SHORT COMMUNICATION

Isolation, cDNA, and Genomic Structure of a Conserved Gene (NOF)at Chromosome 11q13 Next to FAU and Oriented

in the Opposite Transcriptional Orientation

KOEN KAS,*1,2 VANESSA LEMAHIEU,†,2,3 EVA MEYEN,*WIM J. M. VAN DE VEN,*AND JOZEF MERREGAERT†

*Laboratory for Molecular Oncology, Center for Human Genetics, University of Leuven & Flanders Interuniversity Institute forBiotechnology, Herestraat 49, B-3000 Leuven, Belgium; and †Laboratory of Molecular Biotechnology, Department

of Biochemistry, University of Antwerp, Universiteitsplein 1, B-2610 Wilrijk, Belgium

Received November 30, 1995; accepted March 27, 1996

(7), and glomus tumors (6). Moreover, 11q13 is impli-In our effort to characterize a gene at chromosome cated in a number of hematological malignancies, the

11q13 involved in a t(11;17)(q13;q21) translocation best known being the reciprocal translocation t(11;14)in B-non-Hodgkin lymphoma, we have identified a (q13;q32) in B-cell lymphoma, which brings the BCL1/novel human gene, NOF (Neighbour of FAU). It maps CCND1 gene at 11q13 in juxtaposition to the Ig heavyright next to FAU in a head to head configuration gene complex at 14q32 (9, 11). Several cases of involve-separated by a maximum of 146 nucleotides. cDNA ment of the 11q13 region have been described in hema-clones representing NOF hybridized to a 2.2-kb tological diseases where the BCL1/CCND1 region is notmRNA present in all tissues tested. The largest open affected, suggesting the importance of genes other thanreading frame appeared to contain 166 amino acids CCND1 in these neoplasms. Recently, we identified aand is proline rich, and the sequence shows no ho-

t(11;17)(q13;q21) translocation in tumor cells of amology with any known gene in the public data-t(11;14)(q13;q32)-positive B-cell non-Hodgkin lym-bases. The NOF gene consists of 4 exons and 3 in-phoma patient by FISH analysis using a FAU-con-trons spanning approximately 5 kb, and the bound-taining cosmid clone as molecular probe (14). Sincearies between exons and introns follow the GT/AGFAU is not affected in the t(11;17)(q13;q21) transloca-rule. The NOF locus is conserved during evolution,tion, we started a search for another gene(s) involvedwith the predicted protein having over 80% identityin this 11q13 chromosomal rearrangement.to three translated mouse and rat ESTs of unknown

function. Moreover, the mouse ESTs map in the same A human skeletal muscle cDNA library in the Uni-organization, closely linked to the FAU gene, in the ZAP XR vector (Stratagene) was screened with the en-mouse genome. NOF, however, is not affected by the tire FAU-containing cosmid (cosmid 15.1). This led tot(11;17)(q13;q21) chromosomal translocation. q 1996 the isolation of two different clones, one being the FAUAcademic Press, Inc. cDNA and the other being a new clone, putatively

called G22 and consisting of 3080 nucleotides. North-ern blot analysis revealed that the latter clone was

Human chromosome 11 is one of the best character- chimeric. Therefore, we designed amplimer sets forized autosomes, and particular interest in band q13 of both ends of this clone. The first amplimer, specific forthis chromosome is due to the occurrence of multiple the 3* end (G22up, 5*-ATGCCCTATGCTGGCTGTG-3*;disease genes. A number of oncogenes as well as the G22low, 5*-TGGGCACTTTGTTTCCTTG-3*), gener-tumor suppressor gene MEN1 have been mapped to ated a 296-nt fragment when applied to cosmid 15.1this region. Recurrent chromosomal rearrangements DNA and to DNA of a cell line containing human chro-involving 11q13 have been found in a number of solid

mosome 11 as its only human constituent. The secondtumors. These include breast carcinoma, squamous cellamplimer (specific for the 5* end) was not chromosomecarcinomas of head and neck (5), extragonadal germ11 specific (data not shown). Using the former 296-ntcell tumors (10), renal oncocytomas (12), hibernomasPCR probe, we then screened a human lung carcinoidcDNA library in lgt11. After two rounds of hybridiza-Sequence data from this article have been deposited with the

EMBL/GenBank Data Libraries under Accession No. U39400. tion, the four positive clones were sequenced. The lon-1 To whom correspondence should be addressed. Telephone: 32-16- gest clone (PEM14) was found to be identical to the 3*

346082. E-mail: KOEN.KAS@MED.KULEUVEN.AC.BE. end of cDNA clone G22, but lacked the poly(A) tail and2 Both authors contributed equally to this article.a few nucleotides in front of it, which are present in3 Present address: Howard Hughes Medical Institute, Stanford

University Medical Center, Stanford, CA 94305. G22. However, PEM14 was found to contain an extra

433 GENOMICS 34, 433–436 (1996)ARTICLE NO. 0310

0888-7543/96 $18.00Copyright q 1996 by Academic Press, Inc.

All rights of reproduction in any form reserved.

AID Genom 4113 / 6r16$$$481 05-16-96 00:09:49 gnmxa AP: Genomics

SHORT COMMUNICATION434

FIG. 1. (A) Nucleotide and predicted amino acid sequence of the NOF cDNA. The protein sequence is shown below the nucleotidesequence. The asterisks indicate the locations of exon boundaries. The polyadenylation signal is underlined. (B) Genomic organization ofthe human NOF and FAU genes. Black boxes represent the coding regions. H, HindIII.

146 nt at its 5* end. cDNA clones G22 and PEM14 moter region of a maximum of 146 nt. The longest openreading frame starts already at position 14 with ATGwere sequenced (i) after subcloning of cDNA restriction

fragments in plasmids and using vector primers and (ii) and encodes 166 aa. Sequence immediately sur-rounding the first methionine correlates with the con-using cDNA-specific oligonucleotide primers. To obtain

the 5* end of the cDNA we performed 5* RACE experi- sensus sequence for translation initiation in vertebratemRNAs (GCCA/GCCATGG) at the most important po-ments. The longest clone isolated as such extended the

cDNA sequence by 16 nt. Figure 1 shows the entire sitions /4 (G) and 03 (A) only (4). The NOF cDNAcontains an upstream in-frame stop codon precedingsequence of the NOF cDNA (2008 nt) and the amino

acid sequence that it is predicted to encode (166 aa). the presumptive initiating methionine. The open read-ing frame of NOF extends for only a quarter of the full-The first 91 nt of the NOF cDNA match exactly with

the promoter of the FAU gene at positions 0147 to length mRNA sequence, resulting in a long 3*-untrans-lated region. This 3* UTR contains one Alu repeat0237 (3). This places the NOF cDNA in a head to head

configuration, right next to FAU, with a common pro- element (position 778 to 1064) and a polyadenylation

AID Genom 4113 / 6r16$$$481 05-16-96 00:09:49 gnmxa AP: Genomics

SHORT COMMUNICATION 435

FIG. 2. Conservation of the predicted amino acid sequence of the NOF protein among human, rat, and mouse. Amino acids that areidentical in two or more of the clones are boxed. H, human predicted NOF protein; M1, mouse translated EST Z31185; M2, mouse translatedEST N28180; R, rat translated EST H35229.

signal ATTAAA (8) 28 nucleotides upstream from the bp. The intron sizes were determined by exon–exonPCR amplification of genomic DNA (intron 1 and 2) orpoly(A) tail.

The predicted NOF protein has a molecular mass of complete sequencing (intron 3). Sizes are, respectively,{2000, {900, and 110 bp. Two types of introns are19.2 kDa and an isoelectric point of 9.49. One of its

striking features is the high overall proline content present, type 0 (uninterrupted codon (introns 1 and 3)and type 1 (splice site after the first nucleotide) (intron(10% instead of the expected 5%). An abundance of

prolines in a protein sequence would be expected to 2). The 3*-untranslated region is contained within asingle exon.affect significantly the overall structure of the protein

by disrupting a-helices and contributing to an extended The NOF and FAU genes are oriented in a head tohead configuration, separated by a common promoteror highly kinked protein conformation. The introduc-

tion of a run of prolines into a protein sequence could region of maximum 146 bp (Fig. 1). This promoter lacksa TATA box and CAAT element but contains a numberserve to separate physically two functionally distinct

domains of a protein (13). Sequence analysis and data- of GC boxes (Sp1 binding sites). Together with the ubiq-uitous expression of NOF, this promoter is a strongbase searches revealed no significant homology with

any known gene or protein so far and hence provided argument for NOF being a housekeeping gene (2). Thehead to head organization is similar to a growing num-no clues to the function of the NOF gene product. Two

potential protein kinase C (PKC) phosphorylation sites ber of gene pairs that share only a short bidirectionalpromoter.(amino acids 9–11 and 146–148), one potential casein

kinase II (CK-2) phosphorylation site (amino acids 41– Zoo-blot analysis using a 5* NOF cDNA subproberevealed positive fragments in all mammals examined44), and one potential N-myristylation site (amino

acids 16–21) were identified within the NOF protein. (mandril, dog, pig, rat, mouse) (data not shown). Thepredicted protein sequence of the NOF gene showedThat the NOF cDNA represents an almost full-length

transcript is clear from Northern blot experiments per- significant similarity with two mouse cDNA clones (Ac-cession Nos. N28180 and Z31185) and one rat cDNAformed on a broad range of tissue mRNAs (pancreas,

kidney, skeletal muscle, liver, lung, placenta, brain, clone (Accession No. H35229) of as yet unknown func-tion (Fig. 2). The overall amino acid sequence identityheart, spleen, thymus, prostate, testis, ovary, small in-

testine, colon, and peripheral blood leukocyte). The among these four clones is 80% in the published re-gions. The extensive cross-species hybridization of hu-G22uplow probe detects mRNA of approximately 2.2

kb in all tissue types tested, indicating that the NOF man NOF cDNA with mammalian DNA sequences andthe sequence similarity of the putative NOF proteingene is ubiquitously expressed (data not shown).

The positions of splice sites for the NOF gene were with a predicted mouse and rat protein suggest a fun-damental function of the gene product. Sequence align-determined by comparison of the cDNA and genomic

sequences and by identification of consensus sequences ment revealed that mouse EST (Z31185) maps rightnext to the mouse FAU gene at chromosome 19 (1) infrom intron–exon splice junctions. The intron se-

quences for the 5* donor and 3* acceptor splice sites of a head to head configuration. The common promoterregion is 225 bp, slightly larger than that in the humanthis gene correlate well with splice consensus se-

quences of known eukaryotic genes. The exon bound- situation. However, since this mouse cDNA clone isprobably not full length at its 5* end, the common pro-aries are indicated on the cDNA sequence in Fig. 1.

The exon sizes are, respectively, 91, 150, 125, and 1642 moter region will turn out to be comparable if full-

AID Genom 4113 / 6r16$$$481 05-16-96 00:09:49 gnmxa AP: Genomics

SHORT COMMUNICATION436

3. Kas, K., Michiels, L., and Merregaert, J. (1992). Genomic struc-length mouse NOF cDNAs are isolated. Strikingly, theture and expression of the human FAU gene, encoding the ribo-exact match of this mouse EST and the mouse FAUsomal protein S30 fused to a Ubiquitin-like protein. Biochem.

promoter region (1) comes to an end at position 64 of Biophys. Res. Commun. 187: 927–933.Z31185, a perfect exon–exon splice site (TG/AG) at ex- 4. Kozak, M. (1991). Structural features in eukaryotic mRNAsactly the same position where human NOF exon 1 and that modulate the initiation of translation. J. Biol. Chem. 266:

19867–19870.exon 2 are separated by an intron (data not shown).5. Lammie, G. A., and Peters, G. (1991). Chromosome 11q13 ab-The conserved topography of the NOF/FAU gene pair

normalities in human cancer. Cancer Cells 3: 413–420.in human and mice indicates that their organization/6. Mariman, E. C. M., van Beersum, S. E. C., Cremers, C. W. R. J.,conservation is of functional significance and that both

Struycken, P. M., and Ropers, H. H. (1995). Fine mapping of agenes might be coregulated and/or present in the same putatively imprinted gene for familial non-chromaffin paragan-cellular pathway. gliomas to chromosome 11q13.1: Evidence for genetic hetero-

A Southern blot of BamHI-, EcoRI-, HindIII-, and geneity. Hum. Genet. 95: 56–62.7. Mrozek, K., Karakousis, C. P., and Bloomfield, C. D. (1994).PstI-digested genomic DNA of our B-NHL patient was

Band 11q13 is nonrandomly rearranged in hibernomas. Genesscreened with probes representing the 5* part and 3*Chromosomes Cancer 9: 145–147.part of NOF cDNA. No gross rearrangements could be

8. Proudfoot, N. J., and Brownlee, G. G. (1976). 3* non-coding re-detected with these probes (data not shown). Hence wegion sequences in eukaryotic messenger RNA. Nature 263: 211–

conclude that NOF is not affected due to the t(11;17) 214.translocation in our B-NHL patient. However, we have 9. Rosenberg, C. L., Wong, E., Petty, E. M., Bale, A. E., Tsujimoto,preliminary evidence for a third gene contained in this Y., Harris, N. L., and Arnold, A. (1991). PRAD-1, a candidate

BCL1 oncogene: Mapping and expression in centrocyticFAU cosmid, oriented in antisense orientation to NOFlymphoma. Proc. Natl. Acad. Sci. USA 88: 9638–9642.and currently under investigation (Kas et al., unpub-

10. Sinke, R. J., Olde Weghuis, D., Suijkerbuijk, R. F., Tanigami,lished data). Each new gene that becomes character-A., Nakamura, Y., Larsson, C., Weber, G., de Jong, B., Ooster-ized in the 11q13 region is a candidate gene for involve- huis, J. W., Molenaar, W. M., and Geurts van Kessel, A. (1994).

ment in other tumors or hereditary disease. Molecular characterization of a recurring complex chromosomaltranslocation in two human extragonadal germ cell tumors.Cancer Genet. Cytogenet. 73: 11–16.ACKNOWLEDGMENTS

11. Tsujimoto, Y., Yunis, J., Onorato-Showe, L., Erikson, J., Nowell,P. C., and Croce, C. M. (1984). Molecular cloning of the chromo-Koen Kas is a postdoc of the NFWO. Vanessa Lemahieu was sup-somal breakpoint of B-cell lymphomas and leukemias with theported by the Belgisch Werk tegen Kanker. We thank Marianne Vozt(11,14) chromosome translocation. Science 244: 1403–1406.for assistance with the 5* RACE. This work was partially supported

12. van den Berg, E., Dijkhuizen, T., Storkel, S., Brutel de laby the Belgian Fonds voor Geneeskundig Wetenschappelijk Onder-Riviere, G., Dam, A., Mensink, H. J. A., Oosterhuis, J. W., andzoek (FGWO, G.3102.90) and Kom op tegen Kanker (NFWO,de Jong, B. (1995). Chromosomal changes in renal oncocytomas.G.0328.95).Cancer Genet. Cytogenet. 79: 164–168.

13. Vanhoof, G., Goossens, F., De Meester, I., Hendriks, D., andREFERENCES Scharpe, S. (1995). Proline motifs in peptides and their biologi-

cal processing. FASEB J. 9: 736–744.1. Casteels, D., Poirier, C., Guenet, J.-L., and Merregaert, J. 14. Wlodarska, I., Schoenmakers, E., Kas, K., Merregaert, J., Lem-

(1995). The mouse Fau gene: Genomic structure, chromosomal ahieu, V., Weier, U., Van den Berghe, H., and Van de Ven,localization, and characterization of two retropseudogenes. W. J. M. (1993). Molecular mapping of the chromosome 11Genomics 25: 291–294. breakpoint of t(11;17)(q13;q21) in a t(11;14)(q14;q32) positive

B-Non-Hodgkin’s Lymphoma. Genes Chromosomes Cancer 8:2. Dynan, W. S. (1986). Promoters for housekeeping genes. TrendsGenet. 2: 196–197. 224–229.

AID Genom 4113 / 6r16$$$481 05-16-96 00:09:49 gnmxa AP: Genomics