Defining the attributes of individual central pattern-generating (CPG) neurons underlying various rhythmic behaviors are fundamental to our understanding of how the brain controls motor programs, such as respiration and locomotion. To this end, we have explored a simple invertebrate preparation in which the neuronal basis of respiratory rhythmogenesis can be investigated from the whole animal to a single cell level. An identified dopaminergic neuron, termed right pedal dorsal 1 (RPeD1), is a component of the CPG network which controls hypoxia-driven, aerial respiration in the fresh water snail Lymnaea stagnalis. Using intact, semi-intact and isolated brain preparations, we have discovered that in addition to its role as a respiratory CPG neuron, RPeD1 co-ordinates sensory-motor input from the pneumostome (the respiratory orifice) at the water/air interface to initiate respiratory rhythm generation. An additional, novel role of RPeD1 was also found. Specifically, direct intracellular stimulation of RPeD1 induced pneumostome openings, in the absence of motor neuronal activity. To determine further the role of RPeD1 in the respiratory behavior of intact animals, either its axon was severed or the soma selectively killed. Many components of the respiratory behavior in the intact animals were found to be perturbed following RPeD1 axotomy or 'somatomy' (soma removed). Taken together, the data presented provide a direct demonstration that RPeD1 is a multifunctional CPG neuron, which also serves many additional roles in the control of breathing behavior, ranging from co-ordination of mechanosensory input to the motor control of the respiratory orifice.

A Chinese baby girl was born after an uncomplicated pregnancy and a normal spontaneous vaginal delivery to a gravida 2, para 1, 32-year-old mother at 37 weeks' gestation. Neither parent had a history of alcohol or drug ingestion. There was no history of consanguinity. The Apgar scores were 7 at 1 min and 9 at 5 min. Her birth weight was 5 lb 14 oz and her length was 46 cm. At birth, she was noted to have grayish areas on the face and lower back. The infant was seen at 3 months of age because of persistent regurgitations. On average, she regurgitated 7-8 times a day. Physical examination showed that the infant was not in distress. Her weight was 6 lb 4 oz and length 47 cm. She had a Mongolian spot measuring 0.8 x 1.2 cm in the left temporal area (Fig. 1) and another Mongolian spot measuring 4 cm in diameter in the lumbar area. The Mongolian spots were grayish in color and the pigmentation was uniform in intensity. The rest of the examination was essentially normal. The infant was diagnosed to have Mongolian spots and gastroesophageal reflux. The latter was treated with postural therapy, thickened feedings, and metoclopramide 0.35 mg q.i.d. The infant was seen again at 4 months of age for a reassessment. There was no noticeable change in shape, size, or color of the Mongolian spots.

Most information available to date regarding the cellular and synaptic mechanisms of target cell selection and specific synapse formation has primarily come from in vitro cell culture studies. Whether fundamental mechanisms of synapse formation revealed through in vitro studies are similar to those occurring in vivo has not yet been determined. Taking advantage of the regenerative capabilities of adult molluscan neurons, we demonstrate that when transplanted into the host ganglia an identified neuron reestablishes its synaptic connections with appropriate targets in vivo. This synaptogenesis, however, was possible only if the targets were denervated from the host cell. Specifically, the giant dopamine neuron right pedal dorsal 1 (RPeD1) located in the pedal ganglia was isolated from a donor brain and transplanted into the visceral ganglia of the recipient brain. We discovered that within 2-4 days the transplanted RPeD1 exhibited extensive regeneration. However, simultaneous intracellular recordings failed to reveal synapses between the transplanted cell and its targets in the visceral ganglia, despite physical overlap between the neurites. To test whether the failure of a transplanted cell to innervate its target was due to the fact that the targets continued to receive input from the native RPeD1, the latter soma was surgically removed prior to the transplantation of RPeD1. Even after the removal of host soma, the transplanted RPeD1 failed to innervate the targets such as visceral dorsal 4 (VD4)-despite extensive regeneration by the transplanted cell. However, when RPeD1 axon was allowed to degenerate completely, the transplanted RPeD1 successfully innervated all of its targets and these synapses were similar to those seen between host RPeD1 and its targets. Taken together, our data demonstrate that the transplanted cells will innervate their potential targets only if the targets were denervated from the host cell. These data also lend support to the idea that, irrespective of their physical location in the brain, the displaced neurons are able to regenerate, recognize their targets, and establish specific synapses in the nervous system.

Precise neuronal connectivity during development is subservient to all nervous system functions in adult animals. However, the cellular mechanisms that mastermind this neuronal connectivity remain largely unknown. This lack of fundamental knowledge regarding nervous system development is due in part to the immense complexity of mammalian brain, as cell-cell interactions between defined sets of pre- and postsynaptic partners are often difficult to investigate directly. In this study, we developed a novel model system which has allowed us to reconstruct synapses between identified motor neurons and their target heart muscle cell in a soma-muscle configuration. Utilizing this soma-myocardial cell synapse model, we demonstrate that synapses between somata and heart muscle cells can be reconstructed in cell culture. The soma-myocardial cell synapses required 12-24 h to develop and thus differed temporally from conventional neuromuscular synapses (seconds to a few minutes). We also demonstrate that the synapses are target cell-type-specific and are most likely independent of transmitter phenotypic characteristics of presynaptic neurons.

All brain functions, ranging from motor behaviour to cognition, depend on precise developmental patterns of synapse formation between the growth cones of both pre- and postsynaptic neurons. While the molecular evidence for the presence of 'pre-assembled' elements of synaptic machinery prior to physical contact is beginning to emerge, the precise timing of functional synaptogenesis between the growth cones has not yet been defined. Moreover, it is unclear whether an initial assembly of various synaptic molecules located at the extrasomal regions (e.g. growth cones) can indeed result in fully mature and consolidated synapses in the absence of somata signalling. Such evidence is difficult to obtain both in vivo and in vitro because the extrasomal sites are often challenging, if not impossible, to access for electrophysiological analysis. Here we demonstrate a novel approach to precisely define various steps underlying synapse formation between the isolated growth cones of individually identifiable pre- and postsynaptic neurons from the mollusc Lymnaea stagnalis. We show for the first time that isolated growth cones transformed into 'growth balls' have an innate propensity to develop specific and multiple synapses within minutes of physical contact. We also demonstrate that a prior 'synaptic history' primes the presynaptic growth ball to form synapses quicker with subsequent partners. This is the first demonstration that isolated Lymnaea growth cones have the necessary machinery to form functional synapses.