EFFECTS OF'ANTID‘RO'MIC Activity-,1? .7 _ f m, GUSTATORY NERVE Hams ONTASTE RECEPTORS or- THE mos rescue ; - _ A Thesis for the Degree of Ph. ‘D. MICHIGAN STATE UMVERSGTY FRANCIS A. KUTYNA. J‘R.‘ * 1973 ' LIBRARY Michigan State University This is to certify that the thesis entitled Effects of Antidromic Activity in Gustatory Nerve Fibers on Taste Receptors of the Frog Tongue presented by Francis Anthony Kutyna: Jr. has been accepted towards fulfillment of the requirements for Ph.D. degree in Physio logz _ZE;L;A1/¥,_Zg14vw4/uA:__ Major professor Date May 3, 1973 0-7639 2:: ‘ V ‘ emome BY 3 MM; & sous ; 800K amateur me. it Llsmnv BINDERS A 39394! 992?. MICHIGA] l ABSTRACT EFFECTS OF ANTIDROMIC ACTIVITY IN GUSTATORY NERVE FIBERS ON TASTE RECEPTORS OF THE FROG TONGUE BY Francis A. Kutyna Jr. It has been previously shown that interactions between individual gustatory units with overlapping peripheral recep- tive fields modify the stimuluSrreSponse functions of these taste fibers. Antidromic activity in peripheral collaterals of taste fibers can produce lateral interactions between sensory units innervating a common fungiform papilla on the frog tongue. Antidromic electrical stimulation of the lingual branch of the IX nerve of the frog was conducted while recording intracellular potentials of taste disc cells in order to assess the possible role of antidromic sensory fiber activity in the modification of receptor cell bioelectric properties. Antidromic activation of sensory fibers resulted in depolarization of cells of the papilla surface and hyperpolari- zation of the subsurface receptor cells. These potential changes exhibited latencies greater than 1 second which could not be ascribed to the conduction times of any of the fibers in the IX nerve. They also showed summation, adaptation and D... ‘ I I'D. . -|.. u I 'o-v .wu- ucvu ‘n: O (I) .-. "Pu- .- units . fiA u b.-‘ i u. ‘ml N 7.; l.‘ Francis A. Kutyna Jr. post-stimulus rebound. This rebound was of a polarity oppo- site to the initial change produced by antidromic stimulation and of the same direction as the change accompanying adapta- tion. Depending on stimulus conditions antidromic activity was able to produce depression or enhancement of chemosensory fiber discharge in response to taste stimuli. Different stimuli were found to have potentiating or depressing effects on the antidromically elicited potential changes of taste disc cells. The results of these experiments favor the model of lateral interaction between sensory units of the tongue as functionally mediated by bioelectric effects on the receptor cells consequent to antidromic activity in taste fibers inner- vating them. EFFECTS OF ANTIDROMIC ACTIVITY IN GUSTATORY NERVE FIBERS ON TASTE RECEPTORS OF THE FROG TONGUE By. gar Francis Af‘Kutyna Jr. A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Physiology 1973 f (A .fi-uzo "L,‘1 / “‘3 ACKNOWLEDGMENT An encomium in appreciation for the guidance, instruc— tion and friendship unselfishly given by my advisor is a humble return for the years of conscientious and exuberant support I have received from Dr. Rudy A. Bernard throughout the period of personal and professional association which led to the completion of this thesis. As a man of nous and per— spicacity he directed my development as a person and scienv tist, not only aiding but also forming the very character of my attainment. I look forward to a lifetime of friendship and association with him and his family. To my graduate guidance committee I extend my appreci- ation for their excellence and their concern in dealing with the exigencies I have often submitted to them. Drs. J. Hoffert, J. I. Johnson, E. P. Reineke and L. Wolterink have served to cultivate and, by their example, instill in me the foresight that is the characteristic of a good preceptor. The kind aid of Dr. P. Fromm, called upon with such short notice,.contributed much to the final form of this thesis. The guardianship and encouragement of Dr. E. P. Reineke has been especially beneficent throughout the stress encountered in the past years. His sagacious character is one which I will always strive to emulate. ii I am indebted to my friends, Drs. T. Adams, K. Holmes, M. Kadekaro, W. S. Hunter, and M. Morgan for their aid. Often bordering on thaumaturgy, it contributed much to the commence- ment and conclusion of this work. Suffice it to say that they have often rescued this researcher by their sapient preteri- tion of my demerits as well as their discernment of my attributes. To my immediate co-workers D. Samanen and W. Kryda I offer an obeisant thanks for their unselfish and efficacious support. My friends and associates who have in many and indescrib- able ways contributed to the conciliation of this author's aspirations with his abilities are not all named here due to limitations of space. I am confident that they will recognize my heartfelt thanks and indebtedness for their loyalty. This work is dedicated to my family, who have held unswerving faith in me throughout the years, to my grandfather who instilled in me the love of science, and to my grandmother who encouraged its pursuance. iii I. II. III. TABLE OF CONTENTS LITERATURE REVIEW. . . o . . . . . . Anatomy of Frog Tongue Taste Receptors Fungiform Papillae . . . . . . Innervation. . . . . . . . . . Taste Disc . . . . . Nerve Branchings . . . . . . . Receptor-Nerve Junctions . . . Receptor-Receptor Junctions. . Receptor-Goblet Cell Junctions Goblet Cell-Goblet Cell Junctions. Frog IX Nerve Responses to Tongue Stimulation Chemosensory Nerve Fiber Activity. Tactile and Thermal. . . . . . Peripheral Interactions of Stimuli . . . . Temperature Effects. . . . . . Pharmacologic Actions on Chemosensory Discharge . . . . . . . . . Sympathetic Enhancement. . . . Autonomic Effects on Taste . . Taste Disc Influence on Nerve Function . Single Gustatory Fiber Sensitivities . . Autonomic Effects on Peripheral Receptors . Intracellular Recordings From Taste Receptors Receptor Resting Membrane Potential. . Receptor Potentials. . . . . . Extracellular Potentials . . . Effects of H20 and Quinine . . Taste Receptor Transmembrane Conductance Active Transport Electrogenesis . Hyperpolarization. . . . . . . Antidromic Mechanisms . . . . . . Single Fiber Effects . . . . . Antidromic Motor Effects . . . Antidromic Activity Effects Across STATEMENT OF PROBLEM . . . . . . . . STATEMENT OF OBJECTIVES. o . . . . . iv 0 O 0 O Page \DKOKOQQOJNl—‘H H HHHHH wwwHH 15 16 17 17 17 18 20 20 21 24 25 27 28 28 31 31 33 33 36 39 TABLE OF CONTENTS--Continued Page IV. MATERIALS AND METHODS. . . . . . . . . . . . . . 40 Experimental Animals . . . . . . . . . . . 40 Electrical Nerve Stimulation . . . . . . . 40 Electrical Stimulation of Fungiform Papillae. . . . . . . . . . . . . . . . 41 Chemical Stimulation . . . . . . . . . . . 42 Chemical Solutions . . . . . . . . . . . . 42 Glossopharyngeal Sensory Nerve Recording . 44 D. C. Microelectrode Recording . . . . . . 45 Manipulation of Cell Membrane Potential. . 49 V. EXPERIMENTAL PROCEDURES. . . . . . . . . . . . . 51 Glossopharyngeal Nerve Stimulation and Taste Papilla Responses. . . . . . . . . . . . Sl Characterization of Sources of Slow Poten— tials in Papillae. . . . . . . . . . . . . 55 Characterization of Membrane Potential- Responses. . . . . . . . . . . . . . . . . 55 Efferent Electrical Stimulation and Afferent Sensory Activity in IX Nerve . . . . . . 56 Interaction Between Taste Stimulation and Antidromic IX Nerve Stimulation on Taste Disc Cell Potentials . . . . . . . . . . . 57 Electrical Stimulation of Taste Disc and Recording of Effects on Response of Neigh- boring Papillary Cells to Antidromic IX Nerve Stimulation. . . . . . . . . . . . . 58 Displacement of Membrane Potentials During Antidromic Stimulation of the Glosso— pharyngeal Nerve . . . . . . . . . . . . . 58 VI 0 DEFINITIONS 0 O 0 O O O O O O O O 0 O O O .- O O O 6 0 VII. RESULTS O O O O O 0 O o O O O O O O O O O Q 0 O O 64 Glossopharyngeal Nerve Stimulation and Slow Potentials in the Fungiform Papillae . . . 64 Depolarization and Hyperpolarization in Cells Of Fungifom Papillaeo o o o o ‘ o o o o o o 7]. Characteristics of Slow Potential Responses of Taste Disc. . . . . . . . . . . . 76 Antidromic Stimulation and Responses of Af- ferent Nerve Fibers to Taste Stimuli . . . 86 TABLE OF CONTENTS--Continued Intracellular Response of Taste Disc Cells to Direct Chemical Stimuli and Its Effect on Antidromically Elicited Slow Potentials. . Effect of Electrical Stimulation of Neighbor- ing Papillae on the Response of Taste Disc Cells to Antidromic IX Nerve ACtivity. . . Effect of Taste Cell Membrane Potential on the ReSponse to Antidromic Stimulation of the Glossopharyngeal Nerve . . . . . . . . VIII. DISCUSSION 0 o o o o o o o o o o o o o o o o o o Antidromic Activity in IX . . . . . . . . . . Inhibitory Effects of Antidromic Activity in IX. . . . . . . . . . . . . . . . . . Antidromic IX Nerve Activity and Taste Receptor Cell Bioelectric Changes. . . . . Taste Receptor Potentials . . . . . . . . . . Evidence for a Metabolic Component of Taste Cell Response. . . . . . . . . . . . . . Significance of This Work . . . . . . . . . . Ix. CONCLUSIONS 0 O O 0 O O O O O O O O O I O O O O O BIBLIOGRAPHY O Q 0 o o o o o o o o o o o o o o o o o 0 vi Page 88 90 98 102 102 106 109 114 116 119 122 124 FIGURE 1. 2. LIST OF FIGURES Schematic representation of peripheral taste structures in the frog tongue. . . . . . . . . . Chamber used for immobilization of frog tongue during microelectrode recording from cells of fungiform papillae and recording sensory dis- charge of lingual branch of the glossopharyngeal nerve. . . . . . . . . . . . . . . . . . . . . . Current-voltage plot for microelectrOdes of various resistances. . . . . . . . . . . . . . . Diagrammatic representation of preparation used for microelectrode recording from cells of fungi- fOrm papilla taste disc and antidromic stimula- tion of lingual branch of the IX nerve . . . . . Representation of preparation used for recording action potentials invading fungiform papilla on electrical stimulation of the IX nerve . . . . . Graphic representation of parameters character- izing temporal and scalar properties of stimulus and response records used for analysis . . . . . Three representative records of slow-wave bio— electric currents recorded from fungiform papil- lae using the gross chamber electrode illus- trated in Figure 4 . . . . . . . . . . . . . . . Positive potential recorded from surface of fungiform papilla taste disc by microelectrode . Oscilloscope tracing of 5 superimposed sweeps showing compound action potentials recorded in fungiform papilla at 3 and 4 milliseconds after electrical antidromic stimulation of the glosso- pharyngeal nerve containing sensory fibers inner- vating that papilla. . . . . . . . . . . . . . . vii Page 10 43 50 52 54 63 65 67 70 LIST OF FIGURES--Continued FIGURE 10. Depolarizing and hyperpolarizing changes in membrane potentials of two different cells in the taste disc . . . . . . . . . . . . . . . . . ll. Photomicrograph of a saggital section through a taste disc of a fungiform papilla used in micro- electrode recording. . . . . . . . . . . . . . . 12. Response potentials (Ep) of cells.penetrated on the surface.of the taste disc and adjacent non- gustatory epithelium surrounding a fungiform papilla. . . . . . . . . . . . . . . . . . . . . 13. Taste disc.ce11 response increment per anti“ dromic stimulus pulse plotted as a function of the number of stimulus pulses delivered to the IX nerve over a range of l to 80 pulses. . . . . 14. Changes in the depolarizing response potential of a taste disc cell when the IX nerve is anti— dromically stimulated with one (a), three (b) and four (c) pulses. . . . . . . . . . . . . . . 15. Typical hyperpolarizing responses to repeated stimulation of the IX nerve. . . . . . . . . . . 16. Overshoot during adaptation to a prolonged stim- ulation of the IX nerve and rebound after stimu— 1us cessation frequently seen in responses of taste disc cells . . . . . . . . . . . . . . . . 17. Frequency of gustatory nerve response to chemi- cal stimulation of the tongue and the effects of antidromic electrical stimulation of these nerve fibers on their activity . . . . . . . . . . . . 18. Change in membrane potentials (Em) of cells in the taste disc produced by application of chemi— cal solutions to the surface . . . . . . . . . . 19. Effect of prior application of sapid solutions on the amplitude of antidromically elicited hyperpolarizations of taste disc cells . . . . . viii Page 72 74 77 79 80 81 85 87 89 91 LIST OF FIGURES--Continued FIGURE 20a. 20b. 21. 22. 23. 24. 25. 26. Small depolarizations produced in a cell of the taste disc by electrical stimulation of a taste disc on a neighboring fungiform papilla. . . . . Hyperpolarization produced by antidromic Ix nerve stimulation. . . . . . . . . . . . . . . . Effect of electrical stimulation of neighboring papilla on the antidromically elicited response of taste disc cells. . . . . . . . . . . . . . . Time course of effects of neighboring papillae stimulation on the Ep and Tp of the depolarizing response to antidromic stimulation of the IX nerve. . . . . . . . . . . . . . . . . . . . . . Time course of effects of neighboring papilla stimulation on the Ep and Tp of the hyperpolar— izing response to antidromic IX nerve electrical stimulation. . . . . . . . . . . . . . . . . . . Effect of membrane potential (Em) on the re- sponse of three normally depolarizing cells to the antidromic stimulation of the IX nerve . . . Effect of membrane potential (Em) on the hyper— polarizing response (Ep) of two cells. . . . . . Representation of the pathways for orthodromic and antidromic conduction of action potentials in a single taste fiber innervating three papil- lae on the surface of the tongue . . . . . . . . ix Page 94 94 95 96 97 100' 101 103 I. LITERATURE REVIEW Anatomy_of Frog Tongue Taste Receptors Fungiform Papillae The localization of the sense of taste in the receptors of the frog tongue fungiform papillae has been suggested by anatomists (Gaupp, 1904; Hammerman, 1969; Kolmer, 1910). These receptor cells have been described as located in the specialized end disc forming the dorsal surface of the papilla (Ecker, 1889; Gaupp, 1904; Hammerman, 1969). Physiological studies have corroborated this view (Pumphrey, 1935; Kusano and Sato, 1957; Zotterman, 1949). The structures responsible for taste reception are lo- cated in the fungiform papillae of the tongue (Ecker, 1889; Gaupp, 1904; Hammerman, 1969) as well as in discrete buds dispersed throughout the epithelium of the buccal and pharyn- geal cavities. Those of the tongue are solely in the taste disc of the fungiform papillae while those located in other areas are of different morphology, and histologically resemble the lateral line receptors of the lower vertebrates (Ecker, 1889; Pumphrey, 1935). The fungiform papillae have been reported to number 500- 600 on a frog tongue (Kusano and Sato, 1957) and 400-500 by Rapuzzi and Casella (1965). Robbins reports the density of such structures to be on the average 4.9 per square millimeter of tongue surface (Robbins, 1967a). The interpapillary dis- tance is between 100 and 400 microns (Kusano and Sato, 1957; Rapuzzi and Casella, 1965). Hammerman (1969) in a histologi- cal study of their development reports the size of a single papilla to be 142-160 microns in diameter and the height above the tongue surface 200-230 microns. The papillae studied by Kusano and Sato were 100-200 micra in diameter (Kusano and Sato, 1957) and those reported by Rapuzzi and Casella (1965) averaged 100 micra across the surface of the taste disc. Innervation Ariens Kappers §£_gl. (1936) report the amphibian tongue to be innervated by cranial nerves V, VII, IX. The sensory components responsible for taste are primarily located in the IX nerve (Ecker, 1889; Gaupp, 1904; Kusano and Sato, 1957; Pumphrey, 1935; Rapuzzi and Casella, 1965; Strong, 1895), although .Robbins (1967a) mentions a contribution to the base of the tongue by cranial nerve VII. No taste components have been reported in the branches of cranial nerve V innervating the frog tongue. Autonomic fibers are described in the IX nerve of the frog (Chernetski, 1965; Herrick, 1925; Strong, 1895) but these have been considered vasomotor (Pumphrey, 1935). Chernetski (1965) has reported their origin in the anterior sympathetic ganglion. They have light, or no myelin sheaths, and their conduction velocities range from 0.2-0.6 meters per second (Chernetski, 1965) as compared to 1-15 meters per second for sensory fibers (Rapuzzi and Casella, 1965). Sensory fibers in the IX nerve each divide into 4-6 col- laterals (Rapuzzi and Casella, 1965) before entering the fungiform papillae. Each papilla receives 5-10 (Gaupp, 1904), 7-14 (Rapuzzi and Casella, 1965), 6-7 (Robbins, 1967a) afferent fibers, collaterals traveling to and innervating neighboring fungiform papillae. Rapuzzi and Casella (1965) confirm anatomical assertions (Ecker, 1889; Gaupp, 1904) that each papilla contains only one nerve fiber serving the sense of touch. Although Pumphrey (1935) reported sensitivity to touch over the entire tongue surface, Rapuzzi and Casella (1965) and Taglietti £3 31. (1969) describe it as residing only in the fungiform papillae. The nerve fibers entering the papillae lose their myelin sheaths and undergo tortuous convolutions before ascending to the region of the cells making up the taste disc of the fungiform papilla (Gaupp, 1904; Hammerman, 1969). Taste Disc The three morphological cell types composing the major- ity of the taste disc of the fungiform papillae in the frog have been described and differently labeled by various authors. Although the cells of the tOpmost layer appear homogeneous they have been variously called goblet (Ecker, 1889; Hammerman, 1969), cylinder (Gaupp, 1904; Robbins, 1967a), associate (DeHahn and Graziadei, 1971), supporting (Uga, 1966; Uga and Hama, 1967), and mucous cells (Rapuzzi and Casella, 1965; Stensaas, 1971). They apparently contain mucous granules and do not serve a direct sensory function (Hammerman, 1969). The layer of cells below these is composed of two general morphologic types. One has been labeled the forked cell (Ecker, 1889; Gaupp, 1904; Hammerman, 1969) because it di- vides into at least two peripheral processes which extend upward toward the surface of the taste disc. Rapuzzi and Casella (1965) refer to this type as supporting cell while Stensaas names it sustentacular (Stensaas, 1971). The other major cell type has its cell body located below the surface and on a level with that of the forked cell. It has a process extending up to the surface of the taste disc and due to its similarity to the visual receptor cell of the frog retina has been called the rod cell (Hammerman, 1969, Stensaas, 1971). It has been considered the taste receptor of the frog tongue (Beale, 1869; Ecker, 1889; Kolmer, 1910), a view supported by modern microanatomists (Hammerman, 1969; DeHahn and Graziadei, 1971; Stensaas, 1971) and physiologists (Kusano and Sato, 1957; Rapuzzi and Casella, 1965). It has variously been called rod (Hammerman, 1969), cylindrical (Ecker, 1889), goblet (Gaupp, 1904), sensory cell (Rapuzzi and Casella, 1965) and bipolar rod cell (Stensaas, 1971). Without discriminating between rod and forked cells the layer of cells containing these two types has been described as that of the gustatory cell (Kusano and Sato, 1957; Uga, 1966; Uga and Hama, 1967) or sensory cell (DeHahn and Graziadei, 1971). For the purposes of the present work the terminology of Hammerman (1969) seems most suitable, and will be used. The t0pmost layer of cells Will be referred to as goblet or mucous cells, while the lower layer will be con- sidered to contain rod and forked cells. Hammerman (1969) describes a total of seven cell types composing the taste disc, all but the previously mentioned three forming a ring around the periphery of the disc and not directly related to the functional aspects of the sense of taste in the frog. Because of their implication in the peripheral process of taste reception by different authors the anatomy of the three cell types most populous in the taste disc of the frog will be reviewed. Of the average 700 epithelial cells form- ing the taste disc on the surface of the fungiform papilla, 200 are estimated to be goblet cells (Robbins, 1967a). They are described as 10-20 micra in diameter and 20-24 micra long (Ecker, 1889), 19-22 micra long and 10-12 u in diameter (Hammerman, 1969) and Stensaas (1971), reports them to be 6-10 microns in diameter and extending from the surface of the taste disc where they compose most of the surface down to a depth of 25-30 microns, tapering as they descend. Between the basal processes of these goblet cells reside the bodies of the rod and forked cells which send thin processes upward between the goblet cells to the surface of the taste disc. The forked cells are somewhat smaller than most of the other cells of the taste disc. Ecker (1889) gives their dimensions as 6-8 p in diameter with 1-2 p apical processes. This is in agreement with the dimensions of Hammerman (1969) and Stensaas (1971) who report dimensions of 5-8 micra for the cell body and 1-2 micra for the diameter of the apical processes. Rapuzzi and Casella (1965) describe a taste disc as con- taining 40-50 rod cells. Ecker (1889) reports a taste disc as having several hundred of these cells. From the tip of their processes (that project to the surface) to their base these cells are 32-38 micra long. At their widest part they are reported to be 5-7 micra (Hammerman, 1969; Kusano and Sato, 1957) and 7-9 micra (Stensaas, 1971). The rod-like processes taper from four micra at the base to one micron at the apex (Hammerman, 1969; Stensaas, 1971). Nerve Branchings It is well-established that single nerve fibers of the glossopharyngeal nerve branch both within a single fungiform papilla and before entering the papillary stalk (Ecker, 1889; Gaupp, 1904; Herrick, 1925). Rapuzzi and Casella (1965), estimated that a single sensory fiber branched 4-6 times before entering the fungiform papillae and that each branch furthermore formed 5 terminals within a single papilla. If each branch innervated a different sensory cell, a single fiber was estimated to be in contact with about 30 gustatory receptors. Kusano and Sato (1957) add that each taste receptor is apparently innervated by more than one gustatory fiber. Rapuzzi and Casella (1965) tracing the neural inter- connections between papillae on the frog tongue by electrical- ly stimulating one papilla and recording the action potentials of collateral branches of nerve fibers innervating this and neighboring papillae found that from 6-29 papillae may be interconnected with any one papilla. They also state that most of these interconnected papillae have 2-4 nerve fibers in common with each other. Taglietti gt 31, (1969) reported that chemical stimulation of a single papilla with CaCl2 would lead to collateral nerve fiber discharge in 2-3 commonly innervated fungiform papillae nearby. Thus, it is evident that single nerve fibers not only innervate multiple papillae and receptor cells, but that their receptive fields overlap those of other gustatory fibers both within a single and among several fungiform papillae on the surface of the frog tongue. Such an anatomical arrangement is amenable to functional inter- action which may be occurring between gustatory units on the surface of the tongue. Receptor-Nerve Junctions, Hammerman (1969), Rapuzzi and Casella (1965) and Stensaas (1971) report patches of contact between sensory fibers and the rod cells of the taste disc which include loose intertwin- ings, apposition, and increased density of cell membranes in the areas of such contacts. DeHahn and Graziadei (1971) and Uga (1966), Uga and Hama (1967) describe such contacts as synaptic, having tight junctions, 200 angstrom synaptic c1efts,. and thickened membranes in the area of synapsis (Uga, 1966; Uga and Hama, 1967). Such junctions are polarized in their microanatomy. The sense cells of the taste disc contain dense granules aggregated near the synapse. These have been reported by Rapuzzi and Casella (1965), Stensaas (1971) and Uga (1966), Uga and Hama (1967). These vesicles are considered to contain a transmitter substance and have been reported to be 500-900 angstroms (Uga and Hama, 1967) or 800-1000 angstroms (DeHahn and Graziadei, 1971) in diameter. DeHahn and Graziadei (1971) further report that these vesicles contain norepinephrine as identified by microhistochemical techniques. This supports the suggestions of adrenergic functions proposed for these vesicles by Uga and Hama (1967). Another aggregation is that of clear vesicles which are found primarily in the nerve terminals and serve to distinguish sense cell from nerve. These vesicles are 200-600 angstroms in diameter (Uga and Hama, 1967), found only in the nerve terminals (Stensaas, 1971) and have been found by histochemical methods to be cholinergic (DeHahn and Graziadei, 1971). These structures suggest a bidirectional function of such synapses similar to those found in rat olfactory bulb (Andres, 1965). Receptor-Receptor Junctions DeHahn and Graziadei (1971) describe tight and gap junc- tions between peripheral processes of taste disc receptor cells. These have been described as appositional by Hammer- man (1969) but the former authors suggest that they may serve to provide a pathway for interaction between peripheral pro- cesses of receptor cells. Receptor-Goblet Cell Junctions Described as unspecialized by DeHahn and Graziadei (1971), the presence of tight junctions (zonulae occludentes) between the goblet cells and peripheral rod-like processes of the presumed receptor cells of the taste disc has been sug- gested to subserve a functional alteration of receptor func- tion by providing a low resistance pathway for ionic currents between receptor process and goblet cell (Stensaas, 1971). Goblet Cell-Goblet Cell Junctions The many tight junctions between adjacent mucous cell membranes of the taste disc are similar to those described above for receptor-goblet cell junctions. Though not morpho- logically similar to those believed to subserve chemical 10 .wsmsou moum 05p :w mousuosuum mummu Hmnmnmflnmm mo coaumuswmwumou oaumsmsom .H wusmfim Run; 1030... 3 ¢\uor_umc xH on» no access Hmsmcwa mo cowumasswun oesouoflpcm one omfio memes oaaflmmm snowflmcsm mo maaoo Scum mcflouoomu moouuomamouofifi now owns coaumummmum mo coausucwmmummu owusfifimnmcwo .v ousmwm usmazsz to uofeam .2529 / .. .........|ll..ll|.l|.\1 / ..n Isl eve». NH! R 1.6 mofmam .388 Mm... ..z immodest: .So Idl\\ ...........H...................... .H . r all“!- majafi 23.52:“. ................n...,....w....,u 5sz y 1% 3056335.: J . $35.6me 62:44:: ....m 232;; 53 with a metal electrode which serves to sample electric cur- rents produced in the chamber (Figure 5). Substitution of chlorided silver for the platinum used by Rapuzzi and Casella facilitated recording of both fast and slow potential changes. The glossopharyngeal nerve was stimulated electrically and action potentials were recorded from the nerve fibers supplying the papilla in the chamber electrode. This allowed determination of stimulus thresholds and most effective pulse durations and frequencies for adequate excitation of fibers innervating the taste papillae. Conduction velocities for these fibers were also calculated. In addition, under some stimulus parameters slow poten- tial changes could be measured from the whole papilla. These were recorded on the strip-chart oscillograph after the high frequency action potentials were electronically filtered out in the input stage of the amplifier. These slow waves were then related to the excitation of the various fiber groups recorded earlier. Other experiments substituted microelectrodes for the chamber recording electrode. Changes in the electrical cur- rents measured in single cells were likewise related to the stimulus parameters found adequate to elicit both the nerve fiber action potentials and the whole-papilla slow potentials. 54 SUCTION svamct Rte-cause ELEcfllooE \ #4 ca 4 _ l ”a (*3. CLASS Rseonumc CNMBE R 0| an I Figure 5. ' 1m Representation Of preparation used for recording action potentials invading fungiform papilla On electrical stimulation Of the IX nerve. With high-frequency components of the signal filtered out this preparation was also used to record slow- wave activity elicited in papilla by antidromic stimulation of the glossopharyngeal nerve. 55 Characterization Ofggpprces of Slow Potentialsgin Papillae Impaling cells on the surface Of the taste disc with microelectrodes was performed as described earlier. Once a cell had been penetrated, as evidenced by a sudden potential change and stable resting level, the stimulation of the ipsi- lateral glOSSOpharyngeal nerve was commenced and the current flow in the impaled cell measured and displayed on the strip- chart recorder. During the course Of the experiments the electrode was Often advanced through several membrane layers, some forming a seal around the electrode tip adequate to prevent great leakage Of intracellular fluid to the area out- side the cell. The depth Of the electrode as indicated on the micrometer advance mechanism would be noted for those cells which maintained their integrity sufficiently to respond to the stimulation of the IX nerve. Cells of the tongue sur- face outside the area Of the taste disc and fungiform papilla were also examined and their responses compared to membrane potential changes recorded within the papilla. Characterization Of Membrane Potential Responses Records were analyzed for resting membrane potentials (Em), amplitudes of membrane potential responses to nerve stimulation (Ep), polarity of these potentials (negative or positive), latency of onset of response (Lr) and in cases 56 where nerve stimulation was prolonged the potential at time of stimulus cessation was measured (Eacc). The exponential time constant (time for the response to reach 63% of its peak value) was measured for the initial change in potential and designated T p. The values for Ep were plotted against T p for negative (hyperpolarizing) and positive (depolarizing) potential changes. Stimulus-response functions were plotted for Ep vs. stimulus frequency, and Ep vs. number of stimulus pulses delivered at a frequency of one per second. EfferentElectrical Stimulatipn and Afferent Sensory Activity‘in IX Nerve The bipolar electrodes on which the peripheral trunk Of the severed glossopharyngeal nerve was laid could be con- nected in rapid succession to either the electronic stimulator or the input amplifier of the oscilloscope utilizing a rocker arm switch. For these experiments the nerve was first stimu- lated at frequencies and voltages known to initiate action potentials in fibers which innervated the taste papillae. Immediately after the period of stimulation (usually 15 seconds) recording of afferent action potential activity in this same nerve was begun. Concomitantly a chemical stimulus was applied to the surface Of the tongue. The afferent activity was electronically analyzed by an event/time histo- gram generator. The output Of this analyzer was recorded on 57 an oscillograph as number of action potentials for successive one second intervals. This enabled the comparison Of sensory nerve reSponse to chemical stimulation of the tongue with and without antecedent antidromic activation of these sensory fibers. Comparison was made of sensory response to chemical stimulation of the tongue and the effect of antidromic nerve stimulation on this response. In one experiment, nerve stimu- lation and chemical tongue stimulation were applied simultane- ously and the subsequent nerve activity was compared to activity from chemical stimulation alone. This was performed to test the effect of collision of afferent sensory and effer- ent antidromic action potentials on the response characteris- tics Of the tongue sensory units. Interaction Betgeen Taste Stimulation and Antidromic IX Nerve Stimulation On Taste Disc CeIl.Potentials Single cells of the taste disc were impaled with micro- electrodes. After a suitable baseline potential was estab- lished the papilla was stimulated by a single drOp of either Harris-Ringers saline, distilled water, 0.5-1.0 M NaCl, 0.03 M HCl, or 0.03 M quinine hydrocholride (NO. Q1125, Sigma Chemical Co.). The IX nerve was then electrically stimulated as previously described and changes in the cell potentials were recorded. Any effect of previous chemical stimulation of the papilla on the response of the impaled cell to nerve stimulation was noted. 58 Electrical Stimulation Of Taste Disc and IfReEOrding Of Effects on Response of Neighboring Papillary Cells to Antidromic IX Nerve StimulatiOn These experiments utilized a preparation similar to that described above. Instead Of chemical stimulation of the papilla under the recording microelectrode, however, electric- al stimulation of adjacent fungiform papillae was performed. Using a fine-tipped monopolar stainless steel electrode, bi- phasic electric currents were applied to the taste disc be- tween pairs of IX nerve antidromic stimulation periods. Monitoring of potential changes in cells Of the papilla next to the one under the monopolar stimulating electrode was accomplished with micropipette electrodes as described before. Ep was plotted against T p for pre- and post-stimulus periods. Displacement of Membrane Potentials During Antidromic Stimulation of the Glossopharyngeal Nerve Cells of the taste disc were impaled by the recording micropipette and currents passed through the electrode utiliz- ing the constant current source of the M-4A electrometer. The membrane potential could be thus displaced in either a hyperpolarizing (inside made more negative) or depolarizing (inside more positive) direction. The currents available were a function of the electrode-cell resistance. The magnitude of displacement was therefore variable from cell to cell. 59 The effect of membrane potential on Ep of the response to antidromic stimulation of IX was plotted in terms of magnitude and sign of Ep at positive and negative values of the membrane potential Em. VI. DEFINITIONS EO: Zero Potential. That baseline established when both the active recording electrode and indifferent electrode were electrically continuous in the bathing medium of the preparation. Em: Membrane Penetration Potential. The potential difference between the indifferent electrode (referred to E0) and the recording electrode after the recording electrode had penetrated a membrane (presumably a cell membrane) separating an area Of electric potential from the extra- cellular compartment. Ep: Peak Response Potential. The maximal change in Em pro- duced in response to electrical or chemical stimulation of the sensory structures of the tongue. Eacc: Accommodation Potential. The value of Em after a pro- longed period of electrical stimulation to the IX nerve. Lr: Response Latency. The time period from the onset of a stimulus to the time an incipient change in Em became evident. tp: Time to Peak. The time period from incipient post- stimulus change in Em to the time the response was maximal (Ep). 60 61 tacc: .Time of Accommodation. The time period from Ep to Eacc . Tp: Time Constant of Ep. The time for a change in Em to reach 63% Of Ep. Tacc: Time Constant of Eacc. The time for Em to fall to a value of 37% (Ep - Eacc.) Slope Eacc: Slope of Accommodation. The ratio of (Ep - Eacc) to the duration (Ds) Of a prolonged stimulus. The change in Em after the peak response Ep has been Obtained while stimulation is still being applied. ggEEEp: A ratio comparing the value of the response Of taste disc cells to antidromic IX nerve electrical stimulation before (pre) and after (post) subsequent electrical stimulation of a neighboring papilla. Post Pre Tp: A ratio of the value of the time constant of the rising phase Of the response of taste disc cells to anti- dromic IX nerve electrical stimulation before (pre) and after (post) subsequent electrical stimulation of a neighboring papilla. % Response: The value of afferent sensory nerve discharge elicited in response to a stimulus, expressed as a per- centage Of the frequency of action potentials recorded in a basal or non-stimulated condition. 62 Stimulus Pulse Train Duration. The time period over which continuous electrical pulses were delivered by stimulus electrodes. Stimulus Pulse Duration. The duration of the individual pulses composing a stimulus pulse train. Stimulus Pulse Frequency. The rate of delivery of single stimulus pulses within a train. Stimulus Voltage. The amplitude of the individual pulses delivered by the stimulating electrodes. 63 .mflmmamco How owns mouooou mmcommmu one msasaflpm mo mmwusomoum season ocm Hmuomfimu mcHNHuouomumao mumumamumm mo cowumucomonmou owsmmno .m ousmflm MZHB _ nA‘ _ _ nAI. _ . _ 1 Tu “A 3: com... -lIIIIIuLAII o3. (IL V . _ _ mm _ . _ u 4 _ _ .me _ allliil.oomo. IIIIlY. _ _ Em . _ - I, 1 . . u _ _ _ _ .IJ “ _ n 7.3:; . _ _ _ _ om _ . ozfianomm A: 0: mg .....III :01 |I MMMMflE msqnfifiem VI I . RESULTS Glossopharyngeal Nerve Stimulation and Slow Potentials in the Fungiform Papillae If antidromic activity in sensory fibers Of the IX nerve exerts an effect on the bioelectric properties of the taste disc cells innervated by these fibers, such an effect should be reflected in currents generated in the fungiform papilla. Currents of physiological significance in the function of the taste receptors would be expected to be localized in the sensory cells of the fungiform papilla and not likely found in cells of other tongue structures not innervated by taste nerve fibers. The properties of the fibers responsible for effects on the sensory cells should correspond to those of sensory afferent nerve fibers if the mechanism is to be considered an antidromic one. Two easily determined qualities differentiat- ing such fibers would be threshold voltage and conduction velocity. The experimental arrangement of Figure 5 yielded records similar to those shown in Figure 7. The polarities of such currents were variable in amplitude and sign. Different papil- lae would produce differing waveforms even though stimulus parameters effective in eliciting such currents were held 64 Figure 7. 65 I I t 1 vi ;1 J__l ‘9 i? (3L T'" INSTRUMENTS ... : J“ , . ‘f lNCORPORATEDtHO Three representative records of slow-wave bio- electric currents recorded from fungiform papillae using the gross chamber electrode illustrated in Figure 4. Parameters Of electrical stimulation of IX nerve: V = 1 volt; f = 2/second; Dp = l millisecond. 66 constant. The one factor which did relate to these currents was the stimulus parameters which most readily produced them. Application of relatively low-voltage (1-3 V.) square-wave (duration 0.1-1 msec.) pulses to the IX nerve at frequencies of 2-60 per second was usually adequate to elicit such slow wave responses. These parameters are similar to those re- ported by others (Esakov and Byzov, 1971; Rapuzzi and Ricagno, 1970). In experiments using electrolyte-filled micropipettes for intracellular recording, sharply defined currents of similar nature were seen on the surface of fungiform papillae and nearby. Large-tipped electrodes (as in some cases when the micropipette had been inadvertently broken Off at the tip) registered currents of more variability than those seen extra- cellularly with small-tipped (and presumably more selective) recording electrodes. Small-tipped electrodes were seen in all cases to display positive potentials (electrode tip more positive than indif- ferent) when placed on the tongue surface in the presence of stimulation of the glossopharyngeal nerve. Figure 8 shows the time and amplitude characteristics of such a current. The evidence of these slow potentials on the surface of a fungi- form papilla was used as an indicator of adequate nerve stimu- lation in all of the experiments utilizing microelectrode recording. Papillae damaged in preparation, poisoned by toxic substances (alcohol, tissue fixative, strong acids or salts) or with no blood flow rarely displayed these slow potentials 67 .ocooomfiaaflfi H n ma «ocoomm\om u m auHO> H n > "mumumfimumm mSHSEHum .wauofi asap oncommla m ma moms» Eouuon on» one .cofluoHSEflum OHEOHoHucm O» aneucouom succeeds. ecu ma wcflomnu wHOOflE 0:9 .m>umc xH was NO coaumasfiwum Hmowuuoon mo coaumuso may msonm msfiomnu mos one .ooouuooHOOHOHE an Omwo mummy mHmemm EHONwOGSM mo mommusm Eoum omouooou Hsflucmuom O>Huwmom .w shaman .Umm H _>E v .1 1 1 *1 11 t1 _1 .1 1‘ “1 _1 1 m1 ‘ s 1 .1 .1 w r _ L l 1 s _ _ 0021 ......qu_ _ _ _ _ — _ — _ .1 _‘ —‘ m‘ _1 .1 ~‘ ‘1 1 . _ _ _ I ~ 1 J _ . . I . . _. 1 n _ M . —l a! T El 74 i. ‘ A]: 1* (fl “ ~ . M L .1. m: ...L. 11.. (willful? . . . . .1 ,. I . . . . . l . _ ... T WI v.7..ll. ... y 1. .H . 1 1. ‘ s l 1.. 1 a“ r , t 2 --.. _ m l_ Ilml. t a l..¢((. . a _. _ I . A .1 s 1 . + L 1 . I .0U.1. w.ZI_LE:~x_n.Z.4,fl<HOG Heemcmueemommoam was no cOHueHSEHum Oflaonofluee an oouHOHHe on on ocsom memcommeu Haoo enmeu mo memmeao 03H we» omensmhu noncommeu emeea .OmHo same» we» cs mHHeo seemeMMflo 03» mo maewuceuom eceunfiefi ca meoseso Amy msHNflHeHomHemmn oce Adv mafiuflueaomeo i -J.-- ._ “It- ..-L...‘ y . : i - 1 ' _. ...‘. ___ . .' 1 . 1 ‘ - ‘— _ . ' .g . .. '_- ._ .' ' n i _ _. _ J—Agib _L_.4_.._ _ . i . 1 I r ‘ 2 e I m l 1 | 1 :JEJ i __i ! ‘1 4.1—4.? .' . . +—+— . I ‘T‘ A-v- I i T cc .n.eoa ousmflm “WZW!3I>IZ. 34 “NF . . m 73 .mcouoflfi oH ”mamom .Amv Hmmma Hamo @mamaumuumocswmm Hmmmmw msu on maflwaommmuuoo msummw mwouu nomam um cwmm waco mum3 coaumasEHum m>umc oaaouvflucm on mmmcommmn Hamo mcfluflumaomummwm .A¢V Hmmma Hafioflmummsm #mOE may ca UGSOM MHCOEEOO umoE mumB coaumasafium m>ch OHEOHUHucm on mmmcommmn Hamo mcwuflumaommo .cmmm maflmmm ma >68 coaumoHMHumuum Haasaamo may mafluocmw Hmaosc mo mummma 03¢ Mae .mou map so ma omHU may no mommHSm mna .mcflwuoomu mUOHuUmeOHOHE CH wmmn maaflmmm EHOMflmcsm m we omflo mummy m nmsonxu cowuomm Hmuflmmmm m mo ammumouoflfiouonm .HH musmflm 74 75 of each cell was recorded following electrical stimulation of the Ix nerve. There were 98 cells which gave a depolarizing response to nerve stimulation. The resting potential (Em) of these cells had a mean value of -10.46 i 8.4 millivolts. Hyperpolarizing cells had a mean Em of -8.86‘:_5.5 millivolts (79 cells). Depolarizing cells were always found in the surface layer of a papilla showing a response to gloSSOpharyngeal nerve stimulation. Occasionally depolarization was noticed in cells at depths greater than that believed to be occupied by goblet cells. Hyperpolarizing cells were never found at the surface of a responding papilla. The mean amplitude of the depolarizing potential (Ep) in twelve cells of the tOp layer of the taste disc was +16 mV i 2.7 and that of twelve hyperpolarizing cells penetrated at depths below 15 micra from the surface was -12.6 mv :_3.5. These values (disregarding sign) were not significantly dif- ferent (p = .05). The relative magnitude of the re3ponse potentials varied among preparations. In any one papilla, however, using the same electrode for all measurements, relative comparisons could be made with some assurance. In the course of this investigation it became evident that the hyperpolarizing cells were found only within the taste disc, and at some depth beneath the surface. Depolariz- ing cells were seen on the surface of the disc, at depths 76 within the disc, and in areas not considered to contain taste receptors.n These extrapapillary depolarizations were seen in cells located near and around the fungiform papillae. Figure 12 illustrates the data obtained from an experi- ment planned to make comparative measurements of these re- sponses in and around a fungiform papilla. The response amplitudes (Ep) for twenty-one cells exhibiting depolarization on Ix nerve stimulation are plotted with respect to their distance from the center of one fungiform papilla. The two vertical bars delineate the area of the taste disc. As can be seen, the depolarization exists in cells within and to either side of the fungiform papilla but the relative ampli- tude of this depolarization falls off rapidly as the electrode is moved from the border of the taste disc. From this data one would expect that either these cells are also innervated by fibers of the IX nerve, or that they serve as passive sources for a current sink located within the papilla. Characteristics of Slow Potential Responses 1n Taste Disc Both hyperpolarizing and depolarizing slow potentials were recorded in taste disc cells of the fungiform papillae. If these two waveforms are different reflections of a common mechanism they would be expected to display common character- istics in terms of latencies, stimulus-response functions, adaptation or fatigue, thresholds and risetimes. 77 DISTRIBUTION OF POTENTIALS EVOKED BY ANTIDROMIC STIMULATION OF IX PLOTTED WITH RESPECT TO SURFACE OF PAPILLA AND ADJACENT TONGUE AMPLITUDE OF Ep (millivolts) Figure 12. area of : taste disc 20» 9 19- 18F O 17L 0 14 - 13 F 12 ‘ ll 10 t 1- 0 Q 1-0 4 l I L . L L I 1 1 4 l... 300 250 200 150 100 50 0 50 .100 150 200 250 300 DISTANCE (micra) Response potentials (Ep) of cells penetrated on the surface of the taste disc and adjacent non- gustatory epithelium surrounding a fungiform papilla. The two vertical lines delineate the area of the taste disc. 78 Latencies of initial response as measured in 75 depolar- izing cells and 67 hyperpolarizing cells had identical means, 1.3 seconds. This is well above the conduction time measured for nerve impulses to travel from the stimulating electrodes to the taste disc (3—4 msec.). These latencies compare with similar values published by Esakov and Byzov (1971). Thresholds to nerve stimulation were also identical. At a frequency of one stimulus per second the peak ampli- tude of both depolarizing and hyperpolarizing responses in- creased with the increasing number of pulses delivered in a stimulus train up to about ten pulses. However, the response increment per stimulus pulse (alEpl/pulse) decreased as the number of pulses delivered to the glossopharyngeal nerve in- creased. In a train of pulses, therefore, each successive pulse was less effective than the preceding in producing a voltage change in the cells of the taste disc. Figure 13 illustrates this relationship. The summation effectiveness of repetitive pulses in nerve fibers innervating the papilla appear to be optimal at low rates. Frequencies of this order are similar to those elicited in gustatory sensory nerve fibers by weak, near threshold taste stimuli. Figure 14, a,b,c shows responses of a depolarizing cell to 1,3,4 stimulus pulses delivered to the IX nerve respectively. Figure 15 illustrates the latency, peak response time, adaptation and habituation characteristics of hyperpolariz— ing cells. In most cases these were similar to those qualities 79 .cmwm msfloummmumflo .Hmnummou omuuon mum mmmcommmn mCHNflHmHomummmn m can momcommmu msfluwumaommo Ha mo msHm> one .mmmasm ow ou a mo wmcmu m Hm>o m>ums xH was 0» owum>flamo mmmasm msaofiflum mo Hogan: may mo coeuosdm n no oopuoam wmasm msasafium UHEOHUHMCM mom ucmeuosfl oncommmn Hamo omwo mpmma .ma musmflm Aonaun woav o>uoz xH cu vouo>uaon condom uaaaaaum no nopaaz on oe on o~ ‘ OH H ‘ JI ‘ I ‘ d 4 I i H .i- O 1 1 0 o o o a one. :::9; C IoIOuOIlOOIOI I. '03.... 0000...! I mug—NZ NH NIH. OH. EEK—awn mxuomm UHMHUNAN ho mama‘s»; 55:55 UZHSOAAOK mmqu mbghm mum 350m mag: 2H NUSU mg no MDA<> meaqoma 80 Figure l4a,b,c. Changes in the depolarizing response potential of a taste disc cell when the IX nerve is antidromically stimulated with one (a), three (b) and four (c) pulses. Time to peak (tp) for each response as measured from onset of first stimulus pulse is 4.5 seconds. 81 .m>uos xH mnu mo sesamasfiwum woumumwu ou noncommwu mswuwumaomnmmhn HMOfimme .mH ousmfim 82 observed in depolarizing cells except for polarity. Regardless of stimulus strength the latencies of the initial response in both cell types were similar. Latencies of 75 depolarizing cells in 9 experiments gave an average of 1.3 i 0.32 seconds, including the conduction time for Ix nerve impulses of 3-4 msec. Latency of response for hyperpolarizing cell reSponses as measured in 67 cells averaged 1.3 i 0.35 seconds including nerve conduction time. Because time intervals could only be measured in most cases with an accuracy of 0.3 seconds, the latency values can be considered identical, the differences falling well within the experimental error. Stimulus train durations longer than 4—5 seconds did not enhance response potential amplitudes over the value reached in the first 1.5-4 seconds. Thus, response Ep reached a peak in 4-5 seconds, and despite continuing stimulation of the nerve fell off to some relatively constant level in the face of nerve stimulation. The Ep continued to rise in cases where stimulus train durations were less than the character- istic response times of 4-5 seconds. In these cases, Ep values were reached with the same interval required as when stimuli were of longer duration for any given cell. A more accurate estimate of rise times is the time constant, Tp. Values of Tp were calculated for 43 antidromically evoked slow wave responses arbitrarily selected from five different experiments. The mean rise time constant for hyperpolarizing reSponses (25 observations) was 1.7 i 0.6 while that for 19 83 depolarizing responses was 1.6 i 0.4 seconds. The value of the Ep in these cells was not a determinant of the rise time of the response. Within any one cell the antidromically elicited hyperpolarization or depolarization reached its peak in approximately the same time from its initiation regardless of the amplitude of the response. The lepe of the rising phase of the Ep was, therefore, a function of the magnitude of the Ep. This may be appreciated by referring to the responses displayed in Figure 14. Figure 15 illustrates another characteristic of the taste cell response to IX nerve stimulation. Repetitive trains of high frequency stimulation delivered in short intervals led to a progressive decrease in response magnitude Ep. Typical of both hyperpolarizing as well as depolarizing responses the effect of such stimulation was to drive the resting or post stimulus membrane potential in the direction of the Ep. Given sufficient recovery time, the Em of these cells would often return to values near those recorded on penetration of the cell, but due to the technical difficulty of remaining in the cell with the recording electrode for such extended periods (5-10 minutes) few measurements of such long term processes were successful. A consistent characteristic of both hyperpolarizing and depolarizing taste disc cells is the decrease in response potential level shortly after attainment of peak amplitude in the presence of continuing stimulation of the glossopharyngeal 84 nerve. This may be consonant with adaptation. With very long stimulus train durations such adaptation often was seen to return the intracellular potential not only to the baseline level but beyond. The rate of adaptation in hyperpolarizing cells was somewhat more rapid than in depolarizing cells but could not be shown to be statistically different. The slope BE/Bt for 20 measurements of hyperpolarizing responses in two experiments had a mean of 0.908 millivolts/second and 16 measurements of depolarizing responses in three frogs had a mean of 0.214 millivolts/second. The rate of adaptation varied among cells sampled. Rebound of the membrane potential occurred in both cell types at the end of a period of stimula- tion during which it had reached a steady state level. The latency of this rebound was approximately the same order of magnitude as the Ep but its polarity was opposite. That is, if the adapted cell was a depolarizing cell, the rebound phenomenon changed the membrane potential in a hyperpolariz- ing direction and in hyperpolarizing adapted cells, the rev bound tended to produce depolarization of the cell beyond the adapted level of the Em. Figure 16 illustrates both the adaptation and rebound overshoot phenomena as recorded from a hyperpolarizing cell. Figure 16. 85 Overshoot during adaptation to a prolonged stimulation of the IX nerve and rebound after stimulus cessation frequently seen in responses of taste disc cells. The hyperpolarization produced in this cell by antidromic nerve stimulation results in an adaptation overshoot and post-stimulation rebound in a depolarizing direction. 86 Antidromic Stimulation and Responses;of‘Afferent Nerve Fibers to Taste Stimuli . Antidromic stimulation of the frog glossopharyngeal nerve to the tongue produces hyperpolarization and depolarization in different cells of the taste discs. If these antidromic effects have significance in modification of receptor cell sensitivities to taste stimuli, such an effect should be re- flected in the afferent discharge rates of sensory fibers responding to chemical stimulation of the tongue. The values for two preparations are shown in Figure 17 a,b. They show that antidromic nerve activity can depress sensory fiber responses to taste stimuli. Similar results for single papillae have been reported by Filin and Esakov (1968), Macdonald (1971), and Taglietti gt El° (1969). None, however, have correlated this inhibition of afferent response with the effects of antidromic impulses on cells of the receptor sur- face of the taste disc. Figure 17 c shows the results of an experiment performed in one frog. In this series, hydrochloric acid was applied to the tongue as a chemical stimulus concurrently with elec- trical antidromic stimulation of the IX nerve. Because stimu- lation and recording could not be carried out simultaneously in this preparation, activity values in the first response interval are not shown. Chemical stimulation alone produced (an.increase in discharge rate of 191%‘(referred to prestimulus 87 .mua>auom Hams“ so mnmnflm w>nms mmmnu mo QOHMMH :sEHum Havauuomam oweouofipsw mo muomwmm on» can msmcou on» no coaumHSEaum HMUHEQno on uncommon m>umg >H0pmumsm mo zoomswmum .ha wusmflm «Haauum Huuwaono mo acuunoaammd nouw< avowuum mafia can u u u A as nNV k!«, A on nu so as... \\ I»... w .6oauum «use. cos magnum mausooswuasaam voguaa< Huaaaum oo~ ”sowuuooam vow Hogaaoso . con #02 ea ~ 0 cod Amuw>uuuw usasaaumnuun 0 com m 5 upsuwswo: oasomuom com m 00H aowumassfium Hauwuuoofim cow opuoz xH uuumw nsdzawum Hmousoco can .( manaanum aaoaaosu r ZOHH¢ADZHHm NHmmmz NH ho NmZGmmmm zo rHH>HHU< UHZONQHHZ< mo maomhhm 88 baseline rate). Chemical stimulation applied in the presence of antidromic IX nerve stimulation resulted in a post-stimula- tion discharge of 227%. Macdonald (1971) reported a period of enhanced excitability following antidromic stimulation of sensory fibers in the bullfrog tongue but this enhancement was only seen for a period of 5-20 milliseconds after stimula— tion. The enhancement seen in Figure 17c occurs at a time post-stimulation that compares with the rebound depolarization seen in the membrane potential of taste cells following pro- longed antidromic stimulation (Figure 16). Intracellular Responge of Taste-Disc.Cells.to Direct Chemical Stimuli.and.Its Effect On Antidromically Elicited Slow PotentiaIs., If there is a mechanism for peripheral interaction among sensory units of the taste disc, and if such a mechanism in- volves changes in the bioelectric characteristics of the receptor cells, chemically elicited activity in the taste disc might be expected to modify the intracellular responses of receptor cells to antidromic stimulation of the glOSSOpharyn- geal nerve. Cells of the taste disc of the fungiform papilla were impaled by microelectrode and records were obtained of their electrical response to chemical stimuli applied by micro- syringe. The results of 72 observations in four frogs are shown in Figure 18. Records obtained at various depths within 89 .sEsHoo some mo Hmouon puma mcon omxumfi mum uncommon some manasoamo on own: musaom mama .Hmaflfiflm mam>flumuflamsw mums omflo mummy was CH mzummo Eoocmu um moouuomeOHOAE an omammEH maamo mo mmwsommmm .mommnsm on» on uncannaom HMOHEmno mo cowuwoflammm an omosooum omflo mummy may cw maamo mo Asmv mamwugmuom msmnafimfi CH wocmcu HADZHHm GHA36- . 8 55 .4338 $362 .ma musmflm 90 the papilla were not qualitatively different with respect to depth. Not all cells penetrated at depths below 15 micra responded to chemical stimuli, but no effort was made to identify these cells. The IX nerve was antidromically stimulated in these preparations in the presence and absence of chemical stimula— tion of the papilla containing the cell impaled by the micro- electrode. The amplitude of the slow wave response (Ep) of the cell was compared in conditions of taste stimulation and no taste stimulation. Figure 19 illustrates the average responses from 51 observations in 4 frogs. Salt solutions and quinine usually produced a depression of the antidromic response, while double distilled water was seen often to lead to an augmented antidromic response. If the antidromic effect was dependent on the cell membrane potential, one would not expect to see the opposite effects of quinine and water on the Ep since both these substances cause hyperpolarization of the receptor cells. The hyperpolarizing effect of water has been considered a result of leaching out intracellular cations from the receptors (Eyzaguirre 32 31., 1972). Effect of Electrical Stimulation of Neighboring Papillae on the Response of Taste Disc Cells to Antidromic IX Nerve Activity In the frOg action potentials elicited in a sensory fiber :innervating one papilla can travel antidromically up a col- ILateral branch into a neighboring papilla (Rapuzzi and Casella, D E P o L M A E R M I B Z R A A T N I E O N P 0 u T Y E P N E T R I P A L 0 L A R I Z A T I o N Figure 12 . -7-1? -8.L -9 JL 91 u.“ (Values greater than 12) U) a: 3 2 H ad QUININE DISTILLm 1120 (Values greater than -12) - RESPONSE OF TASTE DISC CELLS TO CHEMICAL STIMULATION OP SINGLE PAPILLAE Effect of prior application of sapid solutions on the amplitude of antidromically elicited hyperpolarizations of taste disc cells. Individual data points for each stimulus class are are marked along left border of column depict- ing mean effect. 92 1965) and are capable of depressing afferent activity of other sensory fibers innervating the antidromically invaded papilla (Filin and Esakov, 1968). If such depression is mediated via an effect on the bioelectric properties of taste disc cells, it could be reflected in changes of the transmembrane poten- tials recorded from these cells. The tracings shown in Figure 20 illustrate small poten- tials which could be recorded from depolarizing (Figure 20a) and hyperpolarizing (Figure 20b) cells of a fungiform papilla when an appropriate neighboring taste disc was electrically stimulated via a fine-tipped stainless steel monopolar elec- trode. The direction of potential change in each of these two classes of cells is the same as that produced by anti- dromic electrical stimulation of the IX nerve. The large initial potential deflections of each tracing identify the cell response to antidromic IX nerve stimulation. Neighboring papilla stimulation can increase or decrease the response of taste disc cells to antidromic IX nerve stimu- lation (Figure 21). The data from two of the preparations which were found to reSpond in this way is plotted in Figures 22 and 23. In these figures the Ep and Tp are compared for the period immediately before and at varying times after stimu- lation of the neighboring papilla, and plotted as the ratio of the pre-stimulus:post—stimulus value at various tbmes followu ing stimulation. The ratios of the post to pre-stimulus Ep Figure 20a. 20b. 93 Small depolarizations produced in a cell of the taste disc by electrical stimulation of a taste disc on a neighboring fungiform papilla. The initial depolarization under the solid arrow indicates the response of this cell to electrical antidromic stimulation of the IX nerve. The two smaller depolarizations under the white arrows are the result of direct electrical stimulation of the surface of a neighboring papilla. The hyperpolarization indicated by the solid arrow was produced by antidromic IX nerve stimulation and characterizes this cell as a hyperpolarizing cell, as referred to in the text. Direct stimulation of a neighboring papilla could produce the small hyperpolariza- tion indicated under the white arrow. 94 Y . . I-.. i ”+M. 2 i 1 _. .11'_J__....' . '11:. I -11 1.1--1 .11... ,__v 1.. I ... 9 % _.,..+_.4 A- .. 1 ‘4 1 —. 1 . 1 -..1 . , . --.“ ......— . -_.1.._-_. _~. ‘ . Figure 21. 95 Effect of electrical stimulation of neighboring papilla on the antidromically elicited response of taste disc cells. Black arrows indicate the response of taste cell to antidromic stimula- tion of the IX nerve. White arrows indicate periods of adjacent papilla electrical stimula- tion. 96 .mwxwhHmeH VAHH may mo coaumaseflum oeeouoflusm ou mmcommmn msfluflumaomoo map mo me can mm map so soaumHsfifium maaflmwm mnemonnmflms mo muommmm mo wmuooo mafia .NN musmwm Auoov a-«nqe mo ceauqaaawum u boom gawk 0n o~ DH N l/ A n . . . . . . . . . v I M nu ouuuu mm x canon m6 (AAHA onHfiummos w 3onm mw>uso 03¢ one .omupon mammaflfiflm mum mwmsoammu mmwnu mo mosam> my one .omummp mmB coaumHDEHumumom mm Donn soHumHDEHpm waaflmmm mcfluonnmflms “opus ommmmam eBay or» msmum> omuuon mum waaflmmm mnemonnmwmc 6 mo cofluoasfiflummumuumom mm mo mmowumu was .sowumHSEHum Hmoaupomam w>umc xH UHEOHUwucm OD mmsommmu mcfluflumaomnmmms map mo me one mm wnu so COHumasEHum waaflmmm mnemonnmflms mo muomwmm mo mmusoo mEHB nownv coaucaaawum I umom «Bah om ov on o~ OH H “I!!! o 3:: an / x x aqua» err )II C) L_l1 (AAHm onHuos xH onu mo coeuoasfiflum erOMoHDco onu 0p mHHoo mafiufluoaomoo maaofiuos oounu mo omsommou onu so Afimv Hoflusouom osoHQEoE mo uoommm .vm ousmfim O m a :oo o N e Haoo >3 0' 1 l. O H a HHoU mAAmU UZHNHgome 0.1. On... ON... OH... CHI ON! Om... Gel on! 00.. >805! I m H n ”I L. P am 0+ 1 I 0 o .\ O ‘ \\ 0 0 but 0‘ 000000 . \\ n‘ .......................... O o. \ O O ..................... \q\ 0 O O mH+ 1 T O O 0 >5 «N... 0 am 101 .o>uon xH mo noHuoHsEHum oeaonofluno en ooosooum oHoB Ammv maoeunouom ooxo>o on» ono ooouuooaoouoefi mnHoHooon onu nmsounu unonnso mnemmom an ooeuo> mo3 Hoeunouom onoHnEoE one .meaoo OBu mo name omnommon mnemeuoeomuommn onu no Afimv Hoeunonom onounfiofi mo uoommm .mm ousmem S >Evl 11 o ‘I \I‘ \?\ ‘O‘\“ ‘\ I O]... O \\\du\ m \‘\ I I I luflillulilnlnluululllull. «.11 . 71: n... .1 m+ ~+ T u u u u .. r . . . r u w m N ..m . . . . (low H+xr O n s :3 . s . Hnwu ~+.11 mnnmo quNHmjomdmme: >a~+ 11 VIII . DISCUSSION Peripherally originating interactions among gustatory units have been reported (Filin and Esakov, 1968; Macdonald, 1971; Miller, 1971; Taglietti et al., 1969; Wang and Bernard, 1969). The functional element subserving communication be- tween interacting sensory structures has in all cases been indicated to be the afferent sensory nerve fiber. Because one fiber innervates several sensory papillae via collateral branches (Ecker, 1889; Gaupp, 1904; Herrick, 1925: Rapuzzi and Casella, 1965) it was proposed that such peripheral inter- action required the antidromic conduction of action potentials along these collaterals to other terminals of the sensory unit (Bernard, 1971a,b; Filin and Esakov, 1968; Macdonald, 1971; Rapuzzi and Casella, 1965; Taglietti 33 al., 1969) or local, non-propagated currents in the collaterals (Miller, 1971). Figure 26 shows how such antidromic action potential activity may invade neighboring papillae in a sensory unit (Bernard, 1971 a,b). Antidromic Activity in Ix The lingual branch of the IX nerve in the frog has been found to contain sensory (afferent) and autonomic (efferent) 102 103 TWO-WAY CONDUCTION IN TASTE FIBER COLLATERALS Stimulation Taste Papillae ORTHODROMIC ED Figure 26. Representation of the pathways for orthodromic and antidromic conduction of action potentials in a single taste fiber innervating three papillae on the surface of the tongue. (Bernard, 1971). 104 nerve fibers (Ariens Kappers e£_§l., 1936; Herrick, 1925; Strong, 1895). Sympathetic efferent activity has been found to enhance the afferent sensory activity of fibers from the tongue (Chernetski, 1964; Halpern, 1967; Kimura, 1961) and to moderate blood flow in tongue structures (Erici, Folkow and Uvnas, 1951; Erici and Uvnas, 1951). Other autonomic centrifugal fibers were found to enhance or inhibit afferent taste fiber discharges from both the ipsi and contralateral tongue receptors (Brush and Halpern, 1970; Esakov, 1970; Halpern, 1967). The conduction velocities in these efferent fibers is found to be in the range from 0.2-0.8 meters per second (Chernetski, 1965). Taste afferents in the frog have conduction velocities of l-lS meters per second (Rapuzzi and Casella, 1965), and do not cross the midline to the contra- lateral side of the tongue to any great extent (Strong, 1895). The slow potentials recorded from taste papillae pre- sented in the results of the present work are most likely, therefore, to be the result of primarily antidromic activity in sensory fibers of IX, because the stimulus parameters used to stimulate the cut peripheral stump of IX initiated action potentials in fibers with conduction velocities of 6-14 meters per second, and did not affect blood flow in the tongue. In addition, the effects produced were ipsilateral and no papil- lary bioelectric activity could be elicited on the contra- lateral side of the tongue on stimulation of the IX nerve with parameters effective in producing responses in ipsilateral 105 papillae. The nerve activity was not able to produce changes in the sensory discharge of the contralateral IX nerve as described by Esakov (1961) when he stimulated efferents in the glossopharyngeal of the frog. The relatively long latencies of the papillary cell responses to this antidromic stimulation of the Ix nerve are too great to be accounted for even by the slow conduction velocities in the autonomic fibers. Assuming even the slowest of these fibers were responsible for the effect, with a con- duction velocity of 0.2 meters per second and a maximal nerve length of 5 centimeters the latency due to conduction in the slowest fibers would only be 25 milliseconds. The argument, therefore, that the long latencies indicate activation of autonomic fibers is not a likely one. It is known that touch fibers may innervate more than one papilla per fiber. Rapuzzi and Casella (1965) found that a tactile sensory fiber innervated an average of 2.7 papillae. Each papilla, however, had only one tactile fiber in it, and the fiber did not branch within the papilla. The diameter of these fibers is about the same as the gustatory fibers and their conduction velocities fall within the range observed for those fibers activated in the present experiments. However, the gustatory fibers each innervate from 5-6 papillae, and are estimated to branch again within each papilla about five times (Rapuzzi and Casella, 1965). Therefore, one gustatory fiber can be expected to have about 25-30 peripheral terminals as 106 compared to only 2-4 for a tactile fiber. It is reasonable to assume that the relatively large currents associated with the slow potentials in the whole papilla (Figure 7) could not be produced by the single tactile fiber innervating it, and that the responses observed in the papilla are more likely a result of activity in the chemosensory fibers, and that these fibers were conducting action potentials centri- fugally, a direction antidromic to their afferent sensory conduction. Inhibitory_Effects of Antidromic ' Activity in Ix Antidromic activity in the glossoPharyngeal nerve was found to decrease the afferent activity of chemosensory fibers from the frog tongue (Figure 17). Such inhibitory effects of activity in peripheral taste fibers have been described for the frog tongue by Filin and Esakov (1968), Macdonald (1971), Miller (1971), and Taglietti et 31. (1969). There appears to be a time-dependency of such inhibition. Initial effects (2-20 milliseconds) were facilitatory and were followed by a period of depression (Macdonald, 1971). The depression had a prolonged time course of up to 5 minutes (Filin and Esakov, 1968). Two sites for such depression effects are evident. First, nerve terminals of the sensory fiber collaterals may exhibit a refractoriness due to depolarization produced by antidromically 107 traveling action potentials. These effects would be expected to have their major influence over a period of not more than 400-500 milliseconds at the most since the recovery of these nerve fibers following activity does not exceed this time (Gasser e£_gl., 1938). Such depression times are in agreement with the findings of Macdonald (1971). Similar effects of antidromic depression of nerve terminals for cold receptors in the tongue of the dog are reported by Dodt and Walther (1957). The inhibition of peripheral tactile units by anti- dromic invasion (Lindblom, 1958) follows a similar time course. The initial facilitation reported by Macdonald (1971) is in agreement with the supranormal periods of nerve fibers of this size (Gasser et.gl., 1938). The long time course of depression found by Filin and Esakov (1968), and this study, cannot be explained by the refractory periods of the nerve terminals and must therefore be dependent upon another func— tional mechanism. The second site where depression may be manifest is at the peripheral receptor itself. Such a mechanism would require the postulation of a trans-synaptic effect which would outlast the relatively short depression period seen in the nerve terminals. In addition, since antidromic action potentials in afferent fibers are responsible for the depression, either an inactivation of the neuronal post-synaptic membrane or an inhibition (synaptic) of the receptor must be assumed. 108 No evidence for the former hypothesis is available and con- sideration of the latter would require either the presence of an inhibitory interneuron which would exert a depression on the receptor or a population of both afferent and efferent synaptic structures on the same sensory neuron. For the frog taste papilla both afferent and efferent types of synapses between nerve fibers and taste receptor cells have been anatomically described (DeHan and Graziadei, 1971; Uga, 1967; Uga and Hama, 1967). These have been con- sidered junctions between receptor and sensory neuron without mention of the known efferent fibers to the papilla. However, since it has been shown that antidromic activity in sensory fibers can produce a change in the bioelectric properties of the receptors (Figures 10,13, 14, 15, 19, 20) and that ortho- dromic activity in chemosensory fibers is dependent on chemical transmission between receptor cells and taste fiber endings (Landgren et_§1,, 1954; Rapuzzi and Ricagno, 1970), the func- tional possibility of reciprocal transmission at the receptor- neuron junction is an acceptable hypothesis. Anatomical and physiological evidence has been presented for the existence of such a dual system in the rat olfactory bulb (Andres, 1965; Rall and Shepherd, 1968; Reese, 1965; Reese and Brightman, 1954). It is also known that the same transmitter may be respon- sible for excitatory as well as inhibitory effects on the same postsynaptic cell, the predominant effect being a function of 109 the relative activity of inhibitory and excitatory postsynap— tic sites (Wachtel and Kandel, 1967). Such a system does not violate the principle of Dale (Eccles, 1957) which states that a neuron may liberate only one species of transmitter at all its synapses. Postsynaptic inhibition by the same cell which also elicited excitation of the postsynaptic neuron in the trigeminal nucleus of the cat was shown to be a function of the frequency of the presynaptic discharge (Dubner, 1967). Wachtel and Kandel (1967) showed that a single neuron could be excitatory at some of its synaptic terminals and inhibitory at others. They also demonstrated that a single presynaptic element can exert both excitatory and inhibitory postsynaptic effects at the same site on the postsynaptic membrane. Such mechanisms may be utilized in the peripheral interactions among sensory units of the tongue. Wang and Bernard (1969) found that afferent nerve discharge of chemosensory units in the chorda tympani nerve from the tongue of the cat showed decreasing activity with increasing stimulus concentration. These results can be explained by invoking mechanisms of peripheral modulation of activity among receptor-nerve elements of the tongue. Antidromic IX Nerve Activity and Taste Receptor Cell Bioelectric Changes Hyperpolarization of nerve or receptors has generally been shown to result in inhibition of activity in these 110 elements while depolarization is associated with a state of enhanced excitation (Boeckh et_gl., 1965; Jenerick and Gerard, 1953; Kuffler and Eyzaguirre, 1955; Lorente de No, 1947). Nomura and Sakada (1969) found that afferent activity and slow depolarization of tactile nerve fibers in the papillae of the frog were directly related. However, taste fibers did not produce such a reliable correlation between afferent im- pulse activity and terminal depolarization on stimulation of the papilla with water, sometimes showing action potential generation without a concomitant "generator potential." The slow potentials recorded by Nomura and Sakada were considered to be changes in sensory nerve fiber currents and not currents produced in the receptor cells of the taste disc. Figuresllun 16, 16, 20b, 21 illustrate that hyperpolari- zation was produced in cells of the taste disc by antidromic stimulation of sensory fibers of the glOSSOpharyngeal nerve. Figure l7a,b shows the inhibition of sensory activity in nerve fibers produced by such antidromic stimulation. Salt solu- tions were found to produce depolarization of cells of the taste disc, as did HCl (Figure 18). This finding supports those reported by Eyzaguirre 25 El. (1972) and Sato (1969) for the toad and frog, respectively. Similar effects were reported in intracellular studies of the rat taste bud (Kimura and Beidler, 1961; Ozeki, 1970, 1971; Tateda and Beidler, 1964). Figure 19 shows that depolarizing stimuli such as these were able to lessen the hyperpolarizing effects of antidromic nerve 111 stimulation. Eyzaguirre et 31. (1972) showed that pure water exerted a hyperpolarizing effect on cells of the toad taste papillae. Figure 18 shows the similar effects obtained for water in the frogs used for the present experiments. In Figure 19, furthermore, it is seen that previous application of water to the taste papilla of the frog potentiated the hyperpolarizing effects of antidromic nerve stimulation. On the other hand, Figure 170 shows that the simultaneous appli- cation of a depolarizing chemical stimulus (HCl) to the papilla with antidromic stimulation of the IX nerve led to a small potentiation of the chemosensory discharge. It would seem that the potentiating effects of antidromic activity on the response to HCl is not reconcilable with the concept that afferent activity in chemosensory fibers is a simple function of receptor cell membrane potential. By studying the adaptation of a taste papilla cell to prolonged antidromic stimulation (Figure 16) it can be seen that the hyperpolarization decreased in spite of the continu— ing antidromic activity. The membrane potential of this cell is seen to overshoot the initial resting potential in a more depolarizing direction. If the membrane potential is an accurate reflection of the‘bioelectric state of the cell, it could be considered to be completely adapted in about 90 seconds where the slope of the change in potential is seen to be practically flat. If such adaptation was a function merely of changes in ionic conductances, cessation of the antidromic 112 activity would be expected to allow either a return to the original resting membrane potential, or the cell would remain at nearly the same potential as found when completely adapted. Figure 16 shows, however, that at the termination of anti- dromic activity the cell membrane potential bounces upward, further depolarizing it. This "rebound" phenomenon indi- cates that even after the membrane potential adapted to a steady state, the antidromic nerve activity was producing an effect on the cell, recovery from which involves changes in electrogenic components of the cell that produce a depolariz- ing change in the membrane potential. Figurelln>shows another example of this rebound phenomenon where the adaptation is sufficient to unmask this interesting phenomenon. Apparently when antidromic rates are high, as in Figure 15, the ability of the cell to adapt is unequal to the task and the rebound appears to be an approach to prerstimulation potential. The observation that the waveform of the Ep varies with stimulus conditions may indicate at least a two-component mechanism of antidromic effect, one a fairly rapid change (5-20 seconds) which is reflected in the change in membrane potential, and a more prolonged effect (5-10 minutes) that is less labile. This is supported by the fact that the cell which has exhibited a depolarizing rebound will begin to return to its initial resting potential level. In the cases where this return has been followed, it was observed to approach asymptotically the same Em as it had prior to antidromic activation, the time 113 course of this return being several minutes. This effect on the taste cells by antidromic activity in the sensory fibers does not appear to have an equilibrium potential and changes in resting cell membrane polarization even to the complete reversal of the Em (Figure 25) had little effect on the Ep. The modulation of the antidromic Ep by taste stimuli (Figure 19) therefore is not likely to be mediated by changes in Em produced by these chemicals. If, however, sensory fiber discharge is a function of receptor membrane potential change, then the effects produced by antidromic activity on Em may serve to modulate the dis- charge rates of afferent chemosensory fibers. The hyper— polarizing as well as depolarizing components of these anti- dromically activated potentials could be seen to participate in both depression and enhancement effects of stimuli at the peripheral level. An example of these opposite effects on the sensory response to HCl is seen in Figure l7b,c. Wang and Bernard (1969) have reported peripheral enhancement and depression to chemical stimulation in the tongue of the cat, and Miller (1971) and Beidler (1969) have shown this effect in the rat. The effect of a stimulus on the receptor cell Em, the effects on antidromic Ep, Eacc, Tp and race as well as the latent electrogenic component responsible for the "rebound phenomenon" may be considered additional factors in the coding 114 of stimulus quality which classically has been ascribed to the spatio-temporal effects of the stimulus. Taste Receptor Potentials The fungiform papilla taste disc contains two layers of cells which may be reSponsible for the slow potentials re- corded upon antidromic stimulation of the sensory nerve fibers of the IX nerve. The surface layer of goblet shaped cells is not considered to have a sensory function but the ”rod" cells of the deeper layer have been classically reported as the sensory receptors of the taste system in the frog (Beale, 1869; DeHan and Graziadei, 1971; Ecker, 1889; Hammer- man, 1969; Kolmer, 1910; Stensaas, 1971). Synaptic junctions between cells of the taste disc and nerve fibers have been reported only for the "rod" cells (DeHan and Grasiadei, 1971; Stensaas, 1971; Uga and Hama, 1967; Uga, 1966). Figures 8 and 10a,b show that antidromic stimulation initiated depolarization in surface cells and hyperpolariza- ' tion in cells at a depth corresponding to that of the "rod" cells of the fungiform papilla. In addition, Figure 12 shows that the depolarizing responses were not confined to cells of the taste disc, but could be found in cells of the tongue epithelium not associated with taste sensibility. These epithelial cells, however, are joined by "zonae occludentes" or tight junctions among themselves and with the "rod" cell 115 processes they envelop within the taste disc itself (Dehan and Graziadei, 1971; Stensaas, 1971). The fact that the depolarization of these surface cells followed the potential changes of the ”rod" cells with little difference in latency (Figures 22, 23) and similar stimulus-response characteristics may indicate a passive role in current flow within the papilla. Such a hypothesis is suggested for papilla elements in amphibia by Herrick (1925) and for the toad by Eyzaguirre £2 31. (1972). The theoretical considerations on the effects of passive ele- ments by Llinas and Bloedel (1969), and Lorente de No (1947) are applicable to the depolarizing responses recorded from surface cells within and without the fungiform papillae of the frog tongue. If they serve as passive current sources for the receptor elements within the taste disc,the depolarizing cells would contribute to the peripheral modulation of the effects of taste stimuli on the receptors. Such functions have been considered for smooth muscle by Tomita (1966) and in the inte- gration of activity in the olfactory bulb of rat, rabbit and cat (Rall and Shepherd, 1968). This is especially significant in view of the findings that epithelial cells respond to chem- ical stimuli by a change in electrical potential (Eyzaguirre gt_gl., 1972). While time courses of changes in potential and 1p in epithelial cells follow each other rather closely (Figure 22), it can be seen that cells which are in the receptor region of the taste disc (Figure 23) do not exhibit such a relationship in the initial phases of response to 116 antidromic invasion. The passive elements of a volume con- ductor would be expected to follow the potential changes of the electrotonically-coupled active elements. Potential changes in the depolarizing cells of the papilla show such a relationship to the hyperpolarizing cells. A metabolic effect of antidromic stimulation which affects the time con- stant, Tp of current changes in the cells directly affected by such activity would not be expected to be passed to the passive element of the volume conductor. Figure 23 shows that the time constant changes in the hyperpolarizing cells did not follow the Ep changes of these cells. HoweVer, for the same time periods (6-17 seconds post antidromic activation), the Ep and 1p changes in depolarizing cells are similar and follow the Ep changes of the hyperpolarizing cells. Such an effect would be expected in a system where cells are electrotonically coupled but not metabolically coupled. The taste disc of the frog fungiform papilla appears to satisfy the theoretical requisites for such a system. Evidence for a Metabolic Component of Taste Cell Response The long latencies of taste cell reSponse to antidromic nerve stimulation indicate a delayed effect not consonant with conductance changes produced by synaptic transmission in electrochemical systems (Hubbard, Llina's and Quastel, 1969). 117 Single electrical shocks to the IX nerve did not elicit a potential change in the taste disc cells until l-l.5 seconds post-stimulus (Figure 14). In some cases these effects persisted for several minutes beyond cessation of stimulation. The time course of the effects of repetitive stimulation shows that relatively little augmentation of response Ep occurred after the first four seconds of stimulation (Figures 8, 10, 15, 16). The Ep was more sensitive to the number of low frequency pulses (1-5 per second) than to the frequency of the stimula- tion 225 ES (Figures 13, 14). Blood flow in the papilla was important for the preservation of the antidromic effects. Hellkant (1971) found that rat taste bud function was also dependent on adequate arterial blood flow. Activation of a metabolically driven sodium pump has been described for synaptic activity in bullfrog sympathetic ganglia (Nishi and Koketsu, 1969), mammalian sympathetic gang- lia (Libet and Tosaka, 1969) and mammalian smooth muscle (Burnstock st 21., 1963). The post-synaptic effect of nerve stimulation in these systems was hyperpolarizing, inhibitory, and of a latency on the order of seconds. Gorman et 31. (1967) reported that the hyperpolarization produced in gang- lion cells of a marine mollusc by antidromic nerve stimulation was the result of activation of such a pump, and that this pump was not sensitive to the membrane potential of the cell. Marmor and Gorman further showed that this metabolic component of the cell was temperature dependent, warming of the cell 118 ' producing an enhanced hyperpolarization. Similar metabolic pumps have been demonstrated in Purkinje cells of the rat cerebellum (Siggins and Oliver, 1971), in the medulla of the cat (Hosli and Haas, 1972) and in the superior cervical gang- lion of the rabbit (Torda, 1972). Bourgoignie et_al. (1969) showed that in cells of the toad bladder this hyperpolarizing pump was dependent on cyclic AMP and that the change in mem- brane potential did not follow the increase in transmembrane current produced by the metabolic sodium pump. The cyclic AMP activated pump is not dependent on the transmembrane po- tential and produces hyperpolarization which may persist for several minutes (Siggins et_§l., 1971, Torda, 1972). The activation of C-AMP is intrapostsynaptic (Torda, 1972) and therefore a result of activity initiated by release of some transmitter from the presynaptic terminals (Hosli and Haas, 1972). Reviewing the characteristics of the hyperpolarizing response of taste disc cells to antidromic nerve stimulation, the similarity of this response with those described above for cells utilizing a metabolically activated electrogenic pump seems evident. The long latency and prolonged duration out- lasting the nerve activity by several orders of magnitude; the slow following rate to nerve stimulation; the dependency on adequate circulation; the depolarizing rebound after adap- tation to long stimulus trains; the independence of the Ep from the membrane potential, Em; the differential change in 119 1p and Ep over time; the Opposite effects of some chemical stimuli on the Em and Ep of the taste receptors--all these are evidence for a mechanism utilizing an electrogenic meta- bolic pump for the peripheral modulation of sensory informa- tion in the gustatory system of the frog. Significance of This Work On theoretical grounds Grundfest (1965) suggests of re- ceptors: "Occurrence of a response of indefinite amplitude and duration depending upon the stimulus, sometimes of depolarizing and other times of hyperpolarizing potentials, depending upon the cell and/or the stimulus represents a group of properties which also provide indirect evidence (of re- ceptor function as being more than that able to be accounted for by purely_passive ionic components).“* Considering the receptor potential: "It should not be forgotten however that the potential need have no function, and indeed that the activity of the receptor cell need have no electrical sign." (Grundfest, 1969.) The interpretation of the data presented in this disser—’ tation supports Grundfest's hypothesis as applied to the receptors of the frog taste disc. Furthermore, in addition to the factors affecting receptor response described by Grundfest, this work indicates that a metabolic component, *Italics mine. 120 activated by antidromic activity in chemoreceptor nerve fibers can exert an important effect on the receptor, and contribute to the peripheral processing of sensory information in the frog gustatory system. If similar peripheral mechanisms are functional in other vertebrate taste systems, the findings presented in the pres- ent work may serve as a model for taste interactions in general. Cognizance of this effect may help to explain some of the differential effects of similar taste substances (NaCl and KCl) as well as similar tastes of different molecular species (sucrose and lead salts). This model is attractive in that it offers a mechanism for the peripheral enhancement or depression of taste fiber response as well as for the lateral effects seen in taste afferents with overlapping re- ceptive fields. The change in apparent taste qualities over the course of adaptation to prolonged stimulation becomes more clear when one recognizes the prolonged effects of rela- tively small amounts of nerve activity on the receptor ap- paratus. This obviates the necessity of postulating nerve fiber interactions and behavior which have not been found in any other neural sensory system. In addition, application of the concepts of this model for peripheral interaction can help amend the discrepancy between theoretical transform functions and observed results for the initial phasic and pro- longed tonic components of afferent chemosensory nerve responses. The relative independence of this peripheral 121 effect from the receptor transmembrane potential may help absolve sensory neurophysiologists studying taste from the apparent inconsistencies in coupling of receptor potential changes with sensory nerve fiber discharges. It is believed that this model can serve as a useful tool for understanding the sense of taste and some components of the disease pro- cesses in those afflictions such as cystic fibrosis, adrenal dysfunction, vitamin deficiencies and familial dysautonomia which are reflected in taste abnormalities. IX. CONCLUSIONS Electrical stimulation of the lingual branch of the IX nerve of the frog produced changes in the membrane poten- tial of cells forming the taste disc of the fungiform papillae. These effects were due to the antidromic acti- vation of sensory fibers and not associated with efferent activity in sympathetic fibers. Currents in cells of the taste disc surface were depolar- izing and appeared to be electrotonically coupled to the hyperpolarizing potentials recorded from the cells of the receptor layer. Antidromically elicited potential responses exhibited latency, summation, adaptation and post-stimulation re- bound. This rebound was of a polarity opposite to the initial change produced by antidromic stimulation and of the same direction as the change accompanying adaptation. Depending on stimulus conditions, antidromic nerve activ— ity was able to produce depression or enhancement of chemo- sensory fiber discharge in response to taste stimuli. Different taste stimuli were found to have potentiating or depressing effects on the antidromically elicited poten- tials of taste disc cells. 122 123 6. Electrical stimulation of single taste papillae led to potential changes in taste disc cells of a neighboring papilla belonging to a common receptive field. These potentials were similar to those produced by antidromic stimulation of the whole lingual nerve but were of less amplitude, presumably due to the fewer number of nerve fibers antidromically activated. The antidromically elicited potentials did not exhibit a dependency on the resting level of polarization of the taste disc cells. The properties and behavior of these antidromically eli- cited responses were characteristic of those postulated for sensory receptors by Grundfest and demonstrated in other neural systems to be produced by a metabolically activated electrogenic pump. BIBLIOGRAPHY Ariens-Rappers, C. U., G. C. Huber, and E. C. Crosby. 1936. The Comparative Anatomy of the Nervous System of Verte- ‘brates, IncludingéMan. Macmillan Co., New York. Andres, K. H. 1965. Der Feinbau des Bulbus Olfactorius der Ratte unter besonderer Berucksichtigung der synaptischen Verbindungen. Z, Zellforsgh. 65:530-561. Beale, L. S. 1869. New observations upon the minute anatomy of the papillae of the frog's tongue. Quart. J. Micr. Sci. (New Series). 9:1-18. i Beidler, L. M. 1953. Properties of chemoreceptors of tongue of rat. J. Neurophysiol. 16:595—607. Beidler, L. M. 1961. Mechanisms of gustatory and olfactory receptor stimulation. In Sensory Communication. Ed. W. A. Rosenblith. Wiley and MIT Press, New York. Bernard, R. A. 1971. Antidromic inhibition: a new theory to account for taste interactions. In Research in Physi- ology. Eds. F. F. Kao, K. Koizumi and M. Vassalle. Aulo Gaggi, Bologna. Bernard, R. A. 1972. Antidromic inhibition: a new model of taste receptor function. In Olfaction and Taste IV. Proc. 4th Int. Symp. Wissenschaftliche Verlagsgesell- schaft MBH. Suttgart. Boeqkh, J., K. E. Kaissling and D. Schneider. 1965. Insect olfactory receptors. In Cold Harb. Symp. on Quant. Biol. 30:263-280. Cold Spring Harbor, Long Island, N. Y. Bourgoignie, J., S. Guggenheim, D. Kipnis, and S. Klahr. 1969. Cyclic guanosine monophosphate: effects on short- circuit current and water permeability. Science. 165: 1362-1363. Brush, A. D., B. P. Halpern. 1970. Centrifugal control of gustatory responses. Physiol. Behav. 5:743-746. 124 125 Burnstock, G., G. Campbell, M. Bennett and M. E. Holman. 1963. Inhibition of the smooth muscle of the taenia coli. Nature (London). 200:No. 4906:581-582. Casella, C. and G. Rapuzzi. 1963. Interactions fonction- nelles entre les papilles linguales de la grenouille. Etude electrophysiologique. J. Physiol. (Paris). 55: 219-220 0 Chernetski, K. E. 1964. Sympathetic enhancement of peri- pheral sensory input in the frog. J. Neurophysiol. 27:493-515. Chernetski, K. E. 1965. Cephalic sympathetic fibers in the frog. J. Comp. Neurol. 122:173-179. DeHan, R. and P. P. C. Graziadei. 1971. Functional anatomy of frog's taste organs. Experientia. 27:fasc. 7:823-826. Dodt, E. and J. B. Walther. 1957. Wirkungen zentrifugaler Nervenreizung auf Thermoreceptoren. Pfl. Arch. ges. Physiol. 265:355-364. Dubner, R. 1967. Interaction of peripheral and central in- put in the main sensory trigeminal nucleus of the cat. Exptl. Neurol. 17:186-202. Eccles, J. C. 1957. The Physiology of Nerve Cells. Johns Hopkins Press. Baltimore. Ecker, A. 1889. The Anatomy of the Frog. Transl. G. Haslam. Clarendon Press. Oxford. Eldred, E., H. N. Schnitzlein and J. Buchwald. 1960. Response of muscle spindles to stimulation of the sympa- thetic trunk. Exptl. Neurol. 2:12-25. Erici, I., B. Folkow and B. Uvnas. 1951. Sympathetic vaso- dilator nerves to the tongue of the cat. Acta. Phys. Scand. 25:1-9. Erici, I. and B. Uvnas. 1951. Efferent and antidromic vaso- dilator impulses to the tongue in the chorda-lingual nerve of the cat. Acta Physiol. Scand..25:IO-l4. Esakov, A. I., 1961. The efferent control of receptors (on the example of the chemoreceptors of the tongue). Bull. Exptl. Biol. Med. (Russ). 51:257-262. New York Consultants Bureau, New York. 126 Esakov, A. I. and A. L. Byzov. 1971. Electric reaction of the frog taste cells to centrifugal stimulation. Bull. Exptl. Biol. Med. (Russ). 72:723-726. New Yerk Con- sultants Bureau, New York. Eyzaguirre, C., S. Fidone and P. Zapata. 1972. Membrane potentials recorded from the mucosa of the toad's tongue during chemical stimulation. J. Physiol. 221:No. 3:515- 532. Falk, G. 1961. Electrical activity of skeletal muscle. In Biophysics o£:Physiological and Pharmacological Ac- tions. Ed. M. Shanes. Am. Assn. Adv. Sci., Washington. Faull, J. R. and B. P. Halpern. 1972. Taste stimuli: time course of peripheral nerve response and theoretical models. Science. 178:73-75. Filin, V. A. and A. I. Esakov. 1968. Interaction between taste receptors. Bull. Exptl. Biol. Med. (Russ). 65:9-11. New York Consultants Bureau, New York. Flock, A. 1965. Transducing mechanisms in the lateral line canal organ receptors. In Cold Spr. Harb. Symp. Quant. Bio. 30:133-144. Fons, M. 1969. Electrogustometry. 3. Adaptation in elec- trical taste. 'J. Laryng. 83:973—980. Fuortes, M. G. F. 1971. Generation of responses in recep- tors. In Handbook of Sensory_Physiology. Vol. I. Ed. W. R. Lowenstein. Springer-Verlag, Berlin. Gannon, R. P. and C. A. Laszlo. 1968. Effect of atropine on the latency of cochlear potentials. J. Neurophysiol. 31:419-427. Gasser, H. 1935. Changes in nerve potentials produced by rapidly repeated stimuli and their relation to the re— sponsiveness of nerve stimulation. Am. J. Physiol. 111:35-50. Gasser, H. S., C. H. Richards and H. Grundfest. 1938. Properties of the nerve fibers of slowest conduction in the frog. Am. J. Physiol. 123:299-306. Gaupp, E. 1904. A. Ecker's and R. Wiedersheim's Anatomis ges Frosches. Vol. II. Verlag F. Vieweg und Sohn, Braunschweig. 127 Gorman, A. L. F., M. Mirolli, and G. C. Salmoiraghi. 1967. Unusual characteristics of an inhibitory potential recorded intracellularly from a molluscan nerve cell. Fed. Proc. 26:329. Graham, H. T. 1935. The subnormal period of nerve response. Am. J. Physiol. plll:452—465. Grundfest, H. 1959.. Synaptic and ephaptic transmission. In Handbook of Physiology. Vol. I. Neurophysiology. Am. Physiol. Soc., Washington, D. C. Grundfest, H. 1961. Excitation by hyperpolarizing poten- tials. A general theory of receptor activities. In 'Nervous Inhibition. Pergamoanress, London. Grundfest, H. 1965. Evolution of electrophysiological varieties among sensory receptor systems. In Essays on ‘Physiological Evolution.. Ed..J. W. S. Pringle. Pergamon Press, Oxford. Halpern, B. 1967. Some relationships between electrophysi- ology and behavior in taste. In The Chemical Senses and ' Nutrition. Eds. M. Kare and O. Maller. JEhns Hopkins Press, Baltimore. ' Hammerman, D. L. 1969. The frog tongue: II. Acta Zool. (Scand.) 50:2:25-33. Hellekant, G. 1971. On the relation between chorda tympani nerve response, arterial tension and blood flow in the tongue of the rat. Acta Physiol. Scand. 82:453—459. Herrick, C. J. 1925. The innervation of palatal taste buds and teeth of Amblystoma. J. Comp. Neurol. 38:389—397. Hosli, L. and H. L. Haas. 1972. The hyperpolarization of neurons of the medulla oblongata by glycine. ' ' Experientia. 28:fasc. 9:1057-1958. Hubbard, J. I., R. Llinas and D. M. J. Quastel. 1969. Electrophysiolggical Analysis of Synaptic Transmission. Williams and Wilkins Co., Baltimore. Hunt, C. C. 1960. The effect of sympathetic stimulation on mammalian muscle spindles. J. Physiol. 151:332-341. Jenerick, H. P. and W. Gerard. 1953. Membrane potential and threshold of single muscle fibers. J. Cell. Comp. Physiol. 42:79-95. 128 Kimura, K. 1961. Factors affecting the response of taste receptors of rat. Kumamoto Med. J. (Japan). 14:95-99. Kimura, K. and L. M. Beidler. 1956. Microelectrode study of taste bud of the rat. Am. J. Physiol. 187:610. Kimura, K. and L. M. Beidler. 1961. Microelectrode study of taste receptors of rat and Hamster. J. Cell. and Comp. Physiol. 58:131-140. Kolmer, W. 1910. Ueber Strukturen im Epithel der Sinnesor- ganeo Anato A112. 36:5-2810 Kuffler, S. W. and C. Eyzaguirre. 1955. Synaptic inhibition in an isolated nerve cell. J. Gen. Physiol. 39:155-184. Kusano, K. 1960. Analysis of the single unit activity of gustatory receptors in the frog tongue. Jap. J. Physiol. 10:620-633. Kusano, K. and M. Sato. 1957. Properties of.fungiform papillae in frog's tongue. “Jap. J. Physiol. 7:324-338. Kuwasawa, K. and R. B. Hill. 1973. Interaction of inhibitory and excitatory junctional potentials in the control of.a myogenic myocardium:. the ventricle of Busycon canalicu- 'latUm. Experientia. 28:fasc. 7:800-801. Landgren, S., G. Liljestrand, and Y. Zotterman. 1954. Chemi- cal transmission in taste fiber endings. Acta Physiol. ‘Scand. 30:105-116. Lewis, T. 1942. Observations upon the vascular axon reflex in human skin, as exhibited by a case of urticaria with remarks upon the nocifensor nerve hypothesis. Clin. 'Sci., Inc. Heart. 4:365-384. Libet, B. and T. Tosaka. 1969. Slow inhibitory and excita- tory postsynaptic responses in single cells of mammalian sympathetic ganglia. J. Neurophysiol. 32:43-50. Lindblom, U. F. 1958. Excitability and functional organiza- tion within a peripheral tactile unit. Acta Physiol. Scand. 44:8upp1. 153:1-83. Lindblom, U. F. and D. N. Tapper. 1967. Terminal pr0perties of a vibro-tactile sensor. Exptl. Neurol. 17:1-15. 129 Llinas, R. and J. Bloedel. 1969. Extracellular field poten— tials in the CNS. In Electrophysiolggical Analysis of Synaptic Transmission. Hubbard, J. I., R. Llinas, and D. M. J. Quastel. Williams and Wilkins Co., Baltimore. Lorente de No,, R. -1947. A study of nerve physiology. Studies from the Rockefeller Institute. 132:16. Macdonald, J. A. 1971. Interaction between gustatory papil- lae of the bullfrog tongue. Proc. Int. Union of Physiol. Sci. 9:No. 1056. Marmor, M. F. and A. L. F. Gorman. 1970. Membrane potential as the sum of ionic and metabolic components. Science. 167:65-67. Miller, I. J. Jr. 1971. Peripheral interactions among single papilla inputs to gustatory nerve fibers. J. Gen. Physiol. 57:1-25. Nishi, S. and K. Koketsu. 1968. Analysis of slow inhibitory postsynaptic potential of bullfrog sympathetic ganglion. J. Neuro. 31:717-728. Nomura, H. and S. Sakada. 1969. Local potential changes at sensory nerve.fiber terminals of the frog tongue. In Olfaction and Taste III. Proc. 3rd Int. Symp. of 01f. and Taste. Ed. D. Pfaffmann. Rockefeller University Press, New York. Ozeki, M. 1970. Hetero-electrogenesis of the gustatory cell membrane in rat. Nature. (London). 228:868-869. Ozeki, M. 1971. Conductance change associated with receptor potentials of gustatory cells in rat. J. Gen. Physiol. 58:688-699. Perri, V. 0., Sacchi and G. Rapuzzi. 1969. Influence of some thiamine antagonists on frog taste receptors. Experientia. 25:fasc. 1:40-41. Pfaffmann, C. 1959. The sense of taste. In Handbook of Physiology. Vol. I. Neurophysiology. Am. Physiol. Soc., Washington, D. C. Pumphrey, R. J. 1935. Nerve impulses from the receptors in the mouth of the frog. J. Cell. and Comp. Physiol. 6:457-467. 130 Reese, T. S. 1966. Further studies on dendro-dendritic synapses in the olfactory bulb. Anat. Rec. 154:408. Reese, T. S. and M. W. Brightman. 1965. Electron microscopic studies on the rat olfactory bulb. Anat. Rec. 151:492. Rall, W. and G. M. Shepherd. 1968. Theoretical reconstruc- tion of field potentials and dendrodendritic synaptic interactions in olfactory bulb. J. Neurophysiol. 31: 884-915. Rall, W., G. M. Shepherd, and T. S. Reese. 1966. Dendro- dendritic synaptic pathway for inhibition in the olfac- tory bulb. 'Exptl. Neurol. 14:44-56. Rapuzzi, G. and C. Casella. 1965. Innervation of the fungi- gorm papillae in the frog tongue. J. Neurophysiol. 28:154-165. Rapuzzi, G. and G. Ricagno. 1970. Slow electric potentials of lingual surface during activation of gustatory receptors in frogs. J. Physiol. (Paris). 62:supp1. 1: 206-207. Rapuzzi, G. and G. Ricagno. 1970. Potentials electriques lents de la surface de la langue et mediation chimique au niveau des recepteurs gustatifs de la grenouille. J. Physiol. (Paris). 62:supp1. 1:207-208. Rapuzzi, G., B. Taccardi and C. Casella. 1961. L'action du BaC1 sur les recepteurs de la langue de grenouille. J. pfiysiol. (Paris). 53:669-678. Robbins, N. 1967a. The role of the nerve in maintenance of frog tongue taste buds. Exptl. Neurol. 17:364-380. Robbins, N. 1967b. Peripheral modification of sensory nerve responses after cross-regeneration. J. Physiol. 192: 493-504. Sato, T. 1969. The response of frog taste cells (Ranaw nigromaculata and Rana catesbeans). Experientia. 25: fasc. 7:709-710. Siggins, G. R., A. P. Oliver, B. J. Hoffer, F. E. Bloom. 1971. Cyclic adenosine monophosphate and norepinephrine: effects on transmembrane.properties of cebellar Purkinje cells. Science. 171:192-194. 131 Spoendlin, H. 1966.. The Organization of the Cochlear Receptor. Karger Publ. Basel. Stensaas, L. J. 1971. The fine structure.of fungiform papillae and epitheliumwof the tongue of a.South.Ameri- can toad, Calyptocephalella'gayi. 'Am. J. Anat. 131: 443-462. Strong, 0. S. 1895. The cranial nerves of the amphibia. A contribution to the morphology of.the.vertebrate nervous system. J. Morph. 10:101-230. Taglietti, V., C. Casella and E. Ferrari. 1969. Interac- tions between taste-receptors in the frog tongue. Pfl. Arch. Physiol. 312:139-148. Tasaki, I., E. J. Polley and F. Orrego. 1954. Action poten- tials from individual elements in cat geniculate striate cortex. J. Neurophysiol. 17:454—474. Tasaki, I. and C. S. Spyropoulos. 1957. The influence of changes in temperature and pressure on the.nerve fiber. In Influence of Temperature on Biological Systems. Ed. A. S. Johnson. Waverly Press, Baltimore. Tateda, H. and L. M. Beidler. 1964. The receptor potential of the taste cell of the rat. J. Gen. Physiol. 47: 479‘4860 Tomita, T. 1966. Membrane capacity and resistance of mammal— ian smooth muscle. J. Theor. Biol. 12:216. Torda, C. 1972. .Hyperpolarization by cyclic AMP. Experientia. 28:1438-1439. Tucker, D. and L. M. Beidler. 1956. Efferent impulses to the nasal area. Fed. Proc. 15:188. Uga, S. 1966.? The fine structure of gustatory receptors and their synapses in frog's tongue. Saibo Kagaku Shimpo— jumu. 16:75-86. Uga, S., andK. Hama. 1967. Electron microsc0pic studies on the synaptic region of the taste.organ of carps and frogs. J. Electron Microscop. (Japan). 16:269-277. Wachtel, H. and E. Kandel. 1967. A direct.synaptic connec- tion mediating both excitation and inhibition. Science. 158:1206-1208. 132 Wang, M. and R. A. Bernard. 1969. CharaCterization and interaction of taste responses in chorda tympani fibers of the cat. Brain Res. 15:567-570. Yamashita, S. 1963. Stimulating effectiveness of cations and anions.on.chemoreceptors in the frog tongue. Jap. J. PhYSiOlo 13:54-63. Yamashita, S. 1964. Chemoreceptor response in.frog, as modi- fied by temperature change. Jap. J. Physiol. 14:488-504. Zotterman, J.~ 1949. The response of the frog's taste fibers to the application of pure water. Acta Physiol. Scand. 18:181-189. WWIGIWNSIMIWlulIWI‘VIWIWW“WIT!“ 31293 03015 5054