Molecular mechanisms controlling midline crossing by precerebellar neurons.

Precerebellar neurons of the inferior olive (IO) and lateral reticular nucleus (LRN) migrate tangentially from the rhombic lip toward the floor plate following parallel pathways. This process is thought to involve netrin-1 attraction. However, whereas the cell bodies of LRN neurons cross the midline, IO neurons are unable to do so. In many systems and species, axon guidance and cell migration at the midline are controlled by Slits and their receptor Robos. We showed previously that precerebellar axons and neurons do not cross the midline in the absence of the Robo3 receptor. To determine whether this signaling by Slits and the two other Robo receptors, Robo1 and Robo2, also regulates precerebellar neuron behavior at the floor plate, we studied the phenotype of Slit1/2 and Robo1/2/3 compound mutants. Our results showed that many IO neurons can cross the midline in absence of Slit1/2 or Robo1/2, supporting a role for midline repellents in guiding precerebellar neurons. We also show that these molecules control the development of the lamellation of the inferior olivary complex. Last, the analysis of Robo1/2/3 triple mutants suggests that Robo3 inhibits Robo1/2 repulsion in precrossing LRN axons but not in IO axons in which it has a dominant and distinct function.

Origin of the earliest correlated neuronal activity in the chick embryo revealed by optical imaging with voltage-sensitive dyes.

Spontaneous correlated neuronal activity during early development spreads like a wave by recruiting a large number of neurons, and is considered to play a fundamental role in neural development. One important and as yet unresolved question is where the activity originates, especially at the earliest stage of wave expression. In other words, which part of the brain differentiates first as a source of the correlated activity, and how does it change as development proceeds? We assessed this issue by examining the spatiotemporal patterns of the depolarization wave, the optically identified primordial correlated activity, using the optical imaging technique with voltage-sensitive dyes. We surveyed the region responsible for the induction of the evoked and spontaneous depolarization waves in chick embryos, and traced its developmental changes. The results showed that the wave initially originated in a restricted area near the obex and was generated by multiple regions at later stages. We suggest that the upper cervical cord/lower medulla near the obex is the kernel that differentiates first as the source of the correlated activity, and that regional and temporal differences in neuronal excitability might underlie the developmental profile of wave generation in early chick embryos.

The development of descending projections from the brainstem to the spinal cord in the fetal sheep.

BACKGROUND: Although the fetal sheep is a favoured model for studying the ontogeny of physiological control systems, there are no descriptions of the timing of arrival of the projections of supraspinal origin that regulate somatic and visceral function. In the early development of birds and mammals, spontaneous motor activity is generated within spinal circuits, but as development proceeds, a distinct change occurs in spontaneous motor patterns that is dependent on the presence of intact, descending inputs to the spinal cord. In the fetal sheep, this change occurs at approximately 65 days gestation (G65), so we therefore hypothesised that spinally-projecting axons from the neurons responsible for transforming fetal behaviour must arrive at the spinal cord level shortly before G65. Accordingly we aimed to identify the brainstem neurons that send projections to the spinal cord in the mature sheep fetus at G140 (term = G147) with retrograde tracing, and thus to establish whether any projections from the brainstem were absent from the spinal cord at G55, an age prior to the marked change in fetal motor activity has occurred. RESULTS: At G140, CTB labelled cells were found within and around nuclei in the reticular formation of the medulla and pons, within the vestibular nucleus, raphe complex, red nucleus, and the nucleus of the solitary tract. This pattern of labelling is similar to that previously reported in other species. The distribution of CTB labelled neurons in the G55 fetus was similar to that of the G140 fetus. CONCLUSION: The brainstem nuclei that contain neurons which project axons to the spinal cord in the fetal sheep are the same as in other mammalian species. All projections present in the mature fetus at G140 have already arrived at the spinal cord by approximately one third of the way through gestation. The demonstration that the neurons responsible for transforming fetal behaviour in early ontogeny have already reached the spinal cord by G55, an age well before the change in motor behaviour occurs, suggests that the projections do not become fully functional until well after their arrival at the spinal cord.

Generation of a novel functional neuronal circuit in Hoxa1 mutant mice.

Early organization of the vertebrate brainstem is characterized by cellular segmentation into compartments, the rhombomeres, which follow a metameric pattern of neuronal development. Expression of the homeobox genes of the Hox family precedes rhombomere formation, and analysis of mouse Hox mutations revealed that they play an important role in the establishment of rhombomere-specific neuronal patterns. However, segmentation is a transient feature, and a dramatic reconfiguration of neurons and synapses takes place during fetal and postnatal stages. Thus, it is not clear whether the early rhombomeric pattern of Hox expression has any influence on the establishment of the neuronal circuitry of the mature brainstem. The Hoxa1 gene is the earliest Hox gene expressed in the developing hindbrain. Moreover, it is rapidly downregulated. Previous analysis of mouse Hoxa1(-/-) mutants has focused on early alterations of hindbrain segmentation and patterning. Here, we show that ectopic neuronal groups in the hindbrain of Hoxa1(-/-) mice establish a supernumerary neuronal circuit that escapes apoptosis and becomes functional postnatally. This system develops from mutant rhombomere 3 (r3)-r4 levels, includes an ectopic group of progenitors with r2 identity, and integrates the rhythm-generating network controlling respiration at birth. This is the first demonstration that changes in Hox expression patterns allow the selection of novel neuronal circuits regulating vital adaptive behaviors. The implications for the evolution of brainstem neural networks are discussed.

Anatomic relationships of the human arcuate nucleus of the medulla: a DiI-labeling study.

The arcuate nucleus (ARC) at the ventral surface of the human medulla has been historically considered a precerebellar nucleus. More recently, it has been implicated in central chemoreception, cardiopulmonary coupling and blood pressure responses. A deficiency of the ARC has been reported in a subset of putative human developmental disorders of ventilatory function. To investigate anatomic relationships of the ARC with brainstem regions involved in cardiorespiratory control, we applied crystals of DiI, a lipophilic dye which labels cells and cell processes by lateral diffusion along cell membranes, to 23 paraformaldehyde-fixed human fetal brainstems at 19 to 22 weeks postconceptional age. After 7 to 15.5 months diffusion, serial frozen sections were examined by florescence microscopy. DiI diffusion from the ARC labeled fibers and cell bodies in the medullary raphé, and the external arcuate fibers. Diffusion from the medullary raphé [corrected] labeled the reticular formation, medullary raphé, and the ARC. Diffusion from the pyramid and the basis pontis (negative control) labeled the corticospinal tract, with no labeling of the medullary raphé or ARC. The results suggest the existence of cellular connections between the ARC and the caudal raphé, a region implicated in cardiorespiratory control.

Neurogenesis in the epithalamus, dorsal thalamus and ventral thalamus of the rat: an autoradiographic and cytological study.

Times of final mitotic division for neurons of the epithalamic, dorsal thalamic and subthalamic nuclei of the rat were determined with the aid of thymidine-H3 autoradiography. Intensely labelled neurons were observed in the brains of animals injected with radiochemical from days 13 to 19 of gestation. The pattern of distribution of the labelled neurons indicated that neurogenesis in the regions followed caudorostral, lateromedial and ventrodorsal neurogenetic gradients, all of which were found to operate simultaneously. Since neurogenesis in the epithalamus, subthalamus and caudolateral thalamic regions began on days 13 and 14 of gestation, the ventrodorsal and lateromedial proliferative gradients were clearly discerned only within the ventral and dorsal thalamus exclusive of the epithalamus. These directional neurogenetic gradients were apparent throughout the entire thalamus and within individual thalamic nuclei. No neurogenetic pattern based upon neuronal size was observed, i.e., large neurons were not preferentially formed earlier than smaller ones. Detailed information has also been provided on the cytological character of each thalamic nucleus.

Identification of early neurons in the brainstem and spinal cord: I. An autoradiographic study in the chick.

Early stages in chick neurogenesis were investigated with tritiated thymidine (3H-Tdr) autoradioraphy to determine the location and identity of the first neurons produced for the central nervous system. These cells have been shown to arise prior to neural tube closure (Sechrist, '75). Chicks were treated at selected intervals between 20 and 72 hours of incubation with 3H-Tdr in a modified pulse-labeling technique, and terminated on the 18th day of embryonic development (E18), when neuronal types could be determined. Some of the earliest neurons start their final DNA synthesis before 20 hours of incubation (head process, Hamburger-Hamilton stage 5). These are primarily medium-sized cells of the reticular formation in the medulla and at the diencephalic-mesencephalic junction, but also in the intermediate zone of the spinal cord. Motor neurons of the brainstem and spinal cord begin to appear next, after 26-28 hours incubation; the first sensory neurons arise after 32 hours. Other workers (Ramon y Cajal, '60; Tello, '23; Windle and Austin, '36) found that neurons of the reticular formation were the first to differentiate neurofibrils, during the latter part of E2, indicating that fibrillogenesis in these cells begin about 24 hours after the initial cessation of DNA replication.

Development of the brain stem in the rat. II. Thymidine-radiographic study of the time of origin of neurons of the upper medulla, excluding the vestibular and auditory nuclei.

Groups of pregnant rats were injected with two successive daily doses of 3H-thymidine from gestational days 12 and 13 (E12 + 13) until the day before birth (E21 + 22). In radiographs from adult progeny of these rats the proportion of neurons generated on specific days was determined in the major nuclei of the upper medulla, with the exception of the vestibular and auditory nuclei. The neurons of the motor nuclei are generated over a brief period. Neurons of the retrofacial nucleus are produced first, with more than 60% of the cells arising on day E11 or earlier. Peak generation time of abducens neurons is day E12 and of the neurons of the facial nucleus is day E13. In contrast, the neurons of the superior salivatory nucleus are produced late, predominantly on day E15 and some on day E16. The generation of the (sensory relay) neurons of the nucleus oralis of the trigeminal complex takes place over an extended period between days E12 and E15; the last generated cells include the largest neurons of this nucleus. Neurons of the raphe magnus are produced between days E11 and E14, the neurons of the rostral medullary reticular formation between days E12 and E15. The latest generated neurons of the upper medulla (excluding the cochlear nuclei) belong to a structure identified as the granular layer of the raphe. Combining these results with those of the preceding paper (Altman and Bayer, '80a) and with additional data, it is postulated that the laterally and ventrally situated motor nucleus of the trigeminal, the facial nucleus, and the nucleus ambiguous form a single longitudinal zone of branchial motor neurons with a rostral-to-caudal cytogenetic gradient. In contrast, the medially and dorsally situated (juxtaventricular) hypoglossal nucleus and abducens nucleus (together with the other nuclei of the ocular muscles) form a longitudinal somatic motor zone with a caudal-to-rostral gradient. The dorsal nucleus of the vagus and the superior salivatory nucleus may constitute a preganglionic motor zone, also with a caudal-to-rostral cytogenetic gradient.

Development of the brain stem in the rat. IV. Thymidine-radiographic study of the time of origin of neurons in the pontine region.

Groups of pregnant rats were injected with two successive daily doses of 3H-thymidine from gestational day 12 and 13 (E12 + 13) until the day before parturition (E21 + 22) in order to label in their embryos the proliferating precursors of neurons. At 60 days of age the proportion of neurons generated (or no longer labeled) on specific embryonic days was determined quantitatively in 14 nuclei of the pontine region. Peak production time of neurons of the trigeminal mesencephalic nucleus was on day E11 or earlier, with a small proportion generated on day E12. Peak production time of the trigeminal motor neurons was on day E12, with a small proportion produced earlier. Neurons of the principal sensory nucleus were generated between days E13 and E16, with a peak on day E14; the late-produced neurons tended to belong to a class of intermediate and large cells. The bulk of the neurons of the supratrigeminal and infratrigeminal nuclei arose on days E15 and E16. Neurons of the locus coeruleus are produced mostly on day E12, with about 20% of the cells arising on day E13. The bulk of the neurons of the dorsal tegmental nucleus (Gudden's) are produced between days E13 and E15, whereas most of the neurons of the deep (ventral) tegmental nucleus are produced on day E15. A dorsal-to-caudal gradient was also obtained between the dorsal and ventral nuclei of the lateral lemniscus, the neurons of the former being generated between days E12 and E15; the latter, between days E13 and E17. The neurons of both the pars lateralis and the pars medialis of the parabrachial nucleus were produced simultaneously between days E13 and E15, with a peak on day E13. The heterogeneous collection of neurons of the pontine paramedial reticular formation was produced for day E11 (or earlier) until day E15. Finally, the neurons of the raphe pontis parvicellularis were generated at an even rate between days E13 and E15, whereas the bulk of the neurons of the raphe pontis magnocellularis were produced on days E15 and E16. On the basis of datings obtained for 9 subdivisions of the entire brain stem trigeminal complex, hypotheses were offered of the cytogenetic components of the system. The sequence of neuron production in the dorsal and deep tegmental nuclei was related to their connections with divisions of the mammillary and habenular nuclei on a "first come-first serve" basis.

The onset and development of descending pathways to the spinal cord in the chick embryo.

The ontogenetic development of afferent (supraspinal and propriospinal) as well as efferent (ascending) fiber connections of the spinal cord was examined following the injection of horseradish peroxidase (HRP) or wheat germ agglutinin HRP (WGA-HRP) into the cervical and lumbar spinal cords (or brains) of embryos ranging in age from 4 to 14 days of incubation. A few cells were first reliably retrogradely labelled in the pontine reticular formation on embryonic day (E) 4 and E5 following the injection of WGA-HRP into the cervical and lumbar spinal cord, respectively. Propriospinal projections to the lumbar spinal cord, originating from brachial spinal cord, were found by E5, and from the cervical spinal cord by E5.5. Ascending fibers arising from neurons in the lumbar spinal cord could be followed to rostral mesencephalic levels in E5 embryos. Thus, the earliest supraspinal, propriospinal, and ascending fiber connections appear to be formed almost simultaneously. Retrogradely labelled cells were found in the raphe, reticular, vestibular, interstitial, and hypothalamic nuclei in E5.5 embryos following lumbar injections of WGA-HRP. Except for neurons in cerebellar nuclei, all the cell groups of origin that project to the cervical spinal cord of posthatching chicks were also retrogradely labelled by E8. There was a delay in the time of appearance of the projections from various regions of the brain stem to the lumbar versus the cervical spinal cord, ranging from 0.5 to 7 days, but typically of about 3 days duration. A large number of cells located in the ventral hypothalamic region, just dorsal to the optic chiasma, were found to be labelled following cervical HRP injection between E6 and E10. These cells may represent transient projections that are present only during embryonic stages since no labelled cells were found in this region in the newly-hatched chick.

Development of the brain stem in the rat. I. Thymidine-radiographic study of the time of origin of neurons of the lower medulla.

Groups of pregnant rats were injected with two successive daily doses of 3H-thymidine from gestational days 12 and 13 (E12 + 13) until the day before birth (E21 + 22). In adult progeny of the injected rats the proportion of neurons generated on specific days was determined quantitatively in the major nuclei of the lower medulla. The earliest generated cells form two motor nuclei: the hypoglossal and dorsal vagal nuclei. The bulk of hypoglossal neurons are produced on day E12, with a small proportion earlier; the bulk of dorsal vagal neurons are produced, likewise, on day E12, with a small proportion on day E13. The neurons of the third motor nucleus of the region, the ambiguous, are generated later, with a peak on day E15. Neurons of the sensory relay nuclei, the gracilis, cuneatus, and solitarius are produced over a more extended period, with peaks on day E13; the exception was the external cuneate nucleus in which peak generation time was on day E15. In the caudal nucleus of the trigeminal complex, neurons of the subnucleus magnocellularis arise earliest, with a peak on day E14, and those of the subnucleus marginalis last, with a peak on day E15, and extending into day E16. The neurons of the nuclei raphe pallidus and obscurus, and of the dorsal and ventral portions of the caudal medullary reticular formation, are produced between days E12 and E15, without any obvious peaks. The neurons of the nucleus parasolitarius and the nucleus of Roller are produced relatively late, and the area postrema contains a germinal cell population throughout the embryonic period, presumably supplying cells to the choroid plexus of the fourth ventricle. On the basis of absolute datings, duration of neuron production, intranuclear and internuclear gradients, and other criteria, it is postulated that the neurons of the lower medulla are derived from at least eight different cytogenetic zones.

Ontogeny of monoamine neurons in the locus coeruleus, raphe nuclei and substantia nigra of the rat. II. Synaptogenesis.

Synaptogenesis was studied in the monoamine (MA) cell groups locus coeruleus (LC), dorsal and medial raphe nuclei (RN) and substantia nigra, zona compacta (SN) between day 18 of gestation and postnatal day 60 using ethanolic phosphotungstic acid (E-PTA) to visualize synaptic profiles. Nuclear area, and cellular packing density (inversely proportional to area of neuropil) were also determined. As determined using the E-PTA method, synaptogenesis begins in the neuropil of the SN first, on or before 18 days of gestation, and in the LC and RN at 19 days. Synaptogenesis on MA cell perikarya is first observed in the SN, on or before 18 days, and in the LC and RN at 20 days. The onset of somatic synaptogenesis coincides with the beginning of nuclear growth and development of the neuropil (decrease in cellular packing density) in all MA cell areas, raising the possibility of common factors in the initiation of these processes. Nonsynaptic contacts precede the appearance of synaptic profiles both in the neuropil and on the somata of the MA cells of the LC, RN and SN, and may represent precursors of mature synapses or desmosome-like contacts. Somatosomatic nonsynaptic contacts occur only prenatally between adjacent MA neurons in the LC, RN and SN. Although some synaptogenesis occurs prenatally in these MA cell groups (indiciating that these parts of the MA circuitry may be functional before birth), most of this synaptogenesis occurs postnatally and continues into adulthood. Since such synaptogenesis does not begin until 2-4 days prior to birth, whereas these neurons and their processes exhibit MA fluorescence as early as 12-14 days of gestation, they apparently are capable of synthesizing transmitter and proliferating terminals before they themselves are innervated.

Pathway specificity of reticulospinal and vestibulospinal projections in the 11-day chicken embryo.

The organization of the axonal pathways of reticulospinal and vestibulospinal projections in the 11-day chicken embryo was ascertained through retrograde tracing experiments. An in vitro preparation of the brainstem and cervical spinal cord facilitated precisely localized tracer applications. Single- and double-labelling experiments involving high cervical injections of tracers in combination with selective lesions defined the specific pathways by which different brainstem neurons project to the spinal cord. Coherent, and in many cases distinct, groups of reticulospinal and vestibulospinal neurons could thus be identified on the basis of their position and projection pathway. The organization of these groups and their projections in the 11-day chicken embryo is similar to that in avian and other vertebrate adults and therefore serves as a reference point for studies of pathfinding by bulbospinal axons during early development.

Development of the precerebellar nuclei in the rat: III. The posterior precerebellar extramural migratory stream and the lateral reticular and external cuneate nuclei.

Sequential thymidine radiograms from rats injected on day E15 and killed thereafter at daily intervals up to day E22 were analyzed to trace the migratory routes and settling patterns of neurons of the lateral reticular nucleus and the external cuneate nucleus. The neurons of the lateral reticular and external cuneate nuclei originate in the primary precerebellar neuroepithelium at the same site as the inferior olivary neurons but follow a different migratory route. The labeled young neurons that are produced on day E15 (the last one-third of the total) join the posterior precerebellar extramural migratory stream. The cells move circumferentially over the wall of the medulla in a ventral direction and by day E17 reach the midline and cross it beneath the inferior olive. The crossing cells apparently continue to migrate circumferentially on the opposite side. One complement of these cells begins to form a ventrolateral extramural condensation on day E19. By day E20 some cells begin to penetrate the parenchyma and settle as neurons of the lateral reticular nucleus. The settling of the lateral reticular neurons continues on the following day, and by day E22 all the cells destined for the lateral reticular nucleus have penetrated the parenchyma. A dorsomedial-to-ventrolateral neurogenetic gradient is indicated for the settling lateral reticular neurons. Another complement of migrating cells continues dorsally and forms a condensation on day E19 that we interpret as the external cuneate component of the crossed stream. These cells begin to penetrate the parenchyma on day E20, and by days E21 and E22 two components of the external cuneate nucleus are identifiable-the dorsal and ventral external cuneate nuclei. The neurons of the lateral reticular and external cuneate nuclei differ from neurons of all the other precerebellar nuclei in that their cerebellar projection is predominantly ipsilateral. We speculate that the axons of all precerebellar neurons are genetically specified to cross the midline ventrally to provide a contralateral efferent projection, but this is modified in the case of the ipsilaterally projecting lateral reticular and external cuneate neurons by the cell bodies following their neurites to the opposite side.

Neurogenesis in the vocalization pathway of Xenopus laevis.

We examined possible contributions of neurogenesis to sex differences in the vocalization pathway of the South African clawed frog, Xenopus laevis. Birthdates of neurons were obtained from autoradiograms of animals receiving tritiated thymidine from gastrulation through 1 month after metamorphosis. Thymidine availability studies showed that 80% of the [3H]-thymidine injected into embryos and tadpoles was incorporated into the DNA of dividing cells within 3 hours. We observed 3 patterns of neurogenesis: late-short, a short burst of proliferation occurred late in development in the anterior preoptic area, the ventromedial nucleus of the thalamus, and the pretrigeminal nucleus of the dorsal tegmental area of the medulla; protracted-bimodal, a prolonged period of proliferation with an early and a late peak in the number of labeled cells occurred in the ventral striatum and in the ventrolateral and posterior nuclei of the thalamus; protracted-unimodal, a prolonged period of proliferation with a single early peak occurred in the inferior reticular formation and in the medial and lateral nucleus IX-X (containing laryngeal motor neurons). There were no differences between sexes in the number of tritiated thymidine labeled cells in any nucleus. The difference in nucleus IX-X neuron number in adults does not appear to result from sex differences in the proliferation of these cells during development. Since neurons in the vocalization pathway do not exhibit androgen receptors until after neurogenesis is complete, we also conclude that androgen probably does not regulate the genesis of these cells.

Stimulation of the brainstem reticular formation evokes locomotor activity in embryonic chicken (in ovo).

This study was designed to examine the period of embryonic chick development during which descending brainstem-spinal projections, originating from defined avian brainstem locomotor regions, become functionally active. Locomotor activity was examined using a new in ovo preparation for the focal electrical stimulation of embryonic brainstem locomotor regions. Embryos or hatchlings were anesthetized and mounted in a stereotaxic apparatus. Leg and wing muscle electromyographic (EMG) recordings were used to monitor any brainstem-stimulated motor activity. At present, we have been successful in demonstrating coordinated brainstem-evoked locomotion in embryos as early as embryonic day 15. The patterns of evoked locomotor activity were similar to locomotion evoked in hatchling chicks and were of 4 types: (1) alternating hindlimb movements ('stepping'), (2) synchronous (in-phase) hindlimb movements ('hatching'), (3) synchronous wing movements ('flapping'), and (4) simultaneous 'stepping' and 'flapping'. The cycle durations of evoked embryonic hindlimb movements are shorter than those observed for hatchling chicks. The present results are the first direct demonstration of functional connections between descending supraspinal neurons and spinal locomotor circuits at such an early stage of embryonic development. With modifications in technique, it may be possible to demonstrate functional connections at even earlier stages of embryonic development.

[3H]thymidine autoradiographic study in the transit part from the spinal cord to the medulla oblongata of the chick embryo--the ontogenetic relation between the reticular formation and the spinal cord.

In order to examine the ontogenetic structural continuity of the reticular formation to the spinal gray matter, the time of origin of neurons was investigated in the transit part from the spinal cord to the medulla of chick embryos with tritriated thymidine autoradiography. The reticular formation was found to be neurogenetically divided into parvocellular reticular formation (RP), dorsal part of the magnocellular reticular formation (RME) and ventral part of the magnocellular reticular formation (RMI). Neurons in the RP differentiated synchronously with those in the neck and base of the dorsal horn, neurons in the RME with those in the zona intermedia, and neurons in the RMI with those in the ventral horn, suggesting that each synchronously developing region forms an ontogenetic structural unit.

Compartments in the lamprey embryonic brain as revealed by regulatory gene expression and the distribution of reticulospinal neurons.

The vertebrate neural tube consists of a series of neuromeres along its anteroposterior axis. Between amphioxus that possesses no neuromeres and gnathostomes, the lamprey occupies a critical position in the phylogeny for the origin of the segmented brain. To clarify the rhombomeric configuration of the Japanese lamprey, Lampetra japonica, we injected rhodamine- and fluorescein-labeled dextrans into the larval spinal cord, and retrogradely labeled the reticulospinal neurons. We also isolated prosomere marker genes from the embryonic cDNA library of L. japonica, and performed in situ hybridization on the embryonic brain. Of the genes examined, LjOtxA, LjPax6, LjPax2/5/8, LjDlx1/6, and LjTTF-1 were expressed in clearly demarcated polygonal domains. In the telencephalon, LjDlx1/6, LjPax6, and a putative paralogue of LjEmx were expressed in different domains; the LjEmx paralogue was expressed in the dorsal region, and LjDlx1/6 and LjPax6 in a complimentary fashion of the middle part. These expression patterns implied existence of a tripartite configuration of the lamprey telencephalon similar to that in gnathostomes. All these evidences strongly suggest that the segmental and compartmental architecture of the vertebrate brain was already established before the divergence of agnathans and gnathostomes.

Observations on the development of descending pathways from the brain stem to the spinal cord in the clawed toad Xenopus laevis.

Anurans such as the clawed toad Xenopus laevis offer a unique opportunity to study the ontogeny of descending pathways to the spinal cord. Their transition from aquatic limbless tadpole to juvenile toad occurs over a protracted period time during which the animal is accessible for experimental studies. In Xenopus laevis tadpoles the development of descending pathways has been studied from early limb-bud stage on (stage 50) with the aid of HRP slow-release gels. In stage 50, cells of origin of descending supraspinal pathways were already present throughout the reticular formation (including the interstitial nucleus of the fasciculus longitudinalis medialis) and in the vestibular nuclear complex. Also the giant Mauthner cells project to the cord at this stage. A spinal projection from the anuran homologue of the nucleus ruber of higher vertebrates does not appear before stage 58, i.e., when the hindlimbs are used for locomotion. Hypothalamospinal projections appear for the first time at stage 57. These observations in Xenopus laevis tadpoles suggest that reticulospinal and vestibulospinal projections innervate spinal segments very early in development, whereas the anuran red nucleus projects spinal ward definitely later in development.

Observations on the development of cerebellar afferents in Xenopus laevis.

The development of cerebellar afferents has been studied in the clawed toad, Xenopus laevis, from stage 46 to 64, with the horseradish peroxidase retrograde tracer technique. Already in stage 48 tadpoles, i.e. before the formation of the limbs, a distinct set of cerebellar afferents was found. Vestibulocerebellar (mainly arising bilaterally in the nucleus vestibularis caudalis) and contralateral olivo-cerebellar projections dominate. Secondary trigeminocerebellar (from the descending nucleus of the trigeminal nerve) and reticulocerebellar connections were also found. At stage 50, spinocerebellar projections appear originating from cervical and lower thoracic/upper lumbar levels. The cells of origin of the spinocerebellar projection can be roughly divided in two neuronal types: ipsilaterally projecting large cells, which show a marked resemblance to primary motoneurones ('spinal border cells') and smaller contralaterally projecting neurons. Primary spinocerebellar projections from spinal ganglion cells could not be demonstrated. At stage 50, a possible anuran homologue of the mammalian nucleus prepositus hypoglossi was found to project to the cerebellum. In only one of the experiments labeled neurons were found in the contralateral mesencephalic tegmentum. At none of the studied stages a raphecerebellar projection could be demonstrated. It appears that already early in cerebellar development, before the formation of the limbs, most of the cerebellar afferents as found in adult Xenopus laevis are present.