Loss of AKR1C1 is a good prognostic factor in advanced NPC cases and increases chemosensitivity to cisplatin in NPC cells

Abstract Cisplatin resistance is one of the main obstacles in the treatment of advanced nasopharyngeal carcinoma (NPC). AKR1C1 is a member of the Aldo‐keto reductase superfamily (AKRs), which converts aldehydes and ketones to their corresponding alcohols and has been reported to be involved in chemotherapeutic resistance of multiple drugs. The expression and function of AKR1C1 in NPC have not been reported until now. The aim of this research was to investigate the expression of AKR1C1 and it is role in cisplatin resistance in NPC. AKR1C1 protein expression was detected by immunohistochemistry in human NPC tissues and by Western blot assays in NPC and immortalized nasopharyngeal epithelial cells. The effects of AKR1C1 knock‐down by siRNA on proliferation, migration and invasion in NPC cells were evaluated by CCK8, wound healing and transwell assays. To evaluate the effects of AKR1C1 silencing on cisplatin sensitivity in NPC cells, CCK8 assays were used to detect cell proliferation, flow cytometry was used to detect cell cycle distribution, and flow cytometry and DAPI staining were used to detect cell apoptosis. AKR1C1 down‐regulation was associated with advanced clinicopathological characters such as larger tumor size, more lymphatic nodes involvement, with metastasis and later clinical stages, while AKR1C1 down‐regulation was a good prognostic factor for overall survival (OS) in NPC patients. In vitro study showed that AKR1C1 was not directly involved in the malignant biological behaviours such as proliferation, cell cycle progression and migration of NPC cells, whereas AKR1C1 knock‐down could enhance cisplatin sensitivity of NPC cells. These results suggest that AKR1C1 is a potential marker for predicting cisplatin response and could serve as a molecular target to increase cisplatin sensitivity in NPC.


| INTRODUC TI ON
Nasopharyngeal carcinoma (NPC) is a endemic malignant tumour which has a high incidence in Southern China. 1 In recent years, improved radiotherapy techniques have given satisfactory outcome in early stage of NPC. 2 But most of patients were diagnosed as locally advanced NPC at the first diagnosis and their 5-year survival rate was still only about 30%-80%. [3][4][5] As indicated by the National Comprehensive Cancer Network (NCCN) Guidelines, locoregionally advanced disease requires cisplatin-based concurrent chemoradiotherapy. 1 Cisplatin-based regimen has been proposed as the optimal protocol by a meta-analysis of eight randomized trials including 1753 patients. 6 Unfortunately, cisplatin resistance occurs in some NPC cases and becomes a major obstacle of chemotherapy success for NPC. 7 Thus, understanding the mechanism of cisplatin resistance in NPC may enable the development of new strategies to overcome chemoresistance and improve clinical outcome in locally advanced NPC cases.
Human 20-keto reductase family 1 member C1 (AKR1C1) is a member of the aldehyde ketone reductase superfamily (AKRs). 8 AKR family members catalyse the conversion of aldehydes and ketones to their corresponding alcohols. 9 Their substrates include endogenous and xenobiotic non-steroidal carbonyl compounds. 10 Moreover, chemotherapeutic drugs containing carbonyl can be converted to inactivated reductive metabolite, leading to the chemotherapy resistance. 10 Actually, cumulative data indicated that AKR1C1 plays an important role in chemotherapy resistance in several cancers. [11][12][13][14] Thus, targeting AKR family members provide a novel therapeutic strategy for overcoming chemoresistance in malignant tumours.
Up-regulation of the AKR1C1 gene in multiple cancer cells was reported to be associated with resistance against several anticancer agents including cisplatin. [11][12][13][14] However, the expression and role of AKR1C1 in nasopharyngeal carcinoma has not been reported so far. In the present study, we found that AKR1C1 down-regulated in advanced NPC tissues, but down-regulated AKR1C1 was a good prognostic factor for overall survival (OS) in NPC patients. In vitro study indicated that AKR1C1 did not directly contribute to the malignant biological behaviours and knock-down of AKR1C1 by siRNA increased the cisplatin sensitivity in NPC cells. Hence, AKR1C1targeted strategy may be a novel therapeutic candidate for overcome cisplatin resistance in NPC patients.

| Patients and samples
Patients and samples in this study were shared with one of our previous studies, and their information was previously fully described. 15

| Immunohistochemical staining and microscopic analysis
Immunohistochemical assays were previously fully described. 15 The dilution of anti-AKR1C1 antibody (sc-166297, SANTA CRUZ BIOTECHNOLOGY, USA) was 1:100. The secondary antibody (goat anti-mouse) (Cat. No. KIT-5001; MXB Biotechnologies) was ready to use. IHC scores were evaluated by two pathologists in our department. Staining intensity was classified as 0 (negative); 1 (weak); 2 (moderate); and 3 (strong). The scores 0 and 1 were defined as low, and the scores 2 and 3 were defined as high. 15

| AKR1C1-siRNA transfection
The CNE1 and CNE2 cells were seeded in six-well culture plates and cultured until the NPC cells reached 60%-70% confluency.
Si-AKR1C1 and si-ctrl (RiboBio) were transfected into the CNE1  Finally, blots were detected by ECL substrate (Beyotime Inc.), and Image J software was used to analysis the expression of protein.

| Cisplatin treatment
To obtain an appropriate concentration of cisplatin for subsequent chemotherapeutic sensitivity assays, different concentrations (0.2, 0.4, 0.8, 1.6, 4.0 μg/mL) of cisplatin (QiLu Pharmaceutical Co) were used to determine the IC50 in NPC cells. In brief, CNE1 and CNE2 (4000 cells/well) were plated on 96-well plates (Costar) with 5 wells/ group and were cultured overnight. Then, the cells were treated with cisplatin for 48 hours and CCK8 assays were employed to detect cell growth. According to the IC50 ( Figure S3), 0.5 μg/mL of cisplatin was selected for subsequent chemotoxicity assays. To detect the sensitized effects of AKR1C1 knock-down on cisplatin (DDP) treatment in NPC cells, CNE1 and CNE2 were all divided into four groups, which including si-ctrl, si-AKR1C1, si-ctrl + DDP and si-AKR1C1 + DDP.
After transfected with siRNA overnight, CNE1 and CNE2 cells were treated with DDP for 48 hours and then submitted for CCK8 assays.

| Wound healing assay
After the cells were transfected with siRNA, cell migratory ability was evaluated by wound healing assays. Wound healing assays were previously fully described. 15 The wound healing distance was documented at 48 and 72 hours.

| Transwell migration and invasion assays
Transwell assays were previously fully described. 15 In this study, 2 × 10 4 cells/well (migration assay) or 4 × 10 4 cells/well (invasion assay) in 200 μl serum-free RPMI 1640 were seeded on top of the filter. After 24 hours (migration assay) or 48 hours (invasion assay), the cells were fixed, stained and then calculated under a microscope.

| Flow cytometry-based cell cycle
After treated with cisplatin for 48 hours, each group of cells were washed with PBS for 3 times and fixed with 70% cold ethanol at 4°C overnight. Then, the cells were washed, centrifuged, resuspended and then stained with 200 μL PI/RNase Staining Buffer (BD, Biosciences).
The cell cycle was measured using a FACScan flow cytometer (BD Biosciences) and analysed by ModFit software (BD Biosciences).

| Flow cytometry-based apoptosis
After treated with cisplatin for 48 hours, each group of cells was

| DAPI staining
After treated with cisplatin for 48 hours, the cells were fixed with 4% paraformaldehyde for 10 minutes, permeabilized with 0.5% Triton X-100 (ZSGB-BIO) for 15 minutes and then counterstained with DAPI (Solarbio) 5 minutes. Finally, the cell slides were gently removed and sealed, and the images were observed under a fluorescence microscope.

| AKR1C1 expression decreased in NPC tissues and frequently lost in NPC cells
To determine AKR1C1 expression levels in NPC tissues, 177 NPC tissues and 61 non-cancerous epithelial tissues were detected by IHC.
The IHC staining results revealed that AKR1C1 protein significantly down-regulated in NPC specimens ( Figure 1A and B and Table 1). When AKR1C1 expression evaluated by Western blotting in NPC cell lines (namely CNE1, HK1-EBV, CNE2, SUNE1, HONE1, 5-8F and S18) and immortalized nasopharyngeal epithelial cells (NP69, SWSX-1489 and HNEpC), only three NPC cells (CNE1, CNE2 and S18) expressed AKR1C1 ( Figure 1C). AKR1C1 strongly stained in normal ciliated columnar epithelia of nasopharyngeal tissue while lost in immortalized nasopharyngeal epithelial cells, the underlined mechanism needs to be elucidated. We found that AKR1C1 significantly down-regulated in squamous metaplastic epithelia of nasopharyngeal by IHC ( Figure S1 and Table S1). So the phenotype changes of immortalized nasopharyngeal epithelial cells in vitro still need to be identified. Anyway, the data above showed that AKR1C1 expression decreased in NPC tissues and frequently lost in NPC cells, which implied that loss of AKR1C1 was a key molecular event in the development of NPC.

| Loss of AKR1C1 was associated with advanced TNM stage while was a good prognostic factor in NPC patients
The relationships between AKR1C1 expression and the clinicopathologic characteristics of NPC patients were summarized in Table 2. No significant association was identified between AKR1C1 expression and the age (P = 1.000) or sex (P = .755) of 177 NPC patients (Table 2) Table 2). Although AKR1C1 loss was associated with advanced clinicopathological characters, high AKR1C1 expression was associated with a poor prognosis in NPC patients ( Figure 2C). These data implied that AKR1C1 loss may be an accompanied molecular event, while not a drive event in the progression of NPC.

| Knock-down of AKR1C1 by siRNA showed no impact on cell proliferation, migration and invasion in NPC cells
To investigate the function of AKR1C1 in NPC cells, CNE1 and CNE2 cells were transiently transfected with siRNA. According to Western blot assay results, AKR1C1 was successfully knocked down by siRNA in NPC cells ( Figure 3A). The CCK8 assay showed that AKR1C1 silencing did not promote proliferation of NPC cells ( Figure 3B and   3C). In addition, wound healing assays ( Figure S2) and transwell assays ( Figure 3D and 3E) showed that depleting endogenous AKR1C1 by siRNA did not increase the invasion and migration of NPC cells.
These data indicated that AKR1C1 was not directly involved in the malignant biological behaviours of NPC cells.

| Knock-down of AKR1C1 by siRNA enhanced cisplatin-induced inhibition of cell proliferation and cell cycle arrest
To determine whether knock-down of AKR1C1 by siRNA may increase cisplatin sensitivity or not, CNE1 and CNE2 cells were treated with cisplatin for 48 hours after transfected with siRNA. Figure 4A

| Knock-down of AKR1C1 by siRNA increased cisplatin-induced apoptosis
To investigate the effect of AKR1C1 knock-down on cisplatin-induced cell apoptosis, si-ctrl and si-AKR1C1 transfected NPC cells (CNE1 & CNE2) were treated with cisplatin for 48 hours and then assayed by flow cytometry for calculating apoptotic cells ( Figure 5A and 5B). Meanwhile, morphology of apoptotic cells' nuclei was assayed by DAPI staining ( Figure S4). Both flow cytometry and nuclear staining showed that AKR1C1 knock-down increased cisplatin-induced cell apoptosis. These results suggested that AKR1C1 attenuated cisplatin-induced apoptosis in NPC cells. Most NPC cases are advanced cancer at first diagnosis, and chemotherapy combined with radiotherapy is the standard regime for these patients. [3][4][5][6] Cisplatin-based chemotherapy is the first-line drug for chemoradiotherapy in NPC. 6 Unfortunately, some NPC patients are cisplatin resistance, which leads to therapeutic failure and increases the risk of tumour relapse or distant metastasis. [18][19][20] The major factors involved in the mechanism of cisplatin resistance include disrupted a new player in chemotherapy resistance recently. 16,22 The mechanism of chemotherapy resistance of AKR1C1 included reducing the production of reactive oxygen species, eliminating free radicals and inactivating anticancer drugs, thereby reducing DNA damage and inhibiting apoptosis. [11][12][13][14]23 Accumulating data showed that AKR1C1 contributed to cisplatin resistance in multiple tumours. Overexpression of AKR1C1 in bladder, 13 gastric, 17 ovarian, 24-26 cervical and lung cancer 26 reduced the sensitivity to cisplatin. AKR1C1 is a poor prognostic factor for recurrence and death of HNSCC patients. 16 Our data showed that AKR1C1 protein down-regulated in cancerous tissue when compared with non-cancerous nasopharyngeal columnar epithelia ( Figure 1A and 1B Figure S4). These data implied that AKR1C1

| D ISCUSS I ON
was not a driver factor in the process of carcinogenesis and progression in NPC, AKR1C1 loss maybe an accompanying molecular event which contributed only to the chemotherapeutic sensitivity to cisplatin while not the malignant biological behaviours of NPC.
Although AKR1C1 is reported to be involved in tumour metastasis through interaction with STAT3 in NSCLC 27 and to promote Actually, although therapeutic methods improved greatly in recent years, histological subtype remains a prognostic factor determining the long-term survival outcome of patients with NPC. [28][29][30] Moreover, the non-keratinized carcinoma subtype has the best prognosis, while the keratinized squamous cell carcinoma subtype has the worst prognosis. 30 However, there is not a consensual therapeutic protocol for the different specific histological subtypes of NPC so far. 31 Our data showed that AKR1C1 more frequently lost in undif- In summary, our study demonstrates that loss of AKR1C1 could enhance cisplatin sensitivity in NPC cells. Increased chemotherapeutic sensitivity corroborated AKR1C1 loss as a good prognostic factor in NPC patients.

| CON CLUS IONS
AKR1C1 loss is identified as a good prognostic factor in the advanced staged cases of NPC in this study. Cisplatin sensitivity increased in AKR1C1-silenced NPC cells warrants further validating AKR1C1 as a potential marker for predicting cisplatin response in NPC patients.
New therapeutic strategies must take into consideration when managing different histological subtype and molecular subtype of NPC with differentially expressed chemotherapy-associated genes such as AKR1C1. 2015GXNSFAA139131 to SJXiao).

CO N FLI C T O F I NTE R E S T
The authors report no conflicts of interest in this work.

DATA AVA I L A B I L I T Y S TAT E M E N T
All data that support the findings of this study are available from the corresponding author upon reasonable request.