Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Developmental Biology 248, 319 –330 (2002) doi:10.1006/dbio.2002.0737 Development of Dorsal–Ventral Polarity in the Optic Vesicle and Its Presumptive Role in Eye Morphogenesis As Shown by Embryonic Transplantation and in Ovo Explant Culturing Tomoko Uemonsa,* Kiyo Sakagami,† Kunio Yasuda,† and Masasuke Araki* ,1 *Developmental Neurobiology Laboratory, Faculty of Science, Nara Women’s University, Nara 630-8506, Japan; and †Laboratory of Molecular and Developmental Biology, Nara Institute of Science and Technology, Ikoma 630-0101, Japan Dorsal and ventral specification in the early optic vesicle appears to play a crucial role in the proper development of the eye. In the present study, we performed embryonic transplantation and organ culturing of the chick optic vesicle in order to investigate how the dorsal–ventral (D-V) polarity is established in the optic vesicle and what role this polarity plays in proper eye development. The left optic vesicle was cut and transplanted inversely in the right eye cavity of host chick embryos. This method ensured that the D-V polarity was reversed while the anteroposterior axis remained normal. The results showed that the location of the choroid fissure was altered from the normal (ventral) to ectopic positions as the embryonic stage of transplantation progressed from 6 to 18 somites. At the same time, the shape of the optic vesicle and the expression patterns of Pax2 and Tbx5, marker genes for ventral and dorsal regions of the optic vesicle, respectively, changed concomitantly in a similar way. The crucial period was between the 8- and 14-somite stages, and during this period the polarity seemed to be gradually determined. In ovo explant culturing of the optic vesicle showed that the D-V polarity and choroid fissure formation were already specified by the 10-somite stage. These results indicate that the D-V polarity of the optic vesicle is established gradually between 8- and 14-somite stages under the influence of signals derived from the midline portion of the forebrain. The presumptive signal(s) appeared to be transmitted from proximal to distal regions within the optic vesicle. A severe anomaly was observed in the development of optic vesicles reversely transplanted around the 10-somite stage: the optic cup formation was disturbed and subsequently the neural retina and pigment epithelium did not develop normally. We concluded that establishment of the D-V polarity in the optic vesicle plays an essential role in the patterning and differentiation of the neural retina and pigment epithelium. © 2002 Elsevier Science (USA) Key Words: optic vesicle; dorsal–ventral axis; chick; transplantation; choroid fissure; Pax2; Tbx5. INTRODUCTION Development of the vertebrate eye depends on coordinated interactions between three distinct tissues: the neuroepithelium, overlying surface ectoderm, and mesenchyme (Mikami, 1939; Hyer et al., 1998; Fuhrmann et al., 2000). In the early stage of eye development, the optic vesicle arises as a lateral outgrowth of the neural tube in the forebrain region and extends until it makes contact with To whom correspondence should be addressed. Fax: ⫹81-74220-3411. E-mail: [email protected]. 1 0012-1606/02 $35.00 © 2002 Elsevier Science (USA) All rights reserved. the surface ectoderm. Subsequently, the optic vesicle invaginates to form the double-layered structure of the optic cup, while the surface ectoderm, which contacts the optic vesicle, thickens into a placode that will form a lens vesicle (Grainger, 1992; Saha et al., 1992). The invagination of the optic vesicle begins at a point displaced somewhat toward its ventral surface, rather than beginning at the most lateral point in the optic vesicle, and is directed mediodorsally (Romanoff, 1960; Bellairs and Osmond, 1998). Consequently, the distal and ventral portions of the optic vesicle are considered to organize the inner layer of the optic cup fated to become the neural 319 320 Uemonsa et al. FIG. 1. (A) Schematic drawings of developing optic vesicles of chick embryos. Transplantation of optic vesicle was performed on embryos between the 6- and 17-somite stages. The lateral walls of the forebrain evaginate bilaterally to form optic vesicles at the 6-somite stage. At the 10-somite stage, the optic vesicle makes contact with the surface ectoderm, which thickens into the lens placode at the 17-somite stage. The optic vesicle then starts invagination to form the optic cup. (B) Diagram of surgical operation. The left optic vesicle from the donor embryo was transplanted inversely in the place of the host right optic vesicle. The D-V polarity was reversed, while the A-P polarity remained normal. Transplantation was usually carried out between embryos at the same stage. © 2002 Elsevier Science (USA). All rights reserved. 321 D-V Polarity in Optic Cup Formation FIG. 3. The location of the optic stalk in transplanted eyes observed on day E2. Transplanted eyes are on the left side, and the upper side corresponds to the dorsal direction. Arrows indicate the presumptive optic stalk. (A) Transplanted at the 6-somite stage. The optic cup connects to the ventral forebrain by the ventrally formed optic stalk (arrow). (B) Transplanted at the 10-somite stage. The optic cup connects to the lateral forebrain. (C) Transplanted at the 14-somite stage. The optic cup connects to the more dorsal region of the forebrain. FIG. 4. Expression of Pax2 and Tbx5 in transplanted optic vesicles. Whole-mount in situ hybridization was performed on day E2.5–E3. (A, B) Expression patterns of Pax2 and Tbx5 in normal embryos. In (A), Pax2 is expressed in the ventral region of the optic cup around the choroid fissure (arrow), while in (B), Tbx5 is expressed dorsally. (C, D) After transplantation at the 6-somite stage, the expression of these genes is similar to that in normally developing eyes. Arrows indicate choroid fissures formed ventrally. (E, F) Transplanted at the 10-somite stage. Pax2 (E) and Tbx5 (F) are expressed in the anterior and posterior sides of the optic cup, respectively. (G, H) Transplanted at the 17-somite stage. Arrows indicate choroid fissures formed dorsally. In (G), Pax2 is expressed around the choroid fissure, and in (H), Tbx5 is expressed in the ventral side of the optic cup. retina (NR), while the dorsal-most region is considered to organize the outer layer fated to become the retinal pigmented epithelium (RPE). The optic cup is connected to the ventral diencephalon by the optic stalk. The infolding process of optic cup formation is continued from the ventral side of the optic cup to the region of the optic stalk. As a result, the ventral surface of the optic cup and stalk become grooved, forming the future choroid fissure, in which the optic nerves and blood vessels come to lie later in development (Bellairs and Osmond, FIG. 2. Choroid fissure formation in transplanted optic vesicles. Embryos were examined at around day E2.5–E3 for the location of the choroid fissure. (A) Stage of transplantation and location of the choroid fissure. As the stage of grafting became later, the location of the choroid fissure changed from ventral to dorsal. When transplantation was performed between the 8- and 14-somite stages, choroid fissure formation was seen in only a few cases, and was seen particularly rarely in 10-somite embryos. (B–F) Lateral views of the transplanted side, showing the location of choroid fissure. Schematic drawings under the photographs show the configuration and direction of the transplanted optic cups. Transplantation was made at the 6-, 8-, 10-, 14-, and 17-somite stages, as shown in (B–F), respectively. In (B), the choroid fissure formed ventrally (arrow). In (C), the choroid fissure was not formed, and the developing optic cup appears to show a normal configuration with the normal direction. In (D), the choroid fissure was not formed, and most of the optic cups show an oval shape along the A-P direction. In (E), the choroid fissure was not formed, and the shape of the optic cup shows a reversed direction. In (F), the choroid fissure was formed dorsally (arrow). © 2002 Elsevier Science (USA). All rights reserved. 322 Uemonsa et al. 1998), and retinal axons project in an organized topographic manner through the choroid fissure (Udin and Fawcett, 1988; Holt and Harris, 1993). To accomplish this, the cells within the developing NR need to acquire distinct positional cues specifying the anterior–posterior (A-P) and dorsal–ventral (D-V) polarities. Thus, for proper eye development, regional specialization must occur in the optic vesicle according to the axial patterns. In the present study, we performed embryonic transplantation of the chick optic vesicle to investigate how the D-V polarity is formed and what role it plays in eye development. For this purpose, the left optic vesicle was grafted inversely in the place of the right cavity of host embryos, resulting in reversed D-V polarity with normal A-P polarity. The position of the choroid fissure, which is normally formed ventrally, was examined with respect to whether it formed normally or ectopically. At the same time, we made in ovo explant cultures of the isolated optic vesicle to determine the time when the D-V regions are specified. In addition to morphologic observation, we examined the expression patterns of Pax2 and Tbx5 as marker genes for ventral and dorsal regions, respectively. Pax2, a member of the paired box gene family, is first expressed at stage 10 in the ventral half of the optic vesicle (Nornes et al., 1990; Krauss et al., 1991; Puschel et al., 1992). The chick Tbx5 gene, a member of the T-box transcription factor family, is expressed in the dorsal side of the developing eye and is homologous to the Drosophila optomotor blind (omb) gene, which regulates optic lobe development. Tbx5 expression in the chick eye was first detected at stage 11 throughout the retina, with the strongest signal in the dorsal retina. The Tbx5 gene is considered to be a strong candidate for a determinant of the D-V axis of the eye (Gibson-Brown et al., 1998; Chapman et al., 1996; Jeremy et al., 1998; KoshibaTakeuchi et al., 2000). The present results showed that the D-V polarity of the optic vesicle is gradually established during early optic vesicle development and that the inversion of the D-V polarity at particular stages causes the failure of optic cup formation, followed by a loss of organized structures of the NR and RPE. This suggests that the establishment of the D-V polarity is a prerequisite for the proper development and organization of the NR and RPE. Our results also suggest that presumptive signal(s) derived from the midline portion of the embryo and transmitted within the optic vesicle play a significant role in the determination of the D-V polarity in the optic vesicle. MATERIALS AND METHODS Surgical Manipulation of Embryos Fertilized eggs were incubated in a humidified atmosphere at 37.6°C. All operations were carried out according to the procedures previously described (Alvarado-Mallart and Sotelo, 1984; BronnerFraser, 1996; Araki et al., 2002). A small window was made in the shell, and India ink diluted in Hanks’ solution was injected into the yolk beneath embryos to visualize the embryonic structures. Embryos were staged according to Hamburger and Hamilton (1951) (HH). Transplantation of the optic vesicle was performed in embryos from the 6- to 17-somite stages (Fig. 1A; 6-somite stage corresponds to HH8, 8-somite to HH9, 10-somite to HH10, 14-somite to HH11, and 17-somite to HH12). The left optic vesicle of the donor embryo was truncated at the most proximal part close to the forebrain and was transplanted to the right cavity of the host embryo, from which the counterpart optic vesicle had previously been removed. Transplantation was always carried out between embryos of the same stages. The surgical operation was performed as follows: donor embryos were placed into round-bottomed dishes filled with Hanks’ solution to wash out the yolk and India ink. The embryos were then pinned over a black silicone plate, and the left optic vesicle was cut out by using a fine sharpened needle with the covering ectoderm and the mesenchymal tissues (Fig. 1B). The orientation of the graft was clearly distinguishable by the shape. Grafts were kept in ice-chilled Hanks’ solution until transplantation. Subsequently, the right optic vesicle of the host embryo was excised in ovo after the vitelline membrane was opened. The graft was placed in the corresponding cavity of the host embryo (Fig. 1B). A close match in the size between the graft and the host cavity was very important for the successful integration of the graft. Particular care was taken to maintain the correct anteroposterior orientation of the graft. After transplantation, the windows of the host eggs were sealed with sealing tape, and the eggs were incubated at 37.6°C. One hour later, the sealing tape was removed to confirm that the graft was firmly attached to the host brain vesicle. The egg was sealed again and incubated until embryonic day 2–3 (day E2–E3, corresponding to stages 15–18), at which stage the choroid fissure is already formed in unoperated eyes. In some cases, embryos were incubated until day E4 –E 5 (stage 21–23). In some graft experiments, the optic vesicles of embryos at 10-somite stage were truncated at more distal locations and inversely transplanted in a similar way, and the results were compared with those obtained for embryos truncated at a more proximal location. For in ovo explant cultures, optic vesicles were cut from embryos at the 8-, 10-, 14-, and 17-somite stages and placed into the amnion of host embryos. They were allowed to develop further in ovo, and optic cup formation as well as choroid fissure formation were observed. To allow determination of the orientation of explants, charcoal-activated powder was sprinkled on the most dorsal part of the graft. In Situ Hybridization Plasmids containing Tbx5 and Pax2 cDNAs in pBluescript SK(⫺) were linearized with EcoRV and EcoRI and transcribed in vitro with T3 RNA polymerase to generate digoxigenin-labeled antisense and sense RNA probes, respectively (Promega). The Tbx5 cDNA was obtained from Dr. T. Ogura (Nara Institute of Science and Technology) and the Pax2 cDNA from Dr. H. Nakamura (Tohoku University). Whole-mount in situ hybridization was performed as described previously (Henrique et al., 1995). After hybridization, embryos were incubated with 1/2000-diluted alkaline phosphatase-conjugated anti-DIG antibody (Roche) overnight, and NBT (nitroblue tetrazolium)/BCIP (5-bromo-4-chloro-3-indoyl phosphate) staining was employed for the detection of hybridiza- © 2002 Elsevier Science (USA). All rights reserved. 323 D-V Polarity in Optic Cup Formation tion signals. Embryos were photographed with a camera attached to a stereomicroscope (Olympus). Histological Preparation For the histological observation, embryos were fixed with Bouin fixative for 3–10 h (depending on their size), dehydrated in a graded series of ethanol, and finally embedded in paraffin. Serial sections of 6-m thickness were cut in a transverse plane and stained with hematoxylin and eosin. RESULTS D-V Polarity-Reversed Transplantation Choroid fissure formation in the D-V polarity-reversed transplantation. The location of a choroid fissure in the graft was examined at day E2.5 in embryos that underwent reverse transplantation of the optic vesicle between the 6and 17-somite stages. The choroid fissure was formed ventrally when the transplantation was done at earlier stages (6- and 8-somite stages), while it was formed dorsally when transplantation was done at later stages (14- and 17-somite stage; Fig. 2A). When transplantation was performed at the 6-somite stage (n ⫽ 12), the grafted optic vesicle developed to an optic cup with a ventral choroid fissure in all cases (Figs. 2A and 2B). When grafting was performed at the 8-somite stage (n ⫽ 11), a normally appearing choroid fissure or, in some cases, a groove-like choroid fissure was formed ventrally in a small fraction (36%) of embryos (Figs. 2A and 2C), while in most cases, no choroid fissure was formed. At both stages (6- and 8-somite stages), the optic cup was connected to the ventral forebrain by the optic stalk, which was formed ventrally (Fig. 3A). It is interesting to note that no choroid fissure was formed in most cases (95% of embryos) when grafted at the 10-somite stage (n ⫽ 18) (Figs. 2A and 2D). In one exceptional case out of 18 specimens, a choroid fissure was formed anteriorly (data not shown). All of the transplanted eyes were connected to the lateral forebrain (data not shown). Invagination, a necessary step for optic cup formation, was not obvious in half the cases, and in the rest of the cases, the grafts developed to optic cups, most of which showed an oval shape extending anteroposteriorly rather than in the D-V direction. It appeared as if the D-V axis of the optic vesicle had rotated 90° anteriorly. In embryos grafted at the 14-somite stage (n ⫽ 12), similar results were obtained to those for embryos grafted at the 10-somite stage: a choroid fissure was not formed in most cases (83% of embryos) (Figs. 2A and 2E), and in the rest of the cases (17% of embryos), either a normal choroid fissure or a groove-like choroid fissure was formed at the dorsal– anterior location of the optic cup. When embryos were grafted at the 17-somite stage (n ⫽ 15), a choroid fissure was formed dorsally in most cases (93% of embryos) (Figs. 2A and 2F). The optic cups of embryos transplanted at the 14- and 17-somite stages were connected to the dorsal forebrain by a dorsally formed optic stalk (Fig. 3C). To see whether the procedure of truncation affects the development of the optic cup and choroid fissure, truncated optic vesicles were transplanted to host embryos at the 10-somite stage without inversion. The eyes developed normally and no abnormal features were noticed, indicating that the truncation procedure had no apparent effect on eye development in agreement with the results in a previous study (Araki et al., 2002). Expression patterns of Pax2 and Tbx5. We next examined the expression patterns of two genes, Pax2 and Tbx5, markers for the ventral and dorsal regions of the optic cup, respectively. Pax2 is expressed in ventral structures, such as the optic stalk and early retinal cells around the choroid fissure (Fig. 4A), while Tbx5 is restricted to the dorsal region of the optic (Fig. 4B). We performed whole-mount in situ hybridization of these genes on transplanted embryos at day E2.5. When grafting was done at the 6-somite stage (n ⫽ 4), Pax2 was expressed in the ventral region of the optic cup and Tbx5 was expressed in the dorsal region (Fig. 4D), as seen in the normally developing eye. When grafting was done at the 10-somite stage (n ⫽ 12), Pax2 was expressed in the anterior region of the optic cup, and Tbx5 was expressed in the posterior region (Fig. 4E), and in embryos grafted at the 17-somite stage (n ⫽ 4), Pax2 was expressed around the choroid fissure formed dorsally and Tbx5 was expressed in the ventral region (Fig. 4H). It was confirmed again that in most cases of transplantation at 10-somite stage no choroid fissure was formed and the optic cup showed an oval shape extending anteroposteriorly. Pax2 was expressed in the anterior region, while Tbx5 expression was seen in the posterior region, indicating that segregated localization of the two gene transcripts was achieved within the optic cup. Thus, the expression of the two marker genes changed in the direction from the normal D-V to the inverted V-D location via the A-P location. Effect of the truncation site along the proximodistal length. To clarify whether there is any temporal delay in the determination of the D-V polarity related to the proximodistal location of the optic vesicle, optic vesicles were truncated in 10-somite-stage embryos at increasingly distal sites and transplanted reversely into the corresponding cavity of the host embryos (Fig. 5). When optic vesicles were truncated in the middle of the proximodistal length (Fig. 5A, indicated by line b), a choroid fissure was formed ventrally in some cases (38% of embryos, n ⫽ 8), while in the rest of the cases, no choroid fissure was formed. When the vesicles were truncated at a more distal location (Fig. 5A, line c), a choroid fissure was formed ventrally in half of the cases (58% of embryos, n ⫽ 12), and in the rest of the cases, no choroid fissure was formed. These results indicate that optic vesicles cut at a more distal location form a ventral choroid fissure more frequently and suggest that development of the choroid fissure is affected by some © 2002 Elsevier Science (USA). All rights reserved. 324 Uemonsa et al. FIG. 5. Effect of the truncation site along the proximodistal length of the optic vesicle. (A) Diagram of operation: optic vesicles were truncated at three different sites. (a) the most proximal, (b) the middle, and (c) the most distal sites. They were transplanted in reverse in host embryos in which the corresponding regions had been removed. (B) Summary of the results. When transplanted at the most proximal site, no choroid fissure was formed in any of the cases. factor(s) in the optic vesicle that are present in a proximodistal gradient. In Ovo Explant Culture We next performed in ovo explant culturing of optic vesicles to determine the time when the D-V regions are specified in the optic vesicle. Optic vesicles truncated at the 6-, 8-, 10-, 14-, and 17-somite stages were placed within the amnion of host embryos and cultured in ovo for 2–3 days and examined for the location of a choroid fissure and Pax2 expression site (Fig. 6). The original dorsal position had been marked with charcoal-activated powder before grafting. Choroid fissure formation in the in ovo-cultured optic vesicles. Optic vesicles truncated at the 6-somite stage (n ⫽ 10) underwent a slight but incomplete invagination, and no choroid fissure was formed in any of the cases (Fig. 6B). Optic vesicles truncated at the 10-somite stage (n ⫽ 25) developed into an optic cup with a round shape. In half of the specimens, optic vesicles developed into optic cups with a ventrally formed choroid fissure or groove-like choroid fissure (Fig. 6C), and in the rest of the cases, optic cups had no choroid fissure. Optic vesicles truncated at the 14-somite stage (n ⫽ 18) and cultured developed into optic cups with a ventral choroid fissure or a groove-like choroid fissure in most cases (Fig. 6D). In some cases, no choroid fissure was formed in the optic cup. Optic vesicles truncated at the 17-somite stage (n ⫽ 10) developed into optic cups with a ventral choroid fissure in most cases. These optic vesicles truncated at the 14- and 17-somite stages developed into oval-shaped optic cups extending in the D-V direction, while those truncated at the 10-somite stage formed a round-shaped optic cup, if any. These results indicate that optic vesicles at the 10-somite stage develop into optic cups autonomously without any further influence from the neighboring brain regions. The results also suggest that the distinct capacity for choroid fissure formation is gradually acquired in the ventral region between the 10- and 14-somite stages under the influence of the neighboring tissues. Expression pattern of Pax2 and Tbx5 as marker genes for D-V regions. In explants taken from embryos at the 6-somite stage (n ⫽ 3), Pax2 was expressed in the ventral region of developing optic cups in all cases (Fig. 6B). The same result was obtained in all cases (n ⫽ 4) when explants were made at the 8- and 10-somite stages. On the other hand, no Tbx5 expression was seen in any cases of explants © 2002 Elsevier Science (USA). All rights reserved. 325 D-V Polarity in Optic Cup Formation FIG. 6. Specification of the choroid fissure formation and the expression of Pax2 and Tbx5. Optic vesicles from various embryonic stages were cultured in ovo to examine the stage when D-V polarity is specified intrinsically. (A) Summary of the results. In explant cultures of optic vesicles from the 6- and 8-somite stages, no choroid fissure is formed. At the 10-somite stage, choroid fissures are formed ventrally in half of the cases. (B, C, E, F) Expression of Pax2/Tbx5 in explant cultures. (B) In explants cultured at the 6-somite stage, Pax2 is expressed in the ventral half. Invagination of the optic vesicle is incomplete and no choroid fissure is formed. (C) In explants cultured at the 10-somite stage, Tbx5 is expressed in the dorsal half. (D–F) Explants cultured at the 17-somite stage. In (D), a choroid fissure is formed ventrally (arrow). In (E), Pax2 is expressed around the ventral choroid fissure, and in (F), Tbx5 is expressed in the dorsal half. Charcoal powder (arrowheads in C and E) indicates the original dorsal region of explants. L, lens vesicle; cf, choroid fissure. taken from embryos at the 6- (n ⫽ 2) and 8- (n ⫽ 3) somite stages. In explants taken from embryos at the 10-somite stage (n ⫽ 2), Tbx5 expression was observed in the dorsal region of developing optic cups, but only at a low level (Fig. 6C). In explants taken at the 14-somite stage (n ⫽ 3), Tbx5 was normally expressed. These results suggest that Pax2 expression is already specified in the optic vesicle prior to the 6-somite stage, while Tbx5 expression is specified later, at around the 10-somite stage. Tbx5 expression in the dorsal region also appears to be correlated with the development of morphological polarity of the eyecup. Histological Observation of Reversely Transplanted Optic Vesicle Reversely transplanted eyes were fixed at day E4.5 (stage 23) and examined for the development of the neural retina (NR) and pigment epithelium (RPE) (Fig. 7). At this stage, a choroid fissure can be recognized as a white line at the ventral location (Fig. 7A), and the inner layer of the optic cup has clearly differentiated as the NR, and the outer layer as the RPE (Fig. 7E). When optic vesicles were reversely transplanted at the 6-somite stage (n ⫽ 10), the optic cup developed normally in most cases (90% of embryos; data not shown), and both the NR and RPE differentiated normally. When transplanted at the 8- (n ⫽ 11) and 10- (n ⫽ 16) somite stages, approximately half of the grafts (55% from 8-somite embryos and 44% from 10-somite embryos) developed into optic cups without a choroid fissure, and both the NR and RPE were normally developed. In the rest of the cases (45% from 8-somite embryos; 56% from 10-somite embryos), invagination of optic vesicles was incomplete and the bilayered optic cup failed to develop (Fig. 7D) and lacked the normal organiza- © 2002 Elsevier Science (USA). All rights reserved. 326 Uemonsa et al. tion of the NR and RPE. Instead, the wall of the optic cup consisted of a multistratified cell layer which often possessed pigmented granules (Figs. 7F and 7G). When transplanted at the 14- (n ⫽ 14) and 17- (n ⫽ 6) somite stages, optic cups developed normally with a dorsal choroid fissure (86% from 14-somite embryos; 100% from 17-somite embryos). The morphology of these eyes, including the NR and RPE, was normal except for the location of the choroid fissure. DISCUSSION The structures of the vertebrate eye are organized according to three distinct axes, and these axes are determined during early embryonic development under the influence of the neighboring regions. The A-P and D-V axes appear to play important roles not only in development of the neural retina (NR) and pigment epithelium (RPE), but in the establishment of the retinotectal projection. Because axis formation is a crucial process for the early development of the eye, we addressed this issue by using an embryonic transplantation technique and in ovo organ culture combined with the expression of region-specific marker genes. With these approaches, we investigated how the D-V polarity is specified and determined during the development of the optic vesicle. We also considered what role this polarity plays in the development of the NR and RPE. Development of the D-V Polarity in the Optic Vesicle When the optic vesicle was reversely transplanted at the 6-somite stage, a choroid fissure was formed ventrally (in the normal position), and the expression of both Pax2 and Tbx5 was seen in a region-specific manner similar to that in the normally developing eye. In embryos with optic vesicles reversely transplanted at between the 8- and 14-somite stages, a choroid fissure was developed in only a few cases and was located ectopically, depending on the stage. It is of particular interest that, in embryos with optic vesicles reversely transplanted at the 10-somite stage, no choroid fissure was formed, except in one case, where it was formed anteriorly. When vesicles were transplanted at the 17somite stage, a choroid fissure was always formed dorsally. The expression pattern of Pax2 and Tbx5 was also changed concurrently with the change of the choroid fissure location. These results indicate that the D-V polarity in the optic vesicle is formed gradually during early embryonic development, and the crucial period seems to be between the 8- and 14-somite stages; at the 6-somite stage, the D-V polarity of the optic vesicle appears not to be yet determined, while it appeared to be completely determined at the 17-somite stage. Separate experiments using in ovo explant cultures indicated that Pax2 expression is already specified by the 6-somite stage and that choroid fissure formation is speci- fied around the 10-somite stage. It seems that the expression of a ventral marker gene is specified first, and during subsequent development the ventral region is specified to form the choroid fissure by the 10-somite stage. The questions then arise as to why the optic vesicle development is severely affected and the choroid fissure formation is disrupted when the optic vesicle is reversely transplanted between the 8- and 14-somite stages. Our presumption is that some factor(s) derived from the midline region of the forebrain may have a substantial role in the development of the D-V polarity in the optic vesicle. By reverse transplantation, the ventral part of an optic vesicle comes into contact with the dorsal part of the host forebrain and vice versa. As a result, regional signals from the host embryo may be transmitted to the opposite region of the graft, which has in part already been specified, counteracting the D-V polarity. Subsequently, the presumed regionalizing factor in the transplanted optic vesicle may be neutralized to a certain extent, causing the disorder of D-V polarity and failure of choroid fissure formation. It is significant to note that the shape of the optic cup was oval and oriented in an A-P direction when transplanted at the 10-somite stage, and that the expression pattern of Pax2 and Tbx5 also changed as if the D-V axis had rotated 90° anteriorly. Since the A-P axis appears to be determined at or prior to the 10-somite stage (Matsuno et al., 1992; Dutting et al., 1995), our findings suggest that formation of the D-V polarity may be coordinated with the A-P axis formation, although the mechanism is totally unknown at the moment. In a different series of transplantation, a more distal portion of the optic vesicle was truncated and transplanted reversely at the 10-somite stage. The results showed that the more distally it was truncated, the less disrupted was the optic vesicle in the development of the optic cup and the formation of the choroid fissure. As discussed above, we suppose that regionalizing signals may come from the host forebrain toward the optic vesicle. If the signals are diffusible and transmitted gradually through the optic vesicle in the proximodistal direction, the distal portion of the optic vesicle might be less specified than the proximal portion. A similar embryonic transplantation study in chick embryos was previously reported, in which an optic vesicle from various stages between HH stages 9 and 17 was transplanted reversely into a host embryo at stage 11 (Silver, 1977). In that study, when the grafts were from younger embryos, a choroid fissure was formed ventrally, whereas it was formed dorsally when grafts were from older embryos. Goldberg (1976) also performed a similar work and obtained results similar to the present observations. However, our study is the first to reveal that the D-V polarity is gradually specified and determined around at the 10-somite stage under the influence of the presumed midline signals and that the polarity plays an essential role in the subsequent eye morphogenesis. The positional specificity along the A-P axis also appears to be determined in the optic vesicle at or prior to HH stages 10 –11 (Matsuno et al., © 2002 Elsevier Science (USA). All rights reserved. 327 D-V Polarity in Optic Cup Formation 1992; Dutting and Meyer, 1995). Hence, that stage seems to be very critical for the early development of the eye. Presumptive Regionalizing Signals Which Determine the D-V Polarity Although the precise mechanism of the determination of the D-V polarity in the optic vesicle is unknown, molecules emanating from the midline of the forebrain are thought to affect the D-V patterning of the eye. It has been reported that signals from the prechordal mesoderm control the patterning of the ventral forebrain (Pera and Kessel, 1997; Dale et al., 1997, 1999) and that the notochord induces floor plate cells caudally. Shh is a signaling molecule that is expressed in cells along the ventral midline of the CNS (Fietz et al., 1994; Johnson and Tabin, 1995), including cells at the base of the optic stalks (Krauss et al., 1993). The prechordal mesoderm expresses Shh and BMP7, which appear to induce the differentiation of the ventral midline cells in the diencephalon (Dale et al., 1997). In cyclops mutant embryos of zebrafish, in which midline signaling is severely perturbed, an absence of ventral midline CNS tissue occurs, which subsequently results in the fusion of the bilateral eyes (Krauss et al., 1993; Macdonald et al., 1995). In these embryos, Shh transcripts are absent in the neuroepithelium at the early developmental stages. In mouse embryos, Shh injection expands the Vax1 and Pax2 territory at the expense of the Pax6 and Rx region (Hallonet et al., 1999). These facts suggest that the midline signaling is involved in the determination of the D-V patterning and the development of eye structures. With respect to the dorsalizing factors, it has been reported that bone morphogenetic proteins (BMPs) play crucial roles in regional morphogenesis of the dorsal forebrain in mouse embryos by regulating specific gene expression, cell proliferation, and local cell death (Furuta et al., 1997). Expression of BMP4 or BMP7 in the ectoderm activates dorsal genes in the neural tube (Liem et al., 1995). BMP4 is expressed in the dorsal region of the optic cup in the mouse embryo (Furuta et al., 1997). Ectopic expression of BMP5 and BMP4 in the chicken forebrain leads to cyclopia (Golden et al., 1999). It has also been reported that when mouse BMP4 is misexpressed in the ventral half of the optic cup in chick embryos, round eyes are formed with expansion of Tbx5 expression in the ventral half, and that expression of Vax and Pax2 is repressed (Koshiba-Takeuchi et al., 2000). These genes involved in the D-V patterning of the rostral neural tube are also expected to have important roles in the D-V patterning of the optic vesicle. Role of the D-V Polarity and Tissue Interaction in Eye Organogenesis As mentioned above, when the optic vesicle was reversely transplanted at around the 10-somite stage, development of the vesicle was disrupted in half of all cases: the optic cup formation did not proceed, and instead, the optic vesicle developed to an abnormal pigmented vesicle composed of a multistratified epithelium. These pigmented vesicles seem to be identical to the microphthalmia that was observed by Goldberg (1976) in a similar transplantation work. Our histological observations suggested that the NR and RPE cells might have intermingled within the epithelium. Our preliminary experiments showed that Rx, a marker gene for NR differentiation, was expressed in that pigmented epithelium (unpublished observations). The establishment of the D-V polarity in the optic vesicle appears to be closely related to the determination and separation of the NR and RPE domains within the optic vesicle. Tissue interaction between the optic vesicle and surrounding tissues is crucial for the patterning of the eye. The optic vesicle is in contact with both the head ectoderm and mesenchymal tissue, and classical experiments suggest that interactions among these tissues are essential for early eye development. For instance, the overlying surface ectoderm is considered to polarize the distal optic vesicle so that it becomes the NR (Dragomirov, 1937; Lopashov, 1963). Recent studies have shown that fibroblast growth factors (FGFs) are candidate factors which induce NR differentiation and suppress Mitf and pigmentation in the distal optic vesicle (Hyer et al., 1998; Pittack et al., 1997; Nguyen and Arnheiter, 2000). Extraocular mesenchyme appears to be necessary for the induction and maintenance of expression of RPE-specific genes Mitf and Wnt13 (Fuhrmann et al., 2000). When optic vesicles are subjected to reverse transplantation, the local interaction with surrounding tissues must be altered, because the mesenchymal cells are largely distributed in the dorsal portion rather than the ventral part. The dorsal mesenchymal cells are the neural crest origin, and numerous additional crest cells migrate to the optic vesicle during subsequent development (Araki et al., 2002). Therefore, it is plausible that alteration of the mesenchymal cell distribution after reverse transplantation at the 10-somite stage may cause the disruption of the D-V polarity and severely affect the determination and development of the NR and RPE. The expression patterns of marker genes for NR and RPE differentiation have been determined in the anlagen of the eye in early development. The expression of Six3, Rx, and Optx2 is first observed in the entire optic vesicle, but soon becomes restricted to the distal ventral region, which is thought to differentiate into the NR (Bovolenta et al., 1998; Furukawa et al., 1997; Ohuchi et al., 1999; Toy et al., 1998). Otx2 expression is diffusely distributed throughout the optic vesicles at first but becomes restricted to the dorsal region when the optic vesicle makes contact with the surface ectoderm. As optic cup formation proceeds, Otx2 expression becomes restricted to the outer layer, which gives rise to the RPE (Bovolenta et al., 1997). These expression patterns of NR/RPE-specific genes within the optic vesicle clearly indicate that the developmental fate to become either NR or RPE is regionally determined within the optic vesicle. The present observations indicate that disruption or loss of the D-V regionalization in the optic © 2002 Elsevier Science (USA). All rights reserved. 328 Uemonsa et al. FIG. 7. Optic cup formation in the transplanted eyes and their histological observations. (A) Whole-mount view of a normal embryo at day E4. A choroid fissure can be recognized as a white line in a ventral location (arrow). (B) Eye reversely transplanted at the 10-somite stage. Optic cup formation is disrupted and the eye is considerably smaller than the normal eye. (C) Reversely transplanted eye at the 17-somite stage. The choroid fissure is seen dorsally (arrow), and the optic cup develops normally. cf, choroid fissure; mb, midbrain. (D) Transverse view of an operated embryo like that shown in (B) fixed on day E4. The transplanted eye is on the right side. Invagination is incomplete and the eye fails to form a bilayered optic cup. The parts indicated by rectangles are shown in (E–G) at a higher magnification. (E) In the unoperated eye, the NR and RPE are well developed. (F, G) The transplanted eye. The wall of the optic vesicle consists of a multistratified epithelium which often possesses pigment granules (arrowheads). (H) Summary of the optic cup formation in the reversely transplanted eyes. Eyes were examined to determine whether the optic cup is developed normally (as shown in A or C) or anomalously (as seen in B). In the latter case, the optic cup formed a pigmented multistratified epithelium and is referred to as a pigmented vesicle. Optic cup formation was disrupted in half of the cases transplanted reversely at 8- and 10-somite stages. When transplanted at the 6- and 14-somite stages, such anomalies were seldom observed. L, lens; NR, neural retina; RPE, retinal pigmented epithelium. © 2002 Elsevier Science (USA). All rights reserved. 329 D-V Polarity in Optic Cup Formation vesicle causes severe failure of fate determination of the NR and RPE. Since Pax2 and Tbx5 appear to be localized in a segregated manner along the A-P length, even in such anomalous optic vesicles, localized expression of the two genes is not sufficient for the determination of the NR and RPE domains. Further investigations are needed to clarify the molecular mechanisms involved in such inductive sequences during eye development. ACKNOWLEDGMENTS We thank Dr. H. Nakamura (Tohoku University) and Dr. T. Ogura, J. K. Takeuchi, and K. K. Takeuchi (Nara Institute of Science and Technology) for providing plasmids. We also thank N. Shimada for technical assistance. This work was supported in part by a Grant-in-Aid and Special Coordination Funds for Brain Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan (to M.A.). REFERENCES Alvarado-Mallart, R. M., and Sotelo, C. (1984). Homotopic and heterotopic transplantation of quail tectal primordia in chick embryos: Organization of the retinotectal projections in the chimeric embryos. Dev. Biol. 103, 378 –398. Araki, M., Takano, T., Uemonsa, T., Nakane, Y., Tsudzuki, M., and Kaneko, T. (2002). Epithelia-mesenchyme interaction plays an essential role in transdifferentiation of retinal pigment epithelium of silver mutant quail: Localization of FGF and related molecules and aberrant migration pattern of neural crest cells during eye rudiment formation. Dev. Biol. 244, 358 –371. Bellairs, R., and Osmond, M. (1998). “The Atlas of Chick Development.” Academic Press, London. Bovolenta, P., Mallamaci, A., Briata, P., Corte, G., and Boncinelli, E. (1997). Implication of OTX2 in pigment epithelium determination and neural retina differentiation. J. Neurosci. 17, 4243– 4252. Bovolenta, P., Mallamaci, A., Puelles, L., and Boncinelli, E. (1998). Expression pattern of cSix3, a member of the Six/sine oculis family of transcription factors. Mech. Dev. 79, 201–203. Bronner-Fraser, M. (1996). Manipulation of neural crest cells or their migratory pathways. Methods Cell Biol. 51, 61– 80. Chapman, D. L., Garvey, N., Hancock, S., Alexiou, M., Agulnik, S. I., Gibson-Brown, J. J., Cebra-Thomas, J., Bollag, R. J., Silver, L. M., and Papaioannou, V. E. (1996). Expression of the T-box family genes, Tbx1-Tbx5, during early mouse development. Dev. Dyn. 206, 379 –390. Dale, J. K., Vesque, C., Lints, T. J., Sampath, K., and Placzek, M. (1997). Cooperation of BMP7 and SHH in the induction of forebrain ventral midline cells by prechordal mesoderm. Cell 90, 257–269. Dale, J. K., Sattar, N., Heemskerk, J., Placzek, J. D. W., Clarke, M., and Dodd, J. (1999). Differential patterning of ventral midline cells by axial mesoderm is regulated by BMP7 and chordin. Development 126, 397– 408. Dragomirov, N. I. (1937). The Influence of the neighboring ectoderm on the organization of the eye rudiment. Dokl. Akad. Nauk. 15, 61– 64. Dutting, D., and Meyer, S. U. (1995). Transplantation of the chick eye anlage reveal an early determination of nasotemporal polarity. Int. J. Dev. Biol. 39, 921–931. Fietz, M. J., Concordet, J.-P., Barbosa, R., Johnson, R., Krauss, S., McMahon, A., Tabin, P. C., and Ingham, P. W. (1994). The hedgehog gene family in Drosophila and vertebrate development. Dev. Suppl., 43–51. Fuhrmann, S., Levine, E. M., and Reh, T. A. (2000). Extraocular mesenchyme patterns the optic vesicle during early eye development in the embryonic chick. Development 127, 4599 – 609. Furukawa, T., Kozak, C. A., and Cepko, C. L. (1997). rax, a novel paired-type homeobox gene, shows expression in the anterior neural fold and developing retina. Proc. Natl. Acad. Sci. USA 94, 3088 –3093. Furuta, Y., Piston, D. W., and Hogan, B. L. M. (1997). Bone morphogenetic proteins (BMPs) as regulators of dorsal forebrain development. Development 124, 2203–2212. Gibson-Brown, J. J., Agulnik, S., Silver, L. M., and Papaioannou, V. E. (1998). Expression of T-box genes Tbx2–Tbx5 during chick organogenesis. Mech. Dev. 74, 165–169. Goldberg, S. (1976). Polarization of the avian retina. Ocular transplantation studies. J. Comp. Neurol. 168, 379 –392. Golden, J. A., Bracilovic, A., Mcfadden, K. A., and Beesley, J. S. (1999). Ectopic bone morphogenetic proteins 5 and 4 In the chicken forebrain lead to cycropia and holoprocencephaly. Proc. Natl. Acad. Sci. USA 96, 2439 –2444. Grainger, R. M. (1992). Embryonic lens induction: Shedding light on vertebrate tissue determination. Trends Genet. 8, 349 –355. Hallonet, M., Hollemann, T., Pieler, T., and Gruss, P. (1999). Vax1, a novel homeobox-containing gene, directs development of the basal forebrain and visual system. Genes Dev. 13, 3106 –3114. Hamburger, V., and Hamilton, H. L. (1951). A series of normal stages in the development of the chick embryo. J. Morphol. 88, 49 –92. Henrique, D., Adam, J., Myat, A., Chitnis, A., Lewis, J., and Ish-Horowicz, D. (1995). Expression of a Delta homologue in prospective neurons in the chick. Nature 375, 787–790. Holt, C. E., and Harris, W. A. (1993). Position, guidance, and mapping in the developing visual system. J. Neurosci. 24, 1400 – 1422. Hyer, J., Mima, T., and Mikawa, T. (1998). FGF1 patterns the optic vesicle by directing the placement of the neural retina domain. Development 125, 869 – 877. Jeremy, J., Gibson-Brown, J. J., Sergei, I., Agulnik, S., Silver, L. M., Virginia, E., and Papaioannou, V. E. (1998). Expression of T-box genes Tbx2–Tbx5 during chick organogenesis. Mech. Dev. 74, 165–169. Johnson, D. E., and Tabin, C. (1995). The long and short of hedgehog signalling. Cell 81, 313–316. Koshiba-Takeuchi, K., Takeuchi, J. K., Matsumoto, K., Momose, T., Uno, K., Hoepker, V., Ogura, K., Takahashi, N., Nakamura, H., Yasuda, K., and Ogura, T. (2000). Tbx5 and the retinotectum projection. Science 287, 134 –137. Krauss, S., Johansen, T., Korzh, V., and Fjose, A. (1991). Expression of the zebrafish paired box gene pax [zf-b] during early neurogenesis. Development 113, 1193–1206. Krauss, S., Concordet, J. P., and Ingham, P. W. (1993). A functionally conserved homolog of the Drosophila segment polarity gene hh is expressed in tissues with polarizing activity in zebrafish embryos. Cell 75, 1431–1444. © 2002 Elsevier Science (USA). All rights reserved. 330 Uemonsa et al. Liem, K. F., Jr., Tremml, G., Roelink, H., and Jessel, T. (1995). Dorsal differentiation of neural plate cells induced by BMPmediated signals from epidermal ectoderm. Cell 82, 969 –979. Lopashov, G. V. (1963). “Developmental Mechanisms of Vertebrate Eye Rudiments.” Macmillan, New York. Macdonald, R. K., Barth, A., Xu, Q., Holder, N., Mikkola, I., and Wilson, S. W. (1995). Midline signaling is required for Pax gene regulation and patterning of the eyes. Development 121, 3267– 3278. Matsuno, T., Itasaki, N., Ichijo, H., and Nakamura, H. (1992). Retinotectal projection after partial ablation of chick optic vesicles. Neurosci. Res. 15, 96 –101. Mikami, Y. (1939). Reciprocal transformations of the parts in the developing eye-vesicle, with special reference to the inductive influence of lens-ectoderm on the retinal differentiation. Zool. Mag. (Japan) 51, 253–256. Nguyen, M.-T. T., and Arnheiter, H. (2000). Signaling and transcriptional regulation in early mammalian eye development: A link between FGF and MITF. Development 127, 3581–3591. Nornes, H. O., Dressler, G. R., Knapik, E. W., Deutsch, U., and Gruss, P. (1990). Spatially and temporally restricted expression of Pax2 during murine neurogenesis. Development 109, 797– 809. Ohuchi, H., Tomonari, S., Itoh, H., Mikawa, T., and Noji, S. (1999). Identification of chick rax/rx genes with overlapping patterns of expression during early eye and brain development. Mech. Dev. 85, 193–195. Pera, E. M., and Kessel, M. (1997). Patterning of the chick forebrain anlage by prechordal plate. Development 124, 4253– 4262. Pittack, C. B., Grunwald, G. B., and Reh, T. A. (1997). Fibroblast growth factors are necessary for neural retina but not pigmented epithelium differentiation in chick embryo. Development 124, 805– 816. Puschel, A. W., Westerfield, M., and Dressler, G. R. (1992). Comparative analysis of Pax-2 protein distributions during neurulation in mice and zebrafish. Mech. Dev. 38, 197–208. Romanoff, A. L. (1960). “The Avian Embryo.” The Macmillan Company, New York. Saha, M. S., Servetnick, M., and Grainger, R. M. (1992). Vertebrate eye development. Curr. Opin. Genet. Dev. 2, 582–588. Silver, P. H. S. (1977). Experiments concerning the initiation of the choroid fissure in Gallis domesticus. J. Anat. 123, 219 –225. Toy, J., Yang, J.-M., Leppert, G. S., and Sundin, O. H. (1998). The optx2 homeobox gene is expressed in early precursors of the eye and activates retina-specific genes. Proc. Natl. Acad. Sci. USA 95, 10643–10648. Udin, S. B., and Fawcett, J. W. (1988). Formation of topographic maps. Annu. Rev. Neurosci. 11, 289 –327. © 2002 Elsevier Science (USA). All rights reserved. Received for publication March Revised May Accepted May Published online July 12, 13, 24, 22, 2002 2002 2002 2002