j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 5 ) 1 e1 0
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.JournalofSurgicalResearch.com
Celecoxib inhibits Ewing sarcoma cell migration via actin modulation Christopher A. Behr, MD,a,b,* Anthony J. Hesketh, MD, MS,a,b,c Meade Barlow, MD,a,b Richard D. Glick, MD,a Marc Symons, PhD,b,c Bettie M. Steinberg, PhD,b,c,1 and Samuel Z. Soffer, MDa,b,1 a
Division of Pediatric Surgery, Department of Surgery, Hofstra North Shore-LIJ School of Medicine, New Hyde Park, New York b The Feinstein Institute for Medical Research, Center for Oncology and Cell Biology, North Shore-LIJ Health System, Manhasset, New York c The Elmezzi Graduate School of Molecular Medicine, The Feinstein Institute for Medical Research, North Shore-LIJ Health System, Manhasset, New York
article info
abstract
Article history:
Background: Ewing sarcoma (ES) is an aggressive childhood solid tumor in which 30% of
Received 2 January 2015
cases are metastatic at presentation, and subsequently carry a poor prognosis. We have
Received in revised form
previously shown that treatment with celecoxib significantly reduces invasion and
16 March 2015
metastasis of ES cells in a cyclooxygenase-2-independent fashion. Celecoxib is known to
Accepted 27 March 2015
downregulate b-catenin independently of cyclooxygenase-2. Additionally, the actin cyto-
Available online xxx
skeleton is known to play an important role in tumor micrometastasis. We hypothesized that celecoxib’s antimetastatic effect in ES acts via modulation of one of these two targets.
Keywords:
Methods: ES cells were treated with celecoxib, and the levels of b-catenin and total actin were
Ewing sarcoma
examined by Western blot and quantitative polymerase chain reaction. Cells were trans-
Celecoxib
fected with small interfering RNA targeting b-catenin, and invasion assays were performed.
Metastasis
Immunofluorescence staining for b-catenin and F-actin was performed on treated and un-
Actin
treated cells. Additionally, cells were subjected to a wound healing assay to assess migration.
b-catenin
Results: Celecoxib had no effect on the messenger RNA or protein levels of b-catenin but did
Cytoskeleton
significantly decrease the amount of total actin within ES cells. Reduction of b-catenin by small interfering RNA had no effect on invasion, and celecoxib treatment of the b-catenin depleted cells continued to inhibit invasion. Immunofluorescence staining demonstrated no change in b-catenin with treatment but did show a significant reduction in the amount of F-actin, as well as morphologic changes of the cells. Wound healing assays demonstrated that celecoxib significantly inhibited migration. Conclusions: Celecoxib does not exert its antimetastatic effects in ES through alteration of b-catenin but does significantly modulate the actin cytoskeleton. ª 2015 Elsevier Inc. All rights reserved.
* Corresponding author. The Feinstein Institute for Medical Research, Center for Oncology and Cell Biology, 350 Community Drive, Lab 2262, Manhasset, NY 11030. Tel.: þ1 516 562 1053; fax: þ1 516 562 1022. E-mail address: [email protected] (C.A. Behr). 1 Co-senior authors. 0022-4804/$ e see front matter ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jss.2015.03.085
2
1.
j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 5 ) 1 e1 0
Introduction
Ewing sarcoma (ES) and the ES family of tumors are a group of malignant tumors of the bone and soft tissue. They are characterized as small, round blue cell tumors, which typically affect children between the ages of 5e15 and are often quite aggressive [1]. Although the overall 5-y survival for ES is between 60 and 70%, approximately 25% of patients present with metastatic disease, which carries a poor prognosis [2e4]. Fewer than 40% of patients with distant metastasis on initial diagnosis survive past 5 y [5,6]. Strategies to treat and prevent metastasis continue to be the focus of significant research. We have previously shown that the selective cyclooxygenase-2 (COX-2) inhibitor celecoxib prevents lung metastasis in a murine model of ES, with minimal effect on the primary tumor [7,8]. Furthermore, celecoxib inhibits invasion by ES cells in a COX-2-independent fashion [9]. Recent clinical studies have shown that the addition of celecoxib and vinblastine to standard treatment protocols (alternating cycles of vincristine/doxorubicin/cyclophosphamide and ifosfamide/etoposide) in ES doubled the 2-y event-free survival for patients with lung metastases, from 36%e71% [10]. The mechanism by which celecoxib prevents malignant progression of ES has yet to be elucidated. Celecoxib exhibits many of its antitumor effects via a decrease in COX-2 activity, particularly in colon cancer [11e15]. Celecoxib inhibition of COX-2 also plays a role in treating various other malignancies including breast, endometrial, and pancreatic cancers [16e18]. However, myriad off-target effects of celecoxib treatment are being evaluated for their antitumor properties. A 2006 review by Gro¨sch et al. [19] identified over 40 different targets for the COX-2independent effects of celecoxib on apoptosis, cell cycle regulation, and angiogenesis and/or metastasis. In particular, it alters the Wnt signaling pathway through modulation of b-catenin. Studies in colorectal cancer models have demonstrated that celecoxib works in part by inducing the rapid translocation of predominantly membrane-bound b-catenin to the cytoplasm and nucleus, followed by its degradation [20e23]. Other cancers, such as the lung, liver, and glioblastoma, have likewise shown a b-catenin response to celecoxib treatment [24e27]. Furthermore, b-catenin has been shown to specifically influence the metastatic potential of ES [28,29]. Another potential mechanism through which celecoxib may exert its antitumor effects is through modification of the actin cytoskeleton. Actin dynamics are known to be vital for cell motility, migration, shape, and cellecell junctions and interactions [30,31]. b-actin is overexpressed in various cancers [32,33] and is especially overexpressed in samples taken from metastases [32e34]. Relative invasiveness of cancer cells has been associated with increases in b-actin, as well as increases in the proportion of polymerized F-actin to monomeric G-actin, resulting in the reorganization of actin within the cell [35e38]. Although some have suggested the importance of the actin cytoskeleton in ES [39e42], little is known about celecoxib’s influence on this integral cellular protein, which may have particular importance in tumor micrometastasis.
2.
Materials and methods
2.1.
Cell lines
SK-NEP1 (ATCC, Manassas, VA, HTB-48) cells, a well-described ES cell line, were maintained in McCoy’s media (Gibco [Invitrogen], Grand Island, NY) supplemented with 15% fetal clone II serum (Hyclone; Thermo Scientific, Waltham, MA) and 1% penicillin-streptomycin, and incubated at 37 C and 5% CO2. Additionally, a portion of the cells were transfected with small interfering RNA (siRNA) oligomers targeting b-catenin or with nontargeting control siRNA (Dharmacon, Lafayette, CO). Transfection was accomplished using the Lipofectamine 2000 (Invitrogen, Carlsbad, CA) transfection reagent. Caco-2 (ATCC, HTB-37) cells, a well-described colon cancer cell line, were maintained in Delbecco Modified Eagle medium (Hyclone; Thermo Scientific) supplemented with 10% fetal clone II serum (Hyclone; Thermo Scientific) and 1% penicillinstreptomycin, and incubated at 37 C and 5% CO2.
2.2.
Reagents
Celecoxib (TSZ Scientific, Framingham, MA) was dissolved in 100% dimethyl sulfoxide and stored at 4 C. Cultrex (Trevigen Inc, Gathersburg, MD) basement membrane extract (BME) was aliquoted and stored at 20 C. Antibodies used included antieb-catenin (Cell Signaling, Boston, MA), antieactin C-2 (Santa Cruz Biotechnology, Dallas, TX), and anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) FL-335 (Santa Cruz Biotechnology). Fluorescent antibodies for use with the LiCor Odyssey imaging system were obtained from LiCor (Lincoln, NE). Rhodamine phalloidin (Invitrogen) dissolved in methanol and DAPI (4’,6-diamidino-2-phenylindole) (Invitrogen) were used for immunofluorescence.
2.3.
Western blotting
SK-NEP1 cells were treated for 72 h with 5 mM of celecoxib for b-catenin analysis and with 5, 15, or 20 mM of celecoxib for total actin analysis. Caco-2 cells were treated for 20 h with 5 and 50 mM of celecoxib. Cells were then lysed using TNE lysis buffer (dH2O, 1 M of Tris [pH 8.0], 5 M of NaCl, 0.6 M of NaF, NP-40, protease inhibitor, and sodium orthovanadate), and protein contents were measured using the Pierce bicinchoninic acid protein assay kit (Thermo Scientific). Equal amounts of protein were loaded into Mini-PROTEAN TGX Gels (Bio-Rad, Hercules, CA) and subjected to electrophoresis. Proteins were then transferred to polyvinylidene difluoride transfer membranes (Thermo Scientific). Membranes were incubated with primary antibodies diluted in Odyssey Blocking Buffer (LiCor) at concentrations of 1:1000 (antieb-catenin), 1:5000 (antiebactin), or 1:500 (antieactin C-2 and antieGAPDH FL-335) overnight at 4 C, washed with phosphate-buffered saline (PBS)-Tween-20 thrice and incubated with their corresponding secondary antibodies (LiCor) for 1 h at room temperature. Membranes were imaged using the LiCor Odyssey System (LiCor).
j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 5 ) 1 e1 0
2.4.
Transfection
Oligomers of siRNA were diluted in serum-free McCoy’s media. Lipofectamine 2000 was likewise diluted in serum-free McCoy’s media, mixed gently, and incubated at room temperature for 5 min. After this, the siRNA and Lipofectamine were combined and incubated for 20 min at room temperature. This mixture was added to wells containing SK-NEP1 cells to a final siRNA concentration of 25 nM, and then incubated at 37 for 72 h. Knockdown was verified using Western blotting.
2.5.
Invasion assays
Invasion assays were performed using a modified protocol from Chan et al. that we have previously described [9,43]. Cultrex BME was thawed overnight on ice. Transwells with 8-mM pores (BD Falcon, Franklin Lakes, NJ) were coated with fibronectin (Corning, Bedford, MA) diluted 1:1000 in PBS, incubated at 37 C for 1 h prior, then washed with PBS thrice. SK-NEP1 cells were spun down and resuspended in cold serum-free media, then added to the cold BME. Celecoxib (5 mM) or an equal amount of vehicle (dimethyl sulfoxide) was added to each sample. The BME-cell suspension was added to each well in a 24-well plate for a total of 50,000 cells per well and was then incubated at 37 C for 30 min to allow the BME to polymerize. McCoy’s medium with serum was added to the wells beneath the transwell, whereas serum-free media were added to the top well. Plates were incubated for 72 h. Transwells were fixed with 4% formaldehyde for 15 min, then stained with crystal violet for 15 min. After rinsing with distilled water, the layer of BME was gently removed, and the invaded cells were counted under the microscope. All experiments were performed a minimum of three times, with wells in triplicate for each.
2.6.
3
For b-catenin staining, cells were grown in 4-well chamber slides (Nunc; Thermo Scientific), and a portion was treated with 5 mM of celecoxib for 72 h. Before staining, they were fixed with 4% formaldehyde for 10 min at room temperature and rinsed with PBS:glycine thrice. Cells were permeabilized with 0.1% PBSTriton X-100 for 10 min, then rinsed with immunofluorescence wash (IF-wash; NaCl, Na2HPO4, NaN3, bovine serum albumin, Triton X-100, Tween-20, pH 7.4) thrice. They were then incubated in Odyssey Blocking Buffer (LiCor) for 1 h, then incubated with antieb-catenin (Cell Signaling) diluted in blocking buffer at a concentration of 1:100 for 1 h at room temperature. Cells were then washed with IF-wash thrice and incubated with fluorescent secondary antibody 1:1000 in blocking buffer for 1 h, then rinsed again in IF-wash thrice. Slides were mounted with Prolong Gold Antifade Agent with DAPI (Life Technologies, Thermo Scientific). The cells were imaged using the Zeiss Axiovert 200M Motorized Inverted Microscope and AxioVision 4.7 software.
2.7.
Quantitative polymerase chain reaction
RNA was extracted from treated cells using the Roche High Pure Isolation Kit (Roche, Indianapolis, IN). The LightCycler 480 RNA Master Hydrolysis Probes kit was used to prepare the quantitative polymerase chain reaction (qPCR) mixtures, and 17.5 mL of the mixture was added to each well, followed by 2.5 mL of RNA. The probes and primers used included b-catenin
Immunofluorescence
For phalloidin staining, cells were grown on poly-L-lysine coverslips (Corning) coated with 50 mM of fibronectin (Corning) for 1 h. Cells were treated with celecoxib at concentrations of 5, 15, 20 mM, or vehicle for 72 h and then fixed with 4% formaldehyde microfiltered solution (Ted Pella Inc, Redding, CA) for 10 min at room temperature. Cells were permeabilized with 0.1% PBS-Triton X-100 (Eastman, Rochester, NY) for 10 min. Coverslips were then washed and transferred to parafilm-covered slides. Rhodamine phalloidin in 0.1% PBSTween-20 was added at a concentration of 1:250 for 25 min. Cells were washed with PBS-Tween-20 five times, stained with DAPI in PBS-Tween-20 at a dilution of 1:10,000 for 3 min, then washed again five times. Finally, coverslips were washed in PBS and placed onto slides with 75% Vectashield mounting medium (Vector Laboratories, Burlingame, CA) diluted with H2O, then sealed with nitrocellulose in ethyl acetate. The cells were imaged using the Zeiss Axiovert 200M Motorized Inverted Microscope (Carl Zeiss Microimaging, Thornwood, NY) and AxioVision 4.7 software (Carl Zeiss Microimaging). Image analysis was performed using ImageJ software version 1.48v (National Institutes of Health, Bethesda, MD).
Fig. 1 e (A) Western blot analysis of transfected cells shows successful knockdown with a significant decrease in bcatenin for both sets of oligomers. (B) Relative amount of bcatenin in nontargeting control siRNA transfected cells (controldnormalized to 1) versus b-catenin knockdown cells (siRNA A [ 0.34 ± 0.03 and siRNA C [ 0.29 ± 0.10). Asterisks denote values of P < 0.001.
4
j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 5 ) 1 e1 0
analyzed using the LightCycler 480 software version 1.5.0 SP3 (Roche).
2.8.
Fig. 2 e Knockdown of b-catenin produces no significant difference in number of cells invading through transwells in Boyden chamber assays. Celecoxib treatment (5 mM) significantly reduces SK-NEP1 invasion, regardless of bcatenin levels within the cells. (control [ nontargeting siRNA oligomer transfected cells, siRNA A [ cells transfected with b-catenin knockdown oligomer “A,” siRNA C [ cells transfected with b-catenin knockdown oligomer “C”). Results shown are mean ± standard deviation of four experiments done in triplicate, *P < 0.0001.
(forward primerdAACCTTTCCCATCATCGTGAG, reverse primerdTGAACCAAGCATTTTCACCAG, probedACTGGCCATCT TTAAGTCTGGAGGCATTC), GAPDH (forward primerdCCTGC ACCACCAACTGCTTAG, reverse primerdTGAGTCCTTCCACG ATACCAA, probedCCCTGGCCAAGGTCATCCA). Real-time qPCR was performed using the LightCycler 480 (Roche) and
Wound healing assay
Cell culture dishes with built-in 2-mm grids on the underside (Nunc; Thermo Scientific) were coated with 25 mg/mL of human fibronectin (Corning) for 1 h, then washed with PBS twice. SK-NEP1 cells were then plated, treated with celecoxib, 15 mM, or vehicle for 72 h and allowed to grow to complete confluency. A 200-mL pipette tip was used to scratch a straight line across the plate, and images were taken at time point zero and at 4-h intervals thereafter at 10 phase contrast using the Zeiss Axiovert 200M microscope. Measurements were made using AxioVision 4.7 software. All experiments were performed a minimum of three times.
2.9.
Statistical analysis
Statistical analysis was performed using GraphPad Prism software version 6.05 (GraphPad Software Inc, La Jolla, CA). Analysis was performed using the Student unpaired t-test and one-way and two-way analysis of variance tests with post hoc multiple comparisons using Tukey procedure. Differences were considered statistically significant at P < 0.05.
3.
Results
3.1.
Knockdown of b-catenin does not inhibit invasion
Significant knockdown of b-catenin was accomplished with multiple siRNA’s, verified through Western blotting (Fig. 1).
Fig. 3 e Western blot analysis of control- and celecoxib-treated cells. (A) In SK-NEP1 cells, treatment with celecoxib shows no significant difference in b-catenin levels, P [ 0.829. U [ untreated, C [ 5 mM of celecoxib-treated cells. (B) In Caco-2 cells, treatment with 50 mM of celecoxib resulted in a significant decrease in b-catenin levels, confirming that the drug functioned as expected, *P < 0.005. U [ untreated, 5 mM and 50 mM [ celecoxib-treated doses. Results in graphs are mean ± standard deviation of three experiments.
j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 5 ) 1 e1 0
Boyden chamber invasion assay analysis revealed no significant difference in the number of invaded cells for either siRNA versus the nontargeting control siRNA oligomer (Fig. 2). The group of control siRNA cells had 181.2 49.2 cells invade through the transwell. The b-catenin knockdown SK-NEP1’s (“siRNA A” and “siRNA C”) had 178.2 43.0 cells and 185.7 57.4 cells invade, respectively (P ¼ 0.92).
3.2. Celecoxib significantly inhibits invasion independent of b-catenin knockdown We treated both control siRNA cells and b-catenin siRNA transfected cells with 5 mM concentrations of celecoxib for 72 h. Celecoxib inhibited cellular invasion independent of the level of b-catenin within the cells (Fig. 2). Two-way analysis of variance analysis of these results confirmed that celecoxib treatment significantly inhibited invasion, regardless of bcatenin status (P < 0.0001). Control siRNA transfected cells not exposed to celecoxib had an average of 181.2 49.2 cells invade compared with 123.3 20.0 for those treated with celecoxib. The wells with untreated siRNA A cells had 178.2 43.0 invade, compared with 130.6 18.9 for treated
5
wells. For siRNA C, untreated wells had 185.7 57.4 cells invade, and celecoxib-treated wells had 149.6 37.2.
3.3. Celecoxib does not alter b-catenin messenger RNA, protein levels, or cellular location in ES cells at the doses that inhibit invasion qPCR of b-catenin messenger RNA levels of celecoxib-treated cells showed no significant change compared with those of untreated cells (data not shown). Western blotting of SK-NEP1 cells treated with 5 mM of celecoxib (the concentration necessary to inhibit invasion, Fig. 2) for 72 h showed no significant reduction in the amount of b-catenin protein present (Fig. 3A). Celecoxib did suppress levels of b-catenin in Caco-2 colon cancer cells at the doses previously reported by Maier et al. [20], confirming that our drug functioned as expected (Fig. 3B). Immunofluorescence staining for b-catenin in untreated ES cells was compared with those treated for 72 h and revealed no appreciable change in cellular location of b-catenin (Fig. 4A and B).
3.4.
Celecoxib decreases total actin
Because actin plays a key role in cell motility, as discussed previously, we asked whether it might be a target for celecoxib. Western blotting of SK-NEP1 treated and untreated cells showed a significant decrease in the total amount of
Fig. 4 e Treatment with celecoxib does not alter the location of b-catenin within the cell. In both the untreated control (A) and celecoxib-treated (B) cells, b-catenin (rhodamine [red] staining) remains localized to the cell membrane. (Color version of the figure is available online.)
Fig. 5 e Western blot analysis of total actin. Celecoxib treatment significantly decreases the total amount of actin within SK-NEP1 cells. Graph: Actin levels were normalized to GAPDH levels and expressed relative to untreated cells. Results shown are mean ± standard deviation of three experiments. Asterisks denote P values of
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.JournalofSurgicalResearch.com
Celecoxib inhibits Ewing sarcoma cell migration via actin modulation Christopher A. Behr, MD,a,b,* Anthony J. Hesketh, MD, MS,a,b,c Meade Barlow, MD,a,b Richard D. Glick, MD,a Marc Symons, PhD,b,c Bettie M. Steinberg, PhD,b,c,1 and Samuel Z. Soffer, MDa,b,1 a
Division of Pediatric Surgery, Department of Surgery, Hofstra North Shore-LIJ School of Medicine, New Hyde Park, New York b The Feinstein Institute for Medical Research, Center for Oncology and Cell Biology, North Shore-LIJ Health System, Manhasset, New York c The Elmezzi Graduate School of Molecular Medicine, The Feinstein Institute for Medical Research, North Shore-LIJ Health System, Manhasset, New York
article info
abstract
Article history:
Background: Ewing sarcoma (ES) is an aggressive childhood solid tumor in which 30% of
Received 2 January 2015
cases are metastatic at presentation, and subsequently carry a poor prognosis. We have
Received in revised form
previously shown that treatment with celecoxib significantly reduces invasion and
16 March 2015
metastasis of ES cells in a cyclooxygenase-2-independent fashion. Celecoxib is known to
Accepted 27 March 2015
downregulate b-catenin independently of cyclooxygenase-2. Additionally, the actin cyto-
Available online xxx
skeleton is known to play an important role in tumor micrometastasis. We hypothesized that celecoxib’s antimetastatic effect in ES acts via modulation of one of these two targets.
Keywords:
Methods: ES cells were treated with celecoxib, and the levels of b-catenin and total actin were
Ewing sarcoma
examined by Western blot and quantitative polymerase chain reaction. Cells were trans-
Celecoxib
fected with small interfering RNA targeting b-catenin, and invasion assays were performed.
Metastasis
Immunofluorescence staining for b-catenin and F-actin was performed on treated and un-
Actin
treated cells. Additionally, cells were subjected to a wound healing assay to assess migration.
b-catenin
Results: Celecoxib had no effect on the messenger RNA or protein levels of b-catenin but did
Cytoskeleton
significantly decrease the amount of total actin within ES cells. Reduction of b-catenin by small interfering RNA had no effect on invasion, and celecoxib treatment of the b-catenin depleted cells continued to inhibit invasion. Immunofluorescence staining demonstrated no change in b-catenin with treatment but did show a significant reduction in the amount of F-actin, as well as morphologic changes of the cells. Wound healing assays demonstrated that celecoxib significantly inhibited migration. Conclusions: Celecoxib does not exert its antimetastatic effects in ES through alteration of b-catenin but does significantly modulate the actin cytoskeleton. ª 2015 Elsevier Inc. All rights reserved.
* Corresponding author. The Feinstein Institute for Medical Research, Center for Oncology and Cell Biology, 350 Community Drive, Lab 2262, Manhasset, NY 11030. Tel.: þ1 516 562 1053; fax: þ1 516 562 1022. E-mail address: [email protected] (C.A. Behr). 1 Co-senior authors. 0022-4804/$ e see front matter ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jss.2015.03.085
2
1.
j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 5 ) 1 e1 0
Introduction
Ewing sarcoma (ES) and the ES family of tumors are a group of malignant tumors of the bone and soft tissue. They are characterized as small, round blue cell tumors, which typically affect children between the ages of 5e15 and are often quite aggressive [1]. Although the overall 5-y survival for ES is between 60 and 70%, approximately 25% of patients present with metastatic disease, which carries a poor prognosis [2e4]. Fewer than 40% of patients with distant metastasis on initial diagnosis survive past 5 y [5,6]. Strategies to treat and prevent metastasis continue to be the focus of significant research. We have previously shown that the selective cyclooxygenase-2 (COX-2) inhibitor celecoxib prevents lung metastasis in a murine model of ES, with minimal effect on the primary tumor [7,8]. Furthermore, celecoxib inhibits invasion by ES cells in a COX-2-independent fashion [9]. Recent clinical studies have shown that the addition of celecoxib and vinblastine to standard treatment protocols (alternating cycles of vincristine/doxorubicin/cyclophosphamide and ifosfamide/etoposide) in ES doubled the 2-y event-free survival for patients with lung metastases, from 36%e71% [10]. The mechanism by which celecoxib prevents malignant progression of ES has yet to be elucidated. Celecoxib exhibits many of its antitumor effects via a decrease in COX-2 activity, particularly in colon cancer [11e15]. Celecoxib inhibition of COX-2 also plays a role in treating various other malignancies including breast, endometrial, and pancreatic cancers [16e18]. However, myriad off-target effects of celecoxib treatment are being evaluated for their antitumor properties. A 2006 review by Gro¨sch et al. [19] identified over 40 different targets for the COX-2independent effects of celecoxib on apoptosis, cell cycle regulation, and angiogenesis and/or metastasis. In particular, it alters the Wnt signaling pathway through modulation of b-catenin. Studies in colorectal cancer models have demonstrated that celecoxib works in part by inducing the rapid translocation of predominantly membrane-bound b-catenin to the cytoplasm and nucleus, followed by its degradation [20e23]. Other cancers, such as the lung, liver, and glioblastoma, have likewise shown a b-catenin response to celecoxib treatment [24e27]. Furthermore, b-catenin has been shown to specifically influence the metastatic potential of ES [28,29]. Another potential mechanism through which celecoxib may exert its antitumor effects is through modification of the actin cytoskeleton. Actin dynamics are known to be vital for cell motility, migration, shape, and cellecell junctions and interactions [30,31]. b-actin is overexpressed in various cancers [32,33] and is especially overexpressed in samples taken from metastases [32e34]. Relative invasiveness of cancer cells has been associated with increases in b-actin, as well as increases in the proportion of polymerized F-actin to monomeric G-actin, resulting in the reorganization of actin within the cell [35e38]. Although some have suggested the importance of the actin cytoskeleton in ES [39e42], little is known about celecoxib’s influence on this integral cellular protein, which may have particular importance in tumor micrometastasis.
2.
Materials and methods
2.1.
Cell lines
SK-NEP1 (ATCC, Manassas, VA, HTB-48) cells, a well-described ES cell line, were maintained in McCoy’s media (Gibco [Invitrogen], Grand Island, NY) supplemented with 15% fetal clone II serum (Hyclone; Thermo Scientific, Waltham, MA) and 1% penicillin-streptomycin, and incubated at 37 C and 5% CO2. Additionally, a portion of the cells were transfected with small interfering RNA (siRNA) oligomers targeting b-catenin or with nontargeting control siRNA (Dharmacon, Lafayette, CO). Transfection was accomplished using the Lipofectamine 2000 (Invitrogen, Carlsbad, CA) transfection reagent. Caco-2 (ATCC, HTB-37) cells, a well-described colon cancer cell line, were maintained in Delbecco Modified Eagle medium (Hyclone; Thermo Scientific) supplemented with 10% fetal clone II serum (Hyclone; Thermo Scientific) and 1% penicillinstreptomycin, and incubated at 37 C and 5% CO2.
2.2.
Reagents
Celecoxib (TSZ Scientific, Framingham, MA) was dissolved in 100% dimethyl sulfoxide and stored at 4 C. Cultrex (Trevigen Inc, Gathersburg, MD) basement membrane extract (BME) was aliquoted and stored at 20 C. Antibodies used included antieb-catenin (Cell Signaling, Boston, MA), antieactin C-2 (Santa Cruz Biotechnology, Dallas, TX), and anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) FL-335 (Santa Cruz Biotechnology). Fluorescent antibodies for use with the LiCor Odyssey imaging system were obtained from LiCor (Lincoln, NE). Rhodamine phalloidin (Invitrogen) dissolved in methanol and DAPI (4’,6-diamidino-2-phenylindole) (Invitrogen) were used for immunofluorescence.
2.3.
Western blotting
SK-NEP1 cells were treated for 72 h with 5 mM of celecoxib for b-catenin analysis and with 5, 15, or 20 mM of celecoxib for total actin analysis. Caco-2 cells were treated for 20 h with 5 and 50 mM of celecoxib. Cells were then lysed using TNE lysis buffer (dH2O, 1 M of Tris [pH 8.0], 5 M of NaCl, 0.6 M of NaF, NP-40, protease inhibitor, and sodium orthovanadate), and protein contents were measured using the Pierce bicinchoninic acid protein assay kit (Thermo Scientific). Equal amounts of protein were loaded into Mini-PROTEAN TGX Gels (Bio-Rad, Hercules, CA) and subjected to electrophoresis. Proteins were then transferred to polyvinylidene difluoride transfer membranes (Thermo Scientific). Membranes were incubated with primary antibodies diluted in Odyssey Blocking Buffer (LiCor) at concentrations of 1:1000 (antieb-catenin), 1:5000 (antiebactin), or 1:500 (antieactin C-2 and antieGAPDH FL-335) overnight at 4 C, washed with phosphate-buffered saline (PBS)-Tween-20 thrice and incubated with their corresponding secondary antibodies (LiCor) for 1 h at room temperature. Membranes were imaged using the LiCor Odyssey System (LiCor).
j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 5 ) 1 e1 0
2.4.
Transfection
Oligomers of siRNA were diluted in serum-free McCoy’s media. Lipofectamine 2000 was likewise diluted in serum-free McCoy’s media, mixed gently, and incubated at room temperature for 5 min. After this, the siRNA and Lipofectamine were combined and incubated for 20 min at room temperature. This mixture was added to wells containing SK-NEP1 cells to a final siRNA concentration of 25 nM, and then incubated at 37 for 72 h. Knockdown was verified using Western blotting.
2.5.
Invasion assays
Invasion assays were performed using a modified protocol from Chan et al. that we have previously described [9,43]. Cultrex BME was thawed overnight on ice. Transwells with 8-mM pores (BD Falcon, Franklin Lakes, NJ) were coated with fibronectin (Corning, Bedford, MA) diluted 1:1000 in PBS, incubated at 37 C for 1 h prior, then washed with PBS thrice. SK-NEP1 cells were spun down and resuspended in cold serum-free media, then added to the cold BME. Celecoxib (5 mM) or an equal amount of vehicle (dimethyl sulfoxide) was added to each sample. The BME-cell suspension was added to each well in a 24-well plate for a total of 50,000 cells per well and was then incubated at 37 C for 30 min to allow the BME to polymerize. McCoy’s medium with serum was added to the wells beneath the transwell, whereas serum-free media were added to the top well. Plates were incubated for 72 h. Transwells were fixed with 4% formaldehyde for 15 min, then stained with crystal violet for 15 min. After rinsing with distilled water, the layer of BME was gently removed, and the invaded cells were counted under the microscope. All experiments were performed a minimum of three times, with wells in triplicate for each.
2.6.
3
For b-catenin staining, cells were grown in 4-well chamber slides (Nunc; Thermo Scientific), and a portion was treated with 5 mM of celecoxib for 72 h. Before staining, they were fixed with 4% formaldehyde for 10 min at room temperature and rinsed with PBS:glycine thrice. Cells were permeabilized with 0.1% PBSTriton X-100 for 10 min, then rinsed with immunofluorescence wash (IF-wash; NaCl, Na2HPO4, NaN3, bovine serum albumin, Triton X-100, Tween-20, pH 7.4) thrice. They were then incubated in Odyssey Blocking Buffer (LiCor) for 1 h, then incubated with antieb-catenin (Cell Signaling) diluted in blocking buffer at a concentration of 1:100 for 1 h at room temperature. Cells were then washed with IF-wash thrice and incubated with fluorescent secondary antibody 1:1000 in blocking buffer for 1 h, then rinsed again in IF-wash thrice. Slides were mounted with Prolong Gold Antifade Agent with DAPI (Life Technologies, Thermo Scientific). The cells were imaged using the Zeiss Axiovert 200M Motorized Inverted Microscope and AxioVision 4.7 software.
2.7.
Quantitative polymerase chain reaction
RNA was extracted from treated cells using the Roche High Pure Isolation Kit (Roche, Indianapolis, IN). The LightCycler 480 RNA Master Hydrolysis Probes kit was used to prepare the quantitative polymerase chain reaction (qPCR) mixtures, and 17.5 mL of the mixture was added to each well, followed by 2.5 mL of RNA. The probes and primers used included b-catenin
Immunofluorescence
For phalloidin staining, cells were grown on poly-L-lysine coverslips (Corning) coated with 50 mM of fibronectin (Corning) for 1 h. Cells were treated with celecoxib at concentrations of 5, 15, 20 mM, or vehicle for 72 h and then fixed with 4% formaldehyde microfiltered solution (Ted Pella Inc, Redding, CA) for 10 min at room temperature. Cells were permeabilized with 0.1% PBS-Triton X-100 (Eastman, Rochester, NY) for 10 min. Coverslips were then washed and transferred to parafilm-covered slides. Rhodamine phalloidin in 0.1% PBSTween-20 was added at a concentration of 1:250 for 25 min. Cells were washed with PBS-Tween-20 five times, stained with DAPI in PBS-Tween-20 at a dilution of 1:10,000 for 3 min, then washed again five times. Finally, coverslips were washed in PBS and placed onto slides with 75% Vectashield mounting medium (Vector Laboratories, Burlingame, CA) diluted with H2O, then sealed with nitrocellulose in ethyl acetate. The cells were imaged using the Zeiss Axiovert 200M Motorized Inverted Microscope (Carl Zeiss Microimaging, Thornwood, NY) and AxioVision 4.7 software (Carl Zeiss Microimaging). Image analysis was performed using ImageJ software version 1.48v (National Institutes of Health, Bethesda, MD).
Fig. 1 e (A) Western blot analysis of transfected cells shows successful knockdown with a significant decrease in bcatenin for both sets of oligomers. (B) Relative amount of bcatenin in nontargeting control siRNA transfected cells (controldnormalized to 1) versus b-catenin knockdown cells (siRNA A [ 0.34 ± 0.03 and siRNA C [ 0.29 ± 0.10). Asterisks denote values of P < 0.001.
4
j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 5 ) 1 e1 0
analyzed using the LightCycler 480 software version 1.5.0 SP3 (Roche).
2.8.
Fig. 2 e Knockdown of b-catenin produces no significant difference in number of cells invading through transwells in Boyden chamber assays. Celecoxib treatment (5 mM) significantly reduces SK-NEP1 invasion, regardless of bcatenin levels within the cells. (control [ nontargeting siRNA oligomer transfected cells, siRNA A [ cells transfected with b-catenin knockdown oligomer “A,” siRNA C [ cells transfected with b-catenin knockdown oligomer “C”). Results shown are mean ± standard deviation of four experiments done in triplicate, *P < 0.0001.
(forward primerdAACCTTTCCCATCATCGTGAG, reverse primerdTGAACCAAGCATTTTCACCAG, probedACTGGCCATCT TTAAGTCTGGAGGCATTC), GAPDH (forward primerdCCTGC ACCACCAACTGCTTAG, reverse primerdTGAGTCCTTCCACG ATACCAA, probedCCCTGGCCAAGGTCATCCA). Real-time qPCR was performed using the LightCycler 480 (Roche) and
Wound healing assay
Cell culture dishes with built-in 2-mm grids on the underside (Nunc; Thermo Scientific) were coated with 25 mg/mL of human fibronectin (Corning) for 1 h, then washed with PBS twice. SK-NEP1 cells were then plated, treated with celecoxib, 15 mM, or vehicle for 72 h and allowed to grow to complete confluency. A 200-mL pipette tip was used to scratch a straight line across the plate, and images were taken at time point zero and at 4-h intervals thereafter at 10 phase contrast using the Zeiss Axiovert 200M microscope. Measurements were made using AxioVision 4.7 software. All experiments were performed a minimum of three times.
2.9.
Statistical analysis
Statistical analysis was performed using GraphPad Prism software version 6.05 (GraphPad Software Inc, La Jolla, CA). Analysis was performed using the Student unpaired t-test and one-way and two-way analysis of variance tests with post hoc multiple comparisons using Tukey procedure. Differences were considered statistically significant at P < 0.05.
3.
Results
3.1.
Knockdown of b-catenin does not inhibit invasion
Significant knockdown of b-catenin was accomplished with multiple siRNA’s, verified through Western blotting (Fig. 1).
Fig. 3 e Western blot analysis of control- and celecoxib-treated cells. (A) In SK-NEP1 cells, treatment with celecoxib shows no significant difference in b-catenin levels, P [ 0.829. U [ untreated, C [ 5 mM of celecoxib-treated cells. (B) In Caco-2 cells, treatment with 50 mM of celecoxib resulted in a significant decrease in b-catenin levels, confirming that the drug functioned as expected, *P < 0.005. U [ untreated, 5 mM and 50 mM [ celecoxib-treated doses. Results in graphs are mean ± standard deviation of three experiments.
j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 5 ) 1 e1 0
Boyden chamber invasion assay analysis revealed no significant difference in the number of invaded cells for either siRNA versus the nontargeting control siRNA oligomer (Fig. 2). The group of control siRNA cells had 181.2 49.2 cells invade through the transwell. The b-catenin knockdown SK-NEP1’s (“siRNA A” and “siRNA C”) had 178.2 43.0 cells and 185.7 57.4 cells invade, respectively (P ¼ 0.92).
3.2. Celecoxib significantly inhibits invasion independent of b-catenin knockdown We treated both control siRNA cells and b-catenin siRNA transfected cells with 5 mM concentrations of celecoxib for 72 h. Celecoxib inhibited cellular invasion independent of the level of b-catenin within the cells (Fig. 2). Two-way analysis of variance analysis of these results confirmed that celecoxib treatment significantly inhibited invasion, regardless of bcatenin status (P < 0.0001). Control siRNA transfected cells not exposed to celecoxib had an average of 181.2 49.2 cells invade compared with 123.3 20.0 for those treated with celecoxib. The wells with untreated siRNA A cells had 178.2 43.0 invade, compared with 130.6 18.9 for treated
5
wells. For siRNA C, untreated wells had 185.7 57.4 cells invade, and celecoxib-treated wells had 149.6 37.2.
3.3. Celecoxib does not alter b-catenin messenger RNA, protein levels, or cellular location in ES cells at the doses that inhibit invasion qPCR of b-catenin messenger RNA levels of celecoxib-treated cells showed no significant change compared with those of untreated cells (data not shown). Western blotting of SK-NEP1 cells treated with 5 mM of celecoxib (the concentration necessary to inhibit invasion, Fig. 2) for 72 h showed no significant reduction in the amount of b-catenin protein present (Fig. 3A). Celecoxib did suppress levels of b-catenin in Caco-2 colon cancer cells at the doses previously reported by Maier et al. [20], confirming that our drug functioned as expected (Fig. 3B). Immunofluorescence staining for b-catenin in untreated ES cells was compared with those treated for 72 h and revealed no appreciable change in cellular location of b-catenin (Fig. 4A and B).
3.4.
Celecoxib decreases total actin
Because actin plays a key role in cell motility, as discussed previously, we asked whether it might be a target for celecoxib. Western blotting of SK-NEP1 treated and untreated cells showed a significant decrease in the total amount of
Fig. 4 e Treatment with celecoxib does not alter the location of b-catenin within the cell. In both the untreated control (A) and celecoxib-treated (B) cells, b-catenin (rhodamine [red] staining) remains localized to the cell membrane. (Color version of the figure is available online.)
Fig. 5 e Western blot analysis of total actin. Celecoxib treatment significantly decreases the total amount of actin within SK-NEP1 cells. Graph: Actin levels were normalized to GAPDH levels and expressed relative to untreated cells. Results shown are mean ± standard deviation of three experiments. Asterisks denote P values of
