SEARCH

SEARCH BY CITATION

Keywords:

  • retinoids;
  • head development;
  • forebrain;
  • epiblast;
  • yolk sac;
  • vascular development;
  • endothelium;
  • CYP26A1;
  • CYP26B1

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Retinoic acid (RA) has been implicated as one of the signals providing a posterior character to the developing vertebrate central nervous system. Embryonic RA first appears in the posterior region of the gastrulating embryo up to the node level, where it may signal within the adjacent epiblast and/or newly induced neural plate to induce a hindbrain and spinal cord fate. Conversely, rostral head development requires forebrain-inducing signals produced by the anterior visceral endoderm and/or prechordal mesoderm, and there is evidence that RA receptors must be in an unliganded state to ensure proper head development. As RA is a diffusible lipophilic molecule, some mechanism(s) must therefore have evolved to prevent activation of RA targets in anterior regions of the embryo. This might result from RA catabolism mediated by the CYP26A1 oxidizing enzyme, which is transiently expressed in anteriormost embryonic tissues; however, previous analysis of Cyp26a1−/− mouse mutants did not clearly support this hypothesis. Here we show that Cyp26a1−/− null mutants undergo head truncations when exposed to maternally-derived RA, at doses that do not affect wild-type head development. These anomalies are linked to a widespread ectopic RA signaling activity in rostral head tissues of CYP26A1-deficient embryos. Thus, CYP26A1 is required in the anterior region of the gastrulating mouse embryo to prevent teratological effects that may result from RA signaling. We also report a novel role of CYP26A1 during early development of the intra- and extra-embryonic vascular networks. Developmental Dynamics 236:644–653, 2007. © 2007 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Half a century ago, P. Nieuwkoop proposed an “activation/transformation” model for regionalization of the vertebrate central nervous system (CNS), in which the early neurectoderm is first endowed with an anterior (forebrain) fate, and additional (“transforming”) signals emanating from the organiser (Hensen's node) and/or the embryonic mesoderm are required to induce more posterior (hindbrain and spinal cord) fates (Nieuwkoop et al.,1952; Nieuwkoop and Nigtevecht,1954). This model has been challenged by recent studies indicating that, in mouse and rabbit, the anterior visceral endoderm (AVE) might act as a distinct head and forebrain organiser (reviewed in Beddington and Robertson,1998; Knoetgen et al.,1999). A characterization of the effects of the equivalent tissue in chick (the hypoblast) argued, however, against a role as a true forebrain organizer and indicated that an important function of the hypoblast is to control cell movements driving the early neural-induced (prospective forebrain) cells away from the influence of posteriorizing signals produced in the node region (Foley et al.,2000). These various data have been reconciled in a modified version of Nieuwkoop's model (see Foley et al.,2000; Stern,2001).

In 1989, Nieuwkoop and colleagues were the first to implicate retinoic acid (RA), the active vitamin A derivative, as a potential “posteriorizing” molecule (Durston et al.,1989). Their study, as well as subsequent reports (Sive et al.,1990; Holder and Hill,1991; Simeone et al.,1995; Zhang et al.,1996; and refs. therein), showed that, when administered during gastrulation to Xenopus, zebrafish, or mouse embryos, RA can induce a dose-dependent loss of anterior (forebrain and midbrain) structures and an expansion of posterior (hindbrain) structures. The endogenous distribution of embryonic RA is consistent with a role as a posteriorizing signal, as it is first synthesized at the primitive streak-stage within the mesoderm up to the node level (Ulven et al.,2000; Niederreither et al.,1997; Swindell et al.,1999; Blentic et al.,2003), from which it could diffuse towards the adjacent posterior epiblast (prospective neural plate), which is being specified as hindbrain and spinal cord. Studies in avian and mouse models defective in RA synthesis or signaling have demonstrated its role in generating posterior cell fates within the hindbrain and controlling neurogenesis within the spinal cord (Gale et al.,1999; Niederreither et al.,2000; Dupé and Lumsden,2001; Diez del Corral et al.,2003). More generally, the induction of caudal cell fates in the CNS involves a complex interplay between the Wnt, FGF, and retinoid pathways (Muhr et al.,1999; Nordstrom et al.,2006; reviewed in Diez del Corral and Storey,2004).

Most families of signaling molecules (including FGFs, Wnts, and BMPs) have their range of activity controlled by natural endogenous inhibitors (e.g., Balemans and Van Hul,2002; Kawano and Kypta,2003; Kim and Bar-Sagi,2004). As RA is a small lipophilic, diffusible molecule, its role as a regional determinant may not be properly achieved without additional mechanism(s) that would restrict its activity in the early embryo, e.g., by protecting the anterior epiblast and/or neurectoderm from its diffusion from nearby mesoderm. A combination of biochemical approaches has provided evidence that in Xenopus, proper head formation requires RA receptors to be in an unliganded repressing state (Koide et al.,2001). The P450 CYP26 cytochromes have been identified as enzymes that metabolize RA in vivo (White et al.,1997; Fujii et al.,1997; reviewed in Luu et al.,2001). Among these, CYP26A1 has previously been shown to be required during mouse tail bud development, most likely to prevent inappropriate RA signaling in the caudal stem zone (Sakai et al.,2001; Abu-Abed et al.,2001,2003). During gastrulation, Cyp26a1 is also strongly, although transiently, expressed in the anterior epiblast and early neural plate (Fujii et al.,1997). This rostral expression disappears at the neural fold stage, and persists only in prospective rhombomere 2 thereafter.

The functional significance of CYP26A1 early anterior expression has remained unclear in the mouse, as Cyp26a1−/− mutants do not consistently exhibit head defects except for a relatively subtle posteriorization of rhombomeres 2–3 and sometimes an exencephaly (Sakai et al.,2001; Abu-Abed et al.,2001). However, Cyp26a1 early rostral expression is conserved among vertebrate embryos (Swindell et al.,1999; Kudoh et al.,2002; Blentic et al.,2003). Interestingly, a morpholino “knock-down” of zebrafish Cyp26a1 led to anteroposterior shifts in gene expression reminiscent of those caused by excess RA exposure (Kudoh et al.,2002), although the recently described zebrafish Cyp26a1 loss of function mutant Giraffe does not exhibit obvious defects of rostral head structures (Emoto et al.,2005).

Here we have analyzed Cyp26a1−/− mouse mutants and have used subteratogenic maternal RA supplementation as a system to increase, in a dose- and stage-controlled manner, the RA levels reaching the embryo in order to unveil possible phenotypic defects that may not have been detected (or may occur with a low penetrance) under normal retinoid dietary conditions. We show that CYP26A1 has a critical function during gastrulation, to protect the anterior region of the embryo from teratological effects of RA. We also provide evidence for a role of CYP26A1 in regulating the development of the intra- and extra-embryonic vasculature.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

CYP26A1 and Early Head Development

To investigate whether Cyp26a1−/− embryos could be abnormally sensitive to an elevation of RA levels during gastrulation, pregnant mice from Cyp26a1+/− intercrosses were administered a single dose of RA at various concentrations by oral gavage (see Experimental Procedures), and embryos were collected at E9.5 for analysis (Fig. 1). As expected (Sakai et al.,2001; Abu-Abed et al.,2001), in control litters that were not administered RA, Cyp26a1−/− embryos had a characteristic abnormal tail region (Fig. 1C, arrowhead) but otherwise were similar to their wild-type (WT) and Cyp26a1+/− littermates, including at the cephalic level (compare to Fig. 1A,B). It has been reported that RA administration to WT embryos at gestational day (E)7.5 at a dose of 20 mg/kg body weight (b.w.) generates abnormalities affecting the entire head and rostral brain (Simeone et al.,1995; and refs. therein). When compared to WT and Cyp26a1+/− littermates (Fig. 1D,E, and data not shown), Cyp26a1−/− null mutants exposed to RA at 20 mg/kg b.w. (not shown) or 10 mg/kg b.w. (Fig. 1F) exhibited an overall growth deficiency, as well as a severe head truncation. Reducing the RA dose to 5 mg/kg b.w. led to externally normal WT and Cyp26a1+/− embryos (Fig. 1G,H), whereas the same severe head phenotype was observed in Cyp26a1−/− mutants (Fig. 1I). By further reducing the RA dose (1 mg/kg b.w.), homozygous Cyp26a1−/− embryos of almost normal size were obtained (Fig. 1J–L). However, in addition to their abnormal caudal phenotype (Fig. 1L, arrowhead), the mutant embryos exhibited an underdeveloped head.

thumbnail image

Figure 1. Administration of RA at subteratogenic doses generates head truncation in Cyp26a1 mouse mutants. Comparison of wild-type (WT), heterozygous (Cyp26a1+/−), and homozygous mutant (Cyp26a1−/−) embryos collected at E9.5 after a single maternal RA administration by oral gavage at E7.5, at various doses (as indicated on the right side) or in the absence of RA treatment (A–C). A dose (10 mg/kg body weight) that leads to relatively mild brain anomalies in WT (D) and heterozygous (E) embryos generates a severe head truncation and an overall growth deficiency in null mutants (F). A 5-mg/kg dosage results in morphologically normal WT (G) and Cyp26a1+/− (H) embryos, while the same severe phenotype is observed in homozygous mutants (I). A lower (1 mg/kg) dose leads to milder head defects in null mutants (J–L).

Download figure to PowerPoint

To further characterize the developmental stages at which head development is RA-sensitive in Cyp26a1−/− embryos, we used another mode of supplementation in which pregnant mothers are given a RA-supplemented food during a defined time window (Niederreither et al.,2002a; see Experimental Procedures). Typical head truncations were observed in Cyp26a1−/− embryos when RA was supplied from E7.5 to E8.5, i.e., during late gastrulation and neurulation. Severe head defects were observed in null mutants (Fig. 2B,C) when using a dosage (100 μg RA/g of maternal food), which does not lead to detectable anomalies in WT embryos (Fig. 2A; see Niederreither et al.,2002a,b). Cyp26a1−/− embryos also exhibited a dilated pericardial sac, as well as an abnormal allantoic bud (Fig. 2B,C; and see below). Interestingly, in these litters some of the heterozygous Cyp26a1+/− mutants exhibited a mild head hypoplasia, suggesting that a decreased CYP26A1 activity due to haploinsufficiency cannot clear all excess RA in the cephalic region (Fig. 2D). Dose-dependent effects were also seen in null mutant embryos, as progressively lower RA doses led to less severe head defects, i.e., to a microcephalic phenotype affecting all rostral brain regions (Fig. 2E,F). RA supplementation between E6.5 and E7.5 (20 μg/g food) yielded even more drastic effects, such that all Cyp26a1+/− embryos were being resorbed by E9.5 (Fig. 2E, inset).

thumbnail image

Figure 2. Stage- and dose-dependent effects of RA-supplemented maternal food on Cyp26a1 mutants. Embryos were collected at E9.5, following supplementation of the maternal food during defined 24-h intervals. RA supplementation from E7.5 to E8.5 at a dose (100 μg/g food) that is not detectably harmful for WT embryos (A) generates head truncation in Cyp26a1−/− mutants (B,C). Note also the overall growth deficiency, dilated pericardial sac (ps), and abnormal allantoic bud (ab) in mutants. Under such conditions, some of the heterozygous mutants exhibited a mild truncation of rostral head structures, as well as pericardial dilation (D). Supplementation with lower RA doses (20 and 5 μg/g) led to normal WT and heterozygous embryos (not shown), and to progressively less severe head truncation in Cyp26a1−/− null mutants (E,F). RA supplementation from E6.5 to E7.5 (20 μg/g) led to almost fully resorbed Cyp26a1−/− embryos (inset in E).

Download figure to PowerPoint

We next investigated whether the abnormal cephalic phenotype observed in Cyp26a1−/− embryos correlated with ectopic RA signaling in the head region, by analyzing mice carrying the RA-responsive RARE-hsp68-lacZ reporter transgene (Rossant et al.,1991). The pattern of activity of this transgene in WT embryos matches the distribution of endogenous RA (Ulven et al.,2000) and of its synthesizing enzyme Raldh2 (Niederreither et al.,1997). No activity of the RA-reporter transgene was seen in rostral head tissues of WT embryos at early somite-stages (Rossant et al.,1991). Under normal dietary conditions, Cyp26a1−/− embryos displayed a relatively limited rostral expansion of RARE-hsp68-lacZ transgene activity in the hindbrain neurepithelium, which did not extend beyond the pre-otic sulcus (Sakai et al.,2001; Niederreither et al.,2002c). Embryos from Cyp26a1+/− intercrosses carrying the RARE-hsp68-lacZ transgene were collected at E8.5 after administration of a single RA dose (1 mg/kg) by oral gavage at E7.5, and analyzed by X-gal assay (Fig. 3A,B). This dose did not lead to detectable activation of the reporter transgene rostral to the pre-otic sulcus in WT littermates (Fig. 3A). In Cyp26a1−/− embryos, transgene activity was seen in most of the head, both in mesoderm and neuroepithelium (Fig. 3B). Litters were also analyzed at E9.5, i.e., after the normal endogenous activation of the RARE-hsp68-lacZ transgene in frontonasal tissues (Rossant et al.,1991; Wagner et al.,2000; Fig. 3C). Cyp26a1−/− embryos exhibited an ectopic β-galactosidase activity throughout most of the head tissues (Fig. 3D). These results show that CYP26A1 activity is required to prevent exogenous RA to signal ectopically in rostral head tissues. Ectopic RARE-hsp68-lacZ activity can be induced throughout the head of WT embryos by using higher RA doses (Rossant et al.,1991; Conlon and Rossant,1992), consistent with the idea that CYP26A1 normal metabolizing activity is saturable.

thumbnail image

Figure 3. Cyp26a1 function is required to prevent ectopic RA-induced signaling and Hoxb1 expression in rostral head tissues. A–D: WT and Cyp26a1−/− embryos (genotypes as indicated) harboring a RARE-hsp68-lacZ reporter transgene were collected at E8.5 (A,B) and E9.5 (C,D) after maternal RA administration by oral gavage (1 mg/kg) at E7.5, and analyzed by X-gal staining. Asterisks indicate ectopic transgene activity in the head of mutant embryos, which is not seen in this condition in the control embryos. Arrowheads in A,B point to the pre-otic sulcus. fn, frontonasal region; ot, otocyst. E–G: Whole-mount analysis of Hoxb1 expression in WT and Cyp26a1−/− E8.5 embryos (genotypes as indicated), after two RA dosages (0.2 and 1 mg/kg, maternal gavage) at E7.5. Main panels: profile views; insets: dorsal views of the hindbrain region. The lower RA dose leads to the same molecular phenotype observed in untreated Cyp26a1−/− mutants, i.e., to an ectopic expression in some cells of the prospective rhombomere 2 (arrowheads in F), whereas RA at 1 mg/kg leads to wide ectopic expression up to the prospective midbrain level (H). None of these conditions generates an ectopic Hoxb1 expression in WT embryos (E,G).

Download figure to PowerPoint

Several studies have shown that RA treatment of WT embryos can lead to a rostral ectopic expression of specific hindbrain markers, including the Hoxb1 gene (Conlon and Rossant,1992; Simeone et al.,1995; and refs. therein). Under normal dietary conditons, Cyp26a1−/− embryos exhibited relatively subtle ectopic Hoxb1 expression within the hindbrain, which did not extend rostrally to the prospective rhombomere (r)2 (Sakai et al.,2001; Abu-Abed et al.,2001). We compared the response of Hoxb1 in WT and Cyp26a1−/− embryos, which were given RA at various doses through maternal gavage at E7.5. Doses up to 0.2 mg/kg led in Cyp26a1−/− embryos to the same Hoxb1 pattern as that seen in non-treated mutants, i.e., to a patchy ectopic expression in the r2 territory (Fig. 3F; see Abu-Abed et al.,2001). However, at 1 mg/kg RA, widespread ectopic Hoxb1 expression was seen in Cyp26a1−/− embryos throughout the hindbrain and part of the midbrain region (Fig. 3H). These RA doses did not lead to abnormal Hoxb1 expression in the WT littermate embryos (Fig. 3E,G). Thus, a lack of CYP26A1 activity leads to an ectopic Hoxb1 response to exogenous RA at doses 20 times lower than those affecting WT embryos at E7.5 (see Simeone et al.,1995).

CYP26A1 and Vascular Development

Cyp26a1 expression has been detected in the developing embryonic and extra-embryonic (allantoic) vascular networks of the WT embryo (MacLean et al.,2001). Cyp26a1−/− embryos collected at E8.5 after maternal RA treatments (whether by gavage or through the food supply) often exhibited cysts within the allantois and a dilated pericardial cavity, which may indicate a defective cardiac and/or vascular function (Fig. 4A). At 9.5, a network of extra-embryonic blood vessels had developed in WT embryos (Fig. 4B). Cyp26a1−/− embryos collected after low maternal RA supplementation had a defective yolk sac vascularization, and their yolk sacs contained a mesh of dilated blood sinuses with no formed blood vessels (Fig. 4C). Yolk sac vascular development was further impaired following administration of higher maternal RA doses (Fig. 4D, and data not shown).

thumbnail image

Figure 4. Cyp26a1 function is required for proper vascular development. A: Dilation of the allantoic (al) blood vessels and of the pericardial sac (ps) indicate a defective cardiovascular function in an E8.5 Cyp26a1−/− embryo after maternal dietary RA supplementation (20 μg/g food) from E7.5–E8.5. Note also the severe head (hd) truncation. B–D: Details of the yolk sacs of a WT (B) and two Cyp26a1−/− embryos (C,D) collected at E9.5 after maternal RA supplementation from E7.5 to E8.5 at two dosages (20 and 100 μg/g). Instead of an organized network of vitelline vessels (vv), the yolk sacs of Cyp26a1−/− mutants contain a more or less developed array of unorganized vascular plexi.

Download figure to PowerPoint

A molecular analysis was performed to better characterize the vascular defect of Cyp26a1−/− embryos, following maternal RA food supplementation between E7.5 and E8.5. Flk1, the gene encoding the vascular endothelial growth factor receptor VEGFR2, is one of the earliest markers of vascular progenitor cells (Ema et al.,2006; Fig. 5A). After a 20- or 40-mg/kg RA supplementation, the yolk sacs of E9.5 Cyp26a1−/− embryos contained scattered Flk1-positive cells, which did not form an organized endothelial network (Fig. 5B). PECAM1 immunostaining was used for further analysis of the intra- and extra-embryonic vascular networks (Fig. 5C–H). This analysis showed that the yolk sacs of E9.5 Cyp26a1−/− mutants contained a network of small capillary-like vessels, but lacked large vitelline vessels (Fig. 5D). Intra-embryonic blood vessels had developed in the mutants (Fig. 5F,H). However, the network of small growing vessels was poorly developed, as seen for instance within the head mesenchyme (compare Fig. 5E and F) or the trunk vasculature (compare Fig. 5G and H, arrowheads). PECAM1 labeling also indicated a defective development of the heart endocardium (Fig. 5H).

thumbnail image

Figure 5. Defective blood vessel formation in RA-supplemented Cyp26a1−/− embryos. A,B: Details of the yolk sacs of E9.5 WT (A) and Cyp26a1−/− (B) embryos, after maternal RA dietary supplementation (20 μg/g) from E7.5 to E8.5. Whole-mount in situ hybridization with a Flk1 probe. C–H: Whole-mount PECAM1 immunodetection in E9.5 WT and Cyp26a1−/− embryos (genotypes as indicated), following maternal RA supplementation (10 μg/g) from E7.5 to E8.5 (C,D: yolk sacs; inset shows a detail at higher magnification; E,F: profile views of the head; G,H: profile views of the heart [ht] and trunk). See main text for details.

Download figure to PowerPoint

Whereas murine Cyp26a1 is transiently expressed by E7.5–E8 in the developing endothelial vascular network, its homologue Cyp26b1 has been detected at a later stage (E8–E9.5) and at low levels in some vascular elements, such as the intersomitic vessels (MacLean et al.,2001). Expression of Cyp26b1 is even more prominent in the developing vascular endothelia of the chick embryo, and in this species its expression is upregulated following excess RA administration (Reijntjes et al.,2003,2005). We, therefore, analyzed whether Cyp26b1 expression might be upregulated in the Cyp26a1−/− mutants, either spontaneously or following RA supplementation. Consistent with previous observations (MacLean et al.,2001), Cyp26b1 transcripts were detected at low levels in the developing intersomitic vessels and the heart outflow tract endocardium of E9.5 WT embryos in the absence of RA supplementation (Fig. 6A, compare with the strong labeling seen in hindbrain rhombomeres or forelimb bud). The same expression pattern was observed in WT embryos obtained from RA-supplemented mice, indicating that our supplementation method does not lead to a detectable Cyp26b1 upregulation (Fig. 6B). Cyp26a1−/− embryos in the absence of RA supplementation showed the same expression pattern as WT embryos (Fig. 6C), further indicating that lack of CYP26A1 function does not lead to an upregulation of Cyp26b1 expression under “basal” conditions. However, the RA-treated Cyp26a1−/− mutants consistently showed elevated Cyp26b1 expression in ventral regions of the trunk (Fig. 6D, bracket). This abnormal expression was strongest in the midgut mesoderm, but was also clearly seen along the walls of the developing blood vessels (Fig. 6E,F).

thumbnail image

Figure 6. Abnormal Cyp26b1 expression in intra- and extra-embryonic tissues of RA-supplemented Cyp26a1−/− mutants. A–F: Wild-type (A,B) and non RA-treated E9.5 Cyp26a1−/− embryos (C) show comparable Cyp26b1 transcript patterns, whereas mutants after RA treatment show increased expression in the ventral trunk region (bracket in D). A–D: profile views; E,F: transverse sections at upper thoracic and forelimb levels, respectively. bv, blood vessels; fl, forelimb bud; ht, heart; mg, midgut; r, rhombomeres; sc, spinal cord; sv, somitic vessels. G–J: Details of Cyp26b1 expression in the yolk sacs of E9.5 WT and Cyp26a1−/− embryos (genotypes and RA treatments as indicated above). Insets show the presence of expressing cells in the capillary network. No Cyp26b1-positive cells are detected in the abnormal yolk sacs of the RA-treated Cyp26a1−/− mutants (J).

Download figure to PowerPoint

Examination of the yolk sacs of the corresponding embryos revealed a punctate expression of Cyp26b1 in scattered cells of both the large blood vessels (Fig. 6G–I, main panels) and capillary vascular network (Fig. 6G–I, insets). This expression had not been previously described. The same expression pattern was found in control (non RA-treated) WT embryos (Fig. 6G), as well as in RA-supplemented WT embryos (Fig. 6H) and non RA-treated Cyp26a1−/− mutants (Fig. 6I). However, no Cyp26b1-expressing cells were detected in the yolk sacs of Cyp26a1−/− mutants following RA supplementation (Fig. 6J). Considering our other molecular data (Fig. 5), the most likely explanation is that prospective vascular endothelial cells have not differentiated enough in these abnormal yolk sacs to induce Cyp26b1 expression.

Finally, we also controlled Cyp26a1 expression in both WT and Cyp26a1−/− embryos at E9.5. No expression was detected in yolk sac cells, whatever the genotype or RA-treatment condition (data not shown). Furthermore, there was no evidence at this stage for Cyp26a1 expression in intra-embryonic vascular structures (either in WT or mutants), although it should be mentioned that the in situ hybridization signals were consistently weaker in Cyp26a1−/− embryos, likely due to the lack of several Cyp26a1 exons in the targeted allele (Abu-Abed et al.,2001). Altogether, these results support the conclusion that CYP26A1 function is restricted to early steps of vascular development prior to E9, when it is eventually superseded by CYP26B1, at least in some of the developing vessels (Fig. 6).

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

We have shown here that the RA-metabolizing enzyme CYP26A1 has an important function in the prospective head region of the gastrulating mouse embryo. This function had not been inferred from earlier studies of Cyp26a1−/− knockout mice, as these mutants develop almost normal head structures (Abu-Abed et al.,2001; Sakai et al.,2001). In marked contrast, CYP26A1-deficient embryos undergo head truncations when exposed to maternally-derived RA at doses well below teratogenic levels in WT embryos. These effects are both stage- and gene dosage-dependent, correlating with the transient phase of Cyp26a1 expression in the rostral epiblast and neural plate. These findings provide a functional basis to the widely documented teratological effects that exogenous RA can exert on head development in WT embryos from various species (Durston et al.,1989; Holder and Hill,1991; Simeone et al.,1995; Zhang et al.,1996, and refs. therein). RA affects the development of rostral brain structures when administered during the same critical period (E7.2–7.4 in the mouse; Simeone et al.,1995) and its effects are dose-dependent, with “moderate” RA doses leading to a loss of the caudal midbrain-rostral hindbrain area (Holder and Hill,1991) and higher doses leading to severe microcephaly affecting all rostral brain structures (Durston et al.,1989; Simeone et al.,1995). These different outcomes are consistent with the idea that moderate RA doses lead to partial saturation of CYP26A1 activity, and therefore to a situation of RA excess in the region of the epiblast/neural plate located at the margin between the Cyp26a1-expressing domain and the more posterior RA-producing region (see below), whereas high doses lead to a full saturation of the CYP26A1 enzymatic activity throughout the rostral epiblast, and therefore to drastic RA excess effects.

What is then the functional significance of the CYP26A1 activity that is present in anterior tissues of primitive streak-stage embryos? Embryonic RA is first detected at this stage in WT embryos (Ulven et al.,2000), and its synthesis by RALDH2 in embryonic mesoderm does not extend rostrally to the node (Niederreither et al.,1997, 1999). In avian embryos, RALDH3 may additionally produce RA in the epiblast adjacent to the node (Halilagic et al.,2003; Blentic et al.,2003). Such a distribution is consistent with a role of RA as being one of the signals specifying a posterior character in the epiblast (prospective neural plate) rostrally adjacent to the node, which is fated to become hindbrain and spinal cord. At this stage, the embryo is composed of a few hundred cells, and a small lipophilic molecule such as RA could diffuse towards anterior regions of the epiblast. Furthermore, RA signaling begins while the embryo is undergoing substantial morphogenetic movements. During this phase, signaling from the AVE (equivalent to the chick hypoblast) is likely to mediate cell movements in the epiblast that will drive putative forebrain cells away from the influence of “posteriorizing” signals generated at the node level (Kimura et al.,2000; Foley et al.,2000). There is evidence that rostral head development requires RA receptors to be in an unliganded, repressing state (Koide et al.,2001). CYP26A1 function could be required within anterior epiblast and/or early mesodermal cells before they are protected from the posterior source of RA, in order to avoid RA-mediated gene activation to interfere with the response to signals produced by the hypoblast, and later the prechordal mesendoderm, which are necessary for the induction of a forebrain fate (Foley et al.,1997,2000). Recently, Uehara et al. (2006) have reported that Cyp26a1;Cyp26c1 double mouse mutants exhibit severe head truncations consistent with those described herein, although on its own Cyp26c1 loss of function does not lead to any patterning defect. Thus, both CYP26 enzymes functionally cooperate in regulating RA metabolism in rostral head tissues, although CYP26A1 appears to play a dominant role (Uehara et al.,2006).

Our work has also unveiled a function of CYP26A1 in the development of the mouse extra-embryonic and intra-embryonic vascular networks. CYP26A1 is likely to regulate RA levels in the early vascular endothelium, in which it is specifically expressed between E7.5 and E8.5 (McLean et al.,2001). Expression of its homologue CYP26B1 is detected at later stages (E8.5–9.5) and at low levels, at least in some of the intra-embryonic endothelia (MacLean et al.,2001; Fig. 6). In the course of the present work, we also found a punctate expression of Cyp26b1 along the walls of the yolk sac blood vessels and capillary network (Fig. 6). Cyp26b1 expression in vascular endothelia is actually more prominent in the chick embryo (Reijntjes et al.,2003). In this species, Cyp26a1 expression has not been detected in vascular endothelia under “spontaneous” conditions, but was induced (like that of Cyp26b1) following exposure to excess RA (Reijntjes et al.,2003,2005). Whether these differences in expression patterns do reflect a more prominent role of Cyp26b1 during development of the chick vascular system remains to be demonstrated. Some vascular defects have been described in an ENU-induced zebrafish Cyp26a1 mutant (Emoto et al.,2005). It will be of interest to eventually study vascular development in double Cyp26a1;Cyp26b1 mouse mutants, as the function of both genes in the vascular lineage may be spatially and/or temporally overlapping.

There has been previous evidence that regulation of RA metabolism is important for proper vascular development. Knockout of the Por gene, which encodes an oxidoreductase required for the function of all P450 enzymes, has been found to be early embryonic lethal due to defective development of the vascular system (Otto et al.,2003). It was shown that Por−/− embryos have abnormally high endogenous RA levels, and that their vascular development can be improved by feeding the pregnant mothers with a vitamin A–deficient diet (Otto et al.,2003). We have been further studying Por−/− mutants, and used various approaches to demonstrate that both their vascular and other phenotypic traits are the result of elevated endogenous RA activity (Ribes et al.,2006). Other work performed on Raldh2−/− mouse mutants has led to the conclusion that RA may act during the early steps of vascular development to regulate the proliferation and survival of endothelial precursors, most likely through the TGFβ and fibronectin pathways (Lai et al.,2003; Bohnsack et al.,2004). In the chick embryo, both Cyp26a1 and Cyp26b1 gene expression is strongly upregulated in the vascular endothelium following excess RA application (Reijntjes et al.,2005). Collectively with these data, our results point to a model in which Cyp26a1 and Cyp26b1 gene expression would be induced during normal development of the vascular endothelium after the onset of RA signaling, in order to regulate the intracellular RA levels in a negative feedback mechanism that would allow the sustained branching and growth of both the extra-embryonic and intra-embryonic vascular networks.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Cyp26a1 null mutant mice used in this study have been previously described (Abu-Abed et al.,2001). Two procedures were used for maternal RA supplementation. All-trans-RA (Sigma, St. Louis, MO) was either suspended in ethanol, diluted in sunflower oil (0.5 mg/ml), and administered by oral gavage (1 to 20 mg/kg body weight) to pregnant mice, or RA in ethanol suspension was mixed with powdered food (5 to 100 μg/g food). The RA-containing food mixture (protected from light by aluminium foil) was left for the mice to feed ad libitum for 24 h, after which the mice were placed in a clean cage with non-supplemented food. Whole-mount in situ hybridization (ISH) with digoxigenin-labelled riboprobes was performed with an Intavis InSituPro robot (for a detailed procedure, see http://www.eumorphia.org/EMPReSS/servlet/EMPReSS.Frameset, gene expression section). Whole-mount X-gal assays were performed as described (Rossant et al.,1991). Immunolabelling with rat monoclonal anti-PECAM-1 antibody (Pharmingen) was performed as described (Scardigli et al.,2003), using an alkaline phosphatase conjugated goat anti-rat (Jackson Immuno-research) secondary antibody.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

We thank Prof. P. Chambon for his support and critical comments on the manuscript, as well as Prof. J. Rossant for providing RARE-hsp68-lacZ mice. We thank B. Schuhbaur for expert technical assistance. This work was supported by funds from the CNRS, the INSERM, the Hôpitaux Universitaires de Strasbourg, the Ministère Français de la Recherche (ACI 03-2-490), the Collège de France, and the Institut Universitaire de France (P.D.), and the CIHR and the NCIC (M.P.).

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  • Abu-Abed S, Dollé P, Metzger D, Beckett B, Chambon P, Petkovich M. 2001. The retinoic acid-metabolizing enzyme, CYP26A1, is essential for normal hindbrain patterning, vertebral identity, and development of posterior structures. Genes Dev 15: 226240.
  • Abu-Abed S, Dollé P, Metzger D, Wood C, MacLean G, Chambon P, Petkovich M. 2003. Developing with lethal RA levels: genetic ablation of Rarg can restore the viability of mice lacking Cyp26a1. Development 130: 14491459.
  • Balemans W, Van Hul W. 2002. Extracellular regulation of BMP signaling in vertebrates: a cocktail of modulators. Dev Biol 250: 231250.
  • Beddington RS, Robertson EJ. 1998. Anterior patterning in mouse. Trends Genet 14: 277284.
  • Blentic A, Gale E, Maden M. 2003. Retinoic acid signalling centres in the avian embryo identified by sites of expression of synthesising and catabolising enzymes. Dev Dyn 227: 114127.
  • Bohnsack BL, Lai L, Dolle P, Hirschi KK. 2004. Signaling hierarchy downstream of retinoic acid that independently regulates vascular remodeling and endothelial cell proliferation. Genes Dev 18 13451358.
  • Conlon RA, Rossant J. 1992. Exogenous retinoic acid rapidly induces anterior ectopic expression of murine Hox-2 genes in vivo. Development 116: 357368.
  • Diez del Corral R, Storey K G. 2004. Opposing FGF and retinoid pathways: a signalling switch that controls differentiation and patterning onset in the extending vertebrate body axis. Bioessays 26: 857869.
  • Diez del Corral R, Olivera-Martinez I, Goriely A, Gale E, Maden M, Storey K. 2003. Opposing FGF and retinoid pathways control ventral neural pattern, neuronal differentiation, and segmentation during body axis extension. Neuron 40: 6579.
  • Dupé V, Lumsden A. 2001. Hindbrain patterning involves graded responses to retinoic acid signalling. Development 128: 21992208.
  • Durston AJ, Timmermans JP, Hage WJ, Hendriks HF, de Vries NJ, Heideveld M, Nieuwkoop PD. 1989. Retinoic acid causes an anteroposterior transformation in the developing central nervous system. Nature 340: 140144.
  • Ema M, Takahashi S, Rossant J. 2006. Deletion of selection cassette but not cis-acting elements in targeted Flk1-lacZ allele reveals Flk1 expression in multipotent mesodermal progenitors. Blood 107: 111117.
  • Emoto Y, Wada H, Okamoto H, Kudo A, Imai Y. 2005. Retinoic acid-metabolizing enzyme Cyp26a1 is essential for determining territories of hindbrain and spinal cord in zebrafish. Dev Biol 278: 415427.
  • Foley AC, Storey KG, Stern CD. 1997. The prechordal region lacks neural inducing ability, but can confer anterior character to more posterior neuroepithelium. Development 124: 29832996.
  • Foley AC, Skromne I, Stern CD. 2000. Reconciling different models of forebrain induction and patterning: a dual role for the hypoblast. Development 127: 38393854.
  • Fujii H, Sato T, Kaneko S, Gotoh O, Fujii-Kuriyama Y, Osawa K, Kato S, Hamada H. 1997. Metabolic inactivation of retinoic acid by a novel P450 differentially expressed in developing mouse embryos. EMBO J 16: 41634173.
  • Gale E, Zile M, Maden M. 1999. Hindbrain respecification in the retinoid-deficient quail. Mech Dev 89: 4354.
  • Halilagic A, Zile MH, Studer M. 2003. A novel role for retinoids in patterning the avian forebrain during presomite stages. Development 130: 20392050.
  • Holder N, Hill J. 1991. Retinoic acid modifies development of the midbrain-hindbrain border and affects cranial ganglion formation in zebrafish embryos. Development 113: 11591170.
  • Kawano Y, Kypta R. 2003. Secreted antagonists of the Wnt signalling pathway. J Cell Sci 116: 26272634.
  • Kim HJ, Bar-Sagi D. 2004. Modulation of signalling by Sprouty: a developing story. Nat Rev Mol Cell Biol 5: 441450.
  • Kimura C, Yoshinaga K, Tian E, Suzuki M, Aizawa S, Matsuo I. 2000. Visceral endoderm mediates forebrain development by suppressing posteriorizing signals. Dev Biol 225: 304321.
  • Knoetgen H, Teichmann U, Kessel M. 1999. Head-organizing activities of endodermal tissues in vertebrates. Cell Mol Biol 45: 481492.
  • Koide T, Downes M, Chandraratna RA, Blumberg B, Umesono K. 2001. Active repression of RAR signaling is required for head formation. Genes Dev 15: 21112121.
  • Kudoh T, Wilson SW, Dawid IB. 2002. Distinct roles for Fgf, Wnt and retinoic acid in posteriorizing the neural ectoderm. Development 129: 43354346.
  • Lai L, Bohnsack BL, Niederreither K, Hirschi KK. 2003. Retinoic acid regulates endothelial cell proliferation during vasculogenesis. Development 130: 64656474.
  • Luu L, Ramshaw H, Tahayato A, Stuart A, Jones G, White J, Petkovich M. 2001. Regulation of retinoic acid metabolism. Adv Enzyme Regul 41: 159175.
  • MacLean G, Abu-Abed S, Dollé P, Tahayato A, Chambon P, Petkovich M. 2001. Cloning of a novel retinoic-acid metabolizing cytochrome P450, Cyp26B1, and comparative expression analysis with Cyp26A1 during early murine development. Mech Dev 107: 195201.
  • Muhr J, Graziano E, Wilson S, Jessell TM, Edlund T. 1999. Convergent inductive signals specify midbrain, hindbrain, and spinal cord identity in gastrula stage chick embryos. Neuron 23: 689702.
  • Niederreither K, McCaffery P, Dräger UC, Chambon P, Dollé P. 1997. Restricted expression and retinoic acid-induced downregulation of the retinaldehyde dehydrogenase type 2 (RALDH-2) gene during mouse development. Mech Dev 62: 6778.
  • Niederreither K, Vermot J, Schuhbaur B, Chambon P, Dollé P. 2000. Retinoic acid synthesis and hindbrain patterning in the mouse embryo. Development 127: 7585.
  • Niederreither K, Vermot J, Fraulob V, Chambon P, Dollé P. 2002a. Retinaldehyde dehydrogenase 2 (RALDH2)- independent patterns of retinoic acid synthesis in the mouse embryo. Proc Natl Acad Sci USA 99: 1611116116.
  • Niederreither K, Vermot J, Schuhbaur B, Chambon P, Dollé P. 2002b. Embryonic retinoic acid synthesis is required for forelimb growth and anteroposterior patterning in the mouse. Development 129: 35633574.
  • Niederreither K, Abu-Abed S, Schuhbaur B, Petkovich M, Chambon P, Dollé P. 2002c. Genetic evidence that oxidative derivatives of retinoic acid are not involved in retinoid signaling during mouse development. Nat Genet 31: 8488.
  • Nieuwkoop PD, Nigtevecht GV. 1954. Neural activation and transformation in explants of competent ectoderm under the influence of fragments of anterior notochord in urodeles. J Embryol Exp Morphol 2: 175193.
  • Nieuwkoop PD, Botterenbrood EC, Kremer A, Bloesma FFSN, Hoessels EMLJ, Meyer G, Verheyen FJ. 1952. Activation and organization of the central nervous system in amphibians. J Exp Zool 120: 1108.
  • Nordstrom U, Maier E, Jessell TM, Edlund T. 2006. An early role for Wnt signaling in specifying neural patterns of Cdx and Hox gene expression and motor neuron subtype identity. PLoS Biol 4: e252. [epub ahead of print].
  • Otto DM, Henderson CJ, Carrie D, Davey M, Gundersen TE, Blomhoff R, Adams RH, Tickle C, Wolf CR. 2003. Identification of novel roles of the cytochrome p450 system in early embryogenesis: effects on vasculogenesis and retinoic acid homeostasis. Mol Cell Biol 23: 61036116.
  • Reijntjes S, Gale E, Maden M. 2003. Expression of the retinoic acid catabolising enzyme CYP26B1 in the chick embryo and its regulation by retinoic acid. Gene Express Patterns 3: 621627.
  • Reijntjes S, Blentic A, Gale E, Maden M. 2005. The control of morphogen signalling: regulation of the synthesis and catabolism of retinoic acid in the developing embryo. Dev Biol 285: 224237.
  • Ribes V, Otto DE, Dickmann L, Schmidt K, Schuhbaur B, Henderson C, Blomhoff R, Wolf CR, Tickle C, Dollé P. 2006. Rescue of cytochrome P450 oxidoreductase (Por) mouse mutants reveals functions in vasculogenesis, brain and limb patterning linked to retinoic acid homeostasis. Dev Biol [epub ahead of print].
  • Rossant J, Zirngibl R, Cado D, Shago, M, Giguère V. 1991. Expression of a retinoic acid response element-hsplacZ transgene defines specific domains of transcriptional activity during mouse embryogenesis. Genes Dev 5: 13331344.
  • Sakai Y, Meno C, Fujii H, Nishino J, Shiratori H, Saijoh Y, Rossant J, Hamada H. 2001. The retinoic acid-inactivating enzyme CYP26 is essential for establishing an uneven distribution of retinoic acid along the anterio-posterior axis within the mouse embryo. Genes Dev 15: 213225.
  • Scardigli R, Baumer N, Gruss P, Guillemot F, Le Roux I. 2003.. Direct and concentration-dependent regulation of the proneural gene Neurogenin2 by Pax6. Development 130: 32693281.
  • Simeone A, Avantaggiato V, Moroni MC, Mavilio F, Arra C, Cotelli F, Nigro V, Acampora D. 1995. Retinoic acid induces stage-specific antero-posterior transformation of rostral central nervous system. Mech Dev 51: 8398.
  • Sive HL, Draper BW, Harland RM, Weintraub H. 1990. Identification of a retinoic acid-sensitive period during primary axis formation in Xenopus laevis. Genes Dev 4: 932942.
  • Stern CD. 2001. Initial patterning of the central nervous system: how many organizers? Nat Rev Neurosci 2: 9298.
  • Swindell EC, Thaller C, Sockanathan S, Petkovich M, Jessell TM, Eichele G. 1999. Complementary domains of retinoic acid production and degradation in the early chick embryo. Dev Biol 216: 282296.
  • Uehara M, Yashiro K, Mamiya S, Nishino J, Chambon P, Dollé P, Sakai Y. 2006. CYP26A1 and CYP26C1 cooperatively regulate anterior-posterior patterning of the developing brain and the production of migratory cranial neural crest cells in the mouse. Dev Biol [epub ahead of print].
  • Ulven SM, Gundersen TE, Weedon MS, Landaas VO, Sakhi AK, Fromm SH, Geronimo BA, Moskaug JO, Blomhoff R. 2000. Identification of endogenous retinoids, enzymes, binding proteins, and receptors during early postimplantation development in mouse: important role of retinal dehydrogenase type 2 in synthesis of all-trans-retinoic acid. Dev Biol 220: 379391.
  • Wagner E, McCaffery P, Dräger UC. 2000. Retinoic acid in the formation of the dorsoventral retina and its central projections. Dev Biol 222: 460470.
  • White JA, Beckett-Jones B, Guo YD, Dilworth FJ, Bonasoro J, Jones G, Petkovich M. 1997. cDNA cloning of human retinoic acid-metabolizing enzyme (hP450RAI) identifies a novel family of cytochromes P450. J Biol Chem 272: 1853818541.
  • Zhang Z, Balmer JE, Lovlie A, Fromm SH, Blomhoff R. 1996. Specific teratogenic effects of different retinoic acid isomers and analogs in the developing anterior central nervous system of zebrafish. Dev Dyn 206: 7386.