Search

In order to characterize the neuronal pathways underlying central nervous processing of odor signals, the synaptic circuitry of identified neuron individuals or types within cockroach antennal lobe glomeruli has been studied in the electron microscope. The anatomical findings are consistent with physiological data from other investigations. For the characterization of the synaptic connections, different combinations of neuron labelings have been applied including (a) horseradish peroxidase injection of local interneurons (INs) or uniglomerular projection neurons (uPNs), (b) gamma-aminobutyric acid (GABA)-immunogold labeling, and (c) experimentally induced degeneration of antennal receptor neuron projections (RNs). The following pathways between glomerular input and output neurons are proposed: (1) A monosynaptic excitatory route, and (2) at least two polysynaptic routes, one of them probably inhibitory and the other disinhibitory. (3) Inhibitory feedback connections exist from these two paths via GABA-positive INs onto both RNs as well as uPNs. (4) Since a given GABA-positive IN innervates a large number of antennal lobe glomeruli, inhibitory signals could also be passed to neighbored glomeruli.

The distribution of odor-evoked neuronal activity in the antennal lobes of the blowfly Calliphora vicina has been studied by means of [3H] 2-deoxyglucose (2-DG) uptake. Stimulation with natural attractants like meat or cheese induced uptake of 2-DG in a large number of glomeruli, stimulation with the single odorant butyric acid only in a small number of glomeruli. In control brains, the majority of glomeruli showed low 2-DG uptake. The distribution of labeled glomeruli was odor-specific and consistent in different specimen. The labeling patterns evoked by different odors overlapped partially but were clearly distinct. The highest uptake within the glomeruli was found in the axons of antennal receptor neurons. The present data provide evidence that different odors are represented as defined spatial patterns of activity across the antennal lobe glomeruli. The overlapping patterns suggest that certain glomeruli participate in the nervous processing of different odors., (Copyright 1998 Elsevier Science B.V.)

Synapses within deutocerebral glomeruli between GABA-immunoreactive, putatively inhibitory local interneurons and uniglomerular projection (output) neurons were demonstrated by means of a combination of GABA-immunogold labeling and intracellular HRP injection. The following connections were identified. 1) GABA-immunoreactive (GABAir) neurons form output synapses in a dyadic fashion onto a uniglomerular projection neuron and, in addition, a second GABAir neuron. A uniglomerular projection neuron in turn forms dyadic output synapses onto two GABAir neurons. Several examples of reciprocal connections have been identified between, first, GABAir neurons and uniglomerular projection neurons, and, second, GABAir neurons themselves. 2) GABAir neurons are serially connected with uniglomerular projection neurons via interposed GABAir processes. In some cases, also the first GABAir process of such a polysynaptic connection formed an output synapse onto the projection neuron. Such serial connections may form the structural basis for both, the feedforward inhibition as well as the feedforward disinhibition of uniglomerular projection neurons by GABAergic neurons. The reciprocal contacts may serve as control devices that modulate the output activity of the projection neurons.

Synapses between three types of antennal lobe neurons, namely, local multiglomerular interneurons, antennal receptor neurons, and gamma-aminobutyric acid (GABA)-immunoreactive neurons, were studied by means of a combination of three different markers. The interneurons were labeled by intracellular injection of horseradish peroxidase (HRP) into a single soma or a small group of neurons. Antennal receptor cells were marked by experimentally induced anterograde degeneration, and GABA-containing neurons were identified by postembedding immunogold staining. The following types of connections were found: Local interneurons receive input synapses from 1) degenerated receptor neuron axons, 2) GABA-immunogold-labeled neurites, and 3) non-GABA-immunoreactive neurons. The interneurons form output synapses onto the same three neuron groups. Contacts were also found between HRP-labeled interneurons themselves. The majority of synapses were dyadic. In most cases, only one postsynaptic neuronal process of the dyads was labeled and, thus, was identified. Polysynaptic connections were found between GABA-immunoreactive neurites, HRP-labeled interneuron processes, and nonlabeled neurites or between HRP-labeled interneuron processes and two interconnected GABA-immunoreactive processes. The present findings provide anatomical evidence for an earlier suggested monosynaptic connection between afferent receptor fibers and local, at that time putative, GABAergic interneurons. They further reveal that local multiglomerular interneurons are synaptically interconnected. The interneurons, in addition, form serial connections via more than one GABA-immunoreactive neuron with non-GABA-immunoreactive and putative projection neurons. Such polysynaptic connections would be a substrate for a feed-forward "disinhibition" of projection neurons, which has been suggested on the basis of electrophysiological findings.

A combination of three different labels was used to demonstrate synapses between three types of neurons within the glomeruli: 1) antennal receptor cells, 2) gamma-aminobutyric acid (GABA)-immunoreactive neurons, and 3) uniglomerular projection neurons. Receptor cell axons were experimentally severed and caused to degenerate; uniglomerular projection neurons, a subgroup of glomerular output neurons, were labeled by intracellular horseradish peroxidase (HRP) injection and GABA-containing neurons by postembedding immunogold staining. The following synaptic connections were identified: 1) Receptor cell axons form monosynaptic contacts in a dyadic fashion onto a dendritic process of a uniglomerular projection neuron and in addition onto a GABA-immunoreactive neuron. 2) Receptor cell axons form polysynaptic connections with dendrites of uniglomerular projection neurons via GABA-immunoreactive neurons. 3) GABA-immunoreactive neurons form dyadic output synapses onto receptor cell axons and in addition onto projection neuron dendrites. These findings provide further evidence that signal transfer from receptor cells onto uniglomerular projection neurons is mediated by two different paths: first, a monosynaptic and presumably excitatory route and, second, an inhibitory polysynaptic route via GABAergic, most likely multiglomerular interneurons. The output synapses of GABA-immunoreactive neurons onto both receptor cells and uniglomerular projection neurons are assumed to exert control functions in regulating the neuronal activity within the glomeruli.

Both antennal receptor cell axons and uniglomerular projection neurons of the antennal lobe were specifically labeled, and their synaptic relationship was studied at the fine structural level. The labelings were applied in different combinations: i) Experimentally induced anterograde degeneration of sensory-afferent axons was combined with injection of horseradish peroxidase into uniglomerular projection neurons. ii) Lucifer Yellow was injected into uniglomerular projection neurons, and receptor cell axons were anterogradely labeled with the lipophilic dye DiI. The fluorescent dyes were transformed by immuno- or photochemical treatment into electron-dense markers. In both types of preparations, a considerable number of monosynaptic output synapses from antennal receptor neurons onto processes of uniglomerular projection neurons were identified within the glomeruli of the lobe. In most cases, the receptor axon was connected in a dyadic fashion firstly to a process of a projection neuron and secondly to a nonlabeled process. The results clearly demonstrate a direct connection between receptor cells and output neurons of the cockroach antennal lobe which exists in parallel to the already proposed and demonstrated polysynaptic connection via inhibitory local interneurons.

Co-cultures of rat basal forebrain and cerebral cortex were maintained from 1 to 5 weeks in vitro with serum-free defined medium. The formation of synaptic connections between basal forebrain afferent fibres and cortical neurons was studied by specific labelling with three staining techniques, including (i) neuronal tract tracing with the fluorescent dye 1,1'-dioctodecyl-3,3,3'3'- tetramethylindocarbocyanine perchlorate, (ii) acetylcholinesterase histochemistry, and (iii) choline acetyltransferase immunocytochemistry. Both basal forebrain and cerebral cortex tissue displayed organotypic characteristics in culture. Cerebral cortex revealed a dense innervation by axonal projections from the basal forebrain. All three labelling techniques produced similar results at the light microscopic level, with densest innervation located in the marginal zone. At the fine structural level, the 1,1'-dioctodecyl-3,3,3'3'-tetramethylindocarbocyanine perchlorate-, acetylcholinesterase- and choline acetyltransferase-stained basal forebrain afferents all revealed a number of synaptic contacts with cortical neurons. The contacts displayed consistent synaptic features, including presynaptic accumulation of small round vesicles, cleft widening, and postsynaptic densities forming symmetric synapses. These morphological characteristics of connections formed in vitro are similar to basal forebrain cholinergic projections to cerebral cortex in normal brain. Based on these results, this tissue culture model appears to be an useful tool for investigations of the development of cholinergic innervation of cerebral cortex.

Development of the innervation of the cerebral cortex by acetylcholinesterase (AChE)-stained basal forebrain neurons was studied in vitro using the roller tube technique. Slice cultures were maintained from 3 days to 4 weeks either in serum based medium or in chemically defined medium, each supplemented in some cases with nerve growth factor (NGF). The distribution of AChE and choline acetyltransferase (CAT)-containing neurons was investigated using histo- and immunocytochemical techniques. Slice cultures of basal forebrain revealed the presence of large and medium sized AChE-positive neurons. Within one week of cultivation, numerous AChE-labeled fibers could be seen growing out from the basal forebrain toward the cortex. After entering cortical tissue most of the afferent basal forebrain fibers projected either radially or obliquely into the cortical layers. Many afferent axons initially also travelled tangentially within the white matter, and turned then to grow into the cortical layers. Cerebral cortex tissue maintained a coarse laminar organization. Ramifications of basal forebrain fibers were visible within the subplate region, the deep and superficial cortical layers, and within the marginal zone; greatest density occurred in the subplate region and in marginal zones. Many of these processes exhibited branching patterns markedly similar to those observed during cortical development in vivo. Cortex slices placed with the pial surface adjacent to the basal forebrain revealed AChE-stained fibers that entered the cortical tissue through the marginal surface and gave off ramifications within the superficial layers and, less frequently, the deeper cortical layers. CAT-immunostaining revealed labeled cell bodies and neurites only in the basal forebrain, not in the cortex tissue. Control experiments with co-cultures of basal forebrain and cerebellum slices showed no AChE-positive fiber ingrowth into the cerebellum tissue. The results of these studies demonstrate that basal forebrain projections to cerebral cortex in vitro appear similar to the projections that develop in vivo, and indicate that organotypic co-cultures provide a valuable model for studies of developing cortical afferents.