Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

N‐myc enhances the expression of a large set of genes functioning in ribosome biogenesis and protein synthesis

N‐myc enhances the expression of a large set of genes functioning in ribosome biogenesis and... The EMBO Journal Vol. 20 No. 6 pp. 1383-1393, 2001 N-myc enhances the expression of a large set of genes functioning in ribosome biogenesis and protein synthesis 1 2 1 3 follow a very aggressive course (Schwab et al., 1983; Kathy Boon • , Huib N.Caron , , 1 1 Seeger et al., 1985). Overexpression of transfected N-myc Ronald van Asperen , Linda Valentijn , 1 1 genes in neuroblastoma cell lines strongly increased Marie-Christine Hermus , Peter van Sluis , 1 4 3 proliferation rates (Bemards et al., 1986; Lutz et al., llja Roobeek , Isabel Weis , P.A.Vo0te , 3 1 5 1996). Transgenic mice overexpressing N-myc in neural Manfred Schwab and Rogier Versteeg , crest-derived tissues showed frequent development of Department of Human Genetics, Academic Medical Center, neuroblastomas (Weiss et al., 1997). Numerous compar­ University of Amsterdam, PO Box 22700, ll00 DE Amsterdam, able observations have implicated c-myc and L-myc in the Department of Paediatric Oncology and Haematology, Emma Kinder pathogenesis of many other tumor types (Cole, 1986; Ziekenhuis, Academic Medical Center, Amsterdam, The Netherlands Henriksson and Luscher, 1996). While c-myc and N-myc and Division of Cytogenetics, Deutsches Krebsforschungszentrum, Heidelberg, Germany homozygous knockout mice are embryonic lethal, trans­ genic mice in which N-myc replaced c-myc showed a gross Present address: Department of Pathology, Duke University Medical Center, Durham, NC 27710, USA normal development, indicating that both proteins have largely overlapping functions (Malynn et al., 2000). Corresponding author Many experiments have suggested a role for myc genes e-mail: [email protected] in cell cycle control, metastasis, blocking of differenti­ ation, apoptosis and proliferation rate (Henriksson and The myc oncogenes are frequently activated in human Luscher, 1996; Dang, 1999; Schmidt, 1999). The myc tumors, but there is no comprehensive insight into the family members are transcription factors with a basic/ target genes and downstream cellular pathways of helix-loop-helix/leucine zipper (bHLHzip) domain. They these transcription factors. We applied serial analysis form heterodimers with Max proteins and bind to the of gene expression (SAGE) to identify targets of E-box motif CACGTG to activate target gene transcription N-myc in neuroblastomas. Analysis of 42 000 mRNA (Alex et al., 1992; Blackwood et al., 1992). The limited transcript tags in SAGE libraries of N-myc­ number of initially identified targets of c-myc gave little transfected and control neuroblastoma cells revealed insight into the mechanism of how c-myc induces the 114 up-regulated genes. The majority of these genes variety of phenotypes. Examples of myc target genes have a role in ribosome assembly and activity. are prothymosin a, omithine decarboxylase, CAD and Northern blot analysis confirmed up-regulation of all the DEAD-box gene MrDb (Eilers et al., 1991; tested transcripts. Induction was complete within 4 h Bello-Fernandez et al., 1993; Rosenwald et al., 1993; after N-myc expression. The large majority of the Grandori et al., 1996; Jones et al., 1996; Boyd and ribosomal proteins were induced, as well as genes con­ Farnham, 1997). Induction of cyclins Dl, E and A and trolling rRNA maturation. Cellular rRNA content was cdc25A were found in some but not all model systems 45% induced. SAGE libraries and northern blot (reviewed in Obaya et al., 1999). Induction of cyclin D2 analysis confirmed up-regulation of many of these was consistently found and provides a direct link to the cell genes in N-myc-amplified neuroblastomas. As N-myc cycle (Bouchard et al., 1999; Perez-Roger et al., 1999; can functionally replace c-myc, we analyzed whether Coller et al., 2000). Furthermore, c-myc and N-myc also N-myc targets were induced by c-myc as well. induce Id2 transcription, thus stimulating Rb inactivation Approximately 40% of these N-myc targets were up­ regulated in a c-myc-transfected melanoma cell line. and cell cycle progression (Lasorella et al., 2000). These data suggest that myc genes function as major However, in addition to these direct effects on prolifer­ regulators of the protein synthesis machinery. ation, several findings suggest that c-myc also promotes Keywords: c-myc/N-myc/neuroblastoma/protein the growth of cells, thereby inducing an increase in cell synthesis/ribosomes mass that may be a prerequisite for rapid cell proliferation. Impaired in viva expression of Drosophila dmyc results in adult flies half the normal size (Johnston et al., 1999). Both the volume and proliferation rate of their cells are reduced, Introduction and could be restored by overexpression of dmyc. B cells The members of the myc oncogene family play a of transgenic mice with overexpression of c-myc show for prominent role in cancer. N-myc, c-myc and L-myc are all differentiation stages an increased cell size and rearranged, amplified, mutated and/or overexpressed in increased protein synthesis rate (Iritani and Eisenman, many human tumor types. The c-myc gene is expressed in 1999). Furthermore, fibroblasts with a homozygous a wide variety of tissues and tumors, while N-myc inactivation of c-myc have a reduced proliferation rate as expression is largely restricted to embryonic tissues and well as a reduced protein synthesis rate (Mateyak et al., neuroendocrine tumors. Approximately 20% of neuro­ 1997). Conversely, activation of c-myc in fibroblasts blastomas have N-myc amplification, and these tumors activates protein synthesis (Schmidt, 1999). Furthermore, © European Molecular Biology Organization 1383 K.Boon et al. a role for c-myc in protein synthesis is suggested by the Table I. Summary of neuroblastoma SAGE libraries finding that c-myc induces the translation initiation factors Total tags Different transcripts elF-4E, eIF-2-a (Rosenwald et al., 1993; Jones et al., 1996), eIF5A and eIF4G (Coller et al., 2000). Very Neuroblastoma cell lines recently, ectopic in viva expression of c-myc was found to SHEP-2 20 950 10 154 result in increased expression of six ribosomal protein SHEP-21N (N-myc transfected) 20 938 8566 genes in mouse liver (Kim et al., 2000). A role for myc Neuroblastoma tumors N52 19 597 9356 genes in growth regulation is in line with their effect on the N159 (N-myc amplified) 20 001 10 262 cell cycle. Inactivation of c-myc in fibroblasts prolonged Total 81 486 the G and G phases of the cell cycle, but not the S phase, while high expression of myc genes accelerated transition through G (Steiner et al., 1995; Lutz et al., 1996). A similar effect was found for Drosophila dmyc (Johnston functions in ribosome biogenesis and activity and in later et al., 1999). steps of the protein synthesis and protein turnover Here we describe the use of SAGE (serial analysis of machinery. gene expression; Velculescu et al., 1995) to identify the downstream genes that are activated by N-myc in human neuroblastoma. To date, only prothymosin a, ornithine Induction of genes involved in ribosome decarboxylase and Id2 have been identified as targets of biogenesis and protein synthesis N-myc (Lutz et al., 1996; Lasorella et al., 2000). The The first functional group consists of 56 ribosomal protein analysis of the expression level of >40 000 transcripts genes, which were induced up to 37-fold (P <0.01; identified 114 genes up-regulated in N-myc-expressing Table II). Tags for 10 more ribosomal protein genes were cells. Our results indicate that N-myc functions as a less strongly up-regulated (0.01> P >0.05), with induction regulator of cell growth by stimulating genes functioning levels from 1.4- to 8-fold (Table II, lower part). These 66 in ribosome biogenesis and protein synthesis. Several of tags correspond to 82% of the human ribosomal proteins the identified genes are induced by N-myc as well as by (Wool and Chan, 1996). Northern blot analysis of SHEP-2 c-myc. and SHEP-21N with probes for seven ribosomal protein genes (Sl2, S27, Fau-S30, L8, S6, S19 and PPARP0) confirmed their up-regulation (Figure 1). The fraction of Results tags for ribosomal protein mRNAs was increased from SAGE libraries of N-myc-transfected 4.1 % in SHEP-2 to 12.6% in SHEP-21N. The level of neuroblastoma cell lines induction of individual ribosomal protein genes is a To identify the downstream target genes of N-myc, we function of their basal expression levels in SHEP-2. applied the SAGE technique to an N-myc-transfected Highly expressed genes are less induced than genes with a neuroblastoma cell line. The SHEP cell line has no N-myc low basic expression in SHEP-2 (Figure 2). amplification and expression, or c-myc expression. A A second functional group of genes up-regulated in tetracycline-dependent N-myc expression vector has been SHEP-21N consists of two nucleolar protein genes introduced into these cells, resulting in the SHEP-21N (Table III). Nucleophosmin (or B23) is induced from 26 clone (Lutz et al., 1996). The SHEP-21N cells have to 55 tags (P <0.001), which was confirmed by northern constitutive exogenous N-myc expression that can be blot analysis (Figure 1). Nucleophosmin is an abundant switched off by tetracycline. N-myc expression in the nucleolar protein that processes rRNA by cleavage of the SHEP-21N cells was shown to increase the rate of cell 5' end of the 5.8S pre-rRNA (Zhang et al., 1997) and division, shorten the G phase of the cell cycle and render functions in assembly and nuclear-cytoplasmic shuttling the cells more susceptible to apoptotic triggers (Lutz et al., of pre-ribosomal particles (Borer et al., 1989; Olson, 1996; Fulda et al., 1999). SAGE libraries were constructed 1991). Transcription of another nucleolar protein is also induced: nucleolin, which has two tags due to alternative from SHEP-21N cells expressing N-myc and from SHEP-2 control cells. The SHEP-2 clone was transfected with the transcripts, is induced from 4 to 12 tags in total empty expression vector. From each library, we sequenced (P = 0.035). This induction was confirmed by northern ~21 000 transcript tags, corresponding to 8566 blot analysis (Figure 1). Nucleolin processes pre-rRNA to (SHEP-21N) and 10 154 (SHEP-2) different transcripts mature 18S rRNA (Ginisty et al., 1998, 1999). It binds to (Table I). A tag for the transfected N-myc construct has a nucleophosmin and is also involved in the assembly of frequency of 0 and 8 in SHEP-2 and SHEP-21N, pre-ribosomal particles and their nucleo-cytoplasmic respectively. Comparison of the SAGE libraries yielded transport. The induction of nucleolin and nucleophosmin 114 significantly (P <0.01) up-regulated tags in N-myc­in SHEP-21N suggests that rRNA and ribosome biosynth­ expressing cells, with induction levels of up to 37-fold esis are targets of N-myc stimulation. (Tables II-IV). Another 70 tags were significantly down­ Additionally, tags corresponding to nine translation regulated. Here we focused on the analysis of a series of initiation and elongation factors were induced (Table III). tags induced in the N-myc-transfected cells. The tran­ They are eukaryotic translation initiation factors eIF3s8, scripts corresponding to these tags were identified using eIF4A and eIF5A, and the subunits a, 13, y and 8 of the SAGEmap database (Lal et al., 1999) and our own tag translation elongation factor 1 (EEFl). Furthermore, assignment program (Caron et al., 2001) and checked by elongation factor 2 and the mitochondrial elongation mRNA and expressed sequence tag (EST) sequence factor Tu (tuFM) are up-regulated. Northern blot analysis analyses. A comprehensive set of up-regulated genes of SHEP-21N and SHEP-2 confirmed the induction of 1384 1.9 2.9 4.5 N-myc target genes function in ribosome biogenesis Table II. Downstream targets induced by N-myc: ribosomal proteins Tag sequence SHEP-2 SHEP-21N Fold induction P value Unigene Hs. Gene GCCGAGGAAG 1 37.0 <0.001 82148 ribosomal protein S12 GCTTTTAAGG 1 29 29.0 <0.001 8102 ribosomal protein S20 CCCATCCGAA 1 23 23.0 <0.001 91379 ribosomal protein L26 GGCCGCGTTC 0 23 >23 <0.001 5174 ribosomal protein S 17 CCAGTGGCCC 0 21 >21 <0.001 180920 ribosomal protein S9 GTGTTGCACA 1 16 16.0 165590 ribosomal protein S 13 <0.001 GATGCTGCCA 1 16 16.0 <0.001 99914 ribosomal protein L22 CCGTCCAAGG 2 31 15.5 <0.001 80617 ribosomal protein S 16 GGAGTGGACA 1 14 14.0 <0.001 75458 ribosomal protein L18 GCCTGTATGA 2 27 13.5 <0.001 180450 ribosomal protein S24 GTTCCCTGGC 2 26 13.0 <0.001 177415 ribosomal protein Fau-S30 ATGGCTGGTA 6 72 12.0 <0.001 182426 ribosomal protein S2 GTGTTAACCA 1 11 11.0 <0.001 74267 ribosomal protein L10 CACAAACGGT 4 10.8 <0.001 195453 ribosomal protein S27 CTCAACATCT 3 32 10.7 <0.001 73742 ribosomal protein, large, PO GTTCGTGCCA 2 18 9.0 <0.001 179666 ribosomal protein L35a GACGACACGA 4 30 7.5 <0.001 153177 ribosomal protein S28 TCGTCTTTAT 3 21 7.0 <0.001 75538 ribosomal protein S7 GGACCACTGA 5 34 6.8 <0.001 119598 ribosomal protein L3 CCTCGGA A AA 4 24 6.0 <0.001 2017 ribosomal protein L38 AATCCTGTGG 8 48 6.0 <0.001 178551 ribosomal protein L8 ATCAAGGGTG 4 21 5.3 <0.001 157850 ribosomal protein L9 GGGCTGGGGT 20 101 5.1 <0.001 183698 ribosomal protein L29 AAGGAGATGG 5 25 5.0 <0.001 184014 ribosomal protein L3 l/tag matches mitochondrial sequences AAGGTGGAGG 11 55 5.0 <0.001 163593 ribosomal protein L18a TTACCATATC 10 49 4.9 <0.001 177461 ribosomal protein L39 GTGAAGGCAG 6 27 4.5 <0.001 77039 ribosomal protein S3A GAACACATCC 4 18 0.002 75879 ribosomal protein L19 CGCCGCCGGC 10 40 4.0 <0.001 182825 human ribosomal protein L35 mRNA GCCGTGTCCG 5 20 4.0 0.002 119213 ribosomal protein S6 AGG A AAGCTG 13 52 4.0 <0.001 76437 ESTs, highly similar to 60S rpL36 (Rattus norvegicus) CCCCAGCCAG 7 27 3.9 <0.001 75459 ribosomal protein S3 GCAGCCATCC 13 48 3.7 <0.001 4437 ribosomal protein L28 GGCAAGAAGA 7 25 111611 3.6 <0.001 ribosomal protein L27 CCCGTCCGGA 19 65 3.4 <0.001 ribosomal protein L13 15 51 179943 ribosomal protein Lll CGCTGGTTCC 3.4 <0.001 TAAGGAGCTG 9 30 3.3 <0.001 77904 ribosomal protein S26 CCTTCGAGAT 8 26 3.3 0.001 76194 ribosomal protein SS GGATTTGGCC 33 103 3.1 <0.001 119500 ribosomal protein, large, P2 AGGCTACGGA 20 63 3.2 <0.001 119122 60S ribosomal protein L13A CTGCTATACG 7 22 3.1 <0.004 180946 ribosomal protein LS TGTGCTAAAT 12 35 <0.001 179779 ribosomal protein L37 GAGGGAGTTT 34 2.9 <0.001 76064 ribosomal protein L27 a AAGAAGATAG 8 22 2.8 0.007 184776 ribosomal protein L23a ACATCATCGA 17 46 2.7 <0.001 182979 ribosomal protein L 12 CTGTTGGTGA 12 31 2.6 0.003 3463 ribosomal protein S23 AAGACAGTGG 26 63 2.4 <0.001 184109 ribosomal protein L37 a TTGGTCCTCT 46 108 2.3 <0.001 108124 ribosomal protein L41 CTCCTCACCT 12 28 2.3 0.008 119122 60S ribosomal protein L13A AATAGGTCCA 22 50 2.3 <0.001 113029 ribosomal protein S25 ACTCCAAAAA 23 46 2.0 0.004 133230 ribosomal protein S 15 CTGGGTTAAT 47 87 <0.001 126701 ribosomal protein S 19 TCAGATCTTT 38 74 1.9 <0.001 75344 ribosomal protein S4, X-linked AGCTCTCCCT 38 56 1.9 0.003 82202 ribosomal protein L17 TAATAAAGGT 45 78 1.7 0.002 118690 ribosomal protein S8 TTCAATAAAA 56 89 1.6 0.004 177592 ribosomal protein, large, Pl Additional tags with P values >0.01 and ,e;Q.05 AAGGTCGAGC 1 8 8.0 0.022 184582 ribosomal protein L24 CTCGAGGAGG 0 6 >6 0.016 3254 ribosomal protein L23-like GCTCCGAGCG 0 5 >5 0.028 80617 ribosomal protein S 16 TACAAGAGGA 5 16 3.2 0.014 174131 ribosomal protein L6 CCATTGCACT 7 17 2.4 0.032 53798 ESTs, highly similar to 60S RP L18A CGCCGGAACA 12 27 2.3 0.012 286 ribosomal protein L4 ATTATTTTTC 8 18 2.3 0.037 153 ribosomal protein L 7 CAATAAATGT 34 53 1.6 0.027 179779 ribosomal protein L37 CCAGAACAGA 40 60 1.5 0.030 111222 ribosomal protein L30 GCATAATAGG 38 55 1.4 0.048 184108 ribosomal protein L21 Data obtained from the comparison of the SAGE libraries of the transfected neuroblastoma cell lines (SHEP-2 versus SHEP-21N). The transcripts are ordered by fold induction. P values were calculated by Monte Carlo simulations according to the SAGE 300 software package (see Materials and methods; Velculescu et al., 1995). Tag frequencies are given for the total SAGE libraries of SHEP-2 and SHEP-21N, -21 000 tags each. 97 >5 K.Boon et al. Table III. Downstream targets induced by N-myc: protein synthesis, protein degradation and ribosome biogenesis Tag sequence SHEP-2 SHEP-21N Fold induction P value Unigene Hs. Gene GCATAGGCTG 0 12 >12 <0.001 198304 Tu translation elongation factor, mitochondrial (tufM) GCCCAGCTGG 1 11 11.0 0.003 223241 translation elongation factor lo (EEFlo) TGTGTTGAGA 12 111 9.3 <0.001 181165 eukaryotic translation elongation factor lal (EEFlal) TGGGCAAAGC 5 7.8 <0.001 2186 eukaryotic translation elongation factor 1 y (EEFl y) CAGTCTAAAA 0 8 >8 0.004 76118 ubiquitin C-terminal esterase LI (ubiquitin thiolesterase) GAGCGGGATG 0 >8 0.004 77060 proteasome subunit 6 (13 type) GGCTCCCACT 16 0.002 74335 90-kDa heat-shock protein (HSP90) GCATTTAAAT 86 138 1.6 <0.001 261802 eukaryotic translation elongation factor 113 (EEFl 13) AGCACCTCCA 22 66 3.0 <0.001 75309 eukaryotic translation elongation factor 2 TGAAATAAAA 26 55 2.1 <0.001 173205 nucleophosmin (B23) Additional tags with P values > 0.01 and ,e,0.05 CTGGCGAGCG 1 9 9.0 0.011 174070 human ubiquitin carrier protein (E2-EPF) GGGGCAGGGC 1 8 8.0 0.022 119140 eukaryotic translation initiation factor SA GGCCCTGAGC 2 11 5.5 0.012 71618 human RNA polymerase II subunit (hsRPBlO) TACCAGTGTA 0 5 0.028 79037 60-kDa heat-shock protein 1 (HSP60) TGGCTAGTGT 2 10 >5 0.019 118065 proteasome subunit, 13 type, 7 TACAAAACCA 1 4 4.0 0.202 79110 nucleolin• GTTTTTGCTT 3 8 2.7 0.110 79110 nucleolin" CAGATCTTTG 3 11 3.7 0.029 119502 proteasome subunit, a type, 7 TCACAAGCAA 4 15 3.8 0.010 146763 aNAC mRNA CGCCGCGGTG 5 16 3.2 0.014 4835 eukaryotic translation initiation factor 3, subunit 8 (e1Fs8) GTGACAGAAG 5 13 2.6 0.047 129673 eukaryotic translation initiation factor 4A, isoform 1 CCATTGCACT 7 17 2.4 0.032 173694 ESTs, highly similar to probable ubiquitin C-terminal hydrolase AACTAAAAAA 71 97 1.4 0.028 3297 ubiquitin Genes are listed by fold induction. •Two reliable tags were found for this gene due to alternative polyadenylation. eIF3s8, EEFlal and tuFM (Figure 1). These data further with a role in protein synthesis are induced in SHEP-21N support a role for N-myc as a regulator of protein synthesis. therefore raises the question of whether their up-regulation Genes involved in routing, folding and degradation of is an indirect and late effect consequent on N-myc-induced proteins were also up-regulated. The nascent polypeptide­ growth or whether these genes are early targets of associated complex a (NAC) mRNA was induced induction by N-myc and the cause of myc-mediated cell (Table III, lower part). NAC protects nascent cytosolic growth. We therefore tested in a time course experiment proteins from translocation to the endoplasmatic reticulum whether the genes of the protein synthesis machinery are (Wiedmann et al., 1994). Induction of the chaperones early or late targets of N-myc-mediated induction. N-myc HSP60 (from Oto 5 tags) and HSP90 (from 3 to 12 tags) expression can be switched off reversibly in SHEP-21N further suggested an increased cellular capacity for cells by tetracycline. SHEP-21N cells were treated for 24 h protein folding and maturation. Additionally, the cellular with tetracycline. Northern blot analysis showed that the capacity for protein degradation was possibly induced. N-myc mRNA expression is switched off within 8 h of Three ubiquitin pathway proteins (ubiquitin, ubiquitin tetracycline treatment (Figure 3A, lanes 1-3). After 24 h, C-terminal esterase LI and ubiquitin carrier protein) and cells were washed and grown for an additional 2-36 h three proteasome subunits (P type 6, p type 7 and a type 7) without tetracycline. N-myc mRNA expression is restored showed increased tag frequencies. Northern blot analysis between 2 and 4 h after tetracycline removal (Figure 3A, confirmed induction of HSP60, proteasome subunit P type 6 lanes 5 and 6). Western blot analysis showed that N-myc and ubiquitin in SHEP-21N (Figure 1). protein expression closely follows N-myc mRNA expres­ sion (Figure 3B). The northern blot filter was hybridized Up-regulation of glycolysis genes with probes for the N-myc downstream targets nucleolin, Another functional group of N-myc-induced genes en­ nucleophosmin and the ribosomal protein genes RPS6 and coded key enzymes in the glycolytic pathway (Table IV, RPS12 (Figure 3A). After repression of N-myc by upper part). Tags for aldolase A fructose-bisphosphate tetracycline, the mRNA levels of these genes were (ALDOA), triosephosphate isomerase 1 (TPil), glycer­ unaffected at 0 and 8 h, but were reduced to low basic aldehyde-3-phosphate dehydrogenase (GAPDH) and levels at 24 h. Within 2-4 h after re-expression of N-myc pyruvate kinase are all increased (Table IV). A series of mRNA and protein, the mRNA expression of all four other metabolic enzymes was also induced. Northern blot genes was strongly re-induced (Figure 3B, lanes 6 and 7). analysis confirmed the mRNA induction of ALDOA, Similar results were obtained for EEFlAl, TPil and pyruvate kinase, TPil and GAPDH (Figure 1). eIF3s8 (data not shown). The expression level of cofilin that we used as a control does not change significantly N-myc activates downstream targets within 4 h during the time course. To exclude a direct effect of In several systems it was observed that c-myc can induce tetracycline on nucleolin or nucleophosmin expression, we cell growth and cell mass. The finding that many genes conducted the same experiment with SHEP-2 cells, but no 5.3 >5 >5 N-myc target genes function in ribosome biogenesis Table IV. Downstream targets induced by N-myc: glycolysis Tag sequence SHEP-2 SHEP-21N Fold induction P value Unigene Hs. Gene GCGACCGTCA 1 14 14.0 <0.001 183760 aldolase A fructose-bisphosphate (ALDOA) TAGCTTCTTC 0 7 >7 0.008 76392 aldehyde dehydrogenase 1, soluble• TCTGCTTGTC 0 5 >5 0.028 aldehyde dehydrogenase 1, soluble" TGGCCCCACC 3 18 6.0 <0.001 198281 pyruvate kinase TGAGGGAATA 4 21 83848 triosephosphate isomerase 1 (TPil) 5.3 <0.001 TACCATCAAT 17 3.5 <0.001 195188 glyceraldehyde 3-phosphate dehydrogenase (GAPDH) Additional tags with P value >0.01 and ,;;;0.05 TGACTGAAGC 0 5 0.028 3343 3-phosphoglycerate dehydrogenase mRNA CGGCTGAATT 0 5 0.028 75888 ESTs, highly similar to 6-P-gluconate dehydrogenase, decarboxylating ACCTTGTGCC 0 5 >5 0.028 878 sorbitol dehydrogenase Transcripts are listed by fold induction. "Two reliable tags were found for this gene. SHEP-21N at 0 and 24 h of tetracycline treatment and at 7 .5 h after tetracycline removal showed no change in cell volume, formally excluding the possibility that induction of the genes of the protein synthesis machinery is a result of increased cell volume ( data not shown). These results Nucleolin show that the genes of the protein synthesis machinery are early targets in the N-myc downstream pathway, although Nucleophosmin not necessarily direct targets of N-myc. Induction of these genes by N-myc is highly versatile: expression drops after N-myc abrogation and is restored swiftly after N-myc re­ EEF1a1 expression. N-myc elF3s8 Effect at the protein level and induction of rRNA synthesis tuFM To analyze further the effect of N-myc on the protein APSl2 synthesis machinery, we analyzed the protein levels of Proteasome nucleolin, nucleophosmin, EEFly and EEFl� by western subunit� 6 blotting. All four proteins are more strongly expressed in RPS27 Ubiquitin SHEP-21N cells compared with SHEP-2 (Figure 4, lanes 1 and 2). As a further test that these expression levels are RPLB controlled by N-myc, we treated the SHEP-21N cells for HSP60 1-8 days with tetracycline, which suppressed N-myc RPSi9' expression (Figure 4, lanes 3-8). After 2-3 days, the high Protein sy nthes is protein expression levels of the four N-myc-induced genes Fau-S30 dropped to the basic expression level observed in SHEP-2 cells. Regulation of the mRNA level of these genes by ALDOA N-myc is therefore effective at the protein level. Finally, PPARPO when tetracycline was washed away after 8 days and cells Pyruvate were cultured for 1 or 7 days without tetracycline, N-myc kinase RPS6 expression was restored and expression levels of the four target proteins were increased (Figure 4, lanes 9 and 10). TPl1 The induction by N-myc of nucleolin and nucleo­ Rib osomal proteins phosrnin, two genes with a key role in rRNA processing GAPDH and ribosome biogenesis, urged us to analyze whether SHEP-21N cells have a higher rRNA content than SHEP-2 Glycolysls cells. Total RNA was isolated from 10 samples of 10 cells of SHEP-2 and SHEP-21N. Spectrophotometric analysis Fi . 1. Northern blot analysis of N-myc downstream target genes. showed that SHEP-21N cells on average have a 45% Equal amounts of total RNA from exponentially growing SHEP-2 and higher yield than SHEP-2 cells (P <0.001, Student's t-test SHEP-21N cells were loaded. Northern blots were hybridized with probes for the 19 indicated N-myc targets. RNA quantification was for independent samples; Figure 4B). Triplicate experi­ checked by ethidium bromide staining; the 28S band is shown. ments on independently cultured cells gave the same results. Densitometric quantification of the 18S and 28S effect on gene expression was observed (data not shown). rRNA bands fractionated on agarose gels confirmed that Fluorescence activated cell sorting (FACS) analysis of this increase is caused by rRNA (data not shown). 59 K.Boon et al. Fold Fold Induction • Increase (SHE P-21 N/SHE P-2) (N1 59/N52) ! � .. . ·· ,: ·-. . .... .. , ... . .... 1 ' A B 10 100 1 10 100 Tag frequencies SHEP-2 Tag frequencies N52 Fig. 2. Level of induction of the 56 ribosomal protein genes identified as N-myc targets (P <0.01) in SHEP-21N cells. (A) Fold induction by N-myc in SHEP-21N cells as a function of the basic expression levels in SHEP-2. x-axis, basic expression level in SHEP-2 cells normalized per 10 000 tags; y-axis, fold induction in SHEP-21N cells. (B) Increase in the same 56 ribosomal protein genes in the N-myc-amplified neuroblastoma Nl59 as a function of the basic expression level in N-myc single-copy neuroblastoma N52. x-axis, expression level in N52 normalized per 10 000 tags; y-axis, fold increase in Nl59 relative to N52. We also measured protein content and the rate of protein and EEFly (from 18.4 to 31 tags). There is almost no synthesis. Lysates of 10 induction of the genes involved in glycolysis. The SHEP-2 and SHEP-21N cells contained equivalent amounts of total protein (data not expression levels of nucleolin, nucleophosmin and ribo­ shown). Protein synthesis rates were analyzed by somal protein S6 were confirmed by hybridization of [ S] methionine incorporation, but no differences were northern blots with total RNA from Nl59 and N52 observed between SHEP-2 and SHEP-21N cells. Also, (Figure 5B). These results show that the expression levels when N-myc expression was switched off by a 48 h of many of the N-myc target genes identified in SHEP-21N tetracycline treatment, no differences in incorporation cells also correlate in vivo with N-myc amplification and could be observed (data not shown). N-myc therefore overexpression. strongly induces the rRNA content of SHEP-21N cells, but not the protein synthesis rate. The protein synthesis in N-myc target gene expression in neuroblastoma SHEP-21N cells may be limited by a factor not induced by cell lines and tumors N-myc, or may have been maximal already in the SHEP-2 We further analyzed the expression of N-myc downstream cells and beyond a level that can be boosted by N-myc. genes in a panel of neuroblastoma cell lines and tumors. Hybridization of a northern blot of total RNA from 21 neuroblastoma cell lines showed a fair correlation between SAG E libraries of neuroblastomas with and expression of N-myc, nucleolin, nucleophosmin and the without amplifica tion of endogenous N-myc ribosomal protein PPARP0 (Figure 5A). Cell line SJNB12 To analyze whether genes of the protein synthesis shows high expression of the N-myc target genes machinery are also induced in neuroblastomas with (Figure 5A, lane 7). This cell line has no N-myc N-myc amplification, we generated SAGE libraries of expression, but has c-myc amplification and overexpres­ two neuroblastomas. Neuroblastoma N159 has N-myc sion (Cheng et al., 1995), suggesting that c-myc may amplification and expression, and neuroblastoma N52 is induce the same target genes as N-myc (see below). an N-myc single-copy tumor without N-myc expression As cell lines are not fully representative of neuroblas­ (Figure 5B, lanes 9 and 10). We sequenced 39 598 tags of toma tumors in vivo, we analyzed 16 fresh neuroblastomas the two libraries (Table I). The tag frequencies were of all stages. A northern blot analysis showed a rather good normalized per 20 000 tags and compared. N-myc was overall correlation between expression of N-myc, nucleo­ represented by 16 tags in N159 and 0 tags in N52. There lin and nucleophosmin (Figure 5B). There are some are 52 tags differentially expressed (P <0.01) in the exceptions, but the results suggest that nucleolin and libraries. We analyzed which of the N-myc target genes nucleophosmin are also in vivo targets of N-myc induc­ identified in the SHEP cells correlated with N-myc in the tion. Ribosomal protein S6 (RPS6) expression showed a two tumors. The 56 significantly (P <0.01) induced less consistent relationship with N-myc, indicating that ribosomal protein genes detected in SHEP-21N cells besides N-myc, other factors may also modulate its produce a total of 988 tags in N52 and 1600 tags in N159 expression. (per 20 000 tags). The N-myc-amplified N159 tumor therefore has 62% higher ribosomal protein gene expres­ sion (Figure 2B). This strongly suggests that N-myc Several N-myc target genes are also induced by induces ribosomal protein gene expression in vivo. Other c-myc genes functioning in protein synthesis are also up­ N-myc belongs to the same family of proto-oncogenes as regulated. Increased expression in N159 compared with c-myc. Since N-myc can replace c-myc in transgenic mice N52 is seen for nucleophosmin (from 4 to 19 tags), without inducing gross phenotypic defects (Malynn et al., nucleolin (from 3 to 9 tags), eukaryotic translation 2000), and since both myc proteins share the same target initiation factor 4A, isoform 1 (from 4 to 8 tags) and the recognition sequence, we analyzed whether the N-myc translation elongation factors EEFlal (from 50 to 98 tags) downstream targets identified in this study are also 1388 N-myc target genes function in ribosome biogenesis A SHE P-21N A Tc hours afte r Tc remova l days ri. cl .il'J' S Wllh Telracyciln wilho Tc i----, o a 2.4 1 2 4 6 a 10 12 24 36 0 2 4 3 6 8 1 7 N-myc N-myc .. - Nucleophosm in Nuc le olin Nucle olin -- - Nu c l eo phosm i n -- - EEF1)' RPS6 EE F1 J3 RPS12 control 2 3 4 5 6 7 B e 10 cofl lln ,7 28 S ,6 tota l RNA • al B Tc hours afte r Tc re mo v ,4 0 8 24 1 2 4 6 8 10 12 24 36 ,3 ,2 1 2 3 4 5 6 7 I 9 10 11 12 Fig. 3. Time course analysis of N-myc and downstream target gene SHEP-2 SH EP-2 1 N induction in SHEP-21N cells. SHEP-21N cells were treated for 24 h with 10 ng/ml tetracycline, washed and grown for an additional Fig. 4. Expression of N-myc, nucleolin, nucleophosmin, translation 36 h without tetracycline. Cells were harvested at 0, 8 and 24 h of elongation factors EEF ly and EE FlP and total RNA content of tetracycline treatment. Subsequent samples were taken at 1, 2, 4, 8, SHEP-2 and SHEP-21N cells. (A) Western blot analysis. Total cell 10, 12, 24 and 36 h after removal of the antibiotic. (A) Northern blot extracts (10 µg) were fractionated through an acrylamide gel, blotted analysis of total RNA at the indicated time points. (B) Western blot and probed with monoclonal antibodies against N-myc and nucleo­ analysis of N-myc protein at the indicated time points. A 10 µg phosmin, and with polyclonal antibodies against nucleolin, EEFl y and aliquot of total protein samples of the time course experiment were EE Flp. SHEP-21N cells were treated for 0-8 days with tetracycline fractionated through a 10% SDS- polyacrylamide gel, blotted on an (lanes 2-8) and subsequently cultured for 1 or 7 days without Immobilon membrane and probed with a monoclonal anti-N-myc tetracycline (lanes 9 and 10). A Coomassie Blue staining is shown as antibody. control for loading. (B) Total RNA content of SHEP-2 and SHEP-21N cells. RNA was isolated from 10 samples of 10 cells of each cell line and analyzed spectrophotomerically. Error bars give the SD. induced by c-myc. We analyzed the melanoma cell line IGR39D and a c-myc-transfected clone of this cell line (clone 3; Versteeg et al., 1988). Northern blots with total Discussion RNA of these cell lines were hybridized with the 19 probes tested on the SHEP-2 and SHEP-21N cells. Eight of the One of the surprising aspects of myc oncogenes is their N-myc targets appeared to be induced by c-myc as well multitude of phenotypic effects. They are known to induce (Figure 6). They are the ribosomal protein genes S12, S27, growth, cell division, metastasis and apoptosis. A series of S 19 and S6, and nucleolin, nucleophosmin, ubiquitin and target genes of myc transcription factors has been GAPDH. The remaining 11 genes showed no induction by identified, some of which can be related to specific c-myc. Therefore, c-myc and N-myc share >40% of their phenotypes. However, our knowledge of myc target genes target genes in the cell systems tested here. Interestingly, is probably still fragmentary and insufficient to explain the nucleophosmin, nucleolin and most ribosomal protein full range of phenotypes. As a step towards a complete genes are among them. inventory of the myc downstream pathway, we applied the 1389 K.Boon et al. Neu roblastoma cel l li nes tl .- my c :ai ngle copy 111 - my camplltled 0. c;, ,( , ID I=' z a, "' N N .. :;; "' m 0:, z "' "' z � m u z z z r!t ;: "' .; ::; :z: ,,;, ... ::; -, "' "' I� '! "'i :3 ,,, "' z :3 :::, "' II) ':'J "' ()) z "' ... N-myc Nucleolli l"I N uc leopho ,1m ln PPARPO c-myc 28S B Neu roblastoma tu mors N-my c am plll jed 111-myc sing le, copy 0, c:, ... "' 0) 0, 0, ,... 0 :s 8 1, ., N ... i z z i z z z j i z 12 I I z z - - - -� - ' ' . ' ' • - ' N-myc .&� j� jj, Nucl eolli n Nuc leophosm ln RPS6 B-11ctln 1 Z 3 4 5 6 7 8 9 10 11 1,7 1J 14 15 16 Fig. 5. Northern blot analysis of total RNA from neuroblastoma cell lines and tumors. Filters were hybridized with the indicated probes. RNA quantification was checked by ethidium bromide staining; the 28S band is shown. (A) Panel of 21 neuroblastoma cell lines. (B) Panel of 16 fresh tumors. Tumors in lanes 1-9 are N-myc amplified. SAGE technology to N-myc-transfected cells. We have the analysis of the full myc-induced transcription shift. chosen to compare SHEP-21N with SHEP-2 cells, as we N-myc turns out to have a massive effect on genes with a aimed to identify all genes that are up-regulated in a role in protein synthesis. Approximately 80% of ribosomal situation of stable and enduring N-myc expression, rather protein genes turned out to be enhanced, as well as some than in a transition period after induction of N-myc. SAGE key genes in rRNA maturation and ribosome assembly. provides an integral gene expression profile of a tissue or Furthermore, expression of many translation initiation and cell line. Comparison of the SAGE libraries of the N-myc­ elongation factors is considerably enhanced. We detected expressing SHEP-21N and control SHEP-2 cells identified induction of 89 genes involved in protein synthesis 114 genes significantly (P <0.01) induced in SHEP-21N. (Tables II and III). Together, they produced 1119 of the Moreover, since SAGE is quantitative, the libraries permit 20 950 sequenced tags in SHEP-2. These 89 genes 1390 N-myc target genes function in ribosome biogenesis show an equal overall induction by, for example, a factor of two or three, but induction appears to be related to the basic level of expression in SHEP-2. The strongest induction is observed for genes with the weakest basal expression level (Figure 2). This could suggest that the RP S6 ribosomal protein genes in SHEP-2 cells are restricted in their expression to a variable extent and that N-myc can relieve this restriction. c - myc Nucleolin Induction of genes of the protein synthesis machinery is likely to be a general effect of N-myc in neuroblastomas. Nucleophos m i n Comparison of SAGE libraries of neuroblastoma cell lines RPS12 B23 with and without N-myc amplification shows a 62% increase of ribosomal protein gene transcripts in N-myc­ expressing cells. Moreover, northern blot analysis of 37 RPS27 neuroblastomas and neuroblastoma cell lines showed an overall induced expression level of nucleolin, nucleophos­ min and ribosomal protein genes in N-myc-amplified cases GA PDH RP S19 (Figure 5). In addition, c-myc was found to induce a series of N-myc target genes as well. Of the 19 targets of N-myc Fig. 6. Northern blot analysis of induction of N-myc target genes in a that we tested on northern blots, eight were induced in c-myc-transfected melanoma cell line. Clone 3 is a c-myc-transfected clone of the IGR39D melanoma cell line. Equal amounts of total RNA melanoma cells with ectopic expression of c-myc. of IGR39D and clone 3 were loaded. Filters were hybridized with the Amongst them are the ribosomal protein genes Sl2, S27, indicated probes. S19 and S6, and nucleolin and nucleophosmin. These data suggest that induction of the protein synthesis machinery is a major function of both c-myc and N-myc. therefore contributed 5.3% of the total number of transcripts in SHEP-2. In SHEP-21N, these genes give Myc genes are general inducers of the protein rise to 3327 transcript tags, or 15.9% of all transcripts. synt hesis machinery Two of the up-regulated genes, nucleolin and nucleophos­ Our data are well in line with some of the recently min, function in rRNA maturation and ribosome assembly identified target genes of c-myc and with phenotypic (reviewed in Ginisty et al., 1999). We therefore analyzed effects of myc observed in viva and in vitro. Early analyses whether N-myc expression results in higher rRNA levels. identified two translation initiation factors as targets of We found a striking 45% higher rRNA content in c-myc (Rosenwald et al., 1993; Jones et al., 1996), while SHEP-21N than in SHEP-2 cells on a per cell basis. recent microarray analyses revealed induction by c-myc of Somewhat surprisingly, there was no overall increase in two more translation initiation factors, nucleolin and one the rate of protein synthesis in SHEP-21N cells. One ribosomal protein gene (RPSll) (Coller et al., 2000). interpretation is that some rate-limiting components of the Nucleolin was also identified as a c-myc target by protein synthesis machinery are not induced in SHEP-21N Greasley et al. (2000). The microarray analyses of Coller cells. Alternatively, protein synthesis may already have et al. (2000) do not reveal induction of 32 other ribosomal been maximal in SHEP-2, beyond a level that can be protein genes that were represented on their chips. boosted further. However, it was observed recently that 4 days after Crucial to the interpretation of these data is the finding in viva transduction of a c-myc-expressing retrovirus in that genes of the protein synthesis machinery are early mouse, liver cells expressing ectopic c-myc are greatly targets of the N-myc pathway. Activation of N-myc results enlarged and have increased expression of six ribosomal within 2-4 h in full induction of expression of the tested protein genes as well as nucleolin and nucleophosmin target genes, amongst which were nucleolin, nucleophos­ (Kim et al., 2000). Together with our SAGE analyses in min, two ribosomal protein genes, a translation initiation which the qualitative and quantitative induction of genes factor and a translation elongation factor. This leads us to of the protein synthesis machinery were established, these the conclusion that the massive induction of genes of the data implicate the protein synthesis machinery as a major protein synthesis pathway is an early effect of N-myc. target of induction by myc proteins. These findings are in However, we have not addressed the question of whether agreement with phenotypic effects of myc genes observed these genes are direct targets of N-myc or, in contrast, part in several experiments (Schmidt, 1999). Rat fibroblasts of a hierarchical pathway with myc at the top. Although with inactivated c-myc alleles showed a slower growth rate nucleolin was identified previously as a direct target of and a reduced protein synthesis rate (Mateyak et al., 1997), c-myc (Greasley et al., 2000), the time course experiments while induction of protein synthesis in fibroblasts was do not exclude the possibility that other genes of the observed after c-myc activation (Schmidt, 1999). B cells protein synthesis machinery are induced by an intermedi­ with ectopic c-myc expression in transgenic mice are ary transcription factor that is up-regulated by N-myc. The larger at any stage of differentiation and have an increased quantitative character of SAGE enabled a further analysis protein synthesis rate (Iritani and Eisenman, 1999). of the induction. All N-myc downstream targets have a Drosophila with a mutated dmyc grow more slowly and fair basal expression level in SHEP-2 cells, which is only attain a tiny body volume (Johnston et al., 1999). The not surprising in view of their essential role in protein data in animal model systems, in normal fibroblasts and in synthesis. Interestingly, the ribosomal protein genes do not neuroblastoma tumor cells all suggest that induction of the 139 1 K.Boon et al. Anti-N-myc was obtained from Pharmlngen (Clone B8. 4.B). Rabbit anti­ protein synthesis machinery is a major function of myc human EEF ly and anti-human EE Fl� antibodies (Sanders et al., 1996) genes. In addition to their direct effect on the cell cycle by were a gift of Dr J.Dijk (Sylvius Laboratories, LUMC, Leiden, The inducing cyclin D2 and Id2, this induction of the protein Netherland s). synthesis machinery may provide the increase in cell mass Total rRNA content required to keep the cell volume in step with proliferation. Total RNA of 1 X 10 exponentially growing cells was extracted by guanidium isothiocyanate (Chomczynski and Sacchi, 1987) and quanti­ fied spectrophotometr ically. Results of 10 isolations of each of the cell lines SHEP-2 and SHEP-21N were statistically analyzed with the Materials and methods Studen t's t-test for independent samples. Aliquots on a per cell basis Cell lines were subj ected to agarose gel electrophoresis and stained with ethidium Neuroblastoma cell lines and culture conditions were as described before bromide. The relative fluorescence of the rRNA bands was quantified (Cheng et al., 1995). The melanoma cell lines IGR39D and clone 3 were using the Kodak Digital Science lD Image Analysis Software package described earlier (Versteeg et al., 1988). The SHEP cell lines were (EDAS 120). maintained in RPMI 1640 medium supplemented with 10% feta! calf serum, 4 mM L-glutamine, 100 U/ml penicillin and 100 µg/ml strepto­FA CS analysis mycin (Lutz et al., 1996). Tetracycline (Sigma) was used at a SHEP-21N cells treated or not treated with tetracycline were trypsinized, concentration of 10 ng/ml to inhibit N-myc expression. stained with propidium iodine and analyzed on a Beckman FACScan flow cytometer. Forward scatter (FSC) was used as a means for cell mass. FSC Generation of SAG E libraries was measured for the total cell population or for the G /G fraction, and 0 1 SAGE was performed as described (Velculescu et al., 1995) with minor did not differ for SHEP-21N cells treated for O or 24 h with tetracycline adaptations. Total RNA was extracted by guanidium thiocyanate and for cells subsequently cultured for 7 .5 h without tetracycline. (Chomczynski and Sacchi, 1987). Poly(A) RNA was isolated using the Mess ageMaker kit (Gibco-BRL) according to the manufacturer 's instructions. SAGE libraries were generated using minimally 4 µg of Ackn owl edgem ents RNA. The cDNA was synthesized according to the Superscript poly(A) We thank Dr Pui K.Chan and Dr Phillipe Bouvet for their kind gifts of Choice System (Gibco-BRL), digested with NlaIII and bound to anti-nucleophosmin and anti-nucleolin antibodies, respectively, and Dr streptavidin-coated magnetic beads (Dynal). Linkers containing recogni­ Jan Dijk for antibodies to translation elongation factors. We thank Adam tion sites for BsmFI were ligated to the cDNA. Linker tags including a Benham and Ineke Braakman for their kind help in the protein synthesis cDNA tag were released by BsmFI digestion, ligated to one another and experiments and for their hospitality, and Alvin Chan, Jan Molenaar and amplified. The PCR products were heated for 5 min at 65 C before Danielle Veenma for help in some of the experimen ts. This research preparative analysis on a polyacrylamide gel. After the ligation into was supported by grants from the Stichting Kindergeneeskundig concatamers, a second heating step was included (15 min at 65 C) and Kankeronderzoek (SKK), the Dutch Cancer Foundation (NKB/KWF) fragments between 800 and 1500 bp were purified and cloned in pZer o-1 and the A.Meelmeijer Fund. (lnvitrogen). Colonies were screened with PCR using Ml3 forward and reverse primers. Inserts >300 bp were sequenced with a BigDye terminator kit and analyzed on a 377 ABI automated sequencer (Perkin Elmer). Referen ces Alex,R, Sozer i,O., Meyer, S. and Dildrop,R (1 992) Determination of the Analysis of the SAGE database DNA sequence recognized by the bHLH-zip domain of the N-Myc The SAGE libraries were analyzed using the SAGE 300 program software protein. Nucleic Acids Res. , 20, 2257-2 263 . package (Velculescu et al., 199 7). P values were calculated using Monte Bello-F ernandez ,C., Packham,G. and Cleveland ,J.L. (1 993) The Carlo simulations. Transcripts were identified by comparison of the tags ornithine decarboxylase gene is a transcriptional target of c-Myc. in the database with the 'tag to gene map' (SAGEmap) from the Cancer Proc. Natl Acad. Sci. USA, 90, 7804- 7808. Genome Anatomy Proj ect at the NCBI (http:// www.ncbi.nlm.nih.gov/ Bernards,R, Dessain,S.K. and Weinberg ,RA. (19 86) N-myc SAGE). This database links Unigene clusters to SAGE tags (Lal et al., amplification causes down-modulation of MHC class I antigen 1999). The gene assignments were subsequently checked by hand for expression in neuroblastoma. Cell, 47, 667-674. sequencing errors causing incorrect tags and for erroneous gene Blackwo od,E.M ., Kretzner,L. and Eisenman,RN. (1 992) Myc and Max assignments based on hybrid Unigene clusters. Other database analyses function as a nucleoprotein complex. Curr. Opin. Genet. Dev., 2, and generation of specific primers utilized the Wisconsin GCG package 227- 235. software. Borer.RA., Lehner, C.F., Eppenberge r,H.M. and Nigg,E.A. (19 89) Major nucleolar proteins shuttle between nucleus and cytoplasm. Cell, 56, Northern blot analysis 379-390. Total RNA (20 µg per lane) was electrophoresed through a 0.8% agarose Bouchard,C. et al. (1 999) Direct induction of cyclin D2 by Myc gel in the presence of 6.7% formaldehyde and blotted on Hybond N contributes to cell cycle progression and sequestration of p27. membranes (Amersham) in !OX SSC. Hybridization was carried out EMBO J., 18, 5321 -5333. overnight in 0.5 M NaHPO pH 7.0 , 7% SDS, 1 mM EDTA at 65 C. Bo yd,K.E. and Farnham,P.J. (1 997) Myc versus USF: discrimination at Filters were washed in 40 mM NaHPO , 1 % SDS at 65 C. Probes were the cad gene is determined by core promoter elements. Mol. Cell. labeled by random priming of sequence-verified PCR products. A Biol., 17, 2529-2537. complete list of all the primers used in RT-PCRs is available on request. Caron,H. et al. (200 1) The human transcriptome map: clustering of highly expressed genes in chromosomal domains. Science, 291, Total protein content 1289-1292. Exponentially growing cells were harvested and cell number was determined using a Coulter counter. Cells (1 X 10) were lysed in Chen g,N.C., Van Roy,N., Chan,A., Beitsma,M., Westerve ld, A., Speleman,F. and Versteeg,R (19 95) Deletion mapping in 20 mM Tris-HCI pH 8.0 , 137 mM NaCl, 10% glycerol, 1 % NP-40 and protease inhibitors (protease cocktail, Roche). Samples were assayed with neuroblastoma cell lines suggests two distinct tumor suppressor genes in the lp35-36 region, only one of which is associated with the Bio-Rad Protein assay. Assays were performed at least in duplicate. N-myc amplification. Oncogene, 10, 291-297. Western blots Chomczynski,P. and Sacchi,N. (1 987) Single-step method of RNA Cell lysates were separated on an SDS-polyacrylarnide gel and isolation by acid gnanidinium thiocyanate--phenol-chloroform electroblotted onto Inunobilon-P transfer membrane (Millipore). extraction. Anal. Biochem., 162, 156 -159. Blocking of the membrane and incubation with antibodies involved Cole,M.D. (19 86) The myc oncogene: its role in transformation and standard procedures. Proteins were visualized using the ECL detection differenti ation. Annu. Rev. Genet., 20, 36 1-3 84. system (Amersham). Anti-nucleophosmin monoclonal antibody was a Colle r,H.A., Grandori ,C., Tamayo ,P., Colber t,T., Lander ,E.S ., gift of Dr P.K.Chan (Baylor College of Medicine). The antibody against Eisenman.RN. and Golub ,T.R (2000) Expression analysis with nucleolin was a gift of Dr P.B ouvet (CNRS, IPBS, Toulouse, France). oligonucleotide microarrays reveals that MYC regulates genes 1392 N-myc target genes function in ribosome biogenesis involved in growth, cell cycle, signaling and adhesion. Proc. Natl Seeger, R.C., Brodeur, G.M., Sather,H., Dalton,A., Siegel, S.E., Acad. Sci. USA, 97, 3260-3265. Wong,K.Y. and Hammond,D. (1985) Association of multiple copies Dang ,C.V. (1999) c-Myc target genes involved in cell growth, apoptosis of the N-myc oncogene with rapid progression of neuroblastomas. and metabolism. Mol. Cell. Biol. , 19, 1-11. N. Engl. J. Med. , 313, 1111-1116. Eiler s,M., Schirm,S. and Bishop ,J.M. (1991) The MYC protein activates Steiner,P., Philipp .A., Lukas,J., Godden-Kent, D., Pagano, M., transcription of the a-prothymosin gene. EMB O J. , 10, 133-141. Mittnacht,S., Bartek,J. and Eilers,M. (1995) Identification of a Fulda,S., Lutz,W., Schwab,M. and Debatin, K.M. (1999) MycN Myc-dependent step during the formation of active G cyclin-cdk sensitizes neuroblastorna cells for drug-induced apoptosis. complexes. EMBO J. , 14, 4814-4826. Oncogene, 18, 1479-1486. Velculesc u,V.E., Zhang,L., Vogelstein,B. and Kinzler,K.W. (1995) Ginisty, H., Amalric,F. and Bouvet,P. (1998) Nucleolin functions in the Serial analysis of gene expression. Science, 270, 484-487. first step of ribosomal RNA processing. EMBO J. , 17, 1476-1486. Velculesc u,V.E., Zhang,L., Zhou,W., Vogelstein,J., Basrai, M.A., Ginisty, H., Sicard,H., Roger,B. and Bouvet,P. (1999) Structure and Bassett, D.E., Jr, Hieter,P., Vogelstein,B. and Kinzler,K.W. (1997) functions of nucleolin. J. Cell Sci. , 11 2, 761-772. Characterization of the yeast transcriptome. Cell, 88, 243-251. Grandori,C., Mac,J., Siebelt,F., Ayer,D.E. and Eisenman, R.N. (1996) Versteeg,R., Noordermeer, I.A., Kruse-Wolters, M., Ruiter,D.J. and Myc-Max heterodimers activate a DEAD box gene and interact with Schrier,P.I. (1988) c-myc down-regulates class I HLA expression in multiple E box-related sites in vivo. EMBO J. , 15, 4344-4357. human melanomas. EMBO J. , 7, 1023-1029. Greasley, P.J., Bonnard,C. and Amati,B. (2000) Myc induces the Weiss,W.A., Aldape,K., Mohapatr a,G., Feuerstein, B.G. and Bishop,J.M. nucleolin and BN5 l genes: possible implications in ribosome (1997) Targeted expression of MYCN causes neuroblastoma in biogenesis. Nucleic Acids Res. , 28, 446-453. transgenic mice. EMB O J. , 16, 2985-2995. Henriksson,M. and Luscher,B. (1996) Proteins of the Myc network: Wiedmann,B., Sakai,H., Davis ,T.A. and Wiedmann,M. (1994) A protein essential regulators of cell growth and differ entiation. Adv. Cancer complex required for signal-sequence-specific sorting and trans­ Res. , 68, 109-182. location. Na ture, 370, 434-440. Iritan i,B.M. and Eisenman, R.N. (1999) c-Myc enhances protein Wool,I., Chan,Y.L. and Gliick,A. (1996) Mammalian ribosomes: the synthesis and cell size during B lymphocyte development. Proc. structure and the evolution of proteins. In Hershey,J., Mathews,M. and Sonenberg,N. (eds), Tra nslational Control. Cold Spring Harbor Natl Acad. Sci. USA, 96, 13180--13185. Johnston, L.A., Prober, D.A., Edga r,B.A., Eisenman,R.N. and Gallant,P. Laboratory Press, Cold Spring Harbor, NY, pp. 685-732. (1999) Drosophila myc regulates cellular growth during development. Zhang,L., Zhou,W., Velculesc u,V.E., Kem, S.E., Hruban ,R.H., Cell, 98, 779-790. Hamilton, S.R., Vogelstein,B. and Kinzler, K.W. (1997) Gene Jones ,R.M., Branda,J., Johnston, K.A., Polymenis,M., Gadd,M., expression profiles in normal and cancer cells. Science, 276, Rustgi,A., Callanan,L. and Schmidt, E.V. (1996) An essential E box 1268-1272. in the promoter of the gene encoding the mRNA cap-binding protein (eukaryotic initiation factor 4E) is a target for activation by c-myc. Received May 16, 2000; revised January 9, 2001 ; Mol. Cell. Biol. , 16, 4754-4764. accepted January 18 , 2001 Kim,S., Li,Q., Dang,C.V. and Lee,L.A. (2000) Induction of ribosomal genes and hepatocyte hypertrophy by adenovirus-mediated expression of c-Myc in vivo. Proc. Natl Acad. Sci. USA, 97, 11198-11202. Lal,A. et al. (1999) A public database for gene expression in human cancers. Cancer Res. , 59, 5403-5407. Lasorell a,A., Noseda, M., Beyna,M. and Iavarone.A. (2000) Id2 is a retinoblastoma protein target and mediates signalling by Myc oncoproteins. Na ture, 407, 592-598. Lutz,W., Stohr,M., Schurmann,J., Wenzel,A., Lohr,A. and Schwab,M. (1996) Conditional expression of N-myc in human neuroblastoma cells increases expression of a-prothymosin and omithine decarboxylase and accelerates progression into S-phase early after mitogenic stimulation of quiescent cells. Oncogene, 13, 803-812. Malynn ,B.A., de Alboran,I .M., O' Hagan ,R.C., Bronson,R., Davidson,L., DePinho ,R.A. and Alt,F.W. (2000) N-myc can functionally replace c-myc in murine development, cellular growth and differentiation. Genes Dev. , 14, 1390--1399. Mateyak, M.K., Obaya ,A.J., Adachi,S. and Sedivy,J .M. (1997) Phenotypes of c-Myc-deficient rat fibroblasts isolated by targeted homologous recombination. Cell Growth Differ. , 8, 1039-1048. Obay a,A.J., Mateyak, M.K. and Sedivy, J.M. (1999) Mysterious liaisons: the relationship between c-Myc and the cell cycle. Oncogene, 18, 2934-2941. Olson, M.O. (1991) The role of protein in nucleolar structure and function. In Strauss,R. and Wilson, S.H. (eds), The Eukaryotic Nucleus. Telford Press, Caldwell, NJ, pp. 541-546. Perez-Roger,!., Kim,S.H., Grif fiths,B., Sewing.A. and Land,H. (1999) Cyclins Dl and D2 mediate myc-induced proliferation via sequestration of p27(Kipl) and p21(Cipl). EMB O J. , 18, 5310-5320. Rosenwald, I.B., Rhoads ,D.B., Callanan, L.D., Isselbache r,K.J. and Schmidt, E.V. (1993) Increased expression of eukaryotic translation initiation factors eIF-4E and eIF-2a in response to growth induction by c-myc. Proc. Natl Acad. Sci. USA, 90, 6175-6178. Sanders,J., Brandsma ,M., Janssen, G.M., Dijk,J. and Moller,W. (1996) Immunofluorescence srudies of human fibroblasts demonstrate the presence of the complex of elongation factor-l�yo in the endoplasmic reticulum. J. Cell Sci., 109, 1113-1117. Schmidt, E.V. (1999) The role of c-myc in cellular growth control. Oncogene, 18, 2988-2996. Schwab,M., Alitalo,K., Klempnauer, K.H., Varmus, H.E., Bishop,J .M., Gilbert,F., Brodeur, G., Goldstein,M. and Trent,J. (1983) Amplified DNA with limited homology to myc cellular oncogene is shared by human neuroblastoma cell lines and a neuroblastoma tumour. Nature, 305, 245-248. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The EMBO Journal Springer Journals

N‐myc enhances the expression of a large set of genes functioning in ribosome biogenesis and protein synthesis

Download PDF
11 pages

Loading next page...
 
/lp/springer-journals/n-myc-enhances-the-expression-of-a-large-set-of-genes-functioning-in-ObwCrJS122

References (50)

Publisher
Springer Journals
Copyright
Copyright © European Molecular Biology Organization 2001
ISSN
0261-4189
eISSN
1460-2075
DOI
10.1093/emboj/20.6.1383
Publisher site
See Article on Publisher Site

Abstract

The EMBO Journal Vol. 20 No. 6 pp. 1383-1393, 2001 N-myc enhances the expression of a large set of genes functioning in ribosome biogenesis and protein synthesis 1 2 1 3 follow a very aggressive course (Schwab et al., 1983; Kathy Boon • , Huib N.Caron , , 1 1 Seeger et al., 1985). Overexpression of transfected N-myc Ronald van Asperen , Linda Valentijn , 1 1 genes in neuroblastoma cell lines strongly increased Marie-Christine Hermus , Peter van Sluis , 1 4 3 proliferation rates (Bemards et al., 1986; Lutz et al., llja Roobeek , Isabel Weis , P.A.Vo0te , 3 1 5 1996). Transgenic mice overexpressing N-myc in neural Manfred Schwab and Rogier Versteeg , crest-derived tissues showed frequent development of Department of Human Genetics, Academic Medical Center, neuroblastomas (Weiss et al., 1997). Numerous compar­ University of Amsterdam, PO Box 22700, ll00 DE Amsterdam, able observations have implicated c-myc and L-myc in the Department of Paediatric Oncology and Haematology, Emma Kinder pathogenesis of many other tumor types (Cole, 1986; Ziekenhuis, Academic Medical Center, Amsterdam, The Netherlands Henriksson and Luscher, 1996). While c-myc and N-myc and Division of Cytogenetics, Deutsches Krebsforschungszentrum, Heidelberg, Germany homozygous knockout mice are embryonic lethal, trans­ genic mice in which N-myc replaced c-myc showed a gross Present address: Department of Pathology, Duke University Medical Center, Durham, NC 27710, USA normal development, indicating that both proteins have largely overlapping functions (Malynn et al., 2000). Corresponding author Many experiments have suggested a role for myc genes e-mail: [email protected] in cell cycle control, metastasis, blocking of differenti­ ation, apoptosis and proliferation rate (Henriksson and The myc oncogenes are frequently activated in human Luscher, 1996; Dang, 1999; Schmidt, 1999). The myc tumors, but there is no comprehensive insight into the family members are transcription factors with a basic/ target genes and downstream cellular pathways of helix-loop-helix/leucine zipper (bHLHzip) domain. They these transcription factors. We applied serial analysis form heterodimers with Max proteins and bind to the of gene expression (SAGE) to identify targets of E-box motif CACGTG to activate target gene transcription N-myc in neuroblastomas. Analysis of 42 000 mRNA (Alex et al., 1992; Blackwood et al., 1992). The limited transcript tags in SAGE libraries of N-myc­ number of initially identified targets of c-myc gave little transfected and control neuroblastoma cells revealed insight into the mechanism of how c-myc induces the 114 up-regulated genes. The majority of these genes variety of phenotypes. Examples of myc target genes have a role in ribosome assembly and activity. are prothymosin a, omithine decarboxylase, CAD and Northern blot analysis confirmed up-regulation of all the DEAD-box gene MrDb (Eilers et al., 1991; tested transcripts. Induction was complete within 4 h Bello-Fernandez et al., 1993; Rosenwald et al., 1993; after N-myc expression. The large majority of the Grandori et al., 1996; Jones et al., 1996; Boyd and ribosomal proteins were induced, as well as genes con­ Farnham, 1997). Induction of cyclins Dl, E and A and trolling rRNA maturation. Cellular rRNA content was cdc25A were found in some but not all model systems 45% induced. SAGE libraries and northern blot (reviewed in Obaya et al., 1999). Induction of cyclin D2 analysis confirmed up-regulation of many of these was consistently found and provides a direct link to the cell genes in N-myc-amplified neuroblastomas. As N-myc cycle (Bouchard et al., 1999; Perez-Roger et al., 1999; can functionally replace c-myc, we analyzed whether Coller et al., 2000). Furthermore, c-myc and N-myc also N-myc targets were induced by c-myc as well. induce Id2 transcription, thus stimulating Rb inactivation Approximately 40% of these N-myc targets were up­ regulated in a c-myc-transfected melanoma cell line. and cell cycle progression (Lasorella et al., 2000). These data suggest that myc genes function as major However, in addition to these direct effects on prolifer­ regulators of the protein synthesis machinery. ation, several findings suggest that c-myc also promotes Keywords: c-myc/N-myc/neuroblastoma/protein the growth of cells, thereby inducing an increase in cell synthesis/ribosomes mass that may be a prerequisite for rapid cell proliferation. Impaired in viva expression of Drosophila dmyc results in adult flies half the normal size (Johnston et al., 1999). Both the volume and proliferation rate of their cells are reduced, Introduction and could be restored by overexpression of dmyc. B cells The members of the myc oncogene family play a of transgenic mice with overexpression of c-myc show for prominent role in cancer. N-myc, c-myc and L-myc are all differentiation stages an increased cell size and rearranged, amplified, mutated and/or overexpressed in increased protein synthesis rate (Iritani and Eisenman, many human tumor types. The c-myc gene is expressed in 1999). Furthermore, fibroblasts with a homozygous a wide variety of tissues and tumors, while N-myc inactivation of c-myc have a reduced proliferation rate as expression is largely restricted to embryonic tissues and well as a reduced protein synthesis rate (Mateyak et al., neuroendocrine tumors. Approximately 20% of neuro­ 1997). Conversely, activation of c-myc in fibroblasts blastomas have N-myc amplification, and these tumors activates protein synthesis (Schmidt, 1999). Furthermore, © European Molecular Biology Organization 1383 K.Boon et al. a role for c-myc in protein synthesis is suggested by the Table I. Summary of neuroblastoma SAGE libraries finding that c-myc induces the translation initiation factors Total tags Different transcripts elF-4E, eIF-2-a (Rosenwald et al., 1993; Jones et al., 1996), eIF5A and eIF4G (Coller et al., 2000). Very Neuroblastoma cell lines recently, ectopic in viva expression of c-myc was found to SHEP-2 20 950 10 154 result in increased expression of six ribosomal protein SHEP-21N (N-myc transfected) 20 938 8566 genes in mouse liver (Kim et al., 2000). A role for myc Neuroblastoma tumors N52 19 597 9356 genes in growth regulation is in line with their effect on the N159 (N-myc amplified) 20 001 10 262 cell cycle. Inactivation of c-myc in fibroblasts prolonged Total 81 486 the G and G phases of the cell cycle, but not the S phase, while high expression of myc genes accelerated transition through G (Steiner et al., 1995; Lutz et al., 1996). A similar effect was found for Drosophila dmyc (Johnston functions in ribosome biogenesis and activity and in later et al., 1999). steps of the protein synthesis and protein turnover Here we describe the use of SAGE (serial analysis of machinery. gene expression; Velculescu et al., 1995) to identify the downstream genes that are activated by N-myc in human neuroblastoma. To date, only prothymosin a, ornithine Induction of genes involved in ribosome decarboxylase and Id2 have been identified as targets of biogenesis and protein synthesis N-myc (Lutz et al., 1996; Lasorella et al., 2000). The The first functional group consists of 56 ribosomal protein analysis of the expression level of >40 000 transcripts genes, which were induced up to 37-fold (P <0.01; identified 114 genes up-regulated in N-myc-expressing Table II). Tags for 10 more ribosomal protein genes were cells. Our results indicate that N-myc functions as a less strongly up-regulated (0.01> P >0.05), with induction regulator of cell growth by stimulating genes functioning levels from 1.4- to 8-fold (Table II, lower part). These 66 in ribosome biogenesis and protein synthesis. Several of tags correspond to 82% of the human ribosomal proteins the identified genes are induced by N-myc as well as by (Wool and Chan, 1996). Northern blot analysis of SHEP-2 c-myc. and SHEP-21N with probes for seven ribosomal protein genes (Sl2, S27, Fau-S30, L8, S6, S19 and PPARP0) confirmed their up-regulation (Figure 1). The fraction of Results tags for ribosomal protein mRNAs was increased from SAGE libraries of N-myc-transfected 4.1 % in SHEP-2 to 12.6% in SHEP-21N. The level of neuroblastoma cell lines induction of individual ribosomal protein genes is a To identify the downstream target genes of N-myc, we function of their basal expression levels in SHEP-2. applied the SAGE technique to an N-myc-transfected Highly expressed genes are less induced than genes with a neuroblastoma cell line. The SHEP cell line has no N-myc low basic expression in SHEP-2 (Figure 2). amplification and expression, or c-myc expression. A A second functional group of genes up-regulated in tetracycline-dependent N-myc expression vector has been SHEP-21N consists of two nucleolar protein genes introduced into these cells, resulting in the SHEP-21N (Table III). Nucleophosmin (or B23) is induced from 26 clone (Lutz et al., 1996). The SHEP-21N cells have to 55 tags (P <0.001), which was confirmed by northern constitutive exogenous N-myc expression that can be blot analysis (Figure 1). Nucleophosmin is an abundant switched off by tetracycline. N-myc expression in the nucleolar protein that processes rRNA by cleavage of the SHEP-21N cells was shown to increase the rate of cell 5' end of the 5.8S pre-rRNA (Zhang et al., 1997) and division, shorten the G phase of the cell cycle and render functions in assembly and nuclear-cytoplasmic shuttling the cells more susceptible to apoptotic triggers (Lutz et al., of pre-ribosomal particles (Borer et al., 1989; Olson, 1996; Fulda et al., 1999). SAGE libraries were constructed 1991). Transcription of another nucleolar protein is also induced: nucleolin, which has two tags due to alternative from SHEP-21N cells expressing N-myc and from SHEP-2 control cells. The SHEP-2 clone was transfected with the transcripts, is induced from 4 to 12 tags in total empty expression vector. From each library, we sequenced (P = 0.035). This induction was confirmed by northern ~21 000 transcript tags, corresponding to 8566 blot analysis (Figure 1). Nucleolin processes pre-rRNA to (SHEP-21N) and 10 154 (SHEP-2) different transcripts mature 18S rRNA (Ginisty et al., 1998, 1999). It binds to (Table I). A tag for the transfected N-myc construct has a nucleophosmin and is also involved in the assembly of frequency of 0 and 8 in SHEP-2 and SHEP-21N, pre-ribosomal particles and their nucleo-cytoplasmic respectively. Comparison of the SAGE libraries yielded transport. The induction of nucleolin and nucleophosmin 114 significantly (P <0.01) up-regulated tags in N-myc­in SHEP-21N suggests that rRNA and ribosome biosynth­ expressing cells, with induction levels of up to 37-fold esis are targets of N-myc stimulation. (Tables II-IV). Another 70 tags were significantly down­ Additionally, tags corresponding to nine translation regulated. Here we focused on the analysis of a series of initiation and elongation factors were induced (Table III). tags induced in the N-myc-transfected cells. The tran­ They are eukaryotic translation initiation factors eIF3s8, scripts corresponding to these tags were identified using eIF4A and eIF5A, and the subunits a, 13, y and 8 of the SAGEmap database (Lal et al., 1999) and our own tag translation elongation factor 1 (EEFl). Furthermore, assignment program (Caron et al., 2001) and checked by elongation factor 2 and the mitochondrial elongation mRNA and expressed sequence tag (EST) sequence factor Tu (tuFM) are up-regulated. Northern blot analysis analyses. A comprehensive set of up-regulated genes of SHEP-21N and SHEP-2 confirmed the induction of 1384 1.9 2.9 4.5 N-myc target genes function in ribosome biogenesis Table II. Downstream targets induced by N-myc: ribosomal proteins Tag sequence SHEP-2 SHEP-21N Fold induction P value Unigene Hs. Gene GCCGAGGAAG 1 37.0 <0.001 82148 ribosomal protein S12 GCTTTTAAGG 1 29 29.0 <0.001 8102 ribosomal protein S20 CCCATCCGAA 1 23 23.0 <0.001 91379 ribosomal protein L26 GGCCGCGTTC 0 23 >23 <0.001 5174 ribosomal protein S 17 CCAGTGGCCC 0 21 >21 <0.001 180920 ribosomal protein S9 GTGTTGCACA 1 16 16.0 165590 ribosomal protein S 13 <0.001 GATGCTGCCA 1 16 16.0 <0.001 99914 ribosomal protein L22 CCGTCCAAGG 2 31 15.5 <0.001 80617 ribosomal protein S 16 GGAGTGGACA 1 14 14.0 <0.001 75458 ribosomal protein L18 GCCTGTATGA 2 27 13.5 <0.001 180450 ribosomal protein S24 GTTCCCTGGC 2 26 13.0 <0.001 177415 ribosomal protein Fau-S30 ATGGCTGGTA 6 72 12.0 <0.001 182426 ribosomal protein S2 GTGTTAACCA 1 11 11.0 <0.001 74267 ribosomal protein L10 CACAAACGGT 4 10.8 <0.001 195453 ribosomal protein S27 CTCAACATCT 3 32 10.7 <0.001 73742 ribosomal protein, large, PO GTTCGTGCCA 2 18 9.0 <0.001 179666 ribosomal protein L35a GACGACACGA 4 30 7.5 <0.001 153177 ribosomal protein S28 TCGTCTTTAT 3 21 7.0 <0.001 75538 ribosomal protein S7 GGACCACTGA 5 34 6.8 <0.001 119598 ribosomal protein L3 CCTCGGA A AA 4 24 6.0 <0.001 2017 ribosomal protein L38 AATCCTGTGG 8 48 6.0 <0.001 178551 ribosomal protein L8 ATCAAGGGTG 4 21 5.3 <0.001 157850 ribosomal protein L9 GGGCTGGGGT 20 101 5.1 <0.001 183698 ribosomal protein L29 AAGGAGATGG 5 25 5.0 <0.001 184014 ribosomal protein L3 l/tag matches mitochondrial sequences AAGGTGGAGG 11 55 5.0 <0.001 163593 ribosomal protein L18a TTACCATATC 10 49 4.9 <0.001 177461 ribosomal protein L39 GTGAAGGCAG 6 27 4.5 <0.001 77039 ribosomal protein S3A GAACACATCC 4 18 0.002 75879 ribosomal protein L19 CGCCGCCGGC 10 40 4.0 <0.001 182825 human ribosomal protein L35 mRNA GCCGTGTCCG 5 20 4.0 0.002 119213 ribosomal protein S6 AGG A AAGCTG 13 52 4.0 <0.001 76437 ESTs, highly similar to 60S rpL36 (Rattus norvegicus) CCCCAGCCAG 7 27 3.9 <0.001 75459 ribosomal protein S3 GCAGCCATCC 13 48 3.7 <0.001 4437 ribosomal protein L28 GGCAAGAAGA 7 25 111611 3.6 <0.001 ribosomal protein L27 CCCGTCCGGA 19 65 3.4 <0.001 ribosomal protein L13 15 51 179943 ribosomal protein Lll CGCTGGTTCC 3.4 <0.001 TAAGGAGCTG 9 30 3.3 <0.001 77904 ribosomal protein S26 CCTTCGAGAT 8 26 3.3 0.001 76194 ribosomal protein SS GGATTTGGCC 33 103 3.1 <0.001 119500 ribosomal protein, large, P2 AGGCTACGGA 20 63 3.2 <0.001 119122 60S ribosomal protein L13A CTGCTATACG 7 22 3.1 <0.004 180946 ribosomal protein LS TGTGCTAAAT 12 35 <0.001 179779 ribosomal protein L37 GAGGGAGTTT 34 2.9 <0.001 76064 ribosomal protein L27 a AAGAAGATAG 8 22 2.8 0.007 184776 ribosomal protein L23a ACATCATCGA 17 46 2.7 <0.001 182979 ribosomal protein L 12 CTGTTGGTGA 12 31 2.6 0.003 3463 ribosomal protein S23 AAGACAGTGG 26 63 2.4 <0.001 184109 ribosomal protein L37 a TTGGTCCTCT 46 108 2.3 <0.001 108124 ribosomal protein L41 CTCCTCACCT 12 28 2.3 0.008 119122 60S ribosomal protein L13A AATAGGTCCA 22 50 2.3 <0.001 113029 ribosomal protein S25 ACTCCAAAAA 23 46 2.0 0.004 133230 ribosomal protein S 15 CTGGGTTAAT 47 87 <0.001 126701 ribosomal protein S 19 TCAGATCTTT 38 74 1.9 <0.001 75344 ribosomal protein S4, X-linked AGCTCTCCCT 38 56 1.9 0.003 82202 ribosomal protein L17 TAATAAAGGT 45 78 1.7 0.002 118690 ribosomal protein S8 TTCAATAAAA 56 89 1.6 0.004 177592 ribosomal protein, large, Pl Additional tags with P values >0.01 and ,e;Q.05 AAGGTCGAGC 1 8 8.0 0.022 184582 ribosomal protein L24 CTCGAGGAGG 0 6 >6 0.016 3254 ribosomal protein L23-like GCTCCGAGCG 0 5 >5 0.028 80617 ribosomal protein S 16 TACAAGAGGA 5 16 3.2 0.014 174131 ribosomal protein L6 CCATTGCACT 7 17 2.4 0.032 53798 ESTs, highly similar to 60S RP L18A CGCCGGAACA 12 27 2.3 0.012 286 ribosomal protein L4 ATTATTTTTC 8 18 2.3 0.037 153 ribosomal protein L 7 CAATAAATGT 34 53 1.6 0.027 179779 ribosomal protein L37 CCAGAACAGA 40 60 1.5 0.030 111222 ribosomal protein L30 GCATAATAGG 38 55 1.4 0.048 184108 ribosomal protein L21 Data obtained from the comparison of the SAGE libraries of the transfected neuroblastoma cell lines (SHEP-2 versus SHEP-21N). The transcripts are ordered by fold induction. P values were calculated by Monte Carlo simulations according to the SAGE 300 software package (see Materials and methods; Velculescu et al., 1995). Tag frequencies are given for the total SAGE libraries of SHEP-2 and SHEP-21N, -21 000 tags each. 97 >5 K.Boon et al. Table III. Downstream targets induced by N-myc: protein synthesis, protein degradation and ribosome biogenesis Tag sequence SHEP-2 SHEP-21N Fold induction P value Unigene Hs. Gene GCATAGGCTG 0 12 >12 <0.001 198304 Tu translation elongation factor, mitochondrial (tufM) GCCCAGCTGG 1 11 11.0 0.003 223241 translation elongation factor lo (EEFlo) TGTGTTGAGA 12 111 9.3 <0.001 181165 eukaryotic translation elongation factor lal (EEFlal) TGGGCAAAGC 5 7.8 <0.001 2186 eukaryotic translation elongation factor 1 y (EEFl y) CAGTCTAAAA 0 8 >8 0.004 76118 ubiquitin C-terminal esterase LI (ubiquitin thiolesterase) GAGCGGGATG 0 >8 0.004 77060 proteasome subunit 6 (13 type) GGCTCCCACT 16 0.002 74335 90-kDa heat-shock protein (HSP90) GCATTTAAAT 86 138 1.6 <0.001 261802 eukaryotic translation elongation factor 113 (EEFl 13) AGCACCTCCA 22 66 3.0 <0.001 75309 eukaryotic translation elongation factor 2 TGAAATAAAA 26 55 2.1 <0.001 173205 nucleophosmin (B23) Additional tags with P values > 0.01 and ,e,0.05 CTGGCGAGCG 1 9 9.0 0.011 174070 human ubiquitin carrier protein (E2-EPF) GGGGCAGGGC 1 8 8.0 0.022 119140 eukaryotic translation initiation factor SA GGCCCTGAGC 2 11 5.5 0.012 71618 human RNA polymerase II subunit (hsRPBlO) TACCAGTGTA 0 5 0.028 79037 60-kDa heat-shock protein 1 (HSP60) TGGCTAGTGT 2 10 >5 0.019 118065 proteasome subunit, 13 type, 7 TACAAAACCA 1 4 4.0 0.202 79110 nucleolin• GTTTTTGCTT 3 8 2.7 0.110 79110 nucleolin" CAGATCTTTG 3 11 3.7 0.029 119502 proteasome subunit, a type, 7 TCACAAGCAA 4 15 3.8 0.010 146763 aNAC mRNA CGCCGCGGTG 5 16 3.2 0.014 4835 eukaryotic translation initiation factor 3, subunit 8 (e1Fs8) GTGACAGAAG 5 13 2.6 0.047 129673 eukaryotic translation initiation factor 4A, isoform 1 CCATTGCACT 7 17 2.4 0.032 173694 ESTs, highly similar to probable ubiquitin C-terminal hydrolase AACTAAAAAA 71 97 1.4 0.028 3297 ubiquitin Genes are listed by fold induction. •Two reliable tags were found for this gene due to alternative polyadenylation. eIF3s8, EEFlal and tuFM (Figure 1). These data further with a role in protein synthesis are induced in SHEP-21N support a role for N-myc as a regulator of protein synthesis. therefore raises the question of whether their up-regulation Genes involved in routing, folding and degradation of is an indirect and late effect consequent on N-myc-induced proteins were also up-regulated. The nascent polypeptide­ growth or whether these genes are early targets of associated complex a (NAC) mRNA was induced induction by N-myc and the cause of myc-mediated cell (Table III, lower part). NAC protects nascent cytosolic growth. We therefore tested in a time course experiment proteins from translocation to the endoplasmatic reticulum whether the genes of the protein synthesis machinery are (Wiedmann et al., 1994). Induction of the chaperones early or late targets of N-myc-mediated induction. N-myc HSP60 (from Oto 5 tags) and HSP90 (from 3 to 12 tags) expression can be switched off reversibly in SHEP-21N further suggested an increased cellular capacity for cells by tetracycline. SHEP-21N cells were treated for 24 h protein folding and maturation. Additionally, the cellular with tetracycline. Northern blot analysis showed that the capacity for protein degradation was possibly induced. N-myc mRNA expression is switched off within 8 h of Three ubiquitin pathway proteins (ubiquitin, ubiquitin tetracycline treatment (Figure 3A, lanes 1-3). After 24 h, C-terminal esterase LI and ubiquitin carrier protein) and cells were washed and grown for an additional 2-36 h three proteasome subunits (P type 6, p type 7 and a type 7) without tetracycline. N-myc mRNA expression is restored showed increased tag frequencies. Northern blot analysis between 2 and 4 h after tetracycline removal (Figure 3A, confirmed induction of HSP60, proteasome subunit P type 6 lanes 5 and 6). Western blot analysis showed that N-myc and ubiquitin in SHEP-21N (Figure 1). protein expression closely follows N-myc mRNA expres­ sion (Figure 3B). The northern blot filter was hybridized Up-regulation of glycolysis genes with probes for the N-myc downstream targets nucleolin, Another functional group of N-myc-induced genes en­ nucleophosmin and the ribosomal protein genes RPS6 and coded key enzymes in the glycolytic pathway (Table IV, RPS12 (Figure 3A). After repression of N-myc by upper part). Tags for aldolase A fructose-bisphosphate tetracycline, the mRNA levels of these genes were (ALDOA), triosephosphate isomerase 1 (TPil), glycer­ unaffected at 0 and 8 h, but were reduced to low basic aldehyde-3-phosphate dehydrogenase (GAPDH) and levels at 24 h. Within 2-4 h after re-expression of N-myc pyruvate kinase are all increased (Table IV). A series of mRNA and protein, the mRNA expression of all four other metabolic enzymes was also induced. Northern blot genes was strongly re-induced (Figure 3B, lanes 6 and 7). analysis confirmed the mRNA induction of ALDOA, Similar results were obtained for EEFlAl, TPil and pyruvate kinase, TPil and GAPDH (Figure 1). eIF3s8 (data not shown). The expression level of cofilin that we used as a control does not change significantly N-myc activates downstream targets within 4 h during the time course. To exclude a direct effect of In several systems it was observed that c-myc can induce tetracycline on nucleolin or nucleophosmin expression, we cell growth and cell mass. The finding that many genes conducted the same experiment with SHEP-2 cells, but no 5.3 >5 >5 N-myc target genes function in ribosome biogenesis Table IV. Downstream targets induced by N-myc: glycolysis Tag sequence SHEP-2 SHEP-21N Fold induction P value Unigene Hs. Gene GCGACCGTCA 1 14 14.0 <0.001 183760 aldolase A fructose-bisphosphate (ALDOA) TAGCTTCTTC 0 7 >7 0.008 76392 aldehyde dehydrogenase 1, soluble• TCTGCTTGTC 0 5 >5 0.028 aldehyde dehydrogenase 1, soluble" TGGCCCCACC 3 18 6.0 <0.001 198281 pyruvate kinase TGAGGGAATA 4 21 83848 triosephosphate isomerase 1 (TPil) 5.3 <0.001 TACCATCAAT 17 3.5 <0.001 195188 glyceraldehyde 3-phosphate dehydrogenase (GAPDH) Additional tags with P value >0.01 and ,;;;0.05 TGACTGAAGC 0 5 0.028 3343 3-phosphoglycerate dehydrogenase mRNA CGGCTGAATT 0 5 0.028 75888 ESTs, highly similar to 6-P-gluconate dehydrogenase, decarboxylating ACCTTGTGCC 0 5 >5 0.028 878 sorbitol dehydrogenase Transcripts are listed by fold induction. "Two reliable tags were found for this gene. SHEP-21N at 0 and 24 h of tetracycline treatment and at 7 .5 h after tetracycline removal showed no change in cell volume, formally excluding the possibility that induction of the genes of the protein synthesis machinery is a result of increased cell volume ( data not shown). These results Nucleolin show that the genes of the protein synthesis machinery are early targets in the N-myc downstream pathway, although Nucleophosmin not necessarily direct targets of N-myc. Induction of these genes by N-myc is highly versatile: expression drops after N-myc abrogation and is restored swiftly after N-myc re­ EEF1a1 expression. N-myc elF3s8 Effect at the protein level and induction of rRNA synthesis tuFM To analyze further the effect of N-myc on the protein APSl2 synthesis machinery, we analyzed the protein levels of Proteasome nucleolin, nucleophosmin, EEFly and EEFl� by western subunit� 6 blotting. All four proteins are more strongly expressed in RPS27 Ubiquitin SHEP-21N cells compared with SHEP-2 (Figure 4, lanes 1 and 2). As a further test that these expression levels are RPLB controlled by N-myc, we treated the SHEP-21N cells for HSP60 1-8 days with tetracycline, which suppressed N-myc RPSi9' expression (Figure 4, lanes 3-8). After 2-3 days, the high Protein sy nthes is protein expression levels of the four N-myc-induced genes Fau-S30 dropped to the basic expression level observed in SHEP-2 cells. Regulation of the mRNA level of these genes by ALDOA N-myc is therefore effective at the protein level. Finally, PPARPO when tetracycline was washed away after 8 days and cells Pyruvate were cultured for 1 or 7 days without tetracycline, N-myc kinase RPS6 expression was restored and expression levels of the four target proteins were increased (Figure 4, lanes 9 and 10). TPl1 The induction by N-myc of nucleolin and nucleo­ Rib osomal proteins phosrnin, two genes with a key role in rRNA processing GAPDH and ribosome biogenesis, urged us to analyze whether SHEP-21N cells have a higher rRNA content than SHEP-2 Glycolysls cells. Total RNA was isolated from 10 samples of 10 cells of SHEP-2 and SHEP-21N. Spectrophotometric analysis Fi . 1. Northern blot analysis of N-myc downstream target genes. showed that SHEP-21N cells on average have a 45% Equal amounts of total RNA from exponentially growing SHEP-2 and higher yield than SHEP-2 cells (P <0.001, Student's t-test SHEP-21N cells were loaded. Northern blots were hybridized with probes for the 19 indicated N-myc targets. RNA quantification was for independent samples; Figure 4B). Triplicate experi­ checked by ethidium bromide staining; the 28S band is shown. ments on independently cultured cells gave the same results. Densitometric quantification of the 18S and 28S effect on gene expression was observed (data not shown). rRNA bands fractionated on agarose gels confirmed that Fluorescence activated cell sorting (FACS) analysis of this increase is caused by rRNA (data not shown). 59 K.Boon et al. Fold Fold Induction • Increase (SHE P-21 N/SHE P-2) (N1 59/N52) ! � .. . ·· ,: ·-. . .... .. , ... . .... 1 ' A B 10 100 1 10 100 Tag frequencies SHEP-2 Tag frequencies N52 Fig. 2. Level of induction of the 56 ribosomal protein genes identified as N-myc targets (P <0.01) in SHEP-21N cells. (A) Fold induction by N-myc in SHEP-21N cells as a function of the basic expression levels in SHEP-2. x-axis, basic expression level in SHEP-2 cells normalized per 10 000 tags; y-axis, fold induction in SHEP-21N cells. (B) Increase in the same 56 ribosomal protein genes in the N-myc-amplified neuroblastoma Nl59 as a function of the basic expression level in N-myc single-copy neuroblastoma N52. x-axis, expression level in N52 normalized per 10 000 tags; y-axis, fold increase in Nl59 relative to N52. We also measured protein content and the rate of protein and EEFly (from 18.4 to 31 tags). There is almost no synthesis. Lysates of 10 induction of the genes involved in glycolysis. The SHEP-2 and SHEP-21N cells contained equivalent amounts of total protein (data not expression levels of nucleolin, nucleophosmin and ribo­ shown). Protein synthesis rates were analyzed by somal protein S6 were confirmed by hybridization of [ S] methionine incorporation, but no differences were northern blots with total RNA from Nl59 and N52 observed between SHEP-2 and SHEP-21N cells. Also, (Figure 5B). These results show that the expression levels when N-myc expression was switched off by a 48 h of many of the N-myc target genes identified in SHEP-21N tetracycline treatment, no differences in incorporation cells also correlate in vivo with N-myc amplification and could be observed (data not shown). N-myc therefore overexpression. strongly induces the rRNA content of SHEP-21N cells, but not the protein synthesis rate. The protein synthesis in N-myc target gene expression in neuroblastoma SHEP-21N cells may be limited by a factor not induced by cell lines and tumors N-myc, or may have been maximal already in the SHEP-2 We further analyzed the expression of N-myc downstream cells and beyond a level that can be boosted by N-myc. genes in a panel of neuroblastoma cell lines and tumors. Hybridization of a northern blot of total RNA from 21 neuroblastoma cell lines showed a fair correlation between SAG E libraries of neuroblastomas with and expression of N-myc, nucleolin, nucleophosmin and the without amplifica tion of endogenous N-myc ribosomal protein PPARP0 (Figure 5A). Cell line SJNB12 To analyze whether genes of the protein synthesis shows high expression of the N-myc target genes machinery are also induced in neuroblastomas with (Figure 5A, lane 7). This cell line has no N-myc N-myc amplification, we generated SAGE libraries of expression, but has c-myc amplification and overexpres­ two neuroblastomas. Neuroblastoma N159 has N-myc sion (Cheng et al., 1995), suggesting that c-myc may amplification and expression, and neuroblastoma N52 is induce the same target genes as N-myc (see below). an N-myc single-copy tumor without N-myc expression As cell lines are not fully representative of neuroblas­ (Figure 5B, lanes 9 and 10). We sequenced 39 598 tags of toma tumors in vivo, we analyzed 16 fresh neuroblastomas the two libraries (Table I). The tag frequencies were of all stages. A northern blot analysis showed a rather good normalized per 20 000 tags and compared. N-myc was overall correlation between expression of N-myc, nucleo­ represented by 16 tags in N159 and 0 tags in N52. There lin and nucleophosmin (Figure 5B). There are some are 52 tags differentially expressed (P <0.01) in the exceptions, but the results suggest that nucleolin and libraries. We analyzed which of the N-myc target genes nucleophosmin are also in vivo targets of N-myc induc­ identified in the SHEP cells correlated with N-myc in the tion. Ribosomal protein S6 (RPS6) expression showed a two tumors. The 56 significantly (P <0.01) induced less consistent relationship with N-myc, indicating that ribosomal protein genes detected in SHEP-21N cells besides N-myc, other factors may also modulate its produce a total of 988 tags in N52 and 1600 tags in N159 expression. (per 20 000 tags). The N-myc-amplified N159 tumor therefore has 62% higher ribosomal protein gene expres­ sion (Figure 2B). This strongly suggests that N-myc Several N-myc target genes are also induced by induces ribosomal protein gene expression in vivo. Other c-myc genes functioning in protein synthesis are also up­ N-myc belongs to the same family of proto-oncogenes as regulated. Increased expression in N159 compared with c-myc. Since N-myc can replace c-myc in transgenic mice N52 is seen for nucleophosmin (from 4 to 19 tags), without inducing gross phenotypic defects (Malynn et al., nucleolin (from 3 to 9 tags), eukaryotic translation 2000), and since both myc proteins share the same target initiation factor 4A, isoform 1 (from 4 to 8 tags) and the recognition sequence, we analyzed whether the N-myc translation elongation factors EEFlal (from 50 to 98 tags) downstream targets identified in this study are also 1388 N-myc target genes function in ribosome biogenesis A SHE P-21N A Tc hours afte r Tc remova l days ri. cl .il'J' S Wllh Telracyciln wilho Tc i----, o a 2.4 1 2 4 6 a 10 12 24 36 0 2 4 3 6 8 1 7 N-myc N-myc .. - Nucleophosm in Nuc le olin Nucle olin -- - Nu c l eo phosm i n -- - EEF1)' RPS6 EE F1 J3 RPS12 control 2 3 4 5 6 7 B e 10 cofl lln ,7 28 S ,6 tota l RNA • al B Tc hours afte r Tc re mo v ,4 0 8 24 1 2 4 6 8 10 12 24 36 ,3 ,2 1 2 3 4 5 6 7 I 9 10 11 12 Fig. 3. Time course analysis of N-myc and downstream target gene SHEP-2 SH EP-2 1 N induction in SHEP-21N cells. SHEP-21N cells were treated for 24 h with 10 ng/ml tetracycline, washed and grown for an additional Fig. 4. Expression of N-myc, nucleolin, nucleophosmin, translation 36 h without tetracycline. Cells were harvested at 0, 8 and 24 h of elongation factors EEF ly and EE FlP and total RNA content of tetracycline treatment. Subsequent samples were taken at 1, 2, 4, 8, SHEP-2 and SHEP-21N cells. (A) Western blot analysis. Total cell 10, 12, 24 and 36 h after removal of the antibiotic. (A) Northern blot extracts (10 µg) were fractionated through an acrylamide gel, blotted analysis of total RNA at the indicated time points. (B) Western blot and probed with monoclonal antibodies against N-myc and nucleo­ analysis of N-myc protein at the indicated time points. A 10 µg phosmin, and with polyclonal antibodies against nucleolin, EEFl y and aliquot of total protein samples of the time course experiment were EE Flp. SHEP-21N cells were treated for 0-8 days with tetracycline fractionated through a 10% SDS- polyacrylamide gel, blotted on an (lanes 2-8) and subsequently cultured for 1 or 7 days without Immobilon membrane and probed with a monoclonal anti-N-myc tetracycline (lanes 9 and 10). A Coomassie Blue staining is shown as antibody. control for loading. (B) Total RNA content of SHEP-2 and SHEP-21N cells. RNA was isolated from 10 samples of 10 cells of each cell line and analyzed spectrophotomerically. Error bars give the SD. induced by c-myc. We analyzed the melanoma cell line IGR39D and a c-myc-transfected clone of this cell line (clone 3; Versteeg et al., 1988). Northern blots with total Discussion RNA of these cell lines were hybridized with the 19 probes tested on the SHEP-2 and SHEP-21N cells. Eight of the One of the surprising aspects of myc oncogenes is their N-myc targets appeared to be induced by c-myc as well multitude of phenotypic effects. They are known to induce (Figure 6). They are the ribosomal protein genes S12, S27, growth, cell division, metastasis and apoptosis. A series of S 19 and S6, and nucleolin, nucleophosmin, ubiquitin and target genes of myc transcription factors has been GAPDH. The remaining 11 genes showed no induction by identified, some of which can be related to specific c-myc. Therefore, c-myc and N-myc share >40% of their phenotypes. However, our knowledge of myc target genes target genes in the cell systems tested here. Interestingly, is probably still fragmentary and insufficient to explain the nucleophosmin, nucleolin and most ribosomal protein full range of phenotypes. As a step towards a complete genes are among them. inventory of the myc downstream pathway, we applied the 1389 K.Boon et al. Neu roblastoma cel l li nes tl .- my c :ai ngle copy 111 - my camplltled 0. c;, ,( , ID I=' z a, "' N N .. :;; "' m 0:, z "' "' z � m u z z z r!t ;: "' .; ::; :z: ,,;, ... ::; -, "' "' I� '! "'i :3 ,,, "' z :3 :::, "' II) ':'J "' ()) z "' ... N-myc Nucleolli l"I N uc leopho ,1m ln PPARPO c-myc 28S B Neu roblastoma tu mors N-my c am plll jed 111-myc sing le, copy 0, c:, ... "' 0) 0, 0, ,... 0 :s 8 1, ., N ... i z z i z z z j i z 12 I I z z - - - -� - ' ' . ' ' • - ' N-myc .&� j� jj, Nucl eolli n Nuc leophosm ln RPS6 B-11ctln 1 Z 3 4 5 6 7 8 9 10 11 1,7 1J 14 15 16 Fig. 5. Northern blot analysis of total RNA from neuroblastoma cell lines and tumors. Filters were hybridized with the indicated probes. RNA quantification was checked by ethidium bromide staining; the 28S band is shown. (A) Panel of 21 neuroblastoma cell lines. (B) Panel of 16 fresh tumors. Tumors in lanes 1-9 are N-myc amplified. SAGE technology to N-myc-transfected cells. We have the analysis of the full myc-induced transcription shift. chosen to compare SHEP-21N with SHEP-2 cells, as we N-myc turns out to have a massive effect on genes with a aimed to identify all genes that are up-regulated in a role in protein synthesis. Approximately 80% of ribosomal situation of stable and enduring N-myc expression, rather protein genes turned out to be enhanced, as well as some than in a transition period after induction of N-myc. SAGE key genes in rRNA maturation and ribosome assembly. provides an integral gene expression profile of a tissue or Furthermore, expression of many translation initiation and cell line. Comparison of the SAGE libraries of the N-myc­ elongation factors is considerably enhanced. We detected expressing SHEP-21N and control SHEP-2 cells identified induction of 89 genes involved in protein synthesis 114 genes significantly (P <0.01) induced in SHEP-21N. (Tables II and III). Together, they produced 1119 of the Moreover, since SAGE is quantitative, the libraries permit 20 950 sequenced tags in SHEP-2. These 89 genes 1390 N-myc target genes function in ribosome biogenesis show an equal overall induction by, for example, a factor of two or three, but induction appears to be related to the basic level of expression in SHEP-2. The strongest induction is observed for genes with the weakest basal expression level (Figure 2). This could suggest that the RP S6 ribosomal protein genes in SHEP-2 cells are restricted in their expression to a variable extent and that N-myc can relieve this restriction. c - myc Nucleolin Induction of genes of the protein synthesis machinery is likely to be a general effect of N-myc in neuroblastomas. Nucleophos m i n Comparison of SAGE libraries of neuroblastoma cell lines RPS12 B23 with and without N-myc amplification shows a 62% increase of ribosomal protein gene transcripts in N-myc­ expressing cells. Moreover, northern blot analysis of 37 RPS27 neuroblastomas and neuroblastoma cell lines showed an overall induced expression level of nucleolin, nucleophos­ min and ribosomal protein genes in N-myc-amplified cases GA PDH RP S19 (Figure 5). In addition, c-myc was found to induce a series of N-myc target genes as well. Of the 19 targets of N-myc Fig. 6. Northern blot analysis of induction of N-myc target genes in a that we tested on northern blots, eight were induced in c-myc-transfected melanoma cell line. Clone 3 is a c-myc-transfected clone of the IGR39D melanoma cell line. Equal amounts of total RNA melanoma cells with ectopic expression of c-myc. of IGR39D and clone 3 were loaded. Filters were hybridized with the Amongst them are the ribosomal protein genes Sl2, S27, indicated probes. S19 and S6, and nucleolin and nucleophosmin. These data suggest that induction of the protein synthesis machinery is a major function of both c-myc and N-myc. therefore contributed 5.3% of the total number of transcripts in SHEP-2. In SHEP-21N, these genes give Myc genes are general inducers of the protein rise to 3327 transcript tags, or 15.9% of all transcripts. synt hesis machinery Two of the up-regulated genes, nucleolin and nucleophos­ Our data are well in line with some of the recently min, function in rRNA maturation and ribosome assembly identified target genes of c-myc and with phenotypic (reviewed in Ginisty et al., 1999). We therefore analyzed effects of myc observed in viva and in vitro. Early analyses whether N-myc expression results in higher rRNA levels. identified two translation initiation factors as targets of We found a striking 45% higher rRNA content in c-myc (Rosenwald et al., 1993; Jones et al., 1996), while SHEP-21N than in SHEP-2 cells on a per cell basis. recent microarray analyses revealed induction by c-myc of Somewhat surprisingly, there was no overall increase in two more translation initiation factors, nucleolin and one the rate of protein synthesis in SHEP-21N cells. One ribosomal protein gene (RPSll) (Coller et al., 2000). interpretation is that some rate-limiting components of the Nucleolin was also identified as a c-myc target by protein synthesis machinery are not induced in SHEP-21N Greasley et al. (2000). The microarray analyses of Coller cells. Alternatively, protein synthesis may already have et al. (2000) do not reveal induction of 32 other ribosomal been maximal in SHEP-2, beyond a level that can be protein genes that were represented on their chips. boosted further. However, it was observed recently that 4 days after Crucial to the interpretation of these data is the finding in viva transduction of a c-myc-expressing retrovirus in that genes of the protein synthesis machinery are early mouse, liver cells expressing ectopic c-myc are greatly targets of the N-myc pathway. Activation of N-myc results enlarged and have increased expression of six ribosomal within 2-4 h in full induction of expression of the tested protein genes as well as nucleolin and nucleophosmin target genes, amongst which were nucleolin, nucleophos­ (Kim et al., 2000). Together with our SAGE analyses in min, two ribosomal protein genes, a translation initiation which the qualitative and quantitative induction of genes factor and a translation elongation factor. This leads us to of the protein synthesis machinery were established, these the conclusion that the massive induction of genes of the data implicate the protein synthesis machinery as a major protein synthesis pathway is an early effect of N-myc. target of induction by myc proteins. These findings are in However, we have not addressed the question of whether agreement with phenotypic effects of myc genes observed these genes are direct targets of N-myc or, in contrast, part in several experiments (Schmidt, 1999). Rat fibroblasts of a hierarchical pathway with myc at the top. Although with inactivated c-myc alleles showed a slower growth rate nucleolin was identified previously as a direct target of and a reduced protein synthesis rate (Mateyak et al., 1997), c-myc (Greasley et al., 2000), the time course experiments while induction of protein synthesis in fibroblasts was do not exclude the possibility that other genes of the observed after c-myc activation (Schmidt, 1999). B cells protein synthesis machinery are induced by an intermedi­ with ectopic c-myc expression in transgenic mice are ary transcription factor that is up-regulated by N-myc. The larger at any stage of differentiation and have an increased quantitative character of SAGE enabled a further analysis protein synthesis rate (Iritani and Eisenman, 1999). of the induction. All N-myc downstream targets have a Drosophila with a mutated dmyc grow more slowly and fair basal expression level in SHEP-2 cells, which is only attain a tiny body volume (Johnston et al., 1999). The not surprising in view of their essential role in protein data in animal model systems, in normal fibroblasts and in synthesis. Interestingly, the ribosomal protein genes do not neuroblastoma tumor cells all suggest that induction of the 139 1 K.Boon et al. Anti-N-myc was obtained from Pharmlngen (Clone B8. 4.B). Rabbit anti­ protein synthesis machinery is a major function of myc human EEF ly and anti-human EE Fl� antibodies (Sanders et al., 1996) genes. In addition to their direct effect on the cell cycle by were a gift of Dr J.Dijk (Sylvius Laboratories, LUMC, Leiden, The inducing cyclin D2 and Id2, this induction of the protein Netherland s). synthesis machinery may provide the increase in cell mass Total rRNA content required to keep the cell volume in step with proliferation. Total RNA of 1 X 10 exponentially growing cells was extracted by guanidium isothiocyanate (Chomczynski and Sacchi, 1987) and quanti­ fied spectrophotometr ically. Results of 10 isolations of each of the cell lines SHEP-2 and SHEP-21N were statistically analyzed with the Materials and methods Studen t's t-test for independent samples. Aliquots on a per cell basis Cell lines were subj ected to agarose gel electrophoresis and stained with ethidium Neuroblastoma cell lines and culture conditions were as described before bromide. The relative fluorescence of the rRNA bands was quantified (Cheng et al., 1995). The melanoma cell lines IGR39D and clone 3 were using the Kodak Digital Science lD Image Analysis Software package described earlier (Versteeg et al., 1988). The SHEP cell lines were (EDAS 120). maintained in RPMI 1640 medium supplemented with 10% feta! calf serum, 4 mM L-glutamine, 100 U/ml penicillin and 100 µg/ml strepto­FA CS analysis mycin (Lutz et al., 1996). Tetracycline (Sigma) was used at a SHEP-21N cells treated or not treated with tetracycline were trypsinized, concentration of 10 ng/ml to inhibit N-myc expression. stained with propidium iodine and analyzed on a Beckman FACScan flow cytometer. Forward scatter (FSC) was used as a means for cell mass. FSC Generation of SAG E libraries was measured for the total cell population or for the G /G fraction, and 0 1 SAGE was performed as described (Velculescu et al., 1995) with minor did not differ for SHEP-21N cells treated for O or 24 h with tetracycline adaptations. Total RNA was extracted by guanidium thiocyanate and for cells subsequently cultured for 7 .5 h without tetracycline. (Chomczynski and Sacchi, 1987). Poly(A) RNA was isolated using the Mess ageMaker kit (Gibco-BRL) according to the manufacturer 's instructions. SAGE libraries were generated using minimally 4 µg of Ackn owl edgem ents RNA. The cDNA was synthesized according to the Superscript poly(A) We thank Dr Pui K.Chan and Dr Phillipe Bouvet for their kind gifts of Choice System (Gibco-BRL), digested with NlaIII and bound to anti-nucleophosmin and anti-nucleolin antibodies, respectively, and Dr streptavidin-coated magnetic beads (Dynal). Linkers containing recogni­ Jan Dijk for antibodies to translation elongation factors. We thank Adam tion sites for BsmFI were ligated to the cDNA. Linker tags including a Benham and Ineke Braakman for their kind help in the protein synthesis cDNA tag were released by BsmFI digestion, ligated to one another and experiments and for their hospitality, and Alvin Chan, Jan Molenaar and amplified. The PCR products were heated for 5 min at 65 C before Danielle Veenma for help in some of the experimen ts. This research preparative analysis on a polyacrylamide gel. After the ligation into was supported by grants from the Stichting Kindergeneeskundig concatamers, a second heating step was included (15 min at 65 C) and Kankeronderzoek (SKK), the Dutch Cancer Foundation (NKB/KWF) fragments between 800 and 1500 bp were purified and cloned in pZer o-1 and the A.Meelmeijer Fund. (lnvitrogen). Colonies were screened with PCR using Ml3 forward and reverse primers. Inserts >300 bp were sequenced with a BigDye terminator kit and analyzed on a 377 ABI automated sequencer (Perkin Elmer). Referen ces Alex,R, Sozer i,O., Meyer, S. and Dildrop,R (1 992) Determination of the Analysis of the SAGE database DNA sequence recognized by the bHLH-zip domain of the N-Myc The SAGE libraries were analyzed using the SAGE 300 program software protein. Nucleic Acids Res. , 20, 2257-2 263 . package (Velculescu et al., 199 7). P values were calculated using Monte Bello-F ernandez ,C., Packham,G. and Cleveland ,J.L. (1 993) The Carlo simulations. Transcripts were identified by comparison of the tags ornithine decarboxylase gene is a transcriptional target of c-Myc. in the database with the 'tag to gene map' (SAGEmap) from the Cancer Proc. Natl Acad. Sci. USA, 90, 7804- 7808. Genome Anatomy Proj ect at the NCBI (http:// www.ncbi.nlm.nih.gov/ Bernards,R, Dessain,S.K. and Weinberg ,RA. (19 86) N-myc SAGE). This database links Unigene clusters to SAGE tags (Lal et al., amplification causes down-modulation of MHC class I antigen 1999). The gene assignments were subsequently checked by hand for expression in neuroblastoma. Cell, 47, 667-674. sequencing errors causing incorrect tags and for erroneous gene Blackwo od,E.M ., Kretzner,L. and Eisenman,RN. (1 992) Myc and Max assignments based on hybrid Unigene clusters. Other database analyses function as a nucleoprotein complex. Curr. Opin. Genet. Dev., 2, and generation of specific primers utilized the Wisconsin GCG package 227- 235. software. Borer.RA., Lehner, C.F., Eppenberge r,H.M. and Nigg,E.A. (19 89) Major nucleolar proteins shuttle between nucleus and cytoplasm. Cell, 56, Northern blot analysis 379-390. Total RNA (20 µg per lane) was electrophoresed through a 0.8% agarose Bouchard,C. et al. (1 999) Direct induction of cyclin D2 by Myc gel in the presence of 6.7% formaldehyde and blotted on Hybond N contributes to cell cycle progression and sequestration of p27. membranes (Amersham) in !OX SSC. Hybridization was carried out EMBO J., 18, 5321 -5333. overnight in 0.5 M NaHPO pH 7.0 , 7% SDS, 1 mM EDTA at 65 C. Bo yd,K.E. and Farnham,P.J. (1 997) Myc versus USF: discrimination at Filters were washed in 40 mM NaHPO , 1 % SDS at 65 C. Probes were the cad gene is determined by core promoter elements. Mol. Cell. labeled by random priming of sequence-verified PCR products. A Biol., 17, 2529-2537. complete list of all the primers used in RT-PCRs is available on request. Caron,H. et al. (200 1) The human transcriptome map: clustering of highly expressed genes in chromosomal domains. Science, 291, Total protein content 1289-1292. Exponentially growing cells were harvested and cell number was determined using a Coulter counter. Cells (1 X 10) were lysed in Chen g,N.C., Van Roy,N., Chan,A., Beitsma,M., Westerve ld, A., Speleman,F. and Versteeg,R (19 95) Deletion mapping in 20 mM Tris-HCI pH 8.0 , 137 mM NaCl, 10% glycerol, 1 % NP-40 and protease inhibitors (protease cocktail, Roche). Samples were assayed with neuroblastoma cell lines suggests two distinct tumor suppressor genes in the lp35-36 region, only one of which is associated with the Bio-Rad Protein assay. Assays were performed at least in duplicate. N-myc amplification. Oncogene, 10, 291-297. Western blots Chomczynski,P. and Sacchi,N. (1 987) Single-step method of RNA Cell lysates were separated on an SDS-polyacrylarnide gel and isolation by acid gnanidinium thiocyanate--phenol-chloroform electroblotted onto Inunobilon-P transfer membrane (Millipore). extraction. Anal. Biochem., 162, 156 -159. Blocking of the membrane and incubation with antibodies involved Cole,M.D. (19 86) The myc oncogene: its role in transformation and standard procedures. Proteins were visualized using the ECL detection differenti ation. Annu. Rev. Genet., 20, 36 1-3 84. system (Amersham). Anti-nucleophosmin monoclonal antibody was a Colle r,H.A., Grandori ,C., Tamayo ,P., Colber t,T., Lander ,E.S ., gift of Dr P.K.Chan (Baylor College of Medicine). The antibody against Eisenman.RN. and Golub ,T.R (2000) Expression analysis with nucleolin was a gift of Dr P.B ouvet (CNRS, IPBS, Toulouse, France). oligonucleotide microarrays reveals that MYC regulates genes 1392 N-myc target genes function in ribosome biogenesis involved in growth, cell cycle, signaling and adhesion. Proc. Natl Seeger, R.C., Brodeur, G.M., Sather,H., Dalton,A., Siegel, S.E., Acad. Sci. USA, 97, 3260-3265. Wong,K.Y. and Hammond,D. (1985) Association of multiple copies Dang ,C.V. (1999) c-Myc target genes involved in cell growth, apoptosis of the N-myc oncogene with rapid progression of neuroblastomas. and metabolism. Mol. Cell. Biol. , 19, 1-11. N. Engl. J. Med. , 313, 1111-1116. Eiler s,M., Schirm,S. and Bishop ,J.M. (1991) The MYC protein activates Steiner,P., Philipp .A., Lukas,J., Godden-Kent, D., Pagano, M., transcription of the a-prothymosin gene. EMB O J. , 10, 133-141. Mittnacht,S., Bartek,J. and Eilers,M. (1995) Identification of a Fulda,S., Lutz,W., Schwab,M. and Debatin, K.M. (1999) MycN Myc-dependent step during the formation of active G cyclin-cdk sensitizes neuroblastorna cells for drug-induced apoptosis. complexes. EMBO J. , 14, 4814-4826. Oncogene, 18, 1479-1486. Velculesc u,V.E., Zhang,L., Vogelstein,B. and Kinzler,K.W. (1995) Ginisty, H., Amalric,F. and Bouvet,P. (1998) Nucleolin functions in the Serial analysis of gene expression. Science, 270, 484-487. first step of ribosomal RNA processing. EMBO J. , 17, 1476-1486. Velculesc u,V.E., Zhang,L., Zhou,W., Vogelstein,J., Basrai, M.A., Ginisty, H., Sicard,H., Roger,B. and Bouvet,P. (1999) Structure and Bassett, D.E., Jr, Hieter,P., Vogelstein,B. and Kinzler,K.W. (1997) functions of nucleolin. J. Cell Sci. , 11 2, 761-772. Characterization of the yeast transcriptome. Cell, 88, 243-251. Grandori,C., Mac,J., Siebelt,F., Ayer,D.E. and Eisenman, R.N. (1996) Versteeg,R., Noordermeer, I.A., Kruse-Wolters, M., Ruiter,D.J. and Myc-Max heterodimers activate a DEAD box gene and interact with Schrier,P.I. (1988) c-myc down-regulates class I HLA expression in multiple E box-related sites in vivo. EMBO J. , 15, 4344-4357. human melanomas. EMBO J. , 7, 1023-1029. Greasley, P.J., Bonnard,C. and Amati,B. (2000) Myc induces the Weiss,W.A., Aldape,K., Mohapatr a,G., Feuerstein, B.G. and Bishop,J.M. nucleolin and BN5 l genes: possible implications in ribosome (1997) Targeted expression of MYCN causes neuroblastoma in biogenesis. Nucleic Acids Res. , 28, 446-453. transgenic mice. EMB O J. , 16, 2985-2995. Henriksson,M. and Luscher,B. (1996) Proteins of the Myc network: Wiedmann,B., Sakai,H., Davis ,T.A. and Wiedmann,M. (1994) A protein essential regulators of cell growth and differ entiation. Adv. Cancer complex required for signal-sequence-specific sorting and trans­ Res. , 68, 109-182. location. Na ture, 370, 434-440. Iritan i,B.M. and Eisenman, R.N. (1999) c-Myc enhances protein Wool,I., Chan,Y.L. and Gliick,A. (1996) Mammalian ribosomes: the synthesis and cell size during B lymphocyte development. Proc. structure and the evolution of proteins. In Hershey,J., Mathews,M. and Sonenberg,N. (eds), Tra nslational Control. Cold Spring Harbor Natl Acad. Sci. USA, 96, 13180--13185. Johnston, L.A., Prober, D.A., Edga r,B.A., Eisenman,R.N. and Gallant,P. Laboratory Press, Cold Spring Harbor, NY, pp. 685-732. (1999) Drosophila myc regulates cellular growth during development. Zhang,L., Zhou,W., Velculesc u,V.E., Kem, S.E., Hruban ,R.H., Cell, 98, 779-790. Hamilton, S.R., Vogelstein,B. and Kinzler, K.W. (1997) Gene Jones ,R.M., Branda,J., Johnston, K.A., Polymenis,M., Gadd,M., expression profiles in normal and cancer cells. Science, 276, Rustgi,A., Callanan,L. and Schmidt, E.V. (1996) An essential E box 1268-1272. in the promoter of the gene encoding the mRNA cap-binding protein (eukaryotic initiation factor 4E) is a target for activation by c-myc. Received May 16, 2000; revised January 9, 2001 ; Mol. Cell. Biol. , 16, 4754-4764. accepted January 18 , 2001 Kim,S., Li,Q., Dang,C.V. and Lee,L.A. (2000) Induction of ribosomal genes and hepatocyte hypertrophy by adenovirus-mediated expression of c-Myc in vivo. Proc. Natl Acad. Sci. USA, 97, 11198-11202. Lal,A. et al. (1999) A public database for gene expression in human cancers. Cancer Res. , 59, 5403-5407. Lasorell a,A., Noseda, M., Beyna,M. and Iavarone.A. (2000) Id2 is a retinoblastoma protein target and mediates signalling by Myc oncoproteins. Na ture, 407, 592-598. Lutz,W., Stohr,M., Schurmann,J., Wenzel,A., Lohr,A. and Schwab,M. (1996) Conditional expression of N-myc in human neuroblastoma cells increases expression of a-prothymosin and omithine decarboxylase and accelerates progression into S-phase early after mitogenic stimulation of quiescent cells. Oncogene, 13, 803-812. Malynn ,B.A., de Alboran,I .M., O' Hagan ,R.C., Bronson,R., Davidson,L., DePinho ,R.A. and Alt,F.W. (2000) N-myc can functionally replace c-myc in murine development, cellular growth and differentiation. Genes Dev. , 14, 1390--1399. Mateyak, M.K., Obaya ,A.J., Adachi,S. and Sedivy,J .M. (1997) Phenotypes of c-Myc-deficient rat fibroblasts isolated by targeted homologous recombination. Cell Growth Differ. , 8, 1039-1048. Obay a,A.J., Mateyak, M.K. and Sedivy, J.M. (1999) Mysterious liaisons: the relationship between c-Myc and the cell cycle. Oncogene, 18, 2934-2941. Olson, M.O. (1991) The role of protein in nucleolar structure and function. In Strauss,R. and Wilson, S.H. (eds), The Eukaryotic Nucleus. Telford Press, Caldwell, NJ, pp. 541-546. Perez-Roger,!., Kim,S.H., Grif fiths,B., Sewing.A. and Land,H. (1999) Cyclins Dl and D2 mediate myc-induced proliferation via sequestration of p27(Kipl) and p21(Cipl). EMB O J. , 18, 5310-5320. Rosenwald, I.B., Rhoads ,D.B., Callanan, L.D., Isselbache r,K.J. and Schmidt, E.V. (1993) Increased expression of eukaryotic translation initiation factors eIF-4E and eIF-2a in response to growth induction by c-myc. Proc. Natl Acad. Sci. USA, 90, 6175-6178. Sanders,J., Brandsma ,M., Janssen, G.M., Dijk,J. and Moller,W. (1996) Immunofluorescence srudies of human fibroblasts demonstrate the presence of the complex of elongation factor-l�yo in the endoplasmic reticulum. J. Cell Sci., 109, 1113-1117. Schmidt, E.V. (1999) The role of c-myc in cellular growth control. Oncogene, 18, 2988-2996. Schwab,M., Alitalo,K., Klempnauer, K.H., Varmus, H.E., Bishop,J .M., Gilbert,F., Brodeur, G., Goldstein,M. and Trent,J. (1983) Amplified DNA with limited homology to myc cellular oncogene is shared by human neuroblastoma cell lines and a neuroblastoma tumour. Nature, 305, 245-248.

Journal

The EMBO Journal – Springer Journals

Published: Mar 15, 2001

Keywords: c‐myc; N‐myc; neuroblastoma; protein synthesis; ribosomes

There are no references for this article.