Molecular Genetic Studies of Sporadic and MEN1-Associated Endocrine Pancreatic Tumors

Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 231

DANIEL LINDBERG

ACTA UNIVERSITATIS UPSALIENSIS ISSN 1651-6206 UPPSALA ISBN 978-91-554-6810-1 2007 urn:nbn:se:uu:diva-7595 !" # $ %& %''( )&*'' + + + , -# + $ ./ 0 1 21 /

3 / %''(/ $ 4 2 + 2 $56)7 5 , 0 / / %&)/ 8" / / 92:6 !(;7!)7""<78;)'7)/

, -,50. + 7+ 1 / ,50 1 ) -$56). = 73 / $ 1 + + / 0 + 1 + ++ + + + ,50/ 0 $56) ))>)& 1 + ,50/ 01 + 1 + / + ))>)& & %" 1 ,50/ );>%) 1 + + $56)7 ,50 + 2 < + / 9 ,50 1 & %" + ?60( =@)) + / $ $56) 1 + 7 @6%@A ); @6):A %( @6%:A )" / = 6 + 1 $56) ,50/ @ 7 < -@<. 7$ 1 + + $56) + ,50/ 0 @< 1 + / 0 + @< + ,50 + /

$56) 3B= ?60( =@)) @< @6%:A )" @6):A %( @6%@A ); 7$ 2 < >

!" " " #$ $ %$" " &'()*+) "

C 3 %''(

9226 )8")78%'8 92:6 !(;7!)7""<78;)'7) * *** 7("!" - *AA //A D E * *** 7("!". To Ann-Kristin and Kjell

List of papers

This thesis is based on the following papers, which will be referred to in the text by their roman numerals.

I Ola Hessman*, Daniel Lindberg*, Britt Skogseid, Tobias Car- ling, Per Hellman, Jonas Rastad, Göran Åkerström & Gunnar Westin (1998). Mutation of the Multiple Endocrine Neoplasia Type 1 gene in nonfamilial, malignant tumors of the endocrine pancreas. Cancer Research, 58:377-379.

II Ola Hessman, Daniel Lindberg, Annika Einarsson, Peter Lillha- ger, Tobias Carling, Lars Grimelius, Barbro Eriksson, Göran Åkerström, Gunnar Westin & Britt Skogseid (1999). Genetic alterations on 3p, 11q13, and 18q in nonfamilial and MEN 1- associated pancreatic endocrine tumors. Genes, Chromosomes & Cancer, 26:258-264.

III Daniel Lindberg, Göran Åkerström & Gunnar Westin (2007). Mutational analyses of WNT7A and HDAC11 as candidate tumour suppressor genes in sporadic malignant pancreatic endocrine tumours. Clinical Endocrinology, 66:110-114.

IV Daniel Lindberg, Göran Åkerström & Gunnar Westin (2007). Evaluation of CDKN2C/p18, CDKN1B/p27, and CDKN2B/p15 mRNA expression and CpG methylation status in sporadic and MEN1-associated pancreatic endocrine tumors. Manuscript.

V Daniel Lindberg, Ola Hessman, Göran Åkerström & Gunnar Westin (2007). CDK4 is overexpressed in pancreatic endocrine tumors regardless of MEN1 mutational status. Manuscript.

* These authors contributed equally. Reprints were made with permission from the publishers.

Contents

Introduction...... 11 Pancreatic endocrine tumors ...... 11 The MEN1 syndrome...... 12 Cell cycle regulation...... 12 The CDK4/cyclin D complex ...... 14 The INK4 group of CKIs...... 15 The CIP/KIP group of CKIs ...... 16 The TGFȕ pathway...... 17 The c-Myc protooncogene...... 18 The MEN1 tumor suppressor gene...... 19 WNT7A – Activator of the JNK- and/or WNT pathways ...... 21 HDAC11 – Histone deacetylase ...... 21 Pancreatic endocrine tumor genetics...... 22 Aims of the study...... 29 Materials and Methods...... 30 Summary of materials and methods ...... 30 Patients and tumors...... 30 DNA and cDNA preparation ...... 30 LOH studies...... 31 Direct sequencing ...... 31 Statistical analysis...... 33 Immunohistochemistry ...... 33 Real-time quantitative PCR ...... 34 Pyrosequencing methylation analysis...... 34 Western blot analysis...... 34 Results and discussion ...... 35 Paper I ...... 35 Paper II ...... 37 Paper III...... 39 Paper IV ...... 40 Paper V...... 42 Concluding remarks...... 44 Acknowledgements...... 46 References...... 47 Abbreviations

28S Mitochondrial ribosomal protein S28 AKT Protein kinase B ARF p14ARF (human), p19ARF (mouse) bcl B-cell lymphoma CDK Cyclin dependent kinase CDKN CDK inhibitor CDT1 Chromatin licensing and DNA replication factor 1 CGH Comparative genome hybridization CIP CDK interacting protein CKI CDK inhibitor c-Myc v-myc avian myelocytomatosis viral oncogene homolog DLK1 į-like protein 1/ preadipocyte factor-1 DNA Deoxyribonucleic acid EGF Epidermal growth factor EGFR Epidermal growth factor receptor Erb Avian erythroblastic leukemia vital oncogene homolog ERK Extra-cellular signal-regulated kinase FANCD2 Fanconi Anemia, complementation group D2 FZD Frizzled GSK3 Glycogen synthase kinase 3 HDAC Histone deacetylase HER-2/neu ErbB2, a EGF receptor IGF Insulin-like growth factor IGFBP IGF-binding protein INK4 Inhibitor of CDK4 JNK c-Jun-N-terminal kinase JunD The JunD oncogene and transcriptional factor KIP Kinase inhibitor proteins LOH Loss of heterozygosity MARCKS Myristoylated alanine-rich protein kinase C substrate MEN1 Multiple endocrine neoplasia type 1 MIRN MicroRNA MSI Microsatellite instability mTOR Mammalian target of Rapamycin nm23 Nonmetastatic protein 23 p107 Retinoblastoma-like 1 p130 Retinoblastoma-like 2 p15 CDKN2B, INK4B p16 CDKN2A, INK4A p18 CDKN2C, INK4C p19 CDKN2D, INK4D p21 CDKN1A, CIP1 p27 CDKN1B, KIP1 p38 Mitogen-activated protein kinase p53 Transformation-related protein 53 p57 CDKN1C, KIP2 PCNA Proliferating cell nuclear antigen PET Pancreatic endocrine tumor PP Pancreatic polypeptide QM/Jif-1 Ribosomal protein L10 Rb Retinoblastoma-1, RB1 RNA Ribonucleic acid RPA2 Replication protein A2 Skp2 S-phase kinase associated protein 2 TERT Telomerase reverse transcriptase TGFȕ Transforming growth factor ȕ TGFȕR TGFȕ receptor Thr Threonine VHL von Hippel-Lindau Introduction

Pancreatic endocrine tumors The islets of Langerhans are embedded in the exocrine pancreas, with coher- ent clusters of endocrine cells producing a variety of hormones. ȕ-cells secrete insulin, Į-cells glucagon, į-cells somatostatin, and PP-cells pancreatic polypeptide (PP). Pancreatic endocrine tumors (PETs) represent 1-2% of all of pancreatic tumors and occur sporadically, or in association with relatively uncommon familial syndromes, namely the Multiple endocrine neoplasia type 1 (MEN1) syndrome, caused by germline mutations of the MEN1 gene, and the von Hippel-Lindau (VHL) syndrome, caused by VHL gene mutations (1, 2). PETs may cause typical symptoms of hormone excess or arise as non- functioning lesions without hormonal symptoms (3). Insulinomas are the most common of the functioning tumors where the excess of insulin/proinsulin causes hypoglycemia. Since 90% of insulinomas are small and benign tumors are patients generally efficiently cured by surgery. Malignant insulinomas are treated with chemotherapy, but surgery may sometimes be done to reduce the tumor burden and alleviate hormonal symptoms. Gastrinomas secrete gastrin causing the Zollinger-Ellison syndrome, with often severe peptic ulcer disease, esophagitis, and diarrhea. Gastrino- mas occur in the pancreas or in the duodenum. The duodenal tumors often appear as very tiny submucosal tumors with early lymph gland metastases, but with liver metastases occurring late. Gastrinomas are considered malignant, but progression rate varies, with more favorable survival for duodenal than pancreatic gastrinomas. Glucagonomas are uncommon and often malignant pancreatic tumors causing a typical skin disease, necrolytic migrating erythema, sometimes cachexia, and symptoms from local growth. The tumors are often large but generally slow-growing, and patients often need repeated surgery for re- moval of metastases during an extended disease course. Non-functioning tumors secrete hormones without clinical symptoms (often PP), or lack hormone secretion. Patients with these tumors have con- siderably better survival prospects than patients with exocrine pancreatic tumors. The endocrine origin is secured by demonstrating hypervascularity at contrast enhanced computerized tomography, octreotide uptake by octreo-

11 scan, raised serum levels of hormones (commonly PP and/or chromogranin A), or by needle biopsy stained with chromogranin A or synaptophysin. Non-functioning tumors are generally malignant but with variable rate of progression related to proliferation (Ki-67 index). Many present with liver metastasis. More uncommon tumors are VIPomas secreting vasoactive intestinal peptide with severe diarrhea, and somatostatinomas with often less obvious or no symptoms (2, 3).

The MEN1 syndrome The MEN1 syndrome is caused by germline mutations of the MEN1 gene on chromosome 11q13. The classical tumor sites in MEN1 are the pancreas, the anterior pituitary, and the parathyroid glands (4). There is also an increased risk for endocrine tumors in the stomach, bronchi, thymus, and duodenum, as well as adrenocortical lesions, lipomas, and facial angiofibromas. Ap- proximately half (46%) of MEN1 patients die from the disease, and the most common causes of death are tumors of the pancreas and thymus (4). Pancreatico-duodenal tumors are encountered in 30-75% of MEN1- patients. Gastrinomas and the Zollinger-Ellison syndrome occur in 20-60% of MEN1 patients, non-functioning tumors affect approximately 50%, whereas other tumor types are less common (2). The MEN1 pancreas typi- cally contains multiple microadenomas, only few of which will grow to tumors causing clinical manifestations. The relative potential for each individual tumor to set metastases is impossible to estimate. Lymph gland metastases are common with duodenal gastrinomas, and approximately 25% develop liver metastases (2).

Cell cycle regulation The cell cycle includes the S-phase for DNA replication and M-phase for mitosis. The G1-phase precedes and the G2-phase follows the S-phase. After passage through the M-phase, cells can stop proliferating or repeat the steps of the cell cycle (5). Cell cycle transition is regulated by members of the cyclin-family (n=29) and cyclin dependent kinases (CDKs) (n=20). In non- proliferating cells, proteins of the Rb family (Rb, p107, and p130) bind to members of the E2F family of transcription factors. Progression from G1 to S is determined by Rb phosphorylation, which occurs in two steps. First, cyclin D1, D2, or D3 forms a complex with CDK4 or CDK6 to phosphorylate Rb at certain residues (Fig. 1). Second, cyclin E forms a complex with CDK2 and phosphorylates other Rb amino acid residues. E2F pro-

12 teins thereby dissociate from Rb, and function on their many target genes, and initializing DNA replication (6). Inactivation of Rb also promotes transcription of cyclin A and cyclin B, which are involved in the G2-M transition. During the S-phase CDK2 molecules bind to cyclin A, but as the cell cycle continues, cyclin A switch part- ner to CDK1. Formation of CDK1/cyclin B-complexes starts in the G2- phase. CDK1 has more than 70 targets that are thought to lead the cell through the M-phase (6).

Figure 1. Cell cycle regulation with suggested roles of menin. See text for references.

Phosphorylation of Rb can be inhibited by two groups of CDK inhibitors (CKIs). The INK4 group, which consists of p15, p16, p18, and p19, suppresses the kinase activity of CDK4 (and CDK6). The CIP/KIP group, which can restrict the CDK2 kinase activity, consists of p21, p27, and p57 (6). Homozygous Rb knock-out mice die in utero, before the pancreatic islets of Langerhans develop, while heterozygous deletion of Rb leads to a small increase in ȕ-cell cell mass and cell number (7), and to islet tumors when

13 combined with a p53 knock-out (8, 9). In mice, loss of CDK4 and cyclin D1 lead to islet cell hypoplasia, and subsequent diabetes (10, 11). Knock-out CDK2 mice survive for more than two years, without other abnormalities than defect meiosis (12).

The CDK4/cyclin D complex Activation of the CDK4/cyclin D complex is initiated by phosphorylation of CDK4 by CDK activating kinase (13). Cyclin D1 is the most studied of the three D-cyclins, and found to be upregulated in a variety of tumors, with stabilizing mutations revealed in endometrial and esophageal tumors (14, 15). Downregulation of cyclin D1 occurs through phosphorylation by glycogen synthase kinase 3ȕ (GSK3ȕ), and subsequent ubiquination. Inhibition of GSK3ȕ leads to increased cell proliferation in a rat insulinoma cell line (16). An alternative spliceform of cyclin D1, cyclin D1b, has lost the recognition signal for nuclear export and GSK3ȕ binding (13). Cyclin D1b retains the ability to activate CDKs and is therefore believed to have increased oncogenic potential. Cyclin D1b cannot be detected in normal tissues, but mRNA and protein expression has been found in tumors (17). Cyclin D1 knock-out mice are small, with retina defects (18), while cyclin D1 overexpression in mouse pancreatic ȕ-cells results in hyperplasia, but not tumor development (19). When in a complex with a D-cyclin, CDK4 is able to function as a kinase and phosphorylate its targets: the Rb family (Rb, p107, p130), Smad2, Smad3, MARCKS, and CDT1 (6). The MARCKS protein is substrate for PLC phosphorylation, which initi- ates a process where MARCKS can be phosphorylated by both CDK2 and CDK4, probably at separate sites. The effect of this phosphorylation is not known (20). MARCKS has been found to be truncated in colorectal tumors with microsatellite instability (MSI) (21). DNA replication is supposed to occur once during each cell cycle. Regu- lated CDT1 degradation, by ubiquination through Skp2, is important to restrict DNA replication. CDK-induced phosphorylation of CDT1 results in Skp2 binding and subsequent degradation (22). CDT1 overexpression has been proposed to contribute to tumor development by causing genomic instability (23). Smad2, Smad3, and Smad4 are regulating proteins of the transforming growth factor ȕ (TGFȕ) pathway. Smad2 and Smad3, but not Smad4, can be phosphorylated by CDK2 and CDK4, and this decreases TGFȕ-induced transcriptional activity (24). Vice versa, TGFȕ stimulation of T-cells induces CDK4 down-regulation, and Smad3 is necessary for this process (25). Re- pression of proliferation by TGFȕ-stimulation resulted in decrease of CDK4 expression (26). The CDK4/cyclin D complex interacts with both groups of CKIs. The INK4 group inhibits the kinase activity, while p21 and p27 from the CIP/KIP

14 group are necessary for the physical interaction between CDK4/6 and the corresponding cyclin D (27). Germline mutations of CDK4, inhibiting p16 interaction, cause familial melanoma in humans (28). The same germline mutation in mice causes hemangiosarcomas and tumors in endocrine organs, predominantly in the endocrine pancreas (29). The Protein kinase B (AKT) pathway has been shown to interact with the CDK4/cyclin D1 function. Decreased AKT function leads to decreased translation of cyclin D1 (30). In BON cells (a human cell line from a sero- tonin producing PET), insulin-like growth factor I (IGF-I) has been found to induce cyclin D1 expression through the AKT pathway (31). A constitu- tively active AKT signaling in mouse ȕ-cells leads to increased proliferation through increased levels of cyclin D1, cyclin D2, CDK4, and p21. Combined with CDK4 knock-out, the proliferative effect of AKT on ȕ-cells is lost (32).

The INK4 group of CKIs p15 transcription is induced by TGFȕ in a variety of cells (33), and it interacts with CDK4. By binding to CDK4, p15 inhibits interaction with cyclin D and hence the kinase activity (34). Additionally, when bound to p15, CDK4 can no longer sequester p21/p27, which become free to inhibit the actions of the cyclin E/CDK2 complex (35). Therefore when expressed, p15 silences the kinase activity of both CDK2 and CDK4. However, high levels of c-Myc can inhibit TGFȕ-induction of p15 (36), and inactivation of c-Myc through destabilization can lead to subsequent p15 reexpression (37). Using an expression microarray, p15 was found to be upregulated by TGFȕ signaling, in T-cells expressing Smad3, compared to Smad3 knock- out cells. Any effect of p15 on proliferation, or actions on CDK4, could not be established (25). The Smad3 transcription factor regulates p15 expression, but CDK2 and CDK4 phosphorylation of Smad3 inhibit this function (24). c- Myc can form a complex with Smad2 and Smad3, and repress p15 expression (38). By repressing c-Myc, TGFȕ signaling also indirectly activates p15 expression (39). p15 knock-out mice develop normally without tumors, as do p15/p18 double knock-out mice (40). p16 inhibits phosphorylation of Rb by binding CDK4/6 and inhibiting the kinase activity (41). p16 has also been shown to inhibit CDK4 phosphorylation of Smad3, supposedly in the same fashion (24). In pancreatic ȕ-cells, p16 expression increases with age, while the other CKI expression levels remain unaltered (42). p15 and p16 show a 85% amino acid similarity, and are therefore believed to have similar functions (41). The cellular localization and phosphorylation status of p16 has been proposed to modify the functional interaction between p16 and CDK4/6 (43). Similarly to germline CDK4 mutations that prevent p16 inhibition, germline inactivating mutations of p16 cause melanomas in humans (41).

15 It has been suggested that the primary function of p18 is inhibition of CDK4. p18 knock-out mice show widespread hyperplasia and or- ganomegaly, that do not occur in p18/CDK4 double knock-out mice. Also p27 (see below) knock-out mice show organ hyperplasia which, however, is not reversed in double p27/CDK4 knock-out mice (44). Haploinsufficiency of tumor suppression has been proposed for p18 and for p27 (see below) (45). p19 has been suggested to be involved in histone deacetylase (HDAC) regulated cell cycle arrest (46). Menin knock-out mice did not display altered p19 mRNA expression in pancreatic islets (47). Often associated with the INK4 CKIs, ARF (p14ARF in humans and p19ARF in mice, not to confuse with p19INK4d) is situated at 9p21 together with p15 and p16. While sharing the second and third exon with p16, different reading frames apply, so no amino acids are shared between the two proteins. The whole ARF, p15, p16 locus is often homozygously deleted in cancer, or inactivated by other means (41). ARF functions as a repressor of MDM2, and loss will result in an increased activity of MDM2 on p53 stability, and thereby deregulated cell cycle arrest. Furthermore, ARF is involved in the regulation of c-Myc, and interacts with hypoxia-inducible factor 1, a target protein of VHL (41).

The CIP/KIP group of CKIs In fibroblasts, functional CDK4/cyclin D complexes contain p21 and p27, and these proteins have been proposed to be essential for CDK4/cyclin D assembly (27). p15 and p16 share binding site with p21 and p27 on CDK4, implying that p15/p16 expression inhibits p21/p27 binding to CDK4 (33). p27 interacts with the CDK2/cyclin E complex, where it inhibits the kinase function and subsequent Rb phosphorylation (48). Mice lacking one p27 allele displayed increased tumor formation when exposed to Ȗ-radiation, but tumor development was even more pronounced when both alleles were deleted. Therefore, p27 is considered haploinsuffi- cient for tumor suppression. Tumors of the heterozygous mice had retained p27 expression (49). Likewise, while mice with a single functional p27 allele are bigger than wild-type mice, homozygous p27 deletion leads to further increase in body weight. The most prominent oncogenic feature of p27 knock-out mice is tumors of the pituitary gland (50). Activation of AKT leads to decreased p27 inhibition of the cyclin E/CDK2 complex. AKT phosphorylates p27 at Thr157, thereby impairing the nuclear import of p27, and hence the possible interaction of p27 with the cyclin E/CDK2 complex (51, 52). In rats, a MEN1-like syndrome called MENX, has been proposed to be caused by an inactivating germline p27 mutation (53). Lately, an indication of another kinase phosphorylating Rb, parallel to CDK2, has been suggested. Double p27/CDK2 knock-out mice

16 were found to display the same phenotype as p27 mice, i.e. increased body weight and tumors of the pituitary (54). Furthermore, p27 has been suggested to be involved in cell migration, in association with cyclin D1 and stathmin (55, 56). Skp2-induced ubiquination, and subsequent degradation of p27, is induced by phosphorylation of p27 at Thr187 by the cyclin E/CDK2 complex (57). Thr187 mutation of p27 makes CDK2 initiated degradation ineffective, but cell cycle progression was not inhibited in knock-in mice, which instead increased in size compared to wild-type. Another pathway, where p27 ubiquination can be triggered by Ras and c-Myc, was suggested (58). Thr198 phosphorylation protects p27 from ubiquination, and lack of Thr198 phosphorylation promotes CDK2-binding (59). p27 can also be stabilized by Rb in an E2F-independent manner, where Rb enhances Skp2-downregulation through Skp2-ubiquination, which causes cell cycle arrest (60). Inhibition of mTOR, which is activated by AKT, reduces Skp2 transcription and thereby facilitates p27 stabilization in breast cancer cell lines (61). Skp2 have been found to be upregulated in many tumors, and is considered a protooncogene (62). Overall, the p21 regulation and function is similar to p27 in many aspects. p21 has been shown to inhibit CDK4 phosphorylation of Smad3 (24). Ex- pression of p21 is induced by the TGFȕ pathway in keratinocytes, but c-Myc inhibits this process (63). However, unlike the p27 gene, knock-out of the p21 gene in mice did not result in altered ȕ-cell function or proliferation (64). In the islets of Langerhans of menin knock-out mice, p21 expression was found to decrease over time (47). Loss of p27 is correlated to MSI in colorectal tumors, while inverse correlation is seen for p21 (65, 66). p57 has been found to be expressed in human islets of Langerhans. p57 is important during embryogenesis, but deletion does not induce ȕ-cell hyperplasia (67).

The TGFȕ pathway TGFȕ factors signal by binding to their receptors, TGFȕR1 and TGFȕR2. TGFBR1 becomes activated and phosphorylates Smad2 and Smad3, which in turn bind to Smad4. The three Smad proteins form a complex that can activate or repress target gene expression (33, 68), and for example p15 transcription is induced by TGFȕ in a variety of cells (33). c-Myc expression is suppressed by TGFȕ, and low levels of c-Myc is a prerequisite for p15 induction (36). Phosphorylation of Smad3 by CDKs inhibits the transcriptional activity of the TGFȕ pathway, including p15 transcription and c-Myc repression (24). In tumors, mutational events occur in several of the genes in the TGFȕ pathway. Truncating mutations in the TGFȕR2 gene is seen in some tumor types with MSI, i.e. colorectal and gastric tumors, while other tumor types

17 with MSI retain TGFȕR2 expression, i.e. tumors of the exocrine pancreas. Ovarian tumors often display mutations in the TGFȕR1 gene, but not in the TGFȕR2 gene (68). Biallelic inactivation of the Smad4 gene has been seen in exocrine pancreatic and colorectal tumors. Missense and nonsense mutations of the Smad2 gene has been found in various tumors types (68), resulting in incapability of the Smad complex to function at target genes. In mouse models, homozygous Smad2 gene or Smad4 gene deletions result in uterine death, whereas heterozygous Smad4 gene deletion results in intestinal pol- yps, and tumors (33). A transgenic mouse model expressing a defect Smad4 gene, under the elastase promoter, develops islet cell hyperplasia (69), whereas overactivation of the TGFȕ pathway leads to pancreatic islet hypoplasia (70). So far, genetic Smad3 inactivation has not been detected in any tumor type, but homozygous Smad3 gene deletion in mice leads to development of colorectal tumors (71).

The c-Myc protooncogene c-Myc was one of the first protooncogenes discovered, and overexpression is common in many different tumors. c-Myc has been proposed to promote proliferation through three distinct pathways: control of cyclin/CDK kinase activity, control of E2F activity, and cell growth (72). c-Myc upregulation results in CDK activation and p27 downregulation in fibroblasts (73). In mouse fibroblasts, knock-out of the three D-cyclins together inhibits the proliferative actions of c-Myc, while any single D-cyclins alone is capable of driving c-Myc induced proliferation (74). c-Myc expression is decreased by Smad3 transcriptional activity, and CDK2 and CDK4 phosphorylation of Smad3 results in increased c-Myc expression (24, 75). c-Myc directly induces both cyclin D2 and CDK4 (76, 77), but activation of cyclin D2 requires an activated AKT-pathway (78). Furthermore, c-Myc can directly activate the E2F2 gene promoter (79). E2F1/E2F2 double knock-out mice develop diabetes due to hypoplasia of the exocrine and endocrine pancreas (80). ARF, p53, and bcl-xL deregulation have been found to interact with c-Myc to increase cell growth, by inhibiting the pro-apoptotic pathways induced by c-Myc (81). In a mouse transgenic model of ȕ-cell tumorigenesis, c-Myc has been found to initiate angiogenesis through interleukin-1ȕ activation (82). Many different genes have been found to be regulated by c-Myc in this model. However, p21 and p16 alterations were not observed, and this was thought to depend on bcl-xL overexpression used in this model (83). TGFȕ represses c-Myc expression through a Smad-responsive element in the promoter region (39). Interestingly, Skp2 is both a transcriptional cofactor of c-Myc, as well as a mediator of its ubiquination. In normal cells, this could be a way to restrict c-Myc function, so that the protein is degraded directly after activating transcription (84). Putting c-Myc under the control

18 of the elastase promoter, and thereby overexpressing c-Myc during pancreas development, leads to PET formation (85).

The MEN1 tumor suppressor gene The 67 kDa protein menin is the product of the MEN1 gene situated at 11q13 (86, 87). Several splice forms can be translated from the nine coding exons, with the largest reading frame translating into 615 amino acids. Two known amino acid residues can be phosphorylated, but with unknown implications for interactions with other proteins or other functions (88). The function of the different splice forms has not been evaluated. Germline mutations have been found in 70% of MEN1-kindreds (89). In one kindred, where no germline mutation was found, tumors still lacked menin expression (90), indicating genetic events not detectable by exon sequencing. In another kindred where sequencing did not reveal mutations, whole-gene deletion was discovered (91). Almost 500 germline and somatic MEN1 gene mutations have been discovered in MEN1-related and sporadic tumors without obvious hotspots for tumor development (92). Somatic mutations have been found in sporadic tumors of the classical MEN1 malignancies (93-99). Menin locates to the cell nucleus, regardless of phosphorylation status (88, 100). Menin interacts with many proteins, such as the activator of S-phase kinase, metastasis suppressor nm23, DNA damage repair protein FANCD2, glial fibrillary acidic protein, vimentin, and the myosin heavy chain (101). Interaction with JunD leads to growth suppression (102, 103). This binding to JunD involves recruitment of mSin3A, HDAC1, and HDAC2 (104, 105). Menin also inhibits JunD phosphorylation by the JNK pathway (106). RPA2 is required for DNA replication, recombination, and repair, and is involved in apoptosis (107). Missense menin mutations of amino acid residue 139, 160, and 176 abolished interaction with both RPA2 and JunD, while menin missense mutations of amino acid residues 164, 183, 436, and 447 did not affect RPA2 and JunD binding (108). Eight other menin missense mutations altered binding to JunD and RPA2 in various degrees. IGF signaling involves interaction of several IGF-binding proteins (IGFBPs), and wild-type menin has been found to repress expression of IGFBP2 (109). Mutation of amino acid 119, but not of 418, of menin led to decreased repression. Menin interacts with three members of the NF-țB-family (110). NF-țB normally regulates genes that are involved in the stress response, and menin represses these actions. The NF-țB pathway is often activated in tumors (111). Interacting with the TGFȕ pathway, menin directly binds Smad3, where menin inactivation results in decreased binding to DNA by Smad3/Smad4 (112). Also, menin inactivation leads to decreased TGFȕR2 expression (113). In cell lines from bone forming cells, menin inhibits Smad1 and Smad5 which in turn leads to decreased TGFȕ signaling (114).

19 Several reports conclude that menin is part of a complex together with MLL1 and MLL2 (115-117), that function by histone methyltransferase activity, to epigenetically regulate gene expression. Different Hox genes have been found to be regulated in this matter in leukemia (118). Regulation of p18 and p27 has also been suggested to occur via histone methylation (47, 117). In heterozygous MEN1 knock-out mice, menin, p18, and p27 mRNA expression were found to be reduced in the islets of Langerhans (47), already after 18 weeks. After 40 weeks, CDK4 expression was increased and p15 and p21 expression was reduced. Also other studies have showed CDK4 upregulation in association with reduced menin expression (113). In a genome-wide screening for menin binding sites, decreased p18 expression was confirmed in menin-ablated islet cells, but alterations in p27 expression could not be detected. Both MLL1 and menin were found binding to both promoters (119). Likewise, p27 was reduced in a majority of insulinomas in a mouse MEN1 knock-out model, but normal p27 expression was found in 23% of the tumors. Hyper- and dysplastic islets from four MEN1 knock-out mice were examined without evidence of altered p27 expression (120). Conditional MEN1 gene knock-out in ȕ-cells led to decreased p18 and p27 expression, and a subsequent increase of CDK2 activity, while no alterations of p21 and p16 was observed. Neither the p18 nor the p27 expression was totally abolished (121). Double knock-out mouse models have displayed that combined menin and p27 loss did not increase islet tumor development, while mice with combined p18 and menin loss develop tumors with an increased rate (122). Double p27/MEN1 knock-out mice instead displayed decreased numbers and size of the pancreatic islets, as compared to MEN1 knock-out mice. E-cadherin and ȕ-catenin expression was altered in a conditional ȕ-cell MEN1 gene knock-out mouse model. Normally, these proteins are expressed at the cell membrane, but cytoplasmatic accumulation was found in the tumors (123). In another ȕ-cell specific tumor model, E-cadherin expression was lost in the transition from normal cells to carcinoma (124). Both missense and nonsense mutations of the MEN1 gene are common in tumors (89). Nonsense mutations cause truncation, leading to impaired interaction with other proteins, and deletion of the nuclear localization sequences (107). Besides causing altered interactions with other molecules, missense mutations often lead to rapid degradation via increased ubiquination (125). Polymorphisms resulting in amino acid exchanges, without impaired protein function, did not initiate increased degradation. Eight out of nine amino acid exchanges caused increased menin degradation, and the one stable mutant (E255E) protein causes familial isolated hyperparathyroidism (125).

20 WNT7A – Activator of the JNK- and/or WNT pathways The WNT7A gene at 3p25 encodes a protein of the WNT family of hormones, which consists of 19 genes. During development WNT7A is involved in the formation of limbs and the female reproductive tract (126, 127). WNT7A has been proposed to function as a tumor suppressor in lung tumors, where downregulation has been observed. This has been proposed to incorrectly activate the WNT and JNK pathways (128, 129). E-cadherin downregulation has been observed in tumors of the lung, and stimulation with WNT7A leads to increased ȕ–catenin activation, and concurrent E- cadherin expression (129). HOXA1 has been found to be concordantly deregulated together with WNT7A in lung tumors (130). WNT7A induction of E-cadherin has also been proposed to occur through activation of PPARȖ, through binding of WNT7A to frizzled (FZD) 9 (131), and FZD9 has been found to be upregulated by c-Myc activation in a ȕ-cell model (83). In PC12, a rat pheochromocytoma cell line, WNT7A has been found to stabilize ȕ- catenin by binding FZD5 (132). Hypermethylation of the CpG island of WNT7A has been found in 71% of tumors from the exocrine pancreas (133). Furthermore, WNT7A has been shown to be frequently downregulated in uterine leiomyomas (134). Shortly, WNT pathway activation is caused by GSK3ȕ inactivation, with resulting nuclear ȕ-catenin accumulation due to decreased ubiquination. ȕ- catenin functions as a transcription factor for several genes, i.e. cyclin D1 and c-Myc (135). GSK3 inactivation leads to increased proliferation in a rat insulinoma cell line (16).

HDAC11 – Histone deacetylase The involvement of different HDACs in the development of cancer has been established. Truncating mutations of HDAC2 has been found in MSI positive colorectal and endometrial tumors (136), and HDAC6 mRNA expression was reduced in breast tumors larger than 20 mm (137). HDAC11 is a nuclear protein that is encoded on chromosome 3p25. HDAC11 does not associate with mSin3A, but has instead been found to associate with HDAC6 (138). Treatment of a leukemia cell line with the HDAC inhibitor valproic acid, leads to an increase of HDAC11 mRNA levels (139), indicating that HDAC11 is regulated differently than other HDACs.

21 Pancreatic endocrine tumor genetics In PETs without metastasis or obvious local invasion, histopathology cannot identify certain signs of malignancy. Immunohistochemical staining with proliferation marker Ki-67 correlates to survival in non-functioning PETs and insulinomas (140, 141). The occurrence of chromosomal gains and/or losses has been extensively studied in PETs, by loss of heterozygosity analysis (LOH), comparative genome hybridization (CGH), and array CGH analyses. These results are summarized in Table 1 and 2. Loss of 11q, where the MEN1 gene is situated, is a common event, with MEN1 gene mutations revealed in sporadic gastrinomas, insulinomas, glucagonomas, VIPomas, and non-functioning tumors, and with an overall incidence of mutations in sporadic PETs varying between 13-38% (94, 96-99). Generally, all MEN1 gene mutations are believed to abolish the function of menin (see above). Dele- tions on chromosome 1, 3p, 6, 11q, 17p, 21q, and 22q (97, 142-151), as well as gains on chromosomes 4, 7, 14q, and Xq have been associated to malignancy in sporadic PETs (144, 149). Rigaud et al found many chromosomal losses in regions where others found chromosomal gains (152). Whether this represents a systematic error, or results from careful examination of many sites on each chromosome is unclear. Our group revealed loss on chromosomes 3, 6, 8, 10, 18, and 21 in more than 30 % of examined MEN1 tumors (153). Heterogeneity of the tumors was also demonstrated with variable pat- terns of allelic loss in different parts of the same tumor and similar results has been presented by others (154). Genome-wide expression microarray studies of PETs revealed >3-fold overexpression of 66 transcripts including IGFBP3, which is deregulated in many tumor types (155). Also detected were 119 transcripts with >3-fold underexpression, including p21, O6-MGMT, and JunD (155). When compar- ing benign and malignant PETs, O6-MGMT was downregulated in malignant tumors and the protooncogene MET was upregulated together with IGFBP1 and IGFBP3 (156). Menin overexpression in BON cells leads to increased expression of JunD, whereas DLK1, PCNA, and QM/Jif-1 are downregulated. These findings were confirmed in tumors with different MEN1 mutation status (157). Decreased menin expression in the BON cell line caused > 2 times mRNA upregulation in 66 genes, including CDK2, and >2 times mRNA downregualtion of 22 genes (158). Expression of microRNAs has also been evaluated in PETs (159), with MIRN103 and MIRN107 expressed in tumors but not in normal pancreas, while MIRN155 on chromosome 21q21 appeared downregulated in tumors. Some patients with the VHL syndrome are affected by PETs, and this suggests that deregulation of the VHL protein can cause altered proliferation and apoptosis in the islets of Langerhans (1). The tumors have virtually in- variably been non-functioning, and have displayed allelic loss of the normal VHL allele (1, 160). The VHL gene is situated at 3p25-26, just telomeric of

22 a region found to be deleted in sporadic PETs, and where deletion has been suggested to correlate to malignant behavior. However, no mutations was found when the gene was sequenced in sporadic PETs (143). Another potential tumor suppressor, PPARȖ, is situated adjacent to the VHL gene at 3p25. Sequencing of the PPARȖ gene in tumors with allelic loss at 3p, did not reveal mutations (161). Many proteins directly involved in the cell cycle control have been investigated in PETs. While CDK6 has been found to be downregulated in a subset of PETs (162), cyclin D1 was overexpressed in 43% and 65% of examined tumors (163, 164). Proteins of the INK4 group bind and inhibit the kinase activity of CDK4/6, while p21 and p27 from the CIP/KIP family have been found to stabilize the CCK4/6-cyclin D complex, and thereby enable the kinase function (see above). Instead, the CIP/KIP family decreases the kinase activity of CDK2 (33). In PETs, several studies have investigated the genes of the INK4/ARF locus. Altogether 92% of gastrinomas and non-functioning PETs displayed homozygous deletion of p16 (42%) and/or promoter hypermethylation (58%) (165). Another study of gastrinomas did not reveal any homozygous deletions, while the promoter region was hypermethylated in 52% of the tumors (166). In insulinomas, only three out of seventeen tumors displayed loss of p16 protein expression (167). In three other studies, the p16 promoter was found to be hypermethylated in three out of sixteen (19%) (168), nineteen out of 48 (40%) (169) and one out of eleven (9%) (170) PETs. Immunohistochemical studies of 24 PETs showed abolished p16 expression in three insulinomas (162). A following study of 41 PETs did not detect p16 promoter hypermethylation in any tumor. One tumor displayed a sporadic missense mutation in one of the alternative p16 transcripts (171). The CpG island of ARF has been found to be hypermethylated in one out of eleven (9%) (170) and seven out of sixteen (44%) (168) PETs. In mRNA expression studies of ARF, p15, p16, and p21 in 26 PETs, was p21 expressed in all tumors, and p16 could be found in all tumors except one non- functioning tumor. ARF was lost in three non-functioning PETs and clearly decreased in one insulinoma. p15 expression was undetectable in five tumors, and low in another four tumors (172). Several publications have investigated the p27 expression in PETs. Since p27 has been found to be downregulated in many tumor types, a similar downregulation was expected also in PETs. However, upregulation of p27 was found, when protein expression was investigated by western blot, in 28 PETs and ȕ-cell cell lines from mouse and rat (173). An immunohistochemical investigation of 46 PETs displayed increased p27 expression in 72%, without any correlation to malignancy (174). Another study of 42 PETs revealed decreased p27 expression in benign tumors but retained expression in malignant tumors (175), and still another study of 76 PETs revealed p27 expression in all tumors. Benign tumors displayed a stronger p27 staining

23 than malignant ones, and cytoplasmic expression was found in 38%, without correlation to malignancy (176). In a study of seven MEN1 patients a decreased portion of cells were found to stain positively for p27 in the tumors, compared to the islets of Langerhans, and p18 staining was minimal in all endocrine tissue examined (117). p19 and p57 have to our knowledge not been studied in PETs. CDKs phosphorylates Rb, a tumor suppressor in many tumor types (177). The Rb gene has been excluded as a tumor suppressor gene in PETs, by the lack of LOH at 13q14 (178). The other two members of the Rb family have not been examined. Exons 5-10 of p53 have been evaluated by single strand conformation polymorphism analysis in eleven PETs, but no mutation was found (179), and neither did immunohistochemical evaluation of p53 in 51 PETs reveal deregulation in any tumor (180). Missense mutation of p53 was found in one out of 41 studied sporadic PETs (171). Bartz et al did not find p53 mutation in nineteen PETs, but reported protein overexpression (181). Telomerase reverse transcriptase (TERT) is a protein responsible for maintaining the length of chromosomal telomeres during mitosis. This protein is normally not expressed in islets of Langerhans (182). Even though menin has been found to be a direct repressor of TERT (183), only three non-functional PETs out of 30 studied displayed increased TERT expression (182). Neither does ȕ-catenin seem to play an important role in the tumorigenesis of PETs, since only two out of 108 studied PETs displayed nuclear staining when examined immunohistochemically (184). Sequencing of exon 3 of ȕ-catenin in these tumors did not reveal any mutations, neither did another study of 33 PETs (185). Immunohistochemical studies show that c-Myc is expressed in all examined PETs (186-188). c-Myc expression may not only lead to proliferation, but also triggers apoptosis. In a mouse model that expressed c-Myc, pancreatic endocrine tumor growth was increased when apoptosis was inhibited by anti-apoptotic bcl-xL (189). Bcl-2 functions similarly to bcl-xL, and bcl-2 overexpression has been found in 39 out of 87 (45%) immunohistochemically examined PETs (188). Divergent results have been found regarding K- ras involvement. Four studies examining altogether 96 PETs discovered K- ras mutation in one non-functioning tumor (171, 190-192). Two studies reported K-ras mutations in seven out of twelve malignant insulinomas and two out of eight benign insulinomas (186, 187), and another study of a single non-functioning tumor revealed K-ras mutation (193). When CpG methylation status of eleven candidate tumor suppressor genes was investigated in 48 PETs, hypermethylation was detected in RASSF1A (75%), p16 (40%), O6-MGMT (40%), RAR-ȕ (25%), and hMHL1 (23%) (169). RASSF1A at 3p21 is hypermethylated and downregulated in many different tumor types, and RASSF1A loss results in decreased degradation of cyclin A and cyclin B (194). O6-MGMT plays a part in the

24 DNA repair, and has also been studied without detectable promoter hypermethylation in eleven PETs (170). Neither did this study revealed any evidence of CpG hypermethylation of the RAR-ȕ promoter, encoding a protein involved in apoptosis (170). The promoters of RASSF1A and O6-MGMT were found to be methylated in 57% and 17%, respectively in yet another study (168). hMLH1 inactivation can lead to MSI, and eleven out of 48 PETs had hypermethylated hMLH1 promoters (169). Five of these tumors showed signs of MSI, comprising 10% of all investigated PETs (195). The occurrence of MSI was correlated to increased survival. Another investigation of 20 PETs, revealed no signs of MSI (142). In an investigation of the promoters of the cyclooxygenase 2, thrombospondin 1, T-type calcium channel, estrogen receptor and MEN1 genes, the CpG island of estrogen receptor was found to be commonly hypermethylated (60% versus 20% in normal pancreatic tissue), while the promoter of the T-type calcium channel gene was hypermethylated in 9% of the tumors (170). Furthermore, both the epidermal growth factor receptor (EGFR) and hepatocyte growth factor receptor were found to be overexpressed in a subset of gastrinomas, correlating with aggressive growth (196). EGFR was not detected in insulinomas or non-functioning PETS (197). HER-2/neu, an EGF receptor, was found to be upregulated in seven out of 29 PETs. c-Jun was also overexpressed in 24%, while another EGF receptor, c-erbB3, was downregulated in 27 out of 29 PETs (188). Two genes involved in the TGFȕ pathway, Smad3 and Smad4, have also been investigated in PETs. Smad3 can interact with menin (112), and thereby inactivate TGFȕ signaling. Four out of twenty studied PETs displayed allelic loss in the area of Smad3, but sequencing did not reveal mutations (198). Smad4 has been investigated in several studies. In a study of 25 PETs inactivating mutations of Smad4 were found in five out nine non- functioning PETs (199). In three other studies, composing altogether 94 tumors, Smad4 inactivation was found in only one tumor (171, 200, 201). mRNA coding for TGFȕR1 and TGFȕR2 have been found in gastrinomas and insulinomas, but not detected in a subset of non-functioning PETs (197) Some of the proteins in the AKT signaling pathway have been investigated in PETs. An activated AKT pathway, through AKT phosphorylation, was revealed in fourteen out of twenty examined PETs, and in the same study was the p38 pathway found to be upregulation in ten, and the ERK pathway downregulated in fifteen out of the twenty PETs (163). Activation of AKT can occur through the actions of phosphatidylinositol, and this protein is regulated by PTEN and PIK3CA (202). The PTEN gene is a tumor suppressor gene, and mouse ȕ-cell mass is increased when the PTEN gene is inactivated (203). When the PTEN gene was investigated, one out of 33 PETs harbored a missense mutation, and PTEN protein was expressed in 23 out of 24 PETs (204). We have sequenced two mutation hot-spots (exon 9

25 and 20) of the protooncogene PIK3CA in fifteen PETs, without detecting mutations (unpublished results). PET tumorigenesis may also include ȕ-cell specific signaling pathways. Ghrelin and its receptor have both been found to be expressed in PETs (205), and ghrelin has been suggested to function through activation of the AKT pathway in ȕ-cells (206). IGF-II has been found to repress apoptosis in a mouse tumor cell line (207) and is expressed in PETs (208), with development of islet hyperplasia when overexpressed in transgenic mice (209).

26 Table 1. Chromosomal losses in sporadic PETs as observed by LOH, CGH, and CGH microarray analyses. Only results with LOH in >20% of the tumors are shown. Chromosome Proportion of PETs with allelic loss 1p 44% (210), 25% (211), 75% (97), 27% (144), 40% (152), 30% (145), 31% (146), 23% (212) 1q 88% (147), 75% (97), 20% (144), 35% (152), 44% (145), 26% (146) 2p 20% (144), 44% (152), 36% (146) 2q 44% (210), 22% (144), 31% (152), 31% (146) 3p 59% (148), 30% (149), 35% (152), 25% (142), 26% (212), 62% (213), 33% (143) 3q 27% (148), 30% (149), 22% (144), 25% (152), 33% (212) 4q 38% (152) 5q 38% (152) 6p 27% (149), 31% (152), 26% (146), 20% (212) 6q 39% (149), 24% (144), 68% (152), 62% (150), 20% (212) 7q 31% (152) 8p 25% (152) 8q 23% (149), 20% (144), 31% (152) 9p 30% (152), 33% (171) 10p 38% (152) 10q 25% (149), 25% (152), 53% (204), 26% (146) 11p 24% (214), 30% (149), 29% (144), 52% (152), 30% (146), 35% (212) 11q 32% (214), 50% (97), 36% (149), 33% (144), 66% (152), 43% (146), 45% (142), 55% (215), 29% (212), 28% (213), 47% (98), 79% (96) 12p 25% (152) 12q 25% (152), 26% (146) 13p 38% (152) 13q 38% (152) 14p 25% (152) 14q 25% (152) 15p 44% (152), 20% (198) 15q 44% (152), 27% (146) 16p 33% (211), 38% (152), 33% (146), 33% (212) 16q 44% (152), 29% (146) 17p 38% (152), 35% (142), 38% (171) 18p 27% (146) 18q 30% (146), 35% (142), 33% (171), 22% (152) 19q 38% (152) 20p 25% (152) 20q 56% (152) 21p 50% (152) 21q 50% (152) 22p 38% (152) 22q 93% (151), 38% (152), 44% (146), 29% (212) Xp 24% (214), 32% (146), 38% (216), 82% (217) Xq 24% (214), 25% (149), 20% (144), 27% (146), 31% (216), 100% (217) Y 23% (149), 36% (144)

27 Table 2. Chromosomal gains in sporadic PETs as observed by CGH and CGH microarray analyses. Only results with allelic gain in >20% of the tumors are shown.

Chromosome Proportion of PETs with allelic gain 4p 50% (211), 22% (144) 4q 50% (211), 22% (144) 5p 40% (214), 50% (211), 23% (149), 24% (144), 30% (146) 5q 44% (210), 40% (214), 50% (211), 32% (149), 31% (144) 7p 50% (211), 27% (149), 33% (144), 30% (146) 7q 50% (211), 43% (149), 40% (144), 29% (146) 9p 33% (211), 47% (146) 9q 44% (214), 42% (211), 27% (149), 31% (144) 12p 56% (210), 58% (211), 26% (146) 12q 56% (210), 66% (211), 25% (149), 24% (144) 13p 33% (211) 13q 33% (211) 14p 56% (214), 50% (211) 14q 44% (210), 56% (214), 50% (211), 32% (149), 24% (144), 26% (146) 15q 44% (210) 17p 44% (210), 40% (214), 42% (211), 27% (149) 17q 40% (214), 75% (211), 41% (149), 31% (144) 18p 32% (214), 25% (211) 18q 44% (210), 32% (214), 67% (211) 19p 44% (210), 60% (214), 33% (211) 19q 44% (210), 60% (214), 33% (211), 26% (146) 20p 56% (210), 25% (211), 32% (146) 20q 56% (210), 48% (214), 25% (211), 27% (149), 22% (144) 21q 26% (146) 22p 25% (211) 22q 25% (211) Xp 25% (149) Xq 23% (149)

28 Aims of the study

The specific aims of the study were:

x to investigate whether the MEN1 gene was mutated in sporadic PETs. (Paper I)

x to investigate whether the Smad4 gene was mutated in sporadic and MEN1-related PETs, and to determine the frequency of LOH on chromosome 3p, 11q, and 18q in relation to malignancy. (Paper II)

x to investigate whether the WNT7A and HDAC11 constitute potential tumor suppressor genes in sporadic PETs. (Paper III)

x to investigate whether the mRNA expression of p15, p18, and p27 in sporadic PETs is related to MEN1 mutational status. To investigate whether the p15 gene was mutated or hypermethylated in the tumors. (Paper IV)

x to investigate CDK4 mRNA and protein expression in relation to MEN1 mutational status. To look for mutations in coding parts of the CDK4 gene in sporadic PETs. To investigate the protein expression of c-Myc and CDK4. (Paper V)

29 Materials and Methods

Summary of materials and methods

Patients and tumors

(Paper I-V) All patients, tissue specimen, and clinical data were recruited and collected in the clinical routine at the Department of Surgery, Uppsala University Hospital, with informed consent and approval of ethical committee. Twenty-two PETs from twenty patients with sporadic tumors, and sixteen PETs from eleven patients with the MEN1 syndrome, were collected for paper I-V. In paper V, fifteen additional sporadic PETs were investigated for sequencing of the CDK4 gene. All patients have been carefully examined for features of the MEN1 syndrome, and specimens for signs of malignancy. The sporadic benign tumors consisted of insulinomas, and all sporadic malignant tumors had either metastases or evident growth into adjacent organs. Two tumors in paper II were classified as equivocal, since these criteria for malignancy was not entirely met. According to the WHO-classification these tumors would be classified as low grade malignant (218). Two of the MEN1-related patients displayed signs of malignancy. The pancreatic specimens of MEN1-patients often harbored multiple adenomas and microtumors (2), and which of the tumors that was the origin of metastases was not obvious. The PETs from MEN1 patients, without signs of malignancy at the time of the operation, were considered benign, but these patients may develop malignant tumors during follow-up, and all MEN1 tumors have malignant potential, including insulinomas.

DNA and cDNA preparation

(Paper I-V) Patients with PETs were operated on, and tumors were snap-frozen in liquid nitrogen in connection to the operation, and blood was collected. Tissue was stored in -70 °C and blood in -20 °C. The specimens were microdissected in

30 a cryostat to avoid gross contamination. DNA was extracted using proteinase K/sodium dodecyl sulfate and phenol/chloroform extraction. Total mRNA was prepared from frozen tissue using TRIzol Reagent (Gibco BRL, Life Technologies) according to the manufacturer’s instructions. mRNA was treated with RQ1 DNase I (Promega Corp) and proteinase K. Reverse transcriptase PCR was performed with random nucleotide primers using First-strand cDNA synthesis kit (GE Healthcare) according to the manufacturer’s instructions.

LOH studies

(Paper I & II) DNA from tumors and tumor-free tissue (normal pancreas, spleen or blood) from the same patient were examined. Different polymorphic markers in the areas of interest were used, and the primers were marked with the radionu- cleotide 32P for analyses of chromosome 3p and 11q, and end-labeled with fluorescent dye for 18q studies. PCR reaction was carried out with 20ng of genomic DNA, 0.2 U Taq Polymerase, 1xPCR buffer, 1.5 mM MgCl2, 0.1 mM of each dNTP, and 2 pmoles of each primer. After an initial denaturation at 95 °C for 4 minutes, 27 cycles of annealing at 55 °C for 30s, extension at 72 °C and denaturation at 95 °C for 30s with a final extension at 72 °C for 7 minutes was performed. 32P marked PCR products were mixed with formamide and denaturated for 5 minutes at 80°C, electrophoresed on a 4.5% polyacrylamid gel, and quantified on a PhosphorImager (Molecular Dynamics). PCR with fluorescent marked primers were conducted with the same conditions as with 32P marked primers. The fluorescent PCR fragments were resolved on an ABI 310 instrument, with GeneScan 350 TAMRA as size marker, and quantified with GeneScan software (Applied Biosystems Inc.). PCR products from tumor DNA and constitutional DNA were compared. In paper I allelic signal decrease of at least 70% was considered as LOH, while in paper II decreased allelic signal of 50% or more was considered as LOH.

Direct sequencing

(Paper I-V) Previously published primers for paper I and paper II are no longer available at corresponding internet sites. Therefore, these are included in Table 3. The internet-based software program Primer3 was used for designing primers for paper V (http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi). All coding exons, as well as splice sites, of all genes of interest were investigated.

31 Table 3. Primers used for MEN1 and Smad4 gene sequencing in Paper I and II Sequence MEN1 exon 2 Fw TTA GCG GAC CCT GGG AGG AG Rev GGC GGC GAT GAT AGA CAG GTC MEN1 exon 2 Fw CTG GCG GCC TCA CCT ACT TTC Rev CAT GGA TAA GAT TCC CAC CTA CTG G MEN1 exon 3 Fw GCA CAG AGG ACC CTC TTT CAT TAC Rev TGG GTG GCT TGG GCT ACT ACA G MEN1 exon 4 Fw GGG CCA TCA TGA GAC ATA ATG Rev CTG CCC CAT TGG CTC AG MEN1 exon 5 and 6 Fw CCT GTT CCG TGG CTC ATA ACT C Rev CTC AGC CAC TGT TAG GGT CTC C MEN1 exon 7 Fw GGC TGC CTC CCT GAG GAT C Rev CTG GAC GAG GGT GGT TGG MEN1 exon 8 Fw GTG AGA CCC CTT CAG ACC CTA C Rev TGG GAG GCT GGA CAC AGG MEN1 exon 9 Fw GGG TGA GTA AGA GAC TGA TCT GTG C Rev TGT AGT GCC CAG ACC TCT GTG MEN1 exon 10a Fw CGG CAA CCT TGC TCT CAC C Rev GCC CTT CAT CTT CTC ACT CTG G MEN1 exon 10b Fw TGC CAG CAC CCG CAG CAT C Rev CCC ACA AGC GGT CCG AAG TCC

Smad4 Exon 2 Fw CGT TAG CTG TTG TTT TTC ACT G Rev AGA GTA TGT GAA GAG ATG GAG Smad4 Exon 3 Fw TGT ATG ACA TGG CCA AGT TAG Rev CAA TAC TCG GTT TTA GCA GTC Smad4 Exon 4 Fw CTG AAT TGA AAT GGT TCA TGA AC Rev GCC CCT AAC CTC AAA ATC TAC Smad4 Exon 5 Fw TTT TGC TGG TAA AGT AGT ATG C Rev CTA TGA AAG ATA GTA CAG TTA C Smad4 Exon 6 and 7 Fw CAT CTT TAT AGT TGT GCA TTA TC Rev TAA TGA AAC AAA ATC ACA GGA TG Smad4 Exon 8 Fw TGA AAG TTT TAG CAT TAG ACA AC Rev TGT ACT CAT CTG AGA AGT GAC Smad4 Exon 9 Fw TGT TTT GGG TGC ATT ACA TTT C Rev CAA TTT TTT AAA GTA ACT ATC TGA Smad4 Exon 10 Fw TAT TAA GCA TGC TAT ACA ATC TG Rev CTT CCA CCC AGA TTT CAA TTC Smad4 Exon 11 Fw AGG CAT TGG TTT TTA ATG TAT G Rev CTG CTC AAA GAA ACT AAT CAA C Smad4 Exon 12 Fw CCA AAA GTG TGC AGC TTG TTG Rev CAG TTT CTG TCT GCT AGG AG

PCR reamplification was conducted if the primary PCR reaction resulted in low yield, and dimethylsulfoxide addition to the PCR reaction was used if necessary. One strand was examined for mutations, but in case of suspected mutations both strands were sequenced. Verification of detected mutations

32 was conducted through repetition of the PCR reaction and subsequent re- evaluation of the sequence. For paper I, ABI Prism Dye Terminator Cycle Sequencing Ready Reac- tion kit was used for the sequence reaction, and analyses were done on an ABI373 (Applied Biosystems). In paper II, an ABI Prism BigDye Termina- tor Cycle Sequencing Ready Reaction was used for the sequence reaction and analyzes were conducted on an ABI 310 (Applied Biosystems). For paper III-V, amplified PCR fragments were subjected to sequencing on a 3130XL Genetic Analyzer using a BigDye Terminator v3.1 Cycle Sequenc- ing Ready Reaction Kit (Applied Biosystems).

Statistical analysis

(Paper II) Statistical analyses were conducted with Statistica for Windows 5.1 (Statsoft, Inc.) in paper II. Fisher’s exact test was used for statistical analyses and p<0.05 was considered significant.

Immunohistochemistry

(Paper III & V)

Six ȝm tissue cryosections were acetone fixed, and blocked with 0.3% H2O2 in PBS to reduce endogenous peroxidase activity, followed by blocking with avidin-biotin blocking kit and normal goat or horse serum. In paper III, WNT7A goat polyclonal antibody sc-26360 (Santa Cruz Biotechnology Inc) and HDAC11 rabbit polyclonal antibody ab18973 (Abcam) were diluted 1:40 and incubated for 90 minutes. In paper V, CDK4 rabbit polyclonal antibody sc-601(Santa Cruz Biotechnology Inc) and c-Myc rabbit polyclonal antibody sc-789 (Santa Cruz Biotechnology Inc) were diluted 1:160 and 1:50, respectively, and incubated for 90 minutes. Polyclonal rabbit synaptophysin antibody was diluted 1:100 and incubated for 90 minutes. Bioti- nylated goat anti-rabbit antibody was used as a secondary antibody against rabbit polyclonal antibodies, whereas biotinylated horse anti-goat antibody was used as a secondary antibody against the goat WNT7A antibody. Sec- ondary antibodies were applied for 30 minutes, and then developed with 3- amino-9-ethylcarbazole followed by counterstaining with Mayer’s hema- toxylin. Negative controls, where slides with tumors/tumor-free pancreas were treated with all steps, except with primary antibodies, did not reveal falsely positive staining with any of the used antibodies. Additional control of the CDK4-staining, where the antibody had been treated with an excess of im- munizing peptide, did not reveal falsely positive staining either.

33 Real-time quantitative PCR

(Paper IV & V) Primers were constructed using Primer3 (http://frodo.wi.mit.edu/cgi- bin/primer3/primer3_www.cgi). PCR was carried out using iQ SYBR Green Supermix on MyiQ Single-color Real-time PCR detection System (Bio-Rad Laboratories). Each sample was analyzed in triplicate. For cDNA studies, standard curves for the genes of interest were constructed by dilution of a PCR fragment. The values for the gene of interest were divided with those of 28S rRNA that was used as comperative control. The results for gene copy number of the p15 and CDK4 genes were related to copy numbers of the FCİRI gene at 11q13 and the NPC1 gene at 18q11.

Pyrosequencing methylation analysis

(Paper IV) DNA was bisulphite treated. Primers and sequencing primer were constructed by Biotage AB according to their standards. PCR was conducted for 45 cycles with 10 pmoles of primer, 200 ȝM of dNTP, and Platinum Taq Polymerase (Gibco BRL, Life Technologies Inc) in a 50 ȝl reaction. Anneal- ing temperature varied between 51-58 °C, and MgCl2 concentration varied between 1.5-3.0 mM. Pyrosequencing methylation analysis was conducted according to the manufacturer’s specifications, on the PSQ HS 96 system (Biotage AB). The complete experiment was performed twice, and similar results were obtained. Two CpG sites with more than 5% methylation were considered positive.

Western blot analysis

(Paper V) Protein extracts were prepared from 20 consecutive frozen tissue sections in Cytobuster Protein Extract Reagent (Novagen Inc.) supplemented with Complete protease inhibitor cocktail (Roche Diagnostics). Polyclonal rabbit antibodies against CDK4 (sc-601) and ȕ–tubulin (sc-9104) were used (Santa Cruz Biotechnology Inc). Anti-rabbit antibody from donkey (GE Healthcare) was used as secondary antibody, and bands were visualized using the enhanced chemiluminescence system (GE Healthcare).

34 Results and discussion

Paper I

Mutation of the Multiple Endocrine Neoplasia Type 1 gene in nonfamilial, malignant tumors of the endocrine pancreas The MEN1 gene has been localized to 11q13, and was cloned in 1997 (86, 87). Since PETs are one of the characteristic manifestations of the MEN1 syndrome, we set out to study eleven sporadic malignant tumors. The nine coding MEN1 gene exons were sequenced in two glucagonomas and nine non-functioning tumors, and LOH was done on the same specimens except for one non-functioning tumor. Both glucagonomas and five out of eight non-functioning tumors displayed LOH. Two nonsense mutations were found: E397Stop (GAG/TAG) and R420Stop (CGA/TGA), formerly named E392Stop and R415Stop, and one missense mutation: L89R (CTC/CGC). Since the publication of our results, the amino acid numbering has changed because alternative transcripts have been shown to include five additional amino acids between former amino acid 148 and 149. In the study we also detected two previously known polymorphisms. All tumors with mutation displayed LOH. The primer design did not allow for analysis of the intron branch point sequence for exons 3, 4, 5, 7, 8, 9, and 10. We also could not evaluate the last two amino acids and the consensus 5´ splice site of exon 2, since the presence of an alternative transcript, which included five extra amino acids in the end of exon 2, was unknown at the time. Consequently, inactivating mutations in these parts of the MEN1 gene can not be excluded. Today, almost 500 unique somatic and germline MEN1 gene mutations are known. These are evenly distributed, with no mutational hot-spots, on the protein coding region and its exon/intron borders, most commonly resulting in a truncated protein (89, 92). The three mutations described here are presumed to restrict menin's interaction with other proteins, in accordance with menin functioning as a tumor suppressor. Both the E397Stop and R420Stop mutations truncate menin, leaving out the nuclear localization signals and interaction sites for many of menin’s known partners (107). The L89R mutation exchanges a hydrophobic leucine residue for a basic argin-

35 ine, which may alter the tertiary structure of menin in a part that is important for binding to Smad3. In addition, most missense mutations lead to greatly decreased menin expression in affected cells, possibly as a consequence of increased degradation by the ubiquitin-proteasome pathway (107). Sequestering of menin in the cytoplasm, at the S and early G2 phase, has been suggested to occur through binding to the intermediate filament network (219). When localized to the cell nucleus, menin interacts with several partners, including JunD, HDAC, NF-țB, Smad3, and double-stranded DNA, to regulate transcription (107). Regulation of proteins, such as CDK4, has been shown, as well as transcriptional activation of the p18 and p27 genes (92, 107, 113, 220). Smad3 is involved in the TGFȕ pathway. Menin-ablated cells, exhibiting decreased TGFȕ2 receptor expression, have been shown to lose sensitivity to TGFȕ1-mediated inhibition of proliferation (113). Menin inhibits the binding of the Smad3/Smad4 complex to DNA, subsequently inhibiting transcription of their target genes (112). NF-țB target genes are often up-regulated in tumors (107, 111). Menin interacts with different members of the NF-țB family, which normally are responsible for stress induced transcription. Menin has been proposed to function as a repressor, or recruiter of repressors, of these proteins, e.g. by recruiting HDACs (110). JunD is another target for menin repression (104) where the proposed mechanism of repression includes recruitment of HDACs through association with mSin3a (105). Menin also interacts with CHES1, a member of a transcriptional repressor complex that also include mSin3a, HDAC1, and HDAC2 (221). Two JunD transcripts exist, where one is capable of binding menin. In its normal state, bound to menin, JunD is as a growth repressor, but when detached from menin, JunD acts as a growth promoter (107). In hematopoiesis, menin has been proposed to function in a complex with MLL, a protein family that is deregulated in leukemia. The MLL/menin complex regulates several HOX-genes through histone H3 Lysine 4 methylation (118). MLL and menin have also been shown to regulate expression of the CKIs p18 and p27, by increasing methylation of histone H3 Lysine 4 (47, 117). Mutation of the MEN1 gene may thus interfere with several signaling pathways during tumor development.

36 Paper II

Genetic alterations on 3p, 11q13, and 18q in nonfamilial and MEN1- associated pancreatic endocrine tumors The incidence of allelic loss on chromosome 3p, 11q, and 18q was studied in 38 PETs from 31 patients, where 20 had sporadic tumors and eleven the MEN1 syndrome. Sporadic tumors from nine out of twelve and ten out of twelve patients displayed allelic loss of 3p and 11q13, respectively. All six benign sporadic insulinomas displayed retained alleles at these loci. Allele loss at 18q was present in one benign insulinoma, and in tumors from three patients with sporadic malignant disease. One out two patients with a sporadic disease with equivocal signs of malignancy displayed loss at 11q and 18q, whereas the other patient showed retained alleles in all examined loci. Ten out of eleven patients with the MEN1 syndrome showed LOH at 11q13. One MEN1-patient with metastases displayed LOH on 3p in two out of three examined tumors, whereas 18q alleles were retained in all tumors from both patients with a malignant MEN1 syndrome. We observed an increased LOH rate at 3p in malignant tumors compared to benign tumors (p<0,01), which also has been reported by other groups (143, 148, 149). We suggest that the high incidence of allelic loss of this area is associated with the existence of a tumor suppressor gene involved in PET formation. One tumor (S31, 5773) had allele loss on only one of the microsatellites at 3p, namely D3S1038, whereas another tumor (M5, 721.2) had retained alleles at the D3S1317 locus, but displayed allelic loss centromeric of this microsatellite. This restricts the area of interest to 12.1 Mb between markers D3S1317 and D3S1293. Chung et al presented data from a subset of PETs, two of which have properties that makes it possible to further restrict this area of interest. One tumor displayed LOH telomeric of D3S1110, and another tumor had LOH between RAF1 and D3S1339 (143). This means that there are two areas of interest at 3p: one between RAF1 and D3S1293, and one between D3S1317 and RAF1, encompassing approximately 9.8 Mb and 2.3 Mb respectively. In the latter area, the PPARȖ gene has been excluded from functioning as a potential tumor suppressor gene in PETs (161). VHL, situated telomeric of these regions, has also been excluded (143). Speel et al (149) described loss at 11q (36%) and 11p (30%) in 44 PETs, and Stumpf et al (214) reported loss at 11q in eight and 11p in six out of 25 investigated PETs. There are several chromosomal regions of interest in addition to the MEN1 locus on chromosome 11q13. A region just telomeric of MEN1 has been suggested to harbor a tumor suppressor gene (215). In our study, microsatellites covering 24 Mb of chromosome 11q were used. Three tumors displayed partial deletions in this region. One of the tumors from patient M3 (382.22) retained the INT-2 marker, while the PYGM microsatel-

37 lite marker displayed allelic loss. Likewise, tumor 3537 from patient S22 showed loss only at the MEN1 locus, and tumor 5773 from patient S31 retained both alleles at the D11S906 marker, telomeric to the MEN1 gene. All tumors with LOH at 11q13 in our study showed deletions that encompassed the MEN1 gene. Only two out of fourteen malignant sporadic tumors displayed retained 11q alleles, while all six benign sporadic tumors had retained alleles. There was a significant difference between allelic loss at 11q13 in benign versus malignant sporadic tumors (p<0.01). Low incidence of MEN1 gene involvement in benign PETs has also been reported by Cupisti et al, who found no evidence of inactivating mutations in 24 benign insulinomas (222). We found LOH at 18q to be rarer than LOH at 3p and 11q. Only five sporadic tumors (one benign and four malignant) and four MEN1-related tumors (from three patients without signs of malignancy) had 18q loss. All tumors from the two MEN1 patients with metastases retained 18q. Allelic loss at 18q is not common in PETs. Less than 20 % of the studied PETs displayed LOH at 18q in three CGH studies (149, 210, 214). A BAC array CGH showed deletion of 1.65 Mb at 18q23 in 30% of 27 insulinomas (146), while two LOH studies displayed LOH of 18q in 33% and 35%, respectively (142, 171). The TGFȕ pathway is involved in several cellular functions, i.e. induction of apoptosis, inhibition of proliferation and induction of mesenchymal transition which may promote invasiveness (33). Activation of the TGFȕ pathway results in the formation of a Smad protein complex that induces, or represses, transcription of hundreds of genes. Up-regulated production of factors from the TGFȕ family and/or mutations in the downstream effectors are common in many tumor types (68). Menin has been found to interact with Smad3, but sequencing of the Smad3 gene did not reveal any mutations in PETs (198). Both the Smad2 and Smad4 genes are tumor suppressor genes situated on 18q21 (33). Exocrine pancreatic tumors, as well as other tumor types, exhibit Smad4 inactivation in a large proportion of the tumors (68). We performed sequencing of all eleven coding exons of the Smad4 gene in all tumors with identified LOH at 18q, and no mutations were found. These results were presented in parallel to Bartsch et al (199) that published a study on 25 PETs where five of nine non-functioning tumors harbored inactivating mutation in the Smad4 gene (two missense mutations, one nonsense mutation, one intragenic deletion, and one homozygous deletion). There are also several other publications on Smad4 in PETs. Moore et al found LOH at 18q in thirteen out of 40 (32%) PETs, with normal Smad4 expression and unaltered Smad4 gene coding sequence (171). Scarpa et al (201) studied PETs from 20 patients and found quenched Smad4 expression in only one tumor. Perren et al (200) examined 34 PETs using immunohistochemistry, LOH, and fluorescence in situ hybridization analysis without finding any Smad4 aberrations. There are several possible reasons why the

38 different studies report incongruous results regarding the role of Smad4 in PET tumorigenesis: 1) Bartsch et al accidentally included exocrine pancreatic tumors in their study 2) The true incidence of Smad4 inactivation in PETs is 6/150 (4%), and of altogether 150 examined tumors five of the six tumors with inactivation were included in the study by Bartsch et al. 3) studies may have been performed on endocrine tumors of different origin. The poorly differentiated neuroendocrine carcinoma is an uncommon form of PET with more aggressive features than the classical PETs. In our set of tumors we found no evidence indicating a tumor suppressor function of Smad4. Nevertheless, the observed LOH at 18q indicates that a hitherto unidentified tumor suppressor gene is situated in this region. Several potential tumor suppressor genes, identified in other tumor types, are situated at 18q21. The deleted region encompasses 25.4 Mb, reaching from D18S460 to D18S467, which involve tumor suppressors (the Smad2 and Smad4 genes), oncogenes (the bcl2 and Smad7 genes) as well as more than a hundred genes with unknown tumorigenic potential.

Paper III

Mutational analyses of WNT7A and HDAC11 as candidate tumour suppressor genes in sporadic malignant pancreatic endocrine tumours As described above, LOH at 3p25 correlates to a malignant phenotype in PETs. This region covers approximately 9.3 million nucleotides between D3S1317 and D3S1293, including 63 genes according to the present consensus coding sequences. Only a few of these genes are known tumor suppressors, e.g. the VHL and PPARȖ genes that lies on the telomeric border of this area. Neither of these appears to function as tumor suppressor genes in PETs according to previous studies (143, 161). We have examined the CpG island of the VHL gene by methylation-specific PCR in ten sporadic tumors from eight patients, but no aberrant methylation could be detected (results not shown). Here, we investigate two additional potential tumor suppressor genes in this chromosomal region: the WNT7A and HDAC11 genes. WNT7A has been shown to function as tumor suppressor gene in lung tumors, where WNT7A regulates the E-cadherin expression (128, 129, 131). The HDAC11 gene is situated telomeric of the WNT7A gene. HDAC11 functions in a complex with HDAC6 (138), and loss of HDAC6 mRNA expression has been correlated to breast tumor size. To investigate whether WNT7A and HDAC11 constitutes potential tumor suppressor genes in PETs, we examined ten tumors from eight patients with LOH at 3p25. WNT7A and HDAC11 were expressed in both exocrine pancreas and in islets of Langerhans when tumor-free pancreas was examined immunohistochemically. The CpG island of WNT7A did not display hyper-

39 methylation in any tumor. Sequencing of both genes, including all coding exons and RNA splice sites, did not reveal any mutations in the tumors. To further exclude WNT7A and HDAC11 from being downregulated in tumors, all tumors were examined immunohistochemically. All tumors exhibited positive staining for both WNT7A and HDAC11. Although haploinsufficiency for WNT7A or HDAC11 may in principle contribute to PET tumorigenesis, our findings indicate that neither WNT7A nor HDAC11 function as a classical tumor suppressor in PETs. Consequently, a putative tumor suppressor gene has yet to be identified on chromosome 3p25. Other candidate genes in this chromosomal region include the ZNF659 gene, which has been found to carry missense mutations in two colorectal tumors (223), and microRNA 563, a microRNA with unknown properties. We have sequenced the gene for microRNA 563 in 32 PETs, and found no mutations (unpublished results). ZNF659 is a zinc finger protein, and has likely capacity to bind DNA and/or RNA. This gene consists of eight coding exons, and both mutations identified so far are located in the last exon and does interfere with the zinc finger consensus sites.

Paper IV

Evaluation of CDKN2C/p18, CDKN1B/p27, and CDKN2B/p15 mRNA expression and CpG methylation status in sporadic and MEN1- associated pancreatic endocrine tumors Menin has been found to regulate p18 and p27 expression. p18 inhibits the phosphorylation of Rb by the cyclin D-CDK4/6 complex (44), and p27 by the cyclin E/CDK2 complex (48). Pancreatic islet hyperplasia has been described both in p18 and p27 knock-out mice, while double knock-out mice developed a variety of endocrine tumors, mimicking the MEN1-syndrome (45, 122). In mouse fibroblasts, Milne et al showed that menin, in complex with MLL, binds the promoter regions of p18 and p27. They also reported that reduced expression of either MLL or of menin caused decreased p18 and p27 expression (117). Karnik et al also described association of menin to the p18 and p27 promoters (47). They harvested islets of Langerhans, from mice with a single functional MEN1 gene, at different time points, and found that islet hyperplasia developed after seven months. The mRNA levels of p18 and p27 were decreased at 18 weeks, and totally extinguished at 40 weeks. At that time, p19 mRNA expression was unaltered while p15 and p21 were decreased, and CDK4 greater than before. Increased lysine Histone 3 Lysine 4 methylation at the p18 and p27 promoters was detected in the presence of menin.

40 Expression levels of p15, p18, and p27 have not been compared previously in PETs, in relation to MEN1 gene mutational status. We therefore determined relative mRNA expression levels of p15, p18, and p27 in eighteen PETs from sixteen patients with known MEN1 mutational status, including LOH at 11q13. None of the eighteen tumors showed hypermethylation of the p18 promoter, and p18 mRNA expression was detected in all but three tumors. One of these was a sporadic malignant insulinoma with LOH at 11q13 but no detected MEN1 gene mutation, another was a metastasis from a sporadic non-functioning tumor with 11q13 loss with no detected MEN1 mutation in the primary tumor, and the third was an insulinoma with 11q13 loss from a MEN1-patient without signs of malignancy at the time of surgery. No gen- eral correlation between MEN1 gene/11q13-status and p18 mRNA expression was found, but menin deregulation might still be involved in the extinguished expression of p18 in these three tumors. This is the first study re- porting on p18 expression in human PETs. p27 mRNA expression was detected, and even higher, in all eighteen tumors, compared to normal pancreatic tissue. No correlation between MEN1 gene/11q13-status and p27 mRNA expression was found, and hypermethylation of the p27 gene CpG island was not present. Several previous studies have investigated p27 expression in PETs, albeit not correlating the findings to the MEN1 gene status. Metastases commonly show p27 expression, while more indolent tumors have been reported to lack p27 expression (175). Overexpression of p27 in PETs has also been reported, which is in tune with our findings (173). Chang et al however reported decreased p27 mRNA expression in malignant PETs as compared to apparently benign non- functioning PETs (176), and Milne et al reported somewhat decreased p27 expression in PETs from MEN1-patients as compared to non-neoplastic islets of Langerhans from the same seven individuals (117). In another study of MEN1 knock-out mice, the p27 expression was normal in 23 % of the examined insulinomas (120). These results implicate that the p27 downregulation is not obligatory in MEN1-related tumorigenesis, and other pathways consequently need to be involved. Pellegata et al suggested that germline p27 mutations, rather than MEN1 gene mutations can predispose to the development of a MEN1-like phenotype (53). Eight of the eighteen tumors did not express p15 mRNA, and this did not generally relate to the MEN1 gene/11q13-status. In two cases of non- functioning sporadic tumors, p15 mRNA expression was detected in the primary tumors, but not in the corresponding metastases. Two out of three tumor-free pancreatic tissues did not express p15. Thus, p15 is silent in some PETs and clearly expressed in others, regardless of MEN1 gene mutational status. Biallelic deletion has been observed for p16 in some PETs (165), but the p15 gene was not deleted in any of our tumors as demonstrated by quantitative PCR. We found p15 CpG island hypermethylation in two tumors

41 with silent expression, one benign and one malignant sporadic insulinoma. Non-detectable p15 mRNA expression has been reported in PETs (172), and treatment of the human endocrine cell line QGP-1 with the demethylating agent decitabine activated p15 mRNA expression (172). Whether this was due to CpG- or lysine H3K9 demethylation, or other epigenetic mechanisms, is not known. Sequencing of the p15 gene in the eighteen tumors did not reveal inactivating mutations. Furthermore, the three tumors lacking p18 mRNA expression also showed undetectable p15 expression. Whether p18 and p15 functionally cooperate in the repression of cyclin/CDK-complexes is not known. Sotillo et al suggested that the only tumorigenic function of p18 is inhibition of CDK4, based on results from a mouse knock-out model (224). When combining p18 or p27 knock-out mice with MEN1 knock-outs, it was found that deletion of the p18 gene, but not of the p27 gene, led to increased tumor formation. The p27 gene deletion was even found to have a protective function on ȕ-cell proliferation when introduced into mice lacking menin (122). In conclusion, p15, p18, and p27 were expressed in many or all of the investigated tumors, and the undetected expression of p18 in three tumors could be related to menin inactivation. Undetectable p15 expression could be caused by menin inactivation in some tumors and/or promoter hypermethylation in others.

Paper V

CDK4 is overexpressed in pancreatic endocrine tumors regardless of MEN1 mutational status As previously outlined, mouse pancreatic islets with deleted MEN1 gene have increased CDK4 mRNA levels (47), and shutting down MEN1 mRNA expression in a rat duodenal cell line resulted in increased mRNA levels of CDK4 and cyclin D1, but unaltered cyclin D3 levels (113). CDK4 binds to a D-cyclin, and subsequently phosphorylates Rb. CDK6 has been suggested to function similarly to CDK4 but immunohistochemistry revealed low CDK6 staining in PETs (162). When we determined the CDK4 mRNA levels in eighteen PETs, we found similar or increased CDK4 mRNA levels as compared to the pancreatic control tissue. There was however no difference in CDK4 mRNA levels when PETs with different MEN1/11q13-status were compared. CDK4 is not expressed in normal pancreatic islets as reported by western blot analysis and immunohistochemistry (225). Immunohistochemistry revealed CDK4 expression in all eighteen tumors, and not in normal islets as expected. CDK4 expression was confirmed by

42 Western blot analysis of a subset of the tumors. CDK4 gene amplification was excluded by using quantitative PCR. A germline point mutation in the CDK4 gene, where an amino acid in po- sition 24 is altered, impedes binding to p16 and induces PETs in mice (29). The corresponding human germline mutation is associated with familial melanoma (28). An increased incidence of pancreatic tumors in families with this mutation has not been observed (226). However, MEN1 mutations have been detected in sporadic melanomas (227). We sequenced all coding CDK4 exons in fifteen PETs, and found no mutations. The identified increased CDK4 expression in PETs is possibly a result of altered regulation by menin, or more likely by other factors. For example, c- Myc is overexpressed in all examined PETs (186-188), and regulates CDK4 expression directly by binding to the promoter region (76). Therefore, we investigated c-Myc expression in our PETs and found overexpression in all tumors. Immunohistochemically stained successive tumor sections showed that c-Myc and CDK4 expression was clearly overlapping. In summary, we find that CDK4 is overexpressed in all the eighteen PETs, possibly caused by c-Myc overexpression. CDK4 overexpression could result in augmented kinase activity, increased Rb phosphorylation, and subsequent initiation of DNA replication. Acetylation of a lysine residue in the CDK docking site of Rb in response to DNA damage has been shown to hinder Rb phosphorylation (228). The acetylation status of this residue is not known in PETs. Whether increased CDK4 expression, and in some PETs also increased cyclin D1 expression, is sufficient to induce cell proliferation is not known.

43 Concluding remarks

The MEN1 syndrome, well-known since long time, is caused by inactivating mutations of the MEN1 gene on chromosome 11q13 (2). In paper I, we established the fact that mutations of the MEN1 gene is involved also in the tumorigenesis of sporadic PETs. Large effort has been invested in research of the MEN1 gene, since the cloning in 1997 (86), and several functions have been ascribed to the menin protein (92). Truncating mutations of menin are most common in tumor development, but no mutational hot-spots can be found (89). Some menin mutations interfere with binding to different regulatory factors that have been found to promote proliferation on their own. For example, mutation of amino acid residue 255 potentially disrupts interaction with Smad3, nm23, FANCD2, and HDAC1, as well as disrupting a leucine zipper-like motif with unknown properties (107). In addition, many missense mutations cause increased and rapid protein degradation by the ubiquitin- proteasome pathway (125). Whether the missense mutation observed in the PETs (paper I) lead to rapid degradation or interfere with the binding to specific factors remain to be established. Inactivation of menin leads to inhibition of the TGFȕ growth-inhibitory response, through diminished Smad3 interaction (112), and decreased TGFȕ signaling may in turn lead to increased c-Myc expression (68), which is commonly observed in PETs (paper V). Furthermore CDK4, which is overexpressed in PETs (paper V), can phosphorylate Smad3 and thereby inhibit TGFȕ signaling (24). Altogether, aberrant inactivation of TGFȕ signaling may play a major role in PET development. We observed LOH in 18q21 in PETs, and the Smad4 gene was excluded as potential tumor suppressor gene (paper II). Now it is clear that the Smad2 gene locates within the observed deleted region at 18q21, and therefore should be regarded as a candidate tumor suppressor gene. That menin regulates p18 and p27 expression is undisputable (47, 117), but the results presented here indicate that the mechanisms behind tumorigenesis are likely not identical in human PETs and MEN1 mice knock-out models. In these models, the expression of p18 and p27 are commonly totally extinguished. This appears not to be the case in human PETs. These genes are regulated at several levels and under the influence of a range of regulatory proteins. For example, p27 is subject to phosphorylation by AKT1 (229), and the effects of this modification is largely unknown. AKT activation can induce ȕ-cell hyperplasia, a process counteracted by CDK4

44 deletion (32). p18 deletion and CDK4 overexpression have also been suggested to activate AKT signaling (230). AKT1 has been found to be activated in a subset of PETs (163). Furthermore, cyclin D1 has been shown to be able to induce cell migration in breast cancer, but this process requires p27 (56). Increased quantities of cyclin D/CDK-complexes, as well as c- Myc, has also been found to sequester p21 and p27 (231), thereby functionally removing p27 from inhibiting the cyclin E/CDK2 complex to phosphorylate Rb (232). c-Myc seems to be overexpressed in all examined PETs (186-188). Cyclin D1 has been studied in PETs, and was in two studies reported to be upregulated in 43% and 65% of PETs (163, 164), and in paper V we show that CDK4 is overexpressed in all investigated tumors. It is possible that p27 is beneficial for tumor growth in PETs, but that the inhibition of Rb- phosphorylation by p27 is restricted by sequestration, phosphorylation and/or by other mechanisms. Indeed, MEN1/p27 double knock-out mice had decreased number of pancreatic islets compared to MEN1 single knock-out mice (122). The mystery deepens as it turns out that CDK2 knock-out mice develop almost normally (12), and that CDK2/p27 double knock-out mice still developed tumors, although the obvious p27 substrate was missing (54). CDK4 needs coupling to a corresponding D-cyclin, for activation of its kinase activity. Since we found overexpressed CDK4 in all studied PETs, maybe overexpression of other D-cyclins also occur. While there may be a D-cyclin surplus due to decreased CDK6 sequestration, caused by the CDK6 downregulation in PETs (162), it is uncertain if this, on its own, is enough to supply all CDK4s with cyclins. In rat islets of Langerhans, both cyclin D1 and cyclin D2 are essential for development, whereas cyclin D3 mRNA was nearly undetectable and does not appear to play an important role (10). AKT1 has been shown to induce both cyclin D1 and cyclin D2 in mice (32), and cyclin D2 is overexpressed in insulinomas from heterozygous menin knock-out mice (233). Furthermore, c-Myc has been found to directly induce cyclin D2 (77). So far, there are no studies on the cyclin D2 and cyclin D3 expression in human PETs. The function of CDK4 seems important for tumor development, a notion which has been picked up by the pharmaceutical industry. Several CDK inhibitors, both CDK4-specific (234, 235) and agents with broader CDK- specificity (236, 237), have been constructed and are able to arrest the cell cycle in tumor cells. In summary, we have investigated menin, Smad4, WNT7A, HDAC11, p15, p18, p27, c-Myc, and CDK4 in PETs, and found that menin, p15, p18, c-Myc, and CDK4 show different signs of deregulation. Furthermore, we have found allelic loss on chromosome 3p25, 11q13, and 18q21 in different PETs, and related allelic loss at 3p25 and 11q13 to malignancy.

45 Acknowledgements

This thesis work was carried out at the Department of Surgical Sciences, Uppsala University Hospital. I would like to express my sincere gratitude to colleagues, friends and family members who have contributed to the research and encouraged me in this work.

My supervisors Gunnar Westin for endless enthusiasm, inspiration and energy, and Göran Åkerström for guidance, availability, and support.

All co-authors, especially Ola Hessman for continuous stimulating collaboration.

Colleagues and friends in the research group Carin Backlin, Peyman Björklund, Birgitta Bondeson, Pamela Buchwald, Tobias Carling, Katarina Edfeldt, Per Hellman, Anders Knutson, Peter Lillhager, Wei Liu, Jonas Rastad, Johanna Sandgren, Ulrika Segersten, Jessica Sved- lund,and Eva Szabo.

My parents Ann-Kristin and Kjell and my sister Johanna for all the love and support during my last 33 years.

Carolina, Cesar, Svea, and Jussi for being the most important persons in my life.

46 References

1. Chetty R, Kennedy M, Ezzat S, Asa SL 2004 Pancreatic endocrine pathology in von Hippel-Lindau disease: an expanding spectrum of lesions. Endocr Pathol 15:141-8 2. Öberg K, Eriksson B 2005 Endocrine tumours of the pancreas. Best Pract Res Clin Gastroenterol 19:753-81 3. Åkerström G, Hellman P, Hessman O, Osmak L 2004 Surgical treatment of endocrine pancreatic tumours. Neuroendocrinology 80 Suppl 1:62- 6 4. Skogseid B 2003 Multiple endocrine neoplasia type 1. Br J Surg 90:383-5 5. Park MT, Lee SJ 2003 Cell cycle and cancer. J Biochem Mol Biol 36:60-5 6. Malumbres M, Barbacid M 2005 Mammalian cyclin-dependent kinases. Trends Biochem Sci 30:630-41 7. Vasavada RC, Cozar-Castellano I, Sipula D, Stewart AF 2007 Tissue- Specific Deletion of the Retinoblastoma Protein in the Pancreatic Beta-Cell Has Limited Effects on Beta-Cell Replication, Mass, and Function. Diabe- tes 56:57-64 8. Harvey M, Vogel H, Lee EY, Bradley A, Donehower LA 1995 Mice deficient in both p53 and Rb develop tumors primarily of endocrine origin. Cancer Res 55:1146-51 9. Williams BO, Remington L, Albert DM, Mukai S, Bronson RT, Jacks T 1994 Cooperative tumorigenic effects of germline mutations in Rb and p53. Nat Genet 7:480-4 10. Kushner JA, Ciemerych MA, Sicinska E, Wartschow LM, Teta M, Long SY, Sicinski P, White MF 2005 Cyclins D2 and D1 are essential for postnatal pancreatic beta-cell growth. Mol Cell Biol 25:3752-62 11. Rane SG, Dubus P, Mettus RV, Galbreath EJ, Boden G, Reddy EP, Barbacid M 1999 Loss of Cdk4 expression causes insulin-deficient diabetes and Cdk4 activation results in beta-islet cell hyperplasia. Nat Genet 22:44-52 12. Ortega S, Prieto I, Odajima J, Martin A, Dubus P, Sotillo R, Barbero JL, Malumbres M, Barbacid M 2003 Cyclin-dependent kinase 2 is essential for meiosis but not for mitotic cell division in mice. Nat Genet 35:25-31 13. Gladden AB, Diehl JA 2005 Location, location, location: the role of cyclin D1 nuclear localization in cancer. J Cell Biochem 96:906-13 14. Benzeno S, Lu F, Guo M, Barbash O, Zhang F, Herman JG, Klein PS, Rustgi A, Diehl JA 2006 Identification of mutations that disrupt phosphorylation-dependent nuclear export of cyclin D1. Oncogene 25:6291-303 15. Moreno-Bueno G, Rodriguez-Perales S, Sanchez-Estevez C, Hardisson D, Sarrio D, Prat J, Cigudosa JC, Matias-Guiu X, Palacios J 2003 Cy-

47 clin D1 gene (CCND1) mutations in endometrial cancer. Oncogene 22:6115-8 16. Mussmann R, Geese M, Harder F, Kegel S, Andag U, Lomow A, Burk U, Onichtchouk D, Dohrmann C, Austen M 2007 Inhibition of glycogen synthase kinase (GSK) 3 promotes replication and survival of pancreatic beta cells. J Biol Chem. Epub ahead of print. 17. Knudsen KE 2006 The cyclin D1b splice variant: an old oncogene learns new tricks. Cell Div 1:15 18. Fantl V, Stamp G, Andrews A, Rosewell I, Dickson C 1995 Mice lacking cyclin D1 are small and show defects in eye and mammary gland development. Genes Dev 9:2364-72 19. Zhang X, Gaspard JP, Mizukami Y, Li J, Graeme-Cook F, Chung DC 2005 Overexpression of cyclin D1 in pancreatic beta-cells in vivo results in islet hyperplasia without hypoglycemia. Diabetes 54:712-9 20. Manenti S, Yamauchi E, Sorokine O, Knibiehler M, Van Dorsselaer A, Taniguchi H, Ducommun B, Darbon JM 1999 Phosphorylation of the myristoylated protein kinase C substrate MARCKS by the cyclin E-cyclin- dependent kinase 2 complex in vitro. Biochem J 340 (Pt 3):775-82 21. Yamaguchi T, Iijima T, Mori T, Takahashi K, Matsumoto H, Miya- moto H, Hishima T, Miyaki M 2006 Accumulation profile of frameshift mutations during development and progression of colorectal cancer from patients with hereditary nonpolyposis colorectal cancer. Dis Colon Rectum 49:399-406 22. Liu E, Li X, Yan F, Zhao Q, Wu X 2004 Cyclin-dependent kinases phosphorylate human Cdt1 and induce its degradation. J Biol Chem 279:17283- 8 23. Seo J, Chung YS, Sharma GG, Moon E, Burack WR, Pandita TK, Choi K 2005 Cdt1 transgenic mice develop lymphoblastic lymphoma in the absence of p53. Oncogene 24:8176-86 24. Matsuura I, Denissova NG, Wang G, He D, Long J, Liu F 2004 Cyclin- dependent kinases regulate the antiproliferative function of Smads. Nature 430:226-31 25. Wolfraim LA, Walz TM, James Z, Fernandez T, Letterio JJ 2004 p21Cip1 and p27Kip1 act in synergy to alter the sensitivity of naive T cells to TGF-beta-mediated G1 arrest through modulation of IL-2 responsive- ness. J Immunol 173:3093-102 26. Ewen ME, Sluss HK, Whitehouse LL, Livingston DM 1993 TGF beta inhibition of Cdk4 synthesis is linked to cell cycle arrest. Cell 74:1009-20 27. Cheng M, Olivier P, Diehl JA, Fero M, Roussel MF, Roberts JM, Sherr CJ 1999 The p21(Cip1) and p27(Kip1) CDK 'inhibitors' are essential acti- vators of cyclin D-dependent kinases in murine fibroblasts. Embo J 18:1571-83 28. Zuo L, Weger J, Yang Q, Goldstein AM, Tucker MA, Walker GJ, Hayward N, Dracopoli NC 1996 Germline mutations in the p16INK4a binding domain of CDK4 in familial melanoma. Nat Genet 12:97-9

48 29. Sotillo R, Dubus P, Martin J, de la Cueva E, Ortega S, Malumbres M, Barbacid M 2001 Wide spectrum of tumors in knock-in mice carrying a Cdk4 protein insensitive to INK4 inhibitors. Embo J 20:6637-47 30. Muise-Helmericks RC, Grimes HL, Bellacosa A, Malstrom SE, Tsichlis PN, Rosen N 1998 Cyclin D expression is controlled post-transcriptionally via a phosphatidylinositol 3-kinase/Akt-dependent pathway. J Biol Chem 273:29864-72 31. von Wichert G, Haeussler U, Greten FR, Kliche S, Dralle H, Bohm BO, Adler G, Seufferlein T 2005 Regulation of cyclin D1 expression by autocrine IGF-I in human BON neuroendocrine tumour cells. Oncogene 24:1284-9 32. Fatrai S, Elghazi L, Balcazar N, Cras-Meneur C, Krits I, Kiyokawa H, Bernal-Mizrachi E 2006 Akt induces beta-cell proliferation by regulating cyclin D1, cyclin D2, and p21 levels and cyclin-dependent kinase-4 activity. Diabetes 55:318-25 33. Massague J, Blain SW, Lo RS 2000 TGFbeta signaling in growth control, cancer, and heritable disorders. Cell 103:295-309 34. Sandhu C, Garbe J, Bhattacharya N, Daksis J, Pan CH, Yaswen P, Koh J, Slingerland JM, Stampfer MR 1997 Transforming growth factor beta stabilizes p15INK4B protein, increases p15INK4B-cdk4 complexes, and inhibits cyclin D1-cdk4 association in human mammary epithelial cells. Mol Cell Biol 17:2458-67 35. Reynisdottir I, Polyak K, Iavarone A, Massague J 1995 Kip/Cip and Ink4 Cdk inhibitors cooperate to induce cell cycle arrest in response to TGF-beta. Genes Dev 9:1831-45 36. Warner BJ, Blain SW, Seoane J, Massague J 1999 Myc downregulation by transforming growth factor beta required for activation of the p15(Ink4b) G(1) arrest pathway. Mol Cell Biol 19:5913-22 37. Buschbeck M, Uribesalgo I, Ledl A, Gutierrez A, Minucci S, Muller S, Di Croce L 2006 PML4 induces differentiation by Myc destabilization. Oncogene. Epub ahead of print. 38. Feng XH, Liang YY, Liang M, Zhai W, Lin X 2002 Direct interaction of c-Myc with Smad2 and Smad3 to inhibit TGF-beta-mediated induction of the CDK inhibitor p15(Ink4B). Mol Cell 9:133-43 39. Yagi K, Furuhashi M, Aoki H, Goto D, Kuwano H, Sugamura K, Mi- yazono K, Kato M 2002 c-myc is a downstream target of the Smad pathway. J Biol Chem 277:854-61 40. Latres E, Malumbres M, Sotillo R, Martin J, Ortega S, Martin- Caballero J, Flores JM, Cordon-Cardo C, Barbacid M 2000 Limited overlapping roles of P15(INK4b) and P18(INK4c) cell cycle inhibitors in proliferation and tumorigenesis. Embo J 19:3496-506 41. Kim WY, Sharpless NE 2006 The regulation of INK4/ARF in cancer and aging. Cell 127:265-75 42. Krishnamurthy J, Ramsey MR, Ligon KL, Torrice C, Koh A, Bonner- Weir S, Sharpless NE 2006 p16INK4a induces an age-dependent decline in islet regenerative potential. Nature 443:453-7

49 43. Nilsson K, Landberg G 2006 Subcellular localization, modification and protein complex formation of the cdk-inhibitor p16 in Rb-functional and Rb-inactivated tumor cells. Int J Cancer 118:1120-5 44. Pei XH, Bai F, Tsutsui T, Kiyokawa H, Xiong Y 2004 Genetic evidence for functional dependency of p18Ink4c on Cdk4. Mol Cell Biol 24:6653-64 45. Franklin DS, Godfrey VL, O'Brien DA, Deng C, Xiong Y 2000 Func- tional collaboration between different cyclin-dependent kinase inhibitors suppresses tumor growth with distinct tissue specificity. Mol Cell Biol 20:6147-58 46. Yokota T, Matsuzaki Y, Miyazawa K, Zindy F, Roussel MF, Sakai T 2004 Histone deacetylase inhibitors activate INK4d gene through Sp1 site in its promoter. Oncogene 23:5340-9 47. Karnik SK, Hughes CM, Gu X, Rozenblatt-Rosen O, McLean GW, Xiong Y, Meyerson M, Kim SK 2005 Menin regulates pancreatic islet growth by promoting histone methylation and expression of genes encoding p27Kip1 and p18INK4c. Proc Natl Acad Sci U S A 102:14659-64 48. Ortega S, Malumbres M, Barbacid M 2002 Cyclin D-dependent kinases, INK4 inhibitors and cancer. Biochim Biophys Acta 1602:73-87 49. Fero ML, Randel E, Gurley KE, Roberts JM, Kemp CJ 1998 The murine gene p27Kip1 is haplo-insufficient for tumour suppression. Nature 396:177-80 50. Fero ML, Rivkin M, Tasch M, Porter P, Carow CE, Firpo E, Polyak K, Tsai LH, Broudy V, Perlmutter RM, Kaushansky K, Roberts JM 1996 A syndrome of multiorgan hyperplasia with features of gigantism, tumorigenesis, and female sterility in p27(Kip1)-deficient mice. Cell 85:733-44 51. Liang J, Zubovitz J, Petrocelli T, Kotchetkov R, Connor MK, Han K, Lee JH, Ciarallo S, Catzavelos C, Beniston R, Franssen E, Slingerland JM 2002 PKB/Akt phosphorylates p27, impairs nuclear import of p27 and opposes p27-mediated G1 arrest. Nat Med 8:1153-60 52. Shin I, Yakes FM, Rojo F, Shin NY, Bakin AV, Baselga J, Arteaga CL 2002 PKB/Akt mediates cell-cycle progression by phosphorylation of p27(Kip1) at threonine 157 and modulation of its cellular localization. Nat Med 8:1145-52 53. Pellegata NS, Quintanilla-Martinez L, Siggelkow H, Samson E, Bink K, Hofler H, Fend F, Graw J, Atkinson MJ 2006 Germ-line mutations in p27Kip1 cause a multiple endocrine neoplasia syndrome in rats and humans. Proc Natl Acad Sci U S A 103:15558-63 54. Martin A, Odajima J, Hunt SL, Dubus P, Ortega S, Malumbres M, Barbacid M 2005 Cdk2 is dispensable for cell cycle inhibition and tumor suppression mediated by p27(Kip1) and p21(Cip1). Cancer Cell 7:591-8 55. Baldassarre G, Belletti B, Nicoloso MS, Schiappacassi M, Vecchione A, Spessotto P, Morrione A, Canzonieri V, Colombatti A 2005 p27(Kip1)- stathmin interaction influences sarcoma cell migration and invasion. Cancer Cell 7:51-63 56. Li Z, Jiao X, Wang C, Ju X, Lu Y, Yuan L, Lisanti MP, Katiyar S, Pestell RG 2006 Cyclin D1 induction of cellular migration requires p27(KIP1). Cancer Res 66:9986-94

50 57. Montagnoli A, Fiore F, Eytan E, Carrano AC, Draetta GF, Hershko A, Pagano M 1999 Ubiquitination of p27 is regulated by Cdk-dependent phosphorylation and trimeric complex formation. Genes Dev 13:1181-9 58. Malek NP, Sundberg H, McGrew S, Nakayama K, Kyriakides TR, Roberts JM 2001 A mouse knock-in model exposes sequential proteolytic pathways that regulate p27Kip1 in G1 and S phase. Nature 413:323-7 59. Kossatz U, Vervoorts J, Nickeleit I, Sundberg HA, Arthur JS, Manns MP, Malek NP 2006 C-terminal phosphorylation controls the stability and function of p27kip1. Embo J 25:5159-70 60. Binne UK, Classon MK, Dick FA, Wei W, Rape M, Kaelin WG, Jr., Naar AM, Dyson NJ 2007 Retinoblastoma protein and anaphase- promoting complex physically interact and functionally cooperate during cell-cycle exit. Nat Cell Biol 9:225-232 61. Shapira M, Kakiashvili E, Rosenberg T, Hershko DD 2006 The mTOR inhibitor rapamycin down-regulates the expression of the ubiquitin ligase subunit Skp2 in breast cancer cells. Breast Cancer Res 8:R46 62. Gstaiger M, Jordan R, Lim M, Catzavelos C, Mestan J, Slingerland J, Krek W 2001 Skp2 is oncogenic and overexpressed in human cancers. Proc Natl Acad Sci U S A 98:5043-8 63. Claassen GF, Hann SR 2000 A role for transcriptional repression of p21CIP1 by c-Myc in overcoming transforming growth factor beta - induced cell-cycle arrest. Proc Natl Acad Sci U S A 97:9498-503 64. Cozar-Castellano I, Haught M, Stewart AF 2006 The cell cycle inhibitory protein p21cip is not essential for maintaining beta-cell cycle arrest or beta-cell function in vivo. Diabetes 55:3271-8 65. Ogino S, Kawasaki T, Kirkner GJ, Ogawa A, Dorfman I, Loda M, Fuchs CS 2006 Down-regulation of p21 (CDKN1A/CIP1) is inversely associated with microsatellite instability and CpG island methylator phenotype (CIMP) in colorectal cancer. J Pathol 210:147-54 66. Ogino S, Kawasaki T, Kirkner GJ, Yamaji T, Loda M, Fuchs CS 2007 Loss of nuclear p27 (CDKN1B/KIP1) in colorectal cancer is correlated with microsatellite instability and CIMP. Mod Pathol 20:15-22 67. Cozar-Castellano I, Fiaschi-Taesch N, Bigatel TA, Takane KK, Garcia- Ocana A, Vasavada R, Stewart AF 2006 Molecular control of cell cycle progression in the pancreatic beta-cell. Endocr Rev 27:356-70 68. Pardali K, Moustakas A 2007 Actions of TGF-beta as tumor suppressor and pro-metastatic factor in human cancer. Biochim Biophys Acta 1775:21- 62 69. Simeone DM, Zhang L, Treutelaar MK, Zhang L, Graziano K, Logsdon CD, Burant CF 2006 Islet hypertrophy following pancreatic disruption of Smad4 signaling. Am J Physiol Endocrinol Metab 291:E1305-16 70. Moritani M, Yamasaki S, Kagami M, Suzuki T, Yamaoka T, Sano T, Hata J, Itakura M 2005 Hypoplasia of endocrine and exocrine pancreas in homozygous transgenic TGF-beta1. Mol Cell Endocrinol 229:175-84 71. Zhu Y, Richardson JA, Parada LF, Graff JM 1998 Smad3 mutant mice develop metastatic colorectal cancer. Cell 94:703-14

51 72. Beier R, Burgin A, Kiermaier A, Fero M, Karsunky H, Saffrich R, Moroy T, Ansorge W, Roberts J, Eilers M 2000 Induction of cyclin E- cdk2 kinase activity, E2F-dependent transcription and cell growth by Myc are genetically separable events. Embo J 19:5813-23 73. Steiner P, Philipp A, Lukas J, Godden-Kent D, Pagano M, Mittnacht S, Bartek J, Eilers M 1995 Identification of a Myc-dependent step during the formation of active G1 cyclin-cdk complexes. Embo J 14:4814-26 74. Yu Q, Ciemerych MA, Sicinski P 2005 Ras and Myc can drive oncogenic cell proliferation through individual D-cyclins. Oncogene 24:7114-9 75. Coffey RJ, Jr., Bascom CC, Sipes NJ, Graves-Deal R, Weissman BE, Moses HL 1988 Selective inhibition of growth-related gene expression in murine keratinocytes by transforming growth factor beta. Mol Cell Biol 8:3088-93 76. Hermeking H, Rago C, Schuhmacher M, Li Q, Barrett JF, Obaya AJ, O'Connell BC, Mateyak MK, Tam W, Kohlhuber F, Dang CV, Sedivy JM, Eick D, Vogelstein B, Kinzler KW 2000 Identification of CDK4 as a target of c-MYC. Proc Natl Acad Sci U S A 97:2229-34 77. Bouchard C, Thieke K, Maier A, Saffrich R, Hanley-Hyde J, Ansorge W, Reed S, Sicinski P, Bartek J, Eilers M 1999 Direct induction of cyclin D2 by Myc contributes to cell cycle progression and sequestration of p27. Embo J 18:5321-33 78. Bouchard C, Marquardt J, Bras A, Medema RH, Eilers M 2004 Myc- induced proliferation and transformation require Akt-mediated phosphorylation of FoxO proteins. Embo J 23:2830-40 79. Sears R, Ohtani K, Nevins JR 1997 Identification of positively and negatively acting elements regulating expression of the E2F2 gene in response to cell growth signals. Mol Cell Biol 17:5227-35 80. Iglesias A, Murga M, Laresgoiti U, Skoudy A, Bernales I, Fullaondo A, Moreno B, Lloreta J, Field SJ, Real FX, Zubiaga AM 2004 Diabetes and exocrine pancreatic insufficiency in E2F1/E2F2 double-mutant mice. J Clin Invest 113:1398-407 81. Finch A, Prescott J, Shchors K, Hunt A, Soucek L, Dansen TB, Swigart LB, Evan GI 2006 Bcl-xL gain of function and p19 ARF loss of function cooperate oncogenically with Myc in vivo by distinct mechanisms. Cancer Cell 10:113-20 82. Shchors K, Shchors E, Rostker F, Lawlor ER, Brown-Swigart L, Evan GI 2006 The Myc-dependent angiogenic switch in tumors is mediated by interleukin 1beta. Genes Dev 20:2527-38 83. Lawlor ER, Soucek L, Brown-Swigart L, Shchors K, Bialucha CU, Evan GI 2006 Reversible kinetic analysis of Myc targets in vivo provides novel insights into Myc-mediated tumorigenesis. Cancer Res 66:4591-601 84. von der Lehr N, Johansson S, Wu S, Bahram F, Castell A, Cetinkaya C, Hydbring P, Weidung I, Nakayama K, Nakayama KI, Soderberg O, Kerppola TK, Larsson LG 2003 The F-box protein Skp2 participates in c- Myc proteosomal degradation and acts as a cofactor for c-Myc-regulated transcription. Mol Cell 11:1189-200

52 85. Lewis BC, Klimstra DS, Varmus HE 2003 The c-myc and PyMT oncogenes induce different tumor types in a somatic mouse model for pancreatic cancer. Genes Dev 17:3127-38 86. Chandrasekharappa SC, Guru SC, Manickam P, Olufemi SE, Collins FS, Emmert-Buck MR, Debelenko LV, Zhuang Z, Lubensky IA, Liotta LA, Crabtree JS, Wang Y, Roe BA, Weisemann J, Boguski MS, Agar- wal SK, Kester MB, Kim YS, Heppner C, Dong Q, Spiegel AM, Burns AL, Marx SJ 1997 Positional cloning of the gene for multiple endocrine neoplasia-type 1. Science 276:404-7 87. Larsson C, Skogseid B, Öberg K, Nakamura Y, Nordenskjöld M 1988 Multiple endocrine neoplasia type 1 gene maps to chromosome 11 and is lost in insulinoma. Nature 332:85-7 88. MacConaill LE, Hughes CM, Rozenblatt-Rosen O, Nannepaga S, Mey- erson M 2006 Phosphorylation of the menin tumor suppressor protein on serine 543 and serine 583. Mol Cancer Res 4:793-801 89. Marx SJ 2005 Molecular genetics of multiple endocrine neoplasia types 1 and 2. Nat Rev Cancer 5:367-75 90. Naito J, Kaji H, Sowa H, Kitazawa R, Kitazawa S, Tsukada T, Hendy GN, Sugimoto T, Chihara K 2006 Expression and functional analysis of menin in a multiple endocrine neoplasia type 1 (MEN1) patient with somatic loss of heterozygosity in chromosome 11q13 and unidentified germline mutation of the MEN1 gene. Endocrine 29:485-90 91. Fukuuchi A, Nagamura Y, Yaguchi H, Ohkura N, Obara T, Tsukada T 2006 A Whole MEN1 Gene Deletion Flanked by Alu Repeats in a Family with Multiple Endocrine Neoplasia Type 1. Jpn J Clin Oncol 36:739-44 92. Agarwal SK, Kennedy PA, Scacheri PC, Novotny EA, Hickman AB, Cerrato A, Rice TS, Moore JB, Rao S, Ji Y, Mateo C, Libutti SK, Oliver B, Chandrasekharappa SC, Burns AL, Collins FS, Spiegel AM, Marx SJ 2005 Menin molecular interactions: insights into normal functions and tumorigenesis. Horm Metab Res 37:369-74 93. Carling T, Correa P, Hessman O, Hedberg J, Skogseid B, Lindberg D, Rastad J, Westin G, Åkerström G 1998 Parathyroid MEN1 gene mutations in relation to clinical characteristics of nonfamilial primary hyperparathyroidism. J Clin Endocrinol Metab 83:2960-3 94. Wang EH, Ebrahimi SA, Wu AY, Kashefi C, Passaro E, Jr., Sawicki MP 1998 Mutation of the MENIN gene in sporadic pancreatic endocrine tumors. Cancer Res 58:4417-20 95. Zhuang Z, Ezzat SZ, Vortmeyer AO, Weil R, Oldfield EH, Park WS, Pack S, Huang S, Agarwal SK, Guru SC, Manickam P, Debelenko LV, Kester MB, Olufemi SE, Heppner C, Crabtree JS, Burns AL, Spiegel AM, Marx SJ, Chandrasekharappa SC, Collins FS, Emmert-Buck MR, Liotta LA, Asa SL, Lubensky IA 1997 Mutations of the MEN1 tumor suppressor gene in pituitary tumors. Cancer Res 57:5446-51 96. Zhuang Z, Vortmeyer AO, Pack S, Huang S, Pham TA, Wang C, Park WS, Agarwal SK, Debelenko LV, Kester M, Guru SC, Manickam P, Olufemi SE, Yu F, Heppner C, Crabtree JS, Skarulis MC, Venzon DJ, Emmert-Buck MR, Spiegel AM, Chandrasekharappa SC, Collins FS,

53 Burns AL, Marx SJ, Lubensky IA, et al. 1997 Somatic mutations of the MEN1 tumor suppressor gene in sporadic gastrinomas and insulinomas. Cancer Res 57:4682-6 97. Guo SS, Wu AY, Sawicki MP 2002 Deletion of chromosome 1, but not mutation of MEN-1, predicts prognosis in sporadic pancreatic endocrine tumors. World J Surg 26:843-847 98. 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. Am J Pathol 154:429-436 99. Toliat MR, Berger W, Ropers HH, Neuhaus P, Wiedenmann B 1997 Mutations in the MEN I gene in sporadic neuroendocrine tumours of gastroenteropancreatic system. Lancet 350:1223 100. 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. Proc Natl Acad Sci U S A 95:1630-4 101. Dreijerink KM, Hoppener JW, Timmers HM, Lips CJ 2006 Mecha- nisms of disease: multiple endocrine neoplasia type 1-relation to chromatin modifications and transcription regulation. Nat Clin Pract Endocrinol Me- tab 2:562-70 102. Agarwal SK, Guru SC, Heppner C, Erdos MR, Collins RM, Park SY, Saggar S, Chandrasekharappa SC, Collins FS, Spiegel AM, Marx SJ, Burns AL 1999 Menin interacts with the AP1 transcription factor JunD and represses JunD-activated transcription. Cell 96:143-52 103. 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. Proc Natl Acad Sci U S A 100:10770-5 104. Gobl AE, Berg M, Lopez-Egido JR, Öberg K, Skogseid B, Westin G 1999 Menin represses JunD-activated transcription by a histone deacetylase-dependent mechanism. Biochim Biophys Acta 1447:51-6 105. 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 Res 63:6135-9 106. Gallo A, Cuozzo C, Esposito I, Maggiolini M, Bonofiglio D, Vivacqua A, Garramone M, Weiss C, Bohmann D, Musti AM 2002 Menin uncou- ples Elk-1, JunD and c-Jun phosphorylation from MAP kinase activation. Oncogene 21:6434-45 107. Balogh K, Racz K, Patocs A, Hunyady L 2006 Menin and its interacting proteins: elucidation of menin function. Trends Endocrinol Metab 17:357- 64 108. Sukhodolets KE, Hickman AB, Agarwal SK, Sukhodolets MV, Obungu VH, Novotny EA, Crabtree JS, Chandrasekharappa SC, Collins FS, Spiegel AM, Burns AL, Marx SJ 2003 The 32-kilodalton subunit of replication protein A interacts with menin, the product of the MEN1 tumor suppressor gene. Mol Cell Biol 23:493-509

54 109. La P, Schnepp RW, C DP, A CS, Hua X 2004 Tumor suppressor menin regulates expression of insulin-like growth factor binding protein 2. Endo- crinology 145:3443-50 110. Heppner C, Bilimoria KY, Agarwal SK, Kester M, Whitty LJ, Guru SC, Chandrasekharappa SC, Collins FS, Spiegel AM, Marx SJ, Burns AL 2001 The tumor suppressor protein menin interacts with NF-kappaB proteins and inhibits NF-kappaB-mediated transactivation. Oncogene 20:4917-25 111. Basseres DS, Baldwin AS 2006 Nuclear factor-kappaB and inhibitor of kappaB kinase pathways in oncogenic initiation and progression. Oncogene 25:6817-30 112. Kaji H, Canaff L, Lebrun JJ, Goltzman D, Hendy GN 2001 Inactivation of menin, a Smad3-interacting protein, blocks transforming growth factor type beta signaling. Proc Natl Acad Sci U S A 98:3837-42 113. Ratineau C, Bernard C, Poncet G, Blanc M, Josso C, Fontaniere S, Calender A, Chayvialle JA, Zhang CX, Roche C 2004 Reduction of menin expression enhances cell proliferation and is tumorigenic in intestinal epithelial cells. J Biol Chem 279:24477-84 114. Sowa H, Kaji H, Hendy GN, Canaff L, Komori T, Sugimoto T, Chi- hara K 2004 Menin is required for bone morphogenetic protein 2- and transforming growth factor beta-regulated osteoblastic differentiation through interaction with Smads and Runx2. J Biol Chem 279:40267-75 115. Hughes CM, Rozenblatt-Rosen O, Milne TA, Copeland TD, Levine SS, Lee JC, Hayes DN, Shanmugam KS, Bhattacharjee A, Biondi CA, Kay GF, Hayward NK, Hess JL, Meyerson M 2004 Menin associates with a trithorax family histone methyltransferase complex and with the hoxc8 locus. Mol Cell 13:587-97 116. 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. Mol Cell Biol 24:5639-49 117. Milne TA, Hughes CM, Lloyd R, Yang Z, Rozenblatt-Rosen O, Dou Y, Schnepp RW, Krankel C, Livolsi VA, Gibbs D, Hua X, Roeder RG, Meyerson M, Hess JL 2005 Menin and MLL cooperatively regulate expression of cyclin-dependent kinase inhibitors. Proc Natl Acad Sci U S A 102:749-54 118. Yokoyama A, Somervaille TC, Smith KS, Rozenblatt-Rosen O, Meyer- son M, Cleary ML 2005 The menin tumor suppressor protein is an essential oncogenic cofactor for MLL-associated leukemogenesis. Cell 123:207- 18 119. Scacheri PC, Davis S, Odom DT, Crawford GE, Perkins S, Halawi MJ, Agarwal SK, Marx SJ, Spiegel AM, Meltzer PS, Collins FS 2006 Ge- nome-wide analysis of menin binding provides insights into MEN1 tumorigenesis. PLoS Genet 2:e51 120. Fontaniere S, Casse H, Bertolino P, Zhang CX 2006 Analysis of p27(Kip1) expression in insulinomas developed in pancreatic beta-cell specific Men1 mutant mice. Fam Cancer 5:49-54

55 121. Schnepp RW, Chen YX, Wang H, Cash T, Silva A, Diehl JA, Brown E, Hua X 2006 Mutation of tumor suppressor gene Men1 acutely enhances proliferation of pancreatic islet cells. Cancer Res 66:5707-15 122. Bai F, Pei XH, Nishikawa T, Smith MD, Xiong Y 2007 p18Ink4c, but Not p27Kip1, Collaborates with Men1 To Suppress Neuroendocrine Organ Tumors. Mol Cell Biol 27:1495-504 123. Bertolino P, Tong WM, Herrera PL, Casse H, Zhang CX, Wang ZQ 2003 Pancreatic beta-cell-specific ablation of the multiple endocrine neoplasia type 1 (MEN1) gene causes full penetrance of insulinoma development in mice. Cancer Res 63:4836-41 124. Perl AK, Wilgenbus P, Dahl U, Semb H, Christofori G 1998 A causal role for E-cadherin in the transition from adenoma to carcinoma. Nature 392:190-3 125. 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. Mol Cell Biol 24:6569-80 126. Adamska M, MacDonald BT, Sarmast ZH, Oliver ER, Meisler MH 2004 En1 and Wnt7a interact with Dkk1 during limb development in the mouse. Dev Biol 272:134-44 127. Sassoon D 1999 Wnt genes and endocrine disruption of the female reproductive tract: a genetic approach. Mol Cell Endocrinol 158:1-5 128. Ohira T, Gemmill RM, Ferguson K, Kusy S, Roche J, Brambilla E, Zeng C, Baron A, Bemis L, Erickson P, Wilder E, Rustgi A, Kitajewski J, Gabrielson E, Bremnes R, Franklin W, Drabkin HA 2003 WNT7a induces E-cadherin in lung cancer cells. Proc Natl Acad Sci U S A 100:10429-34 129. Winn RA, Marek L, Han SY, Rodriguez K, Rodriguez N, Hammond M, Van Scoyk M, Acosta H, Mirus J, Barry N, Bren-Mattison Y, Van Raay TJ, Nemenoff RA, Heasley LE 2005 Restoration of Wnt-7a expression reverses non-small cell lung cancer cellular transformation through frizzled-9-mediated growth inhibition and promotion of cell differentiation. J Biol Chem 280:19625-34 130. Calvo R, West J, Franklin W, Erickson P, Bemis L, Li E, Helfrich B, Bunn P, Roche J, Brambilla E, Rosell R, Gemmill RM, Drabkin HA 2000 Altered HOX and WNT7A expression in human lung cancer. Proc Natl Acad Sci U S A 97:12776-81 131. Winn RA, Van Scoyk M, Hammond M, Rodriguez K, Crossno JT, Jr., Heasley LE, Nemenoff RA 2006 Antitumorigenic effect of Wnt 7a and Fzd 9 in non-small cell lung cancer cells is mediated through ERK-5- dependent activation of peroxisome proliferator-activated receptor gamma. J Biol Chem 281:26943-50 132. Caricasole A, Ferraro T, Iacovelli L, Barletta E, Caruso A, Melchiorri D, Terstappen GC, Nicoletti F 2003 Functional characterization of WNT7A signaling in PC12 cells: interaction with A FZD5 x LRP6 receptor complex and modulation by Dickkopf proteins. J Biol Chem 278:37024-31

56 133. Sato N, Fukushima N, Maitra A, Matsubayashi H, Yeo CJ, Cameron JL, Hruban RH, Goggins M 2003 Discovery of novel targets for aberrant methylation in pancreatic carcinoma using high-throughput microarrays. Cancer Res 63:3735-42 134. Li S, Chiang TC, Davis GR, Williams RM, Wilson VP, McLachlan JA 2001 Decreased expression of Wnt7a mRNA is inversely associated with the expression of estrogen receptor-alpha in human uterine leiomyoma. J Clin Endocrinol Metab 86:454-7 135. Lustig B, Behrens J 2003 The Wnt signaling pathway and its role in tumor development. J Cancer Res Clin Oncol 129:199-221 136. Ropero S, Fraga MF, Ballestar E, Hamelin R, Yamamoto H, Boix- Chornet M, Caballero R, Alaminos M, Setien F, Paz MF, Herranz M, Palacios J, Arango D, Orntoft TF, Aaltonen LA, Schwartz S, Jr., Esteller M 2006 A truncating mutation of HDAC2 in human cancers con- fers resistance to histone deacetylase inhibition. Nat Genet 38:566-9 137. Zhang Z, Yamashita H, Toyama T, Sugiura H, Omoto Y, Ando Y, Mita K, Hamaguchi M, Hayashi S, Iwase H 2004 HDAC6 expression is correlated with better survival in breast cancer. Clin Cancer Res 10:6962-8 138. Gao L, Cueto MA, Asselbergs F, Atadja P 2002 Cloning and functional characterization of HDAC11, a novel member of the human histone deacetylase family. J Biol Chem 277:25748-55 139. Bradbury CA, Khanim FL, Hayden R, Bunce CM, White DA, Drayson MT, Craddock C, Turner BM 2005 Histone deacetylases in acute mye- loid leukaemia show a distinctive pattern of expression that changes selec- tively in response to deacetylase inhibitors. Leukemia 19:1751-9 140. Pelosi G, Bresaola E, Bogina G, Pasini F, Rodella S, Castelli P, Iacono C, Serio G, Zamboni G 1996 Endocrine tumors of the pancreas: Ki-67 immunoreactivity on paraffin sections is an independent predictor for malignancy: a comparative study with proliferating-cell nuclear antigen and progesterone receptor protein immunostaining, mitotic index, and other clinicopathologic variables. Hum Pathol 27:1124-34 141. La Rosa S, Sessa F, Capella C, Riva C, Leone BE, Klersy C, Rindi G, Solcia E 1996 Prognostic criteria in nonfunctioning pancreatic endocrine tumours. Virchows Arch 429:323-33 142. Beghelli S, Pelosi G, Zamboni G, Falconi M, Iacono C, Bordi C, Scarpa A 1998 Pancreatic endocrine tumours: evidence for a tumour suppressor pathogenesis and for a tumour suppressor gene on chromosome 17p. J Pathol 186:41-50 143. Chung DC, Smith AP, Louis DN, Graeme-Cook F, Warshaw AL, Ar- nold A 1997 A novel pancreatic endocrine tumor suppressor gene locus on chromosome 3p with clinical prognostic implications. J Clin Invest 100:404-10 144. Zhao J, Moch H, Scheidweiler AF, Baer A, Schaffer AA, Speel EJ, Roth J, Heitz PU, Komminoth P 2001 Genomic imbalances in the progression of endocrine pancreatic tumors. Genes Chromosomes Cancer 32:364-72

57 145. Chen YJ, Vortmeyer A, Zhuang Z, Huang S, Jensen RT 2003 Loss of heterozygosity of chromosome 1q in gastrinomas: occurrence and prognostic significance. Cancer Res 63:817-23 146. 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 tu- mourigenesis. J Pathol 210:450-8 147. Yang YM, Liu TH, Chen YJ, Jiang WJ, Qian JM, Lu X, Gao J, Wu SF, Sang XT, Chen J 2005 Chromosome 1q loss of heterozygosity frequently occurs in sporadic insulinomas and is associated with tumor malignancy. Int J Cancer 117:234-40 148. Barghorn A, Komminoth P, Bachmann D, Rutimann K, Saremaslani P, Muletta-Feurer S, Perren A, Roth J, Heitz PU, Speel EJ 2001 Dele- tion at 3p25.3-p23 is frequently encountered in endocrine pancreatic tumours and is associated with metastatic progression. J Pathol 194:451-8 149. Speel EJ, Richter J, Moch H, Egenter C, Saremaslani P, Rutimann K, Zhao J, Barghorn A, Roth J, Heitz PU, Komminoth P 1999 Genetic dif- ferences in endocrine pancreatic tumor subtypes detected by comparative genomic hybridization. Am J Pathol 155:1787-94 150. Barghorn A, Speel EJ, Farspour B, Saremaslani P, Schmid S, Perren A, Roth J, Heitz PU, Komminoth P 2001 Putative tumor suppressor loci at 6q22 and 6q23-q24 are involved in the malignant progression of sporadic endocrine pancreatic tumors. Am J Pathol 158:1903-11 151. Wild A, Langer P, Ramaswamy A, Chaloupka B, Bartsch DK 2001 A novel insulinoma tumor suppressor gene locus on chromosome 22q with potential prognostic implications. J Clin Endocrinol Metab 86:5782-7 152. Rigaud G, Missiaglia E, Moore PS, Zamboni G, Falconi M, Talamini G, Pesci A, Baron A, Lissandrini D, Rindi G, Grigolato P, Pederzoli P, Scarpa A 2001 High resolution allelotype of nonfunctional pancreatic endocrine tumors: identification of two molecular subgroups with clinical implications. Cancer Res 61:285-92 153. Hessman O, Skogseid B, Westin G, Åkerström G 2001 Multiple allelic deletions and intratumoral genetic heterogeneity in men1 pancreatic tumors. J Clin Endocrinol Metab 86:1355-61 154. Katona TM, Jones TD, Wang M, Abdul-Karim FW, Cummings OW, Cheng L 2006 Molecular evidence for independent origin of multifocal neuroendocrine tumors of the enteropancreatic axis. Cancer Res 66:4936- 42 155. Maitra A, Hansel DE, Argani P, Ashfaq R, Rahman A, Naji A, Deng S, Geradts J, Hawthorne L, House MG, Yeo CJ 2003 Global expression analysis of well-differentiated pancreatic endocrine neoplasms using oli- gonucleotide microarrays. Clin Cancer Res 9:5988-95 156. Hansel DE, Rahman A, House M, Ashfaq R, Berg K, Yeo CJ, Maitra A 2004 Met proto-oncogene and insulin-like growth factor binding protein 3 overexpression correlates with metastatic ability in well-differentiated pancreatic endocrine neoplasms. Clin Cancer Res 10:6152-8

58 157. Stålberg P, Grimfjärd P, Santesson M, Zhou Y, Lindberg D, Gobl A, Oberg K, Westin G, Rastad J, Wang S, Skogseid B 2004 Transfection of the multiple endocrine neoplasia type 1 gene to a human endocrine pancreatic tumor cell line inhibits cell growth and affects expression of JunD, delta-like protein 1/preadipocyte factor-1, proliferating cell nuclear antigen, and QM/Jif-1. J Clin Endocrinol Metab 89:2326-37 158. Stålberg P, Santesson M, Ekeblad S, Lejonklou MH, Skogseid B 2006 Recognizing genes differentially regulated in vitro by the multiple endocrine neoplasia type 1 (MEN1) gene, using RNA interference and oligonu- cleotide microarrays. Surgery 140:921-9; discussion 929-31 159. Roldo C, Missiaglia E, Hagan JP, Falconi M, Capelli P, Bersani S, Calin GA, Volinia S, Liu CG, Scarpa A, Croce CM 2006 MicroRNA expression abnormalities in pancreatic endocrine and acinar tumors are associated with distinctive pathologic features and clinical behavior. J Clin Oncol 24:4677-84 160. Lubensky IA, Pack S, Ault D, Vortmeyer AO, Libutti SK, Choyke PL, Walther MM, Linehan WM, Zhuang Z 1998 Multiple neuroendocrine tumors of the pancreas in von Hippel-Lindau disease patients: histopa- thological and molecular genetic analysis. Am J Pathol 153:223-31 161. Costa-Guda J, Rosen ED, Jensen RT, Chung DC, Arnold A 2005 Muta- tional analysis of PPARG as a candidate tumour suppressor gene in enteropancreatic endocrine tumours. Clin Endocrinol (Oxf) 62:603-6 162. Tomita T 2004 Cyclin-dependent kinase (cdk6) and p16 in pancreatic endocrine neoplasms. Pathology 36:566-70 163. Guo SS, Wu X, Shimoide AT, Wong J, Moatamed F, Sawicki MP 2003 Frequent overexpression of cyclin D1 in sporadic pancreatic endocrine tumours. J Endocrinol 179:73-9 164. Chung DC, Brown SB, Graeme-Cook F, Seto M, Warshaw AL, Jensen RT, Arnold A 2000 Overexpression of cyclin D1 occurs frequently in human pancreatic endocrine tumors. J Clin Endocrinol Metab 85:4373-8 165. Muscarella P, Melvin WS, Fisher WE, Foor J, Ellison EC, Herman JG, Schirmer WJ, Hitchcock CL, DeYoung BR, Weghorst CM 1998 Ge- netic alterations in gastrinomas and nonfunctioning pancreatic neuroendocrine tumors: an analysis of p16/MTS1 tumor suppressor gene inactivation. Cancer Res 58:237-40 166. Serrano J, Goebel SU, Peghini PL, Lubensky IA, Gibril F, Jensen RT 2000 Alterations in the p16INK4a/CDKN2A tumor suppressor gene in gastrinomas. J Clin Endocrinol Metab 85:4146-56 167. Bartsch DK, Kersting M, Wild A, Ramaswamy A, Gerdes B, Schuer- mann M, Simon B, Rothmund M 2000 Low frequency of p16(INK4a) alterations in insulinomas. Digestion 62:171-7 168. Liu L, Broaddus RR, Yao JC, Xie S, White JA, Wu TT, Hamilton SR, Rashid A 2005 Epigenetic alterations in neuroendocrine tumors: methylation of RAS-association domain family 1, isoform A and p16 genes are associated with metastasis. Mod Pathol 18:1632-40 169. House MG, Herman JG, Guo MZ, Hooker CM, Schulick RD, Lillemoe KD, Cameron JL, Hruban RH, Maitra A, Yeo CJ 2003 Aberrant hy-

59 permethylation of tumor suppressor genes in pancreatic endocrine neoplasms. Ann Surg 238:423-31; discussion 431-2 170. Chan AO, Kim SG, Bedeir A, Issa JP, Hamilton SR, Rashid A 2003 CpG island methylation in carcinoid and pancreatic endocrine tumors. On- cogene 22:924-34 171. Moore PS, Orlandini S, Zamboni G, Capelli P, Rigaud G, Falconi M, Bassi C, Lemoine NR, Scarpa A 2001 Pancreatic tumours: molecular pathways implicated in ductal cancer are involved in ampullary but not in exocrine nonductal or endocrine tumorigenesis. Br J Cancer 84:253-62 172. Lubomierski N, Kersting M, Bert T, Muench K, Wulbrand U, Schuermann M, Bartsch D, Simon B 2001 Tumor suppressor genes in the 9p21 gene cluster are selective targets of inactivation in neuroendocrine gastroenteropancreatic tumors. Cancer Res 61:5905-10 173. Guo SS, Wu X, Shimoide AT, Wong J, Sawicki MP 2001 Anomalous overexpression of p27(Kip1) in sporadic pancreatic endocrine tumors. J Surg Res 96:284-8 174. Canavese G, Azzoni C, Pizzi S, Corleto VD, Pasquali C, Davoli C, Crafa P, Delle Fave G, Bordi C 2001 p27: a potential main inhibitor of cell proliferation in digestive endocrine tumors but not a marker of benign behavior. Hum Pathol 32:1094-101 175. Rahman A, Maitra A, Ashfaq R, Yeo CJ, Cameron JL, Hansel DE 2003 Loss of p27 nuclear expression in a prognostically favorable subset of well-differentiated pancreatic endocrine neoplasms. Am J Clin Pathol 120:685-90 176. Chang HJ, Batts KP, Lloyd RV, Sebo TJ, Thompson GB, Lohse CM, Pankratz SV 2000 Prognostic Significance of p27, Ki-67, and Topoisom- erase lla Expression in Clinically Nonfunctioning Pancreatic Endocrine Tumors. Endocr Pathol 11:229-241 177. Foijer F, Te Riele H 2006 Restriction beyond the restriction point: mitogen requirement for G2 passage. Cell Div 1:8 178. Chung DC, Smith AP, Louis DN, Graeme-Cook F, Warshaw AL, Ar- nold A 1997 Analysis of the retinoblastoma tumour suppressor gene in pancreatic endocrine tumours. Clin Endocrinol (Oxf) 47:523-8 179. Yoshimoto K, Iwahana H, Fukuda A, Sano T, Saito S, Itakura M 1992 Role of p53 mutations in endocrine tumorigenesis: mutation detection by polymerase chain reaction-single strand conformation polymorphism. Can- cer Res 52:5061-4 180. Lam KY, Lo CY 1998 Role of p53 tumor suppressor gene in pancreatic endocrine tumors of Chinese patients. Am J Gastroenterol 93:1232-5 181. Bartz C, Ziske C, Wiedenmann B, Moelling K 1996 p53 tumour suppressor gene expression in pancreatic neuroendocrine tumour cells. Gut 38:403-9 182. Tang SJ, Dumot JA, Wang L, Memmesheimer C, Conwell DL, Zuc- caro G, Goormastic M, Ormsby AH, Cowell J 2002 Telomerase activity in pancreatic endocrine tumors. Am J Gastroenterol 97:1022-30 183. Lin SY, Elledge SJ 2003 Multiple tumor suppressor pathways negatively regulate telomerase. Cell 113:881-9

60 184. Hervieu V, Lepinasse F, Gouysse G, Guillaud O, Barel C, Chambon- niere ML, Bringuier PP, Poncet G, Lombard-Bohas C, Partensky C, Chayvialle JA, Scoazec JY 2006 Expression of {beta}-catenin in gastroenteropancreatic endocrine tumours: a study of 229 cases. J Clin Pathol 59:1300-4 185. Gerdes B, Ramaswamy A, Simon B, Pietsch T, Bastian D, Kersting M, Moll R, Bartsch D 1999 Analysis of beta-catenin gene mutations in pancreatic tumors. Digestion 60:544-8 186. Pavelic K, Hrascan R, Kapitanovic S, Karapandza N, Vranes Z, Be- licza M, Kruslin B, Cabrijan T 1995 Multiple genetic alterations in malignant metastatic insulinomas. J Pathol 177:395-400 187. Pavelic K, Hrascan R, Kapitanovic S, Vranes Z, Cabrijan T, Spaventi S, Korsic M, Krizanac S, Li YQ, Stambrook P, Gluckman JL, Pavelic ZP 1996 Molecular genetics of malignant insulinoma. Anticancer Res 16:1707-17 188. Wang DG, Johnston CF, Buchanan KD 1997 Oncogene expression in gastroenteropancreatic neuroendocrine tumors: implications for pathogenesis. Cancer 80:668-75 189. Pelengaris S, Khan M, Evan GI 2002 Suppression of Myc-induced apoptosis in beta cells exposes multiple oncogenic properties of Myc and triggers carcinogenic progression. Cell 109:321-34 190. Sato T, Konishi K, Kimura H, Maeda K, Yabushita K, Tsuji M, Miwa A 2000 Evaluation of PCNA, p53, K-ras and LOH in endocrine pancreas tumors. Hepatogastroenterology 47:875-9 191. Yashiro T, Fulton N, Hara H, Yasuda K, Montag A, Yashiro N, Straus F, 2nd, Ito K, Aiyoshi Y, Kaplan EL 1993 Comparison of mutations of ras oncogene in human pancreatic exocrine and endocrine tumors. Surgery 114:758-63; discussion 763-4 192. Yoshimoto K, Iwahana H, Fukuda A, Sano T, Katsuragi K, Kinoshita M, Saito S, Itakura M 1992 ras mutations in endocrine tumors: mutation detection by polymerase chain reaction-single strand conformation polymorphism. Jpn J Cancer Res 83:1057-62 193. Okai T, Kawashima A, Watanabe H, Takahashi Y, Sakai J, Ohtsubo K, Motoo Y, Matsui O, Murakami S, Nakabayashi H, Sawabu N 2000 Nonfunctioning islet cell carcinoma of the pancreas with high serum CEA & CA19-9, K-ras codon 12 mutation, and microsatellite instability. J Clin Gastroenterol 30:307-10 194. Ji L, Minna JD, Roth JA 2005 3p21.3 tumor suppressor cluster: prospects for translational applications. Future Oncol 1:79-92 195. House MG, Herman JG, Guo MZ, Hooker CM, Schulick RD, Cameron JL, Hruban RH, Maitra A, Yeo CJ 2003 Prognostic value of hMLH1 methylation and microsatellite instability in pancreatic endocrine neoplasms. Surgery 134:902-8; discussion 909 196. Peghini PL, Iwamoto M, Raffeld M, Chen YJ, Goebel SU, Serrano J, Jensen RT 2002 Overexpression of epidermal growth factor and hepatocyte growth factor receptors in a proportion of gastrinomas correlates with aggressive growth and lower curability. Clin Cancer Res 8:2273-85

61 197. Wulbrand U, Wied M, Zofel P, Goke B, Arnold R, Fehmann H 1998 Growth factor receptor expression in human gastroenteropancreatic neuroendocrine tumours. Eur J Clin Invest 28:1038-49 198. Shattuck TM, Costa J, Bernstein M, Jensen RT, Chung DC, Arnold A 2002 Mutational analysis of Smad3, a candidate tumor suppressor implicated in TGF-beta and menin pathways, in parathyroid adenomas and enteropancreatic endocrine tumors. J Clin Endocrinol Metab 87:3911-4 199. Bartsch D, Hahn SA, Danichevski KD, Ramaswamy A, Bastian D, Galehdari H, Barth P, Schmiegel W, Simon B, Rothmund M 1999 Mu- tations of the DPC4/Smad4 gene in neuroendocrine pancreatic tumors. On- cogene 18:2367-71 200. Perren A, Saremaslani P, Schmid S, Bonvin C, Locher T, Roth J, Heitz PU, Komminoth P 2003 DPC4/Smad4: no mutations, rare allelic imbalances, and retained protein expression in pancreatic endocrine tumors. Di- agn Mol Pathol 12:181-6 201. Scarpa A, Orlandini S, Moore PS, Lemoine NR, Beghelli S, Baron A, Falconi M, Zamboni G 2002 Dpc4 is expressed in virtually all primary and metastatic pancreatic endocrine carcinomas. Virchows Arch 440:155-9 202. Bader AG, Kang S, Zhao L, Vogt PK 2005 Oncogenic PI3K deregulates transcription and translation. Nat Rev Cancer 5:921-9 203. Stiles BL, Kuralwalla-Martinez C, Guo W, Gregorian C, Wang Y, Tian J, Magnuson MA, Wu H 2006 Selective deletion of Pten in pancreatic beta cells leads to increased islet mass and resistance to STZ-induced diabetes. Mol Cell Biol 26:2772-81 204. Perren A, Komminoth P, Saremaslani P, Matter C, Feurer S, Lees JA, Heitz PU, Eng C 2000 Mutation and expression analyses reveal differen- tial subcellular compartmentalization of PTEN in endocrine pancreatic tumors compared to normal islet cells. Am J Pathol 157:1097-103 205. Ekeblad S, Lejonklou MH, Grimfjärd P, Johansson T, Eriksson B, Grimelius L, Stridsberg M, Stålberg P, Skogseid B 2007 Co-expression of ghrelin and its receptor in pancreatic endocrine tumours. Clin Endocrinol (Oxf) 66:115-22 206. Granata R, Settanni F, Biancone L, Trovato L, Nano R, Bertuzzi F, Destefanis S, Annunziata M, Martinetti M, Catapano F, Ghe C, Is- gaard J, Papotti M, Ghigo E, Muccioli G 2007 Acylated and Unacylated Ghrelin Promote Proliferation and Inhibit Apoptosis of Pancreatic {beta}- Cells and Human Islets: Involvement of 3',5'-Cyclic Adenosine Monophos- phate/Protein Kinase A, Extracellular Signal-Regulated Kinase 1/2, and Phosphatidyl Inositol 3-Kinase/Akt Signaling. Endocrinology 148:512-29 207. Burtscher I, Compagni A, Lamm GM, Christofori G 1999 An insulin- like growth factor-mediated, phosphatidylinositol 3' kinase-independent survival signaling pathway in beta tumor cells. Cancer Res 59:3923-6 208. Hoog A, Kjellman M, Nordqvist AC, Hoog CM, Juhlin C, Falkmer S, Schalling M, Grimelius L 2001 Insulin-like growth factor-II in endocrine pancreatic tumours. Immunohistochemical, biochemical and in situ hybridization findings. Apmis 109:127-40

62 209. Petrik J, Pell JM, Arany E, McDonald TJ, Dean WL, Reik W, Hill DJ 1999 Overexpression of insulin-like growth factor-II in transgenic mice is associated with pancreatic islet cell hyperplasia. Endocrinology 140:2353- 63 210. Tonnies H, Toliat MR, Ramel C, Pape UF, Neitzel H, Berger W, Wiedenmann B 2001 Analysis of sporadic neuroendocrine tumours of the enteropancreatic system by comparative genomic hybridisation. Gut 48:536-41 211. Terris B, Meddeb M, Marchio A, Danglot G, Flejou JF, Belghiti J, Ruszniewski P, Bernheim A 1998 Comparative genomic hybridization analysis of sporadic neuroendocrine tumors of the digestive system. Genes Chromosomes Cancer 22:50-6 212. Chung DC, Brown SB, Graeme-Cook F, Tillotson LG, Warshaw AL, Jensen RT, Arnold A 1998 Localization of putative tumor suppressor loci by genome-wide allelotyping in human pancreatic endocrine tumors. Can- cer Res 58:3706-11 213. Nikiforova MN, Nikiforov YE, Biddinger P, Gnepp DR, Grosembacher LA, Wajchenberg BL, Fagin JA, Cohen RM 1999 Frequent loss of heterozygosity at chromosome 3p14.2-3p21 in human pancreatic islet cell tumours. Clin Endocrinol (Oxf) 51:27-33 214. Stumpf E, Aalto Y, Hoog A, Kjellman M, Otonkoski T, Knuutila S, Andersson LC 2000 Chromosomal alterations in human pancreatic endocrine tumors. Genes Chromosomes Cancer 29:83-7 215. 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 Cancer 22:130-7 216. Chen YJ, Vortmeyer A, Zhuang Z, Gibril F, Jensen RT 2004 X- chromosome loss of heterozygosity frequently occurs in gastrinomas and is correlated with aggressive tumor growth. Cancer 100:1379-87 217. Azzoni C, Bottarelli L, Pizzi S, D'Adda T, Rindi G, Bordi C 2006 Xq25 and Xq26 identify the common minimal deletion region in malignant gastroenteropancreatic endocrine carcinomas. Virchows Arch 448:119-26 218. Kloppel G, Perren A, Heitz PU 2004 The gastroenteropancreatic neuroendocrine cell system and its tumors: the WHO classification. Ann N Y Acad Sci 1014:13-27 219. Lopez-Egido J, Cunningham J, Berg M, Öberg K, Bongcam-Rudloff E, Gobl A 2002 Menin's interaction with glial fibrillary acidic protein and vimentin suggests a role for the intermediate filament network in regulating menin activity. Exp Cell Res 278:175-83 220. 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 25:3537-46 221. Busygina V, Kottemann MC, Scott KL, Plon SE, Bale AE 2006 Multiple Endocrine Neoplasia Type 1 Interacts with Forkhead Transcription Factor CHES1 in DNA Damage Response. Cancer Res 66:8397-403

63 222. Cupisti K, Hoppner W, Dotzenrath C, Simon D, Berndt I, Roher HD, Goretzki PE 2000 Lack of MEN1 gene mutations in 27 sporadic insulinomas. Eur J Clin Invest 30:325-9 223. Sjöblom T, Jones S, Wood LD, Parsons DW, Lin J, Barber TD, Man- delker D, Leary RJ, Ptak J, Silliman N, Szabo S, Buckhaults P, Farrell C, Meeh P, Markowitz SD, Willis J, Dawson D, Willson JK, Gazdar AF, Hartigan J, Wu L, Liu C, Parmigiani G, Park BH, Bachman KE, Papadopoulos N, Vogelstein B, Kinzler KW, Velculescu VE 2006 The consensus coding sequences of human breast and colorectal cancers. Sci- ence 314:268-74 224. Sotillo R, Renner O, Dubus P, Ruiz-Cabello J, Martin-Caballero J, Barbacid M, Carnero A, Malumbres M 2005 Cooperation between Cdk4 and p27kip1 in tumor development: a preclinical model to evaluate cell cycle inhibitors with therapeutic activity. Cancer Res 65:3846-52 225. Cozar-Castellano I, Takane KK, Bottino R, Balamurugan AN, Stewart AF 2004 Induction of beta-cell proliferation and retinoblastoma protein phosphorylation in rat and human islets using adenovirus-mediated transfer of cyclin-dependent kinase-4 and cyclin D1. Diabetes 53:149-59 226. Goldstein AM, Chan M, Harland M, Gillanders EM, Hayward NK, Avril MF, Azizi E, Bianchi-Scarra G, Bishop DT, Bressac-de Paillerets B, Bruno W, Calista D, Cannon Albright LA, Demenais F, Elder DE, Ghiorzo P, Gruis NA, Hansson J, Hogg D, Holland EA, Kanetsky PA, Kefford RF, Landi MT, Lang J, Leachman SA, Mackie RM, Magnus- son V, Mann GJ, Niendorf K, Newton Bishop J, Palmer JM, Puig S, Puig-Butille JA, de Snoo FA, Stark M, Tsao H, Tucker MA, Whitaker L, Yakobson E 2006 High-risk melanoma susceptibility genes and pancreatic cancer, neural system tumors, and uveal melanoma across GenoMEL. Cancer Res 66:9818-28 227. Nord B, Platz A, Smoczynski K, Kytola S, Robertson G, Calender A, Murat A, Weintraub D, Burgess J, Edwards M, Skogseid B, Owen D, Lassam N, Hogg D, Larsson C, Teh BT 2000 Malignant melanoma in patients with multiple endocrine neoplasia type 1 and involvement of the MEN1 gene in sporadic melanoma. Int J Cancer 87:463-7 228. Markham D, Munro S, Soloway J, O'Connor DP, La Thangue NB 2006 DNA-damage-responsive acetylation of pRb regulates binding to E2F-1. EMBO Rep 7:192-8 229. Nacusi LP, Sheaff RJ 2006 Akt1 sequentially phosphorylates p27kip1 within a conserved but non-canonical region. Cell Div 1:11 230. Bai F, Pei XH, Pandolfi PP, Xiong Y 2006 p18 Ink4c and Pten constrain a positive regulatory loop between cell growth and cell cycle control. Mol Cell Biol 26:4564-76 231. Latham KM, Eastman SW, Wong A, Hinds PW 1996 Inhibition of p53- mediated growth arrest by overexpression of cyclin-dependent kinases. Mol Cell Biol 16:4445-55 232. Vlach J, Hennecke S, Alevizopoulos K, Conti D, Amati B 1996 Growth arrest by the cyclin-dependent kinase inhibitor p27Kip1 is abrogated by c- Myc. Embo J 15:6595-604

64 233. Fontaniere S, Tost J, Wierinckx A, Lachuer J, Lu J, Hussein N, Busato F, Gut I, Wang ZQ, Zhang CX 2006 Gene expression profiling in insulinomas of Men1 {beta}-cell mutant mice reveals early genetic and epigenetic events involved in pancreatic {beta}-cell tumorigenesis. Endocr Relat Cancer 13:1223-36 234. Mahale S, Aubry C, James Wilson A, Jenkins PR, Marechal JD, Sut- cliffe MJ, Chaudhuri B 2006 CA224, a non-planar analogue of fascaplysin, inhibits Cdk4 but not Cdk2 and arrests cells at G0/G1 inhibiting pRB phosphorylation. Bioorg Med Chem Lett 16:4272-8 235. Mahale S, Aubry C, Jenkins PR, Marechal JD, Sutcliffe MJ, Chaud- huri B 2006 Inhibition of cancer cell growth by cyclin dependent kinase 4 inhibitors synthesized based on the structure of fascaplysin. Bioorg Chem 34:287-97 236. Toogood PL, Harvey PJ, Repine JT, Sheehan DJ, VanderWel SN, Zhou H, Keller PR, McNamara DJ, Sherry D, Zhu T, Brodfuehrer J, Choi C, Barvian MR, Fry DW 2005 Discovery of a potent and selective inhibitor of cyclin-dependent kinase 4/6. J Med Chem 48:2388-406 237. Baughn LB, Di Liberto M, Wu K, Toogood PL, Louie T, Gottschalk R, Niesvizky R, Cho H, Ely S, Moore MA, Chen-Kiang S 2006 A novel orally active small molecule potently induces G1 arrest in primary mye- loma cells and prevents tumor growth by specific inhibition of cyclin- dependent kinase 4/6. Cancer Res 66:7661-7

65 Acta Universitatis Upsaliensis Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 231

Editor: The Dean of the Faculty of Medicine

A doctoral dissertation from the Faculty of Medicine, Uppsala University, is usually a summary of a number of papers. A few copies of the complete dissertation are kept at major Swedish research libraries, while the summary alone is distributed internationally through the series Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine. (Prior to January, 2005, the series was published under the title “Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine”.)

ACTA UNIVERSITATIS UPSALIENSIS Distribution: publications.uu.se UPPSALA urn:nbn:se:uu:diva-7595 2007