Multiple Endocrine Neoplasia Type 1 (MEN1) 5′UTR Deletion, in MEN1 Family, Decreases Menin Expression

Multiple endocrine neoplasia type 1 (MEN1) is an autosomal dominant disorder characterized by the occurrence of parathyroid, pancreatic and pituitary tumors, and is due to mutations in the coding region of the MEN1 gene, which encodes menin. We investigated a family with identical twins that had MEN1, with different MEN1 tumors. DNA sequence analysis of the MEN1 coding region had not identified any abnormalities and we hypothesized that deletions and mutations involving the untranslated regions may be involved. Informed consent and venous blood samples were obtained from five family members. Sanger DNA sequencing and multiplex ligation‐dependent probe amplification (MLPA) analyses were performed using leukocyte DNA. This revealed a heterozygous 596bp deletion (Δ596bp) between nucleotides −1087 and −492 upstream of the translation start site, located within the MEN1 5′ untranslated region (UTR), and includes the core promoter and multiple cis‐regulatory regions. To investigate the effects of this 5′UTR deletion on MEN1 promoter activity, we generated luciferase reporter constructs, containing either wild‐type 842bp or mutant 246bp MEN1 promoter, and transfected them into human embryonic kidney HEK293 and pancreatic neuroendocrine tumor BON‐1 cells. This revealed the Δ596bp mutation to result in significant reductions by 37‐fold (p < 0.0001) and 16‐fold (p < 0.0001) in luciferase expression in HEK293 and BON‐1 cells, respectively, compared to wild‐type. The effects of this 5′UTR deletion on MEN1 transcription and translation were assessed using qRT‐PCR and Western blot analyses, respectively, of mRNA and protein lysates obtained from Epstein‐Barr‐virus transformed lymphoblastoid cells derived from affected and unaffected individuals. This demonstrated the Δ596bp mutation to result in significant reductions of 84% (p < 0.05) and 88% (p < 0.05) in MEN1 mRNA and menin protein, respectively, compared to unaffected individuals. Thus, our results report the first germline MEN1 5′UTR mutation and highlight the importance of investigating UTRs in MEN1 patients who do not have coding region mutations. © 2020 The Authors. Journal of Bone and Mineral Research published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research (ASBMR).


Introduction
M ultiple endocrine neoplasia type 1 (MEN1) is an autosomal dominant disorder characterized by the combined occurrence of parathyroid, pancreatic and pituitary tumors. (1,2) The majority (>95%) of MEN1 patients will develop parathyroid tumors, 35% to 75% of patients will develop neuroendocrine tumors (NETs) of the pancreas (PNETs), <30% will have pituitary adenomas, and <25% will develop other endocrine tumors such as adrenal adenomas and carcinomas, and NETs of the thymus and lungs. (1) The majority (>90%) of MEN1 patients have germline mutations affecting the coding region of the MEN1 gene (2)(3)(4) (Fig. 1), which is located on chromosome 11q13. (5) The MEN1 gene spans 9 kilobases (kb) of genomic DNA that contain the 10 exons which encode a ubiquitously expressed 610 amino acid tumor suppressor protein, menin (NM_130799.2). The coding region of the MEN1 gene spans exons 2 to 10, while exon 1, the 5 0 region upstream of exon 2 and the 3 0 region of exon 10 are noncoding (6) (Fig. 1A). The 1.4 kb region upstream of exon 2 has been reported to have strong promoter activity (7) (Supplemental Fig. 1). More than 1800 germline and somatic MEN1 mutations have been reported in both familial and sporadic cases of MEN1, (3,4,8) and these are scattered throughout the nine coding exons of the MEN1 gene and show no genotype-phenotype correlation. (2,4,8) Approximately 25% of the mutations are nonsense mutations, 40% are frameshift mutations, 20% are missense mutations, 10% are splice-site mutations, (3,4) and 5% are rarer mutations, which include partial (9)(10)(11) or complete deletions (12,13) of the MEN1 gene. Frameshift, nonsense, and splice-site mutations are predicted to result in a truncated menin protein that lacks functional domains, or complete loss of protein due to nonsense mediated mRNA decay. (4) In contrast, missense mutations may affect functionally important amino acid residues involved in protein interactions, transcriptional activity, or protein stability. (4,8) To date, all reported mutations affect the 1830 bp protein-coding region. (4,8) However, 5% to 10% of MEN1 patients do not have mutations within the coding region of the MEN1 gene, (3,4,8) suggesting that these patients may harbor mutations in the untranslated regions (UTRs) of the MEN1 gene or they may represent phenocopies with mutations of other genes. (14) Here, we report a deletion involving the minimal core promoter region within the MEN1 5 0 UTR (Fig. 1B) in a MEN1 family with identical twins, who had phenotypic differences (15) (Fig. 2A).

Patients
The clinical details of the family, F/92 ( Fig. 2A), with MEN1 have been previously reported. (15) Briefly, the clinical findings of the three affected members were as follows. The father (I.1) of the identical twins (II.1 and II.2) had parathyroid adenomas, PNETs and, unusually for MEN1, a phaeochromocytoma. (15) One of the twins (II.1) was reported to have primary hyperparathyroidism (PHPT), due to parathyroid adenomas, at 16 years of age, and the other twin (II.2), who had a prolactinoma and presented with delayed puberty at the age of 15 years, was diagnosed to have PHPT at the age of 16 years. (15) During the next 25 years, the clinical history of these twins has been as follows. Twin II.1 had a parathyroidectomy with autotransplantation of parathyroids into the forearm at age of 22 years. He had recurrent PHPT that required surgery four times for removal of parathyroid tumors in the neck and the transplanted parathyroids in the forearm, . The introns are represented by broken lines connecting the boxes. The coding region (gray) of the MEN1 gene spans exons 2 to 10 while the 5 0 region upstream of exon 2, exon 1 and the 3 0 region of exon 10 are non-coding (white). The nucleotide positions of the translation start site (ATG), the translation stop site (TGA) and the intron-exon boundaries are indicated relative to the translation start site. Previously reported large germline deletions of the MEN1 gene that affect the protein-coding and promoter regions, including the 5 0 UTR and the translation start-site (ATG), are shown underneath by black lines with the references reporting these shown in parenthesis and the asterisk refers to this report. (10,11,24) (B) Schematic representation of the wild-type 842 bp MEN1 5 0 UTR from nucleotides −1158 to −317 (shown as a gray box) upstream of the translation start site (ATG), and the mutant 246 bp MEN1 5 0 UTR which lacks 596 bp from nucleotides −1087 to −492 (represented by a broken line). The positions of the oligonucleotide primers 5 0 UTRMEN1F and 5 0 UTRMEN1R, exon 1, the minimal core promoter and the translation start site are shown. between ages 27 and 41 years, after which he became hypoparathyroid and required treatment with oral alfacalcidiol. At the age of 25 years he was found to have a 9 mm PNET and at 30 years of age he had hypoglycaemia due to an insulinoma that was treated by a partial pancreatectomy. There has been no recurrence of the hypoglycemia but he has developed multiple pancreatic microadenomata of~3 mm size. Pituitary imaging studies indicate that he has not developed a pituitary tumor, and his pituitary hormones remain with the normal range. He developed a lipoma at the age of 28 years, which was excised.
In the other twin (II.2), the prolactinoma was treated with cabergoline and the PHPT by partial parathyroidectomy and with autotransplantation of parathyroids into the forearm, at 24 years of age. He suffered from recurrent PHPT that required parathyroidectomy and three further neck explorations between the ages of 30 and 40 years, following which he is normocalcemic. His only medication is cabergoline and he is normo-prolactemic. At the ages of: 34 years, he was found to have a 7 mm non-functioning PNET at 34 years; 36 years a facial angiofibroma; and at 39 years had excision of an abdominal lipoma. He has not had pancreatic surgery and remains well. Blood samples for DNA sequence analysis of the MEN1 gene were obtained from these three affected members (I.1, II.1 and II.2; Fig. 2A) and two unaffected relatives (I.2 and II.3; Fig. 2A). Informed consent was obtained from individuals, using protocols approved by the local and national ethics committees (MREC/02/2/93).

Genetic analysis
Leukocyte DNA was extracted and used to amplify by the polymerase chain reaction (PCR) the coding regions and intron-exon boundaries of the MEN1 gene, using 15 primer pairs that span the nine coding exons of the MEN1 gene and the 18 intron/exon boundaries utilizing conditions, as described. (14) DNA sequence analysis of the gel-purified PCR products was performed using the BigDye Terminator Cycle Sequencing Kit (Life Technologies, Paisley, UK) and an ABI automated capillary sequencer (Applied Biosystems, Loughborough, UK), as described. (16) Sequences were compared with the MEN1 Ensembl reference sequence ENSG00000133895. Multiplex ligation-dependent probe amplification (MLPA) analysis was used to analyze variations in MEN1 gene copy number. An MLPA kit P244-C1 (MRC Holland, Amsterdam, The Netherlands) was used to detect deletions or duplications of exons of MEN1, as reported. (17) The DNA sequence of the 5 0 UTR of the MEN1 gene was determined using specific PCR primers (forward, 5 0 UTRMEN1F: 5 0 -AAAATTAGGC GGGAGTGGTG-3 0 ; and reverse 5 0 UTRMEN1R: 5 0 -CCGGGGC TGTAAAGTCTGT-3 0 ) (Sigma-Aldrich, Cambridge, UK) ( Fig. 1B;  Supplemental Fig. 1). The frequency of the MEN1 5 0 UTR deletion was assessed in: 1284 patients who were clinically diagnosed to have MEN1; and in 74,180 participants, who comprised affected and unaffected family members from all diagnostic categories within the rare disease arm of the 100,000 (100K) Genomes Project (100KGP). Data from the 100KGP were accessed using a custom shell script and bcftools (v1.9). The set of 74,180 structural variant call files (vcf), which had been called by a combination of Manta and Canvas algorithms, were filtered and scanned for deletions that intersect MEN1 and the immediately upstream region (chr11:64,803,000-64,814,000, B38).
Quantitative real-time PCR (qRT-PCR) Total RNA was extracted from EBV transformed lymphoblastoid cells using the RNeasy Mini Kit (Qiagen, Manchester, UK), as previously reported. (19) RNA concentration and quality was determined using a NanoDrop ND-1000 Spectrophotometer (NanoDrop Technologies, Loughborough, UK), and RNA integrity determined by agarose gel electrophoresis. Up to 1 μg of total RNA was converted to cDNA using the QuantiTect Reverse Transcription Kit (Qiagen), as previously reported. (20) qRT-PCR reactions were performed using QuantiTect MEN1 and GAPDH primer assays (Qiagen) and QuantiTect SYBR Green PCR Kit (Qiagen), on a Rotor-Gene Q Cycler (Qiagen), as previously described. (20) The relative expression of the target MEN1 cDNA, normalized to the housekeeper GAPDH mRNA, was determined using described methods. (21) Western blot Total protein was extracted from EBV transformed lymphoblastoid cells using 500 μL of ice-cold NP40 lysis buffer (150mM NaCl, 50mM Tris (pH 8.0); 1% Triton X-100 [vol/vol], and 1× Protease inhibitor tablet (Roche, Hertfordshire, UK), maintained in constant agitation at 4 C for 30 min, centrifuged for 10 min at 10,000 rpm (7840 g) and the supernatant collected. Protein concentration was determined using the Bradford assay, as previously described. (20) Protein samples were prepared in 4× Laemmli loading dye (BioRad, Hertfordshire, UK), boiled at 95 C for 5 min and resolved using SDS-PAGE gel electrophoresis. Samples were transferred onto PVDF membrane (PerkinElmer, Cambridge, UK), blocked in 5% non-fat dried milk powder (commercially available as Marvel) dissolved in PBS and incubated with 1:1000 rabbit anti-menin antibody (AB-303203; Abcam, Cambridge, UK) or 1:2500 rabbit anti-GAPDH (AB-9485; Abcam) in 5% milk/PBS-T overnight. Membranes were subsequently incubated with anti-rabbit horseradish peroxidase (HRP)-conjugated secondary antibodies (SC-2004; Santa Cruz Biotechnology, Dallas, TX, USA) and visualized using ECL Western blotting substrate (BioRad) on a Chemidoc XRS+ system (BioRad) and densitometry analysis performed using Image J software (NIH, Bethesda, MD, USA; https://imagej.nih.gov/ij/), as previously described. (20) Statistical analysis Data are expressed as mean AE standard error of the mean (SEM) and statistical analysis was performed using one-way ANOVA in which the Bonferroni correction for multiple testing was applied. (20) A value of p < 0.05 was considered significant for all analyses.

Results
Identification of a 596 bp deletion in the MEN1 5 0 UTR Mutational analysis of leukocyte DNA from individual II.2, using conventional Sanger DNA sequencing of the MEN1 coding region (exons 2 to 10) and intron/exon boundaries, did not identify any pathogenic variants. However, a heterozygous deletion of two probe sites within the 5 0 UTR of the MEN1 gene was identified using MLPA gene copy number analysis (Supplemental Fig. 1). The presence of this deletion was confirmed using PCR amplification across the two MLPA probe sites ( Fig. 1B; Supplemental Fig. 1), which produced a 842 bp wildtype fragment and 246 bp mutant fragment. Only the 842 bp wild-type fragment was observed in the two unaffected members of the family (I.2 and II.3) and two unrelated normal controls (N1 and N2) ( Fig. 2A). In contrast, the 246 bp mutant fragment was observed, and co-segregated with MEN1 in the three affected individuals (I.1, II.1 and II.2) from the family ( Fig. 2A). Sanger DNA sequencing of this region of the MEN1 5 0 UTR further confirmed the deletion to be 596 bp in size, and to be located between nucleotides −1087 and −492 upstream of the MEN1 gene translation start site (c.−590_−24+29del; Fig. 2B). In addition, this deleted region contained the whole of exon 1 (nucleotides −607 to −521; c.−110_−24), the minimal core promoter region (nucleotides −591 to−491; c.−94_−24 +30) and multiple cis-regulatory and initiator elements (located between nucleotides −700 to −592; c.−203_−95) ( Fig. 1B; Supplemental Fig. 1). This 5 0 UTR 596 bp deletion was not found in 75,464 individuals, which comprised 74,180 participants of the 100KGP, 135 MEN1 patients who had pathogenic MEN1 variants, and 1149 patients with MEN1-associated tumors, thereby demonstrating an allele frequency of 0% for this deletion in these cohorts.

Discussion
We report the first MEN1 mutation that does not involve an abnormality within the protein-coding region, but instead occurs in the promoter region and comprises a 596 bp deletion of the minimal core promoter within the MEN1 5 0 UTR (Figs. 1 and 2;  Supplemental Fig. 1). This deletion, which co-segregates with MEN1 in the family (Fig. 2), results in reduced MEN1 transcription and translation of the encoded protein, menin (Fig. 4), and provides interesting insights in promoter misregulation. Thus, our findings highlight the importance of examining the 5 0 UTR when investigating patients with MEN1, who do not have germline mutations in the protein-coding region of the MEN1 gene, to establish a genetic diagnosis.
Large germline deletions of MEN1 that affect the proteincoding and promoter regions, including the 5 0 UTR, have been previously reported. (10,11,24) These large deletions include: a deletion located upstream of the MEN1 gene that results in loss of the translation start site and exons 2 to 5, in an Australian MEN1 kindred (24) ; a deletion of exons 1 and 2 in a Spanish MEN1 family (11) ; a deletion involving exons 1 to 2 in a Saudi MEN1 family (10) ; and a deletion of exons 1 to 3 in an Italian MEN1 family. (11) The MEN1 deletions reported in these families included both the MEN1 5 0 UTR as well as the ATG initiation codon, and part of the open reading frame (ORF), which could explain the loss of menin expression in the MEN1 tumors, as demonstrated by studies of parathyroid tumor and PNETs from the Italian MEN1 family. (11) However, the consequences of the sole loss of the minimal core promoter of the MEN1 5 0 UTR found in family F/92 (Fig. 2) on transcription and translation were difficult to predict as the region contains multiple cis-regulatory and initiator elements (Figs. 1 and 2; Supplemental Fig. 1). In addition, the 5 0 UTR region of MEN1 transcripts have been shown to be highly variable, as transcription can be initiated simultaneously from multiple initiator elements present in the 5 0 UTR in the absence of a TATA box, (7,25) although the heterogeneity of the 5 0 end of the MEN1 transcripts does not affect the menin protein because all transcript variants are reported to be correctly spliced to exon 2 and therefore contain the same ORF. (25) The location of the minimal promoter and the multiple cis-regulatory regions have been determined by expressing multiple 5 0 and 3 0 serial deletions of a 2.0 kb MEN1 gene promoter in different endocrine and non-endocrine cell lines and assessing responses by use of luciferase reporter assays. (7) The regulatory elements were predicted to have potential binding sites (eg, the CCAAT box) for transcription factors such as SP1, which is normally present in TATA-less promoters to activate transcription through the direct interaction with basic transcription machinery factors. (7,26,27) Furthermore, menin has also been reported to autoregulate transcription from the MEN1 promoter, (7) and our findings demonstrating that the loss of the minimal core promoter, due to the Δ596bp MEN1 5 0 UTR in the MEN1 family (Fig. 2), resulted in significantly decreased MEN1 promoter activity (Figs. 3 and 4), is consistent with observations from these previous studies. (7) However, the MEN1 gene is under complex transcriptional regulation, (7) and it seems possible that other transcription factor binding sites and regulatory elements within the 5 0 UTR of the MEN1 gene (Supplemental Fig. 1) may also have roles that could explain the observed >50% reduction in MEN1 RNA and menin expression in the patient derived EBV transformed cell lines. For example, the MEN1 5 0 UTR sequence contains two binding sites for paired box 5 (PAX5) (Supplemental Fig. 1), which is a master regulator of B-cell fate, (28) and EBV expresses a number of genes during infection, including the BamHI Z fragment leftward open reading frame 1 (BZLF1) gene that encodes a protein (Z), which is a transcription factor that directly interacts with PAX5 and inhibits the ability of PAX5 to activate promoters. (29)(30)(31) PAX5 may be acting as a repressor of MEN1, and its action may be lifted by EBV induced Z expression, which would result in an increase in menin expression in the wild-type MEN1 expressing cell lines relative to the heterozygous MEN1 expressing cell lines, thereby potentially contributing to the observed marked differences in MEN1 expression (Figs. 4 and 5). In addition, menin and EBV proteins may have opposite actions on tumor protein 53 (p53) and retinoblastoma (Rb) activity, (32,33) as expected of tumor suppressors and oncogenes, respectively, and investigations of the possible roles of these pathways and their components in controlling menin expression, may help to advance knowledge of the mechanisms regulating MEN1 transcription and translation.
Reduced (or absent) MEN1 transcription and translation could potentially be due to deletion of the wild-type MEN1 allele, which would lead to a loss of heterozygosity (LOH), in the patientderived EBV transformed cell lines. However, this would seem an unlikely possibility to explain the >50% reduction in MEN1 and menin expression that was observed in association with the MEN1 mutations in the EBV transformed cell lines from the patients (Figs. 4 and 5), because of the following three reasons: (i) the EBV transformed cell lines had undergone eight or fewer passages, thereby yielding a negligible likelihood of any secondary rearrangements; (ii) the probability of all eight patientderived EBV transformed lymphoblastoid cell lines acquiring an identical second genetic abnormality with similar levels of mosaicism would be extremely low; and (iii) MEN1 RNA and menin expression, which is still detected in EBV transformed lymphoblastoid cell lines from affected members of the MEN1 family (Fig. 4), would not be observed if there was LOH of the wild-type MEN1 allele together with the 596 bp MEN1 promoter deletion that completely abolishes transcription and translation (Fig. 3). These observations indicate that the residual MEN1 transcription of >15% (Figs. 4A and 5A) and menin expression of >10% to 35% (Figs. 4B,C and 5B,C) in these EBV lymphoblastoid cells likely comes from an intact wild-type allele.
Mutations affecting the 5 0 UTR of other tumor suppressor genes have also been reported in other endocrine cancer syndromes such as the hyperparathyroid-jaw tumor (HPT-JT) syndrome and MEN4, which are due to mutations of the cell division cycle 73 (CDC73) and cyclin-dependent kinase inhibitor 1B (CDKN1B) genes, respectively. Thus, a heterozygous insertion (c.−4_−11insG) in the CDC73 5 0 UTR occurring in cis with a loss of CDC73 exons 4 to 10 in a family with HPT-JT, revealed it to impair luciferase reporter activity and reduce in vitro translation of the parafibromin protein. (34) However, it was the large intragenic deletion of exons 4 to 10 rather than the 5 0 UTR variant that was likely responsible for the phenotype in this family. (34) In another example involving a heterozygous 4-bp deletion (c. −32_−29delGAGA) in the 5 0 UTR of the CDKN1B gene, the deletion was reported to significantly reduce luciferase reporter activity and decrease transcription of CDKN1B in the MEN4 patient's leukocytes. (35) However, another heterozygous 4-bp deletion (c.−456_−453delCCTT) in the 5 0 UTR of the CDKN1B gene and a point mutation (c.−469C>T) that affected the consensus Kozak sequence for initiation of translation of CDKN1B mRNA in MEN4 patients reduced the level of p27, encoded by CDKN1B, without affecting CDKN1B mRNA levels. (36,37) These findings suggest that germline 5 0 UTR non-coding mutations that do not affect the ATG initiation codon can affect transcription and/or translation and can have a role in cancer susceptibility.
Both of the identical monozygotic twins in our study (Table 1, study A) had the 596 bp deletion (c.−590_−24+29del) affecting the MEN1 5 0 UTR, but one twin (II.1) had parathyroid adenomas, insulinoma, a non-functioning PNET, and a lipoma, whilst the other (II.2) had parathyroid adenomas, a prolactinoma, a nonfunctioning PNET, a lipoma, and an angiofibroma ( Fig. 2A; Table 1). (15) Such phenotypic variability in the MEN1 syndrome is commonly observed and has also been reported to occur in monozygotic twins from six other studies (Table 1, studies B-G). (38)(39)(40)(41)(42)(43) Thus, of the seven studies (Table 1, studies A-G) of monozygotic twins, six (studies A, B, C, D, F and G) report differences in the occurrence of tumors between the identical twins, thereby indicating a lack of correlation between the phenotype and genotype in the majority (>85%) of twin pairs. (15,(38)(39)(40)42,43) Six of these seven studies (A, B, C, D, E and F, Table 1) reported an identified MEN1 mutation, and three of these mutations comprised deletions of 1 bp to 10 bp and were predicted to result in frameshifts with premature truncations (Table 1, studies B, C and F); one resulted in a nonsense mutation (Lys559Stop, study E, Table 1); another lead to a missense mutation (Tyr351His, study D, Table 1); in addition to the 5 0 UTR deletion of 596 bp described here (Table 1, study A). (15,(38)(39)(40)(41)(42) Moreover, these six MEN1 mutations are scattered throughout the 10 exons of the MEN1 gene and highlight the lack of genotype-phenotype correlation in the MEN1 syndrome, suggesting that genetic factors might not be the sole contributors affecting the expression of MEN1 but that other non-heredity factors, such as epigenetics, may be involved. The one study (Table 1, study G) not reporting a MEN1 mutation was published 10 years prior to the identification of the MEN1 gene. (43) (38) c.1561_1571del (exon 10) Thr521fs PHPT a , GAS PHPT a (C) Concolino and colleagues (39) c.292delC (exon 2) Arg98fs PHPT a , INS AS (D) Rix and colleagues (40) c.1051T>C (exon 8) Tyr351His PHPT a , ACTH PHPT a (E) Tso and colleagues (41) c.1675A>T (exon 10) Lys559Stop PHPT a , INS PHPT a , INS (F) Namihira and colleagues (42) c.459delC (exon 3) Asp153fs PHPT a , PNET(NF) PHPT a (G) Bahn and colleagues (43) None reported b Unknown PHPT, PRL, ACTH, GAS PHPT a , PRL Study published 10 years prior to identification of MEN1 gene in 1996. (9) The morbidity and mortality amongst MEN1 patients has reduced considerably since the introduction of DNA testing in 1997, highlighting the importance of identifying patients at higher risk of developing MEN1 and to provide them with regular screening programs with early appropriate clinical interventions. (4,8,44) There are over 1800 germline and somatic MEN1 mutations that have been described in MEN1 patients, all of which affect the protein-coding part of the MEN1 gene. (4,8) However, there are also 10% to 20% of MEN1 patients in whom mutations have not been detected within the protein-coding region of the MEN1 gene, (4,8,44,45) and our results indicated that mutational analysis of non-coding regions of the MEN1 gene is warranted in such patients. Thus, the detection of a MEN1 mutation in such patients will help confirm the diagnosis of MEN1, and also identify those relatives who have inherited the MEN1 mutation and are therefore at high risk of developing tumors that can be found earlier by screening programs that use plasma biochemical analysis and radiological imaging studies. (44) In summary, our study reports the first germline MEN1 5 0 UTR mutation, which causes promoter misregulation, in a family with MEN1 and highlights the importance of investigating UTRs in MEN1 patients who do not have coding region mutations.

Disclosures
The authors have nothing to disclose and all authors state that they have no conflict of interest.