p53R2 overexpression in cervical cancer promotes AKT signaling and EMT, and is correlated with tumor progression, metastasis and poor prognosis

ABSTRACT

p53R2 is a p53-inducible ribonucleotide reductase subunit involved in deoxyribonucleotide biosynthesis and DNA repair. Although p53R2 has been linked to human cancer, its role in cervical cancer remains unknown. In this study, we investigated the expression and clinical significance of p53R2 in early-stage cervical cancer. p53R2 expression is significantly upregulated at both mRNA and protein levels in cervical cancer cells and tissues, compared with that in matched normal cervical cells and tissues, respectively. p53R2 overexpression is associated with increased risk of pelvic lymph node metastasis (PLNM, p = 0.001) and cancer relapse (p = 0.009). Patients with high p53R2 expression have a shorter overall survival (OS) and disease-free survival (DFS). p53R2 is an independent factor for predicting OS and DFS of cervical cancer patients. We further show that p53R2 is important for oncogenic growth, migration and invasion in cervical cancer cells. Mechanistically, p53R2 promotes Akt signaling and epithelial–mesenchymal transition (EMT). In conclusion, our study demonstrates for the first time that p53R2 protein is overexpressed in early-stage cervical cancer and unravels some unconventional oncogenic functions of p53R2. p53R2 may be a useful prognostic biomarker and therapeutic target for cervical cancer.

KEYWORDS:

• Akt
• cell growth
• cervical cancer
• epithelial-mesenchymal transition
• p53R2
• prognosis

This article is referred to by:

p53R2 is oncogenic in human cervical cancer

Introduction

Cervical cancer is the fourth most frequent cancer among women worldwide. In 2012, there was an estimated 527,600 new cases and 265,700 deaths from cervical cancer in the world.¹ Treatment options for cervical cancer include surgery, chemotherapy, radiation therapy or a combination of these strategies. Although the clinical outcome for early stages of cervical cancer is excellent after treatments, the survival rate is low for late stage diseases even with treatment.² Therefore, it is necessary to identify reliable biomarkers for prognosis and new therapeutic targets of cervical cancer.

Ribonucleotide reductase (RNR) catalyzes the conversion of ribonucleotides into 2′-deoxyribonucleotides (dNDPs), which is important for DNA synthesis and repair.³ Human RNR consists of 3 subunits, the large subunit M1 (RRM1) and 2 small subunits M2 (RRM2) and M2B (p53R2). RRM1 is the catalytic subunit consisting of the catalytic site, 2 allosteric effector-binding sites, and redox-active disulfides that participate in the reduction of substrates. RRM2 and p53R2 are the regulatory subunits containing the diferric tyrosyl radical cofactor required for the enzyme activity. RRM1 interacts with either RRM2 or p53R2 to become the catalytically active form of eukaryotic RNR.^3-8

Consistent with a vital role of RNR in DNA synthesis and cell proliferation, dysregulation of RNR has been reported to be involved in tumorigenesis of several types of cancers. However, the 3 subunits of RNR seem to play different roles in different cancers. RRM1 appears to act as a tumor suppressor that inhibits the progression and metastasis of many cancers.^9-15 For example, in human colorectal cancer (CRC), ectopic overexpression of RRM1 prevents the migration and invasion of CRC cells by promoting PTEN expression.⁹ On the other hand, RRM2 is oncogenic by promoting cancer cell proliferation.^16-20 A high level of RRM2 expression is reported to correlate with tumor growth,¹⁷ invasiveness and metastasis,¹⁸ tumor angiogenesis¹⁹ and poor outcome.²¹

Although p53R2 and RRM2 share 80% amino acid sequence similarity, they have distinct functions in human cancers.²¹ p53R2 is a direct target of the tumor suppressor p53 and is involved in DNA replication and repair, cell cycle and mitochondrial homeostasis. p53R2 overexpression promotes the accumulation of p21, G1 or G2 arrest and DNA repair.^22-24 It appears to have a complex role in human cancer. In CRC cancer cells, high expression of p53R2 is negatively correlated with cancer cell proliferation, invasion and metastasis.^25,26 Disruption of the p53R2-mediated DNA repair in ulcerative colitis is implicated in the initiation of colon tumorigenesis.²⁷ In liver cancer, overexpression of p53R2 suppresses cell migration and spreading by modulating the Egr-1/PTEN/Akt1 pathway.²⁸ On the other hand, p53R2 appears to act as an oncogene with elevated expression in lung cancer and esophageal squamous cell carcinoma, and is associated with the progression of these cancers.^22,29,30 Furthermore, downregulation of p53R2 by siRNA or mutation inhibits cancer cell growth and sensitizes cancer cells to anticancer agents or ionizing radiation.^31-33

To date, the role of p53R2 in cervical cancer has not been studied. Because of its important role in tumorigenesis and therapy, we investigated the expression and clinical significance of p53R2 in a large cohort of human cervical cancer samples, and explored its biologic functions and potential regulatory mechanism in cervical cancer models.

Results

p53R2 is overexpressed in cervical cancer

To investigate the potential role of p53R2 in cervical cancer, we analyzed p53R2 expression in immortalized normal cervical cell line Ect1/E6E7 and cervical cancer cell lines including HeLa, SiHa, HeLa229, HCC-94, C-33 A and CaSki. Both p53R2 mRNA and protein levels were higher in cancer than the immortalized cells (Fig. 1A–1C). Consistently, both p53R2 mRNA and protein expression were also much higher in human cervical tumors compared with the paired adjacent noncancerous tissues, as detected by qRT-PCR and Western Blot, respectively (Fig. 1D–1F). Interestingly, the increase in p53R2 mRNA was much greater than that in p53R2 protein, suggesting that p53R2 expression is differentially regulated at the transcriptional and translational levels.

Figure 1. p53R2 expression is elevated in cervical cancer cell lines and tissues. (A) Relative p53R2 mRNA expression in normal cervical cell and cervical cancer cell lines. (B) Western blotting analysis of p53R2 protein expression in immortalized cervical cell and cervical cancer cell lines. (C) Quantification of results from Fig. 1B. (D) Relative p53R2 mRNA expression level in 8 pairs of matched cervical cancer tissues (T) and adjacent noncancerous cervical tissues (ANT). (E) Western blotting analysis of p53R2 protein expression in 8 pairs of matched cervical cancer tissues (T) and adjacent noncancerous cervical tissues (ANT). (F) Quantification of results from Fig. 1E. The expression of p53R2 was normalized against GAPDH.

To further explore the clinical significance of p53R2 overexpression in cervical cancer, we performed IHC staining for p53R2 expression in 275 paraffin-embedded cervical tumor samples. p53R2 was highly expressed in over half of the patients (58.5%, 161/275; see Material and Methods for the IHC score criteria). In contrast, p53R2 was not detectable or marginally detectable in the normal cervical tissues or the areas surrounding the cancerous tissues (Fig. 2A). There was considerable variability in the level of p53R2 among different tissue samples, ranging from negative, low, moderate, to high IHC staining (Fig. 2B). Interestingly, p53R2 was distributed in both the cytoplasm and nucleus (Fig. 2B), and in some instances, prominently concentrated in the nuclei of tumor tissues (e.g. Fig. 2B, moderate panel). The nuclear p53R2 nuclear localization was previously observed, but the significance remains largely unknown.²⁵

Figure 2. Immunohistochemistry (IHC) staining of p53R2 in matched cervical cancer tissues (T) and adjacent noncancerous cervical tissues (ANT). (A) Representative IHC staining of p53R2 in matched cervical cancer tissues (T) and adjacent noncancerous cervical tissues (ANT). (B) IHC staining of p53R2 expression in adjacent normal cervical tissues and cervical cancer tissues with different staining intensity.

p53R2 overexpression is associated with tumor progression, metastasis and poor survival

To evaluate the clinical significance of p53R2, we investigated the relationship between p53R2 expression and clinicopathological features in cervical cancer. p53R2 overexpression was significantly associated with pelvic lymph node metastasis (PLNM, p = 0.001) and cancer relapse (p = 0.009) (Table 1). A log-rank test and Kaplan-Meier analysis were used to calculate the effect of p53R2 on cervical cancer patients' survival. Patients with high p53R2 expression had a significantly shorter overall survival (OS) (HR 2.727, 95% CI 1.190–6.246, p = 0.018) and disease free survival (DFS) (HR 2.458, 95% CI 1.358–4.451, p = 0.003) than those with low p53R2 expression (Fig. 3A–3B). In our study cohort, 207 patients received adjuvant chemoradiotherapy after radical hysterectomy and 68 patients did not. Higher p53R2 expression was also significantly correlated with a shorter OS (HR 2.709, 95% CI 1.081–6.794, p = 0.034) and a shorter DFS (HR 2.517, 95% CI 1.294–4.894, p = 0.007) in this subgroup of patients with adjuvant therapy (Fig. 3C and 3D). Additionally, multivariate analysis indicated that p53R2 expression is an independent prognostic factor for OS (Table 2) and DFS (Table 3) in the patients.

Table 1. Correlation between p53R2 expression and clinicopathological characteristics in cervical cancer patients.

		p53R2 Expression
Clinicopathological characteristics	N (%)	High Exp.	Low Exp.		P value^*
Age (years)					0.270
≤46	153 (55.64%)	85		68
>46	122 (44.36%)	76		46
FIGO staging system					0.241
Stage Ia	1 (0.36%)	0		1
Stage Ib	185 (67.27%)	105		80
Stage IIa	82 (29.82%)	50		32
Stage IIb	7 (2.55%)	6		1
Pathological grade					0.188
G1	27 (9.82%)	16		11
G2	137 (49.82%)	87		50
G3	111 (40.36%)	58		53
Carcinoma diameter (cm)					0.432
≤4	166 (60.36%)	96		70
>4	109 (39.64%)	65		44
Cancer Embolus					0.390
No	234 (85.09%)	134		100
Yes	41 (14.91%)	27		14
Myometrium invasion					0.112
<1/2	131 (47.64%)	70		61
≥1/2	144 (52.36%)	91		53
Pelvic lymph node metastases					0.001
No	215 (78.18%)	115		100
Yes	60 (21.82%)	46		14
Relapse					0.009
No	241 (87.64%)	134		107
Yes	34 (12.36%)	27		7
Distant Metastasis					0.128
No	251 (91.27%)	143		108
Yes	24 (8.73%)	18		6
Adjuvant Therapy					0.202
No	68 (24.73%)	35		33
Yes	207 (75.27%)	126		81

* p values are calculated by χ 2 test or Fisher's exact test.

Figure 3. Kaplan–Meier analysis of cervical cancer patients' survival rates in association with p53R2 expression. (A) and (B). Overall survival rate (OS) and Disease free survival rate (DFS) in 275 patients with high or low p53R2 expression. (C) and (D). Overall survival rate (OS) and Disease free survival rate (DFS) in patient subgroups with or without adjuvant therapy.

Table 2. Univariate and multivariate analysis of p53R2 expression and clinicopathological features on the overall survival in cervical cancer patients.

	Univariate cox analysis		Multivariate cox analysis
Clinicopathologic features	HR (95% CI)	P	HR (95% CI)	P
Age (>46 years vs. ≤46 years)	0.933 (0.380–2.287)	0.879
Pathological grade (III, II, I)	0.881 (0.444–1.750)	0.718
Carcinoma diameter (>4 cm vs. ≤4cm)	1.191 (0.507–2.799)	0.688
Myometrium invasion (Yes vs. No)	1.516 (0.807–2.845)	0.196
Cancer Embolus (Yes vs. No)	1.374 (0.482–3.910)	0.552
Pelvic lymphnode metastases (Yes vs. No)	1.619 (0.670–3.913)	0.285
FIGO staging system (IIb, IIa, Ib, Ia)	1.773 (0.900–3.495)	0.098
P53R2 Expression (High vs. Low)	3.889 (1.130–13.378)	0.031	4.342 (1.288–14.635)	0.018

Table 3. Univariate and multivariate analysis of the effects of P53R2 expression and clinicopathological features on disease free survival in patients with cervical cancer.

	Univariate cox analysis		Multivariate cox analysis
Clinicopathologic features	HR (95% CI)	P	HR (95% CI)	P
Age (>46 years vs. ≤46 years)	0.550 (0.281–1.078)	0.082
Pathological grade (III, II, I)	0.970 (0.594–1.586)	0.905
Carcinoma diameter (>4 cm vs. ≤4 cm)	1.863 (1.011–3.433)	0.046
Myometrium invasion (Yes vs. No)	1.417 (0.944–2.125)	0.092
Cancer Embolus (Yes vs. No)	1.398 (0.645–3.033)	0.396
Pelvic lymphnode metastases (Yes vs. No)	1.766 (0.913–3.419)	0.091
FIGO staging system (IIb, IIa, Ib, Ia)	1.422 (0.845–2.394)	0.185
P53R2 Expression (High vs. Low)	2.434 (1.182–5.015)	0.016	2.604 (1.277–5.310)	0.008

p53R2 is crucial for the proliferation of cervical cancer cells

p53R2 is overexpressed in cervical cancer, which is correlated with tumor progression and poor survival, suggesting that p53R2 is important for this cancer. To understand the role of p53R2, we used previously validated siRNAs³⁴ to knock down p53R2 in HeLa and SiHa, 2 cervical cancer cell lines with high p53R2 protein expression. p53R2-specific siRNAs resulted in significant reduction in p53R2 expression as determined by Western blot (Fig. 4A). Down-regulation of p53R2 markedly inhibited the growth of these cells (Fig. 4B). Consistently, p53R2 knockdown also attenuated colony formation of these cervical cancer cells, as indicated by the decrease in colony size and number (Fig. 4C). These results demonstrate that p53R2 plays a vital role in promoting cell growth and proliferation of cervical cancer cells.

Figure 4. p53R2 is important for the growth, proliferation, migration and invasion of cervical cancer cells. (A) HeLa and SiHa cells were transfected with siRNAs and analyze for p53R2 expression by Western blot. (B) The growth of HeLa and SiHa cells transfected with siRNAs was measured by the MTS assay. (C) HeLa and SiHa cells transfected with siRNAs was measured for their ability to form colonies. (D) HeLa and SiHa cells transfected with siRNAs was measured for their mobility by the transwell migration assay. (E) HeLa and SiHa cells transfected with siRNAs was measured for their invasive ability by the transwell invasion assay. NC: cells transfected with scramble siRNAs; siRNA#1 and #2: cells transfected with 2 human siRNA sequences to repress p53R2 expression, respectively. *p < 0.05, **p < 0.01, ***p < 0.001.

p53R2 is required for the migration and invasion of cervical cancer cells

Our study of clinical samples indicates that p53R2 overexpression is associated with metastasis. To verify this, we used the transwell assay to assess the role of p53R2 in cervical cancer cell migration and invasion. p53R2 knockdown by siRNAs in HeLa and Siha cells attenuated the migration and invasion of cervical cancer cells (Fig. 4D and E). These results show that p53R2 is important for the motility and invasiveness of cervical cancer cells.

Down-regulation of p53R2 leads to inhibition of Akt signaling and reversal of epithelial-mesenchymal transition (EMT)

PI3K-Akt signaling pathway is frequently deregulated in human cancer, including cervical cancer.^35-39 Aberrant activation of the PI3K-Akt signaling contributes to cervical cancer cell proliferation, survival and angiogenesis.⁴⁰ To investigate the mechanism by which p53R2 promote cervical cancer cell growth and proliferation, we examined the effect of p53R2 knockdown on Akt signaling. Down-regulation of p53R2 expression inhibited Akt phosphorylation, but not Akt protein expression (Fig. 5A), suggesting that p53R2 promotes Akt signaling in cervical cancer cells. PTEN is an inhibitor of the PI3K/Akt pathway by antagonizing the PI3K action upstream of Akt.^28,41 Interestingly, p53R2 knockdown leads to upregulation of PTEN expression, suggesting that p53R2 blocks Akt signaling through PTEN (Fig. 5A).

Figure 5. p53R2 promotes Akt signaling and reversed EMT in cervical cancer cells. (A) Western blotting analysis of the expression of p53R2, P-Akt, Akt and PTEN. (B) The expression of EMT markers as determined by Western blot. NC: cells transfected with scramble siRNAs; siRNA#1 and #2: cells transfected with 2 human siRNA sequences to repress p53R2 expression, respectively.

The PI3K/Akt pathway plays an important role in EMT,⁴² which stimulates cancer cell invasion and metastasis through reduced cell adhesion and enhanced cancer cell dissemination. The key characteristic of EMT is the loss of epithelial markers such as E-cadherin and α-catenin, and the increase of mesenchymal markers such as N-cadherin, vimentin and β-catenin. To investigate whether p53R2 regulates the EMT process, we examined the expression of EMT markers in HeLa and SiHa cells after p53R2 knockdown. The loss of p53R2 decreased the expression of mesenchymal markers N-cadherin, β-catenin and vimentin, while increased the expression of epithelial markers E-cadherin and a-catenin (Fig. 5B). Together, these results show that p53R2 activates Akt signaling and EMT.

Discussion

The role of p53R2 in human cancer is complex, which is dependent on the tumor type and stages.^43,44 While it has been reported to be a tumor suppressor in certain cancers such as human colon adenocarcinoma and lung cancer, it appears to act as an oncogene in other cancers such as esophageal squamous cell carcinoma and melanoma. Previous studies primarily focused on the DNA repair functions using in vitro cell models. The clinical significance of p53R2 is largely unknown. Only a small cohort of human non-small cell lung cancer (NSCLC) has been previously investigated.²² In this study, we investigated the potential role of p53R2 in cervical cancer and found that p53R2 expression is upregulated in cervical cancer and that overexpression of p53R2 was significantly correlated with pelvic lymph node metastasis (PLNM, p = 0.001) and relapse (p = 0.009). Multivariate and univariate analysis demonstrate that high p53R2 expression is an independent prognostic factor for poor OS and DFS in patients with early-stage cervical cancer, indicating that p53R2 is a potentially useful biomarker for cervical cancer.

p53R2 is essential for cervical cancer cell proliferation, suggesting that it is a potential therapeutic target. Consistently, it was reported that pharmacological inhibition of RNR with 3-aminopyridine-2-carboxaldehyde thiosemicarbazone significantly enhances radiation-related cytotoxicity in cervical and colon cancer cells in a p53-independent mechanism,⁴⁵ and that down-regulating p53R2 by siRNA also increases radiosensitivity in esophageal squamous cell carcinoma.³⁰ It was shown that silencing p53R2 expression sensitizes CRC cancer cells to undergo apoptotic cell death in the presence of DNA damage agent.⁴⁶ These findings indicate that targeting p53R2 could enhance the effectiveness of ionizing radiation or DNA-damaging chemotherapy. Interestingly, in some tumors, p53R2 nuclear localization is much stronger than the cytoplasmic localization (Fig. 2B). A previous study shows that p53R2 is normally localized in the cytoplasm of several cancer cell lines, but becomes enriched in the nucleus upon radiation-induced DNA damage.^24,47 These studies suggest that nuclear localization of p53R2 directs localized reduction of ribonucleotides to generate deoxyribonucleotides that are important for supporting DNA repair. It will be interesting to determine in the future studies whether nuclear enrichment of p53R2 is associated with radiation and/or chemotherapy resistance.

Although p53R2 is clearly important for DNA repair, there are some evidence that p53R2 is involved in growth regulation. For example, p53R2 knockout mice show growth retardation.⁴⁸ However, such non-DNA repair functions remain obscure. We have investigated the biologic function and mechanism of p53R2 in cervical cancer. Here we show that p53R2 promotes cervical cancer through at least 2 distinct mechanisms. First, p53R2 stimulates PI3K-Akt signaling as evidenced by the dependency of Akt phosphorylation and repression of PTEN expression on p53R2. Interestingly, Akt phosphorylation is also regulated by p53R2 in human chondrocytes in response to mechanical stress,⁴⁹ suggesting that p53R2-dependent regulation of Akt is a common mechanism to control cell growth and survival. Second, p53R2 enhances EMT process. EMT is known to initiate the invasiveness of cancer cells, causing the primary tumor to lose cell-cell adhesion through E-cadherin repression. This provides a mechanistic explanation for increased metastasis in high p53R2 expressing cervical cancer patients. In conclusion, we demonstrate that p53R2 is overexpressed in cervical cancer, and meanwhile it may be a potential prognostic biomarker of cervical cancer. Inhibition of p53R2 may improve the treatment effect of cervical cancer. p53R2 is a potential novel molecular target for cervical cancer therapy.

Materials and methods

Clinical samples

Current study used tumor samples surgically removed from 275 early-stage cervical cancer patients who received radical hysterectomy and lymphadenectomy in the Department of Gynecologic Oncology in Sun Yat-sen University Cancer Center (SYSUCC, Guangzhou, China) enrolled between January 2000 to December 2010. They were pathologically diagnosed as cervical cancer without radiotherapy, chemotherapy, or hormonal therapy before surgery. The clinical staging and pathological grading were determined according to the 2009 FIGO criteria. Clinical/pathological features of the cervical cancer cohort are summarized in Table 1. 60 samples of paired adjacent noncancerous cervical tissues or chronic cervicitis tissues from hysteromyoma patients were obtained as controls by simple hysterectomy. 15 pairs of fresh cervical cancer tissues and paired adjacent noncancerous cervical tissues were obtained from the Department of Tumor Resource Library of SYSUCC for qPCR and Western blotting analysis. All the paraffin embedded tissues and fresh tissues were taken obtained the written informed consent of each patient.

This study was approved by the Institutional Review Boards of SYSUCC. All the patients were followed up regularly with the last follow-up in November 2016. Overall survival (OS) is defined as the time from surgery to the date of cervical cancer-associated death or last follow-up. Disease-free survival (DFS) is calculated from the date of surgery to the date of first cervical cancer-associated relapse, metastasis, death or the last follow-up.

Cell lines and antibodies

Immortalized human normal cervical cell line Ect1/E6E7 and cervical cancer cell lines HeLa, SiHa, HeLa229 and C-33A were cultured in Dulbecco's Modified Eagle Medium (GIBCO BRL). CaSki and HCC-94 were cultured in RPMI 1640 medium (GIBCO BRL). All cultures were supplemented with 10% fetal bovine serum (GIBCO BRL). Cell cultures were maintained in a humidified 95% air/5% CO₂ atmosphere at 37°C. Rabbit anti-p53R2 (#ab154194) was purchased from Abcam; antibody for Akt (#4691), p-Akt(#4060), PTEN (#9188) and GAPDH (#5174) were obtained from Cell Signaling Technology. Mouse Anti-E-cadherin (#610181), N-cadherin (#610921), α-catenin (#610194), β-catenin (#610154) and vimentin (#550513) were purchased from BD Biosciences.

Immunohistochemistry (IHC) staining

IHC staining was performed on 4 μm thick paraffin-embedded tissue sections using rabbit anti-p53R2 antibody (1:750 dilution, #ab8105, Abcam) as previously reported.³⁴ Tissue sections omitting the primary antibody were used as the negative control. Two independent pathologists, blinded to the histopathological features and patients' data of the samples, scored the IHC staining intensity based on the percentage of positive tumor cells multiplied by the intensity of staining. The percentage of positive tumor cells in the sections was scored as 1 (< 10%), 2 (11–50%), 3 (51–75%) and 4 (> 75%). The staining intensity was graded as 0 (no staining), 1 (weak), 2 (moderate) or 3 (strong). The final immunostaining score for each section ranged from 0 to 12. Tissue sections scored 0–3 were considered as low expression, and 4–12 was considered as high expression.

RNA preparation and quantitative reverse-transcription polymerase chain reaction (qRT-PCR)

Total RNA from cultured cells and fresh tissues were extracted using the Trizol reagent (15596018, Invitrogen) according to the manufacturer's instruction. Two-step RT-qPCR was performed to assess p53R2 mRNA level as described previously.³⁴ GAPDH was used as the internal control and all qRT-PCR reactions were done in triplicate. The primer sequences used as follows: p53R2, 5′-AGGAGGTGCAGGTTCCAGAG-3′ (forward) and 5′- CTGCTATCCATCGCAAGGC-3′ (reverse); GAPDH, 5′-TGCACCACCAACTGCTTAGC-3′ (forward) and 5′-GGCATGGACTGTGGTCATGAG-3′ (reverse). Relative mRNA expression level was presented by the 2^−ΔΔCT method.

Transfection of p53R2 siRNA

siRNA transfection was performed as described previously.³⁴ Briefly, cervical cancer cells were grown to 40–60% confluence, and then were incubated with siRNAs at a final concentration of 0.1μM using LipofectamineTM 3000 (L3000015, Invitrogen) according to the manufacturer's protocol. Silencer select siRNAs against p53R2 were purchased from Gene Pharma. The siRNA sequences are as follows: siRNA-1, 5′-GUACCCUGAUAUUUGGAAATT-3′; siRNA-2, 5′-GAGAUGUACAGUUUGCUGATT-3′; The negative control siRNA, 5′-UUCUCCGAACGUGUCACGUTT-3′.

Proliferation assay

Cell proliferation was assessed by the MTS method using the CellTiter 96® AQueous One Solution Cell Proliferation Assay kit (G3580, Promega) according to the manufacturer's instructions. Briefly, cells were plated in 96-well plates at a concentration of 2000 cells per well in 200μL of culture medium. Each plate was taken every 24 h for the MTS assay. The absorbance was measured at 490 nm using a multifunctional microplate reader.

Colony formation assay

Cells transfected with p53R2 siRNAs or a control siRNA were plated in a fresh 6-well plate (1000 cells/well) and maintained in culture medium containing 10% FBS for 2 wk until visible clones were formed. Colonies were fixed with 4% paraformaldehyde for 10 min and stained with 0.1% crystal violet solution for 30 min. After washing with PBS, the number of colonies (> 50 cells/colony) was counted with a microscope.

Cell migration and invasion assays

Transwell migration and invasion assays were performed in Boyden chambers without (for migration) or with (for invasion) Matrigel according to the manufacturer's protocol (Invitrogen). Briefly, 5 × 10⁴ cells in 250μL serum-free medium were seeded in the upper chamber. 750μL medium with 10% FBS was added to the lower chamber in a 24 well plate. The plates were then incubated at 37°C in a 5% CO₂ incubator for 24h. The migrated or invaded cells through the pores were fixed and stained using 0.1% crystal violet, and counted in 5 random microscopic fields. Each experiment was done in triplicate.

Protein extracts and western blotting

Tumor cells or tissues were lysed using RIPA buffer. Protein concentrations were determined using the BCA Protein Assay Kit (KGP902, KeyGEN BioTECH) according to the manufacturer's instructions. Equal amount of protein lysates were mixed with the loading buffer, resolved on SDS-PAGE gels and transferred to PVDF membranes (Millipore). After incubation with a specific primary antibody at 4°C overnight and then a secondary antibody, the signals were detected using an ECL Kit (35055, Thermo Scientific). The Western blot signal intensity was quantified using image analysis software (Image-J, ver. 1.44; NIH) and normalized by the GAPDH intensity.

Statistical analysis

All statistical analysis was performed using the SPSS statistical software program (Version 17.0 SPSS Inc.) and GraphPad Prism 6.0 Software. Paired Student's t test was used to analyze the difference in mRNA and protein expression between cervical cancer and paired adjacent noncancerous cervical tissues. The relationship between p53R2 expression and clinicopathological features was assessed using Chi-square test and Fisher's exact tests. Survival curves were calculated using the Kaplan-Meier method, and differences were assessed using a log-rank test. Univariable and multivariable survival analysis was performed using the Cox regression analysis to analyze the prognostic factors. The criterion for statistical significance was p < 0.05.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Funding

This work was supported by the National Institutes of Health R01 grants (No: CA123391), the National Natural Science Foundation of China (No: 81572440 and 81372600), the Recruitment Program of Global Experts, the Leading Talent of Guangdong Province, the Natural Science Foundation of Guangdong Province for Distinguished Young Scholar (No: 2015A030306047) and the Research Fund of State Key Laboratory of Oncology in South China.