CBP signaling inhibits human airway epithelial–mesenchymal transition and repair

Accepted Manuscript Title: Disruption of ␤-Catenin/CBP signaling inhibits human airway epithelial-mesenchymal transition and repair Author: Fatemeh Moheimani Hollis M. Roth Jennifer Cross Andrew T. Reid Furquan Shaheen Stephanie M. Warner Jeremy A. Hirota Anthony Kicic Teal S. Hallstrand Michael Kahn Stephen M. Stick Philip M. Hansbro Tillie-Louise Hackett Darryl A. Knight PII: DOI: Reference:

S1357-2725(15)30009-1 http://dx.doi.org/doi:10.1016/j.biocel.2015.08.014 BC 4686

To appear in:

The International Journal of Biochemistry & Cell Biology

Received date: Revised date: Accepted date:

17-3-2015 19-8-2015 19-8-2015

Please cite this article as: Moheimani, F., Roth, H. M., Cross, J., Reid, A. T., Shaheen, F., Warner, S. M., Hirota, J. A., Kicic, A., Hallstrand, T. S., Kahn, M., Stick, S. M., Hansbro, P. M., Hackett, T.-L., and Knight, D. A.,Disruption of rmbeta-Catenin/CBP signaling inhibits human airway epithelial-mesenchymal transition and repair, International Journal of Biochemistry and Cell Biology (2015), http://dx.doi.org/10.1016/j.biocel.2015.08.014 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Title page Disruption of β-Catenin/CBP signaling inhibits human airway epithelial-mesenchymal transition and repair Fatemeh Moheimania*, Hollis M. Rothb*, Jennifer Crossb, Andrew T Reida, Furquan Shaheenb, Stephanie M. Warnerb, Jeremy A. Hirotab, Anthony Kicicc-g, Teal S. Hallstrandh, Michael Kahni, Stephen M. Stickd-g, Philip M Hansbroa, Tillie-Louise Hackettb,c, Darryl A. Knighta,c

ip t

*

These authors contributed equally to this work.

a

School of Biomedical Sciences and Pharmacy, University of Newcastle, Callaghan, NSW, Australia;

c

UBC Centre for Heart Lung Innovation, University of British Columbia, Vancouver, Canada;

cr

b

Department of Anesthesiology, Pharmacology and Therapeutics, University of British Columbia,

us

Vancouver, Canada; d

Telethon Kids Institute, Centre for Health Research, The University of Western Australia, Nedlands,

an

6009, Western Australia, Australia; e

Department of Respiratory Medicine, Princess Margaret Hospital for Children, Perth, 6001, Western

Australia, Australia;

M

f

School of Paediatrics and Child Health, Centre for Health Research, The University of Western Australia,

Nedlands, Australia;

Centre for Cell Therapy and Regenerative Medicine, School of Medicine and Pharmacology, The

d

g

h

te

University of Western Australia, Nedlands, 6009, Western Australia, Australia; Department of Medicine, Division of Pulmonary and Critical Care, University of Washington, Seattle,

i

Ac ce p

Washington, USA;

Norris Comprehensive Cancer Center, Department of Biochemistry and Molecular Biology, and

Department of Molecular Pharmacology and Toxicology, University of Southern California, Los Angeles, California;

Corresponding author- refereeing and publication: Dr Fatemeh Moheimani, School of Biomedical Sciences and Pharmacy, Faculty of Health and Medicine, HMRI building, The University of Newcastle, Callaghan NSW 2308 Australia Tel: +61 2 40420363, e-mail: [email protected] Corresponding author-post publication: Prof Darryl A. Knight, School of Biomedical Sciences and Pharmacy, Faculty of Health and Medicine, MS605 Medical Sciences Building, The University of Newcastle, Callaghan NSW 2308 Australia Tel: +61 2 4921 7485, Fax: +61 2 4921 7403, e-mail: [email protected] 1 Page 1 of 33

Abstract The epithelium of asthmatics is characterized by reduced expression of E-cadherin and increased expression of the basal cell markers ck-5 and p63 that is indicative of a relatively undifferentiated

ip t

repairing epithelium. This phenotype correlates with increased proliferation, compromised wound healing and an enhanced capacity to undergo epithelial-mesenchymal transition (EMT). The

cr

transcription factor β-catenin plays a vital role in epithelial cell differentiation and regeneration,

us

depending on the co-factor recruited. Transcriptional programs driven by the β-catenin/CBP axis are critical for maintaining an undifferentiated and proliferative state, whereas the β-catenin/p300 axis is

an

associated with cell differentiation. We hypothesized that disrupting the β-catenin/CBP signaling axis would promote epithelial differentiation and inhibit EMT. We treated monolayer cultures of human

M

airway epithelial cells with TGFβ1 in the presence or absence of the selective small molecule ICG-001 to inhibit β-catenin/CBP signaling. We used western blots to assess expression of an EMT signature, CBP,

d

p300, β-catenin, fibronectin and ITGβ1 and scratch wound assays to assess epithelial cell migration.

te

Snai-1 and -2 expressions were determined using q-PCR. Exposure to TGFβ1 induced EMT, characterized

Ac ce p

by reduced E-cadherin expression with increased expression of α-smooth muscle actin and EDAfibronectin. Either co-treatment or therapeutic administration of ICG-001 completely inhibited TGFβ1induced EMT. ICG-001 also reduced the expression of ck-5 and -19 independent of TGFβ1. Exposure to ICG-001 significantly inhibited epithelial cell proliferation and migration, coincident with a down regulation of ITGβ1 and fibronectin expression. These data support our hypothesis that modulating the β-catenin/CBP signaling axis plays a key role in epithelial plasticity and function.

Keywords: airway epithelium, β-Catenin/CBP, ICG-001, epithelium-mesenchymal transition, wound repair

2 Page 2 of 33

1. Introduction The airway epithelium is the interface between the inhaled environment and the sub-mucosa and forms the structural barrier against inhaled exogenous agents. Damage to the epithelium triggers a cascade of inflammatory and cell signaling events that can lead to regeneration and/or repair.

ip t

Regeneration is the outcome of processes that returns the tissue to its normal structure and function. By contrast, repair regulates the stability of a tissue, but fails to restore full structural or functional

cr

capacity, and in some cases results in excessive wound healing that can lead to pathological remodeling

us

and fibrosis. Thus, full regeneration is critical in maintaining barrier integrity and normal function of the epithelium.

an

We and others showed that the airway epithelium of both children and adults with asthma is

M

relatively undifferentiated characterized by a significantly increased proportion of progenitor and basal cells and reduced expression of key junction proteins (1-5). This altered phenotype (3, 6) has potential

d

implications for many aspects of airway epithelial homeostasis including compromised barrier function

te

(7) and wound repair capacity (6, 8). Importantly, these abnormalities are maintained when the cells are

Ac ce p

grown at air-liquid interface (ALI) culture (9).

It is likely that this phenotypic immaturity also facilitates cell plasticity. In this context we have demonstrated that epithelial cells from asthmatics exhibit a greater capacity to undergo epithelialmesenchymal transition (EMT) in response to TGFβ1 (5). EMT is a process by which epithelial cells lose polarity and intercellular contacts, and assume a mesenchymal phenotype (10, 11). These phenotypic changes are paralleled by transcriptional repression of epithelial genes such as E-cadherin and ZO-1 coordinated with increased expression of mesenchymal proteins, including EDA-fibronectin (EDA-FN), vimentin, and α-smooth muscle actin (11). Thus, increased EMT would have a disruptive effect on the capacity to regenerate an effective epithelial barrier, which in turn facilitates a chronic cycle of inflammation and airway remodeling (12).

3 Page 3 of 33

β-catenin signaling is integral to many facets of epithelial development and differentiation (1317). When not bound to E-cadherin or activated, β-catenin is associated with a degradation complex, which keeps it phosphorylated and targeted for ubiquitination and proteolytic destruction. Disruption of this complex prevents β-catenin phosphorylation and promotes its nuclear translocation, where it forms

ip t

a complex with members of the TCF/LEF family of transcription factors (18). The specific gene program induced by β-catenin depends on the recruitment of transcriptional co-activators, including CREB

cr

binding protein (CBP) or its closely related homolog p300. Transcriptional programs driven by β-

us

catenin/CBP are critical for maintenance of an undifferentiated and proliferative state. Inhibiting the βcatenin/CBP interaction facilitates the binding of β-catenin to its alternate co-activator, p300, resulting

an

in expression of β-catenin/p300-dependent target genes associated with cell differentiation (19).

M

We previously demonstrated that the novel small molecule inhibitor ICG-001, which selectively inhibits the association between β-catenin and CBP (20) induces a mesenchymal-epithelial transition in

d

human fibroblasts (21). In this study, we aimed to elucidate whether this pathway was involved in EMT

te

of human tracheobronchial epithelial cells. We show that ICG-001 ameliorates and therapeutically reverses TGFβ1-induced EMT in human airway epithelial cells, inhibits expression of the basal cell

Ac ce p

markers cytokeratin (ck)-5, ck-14 and ck-19, but reduces expression of β1 integrin (ITGβ1) and fibronectin and suppresses epithelial cell wound repair. These data highlight the complexity of β-catenin signaling in determining epithelial cell fate and differentiation and the therapeutic potential of inhibiting the β-catenin/CBP pathway.

2. Materials and methods 2.1. Human airway epithelial cells and cell culture This study was approved by the Research Ethics Boards of the University of British Columbia and the University of Newcastle. Primary human airway epithelial cells (AECs) were isolated from non4 Page 4 of 33

asthmatic and asthmatic donor lungs deemed unsuitable for transplant, as described previously (1). In addition, cells were obtained by bronchial brushing of non-asthmatic and asthmatic subjects (22, 23). AECs were maintained in bronchial epithelial growth media (Lonza) and supplemented with antibioticsantimycotics (Gibco, Burlington, ON, Canada). Experiments were conducted on cells at passage 2 or 3.

ip t

Minimally-immortalized bronchial epithelial cells (HBEC6-KT) were generously provided by Dr. John Minna (24) and maintained in Keratinocyte Serum-Free Media (KSFM; Invitrogen) with growth

cr

supplements and antibiotics as described in on-line supplement. We have previously characterized

us

these cells for E-cadherin and actin expression and show that they form confluent monolayers with

an

defined intercellular junctions (25). Experiments were conducted on cells at passage 12 and 13.

2.2. Induction of EMT

M

Cells were cultured in 6-well tissue culture plates (BD Falcon, NJ) to reach 60-70% confluence. ICG-001 (10 μM) obtained from Dr. Michael Kahn (University of Southern California, CA) was then

te

d

added, followed by addition of recombinant human TGFβ1 treatment (10 ng/mL, Pepro Tech, NJ) 30 minutes later. Protein samples were obtained 72 hours after treatment (5).

Ac ce p

Therapeutic administration experiments were carried out when cells were at 60-70% confluence. TGFβ1 exposure was carried out as described above, with addition of ICG-001 (10 μM) at 12, 18, 24, or 48 hours after addition of TGFβ1. Protein samples were obtained 72 hours after addition of TGFβ1.

2.3. Western Blot

Cells were lysed in lysis-buffer (25 mM Tris, 150 mM NaCl, 5 mm EDTA, 1% triton X-100, 0.1% SDS, 1% Sodium deoxycholate, pH 7.6) containing a cocktail of protease inhibitors and centrifuged at 13,000 rpm for 10 min. The protein content of the supernatant was determined by the BCA protein assay method. Equal amounts of protein (10 µg) were added to SDS-PAGE buffer, boiled for 5 min and electrophoresed on a 4-15% gradient gel and then transferred to PVDF membranes. Membranes were 5 Page 5 of 33

incubated with Tris buffered saline with 0.5% tween-20 (TBST) containing 5% skim milk powder for 1 hour to block non-specific binding and then incubated with primary antibodies over night at 4° C. After washing in TBST, membranes were incubated with horseradish peroxidase (HRP)-conjugated secondary antibody in TBST for 1 Hour, washed and developed using the ECL detection system. The following

ip t

antibodies were used for immunoblotting: E-cadherin (mouse; Santa Cruz Biotechnology: sc-8426, 1 μg/mL), α-smooth muscle actin (mouse; Sigma-Aldrich: A5228), EDA-fibronectin (EDA-FN: mouse;

cr

Millipore: MAB 1940, 0.1 μg/mL), vimentin (rabbit; Abcam: ab-45939, 2 μg/mL), ZO-1 (rabbit; Santa Cruz

us

Biotechnology: sc-10804, 1 μg/mL), p300 (rabbit; Santa Cruz Biotechnology: sc-585, 1 μg/mL), CBP (rabbit; Santa Cruz Biotechnology: sc-369, 1 μg/mL), ITGβ1 (mouse; R & D: MAB 17783), plasma

an

fibronectin (mouse; R & D: MAB 1918), phospho-Smad3 S423+S425 (Rabbit; Abcam: ab52903, 0.015 μg/mL), Stat3 (rabbit; Abcam: ab15523, 0.1 μg/ml), p-Stat3 (Tyr705) (rabbit: Cell Signaling: 9145, 56

M

μg/ml), p-Stat3 (Ser727) (rabbit; Abcam: ab30647, 0.5 μg /ml), cytokeratin 14 (mouse; Abcam: ab-7800, 0.008 μg/mL), cytokeratin 5 (rabbit; Abcam: ab-52635, 0.002 μg/mL), cytokeratin 19 (rabbit; Abcam:

te

d

ab15463, 1 μg/ml), β-tubulin (mouse; Millipore: 05-661, 0.356 μg/mL), and hsp90 (mouse; BD Biosciences, Mississauga, ON, Canada, 0.25 μg/mL).

Ac ce p

All densitometry results were obtained using Image lab software, version 4.1 and have been normalized to β-tubulin expression levels, with the exception of ITGβ1 and plasma fibronectin results which were run in non-reducing, non-denaturing conditions and normalized to hsp90. Ck-14 and pSMAD3 protein expressions were also normalized to hsp90. All results were then further normalized to the vehicle control.

2.4. Gene expression analysis RNA was collected at 24 hours in all experiments and was extracted using an RNeasy Mini Kit (Qiagen, CA). Gene expression for Snai-1 and -2 was determined by Taqman real-time quantitative PCR

6 Page 6 of 33

according to the manufacturer’s instructions (Applied Biosystems). Refer to the on-line supplement for primers and conditions.

2.5. Wound Repair Kinetics

ip t

Epithelial cells were seeded in 6-well culture plate and grown to 70% confluence. After overnight quiescence in basal media lacking growth supplements, cells were incubated with mitomycin C (30 μM)

cr

for 2 hours to arrest cell proliferation. Cells were then washed in PBS and wounded in a cross-hatch

us

pattern using a p200 pipet tip, rinsed once and either exposed to DMSO (vehicle control) or ICG-001 (10 μM). Wounds were photographed at time 0 and after 4, 8, 24 and 48 hours and repair quantified as

M

2.6. Immunofluorescence Microscopy

an

percent wound closure using ImageJ software.

Epithelial cells were cultured on 8-well chamber slides (1x104 cells/well), fixed in cold ethanol for

te

d

15 minutes, re-hydrated and permeabilized in 0.2% Triton X-100 for 3 mins. After rinsing cells in PBS, cells were blocked with 5% goat serum at room temperature for 1 hour, and then stained with

Ac ce p

monoclonal mouse antibodies against human E-cadherin and EDA-FN for 2 hours at room temperature. After further washes in PBS, slides were exposed to Alexa Fluor-594 or -488-conjugated secondary goat anti-mouse antibody (Invitrogen). Cells were counterstained with 4,6-diamidino-2-phenylindole (DAPI) to visualize nuclei. All images were obtained using a Leica AOBSTM SP2 confocal microscope (Leica Microsystems GmbH, Heidelberg, Germany).

2.7. Statistics Data are presented as the mean ± SEM. Student’s unpaired t-test was used for pair-wise comparisons and a one-way analysis of variance with Bonferroni post-test was completed for comparison of group data. P values less than 0.05 were considered statistically significant. Statistical 7 Page 7 of 33

analysis was carried out using GraphPad Prism Software (version 5.04). Each experiment was repeated three times or more (n ≥ 3).

3. Results

ip t

3.1. ICG-001 prevents TGFβ1-induced EMT To evaluate the influence of ICG-001 on TGFβ1-induced EMT, cell morphology and expression of

cr

relevant epithelial and mesenchymal markers were evaluated. Following exposure to TGFβ1 for 72

us

hours, primary AEC from both asthmatics and non-asthmatics exhibited a marked alteration in cell morphology, changing from the characteristic organized ‘cobblestone’ appearance of epithelial cell

an

monolayers to an elongated fibroblast-like phenotype, indicative of EMT. Fig. 1A represents these alterations in morphology of HBEC6-KT cells. As the particular subset of β-catenin target genes

M

expressed in a given context depends on the specific co-activator recruited to the β-catenin/TCF complex, we examined whether β-catenin/p300- or β-catenin/CBP-dependent gene transcription

te

d

mediates EMT in this model. Our data indicate that ICG-001 inhibits the morphological changes induced by TGFβ1 (Fig. 1A). Furthermore, co-treatment of cells with TGFβ1 and ICG-001 suppressed TGFβ1-

Ac ce p

induced gene expression of CBP and EP300 (Fig. 1B). However this effect on CBP and EP300 protein level was not observed (Fig. 1C).

To corroborate these observations, we assessed changes in expression of epithelial (E-cadherin) and mesenchymal markers (α-SMA and EDA-FN) by Western blot. Following exposure to TGFβ1, AECs exhibited down-regulation of E-cadherin relative to cells under control conditions (Fig. 2A), whereas expression of α-SMA and EDA-FN dramatically increased (Fig. 2B-C). Simultaneous exposure to both ICG001 and TGFβ1 led to preservation of E-cadherin expression and suppression of α-SMA and EDA-FN (Fig. 2A-C). Exposure to ICG-001 alone had no effect on expression of these markers. We also investigated the impact of ICG-001 on the expression and localization of E-cadherin and EDA-FN during TGFβ1induced EMT using confocal microscopy (Fig. 2D). As expected, exposure to TGFβ1 resulted in down8 Page 8 of 33

regulation of E-cadherin relative to control conditions, with a concomitant increase in expression of EDA-FN. Exposure to both ICG-001 and TGFβ1 led to maintenance of E-cadherin immunoreactivity along cell borders and no increase in EDA-FN expression (Fig. 2D).

ip t

To determine potential mechanisms by which ICG-001 modifies TGFβ1 signaling to inhibit EMT, we examined Smad3 phosphorylation, total Stat3 and Stat3 phosphorylation at Tyr705 or Ser727, and

cr

expression of the E-cadherin transcriptional repressors Snai-1 and Snai-2. Somewhat surprisingly, ICG001 enhanced Smad3 phosphorylation over that of TGFβ1 alone (Fig. 3A). In contrast, ICG-001 treatment

us

alone or in combination with TGFβ1 had no effect on total Stat3 (data are not shown) or p-Stat3

an

(Ser727) or (Tyr705) compared with TGFβ1 (Fig. 3B-C). However, exposure of AEC to ICG-001 significantly represses expression of Snai-1 over a 24 hours period (Fig. 3D). In contrast, expression of

3.2. ICG-001 reverses established EMT

d

M

Snai-2 was very low and not influenced by exposure to ICG-001 (Fig. 3D).

te

While many compounds have been shown to prevent EMT, relatively few have either been

Ac ce p

studied or been shown to have the capacity to also reverse this process when administered therapeutically. To assess this, we added ICG-001 at varying periods of time after the induction of EMT by TGFβ1. Fig. 4 shows that addition of ICG-001 between 12 hours and 48 hours after TGFβ1 effectively reverses the TGFβ1-induced decrease in E-cadherin expression (Fig. 4A). Similarly, this treatment regimen reverses the TGFβ1-induced increased in EDA-FN and vimentin expression (Fig. 4B-C). Taken together, these data show that modifying the specific co-activator recruited to the β-catenin/TCF complex using the small molecule ICG-001 significantly impacts the molecular reprogramming induced by TGFβ1.

9 Page 9 of 33

3.3. ICG-001 inhibits repair of airway epithelial wounds in vitro Activation of β-catenin has been previously shown to play an important role in alveolar epithelial repair (26, 27). Thus, we next examined whether disrupting β-catenin/CBP-dependent gene transcription is involved in repair of airway epithelial cells using a scratch wound model. Treatment with

ip t

ICG-001 significantly inhibited wound healing after mechanical injury (Fig. 5A). To determine whether ICG-001 inhibited cell migration and/or proliferation, we treated cells with Mitomycin C (30 μM) to

cr

induce cell cycle arrest. Exposure to Mitomycin C significantly inhibited but did not completely prevent

us

wound healing, whereas addition of both Mitomycin C and ICG-001 together completely prevented wound repair (Fig. 5B). We then sought to determine the potential mechanisms underlying the

an

inhibitory effect of ICG-001 on epithelial cell migration. Since AECs require β1 integrins (ITGβ1) to migrate and a matrix of fibronection (FN) to migrate on (8), we examined expression of these molecules

M

before and after TGFβ1 exposure and in the presence or absence of ICG-001. As shown in Fig. 5C, expression of both FN and ITGβ1 was increased by exposure to TGFβ1 and prevented by ICG-001. These

te

d

data suggest that the primary mode of epithelial repair of mechanical wounds involves both

Ac ce p

proliferation and migration and that β-catenin/CBP-dependent gene transcription is involved in both.

3.4. Effect of ICG-001 on expression of basal cell markers Given the potent effect of ICG-001 on proliferation and its capacity to modulate differentiation, we next examined its effect on expression of basal cell markers ck-5, ck-14 and ck-19. As shown in Fig. 6, expression of ck-5 (Fig. 6A) and ck-19 (Fig. 6C) were significantly suppressed by ICG-001 irrespective of the presence of TGFβ1. There was a similar trend for expression of ck-14 (Fig. 6B), but this just failed to reach statistical significance. These findings suggest that blocking CBP/β-catenin interactions may contribute to restoration of basal cell homeostasis in settings of basal cell hyperplasia such as asthma and that expression of these cytokeratins are not TGFβ1-dependent.

10 Page 10 of 33

4. Discussion Molecular reprogramming of epithelial cells through EMT has been well described in normal organ development (type I), wound healing and fibrosis (type II) and cancer (type III). In the current study, we show that by blocking the interaction between β-catenin and its transcriptional co-activator

ip t

CBP, we can both prevent and reverse TGFβ1-induced EMT in primary cultures of human AECs. We found that ICG-001-mediated inhibition of EMT was related to down-regulation of Snai-1, was not due

cr

to decreased Smad3 activation and was independent of Stat3 activation. Exposure of AEC to ICG-001

us

also significantly inhibited wound repair and this appeared to occur via effects on proliferation and migration as well as by inhibiting expression of β1 integrins and the production of the ECM protein

an

fibronectin. These data suggest that appropriate β-catenin signaling is crucial for epithelial cell repair through the recruitment of specific cofactors, and that manipulating these interactions has profound

M

effects on epithelial cell homeostasis.

There is mounting evidence from both in vivo and in vitro models that EMT contributes to airway

te

d

and lung remodeling. For example, using a mouse model of HDM-induced allergic airway disease, Johnson and coworkers show that EMT both drives phenotypic changes in airway epithelial cells and

Ac ce p

facilitates their migration to the sub-mucosa (28). Moreover, the authors demonstrated a significant overlap between nuclear staining of phospho-Smad3 and the transcriptional repressor of E-cadherin, Snai-1 in epithelial cells suggesting this complex is involved in promoting EMT during HDM-induced inflammation. The induction of EMT is regulated by complex networks of pro-mesenchymal transcription factors, many of which are responsive to the direct actions of TGFβ1. To this end, we have demonstrated that exposure of A549 cells to TGF β1 induced a potent EMT response (29) and more recently showed that EMT occurred in basal cells population of differentiated ALI-primary AEC cultures from asthmatic and non-asthmatic donors (5) that is equivalent to monolayer cultures of primary AECs used in this study. In this study, we reproduce the same findings in airway epithelial cells from a different cohort of asthmatics and show that Snai-1, but not Snai-2 is increased by TGF β1. 11 Page 11 of 33

β-catenin signaling is integral to many facets of epithelial development and differentiation. To generate a transcriptionally active complex, β-catenin recruits the transcriptional co-activators, CBP or its closely related homolog p300, as well as other transcriptional machinery (30, 31). Traditionally, p300 and CBP have been regarded as interchangeable. However, the differential effects of β-catenin signalling

ip t

on target genes may in fact be determined by different co-activator usage. Thus, transcription driven by β-catenin/CBP appears critical for an undifferentiated and pro-proliferative state, whereas β-

cr

catenin/p300 initiates a program that drives differentiation. Data generated by our group showed that

us

the β-catenin/p300 axis plays a central role in epithelial regeneration following exposure to bleomycin by binding to and regulating transcriptional activity of β-catenin (21). In the current study, we

an

interrogate this process further and show that blocking the interaction between CBP and β-catenin prevents TGFβ1-induced EMT in human bronchial epithelial cells, as reflected by restoration of E-

M

cadherin expression coincident with a reduction in expression of α-SMA and EDA-FN (Fig. 2). Critically,

d

ICG-001 prevented the expression of an EMT signature even when administered 48 hours after TGFβ1

te

addition (Fig. 4). Our data indicate that the inhibitory effect on EMT by ICG-001 was likely mediated by a reduction in expression of the E-cadherin transcriptional repressor Snai-1, given the impact of ICG-001

Ac ce p

on TGFβ1-induced expression (Fig. 3). Expression of phospho-Smad3 was increased by ICG-001 suggesting the inhibitory effect on Snai-1 is not Smad3-dependent. Saitoh et al. reported that TGFβ1 dissociates phospho-Stat3 from its inhibitor; protein inhibitor of activated Stat3 (PIAS3), and free phospho-STAT3, in cooperation with Ras, subsequently increases Snai-1 expression (32). While we did not investigate Ras signaling per se, our data showing that TGFβ1 reduced p-Stat (Tyr705) expression suggests that the inhibitory effect of ICG-001 on TGFβ1-induced Snai-1 expression is not dependent on Stat3 activation either. The reasons underlying these disparate findings are unknown but may relate to cell type since Saitoh and colleagues used human cancer cell-lines Panc-1 and Hela, whereas we used primary cultures and minimally immortalized AEC.

12 Page 12 of 33

Our findings are in general agreement with previous studies showing that targeted inhibition of β-catenin/CBP signaling with this small molecule inhibitor attenuates ECM expression, reduces myofibroblast activation, and inhibits renal and lung fibrosis in vivo (33). The action of ICG-001 in disrupting β-catenin-mediated gene transcription is unique as it selectively antagonizes the β-

ip t

catenin/CBP interaction by binding to CBP rather than β-catenin and it does so without interfering with β-catenin/p300 interactions. Thus, while p300/β-catenin signaling is involved in initiating normal cell

cr

differentiation, activation of the CBP/β-catenin pathway initiates a de-differentiation signaling cascade

us

that also directly regulates the expression of a large number of pro-fibrotic genes. By selectively

thereby leading to amelioration of de-differentiation.

an

targeting CBP/β-catenin signaling, ICG-001 effectively inhibits the expression of these fibrogenic genes,

Activation of β-catenin is strongly associated with epithelial cell proliferation and migration. The

M

results from our study suggest that the hyperproliferative and migratory phenotype is driven by β-

d

catenin/CBP since ICG-001 completely inhibited epithelial wound repair and this effect was mediated

te

through effects on proliferation and migration (Fig. 5). These findings are at odds with Zemans and colleagues (26) who showed that ICG-001 accelerated epithelial repair following damage by neutrophil

Ac ce p

transmigration and scratch wounding. The reasons underlying these disparate findings are not readily apparent, although it is worth noting that Zemans et al. used Calu-3 cells, a neoplastic epithelial cell line, whereas we used primary and minimally immortalized airway epithelial cells. Consistent with the findings presented herein, we and others have shown that preventing the association of β-catenin with CBP terminates proliferation and induces a differentiation program in respiratory epithelial cells. By inhibiting epithelial cell migration, the impact of disrupting the β-catenin/CBP interaction extends further, given that epithelial cancers often acquire a mesenchymal and migratory phenotype to metastasis. We have previously shown that the epithelium of asthmatics contains an expanded population of basal cells characterized by expression of ck-5 and -14 (2, 5). Intriguingly, despite TGFβ1 itself having 13 Page 13 of 33

no effect on expression of ck-5, -14 or other epithelial cytokeratins such as ck-19, ICG-001 inhibited their expression (Fig. 6). These data further suggests that CBP/β-catenin signaling promotes a less differentiated epithelial phenotype. These data also suggest that ICG-001 does not interfere with TGFβ β1 signaling in regulating expression of these intermediate filament proteins. This hypothesis is supported

ip t

by the lack of effect of ICG-001 on TGFβ1-induced Smad-3 phosphorylation (Fig. 3A). Our findings are seemingly at odds with Zhou et al. (34), who showed that TGFβ1-induced αSMA expression in rat lung

cr

epithelial cells was dependent on the formation of a complex of Smad3, β-catenin and CBP. While we

us

did not examine the formation of multi-factor complexes in our model system, Zhou and colleagues did not examine Smad3 phosphorylation in rat lung epithelial cells, leading us to speculate that ICG-001

an

does not bind at or near Ser-423/425 of the Smad molecule and inhibits the pro-fibrotic actions of TGFβ1 in a Smad3 independent manner.

M

In conclusion, we provide strong evidence that TGFβ1-induced EMT in human AECs is β-

d

catenin/CBP dependent. This pathway is involved in maintaining epithelial cells in a relatively undifferentiated state and is also critical for a pro-proliferative and migratory phenotype during

te

epithelial repair (35, 36). Our findings support previous studies establishing a key role for β-catenin in

Ac ce p

epithelial repair and differentiation and suggest that specific targeting of β-catenin binding partners may enhance the capacity of the epithelium to properly regenerate to restore an effective barrier.

Conflict of interest statement

There is no conflict of interest.

Acknowledgments This research was supported by funding from the Canadian Institutes of Health Research (MOP119374 ); the National Health and Medical Research Council (NHMRC: APP 1064405), Australia; and Early Career Researcher (ECR) Grant to FM, the University of Newcastle, Australia. DAK was supported 14 Page 14 of 33

by the Canada Research Chair (2004-2012) program and Michael Smith Foundation for Health Research

Ac ce p

te

d

M

an

us

cr

ip t

(2007-2012). The authors kindly thank Dr. John Minna for providing us with HBEC6-KT cells.

15 Page 15 of 33

Ac ce p

te

d

M

an

us

cr

ip t

References 1. Hackett TL, Shaheen F, Johnson A, Wadsworth S, Pechkovsky DV, Jacoby DB, et al. Characterization of side population cells from human airway epithelium. Stem Cells. 2008;26(10):257685. 2. Kicic A, Sutanto EN, Stevens PT, Knight DA, Stick SM. Intrinsic Biochemical and Functional Differences in Bronchial Epithelial Cells of Children with Asthma. Am J Respir Crit Care Med. 2006;174(10):1110-8. 3. de Boer WI, Sharma HS, Baelemans SM, Hoogsteden HC, Lambrecht BN, Braunstahl GJ. Altered expression of epithelial junctional proteins in atopic asthma: possible role in inflammation. Canadian journal of physiology and pharmacology. 2008;86(3):105-12. 4. Hackett TL, Shaheen F, Johnson A, Wadsworth S, Petchkovski DV, Jacoby DB, et al. Characterization of Side Population Cells from Human Airway Epithelium. Stem Cells. 2008. 5. Hackett TL, Warner SM, Stefanowicz D, Shaheen F, Pechkovsky DV, Murray LA, et al. Induction of epithelial-mesenchymal transition in primary airway epithelial cells from patients with asthma by transforming growth factor-beta1. American journal of respiratory and critical care medicine. 2009;180(2):122-33. 6. Stevens PT, Kicic A, Sutanto EN, Knight DA, Stick SM. Dysregulated repair in asthmatic paediatric airway epithelial cells: the role of plasminogen activator inhibitor-1. Clinical and experimental allergy : journal of the British Society for Allergy and Clinical Immunology. 2008;38(12):1901-10. 7. Xiao C, Puddicombe SM, Field S, Haywood J, Broughton-Head V, Puxeddu I, et al. Defective epithelial barrier function in asthma. The Journal of allergy and clinical immunology. 2011;128(3):549-56 e1-12. 8. Kicic A, Hallstrand TS, Sutanto EN, Stevens PT, Kobor MS, Taplin C, et al. Decreased Fibronectin Production Significantly Contributes to Dysregulated Repair of Asthmatic Epithelium. Am J Respir Crit Care Med. 2010;181(9):889-98. 9. Hackett T-L, Singhera GK, Shaheen F, Hayden P, Jackson GR, Hegele RG, et al. Intrinsic Phenotypic Differences of Asthmatic Epithelium and its Inflammatory Responses to RSV and Air Pollution. Am J Respir Cell Mol Biol. 2011;45:1090-100. 10. Hackett TL. Epithelial-mesenchymal transition in the pathophysiology of airway remodelling in asthma. Current opinion in allergy and clinical immunology. 2012;12(1):53-9. 11. Kalluri R. EMT: when epithelial cells decide to become mesenchymal-like cells. The Journal of clinical investigation. 2009;119(6):1417-9. 12. Knight DA, Stick SM, Hackett TL. Defective function at the epithelial junction: a novel therapeutic frontier in asthma? The Journal of allergy and clinical immunology. 2011;128(3):557-8. 13. Ritchie TC, Zhou W, McKinstry E, Hosch M, Zhang Y, Nathke I, et al. Developmental expression of catenins and associated proteins during submucosal gland morphogenesis in the airway. Experimental lung research. 2001;27(2):121-41. 14. Schmidt-Ott KM, Barasch J. WNT/beta-catenin signaling in nephron progenitors and their epithelial progeny. Kidney international. 2008;74(8):1004-8. 15. Driskell RR, Goodheart M, Neff T, Liu X, Luo M, Moothart C, et al. Wnt3a regulates Lef-1 expression during airway submucosal gland morphogenesis. Developmental biology. 2007;305(1):90102. 16. Filali M, Liu X, Cheng N, Abbott D, Leontiev V, Engelhardt JF. Mechanisms of submucosal gland morphogenesis in the airway. Novartis Foundation symposium. 2002;248:38-45; discussion -50, 277-82. 17. Zhang Y, Goss AM, Cohen ED, Kadzik R, Lepore JJ, Muthukumaraswamy K, et al. A Gata6-Wnt pathway required for epithelial stem cell development and airway regeneration. Nature genetics. 2008;40(7):862-70. 18. Hart MJ, de los Santos R, Albert IN, Rubinfeld B, Polakis P. Downregulation of beta-catenin by human Axin and its association with the APC tumor suppressor, beta-catenin and GSK3 beta. Current biology : CB. 1998;8(10):573-81. 16 Page 16 of 33

Ac ce p

te

d

M

an

us

cr

ip t

19. Lenz HJ, Kahn M. Safely targeting cancer stem cells via selective catenin coactivator antagonism. Cancer science. 2014;105(9):1087-92. 20. Ma H, Nguyen C, Lee KS, Kahn M. Differential roles for the coactivators CBP and p300 on TCF/beta-catenin-mediated survivin gene expression. Oncogene. 2005;24(22):3619-31. 21. Henderson WR, Chi EY, Ye X, Nguyen C, Tien Y-t, Zhou B, et al. Inhibition of Wnt/B-catenin/CREB binding protein (CBP) signaling reverses pulmonary fibrosis. Proc Natl Acad Sci,. 2010;107(32):14309-14. 22. Hallstrand TS, Wurfel MM, Lai Y, Ni Z, Gelb MH, Altemeier WA, et al. Transglutaminase 2, a novel regulator of eicosanoid production in asthma revealed by genome-wide expression profiling of distinct asthma phenotypes. PLoS One.5(1):e8583. 23. Hallstrand TS, Wurfel MM, Lai Y, Ni Z, Gelb MH, Altemeier WA, et al. Transglutaminase 2, a novel regulator of eicosanoid production in asthma revealed by genome-wide expression profiling of distinct asthma phenotypes. PLoS One. 2010;5(1):e8583. 24. Ramirez RD, Sheridan S, Girard L, Sato M, Kim Y, Pollack J, et al. Immortalization of human bronchial epithelial cells in the absence of viral oncoproteins. Cancer Res. 2004;64(24):9027-34. 25. Hirota JA, Alexis NE, Pui M, Wong S, Fung E, Hansbro P, et al. PM10-stimulated airway epithelial cells activate primary human dendritic cells independent of uric acid: application of an in vitro model system exposing dendritic cells to airway epithelial cell-conditioned media. Respirology. 2014;19(6):88190. 26. Zemans RL, Briones N, Campbell M, McClendon J, Young SK, Suzuki T, et al. Neutrophil transmigration triggers repair of the lung epithelium via B-catenin signaling. Proc Natl Acad Sci,. 2011. 27. Sun Y, Zhang J, Ma L. alpha-catenin. Cell Cycle. 2014;13(15):2334-9. 28. Johnson JR, Roos A, Berg T, Nord M, Fuxe J. Chronic Respiratory Aeroallergen Exposure in Mice Induces Epithelial-Mesenchymal Transition in the Large Airways. PLoS ONE.6(1):e16175. 29. Murray LA, Hackett TL, Warner SM, Shaheen F, Argentieri RL, Dudas P, et al. BMP-7 does not protect against bleomycin-induced lung or skin fibrosis. PloS one. 2008;3(12):e4039. 30. Kahn M. Can we safely target the WNT pathway? Nature reviews Drug discovery. 2014;13(7):513-32. 31. Ring A, Kim YM, Kahn M. Wnt/catenin signaling in adult stem cell physiology and disease. Stem cell reviews. 2014;10(4):512-25. 32. Saitoh M, Endo K, Furuya S, Minami M, Fukasawa A, Imamura T, et al. STAT3 integrates cooperative Ras and TGF-beta signals that induce Snail expression. Oncogene. 2015. 33. Hao S, He W, Li Y, Ding H, Hou Y, Nie J, et al. Targeted Inhibition of B-Catenin/CBP Signaling Ameliorates Renal Interstitial Fibrosis. Journal of the American Society of Nephrology. 2011;22(9):164253. 34. Zhou B, Liu Y, Kahn M, Ann DK, Han A, Wang H, et al. Interactions between beta-catenin and transforming growth factor-beta signaling pathways mediate epithelial-mesenchymal transition and are dependent on the transcriptional co-activator cAMP-response element-binding protein (CREB)-binding protein (CBP). The Journal of biological chemistry. 2012;287(10):7026-38. 35. Teo J-L, Kahn M. The Wnt signaling pathway in cellular proliferation and differentiation: A tale of two coactivators. Advanced Drug Delivery Reviews.62(12):1149-55. 36. McMillan M, Kahn M. Investigating Wnt signaling: a chemogenomic safari. Drug discovery today. 2005;10(21):1467-74.

17 Page 17 of 33

Figure legends Fig. 1. Modulating β-Catenin/CBP interaction by ICG-001 inhibits TGFβ1-induced EMT, CBP and p300 gene expression. Epithelial cells were incubated with ICG-001 (10 μM) or DMSO (vehicle control) in the

ip t

presence or absence of recombinant human TGFβ1 treatment (10 ng/mL). (A) Phenotypic changes in HBEC6-KT cells after 72 hours treatment. An elongated fibroblast-like phenotype is indicated with an

cr

arrow. Scale bar is equal to 100 μm. (B) Gene expression of CBP and EP300 in primary airway epithelial

us

cells (AECs) after 24 hours treatment, using quantitative RT-PCR. (C) Representative immunoblots and densitometric analysis of CBP protein (265 kDa) and EP 300 protein (300 kDa) in AECs after 72 hours

an

treatment. Values were normalized to β-tubulin (50 kDa) as a loading control. C and D indicate control

M

(media only) and DMSO (vehicle control).

Fig. 2. Inhibiting β-Catenin/CBP interaction by ICG-001 induces epithelial cell marker and suppresses

te

d

mesenchymal markers in TGFβ1-induced EMT of primary human airway epithelial cells (AECs). AECs were treated with ICG-001 (10 μM) or DMSO (vehicle control) with or without recombinant human

Ac ce p

TGFβ1 treatment (10 ng/mL) for 72 hours. Densitometric quantification of immunoblots for (A) Ecadherin, (B) α-SMA and (C) EDA-FN proteins. Values were normalized to β-tubulin as a loading control. Data are from at least four different patients. * indicates data significantly different compared to TGFβ1 + DMSO treated results and # indicates data significantly different from vehicle control data. (D) Immunofluorescence staining for EDA-FN (green), E-cadherin (red) and nuclei (stained with 4,6diamidino-2-phenylindole; blue). Scale bar is equal to 20 μm.

Fig. 3. Disturbing β-Catenin/CBP interaction by ICG-001 has no effect on TGFβ1-induced Smad-3 or Stat3 signalling but suppresses TGFβ1-induced Snai-1 gene expression. Epithelial cells were incubated with ICG-001 (10 μM) or DMSO (vehicle control) with or without TGFβ1 treatment (10 ng/mL). 18 Page 18 of 33

Representative immunoblots of total cell lysate and densitometry analysis of (A) p-Smad3 (48 kDa), (B) p-Stat3 (Ser727: 88 kDa) and (C) p-Stat3 (Tyr705: 86 kDa) after 72 hours. Values were normalized to hsp90 (90 kDa) or β-tubulin (50 kDa) as a loading control. (D) Snai-1 and Snai-2 gene expression after 24 hours of treatment, using quantitative polymerase chain reaction. These data obtain from n=4

ip t

experiments. * indicates data significantly different compared to TGFβ1 + DMSO treated results, and #

cr

indicates data significantly different from vehicle control data.

us

Fig. 4. ICG-001 reverses established EMT in primary AECs. Primary AECs were incubated with ICG-001 between 12 hours and 48 hours after TGFβ1-induced EMT. Representative immunoblot and

an

densitometry analysis of (A) E-cadherin (120 kDa), (B) EDA-FN (260 kDa) and (C) vimentin (54 kDa). Values were normalized to β-tubulin (50 kDa) as a loading control. * indicates data significantly different

M

compared to TGFβ1 + DMSO treated results, and # indicates data significantly different from vehicle

te

d

control data.

Fig. 5. ICG-001 inhibits repair of airway epithelial wounds through inhibition of β1 integrin and

Ac ce p

fibronectin. (A) Photographs of confluent monolayer cultures of HBEC6-KT cells pre-treated with mitomycin C (Mit-C) at time of scratch wounding (0 h) and after 4, 8, 24 and 48 hours in the presence of ICG-001 (10 μM). (B) Analysis of wound repair in n = 4 independent experiments at 0, 4, 8, 24 and 48 hours after scratch wounding, using ImageJ software. Wound repair was calculated as percent area reconstituted compared with the initial wound area (time 0). Scale bar is equal 100 μm. (C) Densitometry analysis of immunoblots for β1 integrin (ITGβ1) and fibronectin (FN) in AECs after incubation with ICG-001 (10 μM) or DMSO (vehicle control) in the presence or absence of recombinant human TGFβ1 treatment (10 ng/mL). * indicates data significantly different compared to TGFβ1 + DMSO treated results, and # indicates data significantly different from vehicle control data.

19 Page 19 of 33

Fig. 6. Modulating β-Catenin/CBP interaction by ICG-001 regulates markers of epithelial cell differentiation. Representative results from densitometric analysis for (A) cytokeratin-5, (B) cytokeratin14 and (C) cytokeratin-19 show the effect of ICG-001 treatment on primary cultures of human AECs. Data represents four different patients. * indicates data significantly different compared to TGF-β1 +

cr

ip t

DMSO treated results; # indicates data significantly different from vehicle control data.

Ac ce p

te

d

M

an

us

Fig. 1

20 Page 20 of 33

21

Page 21 of 33

d

te

Ac ce p us

an

M

cr

ip t

Ac ce p

te

d

M

an

us

cr

ip t

Fig. 2

22 Page 22 of 33

23

Page 23 of 33

d

te

Ac ce p us

an

M

cr

ip t

Ac ce p

te

d

M

an

us

cr

ip t

Fig. 3

24 Page 24 of 33

25

Page 25 of 33

d

te

Ac ce p us

an

M

cr

ip t

26

Page 26 of 33

d

te

Ac ce p us

an

M

cr

ip t

Fig.

Ac ce p

te

d

M

an

us

cr

ip t

4

27 Page 27 of 33

28

Page 28 of 33

d

te

Ac ce p us

an

M

cr

ip t

Ac ce p

te

d

M

an

us

cr

ip t

Fig. 5

29 Page 29 of 33

30

Page 30 of 33

d

te

Ac ce p us

an

M

cr

ip t

Ac ce p

te

d

M

an

us

cr

ip t

Fig. 6

31 Page 31 of 33

32

Page 32 of 33

d

te

Ac ce p us

an

M

cr

ip t

33

Page 33 of 33

d

te

Ac ce p us

an

M

cr

ip t