Alan Gelperin

Differential conditioning to a compound stimulus and its components in the terrestrial mollusc Limax maximus

The neuronal mechanism of associative learning and memory storage operating in the central nervous system of the terrestrial mollusc Limax maximus has been modeled as a computer simulation called LIMAX (Gelperin, J.J. Hopfield, & Tank, 1985). One test of the LIMAX model is to determine whether Limax can learn to avoid a compound stimulus composed of the mixture of two innately attractive odors without simultaneously learning an aversion to the individual odors comprising the mixture

Distribution of FMRFamide-like immunoreactivity in the nervous system of the slug Limax maximus.

SummaryThe distribution of FMRFamide-like immunoreactive neurons in the nervous system of the slug Limax maximus was studied using immunohistochemical methods. Approximately one thousand FMRFamide-like immunoreactive cell bodies were found in the central nervous system. Ranging between 15 μm and 200 μm in diameter, they were found in all 11 ganglia of the central nervous system. FMRFamide-like immunoreactive cell bodies were also found at peripheral locations on buccal nerve roots. FMRFamide-like immunoreactive nerve fibres were present in peripheral nerve roots and were distributed extensively throughout the neuropil and cell body regions of the central ganglia. They were also present in the connective tissue of the perineurium, forming an extensive network of varicose fibres. The large number, extensive distribution and great range in size of FMRFamide-like immunoreactive cell bodies and the wide distribution of immunoreactive fibres suggest that FMRFamide-like peptides might serve several different functions in the nervous system of the slug.

Olfactory Processing and Associative Memory: Cellular and Modeling Studies

Publisher Summary For understanding the operation of some types of neural circuits, the computational approach is an essential complement to the traditional strategies of cellular analysis. Olfactory processing systems seem likely to operate as collective neural systems. Computer simulations have shown that collective networks are adept at classifying patterns of input activity based on imprecise or incomplete data. The spatial and temporal patterns of sensory neuron activity resulting from repeated presentations of a given odor are not likely to be highly specific and thus informationally imprecise from the viewpoint of any particular olfactory sensory cell. The sensory input pattern will be constant only when viewed over the entire ensemble of sensory cells. The analysis of olfactory processing and associative memory in the mollusk Umax maximus is being carried out at several levels. Behavioral experiments define the computational tasks accomplished by the olfactory system. Electrophysiological measurements and anatomical studies delimit the pathways of information flow and constrain the universe of plausible models.

Odors can induce feeding motor responses in the terrestrial mollusc Limax maximus.

Highly developed odor learning was shown in the terrestrial slug Limax maximus. In addition, several key cellular elements of the neural network that controls ingestive feeding have been identified. The results of 3 experiments demonstrate an interaction between odor input and ingestive feeding in that olfactory stimulation with behaviorally attractive odors summed with tactile stimulation from plain agar to produce ingestion of plain agar. Agar ingestion did not occur in the absence of attractive odor stimulation. The adequacy of odor stimulation to trigger agar ingestion was altered by associative learning. Innately attractive odors rendered repellent by associative learning no longer triggered agar ingestion, whereas innately repellent odors rendered attractive by conditioning triggered agar ingestion. The newly discovered feeding command cells in the Limax cerebral ganglion are a logical cellular locus for this interaction.

Opiate agonists activate feeding in Limax : comparison of in vivo and in vitro effects

The neural control system for feeding in the terrestrial mollusc Limax maximus is modulated by at least two major families of peptides. Sequence homology between one of the peptides known to modulate Limax feeding and some members of the opioid peptide family suggested that opioid peptides might also modulate Limax feeding. Experiments with the mu agonist morphine and the kappa agonist U50,488H showed that the probability of feeding, but not meal size, was increased by morphine injection into intact animals, whereas the length of feeding motor program responses elicited from the isolated lip-brain preparation of Limax was augmented by U50,488H. The behavioral effect of morphine was blocked by naltrexone injection, whereas the physiological effect of U50,488H was blocked by naloxone. Factors that influence the behavioral and electrophysiological effects of opioids on mollusc feeding are discussed.

Nitric Oxide, Odour Processing and Plasticity

We are pursuing a cellular and computational analysis of odour recognition and odour learning in the terrestrial gastropod Limax maximus. Two design features in the olfactory system of Limax may be common elements of generalist olfactory systems. They are 1) coherent oscillations in the second-order circuitry processing the sensory input and 2) involvement of the intercellular messenger nitric oxide in both sensory responses and the circuit dynamics of the oscillating olfactory network. The procerebral (PC) lobe of the Limax cerebral ganglion contains of order 105 local interneurons and receives direct input from olfactory receptors. Field potential recordings in the PC show a 0.7 Hz oscillation which is altered by odour input. Optical recordings of voltage changes in local regions of the PC show waves of depolarization originating at the distal pole and propagating to its base. Weak odour stimulation transiently switches PC activity from wave propagating mode to a spatially uniform activity mode. Both attractive and repellent odours cause this mode switch in PC activity. The field potential oscillation in the PC lobe depends on intercellular communication via nitric oxide (NO), based on effects of agents which eliminate or increase NO levels in the PC. Application of oxyhemeprotein or NO synthase inhibitor stops the field potential oscillation, reversibly. Application of exogenous NO using several NO donors increases the oscillation frequency. Liberation of caged NO by near UV flashes applied to preparations treated with dipotassium nitrosylpentachlororuthenate (NPR) also speeds the field potential oscillation. Nystatin perforated-patch recordings from the bursting cells in the PC presumed to drive the oscillation show that NO increases the burst frequency in these cells. These design features of the PC lobe odour processing circuitry may relate to the highly developed odour sensitivity and odour learning ability of Limax.

Localization and synthesis of monoamines in regions of Limax CNS controlling feeding behavior

Localization and synthesis of dopamine and serotonin in the cerebral and buccal ganglia of Limax maximus were studied. A combination of fluorescence histochemistry, immunocytochemistry, and microchemical analysis showed that both amines were localized to particular cell groups and fiber tracts within and between the two sets of ganglia. Since these ganglia control feeding behavior, which is readily modified by associative learning, these studies have direct bearing on analysis of both motor control and learning mechanisms.

Amino acids and serotonin in Limax maximus after a tryptophan devoid diet

1. Animals avoid diets lacking an essential amino acid, such as tryptophan (TRP), the precursor for serotonin (5-HT). 5-HT is important in the control of feeding. 2. To study the effects of TRP deprivation, slugs were fed TRP-devoid (DEV) or control (COR) diets. 3. Food intake was depressed in DEV, as expected, but after 2 weeks, the serontonergic metacerebral giant cell in DEV was still functional. 4. Neither brain 5-HT nor plasma TRP concentration was affected. 5. Compared with food-restricted animals that had reductions in most amino acids, the DEV group sustained a marked plasma amino acid imbalance.

Release of dopamine and serotonin from Limax ganglia in vitro

Abstract 1. We wish to establish the kinetics of serotonin and dopamine release from Limax cerebral and buccal ganglia and find selective treatments to modify their release kinetics. 2. The release of dopamine and serotonin from isolated ganglia was stimulated by high potassium exposure with and without prior treatment of ganglia with 6-hydroxydopamine (6-OHDA). 3. Single ganglia release significant quantities of monoamines during a single 5 min high K + exposure. Multiple high K + exposures deplete a readily releasable transmitter store with little effect on storage pools. 4. 6-OHDA exposure depletes readily releasable DA with little effect on total ganglion DA content or on serotonin. 5. Feeding motor program responsiveness is suppressed reversibly by whole ganglion high K + treatment.