Inhibitory Effects of Flufenamic Acid on YAP-TEAD Activity through the Hippo Signaling Pathway in Murine Pancreatic Adenocarcinoma Cells (LTPA)

Robin Hu

Abstract
Pancreatic adenocarcinoma, accounting for about 85% of pancreatic cancer cases, has poor prognosis and survival rates; novel therapies are needed [1,2]. A core pathway in cancer development is the Hippo Signaling Pathway: a signal cascade that inhibits protein kinases such as the Yes-Associated Protein (YAP) and transcription cofactors such as TEAD, to control organ size and growth through regulation of cell proliferation and apoptosis [3,4]. Deactivation of the Hippo Signaling Pathway leads to uncontrolled cell proliferation, and hence cancer. Previous studies have identified Flufenamic Acid, an anti-inflammatory medicine, as an effective drug in modulating growth in aortic smooth muscle cells [5]. This study illustrates that Flufenamic Acid is also effective in murine pancreatic adenocarcinoma cells (LTPA cell line) by inhibiting the YAP- TEAD interactions that are key to the deactivation of the Hippo Signaling Pathway. This study confirms the efficacy of Flufenamic Acid in inhibiting YAP-TEAD activity, indicating the pathway’s direct role in cancer pathogenesis. YAP-TEAD complexes are promising molecular targets and Flufenamic Acid has clinical potential for pancreatic cancer treatment development.
Introduction
Pancreatic cancer is the fourth most common cause of death from cancer [1]. Because there are no specific, cost-effective tests that can easily and reliably detect early-stage pancreatic cancers, pancreatic cancer is difficult to diagnose [2]. Symptoms normally do not appear until later stages when the cancer has already progressed to a stage that makes the few options available impossible [2]. While more therapies and clinical studies are being implemented to improve patient outcomes and find a cure for this difficult-to-treat cancer, novel treatments are sorely needed.
The Hippo Signaling Pathway is a key signaling cascade involved in the regulation of organ size and tissue growth [3,4]. Originally discovered in Drosophila, its core protein regulators are highly conserved in mammals [4]. The pathway consists of a protein kinase cascade where the
transcription coactivators YAP and TAZ are phosphorylated, sequestered in the cytoplasm and degraded [7]. However, once the Hippo Signaling Pathway is dysregulated, YAP and TAZ are dephosphorylated and translocate into the nucleus where they interact with TEAD1-4 and other transcription factors to induce gene expression [7] (Figure 1). The YAP-TEAD heterodimer complex activates expression of transcription genes crucial for tissue growth [7].

Figure 1. The Hippo Signaling Pathway: upper half illustrates normal functioning of pathway inhibiting cell proliferation; bottom half illustrates dysregulation of pathway leading to uncontrolled growth
Flufenamic Acid is an inexpensive, non-steroidal anti-inflammatory drug (NSAID) first known to have anti-inflammatory properties in the 1960s [8]. Since then, it has been shown to act as an ion channel modulator [8]. Additionally, Flufenamic Acid has been proven effective in inhibition of cell proliferation, clonogenic activity, migratory ability and gene expression in aortic smooth muscle cells [5]. Due to the high rate of gastrointestinal side effects, Flufenamic Acid has not been used in the US. However recent research studies confirm the efficacy of Flufenamic Acid in bladder cancers in mice, re-paving the way for a new path for cancer therapy research [9].
Despite these studies, the efficacy of Flufenamic Acid against pancreatic cancer cells has not previously been researched. This study shows that rather than binding directly to the YAP binding site on TEAD transcription factors and preventing YAP from interacting with TEAD, Flufenamic Acid has a higher affinity for a separate pocket on the TEAD surface, disrupting YAP-TEAD from activating target transcription genes after YAP and TEAD have already interacted. This study examines the potential of the Hippo Signaling Pathway by targeting the YAP- TEAD complex with Flufenamic Acid and establishes a foundation for future treatment studies.
Materials and Methods
Cell culture and cell lines: The murine pancreatic adenocarcinoma cell LTPA (ATCC CRL-2389) line was purchased from American Type Culture Collection (ATCC, www.atcc.org). The cell line was cultured in EMEM media (Invitrogen/ThermoFisher, Carlsbad, CA) with 10% FBS (Invitrogen/ThermoFisher, Carlsbad, CA), 2mM glutamine (Invitrogen, Carlsbad, CA) and 100 U penicillin/0.1 mg/ml streptomycin (Invitrogen, Carlsbad, CA). Cell line was maintained at 37 °C in 5% CO2.
Antibodies and therapeutic reagents: The antibodies used include: YAP mouse monoclonal antibody 63.7 (cat. sc-101199, Santa Cruz Biotechnology, Santa Cruz, CA), Alexa Fluor 488 conjugated anti-mouse antibody G-6 (Jackson, ImmunoResearch, West Grove, PA), biotinylated goat anti-mouse IgG secondary antibody (Vector Laboratories, Burlingame, CA), and DAPI cat. D1306 counterstain (Invitrogen/ ThermoFisher, Carlsbad, CA). Flufenamic Acid was purchased from Sigma. An 320 μM stock solution of Flufenamic Acid was prepared by dissolving in DMSO (Sigma) and stored at 4 °C.
Cell viability assays: To determine percent cell viability, a WST-8 Cell Viability assay (CCK-8 Dojindo Molecular Technologies, Rockville, MD) was used. 100 μL of cell suspension (5000 cells/well) were incubated in a 96-well plate for 24 hours at 37 °C, 5% CO2.
Eight two-fold dilutions were carried out from the Flufenamic Acid stock solution and was added to each specified well with one control well containing cells with no Flufenamic Acid, and another control well containing media only. After 48 hours of continuous drug exposure at 37 °C, 10 μL of WST-8 Viability Assay Reagent (Dojindo Molecular Technologies, Rockville, MD) was added into each well and the plate was incubated for 2 hours at 37 °C. In order to calculate relative cell viability, absorbance value from the wells with WST-8 and no cells was subtracted from the absorbance values of all other wells and these values were divided by the absorbance value of the negative control wells (cells only). Values were obtained from iMark Microplate Reader. Each concentration was tested 3 times in replicate wells and average cell viability assay percentages were calculated.
Immunofluorescence: LTPA cells were fixed with 4% formaldehyde diluted in PBS for 15 minutes at room temperature and rinsed. After permeabilization with 0.1% Triton X-100 and blocking in 1% BSA in PBS for 1 hour, diluted 1:200 primary mouse YAP antibody (cat. sc 101199, Santa Cruz) was applied and incubated overnight at 4 °C. Cells were washed and incubated with Alexa Fluor 488 conjugated anti-mouse antibodies G-6 (Jackson, ImmunoResearch) diluted 1:1000 for 2 hours at room temperature in the dark. DNA was stained with DAPI counterstain cat. D1306 (Invitrogen) diluted 1:1000 in PBS. Cells were washed and mounted. Slides were analyzed using Olympus fluorescence microscope (DP 80).
Immunocytochemistry: In a 6 well plate containing cover slips, LTPA cells were trypsinized, centrifuged, and resuspended in 500 μL EMEM media with approximately 5000 cells/well. Cells were fixed onto coverslips using 400 μL of 4% formaldehyde, washed with PBS, covered with 400 μL of wash buffer and permeabilized with PBS containing 0.1% Triton X-100. After rewashing, 400 μL of blocking buffer (10% normal goat serum, 0.3% Triton X-100) was added and stored for 45 mins at room temperature before removing blocking buffer. 400 μL of diluted 1:50 YAP mouse monoclonal primary antibody 63.7 (cat. sc-101199, Santa Cruz Biotechnology, Santa Cruz, CA) was added and incubated overnight at 4 °C. Coverslips were washed and incubated for 1 hour with biotinylated goat anti-mouse IgG secondary antibody (Vector Laboratories, Burlingame, CA). Coverslips were rinsed with PBS and Streptavidin- HRP (ThermoFisher) was added to each well and incubated for 30 minutes at room temperature before washed again. DAB peroxidase substrate solution (Vector Laboratories, Burlingame, CA) was added and incubated for 5 minutes at room temperature. Samples were washed with PBS to stop development of color reaction and mounted for light microscopy analysis.
Computational Protein and Molecular Modeling: Flufenamic Acid and TEAD/YAP structures were modeled and analyzed using the UCSF Chimera Modeling Application with PDB, PubChem, and UniProt molecular codes. Druggability probability was calculated using PockDrug. PDB codes used for computational modeling: 3KYS (YAP-TEAD), 3L15 (TEAD2), 5DQ8 (Flufenamic Acid-TEAD) [16,17].
 
Results
Target-ability of YAP-TEAD complexes with Flufenamic Acid Monomer:
YAP-TEAD binding activity involves certain pocket surfaces and residue structures (Figure 2A). There are three major binding interfaces on the TEAD transcription factor structure in the YAP binding domain [13]. Interface 1 binds YAP residues 51-58 to TEAD residues 318-324; Interface 2 binds YAP residues 65-69 to TEAD residues 346-370 and 314; Interface 3 binds YAP residues 86-99 to TEAD residues 240-249, 404-406, 272-276, and 391 (Figure 2B). There are 4 possible methods of inhibiting YAP-TEAD complexes in the dysregulated pathway: inhibition at one of the three interfaces or the central pocket of the YAP binding domain.
Structural Models of YAP-TEAD binding domains:

Figure 2. YAP-TEAD complex structural models: (A) Overall structure of YAP-TEAD complex with residues labeled (B) YAP-TEAD complex with YAP outlined in blue, Interface 1 indicated in white, Interface 2 in red, Interface 3 in green, and central pocket of YAP binding domain in yellow
Druggability Probabilities of YAP Binding Domains on TEAD factors:
Two factors considered were volume of the pocket and druggability probability. The volume of the central pocket was significantly larger than the other surfaces, and thus contains more surface residues to bind an inhibiting small molecule drug (Figure 3). The druggability probability is defined as the the likelihood for a small molecule drug to bind to a pocket surface in order to inhibit protein function. Residues were matched between models created in Chimera and the TEAD2 in PockDrug in order to determine which four of the 20+ possible pockets generated on PockDrug contained the corresponding YAP binding domains in Chimera (Figure 4). The druggability probabilities of the four were formulated. The central pocket surface (Pocket 1) had the highest possible druggability probability value of 1.0 with standard deviation value of 0.0. The other three pockets had smaller druggability probabilities and larger standard deviation values.

Figure 3. Pocket Surfaces for YAP binding data: Pocket number, druggability probability value, standard deviation for the probability, volume hull, and number of residues for each pocket surface (interface area)

 
Figure 4. PockDrug (Left Panels) and Chimera (Right Panels) Structures of binding domains: Left panel contains figures of the pocket surfaces that contain YAP binding residues; Right panel contains figures of the actual surface of TEAD that binds to YAP (A) Interface (IF) 1 bright green pocket area (B) IF 1 actual YAP binding surface dark blue (C) IF 2 forest green pocket area (D) IF 2 actual YAP binding surface pink (E) IF 3 light blue pocket area (F) IF 3 actual YAP binding surface light green (G) Central Pocket dark yellow pocket area (H) Central Pocket actual YAP binding surface white
Flufenamic Acid Inhibits YAP-TEAD Function through Interface 3 or Central Pocket
The above data indicates higher potential for druggability in Interface 3 and the central pocket. Flufenamic Acid structure fits both locations when modeled in Chimera; however Flufenamic Acid more probably binds to the central pocket instead of Interface 3 based on the central pocket’s higher druggability probability and larger volume hull (Figure 5). Interface 3 has smaller volume available, 736.55, that is on the outer surface of the TEAD2 factor. This would not fit the Flufenamic Acid drug molecule structurally as effectively as a pocket surface embedded in the center of the TEAD2 factor with larger binding volume. The degree of significance of Interface 3 in Flufenamic Acid-TEAD-YAP complex binding is further researched with immunostaining processes and cell viability assay below.
Flufenamic Acid Inhibits LTPA Cell Proliferation without Disrupting YAP Expression:
Flufenamic Acid effectively inhibited LTPA cell proliferation seen in the cell viability assay results (Figure 6A). At a 160 μM concentration, approximately 52.7% of the cells survived. Among the three trials tested, this concentration had a low standard deviation of 4.15. The dose- response curve generated for the viability assay is similar to the first half of the expected backwards S-shaped curve. A top plateau, sharp decline in the middle, and approach towards a bottom plateau is seen (Figure 6B). In all three trials, cell viability noticeably decreased with Flufenamic Acid treatment, while YAP expression locations did not change.

Figure 5. Structural models of Flufenamic Acid bound to TEAD at the two potential sites: (A) Focus: Flufenamic Acid bound at Interface 3 in the cyan and magenta (B) Focus: Flufenamic Acid bound at central pocket in green and magenta (C) Surface model of Flufenamic Acid bound only at Interface 3 (D) Surface model sliced and capped of Flufenamic Acid bound only at the central pocket (E) Sequence with residues for Flufenamic Acid- TEAD2 binding at Interface 3 highlighted in cyan (F) Sequence with residues for Flufenamic Acid-TEAD2 binding at central pocket highlighted in green

Figure 6. Cell Viability Assay and Dose Response Curve: (A) Two fold dilutions of stock solution of Flufenamic Acid and % cell viability data of 3 different trials including standard deviation of overall results for each concentration (B) Dose-response curve plotting concentration with mean % cell viability and standard deviation bars and modeled mathematical equation for curve
Immunocytochemistry was performed to compare the pattern and levels of YAP in LTPA cells with and without Flufenamic Acid treatment, after observing a dramatic decrease in cell viability after treatment dropping as low as 38.57%. Increased levels of YAP in the pancreatic adenocarcinoma cells before and after Flufenamic Acid treatment are indicated by the arrowheads (Figure 7A and 7B). Pattern of staining with Flufenamic Acid treatment indicate the specific TEAD2-Flufenamic Acid binding analyzed in Discussion. Immunofluorescence was then performed to determine whether YAP remained in the cytoplasm, was translocated to the nucleus, or both, with Flufenamic Acid treatment. YAP staining is present in both but appears in higher concentrations in the nuclear area (Figure 7E and 7H). Density of YAP presence increased after drug treatment as seen by the brighter green fluorescence in the nucleus.

Figure 7. Immunostainings of YAP in LTPA cells: (A) Immunocytochemistry without Flufenamic Acid treatment- elevated patterns of YAP indicated by yellow arrows (B) Immunocytochemistry after Flufenamic Acid treatment- elevated patterns of YAP still apparent with minimal to no change (C) Immunofluorescence stain of YAP (green) in LTPA cells without Flufenamic Acid treatment (D) Immunofluorescence stain of DAPI (blue) nucleus in LTPA cells without Flufenamic Acid treatment (E) Merged image of C and D- YAP detected in LTPA cells with concentration in the nuclear areas but still maintained in the cytoplasm (F) Immunofluorescence stain of YAP (green) in cells after Flufenamic Acid treatment (G) DAPI immunofluorescence of cells after Flufenamic Acid treatment (H) Merged image of G and H- YAP expressed in both nuclear areas and cytoplasm
Discussion
Computational and structural analysis of the YAP-TEAD complex in relation to Flufenamic Acid presented two possibilities:
1) Flufenamic Acid attaches to TEAD2 at Interface 3, preventing YAP from completely binding to TEAD2 and turning off transcription.
2) Flufenamic Acid attaches to a separate site- the central pocket- and YAP still binds to TEAD2 but protein function is inhibited and transcription is turned off.
The former possibility inhibits YAP binding to TEAD2 while the latter still allows YAP-TEAD2 binding but inhibits its functioning afterwards. The latter was more probable due to the larger volume and more residues available for binding in the central pocket, in addition to its higher druggability probability.
The cell viability assay illustrated the efficacy of Flufenamic Acid in inhibiting pancreatic adenocarcinoma cell proliferation by approximately 50% with a concentration of 160 μM. The immunocytochemistry results confirmed the elevated presence of YAP and its staining pattern [10]. The immunofluorescence results elucidate the differences in nuclear and cytoplasmic retention of YAP that were not apparent from the immunocytochemistry. If Flufenamic Acid only attached at Interface 3, YAP would not bind to TEAD in the first place and increased unbound YAP would be present in the nucleus with an appreciable amount of YAP remaining in the cytoplasm. If Flufenamic Acid only attached to the central pocket, YAP would still translocate to the nucleus and bind to TEAD but transcription would not be turned on and there would be a much higher density of YAP retained in the nucleus. The concentrated clusters of YAP apparent in the nucleus from the immunofluorescence in comparison to the cytoplasm after drug treatment suggest that Flufenamic Acid interacts with the central pocket. Additionally the increase in density after drug treatment confirms the more probable dominant manner that Flufenamic Acid would act on the central pocket.
This suggests a novel role that Flufenamic Acid plays in the Hippo Signaling Pathway, preventing cancer cell proliferation in pancreatic cells. By binding to the central pocket, Flufenamic Acid inactivates proper functioning of the YAP-TEAD complexes. In turn, although this would not reverse the dysregulated pathway completely, it would assist in preventing metastasis and pancreatic cancer progression by inhibiting the unchecked cell proliferation that occurs after YAP-TEAD functioning [15]. Some other targets of the Hippo Signaling Pathway in past research include inhibition of YAP nuclear translocation, upstream regulators and YAP binding to other transcription factors [11,12]. However the small volume of space available at the YAP binding domains poses a difficulty in developing drugs that can effectively inhibit binding activity. Flufenamic Acid takes advantage of the larger and highly druggable central pocket space to inhibit YAP-TEAD in a different manner. Many common drugs and their analogs have been tested to regulate the Hippo Signaling Pathway but have failed due to limitations.
Conclusion
This study illustrates that Flufenamic Acid is effective in murine pancreatic adenocarcinoma cells by inhibiting the YAP-TEAD interactions that are key to the deactivation of the Hippo Pathway. The exact underlying mechanisms of how Flufenamic Acid acts on the pathway are poorly understood and yet to be thoroughly researched. The Hippo Signaling Pathway itself is also poorly understood in mammals. Structural analogs of Flufenamic Acid could be studied to determine how to reduce side effects by altering the drug’s mechanism of inhibiting pancreatic cancer progression or investigating a synergistic effect for more targeted inhibition in the Hippo Signaling Pathway. It’s impossible to completely rule out the possibility that Flufenamic Acid partially binds to Interface 3 to inhibit YAP-TEAD binding when the mechanisms of Flufenamic Acid and the pathway are not fully understood. However the efficacy of Flufenamic Acid points to an unresearched potential for targeting pancreatic cancer. With the growing interest in the Hippo Pathway, Flufenamic Acid is a novel avenue for future studies.
References

1. Jemal A., Bray F., Center M. M., Ferlay J., Ward E. & Forman D. Global cancer statistics. CA Cancer J

Clin 61, 69–90 (2011).

2. Witt H., Apte M. V., Keim V. & Wilson J. S. Chronic pancreatitis: challenges and advances in pathogenesis, genetics, diagnosis, and therapy. Gastroenterology 132, 1557–1573 (2007).
3. Yu, F.X.; Zhao, B.; Guan, K.L. Hippo pathway in organ size control, tissue homeostasis, and cancer. Cell 2015, 163, 811–828.
4. Zhang, L.; Ren, F.; Zhang, Q.; Chen, Y.; Wang, B.; Jiang, J. The tead/tef family of transcription factor

scalloped mediates hippo signaling in organ size control. Dev. Cell 2008, 14, 377–387.

5. Schober W, Wiskirchen J, Kehlbach R, et al: Flufenamic acid: growth modulating effects on human aortic smooth muscle cells in vitro. J Vasc Interv Radiol. 13:89–96. 2002.
6. Zeng Q, Hong W. The emerging role of the hippo pathway in cell contact inhibition, organ size control,

and cancer development in mammals. Cancer Cell 2008; 13: 188–192. doi: 10.1016/j.ccr.2008.02.011. pmid:18328423

7. Hong W, Guan K-L. The YAP and TAZ transcription coactivators: key downstream effectors of the mammalian Hippo pathway. Seminars in Cell & Developmental Biology. 2012;23(7):785-793. doi: 10.1016/j.semcdb.2012.05.004.
8. Guinamard R, Simard C, Negro CD. Flufenamic acid as an ion channel modulator. Pharmacology &

therapeutics. 2013;138(2):272-284. doi:10.1016/j.pharmthera.2013.01.012.

9. \”Flufenamic Acid Slows Spread of Bladder Cancer in Mouse Model.\” Biotechdaily.com. Globe Tech

Media, 26 Oct. 2016. Web. 16 Dec. 2016. <http://www.biotechdaily.com/genomics-proteomics/articles/ 294766558/flufenamic-acid-slows-spread-of-bladder-cancer-in-mouse-model.html>.

10. Morvaridi S, Dhall D, Greene MI, Pandol SJ, Wang Q. Role of YAP and TAZ in pancreatic ductal

adenocarcinoma and in stellate cells associated with cancer and chronic pancreatitis. Scientific Reports. 2015;5:16759. doi:10.1038/srep16759.

11. Zhao B, Wei X, Li W, et al. Inactivation of YAP oncoprotein by the Hippo pathway is involved in cell contact inhibition and tissue growth control. Genes & Development. 2007;21(21):2747-2761. doi:10.1101/ gad.1602907.
12. Pobbati AV, Han X, Hung AW, et al. Targeting the Central Pocket in Human Transcription Factor TEAD as a Potential Cancer Therapeutic Strategy. Structure (London, England  : 1993). 2015;23(11):2076-2086. doi: 10.1016/j.str.2015.09.009.
13. Li Z, Zhao B, Wang P, Chen F, Dong Z, Yang H, Guan KL, Xu Y: Structural insights into the YAP and TEAD complex. Genes Dev 24: 235-240, 2010.
14. Kang W, Cheng AS, Yu J, To KF. Emerging role of Hippo pathway in gastric and other gastrointestinal cancers. World Journal of Gastroenterology. 2016;22(3):1279-1288. doi:10.3748/wjg.v22.i3.1279.
15. George NM, Day CE, Boerner BP, Johnson RL, Sarvetnick NE. Hippo Signaling Regulates Pancreas

Development through Inactivation of Yap. Molecular and Cellular Biology. 2012;32(24):5116-5128. doi: 10.1128/MCB.01034-12.

16. Hiba Abi Hussein, Alexandre Borrel, Colette Geneix, Michel Petitjean, Leslie Regad, Anne-Claude Camproux. PockDrug-Server: a new web server for predicting pocket druggability on holo and apo proteins. Nucleic Acids Research. May 8, 2015.
17. Sheridan, R.P., Maiorov, V.N., Holloway, M.K., Cornell, W.D., and Gao, Y.D. Drug-like density: a method of quantifying the “bindability” of a protein target based on a very large set of pockets and drug-like ligands from the protein data bank. Journal of Chemical Information and Modeling. 2010; 50: 2029-2040.