High Glucose–Repressed CITED2 Expression Through miR-200b Triggers the Unfolded Protein Response and Endoplasmic Reticulum Stress

Diabetes induces cell apoptosis, a causal event among many diabetes complications (1–3). Our previous studies have demonstrated that maternal diabetes in pregnancy triggers program cell death in the developing neuroepithelium, which leads to neural tube defects (NTDs) in offspring (3–7). Human studies have shown that as the degree of maternal hyperglycemia increases, so does the incidence of NTDs (8,9). In addition, high glucose directly induces neural stem cell apoptosis and NTD formation in cultured rodent embryos (5,6), supporting the hypothesis that high glucose is a primary causative event in diabetic embryopathy. Although it has been shown that endoplasmic reticulum (ER) stress, gene dysregulation, and caspase activation (10–12) are involved in maternal diabetes in vivo and high glucose in vitro–induced apoptosis (3,13–15), it is unknown whether altered gene expression contributes to unfolded protein response (UPR) and ER stress. It is also unclear how high glucose alters development gene expression that leads to NTD formation.

microRNAs (miRNAs) are small endogenous noncoding RNAs that play important gene-regulatory roles (16–18). miRNAs suppress gene expression by binding to the 3′-untranslated region (3′-UTR) of mRNAs with imperfect complementation, resulting in direct translational repression, mRNA destabilization, or a combination of the two (19). Primary miRNA transcripts are cleaved by the endonuclease Drosha to produce precursor miRs, which are 60- to 70-nt-long imperfect hairpin structures (20). In the cytoplasm, precursor miRNAs are further processed by the endonuclease Dicer, resulting in a mature miRNA, the guide strand for gene regulation, and a passenger strand, which is degraded and does not play a role in gene regulation (16). miRNAs exert pleiotropic actions under physiological and pathophysiological conditions (17,21). Differential miRNA expression is observed during embryonic neurulation and in maternal sera of NTD-affected pregnancies (22,23), suggesting a possible role of miRNAs in diabetes-induced NTDs.

ER stress is indispensible for high glucose–induced neural stem cell apoptosis. Misfolded protein accumulation in the ER lumen leads to the activation of the UPR (24–26). The ER transmembrane proteins, IRE1α (inositol-requiring enzyme 1 α) and PERK (protein kinase RNA-like endoplasmic reticulum kinase), act as sensors of UPR signaling (24). Prolonged UPR signaling causes a severe form of ER stress, which induces apoptosis (27). A recent study has demonstrated that decreased miRNA-322 leads to ER stress (28); therefore, it is highly possible that there is a link between miRNAs and ER stress and that both participate in high glucose–induced UPR, ER stress, and apoptosis.

CBP/p300-interacting transactivator with ED-rich tail 2 (CITED2) is a transcription coactivator that is essential for embryogenesis, and its deletion causes NTDs by increasing neural stem cell apoptosis (29,30). However, information related to the regulation of CITED2 expression and its cellular functions is scarce. Because high glucose induces NTDs similar to those seen in Cited2^−/− embryos (3–5), we hypothesize that altered CITED2 expression mediates the effect of high glucose on UPR activation and ER stress leading to neural stem cell apoptosis.

In the current study, we found that high glucose upregulates miR-200b expression and that elevated miR-200b promotes ER stress and apoptosis through silencing CITED2. It has been demonstrated that CITED2 is required for the survival of neural stem cells of the developing neural tube (29,30). Therefore, we reveal a new miR-200b–CITED2–ER stress circuit that may play a critical role in high glucose–induced cell apoptosis that is the etiology of the initiation of diabetes and a range of diabetes complications.

Wild-type (WT) C57BL/6J mice were purchased from The Jackson Laboratory (Bar Harbor, ME). Superoxide dismutase 1 (SOD1)-transgenic (Tg) mice in C57BL/6J background were revived from frozen embryos by The Jackson Laboratory (stock no. 002298). Streptozotocin (STZ) from Sigma-Aldrich (St. Louis, MO) was dissolved in sterile 0.1 mol/L citrate buffer (pH 4.5). The SOD1 mimetic tempol was purchased from Alexis Italia (Vinci, Florence, Italy). Sustained-release insulin pellets were purchased from Linshin, Canada. The procedures for animal use were approved by the University of Maryland School of Medicine Institutional Animal Care and Use Committee.

Our mouse model of diabetic embryopathy has previously been described (3,5). Briefly, 10-week old WT female mice were injected daily with 75 mg/kg i.v. STZ over 2 days to induce diabetes. Using STZ to induce diabetes is not a complicating factor because STZ is cleared from the bloodstream rapidly (STZ serum half-life, 15 min) (31) and pregnancy is not established until 1–2 weeks after STZ injection (7). Diabetes was defined as a 12-h fasting blood glucose level of ≥14 mmol/L. Insulin pellets were subcutaneously implanted in these diabetic mice to restore euglycemia prior to mating. To generate SOD1 embryos, we crossed SOD1-Tg male mice with nondiabetic or diabetic WT female mice. Male and female mice were paired at 3:00 p.m., and pregnancy was established by the presence of the vaginal plug the next morning, and noon of that day was designated as embryonic day 0.5 (E0.5). On E5.5, insulin pellets were removed to ensure that the developing embryos would be exposed to a hyperglycemic environment during neurulation (E8–E10.5). WT female mice were treated with vehicle injections and sham operations on insulin pellet implantation/removal as nondiabetic controls. On E8.75 (at 6:00 p.m.), mice were euthanized and conceptuses were dissected out of the uteri, and then yolk sacs were removed from the embryos. The embryos were used for analyses.

C17.2 mouse neural stem cells, obtained from European Collection of Cell Culture, were maintained in DMEM (5 mmol/L glucose) supplemented with 10% FBS, 100 units/mL penicillin, and 100 µg/mL streptomycin. The C17.2 cells are newborn mouse cerebellar progenitor cells transformed with retroviral v-myc (32). Lipofectamine RNAiMAX (Invitrogen, Grand Island, NY) was used according to the manufacturer’s protocol for transfection of small interfering (si)RNA into the cells, using 1% FBS culture conditions. The mirVana miRNA mimic and the miRNA inhibitor of miR-200b, the negative control oligo (cat. no. 4464058) for the miR-200b mimic, and the negative control oligo (cat. no. 4464076) for the miR-200b inhibitor were purchased from Ambion (Austin, TX). The miR-200b inhibitor is a single-stranded RNA-based oligonucleotide that is designed to bind to and inhibit endogenous miR-200b. The biotin-labeled miR-200b and the biotin-labeled negative control (Caenorhabditis elegans miR-67) were custom made by Dharmacon (Lafayette, CO). CITED2-siRNA (sc-35960) and control siRNA-A (sc-37007) were obtained from Santa Cruz Biotechnology (Dallas, TX).

The full-length CITED2 coding region (CR) or its 3′-UTR and different 3′-UTR fragments with or without the predicted miR-200b binding sites were amplified and subcloned into the pmirGLO Dual-Luciferase miRNA Target Expression Vector (Promega, Madison, WI) to generate the pmirGLO-Luc-CITED2-CR and pmirGLO–CITED2–3′-UTR. The sequence and orientation of the fragments in the luciferase reporters were confirmed by DNA sequencing and enzyme digestions. Luciferase activity was measured using the Dual-Luciferase Assay System (Promega), and the levels of pmirGLO-Luc-CITED2-CR or pmirGLO–Luc–CITED2–3′-UTR luciferase activities were normalized to Renilla luciferase activity and were further compared with the levels of luciferase mRNA in every experiment. All primer sequences for generating these constructs are provided in Supplementary Table 1.

Immunoblotting were performed as previously described (3,33). The following primary antibodies were used at dilutions of 1:1,000 to 1:2,000 in 5% nonfat milk: CITED2 (Santa Cruz Biotechnology), CHOP, phospho-(p-)PERK, (p-)IRE1α, (p-)eIF2α (eukaryotic translation initiation factor 2A), and cleaved caspase 3 (Cell Signaling Technology). Signals were detected using the SuperSignal West Femto Maximum Sensitivity Substrate kit (Thermo Scientific). All experiments were repeated in triplicate. Antibody information is listed in Supplementary Table 2.

Total RNA was isolated from embryos or cultured cells using the mirVana miRNA isolation kit (Ambion) and reverse transcribed using the NCode VILO miRNA cDNA synthesis kit (Invitrogen). Real-time PCR for CITED2, β-actin, miR-200b, and small nuclear RNA U6 was performed using Maxima SYBR Green/ROX qPCR Master Mix assay (Thermo Scientific). Real-time PCR and subsequent calculations were performed by a StepOnePlus Real-Time PCR System (Applied Biosystem). Real-time PCR primer sequences are listed in Supplementary Table 3.

The biotin-labeled miR-200b or the biotin-labeled negative control (C. elegans miR-67) was transfected into cells for 48 h, and then whole-cell lysates were collected. Cell lysates were mixed with streptavidin-coupled Dynabeads (Invitrogen) and incubated at 4°C on a rotator overnight. After the beads were washed thoroughly, the bead-bound RNA was isolated and subjected to reverse transcription followed by real-time PCR analysis. Input RNA was extracted and served as a control.

The whole-embryo culture system was previously described (5,6). E8.5 embryos were cultured in 4 mL rat serum at 38°C in 30 revolutions/min rotation in the roller bottle system. The culture bottles were gassed with 5% O₂/5% CO₂/90% N₂ for the first 24 h and 10% O₂/5% CO₂/85% N₂ for the last 12 h.

The TUNEL assay was performed by using the In Situ Cell Death Detection kit (Millipore) (5). After transfection and high glucose treatment, cells were fixed with 4% paraformaldehyde in PBS and incubated with TUNEL reagent, counterstained with DAPI, and mounted with aqueous mounting medium (Sigma). TUNEL-positive cells in each well were counted. The percentage of apoptotic cells was calculated as the number of TUNEL-positive (apoptotic) cells, divided by the total number of cells in a microscopic field from three separate experiments.

Data are presented as means ± SE. Three embryos from three separate dams were used for the in vivo studies, and cell culture experiments were repeated three times. One-way ANOVA was performed using SigmaStat 3.5 software, and a Tukey test was used to estimate the significance. Statistical significance was accepted when P < 0.05.

The levels of miR-200b were significantly increased in WT embryos exposed to maternal diabetes compared with WT embryos from nondiabetic dams (Fig. 1A). SOD1-overexpressing embryos had significantly lower levels of miR-200b than those in their WT littermates (Fig. 1A). For ascertainment of whether high glucose of maternal diabetes is the primary factor that stimulates miR-200b expression, embryos from insulin-treated dams were analyzed. Insulin treatment significantly lowered blood glucose levels in diabetic dams from 22.7 ± 1.2 mmol/L to 13.6 ± 0.6 mmol/L glucose and blunted maternal diabetes–increased miR-200b expression (Fig. 1A).

To reveal whether high glucose in vitro has a similar effect on miR-200b expression, neural stem cells were cultured under normal glucose (5 mmol/L glucose) or high glucose (16.6, 25, and 33.3 mmol/L glucose) conditions. High glucose increased miR-200b expression in a dose-dependent manner, and the increased miR-200b expression reached a plateau at 25 mmol/L glucose (Fig. 1B). The 25 mmol/L glucose level was comparable with the high blood glucose levels seen in the diabetic dams (glucose levels 22.7 ± 1.2 mmol/L). In a time course study, 25 mmol/L glucose stimulated miR-200b expression at 12, 24, 48, 72, and 96 h, with the highest increase observed at 24 h (Fig. 1C and Supplementary Fig. 1A). Treatment with the SOD1 mimetic tempol abolished high glucose–increased miR-200b expression (Fig. 1D) Mannitol was used as an osmotic control of glucose. High mannitol concentrations did not affect miR-200b expression (Fig. 1E and F).

A miRNA target prediction algorithm (miRanda) revealed CITED2 as a potential target of miR-200b. There are two predicted binding sites of miR-200b in the 3′-UTR of CITED2 mRNA (Fig. 2A). To test whether CITED2 is a true target of miR-200b, we examined the association of miR-200b with the CITED2 mRNA by an RNA pulldown assay using biotin-labeled miR-200b. To ensure the transfection efficiency of miR-200b, we examined the levels of miR-200b and small nuclear RNA U6 (which served as the control) after 48 h transfection. Cells transfected with the biotin-labeled miR-200b displayed elevated miR-200b levels along with no change in RNA U6 levels (Fig. 2B).

We examined the abundance of CITED2 mRNA in the RNA pulldown assay and found an eleven-fold greater enrichment of CITED2 mRNA in the pulldown beads of cell samples transfected with the biotin-labeled miR-200b compared with cells transfected with biotin-labeled control miRNA (Fig. 2C). Additionally, increasing the levels of miR-200b by biotin-labeled miR-200b transfection did not alter the total levels of CITED2 mRNA (Fig. 2C). Furthermore, the association of miR-200b with CITED2 mRNA was completely inhibited when unlabeled synthetic RNA sense strands (RSS1 and RSS2) against CITED2 3′-UTR (positions from 1514 to 1536 and 1846 to 1867, encompassing the two miR-200b binding sites) were added to the cell culture medium together with biotin-labeled miR-200b (Fig. 2D). Treatment with RSS1 and RSS2 had no effect on the levels of total CITED2 mRNA (Fig. 2D). These results strongly suggest that miR-200b directly interacts with the CITED2 mRNA and forms the miR-200b/CITED2 mRNA complex.

For investigation of the functionality of the miR-200b/CITED2 mRNA complex, luciferase reporter constructs were used to examine whether miR-200b regulates CITED2 expression. miRNAs repress translation and/or degrade mRNA by binding to seed site sequences located within the 3′-UTR of mRNA (24). The CR, 3′-UTR, or the franking regions (F1, F2) of the miR-200b binding sites in CITED2 mRNA were subcloned into the pmirGLO dual-luciferase miRNA target expression vector to generate CR-Luc, UTR-Luc, F1-Luc, and F2-Luc reporter constructs (Fig. 3A). Ectopic miR-200b overexpression decreased the 3′-UTR luciferase reporter activity (Fig. 3A), but it failed to inhibit the CR reporter activity (Fig. 3A), indicating that miR-200b represses CITED2 mRNA translation through interaction with the CITED2 3′-UTR rather than with its CR. Moreover, the luciferase activities of the F1-Luc and F2-Luc constructs were significantly inhibited by the miR-200b mimic (Fig. 3A).

To further characterize the specific miR-200b binding sites in the CITED2 3′-UTR, deletion mutations of the miR-200b binding sites in F1 or F2 fragments of the CITED2 3′-UTR were also performed, in which the nucleotides spanning positions 1514–1536 or 1846–1867 of the CITED2 3′-UTR were eliminated (Fig. 3B). The mutant constructs of F1 and F2 lacking the miR-200b binding sites were refractory to the decrease in luciferase activities by miR-200b (Fig. 3B).

The repression of CITED2 expression by miR-200b was experimentally further validated. CITED2 protein expression was diminished by 37.5 or 75 nmol/L miR-200b mimic (Fig. 3C). On the other hand, CITED2 protein levels increased accordingly when cells were transfected with 37.5 or 75 nmol/L miR-200 inhibitor (Fig. 3D). When miR-200b levels were increased remarkably by transfection with the miR-200b mimic (Fig. 3E), CITED2 mRNA levels were significantly decreased (Fig. 3F). The miR-200b inhibitor suppressed endogenous miR-200b levels and, thus, increased CITED2 mRNA levels (Fig. 3G and H).

Altogether, these results indicate that miR-200b interacts with CITED2 mRNA via the specific binding sites located in its 3′-UTR, thus silencing CITED2 expression by degrading mRNA and blocking translation.

For determination of whether high glucose affects CITED2 expression, C17.2 cells were treated with 25 mmol/L glucose for 12, 24, 48, and 72 h. High glucose downregulated CITED2 at both protein and mRNA levels (Fig. 4A and B and Supplementary Fig. 1B). In contrast, the glucose osmotic control, mannitol, did not affect CITED2 expression (Fig. 4A and B). Consistent with these in vitro findings, CITED2 protein and mRNA levels were significantly decreased in WT embryos exposed to maternal diabetes compared with WT embryos from nondiabetic dams (Fig. 4C and D). Under maternal diabetic conditions, SOD1-overexpressing embryos had significantly higher levels of CITED2 protein and mRNA than those in their WT littermates (Fig. 4C and D). Moreover, the SOD1 mimetic tempol abolished high glucose–induced CITED2 reduction at both protein (Fig. 4E and Supplementary Fig. 2) and mRNA (Fig. 4F) levels.

To explore whether miR-200b mediates the inhibitory effect of high glucose on CITED2 expression, cells were transfected with the miR-200b mimic or inhibitor under normal glucose (5 mmol/L) or high glucose (25 mmol/L) conditions. The miR-200b mimic as well as high glucose inhibited CITED2 expression (Fig. 4G). The miR-200b inhibitor abrogated high glucose–suppressed CITED2 protein expression (Fig. 4H).

Maternal diabetes induces ER stress in the developing embryos, which causes apoptosis in the neural tube leading to NTDs (5). For testing of whether miR-200b induces UPR signaling and ER stress, the phosphorylated levels of the two major UPR sensors, IRE1α and PERK, were assessed. Protein levels of phosphorylated (p)-IRE1α, p-PERK, and p-eIF2α were significantly increased by the miR-200b mimic (Fig. 5A and B). The ER stress marker CHOP was significantly higher in cells transfected with the miR-200b mimic compared with the control mimic (Fig. 5A and B). Additionally, UPR activation and ER stress markers induced by high glucose were abrogated by the miR-200b inhibitor (Fig. 5C and D). Taken together, these data indicate that miR-200b is involved in the induction of UPR and ER stress by high glucose.

Because miR-200b overexpression induces UPR and ER stress and because CITED2 is a downstream effector of miR-200b, we tested whether reduced CITED2 expression promotes UPR and ER stress. To this end, CITED2 protein and mRNA were transiently inhibited by siRNA. Treatment with 35 nmol/L CITED2 siRNA reduced CITED2 protein and mRNA levels to ∼75% and 50% of those in controls, respectively (Fig. 6A and B). When CITED2 was silenced, the phosphorylation of IRE1α, PERK, and eIF2α and the expression of CHOP were upregulated (Fig. 6C).

Because both increasing miR-200b and decreasing CITED2 activate UPR and ER stress, we further investigated whether both high glucose and miR-200b induce UPR and ER stress by inhibiting CITED2. Cells were cotransfected with the CITED2 plasmid and the miR-200b mimic. As expected, phosphorylation of IRE1α, PERK, eIF2α, and the ER stress marker CHOP were significantly increased, and CITED2 was decreased by the miR-200b mimic (Fig. 7A and B). The induction of UPR signaling and ER stress by the miR-200b mimic was attenuated when CITED2 expression, which was suppressed by the miR-200b mimic, was restored by the CITED2 plasmid (Fig. 7A and B).

CITED2 overexpression prevented UPR and ER stress induced by miR-200b, which mediates a pro-ER stress effect of high glucose. Next, we tested whether CITED2 inhibits high glucose–induced UPR and ER stress. Indeed, CITED2 overexpression blocked high glucose–triggered phosphorylation of IRE1α, PERK, eIF2α, and CHOP upregulation (Fig. 7C and D).

To define the biological consequence of high glucose–induced miR-200b upregulation and consequent CITED2 downregulation, we assessed the number of apoptotic cells and caspase cleavage. When cells were exposed to high glucose, apoptotic cells were robustly present, whereas the miR-200b inhibitor protected cells from undergoing high glucose–induced apoptosis (Fig. 8A and B). Likewise, CITED2 overexpression significantly reduced high glucose–increased TUNEL-positive cell numbers (Fig. 8C and D). Moreover, the ER stress inhibitor 4-phenylbutyric acid (4-PBA) blocked CITED2 knockdown-induced apoptosis (Fig. 8E and F). High glucose increased the abundance of cleaved caspase 3, and both the miR-200b inhibitor and CITED2 overexpression blocked high glucose–induced caspase cleavage (Fig. 8G and H). 4-PBA treatment prevented CITED2 knockdown–induced caspase 3 cleavage (Fig. 8I).

We previously found that the ER stress inhibitor, 4-PBA, blocked high glucose–induced apoptosis (5). Together with our previous findings, our data suggest that the miR-200b–CITED2 circuit mediates the proapoptotic effect of high glucose by activating UPR and ER stress.

To test whether the miR-200b inhibitor has any effect on high glucose–induced NTDs, we used the ex vivo whole-embryo culture system as previously described (5). As observed in cultured cells (Fig. 3G), the miR-200b inhibitor suppressed endogenous miR-200b expression in cultured embryos (Fig. 8J) and suppressed high glucose–induced miR-200b upregulation (Fig. 8J). Consistent with the blockade of high glucose–induced miR-200b expression, the miR-200b inhibitor reversed the downregulation of CITED2 mRNA and protein expression (Fig. 8K and L). Furthermore, the miR-200b inhibitor suppressed high glucose–induced ER stress by decreasing ER stress markers, immunoglobulin heavy-chain-binding protein (BiP), and CHOP (Fig. 8K and L), leading to amelioration of NTDs (Fig. 8M).

Hydrogen peroxide, one of the biological reactive oxygen species, increases miR-200b expression (34). High glucose enhances reactive oxygen species production and simultaneously impairs endogenous antioxidant capacity, leading to oxidative stress (35–38). Overexpression of SOD1, an antioxidant enzyme, in SOD1-Tg mice blocks maternal diabetes–induced oxidative stress (39–41). We found that SOD1 abrogated miR-200b upregulation induced by maternal diabetes. Furthermore, we showed that high glucose–increased miR-200b was blunted in cells treated with the SOD1 mimetic, tempol. Consistent with a prior report (34), our findings support the hypothesis that oxidative stress mediates the stimulative effect of high glucose on miR-200b expression.

The current study revealed CITED2 as a new target of miR-200b. The miR-200b mimic suppressed CITED2 expression, whereas a miR-200b inhibitor reversed high glucose–repressed CITED2 expression. Biotin-labeled miR-200b directly binds to CITED2 mRNA. We identified two miR-200b binding sites in the CITED2 3′-UTR. Previous global gene expression studies using DNA microarray (10) could not reveal miRNA expression. For protein coding genes, consistent with our finding, a gene profiling study in neurulation stage embryos (10) revealed that maternal diabetes suppressed CITED2 expression. Our study further uncovers the mechanism underlying CITED2 downregulation that is due to miR-200b upregulation. CITED2 deletion causes excessive apoptosis in the developing neuroepithelium leading to NTDs (29,42). Consistent with the findings in the CITED2 knockout mouse model (29,42), we have found that silencing CITED2 triggers apoptosis. The findings in our study support a possible mechanism for CITED2 deficiency–induced apoptosis: CITED2 deficiency seems to promote ER stress, which causes apoptosis. This is in agreement with our previous findings that ER stress mediates the proapoptotic effects of high glucose (5). Thus, during normal development, CITED2 may protect the embryos against ER stress caused by environmental perturbations or stress conditions. Our ex vivo experiments demonstrated that miR-200b inhibition reversed high glucose–repressed CITED2 and, thereby, ameliorated ER stress and NTDs in the developing embryo. Although these ex vivo findings support the critical involvement of the miR-200b–CITED2–ER stress pathway in high glucose–induced NTDs, one shortcoming of our study is lack of in vivo tests of this proposed pathway in diabetes-induced NTDs.

An intriguing question is whether miRNA dysregulation can directly cause disruption of ER homeostasis. Indeed, the results in the current study support this hypothesis that miRNA dysregulation causes UPR activation and ER stress. Increased miR-200b and decreased CITED2 act upstream of ER imbalance and UPR activation because both ectopic miR-200b overexpression and CITED2 silencing activate UPR leading to ER stress. However, it is still unclear how the miR-200b–CITED2 circuit causes UPR and ER stress. One possibility is that CITED2 is required for gene expression essential for ER homeostasis. CITED2 interacts with a group of transcription factors and transcription coactivators including PGC-1α (43). CITED2 enhances PGC-1α activity (44), which resolves ER homeostatic imbalance through activating transcription factor 6 (45).

The C17.2 cells resemble cells in the developing neuroepithelium in responding to high glucose in inducing ER stress and apoptosis (46,47). Our experiments that were conducted in C17.2 cells are in vitro in nature, and the 17.2 cell line is currently among the best in vitro model in research of diabetes-induced NTDs. Although factors in early embryonic development may contribute to late NTD formation, the key period for mouse neurulation is from E8.0 to E10 (total 48 h) (3). Nevertheless, high glucose mimics maternal diabetes to increase miR-200b expression and decrease CITED2 expression in C17.2 cells even after 72 h and 96 h. Thus, these evidences collectively support the C17.2 cell line as a suitable model in our studies.

We recently discovered the apoptosis signal-regulating kinase 1 (ASK1)–forkhead transcription factor 3a (FoxO3a)–caspase 8 pathway as the major teratogenic pathway (3). FoxO3a functions as a pleiotropic regulator of apoptosis. FoxO3a downregulation decreases miRNA expression in cancer cells (48), suggesting that FoxO3a stimulates miRNA expression. Future studies may reveal whether FoxO3a mediates the stimulatory effect of maternal diabetes on miR-200b expression. The miR-200b–CITED2 pathway induces ER stress, which causes caspase 8 activation (46,49). Therefore, the ASK1–FoxO3a–caspase 8 pathway and the miR-200b–CITED2 pathway may converge on caspase 8 activation.

In summary, we have found that hyperglycemia-induced oxidative stress increases miR-200b expression leading to CITED2 repression. The miR-200b–CITED2 circuit promotes UPR and ER stress and, thus, mediates the proapoptotic effect of high glucose. We, therefore, have revealed a new link between high glucose–induced ER stress and the miR-200b–CITED2 circuit, which may play an important role in diabetic embryopathy.

Acknowledgments. The authors are grateful for the support from the Office of Dietary Supplements, National Institutes of Health. The authors thank Dr. Julie Wu at the University of Maryland School of Medicine for critical reading and editing assistance.

Funding. This study was supported by National Institutes of Health R01-DK-083243, R01-DK-101972, R56-DK-095380, and R01-DK-103024 and the American Diabetes Association Basic Science Award (1-13-BS-220).

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. H.G., J.Y., and D.D. researched data. Q.Z., J.-Y.W., and S.F. participated in data analyses and experimental design. P.Y. conceived the project, designed the experiments, and wrote the manuscript. All authors approved the final version of the manuscript. P.Y. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.