| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
EXTENSIVE CLINICAL EXPERIENCE |
Department of Molecular Medicine and Surgery (E.T., M.N.), Karolinska Institutet, 171 76 Stockholm, Sweden; Department of Clinical Genetics (U.G.), Karolinska University Hospital Solna, 171 76 Stockholm, Sweden; Department of Endocrinology, Metabolism and Diabetes (E.L.), Karolinska University Hospital Huddinge, 141 86 Stockholm, Sweden; Department of Medicine (G.T.), Section of Endocrinology, University Hospital, 581 83 Linköping, Sweden; and Department of Medical Sciences (B.S.), Uppsala University, Uppsala University Hospital, 75185 Uppsala, Sweden
Address all correspondence and requests for reprints to: Emma Tham, Department of Molecular Medicine and Surgery, Center of Molecular Medicine L8:02, Karolinska University Hospital Solna, 171 76 Stockholm, Sweden. E-mail: emma.tham{at}karolinska.se.
| Abstract |
|---|
|
|
|---|
Objective: The goal of this study was to determine the MEN1 mutation spectrum and detection rate among Swedish patients and identify which patient categories should be tested for MEN1 mutations.
Design/Setting/Patients: DNA sequences and referral forms from patients referred to the Department of Clinical Genetics at Karolinska University Hospital, Sweden, for clinical MEN1 mutation screening were analyzed. The mutation status of 371 patients (including 200 probands) was ascertained, and the multiplex ligation-dependent probe amplification (MLPA) assay was evaluated for the detection of large deletions.
Main Outcome Measure: The main outcome measure was MEN1 genotypes.
Results: Forty-eight of 200 index cases (24%) shared 40 different mutations (18 novel). A total of 69% of all mutations resulted in a truncated protein. Two large deletions were detected by MLPA. A total of 94% of all MEN1 families had a mutation in the coding region of the MEN1 gene. A total of 6% of sporadic cases had MEN1 mutations. There was no correlation between severe disease and mutation type or location.
Conclusions: A total of 4% of all mutations were large deletions, and MLPA is now included in our standard MEN1 mutation screening. Individuals with at least one typical endocrine tumour and at least one of the following: 1) a first-degree relative with a major endocrine tumor; 2) an age of onset less than 30 yr; and/or 3) multiple pancreatic tumors/parathyroid hyperplasia were most likely to harbor a mutation; thus these patients should be screened for MEN1 mutations.
| Introduction |
|---|
|
|
|---|
Only a couple of reports on testing for MEN1 mutations in the clinical setting have been published (4, 5, 6). Here we report the results of clinical testing for MEN1 mutations in Sweden.
| Subjects and Methods |
|---|
|
|
|---|
During the period 1997 to July 2006, a total of 371 patient blood samples were referred to the Department of Clinical Genetics at the Karolinska University Hospital for MEN1 mutation testing as part of the clinical health service. The samples were paid for by the tax-financed health care insurance. Two samples were excluded from this study because mutations were found in the CASR (RefSeq NM_000388) and VHL (RefSeq NM_000551) genes, respectively. Of the 369 samples, 200 were from nonrelated probands. Clinical information was obtained by the accompanying referral form with additional information provided by the referring physician if needed (i.e. retrospectively). Clinical information was available for 366 individuals. No systematical clinical screening was performed, thus occult tumors, the presence of multiple tumors, and exact tumor hormone production may have been missed. The patients age is age at diagnosis of their first lesion, or if that was not available, age at referral. This study was reviewed by the local ethics committee.
Because all samples analyzed at the Department of Clinical Genetics must fulfill clinical quality standards, the first family member referred to the clinic was denoted "proband" in this study, even if another proband had been previously analyzed as part of a research project. In three cases, these "probands" were presymptomatic carriers of the family mutation.
Mutation testing
The coding region (exons 2–10) of the MEN1 gene (RefSeq NM_130799) was screened for mutations by DNA sequencing using standard procedures in all index cases with an unknown mutation. If the family mutation was known, only the relevant exon was sequenced. All mutations were confirmed by bidirectional sequencing on two independent PCR samples and, if possible, confirmed in another affected family member. Missense mutations that had not been previously reported were considered deleterious if they altered highly conserved base pairs and segregated with the disease in those cases for which more than one family member was available for study. Detailed methods and all primer sequences are available upon request.
Multiplex ligation-dependent probe amplification (MLPA) (MRC-Holland, Amsterdam, The Netherlands) for MEN1 was performed according to the kit instructions to screen for larger genetic alterations within the gene (7).
All mutations are named in accordance with the Human Genome Variation Society (http://www.hgvs.org/mutnomen), with position +1 at ATG of RefSeq NM_130799.
Screening for polymorphisms
DNA was isolated from peripheral blood from 95 Swedish blood donors of unknown identity using standard procedures. Both alleles from exons 1, 3, and 10 were sequenced to determine the population frequency of nonpathogenic amino acid substitutions.
Allele analysis
Markers D11S4946 (intragenic in 5'UTR) and D11S4940 (93-kb 5' of the MEN1 gene) were used to detect allele sharing between index cases with the same mutation (8). PCR products were run on a POP4 sequencing gel (Applied Biosystems, Foster City, CA) together with a TAMRA-labeled DNA ladder (Applied Biosystems). Allele sizes were determined using GeneMapper v3.7 (Applied Biosystems).
| Results |
|---|
|
|
|---|
Mutations in the MEN1 gene were found in 48 of the 200 index cases (24%), including 32 of our 34 MEN1 families (94%) (Fig. 1
; clinical information can be found in supplemental Table 1, which is published as supplemental data on The Endocrine Societys Journals Online web site at http://jcem.endojournals.org). Of these, 40 were independent mutations, spread across the gene. There were 18 mutations that had not been previously reported [The Human Gene Mutation Database at the Institute of Medical Genetics in Cardiff (www.hgmd.org)]. A total of 33 mutations (69%) caused a truncated or absent protein. All 11 missense mutations altered conserved amino acids (Fig. 1
). In one family, all affected individuals had two amino acid substitutions in exon 3. Neither substitution was found in normal individuals or healthy family members. Of our families, 12 have been previously published, although the c.207_208insC and Gln166X mutations were incorrectly denoted as c.311insG and Ala160Thr in Teh and colleagues (9, 10, 11).
|
|
In five cases, initial mutation screening failed to identify a mutation in the MEN1 gene, in part due to human error and in part due to suboptimal quality of the DNA sequences. After reanalysis and/or resequencing, the mutation was unequivocally identified [Glu45Lys, 312_315delCCTC (n = 2), 1050–1G>A and 1351–1_-4del6]. Since 2005, all sequences are scanned both by a computer program and manually, and so far, no mutations have been missed.
Clinical characteristics of the index cases
Of the 200 index cases, 62% were female, and the median age was 44 yr (range 7–86). Clinical information was available on 199. The index cases with a MEN1 mutation or genetically linked to the MEN1 locus (16) tended to be younger (median age 33 yr, range 9–71, n = 49) compared with those without a mutation (median age 50 yr, range 7–86, n = 150). A total of 58 probands had heredity for endocrine tumors. Of these, 34 belonged to a MEN1 family (94% with mutation), 22 had a first-degree relative with HPT (36% with mutation), and two had a first-degree relative with an endocrine pancreas or pituitary tumor (0% with mutation). Of the 141 sporadic index cases, eight had a MEN1 mutation (6% of all sporadic cases, representing 17% of all mutations). DNA was available from both parents in two cases; none had the mutation. In one additional case, thorough clinical screening did not detect any affected family members. Thus, at least three mutations (6%) were most likely de novo.
As expected, probands with three major tumors, HPT + EPT, or heredity had a higher risk of harboring a MEN1 mutation (Table 2
). The eight sporadic cases with a MEN1 mutation all had an age of onset younger than 30 yr and/or multiple tumors (i.e. confirmed parathyroid hyperplasia and/or multiple pancreatic islet/duodenal tumors).
|
A total of 169 relatives to the MEN1 probands were tested for mutation carrier status. Clinical information was available for 167 of them. There were 12 patients related to probands who lacked a MEN1 mutation, and as expected, none had mutations. A total of 119 relatives were presymptomatic, and 18% had MEN1 mutations. There were 37 younger than 18 yr, and eight of them had MEN1 mutations (22%). Of the 36 affected patients, 34 (94%) harbored MEN1 mutations (three were younger than 18 yr). The two patients in whom no MEN1 mutations were found had first-degree relatives with MEN1 mutations and an isolated increase in pancreatic polypeptide or chromogranin A with no other alterations reported.
Clinical characteristics of all referred patients with MEN1 mutations
A total of 106 tested individuals had MEN1 mutations (28%). Clinical information was available for 104, of which 87 were affected (Fig. 2A
). A total of 115 patients had tumors in at least two major locations (clinical MEN1), of whom 54 (47%) had a MEN1 mutation (Fig. 2B
).
|
At least one individual had three-organ disease in 18 families. A total of 79% had truncating mutations or large deletions, and 21% had missense mutations, all spread across the gene. There were 17 families that had metastatic disease, and 71% had truncating mutations in exons 2, 3, and 10. Four had missense mutations, and one had a splice mutation. Thus, there was no obvious correlation between mutation type and more severe disease. In addition, loss of the entire MEN1 gene did not cause a more severe phenotype than loss of nuclear localization sequence 2 only.
There was considerable variation in expression within families. For example, in one family, one individual died at 48 yr old due to malignant gastrinoma, whereas two first-degree relatives displayed only HPT at 41 and 63 yr of age. Two families were initially characterized as familial isolated hyperparathyroidism (FIHP), with six and seven members with HPT when the proband was referred for mutation screening. In one family, two patients later developed EPT (aged 67 and 73 yr), and one (52 yr) developed prolactinoma. The other family was previously published by Villablanca et al. as a FIHP family (11). The proband later developed bronchial carcinoid at 37 yr of age and two relatives with HPT and suspected EPT and one with HPT and PIT were later referred for mutation screening.
| Discussion |
|---|
|
|
|---|
In this study DNA sequencing of the coding sequence and MLPA have detected MEN1 mutations in 94% of our MEN1 families. There are 18 novel mutations and four represent mutational "warm-spots" that have been previously reported in over 10 different families (reference list published as supplemental data on The Endocrine Societys Journals Online web site at http://jcem.endojournals.org). Of all mutations, 4% were large deletions detected by MLPA. Similar frequencies of large genetic alterations in the MEN1 gene have been previously reported (13), and, therefore, MLPA is now included in our clinics routine analysis for MEN1 mutations. One family had two amino acid substitutions (Gly156Ser and Ala160Pro). Only Ala160Pro has been previously reported (17), although other substitutions of Gly156 have been found [www.hgmd.org (6)], and we later identified an additional family with a Gly156Cys mutation (Fig. 1
). Neither substitution was present in healthy individuals, thus, this family may have a double missense mutation. The two mutation-negative MEN1 families [one previously genetically linked to 11q13 (16)] most likely have mutations in the regulatory regions of the MEN1 gene. To date, few mutations have been reported in the promoter (18), 5' untranslated region (14) or 3' UTR (19) of the MEN1 gene, but these regions have rarely been screened.
MEN1 mutation frequencies compared with other studies
The overall mutation rate among our 200 index cases was only 24%. This is largely due to the low frequency of MEN1 mutation in sporadic cases (Table 2
). Approximately half the sporadic cases only had one endocrine tumor at referral. Of these, the three with a MEN1 mutation were 15, 17, and 43 yr of age, whereas the remaining 71 cases tended to be older (median 44 yr, range 7–82). The sporadic mutation-negative patients with HPT and EPT were also older (median 53 yr, range 25–86) compared with the single patient with a MEN1 mutation (17 yr). Of the nine mutation-negative patients with three major tumors, only eight had two confirmed tumors. They were also older (median 62 yr, range 40–71) compared with those with a mutation (median 32 yr, range 13–50). Thus, the majority of the sporadic patients may be phenocopies, although mutations in the noncoding regions of the MEN1 gene or perhaps in a downstream gene such as p27 (2) cannot be excluded. Of note, the combination of HPT + PIT, especially GH-producing PIT, has occurred as a phenocopy to MEN1 (20, 21). Indeed, none of our sporadic cases with HPT + PIT and only 50% of those with heredity had MEN1 mutations.
All eight sporadic cases with a mutation had an age of onset younger than 30 yr and/or multiple tumors (parathyroid hyperplasia and/or multiple pancreatic islet/duodenal tumors). Therefore, sporadic patients with these characteristics should be tested for MEN1 mutations. Similar guidelines have been suggested by the MEN1 consensus statement (3) and tested by Roijers et al. (22) on a much smaller set of patients in whom mutations were found in nine of 15 (60%) instead of the expected 5%.
Among the nonindex cases, MEN1 mutations were detected in 94% of the affected relatives. The two mutation-negative relatives are most likely not truly affected. Only 18% of the presymptomatic relatives harbored MEN1 mutations instead of the expected 50%. This may be because many mutation carriers had already developed symptoms, and a few of the relatives tested were not first-degree relatives.
Clinical characteristics of our MEN1 patients and families
Within our MEN1 families, there was considerable variation in expression as has been described previously (23, 24). Thus, all major MEN1 manifestations should be screened for, and not only the predominant family tumor type. This also applied to our two FIHP families, supporting the notion that FIHP sometimes represents a variant of MEN1 with reduced expression of other endocrine tumors (25).
Our youngest patient was operated on for insulinoma at 13 yr of age, supporting the suggestion by the MEN1 consensus statement to begin clinical screening from 5 yr of age (3). Of our patients, 81% had HPT as their first diagnosis, although this probably reflects the common use of serum calcium measurements in general health care practice (26). Of note, EPT were the presenting diagnosis in four (5%) of all patients with MEN1 mutations. In addition, other studies have reported mutations in up to 33% of familial cases with only EPT or only PIT (5, 6). However, none of our probands with only EPT or only PIT had MEN1 mutations. Overall, the proportions of tumor types in our MEN1 patients (Fig. 2
, A and B) were similar to previous studies (3, 13), although we had a higher frequency of insulinoma in the mutation-negative group.
Genotype-phenotype correlations
Menin is ubiquitously expressed and has a number of interacting partners (Fig. 1
), including transcriptional regulatory proteins, cytoskeletal proteins, and DNA processing and repair proteins. Our mutations were spread across the gene and do not perturb one particular partner of menin, as is the case for the more than 400 reported MEN1 mutations (27, 28). In addition, deletion of the whole MEN1 gene did not result in a more severe phenotype than single missense mutations. This may be because all mutations result in rapid degradation of menin (29, 30).
| Acknowledgments |
|---|
| Footnotes |
|---|
Disclosure Statement: The authors have nothing to disclose.
First Published Online July 10, 2007
Abbreviations: EPT, Enteropancreatic tumor; FIHP, familial isolated hyperparathyroidism; HPT, hyperparathyroidism; MEN1, multiple endocrine neoplasia type 1; MLPA, multiplex ligation-dependent probe amplification; PIT, tumor in the anterior pituitary.
Received March 1, 2007.
Accepted June 28, 2007.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
D M Lourenco Jr, R A Toledo, I I Mackowiak, F L Coutinho, M G Cavalcanti, J E M Correia-Deur, F Montenegro, S A C Siqueira, L C Margarido, M C Machado, et al. Multiple endocrine neoplasia type 1 in Brazil: MEN1 founding mutation, clinical features, and bone mineral density profile Eur. J. Endocrinol., September 1, 2008; 159(3): 259 - 274. [Abstract] [Full Text] [PDF] |
||||
Read all eLetters
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |