Determinants of Orofacial Clefting I: Effects of 5-Aza-2′-deoxycytidine on Cellular Processes and Gene Expression during Development of the First Branchial Arch

Abstract

In this study, we identify gene targets and cellular events mediating the teratogenic action(s) of 5-Aza-2′-deoxycytidine (AzaD), an inhibitor of DNA methylation, on secondary palate development. Exposure of pregnant mice (on gestation day (GD) 9.5) to AzaD for 12 hrs resulted in the complete penetrance of cleft palate (CP) in fetuses. Analysis of cells of the embryonic first branchial arch (1-BA), in fetuses exposed to AzaD, revealed: 1) significant alteration in expression of genes encoding several morphogenetic factors, cell cycle inhibitors and regulators of apoptosis; 2) a decrease in cell proliferation; and, 3) an increase in apoptosis. Pyrosequencing of selected genes, displaying pronounced differential expression in AzaD-exposed 1-BAs, failed to reveal significant alterations in CpG methylation levels in their putative promoters or gene bodies. CpG methylation analysis suggested that the effects of AzaD on gene expression were likely indirect.

Graphical Abstract

1. INTRODUCTION

In the United States approximately 6,800 babies are born annually with orofacial clefts, such as cleft lip with or without cleft palate (CL/P), and isolated cleft palate (CP) [1,2]. Without proper treatment, such orofacial anomalies often result in feeding and speech impediments, dental abnormalities, hearing loss, ear infections and breathing difficulties. In addition, the affected individuals, owing to their appearance, endure diverse social, emotional and psychological challenges [3].

Palatal clefting primarily occurs as the result of anomalous morphogenesis of the secondary palate. The maxillomandibular prominences of the first branchial arch (1-BA) give rise to various structures of the embryonic orofacial region (e.g. maxilla, mandible, and lateral portions of upper lip) including the palatal processes, precursors of the secondary palate or roof of the oral cavity [4]. The embryology of this process is similar in humans and mice. The critical events include emergence of the bilateral palatal processes as oral projections from the maxillary prominences, reorientation of these processes from a vertical position, lateral to the tongue, to a horizontal position above the tongue, and finally, fusion of these processes to each other along their anterior–posterior length forming the definite secondary palate [5].

Morphogenesis of the orofacial region (including the secondary palate) is reliant, in part, on the migration of neural crest cells derived from the neuroectoderm of rhombomeres 1–3 [6] into the first two branchial arches, and diversification of neural crest cell fates [7]. Since the 1-BAs serve as the palatal precursor tissue, precise regulation of expression of genes and the downstream cellular processes they guide within these tissues is essential for normal development of the secondary palate. Numerous studies using animal models have documented that anomalous 1-BA development results in CP [8–12]. Indeed, altered Wnt [11] or Shh [13] signaling in the 1-BA, both pivotal for normal development of the 1-BA, results in CP. Similarly, defects in development of the 1-BAs are also thought to contribute to craniofacial dysmorphologies that accompany Apert, Pierre Robin and Treacher Collins syndromes [14,15].

Numerous environmental factors such as nutritional deficiencies, exposure to cigarette smoke, alcohol, drugs, pesticides, and microbial infection, have been implicated in the etiology of orofacial dysmorphologies [16–21]. This allows for the hypothesis that a potential epigenetic mechanism by which some of these environmental factors elicit tissue dysmorphology is aberrant gene methylation. Support for such a notion comes from documentation of an epigenetic role of the clf2 gene in the multifactorial etiology of cleft lip and palate [22], and significant CpG island- and global DNA hypomethylation during retinoic acid-induced CP [23].

Epigenetic processes alter gene expression during embryogenesis and mediate heritable activation or silencing of genes in specific cell lineages [24–26] without modifying DNA sequences. Such processes include, but are not limited to, DNA methylation, microRNA (miRNA) function, and histone modification. Mammalian DNA methylation, occurring predominantly at cytosine residues within CpG dinucleotides, typically regulates the expression of specific genes through a transcriptional repression process catalyzed by a family of active DNA methyltransferases (DNMTs) and transcriptional regulators, collectively known as methyl-CpG-binding proteins [27,28]. The cytidine analogue, 5-Aza-2′-deoxycytidine (AzaD), is a potent DNA methylation inhibitor, and teratogen. By forming covalent adducts with cellular DNMT1, AzaD-substituted DNA inhibits DNMT1 activity and triggers demethylation of genomic DNA [29,30].

AzaD has been extensively employed to investigate the role of DNA methylation in regulating gene expression during embryogenesis [31–35]. AzaD treatment of murine fetuses on GD 9.0 has been reported to induce palatal clefting [36,37]. The current studies examining presumptive epigenetic and cellular mechanisms underlying the pathogenesis of AzaD-induced CP, identified key genes encoding several morphogenetic factors, signaling mediators, cell cycle inhibitors and pro-apoptotic factors targeted by AzaD in the developing 1-BAs.

2. MATERIALS AND METHODS

2.1 Animals

Mouse (ICR strain) husbandry and developmental staging of embryos was conducted as described previously [38]. All procedures for the humane use and handling of mice were approved by the University of Louisville Institutional Animal Care and Use Committee and encompass guidelines as set out in the European Commission Directive 86/609/EEC for animal experimentation.

2.2 5-Aza-2′-deoxycytidine Dosing Regimen and Tissue Collection

5-Aza-2′-deoxycytidine was obtained from Sigma Chemical Co., St. Louis, MO. AzaD dosage (1 mg/kg; [37]) was based on the average dam weight on GD 9.5 and was administered in sterile PBS solution. Control embryos were dosed with an equal volume of sterile PBS (vehicle). AzaD or vehicle was delivered via single intraperitoneal injection to pregnant females on GD 9.5 and embryos were collected on GD-15.5 and -17.5. For individual TaqMan quantitative real-time PCR (qRT-PCR) and for TaqMan array card-based gene expression profiling, 1-BAs of AzaD- or vehicle exposed embryos were microdissected 6, 9 and 12 hours post-exposure and pooled for analyses (Figure 6). These three treatment periods were selected on the basis of preliminary data (TaqMan qRT-PCR and immunocytochemistry) that demonstrated: (1) significant, temporal, and differential expression of several genes, and (2) a striking temporal decrease in cell proliferation and marked increase in apoptosis within tissue of the 1-BA. 1-BA tissue from three independent pools of 12 to 15 staged embryos was collected, total RNA and genomic DNA extracted (AllPrep DNA/RNA Mini Kit; Qiagen Inc., Valencia, CA), and stored at −80°C. Whole embryos were also collected 6, 9 and 12 hours post-treatment, fixed in 4% paraformaldehyde (PFA), embedded in OCT (optimal cutting temperature compound, Tissue-Tek 4583, Sakura Finetek USA, Inc., Torrance, CA), frozen at −80°C, cryosectioned and processed as described by Oh et al. [39], either for staining with hematoxylin and eosin, or immunostaining with phospho-histone-H3 or activated caspase-3 antibody, to assess cell proliferation or apoptosis, respectively.

Figure 6. Photomicrographs of embryos exposed in utero on GD 9.5 to 1 mg/kg AzaD or vehicle, for 6, 9 and 12 hours.

(A) The region demarcated by the black line was excised from AzaD- or vehicle-exposed embryos for extraction of total RNA and genomic DNA. (B), (D) and (F) represent 6, 9 and 12 hour AzaD-exposed embryos, respectively; (C), (E) and (G) represent 6, 9 and 12 hour vehicle-exposed embryos, respectively. 1-BA = first branchial arch; 2-BA = second branchial arch.

2.3 “Fetal Phenotyping” and Identification of Palate Phenotype

On GD 15.5 and 17.5, dams exposed to AzaD or vehicle, were euthanized by carbon dioxide asphyxiation followed by cervical dislocation. Uterine horns were exteriorized, and the number and location of implantations and resorptions were recorded. Fetuses were removed from the uterus, blotted, weighed, measured for crown–rump length, and inspected for gross external anomalies. Palatal development and/or degree of palate fusion was assessed by removing the lower jaw for direct visualization of the palate.

2.4 Immunohistochemistry

Embryos exposed in utero (on GD 9.5 for 6, 9 and 12 hours) to either AzaD (1mg/kg) or vehicle, were fixed in 4% PFA, washed in PBS, cryoprotected in sucrose (5–10% in PBS), incubated overnight at 4°C in 20% sucrose, immersed and embedded in OCT. Eight μm coronal sections of the embryos were cut using a Leica CM1900 cryostat (Leica Inc., Bannockburn, IL) and mounted on Superfrost/Plus microscope slides (Fisher Scientific, Pittsburgh, PA, USA). Tissue sections were washed, permeabilized in 0.1% Tween 20, and blocked in 0.1% Tween 20/1.0% BSA/5% FBS (in PBS) for 30 min at 37°C. Sections were immunostained overnight at 4°C with a 1:500 dilution of rabbit anti-phospho Histone H3 (Upstate-Millipore, Billerica, MA) for evaluation of proliferation, or rabbit anti-activated caspase 3 (Cell Signaling Technology Inc., Danvers, MA) for evaluation of apoptosis. Sections were then washed in PBS and goat anti-rabbit AF-594 secondary antibody (Molecular Probes - Life Technologies, Grand Island, NY) at a 1:200 dilution and blocking solution was applied to the sections for 60 min at 37°C. Cellular DNA was counterstained with 300 nM 4′-6′-diamidino-2-phenylindole (DAPI) (Molecular Probes - Life Technologies) and mounted under Fluoromount-G (Southern Biotech, Birmingham, Alabama) to preserve fluorescence. Tissue sections were visualized and photographed with a Nikon Eclipse TE 2000-U microscope equipped with epifluorescence optics. Fluorescent signal intensity was measured using the MetaMorph 6.5 software (Universal Imaging Inc., Downingtown, PA). To assess proliferation and apoptosis in the 1-BAs, the integrated intensity of fluorescence was measured (within the same viewing field using the MetaMorph software) utilizing triplicate sections per embryo. These experiments were performed using independent and comparable sections (in triplicate) derived from AzaD - or vehicle-exposed embryos from each treatment group (6, 9 and 12 hr).

2.5 Gene Expression Profiling and TaqMan^® Quantitative Real-Time PCR

In order to establish a temporal gene expression signature of developing 1-BA tissue exposed in utero to AzaD, total RNA was extracted from microdissected 1-BAs (the region demarcated by the black line in Figure 6A) of embryos exposed to AzaD or PBS (vehicle) for 6, 9 or 12 hours. The quality and quantity of total RNA were determined using an Agilent 2100 Bioanalyzer (Agilent Technologies, Foster City, CA, USA). RNA Integrity Numbers of all the RNA samples were between 9.0 and 10.0. cDNA, synthesized from total RNA using the SuperScript® VILO^™ cDNA Synthesis Kit (Life Technologies, Grand Island, NY, USA) was hybridized to custom-made, 384-well TaqMan Low-Density Array (TLDA) fluidic cards and gene expression profiling was performed using the ViiA7 Real-Time PCR System according to the manufacturer’s instructions (Applied Biosystems - Life Technologies, Grand Island, NY, USA). These array cards contain probe-primers for 379 genes (Table S2-“TLDA plate-Gene list”) encoding regulators of cell cycle/proliferation, apoptosis, signal transduction, growth/differentiation and transcription. Analysis was performed on 3 independent sets of cDNA (total RNA) and gene expression levels were normalized to the expression of the housekeeping gene β-actin. Raw data were acquired and processed with the DataAssist Software utilizing the comparative CT (ΔΔCT) method (Applied Biosystems - Life Technologies). Individual gene expression analysis was performed by TaqMan qRT-PCR (to validate the results of TLDA-card based gene expression profiling) as described [40].

2.6 Methylation Analysis

Pyrosequencing was performed as previously described using the Qiagen PyroMark Q24 instrument [41]. Methylation levels of CpG residues in putative promoters (regions extending up to ~1 kb from the translational start site) and gene bodies of 10 selected genes (Cdkn1a, Caspase 3, Fas, Tnf, E2f1, E2f2, FoxC2, FoxE1, Ptch1 and Shox2) exhibiting significant differential expression were determined in AzaD- and vehicle-treated 1-BAs. Samples for pyrosequencing were derived from no less than 2 independently amplified PCR templates. A list of primers used for DNA amplification and pyrosequencing of the selected genes is presented in Table S2 (“Primers for pyrosequencing”).

2.7 Pathway Analysis

In order to establish an overview of different signaling and biological pathways impacted in 1-BA tissues by in utero AzaD exposure, Ingenuity Pathway Analysis (IPA; Ingenuity Systems, Redwood City, CA) was performed, as previously reported [38], on differentially expressed genes within AzaD-exposed (12-hr) 1-BAs.

2.8 Statistical Analyses

The data set from TLDA-card based gene expression analysis contains 379 genes. Each gene has 9 and 8 samples for AzaD and vehicle (PBS), respectively; i.e., three replicates at 6, 9, and 12 hours from murine embryonic orofacial tissues treated with AzaD and vehicle (except for the 9 hour vehicle sample which had only two replicates). Before identifying differentially expressed genes between AzaD and vehicle at each time point, data pre-processing was performed. This included normalization and missing value imputation of gene expression data. For normalization, gene expression levels (Cycle threshold; CT) were normalized using the internal control gene (β-actin). Normalized expression levels were calculated by subtracting the CT of β-actin from the CT of the target gene. Genes with more than 3 missing entries in each treatment were removed and the remaining missing values in the data set were imputed using the K-nearest neighbor imputation scheme [42]. This resulted in, a total of 373 genes. After data pre-processing, genes that were differentially expressed when comparing AzaD- and vehicle-treated samples at each time point were identified by fitting to a hierarchical linear model using the limma package [43] and performing hypothesis tests of interest using contrasts (e.g., AzaD vs. vehicle at 6 hours, 9 hours, and 12 hours) for each gene. Fold change, moderated t-statistic, and unadjusted and FDR adjusted p-values were calculated for each gene for each comparison. Genes with unadjusted p-values below 0.05 were considered as being differentially expressed. To classify the patterns of gene expression levels between AzaD and vehicle, overall F-test was performed to detect differentially expressed genes with significant change between AzaD and vehicle across three time points. Differentially expressed genes were classified according to their patterns of gene expression levels over time. The R package limma [44] from the Bioconductor project (http://www.bioconductor.org) was used for all statistical analysis.

Independent single-tube TaqMan qRT-PCR assays were performed as described earlier to assess the expression of a selected set of genes in order to validate the results of the array cards. P-values of <0.05 were considered as indicative of statistical significance. Each experiment was conducted at least three times with comparable results.

3. RESULTS

3.1 Effect of in utero AzaD exposure on embryo growth and development of the murine secondary palate

Treatment of pregnant females with 1 mg/kg AzaD on GD 9.5 did not result in any adverse maternal outcomes as assessed by absence of both weight loss and lethality (data not shown). Similar to studies conducted by others [36,37], statistically significant decreases in average fetal weight and crown-rump length were observed at later stages of embryonic development (GD 15.5 and 17.5, data not shown). Importantly, a single dose of 1 mg/kg AzaD administered on GD 9.5 resulted in 100% cleft palate (Figure 1).

Figure 1. Photomicrographs of ventral views and coronal sections of the developing orofacial region of GD-15.5 mouse fetuses exposed in utero on GD 9.5 to AzaD.

Photomicrographs of ventral views: (A) 1 mg/kg, 5-Aza-2′-deoxycytidine (AzaD); (B) vehicle (PBS). UL = upper lip; P1 = primary palate; P2 = secondary palate. H&E stained coronal sections: (C) displays a failure of palate shelf elevation in AzaD-exposed GD 15.5 fetuses; (D) displays palate fusion in vehicle-exposed GD 15.5 fetuses (control). PS = palatal shelf; OC = oral cavity; T = tongue; P = palate.

3.2 Effect of in utero AzaD exposure on cellular proliferation within the developing first branchial arches

Since AzaD is known to inhibit cell proliferation [36, 33], we investigated whether the decreased fetal weight and crown-rump length in AzaD-exposed embryos was due to reduced cellular proliferation. Phosphorylation of histone H3 is tightly correlated with chromosome condensation during mitosis [45,46] and therefore a good indicator of cell proliferation. Immunochemical staining with the anti-phospho-histone H3 (aPH3) antibody revealed a marked reduction in cell proliferation (PH3 positive cells) in embryos exposed to AzaD when compared to those exposed to vehicle (Figure 2). Pronounced and time-dependent reductions in cellular proliferation within the developing 1-BAs of 6-, 9- and 12-hr AzaD-exposed embryos were also evident (Figures 3A and 3B). Quantification of integrated intensity of PH3 fluorescence within the 1-BAs revealed a statistically significant diminution in cell proliferation in embryos exposed to AzaD in comparison to their vehicle-exposed counterparts at each time point examined (Figure 3B). This suggests that the observed growth retardation in AzaD-exposed embryos could be due to reduction in cellular proliferation.

Figure 2. In utero exposure to AzaD for 12 hours markedly reduces cell proliferation throughout the developing embryo.

Embryos exposed in utero on GD 9.5 to 1 mg/kg AzaD or vehicle (PBS) for 12 hours, were collected and cryopreserved for frontal/coronal sectioning. Cellular proliferation was evaluated via immunostaining with the anti-phospho-histone H3 (aPH3) antibody. Representative photomicrographs of embryos labeled with aPH3 antibody (in red) are shown. Nuclei were counterstained with DAPI (blue). 1-BA-mx = maxillary component of the 1-BA; 1-BA-mb = mandibular component of the 1-BA. Scale bar=200μm.

Figure 3. In utero exposure of embryos to AzaD reduces cell proliferation within the first branchial arches.

Embryos exposed in utero on GD 9.5 to 1 mg/kg AzaD or vehicle for 6, 9 or 12 hours, were collected and cryopreserved for frontal/coronal sectioning. Cellular proliferation was evaluated via immunostaining with the anti-phospho-histone H3 (aPH3) antibody. (A) Representative photomicrographs of 1-BAs labeled with aPH3 antibody (red); nuclei were counterstained with DAPI (blue). Embryos were exposed to AzaD or vehicle for: 6 hours (A1); 9 hours (A2); and 12 hours (A3). (B) Bar graphs represent quantitative assessment of cellular proliferation within the 1-BAs, as determined by measuring the intensities of PH3 fluorescence using the Metamorph 6.5 software: embryos exposed to AzaD or vehicle for: 6 hours (B1); 9 hours (B2); and 12 hours (B3). Data represent analysis of 3 sections per embryo for 3 embryos per treatment group, per time of exposure. 1-BA-mx = maxillary component of the 1-BA; 1-BA-mb = mandibular component of the 1-BA. * p<0.05, Scale bar=200μm.

3.3 Effect of in utero AzaD exposure on apoptosis within the developing first branchial arches

To assess AzaD-induced apoptosis within 1-BAs, frontal/coronal sections of embryos from AzaD- or vehicle-injected dams for 6-, 9- or 12-hours, were immunostained with anti-activated caspase-3 antibodies. These antibodies recognize the active form of caspase 3 which plays an essential role in apoptosis. Figure 4 illustrates that AzaD significantly increased apoptosis throughout the developing embryo including the 1-BAs where the extent of apoptosis was markedly increased with time of exposure (Figures 5A and 5B).

Figure 4. In utero exposure to AzaD for 12 hours markedly enhances apoptosis throughout the developing embryo.

Embryos exposed in utero on GD 9.5 to 1 mg/kg AzaD or vehicle for 12 hours, were collected and cryopreserved for frontal/coronal sectioning. Cellular apoptosis was evaluated via immunostaining with the anti-activated caspase 3 (aCasp3) antibody. Representative photomicrographs of embryos labeled with aCasp3 antibody (in red) are shown. Nuclei were counterstained with DAPI (blue). 1-BA-mx = maxillary component of the 1-BA; 1-BA-mb = mandibular component of the 1-BA. Scale bar=200μm.

Figure 5. In utero exposure of embryos to AzaD enhances apoptosis within the first branchial arches.

Embryos exposed in utero on GD 9.5 to 1 mg/kg AzaD or vehicle (PBS) for 6, 9 or 12 hours, were collected and cryopreserved for frontal/coronal sectioning. Apoptosis was evaluated via immunostaining with the anti-activated caspase 3 (aCasp3) antibody. (A) Representative photomicrographs of 1-BAs labeled with aCasp3 antibody (in red). Embryos were exposed to AzaD or vehicle for: 6 hours (A1); 9 hours (A2); and 12 hours (A3). (B) The bar graphs represent quantitative assessment of apoptosis within the 1-BA, as determined by measuring the intensities of aCasp3 fluorescence using the Metamorph 6.5 software: embryos exposed to AzaD or vehicle for: 6 hours (B1); 9 hours (B2); and 12 hours (B3). Data represent analysis of 3 sections per embryo for 3 embryos per treatment group, per time of exposure. 1-BA-mx = maxillary component of the 1-BA; 1-BA-mb = mandibular component of the 1-BA. * p<0.05, Scale bar=200μm.

3.4 Effect of in utero AzaD exposure on the expression of genes encoding regulators of proliferation, apoptosis and palate development within the first branchial arches

AzaD altered the expression of ~40 genes after 6- and 9-hr and >180 genes after 12-hr exposure (Tables 1, 2, and 3). The number of upregulated genes increased with the time of exposure to AzaD. Specifically, expression of 10, 27 and 148 separate genes was upregulated following 6-, 9- and 12 hr exposures, respectively. Expression profiling analysis, particularly following the 12-hr AzaD exposure, revealed that genes encoding several members of the TGFβ superfamily, including Tgfβr2, Tgfβr3, Acvr1c, Acvr2b, Bmp-4,-6,-7,-8b, Gdf-7/-10/-11/-15, and Smad-2/-3/-5/-9, and members of the Wnt-β-catenin family (e.g. Wnt-1/-3/-3a/-7a/-8a/-8b/-9a/-16, Fzd-1/-2/-3/-4/-7/-8/-10, Dvl-1/-2/-3, Tcf7, and Sfrp5) were significantly upregulated in the 1-BAs (Table 3). These two signaling cascades are known to play critical roles in orofacial morphogenesis [4]. AzaD also induced differential regulation of several genes encoding regulators of cell cycle/proliferation (e.g. Ccnd1/-d2/-e1, Cdkn1a/-1b/-2a/-2b/-2d, Cdk1/-2/-6/-7, Cdc25a, E2f1/-2/-3/-4, Rbl1, and Myc) and apoptosis (e.g. Apaf1, Bax, Bid, Fas, Ikbkg, Nfkb2, Rela, Tnf, Tnfrsf1a, Tnfrsf10b, Trp53, Mapk8 (Jnk1) and Mapk9 (Jnk2)) within the 1-BAs (Tables 1, 2, and 3). Genes encoding three cyclin dependent kinase (Cdk) inhibitors (Cdkn1a, -2a and -2b; inhibitors of cell proliferation) were markedly upregulated in 1-BAs following AzaD exposure. Cdkn1a displayed a consistent ~7.0-fold upregulation following 6-, 9- and 12-hr exposures, and Cdkn2b exhibited 3.5- and 9.2-fold upregulation following 9- and 12-hr exposures, respectively (Tables 1, 2, and 3). Genes encoding a number of pro-apoptotic factors were also strikingly upregulated in 1-BAs following AzaD exposure. Specifically, Fas (Tnfr6), Tnf, and Tnfrsf10b demonstrated 8.4- and 16.5-fold, 3.6- and 19.4-fold, and 3.6- and 6.5-fold upregulation following 9- and 12-hr exposures, respectively (Tables 2, and 3). This activation of genes encoding various inhibitors of proliferation and pro-apoptotic factors is consistent with the immunohistochemical demonstration of reduction in cell proliferation and enhanced apoptosis within the developing 1-BA tissues following in utero AzaD exposure (Figures 3 and 5, respectively). Ingenuity Pathway Analysis further supported these findings by highlighting cellular processes and signaling cascades (cell cycle, apoptosis, FGF, Wnt, SHH and TGFβ superfamily) (Table S1) populated by differentially expressed genes within the AzaD-exposed (12-hr) 1-BAs (Table 3). Finally, AzaD significantly altered expression of genes encoding a number of transcriptional regulators, documented to be crucial for normal palate development: examples include Meox2, Shox2, Foxc2, Foxe1, Vax1, Lhx6, Tbx22, and Satb2 (Tables 1, 2, and 3).

Table 1.

Genes Differentially Expressed within 1^st Branchial Arches Following 6 hour in utero Exposure to 1 mg/kg AzaD on GD 9.5.

Gene Symbol	Entrez gene name	TaqMan Assay ID	Fold Changea AzaD vs. Vehicle	p-value
Cdkn1a	Cyclin-dependent kinase inhibitor 1A (p21, Cip1)	Mm04205640_g1	6.53	0.00
Ngfr	Nerve growth factor receptor	Mm01309638_m1	4.69	0.00
Ltbp2	Latent transforming growth factor beta binding protein 2	Mm00522058_m1	3.61	0.01
Ltbp4	Latent transforming growth factor beta binding protein 4	Mm00723631_m1	3.44	0.00
Fst	Follistatin	Mm00514982_m1	3.17	0.00
Pygo2	Pygopus family PHD finger 2	Mm03031056_s1	3.16	0.01
Gdf5	Growth differentiation factor 5	Mm00433564_m1	2.83	0.01
Mmp15	Matrix metallopeptidase 15 (membrane-inserted)	Mm00485062_m1	2.82	0.01
E2f1	E2F transcription factor 1	Mm00432936_m1	2.36	0.02
Ltbp3	Latent transforming growth factor beta binding protein 3	Mm00521855_m1	2.20	0.03
Bmpr1b	Bone morphogenetic protein receptor, type IB	Mm00432117_m1	−2.12	0.03
Flnb	Filamin B, beta	Mm01311723_m1	−2.13	0.04
Cdh1	Cadherin 1, type 1, E-cadherin (epithelial)	Mm01247357_m1	−2.19	0.03
Bmp5	Bone morphogenetic protein 5	Mm00432091_m1	−2.20	0.04
Lhx6	LIM homeobox 6	Mm01333348_m1	−2.31	0.04
Atr	ATR serine/threonine kinase	Mm01223626_m1	−2.36	0.04
Ephb1	EPH receptor B1	Mm00557961_m1	−2.37	0.04
Btrc	Beta-transducin repeat containing E3 ubiquitin protein ligase	Mm00477680_m1	−2.39	0.02
Acvr1	Activin A receptor, type I	Mm01331069_m1	−2.43	0.02
Dvl2	Dishevelled segment polarity protein 2	Mm00432899_m1	−2.44	0.04
Mmp16	Matrix metallopeptidase 16 (membrane-inserted)	Mm01210646_m1	−2.55	0.04
Crebbp	CREB binding protein	Mm01342452_m1	−2.58	0.04
Dact1	Dishevelled-binding antagonist of beta-catenin 1	Mm00458118_m1	−2.76	0.04
Csnk1g3	Casein kinase 1, gamma 3	Mm00514183_m1	−2.93	0.01
Pbx1	Pre-B-cell leukemia homeobox 1	Mm04207617_m1	−2.95	0.04
Fgfr2	Fibroblast growth factor receptor 2	Mm01269930_m1	−2.99	0.01
Ctgf	Connective tissue growth factor	Mm01192933_g1	−3.18	0.01
Creb1	Camp responsive element binding protein 1	Mm00501607_m1	−3.19	0.03
Apc	Adenomatous polyposis coli	Mm00545872_m1	−3.24	0.02
Sox2	SRY (sex determining region Y)-box 2	Mm03053810_s1	−3.28	0.03
Hdac9	Histone deacetylase 9	Mm01293999_m1	−3.53	0.02
Wif1	WNT inhibitory factor 1	Mm00442355_m1	−3.55	0.00
Efnb2	Ephrin-B2	Mm01215897_m1	−3.63	0.00
Cdkn1b	Cyclin-dependent kinase inhibitor 1B (p27, Kip1)	Mm00438168_m1	−3.90	0.01
Ccnh	Cyclin H	Mm00445740_m1	−4.07	0.04
Cntn1	Contactin 1	Mm00514374_m1	−4.47	0.00
Nfkb1	Nuclear factor of kappa light polypeptide gene enhancer in B-cells 1	Mm00476361_m1	−4.75	0.04
Rock1	Rho-associated, coiled-coil containing protein kinase 1	Mm00485745_m1	−4.77	0.02
Ltbp1	Latent transforming growth factor beta binding protein 1	Mm00498255_m1	−5.07	0.00
Rbl1	Retinoblastoma-like 1	Mm01250721_m1	−5.43	0.00

^aGene expression from first branchial arch (1-BA) tissue from each of 6 hour 5-Aza-2′-deoxycytidine (AzaD)- and vehicle (PBS)-exposed embryos were filtered, and the average fold change for each gene was calculated for AzaD vs. Vehicle 1-BA tissue. Only those genes that demonstrated a statistically significant (adjusted p < 0.05) increase or decrease in expression for the AzaD vs. Vehicle expression comparison, were included in this table. Note that AzaD vs. Vehicle means that expression following 6 hour exposure to vehicle (PBS) was utilized as the baseline. Thus, negative numbers indicate a decrease in expression, while positive figures indicate an increase in expression

Table 2.

Genes Differentially Expressed within 1^st Branchial Arches Following 9 hour in utero Exposure to 1 mg/kg AzaD on GD 9.5.

Gene Symbol	Entrez gene name	TaqMan Assay ID	Fold Changea AzaD vs. Vehicle	p-value
Ltbp2	Latent transforming growth factor beta binding protein 2	Mm00522058_m1	12.62	0.00
Acta1	Actin, alpha 1, skeletal muscle	Mm00808218_g1	8.63	0.00
Fas	Fas cell surface death receptor	Mm01204974_m1	8.42	0.00
Cdkn1a	Cyclin-dependent kinase inhibitor 1A (p21, Cip1)	Mm04205640_g1	6.92	0.00
Il17f	Interleukin 17F	Mm00521423_m1	4.37	0.03
Mmp9	Matrix metallopeptidase 9 (gelatinase B, 92kda gelatinase, 92kda type IV collagenase)	Mm00442991_m1	4.30	0.00
Gdf15	Growth differentiation factor 15	Mm00442228_m1	3.96	0.00
Fst	Follistatin	Mm00514982_m1	3.85	0.00
Tnf	Tumor necrosis factor	Mm00443260_g1	3.63	0.00
Tnfrsf10b	Tumor necrosis factor receptor superfamily, member 10a	Mm00457866_m1	3.58	0.00
Cdkn2b	Cyclin-dependent kinase inhibitor 2B (p15, inhibits CDK4)	Mm00483241_m1	3.50	0.01
Wnt7a	Wingless-type MMTV integration site family, member 7A	Mm00437354_m1	3.43	0.01
Sfrp5	Secreted frizzled-related protein 5	Mm01194236_m1	3.23	0.02
Has1	Hyaluronan synthase 1	Mm03048195_m1	3.21	0.02
Amhr2	Anti-Mullerian hormone receptor, type II	Mm00513847_m1	3.01	0.02
Wnt9a	Wingless-type MMTV integration site family, member 9A	Mm00460518_m1	2.95	0.01
Icam1	Intercellular adhesion molecule 1	Mm00516023_m1	2.81	0.01
Lefty1	Left-right determination factor 1	Mm03053915_s1	2.51	0.04
Amh	Anti-Mullerian hormone	Mm00431795_g1	2.33	0.01
Cd44	CD44 molecule (Indian blood group)	Mm01277163_m1	2.30	0.02
Sfrp4	Secreted frizzled-related protein 4	Mm00840104_m1	2.26	0.02
Rplp2	Ribosomal protein, large, P2	Mm00782638_s1	2.24	0.02
Nodal	Nodal growth differentiation factor	Mm00443040_m1	2.12	0.03
Wisp1	WNT1 inducible signaling pathway protein 1	Mm00457574_m1	2.10	0.05
Thbs3	Thrombospondin 3	Mm00449802_m1	2.09	0.03
Meox2	Mesenchyme homeobox 2	Mm00801881_m1	2.08	0.05
Ngf	Nerve growth factor (beta polypeptide)	Mm00443039_m1	1.79	0.04
Frzb	Frizzled-related protein	Mm00441378_m1	−1.98	0.04
Wnt7b	Wingless-type MMTV integration site family, member 7B	Mm01301717_m1	−2.11	0.03
Prdm16	PR domain containing 16	Mm00712556_m1	−2.31	0.02
Bmp3	Bone morphogenetic protein 3	Mm00557790_m1	−2.83	0.01
Cntn1	Contactin 1	Mm00514374_m1	−3.03	0.01
Dlx5	Distal-less homeobox 5	Mm00438430_m1	−3.21	0.01
Zic3	Zic family member 3	Mm00494362_m1	−5.17	0.00
Tbx22	T-box 22	Mm00467433_m1	−7.18	0.01
Zic1	Zic family member 1	Mm00656094_m1	−9.85	0.00
Vax1	Ventral anterior homeobox 1	Mm00494120_m1	−12.99	0.00
Mmp10	Matrix metallopeptidase 10 (stromelysin 2)	Mm01168399_m1	−20.51	0.02

^aGene expression from first branchial arch (1-BA) tissue from each of 9 hour 5-Aza-2′-deoxycytidine (AzaD)- and vehicle (PBS)-exposed embryos were filtered, and the average fold change for each gene was calculated for AzaD vs. Vehicle 1-BA tissue. Only those genes that demonstrated a statistically significant (adjusted p < 0.05) increase or decrease in expression for the AzaD vs. Vehicle expression comparison, were included in this table. Note that AzaD vs. Vehicle means that expression following 9 hour exposure to vehicle (PBS) was utilized as the baseline. Thus, negative numbers indicate a decrease in expression, while positive figures indicate an increase in expression.

Table 3.

Genes Differentially Expressed within 1^st Branchial Arches Following 12 hour in utero Exposure to 1 mg/kg AzaD on GD 9.5.

Gene Symbol	Entrez gene name	TaqMan Assay ID	Fold Changea AzaD vs. Vehicle	p-value
Acvr1c	Activin A receptor, type IC	Mm03023957_m1	60.33	0.00
Ltbp2	Latent transforming growth factor beta binding protein 2	Mm00522058_m1	26.25	0.00
Fst	Follistatin	Mm00514982_m1	22.03	0.00
Gdf15	Growth differentiation factor 15	Mm00442228_m1	20.55	0.00
Tnf	Tumor necrosis factor	Mm00443260_g1	19.43	0.00
Amhr2	Anti-Mullerian hormone receptor, type II	Mm00513847_m1	16.73	0.00
Fas	Fas cell surface death receptor	Mm01204974_m1	16.53	0.00
Cntn1	Contactin 1	Mm00514374_m1	13.80	0.00
Mmp8	Matrix metallopeptidase 8 (neutrophil collagenase)	Mm00439509_m1	10.57	0.00
Wnt3a	Wingless-type MMTV integration site family, member 3A	Mm00437337_m1	9.84	0.00
Cdkn2b	Cyclin-dependent kinase inhibitor 2B (p15, inhibits CDK4)	Mm00483241_m1	9.16	0.00
Foxc2	Forkhead box C2 (MFH-1, mesenchyme forkhead 1)	Mm00546194_s1	8.38	0.00
Il17f	Interleukin 17F	Mm00521423_m1	8.21	0.00
Spp1	Secreted phosphoprotein 1	Mm00436767_m1	8.01	0.00
Wnt1	Wingless-type MMTV integration site family, member 1	Mm01300555_g1	7.36	0.00
Sfrp5	Secreted frizzled-related protein 5	Mm01194236_m1	7.00	0.00
Cdkn1a	Cyclin-dependent kinase inhibitor 1A (p21, Cip1)	Mm04205640_g1	6.82	0.00
Adamts8	ADAM metallopeptidase with thrombospondin type 1 motif, 8	Mm00479220_m1	6.72	0.00
Tnfrsf10b	Tumor necrosis factor receptor superfamily, member 10a	Mm00457866_m1	6.48	0.00
Mmp9	Matrix metallopeptidase 9 (gelatinase B, 92kda gelatinase, 92kda type IV collagenase)	Mm00442991_m1	6.21	0.00
Nodal	Nodal growth differentiation factor	Mm00443040_m1	5.91	0.00
Wnt8b	Wingless-type MMTV integration site family, member 8B	Mm00442107_m1	5.86	0.00
Wnt9a	Wingless-type MMTV integration site family, member 9A	Mm00460518_m1	5.78	0.00
Amh	Anti-Mullerian hormone	Mm00431795_g1	5.77	0.00
Gli1	GLI family zinc finger 1	Mm00494654_m1	5.74	0.00
Tgfa	Transforming growth factor, alpha	Mm00446232_m1	5.44	0.00
Wnt7a	Wingless-type MMTV integration site family, member 7A	Mm00437354_m1	5.33	0.00
Bid	BH3 interacting domain death agonist	Mm00432073_m1	5.30	0.00
Wnt8a	Wingless-type MMTV integration site family, member 8A	Mm01157914_g1	5.25	0.00
Gdf10	Growth differentiation factor 10	Mm01220860_m1	5.23	0.00
Wisp1	WNT1 inducible signaling pathway protein 1	Mm00457574_m1	5.22	0.00
Nog	Noggin	Mm01297833_s1	4.90	0.00
E2f1	E2F transcription factor 1	Mm00432936_m1	4.34	0.00
Dvl1	Dishevelled segment polarity protein 1	Mm00438592_m1	4.24	0.00
Bax	BCL2-associated X protein	Mm00432051_m1	4.23	0.00
Ptch2	Patched 2	Mm00436047_m1	4.11	0.00
Cdkn2a	Cyclin-dependent kinase inhibitor 2A	Mm00494449_m1	3.99	0.00
Vtn	Vitronectin	Mm00495976_m1	3.95	0.00
Tcof1	Treacher Collins-Franceschetti syndrome 1	Mm00448977_m1	3.90	0.00
Chrd	Chordin	Mm00438203_m1	3.82	0.00
Myc	V-myc avian myelocytomatosis viral oncogene homolog	Mm00487804_m1	3.77	0.00
Ccnd2	Cyclin D2	Mm00438070_m1	3.72	0.00
Inhba	Inhibin, beta A	Mm00434339_m1	3.63	0.00
Ptch1	Patched 1	Mm00436026_m1	3.63	0.00
Bmp8b	Bone morphogenetic protein 8b	Mm00432115_g1	3.62	0.00
Ski	SKI proto-oncogene	Mm00448744_m1	3.53	0.00
Gdf7	Growth differentiation factor 7	Mm04213617_sH	3.50	0.00
Fzd1	Frizzled class receptor 1	Mm00445405_s1	3.45	0.00
Hhip	Hedgehog interacting protein	Mm00469580_m1	3.42	0.00
Shox2	Short stature homeobox 2	Mm00443183_m1	3.38	0.00
Tcf7	Transcription factor 7, T cell specific	Mm00493445_m1	3.29	0.02
Icam1	Intercellular adhesion molecule 1	Mm00516023_m1	3.28	0.00
Fzd8	Frizzled class receptor 8	Mm00433419_s1	3.25	0.00
Cdk6	Cyclin-dependent kinase 6	Mm01311342_m1	3.22	0.00
Bmp4	Bone morphogenetic protein 4	Mm00432087_m1	3.16	0.00
Thbs3	Thrombospondin 3	Mm00449802_m1	3.14	0.00
Gusb	Glucuronidase, beta	Mm00446953_m1	3.13	0.00
Bmper	BMP binding endothelial regulator	Mm01175806_m1	3.10	0.00
Pecam1	Platelet/endothelial cell adhesion molecule 1	Mm01242584_m1	3.07	0.00
Has1	Hyaluronan synthase 1	Mm03048195_m1	3.07	0.02
Wnt3	Wingless-type MMTV integration site family, member 3	Mm00437336_m1	3.02	0.01
Fgf4	Fibroblast growth factor 4	Mm03053741_s1	3.02	0.01
Tnfrsf25	Tumor necrosis factor receptor superfamily, member 25	Mm01263821_m1	2.98	0.01
Cd44	CD44 molecule (Indian blood group)	Mm01277163_m1	2.91	0.00
Rela	V-rel avian reticuloendotheliosis viral oncogene homolog A	Mm00501346_m1	2.91	0.00
Nfkb2	Nuclear factor of kappa light polypeptide gene enhancer in B-cells 2 (p49/p100)	Mm00479810_g1	2.89	0.00
Cdc25a	Cell division cycle 25A	Mm00483162_m1	2.88	0.00
Fzd7	Frizzled class receptor 7	Mm00433409_s1	2.88	0.00
Apaf1	Apoptotic peptidase activating factor 1	Mm01223702_m1	2.88	0.00
Ccne1	Cyclin E1	Mm00432367_m1	2.81	0.00
Rplp2	Ribosomal protein, large, P2	Mm00782638_s1	2.81	0.01
Gapdh	Glyceraldehyde-3-phosphate dehydrogenase	Mm99999915_g1	2.79	0.00
Tnfrsf1a	Tumor necrosis factor receptor superfamily, member 1A	Mm00441883_g1	2.73	0.01
Slc9a3r1	Solute carrier family 9, subfamily A (NHE3, cation proton antiporter 3), member 3 regulator 1	Mm00488865_m1	2.73	0.03
Col2a1	Collagen, type II, alpha 1	Mm01309565_m1	2.68	0.01
Adamts1	ADAM metallopeptidase with thrombospondin type 1 motif, 1	Mm00477355_m1	2.62	0.01
Meox2	Mesenchyme homeobox 2	Mm00801881_m1	2.60	0.01
Fzd2	Frizzled class receptor 2	Mm02524776_s1	2.58	0.01
Bmp7	Bone morphogenetic protein 7	Mm00432102_m1	2.54	0.01
Ikbkg	Inhibitor of kappa light polypeptide gene enhancer in B-cells, kinase gamma	Mm00494927_m1	2.51	0.01
Senp2	SUMO1/sentrin/SMT3 specific peptidase 2	Mm00472489_m1	2.50	0.02
Tnc	Tenascin C	Mm00495662_m1	2.49	0.01
Foxe1	Forkhead box E1 (thyroid transcription factor 2)	Mm00845374_s1	2.47	0.02
Zic1	Zic family member 1	Mm00656094_m1	2.46	0.02
Axin2	Axin 2	Mm00443610_m1	2.46	0.01
Tgfbr3	Transforming growth factor, beta receptor III	Mm00803538_m1	2.45	0.02
Csnk1g1	Casein kinase 1, gamma 1	Mm00557447_m1	2.44	0.01
Cdk2	Cyclin-dependent kinase 2	Mm00443947_m1	2.43	0.01
Axin1	Axin 1	Mm01299063_m1	2.43	0.01
Zic4	Zic family member 4	Mm00657066_m1	2.42	0.04
Chd7	Chromodomain helicase DNA binding protein 7	Mm01219527_m1	2.40	0.01
Dvl2	Dishevelled segment polarity protein 2	Mm00432899_m1	2.40	0.02
Ccnd1	Cyclin D1	Mm00432359_m1	2.38	0.01
Tgfbr2	Transforming growth factor, beta receptor II (70/80kda)	Mm00436977_m1	2.38	0.03
Porcn	Porcupine homolog (Drosophila)	Mm00450406_m1	2.38	0.02
Sp1	Sp1 transcription factor	Mm00489039_m1	2.37	0.03
Smad3	SMAD family member 3	Mm01170760_m1	2.37	0.02
Pvrl1	Poliovirus receptor-related 1 (herpesvirus entry mediator C)	Mm00445392_m1	2.36	0.02
Gdf11	Growth differentiation factor 11	Mm01159973_m1	2.33	0.01
Trp53	Tumor protein p53	Mm01731290_g1	2.33	0.01
Bmp1	Bone morphogenetic protein 1	Mm00802220_m1	2.33	0.04
Acvr2b	Activin A receptor, type IIB	Mm01348449_m1	2.31	0.02
Mmp14	Matrix metallopeptidase 14 (membrane-inserted)	Mm00485054_m1	2.31	0.02
Fbxw11	F-box and WD repeat domain containing 11	Mm00460241_m1	2.30	0.01
Fgfr1	Fibroblast growth factor receptor 1	Mm00438930_m1	2.28	0.01
Dkk3	Dickkopf WNT signaling pathway inhibitor 3	Mm00443800_m1	2.28	0.02
Pygo2	Pygopus family PHD finger 2	Mm03031056_s1	2.25	0.05
Tle4	Transducin-like enhancer of split 4	Mm01195172_m1	2.24	0.01
Itga5	Integrin, alpha 5 (fibronectin receptor, alpha polypeptide)	Mm00439797_m1	2.22	0.03
Col11a1	Collagen, type XI, alpha 1	Mm00483387_m1	2.22	0.02
Cdkn2d	Cyclin-dependent kinase inhibitor 2D (p19, inhibits CDK4)	Mm00486943_m1	2.21	0.01
Dvl3	Dishevelled segment polarity protein 3	Mm00432914_m1	2.19	0.02
Smad2	SMAD family member 2	Mm00487530_m1	2.18	0.03
Ikbkb	Inhibitor of kappa light polypeptide gene enhancer in B-cells, kinase beta	Mm01222247_m1	2.18	0.02
Csnk2b	Casein kinase 2, beta polypeptide	Mm00487216_m1	2.17	0.02
Hprt	Hypoxanthine phosphoribosyltransferase 1	Mm00446968	2.17	0.04
Hdac7	Histone deacetylase 7	Mm00469527_m1	2.17	0.03
Fzd3	Frizzled class receptor 3	Mm00445423_m1	2.16	0.02
Hdac4	Histone deacetylase 4	Mm01299557_m1	2.16	0.01
Mmp15	Matrix metallopeptidase 15 (membrane-inserted)	Mm00485062_m1	2.16	0.03
Ephb6	EPH receptor B6	Mm00432456_m1	2.14	0.02
Csnk1g2	Casein kinase 1, gamma 2	Mm00506098_m1	2.14	0.02
Gsk3a	Glycogen synthase kinase 3 alpha	Mm01719731_g1	2.14	0.02
Kmt2b	Lysine (K)-specific methyltransferase 2B	Mm01175395_g1	2.14	0.05
Bcl2	B-cell CLL/lymphoma 2	Mm00477631_m1	2.09	0.04
Fadd	Fas (TNFRSF6)-associated via death domain	Mm00438861_m1	2.09	0.03
Csnk2a2	Casein kinase 2, alpha prime polypeptide	Mm01243455_m1	2.08	0.03
E2f4	E2F transcription factor 4, p107/p130-binding	Mm00514160_m1	2.07	0.03
E2f3	E2F transcription factor 3	Mm01138833_m1	2.07	0.03
Tle3	Transducin-like enhancer of split 3	Mm00437097_m1	2.07	0.04
Wnt16	Wingless-type MMTV integration site family, member 16	Mm00446420_m1	2.07	0.02
Hdac5	Histone deacetylase 5	Mm01246076_m1	2.06	0.04
Bcor	BCL6 corepressor	Mm00551516_m1	2.06	0.03
Ppp2r1a	Protein phosphatase 2, regulatory subunit A, alpha	Mm00772799_m1	2.04	0.03
Traf2	TNF receptor-associated factor 2	Mm00801978_m1	2.03	0.05
Cdk1	Cyclin-dependent kinase 1	Mm00772472_m1	2.03	0.03
Fzd10	Frizzled class receptor 10	Mm00558396_s1	2.01	0.03
Fn1	Fibronectin 1	Mm01256744_m1	2.01	0.03
E2f2	E2F transcription factor 2	Mm00624964_m1	2.00	0.04
Ctnnd1	Catenin (cadherin-associated protein), delta 1	Mm01334599_m1	1.99	0.03
Bmp6	Bone morphogenetic protein 6	Mm01332882_m1	1.96	0.04
Ppp2cb	Protein phosphatase 2, catalytic subunit, beta isozyme	Mm00479549_m1	1.94	0.04
B2m	Beta-2-microglobulin	Mm00437762_m1	1.93	0.04
Fnta	Farnesyltransferase, CAAX box, alpha	Mm00514973_m1	1.90	0.04
Fzd4	Frizzled class receptor 4	Mm00433382_m1	1.89	0.04
Dcn	Decorin	Mm00514535_m1	1.87	0.04
Timp3	TIMP metallopeptidase inhibitor 3	Mm00441826_m1	1.85	0.04
Flna	Filamin A, alpha	Mm01187533_m1	1.85	0.04
Cdk7	Cyclin-dependent kinase 7	Mm00802284_m1	1.81	0.05
Kcnj2	Potassium inwardly-rectifying channel, subfamily J, member 2	Mm00434616_m1	−2.10	0.04
Ccnb1	Cyclin B1	Mm03053893_gH	−2.18	0.04
Wif1	WNT inhibitory factor 1	Mm00442355_m1	−2.21	0.02
Satb2	SATB homeobox 2	Mm00507331_m1	−2.35	0.03
Fgfr2	Fibroblast growth factor receptor 2	Mm01269930_m1	−2.41	0.03
Apc	Adenomatous polyposis coli	Mm00545872_m1	−2.60	0.03
Rhoa	Ras homolog family member A	Mm00834507_g1	−2.61	0.05
Pygo1	Pygopus family PHD finger 1	Mm01179015_m1	−2.68	0.01
Csnk1g3	Casein kinase 1, gamma 3	Mm00514183_m1	−2.91	0.00
Cxxc4	CXXC finger protein 4	Mm01320305_m1	−2.93	0.02
Ep300	E1A binding protein p300	Mm00625535_m1	−2.98	0.03
Mapk8	Mitogen-activated protein kinase 8	Mm00489514_m1	−3.00	0.02
Ctgf	Connective tissue growth factor	Mm01192933_g1	−3.01	0.01
Tle1	Transducin-like enhancer of split 1 (E(sp1) homolog, Drosophila)	Mm00495643_m1	−3.21	0.01
Mn1	Meningioma (disrupted in balanced translocation) 1	Mm03038457_m1	−3.24	0.00
Pbx1	Pre-B-cell leukemia homeobox 1	Mm04207617_m1	−3.43	0.01
Irf6	Interferon regulatory factor 6	Mm00516797_m1	−3.44	0.01
Sox2	SRY (sex determining region Y)- box 2	Mm03053810_s1	−3.44	0.02
Mstn	Myostatin	Mm01254559_m1	−3.54	0.00
Lhx6	LIM homeobox 6	Mm01333348_m1	−3.73	0.00
Skp1a	S-phase kinase-associated protein 1	Mm00495559_m1	−3.76	0.02
Mapk9	Mitogen-activated protein kinase 9	Mm00444239_m1	−3.86	0.01
Cul1	Cullin 1	Mm00516318_m1	−4.10	0.00
Smad5	SMAD family member 5	Mm01341687_g1	−4.18	0.01
Rbl1	Retinoblastoma-like 1	Mm01250721_m1	−5.91	0.00
Smad9	SMAD family member 9	Mm00649885_m1	−6.07	0.00
Smurf2	SMAD specific E3 ubiquitin protein ligase 2	Mm03024086_m1	−6.97	0.01
Ltbp1	Latent transforming growth factor beta binding protein 1	Mm00498255_m1	−7.16	0.00
Ccnh	Cyclin H	Mm00445740_m1	−7.50	0.00
Cdkn1b	Cyclin-dependent kinase inhibitor 1B (p27, Kip1)	Mm00438168_m1	−8.54	0.00
Rock1	Rho-associated, coiled-coil containing protein kinase 1	Mm00485745_m1	−8.92	0.00
Vax1	Ventral anterior homeobox 1	Mm00494120_m1	−11.78	0.00
Bmp10	Bone morphogenetic protein 10	Mm01183889_m1	−16.28	0.02

^aGene expression from first branchial arch (1-BA) tissue from each of 12 hour 5-Aza-2′-deoxycytidine (AzaD)- and vehicle (PBS)-exposed embryos were filtered, and the average fold change for each gene was calculated for AzaD vs. Vehicle 1-BA tissue. Only those genes that demonstrated a statistically significant (adjusted p < 0.05) increase or decrease in expression for the AzaD vs. Vehicle expression comparison, were included in this table. Note that AzaD vs. Vehicle means that expression following 12 hour exposure to vehicle (PBS) was utilized as the baseline. Thus, negative numbers indicate a decrease in expression, while positive figures indicate an increase in expression.

Results of gene expression profiling were independently validated using TaqMan qRT-PCR [47]. Expression levels of 12 candidate genes, exhibiting differential expression in AzaD-exposed 1-BAs, were quantified and compared to those determined by TaqMan Low-Density Arrays. Expression profiles of all genes tested were found to be in consistent agreement between the two methods (Table 4).

Table 4.

TaqMan Verification of Differentially Expressed Genes in the First Branchial Arches of GD 9.5 Murine Embryos Exposed in utero to 1 mg/kg AzaD or PBS (vehicle) for 6-, 9- and 12-hrs.

Gene Name	Individual TaqMan Assaya,b,c			Array Card Data
	Fold Changed AzaD vs. Vehicle			Fold Changed AzaD vs. Vehicle
	6 hr	9 hr	12 hr	6 hr	9 hr	12 hr
Cdkn1b	−1.30±0.05§	−1.95±0.02	−1.20±0.03	−3.90±0.12	−4.00±0.80§	−8.54±0.08
Fgfr2	NC	−1.52±0.05	−1.54±0.08	−3.00±0.13	−2.11±0.82§	−2.40±0.11
Ltbp2	1.30±0.53§	14.00±0.35	11.00±2.02	3.60±0.44	12.60±1.34	26.30±2.65
Fas	6.63±0.89§	7.23±1.65	34.00±3.14	3.01±1.93§	8.40±1.17	16.50±2.06
Sfrp5	NC	2.35±0.04	7.20±0.84	NC	2.26±0.33	7.00±0.94
Wnt9a	1.20±0.04	1.91±0.12	3.14±0.03	2.11±0.07§	3.00±0.98	5.80±1.03
Lhx6	−1.73±0.11	−4.06±1.05	−6.15±0.87	−2.31±0.02	−2.02±0.99§	−3.73±0.57
Tbx22	−1.35±0.03§	−4.40±0.33	−8.40±1.32	3.40±0.07	−7.20±1.07	−2.60±0.05
Satb2	−1.33±0.15§	−1.74±0.35	−1.90±0.18	−2.30±0.66	−3.01±1.76§	−2.35±0.89
Tnfrsf10b	2.40±0.22§	4.90±0.69	3.90±0.35	2.05±0.04	3.60±0.62	6.50±1.12
Zic1	NC	−8.90±2.02	1.70±0.17	NC	−9.90±2.01	2.50±0.15
Ccnh	−1.34±0.02§	−1.26±0.12	−1.30±0.01	−4.70±0.78	−5.03±2.06§	−7.50±1.11

Gene expression was compared using TaqMan Low-Density Array Fluidic cards and individual (single-tube) TaqMan quantitative real-time PCR as detailed in Materials and Methods.

^acDNA samples were prepared from PBS (vehicle control)- and 1 mg/kg AzaD-exposed first branchial arches and subjected to TaqMan® quantitative real-time PCR (qRT-PCR) for each target gene. Analyses were performed in triplicate using data from three independent experiments.

^bCt values represent the number of cycles during the exponential phase of amplification necessary to reach a predetermined threshold level of PCR product as measured by fluorescence. The more template present at the start of a reaction, the fewer the cycles required to synthesize enough fluorescent product to be recorded as statistically above background. All data were normalized to the amplification signal from the housekeeping gene, 18S rRNA or β-actin. The ΔCt values represent these normalized signals,

ΔCt = C_{t sample} − C_{t18S rRNA/β-actin}. Data presented represent mean ΔCt ± standard deviation over three replicates.

^cNegative methodological control reactions, which lacked reverse transcriptase, did not amplify any detectable product.

^dFold-change (FC) values were determined according to the relationship: FC = 2^−ΔΔCt, where ΔΔCt is the difference in ΔCt values between AzaD-exposed 1-BA samples vs. vehicle (PBS)-exposed 1-BA samples [47]. Statistical analysis comparing the two exposures (vehicle and AzaD) was done with one-way ANOVA of the ΔCt values and adjustment for multiple comparisons using Dunnett’s method. 95% confidence intervals for the FC were calculated by taking the appropriate transformation of the 95% confidence limits for the estimated difference in ΔCt values. A negative fold-change value indicates down regulation of gene expression relative to control samples and a positive fold-change value indicates up regulation. NC: represents no change in gene expression. Only those genes that demonstrated a statistically significant (adjusted p < 0.05) increase or decrease in expression for the AzaD vs. vehicle (PBS) expression comparison, were included in this table.

^§represents fold changes that are not statistically significant (p > 0.05).

3.5 CpG methylation analysis by pyrosequencing

DNA methylation analysis by pyrosequencing was utilized in order to determine if AzaD-induced alterations in CpG methylation levels were associated with altered expression of genes in 12-hr AzaD-treated 1-BAs (Table 3). CpG methylation levels in each of 12 genomic regions (promoters and/or gene bodies) of 10 genes that displayed significant changes in expression levels were determined (Table S2-“pyrosequencing data”). CpG methylation levels in vehicle-treated samples for 9 regions in 8 genes were quite low (<15%), thereby precluding assessment of the effects of AzaD-induced demethylation. However, specific CpG residues in the Cdkn1a promoter and in the promoter and gene body of TNF were moderately to highly methylated in vehicle-treated 1-BAs. Treatment with AzaD had little or no effect on the CpG methylation levels of these residues, except for the methylation of CpG3 in the gene body of TNF which increased from 39% to 52% (highlighted in Table S2-“pyrosequencing data”).

4. DISCUSSION

Previous studies documented clefting of the secondary palate in developing murine fetuses following intrauterine exposure during early gestation (GD 9.0 – 9.5) to the DNA methylation inhibitor and cytidine analog, 5-Aza-2′-deoxycytidine (AzaD) [36,37]. The present study corroborated and extended these results. Murine embryos exposed in utero to 1 mg/kg AzaD on GD 9.5, exhibited nearly 100% cleft palate with palatal processes that were considerably smaller (in comparison to those of the control embryos) and remained vertically oriented on either side of the tongue. Significant temporal (6- to 12-hr) reductions in cellular proliferation and increased apoptosis were clearly evident within the 1-BAs of fetuses exposed in utero to AzaD. Since 1-BAs give rise to structures of the orofacial region including the secondary palate, it is plausible that abnormalities in cellular proliferation and survival within these precursor tissues would adversely affect palatogenesis. Supportive of this premise are data demonstrating that expression of numerous genes encoding regulators of cell proliferation and apoptosis (Table S1), were significantly altered in AzaD-exposed 1-BA tissues (see “Results”). Among these, expression of genes encoding the three Cdk inhibitors (e.g. Cdkn1a, -2a and -2b) was markedly enhanced. This is consistent with the observed AzaD-induced inhibition of cell proliferation within the 1-BAs. Likewise, following in utero AzaD exposure, striking upregulation of genes encoding a number of pro-apoptotic factors (e.g. Fas, Tnf, Tnfrsf10b, Tp53, Bax, Apaf1, Myc and Nfkb2, among others) is also consistent with the observed activation of various apoptosis-inducing canonical cascades (e.g. Apoptosis-, Death receptor-, Myc-mediated apoptosis-, and p53-mediated apoptosis signaling pathways; Table S1). Supportive evidence comes from a previous study reporting AzaD-induced inhibition of proliferation and promotion of apoptosis (via a p53-dependent mechanism) of neural progenitor cells of the fetal CNS [33].

Several earlier studies reinforce the hypothesis that altered cellular proliferation and/or apoptosis within the 1-BA can contribute to palatal clefting [48–50]. Significantly retarded outgrowth (due to compromised cell proliferation) of 1-BA-derived tissues in Wnt9b^−/− mice resulted in a failure of physical contact between the facial processes leading to cleft lip and palate [11]. It is now generally accepted that both FGF [51,52] and Wnt [11,53,54] signaling play a crucial role in proliferation and apoptosis during embryonic orofacial morphogenesis. Single-nucleotide polymorphisms, associated with increased risk for nonsyndromic cleft lip and palate (NSCLP), have been identified in genes of the Fgf/Fgfr family [55,56]. Included among these were Fgf4, Fgfr1 and Fgfr2 whose expression in the 1-BAs was significantly modulated by AzaD in the current study; Table S1). Palatal clefting has also been linked to mutation of genes encoding a number of Wnt signaling modulators/effectors [57–59]. In the present study, expression of genes encoding several such proteins (e.g. Wnt-1/-3/-3a/-7a/-8a/-8b/-9a/-16, Axin-1/-2, Dvl-1/-3, Dkk-3, Fzd-1/-2//-3/-4/-7/-8 and -10, among others) in the 1-BAs, was altered by AzaD (Table 3 and Table S1). Findings from the current study (Table S1) thus reinforce the centrality of the FGF and Wnt signaling pathways in palate development.

In utero AzaD exposure activated genes encoding members of the SHH signaling cascade (Tables S1). This canonical pathway plays fundamental roles in governing critical cellular events, (including proliferation and apoptosis) crucial for development of the 1-BA and palate [4]. Indeed, mutation of the PTCH1 gene (which encodes the principal receptor for the Hedgehog signaling pathway) has been linked to palatal clefting [60]. Moreover, the present study revealed that expression of genes encoding members of the Zic family of zinc-finger transcription factors, that function as Gli co-factors in Hedgehog signaling, was also significantly altered in 1-BAs following AzaD exposure. Zic1 and Zic3 were markedly downregulated (9.9- and 5.2-fold, respectively) at 9-hrs (Table 2), whereas, Zic1 and Zic4 were significantly upregulated (2.5- and 2.4-fold, respectively) at 12-hrs (Table 3) after in utero AzaD exposure. Interestingly, Gli- and Zic family of transcription factors have been implicated in abnormal palatogenesis [61]. Genes encoding four members of the SHH pathway, (e.g. Ptch-1/-2, Hhip and Gli1), whose expression has been reported to be associated with palate development [62], were all significantly upregulated in AzaD-exposed 1-BA tissues (Table 3). These findings underscore the presence of an AzaD-induced deregulation of SHH signaling in 1-BAs — potentially contributing to abnormal palatogenesis.

In utero AzaD exposure also activated genes encoding members of the TGFβ superfamily. This family, comprised of TGFβ, BMP, Nodal, Activin and GDF cytokines, execute diverse physiological functions during orofacial development, governing cell proliferation, differentiation, apoptosis and other tissue remodeling processes [4]. The critical role played by members of the TGFβ superfamily in craniofacial development is documented by the spectrum of craniofacial defects exhibited by mice harboring mutations in genes encoding TGFβ/BMP ligands, receptors or their downstream transcriptional regulators, the Smads [63–65]. A recent study reported that overexpression of the inhibitory Smad, Smad7, in neural crest cells repressed both TGFβ and BMP signaling, augmented apoptosis in branchial arches, and resulted in craniofacial anomalies [66]. Mutations in genes encoding a number of TGFβ superfamily members such as, TGFB3 [67], TGFBR1 (ALK5) or TGFBR2 [68], ALK2/ACVR1 [69], ALK3/BMPR1A [70], BMP2 [71], BMP4 [72], BMP7 [73] and GDF11 [74] have been linked to CP formation. In the present study, genes encoding 40 members of the TGFβ superfamily (noted in “Results” and Table S1), were found to be differentially regulated subsequent to in utero AzaD exposure. Inhibitors/regulators of TGFβ/BMP signaling such as, Follistatin, Noggin, and Chordin were significantly activated. AzaD also stimulated genes encoding the TGFβ-specific R-Smads (Smad-2 and -3) and repressed BMP-specific R-Smads (Smad-5 and -9). These findings underscore contrasting action(s) of AzaD on BMP and TGFβ signaling within 1-BAs, and therefore one can hypothesize that AzaD is potentially inhibiting cellular proliferation by enhancing TGFβ signaling (via upregulation of Smad2 and Smad3) and reducing BMP signal transduction (via downregulation of Smad5 and Smad9). These results highlight the possibility that TGFβ superfamily signaling is yet another potential target of AzaD’s teratogenicity, adversely affecting 1-BA and palate development.

The latent TGFβ binding proteins (LTBPs), regulators of secretion and activation of TGFβs, consist of four isoforms: Ltbp1-4. Although, their precise roles in palate development are unclear, mutations in Ltbp4 have been shown to be associated with craniofacial anomalies [75,76]. In the present study, AzaD exposure had a notable effect on the expression of all four Ltbp genes within 1-BAs: dramatic and consistent upregulation of Ltbp2 (3.6-/12.6-/26.3-fold at 6-/9-/12hr, respectively), significant downregulation of Ltbp1 (5.1- and 7.2-fold at 6- and 12-hr, respectively) and considerable upregulation of Ltbp3 and Ltbp4 (2.2- and 3.4-fold at 6 hr, respectively) (Tables 1, 2 and 3). Significant and substantial differential expression of the 4 Ltbp genes point to abnormal secretion and activation of the TGFβs in AzaD-exposed 1-BA tissues, which could contribute to anomalous palatogenesis.

In addition to the aforementioned growth factors and associated signaling mediators, expression of numerous genes encoding transcriptional regulators was significantly altered in 1-BA tissues following in utero AzaD exposure. Among these, documented to be crucial for palate development, were Foxc2, Foxe1, Shox2, Meox2, Tbx22, and Pbx. Mutations in the genes encoding two Forkhead transcription factors (Foxc2 and Foxe1) have been linked to syndromic and non-syndromic CP, respectively [77,78]. Following 12-hr in utero exposure, AzaD significantly stimulated the expression of both of these genes (Table 3). Proper bone and cartilage formation are critical to normal craniofacial development. Given that bone formation and chondrogenesis are both regulated in part by the functions of both Foxc2 [79] and Foxe1 [80], and that polymorphisms in both of these transcription factors have been linked to facial clefts [81–83], this provides added importance to these two transcription factors as foci for AzaD induced cleft palate. Adding weight to such linkage is the observation that FOXE1 functionally targets MSX1 and TGFβ3 [84], both of which play key roles in development of the palate.

Spatially oriented anterior/posterior gene networks have been shown to exist in the developing palate [85]. Indeed, localized expression and function of several anterior- (Shox2 and Msx1) and posterior- (Barx1, Meox2 and Tbx22) specific transcriptional regulators are thought to be responsible for the anterior-posterior patterning vital to morphogenesis of the secondary palate [86,87]. In the present study, AzaD significantly modulated expression of genes encoding several of these transcription factors, upregulating Shox2 (Table 3) and Meox2 (Tables 2 and 3), and downregulating Tbx22 (Table 2), within 1-BAs. This suggests the ability of AzaD to alter genes involved in patterning during 1-BA and palate development.

AzaD exposure significantly downregulated three other genes (Satb2, Vax1 and Lhx6) (Tables 2, 3 and 4) documented to play pivotal roles during palate development. Loss of function of Satb2 (encoding SATB2, a regulator of transcription and chromatin remodeling) results in increased apoptosis in the craniofacial mesenchyme [88]. Our data demonstrating an AzaD-induced cleft palate and down regulation of Satb2 and a coincident increase in apoptosis are consistent with these results. AzaD-induced clefting and down regulation of two LIM homeodomain transcription factors, Lhx6 and Lhx7, in the 1-BA is also consonant with palatal clefting seen in mice lacking both Lhx6 and Lhx7 [8].

Histone modifications have increasingly been cited as contributory in directing gene expression during orofacial ontogenesis [89,90]. Mice harboring mutations in genes encoding histone acetyltransferases (HATs), such as CBP/CREBBP [90] and MYST4 [91], as well as histone deacetylases (HDACs) such as HDAC4 [92], present with palatal dysmorphologies. Our data, demonstrating AzaD-induced clefting and increase in expression of HDACs (Hdac4, Hdac5, and Hdac7) (Table 3) and diminished expression of HATs (Crebbp, Ep300) (Tables 1 and 3, respectively) are consistent with these findings. Increased expression of HDAC would result in increased removal of acetyl groups from histones, allowing the histones to wrap the DNA more tightly with subsequent decrease in transcriptional activation. Decreased expression of histone acetyltransferases (HATs) would likewise result in a decrease in transcriptional activation by diminishing addition of acetyl groups onto histones. This allows the suggestion that AzaD exposure resulted, via two separate mechanisms, in a generalized diminution of transcriptional activation in the embryonic 1-BA.

Expression of several genes (Wnt3, Fzd10, Dvl1, Dvl3, Axin2 and Cdkn2a, among others) activated within 1-BAs subsequent to in utero AzaD exposure, have been documented by others to be modulated via methylation [93]. In order to ascertain if CpG methylation levels in the putative promoters (or gene bodies) of these genes were altered, CpG methylation analysis (pyrosequencing) was performed on 12 genic regions in 10 selected genes displaying some of the most pronounced gene expression changes. Surprisingly, CpG methylation levels remained essentially unaffected by AzaD treatment (Table S2-“pyrosequencing data”). These observations suggest that the effects of AzaD on the expression of the 182 differentially expressed genes in 1-BAs are likely indirect and/or may be due to the fact that the 1-BAs were exposed to the drug only briefly (12 hr). To explore this idea more comprehensively, Methyl Capture-Seq was performed to assess the epigenome of 1-BAs exposed in utero for 12-hr to AzaD or vehicle. The results of this analysis are presented in a following study [94].

5. CONCLUSIONS

A number of critical genes were found to be differentially expressed in 1-BA tissues following 12-hr in utero AzaD exposure. Data presented in the current study suggest a number of possible molecular targets by which AzaD can result in abnormal development of the secondary palate. These include genes that regulate cellular processes (cellular proliferation, cellular apoptosis) and signaling pathways (Shh and TGFβ) known to play roles in palatal ontogeny. Also included are transcription regulators shown to modulate development of the palate (Foxc2, Foxe1, Shox2, Meox2, Tbx22, Pbx1, Satb2, Vax1 and Lhx6) and genes involved in morphogenetic patterning of the palate (Shox2, Meox2, Tbx22). Since AzaD is a potent inhibitor of global DNA methylation, our initial assumption was that differential expression of these genes was functionally linked to alterations in DNA methylation. Analysis of promoter associated CGIs and/or gene bodies of selected differentially expressed genes, exhibiting the largest changes in fold expression, did not, however, reveal significant alterations of methylation of these regions. This suggests the possibility that these genes could be modulated indirectly – e.g., by epigenetic regulation of upstream regulators (e.g., ncRNAs and transcription factors). Indeed, a recent study by Blattler et al. [95] supports this notion. Thus, to more comprehensively assess the demethylating effects of AzaD on the 1-BA epigenome, MethylCap-Seq (MCS) analysis was performed and the results of AzaD-treated 1-BAs were compared to that of vehicle-treated 1-BAs. Findings from this study are presented separately [94].

Highlights.

1. 5-Aza-2′-deoxycytidine (AzaD) exposure induces cleft palate in mouse embryos.

2. AzaD reduces proliferation and induces apoptosis in first branchial arch (1-BA).

3. AzaD-exposed 1-BA tissues exhibit significant changes in gene expression.

4. Changes in gene expression are likely not due to promoter demethylation.

Abbreviations

aCasp3

anti- activated caspase 3

aPH3

anti-phosphohistone H3

1-BA

First branchial arch

AzaD

5-Aza-2′-deoxycytidine

CGI

CpG island

CP

Cleft palate

CL/P

Cleft lip with or without cleft palate

DAPI

4′-6′-diamidino-2-phenylindole

FBS

Fetal bovine serum

GD

Gestational day

IPA

Ingenuity Pathway Analysis

MCS

MethylCap-Sequencing

NSCLP

Non syndromic cleft lip and cleft palate

PFA

Paraformaldehyde

qRT-PCR

Quantitative real-time PCR

TLDA

Taqman Low-density arrays