Retinoic acid drives aryl hydrocarbon receptor expression and is instrumental to dioxin-induced toxicity during palate development

Abstract

Background: Palate development depends on complex events and is very sensitive to disruption. Accordingly, clefts are the most common congenital malformations worldwide, and a connection is proposed with fetal exposure to toxic factors or environmental contaminants, such as dioxins. There is increasing evidence that dioxin interferes with all-trans-retinoic acid (atRA), a hormone-like signal derived from vitamin A, which plays an essential role during embryonic development. Although similarities have been described between dioxin-induced toxicity and the outcome of altered atRA signaling during palate development, their relationship needs to be clarified.

Objectives: We used a genetic approach to understand the interaction between atRA and dioxin and to identify the cell type targeted by dioxin toxicity during secondary palate formation in mice.

Methods: We analyzed the phenotype of mouse embryos harboring an atRA-sensitive reporter transgene or bearing null mutations for atRA-synthesizing enzymes (RALDH) or atRA receptors (RAR) and maternally exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) at gestation day 10.5.

Results: We found that an intact atRA signal was required to enable TCDD to induce cleft palate. This mandatory atRA signal was generated through the activity of RALDH3 in the nasal epithelium and was transduced by RARγ (RARG) in the nasal mesenchyme, where it notably controlled aryl hydrocarbon receptor (Ahr) transcript levels. TCDD also did not alter the developmental pattern of atRA signaling during palate formation.

Conclusions: TCDD-induced alteration of secondary palate development in the mouse appears to depend on atRA signaling, which controls AHR expression. This mechanism is likely conserved throughout vertebrate evolution and may therefore be relevant in humans.

Figure 1

TCDD and atRA similarly impair palate development: ventral view of palatal regions of GD18.5 WT fetuses treated at GD10.5 with oil vehicle (A), TCDD (B), and atRA (C). Bone was stained with Alizarin red and cartilage with Alcian blue. Both TCDD and atRA induced cleft palates through which one can see the presphenoid and vomer bones. The arrows indicate the length of the palatal processes of maxillary bones. Abbreviations: i, incisive bone; if, incisive foramen; m, maxillary bone; p, palatine bone; ppi, palatal process of incisive bone; ppm, palatal process of maxillary bone; ppp, palatal process of palatine bone; ps, presphenoid bone; v, vomer bone. Bar = 1 mm.

Figure 2

TCDD does not modify the pattern of endogenous atRA signaling. Tg(RARE-Hspa1b/lacZ)12Jrt transgenic embryos were treated at GD10.5 with vehicle (A,C,E) or TCDD (B,D,F), and the outcome on atRA signaling was analyzed after 6 hr (A, B) and 24 hr (C–F). A–D are external views, and E and F are frontal histological sections of embryos after XGal staining (blue) to reveal Tg(RARE-Hspa1b/lacZ)12Jrt activity, which indicates atRA-responsive cells. Vehicle- and TCDD-treated fetuses show identical patterns of atRA activity, indicating that the dynamics and extent of atRA signaling remain unaltered upon TCDD treatment. Abbreviations: e, eye; ec, ectoderm; f, forebrain; lm, lateral mesenchyme of the frontonasal region; lng, lacrimonasal groove; mm, medial mesenchyme of the frontonasal region; mx, maxillary prominence of first branchial arch; ne, nasal epithelium; nr, nasal region. Bar in F = 250 μm for A and B, 360 μm for C and D, 200 μm for E and F.

Figure 3

TCDD does not affect atRA-dependent gene expression, shown by relative mRNA levels for atRA target genes (LacZ, Crabp2, Rara, Rarg, and Rxra) and a TCDD-target gene (Cyp1a1) in GD11.5 nasopalatal regions of Tg(RARE-Hspa1b/lacZ)12Jrt embryos treated with vehicle or TCDD. Data are mean + SD of triplicates from three nasopalatal regions in each experimental condition. *p < 0.05.

Figure 4

TCDD is unable to induce cleft palate when atRA signaling is impaired. (A) Whole-mount in situ hybridization indicates restriction of Aldh1a3 expression to the nasal epithelium region. (B and C) XGal staining (blue) reveals Tg(RARE-Hspa1b/lacZ)12Jrt transgene activity in WT and Aldh1a3^–/– embryos, illustrating the requirement of only RALDH3 for atRA synthesis in the nasal region. (D–I) Ventral view of palatal regions (Alizarin red/Alcian blue staining) of GD18.5 fetuses treated at GD10.5 with TCDD. Rara^–/– mutants (D) displayed cleft palate, whereas Rarg^–/– (E) and Aldh1a3^–/– (F) mutants were resistant to this malformation. Administration of low doses of atRA (2 mg/kg every 12 hr from GD8.5 to GD12.5) failed to induce cleft palate in WT (G) and Aldh1a3^–/– mutant (H) embryos but was sufficient to restore TCDD toxicity for palate development in Aldh1a3^–/– mutants (I). Abbreviations: i, incisive bone; if, incisive foramen; lng, lacrimonasal groove; m, maxillary bone; nr, nasal region; p, palatine bone; ppi, palatal process of incisive bone; ppm, palatal process of maxillary bone; ppp, palatal process of palatine bone; ps, presphenoid bone; v, vomer bone. Bar in I = 300 μm for A–C and 1 mm for D–I.

Figure 5

AHR expression depends upon RALDH3 in the frontonasal region. (A and B) Relative Ahr mRNA levels in GD11.5 nasopalatal regions of WT, Aldh1a3^–/– and atRA-rescued (2 mg/kg) Aldh1a3^–/– embryos (A) and in WT or RAR-deficient MEFs treated with vehicle or atRA (B). Data are the mean + SD of triplicates from at least three samples (nasopalatal regions or cell cultures) in each experimental condition. (C and D) Immunohistochemistry showing RARG protein localization in the mesenchyme surrounding the nasal epithelium in WT embryo (C). An Rarg^–/– embryo was used as a negative control for immunostaining (D). Abbreviations: f, forebrain; lm, lateral mesenchyme of the frontonasal region; lng, lacrimonasal groove; mm, medial mesenchyme of the frontonasal region; mx, maxillary prominence of first branchial arch; ne, nasal epithelium. Bar = 200 μm. *p < 0.05 in ligand-treated versus vehicle-treated samples.