ANTICANCER RESEARCH 26: 2123-2128 (2006)

Notch3 Intracellular Domain Accumulates in HepG2 Cell Line

CATIA GIOVANNINI1,2, MICHELA LACCHINI2, LAURA GRAMANTIERI1,2, PASQUALE CHIECO2 and LUIGI BOLONDI1,2

1Department of Internal Medicine and Gastroenterology, University of Bologna and S. Orsola-Malpighi University Hospital, 40138 Bologna; 2Center for Applied Biomedical Research (CRBA), S. Orsola-Malpighi University Hospital, 40138 Bologna, Italy

Abstract. Background: By mediating local cell-cell epithelial growth factor repeats and a lin-12 Notch repeat interactions, the seems to control a as well as a transmembrane subunit (NTM 97-120 kDa) variety of processes from cell fate decisions during containing a short extracellular fragment and an development, to stem cell renewal and to differentiation in intracellular domain (NICD 65-110 kDa) (1). These non- many adult tissues. Hence, perturbed Notch signaling may covalently associated subunits are presented as a be involved both in the development and the spread of heterodimeric functional at the cell surface. Even cancer. The expression and the functional role of some though -independent activation of the Notch major components of the Notch signaling pathway in signaling pathway has been described (2), Notch receptors human hepatocellular carcinoma (HCC) are poorly are mainly activated by transmembrane ligands expressed characterized. Materials and Methods: Notch3, HES1, on the surface of neighboring cells. Five ligands of Notch Jagged1 and Delta1 were analyzed both at the RNA and receptors have been described in vertebrates: Delta-like 1, levels in the HepG2 liver cell line derived from 3 and 4 and Jagged 1 and 2 (3, 4). Upon ligand binding to human HCC. Results: The results of this study the extracellular domain of Notch receptors, two demonstrated, for the first time, that both Jagged1 and sequential proteolytic events occur. First, cleavage takes Delta1 ligands and the downstream effector HES1 are place twelve amino acids outside the transmembrane expressed in the HepG2 actively proliferating cell line. domain by metalloproteinase TACE/ADAM17 (5). The Moreover, a high expression of Notch3 intracellular domain, resulting Notch COOH-terminal fragment, called NEXT indicative of constitutively activated Notch signaling, was (Notch extracellular truncation), is required for the second the only detectable Notch3 subunit in HepG2. Conclusion: cleavage performed by Á-secretase within the trans- These findings suggest that Notch3 may be involved in membrane region. This last proteolytic event releases the mechanisms controling the differentiation and the spread of Notch intracellular domain (NICD) into the cytoplasm (6). HCC and that Notch3 activation may be dependent on both The NICD translocates into the nucleus, binding to the Jagged1 and Delta1 ligands. transcription factor CBF1/RBP-Jk transactivating target . To date, the HES (hairy/enhancer of split) and Notch genes encode for a family of highly conserved HERP (HES-related repressor protein) families of single-pass transmembrane receptors known as , 2, transcriptional repressors are the only known major targets 3 and 4. Notch receptors are synthesized as ~300 kDa of Notch signaling (7, 8). (pro-Notch) which are then cleaved by a furin-like By mediating local cell-to-cell communication, the Notch convertase in the Golgi. This proteolytic event releases an family of receptors controls either cell fate and spread EC extracellular subunit (N ~200 kDa) containing many during development or stem cell renewal and differentiation in many adult tissues (9, 10). A perturbed Notch signaling pathway has been described in different human diseases, including cancer. Constitutively active Notch1 and Notch4 Correspondence to: Dr. Catia Giovannini, CRBA, Pad. 23, polypeptides were found to be involved in human acute S. Orsola-Malpighi University Hospital, Via Massarenti 9, 40138 T-lymphoblastic leukemia (T-ALL) and mouse mammary Bologna, Italy. Tel: +39-051-6364903, Fax: +39-051-6363902, e-mail: [email protected] tumor development, respectively (11, 12). The induced expression of constitutively active Notch4 has been Key Words: Notch3, Delta1, Jagged1, HES1, HCC. described to promote the expansion of hematopoietic stem

0250-7005/2006 $2.00+.40 2123 ANTICANCER RESEARCH 26: 2123-2128 (2006)

Table I. Primer sequences for RT-PCR analysis.

Gene Primer sequence Product Annealing size temp. ÆC

Notch3 FW 5’-aaggacgtggcctctggt-3’ 91 62 REV 5’-tcaggctctcacccttgg-3’ Notch3 FW 5’-aaggacgtggcctctggt-3’ 2043 60.6 REV 5’-ggcccccaagatctaagaac Delta1 FW 5’-agcaagcgtgacaccaag-3’ 286 61 Figure 1. Notch3 evaluated by RT-PCR. A) Conventional REV 5’-tcctcttcagcagcattcg-3’ PCR performed with Herculase Taq Polymerase. B) Real-time PCR Jagged1 FW 5’-tcgctgtatctgtccacctg-3’ 227 64 products loaded on agarose gel. N.G.: PCR-negative control. REV 5’-agtcactggcacggttgtag-3’ HES1 FW 5’-gctggtgctgtctggatg-3’ 176 62 REV 5’-cattcctgctctcgccttc-3’

Fw, forward; Rev, reverse.

cells in vitro (13). These observations suggest that unregulated NICD expression prevents differentiation, favoring malignant transformation. On the other hand, constitutive activation of Notch1 may function as a tumor suppressor in small cell lung cancer cells, in prostate cancer cells and in mouse skin (14-16) by inducing cell growth arrest. Epidemiological evidence indicates that human hepatocellular carcinoma (HCC) is one of the most prevalent neoplasms worldwide. The molecular mechanisms controlling the differentiation and the spread of HCC are still largely unknown. The Notch1 signaling pathway is activated during rat liver regeneration and overexpression of Notch1 has been found to inhibit the growth of HCC Figure 2. Whole-cell extracts (50 Ìg) of the HepG2 cell line were electrophoresed and immunoblotted with the indicated antibodies. A and cells in vitro and in vivo (17). Notch3 has recently been B in the upper panel designate the two different antibodies used against described to be involved in bile duct development (18) but, Notch3; A (BC4), B (sc-5593). to date, there are no studies concerning Notch3 expression and function in HCC. The aim of the present study was to investigate the expression of some major components of the Notch reaction mixture including: 1X RT buffer, 0.4 mM dNTPs, 5 mM signaling pathway, including Notch3, Jagged1, Delta1 and dithiothreitol (DTT), 0.5 ÌM oligodT, 3 ÌM random primers, 240 U Superscript II (all reagents from Invitrogen). The RT the downstream effector gene, HES1, in the HepG2 tumor reaction was carried out at 42ÆC for 1 h, followed by 5 min at 95ÆC cell line. to inactivate the enzyme. Hes1, Jagged1 and Delta1 gene expressions was assessed by real-time PCR using the iCycler Materials and Methods Thermal Cycler (Bio-Rad, Hercules, CA, USA). Notch3 gene expression was assessed using both real-time and conventional end- Cell lines. The HCC HepG2 cell line was obtained from the point PCR amplification. American Type Culture Collection (HB-8065, ATCC, Rockville, For real-time PCR, all transcripts were amplified in 25 Ìl of MD, USA) and was maintained in Eagle’s minimum essential reaction using 1 Ìl of cDNA, 300 mM of each primer and the iQ media (MEM) supplemented with 10% fetal bovine serum (FBS), SYBER Green Supermix (Bio-Rad). Standard curves were 100 U/ml penicillin and 100 mg/ml streptomycin (all reagents from generated by making appropriate dilutions of samples of cDNA ATCC) at 37ÆC in a 5% CO2 incubator. synthesized from human liver tissues. The real-time products were loaded on agarose gel to rule out aspecific amplifications. Reverse transcriptase-polymerase chain reaction (RT-PCR). RNA was Conventional PCR was performed with Herculase Taq Polymerase isolated from the HepG2 cell line by TRIzol (Invitrogen, Paisley, (Stratagene, La Jolla, CA, USA) as described by the company and Scotland), according to the manufacturer’s instructions. Four Ìg of the PCR product was resolved on 1% agarose gel containing total RNA were treated with DNAse I (Invitrogen) to eliminate ethidium bromide. The primer sequences and the annealing contaminating genome DNA. RT was performed in 30 Ìl of temperature are listed in Table I.

2124 Giovannini et al: Notch3 Expression in HepG2 Cell Line

Figure 3. Expression and subcellular localization of the Notch3 receptor, as detected by the immunoperoxidase method, of the HepG2 cell line. Positive staining was observed in the cytoplasm and at the nuclear level. Nuclei were counterstained with hematoxylin. Original magnification 40X.

Protein extraction. The cultured HepG2 cell line was dissolved in Immunocytochemistry. The HepG2 cells were seeded on sterilized lysis buffer containing 10 mM Tris-HCl pH 7.4, 2.5 mM MgCl2, 1% cover-slips and fixed in 4% paraformaldeyde. The cells were then TritonX 100, 1 mM DTT, 0.1mM phenylmethylsulfonyl fluoride permeabilized in PBS containing 0.1% of saponin and incubated with (PMSF) and protease inhibitors (Sigma Chemical Co., St. Louis, normal goat serum at room temperature for 30 min. Notch3 protein MO, USA). The lysate was then centrifuged at 4ÆC, 15,000 x g, for localization into the cell was assessed with the same antibody used in 15 min and the supernatant was assayed for the protein Western blot ( Santa Cruz), followed by an HRP-rabbit EnVision concentration using the Bio-Rad protein assay (Bio-Rad). system with Vector Red (Vector Laboratories, Burlingame, CA, USA) as chromogen. The cells were then counterstained with Mayer’s SDS-PAGE and Western blot analysis. Fifty Ìg of proteins were hematoxylin and mounted with DPX (BDH Chemical, Poole, UK). boiled at 95ÆC for 10 min in 1X SDS-loading buffer (65 mM Negative controls were obtained by omitting the primary antibody. Tris-HCl, pH 7.5, 65 mM 2-mercaptoethanol, 1% SDS, 10% glycerol and 0.003% bromophenol blue), resolved by 5% or 8% Results polyacrylamide gels and blotted on nitrocellulose membranes (Hybond C Extra, Amersham Pharmacia, Little Chalfont, UK). Semi-quantitative real-time PCR. To investigate Notch3, The membranes were stained with Red Ponceau solution (Sigma), blocked in 5% non-fat dry milk for 50 min in phosphate buffered Jagged1, Delta1 and HES1 gene expressions, semi- saline (PBS) and then incubated with the appropriate primary quantitative real-time PCR was performed on RNA antibody. The primary antibodies and dilutions used are listed as extracted from the HepG2 cell line. The real-time PCR follows: anti-Notch3 polyclonal antibody (sc-5593, Santa Cruz efficiency was almost identical for the genes analyzed. The Biotechnology, Santa Cruz, CA, USA) 1:400, anti-Notch3 cycle thresholds (Ct) were as follows: polyclonal antibody (BC4), kindly provided by Dr. A. Joutel (Faculty of Medicine, Lariboisiere, Paris) 1:10.000, anti-Delta1 Ct = 33, Ct = 34, Ct = 20, Ct = 26, polyclonal antibody (sc-9102, Santa Cruz) 1:200, anti-Jagged1 Notch3 Delta1 Jagged1 HES1 polyclonal antibody (sc-8303, Santa Cruz) 1:100. After repeated washing in PBS containing 0.1% Tween 20 suggesting that the levels of Notch3, Jagged1, Delta1 and (PBST), the membranes were incubated with anti-mouse or anti- HES1 gene expressions were significantly different. Faint rabbit HRP-conjugated secondary antibodies using the EnVision bands were observed for Notch3 and Delta1 real-time PCR dextran polymer visualization system (DAKO, Denmark). The products resolved on agarose gel. A high Notch3 expression membranes were washed and autoradiographies were obtained was observed with conventional PCR where a proper Taq using a chemiluminescence reaction (ECL reagents, Amersham). Polymerase (Herculase) efficient with GC-rich templates, Digital images of the autoradiographies were acquired with a scanner (Fluor-S MultiImager, Bio-Rad) and signals were such as the Notch genes, was used (Figure 1). quantified using specific densitometric software (Quantity-one, Bio-Rad) in absorbance units after light calibration with a Protein expression. Notch3 protein expression, determined reference autoradiography. by immunoblotting, was observed as a band corresponding

2125 ANTICANCER RESEARCH 26: 2123-2128 (2006) to the molecular weight of 76 kDa, matching the cleaved signaling, was the only detectable Notch3 subunit in the Notch3 intracellular domain (NICD). Conversely, both the HepG2 cell line. At the immunocytochemical level, strong 97-kDa and 300 kDa bands, corresponding to the Notch3 staining for Notch3 was observed in the cytoplasm and in a transmembrane subunit and to the Notch3 full-length few nuclei, suggesting that a very low amount of NICD receptor, respectively, were not detected. The same results migrates into the nucleus to activate target genes. It has were obtained by using two different antibodies, as recently been described that different Notch receptors can described above (Figure 2). This evidence was supported have opposing effects in a single tumor type (23). This by immunocytochemical results which showed a strong evidence suggests that a cross-talk among different Notch staining for Notch3 in the cytoplasm and in a few nuclei receptors could exist, giving a possible explanation for the (Figure 3). On the other hand, no positive staining was large amount of activated Notch3 kept in the cytoplasm, observed at the membrane level of HepG2 cells, suggesting where it is probably poly-ubiquitylated and degraded by the complete lack of the transmembrane Notch3 subunit. proteasome (24). In order to detect Notch signaling activity, Both the Jagged1 and Delta1 ligands were detected by HES1 gene expression was investigated by real-time PCR. immunoblotting and Jagged1 was expressed three times HES1 was found to be expressed in the HepG2 cell line and more than Delta1 (Figure 2).The difference in protein levels its expression could be regulated, at least in part, by NICD3 between Jagged1 and Delta1 paralleled their different gene in the nucleus. expressions as evaluated by real-time PCR. Notch3 protein We have no data to demonstrate that Notch3 activation is expression correlated with the mRNA level detected by induced by Delta1 or Jagged1 ligands, both expressed in conventional PCR. these cells. Even though ligand–receptor activation by Jagged versus Delta is modulated by the glycosylation of Discussion residues within the epithelial growth factor repeats of the Notch extracellular domain (25), the higher protein levels By mediating a balance between cell proliferation and of Jagged1 than Delta1 observed imply that Notch3 differentiation, Notch signaling has been suggested to be persistent activation could be mediated by Jagged1. involved in malignant transformation (19). The role of This study provides the first evidence that some major Notch in tumorigenesis is still largely unknown and it seems components of the Notch signaling pathway, including to be tumor type-dependent (12, 15, 17). Notch3, Jagged1, Delta1 and the downstream effector gene Notch3 has been described to be involved in different HES1, are expressed in the HepG2 tumor cell line. A high processes, including bile duct development and vascular expression of endogenous NICD, indicative of persistent smooth muscle cell (VSMCs) growth (18, 20). Mutations in Notch signaling, could be necessary for malignant liver cell the Notch3 receptor are involved in cerebral autosomal proliferation; the same result was obtained in the HCC cell dominant arteriopathy with subcortical infarcts and line Hep3B (data not shown). Our hypothesis is supported by leukoencephalopathy (CADASIL) (21). Altered Notch3 a previous study showing that the constitutive activation of seems to allow the disruption of early thymocyte the Notch3 receptor in VSMCs induces a persistent state of differentiation, favoring the development of T-cell leukemia cell cycle progression by mediating suppression of the cell in the mouse (22). To date, little is known about Notch3 cycle inhibitor p27kip (20). Taking our results into expression and function in human cancer. consideration, along with those from previous studies, it can In the present study, the expression of some major be supposed that deregulation of Notch receptor activity may components of the Notch signaling pathway, including be involved both in the differentiation and spread of HCC. Notch3, Jagged1, Delta1 and the downstream effector gene, In order to investigate whether some genes regulated by HES1, were analyzed in the HepG2 tumour cell line. At the Notch3 are involved in such processes, other studies including protein level, the absence of a 300-kDa band, corresponding interference RNA against Notch3 will be performed. to the Notch3 full-length receptor, could be explained in Western blot by the non-covalent association between the Acknowledgements extracellular subunit and the transmembrane subunit destroyed during protein extraction. On the other hand, The authors are grateful to Dr. A. Joutel for providing the anti- both antibodies, specific for epitopes within the carboxy- Notch3 antibody. This study was supported by Fondazione Cassa di Risparmio in Bologna, Italy. terminus of Notch3, failed to detect the Notch3 trans- membrane subunit, suggesting the complete lack of the References receptor on the cellular membrane. This observation was in line with immunocytochemical results which failed to show 1 Blaumuller CM, Qi H, Zagouras P and Artavanis-Tsakonas S: positive staining at the cellular membrane level. A high Intracellular cleavage of Notch leads to a heterodimeric expression of endogenous NICD, indicative of active Notch receptor on the plasma membrane. Cell 90: 281-291, 1997.

2126 Giovannini et al: Notch3 Expression in HepG2 Cell Line

2 Sanchez-Irizarry C, Carpenter AC, Weng AP, Pear WS, Aster 15 Shou j, Ross S, Koeppen H, De Sauvage FJ and Gao WQ: JC and Blacklow SC: Notch subunit heterodimerization and Dynamics of notch expression during murine prostate prevention of ligand-independent proteolytic activation depend, development and tumorigenesis. Cancer Res 61: 7291-7297, 2001. respectively, on a novel domain and the LNR repeats. Mol Cell 16 Nicolas M, Wolfer A, Raj K, Kummer JA, Mill P, Van Noort M, Biol 24: 9265-9273, 2004. Hui CC, Clevers H, Dotto GP and Radtke F: Notch1 functions as 3 Gray GE, Mann RS, Mitsiadis E, Henrique D, Carcangiu ML, a tumor suppressor in mouse skin. Nat Genet 33: 416-421, 2003. Banks A, Leiman J, Ward D, Ish-Horowitz D and Artavanis- 17 Qi R, An H, Yu Y, Zhang M, Liu S, Xu H, Guo Z, Cheng T Tsakonas S: Human ligands of the Notch receptor. Am J Pathol and Cao X: Notch1 signaling inhibits growth of human 154: 785-794, 1999. hepatocellular carcinoma through induction of cell cycle arrest 4 Yoneya T, Tahara T, Nagao K, Yamada Y, Yamamoto T, and apoptosis. Cancer Res 63: 8323-8329, 2003. Osawa M, Miyatani S and Nishikawa M: Molecular cloning of 18 Flynn DM, Nijjar S, Hubscher SG, de Goyet j de V, Kelly DA, delta-4, a new mouse and human Notch ligand. J Biochem 129: Strain AJ and Crosby HA: The role of Notch receptor 27-34, 2001. expression in bile duct development and diseases. J Path 204: 5 Brou C, Logeat F, Gupta N, Bessia C, LeBail O, Doedens JR, 55-64, 2004. Cumano A, Roux P, Black RA and Israel A: A novel proteolytic 19 Maillard I, HeY and Pear WS: From the yolk sac to the spleen: cleavage involved in Notch signaling: the role of disintegrin- new roles for Notch in regulating hematopoiesis. Immunity 18: metalloprotease TACE. Mol Cell 5: 207-216, 2000. 587-589, 2003. 6 Weinmaster G: Notch signaling: direct or what? Curr Opin 20 Campos AH, Wang W, Pollman MJ and Gibbons GH: Genet Dev 8: 436-442, 1998. Determinants of Notch-3 receptor expression and signaling in 7 Jarriault S, Brou C, Logeat F, Schroeter EH, Kopan R and vascular smooth muscle cells. Circ Res 91: 999-1006, 2002. Israel A: Signalling downstream of activated mammalian Notch. 21 Ruchoux MM, Domenga V, Brulin P, Maciazek J, Limol S, Nature 377: 355-358, 1995. Tournier-Lasserve E and Joutel A: Transgenic mouse 8 Iso T, Kedes L and Hamamori Y: HES and HERP families: expressing mutant Notch3 develop vascular alterations multiple effectors of the Notch signaling pathway. J Cell Physiol characteristic of cerebral autosomal dominant arteriopathy with 194: 237-255, 2003. subcortical infarcts and leukoencephalopathy. Am J Pathol 162: 9 Artavanis-Tsakonas S, Rand MD and Lake RJ: Notch signaling: 329-342, 2003. cell fate control and signal integration in development. Science 22 Bellavia D, Campese AF, Vacca A, Gulino A and Screpanti I: 284: 770-776, 1999. Notch3, another Notch in T cell development. Semin Immunol 10 Stier S, Cheng T, Dombkowiski D, Carlesso N and Scadden DT: 15: 107-112, 2003. Notch1 activation increases hematopoietic stem cell self-renewal 23 Fan X, Mikolaenko I, Elhassan I, Ni X, Wang Y, Ball D, Brat in vivo and favors lymphoid over myeloid lineage outcome. Blood DJ, Perry A and Eberhart CG: Notch1 and Notch2 have 99: 2369-2378, 2002. opposite effects on embryonal brain tumor growth. Cancer Res 11 Aster JC and Pear WS: Notch signaling in leukemia. Curr Opin 64: 7787-7793, 2004. Hematol 8: 237-244, 2001. 24 Aguilar RC and Wendland B: Ubiquitin: not just for 12 Politi K, Feirt N and Kitajewski J: Notch in mammary gland proteasomes anymore. Curr Opin Cell Biol 15: 184-190, 2003. development and breast cancer. Sem Cancer Bio 14: 341-347, 25 Hicks C, Johnston SH, diSibio G, Collazo A, Vogt TF and 2004. Weinmaster G: Fringe differentially modulates Jagged1 and 13 Ye Q, Shieh JH, Morrone G and Moore MA: Expression of Delta1 signalling through Notch1 and Notch2. Nat Cell Biol 2: constitutively active Notch4 (Int-3) modulates myeloid 515-520, 2000. proliferation and differentiation and promotes expansion of hematopoietic progenitors. Leukemia 18: 777-787, 2004. 14 Sriuranpong V, Borges MW, Ravi RK, Arnold DR, Nelkin BD, Baylin SB and Ball DW: Notch signaling induces cell cycle arrest Received January 11, 2006 in small cell lung cancer cells. Cancer Res 61: 3200-3205, 2001. Accepted February 27, 2006

2127