The Wnt Signalling Cascade and the Adherens Junction Complex in ...

Endocr Pathol (2015) 26:1–8 DOI 10.1007/s12022-014-9341-8

The Wnt Signalling Cascade and the Adherens Junction Complex in Craniopharyngioma Tumorigenesis Veronica Preda & Sarah J. Larkin & Niki Karavitaki & Olaf Ansorge & Ashley B. Grossman

Published online: 30 October 2014 # Springer Science+Business Media New York 2014

Abstract Craniopharyngiomas are epithelial, sellar tumours with adamantinomatous (aCP) and papillary (pCP) subtypes. The aCP type usually occurs during childhood and pCP in middle-aged adults; aCPs often contain mutations in CTNNB1, encoding β-catenin, a component of the adherens junction and a mediator of Wnt signalling. No such mutational event has been associated with pCPs, where the BRAF gene appears to be more important. In a large series of 95 craniopharyngiomas, we show that the aCP subtype harbours mutations in CTNNB1 in 52 % of cases, while the pCP subtype does not, with agreement between immunohistochemistry and sequencing methods in the majority of cases. When present, the CTNNB1 mutation is found throughout the aCP tumour, while translocation of β-catenin from membrane to cytosol and nucleus is restricted to small cell clusters near the invading tumour front. We observed translocated βcatenin in 100 % of aCPs, occurring not only in cell clusters but also in individual cells scattered throughout the tumour. We characterised the adherens junction involving α-catenin, β-catenin, γ-catenin, p120 and E-cadherin (cytosolic and membranous components). Although suggested to be S. J. Larkin : O. Ansorge Nuffield Department of Clinical Neurosciences, John Radcliffe Hospital, Headley Way, Oxford OX3 9DU, UK S. J. Larkin : O. Ansorge Department of Neuropathology, John Radcliffe Hospital, Headley Way, Oxford OX3 9DU, UK V. Preda (*) : N. Karavitaki : A. B. Grossman Department of Endocrinology, Oxford Centre for Diabetes, Endocrinology and Metabolism, Churchill Hospital, Old Rd, Headington, Oxford OX3 7LE, UK e-mail: [email protected] V. Preda Kolling Institute, Royal North Shore Hospital, University of Sydney, Sydney, Australia

important in other sellar mass tumourigenesis pathways, there was no disruption of the adherens junction in these tumours, indicating that a loss of junctional integrity is not associated with β-catenin translocation or mutation. We conclude that mutations in CTNNB1 underlie tumourigenesis in the majority of aCPs, which are distinct morphologically and at the molecular level from pCPs. Keywords Craniopharyngioma . β-Catenin . Adamantinomatous . Adherens junction . Papillary

Introduction Craniopharyngiomas (CPs) are complex, epithelial tumours of the sellar region that occur with an incidence of 1.3 per million person years. No gender differences are observed, but age at diagnosis shows a bimodal distribution with peaks at 5–14 and 65–74 years [1]. Two subtypes of tumour have been recognised for over 100 years (reviewed in [2]): adamantinomatous craniopharyngiomas (aCP) occur predomi n a n t l y i n c h i l d h o o d , w h i l e t h e r a r e r p a p i ll a r y craniopharyngioma (pCP) is seen almost exclusively in adults (ratio aCP/pCP cases 9:1). Surgical treatment is usually the optimal first-line therapy; however, this is challenging due to their frequently large size and irregular margins and the adherence of the tumour to vital surrounding structures such as the visual pathways and the hypothalamus. As a result, complete surgical removal is often not possible and indeed may be inadvisable, and partial removal—if possible transsphenoidally—followed by external beam radiotherapy is usually recommended [3–5]. Studies of the pathogenesis of aCPs have highlighted the role of the Wnt pathway mediator and adherens junction component β-catenin [6]. Mutations that ablate critical serine and threonine residues (S33, S37, T41 and S45) of β-catenin

2

allow it to evade phosphorylation by the destruction complex and escape ubiquitin-mediated proteasomal degradation ([7–16] and reviewed in [17]). Consequently, β-catenin accumulates in the cytosol and nucleus [18], switching on Wnt signalling and mediating transcription of Wnt target genes in the absence of a Wnt ligand. Analysis of CTNNB1, the gene that encodes β-catenin, in craniopharyngiomas has revealed mutations at S33, S37, T41 or S45 in aCPs at a frequency of around 60 % (ranging from 16 to 100 %). No mutations have been reported in pCPs [6, 19–24]. In contrast, cytosolic and nuclear accumulation of β-catenin has been reported in 93 % of aCPs (range 78–100 %) but never in pCPs [6, 20, 21, 23, 25–28]. Expression of mutated β-catenin in the early stages of mouse pituitary organogenesis leads to aberrant pituitary development and the appearance of tumours resembling the human aCP, suggesting that β-catenin mutation could be a tumourigenic event [29, 30]. However, the presence of a mutation in CTNNB1 does not lead to β-catenin accumulation throughout the whole tumour [20, 21, 23, 31], and so, the relationship between mutation at β-catenin phosphorylation sites and its accumulation in the nucleus and cytosol remains to be clarified. Conversely, BRAF mutations have recently been associated with pCPs [32] [24]. In addition to its role in Wnt signalling, β-catenin is also a component of the adherens junction complex and facilitates cell-cell adhesion, along with α-catenin, γ-catenin and p120, by tethering the actin cytoskeleton to E-cadherin [33, 34]. Previous studies have demonstrated that the cells in aCPs with cytosolic and nuclear accumulation of β-catenin display increased motility compared with surrounding cells, a property that was reduced by siRNA-mediated knockdown of CTNNB1 [28]. Down-regulation of the expression of components of the adherens junction, including E-cadherin and p120 resulting from mutation, epigenetic silencing and allelic loss, has been implicated in the pathogenesis of several human malignancies including breast [35–37], pancreatic [38], gastric [39] [40], and colorectal cancers [41], and disruption of this complex is a common tumourigenic event. Given the role of adherens junction disassembly in cell migration and tumour progression, we examined the integrity of this complex in both aCPs and pCPs to determine whether the translocation of β-catenin seen in aCPs might be associated with adherens junction disruption and whether mislocalisation of other members of the complex might be associated with tumourigenesis in pCPs. We have therefore examined the relationship between mutation in CTNNB1 and the subcellular location of β-catenin in a large series of 95 craniopharyngiomas. The frequent sites of phosphorylation in exon 3 of CTNNB1 were sequenced, and the subcellular location of β-catenin and other components of the adherens junction complex (Ecadherin, α-catenin, γ-catenin and p120 catenin) were determined in both aCPs and pCPs.

Endocr Pathol (2015) 26:1–8

Materials and Methods We retrieved 131 specimens carrying a neuropathological diagnosis of craniopharyngioma from the surgical neuropathology archive of the Oxford Brain Bank. Diagnostic review confirmed classical features of aCP (defined by the presence of wet keratin nodules, calcification and/or stellate reticulum surrounded by peripheral palisaded epithelium) and pCP (defined by the absence of wet keratin nodules, calcification or stellate reticulum, and the presence of squamous epithelium with fibrovascular cores). Cases with insufficient epithelium for a definitive morphological diagnosis and ‘redo’ surgical specimens on the same patient were excluded, totalling 36 specimens. Studies were conducted under multi-site and local REC approval. Immunohistochemistry Immunohistochemical analysis was performed on 4-μm sections from formalin-fixed paraffin-embedded (FFPE) tumour specimens. Following dewaxing through graded alcohols, endogenous peroxidase activity was blocked (3 % (v/v) H2O2 in phosphate-buffered saline (PBS), pH 7.3, 20 min with orbital shaking). Epitope retrieval was achieved by autoclaving in sodium citrate (10 mM, pH 6.0, 10 min) (βcatenin). Sections were blocked with serum block (HiStar 3000 kit, AbD Serotec (Oxford, UK)) for 15 min and then incubated overnight at 4 °C with primary antibody diluted in PBS (see Table 1 for antibodies, concentrations and suppliers). All subsequent steps were carried out according to the manufacturer’s instructions (AbD Serotec) with the following modifications: HRP polymer was applied for 40 min and DAB for 5 min. Sections were counterstained with Cole’s haematoxylin. Of note, ECAD clone 4A2C7 was used, which is specific and does not stain the nuclei of normal human cortex. Previously published pituitary work has used the non-specific ECAD clone 36, resulting in false interpretation of translocation of ECAD and its role in Wnt signalling in pituitary tumours [42]. Others have also demonstrated non-specificity of this ECAD clone 36 for immunohistochemistry on FFPE sections [43]. Sequencing of CTNNB1 exon 3 DNA was extracted from 5×10-μm sections of FFPE tissue from archival surgical specimens (QiaAmp FFPE DNA kit, Qiagen (Crawley, UK)). PCR was performed to generate a 269-bp amplicon including codons S33, S37, T41 and S45 in exon 3 of CTNNB1. Primers were as follows: sense (5′-GATT TGATGGAGTTGGACATGG-3′) and antisense (5′-TGTTCT TGAGTGAAGGACTGAG-3′). DNA template (50 ng) was added to 10× PCR buffer solution (10 % v/v, Qiagen), MgCl2


3

Table 1 Antibody, concentration, suppliers and dilution factor for immunohistochemistry Antibody

Species

Manufacturer

Dilution

E-cadherin (membrane) E-cadherin (cytosol) α-Catenin β-Catenin γ-Catenin p120 Catenin

Mouse monoclonal (clone 36B5) Mouse monoclonal (clone 4A2C7) Mouse monoclonal (clone 5) Rabbit monoclonal (clone E247) Mouse monoclonal (clone 15) Rabbit monoclonal (clone EPR357(2))

Leica Invitrogen BD Transduction Laboratories Epitomics BD Transduction Laboratories Epitomics

1:50 1:800 1:500 1:1,000 1:1,000 1:1,000

(final concentration 4 mM, Qiagen), dNTPs (200 μM) (Promega, Southampton, UK), 0.5 U of Taq polymerase (HotStarTaq Plus, Qiagen) and 200 nM each of sense and antisense primers in a total volume of 20 μl. Cycling conditions were as follows: 95 °C for 5 min, followed by 35 cycles of 95 °C for 30 s, 60 °C for 30 s and 72 °C for 1 min. DNA template (100 ng) was added to 10× PCR buffer solution (10 % v/v, Qiagen), MgCl2 (final concentration 4 mM, Qiagen), dNTPs (200 μM) (Promega, Southampton, UK), 0.5 U of Taq polymerase (HotStarTaq Plus, Qiagen) and 400 nM each of sense and antisense primers in a total volume of 20 μl. Cycling conditions were as follows: 95 °C for 5 min, followed by 35 cycles of 95 °C for 30 s, 60 °C for 30 s and 72 °C for 40 s, and a final elongation step at 72 °C for 1 min. Products were examined by agarose gel separation and purified (MinElute PCR Purification kit (Qiagen)). Sequencing reactions were performed using BigDye Terminator chemistry and an ABI-3730 sequencer.

Results Nuclear Translocation of β-Catenin Is a Feature of aCPs but Not pCPs All the aCP cases in our series for which there was a sufficient epithelium to make a diagnosis (N=72) contained cells that showed nuclear and cytosolic accumulation of β-catenin (Fig. 1a). This pattern was absent in all of the pCP cases (N=23; Fig. 1b). Accumulation of β-catenin has previously been reported only in whorls of epithelial cells that form near to the infiltrating edge of the tumour. However, in our series, we observed that translocation of β-catenin did not always coincide with these whorl formations and could also occur in individual, isolated cells (Fig. 2).

Translocation of β-Catenin Is Mutation-Independent and Not Associated with Disruption of the Adherens Junction Complex Examination of the adherens junction complex in this series revealed that β-catenin is the only member of this complex to undergo translocation and accumulate in the cytosol and nucleus. No difference in the localisation of E-cadherin, α-catenin, γ-catenin or p120 catenin was observed in any of the aCPs or pCPs (Fig. 3). This was true for both the whole slide and for individual scattered cells showing β-catenin translocation. In aCP cases, the pattern of β-catenin staining was compared to mutations in CTNNB1. Cases were classified according to the presence of epithelial whorls and the pattern of β-catenin translocation as (1) individual-cell pattern of βcatenin accumulation, (2) whorl pattern of β-catenin accumulation or (3) whorl pattern without β-catenin accumulation. No relationship was found between CTNNB1 mutation and the pattern of β-catenin translocation (Fisher’s exact test, P= 0.182, 0.330 and 0.279, respectively). Sequencing of the region of exon 3 of CTNNB1 containing S33, S37, T41 and S45 was possible in 42 specimens, 9 of which were pCPs (the quality of the DNA was insufficient in the remainder of cases). Mutation was found in 52 % of aCPs in this series but in none of the pCPs (Table 2).

Discussion Accumulation of β-Catenin Occurs in All aCPs in Clusters or Individual Cells and Is Unrelated to CTNNB1 Mutation in Bulk Tumour In all of the cases of aCP in our series for which sufficient epithelium was available to make a definitive morphological diagnosis, accumulation of β-catenin was observed in either clusters of cells or individual epithelial cells. This finding

4


a

b a

b

_____


Fig.

1 a Adamantinomatous craniopharyngioma. Classical features include a peripheral palisading epithelium (open arrow), surrounding a loose stellate reticulum (SR) containing nodules of ‘wet keratin’ or ‘ghost cells’ (GC). a Whorls of epithelial cells near the infiltrating tumour edge (solid arrow) may also be visible. Translocation of β-catenin (brown reaction product, c) may be visible in clusters of cells associated with epithelial whorls (arrows, b and c). Haematoxylin and eosin (a, b). βcatenin immunohistochemistry (brown reaction product c). Scale bars a 80 μm, c 100 μm. b Papillary craniopharyngioma. Characterised by clefted (pseudo-)papillae and lacking the stellate reticulum, ghost cells and epithelial whorls characteristic of the adamantinomatous type. βCatenin remains at the membrane throughout these tumours. a Haematoxylin and eosin. b β-catenin immunohistochemistry (brown reaction product). Scale bar 100 μm

confirms the utility of β-catenin translocation as a tool in the diagnosis of this tumour subtype; the absence of observable β-catenin translocation is suggestive of a pCP. Previous studies have reported lower frequencies of observed β-catenin translocation [6, 20, 21, 25–28, 44]; some of these studies do not report individual cells with cytosolic or nuclear βcatenin which may have been present, and so, the number of cases with β-catenin translocation may have been underestimated. Accumulation of β-catenin in the cytosol and nucleus of individual epithelial cells has been reported previously [21, 45]. Cells with β-catenin translocation have been described adjacent to ghost cells [21, 27, 46], and Kato et al. proposed that cells with β-catenin translocation represent a transitional form between epithelial and ghost cells. Sequencing of bulk tumour revealed no relationship between mutations at S33, S37, T41 or S45 of CTNNB1 and the pattern of cytoplasmic or nuclear accumulation of β-catenin. The finding that not all aCPs harbour a mutation in CTNNB1 at these sites suggests that, while mutation at these sites may be a tumourigenic event in this lesion, it is neither sufficient nor necessary for tumour formation. This appears to be in contrast to studies of the effect of CTNNB1 mutation on early pituitary formation in mice, which showed that the expression of mutated β-catenin under the control of transcription factor Hesx1 (required for the formation of Rathke’s pouch) led to the

Fig. 2 Relationship between epithelial whorl formations and β-catenin translocation. Characteristic whorls of epithelial cells may (open arrows a, b) or may not (solid arrows a, b) correspond to translocation of βcatenin from membrane to cytosol/nucleus. Translocation of β-catenin may also occur in isolated cells not associated with epithelial whorls (a, b). Scale bar=80 μm

5

development of lesions containing clusters of cells with cytosolic and nuclear β-catenin reminiscent of those found in human aCPs [30], strongly implying a role for β-catenin mutation in the development of aCP. It should be noted that our study did not search for mutations at loci other than the four conserved phosphorylation sites in CTNNB1 which could have led to an underestimation of the proportion of specimens that harbour CTNNB1 mutations enabling stabilisation of βcatenin, for example mutations at codon 32 [6, 20–23]). However, in studies that did examine mutations at other loci by Sanger dideoxy sequencing methods, the rate of CTNNB1 mutation was still not 100 %, [6, 20–23], suggesting that either this method is not sufficiently sensitive to detect the presence of a mutation in all samples [47, 48] or there may be another tumourigenic event underlying formation of this tumour subtype. The use of FFPE tissue for the study allows direct comparison of findings to previous studies that report the frequency of CTNNB1 mutations [6, 20, 49]. The use of frozen material, along with next-generation sequencing technology, would yield much more robust results and with sufficient read depth would increase the sensitivity of the sequencing data in cases with a low ratio of tumour epithelium to normal tissue. In our series, the majority of mutations in CTNNB1 were at T41 (88 % of mutations) and all were T41I. This is in contrast to other published studies that report a more even distribution of mutations with the majority of mutations at S33 (from 33 to 60 % [6, 20–23]). The reason for the difference in distribution of mutations at the four conserved phosphorylation residues is unclear. Phosphorylation of β-catenin occurs at S45 (by CK1), T41, S37 and S33 (by GSK3β) sequentially [8], and phosphorylation of S33 is necessary for subsequent ubiquitination of β-catenin, so mutation at any of these loci should be sufficient to produce degradation-resistant β-catenin. It therefore seems unlikely that the difference in the site of mutation affects the stabilisation of β-catenin. The confinement of β-catenin accumulation to small clusters of cells or individual cells despite the presence of a

6


Fig. 3 Components of the adherens junction in aCP. Translocation of β-catenin from membrane to cytosol/nucleus occurs in clusters of cells near the infiltrating tumour edge (arrow, a). This translocation is not associated with disruption of the adherens junction in these cells. α-Catenin (b), γ-catenin (c), p120 catenin (d), and E-cadherin extracellular domain (e) and intracellular domain (f) retain their membranous location. Scale bar=100 μm

stabilising CTNNB1 mutation in bulk tumour has not been satisfactorily explained. Other authors have observed that these clusters show pituitary stem cell-like characteristics [29, 45, 50–52] and have lower proliferation indices compared to surrounding epithelium as assessed by Ki-67 immunohistochemistry [21, 25]. These cells also express Sonic hedgehog (SHH), transforming growth factor (TGF) and fibroblast growth factor (FGF) at higher levels than surrounding tumour cells, leading some authors to suggest that they represent a ‘signalling hub’, providing growth signals to the surrounding tumour [53–55]. In a small numbers of cases, analysis of the CTNNB1 sequence in DNA extracted from microdissected

cells with β-catenin translocation showed differences compared to the sequence of CTNNB1 from bulk tumour [20, 21, 23, 31]. A larger series is needed to clarify whether mutation in CTNNB1 may be confined to or different in cells that specifically translocate β-catenin. Translocation of β-Catenin Is Not Associated with Adherens Junction Disruption in aCPs We characterised the adherens junction involving α-catenin, β-catenin, γ-catenin, p120 and E-cadherin (cytosolic and membranous components). Previous studies have implicated

Table 2 Mutation loci of CTNNB1 Exon 3 of CTNNB1 encoding for B-catenin Residue S33

S37 T41

S45 IHC (translocation of B-catenin) and pattern

aCP sequencing possible in 33 specimens

2 (1×S>C, 1×S> 0 F)

15 (T> I)

0

pCP sequencing possible in nine specimens

0

0

0

0

100 % CTNNB1 mutation status and pattern of staining Individual cells (4) Whorls (13) Whorls (0) without 0 (0 %)

Fisher’s exact test p value 0.182 0.330 0.279


the adherens junction in other sellar mass tumourigenesis pathways, such as pituitary adenoma pathogenesis [42]. Using a specific antibody for each of the adherens junction components, we found that the complex was not disrupted in βcatenin-accumulating cells versus bulk tumour, a novel finding in craniopharyngiomas. E-cadherin (both extracellular and intracellular domains), α-catenin, γ-catenin (plakoglobin) and p120 catenin remained at the membrane in all tumour cells. Translocation of β-catenin in clusters or isolated cells is therefore not likely to be attributable to disruption of the adherens junction complex. In this series, β-catenin translocation did not always coincide with epithelial whorls of cells near to the invading tumour edge, and cells with cytosolic and nuclear β-catenin were often found at the edge of these formations or in isolated cells (Fig. 1a). Although often colocalised, epithelial whorls are not a marker for β-catenin translocation, and the relationship between these features remains to be defined. Inhibition of the Wnt pathway has proved challenging, and effective inhibitors are still largely in development [56, 57], but our findings clarify the need to explore other tumourigenesis pathways in order to find novel targeted agents.

Conclusions In our series, cytosolic or nuclear accumulation of β-catenin was present in 100 % of aCP cases in either clusters of cells or individual cells but was never present in pCPs. Mutation in CTNNB1 (S33, S37, T41 or S45) was present in 52 % of aCPs but was not linked to β-catenin translocation. Translocation of β-catenin from membrane to cytosol or nucleus is not associated with altered localisation of other members of the adherens junction complex. Acknowledgments We acknowledge the Oxford Brain Bank, supported by the Medical Research Council (MRC), Brains for Dementia Research (BDR) and the NIHR Oxford Biomedical Research Centre. The research was funded by the National Institute for Health Research (NIHR) Oxford Biomedical Research Centre based at Oxford University Hospitals NHS Trust and University of Oxford. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health.

References 1. Bunin, G.R., et al., The descriptive epidemiology of craniopharyngioma. Journal of neurosurgery, 1998. 89(4): p. 547–51. 2. Lindholm, J. and E.H. Nielsen, Craniopharyngioma: historical notes. Pituitary, 2009. 12(4): p. 352–9. 3. Karavitaki, N., et al., Craniopharyngiomas in children and adults: systematic analysis of 121 cases with long-term follow-up. Clin Endocrinol (Oxf), 2005. 62(4): p. 397–409.

7 4. Aggarwal, A., N. Fersht, and M. Brada, Radiotherapy for craniopharyngioma. Pituitary, 2013. 16(1): p. 26–33. 5. Karavitaki, N., et al., Craniopharyngiomas. Endocr Rev, 2006. 27(4): p. 371–97. 6. Buslei, R., et al., Common mutations of beta-catenin in adamantinomatous craniopharyngiomas but not in other tumours originating from the sellar region. Acta Neuropathol, 2005. 109(6): p. 589–97. 7. Aberle, H., et al., beta-catenin is a target for the ubiquitin-proteasome pathway. The EMBO journal, 1997. 16(13): p. 3797–804. 8. Liu, C., et al., Control of beta-catenin phosphorylation/degradation by a dual-kinase mechanism. Cell, 2002. 108(6): p. 837–47. 9. Filali, M., et al., Wnt-3A/beta-catenin signaling induces transcription from the LEF-1 promoter. The Journal of biological chemistry, 2002. 277(36): p. 33398–410. 10. Brabletz, T., et al., beta-catenin regulates the expression of the matrix metalloproteinase-7 in human colorectal cancer. The American journal of pathology, 1999. 155(4): p. 1033–8. 11. Crawford, H.C., et al., The metalloproteinase matrilysin is a target of beta-catenin transactivation in intestinal tumors. Oncogene, 1999. 18(18): p. 2883–91. 12. Jho, E.H., et al., Wnt/beta-catenin/Tcf signaling induces the transcription of Axin2, a negative regulator of the signaling pathway. Mol Cell Biol, 2002. 22(4): p. 1172–83. 13. Lustig, B., et al., Negative feedback loop of Wnt signaling through upregulation of conductin/axin2 in colorectal and liver tumors. Molecular and cellular biology, 2002. 22(4): p. 1184–93. 14. Zhang, X., J.P. Gaspard, and D.C. Chung, Regulation of vascular endothelial growth factor by the Wnt and K-ras pathways in colonic neoplasia. Cancer research, 2001. 61(16): p. 6050–4. 15. Behrens, J., et al., Functional interaction of beta-catenin with the transcription factor LEF-1. Nature, 1996. 382(6592): p. 638–42. 16. Molenaar, M., et al., XTcf-3 transcription factor mediates betacatenin-induced axis formation in Xenopus embryos. Cell, 1996. 86(3): p. 391–9. 17. Clevers, H., Wnt/beta-catenin signaling in development and disease. Cell, 2006. 127(3): p. 469–80. 18. Morin, P.J., et al., Activation of beta-catenin-Tcf signaling in colon cancer by mutations in beta-catenin or APC. Science, 1997. 275(5307): p. 1787–90. 19. Alatzoglou, K.S., et al., SOX2 haploinsufficiency is associated with slow progressing hypothalamo-pituitary tumours. Hum Mutat, 2011. 32(12): p. 1376–80. 20. Cani, C.M., et al., PROP1 and CTNNB1 expression in adamantinomatous craniopharyngiomas with or without betacatenin mutations. Clinics (Sao Paulo), 2011. 66(11): p. 1849–54. 21. Kato, K., et al., Possible linkage between specific histological structures and aberrant reactivation of the Wnt pathway in adamantinomatous craniopharyngioma. J Pathol, 2004. 203(3): p. 814–21. 22. Oikonomou, E., et al., Beta-catenin mutations in craniopharyngiomas and pituitary adenomas. J Neurooncol, 2005. 73(3): p. 205–9. 23. Sekine, S., et al., Beta-catenin mutations are frequent in calcifying odontogenic cysts, but rare in ameloblastomas. Am J Pathol, 2003. 163(5): p. 1707–12. 24. Brastianos, P.K., et al., Exome sequencing identifies BRAF mutations in papillary craniopharyngiomas. Nat Genet, 2014. 46(2): p. 161–5. 25. Cao, J., et al., Expression of aberrant beta-catenin and impaired p63 in craniopharyngiomas. Br J Neurosurg, 2010. 24(3): p. 249–56. 26. Hassanein, A.M., et al., beta-Catenin is expressed aberrantly in tumors expressing shadow cells. Pilomatricoma, craniopharyngioma, and calcifying odontogenic cyst. Am J Clin Pathol, 2003. 120(5): p. 732–6. 27. Hofmann, B.M., et al., Nuclear beta-catenin accumulation as reliable marker for the differentiation between cystic craniopharyngiomas and rathke cleft cysts: a clinico-pathologic approach. Am J Surg Pathol, 2006. 30(12): p. 1595–603.

8 28. Holsken, A., et al., Tumour cell migration in adamantinomatous craniopharyngiomas is promoted by activated Wnt-signalling. Acta Neuropathol, 2010. 119(5): p. 631–9. 29. Andoniadou, C.L., et al., Identification of novel pathways involved in the pathogenesis of human adamantinomatous craniopharyngioma. Acta Neuropathol, 2012. 30. Gaston-Massuet, C., et al., Increased Wingless (Wnt) signaling in pituitary progenitor/stem cells gives rise to pituitary tumors in mice and humans. Proc Natl Acad Sci U S A, 2011. 108(28): p. 11482–7. 31. Holsken, A., et al., Target gene activation of the Wnt signaling pathway in nuclear beta-catenin accumulating cells of adamantinomatous craniopharyngiomas. Brain Pathol, 2009. 19(3): p. 357–64. 32. Larkin, S.J., et al., BRAF V600E mutations are characteristic for papillary craniopharyngioma and may coexist with CTNNB1-mutated adamantinomatous craniopharyngioma. Acta Neuropathol, 2014. 33. Huber, A.H. and W.I. Weis, The structure of the beta-catenin/Ecadherin complex and the molecular basis of diverse ligand recognition by beta-catenin. Cell, 2001. 105(3): p. 391–402. 34. Nanes, B.A., et al., p120-catenin binding masks an endocytic signal conserved in classical cadherins. The Journal of cell biology, 2012. 199(2): p. 365–80. 35. Droufakou, S., et al., Multiple ways of silencing E-cadherin gene expression in lobular carcinoma of the breast. International journal of cancer. Journal international du cancer, 2001. 92(3): p. 404–8. 36. Batlle, E., et al., The transcription factor snail is a repressor of Ecadherin gene expression in epithelial tumour cells. Nature cell biology, 2000. 2(2): p. 84–9. 37. Berx, G. and F. Van Roy, The E-cadherin/catenin complex: an important gatekeeper in breast cancer tumorigenesis and malignant progression. Breast cancer research : BCR, 2001. 3(5): p. 289–93. 38. Mayerle, J., et al., Up-regulation, nuclear import, and tumor growth stimulation of the adhesion protein p120 in pancreatic cancer. Gastroenterology, 2003. 124(4): p. 949–60. 39. Corso, G., et al., Somatic mutations and deletions of the E-cadherin gene predict poor survival of patients with gastric cancer. Journal of clinical oncology : official journal of the American Society of Clinical Oncology, 2013. 31(7): p. 868–75. 40. Guilford, P., et al., E-cadherin germline mutations in familial gastric cancer. Nature, 1998. 392(6674): p. 402–5. 41. Richards, F.M., et al., Germline E-cadherin gene (CDH1) mutations predispose to familial gastric cancer and colorectal cancer. Human molecular genetics, 1999. 8(4): p. 607–10. 42. Elston, M.S., et al., Nuclear accumulation of e-cadherin correlates with loss of cytoplasmic membrane staining and invasion in pituitary adenomas. J Clin Endocrinol Metab, 2009. 94(4): p. 1436–42. 43. Lewis-Tuffin, L.J., et al., Misregulated E-cadherin expression associated with an aggressive brain tumor phenotype. PLoS One, 2010. 5(10): p. e13665.

Endocr Pathol (2015) 26:1–8 44. Sekine, S., et al., Expression of enamel proteins and LEF1 in adamantinomatous craniopharyngioma: evidence for its odontogenic epithelial differentiation. Histopathology, 2004. 45(6): p. 573–9. 45. Garcia-Lavandeira, M., et al., Craniopharyngiomas express embryonic stem cell markers (SOX2, OCT4, KLF4, and SOX9) as pituitary stem cells but do not coexpress RET/GFRA3 receptors. J Clin Endocrinol Metab, 2012. 97(1): p. E80-7. 46. Park, Y.S., et al., Late development of craniopharyngioma following surgery for Rathke's cleft cyst. Clin Neuropathol, 2009. 28(3): p. 177–81. 47. Rossle, M., et al., Ultra-deep sequencing confirms immunohistochemistry as a highly sensitive and specific method for detecting BRAF V600E mutations in colorectal carcinoma. Virchows Archiv : an international journal of pathology, 2013. 463(5): p. 623–31. 48. Routhier, C.A., et al., Comparison of 2 monoclonal antibodies for immunohistochemical detection of BRAF V600E mutation in malignant melanoma, pulmonary carcinoma, gastrointestinal carcinoma, thyroid carcinoma, and gliomas. Human pathology, 2013. 44(11): p. 2563–70. 49. Oikonomou E, B.D., Soares B, De Marco L, Buchfelder M, Adams EF., Beta-catenin mutations in craniopharyngiomas and pituitary adenomas. J Neurooncol. , 2005. 73(3): p. 205–9. 50. Fauquier, T., et al., SOX2-expressing progenitor cells generate all of the major cell types in the adult mouse pituitary gland. Proc Natl Acad Sci U S A, 2008. 105(8): p. 2907–12. 51. Gleiberman, A.S., et al., Genetic approaches identify adult pituitary stem cells. Proc Natl Acad Sci U S A, 2008. 105(17): p. 6332–7. 52. Holsken, A., et al., Adamantinomatous craniopharyngiomas express tumor stem cell markers in cells with activated Wnt signaling: further evidence for the existence of a tumor stem cell niche? Pituitary, 2013. 53. Ericson, J., et al., Integrated FGF and BMP signaling controls the progression of progenitor cell differentiation and the emergence of pattern in the embryonic anterior pituitary. Development, 1998. 125(6): p. 1005–15. 54. Wesche, J., K. Haglund, and E.M. Haugsten, Fibroblast growth factors and their receptors in cancer. Biochem J, 2011. 437(2): p. 199–213. 55. Yoo, Y.A., et al., Sonic hedgehog pathway promotes metastasis and lymphangiogenesis via activation of Akt, EMT, and MMP-9 pathway in gastric cancer. Cancer Res, 2011. 71(22): p. 7061–70. 56. Smith, M.L., G. Hawcroft, and M.A. Hull, The effect of non-steroidal anti-inflammatory drugs on human colorectal cancer cells: evidence of different mechanisms of action. European journal of cancer, 2000. 36(5): p. 664–74. 57. Dihlmann, S., A. Siermann, and M. von Knebel Doeberitz, The nonsteroidal anti-inflammatory drugs aspirin and indomethacin attenuate beta-catenin/TCF-4 signaling. Oncogene, 2001. 20(5): p. 645–53.