- Email: [email protected]
Isolation and characterization of the promoter region of the human GM3 synthase gene
Biochimica et Biophysica Acta 1578 (2002) 84 – 89 www.bba-direct.com
Promoter paper
Isolation and characterization of the promoter region of the human GM3 synthase gene $ Sang-Wan Kim a, Sang-Hyeon Lee b, Kyoung-Sook Kim a, Cheorl-Ho Kim c, Young-Kug Choo d, Young-Choon Lee a,* a
Division of Biotechnology, Faculty of Natural Resources and Life Science, Dong-A University, 840, Hadan-Dong, Saha-Gu, Busan 604-714, South Korea b Department of Bioscience and Biotechnology, Silla University, Busan 617-736, South Korea c Department of Biochemistry and Molecular Biology, College of Oriental Medicine, Dongguk University, Kyung-Pook 780-350, South Korea d Division of Biological Science, College of Natural Science, Wonkwang University, Chonbuk 570-749, South Korea Received 9 April 2002; received in revised form 1 August 2002; accepted 2 September 2002
Abstract GM3 synthase, which transfers CMP-NeuAc with an a2,3-linkage to a galactose residue of lactosylceramide, plays a key role in the biosynthesis of all complex gangliosides. The expression of this gene is highly restricted in both human fetal and adult tissues. To elucidate the mechanisms that regulate the tissue-specific expression of the human GM3 synthase (hST3Gal V) gene, we have isolated and characterized its 5V-flanking region. Potential transcriptional start site was determined by CapSite hunting. Sequence analysis of the 5Vflanking region revealed that hST3Gal V promoter lacked canonical TATA and CAAT boxes but contained several putative binding sites for transcription factors AP4, MZF1, SP1, ATF/CREB, NFY, IK2 and LYF1, etc. Functional analysis of the 5V-flanking region of the hST3Gal V gene by transient expression method revealed that the 177 to 83 region is important for transcriptional activity of the hST3Gal V gene in SK-N-MC and HepG2 cells. The present results also suggested that both positive and negative regulatory elements are present in this TATAless promoter of the hST3Gal V gene. D 2002 Elsevier Science B.V. All rights reserved. Keywords: Promoter; GM3 synthase; Sialyltransferase; Ganglioside
Gangliosides are a family of sialic acid (NeuAc)-containing glycosphingolipids that are found in high concentration on the central nervous system [1]. They play important roles in a large variety of biological processes, such as cell – cell interaction, adhesion, cell differentiation, growth control and receptor function [2]. GM3 is the first and the simplest of the gangliosides and is known to play important roles in the modulation of cell growth through modified signal transduction and cell differentiation. GM3 inhibits tyrosine phosphorylation of the epidermal growth factor (EGF) receptor and EGF-dependent cell growth, independent of receptor – receptor interaction, whereas DeN-acetyl-GM3 enhances serine phosphorylation of the EGF
$ The nucleotide sequences reported in this paper have been submitted to the GenBank/EMBL with accession number AF488709. * Corresponding author. Tel.: +82-51-200-7591; fax: +82-51-200-6993. E-mail address: [email protected] (Y.-C. Lee).
receptor and stimulates cell proliferation [3]. GM3 induces monocytic differentiation of human myeloid and monocytoid leukemic cell lines such as HL-60 and U937 during macrophage-like cell differentiation [4]. The tumor-associated expression of ganglioside GD1a in NFS60 cells by introduction of an interleukin-3 (IL-3) gene is controlled by the activity of a single glycosyltransferase, GM3 synthase [5]. GM3 is synthesized by CMP-NeuAc:lactosylceramide a2,3-sialyltransferase (GM3 synthase, EC 2.4.99.9) which catalyzes the transfer of NeuAc from CMP-NeuAc to the nonreducing terminal galactose of lactosylceramide. GM3 synthase is a key regulatory enzyme for ganglioside biosynthesis [6] because it catalyzes the first committed step in the synthesis of nearly all gangliosides. Although its cDNA has been cloned from mouse [7] and human tissue [8], its genomic structure and transcriptional regulation has not been elucidated. Very recently, we determined the genomic structure of the human GM3 synthase (hST3Gal V) gene and identified alternatively spliced isoforms in its 5V-
0167-4781/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 7 8 1 ( 0 2 ) 0 0 5 0 5 - 5
S.-W. Kim et al. / Biochimica et Biophysica Acta 1578 (2002) 84 – 89
untranslated region (UTR) [9]. The hST3Gal V gene is abundantly expressed in some tissues such as the brain and placenta [9]. To elucidate the molecular basis of hST3Gal V gene expression, in this study, the genomic region containing the promoter of the hST3Gal V was isolated and functionally characterized. To determine the transcription start site of the hST3Gal V gene, the CapSitek hunting method [10 –13] was used in accordance with the manufacturer’s protocol (Nippon Gene, Tokyo, Japan). CapSite cDNAR dT from human fetal brain, in which the 5V-terminal m7GpppN cap structure of mRNA was removed and recapped by the 3V-end of a specific rOligo primer provided by the company, was subjected to nested PCR. The first round PCR was performed using the rOligo-specific primer 1RDT and hST3Gal V gene-specific primer TGP1 (Table 1). The second round PCR was performed using the rOligo-specific primer 2RDT and hST3Gal V gene-specific primer TGP2. The reactions were performed using the following conditions: 95 jC for 5 min and then 35 cycles of 95 jC for 20 s, 60 jC for 20 s, and 72 jC for 90 s, with a final elongation of 72 jC for 5 min. The PCR product of about 0.34 kb fragment was obtained (Fig. 1), subcloned into pT7Blue(R) T-vector (Novagen, Inc., Madison) and sequenced. The transcription start site was determined by identification of the boundary sequence between rOligo and hST3Gal V mRNA sequence. The result showed that its sequence and 5V-end are completely identical to those of the type 2 isoform [9] obtained by 5Vrapid amplification of cDNA ends (5V-RACE). Therefore, we assigned the major initiation site of transcription of
Fig. 1. Identification of the cap site by nested PCR for the determination of transcription start site of the hST3Gal V. CapSite cDNA from human fetal brain was subjected to nested PCR. The first round PCR was performed using the rOligo-specific primer 1RDT and hST3Gal V gene-specific primer TGP1, whose positions are shown in (A). The second round PCR was performed using the rOligo-specific primer 2RDT and hST3Gal V gene-specific primer TGP2, whose positions are also shown in (A). The resulting product (lane 1) was analyzed by 2% agarose gel electrophoresis and visualized by ethidium bromide staining (B). M, DNA size marker (ØX174 digested with Hae III).
85
hST3Gal V gene to the A residue at position designed as + 1 (Fig. 2). Based on this finding, the upstream genomic region of the hST3Gal V was identified using the Human Genome Resources of NCBI, NLM, and NIH. A 1599 bp fragment of the 5V-flanking sequence of the hST3Gal V gene was obtained by long and accurate PCR (LA-PCR) amplification with LA-Taq polymerase (Takara Shuzo, Shiba, Japan). LAPCR was performed with primers FP1600 and RP1 (Table 1), and human genomic DNA (Clontech, Palo Alto, CA) as a template. The reactions were performed using the following conditions: 94 jC for 1 min and then 30 cycles of 98 jC for 20 s and 68 jC for 3 min, with a final elongation of 72 jC for 10 min. The PCR product was subcloned into pT7Blue(R) T-vector (Novagen) to give pThGM3 and sequenced in both directions by cloning convenient restriction fragments into pUC119 or using primers designed from known sequence. Analysis of the 5V-flanking region of the hST3Gal V gene by the MatInspector v2.2 program (core similarity 1.0, matrix similarity 0.95) using TRANSFAC 4.0 matrices [14,15] revealed that this region lacks canonical TATA and CAAT boxes, but contains several putative transcription factor binding sites such as AP4, MZF1, SP1, ATF/CREB, NFY, IK2 and LYF1, etc. (Fig. 2). This region also contains an apparent G + C-rich region between 1 and 304 (GC content, 81%). To determine whether the genomic DNA flanking exon 1 contains a functional promoter, the 1.6 kb fragment of pThGM3 containing 1600 bp of the 5V-flanking region of hST3Gal V gene and the first 15 bp of exon 1 was inserted into the SacI and BglII sites of pCAT3-Basic plasmid, a chloramphenicol acetyltransferase (CAT) expression vector lacked promoter and enhancer (Promega, Madison, WI). This construct pCAT-1600 (2 Ag) and pCAT3-Basic (2 Ag) as a negative control were transfected into the human neuroblastoma cell line SK-N-MC and human hepatoma cell line HepG2 at 70% confluence by lipofectamine-mediated transfection protocol (Gibco BRL, Life Technologies). SK-N-MC cells were cultured in minimum essential medium (MEM) supplemented with 10% fetal bovine serum, 1 mM sodium pyruvate and 1 MEM nonessential amino acids. HepG2 cells were grown in Dulbecco’s modified Eagle medium (DMEM) containing 10% FBS. Each of the transfection experiments were repeated at least twice, yielding reproducible results. To normalize for the efficiency of transfection, these cells were simultaneously cotransfected with 1 Ag of pCMVh (Clontech). Cells were harvested 48 h after transfections. Lysates were prepared by four cycles of freezing and thawing of the harvested cells followed by centrifugation. CAT assays were carried out using the CAT-enzyme linked immunosorbent assay (ELISA) kit (Boehringer Mannheim). CAT activity was normalized to h-galactosidase activity and expressed as a fold increase over pCAT3-Basic. As shown in Fig. 3, the result of the CAT assay clearly demonstrated that the 5V-flanking region of hST3Gal V
86
S.-W. Kim et al. / Biochimica et Biophysica Acta 1578 (2002) 84 – 89
Fig. 2. Nucleotide sequence of the 5V-flanking region of the hST3Gal V. The sequence is numbered with the first nucleotide of the transcription start site as + 1. The exon 1 sequence is shown in bold lowercase letters, and the start position of intron 1 is marked with an arrow. The putative transcription factor binding sites are described in the text. For the detection of promoter activity, the start point of each construction is indicated by an arrowhead.
S.-W. Kim et al. / Biochimica et Biophysica Acta 1578 (2002) 84 – 89
87
Fig. 3. Promoter activity of the 5V-flanking region of the hST3Gal V gene in SK-N-MC and HepG2 cells. Schematic representation of DNA constructions containing various lengths of the 5V-flanking region of the hST3Gal V linked to the CAT reporter gene is presented. The restriction sites are shown and the transcription start site is indicated as + 1. pCAT3-Basic is the promoterless and enhancerless vector control. Each CAT constructs were co-transfected into SKN-MC and HepG2 cells with pCMVh as the internal control. CAT activity was measured using the CAT-ELISA kit and normalized with h-galactosidase activity derived from pCMVh. CAT activity is expressed as a fold increase over reporter plasmid pCAT3-Basic (negative control). Results presented are the mean F S.E. for three to four plates. * ANOVA P < 0.0001 compared with pCAT3-Basic. Transfections were repeated at least twice, yielding reproducible results.
gene displayed transcription activity in both cell lines. Transcription activity of pCAT-1600 in SK-N-MC cells was 3-fold higher than that in HepG2 cells. This elevated transcription activity in a neuronal cell line is consistent with the recent report that showed higher levels of hST3Gal V mRNA in human fetal and adult brain tissues than human liver tissues [9]. This also coincides with the result that both cell lines express hST3Gal V mRNA, but expression level was approximately 2.5-fold higher in SKN-MC cells than HepG2 cells (data not shown). Therefore, these results suggest that the hST3Gal V promoter reveals a cell-specific manner. To characterize the region regulating the transcription activity of the hST3Gal V, hST3Gal V promoter fragments containing varying lengths of 5V-flanking sequence were
generated by LA-PCR using pThGM3 as template. Forward primers FP1210, FP847, FP432, FP177 and FP83 containing SacI sites, and a common reverse primer RP1 containing a BglII site were used (Table 1). To avoid translation from the first ATG at position + 4 of the genomic fragment, which results in frame-shift reading of CAT cDNA, the A of the first ATG codon was replaced with T in a common reverse RP1 primer. The PCR fragments were subcloned into pT7Blue(R) T-vector (Novagen) and sequenced. Each fragment obtained by digestion with SacI and BglII was inserted into the pCAT3-Basic plasmid. The orientations of each construct were confirmed by restriction enzyme analysis. These constructs were designated as pCAT-1210, pCAT-847, pCAT-432, pCAT-177 and pCAT-83 (Fig. 3). All plasmids used in transfection experiments were prepared
Table 1 Primers used in this study Primer
Sequence
Strand
1RDT TGP1 2RDT TGP2 FP1600 FP1210 FP847 FP432 FP177 FP83 RP1
5V-GATGCTAGCTGCGAGTCAAGTC-3V 5V-CCTTCTGCAAGACTTGCTGAG-3V 5V-CGAGTCAAGTCGACGAAGTGC-3V 5V-TGGTCAGGGTCCACATAATGC-3V 5V-ATCGAGCTCAACAGACCCAGACTTGAGAGAGTCC-3V 5V-ATCGAGCTCCAGCTGGTCTTGAACTCC-3V 5V-ATCGAGCTCGCTCCCAATATCCTCACG-3V 5V-ATCGAGCTCACGTTCAGCTGTGTCCAAG-3V 5V-ATCGAGCTCATCGTGCTCTCCGATT-3V 5V-ATCGAGCTCAGCTGAATGGGCGC-3V 5V-AAGAGATCTTCGTCCGCAAACTAATGAGGGGGC-3V
sense antisense sense antisense sense sense sense sense sense sense antisense
Position + 358 to + 338 + 314 to + 294 1600 to 1576 1210 to 1192 847 to 830 432 to 414 177 to 161 83 to 69 + 15 to 11
The primers used for cloning into the reporter vector, pCAT3-Basic, include the following restriction enzyme sites: SacI, underlined text; BglII, italic underlined text. The A of the first ATG codon replaced with T is shown in boldface in a common reverse RP1 primer.
88
S.-W. Kim et al. / Biochimica et Biophysica Acta 1578 (2002) 84 – 89
by Qiagen Midi-prep kit (Qiagen, Valencia, CA). The promoter activities of these deletion constructs were tested in SK-N-MC and HepG2 cells as described above. In SK-NMC cells, as shown in Fig. 3, deletion from 1600 to 1210 resulted in about 34% decrease in transcription activity compared to pCAT-1600. Further deletion downstream to base 432 caused a gradual decrease of the promoter activity. However, deletion in the region from 432 to 177 resulted in increase of transcription activity, while deletion in the region from 177 to 83 markedly reduced transcription activity to the level of the promoterless and enhancerless control vector pCAT3-Basic. These results suggest that potential positive regulatory elements exist within the 1600 to 432 and 177 to 83 region, while potential negative regulatory elements exist within the 432 to 177 region. On the other hand, in HepG2 cells, no remarkable transcription activities were observed in the deletion mutants in the region from 1600 to 432 compared to pCAT3-Basic. However, deletion in the region from 432 to 177 resulted in significant increase in transcription activity, while deletion in the region from 177 to 83 caused a drastic decrease to the basal level like SK-N-MC cells, suggesting that potential negative and positive regulatory elements exist within the 432 to 177 and 177 to 83 regions, respectively. These results show that the region between 177 and 83 functions as the core promoter essential for transcriptional activation of hST3Gal V in SK-N-MC and HepG2 cells. This also suggests that promoter elements located between nucleotide positions 117 and 83 are essential for basal expression of the hST3Gal V gene in both cell lines. As shown in Fig. 2, this region from 177 to 83 is GC-rich (G-C content, 74%) and contains NFY, ATF/CREB, SP1, EGR3 and MZF1 binding sites. These binding sites are present in the inverted orientation except for ATF/CREB and MZF1. SP1 is known to be important for the transcriptional activation of promoters which lack TATA and CAAT boxes [16,17]. NFY is a ubiquitous transcription factor which binds to CCAAT element in direct or inverted orientation in the proximal promoters of a wide variety of mammalian genes and its binding is frequently essential for gene transcription, particularly for TATA-less promoters [18]. The myeloid zinc finger gene, MZF1, is a hematopoietic transcription factor expressed in developing myeloid cells and is necessary to a granulocyte differentiation of human promyelocytic leukemia cell line HL-60 induced by retinoic acid (RA) [19]. ATF/CREB site is shown to be important for the constitutive expression of the promoter in breast cell lines and HepG2 cells [20]. It is known that GM3 synthase activity and GM3 ganglioside increase in HL-60 cells treated with the phorbol ester 12-O-tetradecanoylphorbol-13-acetate (TPA) that is a monocyte/macrophage differentiation inducer, but not with a granulocyte differentiation inducer, RA [21]. In addition, it has been suggested that protein kinase C (PKC) may specifically activate GM3 synthase, resulting in an increase
in the content of GM3 ganglioside during the differentiation of HL-60 cells induced by TPA, and that PKC activated by TPA may regulate the expression of GM3 synthase gene at the transcriptional level [21]. It has also been reported that IL-3 expression in murine myeloid leukemia cell line NFS60 result in an elevation of GM3 synthase activity, and this upregulation, together with the activities of downstream glycosyltransferase, causes the remarkable expression of GD1a. Very recently, it was reported that GM3 synthase activity and its mRNA were elevated in adipocyte-acquired insulin resistance by tumor necrosis factor-a (TNF-a) [23], speculating that the transcription of the GM3 synthase gene is activated by TNF signaling. The classic PKC activator TPA induced a marked increase in CREB phosphorylation in human IL-3-dependent myeloid cells (TF-1). IL-3 induced phosphorylation of CREB on Ser133 in TF-1 cells [24]. Macrophage differentiation of HL-60 cells induced by TPA resulted in marked increase in c-Jun mRNA [25], and cJun mRNA expression with PKC activation increases during TPA-induced monocytic differentiation of U937 cells [26]. c-Jun is constitutively expressed in HepG2 and NS20Y neuroblastoma cells [27] and has been described in several studies to heterodimerize with ATF/ CREB [28 –30]. Usually, the transcriptional activity of c-Jun is increased following phosphorylation on Ser63 and Ser73 by c-Jun N-terminal kinase (JNK) [31,32]. JNK, a member of mitogen-activated protein (MAP) kinase, is activated by TNF-a [33,34]. MAP kinases regulate AP1 transcriptional activity by multiple mechanisms [35]. In many cells, AP1, which consists of homo- and/or heterodimers of c-Jun and c-Fos, mediates gene expression in response to cell stimulation by TPA [36,37]. AP1 binds to the similar consensus sites as the ATF/CREB family as well as the TPA-response element (TRE) with the consensus sequence TGACTCA [29,37]. Interestingly, the consensus ATF/CREB site (TGACGTA) is completely conserved at position 143 to 136 in hST3Gal V promoter region. Furthermore, computer program-based analysis revealed the presence of CREBP1CJUN, which is recognized by CRE-binding protein and c-Jun heterodimer at the same position. It is possible, therefore, that the expression of hST3Gal V gene may be activated by PKC and MAP kinase signaling pathway. On the other hand, the region between 1600 and 432 seems to be responsible for the cell-type specific expression of hST3Gal V gene, since HepG2 cells had a low activity with any constructs examined. As far as we know, the present study is the first functional analysis of a regulatory region required for the expression of GM3 synthase gene, and we found a positive-regulatory region at positions 177 to 83 bp. There are many other putative regulatory elements in the 5V-flanking region of the hST3Gal V gene, and further studies remain to be conducted. Characterization of these potential negative and positive regulatory elements might provide further information as to the expression of hST3Gal V.
S.-W. Kim et al. / Biochimica et Biophysica Acta 1578 (2002) 84 – 89
Multiple mRNA forms that differ only in the 5V-UTRs have been identified in hST3Gal IV [38,39] and hST3Gal VI [40]. We recently identified four isoforms (types 1– 4) of the hST3Gal V mRNA that differ only in the 5V-UTRs [9]. These transcripts are produced by a combination of alternative splicing and promoter utilization, suggesting that tissue-specific transcriptional regulation of these genes depends on the use of alternative promoters. Further work is required to confirm this mechanism for the hST3Gal V gene.
Acknowledgements This work was supported by grant No. R01-2000-00129 from the Korea Science and Engineering Foundation.
References [1] L. Svennerholm, Gangliosides and synaptic transmission, in: L. Svennerholm, P. Mandel, H. Dreyfus, P. Urban (Eds.), Advances in Experimental Biology and Medicine: Structure and Function of Gangliosides, vol. 125, Plenum, New York, 1980, pp. 533 – 544. [2] A. Varki, Glycobiology 3 (1993) 97 – 130. [3] S. Hakomori, J. Biol. Chem. 265 (1990) 18713 – 18716. [4] H. Nojiri, F. Takaku, Y. Miura, M. Saito, Proc. Natl. Acad. Sci. U. S. A. 83 (1986) 782 – 786. [5] A. Tsunoda, M. Nakamura, K. Kirito, K. Hara, M. Saito, Biochemistry 34 (1995) 9356 – 9367. [6] P.H. Fishman, R.O. Brady, Science 194 (1976) 906 – 915. [7] S. Fukumoto, H. Miyazaki, G. Goto, T. Urano, K. Furukawa, K. Furukawa, J. Biol. Chem. 274 (1999) 9271 – 9276. [8] A. Ishii, M. Ohta, Y. Watanabe, K. Matsuda, K. Ishiyama, K. Sakoe, M. Nakamura, J. Inokuchi, Y. Sanai, M. Saito, J. Biol. Chem. 273 (1998) 31652 – 31655. [9] K.-W. Kim, S.-W. Kim, K.-S. Min, C.-H. Kim, Y.-C. Lee, Gene 273 (2001) 163 – 171. [10] K. Maruyama, S. Sugano, Gene 138 (1994) 171 – 174. [11] T. Murata, M. Yamaguchi, J. Biol. Chem. 274 (1999) 1277 – 1285. [12] M. Takakura, S. Kyo, T. Kanaya, H. Hirano, J. Takeda, M. Yutsudo, M. Inoue, Cancer Res. 59 (1999) 551 – 557. [13] N. Nakamura, Y. Suzuki, Y. Ikeda, M. Notoya, S. Hirose, J. Biol. Chem. 275 (2000) 28276 – 28284.
89
[14] K. Quandt, K. French, H. Karas, E. Wingender, T. Werner, Nucleic Acids Res. 23 (1995) 4878 – 4884. [15] E. Wingender, X. Chen, R. Hehl, H. Karas, I. Liebich, V. Matys, T. Meinhardt, M. PruUb, I. Reuter, F. Schacherer, Nucleic Acids Res. 28 (2000) 316 – 318. [16] W.S. Dynan, Trends Genet. 2 (1986) 196 – 197. [17] B.F. Pugh, R. Tjian, Cell 61 (1990) 1187 – 1197. [18] R. Mantovani, Gene 239 (1999) 15 – 27. [19] R. Hromas, S.J. Collins, D. Hickstein, W. Raskind, L.L. Deaven, P. O’Hara, F.S. Hagen, K. Kaushansky, J. Biol. Chem. 266 (1991) 14183 – 14187. [20] E. Atlas, M. Stramwasser, C.R. Mueller, Oncogene 20 (2001) 7110 – 7114. [21] T. Momoi, M. Shinmoto, J. Kasuya, H. Senoo, Y. Suzuki, J. Biol. Chem. 261 (1986) 16270 – 16273. [22] S. Tagami, J. Inokuchi, K. Kabayama, H. Yoshimura, F. Kitamura, S. Uemura, C. Ogawa, A. Ishii, M. Saito, Y. Ohtsuka, S. Sakaue, Y. Igarashi, J. Biol. Chem. 277 (2002) 3085 – 3092. [23] E. Gubina, X. Luo, E. Kwon, K. Sakamoto, Y.F. Shi, R.A. Mufson, J. Immunol. 167 (2001) 4303 – 4310. [24] R. Gaynor, K. Simon, P. Koeffler, Blood 77 (1991) 2618 – 2623. [25] R. Hass, I. Prudovsky, M. Kruhoffer, Leuk. Res. 21 (1997) 589 – 594. [26] L. SteinmuUller, G. Cibelli, J.R. Moll, C. Vinson, G. Thiel, Biochem. J. 360 (2001) 599 – 607. [27] D.M. Benbrook, N.C. Jones, Oncogene 5 (1990) 295 – 302. [28] T. Hai, T. Curran, Proc. Natl. Acad. Sci. U. S. A. 88 (1991) 3720 – 3724. [29] P.M. Chevray, D. Nathans, Proc. Natl. Acad. Sci. U. S. A. 89 (1992) 5789 – 5793. [30] T. Smeal, B. Binetruy, D.A. Mercola, M. Birrer, M. Karin, Nature 354 (1991) 494 – 496. [31] P. Paradis, W.R. MacLellan, N.S. Belaguli, R.J. Schwartz, M.D. Schneider, J. Biol. Chem. 271 (1996) 10827 – 10833. [32] J.M. Kyriakis, P. Banerjee, E. Nikolakaki, T. Dai, E.A. Rubie, M.F. Ahmad, J. Avruch, J.R. Woodgett, Nature 369 (1994) 156 – 160. [33] H.K. Sluss, T. Barrett, B. Derijard, R.J. Davis, Mol. Cell. Biol. 14 (1994) 8376 – 8384. [34] A.J. Whitmarsh, R.J. Davis, J. Mol. Med. 74 (1996) 589 – 607. [35] M. Karin, T. Smeal, Trends Biochem. Sci. 17 (1992) 418 – 422. [36] P. Angel, M. Karin, Biochim. Biophys. Acta 1072 (1991) 129 – 157. [37] H. Kitagawa, M.-G. Mattei, J.C. Paulson, J. Biol. Chem. 271 (1996) 931 – 938. [38] A. Taniguchi, K. Matsumoto, Biochem. Biophys. Res. Commun. 243 (1998) 177 – 183. [39] A. Taniguchi, R. Kaneta, K. Morishita, K. Matsumoto, Biochem. Biophys. Res. Commun. 287 (2001) 1148 – 1156.
Promoter paper
Isolation and characterization of the promoter region of the human GM3 synthase gene $ Sang-Wan Kim a, Sang-Hyeon Lee b, Kyoung-Sook Kim a, Cheorl-Ho Kim c, Young-Kug Choo d, Young-Choon Lee a,* a
Division of Biotechnology, Faculty of Natural Resources and Life Science, Dong-A University, 840, Hadan-Dong, Saha-Gu, Busan 604-714, South Korea b Department of Bioscience and Biotechnology, Silla University, Busan 617-736, South Korea c Department of Biochemistry and Molecular Biology, College of Oriental Medicine, Dongguk University, Kyung-Pook 780-350, South Korea d Division of Biological Science, College of Natural Science, Wonkwang University, Chonbuk 570-749, South Korea Received 9 April 2002; received in revised form 1 August 2002; accepted 2 September 2002
Abstract GM3 synthase, which transfers CMP-NeuAc with an a2,3-linkage to a galactose residue of lactosylceramide, plays a key role in the biosynthesis of all complex gangliosides. The expression of this gene is highly restricted in both human fetal and adult tissues. To elucidate the mechanisms that regulate the tissue-specific expression of the human GM3 synthase (hST3Gal V) gene, we have isolated and characterized its 5V-flanking region. Potential transcriptional start site was determined by CapSite hunting. Sequence analysis of the 5Vflanking region revealed that hST3Gal V promoter lacked canonical TATA and CAAT boxes but contained several putative binding sites for transcription factors AP4, MZF1, SP1, ATF/CREB, NFY, IK2 and LYF1, etc. Functional analysis of the 5V-flanking region of the hST3Gal V gene by transient expression method revealed that the 177 to 83 region is important for transcriptional activity of the hST3Gal V gene in SK-N-MC and HepG2 cells. The present results also suggested that both positive and negative regulatory elements are present in this TATAless promoter of the hST3Gal V gene. D 2002 Elsevier Science B.V. All rights reserved. Keywords: Promoter; GM3 synthase; Sialyltransferase; Ganglioside
Gangliosides are a family of sialic acid (NeuAc)-containing glycosphingolipids that are found in high concentration on the central nervous system [1]. They play important roles in a large variety of biological processes, such as cell – cell interaction, adhesion, cell differentiation, growth control and receptor function [2]. GM3 is the first and the simplest of the gangliosides and is known to play important roles in the modulation of cell growth through modified signal transduction and cell differentiation. GM3 inhibits tyrosine phosphorylation of the epidermal growth factor (EGF) receptor and EGF-dependent cell growth, independent of receptor – receptor interaction, whereas DeN-acetyl-GM3 enhances serine phosphorylation of the EGF
$ The nucleotide sequences reported in this paper have been submitted to the GenBank/EMBL with accession number AF488709. * Corresponding author. Tel.: +82-51-200-7591; fax: +82-51-200-6993. E-mail address: [email protected] (Y.-C. Lee).
receptor and stimulates cell proliferation [3]. GM3 induces monocytic differentiation of human myeloid and monocytoid leukemic cell lines such as HL-60 and U937 during macrophage-like cell differentiation [4]. The tumor-associated expression of ganglioside GD1a in NFS60 cells by introduction of an interleukin-3 (IL-3) gene is controlled by the activity of a single glycosyltransferase, GM3 synthase [5]. GM3 is synthesized by CMP-NeuAc:lactosylceramide a2,3-sialyltransferase (GM3 synthase, EC 2.4.99.9) which catalyzes the transfer of NeuAc from CMP-NeuAc to the nonreducing terminal galactose of lactosylceramide. GM3 synthase is a key regulatory enzyme for ganglioside biosynthesis [6] because it catalyzes the first committed step in the synthesis of nearly all gangliosides. Although its cDNA has been cloned from mouse [7] and human tissue [8], its genomic structure and transcriptional regulation has not been elucidated. Very recently, we determined the genomic structure of the human GM3 synthase (hST3Gal V) gene and identified alternatively spliced isoforms in its 5V-
0167-4781/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 7 8 1 ( 0 2 ) 0 0 5 0 5 - 5
S.-W. Kim et al. / Biochimica et Biophysica Acta 1578 (2002) 84 – 89
untranslated region (UTR) [9]. The hST3Gal V gene is abundantly expressed in some tissues such as the brain and placenta [9]. To elucidate the molecular basis of hST3Gal V gene expression, in this study, the genomic region containing the promoter of the hST3Gal V was isolated and functionally characterized. To determine the transcription start site of the hST3Gal V gene, the CapSitek hunting method [10 –13] was used in accordance with the manufacturer’s protocol (Nippon Gene, Tokyo, Japan). CapSite cDNAR dT from human fetal brain, in which the 5V-terminal m7GpppN cap structure of mRNA was removed and recapped by the 3V-end of a specific rOligo primer provided by the company, was subjected to nested PCR. The first round PCR was performed using the rOligo-specific primer 1RDT and hST3Gal V gene-specific primer TGP1 (Table 1). The second round PCR was performed using the rOligo-specific primer 2RDT and hST3Gal V gene-specific primer TGP2. The reactions were performed using the following conditions: 95 jC for 5 min and then 35 cycles of 95 jC for 20 s, 60 jC for 20 s, and 72 jC for 90 s, with a final elongation of 72 jC for 5 min. The PCR product of about 0.34 kb fragment was obtained (Fig. 1), subcloned into pT7Blue(R) T-vector (Novagen, Inc., Madison) and sequenced. The transcription start site was determined by identification of the boundary sequence between rOligo and hST3Gal V mRNA sequence. The result showed that its sequence and 5V-end are completely identical to those of the type 2 isoform [9] obtained by 5Vrapid amplification of cDNA ends (5V-RACE). Therefore, we assigned the major initiation site of transcription of
Fig. 1. Identification of the cap site by nested PCR for the determination of transcription start site of the hST3Gal V. CapSite cDNA from human fetal brain was subjected to nested PCR. The first round PCR was performed using the rOligo-specific primer 1RDT and hST3Gal V gene-specific primer TGP1, whose positions are shown in (A). The second round PCR was performed using the rOligo-specific primer 2RDT and hST3Gal V gene-specific primer TGP2, whose positions are also shown in (A). The resulting product (lane 1) was analyzed by 2% agarose gel electrophoresis and visualized by ethidium bromide staining (B). M, DNA size marker (ØX174 digested with Hae III).
85
hST3Gal V gene to the A residue at position designed as + 1 (Fig. 2). Based on this finding, the upstream genomic region of the hST3Gal V was identified using the Human Genome Resources of NCBI, NLM, and NIH. A 1599 bp fragment of the 5V-flanking sequence of the hST3Gal V gene was obtained by long and accurate PCR (LA-PCR) amplification with LA-Taq polymerase (Takara Shuzo, Shiba, Japan). LAPCR was performed with primers FP1600 and RP1 (Table 1), and human genomic DNA (Clontech, Palo Alto, CA) as a template. The reactions were performed using the following conditions: 94 jC for 1 min and then 30 cycles of 98 jC for 20 s and 68 jC for 3 min, with a final elongation of 72 jC for 10 min. The PCR product was subcloned into pT7Blue(R) T-vector (Novagen) to give pThGM3 and sequenced in both directions by cloning convenient restriction fragments into pUC119 or using primers designed from known sequence. Analysis of the 5V-flanking region of the hST3Gal V gene by the MatInspector v2.2 program (core similarity 1.0, matrix similarity 0.95) using TRANSFAC 4.0 matrices [14,15] revealed that this region lacks canonical TATA and CAAT boxes, but contains several putative transcription factor binding sites such as AP4, MZF1, SP1, ATF/CREB, NFY, IK2 and LYF1, etc. (Fig. 2). This region also contains an apparent G + C-rich region between 1 and 304 (GC content, 81%). To determine whether the genomic DNA flanking exon 1 contains a functional promoter, the 1.6 kb fragment of pThGM3 containing 1600 bp of the 5V-flanking region of hST3Gal V gene and the first 15 bp of exon 1 was inserted into the SacI and BglII sites of pCAT3-Basic plasmid, a chloramphenicol acetyltransferase (CAT) expression vector lacked promoter and enhancer (Promega, Madison, WI). This construct pCAT-1600 (2 Ag) and pCAT3-Basic (2 Ag) as a negative control were transfected into the human neuroblastoma cell line SK-N-MC and human hepatoma cell line HepG2 at 70% confluence by lipofectamine-mediated transfection protocol (Gibco BRL, Life Technologies). SK-N-MC cells were cultured in minimum essential medium (MEM) supplemented with 10% fetal bovine serum, 1 mM sodium pyruvate and 1 MEM nonessential amino acids. HepG2 cells were grown in Dulbecco’s modified Eagle medium (DMEM) containing 10% FBS. Each of the transfection experiments were repeated at least twice, yielding reproducible results. To normalize for the efficiency of transfection, these cells were simultaneously cotransfected with 1 Ag of pCMVh (Clontech). Cells were harvested 48 h after transfections. Lysates were prepared by four cycles of freezing and thawing of the harvested cells followed by centrifugation. CAT assays were carried out using the CAT-enzyme linked immunosorbent assay (ELISA) kit (Boehringer Mannheim). CAT activity was normalized to h-galactosidase activity and expressed as a fold increase over pCAT3-Basic. As shown in Fig. 3, the result of the CAT assay clearly demonstrated that the 5V-flanking region of hST3Gal V
86
S.-W. Kim et al. / Biochimica et Biophysica Acta 1578 (2002) 84 – 89
Fig. 2. Nucleotide sequence of the 5V-flanking region of the hST3Gal V. The sequence is numbered with the first nucleotide of the transcription start site as + 1. The exon 1 sequence is shown in bold lowercase letters, and the start position of intron 1 is marked with an arrow. The putative transcription factor binding sites are described in the text. For the detection of promoter activity, the start point of each construction is indicated by an arrowhead.
S.-W. Kim et al. / Biochimica et Biophysica Acta 1578 (2002) 84 – 89
87
Fig. 3. Promoter activity of the 5V-flanking region of the hST3Gal V gene in SK-N-MC and HepG2 cells. Schematic representation of DNA constructions containing various lengths of the 5V-flanking region of the hST3Gal V linked to the CAT reporter gene is presented. The restriction sites are shown and the transcription start site is indicated as + 1. pCAT3-Basic is the promoterless and enhancerless vector control. Each CAT constructs were co-transfected into SKN-MC and HepG2 cells with pCMVh as the internal control. CAT activity was measured using the CAT-ELISA kit and normalized with h-galactosidase activity derived from pCMVh. CAT activity is expressed as a fold increase over reporter plasmid pCAT3-Basic (negative control). Results presented are the mean F S.E. for three to four plates. * ANOVA P < 0.0001 compared with pCAT3-Basic. Transfections were repeated at least twice, yielding reproducible results.
gene displayed transcription activity in both cell lines. Transcription activity of pCAT-1600 in SK-N-MC cells was 3-fold higher than that in HepG2 cells. This elevated transcription activity in a neuronal cell line is consistent with the recent report that showed higher levels of hST3Gal V mRNA in human fetal and adult brain tissues than human liver tissues [9]. This also coincides with the result that both cell lines express hST3Gal V mRNA, but expression level was approximately 2.5-fold higher in SKN-MC cells than HepG2 cells (data not shown). Therefore, these results suggest that the hST3Gal V promoter reveals a cell-specific manner. To characterize the region regulating the transcription activity of the hST3Gal V, hST3Gal V promoter fragments containing varying lengths of 5V-flanking sequence were
generated by LA-PCR using pThGM3 as template. Forward primers FP1210, FP847, FP432, FP177 and FP83 containing SacI sites, and a common reverse primer RP1 containing a BglII site were used (Table 1). To avoid translation from the first ATG at position + 4 of the genomic fragment, which results in frame-shift reading of CAT cDNA, the A of the first ATG codon was replaced with T in a common reverse RP1 primer. The PCR fragments were subcloned into pT7Blue(R) T-vector (Novagen) and sequenced. Each fragment obtained by digestion with SacI and BglII was inserted into the pCAT3-Basic plasmid. The orientations of each construct were confirmed by restriction enzyme analysis. These constructs were designated as pCAT-1210, pCAT-847, pCAT-432, pCAT-177 and pCAT-83 (Fig. 3). All plasmids used in transfection experiments were prepared
Table 1 Primers used in this study Primer
Sequence
Strand
1RDT TGP1 2RDT TGP2 FP1600 FP1210 FP847 FP432 FP177 FP83 RP1
5V-GATGCTAGCTGCGAGTCAAGTC-3V 5V-CCTTCTGCAAGACTTGCTGAG-3V 5V-CGAGTCAAGTCGACGAAGTGC-3V 5V-TGGTCAGGGTCCACATAATGC-3V 5V-ATCGAGCTCAACAGACCCAGACTTGAGAGAGTCC-3V 5V-ATCGAGCTCCAGCTGGTCTTGAACTCC-3V 5V-ATCGAGCTCGCTCCCAATATCCTCACG-3V 5V-ATCGAGCTCACGTTCAGCTGTGTCCAAG-3V 5V-ATCGAGCTCATCGTGCTCTCCGATT-3V 5V-ATCGAGCTCAGCTGAATGGGCGC-3V 5V-AAGAGATCTTCGTCCGCAAACTAATGAGGGGGC-3V
sense antisense sense antisense sense sense sense sense sense sense antisense
Position + 358 to + 338 + 314 to + 294 1600 to 1576 1210 to 1192 847 to 830 432 to 414 177 to 161 83 to 69 + 15 to 11
The primers used for cloning into the reporter vector, pCAT3-Basic, include the following restriction enzyme sites: SacI, underlined text; BglII, italic underlined text. The A of the first ATG codon replaced with T is shown in boldface in a common reverse RP1 primer.
88
S.-W. Kim et al. / Biochimica et Biophysica Acta 1578 (2002) 84 – 89
by Qiagen Midi-prep kit (Qiagen, Valencia, CA). The promoter activities of these deletion constructs were tested in SK-N-MC and HepG2 cells as described above. In SK-NMC cells, as shown in Fig. 3, deletion from 1600 to 1210 resulted in about 34% decrease in transcription activity compared to pCAT-1600. Further deletion downstream to base 432 caused a gradual decrease of the promoter activity. However, deletion in the region from 432 to 177 resulted in increase of transcription activity, while deletion in the region from 177 to 83 markedly reduced transcription activity to the level of the promoterless and enhancerless control vector pCAT3-Basic. These results suggest that potential positive regulatory elements exist within the 1600 to 432 and 177 to 83 region, while potential negative regulatory elements exist within the 432 to 177 region. On the other hand, in HepG2 cells, no remarkable transcription activities were observed in the deletion mutants in the region from 1600 to 432 compared to pCAT3-Basic. However, deletion in the region from 432 to 177 resulted in significant increase in transcription activity, while deletion in the region from 177 to 83 caused a drastic decrease to the basal level like SK-N-MC cells, suggesting that potential negative and positive regulatory elements exist within the 432 to 177 and 177 to 83 regions, respectively. These results show that the region between 177 and 83 functions as the core promoter essential for transcriptional activation of hST3Gal V in SK-N-MC and HepG2 cells. This also suggests that promoter elements located between nucleotide positions 117 and 83 are essential for basal expression of the hST3Gal V gene in both cell lines. As shown in Fig. 2, this region from 177 to 83 is GC-rich (G-C content, 74%) and contains NFY, ATF/CREB, SP1, EGR3 and MZF1 binding sites. These binding sites are present in the inverted orientation except for ATF/CREB and MZF1. SP1 is known to be important for the transcriptional activation of promoters which lack TATA and CAAT boxes [16,17]. NFY is a ubiquitous transcription factor which binds to CCAAT element in direct or inverted orientation in the proximal promoters of a wide variety of mammalian genes and its binding is frequently essential for gene transcription, particularly for TATA-less promoters [18]. The myeloid zinc finger gene, MZF1, is a hematopoietic transcription factor expressed in developing myeloid cells and is necessary to a granulocyte differentiation of human promyelocytic leukemia cell line HL-60 induced by retinoic acid (RA) [19]. ATF/CREB site is shown to be important for the constitutive expression of the promoter in breast cell lines and HepG2 cells [20]. It is known that GM3 synthase activity and GM3 ganglioside increase in HL-60 cells treated with the phorbol ester 12-O-tetradecanoylphorbol-13-acetate (TPA) that is a monocyte/macrophage differentiation inducer, but not with a granulocyte differentiation inducer, RA [21]. In addition, it has been suggested that protein kinase C (PKC) may specifically activate GM3 synthase, resulting in an increase
in the content of GM3 ganglioside during the differentiation of HL-60 cells induced by TPA, and that PKC activated by TPA may regulate the expression of GM3 synthase gene at the transcriptional level [21]. It has also been reported that IL-3 expression in murine myeloid leukemia cell line NFS60 result in an elevation of GM3 synthase activity, and this upregulation, together with the activities of downstream glycosyltransferase, causes the remarkable expression of GD1a. Very recently, it was reported that GM3 synthase activity and its mRNA were elevated in adipocyte-acquired insulin resistance by tumor necrosis factor-a (TNF-a) [23], speculating that the transcription of the GM3 synthase gene is activated by TNF signaling. The classic PKC activator TPA induced a marked increase in CREB phosphorylation in human IL-3-dependent myeloid cells (TF-1). IL-3 induced phosphorylation of CREB on Ser133 in TF-1 cells [24]. Macrophage differentiation of HL-60 cells induced by TPA resulted in marked increase in c-Jun mRNA [25], and cJun mRNA expression with PKC activation increases during TPA-induced monocytic differentiation of U937 cells [26]. c-Jun is constitutively expressed in HepG2 and NS20Y neuroblastoma cells [27] and has been described in several studies to heterodimerize with ATF/ CREB [28 –30]. Usually, the transcriptional activity of c-Jun is increased following phosphorylation on Ser63 and Ser73 by c-Jun N-terminal kinase (JNK) [31,32]. JNK, a member of mitogen-activated protein (MAP) kinase, is activated by TNF-a [33,34]. MAP kinases regulate AP1 transcriptional activity by multiple mechanisms [35]. In many cells, AP1, which consists of homo- and/or heterodimers of c-Jun and c-Fos, mediates gene expression in response to cell stimulation by TPA [36,37]. AP1 binds to the similar consensus sites as the ATF/CREB family as well as the TPA-response element (TRE) with the consensus sequence TGACTCA [29,37]. Interestingly, the consensus ATF/CREB site (TGACGTA) is completely conserved at position 143 to 136 in hST3Gal V promoter region. Furthermore, computer program-based analysis revealed the presence of CREBP1CJUN, which is recognized by CRE-binding protein and c-Jun heterodimer at the same position. It is possible, therefore, that the expression of hST3Gal V gene may be activated by PKC and MAP kinase signaling pathway. On the other hand, the region between 1600 and 432 seems to be responsible for the cell-type specific expression of hST3Gal V gene, since HepG2 cells had a low activity with any constructs examined. As far as we know, the present study is the first functional analysis of a regulatory region required for the expression of GM3 synthase gene, and we found a positive-regulatory region at positions 177 to 83 bp. There are many other putative regulatory elements in the 5V-flanking region of the hST3Gal V gene, and further studies remain to be conducted. Characterization of these potential negative and positive regulatory elements might provide further information as to the expression of hST3Gal V.
S.-W. Kim et al. / Biochimica et Biophysica Acta 1578 (2002) 84 – 89
Multiple mRNA forms that differ only in the 5V-UTRs have been identified in hST3Gal IV [38,39] and hST3Gal VI [40]. We recently identified four isoforms (types 1– 4) of the hST3Gal V mRNA that differ only in the 5V-UTRs [9]. These transcripts are produced by a combination of alternative splicing and promoter utilization, suggesting that tissue-specific transcriptional regulation of these genes depends on the use of alternative promoters. Further work is required to confirm this mechanism for the hST3Gal V gene.
Acknowledgements This work was supported by grant No. R01-2000-00129 from the Korea Science and Engineering Foundation.
References [1] L. Svennerholm, Gangliosides and synaptic transmission, in: L. Svennerholm, P. Mandel, H. Dreyfus, P. Urban (Eds.), Advances in Experimental Biology and Medicine: Structure and Function of Gangliosides, vol. 125, Plenum, New York, 1980, pp. 533 – 544. [2] A. Varki, Glycobiology 3 (1993) 97 – 130. [3] S. Hakomori, J. Biol. Chem. 265 (1990) 18713 – 18716. [4] H. Nojiri, F. Takaku, Y. Miura, M. Saito, Proc. Natl. Acad. Sci. U. S. A. 83 (1986) 782 – 786. [5] A. Tsunoda, M. Nakamura, K. Kirito, K. Hara, M. Saito, Biochemistry 34 (1995) 9356 – 9367. [6] P.H. Fishman, R.O. Brady, Science 194 (1976) 906 – 915. [7] S. Fukumoto, H. Miyazaki, G. Goto, T. Urano, K. Furukawa, K. Furukawa, J. Biol. Chem. 274 (1999) 9271 – 9276. [8] A. Ishii, M. Ohta, Y. Watanabe, K. Matsuda, K. Ishiyama, K. Sakoe, M. Nakamura, J. Inokuchi, Y. Sanai, M. Saito, J. Biol. Chem. 273 (1998) 31652 – 31655. [9] K.-W. Kim, S.-W. Kim, K.-S. Min, C.-H. Kim, Y.-C. Lee, Gene 273 (2001) 163 – 171. [10] K. Maruyama, S. Sugano, Gene 138 (1994) 171 – 174. [11] T. Murata, M. Yamaguchi, J. Biol. Chem. 274 (1999) 1277 – 1285. [12] M. Takakura, S. Kyo, T. Kanaya, H. Hirano, J. Takeda, M. Yutsudo, M. Inoue, Cancer Res. 59 (1999) 551 – 557. [13] N. Nakamura, Y. Suzuki, Y. Ikeda, M. Notoya, S. Hirose, J. Biol. Chem. 275 (2000) 28276 – 28284.
89
[14] K. Quandt, K. French, H. Karas, E. Wingender, T. Werner, Nucleic Acids Res. 23 (1995) 4878 – 4884. [15] E. Wingender, X. Chen, R. Hehl, H. Karas, I. Liebich, V. Matys, T. Meinhardt, M. PruUb, I. Reuter, F. Schacherer, Nucleic Acids Res. 28 (2000) 316 – 318. [16] W.S. Dynan, Trends Genet. 2 (1986) 196 – 197. [17] B.F. Pugh, R. Tjian, Cell 61 (1990) 1187 – 1197. [18] R. Mantovani, Gene 239 (1999) 15 – 27. [19] R. Hromas, S.J. Collins, D. Hickstein, W. Raskind, L.L. Deaven, P. O’Hara, F.S. Hagen, K. Kaushansky, J. Biol. Chem. 266 (1991) 14183 – 14187. [20] E. Atlas, M. Stramwasser, C.R. Mueller, Oncogene 20 (2001) 7110 – 7114. [21] T. Momoi, M. Shinmoto, J. Kasuya, H. Senoo, Y. Suzuki, J. Biol. Chem. 261 (1986) 16270 – 16273. [22] S. Tagami, J. Inokuchi, K. Kabayama, H. Yoshimura, F. Kitamura, S. Uemura, C. Ogawa, A. Ishii, M. Saito, Y. Ohtsuka, S. Sakaue, Y. Igarashi, J. Biol. Chem. 277 (2002) 3085 – 3092. [23] E. Gubina, X. Luo, E. Kwon, K. Sakamoto, Y.F. Shi, R.A. Mufson, J. Immunol. 167 (2001) 4303 – 4310. [24] R. Gaynor, K. Simon, P. Koeffler, Blood 77 (1991) 2618 – 2623. [25] R. Hass, I. Prudovsky, M. Kruhoffer, Leuk. Res. 21 (1997) 589 – 594. [26] L. SteinmuUller, G. Cibelli, J.R. Moll, C. Vinson, G. Thiel, Biochem. J. 360 (2001) 599 – 607. [27] D.M. Benbrook, N.C. Jones, Oncogene 5 (1990) 295 – 302. [28] T. Hai, T. Curran, Proc. Natl. Acad. Sci. U. S. A. 88 (1991) 3720 – 3724. [29] P.M. Chevray, D. Nathans, Proc. Natl. Acad. Sci. U. S. A. 89 (1992) 5789 – 5793. [30] T. Smeal, B. Binetruy, D.A. Mercola, M. Birrer, M. Karin, Nature 354 (1991) 494 – 496. [31] P. Paradis, W.R. MacLellan, N.S. Belaguli, R.J. Schwartz, M.D. Schneider, J. Biol. Chem. 271 (1996) 10827 – 10833. [32] J.M. Kyriakis, P. Banerjee, E. Nikolakaki, T. Dai, E.A. Rubie, M.F. Ahmad, J. Avruch, J.R. Woodgett, Nature 369 (1994) 156 – 160. [33] H.K. Sluss, T. Barrett, B. Derijard, R.J. Davis, Mol. Cell. Biol. 14 (1994) 8376 – 8384. [34] A.J. Whitmarsh, R.J. Davis, J. Mol. Med. 74 (1996) 589 – 607. [35] M. Karin, T. Smeal, Trends Biochem. Sci. 17 (1992) 418 – 422. [36] P. Angel, M. Karin, Biochim. Biophys. Acta 1072 (1991) 129 – 157. [37] H. Kitagawa, M.-G. Mattei, J.C. Paulson, J. Biol. Chem. 271 (1996) 931 – 938. [38] A. Taniguchi, K. Matsumoto, Biochem. Biophys. Res. Commun. 243 (1998) 177 – 183. [39] A. Taniguchi, R. Kaneta, K. Morishita, K. Matsumoto, Biochem. Biophys. Res. Commun. 287 (2001) 1148 – 1156.