Vestibular projections to the thalamus of the pigeon: an anatomical study

Morphology of the "prefrontal" nidopallium caudolaterale in the long-distance night-migratory Eurasian blackcap (Sylvia atricapilla)

Migrating birds have developed remarkable navigational capabilities to successfully master biannual journeys between their breeding and wintering grounds. To reach their intended destination, they need to calculate navigational goals from a large variety of natural directional and positional cues to set a meaningful motor output command. One brain area, which has been associated with such executive functions, is the nidopallium caudolaterale (NCL), which, due to its striking similarities in terms of neurochemistry, connectivity and function, is considered analogous to the mammalian prefrontal cortex. To establish a baseline for further analyses elucidating the neuronal correlates underlying avian navigation, we performed quantitative and qualitative analyses of dopaminergic fibres in the brains of long-distance night-migratory Eurasian blackcaps (Sylvia atricapilla). We identified four regions in the caudal telencephalon, each of which was characterized by its specific dopaminergic innervation pattern. At least three of them presumably constitute subareas of the NCL in Eurasian blackcaps and could thus be involved in integrating navigational input from different sensory systems. The observed heterogeneity and parcellation of the NCL subcompartments in this migratory species could be a consequence of the special demands related to navigation.

Pretecto- and ponto-cerebellar pathways to the pigeon oculomotor cerebellum follow a zonal organization

Both birds and mammals have relatively large forebrains and cerebella. In mammals, there are extensive sensory-motor projections to the cerebellum through the pontine nuclei originating from several parts of the cerebral cortex. Similar forebrain-to-cerebellum pathways exist in birds, but the organization of this circuitry has not been studied extensively. Birds have two nuclei at the base of the brainstem that are thought to be homologous to the pontine nuclei of mammals, the medial and lateral pontine nuclei (PM, PL). Additionally, birds are unique in that they have a pretectal nucleus called the medial spiriform nucleus (SpM) that, like the pontine nuclei, also receives projections from the forebrain and projects to the oculomotor cerebellum (OCb; folia VI to VIII). The OCb also receives input from the pretectal nucleus lentiformis mesencephali (LM), which analyzes visual optic flow information resulting from self-movement. In this study, we used single or double injections of fluorescent tracers to study the organization of these inputs from PM, PL, SpM and LM to the OCb in pigeons. We found that these inputs follow a zonal organization. The most medial zone in the OCb, zone A1, receives bilateral inputs from the lateral SpM, PL and LM. Zones A2 and C receive a bilateral projection from the medial SpM, and a mostly contralateral projection from PM and LM. We discuss how the pathway to zone A1 processes mainly visuo-motor information to spinal premotor areas, whereas the pathways to zone A2/C processes somato-motor and visuo-motor information and may have a feedback/modulatory role.

Thalamostriatal and cerebellothalamic pathways in a songbird, the Bengalese finch

The thalamostriatal system is a major network in the mammalian brain, originating principally from the intralaminar nuclei of thalamus. Its functions remain unclear, but a subset of these projections provides a pathway through which the cerebellum communicates with the basal ganglia. Both the cerebellum and basal ganglia play crucial roles in motor control. Although songbirds have yielded key insights into the neural basis of vocal learning, it is unknown whether a thalamostriatal system exists in the songbird brain. Thalamic nucleus DLM is an important part of the song system, the network of nuclei required for learning and producing song. DLM receives output from song system basal ganglia nucleus Area X and sits within dorsal thalamus, the proposed avian homolog of the mammalian intralaminar nuclei that also receives projections from the cerebellar nuclei. Using a viral vector that specifically labels presynaptic axon segments, we show in Bengalese finches that dorsal thalamus projects to Area X, the basal ganglia nucleus of the song system, and to surrounding medial striatum. To identify the sources of thalamic input to Area X, we map DLM and cerebellar-recipient dorsal thalamus (DTCbN ). Surprisingly, we find both DLM and dorsal anterior DTCbN adjacent to DLM project to Area X. In contrast, the ventral medial subregion of DTCbN projects to medial striatum outside Area X. Our results suggest the basal ganglia in the song system, like the mammalian basal ganglia, integrate feedback from the thalamic region to which they project as well as thalamic regions that receive cerebellar output.

Physiology of central pathways

The relative simplicity of the neural circuits that mediate vestibular reflexes is well suited for linking systems and cellular levels of analyses. Notably, a distinctive feature of the vestibular system is that neurons at the first central stage of sensory processing in the vestibular nuclei are premotor neurons; the same neurons that receive vestibular-nerve input also send direct projections to motor pathways. For example, the simplicity of the three-neuron pathway that mediates the vestibulo-ocular reflex leads to the generation of compensatory eye movements within ~5ms of a head movement. Similarly, relatively direct pathways between the labyrinth and spinal cord control vestibulospinal reflexes. A second distinctive feature of the vestibular system is that the first stage of central processing is strongly multimodal. This is because the vestibular nuclei receive inputs from a wide range of cortical, cerebellar, and other brainstem structures in addition to direct inputs from the vestibular nerve. Recent studies in alert animals have established how extravestibular signals shape these "simple" reflexes to meet the needs of current behavioral goal. Moreover, multimodal interactions at higher levels, such as the vestibular cerebellum, thalamus, and cortex, play a vital role in ensuring accurate self-motion and spatial orientation perception.

Early vestibular processing does not discriminate active from passive self-motion if there is a discrepancy between predicted and actual proprioceptive feedback

Most of our sensory experiences are gained by active exploration of the world. While the ability to distinguish sensory inputs resulting of our own actions (termed reafference) from those produced externally (termed exafference) is well established, the neural mechanisms underlying this distinction are not fully understood. We have previously proposed that vestibular signals arising from self-generated movements are inhibited by a mechanism that compares the internal prediction of the proprioceptive consequences of self-motion to the actual feedback. Here we directly tested this proposal by recording from single neurons in monkey during vestibular stimulation that was externally produced and/or self-generated. We show for the first time that vestibular reafference is equivalently canceled for self-generated sensory stimulation produced by activation of the neck musculature (head-on-body motion), or axial musculature (combined head and body motion), when there is no discrepancy between the predicted and actual proprioceptive consequences of self-motion. However, if a discrepancy does exist, central vestibular neurons no longer preferentially encode vestibular exafference. Specifically, when simultaneous active and passive motion resulted in activation of the same muscle proprioceptors, neurons robustly encoded the total vestibular input (i.e., responses to vestibular reafference and exafference were equally strong), rather than exafference alone. Taken together, our results show that the cancellation of vestibular reafference in early vestibular processing requires an explicit match between expected and actual proprioceptive feedback. We propose that this vital neuronal computation, necessary for both accurate sensory perception and motor control, has important implications for a variety of sensory systems that suppress self-generated signals.

Localization of cerebellin-2 in late embryonic chicken brain: implications for a role in synapse formation and for brain evolution

Cerebellin-1 (Cbln1), the most studied member of the cerebellin family of secreted proteins, is necessary for the formation and maintenance of parallel fiber-Purkinje cell synapses. However, the roles of the other Cblns have received little attention. We previously identified the chicken homolog of Cbln2 and examined its expression in dorsal root ganglia and spinal cord (Yang et al. [2010] J Comp Neurol 518:2818-2840). Interestingly, Cbln2 is expressed by mechanoreceptive and proprioceptive neurons and in regions of the spinal cord where those afferents terminate, as well as by preganglionic sympathetic neurons and their sympathetic ganglia targets. These findings suggest that Cbln2 may demonstrate a tendency to be expressed by synaptically connected neuronal populations. To further assess this possibility, we examined Cbln2 expression in chick brain. We indeed found that Cbln2 is frequently expressed by synaptically connected neurons, although there are exceptions, and we discuss the implications of these findings for Cbln2 function. Cbln2 expression tends to be more common in primary sensory neurons and in second-order sensory regions than it is in motor areas of the brain. Moreover, we found that the level of Cbln2 expression for many regions of the chicken brain is very similar to that of the mammalian homologs, consistent with the view that the expression patterns of molecules playing fundamental roles in processes such as neuronal communication are evolutionarily conserved. There are, however, large differences in the pattern of Cbln2 expression in avian as compared to mammalian telencephalon and in other regions that show the most divergence between the two lineages.

Identification of vestibuloocular projection neurons in the developing chicken medial vestibular nucleus

Biocytin was injected into the oculomotor, trochlear, or abducens nucleus on one side using isolated chicken brainstem preparations or brain slices to identify the medial vestibular nucleus (MVN) neurons projecting to these targets. Oculomotor nucleus injections produced retrogradely labeled neurons in the contralateral ventrolateral MVN (MVN(VL)), with few labeled neurons in the ipsilateral MVN(VL) and rarely in the dorsomedial MVN on either side. Labeled MVN(VL) neurons were identified as stellate (95%) and elongate (5%) cells. Trochlear nucleus injections produced a similar pattern of MVN neuron labeling. Abducens nucleus injections resulted in retrogradely labeled stellate (87%) and elongate (13%) neurons in the MVN(VL), which had smaller cell bodies than those projecting to the oculomotor nucleus. Anteroposteriorly, labeled MVN(VL) neurons were coextensive with the tangential nucleus, with neurons projecting to the oculomotor nucleus distributed lateral to and intermixed with the more medially situated neurons projecting to the abducens nucleus. The fundamental pattern of vestibuloocular projecting neurons was similar at both embryonic ages studied, E16 and E13. In contrast to the case in mammals, where most vestibuloocular projection neurons reside within the MVN, most retrogradely labeled neurons in these chicken preparations were found within the ventrolateral vestibular, descending vestibular, and tangential nuclei. The morphological identification and mapping of vestibuloocular projection neurons in the chicken MVN described here represents the first step in a systematic evaluation of the relationship between avian vestibuloocular neuron structure and function.

Efferent connections of the dorsomedial thalamic nuclei of the domestic chick (Gallus domesticus)

Small iontophoretic injections of the anterograde tracer Phaseolus vulgaris leucoagglutinin were placed in the thalamic anterior dorsomedial nucleus (DMA) of domestic chicks. The projections of the DMA covered the rostrobasal forebrain, ventral paleostriatum, nucleus accumbens, septal nuclei, Wulst, hyperstriatum ventrale, neostriatal areas, archistriatal subdivisions, dorsolateral corticoid area, numerous hypothalamic nuclei, and dorsal thalamic nuclei. The rostral DMA projects preferentially on the hypothalamus, whereas the caudal part is connected mainly to the dorsal thalamus. The DMA is also connected to the periaqueductal gray, deep tectum opticum, intercollicular nucleus, ventral tegmental area, substantia nigra, locus coeruleus, dorsal lateral mesencephalic nucleus, lateral reticular formation, nucleus papillioformis, and vestibular and cranial nerve nuclei. This pattern of connectivity is likely to reflect an important role of the avian DMA in the regulation of attention and arousal, memory formation, fear responses, affective components of pain, and hormonally mediated behaviors.

Visual-field-specific heterogeneity within the tecto-rotundal projection of the pigeon

The organization of the tecto-rotundal projection of the pigeon was investigated by means of anterograde and retrograde tracing techniques. Besides the known organization in tecto-rotundal connectivity, this study additionally demonstrates major variations in the ascending projections of different tectal subfields. We show that the ventral tectum opticum (TO) has significantly more projections onto the nucleus rotundus (Rt) than dorsal tectal areas. This difference coincides with differential innervation densities of afferent fibres within rotundal subregions. While ventral tectal efferents project onto the ventral and central Rt, dorsal tectal efferents mainly arborize within limited areas between the central Rt and its dorsal cap, the nucleus triangularis. Thus, the ventral TO, representing the lower and frontal field of view, exhibits a quantitatively and spatially enhanced projection onto the Rt, as compared with the dorsal TO. The data presented here demonstrate a visual field-dependent projection pattern of ascending tectal outputs onto different rotundal domains. The data are consistent with behavioural studies, demonstrating tectofugal lesions to suppress visual stimulus analysis mainly within the frontal field of view.

Afferent and efferent connections of the caudolateral neostriatum in the pigeon (Columba livia): a retro- and anterograde pathway tracing study

The avian caudolateral neostriatum (NCL) was first identified on the basis of its dense dopaminergic innervation. This fact and data from lesion studies have led to the notion that NCL might be the avian equivalent of prefrontal cortex (PFC). A key feature of the PFC is the ability to integrate information from all modalities needed for the generation of motor plans. By using antero- and retrograde pathway tracing techniques, we investigated the organization of sensory afferents to the NCL and the connections with limbic and somatomotor centers in the basal ganglia and archistriatum. Data from all tracing experiments were compared with the distribution of tyrosine-hydroxylase (TH)-immunoreactive fibers, serving as a marker of dopaminergic innervation. The results show that NCL is reciprocally connected with the secondary sensory areas of all modalities and with at least two parasensory areas. Retrograde tracing also demonstrated further afferents from the deep layers of the Wulst and from the frontolateral neostriatum as well as the sources of thalamic input. Efferents of NCL project onto parts of the avian basal ganglia considered to serve somatomotor or limbic functions. Projections to the archistriatum are mainly directed to the somatomotor part of the intermediate archistriatum. In addition, cells in caudal NCL were found to be connected with the ventral and posterior archistriatum, which are considered avian equivalents of mammalian amygdala. All afferents and projection neurons were confined to the plexus of densest TH innervation. Our results show that the NCL is positioned to amalgamate information from all modalities and to exert control over limbic and somatomotor areas. This organization might comprise the neural basis for such complex behaviours as working memory or spatial orientation.

Calretinin in pretecto- and olivocerebellar projections in the chick: immunohistochemical and experimental study

Calretinin (CaR) is a calcium-binding protein that is distributed extensively in the central nervous system. It is localized in the cell bodies and neurites of specific neuronal populations and serves, therefore, as a reliable anatomical marker. Some components of the pretectocerebellar projection, which connects specific pretectal nuclei to caudal cerebellar folia, are concerned with the cerebellar control of visual reflexes. We investigated the distribution of pretectocerebellar-projecting neurons in relation to cells that show CaR immunoreactivity. Cells that project to the cerebellar cortex in the diencephalic primary visual nuclei and in other grisea, like the nucleus spiriformis medialis and the nucleus dorsofrontalis, colocalized with those that appeared to be immunolabeled intensely with anti-CaR antiserum. To explore the hypothesis of a common developmental origin of these pretectal cerebellopetal neurons, we also investigated the development of CaR-immunopositive cells in the chick pretectum and the arrival of their fibers in the cerebellum, from 10 days of incubation (stage 36) to posthatching stages. Finally, we analyzed the source of CaR+ climbing fibers and found a subpopulation of CaR+ cells in the inferior olivary nucleus. On the whole, these results suggest that there is a common developmental origin of pretectal cerebellopetal neurons, some of which share the property of CaR expression. The functional significance of this correlation needs to be investigated.

The differential distribution of AMPA-receptor subunits in the tectofugal system of the pigeon

The tectofugal system of the pigeon was examined for the distribution of several glutamate-receptor subunits (AMPA Glu R1, Glu R2/3, Glu R4) and the calcium binding protein parvalbumin. With respect to the different antigens, a heterogeneous distribution was observed. Within the optic tectum, the Glu R1 like immunoreactivity was limited to the layers 2-5, 9, 10, and sparsely in layer 13, whereas the antibody to Glu R2/3 stained cell bodies in layers 9, 10, and very heavily in layer 13. In the rotundus only the Glu R4 antigen was expressed, while within the ectostriatal complex a large number of Glu R2/3 and a smaller contingent of Glu R4 positive neurons were stained. Quantitative analysis proved significant heterogeneities of these antigens in the mesencephalic as well as the diencephalic centre of the tectofugal pathway. The number of Glu R2/3 positive neurons undergoes a two-fold increase from the dorsal to the ventral lamina 13 of the optic tectum. Alterations in the amount of immunoreactive neurons were also observed within the rotundus, since the number of Glu R4 positive cells decreased from dorsal to ventral. Morphological differences and their correlation with functional specializations in visual information processing are discussed.

An analysis of sex and breeding stage differences in prolactin binding activity in brain and hypothalamic GnRH concentration in Wilson's phalarope, a sex role-reversed species

Wilson's Phalarope (Phalaropus tricolor) is a sex role-reversed species in which incubation of eggs and care of young is performed exclusively by the male. Plasma levels of prolactin (PRL), the hormone most associated with parental care in birds, are higher in incubating males than in nonincubating males or females. Conversely, plasma testosterone levels are reduced in males during incubation. In an attempt to characterize the physiological basis of this unusual parental care system we used quantitative film autoradiography and densitometry to measure the specific binding in vitro of 125I-ovine PRL to 12 brain regions in females, nonincubating males, and incubating males during the normal breeding season. We also measured hypothalamic chicken gonadotropin-releasing hormone I (cGnRH-I) in three brain areas in these same birds, as well as plasma levels of PRL and testosterone. Analysis revealed that cGnRH-I concentrations in the preoptic area and plasma testosterone levels were significantly lower in incubating males than in nonincubating males. Specific binding of 125I-ovine PRL was detected in choroid plexus and in several diencephalic brain regions of both sexes, with highest binding activity recorded in the dorsolateral thalamus, medial habenula, nucleus subrotundus, and preoptic area. When adjustments were made for the large number of comparisons performed, specific binding did not vary significantly by sex or breeding stage in any single brain region. However, average specific binding values in nonincubating males exceeded those of incubating males in 9 of the 11 PRL-sensitive regions examined. Increased occupancy of the receptor by endogenous PRL during incubation could have contributed to this result, since plasma PRL levels were elevated in incubating males. In addition, PRL binding activity in several of these brain regions tended to correlate negatively with plasma PRL. The two exceptions to this general pattern were the preoptic area and the lateral septum, where mean specific binding was 14-15% higher in incubating males than in nonincubating males. This raises the interesting possibility that PRL sensitivity is up-regulated during incubation in some regions of the male phalarope brain, such as the preoptic area and lateral septum, that have been implicated in PRL-modulated changes in behavior and reproductive activity during this breeding stage.

Intrinsic connections in the anterior dorsal ventricular ridge of the lizard Psammodromus algirus

We have studied the intrinsic connections of the anterior dorsal ventricular ridge (ADVR) in the lacertid lizard Psammodromus algirus by means of retrograde transport of horseradish peroxidase (HRP) and fluorescent labeling with the lipophilic carbocyanine dye DiI. We injected HRP into different regions in the ADVR arrayed in a medial-to-lateral sequence, with each consisting of three distinct superficial-to-deep zones. When HRP was injected into a given region, many labeled neurons (always located ipsilateral to the injection site) were found at all mediolateral regions of ADVR in locations rostrally distant from the injection site. DiI crystals were applied on different superficial-to-deep zones within each region. Two patterns could be recognized: DiI crystals applied on the periventricular (most superficial) zone resulted in a labeling of cells widely distributed throughout the ADVR independently of the mediolateral region of the application site, whereas DiI crystals applied on deeper zones resulted in a staining of cells mostly restricted to a narrow radial area. Results from both types of labeling confirm that the ADVR has a prominent radial component in its intrinsic organization, but they also demonstrate that some areas of the ADVR receive projections from distant, rostrally located neurons in every ipsilateral region of the ridge itself, which establishes a clear non-radial component. This organization may have important functional properties with regard to a putative integration of different sensory modalities conveyed by thalamic afferent fibers to the ADVR. Last, we analyzed some evolutionary implications of our results.

Differential central projections of vestibular afferents in pigeons

The question of whether a differential distribution of vestibular afferent information to central nuclear neurons is present in pigeons was studied using neural tracer compounds. Discrete tracing of afferent fibers innervating the individual semicircular canal and otolith organs was produced by sectioning individual branches of the vestibular nerve that innervate the different receptor organs and applying crystals of horseradish peroxidase, or a horseradish peroxidase/cholera toxin mixture, or a biocytin compound for neuronal uptake and transport. Afferent fibers and their terminal distributions within the brainstem and cerebellum were visualized subsequently. Discrete areas in the pigeon central nervous system that receive primary vestibular input include the superior, dorsal lateral, ventral lateral, medial, descending, and tangential vestibular nuclei; the A and B groups; the intermediate, medial, and lateral cerebellar nuclei; and the nodulus, the uvula, and the paraflocculus. Generally, the vertical canal afferents projected heavily to medial regions in the superior and descending vestibular nuclei as well as the A group. Vertical canal projections to the medial and lateral vestibular nuclei were observed but were less prominent. Horizontal canal projections to the superior and descending vestibular nuclei were much more centrally located than those of the vertical canals. A more substantial projection to the medial and lateral vestibular nuclei was seen with horizontal canal afferents compared to vertical canal fibers. Afferents innervating the utricle and saccule terminated generally in the lateral regions of all vestibular nuclei in areas that were separate from the projections of the semicircular canals. In addition, utricular fibers projected to regions in the vestibular nuclei that overlapped with the horizontal semicircular canal terminal fields, whereas saccular afferents projected to regions that received vertical canal fiber terminations. Lagenar afferents projected throughout the cochlear nuclei, to the dorsolateral regions of the cerebellar nuclei, and to lateral regions of the superior, lateral, medial, and descending vestibular nuclei.

Visual and somatosensory inputs to the avian song system via nucleus uvaeformis (Uva) and a comparison with the projections of a similar thalamic nucleus in a nonsongbird, Columba livia

Nucleus uvaeformis (Uva), previously identified as a component of song control circuitry in songbirds, and nucleus dorsolateralis posterior thalami, pars caudalis (DLPc) in pigeon, were compared with respect to their relative positions in the dorsolateral part of the posterior thalamus, their cell types, and their afferent and efferent projections. Both nuclei are closely related to the habenulointerpeduncular tract, have similar cell types, and receive a dense projection from deep layers of the optic tectum, predominantly ipsilaterally, and a distinct projection from the dorsal column and external cuneate nuclei, predominantly contralaterally. Recordings of multiple unit activity evoked by visual and somatosensory stimuli were used to guide injections of tracer into either DLPc or Uva, and the projections to the telencephalon were charted. Both nuclei were found to have a major terminal field in the medial part of the ipsilateral neostriatum intermedium (NI), known as nucleus interfacialis (NIf) in songbirds, and a minor terminal field in the roof of the neostriatum caudale (NC). In pigeon, the DLPc terminations in NC were within a region known as neostriatum dorsale (Nd), and, in male songbirds, the Uva terminations were in the high vocal center (HVC). Recordings of visual and somatosensory evoked activity were then used to guide injections of tracer into NI, and the afferent and efferent projections were again compared in pigeon and songbirds. The projections from either DLPc or Uva were confirmed, and terminal fields were observed either in Nd in pigeon, the dorsolateral part of NC in female songbirds, or HVC in male songbirds. Injections of tracer into either Nd or HVC confirmed their sources of afferents in DLPc or Uva, respectively, and in NI, but there was incomplete overlap of the distribution of retrogradely labelled cells in NI and the terminal fields of DLPc or Uva. It is concluded that DLPc and Uva are comparable nuclei having similar afferent and efferent projections relaying visual and somatosensory information to the telencephalon. The possible role of this information in vocal control is discussed.

Connections of the basal telencephalic areas c and d in the turtle brain

Tracer substances were injected into the basal telencephalic areas c and d of the turtle brain. These areas (Acd) have recently been shown to be connected reciprocally with the dorsal spino-medullary region, though the particular subregions involved in these projections remained unclear. We demonstrated that the efferent projections of area d terminate predominantly within or immediately adjacent to the trigeminal nuclear complex and in the high cervical spinal gray. The dendritic domain of the vagus-solitarius complex and the dorsal column nuclear complex might also receive some basal telencephalic efferents. The afferent projections to Acd, on the other hand, arise predominantly in the dorsal column nuclei as defined according to cytoarchitectural and hodological criteria. A few retrogradely labeled cells were found in the vagus-solitarius complex, the principal trigeminal nucleus and the high cervical spinal cord. Numerous labeled cells were found in the dorsolateral isthmo-rhombencephalic tegmentum, especially the n. visceralis secundarius, the n. vestibularis superior and parts of the lateral lemniscal complex. Aminergic cell populations projecting to Acd were the n. raphes inferior and superior, the locus coeruleus, the substantia nigra, pars compacta and the ventral tegmental area. Other meso-diencephalic cell groups were the griseum centrale (including the n. laminaris of the torus semicircularis), the n. interpeduncularis dorsalis, the nucleus of the fasciculus longitudinalis medialis, the nucleus and the nucleus interstitialis of flm, the n. interstitialis commissuralis posterior and then n. caudalis. Several hypothalamic regions, the reuniens complex and the perirotundal region of the thalamus also appeared to project heavily to Acd. Telencephalic areas retrogradely labeled after injection of tracer into Acd and its immediate surroundings were the rostral part of the lateral (olfactory) cortex, adjacent regions of the basal dorsal ventricular ridge and the n. centralis amygdalae, the n. tractus olfactorius lateralis as well as the areas g and h. The data suggest that areas c and d may correlate best with the 'extended' amygdala in mammals; further correlation with structures similar to the ventral striopallidum, however, cannot be excluded. Homostrategies are discussed with regard to the processing of higher-order somatovisceral information in turtles, birds and mammals.

The evolution of the dorsal thalamus of jawed vertebrates, including mammals: cladistic analysis and a new hypothesis

The evolution of the dorsal thalamus in various vertebrate lineages of jawed vertebrates has been an enigma, partly due to two prevalent misconceptions: the belief that the multitude of nuclei in the dorsal thalamus of mammals could be meaningfully compared neither with the relatively few nuclei in the dorsal thalamus of anamniotes nor with the intermediate number of dorsal thalamic nuclei of other amniotes and a definition of the dorsal thalamus that too narrowly focused on the features of the dorsal thalamus of mammals. The cladistic analysis carried out here allows us to recognize which features are plesiomorphic and which apomorphic for the dorsal thalamus of jawed vertebrates and to then reconstruct the major changes that have occurred in the dorsal thalamus over evolution. Embryological data examined in the context of Von Baerian theory (embryos of later-descendant species resemble the embryos of earlier-descendant species to the point of their divergence) supports a new 'Dual Elaboration Hypothesis' of dorsal thalamic evolution generated from this cladistic analysis. From the morphotype for an early stage in the embryological development of the dorsal thalamus of jawed vertebrates, the divergent, sequential stages of the development of the dorsal thalamus are derived for each major radiation and compared. The new hypothesis holds that the dorsal thalamus comprises two basic divisions--the collothalamus and the lemnothalamus--that receive their predominant input from the midbrain roof and (plesiomorphically) from lemniscal pathways, including the optic tract, respectively. Where present, the collothalamic, midbrain-sensory relay nuclei are homologous to each other in all vertebrate radiations as discrete nuclei. Within the lemnothalamus, the dorsal lateral geniculate nucleus of mammals and the dorsal lateral optic nucleus of non-synapsid amniotes (diapsid reptiles, birds and turtles) are homologous as discrete nuclei; most or all of the ventral nuclear group of mammals is homologous as a field to the lemniscal somatosensory relay and motor feedback nuclei of non-synapsid amniotes; the anterior, intralaminar and medial nuclear groups of mammals are collectively homologous as a field to both the dorsomedial and dorsolateral (including perirotundal) nuclei of non-synapsid amniotes; the anterior, intralaminar, medial and ventral nuclear groups and the dorsal lateral geniculate nucleus of mammals are collectively homologous as a field to the nucleus anterior of anamniotes, as are their homologues in non-synapsid amniotes. In the captorhinomorph ancestors of extant land vertebrates, both divisions of the dorsal thalamus were elaborated to some extent due to an increase in proliferation and lateral migration of neurons during development.(ABSTRACT TRUNCATED AT 400 WORDS)

The evolution of the dorsal pallium in the telencephalon of amniotes: cladistic analysis and a new hypothesis

The large body of evidence that supports the hypothesis that the dorsal cortex and dorsal ventricular ridge of non-mammalian (non-synapsid) amniotes form the dorsal pallium and are homologous as a set of specified populations of cells to respective sets of cells in mammalian isocortex is reviewed. Several recently taken positions that oppose this hypothesis are examined and found to lack a solid foundation. A cladistic analysis of multiple features of the dorsal pallium in amniotes was carried out in order to obtain a morphotype for the common ancestral stock of all living amniotes, i.e., a captorhinomorph amniote. A previous cladistic analysis of the dorsal thalamus (Butler, A.B., The evolution of the dorsal thalamus of jawed vertebrates, including mammals: cladistic analysis and a new hypothesis, Brain Res. Rev., 19 (1994) 29-65; this issue, previous article) found that two fundamental divisions of the dorsal thalamus can be recognized--termed the lemnothalamus in reference to predominant lemniscal sensory input and the collothalamus in reference to predominant input from the midbrain roof. These two divisions are both elaborated in amniotes in that their volume is increased and their nuclei are laterally migrated in comparison with anamniotes. The present cladistic analysis found that two corresponding, fundamental divisions of the dorsal pallium were present in captorhinomorph amniotes and were expanded relative to their condition in anamniotes. Both the lemnothalamic medial pallial division and the collothalamic lateral pallial division were subsequently further markedly expanded in the synapsid line leading to mammals, along with correlated expansions of the lemnothalamus and collothalamus. Only the collothalamic lateral pallial division--along with the collothalamus--was subsequently further markedly expanded in the non-synapsid amniote line that gave rise to diapsid reptiles, birds and turtles. In the synapsid line leading to mammals, an increase in the degree of radial organization of both divisions of the dorsal pallium also occurred, resulting in an 'outside-in' migration pattern during development. The lemnothalamic medial division of the dorsal pallium has two parts. The medial part forms the subicular, cingulate, prefrontal, sensorimotor, and related cortices in mammals and the medial part of the dorsal cortex in non-synapsid amniotes. The lateral part forms striate cortex in mammals and the lateral part of dorsal cortex (or pallial thickening or visual Wulst) in non-synapsid amniotes. Specific fields within the collothalamic lateral division of the dorsal pallium form the extrastriate, auditory, secondary somatosensory, and related cortices in mammals and the visual, auditory, somatosensory, and related areas of the dorsal ventricular ridge in non-synapsid amniotes.

Connections of the auditory forebrain in the pigeon (Columba livia)

Ascending auditory efferents in birds terminate mainly within Field L2, a cytoarchitectonically distinct region of the caudomedial telencephalon. The organization of Field L2, and that of its flanking regions, L1 and L3, was investigated with 14C-2-deoxyglucose (14C-2-DG), cytochrome oxidase, and both retrograde and anterograde tracing techniques. Field L2 was found to contain a high concentration of cytochrome oxidase. Following auditory stimulation, 14C-2-DG autoradiography revealed that Field L2 consists of two adjacent but seemingly discontinuous zones, designated Field L2a, which lies ventromedially, and Field L2b, which lies dorsolaterally. Termination of thalamic efferents: The thalamic auditory nuclei ovoidalis (Ov) and semilunaris parovoidalis (SPO) project predominantly upon Field L2, and possibly sparsely upon L1, L3 and the overlying hyperstriatum ventrale (HV). Ov subnuclei project upon L2a and SPO projects predominantly upon L2b. The topography of the projections is inverted along the ventromedial-to-dorsolateral axis of L2, and is in accord with an inverted tonotopic representation of frequencies; high frequencies (< 3.5 kHz) being found in the more ventromedial parts of L2a, and low frequencies and broad band responses in L2b. Intra- and extratelencephalic connections: Field L2a also receives a substantial projection from HV, but the efferent projections of L2a appear confined to adjacent "neostriatal" regions. The subsequent projections of L2b were not identified in this study. L1 and L3 project predominantly to the dorsal neostriatum (Nd) caudolateral to Field L, and have fewer projections to the caudomedial paleostriatum and anterior hyperstriatum accessorium. Nd projects massively upon the ventromedial nucleus of the intermediate archistriatum (Aivm), which has bilateral projections upon the caudomedial telencephalon and is the origin of a major descending pathway having dense terminations surrounding the ovoidalis complex (Ov and SPO), MLd, the lateral lemniscal nuclei, and sparse terminations within SPO itself. It is suggested that within the telencephalon the major components of the auditory pathway consist of cell groups which collectively correspond to the populations of neurons found within the auditory cortex of mammals.

Direct and indirect "cortico"-rubral and rubro-cerebellar cortical projections in the pigeon

In birds the red nucleus is the most rostral cell group in the brain having projections to all levels of the spinal cord (Cabot et al., Prog. Brain Res., 57:79-108, 1982), but its sources of afferents are incompletely known. In order to determine these, a series retrograde and anterograde tracing experiments was carried out, largely with cholera toxin B-chain conjugated to horseradish peroxidase. The results show that a sparse and diffuse projection to the red nucleus arises from deep regions of the hyperstriatum accessorium (HA) of the anterior Wulst, and that a much more dense projection arises from the caudal part of the nucleus principalis precommissuralis and the medial part of the medial spiriform nucleus (SpMm). These last two sources were themselves shown to receive a substantial projection from HA of the anterior Wulst. The red nucleus was also shown to project upon the cerebellar cortex of lobule VI, and SpM upon the cerebellar cortex of lobules VI through IX (Karten and Finger, Brain Res., 102:335-338, 1976; Clarke, J. Comp. Neurol., 174:535-552, 1977). Double retrograde labelling experiments with fluorescein and rhodamine labelled latex microspheres injected into the cerebellar cortex and spinal cord showed that the rubrocerebellar cortical neurons are a different population from, although intermixed with, the rubrospinal neurons.

Organization of the cerebellum in the pigeon (Columba livia): II. Projections of the cerebellar nuclei

The projections of the deep cerebellar nuclei in the pigeon have been delineated using autoradiographic and histochemical (WGA-HRP) tracing techniques. A medial (CbM) and lateral (CbL) cerebellar nucleus are recognized and CbM may be further partitioned into internal, intermediate, and intercalate divisions. As in mammals, most extracerebellar projections of CbM travel in the fasciculus uncinatus (FU); the rest travel with those of CbL in the brachium conjunctivum (BC). In the pigeon, both of these pathways are bilaterally but primarily contralaterally projecting systems. FU is a predominantly descending tract, with terminations within (1) the vestibular complex, (2) a column of contiguous medial reticular nuclei from pontine to caudal medullary levels; (3) the plexus of Horsley portion of the parvicellular reticular formation, continuing through the nucleus centralis medullae oblongatae, pars dorsalis, into intermediate layer VII of the cervical spinal cord, down to cervical segment 8-9; (4) the lateral reticular nucleus and the paragigantocellular reticular nucleus; (5) the dorsal lamella of the inferior olive. Rostrally FU terminals are found in the locus ceruleus and dorsal subcerulean nucleus. Minimal FU projections are also seen to the motor trigeminal nucleus and the subnucleus oralis of the descending trigeminal system. A small projection from the intercalate division of CbM travels in BC and projects upon the midbrain central grey, the intercollicular nucleus, the lateral tectal periventricular grey, the stratum cellulare externum and, sparsely, upon the dorsolateral thalamus. The bulk of BC originates from the lateral cerebellar nucleus and consists of a massive ascending and a small descending branch. The ascending system projects upon the red nucleus and the dorsally adjacent interstitial nucleus of Cajal and midbrain central grey, the prerubral fields continuing into the stratum cellulare externum, the nucleus intercalatus thalami, the ventrolateral thalamic nucleus, the medial spiriform nucleus, the nucleus principalis precommissuralis, the nucleus of the basal optic root, the nucleus geniculatus lateralis pars ventralis, the dorsolateral thalamus, including the dorsal intermediate posterior, and the dorsolateral intermediate and anterior nuclei. BC also contains axons from the infracerebellar nucleus, which projects upon the trochlear and the oculomotor nuclei. The descending branch of BC distributes to the papilioform nucleus, the medial pontine nucleus, the gigantocellular and paramedian reticular nuclei, and, minimally, the rostral portions of the medial column and ventral lamella of the inferior olive. Taken in conjunction with data on amphibia and reptiles the present findings suggest that the fundamental ground plan of vertebrate cerebellar organization involves a medial and lateral cerebellar nucleus.(ABSTRACT TRUNCATED AT 400 WORDS)

Organization of the cerebellum in the pigeon (Columba livia): III. Corticovestibular connections with eye and neck premotor areas

The connections of the cerebellar cortex with vestibular premotor neurons of the oculomotor and collimotor systems in the pigeon were delineated in experiments using WGA-HRP as an anterograde and retrograde tracer. Putative premotor neuron pools were identified by injections into the oculomotor (mIII) and trochlear nuclei (mIV) and into the most rostral portion of the cervical neck motor nucleus, nucleus supraspinalis (SSp). The retrograde data indicate that ipsilateral projections upon oculomotor neurons arise from the medial portions of the superior (VeS) and tangential (Ta) nuclei. Contralateral projections originate from the infracerebellar nucleus, the interstitial vestibular region including the main (lateral) portion of the tangential nucleus, and from the descending and medial vestibular nuclei (VeD, VeM). These projections were confirmed in anterograde studies that also defined the connections of these vestibular premotor regions with specific subnuclear divisions of the pigeon's "oculomotor" nuclei (mIII, mIV, mVI). The organization of projections from the vestibular nuclei to the pigeon's extraocular motoneurons is similar to that reported in mammals. Projections upon neck premotor neurons arise primarily from neurons in the interstitial region of the vestibular nuclear complex. After injections in SSp, retrogradely labeled neurons were found, contralaterally, in the lateral part of the tangential and superior vestibular nuclei and in the dorsolateral vestibular nucleus (VDL). Ipsilateral labeling was seen in the medial interstitial region (VeM, VeD, and medial Ta). These projections were confirmed in anterograde experiments. With the exception of VDL, vestibular nuclei projecting to neck motoneurons also project to extraocular motoneurons. Thus the infracerebellar nucleus projects exclusively, and the superior vestibular nucleus predominantly, upon oculomotor (mIII, mIV) nuclei; VDL projects predominantly upon the neck motor nucleus, whereas the interstitial vestibular regions (medial Ta, rostral VeD, intermediate VeM) project upon both collimotor and oculomotor neurons. The pattern of retrograde labeling seen in the cerebellar cortex after injections into vestibular premotor nuclei was used to define the projections of specific cerebellar cortical zones upon vestibular eye and neck premotor neurons. Corticovestibular projections upon these regions arise from the auricle and lateral unfoliated cortex, the posterior lobe components of cortical zones B and E, and from the vestibulocerebellum. Each of these cortical zones projects upon components of the vestibular nuclear complex, which are premotor to either oculomotor nuclei or collimotor nuclei. The hodological findings are related to the functional organization of the oculomotor and collimotor systems in the pigeon and compared with the mammalian data.

Vestibulo-thalamic neurons give off descending axons to the spinal cord

Vestibulo-thalamic (VT) neurons were physiologically studied in the anesthetized cat. Forty-seven VT neurons were recorded extracellularly. More than half of the VT neurons responded monosynaptically to vestibular nerve stimulation while the others responded polysynaptically. They were activated antidromically from one or two sites in the VPL. VPM, VL, VM, SG, and PO of the contralateral thalamus. Four fifths of the VT neurons were activated from the C1 segment of the spinal cord. Half of them were also activated from the C4 segment, but none were activated from the L5 segment. It is suggested that most VT neurons project descending axons to the cervical spinal cord. Axonal branching was shown by means of systematic microstimulation in the thalamus and the ventral horn in the C1 segment. The VT neurons were mainly located in the descending vestibular nucleus.

Vestibulo-ocular projections in the 11-day chicken embryo: pathway specificity

The organization of vestibulo-ocular projections and their spatial relationship to vestibulo-spinal projections were ascertained in the 11-day chicken embryo through retrograde tracing experiments. An in vitro preparation of the brainstem facilitated precisely localized application of tracers. Unilateral labelling of the ascending medial longitudinal fascicle (MLF), just caudal to the trochlear nucleus, labelled coherent groups of vestibular neurons that had an asymmetrical distribution on the two sides of the brainstem. Axons originating from either side followed one of four trajectories, and the groups of neurons could therefore be identified on the basis of their position and their projection pathway. Labelling each side of the MLF with a different tracer showed that although neurons projecting in the ipsi- and contralateral MLF were intermingled in some areas, individual neurons projected on one side only. Labelling the ascending MLF and the high cervical spinal cord with different tracers showed that vestibulo-ocular neurons had a wider rostro-caudal distribution than vestibulospinal neurons. Some vestibulospinal and vestibulo-ocular groups were spatially segregated, others were intermingled. Very few (in most preparations no) neurons were found to project to both targets. Together with our previous study on vestibulospinal projections in the 11-day chicken embryo (Glover and Pétursdóttir, J. Comp. Neurol. 270:25-38, '88) the results show that vestibular neurons in different regions project in characteristic subsets of the available pathways.

Projections of the dorsolateral anterior complex and adjacent thalamic nuclei upon the visual Wulst in the pigeon

The distinctive patterns of thalamic input to the rostral (R), intermediate (I) and caudal (C) divisions of the pigeon Wulst were determined using the retrograde multiple label technique following concomitant injections of various fluorescent tracers into different Wulst loci. The results showed topographical projections from components of the visual thalamic n. dorsolateralis anterior complex. Major ipsilateral and contralateral projections upon the R and I Wulst stem from the pars lateralis ventralis and dorsalis nuclei, respectively. The nuclei pars lateralis rostralis, pars magnocellularis and n. suprarotundus provide weak bilateral projections to all of the Wulst divisions sampled. Lastly, bilateral connections from the n. superficialis parvocellularis upon I and C, and from n. dorsolateralis posterior upon the posterior C Wulst were also demonstrated. Based upon their patterns of terminal distribution upon the Wulst, some of these thalamic nuclei are compared to specific components of the mammalian geniculate and extrageniculate visual systems.

Sensory properties and afferents of the N. dorsolateralis posterior thalami of the pigeon

According to previous studies, the avian n. dorsolateralis posterior thalami (DLP) receives visual and somatosensory afferents. While some authors (e.g., Gamlin and Cohen: J. Comp. Neurol. 250:296-310, '86) proposed a distinction between a visual caudal (DLPc) and a somatosensory rostral (DLPr) part, other authors (e.g., Wild: Brain Res. 412:205-223, '87) could not confirm such a differentiation. The aim of the present experiment was to study with physiological and anatomical methods the proposed parcellation of the DLP into various components dealing with different modalities. The physiological properties of the DLP of the pigeon were analysed with extracellular single unit recordings. With the same approach, neurons of the n. dorsalis intermedius ventralis anterior (DIVA), a somatosensory relay nucleus in the dorsal thalamus, were also analysed. The afferents of the DLP were studied by using anatomical tract tracing techniques with retrograde and anterograde tracers. The sensory properties of DLP cells revealed that somatosensory, visual, and auditory modalities affect the neuronal firing frequency in this nucleus. All three modalities were present throughout the full caudorostral extent of the DLP. Cells recorded in DIVA responded nearly exclusively to somatosensory stimulation. Unlike the DLP, single units in DIVA generally had smaller receptive fields encompassing only one extremity. The analysis of afferent connections of the DLP by using injections of retrograde and anterograde tracers (HRP, WGA-HRP, Fast Blue, and Rhodamine-beta-isothiocyanate) demonstrated extensive projections from the nuclei gracilis et cuneatus (GC) and more sparse projections from the nucleus tractus descendens trigemini (TTD), and the nucleus cuneatus externus (CE). Brainstem afferents of the DLP came from different vestibular nuclei, various areas of the brainstem reticular formation, and the optic tectum. Prosencephalic afferents originated in the n. posteroventralis thalami (PV), the n. ventromedialis posterior thalami (VMP), the n. dorsalis intermedius ventralis anterior (DIVA), and the nucleus reticularis superior pars dorsalis and ventralis (RSd and RSv). Telencephalic afferents of the DLP came from the hyperstriatum accessorium (HA) and a group of cells at the borderline between the hyperstriatum intercalatus superior (HIS) and the hyperstriatum dorsale (HD). The somatosensory afferents of the DLP probably originate from the GC, TTD, and CE, whereas it is likely that the visual input is mediated by the optic tectum. The anatomical source for the acoustic input is unclear. The very long latencies of auditory DLP neurons make it likely that the acoustic input originates at least partly in the reticular formation.(ABSTRACT TRUNCATED AT 400 WORDS)

Autoradiographic localization of progestin-concentrating cells in the brain of the zebra finch

The production of song in passerine birds is under the control of steroid hormones, and brain regions involved in song production have been shown to contain androgen and/or estrogen receptors. Studies to date, however, have not considered the possible role of progestins in this behavior. As one approach to this question, the autoradiographic method was used to investigate the distribution of progestin-concentrating cells in the brain of the adult male zebra finch (Poephila guttata) after injection of the radiolabeled synthetic progestin [17 alpha-methyl-3H]-promegestone. In the telencephalon, identifiable groups of progestin-accumulating cells were found in the hyperstriatum dorsale, at the medial edge of the lobus parolfactorius, and in the medial septum. In the diencephalon, labeled groups of cells were found in the preoptic area, through much of the medial hypothalamus--including nucleus periventricularis magnocellularis, nucleus medialis hypothalami posterialis, and area infundibularis--and in the medial spiriform nucleus and dorsomedial thalamus. In the myelencephalon, labeled cells are described at the dorsal edge of the medulla and scattered lateral to nXII. These findings offer no support for the hypothesis that progestin acts on any of the known song regions, but do suggest areas of progestin action in the avian central nervous system outside of the known song system. Not surprisingly, these include many areas of the medial hypothalamus and other midline structures.

Extracellular recording of vestibulo-thalamic neurons projecting to the spinal cord in the cat

Forty vestibulo-thalamic (VT) neurons were recorded extracellularly in the vestibular nuclei of the anesthetized cat. More than half of the VT neurons responded monosynaptically to vestibular nerve stimulation; the others responded polysynaptically. The VT neurons were activated antidromically from one or two sites in the contralateral VPL, VPM, VL, VM, SG, and PO in the thalamus. Their axonal arborizations in the thalamus were likely restricted in narrow areas. About three quarters of the VT neurons were also activated antidromically from the ventral funiculus in the C1 segment. Axonal branchings were found in the contralateral C1 gray matter. The VT neurons were mainly localized in the descending vestibular nucleus.

Avian somatosensory system: II. Ascending projections of the dorsal column and external cuneate nuclei in the pigeon

The ascending projections of the dorsal column and external cuneate nuclei (DCN/CuE) in the pigeon were investigated in anterograde tracing experiments by using autoradiography or wheat germ agglutinin conjugated to horseradish peroxidase (WGA-HRP). The results show that the majority of ascending projections decussate via internal arcuate fibers to form a contralateral medial lemniscus which ascends in a ventral position. In the brainstem, terminal fields were observed in the ventral lamella of the inferior olive (OI), the parabrachial nuclei (PB) of the dorsolateral pons, the intercollicular nucleus (ICo) of the midbrain, and the nucleus pretectalis diffusus (PD). In the diencephalon there were terminal fields in the strata cellulare externum and internum (SCE and SCI) of the caudal hypothalamus; in the intercalated (ICT), ventrolateral (VLT), and reticular nuclei of the ventral thalamus; in the nuclei principalis precommissuralis (PPC), spiriform medialis (SpM), and dorsolateralis posterior, pars caudalis (cDLP) of the caudal thalamus; and in the nuclei dorsalis intermedius ventralis anterior (DIVA), dorsolateralis posterior, pars rostralis (rDLP), dorsolateralis anterior (DLA), and dorsolateralis anterior, pars medialis (DLM) of the rostrodorsal thalamus. The origins of these projections within the DCN/CuE complex were verified in retrograde tracing experiments with WGA-HRP and were found to be partly differentiable with respect to their targets. The projections to DIVA, rDLP, DLA, DLM, cDLP, and SpM arise from all rostrocaudal levels of the DCN/CuE complex; those to ICo arise from caudomedial nuclear regions, while those to the hypothalamus and ventral thalamus arise from rostrolateral nuclear regions. Projections to PB arise from lamina I neurons of the dorsal horn of upper cervical spinal cord segments and from CuE. No evidence was found of a projection to the cerebellum. The distribution of the cells of origin of the medial lemniscus (ML) within the DCN/CuE complex was found to be largely coextensive with the areas of termination of primary spinal (Wild: J. Comp. Neurol. 240:377-395, '85) and some trigeminal (Dubbledam and Karten: J. Comp. Neurol. 180:661-678, '78) afferents. Furthermore, the areas of termination of the ML within the rostrodorsal and caudal thalamus are also either coextensive or closely associated with nuclei which provide a somatosensory projection to separate regions of the telencephalon (Wild: Brain Res. 412:205-223, '87). There are thus clear similarities in the overall pattern of somatosensory projections in the pigeon and in many mammalian species.