Multiple Endocrine Neoplasia Type 1 (MEN1) Syndrome

Francesca Marini; Alberto Falchetti; Ettore Luzi; Francesco Tonelli; Brandi Maria Luisa

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Riegert-Johnson DL, Boardman LA, Hefferon T, et al., editors. Cancer Syndromes [Internet]. Bethesda (MD): National Center for Biotechnology Information (US); 2009-.

Cancer Syndromes [Internet].

Show details

Contents

< Prev Next >

Multiple Endocrine Neoplasia Type 1 (MEN1) Syndrome

Francesca Marini, PhD, Alberto Falchetti, MD, PhD, Ettore Luzi, PhD, Francesco Tonelli, MD, and Brandi Maria Luisa, MD, PhD.

Author Information and Affiliations

Created: July 18, 2008; Last Update: August 9, 2008.

Introduction and Background

Multiple Endocrine Neoplasia type 1 (MEN1) is a rare hereditary endocrine cancer syndrome characterized primarily by tumors of the parathyroid glands (95% of cases), endocrine gastroenteropancreatic (GEP) tract (30-80% of cases), and anterior pituitary (15-90% of cases) (Figure 1) (1). Other endocrine and non-endocrine neoplasms including adrenocortical and thyroid tumors, visceral and cutaneous lipomas, meningiomas, facial angiofibromas and collagenomas, and thymic, gastric, and bronchial carcinoids also occur (Table 1 and Table 2) (2-8). The phenotype of MEN1 is broad, and over 20 different combinations of endocrine and non-endocrine manifestations have been described (9-13). MEN1 should be suspected in patients with an endocrinopathy of two of the three characteristic affected organs, or with an endocrinopathy of one of these organs plus a first-degree relative affected by MEN1 syndrome.

Figure 1.

Reproduced with permission from Gibril and others (1): “Venn diagram comparing the frequency alone or in combination with various MEN1-related abnormalities occurring in 1009 MEN1/ZES patients from the literature.”

Table 1.

Prevalence of endocrine tumors in MEN1 patients

Table 2.

MEN1-related non-endocrine tumors and their approximate prevalence

MEN1 patients usually have a family history of MEN1. Inheritance is autosomal dominant; any affected parent has a 50% chance to transmit the disease to his or her progeny. MEN1 gene mutations can be identified in 70-95% of MEN1 patients.

Many endocrine tumors in MEN1 are benign and cause symptoms by overproduction of hormones or local mass effects, while other MEN1 tumors are associated with an elevated risk for malignancy. About one third of patients affected with MEN1 will die early from an MEN1-related cancer or associated malignancy. Entero-pancreatic gastrinomas and thymic and bronchial carcinoids are the leading cause of morbidity and mortality. Consequently, the average age of death in individuals with MEN1 is significantly lower (55.4 years for men and 46.8 years for women) than that of the general population.

Epidemiology

MEN1 syndrome occurs in approximately one in 30,000 individuals with an equal sex distribution, and there is no ethnic group or racial predilection (14). Patients have been diagnosed between 8-81 years of age, but diagnosis before the age of 10 is rare.

Historical background

The first case of what is known to be MEN was described by Erdheim in 1903 in a report of an autopsy of a patient with acromegaly and four enlarged parathyroid glands (15). About 20 years later, Cushing and Davidhoff reported the first patient with the classic MEN1 tumor triad (16). In 1953 Underdahl and others published the first review of MEN1 syndrome describing 14 cases (17).

Wermer in 1954 was the first to describe the MEN1 phenotype in an autosomal dominant inheritance pattern in a family in which a father and four of nine children were affected (18). The original name “Wermer’s syndrome” (19) was soon replaced by the currently used term MEN1 syndrome. The MEN1 clinical phenotype was only fully characterized in the 1960s when radioimmunoassays for endocrine hormones were developed (20-22).

In 1988 the MEN1 locus was mapped to chromosome 11q13 using recombinant DNA probes in two brothers with MEN1 syndrome (23). Follow-up linkage analysis in MEN1 families confirmed the 11q13 locus. The MEN1 gene was identified by positional cloning, and mutations in MEN1 were confirmed to cause the MEN1 syndrome in 1997.

Pathophysiology

MEN1 gene consists of ten exons, spanning about 10 kb, and encodes a 610 amino acid protein named menin (Figure 2). The first exon and the last part of exon 10 are not translated. A main transcript of 2.8 kb has been described in a large variety of human tissues (pancreas, thymus, adrenal glands, thyroid, testis, leukocytes, heart, brain, lung, muscle, small intestine, liver, and kidney); an additional transcript of approximately 4 kb has been detected in pancreas and thymus, suggesting a tissue-specific alternative splicing (24). Six alternative MEN1 transcripts with variation in the 5’-untranslated region (exon 1) have been described; none of them affect the coding region (25). The genomic region immediately upstream of MEN1 is saturated with repetitive elements, suggesting its promoter is either restricted to or is compromised. The latter explanation may account for the seven alternative transcription start sites observed. Downstream, there are only 281 base-pairs between the end of the MEN1 3’UTR and the adjacent gene MAP4K2. There is a conservation peak within the MEN1 3’UTR that is probably a regulatory element for MAP4K2.

Figure 2.

Panel a: Schematic representation of the MEN1 gene. White boxes are translated exons, hatched boxes are untranslated. The start (ATG) and stop (TGA) codons in exons 2 and 10, respectively, are indicated. NLS1, NLSa, and NLS2 are the nuclear localization (more...)

The CDKN1B gene/p27 protein and a MEN1-like condition (MENX)

The CDKN1B/p27^Kip1 gene was recently identified as a gene for a MEN1-like condition. CDKN1B was identified as a candidate gene for MEN1 based on animal studies in the MENX rat model with mutations in Cdkn1b (26). Cdkn1b is homologous to CDKN1B, and sequencing of CDKN1B in MEN1 patients without MEN1 mutations has identified a few patients with mutations. CDKN1B is a recessive MEN1-like syndrome presenting with variable clinical manifestations typical of both MEN1 and MEN2 such as pheochromocytomas, paragangliomas, thyroid tumors, and parathyroid and pituitary adenomas. A germline nonsense mutation (W76X) was found in a patient with a MEN1-like phenotype, characterized by acromegaly and primary hyperparathyroidism (PHPT), and no mutation of the MEN1 gene. In another patient with pituitary adenoma and PHPT, a 19-bp duplication in the exon 1 of the CDKN1B gene was found (27). The CDKN1B MEN1-like syndrome is rare. Ozawa and others (28) found no CDKN1B mutations in a total of 34 familial and sporadic cases of the parathyroid/pituitary variant of MEN1, and Owens and others (29) found no CDKN1B mutations in 69 patients with MEN1 phenotype. A review of all published data shows that only 1.6% of all tested cases of MEN1 have CDKN1B. Clinical testing for CDKN1A is available (www.geneclinics.org).

MEN1 follows the Two-Hit Model

MEN1 follows Knudson’s “two-hit” model for tumor suppressor gene carcinogenesis (30). The first hit is a heterozygous MEN1 germline mutation, inherited from one parent (familial cases) or developed in an early embryonic stage (sporadic cases) and present in all cells at birth. The second hit is a MEN1 somatic mutation, usually a large deletion, that occurs in the predisposed endocrine cell as loss of the remaining wild-type allele and gives cells the survival advantage needed for tumor development (31).

Heterozygous Men1^-/Men1⁺ mice develop endocrine tumors similar to those observed in human MEN1 patients between 6-9 months of life (Figure 3 and Figure 4) (32). Homozygous Men1^-/Men1^- mice died in utero at embryonic days 11.5-12.5.

Figure 3.

From Crabtree and others with permission (32): “Pancreatic lesions in Men1 ^TSM/+ mice. Panel a: H&E-stained section (x15) of pancreas from a 12-month-old mouse, showing normal islets (N), hyperplastic islets (1), hyperplastic islets with (more...)

Figure 4.

From Crabtree and others with permission (32): “Other lesions in Men1 ^TSM/+ mice. Panel a: H&E-stained section (x40) of parathyroid from a 12-month-old mouse with thyroid follicles (T), focal parathyroid dysplasia (FD), and parathyroid (more...)

MEN1 mutations in multiple endocrine neoplasia patients

MEN1 gene mutations can be identified in 70-95% of MEN1 patients and in about 20% of familial isolated hyperparathyroidism cases (33-36). Almost all patients are heterozygous for mutations. One affected family has been identified with individuals both homozygous and heterozygous for MEN1 mutations. In this family, there was no difference in disease history between the homozygous and heterozygous mutation carriers (37).

Patients without detectable mutations may have mutations in the regulatory or untranslated regions, or introns that remain to be investigated. A study of patients with sporadic and familial PHPT failed to detect mutations in the MEN1 gene 5’ promoter region (38). The failure to detect MEN1 mutations can also be explained if the mutation is a large deletion causing the loss of the whole gene or a whole exon (35, 39, 40), which is not revealed by polymerase chain reaction (PCR)-based analysis. Deletions or other gross rearrangements represent 1-3% of MEN1 mutations and can be detected by Southern blot analysis or multiplex ligation-dependent probe amplification (MLPA) (41, 42).

Since almost all MEN1 families investigated to date have tight linkage to the 11q13 locus, the presence of another gene associated with MEN1 is unlikely but possible if another tumor suppressor in this region was affected (43). Alternatively, a few such MEN1 mutation-negative patients may represent phenocopies such as the CDKN1B-associated parathyroid/pituitary variant of MEN1.

A recent MEN1 mutation update (44) reviewed over 1100 germline and 200 somatic mutations. Benign polymorphisms in MEN1 are shown in Table 3. MEN1 mutations are located along the entire 1830-bp coding region and flanking splicing sites of the MEN1 gene. There are no mutation hot spots (34-36, 39, 45, 46). Approximately 41% of MEN1 mutations are frameshift insertions and deletions, about 23% are nonsense mutations, 6% are in-frame deletions or insertions, 20% are missense mutations, 9% are splice site defects, and about 1% are whole or partial gene gross deletions (44). About 68% of identified missense mutations occur on an amino acid that is conserved among human, mouse, zebrafish, and Drosophila. Approximately half of the mutations are unique. The other half have been found to recur in apparently unrelated kindreds. The recurrence of these mutations might be due to repeated sequences that may undergo misalignment during replication. However, comparison of the clinical features in patients sharing the same mutations has not identified genotype/phenotype correlations.

Table 3.

Benign polymorphisms of MEN1 gene

More than two-thirds of all identified mutations (frameshifts, nonsense, and some splicing site mutations) predict a loss-of-function of menin, and therefore support the hypothesis that MEN1 is a tumor suppressor gene. Other splicing site mutations can lead to retention of incompletely spliced precursors, complete absence of transcripts, or appearance of aberrantly processed mRNAs from the creation of novel or cryptic splicing sites or from the loss of classical splicing sites. The remaining 20-30% of MEN1 mutations are missense mutations and in-frame deletions, potentially affecting the interaction-site of one or more menin partners, altering the capacity of menin to regulate the target promoters or favouring a rapid proteolytic degradation of menin. Missense menin transcripts have been shown to be present at reduced levels with respect to wild-type and benign polymorphic menin levels (34), suggesting a role for rapid proteolytic cleavage. The diminished levels of the mutant menin are due to rapid degradation via the ubiquitin-proteasome pathway. Mutants, but not wild-type menin, interact with both the molecular chaperone Hsp70 and with the Hps70-associated ubiquitin ligase CHIP that promote the ubiquitination of the menin mutants in vivo.

Menin: the MEN1 protein

Menin is a 610 amino acid (67Kda) nuclear protein, highly conserved from mouse (98%), rat (97%) and, more distantly, zebrafish (75%) and Drosophila (47%) (47-51). Human and mouse MEN1 amino acid sequences share 95.8% identity and 98.4% similarity. The average of amino acid conservation score across species is about 87, as shown in Figure 5. Analysis of menin amino acid sequence did not reveal homologies to any other known human or mammalian protein, sequence motif, or signal peptide. The absence of significant homology to any other protein complicates efforts to elucidate the functions of menin and the mechanisms of its tumor suppressor activity.

Figure 5.

Menin is highly conserved. Human and mouse MEN1 amino acid sequences share 95.8% identity and 98.4% similarity. Its amino acid conservation is shown across seven species. Colors denote amino acids with similar biochemical properties; an asterisk indicates (more...)

Since there are few clues to menin’s functions from its amino acid sequence or its mutation profile, most of what is known about its role is derived from in vitro studies. These studies show that menin is located primarily in the nucleus (52) and that menin has three independent nuclear localization signals (NLS1, NLSa, and NLS2) all in the C-terminus of the protein at amino acids 479-497, 546-572, and 588-608, respectively. None of MEN1 missense mutations or in-frame deletions (4, 35, 36, 39, 40, 45, 53-56) alters either of these NLSs. However, all truncating mutations induce lack of at least one of these NLSs. Menin nuclear localization suggests that this protein may have a role in the regulation of DNA transcription and replication, in cell cycle, or in maintenance of genome integrity. Recent studies have demonstrated that over-expression of menin in Ras-transformed NIH3T3 cell model reversed the transformed phenotype (57), inducing a decreased proliferation, a suppression of growth in soft agar and an inhibition of tumor growth in nude mice. There is increasing evidence that menin may act in DNA repair or synthesis, but the exact mechanism by which it does so is currently unknown. In recent years menin has been shown to interact with several proteins involved in transcription regulation, genome stability, and cell division and proliferation (Figure 2). Nevertheless, none of the menin partners or menin pathways has yet been proved to be critical in MEN1 tumorigenesis.

JunD

The first identified partner of menin was JunD, a transcriptional factor belonging to the AP1 transcription complex family and implicated in negative control of cell proliferation. The interaction between JunD and menin is mediated via the N-terminal transcription activation domain of JunD and the N-terminus (codons 1-40) and central domains (codons 139-242 and 323-428) of menin. The wild-type menin binds to and represses JunD-activated transcription via a histone deacetylase-dependent mechanism through association with mSin3A, a general transcriptional coreppressor (58, 59). These data suggest that menin may inhibit transcription of some genes via recruiting histone deacetylase to the promoters. The central region of menin (codons 371-387) contains an α-helical mSin3A-interacting domain that is important for recruiting the mSin3A-histone deacetylase complex to repress JunD transcriptional activity. This inhibition can be relieved by histone deacetylase inhibitors. Disruption of menin-JunD interaction seems to be a component of the mechanism of tumorigenesis in MEN1 syndrome.

Nuclear Factor-kB (NF-kB)

Menin directly interacts with three members of the nuclear factor NF-kB family of transcription regulators: NF-kB1 (p50), NF-kB2 (p52), and RelA (p65) (60). These proteins modulate the expression of various genes and are involved in oncogenesis of numerous organs. Menin interacts with NF-kBs by their central domain (codons 305-381) and represses NF-kB-mediated transcription. Particularly menin represses p65-mediated transcriptional activation on NF-kB sites in a dose-dependent and specific manner. Inhibition of NF-kB has been directly linked to apoptosis and delayed cell growth.

Transforming Growth Factor Beta (TGFβ) and Smads pathways

Menin interferes also with the Transforming Growth Factor beta (TGFβ) signaling pathways, interacting physically with the C-terminal domain of Smad3 and playing an important role in TGFβ-induced growth inhibition (61). Amino acids 101-195 of menin are crucial for this interaction. Menin seems to play an important role in supporting TGFβ and Smad3 transcriptional control of cell growth, and menin inactivation disrupts TGFβ-mediated transcription and growth inhibition, resulting in tumor formation. Alteration of the TGFβ signaling pathways seem to be involved in anterior pituitary, pancreatic, and parathyroid carcinogenesis (62). In anterior pituitary cells the inactivation of menin blocks TGFβ and activin signaling, antagonizing their growth-inhibitory properties. Activin acts as a potent pituitary cell growth and hormone release inhibitor, and these effects require wild-type menin via the Smad3 signaling pathway (63). Activin suppresses the expression of Pit-1, a pituitary transcription factor that plays an essential role in the development and growth of lactotrope cells and in the induction of prolactin (PRL) and growth hormone (GH) expression. The inactivation of menin by mutations blocks the activin signaling pathways in pituitary cells, resulting in an increased cell growth and an enhanced PRL and GH genes expression. In endocrine pancreatic cells menin appears to suppress insulin promoter activity and inhibits glucose-induced insulin secretion. Moreover, apoptosis seems to be higher in menin-expressing insulinoma cells, and deregulation of the menin-Smad3 interaction might result in apoptosis suppression in insulinoma cells (64). TGFβ is also an important negative regulator of parathyroid cell proliferation and parathyroid hormone (PTH) secretion. Depletion of menin disrupts normal TGFβ signaling, leads to loss of the antiproliferative effects of TGFβ and, thus, probably contributes to parathyroid tumorigenesis (65).

The interaction of menin with Smads pathways is also important for the involvement of menin in bone development and osteoblast commitment and later differentiation. Recent finding demonstrated that menin is important for both early differentiation of osteoblasts and inhibition of their later differentiation, and that it might be crucial for intramembranous ossification. Menin promotes the commitment of multipotential mesenchymal stem cells into osteoblast lineage through the interaction with the BMP-2-Smad1/5-Runx2 cascade at level of BMP-2 and Smad1/5 (66). When wild-type menin is lost, the differentiation of multipotential mesenchymal stem cells into osteoblast lineage is inhibited. Furthermore, the interaction of menin with Smad3 inhibits later osteoblast differentiation by negatively regulating the BMP2-Runx2 cascade, at Smad1/5 and Runx2 level, after the commitment to the osteoblast lineage (67).

Nm23H1

Menin interacts (codons 1-486) with the putative metastasis suppressor Nm23H1 in mammalian cells (68). This interaction enables menin to act as an atypical GTPase stimulated by Nm23H1 and to hydrolyze GTP. At the moment it is not known if the GTP-hydrolyzing activity of menin plays key roles in transcriptional regulation, but it is an interesting possibility. Alteration or instability of menin might lead to loss of GTP hydrolysis activity that is a possible mechanism of MEN1 tumor formation. The binding of menin to Nm23H1 may be relevant also to the control of genomic stability, as Nm23H1 is associated to the centrosome that is involved in the maintenance of chromosome integrity. This may be supported by the fact that normal cells from MEN1 patients demonstrate an elevated level of chromosome alterations (69-72) and that MEN1 tumors have more genome aberrations than equivalent tumors from non-MEN1 patients (73).

Homeobox-containing protein Pem

Even the rodent protein Pem has been shown to bind menin directly (codons 278-476) (74). Pem is a homeobox-containing protein which plays a role in the regulation of transcription. However, since Pem sequence has no known homolog in the human genome its direct relevance to MEN1 in humans is still controversial. Mouse and human menin are very similar and this could suggest the existence of a human protein with a function similar to that of Pem, which binds menin and thus exerts its cellular functions. In adult mice Pem is produced by testicular Sertoli cells during androgen-dependent stages of spermatogenesis. Pem is also expressed in the region of epididymis in which the spermatozoa gain forward mobility and fertilization competence. It’s probable that Pem regulates a subset of androgen-dependent genes in the male reproductive system (75). In addition, temporal-spatial expression of MEN1 is similar to that of Pem and this suggests that in mice, menin may have a role in regulating spermatogenesis and sperm development. A similar role may be hypothesized in humans as infertility has been observed in male MEN1 patients homozygous for MEN1 mutations (76).

Insulin-like Growth Factor Binding Protein 2 (IGFBP-2)

Wild-type, but not mutant, menin is essential for repression of the endogenous Insulin-like Growth Factor Binding Protein 2 (IGFBP-2), a member of the IGFBP family that plays a crucial role in regulating both positive and negative cell proliferation, depending on cell type (77). Menin inhibits the activation of IGFBP-2 promoter by repressing the opening of the chromatin maybe via interaction with histone deacetylase. Both NLS1 and NLS2 are crucial for suppressing the expression of IGFBP-2 and mutations of even one of these NLSs compromise the role of menin in repressing expression of IGFBP-2.

Fanconi Anemia Complementation Group D2 Protein (FANCD2)

Menin specifically interacts (codons 219-395) with endogenous Fanconi Anemia Complementation Group D2 Protein (FANCD2), a protein involved in DNA repair and mutated in patients with Fanconi Anemia (78). γ-irradiation increases the interaction between menin and FANCD2, and loss of wild-type menin expression leads to increased sensitivity to DNA damage. In addition, γ-irradiation–induced DNA damage has been reported to increase menin concentration in the nucleus matrix. All together these data indicate that DNA repair and maintaining genome integrity might be possible functions of menin, as also confirmed by the evidence that lymphocytes from MEN1 patients, after treatment with DNA mutagenic agents, showed extensive chromosomal breakage and a premature centromere division not present in lymphocytes from normal controls (70, 79).

Replication Protein A2 (RPA2)

Menin was shown to interact with Replication Protein A2 (RPA2), a protein that can bind to single strand DNA (ssDNA) and is involved in DNA replication, recombination and repair and in the regulation of apoptosis and gene expression (80). Menin could influence RPA2 ssDNA-binding activity and, consequently, its functions in DNA repair, replication, or recombination. The interaction is mediated by menin N-terminal region (amino acids 1-40) and central region (amino acids 286-448) and the central domain of RPA2 (amino acids 43-171). It has been demonstrated that some MEN1 missense mutations abolish or impair the interaction between menin and RPA2.

Ataxia-Telangiectasia-mutated and Rad3-related (ATR) kinase

Ataxia-Telangiectasia-mutated and Rad3-related (ATR) kinase pathway represents a typical UV-induced DNA damage signaling pathway. Menin has been shown to respond to UV irradiation by localizing to the chromatin. Recently it has been demonstrated that this menin subnuclear localization is mediated through the ATR kinase pathway and its downstream target CHK1 (81). Particularly, an over-expression of active CHK1 that mimics the UV irradiation in the cells has been demonstrated to result in increased menin localization to the chromatin. Nevertheless, the exact molecular mechanism by which CHK1 leads menin to localize to the chromatin and the biological consequences of this localization are still unknown.

Activator of S-phase kinase (ASK)

Menin directly interacts with the Activator of S-phase kinase (ASK), an essential component of the Cdc7/ASK kinase complex that plays a crucial role in DNA replication and is essential for cell proliferation. The menin-ASK interaction is mediated by the C-terminus of menin (codons 558-610) and it is prevented by truncating menin mutations. ASK induces cell proliferation in the absence of menin while its function is inactivated by the presence of wild-type menin, as demonstrated in vitro by the fact that targeted disruption of menin enhances cell proliferation whereas complementation of menin-null cells with menin reduces cell proliferation (82). These data suggest that one way menin suppresses tumorigenesis may be via its interaction with ASK and subsequent inhibition of cell proliferation. However, the exact mechanism by which menin and ASK work together to regulate cell proliferation requires additional study.

Plasminogen activator inhibitor type 2 (PAI-2)

Plasminogen activator inhibitor type 2 (PAI-2) is a gastrin-induced gene. Although typically expressed at low levels in normal gastric mucosa, PAI-2 expression is highly increased in hypergastrinemic conditions. This factor has been implicated in the inhibition of urokinase-type plasminogen activator and subsequent decreasing of cell invasion (83), and it has also been proposed to act as an inhibitor of apoptosis (84). Wild-type menin inhibit PAI-2 induction by gastrin, possibly reducing its anti-invasive and anti-apoptotic effects at gastric mucosa level.

Mixed Lineage Leukemia (MLL)-histone methyltransferase complex

Recently menin has been demonstrated to associate in a histone methyltransferase complex with MML1 and MML2, two trithorax family proteins, other mammalian homologues of the yeast Set1 complex (Ash2L, Rbbp5, WDR5, and hDPY30) and RNA polymerase II (85-87). This MLL family-menin complex is structurally and functionally similar to yeast Set1 complex and it possesses a histone methyltransferase activity specific for histone H₃ lysine 4 (H₃K₄) and thus it exerts an epigenetic transcriptional activity (88-90). Wild-type menin targets the histone methyltransferase complex to methylate specific chromatin regions on H₃K₄, resulting in activation of target genes. The clustered homeobox genes, Hoxa9, Hoxc6, and Hoxc8, were the first identified target genes regulated by menin and MLLs (85, 88, 89). The regulation of Hox genes, particularly Hoxa9, seems to be an autonomous way by which wild-type menin regulates bone marrow hematopoiesis (91).

Moreover, Milne and others (92) demonstrated that menin-MLL complex induces transcriptional activation of the cyclin-dependent kinase (CDK) inhibitors p27^kip1 and p18^Ink4c, revealing a novel menin tumor suppressor function in cell cycle inhibition, particularly in the endocrine pancreas. The authors also evidenced that in vivo wild-type menin binds directly with the p27^kip1 and p18^Ink4c proximal promoters, and that expression of p27^kip1 and p18^Ink4c and H₃K₄ methylation were decreased in tumors from MEN1 patients compared with normal neuroendocrine tissues. More recently, a study evaluated the p27^kip1 protein expression in pancreatic islets from MEN1 mutant mice, confirming the altered p27^kip1 expression reported in tumors from MEN1 patients (93). Studies on knockout mice evidenced that the simultaneous loss of p27^kip1 and p18^Ink4c leads to a tumor spectrum very similar to that of human MEN1 patients including tumors of parathyroids, endocrine pancreas, pituitary, thyroid, stomach, and duodenum (94).

In addition, by increasing H₃K₄ methylation, menin acts as coactivator of estrogen receptor α (ERα)-mediated transcription of the estrogen-responsive TFF1 gene (95). These data establish menin as a critical link between activated ERα and H₃K₄ methylation, which acts as a regulator of ERα function. Inactivation of menin may lead to disruption of ERα-mediated transcription and may be implicated in MEN1 tumorigenesis. In the pituitary gland ERα has a direct effect on prolactin production, and it is expressed in pituitary adenomas.

Chromatin immunoprecipitation (ChIP) studies revealed that menin interacts with thousands of human gene promoters in many tissues, suggesting that menin acts as a global regulator of transcription (96). Menin chromatin occupancy is frequently associated with methylation of H₃K₄; however, menin binds many regions of chromatin independently of histone methyltransferase complex, suggesting that menin may also regulate transcription by cooperating with other currently unknown proteins. Genes whose promoters are bound by wild-type menin correlate with increased mRNA expression. A recent study (96) evaluated menin-occupied chromatin sites, using Serial Analysis of Chromatin Occupancy, founding that 32% of menin-occupied loci were in promoters, near the 3’ end of genes (14%) or inside genes (21%). A large number (33%) of menin-occupied sites were located outside known gene regions. Nevertheless, menin does not possess any known DNA-binding domain, and a specific menin binding sequence in DNA was not found, suggesting that many menin interactions with chromatin were indirect through association with other transcriptional regulators. Functional categories of genes near menin-occupied loci were also analyzed, revealing that the most represented categories were genes involved in cellular metabolism (51%), macromolecule metabolism (33%), nucleotide metabolism (24%), biopolymer metabolism (22%), and regulation of cellular physiological processes (22%). Genes implicated in cell organization and biogenesis (12%) and cell cycle (7%) were also represented but with a lower percentage (97).

Intermediate filament proteins

Although menin has been identified primarily as a nuclear protein, recent studies have reported its interaction with the glial fibrillary acid protein (GFAP) and with vimentin that are components of intermediate filaments. Menin and GFAP colocalize at the S–G2 phase of the cell cycle in glioma cells. This interaction may serve as a cytoplasmic sequestering network for menin at the S and early G2 phase of the cell cycle that binds menin, preventing its translocation to the nucleus and its target genes (98, 99). Menin could have an inhibitory role before the S phase starts, and it must be transferred to the cytoplasm to enable the S phase to proceed (100).

Clinical Genetics

MEN1 syndrome is highly penetrant. Fifty percent of patients develop signs and symptoms by 20 years of age and more than 95% have symptoms by 40 years of age (101). There is significant intra- and inter-familial variability in the age of onset, severity of disease, and tumor types. Despite numerous studies, no genotype-phenotype correlations have been established, suggesting that unknown genetic and environmental modifiers are involved in the expression of the MEN1 phenotype (35, 39, 40).

Laboratories advertise clinical testing of MEN1 by sequencing of all coding regions, mutation scanning, deletion duplication analysis, and linkage analysis. A list of laboratories offering MEN1 testing can be found at www.geneclinics.org.

Natural History

Patients most often present with multiple tumors of the parathyroids, anterior pituitary adenomas, and tumors of the neuroendocrine cells in the GEP tract, which constitute the “typical” clinical features of this syndrome. Other endocrine and non-endocrine lesions can also occur in varying combinations in patients. The only specific clustering of tumors within the MEN1 phenotype is the Burin variant in which the prevalence of prolactinoma is higher (40% vs. 22%, P < 0.01) and the prevalence of gastrinoma is lower (10% vs. 42%, P < 0.01) (102). MEN1-associated tumors are often clinically distinct from their sporadic counterparts due to an earlier age of onset and other unique clinical features (Table 4).

Table 4.

Comparison of clinical features of sporadic and MEN-associated tumors.

Parathyroid glands

PHPT is the most common endocrinopathy in MEN1, affecting nearly 100% of patients by age 50 (100). It is the first endocrine MEN1 manifestation in 90% of patients and may be recognized as early as age 8 in rare cases (102). MEN1 hyperparathyroidism can be differentiated from sporadic hyperthyroidism by its earlier age of onset (typically between 20 and 25 years of age versus 50 years) (101,103). Moreover, unlike the single adenomas of sporadic PHPT, PHPT in MEN1 is characterized by multiglandular hyperplasia and usually all parathyroids are affected. Interestingly, parathyroid carcinoma is less frequent in MEN1 than in sporadic PHPT. The clinical symptoms of the parathyroid hyperfunction in MEN1 are similar to those of sporadic PHPT, with a long period of asymptomatic hypercalcemia and a low morbidity. The most common clinical manifestations of hypercalcaemia are shown in Table 5. Hypercalcaemia may also increase gastrin secretion from a gastrinoma, precipitating or exacerbating symptoms of Zollinger-Ellison syndrome (ZES).

Table 5.

Most common clinical manifestations of hypercalcaemia.

Gastroenteropancreatic (GEP) tract neuroendocrine tumors

Endocrine GEP tumors occur in about 30-80% of MEN1 patients and are the second most frequent clinical manifestation of MEN1. Unlike sporadic GEP tumors, they are characterized by multiple nodular lesions that develop usually a decade earlier than their sporadic counterparts (104,105). The multiple adenomas, scattered throughout the whole pancreas, may be very numerous (up to 100 in some cases) and range in size from microadenomas slightly larger than unaffected islets to macroadenomas larger than 0.5 cm. Two thirds of these tumors produce excessive amounts of hormones (gastrin, insulin, somatostatin, glucagons, neurotensin, or vasoactive intestinal polypeptide (VIP)) and are associated with distinct clinical syndromes. The most common functional pancreatic tumors are gastrinomas (54%) and insulinomas (15%). Non-functional tumors and insulinomas are located within the pancreas, while gastrinomas are often found in the soft tissue around the pancreas and in the duodenal submucosa, but not in the mucosa where the gastrin-producing G cells are located.

Gastrinomas

These gastrin-secreting tumors represent about 54% of all functional GEP endocrine tumors in MEN1. Ninety percent are located in the duodenum. Zollinger-Ellison syndrome (ZES) refers to the constellation of clinical findings associated with increased gastric acid production caused by gastrin. Manifestations of ZES include esophagitis, vomiting, epigastric abdominal pain, chronic diarrhea, duodenal ulcers especially in the usual location of the second and third portion, jejunal ulcers, and weight loss. Approximately 40% of MEN1 patients have a gastrinoma that manifests with ZES symptoms which usually occurs before age 40 years (about one decade earlier than sporadic gastrinomas) (106). Gastrinomas in MEN1 are frequently multiple and most of them are malignant, with half having metastasized before diagnosis (106). Malignant gastrinomas represent the major cause of morbidity and mortality in MEN1 patients, principally due to duodenal and jejunal ulcers that may perforate. Poor prognosis is associated with primary pancreatic gastrinomas (more aggressive than duodenal gastrinomas, as suggested by their larger size and greater risk for hepatic metastasis), liver metastases, ectopic Cushing syndrome, and very high gastrin level. Nodal metastases do not seem to negatively influence prognosis.

Insulinomas

These β-islet cell insulin-secreting tumors arise in about 10% of MEN1 patients, often in association with gastrinomas. They usually occur in the third decade of life, one decade earlier than onset of sporadic insulinomas (106). MEN1 insulinomas can occur as single or multiple macroadenomas of about 1-4 cm in diameter and are almost always benign (107). Patients with MEN1 insulinoma present with hypoglycaemia that develops after a fast or exertion and improves after glucose intake. Unlike many patients with sporadic insulinoma, MEN1 patients with insulinomas are usually not obese.

Glucagonomas

These α-islet cell glucagon-secreting tumors have been reported in few MEN1 patients. They usually are a single macroadenoma larger than 3 cm (107). Glucagonomas can manifest with skin rash (necrolytic migratory erythema; Figure 6), venous thrombosis, anemia, diarrhea, anorexia, weight loss, stomatitis, hyperglycaemia, glucose intolerance, and hyperglucagonaemia.

Figure 6.

Cutaneous manifestation of MEN1 syndrome. Panel a: collagenomas; Panel b: facial angiofibroma; Panel c: lipoma; Panel d: glucagonoma-derived necrolytic migratory erythema.

VIPomas

These vasoactive intestinal peptide (VIP)-secreting tumors occur as WDHA syndrome characterized by watery diarrhea, hypokalaemia, and achlorhydria (108). VIPomas have been reported in only a few MEN1 patients.

PPomas

These tumors secrete pancreatic polypeptide (PP) and have been recognized in some MEN1 cases. Increased PP secretion has no known clinical significance.

Non-functional GEP tract tumors

These tumors are frequent in MEN1 syndrome affecting about 20% of patients.

Anterior pituitary tumors

Anterior pituitary adenomas have been reported to occur in 15 to 90% of MEN1 patients (106). They are the first manifestation of MEN1 in 25% of sporadic and 10% of familial cases. MEN1 anterior pituitary adenomas are usually single. They are invasive only in 10-15% of cases, and malignant degeneration is a very rare event. Symptoms depend on both the secreted pituitary hormone and/or compressive effects due to size of the tumor. Pituitary macroadenomas may compress optic chiasm causing bitemporal hemianopia and other visual field defects, blurred vision and headaches, or they compress the adjacent normal pituitary tissue inducing hypopituitarism. Approximately 60% of MEN1-associated pituitary tumors secrete prolactin (prolactinomas), 25% secrete GH, 3% secrete adenocorticotrophin (ACTH) causing hypercortisolism, and the others seem to be non-functional. The frequency of plurihormonal-secreting tumors is higher in MEN1 than in sporadic isolated pituitary tumors (109). The mean age at the time of diagnosis of MEN1 pituitary adenomas is about 40 years, similar to that for sporadic isolated pituitary tumors (101,109).

Prolactinomas

These prolactin-secreting tumors (with or without simultaneous GH over-secretion) are the most common pituitary tumors in MEN1, and symptoms include galactorrhoea, amenorrhoea, and infertility in women and hypogonadism, sexual dysfunction, reduction of libido, impotence, and, more rarely, gynecomastia in men.

GH-secreting tumors

The GH-secreting tumors are the second most frequent MEN1 anterior pituitary tumors after prolactinomas. The increased secretion of GH is responsible for the development of gigantism in children and acromegaly in adults.

Other MEN1-associated endocrine tumors

Adreno-cortical tumors

Adreno-cortical tumors, involving one or both adrenal glands, affect about 20-40% of MEN1 patients (3), usually occurring later in the course of the disease. The great majority of these tumors are non-functional and exhibit an indolent clinical course. However, rare functional adrenal cortical tumors secrete ACTH, and they are associated with elevated serum concentrations of cortisol, causing primary hypercortisolism, primary hyperaldosteronism, and Cushing’s syndrome.

Pheochromocytoma

This tumor affects less than 1% of all MEN1 patients (106) and it is always unilateral. However, it is appropriate to measure urinary catecholamine values prior to surgery to diagnose and treat a MEN1-associated pheochromocytoma and to avoid dangerous and potentially lethal blood pressure peaks during surgery.

Thyroid tumors

Thyroid tumors, consisting of adenoma, colloid goitres, and carcinomas have been reported to occur in over 25% of MEN1 patients (101). Nevertheless, the prevalence of thyroid disorders in the general population is high; thus, association of thyroid lesions in MEN1 patients may be incidental and not significant.

MEN1-associated non-endocrine tumors

Carcinoid tumors

These tumors are estimated to occur in about 10% of MEN1 patients and may be located in the gastrointestinal tract (type II gastric enterochromaffin-like (ECL) cells), the pancreas, the bronchi, or the thymus. Thymic carcinoids are more prevalent in males than in females (106), while bronchial carcinoids are more prevalent in females than in males. Cigarette smoking appears to be a risk factor for bronchial carcinoids (110). Most carcinoids are clinically silent. Thymic carcinoids may be aggressive and lethal, particularly in males who are smokers. Most bronchial carcinoids behave indolently, but there have been reported cases of local mass effects, metastasis and recurrence after resection (110). Rarely thymic, bronchial and gastric carcinoids over-secrete ACTH, calcitonin, GHRH, serotonin, or histamine and rarely cause the carcinoid syndrome associated with flushing attacks and dyspnoea.

Collagenomas and facial angiofibromas

Collagenomas have been reported in >70% of MEN1 patients (7) and they present as multiple, skin-coloured, sometimes hypopigmented cutaneous nodules. They manifest symmetrically arranged on the trunk, neck, and upper limbs. They are typically asymptomatic, roundish, firm-elastic and can range from few millimetres to several centimetres in size (Figure 6). Multiple facial angiofibromas have been observed in 40-90% of MEN1 patients (6). Half of patients with angiofibromas have five or more. They are benign tumors comprising blood vessels and connective tissue, and they consist of acneiform papules that do not regress and that may extend across the vermillion border of the lips (Figure 6).

Lipomas

Lipomas may occur in 20-30% of MEN1 patients (7). They are generally multiple benign fatty tissue tumors and can be subcutaneous or, rarely, visceral. Lesions, often multiple, can be small or large and cosmetically disturbing (Figure 6). When surgically removed they usually do not recur.

Leiomyomas

Leiomyomas of the esophagus, uterus, or rectum can occasionally occur in MEN1 patients. They are benign neoplasms derived from smooth muscle.

Central nervous system tumors: meningiomas and ependymomas

Meningiomas have been reported in about 8% of MEN1 patients, typically presenting later in life (111). Few cases of spinal ependymomas have been rarely (about 1% of patients) associated with MEN1 syndrome and with abnormalities at the 11q13 locus (112). Whether meningiomas or ependymomas occur with increased frequency in MEN1 patients compared to the general normal population is unclear.

Diagnosis

MEN1 clinical diagnosis is based on detection of MEN1-associated tumors and lesions and includes biochemical hormone evaluation, and endoscopic, nuclear medicine or other imaging studies. Examples of the most commonly used imaging procedures are shown in Figure 7. Biochemical and imaging diagnostic tools generally do not distinguish MEN1-associated from sporadic diseases. MEN1-associated tumors typically arise at a younger age than sporadic counterparts; thus, altered hormone values and clinical manifestations in individuals younger than 40 years of age can be suggestive of MEN1 syndrome. In this context, a careful medical history and strong clinical evidence are essential for correct diagnosis. The presence of cutaneous tumors (angiofibromas, collagenomas, lipomas) may be helpful in the clinical presymptomatic diagnosis of MEN1 patients, as often they appear before any clinical manifestations of MEN1-associated hormone-secreting tumors.

Figure 7.

Most commonly used imaging procedures in MEN1 syndrome clinical diagnosis. Left column: endoscopic ultrasonography (EUS); central column: magnetic resonance imaging (MRI); right column: computed tomography (CT) screenings. Red circles indicate the localization (more...)

The clinical management of MEN1-associated tumors is complicated by the limitations of current imaging techniques to accurately localize tumors at an early stage and the relative lack of specific and sensitive tumor markers. Hormone-secreting tumors can be detected by simple analysis of increased serum-specific hormone concentration or urinary evaluation of elevated hormone metabolites. Particularly, biochemical screening permits the detection of endocrine tumors about 5-10 years before the development of clinical symptoms, allowing for early surgical and/or pharmacological intervention.

Genetic diagnosis

Since the discovery of the MEN1 gene in 1997, early recognition of affected and at-risk individuals, even in the absence of biochemical and/or clinical symptoms, has been possible by DNA testing (106). This early recognition has reduced the morbidity and mortality of MEN1 by early detection and treatment. Mutational analysis of the MEN1 gene is recommended for patients who meet the clinical criteria for MEN1 and for those in whom a diagnosis of MEN1 is suspected.

Identification of a mutation in a patient enables testing for relatives. This allows early identification of asymptomatic mutant gene carriers and provides an indication for them to undergo periodic biochemical and/or imaging screening for MEN1-associated endocrine and non-endocrine tumors (Figure 7). Particularly, genetic testing for MEN1 should be considered around age 10 years. Mutation carriers should be enrolled in a cancer surveillance protocol for early detection of the potentially malignant neuroendocrine tumors that account for most of the disease-related morbidity and mortality (Table 6).

Table 6.

Recommend surveillance program for Multiple Endocrine Neoplasia Type 1 according to the International Guidelines for Diagnosis and Therapy of MEN syndromes (106)

Nevertheless, the lack of a genotype-phenotype correlation means that neither the localization nor manifestations of MEN1-associated tumors can be predicted by specific MEN1 mutations. A wide variability of tumor occurrence, onset age, and clinical behaviour, even in patients sharing the same MEN1 gene mutation, has been described, making it difficult to foresee the clinical phenotype in asymptomatic mutant gene carriers. MEN1 clinical manifestations, age of onset, and natural history, in fact, are widely variable even among members of the same family. The absence of a genotype-phenotype correlation suggests there are environmental and genetic modifiers.

Patients who have a negative genetic test can be spared continued annual screenings and are not at risk to transmit the disease to their children. Most laboratories currently use direct DNA sequencing strategies of the MEN1 gene coding region and intron-exon junctions. This analysis requires a single blood sample, can be performed at any age, and does not need, in theory, to be repeated. However, sequencing analysis fails to detect large deletions causing the loss of the whole gene or an entire exon. Deletions or other gross rearrangements represent 1-3% of MEN1 gene germline mutations and can be detected by Southern blot analysis or by multiplex ligation dependent probe amplification (MLPA) (113,114).

When no MEN1 gene mutation in a MEN1 pedigree is identified, the genetic ascertainment can be based on linkage analysis if there are at least two generations of affected members (110). Haplotype analysis can be performed using specific microsatellite markers flanking the MEN1 locus at 11q13 and reaches a degree of confidence when a substantial number of affected members have been analyzed. Microsatellite-based LOH analysis on MEN1 tumors at 11q13 region showed that the MEN1 gene lies telomeric to PYGM gene in a minimal 600 kb interval between PYGM and D11S449 (115,116). Detailed physical map of human chromosomal region 11q12-13 showed a meiotic recombination rate higher than expected, particularly around the MEN1 locus (116). However, several studies demonstrated no recombinants between MEN1, PYGM, and D11S449 loci (19, 40, 41). Studies on MEN1 families demonstrated that MEN1 11q13 locus private familial haplotype transmission correlated with the disease (118,119).

Parathyroid tumors

Accurate clinical and laboratory screening for parathyroid function promotes early biochemical diagnosis. PHPT is biochemically diagnosed by evaluation of a consistently increased serum concentration of PTH (reference range 10-60 pg/ml) and a consistently increased serum concentration of albumin-corrected calcium (reference range 9.0-10.5 mg/dl or 2.2-2.6 mmol/l) and of ionized calcium (reference range 4.5-5.6 mg/dl or 1.1-1.4 mmol/l) (120). Imaging is not usually used for the diagnosis of parathyroid adenomas.

GEP neuroendocrine tumors

Endoscopic ultrasonography (EUS) examination is the most sensitive imaging procedure for the detection of small (≤10 mm) pancreatic endocrine tumors in asymptomatic MEN1 patients; its sensitivity is higher than 75%. Duodenal gastrinomas can be detected by endoscopy (sensitivity 40-50%) and EUS (sensitivity 50%) (121). In addition, somatostatin receptor scintigraphy ((¹¹¹In)DTPA-octreotide scan or Octreoscan) is a proven pancreatic islet imaging method. The use of EUS in association with Octreoscan scintigraphy increases the pancreatic tumoral detection rate to 90% (122); EUS allows precise localization of the tumors while Octreoscan scintigraphy provides additional information regarding the spread of the disease and detects liver metastases with a sensitivity of 92% (123). Pancreatic non-functional tumors can only be identified by imaging procedures, while hormone-secreting tumors can be detected at an earlier stage through biochemical screening of increased serum-specific hormone concentration. Serum Chromagranin-A analysis can be used as a biochemical diagnostic test for MEN1-associated GEP tumors as all radiologically detectable GEP tumors are associated with elevated serum Chromagranin-A concentrations (124).