The role of Ultrabithorax in the patterning of adult thoracic muscles in Drosophila melanogaster.

Mutations in the homeotic gene, Ultrabithorax (Ubx), result in the transformation of the third thoracic (T3) segment into the second thoracic (T2) segment. Although it has been well established that these mutations have striking effects on adult epidermal structures in T3, the effect of these mutations on the adult musculature has been controversial. In this study, a series of Ubx regulatory mutations, anterobithorax, bithorax, postbithorax, and bithoraxoid, as well as combinations of these alleles were used to reevaluate the role of Ubx in the patterning of the T3 musculature. Homeotic indirect and direct flight muscles (IFMs and DFMs) were identified in the transformed T3 segment of all alleles and allelic combinations with the exception of postbithorax. We critically evaluated the pattern and amount of these muscles and found that while the amount and/or quality of homeotic IFMs increased, the amount of homeotic DFMs did not vary significantly as the severity of the ectodermal transformation increased. Because Ubx is not expressed in the adult mesoderm of T3, these results suggest that inductive cues play a major role in the patterning of adult thoracic muscles. We provide a model that illustrates the central role of inductive cues in generating the final adult muscle pattern in the thorax.

Imaginal pioneers prefigure the formation of adult thoracic muscles in Drosophila melanogaster.

In insects, specialized mesodermal cells serve as templates to organize myoblasts into distinct muscle fibers during embryogenesis. In the grasshopper embryo, large mesodermal cells called muscle pioneers extend between the epidermal attachment points of future muscle fibers and serve as foci for myoblast fusion. In the Drosophila embryo, muscle founder cells serve a similar function, organizing large numbers of myoblasts into larval muscles. During the metamorphosis of Drosophila, nearly all larval muscles degenerate and are replaced by a set of de novo adult muscles. The extent to which specialized mesodermal cells homologous to the founders and pioneers of the insect embryo are involved in the development of adult-specific muscles has yet to be established. In the larval thorax, the majority of imaginal myoblasts are associated with the imaginal discs. We report here the identification of a morphologically distinct class of disc-associated myoblasts, which we call imaginal pioneers, that prefigures the formation of at least three adult-specific muscles, the tergal depressor of the trochanter and dorsoventral muscles I and II. Like the muscle pioneers of the grasshopper, the imaginal pioneers attach to the epidermis at sites where the future muscle insertions will arise and erect a scaffold for developing adult muscles. These findings suggest that a prior segregation of imaginal myoblasts into at least two populations, one of which may act as pioneers or founders, must occur during development.

Different neural pathways coordinate Drosophila flight initiations evoked by visual and olfactory stimuli.

To determine the role played by the giant fiber interneurons (GFs) in coordinating the jumping stages of visually elicited and olfactory-induced fight initiation we have recorded extracellularly from the cervical connective nerve during flight initiation. A spike is recorded from the cervical connective upon brain stimulation that has the same threshold as does activation of the tergotrochanteral muscle (TTM) and dorsal longitudinal muscles (DLMs). A consistent time interval occurs between the spike and activation of the TTM. Thus, the spike probably results from activity in the GFs. The time intervals between the spike and activation of the TTM during GF stimulation and visually elicited flight initiation are similar. These results suggest that the GFs coordinate the activation of the TTM and DLMs during the jumping stage of visually elicited flight initiation. A spike is also recorded from the cervical connective during olfactory-induced flight initiations, but its shape and the time interval between it and activation of the TTM is different from that observed during GF stimulation. Although some olfactory-induced flight initiations exhibit a pattern of muscle activation, olfactory-induced flight initiations exhibit a pattern of muscle activation indistinguishable from that evoked by GF stimulation, our results indicate that regardless of the pattern of muscle activation, olfactory-induced flight initiations are not coordinated by the GF circuit. The sterotypic sequence and timing of activation of TTM and DLMs characteristic of the GF pathway can, therefore, be evoked by neurons other than those constituting the GF pathway.

Flight initiations in Drosophila melanogaster are mediated by several distinct motor patterns.

We have monitored the patterns of activation of five muscles during flight initiation of Drosophila melanogaster: the tergotrochanteral muscle (a mesothoracic leg extensor), dorsal longitudinal muscles #3, #4 and #6 (wing depressors), and dorsal ventral muscle #Ic (a wing elevator). Stimulation of a pair of large descending interneurons, the giant fibers, activates these muscles in a stereotypic pattern and is thought to evoke escape flight initiation. To investigate the role of the giant fibers in coordinating flight initiation, we have compared the patterns of muscle activation evoked by giant fiber stimulation with those during flight initiations executed voluntarily and evoked by visual and olfactory stimuli. Visually elicited flight initiations exhibit patterns of muscle activation indistinguishable from those evoked by giant fiber stimulation. Olfactory-induced flight initiations exhibit patterns of muscle activation similar to those during voluntary flight initiations. Yet only some benzaldehyde-induced and voluntary flight initiations exhibit patterns of muscle activation similar to those evoked by giant fiber stimulation. These results indicate that visually elicited flight initiations are coordinated by the giant fiber circuit. By contrast, the giant fiber circuit alone cannot account for the patterns of muscle activation observed during the majority of olfactory-induced and voluntary flight initiations.

Regulation of feeding behavior in adult Drosophila melanogaster varies with feeding regime and nutritional state.

The regulation of feeding behavior in adult Drosophila melanogaster includes such elements as ingestion responsiveness, volume ingested in a single meal, food storage in the crop and rate of defecation. Our results suggest that feeding behavior varies in a manner dependent on feeding regime (food-deprived or ad-libitum-fed) and nutritional state. Fed flies that are subsequently food-deprived become increasingly more responsive to food stimuli over time and, when offered 1% agar diets containing different concentrations of sucrose, ingest greater amounts of diets that have higher sucrose concentrations. When fed ad libitum for 72 h on these same diets, D. melanogaster maintained much smaller crops on average than food-deprived flies fed a single meal. Additionally, ad-libitum-fed flies are grouped into two categories depending on the concentration of sucrose in the diet. Flies fed for 72 h on 1% agar diets having 50 mmoll-1 sucrose or more are not affected by the concentration of sucrose in the diet, while flies fed on diets of 15 or 25 mmoll-1 sucrose increase ingestion responsiveness, crop size and the rate of defecation with decreasing concentrations of sucrose in the diet. Flies fed on even lower sucrose concentrations (5 or 10 mmoll-1 sucrose) for 27-72 h exhibit both a shift over time to larger crop sizes and increased mortality over those of flies fed 15 mmoll-1 sucrose. These data suggest that flies fed ad libitum are capable of modulating their feeding behavior in response to their nutritional state.

The motor neurons innervating the direct flight muscles of Drosophila melanogaster are morphologically specialized.

The anatomy of the motor neurons innervating six direct flight muscles in Drosophila melanogaster has been investigated by using a horseradish peroxidase backfilling technique. The somata of these motor neurons are arranged in two distinct clusters ipsilateral to the muscle they innervate. One cluster of cell bodies is located in the ventrolateral region between the prothoracic neuromere and the mesothoracic leg-related neuropil and the other is situated dorsally and posteriorly to the mesothoracic leg-related neuropil. Axons from somata in the ventrolateral cluster run in the anterior dorsal mesothoracic nerve, while axons from somata in the other cluster run in the mesothoracic accessory nerve. This distribution of somata and axons is discussed in the light of the morphological similarity and proximity of these functionally related muscles. On the basis of the branching patterns of their neurites, direct flight muscle motor neurons can be classified as stubbly, fibrous or tufted. The terminal arborizations of the motor neurons over the direct flight muscles are also morphologically specialized. Both the central and the peripheral morphological specializations of the direct flight muscle motor neurons correlate with the activity patterns exhibited by their associated muscles during flight and courtship song.

Duplication of the escape-response neural pathway by mutation of the bithorax-complex.

Each Drosophila segment exhibits specific patterns of epidermal cells, muscles, and neurons. Mutations in the homeotic genes of the bithorax-complex cause transformations of these patterns. Whereas abundant information exists concerning homeotic transformation of epidermis, transformations of muscles and motor neurons have been largely unexplored. An important indication of neuromuscular transformation in a segment is the expression of novel behavioral and physiological functions within that segment. We have resolved some of the segmental identities of neuromuscular elements in the transformed metathorax of the bithorax-complex mutant, abx bx3 pbx/Df(3R) P2, and have established the presence of a duplicated neural pathway for the escape-jump response within that segment. Although we observed frequent homeotic transformation of neural elements and the tergotrochanteral ("jump") muscle in mutants, corresponding transformation of flight muscles was infrequent, indicating that the presence of a motor neuron was not always sufficient to induce or determine the development of its target muscle.

Giant fiber activation of an intrinsic muscle in the mesothoracic leg of Drosophila melanogaster.

Cinematographic analysis reveals that an important component of the light-elicited escape response of Drosophila melanogaster is the extension of the femur-tibia joint of the mesothoracic leg. During the jumping phase of the response, this extension works synergistically with extension of the femur. Femur extension is generated by contraction of the tergotrochanteral muscle (TTM), one of four previously described escape response muscles. Femur-tibia joint extension in the mesothoracic leg has been thought to be controlled by contraction of the tibial levator (TLM), an intrinsic leg muscle. We investigated the activation of the TLM during the escape response. Electrical stimulation of the giant fiber interneuron that mediates the escape response results in activation of the TLM with a latency of 1.46 +/- 0.02 ms. The TLM is innervated by a motor neuron (TLMn) with a large cell body in the mesothoracic ganglion. The TLMn has extensive arborizations in the lateral mesothoracic leg neuromere and has a prominent medially directed neurite. To investigate possible presynaptic inputs activating the TLMn during the escape response, we analyzed the muscle responses of two mutants, giant fiber A1 and bendless. Our analysis suggests that the TLMn is activated by a novel pathway.

Bendless alters thoracic musculature in Drosophila.

bendless- (ben-) is an X chromosome mutation in Drosophila melanogaster, known to alter patterns of connections in the CNS and thus modify behavior (Thomas and Wyman, 1984). We report that in addition to its CNS effects, ben- has pleiotropic phenotypes affecting thoracic muscle patterning, pupal mortality, and post-eclosional mobility. The tergal depressor of the trochanter (TDT) normally attaches ventrally to an apodeme on the trochanter and dorsally to the lateral scutum just posterior to the intrascutal suture. In ben- individuals, TDT may attach anywhere within the boundaries of the attachment areas for TDT and dorsoventral muscles I (DVM I) and II (DVM II). Furthermore, TDT may completely lack a dorsal attachment, although it still maintains a ventral attachment. DVMs may also attach abnormally to dorsal sites normally occupied by an adjacent DVM, or may be entirely eliminated. DVM loss occurs independently of the position or presence of TDT dorsal attachment. The cytology of ben- TDT is altered. Muscles may have fibers that are swollen and stain abnormally. Other fibers may have large, axially aligned holes. ben- flies have an increased likelihood of failing to eclose and, upon eclosion, show impaired mobility. We describe several possible mechanisms for the ben- developmental defects and discuss this mutation in light of its evolutionary significance.

Morphometric analysis of thoracic muscles in wildtype and in bithorax Drosophila.

The tergotrochanteral (TTM) "jump" muscles in the second (T2) and third (T3) thoracic segments of the fruit fly, Drosophila melanogaster, were analyzed morphologically and morphometrically in wildtype (Canton-S) and bithorax mutants (abx bx3 pbx/Df(3R)P2). In the transformed T3 segments of mutant flies, the TTMs were greatly increased in fiber number (330% of wildtype), length (141%), and volume (460%), thus manifesting both hyperplasia and hypertrophy. In contrast, TTMs in the "untransformed" T2 segments of mutant flies were both hypoplastic and hypotrophic, in that significant decreases in fiber number (93% of wildtype), length (90%), and volume (80%) were observed. Two relationships emerged from analysis of the morphometric data: 1) Although the fiber numbers and volumes of the transformed T3 TTMs in bithorax flies were greatly increased, the total combined volumes of the TTMs in T2 + T3 remained approximately the same in bithorax compared to wildtype flies. 2) The changes in TTM volumes in bithorax flies compared to those in wildtype were proportional to the relative changes in fiber numbers times the relative changes in muscle lengths. These observations suggest that the genes of the bithorax complex influence the number and the length of tubular muscles fibers of the TTMs, but do not significantly affect the mean cross-sectional areas of these fibers. Fibrillar muscle fibers, which are not found at all in T3 segments in wildtype flies, were observed in the transformed T3 segments of bithorax mutants in 11 of 18 cases (61%), but typically as wisps, not in complete muscles. We suggest that, in the T3 segment of the bithorax flies, the relative differences between the massive transformation of tubular TTMs vs. the minimal appearance of fibrillar muscles may be related, in part, to the relative availability of muscle precursors.

Trans-sexually grafted antennae alter pheromone-directed behaviour in a moth.

When tobacco hornworm moths (Manduca sexta) are tested in a wind tunnel with a source of female pheromones upwind, males but not normal females show pheromone-modulated anemotaxis and a characteristic mate-seeking behavioural sequence. These behaviours are produced by stimulation of sensory neurones found only in male antennae. These neurones project axons only to dendrites of pheromone-specific interneurones in the macroglomerular complex, a region of neuropil in the antennal lobe characteristic of males but not present in normal females. Some interneurones in the antennal lobes of female moths that have received grafts of male antennae (gynandromorphs) respond postsynaptically to stimulation with bombykal, a major component of the pheromone. They branch into a region resembling the macroglomerular complex, like their counterparts in normal males. We show here that gynandromorphic females respond to pheromonal stimulation with anemotaxis. We also find that normal females display a similar sequence in response to the odour of their egg-laying site, the tobacco plant. It is likely that a common motor path is used either by pheromone-specific interneurones in the antennal lobes of males or by tobacco-specific interneurones in females. We assume that the interneurones in gynandromorphic females that branch into the macroglomerular complex induced by a grafted male antenna can activate this pathway.