Two types of rhythmic foregut movements are described in fifth instar larvae of the moth, Manduca sexta. These consist of posteriorly-directed waves of peristalsis which move food toward the midgut, and synchronous constrictions of the esophageal region, which appear to retain food within the crop. We describe these movements and the muscles of the foregut that generate them.The firing patterns of a subset of these muscles, including a constrictor and dilator pair from both the esophageal and buccal regions of the foregut, are described for both types of foregut movement.The motor patterns for the foregut muscles require innervation by the frontal ganglion (FG), which lies anterior to the brain and contains about 35 neurons. Eliminating the ventral nerve cord, leaving the brain and FG intact, did not affect the muscle firing patterns in most cases. Eliminating both the brain and the ventral nerve cord, leaving only the FG to innervate the foregut, generally resulted in an increased period for both gut movements and muscle bursts. This manipulation also produced increases in burst durations for most muscles, and had variable effects on the phasing of muscle activity. Despite these changes, the foregut muscles still maintained a rhythmic firing pattern when innervated by the FG alone.Two nerves exit the FG to innervate the foregut musculature: the anteriorly-projecting frontal nerve, and the posteriorly-directed recurrent nerve. Cutting the frontal nerve immediately and irreversibly stopped all muscle activity in the buccal region, while cutting the recurrent nerve immediately stopped all muscle activity in the pharyngeal and esophageal regions. Recordings from the cut nerves leaving the FG showed that the ganglion was Abbreviations: BC, Buccal Constrictor; BC 1, buccal constrictor motoneuron 1; BC2, buccal constrictor motoneuron 2; BD, Buccal Dilator; BD1, buccal dilator motoneuron 1; EC, Esophageal Dilator; EC1, esophageal dilator motoneuron l; EC2, esophageal dilator motoneuron 2; EC3, esophageal dilator motoneuron 3; ejp, excitatory j unction potential; FG, frontal ganglion; psp, postsynaptic potential Correspondence to: C.I. Miles spontaneously active, with rhythmic activity continuing within the nerves. These observations indicate that all of the foregut muscle motoneurons are located within the FG, and the FG in isolation produces a rhythmic firing pattern in the motoneurons. We have identified several motoneurons within the FG, by cobalt backfills and/or simultaneous intracellular recordings and fills from putative motoneurons and their muscles.

It is generally believed that animals make decisions about the selection of mates, kin or food on the basis of pre-constructed recognition templates. These templates can be innate or acquired through experience. An example of an acquired template is the feeding preference exhibited by larvae of the moth, Manduca sexta. Naive hatchlings will feed and grow successfully on many different plants or artificial diets, but once they have fed on a natural host they become specialist feeders. Here we show that the induced feeding preference of M. sexta involves the formation of a template to a steroidal glycoside, indioside D, that is present in solanaceous foliage. This compound is both necessary and sufficient to maintain the induced feeding preference. The induction of host plant specificity is at least partly due to a tuning of taste receptors to indioside D. The taste receptors of larvae fed on host plants show an enhanced response to indioside D as compared with other plant compounds tested.

Larvae of the cabbage white Pieris rapae are specialists on plants belonging to the family Brassicaceae (Cruciferae). Adult females have been shown to use the glucosinolate gluconasturtiin (phenylethylglucosinolate) as a recognition cue for cruciferous plants, so they can identify an appropriate host for oviposition (Huang and Renwick in J Chem Ecol 20:1025-1037, 1994). Here, we report our results from a study of the role of this glucosinolate in feeding preferences of P. rapae larvae. The larvae were allowed to choose between leaf disks from the non-host cowpea Vigna sinensis (Fabaceae) that were treated with pure gluconasturtiin in solvent, or solvent alone. Our results showed that gluconasturtiin is a feeding stimulant for P. rapae larvae. A series of chemosensory ablations revealed that this response is mediated by one set of taste sensilla, the sensilla styloconica. Electrophysiological tip recordings revealed two neurons in the lateral sensillum styloconicum that are sensitive to gluconasturtiin. These neurons show significantly higher firing frequencies with 4 mM gluconasturtiin added to the recording pipette than for recording solution alone. We propose that the sensitivity to gluconasturtiin shown by these two taste neurons is an important contributor to the animals' behavioral preference for this compound.

At the culmination of each molt, the larval tobacco hornworm exhibits a pre-ecdysis behavior prior to shedding its old cuticle at ecdysis. Both pre-ecdysis and ecdysis behaviors are triggered by the peptide, eclosion hormone (EH). Pre-ecdysis behavior consists of rhythmic abdominal compressions that loosen the old larval cuticle. This behavior is robust at larval molts, but at the larval-pupal molt the only comparable behavior consists of rhythmic dorso-ventral flexions of the anterior body. These flexions appear to be an attenuated version of the larval pre-ecdysis behavior because (1) they show the same EH dependence, and (2) the motor patterns recorded from EH treated, deafferented larval and pupal preparations are similar except that the pupal pattern is much weaker. Both patterns are characterized by rhythmic, synaptically-driven bursts of action potentials in motoneurons MN-2 and MN-3, which occur synchronously in all segments. However, the synaptic drive to the motoneurons and their resultant levels of activity are reduced during the pupal pre-ecdysis motor pattern, especially in posterior abdominal segments. Although the dendritic arbors of both motoneurons regress somewhat during the larval-pupal transformation, this does not appear to be the primary source of diminished synaptic drive because regression is greatest in the segments in which synaptic inputs remain the strongest. The developmental weakening of the pre-ecdysis motor pattern thus may be due to changes at the interneuronal level.