Relationship between p-cofilin and cisplatin resistance in patients with ovarian cancer and the role of p-cofilin in prognosis

Abstract

OBJECTIVE:

This study aims to determine the correlation between p-cofilin expression and cisplatin resistance in patients with ovarian cancer, and also to investigate the role of p-cofilin in prognosis.

PATIENTS AND METHODS:

The ovarian cancer cell line A2780/DDP resistant to cisplatin was prepared. The cell resistance to cisplatin was measured via MTT assay. The cell invasion capacity was identified via Transwell assay. The mRNA expression and protein level was evaluated via semi-quantitative PCR and Western blot, respectively. The tumor tissues of patients with cisplatin-resistant ovarian cancer were collected. The relationship between prognosis and p-cofilin expression was analyzed.

RESULTS:

The growth rate of A2780 was similar to that of A2780/DDP. The sensitivity of A2780 to cisplatin was significantly higher than that of A2780/DDP ( 0.01). The migration capacity of A2780/DDP was significantly increased compared to that of A2780 ( 0.01), indicating that the cisplatin-resistant cell lines were successfully constructed. Both CFL1 mRNA level and p-cofilin level in A2780/DDP was significantly higher than that in A2780 ( 0.01). The p-cofilin level in cancer tissues in patients with cisplatin-resistant ovarian cancer was significantly higher than that in patients with cisplatin-sensitive ovarian cancer ( 0.01). The cisplatin resistance was positively correlated with the p-cofilin expression level ($r=$ 0.802, $p=$ 0.023). The survival time of patients with normal or low level of p-cofilin was significantly longer than that of patients with high expression.

CONCLUSION:

The cisplatin resistance of ovarian cancer is closely related to the expression level of p-cofilin, which affects the prognosis of patients with ovarian cancer.

1. Introduction

The incidence of ovarian cancer ranks the third in the gynecological malignancy at present, which causes high mortality. Cisplatin is extensively used for the treatment of tumors as a kind of chemotherapeutic agent, inefficacy due to the drug resistance remains impede its further application in clinic [1, 2, 3]. Cofilin is a member of actin-binding protein family encoded by CFL1 gene, and it can affect the recombination of actin skeleton via promoting the circulation of actin microfilaments in mammals, which is one of the important proteins of cytoskeleton shaping [4, 5, 6]. Several studies have demonstrated the functional roles of cofilin phospho-regulation in cell migration, tumor invasion, mitosis, neuronal development, and synaptic plasticity [7]. Accumulating evidence demonstrates that the phospho-regulation of cofilin is a key convergence point of cell signaling networks that link extracellular stimuli to actin cytoskeletal dynamics and perturbations in the normal control of cofilin activity underlie many pathological conditions, including cancer metastasis and neurological and cardiovascular disorders [7]. Goyal et al. [8] studied and showed that cofilin can affect the direction of cell movement via inducing the formation of lamellar pseudopodia. Li et al. [9] found that taxo-resistance easily occurs in patients with high expression of p-cofilin, suggesting that it may be related to tumor resistance. But there is no report on the exact relation of p-cofilin expression with cisplatin resistance. This study aims to investigate the relationship between p-cofilin level and cisplatin resistance in patients with ovarian cancer and its effect on prognosis of patients.

2. Patients and methods

2.1. Objects

A total of 70 paraffin-embedded sections of patients with ovarian cancer admitted and diagnosed via pathological examination in Weihai maternal and child health hospital after surgical resection from September 2013 to September 2015 were collected in this study. The above patients were aged from 38–87 years. Patients with other types of diseases were excluded. All patients had the complete clinical and pathological data. Patients enrolled had the 1-year follow-up records and complete treatment regimens. All cancer tissue samples were obtained and stored in liquid nitrogen. The study was approved by the ethics committee of Weihai maternal and child health hospital and all the participants have signed the informed consent.

2.2. Instruments and materials

Human ovarian adenocarcinoma cell A2780 was purchased from Shanghai Cell Bank (Chinese Academy of Sciences, Shanghai, China). Methyl thiazolyl tetrazolium (MTT) and dimethylsulfoxide (DMSO) was from Sigma (Temecula, CA, USA). Cisplatin was provided from Shanghai Aladdin Biochemical Technology Co., Ltd. (Shanghai, China). TRIzol kit, SSIII reverse transcription kit and enhanced chemiluminescence (ECI) solution were offered by Invitrogen (Carlsbad, CA, USA). Rabbit anti-p-cofilin and glyceraldehyde-phosphate dehydrogenase (GAPDH) were purchased from Cell Signaling Technology (Beverly, MA, USA). DreamTaq PCR Master Mix (2X) and inversed fluorescent microscope was made from Thermo Fisher Scientific (Waltham, MA, USA). Cell culture flask was obtained from Corning (Waltham, MA, USA). The pipettor and transwell chamber were from Eppendorf (Hamburg, Germany). PCR instrument was provided by ABI (Waltham, MA, USA). UV imaging system Biometra (Berlin, Germany), electronic balance (BP121S) was purchased from Sartorious (Berlin, Germany).

2.3. Construction of ovarian cancer-resistant cell lines

After human ovarian adenocarcinoma cell A2780 was resuscitated, it was cultured in the incubator, and the in vitro cisplatin-resistant cell line was constructed by stepwise progressive concentration method. The cells in the exponential growth phase were placed into the culture flask, and then cisplatin with concentrations of 0.1, 0.2, 0.5, 1 and 2 $\mu$g/mL was added and the treatment was repeated 5 times for each concentration. After cells were cultured for 48 h, the medium was discarded and the fresh medium was added. After 12 months, the stable ovarian cancer cell line A2780/DDP resistant to cisplatin was constructed at the concentration of 2 $\mu$g/mL, and the cell lines were cryopreserved in liquid nitrogen.

2.4. Determination of cell growth cycle and drug sensitive test

Determination of growth cycle: The cisplatin- resistant ovarian cancer cell lines in the exponential growth phase were prepared into the suspension and placed onto the culture dish at the concentration of 1 $\times$ 10${}^5$ cells/5 mL. The cells were collected and counted every 24 h for a total of 7 days. The growth curve was drawn and the doubling time was measured. And A2780 cell line was used as the control group. Drug sensitive test: A2780/DDP suspension in the exponential growth phase was collected and inoculated onto the 96-well plate (1 $\times$ 10${}^5$ cells/well). The control group was applied according to the experimental requirements. DDP with 8 different concentration gradients was added. After treatment for 72 h, 5% MTT was added and the cells were cultured for another 4 h. Then the culture solution was discarded and 150 $\mu$L dimethyl sulfoxide (DMSO) was added. The absorbance value was measured at 492 nm using the multifunctional microplate reader; and IC${}_{50}$ values of drugs were calculated to determine whether the cisplatin-resistant cell lines were constructed successfully.

2.5. Determination of cell invasion capacity

The cell invasion capacity was detected via Transwell assay. After the cisplatin-resistant ovarian cancer cell line A2780/DDP in good growth status was selected, the cell concentration was adjusted to 5 $\times$ 10${}^5$ cells/mL and added to Transwell chamber 2 h later. The cell line A2780 was used as the blank control group. After staining and fixation, the number of cells migrating through the chamber was counted under the microscope.

2.6. Determination of CFL1 gene expression level via semi-quantitative PCR

The total RNA in cisplatin-resistant ovarian cancer cell line A2780/DDP treated with cisplatin (10 $\mu$g/mL) for 24 h and cell line A2780 was extracted using the kit strictly according to the instructions of TRIzol kit. In brief, cells were scraped and washed with phosphate buffered solution (PBS). Cells were centrifuged at 3000 g for 10 min and the precipitation was treated with Trizol. For tissue samples, p-cofilin in tumor tissues and para-cancer tissues were grinded in liquid nitrogen and treated with Trizol (100 mg tissue: 1 ml Trizol). 200 $\mu$L chloroform was then added. After vibrated for 15 sec, the upper aqueous phase was added with 500 $\mu$L isopropanol for 10 min. After centrifuged at 12000 g for 10 min, the precipitation was added with 1 mL ethanol (75%). After centrifuged at 4${}^{\circ }$C and 7500 g for 5 min, the supernatant was removed and the tube was dried for 10 min. Next, the RNA was solved in diethyl pyrocarbonate (DEPC) water and qualified by 0.8% agarose gel electrophoresis. RNA content and purity were determined by ultraviolet spectrophotometer. The RNA integrity was confirmed via agarose gel electrophoresis. The results of electrophoresis showed that Band 28S, 18S and 5S were clear and the brightness of Band 28S was about twice of that of Band 18S, indicating that the RNA is intact and can be used for subsequent experiments. The reverse transcription was performed to obtain cDNA using the reverse transcription kit. Reverse transcription was performed in a 20 $\mu$L system including 4 $\mu$L 25 mM MgCl${}_2$, 2 $\mu$L Reverse Transcription 10X Buffer, 2 $\mu$L 10 mM dNTP Mixture, 0.5 $\mu$L Ribonuclease Inhibitor, 0.5 $\mu$L AMV Reverse Transcriptase, 1 $\mu$L Random Primers, 1$\mu$g total RNA and Nuclease-Free Water to a final volume of 20 $\mu$L under 42${}^{\circ }$C for 15 min and 85${}^{\circ }$C denature (Promega, Fitchburg, WI, USA). According to the instruction of DreamTaq PCR Master Mix (2X), PCR amplification was performed under the following conditions: 94${}^{\circ }$C per-denature for 5 min, followed by 30 cycles each containing 94${}^{\circ }$C denature for 30 sec, 57${}^{\circ }$C annealing for 30 sec and 72${}^{\circ }$C elongation for 20 sec. Using GAPDH as the internal reference. In the system of 25 $\mu$L, 12.5 $\mu$L Master Mix, 1 $\mu$L forward primers, 1 $\mu$L reverse primers, 2 $\mu$L template were included. Primer sequence was shown in Table 1. Semi-quantitative analysis was performed by gel imaging analysis system to calculate relative absorption value.

Table 1.

PCR primers

	Sequence
p-cofilin	Forward primer: 5’-TGGCCCTCGTAGCCTTGAGGA
	C-3’
	Reverse primer: 5’-CCAGTGCTGCAGGGTCCGAGG
	T-3’
$\beta$-actin	Forward primer: 5’-GATGATTGGCATGGCTTT-3’
	Reverse primer: 5’-CACCTTCCGTTCCAGTTT-3’

2.7. Detection of p-cofilin expression level via Western-blot

The proteins in cisplatin-resistant ovarian cancer cell line A2780/DDP and cell line A2780 were extracted. After centrifugation, all the supernatant was removed and the total protein was obtained. For tissues, the cancer tissue samples of patients with cisplatin-resistant ovarian cancer and patients with cisplatin-sensitive ovarian cancer were taken from the liquid nitrogen. The tissues were cut up with scissors and then the lysis buffer (1:20) was added for homogenate. After centrifugation, all the supernatant was removed and the total protein was obtained. The protein loading samples were prepared via protein quantification into the loading sample system with the same concentration. Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) was performed for loading, and the target band was cut off after sealing and cleaning. The p-cofilin (ab100836, Abcam, Cambridge, MA, USA) or total cofilin (ab42824, Abcam) primary antibody was incubated at 4${}^{\circ }$C overnight. After washing with Tris Buffered Saline Tween (TBST) for three times, the secondary antibody was incubated for 2 h at room temperature. After washing with TBST for three times, enhanced chemiluminescence solution (Millipore, Temecula, CA, USA) was added for color development using a developing machine (Bio-rad, Hercules, CA, USA) to calculate the relative expression level of the corresponding protein. The p-cofilin expression level in ovarian cancer was detected with $\beta$-actin as the internal reference.

2.8. Statistical analysis

The data in this study were presented as mean $\pm$ standard deviation and analyzed using SPSS19.0 software. $t$ test was used for measurement data, chi-square test was used for enumeration data, and Pearson analysis was used for correlation analysis. Bonferronic method was used for the pairwise comparison under homogeneity of variance, while Welch method was used under heterogeneity of variance. Dunnett’s T3 method was used for multiple comparisons. A significant level was defined when 0.05.

3. Results

3.1. Cell growth cycle and doubling time

The cisplatin-resistant ovarian cancer cell line A2780/DDP was constructed and the cell line A2780 was used as the control group. MTT data showed that the growth rate of A2780 was similar to that of A2780/DDP ($p>$ 0.05). The doubling time of A2780/DDP was also identical to that of A2780 ($p>$ 0.05) (Fig. 1).

Figure 1.

Growth curves of A2780/DDP and A2780. The growth rate and doubling time between A2780/DDP and A2780 are similar ($p>$ 0.05).

3.2. Sensitivity of A2780/DDP and A2780 to cisplatin

Eight different concentration gradients of DDP were added into the cisplatin-resistant ovarian cancer cell line A2780/DDP, and A2780 was used as the control group. The sensitivity of A2780/DDP to the above drug was detected via MTT and the IC${}_{50}$ values were calculated. The results are shown in Table 2. We found the IC${}_{50}$ values of A2780 and A2780/DDP were 2.28 $\pm$ 0.21 and 37.27 $\pm$ 0.67, respectively, indicating that the sensitivity of A2780/DDP cells to cisplatin was significantly inhibited ( 0.01).

Table 2.

Sensitivity of A2780/DDP and A2780 to cisplatin

Agent	IC50 (mean $\pm$ SD, $\mu$g/mL)		$p$
	A2780	A2780/DDP
DDP	2.28 $\pm$ 0.21	37.27 $\pm$ 0.67	0.0087

3.3. Determination of cell migration capacity

The migration capacity of drug-resistant cell line A2780/DDP was detected via Transwell assay. Of note, the migration capacity of A2780/DDP was significantly increased compared with that of A2780 ( 0.01) (Fig. 2).

Figure 2.

Migration capacity of A2780/DDP via Transwell assay. The migration capacity of A2780/DDP was significantly increased compared with that of A2780 ( 0.01).

3.4. Determination of CFL1 expression via semi-quantitative PCR

The expression levels of p-cofilin gene CFL1 in A2780/DDP and A2780 cells were detected via semi-quantitative PCR. We observed that the mRNA level of CFL1 in A2780/DDP was significantly higher than that of A2780 cells ( 0.01) (Fig. 3).

Figure 3.

CFL1 gene expression via semi-quantitative PCR. A, Diagram of agarose gel electrophoresis. B, The relative expression level of CFL1. The results showed that the CFL1 mRNA expression level in A2780/DDP was not significantly changed, but that in A2780 was significantly decreased ( 0.01).

3.5. Detection of p-cofilin protein expression via Western-blot

The p-cofilin protein expression in A2780/DDP and A2780 were detected via Western-blot. In similar scenario, the results indicated that the p-cofilin protein expression in A2780/DDP was significantly increased compared to that of A2780 ( 0.01) (Fig. 4).

Figure 4.

p-cofilin protein level. A, Western blot. B, Statistical chart of p-cofilin protein expression level. The results indicated that the p-cofilin protein level in A2780/DDP was significant higher than that in A2780 ( 0.01).

3.6. Detection of p-cofilin expression in cancer tissues in patients with ovarian cancer via Western-blot

Also, we detected the p-cofilin expression levels of cancer tissues from patients with cisplatin-resistant ovarian cancer and cisplatin-sensitive ovarian cancer. We found the level in cancer tissues in patients with cisplatin-resistant ovarian cancer was significantly higher than that in patients with cisplatin-sensitive ovarian cancer ( 0.01) (Fig. 5).

Figure 5.

p-cofilin level in cancer tissues in patients with ovarian cancer. A, Western blot. B, Statistical chart of p-cofilin protein expression level. The p-cofilin protein expression level in cancer tissues in patients with cisplatin-resistant ovarian cancer was significantly higher than that in patients with cisplatin-sensitive ovarian cancer ( 0.01).

3.7. Analysis of the correlation between p-cofilin protein expression level and cisplatin resistance in patients with ovarian cancer

The severity of cisplatin resistance in 78 patients with ovarian cancer was identified and divided into Grade 0–3 according to the clinical resistance criteria. The p-cofilin protein level was detected by western blot, and the correlation analysis of the p-cofilin protein expression and cisplatin resistance was performed. Pearson correlation analysis showed that the cisplatin resistance in patients with ovarian cancer was positively correlated with the p-cofilin protein expression level ($r=$ 0.802, $p=$ 0.023) (Fig. 6).

Figure 6.

Correlation analysis between p-cofilin protein expression and cisplatin resistance in patients with ovarian cancer. Cisplatin resistance in patients with ovarian cancer was positively correlated with the p-cofilin protein expression level ($r=$ 0.802, $p=$ 0.023).

3.8. Relationship between p-cofilin protein expression and prognosis of patients with ovarian cancer

The patients were followed up for 1 year. The clinical data and treatment regimens were recorded in details and the p-cofilin expression level in ovarian cancer tissues was detected. The patients were categorized into high-expression group and normal- or low-expression group according to the p-cofilin expression level. The statistical analysis of survival time among the groups showed that the survival time of patients in p-cofilin high-expression group was significantly lower than that in normal- or low-expression group ( 0.01) (Fig. 7).

Figure 7.

Relationship between p-cofilin protein expression and prognosis of patients with ovarian cancer. The survival time of patients in p-cofilin high-expression group was significantly lower than that in normal-or low-expression group ( 0.01).

4. Discussion

Ovarian cancer seriously threatens women’s health. Ovarian adenocarcinoma is a kind of ovarian cancer with the highest mortality rate clinically, and the 5-year survival rate of epithelial ovarian cancer in the world is less than 30%. No obvious pathological features are presented regarding ovarian cancer in the early phase of pathogenesis and it often developed to the advanced stage when diagnosed [10, 11, 12]. With the continuous development and progress of medical technology and anti-tumor drugs, the cure rate of tumors has been significantly improved through surgical resection combined with chemotherapy and radiotherapy, etc., but the 5-year survival rate of patients with advanced ovarian cancer is still around 15–30%. Drug resistance caused by long-term use of chemotherapy drugs in the treatment of tumors adversely affects the efficacy of therapeutic regimens against ovarian cancer [13, 14]. The drug resistance of tumor cells can be induced by many chemicals, and cofilin is one of the important factors resulting in the drug resistance of tumor cells [15]. Castro et al. [16] found that paclitaxel treatment can reduce the p-cofilin expression level in ovarian cancer cells. However, the p-cofilin expression level in drug-resistant cell lines treated with paclitaxel in the same concentration exhibited no significant change, suggesting that the p-cofilin expression level may be closely related to the paclitaxel resistance [17, 18].

In this study, the cisplatin-resistant ovarian cancer cell line A2780/DDP was constructed by inducing human ovarian adenocarcinoma cell line A2780 via cisplatin gradient. The growth rates of A2780/DDP and A2780 were detected via MTT and no significant difference was found. It suggests that the cell metabolic activity was not dramatically affected by DDP. Nevertheless, the sensitivity of A2780/DDP to cisplatin was significantly reduced, and the migration capacity of drug-resistant cell lines was significantly higher than that of A2780. P-cofilin is a major factor in the cytoskeleton and it is closely related to cell migration. Both in vitro and in vivo tests showed that p-cofilin expressions in patients with cisplatin-resistant ovarian cancer and A2780/DDP cells were significantly induced. The above results indicated indirectly that the p-cofilin expression level may be related to the mechanism of cisplatin resistance. Delorme-Walker et al. [19] found that the maintenance of cell morphology required actin, and dephosphorylation of cofilin activated actin. Nadkarni and Brieher [20] demonstrated in pancreatic cancer cell lines that the p-cofilin level in paclitaxel-sensitive cell lines was significantly lower than that in drug-resistant cell lines. The above results also revealed that the p-cofilin expression level may be associated with cell resistance. In the context, pearson correlation analysis further revealed that cisplatin resistance was positively correlated with p-cofilin expression. The therapeutic regimen and general pathological data of patients with ovarian cancer were analyzed and the relationship between the p-cofilin expression level and the prognosis of patients was investigated. The survival data showed that the survival time of patients with low- or normal-expression p-cofilin was significantly longer than that of patients with high-expression p-cofilin ( 0.05). We propose that the excessive p-cofilin expression leads to the increase of tumor cell proliferation and migration, resulting in tumor progression and metastasis. The growing p-cofilin expression can also give rise to the resistance of tumor cell to chemotherapeutic drugs, thus leading to poor prognosis of patients [21, 22].

5. Conclusion

Our data demonstrate that p-cofilin expression in patients with ovarian cancer is positively correlated with cisplatin resistance, which affects the treatment effect and prognosis of patients. Our results provide new ideas for the clinical treatment of ovarian cancer as well as clues of the mechanism of chemotherapy drug resistance.

Conflict of interest

All author declare that they have no conflict of inter- est.