HMGA2 Regulates Transcription of the Imp2 Gene Via an Intronic Regulatory Element in Cooperation with Nuclear Factor-KB

HMGA2 Regulates Transcription of the Imp2 Gene via an Intronic Regulatory Element in Cooperation with Nuclear Factor-KB

Isabelle Cleynen,1 Jan R. Brants,1 Kristel Peeters,1 Rob Deckers,1 Maria Debiec-Rychter,2 Raf Sciot,3 Wim J.M. Van de Ven,1 and Marleen M.R. Petit1

1Laboratory of Molecular Oncology, Department of Human Genetics, Flanders Interuniversity Institute for Biotechnology (VIB); 2Department of Human Genetics; and 3Department of Pathology, University Hospital, University of Leuven, Leuven, Belgium

Abstract Introduction IMP2 (insulin-like growth factor-II mRNA binding The Imp2 (insulin-like growth factor-II mRNA binding protein 2) is an oncofetal protein that is aberrantly protein 2) gene encodes a member of the VICKZ (Vg1 RBP/ expressed in several types of cancer. We recently Vera, IMP1, IMP2, IMP3, CRD-BP, KOC, and ZBP-1) family identified the Imp2 gene as a target gene of the of highly related mRNA-binding proteins that are implicated in architectural transcription factor HMGA2 (high mobility posttranscriptional processes such as mRNA localization, turn- group A2) and its tumor-specific truncated form over, and translational control. In this way, they are involved in HMGA2Tr. In this study, we investigated the mechanism the temporal and spatial control of gene expression at the level via which HMGA2 regulates Imp2 gene expression. of mRNA rather than at the level of gene transcription. The We show that HMGA2 and HMGA2Tr directly regulate three mammalian VICKZ proteins (IMP1/CRD-BP, IMP2, and transcription of the Imp2 gene by binding to an IMP3/KOC) are primarily expressed during embryonic devel- AT-rich regulatory region located in the first intron. opment, and their expression is generally not detected in adult In reporter experiments, we show that this AT-rich tissues, with the exception of placenta and testes (1-4). In regulatory region mimics the response of the contrast, they are overexpressed in various cancers (reviewed in endogenous Imp2 gene to HMGA2 and HMGA2Tr. refs. 5, 6), classifying them as oncofetal proteins. Concerning Furthermore, we show that a consensus nuclear IMP2, Lu et al. (7) detected overexpression of a splice variant factor-KB (NF-KB) binding site located immediately of IMP2 (p62) in liver cancer (hepatocellular carcinoma and adjacent to the AT-rich regulatory region binds NF-KB cholangiocarcinoma), and Zhang et al. (8) found autoantibodies and that NF-KB and HMGA2 cooperate to regulate in sera of more than 10% of patients with esophageal, Imp2 gene expression. Finally, we provide evidence hepatocellular, lung, lymphoma, pharyngeal, or uterine cancer. that there is a strong and statistically significant Testing for the presence of such autoantibodies against a panel correlation between HMGA2 and IMP2 gene of autoantigens, including IMP proteins, has been proposed as a expression in human liposarcomas. valuable tool for cancer detection and diagnosis (9). Until now, (Mol Cancer Res 2007;5(4):363–72) the mechanisms by which the oncofetal VICKZ proteins are overexpressed in certain cancers remain elusive. IMP1, IMP2, and IMP3 are each encoded by a separate gene, and previous work from our group clearly indicates that the mRNA level of Received 10/5/06; revised 1/19/07; accepted 2/22/07. Grant support: Geconcerteerde Onderzoeksacties grant 0/10, Cancer Research Imp2, but not of Imp1 and Imp3, is developmentally regulated Program of Fortis Verzekeringen, Fund for Scientific Research (Krediet Aan by the architectural transcription factor HMGA2 (high mobility Navorsers) grant 1.5173.05N, and Belgian Federation against Cancer grant group A2; ref. 10). Like IMP2, HMGA2 is an oncofetal protein: A5890. The costs of publication of this article were defrayed in part by the payment of it is expressed at high levels during embryonic development page charges. This article must therefore be hereby marked advertisement in (11, 12), whereas it is hardly detectable in adult tissues (13). In accordance with 18 U.S.C. Section 1734 solely to indicate this fact. many tumor types, the HMGA2 gene is disrupted or aberrantly Note: I. Cleynen and J.R. Brants contributed equally to this work. I. Cleynen is a research assistant of the Fund for Scientific Research, Flanders, expressed, making this gene probably the most frequently Belgium (F.W.O. Vlaanderen). rearranged gene in human neoplasias (reviewed in refs. 14, 15). M. Petit is a postdoctoral fellow of the Fund for Scientific Research, Flanders, Belgium (F.W.O. Vlaanderen). The HMGA2 protein consists of three short basic domains Current address for K. Peeters: Center for Human Genetics, University Hospital (AT-hooks), enabling its binding to the minor groove of AT-rich Leuven, University of Leuven, Leuven, Belgium. DNA, followed by an acidic COOH-terminal tail (16). Disrup- Current address for Jan R. Brants: Human Pathology, Institut de Ge´ne´tique et Biologie Mole´culaire et Cellulaire, Centre National de la Recherche Scientifique/ tion of the gene in tumors, as a result of specific chromosomal Institut National de la Sante et de la Recherche Medicale/Universite´ Louis rearrangements, results in the expression of a truncated pro- Pasteur, Illkirch Cedex, France. tein (HMGA2Tr) only containing the DNA-binding AT-hooks Requests for reprints: Wim J.M. Van de Ven, Department of Human Genetics, Flanders Interuniversity Institute for Biotechnology, University of Leuven, (17, 18). Herestraat 49, Box 602, B-3000 Leuven, Belgium. Phone: 32-16-34-60-80; Fax: Previous work from our group identified Imp2 as a target 32-16-34-60-73. E-mail: [email protected] Copyright D 2007 American Association for Cancer Research. gene of both wild-type and tumor-specific truncated HMGA2 doi:10.1158/1541-7786.MCR-06-0331 proteins, suggesting an HMGA2-dependent Imp2 expression

regulation that plays an important role during embryonic devel- identification of two putative gene-regulatory regions for the opment and in tumorigenesis (10). In this study, we investigated mouse Imp2 gene (Fig. 1B, light gray bars). The first predicted the mechanism via which HMGA2 regulates Imp2 gene regulatory region (A¶, 604 bp) encompasses both the transcrip- expression. We show that HMGA2 and HMGA2Tr directly tion initiation site and the ATG translation initiation codon, regulate transcription of the Imp2 gene via an AT-rich regulatory whereas the second region (B¶, 276 bp) is located in the first region located in the first intron. We show that HMGA2 binds intron. The first region has 64.6% identity, whereas the second to this region both in vitro and in vivo. In reporter experiments, region has 73.2% identity, to the homologous regions of the we show that the AT-rich regulatory region mimics the response hIMP2 genomic sequence, pointing towards evolutionary of endogenous Imp2 to HMGA2 and HMGA2Tr. We also conserved noncoding regions (Fig. 2A). identified a consensus nuclear factor-nB (NF-nB) binding site located immediately adjacent to the AT-rich regulatory region The Intronic Gene-Regulatory Region Mimics the Re- that binds NF-nB and show that NF-nB and HMGA2 cooperate sponse of Endogenous Imp2 to HMGA2 and HMGA2Tr to regulate Imp2 gene expression. Finally, we investigated Previous work from our group identified Imp2 as a target whether there is a relationship between HMGA2 and IMP2 gene of both wild-type and tumor-specific truncated HMGA2 expression in a collection of human well-differentiated and proteins (10). HMGA2 and HMGA2Tr contain three DNA- myxoid liposarcomas and found a positive correlation. Collec- binding domains, called AT-hooks, enabling their binding to tively, our data show that Imp2 gene transcription is directly AT-rich DNA. Interestingly, the in silico predicted regulatory regulated by HMGA2 in cooperation with NF-nB. region B¶ was preceded by three AT-rich stretches, the latter being possible HMGA2-binding sites (Fig. 2A). We cloned Results regions A and B covering the in silico predicted regions A¶ and Imp2 The Mouse Gene Contains Two Putative Gene- B¶, respectively, along with flanking sequences (Fig. 1B) in Regulatory Regions front of the firefly luciferase reporter gene. To measure the We determined the mouse Imp2 genomic organization and influence of HMGA2 and HMGA2Tr proteins on reporter gene searched for possible gene-regulatory regions. The genomic expression, we transfected equal amounts of retrovirally organization of the Imp2 gene was obtained by performing a transduced NIH/3T3 fibroblasts stably expressing wild-type BLAT search against the University of California at Santa Cruz HMGA2, HMGA2Tr proteins, or Mock cells (10) with reporter Genome Browser database using the Imp2 cDNA sequence plasmid A or B. Twenty-four hours after transfection, luciferase (19). As shown in Fig. 1A, the Imp2 gene contains 16 exons and h-galactosidase activities were measured. Figure 2B shows spanning a region of about 100 kb. The transcription start site normalized luciferase values plotted relative to their values in (TSS) region was identified by searching the Database of Mock cells. Both HMGA2 and HMGA2Tr proteins had a slight 4 Transcriptional Start Sites (DBTSS) using Refseq sequence repressing effect on reporter plasmid A. More important, NM_183029 as query. This search revealed a region containing however, HMGA2 overexpression enhanced luciferase expres- 58 potential TSSs corresponding to 58 oligo-capped cDNA sion from reporter plasmid B almost 2-fold, whereas HMGA2Tr clones. The most upstream TSS, corresponding to the clone strongly repressed the activity of the same reporter. This latter with accession no. BY310082, was located 113 bp 5-prime to result reflects the endogenous Imp2 mRNA expression changes the translational start site of Imp2.5¶-RNA ligase-mediated induced by HMGA2 and HMGA2Tr, as seen previously in ¶ rapid amplification of cDNA ends (5 -RLM-RACE) analysis, Northern blot analyses (10). Furthermore, measuring the effect done on RNA isolated from NIH/3T3 cells, confirmed that the of HMGA2 and HMGA2Tr on reporter gene expression via TSS of the Imp2 gene was indeed located in this region, the combination of region A and B, where A is functioning as a ¶ ¶ most 5 TSS obtained being 29 bp downstream of the most 5 promoter and B as an enhancer, gave a similar result. Therefore, TSS in the DBTSS. In what follows, position 113 bp upstream we can conclude that region B is the major regulatory region via of the ATG translation initiation site was chosen as +1 (Fig. 1). which the HMGA2 proteins exert their effects on Imp2 gene To have a first indication of regions that are important in the expression. This region encompasses 419-bp intronic sequence regulation of the mouse Imp2 gene, we analyzed part of the of the mouse Imp2 gene (Fig. 1B) and contains three AT-rich Imp2 genomic sequence using different in silico promoter repeats (Fig. 2A). Deletion of the AT-rich repeats completely prediction programs. The genomic sequence that was analyzed abolished the effect of the HMGA2 proteins on luciferase spanned 7 kb and included 5 kb upstream of the assumed reporter gene expression (Fig. 2B), suggesting that this transcription initiation site, the first and the second coding conserved AT-rich stretch represents HMGA2-binding sites. exons, and the intron in between (intron 1; Fig. 1B). Online promoter prediction programs used included PromoterInspector In vitro and Gene2Promoter from GenomatixSuite,5 grailEXP (Grail HMGA2 and HMGA2Tr Bind to the AT-Rich Imp2 Experimental Gene Discovery Suite),6 and PROSCAN.7 Fragment of the Mouse Intronic Gene-Regulatory Combination of the obtained predictions resulted in the Region To investigate whether HMGA2 proteins are directly involved in Imp2 transcriptional regulation, we evaluated the HMGA2 DNA-binding activity to the Imp2 promoter. In 4 http://dbtss.hgc.jp/ particular, we analyzed an 84-bp region spanning nucleotides 5 http://www.genomatix.de 6 http://compbio.ornl.gov/grailexp/ 935 to 1018 in intron 1 of the murine Imp2 gene (boxed 7 http://thr.cit.nih.gov/molbio/proscan/ sequence in Fig. 2A), containing three AT-rich putative

FIGURE 1. Genomic organization and predicted gene regulatory regions of the mouse Imp2 gene. A. Genomic organization of the mouse Imp2 gene. Introns and genomic sequences (lines) and exons (boxes). Coding sequences (dark gray boxes) and 5¶ and 3¶ untranslated regions (open boxes). B. Schematic outline of a 7-kb mouse Imp2 genomic region containing the first two exons and 5-kb upstream sequence. Nucleotide positions are indicated. The (+1) position is the most 5¶ located TSS, based on expressed sequence tag with Genbank accession no. BY310082, according to the DBTSS. The nucleotide sequence of this 7-kb genomic region was retrieved from the University of California at Santa Cruz mouse genome browser and subjected to several promoter prediction programs. The smallest common region for each of the two (A¶ and B¶) identified putative gene-regulatory regions (light gray bars). Both regions along with flanking sequences were cloned (arrows) to investigate promoter activity.

HMGA2-binding sites (TATTTTT, AATTTTTT, and ATTT- specifically displaced by the incubation with antiserum directed TATT). As shown in Fig. 3A, a biotin-labeled double-strand against the HMGA2 protein (lanes 2 and 5). No complex was Imp2 oligonucleotide probe covering the three AT-stretches formed when the nuclear extracts were incubated with the formed DNA-protein complexes with nuclear extracts of the mutImp2 oligo (lanes 3 and 6). Together, these results show NIH/3T3 cell lines, stably expressing wild-type HMGA2 or that the intronic AT-rich regulatory region of the Imp2 gene HMGA2Tr proteins (lanes 2). No retarded protein-DNA binds HMGA2 proteins in vitro. complexes were formed when a mutant biotin-labeled probe was used (lanes 4). Specificity of these HMGA2-DNA and Hmga2 Binds In vivo to the Intronic Regulatory Region of HMGA2Tr-DNA complexes was confirmed by supershifts of the Mouse Imp2 Gene the bands upon incubation with HMGA2-specific antiserum To verify whether the HMGA2 protein binds to the intronic (lanes 5). Both HMGA2 full length and HMGA2Tr proteins regulatory region in vivo, we did chromatin immunoprecipita- bound to the Imp2 oligonucleotide probe. tion experiments. Therefore, the mouse NIH/3T3 fibroblasts In what follows, we limited our studies to full-length that stably express the wild-type HMGA2 protein were cross- HMGA2 because HMGA2 and HMGA2Tr, both having linked (see Materials and Methods) followed by immunopre- identical AT-hooks, bind the same DNA structures (20, 21). cipitation of the cross-linked protein-DNA complexes with Direct binding of the HMGA2 protein was confirmed by using HMGA2-specific antiserum. As a negative control, we included recombinant HMGA2-His full-length protein (Fig. 3B, lanes a no-antibody reaction. The precipitated double-stranded DNA 2-4). The formation of this DNA-protein complex was was amplified by PCR using the primer set depicted in Fig. 2B. markedly reduced by incubation with 200Â molar excess of As shown in Fig. 4, compared with the no-antibody control, a unlabeled Imp2 oligonucleotide (lane 5) and supershifted when band resulting from amplification of the Imp2 region is clearly incubating with HMGA2-specific antiserum (lane 7). In present. To rule out the possibility of nonspecific precipitation contrast 200Â excess of mutant Imp2 probe (mutImp2), in by the HMGA2 antibody, amplification of region À1789 to which the AT-rich sequences are mutated, did not affect À1606 bp relative to the TSS (Fig. 4B) was assayed as negative retardation of the HMGA2-DNA complex (lane 6). Further- control. We did not observe any association of HMGA2 with more, no complex was formed when incubating the purified this upstream region (Fig. 4A) establishing a specific, in vivo protein with the mutant biotin-probe (Fig. 3B, lane 9). binding of the HMGA2 protein to the AT-rich intronic region of To verify the binding of Hmga2 to the Imp2 promoter in the Imp2 gene, indisputably confirming our in vitro studies. primary cells, we assayed the DNA-binding activity of nuclear extracts prepared from mouse embryonic fibroblasts derived from E12.5 wild-type, heterozygous, or knockout Hmga2 NF-jB Binds to a Consensus Binding Site Immediately mouse embryos. As shown in Fig. 3C, specific DNA-protein Adjacent to the AT-Rich Regulatory Region complexes were present in extracts from both wild-type and Detailed analysis of region B revealed a consensus binding heterozygous mouse embryonic fibroblasts (lanes 1 and 4), site for NF-nB immediately adjacent to the AT-rich regulatory whereas they were absent in extracts derived from homozygous region (Fig. 2A). To investigate whether NF-nB is able to bind Hmga2 null mutant fibroblasts (lane 7). These complexes were to this site, we did an electrophoretic mobility shift assay using

mutated AT-hooks; compare lane 2 and lane 6). Together, these results indicate that NF-nB binds to the NF-nB consensus binding site adjacent to the AT-rich regulatory region, and that HMGA2 enhances this binding.

FIGURE 2. Identification of the Hmga2-regulated region. A. Nucleotide sequence and alignment of part of the intronic sequence containing reporter plasmid B (bold) of the human and mouse IMP2 gene. The 84-bp probe used in electrophoretic mobility shift assay experiments is boxed. Upper and lower primers used to amplify a 180-bp fragment of this genomic region in chromatin immunoprecipitation analysis (dashed arrows). AT-rich regions and the NF-nB binding site (underlined). B. HMGA2 and HMGA2Tr have opposite effects on the second regulatory region of the mouse Imp2 gene. NIH/3T3 cells retrovirally transduced with the empty retroviral vector (Mock), HMGA2, or HMGA2Tr were seeded in 24-well plates. Twenty-four hours after seeding, cells were cotransfected with 400 ng reporter construct or empty pGL3basic plasmid and 100 ng pEL1 h-galactosidase – expressing plasmid to normalize for transfection efficiency. Cells were harvested 24 h after transfection and assayed for luciferase and h-galactosidase activity. The ratio of luciferase/h-galactosidase activity was determined in each case and normalized to the respective luciferase value of the empty pGL3basic vector. Normalized luciferase values were calculated relative to their respective luciferase values in control (Mock) cells. Columns, mean of at least three independent experiments done in duplicate; bars, SE.

FIGURE 3. Hmga2 binding to the AT-rich stretch contained in reporter nuclear extracts from NIH/3T3 cells transiently cotransfected B. A. Biotin-labeled wild-type (WTImp2-bio, lanes 2) or mutant (mutImp2- bio, lanes 4) Imp2 promoter oligonucleotides were used in electrophoretic with equal amounts of p50 and p65 expressing constructs (NIH/ mobility shift assay. Binding assays were done using 5 Ag nuclear extract 3T3+). As shown in Fig. 5 (lane 2), incubation of these extracts of Mock, HMGA2, or HMGA2Tr NIH/3T3 cells with or without HMGA2- specific antiserum (lanes 5). Representative data of two independent with the 85-bp biotin-labeled double-stranded DNA probe experiments. B. Electrophoretic mobility shift assay was done by (Fig. 2A) results in the formation of a protein-DNA complex, incubating biotin-labeled wild-type (lanes 1-7) or mutant (lanes 8-9) which is not present when nuclear extract from nontransfected Imp2 promoter oligonucleotide with 5 ng (lane 2), 20 ng (lanes 3, 5-7, and 9), or 50 ng (lane 4) of the recombinant full-length HMGA2-HIS protein. NIH/3T3 cells is used (lane 1). Specificity of these protein- Where indicated, a 200-fold molar excess of unlabeled wild-type (lane 5) DNA complexes was confirmed by the detection of a or mutant (lane 6) oligonucleotides was added, or HMGA2-specific supershifted band when the mix was incubated with anti-p50 antiserum (lane 7). Representative data of two independent experiments. C. Electrophoretic mobility shift assays were done by incubating 5 Ag or anti-p65 antibody (lanes 3 and 4). Interestingly, the intensity nuclear extracts from mouse embryonic fibroblast derived from wild-type À À À of the complex diminishes when the mix was incubated with (WT), Hmga2 +/ , and Hmga2 / mice with biotin-labeled wild-type (lanes 1, 4, and 7) or mutant (lanes 3, 6, and 9) probes. Where indicated, the HMGA2-specific antiserum (compare lane 2 and lane 5) and samples were preincubated with HMGA2-specific antiserum. Represen- when a mutant biotin-labeled oligo was used (mutImp2-bio, tative data of two independent experiments.

FIGURE 4. In vivo association between HMGA2 and the intronic Imp2 gene-regulatory region. The binding of HMGA2 to region B of the mouse Imp2 gene was evaluated with a chromatin immunoprecipitation (ChIP) assay in NIH/3T3 cells stably expressing the wild-type HMGA2 protein. A. Immunoprecipitated DNA was amplified by PCR using a primer set specific for the Imp2 region (+891 to +1070 bp; right). As a negative control, immunoprecipitated DNA was amplified by a primer set specific to an off-target region of the Imp2 gene (À1789 to À1606 bp; left). DNA prepared from the input chromatin, taken before immunoprecipitation, was used as a positive control for PCR amplification. B. Schematic representation of the position of the regions amplified by PCR after chromatin immunoprecipitation. All sequences are drawn to scale, and nucleotide positions indicated are relative to the most 5¶ position of mouse expressed sequence tag with Genbank accession no. BY310082 (+1). Introns and genomic sequences (lines). Coding sequences (dark gray boxes); 5¶ and 3¶ untranslated regions (open boxes). Amplified PCR regions (light gray boxes; see A).

HMGA2 Enhances NF-jB–Mediated Transcriptional of well-differentiated liposarcoma is the presence of supernu- Activation of the Intronic Gene-Regulatory Region merary ring and giant marker chromosomes (22). These To verify whether HMGA2 is able to enhance transcriptional aberrant chromosomal rings and giant markers often contain activation by NF-nB, we did luciferase experiments in NIH/3T3 an amplification of the 12q13-q15 region, which harbors the cells. First, we looked at the transactivation capacity of NF-nB. HMGA2 gene. Several studies have shown immunopositivity of Cotransfection of a fixed amount of reporter B construct with increasing amounts of both p50- and p65-expressing constructs resulted in a dose-dependent increase of luciferase activity (data not shown). When we cotransfected the reporter B construct with a fixed amount of both p50- and p65-expressing constructs and an HMGA2-expressing construct, a further increase of luciferase activity was seen (Fig. 6A). Deletion of the AT-rich repeats clearly inhibited the enhancing effect of HMGA2 (Fig. 6B). When the NF-nB site was mutated, no induction was seen when p50/p65 was cotransfected. However, HMGA2 cotransfection still resulted in a 2-fold increase in luciferase activity (Fig. 6C). These results indicate that HMGA2 is able to enhance the activity of NF-nB by binding to the AT-rich repeats immediately adjacent to the NF-nB consensus site.

A Positive Correlation between HMGA2 and IMP2 Gene Expression in Liposarcomas Having shown that Imp2 gene expression is directly FIGURE 5. NF-kB binds to the NF-nB binding site immediately regulated by HMGA2 in cultured cells (this work) and during adjacent to the AT-rich regulatory region. Electrophoretic mobility shift A embryonic development (10), we next investigated whether assay was done by incubating 5 g nuclear extract of nontransfected (lane 1) or p50/p65 – transfected (NIH/3T3+, lanes 2-6) NIH/3T3 cells with there is a relationship between HMGA2 and IMP2 gene biotin-labeled wild-type (lanes 1-5) or mutant (lane 6) Imp2 promoter expression in tumors. Therefore, we compared expression oligonucleotides. Where indicated, the samples were preincubated with antibody directed against p50 (lane 3) or p65 (lane 4) or with HMGA2- levels of HMGA2 and IMP2 in human well-differentiated and specific antiserum (lane 5). Representative data of two independent myxoid liposarcomas. The most important cytogenetic feature experiments.

FIGURE 6. Ability of HMGA2 to enhance transcriptional activation by NF-nB. NIH/3T3 cells, seeded in 24-well plates, were transiently cotransfected with 400 ng reporter B (A), Bdel (B), or BmutNF-nB(C) with or without an HMGA2-expressing construct and p50/p65 expression vectors as indicated. pEL1 h- galactosidase – expressing plasmid (100 ng) was added to normalize for transfection efficiency. Cells were harvested 48 h after transfection and assayed for luciferase and h-galactosidase activity. The ratio of luciferase/h-galactosidase activity was determined in each case and normalized to the respective luciferase value of the empty pGL3basic vector. Normalized luciferase values were calculated relative to their respective luciferase values in control (no HMGA2 or p50/p65 cotransfected) cells. Columns, mean of at least three independent experiments; bars, SE.

HMGA2 in these tumors and/or amplification of the HMGA2 Discussion gene by Southern blotting or fluorescence in situ hybridization Here, we present the first data on the direct transcriptional (23-26). Furthermore, Pentimalli et al. (27) have shown that regulation of the Imp2 gene. A first major finding is the HMGA2 protein suppression results in growth inhibition of identification of an AT-rich regulatory region in the first intron HMGA2-overexpressing primary cell cultures of well-differen- of the Imp2 gene. In reporter experiments, this AT-rich region tiated liposarcoma, indicating that HMGA2 plays an important mimics the response of the endogenous Imp2 gene to the role in the growth of these tumors. The cytogenetic hallmark of architectural transcription factor HMGA2 and its truncated myxoid liposarcoma is the presence of t(12;16)(q13;p11) that tumor-specific counterpart HMGA2Tr, as shown by Brants leads to the formation of the FUS-DDIT3 fusion protein (28-31). et al. (10). In addition, we show that the HMGA2 protein binds In contrast to well-differentiated liposarcoma, the HMGA2 gene directly to the AT-rich region and is also associated with this is not amplified in myxoid liposarcoma. We analyzed the expression levels of HMGA2 and IMP2 by quantitative real-time reverse transcription-PCR in 11 samples TABLE 1. Clinical and Cytogenetic Information of the 11 of human well-differentiated liposarcoma and 3 specimens of Well-Differentiated (S4-S14) and 3 Myxoid (S1*-S3*) Lipo- myxoid liposarcoma, which were obtained from 14 different sarcomas patients. The karyotypes of the tumors are depicted in Table 1. As can be seen in Table 1, the 11 well-differentiated lipo- Case no. Gender/age Karyotype sarcoma (S4-S14) were characterized by supernumerary ring S1* F/65 46, XX, del(11)(q23), t(12;16)(q13;p11), chromosomes and/or giant marker chromosomes, whereas the 3 t(14;20)(p11;q11), t(15;17)(q22;q25) myxoid liposarcoma (S1*-S3*) showed the presence of S2* M/43 46, XY, t(12;16)(q13;p11) t(12;16)(q13;p11) translocation, specific for this entity. The S3* M/42 47, XY, +5, t(12;16)(q13;p11) S4 F/57 46-49, XX, +GM, +1-3R[inc9] obtained DCt values for HMGA2 and IMP2, which are in- S5 M/64 43-45, COMPLEX, +1-2R,+GM[inc10] versely related to their expression level, are plotted in Fig. 7. S6 M/44 42-47, XY, +1-2GM[inc11] S7 M/82 42-48, XY, clonal and non-clonal aberrations, A strong and statistically significant relationship was observed +1-4R, +DMIN between HMGA2 and IMP2 gene expression (r = 0.8066, S8 M/76 45-47, XY, +R[inc4] P = 0.000244, Spearman rank test). From Fig. 7, it is also clear S9 M/80 43-48, XY, +1-2R S10 F/49 43-51, XX, +1-2GM, +1-4R, +DMINS that the three myxoid liposarcoma samples that did not S11 M/58 46-47, XY, +1-2GM[inc11] have ring chromosomes and/or giant marker chromosomes S12 M/70 47, XY, +R (S1*-S3*) and thus were not expected to have a high expression S13 F/72 42-46, XX, +GM[inc15] S14 M/64 46, XY, COMPLEX, +R, +GM[inc10] level of HMGA2 indeed had low expression levels (high DCt) of HMGA2 and also of IMP2. Abbreviations: F, female; M, male.

FIGURE 7. Correlation between HMGA2 and IMP2 gene expression in well-differentiated versus myxoid liposarcomas. DC t values that were determined by real-time quantitative reverse transciprion-PCR and are inversely related to the expression level of HMGA2 and IMP2 in the 14 tumor samples listed in Table 1. Data were statistically analyzed using the Spearman rank test. Points, mean DC t of at least two independent experiments for a particular sample. region in vivo. HMGA2 is a member of the HMGA family of mouse expressed sequence tags described above. Both in proteins that comprises four known members: HMGA1a, human and mouse, an alternative in frame ATG start codon is HMGA1b, and HMGA1c, which are encoded by alternative present in exon 2, which is preceded by an in-frame stop splice products of the HMGA1 gene, and HMGA2, which is codon. This suggests that the protein that is translated from the encoded by a different gene (32). HMGA proteins do not shorter alternative transcript lacks the first 67 amino acids, have intrinsic transcriptional activation capacity but instead which correspond to the RNA recognition motif-1 domain, play a role in the formation and stabilization of higher- one of the six RNA-binding domains in the Imp2 protein (4). order nucleoprotein complexes (enhanceosomes) in promoter/ At present, to the best of our knowledge, the function of this enhancer regions of their target genes, as such acting as domain in the Imp2 protein is not known. Further experiments ‘‘architectural’’ transcription factors. The best characterized are needed to investigate whether the intronic AT-rich example of how HMGA proteins modulate gene expression is regulatory region regulates expression of the ordinary longer the transcriptional regulation of the IFN-b gene, where transcript as an enhancer, or of the alternative shorter transcript HMGA1 orchestrates the assembly of different transcription as a promoter, or both. As shown in Fig. 2, the AT-rich factors into an enhanceosome (33). In contrast to HMGA1, the regulatory region is functioning both as promoter element role of HMGA2 as an architectural transcription factor is less (reporter B) and as enhancer element (reporter A + B) in well studied. To our knowledge, to date, besides Imp2, only luciferase reporter assays, thereby each time mimicking the five other genes have been identified as target genes of response of the endogenous Imp2 gene to HMGA2 and HMGA2, being the CDC2, TK1, and cyclin E genes (34); the HMGA2Tr. DNA repair gene ERCC1 (21); and the cyclin A gene, which is A second major finding is that transcription of the Imp2 a key factor in cell cycle control (35). In the two latter cases, gene is regulated by NF-nB, and that this regulation takes place it has been shown that HMGA2 binds to AT-rich regions in the in cooperation with HMGA2. The fact that NF-nB and HMGA promoter of these target genes. Concerning the Imp2 gene, proteins work together in regulating transcription is not novel: we show here that the intronic AT-rich regulatory region is it was first discovered by Thanos and Maniatis (38) who mandatory for the regulation of the gene. This AT-rich showed that HMGA1 is required for NF-nB–dependent virus regulatory region is located in the first intron of the gene, induction of the human IFN-b gene. Later, Mantovani et al. about 1 kb downstream of the promoter, suggesting that this (39) found that NF-nB–mediated transcriptional activation is region has a function as a cis-acting transcriptional enhancer, also enhanced by HMGA2, and in a follow-up study by Noro and that HMGA2, like HMGA1, also regulates transcrip- et al. (20), a physical interaction between HMGA2 and the tion from an enhancer (36, 37). However, by screening the p50/p65 subunits of NF-nB was shown. The sequence-specific DBTSS, a mouse Imp2 expressed sequence tag (accession no. transcription factor NF-nB portrays various dimeric complexes BB643209) was found that has its TSS at À27 bp relative to of members of the Rel protein family. Of the various dimeric the beginning of exon 2. By performing a BLAST (Basic combinations, the p50/p65 dimer, which was used in this study, Local Alignment Search Tool) search, two additional mouse is most common (40). expressed sequence tags (accession nos. CJ171095 and The third and last major finding is a relationship between CJ169368) were identified both containing 144 bp of intron 1 HMGA2 and IMP2 gene expression in well-differentiated sequence located immediately upstream of exon 2. These three versus myxoid liposarcomas. Well-differentiated liposarcomas mouse expressed sequence tags suggest the presence of a seemed to be especially suited to study the question about a second TSS, which is located downstream of the AT-rich possible relationship between HMGA2 and IMP2 expression regulatory region. This implies that the AT-rich regulatory in tumors because they displayed different levels of HMGA2 region could be part of an alternative promoter directing the expression as can be seen in Fig. 7. We found a strong positive, transcription of a shorter alternative Imp2 transcript lacking statistically significant correlation between HMGA2 and IMP2 exon 1. Additional evidence for such an alternative trans- expression levels in these tumors. Together with the other cript was found in human: two human IMP2 expressed results presented in this study, this suggests that IMP2 sequence tags were identified (accession nos. BX491079 and overexpression in tumors is accomplished via HMGA2 in CN398355), which have a similar composition as the three association with NF-nB.

Materials and Methods Cell lines used included NIH/3T3 (mouse embryo fibroblast Cloning of mImp2 Regulatory Regions cells; ATCC CRL-1658) and NIH/3T3 cells stably expressing BAC clone RP23-50H13 covering the mouse genomic HMGA2 or HMGA2Tr, or stably transfected with empty vector region containing the Imp2 gene was selected by using the (mock) as described previously (10). Cells were incubated at 37jC University of California at Santa Cruz Genome Browser (19) and 5% CO2 in a humidified incubator. Plasmid DNA transfections and purchased from Invitrogen/Life Technologies (Carlsbad, were done using Fugene 6 Transfection Reagent (Roche, Basel, CA). Promoter regions A (901 bp) and B (419 bp; Fig. 1B) Switzerland) according to the manufacturer’s protocol. were amplified from this BAC clone using Invitrogen TaqPCRx DNA Polymerase in combination with PCRx Enhancer Luciferase Reporter Assays Solution according to the manufacturer’s protocol for GC-rich Twenty-four to 48 hours after transfection of semiconfluent ¶ templates. Primer sets used were 5 -CCTGTGCATCGTCAC- cells, cells were lysed and assayed for luciferase activity using ¶ ¶ CTCGGCCATC-3 /5 -GGCGTAGCCGGACTTGAGTAG- the Luciferase Assay System (Promega) as described in Crombez ¶ ¶ GACCTGTC-3 and 5 -CCCTCGGTCCTTTCCCGAAGCTC- et al. (42). A h-galactosidase expressing plasmid was cotrans- ¶ ¶ ¶ AG-3 /5 -GGGTGACCCAAAGGTCTCGCCAGC-3 ,respec- fected for normalizing the transfection efficiency. h-Galactosi- tively. The obtained fragments were purified and TA-cloned dase activity was assayed as described in Sambrook et al. (43). in the pGEMTeasy vector (Promega, Madison, WI). To obtain Luciferase activity and h-galactosidase A420 were measured in a promoter region B lacking the three AT-rich stretches (Bdel), Wallac Victor2 1420 Multilabel Counter (Perkin-Elmer Life nucleotides 54 to 89 of promoter region B were removed by Sciences, Wellesley, MA). For each experiment, luciferase and PCR-based site-directed mutagenesis (Stratagene, La Jolla, CA) h-galactosidase activities were determined in duplicate wells. ¶ using primer set 5 -GTCTCGGAAGCGCTCGAAGAACTC- The results are the mean of at least three individual experiments. GGGGCT-3¶/5¶-AGCCCCGAGTTCTTCGAGCGCTTCCGA- GAC-3¶. After verification by nucleotide sequence analysis, inserts were excised from the pGEMTeasy vector and blunt- Preparation of Nuclear Extracts for Electrophoretic ligated in the SmaI site of the pGL3Basic luciferase reporter Mobility Shift Assay vector (Promega). Nuclear extracts were prepared from NIH/3T3 cells accord- To obtain reporter plasmids containing promoter region A ing to the procedure described (44). Aliquots of the nuclear upstream and promoter region B or B downstream of the extracts were immediately frozen on dry ice and stored at del À j luciferase coding sequence in the pGL3Basic vector, region B 80 C. The protein concentration of the nuclear extracts was determined by the Bradford assay (BCA Protein Assay kit, or Bdel was cloned in the BamHI and SalI sites of the region A- containing pGL3Basic vector. Pierce Perbio Science, Rockford, IL). The mutation of the NF-nB site was generated by PCR- based site-directed mutagenesis (Stratagene) using primer set Protein Purification 5¶-GGCTGATAGCTGACCGCCTCCCGTCTCCC-3¶/ The coding sequence of human HMGA2, which was 5¶-GGGAGACGGGAGGCGGTCAGCTATCAGCC-3¶ on the adapted for enhanced bacterial expression, was cloned in the promoter region B pGEMTeasy plasmid. NdeI/XhoI sites of the pET-31b(+) bacterial expression vector (Novagen, Darmstadt, Germany), resulting in a construct expressing COOH-terminally HIS-tagged HMGA2 protein 5¶-RLM-RACE (kindly provided by Torik Ayoubi, University of Maastricht, RNA was isolated from NIH/3T3 cells using the RNeasy total the Netherlands). RNA isolation kit (Qiagen, Hilden, Germany), using the The HMGA2 protein was expressed in the Escherichia coli conditions described by the manufacturer. 5¶-RLM-RACE was BL21 strain in the following way: BL21 was grown in 500 mL done using the FirstChoice RLM-RACE kit (Ambion, Austin, Luria-Bertani medium containing ampicillin at 37jC. When TX) according to the supplier’s instructions. The following the A reached a value of about 0.5, 1 mmol/L isopropyl Imp2-specific nested primers were used: outer, 5¶-CCGGACTT- 600 h-D-thiogalactopyranoside (Roche) was added, and the culture GAGTAGGACCTG-3¶;inner,5¶-GGTTCCCAATGTACAG- was incubated overnight at 30jC. The cells were collected CTTG-3¶. The obtained PCR products were size-fractionated by centrifugation and dissolved in buffer A [50 mmol/L Tris/ by agarose gel electrophoresis, purified, cloned, and sequenced. HCl (pH 8), 500 mmol/L NaCl] containing 10 Ag/mL DNase, The transcription start sites were mapped by aligning the 10 Ag/mL RNase, and 1 mg/mL lysozyme. After incubating 5¶-RLM-RACE sequences with genomic Imp2 sequences. 30 min on ice followed by sonication, the sonicate was centrifuged for 15 min at maximum speed. Purification of the Cell Culture and Transfection HMGA2-HIS protein from the supernatant fraction was carried Mouse embryonic fibroblasts were derived from E12.5 out as follows: after washing in buffer A, 100 AL Ni-beads were embryos (Hmga2+/+, Hmga2+/À, and Hmga2À/À; ref. 17) as added to the clear sonicate and rotated for 1 h at 4jC. Then the described (41). In brief, mouse embryonic fibroblasts were beads were washed twice for 1 min and twice for 5 min in 5 mL explanted, maintained on a 3T9 protocol, and propagated in buffer A and retained in a glass wool column by 1-min low- DMEM (Invitrogen, Carlsbad, CA) containing 10% fetal speed centrifugation. The column was washed once in buffer B bovine serum, 2 mmol/L glutamine, 0.1 mmol/L nonessential [50 mmol/L Tris/HCl (pH 8), 100 mmol/L NaCl] before amino acids, 55 Amol/L 2-mercaptoethanol, and 10 Ag/mL elution of the HIS-protein with 200 AL buffer B containing gentamicin (Invitrogen). 250 mmol/L imidazole. All buffers used contained Complete

protease inhibitors (Roche). Elution was repeated, and the 70% power, duty cycle of 25% during 6 min on ice. Immu- purity of both elutes was checked by SDS-PAGE. The con- noprecipitation was done using 10 AL HMGA2fix antiserum. centration of the purified proteins was determined by the Salmon sperm DNA/Protein A-Sepharose/(50%) slurry used to Bradford assay (BCA Protein Assay kit, Pierce). preclear the lysates and to collect the antibody/histone complex was prepared as follows: 2.5 mL packed protein A-Sepharose CL-4B beads (GE Healthcare) were incubated by overnight Electrophoretic Mobility Shift Assay rotation at 4jC in a final volume of 5 mL TE buffer (pH 8) Electrophoretic mobility shift assays were done using the containing 1 mg sonicated salmon sperm DNA, 2.5 mg bovine LightShift Chemiluminescent EMSA kit (Pierce). Briefly, 5 Ag serum albumin, and 0.05% sodium azide. After elution of the of NIH/3T3 nuclear extracts or different amounts (5, 20, and histone complex and reversing of histone-DNA cross-links, 50 ng) of purified proteins were preincubated in the binding DNA was recovered using the Qiaquick PCR purification buffer containing 5 mmol/L MgCl , 50 ng/AL poly(deoxyino- 2 protocol (Qiagen) followed by elution in 50 AL water before sinic-deoxycytidylic acid), 0.05% NP40, and 0.2 Ag/AL bovine PCR amplification. The amplification was done using 3 AL serum albumin for 10 min at room temperature. After DNA template with the following primer sets. Promoter preincubation, the samples were either directly used in the region B, 5¶-TCGGTCCTTTCCCGAAGCTC-3¶/5¶-CCGAGT- binding assay, incubated for 10 min at room temperature with 6 TGAGACAGGGTACC-3¶; the upstream negative control pmol unlabeled competitor probe, or incubated for 1 h at 4jC À1789/À1606 bp region (Fig. 4B), 5¶-CTCAGAGGCAGTG- with 5 AL HMGA2fix antiserum (10), 2 Ag p50, or 2 Ag p65 TAAAAGAGGGCTCC-3¶/5¶-GTTGGTGTTCTCCCTGG- antibody (Santa Cruz, Santa Cruz, CA). Biotin-labeled probe CTGAAAGG-3¶. The obtained PCR product (20 AL) was (30 fmol) was added in a total volume of 20 AL, and the analyzed by electrophoresis on a 2% agarose gel. DNA was reaction mixture was incubated for 20 min at room temperature. visualized by ethidium bromide staining. The protein-DNA complexes were separated by electrophoresis on a 6% polyacrylamide gel in 0.5Â Tris-borate EDTA buffer at 4jC. After size fractionation, protein-DNA complexes Tumor Samples and RNA Isolation were electrophoretically transferred to Hybond N+ membrane Eleven well-differentiated liposarcomas and three myxoid (GE Healthcare, Uppsala, Sweden) for 40 min at 380 mA liposarcomas were resected from 14 patients. Single-cell suspen- followed by cross-linking at 120 mJ/cm2, using a Stratalinker sions of the tumors were obtained by overnight collagenase treat- (Stratagene). Detection and visualization of the protein-DNA ment, whereafter suspensions were frozen in liquid nitrogen. All complexes was done by incubation with streptavidin-horserad- tumors were karyotyped according to standard procedures (45). ish peroxidase conjugate and detection reagents according to To isolate total RNA, collagenase-treated cells were thawed, the manufacturer’s instructions. washed twice in cold PBS, and pelleted. Total RNA was The sequences of the upper strand of the oligonucleotides isolated using the NucleoSpin RNA II total RNA Isolation kit used are WTImp2-bio, 5¶-GCGCTCCTATTTTTCCCTGAA- (Macherey-Nagel, Du¨ren, Germany), according to conditions TTTTTTGTCCGATTTTATTGAGAAGAACTCGGGG- described by the manufacturer. DNase digestion was done CTGGGGGCTTTCCGCCTCCCG-Biotin-3¶; MutImp2-bio, directly on the purification column before elution of total RNA, 5¶-CGCTCCTGTGTGTCCCTGAGTGTGTTGTCCGA- as described in the manual. GTGTGTTGAGAAGAACTCGGGGCTGGGGGCTT- TCCGCCTCCCG-Biotin-3¶; WTImp2, 5¶-CGCTCCTATTTT- Quantitative Reverse Transcription-PCR TCCCTGAATTTTTTGTCCGATTTTATTGAGAA- cDNA was prepared from 5 Ag DNase I–treated total RNA, GAACTCG-3¶; MutImp2, 5¶-CGCTCCTGTGTGTCCCTGA- using the Superscript II First-Strand system for reverse GTGTGTTGTCCGAGTGTGTTGAGAAGAACTCG-3¶; and transcription-PCR (Invitrogen) according to the conditions coNFnB, 5¶-AGTTGAGGGGACTTTCCCAGGC-3¶. All oligo- described by the manufacturer. Real-time PCR was done using nucleotides were purchased from Eurogentec (Seraing, Belgium). the qPCR Mastermix for SYBR green I (Eurogentec) and was carried out using the ABI Prism 7700 sequence detection Chromatin Immunoprecipitation Assay system (Applied Biosystems, Foster City, CA). Target cDNA Chromatin immunoprecipitation assays were done by was amplified in duplicate using the following conditions: following the chromatin immunoprecipitation assay pro- 2 min at 50jC, 10 min at 95jC, followed by 40 cycles of tocol from Upstate Biotechnology (Millipore, Billerica, MA) amplification (15 s at 95jC and 1 min at 60jC). A non-template with minor modifications. Briefly, NIH/3T3 cells stably reaction was included as negative control. After amplification, overexpressing the wild-type HMGA2 protein were seeded in samples were subjected to a dissociation protocol to verify the 10-cm cell culture dishes, grown to nearly confluence, and presence of a single amplification product and the absence cross-linked with 1% formaldehyde for 10 min at 37jC. After of nonspecific PCR products. GAPDH (glyceraldehyde-3- rinsing twice with ice-cold PBS containing Complete protease phosphate dehydrogenase) gene was used as a reference inhibitors (Roche), cells were collected and pelleted by gene to normalize for variations in the amount of input centrifugation at 350 Â g for 4 min at 4jC. Cell pellets were cDNA.Primersetswerechoseninsuchawaythattheyanneal resuspended in 200 AL chromatin immunoprecipitation SDS in exons separated by large intronic sequences to avoid lysis buffer containing Complete protease inhibitors (Roche) amplification of contaminating genomic DNA and were designed and incubated for 10 min on ice. Pooled lysate from two culture with Primer Express 2.0 (Applied Biosystems). Primers included dishes was sonicated using a Branson 250 sonicator, applying qhHMGA2up, 5¶-TCCCTCTAAAGCAGCTCAAAAGA-3¶;

qhHMGA2low, 5¶-TGTTGTGGCCATTTCCTAGGT-3¶; 19. Kent WJ, Sugnet CW, Furey TS, et al. The human genome browser at UCSC. qhIMP2up, 5¶-CCCCCATCACCAGTTTGG-3¶; qhIMP2low, Genome Res 2002;12:996 – 1006. ¶ ¶ ¶ 20. Noro B, Licheri B, Sgarra R, et al. Molecular dissection of the architectural 5 -CTGGGTTGGGATGAAGAGATTC-3 ; GAPDHup, 5 -ATG- transcription factor HMGA2. Biochemistry 2003;42:4569 – 77. GCCTTCCGTGTTCCT-3¶; and GAPDHlow, 5¶-CAGGCGGC- 21. Borrmann L, Schwanbeck R, Heyduk T, et al. High mobility group A2 ACGTCAGAT-3¶. All primers were purchased from Eurogentec. protein and its derivatives bind a specific region of the promoter of DNA repair gene ERCC1 and modulate its activity. Nucleic Acids Res 2003;31:6841 – 51. 22. Arrigoni G, Doglioni C. Atypical lipomatous tumor: molecular character- Statistical Analysis ization. Curr Opin Oncol 2004;16:355 – 8. The Spearman rank correlation test was used to test a 23. Dei Tos AP, Doglioni C, Piccinin S, et al. Coordinated expression and possible correlation between HMGA2 and IMP2 expression. amplification of the MDM2, CDK4, and HMGI-C genes in atypical lipomatous tumours. J Pathol 2000;190:531 – 6. The P value reported is one tailed. 24. Forus A, Larramendy ML, Meza-Zepeda LA, et al. Dedifferentiation of a well-differentiated liposarcoma to a highly malignant metastatic osteosarcoma: amplification of 12q14 at all stages and gain of 1q22-q24 associated with Acknowledgments metastases. Cancer Genet Cytogenet 2001;125:100 – 11. We thank Steffen Fieuws (Biostatistical Center, University Hospital Leuven, Belgium) for help with the statistical analysis. 25. Tallini G, Dal Cin P, Rhoden KJ, et al. Expression of HMGI-C and HMGI(Y) in ordinary lipoma and atypical lipomatous tumors: immunohistochemical reactivity correlates with karyotypic alterations. Am J Pathol 1997;151:37 – 43. References 26. Trahan S, Erickson-Johnson MR, Rodriguez F, et al. Formation of the 12q14-q15 1. Leeds P, Kren BT, Boylan JM, et al. Developmental regulation of CRD-BP, an amplicon precedes the development of a well-differentiated liposarcoma arising RNA-binding protein that stabilizes c-myc mRNA in vitro. Oncogene 1997;14: from a nonchondroid pulmonary hamartoma. Am J Surg Pathol 2006;30:1326 –9. 1279 – 86. 27. Pentimalli F, Dentice M, Fedele M, et al. Suppression of HMGA2 protein 2. Mori H, Sakakibara S, Imai T, et al. Expression of mouse igf2 mRNA-binding synthesis could be a tool for the therapy of well differentiated liposarcomas protein 3 and its implications for the developing central nervous system. overexpressing HMGA2. Cancer Res 2003;63:7423 – 7. J Neurosci Res 2001;64:132 – 43. 28. Aman P, Ron D, Mandahl N, et al. Rearrangement of the transcription factor 3. Mueller-Pillasch F, Lacher U, Wallrapp C, et al. Cloning of a gene highly gene CHOP in myxoid liposarcomas with t(12;16)(q13;p11). Genes Chromo- overexpressed in cancer coding for a novel KH-domain containing protein. somes Cancer 1992;5:278 – 85. Oncogene 1997;14:2729 – 33. 29. Crozat A, Aman P, Mandahl N, Ron D. Fusion of CHOP to a novel RNA- 4. Nielsen J, Christiansen J, Lykke-Andersen J, Johnsen AH, Wewer UM, binding protein in human myxoid liposarcoma. Nature 1993;363:640 – 4. Nielsen FC. A family of insulin-like growth factor II mRNA-binding proteins 30. Rabbitts TH, Forster A, Larson R, Nathan P. Fusion of the dominant negative represses translation in late development. Mol Cell Biol 1999;19:1262 – 70. transcription regulator CHOP with a novel gene FUS by translocation t(12;16) in 5. Yaniv K, Yisraeli JK. The involvement of a conserved family of RNA binding malignant liposarcoma. Nat Genet 1993;4:175 – 80. proteins in embryonic development and carcinogenesis. Gene 2002;287:49 – 54. 31. Turc-Carel C, Limon J, Dal Cin P, Rao U, Karakousis C, Sandberg AA. 6. Yisraeli JK. VICKZ proteins: a multi-talented family of regulatory RNA- Cytogenetic studies of adipose tissue tumors. II. Recurrent reciprocal translocation binding proteins. Biol Cell 2005;97:87 – 96. t(12;16)(q13;p11) in myxoid liposarcomas. Cancer Genet Cytogenet 1986;23:291–9. 32. Reeves R, Beckerbauer L. HMGI/Y proteins: flexible regulators of 7. Lu M, Nakamura RM, Dent ED, et al. Aberrant expression of fetal RNA- transcription and chromatin structure. Biochim Biophys Acta 2001;1519:13 – 29. binding protein p62 in liver cancer and liver cirrhosis. Am J Pathol 2001;159: 945 – 53. 33. Yie J, Merika M, Munshi N, Chen G, Thanos D. The role of HMG I(Y) in the assembly and function of the IFN-beta enhanceosome. EMBO J 1999;18:3074 – 89. 8. Zhang JY, Chan EK, Peng XX, et al. Autoimmune responses to mRNA binding proteins p62 and Koc in diverse malignancies. Clin Immunol 2001;100: 34. Fedele M, Visone R, De Martino I, et al. HMGA2 induces pituitary 149 – 56. tumorigenesis by enhancing E2F1 activity. Cancer Cell 2006;9:459 – 71. 35. Tessari MA, Gostissa M, Altamura S, et al. Transcriptional activation of the 9. Koziol JA, Zhang JY, Casiano CA, et al. Recursive partitioning as an approach cyclin A gene by the architectural transcription factor HMGA2. Mol Cell Biol to selection of immune markers for tumor diagnosis. Clin Cancer Res 2003;9: 2003;23:9104 – 16. 5120 – 6. 36. Bagga R, Michalowski S, Sabnis R, Griffith JD, Emerson BM. HMG I/Y 10. Brants JR, Ayoubi TA, Chada K, Marchal K, Van de Ven WJ, Petit MM. regulates long-range enhancer-dependent transcription on DNA and chromatin by Differential regulation of the insulin-like growth factor II mRNA-binding protein changes in DNA topology. Nucleic Acids Res 2000;28:2541 – 50. genes by architectural transcription factor HMGA2. FEBS Lett 2004;569:277 – 83. 37. Reeves R. HMGA proteins: flexibility finds a nuclear niche? Biochem Cell 11. Hirning-Folz U, Wilda M, Rippe V, Bullerdiek J, Hameister H. The Biol 2003;81:185 – 95. expression pattern of the Hmgic gene during development. Genes Chromosomes 38. Thanos D, Maniatis T. The high mobility group protein HMG I(Y) is required Cancer 1998;23:350 – 7. for NF-n B-dependent virus induction of the human IFN-beta gene. Cell 1992;71: 12. Chiappetta G, Avantaggiato V, Visconti R, et al. High level expression of the 777 – 89. HMGI (Y) gene during embryonic development. Oncogene 1996;13:2439 – 46. 39. Mantovani F, Covaceuszach S, Rustighi A, et al. NF-nB mediated 13. Zhou X, Benson KF, Przybysz K, et al. Genomic structure and expression of transcriptional activation is enhanced by the architectural factor HMGI-C. the murine Hmgi-c gene. Nucleic Acids Res 1996;24:4071 – 7. Nucleic Acids Res 1998;26:1433 – 9. 14. Hess JL. Chromosomal translocations in benign tumors: the HMGI proteins. 40. Moynagh PN. The NF-nB pathway. J Cell Sci 2005;118:4589 – 92. Am J Clin Pathol 1998;109:251 – 61. 41. Kamijo T, Zindy F, Roussel MF, et al. Tumor suppression at the mouse 15. Tallini G, Dal Cin P. HMGI(Y) and HMGI-C dysregulation: a common INK4a locus mediated by the alternative reading frame product p19ARF. Cell occurrence in human tumors. Adv Anat Pathol 1999;6:237 – 46. 1997;91:649 – 59. 16. Bustin M, Reeves R. High-mobility-group chromosomal proteins: architec- 42. Crombez KR, Vanoirbeek EM, Van de Ven WJ, Petit MM. Transactivation tural components that facilitate chromatin function. Prog Nucleic Acid Res Mol functions of the tumor-specific HMGA2/LPP fusion protein are augmented by Biol 1996;54:35 – 100. wild-type HMGA2. Mol Cancer Res 2005;3:63 – 70. 17. Ashar HR, Fejzo MS, Tkachenko A, et al. Disruption of the architectural 43. Sambrook J, Russell DW. Molecular Cloning: a Laboratory Manual. 3rd. factor HMGI-C: DNA-binding AT hook motifs fused in lipomas to distinct Cold Spring Harbor: Cold Spring Harbor Laboratory Press; 2001. transcriptional regulatory domains. Cell 1995;82:57 – 65. 44. Wadman IA, Osada H, Grutz GG, et al. The LIM-only protein Lmo2 is a 18. Kools PF, Wanschura S, Schoenmakers EF, et al. Identification of the bridging molecule assembling an erythroid, DNA-binding complex which includes chromosome 12 translocation breakpoint region of a pleomorphic salivary gland the TAL1, E47, GATA-1 and Ldb1/NLI proteins. EMBO J 1997;16:3145– 57. adenoma with t(1;12)(p22;q15) as the sole cytogenetic abnormality. Cancer Genet 45. Polito P, Dal Cin P, Debiec-Rychter M, Hagemeijer A. Human solid tumors: Cytogenet 1995;79:1 – 7. cytogenetic techniques. Methods Mol Biol 2003;220:135 – 50.

Mol Cancer Res 2007;5(4). April 2007 Downloaded from mcr.aacrjournals.org on September 28, 2021. © 2007 American Association for Cancer Research. HMGA2 Regulates Transcription of the Imp2 Gene via an Intronic Regulatory Element in Cooperation with Nuclear Factor-κB

Isabelle Cleynen, Jan R. Brants, Kristel Peeters, et al.

Mol Cancer Res 2007;5:363-372.

Updated version Access the most recent version of this article at: http://mcr.aacrjournals.org/content/5/4/363

Cited articles This article cites 44 articles, 9 of which you can access for free at: http://mcr.aacrjournals.org/content/5/4/363.full#ref-list-1

Citing articles This article has been cited by 11 HighWire-hosted articles. Access the articles at: http://mcr.aacrjournals.org/content/5/4/363.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mcr.aacrjournals.org/content/5/4/363. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.