Abstract
Pancreatic endocrine tumors (PETs) may be part of hereditary multiple endocrine neoplasia type 1 (MEN1) syndrome. While MEN1 gene mutation is the only ascertained genetic anomaly described in PETs, no data exist on the cellular localization of MEN1-encoded protein, menin, in normal pancreas and PETs. A total of 169 PETs were used to assess the i) MEN1 gene mutational status in 100 clinically sporadic PETs by direct DNA sequencing, ii) immunohistochemical expression of menin in normal pancreas and 140 PETs, including 71 cases screened for gene mutations, and iii) correlation of these findings with clinical–pathological parameters. Twenty-seven PETs showed mutations that were somatic in 25 patients and revealed to be germline in 2 patients. Menin immunostaining showed strong nuclear and very faint cytoplasmic signal in normal islet cells, whereas it displayed abnormal location and expression levels in 80% of tumors. PETs harboring MEN1 truncating mutations lacked nuclear protein, and most PETs with MEN1 missense mutations showed a strong cytoplasmic positivity for menin. Menin was also misplaced in a significant number of cases lacking MEN1 mutations. In conclusion, the vast majority of PETs showed qualitative and/or quantitative alterations in menin localization. In 30% of cases, this was associated with MEN1 mutations affecting sequences involved in nuclear localization or protein–protein interaction. In cases lacking MEN1 mutations, the alteration of one of the menin interactors may have prevented its proper localization, as suggested by recent data showing that menin protein shuttles between the nucleus and cytoplasm and also affects the subcellular localization of its interactors.
Introduction
Pancreatic endocrine tumors (PETs) arise sporadically or in the context of hereditary cancer syndromes (Capelli et al. 2009), the most common of which is multiple endocrine neoplasia type 1 (MEN1) syndrome, an autosomal dominant disorder caused by germline alterations of the MEN1 gene. The syndrome is characterized by the combination of neoplastic–hyperplastic lesions affecting the parathyroids, enteropancreatic endocrine tissues, and the anterior pituitary gland. Less frequently, patients can develop tumors in other endocrine and non-endocrine tissues (Marx et al. 1999, Pannett & Thakker 1999).
Despite extensive studies, the genetic background of PET is still poorly understood. The most common anomaly also encountered in sporadic PETs is MEN1 gene inactivation due to mutation of one allele and loss of the second allele (Capelli et al. 2009). Indeed, allelic losses at chromosome 11q13, which includes the locus of MEN1 gene (Larsson et al. 1988), are found in 46% of sporadic PETs (Speel et al. 1999, Rigaud et al. 2001), while the reported frequency of MEN1 mutations is lower, thus suggesting that additional oncosuppressor genes lie on this chromosomal arm (Eubanks et al. 1994, Chakrabarti et al. 1998, Gortz et al. 1999, Jonkers et al. 2006). MEN1 gene encodes for menin, a 610 amino acid protein (Chandrasekharappa et al. 1997, Lemmens et al. 1997) that is ubiquitously expressed in adult tissues (Chandrasekharappa et al. 1997) and located prevalently in the nucleus via three nuclear localization signals (NLS1, NLS2, and NLSa) in its C-terminal fragment (Guru et al. 1998, La et al. 2006). Cavallari et al. (2003) provided the first description of the immunohistochemical expression pattern of menin in human pancreas using indirect immunofluorescence (IF) on frozen tissues, showing that the protein is mainly localized into the nucleus of all pancreatic cell types. They also analyzed seven pancreaticoduodenal endocrine tumors showing variable expression levels of menin compared to normal tissue. However, no data exist on the subcellular localization of the protein associated with its mutations.
Menin is likely to work as a scaffold recruiting either positive or negative regulators of transcription to control expression of genes that are involved in different cell functions (La et al. 2004). Indeed, among others, menin interacts with the transcription factor JunD suppressing its growth promoting activity (Agarwal et al. 1999, 2003, Kim et al. 2003), and with Smad3 inhibiting transforming growth factor-β (TGF-β) signaling pathway (Kaji et al. 2001). Additional interactions include those with the metastasis suppressor NM23 that is involved in DNA replication (Ohkura et al. 2001), and the transcription factor p65/nuclear factor-κB (NF-κB) (Heppner et al. 2001). These properties, together with the fact that MEN1 mutations are of the loss of function type and that the wild-type allele is lost in tumors derived from MEN1-affected patients (‘second hit’), identify menin as a tumor suppressor acting through either JunD and TGF-β central pathways or more complex effects through multiple interactions.
Menin also associates with a histone methyltransferase complex that methylates lysine 4 of histone H3, an epigenetic event associated with transcriptional activation (Yokoyama et al. 2004). In line with such a role, it has been reported that menin directly regulates the expression of the cyclin-dependent kinase inhibitors p27kip1 and p18Ink4c. The loss of function of menin leads to the down-regulation of these inhibitors thereby causing a deregulation of cell growth (Milne et al. 2005).
Furthermore, menin is also involved in the regulation of transcription of different genes such as hTERT, IGFBP2, and CASPASE8 (Schnepp et al. 2004), and interacts with proteins involved in the maintenance of genome stability (Jin et al. 2003, Sukhodolets et al. 2003).
In this study, we analyzed a large series of apparently sporadic PETs with the aim to i) assess the frequency and site of mutations in the MEN1 gene, ii) compare MEN1 gene status with immunohistochemical stain for menin, and iii) correlate this data with clinical–pathological information.
Materials and methods
Patients and tissues
All samples were obtained from patients who underwent either explorative or radical surgery. Informed consent for analysis of normal and neoplastic tissues was obtained from patients, in accordance with the approval of the Verona University and Hospital Trust's Ethical Committee. The MEN1 mutational study was performed on 100 PETs comprising 75 non-functioning and 25 functioning tumors (16 insulinoma, 4 gastrinoma, 2 somatostatinoma, 2 VIPoma, and 1 glucagonoma; Table 1 and Supplementary Table 1, see section on supplementary data given at the end of this article). Forty-one cases (31 non-functioning and 10 functioning) had already been analyzed for MEN1 mutations in a previous study (Moore et al. 2001).
The 100 pancreatic endocrine tumors included in the mutational analysis
Tumor type | Patients (n=100) | Sex (M/F) | Age (median) |
---|---|---|---|
Insulinomas | |||
WDET | 14 | 7/7 | 36 |
WDEC | 2 | 1/1 | 46 |
Other functioning tumors | |||
WDEC | 9 | 3/6 | 61 |
Non-functioning tumors | |||
WDET | 31 | 12/19 | 59 |
WDEC | 44 | 20/24 | 54 |
WDET, well-differentiated endocrine tumor; WDEC, well-differentiated endocrine carcinoma; M, male; F, female.
Cell lines
The human PET cell lines were CM, BON, and QGP-1 (see Supplementary Table 2, see section on supplementary data given at the end of this article for further information). QGP-1 cells were grown in RPMI 1640 supplemented with 10% fetal bovine serum (FBS); CM cells were grown in RPMI 1640 supplemented with 5% FBS; BON cells were grown in DMEM and F12K (ratio 1:1) supplemented with 10% FBS. All the culture media were supplemented with 20 mM glutamine and 50 mg/ml of gentamicin sulfate (BioWhittaker, Treviglio, Italy). Cells were incubated at 37 °C with 5% CO2.
MEN1 gene mutational analysis
Genomic DNA was isolated after cryostat enrichment of frozen tissues to obtain a neoplastic cellularity of at least 90% (Achille et al. 1996) using DNeasy Blood and Tissue kit (Qiagen). Matched normal DNA was used to determine whether the mutations identified were somatic or germline. Genomic DNA from PET cell lines BON, QGP-1, and CM was isolated using the AllPrep DNA/RNA/Protein mini kit (Qiagen). The entire coding region (exons 2–10) and the adjacent splice junctions of the MEN1 gene were amplified using primers reported in Supplementary Table 3, see section on supplementary data given at the end of this article by touch-down PCR. Cycling conditions were 94 °C for 3 min, 20 cycles at 94 °C for 20 s, 68 °C for 25 s (0.5 °C decrease for each cycle), followed by 15 cycles at 94 °C for 20 s, 55 °C for 20 s, and a final extension at 72 °C for 3 min. PCR purification and direct sequencing have been described (Balakrishnan et al. 2007). Sequence differences to the NCBI (http://www.ncbi.nlm.nih.gov) reference sequence (GeneID 4221) were identified via manual inspection of aligned electropherograms assisted by the Mutation Surveyor software package (SoftGenetics, State College, PA, USA). The genetic alteration identified was cross-referenced to variant information from the NCBI SNP database (http://www.ncbi.nlm.nih.gov), the Ensemble Genome Browser (http://www.ensembl.org), the Swiss-Prot (http://ca.expasy.org) and GenBank (http://www.ncbi.nlm.nih.gov/Genbank) databases, the COSMIC database (http://www.sanger.ac.uk/genetics/CGP/cosmic), and literature.
Menin immunohistochemistry
Immunohistochemical staining for menin was performed on formalin-fixed paraffin-embedded (FFPE) tissue microarrays (TMAs) using the anti-menin commercial antibody A300-105A of the Bethyl Laboratories (Montgomery, TX, USA) at a dilution of 1:1000 for 1 h at room temperature. In order to verify the specificity of the immunostaining, the antibody was preadsorbed with the menin-blocking peptide (BP300-105, Bethyl Laboratories) in a saturating concentration (10 μg of peptide/ml) for 2 h before use. Heat-induced menin antigen retrieval was performed using microwave oven and 0.01 mmol/l (pH 6.0) citrate buffer for 30 min. All samples were processed using a sensitive ‘Bond polymer Refine’ detection system in an automated Bond immunostainer (Vision-Biosystem, Menarini, Florence, Italy). Sections incubated without the primary antibody served as a negative control.
TMAs contained 140 primary PETs, 38 matched metastasis (22 nodal and 16 liver), and 12 normal pancreas. The 140 PETs included 106 non-functioning and 34 functioning tumors (Supplementary Table 4, see section on supplementary data given at the end of this article). The construction of the TMAs was performed using a tissue arrayer (Beecher Instruments, Silver Spring, MD, USA) as described (Kononen et al. 1998). For most cases, at least three cores of 1 mm diameter per sample were analyzed. Slides were scanned with ScanScope GL System (Aperio Technologies, Vista, CA, USA) and visualized using ImageScope Software (Aperio Technologies). Three independent observers scored menin expression. The staining intensity was classified into four grades as follows: 0 (absent), 1 (weak), 2 (moderate), and 3 (strong), with separate scores for nucleus and cytoplasm. The quality of TMAs is testified by the detection of multiple antigens by immunohistochemistry (IHC) on serial sections, thus excluding the occurrence of significant protein degradation in these samples. Indeed, the very same TMAs used for this study have been extensively used for different immunohistochemical analysis, some of which have also been published (Missiaglia et al. 2010).
Menin immunofluorescence
Tumor cell lines were seeded at a density of 2×104 cells per well in 4-well chamber slides (Nalgene Nunc International Corp., Naperville, IL, USA) for 24 and 48 h. Cells were then rinsed three times with ice-cold PBS and fixed in ice-cold methanol for 10 min. The permeabilization was performed using 0.1% Triton X-100 in PBS for 15 min at 20 °C. Menin was detected with the antibody A300-105A of the Bethyl Laboratories used at a dilution of 1:300. Bound antibody was detected with FITC-conjugated anti-rabbit antibody (Jackson ImmunoResearch Laboratories, West Grove, PA, USA). Nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI). Slides were analyzed on an Olympus BX61 microscope (Olympus, Hamburg, Germany) with the appropriate filter for FITC and a u.v. filter for the DAPI nuclear counterstain. The specificity of the IF pattern obtained was verified by preadsorbing the anti-menin antibody with the menin-blocking peptide (BP300-105, Bethyl Laboratories). All experiments have been repeated twice with comparable results.
Protein extraction from FFPE tissue samples
For each extraction, three serial 15 μm sections (∼25 mm2 each) of FFPE PET tissue were used. After standard deparaffination of tissue sections (15 μm each), proteins were extracted as previously described (Becker et al. 2007), using the Qproteome FFPE tissue kit (Qiagen). Briefly, reference routinely hematoxylin/eosin-stained sections of the tissues were histologically verified, and areas of interest were indicated with a pen on the slides. Subsequently, the areas of interest were scratched from the slides with a needle. The microdissected tissue attached to the needle was transferred into the Qproteome FFPE tissue kit buffer. Proteins were extracted according to the protocol provided by the manufacturer and stored at −20 °C.
As a control for antibodies specificity, proteins were extracted from three previously described PET cell lines (Capurso et al. 2006, Cecconi et al. 2007) using the AllPrep DNA/RNA/Protein mini kit (Qiagen) according to the manufacturer's instructions.
Immunoblotting
Total protein extracts from FFPE tissues (30 μg/lane) and cell lines (15 μg/lane) were subjected to SDS-PAGE (12% polyacrylamide gel), and blotting was performed on nitrocellulose membranes (Bio-Rad Laboratories) for 2 h at 60 V. The following antibodies were used: MEN1 (A300-105A, 1:5000, Bethyl Laboratories); menin (H-300, 1:400, Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA); β-actin (AC15, 1:5000, Sigma–Aldrich). Blots were developed using the ECLPlus Western Blot Detection System (Amersham/GE Healthcare Europe GmbH, Munich, Germany).
Statistical analysis
Association of immunohistochemical staining with clinical–pathological variables was evaluated using Pearson's χ2 tests or Fisher's test when appropriate for categorical variables; Kruskal–Wallis test or Wilcoxon test for continuous variables. All tests were considered significant when P value was <0.05. For all the calculation, the R statistical software package was used (http://www.r-project.org).
Results
The mutational analysis of MEN1 was executed on 100 primary neoplasms and 3 human PET cell lines. Immunohistochemical staining for menin was performed on 140 PETs, 71 of which coincided with those analyzed for mutations.
MEN1 gene mutational analysis in 100 PETs
Twenty-seven PETs were found to harbor non-synonymous MEN1 mutations (27%; Table 2 and Fig. 1). Of these, 25 were somatic and 2 were germline (Table 2, numbers 14 and 16) as assessed by the analysis of peripheral blood DNA. Mutations were 8 missense (30%), 3 nonsense (11%), 11 deletions (41%), 1 small insertion (3%), and 4 splice site mutations (15%). Each mutated tumor carried a different MEN1 mutation.
The 27 multiple endocrine neoplasia type 1 (MEN1) mutations found in 100 pancreatic endocrine tumors
Mutation IDa | Sample ID | Exon | Nucleotide changeb | Predicted effectc |
---|---|---|---|---|
Nonsense | ||||
1 | NF23 | 3 | g.3299 G>A | p.W183stopd |
2 | NF1 | 7 | g.4960 G>T | p.E316stop |
3 | NF38 | 9 | g.6184 G>T | p.E425stop |
Missense | ||||
4 | NF18 | 3 | g.3286 A>T | p.E179Vd |
5 | NF24 | 3 | g.3313 C>T | p.P188L |
6 | NF27 | 3 | g.3336 G>T | p.V196Fd |
7 | NF41 | 4 | g.3685 C>T | p.A242V |
8 | NF46 | 6 | g.4223 C>T | p.A284V |
9 | NF7 | 6 | g.4283 G>T | p.R304Sd |
10 | F13 | 9 | g.6140 T>C | p.F410S |
11 | F18 | 9 | g.6233 T>C | p.V441A |
Deletions | ||||
12 | NF16 | 2 | g.1391delC | fs118stopd |
13 | NF19 | 2 | g.1434_1437delGTCT | fs166stopd |
14 | NF11 | 2 | g.1453delTA | fs115stop |
15 | NF79 | 2 | g.1552_1566 delGATGTCATATGGAAC | In-frame deletion |
16 | F21 | 3 | g.3211_3244del33nt | In-frame deletion |
17 | NF8 | 3 | g.3261delC | fs185stop |
18 | NF53 | 3 | g.3384delA | fs223stop |
19 | NF10 | 4 | g.3683_3684delTG | fs470stopd |
20 | NF13 | 7 | g.4972_4973delCC | fs364stopd |
21 | NF32 | 9 | g.6164delA | fs444stop |
22 | F3 | 10 | g.6573delG | fs557stopd |
Insertions | ||||
23 | NF78 | 10 | g.6674_6675insC | fs530stop |
Splice site | ||||
24 | NF67 | Intron 3 | g.3405delGTA | – |
25 | NF57 | Intron 4 | g.4074 A>G | – |
26 | NF70 | Intron 5 | g.4118 T>G | – |
27 | NF63 | Intron 9 | g.6477 A>T | – |
The mutations are listed by mutation type alongside the exons in which they were identified. The nucleotide position of each mutation corresponds to the position of that change on coding strand. NF, non-functioning tumors; F, functioning tumors.
Mutation ID indicates the number reported in Fig. 1.
Mutations are numbered with reference to the MEN1 genomic sequence U93237 (GenBank); nt, nucleotide; g, genomic sequence; del, deletion; ins, insertion.
p, protein sequence; fs, which denotes frameshift, leads to a premature stop codon at indicated amino acids. The effect of splice site mutations is not determined.
Mutations already found by Moore et al. (2001).
Direct DNA sequencing of the 41 cases already investigated by single-strand conformation polymorphism and conformation-sensitive gel electrophoresis confirmed all the nine alterations previously identified (see Table 2 for details; Moore et al. 2001).
Among the 25 somatic mutations, 3 were in insulinomas (19%) and 22 in non-functioning PETs (30%). With regard to sporadic non-functioning PETs, which are the most representative tumor subgroup in our series, 13/44 (29%) malignant tumors and 9/31 (29%) benign tumors were mutated. Moreover, the genetic alterations identified were cross-referenced with variant information from databases and literature, and interestingly only one mutation (number 4, Table 2) was previously found in PET patients (Bergman et al. 2000).
Four common polymorphisms were also found: D418D (rs2071313) in 43 tumors, H433H (rs540012) in 30 tumors, A541T (rs2959656) in 14 tumors, and R171Q (rs607969) in 7 tumors.
The mutational analysis of the three human PET cell lines revealed no genetic abnormalities and identified the presence of two polymorphisms already found in primary neoplasms: R171Q (in CM cell line) and A541T (in BON cell line).
Menin immunohistochemical expression in 140 PETs
Immunohistochemical expression of menin was evaluated using a series of tissue arrays containing 140 primary PETs, 38 matched metastasis (22 nodal and 16 liver), and 12 normal pancreas. Whenever present, the nuclear and/or cytoplasmic immunopositivity was always uniform in all neoplastic cells.
In normal pancreas, menin expression was detected in all cell types (Fig. 2). Islet cells showed strong nuclear and very faint cytoplasmic immunostaining, while acini and ducts showed a moderately stained cytoplasm. In all normal cell types, nuclei showed a higher level of expression with respect to cytoplasm, whereas in tumors, variable localization and expression levels were observed (Fig. 2).
In tumors, nuclear immunostaining was found in 111 cases, while 29 (21%) had no detectable nuclear menin. Of the former, 44 (31%) had strong nuclear menin signals, 40 (29%) displayed an intermediate level of expression, and 27 (19%) had weak immunoreactivity. Cytoplasmic staining was observed in 135 cases and was absent in 5 cases (4%). Of the former, 44 cases (31%) had strong menin expression, 70 (50%) were moderately stained, and 21 (15%) had weak staining. The five cases lacking cytoplasmic menin were among those that also lacked nuclear staining, while the intensity of cytoplasmic staining was not significantly related to the intensity of nuclear staining. No difference in menin localization or expression level was found between primary and metastatic samples.
Menin expression in human PET cell line
IF was performed on confluent and subconfluent human PET cell lines (BON, QGP-1, and CM). The results of the IF analyses were consistent with the immunohistochemical staining observed for the majority of PET tissues. Figure 3 shows the results of the IF analyses of a subconfluent BON cell culture revealing that menin is expressed in both the nucleus and the cytoplasm of the cells (Fig. 3C, overlay image). Interestingly, this cell line is not affected by MEN1 gene mutations and only presents the A541T polymorphism in a heterozygous state.
Western blot analysis
To confirm the presence of menin suggested by the immunohistochemical positivity in cases harboring different MEN1 gene mutations, we performed western blot analysis of five primary PETs selected on the basis of their immunohistochemical and mutational profile (Table 3). Two different polyclonal anti-menin antibodies were used: the A300-105A from Bethyl Laboratories that is affinity purified for an epitope located at the C-terminus of the protein; and the H-300 from Santa Cruz Biotechnology that is raised against the N-terminus of menin. The specificity of antibodies was tested analyzing protein lysates of three PET cell lines (Supplementary Figures 1 and 2, see section on supplementary data given at the end of this article). For the A300-105A antibody, which is the very same antibody used to obtain menin immunostaining of PET tissues as well as menin IF of cell lines, the specificity was also assessed preadsorbing the antibody with the blocking peptide in a saturating concentration (Supplementary Figure 3, see section on supplementary data given at the end of this article). The western blot analysis of primary tumors shown in Fig. 4 is representative of three independent experiments that gave identical results. Menin was present in all samples except for case F8, the tumor that lacked menin immunohistochemical staining. A normal size protein was detected in wild-type tumor NF61 and tumor F18 carrying a missense mutation. A band of lower molecular mass (∼50 kDa) was observed in case NF38 harboring a nonsense mutation and tumor NF32 that had a frameshift-stop mutation. Noteworthy, these two mutations are likely to give a shortened protein, whose predicted size is about 50 kDa (Table 3). The absence of a band corresponding to the wild-type protein in these two cases is due to the loss of the normal allele, as assessed by the analysis of loss of heterozygosity (LOH) using PCR amplification of microsatellite markers (LOH, Table 3).
The five pancreatic endocrine tumors selected for menin western blot analysis shown in Fig. 4
Sample ID | Mutation typea | IHC scoreb (n/c) | Predicted protein size (kDa)c | LOHd |
---|---|---|---|---|
NF38 | Nonsense | 1/3 | 47 | Yes |
F8 | Wild-type | 0/0 | 67 | Yes |
NF32 | fs-stop | 0/2 | 49 | Yes |
F18 | Missense | 3/2 | 67 | NI |
NF61 | Wild-type | 3/3 | 67 | No |
NF, non-functioning tumors; F, functioning tumors; NI, not informative.
Wild-type, tumor with no mutation of MEN1 genomic sequence; fs, which denotes frameshift, leads to a premature stop codon; see Table 2 for detailed information about the predicted effects of the mutations found in tumors NF38, NF32, and F18.
IHC score, immunohistochemical score; separate scores were assigned for nucleus (n) and cytoplasm (c); see Materials and methods section for details.
The predicted size of the mutant proteins was estimated assuming 110 kDa as the molecular weight of each amino acid.
LOH, loss of heterozygosity. Matched normal and tumor DNAs were analyzed for loss of heterozygosity on chromosome 11q using the microsatellite markers D11S898 and D11S908. Analysis was performed on an ABI 310 DNA sequencer as described in Moore et al. (2001).
Furthermore, as the shortened proteins may miss the C-terminal epitope eventually recognized by the A300-105A antibody, tissue lysates (cases NF32, NF38, NF61, and F18) were independently re-probed with the menin (H-300) antibody directed against the N-terminus of the protein thus ensuring that the bands observed were not artifacts (Fig. 5). The results of this western blot were in line with those previously obtained. Indeed, a band of about 50 kDa was still detected in tumors harboring truncating mutations (Fig. 5, lanes 2 and 4) and therefore displaying a shortened protein.
Menin expression and MEN1 mutations
Immunohistochemical data were available for 71 samples screened for MEN1 mutations (Table 4). A significant association was found between the lack of menin nuclear expression and the presence of truncating mutations (Fisher's test, odds ratio (OR) 5.1, 95% CI 1.0–25.9; P=0.025). A significant correlation was also observed between the intensity of cytoplasmic positivity and the presence of mutation regardless of the mutation type (Fisher's test, P=0.006). In fact, all the mutated cases had moderate to strong cytoplasmic staining (Table 3). We also sought to investigate the combined effect of nuclear and cytoplasmic immunohistochemical staining in relation to the mutational status of MEN1. Interestingly, in a multivariate logistic model, we estimated that an increase of one unit of cytoplasmic score had the effect to double the probability of carrying a mutation (OR 2.2; 95% CI 1.0–4.9; P=0.033), whereas the opposite effect was observed for nuclear score (OR 0.5; 95% CI 0.5–0.9; P=0.021).
Frequency of multiple endocrine neoplasia type 1 (MEN1) mutations by immunohistochemical score and mutation type
Mutation typea | |||||
---|---|---|---|---|---|
IHC scoreb | Missense | Splicing | Truncating | Total mutated (n=21) | Wild-type (n=50) |
Nuclear | |||||
0 | 1 | 0 | 5 | 6 | 6 |
1 | 2 | 0 | 3 | 5 | 13 |
2 | 2 | 3 | 4 | 9 | 18 |
3 | 1 | 0 | 0 | 1 | 13 |
Cytoplasmic | |||||
0 | 0 | 0 | 0 | 0 | 2 |
1 | 0 | 1 | 0 | 1 | 11 |
2 | 4 | 1 | 8 | 13 | 25 |
3 | 2 | 1 | 4 | 7 | 12 |
Splicing, mutation occurring in splice sites; truncating, both nonsense and frameshift mutations that lead to a premature stop codon.
IHC score refers to the staining intensity that was classified into four grades: 0 (absent), 1 (weak), 2 (moderate), and 3 (strong).
Correlation of MEN1 gene status and protein expression with clinical–pathological information
There was no significant correlation between MEN1 mutations or polymorphisms and clinical–pathological parameters, including WHO classification, proliferation index, and progression-free and overall survival. Nuclear immunostaining had no significant correlation with any clinical–pathological parameter.
Discussion
The results of our combined genetic and immunohistochemical study of MEN1 gene and protein status in 169 apparently sporadic PETs may be summarized as follows: i) MEN1 gene mutations were found in 27/100 cases, including 23/75 (30%) non-functioning PETs and 4/25 (16%) functioning PETs; two of the cases revealed to harbor germline mutations in the lack of any familial cancer history; ii) mutations were distributed along the gene and most were of the loss of function type; iii) menin IHC showed that normal islet cells had intense nuclear and very faint cytoplasmic staining, while 56/140 (40%) PETs lacked or had weak nuclear immunostaining and 114/140 (80%) had cytoplasmic intensity higher than normal; iv) about one-third of the cases showing an altered localization and intensity of menin coincided with those harboring a MEN1 gene mutation.
This study involved the largest panel of PETs that has been ever screened for mutations of MEN1, and concluded that about 25% of sporadic PETs harbor a somatic MEN1 mutation. Of the 27 mutations, 2 revealed to be germline, suggesting that a small proportion of cases may indeed have a MEN1 syndrome acquired at the time of zygote formation or inherited from an asymptomatic carrier parent.
MEN1 mutations were scattered throughout the coding sequence and splice sites. Although the majority of mutations were localized in the 5′ half of the gene, no obvious hot spot was identified. The alterations affected predominantly the evolutionary conserved regions of menin, and are likely to result in the alteration of protein function. In fact, 12 of the 25 somatic mutations are predicted to give a truncated menin (numbers 1–3, 12–13, 17–19, and 20–23, Table 2) that would lack at least one of the NLS sites due to premature stop codon, thus preventing the protein to enter the nucleus and exert its function. Six missense mutations (numbers 4–7 and 9–10) involve JunD-binding domains, and are therefore predicted to prevent menin-repressive action on JunD-mediated transcription (Agarwal et al. 1999). The effects of the remaining two missense mutations, one in frame deletion, and four splice site alterations (numbers 8 and 11, 15, and 24–27 respectively) are difficult to predict. However, it has been previously demonstrated by in vitro assays that missense mutations could lead to a loss of function due to an enhanced proteolytic degradation of menin (Yaguchi et al. 2004), whereas the in-frame alteration could delete crucial amino acid residues. Those mutated proteins were also hypothesized to affect the interaction with other known menin interactors (Fig. 1B; Ohkura et al. 2001, Jin et al. 2003, Sukhodolets et al. 2003). Finally, it is likely that the pathogenic effects of MEN1 mutations are shown through disruption of multiple and cross-linked interaction pathways.
We also found four common polymorphisms that are often classified as benign. However, several pieces of evidence suggest that the non-synonymous A541T and R171Q polymorphisms that were found in 14 and 7 cases respectively might be considered as pathogenic missense mutations (Shan et al. 1998), which may alter protein function and exhibit higher clinical expression in some genetic context (Bazzi et al. 2008, De Carlo et al. 2008). The frequencies of these non-synonymous polymorphisms in our cohort cases are 14 and 7% for A541T and R171Q respectively. Based on the data reported in the HapMap Project (http://hapmap.ncbi.nlm.nih.gov/), the frequencies of these polymorphisms in the European population are 1 and 2% for A541T and R171Q respectively; therefore, we can speculate that the expression of these genetic changes has eventually a clinical significance.
All the gene alterations identified did not correlate with clinical–pathological features including tumor stage, suggesting that MEN1 alteration is an early and common event in PET pathogenesis. The fact that a large proportion of PETs (70%) do not harbor MEN1 mutations suggest that other genes or other mechanisms of MEN1 inactivation that have been only marginally investigated might be involved in PET tumorigenesis (Arnold et al. 2007).
Despite numerous data have been accumulated about the mutational status of MEN1 in PET cases, only a limited number of studies deal with the expression pattern of menin in human pancreas and PET. Indeed, Cavallari et al. (2003) provided the first description of the expression pattern of menin in human pancreas using an in-house characterized monoclonal antibody (C126) to perform an indirect IF/laser-scanning microscopy protocol on frozen tissues. They described that the protein is mainly located in the nucleus of all pancreatic cell types. They also analyzed seven pancreaticoduodenal endocrine tumors showing variable expression levels of menin compared to normal tissue.
In a recent study, the same protocol was applied to examine menin expression pattern in normal exocrine pancreas and in pancreatic ductal adenocarcinoma (Cavallari et al. 2009) using interchangeably the mouse monoclonal antibody and the A300-105A polyclonal antibody from Bethyl Laboratories. Their analysis of normal exocrine pancreas confirmed that the protein mainly localizes in the nucleus of both acinar and duct cells, but it is also detected in the cytoplasm. The A300-105A was the very same antibody we used for our immunohistochemical analysis of PET tissues. This antibody gave, in our hands, the most reproducible results compared with other commercially available antibodies tested. Indeed, the specificity of this antibody in immunostaining as well as in western blot experiments was addressed in competitive assays using the commercially available blocking peptide from Bethyl Laboratories.
In the present study, immunohistochemical expression of menin was seen in all cell types of normal pancreas, with the protein mainly localized into the nucleus. Cytoplasmic signal was very faint to absent in islet cells and moderate in acini and ducts. Most tumors expressed menin with a broad spectrum of immunoreactive intensity, which varied from a level comparable to that detected in normal tissue to an almost undetectable signal, and further, with variable levels of protein distribution between nucleus and cytoplasm.
Indeed, immunohistochemical data suggested that a localization defect of menin exists in most cases, including those lacking or with very faint nuclear staining and those with cytoplasmic immunostaining more intense than the faint signal seen in normal islets. Half of the cases lacking nuclear menin and one-third of cases showing either weak nuclear or moderate to intense cytoplasmic protein carried MEN1 mutations. To further confirm the menin expression pattern observed in tumors, we performed an IF analysis of human PET cell lines screened for gene mutations. This analysis revealed that menin signal is still detected in the cytoplasm of the cells despite the absence of evident genetic abnormalities.
Interestingly, the absence of nuclear protein expression correlates with the presence of truncating mutations, while positive cytoplasmic protein intensity is associated with the presence of mutation regardless of its predicted effect on the protein functionality. These results suggest that some of the truncating mutations cause the premature block of transcription generating an incomplete protein that accumulates into the cytoplasm due to the lack of at least one of the NLSs. Non-truncating mutations could partially affect nuclear transport as well, possibly causing an incomplete protein processing or preventing menin interaction with other proteins. This can explain the abundance of cytoplasmic protein observed in mutated cases.
Western blot data, demonstrating protein expression in tumors carrying different types of MEN1 gene mutations, are in line with this hypothesis. Indeed, a band of the expected normal molecular size was detected in both a wild-type tumor (i.e. displaying no sequence abnormalities) and in a tumor carrying a missense mutation, whereas bands of lower molecular mass were observed in tumors affected by truncating mutations and these were of the size predicted by the truncating mutations observed.
At partial support to our hypothesis concerning a localization defect of menin in PETs, there is a recent paper by Cao et al. (2009), which demonstrated in vitro that menin is able to shuttle between nucleus and cytoplasm through newly identified nuclear export signals (NES). More interestingly, menin was shown to directly interact with β-catenin and, when overexpressed, to prevent the nuclear accumulation of β-catenin carrying the protein outside the nucleus. Finally, they demonstrated that mutations within these domains impair the nuclear export function of menin. Although none of the mutations we identified in our study affects the NES domain, the results of this very interesting paper suggest that abnormal interaction involving menin might alter the cellular localization of its physical interactor (e.g. β-catenin). These data allow us to speculate that an inverse situation, which is the mislocation of menin, might occur in the event of alterations affecting one of its physical interactors.
In conclusion, our data suggest that gene mutations affect the subcellular localization of menin causing its accumulation in the cytoplasm. However, the majority of PETs show abnormal menin distribution in the lack of MEN1 gene mutations, suggesting that other genes, either independently or more probably as physical and/or functional partners of MEN1, are altered and are responsible for the impairment of menin functionality.
Supplementary data
This is linked to the online version of the paper at http://dx.doi.org/10.1677/ERC-10-0028.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This work was supported by Associazione Italiana Ricerca Cancro (AIRC), Milan, Italy; Fondazione CariPaRo, Padova, Italy; Fondazione Cariverona, Verona, Italy (A Scarpa); Ministero della Salute, Rome, Italy; Fondazione Giorgio Zanotto, Verona, Italy; Fondazione Monte dei Paschi di Siena; German Federal Ministry of Education and Research (BMBF) grant no. 01GR0805 to K-F Becker; European Community FP VI Program Grant PL018771 (MolDiagPaca) and PN037211 (IMPACTS).
Acknowledgements
We thank Prof. Aurel Perren, Institute of Pathology, University of Bern, Switzerland, for critical and helpful discussion and suggestions.
References
Achille A, Biasi MO, Zamboni G, Bogina G, Magalini AR, Pederzoli P, Perucho M & Scarpa A 1996 Chromosome 7q allelic losses in pancreatic carcinoma. Cancer Research 56 3808–3813.
Agarwal SK, Guru SC, Heppner C, Erdos MR, Collins RM, Park SY, Saggar S, Chandrasekharappa SC, Collins FS & Spiegel AM et al. 1999 Menin interacts with the AP1 transcription factor JunD and represses JunD-activated transcription. Cell 96 143–152.
Agarwal SK, Novotny EA, Crabtree JS, Weitzman JB, Yaniv M, Burns AL, Chandrasekharappa SC, Collins FS, Spiegel AM & Marx SJ 2003 Transcription factor JunD, deprived of menin, switches from growth suppressor to growth promoter. PNAS 100 10770–10775.
Arnold CN, Sosnowski A, Schmitt-Graff A, Arnold R & Blum HE 2007 Analysis of molecular pathways in sporadic neuroendocrine tumors of the gastro-entero-pancreatic system. International Journal of Cancer 120 2157–2164.
Balakrishnan A, Bleeker FE, Lamba S, Rodolfo M, Daniotti M, Scarpa A, van Tilborg AA, Leenstra S, Zanon C & Bardelli A 2007 Novel somatic and germline mutations in cancer candidate genes in glioblastoma, melanoma, and pancreatic carcinoma. Cancer Research 67 3545–3550.
Bazzi W, Renon M, Vercherat C, Hamze Z, Lacheretz-Bernigaud A, Wang H, Blanc M, Roche C, Calender A & Chayvialle JA et al. 2008 MEN1 missense mutations impair sensitization to apoptosis induced by wild-type menin in endocrine pancreatic tumor cells. Gastroenterology 135 1698–1709 e1692.
Becker KF, Schott C, Hipp S, Metzger V, Porschewski P, Beck R, Nahrig J, Becker I & Hofler H 2007 Quantitative protein analysis from formalin-fixed tissues: implications for translational clinical research and nanoscale molecular diagnosis. Journal of Pathology 211 370–378.
Bergman L, Boothroyd C, Palmer J, Grimmond S, Walters M, Teh B, Shepherd J, Hartley L & Hayward N 2000 Identification of somatic mutations of the MEN1 gene in sporadic endocrine tumours. British Journal of Cancer 83 1003–1008.
Cao Y, Liu R, Jiang X, Lu J, Jiang J, Zhang C, Li X & Ning G 2009 Nuclear-cytoplasmic shuttling of menin regulates nuclear translocation of β-catenin. Molecular and Cellular Biology 29 5477–5487.
Capelli P, Martignoni G, Pedica F, Falconi M, Antonello D, Malpeli G & Scarpa A 2009 Endocrine neoplasms of the pancreas: pathologic and genetic features. Archives of Pathology & Laboratory Medicine 133 350–364.
Capurso G, Lattimore S, Crnogorac-Jurcevic T, Panzuto F, Milione M, Bhakta V, Campanini N, Swift SM, Bordi C & Delle Fave G et al. 2006 Gene expression profiles of progressive pancreatic endocrine tumours and their liver metastases reveal potential novel markers and therapeutic targets. Endocrine-Related Cancer 13 541–558.
Cavallari I, D'Agostino DM, Ferro T, Rosato A, Barzon L, Pasquali C, Fogar P, Theodoropoulou M, Esposito G & Boscaro M et al. 2003 In situ analysis of human menin in normal and neoplastic pancreatic tissues: evidence for differential expression in exocrine and endocrine cells. Journal of Clinical Endocrinology and Metabolism 88 3893–3901.
Cavallari I, Silic-Benussi M, Rende F, Martines A, Fogar P, Basso D, Vella MD, Pedrazzoli S, Herman JG & Chieco-Bianchi L et al. 2009 Decreased expression and promoter methylation of the menin tumor suppressor in pancreatic ductal adenocarcinoma. Genes, Chromosomes and Cancer 48 383–396.
Cecconi D, Donadelli M, Rinalducci S, Zolla L, Scupoli MT, Scarpa A, Palmieri M & Righetti PG 2007 Proteomic analysis of pancreatic endocrine tumor cell lines treated with the histone deacetylase inhibitor trichostatin A. Proteomics 7 1644–1653.
Chakrabarti R, Srivatsan ES, Wood TF, Eubanks PJ, Ebrahimi SA, Gatti RA, Passaro E Jr & Sawicki MP 1998 Deletion mapping of endocrine tumors localizes a second tumor suppressor gene on chromosome band 11q13. Genes, Chromosomes and Cancer 22 130–137.
Chandrasekharappa SC, Guru SC, Manickam P, Olufemi SE, Collins FS, Emmert-Buck MR, Debelenko LV, Zhuang Z, Lubensky IA & Liotta LA et al. 1997 Positional cloning of the gene for multiple endocrine neoplasia-type 1. Science 276 404–407.
De Carlo E, Pilon C, Zatelli MC, degli Uberti EC & Fallo F 2008 Isolated R171Q amino acid change in MEN1 gene: polymorphism or mutation? Clinical Endocrinology 69 511.
Eubanks PJ, Sawicki MP, Samara GJ, Gatti R, Nakamura Y, Tsao D, Johnson C, Hurwitz M, Wan YJ & Passaro E Jr 1994 Putative tumor-suppressor gene on chromosome 11 is important in sporadic endocrine tumor formation. American Journal of Surgery 167 180–185.
Gortz B, Roth J, Krahenmann A, de Krijger RR, Muletta-Feurer S, Rutimann K, Saremaslani P, Speel EJ, Heitz PU & Komminoth P 1999 Mutations and allelic deletions of the MEN1 gene are associated with a subset of sporadic endocrine pancreatic and neuroendocrine tumors and not restricted to foregut neoplasms. American Journal of Pathology 154 429–436.
Guru SC, Goldsmith PK, Burns AL, Marx SJ, Spiegel AM, Collins FS & Chandrasekharappa SC 1998 Menin, the product of the MEN1 gene, is a nuclear protein. PNAS 95 1630–1634.
Heppner C, Bilimoria KY, Agarwal SK, Kester M, Whitty LJ, Guru SC, Chandrasekharappa SC, Collins FS, Spiegel AM & Marx SJ et al. 2001 The tumor suppressor protein menin interacts with NF-κB proteins and inhibits NF-κB-mediated transactivation. Oncogene 20 4917–4925.
Jin S, Mao H, Schnepp RW, Sykes SM, Silva AC, D'Andrea AD & Hua X 2003 Menin associates with FANCD2, a protein involved in repair of DNA damage. Cancer Research 63 4204–4210.
Jonkers YM, Claessen SM, Feuth T, van Kessel AG, Ramaekers FC, Veltman JA & Speel EJ 2006 Novel candidate tumour suppressor gene loci on chromosomes 11q23–24 and 22q13 involved in human insulinoma tumourigenesis. Journal of Pathology 210 450–458.
Kaji H, Canaff L, Lebrun JJ, Goltzman D & Hendy GN 2001 Inactivation of menin, a Smad3-interacting protein, blocks transforming growth factor type β signaling. PNAS 98 3837–3842.
Kim H, Lee JE, Cho EJ, Liu JO & Youn HD 2003 Menin, a tumor suppressor, represses JunD-mediated transcriptional activity by association with an mSin3A-histone deacetylase complex. Cancer Research 63 6135–6139.
Kononen J, Bubendorf L, Kallioniemi A, Barlund M, Schraml P, Leighton S, Torhorst J, Mihatsch MJ, Sauter G & Kallioniemi OP 1998 Tissue microarrays for high-throughput molecular profiling of tumor specimens. Nature Medicine 4 844–847.
La P, Schnepp RW, Petersen CD, Silva AC & Hua X 2004 Tumor suppressor menin regulates expression of insulin-like growth factor binding protein 2. Endocrinology 145 3443–3450.
La P, Desmond A, Hou Z, Silva AC, Schnepp RW & Hua X 2006 Tumor suppressor menin: the essential role of nuclear localization signal domains in coordinating gene expression. Oncogene 25 3537–3546.
Larsson C, Skogseid B, Oberg K, Nakamura Y & Nordenskjold M 1988 Multiple endocrine neoplasia type 1 gene maps to chromosome 11 and is lost in insulinoma. Nature 332 85–87.
Lemmens I, Van de Ven WJ, Kas K, Zhang CX, Giraud S, Wautot V, Buisson N, De Witte K, Salandre J & Lenoir G et al. 1997 Identification of the multiple endocrine neoplasia type 1 (MEN1) gene. The European Consortium on MEN1. Human Molecular Genetics 6 1177–1183.
Marx SJ, Agarwal SK, Kester MB, Heppner C, Kim YS, Skarulis MC, James LA, Goldsmith PK, Saggar SK & Park SY et al. 1999 Multiple endocrine neoplasia type 1: clinical and genetic features of the hereditary endocrine neoplasias. Recent Progress in Hormone Research 54 397–438.
Milne TA, Hughes CM, Lloyd R, Yang Z, Rozenblatt-Rosen O, Dou Y, Schnepp RW, Krankel C, Livolsi VA & Gibbs D et al. 2005 Menin and MLL cooperatively regulate expression of cyclin-dependent kinase inhibitors. PNAS 102 749–754.
Missiaglia E, Dalai I, Barbi S, Beghelli S, Falconi M, della Peruta M, Piemonti L, Capurso G, Di Florio A & delle Fave G et al. 2010 Pancreatic endocrine tumors: expression profiling evidences a role for AKT–mTOR pathway. Journal of Clinical Oncology 28 245–255.
Moore PS, Missiaglia E, Antonello D, Zamo A, Zamboni G, Corleto V, Falconi M & Scarpa A 2001 Role of disease-causing genes in sporadic pancreatic endocrine tumors: MEN1 and VHL. Genes, Chromosomes and Cancer 32 177–181.
Ohkura N, Kishi M, Tsukada T & Yamaguchi K 2001 Menin, a gene product responsible for multiple endocrine neoplasia type 1, interacts with the putative tumor metastasis suppressor nm23. Biochemical and Biophysical Research Communications 282 1206–1210.
Pannett AA & Thakker RV 1999 Multiple endocrine neoplasia type 1. Endocrine-Related Cancer 6 449–473.
Rigaud G, Missiaglia E, Moore PS, Zamboni G, Falconi M, Talamini G, Pesci A, Baron A, Lissandrini D & Rindi G et al. 2001 High resolution allelotype of nonfunctional pancreatic endocrine tumors: identification of two molecular subgroups with clinical implications. Cancer Research 61 285–292.
Schnepp RW, Mao H, Sykes SM, Zong WX, Silva A, La P & Hua X 2004 Menin induces apoptosis in murine embryonic fibroblasts. Journal of Biological Chemistry 279 10685–10691.
Shan L, Nakamura Y, Nakamura M, Yokoi T, Tsujimoto M, Arima R, Kameya T & Kakudo K 1998 Somatic mutations of multiple endocrine neoplasia type 1 gene in the sporadic endocrine tumors. Laboratory Investigation 78 471–475.
Speel EJ, Richter J, Moch H, Egenter C, Saremaslani P, Rutimann K, Zhao J, Barghorn A, Roth J & Heitz PU et al. 1999 Genetic differences in endocrine pancreatic tumor subtypes detected by comparative genomic hybridization. American Journal of Pathology 155 1787–1794.
Sukhodolets KE, Hickman AB, Agarwal SK, Sukhodolets MV, Obungu VH, Novotny EA, Crabtree JS, Chandrasekharappa SC, Collins FS & Spiegel AM et al. 2003 The 32-kilodalton subunit of replication protein A interacts with menin, the product of the MEN1 tumor suppressor gene. Molecular and Cellular Biology 23 493–509.
Yaguchi H, Ohkura N, Takahashi M, Nagamura Y, Kitabayashi I & Tsukada T 2004 Menin missense mutants associated with multiple endocrine neoplasia type 1 are rapidly degraded via the ubiquitin–proteasome pathway. Molecular and Cellular Biology 24 6569–6580.
Yokoyama A, Wang Z, Wysocka J, Sanyal M, Aufiero DJ, Kitabayashi I, Herr W & Cleary ML 2004 Leukemia proto-oncoprotein MLL forms a SET1-like histone methyltransferase complex with menin to regulate Hox gene expression. Molecular and Cellular Biology 24 5639–5649.
(V Corbo and I Dalai contributed equally to this work)