The expression and significance of AKR1B10 in laryngeal squamous cell carcinoma

Aldosterone reductase family 1 member B10 (AKR1B10) is a nicotinamide adenine dinucleotide phosphate (reduced coenzyme II)-dependent oxidoreductase, and its biological functions include carbonyl detoxification, hormone metabolism, osmotic adjustment, and lipid synthesis. Studies suggested that AKR1B10 is a new biomarker for cancer based on its overexpression in epithelial tumors, such as breast cancer, cervical cancer, and lung cancer. At present, studies on the expression of AKR1B10 in laryngeal cancer have not been reported. However, we found that AKR1B10 is upregulated in laryngeal carcinoma, and its expression was negatively correlated with the degree of differentiation. In addition, AKR1B10 expression was positively correlated with tumor size; lymph node metastasis; alcohol use; and Ki-67, mutant p53, and matrix metalloproteinase 2 expression. AKR1B10 was overexpressed in Hep-2 laryngeal carcinoma cells. Oleanolic acid inhibited AKR1B10 activity and expression in Hep-2 cells and suppressed Hep-2 cell proliferation, migration, and invasion. Therefore, AKR1B10 may be related to the development of laryngeal carcinoma, suggesting its use as a prognostic indicator for laryngeal cancer.


Results
AKR1B10 is upregulated in laryngeal carcinoma and correlated with tumor size, lymph node metastasis, alcohol use, and differentiation. AKR1B10 expression in laryngeal carcinoma cells was mainly localized in the cytoplasm, and the individual nuclei also displayed positive staining (Fig. 1A). Positive staining in the adjacent squamous epithelial tissues was mainly localized in the nuclei of cells near the basal layer (Fig. 1B). The expression pattern differed from that of laryngeal carcinoma. AKR1B10 expression was higher in LSCC tissues than in adjacent tissues (P < 0.001, Table 1). In addition, AKR1B10 is diffusely expressed in laryngeal carcinoma cells (Fig. 1A). However, Fig. 1B illustrates that AKR1B10 was only expressed in the area near the basal layer of the advanced squamous epithelium. AKR1B10 protein expression was higher in laryngeal carcinoma than in the adjacent squamous epithelium.
The average age of the population was 55 years. AKR1B10 expression in laryngeal carcinoma was correlated with tumor size, lymph node metastasis, tumor differentiation, and alcohol use (all P < 0.05) but not with gender, age, T stage, and smoking status ( Table 2).

Effects of oleanolic acid (OA) on AKR1B10 enzyme activity in Hep-2 cells.
Compared with the findings in the control group, AKR1B10 activity in Hep-2 cells was significantly decreased after 48 h of treatment with 30 μM OA (P < 0.05, Fig. 3A). Conversely, the decrease was not significant after exposure to 10 or 60 μM.

Effects of OA on AKR1B10 mRNA expression in Hep-2 cells.
Reverse-transcription polymerase chain reaction (RT-PCR) was used to detect the effects of 48 h of OA exposure on AKR1B10 mRNA expression in Hep-2 cells. Compared with the control group, the 30 and 60 μM OA groups significantly inhibited the mRNA expression of AKR1B10 (both P < 0.01, Fig. 3B,C).

Effects of OA on AKR1B10 protein expression in Hep-2 cells.
Compared with the control group, the 30 and 60 μM OA groups significantly inhibited the protein expression of AKR1B10 (P < 0.01, Fig. 4A,B).

Effects of OA on the proliferation of Hep-2 cells.
The results of the CCK8 assay demonstrated that the 30 μM OA group significantly decreased Hep-2 cell growth starting on day 4 as compared to the control group (P < 0.05 on day 4, P < 0.01 on day 5, P < 0.001 on day 6, and P < 0.01 on day 7; Fig. 5A).

Effects of OA on the invasiveness of Hep-2 cells.
According to the Transwell assay, 30 μM OA significantly suppressed the invasiveness of Hep-2 cells compared with the control findings (P < 0.01, Fig. 6A,B).

Discussion
Globally, LSCC accounts for almost 2% of all malignancies, and it is one of the most common tumors of the head and neck 16 . AKR1B10 was initially isolated from liver cancer lesions 2 . Matkowskyj et al. found that AKR1B10 is upregulated in hepatocarcinoma, and silencing of the AKR1B10 gene in liver cancer cells increased apoptosis and reduced the formation and size of tumor cell colonies. This study provided the first evidence that AKR1B10 participates in hepatocellular carcinogenesis by regulating cell proliferation and apoptosis, and it can be used to identify hepatocellular carcinoma and benign liver lesions. In addition, the enzyme is an independent biomarker 17 . Fukumoto et al. 10 found that AKR1B10 is overexpressed in non-small cell lung cancer and is closely related to smoking. Ma et al. 13 reported that AKR1B10 overexpression was associated with tumor size, lymph node metastasis, and patient survival in breast cancer. Chung et al. 8 found that AKR1B10 is overexpressed   18 . However, studies on AKR1B10 expression in laryngeal carcinoma have not been reported. This experiment was the first to detect the expression of AKR1B10 in laryngeal and precancerous tissues via immunohistochemistry, and the relationships of AKR1B10 with differentiation, tumor size, lymph node metastasis, and Ki-67, MTp53, and MMP2 expression were analyzed. In addition, the effects of AKR1B10 on the biological behavior of Hep-2 cells were detected via in vitro experiments, and the significance of its expression in laryngeal carcinoma was further explored. The findings of the study were consistent with a study by Yoshitake et al. 19 , who found that AKR1B10 is highly expressed in the cytoplasm or nuclei of cervical cancer cells and occasionally expressed in normal squamous epithelial tissues, wherein it exhibits nucleus-specific expression. We hypothesized that AKR1B10 overexpression may be involved in the development and progression of laryngeal cancer. Kang et al. 20 reported that AKR1B10 is upregulated in lung squamous cell carcinoma, and AKR1B10 mRNA expression in highly differentiated tumors was significantly lower than that in moderately poorly differentiated tumors, in line with our results. However, it has also been reported that AKR1B10 is overexpressed in nasopharyngeal carcinoma, and the expression in moderately well differentiated tumors is higher than that in poorly differentiated tumors 21 . This may be attributable to differences in the role of AKR1B10 in different tumors. The finding that AKR1B10 expression was correlated with tumor size was consistent with the previously reported expression of AKR1B10 in oral squamous cell carcinoma and breast cancer 11,13 . In addition, Ma et al. 13 found that AKR1B10 is overexpressed in breast cancer and is positively associated with lymph node metastasis, as demonstrated in the present analysis. However, studies revealed that AKR1B10 expression in oral squamous cell carcinoma and cervical cancer is not associated with lymph node metastasis 11,22 .
Although AKR1B10 is upregulated in oral squamous cell carcinoma, breast cancer, cervical cancer, and lung cancer, the relationships of AKR1B10 expression with tumor size, lymph node metastasis, and squamous cell carcinoma differentiation are different 11,13,22,23 . In addition, studies revealed that AKR1B10 is overexpressed in breast cancer, oral squamous cell carcinoma, cervical cancer, and other tumors and is associated with prognosis 11,13,22 . To further analyze whether AKR1B10 expression in laryngeal carcinoma also has prognostic significance, we analyzed the relationships of AKR1B10 with the prognostic indicators Ki-67, MTp53, and MMP2.  www.nature.com/scientificreports/ The nuclear protein Ki-67 is closely related to tumor cell proliferation, and its overexpression is positively correlated with tumor malignancy. Ki-67 is a prognostic marker for breast cancer, lung cancer, cervical cancer, and other tumors 22 . Gioacchini et al. 23 demonstrated that Ki-67 is overexpressed in laryngeal carcinoma and is associated with poor tumor cell invasion and prognosis. It has been reported that the expression of Ki-67 in highly differentiated laryngeal carcinoma is significantly lower than that in poorly differentiated laryngeal carcinoma 24 . The expression rate of Ki-67 in laryngeal carcinoma with lymph node metastasis is significantly higher than that in laryngeal carcinoma without lymph node metastasis 25 . Previously, Wei et al. 26 silenced AKR1B10 in MHCC97H liver cancer cells, finding that Ki-67 and the oncogenes c-myc, c-fos, and N-ras were downregulated and the pro-apoptotic protein genes caspase-3 and Bax were upregulated, indicating that AKR1B10 may promote cell proliferation, inhibit cell apoptosis, and induce hepatocyte deterioration by regulating the expression of tumor-associated genes. However, Schmitz et al. 27 reported that the proliferative capacity of AKR1B10-positive hepatocellular carcinoma and its Ki-67 expression rate were significantly lower than those of AKR1B10-negative hepatocellular carcinoma. However, our study only found a positive correlation between AKR1B10 and Ki-67 expression. The specific links among these factors remain to be elucidated.
The tumor suppressor gene p53 protein is lost or mutated in approximately half of human cancers. It is well known that the loss of p53 function affects cell proliferation, migration, and invasion 28 . Wang et al. found that the 5-year survival rate of patients with LSCC and high MTp53 expression was significantly lower than that of patients with LSCC and low MTp53 expression, suggesting that MTp53 is associated with prognosis 29 . Our findings regarding MTp53 were consistent with the results of Ji et al., who reported that MTp53 expression in oral squamous cell carcinoma was associated with lymph node metastasis and tumor differentiation, but not age, gender, or tumor location 30 . Therefore, we speculate that AKR1B10 could be an indicator of poor prognosis in laryngeal cancer.
MMP2 is a gelatinase that degrades basement membrane collagen, which is related to the metastasis of various human tumors 31 . Liu et al. found that MMP2 is highly expressed in laryngeal carcinoma, and its expression in highly differentiated tumors is significantly lower than that in moderately poorly differentiated tumors. Additionally, MMP2 expression in lymph node metastasis is significantly higher than that in non-metastatic tissues. In vitro experiments were performed to confirm the aforementioned findings using the highly selective AKR1B10 inhibitor OA 34 and Hep-2 cells. Our results were consistent with the findings of Zhou et al., who demonstrated that silencing of the AKR1B10 gene inhibited lung cancer cell proliferation and migration 35 . Our findings regarding the inhibitory effects of OA on AKR1B10 expression and activity were consistent with the aforementioned immunohistochemistry results. Therefore, we speculate that the overexpression of AKR1B10 in laryngeal carcinoma may be related to the occurrence and poor prognosis of cancer.
Certain limitations of our work must be addressed. We did not perform a quantitative analysis of AKR1B10 expression at the tissue level. Additional experiments are needed to further clarify the mechanism by which AKR1B10 affects the phenotype of LSCC. We plan to perform in vivo experiments to further study the role of AKR1B10 in laryngeal cancer in the future.
This study was the first to detect the expression of AKR1B10 in laryngeal carcinoma and explore its relationship with tumor differentiation, tumor size, lymph node metastasis and prognosis. Based on the findings, AKR1B10 may represent a new target and theoretical molecular basis for the treatment and prognosis of laryngeal cancer.

Methods
Patient samples and cell culture. In total, 87 cases of LSCC detected between 2015 and 2017 in the second department of Jilin University First Hospital were examined in the study. The female to male ratio was 3.14:1. Sixty-one cases involved lymph node metastasis (Table 4), and 77 cases involved invasion of adjacent tissues. This study was approved by the First Hospital of Jilin University's Institutional Review Board. All methods were performed in accordance with relevant guidelines and regulations. Informed consent was obtained from all patients.
Hep-2 and A549 cells 36 were provided by the Key Laboratory of Pathology and Biology of the Ministry of Education of Jilin University. The cells were cultured in H-DMEM containing 10% fetal bovine serum in an incubator containing 5% CO 2 at 37 °C. Result interpretation. AKR1B10 and MMP2 staining intensity was scored as the percentage of positive cells as follows: < 5%, 0 points; 5-10%, 1 point; 11-50%, 2 points; and > 50%, 3 points Meanwhile, the staining color was scored as follows: no color, 0 points; light yellow, 1 point; dark yellow, 2 points; and yellowish brown, 3 points. The staining score was calculated by summing the intensity and color scores and categorized as follows: < 2, negative; 2-3, weakly positive; 4-5, moderately positive; and ≥ 6, strongly positive 17 . For Ki-67 scoring, the percentage of positive cells among 1000 positive cells was categorized as follows: ≤ 25%, low expression; and > 25%, high expression 37 .
MTp53 expression was scored as the percentage of positive cells among 1000 cells as follows: ≤ 10%, negative; and > 10%, positive 38 .

Reverse-transcription polymerase chain reaction (RT-PCR).
Ribonucleic acid (RNA) was extracted from Hep-2 and A549 cells. RNA was reverse-transcribed into cDNA according to the instructions of the kit. After the PCR reaction system was thoroughly mixed, 30 cycles were performed according to the reaction conditions.
PCR reaction system. For electrophoresis, 10 μL of the PCR product were added to each well and electrophoresed at 100 V for 1 h, and the results were observed using an imager.
Western blotting. To extract protein, Hep-2 and A549 cells were washed with cold phosphate-buffered saline (PBS), after which the culture dishes were placed on ice. Cells were lysed with 500 μL of lysate per dish for 30 min, and cells were scraped into Eppendorf tubes and centrifuged at 15 × 10 rpm for 15 min. The supernatant was retained and used to measure the protein concentrations in the samples.
For electrophoresis, 30 μg of the protein sample were added to each well, electrophoresed at 80 V for 30 min, and then electrophoresed at 120 V for 90 min. Proteins were then transferred to a membrane and incubated sequentially with primary and secondary antibodies. Due to the need to add different antibodies during the experiment, the PDVF membrane was cut.
Immunofluorescence staining. Hep-2 and A549 cells were cultured in 24-well plates. After permitting adherent growth, cells were fixed with paraformaldehyde for 20 min, washed three times with PBS, incubated with 0.3% Triton for 5 min, and washed three times with PBS. Cells were then exposed to 0.5% BSA for 30 min. After aspirating the liquid, cells were incubated with 0.1% BSA-diluted AKR1B10 antibody (1:50) overnight at 4 °C and washed three times with PBS. Cells were incubated with the fluorescent secondary antibody (1:400) at room temperature for 1 h, washed three times with PBS, stained with DAPI for 10 min, and washed with PBS, and the stained cells were observed under a fluorescence microscope.

Effect of OA on the activity and expression of AKR1B10 in Hep-2 cells.
To conduct ELISA, 6 × 10 5 Hep-2 cells were added to each well of a six-well plate. After permitting adherent growth, cells were cultured with 0, 10, 30, or 60 μM OA for 48 h and lysed (approximately 1 million cells/ml). Cells were then centrifuged (3 × 1000 rpm, 10 min). According to the AKR1B10 enzyme activity assay kit instructions, a standard curve was drawn to calculate enzyme activity. For RT-PCR and Western blotting, cells were treated with OA, after which CCK8 experiment. Hep-2 cells were cultured in seven 96-well plates at 2000 cells/100 μL per well. Three replicates each were created for the control and experimental groups. The control group was treated with H-DMEM containing 10% serum. The experimental group was treated with H-DMEM containing 30 μM OA. Cell growth was measured continuously for 7 days by adding 10 μL of CCK8 solution at 37 °C for 1-4 h, measuring the absorbance at 450 nm, and plotting the proliferation curve.
Scratch test. A straight line was drawn on the back of a six-well plate. Hep-2 cells (6.5 × 10 5 ) were added to each well and cultured overnight. Then, the cell monolayer was scratched in a direction perpendicular to the line. The control group was cultured in serum-free H-DMEM. The experimental group was cultured in serum-free H-DMEM containing 30 μM OA. Cell migration was observed under a phase contrast microscope at 0 and 48 h. Cell migration was calculated using the following formula: where A is the area of the scratch at 0 h and B is the area of the scratch at 48 h. Statistical analysis. AKR1B10 expression in laryngeal cancer and paracancerous tissues was compared using t-test. Paracancerous tissue refers to tissue located within 2 cm of cancer tissue. The relationships of clinical indicators, excluding age, with AKR1B10 expression in laryngeal carcinoma was analyzed using Fisher's exact probability test, whereas the correlation between age and AKR1B10 expression was analyzed using the R × C χ 2 test. Nonsmokers comprised never-smokers and people with a past history of light smoking (< 10-pack-year history with at least 20 years of cessation). Alcohol use was defined as the consumption of 10-20 alcohol drinks weekly. Spearman's correlation analysis was used to analyze the correlations of AKR1B10, Ki-67, MTp53, and MMP2 expression and clinical indicators.