Central-complex control of movement in the freely walking cockroach.

To navigate in the world, an animal's brain must produce commands to move, change direction, and negotiate obstacles. In the insect brain, the central complex integrates multiple forms of sensory information and guides locomotion during behaviors such as foraging, climbing over barriers, and navigating to memorized locations. These roles suggest that the central complex influences motor commands, directing the appropriate movement within the current context. Such commands are ultimately carried out by the limbs and must therefore interact with pattern generators and reflex circuits that coordinate them. Recent studies have described how neurons of the central complex encode sensory information: neurons subdivide the space around the animal, encoding the direction or orientation of stimuli used in navigation. Does a similar central-complex code directing movement exist, and if so, how does it effect changes in the control of limbs? Recording from central-complex neurons in freely walking cockroaches (Blaberus discoidalis), we identified classes of movement-predictive cells selective for slow or fast forward walking, left or right turns, or combinations of forward and turning speeds. Stimulation through recording wires produced consistent trajectories of forward walking or turning in these animals, and those that elicited turns also altered an inter-joint reflex to a pattern resembling spontaneous turning. When an animal transitioned to climbing over an obstacle, the encoding of movement in this new context changed for a subset of cells. These results indicate that encoding of movement in the central complex participates in motor control by a distributed, flexible code targeting limb reflex circuits.

Responses of Drosophila giant descending neurons to visual and mechanical stimuli.

In Drosophila, the paired giant descending neurons (GDNs), also known as giant fibers, and the paired giant antennal mechanosensory descending neurons (GAMDNs), are supplied by visual and mechanosensory inputs. Both neurons have the largest cell bodies in the brain and both supply slender axons to the neck connective. The GDN axon thereafter widens to become the largest axon in the thoracic ganglia, supplying information to leg extensor and wing depressor muscles. The GAMDN axon remains slender, interacting with other descending neuron axons medially. GDN and GAMDN dendrites are partitioned to receive inputs from antennal mechanosensory afferents and inputs from the optic lobes. Although GDN anatomy has been well studied in Musca domestica, less is known about the Drosophila homolog, including electrophysiological responses to sensory stimuli. Here we provide detailed anatomical comparisons of the GDN and the GAMDN, characterizing their sensory inputs. The GDN showed responses to light-on and light-off stimuli, expanding stimuli that result in luminance decrease, mechanical stimulation of the antennae, and combined mechanical and visual stimulation. We show that ensembles of lobula columnar neurons (type Col A) and mechanosensory antennal afferents are likely responsible for these responses. The reluctance of the GDN to spike in response to stimulation confirms observations of the Musca GDN. That this reluctance may be a unique property of the GDN is suggested by comparisons with the GAMDN, in which action potentials are readily elicited by mechanical and visual stimuli. The results are discussed in the context of descending pathways involved in multimodal integration and escape responses.

Optic glomeruli and their inputs in Drosophila share an organizational ground pattern with the antennal lobes.

Studying the insect visual system provides important data on the basic neural mechanisms underlying visual processing. As in vertebrates, the first step in visual processing in insects is through a series of retinotopic neurons. Recent studies on flies have found that these converge onto assemblies of columnar neurons in the lobula, the axons of which segregate to project to discrete optic glomeruli in the lateral protocerebrum. This arrangement is much like the fly's olfactory system, in which afferents target uniquely identifiable olfactory glomeruli. Here, whole-cell patch recordings show that even though visual primitives are unreliably encoded by single lobula output neurons because of high synaptic noise, they are reliably encoded by the ensemble of outputs. At a glomerulus, local interneurons reliably code visual primitives, as do projection neurons conveying information centrally from the glomerulus. These observations demonstrate that in Drosophila, as in other dipterans, optic glomeruli are involved in further reconstructing the fly's visual world. Optic glomeruli and antennal lobe glomeruli share the same ancestral anatomical and functional ground pattern, enabling reliable responses to be extracted from converging sensory inputs.

Interaction between descending input and thoracic reflexes for joint coordination in cockroach. II comparative studies on tethered turning and searching.

Tethered cockroaches turn from unilateral antennal contact using asymmetrical movements of mesothoracic (T2) legs (Mu and Ritzmannin J Comp Physiol A 191:1037-1054, 2005). During the turn, the leg on the inside of the turn (the inside T2 leg) has distinctly different motor patterns from those in straight walking. The transformation from walking to inside leg turning could be triggered by descending commands that alter a few critical reflexes that start a cascade of physical changes in leg movement or posture, leading to further alterations. This hypothesis has two implications: First, the descending activities must be able to influence thoracic reflexes. Second, one should be able to initiate the turning motor pattern in the absence of descending signals by mimicking a point farther down in the reflex cascade. We addressed the first implication in the companion paper. To examine the second implication, we compared kinematics and motor activities of the T2 leg during searching with that of inside leg turning. The reaching movements made during searching were found to be similar to the movements made by the inside leg during turning. Moreover, even after disconnecting the brain from the thoracic ganglia the reaching movements were similar. This observation is consistent with the second implication from the hypothesis.

Interaction between descending input and thoracic reflexes for joint coordination in cockroach: I. descending influence on thoracic sensory reflexes.

Tethered cockroaches turn from unilateral antennal contact using asymmetrical movements of mesothoracic (T2) legs (Mu and Ritzmann in J Comp Physiol A 191:1037-1054, 2005). During the turn, the leg on the inside of the turn (the inside T2 leg) has distinctly different motor patterns from those in straight walking. One possible neural mechanism for the transformation from walking to inside leg turning could be that the descending commands alter a few critical reflexes that start a cascade of physical changes in leg movement or posture, leading to further alterations. This hypothesis has two implications: first, the descending activities must be able to influence thoracic reflexes. Second, one should be able to initiate the turning motor pattern without descending signals by mimicking a point farther down in the reflex cascade. We addressed the first implication in this paper by experiments on chordotonal organ reflexes. The activity of depressor muscle (Ds) and slow extensor tibia muscle (SETi) was excited and inhibited by stretching and relaxing the femoral chordotonal organ. However, the Ds responses were altered after eliminating the descending activity, while the SETi responses remain similar. The inhibition to Ds activity by stretching the coxal chordotonal organ was also altered after eliminating the descending activity.

Kinematics and motor activity during tethered walking and turning in the cockroach, Blaberus discoidalis.

When insects turn from walking straight, their legs have to follow different motor patterns. In order to examine such pattern change precisely, we stimulated single antenna of an insect, thereby initiating its turning behavior, tethered over a lightly oiled glass plate. The resulting behavior included asymmetrical movements of prothoracic and mesothoracic legs. The mesothoracic leg on the inside of the turn (in the apparent direction of turning) extended the coxa-trochanter and femur-tibia joints during swing rather than during stance as in walking, while the outside mesothoracic leg kept a slow walking pattern. Electromyograms in mesothoracic legs revealed consistent changes in the motor neuron activity controlling extension of the coxa-trochanter and femur-tibia joints. In tethered walking, depressor trochanter activity consistently preceded slow extensor tibia activity. This pattern was reversed in the inside mesothoracic leg during turning. Also for turning, extensor and depressor motor neurons of the inside legs were activated in swing phase instead of stance. Turning was also examined in free ranging animals. Although more variable, some trials resembled the pattern generated by tethered animals. The distinct inter-joint and inter-leg coordination between tethered turning and walking, therefore, provides a good model to further study the neural control of changing locomotion patterns.