Estradiol Prevents High Glucose-Induced β-cell Apoptosis by Decreased BTG2 Expression

Hyperglycemia stimulates several pathways to induce pancreatic β-cell apoptosis. In our previous study by mRNA analysis, we demonstrated that B-cell translocation gene 2 (BTG2) expression was up-regulated in INS-1 cells cultured under high glucose conditions, but this effect was reversed by estrogen. In the present study, we demonstrated that BTG2 mRNA and protein expressions in both INS-1 cells and mouse pancreatic islets increased under high glucose conditions compared to those cultured under basal glucose conditions, while in the presence of estrogen, the BTG2 mRNA and protein expressions decreased. SiRNA-BTG2 significantly reduced cell apoptosis, cleaved-caspase 3, and Bax, compared to the siRNA-control in INS-1 cultured under high glucose conditions. We further demonstrated that BTG2 promoter activity was activated under high glucose conditions whereas estrogen significantly reduced it. The effects of estrogen on BTG2 expression were inhibited by estrogen receptor inhibitors. Also, under high glucose conditions, p53 and Bax mRNA and protein expressions increased, but they decreased in the presence of estrogen. Again, the effect of estrogen on p53 and Bax expression was inhibited by estrogen receptor inhibitors. Taken together, this study demonstrates that estrogen reduces pancreatic β-cell apoptosis under high glucose conditions via suppression of BTG2, p53, and Bax expressions.

RNA isolation and reverse transcriptase-polymerase chain reaction. The total RNA was extracted from INS-1 cells or mouse pancreatic islets by using the High Pure RNA Isolation Kit (Roche Diagnostics Corporation, USA) and following the manufacturer's instructions. The concentration of total RNA was measured with a ND-1000 Spectrophotometer (Nanodrop, USA). First-strand complementary DNA (cDNA) was generated from 0.5-1 μg of total RNA using the SuperScript III Reverse Transcriptase (RT) and Random Hexamer Primer (Invitrogen, USA) according to the manufacturer's instructions. Primers were synthesized by Sigma-Aldrich (Sigma-Aldrich, USA). The rat primers for real-time PCR were as follows. The BTG 2 forward primer was 5′-GGT TGG AGA AAA TCG GGA AAC-3′, and the reverse primer was 5′-GCC TTC TGA GAA GCC CTC ATC C-3′ 26 . The Bax forward primer was 5′-CCA GGA CGC ATC CAC CAA GAA GC-3′, and the reverse primer was 5′-TGC CAC ACG GAA GAA GAC CTC TCG-3′ 27 . The β-Actin forward primer was 5′-ATG AAG TGT GAC GTT GAC ATC GTC-3′, and the reverse primer was 5′-CCT AGA AGC ATT TGC GGT GCA CGA TG-3′. The real-time PCR for mouse primers were as follows. The BTG 2 forward primer was 5′-GGT TGG AGA AAA TTG GGA AAC-3′, and the reverse primer was 5′-GCC TTC TAA GAA GCC CTC ATC-3′. The real-time PCR was performed to amplify specific DNA sequences with the Brilliant II SYBR Green QPCR Master Mix (Agilent Technologies, USA). The reactions were carried out on the Mx3005P instrument (Stratagene, USA). The quantity of gene expression was calculated by the 2 −∆∆Ct method and was presented as fold changes, compared to those of the control.
Small interference RNA (siRNA) transfection. Transfection of siRNA directed against BTG 2 mRNA (Dharmacon, USA) was performed using Lipofectamine 2000 (Invitrogen, USA), as detailed by the manufacturer. INS-1 cells were seeded into a 6-well plate for 24 h before transfection. The double-stranded siRNAs were transfected. After 6 h, the medium was changed to complete the culture medium. As a control, the cells were treated with siRNA-Control (Dharmacon, USA) under identical conditions. Twenty-four h after the siRNA transfection, the cells were treated with 11.1 mM or 40 mM glucose for 72 h. They were then harvested, and the BTG 2 , cleaved caspase-3 and Bax were determined using Western blotting. As for the cell lysate preparation, apoptotic and adhered cells were extracted with an RIPA buffer. The lysate was subjected to 15% SDS-PAGE, and the protein expression of BTG 2, cleaved caspase-3 and Bax were determined by immunoblotting. BTG 2 was detected by the anti-BTG 2 antibody (Santa Cruz Biotechnology, USA), the anti-cleaved caspase-3 antibody (Cell Signalling, USA), rabbit polyclonal anti-Bax (Santa Cruz Biotechnology, USA), or the anti-β-actin antibody (Santa Cruz Biotechnology, USA) as an internal control. The membrane was probed with horseradish peroxidase-conjugated antibody. The immunoreactive proteins were visualized by SuperSignal West Pico Chemiluminescent Substrate (Thermo Scientific, Rockford, IL, USA), and were detected by using a G:BOX chemiluminescence imaging system (Syngene, Frederick, MD, USA).
ScientiFic REPORTS | (2018) 8:12256 | DOI:10.1038/s41598-018-30698-x Western blotting analysis. The total protein of INS-1 cells and mouse pancreatic islets were extracted by using a RIPA buffer. Nuclei proteins were extracted from the cells by using the Nuclear and Cytoplasmic Extraction Reagent Kit (Pierce, USA). The protein concentrations were then determined by a micro BCA assay. The total protein was separated on a 4-12% (wt/vol) SDS-PAGE. After that, the protein was transferred to a polyvinylidene fluoride (PVDF) membrane (Bio-Rad, USA). The membrane was blocked with 5% skimmed milk before being incubated overnight at 4 °C with one of the following primary antibodies: rabbit polyclonal anti-BTG 2 (Santa Cruz Biotechnology, USA), rabbit polyclonal anti-p53 (Santa Cruz Biotechnology, USA), rabbit polyclonal anti-Bax (Santa Cruz Biotechnology, USA), or mouse monoclonal anti-β-Actin (Santa Cruz Biotechnology, USA). After washing, the membrane was incubated with one of the following secondary antibodies: horseradish peroxidase-conjugated anti-rabbit IgG (Santa Cruz Biotechnology, USA), or horseradish peroxidase-conjugated anti-mouse IgG (Santa Cruz Biotechnology, USA), at room temperature. The protein bands were detected with an enhanced chemiluminescence system (Pierce Biotechnologies, USA) and exposed on x-ray films. The band intensities of proteins were quantified by using ImageJ v 1.43 software. All Western blot results were shown in supplement data.
Promoter assay. The INS-1 BTG 2 promoter (−43 to −1802) was amplified from INS-1 genomic DNA by PCR using Pfu DNA polymerase (Stratagene, La Jolla, CA, USA). The PCR products of the BTG 2 promoter were confirmed by automated DNA sequencing before being separately subcloned into pGL3 reporter vectors to generate INS-1 BTG2 promoter-firefly luciferase reporter plasmids.
The INS-1 cells were transfected with 1 µg luciferase reporter plasmid, pGL3-basic, or pGL3-Btg2 gene promoter together with an internal control renilla luciferase plasmid, pRL-SV40. After transfection and culturing for 24 h, the culture medium was changed into a basal glucose-containing medium or a high glucose-containing medium, with or without 10 −8 M estrogen, before being cultured for 72 h. The firefly luciferase activity was normalized by the internal control renilla luciferase activity. The dual-luciferase reporter assay was performed according to the manufacturer's instructions (Promega Corp., Fitchburg, WI, USA). The experiments were performed in six-plicate and on three independent occasions. Statistical Analysis. Data were analyzed by using SPSS Statistics for Windows, version 17 (SPSS Inc., Chicago, Ill., USA) and expressed as mean ± standard error of mean (S.E.M). The differences between the groups of results were determined by one-way ANOVA, followed by Tukey's post hoc test. A P-value less than 0.05 was considered to be statistically significant.

Results
Estradiol increased pancreatic β-cell viability after culture under high glucose conditions. To examine whether estradiol increased pancreatic β-cell viability under high glucose conditions, INS-1 cells were cultured under different conditions before measuring the apoptotic cell death by the cleaved-caspase 3 activity. INS-1 cells cultured in normal glucose were used as a control, and 10 −8 M 17-β estradiol did not change the cleaved-caspase 3 activity compared to that of the control. As expected, the cleaved-caspase 3 significantly increased in INS-1 cells cultured in a high glucose medium compared to that of the control. In contrast, INS-1 cells cultured in a high glucose medium with 10 −8 M 17-β estradiol significantly reduced the cleaved-caspase 3, suggesting that estrogen increased viable cells when the cells were cultured in high glucose ( Fig. 1A).

High glucose conditions increased BTG 2 expression in pancreatic β-cells, and effect reversed by estradiol.
To identify the signaling pathway of estradiol that decreased pancreatic β-cell death against the high (40 mM) glucose medium, a signaling RT² Profiler PCR Array was performed. The preliminary results suggested that the BTG2 mRNA expression was higher in the high glucose medium than in the normal glucose medium (data not shown). To confirm the RT² Profiler PCR Array results, a conventional real-time PCR was performed for the samples from the experimental conditions. INS-1 cells cultured in the high glucose medium had a significantly increased BTG 2 mRNA expression compared to those cultured in normal glucose. The presence of estradiol in the high glucose medium significantly reduced the BTG 2 mRNA expression (Fig. 1B). The BTG 2 protein expression corresponded with the BTG 2 mRNA expression (Fig. 1C).
To examine the effects of high glucose and estrogen on BTG 2 mRNA and protein expression, mouse pancreatic islets were cultured under experimental conditions, and real-time PCR and Western blot analyses were performed. The BTG 2 mRNA and protein expressions were significantly upregulated by the high glucose. Estrogen significantly reduced the BTG 2 mRNA and protein expressions compared to those cultured in high glucose alone ( Fig. 2A). Thus, a high glucose medium increased BTG 2 mRNA expression in both INS-1 and islets, whereas estradiol reversed BTG 2 mRNA and protein expressions in both INS-1 cells and islets in high glucose conditions (Fig. 2B). BTG 2 knockdown rescued pancreatic β-cells apoptosis from high-glucose conditions. To investigate the role of BTG 2 in protecting pancreatic β-cells apoptosis, BTG 2 silencing was performed in INS-1 cells cultured in basal and high glucose media ( Fig. 3A-C). After SiRNA-BTG 2 knockdown, cellular apoptosis was determined by the detection of cleaved-caspase 3 and Bax using Western blotting analysis. SiRNA-BTG 2 diminished the BTG 2 protein expression in INS-1 cells cultured in basal and high-glucose media, and cleaved-caspase 3 and Bax were significantly decreased in INS-1 cells with SiRNA-BTG 2 knockdown cultured in a high glucose medium. These findings were similar to the results for cells cultured in a basal glucose medium with mock treatment, siRNA-control and siRNA BTG 2 , whereas INS-1 cells cultured in a high glucose medium with mock treatment and siRNA-control showed markedly increased cleaved-caspase 3, BTG 2 and Bax protein levels compared with those cultured in a basal glucose medium. To confirm these results, SiRNA-BTG 2 knockdown was performed and cell apoptosis was assessed by Annexin V/PI staining. SiRNA BTG 2 significantly decreased ScientiFic REPORTS | (2018) 8:12256 | DOI:10.1038/s41598-018-30698-x cell apoptosis when compared to siRNA-control. These results suggest that BTG 2 silencing protects against high-glucose-induced pancreatic β-cell apoptosis.
Estradiol regulated BTG 2 promoter activity. In a breast cancer study, it was demonstrated that estradiol suppressed the BTG 2 promoter in MCF-7 and Hela cells 28 . To examine whether estrogen regulated BTG 2 mRNA expression, the INS-1 BTG 2 promoter (−43 to −1802) was cloned into pGL3 reporter vector. INS-1 cells in high glucose conditions significantly increased BTG 2 promoter activity compared to those cultured under basal glucose conditions. Estradiol in high glucose condition significantly reduced BTG 2 promoter activity compared to the high glucose condition alone (Fig. 3D). The presence or absence of estradiol under the basal glucose conditions did not change BTG 2 promoter activity. This result confirms that high glucose condition induces BTG 2 promoter activity, but the addition of estradiol into INS-1 cells cultured under high glucose conditions decreases BTG 2 promoter activity.  The data IS presented as mean ± S.D. of 3 independent experiments. * P < 0.05, ** P < 0.01, *** P < 0.001, compared to the high-glucose-treated group.

Estradiol decreased BTG
ScientiFic REPORTS | (2018) 8:12256 | DOI:10.1038/s41598-018-30698-x respectively, did not abolish the effect of estradiol in reducing the BTG 2 mRNA and protein expressions. Also, G15, a membrane estrogen receptor inhibitor, did not by itself diminish the effect of estradiol. Only in the presence of both ICI 182,780 and G15 was the effect of estradiol diminished, with no difference in BTG 2 mRNA and protein expressions evident compared with those for the high glucose conditions (Fig. 4A,B).
It has been proposed that BTG 2 induces apoptosis via activated Bax 29 . To correlate BTG 2 expression and pancreatic β-cell apoptosis, Bax mRNA and protein expressions were measured by RT-PCR and Western blot analyses. High glucose conditions significantly increased Bax mRNA and protein expressions compared to basal glucose conditions. INS-1 cells cultured with estradiol and a high glucose medium significantly reduced Bax mRNA and protein expressions compared to those cultured in a high glucose medium alone. To examine whether Bax mRNA and protein expressions responded in a similar manner to BTG 2 in the presence of nuclear and/or membrane estrogen receptor inhibitors, ICI 182,780, 4-HT and G15 were added under the experimental conditions. Comparable with BTG 2 expression, Bax mRNA and protein expression induction under the high glucose conditions were decreased by estradiol. The effect of estradiol in the high glucose conditions was attenuated by a combination of ICI 182,780 and G15 (Fig. 4C,D).
Estradiol reduced p53 protein expression. BTG 2 expression uses either a p53-dependent or a p53-independent pathway in prostate carcinoma cells 9 . To examine whether BTG 2 expression is associated with p53, INS-1 cells cultured under experimental conditions were assessed for nuclear p53 expression by Western blot analysis. High glucose conditions significantly increased the p53 protein expression in the nucleus compared to that under basal glucose conditions. However, estradiol significantly reduced the p53 protein expression in the nucleus compared to the high glucose conditions alone. Neither the nuclear estrogen receptor inhibitor nor the membrane estrogen receptor inhibitor reversed the estradiol effect when co-cultured in a high glucose medium. In the presence of both, the nuclear and membrane estrogen receptor inhibitors attenuated the effects of estradiol on the nuclear p53 expression under high glucose conditions (Fig. 5A).

Estrogen reduced p53 and Bax protein expressions in mouse pancreatic islets.
To confirm the effects of high glucose and estradiol that were observed in INS-1 cells, mouse pancreatic islets were isolated and cultured under basal and high glucose conditions with or without estradiol for 72 h. As observed in the INS-1 cells, the high glucose conditions induced p53 and Bax protein expressions in mouse pancreatic islets compared to those cultured in the basal glucose medium. Estradiol with high glucose significantly reduced the p53 and Bax protein expressions, compared to those cultured in high glucose alone (Fig. 5B,C).

Discussion
Hyperglycemia is a stressful condition that produces both oxidative and ER stress 4,30,31 . Both types of stress cause DNA damage 32,33 , which activates early growth response genes 34 . BTG 2 is one of the early growth response genes 35 . BTG 2 has different effects, depending on the cell type 7 . In our preliminary study by mRNA analysis using the RT 2 PCR array, the results showed that high glucose conditions increased BTG 2 mRNA expression, and estradiol reversed the effect of the high glucose. BTG2 seemed to correlate with high-glucose-induced cell death. This hypothesis was tested by this study, which aimed to demonstrate the association of the BTG 2 level and high-glucose-inducing cell death. The results of this study showed that the high glucose conditions increased cell death and up-regulated the BTG 2 mRNA and protein expressions. The fold of the BTG 2 mRNA expression with conventional real-time PCR was lower than with the RT 2 PCR array. This might be due to the better optimized conditions of the commercial RT 2 PCR array than conventional real-time PCR. Although the specificity of the primers was different, the pattern of BTG 2 mRNA induction was similar. The up-regulation of BTG 2 was found in both rat pancreatic β-cell line (INS-1 cells) and mouse pancreatic islets. BTG 2 is known as an immediate early gene which responds to stress 36 . High glucose levels produced cellular stress in the form of oxidative and endoplasmic reticulum stress 4,30,31 . Thus, the cellular stress produced by high glucose likely stimulated BTG 2 expression. Also, BTG 2 has been proposed as a protein involved in the programed cell death of PC12 37 . On the contrary,  Another beneficial role of BTG2 has been reported that BTG 2 is a co-activator to up-regulating antioxidant 38 . This role is supported our study that estrogen has a trend to increase BTG 2 in basal glucose. Estrogen might up-regulate antioxidant via increase BTG 2 . It is known that BTG 2 plays a role in both physiological and pathological processes 7 . Our knockdown BTG 2 experiment indicated that high-glucose-induced BTG 2 is a pathological process, whereas the up-regulation of BTG 2 by GLP-1 is a role of BTG 2 in physiological processes.
This study also demonstrated that estradiol protected pancreatic β-cell apoptosis against high glucose via decreased BTG 2 mRNA and protein expressions. Again, this finding was found in both rat pancreatic β-cell lines (INS-1 cells) and mouse pancreatic islets. A previous study suggested that estrogen reduced BTG 2 transcription in breast cancer cells 28 . That study also suggested that the estrogen receptor can interact with other transcription factors, including AP-1, Sp1, p53 and NF-kB, which are contained in the BTG 2 promoter. Furthermore, they performed ChIP-on-chip analysis and found that the ERα binding site was present around −2000 upstream of the BTG 2 start site. To examine this possibility in pancreatic β-cells in this present study, the INS-1 BTG 2 promoter was cloned to perform a promoter assay. The promoter assay confirmed that high glucose increased BTG 2 promoter activity, while estrogen significantly decreased BTG 2 promoter activity. The promoter assay results support our previous findings. In breast cancer cells, it was demonstrated that the estrogen receptor alpha plays a role in the reduction of BTG 2 promoter activity 28 . To investigate this observation, 4 HT (the estrogen receptor alpha inhibitor), ICI 182,780 (the nuclear receptor inhibitor), and G15 (the G-protein coupled estrogen receptor inhibitor) were added to the culture experiments. Estradiol effect on BTG 2 expression was ameliorated in the presence of both the nuclear and G-protein coupled estrogen receptor inhibitors. This suggests that estrogen exerts its effect through both the nuclear and G-protein coupled estrogen receptors. In parallel with our previous study, it has been shown that estrogen decreases ER stress and cell apoptosis via the nuclear and membrane estrogen receptors 23,39 . It is worth mentioning that estrogen has been known to protect pancreatic β-cell apoptosis against toxic substances through multiple pathways 16,40,41 . BTG 2 is known to induce cell apoptosis via increased Bax 29 . Our result confirmed that Bax mRNA and protein expressions were altered in response to the BTG 2 expression. The activated Bax bound together to form a homodimer and then inserted pores on the mitochondrial membrane and released cytochrome C 42 . The released cytochrome C triggers the mitochondrial-induced apoptosis pathway 43 . This result provided a mechanism for BTG 2 -induced pancreatic β-cell apoptosis through Bax. It is known that BTG 2 expression can be induced through a p53-dependent or a p53-independent mechanism 13,14 . This study further showed that the nuclear p53 level is increased. Normally, p53 is inactivated in the cytoplasm compartment. When p53 is activated, the activated p53 moves into the nucleus 44 . The activated p53 acts as a transcription factor to activate the expression of apoptotic genes such as Bax 45 . p53-induced cell apoptosis was also found in cardiac myocyte cultured under high glucose conditions 46 . Our results showed that high glucose conditions increased the p53, BTG 2 and Bax in the INS-1 cells and mouse isolated pancreatic islets. Knockdown BTG 2 significantly decreased Bax in high glucose condition. Thus, it is likely that the high glucose condition increased pancreatic β-cell apoptosis through the p53-BTG 2 -Bax pathway. Furthermore, our results showed that estradiol directly suppressed BTG 2 promoter activity. Estradiol might separately suppress both BTG 2 and p53 expression. Estrogen-reduced p53 signaling has been observed in other cells. In breast cancer, induction of p53 increases cell apoptosis, whereas estrogen promotes breast cancer cell proliferation by a decreased p53 pathway 47 . Estrogen-protected ischemia reperfusion induced cardiomyocytes apoptosis by suppression of the p53 pathway 48 . Estrogen prevented mesangial cells apoptosis through inhibition of p53 expression 49 . It is possible that estrogen suppressed BTG 2 expression through reduced p53. However, our results also showed that estradiol directly suppressed transcriptional activation of the BTG2 promoter by the luciferase promoter assay. Thus, our results suggest that estrogen might suppress both the p53 and BTG2 promoters.
In summary, our results show that high glucose conditions induce BTG2, p53 and Bax expressions, which are associated with increased pancreatic β-cell apoptosis (Fig. 6). Estradiol can suppress the BTG2 promoter under high glucose conditions. The protective effect of estradiol against high-glucose-induced cell death through the reduction of BTG2, p53 and Bax expressions is diminished by inhibition of both the nuclear and the membrane estrogen receptors. However, the detailed molecular mechanisms on how estrogen suppresses p53 and BTG2 require further investigation.