Accumulating evidence has strongly suggested that adverse environmental factors contribute to congenital heart disease (CHD) by interfering with regulatory pathways directing embryonic development early in gestation. Maternal exposure to teratogenic drugs has been identified as potential environmental factors for CHD. Valproic acid (VPA) is one of the most frequently prescribed antiepileptic drugs (AEDs) during pregnancy. Although VPA possesses a good side effect profile with little sedation and few behavioral effects, it warrants particular attention owing to more teratogenic than other AEDs. Compelling body of studies and our previous studies have confirmed that in utero exposure to VPA was associated with cardiac anomalies.[3, 4] Yet, on the whole, the teratogenic mechanism of VPA is still largely unexplained. Therefore, exploration of the molecular mechanisms by which regulatory pathways VPA-induced cardiac abnormalities is crucial for prevention of CHD and development of safer drugs.
For further unraveling the teratogenic effect of VPA on the cardiomyocytes, we used H9C2 cells (undifferentiated rat cardiomyoblasts) to be treated with VPA, which demonstrated that VPA exposure resulted in defective cell polarity, migration, and lamellipodia formation (data unpublished). These findings indicated that disruption of cardiomyocyte polarity might be a factor for the cardiac teratogenesis of VPA. It has been well recognized that cardiomyocyte polarity is strictly controlled by planar cell polarity (PCP) signaling pathway.[6, 7] PCP pathway is required for transformation of cardiomyocytes from initially spherically shaped to ultimately rod-shaped polarized cells through modification of cytoskeleton, playing indispensable roles in cellular organization and directed migration within chamber walls.[7, 8] Vangl2, Scrib, and Rac1 are pivotal molecules in the PCP pathway to govern apical-basal polarity and tissue development. Convincing evidence has demonstrated that genetic knockout or mutation of Vangl2, Scrib, and Rac1 in mouse all resulted in disruption of proliferation, differentiation, and migration of embryonic cardiomyocytes, ultimately contributing to CHD.[9, 10, 11, 12, 13] Therefore, we postulated that VPA might induce CHD by interfering with these key genes of PCP pathway.
Apart from its antiepileptic property, VPA is a well-recognized histone deacetylase (HDAC) inhibitor. As a pivotal enzyme involved in histone modifications, HDACs play a major role in removing acetyl groups from histone tails, stabilizing nucleosome structure, and then compacting chromatin, leading to blockage of DNA accessibility for transcriptional activators and repress gene transcription. Recent studies have provided convincing evidence that the teratogenic ability of VPA was positively correlated with its HDAC inhibition activity, by which VPA could disrupt gene expression through affecting the status of histone acetylation/deacetylation, resulting in congenital anomalies.[3, 15, 16, 17, 18] VPA preferentially inhibits Class I (1, 2, 3, and 8) HDACs. Among them, HDAC1, 2, and 3, core members of Class I HDACs, are ubiquitously expressed, proven to have substantial roles in cell differentiation, proliferation, and morphogenesis. Recent studies in transgenic mice showed that HDAC1/2/3 were required for normal growth and morphogenesis of the heart, while specific deletions of these enzymes would result in various cardiac abnormalities.[21, 22, 23] Given the pivotal molecular switches of downstream gene regulation, however, little is known about the mechanism by which pathway HDAC1/2/3 affect cardiac development. In this regard, we hypothesized that CHD caused by VPA might be related to interruption of PCP key genes through HDAC1/2/3 participation. Therefore, the aim of the study was to investigate the effect of the VPA on cardiac morphogenesis, aforementioned HDACs, and PCP molecules, subsequently screening out the specific HDACs regulating PCP pathway, which might preliminarily illuminate molecular mechanism of VPA-induced CHD.
Animals and drug treatment
The healthy and adult SPF class C57BL mice about 6–8 weeks old (weigh 21 to 26 g) were purchased from Sichuan University Animal Institution. Animals were housed in groups of five, had free access to food and water, and maintained on a 12 h light/dark cycle (light phase: 08:00–20:00) in a temperature-controlled environment (21 ± 2°C). All the procedures were performed in accordance with the National Institutes of Health Guide and with the approval of the Sichuan University Committee for the Care and Use of Laboratory Animals. Female mice were mated with the male at 5:00 PM and inspected for the vaginal plug in the next morning. Pregnancy was defined after the presence of a vaginal plug and designated as embryonic day 0.5 (E0.5). Pregnant mice were randomly allotted to treatment group, vehicle group, or blank group: the treatment group received a single intraperitoneal injection of VPA (P4543, Sigma-Aldrich, USA) at 700 mg/kg on embryonic day 10.5 (E10.5), the vehicle group received an equivalent volume of saline, and the blank received no interventions. The time point and dose of the treatment were selected as several studies and our previous studies, which demonstrated that E10.5–15.5 was a pivotal stage of cardiac development in fetal mice, in association with abundant expression of PCP pivotal molecules in the heart,[9, 10, 11, 12, 13, 24] and the one time injection of VPA at 700 mg/kg showed the highest rate of cardiac abnormality with acceptable fetal death/reabsorption rates. The pregnant mice were sacrificed through cervical dislocation on E15.5. The fetuses were euthanized by decapitation. The embryonic hearts were quickly removed and stored at −80°C for further mRNA/protein expression analysis, total HDAC activity assay, or fixed in 10% formalin solution for hematoxylin-eosin staining.
Hematoxylin-eosin staining of embryonic hearts
Embryonic hearts were fixed in 10% formalin solution for at least 12 h. The hearts were dehydrated progressively in 60%, 70%, 80%, and 100% ethanol for 1 h, respectively. After xylene treatment, 100% paraffin was used for tissue embedding. Then, embryonic hearts were serially sectioned at 5 μm from the top of the aortic arch to the apex. The slides were viewed and photographed by the microscope (Nikon, Tokyo, Japan).
Cell line and culture conditions
H9C2 cells (undifferentiated rat cardiomyoblasts) obtained from the Cell Bank of Chinese Academy of Science were cultured in 10% fetal bovine serum-DMEM/F-12 (Thermo Fisher Scientific, USA) supplemented with 100 units/ml penicillin and 100 μg/ml streptomycin (Gibco, USA) at 37°C in a humidified atmosphere with 5% CO2.
Transfection of Hdac1/2/3 small interfering RNA in vitro
An amount of 1 × 105 or 4 × 105 H9C2 cells were seeded initially in 24-well or 6-well plates for real-time quantitative polymerase chain reaction (PCR) or Western blot analysis, respectively. After cells were grown to 30–50% confluence, 80 nmol/L small interfering RNA (siRNA) specific for Hdac1 (siG1379144052, GuangZhou RiboBio. Co., USA)/Hdac2 (siB160628043037, GuangZhou RiboBio. Co.)/Hdac3 (siG160824094833, GuangZhou RiboBio. Co.) or control siRNA (siN05815122147, GuangZhou RiboBio. Co.) was transfected into H9C2 cells using Lipofectamine® RNAiMAX Reagent (13778-150, Invitrogen, USA) according to the manufacturer's instruction. Cells were harvested after 48 h of transfection for next step tests. All experiments were repeated three times.
The siRNA sequences used were as follows:
Hdac1 Sense: 5’-GCAAGUACUACGCC GUUAAdTdT-3’
Hdac1 Anti-sense: 3’-dTdT CGUUCAUGAUGCGGCAA UU-5’
Hdac2 Sense: 5’-GAUGGACUCUUUGAGUU UU dTdT-3’
Hdac2 Anti-sense: 3’-dTdT CUACCUGAGAAACUC AAAA-5’
Hdac3 Sense: 5’-GCCAGAAGCACCCAAU GAA-3’
Hdac3 Anti-sense: 3’-dTdT CGGUCUUCGUGGGUUAC UU-5’.
Cell transfection of Hdac3 expression plasmid and valproic acid treatment
Based on the results of siRNA transfection in vitro experiments, demonstrating only Hdac3 was involved in Vanlg2 and Scrib regulation, Hdac3 expression plasmid or control plasmid was transfected into H9C2 cells using Lipofectamine™ 3000 Transfection Reagent (L3000015, Invitrogen) according to the manufacturer's instruction. The plasmids were kindly provided by Liu lab of Sichuan University (Chengdu, China). Following 48 h posttransfection, cells were mock-treated or treated with 8.0 mmol/L VPA for another 48 h. Subsequently, cells were collected and subjected to real-time quantitative PCR or Western blotting analysis.
The Hdac3 plasmid sequences used were as follows:
Anti-sense: AAGCTCGAGTTATTGCTCCA CATCGCTTTC.
Real-time quantitative polymerase chain reaction analysis
Total RNA was isolated and purified using the Trizol reagent (Invitrogen). The concentration of purified RNA samples was assessed spectrophotometrically using the Nanodrop 2000 instrument (Thermo Scientific). RNA (1 μg) was reverse transcribed using PrimeScript™ RT Reagent Kit with gDNA eraser (RR047A, Takara, Japan) according to the manufacturer's instructions.
Amplification of cDNA was performed with SsoFast EvaGreen Supermixture (Bio-Rad Laboratories, Hercules, CA, USA) using 5 μl reaction mixture, 0.5 μl forward primer, 0.5 μl reverse primer, 3 μl nuclease-free H2O, and 1 μl cDNA in a total volume of 10 μl. Preceded by an initial denaturation of 3 min at 95°C, and followed by a continuous melt curve from 65–95°C, the PCR conditions for mice samples were 39 cycles of 30 s at 95°C, 10 s at 58°C for Hdac1/Hdac2/Hdac3/Vangl2/Scrib/Rac1/Gapdh; for H9C2 samples were 39 cycles of 30 s at 95°C, 10 s at 55°C for Hdac1/Hdac2/Hdac3/Vangl2/Scrib/Rac1/Gapdh. A validation experiment had been undertaken in which equal quantities of cDNA were used. Similar amplification efficiencies and cycle threshold (CT) values of Gapdh were obtained among the controls and treatment group. The stability of Gapdh expression among the controls and treatment group guaranteed its use as an appropriate endogenous control for normalization. In addition, we have ascertained the efficiencies of amplifications for all genes involved in our study (Hdac1/Hdac2/Hdac3/Vangl2/Scrib/Rac1/Gapdh), which were consistent across a range of template concentrations. All the slopes of the amplification efficiency curves were more than 95% and efficiencies for the target genes (Hdac1/Hdac2/Hdac3/Vangl2/Scrib/Rac1) and the endogenous control (Gapdh) were approximately equal (0.956–0.989). All samples were amplified in triplicates. Gene expression was represented for the CT value by the mean of triple tests. Relative expressions of the target genes in each sample were normalized to expression of Gapdh using 2−ΔΔCt.
The primer sequences specific for mouse target genes were as follows:
Hdac1: 5’-GAACTACCCACTGCGAGACG-3’ (forward)
Hdac1: 5’-ACAGGGAATCTGAGCCACAC-3’ (reverse)
Hdac2: 5’-GAAGGTGAAGGAGGTCGTAGG-3’ (forward)
Hdac2: 5’-AGGGTTGCTGAGTTGTTCTGA-3’ (reverse)
Hdac3: 5’-TGGTGGGAAGGA AAGTATGG-3’ (forward)
Vangl2: 5’-GTACCTTCGCACCACCAA AC-3’ (forward)
Vangl2: 5’-GCTCACCAAGGTCCACTGTT-3’ (reverse)
Scrib: 5’-ACGTGGAGTCGGTGGATA A-3’ (forward)
Scrib: 5’-TGTCACTGAGGCCCA ACTTT-3’ (reverse)
Rac1: 5’-CGACACCACTGTCCCAATAC-3’ (forward)
Gapdh: 5’-TTCACCACCATGGAGAAGGC-3’ (forward)
Gapdh: 5’-GGCATGGACTGTGGTCATGA-3’ (reverse)
The primer sequences specific for H9C2 target genes were as follows:
Hdac1: 5’-CACAGCGACGACTACATCAAG-3’ (forward)
Hdac1: 5’-ACAGAACTCAAACAAGCCATCA-3’ (reverse)
Hdac2: 5’-CAGTTGCCCTTGATTGTGAAAT-3’ (forward)
Hdac2: 5’-TCTGGAGTGTTCTGGTTTGTCA-3’ (reverse)
Hdac3: 5’-CGTCCGAAATGT TGC-3’ (forward)
Hdac3: 5’-GAAGTTCCTCACTAATGG-3’ (reverse)
Vangl2: 5’-CGA AACAGCAGCCTTACCAC-3’ (forward)
Vangl2: 5’-CTCGCTCACCAATGTCCAC-3’ (reverse)
Scrib: 5’-ACCGCATCCTAGCAGTGAAT-3’ (forward)
Scrib: 5’-CCTCCGTACAAGCAGACACA-3’ (reverse)
Rac1: 5’-TGCTGTCAA ATACCTGGAGTGCT-3’ (forward)
Rac1: 5’-TTGGCATCAGACGCGAAGACTCw-3’ (reverse)
Gapdh: 5’-CAGGGTCGACCTTGCCCACAGCCTT-3’ (forward)
Gapdh: 5’-GCCAAGGATATCCATGACAACT-3’ (reverse).
Western blotting analysis
Samples were lysed by RIPA (P0013B, Beyotime, China) with complete protease inhibitor cocktail (P8340, Sigma-Aldrich) for 20 min at 4°C, centrifuged at 12,000 g for 5 min at 4°C. Supernatants were collected and the protein concentration was determined by enhanced Bicinchoninic Acid protein assay kit (P0010S, Beyotime, China) according to the manufacturer's protocol. Cell lysates were boiled in ×4 sample buffer for 5 min and 50 μg protein/lane was subjected to 8% sodium dodecyl sulfate-polyacrylamide gel. Proteins were transferred to polyvinylidene difluoride membranes (Millipore, Bedford, MA, USA). The membranes were blocked for 60 min with Tris base buffer containing 0.1% Tween 20 (TBST) and 5% nonfat milk and incubated overnight at 4°C with primary antibodies’ against Hdac1 (10197-1-AP, Proteintech, USA, 1:1000)/Hdac2 (12922-3-AP, Proteintech, 1:1000)/Hdac3 (10255-1-AP, Proteintech, 1:500)/Vangl2 (21492-1-AP, Proteintech, 1:1000)/Scrib (C0613, Santa Cruz, USA, 1:100)/Rac1 (D2115, Santa Cruz, 1:200) and Gapdh (CW0100A, CWBIO, 1:500). Following extensive washing with TBST, the membranes were immunoblotted with horseradish peroxidase-conjugated goat antirabbit/goat antimouse/rabbit antigoat immunoglobulin G secondary antibodies (1:2500 dilution) for 2 h at room temperature. Washed several times in TBST, membranes were exposed to enhanced chemiluminescence detection system. The protein expression levels were quantified by software Gelpro32 (Media Cybernetics, USA) and normalized against the Gapdh as an endogenous control. We have confirmed the specificity of antibodies (Hdac1/Hdac2/Hdac3/Vangl2/Scrib/Rac1) through preincubation with the peptide epitope and prevention of binding at 60,000/55,000/49,000/60,000/220,000/22,000, respectively.
Hdac activity assay
The total Hdac activity was determined by Hdac activity colorimetric Assay Kits (K331-100, BioVision, USA) according to the manufacturer's protocol. The assay was based on a two-step colorimetric reaction. The first step of reaction was deacetylation of the acetylated lysine side chain of Hdac colorimetric substrate by a sample containing Hdac activity. Deacetylation of the substrate sensitized the substrate so that, in the second step, treatment with the lysine developer produced a chromophore, analyzed by the microplate reader, which was proportional to the deacetylation activity of the sample. Briefly, the sample was lysed and protein was extracted using the protocol mentioned above. Following measurement of protein concentration, 100 μg of protein was incubated with Hdac colorimetric substrate for 1 h at 37°C. The lysine developer was added to above reaction mixture and incubated for another 30 min at 37°C. Total Hdac activity was determined by assessing the absorbance at 405 nm using a microplate reader (Varioskan Flash, Thermo scientific), expressed as the relative O.D value per μg protein. Background readings were subtracted from all the samples. Then, based on the prepared standard curve using the known amount of deacetylated standard in the kit, the absolute amount of deacetylated lysine generated in the sample could be determined (μmol/L/μg protein).
Mean ± standard error was calculated for quantitative data, while qualitative data were presented as n/%. All analyses were conducted with SPSS 17.0 version (SPSS, Chicago, IL, USA). Shapiro–Wilk test and homogeneity test of variance were used to confirm that quantitative data from different groups came from a normal distribution and met the homogeneity of variance. The differences between two groups were determined by the independent sample t-test. The differences among different groups were determined by one-way analysis of variance followed by Tukey's honestly significant difference multiple range test. Chi-square test was used to compare proportions and Fisher's exact Test was used if not matched. A two-tailed P < 0.05 was considered statistically significant.
Examination of litters at E15.5 revealed that there were no significant differences in the average number of dams per pregnant mouse among VPA-treated, vehicle, or blank group (VPA-treated vs. blank: q = 2.172, P > 0.05; VPA-treated vs. Vehicle: q = 2.764, P > 0.05; blank vs. Vehicle: q = 1.812, P > 0.05). However, the death and resorption rates of embryos were significantly higher in VPA-treated group (χ2 = 64.21, P < 0.001). Histological examination of cardiac tissues showed VPA-induced dominant increase of cardiac abnormalities compared with the controls (χ2 = 35.90, P < 0.001) [Table 1]. Cardiac malformations, most commonly affecting the interventricular septum and the ventricular myocardium, were seen in the mice treated with VPA. Among them, in contrast to the normal closure of interventricular communication [Figure 1a and 1b], one-fifth (29/145) had ventricular septum defects, accounting for 74.4% of all types of cardiac abnormalities [Table 1 and Figure 1e, 1f]. In addition, by E15.5, the normal right and left ventricle had a thick compact myocardium [Figure 1c and 1d]. In contrast, the VPA-treated embryo had a thinned ventricular wall with little compact layers, the marked trabeculae with deep intratrabecular recesses being seen [Figure 1g and 1h]. These anomalies in the disorganization of the ventricular myocardium resembled those seen in the human cardiomyopathy – noncompaction of ventricular myocardium. In comparison with the treated group, cardiac abnormalities were sparse in control groups, only two cases of atrial septal dysmorphogenesis being observed [Table 1].
As shown in Figure 2, after exposure to VPA, Hdac1/2/3 mRNA and protein levels in E15.5 embryonic hearts were significantly decreased in comparison with the blank and vehicle groups (mRNA: q = 4.523/4.762 and P = 0.027/0.014 for Hdac1, q = 4.018/4.655 and P = 0.034/0.021 for Hdac2, and q = 4.502/4.711 and P = 0.028/0.016 for Hdac3, in comparison with the blank and the vehicle, respectively; protein: q = 5.483/4.806 and P = 0.018/0.017 for Hdac1, q = 7.414/6.052 and P = 0.002/0.008 for Hdac2, and q = 7.499/9.589 and P = 0.002/0.001 for Hdac3, in comparison with the blank and the vehicle, respectively) [Figure 2a and 2b]. Referring to PCP genes, VPA dominantly ablated Vangl2/Scrib mRNA and protein expressions (mRNA: q = 4.296/3.583 and P = 0.011/0.027 for Vangl2 and q = 5.123/4.459 and P = 0.002/0.006 for Scrib, in comparison with the blank and the vehicle, respectively; protein: q = 7.717/6.265 and P = 0.002/0.004 for Vangl2 and q = 6.090/4.574 and P = 0.005/0.018 for Scrib, in comparison with the blank and the vehicle, respectively), whereas no significant differences in Rac1 expression were noted after VPA treatment (mRNA: q = 1.920/1.221 and P > 0.05; protein: q = 0.829/0.945 and P > 0.05) [Figure 2b and 2d]. In addition, VPA treatment significantly repressed total Hdac activity of embryonic hearts (q = 4.885/5.051 and P = 0.018/0.012, in comparison with the blank and vehicle, respectively) [Figure 2e].
As revealed in Figure 3, after transfection of Hdac1/2/3 siRNA into H9C2 cells, endogenous expressions of Hdac1/2/3 were successfully inhibited, respectively, as evidenced by qRT-PCR [Figure 3a] and Western blotting [Figure 3c and 3e] in comparison with the control group (mRNA: t = 5.237, P = 0.006 for Hdac1; t = 6.933, P = 0.002 for Hdac2; and t = 6.771, P = 0.002 for Hdac3; protein: t = 4.682, P = 0.009 for Hdac1; t = 4.858, P = 0.008 for Hdac2; and t = 5.976, P = 0.002 for Hdac3). Compared with the control, the inhibition of Hdac3 could significantly result in repression of either Vangl2/Scrib mRNA [Figure 3b] or protein [Figure 3d and 3e] expressions (mRNA: q = 9.947, P < 0.001 for Vanlg2; q = 7.081, P = 0.005 for Scrib; protein: q = 13.502, P < 0.001 for Vangl2; q = 8.896, P = 0.001 for Scrib); however, knockdown of Hdac1/2 was not (mRNA: q = 1.567/2.062 for Vangl2; q = 3.372/2.709 for Scrib; protein: q = 1.110/2.039 for Vangl2; and q = 1.599/0.795 for Scrib; all P > 005). No significant differences in Rac1 expression among different groups were found after Hdac1/2/3 silencing (mRNA: q = 3.094/3.818/1.056; protein: q = 0.539/0.733/0.795; all P > 0.05) [Figure 3b, 3d and 3e]. To further determine the role of Hdac3 on VPA target gene expression, the total RNAs/proteins were extracted from the control or Hdac3-overexpression H9C2 cells in the presence or absence of VPA treatment and subjected to qRT-PCR and Western blotting analysis. As shown in Figure 4a and 4b, VPA exposure significantly decreased the mRNA [Figure 4a] and protein levels [Figure 4b] of Vanlg2 and Scrib compared with controls (mRNA: q = 6.778, P = 0.001 for Vangl2; q = 13.781, P < 0.001 for Scrib; protein: q = 4.972, P = 0.008 for Vanlg2; and q = 9.967, P < 0.001 for Scrib). Conversely, overexpression of Hdac3 induced Vanlg2 and Scrib mRNA [Figure 4a] together with protein expressions [Figure 4b] (mRNA: q = 4.574, P = 0.023 for Vangl2; q = 3.432, P = 0.041 for Scrib; protein: q = 5.236, P = 0.01 for Vanlg2; q = 8.219, P = 0.001 for Scrib). Notably, Hdac3 overexpression completely reversed VPA-induced Vanlg2 and Scrib reduction [Figure 4a and 4b] (Hdac3 plasmid + VPA vs. control: mRNA, q = 0.962/2.538 for Vangl2/Scrib, respectively, protein, q = 0.066/0.626 for Vangl2/Scrib, respectively; all P > 0.05). However, expression of Rac1 was not alter among different groups (mRNA: q = 1.356/1.161/2.225; protein: q = 1.088/1.075/1.037; all P > 0.05). Moreover, it was found that, in comparison with the control, VPA could robustly ablate Hdac activity (q = 7.477, P = 0.001), whereas Hdac3 overexpression increased that (q = 3.660, P = 0.032) and also partially rescued the inhibition caused by VPA (q = 4.711, P = 0.011) [Figure 4c].
Emerging evidence has proven that epigenetics displays a diverse control on meticulous expression of vital genes requiring for the processes of cardiomyocyte growth and proliferation. Drugs altering these epigenetic processes have been shown to play substantial roles in the development of CHD. Although it is well recognized that VPA is teratogenic, few studies have looked for its effect on epigenetic status and cardiac development-related genes. To our knowledge, our study was the first one to unravel whether VPA-induced CHD was related to disruption of PCP pathway through Class I HDACs. Our results showed that in vivo VPA could induce cardiac malformations at the middle stage of pregnancy; VPA-induced repression of Hdac1/2/3 expression and Hdac activity were accompanied by inhibition of Vanlg2 and Scrib in embryonic hearts. In addition, in vitro Hdac3 knockdown by RNA silencing decreased Vangl2 and Scrib expression; Hdac3 overexpression reversed the VPA-induced inhibition of these two genes, as well as Hdac activity. These findings strongly suggested that VPA might induce CHD by inhibiting Vangl2 and Scrib through Hdac3 participation.
The vital roles of PCP signaling in cardiac development have been evidenced by gene knockout mouse models.[9, 10, 11, 12, 13] Vanlg2 and Scrib are required for septation between ventricular chambers; moreover, Scrib also plays significant roles in the organization of developing cardiomyocytes and formation of the compact ventricular myocardium. Deficiency of Vanlg2 or Scrib in embryonic cardiomyocytes exhibits ventricular septal defects. In addition, loss of Scrib displays a thinned ventricular myocardium and ventricular noncompaction as well, apart from ventricular septal defects.[9, 10, 11, 12, 13] The morphology of cardiac abnormalities observed after VPA exposure in our study was very similar with the types of those found in PCP gene deletion models. Further investigation into the effect of VPA on PCP signaling revealed that VPA treatment dramatically repressed Vangl2/Scrib levels in embryonic hearts. Collectively, these results indicated that VPA-induced cardiac anomalies could be due, at least in part, to disruption of PCP pathway during cardiac development.
Our results also illustrated that VPA-induced repression of Vangl2/Scrib expression was accompanied by inhibition of Hdac1/2/3 together with Hdac activity. We further identified the specific Hdac subtype involving in Vangl2/Scrib regulation by transfection of specific siRNA into H9C2 cells. The results manifested that Hdac3 might act as an activator of Vanlg2/Scrib. Moreover, Hdac3 knockdown decreased Vangl2/Scrib mRNA as well as their protein levels, suggesting Hdac3 might activate expressions of the two genes at the transcriptional level. These findings did not correspond with the best-documented biological function of Hdac3, acting as a transcriptional repressor. Of note, this unexpected phenomenon of Hdac3 has been observed both in vivo and vitro experiments previously. It has been reported that Hdac3 was required for transcriptional activation of retinoic acid response element. Cells derived from Hdac3 knockout mice displayed both downregulation and upregulation of gene expression. Furthermore, gene expression profiling studies comparing the cells treated with HDAC inhibitors versus those untreated revealed that the number of genes downregulated were comparable to those upregulated. One possibility is that Hdac3 may prohibit transcription of transcriptional repressors, leading to derepression of gene expression. Alternatively, Hdac3 may deacetylate and, ultimately, activate transcription activators or suppress the functions of transcription repressors independent of histone modifications. However, the process of Hdac3 regulating Vanlg2/Scrib in the cardiomyocytes still remains to be established.
The teratogenic ability of VPA is closely correlated to its Hdac inhibition activity, while VPA analogs that do not have Hdac inhibition activity are not teratogenic.[15, 16] Yet, on the whole, the teratogenic mechanism of VPA is still largely unexplained. In the present study, we also found that VPA-induced cardiac abnormalities were accompanied with inhibition of Hdac activity, whereas Hdac3 overexpression not only partially reversed VPA-induced Hdac inhibition but also blocked the inhibitory effect of VPA on Vangl2/Scrib expressions, indicating that VPA probably interfered with PCP pathway through inhibition of Hdac activity, consequently producing teratogenic effects, in a Hdac3-dependent manner. Therefore, Hdac3 might be promising target for prevention of VPA-induced CHD, if these findings could be validated in animal models and human studies.
However, some limitations must be considered. First, the action mechanism of VPA on Hdacs and PCP signaling needs to be unveiled to gain a deeper insight into novel prevention for its cardiac teratogenesis and the treatment of diseases linked to the overexpression of Hdacs. Second, given the discordance between gene expression and its function, it is unknown whether the VPA and/or Hdac3 is also involved in the regulation of Vanlg2/Scrib function. In addition, we only looked into the regulation of Hdac1/2/3 on PCP pathway. However, apart from Hdac1/2/3, VPA could inhibit other Hdacs, whether other Hdacs are also involved in the regulation of this pathway or not still remain to be further clarified. Finally, a multifactorial pathogenesis with interplay between inherited and noninherited causes was involved in the development of CHD. In the present study, we merely made a preliminary exploration of the possible molecular mechanisms in VPA-induced CHD, and those findings could not be expanded indiscriminately to the complex basis of cardiac perturbations. However, our study has shown that PCP pathway was critical to cardiogenesis, and disruption of this pathway possibly resulted in the occurrence of cardiac developmental disorders. Those findings might aid in the identification of VPA-specific target genes and offer some references for preventing or ameliorating its teratogenic risk in the context of gene-environment interactions. Taken together, the findings in this study were merely the first step toward the possible molecular mechanism of VPA-induced CHD in the context of PCP pathway from the perspective of epigenetics, which could offer a clue for the prevention of its cardiac teratogenesis linked to the overexpression of Hdac3. The results need to be further validated in animal models and human studies.
Financial support and sponsorship
This work was supported by grants from Science-technology Support Plan Projects in Sichuan province (No. 2016FZ0088, and No. 2017SZ0117), National Natural Science Foundation of China (No. 81602817, No. 81741026, and No. 81571515), and Research Projects of Health and Family Planing Commission of Sichuan Province (No. 17PJ258).
Conflicts of interest
There are no conflicts of interest.
1. Vecoli C, Pulignani S, Foffa I, Andreassi MG. Congenital heart disease
: The crossroads of genetics, epigenetics and environment Curr Genomics. 2014;15:390–9 doi: 10.2174/1389202915666140716175634
2. Somerville ER, Cook MJ, O’Brien TJ. Pregnancy treatment guidelines: Throwing the baby out with the bath water Epilepsia. 2009;50:2172 doi: 10.1111/j.1528-1167.2009.02262.x
3. Ornoy A. Valproic acid
in pregnancy: How much are we endangering the embryo and fetus? Reprod Toxicol. 2009;28:1–10 doi: 10.1016/j.reprotox.2009.02.014
4. Wu G, Nan C, Rollo JC, Huang X, Tian J. Sodium valproate-induced congenital cardiac abnormalities in mice are associated with the inhibition of histone deacetylase J Biomed Sci. 2010;17:16 doi: 10.1186/1423-0127-17-
5. Tomson T, Battino D. Teratogenic effects of antiepileptic medications Neurol Clin. 2009;27:993–1002 doi: 10.1016/j.ncl.2009.06.006
6. Henderson DJ, Chaudhry B. Getting to the heart of planar cell polarity
signaling Birth Defects Res A Clin Mol Teratol. 2011;91:460–7 doi: 10.1002/bdra.20792
7. Wu G, Ge J, Huang X, Hua Y, Mu D. Planar cell polarity
signaling pathway in congenital heart diseases J Biomed Biotechnol. 2011;2011:589414 doi: 10.1155/2011/
8. Davis EE, Katsanis N. Cell polarization defects in early heart development Circ Res. 2007;101:122–4 doi: 10.1161/CIRCRESAHA.107.157446
9. Phillips HM, Murdoch JN, Chaudhry B, Copp AJ, Henderson DJ. Vangl2 acts via RhoA signaling to regulate polarized cell movements during development of the proximal outflow tract Circ Res. 2005;96:292–9 doi: 10.1161/01.RES.0000154912.08695.88
10. Ramsbottom SA, Sharma V, Rhee HJ, Eley L, Phillips HM, Rigby HF, et al Vangl2-regulated polarisation of second heart field-derived cells is required for outflow tract lengthening during cardiac development PLoS Genet. 2014;10:e1004871 doi: 10.1371/journal.pgen.1004871
11. Phillips HM, Rhee HJ, Murdoch JN, Hildreth V, Peat JD, Anderson RH, et al Disruption of planar cell polarity
signaling results in congenital heart defects and cardiomyopathy attributable to early cardiomyocyte disorganization Circ Res. 2007;101:137–45 doi: 10.1161/CIRCRESAHA.106.142406
12. Boczonadi V, Gillespie R, Keenan I, Ramsbottom SA, Donald-Wilson C, Al Nazer M, et al Scrib: Rac1 interactions are required for the morphogenesis of the ventricular myocardium Cardiovasc Res. 2014;104:103–15 doi: 10.1093/cvr/cvu193
13. Leung C, Lu X, Liu M, Feng Q. Rac1 signaling is critical to cardiomyocyte polarity and embryonic heart development J Am Heart Assoc. 2014;3:e001271 doi: 10.1161/JAHA.114.001271
14. Eom GH, Kook H. Posttranslational modifications of histone deacetylases: Implications for cardiovascular diseases Pharmacol Ther. 2014;143:168–80 doi: 10.1016/j.pharmthera.2014.02.012
15. Eikel D, Lampen A, Nau H. Teratogenic effects mediated by inhibition of histone deacetylases: Evidence from quantitative structure activity relationships of 20 valproic acid
derivatives Chem Res Toxicol. 2006;19:272–8 doi: 10.1021/tx0502241
16. Gurvich N, Berman MG, Wittner BS, Gentleman RC, Klein PS, Green JB, et al Association of valproate-induced teratogenesis with histone deacetylase inhibition in vivo
FASEB J. 2005;19:1166–8 doi: 10.1096/fj.04-3425fje
17. Di Renzo F, Broccia ML, Giavini E, Menegola E. Relationship between embryonic histonic hyperacetylation and axial skeletal defects in mouse exposed to the three HDAC inhibitors apicidin, MS-275, and sodium butyrate Toxicol Sci. 2007;98:582–8 doi: 10.1093/toxsci/kfm115
18. Di Renzo F, Cappelletti G, Broccia ML, Giavini E, Menegola E. Boric acid inhibits embryonic histone deacetylases: A suggested mechanism to explain boric acid-related teratogenicity Toxicol Appl Pharmacol. 2007;220:178–85 doi: 10.1016/j.taap.2007.01.001
19. Göttlicher M, Minucci S, Zhu P, Krämer OH, Schimpf A, Giavara S, et al Valproic acid
defines a novel class of HDAC inhibitors inducing differentiation of transformed cells EMBO J. 2001;20:6969–78 doi: 10.1093/emboj/20.24.6969
20. Witt O, Deubzer HE, Milde T, Oehme I. HDAC family: What are the cancer relevant targets? Cancer Lett. 2009;277:8–21 doi: 10.1016/j.canlet.2008.08.016
21. Paige SL, Plonowska K, Xu A, Wu SM. Molecular regulation of cardiomyocyte differentiation Circ Res. 2015;116:341–53 doi: 10.1161/CIRCRESAHA.116.302752
22. Montgomery RL, Davis CA, Potthoff MJ, Haberland M, Fielitz J, Qi X, et al Histone deacetylases 1 and 2 redundantly regulate cardiac morphogenesis, growth, and contractility Genes Dev. 2007;21:1790–802 doi: 10.1101/gad.1563807
23. Trivedi CM, Lu MM, Wang Q, Epstein JA. Transgenic overexpression of Hdac3 in the heart produces increased postnatal cardiac myocyte proliferation but does not induce hypertrophy J Biol Chem. 2008;283:26484–9 doi: 10.1074/jbc.M803686200
24. Xin M, Olson EN, Bassel-Duby R. Mending broken hearts: Cardiac development as a basis for adult heart regeneration and repair Nat Rev Mol Cell Biol. 2013;14:529–41 doi: 10.1038/nrm3619
25. Vallaster M, Vallaster CD, Wu SM. Epigenetic mechanisms in cardiac development and disease Acta Biochim Biophys Sin (Shanghai). 2012;44:92–102 doi: 10.1093/abbs/gmr090
26. Seto E, Yoshida M. Erasers of histone acetylation: The histone deacetylase enzymes Cold Spring Harb Perspect Biol. 2014;6:a018713 doi: 10.1101/cshperspect
27. Bhaskara S, Chyla BJ, Amann JM, Knutson SK, Cortez D, Sun ZW, et al Deletion of histone deacetylase 3 reveals critical roles in S phase progression and DNA damage control Mol Cell. 2008;30:61–72 doi: 10.1016/j.molcel.2008.02.030
28. LaBonte MJ, Wilson PM, Fazzone W, Groshen S, Lenz HJ, Ladner RD, et al DNA microarray profiling of genes differentially regulated by the histone deacetylase inhibitors vorinostat and LBH589 in colon cancer cell lines BMC Med Genomics. 2009;2:67 doi: 10.1186/1755-8794-2-67
Edited by: Li-Shao Guo
Keywords:© 2018 Chinese Medical Association
Congenital Heart Disease; Hdacs; Planar Cell Polarity; Valproic Acid