Homocysteine-induced neural tube defects in chick embryos via oxidative stress and DNA methylation associated transcriptional down-regulation of miR-124

Abstract

Although moderate homocysteine (HCY) elevation is associated with neural tube defects (NTDs), the underlying mechanisms have not been elucidated. In this study, we aimed to investigate that whether HCY-induced NTDs were associated with oxidative stress and methyl metabolism in chick embryos. The potential role of miR-124 in neurogenesis was also investigated. In this study, increased intracellular oxidative species and alterations in DNA methylation were observed following HCY treatment. This alteration coincided with decreases of Mn superoxide dismutase and glutathione peroxidase activities, as well as the expression of anti-rabbit DNA methyltransferase (DNMT) 1 and 3a. In addition, HCY induced significant decreases of S-adenosylmethionine (SAM)/S-adenosylhomocysteine (SAH) (P < 0.05). N-acetyl-L-cysteine and choline ameliorated global DNA hypomethylation induced by HCY. MiR-124 levels were significantly suppressed by HCY (P < 0.05), while elevated by 5-aza-2′-deoxycytidine (5-aza-dC). MiR-124 knockdown resulted in spina bifida occulta. Our research suggests that HCY-induced NTDs were associated with oxidative stress and methyl metabolism in chick embryos. MiR-124 down-regulation may occur via epigenetic mechanisms and contribute to HCY-induced NTDs in chick embryo models.

Introduction

Homocysteine (HCY) is a non-essential sulfur-containing amino acid and an intermediate metabolic product of methionine, a demethylated essential amino acid. Recently, HCY has become widely recognized as an independent and graded risk factor for a variety of debilitating and potentially fatal human diseases. Studies found a significant difference in plasma total homocysteine between mothers of having a neural tube defects (NTDs) offspring and mothers of having a normal offspring [1, 2]. However, the causative role of HCY remains controversial and needs to be further elucidated.

Reactive oxygen species (ROS) are toxic to cells and tissues by damaging lipids, nucleic acids, and proteins [3]. Oxidative stress has been shown to be an important mechanism of HCY toxicity in neuronal and vascular cells [4, 5]. HCY has also been found to induce vascular dysfunction, atherogenesis, and neurological dysfunction via oxidative stress [6].

HCY conversion to S-adenosylhomocysteine (SAH) is another mechanism that may explain deleterious HCY effects [7]. Altered methylation patterns related to increased SAH levels have been demonstrated to modulate protein function and transcriptional control of GC-rich promoters [8]. Global hypomethylation can result in oncogene expression, genomic instability, and imprinting loss [9]. Promoter region hypermethylation is associated with tumor-suppressor gene silencing. Recent studies demonstrate that the same mechanism plays a role in characterized loss of microRNA (miRNA) expression [10, 11]. MiRNAs are short, ~22 nucleotide non-coding RNAs that are considered to regulate gene expression by sequence-specific base pairing to 3′-untranslated regions of target mRNAs that direct degradation or translational inhibition. In addition, miRNA expression patterns are considered to play important roles in cell proliferation, apoptosis, and differentiation [10, 11].

NTDs are severe common birth defects that result from the neural tube failing to close. Genetic defects and environmental factors are considered to be associated with NTDs, though the fundamental mechanism remains limited [12]. Oxidative stress has been widely accepted as a pathogenesis component of many diseases including NTDs [13]. Several studies have reported an association between NTDs and increased oxidative stress as well as decreased antioxidant levels [14, 15]. Study demonstrated that global DNA hypomethylation is associated with NTD-affected pregnancies [16]. Several reports have shown that methylation cycle inhibition induces cranial NTDs at a high frequency in ex vivo mouse and chick embryos [17]. These studies suggest that impaired genomic methylation underlies the complex pathogenesis of NTDs.

The underlying mechanisms of HCY-induced NTDs are still largely unknown. Oxidative stress and altered DNA methylation are observed in HCY-treated animals, although the mechanism remains unclear. Therefore, the aim of study was to determine whether HCY treatment alters ROS levels, which then affects DNA methylation and may contribute to NTDs in chick embryos. We also explored the role of miR-124 in neurogenesis and the possible regulatory mechanism of miR-124 expression in chick embryos. This study may provide new prevention methods for HCY-induced NTDs during early embryogenesis.

Materials and Methods

Embryo treatments

White Leghorn chick eggs (Bovan strain) were purchased from Beijing Merial Vital Laboratory Animal Technology Co., Ltd., and incubated in a roller incubator (Grumbach S84, Germany) at 37°C with 55% humidity. Compounds were directly injected into the egg yolk center via a small hole in the blunt end of the egg using an established protocol. Normal chick developmental stages were based on experimental procedures by Rosenquist et al. [18]. Chick embryos were treated with 50-μl HCY (200 mM, Sigma-Aldrich) with or without N-acetyl-L-cysteine (NAC) (200 mM; Sigma-Aldrich) or choline (CHO) (25 μg/μl at Hamburger-Hamilton (HH) stages 6, 8 and 12 [19]. Embryos were harvested for analysis after 72 h (HH stage 19–20).

Procedures were approved by the Institutional Animal Care and Use Committee and in accordance with the National Institutes of Health guide for the care and use of laboratory animals.

Histology

Embryos harvested at 72 h were fixed in 4% formaldehyde, embedded in paraffin and serially sectioned (4 μM). Neural tube cross-sections were stained with hematoxylin–eosin (H&E) for morphological assessment. Sections were examined and imaged with a Nikon TE 2000-U microscope (Nikon, Japan). Dorsal neural tube dysraphism was defined as spina bifida occulta. The incidence of spina bifida occulta was expressed as the ratio of spina bifida occulta in embryos versus total embryos.

Oxidative stress-related biomarker assay

Determination of intracellular ROS levels was performed with an ROS assay kit (Beyotime Institute of Biotechnology, Jiangsu, China). Live 3-d-old embryos (theoretical stage 19–20) were treated with 10-μM 2′,7′-dichlorofluorescein diacetate (DCFH-DA) supplemented into embryo medium [Dulbecco’s Modified Eagle Medium (high glucose)] for 30 min at 37°C and then washed three times for 5 min with embryo medium. The positive control was pre-treated with rousp for 30 min at 37°C and the negative control was treated with DCFH-DA-free embryo medium.

Harvested 3-d-old embryo and whole-embryo lysates were prepared in a radioimmunoprecipitation assay (RIPA) lysis buffer at 4°C. Homogenates were centrifuged at 10 000 g for 10 min at 4°C, and the supernatants were collected.

Superoxide dismutase (SOD) activity was measured using a CuZn/Mn-SOD assay kit (WST) (Beyotime Institute of Biotechnology, Jiangsu, China). Briefly, total SOD activity was measured by reduction rate inhibitions of 2-(4-iodophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium and monosodium salt (WST-1). The Mn SOD activity was measured by adding 10-mM potassium cyanide to inactivate Cu-Zn SOD activity. The difference between total SOD and Mn SOD activity was considered as the Cu-Zn SOD activity. SOD activity was expressed as units per mg of protein (one unit was defined as the amount of enzyme that inhibited WST-1 reduction by 50%).

Glutathione peroxidase (GPX) activity (nmol NADPH/min/ml) was assayed with a Cellular Glutathione Peroxidase assay kit (Beyotime Institute of Biotechnology, Jiangsu, China). Lysates were mixed with 50-mmol/l Na₂HPO₄/NaHP₂O₄, pH 7.0, 1-mmol/l EDTA, 1-mmol/l NaN₃, 0.2-mmol/l NADPH, 1-mmol/l glutathione and 1-U/ml glutathione reductase in 0.1-mL volumes at 25°C for 5 min. Reactions were initiated by 1.5-mmol/l cumene hydroperoxide, and absorbances were measured at 340 nm for 3 min.

The lipid peroxidation product of malondialdehyde (MDA), which is generated in embryos by free radical injury, was measured by thiobarbituric acid reactivity using a colorimetric assay kit (Beyotime Institute of Biotechnology, Jiangsu, China). MDA concentration was calculated by a calibration curve using 1,1,3,3’tetra-ethoxy propane as the standard and expressed as nmol per mg of protein in the embryo.

Whole-mount immunofluorescence

Whole-mount immunofluorescence was performed as described by Rivera-Pérez et al. [20]. Harvested 3-day-old embryos were rinsed with phosphate buffered saline (PBS)-T (0.1-mol/l PBS + 0.1% Tween20), fixed with 4% paraformaldehyde in 0.1-mol/l PBS and stored at 4°C overnight. Embryos were immersed in 100% methanol and then stored at −20°C. After washing in 100% methanol, embryos were rehydrated sequentially using 100–25% methanol in 0.1-mol/l PBS. Primary antibodies were mouse anti-5-mecthylcystidine (5-mec), goat anti-SCP1 (Sigma-Aldrich Corp., St Louis, MO, USA) and rabbit anti-Nanog (Milipore). After rinsing in PBS, embryos were incubated with a fluorescein isothiocyanate-conjugated goat anti-mouse IgG secondary antibody (Jackson Immunoresearch Laboratories, Inc., West Grove, PA, USA). Fluorescence was observed at 488-nm wavelength excitation.

Western blot analysis

To detect protein levels in embryos, whole-embryo lysates were prepared using RIPA lysis buffer and assayed by western blotting. Proteins were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis. Immunoblotting was performed with anti-rabbit DNA methyltransferase (DNMT)3a (1:1000, Santa Cruz, CA, USA) and anti-human DNMT1 (1:500, New England Biolabs) antibodies. Proteins were detected using horseradish peroxidase-conjugated goat anti-rabbit IgG (Jackson Immunoresearch Laboratories, Inc., West Grove, PA, USA) and chemiluminescence was produced using an ECL kit (Pierce, Appleton, WI, USA). β-Actin detection by a mouse antibody (1:3000, Santa Cruz, CA, USA) was used as a loading control. Experiments were repeated three times. Protein bands were evaluated using a Quantity One analyzing system (Bio-Rad, Hercules, CA, USA).

Endogenous SAM and SAH levels

Separation and quantification of S-adenosylmethionine (SAM) and SAH were performed by modifying the procedure of Yi et al. [21]. Embryos were homogenized in 500-μl ice-cold 50-mM HEPES buffer pH 7.4, 1-mM DTT and 0.1-mM EDTA. Insoluble material was removed by centrifugation at 10 000 g for 5 min at 4°C. Aliquots of 200 μl were deproteinated by adding 40 μl 40% trichloroacetic acid followed by incubation on ice for 30 min. Protein-containing precipitates were removed by centrifugation at 15 000 g for 15 min at 4°C and supernatants were filtered through 0.2-μM millex-FG 13 PTFE filters (Sigma, catalog number Z227536). Aliquots were injected into a 4.6 mm × 25 cm Agilent extend C 18 1100 HPLC column (5-μm beads) (Agilent Technologies, Santa Clara CA) using a UV–VIS detector (UV-2000 detector; Thermo Electron, Corp., Austin, TX, USA) set at 254 nm. The mobile phase consisted of 50-mM sodium mono-phosphate, 10-mM 1-heptane sulfonic acid, and 7.5% methanol, adjusted to pH 3.4 with 85% phosphoric acid. The isocratic flow rate was 1 ml/min (1800–2100 psi). SAH and SAM concentrations were determined by standard curves prepared from known SAH and SAM concentrations (0–350 nmol) (Sigma-Aldrich).

In ovo electroporation

About 0.7-μg/μl miR-124 mimics, pre-miR-124 control, locked nucleic acid (LNA)-modified miR-124 oligonucleotide or anti-miR-124 control was injected into HH stage 6 chick embryos (Table 1). At least five embryos were analyzed for HE staining.

Table 1.

miR-124-related nucleotides

miRNA	Sequence
miR-124 mimics	Sense: 5′-UAAGGCACGCGGUGAAUGCC-3′
	Antisense: 5′-CAUUCACCGCGUGCCUUAUU-3′
pre-miR-124 control	Sense: 5′-UUCUCCGAACGUGUCACGUTT-3′;
	antisense: 5′-ACGUGACACGUUCGGAGAATT-3′;
miR-124 inhibitors	5′-gGcaTtcACCgCgtGccTta-3′ (capitals indicate LNAs)
anti-miR-124 control	5′-CAGUACUUUUGUGUAGUACAA-3′.

Total RNA isolation and miRNA expression assays

Total RNA was isolated using TriZol (Invitrogen) according to the manufacturer’s instructions. RNA concentration and purity were determined by measuring the absorbances at 260 and 280 nm. Reverse transcription (RT) was performed using miRNA stem-loop primers and a Taqman® miRNA RT kit (Applied Biosystems, Australia). Mature miRNA expression was measured by a Taqman® microRNA assay (Applied Biosystems, Australia) according to the manufacturer’s instructions. miRNA expression levels were normalized to those of U6 small nuclear RNA.

Data analysis

Statistical analysis was performed using SPSS 25.0 software. Results represented the means ± standard deviation (SD) of at least three independent experiments. Statistical data were evaluated for significance by an analysis of variance. The incidence of NTDs was analyzed by the Chi Square statistic. P < 0.05 was regarded as statistically significant.

Results

HCY-induced spina bifida is associated with oxidative stress and methyl metabolism

Morphological examination of H&E sections contributed to analysis of hidden NTDs including spina bifida, exencephaly and cranioschisis. In the study, 40.0 ± 20.0% of embryos indicated spina bifida occulta in dorsal regions with HCY treatment alone (Fig. 1); 21.6 ± 2.8% of embryos indicated spina bifida occulta following co-treatment with NAC, a known antioxidant, and statistically significant difference was detected between the two groups; 24.3 ± 7.5% of embryos showed spina bifida occulta by co-treatment with CHO, a methyl donor. These results indicate that oxidative stress and methyl metabolism may closely associate with spina bifida occulta induced by HCY in chick embryos.

Figure 1.

The effect of different treatment on chick neural tube development (10×). Chick embryos were treated with HCY (200 mM) with or without NAC (500mM) or CHO (25 μg/μL) for 72 h. (A) The morphological analysis of spinal bifida by H&E staining. L: lumen; N: notochord; NT: neural tube. Scale bar = 100 μm. (B) Quantification of spina bifida occulta occurrence. Values are expressed as mean ± S.D. (n = 5 embryos/group/experiment in three different experiments). *P < 0.05, compared with the control.

HCY effects on early neural markers during chick embryogenesis

To explore the mechanism of HCY-induced spina bifida, we investigated the level of early neural markers, including Pax6, Tuj1, nanog, and nestin in HCY-treated chick embryos. As shown in Fig. 2A, whole-mount immunofluorescence showed that the level of progenitor marker Pax6 was increased, and the neuronal marker Tuj1 was attenuated in whole embryo, brain, and spine of chick embryo, respectively (P < 0.05). However, the expression of pluripotent stem cell marker nanog and neural stem cell marker nestin were detected with no obvious difference compared with controls. To further investigate the effect of HCY on neural tube development, Pax6, Tuj1, and SCP1 protein expression were analyzed by immunohistochemistry. HCY induced over-expression of Pax6 and SCP1, and down-regulation of Tuj1 (Fig. 2B1–B3).

Figure 2.

Effects of HCY on early neural markers in chick embryos (20×). (A) Early neural marker expression was analyzed by whole-mount immunofluorescence in 200-mM HCY-treated chick embryos. Fluorescence intensity of Pax6, Tuj1, nanog and nestin was visualizedin in whole embryo, brain and spine with the Axiovision Rel. 4.8 Software (Zeiss). Scale bar = 500 μm. (B1-B3) The expression of SCP1, Pax6 and Tuj1 in HCY treatment (200 mM) was detected by imunohistochemistry, respectively. Values are the mean ± SD (n = 3 embryos/group/experiment in three different experiments), *P < 0.05, compared with control.

Protective effects of NAC on HCY-induced oxidative injuries

To further explore the effects of HCY on chick embryos, DCFH-DA, an ROS indicator, was detected by fluorescence at 488-nm-wavelength excitation. DCFH-DA staining was significantly increased following HCY treatment (Fig. 3, P < 0.05). Furthermore, antioxidant enzyme activities, SOD and GPX activities were measured. As shown in Fig. 4A, total SOD activity, which includes Cu-Zn SOD and Mn SOD activities, was significantly decreased following HCY treatment (P < 0.05). Mn SOD activity was also significantly decreased (Fig. 4A, P < 0.05), while no significant difference was detected in Cu-Zn SOD activity. As shown in Fig. 4B, the activity of GPX was also significantly reduced in HCY- treated chick embryos (P < 0.05). MDA, as a major aldehyde product of lipid hydroperoxides, was dramatically increased by induction of HCY (P < 0.05) (Fig. 4C). However, Co-treatment with HCY and NAC resulted in the attenuation of fluorescence signals (Fig. 3). Total SOD activity and MDA levels were also partially rescued (Fig. 4A and C). These results clearly imply that HCY could elevate ROS levels by the attenuation of Mn SOD and GPX activities in chick embryos.

Figure 3.

The effects of HCY on ROS in chick embryos. (A) Chick embryos were treated with HCY (200 mM) with or without NAC (500 mM) for 72 h. DCFH-DA staining increased in HCY-treated embryos indicating HCY-induced ROS. HCY and NAC co-treatment partially decreased HCY-induced ROS. Scale bar = 500 μm. (B) DCFH-DA fluorescence intensities. Fluorescence images were visualized as described. Values are the mean ± SD (n = 3 embryos/group/experiment in three different experiments), *P < 0.05. P: positive control (Rosup pretreatment); N: negative control (DCFH-DA-free embryo medium).

Figure 4.

Effects of HCY on other oxidative stress-related indicators in chick embryos. Chick embryos were treated with HCY (200 mM) with or without NAC (500 mM) or CHO (25 μg/μL) for 72 h. The activity of total SOD, GPX and MDA content were analyzed, respectively (A-C). Values are the mean ± SD (n = 3 embryos/group/experiment in three different experiments), *P < 0.05.

HCY induces hypomethylation via inhibiting DNMT 1 and 3a expression

Studies indicate that HCY associates with DNA methylation [22]. In this study, we investigated whether global DNA methylation was affected by HCY in chick embryos. 5-mec, as an indicator of global DNA methylation, was significantly reduced in whole embryo, brain, and spine in HCY-treated embryos by whole-mount immunofluorescence, respectively (Fig. 5A, P < 0.05). Furthermore, DNMT 1 and 3a, which have close relationship with DNA methylation, was markedly decreased in HCY-treated embryos by 22.3% (P < 0.05) and 36.4% (P < 0.05), respectively (Fig. 5B). Moreover, co-treatment with CHO or NAC partially rescued the expression level of DNMT 1 and 3a.

Figure 5.

The effects of HCY on Global DNA methylation and DNMT 1, 3a expression. (A) Chick embryos were treated with HCY (200 mM) with or without NAC (500 mM) or CHO (25 μg/μL) for 72 h. 5-mec was analyzed by whole-mount immunofluorescence (×20). Scale bar = 500 μm. Fluorescence intensity of 5-mec was visualized in whole embryo, brain and spine with the Axiovision Rel. 4.8 Software (Zeiss). Values are the mean ± SD (n = 3 embryos/group/experiment in three different experiments), *P < 0.01. (B) DNMT 1 and 3a expression was detected by Western blot in various embryo treatments, respectively. β-actin was used as a loading control . Representative images revealed the effects of NAC (500 mM) or CHO (25 μg/μL) on DNMT 1 and 3a protein expression in HCY-induced embryos. Values are the mean ± SD (n = 3 embryos/group/experiment in three different experiments), *P < 0.05.

SAM and SAH changes are induced by HCY and modulated by CHO

As substrate and product of essential cellular methyltransferase reactions, SAM and SAH intracellular concentrations are important metabolic indicators of methylation transferring capacity. We compared SAM and SAH concentrations in various embryo treatments. Table 2 shows that SAM significantly decreased (P < 0.05), while SAH significantly increased and SAM/SAH obviously decreased following HCY treatment (P < 0.05). SAH has a higher affinity for the methyltransferase active site compared with that of the precursor SAM. Co-treatment with CHO partially rescued SAM and SAM/SAH levels. However, there was no significant difference in SAM and SAM/SAH levels with NAC co-treatment. SAH pathological accumulation can decrease the SAM/SAH ratio and inhibit most cellular methyltransferase activity, which may induce DNA hypomethylation. Our results suggest that HCY may also contribute to DNA hypomethylation via reduction of SAM/SAH ratio in chick embryos, and HCY-induced oxidative stress may affect methylation by another mechanism.

Table 2.

The content of SAM, SAH and SAM/SAH in different treatment

	nmol SAH/g	nmol SAM/g	SAM/SAH
Saline(CON)	16.754 ± 6.114	77.860 ± 32.430	4.577 ± 0.488
HCY	22.346 ± 6.090*	67.054 ± 29.423*	3.223 ± 0.935*
HCY + CHO	19.542 ± 1.612	74.451 ± 27.112	3.760 ± 1.067
HCY + NAC	20.902 ± 3.621	64.990 ± 5.872	3.192 ± 0.783

Data presented as mean ± S.D. (n = 3)

^* P < 0.05, compared with control.

HCY induces miR-124 down-regulation

MiR-124 has mature sequences conserved from Caenorhabditis elegans to humans and is the most abundant miRNA in the embryonic and adult central nervous system (CNS). Our preliminary data indicated that miR-124 was significantly down-regulated in fetal rat with NTDs [23]. In this research, miR-124 expression level was detected in various embryo treatments. As shown in Fig. 6B, miR-124 expression significantly decreased in HCY-treated chick embryos (P < 0.05), while co-treatment with NAC or CHO partially rescued miR-124 expression. Studies indicate that miR-124 expression is epigenetically suppressed in a number of tumor types including colorectal and breast cancers [24]. Therefore, we investigated that whether miR-124 expression was affected in chick embryos following HCY treatment with 5-aza-2′-deoxycytidine (5-aza-dC), a DNA methylation inhibitor. Our results showed that DNMT1 and DNMT3a were significantly down-expressed by 5-aza-dC, respectively (P < 0.05, Fig. 6A). miR-124 expression was increased following co-treatment with HCY and 5-aza-dC (Fig. 6B). These data indicate that epigenetic modifications of regulatory sequences in CpG islands may contribute to miR-124 silencing in HCY-treated embryos.

Figure 6.

MiR-124 expression in various embryo treatments. Chick embryos were treated with HCY (200 mM) with or without NAC (500 mM), CHO (25 μg/μL) or 5-AZA (10 μM) for 72 h. (A) DNMT 1 and 3a expression was detected by Western blot in 5-AZA-treated chick embryos. β-actin was used as a loading control . (B) miR-124 expression was detected by a Taqman MicroRNA assay. Values are the mean ± SD (n = 3 embryos/group/experiment in three different experiments), *P < 0.05.

miR-124 down-regulation and over-expression effects on neurogenesis

To further demonstrate miR-124 expression effects on neurogenesis, we transfected miR-124 mimics or LNA-miR-124 oligonucleotides into HH stage 6 chick embryos. Neurogenesis was analyzed by morphological examination of H&E sections. As shown in Fig. 7, there was no significant change of the neural tube with miR-124 over-expression; however, miR-124 knockdown induced spina bifida occulta (26.6 ± 11.5%, P < 0.05). Therefore, in vivo results indicate that miR-124 plays an important role in neurogenesis.

Figure 7.

MiR-124 knockdown and over-expression effects on neural tube closure. (10×). miR-124 mimics or LNA-miR-124 oligonucleotides were transfected into HH stage 6 chick embryos, respectively. (A) The morphological analysis of spinal bifida by H&E staining. Scale bar = 100 μm. (B) Quantification of spina bifida occulta occurrence. Values are expressed as mean ± S.D. (n = 5 embryos/group/experiment in three different experiments), P < 0.05.

The effects of miR-124 on the expression of SCP1

SCP1 is the potential 3′-UTR target for miR-124. The expression of SCP1 was detected by immunohistochemistry following transfection of miR-124 mimics or LNA-miR-124 oligonucleotides into HH stage 6 chick embryos, respectively. The expression of SCP1 is significantly increased after miR-124 down regulation (Fig. 8, P < 0.05). Therefore, we speculate that down-regulation of miR-124 may lead to up-regulation of SCP1, which will result in increase of neural precursor cells and thickening of neural epithelium. This may be the reason why the neural tube cannot be closed normally.

Figure 8.

The effects of miR-124 on the expression of SCP1. The expression of SCP1 was detected by imunohistochemistry following transfection of miR-124 mimics or LNA-miR-124 oligonucleotides into HH stage 6 chick embryos, respectively. Values are the mean ± SD (n = 3 embryos/group/experiment in three different experiments), *P < 0.05, compared with control.

Discussion

Maternal hyperhomocysteinemia is associated with complications, such as repeated miscarriages, placental abruption, in utero fetal death, intrauterine fetal growth restriction and fetal NTDs [25]. High levels of HCY are related to an increased risk of the occurrence of congenital anomalies. Studies indicate that HCY can affect the cell cycle proteins and proteins involved in mesenchymal cell differentiation during limb development in a chicken embryo model [26]. Angiogenesis is a critical process involved in embryo survival and development. HCY exposure inhibits early vascular development, displayed by a significant reduction of vessel area, and altered composition of the vascular beds [27]. However, the HCY mechanism of fetal brain degeneration is mostly uncharacterized. We first investigated HCY effects on early neural markers in chick embryos. In the developing CNS, neurogenesis requires neuronal gene up-regulation and non-neuronal gene down-regulation. Our results demonstrated that HCY induced progenitor marker Pax6 over-expression and neuronal marker Tuj1 down-regulation in chick embryos.

NTDs, emerging between the 3rd and 4th weeks of embryogenesis, are caused by partial or complete failure of neural tube closure. Several studies have shown that NTDs are associated with increased oxidative stress and decreased antioxidant levels [14, 15]. Our study reveals that HCY induces oxidative stress by decreasing Mn SOD and GPX activities. Studies indicate that NAC specifically induces Mn SOD expression and activity [28]. In this study, NAC treatment also partially rescued intracellular ROS levels and antioxidant enzymes in HCY-treated chick embryos.

In mammals, methylation of gene promoters is probably one of the foremost mechanisms responsible for cell differentiation during embryogenesis [29]. Several reports suggest that aberrant DNA methylation is an underlying mechanism of NTDs [30–32]. Our results showed that global DNA methylation was clearly decreased in HCY-treated chick embryos. Global DNA hypomethylation can contribute to genomic instability and has been demonstrated to be associated with NTD-affected pregnancy. Our research indicates that HCY induces NTDs via global DNA hypomethylation, which may associate with the reduction of DNMT 1 and 3a. Additionally, HCY resulted in the increase of an intracellular SAH, and reduction of SAM. CHO, similar to folic acid, is involved in one-carbon metabolic methylation of HCY to methionine. Our study indicates that HCY and CHO co-treatment partially rescued SAM and SAH levels. These results suggest that HCY induces hypomethylation by reducing the SAM/SAH ratio.

Studies indicate that ROS may lead to epigenetic changes that strongly affect gene expression without changing the DNA sequence [33, 34]. Therefore, we explored the relationship between oxidative stress and DNA methylation. In our research, NAC impaired global DNA hypomethylation following HCY treatment. HCY and NAC co-treatment increased DNMT 1 and 3a expression. These data support the suggestion that oxidative stress regulates DNA methylation via modifications of DNA methyltransferase expression. Taken together, our results demonstrated that both oxidative stress and SAM/SAH reduction contribute to HCY-induced global DNA methylation.s.lock embryionrole.

During neurogenesis, neuron-enriched miR-124 is present at undetectable or very low levels in neural progenitors but is highly expressed in differentiating and mature neurons. Our preliminary data indicated that miR-124 was significantly down-regulated in fetal rats with neural tube defects [35]. Zhang et al. [36] reported that miR-124 was down-regulated in the tissues of fetuses with anencephaly. In our research, miR-124 expression was significantly decreased in HCY-treated chick embryos. However, co-treatment with HCY and NAC partially rescued miR-124 expression. These data support the suggestion that HCY may participate in oxiditive stress through miR-124.

In summary, the present study reveals that global DNA hypomethylation is an epigenetic event associated with HCY-induced NTDs in chick embryos. Oxidative stress reduces DNMT 1 and 3a expression. SAM/SAH reduction also inhibits DNMT 1 and 3a activities. Because of these two mechanisms, HCY treatment results in global DNA hypomethylation and may induce miR-124 down-regulation via epigenetic mechanisms. miR-124 knockdown results in SCP1 over-expression and a significant increase in neural progenitors, which may affect neural tube closure. This study provides a novel viewpoint for exploring HCY-induced NTDs. However, it has certain limitations; larger sample size should be done in the further research. Moreover, animal experimentation can be performed to test whether global DNA hypomethylation precedes NTDs. Additional target gene sites of hypo or hypermethylation also need to be explored. For the miRNAs at other stages of development such as proliferation maturation, we will select to explore their function in the further research.

Ethics approval and consent to participate

This study was approved by the Ethics Committee of National Research Institute for Family Planning.

Conflict of interest statement

The authors declare that they have no competing interests.