OC OVCAR3 and OVCAR8 cell lines were purchased from the National Cancer Institute and maintained in RPMI 1640 medium supplemented with 10% FBS (Atlanta Biologicals, GA), 100 U/ml penicillin, and 100 μg/ml streptomycin (Fisher Scientific) in a humidified incubator at 37°C in 5% CO₂. VS6063 and FAK PROTAC Degrader 1 was purchased from MedChemExpress (MCE).

Cell proliferation was examined following instructions of MTT assay kit (ATCC, Manassas, VA). Briefly, OVCAR3 or OVCAR8 cells (3,000/well) were plated into 96-well plates and treated with different doses of FAK PROTAC, VS6063 or vehicle for various time points, and then MTT reagent (5mg/mL) was added to each well. After a 4-hour incubation, the media/MTT was aspirated and 100 μL DMSO was added to each well. The plate was incubated for 20 min and the absorbance was measured at single-wavelength mode (490 nm) using a Multiskan MK3 Microplate Reader (Thermo Fisher Scientific). BrdU proliferation assay was performed by labelling OC cells for 12 h with 10 µM BrdU and cells were fixed with 3.7% formalin and cells were washed and incubated with BrdU antibody (Santa Cruz, Cat.No. SC-32323) for 24h at 4°C and then fluorescent labelled secondary antibody Alexa 488 (Thermo Fisher Scientific. Cat. No. A32723) for 1h at room temperature. Cell proliferation was detected by counting BrdU labelled cells and DAPI labelled cell nuclei from four different fields under fluorescent microscopy.

OVCAR3 or OVCAR8 cells (300 cells/well) were plated into 6-well plates and treated with 0.5 μM FAK PROTAC, VS6063 or vehicle for two weeks and then stained with 0.1% Crystal Violet. Cell colonies were counted using ImageJ software from three different wells.

Cell migration was performed using modified Transwell chambers (BD Falcon™, San Jose, CA) inserted into 24-well culture plates as described previously (3). Briefly, OVCAR3 or OVCAR8 cells were treated with 1 μM FAK PROTAC, VS6063 or vehicle. After 24h cells were seeded into transwells. RPMI 1640 containing 10% FBS was added in the bottom of the chamber as the chemical attractant and incubated for 16 h. The upper chamber medium and non-migrated cells were removed, and migrated cells were fixed with methanol and stained with crystal violet. Images were taken at 10× magnification and migrated cells were counted in at least three randomly selected different fields using the ImageJ software. Cell migration was also determined by wound healing assays as described previously (29). Briefly, cells (3×10⁵ cells per well) were seeded in triplicate into 6-well plates and cultured for 24h, and then cell surface was scratched with a pipette tip. After 24h of culture, the migration rate was calculated using the following formula: (area of the wound area at 0 h - the wound area at 24 h)/the wound area at 0 h.

OVCAR3 or OVCAR8 cells (3×10⁵) were treated with 1 μM FAK PROTAC, VS6063 or vehicle for 24 h and then seeded in serum-free RPMI 1640 media onto Matrigel (BD BioSciences, San Jose, CA) coated transwells. RPMI 1640 containing 10% FBS was added in the bottom chamber as chemoattractant and incubated for 24h. The invaded cells were stained with H&E and counted as described previously (30).

Paraffin-embedded ovarian sections of HGSC were obtained from de-identified ovarian patient biopsy specimens (UTHSC Tissue Services Core). Deidentified ovarian tumor PDX sections and human OC specimens were collected under a UTHSC IRB-approved protocol (14-03113-XP). Sections were subjected to antigen retrieval at 90°C for 30 mins, and then incubated with FAK (Cell Signaling Technology, Inc, Danvers, MA) and PCNA (Santa Cruz, CA, Cat.No. SC-56) antibodies at 4°C overnight. Sections were washed and incubated with secondary antibody Alexa 488 or 594 conjugated goat anti-rabbit (1:200 dilution, Fisher Scientific). Nuclei were counterstained with DAPI.

Paraffin-embedded ovarian tumor sections were subjected to antigen retrieval at 90°C for 30 mins and incubated at 4°C overnight with FAK (Cell Signaling Technology, Inc, Danvers, MA) and PCNA (Santa Cruz, CA) antibodies, and then incubated with biotinylated secondary antibody using the Vectastain Elite kit (Vector Laboratories).

WB was performed as described previously (3). The membranes were blocked with 5% nonfat milk for 1h and incubated with primary antibodies against ASAP1 (1:1,000, Santa Cruz, CA, Cat.No.SC32323), FAK (Cat.No. 3285S), p-FAK, (1:1000, Cell Signaling Technology, Inc, Danvers, MA, Cat. No.8556S) and GAPDH (1:1,000, Sigma, St. Louis, MO, Cat. No. G8795) overnight at 4°C. The membranes were washed three times with PBST and then incubated with HRP-conjugated secondary antibodies (Yeasen) for 1 h. After the final wash with TBST, the proteins were detected using enhanced chemiluminescence (ECL) reagents.

Cells were treated with 1 µM FAK PROTAC, VS6063 and vehicle for 24 h and then cells were collected for IP as described previously (3) using the Pierce Classic IP Kit (Thermo Fisher). Cell lysate was incubated with either 10 µg of IgG or ASAP1 antibody (Santa Cruz, Cat. NO. SC-374410) overnight at 4°C. The immune complex was then captured with Pierce Protein A/G Agarose beads for 1h at 4°C, then washed with lysis buffer and conditioning buffer, and eluted with elution buffer. The immune complex was analyzed by WB using ASAP1 or FAK antibodies.

All animal experiments were conducted in accordance with the approval from the Institutional Animal Care and Use Committee (IACUC) at the University of Tennessee Health Science Center. Luciferase reporter gene labelled OVCAR8 cells were intrabursally injected into ten two-month-old immunocompromised NOD scid gamma (NSG) female mice and randomly divided into two groups. Mice were treated with FAK PROTAC (10 mg/kg body weight) or vehicle every day for three weeks. Tumor growth and metastasis were monitored using a Xenogen live animal imaging system weekly. All mice were sacrificed after three-weeks of treatment, tumors were weighed and collected for WB to detect FAK.

To examine the genetic alterations and expression of PTK2 in OC, we queried the TCGA (https://tcga-data.nci.nih.gov/tcga/tcgaHome2.jsp) and Oncomine database (https://www.oncomine.org). For gene copy number analysis, OC patient data were selected from Pan Cancer Atlas and compared to other cancer types in TCGA database. In Oncomine database, OC patient samples were selected from two TCGA cohorts including normal ovaries, normal blood and HGSC. Copy number or expression level was analyzed by selecting bar-graph format.

Significant differences between independent experiments were determined by one-way ANOVA or Student’s t-test, and the data were expressed as mean ± SD. P<0.05 was considered as statistically significant.

To assess the expression of PTK2 in cancer, we examined gene profiles from dataset in the TCGA’s Pan Cancer Atlas (The Cancer Genome Atlas). PTK2 is most upregulated or amplified in OC, followed by esophageal and breast cancer among 32 cancer types listed in Figure 1A . We then examined the copy numbers of PTK2 in 431 samples of normal blood, 130 samples of normal ovaries and 607 samples of high-grade serous carcinoma (HGSC) and found PTK2 was significantly amplified or upregulated in HGSC compared to normal blood or ovarian samples ( Figure 1B ). We then examined PTK2 expression is 586 samples of HGSC and 8 samples of normal ovaries in the Human Genome U133A Array data (https://www.oncomine.org). PTK2 is significantly higher in HGSC than that in normal ovaries ( Figure 1C ). To validate PTK2 (FAK) expression, we perform immunostaining on sections of human HGSC ( Figure 1D ) and Patient-Derived-Xenograft (PDX) samples ( Figure 1E ) using FAK antibody. FAK is significantly higher in tumors than that in adjacent normal tissues. We also examined the correlation between FAK expression and OC patient survival using the Kaplan Meier Plotter database. High FAK expression is significantly correlated with patient poor overall survival (OS) and progression free survival (PFS) among 1435 OC patients including HGSC and endometroid ( Figure 1F ).

FAK PROTAC was designed and synthesized based on the parent FAK kinase inhibitor VS6063 (12). As shown in Figure 2A , FAK PROTAC links with VS6063 and a ubiquitin E3 ligase. VS6063 as a FAK kinase inhibitor targets FAK kinase activity and E3 ligase transfers ubiquitin to FAK, degrading FAK through polyubiquitination/proteasome pathway. To examine whether FAK PROTAC degrades FAK, we treated OC cells, OVCAR3 and OVCAR8, with different PROTAC doses for 24 h and performed WB to detect FAK protein levels. FAK was significantly degraded in OVCAR3 and OVCAR8 cells with a dose of 50 nM PROTAC and the most effective dose is 1 µM in both cell lines ( Figure 2B ). Next, we examined the time course of FAK degradation by treating both OVCAR3 and OVCAR8 cells with 1 µM PROTAC for 4, 8 and 16 h. FAK was significantly inhibited at 4 h and depleted at 16 h ( Figure 2C ). To compare the efficacy of FAK PROTAC and its parent FAK kinase inhibitor VS6063, we treated both OVCAR3 and OVCAR8 cells with 1 µM PROTAC or 1 µM VS6063 for 24h and then examined the phospho-FAK (pFAK) and total FAK by WB. As expected, VS6063 inhibits FAK activity but not total FAK protein, while FAK PROTAC inhibits FAK activity and depletes total FAK protein ( Figure 2D ).

To compare the functional differences of FAK PROTAC and FAK inhibitor VS6063, we treated OVCAR3 and OVCAR8 cells with FAK PROTAC, VS6063 or vehicle for 48 and 72 h and cell proliferation was determined by MTT assay. Both FAK PROTAC and VS6063 significantly inhibited cell proliferation and FAK PRTOAC is more effective than VS6063 at both time points ( Figure 3A ). To examine how PROTAC affects cell survival, we performed cell clonogenic assay after treating cells for two weeks with 1 µM PROTAC or VS6063. FAK PROTAC was more effective in inhibiting cell colony formation than VS6063 in both OVCAR3 and OVCAR8 cell lines ( Figure 3B ). We also assessed how FAK PROTAC affects cell migration and invasion by treating both OVCAR3 and OVCAR8 cells for 4 h with 1 µM FAK PROTAC and 1 µM VS6063. FAK PROTAC is more effective than VS6063 in inhibiting cell migration ( Figures 4A–C ) and invasion ( Figure 4D ) in both cell lines.

We recently reported that FAK is co-expressed in OC and interacts with ASAP1 in OC cells (3). To understand whether FAK PROTAC blocks FAK kinase-dependent and kinase-independent pathways, we tested whether FAK PROTAC disrupts the interaction between FAK and ASAP1. We treated OVCAR3 cells with 1 µM PROTAC or 1 µM VS6063 for 24 h and then performed IP with ASAP1 antibody, followed by WB with ASAP1 and FAK antibodies. FAK protein was pulled down by ASAP1 in VS6063 treated cells but not in PROTAC-treated cells. Although both PROTAC and VS6063 inhibit pFAK ( Figure 5 ), FAK PROTAC indeed disrupts the interaction between FAK and ASAP1. In contrast, VS6063 inhibits FAK activity as shown by reduced pFAK but does not disrupt the interaction between FAK and ASAP1.

We showed that FAK PROTAC is more effective than VS6063 in inhibiting cell growth, migration, and invasion. To test the efficacy of FAK PROTAC in vivo, we intrabursally injected OVCAR8 cells in immunocompromised mice and treated them with FAK PROTAC for three weeks. Ovarian tumor growth and metastasis was monitored weekly using Xenogen live animal imaging. FAK PROTAC significantly inhibited primary ovarian tumor growth as shown in bioluminescent images ( Figure 6A ). The decrease in dissected ovarian tumors by FAK PROTAC was also shown by tumor weight and bioluminescence ( Figures 6B, C ). To examine the FAK in vivo, we performed immunohistochemical staining. FAK was strongly stained in the cytoplasm of ovarian tumor sections of control mice but was weakly stained in ovarian tumor sections of mice treated with FAK PROTAC ( Figure 6D ). Ovarian tumors were also characterized by H&E staining ( Figure 6D ). The levels of FAK protein were diminished in ovarian tumors of mice treated with FAK PROTAC but was expressed in tumors of vehicle-treated mice with as shown by WB ( Figure 6E ). Tumor metastasis in different organs was also examined when mice were sacrificed, and FAK PROTAC significantly inhibits tumor metastasis as shown in intestine, liver, spleen, kidney, and stomach ( Figure 7 ).