Menin and the Cell Cycle

Menin is a nuclear protein in nondividing (interphase) cells^1,2 and it is only in mitosis when the nuclear membrane has dissolved that menin appears in the cytoplasm.³ At this time it may be associated with cytoskeletal elements. Menin interacts with nonmuscle myosin II-A heavy chain (NMHC II-A) that mediates alterations occurring in cytokinesis and cell shape during cell division⁴ and also interacts with the intermediate filament network proteins, glial fibrillary acidic protein (GFAP) and vimentin.⁵ It is unclear whether this relocalization represents only a sequestering of menin before cytokinesis or that the protein is playing a functional role in this location during late mitosis..

Tumor suppressors BRCA1 and BRCA2 are poorly expressed in quiescent cells. By contrast, we found that menin protein is relatively well expressed in quiescent rat pituitary somatolactotrope GH4C1 cells at G0-G1.² The CDKIs, such as p21 and p27, are also well expressed in quiescent cells and we suggested that menin may function like these CDKIs (or regulate their expression). We found that the levels of menin transiently decrease as Rb protein becomes hyperphosphorylated as cells enter the cell cycle and then increase again as the cells enter S phase from the G1-S-phase boundary onward (see Fig. 1). Others have identified increases in menin mRNA at this time.⁶ It is at this stage that expression of other tumor suppressors such as BRCA1, BRCA2 and p53 increases. Thus menin may play some role at the G1-S-phase checkpoint analogous to BRCA1, BRCA2 and p53. It is to be emphasized, however, that the relative changes in menin expression are modest (2-3-fold) relative to the marked changes in expression noted for other tumor suppressors over the course of the cell cycle. Changes in posttranslational modification that might suggest alterations in activity throughout the cell cycle have yet to be examined.

Figure 1

Menin affects the activity and/or expression of several key cell cycle regulators either by directly interacting with them or modulating transcription of their genes. A) G0/G1 to S phase transition; Menin blocks the transition from G0/G1 to S phase and (more...)

Menin and the Retinoblastoma Protein

Studies with knockout mice have provided evidence that menin and Rb may operate in a common pathway to regulate cell proliferation.⁷ Mice homozygous for either deletion of the Rb1 gene or the Men1 gene die in utero. Mice heterozygous for either deletion of the Rb1 gene⁸ or the Men1 gene develop endocrine tumors.^9,10 In the Rb1^+/−mice, intermediate pituitary and thyroid tumors occur frequently with less frequent development of pancreatic islet hyperplasia and parathyroid lesions. In the Men1^+/−mice, pancreatic islet and anterior pituitary adenomas are common. In mice heterozygous for both Men1 and Rb1 deletion, pancreatic hyperplasia and tumors of the intermediate pituitary and thyroid occur at high frequency. The tumor spectrum in the double heterozygotes is a combination of those for the individual heterozygotes, with no decrease in age of onset. This suggests that menin and Rb function in a common pathway. This would be in contrast with studies of mice heterozygous for deletion of Rb1 and p53 that exhibit accelerated tumorigenesis and a broadening of the spectrum of tumor types observed to encompass all the types exhibited in the individual heterozygotes.¹¹

Menin and CDK Inhibitors

p18-p27 double mutant mice, like Rb heterozygous mice, develop multiple endocrine neoplasias, including pituitary, thyroid, parathyroid and adrenal, providing evidence that p18 and p27 function by regulating Rb's tumor suppressor function. The types of endocrine tumors cover the spectrum seen in MEN1 and MEN2.^12,13 However, mice lacking p53 develop lymphomas and soft tissue sarcomas.¹¹ p53 functions as a checkpoint gene to monitor genomic integrity, whereas Rb functions to integrate mitogenic signals to determine whether the cell will enter S phase or not.

Menin directly regulates the expression of the CDKIs, p27 and p18. Menin does this by recruiting mixed lineage leukemia (MLL) protein complexes to their gene promoters and coding regions. Loss of function of either MLL or menin results in down-regulation of p27 and p18 expression and deregulated cell proliferation.¹⁴ By use of a mouse model with heterozygous deletion of the Men1 gene, it was shown that menin-dependent histone methylation maintained expression of CDKIs and prevented the formation of pancreatic islet tumors.¹⁵ Therefore, menin is involved in cell proliferation control by an epigenetic mechanism. Excision of Men1 in mouse embryonic fibroblasts (MEFs) accelerated entry into S phase accompanied by increased CDK2 activity and decreased expression of p18 and p27. Complementation of menin-null cells with menin repressed S-phase entry.¹⁶

In stable Men1-deficient Leydig tumor cell lines reconstituted menin expression decreased cell proliferation with a block in transition from G0/G1 to S phase and an increase in apoptosis accompanied by increased p18 and p27 expression.¹⁷ Tissue-specific inactivation of Men1 in neural crest precursor cells in the mouse leads to perinatal death with cleft palate and other cranial bone defects associated with decreased p27 expression.¹⁸ The study demonstrated that menin functions in vivo during osteogenesis and is required for palatogenesis, skeletal rib formation and perinatal viability.

Therefore, a recurring theme arising from studies investigating the functions of menin either early in embryogenesis and fetal development, on the one hand and those examining roles that when lost in the growing or adult organism lead to endocrine (and other) neoplasia, on the other, has been the requirement of menin for the proper functioning of some of the CDKIs in cell cycle control.

Menin and GTPases

Ras-transformed murine fibroblasts (NIH3T3 cells) demonstrate increased proliferation, clonal formation in soft agar and tumor growth after inoculation into nude mice. Overexpression of menin in these cells caused them to partially revert to the phenotype of the parent NIH3T3 cells in vitro and in vivo.¹⁹ The studies support menin acting as a tumor suppressor, although the activities being displayed by menin in these experiments may not necessarily involve a direct antagonism of the Ras pathway.

It has been suggested that nucleoside diphosphate (NDP) kinases might act as molecular switches to alter cell fate towards proliferation or differentiation in response to external signals.²⁰ The activity of the NDP kinases would be regulated by small molecular weight G proteins like Rad and Rac that function as guanosine triphosphate (GTP)ases. While menin is not of small molecular weight it has been shown to interact with the tumor suppressor NM23H1/NDP kinase. While neither protein has GTPase activity of its own, their interaction induces GTPase activity by menin.^21,22 It has been proposed that menin has several motifs similar to those in known GTPases although the homology is weak. It would be anticipated that the interaction with menin stimulates GEF activity of nm23 although this remains unknown. In addition NDP kinases function at the plasma membrane and evidence is lacking for menin being in the appropriate cellular localization to fulfil this function.

Menin and JunD

JunD is a member of the activator protein-1 (AP-1) transcription factor family and in contrast to other Jun and Fos proteins has antimitogenic activity. JunD negatively regulates fibroblast proliferation and antagonizes transformation by Ras in vitro and in vivo.²³ Of all the AP-1 family members, menin interacts only with JunD and represses its transcriptional activity by association with an mSin3A-histone deacetylase (HDAC) complex.^24-27 We had pointed out that it appeared paradoxical that one antimitogenic factor would reduce the activity of another.² We suggested that whereas menin is a regulator of JunD action, JunD may not be the main mediator of menin action. Further studies in fibroblasts have suggested that the nature of JunD can change depending upon whether it is bound by menin when it functions as a growth suppressor or it is not bound by menin in which case it acts as a growth promoter like other AP-1 family members.²⁸ However, the result of JunD-menin interaction may be cell-type specific as JunD has a differentiating effect in osteoblasts, an action that is inhibited by menin.²⁹

Menin and Activator of S-Phase Kinase (ASK)

Targeted disruption of the Men1 gene leads to enhanced cell proliferation, whereas complementation of menin-null cells with menin reduces cell proliferation. Menin interacts with activator of S-phase kinase (ASK), a component of the Cdc7/ASK kinase complex. The C-terminal domain of menin interacts with ASK. Wild-type menin completely represses ASK-induced cell proliferation although it does not affect the steady-state cell cycle profile of ASK-infected cells.³⁰ As menin itself represses basal cell proliferation, it is unclear whether menins effects are occurring via its interaction with ASK or in an unrelated manner. Also, ASK itself did not alter the steady-state cell cycle profile. Disease-related C-terminal menin mutants that do not interact with ASK did not repress either ASK-induced or basal cell proliferation. From other studies it appears that ASK is in the nucleus.³¹ The C-terminal menin deletion mutants would not gain access to the nucleus. Further studies need to be done to determine where in the cell any physical interaction between menin and ASK is taking place and what the exact functional link between the two proteins is.

Menin and TGF-Beta Family Members

In most mature tissues the cytokine TGF-beta provides cytostatic signals that limit G1 progression and cell proliferation. Menin is a Smad3-interacting protein and is a facilitator of transcriptional activity of the Smads (see Fig. 2).³² In anterior pituitary cells, inactivation of menin blocks TGF-beta and activin signaling, antagonizing their proliferation-inhibitory properties.^33,34 In cultured parathyroid cells from uremic hemodialysis patients in which the menin signaling pathways are intact, menin inactivation achieved by menin antisense oligonucleotides leads to loss of TGF-beta inhibition of parathyroid cell proliferation and parathyroid hormone (PTH) secretion (see Fig. 3). Moreover, TGF-beta does not affect the proliferation and PTH production of parathyroid cells from MEN1 patients that were devoid of menin protein.^35,36 Antisense inhibition of menin in a rat duodenal crypt-like cell line increased cell proliferation with loss of cell-cycle arrest in G1 and increased expression of cyclin D1 and CDK4 and decreased expression of the TGF-beta type II receptor.³⁷ Hence, menin plays a critical role in mediating the cytostatic effects of TGF-beta ligands.

Figure 2

Role of menin in TGF-beta-mediated cell proliferation and gene transcription facilitated by Smad/DNA interaction. A) TGF-beta stimulates menin expression. Serum-starved GH4C1 cells were cultured in TGF-beta for the indicated times (h) and total cell menin (more...)

Figure 3

Role of menin in TGF-beta-mediated inhibition of parathyroid cell proliferation and parathyroid hormone (PTH) production. Parathyroid cells (from patients with secondary hyperparathyroidism in which menin function is normal) were cultured in chamber slides (more...)

During early embryogenesis and fetal development and in some adult mesenchymal cells, TGF-beta and bone morphogenetic proteins (BMPs) play important roles. Homozygous Men1 inactivation in mice is embryonic lethal and the fetuses exhibit cranial and facial developmental defects. Cranial bones form by intramembranous ossification and menin may play an important role in this type of bone formation. In vitro, menin promotes the initial commitment of multipotential mesenchymal stem cells to the osteoblast lineage through interactions with the BMP-2 signaling molecules, Smad1/Smad5 and the key osteoblast regulator, Runx2, whereas the interaction of menin and the TGF-beta signaling molecule, Smad3, inhibits later osteoblast differentiation by negatively regulating the BMP-Runx2 cascade.^38,39 The focus in these studies was predominantly on bone differentiation markers. However, it would also be important to extend these studies by evaluating menins influence on cell cycle markers. In vivo, in the mouse, tissue-specific inactivation of Men1 in neural crest cells that contribute to cranial bones and the skeletal ribs and other tissues leads to defects in osteogenesis and perinatal death.¹⁸ The fact that mice and humans heterozygous for loss of the Men1 gene develop normally indicates haplosufficiency for all of menins normal functions.^9,10 However, for some functions and in some cell types only, further reduction in menin results in developmental deficits.

The deregulation of the TGF-beta family pathway has also been correlated with Men1 inactivation and altered cell growth in vivo by studying the Leydig cell tumors of heterozygous Men1 mutant mice.⁴⁰ In the cells of the tumors the anti-Mullerian hormone (AMH)/BMP pathway was impaired with reduced expression of AMH receptor type 2, decreased expression of Smad1, -3, -4 and -5 and reduced BMP transcriptional activity. The expression of p18 and p27 was reduced and that of CDK4 increased. In other studies, it was noted that Men1-null MEFs demonstrate reduced expression of extracellular matrix proteins critical for organogenesis and that are induced by TGF-beta.⁴¹ TGF-beta failed to stimulate expression of these proteins in the menin-null MEFs that also had poor responsiveness to TGF-beta induced Smad3-mediated transcription.

Menin and Genomic Instability

A role for menin in the maintenance of genomic stability is suggested. Chromosomal instability (increased chromosomal breakage) was observed in cultured lymphocytes and in fibroblasts derived from skin biopsies of MEN1 patients and hence heterozygous for an MEN1 mutation.⁴² Peripheral lymphocytes from MEN1 patients displayed an increase relative to normal controls in premature centromere division after exposure to the alkylating agent diepoxybutane (DEB) that crosslinks DNA.^43-45 Menin-deficient MEFs were also moderately sensitive to DEB and displayed a high frequency of chromosomal aberrations after exposure to this agent.⁴⁶ Studies in Drosophila showed that flies mutant for the Men1 orthologue are hypersensitive to ionization radiation and are hypermutable.⁴⁶ A genome-wide loss of heterozygosity (LOH) screening of 23 pancreatic lesions from 13 MEN1 patients has shown multiple allelic deletions indicating that MEN1 pancreatic tumors fail to maintain DNA integrity and demonstrate signs of chromosomal instability.⁴⁷ However, in another study, no obvious chromosomal instability was observed in islet cells of Men1 knockout mice and tumors developed in the absence of chromosome or microsatellite instability.⁴⁸

Menin, DNA Binding and ATR-CHK1 Pathway

Menin binds DNA and interacts with proteins implicated in DNA damage pathways. The canonical cellular response to UV-induced damage involves activation of the ATR kinase pathway. Following UV irradiation of human embryonic kidney (HEK293) cells, menin concentration in chromatin increased but was decreased by the ATR inhibitor, caffeine.⁴⁹ Transfection of constitutively active checkpoint kinase 1 (CHK1) increased chromatin-bound menin mimicking the effect of UV irradiation and implicating the involvement of an ATR-CHK1 dependent pathway.

Similar to other tumor suppressors like BRCA1,⁵⁰ that are involved in DNA repair pathways, menin apparently directly binds dsDNA in a sequence independent way.⁵¹ Amino acid sequences located towards the C-terminus of menin within the nuclear localization signals (NLS1 and NLS2) appear to be essential for this binding. The study was conducted in MEFs null for Men1 in which a failure to repress cell proliferation and cell cycle progression at the G2/M phase was noted. An anticipated effect at G1/S as occurs with BRCA1 depletion was not observed. However, the authors noted that the method of immortalization of the MEFs would have blocked the Rb pathway so that effects on this part of the cell cycle would not have been evaluated.

Menin and FANCD2

The FANCD2 protein is involved in DNA repair and mutations in it result in the inherited cancer syndrome, Fanconi's Anemia. Menin interacts with FANCD2; gamma-irradiation enhances the interaction and increases the accumulation of menin in the nuclear matrix.⁵² These results suggest a role for menin, in cooperation with FANCD2, in DNA repair.

More recently, Marek et al⁵³ have compared the mutation frequency and spectra of Men1 and FANCD2 mutants in Drosophila. Men1 mutant flies were extremely prone to single base deletions within a homopolymeric tract, whereas FANCD2 mutants displayed large deletions. Neither overexpression nor loss of Men1 modified the interstrand crosslink (ICL) sensitivity of FANCD2 mutants. The different mutation spectra of Men1 and FANCD2 mutants together with lack of evidence for genetic interaction between these genes indicates that Men1 plays an essential role in ICL repair distinct from the Fanconi anemia genes.

Menin and RPA2

Menin directly binds to and colocalizes in the nucleus with the 32-kDa subunit (RPA2) of replication protein A (RPA), a heterotrimeric protein required for DNA replication, recombination and repair.⁵⁴ The interactive region was mapped to the N-terminal portion of menin and menin bound preferentially in vitro to free RPA2 rather than the RPA heterotrimer. Menin had no effect on RPA-DNA binding in vitro. However, menin antibodies coimmunoprecipitated RPA1 with RPA2 from HeLa extracts suggesting that menin binds to the RPA heterotrimer or a novel RPA1-RPA2-containing complex in vivo. The functional consequences of the interactions are unclear but it is possible that menin acts as a scaffold protein to bridge different components of a cross-link repair network. Menin could also have, by its association with HDACs, a more general effect on chromatin remodeling facilitating access of damaged DNA to the repair machinery.

Menin and FOXN3 (CHES1)

FOXN3 (CHES1), a member of the forkhead/winged-helix transcription factor family, was originally identified by its ability to suppress DNA damage sensitivity phenotypes in checkpoint-deficient yeast strains. By study of integrity of DNA damage checkpoints in mutant Drosophila lacking the Men1 orthologue and in MEFs deficient for Men1, biochemical and genetic interactions between menin and FOXN3 (CHES1) were demonstrated.⁵⁵ FOXN3 (CHES1) is part of a transcriptional repressor complex, that includes mSin3a, HDAC1 and HDAC2 and it interacts with menin in a DNA damage-responsive S-phase checkpoint pathway.

Menin and Transcriptional Regulation

Menin is implicated in the regulation of many genes via interaction with specific transcription factors and with the large subunit of RNA polymerase II.⁵⁶ Menin exerts a dual role, either as a repressor or as an activator depending upon the particular transcription factor involved. This might be explained by a model in which menin serves to link transcription regulation with chromatin modification.

Menin and Histone Deacetylase

The chromatin modification, histone acetylation, is correlated with activation of gene transcription. Histone deacetylation mediated by complexes of the general transcription repressor, Sin3A, with HDACs is related to gene inactivation. Menin interacts with HDAC-mSin3A complexes to repress transcription of JunD (see Fig. 4a).^25,26

Figure 4

Menin regulates gene transcription. A) Menin interacts with JunD and represses JunD-mediated transcription by recruiting a histone-deacetylating complex comprising the mSin3A, HDAC1 and HDAC2 proteins. B) Menin is a component of MLL1 or MLL2 complexes (more...)

Menin and Histone Methyltransferase

Menin, as a component of the MLL chromatin remodeling complexes, is involved in trimethylation of the fourth lysine (K) residue of histone H3 (H3K4 trimethylation) that is strongly associated with transcription activation.^56,57 By recruiting histone methyltransferase activity menin acts as a tumor suppressor by activating transcription of antiproliferative genes like those for CDKIs (see Fig. 4b).^14,15

Menin binds to the N-terminal region of MLL1. Some but not all of the few MEN1 associated missense mutants tested fail to bind MLL.⁵⁶ So it remains unclear exactly how essential the loss of H3K4 methylation is in MEN1 tumorigenesis. Of interest, fusion proteins created by chromosomal translocations of MLL and other proteins in the complex have a causal role in several forms of leukemia. Menin is an essential cofactor for MLL-associated leukemogenesis.^58-60 This has prompted the suggestion that menin, usually considered as a tumor suppressor, under special circumstances may promote oncogenesis.

Homeobox-domain (HOX) genes that control the fundamental body plan during embryogenesis are targets for MLL. Consistent with menin being part of the MLL complexes, the protein is located at the promoters of some HOX genes and shown in some cases to regulate individual HOX gene expression in various cells including MEFs,⁵⁶ mouse bone marrow cells⁵⁹ and HeLa cells.^57,58,61 Deregulation of expression of many and a few HOX genes has been found in parathyroid tumors of patients with MEN1 and sporadic hyperparathyroidism, respectively.⁶² The mechanisms underlying the altered HOX gene expression and whether it is causal of or coincidental to the tumorigenesis remains to be established.

Menin, Chromatin and Gene Expression

By combining chromatin immunoprecipitation (ChIP) assay with gene expression analysis, hundreds of menin-occupied chromatin regions were revealed.⁶³ The majority (67%) were at known genes (promoters, internal, 3' regions) with 33% located outside known gene regions. While this reinforces the notion of menin as a transcriptional regulator, menin likely acts in an indirect fashion (it may not bind DNA directly) by functioning as a scaffold protein within chromatin remodeling complexes. Knowledge of the participation of menin in regulating several genes has come from ChIP analysis and promoter-reporter transfection assays. Additional information has come from microarray analysis of a variety of Men1 expressing versus Men1 deleted cells. Interestingly, this has revealed that menin target genes in one tissue show minimal overlap with targets of other tissues.⁶³

Genomic binding sites of menin and other proteins representing MLL complexes (see above) were mapped to several thousand gene promoters in three different cell types by ChIP coupled with microarray analyses.⁶¹ While menin frequently colocalized with chromatin modifying complexes it also bound other promoters by other (unknown) mechanisms.