Introduction

Endometrial cancer (EC) is the most common gynecologic malignancy of the female reproductive tract in developed countries1. More than 47,000 women are diagnosed with endometrial carcinoma in the United States each year2,3, Numerous epidemiological studies have confirmed that metabolic factors are closely associated with various cancers, particularly EC4,5,6. Hyperinsulinemia has been considered as an EC risk factor independent of estradiol7, and metformin, an insulin-sensitizer, is found to diminish EC proliferation and has positive effects on cancer clinical evolution8,9. The majority of EC cases diagnosed in the early stage have a favorable prognosis. However, women diagnosed with advanced stage, poor differentiation, and progesterone receptor negative EC have a higher risk of metastasis and a poor prognosis3. Hormonal therapy, such as medroxyprogesterone acetate (MPA), has been applied in the conservative treatment of young patients who wish to preserve their fertility, as well as in the palliative treatment of advanced-state patients10. However, the administration of progesterone to EC patients does not work effectively because of de novo or acquired progestin resistance. To date, more than 30% of EC patients with progestin treatment have presented progestin resistance11. Over the past decades, little improvement has been demonstrated in this area. Thus, it is imperative to identify novel, effective anticancer targets with prognostic value to accurately predict the survival of EC patients and to cure this disease efficiently.

3β-Hydroxysteroid-Δ24 reductase (DHCR24), the final enzyme of the cholesterol biosynthetic pathway, catalyzes the reduction of the Δ24 double bond in desmosterol to produce cholesterol12. DHCR24 is involved in multiple cellular functions, such as oxidative stress, cell differentiation, anti-apoptotic function, and anti-inflammatory activity13. In addition, DHCR24 is dysregulated in various tumors including prostate cancer, ovarian cancer, and urothelial carcinoma14,15,16. The up-regulation of DHCR24 is associated with invasiveness and disease recurrence in urothelial cancer, which suggests a crucial role for this protein in tumor progression14. Although DHCR24 has been found to take part in tumor progression, the expression pattern of DHCR24 in EC is still not fully elucidated. Moreover, the regulatory mechanisms by which DHCR24 mediates its functions as well as its involvement in progestin resistance are unclear.

A new era of EC research has begun as clinically relevant genomic information has been gradually elucidated17,18. Herein, we analyzed the expression pattern of DHCR24 at both the mRNA and protein levels in a large set of clinical samples of EC combined with TCGA (the Cancer Genome Atlas, which generates comprehensive, multidimensional maps of the key genomic changes in 33 types of cancer) and GEO datasets (Gene Expression Omnibus, which is a public functional genomics data repository)19. We explored the relationship between DHCR24 expression and the corresponding clinical parameters. We demonstrated that DHCR24 overexpression exhibited characteristics of metastasis and progestin resistance. Furthermore, we found for the first time that insulin, the peptide hormone produced by beta cells of the pancreatic islets, could enhance the transcription of cholesterol synthetase DHCR24 via STAT3 in EC. These experimental data in support of the insulin/STAT3/DHCR24/PGR axis reveal characteristics of an important pathway involved in both metastasis and progesterone response in EC.

Results

DHCR24 is up-regulated in EC

We first analyzed DHCR24 expression in the GEO database of 91 endometrial cancer and 12 normal tissues19. We observed that DHCR24 was significantly over-expressed in EC tissues compared with their normal counterparts (Fig. 1A). Importantly, we found that DHCR24 expression increased in tissues from different grades of EC (Grade 1, Grade 2, and Grade 3) (Fig. 1B). Consistent with the data from the GEO database, DHCR24 expression was also up-regulated in the EC patients of the cohort from our hospital at the mRNA level in EC patients of the cohort from our hospital (Fig. 1C). Subsequently we detected the subcellular localization of DHCR24 by immunohistochemistry (IHC) in two tissue microarrays of 258 EC samples and 42 noncancerous samples. Expectedly, the protein levels of DHCR24 in the tissues of the EC patients were also significantly higher than those of the normal controls (Fig. 1D,E). Furthermore, the Kaplan Meier analysis of the patients’ follow-ups in the corpus endometrioid carcinomas Dataset (n = 548) from TCGA showed that the overall survival in EC patients with the DHCR24 gene alteration was significantly shorter than those without the DHCR24 alteration (p < 0.05) (Fig. 1F).

Figure 1: DHCR24 is up-regulated in endometrial cancer.
figure 1

(A) mRNA expression level of DHCR24 in 91 EC tissues and 12 non-tumor tissues in the GEO database. (B) Differential mRNA expression level of DHCR24 in the histological grading of the GEO database. n, number of analyzed samples. (C) mRNA expression level of DHC24 in freshly frozen endometrial cancer tissues. (D) Representative IHC staining of DHCR24 in the tissue microarray (TMA) of endometrial cancer from histological Grade 1 to Grade 3 of EC specimens. Scale bar, 50 μm. (E) Score of the IHC stained in the TMA of EC. Distribution of DHCR24 in the cytoplasm was quantified in EC. Scoring was conducted based upon the percentage of positively stained cells: 0–10% scored 0; 10–35% scored 1; 36–70% scored 2; and more than 70% scored 3. (F) Overall survival Kaplan-Meier estimate of DHCR24 alterations in TCGA (DHCR24 alteration group: Red line, No DHCR24 alteration group: Blue line).

Full size image

High DHCR24 expression predicts poor prognosis in EC

We next investigated the relationship between the protein levels of DHCR24 and patient clinical parameters in EC (Table 1). High expression of DHCR24 in the EC was associated with histological grade (p < 0.001), FIGO (International Federation of Gynecology and Obstetrics) stage (p < 0.001), vascular invasion (p = 0.025), and LN metastasis (p = 0.006). However, no significant correlation was observed between DHCR24 levels and age, pregnancy, family history, and pathological type (Table 1, p > 0.05).

Table 1 Correlation between DHCR24 expression and clinicopathological parameters in 258 patients with endometrial cancer.
Full size table

Based on its association with aggressive tumor behaviors, DHCR24 may serve as a novel potential prognostic marker. To determine the prognostic value of DHCR24 in EC, the correlation between the expression of DHCR24 and clinical follow-up information was assessed by a Kaplan-Meier analysis and a log-rank test. As shown in Fig. 2A, high DHCR24 expression was associated with poor overall survival (p < 0.001). We further analyzed the relationship between DHCR24 expression and the overall survival in EC patients with the advanced EC who classified as Grades 2 and 3 or in stages II, III, and IV. These results showed that the overall survival was much shorter in the advanced EC patients with high DHCR24 expression than those with low DHCR24 expression (Fig. 2B,C). Similar results were also obtained in EC patients with vascular invasion (Fig. 2D).

Figure 2: High DHCR24 expression predicts poor prognosis in the two tissue microarrays of EC.
figure 2

(A) Comparisons of the overall survival (OS) between DHCR24 low- and high-expression groups in EC. (B) Comparisons of the OS between DHCR24 low- and high-expression groups in the histological Grade 2 and Grade 3 cohort. (C) Comparisons of the OS between DHCR24 low- and high-expression groups in the clinical stage II-IV cohort. (D) Comparisons of the OS between DHCR24 the low- and high-expression groups in the vascular invasion cohort.

Full size image

DHCR24 could be induced by insulin stimulation via STAT3 in EC

DNA copy number gains play an important role in gene overexpression. In an attempt to identify whether gene amplification contributes to the higher expression of DHCR24 in EC, we analyzed the DHCR24 alterations in the genomic-scale sequencing of EC in TCGA. We found that DHCR24 gene amplification was identified in only 8 of 548 samples (1.5%) (Fig. 3A), insufficient evidence to explain the overexpression of DHCR24. It has been reported that tumor microenvironment factors, such as cytokines and growth factors, are involved in cancer cell metastasis and the response to progestin in EC20,21. EGF, insulin, IL6, TGF-β, and estradiol were applied to stimulate EC cells to investigate whether these factors contribute to the overexpression of DHCR24. Our results suggest that insulin can significantly induce the expression of DHCR24 (Fig. 3B). This coincides with the fact that chronic hyperinsulinemia from insulin resistance is involved in endometrial tumorigenesis. Therefore, the alteration of DHCR24 expression in EC could be caused by the tumor microenvironment.

Figure 3: DHCR24 could be induced by insulin stimulation via STAT3 in EC.
figure 3

(A) DHCR24 DNA copy number amplification from the genomic-scale sequencing of EC in TCGA. (B) mRNA expression of DHCR24 stimulated by EGF, insulin, IL6, TGF-β and estradiol evaluated by qPCR. (C) The mRNA expression level of DHCR24 and STAT3 under STAT3 siRNA interference. (D) The protein level of DHCR24 was determined in ECC-1 and HEC-1A cells treated with insulin, GSK1904529A and Stattic by immunoblotting (cropped images are shown here, the non-cropped images are shown in Supplementary data 6). (E) A ChIP assay was performed to confirm the potential STAT3 binding site in the DHCR24 promoter region. IgG and input fraction were used as controls. (F) A ChIP assay was performed in two endometrial cancer cell lines, HEC-1A and ECC-1. (G) In total, 10 pairs of primers were constructed according to the promoter of DHCR24. (H and I) A Luciferase reporter assay was performed using ECC-1 and HEC-1A cells after transfecting the wild type plasmids and mutated plasmids (mutation site: red). The data shown are the mean ± SD. (*p < 0.05; **p < 0.01; ***p < 0.001).

Full size image

To identify which transcription factor was responsible for the insulin induced overexpression of DHCR24, we searched for downstream transcription factors of insulin that are also known to be involved in the regulation of cell metastasis and the response to progestin. Previous studies have shown that STAT3 is found in hepatocytes and multipotent progenitors22,23,19. We subsequently performed analyses on the database from corpus endometrioid carcinomas in the Cancer Genome Atlas (TCGA, http://www.cbioportal.org/).

Cell culture

Human endometrial cancer cell lines KLE, HEC1-A, RL95–2, ISHIKAWA, HEC1-B, AN3CA, and ECC-1 were all preserved at Shanghai Cancer Institute. All these cells were cultured in specific medium (Invitrogen, Carlsbad, CA, USA) supplemented with 10% (v/v) fetal bovine serum (FBS, Invitrogen), 100 μg/ml streptomycin, and 100U/ml penicillin (Invitrogen) at 37 °C with 5% CO2. Cells were serum starved for 12 h prior to EGF (10 ng/ml), insulin (100 nmol/l), TGF-β (5 ng/ml), IL-6 (50 ng/ml), and estradiol (10 nM) treatment for 6 hours36,37,38,39. Culture medium was switched to serum-free and phenol red-free medium to avoid potential interference from serum hormones or phenol red. Human recombinant EGF was obtained from R&D (R&D Minneapolis, MN, USA), and insulin, TGF-β, IL-6, and estradiol were obtained from Sigma (Sigma-Aldrich, St. Louis, MO, USA).

Real-time PCR

The total RNA was extracted from frozen tissue samples or EC cells using TRIzol reagent (Invitrogen) according to the supplier’s protocols. The primers used in this study are shown in Supplementary data 1. Reverse transcription was performed using a Prime-Script RT-PCR Kit (Takara, Tokyo, Japan). The mRNA levels of the detected genes were quantified using an ABI Prism 7500 Sequence Detection System (Applied Biosystems, Inc. USA) with SYBR Green Master Mix (Takara). The 2−∆Ct method was used to quantify the relative expression levels of DHCR24. Experiments were repeated in triplicate independently.

Western blot

Protein samples were homogenized in a lysis buffer (Beyotime, Suzhou, China) containing proteinase inhibitors and phosphatase inhibitors (Selleck, TX, USA). According to the manufacturer’s protocol, the protein concentration was evaluated using a protein assay reagent kit (Beyotime). Equal amounts of proteins were separated on a sodium dodecyl sulfate polyacrylamide gel and transferred onto polyvinylidene fluoride membranes (Merck Millipore, MA, USA). The membranes were then incubated with the primary antibodies anti-DHCR24 (1:500) (Thermo Fisher Scientific, MA, USA), anti-STAT3 (1: 1,000, Proteintech, IL, USA), anti-β-actin (1: 3,000, Sigma). The proteins were detected using ECL Western Blotting Detection Reagents (Millipore) in accordance with the supplier’s protocol.

Immunohistochemistry

Tissue samples embedded in paraffin were deparaffinized and rehydrated using xylene and graded ethanol. The sections were blocked with 5% bovine serum albumin for 1 h after antigen retrieval and the neutralization of endogenous peroxidase. The slides were then incubated overnight at 4 °C with the primary antibodies (DHCR24, Thermo Fisher; PGR, Abcam, MA, USA). After washing in phosphate-buffered saline (PBS) for three times, the slide was labeled by a HRP secondary antibody (rabbit) for 1 h at room temperature and again washed three times with PBS. Visualization was performed by DAB and counterstaining with hematoxylin. Scoring was conducted by the percentage of positively stained cells: 0–10% scored 0; 10–35% scored 1; 36–70% scored 2; and more than 70% scored 3. We designated the final score as a high or low expression group as follows: a score of 0 or 1 was considered low expression, and a score of 2 or 3 indicated high expression. The scoring by two senior pathologists was performed in a blind manner.

Short Interfering RNA–mediated DHCR24 knockdown

Human endometrial carcinoma cell lines KLE and HEC-1-B were maintained in specific medium containing 10% (v/v) FBS and 1% penicillin/streptomycin at 37 °C in humidified atmosphere of 5% CO2. Cells were transiently transfected using the Lipofectamine RNAiMAX reagent (Invitrogen), Opti-MEM reduced-serum medium (Invitrogen), and siRNA oligonucleotides (Supplementary data 2). The siRNA (15 nM) treated with 5 μL of RNAiMAX reagent for 48 h was found to be the best concentration with a maximum transfection efficiency (95%) and a maximum silencing of the gene expression of DHCR24. The siRNA-treated cells were used in subsequent experiments after transfection for 48 hrs.

Cell viability assay

For the drug sensitivity assay, KLE and HEC-1B cells were transfected with DHCR24 siRNA and followed by treatment with MPA at a concentration of 10 μM for 48 hours. Cell viability was measured using a CCK-8 assay (Do**do, Kumamoto, Japan) according to the manufacturer’s instructions. Cell survival was also measured after treatment with MPA (10 μM), insulin (100 nmol/l), GSK1904529A (60 nM), and Stattic (10 μM). MPA, GSK1904529A and Stattic were obtained from Selleck.

Migration and invasion assay

Migration and invasion assays were performed as follows: a 200-μL volume of cells at a density of 2 × 105/mL was seeded into the upper chambers (Corning, NY, USA) with an 8-μm pore in 24-well plates. DMEM (Invitrogen) containing 15% FBS as a stimulatory factor was added to the lower chambers. After 12 hours in culture at 37 °C in a 5% CO2 atmosphere, the cells were fixed and stained with crystal violet. Then, the cell were counted using a microscope. Five random microscopic fields per well were counted with the double-blind method. For the cell invasion assay, BioCoat Matrigel (BD Biosciences, CA, USA) (300 μg/mL, 100 μL per chamber) was applied to the upper chambers before following the procedure for the migration assay described above.

ChIP Assay

A Pierce Agarose ChIP Kit was purchased from Thermo Fisher Scientific. Cells were cross-linked with 1% formaldehyde for 10 minutes at room temperature and terminated by adding glycine (1.25 M). Fixed cells were harvested in SDS buffer with a protease inhibitor and then sonicated to generate DNA of 200–1,000 base pairs (bp) in length. The sheared chromatin-lysed extracts were incubated overnight at 4 °C with the anti-STAT3 antibody or control IgG with rotation. After immunoprecipitation (IP) of the cross-linked protein/DNA, the immunocomplexes were reversed to free DNA. PCR was performed with the input DNA or the immunoprecipitates. The PCR products were separated by agarose gel electrophoresis. The primers used for ChIP are listed in the Supplementary data 3.

Plasmids construction and dual-luciferase reporter assay

DHCR24 promoter-luciferase reporter plasmids containing the DHCR24 promoter region were constructed in the pGL4 plasmid. Wild-type and mutants DHCR24 promoter luciferase constructs were verified by DNA sequencing (Supplementary data 4 and 5). A dual luciferase reporter assay (Promega, WI, USA) was performed according to the manufacturer’s instructions.

Statistical analysis

The statistical analyses were conducted using the SPSS 19.0 software (SPSS). All experiments were repeated independently at least three times. The results are presented as the means ± SD, and the comparisons were evaluated using a two-tailed paired Student’s t-tests. The relationship between DHCR24 expression and clinicopathological parameters was tested by a Chi-Square and Fisher’s exact tests. For survival analysis, the Kaplan-Meier method was carried out to analyze the correlation between OS and other variables in the GraphPad Prism software (GraphPad). Survival curves were analyzed by the log-rank test. For all tests, p-values < 0.05 were considered to be statistically significant.

Additional Information

How to cite this article: Dai, M. et al. Cholesterol Synthetase DHCR24 Induced by Insulin Aggravates Cancer Invasion and Progesterone Resistance in Endometrial Carcinoma. Sci. Rep. 7, 41404; doi: 10.1038/srep41404 (2017).

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.