The Activity of Single Motor Fibres in Arthropods: II. The Australian Yabbie

A preceding paper has described the responses of single motor neurones in an insect following mechanical and electrical stimulation (Knights, 1965). The salient properties of these neurones in the dragonfly nymph were shown to include the presence of a background discharge, the occurrence of convergence and the importance of both inhibition and temporal summation/facilitation. This study of arthropod motor fibres is continued here with a comparable investigation of single dissected units in the yabbie Cherax destructor, an Australian species related to the European and American crayfish.

Considerably more work has been done on crustacean motor junctions than on those of insects. The synaptic properties of the somewhat atypical giant fibres are well known (Furshpan & Potter, 1959) and in a more recent study by Takeda & Kennedy (1964) the connexions of non-giant motor neurones are described from microelectrode recordings. These authors have also correlated these reflex connexions and discharge patterns with peripheral neuromuscular activity (Kennedy & Takeda, 1965 a, b). Hughes & Wiersma (1960a) have observed unit responses in the segmental roots in relation to the control of swimmeret movements and many investigations have dealt with single motor neurones in terms of specialized reflex feed-back systems (Eckert, 1961; Bush, 1962). Nevertheless, it seems profitable to describe the activity of yabbie motor fibres, both in terms of their natural responses and synaptic properties, as a direct comparison with those of the insect. Segmental nerves from the last abdominal ganglion were used exclusively in the present work as these provide a more diverse sensory input, and have also received very little attention from previous workers.

Male and female yabbies of the species Cherax destructor were used throughout the experiments. The animals were placed ventral side uppermost in physiological saline (van Harreveld’s solution) and the sternal exoskeleton was removed from the caudal region. For many experiments isolated abdomens were used as their survival appeared to be as good as that of whole preparations. The ventral blood vessel was carefully dissected free of the nerve cord and the caudal nerves were split longitudinally using electrolytically tapered tungsten needles as described for the dragonfly nymph (Fielden & Hughes, 1962). The paring process was continued until a readily recognizable single-unit response was obtained. These responses were monitored using fine platinum electrodes and displayed after suitable amplification on a cathode-ray oscilloscope. They were heard simultaneously by way of a loud-speaker unit. Tactile and proprioceptive endings were excited by fine brushes or needles and square-wave pulses of 0·1-1·0 msec. duration were delivered to afferent input channels through silver/silver chloride electrodes. The criteria described in the previous paper (Knights, 1965) were used in confirming the efferent nature of over sixty units.

The innervation of the caudal lobes of Cherax destructor is similar to that described for Astacus by Keim (1915). In details, however, the genera show differences in the number and subdivision of the nerves supplying the uropods and telson. The peripheral distribution of the major branches is shown in Fig. 1. The sixth abdominal ganglion possesses six pairs of segmental nerves which, unlike those of the other ganglia, appear to be mixed in function, as both sensory and motor responses have been recorded from them. These nerves, here numbered for convenience N1–N6, showed some variation both in their origins and in their terminal branches. In general, four pairs run laterally and innervate segment six, the exopodite, endopodite and the lateral part of the telson. The remaining two pairs run more posteriorly and these appear to differ slightly from those described by Keim in their innervation of the anus, intestine and the telson and its musculature. The medial nerve, N6, is the thicker of the two and runs parallel to the intestine, sending small branches to both intestine and anus, though its major branches are apparently sensory and innervate dorsal and ventral parts of the median telson. The smaller nerve, N 5, is largely motor and sends numerous branches to the compressor ani and dilator ani muscles and the anterior and posterior telson flexors (nomenclature after Schmidt, 1915). This nerve and N 6 were used predominantly in the current experiments as it was possible both to split the main trunk into small bundles and to record from some of the finer branches supplying the above mentioned muscles. Hence certain motor units could be recognized in different preparations and their efferent nature confirmed by recording from the muscle concerned. Motor axons in these nerves varied considerably in diameter, the largest being up to 30-40 μ.

Afferent fibres enter the last ganglion through all the segmental nerves. The majority could be excited by tactile stimulation of the hairs bordering the uropods and telson. In addition, manipulation of the segmental joints excited proprioceptive endings with axons running to the last ganglion (Fielden, 1960; Kennedy & Preston, 1960). In the yabbie the occurrence of vibration-sensitive endings was also demonstrated. These could be detected in recordings from the third nerve and have been shown to respond synchronously to frequencies of 100/sec. They have not been identified histologically, but possibly resemble those described in the carapace of the crayfish by Mellon (1963). Thus the sixth ganglion receives a variety of afferent fibres responding to tactile, proprioceptive, vibratory and photic stimuli. In the present work responses of motor units were observed following tactile and proprioceptive stimulation, and following electrically induced volleys applied to different afferent channels, in particular the connectives and uropod nerves N2 and N3.

The term spontaneous activity is used here to denote a background discharge occurring in many of the caudal efferent neurones in the absence of any apparent sensory input (Fig. 2). Approximately 60% of the fibres dissected in the segmental nerves of the last ganglion possessed this type of activity. The discharges showed various patterns which closely resembled those described by Preston & Kennedy (1962) in interneurones from the connectives of the crayfish and hence can be dealt with somewhat briefly here. The activity appeared to be independent of the age of the preparation and in the majority of units was unaffected by complete de-afferentation of the last ganglion. In these units the discharges were almost certainly due to endogenous activity in the last ganglion and not to an uncontrolled sensory source.

The impulse patterns ranged from regular continuous firing to irregularly occurring spikes with variable intervals in between. Most motor units showed spontaneous discharge rates of from 2 to 20 impulses/sec (Fig. 2). These are directly comparable to those observed in crayfish interneurones (Preston & Kennedy, 1962; Biederman, 1964) but are slightly lower than those found in insect motor fibres (Knights, 1965). Wiersma (1952) has described rates as high as 50/sec. in whole segmental nerves of the crayfish but these were not encountered in the yabbie. Several units showed paired discharges or ‘doublets’ similar to those described in other abdominal nerves by Hughes & Wiersma (1960a) (Fig. 2 b). The rhythmic discharges described by these authors in the first abdominal roots of other ganglia were not found in any of the caudal nerves, but the last ganglion has been shown not to be involved in swimmeret movement. Indeed, activity characterized by intermittent bursts of impulses was encountered only rarely, unlike that found in caudal efferents of the dragonfly nymph, which are also concerned with phasic respiratory and locomotory muscles.

The effects of transecting pathways from higher centres were not as clear-cut as in the insect where a removal of local inhibition was sometimes observed. The majority of fibres showed only a transient burst of activity on cutting the connectives between the fifth and sixth ganglia and no enhancement or depression of the spontaneous firing rate was noted. Very occasionally (two units) activity could be initiated following repetitive stimulation of some afferent pathway but this effect was seen in preparations with all peripheral and central connexions intact as well as in those in which the connectives were transected. Descending inhibitory influences in Crustacea, although well documented (Wiersma, 1961), may therefore not be as important locally as in the insect.

In general more difficulty was experienced in eliciting responses in a given efferent fibre in the yabbie than in the dragonfly nymph. In addition, the peripheral areas producing excitation were usually more diffuse. For example, although tactile afferents entering the ganglion in the uropod nerves N2 and N3 were the most effective in evoking motor neurone responses in N5 or N6, the manipulation of a wide area of hairs was necessary to produce a response. This contrasts with the situation found in caudal interneurones in the crayfish (Fielden, 1960) but it must be emphasized that it is not possible to determine the number of synapses between afferents and efferents and hence the role of interneurones is unknown. Many of the dissected motor units responded to both tactile and proprioceptive stimulation and the majority were affected by flexion of the telson. Some efferent fibres responded to descending sensory tracts from higher areas of the abdomen but more commonly they were excited by manipulation of the caudal lobes, though even here relatively few responded to bilateral stimulation (7 units).

Stimulation of tactile or proprioceptive endings affected the background discharge of approximately half (18 units) of the isolated motor fibres (Fig. 2 a). The response was most often seen as an acceleration of the resting discharge producing a burst of impulses which was commonly followed by a depression of the spontaneous firing rate. The frequency of impulses within the burst fell normally in the range of 20-100/sec. depending on the intensity of the mechanical stimulation. Occasionally some units showing an irregular low-frequency discharge responded with a sharp increase in frequency to 150 impulses/sec. comparable to that described for opener motor axons in the crab claw by Bush (1962). Surprisingly, inhibition of spontaneous firing following mechanical stimulation of receptors was rarely encountered, in contrast to the responses obtained in the dragonfly nymph.

Motor units without a background discharge showed similar bursts of impulses on manipulation of tactile or proprioceptive endings. These varied in their rates of adaptation and were usually more slowly adapting than were the responses of crayfish interneurones or sensory fibres (Preston & Kennedy, 1960; Hughes & Wiersma, 1960b) in that they rarely responded with only a few impulses. Several of these units were excited by bi-lateral stimulation, particularly flexion of the uropod lobes, in a manner comparable to that described by Eckert (1961) in dorsal nerve efferent fibres in the crayfish.

The most interesting effects of the electrical stimulation of afferent pathways were seen in the responses of spontaneously active fibres. These resembled those of the dragonfly nymph in that excitation was shown either by an acceleration of the resting discharge or by the interpolation of single or multiple evoked impulses within the spontaneous rhythm (Fig. 3). In marked contrast to the insect, however, synchronization of this rhythm with repetitive afferent volleys was commonly found in yabbie motor fibres. Re-setting of the background discharge at a frequency other than its natural one was readily achieved with pre-synaptic stimulation but rarely on direct stimulation of the motor axon. For example the unit shown in Fig. 3,b had a spontaneous discharge rate of 8-10/sec but could be synchronized at frequencies of 10, 30 and 40/sec. by stimulation of the ipsi-lateral connective or uropod nerves. The same unit responded with a burst of impulses to tactile manipulation of the uropod hairs. These units showing synchronization of their discharge rarely responded with more than one impulse to the evoked afferent volley. In other yabbie motor neurones the interpolated responses to stimulation did not re-set the natural rhythm though a wide variation of input frequencies was used (Fig. 2). These effects have been discussed in some detail by Preston & Kennedy (1962) for crayfish interneurones whose rhythm could also often be re-set by both synaptic and direct stimulation.

Inhibition of spontaneous activity was found in surprisingly few motor fibres although varying parameters of electrical stimulation were used. The background discharge was appreciably diminished in only six units at optimum frequencies of 30-40/sec. For example, in Fig. 3 a the larger unit was inhibited at higher frequencies of stimulation which concurrently fired a second spontaneously active fibre. Inhibition sometimes outlasted the period of stimulation, and in two preparations apparently identical units in N5 remained silent for nearly a second following flexion of the exopodite or stimulation of its nerve. The responses of yabbie motor fibres therefore differ significantly from those in the insect where central inhibition of motor discharge as found to be a common occurrence.

Units without a background discharge gave responses varying from single impulses to repetitive trains of spikes depending on the source and intensity of the afferent volley (Fig. 4). Single spike responses, or those showing only a few impulses, were more common than in the dragonfly nymph although high-frequency discharges comparable to those seen on natural stimulation were seen in some units. More than 30% of the fibres could be excited by stimulation of several ganglionic inputs but the connexions of these units did not appear to be as diffuse as in the insect. Delays varied from 1-5 to 20 msec. or more for the different pathways on to N5 and N6 motor neurones and could be shortened more significantly by an increase in the intensity of stimulation than by a change in its frequency (Fig. 4). In this respect the responses closely resemble those of crayfish interneurones and motor neurones but differ from some dragonfly efferents whose latencies were reduced by raising the frequency. Normally in the yabbie an increase in the intensity of the volley increased the length of the train, whereas an increase in repetition rate reduced the number of spikes until only single-spike responses were seen above 5/sec. and the response failed at frequencies of 20-30/sec. An increase in latency was usually evident before the response dropped out completely. Blocking therefore occurred at lower repetition rates than in the insect but comparable low-frequency failures have been described in non-giant crayfish efferents (Takeda & Kennedy, 1964). Repetitive stimulation was not without a facilitating effect on other motor units, however, and this is described in the next section.

In experiments on the dragonfly nymph it was demonstrated that many motor junctions possessed an enhanced responsiveness dependent on repetitive stimulation.

This facilitating effect, which showed various different patterns, was found in a large proportion (50-60%) of the dissected units during or subsequent to frequency increases in the range 10-100/sec. The frequency sensitivity appeared to be a peculiar characteristic of insect motor junctions and hence it was subjected to closer investigation in the yabbie. It must be emphasized that while enhanced excitability has been found at yabbie motor synapses it was of much less frequent occurrence than in the dragonfly nymph, being encountered in only ten units. These were usually facilitated at lower frequencies of 10-40/sec. but showed distinct responses which may be broadly classified in the following categories.

• i. Motor neurones, closely comparable to those found in the dragonfly nymph, which, upon increasing the repetition rate, gave a 1:1 response to volleys subthreshold at lower frequencies. This response subsequently became intermittent with fatigue, but in these units increments in the magnitude or frequency of the volley were almost interchangeable in exciting a post-ganglionic discharge (Fig. 2 c, d). At low intensities a much higher frequency was necessary to elicit a response than at high intensities, where the responses occurred at very low repetition rates. As in the insect, two units showed post-activation potentiation as they were excited by previously subthreshold volleys for a short time following a period of stimuation at higher frequency. Attempts to study the parameters of stimulation quantitatively met with the difficulty that any prior activity at the motor junction affected its responsiveness. However, plots of the relationship between intensity and frequency were made as described in the preceding paper and were almost identical with those shown for the dragonfly nymph. A similar comparison was found for plots of frequency against the latency of the response from the onset of stimulation (Fig. 5). Latencies were considerably longer at the lower frequencies and often very variable as in the insect, but shortened to a minimum at 30−40/sec.

• ii. Units which were dependent on a minimum frequency of stimulation to produce a response which was not synchronized with the afferent volley (Fig. 3 d). These units did not respond at any intensity of stimulation at frequencies below about 10/sec. and an optimum frequency of 30−40/sec. was evident. Some then produced responses of higher or lower frequency than the applied volley while others fired only a few impulses. These yabbie motor fibres possibly resembled those in the crayfish described by Hughes & Wiersma (1960a) where a direct driving of units in the first roots was found at frequencies of 30/sec.

• iii. Units comparable to those described in the above section but which showed a considerable after-discharge subsequent to stimulation (Fig. 2c, d). These after-discharges, which sometimes lasted for a period of 2−3 sec., were the most prominent feature of the response. The frequency of impulses within the discharge was sometimes as high as 300/sec. and the highest rate was achieved 0·5−1·0 sec. post-stimulation. High-frequency after-discharges have also been described in crayfish interneurones by Kennedy & Preston (1963) who discuss other post-activation changes in excitability. In two N5 units showing similar behaviour patterns in different preparations the after-discharge was followed by intermittent bursts of impulses occurring at about 15/min (Fig. 2 c). This constitutes one of the rare cases of burst activity found in a caudal motor fibre. After-discharges and intermittent bursts triggered by high frequencies were far more common in the dragonfly nymph.

In summary, an enhanced responsiveness, dependent on repetitive stimulation does exist at some yabbie motor junctions but it has been found in relatively few units and appears to be of much less significance than in the insect.

The results described here, in conjunction with those presented previously on the dragonfly nymph, show that the main characteristics of arthropod motor neurones include the presence of a background discharge, the occurrence of convergence and a variable degree of both inhibition and temporal summation/facilitation. These have been discussed in the preceding communication (Knights, 1965) and can be dealt with here mainly from a comparative viewpoint. The background discharge shows a wide variety of patterns and in both the crustacean and the insect is comparable to that described for interneurones. The discharges have been shown to be due to endogenous activity in the last ganglion and may therefore by attributed either to intrinsic activity of the motor neurone or to spontaneous pace-maker interneurones, either source presumably resulting in a tonic maintenance of muscular contraction. Continuous tonic activity is normal for arthropod slow fibres and the range of frequencies found here on excitation is compatible with that known to produce tension in the slow system (Horridge, 1965). The most notable differences between the spontaneous motor activity in the two preparations are the occurrence of paired impulses and the re-setting of the rhythm found commonly in the yabbie, and the relative lack of intermittent firing in this animal compared with the dragonfly nymph. The latter may be attributed to the fact that in the crustacean the caudal units are less involved in sustaining phasic muscular activity. The synchronization of the spontaneous discharge by afferent volleys is less easy to explain but may indicate a greater sensitivity of the autorhythmic pacemaker neurone in the yabbie (Preston & Kennedy, 1962).

Responses of ‘silent’ units were very similar in the two arthropods and again resembled those of interneurones, particularly in the crustacean where motor junctions appeared to be more dependent on spatial summation than in the insect. Their most significant difference appears to be in the relatively lower susceptibility of yabbie motor neurones to the timing of afferent impulses, both in terms of inhibition and facilitation. These effects have been discussed in some detail for Aeschna where changes in pulse repetition rate were shown to be capable of producing excitation, inhibition, after-discharge and potentiation at different motor junctions. Other investigators have described positive effects of repetitive pre-synaptic bombardment on the excitability of crustacean interneurones (Kennedy & Preston, 1963) and motor neurones (Wiersma & Ikeda, 1964; Wilson & Davis, 1965) and central inhibition in Crustacea is well documented (Bush, 1962; Kennedy & Takeda, 1965b). Comparable examples have been observed in the yabbie but they were of less frequent occurrence and appear to play a role of lesser importance than in the dragonfly nymph. In the latter it was emphasized that the variability of the relationship between the afferent response and the frequency, intensity and pathway of stimulation has a significance in allowing for a wide range of signals in the very few fibres which supply insect muscle. A similar argument applies in the case of the yabbie but in contrast to the normally dual innervation found in insects, crustaceans often have one or two additional inhibitor axons and there may be more than two motor axons. Hence there is possibly a more important role of peripheral gradation of contraction in the crustacean than in the insect where there is no evidence for inhibition at the neuro-muscular junction. It is suggested therefore that a greater percentage of neurones may require a frequencysensitive system in the central control of motor output in the dragonfly than in the yabbie. However it is interesting that spatial summation, an alternative mechanism for fibre recruitment, was more common in the crustacean. The prevalence of paired discharges in the latter animal may also provide an additional method for augmenting tension, as in the crayfish (Wilson & Davis, 1965). In either case, central mechanisms for producing a variable timing of efferent impulses have been demonstrated in these arthropods where it is well known that the neuromuscular junctions are strongly dependent on facilitation.

I would like to thank Prof. A. K. McIntyre for his constructive criticism of the manuscript.