- Email: [email protected]
DNA hypomethylation of proliferating cell nuclear antigen gene in human hepatocellular carcinoma is not due to cell proliferation
Cancer Letters, 70 (1993) 189- 196 Elsevier Scientific Publishers Ireland
189 Ltd.
DNA hypomethylation of proliferating cell nuclear antigen gene in human hepatocellular carcinoma is not due to cell proliferation Yie-Wen Liu”, King-Jen Changb and Yin-Chang Liu” ‘Institute of Life Science, National Tsing-Hua University, Hsin-Chu, and hDepartment of Surgery, College of Medicine. National Taiwan University, Taipei (Taiwan) (Received 25 January 1993) (Revision received 15 March (Accepted 22 March 1993)
1993)
Summary
Introduction
Proliferating cell nuclear antigen (PCNA) gene expresses preferentially in proliferative cells or tissues. The levels of PCNA mRNA are very low in livers of adult mammals. Expression of PCNA gene in hepatocellular carcinoma tissues was, however, elevated; and 5’-CCGG-3 ’ sequences of the gene in neoplastic tissue were less methylated. Such DNA hypomethylation was concluded, on the basis of two observations, not to be due to the cell proliferation in hepatoma tissues. First, while the expression of PCNA was increased during serum-stimulation of quiescent Hep G2 cells, the DNA methylation pattern of PCNA gene remained unchanged. Second, in rat liver regeneration, the PCNA mRNA level rose and declined, but the DNA methylation status of PCNA gene was unaltered. Therefore, the DNA hypomethylation of the PCNA gene found in hepatocellular carcinoma was not due to cell proliferation, but a possible consequence of cell transformation.
Proliferation cell nuclear antigen (PCNA), also known as cyclin, is a highly conserved non-histone nuclear protein among eucaryotes from yeast to human (for review see [2,3,7]). It is required for quiescent cells to re-enter the cell cycle [ 10,131 and may participate in cellular DNA replication (201 or DNA excision repair 117,181 as an auxiliary protein of DNA polymerase 6 [14,19]. Expression of PCNA gene is well correlated with the proliferation of cells and tissues [ 1,12,15]. Liver of adult mammals such as rat and human, although active in metabolism, is not proliferative and has been found to hardly express PCNA transcription [12,21]. To explore the mechanism that suppresses the PCNA gene transcription in this tissue, we have been studying the gene expression and gene structure in human hepatocellular carcinoma. In this report, we showed that PCNA mRNA level was elevated in human hepatoma tissue, which was accompanied with the apparent decrease of DNA methylation in the PCNA gene. This change in DNA methylation was however, not observed when quiescent hepatocytes were serum stimulated to start proliferation.
Keywords: proliferation cell nuclear antigen; DNA hypomethylation; hepatocellular carcinoma; cell proliferation
Correspondence to: Yin-Chang Liu, Institute of Life Science. National Tsing-Hua University. Hsin-Chu, Taiwan 30043.
0304-3835/93/$06.00 0 1993 Elsevier Scientific Printed and Published in Ireland
Publishers
Ireland
Materials and Methods Tissues
The pathologically-identified Ltd
human
hepato-
obtained from the cell culture center at Taipei Veterans General Hospital and cultured in DMEM supplemented with 10% fetal calf serum. Hep G2 cells were grown to confluency and made quiescent by serum starvation for 4 days in DMEM with 0.5% fetal calf serum. Serum stimulation of quiescent Hep G2 cells were carried out by replacing the serum-deficient DMEM with the medium containing 10% serum.
cellular carcinoma tissues and portions of their neighboring tissues were surgically resected from patients of National Taiwan University Hospital. The tissues were frozen immediately in liquid N:! and stored at -70°C before the subsequent experiments. Animal and partial hepatectomy
Male Wistar rats weighing about 150-180 g were obtained from the Animal Care Center of National Taiwan University Hospital. The animals were anesthetized by ether and partially hepatectomized as described by Higgins and Anderson 191. Portions of regenerating liver tissue specimen were dissected at 21 and 96 h after partial hepatectomy. Resected tissues of regenerating livers and normal livers from the respective rats were washed briefly in phosphate-buffered saline, frozen immediately in liquid Nz and stored at -70°C before the subsequent experiments.
Probes and probe labeling
The human PCNA cDNA clone, Sd [l] was obtained from Dr. R. Bravo (Princeton University, NJ). The rat PCNA cDNA clone was prepared as described in [13]. The probes were 32P-labeled with a random primer labeling system (product of Promega Co.) based on the method described by Feinberg and Volgelstein [8]. RNA isolation and Northern blot analysis
Total cellular RNAs were isolated from tissues and cells according to the procedures described in [S]. To extract RNA from liver tissues, the frozen
Cell culture
Hep G2 cells, a human hepatoma cell line, were
A.
1
2
3
4
5
6
6.
1
2
3
4
5
6
Fig. 1. Northern blot analysis of PCNA mRNA in three pairs of human hepatocellular carcinoma tissues along with their respective non-cancerous liver tissues. (A) The mRNA levels of PCNA in the hepatoma (I, 3, 5) and their neighboring non-cancerous tissues (2,4,6). The probe used in this experiment was linearized human PCNA cDNA clone, S4. (B) The ethidium bromide staining patterns of RNA samples respective to those shown in (A) after electrophoretic separation in I”%agaroselformaldehyde gel. Each sample contained approx. 20 pg RNA (estimated by A2m).
j
‘s
;
6
1’ ‘2
3’ ‘4
;
zj
M
2.3 2.2
4.4
6.6
9.4
‘23,l
kb 1’1
E
/p1\ ii
H 5’8
M ;ici
E
/p2\ H 4;
M
-
2,3 2,2
-23.1
kb
carcinoma tissues (Pl and P2) and their respective non-cancerous liver tissues. Fig. 2. Southern blot analysis of PCNA gene structures in two human hepatocellular (A) Autoradiography of probed genomic DNAs extracted from various tissues and completely digested with restriction enzymes as follows: E - EroRI; H - HP&; M - MspI. Lanes 1, 3, 5: hepatoma DNAs. Lanes 2, 4, 6: non-cancerous liver DNAs. The probe used in the analysis was linearized human PCNA cDNA clone. S4. The DNA size marker in kilobase-pair was indicated in the right. (B) The ethidium bromide staining patterns of DNA samples respective to those shown in (A) after electrophoretic separation in 0.8% agarose gel. Each sample contained approx. 20 pg DNA (estimated by A?&,
A
1’ ‘2
H
E
H
M
P2
PI
192
tissues were first pulverized with a stainless steel mortar and homogenized in denaturing solution (4 M guanidine isothiocyanate, 25 mM sodium citrate (pH 7.0); 0.5% sarcosyl, 0.1 M P-mercaptoethanol) with a polytron. Approximately 20 pg of total RNAs (according to the absorbance values at 260 nm) from each sample were used and electrophoretically separated in a 1% agarose/formaldehyde gel. The RNAs were subsequently transferred to a nylon membrane (Hybond-N from Amersham Co.) and hybridized with the radiolabeled probe. The procedures of hybridization, washing and autoradiography were conducted as described in [ 121. DNA isolation and Southern blot analysis
Genomic DNAs of liver tissues were isolated by the method described by Davis et al. [6], which was derived from the procedure described by Chirgwin et al. [4]. Briefly, the frozen tissues were pulverized as described previously and homogenized in denaturing buffer (4 M guanidine isothiocyanate, 0.1 M /3-mercaptoethanol, 25 mM sodium acetate, pH 6.0). The homogenate was placed on top of CsCl buffer (5.7 M CsCl, 25 mM sodium acetate, pH 6.0) and spun overnight at 35 000 rev./min at 21°C. After centrifugation, the DNA containing portion (approx. lower than one-third of the liquid content) was transferred to a new container for further purification with ethanol precipitation and phenol/chloroform extraction. Portions of genomic DNA (approx. 20 pg as estimated by the absorbance at 260 nm) were completely digested with EcoRI (Promega Co.), HpaII (Promega Co.), and MspI (BioLab Co.) at 37°C for 24 h. DNA fragments were electrophoretically separated on 0.8% agarose and then blotted onto Nylon membrane. 32P-labeled human or rat PCNA cDNAs were used as probes. Procedures of hybridization, washing and autoradiography were performed as described in [ 121. Results and Discussion
Expression of PCNA gene in liver tissues from patients with hepatocellular carcinoma was studied by Northern blot analyses. Thirteen different paired samples (normal and neoplastic),
besides those which did not produce RNA of good quality for examination, were found to have elevated PCNA mRNA levels in cancerous tissues as exemplified in a Northern analysis shown in Fig. 1. To examine the chromosomal alteration in the PCNA gene, genomic DNAs were extracted from the tissues of interest and subjected to Southern blot analysis using the restriction enzymes EcoRI, &a11 and MspI. A typical analysis was shown in Fig. 2. Different restriction length patterns between cancerous and non-cancerous livers were produced using the HpaII enzyme (see lanes 3 and 4 of both Pl and P2, Fig. 2A). HpaII enzyme recognizes 5’-CCGG-3 ‘, but is unable to cleave the sequence when the internal cytosine is methylated. More HpaII bands were found in the cancerous than in the non-cancerous tissues,
A.
PCNA6.
28S18SFig. 3. Northern blot analysis of PCNA mRNA in a human hepatoma cell line Hep G2. (A) The mRNA levels of PCNA in the Hep G2 cell line. Lane I: non-synchronous Hep G2 cells. Lane 2: quiescent Hep G2 cells synchronized by serum deprivation for 4 days. Lane 3: Hep G2 cells synchronized by serum deprivation for 4 days and serum stimulated for 24 h. The probe used in this experiment was linearized human PCNA cDNA clone, S4. (B) The ethidium bromide staining patterns of RNA samples respective to those shown in (A) after electrophoretic separation in 1% agarose/formaldehyde gel. Each sample contained approx. 20 pg RNA (estimated by A,,a).
H
M
-
2,3 2,2
- 23.1
kb
123123123
E H M
23,l
2,3 2,2
-
-
kb
Fig. 4. Southern blot analysis of PCNA gene structure in a human hepatoma cell line Hep G2 during quiescent and activated states. (A) Autoradiography of probed genomic DNAs extracted from various states of Hep G2 cells and fully digested by restriction enzymes: E - EcoRI; H - Hpdl; M - MspI. Lane 1: non-synchronous Hep G2 cells. Lane 2: quiescent Hep G2 cells synchronized by serum deprivation for 4 days. Lane 3: Hep G2 cells synchronized by serum deprivation for 4 days and serum stimulated for 24 h. The probe used in the analysis was linearized human PCNA clone, S4. The DNA size marker in kilobase-pair was indicated in the right. (B) The ethidium bromide staining patterns of DNA samples respective to those shown in (A) after electrophoretic separation in 0.8% agarose gel. Each sample contained approx. 20 pg DNA (estimated by A,,).
A
123123123
E
194
indicating that PCNA gene in the neoplastic tissue is less methylated than its non-cancerous counterpart. Further support for this hypothesis was shown by the same restriction pattern in both tissues when MspI was used (see lanes 5 and 6 of both Pl and P2, Fig. 2A). MS@ recognizes the same sequence as HpaII but is not sensitive to cytosine methylation. Since the hepatoma tissue was considered to be more proliferative than the part of non-cancerous liver, the elevation of PCNA gene expression and the alteration of DNA methylation might be due to cell proliferation. To test this view, methylation in the 5’-CCGG-3 ’ sequence of PCNA gene during the serum stimulation of the quiescent Hep G2 cell line was studied. The Hep G2 cell line is a transformed human hepatocyte cell line. Expression of PCNA mRNA in the serum-stimulated (24 h) Hep G2 cells was significantly elevated (see lanes 2 and 3 of Fig. 3). The HpaII restriction patterns of PCNA gene were, however, not changed (see lanes 3 and 4 of Fig. 4A). The 5’-CCGG-3 ’ sequences were methylated as HpaII and it@1 did not produce identical patterns (see lanes 3,4, 5 and 6 of Fig. 4A). Such absence of difference in HpaII patterns was, therefore, not due to complete demethylation in the 5 ‘-CCGG-3 ’ sequences of the PCNA gene. The results of Hep G2 cells indicated that cell proliferation does not require (or cause) demethylation of PCNA gene. To confirm this observation, a similar analysis was conducted with regenerating rat livers. Liver of adult animal is quiescent; however, it can resume a proliferative state by partial hepatectomy. Figure 5 shows the expression of PCNA during the course of rat liver regeneration following partial hepatectomy, indicating that PCNA mRNA level gradually increased and declined back after reaching a peak level at 21 h post-surgery. The HpaII restriction patterns of PCNA gene during the same course were, however, unchanged (lanes 3 and 4 of Fig. 6A). This result, in accordance with the previous one, indicates that rat liver regeneration does not require (or produce) demethylation of PCNA gene. Therefore, it is clear that the hypomethylation occurring to PCNA gene in hepatocellular carcinoma was not due to proliferation, but probably related to the state of cell transformation.
In consistence with the previous study that PCNA antigen was increased in small hepatocellular carcinoma [16], the study showed that expression of PCNA gene was elevated in human hepatocellular carcinoma and the gene was hypomethylated. The neoplasm of the hepatoma samples used herein was also verified by hepatitis B virus integration in liver genomic DNA and chromosome alteration in ~53 genes (results not shown). DNA hypomethylation in many genes has been found in various kinds of human cancers [ll]. In contrast to conventional thought that DNA methylation is related to gene activity, our results, however, imply that the DNA methylation is not involved in the activation of the PCNA gene.
12
3
4
PCNA-
B.
1234 28S-c 18SFig. 5. Northern blot analysis of PCNA mRNA expression in rat regenerating livers. (A) Upper: The PCNA mRNA levels of normal liver (lane I), regenerating livers resected at 21 h (lane 2) as well as 96 h (lane 3) after partial hepatectomy, and normal spleen (lane 4). The probe used in this analysis was rat PCNA cDNA. Lower: The ethidium bromide staining patterns of RNA samples respective to those shown in Upper after electrophoretic separation in 1% agarose/formaldehyde gel. Each sample contained approx. 20 pg RNA (estimated by A&. (B) A similar analysis as carried out in (A) with more defined time course. I: normal liver. 2, 3,4, 5: regenerating livers resected at I2 h, 21 h, 48 h and 72 h. 6: normal spleen. The PCNA mRNA in spleen was used for positive control [l2].
195
E
H
123123123
M
E kb
123123123
H
M kb
-7,l -6,l -5,l -41 -3,0 -2,0 -1,6
-2,0 -1.6 -1.0
Fig. 6. Southern blot analysis of PCNA gene structures in regenerating livers of Wistar rats. (A) autoradiography of genomic DNAs extracted from normal liver (lane 1) as well as regenerating livers resected at 21 (lane 2) and 96 (lane 3) h, respectively after partial hepatectomy. Extracted DNAs were fully digested by restriction enzymes: E - EcoRI; H - HpaII; M - Mspl. The probe used in this analysis was rat PCNA cDNA. The DNA size marker in kilobase-pair was indicated in the right. (B) The ethidium bromide staining patterns of DNA samples respective to those shown in (A) after electrophoretic separation in 0.8”/ agarose gel. Each sample contained approx. 20 pg DNA (estimated by A,,).
In fact, we found that actin gene was hypomethylated in human hepatocellular carcinema, and yet the expression of this gene was indifferent between normal and cancerous liver in tissues. Therefore, DNA hypomethylation PCNA was more likely to be a consequence of cell transformation.
References 1 Almendral,
2 3
Acknowledgements We are grateful to Dr. R. Bravo for providing human PCNA cDNA and to Mr. Ken Mueller for English editing on this manuscript. This study was supported by the research grant NSCSI-021 lB007-08 from National Science Council, R.O.C.
J.M.. Huebsh, D., Blundell, P.A., MacDonald-Bravo, H. and Bravo, R. (1987) Cloning and sequence of the human nuclear protein cychn: homology with DNA-binding proteins. Proc. Nat]. Acad. Sci., 84. 1575-1579. Baserga, R. (1991) Growth regulation of the PCNA gene. J. Cell Sci., 98, 433-436. Celis, J.E., Madsen, P., Celis, A., Nielsen, H.V. and Gesser, B. (1987) Cychn (PCNA, auxiliary protein of DNA polymerase 6) is a central component of the pathway(s) leading to DNA replication and cell division. FEBS Lett., 220, 1-7. Chirgwin, J.M., Przybyla, A.E., MacDonald, R.J. and Rutter, W.J. (1979) Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry, 18, 5294-5299.
196 Chomczynski, P. and Sacchi, N. (1987) Single-step method of RNA isolation by acid guanidium thiocyanatephenol-chloroform extraction. Anal. Biochem., 162. 156-159. Davis, L.C., Dibner, M.D. and Battey, J.F. (1986) Guanidine isothiocyanate preparation of total RNA. In: Basic Methods in Molecular Biology, pp. 130-135. Editors: L.G. Davis, M.D. Dibner and J.F. Battey. Elsevier, New York. Fairman, M.P. (1990) DNA polymerase UPCNA: actions and interactions. J. Cell Sci., 95, 1-4. Feinberg, A.P. and Volgelstein, B. (1983) A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem., 132, 6-13. Higgins, G.H. and Anderson, R.M. (1931) Experimental pathology of the liver. I. Restoration of liver of white rat following partial surgical removal. Arch. Pathol., 12, 186-202. 10 Jaskulski, D., DeRiel, K., Mercer, W.E., Calabretta, B. and Baserga, R. (1988) Inhibition of cellular proliferation by antisense oligodeoxynucleotide to PCNAicyclin. Science, 240, 1S44- 1546. I1 Jones, P.A. and Buckley, J.D. (1990) The role of DNA methylation in cancer. Adv. Cancer Res. 54. l-27. 12 Liu, Y.C. and Bambara, R.A. (1989) Gene expression of PCNA/cyclin in adult tissues and the R3230AC mammary tumor of rat. B&hem. Biophys. Res. Commun., 161, 873-882. 13 Liu, Y.C., Marracino, R.L., Keng, P.C., Bambara, R.A.. Lord, E.M., Chow, W.G. and Zain, S.B. (1989) Requirement for proliferating cell nuclear antigen expressed during stages of the Chinese hamster ovary cell cycle. Biochemistry, 28, 2967-2974.
14 Mathews, M.B., Bernstein, R.M., Franza. R.R. and Garrels, J.I. (1984) Identity of the proliferating cell nuclear antigen and cyclin. Nature (Lond.), 309, 374-376. IS Matsumoto, K., Moriuchi, T., Koji. T. and Nakane, P.K. (1987) Molecular cloning of cDNA coding for rat proliferating cell nuclear antigen (PCNA)/cyclin. EMBO J., 6, 637-642. 16 Matsuno, Y., Hirohashi, S., Furuya, S., Sakamoto, M., Mukai, K. and Shimosato, Y. (1990) Heterogeneity of proliferative activity in nodule-in-nodule lesions of small hepatocellular carcinoma. Jpn. J. Cancer Res., 81. 1137-1140. 17 Nichols, A.F. and Sancar, A. (1992) Purification of PCNA as a nucleotide excision repair protein. Nucleic Acids Res.. 20, 2441-2446. 18 Shivji, M.K.K., Kenny. M.K. and Wood, R.D. (1992) Proliferating cell nuclear antigen is required for DNA excision repair. Cell, 69, 367-374. 19 Tan, C.K., Castillo. C., So, A.G. and Downey, K.M. (1986) An auxiliary protein for DNA polymerase 6 from fetal calf thymus. J. Biol. Chem., 261, 12310-12316. 20 Tsurimoto, T., Melendy, T. and Stillman, B. (1990) Sequential initiation of lagging and leading strand synthesis by two different polymerase complexes at the SV40 DNA replication origin. Nature (Land.). 346, 534-539. 21 Yamaguchi, M., Hayashi, Y., Hirose. F., Matsoka, S., Moriuchi, T., Shiroishi, T., Moriwaki, K. and Matsukage, A. (1991) Molecular cloning and structural analysis of mouse gene and pseudogenes for proliferating cell nuclear antigen. Nucleic Acid Res.. 19, 2403-2410.
189 Ltd.
DNA hypomethylation of proliferating cell nuclear antigen gene in human hepatocellular carcinoma is not due to cell proliferation Yie-Wen Liu”, King-Jen Changb and Yin-Chang Liu” ‘Institute of Life Science, National Tsing-Hua University, Hsin-Chu, and hDepartment of Surgery, College of Medicine. National Taiwan University, Taipei (Taiwan) (Received 25 January 1993) (Revision received 15 March (Accepted 22 March 1993)
1993)
Summary
Introduction
Proliferating cell nuclear antigen (PCNA) gene expresses preferentially in proliferative cells or tissues. The levels of PCNA mRNA are very low in livers of adult mammals. Expression of PCNA gene in hepatocellular carcinoma tissues was, however, elevated; and 5’-CCGG-3 ’ sequences of the gene in neoplastic tissue were less methylated. Such DNA hypomethylation was concluded, on the basis of two observations, not to be due to the cell proliferation in hepatoma tissues. First, while the expression of PCNA was increased during serum-stimulation of quiescent Hep G2 cells, the DNA methylation pattern of PCNA gene remained unchanged. Second, in rat liver regeneration, the PCNA mRNA level rose and declined, but the DNA methylation status of PCNA gene was unaltered. Therefore, the DNA hypomethylation of the PCNA gene found in hepatocellular carcinoma was not due to cell proliferation, but a possible consequence of cell transformation.
Proliferation cell nuclear antigen (PCNA), also known as cyclin, is a highly conserved non-histone nuclear protein among eucaryotes from yeast to human (for review see [2,3,7]). It is required for quiescent cells to re-enter the cell cycle [ 10,131 and may participate in cellular DNA replication (201 or DNA excision repair 117,181 as an auxiliary protein of DNA polymerase 6 [14,19]. Expression of PCNA gene is well correlated with the proliferation of cells and tissues [ 1,12,15]. Liver of adult mammals such as rat and human, although active in metabolism, is not proliferative and has been found to hardly express PCNA transcription [12,21]. To explore the mechanism that suppresses the PCNA gene transcription in this tissue, we have been studying the gene expression and gene structure in human hepatocellular carcinoma. In this report, we showed that PCNA mRNA level was elevated in human hepatoma tissue, which was accompanied with the apparent decrease of DNA methylation in the PCNA gene. This change in DNA methylation was however, not observed when quiescent hepatocytes were serum stimulated to start proliferation.
Keywords: proliferation cell nuclear antigen; DNA hypomethylation; hepatocellular carcinoma; cell proliferation
Correspondence to: Yin-Chang Liu, Institute of Life Science. National Tsing-Hua University. Hsin-Chu, Taiwan 30043.
0304-3835/93/$06.00 0 1993 Elsevier Scientific Printed and Published in Ireland
Publishers
Ireland
Materials and Methods Tissues
The pathologically-identified Ltd
human
hepato-
obtained from the cell culture center at Taipei Veterans General Hospital and cultured in DMEM supplemented with 10% fetal calf serum. Hep G2 cells were grown to confluency and made quiescent by serum starvation for 4 days in DMEM with 0.5% fetal calf serum. Serum stimulation of quiescent Hep G2 cells were carried out by replacing the serum-deficient DMEM with the medium containing 10% serum.
cellular carcinoma tissues and portions of their neighboring tissues were surgically resected from patients of National Taiwan University Hospital. The tissues were frozen immediately in liquid N:! and stored at -70°C before the subsequent experiments. Animal and partial hepatectomy
Male Wistar rats weighing about 150-180 g were obtained from the Animal Care Center of National Taiwan University Hospital. The animals were anesthetized by ether and partially hepatectomized as described by Higgins and Anderson 191. Portions of regenerating liver tissue specimen were dissected at 21 and 96 h after partial hepatectomy. Resected tissues of regenerating livers and normal livers from the respective rats were washed briefly in phosphate-buffered saline, frozen immediately in liquid Nz and stored at -70°C before the subsequent experiments.
Probes and probe labeling
The human PCNA cDNA clone, Sd [l] was obtained from Dr. R. Bravo (Princeton University, NJ). The rat PCNA cDNA clone was prepared as described in [13]. The probes were 32P-labeled with a random primer labeling system (product of Promega Co.) based on the method described by Feinberg and Volgelstein [8]. RNA isolation and Northern blot analysis
Total cellular RNAs were isolated from tissues and cells according to the procedures described in [S]. To extract RNA from liver tissues, the frozen
Cell culture
Hep G2 cells, a human hepatoma cell line, were
A.
1
2
3
4
5
6
6.
1
2
3
4
5
6
Fig. 1. Northern blot analysis of PCNA mRNA in three pairs of human hepatocellular carcinoma tissues along with their respective non-cancerous liver tissues. (A) The mRNA levels of PCNA in the hepatoma (I, 3, 5) and their neighboring non-cancerous tissues (2,4,6). The probe used in this experiment was linearized human PCNA cDNA clone, S4. (B) The ethidium bromide staining patterns of RNA samples respective to those shown in (A) after electrophoretic separation in I”%agaroselformaldehyde gel. Each sample contained approx. 20 pg RNA (estimated by A2m).
j
‘s
;
6
1’ ‘2
3’ ‘4
;
zj
M
2.3 2.2
4.4
6.6
9.4
‘23,l
kb 1’1
E
/p1\ ii
H 5’8
M ;ici
E
/p2\ H 4;
M
-
2,3 2,2
-23.1
kb
carcinoma tissues (Pl and P2) and their respective non-cancerous liver tissues. Fig. 2. Southern blot analysis of PCNA gene structures in two human hepatocellular (A) Autoradiography of probed genomic DNAs extracted from various tissues and completely digested with restriction enzymes as follows: E - EroRI; H - HP&; M - MspI. Lanes 1, 3, 5: hepatoma DNAs. Lanes 2, 4, 6: non-cancerous liver DNAs. The probe used in the analysis was linearized human PCNA cDNA clone. S4. The DNA size marker in kilobase-pair was indicated in the right. (B) The ethidium bromide staining patterns of DNA samples respective to those shown in (A) after electrophoretic separation in 0.8% agarose gel. Each sample contained approx. 20 pg DNA (estimated by A?&,
A
1’ ‘2
H
E
H
M
P2
PI
192
tissues were first pulverized with a stainless steel mortar and homogenized in denaturing solution (4 M guanidine isothiocyanate, 25 mM sodium citrate (pH 7.0); 0.5% sarcosyl, 0.1 M P-mercaptoethanol) with a polytron. Approximately 20 pg of total RNAs (according to the absorbance values at 260 nm) from each sample were used and electrophoretically separated in a 1% agarose/formaldehyde gel. The RNAs were subsequently transferred to a nylon membrane (Hybond-N from Amersham Co.) and hybridized with the radiolabeled probe. The procedures of hybridization, washing and autoradiography were conducted as described in [ 121. DNA isolation and Southern blot analysis
Genomic DNAs of liver tissues were isolated by the method described by Davis et al. [6], which was derived from the procedure described by Chirgwin et al. [4]. Briefly, the frozen tissues were pulverized as described previously and homogenized in denaturing buffer (4 M guanidine isothiocyanate, 0.1 M /3-mercaptoethanol, 25 mM sodium acetate, pH 6.0). The homogenate was placed on top of CsCl buffer (5.7 M CsCl, 25 mM sodium acetate, pH 6.0) and spun overnight at 35 000 rev./min at 21°C. After centrifugation, the DNA containing portion (approx. lower than one-third of the liquid content) was transferred to a new container for further purification with ethanol precipitation and phenol/chloroform extraction. Portions of genomic DNA (approx. 20 pg as estimated by the absorbance at 260 nm) were completely digested with EcoRI (Promega Co.), HpaII (Promega Co.), and MspI (BioLab Co.) at 37°C for 24 h. DNA fragments were electrophoretically separated on 0.8% agarose and then blotted onto Nylon membrane. 32P-labeled human or rat PCNA cDNAs were used as probes. Procedures of hybridization, washing and autoradiography were performed as described in [ 121. Results and Discussion
Expression of PCNA gene in liver tissues from patients with hepatocellular carcinoma was studied by Northern blot analyses. Thirteen different paired samples (normal and neoplastic),
besides those which did not produce RNA of good quality for examination, were found to have elevated PCNA mRNA levels in cancerous tissues as exemplified in a Northern analysis shown in Fig. 1. To examine the chromosomal alteration in the PCNA gene, genomic DNAs were extracted from the tissues of interest and subjected to Southern blot analysis using the restriction enzymes EcoRI, &a11 and MspI. A typical analysis was shown in Fig. 2. Different restriction length patterns between cancerous and non-cancerous livers were produced using the HpaII enzyme (see lanes 3 and 4 of both Pl and P2, Fig. 2A). HpaII enzyme recognizes 5’-CCGG-3 ‘, but is unable to cleave the sequence when the internal cytosine is methylated. More HpaII bands were found in the cancerous than in the non-cancerous tissues,
A.
PCNA6.
28S18SFig. 3. Northern blot analysis of PCNA mRNA in a human hepatoma cell line Hep G2. (A) The mRNA levels of PCNA in the Hep G2 cell line. Lane I: non-synchronous Hep G2 cells. Lane 2: quiescent Hep G2 cells synchronized by serum deprivation for 4 days. Lane 3: Hep G2 cells synchronized by serum deprivation for 4 days and serum stimulated for 24 h. The probe used in this experiment was linearized human PCNA cDNA clone, S4. (B) The ethidium bromide staining patterns of RNA samples respective to those shown in (A) after electrophoretic separation in 1% agarose/formaldehyde gel. Each sample contained approx. 20 pg RNA (estimated by A,,a).
H
M
-
2,3 2,2
- 23.1
kb
123123123
E H M
23,l
2,3 2,2
-
-
kb
Fig. 4. Southern blot analysis of PCNA gene structure in a human hepatoma cell line Hep G2 during quiescent and activated states. (A) Autoradiography of probed genomic DNAs extracted from various states of Hep G2 cells and fully digested by restriction enzymes: E - EcoRI; H - Hpdl; M - MspI. Lane 1: non-synchronous Hep G2 cells. Lane 2: quiescent Hep G2 cells synchronized by serum deprivation for 4 days. Lane 3: Hep G2 cells synchronized by serum deprivation for 4 days and serum stimulated for 24 h. The probe used in the analysis was linearized human PCNA clone, S4. The DNA size marker in kilobase-pair was indicated in the right. (B) The ethidium bromide staining patterns of DNA samples respective to those shown in (A) after electrophoretic separation in 0.8% agarose gel. Each sample contained approx. 20 pg DNA (estimated by A,,).
A
123123123
E
194
indicating that PCNA gene in the neoplastic tissue is less methylated than its non-cancerous counterpart. Further support for this hypothesis was shown by the same restriction pattern in both tissues when MspI was used (see lanes 5 and 6 of both Pl and P2, Fig. 2A). MS@ recognizes the same sequence as HpaII but is not sensitive to cytosine methylation. Since the hepatoma tissue was considered to be more proliferative than the part of non-cancerous liver, the elevation of PCNA gene expression and the alteration of DNA methylation might be due to cell proliferation. To test this view, methylation in the 5’-CCGG-3 ’ sequence of PCNA gene during the serum stimulation of the quiescent Hep G2 cell line was studied. The Hep G2 cell line is a transformed human hepatocyte cell line. Expression of PCNA mRNA in the serum-stimulated (24 h) Hep G2 cells was significantly elevated (see lanes 2 and 3 of Fig. 3). The HpaII restriction patterns of PCNA gene were, however, not changed (see lanes 3 and 4 of Fig. 4A). The 5’-CCGG-3 ’ sequences were methylated as HpaII and it@1 did not produce identical patterns (see lanes 3,4, 5 and 6 of Fig. 4A). Such absence of difference in HpaII patterns was, therefore, not due to complete demethylation in the 5 ‘-CCGG-3 ’ sequences of the PCNA gene. The results of Hep G2 cells indicated that cell proliferation does not require (or cause) demethylation of PCNA gene. To confirm this observation, a similar analysis was conducted with regenerating rat livers. Liver of adult animal is quiescent; however, it can resume a proliferative state by partial hepatectomy. Figure 5 shows the expression of PCNA during the course of rat liver regeneration following partial hepatectomy, indicating that PCNA mRNA level gradually increased and declined back after reaching a peak level at 21 h post-surgery. The HpaII restriction patterns of PCNA gene during the same course were, however, unchanged (lanes 3 and 4 of Fig. 6A). This result, in accordance with the previous one, indicates that rat liver regeneration does not require (or produce) demethylation of PCNA gene. Therefore, it is clear that the hypomethylation occurring to PCNA gene in hepatocellular carcinoma was not due to proliferation, but probably related to the state of cell transformation.
In consistence with the previous study that PCNA antigen was increased in small hepatocellular carcinoma [16], the study showed that expression of PCNA gene was elevated in human hepatocellular carcinoma and the gene was hypomethylated. The neoplasm of the hepatoma samples used herein was also verified by hepatitis B virus integration in liver genomic DNA and chromosome alteration in ~53 genes (results not shown). DNA hypomethylation in many genes has been found in various kinds of human cancers [ll]. In contrast to conventional thought that DNA methylation is related to gene activity, our results, however, imply that the DNA methylation is not involved in the activation of the PCNA gene.
12
3
4
PCNA-
B.
1234 28S-c 18SFig. 5. Northern blot analysis of PCNA mRNA expression in rat regenerating livers. (A) Upper: The PCNA mRNA levels of normal liver (lane I), regenerating livers resected at 21 h (lane 2) as well as 96 h (lane 3) after partial hepatectomy, and normal spleen (lane 4). The probe used in this analysis was rat PCNA cDNA. Lower: The ethidium bromide staining patterns of RNA samples respective to those shown in Upper after electrophoretic separation in 1% agarose/formaldehyde gel. Each sample contained approx. 20 pg RNA (estimated by A&. (B) A similar analysis as carried out in (A) with more defined time course. I: normal liver. 2, 3,4, 5: regenerating livers resected at I2 h, 21 h, 48 h and 72 h. 6: normal spleen. The PCNA mRNA in spleen was used for positive control [l2].
195
E
H
123123123
M
E kb
123123123
H
M kb
-7,l -6,l -5,l -41 -3,0 -2,0 -1,6
-2,0 -1.6 -1.0
Fig. 6. Southern blot analysis of PCNA gene structures in regenerating livers of Wistar rats. (A) autoradiography of genomic DNAs extracted from normal liver (lane 1) as well as regenerating livers resected at 21 (lane 2) and 96 (lane 3) h, respectively after partial hepatectomy. Extracted DNAs were fully digested by restriction enzymes: E - EcoRI; H - HpaII; M - Mspl. The probe used in this analysis was rat PCNA cDNA. The DNA size marker in kilobase-pair was indicated in the right. (B) The ethidium bromide staining patterns of DNA samples respective to those shown in (A) after electrophoretic separation in 0.8”/ agarose gel. Each sample contained approx. 20 pg DNA (estimated by A,,).
In fact, we found that actin gene was hypomethylated in human hepatocellular carcinema, and yet the expression of this gene was indifferent between normal and cancerous liver in tissues. Therefore, DNA hypomethylation PCNA was more likely to be a consequence of cell transformation.
References 1 Almendral,
2 3
Acknowledgements We are grateful to Dr. R. Bravo for providing human PCNA cDNA and to Mr. Ken Mueller for English editing on this manuscript. This study was supported by the research grant NSCSI-021 lB007-08 from National Science Council, R.O.C.
J.M.. Huebsh, D., Blundell, P.A., MacDonald-Bravo, H. and Bravo, R. (1987) Cloning and sequence of the human nuclear protein cychn: homology with DNA-binding proteins. Proc. Nat]. Acad. Sci., 84. 1575-1579. Baserga, R. (1991) Growth regulation of the PCNA gene. J. Cell Sci., 98, 433-436. Celis, J.E., Madsen, P., Celis, A., Nielsen, H.V. and Gesser, B. (1987) Cychn (PCNA, auxiliary protein of DNA polymerase 6) is a central component of the pathway(s) leading to DNA replication and cell division. FEBS Lett., 220, 1-7. Chirgwin, J.M., Przybyla, A.E., MacDonald, R.J. and Rutter, W.J. (1979) Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry, 18, 5294-5299.
196 Chomczynski, P. and Sacchi, N. (1987) Single-step method of RNA isolation by acid guanidium thiocyanatephenol-chloroform extraction. Anal. Biochem., 162. 156-159. Davis, L.C., Dibner, M.D. and Battey, J.F. (1986) Guanidine isothiocyanate preparation of total RNA. In: Basic Methods in Molecular Biology, pp. 130-135. Editors: L.G. Davis, M.D. Dibner and J.F. Battey. Elsevier, New York. Fairman, M.P. (1990) DNA polymerase UPCNA: actions and interactions. J. Cell Sci., 95, 1-4. Feinberg, A.P. and Volgelstein, B. (1983) A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem., 132, 6-13. Higgins, G.H. and Anderson, R.M. (1931) Experimental pathology of the liver. I. Restoration of liver of white rat following partial surgical removal. Arch. Pathol., 12, 186-202. 10 Jaskulski, D., DeRiel, K., Mercer, W.E., Calabretta, B. and Baserga, R. (1988) Inhibition of cellular proliferation by antisense oligodeoxynucleotide to PCNAicyclin. Science, 240, 1S44- 1546. I1 Jones, P.A. and Buckley, J.D. (1990) The role of DNA methylation in cancer. Adv. Cancer Res. 54. l-27. 12 Liu, Y.C. and Bambara, R.A. (1989) Gene expression of PCNA/cyclin in adult tissues and the R3230AC mammary tumor of rat. B&hem. Biophys. Res. Commun., 161, 873-882. 13 Liu, Y.C., Marracino, R.L., Keng, P.C., Bambara, R.A.. Lord, E.M., Chow, W.G. and Zain, S.B. (1989) Requirement for proliferating cell nuclear antigen expressed during stages of the Chinese hamster ovary cell cycle. Biochemistry, 28, 2967-2974.
14 Mathews, M.B., Bernstein, R.M., Franza. R.R. and Garrels, J.I. (1984) Identity of the proliferating cell nuclear antigen and cyclin. Nature (Lond.), 309, 374-376. IS Matsumoto, K., Moriuchi, T., Koji. T. and Nakane, P.K. (1987) Molecular cloning of cDNA coding for rat proliferating cell nuclear antigen (PCNA)/cyclin. EMBO J., 6, 637-642. 16 Matsuno, Y., Hirohashi, S., Furuya, S., Sakamoto, M., Mukai, K. and Shimosato, Y. (1990) Heterogeneity of proliferative activity in nodule-in-nodule lesions of small hepatocellular carcinoma. Jpn. J. Cancer Res., 81. 1137-1140. 17 Nichols, A.F. and Sancar, A. (1992) Purification of PCNA as a nucleotide excision repair protein. Nucleic Acids Res.. 20, 2441-2446. 18 Shivji, M.K.K., Kenny. M.K. and Wood, R.D. (1992) Proliferating cell nuclear antigen is required for DNA excision repair. Cell, 69, 367-374. 19 Tan, C.K., Castillo. C., So, A.G. and Downey, K.M. (1986) An auxiliary protein for DNA polymerase 6 from fetal calf thymus. J. Biol. Chem., 261, 12310-12316. 20 Tsurimoto, T., Melendy, T. and Stillman, B. (1990) Sequential initiation of lagging and leading strand synthesis by two different polymerase complexes at the SV40 DNA replication origin. Nature (Land.). 346, 534-539. 21 Yamaguchi, M., Hayashi, Y., Hirose. F., Matsoka, S., Moriuchi, T., Shiroishi, T., Moriwaki, K. and Matsukage, A. (1991) Molecular cloning and structural analysis of mouse gene and pseudogenes for proliferating cell nuclear antigen. Nucleic Acid Res.. 19, 2403-2410.