Identification and characterization of the multiple endocrine neoplasia type 1 (MEN1) gene

Journal of Internal Medicine 1998; 243: 433–439

© 1998 Blackwell Science Ltd 433

Introduction

Multiple endocrine neoplasia type 1 (MEN-1) is anautosomal dominant familial cancer syndrome,characterized primarily by tumours of theparathyroid, enteropancreatic tissues and anterior

pituitary. To a lesser extent, foregut carcinoids,lipomas, angiofibriomas, thyroid adenomas,

Identification and characterization of the multiple endocrineneoplasia type 1 (MEN1) gene

S. C. GURU a , P. MANICKAM a , J. S. CRABTREE a , S-E. OLUFEMI a , S. K. AGARWAL b , L . V. DEBELENKO c et al.*From aGenetics and Molecular Biology Branch, NHGRI, bMetabolic Diseases Branch, NIDDK, cLaboratory of Pathology, NCI, dNational Center forBiotechnology Information, NLM, National Institutes of Health, Bethesda, MD; and eDepartments of Chemistry and Biochemistry, University ofOklahoma, Norman; USA

MINISYMPOSIUM MEN & VHL

Abstract. Guru SC, Manickam, Crabtree JS, OlufemiS-E, Agarwal SK, Debelenko LV, Zhuang Z, LubenskyIA, Kester MB, Kim YS, Heppner C, Weisemann JM,Boguski MS, Wang Y, Roe BA, Burns AL, Liotta LA,Speigel AM, Emmert-Buck MR, Marx SJ, Collins FS,Chandrasekharappa SC (National Institutes ofHealth, Bethesda; University of Oklahoma, Norman;USA). Identification and characterization of the mul-tiple endocrine neoplasia type 1 (MEN1) gene(Minisymposium: MEN & VHL). J Intern Med 1998;243: 433–9.

For nearly a decade since the mapping of the multi-ple endocrine neoplasia type 1 (MEN1) locus to11q13 and the suggestion that it is a tumour sup-pressor gene, efforts have been made to identify thegene responsible for this familial cancer syndrome.Recently, we have identified the MEN1 gene by thepositional cloning approach. This effort involved con-struction of a 2.8-Mb physical map(D11S480–D11S913) based primarily on a bacterialclone contig. Using these resources, 20 new polymor-

phic markers were isolated which helped to reducethe interval for candidate genes by haplotype analy-sis in families and by loss of heterozygosity (LOH)studies in approximately 200 tumours, utilizinglaser-assisted microdissection to obtain tumour cellswith minimal or no admixture by normal cells. Theinterval was narrowed by LOH to only 300 kb, andnearly 20 new transcripts that map to this region of11q13 were isolated and characterized. One of thetranscripts was found by dideoxyfingerprinting andcycle sequencing to harbour deleterious germlinemutations in affected individuals from MEN-1 kin-dreds and therefore identified as the MEN1 gene. Thetype of germline mutations and the identification ofmutations in sporadic tumours support theKnudson’s two-hit model of tumorigenesis for MEN-1. Efforts are being made to identify the function ofthe MEN1 gene-encoded protein, menin, and tostudy its role in tumorigenesis.

Keywords: 11q13, contig, LOH, MEN-1, menin,mutations, transcripts.

*Z. Zhuangc, I. A. Lubenskyc, M. B. Kesterb, Y. S. Kimb, C. Heppnerb,J. M. Weisemannd, M. S. Boguskid, Y. Wange, B. A. Roee, A. L.Burnsb, L. A. Liottac, A. M. Spiegelb, M. R. Emmert-Buckc, S. J.Marxb, F. S. Collinsa & S. C. Chandrasekharappaa.

S. C. GURU et al.: IDENTIFICATION AND CHARACTERIZATION OF THE MEN1 GENE434

© 1998 Blackwell Science Ltd Journal of Internal Medicine 243: 433–439

angiomyolipomas and spinal cord ependymomas are also associated with MEN-1 [1]. The prevalenceof MEN-1 is estimated to range from 1 in 10 000 to 1in 100 000 [1, 2].

Larsson et al. [3] showed by linkage analysis thatthe predisposing gene for MEN-1 is linked to themarker PYGM (muscle glycogen phosphorylase)located at 11q13. Several other subsequent studieshave also shown tight linkage to the PYGM locus [4].Loss of heterozygosity (LOH) studies showingtumour-associated deletions, both in familial MEN-1tumours and in sporadic tumours, suggested that theMEN1 gene is likely to be a tumour suppressor gene[5].

Since the initial linkage studies nine years ago,intensive efforts have been made to identify theMEN1 gene. Three years ago, the Fifth InternationalWorkshop on MEN-1 concluded that the disease genelocus is flanked by the markers D11S480 andD11S913, a nearly 5-cM interval at 11q13, based onthe recombinants observed in the linkage analysis of87 MEN-1 kindreds [6]. Recently, we succeeded inidentifying the MEN1 gene from this interval [7].This effort required construction of a clone contig,isolation and mapping of polymorphic markers andtranscripts for the D11S480–D11S913 MEN1 inter-val. The process of positional cloning that led to theidentification of the gene is described here.

Results

Narrowing the MEN-1 candidate gene interval at 11q13

Haplotype analysis of a large North American MEN-1 kindred identified recombinants definingD11S1883 and D11S4907 as the centromeric andthe telomeric boundaries for the MEN1 locus [8]. Acloser telomeric boundary at D11S449, reported bythe European Consortium for MEN-1 [4], narrowedthe recombination-based MEN1 interval to a 2-Mb(D11S1883–D11S449) region (Fig. 1).

We have carried out LOH studies on nearly 200microdissected tumour samples from both familialMEN-1 and sporadic tumours of the same tissues for a large number of polymorphic markers [9–11]. These studies identified PYGM as theproximal and D11S4936 as the telomeric boundaries(Fig. 1). Thus the MEN1 locus was narrowed to a 300-kb PYGM–D11S4936 interval. Since theD11S4936 boundary was identified in a sporadicgastrinoma tumour, it is indicated by a dotted line (Fig. 1).

Both the haplotype analysis and LOH studies were able to narrow the interval as a result ofisolation of new polymorphic markers, and preciseordering of the new and known polymorphicmarkers for the D11S480–D11S913 region(described below).

Fig. 1 The MEN1 interval at 11q13 defined by linkage analysis and loss of heterozygosity (LOH) studies. The MEN1 locus at 11q13 isexpanded to show the location of a few relevant markers on a 2.8-Mb physical map for the D11S480–D11S913 region. The MEN1 intervalby linkage analysis, D11S1883–D11S449, as well as the LOH interval, PYGM–D11S4936, are shown. Since the D11S4936 boundary wasobserved in a sporadic gastrinoma, this boundary is shown by a dotted line.

MINISYMPOSIUM: MEN & VHL 435


Construction of a clone contig for the MEN1 region

A single genomic clone contig for the entireD11S480–D11S913 region was constructed, and thesize of this interval was found to be 2.8 Mb [12].Initial efforts to generate a clone contig based onyeast artificial chromosome (YAC) clones alone wereunsuccessful. In addition to the inherent problemsassociated with the YAC clones such as chimaerismand interstitial deletions, it was soon realized that theMEN1 region was not well represented in YAClibraries. However, the advent of new bacterial clonelibraries, PAC [13], BAC [14] and P1 [15], allowed forthe successful assembly of overlapping clones for theentire 2.8-Mb region. A total of 79 clones wereassembled into a single contig based on the STS-con-tent analysis with 118 STSs [12]. This contig provid-ed reagents not only for the isolation andcharacterization but also for the precise ordering ofpolymorphic markers and transcripts.

Isolation of polymorphic markers for the MEN1 region

We have isolated 16 new short tandem repeat (STR)polymorphic markers, and generated a precisely ordered

map of the new and the six previously knownmicrosatellite polymorphic markers (D11S1883,D11S599, D11S457, PYGM, D11S1783 and D11S449)from the D11S480–D11S913 region [16]. The newmarkers include 10 dinucleotide (D11S4909,D11S4945, D11S4946, D11S4947, D11S4942,D11S4933, D11S4943, D11S4944, D11S4907 andD11S4908), one trinucleotide (D11S4937) and fivetetranucleotide (D11S4940, D11S4939, D11S4938,D11S4936 and D11S4941) repeats. In addition to the22 STR polymorphic markers, our precisely ordered mapof a total of 30 markers within the D11S480–D11S913region includes five biallelic and three previously knownVNTR markers [16]. The biallelic markers included onepreviously known marker (D11S1266), and four intra-genic markers, one each for GC-kinase, ARL2, Kappa andDNA polymerase a. The VNTR markers are D11S427,D11S633 and D11S460. These markers were critical innarrowing the MEN1 interval by haplotype [8] and LOHanalysis [11] as described above.

Generating a transcript map for the MEN1 region

Figure 2 shows an ordered map of 33 transcripts forthe D11S480–D11S913 region, including the 12

Fig. 2 Transcript map of the 2.8-Mb region containing the MEN1 locus in the interval D11S480–D11S913. The MEN1 interval identified bylinkage analysis spanning from D11S1883 to D11S449 is shown, with the region from PYGM to D11S4936 marked by arrows as theminimal interval inferred from LOH data. The arrow with a dotted line represents the boundary determined from a sporadic gastrinoma. The1-Mb region from PYGM to D11S4933 is expanded for clarity. Some relevant markers are indicated above the map and the transcripts areidentified below. * denotes genes which were eliminated as MEN1 candidates by mutation detection analysis. Mutation analysis was notcompleted for Kappa and Eta, marked with **. Identification and mapping of OVO was previously reported [23]. The location of theMEN1(Mu) gene is indicated.



genes previously known to map to this region: VRF,FKBP2, PNG, PLCb3, PYGM, ZFM1, FAU1, NOF1,CAPN1, MLK3, TIP60 and RELA [17]. Our efforts toidentify additional transcripts primarily involvedanalysis of the sequences generated from either theends or sample sequencing of the genomic (YAC,BAC, PAC, P1 and cosmid) clones from this region.Newly identified transcripts include four knowngenes (GC-kinase, ARL2, PP2AB56b and DNA poly-merase a), and three human homologues of knowngenes (HREQ, HPAST and Neurexin are homologuesof mouse Requiem, Drosophila PAST-1 and ratNeurexin II a genes, respectively). Since the remain-ing transcripts did not provide any clues as to theiridentity other than their sequence, they were namedby Greek letters (Alpha, Beta, Gamma, Delta, Epsilon,Zeta, Eta, Theta, Iota, Kappa, Lambda, Mu, Nu andXi); Beta and Epsilon were later found to bepolyadenylation isoforms of the same transcript. Allbut two new transcripts (Kappa and Xi) were identi-fied by the analysis of expressed sequence tags (ESTs)in the dbEST database (http://ncbi.nlm.gov/) thatmatched with the genomic sequences. Kappa and Xi,however, were identified by the analysis of genomicsequences with exon-finding software GRAIL andFEXH.

Mutation analysis of candidate genes

Two methods were employed for the mutation analy-sis of candidate genes; Southern blot and dideoxyfin-gerprinting (ddF) followd by cycle sequencing [17,18]. Southern blots with DNA from 32 independentprobands, digested with HindIII and TaqI, wereprobed with full-length cDNA probes. None of the 11candidate genes analysed identified alterations thatsegregated with the affected individuals in the MEN-1 kindreds [17]. However, three benign polymor-

phisms were observed, a HindIII polymorphism inGC-kinase, and a TaqI polymorphism in HREQ andDNA polymerase a.

On the other hand, deleterious germline mutationswere identified by the ddF method in one of the can-didate genes, Mu, in 15 of the 16 probands tested(Fig. 3). In all instances the mutation segregatedwith the affected individuals in each kindred [7, 19].This initial finding has now been extended to theidentification of germline mutations in additionalMEN-1 kindreds [18, 20, 21].

The MEN1 gene

The MEN1 gene (GenBank Acc. No. U93237)extends across 9 kb and has 10 exons. The gene isexpressed ubiquitously as a 2.8-kb message(GenBank Acc. No. U93236) in all tissues tested; thenature of an additional, slightly larger transcript (3.4kb) expressed in the pancreas and thymus remains tobe characterized [17]. The exon–intron organization,the cDNA sequence and the predicted amino acidsequence are shown in Fig. 4. At present, the aminoacid sequence of the encoded 67-kDa protein, menin,provides few clues to its function.

Discussion

Since the initial linkage analysis demonstrating themapping of the MEN1 locus to 11q13 based on themarker PYGM, several subsequent studies in kin-dreds from divergent ethnic backgrounds haveshown tight linkage to the same locus. No recombi-nants have ever been observed with PYGM [4]. Therehas been no convincing evidence of locus hetero-geneity.

Our meiotic analysis, in combination with thatreported by the European Consortium, helped to

Fig. 3 The 13 different mutations observed in 15 unrelated familial MEN-1 probands. The locations of the frameshift mutations are shownabove a diagram of the MEN1 gene. The in-frame deletions of a single amino acid (K119del and E363del), four nonsense mutations(W198X, W436X, R460X and R527X), and two missense mutations (L22R and W436R) are shown below. The mutations 416delC and512delC were observed in two patients each. The exons of the MEN1 gene are numbered and the cross-hatched areas indicate theuntranslated region.



reduce the MEN1 linkage interval to a 2-MbD11S1883–D11S449 region [4, 8].

LOH studies on MEN-1 and sporadic tumours pro-vided an opportunity to define the MEN1 intervalbased on overlapping deletions. In fact, the applica-tion of microdissection technology to obtain tumourcells with little or no admixture of normal cells, and

the development of a large number of closely placedpolymorphic markers for the MEN1 region, helped tonarrow the interval substantially [9, 10]. The LOHstudies placed the MEN1 gene between PYGM andD11S4936, a 300-kb interval [11].

The construction of a clone contig was an essen-tial part of the positional cloning of the MEN1 gene

Fig. 4 (a) Diagrammaticrepresentation of the exon–intronorganization of the MEN1 gene.The sizes of the exons (filled boxes)and the introns are drawn to scale.(b) The MEN1 cDNA completesequence and the predicted aminoacid sequence of the encodedprotein, menin, are shown. Thefilled triangles point to the locationof the exon–intron junctions. Thetermination codon (TGA) and thepolyadenylation signal (AATACA)are boxed.



[12]. The contig represents complete coverage of theentire 2.8-Mb D11S480–D11S913 region. Thissequence-ready contig provided the precise order ofmore than a hundred STSs which included nearly 30polymorphic markers and more than 30 transcripts.

With the ever-growing list of ESTs in the dbESTdatabase, analysis of sequences generated by thesample sequencing of the genomic (BAC, PAC, P1and cosmid) clones appears to be an efficient methodfor the identification of transcripts. Nearly all thetranscripts we identified in our efforts were represent-ed in the EST database [17], and only two were iden-tified by GRAIL analysis of genomic sequences thatwere not found in dbEST.

The deleterious germline mutations observed inMEN-1 kindreds and the loss of the wild-type alleleobserved in tumours from familial MEN-1 patientssupport the Knudson’s two-hit model of tumorigene-sis for the MEN1 gene [7, 20, 21]. As predicted fromthis model, somatic mutations in sporadic endocrinetumours have also been observed [18, 22]. It remainsto be seen if mutations in the MEN1 gene play a rolein the tumorigenesis of other nonendocrine tumoursas well.

The predicted 610-amino acid sequence of theMEN1 gene-encoded protein, menin, does not showhomology to any previously known proteinsequence. There are no readily identifiable functionalmotifs in the menin sequence. At present, the func-tion of the menin protein in the cell, and its role intumorigenesis, remains unknown. Thus the identifi-cation of the predisposing MEN1 gene has broughtan end to a 9-year-long intensive search. It is now tobe hoped that the functional role of menin will beuncovered, and application of this information toimprove the diagnosis and treatment of MEN-1 willbe forthcoming.

Acknowledgements

C.H. was supported by a grant from the Fritz ThyssenStiftung (Germany).

References1 Metz DC, Jensen RT, Bale AE, Skarulis MC, Eastman RC,

Nieman L et al. Multiple Endocrine Neoplasia type-1: clinicalfeatures and management. In: Bilezekian JP, Levine MA &Marcus R, eds. The Parathyroids. New York: Raven Press, 1994;591–646.

2 Trump D, Farren B, Wooding C, Pang JT, Besser GM,

Buchanan KD et al. Clinical studies of multiple endocrine neo-plasia type 1 (MEN1). Quart J Med 1996; 89: 653–69.

3 Larsson C, Skogseid B, Oberg K, Nakamura Y, Nordenskjold M.Multiple Endocrine Neoplasia type 1 maps to chromosome 11and is lost in insulinoma. Nature 1988; 332: 85–7.

4 Courseaux A, Grosgeorge J, Gaudray P, Pannett AA, ForbesSA, Williamson C et al. Definition of the minimal MEN1 candi-date area based on a 5-Mb integrated map of proximal 11q13.The European Consortium on MEN1. Genomics 1996; 37:354–65.

5 Friedman E, Larsson C, Amorosi A, Bale ML, Metz D, Jensen RTet al. Multiple Endocrine Neoplasia type 1: Pathophysiology,and differential diagnosis. In: Bilezekian JP, Levine MA &Marcus R, eds. The Parathyroids. New York: Raven Press, 1994;647–80.

6 Larsson C, Calender A, Grimmond S, Giraud S, Hayward NK etal. Molecular tools for presymptomatic testing in multipleendocrine neoplasia type 1. J Intern Med 1995; 238: 239–44.

7 Chandrasekharappa SC, Guru SC, Manickam P, Olufemi SE,Collins FS, Emmert-Buck MR et al. Positional cloning of thegene for Multiple Endocrine Neoplasia-type 1. Science 1997;276: 404–7.

8 Debelenko LV, Emmert-Buck MR, Manickam P, Kester M, GuruSC, DiFranco EM et al. Haplotype analysis defines a minimalinterval for the multiple endocrine neoplasia type 1 (MEN1)gene. Cancer Res 1997; 57: 1039–42.

9 Lubensky IA, Debelenko LV, Zhuang Z, Dong Q,Chandrasekharappa S et al. Allelic deletions on chromosome11q13 in multiple tumors from individual MEN1 patients.Cancer Res 1997; 56: 5272–8.

10 Debelenko LV, Zhuang Z, Emmert-Buck MR,Chandrasekharappa SC, Manickam P, Guru SC et al. Allelicdeletions on chromosome 11q13 in multiple endocrine neo-plasia type 1-associated and sporadic gastrinomas and pancre-atic endocrine tumors. Cancer Res 1997; 57: 2238–43.

11 Emmert-Buck MR, Lubensky IA, Dong Q, Manickam P, GuruSC, Kester MB et al. Localization of the multiple endocrine neo-plasia type I (MEN1) gene based on tumor loss of heterozygosi-ty analysis. Cancer Res 1997; 57: 185558.

12 Guru SC, Olufemi SE, Manickam P, Cummings C, Gieser LM,Pike BL et al. A 2.8-Mb clone contig of the multiple endocrineneoplasia type 1 (MEN1) region at 11q13. Genomics 1997; 42:436–45.

13 Ioannou PA, Amemiya CT, Garnes J, Kroisel PM, Shizuya H,Chen C et al. A new bacteriophage P1-derived vector for thepropagation of large human DNA fragments. Nature Genet1994; 6: 84–9.

14 Shizuya H, Birren B, Kim UJ, Mancino V, Slepak T, Tachiiri Y,Simon M. Cloning and stable maintenance of 300-kilobase-pair fragments of human DNA in Escherichia coli using an F-factor-based vector. Proc Natl Acad Sci USA 1992; 89: 8794–7.

15 Shepherd NS, Pfronger BD, Coulby JN, Ackerman SL,Vaidynathan G, Sauer RH et al. Preparation and screening ofan arrayed human genomic library generated with the P1cloning system. Proc Natl Acad Sci USA 1994; 91: 2629–33.

16 Manickam P, Guru SC, Debelenko LV, Agarwal SK, Olufemi S-E, Weissman JM et al. Eighteen new polymorphic markers inthe MEN1 Region. Human Genet 1998; 101: 102–8.

17 Guru SC, Agarwal SK, Manickam P, Olufemi S-E, Crabtree J,Weisemann JM et al. A transcript map for the 2.8 Mb regioncontaining the Multiple endocrine neoplasia type 1 (MEN1)locus. Genome Res 1997; 7: 725–35.



18 Marx SJ, Agarwal SK, Kester MB, Hepner C, Kim YS, Emmert-Buck MR et al. Germline and somatic mutation of the gene forMultiple Endocrine Neoplasia type-1 (MEN1). SixthInternational Workshop on Multiple Endocrine Neoplasia June25–28, The Netherlands, 1997.

19 Olufemi S-E, Green JS, Manickam P, Guru SC, Agarwal SK,Kester MB et al. A common ancestral mutation in the MEN1gene is likely responsible for MEN1Burin in four kindreds fromNewfoundland. Human Mutation 1998; 11: 264–9.

20 Agarwal SK, Kester MB, Debelenko LV, Heppner C, Emmert-Buck MR, Skarulis MC et al. Germline mutations of the MEN1gene in familial multiple endocrine neoplasia type 1 and relat-ed states. Hum Mol Genet 1997; 6: 1169–75.

21 Lemmens I, Van de Ven WJ, Kas K, Zhang CX, Giraud S,Wautot V et al. Identification of the multiple endocrine neopla-sia type 1 (MEN1) gene. The European Consortium on MEN1.Hum Mol Genet 1997; 6: 1177–83.

22 Heppner C, Kester MB, Agarwal SK, Debelenko LV, Emmert-Buck MR, Guru SC et al. Somatic mutation of theMEN1 gene in parathyroid tumors. Nature Genet 1997; 16:375–8.

23 Chidambaram A, Allikmets R, Chandrasekarappa SC, GuruSC, Modi W, Gerrard B, Dean M. Characterization of a humanhomolog (OVOL1) of the Drosophila ovo gene, which maps tochromosome 11q13. Mamm Genome 1997; 8: 950–1.

Received 25 November 1997; accepted 22 January 1998.

Correspondence: Settara C. Chandrasekharappa, Genetics andMolecular Biology Branch, National Human Genome ResearchInstitute, National Institutes of Health, Building 49, Room 3E-13, 49 Convent Drive, MSC 4442, Bethesda, MD 20892-4442, USA (tel: 301 402 2344; fax: 301 402 4929; e-mail:[email protected]).

Identification and characterization of the multiple endocrine neoplasia type 1 (MEN1) gene

Documents

Transcript of Identification and characterization of the multiple endocrine neoplasia type 1 (MEN1) gene