Monitoring Spiking Activity of Many Individual Neurons in Invertebrate Ganglia.

Optical recording with fast voltage sensitive dyes makes it possible, in suitable preparations, to simultaneously monitor the action potentials of large numbers of individual neurons. Here we describe methods for doing this, including considerations of different dyes and imaging systems, methods for correlating the optical signals with their source neurons, procedures for getting good signals, and the use of Independent Component Analysis for spike-sorting raw optical data into single neuron traces. These combined tools represent a powerful approach for large-scale recording of neural networks with high temporal and spatial resolution.

Evidence that the swim afferent neurons of tritonia diomedea are glutamatergic.

The escape swim response of the marine mollusc Tritonia diomedea is a well-established model system for studies of the neural basis of behavior. Although the swim neural network is reasonably well understood, little is known about the transmitters used by its constituent neurons. In the present study, we provide immunocytochemical and electrophysiological evidence that the S-cells, the afferent neurons that detect aversive skin stimuli and in turn trigger Tritonia's escape swim response, use glutamate as their transmitter. First, immunolabeling revealed that S-cell somata contain elevated levels of glutamate compared to most other neurons in the Tritonia brain, consistent with findings from glutamatergic neurons in many species. Second, pressure-applied puffs of glutamate produced the same excitatory response in the target neurons of the S-cells as the naturally released S-cell transmitter itself. Third, the glutamate receptor antagonist CNQX completely blocked S-cell synaptic connections. These findings support glutamate as a transmitter used by the S-cells, and will facilitate studies using this model system to explore a variety of issues related to the neural basis of behavior.

Intrinsic neuromodulation: altering neuronal circuits from within.

There are two sources of neuromodulation for neuronal circuits: extrinsic inputs and intrinsic components of the circuits themselves. Extrinsic neuromodulation is known to be pervasive in nervous systems, but intrinsic neuromodulation is less recognized, despite the fact that it has now been demonstrated in sensory and neuromuscular circuits and in central pattern generators. By its nature, intrinsic neuromodulation produces local changes in neuronal computation, whereas extrinsic neuromodulation can cause global changes, often affecting many circuits simultaneously. Studies in a number of systems are defining the different properties of these two forms of neuromodulation.

Habituation and iterative enhancement of multiple components of the Tritonia swim response.

To understand the relationship between memory storage sites in the brain and learned changes in behavior, the learned behavior must be characterized. However, even simple types of learning may be quite complex. Repeated elicitation of the Tritonia swim produced multiple changes in the response. Several types of acquisition curves were observed in a single experiment depending on the response component measured. Habituation (response decrement) and iterative enhancement (response facilitation) occurred simultaneously in different swim components. The acquisition curve for one component could be modulated by stimulus strength. Because the Tritonia swim neural network is well studied, it may be possible to causally relate experience-dependent behavioral changes to the underlying memory trace in this marine mollusk.

Realistic simulation of the Aplysia siphon-withdrawal reflex circuit: roles of circuit elements in producing motor output.

The circuitry underlying the Aplysia siphon-elicited siphon-withdrawal reflex has been widely used to study the cellular substrates of simple forms of learning and memory. Nonetheless, the functional roles of the different neurons and synaptic connections modified with learning have yet to be firmly established. In this study we constructed a realistic computer simulation of the best-understood component of this network to better understand how the siphon-withdrawal circuit works. We used an integrate-and-fire scheme to simulate four neuron types (LFS, L29, L30, L34) and 10 synaptic connections. Each of these circuit components was individually constructed to match the mean or typical example of its biological counterpart on the basis of group measurements of each circuit element. Once each cell and synapse was modeled, its free parameters were fixed and not subject to further manipulation. The LFS motor neurons respond to sensory input with a brief phasic burst followed by a long-lasting period of tonic firing. We found that the assembled model network responded to sensory input in a qualitatively similar fashion, suggesting that many of the interneurons important for producing the LFS firing response have now been identified. By selectively removing different circuit elements, we determined the contribution of each of the LFS firing pattern. Our first finding was that the monosynaptic sensory neuron to motor neuron pathway contributed only to the initial brief burst of the LFS firing response, whereas the polysynaptic pathway determined the overall duration of LFS firing. By making more selective deletions, we found that the circuit elements responsible for transforming brief sensory neuron discharges into long-lasting LFS firing were the slow components of the L29-LFS fast/slow excitatory postsynaptic potentials. The inhibitory L30 neurons exerted a significant braking action on the flow of excitatory information through the circuit. Interestingly, L30 lost its ability to reduce the duration of LFS firing at high stimulus intensities. This was found to be due to the intrinsic nature of L30's current-frequency relationship. Some circuit elements, including interneuron L34, and the electrical coupling between L29 and L30 were found to have little impact when subtracted from the network. These results represent a detailed dissection of the functional roles of the different elements of the siphon-elicited siphon-withdrawal circuit in Aplysia. Because many vertebrate and invertebrate circuits perform similar tasks and contain similar information processing elements, aspects of these results may be of general significance for understanding the function of motor networks. In addition, because several sites in this network store learning-related information, these results are relevant to elucidating the functional significance of the distributed storage of learned information in Aplysia.

Removal of spike frequency adaptation via neuromodulation intrinsic to the Tritonia escape swim central pattern generator.

For the mollusc Tritonia diomedea to generate its escape swim motor pattern, interneuron C2, a crucial member of the central pattern generator (CPG) for this rhythmic behavior, must fire repetitive bursts of action potentials. Yet, before swimming, repeated depolarizing current pulses injected into C2 at periods similar those in the swim motor program are incapable of mimicking the firing rate attained by C2 on each cycle of a swim motor program. This resting level of C2 inexcitability is attributable to its own inherent spike frequency adaptation (SFA). Clearly, this property must be altered for the swim behavior to occur. The pathway for initiation of the swimming behavior involves activation of the serotonergic dorsal swim interneurons (DSIs), which are also intrinsic members of the swim CPG. Physiologically appropriate DSI stimulation transiently decreases C2 SFA, allowing C2 to fire at higher rates even when repeatedly depolarized at short intervals. The increased C2 excitability caused by DSI stimulation is mimicked and occluded by serotonin application. Furthermore, the change in excitability is not caused by the depolarization associated with DSI stimulation or serotonin application but is correlated with a decrease in C2 spike afterhyperpolarization. This suggests that the DSIs use serotonin to evoke a neuromodulatory action on a conductance in C2 that regulates its firing rate. This modulatory action of one CPG neuron on another is likely to play a role in configuring the swim circuit into its rhythmic pattern-generating mode and maintaining it in that state.

Structure of the network mediating siphon-elicited siphon withdrawal in Aplysia.

1. The network mediating siphon-elicited siphon withdrawal in Aplysia is a useful model system for cellular studies of simple forms of learning and memory. Here we describe three new cells in this circuit, L33, L34, and L35, and several new connections among the following network neurons: LE, L16, L29, L30, L32, L33, L34, and L35. On the basis of these findings we present an updated diagram of the network. Altogether, 100 neurons have now been identified in the abdominal ganglion that can participate in both siphon-elicited and spontaneous respiratory pumping siphon withdrawals. 2. Two features of the interneuronal population may have important behavioral functions. First, the L29 interneurons make fast and slow excitatory connections onto the LFS cells, which may be important for transforming brief sensory neuron discharges into the long-lasting motor neuron firing that underlies withdrawal duration. Second, inhibitory interneurons are prominent in the network. The specific connectivity of certain of these interneurons is appropriate to block potentially interfering inhibitory inputs from other networks during execution of the behavior. 3. Deliberate searches have so far revealed very few excitatory interneuronal inputs to the network interneurons and motor neurons within the abdominal ganglion. These results, together with intracellular studies by others, are more consistent at present with a relatively dedicated rather than a highly distributed organizational scheme for the siphon-elicited siphon withdrawal circuitry.

Intrinsic neuromodulation in the Tritonia swim CPG: the serotonergic dorsal swim interneurons act presynaptically to enhance transmitter release from interneuron C2.

Heterosynaptic enhancement of transmitter release is potentially very important for neuronal computation, yet, to our knowledge, no prior study has shown that stimulation of one neuron directly enhances release from an interneuron. Here, we demonstrate that in the marine mollusk Tritonia diomedea, the serotonergic dorsal swim interneurons (DSIs) heterosynaptically increase the amount of transmitter released from another interneuron, C2. Stimulation of a single DSI at physiological firing frequencies increases the size of synaptic potentials evoked by C2. This increase in synaptic efficacy is correlated with an increase in homosynaptic paired-pulse facilitation by C2. Thus, it is likely to be due to an enhancement of transmitter release from C2, rather than a postsynaptic action on the followers of C2. This is further supported by the fact that DSI stimulation enhances the strengths of all chemical synapses made by C2 within the swim network, regardless of their sign. Furthermore, DSI enhances the amplitude of C2 synaptic potentials recorded in neurons that DSI itself does not synapse with. Finally, DSI differentially modulates different synaptic inputs to the same postsynaptic target; while increasing C2-evoked EPSPs it simultaneously decreases the size of EPSPs evoked by other DSIs. The heterosynaptic facilitation of C2 synaptic potentials by DSI is not caused by a simple depolarization of C2, but may be a direct action on the transmitter release mechanism. This neuromodulatory effect, which is intrinsic to the circuitry of the central pattern generator for escape swimming in Tritonia, may be important for self-reconfiguration of the swim motor network.

Intrinsic neuromodulation in the Tritonia swim CPG: serotonin mediates both neuromodulation and neurotransmission by the dorsal swim interneurons.

1. Neuromodulation has previously been shown to be intrinsic to the central pattern generator (CPG) circuit that generates the escape swim of the nudibranch mollusk Tritonia diomedea; the dorsal swim interneurons (DSIs) make conventional monosynaptic connections and evoke neuromodulatory effects within the swim motor circuit. The conventional synaptic potentials evoked by a DSI onto cerebral neuron 2 (C2) and onto the dorsal flexion neurons (DFNs) consist of a fast excitatory postsynaptic potential (EPSP) followed by a prolonged slow EPSP. In their neuromodulatory role, the DSIs produce an enhancement of the monosynaptic connections made by C2 onto other CPG circuit interneurons and onto efferent flexion neurons. Previous work showed that the DSIs are immunoreactive for serotonin. Here we provide evidence that both the neurotransmission and the neuromodulation evoked by the DSIs are produced by serotonin, and that these effects may be pharmacologically separable. 2. Previously it was shown that bath-applied serotonin both mimics and occludes the modulation of the C2 synapses by the DSIs. Here we find that pressure-applied puffs of serotonin mimic both the fast and slow EPSPs evoked by a DSI onto a DFN, whereas high concentrations of bath-applied serotonin occlude both of these synaptic components. 3. Consistent with the hypothesis that serotonin mediates the actions of the DSIs, the serotonin reuptake inhibitor imipramine prolongs the duration of the fast DSI-DFN EPSP, increases the amplitude of the slow DSI-DFN EPSP, and increases both the amplitude and duration of the modulation of the C2-DFN synapse by the DSIs. 4. Two serotonergic antagonists were found that block the actions of the DSIs. Gramine blocks the fast DSI-DFN EPSP, and has far less of an effect on the slow EPSP and the modulation. Gramine also diminishes the depolarization evoked by pressure-applied serotonin, showing that it is a serotonin antagonist in this system. In contrast, methysergide greatly reduces both the slow EPSP and the modulation evoked by the DSIs, but has mixed effects on the fast EPSP. Methysergide also blocks the ability of exogenous serotonin to enhance the C2-DFN EPSP, demonstrating that it antagonizes the serotonin receptors responsible for this modulation. 5. Taken together with previous work, these results indicate that serotonin is likely to be responsible for all three actions of the DSIs that were examined: the fast and slow DSI-DFN EPSPs and the neuromodulation of the C2-DFN synapse. These results also indicate that the conventional and neuromodulatory effects of the DSIs may be pharmacologically separable. In future work it may be possible to determine the functional role of each in the swim circuit.

Single neuron control over a complex motor program.

While there are many instances of single neurons that can drive rhythmic stimulus-elicited motor programs, such neurons have seldom been found to be necessary for motor program function. In the isolated central nervous system of the marine mollusc Tritonia diomedea, brief stimulation (1 sec) of a peripheral nerve activates an interneuronal central pattern generator that produces the long-lasting (approximately 30-60 sec) motor program underlying the animal's rhythmic escape swim. Here, we identify a single interneuron, DRI (for dorsal ramp interneuron), that (i) conveys the sensory information from this stimulus to the swim central pattern generator, (ii) elicits the swim motor program when driven with intracellular stimulation, and (iii) blocks the depolarizing "ramp" input to the central pattern generator, and consequently the motor program itself, when hyperpolarized during the nerve stimulus. Because most of the sensory information appears to be funneled through this one neuron as it enters the pattern generator, DRI presents a striking example of single neuron control over a complex motor circuit.

Dishabituation of the Tritonia escape swim.

When repeatedly elicited, the oscillatory escape swim of the marine mollusc Tritonia diomedea undergoes habituation of the number of cycles per swim. Although similar in most respects to habituation observed in vertebrates and other invertebrates, one key feature, dishabituation, has been surprisingly difficult to demonstrate. Here we evaluate the hypothesis that this is due to interference from short-term sensitization, which is manifested as a reduction in swim onset latency, that occurs simultaneously during habituation training. Robust dishabituation was obtained using a multisession habituation protocol designed to allow this sensitization to dissipate before the dishabituatory stimulus was applied. These results extend the similarity of habituation in Tritonia to that described in other species, strengthening the usefulness of this preparation as a model system for studies of the cellular basis of habituation.

Parallel processing of short-term memory for sensitization in Aplysia.

How is the short-term memory for a single form of learning distributed among the various elements of a neuronal circuit? To answer this question, we examined the short-term memory for sensitization, using the siphon component of the defensive gill- and siphon-withdrawal reflex. We found that the memory for short-term sensitization is represented by at least four sites of circuit modification, each involving a different type of plasticity. These include (1) presynaptic facilitation of the sensory neuron connections onto both interneurons and motorneurons; (2) presynaptic inhibition at the connections of the L30 inhibitory neurons onto the excitatory interneuron L29; (3) posttetanic potentiation of the excitatory connections made by L29 onto a specific subclass of siphon motorneurons, the LFS cells; and (4) an increase in the tonic firing rate of the LFS siphon motor neurons, resulting in neuromuscular facilitation. Each of the heterosynaptic changes seems to involve a common modulatory transmitter and to utilize a common second messenger system. Moreover, each of these sites seems capable of encoding a different component of the short-term memory. Facilitation of the connections of sensory neurons should contribute to the increase in amplitude of the response; the disinhibition of the L29 interneurons and the posttetanic potentiation at L29 synapses should contribute to an increase in the duration of the response; and the increase in tonic firing of the LFS subclass of siphon motor neurons seems capable of contributing both to an increase in response amplitude and to changes in response topography.

Sensitization of the Tritonia escape swim.

When repeatedly elicited, the oscillatory escape swim of the marine mollusc Tritonia diomedea undergoes habituation of the number of cycles per swim. Previous work has shown that this habituation is accompanied by sensitization of another feature of the behavior: latency to swim onset. Here we focused on the behavioral features of sensitization itself. Test swims elicited 5 min after a strong sensitizing head stimulus differed in several ways from control swims: sensitized animals had shorter latencies for gill and rhinophore withdrawal, a shorter latency for swim onset, a lower threshold for swim initiation, and an increased number of cycles per swim. Sensitized animals did not, however, swim any faster (no change in cycle period). A separate experiment found that swim onset latency also sensitized when Tritonia came into contact with one of their natural predators, the seastar Pycnopodia helianthoides, demonstrating the ecological relevance of this form of nonassociative learning. These results define the set of behavioral changes to be explained by cellular studies of sensitization in Tritonia.

Dynamic neuromodulation of synaptic strength intrinsic to a central pattern generator circuit.

Motor circuits are often thought to be physically separate from their neuromodulatory systems. We report here a counter example, where neurons within a circuit appear to modulate synaptic properties of that same circuit during its normal operation. The dorsal swim interneurons (DSIs) are members of the central pattern generator circuit for escape swimming in the mollusc Tritonia diomedea. However, DSI stimulation also rapidly enhances the synaptic potentials evoked by another neuron in the same circuit onto its follower cells. This modulatory action appears to be mediated by serotonin (5-hydroxytryptamine); the DSIs are serotonin-immunoreactive, and bath-application of serotonin mimics and occludes the effect of DSIs. These results indicate that during the escape swim, circuit connection strengths are dynamically controlled by the activity of neurons within the circuit itself. This 'intrinsic neuromodulation' may be important for the animal's initial decision to swim, the generation of the swim motor programme itself, and certain types of learning.

Prepulse inhibition of the Tritonia escape swim.

Presenting a weak stimulus just before a strong, startle stimulus reduces the amplitude of the ensuing startle response in humans and other vertebrates. This phenomenon, termed "prepulse inhibition" (PPI), appears to function to reduce distraction while processing sensory input. To date, no detailed neural mechanism has been described for PPI. Here we demonstrate PPI in the marine mollusk Tritonia diomedea, which has a nervous system highly suitable for cellular analyses. We found that a 100 msec vibrotactile prepulse prevented the animal's escape swim response to a closely following 1 sec tail shock. This inhibition was highly transient, with a significant effect lasting just 2.5 sec. These findings indicate that the Tritonia escape swim response undergoes a form of PPI phenomenologically similar to that observed in vertebrates. Further tests showed that the vibrotactile stimulus had no inhibitory effect if applied after tail shock, while the animal was preparing to swim, but it acted to terminate swims once they were actively under way. As a first step toward a cellular analysis of PPI, we recorded from neurons of the swim circuit in a semi-intact preparation and found that the vibrotactile stimulus used in the behavioral experiments also prevented the tail shock-elicited swim motor program. These results represent the first explicit demonstration of PPI in an invertebrate and establish Tritonia as a model system for analyzing its physiological basis.

Parametric features of habituation of swim cycle number in the marine mollusc tritonia diomedea.

When repeatedly elicited, the oscillatory escape swim of the marine mollusc Tritonia diomedea undergoes habituation of the number of cycles per swim. Because the neural circuit for this behavior is reasonably well understood, a cellular analyses of habituation in Tritonia is feasible. Since such a study must ultimately relate cellular correlates to behavioral modifications, we have sought to increase our understanding of the parametric features of cycle number habituation in Tritonia. Habituation was compared when using different intertrial intervals, repeated training sessions, and different stimulus locations. Stimulus site generalization of habituation was demonstrated, suggesting that at least one site of plasticity underlying habitation is located postsynaptic to the sensory neurons for the response. Dishabituation from an above-zero baseline response level was not obtained. An isolated brain preparation was tested as a potential simplified system for cellular studies of habituation mechanisms. Repeated stimulation of a nerve containing sensory afferent processes resulted in a progressive reduction of swim motor program cycle number, with a rate similar to that seen in the behavior. Together, these findings: (1) establish a set of parametric features of cycle numbers habituation to be explained by physiological studies; (2) suggest that at least one circuit modification underlying the habituation is located among the circuit interneurons; and (3) indicate that the isolated brain preparation may serve as a useful neural analogue for studies of the cellular mechanisms of cycle number habituation in Tritonia.

Monosynaptic connections made by the sensory neurons of the gill- and siphon-withdrawal reflex in Aplysia participate in the storage of long-term memory for sensitization.

We have found that in the gill- and siphon- withdrawal reflex of Aplysia, the memory for short-term sensitization grades smoothly into long-term memory with increased amounts of sensitization training. One cellular locus for the storage of the memory underlying short-term sensitization is the set of monosynaptic connections between the siphon sensory cells and the gill and siphon motor neurons. We have now also found that these same monosynaptic connections participate in the storage of the memory underlying long-term Sensitization. We examined the amplitudes of the direct synaptic connections produced by siphon sensory neurons on the gill motor neuron L7 in nervous systems removed from control and from long-term sensitized animals 1 day after the end of long-term sensitization training. The connections were significantly larger in long-term sensitized animals than in control animals. The finding that long-term memory occurs at the same synaptic locus as the short-term memory should facilitate study of the cellular and molecular mechanisms involved in the conversion of short-term to long-term memory.

Recovery of tail-elicited siphon-withdrawal reflex following unilateral axonal injury is associated with ipsi- and contralateral changes in gene expression in Aplysia californica.

Behavioral, cellular and molecular changes were examined following axonal injury in the marine mollusc Aplysia californica. Unilateral nerve injury was performed by crushing the pleural-pedal connective and the peripheral pedal nerves innervating one side of the posterior body wall and the tail. The injury procedure severs the axons of the pleural sensory neurons resulting in the blockade of the tail-elicited siphon-withdrawal reflex. Partial reflex recovery is observed within 3 d and reaches 50% of the pretest value by six weeks postinjury. Retrograde staining of injured nerves combined with electrophysiological recordings from siphon motor neurons show that axons can regenerate through the crushed site and reconnect with the tail by three weeks postinjury. Moreover, the behavioral and electrophysiological measurements suggest that the contralateral sensory neurons may contribute to the early recovery of the siphon-withdrawal reflex. The levels of mRNAs coding for actin and calreticulin are elevated while the mRNAs coding for intermediate filament protein, sensorin A, FMRFamide are reduced in the ipsilateral pleural ganglia as detected by Northern blots. In the contralateral pleural ganglia, the levels of mRNAs coding for actin, sensorin A and FMRFamide are elevated. These molecular changes in both the ipsi- and contralateral sides are consistent with the hypothesis that both sides are participating in the behavioral recovery following unilateral axonal injury.

Cell and molecular analysis of long-term sensitization in Aplysia.

We have found that one cellular locus for the storage of the memory underlying short-term sensitization of the gill and siphon withdrawal reflex in Aplysia is the set of monosynaptic connections between the siphon sensory cells and the gill and siphon motor neurons. These connections also participate in the storage of memory underlying long-term sensitization. In animals that have undergone long-term sensitization, the amplitudes of the monosynaptic connections are significantly larger (2.2x) than the ones in control animals. To study the mechanisms of onset and retention of long-term synaptic facilitation that underly long-term sensitization and the role of protein synthesis in long-term memory, we have developed two types of reduced preparations: the intact reflex isolated from the remainder of the animal, and a dissociated cell culture system in which the monosynaptic component (sensory neurons and motor neurons) of the neuronal circuit mediating the withdrawal reflex is reconstituted. We found that protein synthesis inhibitors, such as anisomycin or emetine, and RNA synthesis inhibitors, such as actinomycin D or alpha-amanitin, blocked long-term facilitation without interfering with short-term facilitation. These results suggest that the acquisition of long-term memory may require the expression of genes and the synthesis of proteins not needed for short-term memory.