| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Oncology and Surgical Sciences (I.C., D.M.D., T.F., A.R., G.E., L.C.-B., V.C.), University of Padova, I-35128 Padova, Italy; Department of Medical and Surgical Sciences (L.B., C.P., P.F., E.T., F.F.), I-35123 Padova, Italy; Max-Planck Institute of Psychiatry (M.T., U.P.), D-80804 Munich, Germany; and Endocrinology Clinic (M.B.), University of Ancona, 60020 Torrette-Ancona, Italy
Address all correspondence and requests for reprints to: Vincenzo Ciminale, M.D., Dipartimento di Scienze Oncologiche e Chirurgiche, Universita’ di Padova, Via Gattamelata 64, 35128 Padova, Italy. E-mail: v.ciminale{at}unipd.it.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Genetic analyses have revealed a large number of mutations within the MEN1 gene in MEN1 tumors and in a subset of sporadic parathyroid adenomas, insulinomas, gastrinomas, and pulmonary carcinoids (reviewed in Ref. 3). Most mutations are nonsense or frameshift types resulting in premature truncation of the protein (reviewed in Ref. 3), suggesting that menin acts as a tumor suppressor. In line with such a role, menin was shown to suppress tumorigenicity of Ras in vitro (7). Furthermore, knockout mice carrying germline monoallelic deletions of the MEN1 gene develop endocrine lesions resembling human MEN1 (8).
The MEN1 mRNA was detected by Northern blotting in a wide spectrum of adult human tissues (4, 5). Immunoblot assays of human fetal tissues revealed menin expression in many organs, except in liver, lung, pancreas, and skin (9). The MEN1 mRNA and menin protein were detected by RT-PCR and immunoblotting in primary and secondary hyperparathyroid lesions (10). Immunoblot analyses indicated that expression of menin is increased in sporadic pituitary adenomas, compared with normal pituitary tissue (11). Based on both RT-PCR and immunoblotting, menin expression was reported to be down-regulated, but not abolished, in MEN1 tumors, including cases carrying biallelic ablative mutations of the MEN1 gene (10).
Results obtained using techniques that analyze homogenized tissues (e.g. immunoblot, Northern blot, and RT-PCR) can be difficult to interpret because of the presence of stromal components. In situ RNA hybridization, a technique that overcomes such problems, demonstrated expression of the MEN1 transcript in a wide array of adult human tissues, with the strongest signal detected in placenta (12). To date, limited information is available on the pattern of menin protein expression in situ in normal tissues and in sporadic and MEN1 tumors, mostly because of the lack of an effective antibody and detection protocol. This report describes an antimenin monoclonal antibody and indirect immunofluorescence/laser scanning microscopy (IF/LSM) protocol developed to detect human menin in situ. We provide the first description of the protein’s expression in frozen sections of normal pancreas and endocrine pancreatic tumors, which revealed interesting differences in expression levels among exocrine and islet cell populations.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Nontumoral samples. Nontumoral pancreatic tissues were obtained from a type II diabetes patient (sample no. 3) and from patients who underwent surgical resection of sporadic pancreatic tumors [an insulinoma (sample no. 1), a nonsecreting endocrine tumor (sample no. 2), a well-differentiated sporadic gastrinoma (sample no. 7B), a well-differentiated sporadic nonfunctioning endocrine tumor (sample no. PF28), two ductal pancreatic adenocarcinomas (no. PF14 and PF30), a pseudopapillary carcinoma (no. PF26), and an adenocarcinoma of the papilla of Vater (sample no. PF35)].
Tumor samples. The following tumor samples were analyzed. Case no. 10A was a sporadic gastrinoma (maximum diameter, 1.6 cm) of the uncinate process of the pancreas, microscopically infiltrating the duodenum and peripancreatic fat, resected from a 70-yr-old male with Zollinger-Ellison syndrome; it scored positive in immunohistochemistry for chromogranin, neuron-specific enolase (NSE), and gastrin. Case no. 12A was a massive lymph node metastasis (2.2 cm) arising from a small (<2 mm) gastrinoma in the duodenal wall of a 51-yr-old woman who, in the past, had undergone a total parathyroidectomy and suffered from hypergastrinemia; although her clinical and family history were consistent with MEN1 syndrome, no germ line mutations were found in the coding portion of MEN1. Case no. 14B was a sporadic, well-differentiated insulinoma (maximum diameter, 1.7 cm) resected from the pancreatic head of a 39-yr-old male who presented hypoglycemic syndrome; it was chromogranin +, NSE +, insulin +, Ki 67 less than 0.9%, with no evidence of mitosis. Case no. 19B was a sporadic, well-differentiated insulinoma (maximum diameter, 2.7 cm) resected from the pancreatic head of a 24-yr-old male with hypoglycemic syndrome; it was chromogranin +, NSE +, insulin, Ki 67 = 2.1%, with no signs of mitosis. Case no. 21A was a sporadic, well-differentiated insulinoma (maximum diameter, 0.8 cm) resected from the pancreatic head of a 20-yr-old male with hypoglycemic syndrome; it was chromogranin +, NSE +, somatostatin +, insulin + (only 30% of cells). Case no. 13A was a sporadic duodenal somatostatinoma (maximum diameter, 2.0 cm) that presented infiltration of the duodenal wall and massive lymph node metastases; it was resected from a 53-yr-old male with no clinical symptoms; the sample showed diffuse positivity for somatostatin and NSE (>90% of cells); 20% of the cells were positive for chromogranin. Case no. 15A was a duodenal gastrinoma (maximum diameter < 5 mm) resected from a 37-yr-old male MEN1 patient whose clinical history included a prolactin-secreting pituitary adenoma, parathyroid hyperplasia, hypergastrinemia, stomach and lung carcinoids, pancreatic PPomas, and a family history of MEN1; in addition to the duodenal gastrinoma, the pathologic picture included multiple duodenal microadenomas, multiple gastric neuroendocrine tumors, multiple pancreatic nonfunctioning neuroendocrine tumors, periduodenal lymph node metastases, and liver metastasis of neuroendocrine origin. The duodenal tumor scored NSE + and gastrin + by immunohistochemistry. Genetic analysis revealed a germ-line 11-bp deletion in exon 9 of MEN1 and loss of the second MEN1 allele in the tumor.
Informed consent for tissue analyses was obtained from living patients or from relatives. Tissues were snap-frozen in dry ice immediately after surgery and stored at -80 C.
Genetic analyses
DNA was isolated from frozen tissues by proteinase K digestion and phenol/chloroform extraction. Exons 2–10 of MEN1 and adjacent splice junctions were amplified by PCR as 15 partially overlapping fragments (5, 13). Sequence analysis was performed on both DNA strands using an ABI PRISM 310 DNA sequencer and the ABI PRISM BigDye terminator cycle sequencing kit (Applied Biosystems, Foster City, CA).
Generation of a menin-specific monoclonal antibody
The portion of MEN1 coding for the C-terminal 126 amino acids of menin (C126men) was PCR-amplified from human genomic DNA using primers M1 (5'-ccgggagtccaagccagaggag-3') and M2 (5'-agttcagaggcctttgcgctgc-3') (synthesized by Life Technologies, Glasgow, Scotland). The resulting product was cloned into pBluescript KS+ (Stratagene, La Jolla, CA) and then subcloned as a BamHI-EcoRI fragment into the pGEX-2T vector (Amersham Pharmacia Biotech, Uppsala, Sweden) "in frame" with the glutathione-S-transferase (GST) gene. GST-C126men fusion protein was expressed in Escherichia coli strain BL21 after induction with isopropyl ß-D-thiogalactopyranoside. The bacteria were lysed in 50 mM Tris-HCl (pH 8.0), 25% sucrose, 1 mM EDTA, 0.5% Triton X-100, and 100 µg/ml lysozyme (Roche, Basel, Switzerland) followed by centrifugation at 17,500 x g at 4 C for 10 min. GST-C126men was purified from the resulting supernatant, by affinity chromatography, using glutathione-Sepharose (Amersham Pharmacia Biotech), as described elsewhere (14), followed by SDS-PAGE; pulverized gel bands containing the protein were used to immunize BALB/c mice. Serological reactivity was tested by dot blot using GST-C126men and by IF using a HeLa-derived cell line (HLtat, Ref. 15) transiently transfected with a eukaryotic expression plasmid coding for the C-terminal 125 amino acids of menin (pLs-C125men-AU). Splenocytes from 1 positive animal were fused to NSO mouse myeloma cells, selected in HAT medium and cloned by limiting dilution (16). Culture supernatants were screened for menin reactivity using dot blot, IF, and immunohistochemistry on paraffin-embedded sections of HLtat cells transfected with pLs-C125men-AU. Of over 500 clones, only 1 (i.e. mAb C126) proved to be positive in all these assays and was inoculated ip into severe combined immunodeficiency (SCID) mice to generate ascites. Procedures involving mice and their care were monitored by a University of Padova veterinarian and conformed to institutional guidelines that comply with Italian and international laws and policies.
Immunohistochemistry
HLtat cells were transiently transfected with the menin cDNA expression vector pCMVsport A11 (kindly provided by S. J. Marx and S. C. Chandrasekharappa). One day later, the cells were detached with trypsin, rinsed in PBS, paraffin-embedded, and analyzed by immunohistochemistry as described elsewhere (17). In brief, sections were subjected to three cycles of microwave pretreatment at 750 W for 5 min and then incubated with a 1:10 dilution of normal horse serum for 30 min at room temperature, and treated with the Vector Avidin/Biotin Blocking Kit (Vector Laboratories, Inc., Burlingame, CA). Sections were then incubated overnight at 4 C with mAb C126 (1:100) followed by 30-min incubations at room temperature with biotinylated horse antimouse antibody (1:300; Vector Laboratories, Inc.) and avidin-biotin-peroxidase complex (Vectastain Elite Kit; Vector Laboratories, Inc.). Peroxidase activity was detected using diaminobenzidine (1 mg/ml; Sigma, St. Louis, MO) and 0.01% H2O2. The cells were counterstained with hematoxylin, dehydrated, and mounted with Entellan (Merck, Darmstadt, Germany).
IF/LSM
Frozen tissues were cryosectioned into 5-µm sections, air-dried for 30 min, fixed in phosphate-buffered 4% paraformaldehyde for 5 min, rinsed twice in PBS, and then incubated for 4 min in 70% ethanol and for 5 min in 96% ethanol. The tissues were either analyzed immediately or stored in 96% ethanol at 4 C for no longer than 24 h, because prolonged storage in ethanol after paraformaldehyde treatment resulted in a substantial decrease in menin reactivity. IF was performed by incubating the tissues with normal goat serum (1:70) in PBS for 30 min at room temperature and then overnight at 4 C with a 1:80 dilution of mAb C126, followed by a 1:500 dilution of Alexa 488-conjugated goat antimouse secondary antibody (Molecular Probes, Eugene, OR). Dual labeling experiments were carried out by staining with 0.1 µg/ml propidium iodide or with rabbit antibodies specified in the figure legends, followed by secondary antibodies conjugated to either Texas red (Jackson Laboratories, Bar Harbor, ME) or Alexa 546 (Molecular Probes). LSM was carried out with a Zeiss LSM 510 microscope (Carl Zeiss, Jena, Germany) using Argon (488 nm) and Helium-Neon (543 nm) lasers with objectives and settings as specified in the figure legends. Laser intensity, pinhole aperture, and photomultiplier parameters were standardized to allow comparison of signals obtained in different samples. Fluorescence signals were analyzed using a 505- to 530-nm band-pass filter for Alexa 488 and a long-pass 560-nm filter for the propidium iodide, Texas red, and Alexa 546 signals. Quantitations were carried out with Zeiss’s Profile software tool. After IF analysis, sections were subjected to standard hematoxylin-eosin staining, and fields matching IFs were analyzed by LSM using three lasers (Ar 488 nm, HeNe 543, HeNe 633) and visualized by red, blue, and green false-color overlay.
The specificity of the IF pattern obtained with mAb C126 was verified in a competition assay carried out by preadsorbing mAb C126 ascites with glutathione-Sepharose-attached GST or GST-C126men (approximately 5 µg), for 5 h at 4 C, before incubation with tissue sections.
Immunoblots
The ability of mAb C126 to recognize menin in immunoblots was tested as follows. Hltat cells were transfected with either pCMVsport A11 or a control plasmid (pBluescript KS+) and lysed by shearing five times through a 26-G needle in a solution containing 50 mM Tris (pH 6.8), 2% sodium dodecyl sulfate, 1% ß-mercaptoethanol, 5% glycerol, 0.05% bromophenol blue, and a protease inhibitor cocktail (Complete; Roche). Resulting lysates were loaded in triplicate on a 10% T/2.6% C denaturing polyacrylamide gel along with prestained size markers (Cell Signaling Technology, Beverly, MA), electrophoresed, and electrotransferred to nitrocellulose. Membranes were cut into strips, blocked in 1% nonfat dried milk (Sigma), and incubated with either mAb C126 (diluted 1:1,000) or with commercially available antibodies raised against the N terminus or C terminus of menin (Santa Cruz Biotechnology, Santa Cruz, CA) followed by horseradish peroxidase-conjugated sheep antimouse (Amersham Pharmacia Biotech) or donkey antigoat (Santa Cruz Biotechnology) antibodies. The strips were developed using chemiluminescence reagents (Supersignal Pico; Pierce, Rockford, IL), realigned, and exposed to Hyperfilm MP (Amersham Pharmacia Biotech). For immunoblots of pancreatic tissue samples, frozen sections (25-µm thick) were solubilized as described above, and analyzed by SDS-PAGE/immunoblotting using a mouse antitubulin antibody (Sigma). After balancing the sample quantities based on the tubulin signal, SDS-PAGE/immunoblotting was repeated with mAb C126 as described above, with the exception that Supersignal femto was used as developer. All antigen-antibody reactions and washing steps were carried out in the presence of 0.05% Tween 20.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
A 126-amino-acid C-terminal portion of menin was expressed as a GST-fusion protein, purified, and used as an immunogen to generate a monoclonal antibody (mAb C126, see Materials and Methods) recognizing menin in situ. As shown in Fig. 1
, immunohistochemistry (A) and IF/LSM (B) analyses, carried out using mAb C126, yielded a strong nuclear signal in cells transfected with a menin expression plasmid. The nuclear localization of menin is more evident in B, which shows an overlay IF/LSM image obtained using mAb C126 (green signal) and an antibody against the nuclear export protein CRM-1, added to visualize the nucleus (red signal).
|
As shown in Fig. 1G
, mAb C126 recognized a band of approximately 68 kDa in immunoblots of cells transfected with the menin expression vector pCMVsport-A11 (Hltat, lanes +); this band, whose apparent size is consistent with that published for menin, was also detected by two commercially available antimenin antibodies raised against C-terminal and N-terminal portions of the protein. Immunoblots of lysates prepared from three normal pancreas samples and a sporadic insulinoma confirmed that mAb C126 recognized the 68-kDA menin band in all of the samples (Fig. 1G, pancreatic tissues).
In situ detection of menin in nontumoral pancreas by IF/LSM
Figure 2 shows results of IF/LSM analyses carried out on a frozen section of human nontumoral pancreas (sample no. 1). Menin was detected using mAb C126 and Alexa 488-conjugated antimouse antibody (green signal, A) and nuclei were identified by propidium iodide staining (red signal, B). The area shown includes exocrine cells and three islets (indicated by arrows), whose identification was facilitated by staining the section with hematoxylin/eosin after IF (D). Results showed strong antimenin reactivity in most exocrine cells (A); overlap of the menin and propidium iodide signals produced a yellow-green color in the overlay image (C), thus demonstrating nuclear accumulation of endogenous menin in human tissues, and confirming results obtained with ectopic expression systems (Fig. 1 and Ref. 6). Interestingly, in contrast to adjacent exocrine cells, the islets exhibited substantially lower levels of menin expression, yielding a red-orange signal in the overlay (C). An area spanning the islet-exocrine boundary (white square in C) was examined at higher magnification (E) and analyzed with the Profile software tool (F) to quantitate the menin and propidium iodide signals across several exocrine and islet cells (see arrow in E). Results of this analysis are plotted in a graph showing the relative intensities of the signals in the two channels (y-axis) against the distance, in micrometers, from the origin of the arrow. Results confirmed the overlap of the menin and propidium iodide peaks, and demonstrated that menin expression in islet cells was about 4- to 5-fold lower than in exocrine cells (e.g. compare two exocrine nuclei corresponding to the two peaks at about 5 and 30 µm, with islet cell nuclei at about 40, 55, 62, and 68 µm). Similar results were observed in the majority of the islets in all the other nontumoral tissues examined (samples no. 2, 3, PF14, PF26, PF28, PF30, and PF35; see Materials and Methods). In addition to exocrine/islet borders, we also analyzed exocrine areas more distant from islets and observed no major difference in the expression levels of menin, compared with exocrine cells adjacent to islets (data not shown).
|
, which shows dual labeling of sections of sample no. PF35 with mAb C126 and an antibody against synaptophysin, a marker for neuroendocrine cells, with strong menin reactivity detected in some synaptophysin-positive cells (top panels) and weak reactivity in others (bottom panels). As expected, synaptophysin was detected mainly in the cytoplasm and was excluded from nuclei.
|
, upper panels labeled I, G, and SS, respectively). In contrast, high levels of menin expression were found in PP-positive cells. Interestingly, menin expression among PP-positive cells seemed to be heterogeneous, with higher-expressing cells more frequently located at the islet periphery (see arrows in PP panels E, F, and G). These observations were made for nontumoral samples no. 1, PF26, PF28, and PF35; other samples were not informative, because they did not score positive for all three hormones.
|
Figure 5 shows IF/LSM analysis of a sporadic gastrinoma of the head of the pancreas (sample no. 10A), with menin (A) and propidium iodide (B) signals detected as described above. The field shown in A, B, C, and E spans an area that includes normal exocrine tissue with typical acinar structure (upper left-hand and center portions of field) and the tumor mass consisting of closely packed cells (bottom-center and right-hand portions of the field). An area spanning the boundary between exocrine and tumor cells (white square in C) was analyzed at a higher magnification (D), and menin vs. propidium iodide signals were quantitated across several exocrine and tumor cells (indicated by arrow) as described above. Results (F) showed the nuclei of two exocrine cells (first two peaks at about 10 and 35 µm) with a menin signal of relative intensity of 4,000. In contrast, nuclei of tumor cells (the series of peaks from 45–120 µm) showed a relative intensity of 1,000–2,000. Similar differences in menin expression between exocrine and tumor cells were observed for a sporadic insulinoma (sample no. 14B). Additional IF/LSM analyses of one gastrinoma (sample no. 12A), two insulinomas (samples no. 19B and 21A), and one somatostatinoma (sample no. 13A) revealed that menin was expressed at variable levels; however, the absence of normal exocrine cells in these tissue sections precluded a direct comparison of the relative levels of menin expression.
|
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
) is associated with tumor formation. The observation that menin is differentially expressed in exocrine vs. neuroendocrine pancreatic cells extends previous in situ hybridization data published by Ikeo et al. (12), which suggested differences in expression between exocrine and islet cells at the RNA level.
Ikeo et al. (12) also provided evidence that the MEN1 mRNA is more abundantly expressed in actively proliferating cells. It is thus conceivable that expression of menin in islet cells might be tightly regulated in response to proliferative stimuli. Our observation of high levels of menin expression in a subpopulation of pancreatic peptide-producing cells (Fig. 4) indeed suggests tight regulation of menin expression and a possible functional diversity within this cell population. The recent finding that overexpression of menin in rat insulinoma cells inhibits expression from the insulin promoter, diminishes activation of insulin production in response to glucose, and reduces proliferation (18) indicates that menin might also play a role in determining the secretion status of islet cells.
Our IF/LSM screening of endocrine pancreatic tumors (see Table 1; three gastrinomas, three insulinomas, and one somatostatinoma) revealed menin signals of varying intensities, except in MEN1 -/- gastrinoma no. 15A, in which menin was not detected. Genetic analysis of sporadic insulinomas no. 14B, 19B, and 21A failed to detect biallelic loss of MEN1, which is in line with the presence of a detectable menin signal.
|
Menin is stable when expressed ectopically in tissue culture cells, with a half-life of about 10 h (22). However, our trials with mAb C126 suggest that the protein might be more susceptible to degradation in the context of primary tissues, with its detection depending on prompt freezing of the samples in dry ice, after surgery, and storage at -80 C. Furthermore, frozen sections must be subjected to IF promptly after drying-fixation, because storage of the fixed sections for more than 24 h results in substantial loss of the signal. Quantitative evaluation of menin expression with mAb C126 therefore requires careful, standardized processing of samples, which should include a component of normal exocrine cells as a positive standard within the same section.
In our hands, IF/LSM was a more rapid and robust technique than standard IHC for detecting menin with mAb C126. In addition to propidium iodide counterstaining and dual labelings with islet cell markers employed in the present study, IF/LSM would allow comparison of the levels of menin expression with those of its molecular partners [e.g. JunD (23), Smad3 (24), NF-
B (25), nm23 (26), and RPA-2 (27)]. Furthermore, the quantitative aspect of this approach provides a means of investigating possible correlations between menin expression and the biological/clinical properties of different tumors. Our findings thus provide the first description of the in situ pattern of expression of menin in human pancreas and provide a basis for further studies aimed at defining the pattern of menin expression in other tissues.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
I.C., D.M.D., and T.F. contributed equally to the work.
Abbreviations: C126men, Portion of MEN1 coding for the C-terminal 126 amino acids of menin; GST, glutathione-S-transferase; IF/LSM, immunofluorescence/laser-scanning microscopy; MEN1, multiple endocrine neoplasia type 1; NSE, neuron-specific enolase; PP, pancreatic polypeptide.
Received November 22, 2002.
Accepted April 10, 2003.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  Y M H Jonkers, S M H Claessen, A Perren, S Schmid, P Komminoth, A A Verhofstad, L J Hofland, R R de Krijger, P J Slootweg, F C S Ramaekers, et al. Chromosomal instability predicts metastatic disease in patients with insulinomas Endocr. Relat. Cancer, June 1, 2005; 12(2): 435 - 447. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J W Cardinal, L Bergman, N Hayward, A Sweet, J Warner, L Marks, D Learoyd, T Dwight, B Robinson, M Epstein, et al. A report of a national mutation testing service for the MEN1 gene: clinical presentations and implications for mutation testing J. Med. Genet., January 1, 2005; 42(1): 69 - 74. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
