitis B antigen.
914
carcinoma carcinoma carcinoma carcinoma carcinoma carcinoma
carcinoma carcinoma carcinoma carcinoma
Vol.
182, No. 2, 1992
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
alcohol (25:l). DNA was precipitated with one-tenth volume of 3 M sodium acetate (pH 7.4) and two volumes of ethanol at -8o’C for 1 hour, and then rinsed with 70% ethanol. The precipitated DNA was recovered in 50 ~1 of 10 mM Tris-HCl, 0.1 mM EDTA (pH 8.0). Oligonucleotide primers: Primers used for PCR were synthesized using a model 380B DNA synthesizer (Applied Biosystems, Foster City, CA) and then purified on Oligonucleotide Purification Cartridges (Applied Biosystems). The base sequences of the oligonucleotide primers are reported in the previous paper (( 17)). Primary PCR amplification: Polymerase chain reaction (PCR) amplification was carried out using 2 ~1 of the DNA solution (cu. 20 ng of total cellular DNA), in a final volume of 100 ~1 PCR buffer (10 mM Tris-HCl, pH 8.8, 50 mM KCl, 1.5 mM MgC12 and 0.1% Triton X-100) which included 200 p.M of each dNTP, 2.5 units of Taq DNA polymerase (AmpliTaq, Cetus, Emeryville, CA) and 1 p.M of each primer. For comparative analysis of PCR products, PCR reactions were performed separately by using the same template with two different pairs of primers: one pair (primers L116 and H617) for estimating the amount of normal mtDNA, and another pair (primers L853 and H38) for detecting deleted mtDNAs. The reactions were carried out for a total of 35 amplification cycles with the use of a Thermal Cycler (Perkin-Elmer/Cetus). The cycle times were as follows: denaturation at 94’C for 15 set; annealing at 5o’C for 15 set; primer extension at 72°C for 120 set for the primer pairs L853 and H38 or for 100 set for the primer pairs L116 and H617. Amplified fragments were separated by electrophoresis on 1% agarose gels and were detected fluorographically after staining with ethidium bromide. Primer shift PCR: In PCR, misannealing of primers sometimes results in amplification of abnormal fragments. The primer shift method (15) was applied in order to ascertain that amplified fragments are not derived from misannealing of primers to an unexpected position of mtDNA; the principle of this method is as follows. In the first experiment, a fragment is amplified from the deleted mtDNA by using a pair of primers L853 and H38 surrounding the deletion. The size of deletion can be obtained by subtracting the size of the amplified fragment from the distance between the primers. In the second experiment, another fragment is amplified from the deleted mtDNA by using the second pair of primers L731 and H38. In the third experiment, the fragment is amplified from the deleted mtDNA by using the third pair of primers L853 and H60. The shift in the sizes of the amplified fragments should parallel the shift in the positions of the primers from L853 to L73 1 (1.2 kb), and from H38 to H60 (0.2 kb), respectively. If the deletion sizes calculated from the two experiments are identical, it is concluded that the amplified abnormal fragments are not due to misannealing of the primers but due to the presence of the deleted mtDNA. PCR-Southern analysis: PCR-Southern analysis is another method (16) to confirm the existence of mtDNA deletion. Three different fragments were amplified by PCR from the normal mtDNA and used as the probes: probe A, spanning positions 8,531-8,860 and covering the left-side boundary of the deletion, was amplified using primers L853 and H884; probe B, spanning positions 11,671-l 1,910 and covering the middle part of the deletion, was amplified using primers L1167 and H1189; probe C, spanning positions 16,41 l-140 and covering the right-side boundary of the deletion, was amplified using primers L1641 and H12. If the abnormal PCR fragment is derived from the deleted mtDNA, probes A and C would hybridize to it, but probe B would not . DNA sequencing: DNA was sequenced by the Sanger’s dideoxynucleotide chain termination method using the incorporation of a-[32P]dCTP as the radiolabeling extension method (18).
RESULTS Before we analyzed deleted mtDNA, we estimated the amount of normal mtDNA in each sample by using the primers, L116 (position 1,161-1,180) and H617 (position 6,1716,190), the distance between which was 5.0 kb (Fig. 1). We confirmed that the amounts of the 5.0-kb fragment were approximately the same in both cirrhotic liver (A) and hepatic tumor (B) of all patients. The concentrations of normal mtDNA that served as the template for PCR in each sample are similar between the cirrhotic portion of the liver and the tumors of these patients. 915
Vol.
182, No. 2, 1992
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
Fig. 1. PCR amplification using primers L116 and H617. The normal-sized fragments of 5.0 kb am amplified in all the samples, and the amounts of the fragments are almost the same in each cirrhotic liver (A) and hepatic tumor (B). Fig. PCR detection of deleted mtDNAs using primers L853 and H38 in cirrhotic liver (A) and hepatic tumor (B) of ten patients. Amplified fragments were separated on a 1% agarose gel, and then stained with ethidium bromide. Sizes of amplified fragments are indicated in kb.
Fig. 2 shows analysis of deleted mtDNA in cirrhotic liver and hepatic tumor of the patients by using primers L853 (position 8,531 - 8,550) and H38 (position 381 - 400). Multiple abnormal fragments of 1.4, 1.0 and 0.7-kb, probably derived from deleted mtDNAs, were detected in the cirrhotic portion of the liver of all the patients (Fig. 2A), but abnormal fragments were scarcely detected in the hepatic tumors (Fig. 2B). In order to exclude the possibility that the abnormal fragment resulted from misannealing of primers, we further analyzed these abnormal fragments by the primer shift PCR method and the PCR Southern method. Fig. 3 shows the result of primer shift PCR analysis of the primary PCR products obtained in the cirrhotic liver of patient 4 (Fig. 2A, lane 4). When a pair of primers L853 and H38 were used, l.O-kb and 1.4-kb fragments (Fig. 3, lane A) were amplified. When the distance between primers was shifted from 8.4 kb to 8.6 kb by replacing primer H38 by primer H60, 1.2-kb and 1.6-kb abnormal fragments were amplified (lane B). When the distance between primers was shifted from 8.4 kb to 9.6 kb by replacing primer L853 by primer L731, a 2.2-kb abnormal fragment was amplified (lane C), probably corresponding to the l.O-kb fragment amplified by using primer pairs L853 and H38 (lane A). These result suggest that these abnormal fragments are derived from mutant mtDNAs with deletions of 7.0 kb and 7.4 kb. In order to further confirm that the small fragments am derived from deleted mtDNA, a PCR-Southern analysis was performed (Fig. 4). The PCR products shown in lane A of Fig. 3 were blotted onto a nylon membrane and probed with three different probes. Probe A (spanning nucleotide position 8,531- 8,860) and probe C (spanning 16,411- 140) hybridized to the 1.4kb and l.O-kb fragments, but probe B (spanning 11,671- 11,910) did not hybridize to any of these fragments. These results clearly indicate that 1.4-kb and l.O-kb fragments are not artifacts but are derived from deleted mtDNAs. The 0.7-kb fragment was hybridized to 916
Vol.
182, No. 2, 1992
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
tHJX deletion LS53 PROBE
-
L1167 L t
L1641 A
t H884
Ill189 B
c HI2
C
Fig. 3. Primer shift PCR analysis of deleted mtDNA in cirrhotic liver. PCR amplification was carried out using primers L853 and H38 (lane A), primers L853 and H60 (lane B), and primers L731 and H38 (lane C). Sizes of amplified fragments are indicated in kb. The shift in the sizes of the amplified fragments paralleled the shift in the positions of the primers from L853 to L731 (1.2 kb) and from H60 to H38 (0.2 kb). Fie. The schema and result of the PCR-Southern analysis of deleted mtDNA in cirrhotic liver. The PCR products shown in lane 4 of Fig. 2A were separated on a 1% agarose gel, blotted onto a nylon membrane, and hybridized with three different probes: probe A (spanning positions 8,531- 8,860) amplified with primers L853 and H884, probe B (spanning positions 11,671 - 11,910) amplified with primers L1167 and H1189; and probe C (spanning positions 16,411- 140) amplified with primers L1641 and H12.
probe A, but not to probe B or C. It is not clear whether this fragment is derived from a deleted mtDNA. We analyzed the sequences of the junctional regions of these two abnormal fragments by using the direct sequencing method (18). The crossover sequence in the l.O-kb fragment was demonstrated to be a 1Zbp directly repeated sequence of 5’-CATCAACAACCG-3’, which was located at the boundaries of the deletion between the ATP6 gene and the D-loop region. This deletion spanned 7,436 bp from nucleotide position 8,649 to position 16,084. The crossover sequence in the 1.4-kb fragment was demonstrated to be a single C nucleotide, which was located at the boundaries of the deletion between the ATP6 gene and the D-loop region. This deletion started from position 8,992 and ended at position 16,072 of the mtDNA resulting in a 7,079 bp deletion (9,17). Thus, at least two fragments detected in the cirrhotic livers (Pig. 2A) were proved to be derived from deleted mtDNA. DISCUSSION In our present study, the multiple abnormal fragments were detected in cirrhotic liver of all the patients analyzed. By using the primer shift PCR method, the PCR Southern method, 917
Vol.
182,
No.
2, 1992
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
and the direct DNA sequencing method, we confumed that the detected fragments were derived from the deleted mtDNA. In fact, the junctional regions of two major deleted mtDNA were clarified by sequencing. We have proposed that the accumulation of mitochondrial gene mutations and the subsequent cytoplasmic segregation of deleted mtDNA during life are an important contributor both to the ageing process and to several human degenerative diseases (8). Mutant mtDNA with a 4,977-bp deletion was detected in the striatum of both Parkinson’s patients and aged controls (19). Age-related increase of the same 4,977-bp deletion in human liver (11) and several organs (12) was reported recently. The 7,436-bp deletion in liver mtDNA reported here was commonly detected in the heart of elderly individuals (9), and the proportion of deleted mtDNA to normal mtDNA increased with advancing age (20). We have observed ageassociated accumulation of 8-hydroxydeoxyguanosine and deletions of mtDNA from diaphragmatic muscle (21). These observations suggest that accumulation of deleted mtDNA and consequent mitochondrial dysfunction play a role in the aging process. In the case of degenerative disease, multiple deletions of mtDNA among mitochondrial myopathy (15, 18) and cardiomyopathy (16) were reported. The 7,079-bp deletion in liver mtDNA reported here was found in the myocardial mtDNA of a patient with cardiomyopathy in association with 7,436-bp deletion (17). Liver cirrhosis is a diffuse process characterized by fibrosis and irreversible conversion of normal architecture into structurally abnormal nodules. Morimoto et al. (3) demonstrated that phosphorylative activity per unit of cytochrome a(+us) and the hepatic energy charge level decreased significantly in the carbon tetrachloride induced cirrhotic rat liver. Krahenbiihl et al. (4) also showed that RCI was significantly reduced for hydroxybutyrate, but not for succinate in mitochondria of the cirrhotic rat liver induced by carbon tetrachloride, indicating a damage at the coupling site I of the respiratory chain. Omokawa ef al. (5) reported that respiratory control index (RCI), ADPIO ratio and ATP synthesis were significantly lower in the thioacetamideinduced cirrhotic rat liver than those in normal liver. Diaz et al. (22) observed decreased RCI in mitochondrial isolated from human cirrhotic liver. Accumulated evidence establishes that liver cirrhosis is associated with mitochondrial dysfunction both in human and in rat cirrhotic liver models. It is hypothesized that mitochondrial damage could be a possible route to oncogenesis (23,24). Association of mitochondrial tRNA mutations and tumors has been reported in rat hepatoma cell lines (14). In contrast, Welter et al. (25) reported that no major structural changes were observed between the mtDNA of five human colon cancers and that of adjacent healthy tissues of the same patients when the restriction fragment patterns obtained by ten restriction endonucleases were analyzed. They suggested that a strong selective mechanism exists conserving the primary structure of mtDNA in tumorigenesis. Monnat et al. (26), who analyzed mtDNA from human leukemia cells in detail using a DNA sequencing technique, also reported that nucleotides in mtDNA of human leukemia cells were conserved at a higher level than mtDNA of different individuals and suggested that a mechanism or mechanisms exist that limit the development of nucleotide sequence divergence in mtDNA. In the present study, we 918
Vol.
182, No. 2, 1992
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
have demonstrated that deleted mtDNAs are barely detectable in hepatic tumors, although multiple mutant mtDNAs with deletions are easily detected in cirrhotic portion of the liver of the same patient. These findings suggest that liver cells which acquired somatic mutations in the tumor suppressor gene (6) but bioenergetically active with few deleted mtDNAs could survived during tumor growth. The most common cancer-related genetic change known is mutations in a tumor suppressor gene, ~53, and over 95 % of the mutations are somatically acquired (27). In addition to point mutations, allelic loss, rearrangements, and deletion of the p53 gene have been detected in human tumors. G to T transversions are dispersed among numerous codons in the p53 gene (6). &Hydroxydeoxyguanosine (8-OH-dG), an active oxygen adduct of DNA, was shown to generate G to T transversions (28,29). Thus, oxidative damage of DNA seems to be a plausible cause of mutations in the tumor suppressor gene. We have observed that administration of azidothymidine which have been reported to cause mitochondrial myopathy in patients with AIDS resulted in massive conversion of deoxyguanosine to 8-OHdG in mouse mtDNA (30) and that age-associated accumulation of 8-OH-dG in mtDNA from human diaphragmatic muscle is accompanied with a significant rise in the rate of deletions (21). Impaired mitochondrial electron transport chain due to mutated mtDNA generate free radicals which was shown in vitro experiment (30). Thus, mitochondria with multiple deleted mtDNA in cirrhotic liver reported here could be a major source of active oxygen causing somatic mutations in the nuclear cancer-related genes, and thus one of the important endogenous factors etiologically contributing to human carcinogenesis. REFERENCES ::
Okuda, K., and the liver cancer study group of Japan. (1980) Cancer 45,2663-2669. Vecchia, C. L., Negri, E., D’Avanzo, B., Boyle, P., and Franceschi, S. (1990) Cancer Res. 50, 6274-6277.
3.
Morimoto, T., Tanaka, A., Taki, Y., Noguchi, M., Yokoo, N., Nishihira, T., Nishikawa, K., Yamamoto, S., Nitta, N., Jikkoh, A., Kamiyama, Y., Yamaoka, Y., and Ozawa, K. (1988) Clin. Sci. 74,485-489. Krghenbiihl, S., Stucki, J., and Reichen, J. (1989) Biochem. Pharmacol. 38, 15831588. Omokawa, S., Koyama, K., Suzuki, K., and Asanuma, Y. (1989) Tohoku J. Exp. Med. 158, 265-268. Hollstein, M., Sidransky, D., Vogelstein, B., and Harris, C. C. (1991) Science 253, 49-53.
8. 9.
10. 11. 12.
Anderson, S., Bankier, A. T., Barrell, B. G., de Bruijn, M. H. L., Coulson, A. R., Drouin, J., Eperon, I. C., Nierlich, D. P., Roe, B. A., Sanger, F., Schreier, P. H., Smith, A. J. H., Staden, R., and Young, I. G. (1981) Nature 290, 457-465. Linnane, A. W., Marzuki, S., Ozawa, T., and Tanaka, M. (1989) Luncet i, 642-645. Hattori, K., Tanaka, M., Sugiyama, S., Obayashi, T., Ito, T., Satake, T., Hanaki, Y., Asai, J., Nagano, M., and Ozawa, T. (1991) Am. Heart. J. 1735-1742. Torii, K., Sugiyama, S., Tanaka, M., Takagi, K., Hanaki, Y., Iida, K., Matsuyama, M., Hirabayashi, N., Uno, Y., and Ozawa, T. (1991) Am. J. Respir. Cell. Mol. Biol. (in press). Yen, T. C., Su, J. H., King, K. L., and Wei, Y. H. (1991) Biochem. Biophys. Res. Comnwn. 178, 124-131. Linnane, A. W., Baumer, A., Maxwell, R. J., Preston, H., Zhang, C. F., and Marzuki, S. (1990) Biochem. Int. 22, 1067-1076. 919
Vol.
182, No. 2, 1992
13. 14. 15. 16. 17. 18. 19. 20. 21. 22.
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
Wilkie, D., and Fearon, K. (1985) in : Achievement and perspectives of mitochondrial research. Volume II: Biogenesis (Quagliariello, Slater, Palmieri, Saccone and Kroon ed.) pp 437-444, Elsevier, Amsterdam. Taira, M., Yoshida, E., Kobayashi, M., Yaginuma, K., and Koike, K. (1983) Nucleic Acids Res. 11, 1635-1643. Sato, W., Tanaka, M., Ohno, K., Yamamoto, T., Takada, G., and Ozawa, T. (1989) Biochem. Biophys. Res. Commun. 162, 664-672. Ozawa, T., Tanaka, M., Sugiyama, S., Hattori, K., Ito, T., Ohno, K., Takahashi, A., Sato, W., Takada, G., Mayumi, B., Yamamoto, K., Adachi, K., Koga, Y., and Toshima, H. (1990) Biochem. Biophys. Res. Commun. 170, 830-836. Hattori, K., Ogawa, T., Kondo, T., Mochizuki, M., Tanaka, M., Sugiyama, S., Ito, T., Satake, T., and Ozawa, T. (1991) Am. Heart J. i121, 1735-1742. Tanaka, M., Sato, W., Ohno, K., Yamamoto, T., and Ozawa, T. (1989) Biochem. Biophys. Res. Commun. 164, 156-163. Ikebe, S., Tanaka, M., Ohno, K., Sato, W., Hattori, K., Kondo, T., Mizuno, Y., and Ozawa, T. (1990) Biochem. Biophys. Res. Commun. 170, 1044-1048. Sugiyama, S., Hattori, K., Hayakawa, M., and Ozawa, T. (1991) Biochem. Biophys. Res. Commun. 180, 894-899. Hayakawa, M., Torii, K., Sugiyama, S., Tanaka, M., and Ozawa, T. (1991) Biochem. Biophys. Res. Commun. 179, 1023-1029. Diaz, J., Rossi, I., Escartin, P., Segovia, J., and Gosalvez, M. (1977) Clin. Sci. Mol. Med. 52, 599606.
23.
Corral, M., Kitzis, A., Baffet, G., Paris, B., Tichonicky, L., Kruh, J., GuguenGuillouzo, C., and Defer, N. (1989) Nucleic Acids Res. 17, 5191-5206. 24. Wilkie, D., Evans, I. H., Egilsson, V., Diala, E. S., and Collier, D. (1983) Int. Rev. Cytol. St@. 15, 157-89. 25. Welter, C., and Blin, N. (1988) J. Exp. Clin. Cancer Res. 7, 13-20. Monnat, R., Maxwell, C., and Loeb, L. (1985) Cancer Res. 45, 1809-1814. ;76* Vogelstein, B. (1990) Nature 348,681- 682. 28: Shibutani, S., Takeshita, M., and Grollman, A. P. (1991) Nature. 349,431-434. 29. Wood, M. L., Dizdaroglu, M., Gajewski, E., and Essigmann, J. M. (1990) Biochemistry 349, 7024-7032.
30.
Hayakawa, M., Ogawa, T., Sugiyama, S., Tanaka, M., and Ozawa, T. (1991) Biochem. Biophys. Res. Commun. 176, 87-93.
920
BIOCHEMICAL
AND BIOPHYSICAL
31, 1992
RESEARCH COMMUNICATIONS Pages 913-920
Significant Existence of Deleted Mitochondrial DNA in Cirrhotic Liver Surrounding Hepatic Tumor Hideo Yamamoto, Masashi Tanaka*, Makoto Katayama, Toshihiro Obayashi*, Yuji Nimura, and Takayuki Ozawa* First Department of Surgery and *Department of Biomedical Chemistry, Faculty of Medicine, University of Nagoya, Nagoya 466, Japan Received
December
18,
1991
Summary: To understand the role of mitochondria in carcinogenesis, we compared the amount of deleted mtDNAs between human hepatic tumors and surrounding cirrhotic portion of the liver of ten patients by using polymerase chain reaction (PCR). Multiple mtDNA deletions were detected in cirrhotic portion, but no deletions were detected in the tumor portion. Direct sequencing of the fragments revealed a 7,079-bp deletion (nucleotide position 8,992-16,072) involving no direct repeated sequences and a 7,436-bp deletion (position 8,649-16,084) involving a 12-bp directly repeated sequence of SCATCAACAACCG-3’ exists in both the ATP6 gene and the D-loop region. These mtDNA mutations could be one of the endogenous factors that induce somatic mutations in nuclear genome and etiologically contribute to human carcinogenesis. 0 1992 Academic Press, hc.
Most hepatocellular carcinomas in Japan are associated with cirrhosis or fibrosis in the noncancerous portion of the liver (1). Liver cirrhosis is a chronic liver lesion which involves both diffuse destruction and regeneration of hepatic parenchymal cells, and is one of the risk factors for liver cancer (2). In experimental liver cirrhosis models, reduced function of mitochondria has been reported (2, 3,4, 5). It seems possible that the impaired function of hepatocytes in liver cirrhosis is related to dysfunction of mitochondria. Recently, it was reported (6) that somatic mutations in the p53 tumor suppressor gene are common in diverse types of human cancer including the liver tumor, and that both exogenous and endogenous factors may contribute to the somatic mutations from the differences of the p53 mutational spectrum among the tumors. In this paper, it is intended to clarify whether mutations in mitochondrial DNA (mtDNA), as one of the endogenous factors to carcinogenesis, exists in cirrhotic liver cells or not. Mitochondria are unique organellas containing their own genome, which encodes 13 subunits of the ATP producing system (7). For the maintenance of this system, the entire expression of the whole genome is needed. Moreover, mtDNA shows a high mutation rate and does not possess an adequate DNA repair system. Therefore, any type of mutation in the 0006-291X/92 Copyright All rights
$1.50
0 1992 by Academic Press, of reproduction in any form
Inc. reserved.
913
Vol.
182, No. 2, 1992
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
mtDNA is likely to exert profound effects on mitochondrial function.
It is proposed that
deleted mtDNAs are a major contributor to both the ageing process and degenerative diseases (8). In fact, age-dependent increase in deleted mtDNA has been demonstrated among human hearts (9), diaphragms (lo), liver (11) and several organs (12). In addition, mtDNA is considered as a preferential target for certain exogenous mutagens and carcinogens (13). Several mutations of mtDNA have been demonstrated in rat tumor cell lines (14). However, mtDNA mutations have not been analyzed in human liver carcinomas. In order to investigate whether there are any mutations of mtDNA in cirrhotic liver or whether there is any difference in the amount of mutant mtDNA between cirrhotic liver and liver cancer, we analyzed the amount of deleted mtDNA in both hepatic tumors and the surrounding cirrhotic portion by using a sensitive gene amplification method which enables us to detect a small population of deleted mtDNA (15, 16). The results presented here show that few deletions were detected in the tumor portion, but significant amount of multiple mtDNA deletions exists in cirrhotic portion,
PATIENTS
AND METHODS
Patients: Hepatic tumors and surrounding cirrhotic portion of liver tissues were obtained from the resected specimens of ten patients with the liver cancer bearing cirrhosis who underwent surgery between March 1989 and October 1990. A clinical summary of the patients is presented in Table 1. The patients (eight males and two females) ranged in age from 41 to 67 years, with a mean of 56.7 years. Hepatitis B surface antigen was detected in three of ten patients. Histological examination revealed hepatocellular carcinoma in nine patients and cholangiocellular carcinoma in one patient, all of which were associated with liver cirrhosis. Preparation of DNA: Tissues (30 mg) of hepatic tumor and cirrhotic liver were homogenized using a Physcotron Handy Micro Homogenizer (N&on, Tokyo) for 30 set, and then digested in 1 ml of 10 mM Tris-HCl, 0.1 M EDTA (pH 7.4) containing 0.1 mg/ml proteinase K and 0.5% sodium dodecyl sulfate. DNA was extracted twice with an equal volume of phenol/chloroform/isoamyl alcohol (25:25:1), then once with chloroform/isoamyl
Table 1. The list of patients with the hepatic tumors associated with liver cirrhosis No.
Patient
Sex
1 2 3 4 5 6 7 8 9 10
N.S. J.U. T.H. T.M. S.I. N.H. T.S. S.H. S.H. S.M.
F M M M F M M M M M
Age (yead 41 54 54 55 57 58 59 59 63 67
HBs Ag*
Type of hepatic tumor
Hepatocellular Hepatocellular Hepatocellular Hepatocellular Cholangiocellular Hepatocellular Hepatocellular Hepatocellular Hepatccellular Hepatocellular
-
F, female; M, male; *HBs Ag, hepatitis B antigen.
914
carcinoma carcinoma carcinoma carcinoma carcinoma carcinoma
carcinoma carcinoma carcinoma carcinoma
Vol.
182, No. 2, 1992
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
alcohol (25:l). DNA was precipitated with one-tenth volume of 3 M sodium acetate (pH 7.4) and two volumes of ethanol at -8o’C for 1 hour, and then rinsed with 70% ethanol. The precipitated DNA was recovered in 50 ~1 of 10 mM Tris-HCl, 0.1 mM EDTA (pH 8.0). Oligonucleotide primers: Primers used for PCR were synthesized using a model 380B DNA synthesizer (Applied Biosystems, Foster City, CA) and then purified on Oligonucleotide Purification Cartridges (Applied Biosystems). The base sequences of the oligonucleotide primers are reported in the previous paper (( 17)). Primary PCR amplification: Polymerase chain reaction (PCR) amplification was carried out using 2 ~1 of the DNA solution (cu. 20 ng of total cellular DNA), in a final volume of 100 ~1 PCR buffer (10 mM Tris-HCl, pH 8.8, 50 mM KCl, 1.5 mM MgC12 and 0.1% Triton X-100) which included 200 p.M of each dNTP, 2.5 units of Taq DNA polymerase (AmpliTaq, Cetus, Emeryville, CA) and 1 p.M of each primer. For comparative analysis of PCR products, PCR reactions were performed separately by using the same template with two different pairs of primers: one pair (primers L116 and H617) for estimating the amount of normal mtDNA, and another pair (primers L853 and H38) for detecting deleted mtDNAs. The reactions were carried out for a total of 35 amplification cycles with the use of a Thermal Cycler (Perkin-Elmer/Cetus). The cycle times were as follows: denaturation at 94’C for 15 set; annealing at 5o’C for 15 set; primer extension at 72°C for 120 set for the primer pairs L853 and H38 or for 100 set for the primer pairs L116 and H617. Amplified fragments were separated by electrophoresis on 1% agarose gels and were detected fluorographically after staining with ethidium bromide. Primer shift PCR: In PCR, misannealing of primers sometimes results in amplification of abnormal fragments. The primer shift method (15) was applied in order to ascertain that amplified fragments are not derived from misannealing of primers to an unexpected position of mtDNA; the principle of this method is as follows. In the first experiment, a fragment is amplified from the deleted mtDNA by using a pair of primers L853 and H38 surrounding the deletion. The size of deletion can be obtained by subtracting the size of the amplified fragment from the distance between the primers. In the second experiment, another fragment is amplified from the deleted mtDNA by using the second pair of primers L731 and H38. In the third experiment, the fragment is amplified from the deleted mtDNA by using the third pair of primers L853 and H60. The shift in the sizes of the amplified fragments should parallel the shift in the positions of the primers from L853 to L73 1 (1.2 kb), and from H38 to H60 (0.2 kb), respectively. If the deletion sizes calculated from the two experiments are identical, it is concluded that the amplified abnormal fragments are not due to misannealing of the primers but due to the presence of the deleted mtDNA. PCR-Southern analysis: PCR-Southern analysis is another method (16) to confirm the existence of mtDNA deletion. Three different fragments were amplified by PCR from the normal mtDNA and used as the probes: probe A, spanning positions 8,531-8,860 and covering the left-side boundary of the deletion, was amplified using primers L853 and H884; probe B, spanning positions 11,671-l 1,910 and covering the middle part of the deletion, was amplified using primers L1167 and H1189; probe C, spanning positions 16,41 l-140 and covering the right-side boundary of the deletion, was amplified using primers L1641 and H12. If the abnormal PCR fragment is derived from the deleted mtDNA, probes A and C would hybridize to it, but probe B would not . DNA sequencing: DNA was sequenced by the Sanger’s dideoxynucleotide chain termination method using the incorporation of a-[32P]dCTP as the radiolabeling extension method (18).
RESULTS Before we analyzed deleted mtDNA, we estimated the amount of normal mtDNA in each sample by using the primers, L116 (position 1,161-1,180) and H617 (position 6,1716,190), the distance between which was 5.0 kb (Fig. 1). We confirmed that the amounts of the 5.0-kb fragment were approximately the same in both cirrhotic liver (A) and hepatic tumor (B) of all patients. The concentrations of normal mtDNA that served as the template for PCR in each sample are similar between the cirrhotic portion of the liver and the tumors of these patients. 915
Vol.
182, No. 2, 1992
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
Fig. 1. PCR amplification using primers L116 and H617. The normal-sized fragments of 5.0 kb am amplified in all the samples, and the amounts of the fragments are almost the same in each cirrhotic liver (A) and hepatic tumor (B). Fig. PCR detection of deleted mtDNAs using primers L853 and H38 in cirrhotic liver (A) and hepatic tumor (B) of ten patients. Amplified fragments were separated on a 1% agarose gel, and then stained with ethidium bromide. Sizes of amplified fragments are indicated in kb.
Fig. 2 shows analysis of deleted mtDNA in cirrhotic liver and hepatic tumor of the patients by using primers L853 (position 8,531 - 8,550) and H38 (position 381 - 400). Multiple abnormal fragments of 1.4, 1.0 and 0.7-kb, probably derived from deleted mtDNAs, were detected in the cirrhotic portion of the liver of all the patients (Fig. 2A), but abnormal fragments were scarcely detected in the hepatic tumors (Fig. 2B). In order to exclude the possibility that the abnormal fragment resulted from misannealing of primers, we further analyzed these abnormal fragments by the primer shift PCR method and the PCR Southern method. Fig. 3 shows the result of primer shift PCR analysis of the primary PCR products obtained in the cirrhotic liver of patient 4 (Fig. 2A, lane 4). When a pair of primers L853 and H38 were used, l.O-kb and 1.4-kb fragments (Fig. 3, lane A) were amplified. When the distance between primers was shifted from 8.4 kb to 8.6 kb by replacing primer H38 by primer H60, 1.2-kb and 1.6-kb abnormal fragments were amplified (lane B). When the distance between primers was shifted from 8.4 kb to 9.6 kb by replacing primer L853 by primer L731, a 2.2-kb abnormal fragment was amplified (lane C), probably corresponding to the l.O-kb fragment amplified by using primer pairs L853 and H38 (lane A). These result suggest that these abnormal fragments are derived from mutant mtDNAs with deletions of 7.0 kb and 7.4 kb. In order to further confirm that the small fragments am derived from deleted mtDNA, a PCR-Southern analysis was performed (Fig. 4). The PCR products shown in lane A of Fig. 3 were blotted onto a nylon membrane and probed with three different probes. Probe A (spanning nucleotide position 8,531- 8,860) and probe C (spanning 16,411- 140) hybridized to the 1.4kb and l.O-kb fragments, but probe B (spanning 11,671- 11,910) did not hybridize to any of these fragments. These results clearly indicate that 1.4-kb and l.O-kb fragments are not artifacts but are derived from deleted mtDNAs. The 0.7-kb fragment was hybridized to 916
Vol.
182, No. 2, 1992
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
tHJX deletion LS53 PROBE
-
L1167 L t
L1641 A
t H884
Ill189 B
c HI2
C
Fig. 3. Primer shift PCR analysis of deleted mtDNA in cirrhotic liver. PCR amplification was carried out using primers L853 and H38 (lane A), primers L853 and H60 (lane B), and primers L731 and H38 (lane C). Sizes of amplified fragments are indicated in kb. The shift in the sizes of the amplified fragments paralleled the shift in the positions of the primers from L853 to L731 (1.2 kb) and from H60 to H38 (0.2 kb). Fie. The schema and result of the PCR-Southern analysis of deleted mtDNA in cirrhotic liver. The PCR products shown in lane 4 of Fig. 2A were separated on a 1% agarose gel, blotted onto a nylon membrane, and hybridized with three different probes: probe A (spanning positions 8,531- 8,860) amplified with primers L853 and H884, probe B (spanning positions 11,671 - 11,910) amplified with primers L1167 and H1189; and probe C (spanning positions 16,411- 140) amplified with primers L1641 and H12.
probe A, but not to probe B or C. It is not clear whether this fragment is derived from a deleted mtDNA. We analyzed the sequences of the junctional regions of these two abnormal fragments by using the direct sequencing method (18). The crossover sequence in the l.O-kb fragment was demonstrated to be a 1Zbp directly repeated sequence of 5’-CATCAACAACCG-3’, which was located at the boundaries of the deletion between the ATP6 gene and the D-loop region. This deletion spanned 7,436 bp from nucleotide position 8,649 to position 16,084. The crossover sequence in the 1.4-kb fragment was demonstrated to be a single C nucleotide, which was located at the boundaries of the deletion between the ATP6 gene and the D-loop region. This deletion started from position 8,992 and ended at position 16,072 of the mtDNA resulting in a 7,079 bp deletion (9,17). Thus, at least two fragments detected in the cirrhotic livers (Pig. 2A) were proved to be derived from deleted mtDNA. DISCUSSION In our present study, the multiple abnormal fragments were detected in cirrhotic liver of all the patients analyzed. By using the primer shift PCR method, the PCR Southern method, 917
Vol.
182,
No.
2, 1992
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
and the direct DNA sequencing method, we confumed that the detected fragments were derived from the deleted mtDNA. In fact, the junctional regions of two major deleted mtDNA were clarified by sequencing. We have proposed that the accumulation of mitochondrial gene mutations and the subsequent cytoplasmic segregation of deleted mtDNA during life are an important contributor both to the ageing process and to several human degenerative diseases (8). Mutant mtDNA with a 4,977-bp deletion was detected in the striatum of both Parkinson’s patients and aged controls (19). Age-related increase of the same 4,977-bp deletion in human liver (11) and several organs (12) was reported recently. The 7,436-bp deletion in liver mtDNA reported here was commonly detected in the heart of elderly individuals (9), and the proportion of deleted mtDNA to normal mtDNA increased with advancing age (20). We have observed ageassociated accumulation of 8-hydroxydeoxyguanosine and deletions of mtDNA from diaphragmatic muscle (21). These observations suggest that accumulation of deleted mtDNA and consequent mitochondrial dysfunction play a role in the aging process. In the case of degenerative disease, multiple deletions of mtDNA among mitochondrial myopathy (15, 18) and cardiomyopathy (16) were reported. The 7,079-bp deletion in liver mtDNA reported here was found in the myocardial mtDNA of a patient with cardiomyopathy in association with 7,436-bp deletion (17). Liver cirrhosis is a diffuse process characterized by fibrosis and irreversible conversion of normal architecture into structurally abnormal nodules. Morimoto et al. (3) demonstrated that phosphorylative activity per unit of cytochrome a(+us) and the hepatic energy charge level decreased significantly in the carbon tetrachloride induced cirrhotic rat liver. Krahenbiihl et al. (4) also showed that RCI was significantly reduced for hydroxybutyrate, but not for succinate in mitochondria of the cirrhotic rat liver induced by carbon tetrachloride, indicating a damage at the coupling site I of the respiratory chain. Omokawa ef al. (5) reported that respiratory control index (RCI), ADPIO ratio and ATP synthesis were significantly lower in the thioacetamideinduced cirrhotic rat liver than those in normal liver. Diaz et al. (22) observed decreased RCI in mitochondrial isolated from human cirrhotic liver. Accumulated evidence establishes that liver cirrhosis is associated with mitochondrial dysfunction both in human and in rat cirrhotic liver models. It is hypothesized that mitochondrial damage could be a possible route to oncogenesis (23,24). Association of mitochondrial tRNA mutations and tumors has been reported in rat hepatoma cell lines (14). In contrast, Welter et al. (25) reported that no major structural changes were observed between the mtDNA of five human colon cancers and that of adjacent healthy tissues of the same patients when the restriction fragment patterns obtained by ten restriction endonucleases were analyzed. They suggested that a strong selective mechanism exists conserving the primary structure of mtDNA in tumorigenesis. Monnat et al. (26), who analyzed mtDNA from human leukemia cells in detail using a DNA sequencing technique, also reported that nucleotides in mtDNA of human leukemia cells were conserved at a higher level than mtDNA of different individuals and suggested that a mechanism or mechanisms exist that limit the development of nucleotide sequence divergence in mtDNA. In the present study, we 918
Vol.
182, No. 2, 1992
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
have demonstrated that deleted mtDNAs are barely detectable in hepatic tumors, although multiple mutant mtDNAs with deletions are easily detected in cirrhotic portion of the liver of the same patient. These findings suggest that liver cells which acquired somatic mutations in the tumor suppressor gene (6) but bioenergetically active with few deleted mtDNAs could survived during tumor growth. The most common cancer-related genetic change known is mutations in a tumor suppressor gene, ~53, and over 95 % of the mutations are somatically acquired (27). In addition to point mutations, allelic loss, rearrangements, and deletion of the p53 gene have been detected in human tumors. G to T transversions are dispersed among numerous codons in the p53 gene (6). &Hydroxydeoxyguanosine (8-OH-dG), an active oxygen adduct of DNA, was shown to generate G to T transversions (28,29). Thus, oxidative damage of DNA seems to be a plausible cause of mutations in the tumor suppressor gene. We have observed that administration of azidothymidine which have been reported to cause mitochondrial myopathy in patients with AIDS resulted in massive conversion of deoxyguanosine to 8-OHdG in mouse mtDNA (30) and that age-associated accumulation of 8-OH-dG in mtDNA from human diaphragmatic muscle is accompanied with a significant rise in the rate of deletions (21). Impaired mitochondrial electron transport chain due to mutated mtDNA generate free radicals which was shown in vitro experiment (30). Thus, mitochondria with multiple deleted mtDNA in cirrhotic liver reported here could be a major source of active oxygen causing somatic mutations in the nuclear cancer-related genes, and thus one of the important endogenous factors etiologically contributing to human carcinogenesis. REFERENCES ::
Okuda, K., and the liver cancer study group of Japan. (1980) Cancer 45,2663-2669. Vecchia, C. L., Negri, E., D’Avanzo, B., Boyle, P., and Franceschi, S. (1990) Cancer Res. 50, 6274-6277.
3.
Morimoto, T., Tanaka, A., Taki, Y., Noguchi, M., Yokoo, N., Nishihira, T., Nishikawa, K., Yamamoto, S., Nitta, N., Jikkoh, A., Kamiyama, Y., Yamaoka, Y., and Ozawa, K. (1988) Clin. Sci. 74,485-489. Krghenbiihl, S., Stucki, J., and Reichen, J. (1989) Biochem. Pharmacol. 38, 15831588. Omokawa, S., Koyama, K., Suzuki, K., and Asanuma, Y. (1989) Tohoku J. Exp. Med. 158, 265-268. Hollstein, M., Sidransky, D., Vogelstein, B., and Harris, C. C. (1991) Science 253, 49-53.
8. 9.
10. 11. 12.
Anderson, S., Bankier, A. T., Barrell, B. G., de Bruijn, M. H. L., Coulson, A. R., Drouin, J., Eperon, I. C., Nierlich, D. P., Roe, B. A., Sanger, F., Schreier, P. H., Smith, A. J. H., Staden, R., and Young, I. G. (1981) Nature 290, 457-465. Linnane, A. W., Marzuki, S., Ozawa, T., and Tanaka, M. (1989) Luncet i, 642-645. Hattori, K., Tanaka, M., Sugiyama, S., Obayashi, T., Ito, T., Satake, T., Hanaki, Y., Asai, J., Nagano, M., and Ozawa, T. (1991) Am. Heart. J. 1735-1742. Torii, K., Sugiyama, S., Tanaka, M., Takagi, K., Hanaki, Y., Iida, K., Matsuyama, M., Hirabayashi, N., Uno, Y., and Ozawa, T. (1991) Am. J. Respir. Cell. Mol. Biol. (in press). Yen, T. C., Su, J. H., King, K. L., and Wei, Y. H. (1991) Biochem. Biophys. Res. Comnwn. 178, 124-131. Linnane, A. W., Baumer, A., Maxwell, R. J., Preston, H., Zhang, C. F., and Marzuki, S. (1990) Biochem. Int. 22, 1067-1076. 919
Vol.
182, No. 2, 1992
13. 14. 15. 16. 17. 18. 19. 20. 21. 22.
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
Wilkie, D., and Fearon, K. (1985) in : Achievement and perspectives of mitochondrial research. Volume II: Biogenesis (Quagliariello, Slater, Palmieri, Saccone and Kroon ed.) pp 437-444, Elsevier, Amsterdam. Taira, M., Yoshida, E., Kobayashi, M., Yaginuma, K., and Koike, K. (1983) Nucleic Acids Res. 11, 1635-1643. Sato, W., Tanaka, M., Ohno, K., Yamamoto, T., Takada, G., and Ozawa, T. (1989) Biochem. Biophys. Res. Commun. 162, 664-672. Ozawa, T., Tanaka, M., Sugiyama, S., Hattori, K., Ito, T., Ohno, K., Takahashi, A., Sato, W., Takada, G., Mayumi, B., Yamamoto, K., Adachi, K., Koga, Y., and Toshima, H. (1990) Biochem. Biophys. Res. Commun. 170, 830-836. Hattori, K., Ogawa, T., Kondo, T., Mochizuki, M., Tanaka, M., Sugiyama, S., Ito, T., Satake, T., and Ozawa, T. (1991) Am. Heart J. i121, 1735-1742. Tanaka, M., Sato, W., Ohno, K., Yamamoto, T., and Ozawa, T. (1989) Biochem. Biophys. Res. Commun. 164, 156-163. Ikebe, S., Tanaka, M., Ohno, K., Sato, W., Hattori, K., Kondo, T., Mizuno, Y., and Ozawa, T. (1990) Biochem. Biophys. Res. Commun. 170, 1044-1048. Sugiyama, S., Hattori, K., Hayakawa, M., and Ozawa, T. (1991) Biochem. Biophys. Res. Commun. 180, 894-899. Hayakawa, M., Torii, K., Sugiyama, S., Tanaka, M., and Ozawa, T. (1991) Biochem. Biophys. Res. Commun. 179, 1023-1029. Diaz, J., Rossi, I., Escartin, P., Segovia, J., and Gosalvez, M. (1977) Clin. Sci. Mol. Med. 52, 599606.
23.
Corral, M., Kitzis, A., Baffet, G., Paris, B., Tichonicky, L., Kruh, J., GuguenGuillouzo, C., and Defer, N. (1989) Nucleic Acids Res. 17, 5191-5206. 24. Wilkie, D., Evans, I. H., Egilsson, V., Diala, E. S., and Collier, D. (1983) Int. Rev. Cytol. St@. 15, 157-89. 25. Welter, C., and Blin, N. (1988) J. Exp. Clin. Cancer Res. 7, 13-20. Monnat, R., Maxwell, C., and Loeb, L. (1985) Cancer Res. 45, 1809-1814. ;76* Vogelstein, B. (1990) Nature 348,681- 682. 28: Shibutani, S., Takeshita, M., and Grollman, A. P. (1991) Nature. 349,431-434. 29. Wood, M. L., Dizdaroglu, M., Gajewski, E., and Essigmann, J. M. (1990) Biochemistry 349, 7024-7032.
30.
Hayakawa, M., Ogawa, T., Sugiyama, S., Tanaka, M., and Ozawa, T. (1991) Biochem. Biophys. Res. Commun. 176, 87-93.
920