Hif1α down-regulation is associated with transposition of great arteries in mice treated with a retinoic acid antagonist

Abstract

Background: Congenital heart defect (CHD) account for 25% of all human congenital abnormalities. However, very few CHD-causing genes have been identified so far. A promising approach for the identification of essential cardiac regulators whose mutations may be linked to human CHD, is the molecular and genetic analysis of heart development. With the use of a triple retinoic acid competitive antagonist (BMS189453) we previously developed a mouse model of congenital heart defects (81%), thymic abnormalities (98%) and neural tube defects (20%). D-TGA (D-transposition of great arteries) was the most prevalent cardiac defect observed (61%). Recently we were able to partially rescue this abnormal phenotype (CHD were reduced to 64.8%, p = 0.05), by oral administration of folic acid (FA). Now we have performed a microarray analysis in our mouse models to discover genes/transcripts potentially implicated in the pathogenesis of this CHD.

Results: We analysed mouse embryos (8.5 dpc) treated with BMS189453 alone and with BMS189453 plus folic acid (FA) by microarray and qRT-PCR. By selecting a fold change (FC) ≥ ± 1.5, we detected 447 genes that were differentially expressed in BMS-treated embryos vs. untreated control embryos, while 239 genes were differentially expressed in BMS-treated embryos whose mothers had also received FA supplementation vs. BMS-treated embryos. On the basis of microarray and qRT-PCR results, we further analysed the Hif1α gene. In fact Hif1α is down-regulated in BMS-treated embryos vs. untreated controls (FCmicro = -1.79; FCqRT-PCR = -1.76; p = 0.005) and its expression level is increased in BMS+FA-treated embryos compared to BMS-treated embryos (FCmicro = +1.17; FCqRT-PCR = +1.28: p = 0.005). Immunofluorescence experiments confirmed the under-expression of Hif1α protein in BMS-treated embryos compared to untreated and BMS+FA-treated embryos and, moreover, we demonstrated that at 8.5 dpc, Hif1α is mainly expressed in the embryo heart region.

Conclusions: We propose that Hif1α down-regulation in response to blocking retinoic acid binding may contribute to the development of cardiac defects in mouse newborns. In line with our hypothesis, when Hif1α expression level is restored (by supplementation of folic acid), a decrement of CHD is found. To the best of our knowledge, this is the first report that links retinoic acid metabolism to Hif1α regulation and the development of D-TGA.

Figure 1

Graphic illustration of GO classification of up- and down-regulated genes (FC≥ ± 1.5) in BMS-treated (A) and BMS+FA-treated embryos (B).

Figure 2

Quantitative RT-PCR gene expression analysis in BMS treated and BMS+FA-treated embryos. Fold change in gene expression was analysed by the 2-^ΔΔCt (see methods for details) of six selected genes in untreated controls, BMS- and BMS+FA-treated embryos (A) and in BMS+FA vs. BMS-treated embryos (B). Data are mean ± S.E.M., n = 6; * p < 0.05, ** p < 0.01, *** p < 0.0001.

Figure 3

Immunofluorescence results of Hif1α expression in untreated controls (A'), BMS- (B') and BMS+FA-treated (C') embryos at 8.5dpc. Transverse sections of untreated controls (A), BMS-treated (B) and BMS+FA-treated (C) embryos at 8.5 dpc were counterstained with Hoechst to show nuclei (blue). Positive Hif1α staining is evident in the myocardium of untreated controls (A') and BMS+FA-treated embryos (C'). Bc = bulbus cordis; Da = dorsal aorta; Ec = endocardium or endocardial tissue; Fd = foregut diverticulum; Lhsv = wall of the left horn of sinus venosus; Mc = myocardium; Nt = neural tube; Pv = primitive ventricle

Figure 4

Alignment of a common 476 bp sequence in the human HIF1α (top sequence) and mouse Hif1α (bottom sequence) promoter region. The putative DR2 RARE element in the mouse promoter region is underlined, while the putative DR3 RARE element in the human promoter region is grey coloured.

Figure 5

Cited2 expression in treated embryos. Fold change in gene expression analysed by the 2-^ΔΔCt (see methods for details) of Cited2 in untreated controls, and BMS- and BMS+FA-treated embryos (8.5 dpc). Data are mean ± S.E.M., n = 6; *p < 0.05, **p < 0.01, ***p < 0.0001.

Figure 6

Hif1α and Cited2 mRNA expression pattern during normal mouse embryogenesis. Data are mean ± S.E.M., n = 3.