myofibroblast-induced tumorigenicity of pancreatic ductal

rights reserved. For Permissions, please email: journals.permissions@ © The Author 201 oup.com1 . Published by Oxford University Press. All

Myofibroblast-induced tumorigenicity of pancreatic ductal epithelial cells

is L1CAM-dependent

Heiner Schäfer1, Claudia Geismann1, Carola Heneweer2, Jan-Hendrik Egberts3, Olena Korniienko4, Helena

Kiefel5, Gerhard Moldenhauer5, Max G. Bachem6 Holger Kalthoff4, Peter Altevogt5, Susanne Sebens1,7

1Department of Internal Medicine I, Laboratory of Molecular Gastroenterology & Hepatology, UKSH-

Campus Kiel, Kiel, Germany; 2 Clinic for Diagnostic Radiology, UKSH-Campus Kiel, Kiel, Germany 3

Clinic of General Surgery and Thoracic Surgery, UKSH-Campus Kiel, Kiel, Germany; 4 Institute for

Experimental Cancer Research, UKSH-Campus Kiel, Kiel, Germany; 5German Cancer Research Center,

Department of Translational Immunology D015, Heidelberg, Germany; 6 Department of Clinical Chemistry

and Pathobiochemistry, University of Ulm, Ulm, Germany; 7 Institute for Experimental Medicine c/o

Department of Internal Medicine I, UKSH-Campus Kiel, Kiel, Germany

Short title: L1CAM-mediated tumorigenesis of pancreatic cancer

Keywords: CD171, L1CAM antibody, pancreatic cancer, myofibroblasts, tumorigenicity

Abbreviations: aPSC: activated pancreatic stellate cell; CAF: carcinoma asscociated fibroblast; EMT:

Epithelial-Mesenchymal Transition; MF: myofibroblast; PanIN: Pancreatic Intraepithelial Neoplasia;

PDAC: pancreatic ductal adenocarcinoma; PMF: pancreatic myofibroblast; SCID: severe combined

immunodeficiency; TGF-β1: Transforming Growth Factor-beta1; TRAIL: TNF-related apoptosis-inducing

ligand

Correspondence address:

Susanne Sebens, Ph.D.

Institute for Experimental Medicine

c/o Department of Internal Medicine I, UKHS-Campus Kiel

Arnold-Heller-Str. 3, Haus 6, 24105 Kiel, Germany

Phone: +49-431-597-3835/ Fax: +49-431-597-1427

Email: [email protected]

Carcinogenesis Advance Access published November 17, 2011 at Pennsylvania State U

niversity on September 11, 2016

http://carcin.oxfordjournals.org/D

ownloaded from

mailto:[email protected]

http://carcin.oxfordjournals.org/

Abstract

Pancreatic ductal adenocarcinoma (PDAC) and chronic pancreatitis, representing one risk factor for PDAC,

are characterized by a marked desmoplasia enriched of pancreatic myofibroblasts (PMFs). Thus, PMFs are

thought to essentially promote pancreatic tumorigenesis. We recently demonstrated that the adhesion

molecule L1CAM is involved in epithelial-mesenchymal transition of PMF-cocultured H6c7 human ductal

epithelial cells and that L1CAM is expressed already in ductal structures of chronic pancreatitis with even

higher elevation in primary tumors and metastases of PDAC patients. This study aimed at investigating

whether PMFs and L1CAM drive malignant transformation of pancreatic ductal epithelial cells by

enhancing their tumorigenic potential. Cell culture experiments demonstrated that in the presence of PMFs,

H6c7 cells exhibit a profound resistance against death ligand induced apoptosis. This apoptosis protection

was similarly observed in H6c7 cells stably overexpressing L1CAM. Intrapancreatic inoculation of H6c7

cells together with PMFs (H6c7co) resulted in tumor formation in 7/8 and liver metastases in 6/8 SCID mice

whereas no tumors and metastases were detectable after inoculation of H6c7 cells alone. Likewise, tumor

outgrowth and metastases resulted from inoculation of L1CAM overexpressing H6c7 cells in 5/7 and 3/7

SCID mice, respectively, but not from inoculation of mock transfected H6c7 cells. Treatment of H6c7co

tumor-bearing mice with the L1CAM antibody L1-9.3/2a inhibited tumor formation and liver metastasis in

100% and 50%, respectively, of the treated animals. Overall, these data provide new insights into the

mechanisms of how PMFs and L1CAM contribute to malignant transformation of pancreatic ductal

epithelial cells in early stages of pancreatic tumorigenesis.

Introduction

Pancreatic ductal adenocarcinoma (PDAC) is one of the most malignant tumor diseases exhibiting an

unfavorable prognosis. In Western countries, PDAC ranked 4th in the order of fatal tumor diseases with a

still increasing prevalence (1). Owing to the lack of specific symptoms, PDAC is mostly detected in an

at Pennsylvania State University on Septem

ber 11, 2016http://carcin.oxfordjournals.org/

Dow

nloaded from


3

advanced stage when the tumor has already metastasized and therapeutic options are limited accounting for

an overall 5-year survival rate of 2 % (2).

Most PDACs are thought to arise from the ductal epithelium. One hallmark of PDAC is its pronounced

desmoplasia (tumor stroma) mainly consisting of activated fibroblasts (also designated as myofibroblasts

(MF), activated pancreatic stellate cells (aPCS) or carcinoma-associated fibroblasts (CAF)) and extracellular

matrix proteins. This marked desmoplasia is already present in chronic pancreatitis (3-5) which represents

one risk factor for PDAC (3,4). Besides genetic alterations e.g. in the k-ras, p53 or DPC4/Smad4 genes,

continuous interactions between ductal and, later on, carcinoma cells and the stromal environment are

important for PDAC tumorigenesis (6,7,8). Guerra et al. showed that expression of mutant k-RasG12V in

acinar and centroacinar cells of adult mice did not result in the development of Pancreatic Intraepithelial

Neoplasias (PanINs) and invasive PDAC unless the animals undergo a chronic pancreatitis (9) indicating

that genetic alterations in somatic cells alone do not necessarily lead to neoplastic formation. Moskovitz et

al. demonstrated that chronic exposure to oxidative stress which is likely to occur during chronic

inflammation can induce chromosomal alterations in cultured normal pancreatic ductal epithelial cells (10).

Accordingly, genomic instability in terms of chromosomal and gene abnormalities could be detected already

in earliest PanIN lesions and tissues of chronic pancreatitis (10,11). In addition, the epithelial-mesenchymal

transition (EMT) is supposed to promote tumorigenesis. Upon this process, epithelial and carcinoma cells,

respectively, lose their epithelial phenotype by acquiring a spindle-shaped morphology and exhibiting a

reduced expression of the epithelial specific surface molecule E-cadherin whereas the expression of

mesenchymal marker proteins such as vimentin and N-cadherin is elevated (12,13). These alterations are

thought to be prerequisite for cell dissociation from the epithelial tissue consequently leading to metastasis.

One important inducer of EMT is the transforming growth factor-beta 1 (TGF-β1) which can be released by

both stroma and tumor cells leading to a paracrine loop by which malignancy associated alterations of the

stroma and tumor are promoted. Thus, tumor cell derived TGF-β1 leads to activation of fibroblasts/stellate

cells and promotes desmoplastic reaction. Secretion of TGF-β1 by MFs (aPSCs/CAFs) fosters EMT in the

tumors cells and thereby tumor progression (14-17). Using the human pancreatic ductal cell line H6c7, we

recently demonstrated that coculture with pancreatic myofibroblasts (PMFs) led to TGF-β1-dependent EMT



Dow

nloaded from


4

being associated with an upregulation of the adhesion molecule L1CAM (18). Moreover, as a result of this

elevated L1CAM expression, H6c7 cells acquire a chemoresistant and migratory phenotype (18). L1CAM

(CD171) is a member of the immunoglobulin superfamily and has been detected in a variety of tumors such

as melanoma, colon cancer as well as in PDAC (19-24) being associated with poor prognosis and short

survival times (21,24,25). Several studies demonstrated an involvement of L1CAM in various biological

processes of tumorigenesis such as tumor cell migration and invasion, chemoresistance as well as tumor

growth and survival (18,19,20,22,26) pointing to an important role of this protein in the development and

progression of tumors.

H6c7 cells do not harbor established mutations in oncogenes or tumor suppressor genes (27-29) regarded as

initial events in pancreatic tumorigenesis. Moreover, they are described to be non-tumorigenic and our

recent data support the hypothesis that i) malignant transformation of pancreatic ductal epithelial cells might

be initiated before genetic aberrations occur and ii) PMFs and L1CAM play an essential role in this scenario.

Based on these findings, the present study aimed at elucidating whether the PMF- and L1CAM-mediated

alterations of ductal epithelial cells relates to an increased tumor formation and metastasis in vivo.



Dow

nloaded from


5

Material & Methods

Cell lines and cell culture. The human pancreatic ductal epithelial cell line H6c7 (kindly provided by Prof.

M.S. Tsao, Ontario Cancer Center, Toronto, Canada) was cultured as described (30). For some experiments,

H6c7 cells were retrovirally transduced either with an empty vector (mock) or with a vector encoding human

L1CAM (31). Before transduction, the L1CAM construct was verified by sequencing. The transduction

procedure with retroviral vectors and the selection of transduced clones with puromycin were described

previously (32).

Pancreatic fibroblasts. Murine myofibroblasts (PMFs) were isolated and cultured as described (8). The

myofibroblastic phenotype was confirmed by immunocytochemical staining for vimentin and α-smooth

muscle actin. Since the in vivo experiments were performed with PMFs, the in vitro experiments were

mainly done with murine PMFs and confirmative experiments were performed with human PMFs (see

supplementary data).

Coculture. For coculture, 1x105 H6c7 cells in 2 mL H6c7 medium were seeded into a well of a 6-well-plate,

and 2x105 PMFs in 1.5 mL PMF medium were seeded into a transwell insert (0.4 µm pore size, Greiner,

Frickenhausen, Germany) which was placed into a well with H6c7 cells. After 24 hours, medium was

exchanged. To continue the coculture for four weeks with PMFs (for human PMFs see supplementary data),

cells were seeded as described above. Then, every week cocultured H6c7 cells and cocultured PMFs were

detached, counted and freshly seeded at the respective cell numbers.

Antibodies and reagents. The mouse anti-human L1-9.3/2a antibody binding to the ectodomain of L1CAM

has been described earlier (33). Mouse IgG2a control antibodies (clone Cl.18) were purchased from

BioXCell (West Lebanon NH, USA). Killer TRAIL was purchased from Alexis Biochemicals (Grünberg,

Germany) and the agonistic anti-CD95/Fas antibody CH-11 was obtained from Coulter Immunotech



Dow

nloaded from


6

(Hamburg, Germany). Both reagents were used at a final concentration of 0.1 µg/ml. Human TNF-α was

from Sigma-Aldrich Chemie (Taufkirchen, Germany) and used at a final concentration of 20 ng/ml.

Measurement of apoptosis by determination of caspase-3/-7 activity. For determination of apoptosis,

cells were either left untreated or were treated as indicated. Twenty-four hours after treatment, caspase-3/-7

activity was determined by using a kit from Promega (Mannheim, Germany) according to the

manufacturer’s instructions and as previously described (8,22). All assays were done in duplicates. Data are

expressed as n-fold treatment induced caspase-3/-7 activity (over basal) related to the cell number which

was determined in parallel.

Tumor model and therapy. All animal experiments have been approved by the local ministry and were

performed in the central animal facility of the University Hospital Schleswig-Holstein under pathogen-free

conditions. All experiments were assessed in four to six week old female SCID beige mice purchased from

Charles River Laboratories (Sulzfeld, Germany). For orthotopic cell inoculation, H6c7-mock, H6c7-

L1CAM, H6c7mono or H6c7co cells were cultured as described above, trypsinized and resuspended in 50 µl

Matrigel (BD Biosciences, Heidelberg, Germany) at a concentration of 2x106 cells. H6c7co cells that were

cocultured for 4 weeks in the presence of PMFs were inoculated together with 4x106 murine PMFs in 50 µl

Matrigel. All syringes containing Matrigel and cells were placed on ice until intra-pancreatic inoculation of

the mice (6-8 animals/group). For this purpose, animals were anesthetised and a median laparotomy was

performed. Then, cells were inoculated into the pancreatic body to the left of the organ midline. Tumor

growth was monitored as recently described (34) by high resolution ultrasound using a Vevo 770 small

animal micro-imaging system from VisualSonics (Toronto, Ontario, Canada).

To investigate whether tumor outgrowth of H6c7co cells can be inhibited by blockade of L1CAM, two

month after cell inoculation animals were randomized into two groups (n = 7) and treated either with 10

mg/kg body weight control IgG2a or L1-9.3/2a. All antibodies were diluted in 200 μL 0.9 % NaCl and intra-

peritoneally injected weekly for a time period of 3 months. In all experiments, the last ultrasound

examination was done after 5 months just before animals were sacrificed and pancreatic and liver tissues

were removed for final determination of the tumor size and preservation in liquid nitrogen.



Dow

nloaded from


7

Immunohistochemistry. Six-µm thick consecutive cryostat sections were mounted on uncovered glass

slides, air-dried for at least 1 hour at room temperature (RT) and stored at -20°C. For staining procedure,

sections were dried again for 30 minutes (min) at RT, fixed in acetone (Merck, Darmstadt, Germany) for 10

min and air-dried again for 10 min. Then, slides were washed in PBS and incubated with 4 % bovine serum

albumine (BSA) in PBS (Serva, Heidelberg, Germany) for 20 min followed by incubation with the primary

antibody for 45 minutes at RT. Primary antibodies were diluted at a final concentration of 10 µg/mL

(L1CAM) or 2 µg/mL in PBS + 1 % BSA/PBS: mouse IgG1 anti-human L1CAM (clone UJ127; 35), mouse

IgG1 anti-human pan-Cytokeratin (DakoCytomation, Hamburg, Germany), mouse IgG1 anti-human

vimentin (Santra Cruz Biotechnology, Heidelberg, Germany), rabbit-anti-human E-Cadherin (Cell Signaling

via New England Biolabs, Frankfurt a.M., Germany). Primary antibodies were detected with EnVision

conjugated with peroxidase (DakoCytomation) for 30 min. The substrate reaction for peroxidase was

performed with the AEC substrate from DakoCytomation according to the instructions of the manufacturer.

Afterwards, sections were washed in water, counterstained in 50 % haemalaun (Merck) and mounted with

glycerol-gelatin (Merck). The same protocol was performed for negative controls in which an isotype-

matched control antibody was used, which revealed either no staining or only weak background staining.

Statistical analysis. Data are presented as mean ± standard deviation and were analyzed by student´s t-test.

A p-value < 0.05 was considered statistically significant.



Dow

nloaded from


8

Results

H6c7 cells exhibiting elevated L1CAM expression and EMT-associated alterations during coculture

with PMFs show a decreased apoptosis sensitivity towards death ligands. Our previous results showed

that upon coculture with PMFs H6c7 cells acquire a migratory and chemoresistant phenotype (18) thus

pointing to the capability of PMFs to induce malignant transformation of H6c7 cells. In support of these

findings, H6c7 cells cocultured with PMFs for four weeks were characterized by EMT-associated

alterations. As shown in Figure 1A, H6c7 cells that were positively stained with a pan-cytokeratin antibody

indicating their epithelial origin, exhibited elevated expression levels of L1CAM when cocultured in the

presence of PMFs (H6c7co) in contrast to monocultured H6c7 cells (H6c7mono). Moreover, whereas

H6c7mono cells were characterized by a transmembraneous expression of E-cadherin and the absence of

vimentin, E-cadherin expression was lost in single H6c7co cells and numerous cells exhibited a strong

expression of vimentin (indicated by arrows). Coculture-mediated alterations of L1CAM, E-cadherin and

vimentin expression in H6c7co cells were confirmed by western blotting (Supplementary Figure 1). In

addition, numerous H6c7co cells already exhibited a spindle-like cell shape which is consistent with the

above mentioned alterations. To further elucidate this conversion process we investigated whether coculture

with PMFs also affects proliferation and death ligand-induced apoptosis. Whereas proliferation of H6c7co

cells was rather decelerated than accelerated upon coculture with PMFs (data not shown), the apoptotic

response towards death ligands was considerably altered. As shown in Figure 1B, H6c7mono cells showed a

clear apoptosis induction after stimulation with TRAIL (5.6-fold), TNF-α (2.6-fold) or the CD95 agonistic

antibody CH-11 (5.3-fold) as determined by caspase-3/-7 assay. In contrast, H6c7co cells became much less

apoptotic in response to these stimuli, in particular after treatment with TRAIL (1.9-fold) and CH-11 (2.2-

fold). Similar data were obtained upon coculture with human PMFs (Supplementary Figure 2). In order to

elucidate whether L1CAM might account for protection from death ligand-induced apoptosis, L1CAM

expression was suppressed by siRNA transfection in H6c7co cells during coculture and apoptosis induction

after treatment with TRAIL and CH11 was analyzed. As demonstrated in Figure 1C, caspase-3/-7 activity

was markedly increased in H6c7co cells after treatment with either apoptotic stimuli when L1CAM

expression was inhibited compared to control transfected cells. Overall these data demonstrate that H6c7



Dow

nloaded from


9

cells undergo EMT-associated alterations in the presence of PMFs which is associated with an increased

expression of L1CAM and an L1CAM-dependent apoptosis resistant phenotype.

H6c7 cells exhibiting elevated L1CAM expression by stable transfection partially exhibit EMT-

associated alterations and apoptosis resistance towards death ligands.

To further confirm the importance of L1CAM in malignant transformation of H6c7 cells, stably L1CAM

overexpressing H6c7 cells were analyzed regarding EMT-associated alterations and death ligand-induced

apoptosis. Of the L1CAM transfectants (H6c7-L1CAM) exhibiting strong expression of L1CAM scattered

cells were characterized by an elongated cell shape and a loss of E-cadherin expression similarly to H6c7co

cells (Figure 1D). In contrast, H6c7-mock cells grown in a dense monolayer were all characterized by a

transmembraneous expression of E-cadherin and the absence of L1CAM (Figure 1D). In addition, the

number of vimentin expressing cells was slightly increased in the population of H6c7-L1CAM cells

compared to H6c7-mock cells albeit weaker than in H6c7co cells. These results were confirmed by western

blot analysis demonstrating strong L1CAM expression in H6c7-L1CAM cells. According to the small

differences seen in the staining, the expression of E-cadherin was slightly reduced and the expression of

vimentin was weakly increased in these cells compared to the mock-transfected cells (Supplementary Figure

3). As observed upon coculture with PMFs, L1CAM overexpressing H6c7 cells showed a reduced

sensitivity towards death ligand-induced apoptosis (Figure 1E). In particular, caspase-3-/-7 activity was

reduced after stimulation with TRAIL compared to control transfected cells (5,3-fold in mock versus 4,2-

fold in H6c7-L1CAM) and CH-11 (5,2-fold in H6c7-mock versus 3,1-fold in H6c7-L1CAM). Similar to the

observations upon coculture, L1CAM overexpression did not affect proliferation of H6c7-L1CAM cells

(data not shown). Overall, these data indicate that the presence of PMFs or elevated expression of L1CAM

confers a reduced apoptosis sensitivity towards death ligand-induced apoptosis in pancreatic ductal epithelial

cells.

PMFs increase tumorigenicity of H6c7 cells leading to tumor growth and metastasis in vivo. Since

PMFs induce EMT-associated alterations in H6c7 cells involving L1CAM dependent migration and



Dow

nloaded from


10

aptoptosis resistance (see above and reference 18), we next analyzed whether the tumorigenic potential of

H6c7 cells is increased along with the malignant conversion of the cells. For this purpose, SCID beige mice

were orthotopically inoculated with either H6c7 cells alone (H6c7mono) or with a combination of PMFs and

cocultured H6c7cells (H6c7co). As shown in Table 1, H6c7mono cells were almost non-tumorigenic

confirming previous results (28), only in 2/6 animals tiny lesions of unknown specificity could be detected

by ultrasound examination in the pancreas. In contrast, 7/8 animals exhibited outgrowth of pancreatic tumors

ranging between 8-80 mm3 in size (mean ± SD: 28 ± 30.1 mm3) after inoculation of H6c7co cells. Although

no metastases were detectable in spleens, 6/8 animals inoculated with H6c7co cells exhibited liver

metastases as already detected by high-resolution ultrasound (Figure 2A) and confirmed by post-mortem

analysis (Table 1). As demonstrated by immunhistochemical stainings with human-specific antibodies,

H6c7co tumors were largely composed of cytokeratin expressing cells confirming their human and epithelial

origin thus indicating that tumors are mainly composed of H6c7 cells (Figure 2B). Moreover, the majority of

these H6c7 cells exhibited an elongated cell shape and a strong L1CAM expression. Important to note, E-

cadherin expression was less prominent in the cells than cytokeratin or L1CAM indicating the loss E-

cadherin expression in tumoral H6c7cells. Vimentin could be detected in numerous cells within the primary

tumor albeit not in the frequency of cytokeratin and L1CAM (Figure 2B). Pan-cytokeratin staining of livers

revealed that most cells within metastases were indeed H6c7 cells partially expressing L1CAM.

Interestingly, all metastasized H6c7 cells expressed E-cadherin while vimentin expression was

predominantly detectable in cells at the margins of the metastases (Figure 2C). These data suggest that H6c7

cells become tumorigenic in the presence of PMFs leading to tumor formation and metastasis in vivo.

L1CAM overexpressing H6c7 cells become tumorigenic in vivo. In order to elucidate whether tumor

formation of H6c7 cells depends on L1CAM expression, H6c7 cells stably overexpressing L1CAM (H6c7-

L1CAM) or the respective control transfectants (H6c7-mock) were orthotopically inoculated in SCID beige

mice. Similarly to H6c7co cells, inoculation of H6c7-L1CAM cells (Table 2) led to pancreatic tumor

formation detectable in 71 % of the animals (5/7). In contrast, H6c7mock cells were almost non-tumorigenic

(Table 2) besides one animal bearing a tiny pancreatic lesion of unknown specificity. H6c7-L1CAM tumor



Dow

nloaded from


11

sizes ranged between 4-15 mm3 (mean ± SD: 6.7 ± 5.6 mm3) (Table 2). Thus, H6c7-L1CAM tumors were

much smaller than those arising from inoculation of H6c7co cells and PMFs but could be detected by high-

resolution ultrasound examination (Figure 3A). Inoculation of H6c7-L1CAM cells resulted in liver

metastases in 3/7 animals (Table 2) whereas only in 1/6 animals inoculated with H6c7-mock cells a liver

metastasis was found. The metastases were not detected by high-resolution ultrasound most probably due to

difficult scanning conditions because large parts of the liver are located underneath the rib cage but later on

after removal of the liver. Overall, these findings point to a role of L1CAM in increasing the tumorigenicity

and to a lesser extent the metastatic potential of human pancreatic ductal epithelial cells.

Treatment with L1CAM-antibodies inhibits tumor growth and metastasis of tumorigenic H6c7 cells.

To support the idea that L1CAM expression confers an increased tumorigenic phenotype of H6c7 cells,

H6c7co tumor-bearing animals were treated with the L1CAM-blocking antibody L1-9.3/2a or a control

antibody. Antibody treatment started two months after cell inoculation when small tumors were detectable

by ultrasound-examination. One animal treated with the L1-9.3/2a antibody for 9 weeks died due to

intestinal obstruction but exhibited neither a pancreatic tumor nor liver metastases. As seen in Table 3 and

Figure 3B, weekly administration of L1CAM-blocking antibodies over a period of three months resulted in a

complete remission of H6c7co tumor growth and reduced formation of liver metastases in 50 % of the

animals (Table 3). In contrast, 6/7 H6c7co tumor-bearing mice that were treated with the control antibody

exhibited formation of pancreatic tumors ranging between 6-468 mm3 in size (mean ± SD: 112.6 ± 180.4

mm3) (Table 3). In one animal, the pancreatic lesion was too small for further analysis and was therefore

excluded from the statistics. Metastases in the liver were not detectable by high-resolution ultrasound but

later on during the post-mortem analysis in 5/7 animals (Table 3) underscoring our previous results (Table 1,

Figure 2). Together with our findings from the H6c7-L1CAM cells, these data confirm that tumor growth

and metastasis of pancreatic ductal epithelial cells depends on L1CAM expression and that L1CAM-

blocking antibodies are an appropriate tool for anti-cancer therapy of pancreatic tumors.



Dow

nloaded from


12

Discussion

To date, PDAC is mostly diagnozed in an advanced stage limiting curative therapeutic approaches and

PDAC patient’s prognosis. Thus, an improved understanding of the early steps of pancreatic tumorigenesis

is pivotal to improve diagnostic and therapeutic measures.

The substantial impact of PMFs/CAFs on processes involved in PDAC development such as EMT, cell

migration and chemoresistance has been demonstrated by us and other groups (5,6,8,15,16). In accordance,

expression of the PMF marker protein alpha-smooth muscle actin is highly elevated in the stroma of PDAC

tissues correlating with short survival of the patients (36). We recently showed that H6c7 cells cocultured

with PMFs acquire a migratory and chemoresistant phenotype through PMF-induced L1CAM expression.

Accordingly, elevated L1CAM expression was detected in carcinoma cells in PDAC tissues but also in

ductal structures in stroma-enriched chronic pancreatitis indicating that PMFs might be involved in

malignant transformation of pancreatic ductal epithelial cells (18). We are now showing that H6c7 cells

become tumorigenic in the presence of PMFs resulting in the outgrowth of pancreatic tumors and liver

metastases whereas H6c7 cells alone were non-tumorigenic confirming the results by Furukawa et al. (28).

H6c7co tumors were strongly stained by a human pan-cytokeratin antibody indicating that they are largely

composed of human cells of epithelial origin. In search for the mechanisms leading to pancreatic

tumorigenesis, mutational activation of the k-ras gene has gained much attention since mutant k-ras is

present already in early PanINs and later on in 90% of PDAC (2,4). Hence, Qian et al. could demonstrate

that expression of mutant k-ras in H6c7 cells that per se exhibit none of the mutations commonly detected in

PDAC (27,28) induced tumor formation in 50 % of subcutaneously inoculated SCID mice (37). Similar

results have been demonstrated in rat models (38,39). Moreover, development of human PDAC was

recapitulated in different transgenic mouse models e.g. by expression of endogenous k-rasG12D induced

during early stages of embryonic pancreatic development in the mouse model of Hingorani et al. (40).

However, expression of k-rasG12D in adult mice more closely reflecting the situation in humans did not result

in PanIN formation and invasive PDAC unless a mild chronic pancreatitis was induced (9). Altogether these

data suggest that genetic alterations alone do not necessarily lead to PDAC development and that context-

derived factors such as stromal cells and their secreted factors are likewise important. This is supported by



Dow

nloaded from


13

our own findings showing that already an indirect coculture with PFMs leads to EMT in H6c7 and PDAC

cells along with a migratory and apoptosis resistant phenotype which highly depends on humoral factos such

as TGF-β1, IL-1β and nitric oxide (NO) (8,18,30). Moreover, all these factors have been described as

potential inducers of L1CAM (18,22). The observation that elevated L1CAM expression in H6c7 cells

essentially contributes to the PMF-induced migratory and apoptosis resistant phenotype in vitro points to an

important role of this molecule in the malignant transformation of pancreatic ductal epithelial cells.

In line with this, H6c7 cells in H6c7co tumors exhibited a high L1CAM expression along with a marked

expression of vimentin and a reduced expression of E-Cadherin in numerous cells pointing to a continuation

of EMT in H6c7co tumors. Interestingly, the majority of H6c7 cells in PDAC liver metastases expressed E-

cadherin whilst vimentin expression was mainly detected in cells located at the margins of the metastases.

Similarly, E-cadherin expression in metastases but not in primary tumors has been reported by various

groups (41-43). Overall, these data are consistent with the idea of a mesenchymal-epithelial transition by

which metastasized tumor cells re-express E-cadherin to seed in the new tissue and to establish metastatic

foci (44).

Gavert et al. already demonstrated that subcutaneous inoculation of L1CAM overexpressing NIH3T3 cells

into nude mice results into formation of larger tumors than inoculation of control transfected cells (20). Our

study shows that stable overexpression of L1CAM led to tumor formation of H6c7 cells in 5/7 animals,

similarly seen after coinoculation with PMFs, whereas no pancreatic tumors developed after inoculation of

the control transfectants. However, H6c7-L1CAM tumors were much smaller in size and the incidence of

metastases in the liver was lower compared to tumors and metastases in animals after incoculation of

H6c7co cells. The observation that inoculation of H6c7-mock cells resulted in the formation of metastases at

least in one animal might be explained by the fact that these cells had undergone a double retroviral

transduction procedure (1st time for immortalization and 2nd time for stable transfection with the control

vector for L1CAM) increasing the tumorigenic potential of the cells. Overall, these data indicate that

expression of L1CAM substantially increases the tumorigenicity of H6c7 cells but is not sufficient to start

metastasis in full extent as occurring under the influence of PMFs. Supporting this idea, expression of

vimentin was much less pronounced in H6c7-L1CAM cells compared to H6c7-mock cells. Gavert et al.



Dow

nloaded from


14

recently showed that L1CAM confers a motile phenotype in colon cancer cells resulting in metastasis to the

liver but without inducing EMT and stem cell properties (45,46). These results suggest that the impact of

L1CAM on metastasis is i) either more prominent in the tumorigenesis of the colon than of the pancreas or

ii) is more pronounced in fully transformed cancer cells than in epithelial cells where transformation still

commences. Accordingly, treatment of H6c7co tumor-bearing mice with L1CAM-blocking antibodies

completely abolished formation of primary tumors but reduced liver metastasis only in 50 % of the animals.

Thus, L1CAM seemed to be sufficient for initial tumor formation and growth of pancreatic ductal epithelial

cells but the process of metastasis apparently requires additional alterations and/or signals which are

provided e.g. by the stromal microenvironment. Stromal cells such as PMFs (and later on CAFs) are known

to promote cell migration and metastasis by producing a plethora of cytokines, growth factors and proteases

which enhance the migratory potential of epithelial/carcinoma cells and help to digest the extracellular

matrix (6,8,16,47).

Bao et al. reported elevated L1CAM expression on CD133+ cancer stem cells in gliomas and targeting of

L1CAM in these cells significantly reduced tumor outgrowth in immunocompromised mice (48) supporting

the view that L1CAM is able to confer a tumorigenic phenotype. Moreover, this group showed that L1CAM

is essentially involved in the regulation of DNA damage checkpoint responses thereby determining

radiosensitivity of glioblastoma stem cells (49). Under physiological conditions, these checkpoints are

tightly regulated and exert cytoprotective functions to promote cell survival (50). However, dysregulation of

these checkpoints e.g. as a result of sustained L1CAM expression might promote cell survival of damaged

cells that actually would have been eliminated in order to prevent proliferation of genetically altered cells

and tumorigenesis. Besides enhanced DNA repair, the protection from apoptosis induction might be another

mechanism by which cells harbouring DNA mutations can survive and expand. Accordingly, we could

demonstrate that H6c7 cells either cocultured in the presence of PFMs or stably transduced with

recombinant L1CAM acquired a resistant phenotype towards death ligand-induced apoptosis in particular

after stimulation of CD95 and TRAIL. To underscore the importance of L1CAM in apoptosis resistance of

tumor cells and the suitability of L1CAM antibodies as therapeutic tool, a recent study demonstrated that

L1CAM-dependent chemoresistance can be overcome by antibody-mediated blockade of L1CAM in PDAC



Dow

nloaded from


15

and ovarian carcinoma cells in vivo resulting in significantly reduced tumor growth (unpublished

observations).

In summary, the present study demonstrates that the tumorigenic potential of pancreatic ductal epithelial

cells is highly increased in the presence of PMFs promoting tumor growth and metastasis. Elevated

expression of L1CAM plays an essential role in this scenario probably by conferring a profound apoptosis

protection allowing survival and expansion of genetically altered cells. Hence, L1CAM expression in ductal

structures of chronic pancreatitis as reported previously (18) has to be seriously regarded as first steps of

tumorigenesis.

Funding

This work was supported by the German Research Society DFG [SE-1831/2-1 to S.S.] and the German

Cluster of Excellence “Inflammation at Interfaces”.

Acknowledgements

The authors thank Dagmar Leisner and Maike Witt-Ramdohr for excellent technical assistance.



Dow

nloaded from


16

References

1. Siegel, R., Ward, E., Brawley, O. and Jemal, A. (2011). Cancer statistics, 2011. CA Cancer J. Clin., 61,

212-236.

2. Schneider, G., Siveke, J.T., Eckel, F. and Schmid, R.M. (2005). Pancreatic cancer: basic and clinical

aspects. Gastroenterology, 128, 1606-1625.

3. Farrow, B. and Evers, B.M. (2002). Inflammation and the development of pancreatic cancer. Surg.

Oncol., 10, 153-169.

4. Luttges, J. and Kloppel, G. (2005). Pancreatic ductal adenocarcinoma and its precursors. Pathologe, 26,

12-17.

5. Yen, T.W., Aardal, N.P., Bronner, M.P., Thorning, D.R., Savard, C.E., Lee, S.P. and Bell, R.H. Jr. (2002).

Myofibroblasts are responsible for the desmoplastic reaction surrounding human pancreatic carcinomas.

Surgery, 131, 129-134.

6. Bhowmick, N.A., Neilson, E.G. and Moses, H.L. (2004). Stromal fibroblasts in cancer initiation and

progression. Nature, 432, 332-337.

7. Kleeff, J., Beckhove, P., Esposito, I., Herzig, S., Huber, P.E., Löhr, J.M. and Friess, H. (2007). Pancreatic

cancer microenvironment. Int. J. Cancer, 121, 699-705.

8. Muerkoster, S., Wegehenkel, K., Arlt, A., Witt, M., Sipos, B., Kruse, M.-L., Sebens, T., Kloppel, G.,

Kalthoff, H., Folsch, U.R. and Schafer, H. (2004). Tumor stroma interactions induce chemoresistance in

pancreatic ductal carcinoma cells involving increased secretion and paracrine effects of nitric oxide and

interleukin-1beta. Cancer Res., 64, 1331-1337.

9. Guerra, C., Schuhmacher, A.J.M., Canamero, M., Grippo, P.J., Verdaguer, L., Gallego-Perez, L., Debus,

P., Sandgren, E.P. and Barbacid, M. (2007). Chronic pancreatitis is essential for induction of pancreatic

ductal adenocarcinoma by k-ras oncogenes in adult mice. Cancer Cell, 11, 291-302.

10. Moskovitz, A.H., Linford, N.J., Brentnall, T.A.; Bronner, M.P., Storer, B.E., Potter, J.D., Bell, Jr. R.H.

and Rabinovitch, P.S. (2003). Chromosomal instability in pancreatic ductal cells from patients with

chronic pancreatitis and pancreatic adenocarcinoma. Genes, Chromosomes & Cancer, 37, 201-206.



Dow

nloaded from


17

11. Baumgart, M., Werther, M., Bockholt, A., Scheurer, M., Rüschoff, J., Dietmaier, W., Ghadami, B.M. and

Heinmöller, E. (2010). Genomic instability at both the base pair level and the chromosomal level is

detectable in earliest PanIN lesions in tissues of chronic pancreatitis. Pancreas, 39, 1093-1103.

12. Bhowmick, N.A., Ghiassi, M., Bakin, A., Aakre, M., Lundquist, C.A., Engel, M.E., Arteaga, C.L. and

Moses, H.L. (2001). Transforming growth factor-beta1 mediates epithelial to mesenchymal

transdifferentiation through a RhoA-dependent mechanism. Mol. Biol. Cell, 12, 27-36.

13. Guarino, M. (2007) Epithelial-mesenchymal transition and tumour invasion. Int J Biochem Cell Biol., 39,

2153-60.

14. Löhr, M., Schmidt, C., Ringel, J., Kluth, M., Müller, P., Nizze, H. and Jesnowski R. (2001).

Transforming growth factor-β1 induces desmoplasia in an experimental model of human pancreatic

carcinoma. Cancer Res., 61, 550-555.

15. Bachem, M.G., Schünemann, M., Ramadani, M., Siech, M., Beger, H., Buck, A., Zhou, S., Schmid-

Kotsas, A. and Adler, G. (2005). Pancreatic carcinoma cells induce fibrosis by stimulating proliferation

and matrix synthesis of stellate cells. Gastroenterology, 128, 907-921.

16. Bachem, M.G., Zhou, S., Buck, K., Schneiderhan, W. and Siech, M. (2008). Pancreatic stellate cells –

role in pancreatic cancer. Langenbecks Arch. Surg., 393, 891-900.

17. Pilarsky, C., Ammerpohl, O., Sipos, B., Dahl, E., Hartmann, A., Wellmann, A., Braunschweig, T., Löhr,

M., Jesnowski, R., Friess, H., Wente, M.N., Kristiansen, G., Jahnke, B., Denz, A., Rückert, F.,

Schackert, H.K., Klöppel, G., Kalthoff, H., Saeger, H.D. and Grützmann, R. (2008). Activation of Wnt

signalling in stroma from pancreatic cancer identified by gene expression profiling. J. Cell Mol. Med.,

12, 2823-2835.

18. Geismann, C., Morscheck, M., Koch, D., Ungefroren, H., Arlt, A., Tsao, M.-S., Bachem, M.G., Altevogt,

P., Fölsch, U.R., Schäfer, H. and Sebens Müerköster S. (2009). Up-regulation of L1CAM in pancreatic

duct cells is transforming Growth Factor b1- and Slug-dependent: Role in malignant transformation of

pancreatic cancer. Cancer Res., 69, 4517-4526.



Dow

nloaded from


18

19. Gast, D., Riedle, S., Riedle, S., Schabath, H., Schlich, S., Schneider, A., Issa, Y., Stoeck, A., Fogel, M.,

Joumaa, S., Wenger, T., Herr, I., Gutwein, P. and Altevogt, P. (2005). L1 augments cell migration and

tumor growth but not beta3 integrin expression in ovarian carcinomas. Int. J. Cancer, 115, 658-665.

20. Gavert, N., Conacci-Sorrell, M., Gast, D., Schneider, A., Altevogt, P., Brabletz, T. and Ben-Ze'ev, A.

(2005). L1, a novel target of beta-catenin signaling, transforms cells and is expressed at the invasive

front of colon cancers. J. Cell Biol., 168, 633-642.

21. Kaifi, J.T., Reichelt, U., Quaas, A., Schurr, P.G., Wachowiak, R., Yekebas, E.F., Strate, T., Schneider,

C., Pantel, K., Schachner, M., Sauter, G. and Izbicki, J.R. (2007). L1 is associated with micrometastatic

spread and poor outcome in colorectal cancer. Mod. Pathol., 20, 1183-1190.

22. Sebens Müerköster, S., Werbing, V., Sipos, B., Debus, M.A., Witt, M., Grossmann, M., Leisner, D.,

Kötteritzsch, J., Kappes, H., Klöppel, G., Altevogt, P., Fölsch, U.R. and Schäfer, H. (2007). Drug-

induced expression of the cellular adhesion molecule L1CAM confers anti-apoptotic protection and

chemoresistance in pancreatic ductal adenocarcinoma cells. Oncogene, 26, 2759-2768.

23. Bergmann, F., Wandschneider, F., Sipos, B., Moldenhauer, G., Schniewind, B., Welsch, T., Schirrmacher,

P., Klöppel, G., Altevogt, P., Schäfer, H. and Sebens Müerköster, S, (2010). Elevated L1CAM expression

in precursor lesions and primary and metastastic tissues of pancreatic ductal adenocarinoma. Oncology

Rep., 24, 909-915.

24. Ben, Q.-W., Wang, J.-C., Liu, J., Zhu, Y., Yuan, F., Yao, W.-Y. and Yuan, Y.-Z. (2010). Positive

expression of L1-CAM is associated with perineural invasion and poor outcome in pancreatic ductal

adenocarcinoma. Ann. Surg. Oncol., 17, 2213-2221.

25. Fogel, M., Gutwein, P., Mechtersheimer, S., Riedle, S., Stoeck, A., Smirnov, A., Edler, L., Ben-Arie, A.,

Huszar, M. and Altevogt, P. (2003). L1 expression as a predictor of progression and survival in patients

with uterine and ovarian carcinomas. Lancet, 362, 869-875.

26. Arlt, M.J., Novak-Hofer, I., Gast, D., Gschwend, V., Moldenhauer, G., Grünberg, J., Honer, M.,

Schubiger, P.A., Altevogt, P. and Krüger, A. (2006). Efficient inhibition of intra-peritoneal tumor

growth and dissemination of human ovarian carcinoma cells in nude mice by anti-L1-cell adhesion

molecule monoclonal antibody treatment. Cancer Res., 66, 936-943.



Dow

nloaded from


19

27. Ouyang, H., Mou, L., Luk, C., Liu, N., Karaskova, J., Squire, J. and Tsao, M.S. (2000). Immortal human

pancreatic duct epithelial cell lines with near normal genotype and phenotype. Am. J. Pathol., 157, 1623-

1631.

28. Furukawa, T., Duguld, W.P., Rosenberg, L., Vlallet, J., Galloway, D.A. and Tsao, M.S. (1996). Long-

term culture and immortalization of epithelial cells from normal adult human pancreatic ducts

transfected by the E6E7 gene of human papilloma virus 16. Am. J. Pathol., 148, 1763-1770.

29. Liu, N., Furukawa, T., Kobari, M. and Tsao, M.S. (1998). Comparative phenotypic studies of duct

epithelial cell lines derived from normal human pancreas and pancreatic carcinoma. Am. J. Pathol., 153,

263-269.

30. Muerkoster, S.S., Werbing, V., Koch, D., Sipos, B., Ammerpohl, O., Kalthoff, H., Tsao, M.S., Folsch,

U.R. and Schafer, H. (2008). Role of myofibroblasts in innate chemoresistance of pancreatic carcinoma -

epigenetic downregulation of caspases. Int. J. Cancer, 123, 1751-1760.

31. Riedle, S., Kiefel, H., Gast, D., Bondong, S., Wolterink, S., Gutwein, P. and Altevogt, P. (2009). Nuclear

translocation and signalling of L1-CAM in human carcinoma cells requires ADAM10 and

presenilin/gamma-secretase activity. Biochem. J., 420, 391-402.

32. Stoeck, A., Schlich, S., Issa, Y., Gschwend, V., Wenger, T., Herr, I., Marmé, A., Bourbie, S., Altevogt, P.

and Gutwein, P. (2006). L1 on ovarian carcinoma cells is a binding partner for Neuropilin-1 on

mesothelial cells. Cancer Lett., 239, 212-226.

33. Wolterink, S., Moldenhauer, G., Fogel, M., Pfeifer, M., Lüttgau, S., Gouveia, R., Costa, J., Endell, J.,

Moebius, U. and Altevogt, P. (2010). Therapeutic antibodies to human L1CAM: Functional

characterization and application in a mouse model for ovarian carcinoma. Cancer Res.. 70, 2504-2515.

34. Tiwari, S., Egberts, J.H., Korniienko, O., Köhler, L., Trauzold, A., Glüer, C.C. and Kalthoff, H. (2011).

Assessment of anti-inflammatory tumor treatment efficacy by longitudinal monitoring employing

sonographic micro morphology in a preclinical mouse model. BMC Med. Imaging, 11: 15.

35. Mechtersheimer, S., Gutwein, P., Agmon-Levin, N., Stoeck, A., Oleszewski, M., Riedle, S., Postina, R.,

Fahrenholz, F., Fogel, M., Lemmon, V. and Altevogt, P. (2001). Ectodomain shedding of L1 adhesion

molecule promotes cell migration by autocrine binding to integrins. J. Cell Biol., 155, 661-673.



Dow

nloaded from


20

36. Fujita, H., Ohuchida, K., Mizumoto, K., Nagata, K., Yu, J., Kayashima, T., Cui, L., Manabe, T., Ohtsuka,

T. and Tanaka, M. (2010). alpha-smooth muscle actin expressing stroma promotes an aggressive tumor

biology in pancreatic ductal adenocarcinoma. Pancreas, Epub ahead of print.

37. Qian, J., Niu, J., Li, M., Chiao, P.J. and Tsao, M.-S. (2005). In vitro modeling of human pancreatic duct

epithelial cell transformation defines gene expression changes induced by k-ras oncogenic activation in

pancreatic carcinogenesis. Cancer Res., 65, 5045-5052.

38. Löhr, M., Müller, P., Zauner, I., Schmidt, C., Trautmann, B., Thévenod, F., Capellà, G., Farré, A., Liebe,

S. and Jesnowski, R. (2001). Immortalized bovine pancreatic duct cells become tumorigenic after

transfection with mutant k-ras. Virchows Arch., 438, 581-590.

39. Agbunag, C. and Bar-Sagi, D. (2004). Oncogenic k-ras drives cell cycle progression and phenotypic

conversion of primary pancreatic duct epithelial cells. Cancer Res., 64, 5659-5663.

40. Hingorani, S.R., Petricoin, E.F., Maitra, A., Rajapakse, V., King, C., Jacobetz, M.A., Ross, S., Conrads,

T.P., Veenstra, T.D., Hitt, B.A., Kawaguchi, Y., Johann, D., Liotta, L.A., Crawford, H.C., Putt, M.E.,

Jacks, T., Wright, C.V., Hruban, R.H., Lowy, A.M. and Tuveson, D.A. (2003). Preinvasive and invasive

ductal pancreatic cancer and its early detection in the mouse. Cancer Cell, 4, 437-450.

41. Imai. T., Horiuchi. A., Shiozawa. T., Osada, R., Kikuchi, N., Ohira, S., Oka, K. and Konishi, I. (2004).

Elevated expression of E-cadherin and alpha-, beta-, and gamma-catenins in metastatic lesions compared

with primary epithelial ovarian carcinomas. Hum. Pathol., 35, 1469-1476.

42. Kowalski, P.J., Rubin, M.A. and Kleer, C.G. (2003). E-cadherin expression in primary carcinoma of the

breast and its distant metastases. Breast Cancer Res., 5, R217-R222.

43. Rubin, M.A., Mucci, N.R., Figurski, J., Fecko, A., Pienta, K.J. and Day, M.L. (2001). E-cadherin

expression in prostate cancer: a broad survey using high-density tissue microarray technology. Hum.

Pathol., 32, 690-697.

44. Wells, A., Yates, C. and Shepard, C.R. (2008). E-cadherin as an indicator of mesenchymal to epithelial

reverting transitions during the metastatic seeding of disseminated carcinomas. Clin. Exp. Metastasis,

25, 621-628.



Dow

nloaded from


21

45 Gavert, N., Sheffer, M., Raveh, S., Spaderna, S., Shtutman, M., Brabletz, T., Barany, F., Paty, P.,

Notterman, D., Domany, E. and Ben-Ze'ev, A. (2007). Expression of L1-CAM and ADAM10 in human

colon cancer cells induces metastasis. Cancer Res., 67, 7703-7712.

46. Gavert, N., Vivanti, A., Hazin, J., Brabletz ,T. and Ben-Ze`ev, A. (2011). L1-mediated colon cancer cell

metastasis does not require changes in EMT and cancer stem cell markers. Mol. Cancer Res., 9, 14-24.

47. Hwang, R.F., Moore, T., Arumugam, T., Ramachandran, V., Amos, K.D., Rivera, A., Ji, B., Evans, D.B.

and Logsdon, C.D. (2008). Cancer-associated stromal fibroblasts promote pancreatic tumor progression.

Cancer Res., 68, 918-926.

48. Bao, S., Wu, Q., Li, Z., Sathornsumetee, S., Wang, H., LcLendon, R.E., Hjemeland, A.B. and Rich, J.N.

(2008). Targeting cancer stem cells through L1CAM suppreses glioma growth. Cancer Res., 68, 6043-

6048.

49. Cheng, L., Wu, Q., Huang, Z., Guryanova, O.A.; Huang, Q., Shou, W., Rich, J.N. and Bao, S, (2011).

L1CAM regulates DNA damage checkpoint response of glioblastoma stem cells through NBS1. EMBO,

30,800-813.

50. Kastan, M.B. and Bartek, J. (2004). Cell-cycle checkpoints and cancer. Nature, 432, 316-323.



Dow

nloaded from


22

Figure legends

Fig. 1: H6c7 cells exhibiting elevated L1CAM expression and EMT-associated alterations show a

decreased apoptosis sensitivity towards death ligands. (A-C) H6c7 cells were cultured in the absence

(H6c7mono) or presence of PMFs for 4 weeks (H6c7co) or (D,E) were stably transfected with a control vector

(H6c7-mock) or a vector encoding for L1CAM (H6c7-L1CAM) (A,D) Representative images of the staining of

cytokeratin, L1CAM, E-cadherin and vimentin in (A) H6c7mono and H6c7co cells as well as in (D) H6c7-mock

and H6c7-L1CAM cells. Scale bar = 20 µm. Arrows indicate E-cadherin negative and vimentin positive cells,

respectively. (B) H6c7mono and H6c7co cells as well as in (E) H6c7-mock and H6c7-L1CAM were either left

untreated or treated with TRAIL, TNF-α or the CD95 agonistic antibody CH-11. (C) H6c7co cells were treated

with control siRNA or L1CAM siRNA for 48 hours. Then, cells were either left untreated or treated with TRAIL

or CH-11. (B,C,E) After 24 hours, cells were analyzed for caspase-3/-7 activity. Means ±SD of treatment

induced caspase-3/-7 activity (n-fold of basal) from four to six independent experiments are shown. * p<0.05.

Fig. 2: PMFs increase tumorigenicity of H6c7 cells leading to tumor growth and metastasis in vivo. SCID

mice were intra-pancreatically inoculated with H6c7 cells alone (H6c7mono) or with H6c7 cells that were

cocultured for 4 weeks and inoculated with PMFs (H6c7co). (A) Representative ultrasound images of the

pancreatic area are shown from H6c7mono inoculated animals exhibiting no tumor (A1) or a tumor (A2) as well

as liver metastases (A3) of H6c7co inoculated animals. The arrows indicate the primary tumor and metastases.

Black areas within the tumor are fluid filled cysts. (B+C) Representative images from immunohistochemical

stainings with either control IgG1 or an antibody detecting cytokeratin, L1CAM, E-cadherin or vimentin in

tumor (B) and liver (C) sections of an H6c7co inoculated animal. The frame in the left pictures indicates the area

shown in pictures with higher magnification (right). Scale bar = 20 µm.

Fig. 3: L1CAM overexpressing H6c7 cells become tumorigenic in vivo and treatment with L1CAM-

antibodies inhibits tumor growth and metastasis of tumorigenic H6c7 cells (A) SCID mice were

orthotopically inoculated with H6c7-mock or H6c7-L1CAM cells. Representative ultrasound images of the

pancreatic area are shown from (A1) H6c7-mock inoculated animals exhibiting no primary tumor and from

(A2) H6c7-L1CAM inoculated animals exhibiting pancreatic tumor growth. (B) SCID mice were



Dow

nloaded from


23

orthotopically inoculated with H6c7 cells that were cocultured for 4 weeks and inoculated with PMFs

(H6c7co). Two months after inoculation, animals were randomized and weekly treated with 10 mg/kg body

weight control IgG2a or the L1CAM-antibody L1-9.3/2a. Representative ultrasound images of the

pancreatic area are shown from (B1) control IgG1 treated animals exhibiting tumor formation and from (B2)

L1-9.3/2a treated animals exhibiting no pancreatic tumor growth (w/o). The arrow indicates fluid filled cysts

within primary tumors (black areas).

Supplementary Fig. 1: H6c7 cells cocultured with PMFs exhibit elevated L1CAM expression and

EMT-associated alterations. H6c7 cells were cultured in the absence (H6c7mono) or presence of PMFs for

4 weeks (H6c7co). Whole cell lysates were analysed by western blotting for the expression of L1CAM, E-

cadherin and vimentin. HSP90 was analysed as loading control. One representative western blot is shown.

Supplementary Fig. 2: H6c7 cells cocultured with human PMFs show a decreased apoptosis sensitivity

towards death ligands. H6c7 cells were cultured in the absence (H6c7mono) or presence of human PMFs

(hPMFs) for 2 weeks (H6c7co). H6c7mono and H6c7co cells were either left untreated or treated with

TRAIL, TNF-α or the CD95 agonistic antibody CH-11. After 24 hours, cells were analysed for caspase-3/-7

activity. Means ±SD of treatment induced caspase-3/-7 activity (n-fold of basal) from three independent

experiments are shown. * p<0.05.

Supplementary Fig. 3: H6c7-L1CAM cells exhibit EMT-associated alterations. Whole cell lysates from

H6c7-mock and H6c7-L1CAM cells were analysed by western blotting for the expression of L1CAM, E-

cadherin and vimentin. HSP90 was analysed as loading control. One representative western blot is shown.



Dow

nloaded from


Table 1: PMFs increase tumorigenicity of H6c7 cells leading to tumor growth and

metastasis in vivo. SCID mice were orthotopically inoculated with H6c7 cells alone

(H6c7mono) or with H6c7 cells that were cocultured for 4 weeks and inoculated with PMFs

(H6c7co). (A) Number of animals exhibiting primary tumors and liver metastases. (B)

Volumes (in mm3) of resected pancreatic tumors from H6c7mono and H6c7co inoculated

animals. * indicates lesion detected only by high-resolution ultrasound.

A

no. of animals H6c7mono (% of animals) H6c7co (% of animals)

primary tumor 0/6 (0%) 7/8 (87,5%)

liver metastases 0/6 (0%) 6/8 (75 %)

B

Tumor H6c7mono H6c7co

1 0 80

2 0 8

3 0 48

4 0 12

5 2* 0

6 2* 8

7 0 60

8 - 8

Mean 0,57 28,00

SD 0,98 31,39



Dow

nloaded from


Table 2: L1CAM overexpressing H6c7 cells become tumorigenic in vivo. SCID mice were

intra-pancreatically inoculated with H6c7-mock or H6c7-L1CAM cells. (A) Number of

animals exhibiting primary tumors and liver metastases. (B) Volumes (in mm3) of resected

tumors from H6c7-mock and H6c7-L1CAM inoculated animals. * indicates lesion detected

only by high-resolution ultrasound.

A

no. of animals H6c7-mock (% of animals) H6c7-L1CAM (% of animals)

primary tumor 0/7 (0%) 5/7 (71,4 %)

liver metastases 1/7 (14,3%) 3/7 (42,9 %)

B

Tumor H6c7-mock H6c7-L1CAM

1 0 12

2 0 4

3 0 9

4 1* 15

5 0 6

6 0 0

7 0 1*

Mean 0,14 6,71

SD 0,37 5,58



Dow

nloaded from


Table 3: Treatment with L1CAM-antibodies inhibits tumor growth and metastasis of

tumorigenic H6c7 cells. SCID mice were intra-pancreatically inoculated with H6c7 cells that

were cocultured for 4 weeks and inoculated with PMFs (H6c7co). Two months after

inoculation, animals were randomized and weekly treated with 10 mg/kg body weight control

IgG2a or the L1CAM-antibody L1-9.3/2a. (A) Number of animals exhibiting primary tumors

and liver metastases. (B) Volumes (in mm3) of resected tumors from IgG2a and L1-9.2/2a

treated animals.

A

no. of animals H6c7co + IgG2a

(% of animals)

H6c7co + L1-9.3/2a

(% of animals)

primary tumor 6/7 (85,7 %) 0/6 (0%)

liver metastases 5/7 (71,4 %) 3/6 (50%)

B

Tumor H6c7co + IgG2a H6c7co + L1-9.3/2a

1 468 0

2 12 0

3 252 0

4 30 0

5 19 0

6 1 0

7 6 -

Mean 112,57 0

SD 180,39 0



Dow

nloaded from




Dow

nloaded from


myofibroblast-induced tumorigenicity of pancreatic ductal

Documents