- Email: [email protected]
Molecular characterization of NSD1, a human homologue of the mouse Nsd1 gene
Gene 279 (2001) 197–204 www.elsevier.com/locate/gene
Molecular characterization of NSD1, a human homologue of the mouse Nsd1 gene Naohiro Kurotaki a,b,c, Naoki Harada a,b,e, Koh-ichiro Yoshiura a,b, Sumio Sugano d, Norio Niikawa a,b, Naomichi Matsumoto a,b,* a
Department of Human Genetics, Nagasaki University School of Medicine, Sakamoto 1-12-4, Nagasaki 852-8523, Japan b CREST, Japan Science and Technology Corporation, Honcho 4-1-8, Kawaguchi 332-0012, Japan c Department of Neuropsychiatry, Nagasaki University School of Medicine, Sakamoto 1-12-4, Nagasaki 852-8523, Japan d Department of Virology, Institute of Medical Science, University of Tokyo, Shirokanedai 4-6-1, Minato-ku, Tokyo 108-8639, Japan e Kyusyu Medical Science Nagasaki Laboratory, Hamaguchi-machi 9-9, Nagasaki 852-8107, Japan Received 16 July 2001; received in revised form 20 September 2001; accepted 8 October 2001 Received by T. Sekiya
Abstract NR-binding SET-domain-containing protein (NSD1) is a mouse nuclear protein containing su(var)3-9, enhancer-of-zeste, trithorax (SET), proline-tryptophan-tryptophan-proline (PWWP) and plant homeodomain protein (PHD)-finger domains (Huang et al., EMBO J. 17 (1998) 3398). This protein also has two other distinct nuclear receptor (NR)-interaction domains, called NID 2L and NID 1L, and acts as both a NR corepressor and a coactivator by interacting directly with the ligand-binding domain of several NRs. Thus, NSD1 is a bifunctional, transcriptional, intermediary factor. We isolated the human homologue (NSD1) of the mouse NSD1 gene (Nsd1), mapped it to human chromosome 5q35, and characterized its genomic structure. NSD1 consists of at least 23 exons. Its cDNA is 8552 bp long, has an 8088 bp open reading frame, contains at least six functional domains (SET, PWWP-I, PWWP-II, PHD-I, PHD-II, and PHD-III) and ten putative nuclear localization signals, and encodes 2696 amino acids. NSD1 shows 86% identity with the mouse Nsd1 at the nucleotide level, and 83% at the amino acid level. NSD1 is expressed in the fetal/adult brain, kidney, skeletal muscle, spleen, and the thymus, and faintly in the lung. Two different transcripts (9.0 and 10.0 kb) were consistently observed in various fetal and adult tissues examined. These findings favor the character of NSD1 as a nucleus-localized, basic transcriptional factor and also a bifunctional transcriptional regulator, such as that of the mouse Nsd1. It remains to be investigated whether mutations of NSD1 lead to a specific phenotype in man. q 2001 Elsevier Science B.V. All rights reserved. Keywords: NSD1; Transcriptional cofactor; SET domain; PWWP domain; PHD finger domain
1. Introduction The transcriptional process is very important in gene expression and is regulated by many kinds of nuclear receptors (NRs) in response to cognate ligands, such as steroids, thyroid hormones, vitamins and retinoids. Generally, NRs Abbreviations: BAC, bacterial artificial chromosome; cDNA, complementary DNA; FISH, fluorescence in situ hybridization; NID, NR-interaction domain; NR, nuclear receptor; NSD1, NR-binding SET-domaincontaining protein; ORF, open reading frame; PCR, polymerase chain reaction; PHD, plant homeodomain protein; PWWP, proline-tryptophan-tryptophan-proline; RT, reverse-transcription; SET, su(var)3-9, enhancer-ofzeste, trithorax; UTR, untranslated region * Corresponding author. Department of Human Genetics, Nagasaki University School of Medicine, Sakamoto 1-12-4, Nagasaki 852-8523, Japan. Tel.: 181-95-849-7120; fax: 181-95-849-7121. E-mail address: [email protected] (N. Matsumoto).
activate transcription in the presence of their ligands and cofactors (Chambon, 1996). Cofactors play important roles in interaction between general transcription factors and transcription regulatory factors. Among these, NSD1 protein was identified through a search for proteins that interact with NRs (Huang et al., 1998). This protein is a unique bifunctional cofactor with two distinct NR-interaction domains, called NID 2L and NID 1L, and acts as both a corepressor and a coactivator by interacting directly with the ligand-binding domain of NRs. NSD1 also contains several conserved functional domains, i.e. SET, PWWP, and PHD. These domains are possibly associated with chromatin architecture and functions as a transcriptional regulator (Jenuwein et al., 1998; Aasland et al., 1995; Stec et al., 2000). Thus, aberration of the human NSD1 gene (NSD1) may have significant effects on human phenotype. Here, we
0378-1119/01/$ - see front matter q 2001 Elsevier Science B.V. All rights reserved. PII: S 0378-111 9(01)00750-8
198
N. Kurotaki et al. / Gene 279 (2001) 197–204
report the isolation of human NSD1, a homologue of the mouse Nsd1, characterization of its genomic structure, expression patterns in various human tissues, and chromosomal localization. 2. Materials and methods 2.1. Cloning of human NSD1-cDNA and mouse Nsd1-cDNA To isolate the human NSD1-cDNA, we first adopted a homology search using the mouse Nsd1-cDNA sequence (GenBank Accession number: AF064553) by BLAST (http://www.ncbi.nlm.nih.gov/BLAST/) against non-redundant (nr), high throughput genomic sequences (htgs), and dbEST. A human cDNA whose partial sequence was homologous to Nsd1 was also obtained from an oligo-capped cDNA library (Suzuki et al., 2000), and completely sequenced. To obtain a complete transcript of NSD1, an exon connection strategy by RT-PCR was adopted. Primers were designed from human cDNA and/or genomic sequences showing similarity to the mouse Nsd1-cDNA, if available. RT-PCR was performed using a human brain cDNA pool (Marathon-Ready cDNA, Clontech, Palo Alto, CA) as a template in a 10 ml mixture containing 0.5 U AmpliTaq Gold (PE Applied Biosystems, Foster city, CA), 200 mM of each dNTP, 1 mM each of forward and reverse primers, 1 £ PCR buffer, and 1.5 mM MgCl2. PCR was cycled 35 times at 948C for 30 s, 508C for 30 s, and 728C for 120 s. RTPCR products were purified by a QIAquick PCR purification kit (Qiagen, Chatsworth, CA) and sequenced using a BigDye Terminator Cycle Sequencing kite and analyzed on an ABI PRISM 377 autosequencer (PE Applied Biosystems). To characterize the 5 0 region of the cDNA, IMAGE clones (purchased from Research Genetics, Hunstville, AL) were collected through BLAST search on dbEST using the most 5 0 50 bp sequence of exon 2. DNA of an IMAGE clone was purified with Qiagen Midi-Prep columns (Chatsworth, CA) and sequenced in the same way as above. As the original mouse Nsd1-cDNA did not contain the PWWP-II domain (Huang et al., 1998) which the human NSD1-cDNA surely had, RT-PCR was performed to confirm the mouse Nsd1-cDNA sequence around the PWWP-II domain using a mouse brain cDNA pool (Marathon-Ready cDNA, Clontech) as a template. Primers FNsd1 (5 0 -AAACATGCTAGAAATACCTG-3 0 ; nt 1972–1991) and RNsd1 (5 0 -CACACTTTTGGTTCTTAGAG-3 0 ; nt 2101–2110) were designed from a mouse Nsd1-cDNA sequence (AF064553). RT-PCR and its sequencing were performed as described above. 2.2. Genomic structure of NSD1 BAC clones covering the putative NSD1 were addressed by BLAST (http://www.ncbi.nlm.nih.gov/BLAST/) and the TIGR BAC end database (http://www.tigr.org/tdb/humgen/
bac_end_search/bac_end_intro.html), and purchased from Research Genetics. Exon–intron boundaries were determined in part by a comparison between cDNA sequences and genomic draft sequences, if available, or by direct sequencing using primers designed from the sequence of cDNA and BAC-DNA as a template. The BAC-DNA was extracted by Qiagen Midi-Prep columns. Sequencing reactions were carried out with the ABI PRISMe BigDye Terminator Cycle Sequencing Kit using 2 mg of the BACDNA as a template and 40 pmol primers. Cycle sequencing was performed for 75 cycles at 968C for 10 s, 508C for 5 s, and at 608C for 4 min. 2.3. Mapping with fluorescence in situ hybridization To determine chromosomal localization of NSD1, fluorescence in situ hybridization (FISH) was performed. BACDNA containing NSD1 exons 6–23 was labeled with SpectrumGreene (Vysis, Grove, IL), and hybridized to Rbanded normal metaphase chromosomes (Takahashi et al., 1990). 2.4. Expression studies Expression of NSD1 was studied by RT-PCR on a human multiple tissue cDNA (MTCe) panel (Clontech, Palo Alto, CA) and Northern blot analysis using human multiple tissue Northern (MTNe) blots (Clontech). To avoid contamination of genomic DNA, primers were designed to connect exon 10 with exon 13: a forward primer 5 0 -AGGTGTAGAACACGATCCCG-3 0 (NSD1-cDNA nt 4551–4570, see Fig. 2) and a reverse primer 5 0 -AGCCGACCTTTAGATGCAGA-3 0 (nt 4954–4973). PCR was carried out 30–35 times at 948C for 30 s, 508C for 30 s, and 728C for 60 s in a 10 ml mixture containing 0.5 U of AmpliTaq Gold (Perkin-Elmer Applied Biosystems), 200 mM of each dNTP, 1 mM each of forward and reverse primers, 1 £ PCR buffer, 1.5 mM MgCl2, and cDNA from the tissue panel. PCR product on fetal brain cDNA was confirmed to be truly derived from NSD1 by sequencing. The same NSD1-specific probe made by RT-PCR for nt 4551–4973 of the cDNA, as well as a control probe (human b-actin cDNA), was radio-labeled, hybridized and washed, according to the manufacturer’s protocol. 3. Results and discussion 3.1. Isolation of the human NSD1-cDNA We isolated and characterized the human NSD1-cDNA. By the homology search using the mouse Nsd1 sequence, we first obtained a human draft sequence (GenBank Accession number: AC008570) which was derived from a BAC clone, CTC-549A4. The exon connection study by RT-PCR revealed that the BAC clone contains at least the first ten NSD1 exons, which later turned out to be exon 2 to exon 11
N. Kurotaki et al. / Gene 279 (2001) 197–204
199
Fig. 1. Schematic presentation of the NSD1 gene. The upper row depicts a genomic structure of NSD1, the middle row shows cDNA clones including two IMAGE clones (2117843 and 3908832) and an oligo-capped cDNA (HRC08570), and the lower row indicates human BAC clones. Introns are not proportional in size. Rectangular boxes indicate exons, and hatched boxes an 8088 bp ORF which starts at exon 2 and ends at exon 23.
(Fig. 1). Sequence analysis showed that exon 2 had an initiation codon corresponding to that of the mouse Nsd1, and a Kozak consensus sequence (CCCAGGatgG) (Kozak, 1996). However, most sequence at the 5 0 -UTR of the mouse Nsd1 never had any similarity to human genomic sequences. Instead, the 3 0 -half of Nsd1 showed high homology to the sequence of a human cDNA clone (HRC08475, GenBank Accession number: AK026066). The AK026066 sequence was connected to exon 11 by RT-PCR (Fig. 1). Two IMAGE clones, 2117843 and 3908832 (GenBank Accession numbers: AI5241800 and BE883626, respectively), were obtained, and sequences of these clones unraveled a 5 0 -upstream 175 bp region from the putative initiation site. A termination codon was observed from 15 to 13 bp upstream from the initiation site. Comparison of the upstream 175 bp sequences and the draft sequence, AC008570, uncovered that the 175 bp 5 0 -UTR was separated by an intronic sequence. Exons 1 and 2 were connected by RT-PCR. One of two BAC clones overlapping with CTC-549A4 was obtained through the TIGR BAC end database. The CTC-HSP2301A4 covered a genomic region corresponding to HRC08570. Direct sequencing with primers from the HRC08570 sequence and from the CTCHSP2301A4 revealed an additional 12 exons (Fig. 1). Thus, the NSD1-cDNA was identified to be 8552 bp long, and have an 8088 bp ORF (GenBank Accession number: AF395588, Fig. 2) and a polyadenylation signal (AATAAA) at the putative 3 0 -UTR. We found at least 23 exons in NSD1 (Fig. 1, Table 1). The exon size ranges from 76 bp (exon 9) to 2560 bp (exon 5). ORF starts at exon 2 and ends at exon 23 (Fig. 1). The GT-AG rule was maintained at all introns. The genomic region from exons 2–11 spans 113 kb according to the draft sequence (AC008570). The genomic size of the remaining region after exon 12 was unknown, because draft sequences were not available. 3.2. Chromosomal localization of NSD1 FISH analysis using the BAC clone, CTC-HSP23014ADNA, as a probe gave twin signals on chromosome 5q35 (Fig. 3). In addition, the draft AC008570, which we identi-
fied to cover exon 1 through exon 11, has been assigned to 5q34-5q35 according to the Map viewer web site (http:// www.ncbi.nlm.nih.gov/cgi-bin/Entrez/hum_srch?chr ¼ hum_chr.inf&query). Thus, it was concluded that NSD1 is located at 5q35, to which the loci of hereditary lymphedema (Evans et al., 1999), mental retardation with congenital anomalies (Zhu et al., 2001), and cerebral gigantism (Maroun et al., 1994) were suggested to be mapped. 3.3. Expression of NSD1 RT-PCR showed the expression of NSD1 strongly in the fetal brain, kidney, skeletal muscle, spleen, and the thymus (Fig. 4A) and faintly in the lung. Northern blot analysis confirmed its expression in the fetal brain and kidney (Fig. 4B), and also in the adult brain, skeletal muscle, spleen, kidney, placenta, lung, and peripheral blood leukocytes. Two different NSD1 transcripts (9.0 and 10.0 kb) were consistently observed in various fetal and adult tissues. 3.4. Comparison of gene structure and functional domains between NSD1, the mouse Nsd1, and other related genes RT-PCR analysis of the mouse brain cDNA pools with primers, FNsd1 and RNsd1, yielded a single product. Sequence analysis revealed that the product contained a novel mouse Nsd1-cDNA sequence (GenBank Accession number: AF419220). Thus, the ORFs of human NSD1 and mouse Nsd1 are 8088 and 8073 bp long, respectively, and their nucleotide sequences have 86% identity. Likewise, the two genes encode 2696 and 2691 amino acids, respectively, between which there is 83% identity. In the previous study of mouse Nsd1 (Huang et al., 1998), two transcripts of 8589 and 8277 bp nucleotides with or without a 312 bp segment were recognized. The segment showed 99% identity at the nucleotide level to a 309 bp sequence newly isolated by us, but only 84% identity at the amino acid level. Two such transcripts were also observed in human NSD1 by Northern blot analysis, probably due to alternative splicing of exons 3 and 4 as seen in Nsd1, although only a longer transcript was detected in the human and mouse fetal brains by RT-PCR. Eight functional domains were detected in NSD1 by the
200
N. Kurotaki et al. / Gene 279 (2001) 197–204
N. Kurotaki et al. / Gene 279 (2001) 197–204 201
Fig. 2. The entire 8552 bp sequence of NSD1-cDNA and deduced amino acid sequences. An ORF of 8088 bp is shown in upper case letters, and the 5 0 -/3 0 -UTR is in lower case letters. Termination codon and polyadenylation signals are underlined. Thick underlines indicate putative nuclear localization signals, and open boxes the PWWP and SET domains. PHD finger domains are shown by dotted boxes, and NID 2L and NID 1L by dotted lines.
202
N. Kurotaki et al. / Gene 279 (2001) 197–204
Table 1 Exon–intron boundaries of NSD1 a Exon
Size (bp)
5 0 splice donor
Intron
Size (bp)
3 0 splice acceptor
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
ND 944 135 173 2560 125 271 110 76 119 144 124 201 180 157 206 113 270 117 142 107 205 ND
GGCACGgtaact CCATTTgtaagc TGAAAGgtaata CATAAGgtagga AACCAGgtaagg CACAAGgtatgt CGCCAGgtaagg AAAAAGgtatgt GTGGAGgtgagt CTCAGAGgtatt TGTCAGgtagag GCACAGgtaaag CTAAAGgtatgg CAGAAGgtaaga ATACAGgtaagc AAAAAGgtaact ATAAAGgtgagg AAAAAGgttaga GACAAAgtaagt AAGCAGgtaaga CCAAAGgtacca CAGCAGgttggt
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
1133 55,853 12,100 5343 23,626 2292 1248 4341 2406 1384 ND ND ND ND ND ND ND ND ND ND ND ND
ttccagGTTGAT ctaaagTGTCAA acctagTTTCCA tttaagGTTCCT ttttagCTGTGC tcttagGTAAGT tttcagGAAATT ttatagTGCTAT tcatagGCACTA tgcaagGGAGAA taacagAATTGT tctcagGAATCC ctttagGTCGGC ccacagGAGGCA tcccagGTGGTG tcatagCTCTTA ttccagGTAAAC ttctagGGTGAA ttctagGACCGA atgcagGCACTG ctgtagAATCAA tttcagGGAAAT
a
ND, not determined; nucleotides in upper and lower case letters represent exonic and intronic sequences, respectively.
interPro program (http:/www.ebi.ac.uk/interpro/). They include two NR-interaction domains (NID 2L and NID 1L) and six other domains (PWWP (proline-tryptophan-tryptophan-proline)-II, PHD (plant homeodomain protein)-I, PHD-II, PWWP-I, SET (su(var)3-9, enhancer-of-zeste, trithorax), and PHD-III domains) (Figs. 2 and 5). Furthermore, ten putative nuclear localization signals were identified as were in the mouse Nsd1 (Fig. 2). The PWWP-II domain was not recognized in original mouse Nsd1 (AF064553), but was identified in the new mouse Nsd1 identified by us (GenBank Accession number: AF419220). Thus, all domains were well conserved between NSD1 and Nsd1. Amino acid sequences of both the PWWP-I and PHDII domains perfectly correspond between the two genes,
Fig. 3. FISH analysis using a NSD1 BAC clone (CTC-HSP23014A) on normal R-banded metaphase chromosomes. An arrowhead shows signals at 5q35.
Fig. 4. (A) RT-PCR products for NSD1 and for glyceraldehyde-3-phosphate dehydrogenase gene (G3PDH) as a control were electrophoresed in 4% polyacrylamide gel. Human fetal multiple tissue cDNA (MTCe, Clontech) panels were used as a template. PCR was cycled 35 times for NSD1 and 30 times for G3PDH. Control cDNA attached to the panel (0.2 ng/ml) and distilled water were used as positive and negative controls, respectively. (B) Northern blot analysis of NSD1 expression in human adult and fetal tissues. Each lane contains approximately 2 mg of polyA 1 RNA (Clontech). Two different NSD1 transcripts with sizes of 10 and 9 kb were observed (arrowhead). The human b-actin gene was used as a control.
N. Kurotaki et al. / Gene 279 (2001) 197–204
203
Fig. 5. Amino acid alignment of eight domains of NSD1, mouse Nsd1 modified by us, NSD2, and NSD3 proteins. Shaded boxes show identical amino acids among proteins listed. Pfam (http://www.sanger.ac.uk/Software/Pfam/) Accession numbers are PF00855 for PWWP, PF00628 for PHD-finger, and PF00856 for SET domains. NID domains were based on a previous publication (Huang et al., 1998).
while those of SET, PHD-I, PHD-III, PWWP-II, NID 2L and NID 1L have 99, 97, 97, 95, 88, and 83% identity, respectively. It has been suggested that the SET domain functions as a regulator of transcriptional process, modulates the structure of chromatin, and plays a roll in cell growth and differentiation (Jenuwein et al., 1998). The PHD finger domains may also involve chromatin-mediated transcriptional regulation (Aasland et al., 1995). There are two other human genes, NSD2 and NSD3, with homology to NSD1, and all probably belong to a gene family (Stec et al., 1998; Chesi et al., 1998; Angrand et al., 2001). NSD2
was originally isolated from the Wolf–Hirschhorn syndrome critical region (Stec et al., 1998) and from the 4p-breakpoint of multiple myeloma with a (4;14) translocation, and showed 75% homology to NSD1 at the nucleotide level. NSD3 showed 70% identity to NSD1 and is expressed in several tumor cell lines (Angrand et al., 2001). NSD1 and NSD2 share all but two domains (NID 2L and NID 1L) with NSD1, according to the Pfam database (Bateman et al., 2000) (http://www.sanger.ac.uk/Software/Pfam/) (Fig. 5). These findings suggest that NSD1 is a basic transcriptional factor protein that is localized in the nucleus, and acts as a
204
N. Kurotaki et al. / Gene 279 (2001) 197–204
bifunctional transregulator as does the mouse Nsd1. While this paper was in a process of revision, NSD1 was newly isolated as a fusion transcript with the nucleoporin gene (NUP98) in de novo childhood acute myeloid leukemia (Jaju et al., 2001). It remains to be seen whether mutations of NSD1 alone result in a distinct human phenotype. Acknowledgements We express our gratitude to Ms Kazumi Miyazaki, Naoko Takaki, Yasuko Noguchi, and Naoko Yanai for their excellent technical assistance on this work. This work was supported in part by CREST, JST, Japan, and by Grantsin-Aid (Nos. 08307019, 08283101, and 11470507) from the Ministry of Education, Science, Sports and Culture of Japan. References Aasland, R., Gibson, T., Stewart, A., 1995. PHD finger: implications for chromatin-mediated transcriptional regulation. Trends Biochem. Sci. 20, 56–59. Angrand, P.-O., Apiou, F., Stewart, A.F., Dutrillaux, B., Losson, R., Chambon, P., 2001. NSD3, a new set domain-containing gene, maps to 8q12 and is amplified in human breast cancer cell lines. Genomics 74, 79–88. Bateman, A., Birney, E., Durbin, R., Eddy, R.S., Howe, L.K., Sonnhammer, L.L.E., 2000. The Pfam protein families database. Nucleic Acids Res. 28, 263–266. Chambon, P., 1996. A decade of molecular biology of retinoic acid receptors. Fed. Am. Soc. Exp. Biol. J. 10, 940–954. Chesi, M., Nardini, E., Lim, R., Smith, K., Kuehl, W., Bergsagel, L., 1998. The t(4;14) translocation in myeloma dysregulates both FGFR3 and a novel gene, MMSET, resulting in IgH/MMSET hybrid transcripts. Blood 92, 3025–3034. Evans, A., Brice, G., Sotirova, V., Mortimer, P., Benison, J., Burnand, K., Rosbotham, J., Child, A., Sarfafazi, M., 1999. Mapping of primary
congenital lymphedema to the 5q35.3 region. Am. J. Hum. Genet. 64, 547–555. Huang, N., von Baur, E., Garnier, J.-M., Lerouge, T., Vonesh, J.-L., Lutz, Y., Chambon, P., Losseon, R., 1998. The distinct nuclear receptor interaction domains in NSD1, a novel SET protein that exhibits characteristics of both corepressor and co activators. EMBO J. 17, 3398–3412. Jaju, R.J., Fidler, C., Haas, O.A., Strickson, A.J., Watkins, F., Clark, K., Cross, N.C.P., Cheng, J.-F., Aplan, P.D., Kearney, L., Boultwood, J., Wainscoat, J.S., 2001. A novel gene, NSD1, is fused to NUP98 in the t(5;11)(q35;p15.5) in de novo childhood acute myeloid leukemia. Blood 98, 1264–1267. Jenuwein, T., Laible, G., Dorn, R., Reuter, G., 1998. SET domain proteins modulate chromatin domains in eu- and heterochromatin. Cell. Mol. Life Sci. 54, 80–93. Kozak, M., 1996. Interpreting cDNA sequences: some insights from studies on translation. Mamm. Genome 7, 563–574. Maroun, C., Schmerler, S., Huchen, R., 1994. Child with Sotos phenotype and a 5:15 translocation. Am. J. Med. Genet. 50, 291–293. Stec, I., Wright, T., van Ommen, G.-J.B., de Boer, P.A.G., van Haeringen, A., Moorman, A.F.M., Altherr, M.R., den Dunnen, J.T., 1998. WHISC1, a 90 kb SET domain-containing gene, expressed in early development and homologous to a Drosophila dysmorphy gene maps in the WolfHirschhorn syndrome critical region and is fused to IgH in t(4;14) multiple myeloma. Hum. Mol. Genet. 7, 1071–1082. Stec, I., Nagl, S.B., van Ommen, G.-J.B., den Dunnen, J., 2000. Two PWWP domain: a potential protein-protein interaction domain in nuclear proteins influencing differentiation? FEBS Lett. 473, 1–5. Suzuki, Y., Ishihara, D., Sasaki, M., Nakagawa, H., Hata, H., Tsunoda, T., Watanabe, M., Komatsu, T., Ota, T., Isogai, T., Suyama, A., Sugano, S., 2000. Statistical analysis of the 5 0 untranslated region of human mRNA using “oligo-capped” cDNA libraries. Genomics 64, 286–297. Takahashi, E., Hori, T., O’Connell, P., Leppert, M., White, R., 1990. Rbanding and nonisotopic in situ hybridization: precise localization of human type II collagen gene (COL2A1). Hum. Genet. 86, 14–16. Zhu, G., Gillessen-Kaesbach, G., Wirth, J., Passarge, E., Bartsch, O., 2001. Girl with phenotypic abnormalities and a de novo, apparently balanced translocation 46, XX, t(5;10)(q35.2q11.2). Am. J. Med. Genet. 98, 317– 319.
Molecular characterization of NSD1, a human homologue of the mouse Nsd1 gene Naohiro Kurotaki a,b,c, Naoki Harada a,b,e, Koh-ichiro Yoshiura a,b, Sumio Sugano d, Norio Niikawa a,b, Naomichi Matsumoto a,b,* a
Department of Human Genetics, Nagasaki University School of Medicine, Sakamoto 1-12-4, Nagasaki 852-8523, Japan b CREST, Japan Science and Technology Corporation, Honcho 4-1-8, Kawaguchi 332-0012, Japan c Department of Neuropsychiatry, Nagasaki University School of Medicine, Sakamoto 1-12-4, Nagasaki 852-8523, Japan d Department of Virology, Institute of Medical Science, University of Tokyo, Shirokanedai 4-6-1, Minato-ku, Tokyo 108-8639, Japan e Kyusyu Medical Science Nagasaki Laboratory, Hamaguchi-machi 9-9, Nagasaki 852-8107, Japan Received 16 July 2001; received in revised form 20 September 2001; accepted 8 October 2001 Received by T. Sekiya
Abstract NR-binding SET-domain-containing protein (NSD1) is a mouse nuclear protein containing su(var)3-9, enhancer-of-zeste, trithorax (SET), proline-tryptophan-tryptophan-proline (PWWP) and plant homeodomain protein (PHD)-finger domains (Huang et al., EMBO J. 17 (1998) 3398). This protein also has two other distinct nuclear receptor (NR)-interaction domains, called NID 2L and NID 1L, and acts as both a NR corepressor and a coactivator by interacting directly with the ligand-binding domain of several NRs. Thus, NSD1 is a bifunctional, transcriptional, intermediary factor. We isolated the human homologue (NSD1) of the mouse NSD1 gene (Nsd1), mapped it to human chromosome 5q35, and characterized its genomic structure. NSD1 consists of at least 23 exons. Its cDNA is 8552 bp long, has an 8088 bp open reading frame, contains at least six functional domains (SET, PWWP-I, PWWP-II, PHD-I, PHD-II, and PHD-III) and ten putative nuclear localization signals, and encodes 2696 amino acids. NSD1 shows 86% identity with the mouse Nsd1 at the nucleotide level, and 83% at the amino acid level. NSD1 is expressed in the fetal/adult brain, kidney, skeletal muscle, spleen, and the thymus, and faintly in the lung. Two different transcripts (9.0 and 10.0 kb) were consistently observed in various fetal and adult tissues examined. These findings favor the character of NSD1 as a nucleus-localized, basic transcriptional factor and also a bifunctional transcriptional regulator, such as that of the mouse Nsd1. It remains to be investigated whether mutations of NSD1 lead to a specific phenotype in man. q 2001 Elsevier Science B.V. All rights reserved. Keywords: NSD1; Transcriptional cofactor; SET domain; PWWP domain; PHD finger domain
1. Introduction The transcriptional process is very important in gene expression and is regulated by many kinds of nuclear receptors (NRs) in response to cognate ligands, such as steroids, thyroid hormones, vitamins and retinoids. Generally, NRs Abbreviations: BAC, bacterial artificial chromosome; cDNA, complementary DNA; FISH, fluorescence in situ hybridization; NID, NR-interaction domain; NR, nuclear receptor; NSD1, NR-binding SET-domaincontaining protein; ORF, open reading frame; PCR, polymerase chain reaction; PHD, plant homeodomain protein; PWWP, proline-tryptophan-tryptophan-proline; RT, reverse-transcription; SET, su(var)3-9, enhancer-ofzeste, trithorax; UTR, untranslated region * Corresponding author. Department of Human Genetics, Nagasaki University School of Medicine, Sakamoto 1-12-4, Nagasaki 852-8523, Japan. Tel.: 181-95-849-7120; fax: 181-95-849-7121. E-mail address: [email protected] (N. Matsumoto).
activate transcription in the presence of their ligands and cofactors (Chambon, 1996). Cofactors play important roles in interaction between general transcription factors and transcription regulatory factors. Among these, NSD1 protein was identified through a search for proteins that interact with NRs (Huang et al., 1998). This protein is a unique bifunctional cofactor with two distinct NR-interaction domains, called NID 2L and NID 1L, and acts as both a corepressor and a coactivator by interacting directly with the ligand-binding domain of NRs. NSD1 also contains several conserved functional domains, i.e. SET, PWWP, and PHD. These domains are possibly associated with chromatin architecture and functions as a transcriptional regulator (Jenuwein et al., 1998; Aasland et al., 1995; Stec et al., 2000). Thus, aberration of the human NSD1 gene (NSD1) may have significant effects on human phenotype. Here, we
0378-1119/01/$ - see front matter q 2001 Elsevier Science B.V. All rights reserved. PII: S 0378-111 9(01)00750-8
198
N. Kurotaki et al. / Gene 279 (2001) 197–204
report the isolation of human NSD1, a homologue of the mouse Nsd1, characterization of its genomic structure, expression patterns in various human tissues, and chromosomal localization. 2. Materials and methods 2.1. Cloning of human NSD1-cDNA and mouse Nsd1-cDNA To isolate the human NSD1-cDNA, we first adopted a homology search using the mouse Nsd1-cDNA sequence (GenBank Accession number: AF064553) by BLAST (http://www.ncbi.nlm.nih.gov/BLAST/) against non-redundant (nr), high throughput genomic sequences (htgs), and dbEST. A human cDNA whose partial sequence was homologous to Nsd1 was also obtained from an oligo-capped cDNA library (Suzuki et al., 2000), and completely sequenced. To obtain a complete transcript of NSD1, an exon connection strategy by RT-PCR was adopted. Primers were designed from human cDNA and/or genomic sequences showing similarity to the mouse Nsd1-cDNA, if available. RT-PCR was performed using a human brain cDNA pool (Marathon-Ready cDNA, Clontech, Palo Alto, CA) as a template in a 10 ml mixture containing 0.5 U AmpliTaq Gold (PE Applied Biosystems, Foster city, CA), 200 mM of each dNTP, 1 mM each of forward and reverse primers, 1 £ PCR buffer, and 1.5 mM MgCl2. PCR was cycled 35 times at 948C for 30 s, 508C for 30 s, and 728C for 120 s. RTPCR products were purified by a QIAquick PCR purification kit (Qiagen, Chatsworth, CA) and sequenced using a BigDye Terminator Cycle Sequencing kite and analyzed on an ABI PRISM 377 autosequencer (PE Applied Biosystems). To characterize the 5 0 region of the cDNA, IMAGE clones (purchased from Research Genetics, Hunstville, AL) were collected through BLAST search on dbEST using the most 5 0 50 bp sequence of exon 2. DNA of an IMAGE clone was purified with Qiagen Midi-Prep columns (Chatsworth, CA) and sequenced in the same way as above. As the original mouse Nsd1-cDNA did not contain the PWWP-II domain (Huang et al., 1998) which the human NSD1-cDNA surely had, RT-PCR was performed to confirm the mouse Nsd1-cDNA sequence around the PWWP-II domain using a mouse brain cDNA pool (Marathon-Ready cDNA, Clontech) as a template. Primers FNsd1 (5 0 -AAACATGCTAGAAATACCTG-3 0 ; nt 1972–1991) and RNsd1 (5 0 -CACACTTTTGGTTCTTAGAG-3 0 ; nt 2101–2110) were designed from a mouse Nsd1-cDNA sequence (AF064553). RT-PCR and its sequencing were performed as described above. 2.2. Genomic structure of NSD1 BAC clones covering the putative NSD1 were addressed by BLAST (http://www.ncbi.nlm.nih.gov/BLAST/) and the TIGR BAC end database (http://www.tigr.org/tdb/humgen/
bac_end_search/bac_end_intro.html), and purchased from Research Genetics. Exon–intron boundaries were determined in part by a comparison between cDNA sequences and genomic draft sequences, if available, or by direct sequencing using primers designed from the sequence of cDNA and BAC-DNA as a template. The BAC-DNA was extracted by Qiagen Midi-Prep columns. Sequencing reactions were carried out with the ABI PRISMe BigDye Terminator Cycle Sequencing Kit using 2 mg of the BACDNA as a template and 40 pmol primers. Cycle sequencing was performed for 75 cycles at 968C for 10 s, 508C for 5 s, and at 608C for 4 min. 2.3. Mapping with fluorescence in situ hybridization To determine chromosomal localization of NSD1, fluorescence in situ hybridization (FISH) was performed. BACDNA containing NSD1 exons 6–23 was labeled with SpectrumGreene (Vysis, Grove, IL), and hybridized to Rbanded normal metaphase chromosomes (Takahashi et al., 1990). 2.4. Expression studies Expression of NSD1 was studied by RT-PCR on a human multiple tissue cDNA (MTCe) panel (Clontech, Palo Alto, CA) and Northern blot analysis using human multiple tissue Northern (MTNe) blots (Clontech). To avoid contamination of genomic DNA, primers were designed to connect exon 10 with exon 13: a forward primer 5 0 -AGGTGTAGAACACGATCCCG-3 0 (NSD1-cDNA nt 4551–4570, see Fig. 2) and a reverse primer 5 0 -AGCCGACCTTTAGATGCAGA-3 0 (nt 4954–4973). PCR was carried out 30–35 times at 948C for 30 s, 508C for 30 s, and 728C for 60 s in a 10 ml mixture containing 0.5 U of AmpliTaq Gold (Perkin-Elmer Applied Biosystems), 200 mM of each dNTP, 1 mM each of forward and reverse primers, 1 £ PCR buffer, 1.5 mM MgCl2, and cDNA from the tissue panel. PCR product on fetal brain cDNA was confirmed to be truly derived from NSD1 by sequencing. The same NSD1-specific probe made by RT-PCR for nt 4551–4973 of the cDNA, as well as a control probe (human b-actin cDNA), was radio-labeled, hybridized and washed, according to the manufacturer’s protocol. 3. Results and discussion 3.1. Isolation of the human NSD1-cDNA We isolated and characterized the human NSD1-cDNA. By the homology search using the mouse Nsd1 sequence, we first obtained a human draft sequence (GenBank Accession number: AC008570) which was derived from a BAC clone, CTC-549A4. The exon connection study by RT-PCR revealed that the BAC clone contains at least the first ten NSD1 exons, which later turned out to be exon 2 to exon 11
N. Kurotaki et al. / Gene 279 (2001) 197–204
199
Fig. 1. Schematic presentation of the NSD1 gene. The upper row depicts a genomic structure of NSD1, the middle row shows cDNA clones including two IMAGE clones (2117843 and 3908832) and an oligo-capped cDNA (HRC08570), and the lower row indicates human BAC clones. Introns are not proportional in size. Rectangular boxes indicate exons, and hatched boxes an 8088 bp ORF which starts at exon 2 and ends at exon 23.
(Fig. 1). Sequence analysis showed that exon 2 had an initiation codon corresponding to that of the mouse Nsd1, and a Kozak consensus sequence (CCCAGGatgG) (Kozak, 1996). However, most sequence at the 5 0 -UTR of the mouse Nsd1 never had any similarity to human genomic sequences. Instead, the 3 0 -half of Nsd1 showed high homology to the sequence of a human cDNA clone (HRC08475, GenBank Accession number: AK026066). The AK026066 sequence was connected to exon 11 by RT-PCR (Fig. 1). Two IMAGE clones, 2117843 and 3908832 (GenBank Accession numbers: AI5241800 and BE883626, respectively), were obtained, and sequences of these clones unraveled a 5 0 -upstream 175 bp region from the putative initiation site. A termination codon was observed from 15 to 13 bp upstream from the initiation site. Comparison of the upstream 175 bp sequences and the draft sequence, AC008570, uncovered that the 175 bp 5 0 -UTR was separated by an intronic sequence. Exons 1 and 2 were connected by RT-PCR. One of two BAC clones overlapping with CTC-549A4 was obtained through the TIGR BAC end database. The CTC-HSP2301A4 covered a genomic region corresponding to HRC08570. Direct sequencing with primers from the HRC08570 sequence and from the CTCHSP2301A4 revealed an additional 12 exons (Fig. 1). Thus, the NSD1-cDNA was identified to be 8552 bp long, and have an 8088 bp ORF (GenBank Accession number: AF395588, Fig. 2) and a polyadenylation signal (AATAAA) at the putative 3 0 -UTR. We found at least 23 exons in NSD1 (Fig. 1, Table 1). The exon size ranges from 76 bp (exon 9) to 2560 bp (exon 5). ORF starts at exon 2 and ends at exon 23 (Fig. 1). The GT-AG rule was maintained at all introns. The genomic region from exons 2–11 spans 113 kb according to the draft sequence (AC008570). The genomic size of the remaining region after exon 12 was unknown, because draft sequences were not available. 3.2. Chromosomal localization of NSD1 FISH analysis using the BAC clone, CTC-HSP23014ADNA, as a probe gave twin signals on chromosome 5q35 (Fig. 3). In addition, the draft AC008570, which we identi-
fied to cover exon 1 through exon 11, has been assigned to 5q34-5q35 according to the Map viewer web site (http:// www.ncbi.nlm.nih.gov/cgi-bin/Entrez/hum_srch?chr ¼ hum_chr.inf&query). Thus, it was concluded that NSD1 is located at 5q35, to which the loci of hereditary lymphedema (Evans et al., 1999), mental retardation with congenital anomalies (Zhu et al., 2001), and cerebral gigantism (Maroun et al., 1994) were suggested to be mapped. 3.3. Expression of NSD1 RT-PCR showed the expression of NSD1 strongly in the fetal brain, kidney, skeletal muscle, spleen, and the thymus (Fig. 4A) and faintly in the lung. Northern blot analysis confirmed its expression in the fetal brain and kidney (Fig. 4B), and also in the adult brain, skeletal muscle, spleen, kidney, placenta, lung, and peripheral blood leukocytes. Two different NSD1 transcripts (9.0 and 10.0 kb) were consistently observed in various fetal and adult tissues. 3.4. Comparison of gene structure and functional domains between NSD1, the mouse Nsd1, and other related genes RT-PCR analysis of the mouse brain cDNA pools with primers, FNsd1 and RNsd1, yielded a single product. Sequence analysis revealed that the product contained a novel mouse Nsd1-cDNA sequence (GenBank Accession number: AF419220). Thus, the ORFs of human NSD1 and mouse Nsd1 are 8088 and 8073 bp long, respectively, and their nucleotide sequences have 86% identity. Likewise, the two genes encode 2696 and 2691 amino acids, respectively, between which there is 83% identity. In the previous study of mouse Nsd1 (Huang et al., 1998), two transcripts of 8589 and 8277 bp nucleotides with or without a 312 bp segment were recognized. The segment showed 99% identity at the nucleotide level to a 309 bp sequence newly isolated by us, but only 84% identity at the amino acid level. Two such transcripts were also observed in human NSD1 by Northern blot analysis, probably due to alternative splicing of exons 3 and 4 as seen in Nsd1, although only a longer transcript was detected in the human and mouse fetal brains by RT-PCR. Eight functional domains were detected in NSD1 by the
200
N. Kurotaki et al. / Gene 279 (2001) 197–204
N. Kurotaki et al. / Gene 279 (2001) 197–204 201
Fig. 2. The entire 8552 bp sequence of NSD1-cDNA and deduced amino acid sequences. An ORF of 8088 bp is shown in upper case letters, and the 5 0 -/3 0 -UTR is in lower case letters. Termination codon and polyadenylation signals are underlined. Thick underlines indicate putative nuclear localization signals, and open boxes the PWWP and SET domains. PHD finger domains are shown by dotted boxes, and NID 2L and NID 1L by dotted lines.
202
N. Kurotaki et al. / Gene 279 (2001) 197–204
Table 1 Exon–intron boundaries of NSD1 a Exon
Size (bp)
5 0 splice donor
Intron
Size (bp)
3 0 splice acceptor
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
ND 944 135 173 2560 125 271 110 76 119 144 124 201 180 157 206 113 270 117 142 107 205 ND
GGCACGgtaact CCATTTgtaagc TGAAAGgtaata CATAAGgtagga AACCAGgtaagg CACAAGgtatgt CGCCAGgtaagg AAAAAGgtatgt GTGGAGgtgagt CTCAGAGgtatt TGTCAGgtagag GCACAGgtaaag CTAAAGgtatgg CAGAAGgtaaga ATACAGgtaagc AAAAAGgtaact ATAAAGgtgagg AAAAAGgttaga GACAAAgtaagt AAGCAGgtaaga CCAAAGgtacca CAGCAGgttggt
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
1133 55,853 12,100 5343 23,626 2292 1248 4341 2406 1384 ND ND ND ND ND ND ND ND ND ND ND ND
ttccagGTTGAT ctaaagTGTCAA acctagTTTCCA tttaagGTTCCT ttttagCTGTGC tcttagGTAAGT tttcagGAAATT ttatagTGCTAT tcatagGCACTA tgcaagGGAGAA taacagAATTGT tctcagGAATCC ctttagGTCGGC ccacagGAGGCA tcccagGTGGTG tcatagCTCTTA ttccagGTAAAC ttctagGGTGAA ttctagGACCGA atgcagGCACTG ctgtagAATCAA tttcagGGAAAT
a
ND, not determined; nucleotides in upper and lower case letters represent exonic and intronic sequences, respectively.
interPro program (http:/www.ebi.ac.uk/interpro/). They include two NR-interaction domains (NID 2L and NID 1L) and six other domains (PWWP (proline-tryptophan-tryptophan-proline)-II, PHD (plant homeodomain protein)-I, PHD-II, PWWP-I, SET (su(var)3-9, enhancer-of-zeste, trithorax), and PHD-III domains) (Figs. 2 and 5). Furthermore, ten putative nuclear localization signals were identified as were in the mouse Nsd1 (Fig. 2). The PWWP-II domain was not recognized in original mouse Nsd1 (AF064553), but was identified in the new mouse Nsd1 identified by us (GenBank Accession number: AF419220). Thus, all domains were well conserved between NSD1 and Nsd1. Amino acid sequences of both the PWWP-I and PHDII domains perfectly correspond between the two genes,
Fig. 3. FISH analysis using a NSD1 BAC clone (CTC-HSP23014A) on normal R-banded metaphase chromosomes. An arrowhead shows signals at 5q35.
Fig. 4. (A) RT-PCR products for NSD1 and for glyceraldehyde-3-phosphate dehydrogenase gene (G3PDH) as a control were electrophoresed in 4% polyacrylamide gel. Human fetal multiple tissue cDNA (MTCe, Clontech) panels were used as a template. PCR was cycled 35 times for NSD1 and 30 times for G3PDH. Control cDNA attached to the panel (0.2 ng/ml) and distilled water were used as positive and negative controls, respectively. (B) Northern blot analysis of NSD1 expression in human adult and fetal tissues. Each lane contains approximately 2 mg of polyA 1 RNA (Clontech). Two different NSD1 transcripts with sizes of 10 and 9 kb were observed (arrowhead). The human b-actin gene was used as a control.
N. Kurotaki et al. / Gene 279 (2001) 197–204
203
Fig. 5. Amino acid alignment of eight domains of NSD1, mouse Nsd1 modified by us, NSD2, and NSD3 proteins. Shaded boxes show identical amino acids among proteins listed. Pfam (http://www.sanger.ac.uk/Software/Pfam/) Accession numbers are PF00855 for PWWP, PF00628 for PHD-finger, and PF00856 for SET domains. NID domains were based on a previous publication (Huang et al., 1998).
while those of SET, PHD-I, PHD-III, PWWP-II, NID 2L and NID 1L have 99, 97, 97, 95, 88, and 83% identity, respectively. It has been suggested that the SET domain functions as a regulator of transcriptional process, modulates the structure of chromatin, and plays a roll in cell growth and differentiation (Jenuwein et al., 1998). The PHD finger domains may also involve chromatin-mediated transcriptional regulation (Aasland et al., 1995). There are two other human genes, NSD2 and NSD3, with homology to NSD1, and all probably belong to a gene family (Stec et al., 1998; Chesi et al., 1998; Angrand et al., 2001). NSD2
was originally isolated from the Wolf–Hirschhorn syndrome critical region (Stec et al., 1998) and from the 4p-breakpoint of multiple myeloma with a (4;14) translocation, and showed 75% homology to NSD1 at the nucleotide level. NSD3 showed 70% identity to NSD1 and is expressed in several tumor cell lines (Angrand et al., 2001). NSD1 and NSD2 share all but two domains (NID 2L and NID 1L) with NSD1, according to the Pfam database (Bateman et al., 2000) (http://www.sanger.ac.uk/Software/Pfam/) (Fig. 5). These findings suggest that NSD1 is a basic transcriptional factor protein that is localized in the nucleus, and acts as a
204
N. Kurotaki et al. / Gene 279 (2001) 197–204
bifunctional transregulator as does the mouse Nsd1. While this paper was in a process of revision, NSD1 was newly isolated as a fusion transcript with the nucleoporin gene (NUP98) in de novo childhood acute myeloid leukemia (Jaju et al., 2001). It remains to be seen whether mutations of NSD1 alone result in a distinct human phenotype. Acknowledgements We express our gratitude to Ms Kazumi Miyazaki, Naoko Takaki, Yasuko Noguchi, and Naoko Yanai for their excellent technical assistance on this work. This work was supported in part by CREST, JST, Japan, and by Grantsin-Aid (Nos. 08307019, 08283101, and 11470507) from the Ministry of Education, Science, Sports and Culture of Japan. References Aasland, R., Gibson, T., Stewart, A., 1995. PHD finger: implications for chromatin-mediated transcriptional regulation. Trends Biochem. Sci. 20, 56–59. Angrand, P.-O., Apiou, F., Stewart, A.F., Dutrillaux, B., Losson, R., Chambon, P., 2001. NSD3, a new set domain-containing gene, maps to 8q12 and is amplified in human breast cancer cell lines. Genomics 74, 79–88. Bateman, A., Birney, E., Durbin, R., Eddy, R.S., Howe, L.K., Sonnhammer, L.L.E., 2000. The Pfam protein families database. Nucleic Acids Res. 28, 263–266. Chambon, P., 1996. A decade of molecular biology of retinoic acid receptors. Fed. Am. Soc. Exp. Biol. J. 10, 940–954. Chesi, M., Nardini, E., Lim, R., Smith, K., Kuehl, W., Bergsagel, L., 1998. The t(4;14) translocation in myeloma dysregulates both FGFR3 and a novel gene, MMSET, resulting in IgH/MMSET hybrid transcripts. Blood 92, 3025–3034. Evans, A., Brice, G., Sotirova, V., Mortimer, P., Benison, J., Burnand, K., Rosbotham, J., Child, A., Sarfafazi, M., 1999. Mapping of primary
congenital lymphedema to the 5q35.3 region. Am. J. Hum. Genet. 64, 547–555. Huang, N., von Baur, E., Garnier, J.-M., Lerouge, T., Vonesh, J.-L., Lutz, Y., Chambon, P., Losseon, R., 1998. The distinct nuclear receptor interaction domains in NSD1, a novel SET protein that exhibits characteristics of both corepressor and co activators. EMBO J. 17, 3398–3412. Jaju, R.J., Fidler, C., Haas, O.A., Strickson, A.J., Watkins, F., Clark, K., Cross, N.C.P., Cheng, J.-F., Aplan, P.D., Kearney, L., Boultwood, J., Wainscoat, J.S., 2001. A novel gene, NSD1, is fused to NUP98 in the t(5;11)(q35;p15.5) in de novo childhood acute myeloid leukemia. Blood 98, 1264–1267. Jenuwein, T., Laible, G., Dorn, R., Reuter, G., 1998. SET domain proteins modulate chromatin domains in eu- and heterochromatin. Cell. Mol. Life Sci. 54, 80–93. Kozak, M., 1996. Interpreting cDNA sequences: some insights from studies on translation. Mamm. Genome 7, 563–574. Maroun, C., Schmerler, S., Huchen, R., 1994. Child with Sotos phenotype and a 5:15 translocation. Am. J. Med. Genet. 50, 291–293. Stec, I., Wright, T., van Ommen, G.-J.B., de Boer, P.A.G., van Haeringen, A., Moorman, A.F.M., Altherr, M.R., den Dunnen, J.T., 1998. WHISC1, a 90 kb SET domain-containing gene, expressed in early development and homologous to a Drosophila dysmorphy gene maps in the WolfHirschhorn syndrome critical region and is fused to IgH in t(4;14) multiple myeloma. Hum. Mol. Genet. 7, 1071–1082. Stec, I., Nagl, S.B., van Ommen, G.-J.B., den Dunnen, J., 2000. Two PWWP domain: a potential protein-protein interaction domain in nuclear proteins influencing differentiation? FEBS Lett. 473, 1–5. Suzuki, Y., Ishihara, D., Sasaki, M., Nakagawa, H., Hata, H., Tsunoda, T., Watanabe, M., Komatsu, T., Ota, T., Isogai, T., Suyama, A., Sugano, S., 2000. Statistical analysis of the 5 0 untranslated region of human mRNA using “oligo-capped” cDNA libraries. Genomics 64, 286–297. Takahashi, E., Hori, T., O’Connell, P., Leppert, M., White, R., 1990. Rbanding and nonisotopic in situ hybridization: precise localization of human type II collagen gene (COL2A1). Hum. Genet. 86, 14–16. Zhu, G., Gillessen-Kaesbach, G., Wirth, J., Passarge, E., Bartsch, O., 2001. Girl with phenotypic abnormalities and a de novo, apparently balanced translocation 46, XX, t(5;10)(q35.2q11.2). Am. J. Med. Genet. 98, 317– 319.