- Email: [email protected]
Inhibition of DNA synthesis by downregulation of cyclin A but not Skp 2 overexpression in human hepatocellular carcinoma cells
Cancer Letters 139 (1999) 1±6
Inhibition of DNA synthesis by downregulation of cyclin A but not Skp 2 overexpression in human hepatocellular carcinoma cells Yee Chao a,*, Yung-Luen Shih b, Hsin-Ju Chen b, Shou-Dong Lee a, Tze-Sing Huang b a
Cancer Center and Department of Internal Medicine, Veterans General Hospital (VGH)-Taipei and Institute of Clinical Medicine, National Yang-Ming University, Taipei 112, Taiwan b Cooperative Laboratory at VGH-Taipei, Cancer Research Division, National Health Research Institutes, Taipei 112, Taiwan Received 5 October 1998; received in revised form 9 November 1998; accepted 10 November 1998
Abstract Cyclin A is an S and G2/M phase regulatory protein and associates with Skp 2 in many transformed cells. Our previous results showed that 12 (39%) and 17 (55%) out of 31 hepatocellular carcinoma (HCC) patients exhibited higher protein expression levels of cyclin A and Skp 2, respectively, in their tumorous compared to non-tumorous tissues. In the present study, we used Western blot analysis to show that 3 out of 6 HCC cell lines, HA59T, HA22T and HCC36, exhibited overexpression of cyclin A, among which the HCC36 cell line also expressed a higher Skp 2 protein level. Moreover, we used the antisense oligonucleotide phosphorothioates to down regulate the overexpression of cyclin A and Skp 2 proteins to determine whether or not these two proteins are involved in the mitogenesis of HCC36 cells. After treatment with antisense oligonucleotide phosphorothioates, the gene product of cyclin A or Skp 2 was suppressed dose-dependently as revealed by Western blot analyses. By [ 3H]thymidine incorporation assay, we found that downregulation of cyclin A but not Skp 2 overexpression could inhibit the DNA synthesis ability of HCC36 cells, suggesting that abnormal Skp 2 expression is not directly correlated with the HCC proliferation. q 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Cyclin A; Skp 2; Antisense oligonucleotide phosphorothioate (S-oligo); Hepatocelluar carcinoma (HCC); DNA synthesis
1. Introduction Hepatocellular carcinoma (HCC) is among the most common malignancies in the world. Human cyclin A gene, encoding a nuclear protein with a molecular weight of 60 kDa, was ®rst identi®ed from an integration site of hepatitis B virus DNA in a HCC patient [1]. Cyclin A functions as a regulatory protein required for S and G2/M phases progression via association with CDK 2 and CDC 2 protein kinases, * Corresponding author. Cooperative Laboratory, Veterans General Hospital-Taipei, No. 201, Sec. 2, Shih-Pai Road, Taipei 112, Taiwan. Fax: 1 886-2-28733664.
respectively [2±4]. The cyclin A/CDK 2 regulates cellular DNA replication probably due to its involvement in the phosphorylation of replication protein A [5,6]; it also forms complexes with E2F transcription factor and Rb-family proteins and participates in cell cycle transcriptional control [7±9]. Clinically, 39% of HCC patients exhibited cyclin A overexpression in tumor tissues, and the elevated cyclin A expression was statistically correlated with the increased percentage of tumor cells in the S and G2/M phases [10]. Skp 2 is a cyclin A-associating protein [11]. The level of Skp 2 is greatly increased in many transformed cells [11], and 17 (55%) out of 31 HCC patients also exhibited higher protein levels of Skp 2 in the tumorous
0304-3835/99/$ - see front matter q 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0304-383 5(98)00352-8
2
Y. Chao et al. / Cancer Letters 139 (1999) 1±6
2.3. Oligonucleotide phosphorothioates (S-oligos)
compared to non-tumorous tissues [10]. Therefore, suppression of cyclin A or Skp 2 production might inhibit the growth of HCC cells due to perturbation of the S phase progression. Antisense oligonucleotides are short, basically 15± 25 bases long, single-stranded DNA fragments, which recognize and anneal with the speci®c mRNA as targets. The antisense oligonucleotides can be with the sequence complementary to the region around the initiation codon of target mRNA, and thus interrupt the ribosome/mRNA interaction and protein synthesis [12,13]. In the present studies, we found that the human HCC cell line, HCC36, exhibited the overexpression of both cyclin A and Skp 2 proteins. To explore the relationship of these two proteins with the proliferative activity of HCC36 cells, we used antisense oligonucleotide phosphorothioates to downregulate the overexpression of cyclin A and Skp 2 as well as examine the cellular DNA synthesis level by [ 3H]thymidine incorporation assay. We found that the DNA synthesis activity of human HCC36 cells was suppressed by downregulation of cyclin A but not Skp 2 overexpression by using antisense oligonucleotides. The data supports our previous clinical statistical results that overexpression of cyclin A but not Skp 2 correlated with the proliferative potency and thus the tumor relapse of HCC.
Frozen tumor specimens were minced into minute pieces and further digested into single cell suspensions [14]. Cell suspensions were then ®xed at 2208C and 80% of ethanol for at least 30 min before subsequent Triton X-100 permeabilization and propidium iodide staining [15]. The cellular DNA content was analyzed using a FACStar ¯ow cytometer with an argon laser tuned to the 488-nm line for excitation (Becton-Dickinson).
2. Materials and methods
2.5. Protein extraction and Western blot analysis
2.1. Tumor specimens Tumor specimens were obtained from 31 patients who accepted resection of primary HCC at Veterans General Hospital (VGH)-Taipei during the period 1984±1987. Specimens were frozen in liquid nitrogen immediately after resection. All were kept in liquid nitrogen or in a 2708C freezer until use. 2.2. Cell culture Six HCC cell lines, HCC36, HepG2, Hep3B, Tong, HA22T, and HA59T, were obtained from the Cell Bank of VGH-Taipei. Routinely, these cell lines were cultivated in minimal essential medium (MEM) supplemented with 10% fetal bovine serum (FBS), 2 mM l-glutamine, 100 units/ml penicillin, and 100 mg/ml streptomycin.
The cyclin A and Skp 2-speci®c sense and antisense oligonucleotide phosphorothioates were synthesized and puri®ed by high-performance liquid chromatography by Genset Co. The sequences of oligonucleotide phosphorothioates are as follows: cyclin A sense: 5 0 -GsGAsGCsAGsTGsATsGTsTGsGGsC-3 0 cyclin A antisense: 5 0 -GsCCsCAsACsATsCAsCTsGCsTCsC-3 0 Skp 2 sense: 5 0 -GsCCsTGsCGsGAsCGsCTsATsGCsA-3 0 Skp 2 antisense: 5 0 -TsGCsATsAGsCGsTCsCGsCAsGGsC-3 0 2.4. Flow cytometric analysis
Cell lysates for Western blot analyses were prepared according to the method described previously [16]. Aliquots (30 mg protein) of cell lysates were separated on 10% SDS±polyacrylamide gels, and electrotransferred onto polyvinylidene di¯uoride membranes. After blocking with phosphate-buffered saline plus 0.1% Tween-20 plus 5% non-fat milk, the membranes were incubated with anti-cyclin A, cyclin E, CDK 2, Skp 2 and proliferating cell nuclear antigen (PCNA) antibodies (Santa Cruz Biotechnology), respectively, in PBST plus 5% milk at 48C for 12 h. These membranes were then washed three times with PBST and incubated with horseradish peroxidase-conjugated secondary antibodies for 1 h at room temperature. After washing three times with PBST, the bands representing the antibody-recognized proteins were detected by enhanced chemiluminescence (ECL, Amersham Co.).
Y. Chao et al. / Cancer Letters 139 (1999) 1±6
3
Table 1 Increased S-phase fraction correlated with the poor prognosis of HCC patients
S-phase fraction (%) P a
Disease-free interval (mean: 39:79^40:86 months)
Overall survival period (mean: 51:46^44:34 months)
$ mean
, mean
$ mean
, mean
3.05 ^ 1.82 0.022 a
9.36 ^ 7.54
3.15 ^ 1.74 0.011 a
9.95 ^ 7.74
Results were considered statistically signi®cant if P , 0:05.
2.6. [ 3H]Thymidine incorporation assay The DNA synthesis activity of HCC36 cells was determined by measuring the amount of [ 3H]thymidine incorporation into cellular DNA. Brie¯y, HCC36 cells were incubated with indicated concentrations of cyclin A or Skp 2 antisense or sense oligonucleotide phosphorothioates for 22 h and then pulse-labeled with 1 mCi/ml of [ 3H]thymidine (1 Ci 37 GBq) for another 2 h. Labeled cells were further lysed in
100 mM NaOH, 10 mM EDTA and 0.5% SDS, and mixed with an equal volume of 20% trichloroacetic acid (TCA). The TCA-precipitated macromolecules which were labeled with [ 3H]thymidine were ®ltered using glass ®ber ®lters and the amounts were determined by means of liquid scintillation counting [17]. 3. Results 3.1. Increased S-phase fraction occurred in human HCC specimens We analyzed the S-phase ratio in the tumor cells of 31 HCC patients. The patients whose overall survival periods were shorter than the mean value, 51:46 ^ 44:34 months, exhibited signi®cantly increased Sphase fractions (Table 1; mean, 9:95 ^ 7:74% versus 3:15 ^ 1:74%, P 0:011). If we set a cutoff of Sphase fraction at 6%, the patients who contained the S-phase fractions higher than the cutoff obviously had shorter disease-free intervals (mean, 5:20 ^ 4:97 months versus 49:72 ^ 43:01 months, P 0:000). These data indicate that increased S-phase ratio would account for the poor prognosis of HCC patients. To investigate the molecules involved in the increase of the S-phase fraction in HCC, we chose six HCC cell lines for further studies. 3.2. Elevated cyclin A and Skp 2 expression in HCC cells
Fig. 1. Western blot analyses of cyclin A, cyclin E, CDK 2, Skp 2, and PCNA levels in the lysates of six HCC cell lines, HA59T, HA22T, HCC36, Tong, Hep3B and HepG2.
Five S-phase proteins including cyclin A, cyclin E, Skp 2, PCNA and CDK 2 were examined for their expression levels in six HCC cell lines, HCC36, HA22T, HA59T, Tong, Hep3B and HepG2. As shown in Fig. 1, cyclin A expression could be detected in all six HCC cell lines, among which HA59T,
4
Y. Chao et al. / Cancer Letters 139 (1999) 1±6
Fig. 3. Relative DNA synthesis levels in the HCC36 cells treated with 5 or 10 mM of Skp 2 sense or antisense oligonucleotide phosphorothioates for 22 h, followed by additional [ 3H]thymidine for another 2 h. The data represent the means of three independent experiments. Each experiment was performed in triplicate.
cyclin A and Skp 2 and therefore were chosen for further investigation.
Fig. 2. Inhibition of DNA synthesis in HCC36 cells by downregulation of cyclin A expression. (A) Western blot analyses of cyclin A and CDC 2 levels in the HCC36 cells treated with 1, 5 or 10 mM of cyclin A sense or antisense oligonucleotide phosphorothioates for 24 h. The unchanged level of CDC 2 was noted and used as the internal control. (B) Relative DNA synthesis levels in the HCC36 cells treated with 2.5, 5 or 10 mM of cyclin A sense or antisense oligonucleotide phosphorothioates for 22 h, followed by additional [ 3H]thymidine for another 2 h. The data represent the means of three independent experiments, and each experiment was performed in triplicate.
HA22T and HCC36 cells had much more cyclin A expression than others, if normalized to their CDK 2 levels, respectively. However, cyclin E protein was not so abundant in these three cyclin A-overexpressing cell lines. In contrast to the cyclin A result, cyclin E was expressed mostly in the Tong cells and the highest level of PCNA was exhibited by HepG2 and Hep3B cells (Fig. 1). In addition, higher Skp 2 expression was found only in the HCC36 cells, although we previously reported that 55% of 31 HCC patients had Skp 2 overexpression. Among the six cell lines, only HCC36 cells exhibited elevated expression of both
3.3. Inhibition of HCC36 DNA synthesis by cyclin A but not Skp 2-speci®c antisense oligonucleotide phosphorothioates Antisense oligonucleotide phosphorothioates were used to down regulate the overexpression of cyclin A and Skp 2 and thus to explore the relationship of these two proteins with the proliferative activity of HCC36 cells. Logarithmically growing HCC36 cells were incubated for 24 h with serial concentrations of cyclin A antisense or sense oligonucleotide phosphorothioates. The Western blot result shows that the antisense oligonucleotides suppressed the cyclin A amount in a dose-dependent manner (Fig. 2A). In contrast, the cellular cyclin A level increases when accompanied with increasing concentrations of sense oligonucleotides. By the [ 3H]thymidine incorporation assay, we further observed that the cellular DNA synthesis level was reduced down to 60% when cells were treated with 2.5 mM of cyclin A antisense oligonucleotides (Fig. 2B). In contrast, 2.5 mM of sense oligonucleotides resulted in a 25% increase in [ 3H]thymidine incorporation. When the dosage of antisense oligonucleotides was increased to either 5 or 10 mM, only 30% of the control DNA synthesis level could be
Y. Chao et al. / Cancer Letters 139 (1999) 1±6
detected. Unexpectedly, a 44±48% reduction of DNA synthesis occurred when 5±10 mM of sense oligonucleotides was used, probably because super-overexpression of cyclin A would induce apoptosis in cancer cells (data not shown) [18±20]. On the other hand, when Skp 2 oligonucleotides were used to treat HCC36 cells according to the cyclin A's condition, no signi®cant effect on DNA synthesis was found whether 5 mM of sense or antisense oligonucleotides were used (Fig. 3). However, at 10 mM, 40 and 15% elevation of DNA synthesis was observed in the cells treated with Skp 2 sense and antisense oligonucleotides, respectively. The Skp 2 level was only reduced in the antisense oligonucleotide-treated cells (data not shown). 4. Discussion Uncontrolled cell division cycle is one of the most characteristic features of cancer cells. Among many proteins involved in the regulation of the cell cycle, cyclin A has been found to have promoting activities in DNA replication and cell-cycle transcription control, and thus plays an important role in the progression of both S and G2/M phases [2±4]. In this report, we observed that 3 out of 6 HCC cell lines, HA59T, HA22T and HCC36 exhibited cyclin A overexpression. Our previous investigation and other results using tumor specimens have indicated that 39±42% of HCC patients and 75% of patients with non-small-cell lung cancer showed elevation of cyclin A expression in tumor cells [10,21]. Because no rearrangement of the cyclin A DNA was found and the frequency of allelic loss in the cyclin A gene was relatively low (1/20) in human HCC [22,23], the dysregulation of cyclin A expression can act as an important factor in hepatocarcinogenesis. Our previous report has pinpointed that overexpression of cyclin A correlated with increased S and G2/M phase fractions in human HCC [10]. Moreover, the hepatitis B viral infection and cirrhosis were not factors leading to the tumor recurrence of HCC. The proliferation potency which increased by dysregulated cyclin A was the risk factor linked to tumor relapse after liver resection [10]. Patients with cyclin A overexpression had decreased disease-free interval and median survival time. These data suggest that
5
cyclin A can serve as a good target molecule for antisense gene therapy. In the present study, the DNA synthesis activity of HCC36 cells was reduced after treatment with cyclin A antisense oligonucleotides, which supports the importance of cyclin A in the mitogenesis of cyclin A-overexpressing HCC cells. Although we previously showed that 17 of 31 (55%) HCC patients exhibited higher Skp 2 protein levels, the elevated Skp 2 expression was not correlated with the increase of S-phase cells in tumorous tissues and the patients' decreased disease-free intervals [10]. The present data also demonstrated that the reduction of Skp 2 expression by antisense oligonucleotides had no inhibitory effect on the DNA synthesis activity of HCC36 cells. Although the biochemical function of Skp 2 remains to be determined, these results suggest that abnormal Skp 2 expression has no direct correlation with HCC proliferation. In contrast, it has been reported that many transformed cells have increased Skp 2 expression and microinjection of Skp 2 antibodies or antisense oligonucleotides prevents normal cell cycling, especially entry into S phase [11]. Perhaps, the different outcome from our experiments using Skp 2 antisense oligonucleotides can be attributed to the development of heterogeneity in cancer cells during hepatocarcinogenesis. In addition, we observed that the DNA synthesis level was elevated in cells when treated with a higher dose of Skp 2 sense or antisense phosphorothioate oligonucleotides. The use of higher amounts of phosphorothioate oligonucleotides would increase the possibility of the nuclear entry of phosphorothioate oligonucleotides and thus the formation of nuclear matrix-associated nuclear bodies. Due to the fact that many proteins have been found in the nuclear matrix, many nuclear functions such as transcription and DNA replication could be affected by the formation of nuclear bodies of phosphorothioate oligonucleotides [24]. In summary, abnormal expression of cyclins has been proposed to be associated with the occurrence of several cancers [25,26]. In accordance with the ®nding that suppression of cyclin D1 overexpression can inhibit cancer cell growth and reverse its transformed phenotype in esophageal cancer [27], our results identify the cyclin A molecule as an essential element in the tumorigenesis of some population of HCC. We suggest that cyclin A may be used as a
6
Y. Chao et al. / Cancer Letters 139 (1999) 1±6
suitable target for the exploration of new anti-HCC therapeutics. Acknowledgements This work was partly supported by Grant VGHTaipei No. 300. References [1] J. Wang, X. Chenivesse, B. Henglein, C. Brechot, Hepatitis B virus integration in a cyclin A gene in a hepatocellular carcinoma, Nature (Lond.) 343 (1990) 555±557. [2] D.H. Walker, J.L. Maller, Role of the cyclin A in the dependence of mitosis on completion of DNA replication, Nature (Lond.) 345 (1991) 314±317. [3] M. Pagano, R. Pepperkok, F. Verde, W. Ansorge, G. Draetta, Cyclin A is required at two points in the human cell cycle, EMBO J. 11 (1992) 961±971. [4] J. Rosenblatt, Y. Gu, D.O. Morgan, Human cyclin-dependent kinase 2 is activated during the S and G2 phases of the cell cycle and associates with cyclin A, Proc. Natl. Acad. Sci. USA 89 (1992) 2824±2828. [5] A. Dutta, B. Stillmann, Cdc2 family kinases phosphorylate a human cell DNA replication factor, RPA, and activate DNA replication, EMBO J. 11 (1992) 2189±2199. [6] Y. Xiong, H. Zhang, D. Beach, Subunit rearrangement of cyclin-dependent kinases is associated with cellular transformation, Genes Dev. 7 (1993) 1572±1583. [7] L. Bandara, J. Adamczewski, T. Hunter, N. La Thangue, Cyclin A and retinoblastoma gene product complex with a common transcription factor, Nature (Lond.) 352 (1991) 249±251. [8] M. Mudryj, S. Devoto, S. Hiebert, T. Hunter, J. Pines, J. Nevins, Cell cycle regulation of the E2F transcription factor involves an interaction with cyclin A, Cell 65 (1991) 1243± 1253. [9] S. Shirodkar, M.E. Ewen, J. DeCaprio, J. Morgan, D.M. Livingstone, T. Chittenden, The transcription factor E2F interacts with the retinoblastoma product and a p107-cyclin A complex in a cell cycle regulated manner, Cell 69 (1992) 157±166. [10] Y. Chao, Y.-L. Shih, J.-H. Chiu, G.-Y. Chau, W.-Y. Lui, W.K. Yang, S.-D. Lee, T.-S. Huang, Overexpression of cyclin A but not Skp 2 correlates with the tumor relapse of human hepatocellular carcinoma, Cancer Res. 58 (1998) 985±990. [11] H. Zhang, R. Kobayashi, K. Galaktionov, D. Beach, p19 Skp1
[12] [13] [14]
[15] [16] [17] [18] [19] [20] [21]
[22]
[23] [24] [25] [26] [27]
and p45 Skp2 are essential elements of the cyclin A-CDK2 S phase kinase, Cell 82 (1995) 915±925. J.L. Tonkinson, C.A. Stein, Antisense oligodeoxynucleotides as clinical therapeutic agents, Cancer Invest. 14 (1996) 54±65. L.J. Maher III, Prospects for the therapeutic use of antisense oligonucleotides, Cancer Invest. 14 (1996) 66±82. M.-H. Chen, W.K. Yang, J. Whang-Peng, L.-S. Lee, T.-S. Huang, Differential inducibilities of GFAP expression, cytostasis and apoptosis in primary cultures of human astrocytic tumors, Apoptosis 3 (1998) 171±182. T.-S. Huang, W.K. Yang, J. Whang-Peng, GL331-induced disruption of cyclin B1/CDC 2 complex and inhibition of CDC2 kinase activity, Apoptosis 1 (1996) 213±217. T.-S. Huang, C.-H. Shu, W.K. Yang, J. Whang-Peng, Activation of CDC 25 phosphatase and CDC 2 kinase involved in GL331-induced apoptosis, Cancer Res. 57 (1997) 2974±2978. T.-S. Huang, J. Duyster, J.Y.J. Wang, Biological response to phorbol ester determined by alternative G1 pathways, Proc. Natl. Acad. Sci. USA 92 (1995) 4793±4797. W. Meikrantz, S. Gisselbrecht, S.W. Tam, R. Schlegel, Activation of cyclin A-dependent protein kinases during apoptosis, Proc. Natl. Acad. Sci. USA 91 (1994) 3754±3758. A.T. Hoang, K.J. Cohen, J.F. Barrett, D.A. Bergstrom, C.V. Dang, Participation of cyclin A in Myc-induced apoptosis, Proc. Natl. Acad. Sci. USA 91 (1994) 6875±6879. C.J. Li, D.J. Friedman, C. Wang, V. Metelev, A.B. Pardee, Induction of apoptosis in uninfected lymphocytes by HIV-1 Tat protein, Science 268 (1995) 429±431. M. Volm, R. Koomagi, J. Mattern, G. Stammler, Cyclin A is associated with an unfavorable outcome in patients with nonsmall-cell lung carcinomas, Br. J. Cancer 75 (1997) 1774± 1778. P. Paterlini, J.-F. Flejou, M.S. De Mitri, E. Pisi, D. Franco, C. Brechot, Structure and expression of the cyclin A gene in human primary liver cancer. Correlation with ¯ow cytometric parameters, J. Hepatol. 23 (1995) 47±52. M.S. De Mitri, E. Pisi, C. Brechot, P. Paterlini, Low frequency of allelic loss in the cyclin A gene in human hepatocellular carcinomas: a study based on PCR, Liver 13 (1993) 259±261. P. Lorenz, B.F. Baker, C.F. Bennett, D.L. Spector, Phosphorothioate antisense oligonucleotides induce the formation of nuclear bodies, Mol. Biol. Cell 9 (1998) 1007±1023. T. Hunter, J. Pines, Cyclins and cancer, Cell 66 (1991) 1071± 1074. T. Hunter, J. Pines, Cyclins and cancer II: cyclin D and CDK inhibitors come of age, Cell 79 (1994) 573±582. P. Zhou, W. Jiang, Y. Zhang, S.M. Kahn, I. Schieren, R.M. Santella, I.B. Weinstein, Antisense to cyclin D1 inhibits growth and reverses the transformed phenotype of human esophageal cancer cells, Oncogene 11 (1995) 571±580.
Inhibition of DNA synthesis by downregulation of cyclin A but not Skp 2 overexpression in human hepatocellular carcinoma cells Yee Chao a,*, Yung-Luen Shih b, Hsin-Ju Chen b, Shou-Dong Lee a, Tze-Sing Huang b a
Cancer Center and Department of Internal Medicine, Veterans General Hospital (VGH)-Taipei and Institute of Clinical Medicine, National Yang-Ming University, Taipei 112, Taiwan b Cooperative Laboratory at VGH-Taipei, Cancer Research Division, National Health Research Institutes, Taipei 112, Taiwan Received 5 October 1998; received in revised form 9 November 1998; accepted 10 November 1998
Abstract Cyclin A is an S and G2/M phase regulatory protein and associates with Skp 2 in many transformed cells. Our previous results showed that 12 (39%) and 17 (55%) out of 31 hepatocellular carcinoma (HCC) patients exhibited higher protein expression levels of cyclin A and Skp 2, respectively, in their tumorous compared to non-tumorous tissues. In the present study, we used Western blot analysis to show that 3 out of 6 HCC cell lines, HA59T, HA22T and HCC36, exhibited overexpression of cyclin A, among which the HCC36 cell line also expressed a higher Skp 2 protein level. Moreover, we used the antisense oligonucleotide phosphorothioates to down regulate the overexpression of cyclin A and Skp 2 proteins to determine whether or not these two proteins are involved in the mitogenesis of HCC36 cells. After treatment with antisense oligonucleotide phosphorothioates, the gene product of cyclin A or Skp 2 was suppressed dose-dependently as revealed by Western blot analyses. By [ 3H]thymidine incorporation assay, we found that downregulation of cyclin A but not Skp 2 overexpression could inhibit the DNA synthesis ability of HCC36 cells, suggesting that abnormal Skp 2 expression is not directly correlated with the HCC proliferation. q 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Cyclin A; Skp 2; Antisense oligonucleotide phosphorothioate (S-oligo); Hepatocelluar carcinoma (HCC); DNA synthesis
1. Introduction Hepatocellular carcinoma (HCC) is among the most common malignancies in the world. Human cyclin A gene, encoding a nuclear protein with a molecular weight of 60 kDa, was ®rst identi®ed from an integration site of hepatitis B virus DNA in a HCC patient [1]. Cyclin A functions as a regulatory protein required for S and G2/M phases progression via association with CDK 2 and CDC 2 protein kinases, * Corresponding author. Cooperative Laboratory, Veterans General Hospital-Taipei, No. 201, Sec. 2, Shih-Pai Road, Taipei 112, Taiwan. Fax: 1 886-2-28733664.
respectively [2±4]. The cyclin A/CDK 2 regulates cellular DNA replication probably due to its involvement in the phosphorylation of replication protein A [5,6]; it also forms complexes with E2F transcription factor and Rb-family proteins and participates in cell cycle transcriptional control [7±9]. Clinically, 39% of HCC patients exhibited cyclin A overexpression in tumor tissues, and the elevated cyclin A expression was statistically correlated with the increased percentage of tumor cells in the S and G2/M phases [10]. Skp 2 is a cyclin A-associating protein [11]. The level of Skp 2 is greatly increased in many transformed cells [11], and 17 (55%) out of 31 HCC patients also exhibited higher protein levels of Skp 2 in the tumorous
0304-3835/99/$ - see front matter q 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0304-383 5(98)00352-8
2
Y. Chao et al. / Cancer Letters 139 (1999) 1±6
2.3. Oligonucleotide phosphorothioates (S-oligos)
compared to non-tumorous tissues [10]. Therefore, suppression of cyclin A or Skp 2 production might inhibit the growth of HCC cells due to perturbation of the S phase progression. Antisense oligonucleotides are short, basically 15± 25 bases long, single-stranded DNA fragments, which recognize and anneal with the speci®c mRNA as targets. The antisense oligonucleotides can be with the sequence complementary to the region around the initiation codon of target mRNA, and thus interrupt the ribosome/mRNA interaction and protein synthesis [12,13]. In the present studies, we found that the human HCC cell line, HCC36, exhibited the overexpression of both cyclin A and Skp 2 proteins. To explore the relationship of these two proteins with the proliferative activity of HCC36 cells, we used antisense oligonucleotide phosphorothioates to downregulate the overexpression of cyclin A and Skp 2 as well as examine the cellular DNA synthesis level by [ 3H]thymidine incorporation assay. We found that the DNA synthesis activity of human HCC36 cells was suppressed by downregulation of cyclin A but not Skp 2 overexpression by using antisense oligonucleotides. The data supports our previous clinical statistical results that overexpression of cyclin A but not Skp 2 correlated with the proliferative potency and thus the tumor relapse of HCC.
Frozen tumor specimens were minced into minute pieces and further digested into single cell suspensions [14]. Cell suspensions were then ®xed at 2208C and 80% of ethanol for at least 30 min before subsequent Triton X-100 permeabilization and propidium iodide staining [15]. The cellular DNA content was analyzed using a FACStar ¯ow cytometer with an argon laser tuned to the 488-nm line for excitation (Becton-Dickinson).
2. Materials and methods
2.5. Protein extraction and Western blot analysis
2.1. Tumor specimens Tumor specimens were obtained from 31 patients who accepted resection of primary HCC at Veterans General Hospital (VGH)-Taipei during the period 1984±1987. Specimens were frozen in liquid nitrogen immediately after resection. All were kept in liquid nitrogen or in a 2708C freezer until use. 2.2. Cell culture Six HCC cell lines, HCC36, HepG2, Hep3B, Tong, HA22T, and HA59T, were obtained from the Cell Bank of VGH-Taipei. Routinely, these cell lines were cultivated in minimal essential medium (MEM) supplemented with 10% fetal bovine serum (FBS), 2 mM l-glutamine, 100 units/ml penicillin, and 100 mg/ml streptomycin.
The cyclin A and Skp 2-speci®c sense and antisense oligonucleotide phosphorothioates were synthesized and puri®ed by high-performance liquid chromatography by Genset Co. The sequences of oligonucleotide phosphorothioates are as follows: cyclin A sense: 5 0 -GsGAsGCsAGsTGsATsGTsTGsGGsC-3 0 cyclin A antisense: 5 0 -GsCCsCAsACsATsCAsCTsGCsTCsC-3 0 Skp 2 sense: 5 0 -GsCCsTGsCGsGAsCGsCTsATsGCsA-3 0 Skp 2 antisense: 5 0 -TsGCsATsAGsCGsTCsCGsCAsGGsC-3 0 2.4. Flow cytometric analysis
Cell lysates for Western blot analyses were prepared according to the method described previously [16]. Aliquots (30 mg protein) of cell lysates were separated on 10% SDS±polyacrylamide gels, and electrotransferred onto polyvinylidene di¯uoride membranes. After blocking with phosphate-buffered saline plus 0.1% Tween-20 plus 5% non-fat milk, the membranes were incubated with anti-cyclin A, cyclin E, CDK 2, Skp 2 and proliferating cell nuclear antigen (PCNA) antibodies (Santa Cruz Biotechnology), respectively, in PBST plus 5% milk at 48C for 12 h. These membranes were then washed three times with PBST and incubated with horseradish peroxidase-conjugated secondary antibodies for 1 h at room temperature. After washing three times with PBST, the bands representing the antibody-recognized proteins were detected by enhanced chemiluminescence (ECL, Amersham Co.).
Y. Chao et al. / Cancer Letters 139 (1999) 1±6
3
Table 1 Increased S-phase fraction correlated with the poor prognosis of HCC patients
S-phase fraction (%) P a
Disease-free interval (mean: 39:79^40:86 months)
Overall survival period (mean: 51:46^44:34 months)
$ mean
, mean
$ mean
, mean
3.05 ^ 1.82 0.022 a
9.36 ^ 7.54
3.15 ^ 1.74 0.011 a
9.95 ^ 7.74
Results were considered statistically signi®cant if P , 0:05.
2.6. [ 3H]Thymidine incorporation assay The DNA synthesis activity of HCC36 cells was determined by measuring the amount of [ 3H]thymidine incorporation into cellular DNA. Brie¯y, HCC36 cells were incubated with indicated concentrations of cyclin A or Skp 2 antisense or sense oligonucleotide phosphorothioates for 22 h and then pulse-labeled with 1 mCi/ml of [ 3H]thymidine (1 Ci 37 GBq) for another 2 h. Labeled cells were further lysed in
100 mM NaOH, 10 mM EDTA and 0.5% SDS, and mixed with an equal volume of 20% trichloroacetic acid (TCA). The TCA-precipitated macromolecules which were labeled with [ 3H]thymidine were ®ltered using glass ®ber ®lters and the amounts were determined by means of liquid scintillation counting [17]. 3. Results 3.1. Increased S-phase fraction occurred in human HCC specimens We analyzed the S-phase ratio in the tumor cells of 31 HCC patients. The patients whose overall survival periods were shorter than the mean value, 51:46 ^ 44:34 months, exhibited signi®cantly increased Sphase fractions (Table 1; mean, 9:95 ^ 7:74% versus 3:15 ^ 1:74%, P 0:011). If we set a cutoff of Sphase fraction at 6%, the patients who contained the S-phase fractions higher than the cutoff obviously had shorter disease-free intervals (mean, 5:20 ^ 4:97 months versus 49:72 ^ 43:01 months, P 0:000). These data indicate that increased S-phase ratio would account for the poor prognosis of HCC patients. To investigate the molecules involved in the increase of the S-phase fraction in HCC, we chose six HCC cell lines for further studies. 3.2. Elevated cyclin A and Skp 2 expression in HCC cells
Fig. 1. Western blot analyses of cyclin A, cyclin E, CDK 2, Skp 2, and PCNA levels in the lysates of six HCC cell lines, HA59T, HA22T, HCC36, Tong, Hep3B and HepG2.
Five S-phase proteins including cyclin A, cyclin E, Skp 2, PCNA and CDK 2 were examined for their expression levels in six HCC cell lines, HCC36, HA22T, HA59T, Tong, Hep3B and HepG2. As shown in Fig. 1, cyclin A expression could be detected in all six HCC cell lines, among which HA59T,
4
Y. Chao et al. / Cancer Letters 139 (1999) 1±6
Fig. 3. Relative DNA synthesis levels in the HCC36 cells treated with 5 or 10 mM of Skp 2 sense or antisense oligonucleotide phosphorothioates for 22 h, followed by additional [ 3H]thymidine for another 2 h. The data represent the means of three independent experiments. Each experiment was performed in triplicate.
cyclin A and Skp 2 and therefore were chosen for further investigation.
Fig. 2. Inhibition of DNA synthesis in HCC36 cells by downregulation of cyclin A expression. (A) Western blot analyses of cyclin A and CDC 2 levels in the HCC36 cells treated with 1, 5 or 10 mM of cyclin A sense or antisense oligonucleotide phosphorothioates for 24 h. The unchanged level of CDC 2 was noted and used as the internal control. (B) Relative DNA synthesis levels in the HCC36 cells treated with 2.5, 5 or 10 mM of cyclin A sense or antisense oligonucleotide phosphorothioates for 22 h, followed by additional [ 3H]thymidine for another 2 h. The data represent the means of three independent experiments, and each experiment was performed in triplicate.
HA22T and HCC36 cells had much more cyclin A expression than others, if normalized to their CDK 2 levels, respectively. However, cyclin E protein was not so abundant in these three cyclin A-overexpressing cell lines. In contrast to the cyclin A result, cyclin E was expressed mostly in the Tong cells and the highest level of PCNA was exhibited by HepG2 and Hep3B cells (Fig. 1). In addition, higher Skp 2 expression was found only in the HCC36 cells, although we previously reported that 55% of 31 HCC patients had Skp 2 overexpression. Among the six cell lines, only HCC36 cells exhibited elevated expression of both
3.3. Inhibition of HCC36 DNA synthesis by cyclin A but not Skp 2-speci®c antisense oligonucleotide phosphorothioates Antisense oligonucleotide phosphorothioates were used to down regulate the overexpression of cyclin A and Skp 2 and thus to explore the relationship of these two proteins with the proliferative activity of HCC36 cells. Logarithmically growing HCC36 cells were incubated for 24 h with serial concentrations of cyclin A antisense or sense oligonucleotide phosphorothioates. The Western blot result shows that the antisense oligonucleotides suppressed the cyclin A amount in a dose-dependent manner (Fig. 2A). In contrast, the cellular cyclin A level increases when accompanied with increasing concentrations of sense oligonucleotides. By the [ 3H]thymidine incorporation assay, we further observed that the cellular DNA synthesis level was reduced down to 60% when cells were treated with 2.5 mM of cyclin A antisense oligonucleotides (Fig. 2B). In contrast, 2.5 mM of sense oligonucleotides resulted in a 25% increase in [ 3H]thymidine incorporation. When the dosage of antisense oligonucleotides was increased to either 5 or 10 mM, only 30% of the control DNA synthesis level could be
Y. Chao et al. / Cancer Letters 139 (1999) 1±6
detected. Unexpectedly, a 44±48% reduction of DNA synthesis occurred when 5±10 mM of sense oligonucleotides was used, probably because super-overexpression of cyclin A would induce apoptosis in cancer cells (data not shown) [18±20]. On the other hand, when Skp 2 oligonucleotides were used to treat HCC36 cells according to the cyclin A's condition, no signi®cant effect on DNA synthesis was found whether 5 mM of sense or antisense oligonucleotides were used (Fig. 3). However, at 10 mM, 40 and 15% elevation of DNA synthesis was observed in the cells treated with Skp 2 sense and antisense oligonucleotides, respectively. The Skp 2 level was only reduced in the antisense oligonucleotide-treated cells (data not shown). 4. Discussion Uncontrolled cell division cycle is one of the most characteristic features of cancer cells. Among many proteins involved in the regulation of the cell cycle, cyclin A has been found to have promoting activities in DNA replication and cell-cycle transcription control, and thus plays an important role in the progression of both S and G2/M phases [2±4]. In this report, we observed that 3 out of 6 HCC cell lines, HA59T, HA22T and HCC36 exhibited cyclin A overexpression. Our previous investigation and other results using tumor specimens have indicated that 39±42% of HCC patients and 75% of patients with non-small-cell lung cancer showed elevation of cyclin A expression in tumor cells [10,21]. Because no rearrangement of the cyclin A DNA was found and the frequency of allelic loss in the cyclin A gene was relatively low (1/20) in human HCC [22,23], the dysregulation of cyclin A expression can act as an important factor in hepatocarcinogenesis. Our previous report has pinpointed that overexpression of cyclin A correlated with increased S and G2/M phase fractions in human HCC [10]. Moreover, the hepatitis B viral infection and cirrhosis were not factors leading to the tumor recurrence of HCC. The proliferation potency which increased by dysregulated cyclin A was the risk factor linked to tumor relapse after liver resection [10]. Patients with cyclin A overexpression had decreased disease-free interval and median survival time. These data suggest that
5
cyclin A can serve as a good target molecule for antisense gene therapy. In the present study, the DNA synthesis activity of HCC36 cells was reduced after treatment with cyclin A antisense oligonucleotides, which supports the importance of cyclin A in the mitogenesis of cyclin A-overexpressing HCC cells. Although we previously showed that 17 of 31 (55%) HCC patients exhibited higher Skp 2 protein levels, the elevated Skp 2 expression was not correlated with the increase of S-phase cells in tumorous tissues and the patients' decreased disease-free intervals [10]. The present data also demonstrated that the reduction of Skp 2 expression by antisense oligonucleotides had no inhibitory effect on the DNA synthesis activity of HCC36 cells. Although the biochemical function of Skp 2 remains to be determined, these results suggest that abnormal Skp 2 expression has no direct correlation with HCC proliferation. In contrast, it has been reported that many transformed cells have increased Skp 2 expression and microinjection of Skp 2 antibodies or antisense oligonucleotides prevents normal cell cycling, especially entry into S phase [11]. Perhaps, the different outcome from our experiments using Skp 2 antisense oligonucleotides can be attributed to the development of heterogeneity in cancer cells during hepatocarcinogenesis. In addition, we observed that the DNA synthesis level was elevated in cells when treated with a higher dose of Skp 2 sense or antisense phosphorothioate oligonucleotides. The use of higher amounts of phosphorothioate oligonucleotides would increase the possibility of the nuclear entry of phosphorothioate oligonucleotides and thus the formation of nuclear matrix-associated nuclear bodies. Due to the fact that many proteins have been found in the nuclear matrix, many nuclear functions such as transcription and DNA replication could be affected by the formation of nuclear bodies of phosphorothioate oligonucleotides [24]. In summary, abnormal expression of cyclins has been proposed to be associated with the occurrence of several cancers [25,26]. In accordance with the ®nding that suppression of cyclin D1 overexpression can inhibit cancer cell growth and reverse its transformed phenotype in esophageal cancer [27], our results identify the cyclin A molecule as an essential element in the tumorigenesis of some population of HCC. We suggest that cyclin A may be used as a
6
Y. Chao et al. / Cancer Letters 139 (1999) 1±6
suitable target for the exploration of new anti-HCC therapeutics. Acknowledgements This work was partly supported by Grant VGHTaipei No. 300. References [1] J. Wang, X. Chenivesse, B. Henglein, C. Brechot, Hepatitis B virus integration in a cyclin A gene in a hepatocellular carcinoma, Nature (Lond.) 343 (1990) 555±557. [2] D.H. Walker, J.L. Maller, Role of the cyclin A in the dependence of mitosis on completion of DNA replication, Nature (Lond.) 345 (1991) 314±317. [3] M. Pagano, R. Pepperkok, F. Verde, W. Ansorge, G. Draetta, Cyclin A is required at two points in the human cell cycle, EMBO J. 11 (1992) 961±971. [4] J. Rosenblatt, Y. Gu, D.O. Morgan, Human cyclin-dependent kinase 2 is activated during the S and G2 phases of the cell cycle and associates with cyclin A, Proc. Natl. Acad. Sci. USA 89 (1992) 2824±2828. [5] A. Dutta, B. Stillmann, Cdc2 family kinases phosphorylate a human cell DNA replication factor, RPA, and activate DNA replication, EMBO J. 11 (1992) 2189±2199. [6] Y. Xiong, H. Zhang, D. Beach, Subunit rearrangement of cyclin-dependent kinases is associated with cellular transformation, Genes Dev. 7 (1993) 1572±1583. [7] L. Bandara, J. Adamczewski, T. Hunter, N. La Thangue, Cyclin A and retinoblastoma gene product complex with a common transcription factor, Nature (Lond.) 352 (1991) 249±251. [8] M. Mudryj, S. Devoto, S. Hiebert, T. Hunter, J. Pines, J. Nevins, Cell cycle regulation of the E2F transcription factor involves an interaction with cyclin A, Cell 65 (1991) 1243± 1253. [9] S. Shirodkar, M.E. Ewen, J. DeCaprio, J. Morgan, D.M. Livingstone, T. Chittenden, The transcription factor E2F interacts with the retinoblastoma product and a p107-cyclin A complex in a cell cycle regulated manner, Cell 69 (1992) 157±166. [10] Y. Chao, Y.-L. Shih, J.-H. Chiu, G.-Y. Chau, W.-Y. Lui, W.K. Yang, S.-D. Lee, T.-S. Huang, Overexpression of cyclin A but not Skp 2 correlates with the tumor relapse of human hepatocellular carcinoma, Cancer Res. 58 (1998) 985±990. [11] H. Zhang, R. Kobayashi, K. Galaktionov, D. Beach, p19 Skp1
[12] [13] [14]
[15] [16] [17] [18] [19] [20] [21]
[22]
[23] [24] [25] [26] [27]
and p45 Skp2 are essential elements of the cyclin A-CDK2 S phase kinase, Cell 82 (1995) 915±925. J.L. Tonkinson, C.A. Stein, Antisense oligodeoxynucleotides as clinical therapeutic agents, Cancer Invest. 14 (1996) 54±65. L.J. Maher III, Prospects for the therapeutic use of antisense oligonucleotides, Cancer Invest. 14 (1996) 66±82. M.-H. Chen, W.K. Yang, J. Whang-Peng, L.-S. Lee, T.-S. Huang, Differential inducibilities of GFAP expression, cytostasis and apoptosis in primary cultures of human astrocytic tumors, Apoptosis 3 (1998) 171±182. T.-S. Huang, W.K. Yang, J. Whang-Peng, GL331-induced disruption of cyclin B1/CDC 2 complex and inhibition of CDC2 kinase activity, Apoptosis 1 (1996) 213±217. T.-S. Huang, C.-H. Shu, W.K. Yang, J. Whang-Peng, Activation of CDC 25 phosphatase and CDC 2 kinase involved in GL331-induced apoptosis, Cancer Res. 57 (1997) 2974±2978. T.-S. Huang, J. Duyster, J.Y.J. Wang, Biological response to phorbol ester determined by alternative G1 pathways, Proc. Natl. Acad. Sci. USA 92 (1995) 4793±4797. W. Meikrantz, S. Gisselbrecht, S.W. Tam, R. Schlegel, Activation of cyclin A-dependent protein kinases during apoptosis, Proc. Natl. Acad. Sci. USA 91 (1994) 3754±3758. A.T. Hoang, K.J. Cohen, J.F. Barrett, D.A. Bergstrom, C.V. Dang, Participation of cyclin A in Myc-induced apoptosis, Proc. Natl. Acad. Sci. USA 91 (1994) 6875±6879. C.J. Li, D.J. Friedman, C. Wang, V. Metelev, A.B. Pardee, Induction of apoptosis in uninfected lymphocytes by HIV-1 Tat protein, Science 268 (1995) 429±431. M. Volm, R. Koomagi, J. Mattern, G. Stammler, Cyclin A is associated with an unfavorable outcome in patients with nonsmall-cell lung carcinomas, Br. J. Cancer 75 (1997) 1774± 1778. P. Paterlini, J.-F. Flejou, M.S. De Mitri, E. Pisi, D. Franco, C. Brechot, Structure and expression of the cyclin A gene in human primary liver cancer. Correlation with ¯ow cytometric parameters, J. Hepatol. 23 (1995) 47±52. M.S. De Mitri, E. Pisi, C. Brechot, P. Paterlini, Low frequency of allelic loss in the cyclin A gene in human hepatocellular carcinomas: a study based on PCR, Liver 13 (1993) 259±261. P. Lorenz, B.F. Baker, C.F. Bennett, D.L. Spector, Phosphorothioate antisense oligonucleotides induce the formation of nuclear bodies, Mol. Biol. Cell 9 (1998) 1007±1023. T. Hunter, J. Pines, Cyclins and cancer, Cell 66 (1991) 1071± 1074. T. Hunter, J. Pines, Cyclins and cancer II: cyclin D and CDK inhibitors come of age, Cell 79 (1994) 573±582. P. Zhou, W. Jiang, Y. Zhang, S.M. Kahn, I. Schieren, R.M. Santella, I.B. Weinstein, Antisense to cyclin D1 inhibits growth and reverses the transformed phenotype of human esophageal cancer cells, Oncogene 11 (1995) 571±580.