All subjects were from Shengjing Hospital, which is affiliated with China Medical University. This study was approved by the Ethics Committee of Shengjing Hospital, and all participants provided informed consent.

In total, 180 women with singleton pregnancy, positive oral glucose tolerance test (OGTT), and regular prenatal examinations were selected into the study. The normal group was comprised of 210 cases that were OGTT negative, healthy pregnant women with singleton pregnancy over the same period. Pregnant women with a complication, such as gestational hypertension and pregestational DM were excluded. The GDM group was selected from the outpatients whose glucose was well-controlled and who received systematic examinations during the pregnancy from Nove mber 2012 to September 2014. According to the American Diabetes Association (22), the recommended GDM diagnosis is: i) fasting plasma glucose over 5.1 mmol/l during pregnancy; and ii) blood glucose levels at 1 and 2 h after taking 75-g oral glucose over 10.0 and 8.5 mmol/l during gestational weeks 24 to 28. Either one beyond the diagnostic boundary can be identified as GDM.

The placental tissues of the two groups were collected from each patient immediately after delivery. The tissue was cut into ~0.5 cm³ pieces from placenta on the side of the mother, central and marginal areas. Cold normal saline was used to wash the pieces before they were dissected into small sections, which were blotted dry on filter paper and snap-frozen in liquid nitrogen for 4 h, before being stored at −80°C.

We had statistics on general information in the selected 390 cases, including in pregnant women the age, gestational weeks, birth weight, childbearing history and fasting blood glucose, and the differences were compared between the two groups.

Genomic DNA was extracted from placental samples by a blood/cell/tissue genomic DNA extraction kit (Tiangen Biotech Co., Ltd., Beijing, China) strictly according to the manufacturer's instructions. The test result determination must take the microtiter plate reader as a standard.

PCR primer sequences were as follows: forward primer, 5′-TTCTGGATCCGAGCCACCT-3′ and reverse primer, 5′-AGTGCCTGGTACATTGCTAG-3′ (Invitrogen Life Technologies, Carlsbad, CA, USA). Each PCR reaction contained, DNA template 2 μl, Takara Ex Taq (5 U/μl) 0.25 μl, dNTP mixture (2.5 mM) 4 μl, 1 μl (10 μM) each of SHBG forward and reverse primers, 10X Ex Taq buffer (Mg²⁺ Plus) 5 μl, with ultrapure water added to 50 μl total volume (Takara Bio, Inc., Otsu, Japan). The PCR conditions: initial denaturation at 94°C for 5 min. PCR profiles consisted of 35 cycles with denaturation at 94°C for 10 sec, annealing at 55°C for 30 sec, and extension at 72°C for 60 sec, followed by a final extension at 72°C for 7 min. PCR products were electrophoresed in a 2% agarose gel and visualized by ethidium bromide and UV light (c150; Azure Biosystems, Inc., Dublin, CA, USA).

According to the SNP mutation site gene, we selected the restriction enzyme HinfI (Takara Bio, Inc.) for the treatment of PCR amplified fragments, to carry on the identification of the gene. All of the 180 GDM and 210 in the control group PCR amplified fragments were by HinfI. The reaction system contained Takara HinfI 1 μl, 10X HinfI buffer 2 μl, PCR template 7-8 μl, add ultrapure water to 20 μl, kept in 37°C water for 8 h. The fragments digested by HinfI were electrophoresed in a 3% agarose gel. Genotypes were determined by the results of various bands though UV light (c150; Azure Biosystems, Inc.).

Placental tissue total protein was extracted by RIPA with phenylmethylsulfonyl fluoride (PMSF) (Beyotime Institute of Biotechnology, Shanghai, China). Then put into homogenizer to grind into tissue homogenate and the supernatant was collected and centrifuged at 14,000 × g at 4°C for 5 min. The protein concentration was quantified by BCA assay. Then, samples containing 50 μg were separated by polyacrylamide gel electrophoresis with a 10% separating gel (pH 8.8) and a 5% stacking gel (pH 6.8) (Beyotime Institute of Biotechnology) and the proteins were transferred onto PVDF membranes (Millipore, Billerica, MA, USA). Membranes were incubated with primary antibody: goat anti-human SHBG polyclonal antibody (INC sc-32468, 1:2,000 dilution; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) and GAPDH monoclonal antibody (60004-1-lg, 1:10,000 dilution; Proteintech, Wuhan, China) overnight at 4°C after blocking in 5% skim milk for 2 h. Following three washes with 1X Tris-buffered saline containing 0.1% Tween-20 (TBST), the membranes were incubated in secondary antibody: HBP-conjugated rabbit anti-goat IgG (Ab-204-01, 1:5,000 dilution; Vazyme, Nanjing, China) and peroxidase-conjugated AffiniPure goat anti-mouse IgG (SA00001-1, 1:2,000 dilution; Proteintech) for 120 min then room temperature for 2 h. Immune complexes were detected with enhanced chemiluminescence (ECL) (Beyotime Institute of Biotechnology). A bioanalytical imaging system (c300; Azure Biosystems, Inc.) was used to catch the bands. All experiments were repeated at least three times. Data were expressed as a ratio of SHBG gray value to GAPDH (ImageJ; National Institutes of Health, Bethesda, MA, USA).

The placenta villus trophoblast cells HTR8-SVneo were obtained from Canada Queen's University and cultured conventionally with RPMI-1640 medium, supplemented with 10% heat-inactivated fetal bovine serum (FBS) (Gibco, Carlsbad, CA, USA). Standard culture condition: 37°C, 5% humidified CO₂ incubator. Cell culture supplies were purchased from the Greiner Bio-One GmbH (Frickenhausen, Germany), trypsin (Gibco).

The recombinant lentivirus was designed and synthesized with all of the SBHG genetic sequence labeled by green fluorescence (GenePharma, Shanghai, China). For this part, HTR8-SVneo cells were divided into ten flasks, some of which had upregulated SHBG expression. A and B were the normal groups (without transfection), C and D were the blank control groups (transfected with empty virus LV-5). E-G were negative control groups (respectively transfected with SHBG rs6259 Asp-1, rs6259 Asp-2 and rs6259 Asp-3). H-J were the positive transfection group (respectively transfected with SHBG rs6259 Asn-1, rs6259 Asn-2 and rs6259 Asn-3).

One day before transfection, cells were cultured with 90% RPMI-1640+10% FBS and seeded in 25 cm² flasks (250-500×10³ cells/flasks), cultivated at 37°C, in 5% CO₂ incubator. Twenty-four hours later, the cell growth reached 30-40%, the medium was replaced and the cells were trans-fected with the lentivirus. Cells were incubated at 37°C, morphological structure was observed at 12 h and the culture media was changed at 24 h. After 96 h, transfection efficiency was determined by using fluorescence imaging. Then, the cells were collected, SHBG mRNA and protein were assayed via western blot analysis and RT-qPCR.

Total SHBG mRNA was extracted by TRIzol reagent (Invitrogen Life Technologies) from transfected cells according to the manufacturer's instructions. Amplification of SHGB cDNA fragment was divided into two parts: i) removal of extraneous DNA contamination: 1 μl RNA template, 2 μl 5X g DNA Erase buffer, 1 μl g DNA Erase buffer, 6 μl RNase Free dH₂O, at 42°C for 2 min, then for 4°C; ii) cDNA production: 4 μl 5X g Premix Script buffer 2, 1 μl Premix Script RT enzyme mix, 1 μl RT Primix Mix and 4 μl RNase Free dH₂O into reaction volume from step 1, at 37°C for 15 min, at 85°C for 5 sec, then for 4°C (no. RR047A; Takara Bio, Inc.).

PCR primer sequences were as follows: SHBG mRNA forward primer, 5′-CCTCACCAAGATCACAAAAA-3′ and reverse primer, 5′-TCTCGAAGTCCCAGCATAAACC-3′, giving a fragment length of 120 bp; β-actin forward primer, 5′-AGCACAATGAAGATCAAGATCAT-3′ and reverse primer, 5′-ACTCGTCATACTCCTGCTTGC-3′, giving a fragment length of 127 bp (Invitrogen Life Technologis). Real-time PCR amplification was carried out in 7500 fast thermocycler (Life Technologies) as follows: 10 μl SYBR Premix Ex Taq, 6 μl RNase Free dH₂O, 1 μl each of SHBG and β-actin forward and reverse primers (10 μM), 2 μl cDNA template. Conditions were: 95°C for 5 min (95°C for 10 sec, 60°C for 30 sec) 40 (no. RR820A; Takara Bio, Inc.). Gene expression was determined by 2^−ΔΔCt, where ΔCt = (Ct_SHBG − Ct_GAPDH) and Ct is the threshold cycle.

Expression of SHBG was tested and analyzed in different transfected groups with the same approach as placental tissue.

The direct counting method for % calculation of the gene frequency. The differences with alleles and genotypes distribution compared with Chi-square statistics, P<0.05 was considered statistically significant. Hardy-Weinberg balance check (HWE) the reliability of the survey data (gene frequency, estimation accuracy and reliability) and disease risk with rs6259 through the SHEsis software (http://analysis.bio-x.cn/myAnalysis.php) (23). The correlation between SHBG levels and genotypes were evaluated by Pearson's correlation coefficient. Logistic regression models were used to analyse the correlation of genotype and disease with odds ratios (ORs) and 95% confidence intervals (95% CIs). T-test and LSD t-test were used with α=0.05 considered to be significant. All analyses used SPSS 13.0 software (SPSS, Inc., Chicago, IL, USA).

The statistics and analysis of population characteristics are depicted in Table I, with the mean ± standard deviation. The differences of age distribution, gestational weeks, childbearing history and birth weight had no significant difference, but statistically significant difference wasfound in fasting blood glucose between normal and GDM pregnant womean (P<0.05).

PCR fragment amplification is shown in Fig. 1A. Three genotypes were detected in two groups: the GG genotype represented three segments, while the AA genotype was characterized by two parts and GA genotype was constituted by four segments (Fig. 1B). SHEsis online software detected the Hardy-Weinberg balance, control group: χ²=0.469, P=0.493; GDM group: χ²=0.141, P=0.707. The gene frequency conforms to the laws of genetic balance constantly (P>0.05). Allelic frequencies and genotype distributions are given in Table II. Because of the low proportion of AA homozygotes, they were considered together with the GA heterozygotes. In the GDM pregnancy group, 14.17% had the variant gene, higher than the control group. Compared with the control group, the decreased tendencies were observed in the GDM group with AA and GA group. The result [odd ratio <1, 95% CI=(0-1)] shows that rs6259 is a protective factor for decreased risk of GDM (P=0.043). Pearson's p=0.707 confirms that rs6259 is highly correlated with the morbidity of GDM.

Following the results of PCR-RFLP and enzyme digestion, the placental tissues were divided into two groups. As shown in Fig. 2, the SHBG concentration of the GA heterozygote and AA homozygote (SHBG Asn³²⁷ positive group) are significantly higher than the GG homozygote (SHBG Asp³²⁷/Asp³²⁷ group). The difference between the two groups was statistically significant (t=2.176, P=0.035 <0.001).

HTR8-SVneo cells were incubated in 5% CO₂, 37°C incubator. After repeated exploration the optimal transfection conditions were: MOI=80, 5 μg/ml polybrene, which were successfully transfected with recombinant lentivirus and upregulation of SHBG expression (Fig. 3).

Western blot analysis and gel image software analysis demonstrated that the positive transfection groups showed an increased SHBG level over negative control groups (t=2.675, P=0.011 <0.001), blank control and normal groups (P<0.001). There was no significant difference between the blank control group and the normal group (Table IIIA; Figs. 4 and 5).

According to the LDS-t analysis there were no significant differences between the blank control group and normal cell group (P>0.05), in the positive transfection groups a higher average 2^−ΔΔCt for SHBG mRNA were found than the negative control groups (P<0.001), blank control groups (P<0.001) and normal cell groups (P<0.001), the differences were statistically significant (Table IIIB; Fig. 5).