The influence of neurotransmitters and neuroactive substances on stridulatory behaviour was analysed in two species of acridid grasshoppers (Omocestus viridulus and Chorthippus mollis). Acetylcholine, octopamine, γ-aminobutyric acid and glutamate were applied by pressure injection (0.5–1.0 nl, 10−3 mol l-1) into the protocerebrum. All except octopamine were also applied to the metathoracic ganglion by pressure injection or superfusion (1 ml).

Injection of acetylcholine into the medial dorsal neuropile of the protocerebrum elicited continuous long-lasting species-specific stridulation in both acridid species. All other substances tested had no effect when injected into the brain. Injection of acetylcholine into the medial dorsal neuropile of the metathoracic ganglion enhanced the amplitude of the stridulatory leg movements elicited by electrical brain stimulation. It did not alter the repetition rate or coordination of the movements in O. viridulus; but it decreased the length of stridulatory cycles in C. mollis.

Injection of γ-aminobutyric acid into the medial dorsal metathoracic neuropile in both species suppressed the stridulatory leg movements ipsilateral to the injection site but did not alter those on the contralateral side. Superfusion of the metathoracic ganglion with aminobutyric acid suppressed the movements of both hindlegs.

Pressure injection of glutamate into the metathoracic ganglion had no effect on the stridulatory leg movements, but superfusion enhanced the stridulatory movements.

Acoustic communication is well developed in acridid grasshoppers of the taxon Gomphocerinae. These grasshoppers stridulate sonorously by rhythmically rubbing their hindlegs against the elytra. Every species produces a specific stridulatory sound and movement pattern. The acoustic communication of these grasshoppers is thus the basis for mate recognition and species separation in the biotope (Jacobs, 1953; von Helversen and von Helversen, 1983, 1994).

The main centres for the control of the stridulatory behaviour are the brain and the metathoracic ganglion complex (for a review, see Elsner, 1994). Electrical brain stimulation experiments indicate that the brain controls stridulation by descending activity (Huber, 1964; Wadepuhl, 1983; Hedwig, 1986). Considerable progress has been made in analysing neuronal activity underlying stridulatory behaviour. The cephalothoracic command system that drives the thoracic stridulatory network was recently identified (Hedwig, 1994), and the physiology, structure and function of stridulatory interneurones within the metathoracic network have been described (Gramoll and Elsner, 1987; Hedwig, 1992a,b).

In several invertebrate motor systems, neurotransmitters have a clear impact on rhythmic motor activity. In the leech, swimming activity can be elicited by application of serotonin (Willard, 1981), and in the locust, flight motor activity occurs after local injection of octopamine into the metathoracic ganglion complex (Sombati and Hoyle, 1984). Moreover, stridulation can be evoked in crickets by injection of acetylcholine agonists into the protocerebrum (Otto, 1978). Besides the pharmacological initiation of rhythmic behaviour, there are some reports on modulation of motor activity by classical neurotransmitters (Rajashekhar and Wilkens, 1992; Cazalets et al. 1987). However, nothing is known of the pharmacological aspects of grasshopper stridulation. It was the aim of our experiments to analyse the influence of classical invertebrate neurotransmitters on stridulation in two species of acridid grasshoppers, Omocestus viridulus and Chorthippus mollis. Two species with different movement patterns were chosen for the experiments to permit an analysis of possible species-specific differences in the effects of the neurotransmitters.

Experiments were performed on 84 male acridid grasshoppers Omocestus viridulus (L.) and Chorthippus mollis (Charp.). Specimens were collected in low mountain ranges in the surroundings of Göttingen and Erlangen, Germany, and were kept in the laboratory for several weeks. All experiments were performed at 25–28 °C.

Grasshoppers were fixed with wax at the pronotum in a U-shaped holder. The animals were free to move their hindlegs. Stridulatory movements were recorded with two optoelectronic cameras, one for each leg. The cameras use position-sensitive photodiodes (United Detector Technology, type SL 15) for the measurement of leg movements (von Helversen and Elsner, 1977).

For pharmacological stimulation, either the head capsule was opened frontally to expose the surface of the protocerebrum or the metathoracic sternites were removed to expose the metathoracic ganglion complex. Pharmacological agents were administered either by superfusing the whole ganglion or by pressure injection into the neuropile. During superfusion, about 1 ml of the agent at a concentration of 10−3 mol l-1 dissolved in saline (Robertson and Pearson, 1982) was applied with a pipette to the ganglion and allowed to diffuse within the haemolymph. For pressure injection into the neuropile of the brain or metathoracic ganglion, a micropump (WPI, type PV 820) was connected to a glass microelectrode broken to a tip diameter of 10–15 μm and filled with the agent (10−3 mol l-1 in saline). After penetration of the neuronal tissue, volumes of 0.5–1.0 nl were injected by pressure pulses of 100–500 ms duration and 267 kPa pressure. All neuroactive agents were obtained from Sigma.

In the brain, the substances were injected into the medial dorsal part of the protocerebrum (Fig. 1) and the elicited behaviour was recorded. In the metathoracic ganglion, the anterior dorsal neuropile was chosen as the injection site (Fig. 1). During thoracic application of neuroactive agents, stridulation was elicited by electrical d.c. brain stimulation (Hedwig, 1986), and the behaviour was compared before and after pharmacological stimulation. The number of animals used for every neurotransmitter application is given in the corresponding section. Quantitative data are presented for one representative experiment. The terms ipsilateral and contralateral refer to locations relative to the injection site.

Fig. 1.

Areas used for pressure injection of neuroactive agents into the neuropile of the protocerebrum or metathoracic ganglion. In both species, stridulatory interneurones arborize in the dorsal neuropile regions of the marked areas.

Fig. 1.

Areas used for pressure injection of neuroactive agents into the neuropile of the protocerebrum or metathoracic ganglion. In both species, stridulatory interneurones arborize in the dorsal neuropile regions of the marked areas.

The stridulatory leg movements and markers for the pressure pulse and the brain stimulus were recorded on a tape recorder (Racal Store 4D). After the experiments, data were sampled at

3.3 kHz per channel using an A/D board (Data Translation, DT 2821 F-8DI) and stored as binary files on the hard disc of an IBM-compatible PC. The data were evaluated and representative plots created using the software package NEUROLAB (Hedwig and Knepper, 1992). All frequency distributions were normalized to a single reference, i.e. the number of occurrences per bin was divided by the total number of references. Computations were carried out using Quattro Pro 2.0 (Borland Corp.). All graphic output was produced with a laser printer (Hewlett Packard Laserjet 4).

During courtship behaviour, the acridid grasshoppers Omocestus viridulus (L.) and Chorthippus mollis (Charp.) perform sequences of species-specific stridulatory leg movements which last for 30–60 s. In both species, the movement amplitude gradually increases to a maximum level during a sequence. In O. viridulus, a movement cycle consists of simple up-and-down movements with a cycle length of 70–90 ms. The hindlegs each perform a different movement pattern (patterns I and II) which have different amplitudes and are slightly out of phase. The maximum peak-to-peak amplitudes of patterns I and II are 2–3 mm and about 1 mm, respectively. The hindleg producing pattern II reaches the upper reversal point slightly after the hindleg producing pattern I; the phase shift φ is 0.05–0.1 (Elsner, 1974; R. Heinrich and I. Reis, unpublished data).

The hindlegs also perform two different movement patterns in C. mollis. Pattern I consists of a slow up-and-down movement about 3 mm in amplitude and a following bow-like up-and-down movement 2.5 mm in amplitude, upon which are superimposed rapid leg oscillations of about 50 Hz. In pattern II, there is no slow up-and-down movement and only the bow-like movement with rapid leg oscillations occurs. A complete stridulatory cycle lasts about 600 ms and is therefore almost 10 times longer than in O. viridulus. The cycles start synchronously on both sides and the amplitude of the slow up- and-down movements gradually increases at the beginning of a stridulatory sequence. The rapid oscillations have a phase relationship of 0.2–0.3 and are generally 0.2–0.7 mm in amplitude (Elsner, 1974; R. Heinrich and I. Reis, unpublished data).

Brain application of pharmacological agents elicits stridulation

Descending cephalothoracic interneurones, stimulation of which is sufficient to elicit stridulation, have their main arborization in the medial dorsal neuropile of the protocerebrum (Hedwig, 1994). Guided by the structure of these command neurones, we attempted to elicit stridulation by pharmacological stimulation of the neuropilar regions in which they arborize.

In six different individuals of both O. viridulus and C. mollis, acetylcholine (ACh) was pressure-injected into the medial dorsal neuropile of the protocerebrum (Fig. 1) and the subsequent behaviour recorded. In these experiments, no electrical brain stimulus was applied. In all animals of both species, ACh elicited long-lasting stridulation (Figs 2, 3). In O. viridulus and C. mollis, stridulation started 1–4 s after the pressure pulse (Figs 2A, 3A) and full-amplitude stridulation was reached more quickly than under natural conditions. The timing and amplitude of leg movements, however, were regular and similar to those of unrestrained singing grasshoppers (Elsner, 1974). In O. viridulus, the mean period of the movement cycles was 71 ms, the phase relationship between the two hindleg movements was 0.07 and the mean amplitude of the pattern I movement 2.14 mm (Fig. 2B). In C. mollis, single movement cycles lasted 514 ms, the movements of both legs were coordinated with a phase of 0.02 and the maximum amplitude of the slow up-and-down movements was about 2.73 mm. The rapid leg oscillations had a duration of 22 ms, a phase of 0.22 between the two legs and an amplitude of about 0.61 mm (Fig. 3B).

Fig. 2.

Effect of pressure injection of acetylcholine (ACh) into the dorsal neuropile of the protocerebrum in Omocestus viridulus. (A) A long-lasting species-specific stridulation sequence is elicited after a pulse injection of ACh, indicated by the black bar. Sections of the stridulatory movements are given immediately after the injection and after 8 min. (B) Quantitative evaluation of the stridulatory movements (occurrence is given in relative units): cycle length (left; mean value 71 ms), phase between ipsilateral and contralateral leg movements (middle; mean value 0.07) and amplitude of ipsilateral leg movements (right; mean value 2.14 mm). The x-axis of the phase diagram has been shifted by 0.5. The variables are defined in the inset. ACh, ACh injection; iHL, ipsilateral hindleg; cHL, contralateral hindleg. 452 cycles were evaluated.

Fig. 2.

Effect of pressure injection of acetylcholine (ACh) into the dorsal neuropile of the protocerebrum in Omocestus viridulus. (A) A long-lasting species-specific stridulation sequence is elicited after a pulse injection of ACh, indicated by the black bar. Sections of the stridulatory movements are given immediately after the injection and after 8 min. (B) Quantitative evaluation of the stridulatory movements (occurrence is given in relative units): cycle length (left; mean value 71 ms), phase between ipsilateral and contralateral leg movements (middle; mean value 0.07) and amplitude of ipsilateral leg movements (right; mean value 2.14 mm). The x-axis of the phase diagram has been shifted by 0.5. The variables are defined in the inset. ACh, ACh injection; iHL, ipsilateral hindleg; cHL, contralateral hindleg. 452 cycles were evaluated.

Fig. 3.

Effect of ACh injection into the protocerebrum in Chorthippus mollis. (A) A long-lasting species-specific stridulation sequence is elicited. Sections of the stridulatory movements are given immediately after the injection and after 5 min. (B) Quantitative evaluation of the slow stridulatory up-and-down movements (top): cycle length (left; mean value 514 ms), phase between ipsilateral and contralateral leg movements (middle; mean value 0.02) and amplitude of ipsilateral leg movements (right; mean value 2.73 mm). Quantitative evaluation of the rapid stridulatory oscillations (bottom): cycle length (left; mean value 22 ms), phase between ipsilateral and contralateral oscillations (middle; mean value 0.22) and amplitude of ipsilateral leg oscillations (right; mean value 0.61 mm). The variables are defined in the insets. For abbreviations, see Fig. 2. Twenty-six slow up-and-down movements and 256 rapid oscillations were evaluated.

Fig. 3.

Effect of ACh injection into the protocerebrum in Chorthippus mollis. (A) A long-lasting species-specific stridulation sequence is elicited. Sections of the stridulatory movements are given immediately after the injection and after 5 min. (B) Quantitative evaluation of the slow stridulatory up-and-down movements (top): cycle length (left; mean value 514 ms), phase between ipsilateral and contralateral leg movements (middle; mean value 0.02) and amplitude of ipsilateral leg movements (right; mean value 2.73 mm). Quantitative evaluation of the rapid stridulatory oscillations (bottom): cycle length (left; mean value 22 ms), phase between ipsilateral and contralateral oscillations (middle; mean value 0.22) and amplitude of ipsilateral leg oscillations (right; mean value 0.61 mm). The variables are defined in the insets. For abbreviations, see Fig. 2. Twenty-six slow up-and-down movements and 256 rapid oscillations were evaluated.

In both species, natural stridulation sequences last for 30–60 s. The grasshoppers completed one stridulatory sequence after a single pulse of ACh. In O. viridulus, however, continuous stridulation of up to 8 min could be elicited, and in C. mollis stridulation could last for about 5 min after a single pulse of ACh. After such long stridulatory sequences, the grasshoppers appeared to be exhausted. Even then, further brief sequences could be induced by pressure injection of additional ACh. ACh therefore obviously has a strong activating effect on the cephalic stridulatory system when injected into the medial dorsal protocerebrum.

In five animals, saline, γ-aminobutyric acid (GABA), glutamate and/or octopamine was injected into brain regions corresponding to the sites of acetylcholine injections. None of these substances elicited any significant leg movements.

Modulation of stridulatory leg movements by pharmacological stimulation of the metathoracic neuropile

Although stridulatory behaviour is controlled by the brain, the stridulatory pattern-generating network is located within the metathoracic ganglion complex (Gramoll and Elsner, 1987; Hedwig, 1992a,b). Certain interneurones that are able to induce or modulate stridulatory leg movements arborize close to the medial dorsal tract. The thoracic stridulatory network can be activated by electrical d.c. brain stimulation (Hedwig, 1986). In O. viridulus, a rectangular d.c. current pulse (10 s duration, 15–30 μA amplitude) elicits stridulation for the duration of the stimulus with a reliability of almost 100 %. In C. mollis, the brain stimulus must be increased gradually over many seconds and then kept at a constant level to elicit stridulation. It was effective in about 80 % of animals stimulated. In C. mollis, the d.c. stimulus marker could only be represented as a straight line in the figures so it is not presented.

To analyse the effect of neuroactive substances on the thoracic stridulatory network, the following steps were performed. First, stridulatory leg movements were elicited by brain stimulation and were taken as a reference. A neuroactive substance was then pressure-injected into the medial dorsal metathoracic neuropile (Fig. 1), or the ganglion was superfused with the substance. Thereafter, the brain stimulus was repeatedly applied at intervals of 1–3 min to elicit stridulatory leg movements. Finally, the stridulatory movements were compared before and after the pharmacological stimulus.

Effects of acetylcholine on stridulatory leg movements

In natural stridulation, the amplitude of the stridulatory leg movements in O. viridulus and C. mollis increases gradually (Elsner, 1974). When stridulation was induced in O. viridulus by d.c. brain stimulation, the leg movements also gradually increased in amplitude. The average amplitude was about 2 mm and reached almost 3 mm at the end of the stimulus (Fig. 4A). This increase in amplitude corresponds closely to the natural behaviour of the grasshopper. When the brain stimulus ended, stridulation also stopped.

Fig. 4.

Effect of pressure injection of acetylcholine into the dorsal neuropile of the metathoracic ganglion in O. viridulus. (A) Reference recording of the stridulatory movements elicited by d.c. brain stimulation (top). Stridulation sequence 1 min after ACh injection (bottom). (B) Quantitative evaluation of the stridulatory movements of the reference recording (top): cycle length (left; mean value 71 ms), phase between ipsilateral and contralateral leg movements (middle; mean value 0.11) and amplitude of ipsilateral leg movements (right; mean value 2.1 mm). Stridulatory movements after ACh application (bottom): cycle length (left; mean value 72 ms), phase between ipsilateral and contralateral leg movements (middle; mean value 0.11) and amplitude of ipsilateral leg movements (right; mean value 3.38 mm). BSt, d.c. brain stimulus to elicit stridulation; iHL, ipsilateral hindleg; cHL, contralateral hindleg. 123 stridulatory cyles were evaluated for the control and 127 were evaluated following ACh injection.

Fig. 4.

Effect of pressure injection of acetylcholine into the dorsal neuropile of the metathoracic ganglion in O. viridulus. (A) Reference recording of the stridulatory movements elicited by d.c. brain stimulation (top). Stridulation sequence 1 min after ACh injection (bottom). (B) Quantitative evaluation of the stridulatory movements of the reference recording (top): cycle length (left; mean value 71 ms), phase between ipsilateral and contralateral leg movements (middle; mean value 0.11) and amplitude of ipsilateral leg movements (right; mean value 2.1 mm). Stridulatory movements after ACh application (bottom): cycle length (left; mean value 72 ms), phase between ipsilateral and contralateral leg movements (middle; mean value 0.11) and amplitude of ipsilateral leg movements (right; mean value 3.38 mm). BSt, d.c. brain stimulus to elicit stridulation; iHL, ipsilateral hindleg; cHL, contralateral hindleg. 123 stridulatory cyles were evaluated for the control and 127 were evaluated following ACh injection.

In O. viridulus, pressure injection of ACh (0.5–1 nl, 10−3 mol l-1) into the anterior dorsal neuropile of the metathoracic ganglion had a significant effect on the amplitude of stridulation (tested in nine animals). About 1 min after ACh injection, the mean amplitude of the stridulatory leg movements increased to about 4 mm. However, during the first part of the sequence, single up-and-down movements of both legs reached as much as 6–7 mm in amplitude. In the corresponding part of the reference sequence, the amplitude of the stridulatory leg movements was only 0.5–1.0 mm. Interestingly, ACh did not alter the repetition rate or phase of the movements. The mean cycle length was 71 ms before and 72 ms after pharmacological treatment, and the phase of the legs was 0.11 in both cases (Fig. 4B). The duration of the stridulatory sequence did not change and the leg movements stopped at the end of the brain stimulus. The effect of the injected acetylcholine vanished 3–4 min after the application.

In C. mollis, electrical brain stimulation was generally not as effective as in O. viridulus. There was some variation in the movement pattern compared with natural stridulation. During the reference recording, the brain stimulus elicited a stridulatory sequence with low-amplitude slow up-and-down movements (Fig. 5A). After injection of ACh, the slow up-and-down movements of the leg ipsilateral to the injection site were distinctly increased in amplitude. Some slow up-and-down movements increased even during the first 5 min (see asterisks in Fig. 5A). The effect then stabilized and after 10 min nearly every slow up-and-down movement reached 4 mm in amplitude. In addition, the mean cycle length decreased from 616 to 519 ms. Over the next 10 min, the effects gradually vanished. There was no change in the amplitude of the rapid leg oscillations. Owing to the variability of the leg movements, we did not check the phase relationship between the legs quantitatively.

Fig. 5.

Effect of pressure injection of ACh into the dorsal neuropile of the metathoracic ganglion in C. mollis. (A) Reference recording of the stridulatory movements elicited by d.c. brain stimulation (top). Stridulatory movements 2 min after acetylcholine injection (bottom). Increased slow up-and-down movements are marked by asterisks. (B) Amplitude of the ipsilateral slow up-and-down movements before and after acetylcholine application (arrow). Insets give corresponding original recordings. iHL, ipsilateral hindleg; cHL, contralateral hindleg.

Fig. 5.

Effect of pressure injection of ACh into the dorsal neuropile of the metathoracic ganglion in C. mollis. (A) Reference recording of the stridulatory movements elicited by d.c. brain stimulation (top). Stridulatory movements 2 min after acetylcholine injection (bottom). Increased slow up-and-down movements are marked by asterisks. (B) Amplitude of the ipsilateral slow up-and-down movements before and after acetylcholine application (arrow). Insets give corresponding original recordings. iHL, ipsilateral hindleg; cHL, contralateral hindleg.

In neither species did superfusion of the metathoracic ganglion with 1 ml of 10−3 mol l-1 ACh have any effect (tested in five animals).

Effects of GABA on stridulatory leg movements

The neurotransmitter GABA has a well-established inhibitory function within the central nervous system of arthropods (Kerkut et al. 1969; Pitman and Kerkut, 1970). In the stridulatory network of grasshoppers, descending suboesophageal neurones, which arborize in the metathoracic ganglion and decrease the amplitude of stridulatory leg movements upon intracellular stimulation, exhibit a GABAergic immunoreactivity (Lins and Lakes-Harlan, 1994). We tested the effect of GABA on the stridulatory movement pattern by local pressure injection of GABA into one side of the metathoracic ganglion (14 animals) and by superfusion of the complete ganglion complex (19 animals).

During the reference recording, the amplitude of the stridulatory leg movements was 1.7 mm in O. viridulus (Fig. 6A) and 2.3 mm for the slow up-and-down movements in C. mollis (Fig. 7A). Injection of GABA into the metathoracic ganglion had a significant effect on the amplitude of the stridulatory leg movements. Within 1 min, in both species, movement of the leg ipsilateral to the injection side was completely suppressed. This effect lasted about 3 min. Over the next 15 min, there was a gradual recovery of the stridulatory amplitude (Figs 6B, 7B). In C. mollis, the suppression of the rapid leg oscillations was not as strong as the suppression of the slow up-and-down movements since the recovery of rapid oscillations occurred within 2 min of GABA injection. In both species, the leg contralateral to the injection side continued stridulating, generally without a change in amplitude. The repetition rate of these stridulatory cycles sometimes increased after GABA application. This infrequent effect may be due to mutual inhibitory connections of the hemiganglionic pattern generators (see Discussion).

Fig. 6.

Effect of GABA injection into the dorsal neuropile of the metathoracic ganglion in O. viridulus. (A) Reference recording of the stridulatory movements elicited by d.c. brain stimulation (top). Stridulatory movements 1 min after GABA injection (bottom). (B) Evaluation of the ipsilateral stridulatory movement amplitude before and after GABA application. The application of GABA is marked by the arrow. For abbreviations, see Fig. 4.

Fig. 6.

Effect of GABA injection into the dorsal neuropile of the metathoracic ganglion in O. viridulus. (A) Reference recording of the stridulatory movements elicited by d.c. brain stimulation (top). Stridulatory movements 1 min after GABA injection (bottom). (B) Evaluation of the ipsilateral stridulatory movement amplitude before and after GABA application. The application of GABA is marked by the arrow. For abbreviations, see Fig. 4.

Fig. 7.

Effect of GABA injection into the dorsal neuropile of the metathoracic ganglion in C. mollis. (A) Reference recording of the stridulatory movements elicited by d.c. brain stimulation (top). Stridulatory movements elicited 1 min after GABA injection (bottom). (B) Evaluation of the amplitude of the ipsilateral slow stridulatory up-and-down movements (filled squares) and the amplitude of the rapid oscillations (open squares) before and after GABA application. The application of GABA is marked by the arrow. For abbreviations, see Fig. 5.

Fig. 7.

Effect of GABA injection into the dorsal neuropile of the metathoracic ganglion in C. mollis. (A) Reference recording of the stridulatory movements elicited by d.c. brain stimulation (top). Stridulatory movements elicited 1 min after GABA injection (bottom). (B) Evaluation of the amplitude of the ipsilateral slow stridulatory up-and-down movements (filled squares) and the amplitude of the rapid oscillations (open squares) before and after GABA application. The application of GABA is marked by the arrow. For abbreviations, see Fig. 5.

In both species, stridulation of the ipsilateral hindleg could be completely suppressed. Thus, GABA predominantly exhibited a local inhibitory effect and did not diffuse through the neuropile of the ganglion.

Superfusion of the metathoracic ganglion with GABA solution (1 ml, 10−3 mol l-1) was also effective. In both species, the stridulatory movements of both hindlegs could repeatedly be completely suppressed by bath application. The behaviour always recovered after superfusing the ganglion with saline. During this recovery phase, the d.c. brain stimulus required to elicit stridulation generally had to be increased from 20 μA to about 60–80 μA to be effective.

Effects of glutamate on stridulatory leg movements

Glutamate is the excitatory transmitter at the insect neuromuscular synapse (Piek, 1985) and it may also have central excitatory effects (Sombati and Hoyle, 1984). Its effect on stridulatory leg movements was tested with pressure injection in four animals and with bath application to the metathoracic ganglion in six grasshoppers. Pressure injection of glutamate into the neuropile did not alter the stridulatory leg movements. We did not observe any changes in O. viridulus or in C. mollis. However, bath application of glutamate significantly enhanced the amplitude of the stridulatory leg movements (Figs 8A, 9A). In O. viridulus, the stridulatory amplitude of the two hindlegs increased from about 1.0 mm and 0.5 mm, respectively, to 2.0 mm for both legs. The effect occurred within 2 min of glutamate application and remained at a maximum level for about 3 min. The effect gradually decreased within the next 10 min (Fig. 8B). There was no effect on the cycle length of the stridulatory movement, which was 81 ms before and 82 ms after application.

Fig. 8.

Effect of glutamate (Glu) superfusion of the metathoracic ganglion in O. viridulus. (A) Reference recording of the stridulatory movements elicited by d.c. brain stimulation (top). Stridulatory movements 1 min after glutamate application (bottom). (B) Evaluation of the ipsilateral stridulatory movement amplitude before and after glutamate application. The application of glutamate is marked by the arrow. For abbreviations, see Fig. 4.

Fig. 8.

Effect of glutamate (Glu) superfusion of the metathoracic ganglion in O. viridulus. (A) Reference recording of the stridulatory movements elicited by d.c. brain stimulation (top). Stridulatory movements 1 min after glutamate application (bottom). (B) Evaluation of the ipsilateral stridulatory movement amplitude before and after glutamate application. The application of glutamate is marked by the arrow. For abbreviations, see Fig. 4.

In C. mollis there was a clear alternation of up-and-down movements and rapid oscillations during the reference recording, especially in the right hindleg (Fig. 9A). Within 3 min of bath application of glutamate, the amplitude of the slow up-and-down movements distinctly increased on the right side, from 2.8 to 5.5 mm. This effect remained stable for about 15 min and then gradually vanished. Simultaneously, the amplitude of the rapid leg oscillations decreased from 0.6 to 0.2 mm. Corresponding changes also occurred on the left side, but were less marked. Glutamate application had only a minor effect on the cycle length, which was 468 ms before and 477 ms after the application.

Fig. 9.

Effect of glutamate (Glu) superfusion of the metathoracic ganglion in C. mollis. (A) Reference recording of the stridulatory movements elicited by d.c. brain stimulation (top). Stridulatory movements elicited 5 min after glutamate superfusion (bottom). (B) Evaluation of the amplitude of the ipsilateral slow stridulatory up-and-down movements (open squares) and of the amplitude of the rapid leg oscillations (filled squares) before and after glutamate application. The application of glutamate is marked by the arrow. For abbreviations, see Fig. 5.

Fig. 9.

Effect of glutamate (Glu) superfusion of the metathoracic ganglion in C. mollis. (A) Reference recording of the stridulatory movements elicited by d.c. brain stimulation (top). Stridulatory movements elicited 5 min after glutamate superfusion (bottom). (B) Evaluation of the amplitude of the ipsilateral slow stridulatory up-and-down movements (open squares) and of the amplitude of the rapid leg oscillations (filled squares) before and after glutamate application. The application of glutamate is marked by the arrow. For abbreviations, see Fig. 5.

The stridulatory motor pattern of acridid grasshoppers is generated in the metathoracic ganglion complex and is under the control of descending cephalic neurones. That is, control and motor pattern generation constitute different aspects of stridulatory behaviour and are spatially separated within the central nervous system. Therefore, in our considerations we must distinguish between pharmacological effects on the control of stridulatory behaviour by the brain and effects on the actual generation of the stridulatory motor pattern by the thoracic network.

Pharmacological brain stimulation and the control of stridulation

Within the insect brain, all classical transmitters such as acetylcholine, GABA and glutamate, as well as amines and peptides, have been detected (for a review, see Homberg, 1994). In our experiments, we did not observe an effect on stridulation after injection of GABA, octopamine or glutamate into the medial protocerebrum. Thus, these substances might not be involved in eliciting stridulatory behaviour. Because of the experimental approach used, we are not able to identify any inhibitory effects on stridulation. If any of these substances had an inhibitory action on the control of stridulation, we would not have been able to detect it in these experiments.

Local injection of ACh into the medial dorsal neuropile of the protocerebrum reliably elicited continuous stridulation in both species. It has to be considered that small mechanically induced lesions in the brain neuropile can also elicit long-lasting stridulation (Huber, 1955). However, any mechanical cause was probably only of minor importance since neither saline nor any other tested substance elicited stridulation after pressure injection. Stridulation was always coupled to ACh injection, and ACh is known to be a widespread excitatory neurotransmitter within all regions of the insect brain (Breer, 1987; Homberg, 1994; Kreissl and Bicker, 1989). So far, we are not able to describe precisely the location where the injected acetylcholine was active. As a transmitter, it may have produced tonic activation of the descending cephalothoracic command neurones for stridulation, which are known to arborize in the region of the injection site. This would be in agreement with physiological results, since even tonic activity of the command neurones is sufficient to elicit stridulatory leg movements in both species of grasshopper (Hedwig, 1994 and unpublished data). It is also possible that brain structures presynaptic to the descending pathway may have been activated. However, we have no reliable data on the structure and location of these neuronal networks within the brain.

As in these two species of grasshopper, injection of ACh agonists into the cricket brain also elicited long-lasting stridulation (Otto, 1978). In both orthopteran groups, therefore, ACh plays an important role in the control of stridulatory behaviour by acting on brain structures.

Pharmacological modulation of stridulation

In acridid grasshoppers, the metathoracic ganglion complex contains the pattern-generating network, which consists of two hemiganglionic pattern generators whose activity is coordinated by local interneurones (Ronacher, 1989; Hedwig, 1992a,b). Interneurones that can elicit stridulatory motor activity are located in the dorsal neuropile regions (Hedwig, 1992a). The effect of neurotransmitters on stridulatory pattern generation was tested by superfusion or pressure injection into the dorsal neuropile regions. In these experiments, stridulation was elicited by electrical brain stimulation, so we could detect excitatory as well as inhibitory effects.

None of the substances tested (ACh, GABA, glutamate) elicited stridulatory leg movements after metathoracic application. It may be that we did not stimulate the crucial locations within the neuropile or that the spatial and temporal distribution of neurotransmitters within the network was not that required to elicit the stridulatory motor pattern. In other motor systems, local injection or superfusion of the nerve cord with a specific neurotransmitter may elicit motor activity, such as locust flight (Sombati and Hoyle, 1984; Stevenson and Kutsch, 1987) or leech swimming (Willard, 1981).

In both species of grasshopper, the modulatory effects of the neurotransmitters on stridulatory pattern generation generally corresponded to their established function in insect nervous systems. The elicited modulations were very similar. In O. viridulus and in C. mollis, ACh and glutamate had an excitatory influence, whereas GABA suppressed the stridulatory leg movements. Since almost nothing is known about the function of these neurotransmitters in the stridulatory network, we are unable to relate the elicited effects to specific neurones and their functions. Moreover, we can only discuss the general effects encountered.

Within the metathoracic ganglion of the locust Schistocerca gregaria, acetylcholine receptors are distributed in all neuropile regions (Leitch et al. 1993). Somata of thoracic neurones of the cockroach Periplaneta americana (Kerkut et al. 1969) and the locust Locusta migratoria (Benson, 1992) are excited by ionophoretically applied ACh. Corresponding to its excitatory effect, pressure-injected ACh caused a distinct increase in the amplitude of stridulatory movements in both O. viridulus and C. mollis. In C. mollis, however, only the slow up-and-down movements were enhanced, whereas the rapid leg oscillations remained unchanged. This may be because in this species the slow up-and-down movements and the rapid leg oscillations are controlled by different neuronal networks. ACh did not consistently change the repetition rate of the leg movements or their coordination. It seems that ACh particularly enhanced the output of the pattern-generating network and the motoneurone activity but did not influence the timing within the network. In the locust flight system, however, ACh does influence pattern generation and induces a transient decrease in flight frequency (Kutsch and Murdock, 1973).

GABA functions as an inhibitory neurotransmitter in the invertebrate nervous system (Pitman, 1985; Sattelle, 1990). In the stridulatory network of both grasshopper species, certain suboesophageal neurones with axonal arborizations in the metathoracic ganglion show GABA-like immunoreactivity. Depolarization of these interneurones reduces the amplitude of stridulatory movements of both hindlegs (Lins and Lakes-Harlan, 1994). Coordination of the stridulatory leg movements is controlled by a metathoracic local stridulatory interneurone with bilateral arborizations. Depolarization of the interneurone reduces the amplitude of the contralateral leg movement. This has led to the assumption of mutually inhibitory connections between the hemiganglionic pattern generators (Hedwig, 1992a). This interneurone also seems to be GABAergic (Ocker, 1994). Therefore, in O. viridulus and in C. mollis, the neurotransmitter GABA has an important function in both metathoracic and suboesophageal neurones that are part of the stridulatory network. Effects such as those resulting from stimulation of the GABAergic stridulatory interneurones could be evoked by application of GABA to the metathoracic stridulatory network. In both species, bath application blocked stridulation completely and pressure injection into one hemiganglion suppressed ipsilateral leg movement. Thus, in these experiments, the functions of inhibitory stridulatory interneurones may have been mimicked by pharmacological treatment. Even the increase in the stridulatory repetition rate may not be unusual. If one of the hemiganglionic pattern generators is blocked by GABA, then the inhibition that is transmitted to the contralateral pattern generator via the bilateral interneurones is also reduced. The consequence could be an increase in contralateral motor activity.

Glutamate is the neuromuscular exitatory transmitter in insects (for a review, see Piek, 1985). Data on its function as a central transmitter are partially contradictory because it has been reported to have either excitatory (Sombati and Hoyle, 1984) or inhibitory (Dubas, 1991) effects on thoracic motoneurones in locusts. The effect of glutamate, however, depends on the site of its release since it depolarizes locust motoneurones when injected into the neuropile but may have an inhibitory effect when applied onto their somata (Parker, 1994). Pressure-injected glutamate did not change the stridulatory leg movements in our experiments. However, bath application of the transmitter caused a distinct increase of the stridulatory movement amplitude in both species of grasshopper. These effects occurred 2–10 min after glutamate application. Therefore, we assume that bath-applied glutamate was not effective in the metathoracic ganglion but influenced neuromuscular transmission or the activity of the muscles involved in stridulation.

Future prospects

It was the aim of the present study to start a pharmacological approach to the analysis of stridulatory behaviour in grasshoppers. Some effects of neurotransmitters on the control and modulation of stridulation have been revealed. We are able to relate the effect of experimentally applied GABA to the activity of inhibitory interneurones. Since, in both species, Ach seems to be important as an excitatory transmitter for the control and modulation of stridulatory behaviour, we now must analyse its distribution and function within the stridulatory system.

Mrs A. Thorson corrected the English. W.-G.O. was an associate member of the Göttinger Graduierten Kolleg ‘Organisation und Dynamik Neuronaler Netzwerke’. Supported by Deutsche Forschungsgemeinschaft He 2018/1.

Benson
,
J. A.
(
1992
).
Electrophysiological pharmacology of the nicotinic and muscarinic cholinergic responses of isolated neuronal somata from locust thoracic ganglia
.
J. exp. Biol
.
170
,
203
233
.
Breer
,
H.
(
1987
).
Neurochemical aspects of cholinergic synapses in the insect brain
. In
Arthropod Brain – its Evolution Development Structure and Functions
(ed.
A. P.
Gupta
), pp.
416
437
.
New York
:
Wiley-Interscience Publication
.
Cazalets
,
J. R.
,
Cournil
,
I.
,
Geffard
,
M.
and
Moulins
,
M.
(
1987
).
Suppression of oscillatory activity in crustacean pyloric neurons: Implication of GABAergic inputs
.
J. Neurosci
.
7
,
2884
2893
.
Dubas
,
F.
(
1991
).
Actions of putative amino acid neurotransmitters on the neuropile arborizations of flight motoneurons
.
J. exp. Biol.
155
,
337
356
.
Elsner
,
N.
(
1974
).
Neuroethology of sound production in gomphocerine grasshoppers (Orthoptera: Acrididae). I. Song patterns and stridulatory movements
.
J. comp. Physiol. A
88
,
67
102
.
Elsner
,
N.
(
1994
).
The search for the neural centers of cricket and grasshopper song
.
Fortschr. Zool
.
39
,
167
193
.
Gramoll
,
S.
and
Elsner
,
N.
(
1987
).
Morphology of local ‘stridulation’ interneurons in the metathoracic ganglion of the acridid grasshopper Omocestus viridulus L
.
J. comp. Neurol
.
263
,
593
606
.
Hedwig
,
B.
(
1986
).
On the role in stridulation of plurisegmental interneurons of the acridid grasshopper Omocestus viridulus L. I. Anatomy and physiology of descending cephalothoracic interneurons
.
J. comp. Physiol. A
158
,
413
427
.
Hedwig
,
B.
(
1992a
).
On the control of stridulation in the acridid grasshopper Omocestus viridulus L. I. Interneurons involved in rhythm generation and bilateral coordination
.
J. comp. Physiol. A
171
,
117
128
.
Hedwig
,
B.
(
1992b
).
On the control of stridulation in the acridid grasshopper Omocestus viridulus L. II. Shaping of hindleg movements by spiking and non-spiking interneurons
.
J. comp. Physiol. A
171
,
129
140
.
Hedwig
,
B.
(
1994
).
A cephalothoracic command system controls stridulation in the acridid grasshopper Omocestus viridulus L
.
J. Neurophysiol.
72
,
2015
2025
.
Hedwig
,
B.
and
Knepper
,
M.
(
1992
).
NEUROLAB, a comprehensive program for the analysis of neurophysiological and behavioural data
.
J. Neurosci. Meth.
45
,
135
148
.
Homberg
,
U.
(
1994
).
Distribution of neurotransmitters in the insect brain
.
Prog. Zool.
40
,
1
88
.
Huber
,
F.
(
1955
).
über die Funktion der Pilzkörper (Corpora pedunculata) beim Gesang der Keulenheuschrecke Gomphocerus rufus L. (Acrididae)
.
Naturwissenschaften
42
,
566
567
.
Huber
,
F.
(
1964
).
The role of the central nervous system in Orthoptera during the co-ordination and control of stridulation
. In
Acoustic Behaviour of Animals
(ed.
R. G.
Busnel
), pp.
440
487
. Amsterdam: Elsevier.
Jacobs
,
W.
(
1953
).
Verhaltensbiologische Studien an Feldheuschrecken
.
Z. Tierpsychol. (Suppl. 1)
, pp.
1
228
.
Kerkut
,
G. A.
,
Pitman
,
R. M.
and
Walker
,
R. J.
(
1969
).
Iontophoretic application of acetylcholine and GABA onto insect central neurones
.
Comp. Biochem. Physiol.
31
,
611
633
.
Kreissl
,
S.
and
Bicker
,
G.
(
1989
).
Histochemistry of acetylcholinesterase and immunocytochemistry of an acetylcholine receptor-like antigen in the brain of the honeybee
.
J. comp. Neurol
.
286
,
71
84
.
Kutsch
,
W.
and
Murdock
,
L.
(
1973
).
Locust flight oscillator: Changes in motor output by certain pharmacological agents
.
J. Insect Physiol
.
19
,
1519
1525
.
Leitch
,
B.
,
Watkins
,
B. L.
and
Burrows
,
M.
(
1993
).
Distribution of acetylcholine receptors in the central nervous systems of adult locusts
.
J. comp. Neurol.
334
,
47
58
.
Lins
,
F.
and
Lakes-Harlan
,
R.
(
1994
).
Interneurons with inhibitory effects on stridulation in grasshoppers exhibit GABA-like immunoreactivity
.
Brain Res.
635
,
103
112
.
Ocker
,
W. G.
(
1994
).
Vergleichende pharmakologische, elektrophysiologische und immunhistochemische Untersuchung zur neuronalen Mustergenese der Stridulation bei Feldheuschrecken. Dissertation, University of Göttingen. 158pp
.
Otto
,
D.
(
1978
).
änderung von Gesangsparametern bei der Grille (Gryllus campestris L.) nach Injektion von Pharmaka ins Gehirn
.
Verh. dt. Zool. Ges
.
71
,
245
.
Parker
,
D.
(
1994
).
Glutamatergic transmission between antagonistic motorneurones in the locust
.
J. comp. Physiol. A
175
,
737
748
.
Piek
,
T.
(
1985
).
Neurotransmission and neuromodulation of skeletal muscles
.
Comp. Insect Biochem. Physiol. Pharmac.
11
,
55
118
.
Pitman
,
R.
(
1985
).
Nervous system
.
Comp. Insect Physiol. Biochem. Pharmac.
11
,
5
54
.
Pitman
,
R. M.
and
Kerkut
,
G. A.
(
1970
).
Comparison of the actions of ionophoretically applied acetylcholine and gamma aminobutyric acid with the EPSP and IPSP in cockroach central neurons
.
Comp. gen. Pharmac.
1
,
221
230
.
Rajashekhar
,
K. P.
and
Wilkens
,
J. L.
(
1992
).
Dopamine and nicotine, but not serotonin, modulate the crustacean ventilatory pattern generation
.
J. Neurobiol
.
23
,
680
691
.
Robertson
,
R. M.
and
Pearson
,
K. G.
(
1982
).
A preparation for the intracellular analysis of neuronal activity during flight in the locust
.
J. comp. Physiol. A
146
,
311
320
.
Ronacher
,
B.
(
1989
).
Stridulation of acridid grasshoppers after hemisection of thoracic ganglia: evidence for hemiganglionic oscillators
.
J. comp. Physiol. A
164
,
723
736
.
Sattelle
,
D. B.
(
1990
).
GABA receptors of insects
.
Adv. Insect Physiol
.
22
,
1
113
.
Sombati
,
S.
and
Hoyle
,
G.
(
1984
).
Generation of specific behaviour in a locust by local release into neuropil of the natural neuromodulator octopamine
.
J. Neurobiol.
15
,
481
506
.
Stevenson
,
P. A.
and
Kutsch
,
W.
(
1987
).
A reconsideration of the central pattern generator concept for locust flight
.
J. comp. Physiol. A
161
,
115
129
.
Von Helversen
,
D.
and
Von Helversen
,
O.
(
1983
).
Species recognition and acoustical localisation in acridid grasshoppers: a behavioral approach
. In
Neuroethology and Behavioral Physiology
(ed.
F.
Huber
and
H.
Markel
), pp.
95
107
.
Berlin
:
Springer
.
Von Helversen
,
O.
and
Elsner
,
N.
(
1977
).
The stridulatory movements of grasshoppers recorded with an opto-electronic device
.
J. comp. Physiol. A
122
,
53
64
.
Von Helversen
,
O.
and
Von Helversen
,
D.
(
1994
).
Forces driving coevolution of song and song recognition in grasshoppers
.
Fortschr. Zool.
39
,
253
284
.
Wadepuhl
,
M.
(
1983
).
Control of grasshopper singing behavior by the brain: responses to electrical stimulation
.
Z. Tierpsychol
.
63
,
173
200
.