The transcription factor Pax6 is required for development ... - Biblioteca

Mechanisms of Development 109 (2001) 215–224 www.elsevier.com/locate/modo

The transcription factor Pax6 is required for development of the diencephalic dorsal midline secretory radial glia that form the subcommissural organ Guillermo Estivill-Torru´s a,b, Tania Vitalis a, Pedro Fernańdez-Llebrez b, David J. Price a,* a

Department of Biomedical Sciences, University of Edinburgh Medical School, Hugh Robson Building, George Square, Edinburgh EH8 9XD, UK b Department of Animal Biology, Faculty of Sciences, University of Ma´laga, Ma´laga E-29071, Spain Received 9 February 2001; received in revised form 13 August 2001; accepted 14 August 2001

Abstract During brain development, Pax6 is expressed in specific regions of the diencephalon including secretory cells of the subcommissural organ (SCO), a circumventricular organ at the forebrain–midbrain boundary that originates from the pretectal dorsal midline neuroepithelial cells beneath the posterior commissure (PC). Homozygous small eye (Sey/Sey) mice lack functional Pax6 protein and fail to develop the SCO, a normal PC and the pineal gland. Small eye heterozygotes (Sey/1) show defective development of the SCO’s basal processes which normally penetrate the PC, indicating that normal development of the gland requires normal Pax6 gene-dosage. A correlation between the defects of SCO formation and altered R- and OB-cadherin expression patterns in the SCO is observed in mutants suggesting a role for cadherins in SCO development. q 2001 Elsevier Science Ltd. All rights reserved. Keywords: Cadherins; Diencephalon; Forebrain development; Pax6; Posterior commissure; Small eye mouse; Subcommissural organ

1. Introduction According to the prosomeric model, the developing forebrain in vertebrates can be divided into a series of rostrocaudal subdivisions on the basis of anatomical features and patterns of regulatory gene expression (Rubenstein et al., 1994). At the border between the most caudal prosomere (prosomere 1) and the mesencephalon (i.e. the forebrain– midbrain boundary) is the pretectal region, characterized by an important landmark, the posterior commissure (PC), a conspicuous decussation of fibers originating in the pretectal nuclei serving auxiliary visual functions (Mastick and Easter, 1996; Mastick et al., 1997). Even before the development of the PC, underlying pretectal dorsal midline neuroepithelial cells begin to produce characteristic high molecular weight glycoproteins of the thrombospondin family that later polymerize in a fiber called Reissner’s fiber (RF) which extends along the aqueduct, the fourth ventricle and the spinal central canal (Schoëbitz et al., 1986, 1993; Naumann et al., 1993). This secretory ependymal glial region grows concomitantly with the PC. The roof * Corresponding author. Tel.: 144-131-650-3262; fax: 144-131-6511706. E-mail address: [email protected] (D.J. Price).

of the fully differentiated caudal diencephalon consists almost entirely of the PC and the underlying secretory ependyma that forms a true brain gland, called by its location, the subcommissural organ (SCO) (reviewed by Oksche et al., 1993; Rodrı´guez et al., 1998; Fig. 1A). Throughout development and in adults, the secretory cells of the SCO display long basal cell processes that cross the nerve bundles of the PC and end in a dilatation on the outer limiting membrane or local blood vessels. This has led authors to consider SCO cells as a type of secretory radial glia persisting in the adult brain (Rakic and Sidman, 1968; Viehweg and Naumann, 1996). The close anatomical relationship of SCO and PC, together with in vitro evidence showing that SCO glycoproteins promote neurite outgrowth (Gobron et al., 2000), suggest a developmental relationship between SCO secretions and the formation of the PC. As mentioned above, the establishment of subdivisions in the developing vertebrate central nervous system (CNS) depends on the spatially and temporally restricted expression of regulatory genes encoding transcription factors and signaling molecules (Krumlauf et al., 1993; Puelles and Rubenstein, 1993; Rubenstein et al., 1994; Stoykova and Gruss, 1994; Rubenstein and Beachy, 1998). Pax6 is a developmentally regulated transcription factor, containing a paired domain (Bopp et al., 1986; Treisman et al., 1991)

0925-4773/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved. PII: S 0925-477 3(01)00527-5

216

G. Estivill-Torru´ s et al. / Mechanisms of Development 109 (2001) 215–224

and a paired-like homeodomain (Frigerio et al., 1986), expressed in specific regions of the developing CNS (Walther and Gruss, 1991). In mice Pax6 expression starts at early embryonic stages in the developing eyes, nasal

structures, forebrain, hindbrain and spinal cord (Walther and Gruss, 1991; Stoykova and Gruss, 1994; Grindley et al., 1995, 1997; Mastick et al., 1997; Warren and Price, 1997). In the dorsal forebrain, expression extends caudally to the border between prosomere 1 and the mesencephalon (Walther and Gruss, 1991). Pax6 is mutated in small eye (Sey) mice (Thieler et al., 1978; Hogan et al., 1986; Hill et al., 1991). Sey/Sey mice die at birth with numerous defects including abnormal development of the forebrain germinative neuroepithelium (Schmahl et al., 1993; Caric et al., 1997; Brunjes et al., 1998; Go¨ tz et al., 1998; Kawano et al., 1999; Chapouton et al., 1999; Warren et al., 1999; Vitalis et al., 2000). In the Sey/Sey diencephalon, the PC displays severe defects and the identity of prosomere 1 is altered (Stoykova et al., 1996; Mastick et al., 1997; Warren and Price, 1997; Schwarz et al., 1999). Pax6 appears to be a good candidate for a regulator of the morphogenesis of the SCO and the PC. The availability of antibodies against SCO secretory glycoproteins and Pax6 protein allowed us to study Pax6 expression in the radial glial secretory cells of the developing SCO, identified by immunocytochemistry using antibodies against glycoproteins of RF. We showed that mice lacking Pax6 failed to develop a normal PC and SCO. Analysis of heterozygotes indicated a gene-dosage dependent effect on SCO development. Moreover, since a link between defects elsewhere in Sey/Sey mice and altered expression of cadherins has been reported (Matsunami and Takeichi, 1995; Redies and Takeichi, 1996; Stoykova et al., 1997; Redies et al., 2000), we analyzed the expression of Rand OB-cadherin in the SCO of normal and mutant embryos.

2. Results 2.1. Pax6 is expressed in the dorsal midline secretory glia that constitute the SCO

Fig. 1. (A) Schematic sagittal view of an adult mouse brain indicating the location of the SCO beneath the PC and the RF. lv, lateral ventricle; III.v, third ventricle; IV.v, fourth ventricle; pg, pineal gland. (B–D) Pax6 protein and RF glycoprotein expression in SCO cells, as seen in coronal sections of E18.5 wild-type mouse diencephalon. (B) Rostral part of the organ showing the secretory material immunostained with a polyclonal antibody against bovine RF, AFRU. PC, posterior commissure. The section was counterstained with haematoxylin–eosin. (C) The same level as (B), stained with a monoclonal antibody against Pax6 protein. Note label in ependymal SCO cells as well as in some groups of cells in its vicinity. (D) Detail of SCO stained by double immunofluorescence for Pax6 protein (red) and RF glycoproteins (green). Double labeling is evident in SCO cells. Scale bar: (B, C) 90 mm; (D) 30 mm.

The pretectal dorsal midline neuroepithelial cells started to express Pax6 at E14.5. Expression increased and around E18.5, when Pax6 protein expression had decreased in other non-pretectal regions (Stoykova et al., 1996), most cells in the dorsal pretectal ependyma still displayed high levels of Pax6 immunoreactivity (Fig. 1C,D). In addition to the ependyma, groups of Pax6-positive cells were found in the vicinity of the SCO (Fig. 1C). Expression of Pax6 decreased after birth in the pretectal dorsal midline ependyma and very few labeled cells were found at P4. Polyclonal antibodies against bovine RF glycoproteins (AFRU) selectively stained the neuroepithelial cells lining the dorsal midline of the developing pretectum just beneath the prospective PC. Immunostaining was evident from E14.5 and progressively increased to reach the maximum level of immunoreactivity at around E18.5 (Fig. 1B), when Pax6 immunoreactivity was also highest. We considered the


Fig. 2.

217

218


AFRU-immunoreactive region in the pretectal dorsal neuroepithelium as the developing SCO since it has been repeatedly shown that this antiserum is highly specific for this organ throughout the vertebrate phylogeny (Rodrı´guez et al., 1984). Double immunofluorescence using anti-Pax6 antibodies and AFRU showed that the ependymal and hypendymal cells of the developing SCO expressed Pax6 protein (Fig. 1D). 2.2. Loss of the SCO and pineal gland in Sey/Sey mice To better understand the function of Pax6 in the development of the mouse SCO, we examined the SCO and its relation with PC development in wild-type and Sey/Sey embryos. In addition to morphological features, we analyzed the expression of the RF glycoproteins in the SCO secretory cells using AFRU. PC fibers were also examined by immunostaining against GAP43 (Matsunaga et al., 2000), an axonal growth associated protein highly expressed in developing neurones (Fitzgerald et al., 1991). In E12.5 wild-type mouse embryos, the pretectal dorsal midline was formed by a stratified neuroepithelium with elongated cells having basal processes reaching the outer surface of the brain (Fig. 2A,C). No AFRU-immunoreactivity was detected and GAP43 immunostaining revealed bundles of fibers in the region (Fig. 2E). By contrast, E12.5 Sey/Sey neuroepithelial cells at the mesencephalon/ prosomere 1 border did not display features distinguishing them from the rest of the neuroepithelium (Fig. 2B,D) and GAP43 labeled only a few thin axonal bundles that did not constitute any special commissural structure at the dorsal midline (Fig. 2F). By E14.5, an embryonic SCO was detected anatomically in the wild-type diencephalon by a slight curvature of the ventricular surface (Fig. 2G). Ependymal cells became elongated and displayed conspicuous apical cytoplasms and basal processes that penetrated among the bundles of fibers of the PC. AFRU labeled material in most of the SCO cells (Fig. 2I). At E16.5, the SCO achieved a morphology quite similar to that in adults, with cells displaying AFRU-immunoreactivity at their apical and basal poles and conspicuous basal processes (Fig. 2K,M). By contrast, Sey/Sey embryos

at E14.5 and E16.5 did not show any morphological differentiation of the ependymal cells lining the dorsal pretectum, nor was any AFRU-immunoreactivity detected (Fig. 2H,J,L,N). Moreover, the dorsal region was thinner displaying only a few bundles of nerve fibers but no differentiated PC (Fig. 2J,N). The SCO region was also analyzed using lectins. Previous studies have shown that SCO glycoproteins can be detected by use of Concanavalin A (ConA) and wheat germ agglutinin (WGA) lectins (see Rodrı´guez et al., 1998 for a review). In wild-type embryos, both lectins recognized the SCO secretory material in a pattern similar to AFRU-immunoreactivity. However, only very weak reactivity could be observed in the putative SCO area of the Sey/Sey embryo, similar to that seen in other ependymal regions (data not shown). In wild-types, the pineal gland developed from a dorsal evagination of the diencephalic roof, rostral to the SCO region (Figs. 2G,K and 3B). The pineal gland did not develop in Sey/Sey mice (Figs. 2H,L and 3C). In wild-type embryos, expression of Pax6 was detected in the pineal gland primordium from E16.5 (Fig. 3A). The pineal gland could also be identified by its immunoreactivity to OB-cadherin (not shown) and by the presence of serotonin immunoreactive cells (Fig. 3B). In contrast, Sey/Sey embryos showed no pineal gland primordium and no serotonin immunoreactivity in the developing dorsal diencephalon in any embryonic stage. Serotonin immunoreactive neurons were present in the raphe nuclei of Sey/Sey embryos (Fig. 3C, inset), although they were fewer than in wild-types (Fig. 3B). 2.3. Analysis of heterozygotes The analysis was carried out in E18.5 embryos since defects elsewhere in heterozygous (Sey/1) embryos have been demonstrated to appear later in embryogenesis (Grindley et al., 1995; Mastick et al., 1997) and since wild-types have a well-developed SCO at E18.5. The SCO of Sey/1 mice exhibited well-developed ependymal cells with conspicuous apical cytoplasms. In contrast to all wild-type cases (Fig. 4A; n ¼ 6), however, not all SCO cells in heterozygotes were immunoreactive to AFRU. In sagittal section, strong immunopositive cells were intermingled with immu-

Fig. 2. Development of the SCO in wild-type (wt) (A, C, E, G, I, K, M) and Sey/Sey (B, D, F, H, J, L, N) embryos. Mid-sagittal sections through the brain of E12.5 (A–F), E14.5 (G–J) and E16.5 (K–N) embryos. Sections were stained with an antiserum against the RF and haematoxylin–eosin counterstained (A–D, G–N) or with an antibody against the axonal growth associated protein, GAP43 (E, F). (A) In E12.5 wild-type mice, the dorsal neuroepithelium of the diencephalon had started to differentiate forming the prospective SCO (boxed and enlarged in C). (B) By contrast, E12.5 Sey/Sey embryos had an enlarged ventricle and exhibited no ependymal differentiation in the corresponding area (boxed and enlarged in D). (C) Detail of the zone boxed in (A). The prospective SCO cells showed basal processes among bundles of nerve fibers (see panel E). (D) Detail of the zone boxed in (B): no basal processes were present. (E) The same region as panel C immunostained with anti-GAP43 antibody that stained the bundles of fibers of the early PC. (F) The same region as panel D immunostained with the anti-GAP43 antibody, showing only a few nerve fibers scattered in the marginal region that did not form large bundles. (G) By E14.5, the SCO (boxed) formed a distinguishable ventricular fold, and the PC and the primordium of pineal gland (arrow) were well observed in the wild-type diencephalon. (H) E14.5 Sey/Sey diencephalon showing no signs of a SCO, PC or pineal gland. (I, J) Higher magnification views of the regions boxed in (G) and (H), respectively. Secretory material (brown label) was detected in SCO cells in wild-types but not in mutants. In wild-types (I), thick bundles of PC fibers (arrowheads) were surrounded by the basal processes of SCO cells. In (J), although a proper PC was not visible in the mutants, a few bundles of fibers were present (arrowheads). (K) E16.5 wild-type brain showing the SCO (boxed) displaying strong AFRU-immunoreactivity. The PC and pineal gland primordium (arrow) were evident. (L) E16.5 mutant diencephalon: the SCO was absent, but in the corresponding region (boxed) a tall ependyma lined the ventricular wall. (M, N) Higher magnifications of the boxed zones in (K) and (L), respectively. Scale bar: (A, B) 300 mm; (C–F, I, J, M, N) 80 mm; (G, H, K, L) 400 mm.


219

in the wild-type SCO cells. Expression resembled the pattern reported for the AFRU-immunoreactive secretory material (Fig. 4D,G), delineating clearly the SCO cells and their basal processes, especially in the proximal zone next to the cell body. In Sey/1 SCOs, the numbers of cadherin-immunoreactive basal processes were always lower than in wild-types. Fig. 4E,H shows typical examples: only a few stained basal processes were seen. Staining for R- and OB-cadherin was also evident in hypendymal cells. All Sey/Sey homozygotes exhibited weak immunoreactivity for R- and OB-cadherins in the corresponding region (Fig. 4F,I). 3. Discussion 3.1. Pax6 is involved in SCO ependymal cell differentiation

Fig. 3. Pineal gland in wild-type (A, B) and Sey/Sey (C) mice. (A) Coronal section of the brain of E17.5 wild-type embryo: in situ hybridization for Pax6 expression. The pineal gland (arrow) displayed Pax6 expression. (B, C) Mid-sagittal sections through the brains of E16.5 wild-type (B) and Sey/ Sey (C) embryos immunostained with anti-serotonin. (B) Strong immunoreactivity was evident in the pineal gland (arrow) and in neurons of the raphe of the wild-type brain. Inset in (B) shows a higher magnification of the immunostained pineal gland. (C) In Sey/Sey embryos no immunoreactivity was found in the diencephalic roof while some raphe neurons were labeled (inset in C). SCO, subcommissural organ; pg, pineal gland; aq, cerebral aqueduct; rfn, raphe nuclei. Scale bar: (A) 750 mm; (B, C) 350 mm; inset in (B) 150 mm; inset in (C) 90 mm.

nonegative cells in all the heterozygotes examined (Fig. 4B; n ¼ 12). Thus, the overall immunoreactivity of the SCO decreased. No AFRU-immunostaining was seen in Sey/Sey embryos at E18.5 (Fig. 4C; n ¼ 6). Another striking difference between heterozygotes and wild-types was the drastic reduction in the number and thickness of basal processes and of AFRU-immunoreactive hypendymal cells (compare Fig. 4A and B). Neither basal processes nor hypendymal cells showed AFRU positive material in any of the heterozygotes ðn ¼ 12Þ. Sey/1 mice exhibited an apparently normal PC (Mastick et al., 1997; present study) and the pineal gland was present, although often smaller than normal. Immunocytochemical analysis of cadherin expression showed that both R- and OB-cadherin proteins were present

The present results show that Pax6 is expressed by developing SCO secretory cells. Moreover the onset of both Pax6 expression and AFRU-immunoreactive secretory glycoprotein expression occurs simultaneously, suggesting that Pax6 could control the fate of the neuroepithelial cells that gives rise to the SCO and could be responsible for its distinctive secretory activity and development. This suggestion is reinforced by the fact that the SCO does not develop in mice lacking Pax6. Previous work has implicated Pax6 in the regulation of the proliferative activity of diencephalic precursors (Warren and Price, 1997), the differentiation of radial glial cells (Go¨ tz et al., 1998) and the control of cell fate (Ericson et al., 1997; Go¨ tz et al., 1998; Briscoe et al., 1999; Pituello et al., 1999). Little is known about the mechanism by which Pax6 protein influences neuroepithelial cell development. A possibility is that Pax6 regulates the expression of cadherins, surface glycoproteins involved in cell adhesion that have been shown to control the sorting and differentiation of neuroepithelial cells (Matsunami and Takeichi, 1995). In some locations, cadherin expression depends on the expression of Pax6 (Matsunami and Takeichi, 1995; Stoykova et al., 1997). Since cadherins are also present in the developing SCO (Kimura et al., 1996; Simonneau and Thiery, 1998; present results), it is possible that the absence of SCO formation in Sey/Sey mutants could be related to defective expression of cadherins. Interestingly, Pax6 has been suggested to control cell differentiation by controlling cell-adhesion events not only in this glandular structure, the SCO, but also outside the brain in pancreatic hormone-producing endocrine cells (Turque et al., 1994; St-Onge et al., 1997; Dohrmann et al., 2000). 3.2. The relationship between the development of the SCO and the PC A remarkable feature of the prosencephalon of Sey/Sey mice is the virtual absence of diencephalic structures such as the SCO, PC and pineal gland. Instead, the tectum extends over the region normally occupied by these structures (Stoy-

220


Fig. 4. Sagittal sections of the SCO from wild-type (A, D, G), Sey/1 (B, E, H) and Sey/Sey (C, F, I) E18.5 embryos immunostained with AFRU antiserum (A, B, C), anti-R-cadherin (D, E, F) and anti-OB-cadherin (G, H, I). (A, D, G) In the wild-type SCO, strong labeling of AFRU (A), R-cadherin (D) and OB-cadherin (G) was detected. Conspicuous basal cell processes (arrowheads) reached the meninges through the PC. (B) In heterozygotes, the SCO showed AFRUimmunopositive cells intermingled with immunonegative cells. Also, a significant decrease in the number and thickness of the basal processes was observed as compared to wild-types. (E, H) R- and OB-cadherin expression in heterozygotes was detected in SCO perikarya but there was a drastic reduction of the label in the basal processes of SCO cells. (C) In homozygotes (Sey/Sey), no AFRU-immunoreactive cells were detected in the corresponding region. In this area, however, an ependyma developed and displayed only weak immunoreactivity to R-cadherin (F) and OB-cadherin (I). Scale bar, 60 mm.

kova et al., 1996; Grindley et al., 1997; Mastick et al., 1997). Interestingly, in mice lacking Pax2 and Pax5, which are expressed in the mesencephalic/metencephalic primordium, the midbrain and cerebellum are replaced by an overdeveloped caudal diencephalon characterized by an enlarged PC (Schwarz et al., 1999). Judging by the images in Schwarz et al. (1999), the enlarged neuroepithelium beneath the PC should correspond to the developing SCO, although no specific staining was reported in that paper. We have additional evidence that mice defective in the regulatory gene Msx1, that is also expressed in the dorsal pretectum, lack a SCO and a PC (Ferna´ ndez-Llebrez; unpublished results). Thus, it may be that the development of both the PC and the underlying SCO are tightly related to one another and that these structures are under the control of regulatory genes such as Pax2, Pax5, Pax6 and Msx1. The SCO could influence the development of the PC in a way similar to that in which the floor plate influences the development of the ventral commissures in the spinal cord. In addition to the tight morphological relationship, in vitro studies have shown that SCO secretory glycoproteins promote neurite outgrowth and neural aggregation (Gobron et al., 1996, 2000; Monnerie et al., 1996, 1997a,b, 1998;

Didier et al., 2000). It is thought that this activity could be mediated by the thrombospondin-like repeats present in the RF glycoprotein (Gobron et al., 1996, 2000) that could have an attractive activity on axons forming the PC in a manner similar to F-spondin present in the floor plate of the developing spinal cord (Klar et al., 1992; Higashijima et al., 1997; Burstyn-Cohen et al., 1999). In this regard it is interesting that in zebrafish, F-spondin has been reported to be present in the developing SCO (Higashijima et al., 1997), although we have not observed any F-spondin immunoreactivity in the developing mouse SCO (unpublished results). On the other hand, nerve fibers of the PC could control the development of the SCO and the defects seen in this region in Sey/Sey embryos could be secondary to absence of the PC rather than being cell-autonomous. Studies in rat have demonstrated that the activity of the SCO depends on serotoninergic fibers originated in the raphe nuclei, some of which reach the SCO through the PC (Mikkelsen et al., 1997). However, since this innervation seems to be established at postnatal ages in the rat (Jime´ nez et al., 2001) it is difficult to envisage how it could influence prenatal events such as the formation of the SCO. Moreover, in other species, such as the rabbit and mouse, serotoninergic inner-


vation of the SCO has not been found (Wiklund et al., 1977; Matsuura et al., 1989). Thus we consider it unlikely that the development of the SCO depends on innervation by serotoninergic axons from the PC. A developmental relationship could also exist between the SCO and the pineal gland since both structures have been reported to be functionally and anatomically linked (Diederen, 1975; Oksche and Korff, 1993; Jime´ nez et al., 1993) and require the presence of Pax6 protein (present results). However, in mice mutant for Msx1, the pineal gland develops even though SCO and PC are absent (Ferna´ ndez-Llebrez, unpublished results), indicating that the pineal can develop in the absence of the SCO and PC. Previous studies (Hogan et al., 1986; Glaser et al., 1994; Schedl et al., 1996) have shown that Pax6 has gene-dosage dependent actions. The partial defects observed in the SCO in heterozygous mice, despite the presence of an apparently normal PC (Mastick et al., 1997), suggest that there is a gene-dosage dependent requirement for Pax6 by SCO cells themselves. In conclusion, the present work shows that the expression of Pax6 is essential for the development of the SCO, the PC and the pineal gland from the dorsal midline neuroepithelium at the mesencephalon/prosomere 1 boundary. Although a number of developmentally regulated genes has been reported to be expressed in the pretectal region or its neighbourhood, Pax6 is the first that has been demonstrated to regulate the development of the SCO. 4. Experimental procedures 4.1. Animals For this study, we used mice with a spontaneous point mutation of Pax6, namely Sey, that generates a non-functional Pax6 protein (Hill et al., 1991). The original small eye mutation (Pax6 sey; Roberts, 1967) arose in a stock called ‘CSR’ and was subsequently outcrossed generating the socalled ‘Edinburgh-derived Small eye’ (Kaufman et al., 1995; Quinn et al., 1997). Adult Sey/1 mice (derived from a Swiss background) were distinguished by their abnormally small eyes. Sey/Sey mice die at birth and Sey/ Sey and Sey/1 embryos used for this analysis were derived from Sey/1 £ Sey/1 matings. The day of the first appearance of the vaginal plug was considered day 0.5 (E0.5). Homozygote (Sey/Sey) embryos were recognized by the absence of eyes and a shortened snout (Thieler et al., 1978; Hogan et al., 1986; Hill et al., 1991) and heterozygotes (Sey/1) by their abnormal lower iris margin (Kaufman et al., 1995). Wild-type embryos were obtained from the same litter or derived from 1/1 £ 1 /1 matings (Swiss outbred strain). 4.2. Immunocytochemical analysis Embryos were studied at four ages: E12.5, E14.5, E16.5 and E18.5. Pregnant mice were sacrificed by cervical dislo-

221

cation and embryos dissected. Embryos and embryonic brains were placed into cold 0.1 M phosphate buffered saline (PBS), pH 7.2–7.3 and then fixed overnight in Bouins’s fluid. After being washed in PBS, they were embedded in wax and 10 mm coronal sections were cut through the telencephalon and diencephalon. Six wild-type and six Sey/Sey embryos were used for each age. When Sey/ 1 heteroxygotes were analyzed, 12 embryos were used. For immunocytochemistry, all steps were carried out at room temperature. Sections were first treated for 10 min with PBS containing 10% methanol and 10% hydrogen peroxide to inactivate endogenous peroxidase. After washing, they were exposed to one of the following antibodies for 18 h: (1) AFRU, a rabbit antiserum against bovine RF, raised in the laboratory of Dr Ferna´ ndez-Llebrez, according to Rodrı´guez et al. (1984) (used at 50 mg/ml concentration; affinity purified IgG fraction (Estivill-Torru´ s et al., 1998)); (2) anti-R-cadherin (H-66, rabbit polyclonal, 1:20 dilution, Santa Cruz Biotechnology); (3) anti-OB-cadherin (C-16, goat polyclonal, 1:20 dilution, Santa Cruz Biotechnology); (4) anti-GAP43 (GAP-7B10, mouse monoclonal, 1:1000 dilution, Sigma, St. Louis, MO, USA) and (5) anti-serotonin (rabbit polyclonal, 1:1000 dilution, Sigma). For detection, slides were processed as follows: (i) with rabbit anti-mouse, swine anti-rabbit or rabbit anti-goat (as appropriate) biotin conjugated immunoglobulins (1:400 dilution, all from Dako) for 2 h; (ii) with ExtrAvidinw-peroxidase (1:500 dilution in PBS, Sigma) for 20 min; (iii) with 3,3 0 -diaminobenzidine (0.05%; DAB, Sigma). All the antibodies were diluted in PBS containing 0.5% Triton X-100 and 2.5% normal serum (rabbit or swine normal serum, depending on the source of the secondary antibody used). Omission of the primary antibody resulted in no detectable staining. Sections were haematoxylin counterstained when required. 4.3. Fluorescence immunocytochemistry Embryos were processed according to a procedure previously described by Vitalis et al. (2000). Briefly, embryos were dissected out and frozen in isopentane (2408C). Coronal cryosections (10 mm) were obtained and processed on the same day. After being dried at room temperature, they were fixed for 10 min in methanol/acetone (1:1; 2208C), hydrated in PBS, and blocked in a solution containing 2% bovine serum albumin, 2% sheep serum, 7% glycerol and 0.2% Tween 20 (BS). Sections were incubated overnight with two different anti-PAX6 monoclonal antibodies (AD1.5.6 and AD2.35; 1:50 (Engelkamp et al., 1999)) diluted in BS. After washing in PBS containing 0.2% Tween 20 (PBST), they were incubated for 1 h in TRITC anti-mouse antibody (diluted 1:200 in phosphate buffer; Vector Laboratories, Burlingame, CA, USA). Sections were finally washed in PBST and analyzed with a Leica (Nussloch) confocal microscope. All incubations were carried out at room temperature. When double immunostaining was required, AFRU (developed in rabbit) and

222


monoclonal anti-PAX6 were used as the first antibodies and TRITC anti-mouse and FITC anti-rabbit antibodies (diluted 1:200 in phosphate buffer; Sigma) were used as secondaries. 4.4. Lectin histochemistry Adjacent serial sections to those considered for immunocytochemistry were treated similarly; they were incubated to block endogenous peroxidase and then washed in lectin buffer (0.1 M Tris-saline containing 1 mM CaCl2, pH 6.8). Slides were then incubated for 1 h at room temperature in ConA (with affinity for residues of mannose and glucose) or lectin from Triticum vulgaris WGA (with N-acetyl-glucosamine and sialic acid affinity), peroxidase labeled (1.25 and 0.5 mg/ml, respectively, diluted in lectin buffer; both from Sigma) and then reacted with DAB, as described above. 4.5. In situ hybridization Mouse brains were dissected and fixed overnight in 4% paraformaldehyde at 48C. In situ hybridizations were done as in Warren and Price (1997) and Pratt et al. (2000). The Pax6 antisense digoxigenin-labeled riboprobe was transcribed from a 1.7 kb fragment from a Pax6 cDNA clone (a gift from R. Hill) and generated sense probes from these plasmids were used for controls. Acknowledgements We thank Dr Engelkamp for kindly supplying PAX6 antibodies, Linda Sharp for confocal assistance and K. Gillies, G. Grant and V. Allison for their technical help in this study. Support was from European Commission (BIOMED BMH4 CT97-2412; TMR Programme, Marie Curie Research Training Grants ERB4001 GT972907; FIS 01-0948 and BFI2000-1360, from DGICYT, Spain). G.E.-T. is a Marie Curie Research Training Fellow and member of the Marie Curie Fellowship Association (MCFA). References Bopp, D., Burri, M., Baumgartner, S., Frigerio, G., Noll, M., 1986. Conservation of a large protein domain in the segmentation gene paired and in functionally related genes of Drosophila. Cell 47, 1033–1040. Briscoe, J., Sussel, L., Serup, P., Hartigan-O’Connor, D., Jessel, T.M., Rubenstein, J.L.R., Ericson, J., 1999. Homeobox gene Nkx2.2 and specification of neuronal identity by graded Sonic hedgehog signalling. Nature 398, 622–627. Brunjes, P.C., Fisher, M., Grainger, R., 1998. The small-eye mutation results in abnormalities in the lateral cortical migratory stream. Dev. Brain Res. 110, 121–125. Burstyn-Cohen, T., Tzarfaty, V., Frumkin, A., Feinstein, Y., Stoeckli, E., Klar, A., 1999. F-spondin is required for accurate pathfinding of commissural axons at the floor plate. Neuron 23, 233–246. Caric, D., Gooday, D., Hill, R.E., McConnell, S.K., Price, D.J., 1997. Determination of the migratory capacity of embryonic cortical cells lacking the transcription factor Pax-6. Development 124, 5087–5096. Chapouton, P., Ga¨ rtner, A., Go¨ tz, M., 1999. The role of Pax6 in restricting

cell migration between developing cortex and basal ganglia. Development 126, 5569–5579. Didier, R., Creveaux, I., Meiniel, R., Herber, A., Dastugue, B., 2000. SCOspondin and RF-GlyI: two designations for the same glycoprotein secreted by the SCO. J. Neurosci. Res. 61, 500–507. Diederen, J.H.B., 1975. A possible functional relationship between the subcommissural organ and the pineal complex and lateral eyes in Rana esculenta and Rana temporaria. Cell Tissue Res. 158, 37–60. Dohrmann, C., Gruss, P., Lemaire, L., 2000. Pax genes and the differentiation of hormone-producing endocrine cells in the pancreas. Mech. Dev. 92, 47–54. Engelkamp, D., Rashbass, P., Seawright, A., van Heyningen, V., 1999. Role of Pax6 in development of the cerebellar system. Development 126, 3583–3596. Ericson, J., Rashbass, P., Schedl, A., Brenner-Morton, S., Kawakami, A., van Heyningen, V., Jessell, T.M., Briscoe, J., 1997. Pax6 controls progenitor cell identity and neural fate in response to graded Shh signaling. Cell 90, 169–180. Estivill-Torru´ s, G., Cifuentes, M., Grondona, J.M., Miranda, E., Bermu´ dezSilva, B., Ferna´ ndez-Llebrez, P., Pe´ rez, J., 1998. Quantification of the secretory glycoproteins of the SCO by a sensitive sandwich ELISA with a polyclonal antibody and a set of monoclonal antibodies against the bovine Reissner’s fiber. Cell Tissue Res. 294, 407–413. Fitzgerald, M., Reynolds, M.L., Benowitz, L.I., 1991. GAP-43 expression in the developing rat lumbar spinal cord. Neuroscience 41, 187–199. Frigerio, G., Burri, M., Bopp, D., Baumgartner, S., Noll, M., 1986. Structure of the segmentation gene paired and the Drosophila PRD gene set as part of a gene network. Cell 47, 735–746. Glaser, T., Jepeal, L., Edwrads, J.G., Young, S.R., Favor, J., Maas, R.L., 1994. PAX6 gene dosage effect in a family with congenital cataracts, aniridia, anophtalmia and central nervous system defects. Nat. Genet. 7, 463–471. Gobron, S., Monnerie, H., Meiniel, R., Creveaux, I., Lehmann, W., Lamalle, D., Dastugue, B., Meiniel, A., 1996. SCO-spondin: a new member of the thrombospondin family secreted by the SCO is a candidate in the modulation of neuronal aggregation. J. Cell Sci. 109, 1053– 1061. Gobron, S., Creveaux, I., Meiniel, R., Didier, R., Herbet, A., Bamdad, M., El Bitar, F., Dastugue, B., Meiniel, A., 2000. SCO/Reissner’s fiber complex: characterization of SCO-spondin, a glycoprotein with potent activity on neurite outgrowth. Glia 32, 177–191. Go¨ tz, M., Stoykova, A., Gruss, P., 1998. Pax6 controls radial glia differentiation in the cerebral cortex . Neuron 21, 1031–1044. Grindley, J.C., Davidson, D.R., Hill, R.E., 1995. The role of Pax6 in eye and nasal development. Development 121, 1433–1442. Grindley, J.C., Hargett, L., Hill, R.E., Ross, A., Hogan, B.L.M., 1997. Disruption of Pax6 function in mice homozygous for the Pax6 Sey-Neu mutation produces abnormalities in the early development and regionalization of the diencephalon. Mech. Dev. 64, 111–126. Higashijima, S., Nose, A., Eguchi, G., Hotta, Y., Okamoto, H., 1997. Mindin/F-spondin family: novel ECM proteins expressed in the zebrafish embryonic axis. Dev. Biol. 192, 211–227. Hill, R.E., Favor, J., Hogan, B.L.M., Ton, C.C.T., Saunders, G.F., Hanson, I.M., Prosser, J., Jordan, T., Hastie, N.D., van Heyningen, V., 1991. Mouse small eye results from mutation in a paired-like homeoboxcontaining gene. Nature 354, 522–525. Hogan, B.L.M., Horsburgh, G., Cohen, J., Hetherington, C.M., Fisher, G., Lyon, M.F., 1986. Small eye (Sey): a homozygous lethal mutation on chromosome 2 which affects the differentiation of both lens and nasal placodes in the mouse. J. Embryol. Exp. Morphol. 97, 95–110. Jime´ nez, A.J., Pe´ rez-Fı´gares, J.M., Rodrı´guez, E.M., Ferna´ ndez-Llebrez, P., Oksche, A., 1993. Synapse-like contacts between axons of the pineal tract and the subcommissural organ in Rana perezzi (Anura) and their absence in Carassius auratus (Teleostei): ultrastructural tracer studies. Cell Tissue Res. 273, 317–325. Jime´ nez, A.J., Ferna´ ndez-Llebrez, P., Pe´ rez-Fı´gares, J.M., 2001. Neural input and neural control of the SCO. Microsc. Res. Tech. 52, 520–533.

G. Estivill-Torru´ s et al. / Mechanisms of Development 109 (2001) 215–224 Kaufman, M.H., Chang, H.H., Shaw, J.P., 1995. Craniofacial abnormalities in homozygous small eye (Sey/Sey) embryos and newborn mice. J. Anat. 186, 607–617. Kawano, H., Fukuda, T., Kubo, K., Horie, M., Uyemura, K., Takeuchi, K., Osumi, N., Eto, K., Kawamura, K., 1999. Pax-6 is required for thalamocortical pathway formation in fetal rats. J. Comp. Neurol. 408, 147– 160. Kimura, Y., Matsunami, H., Takeichi, M., 1996. Expression of cadherin-11 delineates boundaries, neuromeres, and nuclei in the developing mouse brain. Dev. Dyn. 206, 455–462. Klar, A., Baldassare, M., Jessell, T.M., 1992. F-spondin: a gene expressed at high levels in the floor plate encodes a secreted protein that promotes neural cell adhesion and neurite extension. Cell 69, 95–110. Krumlauf, R., Marshall, H., Studer, M., Nonchev, S., Sham, M.H., Lumsden, A., 1993. Hox homeobox genes and regionalisation of the nervous system. J. Neurobiol. 24, 1328–1340. Mastick, G.S., Easter Jr, S.S., 1996. Initial organization of neurons and tracts in the embryonic mouse fore- and midbrain. Dev. Biol. 173, 79–94. Mastick, G.S., Davis, N.M., Andrews, G.L., Easter Jr, S.S., 1997. Pax6 functions in boundary formation and axon guidance in the embryonic mouse forebrain. Development 124, 1985–1997. Matsunaga, E., Araki, I., Nakamura, H., 2000. Pax6 defines the di-mesencephalic boundary by repressing En1 and Pax2. Development 127, 2357–2365. Matsunami, H., Takeichi, M., 1995. Fetal brain subdivisions defined by Rand E-cadherin expressions: evidence for the role of cadherin activity in region-specific, cell–cell adhesion. Dev. Biol. 172, 466–478. Matsuura, T., Kumamoto, K., Ebara, S., Sano, Y., 1989. Serotonergic innervation of the mammalian SCO: an immunohistochemical study. Biomed. Res. Suppl. 3 (10), 177–186. Mikkelsen, J.D., Hay-Schmidt, A., Larsen, P.J., 1997. Central innervation of the rat ependyma and SCO with special reference to ascending serotoninergic projections from the raphe nuclei. J. Comp. Neurol. 384, 556–568. Monnerie, H., Boesplfug-Tanguy, O., Dastugue, B., Meiniel, A., 1996. Soluble material from Reissner’s fibre displays anti-aggregative activity in primary cultures of chick cortical neurons. Dev. Brain Res. 96, 120– 129. Monnerie, H., Dastugue, B., Meiniel, A., 1997a. Reissner’s fibre promotes neuronal aggregation and influences neuritic outgrowth in vitro. Cell Tissue Res. 281, 285–295. Monnerie, H., Dastugue, B., Meiniel, A., 1997b. In vitro differentiation of chick spinal cord neurons in the presence of the Reissner’s fibre, an ependymal brain secretion. Dev. Brain Res. 102, 167–176. Monnerie, H., Dastugue, B., Meiniel, A., 1998. Effect of synthetic peptides derived from the SCO-spondin conserved domains on chick cortical and spinal-cord neurons in cell cultures. Cell Tissue Res. 293, 407–418. Naumann, W.W., Lehmann, W., Debbage, P., 1993. The SCO and ontogenetic development of the brain. In: Oksche, A., Rodrı´guez, E.M., Ferna´ ndez-Llebrez, P. (Eds.). The SCO. An Ependymal Brain Gland, Springer, Berlin, pp. 61–69. Oksche, A., Korf, H., 1993. Relationships between the subcommissural organ and the pineal complex. In: Oksche, A., Rodrı´guez, E.M., Ferna´ ndez-Llebrez, P. (Eds.). The SCO. An Ependymal Brain Gland, Springer, Berlin, pp. 201–209. Oksche, A., Rodrı´guez, E.M., Ferna´ ndez-Llebrez, P. (Eds.), 1993. The SCO. An Ependymal Brain Gland Springer, Berlin. Pituello, F., Medevielle, F., Foulquier, F., Duprat, A.-M., 1999. Activation of Pax6 depends on somitogenesis in the chick embryo cervical spinal cord. Development 126, 587–596. Pratt, T., Vitalis, T., Warren, N., Edgar, J.M., Mason, J.O., Price, D.J., 2000. A role for Pax6 in the normal development of dorsal thalamus and its cortical connections. Development 127, 5167–5178. Puelles, L., Rubenstein, J.L.R., 1993. Expression patterns of the homeobox and other putative regulatory genes in the embryonic mouse forebrain suggest a neuromeric organization. Trends Neurosci. 16, 472–479.

223

Quinn, J.C., West, J.D., Kaufman, M.H., 1997. Genetic background effects on dental and the other craniofacial abnormalities in homozygous small eye (Pax6 sey/Pax6 sey) mice. Anat. Embryol. 196, 311–321. Rakic, P., Sidman, R.L., 1968. SCO and adjacent ependyma: autoradiographic study of their origin in the mouse brain. Am. J. Anat. 122, 317– 336. Redies, C., Takeichi, M., 1996. Cadherins in the developing central nervous system: and adhesive code for segmental and functional subdivisions. Dev. Biol. 180, 413–423. Redies, C., Ast, M., Nakagawa, S., Takeichi, M., Martıńez-de-la-Torre, M., Puelles, L., 2000. Morphologic fate of diencephalic prosomeres and their subdivisions revealed by mapping cadherin expression. J. Comp. Neurol. 421, 481–514. Roberts, R.C., 1967. Small eyes – a new dominant mutation in the mouse. Genet. Res. Camb. 9, 121–122. Rodrı´guez, E.M., Oksche, A., Hein, S., Rodrı´guez, S., Yulis, C., 1984. Comparative immunocytochemical study of the SCO. Cell Tissue Res. 237, 427–441. Rodrı´guez, E.M., Rodrı´guez, S., Hein, S., 1998. The SCO. Microsc. Res. Tech. 41, 98–123. Rubenstein, J.L.R., Beachy, P.A., 1998. Patterning of the embryonic forebrain. Curr. Opin. Neurobiol. 8, 18–26. Rubenstein, J.L.R., Martinez, S., Shimamura, K., Puelles, L., 1994. The prosomeric model: a proposal for the organization of the embryonic forebrain. Science 266, 578–580. Schedl, A., Ross, A., Lee, M., Engelkemp, D., Rashbass, P., van Heyningen, V., Hastie, N.D., 1996. Influence of PAX6 gene dosage on development: overexpression causes severe eye abnormalities. Cell 86, 71– 82. Schmahl, W., Knoediseder, M., Favor, J., Davidson, D., 1993. Defects of neural migration and the pathogenesis of cortical malformations are associated with small eye (sey) in mouse, a point mutation at the Pax-6 locus. Acta Neuropathol. 86, 126–135. Scho¨ ebitz, K., Garrido, O., Heinrichs, M., Speer, L., Rodrı´guez, E.M., 1986. Ontogenetical development of the chick and duck SCO. Histochemistry 84, 31–40. Scho¨ ebitz, K., Rodrı´guez, E.M., Garrido, O., del Brio-Leo´ n, M.A., 1993. Ontogenetic development of the SCO with reference to the flexural organ. In: Oksche, A., Rodrı´guez, E.M., Ferna´ ndez-Llebrez, P. (Eds.). The SCO. An Ependymal Brain Gland, Springer, Berlin, pp. 41–49. Schwarz, M., Alvarez-Bolado, G., Dressler, G., Urba´ nek, P., Busslinger, M., Gruss, P., 1999. Pax2/5 and Pax6 subdivide the early neural tube into three domains. Mech. Dev. 82, 29–39. Simonneau, L., Thiery, J.P., 1998. The mesenchymal cadherin-11 is expressed in restricted sites during the ontogeny of the rat brain in modes suggesting novel functions. Cell Adhes. Commun. 6, 431–450. St-Onge, L., Sosa-Pineda, B., Chowdhury, K., Mansouri, A., Gruss, P., 1997. Pax-6 is required for the differentiation of glucagon-producing a-cells in mouse pancreas. Nature 387, 406–409. Stoykova, A., Gruss, P., 1994. Roles of Pax-genes in developing and adult brain as suggested by expression patterns. J. Neurosci. 14, 1395–1412. Stoykova, A., Fritsch, R., Walther, C., Gruss, P., 1996. Forebrain patterning defects in small eye mutant mice. Development 122, 3453–3465. Stoykova, A., Go¨ tz, M., Grusss, P., Price, J., 1997. Pax6-dependent regulation of adhesive patterning, R-cadherin expression and boundary formation in developing forebrain. Development 124, 3765–3777. Thieler, K., Varnum, D.S., Stevens, L.C., 1978. Development of Dickie’s small eye, a mutation in the house mouse. Anat. Embryol. 155, 81–86. Treisman, F., Harris, E., Desplan, C., 1991. The paired box encodes a second DNA-binding domain in the paired homeodomain protein. Genes Dev. 5, 594–604. Turque, N., Plaza, S., Radvanyi, F., Carriere, C., Saule, S., 1994. Pax-QNR/ Pax-6, a paired box- and homeobox-containing gene expressed in neurons, is also expressed in pancreatic endocrine cells. Mol. Endocrinol. 8, 929–938. Viehweg, J., Naumann, W.W., 1996. Radial secretory glia conserved in the postnatal brain: a study in the rat. Anat. Embryol. 194, 355–363.

224


Vitalis, T., Cases, O., Engelkamp, D., Verney, C., Price, D.J., 2000. Defects of tyrosine hydroxylase-immunoreactive neurons in the brains of mice lacking the transcription factor Pax6. J. Neurosci. 20, 6501–6516. Walther, C., Gruss, P., 1991. Pax-6, a murine paired box gene, is expressed in the developing CNS. Development 113, 1435–1449. Warren, N., Price, D.J., 1997. Roles of Pax6 in murine diencephalic development. Development 124, 1573–1582.

Warren, N., Caric, D., Pratt, T., Clausen, J.A., Asavaritikrai, P., Mason, J.O., Hill, R.E., Price, D.J., 1999. The transcription factor, Pax6, is required for cell proliferation and differentiation in the developing cerebral cortex. Cereb. Cortex 9, 627–635. Wiklund, L., Lundberg, J.J., Møllgard, K., 1977. Species differences in serotoninergic innervation and secretory activity of rat, gerbil, mouse and rabbit SCO. Acta Physiol. Scand. Suppl. 452, 27–30.