Sensory adaptation: extracellular recording from locust wing hinge stretch receptor

R. MELDRUM ROBERTSON Department of Biology, Queen’s University, Kingston, Ontario K7L 3N6, Canada

Robertson, R. Meldrum. Sensory adaptation: extracellular recording from locust wing hinge stretch receptor. Am. J. Phys- iol. 263 (Adu. Physiol. Educ. 8): S7-Sll, 1992.-Good student laboratory exercises that do not require much manipulative or technical expertise of the student and that have minor equip- ment demands are hard to find. One experiment that has these desirable characteristics is the description of adaptation of the firing frequency of the locust forewing stretch receptor after elevation of the wing. Unambiguous recordings of the activity of the stretch receptor can be made using a simple monopolar hook electrode inserted into the thoracic cavity of a decapitated lo- cust. Elevation movements of the forewing are simple to per- form and measure. The response of the stretch receptor as a function of time and the stimulus history is monitored. Within a relatively short time it is possible to collect enough data to characterize thoroughly the adequate stimulus of a single sen- sory neuron. There is considerable scope for student innovation, and several important concepts of sensory physiology can be discussed.

teaching; student laboratory; proprioception; mechanoreceptor; insect

EVEN A BRIGHT AND DEDICATED student can become daunted and dispirited by laboratory exercises that re- quire excellent manipulative skills and/or the use of much highly sophisticated equipment that can take months to learn to operate effectively. Too often stu- dents have to stand and watch while a teaching assistant fiddles with the setup and extols the virtues of the par- ticular preparation that has again failed to work. Yet without such attributes of difficulty, many physiology laboratories, especially those investigating phenomena at the cellular level, can become trivial and boring. What is required is an exercise that illustrates important phys- iological principles, requires little exotic equipment, can readily be set up by the most ham-fisted individual, uses a preparation that is cheap, robust, and long-lived, and provides a good scope for extended data collection and independent inquiry. Several of these requirements can be met by using invertebrate preparations (4), and I describe here a preparation of the locust that investi- gates the sensory physiology of a single proprioceptive neuron responsive to elevations of the forewing and that, to my mind, provides an excellent student labora- tory exercise.

The preparation was first described to me by A. N. Spencer (Department of Zoology, University of Alberta). I subsequently incorporated it into a laboratory course in Neurobiology offered in the Department of Biology, McGill University, which I taught from 1984-1988, and into the laboratory component of a course in Compara- tive Animal Physiology offered in the Department of Biology, Queen’s University, which I have taught since 1990. It has been a successful addition to these courses. It has the advantages of needing little equipment while enabling the activity of a single proprioceptive neuron to be monitored with ease. Natural stimulation of the neu-

ron is straightforward and easily quantified. Moreover, the preparation is robust and can last for 12-24 h with- out apparent deterioration. Indeed a single preparation has been known to serve for three different 3-h labora- tory sections (morning, afternoon, and evening) of a class.

I provide here a version of the student handout that I have used, some examples of the results that can be obtained, and some hints that may be useful.

THEHANDOUT

Introduction. Movements and deformations of each of a lo- cust’s wings are monitored by several sense organs located ei- ther on the wing or at the wing hinge. The function of the sense organs is to compensate for externally imposed perturbations of wing movement during flight and for internal variations in the motor system (2). They are also involved in generating the motor pattern by controlling the phase of activity of different muscle groups (10) and, more importantly, by controlling the wingbeat frequency in a cycle-by-cycle fashion (11, 15, 17). One of these sense organs is the wing hinge stretch receptor (SR). At each wing it comprises a single multipolar sensory neuron in- nervating a connective tissue strand (6). This strand spans skel- etal elements at the wing hinge that move relative to each other as the wing is elevated or depressed. The strand is stretched by elevation of the wing, causing an increase in the rate of firing of the sensory neuron (6,9). This information is transmitted to the central nervous system via the axon that runs in nerve 1 of the mesothoracic ganglion (forewing SR) or the metathoracic gan- glion (hindwing SR) (16). The axon of the forewing SR has extensive branching throughout the three thoracic ganglia where it has monosynaptic connections with motor neurons (1, 12) and interneurons (14) in the flight system. Although the dynamic response characteristics of this sense organ, i.e., to sinusoidal movements of the wing at approximately the wing- beat frequency (7, 8), are possibly of greater significance to the animal, in this experiment you will be investigating the static response characteristics of the sense organ to passively imposed elevations and depressions of the wing (9).

Preparation. You will be provided with a locust. First remove the legs by cutting them off at the level of the coxae. Next cut off the head. Grasp the gut with a pair of forceps and pull as much as possible out of the animal. The aperture must be cleared of obstructing air sacs. The animal is then set up as depicted in Fig. 1. Fix the locust to the stand provided @/&in. rod on a magnetic base) using Plasticene (or wax). Ground your prepa- ration with a silver wire inserted into the abdomen and con- nected to the ground socket of the preamplifier. You are pro- vided with a simple device for moving a forewing and measuring the angle through which it has been moved. Clamp this onto a stand and arrange it so that the axis of the rotating arm is in a straight line with the hinge of one of the forewings, the protrac- tor is at right angles to the longitudinal axis of the locust, and the 90” line of the protractor is horizontal. Carefully extend the forewing that you have chosen to use. Place it over the narrow wing support that protrudes from the arm, and fix it in place with a narrow piece of Scotch tape.

It is important to ensure that the locust and the device are lined up correctly so that movements of the arm smoothly ele- vate and depress the wing without twisting it or bending it

1043-4046/92 $2.00 Copyright 0 1992 The American Physiological Society S7

S8 LOCUST STRETCH RECEPTOR

Fig. 1. Diagram of experimental setup. Decapitated locust is mounted on a rod such that a silver wire electrode can be inserted into thoracic cavity through neck. Forewing is extended laterally from thorax and fixed to arm of device that can elevate the forewing through measured angles. Animal is grounded via a silver wire inserted into abdomen.

unduly (this will excite other receptors causing reflex motor activity that will obscure the extracellularly recorded SR spike), and so tha .t a reasonably accurate measure of the angle through which the wing has been moved can be read off the protractor.

To record activity from the nerve trunk that contains the axon of the forewing SR (mesothoracic nerve I), you will use a monopolar silver wire hook electrode. The glass tubing around the silver wire serves to support it and insulate it to some extent. Try not to break it. The electrode should be connected to the Gl input of the preamplifier. Because you are recording with a monopolar electrode, the amplifier is used in its single-ended configuration, and the G2 input is connected to ground.‘A con- necting loop is provided for you to do this. The output from the amplifier is connected to one of the channels (AC coupled) of the oscilloscope so that the voltage signals can be displayed. Fit the rod holding the electrode into a micromanipulator, and ar- range things so that you can extend the electrode horizontally into the body cavity of the locust. The procedure now is to search for SR activity using the electrode. Figure 2 will give some idea of where you should direct your search.

Move the electrode laterally onto the muscles of the wing you - - are working with. Is there activity? (The SR is usually firing tonically at - 10 impulses/s when the wing is extended straight out from the body at OO.) If not, then move the electrode. Now is there activity? If not, move the electrode again, etc. From time to time you could elevate the wing to stimulate the SR in case it is not firing spontaneously (this is rare). Set the pream- plifier to amplify at xl00 with the high- and low-frequency cutoffs at 10 kHz and 300 Hz, respectively. This will give you a better chance of seeing the SR spike. The extracellularly re- corded action potential (Fig. 3) is triphasic (Why?). Once you have found the nerve and are recording large enough SR spikes (i.e., about 3 times the amplitude of the background noise), you can prevent the animal from desiccating by building a wall of Vaseline-mineral oil mixture at the entrance to the body cavity. This is done with a syringe and a hypodermic needle. Be careful not to knock your electrode out of position. The preparation should now last for as long as you need it. The wing hinge is reasonably robust; however, you should not try to elevate or depress the wing much past the 45” from horizontal positions.

Procedure. You are now in a position to investigate the re- sponse characteristics of the SR. Starting with the knowledge that the adequate stimulus for the SR is elevation of the wing,

Fig. 2. Diagram illustrating recording position. The axon of stretch receptor travels in nerve 1 of mesothoracic ganglion (meso Nl; see Ref. 3 for anatomy) from base of forewing to ventral nerve cord. Electrode has to be placed close enough to this nerve to pick up the extracellular field potentials of stretch receptor action potentials.

how exactly does the SR respond to such elevation? Does the response adapt? Does it habituate? Can you document this graphically? What information is coded for in the response?

SAMPLE RESULTS

If the electrode is close enough to nerve 1 it is possible to record SR spikes with little or no ambiguity. It is the largest sensory axon in the nerve trunk, thus having a relatively large extracellularly recorded spike, and it usu- ally fires at a fairly constant frequency (- 10 impulses/s). Activity of the have axons in

dorsal longitudi the same nerve,

nal wil

motor 1 have

neurons, which larger extracel-

lular sp neously

.kes, bu t these motor neurons active. A convenient method

are of

rarely sponta- measuring the

frequency of firing is to trigger the oscilloscope sweep internally from the SR spike and measure the interspike interval (ISI) [ l,OOO/ISI (in ms) = instantaneous fre- quency (in impulses/s)]. For this experiment the measure is of spike frequency while the stimulus parameters are elevation angle, time after stimulus, and stimulus history. Elevation of the forewing causes an immediate and dra- matic increase in the frequency of firing of the SR that is dependent on the extent of elevation (Fig. 4). The ini- tially high frequency adapts over 2-4 min (Fig. 5) to a final frequency that is characteristic of the maintained angle. These results are depicted graphically in Figs. 6 and 7. Pabst (9) describes two periods of adaptation: dur- ing the first second of activity and during the subsequent few minutes. Without specialized equipment it is not pos- sible to document the of the forewing from

initial rapid adaptation. Depression 0” (horizontal) has little effect on

firing frequency (not shown). More interesting is that a return to 0” from elevated angles results in variable peri- ods of silence of the SR (e.g., 2 min on return from 40”, 1

Fig. 3. Extracellular recording of an action potential of the stretch re- ceptor. Note that it is triphasic with a major negative peak (-0.3 mV) because cellular

nature of the space.

of the monopolar recording in unrestricted extra-

LOCUST STRETCH RECEPTOR s9

10" / O.lmV

min on return from 30”, etc.; not shown) followed by a gradual return to the resting frequency. Enterprising stu- dents could investigate 1) hysteresis effects on the fully adapted firing frequencies of different angles of elevation, 2) whether equal increments of elevation cause the equal increments of initial firing frequency independent of po- sition in the range, 3) whether the rate of elevation affects final frequency, or 4) whether the SR continues to fire if the wing is folded.

TIPS AND SUGGESTIONS

The major difficulty in setting up this experiment is finding the correct position for the recording electrode. It is essentially a blind search with little indication that one is getting closer or further away. Nevertheless, it is my experience that with some patience most students will get a satisfactory recording within ~0.5 h. It is good to re- member that the locust is bilaterally symmetrical and

Fig. 4. Activity of stretch receptor after elevation of forewing. Each trace pre- sents 200-ms sample of firing of stretch receptor -2 s after elevating forewing from 0” position (extended horizontally from thorax) to 10,20, 30, and 40”. Each trace has been triggered at position of the only action potential recorded in bottom trace. Note the increase in frequency as a result of increasing the angle of elevation.

J

2Oms

that after searching for a while on one side it is a rela- tively simple matter to switch to the opposite side. I have also found that a small proportion of animals yield noth- ing in spite of the most dedicated efforts. It is better to discard these and dissect a new animal than to waste time with recalcitrant specimens. The size of the signal can be improved by gently drying the interior of the thorax with a Kimwipe before inserting the electrode. This mitigates the problem of shorting the electrode to ground through the hemolymph.

Any motor activity picked up either in the nerve trunk or as electromyographic signals from active muscles will mask the SR activity. This can be a problem in two ways. First, some animals are particularly sensitive and will attempt to fly, especially in response to auditory stimu- lation (sudden, high-pitched noises, such as hissing or rustling, in the laboratory). This problem can be allevi- ated by ablating the ears. Second, in some preparations,

Fig. 5. Adaptation of activity of stretch receptor after elevation of forewing from 0 through 40”. Each trace presents a 100-ms sample of firing of stretch recep- tor ~5, 10, 15, and 30 s after elevating the forewing. Each trace has been trig- gered at position of the first action po- tential recorded in bottom trace.

0.1 mV

10ms

SlO

& 7 W

g 50

E LL-

c-7 7 -

E 25

0

v - 10 degrees

l - 20 degrees

v - 30 degrees cl - 40 degrees

LOCUST STRETCH RECEPTOR

1 I I I I I I 1

0 30 60 90 120 150 180 210 240

TIME (seconds)

Fig. 6. Adaptation of firing frequency (in Hz, or impulses/s) of stretch receptor after elevating forewing through different angles. Adaptation is complete after -3 min. Note the initial high rate of firing, which decays to a final value that is dependent on angle through which wing is elevated.

XT 75 I

w

z 7 W

c-7 7 -

E 25

0

ELEVATION (degrees)

Fig. 7. Initial (after 5 s) and adapted (after 2 min) firing frequency of stretch receptor as a function of angle through which forewing has just been elevated. Activity of stretch receptor contains information about movement and position of forewing.

raising the wing causes reflex motor activity. This is usu- ally because the wing and lifting device are not arranged correctly and raising the wing is also bending, twisting, pushing, or pulling it. Numerous other sense organs will be stimulated by this and generate the reflexes. Proper care and attention to the alignment of the preparation usually avoids this problem. Another way is not to stick the wing to the support with the tape but simply to use the tape to provide a channel that the wing can slide through. Thus slight misalignments will cause the wing to

slide over the support rather than deforming it or the wing hinge.

In more than one-half of the preparations that I have witnessed, maintained elevations ~40” will cause a rapid diminution in the firing rate of the SR to values around the resting value or even to zero firing rate. It seems as if the geometry of the strand and hinge is such that at elevations MO” the strand suddenly becomes relaxed to relieve all stretch in it. Doing this has never appeared to damage the organ, and subsequent performance has been normal.

I have always used specimens of Locusta migratoria. However, I am sure that any species of large locust or grasshopper will work equally wel1.l Crickets and cock- roaches undoubtedly have SRs; however, I have not tried to use them, and I would be hesitant to do so because of the relative lack of open space in the thorax in which to search with the electrode. Furthermore, I have no knowl- edge of whether the organs in these other insects have a spontaneous firing rate; such tonic firing makes the search quicker.

The setup can easily be made slightly more sophis- ticated in two ways. First, movements of the forewing could be driven using a pen motor and a function gen- erator. This would allow one to investigate the dynamic properties of the SR in a controlled fashion, for example by moving the forewing sinusoidally at about the wing- beat frequency. Second, it is easy to mount the pro- tractor and elevating arm on the spindle of a variable resistor. Wiring this to a battery allows the elevation angle to be monitored as a voltage drop across the vari- able resistor.

Pfau et al. (13) have shown that the response of the SR is temperature dependent. Also it is well known that locusts are poikilothermic and have little control of tho- racic temperature during flight. Indeed, during flight thoracic temperature can often exceed ambient by as much as lO”C, and this affects the wingbeat frequency (5). Investigation of the effect of temperature on firing frequency and adaptation rates would be relatively simple

l The locusts I have used were obtained from breeding colonies of Locusta migratoria maintained for research either by myself at McGill University, Montreal, Quebec, or by Dr. G. Wyatt at Queen’s University, Kingston, Ontario. I have no knowledge of the difficulty or ease of obtaining a supply of locusts or other large grasshoppers out- side of North America. Within North America, as far as I am aware, the usual biological supply companies do not supply live grasshoppers or locusts. This leaves the options of catching wild specimens of ap- propriate species when they are available, obtaining specimens from established breeding colonies used for research, or establishing a small breeding colony for teaching purposes. In the area around Kingston, Ontario, a suitable endemic alternative would be the Carolina locust, Dissosteira Carolina. However, this is available only during the late summer months, and Canadians with limited needs would be better advised to contact someone with a research colony of either L. migra- toria or one of the Schistocerca species. L. migratoria is not available in the United States and cannot be imported. However, there are numerous research colonies of Schistocerca scattered throughout the United States. The best way to find them may be to contact the authors of publications using the desired species. Modest quantities will usually be supplied willingly from these sources. For larger num- bers it may be worthwhile establishing a small breeding colony with some seed stock, assuming the proper authorization can be obtained. Such a colony is relatively easy and inexpensive to maintain and has the advantage that the locusts can be used for numerous other labo- ratorv exercises.

LOCUST STRETCH RECEPTOR Sll

using a 250-W heat lamp (supplied for chick brooders) and a small thermocouple inserted into the thorax.

CONCLUSION

In my opinion this is an excellent laboratory exercise because it fulfills most of the criteria outlined in this paper’s introduction. Much can be done with it, and the handout can be tailored to suit particular levels of stu- dent. I have tended to use limited direction of procedure so that the students have to think more about doing an experiment to discover how the organ works rather than completing a list of instructions. The latter approach could be used for less advanced students.

I thank Andy Spencer for suggesting this laboratory exercise to me many years ago. I also thank Chris Gee and Jack Gray for their com- ments on a previous version of this manuscript.

The author’s research on locust flight is funded by the Natural Sci- ences and Engineering Research Council of Canada and by the Faculty of Graduate Studies and Research at Queen’s University.

An abstract of this material has been published (Sot. Neurosci. Abstr. 17: 516, 1991).

Address reprint requests to R. M. Robertson.

Received 29 June 1992; accepted in final form 13 August 1992.

REFERENCES

1. Burrows, M. Monosynaptic connexions between wing stretch receptors and flight motoneurones of the locust. J. Exp. Biol. 62: 189-219, 1975.

2. Camhi, J. Neuroethology. Sunderland, MA: Sinauer Associates, 1984, p. 345-353.

3. Campbell, J. I. The anatomy of the nervous system of the me- sothorax of Locusta migratoria migratorioides R and F. Proc. 2001. Sot. 137: 403-432, 1961.

4. Deyrup-Olsen, I., and T. M. Linder. Use of invertebrate an- imals to teach physiological principles. Am. J. Physiol. 260 (Ada.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

Physiol. Educ. 5): S22-S24, 1991. Foster, J. A., and R. M. Robertson. Temperature dependency of wing-beat frequency in intact and deafferented locusts. J. Exp. Biol. 162: 295-312, 1992. Gettrup, E. Thoracic proprioceptors in the flight system of lo- custs. Nature Lond. 193: 498-499, 1962. Mohl, B. High-frequency discharge of the locust wing hinge stretch receptor during flight. Naturwissenschaften 66: 158-159, 1979. Mohl, B. The role of proprioception in locust flight control. II. Information signalled by forewing stretch receptors during flight. J. Comp. Physiol. 156: 103-116, 1985. Pabst, H. Elektrophysiologische Untersuchung des Streckrezep- tors am Flugelgelenk der Wanderheuschrecke Locusta migratoria. 2. Vgl. Physiol. 50: 498-541, 1985. Pearson, K. G., and J.-M. Ramirez. Influence of input from the forewing stretch receptors on motoneurones in flying locusts. J. Exp. Biol. 151: 317-340, 1990. Pearson, K. G., D. N. Reye, and R. M. Robertson. Phase- dependent influences of wing stretch receptors on flight rhythm in the locust. J. Neurophysiol. 49: 1168-l 181, 1983. Peters, B. H., J. S. Altman, and N. M. Tyrer. Sy- naptic connections between the hindwing stretch receptor and flight motor neurones in the locust revealed by double cobalt la- belling for electron microscopy. J. Comp. Neurol. 233: 269-284, 1985. Pfau, H. K., U. T. Koch, and B. Mohl. Temperature depen- dence and response characteristics of the isolated wing hinge stretch receptor in the locust. J. Comp. Physiol. 165: 247-252, 1989. Reye, D. N., and K. G. Pearson. Projections of the wing stretch receptors to central flight neurons in the locust. J. Neu- rosci. 7: 2476-2487, 1987. Reye, D. N., and K. G. Pearson. Entrainment of the locust central flight oscillator by wing stretch receptor stimulation. J. Comp. Physiol. 162: 77-89, 1988. Tyrer, N. M., and J. S. Altman. Motor and sensory flight neurones in a locust demonstrated using cobalt chloride. J. Comp. Neurol. 157: 117-138, 1974. Wendler, G. The influence of proprioceptive feedback on locust flight coordination. J. Comp. Physiol. 88: 173-200, 1974.