Neuronal microcircuits for decision making in C. elegans.

The simplicity and genetic tractability of the nervous system of the nematode Caenorhabditis elegans make it an attractive system in which to seek biological mechanisms of decision making. Although work in this area remains at an early stage, four basic types paradigms of behavioral choice, a simple form of decision making, have now been demonstrated in C. elegans. A recent series of pioneering studies, combining genetics and molecular biology with new techniques such as microfluidics and calcium imaging in freely moving animals, has begun to elucidate the neuronal mechanisms underlying behavioral choice. The new research has focussed on choice behaviors in the context of habitat and resource localization, for which the neuronal circuit has been identified. Three main circuit motifs for behavioral choice have been identified. One motif is based mainly on changes in the strength of synaptic connections whereas the other two motifs are based on changes in the basal activity of an interneuron and the sensory neuron to which it is electrically coupled. Peptide signaling seems to play a prominent role in all three motifs, and it may be a general rule that concentrations of various peptides encode the internal states that influence behavioral decisions in C. elegans.

Artificial dirt: microfluidic substrates for nematode neurobiology and behavior.

With a nervous system of only 302 neurons, the free-living nematode Caenorhabditis elegans is a powerful experimental organism for neurobiology. However, the laboratory substrate commonly used in C. elegans research, a planar agarose surface, fails to reflect the complexity of this organism's natural environment, complicates stimulus delivery, and is incompatible with high-resolution optophysiology experiments. Here we present a new class of microfluidic devices for C. elegans neurobiology and behavior: agarose-free, micron-scale chambers and channels that allow the animals to crawl as they would on agarose. One such device mimics a moist soil matrix and facilitates rapid delivery of fluid-borne stimuli. A second device consists of sinusoidal channels that can be used to regulate the waveform and trajectory of crawling worms. Both devices are thin and transparent, rendering them compatible with high-resolution microscope objectives for neuronal imaging and optical recording. Together, the new devices are likely to accelerate studies of the neuronal basis of behavior in C. elegans.

Circuit motifs for spatial orientation behaviors identified by neural network optimization.

Spatial orientation behavior is universal among animals, but its neuronal basis is poorly understood. The main objective of the present study was to identify candidate patterns of neuronal connectivity (motifs) for two widely recognized classes of spatial orientation behaviors: hill climbing, in which the organism seeks the highest point in a spatial gradient, and goal seeking, in which the organism seeks an intermediate point in the gradient. Focusing on simple networks of graded processing neurons characteristic of Caenorhabditis elegans and other nematodes, we used an unbiased optimization algorithm to seek values of neuronal time constants, resting potentials, and synaptic strengths sufficient for each type of behavior. We found many different hill-climbing and goal-seeking networks that performed equally well in the two tasks. Surprisingly, however, each hill-climbing network represented one of just three fundamental circuit motifs, and each goal-seeking network comprised two of these motifs acting in concert. These motifs are likely to inform the search for the real circuits that underlie these behaviors in nematodes and other organisms.

The homeobox gene lim-6 is required for distinct chemosensory representations in C. elegans.

The ability to discriminate between different chemical stimuli is crucial for food detection, spatial orientation and other adaptive behaviours in animals. In the nematode Caenorhabditis elegans, spatial orientation in gradients of soluble chemoattractants (chemotaxis) is controlled mainly by a single pair of chemosensory neurons. These two neurons, ASEL and ASER, are left-right homologues in terms of the disposition of their somata and processes, morphology of specialized sensory endings, synaptic partners and expression profile of many genes. However, recent gene-expression studies have revealed unexpected asymmetries between ASEL and ASER. ASEL expresses the putative receptor guanylyl cyclase genes gcy-6 and gcy-7, whereas ASER expresses gcy-5 (ref. 4). In addition, only ASEL expresses the homeobox gene lim-6, an orthologue of the human LMX1 subfamily of homeobox genes. Here we show, using laser ablation of neurons and whole-cell patch-clamp electrophysiology, that the asymmetries between ASEL and ASER extend to the functional level. ASEL is primarily sensitive to sodium, whereas ASER is primarily sensitive to chloride and potassium. Furthermore, we find that lim-6 is required for this functional asymmetry and for the ability to distinguish sodium from chloride. Thus, a homeobox gene increases the representational capacity of the nervous system by establishing asymmetric functions in a bilaterally symmetrical neuron pair.

Pressure polishing: a method for re-shaping patch pipettes during fire polishing.

The resolution of patch-clamp recordings is limited by the geometrical and electrical properties of patch pipettes. The ideal whole-cell patch pipette has a blunt, cone-shaped tip and a low resistance. The best glasses for making patch pipettes are low noise, low capacitance glasses such as borosilicate and aluminasilicate glasses. Regrettably, nearly all borosilicate glasses form pipettes with sharp, cone-shaped tips and relatively high resistance. It is possible, however, to reshape the tip during fire polishing by pressurizing the pipette lumen during fire polishing, a technique we call 'pressure polishing.' We find that this technique works with pipettes made from virtually any type of glass, including thick-walled aluminasilicate glass. We routinely use this technique to make pipettes suitable for whole-cell patch-clamp recording of tiny neurons (1-3 microm in diameter). Our pipettes are made from thick-walled, borosilicate glass and have submicron tip openings and resistances <10 MOmega. Similar pipettes could be used to record from subcellular neuronal structures such as axons, dendrites and dendritic spines. Pressure polishing should also be useful in patch-clamp applications that benefit from using pipettes with blunt tips, such as perforated-patch whole-cell recordings, low-noise single channel recordings and experiments that require internal perfusion of the pipette.

The fundamental role of pirouettes in Caenorhabditis elegans chemotaxis.

To investigate the behavioral mechanism of chemotaxis in Caenorhabditis elegans, we recorded the instantaneous position, speed, and turning rate of single worms as a function of time during chemotaxis in gradients of the attractants ammonium chloride or biotin. Analysis of turning rate showed that each worm track could be divided into periods of smooth swimming (runs) and periods of frequent turning (pirouettes). The initiation of pirouettes was correlated with the rate of change of concentration (dC/dt) but not with absolute concentration. Pirouettes were most likely to occur when a worm was heading down the gradient (dC/dt < 0) and least likely to occur when a worm was heading up the gradient (dC/dt > 0). Further analysis revealed that the average direction of movement after a pirouette was up the gradient. These observations suggest that chemotaxis is produced by a series of pirouettes that reorient the animal to the gradient. We tested this idea by imposing the correlation between pirouettes and dC/dt on a stochastic point model of worm motion. The model exhibited chemotaxis behavior in a radial gradient and also in a novel planar gradient. Thus, the pirouette model of C. elegans chemotaxis is sufficient and general.

Computational rules for chemotaxis in the nematode C. elegans.

We derive a linear neural network model of the chemotaxis control circuit in the nematode Caenorhabditis elegans and demonstrate that this model is capable of producing nematodelike chemotaxis. By expanding the analytic solution for the network output in time-derivatives of the network input, we extract simple computational rules that reveal how the model network controls chemotaxis. Based on these rules we find that optimized linear networks typically control chemotaxis by computing the first time-derivative of the chemical concentration and modulating the body turning rate in response to this derivative. We argue that this is consistent with behavioral studies and a plausible mechanism for at least one component of chemotaxis in real nematodes.

Active currents regulate sensitivity and dynamic range in C. elegans neurons.

Little is known about the physiology of neurons in Caenorhabditis elegans. Using new techniques for in situ patch-clamp recording in C. elegans, we analyzed the electrical properties of an identified sensory neuron (ASER) across four developmental stages and 42 unidentified neurons at one stage. We find that ASER is nearly isopotential and fails to generate classical Na+ action potentials. Rather, ASER displays a high sensitivity to input currents coupled to a depolarization-dependent reduction in sensitivity that may endow ASER with a wide dynamic range. Voltage clamp revealed depolarization-activated K+ and Ca2+ currents that contribute to high sensitivity near the zero-current potential. The depolarization-dependent reduction in sensitivity can be attributed to activation of K+ current at voltages where it dominates the net membrane current. The voltage dependence of membrane current was similar in all neurons examined, suggesting that C. elegans neurons share a common mechanism of sensitivity and dynamic range.

Using reflexive behaviors of the medicinal leech to study information processing.

The interneuronal network that produces local bending in the leech is distributed, in the sense that most of the interneurons involved are activated in all forms of local bending, even those in which their outputs would produce inappropriate movements. Such networks have been found to control a number of different behaviors in a variety of animals. This article reviews three issues: the physiological and modeling observations that led to the conclusion that local bending in leeches is controlled by a distributed system; what distributed processing means for this and other behaviors; and why the leech interneuronal network may have evolved to be distributed in the first place.

The computational leech.

The local bending reflex of the leech computes a well-defined sensorimotor input-output function in which each of several unique patterns of sensory input elicits a unique pattern of motoneuron activity. Interneurons in the reflex respond to most input patterns and contribute to most motor patterns, suggesting a distributed processing mechanism for the reflex. This suggestion is supported by models in which connection strengths are adjusted by a neural network optimization algorithm to reproduce the local bending input-output function. In addition, computational parallels between the local bending network and the perceptron, a major class of artificial neural networks, brings the functional role of local bending interneurons into question and suggests new physiological experiments.

A lower bound on the detectability of nonassociative learning in the local bending reflex of the medicinal leech.

Studies of neural mechanisms of learning and memory have focused on large changes at identified synapses. However, memory in distributed processing reflexes could involve widely distributed engrams characterized by small changes at every synapse in the network. To investigate this possibility, we used a neural network optimization algorithm to construct distributed engrams for nonassociative conditioning in a model of the local bending reflex of the medicinal leech (Hirudo medicinalis). The model comprised 4 sensory neurons, 10 to 40 interneurons, 8 motor neurons, and up to 480 connections. Synaptic connections in the model were first optimized to reproduce the amplitude and time course of motor neuron synaptic potentials recorded during local bending. This network, which represented the naive state before conditioning, was then reoptimized to the habituated or sensitized state. Following reoptimization, the memory for nonassociative learning was encoded by small changes dispersed across the entire network, and each change made only a small contribution to the learning. Moreover, because the changes were small, resolution of a few tenths of a millivolt, or 3-5% of an average synaptic potential, would be needed to account for half of the nonassociative learning. These results show how difficult distributed engrams can be to detect and provide a likely lower bound on the detectability of nonassociative learning in this and related networks.

Distributed processing of sensory information in the leech. III. A dynamical neural network model of the local bending reflex.

The subpopulation of identified interneurons in the local bending reflex receive multiple inputs from dorsal and ventral mechanoreceptors and have outputs to dorsal and ventral motor neurons. Their connections suggest a distributed processing mechanism in which withdrawal from dorsal, ventral, or lateral stimuli is controlled by a single population of approximately 40 multifunctional interneurons, but it is unclear whether additional interneurons dedicated to particular inputs are needed to account for each kind of bend. We therefore asked whether a model could be constructed that reproduced all behaviors without dedicated interneurons. Interneurons in the model were constrained to receive both dorsal and ventral inputs. Connection strengths were adjusted by gradient descent optimization until the model reproduced the amplitude and time course of motor neuron synaptic potentials in intracellular recordings of the response to many different stimuli. After optimization, the similarity between model and identified interneurons showed that additional dedicated interneurons are not necessary to produce all forms of the behavior. Successful optimization of networks with many fewer interneurons showed that the 40-interneuron network is redundant, raising the possibility that the interneurons have additional functions. Finally, optimizing networks with additional constraints produced better matches to some of the identified interneurons and showed that local bending can be produced by two populations of interneurons: one with outputs consistent with dorsal bending, the other with ventral bending. This suggests a simple model in which two principal types of interneurons produce many different behaviors and predicts the type of interneuron that remains to be identified.

Reconstruction of action potential development from whole-cell currents of differentiating spinal neurons.

The duration and ionic dependence of action potentials are developmentally regulated. Voltage-clamp recordings of amphibian spinal neurons have revealed alterations in five currents. To determine whether the changes in the currents are sufficient to produce the change in action potential duration and ionic dependence, we constructed a Hodgkin-Huxley model of electrical excitability of these neurons. The model shows that the equations describing the voltage-clamped currents of young and mature neurons generate action potentials appropriate in duration and ionic dependence for each developmental stage. Moreover, the observed changes in the currents are quantitatively sufficient to produce the changes in the action potential. The effect of the change in each current is detectable in the model. However, the increase in amplitude of the delayed-rectifier potassium current has the largest effect. The model further shows that changes in action potential duration could be achieved with changes in kinetics rather than amplitude of this current, or with changes in amplitudes of other currents. Thus, although increase in amplitude of the delayed rectifier plays a pivotal role in the maturation of excitability, it is not uniquely positioned to govern the action potential duration.

Two forms of sensitization of the local bending reflex of the medicinal leech.

Sensitization of the local bending reflex of the medicinal leech Hirudo medicinalis was studied in a semi-intact preparation in which behavioral and electrophysiological recordings were made simultaneously. 1. Sensitization of local bending could be produced in two ways: by repeated stimulation of the mechanoreceptor sensitive to pressure (the P cell), and by stimulation of the mechanoreceptor sensitive to noxious stimuli (the N cell). 2. Both forms of sensitization produced a central neuronal change, measured as an increase in the number of stimulus-evoked action potentials in cell 3 (an excitor of dorsal longitudinal muscles). 3. Intracellular stimulation of serotonin-containing neurons 21 and 61 mimicked the sensitizing stimuli, but stimulation of the Retzius cell, which also contains serotonin, did not. 4. Stimulation of the Leydig cell, which releases octopamine, decreased the strength of local bending.

Distributed processing of sensory information in the leech. I. Input-output relations of the local bending reflex.

Sensory processing in the local bending reflex of the leech (Hirudo medicinalis) was studied by examining the input-output relations of the reflex. Sensory cells (P cells) were stimulated singly and in pairs and the responses of the longitudinal muscle motor neurons were recorded. Each pattern of single and paired P cell stimulation produced a unique pattern of motor neuron response. In general, motor neuron response patterns were consistent with the behavioral response of the whole animal.

Distributed processing of sensory information in the leech. II. Identification of interneurons contributing to the local bending reflex.

Isolated midbody ganglia of the leech Hirudo medicinalis were surveyed for interneurons contributing to the dorsal component of the local bending reflex, i.e., to the excitation of dorsal excitatory motor neurons that follows stimulation of dorsal mechanoreceptors responsive to pressure (P cells). Nine types of local bending interneuron could be distinguished on physiological and morphological grounds--8 paired and 1 unpaired cell per ganglion. Synaptic latencies from sensory neurons to interneurons were consistent with a direct or possibly disynaptic pathway. Connections between interneurons appeared to be rare and hyperpolarization of individual interneurons during local bending produced small but reliable decrements in motor neuron response, suggesting that multiple parallel pathways contribute to the behavior. Paradoxically, most interneurons received substantial inputs from ventral as well as dorsal mechanoreceptors, indicating that interneurons that were distinguished by their contribution to dorsal local bending were, in fact, active in ventral and lateral bends as well. The capacity to detect a particular stimulus and produce the appropriate response cannot be localized to particular types of interneuron; rather, it appears to be a distributed property of the entire local bending network.

Function of identified interneurons in the leech elucidated using neural networks trained by back-propagation.

Mechanical stimulation of the body surface of the leech causes a localized withdrawal from dorsal, ventral and lateral stimuli. The pathways from sensory to motor neurons in the reflex include at least one interneuron. We have identified a subset of interneurons contributing to the reflex by intracellular recording, and our analysis of interneuron input and output connections suggests a network in which most interneurons respond to more than one sensory input, most have effects on all motor neurons and in which each form of the behaviour is produced by appropriate and inappropriate effects of many interneurons. To determine whether interneurons of this type can account for the behaviour, or whether additional types are required, model networks were trained by back-propagation to reproduce the physiologically determined input-output function of the reflex. Quantitative comparisons of model and actual connection strengths show that model interneurons are similar to real ones. Consequently, the identified subset of interneurons could control local bending as part of a distributed processing network in which each form of the behaviour is produced by the appropriate and inappropriate effects of many interneurons.

Habituation of the shortening reflex in the medicinal leech.

The shortening reflex was elicited in food-sated Hirudo medicinalis by light flashes delivered at 20-s intervals over a 40-trial session. The reflex was enhanced if animals were stored at 5-7 degrees C rather than 20 degrees C. Short-term, that is, within-sessions, habituation was readily observed. Dishabituation could be produced by a single electric shock at Trial 30. However, the shock also enhanced responding when delivered before Trial 1 (sensitization). No long-term, that is, between-sessions, habituation occurred over 6 consecutive days of testing; on the contrary, responding gradually increased.