The green tea polyphenol EGCG alleviates maternal diabetes–induced neural tube defects by inhibiting DNA hypermethylation

Abstract

BACKGROUND

Maternal diabetes increases the risk of neural tube defects in offspring. Our previous study demonstrated that the green tea polyphenol, Epigallocatechin gallate, inhibits high glucose-induced neural tube defects in cultured embryos. However, the therapeutic effect of Epigallocatechin gallate on maternal diabetes–induced neural tube defects is still unclear.

OBJECTIVE

We aimed to examine whether Epigallocatechin gallate treatment can reduce maternal diabetes–induced DNA methylation and neural tube defects.

STUDY DESIGN

Nondiabetic and diabetic pregnant mice at embryonic day 5.5 were given drinking water with or without 1 or 10 μM Epigallocatechin gallate. At embryonic day 8.75, embryos were dissected from the visceral yolk sac for the measurement of the levels and activity of DNA methyltransferases, the levels of global DNA methylation, and methylation in the CpG islands of neural tube closure essential gene promoters. embryonic day 10.5 embryos were examined for neural tube defect incidence.

RESULTS

Epigallocatechin gallate treatment did not affect embryonic development because embryos from nondiabetic dams treated with Epigallocatechin gallate did not exhibit any neural tube defects. Treatment with 1 μM Epigallocatechin gallate did not reduce maternal diabetes–induced neural tube defects significantly. Embryos from diabetic dams treated with 10 μM Epigallocatechin gallate had a significantly lower neural tube defect incidence compared with that of embryos without Epigallocatechin gallate treatment. Epigallocatechin gallate reduced neural tube defect rates from 29.5% to 2%, an incidence that is comparable with that of embryos from nondiabetic dams. Ten micromoles of Epigallocatechin gallate treatment blocked maternal diabetes–increased DNA methyltransferases 3a and 3b expression and their activities, leading to the suppression of global DNA hypermethylation. Additionally, 10 μM Epigallocatechin gallate abrogated maternal diabetes–increased DNA methylation in the CpG islands of neural tube closure essential genes, including Grhl3, Pax3, and Tulp3.

CONCLUSION

Epigallocatechin gallate reduces maternal diabetes–induced neural tube defects formation and blocks the enhanced expression and activity of DNA methyltransferases, leading to the suppression of DNA hypermethylation and the restoration of neural tube closure essential gene expression. These observations suggest that Epigallocatechin gallate supplements could mitigate the teratogenic effects of hyperglycemia on the developing embryo and prevent diabetes–induced neural tube defects.

Currently nearly 60 million women of reproductive age (18–44 years old) worldwide have diabetes, and this number has been estimated to double by 2030.^1–4 Clinical studies and animal model investigations have revealed that maternal diabetes increases the risk of neural tube defects in offspring and that hyperglycemia is a teratogen.^1–3,5–10

Although strict glycemic control by lifestyle and pharmacological treatment can decrease the incidence of hyperglycemia-induced embryonic malformations in pregnancies affected by preexisting maternal diabetes,^1–3,6,7 euglycemia is difficult to achieve and maintain, and even transient exposure to high glucose could lead to abnormal embryonic development.^{1–3,11–13} Thus, diabetes-induced birth defects are significant public health problems, and there is an urgent need for new therapeutic approaches against diabetic embryopathy.

Neural tube defects are common complex congenital malformations of the central nervous system that form during embryogenesis.¹⁴ There are approximately 5 times more neural tube defects in offspring from diabetic mothers than in those from nondiabetic mothers, despite modern preconception care.¹⁵ Studies from our group^{1,5,6,16–25} and others²⁶ have demonstrated that maternal diabetes induces cellular stress, including oxidative stress and endoplasmic reticulum stress, and that those cellular stresses cause apoptosis in the embryonic neural tissue, leading to neural tube defect formation. Recently several studies have suggested that altered DNA methylation disrupts the folate metabolic pathway and causes neural tube defects.^27–29 Therefore, we hypothesize that altered DNA methylation is involved in neural tube defect formation in diabetic pregnancies.

Our previous studies have revealed that naturally occurring polyphenols exert protective effects against high glucose–induced neural tube defects in vitro.^30,31 Epigallocatechin gallate is the major polyphenol in green tea (Camellia sinensis) and makes up to approximately 30% of the solids in green tea.³²

Epigallocatechin gallate is the subject of increasing research interest because it has demonstrated beneficial effects in studies of diabetes, Parkinson’s disease, Alzheimer’s disease, stroke, and obesity.³³ The cancer-preventive effects of Epigallocatechin gallate have been widely reported in epidemiological, cell culture, animal, and clinical studies.³⁴ One of the mechanisms by which Epigallocatechin gallate exerts effects on cancer cells is through the inhibition of DNA methyltransferases and reactivation of DNA methylation–silenced gene expression.³⁵ Thus, in the present study, we investigated whether Epigallocatechin gallate could reduce or prevent neural tube defect formation in embryos from diabetic dams and inhibit maternal diabetes–increased DNA methylation.

Methods and Materials

Animals and reagents

All animal procedures were approved by the University of Maryland School of Medicine Institutional Animal Care and Use Committee. Wild-type C57BL/6J mice were purchased from The Jackson Laboratory (Bar harbor, ME). Streptozotocin from Sigma (St Louis, MO) was dissolved in 0.1 M citrate buffer (pH 4.5). Sustained-release insulin pellets were purchased from Linplant (Linshin, Canada).

The mouse model of diabetic embryopathy

Our mouse model of diabetic embryopathy was described previously.^21,36 Briefly, female mice were intravenously injected daily with 75 mg/kg streptozotocin over 2 days to induce diabetes. Nondiabetic wild-type mice with vehicle injection served as controls. Diabetes was defined as 12 hour fasting blood glucose levels of ≥250 mg/dL, which normally occurred at 3–5 days after streptozotocin injections. Once the level of hyperglycemia indicative of diabetes (≥250 mg/dL) was achieved, insulin pellets were subcutaneously implanted in these diabetic mice to restore euglycemia prior to mating. The mice were then mated with wild-type male mice at 3:00 PM.

We designated that the morning when a vaginal plug was present as embryonic day 0.5. On embryonic day 5.5, insulin pellets were removed to permit frank hyperglycemia (>250 mg/dL glucose level), so the developing conceptuses would be exposed to hyperglycemic conditions. Wild-type, nondiabetic female mice with vehicle injections and sham operation of insulin pellet implants served as nondiabetic controls. On embryonic day 8.75, the mice were euthanized, and conceptuses were dissected out of the uteri for analysis. Embryos were harvested at embryonic day 8.75 for analysis and at embryonic day 10.5 for neural tube defect examination.

At embryonic day 10.5, embryos were examined under a Leica MZ16F stereo-microscope (Leica, Wetzlar, Germany) to identify neural tube defects. Images of embryos were captured by a DFC420 5-megapixel digital camera with software (Leica). Normal embryos were classified as having completely closed neural tube and no evidence of other malformations. Malformed embryos were classified as showing evidence of failed closure of the anterior neural tubes, resulting in exencephaly, a major type of neural tube defect.

Epigallocatechin gallate treatment

Epigallocatechin gallate treatment was performed as described previously.³⁷ Concentrations of either 1 or 10 μM Epigallocatechin gallate (Sigma) were given to wild-type nondiabetic and diabetic pregnant mice at embryonic day 5.5 in drinking water.

Real-time polymerase chain reaction

Using the Trizol (Invitrogen, Carlsbad, CA), messenger RNA was isolated from embryonic day 8.5 embryos and then reversed transcribed using the high-capacity complementary DNA archive kit (Applied Biosystems, Grand Island, NY). Real-time polymerase chain reaction for Dnmt1, Dnmt3a, Dnmt3b, Grhl3 (grainyhead-like-3), Pax3 (paired box gene 3), Tulp3 (tubby-like-3), and β-actin was performed using the Maxima SYBR Green/ROX quantitative polymerase chain reaction master mix assay (Thermo Scientific, Rockford, IL) in the StepOnePlus system (Applied Biosystems, Grand Island, NY). Primer sequences are listed in Table 1.

TABLE 1.

Primers used for RT-PCR

Primer name	Primer sequences
Dnmt1	Forward primer, 5′- AAGAATGGTGTTGTCTACCGAC -3′
	Reverse primer, 5′- CATCCAGGTTGCTCCCCTTG -3′
Dnmt3a	Forward primer, 5′- GATGAGCCTGAGTATGAGGATGG -3′
	Reverse primer, 5′- CAAGACACAATTCGGCCTGG -3′
Dnmt3b	Forward primer, 5′- CGTTAATGGGAACTTCAGTGACC -3′
	Reverse primer, 5′- GGGAGCATCCTTCGTGTCTG -3′
Grhl3	Forward primer, 5′- CCCGGCAAGACCAATACCG -3′
	Reverse primer, 5′- AACCCCATGAATGCTCTCAAAT -3′
Pax3	Forward primer, 5′- TTTCACCTCAGGTAATGGGACT -3′
	Reverse primer, 5′- GAACGTCCAAGGCTTACTTTGT -3′
Tulp3	Forward primer, 5′- CCAAAAACACGGCATCTTGAG -3′
	Reverse primer, 5′- GGGCTATACGCAAAGTCCTCTAA -3′
β-Actin	Forward primer, 5′- GTGACGTTGACATCCGTAAAGA -3′
	Reverse primer, 5′- GCCGGACTCATCGTACTCC -3′

RT-PCR, Real-time polymerase chain reaction.

Zhong et al. The green tea polyphenol EGCG prevents diabetic embryopathy. Am J Obstet Gynecol 2016.

Western blotting

Western blotting was performed as described previously.²¹ To extract proteins, embryos were sonicated in ice-cold lysis buffer (Cell Signaling Technology, Beverly, MA) with protease inhibitor cocktail (Sigma). Equal amounts of protein from different experimental groups and the Precision Plus protein standards (Bio-Rad Laboratories, Hercules, CA) were resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, transferred onto polyvinyl difluoride membranes, and then immunoblotted by primary antibodies at 1:1000 dilutions in 5% nonfat milk. Antibodies to protein DNA methyltransferase-1, DNA methyltransferase-3a, and DNA methyltransferase-3b were purchased from Cell Signaling Technology.

Horseradish peroxidase–conjugated goat antirabbit, goat antimouse (Jackson ImmunoResearch Laboratories, West Grove, PA) or goat antirat (Chemicon, Temecula, CA) secondary antibodies at 1:1000 were used. The intensity of the target protein bands were identified by densitometry and normalized by the densities of β-actin (Abcam, Cambridge, United Kingdom). Signals were detected by SuperSignal West Femto maximum sensitivity substrate kit (Thermo Scientific), and chemiluminescence emitted from bands was captured by a UVP Bioimage EC3 system (Upland, CA). All experiments were repeated 3 times with the use of independently prepared tissue lysates.

Measurement of DNA methyltransferase activities

DNA methyltransferase activities were measured by an EpiQuik DNA methyltransferase activity/inhibition assay kit (Epigentek, Farmingdale, NY) according to the manufacturer’s instructions. Briefly, we first extracted nuclear from embryonic day 8.5 embryos and then incubated nuclear with substrate and assay buffer for 1 hour, later added capture antibody for wash, after the wash added detection antibody, and finally added fluorodeveloping solution for fluorescence development and measurement.

Detection of global DNA methylation level

The global DNA methylation level was detected by a MethylFlash methylated DNA quantification kit (Colorimetric; Epigentek, Farmingdale, NY) according to the manufacturer’s instructions. In brief, we extracted genomic DNA from embryonic day 8.5 embryos and then bound DNA to the assay well, washed the well and added the capture antibody, washed the well again and added detection antibody and enhancer solution, and finally added color developing solution for color development and measurement.

Methylation-specific PCR

DNA methylation patterns in the CpG islands of Grhl3, Pax3, and Tulp3 genes were determined by methylation-specific polymerase chain reaction.³⁸ Methylation-specific polymerase chain reaction distinguishes unmethylated from methylated alleles in a given gene based on sequence changes produced after bisulfite treatment of DNA, which converts unmethylated cytosines to uracil, and subsequent polymerase chain reaction using primers designed for either methylated or unmethylated DNA. Polymerase chain reaction was performed with 3.0 μL of bisulfite-modified DNA template in a 25 μL reaction. The CpG islands were identified on the Li Lab website (www.urogene.org). Primer sequences are listed in Table 2.

TABLE 2.

Primers used for MSP

Primers name	Primer sequences
Grhl3	Left M primer, 5′- TTAAAGCGTAACGTAGAGTAAACGT -3′
	Right M primer, 5′- ACCTCGATATACTAAAAAAACCGAA -3′
	Left U primer, 5′- TTTAAAGTGTAATGTAGAGTAAATGT -3′
	Right U primer, 5′- ACCTCAATATACTAAAAAAACCAAA -3′
Pax3	Left M primer, 5′- GTATTGTGTTCGTTTTTTCGTTTC -3′
	Right M primer, 5′- GCTACGTAAATAATTCTACCCCGA -3′
	Left U primer, 5′- TTGTGTTTGTTTTTTTGTTTTGTTT -3′
	Right U primer, 5′- ACTACATAAATAATTCTACCCCAAAC -3′
Tulp3	Left M primer, 5′- TTTTCGATTTTTTTATTTGTAATGC -3′
	Right M primer, 5′- CAACTCAATTCTAATCCTACTCGTA -3′
	Left U primer, 5′- TTTGATTTTTTTATTTGTAATGTGT -3′
	Right U primer, 5′- CCAACTCAATTCTAATCCTACTCATA -3′

M, methylated; MSP, methylation-specific polymerase chain reaction; U, unmethylated.

Zhong et al. The green tea polyphenol EGCG prevents diabetic embryopathy. Am J Obstet Gynecol 2016.

Statistical analysis

Sample sizes were preestimated based on our previous studies¹⁷ before experiments were performed. Data on neural tube defect rates of each experimental group were analyzed by Fisher exact test or a χ² test. Data on protein and messenger RNA expression were presented as means ± SE. A one-way analysis of variance was performed using the SigmaStat 3.5 software (Systat Software Inc, San Jose, CA) followed by a Tukey test to estimate the significance of results (P <.05).

Results

Epigallocatechin gallate ameliorates maternal diabetes–induced neural tube defects

Mouse embryonic neurulation occurs during embryonic day 8.5–10.5.³⁹ To determine whether Epigallocatechin gallate treatment could ameliorate maternal diabetes–induced neural tube defects, concentrations of either 1 μM or 10 μM Epigallocatechin gallate were given to wild-type nondiabetic and diabetic pregnant mice at embryonic day 5.5 in drinking water.

The neural tube defect rate in embryos from diabetic dams was significantly higher than that in embryos from nondiabetic dams, with or without Epigallocatechin gallate treatment (Table 3). Treatment with 10 μM Epigallocatechin gallate dramatically decreased neural tube defect formation in embryos from diabetic dams, compared with untreated diabetic dams (Table 3). Treatment with 1 μM Epigallocatechin gallate did not reduce maternal diabetes–induced neural tube defects significantly. Therefore, 10 μM Epigallocatechin gallate was used in subsequent experiments.

TABLE 3.

EGCG treatment ameliorates maternal diabetes-induced neural tube defects

Group	Total embryos	Embryos with NTD	Blood glucose levels	NTD rate, %
Nondiabetic, no EGCG (n = 12)	85	0	167.71 ± 7.1	0
Nondiabetic, 1 μM EGCG (n = 4)	27	0	157.3 ± 8.2	0
Nondiabetic, 10 μM EGCG (n = 4)	24	0	152.5 ± 11.3	0
Diabetic, no EGCG (n = 12)	78	23	442.2 ± 14.7	29.5a
Diabetic, 1 μM EGCG (n = 7)	46	11	417.5 ± 18.4	23.9a
Diabetic, 10 μM EGCG (n = 8)	51	1	406.0 ± 20.1	2.0

EGCG, Epigallocatechin gallate; NTD, neural tube defect.

^aSignificant difference compared with the other groups in χ² test (P < .05). The diabetic no Epigallocatechin gallate group is significantly different when compared with the nondiabetic no Epigallocatechin gallate, nondiabetic 1 μM Epigallocatechin gallate, nondiabetic 10 μM Epigallocatechin gallate, and diabetic 10 μM Epigallocatechin gallate groups. The diabetic 1 μM Epigallocatechin gallate group and the diabetic no Epigallocatechin gallate group are not significantly different, and the diabetic 1 μM Epigallocatechin gallate group is significantly different when compared with the nondiabetic no Epigallocatechin gallate, nondiabetic 1 μM Epigallocatechin gallate, nondiabetic 10 μM Epigallocatechin gallate, and diabetic 10 μM Epigallocatechin gallate groups.

Zhong et al. The green tea polyphenol EGCG prevents diabetic embryopathy. Am J Obstet Gynecol 2016.

Epigallocatechin gallate inhibits maternal diabetes–increased DNA methyltransferase expression

To assess whether Epigallocatechin gallate treatment prevents diabetes–induced neural tube defects by regulating embryonic DNA methylation level, we first examined the DNA methyltransferase expression level in embryos from diabetic dams, with or without Epigallocatechin gallate treatment. Messenger RNA expression of Dnmt1 and Dnmt3a did not differ in embryos from both the nondiabetic and diabetic groups, regardless of Epigallocatechin gallate treatment (Figure 1A). However, Dnmt3b messenger RNA expression was significantly higher in embryos from the diabetic group than in embryos from the nondiabetic group, with or without Epigallocatechin gallate treatment. We also observed that Epigallocatechin gallate treatment blocked maternal diabetes–induced Dnmt3b expression (Figure 1A).

FIGURE 1.

Epigallocatechin gallate treatment blocks maternal diabetes–increased DNA methyltransferase expression

A, Messenger RNA levels of Dnmt1, Dnmt3a, and Dnmt3b in E8.75 embryos from nondiabetic and diabetic dams with or without Epigallocatechin gallate treatment. B, Protein levels of DNA methyltransferase-1, DNA methyltransferase-3a, and DNA methyltransferase-3b in E8.75 embryos from nondiabetic and diabetic dams with or without Epigallocatechin gallate treatment. Experiments were performed using 3 embryos from 3 different dams per group (n = 3). Asterisk indicates significant difference compared with the other groups (P < .05).

DM, diabetic mellitus; E, embryonic day; EGCG, Epigallocatechin gallate; NC, nondiabetic control; NC-EGCG, nondiabetic control–Epigallocatechin gallate; DM-EGCG, diabetic mellitus–Epigallocatechin gallate.

Zhong et al. The green tea polyphenol EGCG prevents diabetic embryopathy. Am J Obstet Gynecol 2016.

We next examined the DNA methyltransferaseprotein expression by immunoblotting. The levels of all 3 DNA methyltransferases, DNA methyltransferase-1, DNA methyl transferase-3a, and DNA methyl transferase-3b, were significantly higher in embryos from the diabetic group than that in embryos from the nondiabetic group. Treatment with 10 μM Epigallocatechin gallate suppressed maternal diabetes–increased DNA methyltransferase protein expression, whereas Epigallocatechin gallate treatment did not further suppress DNA methyltransferase protein expression in embryos from the nondiabetic group (Figure 1B). These data suggested that Epigallocatechin gallate treatment inhibits maternal diabetes–increased DNA methyltransferase expression.

Epigallocatechin gallate reduces maternal diabetes–increased DNA methyltransferase activity and global methylation levels

Total DNA methyltransferase activity was determined in embryos from diabetic dams vs nondiabetic dams. DNA methyltransferase activity was dramatically increased in embryos from diabetic dams compared with those in embryos from nondiabetic dams, with or without Epigallocatechin gallate treatment. However, the DNA methyltransferase activity in embryos from diabetic dams treated with 10 μM Epigallocatechin gallate was similar to those seen in embryos from nondiabetic dams, with or without Epigallocatechin gallate treatment, and was significantly lower than that in the diabetic group without Epigallocatechin gallate treatment (Figure 2A).

FIGURE 2.

EGCG inhibits diabetes–increased DNA methyltransferase activity and global DNA methylation

A, Total DNA methyltransferase activity was tested in E8.75 embryos from nondiabetic and diabetic dams with or without Epigallocatechin gallate treatment. B, Global DNA methylation levels were determined in E8.75 embryos from nondiabetic and diabetic dams with or without Epigallocatechin gallate treatment. Experiments were performed using 3 embryos from 3 different dams per group. Asterisk indicates significant differences compared with the other groups (P < .05).

DM, diabetic mellitus; E, embryonic day; EGCG, Epigallocatechin gallate; NC, nondiabetic control; NC-EGCG, nondiabetic control–Epigallocatechin gallate; DM-EGCG, diabetic mellitus-Epigallocatechin gallate.

Zhong et al. The green tea polyphenol EGCG prevents diabetic embryopathy. Am J Obstet Gynecol 2016.

We tested whether Epigallocatechin gallate treatment could affect global DNA methylation levels in embryos from diabetic dams. Consistent with our observation of DNA methyltransferase activity, maternal diabetes increased global DNA methylation levels in embryos from the diabetic group, compared with the nondiabetic group, and Epigallocatechin gallate treatment blocked maternal diabetes–increased global DNA methylation levels (Figure 2B).

Epigallocatechin gallate decreases methylation in the CpG islands of neural tube closure essential genes

Hypermethylation of CpG islands in the promoter regions of genes involved in embryogenesis is an important mechanism to silence gene expression.³⁵ To determine whether maternal diabetes induces hypermethylation of CpG islands in the promoters of neural tube closure essential genes, we measured DNA methylation in the CpG islands of several neural tube closure essential genes, including Grhl3, Pax3, and Tulp3. CpG islands were identified the promoter regions of the Grhl3, Pax3, and Tulp3 gene (Figure 3A). We defined a CpG island using the following criteria: size >200 bp, guanine-cytosine percentage >50%, CpG observed/CpG expected >0.6.⁴⁰

FIGURE 3.

EGCG reduces DNA hypermethylation in gene promoter CpG islands

A, CpG islands (blue) in the promoter regions of neural tube closure–essential genes, including Grhl3, Pax3, and Tulp3. B, Methylation levels in CpG islands of Grhl3, Pax3, and Tulp3 were detected in the E8.75 embryos from nondiabetic and diabetic dams with or without Epigallocatechin gallate treatment. C, Messenger RNA levels of Grhl3, Pax3, and Tulp3 were determined in the E8.75 embryos from nondiabetic and diabetic dams with or without Epigallocatechin gallate treatment. Experiments were performed using 3 embryos from 3 different dams per group. Asterisk indicates significant differences compared with the other groups (P < .05).

DM, diabetic mellitus; DM-EGCG, diabetic mellitus-Epigallocatechin gallate; E, embryonic day; EGCG, Epigallocatechin gallate; MSP, methylation-specific primer; NC, nondiabetic control; NC-EGCG, nondiabetic control–Epigallocatechin gallate; NSP, nonmethylation-specific primer.

Zhong et al. The green tea polyphenol EGCG prevents diabetic embryopathy. Am J Obstet Gynecol 2016.

Through methylation-specific polymerase chain reaction, we observed that DNA methylation levels in the CpG islands of Grhl3, Pax3, and Tulp3 were increased in embryos from diabetic dams (Figure 3B). Epigallocatechin gallate treatment abrogated maternal diabetes–increased methylation in CpG islands of these 3 genes (Figure 3B).

In addition, we measured the messenger RNA levels of Grhl3, Pax3, and Tulp3. Consistent with increased DNA methylation in the promoters of these 3 genes, the messenger RNA levels of these genes were significantly down-regulated by maternal diabetes, and treatment with 10 μM Epigallocatechin gallate reversed maternal diabetes–suppressed Grhl3, Pax3, and Tulp3 expression (Figure 3C).

Comment

Although it is well documented that maternal diabetes induces neural tube defects in offspring^1–3,6 and it has been reported that altered DNA methylation is involved in neural tube defect formation,^27–29,35 the relationship between maternal diabetes–induced neural tube defects and DNA methylation remains unclear. Here we demonstrated that neural tube defects were reduced in association with reduced DNA methylation.

Yet the causal relationship between Epigallocatechin gallate–suppressed DNA methylation and Epigallocatechin gallate–reduced neural tube defects was not established. Future studies on a dose-response effect of Epigallocatechin gallate on both DNA methylation of the NT closure genes and neural tube defect formation is essential for the establishment of this causal relationship.

Epigallocatechin gallate is the major polyphenol in green tea and has beneficial effects in preventing the negative effects of cancer, diabetes, and Parkinson’s disease, among other conditions.³³ Previous studies have indicated that Epigallocatechin gallate inhibits DNA methyltransferase activity, causes CpG island hypomethylation, and reactivates hypermethylation-silenced genes.³⁵

Here we revealed that Epigallocatechin gallate treatment blocks hypermethylation in promoters of neural tube closure–essential genes during embryogenesis and ameliorates diabetes-induced neural tube defects. DNA hypermethylation is a critical epigenetic mechanism for the silencing of many genes, including those essential for neural tube closure.^28,41,42 DNA hypermethylation of CpG islands in the promoters of active genes is a mechanism for locking the chromatin in a repressed state.^43,44

In the present study, we showed that DNA hypermethylation of the CpG islands of several neural tube closure–essential genes, including Grhl3, Pax3, and Tulp3, occurs in response to maternal diabetes. Grhl3 is required for neural tube closure, and its null mutants exhibit neural tube defects similar to those observed in diabetic embryopathy.⁴⁵ Loss of Pax3 also results in neural tube defects by having a negative impact on the folate metabolic pathway.^46,47 Tulp3-null embryos are manifested in neural tube defects through the induction of excessive neuroepithelial cellapoptosis.⁴⁸

The expression of these 3 genes is inhibited by maternal diabetes in neurulation-stage embryos; however, the underlying mechanism of the inhibition of these 3 gene is unknown. The present study demonstrates that DNA hypermethylation in the CpG islands of the promoters of these 3 genes is responsible for their down-regulation in diabetic embryopathy.

Because Epigallocatechin gallate is a DNA methylation inhibitor and reduces high glucose–induced neural tube defect formation in vitro,⁴⁹ we tested 1 and 10 μM Epigallocatechin gallate to determine which dose might have a protective effect against maternal diabetes–induced neural tube defect formation. Both doses of Epigallocatechin gallate reduced neural tube defect incidence in embryos from diabetic dams; however, only 10 μM Epigallocatechin gallate significantly decreased the risk of neural tube defects.

In the present study, we demonstrated that Epigallocatechin gallate treatment blocks maternal diabetes–induced neural tube defects through the inhibition of DNA hypermethylation of the promoters of neural tube closure–essential genes. Future studies that explore Epigallocatechin gallate as a therapeutic intervention against maternal diabetes–induced neural tube defects will need to carefully examine any potential toxicities high-dose Epigallocatechin gallate may have to the developing embryo. However, we did not observe any Epigallocatechin gallate toxicities to the developing embryo.

Epigallocatechin gallate has been used as a dietary supplement for humans. As a stand-alone supplement, Epigallocatechin gallate is inexpensive and comes in a capsule form that can be taken orally.⁵⁰ Epigallocatechin gallate capsules (200 mg) taken daily for 12 weeks in patients with human papilloma virus–infected cervical lesions are safe and effective.⁵¹ Epigallocatechin gallate has also been shown to be safe and effective in other human diseases.³³ Human studies have demonstrated that high Epigallocatechin gallate doses, equivalent to the doses we used, exert beneficial effects.⁵² The Epigallocatechin gallate dosing typically available in current pharmacopoeia⁵² are in a close range of the doses used in the current study and forms the basis for future human clinical trials.

In our previous and present studies, we have demonstrated that maternal diabetes–increased DNA methylation and that maternal diabetes–induced oxidative stress are involved in neural tube defect formation.^1–3,6,53 Understanding how to diminish DNA hypermethylation and oxidative stress in embryos from diabetic dams may be critical to reducing the risk of neural tube defects in the offspring of diabetic mothers.

DNA hypermethylation is involved in many pathological processes during embryogenesis, including adverse fetal programming, systemic sclerosis, systemic inflammatory syndrome, autism spectrum disorders, gestational diabetes, offspring obesity, fetal telomere length, and stem cell epigenetics.^54–62 Similarly, oxidative stress is the pathogenesis of a variety of adverse pregnancy outcomes including preeclampsia, fetal alcohol syndrome, fetal brain inflammation, and maternal inflammation–induced offspring cerebral injury,^61,63–69 and the effects of antioxidant treatments have been the subjective of intensive investigations.^70–72

Recent studies have demonstrated that alterations of epigenetic modifications, including DNA methylation and histone acetylation, are involved in the pathogenesis of an array of adverse pregnancy outcomes.^{41,55,58–62,73} Thus, our finding that Epigallocatechin gallate prevents diabetic embryopathy by inhibiting DNA hypermethylation has a significantly high clinical value. Epigallocatechin gallate may be effective in preventing adverse pregnancy outcomes associated with epigenetic modifications.^{41,55,58–62,73}

In addition to neural tube defects, Epigallocatechin gallate may also be effective in correcting epigenetic alterations associated in other structural birth defects such as congenital heart defects.^1,8,56 Although the efficacy of general antioxidants, including folic acid, in reducing adverse pregnancy outcomes is controversial,^74–76 antioxidants with specific inhibitory effects on DNA methylation, such as Epigallocatechin gallate, may be better therapeutics.

Consistent with our current observation of Epigallocatechin gallate treatment preventing neural tube defects, recent studies have revealed the effectiveness of a group of natural compounds in preventing diabetic embryopathy in animal models.^{30,31,37,49,77} However, the effectiveness of these natural compounds needs be tested in human diabetic pregnancies.

In addition to birth defects, maternal diabetes causes a variety of adverse pregnancy outcomes including preterm birth, small for gestational age, fetal hypertrophic cardiomyopathy, preeclampsia, stillbirth and perinatal deaths.^78–84 Future studies may test the effectiveness of Epigallocatechin gallate on these adverse pregnancy outcomes induced by maternal diabetes.

Previous studies have demonstrated that neural tube defects may result from diabetes as a consequence of oxidative stress, endoplasmic reticulum stress, or apoptosis.^17,85 Neural tube closure–essential genes, which are repressed by DNA methylation, are required for cell survival and cellular homeostasis.⁸⁶ Therefore, future studies will aim to reveal the relationship of these factors with Epigallocatechin gallate treatment.