Allelic Loss from Chromosome 11 in Parathyroid Tumors

[CANCER RESEARCH 52, 6804-6809, December 15. 1992] Allelic Loss from Chromosome 11 in Parathyroid Tumors

Eitan Friedman, ~ Luiz De Marco, 2 Pablo V. Gejman, Jeffrey A. Norton, Allen E. Bale, Gerald D. Aurbach, 3 Allen M. Spiegel, and Stephen J. Marx Molecular Pathophysiology [E. F., .4. M. S.] and Metabolic Diseases [L.D. M., G. D. A., S. J. M.] Branches, National Institute of Diabetes, Digestive, and Kidney Diseases; the Clinical Neurogenetics Branch [P. V. G.], National Institute of Mental Health; and Surgery Branch [J. A. N.], National Cancer Institute, Bethesda, Maryland 20892, and the Department of Human Genetics [.4. E. B.], Yale University School of Medicine, New Haven, Connecticut 06510

ABSTRACT tion mapping") (17, 18). Since the observed allelic losses in parathyroid tumors in the context of MENI have been shown to Parathyroid tumors may occur in a sporadic fashion or, more rarely, be for the most part chromosome 11 specific, and are often as part of a familial syndrome (such as familial multiple endocrine subchromosomal (8, 10), they may be helpful in the fine map- neoplasia type I). The MENI gene has been mapped by linkage analysis to chromosome 11 at band q11-q13, and presumably acts as a tumor ping of the MENI gene. suppressor gene. In the present study, which is an extension of our Sporadic parathyroid adenomas also display allelic losses previous studies, we examined 41 parathyroid tumors from patients with spanning the MENI gene locus as well as other loci on chro- familial multiple endocrine neoplasia type I and 61 sporadic parathyroid mosome 11 (8, 10, 19). Thus, the MENI gene or other tumor tumors with markers on chromosome 11, to assess the extent of allelic suppressor genes on chromosome 11 (20) may be involved in loss in those tumors. Twenty-four of the MENl-associated tumors the development of sporadic parathyroid tumors. We now show (58%) and 16 of the sporadic parathyroid tumors (26%) displayed allelic the extent of allelic losses from chromosome 11 in our series of loss from chromosome 11. The region of overlap of the allelic losses in 41 parathyroid tumors from 31 patients with MENI and from the MENI-associated tumors enables us to place the MENI gene between PGA centromerically and INT2 telomerically, a region spanning 61 parathyroid tumors from 61 patients with sporadic primary about 7.5 cM. Taken together with locus ordering by linkage analysis, hyperparathyroidism. this clearly localizes the MENI gene telomeric to the PGA locus. Our inability to detect allelic loss on chromosome 11 in some parathyroid MATERIALS AND METHODS tumors suggests the existence of other genes involved in the development and/or progression of this subgroup of presumably monocional Patients and Specimens. Parathyroid glands surgically excised to tumors; or that localized events involving the 11q tumor suppressor gene correct primary hyperparathyroidism were obtained: 41 parathyroids have occurred in some parathyroid tumors whose detection is beyond the from 31 patients with FMENI. In 8 patients, 2 glands were separately sensitivity of our analysis; or that at least some of the specimens ana- analyzed: in one case, 3 glands were studied, and in one case, 2 regions lyzed were in fact primarily hyperplastic parathyroid tissue. from one large gland were analyzed individually. Thirteen tumors were from cryopreserved tissue and the rest from fresh surgical specimens. INTRODUCTION Sixty-one sporadic parathyroid adenomas were similarly analyzed. The criteria for classifying a parathyroid tumor as sporadic or as part of FMENI 4 is an autosomal dominant disorder characterized by MENI were reported previously (8). The parathyroid tumors reported hyperfunction of the pancreatic islets, the anterior pituitary, in this communication include the subset of parathyroid tumors previ- and multiple parathyroid glands (1, 2). Primary hyperpara- ously reported by us (8). In the previous report, 16 tumors from 14 thyroidism is the most common clinically apparent manifesta- MENI patients and 34 tumors from patients with sporadic hyperpara- tion of MENI, affecting virtually all gene carriers by age 50 (3, thyroidism were tested with 8 chromosome 11 probes (8). In this report, 4). The MENI-susceptibility locus has been mapped by genetic additional tumors (25 in the MENI category and 27 in the sporadic linkage in families to chromosome 11 at bands qll-13: to category) were tested. Variable numbers of 9 additional chromosome 11 restriction fragment length polymorphism markers PYGM (5) polymorphic markers were used to evaluate allelic losses in the entire and INT2 (6). Further linkage analysis localized the gene to group, and we also applied some of the previously used probes to the within a few cM of the PYGM marker (7). Larsson and col- newly reported tumors. Gland volume was assessed from the gland dimensions recorded at surgery and the formula for an ellipsoid (21). leagues (5) showed that 2 malignant insulinomas had lost chro- Estimated volume was converted to mass by assuming density equal to mosome 11 alleles inherited from the unaffected parent. Sub- that of water (1 g/cm3). sequently, parathyroid tumors from patients with MENI have Statistical Analysis. Data on gland size were analyzed after logarith- been shown frequently to display allelic loss with markers from mic (base 10) transformation. Group means were compared by the chromosome 11 (8-11). These allelic losses in parathyroids Scheffe test (22). were for the most part confined to chromosome 11 (8, 10). DNA Extraction, Southern Blotting, and Probe Hybridization. High The observation of allelic loss in tumors was first outlined in molecular weight DNA was isolated from tumor specimens and peri- familial retinoblastoma (12), and helped confirm the 2-hit hy- pheral blood leukocytes according to standard protocols (8). In tumor pothesis of tumorigenesis (13). Genes that can underlie this tissue processed immediately after surgery, adjacent, clearly distinct, mechanism have been called tumor suppressor genes (14-16). non-tumor tissue (e.g., fat, fibrous tissue) was manually dissected out. By comparing normal and tumor tissue from the same individ- No attempt was made to quantify the tumor cell fractions in histological slides from tumor tissues. DNA samples (5-10 #g) were digested to ual with multiple probes, a region of allelic loss overlapping in completion with restriction endonucleases (Boehringer Mannheim), different tumors can sometimes be delineated (so-called "dele- electrophoresed on 0.8% agarose gels, transferred to nylon filters (GeneScreen Plus; New England Nuclear), and hybridized to radiola- Received 8/14/92; accepted 9/30/92. beled probes as described (8). The chromosome 11 probes used were: The costs of publicationof this article were defrayed in part by the payment of PTH (pPTHm122), INS (pHINS310), CAT (pINT800), H-ras page charges. This article must thereforebe hereby marked advertisement in accord- ance with 18 U.S.C. Section 1734 solely to indicate this fact. (pTBB2), DI 1S147 (HBI18P1), D11S29 (L7), DI 1S144 (MCT128.1), To whom requests for reprints should be addressed, at Department of Clinical D11S288 (p3C7), DllS149 (pTHH26), PGA (phpepl4-21), PYGM Genetics, KarolinskaHospital, P.O. Box 60500, S-104 01, Stockholm, Sweden. (pMCMP1), DllS146 (pHBI59), INT2 (SS6), and DllS84 (p2-7- 2 Supported by CNPq and FAPEMIG, Brazil. 3 Deceased. I D6). In addition, dinucleotide markers (also called microsatellite 4 The abbreviation used is: FMENI, familial multiple endocrine neoplasia probes) D11S35, CD3D, and DllS420 were used (see below). The type I. nomenclature, chromosomal localization, and linear ordering of the 6804 Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 1992 American Association for Cancer Research. ALLELIC LOSS FROM CHROMOSOME I1 IN I'ARATHYROID TUMORS probes is based on a prior analysis (23, 24) and is shown in Fig. 1C. The RESULTS criteria for scoring a signal as allelic loss were described previously (8). PCR for Microsatellite Markers. Two hundred ng each from paired Allelic Loss in Parathyroid Tumors in MENI. Constitu- tumor and peripheral blood DNA samples were used as a template. The tional and tumor samples were compared at 17 loci on chro- PCR was carried out in a final volume of 15 ul in a microtiter plate, mosome 11 for 41 tumors from 31 patients. In 31 tumors, more using the Techne MW2 thermocycler (Cambridge, England). The re- than 10 probes per tumor were tested; in 5 tumors between 5 action included 1.5-ul 10x buffer (50 mM KC1; 10 mMTris HCI, pH 8.3, 0.01% gelatin, and 1.5 mM MgCl2), 0.25 UM spermidine, 8 pmol of each and 9 markers were tested, and in 5 tumors between 1 and 4 oligonucleotide primer, 200 uM each of dATP, dTTP, dGTP, 2.5 UM of markers were tested. All patients were heterozygous (i.e., infor- dCTP, 0.3/~Ci of [a-32p]dCTP (3000 Ci/mmol), and 0.5 unit of Taq mative) with at least one probe. Seventeen tumors (42%) from DNA polymerase (Perkin Elmer, Norwalk, CT). To quantitate the ef- 14 patients did not display allelic loss with any of the chromo- ficiency of the amplification and the correct amount of DNA in each some 11 markers examined (tumors 6, 8B, llA, llB, 13, 14, reaction, a set of primers from a different chromosome (D 19S75 from 16, 16A, 17, 19A-C, 22, 25B, 26B, 27B, and 30). Twenty-four chromosome 19) that amplifies a region with a different size range was of 41 tumors (58%) showed allelic loss with at least one marker also included in the same reaction mixture (multiplexing). Samples were overlaid with 20 ul light mineral oil to prevent evaporation. After from chromosome 11 (Figs. 1 and 2). The extent of these losses an initial denaturation step at 92~ for 5 min, 20 cycles consisting of is presented in ideograms (Fig. 2, A and B). The classification of annealing at 50~ (90 s), extension at 72~ (90 s), and denaturation at each tumor into a specific subgroup (see below) is limited by the 920C (50 s) were carried out followed by an extension step of 4 min at number and chromosomal locations of informative probes 720C. Two- to 4-~1 aliquots of the amplified DNA were mixed with 4 ul tested in that tumor. Given those limitations, the tumors dis- formamide containing DNA gel loading buffer (U.S. Biochemicals, playing allelic loss could be subdivided into 3 groups: in 7 Clevland, OH). Samples were electrophoresed on 6-8% denaturing tumors from 6 patients (tumors 2, 5, 10, 25A, 27A, 28A, and polyacrylamide DNA sequencing gels for 3-3.5 h at 75 W. Gel size 28B) there was allelic loss with every informative marker from standards were dideoxy sequencing ladders (M 13 mpl 8 template). Gels were fixed in a solution containing 5% methanol and 5% acetic acid, chromosome 11. These tumors are therefore considered to have dried, and exposed to X-Omat film for 24-72 h. All priming oligonu- lost alleles from a large portion of chromosome 11 (total allelic cleotides were 18-20 bases in length, and the sequences for these PCR loss). Five tumors from 5 patients showed an allelic loss pattern primers were taken from the published sequences (25-28). possibly extending to one of the telomeres: in 3 tumors (tumors

PB T

AI - ~ PB T (nt) D 11 ~20 --192 -- A 2 A3 I INS (1 lq 23.3-4) J --188

I PTH 9 D 19S75 --150 Fig. 1. Examples of allelic losses from chromosome I l in parathyroid tumors (,4). South- (19q 12-13.1) ] -'148 ern blots are shown for DNA derived from L _ A 1 A2 germline (P.B., peripheral blood) and from the corresponding tumor tissue (T). For each locus, paired DNA samples were chosen to de- A2 pict the presence of 2 alleles in the germline and only one allele in the tumor tissue. Probe 10 5 names and linear order are as determined pre- B A2 -- PB T PATIENT 7 viously (23, 24). The allele numbers are shown / next to the blot in numbered letters. Blots rep- 15.5 resenting allelic loss with probes PTH, INS, 15.4 PYGM 15.3 PYGM, and Dl IS144 were shown previously PB T DllS1~ 15.2 (adapted from data in Ref. 8). B, multiplexing I-t 15.1 with primers from chromosome 11 and chro- A1 _ INT2 14 mosome 19 to detect allelic loss using the di~ nucleotide repeats. Numbers on the right, sizes 13 of the alleles. Patient's number is shown at the A2 - DllS35 ...... 12 bottom, and the primers used and their chro- 7 11.2 mosomal localization are shown on the left. C, II.I relative localization, linear ordering, and no- PB T PB T DllS147 PB T 1 T2 menclature of the probes used. D11S149 and II 12 Dl IS288 are pericentromeric probes, and are / 13.1 shown in Figs. 2 and 3 on the long arm. 13.2 13.3 13.4 13.5 14.1 14.2 14.3 21 22.1 22.2 22.3 ''2 25 23.1 23.2 23.3 24 ' A 25 11 C 6805 Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 1992 American Association for Cancer Research. ALLELIC LOSS FROM CHROMOSOME I1 IN PARATHYROID TUMORS

A TUMOR 26 PROBE 9 7 3 12 A 111,

"" t') " 0 9 o m l) I ~" 0 q) 9 9 q) -, 9 q )

CENTROMERI

0118140 9 9 9 0 I)1182N 9 9 9 9 9 9 0 9 O1181,1e 9 0 O W12 .o DllN4 ~IIIN o Dl18147 Dl1~ C~aO o o I)118144 O I)11~g20 O 11Q

OrWO ALLELESRETAINED IN TUMOR 0 ONE ALLELE LOST IN TUMOR O PROOE NOT INFORMATIVE

B ~R 8 24 29 A 20 18 I 21 23 A B 4 15 A B 11u,

.,, ~) q) ql, q) q ,',. q) q) q) q) 9 q) q) r q) q) q) ql 0 ql q)

-,-- 0 q ) q ) 0 9 9

~0' q) l) 0 Q 9 ) l) 9 0 9 O0 0 O0 ,,,,.,- O O O 0 0 0 Q O O ,,,2 O 0 0 q) O O ql, q) I O O ~"" q ) q ) q ) q ) 0 0 O O ~"- q ) q ) 9 9

1111

Fig. 2. Allelic patterns in MENI-associated parathyroid tumors. For each tumor, all tested loci on chromosome 11 are shown. _Filled grey circles, noninformative loci (i.e., one allele present in germline DNA). e, informative loci (i.e., 2 alleles in the germline) that did not show allelic loss; 9 informative loci with allelic loss. Empty regions, continuous subregions of presumed loss of alleles from one copy of chromosome 11. Solid lines, presumed retained regions. A, tumors that display allelie loss on either the short arm or the long arm of chromosome 11, possibly extending to the telomere and retaining other loci in the tumor tissue ("open ended allelic loss"). B, tumors showing interstitial allelic loss: loci of alle|ic loss flanked by loci of allelic retention. Numbers, patients. A, B, or C, more than one tumor was analyzed from the same patient.

7, 9, and 26A), the retained locus was on the short arm of cally and the DllS288 marker centromerically. Twelve tumors chromosome 11, and in 2 tumors (tumors 3 and 12), on the long from 10 patients displayed an interstitial deletion pattern (tu- arm (open ended allelic loss)(Fig. 2,4). The region of overlap of mors 1, 4, 8A, 15, 18, 20, 21, 23, 24A, 24B, 29A, and 29B); these allelic losses spans between the INT2 marker telomeri- retaining informative markers at both the long arm and the 6806 Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 1992 American Association for Cancer Research. ALLELIC LOSS FROM CHROMOSOME 11 IN PARATHYROID TUMORS

OPEN ENDED short arm, and showing allelic loss of other markers in between A TUMOR TOTAL LOSS ~,,,,,,,~ UM BE R (Fig. 2B). In 2 of these tumors (tumors 1 and 4) the zone of 11/ROBE ~ 103 135 160 allelic loss was noncontinuous. We tested four loci in the 1 lql 1-q13 region: PGA, PYGM, 9 DI 1S146, INT2, and these data are shown from the MENI- PTH :i);i 9 associated tumors (Table 1). In tumors with no allelic loss or CAT total allelic loss, these data from informative markers do not provide information for subchromosomal localization of the CENTROMER .~...... MENI gene. In tumors with subchromosomal allelic loss, one O11S14g '~ PYGM allele was lost in each of 5 informative tumors (tumors 3, 4, 21, 24A, and 24B), and one DllS146 allele was lost in @ 0 @ each of 5 informative tumors (tumors 1, 4, 7, 21, and 23). The D11S14~~ @ data from the PYGM and D11S146 loci contrast those from IN'r2 @ PGA (allelic loss in only 1 of 7 tumors) or INT2 (allelic loss in DllS84 only 3 of 9 tumors) (Fig. 2, A and B). DllS35 In 9 patients we analyzed more than one tumor (tumors 8, 11, o...... 9 9 19, 24, 25, 26, 27, 28, and 29). In 2 cases (tumors 11 and 19), no allelic loss was demonstrated in any tumor. In one case CD-3D (tumor 28), with the only probe that was informative, both tumors displayed allelic loss. In 4 cases (tumors 8, 25, 26, and D11S420 -- 11,1 27), one tumor (usually the smaller one) did not display allelic OTWO ALLELES RETAINED IN TUMOR (~ ONE ALLELE LOST IN TUMOR :i PROBE NOT INFORMATIVE loss while the other tumor did (Fig. 2, A and B). Surprisingly, 2 patients (tumors 24 and 29) displayed an identical subchromosomal pattern of allelic loss in the 2 tumors analyzed from each B ~NIUUM BMOER ...... UOUS .... CO.... UOUS EXC..... O 11q13 (Fig. 2B). These tumors, in both patients, were located in ana- PROBE ~, 128 146 149 141 113 152 125 115 139 110 tomically distinct, separate glands and were removed at the same operation. Allelic Loss in Sporadic Parathyroid Adenoma. Sixteen of 61 sporadic parathyroid tumors (26%) exhibited allelic loss with at least one marker from chromosome 11. In 34 tumors, more than 10 markers were analyzed, in 21 between 5 and 9 markers, and in 6 between 1 and 4 markers. The same 3 cate- gories that were defined in the MENl-associated tumors were noted (Fig. 3, ,4 and B). Three tumors (tumors 103, 135, and 160) showed allelic losses with every informative marker from chromosome 11 (total loss group). In three others (tumors 104, 119, and 155) there was a subchromosomal but presumably continuous zone of allelic loss extending to one telomere (open- ended loss), that included the MENI gene locus. The remaining tumors displayed a more complex pattern of allelic loss: tumors 113, 128, 141, 146, and 149 showed an interstitial-continuous allelic loss pattern that included the MENI locus; tumors 115, 125, 139, and 152 displayed a subchromosomal, discontinuous ~ITWO ALLELES RETAINEDIN TUMOR (~ ONE ALLELELOST IN TUMOR Q PROBENOT INFORMATIVE allelic loss pattern. Finally, tumor 110 exhibited subchromo- Fig. 3. Allelic patterns in sporadic parathyroid tumors. From each tumor, all somal loss that clearly involved a region that is outside the tested loci on chromosome 11 are shown. Only tumors that have displayed allelic loss are shown. Symbols are identical to Fig. 2. A, total loss and open-ended allelic MENI locus. loss. B, interstitial ailelic loss. Tumor 110 has allelic loss that involves a region Allelic Loss Related to Gland Size. After logarithmic (base clearly outside the MENI gene locus and despite the fact that it may represent 10) transformation of gland mass, the MENI-associated glands "open ended loss," is shown with the interstitial loss group. with allelic loss (data from 14 glands available) had a mean weight of 1.48 (range, 0.27-6.28 g), whereas those not demon- strating allelic loss had a mean weight of 0.43 (data from 16 glands; range, 0.125-5.50 g) (Fig. 4). For sporadic parathyroid tumors, the allelic loss group had a mean weight of 1.01 (14 glands; range, 0.125-3.40 g) and the group without demonstrable allelic loss, 1.66 (32 glands; range, 0.275-57 g). In a comparison of multiple means, the estimated mass of MENl-asso- Table 1 Results from informative loci in the l lql 1-q13 region according to tumor category in MENI cases ciated glands not displaying allelic loss was significantly less than that in any of the other 3 groups (at P < 0.05). The smaller No allelic Total allelic Subchromosomal Marker loss loss allelic loss average mass in MENl-associated glands without allelic loss PGA 0 (4) a 100 (1) 14 (7) reflected an excess (n = 7) of glands with a mass below 250 mg. PYGM 0 (4) 100 (2) 100 (5) This excess does not represent a bias from testing too few D11S146 0 (2) 100 (2) 100 (5) markers in small samples: the number of informative loci was INT2 0 (12) 100 (2) 33 (9) similar in MENI glands with or without allelic loss, and was a Table entries show percent of tumors with allelic loss at the indicated locus. Numbers in parentheses, number of tumors in that cell of table. also similar in those above or below 250 mg. 6807 Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 1992 American Association for Cancer Research. ALLELIC LOSS FROM CHROMOSOME I1 IN PARATHYROID TUMORS

DllS146, and DllS97, whereas the evidence for the centromeric border is not as convincing. To complement the data 4O p obtained from linkage analysis and further restrict the MENI 20 region, Bystr6m et al. (10) have used subchromosomal deletions in parathyroid and pancreatic tumors. Based on a single tumor 10 that retained the PYGM marker, the centromeric border of the

A , o MENI gene was concluded to be at the PYGM locus. The data 14 presented here, which are based exclusively on parathyroid tu- 8o a mors, clearly support the exclusion of PGA from the MENI ~ 2 I, locus as it was retained in the majority of tumors, and thus help in further restricting the MENI region. Allelic losses excluding the PGA locus in MENI-related tumors have also been observed 8 == by others. 6 The INT2 locus, shown to be retained in the ma- o = oo jority of tumors in our series, has clearly been excluded from the 0.4 - O ~ 13 eg 13 MENI region both by linkage analysis and tumor deletion map- 0.2 ping (7, 10, 29). Generation of more polymorphic probes from the region of interest and aligning them in the appropriate order o T 0.1 (30-32) could enable the definition of a still smaller region of ~i~E~i~iii!i!iiiriiiiiiiiii~iii~iii!i!~!~!~!;~i~i~i!i!~!ii!!!i!iiii~i;i~i~i~i~i~;~;~!;~;~!;i~!~!~!~!~!i!!iii!iiiiiiii;i~i~i~i~i~i~i~;i~i~ii~i~i~iiiii~iiiiiiiiiiii~!~!~i;!~!~overlap that, in complementation with linkage analysis, would be useful in fine mapping and possibly cloning of the MENI 0.04 i iii iiii iii i iiiiiiiiiiiiiiiiiiiiiiiii iii i i i! ;i i i i i iiii iiiiiiiiiiiiii i iiiii iiiii iii i i ii i iiiiiiiii iiiiiiiiiiiiiiiiiiiiiiiiiiiii iiiiii ii i i Allelic No Allelic No gene. Loss Allelic Loss Allelic Of note is the fact that the majority of allelic losses detected Loss Loss in parathyroids in FMENI are subchromosomal and interstitial Sporadic FMEN-1 (Fig. 2, ,4 and B), whereas those allelic losses in pancreatic Fig. 4. Parathyroid tumor mass in patients with primary hyperparathyroidism tumors of MENI are for the most part "total allelic loss" (10, caused by sporadic adenoma (0, 9 or FMENI (ll, D). Each group is subdivided 33). The different frequency of subchromosomal allelic losses into tumors with allelic losses (ll, 73) and without allelic losses (0, II). Solid horizontal line, upper limit of normal range (0.075gr) (21). Horizontal dashed between parathyroid and pancreatic tumors may merely reflect line, mean mass of each subgroup. the number of probes and tumors analyzed. Alternatively, it may suggest that the mechanism of acquiring the second inac- DISCUSSION tivating mutation is different for the parathyroid and the pancreatic islet cell in MENI. The high frequency of interstitial In the present study, parathyroid tumors either sporadic or as allelic loss in parathyroids may indicate that allelic loss in this part of FMENI have been probed with markers from chromo- type of tumor involves a complex pattern of genetic events like some 11. This report includes the previously reported parathy- multiple cross-overs. In a mouse model, only a minority (about roid glands (8) and represents an expansion of this group with 3%) of clones displaying allelic losses arises through similar more tumors and more markers. Forty-two % of MENl-asso- complex patterns, whereas the majority of clones (97%) showed ciated parathyroid tumors in this series do not display allelic evidence of more simple events (chromosomal loss or one cross- loss with chromosome 11 markers. Some or all of these may ing over) (34). represent a polyclonal expansion of the parathyroid tissue (i.e., About one-fourth of sporadic parathyroid tumors showed true hyperplasia). Alternatively, it is possible in some or all of allelic losses of chromosome 11 markers. This rate of allelic these tumors that allelic losses are too small to be detected with losses in this type of tumors is in agreement with that reported our panel of probes or our detection technique. Indeed, Iwasaki by others (10). Allelic losses in sporadic tumors involving re- and coworkers 5 have found allelic losses in all 18 of their ME- gions shown to be linked to the inherited form of the disease NI-associated parathyroid tumors, using a panel of highly poly- have been previously demonstrated in several tumors, including morphic dinucleotide markers around the MENI gene locus. retinoblastoma (35) and neurofibromatosis type II (36). The More than 50% of parathyroid tumors in MENI in our series finding in sporadic parathyroid adenomas of allelic loss that display allelic loss with markers from chromosome 11, and all generally includes the MENI gene locus supports the notion regions of loss included the MENI locus. Previous studies (8, that development of such tumors may involve an unmasking of 10) have shown that alleic losses in MENI-associated tumors an inactivating mutation on chromosome 11, perhaps of the are, for the most part, chromosome 11 specific. There are sev- MENI gene itself. However, the complicated pattern of allelic eral implications to finding allelic loss in parathyroid tumors in loss in some tumors as well as the apparent deletion in one MENI. (a) It is indirect evidence for the monoclonality of these tumor of a region that is clearly outside the MENI region, raise tumors (8). (b) The fact that the region of loss in the tumors the possibility that other tumor suppressor genes on chromo- contains markers that are tightly linked to the disease in kin- some 11 are involved in benign sporadic parathyroid tumori- dreds (5, 6) supports the hypothesis that the MENI gene acts as genesis. The fact that the mean weight of all sporadic, benign, a tumor suppressor gene (5, 16). (c) The deletions observed in probably monoclonal parathyroid tumors is similar regardless MENI-associated parathyroid tumors with breakpoints within of allelic loss from chromosome 11, may serve to indicate that 11 q l 1-ql 3 may help to localize the disease gene. oncogenic loci outside chromosome 11 have pathological ef- From several linkage analysis studies, the MENI gene could fects on the parathyroid gland similar to effects from chromo- be localized between DllS288/D11S149 centromerically and some 11. INT2 telomerically (5-7, 29). In most studies, the telomeric The data presented in this study have to be taken in the border is well defined with meiotic cross-overs to INT2, context of the number of probes analyzed per tumor and any

5 H. Iwasaki, personal communication.. 6 C. Larsson, personal communication 6808 Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 1992 American Association for Cancer Research. ALLELIC LOSS FROM CHROMOSOME 11 IN PARATHYROID TUMORS conclusions regarding these tumors must be made in a reserved B., Oberg, K., Werner, S., and Nordenskjold, M. Localization of the gene for multiple endocrine neoplasia type I to a small region within chromosome way. We were obviously limited by the amount of DNA from 1 lql3 by deletion mapping in tumor tissue. Proc. Natl. Acad. Sci. USA, 87: tumor tissue in some cases, the quality of the existing DNA in 1968-1972, 1990. others and the informativeness of the markers used. As evident l 1. Radford, D. M., Ashley, S. W., Wells, S. A., Jr., and Gerhard, D. S. Loss of heterozygosity of markers on chromosome 11 in tumors from patients with from Figs. 2 and 3, in certain instances (particularly tumors 13, multiple endocrine neoplasia type I. Cancer Res., 50: 6529-6533, 1990. 14, 28, and 30 in MENI and tumors 103 and 160 in sporadic 12. Cavenee, W. K., Dryja, T. P., Philips, R. A., Benedict, W. F., Godbout, R., parathyroid tumors), the analysis was limited. Thus, some tu- Gallie, B. L., Murphee, A. L., Strong, L. C., and White, R. L. Expression of recessive alleles by chromosomal mechanisms in retinoblastoma. Nature mors classified as showing no allelic loss might show loss if (Lond.), 305: 779-784, 1983. tested with additional probes; some classified as open-ended 13. Knudson, A. G. Mutation and cancer: statistical study of retinoblastoma. loss might be reclassified as interstitial loss; some with contin- Proc. Natl. Acad. Sci. USA, 68." 820-823, 1971. 14. Bishop, J. M. The molecular genetics of cancer. Science (Washington DC), uous loss could be reclassified as discontinuous loss. 235: 305-311, 1987. The finding of an apparently identical pattern of interstitial 15. Klein, G. The approaching era of the tumor suppressor genes. Science (Wash- ington DC), 238: 1539-1545, 1987. allelic loss in 2 tumors from the same patient in 2 cases (tumors 16. Knudson, A. G. Hereditary cancer, oncogenes and antioncogenes. Cancer 24 and 29) is intriguing. Recently, a similar phenomenon of Res., 45: 1437-1443, 1985. apparently identical interstitial deletion patterns in 6 separate 17. Fearon, E. R., Cho, K. R., Nigro, J. M., Kern, S. E., Simons, J. W., Ruppert, J. M., Hamilton, S. R., Preisinger, A. C., Thomas, G., Kinzler, K. W., and tumors from the same individual has been described in colon Vogelstein, B. Identification of a chromosome 18q gene that is altered in cancer (37). If the finding in parathyroid tumors does not rep- colorectal cancers. Science (Washington DC), 247: 49-56, 1990. resent an artifact (such as tissue misidentification or testing 18. Baker, S. J., Fearon, E. R., Nigro, J. M., Hamilton, S. R., Preisinger, A. C., Jessup, J. M., Vantuienen, P., Ledbetter, D. H., Barker, D. F., Nakamura, Y., with insufficient number of probes), it suggests that the somatic White, R., and Vogelstein, B. Chromosome 17 deletions and p53 mutations mutation (second hit) was not random or that it arose in a in colorectal carcinomas. Science (Washington DC), 244:217-221, 1989. common precursor cell. 19. Arnold, A., and Kim, H. G. Clonal loss of one chromosome 11 in parathyroid adenoma. J. Clin. Endocrinol. Metab., 69: 496-499, 1989. Weight comparison of tumors with allelic losses versus the 20. Saxon, P. J., Srivatsan, E. S., and Stanbridge, E. J. Introduction of human ones not displaying it showed that, as a group, the glands with- chromosome 11 via microcell transfer controls tumorigenic expression of out demonstrable allelic loss were significantly smaller than the HeLa cells. EMBO J., 5: 3461-3466, 1986. 21. Marx, S. J., Menczel, J., Campbell, G., Aurbach, G. D., Spiegel, A. M., and ones with allelic loss. This may indicate that monoclonal ex- Norton, J. A. Heterogeneous size of the parathyroid glands in familial mul- pansion in the development of parathyroid tumors in MENI tiple endocrine neoplasia type 1. Clin. Endocrinol., 35: 521-526, 1991. 22. SAS Institute, Inc. SAS Users Guide: Statistics, Ed. 2, p. 584. Cary, NC: SAS may follow a period of polyclonal expansion. Regardless of the Institute, Inc., 1982. mechanism for the development of allelic losses in MENI, it is 23. Julier, C., Nakamura, Y., Lathrop, M., O'Connell, P., Leppert, M., Litt, M., evident that characterization of these losses with increasing Mohandas, T., Lalouel, J-M., and White, R. A detailed genetic map of the long arm of chromosome 11. Genomics, 7: 335-345, 1990. precision is informative. 24. Junien, C., and Van Heyningen, V. Report of the committee on the genetic constitution of chromosome 11. Cytogenet. Cell Genet., 58." 459-554, 1991. 25. Litt, M., Sharma, V. and Luty, J. A. Dinucleotide repeat polymorphism at ACKNOWLEDGMENTS the D11S35 locus. Nucleic Acids Res., 18." 5921, 1990. 26. Luo, Y., Evans, G. A., and Litt, M. Dinucleotide repeat polymorphism at the We wish to thank Dr. Gunter Weber, Dr. Magnus Nordensjk61d, and D1 IS420 locus. Nucleic Acids Res., 18: 2020, 1990. Dr. Catharina Larsson from the Clinical Genetics Laboratory at the 27. Weber, J. L., Kwitek, A. E., and May, P. E. Dinucleotide repeat polymor- Karolinska Institute for reading the manuscript and for their insightful phisms at the Dl IS419 and CD3D loci. Nucleic Acids Res., 18: 4036, 1990. 28. Weber, J. L., Kappel, C., May, P. E., and Kwitek, A. E. Dinucleotide repeat comments. polymorphism at the D19S75 locus. Nucleic Acids Res., 18: 4639, 1990. 29. Nakamura, Y., Larsson, C., Juliet, C., Bystrom, C., Skogseid, B., Wells, S., REFERENCES Oberg, K., Carlson, M., Taggart, T., O'Connel, P., Leppert, M., Lalouel, J-M., Nordenskj61d, M., and White, R. Localization of the genetic defect in 1. Brandi, M. L., Marx, S. J., Aurbach, G. D., and Fitzpatrick, L. A. Familial multiple endocrine neoplasia type I within a small region of chromosome 11. multiple endocrine neoplasia type I: a new look at pathophysiology. Endocr. Am. J. Hum. Genet., 44: 751-755, 1989. Rev., 8: 341-405, 1987. 30. Janson, M., Larsson, C., Werelius, B., Jones, C., Glaser, T., Nakamura, Y., 2. Larsson, C., and Nordenskjold, M. Multiple endocrine neoplasia. Cancer Jones, C. P., and Nordensjk61d, M. Detailed physical map of human chro- Surv., 9: 703-723, 1990. mosome region 1 lq12-13 shows high meiotic recombination rate around the 3. Rizzoli, R., Green, J., III, and Marx, S. J. Primary hyperparathyroidism in MENI locus. Proc. Natl. Acad. Sci. USA, 88: 10609-10613, 1991. familial multiple endocrine neoplasia type I: long term follow up of serum 31. Richard, C. W., Withers, D. A., Meeker, T C., Maurer, S., Evans, G. A., calcium levels after parathyroidectomy. Am. J. Med., 78: 467-474, 1985. Meyers, R. M., and Cox, D. R. A radiation hybrid map of the proximal long 4. Marx, S. J., Vinik, A. I., Santen, R. J., Flyod, J. C., Jr., Mills, J. L., and arm of human chromosome 11 containing the multiple endocrine neoplasia Green, J., lII. Multiple endocrine neoplasia type I: assessment of laboratory type I (MEN-I) and Bcl-l disease loci. Am. J. Hum. Genet., 49:1189-1196, tests to screen for the gene in a large kindred. Medicine, 66." 226-241, 1986. 1992. 5. Larsson, C., Skogseid, B., Oberg, K., Nakamura, Y., and Nordenskj61d, M. 32. Fujimori, M., Wells, S.. A., and Nakamura, Y. Fine scale mapping of the Multiple endocrine neoplasia type I gene maps to chromosome 11 and is lost gene responsible for multiple endocrine neoplasia type 1 (MENI) Am. J. in insulinoma. Nature (Lond.), 332: 85-87, 1988. Hum. Genet., 50: 399-403, 1992. 6. Bale, S. J., Bale, A. E., Stewart, K., Dachowski, L., McBride, O. W., Glaser, 33. Bale, A. E., Norton, J. A., Wong, E. L., Fryburg, J. S., Maton, P. N., T., Green, J. E., Ill, Mulvhill, J. J., Brandi, M. L., Sakaguchi, K., Aurbach, Oldfield, E. H., Streeten, E., Aurbach, G. D., Brandi, M-L., Friedman, E., G. D., and Marx, S. J. Linkage analysis of multiple endocrine neoplasia type Spiegel, A. M., Taggart, R. T., and Marx, S. J. Allelic loss on chromosome I with INT2 and other markers on chromosome 11. Genomics, 4: 320-322, 11 in hereditary and sporadic tumors related to familial multiple endocrine 1989. neoplasia type I. Cancer Res., 51:1154-1157, 1990. 7. Larsson, C., Shepard, J., Nakamura, Y., Blomberg, C., Weber, G., Werelius, 34. Henson, V., Palmer, L., Banks, S., Nadeau, J. H., and Carlson, G. A. Loss of B., Hayward, N., Teh, B., Tokino, T., Seizinger, B., Skogseid, B., Oberg, K., heterozygosity and mitotic linkage maps in the mouse. Proc. Natl. Acad. Sci. and Nordensjk61d, M. Predictive testing for multiple endocrine neoplasia USA, 88: 6486-6490, 1991. type 1 using DNA polymorphisms. J. Clin. Invest., 89: 1344-1349, 1992. 35. Dryja, T. P., Rapaport, J. M., Joyce, J. M., and Peterson, R. A. Molecular 8. Friedman, E., Sakaguchi, K., Bale, A. E., Falchetti, A., Streeten, E., Zimer- detection of deletions involving band q14 of chromosome 13 in retinoblas- ing, M. B., Weinstein, L. S., McBride, W. O., Nakamura, Y., Brandi, M. L., toma. Proc. Natl. Acad. Sci. USA, 83: 7391-7393, 1986. Norton, J. A., Aurbach, G. D., Spiegel, A. M., and Marx, S. J. Clonality of 36. Rouleau, G. A., Wertelecki, W., Haines, J. L., Hobbs, W. J., Trofatter, J. A., parathyroid tumors in familial multiple endocrine neoplasia type I. N. Engl. Seizinger, B. R., Martuza, R. L., Superneau, D. W., Conneally, P. M., and J. Med., 321: 213-218, 1989. Gusella, J. P. Genetic linkage of bilateral acoustic neurofibromatosis to a 9. Thakker, R. V., Bouloux, P., Wooding, C., Chotai, K., Broad, P. M., Spurr, DNA marker on chromosome 22. Nature (Lond.), 329: 246-248, 1987. N. K., Besser, G. M., and O'Riordan, J. L. H. Association of parathyroid 37. Miki, Y., Nishisho, l., Miyoshi, Y., Utsunomiya, J., and Nakamura, Y. tumors in multiple endocrine neoplasia type I with loss of alleles on chro- Interstitial loss of the same region of 5q in multiple adenomas and a carci- mosome 11. N. Engl. J. Med., 321: 218-224, 1989. noma derived from an adenomatous polyposis coli (APC) patient. Genes, 10. Bystrom, C., Larsson, C., Blomberg, C., Sandelin, K., Falkmer, U., Skogseid, Chrom. Cancer, 4: 81-83, 1992.

Eitan Friedman, Luiz De Marco, Pablo V. Gejman, et al.

Cancer Res 1992;52:6804-6809.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/52/24/6804

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://cancerres.aacrjournals.org/content/52/24/6804. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.