Accepted Manuscript

Altered expression of E-Cadherin-related transcription factors

indicates partial epithelial-mesenchymal transition in aggressive
meningiomas

Maren Wallesch, Doreen Pachow, Christina Blcher, Raimund

Firsching, Jan-Peter Warnke, Werner E.K. Braunsdorf, Elmar
Kirches, Christian Mawrin

PII: S0022-510X(17)30443-4
DOI: doi: 10.1016/j.jns.2017.07.009
Reference: JNS 15433
To appear in: Journal of the Neurological Sciences
Received date: 14 June 2017
Revised date: 4 July 2017
Accepted date: 6 July 2017

Please cite this article as: Maren Wallesch, Doreen Pachow, Christina Blcher, Raimund
Firsching, Jan-Peter Warnke, Werner E.K. Braunsdorf, Elmar Kirches, Christian Mawrin ,
Altered expression of E-Cadherin-related transcription factors indicates partial epithelial-
mesenchymal transition in aggressive meningiomas. The address for the corresponding
author was captured as affiliation for all authors. Please check if appropriate. Jns(2017),
doi: 10.1016/j.jns.2017.07.009

This is a PDF file of an unedited manuscript that has been accepted for publication. As
a service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting proof before
it is published in its final form. Please note that during the production process errors may
be discovered which could affect the content, and all legal disclaimers that apply to the
journal pertain.
 ACCEPTED MANUSCRIPT

Altered expression of E-Cadherin-related transcription factors indicates partial

epithelial-mesenchymal transition in aggressive meningiomas -revised
manuscript-

Maren Wallesch1, Doreen Pachow1, Christina Blcher1, Raimund Firsching2, Jan-

Peter Warnke3, Werner EK Braunsdorf4, Elmar Kirches1, and Christian Mawrin1,*

PT
1
Department of Neuropathology, Otto- von-Guericke University, Magdeburg
2
Department of Neurosurgery, Otto- von-Guericke University, Magdeburg

RI
3
Department of Neurosurgery, City Hospital; Magdeburg

SC
4
Department of Neurosurgery, Paracelsus-Hospital Zwickau, Germany
NU
MA
ED

Key words: Meningioma, E-Cadherin, Zo-1, Snail, Slug, Twist, NF2

T

Conflict of interest: None

C EP

*Correspondence:
AC

Christian Mawrin, MD
Department of Neuropathology
Otto-von-Guericke University Magdeburg, Germany
Tel: +49 391 6715825
Fax: +49 391 6713300
e-mail: christian.mawrin@med.ovgu.de
 ACCEPTED MANUSCRIPT

Summary

E-Cadherin has been suggested to be involved in meningioma progression but is

also known as a key player of epithelial to mesenchymal transition (EMT). We

wondered whether the adherens junction protein E-Cadherin, the tight junction

protein Zo-1, and transcription factors suppressing E-Cadherin expression (Slug,

Snail, Twist, Zeb-1) are differentially expressed between histopathological subtypes

PT
of meningioma, and if the expression of these factors is related to biological features

RI
of meningiomas. Analyzing 85 meningiomas of different histopathological subtypes

SC
and grades of malignancy by immunohistochemistry and 50 of them in addition by

real-Time-PCR, we observed significantly reduced expression of Zeb-1, Twist and

NU
Slug, together with slightly increased expression levels for E-Cadherin and Zo-1 in

fibroblastic WHO-grade I tumors compared to meningothelial WHO grade I tumors,

MA

contradicting the hypothesis of EMT in the fibroblastic meningiomas characterized by

mesenchymal appearance. However, comparing aggressive WHO grade II or III

ED

meningiomas with WHO-grade I tumors, we observed altered expression levels (loss

of E-Cadherin and Zo-1, increased expression of Zeb-1 and Slug) indicating

T
EP

molecular features of EMT in aggressive meningiomas. This was supported by

reduced E-Cadherin and increased Slug levels in recurrent compared to non-

C

recurrent meningiomas. The expression levels of E-cadherin and Zo-1 were positively
AC

correlated with expression of NF2 mRNA. In primary meningioma cultures and

IOMM-Lee meningioma cells, EMT induction by TGF- resulted in altered

morphology and increased expression of EMT associated transcription factors.

Meningioma cells with allelic losses of NF2 showed generally higher levels of various

EMT relevant proteins, but were unresponsive to TGF- treatment. Our data indicate

that aggressive meningiomas of WHO grade II/III are characterized by molecular

 ACCEPTED MANUSCRIPT

alterations indicating partial EMT. This might contribute to the aggressive biology of

these tumors.

PT
RI
SC
NU
MA
T ED
C EP
AC
 ACCEPTED MANUSCRIPT

Introduction

Meningiomas are generally benign tumors of the central nervous system, which arise

from the arachnoidal cap cells. They account for more than one-third of all primary

intracranial neoplasms [1, 2]. Being generally more common in women than in men,

meningiomas show a higher incidence with age, in fact they are the most common

type of brain tumors in patients over 70 [1, 3]. Meningiomas may be classified

PT
according to their location. Basically, meningiomas can arise either at the convexity,

RI
or at the skull base. Preferential locations of skull base meningiomas are the lateral

SC
sphenoid wing, the cavernous sinus, clivus, the olfactory groove, or the suprasellar

region, respectively. Intraventricular, intraorbital and spinal meningiomas, which tend

NU
to be located in the thoracic area, are less common [4].

Meningiomas are classified according to the World Health Organization (WHO) into
MA

three grades: The vast majority of tumors are benign grade I meningiomas, but about

20% are classified as atypical or invasive grade II meningiomas while 1-2% are
ED

considered as anaplastic grade III meningiomas with poor clinical outcome [2, 5].

However, even benign WHO I tumors may take a clinically difficult course, with
T
EP

increased recurrence and difficulties to completely remove the tumor surgically [6].

WHO grade I meningiomas can be further subdivided into different histological

C

subtypes. Meningothelial meningiomas are the most common meningiomas and are
AC

characterized by tumor-cell whorls with an epithelial phenotype. Fibroblastic and

transitional tumors represent further common subtypes, which are characterized by

spindle-like cells which grow in bundles, with a high content of reticulin fibres, overall

resembling a mesenchymal phenotype [2].

Common molecular changes found in meningiomas are alterations of the

Neurofibromatosis Type 2 (NF2) gene located on chromosome 22, which can be

 ACCEPTED MANUSCRIPT

found in more than fifty percent of sporadic meningiomas [7]. Interestingly, NF2

alterations are more frequent in fibroblastic than in meningothelial meningiomas [8].

The gene-product of NF2 gene is called Merlin (Moesin-Ezrin-Radixin-like protein) or

Schwannomin and acts as a linker between the actin cytoskeleton and

transmembrane proteins, regulating cell-cell and cell-matrix-adhesion. Merlin is

believed to function as a tumor suppressor protein which inhibits cell invasion [9].

PT
Not only has the role of NF2 alterations for meningioma development been

RI
convincingly demonstrated in animal models [10], it has furthermore been highlighted

SC
how timing and location of NF2 inactivation affect the development of either the

meningothelial or fibroblastic subtype [11]. In human meningiomas, NF2 alterations

NU
are preferentially found in tumors located at the convexity, while meningiomas

located at the skull base are mostly NF2 wild-type, but can show alterations in other
MA

genes, such as Smo, AKT1, and KLF4/TRAF7, and PI3K, respectively [8, 12-14].
ED

For various tumor types, the concept of epithelial to mesenchymal transition (EMT)

has been introduced in the last few years as an important mechanism to explain
T
EP

increased motility and metastatic behaviour. EMT is a complex biochemical process

with the consequence that formerly polarized, membrane bound epithelial cells
C

undergo morphological and biochemical changes with the acquisition of a

AC

mesenchymal phenotype. This phenotype is characterised by the loss of polarity and

increased invasive and migratory properties [15, 16]. EMT has been shown to be

relevant for the promotion of epithelial tumor cell invasiveness [17]. Moreover, the

occurrence of a reverse process called mesenchymal-epithelial transition (MET)

involved in the establishment of solid tumors from metastatic tumor cells has been

proposed [18, 19]. Basically, during EMT epithelial marker proteins (E-cadherin, ZO-

1, and others) are progressively lost, while mesenchymal marker proteins (N-
 ACCEPTED MANUSCRIPT

cadherin, Vimentin and others) gain increased expression levels during acquisition of

a spindle-cell, invasive phenotype (see [16, 20] for review). The transcription of E-

Cadherin is regulated by the transcription factors Snail1 (zinc finger protein), Snail2

(Slug), Zinc-finger E-box binding homebox 1 (Zeb-1), and Twist [21].

Using immunohistochemistry, it has been demonstrated that meningiomas are

PT
characterized by the expression of E-Cadherin and Vimentin as two of the major

RI
proteins. Both of them are used to distinguish meningiomas from other meningeal

SC
neoplasms [2, 22]. Moreover, the involvement of EMT in meningioma biology has

been suggested [23], but detailed studies are lacking so far. Given the important role
NU
of E-Cadherin for meningioma biology, the present study aimed to characterize the

relation between histopathological subtypes, EMT marker protein expression, and

MA

clinicopathological data including NF2 alterations in meningiomas.

T ED
C EP
AC
 ACCEPTED MANUSCRIPT

Materials & Methods

Tumor material

A total of 85 tumors from 85 meningioma patients (30 male, 55 female; median age:

68 years, range. 21-98 years) were investigated in the present study. All tumors were

graded according to the 2007 WHO classification of brain tumors. Brain-invasive

tumors were graded as WHO-grade II, in accordance with the new 2016 WHO

PT
classification system [24]. Formalin-fixed, paraffin-embedded (FFPE) tumor tissue

RI
was available for all tumors. In a fraction of these tumors (n=50), additional tissue

SC
was snap-frozen in liquid nitrogen directly after surgery and stored at -80C (see

demographic data for these samples in Table S2). Solely the snap frozen samples
NU
were used for RNA preparation and subsequent Real-Time-PCR.
MA

Cell culture and differentiation

Different meningioma cells were used to investigate effects of TGFbeta treatment on

ED

EMT-related gene expression. The cell lines Men/Men-NF2 were kindly provided by

Anita Lal, San Francisco (CA) [25]. The malignant meningioma line IOMM-Lee [26]
T
EP

was kindly provided by David Gutmann, St. Louis (MO). Finally, a primary culture

derived from the transitional WHO grade I meningioma of a 63 yearold woman was
C

used. Differentiation using TGFbeta (20ng/ml) was done as reported in [27].

AC

Immunohistochemistry

For immunohistochemistry, we used a tissue microarrays (TMA), which contained 55

samples covering all meningioma WHO grades, as previously described [28].

Immunohistochemical stainings were performed as described previously [28].

Immunostaining was assessed by an experienced observer (C.M.) as either

immunonegative (-) or immunopositive (+). TMAs normally contained two

 ACCEPTED MANUSCRIPT

representative spots from a given FFPE tumor block; however, due to technical

limitations, not all spots were visible on all TMA slides, giving rise to different case

numbers for individual immunostainings. The following antibodies were used

following intensive testing using whole slides: E-Cadherin (dilution 1:100), Snail

(1:100), Slug (1:100), Zo-1 (1:100), Zeb-1 (1:200), and Twist (1:100). All antibodies

were purchased from Cell Signaling Technology. Slides of glioblastoma tissue were

PT
used as positive controls for the antibodies and omission of primary antibody during

RI
staining of TMAs served as negative controls.

SC
SYBR-green based Real Time PCR
NU
Samples from 50 tumors were analyzed by real-time PCR, covering 37 WHO I (21

meningothelial, 7 fibroblastic, 8 transitional, 1 psammomatous) meningiomas, 7

MA

atypical WHO II, and 6 anaplastic WHO III meningiomas, respectively. RNA was

extracted from tumor samples using the RNA Mini Kit (Qiagen) according to the
ED

manufacturers instructions and diluted in 70l RNase free water. One g of the

yielded RNA was submitted to cDNA synthesis with a cloned reverse transcriptase
T
EP

(Bioline, Luckenwalde, Germany) according to the manufacturers instructions in a

total volume of 20 l. One microliter of the cDNA was hence subjected to

C

SybergreenTM (SensiMix SYBR Kit, Bioline, Germany) based real-time PCR (qPCR)
AC

analysis according to manufacturer s instructions using the ABIPrism 7000

Sequence Detection System (Applied Biosystems, Darmstadt, Germany). For data

analysis, the sample cycle threshold (Ct) values were normalized to that of the

housekeeping gene GAPDH. The used primers are listed in the supplementary Table

S1. Results are presented as the respective genes mean relative mRNA expression

(Er), which was calculated according to the formula E r = 1/2Ct, with C t being the
 ACCEPTED MANUSCRIPT

difference of C t- values between gene of interest (minuend) and housekeeping gene

(subtrahend) S.E.M.

LOH studies based on PCR-Amplified Polymorphic Markers

For NF2 LOH studies, available tumor samples with corresponding blood samples

were used. DNA isolation from corresponding tumor and blood was performed using

PT
the NucleoSpin Kit (Macery Nagel, Dren), according to the manufacturers

RI
instructions and then evaluated for LOH using PCR based markers, specific for the

SC
22q12 region as reported previously [29]. The PCR products were analyzed using the

ABI Prism 310 genetic analyser and GeneScan Software (PE Applied Biosystems,
NU
Weiterstadt, Germany). ROX-500 standard served for calibration of fragment lengths.
MA

Statistical Analyses

Comparisons of mRNA expressions in tumors were performed using the Mann-

ED

Whitney-U-test. Correlations were assessed by calculating Spearman correlation

coefficient. Expression levels of treated cells were determined by ANOVA, followed

T
EP

by a Tuckey post-hoc test. Significance was assumed for p 0.05. All calculations

were done using PASW Statistics (SPSS), Version 18.0.

C
AC
 ACCEPTED MANUSCRIPT

Results

First, we determined the expression of EMT-associated genes in WHO-grade I

meningiomas. As shown in Figure 1A, when compared to meningothelial

meningiomas, fibroblastic/transitional meningiomas showed slightly (not significant)

increased mRNA expression of the two epithelial marker proteins E-cadherin and Zo-

1, while in parallel they exhibited a significantly lower expression of the transcription

PT
factors Zeb-1, Twist and Slug, which are able to repress the E-Cadherin promoter

RI
[16]. This suggested a surprising shift to a more epithelial or less mesenchymal

SC
expression pattern of this marker set in the fibroblastic/transitional subtypes. As

reported in Table 1, analyses of immunostained TMAs revealed no major difference

NU
in the fraction of immunopositive tumors between both histological subgroups,

probably explainable by a general expression of these epithelial proteins and

MA

transcription factors above the sensitivity level of immunohistochemistry, not allowing

the detection of major subtype differences with a solely qualitative technique.

ED

Examples from the TMA are shown in Figure 1B. Because only 7 frozen fibroblastic

and 8 frozen transitional meningiomas had been available for quantitative mRNA
T
EP

analysis (see Methods), these two subtypes had been pooled in Figure 1A. Splitting

of these two subsets to allow separate comparisons between the meningothelial (n =

C

21), fibroblastic and transitional meningiomas, did not reveal any statistically
AC

significant group differences and results were not shown in the figure. Especially the

expression signatures for pure epithelial and pure fibroblastic subtypes were not

inverted, thus not giving any hint for EMT in the pure fibroblastic subtype.

To shed light on the altered expression of EMT-related genes in association with

increased malignancy in meningiomas, we compared unselected grade I tumors with

a series of WHO grade II/III meningiomas. Figure 2 shows that mRNA levels for E-

Cadherin and Zo-1 were significantly reduced in grade II/III meningiomas, while Slug
 ACCEPTED MANUSCRIPT

and Zeb-1 showed significant mRNA upregulation (A). Levels of Twist and Snail were

not changed significantly (data not shown). The observed shift was largely in

accordance with EMT in the high grade tumors, except for (not significant) tendency

of lower Twist levels in grade II/III compared to grade I meningiomas. We further

compared grade II and grade III meningiomas with all grade I meningiomas by using

immunohistochemistry. The results in Table 2 show that the fraction of E-Cadherin

PT
positive tumors is reduced in anaplastic meningiomas, while the transcription factors

RI
are fairly evenly expressed in all groups. Figure 2B shows examples from the

SC
stained TMAs.
NU
To assess the relationship between E-Cadherin expression and expression of the

individual transcription factors, we performed correlation analyses using mRNA

MA

expression data from a fraction of tumors with available snap-frozen tissue.

Significant negative correlations were found between E-Cadherin and Slug (r=-0.640)
ED

and between Zo-1 and Slug (r=-0.588). A significant positive correlation was found

between Zeb-1 and Slug (r=0.723). All other correlations remained nonsignificant.
T
EP

Completed EMT has been known to be linked to certain cell properties, such as
C

increased motility or invasiveness. We therefore compared expression levels of the

AC

chosen set of markers with clinical data. A group of meningiomas of WHO-grade I

with known tumor recurrence (n=5) was compared with non-recurrent tumors of

WHO-grade I (n=29), for which the information was available that no recurrence had

occurred. We found a significant loss of E-Cadherin in recurrent meningiomas

accompanied by a significant upregulation of Slug in meningiomas with recurrence

(Figure 3).
 ACCEPTED MANUSCRIPT

Based on the fact that NF2 is altered in approximately half of the sporadic

meningiomas, we studied the relation between the expression of NF2, E-Cadherin,

Zo-1 and the EMT-related transcription factors in those tumors of WHO-grade I in

more detail, for which the LOH-status was available. Because sequencing data of the

complete NF2 exome, including splice sites, were not available in this series, it

remained unknown, if a bi-allelic genetic inactivation had occurred in the tumors with

PT
LOH. As shown in Figure 4A, significantly positive correlations were observed

RI
between NF2 and Zo-1 (r=0.603), E-Cadherin (r=0.560), and Twist (r=0.491),

SC
respectively. By analyzing selectively meningiomas from patients with known allelic

loss at chromosome 22 (LOH) (n=12), NF2 expression was still significantly

NU
correlated with E-cadherin (r=0.637) and Zo-1 expression (r=0.659), indicating a

positive correlation between the expression of NF2 and the two epithelial markers.
MA

The expression of E-Cadherin and Zo-1, which are directly or indirectly linked to

cortical actin fibers and which are essential to form epithelial tight junctions, was thus
ED

found to be correlated in meningiomas with the expression of NF2/Merlin, which links

actin fibers to membrane proteins, is associated with stability of meningeal cell-cell

T
EP

contacts and an accepted meningeal tumor suppressor [30]. No significant

correlations between these mRNAs were found, if the analysis was restricted to the
C

subgroup of meningiomas without LOH at the NF2 locus (n = 11, data not shown).
AC

Finally, we wondered whether induction of EMT can alter the expression of the EMT-

related factors E-Cadherin, Snail, Slug and Twist in human meningioma cells in vitro.

We treated different meningioma cells with the growth factor TGFbeta, which has

been widely used for the induction of EMT. As shown in Figure 5A, a primary

meningioma culture (WHO grade I tumor) showed both, morphological changes

(upper panel), as well as induction of E-Cadherin, Snail, Slug and Twist expression
 ACCEPTED MANUSCRIPT

(lower figure). However, the mRNA expression changes did not reach significance

statistically. By treating the malignant meningioma cell line IOMM-Lee (Figure 5B)

we again observed morphological changes in the cells (upper panel) with elongated

cell bodies (arrows), as well as a clear induction of Snail and Slug mRNA expression,

while low E-Cadherin levels were not changed.

PT
To further elucidate a potential relation between the expression of EMT factors and

RI
NF2, we sought to induce EMT with TGFbeta in a meningioma cell line with

downregulated Merlin [25]. As shown in Figure 5C, we observed that TGFbeta was

SC
able to induce expression changes in cells with intact NF2/merlin, kept as control. We
NU
observed significant increase of Snail mRNA expression. However, to our surprise

the NF2/merlin knockdown cells had a remarkable high basal expression of E-

MA

Cadherin, Snail, Twist, and Slug, while TGFbeta treatment did not result in any

change of the expression level. Comparable data were found using another NF2-
ED

deficient meningioma cell line (data not shown).

T
C EP
AC
 ACCEPTED MANUSCRIPT

Discussion

Evidences for EMT in high-grade and recurrent meningiomas

Meningiomas vary considerably in their histopathological appearance and their

biological features. Higher-grade meningiomas WHO grade II or III for example

display a more aggressive behaviour and can cause significant morbidity and

mortality. We hence wondered whether alterations in the expression of mRNA and

PT
protein for E-Cadherin and Zo-1, constitutive components of epithelial adherens

RI
junctions and tight junctions [31], might be related to histopathological subtypes and

SC
aggressive behaviour of meningiomas. We included the transcription factors Zeb-1,

Slug and Twist, which are able to repress the activity of the E-Cadherin promoter and
NU
are involved in epithelial-to-mesenchymal transition (EMT) in various cancer diseases

together with E-cadherin and Zo-1 [16]. In the course of EMT in epithelial tumors,
MA

downregulation of E-Cadherin during loss of functional cell-cell-contacts is the most

important change and may be accompanied by upregulation of the E-Cadherin

ED

suppressing transcriptions factors and eventually by downregulation of Zo-1, located

at the cytoplasmic surface of tight junctions and being associated with actin filaments
T
EP

of the cortical cytoplasm [32].

Indeed, we observed a significant downregulation of both epithelial markers in

C

meningiomas of grades II and III, accompanied by a significant upregulation of Zeb-1

AC

and Slug (Figure 2). Despite the non-significant tendency for decreased Twist

expression and increased Snail (not shown), these results fit quite well to the

presence of EMT in higher-grade meningiomas, i.e. loss of organized cell-cell-

contacts, including loss of anchoring of actin fibres to these structures, which are

normally realized by direct binding to Zo-1 or binding to E-Cadherin via beta-Catenin,

allowing enhanced motility and invasiveness, as observed in patients and reflected

by brain infiltration and partially by metastasis occurring in high-grade tumors. The

 ACCEPTED MANUSCRIPT

decreased expression of E-Cadherin mRNA in recurrent meningiomas, together with

their increased levels of Slug (Figure 3) further support the notion that loss of

organized cell-cell-contacts in meningiomas allow increased motility and invasion as

a basis of enhanced recurrence risk.

No EMT in fibroblastic meningiomas despite mesenchymal morphology

PT
While the histopathological appearance of meningiomas can be diverse, the majority

RI
of WHO grade I meningiomas belongs either to the meningothelial or

fibrous/fibroblastic subtype [2]. The epithelial-like appearance of the meningothelial

SC
subtype and the mesenchymal-like appearance of the fibrous/fibroblastic subtype
NU
implies different biology which should be reflected by distinct gene expression

characteristics. A shift to a more mesenchymal gene expression signature may be

MA

assumed.

However, with respect to the markers analysed in the present study, a less clear
ED

picture became visible for the comparison of fibroblastic versus meningothelial

meningiomas, as compared to the relatively clear distinction between tumor grades

T
EP

depicted above. No major differential expression of mRNAs for both tight junction

proteins was observed (Figure 1), although a loss might have been expected in the
C

fibroblastic subtype due to the mesenchymal morphology. In contrast, the significant

AC

downregulation of all E-Cadherin-suppressing transcription factors (Figure 1),

together with the minimal upregulation of both epithelial markers, may point exactly

into the opposite direction. These results contradict speculations about a potentially

higher motility of fibroblastic meningioma cells due to EMT. However, the inverse

conclusion of a partial MET (mesenchymal-to-epithelial transition) would be

premature at this point, because the most essential aspect, i.e. an upregulation of E-
 ACCEPTED MANUSCRIPT

Cadherin in the fibroblastic subtype, cannot be safely deduced from the data.

Upregulation was less than 1.5-fold and did not reach statistical significance.

Microarray studies, which were not restricted to EMT-related proteins, but analyzed a

broad spectrum of genes, detected subtype-specific expression signatures, which

may fit with a more mesenchymal pattern in fibroblastic meningiomas. Fevre-

Montange et al. [33] identified by microarray gene profiling an expression signature

PT
specific for fibroblastic meningiomas composed of genes related to extracellular

RI
matrix (ECM) biology (Tenascin C, MMP2, and FBLN1), while meningothelial

SC
meningiomas were characterized by expression of MLPH, DEFB1, and FAT3.

Remarkably, two of the genes with a striking expression difference between the two
NU
subtypes, i.e. the Wnt-pathway regulator APCDD1 and the above mentioned FBLN1,

were meanwhile shown to clearly distinguish these subtypes in a transgenic mouse

MA

model of NF2-dependent meningioma genesis [11]. A lower expression of APCDD1

and relative overexpression of FBLN1 seemed to be a general feature of human

ED

fibroblastic meningiomas and was reproduced in the corresponding histologic

subtype derived from an NF2-loss induced mouse meningioma model [11]. The study
T
EP

suggested that both histologic variants -with distinguishable mRNA signatures - can

be derived by NF2-loss from a common primordial arachnoidal cell, characterized by

C

positivity for prostaglandin-D2-synthase. Besides supporting differential gene

AC

expression as a correlate of morphology, this study in addition argued against two

different cell types of origin.

The association between fibroblastic subtype and ECM-related genes was confirmed

in another study, reporting high levels of MMPs, ADAMs and TIMPs in fibroblastic

subtype meningioma [34]. These patterns fit for example well with a higher capacity

of fibroblastic meningioma cells for proteolytic ECM degradation, a prerequisite for

brain invasion. In a previous cooperation project, we found an enrichment of genes

 ACCEPTED MANUSCRIPT

which are regulated by the transcription factors Sp1 and AGP/EBP in fibroblastic

meningioma as compared to meningothelial tumors [35]. Aarhus et al. [36] identified

another expression signature specific for fibrous meningiomas (BMPR1B, DMD,

RAMP1). Moreover, it was reported that NCAM expression is a feature of fibroblastic

meningiomas, while CD44 and E-Cadherin expression designates the meningothelial

subtype [37, 38].

PT
RI
Clues from previous immunohistochemical studies

SC
Up to this point, E-Cadherin has been well characterized in meningiomas, mainly by

expression studies using immunohistochemistry. E-Cadherin acts as a tumor

NU
suppressor and regulates cell polarity, differentiation, and migration [21]. While some

immunohistochemical studies could not find a relation between meningioma grading

MA

and E-Cadherin expression [22, 39] other studies suggested a reduced E-Cadherin

expression in aggressive meningiomas [40-42], which fits well to the results of the
ED

present study. Moreover, loss of E-Cadherin has been related to invasive

meningioma phenotype [42]. Therefore, it appears that E-cadherin is expressed

T
EP

differentially in meningiomas related to the histopathological subtype, and the grade

of malignancy including the capacity for meningioma cell infiltration. This indicates at
C

least for the aggressive meningioma variants the contribution of EMT-related

AC

mechanisms.

The process of EMT has been attributed to tumor cell invasion and the formation of

metastasis [16]. One of the key molecules in this process is E-Cadherin, whose loss

during EMT is regulated by transcriptional repression through transcription factors

such as Snail, Slug, and Twist [21]. Finally, a participation of EMT-related processes

in meningioma biology has been suspected, mainly based on an altered E-Cadherin

expression [23]. The expression of E-Cadherin-related transcription factors in

 ACCEPTED MANUSCRIPT

meningiomas so far has been reported only in a few papers and was studied

basically by immunohistochemistry. Nagaishi et al. [43] found reduced E-Cadherin

expression in atypical meningioma, but only a few cases with immunoexpression of

Twist, which were mainly of angiomatous subtype, and only 1 meningioma was

immunopositive for Slug. In another study, Twist expression was observed in only 8

out of 103 meningiomas [44].

PT
In the present study we applied a more specific quantitative approach by measuring

RI
mRNA expression levels. Its main message, i.e E-Cadherin loss and EMT in high

SC
grade meningiomas, fits with previous immunohistochemical studies, but further

supports EMT by proving the upregulation of relevant transcription factors.

NU
In the present study, a role of EMT-related transcription factors for the biology of

aggressive meningiomas is supported by the observation of reduced E-Cadherin and

MA

increased Slug expression in recurrent meningiomas. Meningioma recurrence is seen

in high-grade meningiomas or in meningiomas with incomplete resection [45, 46].

ED

Loss of E-Cadherin expression has been shown to promote an aggressive tumor

phenotype including early recurrence in different types of cancer [47-49]. In contrast,

T
EP

upregulation of Slug has been linked to unfavorable prognosis and early tumor

recurrence [50-52]. Thus, our data suggest that low E-Cadherin and high Slug
C

expression might indicate the presence of aggressive meningioma with increased

AC

recurrence risk.

While a reduced expression of NF2 and its protein product merlin has been clearly

linked to tumor invasion [53, 54], reduced Twist expression has not clearly shown to

be associated with tumor cell invasion so far. While a few papers described a

reduced Twist expression together with enhanced metastasis and invasive potential

[55, 56], others showed an association between tumor invasion and high Twist

expression [57, 58].

 ACCEPTED MANUSCRIPT

Role of NF2

Loss of NF2 and its protein merlin is a feature in more than half of sporadic

meningiomas [59], but the biological consequences of this alteration is less clearly

defined [60]. Our finding of significant positive correlations between NF2 expression

and expression of both epithelial markers may indicate a functional relation between

those molecules. This aspect warrants further investigations in the future, because

PT
two of these molecules (E-Cadherin and Zo-1) can be found in adherens/tight

RI
junctions and are both associated with actin filaments [31], while NF2/merlin loss in

SC
meningeal cells is associated with loss of organized tight junctions [30] and the

protein is also connected to the actin cytoskeleton. All three proteins seem to be
NU
involved in contact inhibition and suppression of tumor cell motility. Two of them

(NF2, E-Cadherin) can be regarded as tumor suppressors. Surprisingly, the available

MA

data regarding this potential association are very limited [61].

In this context the hypothesis of a potential continuum between fibroblastic

ED

meningiomas with NF2-inactivation and meningiomas of higher WHO-grade

(progression) may shortly be discussed. Such a hypothesis may be formulated due to

T
EP

the fact of extremely frequent NF2-inactivation in high-grade meningiomas [62, 63],

which is accompanied by a 2 to 3-fold higher percentage of NF2 mutations [64] and

C

of chromosomal losses and true bi-allelic NF2-inactivations [65] in fibroblastic as

AC

compared to meningothelial meningiomas. It must be stated that analysis of EMT

markers in the present study cannot deliver a specific contribution to this debate,

because hints for an EMT were solely found in the high-grade tumors, but not in the

fibroblastic subtype. Our study delivered no clues to define EMT as a link between

NF2-associated tumorigenesis and progression to higher grades. On the other hand,

the mRNA profiles of meningothelial and fibroblastic tumors are distinct with respect

to other genes, not involved in EMT [33], even in mouse models of NF2-dependentt
 ACCEPTED MANUSCRIPT

tumorigenesis [11]. This leaves open the general possibility that some of these

genes, which are differentially expressed among grade-I tumors, may contribute to

progression.

Lessons from cell cultures

The use of human biopsy material limits the experimental options to further elucidate

PT
the role of EMT in meningioma biology. By treating meningioma cell cultures with

RI
TGF- to induce EMT, despite some morphological alterations of the treated cells,

SC
we observed no major changes in E-Cadherin. At least no downregulation occurred

after TGF- treatment, despite a quite systematic increase of most E-Cadherin

NU
suppressing transcription factors (Figure 5). The increase of transcription factors was

observed in a primary meningioma culture derived from a benign meningioma, and in

MA

malignant (IOMM-Lee, Men) meningioma cells. Because malignant IOMM-Lee cells

exhibited a very low basal level of E-Cadherin, an increase of the transcription factors
ED

by TGF- stimulation may allow no further suppression. On the other hand it

remained astonishing that also the primary cells and Men cells did not react with a
T
EP

decrease of the desmosomal protein. Not even Men-shNF2 cells with significantly

higher E-Cadherin expression reacted with any loss of the marker. Surprisingly, the
C

use of Men-shNF2 meningioma cells with downregulated NF2 revealed that the loss
AC

of NF2 resulted in increased expression of E-Cadherin and the E-Cadherin-

associated transcription factors Snail, Slug, and Twist. So far, no studies

investigating the relation between NF2 and Twist are reported. For Snail, one study

showed a functional relation between Snail and Nf2 in mesothelioma [66].

Taken together, our results confirm and extend previous findings of E-Cadherin

downregulation in high grade meningiomas and the suspected role of EMT in high

grade and recurrent meningiomas. It contradicts a role of EMT in the grade I

 ACCEPTED MANUSCRIPT

meningiomas of mesenchymal phenotype (fibroblastic subtype) and it underlines a

positive correlation between the well-recognized meningeal tumor suppressor NF2

and the two tight junction proteins E-Cadherin and Zo-1, which warrants further

investigations.

PT
RI
SC
NU
MA
T ED
C EP
AC
 ACCEPTED MANUSCRIPT

Figure legends

Figure 1: Expression analyses of E-Cadherin and E-Cadherin-dependent

transcription factors in WHO-grade I meningiomas of meningothelial or fibroblastic

subtype (the latter pooled with transitional). A Real-time PCR analysis of mRNA

expression levels reveals increased E-cadherin levels in fibroblastic meningiomas

compared to meningothelial meningiomas. The increased E-Cadherin expression is

PT
associated with reduced expression of Zeb-1, Twist, and Slug, while Zo-1 expression

RI
is nearly unchanged (**p<0.01). B Examples of meningothelial or fibroblastic

SC
meningiomas stained for E-Cadherin, Slug, Zeb-1, Snail, and Twist. Pictures are

taken from stained tissue microarrays. Bars indicate 200m.

NU
Figure 2: Expression analyses of E-Cadherin and E-Cadherin-dependent
MA

transcription factors in meningiomas of different grades of malignancy. A Real-time

PCR analysis of mRNA expression levels show reduced E-Cadherin and Zo-1
ED

expression in WHO grade II/III meningiomas compared to WHO grade I

meningiomas, while Zeb-1 and Slug mRNA expression is significantly increased in

T
EP

aggressive WHO grade II/III meningiomas (**p<0.01, ***p<0.001). B

Immunoexpression of E-Cadherin, Slug, Zeb-1, Snail, and Twist in WHO grade II/III
C

meningiomas compared to WHO grade I tumors. Bars indicate 200m.

AC

Figure 3: Relation between E-Cadherin, E-Cadherin-dependent transcription factors,

and biological features of aggressive meningiomas. Recurrent meningiomas show

reduced mRNA expression of E-Cadherin and increased mRNA expression of Slug

(*p<0.05).
 ACCEPTED MANUSCRIPT

Figure 4: Relation between NF2 and E-Cadherin or E-Cadherin transcription factors.

A Significant positive correlation is found between NF2 mRNA expression and Zo-1,

E-Cadherin, and Twist mRNA, respectively. B In a subset of tumors characterized by

allelic losses (LOH) at the NF2 gene, E-Cadherin and Zo-1 are positively correlated

with NF2 expression (*p<0.05; **p<0.01)

PT
Figure 5: TGFbeta induces expression of E-Cadherin dependent transcription factors

RI
in meningioma cells. A Primary meningioma cells (left) or B IOMM-Lee meningioma

SC
cells (right) were treated with TGFbeta for 48 hours. The upper panel shows

morphologic changes of treated cells with slightly elongated tumors cells following
NU
TGFbeta treatment. The lower panel shows induction of Snail, Slug and Twist mRNA

in treated meningioma cells (*p<0.05). C Expression of E-Cadherin-dependent

MA

transcription factors in relation to NF2. Meningioma cells with loss of NF2/merlin

(Men-NF2) or control cells (Men-Co) were treated with TGFbeta for 48 hours. mRNA
ED

expression analyses show significant induction of E-Cadherin, Snail, Slug and Twist

in NF2-wild type cells. Cells with loss of NF2 have increased basal levels of E-
T
EP

Cadherin, Snail, and Slug which cannot be modulated by TGFbeta treatment. Twist

mRNA is increased in cells with NF2 loss, while TGFbeta treatment does not alter
C

expression levels.
AC

Acknowledgements:

The project was partly supported by grants from the DFG (grant # MA2530/6-1 and

MA2530/8-1), the Wilhelm Sander-Stiftung (grant #2014.092.1), and the Deutsche

Krebshilfe (grant #111853) (all to C.M.).

 ACCEPTED MANUSCRIPT

References

[1] Ostrom QT, Gittleman H, Farah P, Ondracek A, Chen Y, Wolinsky Y, et al. CBTRUS
statistical report: Primary brain and central nervous system tumors diagnosed in the United
States in 2006-2010. Neuro-oncology. 2013;15 Suppl 2(15):ii1-56.
[2] Mawrin C, Perry A. Pathological classification and molecular genetics of
meningiomas. J Neurooncol. 2010;99(3):379-91.
[3] Klaeboe L, Lonn S, Scheie D, Auvinen A, Christensen HC, Feychting M, et al.
Incidence of intracranial meningiomas in Denmark, Finland, Norway and Sweden, 1968-
1997. Int J Cancer. 2005;117(6):996-1001.
[4] Black PM. Meningiomas. Neurosurgery. 1993;32(4):643-57.

PT
[5] Perry A, Scheithauer BW, Stafford SL, Lohse CM, Wollan PC. "Malignancy" in
meningiomas: a clinicopathologic study of 116 patients, with grading implications. Cancer.
1999;85(9):2046-56.

RI
[6] Hasseleid BF, Meling TR, Ronning P, Scheie D, Helseth E. Surgery for convexity
meningioma: Simpson Grade I resection as the goal: clinical article. Journal of neurosurgery.

SC
2012;117(6):999-1006.
[7] Ruttledge MH, Sarrazin J, Rangaratnam S, Phelan CM, Twist E, Merel P, et al.
Evidence for the complete inactivation of the NF2 gene in the majority of sporadic
meningiomas. Nat Genet. 1994;6(2):180-4.
NU
[8] Kros J, de Greve K, van Tilborg A, Hop W, Pieterman H, Avezaat C, et al. NF2 status
of meningiomas is associated with tumour localization and histology. J Pathol.
2001;194(3):367-72.
MA

[9] McClatchey AI. Merlin and ERM proteins: unappreciated roles in cancer
development? Nat Rev Cancer. 2003;3(11):877-83.
[10] Kalamarides M, Niwa-Kawakita M, Leblois H, Abramowski V, Perricaudet M, Janin
A, et al. Nf2 gene inactivation in arachnoidal cells is rate-limiting for meningioma
ED

development in the mouse. Genes & development. 2002;16(9):1060-5.

[11] Kalamarides M, Stemmer-Rachamimov AO, Niwa-Kawakita M, Chareyre F,
Taranchon E, Han ZY, et al. Identification of a progenitor cell of origin capable of generating
diverse meningioma histological subtypes. Oncogene. 2011;30(20):2333-44.
T

[12] Brastianos PK, Horowitz PM, Santagata S, Jones RT, McKenna A, Getz G, et al.
EP

Genomic sequencing of meningiomas identifies oncogenic SMO and AKT1 mutations. Nat
Genet. 2013;45(3):285-9.
[13] Clark VE, Erson-Omay EZ, Serin A, Yin J, Cotney J, Ozduman K, et al. Genomic
C

analysis of non-NF2 meningiomas reveals mutations in TRAF7, KLF4, AKT1, and SMO.
Science. 2013;339(6123):1077-80.
AC

[14] Abedalthagafi M, Bi WL, Aizer AA, Merrill PH, Brewster R, Agarwalla PK, et al.
Oncogenic PI3K mutations are as common as AKT1 and SMO mutations in meningioma.
Neuro-oncology. 2016.
[15] Kalluri R, Neilson EG. Epithelial-mesenchymal transition and its implications for
fibrosis. J Clin Invest. 2003;112(12):1776-84.
[16] Kalluri R, Weinberg RA. The basics of epithelial-mesenchymal transition. J Clin
Invest. 2009;119(6):1420-8.
[17] Thiery JP. Epithelial-mesenchymal transitions in tumour progression. Nat Rev Cancer.
2002;2(6):442-54.
[18] Gunasinghe NP, Wells A, Thompson EW, Hugo HJ. Mesenchymal-epithelial
transition (MET) as a mechanism for metastatic colonisation in breast cancer. Cancer
Metastasis Rev. 2012;31(3-4):469-78.
 ACCEPTED MANUSCRIPT

[19] Wells A, Yates C, Shepard CR. E-cadherin as an indicator of mesenchymal to

epithelial reverting transitions during the metastatic seeding of disseminated carcinomas. Clin
Exp Metastasis. 2008;25(6):621-8.
[20] Roussos ET, Keckesova Z, Haley JD, Epstein DM, Weinberg RA, Condeelis JS.
AACR special conference on epithelial-mesenchymal transition and cancer progression and
treatment. Cancer Res. 2010;70(19):7360-4.
[21] Serrano-Gomez SJ, Maziveyi M, Alahari SK. Regulation of epithelial-mesenchymal
transition through epigenetic and post-translational modifications. Molecular Cancer.
2016;15:18.
[22] Brunner EC, Romeike BF, Jung M, Comtesse N, Meese E. Altered expression of beta-
catenin/E-cadherin in meningiomas. Histopathology. 2006;49(2):178-87.

PT
[23] Pecina-Slaus N, Cicvara-Pecina T, Kafka A. Epithelial-to-mesenchymal transition:
possible role in meningiomas. Frontiers in bioscience (Elite edition). 2012;4:889-96.
[24] Louis DN, Perry A, Reifenberger G, von Deimling A, Figarella-Branger D, Cavenee

RI
WK, et al. The 2016 World Health Organization Classification of Tumors of the Central
Nervous System: a summary. Acta Neuropathologica. 2016:1-18.

SC
[25] Striedinger K, Vandenberg SR, Baia GS, McDermott MW, Gutmann DH, Lal A. The
neurofibromatosis 2 tumor suppressor gene product, merlin, regulates human meningioma cell
growth by signaling through YAP. Neoplasia. 2008;10(11):1204-12.
[26] Lee WH. Characterization of a newly established malignant meningioma cell line of
NU
the human brain: IOMM-Lee. Neurosurgery. 1990;27(3):389-95; discussion 96.
[27] Yang L, Lin C, Liu Z-R. P68 RNA Helicase Mediates PDGF-Induced Epithelial
Mesenchymal Transition by Displacing Axin from -Catenin. Cell. 2006;127(1):139-55.
MA

[28] Pachow D, Andrae N, Kliese N, Angenstein F, Stork O, Wilisch-Neumann A, et al.

mTORC1 inhibitors suppress meningioma growth in mouse models. Clin Cancer Res.
2013;19(5):1180-9.
[29] Petermann A, Haase D, Wetzel A, Balavenkatraman KK, Tenev T, Guhrs KH, et al.
ED

Loss of the Protein-Tyrosine Phosphatase DEP-1/PTPRJ Drives Meningioma Cell Motility.

Brain pathology (Zurich, Switzerland). 2011;21(4).
[30] James MF, Lelke JM, Maccollin M, Plotkin SR, Stemmer-Rachamimov AO, Ramesh
V, et al. Modeling NF2 with human arachnoidal and meningioma cell culture systems: NF2
T

silencing reflects the benign character of tumor growth. Neurobiol Dis. 2008;29(2):278-92.
EP

[31] Hartsock A, Nelson WJ. Adherens and tight junctions: Structure, function and
connections to the actin cytoskeleton. Biochimica et Biophysica Acta (BBA) -
Biomembranes. 2008;1778(3):660-9.
C

[32] Van Itallie CM, Tietgens AJ, Anderson JM. Visualizing the dynamic coupling of
claudin strands to the actin cytoskeleton through ZO-1. Mol Biol Cell. 2017;28(4):524-34.
AC

[33] Fevre-Montange M, Champier J, Durand A, Wierinckx A, Honnorat J, Guyotat J, et al.

Microarray gene expression profiling in meningiomas: differential expression according to
grade or histopathological subtype. Int J Oncol. 2009;35(6):1395-407.
[34] Rooprai HK, Martin AJ, King A, Appadu UD, Jones H, Selway RP, et al. Comparative
gene expression profiling of ADAMs, MMPs, TIMPs, EMMPRIN, EGF-R and VEGFA in
low grade meningioma. Int J Oncol. 2016;49(6):2309-18.
[35] Keller A, Ludwig N, Backes C, Romeike BF, Comtesse N, Henn W, et al. Genome
wide expression profiling identifies specific deregulated pathways in meningioma. Int J
Cancer. 2009;124(2):346-51.
[36] Aarhus M, Bruland O, Bredholt G, Lybaek H, Husebye ES, Krossnes BK, et al.
Microarray analysis reveals down-regulation of the tumour suppressor gene WWOX and up-
regulation of the oncogene TYMS in intracranial sporadic meningiomas. J Neurooncol.
2008;88(3):251-9.
 ACCEPTED MANUSCRIPT

[37] Figarella-Branger D, Roche PH, Daniel L, Dufour H, Bianco N, Pellissier JF. Cell-
adhesion molecules in human meningiomas: correlation with clinical and morphological data.
Neuropathol Appl Neurobiol. 1997;23(2):113-22.
[38] Figarella-Branger D, Pellissier JF, Bouillot P, Bianco N, Mayan M, Grisoli F, et al.
Expression of neural cell-adhesion molecule isoforms and epithelial cadherin adhesion
molecules in 47 human meningiomas: correlation with clinical and morphological data.
Modern pathology : an official journal of the United States and Canadian Academy of
Pathology, Inc. 1994;7(7):752-61.
[39] Shimada S, Ishizawa K, Hirose T. Expression of E-cadherin and catenins in
meningioma: ubiquitous expression and its irrelevance to malignancy. Pathology
international. 2005;55(1):1-7.

PT
[40] Utsuki S, Oka H, Sato Y, Kawano N, Tsuchiya B, Kobayashi I, et al. Invasive
meningioma is associated with a low expression of E-cadherin and beta-catenin. Clin
Neuropathol. 2005;24(1):8-12.

RI
[41] Pecina-Slaus N, Nikuseva Martic T, Deak AJ, Zeljko M, Hrascan R, Tomas D, et al.
Genetic and protein changes of E-cadherin in meningiomas. J Cancer Res Clin Oncol.

SC
2010;136(5):695-702.
[42] Zhou K, Wang G, Wang Y, Jin H, Yang S, Liu C. The Potential Involvement of E-
cadherin and beta-catenins in Meningioma. PLoS One. 2010;5(6):e11231.
[43] Nagaishi M, Nobusawa S, Tanaka Y, Ikota H, Yokoo H, Nakazato Y. Slug, Twist, and
NU
E-Cadherin as Immunohistochemical Biomarkers in Meningeal Tumors. PLoS ONE.
2012;7(9):e46053.
[44] Soini Y, Rauramaa T, Alafuzoff I, Sandell PJ, Karja V. Claudins 1, 11 and twist in
MA

meningiomas. Histopathology. 2010;56(6):821-4.

[45] Champeaux C, Wilson E, Shieff C, Khan AA, Thorne L. WHO grade II meningioma:
a retrospective study for outcome and prognostic factor assessment. Journal of Neuro-
Oncology. 2016;129(2):337-45.
ED

[46] Hwang WL, Marciscano AE, Niemierko A, Kim DW, Stemmer-Rachamimov AO,
Curry WT, et al. Imaging and extent of surgical resection predict risk of meningioma
recurrence better than WHO histopathological grade. Neuro-oncology. 2016;18(6):863-72.
[47] Hsu HP, Shan YS, Jin YT, Lai MD, Lin PW. Loss of E-cadherin and beta-catenin is
T

correlated with poor prognosis of ampullary neoplasms. Journal of surgical oncology.

EP

2010;101(5):356-62.
[48] Huszar M, Pfeifer M, Schirmer U, Kiefel H, Konecny GE, Ben-Arie A, et al. Up-
regulation of L1CAM is linked to loss of hormone receptors and E-cadherin in aggressive
C

subtypes of endometrial carcinomas. J Pathol. 2010;220(5):551-61.

[49] Iso Y, Sawada T, Okada T, Kubota K. Loss of E-cadherin mRNA and gain of
AC

osteopontin mRNA are useful markers for detecting early recurrence of HCV-related
hepatocellular carcinoma. Journal of surgical oncology. 2005;92(4):304-11.
[50] Pulkka OP, Nilsson B, Sarlomo-Rikala M, Reichardt P, Eriksson M, Hall KS, et al.
SLUG transcription factor: a pro-survival and prognostic factor in gastrointestinal stromal
tumour. Br J Cancer. 2017.
[51] Kihara A, Wakana K, Kubota T, Kitagawa M. SLUG expression is an indicator of
tumour recurrence in high-grade endometrial carcinomas. Histopathology. 2016;69(3):374-82.
[52] Cappellesso R, Marioni G, Crescenzi M, Giacomelli L, Guzzardo V, Mussato A, et al.
The prognostic role of the epithelial-mesenchymal transition markers E-cadherin and Slug in
laryngeal squamous cell carcinoma. Histopathology. 2015;67(4):491-500.
[53] Murray LB, Lau YK, Yu Q. Merlin is a negative regulator of human melanoma
growth. PLoS One. 2012;7(8):e43295.
 ACCEPTED MANUSCRIPT

[54] Morrow KA, Das S, Metge BJ, Ye K, Mulekar MS, Tucker JA, et al. Loss of tumor
suppressor Merlin in advanced breast cancer is due to post-translational regulation. J Biol
Chem. 2011;286(46):40376-85.
[55] Nakayama A, Ninomiya I, Harada S, Tsukada T, Okamoto K, Nakanuma S, et al.
Metformin inhibits the radiation-induced invasive phenotype of esophageal squamous cell
carcinoma. Int J Oncol. 2016;49(5):1890-8.
[56] Vand-Rajabpour F, Sadeghipour N, Saee-Rad S, Fathi H, Noormohammadpour P,
Yaseri M, et al. Differential BMI1, TWIST1, SNAI2 mRNA expression pattern correlation
with malignancy type in a spectrum of common cutaneous malignancies: basal cell
carcinoma, squamous cell carcinoma, and melanoma. Clinical & translational oncology :
official publication of the Federation of Spanish Oncology Societies and of the National

PT
Cancer Institute of Mexico. 2017;19(4):489-97.
[57] Kucuksayan H, Ozes ON, Akca H. Downregulation of SATB2 is critical for induction
of epithelial-to-mesenchymal transition and invasion of NSCLC cells. Lung cancer

RI
(Amsterdam, Netherlands). 2016;98:122-9.
[58] Wang Y, Liu J, Ying X, Lin PC, Zhou BP. Twist-mediated Epithelial-mesenchymal

SC
Transition Promotes Breast Tumor Cell Invasion via Inhibition of Hippo Pathway. Scientific
reports. 2016;6:24606.
[59] Ruttledge MH, Xie YG, Han FY, Peyrard M, Collins VP, Nordenskjold M, et al.
Deletions on chromosome 22 in sporadic meningioma. Genes, chromosomes & cancer.
NU
1994;10(2):122-30.
[60] Petrilli AM, Fernandez-Valle C. Role of Merlin/NF2 inactivation in tumor biology.
Oncogene. 2016;35(5):537-48.
MA

[61] Rangwala R, Banine F, Borg JP, Sherman LS. Erbin regulates mitogen-activated
protein (MAP) kinase activation and MAP kinase-dependent interactions between Merlin and
adherens junction protein complexes in Schwann cells. J Biol Chem. 2005;280(12):11790-7.
[62] Linsler S, Kraemer D, Driess C, Oertel J, Kammers K, Rahnenfuhrer J, et al.
ED

Molecular biological determinations of meningioma progression and recurrence. PLoS One.

2014;9(4):e94987.
[63] Lee JY, Finkelstein S, Hamilton RL, Rekha R, King JT, Jr., Omalu B. Loss of
heterozygosity analysis of benign, atypical, and anaplastic meningiomas. Neurosurgery.
T

2004;55(5):1163-73.
EP

[64] Kim JH, Kim IS, Kwon SY, Jang BC, Suh SI, Shin DH, et al. Mutational analysis of
the NF2 gene in sporadic meningiomas by denaturing high-performance liquid
chromatography. Int J Mol Med. 2006;18(1):27-32.
C

[65] Hansson CM, Buckley PG, Grigelioniene G, Piotrowski A, Hellstrom AR,

Mantripragada K, et al. Comprehensive genetic and epigenetic analysis of sporadic
AC

meningioma for macro-mutations on 22q and micro-mutations within the NF2 locus. BMC
Genomics. 2007;8:16.
[66] Cho JH, Lee SJ, Oh AY, Yoon MH, Woo TG, Park BJ. NF2 blocks Snail-mediated
p53 suppression in mesothelioma. Oncotarget. 2015;6(12):10073-85.
 ACCEPTED MANUSCRIPT

PT
RI
SC
NU
MA
T ED
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
SC
NU
MA
T ED
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
SC
NU
MA
T ED
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
SC
NU
MA
T ED
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
SC
NU
MA
T ED
C EP
AC

Table 1: Immunoexpression of E-Cadherin and E-Cadherin-dependent

transcription factors in WHO-grade I meningiomas

WHO grade I
Meningothelial Fibroblastic/Transitional

E-Cadherin 10/10 (100%) 12/14 (86%)

Slug 10/10 (100%) 14/14 (100%)
 ACCEPTED MANUSCRIPT

Zeb-1 7/9 (78%) 9/13 (69%)

Snail 9/10 (90%) 12/14 (86%)
Twist 8/9 (89%) 6/6 (100%)

Table 2: Immunoexpression of E-Cadherin and E-Cadherin-dependent

transcription factors in WHO-grade II and III meningiomas compared to
WHO-grade I meningiomas

WHO WHO WHO

PT
grade I grade II grade III

E-Cadherin 29/31 (94%) 11/12 (92%) 5/9 (56%)

RI
Slug 13/13 (100%) 8/8 (100%) 2/ 4 (50%)
Zeb-1 21/28 (75%) 11/11 (100%) 6/6 (100%)

SC
Snail 21/28 (75%) 9/12 (75%) 5/7 (71%)
Twist 15/17 (88%) 11/11 (100%)
NU 6/6 (100%)
MA
T ED
C EP
AC
 ACCEPTED MANUSCRIPT

Highlights
no hints for EMT (epithelial-mesenchymal transition) is found by comparing
WHO grade I meningioma subtypes (meningothelial versus
fibroblastic/transitional)
aggressive meningiomas of WHO grade II or III show altered expression levels
(loss of E-Cadherin and Zo-1, increased expression of Zeb-1 and Slug)
indicating molecular features of EMT

PT
E-cadherin is reduced in aggressive and recurrent meningioma
E-cadherin and Zo-1 expression levels are correlated with NF2 expression
TGFbeta treatment of meningioma cells induces morphological changes and

RI
altered expression of EMT factors

SC
NU
MA
T ED
C EP
AC