Critical parameters of the spike trains in a cell assembly: coding of turn direction by the giant interneurons of the cockroach

Cockroaches (Periplaneta americana) respond to air displacement produced by an approaching predator by turning and running away. A set of 4 bilateral pairs of ventral giant interneurons is important in determining turn direction. Wind from a given side is known to produce more spikes, an earlier onset of the spike trains, and different fine temporal patterning, in the ipsilateral vs the contralateral set of these interneurons. Here we investigate which of these spike train parameters the cockroach actually uses to determine the direction it will turn.We delivered controlled wind puffs from the right front, together with intracellular injection of spike trains in a left ventral giant interneuron, under conditions where the animal could make normally directed turning movements of the legs and body. In trials where our stimuli caused the left side to give both the first spike and more total spikes than the right, but where our injected spike train included none of the normal fine temporal patterning, 92% of the evoked turns were to the rightopposite of normal (Figs. 4–6). In trials where the left side gave the first spike, but the right side gave more spikes, 100% of the turns were to the left-the normal direction (Figs. 8, 9). Comparable results were obtained when each of the left giant interneurons 1, 2 or 3 were electrically stimulated, and when either weak or stronger wind puffs were used. Stimulating a left giant interneuron electrically in the absence of a wind puff evoked an escape-like turn on 9% of the trials, and these were all to the right (Fig. 9).These results indicate that fine temporal patterning in the spike trains is not necessary, and information about which side gives the first spike is not sufficient, to determine turn direction. Rather, the key parameter appears to be relative numbers of action potentials in the left vs the right group of cells. These conclusions were supported by similar experiments in which extracellular stimulation of several left giant interneurons was paired with right wind (Figs. 11, 12).Abbreviations GI giant interneuron - vGI ventral giant interneuron - dGI dorsal giant interneuron - LY Lucifer yellow - CF carboxyfluorescein     

Distributing coordinated motor outputs to several body segments: escape movements in the cockroach

In the escape behavior of the cockroach, all six legs begin to make directed movements nearly simultaneously. The sensory stimulus that evokes these leg movements is a wind puff. Posterior wind receptors excite giant interneurons that carry a multi-cellular code for stimulus direction — and thus for turn direction-to the three thoracic ganglia, which innervate the three pairs of legs. We have attemptd to discriminate among various possible ways that the directional information in the giant interneurons could be distributed to each leg's motor circuit. Do the giant interneurons, for instance, inform separately each thoracic ganglion of wind direction? Or is there one readout system that conveys this information to all three ganglia, and if so, might the identified thoracic interneurons, which are postsynaptic to the giant interneurons, subserve this function? We made mid-sagittal lesions in one or two thoracic ganglia, thus severing the initial segments of all the known thoracic interneurons in these ganglia, and thus causing their projection axons to the other thoracic ganglia to degenerate. This lesion did not sever the giant interneurons, however (Fig. 5). Following such lesions, the legs innervated by the intact thoracic ganglia made normally directed leg movements (Figs. 4, 6, 7). Thus, the projection axons of the thoracic interneurons are not necessary for normal leg movements. Rather, the giant interneurons appear to specify to each thoracic ganglion in which direction to move the pair of legs it innervates.     

Indirect synaptic inputs from filiform hair sensory neurons contribute to the receptive fields of giant interneurons in the first-instar cockroach

The first-instar cockroach, Periplaneta americana, detects air movements using four filiform hair sensilla, which make synaptic connections to seven pairs of giant interneurons (GIs) in the terminal abdominal ganglion. The directional sensitivities of some of the GIs, predicted from their patterns of monosynaptic inputs, may not be the same as in the second instar or adult. Intracellular recordings were made to determine the contribution of polysynaptic inputs to the receptive fields of first-instar GIs. The ventral GI1, and the dorsal GI5, GI6, and GI7 were all found to have indirect synaptic inputs from filiform afferents. The indirect inputs were excitatory to GI1, GI5, and GI7, and inhibitory to GI6 and GI7. The indirect excitatory input to GI1 was predicted to alter qualitatively its receptive field, allowing it to respond to wind from the side of the animal, as in the adult. Inhibition was predicted to sharpen the receptive fields of GI6 and GI7. The inhibitory postsynaptic potentials reversed 6–8 mV below resting potential and were blocked by picrotoxin, indicating that they are GABAergic. Indirect excitation also altered the predicted receptive field of GI7, one of the inputs being an unusual “off-response” to movement of a filiform hair in its inhibitory direction. Accepted: 19 June 1998     

Modulation of flight by the giant interneurons of the cockroach

In this paper, I have examined the behavioral functions of feedback loops between the cockroach (Periplaneta americana) giant interneurons (GIs) and the flight thoracic rhythm generator.

Synaptic specificity in the first instar cockroach: patterns of monosynaptic input from filiform hair afferents to giant interneurons

Summary Direct evidence for monosynaptic connections between filiform hair sensory axons and giant interneurons (GIs) in the first instar cockroach, Periplaneta americana, was obtained using intracellular recording and HRP injection followed by electron microscopy. GIs 1–6 all receive monosynaptic input from at least one filiform afferent axon. GI1, GI2 and GI5 receive input only from the medial (M) axon, while GI3, GI4 and GI6 receive input from both M and lateral (L) axons. The dendrites of GI3 and GI6 which are contralateral to the cell bodies receive input from both axons whereas the smaller ipsilateral dendritic fields have synapses only from the L axon. GI5 has M axon input only onto its contralateral dendrites. In 50% of preparations GI7 receives weak input from the ipsilateral L axon. There is no obvious relationship between the morphology of the giant interneurons and the pattern of input they receive from the filiform afferents.Abbreviations GI giant interneuron - HRP horseradish peroxidase - L lateral axon - M medial axon     

Biogenic amines modulate synaptic transmission between identified giant interneurons and thoracic interneurons in the escape system of the cockroach

In the escape system of the cockroach, Periplaneta americana, a population of uniquely identifiable throacic interneurons (type A or TI_As) receive information about wind via chemical synapses from a population of ventral giant interneurons (vGIs). The TI_As are involved in the integration of sensory information necessary for orienting the animal during escape. It is likely that there are times in an animal's life when it is advantageous to modify the effectiveness of synaptic transmission between the vGIs and the TI_As. Given the central position of the TI_As inthe escape system, this would greatly alter associated motor outputs. We tested the ability of octopamine, serotonin, and dopamine to modulate synaptic transmission between vGIs and TI_As. Both octopamine and dopamine significantly increased the amplitude of vGI-evoked excitatory postsynaptic potentials (EPSPs) in TI_As at 10^?4?10^?2 M, and 10^?3 M, respectively. On the other hand, serotonin significantly decreased the vGI-evoked EPSPs in TI_As at 10^?4?10^?3 M. These results indicate that octopamine, serotonin, and dopamine are capable of modulating the efficacy of transmission of important neural connections within this circuit. © 1992 John Wiley & Sons, Inc.     

Postembryonic changes in the response properties of wind-sensitive giant interneurons in cricket

The intensity-response (I-R) relations for four wind-sensitive giant interneurons (GIs 8-1, 9-1, 9-2 and 9-3) in the fourth-, sixth- and last-instar nymphs of the cricket, Gryllus bimaculatus, were investigated using a unidirectional air current stimulus in order to explore the functional changes of GIs during postembryonic development. Contrary to our expectations, the response properties of GIs in nymphs were largely different from those in adults. The response magnitude of GI 8-1 in an intact cricket decreased during development, i.e. the GI in younger insects showed a larger response magnitude. Although the response magnitudes of GIs 9-1 and 9-2 were almost identical during the nymphal period, a significant decrease was observed after the imaginal ecdysis. During the nymphal period, the response magnitude of GI 9-3 increased according to the developmental stage. However, it decreased significantly after the imaginal ecdysis. We also investigated the response magnitudes of the GIs in nymphs after unilateral cercal ablation. From the results of ablation experiments, the changes in excitatory and/or inhibitory connections between filiform hairs and each GI during postembryonic development were revealed.     

Neural circuitry underlying linear representation of wind information in a nonspiking local interneuron of the cockroach

In the cercal system of the cockroach Periplaneta americana, primary sensory interneurons exhibiting a sharp directional sensitivity respond to wind in a linear manner whereas those exhibiting an omnidirectional sensitivity respond nonlinearly. For example, the wind-evoked response in an identifiable, nonspiking local interneuron, 101, which responds preferentially to wind from the left versus the right, is characterized exclusively by a differential first-order (linear) kernel. However, the slow potential response in a cercal giant interneuron, GI-1, is omnidirectional, and characterized by a second-order (nonlinear) kernel with an elongated depolarizing peak on the diagonal with two off-diagonal valleys. We here examined the neural circuitry underlying the linear and nonlinear representations of wind information by the deprivation of inputs from particular sets of cercal hair afferents. Electrical stimulation of the ipsilateral (related to the soma) cercal nerve elicited a depolarizing potential in 101, which was followed by delayed hyperpolarization. A continuous flow of 10^–4 M picrotoxin, which selectively blocked this delayed hyperpolarization, resulted in a significant change in the 101 response from linear to nonlinear. Because no frequency-doubling response was observed, the nonlinearity is due to signal compression (or rectification) that reflects the mechanical property of cercal afferents. This is consistent with the hypothesis that the linear representation in 101 is based on a subtraction process between two subsets of particular column hairs, whose best optimal directions are opposite to each other.Abbreviations GABA -aminobutyric acid - GI(s) giant interneuron(s) - GI-1, GI-2, GI-3, GI-4 giant interneuron 1,2,3,4 - ipsi ipsilateral - cont contralateral - MSE(s) mean square error(s)     

A model of antennal wall-following and escape in the cockroach

Cockroaches exploit tactile cues from their antennae to avoid predators. During escape running the same sensors are used to follow walls. We hypothesise that selection of these mutually exclusive behaviours can be explained without representation of the stimulus or an explicit switching mechanism. A neural model is presented that embodies this hypothesis. The model incorporates behavioural and neurophysiological data and is embedded in a mobile robot in order to test the response to stimuli in the real world. The system is shown to account for data on escape direction and high-speed wall-following in the cockroach, including the counter-intuitive observation that faster running cockroaches maintain a closer distance to the wall. The wall-following behaviour is extended to include discrimination of tactile escape cues according to behavioural context. We conclude by highlighting questions arising from the robot experiments that suggest interesting hypotheses to test in the cockroach.     

A gradient of synaptic efficacy and its presynaptic basis in the cercal system of the cockroach

1. The cerci of the cockroach Periplaneta americana bear longitudinal columns of wind-sensitive receptors which provide excitatory inputs to the giant interneurons (GIs) of the abdominal nerve cord. By using sound stimuli, we showed that spikes were more easily induced in the GIs from the most proximal than from the most distal receptors of the same column. 2. This was not due to a greater responsiveness of proximal sensilla to tones but to stronger synaptic connections; for the 3 largest GIs, the amplitude of the monosynaptic unitary EPSP tended to be all the higher as the stimulated sensillum was more proximal in each column. 3. The differences in EPSP size were due, at least partly, to presynaptic factors: a statistical analysis of the amplitude fluctuations of single-fibre EPSPs, showed that the amount of transmitter released per presynaptic impulse was larger for proximal than for distal sensory neurons in each column. 4. These differences in synaptic strength were correlated with differences in the structure of the afferent terminals. The location, the size and the shape of the axonal arbors are nearly the same for all sensory neurons of the same column, but proximal neurons arborize more profusely, and the terminal arbor of distal neurons is generally characterized by dorsal clusters of varicosities. 5. During postembryonic development, a decrease in the connection strength of 2 identified cercal neurons was accompanied by a retraction of ramifications on the medial side of their axonal arbor. 6. Possible mechanisms involved in the genesis and the remodelling of the gradient of synaptic strength are discussed in the light of available data and hypotheses relative to the development of ordered afferent connections.     

Monosynaptic excitation of lateral giant fibres by proprioceptive afferents in the crayfish

Giant interneurones mediate a characteristic `tail flip' escape response of the crayfish, Procambarus clarkii, which move it rapidly away from the source of stimulation. We have analysed the synaptic connections of proprioceptive sensory neurones with one type of giant interneurone, the lateral giant. Spikes in sensory neurones innervating an exopodite-endopodite chordotonal organ in the tailfan, which monitors the position and movements of the exopodite, are followed at a short and constant latency by excitatory postsynaptic potentials in a lateral giant interneurone (LG) recorded in the terminal abdominal ganglion. These potentials are unaffected by manipulation of the membrane potential of LG, by bath application of saline with a low calcium concentration, or by one containing the nicotinic antagonist, curare. The potentials evoked in LG by chordotonal organ stimulation are thus thought to be monosynaptic and electrically mediated. This is the first demonstration that LG receives input from sensory receptors other than exteroceptors in the terminal abdominal ganglion. Accepted: 7 April 1997     

Activation of DUM cell interneurons by ventral giant interneurons in the cockroach, Periplaneta americana

Dorsal unpaired median (DUM) cells in orthopteran insects are known to contain the neuromodulatory substance octopamine, and DUM cells with peripheral axons augment synaptic activity at neuromuscular junctions. One of the most studied systems in the cockroach is the giant interneuron (GI) system which controls the initial movements of a wind-mediated escape response. Our data demonstrate that DUM cells that are restricted to the central nervous system (DUM interneurons) receive inputs from ventral giant interneurons (vGIs) but not from dorsal giant interneurons (dGIs). In contrast, DUM cells that have peripheral axons consistently fail to be excited by any giant interneurons. The DUM interneurons are excited by vGIs on both sides of the CNS and, when the vGIs are excited in pairs, summation occurs. Wind fields that have been generated for two of the DUM interneurons are omnidirectional. These data, taken along with the known association of DUM cells with the neuromodulatory substance octopamine, suggest that the DUM interneurons may act to modulate central synapses.     

White noise analysis of graded response in a wind-sensitive,nonspiking interneuron of the cockroach

1. A novel approach using a Gaussian white noise as stimulus is described which allowed quantitative analysis of neuronal responses in the cercal system of the cockroach, Periplaneta americana. Cerci were stimulated by air displacement which was modulated by a sinusoidal and a white noise signal. During the stimulation, intracellular recordings were made from a uniquely identifiable, nonspiking, local interneuron which locates within the terminal abdominal ganglion. The white noise stimulation was cross-correlated with the evoked response to compute first- and second-order kernels that could define the cell's response dynamics. 2. The interneuron, cell 101, has an exceptionally large transverse neurite that connects two asymmetrical dendritic arborizations located on both sides of the ganglion. 3. The first-order Wiener kernels in cell 101 were biphasic (differentiating). The waveforms of the kernels produced by the ipsilateral and contralateral stimulations were roughly mirror images of each other: the kernels produced by wind stimuli on the side ipsilateral to the cell body of the interneuron are initially depolarized and then hyperpolarized, whereas those on the other side are initially hyperpolarized. The polarity reversal occurred along the midline of the animal's body, and no well-defined kernel was produced by a stimulus directed head on or from the tail. 4. Mean square error (MSE) between the actual response and the model prediction suggests that the linear component in cell 101 comprises half of the cell's total response (MSEs for the linear models were about 50% at preferred directions), whereas the second-order, non-linear component is insignificant. The linear component of the wind-evoked response was bandpass with the preferred frequency of 70-90 Hz. 5. Accounting for a noise, we reasonably assumed that at high frequencies the graded response in cell 101 is linearly related to a modulation of the air displacement and sensitive to the rate of change of the signal (i.e., wind velocity) and the direction of its source. It is suggested that the dynamics of the first-order kernel simply reflect the dynamics of sensory receptors that respond linearly to wind stimulation.     

A circadian rhythm in neural activity can be recorded from the central nervous system of the cockroach

Summary Evidence presented in this paper indicates that a robust circadian rhythm in the frequency of neural activity can be recorded from the central nervous system of intact cockroaches, Leucophaea maderae. This rhythmicity was abolished by optic lobe removal. Spontaneous neural activity was then used as an assay to demonstrate that the optic lobe is able to generate circadian oscillations in vitro. These results provide direct evidence that the cockroach optic lobe is a self-sustained circadian oscillator capable of generating daily rhythms in the absence of neural or hormonal communications with the rest of the organism.Abbreviations CNS central nervous system - DD constant dark - LD light/dark cycle - SCN suprachiasmatic nucleus - ZT Zeitgeber time     

Electrophysiology and anatomy of medulla interneurons in the optic lobe of the cockroach,Periplaneta americana

1. Medulla interneurons of the optic lobe of P. americana were studied to determine their spectral properties. These neurons exhibited tonic firing which changed with monochromatic broadfield illumination of the ipsilateral eye. The response patterns of these neurons were analyzed by inferring their relation to the ultraviolet (UV) and green (G) photoreceptor groups of the eye. Their anatomy was described after injection of Lucifer yellow. 2. Broadband neurons received either excitatory or inhibitory input from both UV and G receptors. These neurons were not strictly sensitive to luminosity levels and had large cell bodies in the central rind of the medulla and wide dendritic arbors in the medulla neuropil. 3. Narrow band neurons received input from predominantly one receptor type. Their spectral sensitivity curves were more finely tuned than those of the primary receptors presumably due to neural interactions within the optic lobe. 4. Color opponent neurons were inhibited by UV and excited by G inputs in their sustained response. Under certain conditions, some of these neurons also showed G inhibition. These neurons suggested the presence of a subsystem involved in color vision. 5. Broadband, narrow band and color opponent properties were seen in some single neurons when tested over a 5-6 log unit range of intensity. The responses of some of these neurons changed when stimulus duration was increased. These findings indicated that functional classification for these neurons was dependent on stimulus intensity and duration. 6. Polarizational sensitivity was tested in preliminary experiments. Two neurons responded to the movement and direction of polarized light.     

Morphology and physiology of peripheral giant interneurons in the forelegs (whips) of the whip spider Heterophrynus elaphus Pocock (Arachnida: Amblypygi)

Summary The tarsi of the modified front legs (whips) of the whip spider Heterophrynus elaphus contain two afferent giant fibers, GN1 and GN2, with diameters at the tibia-tarsus joint of ca. 21 m and 14 m, respectively. The somata of these two neurons lie in the periphery, about 25 cm away from the CNS. These two neurons are interneurons which receive mechanoreceptive inputs from approximately 750 and 1500 bristles, respectively. The receptive fields of GN1 and GN2 overlap; they extend for 40 mm (GN1) and 90 mm (GN2) along the length of the tarsus. About 90% of the synapses onto the giant fibers are axo-axonic. Mechanical stimulation of a single bristle is sufficient to elicit action potentials in one or both interneurons. The response of the interneurons adapts quickly. Average conduction time from the soma to the CNS is 45 ms for GN1 and 55 ms for GN2. Mean conduction velocities are 5.5 and 4.2 m/s, respectively. Activity in the giant fibers does not elicit a motor response; hence the giant fibers do not mediate an escape response. Possible functions of these giant fibers are discussed and compared to those of giant fiber systems in other arthropods.Abbreviations GN giant neuron - S segment     

Activity modulation in cockroach sensillum: the role of octopamine

The plasticity of sensory perception is provided partially by modulation of receptor cells. The electrical activity of American cockroach chemoreceptor cells in response to sex pheromone was measured under the influence of octopamine treatment and tracheal anoxia. Both experimental procedures caused decreased electroantennograms but affected spike activity differently: octopamine treatment increased firing rate, whereas anoxia decreased it. Spike frequency under octopamine treatment was elevated in response to pheromone stimulation and at background activity. Experiments with perfusion of isolated antennae showed a direct effect of octopamine on spike activity of pheromone sensilla, and excluded the possibility of indirect effects via octopamine-dependent release of other biologically active substances. The suggested mechanism of octopamine action is receptor cell membrane depolarization.     

Neuromodulation of flight initiation by octopamine in the cockroach Periplaneta americana

We have tested the effect of a known insect neuromodulator, octopamine, on flight initiation in the cockroach. Using minimally dissected animals, we found that octopamine lowered the threshold for windevoked initiation of flight when applied to either of two major synaptic sites in the flight circuitry: 1) the last abdominal ganglion, where wind-sensitive neurons from the cerci excite dorsal giant interneurons, or 2) the metathoracic ganglion, where the dorsal giant interneurons activate interneurons and motoneurons which are involved in producing the rhythmic flight motor pattern in the flight muscles (Fig. 2).Correlated with this change in flight initiation threshold, we found that octopamine applied to the last abdominal ganglion increased the number of action potentials produced by individual dorsal giant interneurons when recruiting the cereal wind-sensitive neurons with wind puffs (Figs. 3, 4, 5) or with extracellular stimulation of their axons (Fig. 6). Octopamine increases the excitability of the giant interneurons (Figs. 7, 8). Also, when we stimulated individual dorsal giant interneurons intracellularly, the number of action potentials needed to initiate flight was reduced when octopamine was applied to the metathoracic ganglion (Fig. 9).Abbreviations EMG electromyogram - dGIs dorsal giant interneurons - GI giant interneuron - A6 sixth abdominal ganglion - T3 third thoracic ganglion - EPSP excitatory postsynaptic potential     

Synchronous firing by specific pairs of cercal giant interneurons in crickets encodes wind direction

One prominent stimulus to evoke an escape response in crickets is the detection of air movement, such as would result from an attacking predator. Wind is detected by the cercal sensory system that consists of hundreds of sensory cells at the base of filiform hairs. These sensory cells relay information to about a dozen cercal giant and non-giant interneurons. The response of cercal sensory cells depends both, on the intensity and the direction of the wind. Spike trains of cercal giant interneurons then convey the information about wind direction and intensity to the central nervous system. Extracellular recording of multiple cercal giant interneurons shows that certain interneuron pairs fire synchronously if a wind comes from a particular direction. We demonstrate here that directional tuning curves of synchronously firing pairs of interneurons are sharper than those of single interneurons. Moreover, the sum total of all synchronously firing pairs eventually covers all wind directions. The sharpness of the tuning curves in synchronously firing pairs results from excitatory and inhibitory input from the cercal sensory neurons. Our results suggest, that synchronous firing of specific pairs of cercal giant interneurons encodes the wind direction. This was further supported by behavioral analyses.