Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Electrophysiological Experiments on the Isolated Sur- viving Labyrinth of Elasmobranch Fish to Analyze the Responses to Linear Accelerations OTTO E. LOWENSTEIN University of Birmingham, England SUMMARY The classical assumption that semicircular canals respond exclusively to angular acceleration and cannot, therefore, be involved in the elicitation of responses to linear acceleration has recently been challenged by a number of observations. Experiments are described in which the isolated surviving labyrinth of elasmobranch fishes (dogfish and ray) was subjected to linear acceleration. Recordings from horizontal and vertical canals as well as from the utriculus and lagena indicate that semicircular canalsâunder these physiological conditions (interruption of blood supply and opening into the perilymphatic space)âdo yield responses to tilting and to rotating vectors of linear acceleration which resemble those obtained from the otolith organs. As compared with the latter, however, the responses from the semicircular canal have a significantly higher threshold. INTRODUCTION There is no need for a detailed restatement of the classical concepts concerning the distri- bution of function between semicircular canals and otolith organs. Whereas it is generally assumed that the semicircular canal is stimulated exclusively by angular acceleration and pro- vides the central nervous system with integrated information relating to angular velocity at any given moment, the otolith organ, although capa- ble of being stimulated by changes in angular velocity, is chiefly concerned with the monitor- ing of linear accelerations in the form of gravita- tional, centrifugal, translatory, and oscillatory stimuli, the last belonging to the field of vibration and sound. The assumption that semicircular canals are exclusively stimulated by angular acceleration, and not at all by linear acceleration, has been challenged in the past, but has recently come under renewed and serious scrutiny in the light of a number of experiments in which it was found that responses to angular acceleration, i.e., perrotatory and postrotatory nystagmus, per- rotatory and postrotatory impulse activity chiefly in second-order neurons in vestibular centers, as well as the subjective experience of human ob- servers, are significantly modified by centrifugal effects and by simultaneous changes of head position in space. For detailed references and discussion, see references 1 to 7. In a number of these cases the authors have quite rightly taken into account that these modi- fications in the canal responses could be ac- counted for by central integration of canal re- sponse with simultaneous information derived from otolithic inputs with or without a contri- bution from vertical canals. Money, Graybiel, and Johnson (see ref. 2, pp. 196 and 197) and Benson (personal communication) have tried to deal with such ambiguities by an ingenious device which makes it possible to subject the hori- zontal semicircular canal to a rotating vector of linear acceleration without any concomitant posi- tional change of the subject. However, here, 161
162 THE ROLE OF THE VESTIBULAR ORGANS IN SPACE EXPLORATION too, the possibility of interference by simul- taneous signals from vertical canals cannot be excluded (refs. 1, 8, and 9). Nor can one be sure that kinesthetic information and otolith- derived responses to centrifugal effects may not have contributed to the observed modification of perrotatory and postrotatory responses of the horizontal canals. If we are interested in the functional range of the peripheral organ as such, even if this may be considered of rather academic importance, the only conclusive method is to test and evaluate the responses of such peripheral structures in complete isolation. Material and Methods Preparations that would lend themselves to such a study in isolation are not easy to obtain. Attempts to record from first-order vestibular neurons or their nerve processes have only rarely been reliably successful, especially in birds and mammals. Gernandt (ref. 10) reports having recorded from the vestibular branch of the eighth nerve where it leaves the auditory meatus. The labyrinths of elasmobranch fishes (refs. 8, 9, and 11) and of the frog (ref. 12) have yielded much of the information on which our qualitative notions about the functional behavior of semi- circular canals and otolith organs are at present based. When the emphasis is on quantitative aspects, the study of single units may be con- sidered preferable, if not imperative, although useful results have been obtained by the analysis of the electronic integration of responses based on the recordings of massive multifiber response pictures. Single-unit peripheral preparations are rela- tively easily obtainable from the dendriti<- branches of the primary sensory neurons in- nervating the ampullae of the horizontal and anterior vertical canals of the isolated surviving labyrinth of the dogfish, Scyllium canicula, and of the thornback ray, Raja clavata. Greater difficulties are encountered in attempts to obtain such preparations from a utriculus, a sacculus, or lagena. Having developed the tech- nique for the isolation of these recording sites, I felt it incumbent on me to try to contribute to the search for the site of origin of responses to linear acceleration, especially of those that have been postulated to modify the basic response picture of the semicircular canals. It is imperative at the outset to establish under what conditions the exposure of isolated units from a semicircular canal to linear accelerations is capable of yielding truly crucial results. The isolated labyrinth preparation, although eminently manageable, is cut off from the circulatory sources of oxygen and other metabolic materials. It is true that useful response pictures can be ob- tained from such a preparation from a cold- blooded animal for prolonged periods of time, in fact for an hour or more, if the ambient tem- perature and air humidity are kept within certain limits. When deterioration becomes recogniz- able in a canal preparation, it manifests itself in ways usually compatible with the assumption that it has its origin in a change in the turgor or elasticity of the cupula. The threshold to angu- lar acceleration becomes lower, the preparation becomes highly sensitive to vibration of the substrate, and, finally, in excessively old prepara- tions, say 3 to 4 hours after isolation from the animal, the strict directionality of the response may disappear before the final disappearance of discharge activity in the nerve. This well-known sequence of events yields useful signposts to the experimenter from which he learns when the further use of a certain preparation is inadvisable in accordance with this or that objective of the experiment, especially so far as its quantitative evaluation is concerned. It must be conceded here that ideally the use of the isolated prepara- tion, as elsewhere in physiological experimenta- tion, is inadvisable for the evaluation of absolute as opposed to relative quantitative parameters. The second important consideration arises from the fact that access to the nerve strands innervating the individual labyrinthine end organs is gained by the removal of part, however small, of the cartilaginous wall of the labyrinth capsule. This results in the opening of the perilymphatic space and in a certain loss of perilymphatic fluid. As this fluid may be con- sidered to act as a shock absorber and, together with anchoring strands of connective tissue, nor- mally helps to keep the membranous labyrinth in place, such loss of support may easily con-
EXPERIMENTS ON THE LABYRINTH OF ELASMOBRANCH FISH 163 tribute to dislocations of canal and ampulla under the impact of linear accelerations. The rele- vance of this can be seen in the fact that experi- menters observing effects which they feel might be due to canal responses to linear acceleration have suspected that they might be brought about by such dislocations within the perilymphatic space, even in the intact unopened bony laby- rinth, especially when exposed to linear accelera- tion of high g-values outside the normal range with which the labyrinth is evolutionarily adapted to cope (ref. 13). Both circumstances might greatly impair the significance of results in which linear accelerations are seen either to modify the responses to simultaneously applied angular accelerations or to lead to a significant modulation of the resting discharge purely on linear acceleration, be this by gravity in positions deviating from the normal, during constant-speed rotations, or during rotations in which vectors of linear acceleration are made to sweep over the end organ. If, however, a canal preparation showing the full range of responses characteris- tically occurring during angular acceleration failed to respond to any of the above-mentioned types of linear acceleration, such behavior would have to be considered significant evidence against the susceptibility of canals to such stimuli. The ideal solution would be if it were found possible to record single-unit first-order activity of definitely known origin from the anterior ramus of the eighth nerve just outside the foramen through which it leaves the labyrinth capsule. On their passage through the foramina of the elasmobranch labyrinth capsule, the branches of the eighth nerve are packed in a highly viscous jelly which obviously serves the purpose of estab- lishing a leakproof barrier between perilymphatic space and the brain case. Centrally to this pas- sage the nerve disperses into a brain-type tissue organization which makes it quite impossible to isolate strands containing recordable single units. Moreover, a scattered assembly of cell bodies of the first-order neurons are found here, and any record taken from here could not be held to be above suspicion with regard to possi- ble synaptic cross-modification of responses originating in neighboring end organs. The only reliable pickup that can be held for protracted periods of time, especially under ex- perimental conditions of the preparation's rapid displacement in space, is the attachment of small, whittled-down nerve twigs in a forceps electrode. The application of microelectrodes such as tung- sten needles coated with insulating material up to the tip does not generally yield lasting pickup from single units, unless balanced floating micro- electrodes of the type developed by Gualtierotti (personal communication) are used. These sophisticated electrodes may in fact provide the answer to the problem, but they are difficult to make and consequently very expensive. These thoughts crystallized in part during a conversation with A. J. Benson and, with them in mind, I had the dual-purpose accelerator de- signed and built in my department by my chief technical officer, S. V. Hill, whose ingenuity has again, as on previous occasions, made a signifi- cant contribution to the work to be described. Technique The preparation can be mounted on the plat- forms in various orientations such as normal, upside-down, nose-up, nose-down, side-up, or side-down longitudinally or transversely to the direction of movement. The movement is monitored photoelectrically and transmitted via radio link to the second oscilloscope channel. The impulse discharges after pre- amplification are similarly transmitted to the first channel of the oscillograph and to a tape recorder or, if necessary, straight through an integrating frequency meter to a pen recorder. Platform I is rigidly fixed to a rotating arm at a distance of 30 cm from the center. Prepara- tions mounted on it are stimulated by rotatory eccentric torque; i.e., angular acceleration at the beginning and end of prolonged periods of constant-speed rotations up to 90 rpm with a slowest useful constant speed of 2 rpm. Again the impulse discharge from a preparation is preamplified on the platform and transmitted by radio link to the first channel of the oscillo- scope to tape recorder, frequency meter, and pen recorder. The rotation is monitored photo- electrically. Platform II (cf. ref. 2, pp. 19 6 and 197) rotates freely on its central spindle at the end of the rotat-
164 THE ROLE OF THE VESTIBULAR ORGANS IN SPACE EXPLORATION ing arm. By means of a chain drive to the center spindle, it is made to counterrotate at the same angular velocity and consequently keeps facing in the same direction during the rotation of the arm. A preparation mounted on it will therefore not be subjected to torque but to vectors of linear acceleration which sweep over it once per full cycle rotation of the platform. Here transmission by radio link was found unnecessary and takes place by wire after preamplification on the plat- form. The machine is at present installed at the Marine Biological Laboratory at Plymouth, and I am only in a position to report on preliminary results gained during a period of 3 months' experimentation. The first test to which a successful single- or few-fiber preparation of either the horizontal or vertical ampulla is subjected is a tilting test on a tilting gantry. Its responsiveness to swings in the plane of the canal and to swing about hori- zontal axes is tested and put on record. Pickup of the nerve was in all cases by means of a forceps electrode through a minimal opening in the cartilaginous capsule as near as possible to where the nerve twig joins the anterior ramus and as far as possible from the ampullary wall itself. So far, attempts at picking up fibers after removal of the jelly outside the unopened capsule have not met with success. During earlier work with this type of prepara- tion, postional responses shown by the ampullae of the horizontal or anterior vertical canal were interpreted as signs of functional deterioration. In the present series of experiments, tilting tests were carried out as soon as possible after the isolation of the labyrinth, and clear static re- sponses were found in both types of canals with disturbing regularity as early as 15 minutes after decapitation of the animal. The difference in frequency in the horizontal and in 90Â° head-up and head-down positions is striking, and we are dealing with a clear response of the organ to a gravitational stimulus; i.e., to linear acceleration (see also ref. 12). Whether this is entirely due to the interruption of the blood supply or to perilymph loss around the ampullae, or to both, cannot be ascertained. Maximum discharge frequencies are found near the 90Â° nose-down or side-down positions, with minimum in the nose-up and side-up positions. The same holds, but to a much lesser extent, for the horizontal ampulla. There is a distinct possibility that these responses may be the result of a dislocation forward of the whole membranous labyrinth as a consequence of the disturbance in the anchorage and/or loss of perilymph associated with the opening made in the wall of the labyrinth capsule. On the other hand, I have recently observed similar static responses from the ampullae of the un- opened cyclostome labyrinth. PRELIMINARY RESULTS The account of the preliminary findings in the first 3 months of the present series of experi- ments may conveniently start with the response picture obtained from otolith organs, i.e., utric- ulus and lagena, to stimulation with the rotating vector. The strong increases and decreases in the discharge activity are confined to a certain direction of the vector, each wave in the inte- grated total voltage occurring once per revolution of the stimulator and in a fixed sector in space. The utriculus preparation was mounted in the normal horizontal position; the lagena prepara- tions were also mounted horizontally. Thresh- olds of such responses from otolith organs were found to vary, but in some instances were as low as 2 rpm or below 0.01 g. This then is what one has to look out for in search of a response to linear acceleration. Turning now to the semicircular canals, we are faced with a puzzling situation. Although, with the preparative methods used so far, a large proportion of the vertical ampullae and somewhat fewer of the horizontal ones proved to be position- sensitive immediately after isolation and there- fore to react to gravitational stimulation, only a small percentage have so far yielded significant responses to stimulation by a rotating linear vector. The threshold was high and only in one case did the response amount to much at 30 rpm (0.3 g). In the material analyzed so far, I have not found an impressive response of this kind from a vertical ampulla, despite their very strong positional responses. This is puzzling, especially as in a number of experiments the resultant be- tween gravity and the linear vector ought to have
EXPERIMENTS ON THE LABYRINTH OF ELASMOBRANCH FISH 165 summed. I have to leave this question open at present. There are, however, not a few instances of the total absence of a response to the rotating linear vector both in horizontal and vertical ampullae. Finally, there is yet another type of situation in which linear acceleration has been claimed to have modified the response from the semicircular canal. This is claimed to happen on eccentric torque. Here the time constant of the decay of the perrotatory nystagmus is said to have been significantly altered, i.e., shortened (ref. 3), under the influence of the constant centrifugal effect on the semicircular canal during and after a period of constant acceleration followed by constant-veloc- ity rotation. This situation too can be easily re- produced on platform I of the rotary stimulator. The decay of perrotatory and postrotatory re- sponses appears to fall well within the decay periods found by Groen, Vendrik, and myself (ref. 14). I have had an opportunity to test one and the same preparation quantitatively during centric and eccentric rotation. This would obviously be the necessary condition for an as- sessment of the significance levels of any differ- ences in the time constant under these conditions of stimulation. As yet no significant difference has been observed at 45 rpm. CONCLUSIONS' The important task now is to elaborate a technique of recording from the unopened laby- rinth as soon as possible after its isolation from the animal. It will be seen whether under these conditions the rather puzzling effect of spatial orientation on the basic discharge level of semi- circular canal units can be minimized or com- pletely prevented. In such preparations, linear accelerations of any kind should be without effect, at least atg-values. REFERENCES 1 The results discussed in this paper are tentative and un- published. The illustrations shown at its presentation are therefore not included in this publication. 1. CRAMPTON, GEORGE H.: Does Linear Acceleration Modify Cupular Deflection? The Role of the Vestib- ular Organs in Space Exploration, NASA SPâ115, 1966, pp. 169-184. 2. GUEDRY, FRED E., JR.: Influence of Linear and Angular Accelerations on Nystagmus. The Role of the Ves- tibular Organs in Space Exploration, NASA SP-115, 1966, pp. 185-196. Discussion, pp. 196-198. 3. BENSON, ALAN J.: Modification of Per- and Post-Rota- tional Responses by the Concomitant Linear Accelera- tion. The Role of the Vestibular Organs in Space Exploration, NASA SP-115, 1966, pp. 199-213. 4. LANSBERG. MARTIN P.; GUEDRY, FRED E., JR.; AND GRAYBIEL, ASHTON: The Effect of Changing the Re- sultant Linear Acceleration Relative to the Subject on Nystagmus Generated by Angular Acceleration. Aerospace Med., vol. 36, 1965, pp. 456-460. 5. BENSON, ALAN J.; AND BODIN, M. A.: Interaction of Linear and Angular Accelerations on Vestibular Re- ceptors in Man. FPRC Report 1243. Ministry of Defence (Air), London. 1965. 6. BENSON, ALAN J.; GUEDRY, FRED E., JR.; AND JONES, G. MELVILLE: Responses of Lateral Semicircular Canal Units in Brain Stem to a Rotating Linear Acceleration Vector. J. Physiol., vol. 191, 1967, pp. 26P-27P. 7. BODIN, M. A.: The Effect of Gravity on Human Vestib- ular Responses During Rotation in Pitch. J. Physiol., vol. 196, 1968, pp. 74P-75P. 8. LoWENSTElN, O.; AND SAND, A.: The Individual and Integrated Activity of the Semicircular Canals of the Elasmobranch Labyrinth. J. Physiol., vol. 99, 1940, pp. 89-101. 9. LoWENSTElN, O.; AND SAND, A.: The Mechanism of the Semicircular Canal. A Study of the Responses of Single-Fibre Preparations to Angular Accelerations and to Rotation at Constant Speed. Proc. Roy. Soc. Med. B, vol. 129,1940, pp. 256-275. 10. GERNANDT, B.: The Effect of the Centrifugal Force Upon the Nerve Discharge from the Horizontal Canal. Acta Physiol. Scand., vol. 21,1950, pp. 61-72. 11. LOWENSTEIN. O.: AND ROBERTS, T. D. M.: The Equilib- rium Function of the Otolith Organs of the Thornback Ray (Raja clavata). J. Physiol., vol. 110, 1949, pp. 392-415. 12. LEDOUX, A.: Activite Electrique des Nerfs des Canaux Semicirculaires du Saccule et de 1'Utricle chez la Grenouille. Acta Oto-Rhino-Laryngol., vol. 3, 1949, pp. 335-349. 13. MONEY. K. E.: SOKOLOFF, M.; AND WEAVER, R. S.: Specific Gravity and Viscosity of Endolymph and Perilymph. The Role of the Vestibular Organs in Space Exploration, NASA SP-115, 1966, pp. 91-98. 14. GROEN, J. J.: LOWENSTEIN. O.; AND VENDRIK, A. J. H.: The Mechanical Analysis of the Responses From the End-Organs of the Horizontal Semicircular Canal in the Isolated Elasmobranch Labyrinth. J. Physiol., vol. 117, 1952, pp. 329-346.
166 THE ROLE OF THE VESTIBULAR ORGANS IN SPACE EXPLORATION DISCUSSION Money: Did I see what you called a torque response at the Lowenstein: No. beginning of the counterrotating stimuli? w . â. . Anliker: We are engaged in the mechanical analysis of l.owenstem: Ihe torque response was a control expen- . . , , the semicircular canals, and we predict that the semicircular ment. I wanted to see how the preparation responds to . , , ... . canals not only respond to linear acceleration but also to torque in the plane ol the canal. ,. . TL i r L â¢ â¢) circular translation or counterrotation. If we have lime. 1 Money: 1 see. 1 hen that was not part ol the picture: ..... , , could show how this works. Lowenstem: No: it was not part of the picture. Money: I thought it was the acceleration of the counter- Lowenstein: I was told about this by Dr. Anliker. and il rotation. sounded convincing at first hearing.