A total of 100 Kunming female mice (weight, 22±1.8 g; age, 6 weeks) were used to generate this animal model. Ethical approval for the use of animals in this study was obtained from Chongqing Medical University (Chongqing, China). Care and handling of these mice was conducted in accordance with the animal research committee guidelines established by the Ethics Committee of the Institute of Zoology, Chongqing Medical University. An insulin intervention mouse model was established as previously described by Ou et al (11). Mice were randomly divided into control and insulin-treated groups (n=50/group). The treatment protocol was composed of 2 subcutaneous injections of saline for the control group or insulin for insulin-treated group, in order to induce hyperinsulinemia. Human recombinant insulin (Insulin glargine) was purchased from Sanofi (Shanghai, China). For the insulin-treated group, 0.05 IU insulin was injected until day 16, and from day 17 on it was gradually increased until reaching 0.8 IU/day by day 23 (20,21) (day 17, 0.2 IU; day 18, 0.35 IU; day 19, 0.5 IU; day 20, 0.65 IU; day 21, 0.8 IU; day 22, 0.8 IU; day 23, 0.8 IU). All mice were then mated with fertile males (female: male, 2:1) overnight and the female mice were checked for vaginal plugs the following morning. The day at which a positive vaginal plug was identified was considered to be day 1 (D1) of pregnancy.

Blood samples (~1 ml) were collected from the eye socket on insulin treated day 23 and incubated at room temperature for 3 h to obtain the serum. The serum levels of insulin, P4 and E2 were detected using enzyme-linked immunosorbent assay kits (YH1332, YH4016 and YH3997; Shanghai Yan Hui Biological Technology, Co., Ltd., Shanghai, China) according to the manufacturer's protocol. Absorbance was read at 450 nm. An ACCU-CHEK Advantage blood glucose meter (Roche Applied Science, Mannheim, Germany) was used to determine the glucose concentrations in the blood, which were taken from the cut tail tip. Homeostasis model assessment (HOMA) were calculated as: Plasma glucose level×serum insulin level/22.5.

In order to examine attachment and implantation, pregnant dams were sacrificed by cervical dislocation on D5 and following Trypan blue (10 mg/ml) injection into the tail veins (22). The number of implantation sites was identified by distinct blue bands. In addition, endometrial tissues at D4-D6 were collected and squeezed from the uterine horns by the use of fine forceps under an anatomical microscope (23). The isolated endometrial tissues were immediately stored in liquid nitrogen.

TRIzol reagent (Takara Biotechnology Co., Ltd., Dalian, China) was used for total RNA extraction from 50 mg endometrium obtained from D4-D6, according to the manufacturer's protocol. RNA concentration and purity was determined by spectrophotometric determination of OD_260/280 ratio. Total RNA (0.5 mg) was used for cDNA synthesis in a 10 ml reaction system using the PrimeScript RT reagent kits (Takara Biotechnology Co., Ltd.). The specific primers for estrogen receptor (Esr) 1, Esr2, progesterone receptor (Pgr), homeobox A10 (Hoxa10) and β-actin are presented in Table I. qPCR was performed using the CFX96 Real-Time PCR Detection system (Bio-Rad Laboratories, Inc., Hercules, CA, USA) and SYBR Premix Ex Taq (Takara Biotechnology Co., Ltd.). The reaction mixture consisted of 12.5 µl SYBR Premix Ex Taq (2X), 0.5 µl of each specific primer (10 µM) and 2 µl cDNA. Water was then added to a final volume of 25 µl. Following initial denaturation at 95°C for 30 sec, PCR was performed for 40 cycles according to the following parameters: Denaturation at 95°C for 10 sec; annealing and extension at 60°C for 30 sec. Experiments were performed in triplicate for each sample. Gene expression was obtained by normalizing cDNA quantity to that of β-actin and calculated using the 2^−ΔΔCq method (24).

Proteins were extracted from 200 mg endometrium obtained from D4-D6 mice. Endometrial tissues were lysed using RIPA lysis buffer (Beyotime Institute of Biotechnology, Jiangsu, China). The concentration of protein was determined using a bicinchoninic acid protein assay kit (Beyotime Institute of Biotechnology). Proteins were boiled in 5X SDS sample loading buffer for a minimum of 10 min and subsequently stored at −80°C prior to use. 50 µg proteins were separated on 10% SDS-PAGE gels. Proteins were then transferred onto polyvinylidene difluoride membranes (Bio-Rad Laboratories, Inc.). Membranes were subsequently blocked with 5% non-fat milk at room temperature, before they were incubated with rabbit monoclonal anti-Hoxa10 (SC17159; 1:500; Santa Cruz Biotechnology, Inc., Dallas, TX, USA), mouse monoclonal anti-Esr1 (ab66102; 1:1,000; Abcam, Cambridge, UK), mouse monoclonal anti-Pgr (ab2765; 1:500; Abcam), mouse monoclonal anti-Esr2 (ab288; 1:1,000; Abcam), rabbit monoclonal anti-mechanistic target of rapamycin (mTOR; 2983; 1:500; Cell Signaling Technology, Inc., Danvers, MA, USA), rabbit monoclonal anti-phosphorylated (p)-mTOR (5536; 1:500; Cell Signaling Technology, Inc.), rabbit monoclonal anti-ribosomal protein S6 kinase β-1 (p70S6K; 2708; 1:1,000; Cell Signaling Technology, Inc.), rabbit monoclonal anti-p-p70S6K (9234; 1:500; Cell Signaling Technology, Inc.) and mouse monoclonal anti-β-actin (A5441; 1:1,500; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) overnight at 4°C. The membranes were washed 3 times with Tris-buffered saline containing Tween-20 (0.2%), and then incubated for 1 h on room temperature with a horseradish peroxidase (HRP)-conjugated secondary antibody (TA130003 and TA130023; 1:1,000; OriGene Technologies, Inc., Beijing, China). The positive bands were detected using HRP-enhanced chemiluminescence reagents (WBKLS0500; Merck KGaA) and quantified by densitometry analysis using Quantity One software version 4.6.2 (Bio-Rad Laboratories, Inc.). Protein expression was normalized to β-actin.

H&E staining was performed according to the manufacturer's protocol (Nanjing Jiancheng Bioengineering Institute, Nanjing, China). Briefly, sections (4 µm in thickness) of uterine tissue sacrificed on D5 were deparaffinized in xylene and rehydrated in decreasing concentrations of ethanol. Subsequently, the sections were stained with hematoxylin (0.45%) for 3 min and with eosin (1%) for 30 sec at room temperature, followed by dehydration until they were cleared. Finally, the tissue sections were mounted with neutral gum.

Mice uteri at D5 were fixed in 4% paraformaldehyde and embedded in paraffin. The tissue was then cut to 5 µm sections. Immunohistochemistry was performed using DAB color reagent kit (SP-9000; Zhongshan Biosciences Inc. China) according to the manufacturer's protocols. The sections were examined and imaged under a microscope (BX43, Olympus Corporation, Tokyo, Japan). Non-immune goat serum (Zhongshan Biosciences Inc. China) was used instead of the primary antibody as a negative control.

All experiments were repeated at least three times. The data were analyzed using SPSS software (version, 16.0; SPSS Inc., Chicago, IL, USA). The Student's t-test was used to determine the differences in serum insulin, P4 and E2 levels, mRNA and protein expression levels and the number of implantation sites between the control and insulin-treated groups. P<0.05 was considered to indicate a statistically significant difference.

In order to determine the effects of maternal insulin resistance on embryo implantation, a mouse model of hyperinsulinemia was established through exogenous insulin injection (11,21). The effects of chronic treatment with insulin on mouse weight, and the serum levels of insulin, E2 and P4 are presented in Table II. In addition, serum glucose levels and the HOMA values were significantly higher in the insulin-treated group. The results demonstrated that the insulin-treated group developed chronic insulin resistance and hyperinsulinemia.

A previous study demonstrated that endometrial receptivity is maintained on day 5 of pregnancy in mice (25). In order to determine the potential role of hyperinsulinemia on endometrium receptivity, the present study recorded the number of embryo implantation sites on D5 by counting the distinctive blue bands on the mice uteri as a result of Trypan blue staining. No significant difference between the number of implantation sites in the insulin-treated group was observed when compared with the control group (Fig. 1A). In addition, no alterations in the morphology of the uteri between the two groups were observed (Fig. 1B).

Although no significant difference was evident between the two groups when examined visually, it was unclear whether endometrium receptivity was altered at the microscopic scale. In order to investigate this further, morphological examination of uterine cross-sections was performed at D5, and no marked difference between the control and insulin-treated groups was observed (Fig. 2). During the implantation window, the process of embryo implantation is divided into the following three steps: Pre-implantation on day 4, peri-implantation on day 5 and post-implantation on day 6 of gestation in mice (26,27). The expression levels of molecules associated with endometrial receptivity, such as Esr1 and Pgr may vary, depending on their functions during the different steps of the embryo implantation process. The present study determined that expression levels of Esr1, Pgr and Hoxa10 were significantly upregulated (P<0.05, Fig. 3) and Esr2 levels were downregulated in the insulin-treated group when compared with the control group on D4 (P<0.05, Fig. 3). However, no obvious differences in Pgr expression levels were observed during peri-implantation on D5 (Fig. 3). Unlike Pgr, Esr1, Esr2 and Hoxa10 were significantly imbalanced in mice with hyperinsulinemia (P<0.05; Fig. 3B, D and E). In addition, a marked reduction in the levels of these molecules was recorded during post-implantation on D6 of gestation in mice (Fig. 3C and D). Therefore, it is possible that the establishment process of endometrium receptivity was altered by maternal hyperinsulinemia, even though the embryo was able to implant into endometrium.

A previous study revealed that mTOR was essential for endometrial receptivity (18), and previous studies have demonstrated that insulin could serve its roles though mTOR (28,29). Therefore, the present study investigated mTOR signaling in the endometrium following insulin treatment. During the implantation window, the vital time for endometrial receptivity, western blotting of endometrial tissues revealed a significant increase in endometrial p-mTOR and p-p70S6K protein expression levels in the insulin-treated group when compared with the control group (P<0.05; Fig. 4). No marked difference in the levels of total mTOR and p70S6K protein expression levels was identified in the insulin-treated group when compared with the control group (Fig. 4).