Search

Calcitonin gene-related peptide (CGRP) is a widespread neuropeptide with multiple central and peripheral targets. In an analysis on the expression of this peptide throughout the rat brain during postnatal development, we observed a discrepancy between results obtained by immunohistochemistry and by in situ hybridization. In the superior colliculus (SC), only the immunohistochemical signal could be detected (Terrado et al. [1997] Neuroscience 80:951-970). Here we focus our attention on this structure because the temporal pattern of CGRP immunoreactivity observed in the SC suggested the participation of this peptide in the postnatal maturation of the SC. In the present study, we describe in detail the postnatal development of collicular CGRP-immunoreactive structures and their spatiotemporal relationship with cholinergic modules and definitively demonstrate the local expression of CGRP in the SC. CGRP-immunopositive axons and neurons were distributed within the most ventral part of superficial strata and in the intermediate strata of the SC, showing a peak in staining intensity and density at the end of the first postnatal week. At P14, CGRPergic terminal fibers are arranged in small, clearly defined patches in a complementary manner with respect to the cholinergic modules, which start forming at this stage. By using Western blot and RT-PCR analyses, and by means of injections of antisense oligonucleotides, both the presence of CGRP peptide in the SC and the local expression of alpha-CGRP transcripts in collicular neurons were demonstrated. A possible role of CGRP is discussed in the context of postnatal modular compartmentalization of collicular afferents.

Synaptic plasticity is defined as any sort of change in intensity of the synaptic response or in the number of synapses that occurs in nervous structures. Long-lasting increase or decrease of synaptic efficacy — termed long-term potentiation (LTP) and long-term depression (LTD), respectively — has been related to the establishment of memory traces. Experimental induction of both types of long-lasting changes has contributed effectively to the study of the mechanisms that underlie the learning processes. Although the wealth of mechanisms related to synaptic plasticity still remains partially unveiled, at the core of this phenomenon lies the glutamate receptor system together with neurotrophins and their receptors, second-messenger cascades, and changes of gene expression. An increase in the numbers of synapses is termed structural plasticity, and is governed mainly by molecules belonging to the neurotrophin and cytokine families. This form of plasticity participates in the establishment of the intricate arrays of nervous connections during normal development, and in rearrangements of the neuronal circuitries that occur after nervous lesions.

Somatosensory stimuli from the body to deep and intermediate strata of the superior colliculus (SC) are relayed from the dorsal column nuclei (DCN), gracile (GrN) and cuneate (CuN). Electrophysiological studies have shown that the somatosensory representation in SC is arranged into a map-like pattern. However, there is a lack of studies confirming a morphological correlate of such an organization. On the other hand, after neonatal enucleation in rodents, somatosensory inputs ascend from their normal termination territory in intermediate and deep collicular strata to invade the more dorsally located visual strata. However, the origin of these reactive afferents has not been specified. By using anterograde (biotinylated dextran amine 10,000; BDA) and retrograde (Fluoro-Gold; FG) tracers, we studied separately the connection from GrN and CuN to the intact and neonatally deafferented SC. GrN-collicular afferents were found to terminate mainly within the periphery of the caudomedial SC quadrant, whereas CuN-collicular fibers innervated primarily the lateral part of the rostrolateral and caudolateral collicular quadrants, in a way consistent with previously described functional data. Retrograde tracing experiments using FG injected in SC confirmed this topographical arrangement. Injections of BDA in GrN or CuN of neonatally enucleated rats showed that reactive fibers reaching superficial strata are only those CuN-collicular fibers innervating the caudolateral SC quadrant, where the forelimb is represented. The present results provide an anatomical substrate for the known somatotopic organization of tactile representation in SC and further reinforce the previous proposal that the plastic reorganization of DCN-collicular afferents following neonatal enucleation constitutes a functional compensatory response.

Quantification of presynaptic terminals often requires laborious techniques that involve tissue preparation for ultrastructural analysis [Science 171 (1971) 804; Vision Res. Suppl. 3 (1971) 281]. Modern preembedding immunohistochemical techniques provide a high morphological resolution at the light microscope level, thus allowing us to identify immunostained presynaptic boutons using specific antibodies. When absolute density of boutons (Da) is analysed for comparison between control and deafferented nervous tissue, quantification may be distorted due to tissue shrinkage that follows deafferentiation. The magnitude of this effect must be, therefore, estimated to correct quantitative data. Using the superior colliculus (SC) as a model, an easily applicable protocol to quantify the density of small size labelled particles in control and deafferented nervous tissue is described. This protocol was used to analyse the effect of neonatal and adult enucleation on the adult pattern of cholinergic input to the rat SC. Statistical treatment of data demonstrated that neonatal enucleation caused a drastic increase in bouton density in the visual collicular layers, stratum zonale (SZ) and stratum griseum superficiale (SGS). The same lesion carried out in adult animals caused an increase in the bouton density exclusively in the SZ.

The effects of neonatal or adult enucleation on the final adult pattern of the rat visual corticocollicular (C-Co) connection were studied using the anterograde tracer biotinylated dextranamine 10,000 (BDA) iontophoretically injected in the primary visual cortex. In control animals, column-shaped terminal fields limited to a small portion of the collicular surface were observed. Synaptic boutons were present in all superficial strata of the superior colliculus (SC), with the highest density in the ventral part of the stratum griseum superficiale (SGS). Neonatal enucleation caused a considerable expansion of the contralateral visual C-Co terminal fields, which occupied almost the entire collicular surface, suggesting that axonal sprouting had occurred. In addition, terminal boutons tended to localize more dorsally in these cases compared with controls. Following enucleation in adult animals, no changes were observed with respect to the extension of the terminal fields, although a plastic reaction leading to an increase in the bouton density in the stratum zonale (SZ) and upper SGS was found, reflecting a process of reactive synaptogenesis at these levels. These results show that both neonatal and adult visual C-Co fibers react in response to retinal ablation, although this reaction shows distinct characteristics. Molecular factors, such as growth-associated cytoskeletal proteins operating in the cortical origin, and extracellular matrix components and myelin-associated axonal growth inhibitors acting on the collicular target very likely account for these differences.

The effects of neonatal and adult enucleation on the adult pattern of cholinergic inputs to the rat superior colliculus (SC) was analysed. In the superficial layers immunohistochemical labelling revealed that choline acetyltranferase (ChAT) was predominantly confined to single boutons which were almost continuously distributed throughout the rostrocaudal and lateromedial axes. In these layers a higher density of boutons was observed in the stratum zonale (SZ) and lower stratum griseum superficiale (SGSl) than in the upper stratum griseum superficiale (SGSu) and stratum opticum (SO). In intermediate collicular layers ChAT-immunostaining was mainly found in axonal profiles which were arranged in a patchy fashion. Neonatal enucleation caused a drastic increase in bouton density in the SZ, SGSu and SGSl. The density of boutons was particularly high in the SGSu, giving the appearance of an almost homogeneous distribution of boutons from the collicular surface down to the upper limit of SO. Visual deafferentiation at the adult stage was followed by an increase in the bouton density exclusively in the SZ. Neonatal enucleation produced a dorsoventral enlargement of the region containing patches of ChAT staining which was slightly greater following adult deafferentiation. The results described here show that after visual deafferentiation an increase in ChAT innervation to superficial and intermediate collicular layers occurs, providing new information regarding plasticity in the visual system. In view of previous data on cholinergic function in the central nervous system, such an increase could compensate for the loss of retinal excitatory input by facilitating neuronal responses in the SC.

Throughout rat embryogenesis we analysed the expression patterns of the three mature transcripts generated from the two calcitonin gene-related peptide genes: calcitonin, alpha-calcitonin gene-related peptide, and beta-calcitonin gene-related peptide messenger RNAs. In addition, we examined in parallel the distribution of calcitonin gene-related peptide and calcitonin immunoreactivity. Of the three transcripts, beta-calcitonin gene-related peptide messenger RNA was first detected in sensory ganglia on embryonic day 14, and by embryonic day 15 was seen to a lesser degree in motor neurons and autonomic ganglia. Starting at embryonic day 16, however, the highest levels of beta-calcitonin gene-related peptide messenger RNA were found in motor neurons rather than sensory ganglia. Alpha-calcitonin gene-related peptide messenger RNA was first detected on embryonic day 16 in both sensory ganglia and motor neurons, but at lower levels than beta-calcitonin gene-related peptide, particularly in the motor neurons of the spinal cord. By embryonic day 20, transcripts for alpha- and beta-calcitonin gene-related peptide were expressed in distinct brain regions. High levels of alpha-calcitonin gene-related peptide messenger RNA were detected in hypoglossal, facial, and parabrachial nuclei, and moderate levels in the trigeminal motor and ambiguus nuclei. By contrast, beta-calcitonin gene-related peptide messenger RNA was detected at low levels in hypoglossal, ambiguus, facial, and parabrachial nuclei, and at high levels in the trigeminal nucleus. In the oculomotor-trochlear nucleus, beta-calcitonin gene-related peptide messenger RNA was the sole isotype expressed. Low levels of messenger RNA for both calcitonin gene-related peptide transcripts were appreciated in the inferior olive. Outside the nervous system, alpha-calcitonin gene-related peptide messenger RNA was weakly expressed in the thyroid gland and beta-calcitonin gene-related peptide messenger RNA in the thymus. Throughout embryogenesis, calcitonin gene-related peptide immunoreactivity usually followed the expression of either alpha- or beta-calcitonin gene-related peptide messenger RNA. Calcitonin messenger RNA and protein were detected only in the thyroid gland from embryonic day 18 onward. This work shows that of the three mature transcripts produced by the two calcitonin gene-related peptide genes, beta-calcitonin gene-related peptide messenger RNA is the predominant transcript produced early in rat embryogenesis. However, by perinatal stages alpha-calcitonin gene-related peptide shows the highest expression in the brain and spinal cord. In autonomic ganglia, beta-calcitonin gene-related peptide is either the sole or the predominant transcript. Unlike the chick embryo in which calcitonin messenger RNA is expressed early in the CNS, in rat it was only expressed outside the nervous system in the thyroid gland during the last days of embryogenesis.