Frequent epigenetic inactivation of the RASSF1A gene in hepatocellular carcinoma

Article metrics


Aberrant promoter methylation is a fundamental mechanism of inactivation of tumor suppressor genes in cancer. The Ras association domain family 1A gene (RASSF1A) is frequently epigenetically silenced in several types of human solid tumors. In this study, we have investigated the expression and methylation status of the RASSF1A gene in hepatocellular carcinoma (HCC). In two HCC cell lines (HepG2 and Hep3B) RASSF1A was inactivated and treatment of these cell lines with a DNA methylation inhibitor reactivated the transcription of RASSF1A. The methylation status of the RASSF1A promoter region was analysed in 26 primary liver tissues including HCC, hepatocellular adenoma (HCA), liver fibrosis, hepatocirrhosis. Out of 15, 14 (93%) HCC were methylated at the RASSF1A CpG island and hypermethylation was independent of hepatitis virus infection. RASSF1A was also methylated in two out of two fibrosis and in three (75%) out of four cirrhosis; the latter carries an increased risk of developing HCC. Additionally, we analysed the methylation status of p16INK4a and other cancer-related genes in the same liver tumors. Aberrant methylation in the HCC samples was detected in 71% of samples for p16, 25% for TIMP3, 17% for PTEN, 13% for CDH1, and 7% for RARβ2. In conclusion, our results demonstrate that RASSF1A and p16INK4a inactivation by methylation are frequent events in hepatocellular carcinoma, but not in HCA, which is in contrast to HCC without cirrhosis, viral hepatitis, storage diseases, or genetic background. Therefore, this study gives additional evidence against a progression of adenoma to carcinoma in the liver. Thus, RASSF1A hypermethylation could be useful as a marker of malignancy and to distinguish between the distinct forms of highly differentiated liver neoplasm.


Hepatocellular carcinoma (HCC) is one of the most frequent malignancies worldwide, with Eastern Asia and sub-Saharan Africa being the most prevalent regions (Schafer and Sorrell, 1999; Flemming et al., 2001). Chronic exposure to aflatoxin B1 is thought to be a major etiological mechanism in these geographic regions. Epidemiologic data indicate that hepatitis B (HBV) and C (HCV) virus are major risk factors for the development of HCC (Schafer and Sorrell, 1999). In Europe and the United States chronic alcoholism may contribute to hepatocellular carcinogenesis. The development of HCC is a multistep process, which is also suggested by histological studies (Feitelson et al., 2002). HCC appears within the context of chronic viral hepatitis, cirrhosis, and distinct storage diseases as tyrosinemia, glycogenoses, and genetic hemochromatosis.

Inactivation of tumor suppressor genes occurs by mutations, loss of heterozygosity (LOH), and/or epigenetic silencing (Esteller and Herman, 2002; Jones and Baylin, 2002). In HCC, inactivation of p16 (Wong et al., 1999; Shen et al., 2002) and GSTP1 gene (Zhong et al., 2002) is common. Aberrant promoter methylation and LOH were detected in noncancerous hepatitis and cirrhosis from patients with HCC (Kondo et al., 2000; Roncalli et al., 2002). In HCC, LOH on several chromosomal arms including 1p, 4q, 5q, 6q, 8p, 10q, 11p, 13q, 16q, and 17p has been reported (Boige et al., 1997; Marchio et al., 1997; Wilkens et al., 2001) and in recent studies, LOH on chromosome 3p was identified in cirrhosis and HCC (Roncalli et al., 2000; Li et al., 2001). In several types of human carcinoma, LOH of 3p is one of the most frequent and earliest events in the pathogenesis of cancer (Kok et al., 1997; Wistuba et al., 2000).

Recently, we and others have cloned and characterized the Ras association domain family 1A gene (RASSF1) from the lung cancer tumor suppressor locus 3p21.3 (Dammann et al., 2000; Lerman and Minna, 2000; Burbee et al., 2001). The two main forms (RASSF1A and RASSF1C) are transcribed from distinct CpG island promoters. The RASSF1A promoter is intensively hypermethylated in a variety of primary human cancers, for example, in lung, breast, ovarian, and thyroid carcinomas (Dammann et al., 2000,2001a,2001b; Agathanggelou et al., 2001; Burbee et al., 2001; Lehmann et al., 2002; Schagdarsurengin et al., 2002). Reinsertion of RASSF1A into different cancer cell lines inhibits tumorigenicity in vitro and in vivo (Dammann et al., 2000; Burbee et al., 2001). The biological function of RASSF1A is still under investigation. First, it was shown that RASSF1C binds Ras and overexpression of RASSF1C induces apoptosis (Vos et al., 2000). However, our data indicated that RASSF1 binds to Ras only very weakly (Ortiz-Vega et al., 2002) and requires heterodimerization with the Ras effector NORE1 to stimulate a novel Ras-regulated proapoptotic pathway (Khokhlatchev et al., 2002). A recent report suggests, that RASSF1A blocks cell cycle progression by inhibiting cyclin D1 accumulation and cannot induce any apoptotic response (Shivakumar et al., 2002). However, which of these functional pathways may inhibit tumorigenesis remains to be determined.

In this study, we analysed the expression of RASSF1A in HCC cell lines and we used bisulfite methylation-specific PCR (MSP) to investigate the aberrant hypermethylation of RASSF1A and other tumor suppressor genes (p16INK4A, TIMP3, PTEN, CDH1, and RARβ2) in primary HCC, hepatocellular adenoma and in HCC cancer cell lines. All 26 nonmicrodissected liver tissues were obtained from the Institut für Pathologie of the Medizinische Hochschule Hannover and hepatitis B and C infection status was determined with serum hepatitis B surface antibodies, hepatitis B core antibodies and measurement of HCV. To investigate the role of RASSF1A during liver cancer pathogenesis, we analysed the methylation status of the RASSF1A promoter in two HCC cell lines (Hep3B and HepG2) by combined bisulfite restriction analysis (COBRA). In both cell lines, the RASSF1A promoter CpG island was completely methylated (Figure 1a). These cell lines were treated with different concentrations of 5-aza-2′-deoxycytidine (5-Aza-dCR) a drug, which blocks methylation of the DNA after replication (Jones and Taylor, 1980). After 4 days of treatment, the appearance of unmethylated CpGs was detected in the RASSF1A promoter sequence (Figure 1a). To investigate the effect of the aberrant methylation on RASSF1A transcription, we performed RT–PCR with the RNA of these cell lines (Figure 1b). RASSF1A transcription was highly reduced in the methylated HCC cell lines compared to normal human fibroblast with an unmethylated RASSF1A promoter. Treatment with 5-Aza-dCR partially restored the inactivation of the promoter and RASSF1A transcripts were detected (Figure 1b). This indicates that hypermethylation of the RASSF1A promoter is responsible for lack of transcription. The RASSF1C transcript, which initiates from an unmethylated downstream CpG island, was detected in both cell lines (Figure 1b). Furthermore, we investigated the aberrant promoter methylation of RASSF1A in primary HCC by MSP (Figure 2 and and Table 1). In 14 out of 15 (93%) HCC RASSF1A methylation was detected and was detected in all gradings of HCC (Table 2). RASSF1A inactivation was independent of the viral hepatitis infection status (Table 2). However, RASSF1A methylation was not detected in a fibrolamellar HCC (FLC), three hepatocellular adenomas (HCA), two normal liver tissues and six unrelated blood samples (Figure 2 and Table 2). Interestingly, RASSF1A inactivation was also detected in both liver fibrosis and in three out of four (75%) liver cirrhosis probes.

Figure 1

(a) Methylation analysis of RASSF1A by COBRA in HCC cell lines. Hep3B and HepG2 were treated for 4 days with the indicated concentration of 5-Aza-CdR. PCR products (205 bp) of bisulfite-treated DNA were digested (+) or mock digested (−) with Taql. (b) Expression analysis of RASSF1A in HCC. The cell lines (Hep3B and HepG2) were treated for 4 days with the indicated concentration of 5-Aza-CdR. Expression of RASSF1A and RASSF1C transcripts were analysed by isoform-specific RT–PCR. The RASSF1A fragment was 239 bp and the RASSF1C fragment was 354 bp long. RNA from untreated human fibroblasts (HF) was used as a control. Expression of GAPD was determined as a control for RNA integrity. Methylation of the RASSF1A promoter region was determined by bisulfite modification of genomic DNA (Clark et al., 1994; Dammann et al., 2000). COBRA was performed as described previously (Xiong and Laird, 1997; Dammann et al., 2001a). Briefly, 100 ng of bisulfite-treated DNA were amplified with primers MU379 and ML730 in 25 μl reaction buffer containing 0.2 mM dNTP mix, 1.5 mM MgCl2, 10 pmol of each primer and Taq polymerase (InViTek GmBH, Berlin, Germany) at 92°C for 30 s, 55°C for 30 s, and 72°C for 30 s for 20 cycles. A seminested PCR was performed using an internal primer ML561 and primer MU379 with similar conditions as described for the preceding PCR amplification, but for 30 cycles. For the restriction enzyme analysis of PCR products from bisulfite-treated DNA, 20–50 ng of the PCR products were digested with 10 units of Taql (New England Biolabs; Beverly, MA, USA) according to conditions specified by the manufacturer of the enzyme and analysed on a 2% Tris-borate EDTA agarose gel. HCC cell lines were treated with 5-aza-2′-deoxycytidine (5-Aza-CdR; Sigma). In all, 2 × 106 cells each were grown for 4 days in the presence of different concentrations of 5-Aza-CdR (0, 5 and 10 μ M). RNA was isolated and RT–PCR analysis was performed as described previously (Schagdarsurengin et al., 2002). Briefly, 1 μg of RNA from HCC cell lines and HF was reverse transcribed with primer LHE4 in 25 μl of RT-mix with AMV-RT (Promega GmBH, Heidelberg, Germany). cDNA (1 μl) was used for PCR amplification. The primers used for RASSF1A: L27111 (5′TCCTGCAAGG AGGGTGGCTTC) and UHE2αβ (5′GGCTGGGAACCCGCG GTG), and for RASSF1C: L27111 and U2γ (5′GGCTATGGG CGAGGCGGAGGCG). PCR products were separated on 2% Tris-borate EDTA agarose gels

Figure 2

Methylation analysis of RASSF1A and four cancer-related genes by MSP in HCC. DNA of primary HCC, FLC, HCA, hepatocirrhosis (C), matching normal liver tissue (N) and HCC cell lines (Hep3B and HepG2) were modified with bisulfite and analysed by MSP on 2% Tris-borate EDTA agarose gels. The gene studied is given at the left of each panel. Methylation-specific (m) and unmethylation-specific (u) primers were used for MSP. HF and mesothelium were used as a control. In vitro methylated DNA (Meth.) was obtained by modification of genomic HeLa DNA with the CpG methylase Sssl (New England Biolabs; Beverly, MA, USA) according to conditions specified by the manufacturer of the enzyme. Methylation of the CpG islands was determined by bisulfite modification of genomic DNA and MSP (Clark et al., 1994; Herman et al., 1996). Primer sequences of all genes for the methylated and the unmethylated forms, annealing temperatures and the expected PCR product sizes are summarized in Table 1.Bisulfite-modified DNA (100 ng) was amplified in 50 μl reaction buffer containing 0.2 mM dNTP mix, 1.5 mM MgCl2, 10 pmol of each primer and Taq polymerase (InViTek GmBH, Berlin, Germany) and analysed on 2% Tris-borate EDTA agarose gels

Table 1 Summary of primer sequencea, annealing temperatures (°C) and PCR product sizes (bp) used for MSP
Table 2 Summary of the methylation analysis

The growing list of tumor suppressor genes inactivated by promoter hypermethylation provides an opportunity to examine their silencing in HCC. Epigenetic silencing of p16INK4a, PTEN, CDH1, RARβ2, and TIMP3 is a common and frequent event in solid tumors (Herman et al., 1996; Graff et al., 1997; Cote et al., 1998; Bachman et al., 1999; Soria et al., 2002). To elucidate their inactivation in the pathogenesis of HCC, we investigated the aberrant CpG island methylation status by MSP (Figure 2). These results are summarized in Table 2. In the two HCC cell lines three of these genes (p16, CDH1 and RARβ2) were unmethylated. Amplification of the PTEN CpG island failed in the MSP reaction and TIMP3 was the only gene methylated additionally to RASSF1A. Furthermore, we analysed the inactivation of these tumor suppressor genes in the primary liver tissues. The p16 gene was hypermethylated in 10 out of 14 (71%) HCC and was also inactivated in two out of four cirrhosis and in one out of two normal liver tissues. This is consistent with 42–82% of silencing in similar studies (Liew et al., 1999; Wong et al., 1999; Kondo et al., 2000; Roncalli et al., 2002; Shen et al., 2002). No p16 inactivation was found in HCA, FLC, fibrosis and blood samples. The methylation of the other cancer-related genes was less pronounced in the HCC samples and was detected in 25% (1/4) for TIMP3, 17% (1/5) for PTEN, 13% (2/15) for CDH1, and 7% (1/15) for RARβ2 (Figure 2 and Table 2). CDH1 and RARβ2 methylation were detected in one (HCA21) out of three HCA. Neither p16 nor RASSF1A were inactivated in HCA and FLC. In three out of four hepatocirrhosis methylation of RASSF1A was detected and several probes also showed aberrant methylation of p16 (2/4), PTEN (1/1), CDH1 (1/3), and RARβ2 (1/4), indicating that promoter hypermethylation might be an early event during hepatocarcinogenesis. However, the study was carried out on a small sample set and therefore, it will be interesting to analyse RASSF1A inactivation in more detail in a larger sample set of nonmalignant tissues. RASSF1A hypermethylation was detected in 93% of HCC and is the highest frequency of inactivation reported in primary tumors so far. Such intensive methylation of RASSF1A has only been observed in few other primary tumors, for example in 79% of small cell lung cancers (Dammann et al., 2001b), 91% of renal cell carcinomas (Dreijerink et al., 2001), and in 80% of medullary thyroid cancer (Schagdarsurengin et al., 2002). It was previously reported, that RASSF1A methylation was significantly higher in Simian Virus 40 positive malignant mesothelioma (Toyooka et al., 2001) and in Epstein–Barr virus positive gastric carcinoma (Kang et al., 2002). However, in our study no correlation between the hepatitis virus infection and promoter methylation was found in HCC and cirrhosis (Table 2). In nonsmall cell lung carcinoma, RASSF1A methylation is associated with impaired patient survival (Burbee et al., 2001; Tomizawa et al., 2002). In thyroid cancer (Schagdarsurengin et al., 2002) and prostate cancer (Liu et al., 2002), we reported that RASSF1A methylation was higher in more aggressive tumors compared to less malignant forms. Here, no RASSF1A methylation was found in a fibrolamellar carcinoma, a clinicopathologic type of HCC with a favorable prognosis with long-term survival. In contrast to the variable morphology of the common HCC, FLC displays a characteristic reproducible histology with large oncocytic cells, prominent nucleoli of the nuclei and a lamellar fibrous stroma pattern, which argues for a unique trait of carcinogenesis of a single cell, probably without an involvement of RASSF1A. Furthermore, no aberrant RASSF1A methylation was detected in HCA, a rare tumor, which may be difficult to distinguish from its malignant counterpart, low-grade HCC. Therefore, detection of aberrant methylation of RASSF1A in highly differentiated hepatocellular neoplasm with an ‘adenoma-like’ pattern could confirm it as a highly differentiated HCC and discriminate it from HCA. In summary, our results suggest that the RASSF1A gene plays an important role in the pathogenesis of HCC and might become useful as a marker of malignancy and to distinguish between the distinct forms of highly differentiated liver neoplasm.



Ras association domain family 1 gene


hepatocellular carcinoma


hepatocellular adenoma


fibrolamellar hepatocellular carcinoma


loss of heterozygosity




methylation-specific PCR


combined bisulfite restriction analysis


phosphatase and tensin homologue deleted on chromosome 10




retinoic acid receptor beta2

T1MP3 :

tissue inhibitor of metallproteinase-3


hepatitis B virus


hepatitis C virus


  1. Agathanggelou A, Honorio S, Macartney DP, Martinez A, Dallol A, Rader J, Fullwood P, Chauhan A, Walker R, Shaw JA, Hosoe S, Lerman MI, Minna JD, Maher ER and Latif F . (2001). Oncogene, 20, 1509–1518.

  2. Bachman KE, Herman JG, Corn PG, Merlo A, Castello JF, Cavenee WK, Baylin SB and Graff JR . (1999). Cancer Res., 59, 798–802.

  3. Boige V, Laurent-Puig P, Fouchet P, Flejou JF, Monges G, Bedossa P, Bioulac-Sage P, Capron F, Schmit A, Olschwang S and Thomas G . (1997). Cancer Res., 57, 1986–1990.

  4. Burbee DG, Forgacs E, Zochbauer-Muller S, Shivakumar L, Fong K, Gao B, Randle D, Kondo M, Virmani A, Bader S, Sekido Y, Latif F, Milchgrub S, Toyooka S, Gazdar AF, Lerman MI, Zabarovsky E, White M and Minna JD . (2001). J. Natl. Cancer Inst., 93, 691–699.

  5. Clark SJ, Harrison J, Paul CL and Frommer M . (1994). Nucleic Acids Res., 22, 2990–2997.

  6. Cote S, Sinnett D and Momparler RL . (1998). Anticancer Drugs, 9, 743–750.

  7. Dammann R, Li C, Yoon JH, Chin PL, Bates S and Pfeifer GP . (2000). Nat. Genet., 25, 315–319.

  8. Dammann R, Takahashi T and Pfeifer GP . (2001b). Oncogene, 20, 3563–3567.

  9. Dammann R, Yang G and Pfeifer GP . (2001a). Cancer Res., 61, 3105–3109.

  10. Dreijerink K, Braga E, Kuzmin I, Geil L, Duh FM, Angeloni D, Zbar B, Lerman MI, Stanbridge EJ, Minna JD, Protopopov A, Li J, Kashuba V, Klein G and Zabarovsky ER . (2001). Proc. Natl. Acad. Sci. USA., 98, 7504–7509.

  11. Esteller M and Herman JG . (2002). J. Pathol., 196, 1–7.

  12. Feitelson MA, Sun B, Satiroglu Tufan NL, Liu J, Pan J and Lian Z . (2002). Oncogene, 21, 2593–2604.

  13. Flemming P, Wilkens L and Kreipe HH . (2001). Pathologe, 22, 184–190.

  14. Graff JR, Herman JG, Myohanen S, Baylin SB and Vertino PM . (1997). J. Biol. Chem., 272, 22322–22329.

  15. Herman JG, Graff JR, Myohanen S, Nelkin BD and Baylin SB . (1996). Proc. Natl. Acad. Sci. USA, 93, 9821–9826.

  16. Jones PA and Baylin SB . (2002). Nat. Rev. Genet., 3, 415–428.

  17. Jones PA and Taylor SM . (1980). Cell, 20, 85–93.

  18. Kang GH, Lee S, Kim WH, Lee HW, Kim JC, Rhyu MG and Ro JY . (2002). Am. J. Pathol., 160, 787–794.

  19. Khokhlatchev A, Rabizadeh S, Xavier R, Nedwidek M, Chen T, Zhang X, Seed B and Avruch J . (2002). Curr. Biol., 12, 253–265.

  20. Kok K, Naylor SL and Buys CH . (1997). Adv. Cancer Res., 71, 27–92.

  21. Kondo Y, Kanai Y, Sakamoto M, Mizokami M, Ueda R and Hirohashi S . (2000). Hepatology, 32, 970–979.

  22. Lehmann U, Langer F, Feist H, Glockner S, Hasemeier B and Kreipe H . (2002). Am. J. Pathol., 160, 605–612.

  23. Lerman MI and Minna JD . (2000). Cancer Res., 60, 6116–6133.

  24. Li SP, Wang HY, Li JQ, Zhang CQ, Feng QS, Huang P, Yu XJ, Huang LX, Liang QW and Zeng YX . (2001). J. Hepatol., 34, 840–849.

  25. Liew CT, Li HM, Lo KW, Leow CK, Chan JY, Hin LY, Lau WY, Lai PB, Lim BK, Huang J, Leung WT, Wu S and Lee JC . (1999). Oncogene, 18, 789–795.

  26. Liu L, Yoon JH, Dammann R and Pfeifer GP . (2002). Oncogene, 21, 6835–6840.

  27. Lo KW, Kwong J, Hui AB, Chan SY, To KF, Chan AS, Chow LS, Teo PM, Johnson PJ and Huang DP . (2001). Cancer Res., 61, 3877–3881.

  28. Marchio A, Meddeb M, Pineau P, Danglot G, Tiollais P, Bernheim A and Dejean A . (1997). Genes Chromosomes Cancer, 18, 59–65.

  29. Ortiz-Vega S, Khokhlatchev A, Nedwidek M, Zhang XF, Dammann R, Pfeifer GP and Avruch J . (2002). Oncogene, 21, 1381–1390.

  30. Roncalli M, Bianchi P, Bruni B, Laghi L, Destro A, Di Gioia S, Gennari L, Tommasini M, Malesci A and Coggi G . (2002). Hepatology, 36, 427–432.

  31. Roncalli M, Bianchi P, Grimaldi GC, Ricci D, Laghi L, Maggioni M, Opocher E, Borzio M and Coggi G . (2000). Hepatology, 31, 846–850.

  32. Schafer DF and Sorrell MF . (1999). Lancet, 353, 1253–1257.

  33. Schagdarsurengin U, Gimm O, Hoang-Vu C, Dralle H, Pfeifer GP and Dammann R . (2002). Cancer Res., 62, 3698–3701.

  34. Shen L, Ahuja N, Shen Y, Habib NA, Toyota M, Rashid A and Issa JP . (2002). J. Natl. Cancer Inst., 94, 755–761.

  35. Shivakumar L, Minna J, Sakamaki T, Pestell R and White MA . (2002). Mol. Cell. Biol., 22, 4309–4318.

  36. Soria JC, Lee HY, Lee JI, Wang L, Issa JP, Kemp BL, Liu DD, Kurie JM, Mao L and Khuri FR . (2002). Clin. Cancer Res., 8, 1178–1184.

  37. Tomizawa Y, Kohno T, Kondo H, Otsuka A, Nishioka M, Niki T, Yamada T, Maeshima A, Yoshimura K, Saito R, Minna JD and Yokota J . (2002). Clin. Cancer Res., 8, 2362–2368.

  38. Toyooka S, Pass HI, Shivapurkar N, Fukuyama Y, Maruyama R, Toyooka KO, Gilcrease M, Farinas A, Minna JD and Gazdar AF . (2001). Cancer Res., 61, 5727–5730.

  39. Vos MD, Ellis CA, Bell A, Birrer MJ and Clark GJ . (2000). J. Biol. Chem., 275, 35669–35672.

  40. Wilkens L, Bredt M, Flemming P, Becker T, Klempnauer J and Kreipe HH . (2001). J. Pathol., 193, 476–482.

  41. Wistuba II, Behrens C, Virmani AK, Mele G, Milchgrub S, Girard L, Fondon III JW, Garner HR, McKay B, Latif F, Lerman MI, Lam S, Gazdar AF and Minna JD . (2000). Cancer Res., 60, 1949–1960.

  42. Wong IH, Lo YM, Zhang J, Liew CT, Ng MH, Wong N, Lai PB, Lau WY, Hjelm NM and Johnson PJ . (1999). Cancer Res., 59, 71–73.

  43. Xiong Z and Laird PW . (1997). Nucleic Acids Res., 25, 2532–2534.

  44. Zhong S, Tang MW, Yeo W, Liu C, Lo YM, and, Johnson PJ . (2002). Clin. Cancer Res., 8, 1087–1092.

Download references


This work was supported by NBL3 Grant (FKZ 01ZZ0104) from the BMBF to R Dammann, and by NIH Grant CA88873 to GP Pfeifer.

Author information

Correspondence to Reinhard Dammann.

Rights and permissions

Reprints and Permissions

About this article


  • hepatocellular carcinoma
  • hepatocellular adenoma
  • methylation
  • RASSF1
  • tumor suppressor gene

Further reading