University of ZurichZurich Open Repository and Archive
Winterthurerstr. 190
CH-8057 Zurich
http://www.zora.uzh.ch
Year: 2008
Evidence for a role of epithelial mesenchymal transition duringpathogenesis of fistulae in Crohn's disease
Bataille, F; Rohrmeier, C; Bates, R; Weber, A; Rieder, F; Brenmoehl, J; Strauch, U;Farkas, S; Fürst, A; Hofstädter, F; Schölmerich, J; Herfarth, H; Rogler, G
Bataille, F; Rohrmeier, C; Bates, R; Weber, A; Rieder, F; Brenmoehl, J; Strauch, U; Farkas, S; Fürst, A; Hofstädter,F; Schölmerich, J; Herfarth, H; Rogler, G (2008). Evidence for a role of epithelial mesenchymal transition duringpathogenesis of fistulae in Crohn's disease. Inflammatory Bowel Diseases, 14(11):1514-1527.Postprint available at:http://www.zora.uzh.ch
Posted at the Zurich Open Repository and Archive, University of Zurich.http://www.zora.uzh.ch
Originally published at:Inflammatory Bowel Diseases 2008, 14(11):1514-1527.
Bataille, F; Rohrmeier, C; Bates, R; Weber, A; Rieder, F; Brenmoehl, J; Strauch, U; Farkas, S; Fürst, A; Hofstädter,F; Schölmerich, J; Herfarth, H; Rogler, G (2008). Evidence for a role of epithelial mesenchymal transition duringpathogenesis of fistulae in Crohn's disease. Inflammatory Bowel Diseases, 14(11):1514-1527.Postprint available at:http://www.zora.uzh.ch
Posted at the Zurich Open Repository and Archive, University of Zurich.http://www.zora.uzh.ch
Originally published at:Inflammatory Bowel Diseases 2008, 14(11):1514-1527.
Evidence for a role of epithelial mesenchymal transition duringpathogenesis of fistulae in Crohn's disease
Abstract
BACKGROUND: The pathogenesis of fistulae in Crohn's disease (CD) patients is barely understood.We recently showed that more than two-thirds of CD fistulae are covered with flat, mesenchymal-likecells (transitional cells [TC]) forming a patchy basement membrane. Epithelial-to-mesenchymaltransition (EMT) is a process of reprogramming epithelial cells, allowing them to migrate moreeffectively and giving epithelial cells an "invasive" potential. EMT has been suggested to be crucial infibrosis found in different tissues and diseases. We therefore investigated whether EMT could beinvolved in the pathogenesis of fistulae formation in CD. METHODS: In all, 18 perianal fistulae, 2enteroenteric, and 1 enterovesical fistulae from 17 CD patients were analyzed. In addition 2 perianalfistulae of non-CD patients were studied. Hematoxylin and eosin staining, immunohistochemistry forthe expression of cytokeratins 8 and 20, beta6-integrin, E-cadherin, beta-catenin, vimentin, andTGF-beta1 and 2 were performed according to standard techniques. RESULTS: The TC coveringperianal or enteroenteric fistulae were strongly positive for cytokeratins 8 and 20 but negative forvimentin, indicating their epithelial origin. beta6-Integrin and TGF-beta had the highest stainingintensities in the transitional zone between the epithelium and the TC. Expression of junctional proteinssuch as E-cadherin was reduced in TC as compared to regular fistulae epithelium. In addition, atranslocation of beta-catenin from the membrane to the cytoplasm was observed. CONCLUSIONS: Ourdata for the first time indicate an expression pattern of epithelial and mesenchymal markers in TCassociated with fistulae formation that is characteristic for EMT. Studying the pathways of EMT duringintestinal fistulae formation may help to develop new therapeutic strategies.
Evidence for a role of epithelial mesenchymal transition during
pathogenesis of fistulae in Crohn’s disease
Frauke Bataille*, Christian Rohrmeier+, Richard Batesx, Achim Weber$, Florian
Rieder+, Julia Brenmoehl+, Ulrike Strauch+, Stefan Farkas§, Alois Fürst#, Ferdinand
Hofstädter*, Jürgen Schölmerich+, Hans Herfarth+b, Gerhard Rogler+a
* Institute of Pathology + Department of Internal Medicine I § Department of Surgery
University of Regensburg 93042 Regensburg Germany # Department of Surgery Krankenhaus St. Josef 93042 Regensburg Germany X Department of Cancer Biology University of Massachusetts Medical School Lazare Research Building - 470Y 364 Plantation Street Worcester MA 01605 USA $ Department of Pathology, Institute of Surgical Pathology
University Hospital of Zürich CH-8091 Zürich Switzerland
a present address:
Clinic of Gastroenterology and Hepatology Department of Internal Medicine
University Hospital of Zürich CH-8091 Zürich Switzerland
b present address: Department of Medicine Division of Gastroenterology and Hepatology University of North Carolina at Chapel Hill Chapel Hill, NC 27599-7555 USA
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 2
Word count: 3399 Key words: Crohn’s disease, fistulae, epithelial mesenchymal transition Short title: EMT and fistula formation in CD Address for correspondence: Gerhard Rogler, MD, PhD Clinic of Gastroenterology and Hepatology Department of Internal Medicine University Hospital of Zürich CH-8091 Zürich Switzerland Tel. +41-44-255-9477 Fax. +41-44-255-4503 E-mail: [email protected]
Word count: 3596 words
No conflicts of interest exist.
Abbreviations: 5-ASA: 5-aminosalicylic acid; CD: Crohn´s disease; DAB:
diaminobenzidine chromogen; EMT: epithelial-to-mesenchymal transition; IBD:
inflammatory bowel disease; H&E: haematoxylin and eosin; TC: transitional cells
Acknowledgements:
This work was supported by the Bundesministerium für Bildung und Forschung,
BMBF (Kompetenznetz CED), and the Deutsche Forschungsgemeinschaft (DFG
Ro1236/15-1).
We thank Paul Weinreb and Shelia Violette at Biogen-Idec Inc. Cambridge MA, USA,
for providing the integrin beta6 antibody.
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 3
Abstract
BACKGROUND: The pathogenesis of fistulae in Crohn’s disease (CD) patients is
barely understood. We could recently show that more than 2/3 of CD fistulae are
covered with flat, mesenchymal-like cells (transitional cells; TC) forming a patchy
basement membrane. Epithelial-to-mesenchymal transition (EMT) is a process of re-
programming epithelial cells allowing them to migrate more effectively and giving
epithelial cells an “invasive” potential. EMT has been suggested to be crucial in
fibrosis found in different tissues and diseases. We therefore investigated, whether
EMT could be involved in the pathogenesis of fistulae formation in CD.
METHODS: 18 perianal fistulae, two entero-enteric and one entero-vesical fistulae
from 17 CD patients were analysed. In addition 2 perianal fistulae of non-CD patients
were studied. H&E staining, immunohistochemistry for the expression of cytokeratins
8 and 20, β6-integrin, E-cadherin, β-catenin, vimentin and TGF-β1 and 2 were
performed according to standard techniques.
RESULTS: The TC covering perianal or entero-enteric fistulae were strongly positive
for cytokeratins 8 and 20, but negative for vimentin indicating their epithelial origin.
ß6-integrin and TGF-β had highest staining intensities in the transitional zone
between the epithelium and the TC. Expression of junctional proteins such as E-
cadherin was reduced in TC as compared to regular fistulae epithelium. In addition a
translocation of β-catenin from the membrane to the cytoplasm was observed.
CONCLUSION: Our data for the first time indicate an expression pattern of
epithelial and mesenchymal markers in TC associated with fistulae formation which is
characteristic for EMT. Studying the pathways of EMT during intestinal fistulae
formation may help to develop new therapeutic strategies.
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 4
Introduction
Fistulae are a frequent and serious complication of patients with Crohn’s disease
(CD). The cumulative incidence in population based studies ranges from 17 up to 50
% 1-4. The clinical management of CD-fistulae remains a major therapeutic challenge.
Medical treatment is mainly based on antibiotics and immunosuppressants such as
azathioprine, cyclosporine A, tacrolimus or infliximab 3, 5-8. However, an initially
successful closure of fistulae can only be maintained in about 30 to 40 % of patients
after 12 months. Despite the clinical importance of the problem the pathogenesis of
fistulae-formation is poorly understood and investigations focussed on the etiology of
fistulae are sparse. A better knowledge of the pathophysiology of fistulae-formation
would be crucial for the development of new and more effective therapeutic options.
It is generally assumed that tissue destruction is the first step in the pathogenesis of
fistula formation. To gain further insights into fistula formation we
histomorphologically characterized fistulae resected from CD patients and patients
without the diagnosis of inflammatory bowel disease (IBD) 9. Ninety-seven fistulae of
an unselected patient cohort of 78 patients were investigated. Eighty-four fistula
specimens were derived from 67 CD patients. The results were compared to 13
fistulae of 11 patients without IBD. Histologically all fistulae showed a central fissure
penetrating through the lamina propria and the muscularis mucosae into the deeper
layers of the underlying tissue. All fistulae were surrounded by a granulation tissue
with histiocytes and a tight network of capillaries. In patients with CD, the interior
wall of the studied fistulae usually was infiltrated by CD45RO positive T cells,
followed by a small band of CD68 positive macrophages. At the outer fistulae wall,
there was a dense infiltrate of CD20 positive B cells.
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 5
31 % of the „control fistulae“ (i.e. non-IBD) and 27 % of the CD-fistulae showed a
layer with easily recognizable epithelial cells. Those cells showed tight junctions and
a basement membrane. Interestingly the „non-epithelialised“ CD-fistulae - present in
more than 2/3 of all cases - were covered by a thin layer of myofibroblast-like
“transitional cells” (TC) with gap junctions. These cells also formed a basement
membrane-like structure. In a subgroup of specimens a region could be identified in
which the mucosal epithelial cells seemed to continuously transform into the TC-layer
9.
Under physiologic conditions fibroblasts are recruited to the sites of tissue injury 10,
11. However, we recently showed that the migratory potential of CD mucosal
fibroblasts is reduced compared to controls 12-14 . An even stronger reduction of
migratory ability was found in fibroblasts derived from CD fistula tissue (unpublished
data) indicating a disturbance in wound healing. Epithelial cell migration is an
another hallmark of the attempt of the intestinal mucosa to rapidly close defects of
the intestinal barrier 15-21. It is induced by a number of different growth factors 20.
However, epithelial cells migrate slowly. If a superficial tissue defect of the mucosa
cannot be closed by intestinal fibroblasts epithelial cells might migrate towards the
defect.
Recent evidence suggests that epithelial cells do not represent a final and irreversible
state of differentiation. Conversion of an epithelial cell to a mesenchymal cell has
been shown to be critical during embryogenesis and is a structural feature of organ
development. Current interest in this process – called epithelial–to-mesenchymal
transition (EMT) - has increased due to its involvement in adult pathologies 22, 23. It
has been proposed that epithelial tumours undergo EMT, facilitating their invasion.
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 6
EMT is also an essential component of tissue remodelling, and wound repair 23-27.
During this transition, the epithelial cells - characterized by strong cell-cell junctions
and polarity - are replaced by a mesenchymal phenotype, with reduced cell-cell
adhesions, a fibroblast morphology and function 22, 23, 28. There are several molecular
markers for the detection of EMT in vivo 22, 23. These include decreased E-cadherin
and β-catenin expression, and increased expression of β6 integrin 29, 30. During EMT
β-catenin translocates to the nucleus. TGF-β has been shown to be an inducer of
EMT.
We hypothesized that EMT could be important in the pathophysiology of fistula
development in CD. Adhesions between mesenchymal cells are less tight than
between epithelial cells facilitating an increased migratory capacity. Epithelial cells
undergoing EMT have an increased migratory potential independently of cell-cell
contacts 22. To test our hypothesis we used fistula specimens of patients with CD for
immunohistochemical staining of EMT markers. In addition we established primary
fibroblast cultures from CD patients with and without fistulae. Here we present first
time evidence for EMT in intestinal wound healing and the pathogenesis of fistulae in
CD.
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 7
Materials and Methods
Patients
18 entero-cutaneous fistula specimens from 15 patients were examined
retrospectively. In addition we studied 2 perianal fistulae from patients without CD.
This series was obtained from unselected cases at the Institute of Pathology of the
University of Regensburg. Specimens had been surgically removed between August
1993 and May 2003. 14 CD fistula specimens showed flat TC covering the fistula
walls, 4 fistulae had a squamous cell layer. The two non-CD fistulae also showed flat
TC cells covering the fistula tract. In addition, two entero-enteric and one entero-
vesical fistulae from 2 patients were investigated. Those two patients underwent
surgery in December 2008 at the University Hospital of Zurich. Those fistulae had a
columnar epithelium at the opening and flat mesenchymal like cells along the fistula
tracts.
The diagnosis of CD was based on established clinical, endoscopic, histological and
radiological parameters 31, 32. Details on fistula distribution, patient age at
presentation and patient gender are shown in table 1.
The degree of inflammation was graded microscopically by determination of the
inflammatory infiltrate of neutrophils, eosinophils and lymphocytes: 0 = no
inflammation, 1 = low degree of inflammation, 2 = severe inflammation.
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 8
This study was approved by the Ethics Committee of the University of Regensburg
and performed according to the declaration of Helsinki.
Specimen preparation
Tissue specimens were fixed in 4% buffered formalin for at least 24 h and embedded
in paraffin. Sections of approximate 2 – 3 µm thickness were cut from tissue blocks
and stained with H&E (haematoxylin and eosin) according to standard protocols.
Immunohistochemistry
Immunohistochemical studies for the expression of cytokeratins 8 and 20, β6-
integrin, TGF-β, E-cadherin, β-catenin and vimentin were performed on 17 fistulae of
15 patients with CD by using an avidin-biotin peroxidase method with
diaminobenzidine (DAB) chromogen. After antigen retrieval (microwave treatment of
formalin-fixed, paraffin-embedded 2-3 µm tissue sections for 40 min at 240 W in
citrate buffer, pH 6.0) immunohistochemistry was carried out in a NEXES
immunostainer (Ventana Medical System, Tucson, AZ) according to the
manufacturer´s instructions. As primary antibodies mouse monoclonal and rabbit
polyclonal antibodies were used (ß-catenin [Santa Cruz, SC-7963, mouse monoclonal
(E-5), dilution 1:50]; E-cadherin [Santa Cruz, SC-8426, mouse monoclonal (G-10),
dilution 1:75]; TGF-ß2 [Santa Cruz (SC-90), rabbit polyclonal, dilution 1:30]; TGF-ß1
[Acris (DM1047), mouse (TB21) monoclonal, dilution 1:200]; CK8 [Dako M0631),
mouse (35ßH11) monoclonal, dilution 1:50], CK20 [Progen (61026), mouse
(ITKS20.8) monoclonal, dilution 1:10]). After incubation for 24 min at 37° C, the
slides where rinsed in PBS, and incubated with the secondary antibody (rabbit-anti-
mouse; Ventana Medical System, 1:500 dilution in PBS) for 2 h at room temperature.
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 9
Antibody binding was visualized with 0.05% DAB (Ventana Medical System), and
0.01% hydrogen peroxide. The material was rinsed in PBS and counterstained with
haematoxylin. For integrin β6 staining, antigen retrieval was performed using
incubation in pepsin solution [Zymed Laboratories Inc.] at 37°C, prior to overnight
incubation with the monoclonal antibody 2G2 (Biogen-Idec, 2 ug/ml).
Three different pathologists (F. Bataille, R. Bates, A Weber) evaluated the specimens
according to subjective criteria and graded the expression into three different
categories (0 = no expression; 1 = weak expression and 2 = strong expression). The
results of this evaluation were consistent.
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 10
Results
We evaluated differences in the expression of specific antigens between the cells of
origin (intestinal columnar epithelium or squamous cell epithelium) and the TC by
immunohistochemistry. 14 perianal fistulae with intestinal epithelium, two entero-
enteric, one entero-vesical fistulae, four perianal fistulae with squamous cell
epithelium from CD patients and two perianal fistulae from non-IBD patients were
investigated. The expression of the respective antigen staining intensity was semi-
quantitatively rated as “absent” (0), „weak“ (1) or „strong“ (2). A median was
calculated and used for the comparison of the two fistula areas.
E-Cadherin
During the process of EMT a loss or reduction of E-cadherin expression is well known
to occur 33-37. E-cadherin is involved in homophilic interactions between epithelial
cells and is necessary for the formation of zonulae adherentes. In the normal
intestinal epithelial cells E-cadherin staining was found at the lateral cell membrane
at the cell-cell contact sites. In the fistulae lining cells a decrease in the intensity of
staining was found (Figure 1). In 64.3% of the specimens additionally a
redistribution of the membranous E-cadherin was found giving the staining at the cell
wall a scattered appearance. Comparable observations were made in fistulae
containing a squamous cell layer. A decrease in the median from weak staining to
virtually absent staining was noted. With higher magnification it became evident that
the pattern of E-cadherin expression in the fistula lining cells was dependent on the
distance of the cells from the origin of the fistula in the gut lumen. Figure 2 clearly
shows that cells closer to the mucosa (Figure 2A and B) have a more regular
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 11
expression of E-cadherin than cells from deeper areas of the fistula (Figure 2C and
D). In Figure 2E this change in localization of E-cadherin can be seen in a typical
“transition zone”.
β-Catenin
The process of EMT is characterized by a re-distribution of β-catenin 38-42. Initially
during EMT β-catenin expression is increased, however, the protein is no longer
membrane associated but localized in the cytoplasm or even in the nucleus. In the
later stages of EMT the synthesis of β-catenin is reduced. As expected we found a
strong staining of the lateral cell membrane of the columnar normal mucosal
intestinal epithelial cells (Figure 3 A, B and D). In the TC a diffuse and much
weaker expression was found, which was localized cytoplasmatically (Figure 3 B, C
and D). Comparable findings were obtained in the four fistulae with a squamous cell
epithelial layer (Figure 3 E). The staining was much stronger in the regular epithelia
compared to TC (Figure 3 G and H).
Cytokeratin 8 and 20
Cytokeratin 8 (CK 8) is an intermediate filament and part of the cytoskeleton of
intestinal epithelial cells 43. We used this typical epithelial marker to evaluate its
expression in the flat TC covering the deeper areas of fistulae. In the 14 fistulae
containing intestinal epithelial cells and TC no difference between the two cell
populations with respect to CK 8 expression was observed (Figure 4). In the four
specimens that contained a squamous cell layer a difference was noted. The
squamous epithelial cells did not express CK 8 whereas in the fistula lining cells a
median staining intensity of 1.5 was found.
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 12
Comparable results were found for cytokeratin 20 (CK 20). Intestinal epithelial cells
as well as TC were strongly positive for CK 20 (data not shown).
Vimentin
Vimentin is an intermediate filament and one major component of the cytoskeleton.
It is abundantly expressed in fibroblasts and endothelial cells 44. We used this typical
mesenchymal marker to evaluate its expression in the flat TC covering the deeper
areas of fistulae. In our staining all epithelia – columnar and squamous epithelia as
well as TC – were negative for vimentin (data not shown).
Integrin β6
The β6 integrin-chain is restricted to epithelial cells and expressed during embryonic
development and organogenesis. Re-induction of αvβ6 indicates an important role of
this receptor during intestinal EMT. Six fistulae were stained with a specific antibody
for β6 integrin. A clear difference in the intensity of staining between normal
epithelial and fistulae covering cells could be noted. As expected virtually no protein
expression of β6 integrin could be found in normal intestinal epithelial cells.
Especially in the „transitional zones“, where a stepwise flattening of the epithelial
cells occurs, a strong staining pattern (Figure 5) was detected.
TGFβ1 and TGFβ2
TGF is known to play a major role in the induction of EMT and its induction is linked
to β6 integrin-expression 30, 45-53. Immunohistochemistry for TGFβ1 showed a weak
staining in the fistula lining cells whereas normal mucosal intestinal epithelial cells
were negative for TGFβ1. This indicates an induction of TGFβ1 expression in the
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 13
fistula tract. A clear difference of the medians of expression was found. In specimens
containing squamous cells a comparable observation was made (data not shown).
Expression of TGFβ2 was weak in normal intestinal epithelial cells. The expression in
the TC was stronger (Figure 6). Similar observations were made in the four
specimens containing squamous cells.
Entero-enteric fistulae
To investigate whether the described features of EMT are generally found in
“mesenchymal cell” covered CD fistula of specific for perianal fistula we investigated
two enter-enteric fistulae obtained by surgical resection. The first patient suffered
from ileo-sigmoidal and ileo-vesical fistulae and the second patient from ileo-ileal
fistula formation. The Immunohistochemical results in these two fistula patients were
identical to entero-cutaneous fistulae described above (Figure 7). Cytokeratin 8 and
cytokeratin 20 were found to be expressed in all fistula lining cells independent of
whether they had epithelial or mesenchymal morphology (Figure 7 B). E-Cadherin
expression clearly was associated to cell membranes and cell-cell contacts in the
beginning of the entero-enteric fistulae tracts. However in the transition zone E-
cadherin expression clearly was reduced or almost absent (Figure 7C, D and E).
Whereas β-catenin expression was located at cell borders and cell membranes at the
luminal end of the fistulae (Figure 7 F and G) in the transition zone β-catenin was
localized in the cytoplasm or in the nucleus (Figure 7H) as described to be typical
for EMT. These data are indicative for a role of EMT in the formation of entero-
enteric and entero-vesical fistulae.
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 14
Non-CD fistulae.
Two specimens from patients without CD were investigated by
immunohistochemistry with the same antibodies as used above (with the exception
of β6 integrin). Very similar to the findings indicated above we found evidence for
EMT in these specimens (data not shown).
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 15
Discussion
In the present manuscript we show clear first time evidence for EMT in the epithelial
transitional zone close to the luminal origin of perianal, entero-cutaneous, entero-
enteric as well as entero-vesical fistulae tracts in CD patients. We provide data that
the flat TC, that used to be morphologically classified as fibroblasts retain epithelial
markers such as CK 8 and CK 20, are vimentin negative, show a redistribution of E-
cadherin and β-catenin as described earlier to be typical for EMT and illustrate a re-
induction of markers of early morphogenesis such as β6-integrin. TGF-β is highly
expressed in the transitional zone between normal intestinal epithelium and fistula
lining “mesenchymal cells” indicating a potential role during fistula-EMT. We
therefore hypothesize, that CD fistula form via the TGF driven initiation of EMT which
so far has only been described in cancer metastasis and tissue fibrosis.
What may be the reason for epithelial cells to undergo EMT during fistula formation?
Normally regularly spaced cell–cell junctions and adhesions between neighbouring
epithelial cells hold them together in a tight formation preventing the movement of
individual cells away from the epithelial monolayer. In contrast, mesenchymal cells
lack tight intercellular adhesions making them much more mobile followed by an
increased migratory capacity. Whereas epithelial cells usually move as a sheet en
block, mesenchymal migration is much more dynamic. Cells move individually and
can leave part of the trailing region behind 54.
Turning an epithelial cell into a mesenchymal cell requires alterations in morphology,
cellular architecture, downregulation of cell-cell adhesion systems, and migratory
capacity. Commonly used molecular markers for EMT include β6-integrin, vimentin
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 16
and cytokeratin expression, cadherins and nuclear localization of β-catenin.
A defining feature of EMT is a reduction in E-cadherin levels (E for epithelial
cadherin). Cadherins are transmembrane proteins whose homotypic interaction
between neighbouring cells creates adherens junctions 26, 55. Alteration of cadherin-
based adhesion has a key role in modulating development, organogenesis or
turnover of rapidly growing tissues. At the cell membrane, cadherin proteins are
found as homodimers tethered to the actin cytoskeleton by a multiprotein complex
that includes α-catenin, β-catenin, and p120-catenin. The presence of E-cadherin in
epithelial cells allows for greater cell–cell adhesive strength compared with that of
the N-cadherin–expressing mesenchyme.
β-catenin is the direct physiological link between cadherins and the actin
cytoskelleton at the adherens junctions. In addition β-catenin has a role as a signal-
transducing molecule influencing the state of the actin cytoskeleton 56. Catenins
directly modulate the adhesive state of the cadherin extracellular homophilic
adhesive binding domain and therefore control epithelial adhesion and junctional
formation in a way similar to integrins 55, 57. The nuclear translocation of β-catenin
from the cytoplasm is considered a key molecular step in EMT 23. E-cadherin as well
as β-catenin was down regulated in TC compared to mucosal epithelial cells and
squamous cells. In addition in 2/3 of the specimens a redistribution of E-cadherin
with a scattered appearance was detected. β-catenin in TC showed a diffuse
cytoplasmatic localization in comparison to localization to the lateral cytoplasmic cell
membrane in squamous epithelial or mucosal epithelial cells. This is clear evidence
for a disaggregation of the epithelial units and a reshaping for movement.
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 17
The integrin complexes represent the major receptors that mediate attachment to
the extracellular matrix, with ligand occupancy triggering intracellular signalling
pathways 58. The expression of β6-integrin is restricted to epithelial cells. The β6-
chain only associates with the αV subunit. Among the ligands of the aVβ6 complex
are fibronectin, tenascin, and “TGF-β latency-associated peptide” (LAP). αvβ6 is
expressed during embryonic development and organogenesis. In adults αvβ6
expression is only found in few epithelial tissues. Expression of αvβ6 can be re-
induced during specific morphogenetic processes such as inflammation and wound
healing. Re-induction of αvβ6 during intestinal EMT indicates an important role of
this receptor in that process. It could be shown that the re-expression of αvβ6 in
colon carcinoma cell lines is correlated with aggressiveness of metastases and the
extent of EMT 29, 30. In CD fistulae αvβ6 was solely expressed in TC with the highest
staining intensity in the transitional zone between regular intestinal epithelium and
TC. The invasive potential of the EMT cells could explain why fistulae similar to
carcinoma cells are relatively aggressive. EMT in the context of fistula formation,
however, differs from EMT found during tumour metastasis as in the case of fistulae
normal, non-transformed cells undergo this change in differentiation.
TGF-β has been shown to be a key inducer and an important regulator of EMT. The
TGF-β effect on EMT activation depends on β-integrin transduction 23. The effect of
TGFβ typically is either mediated via a Smad3-dependent regulation of transcription
or a Smad3-independent, p38MAP-kinase-activation and GTPase-mediated signalling
45, 50, 53, 59, 60. The induction of β6 integrin expression is likely to be involved in the
induction of autocrine TGF secretion 30, 49. In concordance with this TGF was stronger
expressed in the TC versus squamous cells or mucosal epithelial cells.
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 18
Usually an induction of vimentin expression has been associated with EMT. Thus, the
characteristics of the TCs fit most but not all criteria for EMT. However, there is no
clear consensus in the literature defining the use of the term EMT. Therefore, we
suggest that our lesser stringent definition of EMT is appropriate.
The specific pathogenesis of fistulae in CD is not known. There is a body of data
suggesting a persistent functional change of intestinal myofibroblasts isolated from
inflamed CD compared to controls. This is accompanied by a reduction of FAK protein
and FAK phosphorylation, which may be responsible for the reduced migratory
capacity of CD-myofibroblasts and TNF/IFN-γ treated cells 13. The reduced ability of
inflammation-modified CD myofibroblast to close ulcers and deep tissue defects may
force epithelial cells to undergo EMT and become rapidly moving and “invasive” to
re-establish an epithelial layer and consecutively the mucosal barrier.
In summary our data indicate a change in phenotype of intestinal mucosal and
squamous epithelial cells into TC. Staining for markers of EMT provides evidence for
this mechanism in perianal, entero-cutaneous, entero-enteric and entero-vesical
fistulae. The functional characteristics of EMT, such as increased capacity for
migration and three-dimensional invasion by downregulation of molecules, which
promote epithelial integrity, might be an attempt of the intestinal non-immune cells
to close the mucosal defects, however in this case with a detrimental effect. The
presence of similar results in non-CD fistulae implies that EMT may not only be
involved in fistula formation in CD. This indicates that EMT may be a general feature
of histologically similar fistulae, regardless of their etiology.
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 19
It will be important to further understand the mechanisms of EMT in the
pathophysiology of fistula formation. This study opens a new area of research.
Interfering with EMT may finally provide new therapy targets in the treatment of CD
fistulae.
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 20
Figure legends
Figure 1. Expression of E-cadherin in CD fistulae
The regular mucosal epithelial cells (black arrow) showed a stronger expression of E-
cadherin, compared to the transitional zone and the TC (red arrow; A, B). 2/3 of the
specimens showed a scattered appearance of the fistula epithelium (B). In the TC
originating from the squamous epithelial cells no E-cadherin expression was noted
(C). The negative control is shown in figure D. Semi-quantitative analysis of the
staining intensity revealed a clearly higher expression of E-cadherin in the mucosal
epithelial cells versus the TC (E). Epithelial cells from the specimen with the
squamous epithelial lining expressed E-cadherin with a lower intensity, which was
even lower in TC (F).
Figure 2. Change of the pattern of expression of E-cadherin in CD fistulae
The pattern of E-cadherin expression in the fistula lining cells was dependent on the
distance of the cells from the origin of the fistula in the gut lumen. Cells closer to the
mucosal surface show a more regular expression of E-cadherin (A and B) than cells
from deeper areas of the fistula (C and D). E: typical “transition zone” showing this
change of localization of E-cadherin staining. (A, C, D and E: magnification x 200; C:
magnification x 400).
Figure 3. Expression of β-catenin in CD fistulae
The regular mucosal epithelial cells (black arrow) showed a strong expression of β-
catenin, which was located at the cell membrane. The TC however (red arrow)
express lower amount of β-catenin, which was mainly fragmented in the cytoplasm
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 21
(A – D). The same observation could be made for fistulae with a squamous cell
epithelial origin (E). The negative control is shown in figure F. Semi-quantitative
analysis of the staining intensity revealed a higher expression of β-catenin in the
mucosal epithelial cells versus the TC (G). The same observation could be made for
fistulae with a squamous cell epithelial origin (H).
Figure 4. Expression of cytokeratin 8 in CD fistulae
The mucosal epithelial cells (black arrow) as well as the TC lining the fistula lumen
(red arrow) are strongly positive for cytokeratin 8 (A - D; C: same specimen as
shown in Figure 3C). In D the transition zone with more flat cells to the lower right
corner is clearly visible without reduction of CK 8 expression. The TC, which
originated from squamous epithelial cells were also positive for cytokeratin 8 (E). The
negative control is shown in figure F. Squamous epithelial cells did not express
cytokeratin 8 versus a median expression of 1.5 in TC.
Figure 5. Expression of β6 integrin in CD fistulae
The regular mucosal epithelial cells (black arrow) did not express β6 integrin,
whereas the staining in the transitional zone (red arrow) is strongly positive (A - D).
A higher magnification (areas from panel D) clearly shows the difference in staining
intensity between TC cells (left) and normal epithelium (right). The isotype control
did not show any staining (data not shown). Semi-quantitative analysis of the
staining intensity revealed a clearly higher expression of β6 integrin at the
transitional zone versus the regular mucosal epithelium (F).
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 22
Figure 6. Expression of TGFβ2 in CD fistulae
The regular mucosal epithelial cells (black arrow) showed a weak expression of
TGFβ2, whereas the staining of the TC (red arrow) was strongly positive for TGFβ2
(A). In panel B a higher magnification of the fistula lining cells and in C of a normal
crypt is shown to clearly demonstrate the difference in immunohistochemical
staining. The same observation was made for fistulae with a squamous cell epithelial
origin (D). Semi-quantitative analysis of the staining intensity revealed a higher
expression of TGFβ2 at the transitional zone versus the regular mucosal epithelium
and the squamous epithelium (E, F).
Figure 7. Evidence for EMT in an entero-enteric fistula
A: H&E staining of an entero-enteric fistula. The fistula origin is to the upper left. B:
Cytokeratin 8 is expressed in all fistula lining cells, even in regions were they are flat
and no longer columnar with a more mesenchymal-like morphology. E-Cadherin
expression was associated to cell membranes and cell-cell contacts in the beginning
of the fistula. However in the transition zone E-cadherin expression clearly was
reduced (C). Higher magnifications (D and E) indicate this difference. Whereas b-
catenin expression also could be located to cell borders and cell membranes at the
beginning of the fistula (F and higher magnification in G) in the transition zone β-
catenin was localized in the cytoplasm or even in the nucleus (H) as described to be
typical for EMT.
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 23
Tables Table 1 Characteristics of the fistulae investigated in this study. Age of patient
at presentation
gender localisation of fistula
Entero-cutaneous, CD patients 46 f ileum 67 f ileum 17 m left sided colon 24 m terminal ileum 37 m terminal ileum 23 f cecum 20 m sigmoid colon 20 m sigmoid colon 48 f colon 42 f ileum 24 m ileum 40 m ileum 37 f perianal 27 f ileum 71 m perianal 39 f perianal 24 f perianal 35 m perianal
Entero-enteric and entero-vesical, CD patients 52 m Ileum – sigmoid colon (entero-
enteric) and ileum-bladder (entero-vesical)
19 m Ileum-ileum (entero-enteric) Entero-cutaenous, non IBD patients
48 m perianal 56 m perianal
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 24
References 1. Hellers G, Bergstrand O, Ewerth S, Holmstrom B. Occurrence and outcome
after primary treatment of anal fistulae in Crohn's disease. Gut 1980;21:525-7.
2. Michelassi F, Stella M, Balestracci T, Giuliante F, Marogna P, Block GE.
Incidence, diagnosis, and treatment of enteric and colorectal fistulae in
patients with Crohn's disease. Ann Surg 1993;218:660-6.
3. Solomon MJ. Fistulae and abscesses in symptomatic perianal Crohn's disease.
Int J Colorectal Dis 1996;11: 222-6.
4. Loftus EV, Jr., Schoenfeld P, Sandborn WJ. The epidemiology and natural
history of Crohn's disease in population- based patient cohorts from North
America: a systematic review. Aliment Pharmacol Ther 2002;16:51-60.
5. Bell SJ, Kamm MA. Review article: the clinical role of anti-TNFalpha antibody
treatment in Crohn's disease. Aliment Pharmacol Ther 2000;14:501-14.
6. Present DH, Rutgeerts P, Targan S, Hanauer SB, Mayer L, van Hogezand RA,
Podolsky DK, Sands BE, Braakman T, DeWoody KL, Schaible TF, van Deventer
SJ. Infliximab for the treatment of fistulas in patients with Crohn's disease. N
Engl J Med 1999;340:1398-405.
7. Givel JC, Hawker P, Allan R, Keighley MR, Alexander-Williams J. Entero-enteric
fistula complicating Crohn's disease. J Clin Gastroenterol 1983;5:321-3.
8. Sandborn WJ, Fazio VW, Feagan BG, Hanauer SB. AGA technical review on
perianal Crohn's disease. Gastroenterology 2003;125:1508-30.
9. F. Bataille, F. Klebl, R. Rummele, J. Schroeder, S. Farkas, P.J. Wild, A. Furst,
F. Hofstädter, J. Schölmerich, H. Herfarth, G. Rogler. Morphological
characterisation of Crohn's disease fistulae. Gut 2004;53:1314-21
10. Postlethwaite AE, Shigemitsu H, Kanangat S. Cellular origins of fibroblasts:
possible implications for organ fibrosis in systemic sclerosis. Curr Opin
Rheumatol 2004;16:733-8.
11. Postlethwaite AE, Seyer JM. Stimulation of fibroblast chemotaxis by human
recombinant tumor necrosis factor alpha (TNF-alpha) and a synthetic TNF-
alpha 31-68 peptide. J Exp Med 1990;172:1749-56.
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 25
12. Leeb SN, Vogl D, Grossmann J, Falk W, Scholmerich J, Rogler G, Gelbmann
CM. Autocrine fibronectin-induced migration of human colonic fibroblasts. Am
J Gastroenterol 2004;99:335-40.
13. Leeb SN, Vogl D, Gunckel M, Kiessling S, Falk W, Goke M, Scholmerich J,
Gelbmann CM, Rogler G. Reduced migration of fibroblasts in inflammatory
bowel disease: role of inflammatory mediators and focal adhesion kinase.
Gastroenterology 2003;125:1341-54.
14. Leeb SN, Vogl D, Falk W, Scholmerich J, Rogler G, Gelbmann CM. Regulation
of migration of human colonic myofibroblasts. Growth Factors 2002;20:81-91.
15. Heath JP. Epithelial cell migration in the intestine. Cell Biol Int 1996;20:139-
46.
16. Goke M, Podolsky DK. Regulation of the mucosal epithelial barrier. Baillieres
Clin Gastroenterol 1996;10:393-405.
17. Podolsky DK. Healing the epithelium: solving the problem from two sides. J
Gastroenterol 1997;32:122-6.
18. Jones MK, Tomikawa M, Mohajer B, Tarnawski AS. Gastrointestinal mucosal
regeneration: role of growth factors. Front Biosci 1999;4:D303-9.
19. Thompson JS, Saxena SK, Sharp JG. Regulation of intestinal regeneration:
new insights. Microsc Res Tech 2000;51:129-37.
20. Dignass AU. Mechanisms and modulation of intestinal epithelial repair.
Inflamm Bowel Dis 2001;7:68-77.
21. Tarnawski AS. Cellular and molecular mechanisms of gastrointestinal ulcer
healing. Dig Dis Sci 2005;50 Suppl 1:S24-33.
22. Lee JM, Dedhar S, Kalluri R, Thompson EW. The epithelial-mesenchymal
transition: new insights in signaling, development, and disease. J Cell Biol
2006;172:973-81.
23. Kalluri R, Neilson EG. Epithelial-mesenchymal transition and its implications for
fibrosis. J Clin Invest 2003;112:1776-84.
24. Arias AM. Epithelial mesenchymal interactions in cancer and development. Cell
2001;105:425-31.
25. Thiery JP, Chopin D. Epithelial cell plasticity in development and tumor
progression. Cancer Metastasis Rev 1999;18:31-42.
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 26
26. Thiery JP. Epithelial-mesenchymal transitions in development and pathologies.
Curr Opin Cell Biol 2003;15:740-6.
27. Thiery JP, Sleeman JP. Complex networks orchestrate epithelial-mesenchymal
transitions. Nat Rev Mol Cell Biol 2006;7:131-42.
28. Zeisberg M, Shah AA, Kalluri R. Bone morphogenic protein-7 induces
mesenchymal to epithelial transition in adult renal fibroblasts and facilitates
regeneration of injured kidney. J Biol Chem 2005;280:8094-100.
29. Bates RC. Colorectal cancer progression: integrin alphavbeta6 and the
epithelial-mesenchymal transition (EMT). Cell Cycle 2005;4:1350-2.
30. Bates RC, Bellovin DI, Brown C, Maynard E, Wu B, Kawakatsu H, Sheppard D,
Oettgen P, Mercurio AM. Transcriptional activation of integrin beta6 during the
epithelial-mesenchymal transition defines a novel prognostic indicator of
aggressive colon carcinoma. J Clin Invest 2005;115:339-47.
31. Gasche C, Scholmerich J, Brynskov J, D'Haens G, Hanauer SB, Irvine EJ,
Jewell DP, Rachmilewitz D, Sachar DB, Sandborn WJ, Sutherland LR. A simple
classification of Crohn's disease: report of the Working Party for the World
Congresses of Gastroenterology, Vienna 1998. Inflamm Bowel Dis 2000;6:8-
15.
32. Modigliani R. Endoscopic severity index for Crohn's disease. Gastrointest
Endosc 1990;36:637.
33. Lombaerts M, van Wezel T, Philippo K, Dierssen JW, Zimmerman RM, Oosting
J, van Eijk R, Eilers PH, van de Water B, Cornelisse CJ, Cleton-Jansen AM. E-
cadherin transcriptional downregulation by promoter methylation but not
mutation is related to epithelial-to-mesenchymal transition in breast cancer
cell lines. Br J Cancer 2006.
34. Tanaka H, Shirkoohi R, Nakagawa K, Qiao H, Fujita H, Okada F, Hamada J,
Kuzumaki S, Takimoto M, Kuzumaki N. siRNA gelsolin knockdown induces
epithelial-mesenchymal transition with a cadherin switch in human mammary
epithelial cells. Int J Cancer 2006;118:1680-91.
35. Peinado H, Portillo F, Cano A. Transcriptional regulation of cadherins during
development and carcinogenesis. Int J Dev Biol 2004;48:365-75.
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 27
36. Lu Z, Ghosh S, Wang Z, Hunter T. Downregulation of caveolin-1 function by
EGF leads to the loss of E-cadherin, increased transcriptional activity of beta-
catenin, and enhanced tumor cell invasion. Cancer Cell 2003;4:499-515.
37. Janda E, Nevolo M, Lehmann K, Downward J, Beug H, Grieco M. Raf plus
TGFbeta-dependent EMT is initiated by endocytosis and lysosomal degradation
of E-cadherin. Oncogene 2006.
38. Bellovin DI, Bates RC, Muzikansky A, Rimm DL, Mercurio AM. Altered
localization of p120 catenin during epithelial to mesenchymal transition of
colon carcinoma is prognostic for aggressive disease. Cancer Res
2005;65:10938-45.
39. Brabletz T, Hlubek F, Spaderna S, Schmalhofer O, Hiendlmeyer E, Jung A,
Kirchner T. Invasion and metastasis in colorectal cancer: epithelial-
mesenchymal transition, mesenchymal-epithelial transition, stem cells and
beta-catenin. Cells Tissues Organs 2005;179:56-65.
40. Masszi A, Fan L, Rosivall L, McCulloch CA, Rotstein OD, Mucsi I, Kapus A.
Integrity of cell-cell contacts is a critical regulator of TGF-beta 1-induced
epithelial-to-myofibroblast transition: role for beta-catenin. Am J Pathol
2004;165:1955-67.
41. Eger A, Stockinger A, Park J, Langkopf E, Mikula M, Gotzmann J, Mikulits W,
Beug H, Foisner R. beta-Catenin and TGFbeta signalling cooperate to maintain
a mesenchymal phenotype after FosER-induced epithelial to mesenchymal
transition. Oncogene 2004;23:2672-2680.
42. Muller T, Bain G, Wang X, Papkoff J. Regulation of epithelial cell migration and
tumor formation by beta-catenin signaling. Exp Cell Res 2002;280:119-33.
43. Owens DW, Lane EB. Keratin mutations and intestinal pathology. J Pathol
2004;204:377-85.
44. Wang N, Stamenovic D. Mechanics of vimentin intermediate filaments. J
Muscle Res Cell Motil 2002;23:535-40.
45. Roberts AB, Tian F, Byfield SD, Stuelten C, Ooshima A, Saika S, Flanders KC.
Smad3 is key to TGF-beta-mediated epithelial-to-mesenchymal transition,
fibrosis, tumor suppression and metastasis. Cytokine Growth Factor Rev
2006;17:19-27.
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 28
46. Zavadil J, Bottinger EP. TGF-beta and epithelial-to-mesenchymal transitions.
Oncogene 2005;24:5764-74.
47. Kasai H, Allen JT, Mason RM, Kamimura T, Zhang Z. TGF-beta1 induces
human alveolar epithelial to mesenchymal cell transition (EMT). Respir Res
2005;6:56.
48. Nawshad A, Lagamba D, Polad A, Hay ED. Transforming growth factor-beta
signaling during epithelial-mesenchymal transformation: implications for
embryogenesis and tumor metastasis. Cells Tissues Organs 2005;179:11-23.
49. Bates RC, Mercurio AM. The Epithelial-Mesenchymal Transition (EMT) and
Colorectal Cancer Progression. Cancer Biol Ther 2005;4:365-70.
50. Valcourt U, Kowanetz M, Niimi H, Heldin CH, Moustakas A. TGF-beta and the
Smad signaling pathway support transcriptomic reprogramming during
epithelial-mesenchymal cell transition. Mol Biol Cell 2005;16:1987-2002.
51. Margetts PJ, Bonniaud P, Liu L, Hoff CM, Holmes CJ, West-Mays JA, Kelly MM.
Transient overexpression of TGF-{beta}1 induces epithelial mesenchymal
transition in the rodent peritoneum. J Am Soc Nephrol 2005;16:425-36.
52. Yao HW, Xie QM, Chen JQ, Deng YM, Tang HF. TGF-beta1 induces alveolar
epithelial to mesenchymal transition in vitro. Life Sci 2004;76:29-37.
53. Zavadil J, Cermak L, Soto-Nieves N, Bottinger EP. Integration of TGF-
beta/Smad and Jagged1/Notch signalling in epithelial-to-mesenchymal
transition. Embo J 2004;23:1155-65.
54. Rieder F, Brenmoehl J, Leeb S, Scholmerich J, Rogler G. Wound healing and
fibrosis in intestinal disease. Gut 2007;56:130-9.
55. Gumbiner BM. Regulation of cadherin-mediated adhesion in morphogenesis.
Nat Rev Mol Cell Biol 2005;6:622-34.
56. Perez-Moreno M, Jamora C, Fuchs E. Sticky business: orchestrating cellular
signals at adherens junctions. Cell 2003;112:535-48.
57. Gumbiner BM. Coordinate gene regulation by two different catenins. Dev Cell
2005;8:795-6.
58. Hynes RO. Integrins: versatility, modulation, and signaling in cell adhesion.
Cell 1992;69:11-25.
59. Davies M, Robinson M, Smith E, Huntley S, Prime S, Paterson I. Induction of
an epithelial to mesenchymal transition in human immortal and malignant
Bataille Evidence for a role of EMT in the pathogenesis of CD fistulae 29
keratinocytes by TGF-beta1 involves MAPK, Smad and AP-1 signalling
pathways. J Cell Biochem 2005;95:918-31.
60. Saika S, Kono-Saika S, Tanaka T, Yamanaka O, Ohnishi Y, Sato M, Muragaki
Y, Ooshima A, Yoo J, Flanders KC, Roberts AB. Smad3 is required for
dedifferentiation of retinal pigment epithelium following retinal detachment in
mice. Lab Invest 2004;84:1245-58.