Background
Ovarian cancer (OC) is the third commonly occurring gynaecological cancer and the most lethal malignancy worldwide, associated with significant mortality [
1]. Early age at menarche, late menopause, nulliparity and low parity, use of hormone replacement therapy and oral contraceptive use are well-established risk factors for OC [
2]. Patients with OC often have a poor survival rate mainly due to the resistance to the chemotherapies, and drug resistance is a complex event that involves numerous genes and interactions between pathways [
3]. Thus, identifying and understanding the underpinning molecular mechanisms relevant to chemoresistance are crucial for the management of treatment and development of novel and effective drug targets.
Forkhead box D3 (FOXD3) is a member of the forkhead box transcription factor family and often functions as a transcriptional repressor in the tumorigenesis of several types of cancers [
4,
5]. FOXD3 could act as a tumor suppressor to impair cell viability and colony formation of OC A2780 cells [
6]. The inhibiting effect of FOXD3 on the drug resistance has also been extensively reported in multiple cancers, including nasopharyngeal carcinoma and lung cancer [
7,
8]. However, its potential effect on the OC chemoresistance has been rarely reported. microRNAs (miRNAs or miRs) have been highlighted to be a potential biomarker to predict the response of malignant tumor cells to chemotherapy owing to their roles in overcoming resistance or strengthening the sensitivity of tumor cells to chemotherapeutic agents [
9]. A previously published report revealed that miR-335 was downregulated in drug-resistant OC cell lines and linked to the development of chemoresistance, representing a prognostic tool to monitor the chemotherapeutic outcomes [
10].
Interestingly, miR-335 has been documented to post-transcriptionally target disheveled-associated activator of morphogenesis 1 (DAAM1) [
11]. DAAM1 is an actin-associated regulatory factor and its overexpression results in promotion of OC cell migration and invasion [
12]. Moreover, DAAM1 can activate Rho-ROCK1/2-myosin II and thus plays pivotal roles in modulating tumor-initiating potential and local invasion as well as distant metastasis [
13]. Drug-resistant cancer cells are known to be mechanically heterogeneous, with subtypes of resistant cells exhibiting enhanced stiffness relative to their drug-sensitive cells, and myosin II has been involved in the mechanical alteration of drug-resistant cancer cells [
14]. The above discussion revealed a possible network among FOXD3, miR-335, DAAM1, and myosin II in the OC chemoresistance development. Herein, our study was conducted in a bid to reveal the specific mechanism of FOXD3/miR-335/DAAM1/myosin II axis in OC.
Discussion
Chemoresistance often causes treatment failure and is the main obstacle for OC treatment during clinical therapy [
16]. Development of new strategies for overcoming chemoresistance is thus the central goal in OC research. The findings gained from the present study uncovered the inhibiting property of FOXD3 in the malignant characteristics of OC cells and chemoresistance to CP via miR-335-mediated DAAM1 inhibition and the resultant myosin II inactivation.
Initial results indicated that overexpression of FOXD3 could promote the apoptosis while inhibiting the proliferation and chemoresistance of OC cells. FOXD3 is poorly expressed in OC tissues while its overexpression weakens OC cell proliferating and migrating ability and increases cell apoptosis while restraining tumor growth
in vivo [
17]. FOXD3 expression is downregulated in CP-resistant nasopharyngeal carcinoma cells and it can promote the expression of miR-26b which sensitizes nasopharyngeal carcinoma cells to CP [
7]. In addition, FOXD3 can repress the drug resistance of lung cancer cells
in vitro and
in vivo [
8], which is in line with our results that overexpression of FOXD3 could attenuate the chemoresistance of OC cells to CP
in vivo.
FOXD3 was previously unveiled to positively regulate the expression of miRNAs, such as miR-214 and miR-26b [
7,
18]. This study first demonstrated that FOXD3 upregulated miR-335 by enriching in the promoter region of miR-335. miR-335-5p is weakly expressed in CP-resistant A2780 cells while its overexpression can reduce cell viability and enhance apoptosis and the sensitivity of OC cells to CP by suppressing its target gene BCL2L2 [
19]. Therefore, the suppressing impact of FOXD3 on the growth of OC cells and chemoresistance to CP was associated with miR-335 upregulation.
Further investigation revealed that DAAM1 was a downstream target inhibited by miR-335. Consistently, miR-335-5p can directly target the 3’UTR of DAAM1 and consequently result in the repression of the DAAM1 expression [
20]. Meanwhile, increased expression of DAAM1 is closely associated with the distant metastasis in OC and siRNA-induced silencing of DAAM1 impairs the migration and invasion potentials of OC cells [
12]. Thus, the miR-335-mediated DAAM1 suppression may serve as therapeutic strategies for the management of OC chemoresistance. Further mechanistic findings suggested that DAAM1 silencing repressed myosin II activation in drug-resistant OC cells. Consistently, DAAM1 is capable of activating Rho-ROCK1/myosin II, and the Wnt11/5B-FZD7-DAAM1 axis could upregulate MLC2 phosphorylation and myosin II activity by activating RhoA-ROCK1/2 signal [
13]. Therefore, we speculate that silencing of DAAM1 in A2780CP cells may reduce MLC2 phosphorylation by inhibiting RhoA-ROCK1/2 signal, which warrants further validation. Meanwhile, myosin II is overexpressed in drug-resistant OC cells and essential for the migration of drug-resistant cells via modulating nuclear deformability and protease activity [
21], suggesting the potential of myosin II activation to enhance the drug resistance of OC cells. Additionally, the restored activity of myosin II in therapy-resistant melanoma cells can increase cell survival while its ablation specifically kills resistant cells through intrinsic lethal reactive oxygen species and unresolved DNA damage [
22]. Overall, the aforementioned results provided evidence that FOXD3 facilitated the miR-335-mediated DAAM1 inhibition and disrupted myosin II activity, thus impeding the growth and chemoresistance of OC cells. However, owing to the limited supportive data for the interaction between FOXD3, myosin II and DAAM1, miR-335 and myosin II, our subsequent endeavors are necessary for validating the potential of this newly discovered mechanism.
Materials and methods
Ethics statement
The study was approved by the Animal Ethics Committee of The First Affiliated Hospital of University of South China and strictly conformed to the Guidelines for the Care and Use of Laboratory Animals, with minimal suffering ensured.
Microarray-based gene expression profiling
The target genes of miR-335 was predicted using the miRDB, mirDIP, and TargetScan databases, followed by intersection analysis. The STRING database was used for interaction analysis of miR-335 target genes and interaction networks were constructed with cytoscape v3.7.1, with the degree value of core genes calculated. Finally, KEGG enrichment analyses were followed with the assistance of the KOBAS3.0 database.
Cell culture and treatment
Normal human ovarian epithelial cell line IOSE80, human OC cell lines A2780S and SKOV3 were all purchased from Meisen (Hangzhou, China). The corresponding CP-resistant OC cell lines A2780CP and SKOV3/CDDP were constructed by our laboratory. Sensitive parental A2780S cells and SKOV3 cells were co-cultured with 10–80 μM CP (Sigma-Aldrich, St. Louis, MO) for 48 h, during which the concentration of CP was gradually increased [
23,
24]. After resuscitation, these cells were cultured in DMEM (Gibco, Carlsbad, CA) containing 10% FBS (Gibco), 100 U/mL penicillin and 100 μg/mL streptomycin in a 5% CO
2 incubator (BB15, Thermo Fisher Scientific Inc., Waltham, MA) at 37℃. The medium was renewed every 24 h, and cells were treated with 0.25% trypsin (Hyclone Laboratories, Logan, UT) every 72 h and sub-cultured. Before the experiment, A2780CP cells were treated with 9 μg/mL CP (Sigma-Aldrich) for one week to maintain drug resistance.
The log-phase cells were trypsinized, seeded into 6-well plates, and cultured for 24 h. Upon reaching approximately 60% cell confluence, transfection was conducted as per the Lipofectamine 3000 transfection reagent (Thermo Fisher Scientific) with the following plasmids: oe-NC, oe-FOXD3, si-NC, si-FOXD3, mimic NC, miR-335 mimic, inhibitor NC, miR-335 inhibitor, oe-NC + inhibitor NC, oe-FOXD3 + inhibitor NC, oe-FOXD3 + miR-335 inhibitor, si-DAAM1, si-DAAM1 + si-FOXD3, oe-DAAM1 and oe-DAAM1 + oe-FOXD3. These aforementioned plasmids were synthesized by GenePharma (Shanghai, China). The medium was replaced with complete medium after 6 – 8 h of culture, followed by continued culture till 48 h. The cells were collected and used for subsequent experimentation. Then, 10 μM Myosin II inhibitor blebbistatin (Cat#120,491, Abcam Inc., Cambridge, UK) was prepared by dissolving in DMSO and subsequently added to the cells 4 h after cell seeding to facilitate adherence of the cell matrix [
21].
CCK-8 assay
Cells were seeded into 96-well plates with 1 × 104 cells per well and cultured for 24 h. When IC50 was detected, OC cells were incubated with 0, 5, 10, 20, 40 and 80 μmol/L CP (Sigma-Aldrich) for 48 h. CP was soluble in water, and the mother solution was thus prepared with distilled water before the experiment and cryopreserved at—20℃ for subsequent cell viability test. The concentration of CP was 25 μg/mL and the cells were incubated with CP for 48 h. Each well was subsequently supplemented with 10 µL of CCK-8 solution (Sigma-Aldrich) for 1-h incubation in a 37℃ incubator for 1 h. The OD value of each well was measured using an Epoch microplate spectrophotometer (Omega Bio-tek Inc, Norcross, GA) at 450 nm.
Log-phase cells were digested with 0.25% trypsin and resuspended. The cells in 6-well plates (1 × 103 cells/well) were cultured to form cell colonies within two weeks. During the transfection experiment, the medium was renewed every 2 days following 12 h of transfection. On day 6, the cells were transfected again. After another 6 days, the cells were fixed in 4% paraformaldehyde and dyed by 0.5% crystal violet (0.5% w/v, Solarbio, Beijing, China) for 15 min. Cell colonies (over 50 cells) were counted under a stereomicroscope.
Flow cytometry
Annexin V-FITC/PI double staining kit (70-AP101-100, Lanke Biotechnology Co., Ltd., Hangzhou, Zhejiang, China) was employed to examine cell apoptosis. The cells were detached with 0.25% trypsin without EDTA, centrifuged at 300 × g for 5 min, and resuspended in 500 µL of binding buffer. Next, 3 μL Annexin V-FITC and 2.5 μL PI were supplemented to the cells for 20-min incubation at 37℃ under dark conditions. A FACSCalibur flow cytometer (BD Bioscience, Franklin Lakes, NJ) was utilized for analysis.
RNA extraction and quantification
Cellular total RNA and total RNA in tissues were isolated with TRIzol reagent (15596026, Invitrogen, Carlsbad, CA). Next, the extracted RNA was reversely transcribed into cDNA with PrimeScript™ RT reagent Kit (RR047A, Takara, Japan) or PolyA tailing kit (B532451, Sangon, Shanghai, China). RT-qPCR was then performed using Fast SYBR Green PCR Kit (Applied Biosystems, Foster City, CA) and the ABI PRISM 7300 RT-PCR system (Applied Biosystems). U6 was employed as an internal reference for miR-335 and GAPDH for the remaining genes (Supplementary Table
1). The fold changes were calculated by means of relative quantification (2
−△△Ct method).
Western blot analysis
Cellular total protein and total protein in tissues were obtained using PMSF-containing RIPA lysis buffer (Beyotime Biotechnology Co., Shanghai, China) and then quantified using a BCA protein assay kit (Pierce, Rockford, IL). Afterwards, 50 μg protein following electrophoresis were transferred onto a PVDF membrane. The membrane blocked by 5% skim milk powder was then probed overnight at 4℃ with primary antibodies against FOXD3 (mouse, sc-517206, 1:500, Santa Cruz Biotechnology, Inc, Santa Cruz, CA), DAAM1 (mouse, sc-100942, 1:500, Santa Cruz Biotechnology), phosphorylated myosin light chain 2 (p-MLC2; Cat#3674S, 1:1000, Cell Signaling Technologies, Beverly, MA), MLC2 (Cat#3672, 1:1200, Cell Signaling Technologies) and GAPDH (mouse, sc-47724, 1:100–1:1000, Santa Cruz Biotechnology, serving as a loading control). HRP-conjugated secondary goat anti-mouse IgG (1:10,000, Boster Biological Technology Co., Ltd., Wuhan, Hubei, China) was supplemented for 1 h of re-probing. The immunocomplexes on the membrane were visualized using enhanced chemiluminescence reagent, and quantitative analysis of band intensities was done using Image J software, with the expression of GAPDH as the internal reference.
Dual luciferase reporter assay
The sequence containing the targeted binding site between miR-335 and DAAM1 was obtained through TargetScan database. The DAAM1-3’UTR-WT or DAAM1-MUT sequence was synthesized and cloned into pmiR-RB-Report vector (RiboBio, Guangzhou, China). All the plasmids were extracted employing Omega plasmid small amount extraction kit (D1100-50 T, Solarbio), and the constructed recombinant plasmids were transformed and amplified. After the cell adherence, DAAM1-3'UTR-WT and DAAM1-3'UTR-MUT reporter plasmids were co-transfected with miR-335-5p mimic/NC mimic into HEK-293 T cells (2 × 105 cells/well) in 6-well plates. After 48 h of culture, the cells were collected for luciferase activity detection.
The JASPAR website was adopted to acquire the putative binding sites of FOXD3 in miR-335-5p promoter. The 2 kb region upstream of miR-335-5p was amplified by PCR, and then the product was cloned into the vector. The binding of FOXD3 to the promoter of miR-335-5p was detected by transient transfection of pGL3 promoter/oe-FOXD3/pRL-TK, pGL3 promoter/vector/PRL-TK or pGL3-basic/pRL-TK into HEK293T cells. Following 48 h of transfection, the changes in luciferase activity were detected using a dual-luciferase assay detection kit (D0010, Solarbio) on a Glomax 20/20 luminometer (E5311, Shaanxi Zhongmei Biotechnology Co., Ltd., Shaanxi, China).
ChIP
Cells were fixed with 1% formaldehyde for 10 min to produce DNA–protein cross-linking. Then, the cells lysate was subjected to ultrasonic treatment to generate 200—1000 bp chromatin fragments. An overnight incubation was followed at 4℃ with antibodies against FOXD3 (sc-517206, Santa Cruz Biotechnology) and IgG (serving as NC). The Pierce protein A/G beads (88,803, Thermo Fisher Scientific) were used to immunoprecipitate the DNA that could bind to FOXD3. After 5-min centrifugation at 12,000 × g and washing, the DNA–protein complex was incubated at 65℃ overnight to relieve cross-linking. Ultimately, the precipitated DNA was analyzed by RT-qPCR using Bio-Rad iQ SYBR green supermix.
OC xenografts in nude mice
Thirty female BALB/c nude mice (aged 4–5 weeks old and weighing 18—22 g) were purchased from Shanghai SLAC Laboratory Animal Co., Ltd. (Shanghai, China). The mice were housed at 25—27℃ with 45—50% humidity for one week with a 12-h light/dark cycle. The mice were given ad libitum access to food and water but fasted for 12 h prior to subcutaneous administration. The mice were randomly grouped into 3 groups, 10 mice in each group: oe-NC + PBS, oe-NC + CP and oe-FOXD3 + CP. The A2780CP cells stably transfected with oe-NC and oe-FOXD3 were subcutaneously injected into the location 1–2 cm under the right armpit of mice at a density of 1 × 107 cells/mouse (0.2 mL). When the tumor volume reached 80 mm3, CP (2 mg/kg) was injected into the tumor with a micro-syringe, once every 5 days. The weight of mice was weighed before each administration, and the tumor growth was monitored by measuring the tumor width (W) and length (L) with Vernier calipers, and calculated using the formula: tumor growth = (W2 × L)/2. After 30 days, the mice were euthanized and the tumor tissue was removed. The isolated tumors were photographed with a camera after weighting.
HE staining
OC tissues were fixed with 4% paraformaldehyde, paraffin-embedded and sectioned. The sections were heated at 60℃ for 1 h, dewaxed with xylene, hydrated with gradient alcohol and stained with hematoxylin (Solarbio) for 2 min. Following tap water washing for 10 s, the sections were immersed in 1% hydrochloric acid–ethanol for 10 s, and colored with eosin (Solarbio) for 1 min. The sections were lastly subjected to observation with the XP-330 optical microscope (Shanghai Bingyu Optical Instrument Co., Ltd., Shanghai, China).
Immunohistochemical staining
OC tissues collected from nude mice were fixed with 4% paraformaldehyde, paraffin-embedded and cut into 4-μm-thick sections. The sections were subjected to microwave-stimulated antigen retrieval in 0.1 M citric acid buffer (pH 6.0) for 10 min and then immunostained with primary antibody against MLC2 (phosphor-Ser 19) (11114, Signalway Antibody Co., College Park, MD, USA) at 4℃ overnight. Goat anti-rabbit (ab6721, Abcam) labeled by HRP was utilized as secondary antibody. Following that, 0.05% DAB containing 0.01% hydrogen peroxide (Beijing Bioss Biotechnology Co. Ltd., Beijing, China) was employed for color development. Thereafter, the sections were stained with hematoxylin for 5 min, soaked in 1% hydrochloric acid–ethanol for 4 s, and blued with tap water for 20 min. The cells with brownish-yellow nuclei were normal positive cells. Image-Pro Plus software (Version X; Media Cybernetics, Silver Springs, MD) was adopted to detect the average OD value of positive cells under a high power microscope, followed by quantitative analysis of protein expression: five high power microscope fields/section and 200 cells/field were observed.
Statistical analysis
All data were analyzed using SPSS 22.0 statistical software (IBM Corp. Armonk, NY) Graphpad Prism 7.0 (Graphpad software, La Jolla, CA). The measurement data were described as mean ± standard deviation. Data between two groups were compared by unpaired t-test. Differences among multiple groups were statistically analyzed employing one-way ANOVA and Tukey’s multiple comparisons test. Data among multiple groups depending on time points were compared with repeated measures ANOVA with Tukey’s post hoc test. A value of p < 0.05 was statistically significant.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.