Radial glial elements in the cerebral cortex of the lesser hedgehog tenrec.

We investigated astroglial cells in several areas of the telencephalic cortex of the lesser hedgehog tenrec (Echinops telfairi). Compared to other mammals, the cortex of the tenrec has a relatively large paleocortex and a low encephalization index. We stained sections from tenrec forebrains with structural and functional glia markers focusing on selected cortical areas, the paleocortex, rhinal cortex, neocortex and the dentate gyrus of the hippocampal formation. We found that in all parts of the tenrec forebrain cortex, radial processes exist which are positive for glial fibrillary acidic protein (GFAP) although with differential localization: in the rhinal cortex and neocortical region radial glial fibers are located in the subventricular regions, whereas in the dentate gyrus and paleocortex they appear to arise from the cells in the respective granular layers. The relatively high abundance of the radial fibers in layer III of the paleocortex was very conspicuous. Only few of these radial processes were also co-labeled with doublecortin (DCX), yet most of the DCX-positive cells were negative for GFAP. The GFAP-positive radial fibers were in turn neither positive for glutamine synthetase, nor did they show immunoreactivity for the astroglia-specific water channel aquaporin-4 (AQP4). Star-shaped astrocytes, however, displayed the typical perivascular and subpial expression patterns for AQP4. We conclude that the radial glia in the adult tenrec represents an immature form of astroglia that persists in these animals throughout life.

Amygdalar connections in the lesser hedgehog tenrec.

The present study analyses the overall extrinsic connectivity of the non-olfactory amygdala (Ay) in the lesser hedgehog tenrec. The data were obtained from tracer injections into the lateral and intermediate portions of the Ay as well as several non-amygdalar brain regions. Both the solitary and the parabrachial nucleus receive descending projections from the central nucleus of the Ay, but only the parabrachial nucleus appears to project to the Ay. There is one prominent region in the ventromedial hypothalamus connected reciprocally with the medial and central Ay. Amygdalar afferents clearly arise from the dorsomedial thalamus, the subparafascicular nuclei and the medial geniculate complex (GM). Similar to other subprimate species, the latter projections originate in the dorsal and most caudal geniculate portions and terminate in the dorsolateral Ay. Unusual is the presence of amygdalo-projecting cells in the marginal geniculate zone and their virtual absence in the medial GM. As in other species, amygdalo-striatal projections mainly originate in the basolateral Ay and terminate predominantly in the ventral striatum. Given the poor differentiation of the tenrec's neocortex, there is a remarkable similarity with regard to the amygdalo-cortical connectivity between tenrec and rat, particularly as to prefrontal, limbic and somatosensorimotor areas as well as the rhinal cortex throughout its length. The tenrec's isocortex dorsomedial to the caudal rhinal cortex, on the other hand, may not be connected with the Ay. An absence of such connections is expected for primary auditory and visual fields, but it is unusual for their secondary fields.

Slow age-dependent decline of doublecortin expression and BrdU labeling in the forebrain from lesser hedgehog tenrecs.

In addition to synaptic remodeling, formation of new neurons is increasingly acknowledged as an important cue for plastic changes in the central nervous system. Whereas all vertebrates retain a moderate neuroproliferative capacity, phylogenetically younger mammals become dramatically impaired in this potential during aging. The present study shows that the lesser hedgehog tenrec, an insectivore with a low encephalization index, preserves its neurogenic potential surprisingly well during aging. This was shown by quantitative analysis of 5-bromo-2'-deoxyuridine (BrdU) immunolabeling in the olfactory bulb, paleo-, archi-, and neocortices from 2- to 7-year-old animals. In addition to these newly born cells, a large number of previously formed immature neurons are present throughout adulthood as shown by doublecortin (DCX) immunostaining in various forebrain regions including archicortex, paleocortex, nucleus accumbens, and amygdala. Several ventricle-associated cells in olfactory bulb and hippocampus were double-labeled by BrdU and DCX immunoreactivity. However, most DCX cells in the paleocortex can be considered as persisting immature neurons that obviously do not enter a differentiation program since double fluorescence labeling does not reveal their co-occurrence with numerous neuronal markers, whereas only a small portion coexpresses the pan-neuronal marker HuC/D. Finally, the present study reveals tenrecs as suitable laboratory animals to study age-dependent brain alterations (e.g., of neurogenesis) or slow degenerative processes, particularly due to the at least doubled longevity of tenrecs in comparison to mice and rats.

Transposition and Intermingling of Galphai2 and Galphao afferences into single vomeronasal glomeruli in the Madagascan lesser Tenrec Echinops telfairi.

The vomeronasal system (VNS) mediates pheromonal communication in mammals. From the vomeronasal organ, two populations of sensory neurons, expressing either Galphai2 or Galphao proteins, send projections that end in glomeruli distributed either at the rostral or caudal half of the accessory olfactory bulb (AOB), respectively. Neurons at the AOB contact glomeruli of a single subpopulation. The dichotomic segregation of AOB glomeruli has been described in opossums, rodents and rabbits, while Primates and Laurasiatheres present the Galphai2-pathway only, or none at all (such as apes, some bats and aquatic species). We studied the AOB of the Madagascan lesser tenrec Echinops telfairi (Afrotheria: Afrosoricida) and found that Galphai2 and Galphao proteins are expressed in rostral and caudal glomeruli, respectively. However, the segregation of vomeronasal glomeruli at the AOB is not exclusive, as both pathways contained some glomeruli transposed into the adjoining subdomain. Moreover, some glomeruli seem to contain intermingled afferences from both pathways. Both the transposition and heterogeneity of vomeronasal afferences are features, to our knowledge, never reported before. The organization of AOB glomeruli suggests that synaptic integration might occur at the glomerular layer. Whether intrinsic AOB neurons may make synaptic contact with axon terminals of both subpopulations is an interesting possibility that would expand our understanding about the integration of vomeronasal pathways.

Tracing thalamo-cortical connections in tenrecA further attempt to characterize poorly differentiated neocortical regions, particularly the motor cortex.

The hedgehog tenrec (Afrosoricidae) has a very poorly differentiated neocortex. Previously its primary sensory regions have been characterized with hodological and electrophysiological techniques. Unlike the marsupial opossum the tenrec may also have a separate motor area as far as there are cortico-spinal cells located rostral to the primary somatosensory cortex. However, not knowing its thalamic input it may be premature to correlate this area with the true (mirror-image-like) primary motor cortex in higher mammals. For this reason the tenrec's thalamo-cortical connections were studied following tracer injections into various neocortical regions. The main sensory areas were confirmed by their afferents from the principal thalamic nuclei. The dorsal lateral geniculate nucleus, in addition, was connected with the retrosplenial area and a rostromedial visual region. Unlike the somatosensory cortex the presumed motor area did not receive afferents from the ventrobasal thalamus but fibers from the cerebello-thalamic target regions. These projections, however, were not restricted to the motor area, but involved the entire somatosensorimotor field as well as adjacent regions. The projections appeared similar to those arising in the rat thalamic ventromedial nucleus known to have a supporting function rather than a specific motor task. The question was raised whether the input from the basal ganglia might play a crucial role in the evolution of the mammalian motor cortex? Certainly, in the tenrec, the poor differentiation of the motor cortex coincides with the virtual absence of an entopeduncular projection to the ventrolateral thalamus.

The presence and absence of prosencephalic cell groups relaying striatal information to the medial and lateral thalamus in tenrec.

Although there are remarkable differences regarding the output organization of basal ganglia between mammals and non-mammals, mammalian species with poorly differentiated brain have scarcely been investigated in this respect. The aim of the present study was to identify the pallidal neurons giving rise to thalamic projections in the Madagascar lesser hedgehog tenrec (Afrotheria). Following tracer injections into the thalamus, retrogradely labelled neurons were found in the depth of the olfactory tubercle (particularly the hilus of the Callejal islands and the insula magna), in subdivisions of the diagonal band complex, the peripeduncular region and the thalamic reticular nucleus. No labelled cells were seen in the globus pallidus. Pallidal neurons were tentatively identified on the basis of their striatal afferents revealed hodologically using anterograde axonal tracer substances and immunohistochemically with antibodies against enkephalin and substance P. The data showed that the tenrec's medial thalamus received prominent projections from ventral pallidal cells as well as from a few neurons within and ventral to the cerebral peduncle. The only regions projecting to the lateral thalamus appeared to be the thalamic reticular nucleus (RTh) and the dorsal peripeduncular nucleus (PpD). On the basis of immunohistochemical data and the topography of its thalamic projections, the PpD was considered to be an equivalent to the pregeniculate nucleus in other mammals. There was no evidence of entopeduncular (internal pallidal) neurons being present within the RTh/PpD complex, neuropils of which did not stain for enkephalin and substance P. The ventrolateral portion of RTh, the only region eventually receiving a striatal input, projected to the caudolateral rather than the rostrolateral thalamus. Thus, the striatopallidal output organization in the tenrec appeared similar, in many respects, to the output organization in non-mammals. This paper considers the failure to identify entopeduncular neurons projecting to the rostrolateral thalamus in a mammal with a little differentiated cerebral cortex, and also stresses the discrepancy between this absence and the presence of a distinct external pallidal segment (globus pallidus).

Pyramidal neurons in the septal and temporal CA1 field of the human and hedgehog tenrec hippocampus.

The present study examines comparatively the cellular density of disector-counted/Nissl-stained CA1 pyramidal neurons and the morphometric characteristics (dendritic number/length, spine number/density and Sholl-counted dendritic branch points/20 microm) of the basal and apical dendritic systems of Golgi-impregnated CA1 neurons, in the septal and temporal hippocampus of the human and hedgehog tenrec brain. The obtained results indicate that in both hippocampal parts the cellular density of the CA1 pyramidal neurons is lower in human than in tenrec. However, while the human pyramidal cell density is higher in the septal hippocampal part than in the temporal one, in the tenrec the density of these cells is higher in the temporal part. The dendritic tree of the CA1 pyramidal cells, more developed in the septal than in temporal hippocampus in both species studied, is in general more complex in the human hippocampus. The basal and the apical dendritic systems exhibit species related morphometric differences, while dendrites of different orders exhibit differences in their number and length, and in their spine density. Finally, in both species, as well as hippocampal parts and dendritic systems, changes of dendritic morphometric features along ascending dendritic orders fluctuate in a similar way, as do the number of dendritic branch points in relation to the distance from the neuron soma.

Thalamo-striatal projections in the hedgehog tenrec.

Unlike the basal ganglia input from the midline and intralaminar nuclei, the origin and prominence of striatal projections arising in the lateral thalamus varies considerably among mammals being most restricted in the opossum and monkey, most extensive in the rat. To get further insight into the evolution of thalamo-striatal pathways the Madagascar lesser hedgehog tenrec (Afrotheria) was investigated using anterograde and retrograde flow techniques. An extensive medial thalamic region (including presumed equivalents to the paraventricular, parataenial and dorsomedial nuclei as well as the reuniens complex), the rostral (central) and caudal (parafascicular) intralaminar nuclei were shown to give rise to striatal projections. Additional projections originated in the ventral anterolateral nuclear group and regions within and around the medial geniculate complex. Similar to the rat there was also substantial projections from the lateral posterior-pulvinar complex and the ventral posterior nucleus. The fibers terminated extensively across the striatum in a mainly homogeneous fashion. Isolated patches of low-density terminations were found in the caudoputamen. This inhomogeneous labeling pattern appeared similar to one described in the cat with the unlabeled islands showing features of striosomes. The medial and intralaminar nuclei also projected heavily upon the olfactory tubercle. Differential innervation patterns were noted in the polymorphous layer, the deep and the superficial molecular layer.

Organization of the main olfactory bulb of lesser hedgehog tenrecs.

Using a confocal laser scanning microscope (CLSM) and an electron microscope, we investigated the organization of the main olfactory bulb (MOB) of tenrecs, which were previously included into insectivores but now considered to be in a new order "Afrosoricida" in the superclade 'Afrotheria'. We confirmed that the overall structural organization of the tenrec MOB was similar to that of rodents: (1) the compartmental organization of glomeruli and two types of periglomerular cells we proposed as the common organizational principles were present; (2) there were characteristic dendrodendritic and axo-dendritic synapses in the glomerulus and external plexiform layer (EPL) and gap junctions in glomeruli; and (3) no nidi, particular synaptic regions reported only in laboratory musk shrew and mole MOBs, were encountered. However, instead of nidi, we often observed a few tangled olfactory nerves (ONs) with large irregular boutons in the glomerular-external plexiform layer border zone, with which dendrites of various displaced periglomerular cells were usually found to be intermingled. Electron microscopic (EM) examinations confirmed characteristic large mossy terminal-like ON terminals making asymmetrical synapses to presumed mitral/tufted cell and displaced periglomerular cell dendrites. In addition, gap junctions were also encountered between dendritic processes in these tiny particular regions, further showing their resemblance to glomeruli.

An extrahippocampal projection from the dentate gyrus to the olfactory tubercle.

BACKGROUND: The dentate gyrus is well known for its mossy fiber projection to the hippocampal field 3 (CA3) and its extensive associational and commissural connections. The dentate gyrus, on the other hand, has only few projections to the CA1 and the subiculum, and none have clearly been shown to extrahippocampal target regions. RESULTS: Using anterograde and retrograde tracer techniques in the Madagascan lesser hedgehog tenrec (Afrosoricidae, Afrotheria) it was shown in this study that the dentate hilar region gave rise to a faint, but distinct, bilateral projection to the most rostromedial portion of the olfactory tubercle, particularly its molecular layer. Unlike the CA1 and the subiculum the dentate gyrus did not project to the accumbens nucleus. A control injection into the medial septum-diagonal band complex also retrogradely labeled cells in the dentate hilus, but these neurons were found immediately adjacent to the heavily labeled CA3, while the tracer injections into the rostromedial tubercle did not reveal any labeling in CA3. CONCLUSION: The dentate hilar neurons projecting to the olfactory tubercle cannot be considered displaced cells of CA3 but represent true dentato-tubercular projection neurons. This projection supplements the subiculo-tubercular projection. Both terminal fields overlap among one another as well as with the fiber terminations arising in the anteromedial frontal cortex. The rostromedial olfactory tubercle might represent a distinct ventral striatal target area worth investigating in studies of the parallel processing of cortico-limbic information in tenrec as well as in cat and monkey.

The striatum in the hedgehog tenrec: histochemical organization and cortical afferents.

In order to get insight into the striopallidal organization in mammals with little differentiated brain the striatum of the lesser hedgehog tenrec (Afrotheria) was characterized histochemically and analysed with regard to its cortical afferents using axonal tracer substances. The majority of neocortical cells projecting to the striatum were found bilaterally in the layers 2 and 3 of the frontal hemisphere; caudalwards the relative number of cells increased somewhat in the upper layer 5. There was a topographical organization as far as the allocortical projections appeared confined to the ventral striatum, and the efferents from hippocampal, posterior paleocortical, somatosensory and audiovisual areas were distributed in largely different striatal territories. Projections from the anterior frontal cortex, on the other hand, terminated extensively upon the caudate-putamen and also involved the nucleus accumbens and the olfactory tubercle. In the latter region the molecular layer was especially involved. The entorhinal cortex also projected heavily to the olfactory tubercle but unlike other species it scarcely involved the nucleus accumbens. The cortical fibers were distributed in a relatively homogenous fashion within their striatal territory and there was little evidence for patches of high density terminations. Islands of low density labeling, however, were noted occasionally in the caudate-putamen. These islands were partly similar in size as the patches of neuropil staining obtained with anti-calretinin and anti-substance P. There were also hints for the presence of a shell-like region in the nucleus accumbens stained with anti-dopamine transporter and NADPh-diaphorase. The classical striosome-matrix markers such as calbindin, acetylcholinesterase and enkephalin, however, failed to reveal any compartmental organization.

Cutaneous eccrine glands of the foot pads of the small Madagascan tenrec ( Echinops telfairi, Insectivora, Tenrecidae): skin glands in a primitive mammal.

In order to find correlations between skin gland morphology and specific ethological features, the cutaneous glands of the foot pads of the primitive mammal the Madagascan tenrec, Echinops telfairi, were studied by histological and various histochemical methods as well as by electron microscopy. In the foot pads specific eccrine skin glands occurred consisting of coiled ducts and tubular secretory portions, the lumina of which were considerably wider than in primate sweat glands. The secretory tubules were composed of branched myoepithelial cells and glandular cells. The latter contained abundant mitochondria, large amounts of glycogen particles and few secretory granules as well as individual heterolysosomes and myelin bodies. The lateral cell membrane was marked by extensive interdigitations. The apical membranes of all glandular cells contained proteoglycans with sulfated and carboxylated groups containing N-acetyl-glucosamine, N-acetyl-galactosamine, galactose and mannose. The expression pattern of cytokeratins of the glandular epithelium was variable and showed similarities to that of the human eccrine glands. Tubulin, vinculin and actin were expressed in the glandular epithelium. The secretory cells showed positive reactions with antibodies against antimicrobial peptides and IgA. A positive reaction was observed with antibodies against the androgen receptor. The PCNA and TUNEL reactions indicated that the tubular skin glands of Echinops are made up of a slowly renewing tissue. We conclude that the glands fulfill several functions: production of a fluid-rich secretory product, which may prevent slipping of the foot pads on the substrate during running or climbing, secretion of antimicrobial peptides and proteins, and playing a role in thermoregulation.

Zebrin II compartmentation of the cerebellum in a basal insectivore, the Madagascan hedgehog tenrec Echinops telfairi.

The mammalian cerebellum is histologically uniform. However, underlying the simple laminar architecture is a complex arrangement of parasagittal stripes and transverse zones that can be revealed by the expression of zebrin II/aldolase C. The cerebellar cortex of rodents, for example, is organized into four transverse zones: anterior, central, posterior and nodular. Within the anterior and posterior zones, parasagittal stripes of Purkinje cells expressing zebrin II alternate with those that do not. Zonal boundaries appear to be independent of cerebellar lobulation. To explore this model further, and to broaden our understanding of the evolution of cerebellar patterning, zebrin II expression has been studied in the cerebellum of the Madagascan hedgehog tenrec (Echinops telfairi), a basal insectivore with a lissiform cerebellum with only five lobules. Zebrin II expression in the tenrec reveals an array of four transverse zones as in rodents, two with homogeneous zebrin II expression, two further subdivided into stripes, that closely resembles the expression pattern described in other mammals. We conclude that a zone-and-stripe organization may be a common feature of the mammalian cerebellar vermis and hemispheres, and that zonal boundaries and cerebellar lobules and fissures form independently.

Distribution of perihippocampo-hippocampal projection neurons in the lesser hedgehog tenrec.

The entorhinal cortex in the Madagascan lesser hedgehog tenrec is thought to be part of the three-layered subrhinal paleocortex (PCx) but cyto- and chemoarchitectural studies have failed so far to identify the area. To reach this goal tracer injections were made into the tenrec's hippocampus. Retrogradely labeled cells were found in dorsal portion of the posterior PCx, the adjacent rhinal cortex (RCx) and the so-called area XCx. The main paleocortical portion in the ventral PCx, however, remained unlabeled with the exception of a caudal region possibly equivalent to the amygdalo-piriform transition area. The labeled neurons showed a bilaminar distribution with the cells in the layer 2A giving rise to fibers to predominantly the dentate area and the cells in the layer 3A mainly projecting to the cornu ammonis and the subiculum. The latter regions, in addition, gave rise to a feedback projection to the layer 3B of especially the caudal RCx and the XCx. The analysis of the terminal projections, however, was hampered by the fact, that under certain conditions retrogradely transported biotinylated dextran was also transported in anterograde direction via collaterals of the entorhino-dentate fibers. The findings are compared with equivalent regions in more differentiated mammals particularly with regard to the perirhinal area showing little if any connections with the dentate gyrus.

Parabrachio-cortical connections with the lateral hemisphere in the madagascan hedgehog tenrec: prominent projections to layer 1, weak projections from layer 6.

The present study was undertaken to further characterize and subdivide the rhinal cortex (insular and perirhinal areas) in the hedgehog tenrec (Echinops telfairi), a placental mammal with a rather low encephalisation index. Injections of wheat germ agglutinin-horseradish peroxidase into the dorsolateral pontine tegmentum revealed a prominent layer 1 projection to several rhinal target areas, while the rhinal cortex only stained weakly for the calcitonin gene-related peptide. Among the regions retrogradely labeled following tracer injections into the rhinal cortex, the parabrachial nucleus was considered the main origin of the tegmento-cortical projection. This conclusion was based on the circumscribed pattern of termination, as well as the differences noted between the pattern of anterograde labeling and the pattern obtained by thyrosine hydroxylase immunohistochemistry. The tracer injections into the dorsolateral tegmentum also revealed numerous retrogradely labeled cells in the layer 5 of the dorsomedial frontal cortex. In contrast, the rhinal cortex only showed few labeled cells and most of these cells were located in the layer 6/7. A comparison with other species indicates that the tenrec's parabrachial nucleus gives rise to the most extensive cortical projections but receives the least prominent input from the lateral cerebral hemisphere. The layer 6/7 projection may be a common mammalian feature but it is overshadowed by the layer 5 projection in higher mammals.