deficiency and male sterility due to loss of function of RXRb (Krezel et al., 1996). In contrast, mutants exhibited the same cardiac and ocular defects found in RXRa null mutants, with no additional abnormalities observed, suggesting that one copy of RXRa is sufficient for most of the functions of the RXRs. Compound null mutations of RXRa with RARs display a marked synergistic effect, as a large number of developmental defects, found mainly in RAR single and compound mutants, and are recapitulated in specific RXRa/RAR compound mutants. Several malformations are observed only in one type of RXRa/RAR mutant combination, whereas others are observed in several types of RXRa/RAR double knockout mutants. But no such synergy is observed when RXRb or RXRg mutations are combined with any of the RAR mutations (Kastner et al., 1994, 1997a) (Table I). These data suggest that RXRa/RAR heterodimers are essential for most of the events during embryogenesis and fetal development and that RXRa is the main RXR implicated in the developmental functions of RARs.

The requirement of embryonic synthesis of RA for early heart development has also been demonstrated by targeted disruption of the retinaldehyde dehydrogenase 2 (Raldh2) gene (Niederreither et al., 1999). Raldh2 deficiency results in a block in early embryonic RA production, and induces a complete failure of embryo survival and early morphogenesis. The heart of mutant embryos consists of a single, dilated ventricle-like cavity, lacking heart looping and chamber morphogenesis. Maternal RA administration rescued the early morphogenesis. These data demonstrate that local embryonic RA synthesis by Raldh2 is essential for early postimplantation mouse development.

Together the molecular and genetic identification of the RA-signaling pathway has provided valuable insights into the pleiotropic effects of RA, demonstrating a critical role of retinoid receptors in embryonic heart malformation. Interpretation of the data has been hampered by receptor redundancy, and thus the receptor knockouts alone cannot provide definitive answers to RA function.

The significance of RA for early embryogenesis and heart development is most clearly demonstrated in the VAD avian embryo (Dersch and Zile, 1993; Dong and Zile, 1995; Heine et al., 1985; Kostetskii et al., 1998; Thompson et al., 1969). With the VAD avian model, which is devoid of any form of vitamin A from the beginning of fertilization, it is possible to examine morphological, anatomical, and molecular biological aspects that are attributable to RA. The ability to rescue the VAD embryo at a precise time during development makes this model a powerful tool for the elucidation of the physiological functions of RA during early heart development. The VAD quail embryo is grossly abnormal in many developmental aspects, with the misshapen heart positioned to the left side. The heart does not loop, is enlarged and ballooned, has no chambers, is closed at the site of the inflow tract, and the formation of the extraembryonal circulatory system was blocked (Zile, 2004; Zile et al., 2000). An essential aspect of heart development is the establishment of proper heart sidedness. The asymmetry is set up early in embryogenesis and is regulated by many stage-specific genes. The cardio-genic cells do not function appropriately in the absence of vitamin A and migration to either left or right of the midline is random. In the VAD embryo the orientation of the heart is often abnormal, as 72% of embryos had reversed cardiac situs, with only 28% of the embryos demonstrating normal heart positioning. These studies demonstrate a requirement of retinoids for avian heart development and in establishing cardiac left/right asymmetry. Other studies have shown that RA-induced activation of retinoid receptors is required during early avian heart development (Heine et al., 1985; Romeih et al., 2003; Zile et al., 2000). A specific role for RARa2 in cardiac inflow tract morphogenesis and for RARg in cardiac left/right orientation and looping morphogenesis has been demonstrated (Romeih et al., 2003). Blocking the function of RARa2, RARg, and RXRa recapitulates the VAD pheno-type. These studies provide strong evidence that critical RA-requiring developmental events in the early avian embryo are regulated by distinct retinoid receptor-signaling pathways.

2. Retinoic Acid-Induced Heart Malformations

Although RA is required for normal embryonic and fetal development and cardiogenesis, embryonic exposure to an excess of RA leads to abnormal development. The teratogenic effects of a high dose of vitamin A were first demonstrated in pregnant rats (Cohlan, 1953, 1954). RA was shown to be a more potent teratogen than retinol in several animal models (Creech Kraft et al., 1989; Morriss and Steele, 1977; Shenefelt, 1972a). RA has previously been shown to have teratogenic effects on heart development in mammalian embryos (Fantel et al., 1977; Kalter and Warkany, 1961; Robens, 1970; Shenefelt, 1972b; Taylor et al., 1980). Dependent on the species, stage, and mode of administration, excess RA can result in different types of heart malformations. During the early stage of development, excess RA exposure restricted the cardiac progenitor pool and cardiac specification in the zebrafish embryo (Keegan et al., 2005). In the chick embryo, excess RA inhibits normal precardiac mesoderm migration and the formation of the normal heart tube. Similarly, local application of RA to the heart-forming area disrupts the formation of the cardiogenic crescent and subsequent development of a single midline heart tube (Osmond et al., 1991). Late primitive streak-stage chick embryos exposed to excess RA result in cardiac bifida and clustered heart tissue formation (Dickman and Smith, 1996). Exposure of Xenopus embryos to continuous low levels of RA (1 mM), starting at the time of neural fold closure, blocks expression of myocardial differentiation markers and the heart tube fails to loop during subsequent development, never developing into beating tissue (Drysdale et al., 1997). All vertebrates develop with left/right asymmetry with formation of the left/right body axis being a critical early step in embryogenesis. The heart loop is one of the first clearly recognizable morphological asymmetries. RA-induced asymmetry defects in mammalian (hamster, rat, and mouse) embryos have been reported (Fujinaga, 1997).

Excess RA administration can cause situs inversus in avian embryos, resulting in randomization of heart looping and defects in anteroposterior patterning in the mouse embryo (Chazaud et al., 1999; Smith et al., 1997; Wasiak and Lohnes, 1999). These results suggest that alterations of RA signaling affect the left/right situs, as well as heart morphogenesis in embryonic heart development. Excess RA exposure after heart specification results in congenital heart malformations, including ventricular septal defects, double outlet right ventricle, and persistent truncus arteriosus. Transgenic mice that overexpress a constitutively active RARa in fetal ventricles developed a dilated cardiomyopathy. Lesions included biventricular chamber dilation and left atrial thrombosis, the incidence and severity increasing with copy number. Hypertrophic markers (a-skeletal actin and atrial natriuretic factor) were also upregulated. In contrast, animals that overexpressed a constitu-tively active RARa in developing atria and/or in postnatal ventricles developed no signs of malformations (Colbert et al., 1997). The overlap of the teratological symptoms of vitamin A deficiency and excess suggests common targets and an important role for RA in the development of many organs, including the cardiovascular system.

The above observations suggest that RA-mediated signaling pathways are required at early stages of cardiac development to prevent differentiation, support cell proliferation, and control the shape of ventricular myocytes. Both RXRs and RARs participate in the mediation of these functions. In the postdevelopment period, RA-dependent signal transduction appears to preserve the normal differentiated phenotype of cardiomyocytes by antagonizing the effect of various hypertrophic stimuli (Wang et al., 2002; Wu et al., 1996; Zhou et al., 1995).

0 0

Post a comment