Developmental origin and kit-dependent development of the interstitial cells of cajal in the mammalian small intestineкод для вставкиСкачать
DEVELOPMENTAL DYNAMICS 211:35?51 (1998) Rostral Truncation of a Cyclostome, Lampetra japonica, Induced by All-Trans Retinoic Acid Defines the Head/Trunk Interface of the Vertebrate Body SHIGERU KURATANI,1* TATSUYA UEKI,1 SHIGEKI HIRANO,2 AND SHINICHI AIZAWA1 of Morphogenesis, Institute of Molecular Embryology and Genetics, Kumamoto University School of Medicine, Kumamoto, Japan 2Department of Anatomy, Niigata University School of Medicine, Niigata, Japan 1Department ABSTRACT The effect of all-trans retinoic acid on embryogenesis was studied in a cyclostome, Lampetra japonica. Treatment with 0.05? 0.5 mM retinoic acid on early gastrula and early neurula resulted in loss of the pharynx and in the rostral truncation of the neural tube. The mouth, pharynx, esophagus, heart, endostyle, and rostral brain were missing with graded severity. In the severest case, the embryo consisted only of trunk segments, especially myotomes that extended to the rostral end of the axis. The effect appeared to be dose- and stage-dependent: Rostral pharyngeal arches were more vulnerable to a lower amount of retinoic acid, and earlier treatment resulted in severer defects. The initial protrusion of the anterior axis started equally in control and retinoic acid-treated embryos, implying that the head morphogenesis is omitted in treated embryos. By identifying the number of myotomes based on the differentiation of hypobranchial muscles, there seemed to be no myotomes lost by retinoic acid-induced truncation. The rostral truncation, therefore, was not simply a limitation of the anterior axis but was restricted to the ventral portion; only the branchial arches disappeared with normally developing myotomes dorsally. The absent region can be defined as the vertebrate head in a morphological sense, including the branchiomeric and preotic paraxial regions as well as the heart. The results suggest the presence of distinct programs between somitomeric and branchiomeric portions of the body, providing a developmental basis for the dualmetamerical body plan of vertebrates. Dev. Dyn. 1998;211:35-51. r 1998 Wiley-Liss, Inc. Key words: branchial region; metamerism; cranial nerves; neural crest; all-trans retinoic acid; lampreys comprise the most basic body plan of the common ancestor of cephalochordates and craniates. Among the classes of vertebrates, lampreys belong to a sister group of gnathostomes. Fossil records suggest that the dichotomy of these two major lines may have taken place quite early in the vertebrate evolution (for review, see Janvier, 1993). Regardless of the specific characteristics evolved during their own history, lampreys share common morphological features with gnathostomes, i.e., anteroposterior as well as dorsoventral dissociation of branchiomerism and somitomerism and transient neuromeres in the rostral neural tube (Kuratani, 1997; Kuratani et al., 1997a). These synapomorphies are apparently missing in amphioxus, in which the two metamerisms are separated only dorsoventrally. The metamerical body plan becomes distinct through the organized and hierarchical processes of cell growth and differentiation, for which spatiotemporally regulated gene expression is a prerequisite. Alternatively, this body plan is profoundly related to such gene expression patterns. All-trans retinoic acid (RA) has been shown to affect diverse aspects of cellular development (for review, see Lotan, 1980) as well as the patterning processes of vertebrates (for review, see Hofmann and Eichele, 1994): In several different species, administration of RA results in the truncation of the rostral brain and posteriorization of structures caudal to a certain level of the hindbrain (Holder and Hill, 1991; Morriss-Kay et al., 1991; Papalopulu et al., 1991; for reviews, see Morriss-Kay, 1993; OsumiYamashita, 1996). Advances in molecular genetics have shed light on regulatory gene functions in embryonic patterning. In this respect, the effect of RA on axial development has drawn the attention of molecular embryologists, because this molecule appears to alter the axial value of the body by shifting the genetic code, as exemplified by Hox gene expression pattern: Morphological changes depending on RA dosage roughly coin- INTRODUCTION The vertebrate body is characterized by possession of segmentally arranged units along the anteroposterior axis: myotomes, pharyngeal arches and peripheral nerves. All of these characters are also present in amphioxus, a close relative of vertebrates, and probably r 1998 WILEY-LISS, INC. Grant sponsor: Ministry of Education, Science, and Culture of Japan (Specially Promoted Research); Grant number: 08102009. *Correspondence to: Shigeru Kuratani, Department of Morphogenesis, Institute of Molecular Embryology and Genetics, Kumamoto University School of Medicine, 4-24-1 Kuhonji, Kumamoto 862, Japan. E-mail: firstname.lastname@example.org Received 14 August; Accepted 25 September 1997 36 KURATANI ET AL. Fig. 1. External appearance of retinoic acid (RA)-treated embryos in experiment 1 at 2 days (A?C), 3 days (D?F), and 4 days (G?I) of RA treatment. A: Control embryo has grown to stage 21 with the protrusion of the head (arrowhead). B: The 0.01 然 RA-treated embryos. The morphology of the embryos is similar to that of the control. The arrowhead indicates the head protrusion. C: The anterior axis begins to protrude in 0.1 然 RA-treated embryos (arrows). D,E: The control and 0.01 然 RA-treated embryos are at stage 23, showing the rolling of the head (arrowhead). F: The difference is clear in the thickness of the rostral protrusion (arrows) of 0.1 然 RA-treated embryos compared with the embryos shown in D and E. G, H: On day 4 of RA treatment, control and 0.01 然 RA-treated embryos have extended their head and neck in a straight line. I: Due to the lack of the pharynx, the rostral portion of the 0.1 然 RA-treated embryos rolls downward and is much thinner than in the other two. Scale bar 5 1 mm. cide with the shift of the Hox code as well as the expression domains of other regulatory genes (Kessel, 1992; Lo?pez et al., 1995; Marshall et al., 1992; MorrissKay et al., 1991; for review, see Hofmann and Eichele, 1994). To investigate the evolutionary process of the vertebrate body plan, comparative analyses are essential not only by looking at the normal developmental processes but also by examining the effects of shifts in patterning programs. The effects of RA have been examined in a variety of gnathostomes and amphioxus (Holland and Holland, 1996), but nothing has been reported on agnathan animals. It is intended here to evaluate the morphological change in lamprey embryos after the administration of RA, thereby comparing the developmental plan of this animal with that of gnathostomes and cephalochordates. By controlled treatment with various concentrations of RA, we successfully obtained embryos with altered patterns. Here, we demonstrate that RA affects the lamprey development in a similar but distinct way to that shown by the experiments in gnathostomes. Long-term viability after RA treatment and the many branchial arches possessed by this animal helped to elucidate some previously unrecognized aspects of vertebrate body architecture. RESULTS Experiment 1 External features. Development of the embryos treated with 1 然 and 5 然 RA at early gastrula stopped within the first 2 hours. Other embryos treated with lower concentrations of RA survived. No substantial difference was detected morphologically between the control embryos and those treated with 0.01 然 RA. On the second day of treatment, the anterior protrusion began to appear in both control and RA-treated embryos (Fig. 1A?C). The protrusions of 0.1 然 RAtreated embryos were more pointed at the tip than the RETINOIC ACID AND LAMPREY DEVELOPMENT control embryos, and, on day 3, the head portion was more slender (Fig. 1D?F). There were no apparent differences between the control and 0.01 然 RAtreated embryos (Fig. 1D,E). By 4 days of treatment (around stage 25), control and 0.01 然 RA-treated embryos had developed normally, whereas 0.1 然 RA-treated embryos showed a truncated head and strong ventral bending of the body axis, which lacked the pharynx (Fig. 1G?I). At 5 days, roughly corresponding to stage 26, melanocytes were first seen in the embryos. Normally, they appeared around the otocyst, forming a couple of cell streams both rostral and caudal to the otocyst (Fig. 2) that corresponded to rhombomere 4 (r4) on the neuraxis (Kuratani et al., 1997b). In the 0.1 然 RA-treated embryos, the melanocyte also appeared and was distributed in the rostralmost portion of the body (Fig. 2C). In 5 of 16 embryos of this group at all stages, an epithelial cyst that was reminiscent of an otocyst was associated with the melanocytes, showing that the neuraxis of these embryos was truncated at the level of the anterior hindbrain. Anatomical features. At day 3 of RA treatment, when the control embryo had grown to stage 21 (Fig. 3B), nervous tissue was detected by using the monoclonal antibody (MAb), T-6793, which recognizes the acetylated tubulin. In the neural tube of control and 0.01 然 RA-treated embryos, Rohon-Beard cells formed the dorsolateral fasciculus (DLF), which had a rostral tip that reached the caudal hindbrain, and commissure neurons grew axons ventrally to form the ventrolateral fasciculus (VLF); in the middiencephalic region, the interstitial nucleus, the anlage of the medial longitudinal fasciculus, was observed (Fig. 3B; Kuratani et al., 1997b). The otocyst was already present, and the DLF reached a level caudal to the otocyst (Fig. 3B). In all of the 0.1 然 RA-treated embryos (3 of 3), the DLF and VLF were found up to the anterior end of the neural tube (Fig. 3A,C). An otocyst-like cyst was often observed nearby (see below). The interstitial nucleus, on the other hand, could not be identified. Consistent with the RA-induced changes at stage 21, older embryos also lacked a number of head-specific structures in the nervous system. At day 3.5 of RA treatment, when control embryos grew to stage 24, the posterior commissure and some neuromeres were recognized in the neural tube in control and 0.01 然 RA-treated embryos. The eye primordium was visible, and the maxillomandibular and facial nerve roots were present. Around the eye stalk, supraoptic and postoptic tracts were developing. In the anterior head ectoderm, olfactory epithelium had appeared as a compact sheet of cells showing strong acetylated tubulin (AT) immunoreactivity (not shown). In the 0.1 然 RA-treated embryos, on the other hand, no neuronal structures rostral to the otocyst could be identified, including the eye, the cranial nerve roots, the nerve tracts, or the olfactory epithelium (2 of 2 embryos). 37 Fig. 2. Initial distribution patterns of melanocytes (experiment 1). Wholemount fixed embryos on day 5 of RA treatment. Myotubules are illuminated by the polarized light. A: Control embryo. There are two melanocyte populations (arrows) rostral and caudal to the otocyst (ot). B: An embryo treated with 0.01 然 RA. The distribution pattern of melanocytes (arrows) is similar to that in the control embryo. C: An embryo treated with 0.1 然 RA. The melanocyte population is accumulated in the rostralmost part of the axis (small arrows). A large arrow indicates the position of an epithelial cyst. Scale bar 5 200 痠. Defects of the peripheral nervous system (PNS) in 0.1 然 RA-treated embryos were correlated with the truncation of the central nervous system (CNS) as well as with the absence of the pharynx. By day 9 of RA administration, the control and 0.01 然 RA-treated embryos had grown to stages 27?28, and all of the branchiomeric nerves were present (Fig. 4); in contrast, the 0.1 然 RA-treated embryos apparently lacked the branchiomeric nerves (Fig. 5B), i.e., all of the peripheral nerves were developing either between or within myotomes, a feature of the spinal nerves of this animal 38 KURATANI ET AL. Fig. 3. Comparison of the nervous system (experiment 1). A: An embryo treated with 0.1 然 RA at 3 days of RA treatment wholemounted and stained with the antiacetylated tubulin antibody. Note the rostral extension of the dorsolateral fasciculus (DLF) and the presence of Rohon-Beard cell-like immunostaining (arrowheads). Pronephric ducts (prd) are stained with the antibody. B,C: Illustrations based on similarly stained wholemount embryos. An embryo treated with 0.01 然 RA (B) shows the normal development of the nervous system. The DLF and ventrolateral fasciculus (VLF) are present caudal to the posterior hindbrain level. Note the development of the otocyst (ot) as well as the interstitial neurons (isn). An embryo treated with 0.1 然 RA is shown in C. Otocyst was not found in this specimen. The VLF and DLF extend almost the entire length of the neural tube. In B and C, compare the levels at which the rostral end of each tracts (arrows and arrowheads in B and C). Some axonal bundles are sprouting from the ventral aspect of the neural tube, reminiscent of the ventral spinal nerves (asterisks). Note that, in B and C, that the anterior protrusions are of similar length. Scale bars 5 100 痠. (Kuratani et al., 1997a). Of those, three anteriormost nerves passed the intermyotomic space and fasciculated distally to supply the gut and the putative pharyngeal region (Fig. 5B). Concomitant with the loss of the head, the rostral end of the notochord bent ventrally and reached the rostral end of the body (Figs. 5B, 6). The pharyngeal pouches were never visible, but endodermal tissue extended rostrally to the rostral tip of the notochord, where two tissues were mixed (Figs. 5B, 7A). No mouth opening or stomodaeum was apparent. Myotomes in the lamprey developed caudal to the otocyst and grew secondarily in a rostral direction (Fig. 4), whereas, in the 0.1 然 RA-treated embryo, the rostralmost one developed slightly caudal to the rostral end of the body (Figs. 6, 7). The space between the otocyst and the first myotome was filled with the rostral endodermal tissue, which, by this stage, had become greenish, similar to the tint of the liver primordium in the normal embryo of this stage (not shown). In the histological sections, the heart was lacking in 0.1 然 RA-treated embryos, although pericardium as well as pronephric ducts were present (Fig. 7A,B; see also Fig. 3A). The number of the ducts varied in each RA-treated embryo. In the total of 18 embryos treated with 0.1 然 RA, 15 embryos lacked the entire pharynx as well as the esophagus, three embryos possessed only the esophagus and the posterior portion of the pharynx, and one embryo developed a normal endodermal configuration (Table 1). For the analyses of muscle morphology, embryos were stained with CH-1 MAb, which recognizes tropomyosin. In normal development, the first postotic myotome becomes the infraoptic myotome, and the second becomes the supraoptic myotome (Fig. 4). Caudally, a certain number of occipital level myotomes were involved in the formation of the hypobranchial muscle, which is homologous with gnathostome tongue muscles (Fig. 4; Kuratani et al., 1997a). The myotomic origin of the hypobranchial muscle could not always be determined in the normal embryos due to the dorsocaudal expansion of the pharynx. In the RA-treated embryos, the first myotome was found slightly caudal to the rostral tip of the notochord (3 of 3), and the ventral portions of myotomes 7?12 were detached and had migrated ventrally, apparently forming the hypobranchial and hypaxial muscles (Fig. 6A). Experiment 2 In the second series of experiments, RA treatment was performed at the early-neurula stage (Tahara?s stage 18). The morphology of the treated embryos showed a graded series of severity, i.e., more extensive loss of the head structures with a higher dose of RA. The nervous system was the main focus of the analyses. The development of the nervous system at stages 26 and 27 was consistent with regression of the pharynx (Figs. 8, 9). In the control embryos at stage 26, pharyngeal pouches 2?7 were present, and the first pouch was made vestigial by the modification of the second pharyn- RETINOIC ACID AND LAMPREY DEVELOPMENT 39 Fig. 4. Normal development of the lamprey embryo at stage 28. A: Microphotograph of a wholemounted, stained normal embryo with the antiacetylated tubulin antibody. Note in A that the pronephric duct (prd) stains with the antibody. B: Illustration of the nervous system and myotomes of a similar embryo. Myotomes are numbered. Rostral myotomes (1?3) have grown rostrally over cranial nerves. On the ventrolateral aspect of the pharynx, the hypobranchial muscle (HBM) is forming. Due to the normal development of the pharynx, the origin of the myotubules of the HBM cannot be determined precisely. ds, spinal dorsal nerves; e, eye; lat, posterior lateral line nerve; llp, lower lip; oph, ophthalmic nerve; ot, otic vesicle; p1?p8, pharyngeal pouches; ulp, upper lip; vs, spinal ventral nerves; IX, glossopharyngeal nerve; V2,3, maxillomandibular nerve; VII, facial nerve; X, vagus nerve; x1?x6, vagal nerve branches innervating each branchial arch; XII, hypoglossal nerve. Scale bar 5 100 痠. geal arch (Fig. 8A). The notochord rostrally ended at the level of diencephalon. In the PNS, each pharyngeal arch was innervated by a branchiomeric nerve branch. When it was stained with T-6793 antibody, the olfactory epithelium was seen at the tip of the head, and the endostyle was darkly stained at the floor of the pharynx at stage 27 (Figs. 8A). In some of the 0.05 然 RA-treated embryos (Figs. 8B, 9C,D), three rostral nerves, the ophthalmic, maxillomandibular, and facial nerves, were developmentally altered, but others showed normal PNS morphology. The ophthalmic nerve was apparently lost or fused with the trigeminal nerve, and the development of the facial nerve was arrested. The endostyle was always present (4 of 4 embryos; Fig. 9C). In 2 of 4 embryos, the lower lip failed to differentiate (Fig. 9D), whereas, in the other two embryos, the endodermal portion corresponding to the first pharyngeal pouch was missing (not shown). At stage 26, the vagal nerve root grew more caudally on the hindbrain than in the control (Figs. 8B). The eye was lacking in 1 of 4 embryos. Olfactory epithelium was always present. The pharyngeal defects of embryos treated with 0.1 然 RA extended caudally to the third pharyngeal arch (1 of 3 embryos; Figs. 6B, 8E,F). The second pouch was usually absent (3 of 4). Concomitantly, the facial and glossopharyngeal nerves were fused distally. The second pharyngeal arch usually failed to develop, and several irregular connections were seen between the glossopharyngeal nerve and the vagus nerve. The olfactory epithelium, heart, and otocyst were always present (3 of 3), but the endostyle was missing (Fig. 8E). 40 KURATANI ET AL. Fig. 5. Development of the nervous system and myotomes in an RA-treated embryo (experiment 1). A: Microphotograph of a wholemounted, stained 0.1 然 RA-treated embryo with the antiacetylated tubulin antibody on the 9th day of RA treatment, when a control embryo has grown to stage 27/28. B: Illustration of the nervous system of the same embryo shown in A. The nerve roots are drawn with heavy lines. Note that typical ventral spinal nerves that pass the middle part of myotomes are seen caudal to the third myotome, whereas the first three roots associated with the rostral two myotomes are passing along the intermyotomic pathway, reminiscent of the vagus nerve in the normal embryo (Kuratani et al., 1997a). Consistently, they are fasciculated distally to innervate the gut. Endodermal portions are shaded by dots. Note that the rostral end of the notochord is closely associated with the endoderm. Scale bar 5 100 痠. In the 0.5 然 RA-treated embryos, most of the pharynx as well as the endostyle were missing (3 of 4), except for the presence of a few pharyngeal arches belonging to the vagal segment in one embryo (Figs. 8D, 9G). The esophagus could be seen in 4 of 5 embryos. Due to the similar appearances of the postotic arches, the numbering of these pouches was not always possible (see Table 1). Nevertheless, the nerve innervating these pouches resembled the vagus of the control embryo, in that the nerve root formed an arch rostral to myotomes after issuing from the hind brain (Fig. 8D). The vagal branch was distributed on the dorsal aspect of the gut, like the intestinal branch of the vagus nerve (Kuratani et al., 1997a). The rostral end of the notochord was RETINOIC ACID AND LAMPREY DEVELOPMENT 41 Fig. 6. Changes in myotomes (experiment 1). A: Illustration of a 0.1 然 RA-treated embryo 9 days after the treatment. The rostral head, as well as the entire pharynx are missing. Note the positions of the rostral tip of the notochord (double arrows) and the otocyst (ot) with those structures. Myotomes are numbered from rostral to caudal direction. Seventh through twelfth myotomes are releasing myotubules ventrally (small arrows; compare with Fig. 4B). They appear to represent the autonomous development of the HBM in the absence of the pharynx. The line indicates the plane of section shown in Figure 7A. B: Microphotograph of the same embryo as A. Wholemounted and stained with the antibody, CH-1. Hypobranchial muscle-like myotubules are indicated by arrowheads. Scale bar 5 100 痠. found almost at the rostral tip of the body, and there was no olfactory epithelium except in 1 of 4 embryos. The otocyst was missing in some embryos. The pericardium and pronephros were present in all of the embryos, but the heart was missing in 1 of 3 examined. 1992; Marshall et al., 1992; for review, see Marshall et al., 1996). Ectopic expression of a Hox gene in a more rostral region can actually mimic the RA effect in the head of zebrafish (Alexander et al., 1996). Furthermore, several models of Hox gene regulation mediated by endogenous RA have been proposed (Dekker et al., 1993; Hogan et al., 1992; Simeone et al., 1990, 1991; for reviews, see Hofmann and Eichele, 1994; Langston and Gudas, 1994). It must be noted, however, that the above-proposed scheme inherently tends to deal with the vertebrate body too simply, as though it had only a single metamerical body axis. To circumvent the idiosyncrasy obtained from a rather compact group of animals (gnathostomes), it is necessary to examine a sister group of gnathostomes, the lamprey. DISCUSSION Many studies have described the developmental changes caused by exogenous all-trans RA in various embryos. The consensus on how RA affects embryogenesis can be summarized as rostral truncation of the axis and concomitant posteriorization of segmental values. Typically, this phenomenon is realized by homeotic transformation of segments (e.g., vertebrae), which accompanies the anteriorly shifted Hox code (Kessel, 42 KURATANI ET AL. Fig. 7. Histology of RA-treated embryos (experiment 1). A: Semihorizontal section of the same experimental embryo shown in Figure 6 (for the plane of sectioning, see Fig. 6B). The neural tube ends slightly rostral to the otocyst (ot). A crest cell population is accumulated on the dorsal aspect of the neural tube (arrows). The rostral end of the notochord is sectioned. Note the close association of endodermal cells (containing yolk granules; arrowheads) and notochordal cells (n; vacuolated) in this region. my, myotome. B: Transverse section of the control embryo at the corresponding stage cut at the level of the pericardial cavity (pc). The pronephric duct (prd); the heart, consisting of myocardium (mc) and endocardium (ec); the dorsal aorta (da); and the esophagus (eso) are recognized. C: The same 0.1 然 RA-treated embryo shown in A sectioned at the corresponding level to B. Note the absence of the heart and esophagus and expansion of the dorsal aorta. Scale bars 5 100 痠. RA Eliminates the Rostral Identity With a high dose of RA at the time of gastrulation, the rostral neural tube of the lamprey embryo was lost. The apparent rostral extension of the notochord in these embryos is consistent with the rostral truncation. Based on the frequent development of the otocyst and melanocyte distribution pattern, the caudal limit of the truncation seemed to fall close to r4. 43 RETINOIC ACID AND LAMPREY DEVELOPMENT TABLE 1. Results Obtained in Experiments 1 and 2a RA-affected site Olfactory placode Eye Otocyst Pronephric duct Heart Mouth Pharynx Endostyle Pouch 1 Pouch 2 Pouch 3 Pouch 4 Pouch 51 Esophagus Experiment 1: Tahara?s stage 12 (early gastrula) Control 0.01 然 0.1 然 10/10 8/8 1/16 5/5 4/4 0/18 6/6 4/4 5/16 6/6 4/4 8/8 11/11 8/8 2/17 11/11 8/8 1/18 11/11 11/11 11/11 11/11 11/11 11/11 11/11 8/8 8/8 8/8 8/8 8/8 8/8 8/8 1/18 1/18 1/18 1/18 1/18 3/18 3/18 Experiment 2: Tahara?s stage 18 (early neurula) Control 0.05 然 0.1 然 0.5 然 2/2 4/4 3/3 1/4 2/2 3/4 0/3 0/4 2/2 4/4 3/3 0/2 2/2 4/4 3/3 4/4 2/2 4/4 3/3 1/3 2/2 4/4 0/4 0/4 2/2 2/2 2/2 2/2 2/2 2/2 2/2 4/4 2/4 4/4 4/4 4/4 4/4 4/4 1/4 0/4 1/4 3/4 4/4 4/4 4/4 0/4 0/4 0/4 0/4 0/4b 1/4b 4/5 aThe table shows the ratio of embryos among all specimens in which the presence or absence of each structure was able to be determined. Not all structures were able to be observed in each embryo. RA, retinoic acid. bPresence of pouches 5 and 6 was assumed in one embryo in which two posterior pouches were innervated by the vagus nerve. No pharyngeal pouch was found in the other embryos at this dose. The rostral truncation of the CNS is apparently analogous with the results in gnathostome species. The severest phenotype obtained in the lamprey resembles the zebrafish embryo treated with 9-cis RA (Zhang et al., 1996); only a mild truncation can be obtained in the zebrafish by all-trans RA (Hill et al., 1995; Holder and Hill, 1991). In mouse and Xenopus, similar truncation has been obtained with RA (Morriss-Kay et al., 1991; Papalopulu et al., 1991) in which the brain rostral to the anterior hindbrain is truncated. RA has been shown to down-regulate some rostrally expressed genes, resulting in the loss of anterior brain identities (Bally-Cuif et al., 1995; Simeone et al., 1995; Taira et al., 1994). Being expressed in the chordal- and mesodermal substrate, Otx2 appears to be one such gene responsible for anterior specification of the neurectoderm (Ang et al., 1994). Actually, homozygous disruption of the Otx2 gene results in the rostral truncation of the CNS at a similar neuraxial level (Acampora et al., 1995; Ang et al., 1996; Matsuo et al., 1995). In the disruption of Lim1 as well, a similar level of the rostral head is lost (Shawlot and Behringer, 1995). Obviously, the next step is to examine the normal and shifted expression patterns of these anteriorly expressed regulatory genes. The heart of a lamprey embryo was often lost by a rather high concentration of RA (Table 1). In gnathostomes as well, the heart is included in the target of teratogenic effects of retinoids or vitamin A deficiency (Wilson and Warkany, 1949), involving the malformation of mesoderm-derived myocardium and aorticopulmonary septation to which the neural crest contributes (Kirby and Waldo, 1990; Kirby et al., 1983). The heart is known to be affected by exogenous RA in the chick and zebrafish (Osmond et al., 1991; Stainier and Fishman, 1992). In the mouse, retinoid receptor-mediated path- ways have been shown to be prerequisite for normal cardiogenesis (Sucov et al., 1994; for review, see OsumiYamashita, 1996). The intimate relationships between the heart and the pharynx during embryogenesis (for review, see Kirby and Waldo, 1990) might explain the codistribution of defects in the two structures in the lamprey embryos. Also, the pharyngeal endoderm, which is sensitive to RA in the lamprey (see below), has been suggested to be responsible for myocardial differentiation (Sugi and Lough, 1994). RA Induces the Loss of Pharyngeal Arches Most intriguing was the graded loss of the pharynx; the more rostral part tends to disappear with lower concentration of RA (Figs. 8, 9), indicating that more anterior pharyngeal pouches are more sensitive to RA. Pharyngeal pouches in the lampreys develop from anterior to posterior, as in gnathostomes (Damas, 1944; Kuratani et al., 1997b; Veit, 1939). The treatment at early stages seemed to have more impact than a higher concentration given at later stages; the former phenotype was more severe than the latter (Figs. 5, 6, 8, 9). In particular, the heart was usually present in 0.5 然 RA-treated embryos in the second series, whereas it was totally absent from the 0.1 然 RA-treated embryos in the first series (Figs. 5, 7). Therefore, the sensitivity of the pharyngeal arches and also of the heart to RA is stage-dependent as well as dose-dependent (Fig. 10). The possible homeotic transformation of pharyngeal arches has been observed in gnathostomes (Lee et al., 1995; Marshall et al., 1992). Unlike vertebrates, branchial skeletons are hard to transform homeotically by RA; they are more likely to disappear (see, e.g., Mallo, 1997). In the lamprey, a similar change would only be found, if present, in the second arch; it failed to expand as the lower lip in a few specimens with low-dose 44 KURATANI ET AL. Fig. 8. treatment at the late stage (Fig. 9D). Even if it was the case, the developmental window and the RA concentration that allow the transformation would be very narrow. It has not yet been demonstrated in any experiment using gnathostomes that pharyngeal arches disappear as dramatically as they do in the lamprey. This may be RETINOIC ACID AND LAMPREY DEVELOPMENT partly because gnathostome embryos cannot survive long enough to develop the pharyngeal arch. Lee et al. (1995) found that RA treatment at the neural plate stage of the rat results in the fusion of the first and second arches. The developmental stage at which they administered RA was similar to experiment 2 of the present study. The two results, however, are probably of a different nature: The fused branchial arch of the rat still contains a pharyngeal pouch. The loss of endodermal pharyngeal pouches has never been reported in gnathostomes, although it has been reported in the amphioxus (see below). The branchiomeric nerve anomalies (Figs. 8, 9) might be secondary to those of the pharynx. Among gnathostomes, the RA treatment at midgastrula results in the partial posteriorization of the branchiomotor neurons in the mouse, with no substantial changes in the peripheral morphology (Kessel, 1993; Marshall et al., 1992). In Xenopus (Papalopulu et al., 1991) and zebrafish (Holder and Hill, 1991), however, a disturbed pattern of the PNS has been illustrated. Especially in Xenopus, cranial nerves tended to fuse with each one another, as in the lamprey, although the relationship between the nerve and the pharyngeal arches has not been reported (Papalopulu et al., 1991). The morphology of the PNS depends largely on the distribution pattern of neural crest cells in amniotes (Kuratani and Eichele, 1993; Loring and Erickson, 1987; Noden, 1975) and also in the lamprey (Kuratani Fig. 8. Development of the cranial nerves (experiment 2). Wholemount embryos stained with the antiacetylated tubulin antibody to show the nervous system. The developmental stage of these embryos corresponds to stage 26 of normal development. Anterior is to the right except in D. Semidiagrammatic drawings of the same embryos are shown at right (that in D is reversed horizontally for comparison). A: In a control embryo, all of the branchiomeric nerves have developed, growing axons to each pharyngeal arch. Pharyngeal arches are divided by pharyngeal pouches (p2?p7). The mouth (arrow) is formed between the upper (ulp) and lower (llp) lips. The vagus nerve (x) is distally innervating the posterior pharyngeal arches caudal to pouch 3, and, dorsally, it is associated with the posterior lateral line nerve (lat) at the exit point from the hindbrain (arrowheads). Olfactory epithelium (ol) stains dark. B: In a 0.05 然 RA-treated embryo, the first and second pharyngeal arches are deformed. Note the shift of the mouth opening (large arrow) and the change of the upper lip (ulp). The trigeminal nerve (v) is partly fused with the facial nerve (vii). In the vagus nerve, an additional nerve root is visible (small arrows). Note the caudal elongation of the vagus nerve root on the hindbrain (arrowheads). C: In a 0.1 然 RA-treated embryo, the fusion and disturbance of the cranial nerves extend to the glossopharyngeal nerve (ix). Identification of the cranial nerves was based on the position of the otocyst (ot). The second pharyngeal pouch has been diminished, and the preliminary pouches are numbered by asterisks (p3*?p6*). The mouth opening is absent. The proximal root of the vagus nerve (x) has been extensively disturbed, and its nerve root extended farther posteriorly along the hindbrain (arrowheads). Arrows indicate vagus nerve trunks developing between myotomes. D: In a 0.5 然 RA-treated embryo, the only cranial nerves left are the vagus nerve (x) and vestigial nerve fibers, possibly representing the glossopharyngeal nerve (ix). The vagus nerve is associated with only two pharyngeal arches. In the rostralmost portion of the head, there is an unorganized reticulum of neurites, in which segmental identity is no longer distinguishable. The vagus nerve root has extended more caudally than other embryos (arrowheads). Arrows indicate vagal nerve trunks developing between two successive myotomes. Scale bar 5 100 痠. 45 et al., 1997a,b). In the pharynx and hindbrain region, cephalic crest cells in the chick embryo adhere to even-numbered rhombomeres, providing the bases for the nerve root metamerism (Kuratani and Eichele, 1993). In the lamprey, a similar, shared pattern seems to be present; the rhombomere develops only transiently around stage 23, and nerve roots arise on even-numbered rhombomeres (Kuratani et al., 1997b). RA treatment at stage 10 chick embryo results in the disturbed migration of both crest cells and cranial nerve morphology (Gale et al., 1996). A similar alteration may be involved in the RA-treated lamprey embryos, especially in those cranial nerves that were associated with the lost pharyngeal arches (Fig. 8). Myotomes Do Not Disappear but Shift Rostrally The initial rostral protrusion took place similarly in control and RA-treated embryos (Fig. 1), but the protrusion contained neural elements of quite different axial natures (Fig. 3). It might be because of the so-called posteriorization of embryonic materials that RA treatment leads to the elimination of anterior identities, in the sense that trunk-type specification is executed by skipping the accomplishment of the head. In the preotic region of cyclostomes as well, there is the cephalic mesoderm, which is not segmented and does not develop into myotomes (Damas, 1944; Veit, 1939; our unpublished data on the lamprey; see Fig. 11A). This mesoderm, as a result of a posteriorizing effect, appears specifically to be truncated by a high concentration of RA (Figs. 5B, 6A). On the other hand, as indicated by the hypobranchial muscle formation, the postotic elements are likely to be simply posteriorized by RA treatment. The first myotome, as the result of rostral truncation of the lamprey, was found near the rostral end of the head, as has been observed in RA-treated gnathostome embryos (Morriss-Kay et al., 1991; Sundin and Eichele, 1992; Sundin et al., 1993). This raises the question of what numbers of segments were actually shifted rostrally. In the trunk of the lamprey, myotomes go through developmental modifications in a segmental, levelspecific manner; myotomes 1 (m1)?m3 grown secondarily into the preotic region, and some occipital level myotomes differentiate into hypobranchial muscles (Fig. 4). The numbers and levels of myotomes involved in the hypobranchial muscle are not clear in the control embryos due to the expansion of the pharynx, and observation of slightly younger normal embryos did not allow us to see clearly the developing hypobranchial muscles. In light of the number of spinal motor nerves forming the hypoglossal nerve, myotomes involved in the hypobranchial muscle formation probably range from m5 through m10 in normal development (Kuratani et al., 1997a,b). Importantly, the segmental levels of myotomes involved in the putative hypobranchial muscle in RA-treated embryos are very close to those of the control embryo (Fig. 11). Therefore, the rostral Fig. 9. Development of the cranial nerves (experiment 2). Photomicrographs of the control (A,B), 0.05 然 RA-treated (C,D), 0.1 然 RA-treated (E,F), and 0.5 然 RA-treated (G,H) embryos correspond to stage 27 of normal development. Myotomes are illuminated by the polarized lighting in B, F, and H. A,B: The first myotome (arrow in B) develops ventral to the otocyst (asterisk in A). Note the development of the olfactory epithelium (ol) and the endostyle (es). The mouth opens between the upper (ulp) and the lower (llp) lips (arrow in A). C,D: The lower lip is often obliterated (D), and the mouth (arrow in D) opens widely caudal to the upper lip (ulp). E?H: The morphology of 0.1 然 and 0.5 然 RA-treated embryos can be very similar. In both groups, only the vagus nerve is seen in the head (arrows in E and G). In G, the vagus nerve roots are arranged metamerically with the rostral myotomes (arrows; compare with F). Only two pharyngeal pouches (p) are present in E, whereas none are seen in G. The distal portion of the vagus nerve in G innervates the putative pharyngeal region. In both embryos, the anteriormost myotome develops close to the rostral end (single arrow in F,H). The rostral tip of the notochord is indicated by double arrows (F,H). Note the absence of endostyle and otocyst in these embryos. Scale bar 5 100 痠. RETINOIC ACID AND LAMPREY DEVELOPMENT Fig. 10. Graded sensitivity of pharyngeal pouches to exogenous RA. Three-dimensional graph shows the ratio by number of embryos that lacked each pharyngeal pouch at each RA concentration in experiment 2. Note that the more anterior pouches are apt to disappear more frequently, and the absence of the pouches becomes more frequent as the concentration of RA rises. myotomes seem to develop normally in RA-treated embryos, whereas the branchial arches disappear ventrally at the same axial levels. In summary, RA induces what is known as rostral truncation in the lamprey, but it is not a simple posteriorization of segmental values. The RA-induced anomalies are not merely axis-specific but, rather, are body part-specific, i.e., even when all of the pharyngeal pouches disappeared, all myotomes were present, and there appeared to be none clearly lost by the RA treatment. The anomalies in the pharyngeal structures, on the other hand, are more likely to be the simple disappearance of endodermal pharyngeal pouches. Thus, the developmental disturbance of segmental units is not equally coextensive for the paraxial and pharyngeal structures, but RA appears to affect different axial levels in each system. It seems more appropriate to postulate that independent developmental programs may exist for paraxial and endodermal systems that are isolated from one another. RA and the Vertebrate Body Plan Some comparative morphologists and embryologists have assumed a complete set of head segments, each possessing one branchiomere and one somitic metamere (Balfour, 1878; Bjerring, 1977; Goodrich, 1930; Jarvik, 1980; van Wijhe, 1882). Others have realized, however, that the branchiomerism and somitomerism do not develop in a coordinated manner but are shifted away from each other in actual developmental processes (Hatschek, 1892; Damas, 1944; de Beer, 1922). The distribution pattern of crest cells as well as the actual morphology of paraxial mesoderm are consistent 47 with the latter hypothesis, the dual metamerism put forth by Romer (1972; see also Jefferies, 1986; Kuratani, 1997; Starck, 1975). The dual-metamerical model is well reflected in the developmental pattern of the PNS in vertebrate embryos as well, i.e., the branchiomeric nerves are associated segmentally with rhombomeres and pharyngeal arches, whereas each spinal nerve is associated with somite derivatives. Different embryonic environments are shown experimentally to function as distinct constraints in patterning of each group of nerves (Detwiler, 1934; Keynes and Stern, 1984; Keynes et al., 1991; Kuratani and Eichele, 1993; Rickmann and Fawcett, 1985; Teillet et al., 1987). The early developmental pattern of the lamprey PNS has shown that the dualmetamerical scheme is also applicable in this animal (Kuratani et al., 1997a). Interestingly, in this dualmetameric model, it is the head of the lamprey that was almost selectively lost by RA treatment (Fig. 11A,B). The interface between the affected and nonaffected regions corresponds roughly, if not exactly, to the head/ trunk interface based on the distribution pattern of cephalic crest cells (Fig. 11C; Kuratani, 1997). The evolutionary origin of the dual-metamerical developmental plan is an intriguing subject for speculation. In this context, tunicate embryos have been treated with RA, and rostral truncation has been observed in the tadpole larvae (Katsuyama et al., 1995). Because the possible somitomeric and branchiomeric segmentations are found in this organism (Wada et al., 1996; for review, see Gee, 1996), it would be interesting to know whether there would be a condition in which only the endodermal derivatives were affected. The configuration of the notochord in RA-treated embryos (Figs. 4, 5, 8E?H) as well as the apparent absence of nerves with typical branchiomeric features are reminiscent of the anatomy of amphioxus (Fig. 6; Franz, 1927; Hatschek, 1892). The development of the amphioxus PNS is not well known, although the adult anatomy has shown the absence of distinction between the branchiomeric and spinal nerves, and all of the segmental dorsal nerves pass along the intermyotomic pathway (for reviews, see Franz, 1927; Fritzsch and Northcutt, 1993; Jefferies, 1986). Moreover, the brain of this animal is reported to have some features common to the vertebrate rostral brain (Lacalli et al., 1994) that are clearly lost by RA treatment (Fig. 11A,B). Therefore, the resemblance of the RA-treated lamprey embryo to the amphioxus larva may be no more than superficial and may be caused merely by truncation of the unsegmented head mesoderm. The idea of selective loss of the morphological head by RA is strengthened by a similar experiment made on the amphioxus. Holland and Holland (1996) have shown that, in this animal, RA induces the absence of anterior endodermal structures, the mouth and pharyngeal slits, a result that is reminiscent of the present study. Although the pharyngeal and mouth formation processes differ greatly between the two animals (Willey, 48 KURATANI ET AL. Fig. 11. Diagram summarizing the effects of RA on lamprey development. A: Pharyngula stage of normal embryo. On the neural tube, the posterior commissure (pc) divides the forebrain and midbrain, and the midhind brain border (prm) is located between midbrain (m) and hindbrain. The notochord extends rostrally to the level of diencephalon. The dorsolateral fasciculus (DLF) terminates rostrally at the level of the rostral hindbrain. The rostral end of the forebrain is associated with the olfactory epithelium (olep). Paraxial mesoderm is segmented as myotomes caudal to the otocyst but is not clearly segmented preotically (phm). Myotomes caudal to m7 are likely to differentiate into hypobranchial musculature (HBM). Endodermal pharyngeal pouches (solid circles) penetrate the pharyngeal wall, resulting in nine pharyngeal arches. The rostralmost pouch is vestigial. The endostyle (es) develops on the floor of the pharynx. The heavy dashed line indicates the border between the truncated and unaffected regions after RA treatment (myotomes 1?3 and HBM second- arily grow rostrally beyond the border). ep, epiphysis; oc, oral cavity; ot, otocyst; phm, preotic head mesoderm. B: RA-treated embryo, representing a generalized severest case. The dorsolateral fasciculus is associated with almost the entire length of the neural tube. The otocyst can be present at the rostral tip of the head, close to the rostral tip of the notochord. The notochord is associated rostrally with the detached endoderm (arrow). The pharynx and esophagus are missing. The paraxial space of this embryo is occupied by myotomes, the seventh and caudal of which differentiate into presumptive HBM. The pericardium develops but does not contain the heart. C: A pharyngula embryo of a shark (modified from Goodrich, 1930). There appears to be an S-shaped ??head/trunk interfaces?? (dashed line) along the caudal limit of the cephalic crest cell distribution in this animal as well as in other gnathostome embryos (Kuratani, 1997). Note that the dashed line resembles the border shown in A. 1891), both depend on normal pharyngeal arch development. Due to the small amount of yolk, RA-treated, mouthless amphioxus larvae cannot live longer than 4 days, and there is no knowing how normally their somitomeric part can develop or whether there is dosedependent sensitivity at each pharyngeal level. Nevertheless, the apparent specific loss of pharyngeal RETINOIC ACID AND LAMPREY DEVELOPMENT structures commonly observed in the lamprey and amphioxus might represent shared mechanisms of dual-metamerical patterning, like the synplesiomorphies that may be obscured in many gnathostomes. EXPERIMENTAL PROCEDURES Embryos Adult male and female lampreys (Lampetra japonica) were collected in Miomote River (Niigata, Japan) during the breeding season (late May through June) in 1995. They were brought into the laboratory, where eggs were artificially fertilized and kept in fresh water at 18蚓. Embryos were fixed either with 4% paraformaldehyde in 0.1 M phosphate-buffered saline (PFA/PBS) for immunohistochemistry or with Bouin?s fixative for histological preparations. For the staging of embryos, developmental sequence of a brook species, L. reissneri (Tahara, 1988), was applied. RA Administration A stock solution of the all-trans RA (R-2625, Sigma, St. Louis, MO) was made at a concentration of 0.01 M in dimethylsulfoxide (DMSO) and was stored at 220蚓 in the dark. The solution was diluted with DMSO, and the same amount of RA/DMSO solution was added to 50 ml of Steinberg?s solution (Steinberg, 1957) diluted 1/10, into which the lamprey embryos were transferred. For the control treatment, the same amount of DMSO alone was added. Embryos were left in the solution for 1 hour at 23蚓 in the dark. After extensive washing with 1/10 Steinberg?s solution, they were allowed to grow at room temperature in the same buffer. Two series of experiments were made. In the first, embryos at Tahara?s stage 12 (early gastrula) were treated with different concentrations of RA. Concentrations were 0.01 然, 0.1 然, 1 然, and 5 然. The second experiment was made at stage 18 (early neurula) with concentrations of 0.05 然, 0.1 然, and 0.5 然. Wholemount Immunostaining Wholemount embryos were prepared as described by Kuratani et al. (1997a) with minor modifications. After fixation with PFA/PBS at 4蚓 for 1 day, embryos were washed in 0.9% NaCl/distilled water, dehydrated in a graded series of methanol (50%, 80%, 100%), and stored at 220蚓. The samples to be stained were placed on ice in 2 ml DMSO/methanol until they sank. Five tenths of a milliliter of 10% Triton X-100/distilled water was added, and the embryos were incubated for another 30 minutes at room temperature. After washing in TrisHCl-buffered saline (20 然 Tris-HCl, pH 8.0; 150 然 NaCl; and 0.1% Triton X-100; TST), the samples were sequentially blocked with aqueous 1% periodic acid and with 5% dry nonfat milk in TST (TSTM). For wholemount immunostaining of the nervous system, a MAb raised against acetylated tubulin (monoclonal antiacetylated tubulin; no. T-6793; Sigma) was used. Embryos were incubated in primary antibody (diluted 1/1,000 in spin-clarified TSTM containing 0.1% 49 sodium azide and 5% DMSO) for 2?3 days at room temperature while being gently agitated on a shaking platform. The secondary antibody used was horseradish peroxidase (HRP)-conjugated goat antimouse IgG (Zymed Laboratories, San Francisco, CA) diluted 1/200 in TSTM. After washing in TST, the embryos were preincubated with peroxidase substrates 3,38-diaminobenzidine (DAB; 100 mg/ml) in Tris-HCl-buffered saline (TS) for 1 hour and allowed to react in TS with the same concentration of DAB and hydrogen peroxide dissolved at 0.01% for 20?40 minutes. After stopping the reaction with TS, some embryos were clarified in 30% glycerol containing 0.5% potassium hydroxide (KOH). These embryos were stored in 60% glycerol/water. Others were dehydrated in a graded series of methanol and cleared in benzyl alcohol/benzyl benzoate mixture (1:2). Embryos of both groups were mounted on depression slide glass for observation under the light microscope with Nomarski optics. Another MAb, CH-1 (purchased from Developmental Studies Hybridoma Bank, Iowa City, IA) raised against tropomyosin, was used to stain myotomes in wholemount embryos. Histology Embryos fixed with Bouin?s fixative were dehydrated and embedded in paraffin. Sections were cut at 5 痠 and stained with Gill?s hematoxylin and eosin. ACKNOWLEDGMENTS We are grateful to Nobutoshi Maeda and Miyuki Yamamoto at the Institute for Laboratory Animals, Niigata University School of Medicine for maintenance of animals and to Atsuo Nishino and Naoto Horigome for their technical assistance. We also thank Dr. Norimasa Miyamoto for technical advice and Drs. Noriko Osumi and Kazuhiko Umesono for critical reading of the paper. The CH-1 antibody developed by Dr. J. Lin was obtained from the Developmental Studies Hybridoma Bank maintained by the Department of Pharmacology and Molecular Sciences, Johns Hopkins University School of Medicine, Baltimore, MD 21205, and the Department of Biological Sciences, University of Iowa, Iowa City, IA 52242, under contract N01-HD-2-3144 from the NICHD. REFERENCES Acampora D, Mazan S, Lallemand Y, Avantaggiato V, Maury M, Simeone A, Bru?let P. Forebrain and midbrain regions are deleted in Otx22/2 mutants due to a defective anterior neurectoderm specification during gastrulation. Development 1995;12:3279?3290. Alexander D, Clarke JDW, Oxtoby E, Yan Y-L, Jowett T, Holder N. Ectopic expression of Hoxa-1 in the zebrafish alters the fate of the mandibular arch crest and phenocopies a retinoic acid-induced phenotype. Development 1996;122:735?746. Ang S-L, Conlon RA, Jin O, Rossant J. Positive and negative signals from mesoderm regulate the expression of mouse Otx2 in ectoderm explants. Development 1994;120:2979?2989. Ang S-L, Jin O, Rhinn M, Daigle N, Stevenson L, Rossant J. A targeted mouse Otx2 mutation leads to severe defects in gastrulation and formation of axial mesoderm and to deletion of rostral brain. Development 1996;122:243?252. 50 KURATANI ET AL. Balfour FM. The development of the elasmobranchial fishes. J. Anat. Physiol. 1878;11:405?706. Bally-Cuif L, Gulisano M, Broccoli V, Boncinelli E. c-otx2 Is expressed in two different phases of gastrulation and is sensitive to retinoic acid treatment in chick embryo. Mech. Dev. 1995;49:49?63. Bjerring HCA. Contribution to structural analysis of the head of craniate animals. Zool. Scrpt. 1977;6:127?183. Damas H. Recherches sur la de?velopment de Lampetra fluviatilis L. Contribution a? l?e?tude de la ce?phalogene?se des verte?bre?s. Arch. Biol. 1944;55:5?284. de Beer G. The segmentation of the head in Squalus acanthias. Quart. J. Microsc. Sci. 1922;66:457?474. Dekker EJ, Pannese M, Houtzager E, Boncinelli E, Durston A. Colinearity in the Xenopus laevis Hox-2 complex. Mech. Dev. 1993;40:3?12. Detwiler SR. An experimental study of spinal nerve segmentation in Amblystoma with reference to the plurisegmental contribution to the brachial plexus. J. Exp. Zool. 1934;67:395?441. Franz V. Morphologie der Akranier. Z. Ges. Anat. 1927;27:465?692. Fritzsch B, Northcutt G. Cranial and spinal nerve organization in amphioxus and lampreys: Evidence for an ancestral craniate pattern. Acta Anat. 1993;148:96?109. Gale E, Prince V, Lumsden A, Clarke J, Holder N, Maden M. Late effects of retinoic acid on neural crest and aspects of rhombomere identity. Development 1996;122:783?793. Gee H. Before the Backbone. London: Chapman and Hall, 1996. Goodrich ES. Studies on the Structure and Development of Vertebrates. London: McMillan, 1930. Hatschek B. Die Metamerie des Amphioxus und des Ammocoetes. Verh. Anat. Ges. 1892;6:136?162. Hill J, Clarke JDW, Vargesson N, Jowett T, Holder N. Exogenous retinoic acid causes specific alterations in the development of the midbrain and hindbrain of the zebrafish embryo including positional respecification of the Mauthner neuron. Mech. Dev. 1995;50: 3?16. Hofmann C, Eichele G. Retinoids in development. In: Sporn MB, Roberts AB, Goodman DS, eds. The Retinoids: Biology, Chemistry, and Medicine, 2nd Ed. New York: Raven Press Ltd., 1994:387?441. Hogan BL, Thaller C, Eichele G. Evidence that Hensen?s node is a site of retinoic acid synthesis. Nature 1992;359:237?241. Holder N, Hill J. Retinoic acid modifies development of the midbrainhindbrain border and affects cranial ganglion formation in zebrafish embryos. Development 1991;113:1159?1170. Holland LZ, Holland ND. Expression of AmphiHox-1 and Amphipax-1 in amphioxus embryos treated with retinoic acid: Insights into evolution and patterning of the chordate nerve cord and pharynx. Development 1996;122:1829?1838. Janvier P. Patterns of diversity in the skull of jawless fishes. In: Hanken J, Hall BK, eds. The Skull, Vol. 2. Chicago: University of Chicago Press, 1993:131?188. Jarvik E. Basic Structure and Evolution of Vertebrates, Vol. 2. New York: Academic Press, 1980. Jefferies RPS. The Ancestry of the Vertebrates. London: British Museum (Natural History), 1986. Katsuyama Y, Wada S, Yasugi S, Saiga H. Expression of labial group Hox gene HrHox-1 and its alteration induced by retinoic acid in development of the ascidian Halocynthia roretzi. Development 1995;121:3197?3205. Kessel M. Respecification of vertebral identities by retinoic acid. Development 1992;115:487?501. Kessel M. Reversal of axonal pathways from rhombomere 3 correlates with extra Hox expression domains. Neuron 1993;10:379?393. Keynes RJ, Stern CD. Segmentation in the vertebrate nervous system. Nature 1984;310:786?789. Keynes RJ, Jaques KF, Cook GMW. Axon repulsion during peripheral nerve segmentation. Development 1991;2(Suppl.):131?139 Kirby ML, Waldo KL. Role of neural crest in congenital heart disease. Circulation 1990;82:332?340. Kirby ML, Gale TF, Stewart DE. Neural crest cells contribute to normal aorticopulmonary septation. Science 1983;220:1059?1061. Kuratani S. Spatial distribution of postotic crest cells defines the head/trunk interface of the vertebrate body: Embryological interpre- tation of peripheral nerve morphology and evolution of the vertebrate head. Anat. Embryol. 1997;195:1?13. Kuratani SC, Eichele G. Rhombomere transplantation repatterns the segmental organization of cranial nerves and reveals cell-autonomous expression of a homeodomain protein. Development 1993;117: 105?117. Kuratani S, Ueki T, Aizawa S, Hirano S. Peripheral development of cranial nerves in a cyclostome, Lampetra japonica: Morphological distribution of nerve branches and the vertebrate body plan. J. Comp. Neurol. 1997a;384:483?500 Kuratani S, Horigome N, Hirano S, Ueki T, Aizawa S. Stereotyped axonal bundle formation and neuromeric patterns in embryos of a cyclostome, Lampetra japonica. J. Comp. Neurol. 1997b (in press). Lacalli TC, Holland ND, West JE. Landmarks in the anterior nervous system of amphioxus larvae. Phil. Trans. R. Soc. London B 1994;344: 165?185. Langston AW, Gudas LJ. Retinoic acid and homeobox gene regulation. Curr. Opin. Genet. Dev. 1994;4:550?555. Lee YM, Osumi-Yamashita N, Ninomiya Y, Moon CK, Eriksson U, Eto K. Retinoic acid stage-dependently alters the migration pattern and identity of midbrain neural crest cells. Development 1995;121:825? 837. Lo?pez SL, Dono R, Zeller R, Carrasco AE. Differential effects of retinoic acid and a retinoid antagonist on the spatial distribution of the homeoprotein Hoxb-7 in vertebrate embryos. Dev. Dyn. 1995;204: 457?471. Loring JF, Erickson CA. Neural crest cell migratory pathways in the trunk of the chick embryo. Dev. Biol. 1987;121:220?36. Lotan R. Effects of vitamin A and its analogous (retinoids) on normal and neoplastic cells. Biochem. Biophys. Acta 1980;605:33?93. Mallo M. Retinoic acid disturbs mouse middle ear development in a staged-dependent fashion. Dev. Biol. 1997;184:175?186. Marshall H, Nonchenv S, Sham MH, Muchamore I, Lumsden A, Krumlauf R. Retinoic acid alters hindbrain Hox code and induces transformation of rhombomeres 2/3 into a 4/5 identity. Nature 1992;360:737?741. Marshall H, Morrison A, Studer M, Po?pperl H, Krumlauf R. Retinoids and Hox genes. FASEB J. 1996;10:969?978. Matsuo I, Kuratani S, Kimura C, Takeda N, Aizawa S. Mouse Otx2 functions in the formation and patterning of rostral head. Genes Dev. 1995;9:2646?2658. Morriss-Kay G. Retinoic acid and craniofacial development: Molecules and morphogenesis. Bioessays 1993;15:9?15. Morris-Kay G, Murphy M, Hill P, Davidson D. Effects of retinoic acid excess on expression of Hox-2.9 and Krox-20 and on morphological segmentation of the hindbrain of mouse embryo. EMBO J. 1991;10: 2985?2995. Noden DM. An analysis of the migratory behavior of avian cephalic neural crest cells. Dev. Biol. 1975;42:106?130. Osmond MK, Butler AJ, Voon FCT, Bellairs R. The effects of retinoic acid on heart formation in early chick embryo. Development 1991; 113:1405?1417. Osumi-Yamashita N. Retinoic acid and mammalian craniofacial morphogenesis. J. Biosci. 1996;21:313?327. Papalopulu N, Clarke JDW, Bradley L, Wilkinson D, Krumlauf R, Holder N. Retinoic acid causes abnormal development and segmental patterning of the anterior hindbrain in Xenopus embryos. Development 1991;113:1145?1158. Rickmann M, Fawcett JW, Keynes R. The migration of crest cells and the growth of motor axons through the rostral half of the chick somite. J. Embryol. Exp. Morphol. 1985;90:437?455. Romer AS. The vertebrate as a dual animal?Somatic and visceral. Evol. Biol. 1972;6:121?156. Shawlot W, Behringer RR. Requirement for Lim1 in head-organizer function. Nature 1995;374:425?430. Simeone A, Acampora D, Arcioni L, Andrews PW, Boncinelli E, Mavilio F. Sequential activation of HOX2 homeobox genes by retinoic acid in human embryonal carcinoma cells. Nature 1990;346:763?766. Simeone A, Acampora D, Nigro V, Faiella A, D?Esposito M, Stornaiuolo A, Mavilio F, Boncinelli E. Differential regulation by retinoic acid of the homeobox genes of the four HOX loci in human embryonal carcinoma cells. Mech. Dev. 1991;33:215?227. RETINOIC ACID AND LAMPREY DEVELOPMENT Simeone A, Avantaggiato V, Moroni MC, Mavilio F, Arra C, Cotelli F, Nigro V, Acampora D. Retinoic acid induces stage-specific anteroposterior transformation of rostral central nervous system. Mech. Dev. 1995;51:83?98. Stainier DYR, Fishman MC. Patterning the zebrafish heart tube: Acquisition of anteroposterior polarity. Dev. Biol. 1992;153:91?101. Starck D. Embryologie?Ein Lehrbuch auf Allgemein Biologischer Grundlage, 3rd Ed. Stuttgart: Georg Thieme Verlag, 1975. Steinberg M. A nonnutrient culture medium for amphibian embryonic tissues. Carnegie Inst. Washington Year Book 1957;56:347?348. Sucov HM, Dyson E, Gumeringer CL, Price J, Chien KR, Evans RM. RXRa mutant mice establish a genetic basis for vitamin A signaling in heart morphogenesis. Genes Dev. 1994;8:1007?1018. Sugi Y, Lough J. Anterior endoderm is a specific effector of terminal cardiac myocyte differentiation of cells from the embryonic heart forming region. Dev. Dyn. 1994;200:155?162. Sundin O, Eichele G. An early marker of axial pattern in the chick embryo and its respecification by retinoic acid. Development 1992; 114:841?852. Sundin O, Janocha R, Eichele G. Retinoids and pattern formation in vertebrate embryos. In: Robertson E, Maxfield FR, Vogel HJ, eds. Cell-Cell Signalling in Vertebrate Development. New York: Academic Press, 1993:157?178. Tahara Y. Normal stages of development in the lamprey, Lampetra reissneri (Dybowski). Zool. Sci. 1988;5:109?118. 51 Taira M, Otani H, Jamrich M, David IB. Expression of LIM class homeobox gene Xlim-1 in pronephros and CNS cell lineages of Xenopus embryo is affected by retinoic acid and exogastrulation. Development 1994;120:1525?1536. Teillet AM, Kalcheim C, Le Douarin NM. Formation of the dorsal root ganglia in the avian embryo: Segmental origin and migratory behavior of neural crest progenitor cells. Dev. Biol. 1987;120:329? 347. van Wijhe JW. U?ber die Mesodermsegmente und die Entwicklung der Nerven des Selachierkopfes. Ver. Akad. Wiss. Amsterdam, Groningen 1882;1882:1?50. Veit O. Beitra?ge zur Kenntnis des Kopfes der Wirbeltiere. III. Beobachtungen zur Fru?hentwicklung des Kopfes von Petromyzon planeri. Morphol. Jb. 1939;84:86?107. Wada H, Holland PWH, Satoh N. Origin of patterning in neural tubes. Nature 1996;384:123. Willey A. The later larval development of amphioxus. Q. J. Microsc. Sci. 1891;32:183?234. Wilson JG, Warkany J. Aortic-arch and cardiac anomalies in the offspring of vitamin A deficient rats. Am. J. Anat. 1949;85:113?155. Zhang Z, Balmer JE, L黲lie A, Fromm SH, Blomhoff R. Specific teratogenic effects of different retinoic acid isomers and analogs in the developing anterior central nervous system of zebrafish. Dev. Dyn. 1996;206:73?86.