In mammals, the suprachiasmatic nuclei are involved in the generation of biological rhythms and are synchronized by light input coming from the retina. The targets of retinal afferents and the involvement of neurons containing gastrin-releasing and vasoactive intestinal peptides in photic reception were investigated in the suprachiasmatic nuclei of the Syrian hamster by using light- and electron-microscopic immunocytochemistry. Cholera toxin was used to trace retinal fibers and Fos immunoreactivity to visualize cellular response to light stimulation. Ultrastructural observations were made in the intermediate third of the nuclei, the area of highest overlap for the immunoreactivities investigated. Gastrin-releasing peptide and vasoactive intestinal peptide cell bodies were localized in the ventral part of the nuclei; their dense immunoreactive fiber network often displayed synaptic contacts. Both neuropeptides were colocalized in elongated cells observed near the optic chiasm. Following a light pulse in the middle of the subjective night, Fos protein was expressed in most gastrin-releasing peptide perikarya and in some vasoactive intestinal peptide cells. Retinal terminals mostly occurred in the midline zone between the suprachiasmatic nuclei. Symmetrical or asymmetrical retinal synapses were observed on gastrin-releasing peptide-immunoreactive dendrites and somata, but never on vasoactive intestinal peptide neurons. These results are discussed in relation to the photic entrainment of the circadian clock.

No abstract

The objective of this work is to describe the glomerular organization of the deutocerebrum in Blaberus craniifer and to test the hypothesis that the glomeruli are identifiable. The problem is studied using the techniques of analytical geometry, i.e., by measuring the location of the glomeruli in cartesian coordinate systems. Computerized geometrical and statistical techniques are described for the three-dimensional reconstruction and quantitative study of these brain structures. The invariance of the glomerular organization, and consequently the identifiability of the glomeruli, is based on three criteria: (1) the number of glomeruli per hemideutocerebrum (hDTC) is a constant (109); (2) the position of the glomeruli is symmetrical in the two sides of one individual and identical in the same side of two individuals, except for small local variations,(3) the dimensions of two positionally homologous glomeruli are statistically equal in the same and in different individuals, except for a macroglomerulus which is found only in the male. The causes of variability, from experimental and biological origin, are discussed and quantitatively evaluated. The interindividual biological variability, not ascribable to differences in size between brains, expressed as a 95% interval, is estimated to be 27 micrometer for the location of glomeruli and 12 micrometer for the diamter of glomeruli.

The glomerular organization of the hemideutocerebrum is analyzed quantitatively, using only spatial position, in four individuals (eight hDTCs) after a visual identification of glomeruli on graphic reconstructions. In order to assess directly the invariance of the neuropil in an insect brain the following is done: (1) The position of each glomerulus is compared to the position A' it should occupy if the hDTCs were identical. It is shown that in 80 and 71% of the cases, respectively, intra- and interindividual comparisons in the studied glomerulus is the nearest to A'. (2) The actual position is equally compared to the theoretical location A" each glomerulus should occupy if its absolute position could be changed but not its relative position with respect to its neighbors. The calculation of A" is based on process (3). In 86 and 80% of the cases, respectively, it is found that the actual position is that which is nearest to A". (3) An automated identification process, based on absolute and relative locations, but completely independent of visual identification, is described. It allows the identification of 77 and 74% of glomeruli, respectively. These matchings are in 96 and 90% of the cases identical to the visual matchings. The location predictability of most glomeruli is discussed to show the existence, nature, and limits of the hemideutocerebral invariance.