INFORMATION TO USERS
This manuscript has been reproduced fiom the microfilm master. UMI films the text directly firom the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be fi’om any type of computer printer.
The quality of this reproduction is dependent upon the quality of the copy subm itted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely afifect reproduction.
In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing fi~om left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & £towell Infonnation Company 300 North Zed) Road, Ann Arbor NO 48106-1346 USA 313/761-4700 800/521-0600
EFFECTS OF ACUTE AND CHRONIC CHORDA TYMPAN! DENERVATION ON TASTE RESPONSES IN THE NUCLEUS OF THE SOLITARY TRACT
DISSERTATION
Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University
By
Mark E. Dinkins, B.S., D.D.S
*****
The Ohio State University 1999
Dissertation Committee: Approved by Susan Travers, Ph.D., Adviser
Keith Alley. D.D.S., Ph.D.
Michael Beattie, Ph.D. Susan Travers, Ph.D. Adviser Scott Herness, Ph.D. Oral Biology Graduate Program ÜMX Number: 9919858
UMI Microform 9919858 Copyright 1999, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized copying under Title 17, United States Code.
UMI 300 North Zeeb Road Ann Arbor, MI 48103 ABSTRACT
The central effects of peripheral deafferentatlcn have been well described in visual, somatosensory and auditory systems. However little is known of central plasticity in the gustatory system. A limited number of anatomical, neurophysiological, behavioral and psychophysical studies suggest that changes do occur in the central taste system after peripheral deafferentation. The purpose of this project was to determine whether physiological changes occur in the first order central taste relay, the nucleus of the solitary tract (NST) after acute or chronic chorda tympani (CT) denervation. Ten percent of multi- and single-unit sites in NST increased their response to taste stimulation by an average of 33% during acute CT denervation (anesthesia). In addition, we noted a decrease in the mean spontaneous rate of NST cells during anesthesia. This effect was robust for cells that responded to taste mixture stimulation of the anterior tongue and hard palate. The CT innervates taste buds located on the anterior tongue and the greater superficial petrosal nerve innervates taste buds on the hard palate. Since the taste response elicited by palatal stimulation was affected little by anesthesia, but the spontaneous rate greatly decreased, in essence, there was an increase in the signal-to-noise ratio for these cells. It appears that taste compensation after CT anesthesia may be partly due to
disinhibition and an increase in the signal-to-noise ratio of a subset of NST cells.
Chronic CT denervation (transection) resulted in increased posterior tongue-
elicited taste responses of NST cells. Also, there was a trend for increased
palatal-elicited taste responses of NST cells. However, the orotopic organization of taste and tactile responses in the NST was unaltered after denervation.
These results suggest that intact nerves increase their functional input in their normal terminal fields in the NST. Based on these studies, the gustatory system at the level of the NST has the potential for neural plasticity to compensate for lost input, but appears more resistant to the large-scale changes that have been found in other sensory systems. More importantly, these results provide us with more information concerning the normal processing of taste input in the central nervous system.
Ill First of all, I would like to dedicate this to my parents, Susan and William, my stepfather, Paul, and my sister, Lori, who through hard work impressed upon me the values and principles that I believe in. Secondly, to my grandmother, Macil, who allowed me the opportunity to pursue higher education at The Ohio State
University. Finally, to my wife, Susan, and my son, Jake, who have made my life so fulfilling. I thank all of you for your love and encouragement.
IV ACKNOWLEDGMENTS
I would like to thank my adviser, Dr. Susan Travers, for her tireless mentorship and dedication to this project. I have learned a great deal from this experience due to her expertise and knowledge.
I thank Dr. Joe Travers for his many insightful comments on several manuscripts that I have written. I am also grateful for his guidance in this project.
Many individuals associated with the lab have helped me immensely in this endeavor and I am sincerely grateful. I have had many interesting conversations with Drs. Chris Halsell, Hamid Karimnimazi, and Lisa DiNardo who have aided me in the direction of this project. I am fortunate to have had help from the following research assistants/associates who have done an excellent job: Mrs. Elizabeth Hauswirth, Dr. Hecheng Hu, and Mr. Kevin Urbaneck.
I would like to thank the members of the Dissertation Committee for their commitment and helpful comments that have improved the quality of this dissertation as well as enriching my educational experience.
I thank Dr. Ilene Bernstein and Mr. Mitch Roitman for their role in the second study.
This research was supported by a Dentist-Scientist Award from the
National Institutes of Health (NIDCR). VITA
February 25, 1967 ...... Bom-Canton, Ohio
1992...... B.S. in absentia. Microbiology, The Ohio State
University
D.D.S., Dentistry, The Ohio State University
1992-present...... Graduate Teaching and Research Associate,
Endodontics and Oral Biology, The Ohio State
University
PUBLICATIONS
Manuscripts 1. Zwilling, B.S., M. Dinkins, R. Christner, M. Paris, A. Griffin, M. Hilburger, M. McPeek, D. Pearl. Restraint stress-induced suppression of major histocompatibility class II expression by murine peritoneal macrophages. J. Neuroimmunology. 29:125-130. 1990.
Abstracts 2. Dinkins, M.E. and C.F. Shuler. Keratin analysis of full thickness oral mucosal equivalents. J. Dental Research. 69(special issue):225. March 1990. Abstract 1174.
3. Dinkins, M. and E.L. Dabreo. Color change of two maxillofacial elastomers during polymerization. J. Dental Research. 70(special issue):290. April 1991. Abstract 200.
VI 4. Harrer, M. I., M. Dinkins, J. B. Travers, and S. P. Travers. Gustatory elicited c-fos expression in brainstem nuclei. Society for Neuroscience Abstracts, 1993
5. Dinkins, M.E. and S.P. Travers. Neurophysiologic study of taste compensation following nerve anesthetization in rats. J D ental Research, 75(special issue). 1996. Abstract 743.
6. Dinkins, M.E. and S.P. Travers. Alternative Mechanism for Taste Compensation Following Chorda Tympani Anesthetization. Chemical Senses, abstract, 1996.
7. Dinkins, M.E., I. Bernstein, M. Roitman and S.P. Travers. Lack of Orotopic Reorganization in the Adult NST Following Neonatal Chorda Tympani Transection. J D ental Research, abstract, 1998.
8. Dinkins, M.E. and S.P. Travers. Chorda Tympani Nerve Transection in Neonatal Rats Results in Altered Taste Responses of Adult NST Neurons. Society for Neuroscience Abstracts, 1998.
FIELDS OF STUDY
Major Field: Oral Biology
VII TABLE OF CONTENTS
ABSTRACT...... ii
DEDICATION...... iv
ACKNOWLEDGMENTS...... v
VITA...... Vi
LIST OF FIGURES...... xi
CHAPTERS:
1. INTRODUCTION...... 1
1.1 Plasticity in sensory system s ...... 1 1.1.1 Overview...... 1 1.1.2 Visual sy stem ...... 2 1.1.3 Somatosensory system ...... 2 1.1.4 Trigeminal system ...... 6 1.1.5 Mechanisms of plasticity ...... 7 1.2 Plasticity in the gustatory system ...... 7 1.2.1 Developmental plasticity ...... 7 1.2.2 Clinical observations ...... 8 1.2.3 Psychophysical studies ...... 10 1.2.4 Physiological studies ...... 10 1.2.5 Taste phantoms ...... 11 1.3 Phantom limb pain ...... 11 1.4 Objectives...... 13
2. EFFECTS OF CHORDA TYMPANI ANESTHESIA ON TASTE RESPONSES IN THE N S T ...... 15
2.1 Introduction ...... 15 2.2 Materials and Methods ...... 17
VIII 2.2.1 Subjects and A nesthesia ...... 17 2.2.2 Neurophysiologic surgical preparation...... 18 2.2.3 Neurophysiologic recording sessio n ...... 19 2.2.4 Taste stimulation ...... 19 2.2.5 Chorda tympani anesthetization ...... 21 2.2.6 Histologic reconstruction of recording sites ...... 22 2.2.7 Quantification of neural activity ...... 22 2.2.8 Stability of taste responses ...... 25 2.2.9 Statistical Analyses ...... 26 2.2.10 Analysis of Individual Recording S ite s ...... 27 2.3 Results ...... 28 2.3.1 Anatomical Location ...... 28 2.3.2 Classification of recording sites ...... 29 2.3.3 Taste responses before CTN anesthesia ...... 29 2.3.4 The effect of CTN anesthesia on taste responses: averaged responses ...... 30 2.3.5 The effect of CTN anesthesia on taste responses: an individual b a s is ...... 32 2.3.6 The effect of CTN anesthesia on spontaneous activity .. 34 2.3.7 Relative taste responses ...... 35 2.4 Discussion ...... 37 2.4.1 Comparison with Previous Neurophysiological Studies .. 38 2.4.2 Relation to Psychophysical Effects of CTN Anesthesia or Section ...... 41 2.4.3 An increase in signal-to-noise ratio ...... 43
3. EFFECTS OF CHORDA TYMPANI TRANSECTION ON TASTE RESPONSES IN THE N S T ...... 65
3.1 Introduction ...... 65 3.2 Materials and Methods ...... 69 3.2.1 Subjects and anesthesia ...... 69 3.2.2 Chorda tympani transection ...... 69 3.2.3 Neurophysiologic surgical preparation...... 70 3.2.4 Neurophysiologic recording se ssio n ...... 71 a. Multi-unit m apping ...... 72 b. Single cell isolation ...... 73 3.2.5 Taste and tactile stimulation ...... 73 3.2.6 Histologic reconstruction of recording sites ...... 75 3.2.7 Analysis of orotopic representation ...... 76 3.2.8 Quantification of neural activity...... 77 3.2.9 Statistical analyses ...... 78 3.2.10 Neuron ty p es ...... 79 3.2.11 Taste pores and fungiform papillae ...... 80
ix 3.3 Results ...... 80 3.3.1 Orotopic representation of multi-unit resp o n ses ...... SO 3.3.2 General description and orotopic organization-single cells 83 3.3.3 Receptive field organization ...... 86 3.3.4 Altered neural responses in the N S T ...... 88 a. Spontaneous activity ...... 88 b. Taste responses to mixture stimulation ...... 89 c. Taste responses to individual tastant stimulation .... 91 3.3.5 Neuron types ...... 93 3.3.6 Taste pores and fungiform papillae ...... 94 3.4 Discussion ...... 94 3.4.1 Orotopic representation ...... 94 3.4.2 Neural responses ...... 98 a. Spontaneous activity ...... 98 b. Taste responses ...... 99 3.4.3 Sodium and ammonium chloride responses ...... 100
4. GENERAL DISCUSSION ...... 131
4.1 Chorda tympani anesthesia ...... 131 4.2 Chorda tympani transection ...... 133 4.3 Factors influencing our results ...... 134 4.4 Model ...... 137 4.5 Taste phantoms ...... 141
LIST OF REFERENCES...... 148
APPENDIX...... 158 LIST OF FIGURES
2.1 Illustration of chorda tympani anesthetization procedure ...... 45
2.2 Individual whole mouth taste responses before, during and after recovery from CT anesthesia by group...... 47
2.3 Horizontal schematic of the NST depicting location of recording sites. 49
2.4 Taste bud subpopulation responses categorized by group ...... 51
2.5 Mean (± SEM) taste responses for each of the three groups of recording sites for each taste bud subpopulation and anesthetic condition 53
2.6 Mean (± SEM) taste responses for the circumvallate papilla responsive sites...... 55
2.7 Changes in individual taste responses for each taste bud subpopulation during chorda tympani anesthesia ...... 57
2.8 Increased taste responses for individual recording sites during CT anesthesia ...... 59
2.9 Mean (± SEM) relative taste responses for sites in the CT-mixed group. 61
2.10 Peristimulus time histograms of taste responses of a CT-mixed cell. . 63
3.1 Horizontal schematic of the NST depicting the location of 109 multi-unit taste and tactile-responsive sites on intact and cut sides ...... 103
3.2 The number of sites that responded to AO, PO or AO/PO taste or tactile stimulation on each side ...... 105
3.3 Photomicrographs of symmetrically located recording sites made on both sides of the NST ...... 107
XI 3.4 Horizontal schematic of the NST depicting the location of 78 single cells responsive to taste stimulation on intact and cut sides ...... 109
3.5 The number of cells that responded to AO, PO, AO/PO taste stimulation...... Ill
3.6 The number of cells on each side that responded to particular receptive fields ...... 113
3.7 The mean (± SEM) spontaneous rates for all cells and subsets of cells that responded best to a particular receptive field for each side 115
3.8 Taste mixture responses of all cells on each side to individual receptive fields ...... 117
3.9 Individual responses to individual receptive field stimulation with taste mixture ...... 118
3.10 Individual receptive field responses for cells which responded to that receptive field...... 120
3.11 Mean (± SEM) responses to whole mouth stimulation with individual tastants ...... 121
3.12 Response profiles of neuron types found in NST between CT-intact and cut sides ...... 123
3.13 Neural trace of a single cell on the CT-cut side responsive to ammonium chloride but not to sodium chloride ...... 125
3.14 Mean (± SEM) number of taste pores and fungiform papillae ...... 127
3.15 Mean (± SEM) spontaneous rates of nasoincisor duct-responsive (NID) cells from the two studies ...... 129
4.1 Individual responses to individual receptive field stimulation with taste mixture (study 2) ...... 144
4.2 Individual responses to individual receptive field stimulation with taste mixture (study 1) ...... 146
XII CHAPTER 1
INTRODUCTION
Plasticity in sensory systems
a. Overview
Neural plasticity can be defined as a property of nervous tissue to change
its responsivity (Weinberger, 1995). Plasticity has been well documented in
visual, somatosensory and auditory systems. The significance of studying this
phenomenon is several fold. In young animals, alterations in the nervous system after peripheral deafferentation helps us to explain how sensory systems may develop and emphasizes the importance of sensory experience in their normal development. In adult animals, peripheral deafferentation provides us with information concerning the normal processing of sensory input. Even though this dissertation focuses primarily on denervation-induced plasticity, other forms of plasticity such as experience-dependent plasticity may contribute significantly to our understanding of learning and memory. Although well documented in other systems, neural plasticity in the central gustatory system has been studied very little. b. Visual system
Classical studies by Wiesel and Hubei (1963a, 1963b) documented
developmental plasticity induced by sensory deprivation in the visual system of
kittens. In their initial studies, they found that after suturing one eyelid closed,
the number of cells in the primary visual cortex that responded to the other eye
increased greatly. The change in response of these cells appeared to be
cortically-mediated because lateral geniculate nucleus (LGN) cells responded
normally even though there was a significant degree of cell atrophy in LGN
layers responsive to the closed eye. Sensory deprivation had a profound effect
in the first 3 months of life, but was not effective in causing similar changes in adult cats. This work revealed the importance of sensory input in the normal development of the central nervous system and the robust changes that could occur when peripheral deafferentation was performed during a critical period’
(Hubei and Wiesel, 1970). Their results contributed significantly to our understanding of normal central nervous system development as well as experience-dependent plasticity. In general, even though plasticity seem s to be greater in young animals, more recent studies in the visual system have found plasticity in adult animals as well (e.g., Gilbert and Wiesel, 1992).
c. Somatosensory system
Cortical reorganization has been studied in depth in the somatosensory system of adult animals. Reorganization refers to plasticity in the central representation of a receptor epithelium, that is, a change of a representational
2 map (Weinberger, 1995). Merzenlch et al. (1983a) discovered cortical representational plasticity in adult monkeys who had received median nerve transection. After several months of denervation, cortical cells in primary somatosensory cortex that would normally respond to the ventral part of the hand responded to adjacent digits and the dorsum of the hand. These changes progressed over a long period of denervation (Merzenich, et al., 1983b). A number of additional studies documented similar types of changes in the primary somatosensory cortex after deafferentation (reviewed in Buonomano and
Merzenich, 1998). A study by Pons et al. (1991) emphasized the degree to which these changes could occur. Following very long-term deafferentation involving the upper limbs of monkeys, large areas of cortex that would normally respond to the limbs responded to tactile stimulation of the face. Astonishingly, unresponsive areas were not found. These authors proposed that anatomical changes such as axonal sprouting might account for such large scale changes.
Collectively, these studies demonstrate that the somatosensory cortex undergoes progressive changes after denervation, which appear to be a common reaction to lost peripheral input.
The results obtained from cortical reorganizational studies suggest potential mechanisms of normal somatosensory processing. For example, in the raccoon, plasticity occurs immediately after digit amputation (Rasmusson and
Turnbull, 1983). This result suggests that the mechanism responsible (e.g., disinhibition-a release in lateral inhibition caused by denervation) is already in place and probably important in the normal maintenance of receptive field
3 organization. Other Investigators have documented both Immediate and chronic changes In digit representation after amputation In the same animal model
(Calford and Tweedale, 1988). The progressive changes that occur suggest that either the same mechanism results in further cortical changes or additional mechanisms come Into play. In addition to understanding how the central nervous system changes and compensates for lost Input, these studies make
Important contributions In understanding how the somatosensory system normally processes Information. For example, thalamocortical projections extend much further laterally under normal circumstances than what had been detected using standard physiological experimental methods (e.g., extracellular electrophyslology, glucose metabolism, non-lnvaslve Imaging). Inhibitory networks which utilize GABA as a neuromodulator appear to be significant factors In restricting physiologic receptive fields In the cortex and could be a substrate for dislnhibitlon (Jones, 1993). Experience-dependent plasticity lends further information on the normal processing of Input. For example, string players have greater areas of somatosensory cortex dedicated to fingering digits than non-fingering digits or digits from control subjects (Elbert, et al., 1995).
In the current decade, evidence has accumulated suggesting that cortical representational plasticity Is due to changes that occur at lower levels of the neuraxis as well as Intracortlcally. Panetsos et al. (1995) provides us with compelling evidence that cortical changes arise from lower levels of the nervous system. These authors recorded from single cells in nucleus gracilis and primary somatosensory cortex simultaneously. Pairs of cells with overlapping receptive
4 fields were targeted. Evidence from 10 such pairs suggest that immediate
receptive field changes that occur after local anesthesia are due to changes in
nucleus gracilis. After injecting lidocaine subcutaneously into the hindpaw
receptive field, novel receptive fields for the pair of cells were found and they still partially overlapped. Although the previous authors did not rule out descending cortical influence, a study by Pettit and Schwark (1993) does. They provide evidence that disinhibition can occur in the brainstem without cortical influence.
Single cells were recorded in the dorsal column nuclei responsive to hindlimb tactile stimulation before and after infiltrating the cutaneous receptive field with lidocaine. The emergence of novel receptive fields was found immediately in all of these cells (13 of 13). The mechanism responsible appears to be local disinhibition because the same result was found in 10 cells after descending cortical input was removed. Many other studies have confirmed changes in lower levels of the neuraxis after a manipulation that was shown previously to cause cortical changes. For example, monkeys who received therapeutic hand or hand/forelimb amputation demonstrated cortical representational changes. In addition, there were increased terminal fields of intact nerves in the spinal cord and cuneate nucleus (Florence and Kaas, 1995). Similar somatotopic representational changes have been found at the level of the thalamus after median and ulnar nerve transection in monkey (Garraghty and Kaas, 1991).
Therefore, it appears that much of the reorganization seen in the somatosensory cortex reflects cumulative subcortical changes. d. Trigeminal system
Closely related to the somatosensory system, changes after peripheral denervation have been well-documented in the trigeminal system. Whiskers are represented in the cortex in the same pattern that they appear on the whiskerpad of mice. That is, there are rows and columns of whiskers which are represented in a one to one fashion in the cortex. The cortical cells which respond to a single whisker form a cortical barrel. Removing a row of whiskers in neonatal mice results in the formation of a band of cells instead of a row of discrete barrels
(Van der Loos and Woolsey, 1973). Melzer and Smith (1998a, 1998b) have studied the effects of whisker follicle removal throughout the neuraxis in adult mice. Using glucose utilization (quantitative autoradiographic [14C]2- deoxyglucose method), increased activity of denervated barreloids in the brainstem was found after 2 days. These results suggest that adjacent intact whisker representations filled in the denervated area. Since metabolic representational changes occurred at a much slower rate in the cortex, they concluded that unmasking of suppressed inputs in the brainstem accounted for early changes but these were not immediately transferred to the cortex. Instead regrowth of primary neurons may be necessary for cortical changes. Therefore, similar to that found in the somatosensory system, cortical changes in the trigeminal system are largely due to subcortical changes. e. Mechanisms of plasticity
Only a brief overview of mechanisms that may be responsible for neural plasticity will be presented here, since the aim of this research project was to determine whether plasticity occurred in the gustatory system, not to identify specific mechanisms involved. In general, plasticity is thought to occur by morphological (e.g., axonal sprouting, changes in dendritic arborization) and functional (e.g., disinhibition, changes in synaptic strength) changes in the nervous system that result from peripheral perturbation. The molecular changes that can occur to account for possible functional changes are numerous.
Receptor characteristics and expression, neurotransmitter synthesis, storage, release and uptake, and second messenger cascades could all be affected by deafferentation which may result in functional synaptic plasticity (Shaw, et al.,
1994). For example, GABA, a widespread neuromodulator is thought to be important in the maintenance of receptive field size by inhibiting the extensive lateral connections in the somatosensory cortex (Jones, 1993). Disrupting the normal activity of afferent input in the somatosensory system, has been found to decrease GABAergic activity which may account for the increase in receptive field size found after denervation (e.g., Garraghty, et al., 1991).
Plasticity in the gustatory system a. Developmental plasticity
Denervation-induced plasticity has been well studied in the somatosensory and trigeminal systems. Yet, little is known about it in the
7 gustatory system. Most studies that have been done thus far have been aimed
at elucidating the normal development of the taste system rather than
investigating whether the central nervous system can compensate for lost input.
For example, Hill and Przekop (1988) found that matemal sodium deprivation on
or before embryonic day 8 reduced the responses of the chorda tympani (CT)
nerve in the offspring. The CT nerve which innervates taste buds on the anterior
tongue normally responds robustly to sodium chloride. These authors found a
40% decrease in sodium chloride responsiveness compared to controls. An
earlier study demonstrated a decrease in preference for sodium chloride as well
as decreased CT responses (Hill, et al., 1986). Therefore, sodium deprivation
during a sensitive period’ of development decreases the physiologic responses
and dampens preference to a normally preferred stimulus, sodium chloride.
However, unlike the developing visual system, changes responsible for this
effect are probably peripherally mediated, because the effect may be due to a
lack of functional development of amiloride sensitive sodium channels in the taste receptor cell (Hill and Przekop, 1988; Hill and Mistretta, 1990).
b. Clinical observations
Interestingly, humans do not appear to experience long-lasting taste alterations after taste nerve damage. In fact, after CT damage, humans do not complain of taste deficits (Rice, 1963; Bull, 1965). For example, Carl Pfaffmann, who was a preeminent scientist in taste research, had the unfortunate experience of knowing what it was like to suffer taste nerve damage (Miller and
8 Bartoshuk, 1991). After being inflicted with herpes zoster, he experienced a
unilateral loss of taste sensation. Both the CT and the lingual-tonsillar branch of the glossopharyngeal nerve which innervate taste buds on the anterior and
posterior tongue, respectively, were affected by the infection. Essentially, he lost
50% of the taste input that normally arises from the tongue. Despite this profound loss, he did not experience a significant decrease in perceived taste intensity. When he was tested with psychophysical techniques, he did not appreciate taste on the affected side of his mouth; however whole mouth taste intensity seemed normal. Therefore, even though a significant amount of taste input to the CNS was abolished, real world' taste was essentially unaffected.
His case is supported by others who have been studied following more limited taste nerve damage, that is damage to the CT nerve only. After CT nerve damage caused by infections, middle ear surgery, third molar extractions, ablative/reconstructive surgery or trauma, subjects do not typically complain of taste loss. Testing the denervated area reveals a loss of taste (ageusia), but whole mouth taste using the sip and spit’ technique is unaffected (Miller and
Bartoshuk, 1991). Most subjects experience abnormal sensations (dysesthesia, parasthesia) or tastes (dysgeusia, taste phantom) but these typically improve over several months (Bull, 1965). The apparent compensatory phenomenon that occurs after taste nerve damage has been further studied psychophysically using an anesthesia model. c. Psychophysical studies
Lehman et al. (1995) anesthetized the CT nerve of human subjects and
tested receptive fields intraorally with different tastants to determine whether
there was any evidence for taste compensation due to increased taste from
unanesthetized receptive fields. Indeed when the taste buds located in the
trenches of the circumvallate papillae were tested, quinine seemed stronger.
This effect was even more pronounced contralaterally. The authors
hypothesized that this was due to disinhibition in the central nervous system
(CNS). Specifically, they proposed that input from the CT nerve normally inhibits
taste responses In the CNS mediated by the glossopharyngeal nerve. When the
CT is damaged or in this instance anesthetized, the glossopharyngeal-mediated
taste responses are released from inhibition, therefore perceptually the
magnitude of the stimulus increases.
d. Physiological studies
Physiologic evidence for disinhibition in the gustatory system arose from a
study of taste responses in the nucleus of the solitary tract (NST) by Halpern and
Nelson (1965). While recording from NST cells in rats these investigators
anesthetized the CT nerve to determine whether the response originated from
CT or glossopharyngeal input. A curious finding was uncovered when they
recorded taste responses from an area of the NST that received input from both of these nerves. That is, during CT nerve anesthesia the remaining taste response increased in magnitude. Naturally it was assumed that the response
10 which was resistant to anesthesia was mediated by the glossopharyngeal nerve since palatal taste buds were not known at that time. The authors proposed that
CT input normally inhibits responses of the glossopharyngeal nerve and under
CT anesthesia inhibition is released and the glossopharyngeal responses increase in magnitude. NST seems to be a reasonable place to find disinhibition since there is an extensive GABAergic inhibitory interneuron network present
(Liu, et al., 1993). Since that time, many researchers have proposed this as a mechanism for taste compensation after CT nerve damage or anesthesia. The aim of the first study in this dissertation was to test this hypothesis. That is, to determine more definitively whether CT anesthesia results in increased taste responses from other receptive fields in the NST of rats.
e. Taste phantoms
In addition to taste compensation, during CT anesthesia human subjects have reported "taste phantoms” (Yanagisawa, et al., 1998). A taste phantom, as the name implies, is a taste sensation for which there is no apparent physical stimulus (Miller and Bartoshuk, 1991). Bull (1965) reported that 80% of patients that had chorda tympani nerve damage, experienced adverse symptoms and some of these were taste phantoms.
Phantom limb pain
The cause of taste phantoms after nerve damage is unknown. However, a better understanding of phantom limb pain may also be helpful in
11 understanding taste phantom. Of the possible peripheral and central
mechanisms that have been proposed to cause phantom limb pain, central
reorganization of somatosensory input, has been supported by several recent
studies by Flor et al. (1995, 1998). These authors noted a strong correlation
between the degree of cortical reorganization and the degree of phantom limb
pain. Specifically, the greater the movement of the lip representation toward the
denervated area of the primary somatosensory cortex, the greater degree of pain the patient experienced. Although this finding does not prove that reorganization causes phantom pain, it suggests that phantom pain may be due in part to cortical neurons receiving novel input and the behavioral sequelae to this physiological mechanism is experienced as pain. Changes which occur at lower levels, such as the spinal cord, may also be critical initiators of phantom pain.
For example, sprouting of large myelinated fibers occurs in the dorsal horn of the spinal cord after nerve damage (Woolf, et al., 1992; Wilson and Kitchener,
1996). Specifically, the central endings of type A fibers in which the peripheral component was transected, sprout into superficial lamina (layer II) which is normally restricted to smaller myelinated and unmyelinated fibers that have an important role in pain transmission. Since the cells in the superficial lamina receive novel input from fibers which usually carry information about innocuous stimuli, these stimuli may now be misinterpreted as noxious. Similar information concerning central reorganization in the taste system is not known. It Is tempting to speculate that changes similar to those that occur in the somatosensory system occur in taste as well.
12 Based on behavioral and psychophysical observations it appears that
compensation occurs in the taste system after peripheral deafferentation.
However, only a few studies have focused their attention toward determining the
neural consequences of deafferentation. Adult CT terminal field volume is
decreased after taste receptor damage when performed in 2 day old rats (Lasiter
and Kachele, 1990). However, the physiologic consequences after denervation
are unknown.
Objectives
Given the limited amount of information that is available concerning denervation-induced plasticity in the taste system, it was our objective to determine whether peripheral deafferentation leads to physiologic changes in the taste system of rat. Specifically, Study 1 was designed to determine whether CT anesthesia results in altered responses of NST neurons suggestive of a disinhibitory mechanism or other changes that could account for taste compensation after nerve anesthesia/damage. Study 2 investigated the effects of chronic CT denervation (transection) in a model similar to one which resulted in behavioral changes (Sollars and Bernstein, 1996). That is, the adult orotopic representation of taste responses in the NST was studied after CT transection in
10 day old rats. Changes in neural responsiveness were also analyzed to determine whether novel responses, increased convergence or increased magnitude of taste responses occurred following long-term denervation.
13 These studies had the potential of contributing information about the normal processing of taste input in the NST as well as determining the system’s ability to compensate for lost input. Although these experiments were not aimed at elucidating specific mechanisms, the results could be suggestive of mechanisms that have been proposed in other systems. This information could be useful in formulating hypotheses regarding how compensation as well as adverse sequelae occur after denervation.
14 CHAPTER 2
Effects of Chorda Tympani Anesthesia on Taste Responses in the NST
Introduction
Patients who suffer from chorda/lingual nerve damage due to trauma, surgery, Infection or pathosis do not typically report deficits In taste (Rice, 1963;
Bull, 1965; reviewed In Miller and Bartoshuk, 1991). Although testing discrete taste bud subpopulations reveals an absence of taste sensation on the denervated areas, whole mouth taste perception appears to be normal. In fact, when tested using the “sip and spit” technique. Intensity ratings for a variety of testants are only modified slightly following anesthetization of the chorda/lingual or chorda tympani nerve (CTN) (Catalanotto, et al., 1993; Lehman, et al., 1995;
Miller and Bartoshuk, 1991). This phenomenon Is referred to as “taste constancy" (Lehman, et al., 1995) and based on early neurophyslologlcal observations (Halpern and Nelson, 1965), has been hypothesized to result from the removal of putative Inhibitory CTN Influences on cells that receive excitatory inputs from other gustatory nerves within the NST (Catalanotto, et al., 1993;
Lehman, et al., 1995; Miller and Bartoshuk, 1991). Additional psychophysical evidence for “release of Inhibition” was obtained In a recent experiment In which
15 CTN anesthesia produced increases in the perceived intensity of quinine applied to the circumvailate papillae (Lehman, et al., 1995).
The consequences of CTN anesthesia in humans appear to fit nicely with neurophysiological findings in rats reported more than 30 years ago (Halpern and Nelson, 1965). In a classic study of the first-order gustatory relay, the nucleus of the solitary tract (NST), Halpern and Nelson (1965) reported that posterior tongue taste responses increased in magnitude after CTN anesthesia.
These observations would suggest that plasticity capable of compensating for partial taste loss exists in the initial stages of central processing. However, because the main purpose of their study was to investigate gustatory topography and chemosensitivity, many questions regarding anesthetic effects persisted.
Importantly, neither the frequency nor magnitude of the residual response increases was clear because CTN anesthetic effects were tested only at a few recording sites. In addition, the identity of the taste receptors giving rise to the enhanced responses was, in fact, unknown. The authors used an anterior tongue chamber to specifically stimulate only taste buds on the anterior tongue.
However, taste buds outside the anterior tongue chamber were non-specifically stimulated. Because the distribution of palatal taste buds was not considered at that time, it was presumed that the taste buds which were stimulated outside the chamber were foliate and circumvailate receptors on the posterior tongue.
However, responses could also have arisen from palatal taste buds, which comprise 17% of oral taste buds in the rat (reviewed in Travers and Nicklas,
16 1990). Thus, taste responses arising from the palate may increase during CTN anesthesia in addition to those from the posterior tongue.
The aim of the present study was to more fully investigate the acute effects of CTN anesthesia on gustatory responses in the rat NST. Specifically, we wanted to determine the frequency of disinhibition, quantify its magnitude, and characterize which taste bud subpopulations were involved. On the basis of previous neurophysiological and psychophysical results, we hypothesized that disinhibition would occur frequently, be robust, and arise from posterior tongue stimulation.
Materials and Methods
Subjects and Anesthesia
Thirty-seven adult male Sprague Dawley rats (287-601 gm) were used in this study. Animals were anesthetized with ethyl carbamate (urethane, 1 g m/kg,
IP) and sodium pentobarbital (Nembutal, 25 mg/kg, IP) to achieve a surgical level of anesthesia. This was characterized by an absence of pedal withdrawal upon pinching and lack of corneal blink reflex, and was maintained throughout the experiment with supplemental doses of Nembutal. Animal procedures were approved by the Ohio State University’s Institutional Laboratory Animal Care and
Use Committee.
17 Neurophysiologic surgical preparation
Surgical preparatory procedures for acute neurophysiologic recording were similar to those described in previous work from this laboratory (e.g.,
Travers, et al., 1986; Travers and Norgren, 1995). An exception to this was the initial perforation of the tympanic membrane ventral to the malleus using a sharp retraction needle. Great care was taken during this procedure to avoid damaging the chorda tympani nerve which lies medial to the malleus. The perforation allowed for rapid diffusion of lidocaine and saline to the chorda tympani nerve (CTN). Animals were placed on a heating pad to keep them near a constant rectal temperature of 37® 0. Animals were stabilized on a stereotaxic apparatus using a mouthpiece and atraumatic earbars (Kopf instruments) which were modified so that a plastic cannula (PE-50, Becton Dickinson, NJ) could be introduced into the external auditory meatus while the head remained stable in the stereotaxic (see Figure 2.1). The hollow earbars were a modified version from a schematic drawn by Norgren {personal communication). The animal's head was leveled with respect to lambda and bregma landmarks in the horizontal plane. A head holder device was fastened to one earbar and attached to the skull via small bone fixation screws and secured with methyl methacrylate. The advantage of this head holder is to stabilize the rat's head during recording while being able to stimulate discrete taste bud subpopulations in the oral cavity
(described in Travers, et al., 1986). A tracheal cannula was placed to allow for unimpeded respiration during fluid delivery. An oral drain tube, used to evacuate excess fluid, was placed exiting the same ventral incision (modified from Halpern
1 8 and Nelson, 1965). The superior laryngeal nerves were routinely transected: the hypoglossal nerves were transected some of the time. Sutures were placed at four sites around the oral cavity and through the tongue to allow adequate access for the stimulation of different taste bud subpopulations (Travers, et al.,
1986; Halsell, et al., 1993). A craniotomy was performed posterior to lambda in order to access the brain for microelectrode penetration. Physiologic saline was applied to the exposed area of the cerebellum.
Neurophysiologic recording session
Glass-coated tungsten microelectrodes (0.4-2.4 Mohm) were used to record multi- and single-unit neural activity. Neural activity was amplified, observed on an oscilloscope and recorded on VMS tapes for off-line analysis.
Recording sites were marked with electrolytic lesions made with anodal current
(3pA, 3 sec, Grass stimulator) at the recording site or at a site that was typically
200-400pm ventral to it.
Taste stimulation
In the main set of experiments (n=36) responses to gustatory stimulation of the whole mouth, anterior tongue, nasoincisor ducts and foliate papillae were tested. Occasionally, other taste bud subpopulations (soft palate, sublingual organ, retromolar mucosa) were also stimulated. Testing commenced with stimulating the whole mouth with a mixture of testants (0.3M sucrose, 0.3M
19 NaCI, 0.01 M HCI, 0.003M quinine hydrochloride) and then individual taste bud
subpopulations were tested. Whole mouth stimulation consisted of sequentially
flowing 2 ml of water, 2 ml of taste mixture, and then 4 ml of water rinse over the
lingual, palatal, and buccal mucosa using a syringe. Individual taste bud
subpopulations were stimulated in a similar water-stimulus-rinse sequence.
Small amounts of water and then mixture were applied to the taste buds of interest with a nylon brush and then the whole mouth was rinsed with water from a syringe (Travers, et al., 1986; Travers and Norgren, 1995).
In a separate subset of animals (n=5), sites responsive to circumvailate gustatory stimulation were identified by placing a modified glass pipette in the trench surrounding this papilla to provide for adequate stimulation of these receptors, which are located in the walls of the trench (Frank, 1991; Halsell and
Travers, 1997). Because the pipette assembly made it awkward to stimulate other taste bud groups, only whole mouth and circumvailate responses were routinely tested and recorded in these preparations. However, before formal testing commenced, additional taste bud subpopulations were always screened for a response. Sometimes responses to foliate stimulation were present, and in these cases, foliate responsiveness was also tested.
For each stimulus trial, spontaneous activity was recorded for 5-10 seconds preceding stimulation and the water, tastant, and rinse applications were of an equal duration. Neural activity was allowed to return to baseline before the next stimulation (usually at least 60 seconds). Mechanoreceptive responses were noted but not systematically tested.
2 0 Chorda tympani anesthetization
The CTN was anesthetized by administering approximately 0.02 ml of 2%
lidocaine into the external auditory meatus which would then diffuse across the
perforated tympanic membrane and anesthetize the nerve, commonly within 10-
20 sec. In order to hasten recovery from anesthesia, 1-2 ml of physiologic saline
was delivered via the same route after testing. The duration of CTN anesthesia
exceeded the duration of the stimulation protocol (approximately 15 min or less)
and recovery typically required 5-10 min once the nerve was rinsed with saline.
The introduction of lidocaine and physiologic saline was via a polyethylene tube
attached to a 1 ml syringe in which the tube was advanced through a hollow ear
bar until the end of the tube was flush with the blunt end of the earbar (see
Figure 2.1). For each preparation, the anesthesia and recovery procedure was
initially tested at a site within the nucleus of the solitary tract (NST) responsive to gustatory stimulation of the anterior tongue to establish the parameters (volumes of solutions and time) for reliable anesthetization and recovery. Responses to stimulation of the whole mouth and individual taste bud subpopulations with taste mixture before, during and after recovery from chorda tympani anesthesia were tested at each site. The post-anesthesia testing was important for establishing the variability of the responses in the unanesthetized state. Whenever possible, the entire stimulus protocol was repeated.
21 Histologie reconstruction o f recording sites
After the recording session, animals received a lethal dose of sodium pentobarbital (150 mg/kg). They were then perfused intracardially with physiologic saline (300-400 ml) and fixed with 10% buffered formalin (200-300 ml). The brain was dissected from the cranium and stored in a 10% formalin
20% sucrose mixture for cryoprotection. Brains were sectioned at 52 pm on a freezing microtome, mounted on chrome-alum coated slides, and alternate sections were stained for NissI substance (cresylecht violet) or myelin (Weil).
Recording sites were reconstructed by tracing brainstem sections through the microscope, which was interfaced with a computer using commercially-available hardware and software (Vidlucida, Microbrightfield). Electrolytic lesions
(approximately 100-150 pm in diameter) were identified on cresylecht violet and
Weil-stained sections and traced relative to the NST, solitary tract, vestibular nuclei, spinal trigeminal tract and brainstem outline. The location of the recording sites in the antero-posterior and medio-lateral dimensions were transposed to a schematic outline of the NST in the horizontal plane (adapted from Hamilton and Norgren, 1984).
Quantification of neural activity
Recorded neural activity was analyzed off-line. Single-unit activity was differentiated with a window discriminator using consistency of amplitude and waveform as criteria. Multi-unit activity was differentiated by setting the lower
2 2 level of the window discriminator just above the background level (see Halsell, et
al., 1993; DIckman and Smith, 1989; Halsell and Frank, 1992). Because the
analyses in this study involved comparing responses at the same recording sites during different anesthetic states, normalization was not necessary for the multi
unit responses. Both single- and multi-unit activity were quantified by converting action potentials to digital pulses and accumulating these in 500 ms bins in peristimulus time histograms.
Net-evoked activity was quantified by using a standard response measure, defined as the number of spikes over a 5 or 10 sec period (10 sec. were used when available) during taste stimulation minus the number of spikes that occurred during the preceding water stimulation. This measure was then converted to spikes/sec. An exception to this definition was for circumvallate- elicited responses. These were calculated as the mixture response minus spontaneous, rather than water-evoked activity due to the frequent occurrence of a large but transient mechanoreceptive and/or thermal response to fluid onset.
The initial water flow typically evoked a response but it adapted quickly. If a second water stimulation immediately followed the initial water stimulation, a second response was not evident. Thus, subtracting the water response would have underestimated the gustatory contribution. This situation was unique for circumvailate stimulation through the pipette. Transient responses to fluid stimulation sometimes occurred when stimulating other taste bud groups, but in these cases the water and gustatory stimulations were discontinuous, preventing somatosensory adaptation. The criteria for a suprathreshold taste response
23 were defined as a minimum 1 spike/sec change in activity, which also had to be
greater than 2.5 times the standard deviation of the spontaneous rate (Travers
and Norgren, 1995; Travers and Norgren, 1991; Travers and Smith, 1984;
Travers, et al., 1986).
In addition to the standard measure just described, we used a second
measure of taste-evoked activity that we have termed the relative taste
response. The relative taste response was the standard response divided by
spontaneous activity. Since CTN anesthesia often caused marked decrements
In spontaneous activity, analyzing relative responses had the potential to reveal
response changes otherwise unapparent using the standard measure analysis.
Although the standard measure also incorporates changes in spontaneous activity, it represents net-evoked spikes, and does not fully reflect certain proportional changes in responsiveness that occurred during anesthetization.
For example, at recording sites responsive to both anterior tongue and nasoincisor duct stimulation, the spontaneous rate sometimes decreased to near zero. Using the standard measure, net nasoincisor-evoked activity usually remained unchanged. Viewed from another perspective, however, it could be argued that the nasoincisor response actually increased during anesthesia, since the same number of spikes were evoked relative to a lower baseline. The relative response measure quantifies this putative increase.
24 stability of taste responses
It was critical to determine that changes observed during CTN anesthesia were due to anesthetization and not simply to variation over the course of testing. To this end, we compared responses in the anesthetized state with those both prior to and after recovery from anesthesia. As a result, we excluded sites with responses that we defined as unstable. Whole-mouth responses were used to determine stability, except at circumvallate-responsive sites where circumvailate responses were used since they were more reliable. Unstable whole mouth responses were defined as sites where responses occurred before anesthesia but not after recovery, or where the percent change in the unanesthetized state exceeded 50% of the mean unanesthetized response:
% change in unanesthetized state = (Pre-Post)/((Pre+Post)/2)*100
where Pre= response before CTN anesthesia, Post= response after recovery from CTN anesthesia.
Using this criterion, the recording sites retained exhibited a mean change in responsiveness in the unanesthetized state of 18.5 ± 12.4% (SD), with a quarter of the sites varying by 10% or less and over half of the sites (61%) by
20% or less. Figure 2.2 depicts individual responses before, during, and after recovery from anesthesia, and suggests a high degree of response stability prior to and after recovery from anesthesia. Across sites, pre- and post- anesthesia responses were very highly correlated (r=+.9B, P < .0005).
25 Statistical Analyses
Sites were categorized into 1 of 3 groups based upon their gustatory receptive field response prior to CTN anesthesia since chorda tympani-mediated responses were expected to be eliminated whereas non-chorda tympani- mediated responses were expected to remain the same or increase during CTN anesthesia. Sites which responded only to stimulation of taste bud subpopulations innervated by the CTN (e.g., anterior tongue, sublingual organ, retromolar mucosa) were placed in the CT group. Sites which responded to taste bud subpopulations innervated by the CTN plus another nerve (i.e., glossopharyngeal or greater superficial petrosal) were included in the CT-mixed group. Sites which responded to a taste bud subpopulation innervated solely
(nasoincisor duct, circumvailate papilla, soft palate) or principally (foliate papillae) by a nerve other than the CTN were classified as “non-chorda tympani” (non-
CT). A subset of sites (n=7) within the non-CT group which responded to circumvailate stimulation were also referred to as non-CT^,,. One potential complication for this scheme is that the foliate papillae also receive a minor innervation from the CTN (Whiteside, 1927; Miller, et al., 1978; Yamamoto and
Kawamura, 1975). This issue is addressed in the Discussion.
Repeated measures ANOVAs were performed to compare the three anesthetic conditions. Separate analyses were done for each taste bud subpopulation and spontaneous activity for each group of recording sites. Multi- and single-unit sites were collapsed for these analyses, because nearly identical results were obtained when the analysis was restricted to single units. ANOVAs
26 were followed by post-hoc contrasts comparing pre- and post-anesthesia responses with each other and with the responses in the anesthetized condition.
Probability values for contrasts were Bonferonni-adjusted, and significance levels set at P s0.05. Unless noted, the P values in the text are the adjusted P values for the contrasts, which assume a significant main effect for anesthetic condition.
In a few instances it was of interest to compare the magnitude of responses for different types of recording sites. These comparisons were performed using t- tests and restricted to single-unit sites. Unless stated otherwise, variances are reported as S.E.M.s.
Analysis of Individual Recording Sites
The strength of analyzing average responses was that enough data were available to use inferential statistical methods. However, a potential limitation was that we might have missed effects that occurred in specific subsets of the sample. For example, if responses at one-third of the recording sites increased while another third decreased by a similar amount during anesthesia, it would have appeared as though no changes had occurred overall, even though they had in a significant percentage of the population. Because insufficient data for standard statistics were available for individual sites, an alternative criterion was developed for making reasonable judgements about whether a change in responsiveness occurred during anesthesia. The criterion used the standard deviation of the gustatory responses before and after recovery from CTN anesthesia as a measure of variability in the unanesthetized state. If the
27 response during anesthesia deviated from the mean of the pre- and post
anesthesia responses by more than two standard deviations, the response was
considered to have been altered during the anesthetic state. For sites which met
this criterion, the change in response (%) was calculated using the following
formula:
% change during anesthesia = ((Anesth-(Pre + Post)/2)/(Pre + Post)/2)*100
where Anesth= response during CTN anesthesia, Pre= response before CTN anesthesia , Post= response after recovery from CTN anesthesia.
Results
Anatomical Location
Gustatory responses from fifty-nine multi- and single-unit sites were recorded before, during and after recovery from chorda tympani nerve (CTN) anesthesia. Subsequent to recording, electrolytic lesions were made either at the site of recording or 200-400 pm ventral to it. Based on histologic reconstruction of 41 sites, all appeared to be within the boundaries of the nucleus of the solitary tract (NST) (see Figure 2.3). A topographic organization was observed, with CT and CT-mixed sites predominantly anterior and lateral to non-CT sites. Since all CT and most CT-mixed sites responded mainly to anterior oral cavity stimulation whereas a majority of non-CT sites responded to
2 8 posterior oral cavity stimulation (see below), this organization is similar to
previous descriptions of NST orotopy from this lab (Travers, et al., 1986; Travers
and Norgren, 1995).
Classification of recording sites
Approximately equal numbers of recordings were obtained from multi-
(n=30) and single- (n=29) unit sites; however, these sites were unevenly
dispersed among groups. Seven sites were classified as CT; each was a single
unit site. Twenty-six sites were placed in the CT-mixed group including 10 single
cells. The remaining 26 sites were non-CT sites; 12 were single units. All of the
CT sites responded to gustatory stimulation of the whole mouth and anterior
tongue only. The sites categorized as CT-mixed and non-CT were more
complex. Most CT-mixed sites (n=21) responded to whole mouth, anterior tongue and nasoincisor duct stimulation. Many non-CT sites responded to stimulation of the posterior tongue, 12 to whole mouth and foliate stimulation and another seven to circumvailate stimulation. The receptive fields for all sites are summarized in Figure 2.4. Note that only three of the 59 sites responded to both anterior and posterior tongue stimulation.
Taste responses before CTN anesthesia
Similar to previous investigations (Travers, et al., 1986; Travers and
Norgren, 1995), we noted differences in gustatory responses and spontaneous
29 rate at recording sites with different peripheral inputs. Because this analysis
compared responses at different recording sites and multi-unit activity reflects
the number of recorded units as well as their firing rate, only single-unit
responses were used for these analyses. Prior to anesthesia, the mean whole-
mouth gustatory responses for CT and CT-mixed sites were similar (20.6 ± 6.4
versus 21.2 ± 5.0 spikes/sec respectively, P=0.94, t-test) but responses at CT-
mixed sites were significantly larger that those of the non-CT group (10.0 ± 2.7
spikes/sec, P=0.05, t-test for non-CT vs CT-mixed). A comparable pattern was
observed for spontaneous activities before CTN anesthesia. The average
spontaneous activities were 3.0 ±1.2 spikes/sec for the CT sites, 3.1 ±1.2 spikes/sec for the CT-mixed sites and 0.29 ± .07 spikes/sec for the non-CT sites
(CT versus CT-mixed, P=0.96; non-CT versus CT-mixed, P=0.038, t-tests). In summary, single units in the CT and CT-mixed groups had comparable rates of spontaneous and evoked activity, but non-CT single units were less active.
The effect of CTN anesthesia on taste responses: averaged responses
In the proceeding analyses, we will combine multi- and single-unit data to compare responses before, during and after CTN anesthesia unless noted othen/vise. The mean responses (multi- and single-unit sites combined) for whole mouth, anterior tongue, nasoincisor duct, and foliate papillae stimulation elicited before, during, and after recovery from anesthesia are depicted in Figure
2.5 for the CT, CT-mixed, and non-CT sites not tested for circumvailate stimulation, respectively. Whole mouth and circumvailate responses for non-CT
30 sites responsive to circumvailate stimulation (non-CT^^) appear in Figure 2.6.
Whole mouth and anterior tongue responses for CT sites (Figure 2.5a) remained
stable before and after recovery from CTN anesthesia (P >.1 for both
responses). CTN anesthesia abolished whole mouth and anterior tongue
responses for CT sites (P<0.05 for both responses). Similarly, whole mouth and
anterior tongue responses for CT-mixed sites (Figure 2.5b) remained stable in
the unanesthetized state (P>0.1 for both responses), and decreased during
anesthesia (P<0.001 for both). Nasoincisor duct responses did not change
during CTN anesthesia for CT-mixed sites (main effects: P>0.05). Somewhat
surprisingly, whole-mouth responses for the non-CT sites (Figure 2.5c) were
significantly, albeit slightly, decreased during anesthesia (P<0.01), but remained
stable in the unanesthetized state (P>0.1). The whole mouth decrement at non-
CT sites, however, was not reflected in nasoincisor duct- or foliate papillae- elicited responses for these sites (main effects for both: P>0.1). With minor exceptions, these results were the same when the analysis was restricted to single units in the CT-mixed and non-CT groups. For non-CT^^ sites
(n=7)(Figure 2.6), there were no main effects of anesthesia for either the whole mouth or CV responses (for both P>0.1). In summary, except for the small decrease in the whole mouth response at non-CT sites, the only effect of CTN anesthesia was an abolition of activity evoked by gustatory stimulation of the anterior tongue, which was reflected in an abolition or decrement of whole-mouth
31 responses at CT and CT-mixed sites, respectively. Contrary to our original
hypothesis, no increases in average responsiveness occurred during anesthesia.
The effect of CTN anesthesia on taste responses: an individual basis
As discussed in the Materials and Methods, we analyzed individual as well as averaged responses to avoid missing effects that might occur only for a subset of sites. The results of the individual analysis appear in Figure 2.7, which lists the number of suprathreshold taste responses from multi- and single-unit sites which increased, decreased or did not change during CTN anesthesia, categorized by taste bud subpopulation. With this analysis, six of 59 (10.2%) sites did exhibit response increases during anesthesia that exceeded the criteria for a reliable change; i.e., they were twice as great as the standard deviation of the responses in the unanesthetized state. Figure 2.8a depicts mean responses in the unanesthetized state (±SDs) compared to the responses that occurred during anesthesia, for these six sites. In general, the increases were small and most frequently (4/6) involved nasoincisor duct responses. When responses were averaged across these six sites, responses in the unanesthetized state
(pre- versus post-anesthesia) changed by 8.9%, compared to an increase of
32.9% during anesthesia. The responses for one neuron with an augmented response are shown in more detail in Figure 2.8b. Prior to and after recovery from anesthesia, this neuron responded to whole mouth and nasoincisor duct, but not anterior tongue or foliate papillae stimulation. The responses to palatal
32 stimulation before and after recovery from anesthesia were nearly identical, 14.9
vs 14.3 spikes/sec, but during anesthesia, the response increased to 17.6
spikes/sec, a 20.5% increase. The small magnitude of this increase seems even
less impressive since there is not a similar increase in the whole mouth
response, despite that fact that the cell apparently received no CTN input. The
largest absolute increase in response during CTN anesthesia was for a foliate
papillae response which increased by 10 spikes/sec (extreme right example in
Figure 2.8a).
As predicted by the averaged data, rather than increasing during CTN
anesthesia, most responses decreased (79 of 141, 56%) or did not change (56
of 141, 40%). That the effects of anesthesia were consistent and our criterion
sensitive is supported by the fact that nearly all responses elicited by stimulating
CTN-innervated receptor subpopulations— i.e., 29/30 anterior tongue responses
and the single sublingual response, met the criterion for a decrease. In contrast to the increases just discussed, the average decrement for the anterior tongue responses was nearly complete (x= 99.5%). The consistency of this anesthetic effect is apparent in the individual whole mouth responses depicted in figure 2.2.
Anesthetizing the CTN abolished responses to gustatory stimulation of the whole mouth for all CT sites, and produced decrements in the whole-mouth responses for nearly all of the CT-mixed sites. In addition to these expected decrements, it was interesting that response decrements were also observed for receptor subpopulations innervated solely or principally by a nerve other than the CTN.
Not surprisingly these decrements were smaller and less frequent. Thus, 7/23
33 nasoincisor, 1/7 circumvailate, and 1/4 soft palate responses exhibited
decrements during CTN anesthesia (x= 41.6%). Even for foliate-elicited
responses, decrements were not common. These papillae are principally
innervated by the glossopharyngeal nerve but also receive minor CTN
innervation. However, only 3/16 foliate responses exhibited decrements during
anesthesia, and they were also small (x= 18.6%).
The effect o f CTN anesthesia on spontaneous activity
In addition to abolishing responses evoked by anterior tongue stimulation,
another salient effect of CTN anesthesia was a decrease in spontaneous activity.
This effect occurred for all types of recording sites but was more pronounced for some (Figure 2.5). Relative to the average spontaneous activity prior to anesthesia, spontaneous activity during CTN anesthesia decreased by 100% for
CT, 65.2% for CT-mixed, and 13.1% for non-CT sites. The average spontaneous activity for the 7 cells in the CT group dropped from 3.0 spikes/sec to 0.0 spikes/sec, although this decrease only approached significance
(P=0.068), probably due to a floor effect and the small number of cells. Although decreases in spontaneous activity for the CT-mixed and non-CT multi- and single-unit sites were smaller, both were significant (P<0.001, P=0.05, respectively). The ten single units in the CT-mixed group also reflected the overall decrease in spontaneous activity during CTN anesthesia (P=0.05), although this was not true of the non-CT single units. In addition, no change during anesthesia was noted for sites responsive to circumvailate stimulation.
34 These effects on spontaneous activity were also evident when individual
multi- and single-unit recording sites were analyzed. Most sites exhibited
decreases in spontaneous firing during CTN anesthesia (29 of 59, 49%) or did
not change (27 of 59, 46%), whereas only a few exhibited increases (3 of 59,
5%). Decrements occurred most frequently (6 of 7, 86%) for CT sites and spontaneous activity was nearly abolished (x=98.1%). Many CT-mixed sites (19 of 26, 73%) also had decrements in spontaneous activity during anesthesia and these were fairly large (x= 68.1%). Fewer non-CT sites (4 of 26, 15%) decreased and the change was smaller but notable (x=22.5%).
Relative taste responses
The relative response as a measure of quantifying the present data was conceived because of the widespread decrease in spontaneous activity that occurred during CTN anesthesia. The relative response was calculated as the standard gustatory response (net-evoked activity) divided by spontaneous activity. Except for CT sites, which were virtually silent during anesthesia, relative responses were calculated for all CT-mixed and non-CT multi- and single-unit sites, except in the few cases (4/26 CT-mixed and 3/26 non-CT sites) where this was not possible because of the total lack of spontaneous activity.
Similar to what was apparent for responses calculated in the standard fashion, anterior tongue relative responses at CT-mixed sites decreased during anesthesia (anesthetized versus unanesthetized, P<0.001). However, the
35 effects of anesthesia were different for whole mouth and nasoincisor duct
responses, calculated using the relative (Figure 2.9) versus the standard (Figure
2.5) measures. In contrast to the decrease apparent for standard whole-mouth
responses at these sites (Figure 2.5), the relative response did not change
during anesthesia (main effect: P>0.1, Figure 2.9). Most strikingly, nasoincisor
duct relative responses actually increased in the anesthetized state
(anesthetized versus unanesthetized, P<0.005, Figure 2.9), in contrast to the
lack of an anesthetic effect when responses were calculated using the standard
measure (Figure 2.5). To illustrate this point further. Figure 2.10 displays
histograms of neural activity (spikes/sec) before, during and after recovery from
anesthesia for a cell in the CT-mixed group. During anesthesia there is an
obvious decline in whole mouth taste response and a complete abolishment of
anterior tongue response. However, the evoked nasoincisor duct taste response
is unaltered during anesthesia while the spontaneous activity decreased
profoundly. Therefore, the remaining nasoincisor duct response relative to a
decreased spontaneous activity has increased. This was a common finding for
individual CT-mixed sites and is supported by the mean data (figure 2.9). In
contrast to what was found for CT-mixed sites, analysis of relative responses for the non-CT and non-CT^^ sites revealed a pattern of results very similar to those obtained when standard responses were analyzed. Anesthesia did not affect averaged relative responses for these sites.
36 Discussion
Contrary to our original hypothesis, the present investigation did not reveal
an average increase during chorda tympani nerve (CTN) anesthesia in net firing
evoked by stimulation of any taste bud subpopulation for any group of recording
sites. Instead, anterior tongue responses were abolished and whole mouth
responses abolished or diminished, respectively, at CT and CT-mixed sites,
demonstrating the efficacy of our anesthetization procedure. Although mean
responses were not enhanced, there were six individual sites where responses
increased during anesthesia. Specifically, six sites exhibited responses that
exceeded the average response in the unanesthetized state by standard
deviations. Although there were insufficient data at individual recording sites for
statistical analysis, this criterion is similar to using a 95% confidence interval,
making it reasonable to suggest that these increases do not represent random
variability. Although apparently reliable, increases occurred less frequently than
predicted. It is, of course, possible that more subtle excitatory effects of CTN anesthesia were missed.
In addition to their infrequent occurrence, increases were small and the receptive fields of the augmented responses were not entirely consistent with our original hypothesis. We predicted that responses arising from taste bud subpopulations innervated by the glossopharyngeal nerve would increase following CTN anesthesia. However, of the 17 foliate papillae and seven circumvallate-responsive sites tested, only one with a foliate input increased
37 during anesthesia. Of the five remaining sites where response increases
occurred, four were CT-mixed sites, and the enhanced responses occurred
following nasoincisor duct stimulation. Therefore, disinhibition of taste
responses actually predominated for gustatory signals arising from taste buds
innervated by a branch of the Vllth, not the IXth nerve. A complication of
interpreting the negative data for IXth-nerve mediated responses arises for
foliate papillae stimulation. That is, foliate taste buds receive some CTN
innervation, mainly to those receptors in the anterior two folds of these papillae
(Whiteside, 1927; Miller, et al., 1978; reviewed in Travers and Nicklas, 1990).
Thus, it is conceivable that response increases occurred for the
glossopharyngeal component of the foliate response, but were masked by decrements occurring for the CTN component. However, such an explanation seems unlikely given the minimal amount of CTN-foliate innervation. There is no complication for interpreting the lack of increased IXth-nerve mediated responses for circumvallate stimulation, since these receptors are entirely innervated by the IXth nerve (Whiteside, 1927).
Comparison with Previous Neurophysiological Studies
The increases in gustatory responsiveness that occurred following CTN anesthesia appear to agree with the results of Halpern and Nelson (1965) in their magnitude, although they seem to differ in their frequency of occurrence. At
“composite” recording sites, i.e., those sites responsive to stimulation of the anterior tongue and to receptors outside the tongue chamber, Halpern and
38 Nelson found that instilling 3% mepivicaine in the ipsilateral external auditory
meatus abolished anterior tongue responses but that responses evoked by
stimulating receptors outside the chamber “tended to increase”. Although the
previous study was primarily qualitative, the authors quantified the response at
one site before and during CTN anesthesia. Integrated responses from
stimulation of receptors outside the chamber increased from 9.5 to 12 units
during anesthesia, a 20.8% increase. The increases in response magnitudes
that we observed were in a similar range, however the frequency with which we
encountered them appears much lower. The previous authors apparently found
consistent increases in five animals (we are assuming that this is 5 of 5 animals
since the authors did not report the total number of animals tested in this way).
In contrast, we observed that anesthetic-induced increments occurred only at a
minority (3/24) of recording sites that appear analogous to theirs; i.e., most of our
CT-mixed sites. Indeed, at seven of our sites, we found decrements in non-
anterior tongue responses. The reason for the lower proportion of anesthetic-
induced increases in our study is not clear although we tested a larger sample of
recording sites and compared responses in the anesthetized state to those both
prior to and after recovery from anesthesia. Although the previous authors
attributed response increases to posterior tongue stimulation, we found that
responses following palatal stimulation increased more frequently during
anesthesia. However, the interpretation by the previous authors is probably
merely a result of the fact that palatal taste buds had not been described at that time, in combination with the non-specific stimulation techniques used.
39 The effects of chorda tympani anesthesia or acute nerve cuts have also
been studied in the major synaptic target of nucleus of the solitary tract (NST)
efferents, the parabrachial nucleus (PBN) (Norgren and Pfaffmann, 1975;
Miyaoka, et al., 1997). In agreement with the present study, neither investigation
provides evidence for release of inhibition, although the posterior tongue was not
specifically stimulated in either case. Norgren and Pfaffmann (1975) recorded
from single PBN neurons before, during, and after recovery from CTN
anesthesia with 2% lidocaine. Ten cells responded to stimulation of receptors
outside an anterior tongue chamber and of these, seven responses persisted
during CTN anesthesia. However, none of the remaining responses increased.
Instead, most declined. Similarly, a recent study used an across-animal design to compare responses in a sizeable sample of PBN neurons in CTN denervated and intact animals. Quinine-elicited taste responses did not change between the two groups, suggesting a lack of compensatory change (Miyaoka, et al., 1997).
The non-additive effects of simultaneous stimulation have also been cited as evidence for opposing peripheral interactions in the first-order gustatory relay
(NST) (Grabauskas and Bradley, 1996; Lehman, et al., 1995; Miller and
Bartoshuk, 1991; Sweazey and Smith, 1987). In the hamster NST (Sweazey and Smith, 1987), 11 single neurons responsive to receptors on the anterior tongue and outside the anterior tongue chamber were studied. The response to individually stimulating each region was excitatory, but simultaneous stimulation usually produced responses the same or slightly greater than the largest individual response. Although such effects have been interpreted as
40 excitatory/inhibitory interactions (Grabauskas and Bradley, 1996; Lehman, et a!.,
1995; Miller and Bartoshuk, 1991; Sweazey and Smith, 1987), lack of summation
is not equivalent to inhibition. A recent in vitro slice preparation similarly reported
that stimulation of the solitary tract at sites consisting primarily of Vllth or IXth
nerve fibers usually resulted in both sites evoking excitatory postsynaptic
potentials (Grabauskas and Bradley, 1996). Thus, interactions between
peripheral gustatory influences are ubiquitous in NST but do not appear to be predominantly characterized by opposing inputs from the Vllth versus IXth nerves.
Relation to Psychophysical Effects of CTN Anesthesia or Section
A more complex problem exists in attempting to reconcile our results with human psychophysical data and clinical observations which provide evidence for compensatory mechanisms following CTN anesthesia or damage. Unless subjected to spatial testing, CTN anesthesia or damage does not produce notable changes in perceived intensity (Rice, 1963; Bull, 1965; reviewed in Miller and Bartoshuk, 1991). Although experiments in rodents have demonstrated important deficits in threshold detection (Spector, et al., 1990) or qualitative discrimination (Spector and Grill, 1992) for specific combinations of nerve cuts and tastants, the resistance of the gustatory system to partial denervation is a salient feature of its organization in the rat as well. Indeed, one intensive aspect of gustatory-guided behavior, namely, concentration-dependent increases or decreases in lick rate, appear mostly resistant to partial gustatory denervation
41 (Spector, et al., 1993; St. John, et al., 1994). A compensatory increase in
responses to stimulation of residual taste buds is a simple hypothesis which
would explain this resilience. Indeed, Lehman et al. (1995) provided direct
psychophysical support for such a mechanism in humans, demonstrating that
CTN anesthesia induced an increase in the perceived intensity of quinine applied
to the circumvallate papillae. However, we did not observe an increase in
circumvallate-elicited responses at the level of the first-order gustatory relay in
rat. This discrepancy could be due to a species difference, although such a
possibility is difficult to address.
As discussed above, data from the other brainstem taste nucleus, the
PBN, also provide little evidence for response increases following denervation or
anesthesia. However, in other sensory systems, the most striking evidence for anesthesia- or denervation-induced plasticity is for the forebrain, particularly the cortex, although smaller changes have been reported for lower levels (reviewed in Kaas, 1991). Indeed, it is interesting that the psychophysical effects for circumvallate stimulation were observed to be greatercontralateral to CTN anesthesia, although bilateral changes were observed (Lehman, et al., 1995).
The ascending gustatory system is primarily ipsilateral, but some bilateral ascending projections occur at levels rostral to NST and there are opportunities for bilateral interactions via descending projections, e.g., from the cortex to NST
(van der Kooy, et al., 1984; Norgren and Grill, 1976; reviewed in Norgren, 1993).
Thus, compensatory changes originating or occurring at forebrain levels are possible. If cortical influences are necessary for compensatory changes to occur
42 In NST, they could have been dampened in the present study. In this study,
animals were anesthetized with urethane and sodium pentobarbital, and the
latter agent, in particular, decreases cortical activity (Clark and Rosner, 1973)
which in turn can alter NST taste cell responsiveness (Hayama, et al., 1985;
Nakamura and Norgren, 1991). Although we attempted to minimize the total
amount of pentobarbital by combining it with urethane, cortical suppression of activity undoubtedly occurred. Thus, it might be fruitful to reexamine CTN anesthetic effects in NST in a chronic preparation or one which uses a different anesthetic regimen.
An increase in signal-to-noise ratio
Although there was minimal evidence for compensatory increases when responses were quantified using net firing rate, a different perspective is suggested by our alternative “relative” response measure, which quantifies evoked activity on a proportional basis. Specifically, at recording sites receiving both anterior tongue and nasoincisor duct inputs, anesthesia eliminated lingual responses, reduced whole mouth responses, but left net nasoincisor duct- evoked responses unaltered (Figure 2.5). At these sites, CTN anesthesia also reduced spontaneous firing rate by more than 50%. As a consequence, although the relative measure still revealed anterior tongue responses to be eliminated, average nasoincisor duct responses increased by threefold and whole mouth taste responses were unaltered (Figure 2.9). Thus, if information about stimulus intensity is conveyed by proportional increases in neural activity,
43 intensity may be conserved for these recording sites. A similar suggestion has been made with regard to the action of dopamine in the frog olfactory bulb.
Dopamine reduced the spontaneous rate of mitral cells but did not alter olfactory- evoked responses, prompting the hypothesis that this neuromodulator increased the “signal-to-noise” ratio via its effect on spontaneous activity (Duchamp-Viret, et al., 1997). We suggest that the reduction in spontaneous rate induced by
CTN anesthesia might have a similar function in increasing the signal-to-noise ratio of nasoincisor duct-evoked gustatory responses in the NST. It should be kept in mind, however, that if this type of compensation does occur, it is limited to a specific subset of neurons, those receiving convergent inputs from the CTN and other peripheral sources.
Aside from these considerations, CTN anesthetic effects on spontaneous rate are interesting in their own right. Under our experimental conditions, nearly all the spontaneous activity in neurons that respond only to anterior tongue stimulation can be blocked by CTN anesthesia, suggesting that peripheral inputs contribute importantly to baseline activity. Since many CTN afferents are sensitive to stimulation with sodium salts (Frank, et al., 1983), activation by salivary sodium ions may contribute to the spontaneous activity in central neurons receiving inputs from this source. A relatively high spontaneous activity in anterior-tongue responsive fibers and their central counterparts may serve an important function in allowing the system to respond to both increases and decreases in sodium concentration, either from saliva or external sources.
44 Figure 2.1
Illustration of chorda tympani anesthetization procedure
Modified earbars were placed in the external auditory meatus which allowed for convenient and reliable anesthesia of the chorda tympani nerve. A
PE-50 cannula was used to administer a small amount (.02 ml) of lidocaine to the external auditory meatus. The close proximity of the cannula orifice and tympanic membrane allow for rapid diffusion of anesthetic through the perforation to anesthetize the CTN.
45 Figure 2.1
46 Figure 2.2 individual whole mouth taste responses before, during and after recovery from
CT anesthesia by group.
The responses (splkes/sec) for all 59 sites are depicted. Note the reliability In responses across the unanesthetized state (pre- versus post anesthesia) and the efficacy of anesthesia apparent as total or partial response decrements for the CT and CT-mlxed sites, respectively.
47 160 ------pre-anesthesia 140 during anesthesia ------post-anesthesia 120 I To © 100 CT CT-mixed / non-CT non- Q. CTcv 0) 80 © (0 60 c 0 Q. 40 1 1 20 0 j\ •20 10 20 30 40 50 60 Site
Figure 2.2
48 Figure 2.3
Horizontal schematic of the NST depicting location of recording sites.
A total of 41 sites are depicted. Symbols for the sites are based upon the response categories described in the text. Note that an orotopic organization of taste responses was found as described previously by Travers and Norgren
(1995).
49 # CT site
■ CT-mixed site
♦ non-CT site ♦ ♦
♦
rostral
medial
.3mm
Figure 2.3
50 Figure 2.4
Taste bud subpopulation responses categorized by group.
All of the sites are described by taste bud subpopulations and categorized accordingly into CT, CT-mixed and non-CT groups. The first number indicates multi- and single-unit sites combined, whereas the number in parentheses indicates single cells only. The following abbreviations are used; WM=whole mouth, AT=anterior tongue, NID=nasoincisor duct, FOL=foliate papillae, SP=soft palate, CV=circumvallate papilla, RM=retromolar mucosa, SLO=sublingual organ.
51 Taste bud subpopulation responses categorized by group Taste bud CT CT-mixed non-CT Total subpopulation WM/AT 7(7) 7(7) WM/AT/NID 21(8) 21(8) WM/AT/NID/ 2(0) 2(0) POL WM/RM/SP 1(1) 1(1) WM/SLO/NID 1(1) 1(1) WM/AT/NID/ 1(0) 1(0) FOL/SP WM/FOL 12(7) 12(7)
WM/SP 3(1) 3(1) WM/NID 2(2) 2(2) WM 1(0) 1(0) WM/NID/FOL 1(1) 1(1) WM/CV 4(0) 4(0) CV 2(1) 2(1) WM/FOL/CV 1(0) 1(0) Total 7(7) 26(10) 26(12) 59(29)
Figure 2.4
52 Figure 2.5
Mean (± SEM) taste responses for each of the three categories of recording sites for each taste bud subpopulation and anesthetic condition.
The abbreviations are as follows for the taste bud subpopulations; wm=whole mouth, at=anterior tongue, nid=nasoincisor duct, fol=foliate papillae, spon=spontaneous rate. Responses appear for pre-, during and post-anesthesia conditions. Note the difference in scales used for each graph and the different orientations used to differentiate each column more clearly. Statistically significant differences are shown by asterisks for anesthetized versus unanesthetized responses when P<0.05 as determined by ANOVAs.
53 a. Mean taste responses for CT sites (n=7)
O ,oi
wm^ post d itr tn g X
b. Mean taste responses for CT-mbced sites (n=26)
S
20 g.
pOM
pxv<
c. Mean taste responses for non-CT sites (n=19)
d u h m g
Figure 2.5
54 Figure 2.6
Mean (± SEM) taste responses for the circumvallate papilla responsive sites.
The whole mouth (WM)- and circumvallate papilla (CV)- elicited responses are shown with standard error bars before (pre-), during and after recovery (post-) from CTN anesthesia. No significant changes were found as determined by ANOVAs.
55 Mean taste responses for non-CTcv sites (n=7)
18 ■ ■ 1 pre-anesthesia 16 during anesthesia ■■■ post-anesthesia g ' 14 To ® 12
« 10 O 8 c(0 0 a 1
WM CV Taste bud subpopulation
Figure 2.6
56 Figure 2.7
Changes in individual taste responses for each taste bud subpopulation during
CT anesthesia.
The following abbreviations are used: WM=whole mouth, AT=anterior tongue, NID=nasoincisor duct, FOL=foliate papillae, CV=circumvallate papilla,
SP=soft palate, SLO=sublingual organ.
The number in parentheses indicate the percent of cases that respond in a particular way for the total number of responses for each taste bud subpopulation.
*The during CTN anesthesia response was 0 spikes/sec, but due to the large variability in responses in the unanesthetized state, the lower value of the criterion was a negative number. Therefore, using this type of analysis we could not conclude that the response was decreased even though it was 0 spikes/s.
57 Changes in individual taste responses for each taste bud subpopulation during CTN anesthesia Taste bud Increased Decreased No change in Total subpopulation response response response WM 1(2) 37(66) 18(32) 56 AT 0 29(97) r(3) 30 NID 4(15) 7(26) 16(59) 27 POL 1(6) 3(19) 12(75) 16 CV 0 1(14) 6(86) 7 SP 0 1(25) 3(75) 4 SLO 0 1(100) 0 1 Total 6(4) 79(56) 56(40) 141
Figure 2.7
58 Figure 2.8
Increased taste responses for Individual recording sites during CT anesthesia.
a. Taste responses increased for 6 individual sites during anesthesia.
The single response in the anesthetized state is compared to the mean (±SD) of the responses before and after recovery from anesthesia ( pre- and post anesthesia responses). The receptive field for the response which increased is listed next to the site number.
b. An individual site which displayed an increased nasoincisor duct taste response during anesthesia, but the whole mouth response remained the same.
59 a. » ■ ■ mean unanesthetized taste response 30 taste response during anesthesia
25
IO. 20 0)
12-NID 108-NID 31-NID 40-NID 41-WM 46-rOL
Site-Taste bud subpopulation
cn I a 10 s c 0 Q. 1
p r e during p o s t
Anesthetic condition
# wtme mouth response O anterior tongue response ipr nasoincisor duct response foliate response ■ spontaneous activity
Figure 2.8
60 Figure 2.9
Mean (± SEM) relative taste responses for sites in the CT-mixed group.
The relative response is defined as the standard taste response divided
by spontaneous activity for a given response. The whole mouth (WM)-, anterior tongue (AT)- and nasoincisor duct (NID)- elicited responses are shown to
highlight the constancy of the whole mouth response during CTN anesthesia.
This is apparently due to the increase in relative NID response versus the decrease in AT response. The statistically significant changes during CTN anesthesia as determined by ANOVAs are indicated by an asterisk.
61 16 pre-anesthesia 14 during anesthesia O post-anesthesia % 12
Q) 10 S 0 CL 8 £ > m 1
WM AT NID Taste bud subpopulation
Figure 2.9
62 Figure 2.10
Peristimulus time histograms for taste responses of a CT-mixed cell.
Filled triangles depict the onset of taste stimulation and open triangles indicate water onset. Note the decreases in net spikes elicited by whole mouth and anterior tongue stimulation, as well as the decrease in spontaneous rate.
However, the net nasoincisor duct response is unaltered.
63 u 25 0) ▼ WM: BEFORE T WM: DURING WM: AFTER V) E o 0 1 '5. CO hliiAUJLiil \Jàu»i ■■
u 0) AT: BEFORE AT: DURING AT: AFTER w E T o o V V
& ill 1II 1 111
o 0) NID: BEFORE NID: DURING (/) NID: AFTER E o 0 1 I u il Ml II 70 0 Time (sec) Time (sec) Time (sec) Figure 2.10 CHAPTER 3
Effects of Chorda Tympani Transection on Taste Responses in the NST
Introduction
Central neural reorganization has been well documented in visual, somatosensory and auditory systems after peripheral deafferentation (reviewed in Weinberger, 1995; Donoghue, 1995; Buonomano and Merzenich, 1998).
However, little is known of the central effects of peripheral deafferentation in the gustatory system. Recent evidence, however, suggests that denervation in young animals may cause central reorganization in the central taste system.
Following neonatal denervation (10 days postnatal), adult rats (60 days of age) demonstrated a behavioral preference for ammonium chloride, a tastant that is not normally preferred (Sollars and Bernstein, 1996). In contrast, rats that received CT nerve transection as adults and were tested 50 days later, did not show this behavioral alteration. Sollars and Bernstein proposed that the difference between neonatal and adult CT-transected animals may be attributed to central reorganization of the remaining afferent input in the first order taste relay, the nucleus of the solitary tract (NST). This hypothesis is plausible, because reorganizational changes in other sensory systems following peripheral
65 deafferentation are more robust in developing animals (e.g., Kalaska and
Pomeranz, 1979; reviewed in Kaas, et al., 1983; O'Leary, et al., 1994; Wilson
and Kitchener, 1996).
Anatomical consequences of peripheral deafferentation have been
studied in the NST of young, developing rodents as well as adults. Lasiter and
Kachele (1990) found a decrease in the CT terminal field volume in the NST
following anterior tongue cautery in 2 day old rat pups. In addition,
transganglionic degeneration was demonstrated in the NST after CT transection
in adult hamsters (Whitehead, et al., 1995). Therefore, peripheral deafferentation results in anatomical changes in the NST of developing and adult
animals that could allow for central reorganization. Specifically, the glossopharyngeal nerve may increase its functional input via increased synaptic efficacy in the CT terminal field because of the loss of effective synaptic sites from the CT.
Increased functional input from the glossopharyngeal nerve after CT denervation is also reasonable considering the rapid developmental changes which occur in the first several weeks of life. Although CT fibers begin to terminate in the NST prenatally, the CT terminal field changes rapidly between birth and postnatal (P) day 30 (Lasiter, 1993). In contrast, development of the glossopharyngeal terminal field distinctly lags that of the CT; its fibers begin to terminate in the NST at 9/10 days of age caudal to the CT terminal field and continue developing to P45-P50 (Lasiter, 1992; Lasiter, 1993). Therefore, CT terminal field changes that occur before the period in which glossopharyngeal
6 6 fibers normally arrive in the rostral NST, may be sufficient to allow these fibers to
Increase synaptic input in the denervated CT terminal field.
If these proposed changes occurred, how might they account for the behavioral observations of Sollars and Bemstein (1996)? Electrophysiological and behavioral data show that the CT and glossopharyngeal nerves convey information about salts differently. The CT innervates taste buds on the anterior tongue and is highly sensitive to electrolytes including the normally preferred salt, sodium chloride, as well as ammonium chloride. Importantly, however, a subset of CT fibers respond selectively to sodium salts (Frank, et al., 1983; Hill, et al.,
1982; Dahl, et al., 1997). The glossopharyngeal nerve, which innervates posterior tongue taste buds, also responds to sodium and ammonium chloride, but these responses are less specific than those in the CT (Frank, 1991). In contrast to CT transection, cutting the glossopharyngeal nerve in adult rats does not result in altered preference or discrimination for sodium (see Markison, et al.,
1995; Spector and Grill, 1992; but also St John, et al., 1997b).
If central reorganization occurred in the NST following CT transection, increased glossopharyngeal input in the denervated CT terminal field could cause a switch in afferent input from sodium-selective CT fibers to less selective electrolyte-sensitive glossopharyngeal fibers. Therefore, ammonium responses in rostral NST cells might be misinterpreted as sodium responses, accounting for an increased preference to ammonium. Given the orotopic arrangement of taste responses in the NST (Hamilton and Norgren, 1984; Travers and Norgren,
67 1995), this type of central reorganization would be reflected as a shift in orotopic
representation of glcssopharyngeal-mediated responses.
The objective of the current study was to determine whether neonatal CT transection leads to adult central reorganization in the NST as reflected in
neurophysiological responses to gustatory stimulation. Specifically, we sought to determine whether there is an overall shift in orotopic representation of taste, in particular a rostral shift of glossopharyngeal-mediated responses. In addition, we investigated whether other forms of plasticity were evident, that is, whether individual neurons increase their response magnitude to stimulation of certain chemicals or receptive fields (i.e., taste bud subpopulations), and whether novel receptive field responses appear. A shift in orotopy, an increase in response to certain taste stimuli (e.g., ammonium chloride) or receptive fields (e.g., a glossopharyngeal-mediated field), and/or appearance of novel receptive field responses could signify increased or altered functional input from the remaining gustatory nerves. Although not all these possibilities simply explain the behavioral changes noted above, they would be consistent with denervation- induced changes in other sensory systems (e.g., Merzenich, et al., 1983a;
Schwaber, et al., 1993; Gilbert and Wiesel, 1992; Pons, et al., 1991). To assess whether a shift in orotopy had occurred, multi-unit recordings were made to systematically map a large area of the NST. Changes in neural responsiveness were determined by recording from individual neurons.
6 8 Materials and Methods
Subjects and anesthesia
Twenty-two male and eight female Long-Evans rats were used In this
study. Five were used for multi-unit mapping and the remaining twenty-five for
single cell recording. Procedures for multi- and single-unit recording were similar
unless noted otherwise. Animals underwent two surgical procedures during the
course of this study. At 10 days of age, rat pups were anesthetized with a
mixture of ketamine and xylazine (30 and 150 mg/Kg, respectively, IP) in order to
achieve a surgical level of anesthesia for CT nerve transection. For
neurophysiologic recording sessions, adult rats (at least 60 days of age, 232-
498 gm, males x=406 gm and females x=262 gm) were anesthetized with sodium pentobarbital alone (Nembutal, 50 mg/kg, IP) or in combination with ethyl carbamate (urethane, 1 gm/kg, IP; Nembutal, 25 mg/kg, IP). A surgical level of anesthesia was characterized by an absence of pedal withdrawal upon pinching the hindlimb and a lack of corneal blink reflex. Supplemental doses of Nembutal were given as needed. Animal procedures were approved by the Ohio State
University’s Institutional Laboratory Animal Care and Use Committee.
Chorda tympani transection
At 10 days of age, rat pups underwent a surgical procedure similar to that of Sollars and Bernstein (1996) except that CT nerve transection was unilateral.
Rat pups were anesthetized (see above) and placed on a heating pad. The
69 surgical site was swabbed with 10% povidone/iodine and a midline longitudinal
ventral neck incision was made using the angle of the mandible as a landmark to
center the incision anterior-posteriorly. Blunt dissection was performed with
modified #5 forceps in order to trace the lingual nerve to the site where the
chorda tympani nerve anastomoses with it. Once the CT nerve was visualized, a
forceps was used to tease its distal end from the lingual nerve and the remaining
proximal portion was removed. Typically, a 5-7 mm piece of CT nerve was
removed, insuring that regeneration would be very unlikely. The wound was
approximated and closed with absorbable 5-0 gut suture. Pups were returned to
the dam after approximately 2 hours. There were no instances of maternal
rejection and it was common to observe the pups suckling from the dam
immediately after their return. The pups were weighed and observed daily for a
1-2 week period. Pups typically gained weight the 1st post-op day and this
continued throughout the observation period. The sutures were commonly
absent within 1-2 days after surgery while the wounds healed uneventfully. One
rat pup, which did not gain weight during the first 3 days post-op, was euthanized.
Neurophysiologic surgical preparation
Surgical preparatory procedures for acute neurophysiologic recording were similar to those previously described (Travers, et al., 1986; Travers and
Norgren, 1995). Briefly, adult animals (at least 60 days of age) were anesthetized (see above) and placed on a heating pad to keep them near a
70 constant rectal temperature of 37®C. Animals were stabilized on a stereotaxic
apparatus using a mouthpiece and atraumatic earbars (Kopf instruments,
Tujunga, CA). The animal’s head was leveled with respect to lambda and
bregma landmarks on the skull in the horizontal plane. A headholder device was fastened to one earbar and secured to the skull via small bone fixation screws surrounded by methyl methacrylate. The advantage of this head holder is to stabilize the rat's head during recording while eliminating the need for a mouthpiece so that one can stimulate discrete taste bud subpopulations in the oral cavity without restriction (described in Travers, et al., 1986). A tracheal cannula was placed to allow unimpeded respiration during fluid delivery. An oral drain tube, used to evacuate excess fluid, was placed exiting the same ventral incision (modified from Halpem and Nelson, 1965). The superior laryngeal nerves were transected bilaterally. Retraction sutures were placed at four sites around the oral cavity and through the tongue to allow adequate access for stimulating different taste bud subpopulations (Travers, et al., 1966; Halsell, et al., 1993). A craniotomy was performed posterior to lambda in order to access the brain for microelectrode penetration. Physiologic saline was applied to the exposed area of the cerebellum.
Neurophysiologic recording session
Neural responses were recorded with glass-coated tungsten microelectrodes (150-700 kQ for multi-unit mapping and 1.0-1.5 MQ for single cell isolation) on the CT-cut and intact sides. Neural activity was amplified and
71 observed on an oscilloscope and audio monitor. Anterior-posterior and medio-
lateral coordinates relative to lambda were noted for each track. Recording sites
(multi- and single-unit) were marked with electrolytic lesions made with anodal
current (3pA, 3 sec, Grass stimulator) at the recording site or at a site that was typically 200-300pm ventral to it.
a. Multi-unit mapping
Many tracks (10-18 per side, x=13.4 intact and x=14.6 cut side) were made in order to construct a detailed map of multi-unit taste and tactile responsiveness in each rat. Electrode tracks were made in a systematic manner usually 200pm apart. Typically, 2-3 tracks at different medio-lateral locations were made per anterior-posterior level. Responses were determined at 50pm intervals dorso-ventrally for each track starting ventral to spontaneous activity characteristic of vestibular nuclei and ending in strong jaw stretch activity characteristic of reticular formation. The receptive field(s) for oral taste and tactile responses were classified qualitatively on the basis of a clear increase in activity using a storage oscilloscope and audio monitor. This qualitative procedure was considered sufficient, based upon agreement between similar qualitative and quantitative categorizations from a previous study (Dinkins and
Travers, in press).
72 b. Single cell Isolation
The CT-intact side was sampled first since it was more efficient to locate taste-responsive neurons on this side and then to use similar coordinates to
locate taste-responsive neurons on the CT-cut side. Also, by using similar
coordinates, similar areas of the NST were sampled on either side. Responses from single cells were recorded on VMS tapes for off-line quantitative analysis.
Taste and tactile stimulation
Neural responses to taste stimulation of the whole mouth, anterior tongue, nasoincisor ducts, foliate papillae and soft palate were assessed. On occasion, the circumvallate papilla was also stimulated. Taste buds on the anterior tongue are innervated by the CT and those within the nasoincisor duct are innervated by the greater superficial petrosal nerve, both are branches of the facial nerve.
Taste buds associated with the foliate and circumvallate papillae are innervated by the lingual-tonsillar branch of the glossopharyngeal and those on the soft palate are innervated by the greater superficial petrosal nerve. The anterior tongue and nasoincisor duct are located within the anterior oral cavity (AO) and the foliate, circumvallate, and soft palate are located within the posterior oral cavity (PC). The testing session started by stimulating the whole mouth with a mixture of testants (0.3M sucrose, 0.3M sodium chloride, 0.01 M hydrochlorous,
0.003M quinine hydrochloride) and then individual receptive fields were tested.
Whole mouth stimulation consisted of sequentially flowing 2 ml of water, 2 ml of taste mixture, and then 4 ml of water rinse over the lingual, palatal, and buccal
73 mucosa using a syringe. Individual receptive fields were stimulated in a similar water-stimulus-rinse sequence. However, instead of delivering stimuli with a syringe, different sized sable hair brushes were used to apply small amounts of water and then taste mixture to the taste buds of interest. Following taste stimulation the whole mouth was rinsed with water from a syringe (Travers, et al.,
1986; Travers and Norgren, 1995). Different sized brushes were used to deliver an adequate amount of fluid, yet constrict the degree of fluid diffusion to other areas. Accordingly, taste buds confined to a small area (foliate and nasoincisor duct) were stimulated with a small brush (#00) and those occupying larger areas
(soft palate and anterior tongue) with larger brushes (#2 and #5, respectively).
Fluid delivery was observed through an operating microscope to insure accurate stimulus application. For each stimulus trial, spontaneous activity was recorded for 10 seconds preceding stimulation and the water, tastant, and rinse applications were of an equal duration. Single cells were also tested using the same protocol for stimulating the whole mouth and individual receptive fields, with individual tastants; 0.3M sucrose, 0.3M sodium chloride, 0.3M ammonium chloride, 0.01 M hydrochlorous and 0.003M quinine hydrochloride. Both multi- and single-unit sites were tested for tactile responsivity using a blunt glass probe which was applied to the buccal mucosa, anterior tongue, foliate papillae, circumvallate papilla, soft and hard palate. Stimulus onset was marked with verbal comments.
74 Histologie reconstruction of recording sites
After the recording session, rats received a lethal dose of sodium
pentobarbital (150 mg/kg). They were perfused intracardially with physiologic
saline (300-400 ml) and fixed with 10% buffered formalin (200-300 ml). The
brain was dissected from the cranium and stored in 10% formalin. The brain
was transferred to a 20% sucrose/10% buffered formalin solution for
cryoprotection several days before cutting. Brains were cut at 52 pm on a freezing microtome, sections mounted on chrome-alum coated slides, and alternate sections were stained for NissI substance (cresylecht violet) or myelin
(Weil). The location of the taste responsive multi- and single-unit sites were reconstructed by tracing brainstem sections through a light microscope interfaced with a computer using commercially-available hardware and software
(Vidlucida, Microbrightfield, Colchester, VT). Electrolytic lesions (approximately
100-150 pm in diameter) were identified on cresylecht violet- and Weil-stained sections and traced relative to the borders of the rostral NST and solitary tract.
For multi-unit mapping, approximately half of the tracks were marked with a lesion so that reconstruction would be as accurate as possible without confusing adjacent lesions. The other half were interpolated from the closest lesion, typically 200pm away but some were 400pm away at the same anterior- posterior level. Almost every single-unit (90/94) was marked with a lesion. The medio-lateral location of multi- and single-unit sites were expressed as a proportion of the distance from the medial border compared to the medio-lateral width of the NST. The anterior-posterior location of sites were expressed as a
75 proportion of the rostral NST, that Is, the distance from the rostral pole of the
NST to where the nucleus is adjacent to the IV ventricle. Proportions of medio-
lateral and anterior-posterior locations were used to plot multi-unit sites on one
horizontal schematic of the NST and single-unit sites on another for both sides
(modified from Travers and Norgren, 1995). The distances of the lesions from the dorsal border of the NST were also determined to confirm that multi- and single-unit sites were within the NST.
Analysis o f orotopic representation
Both multi- and single-unit sites were used to determine whether a shift occurred in orotopic organization. A specific question of interest was whether glossopharyngeal-mediated responses were located further rostrally. Multi-unit responses were the most useful in this regard because more tracks were made to systematically map the entire extent of the gustatory NST. The location of single cell responses in the NST were useful in corroborating multi-unit sites because, although fewer tracks were made, the neurophysiologic data was quantitatively analyzed. Multi-unit sites were classified based on anterior and/or posterior oral cavity taste and tactile responsiveness, whereas single cells were plotted based only on taste responsiveness because tactile responses were not quantified. Single cells were classified as responding best to anterior or posterior oral cavity stimulation using the following criterion; one response had to be 50% greater than another or they were considered equal. The rostral NST was divided into 10 equal anterior-posterior divisions and the caudal 5 segments
76 were collapsed Into one because this area was Infrequently sampled. The
number of sites (multi- and single-unit sites compared separately) which
responded to anterior and/or posterior oral cavity taste and tactile stimulation
was compared for each division. The rostral extent of glossopharyngeal-
mediated taste responses was compared for each side.
Quantification of neural activity
Recorded single-unit activity was analyzed off-line. Single cells were
differentiated with a window discriminator using consistency of amplitude and
waveform as criteria. Slngle-unlt activity was quantified by converting action
potentials to digital pulses and accumulating these In 500 ms bins In perlstlmulus time histograms using Mil hardware and software (Modular Instruments). Net- evoked activity was quantified by using a standard response measure, defined as the number of spikes over a 10 sec period during taste stimulation minus the number of spikes that occurred during the preceding water stimulation. The criteria for a suprathreshoid taste response were defined as a minimum 1 spIke/sec Increase In activity, which also had to be greater than 2.5 times the standard deviation of the spontaneous rate (Travers and Norgren, 1995; Travers and Norgren, 1991; Travers and Smith, 1984; Travers, et al., 1986; Dinkins and
Travers, In press).
77 Statistical analyses
Comparisons were made between CT intact and cut sides. These determined whether the sides were different, but did not provide much insight regarding whether differences were due to central reorganization (e.g., unmasking) since differences would be expected based solely on removing CT input. To address this, we performed the same analyses, but instead of using all of the cells, we compared whole mouth responses after removing cells responsive to anterior tongue stimulation on the intact side or compared responses to specific stimulation of non-anterior tongue receptive fields, that is, the nasoincisor duct, foliate or soft palate. This allowed us to directly compare responses from a peripheral source other than the CT. The disadvantage was the low number of cells and subsequent decrease in power for each group.
Chi-square was used to determine the difference in number of anterior oral cavity multi- and single-unit responses between sides (significance set at
P<0.05). Independent f-tests were used to determine differences in mean anterior-posterior and medio-lateral locations of anterior or posterior oral cavity multi- and single-unit responses within and between sides (significance set at
P<0.05). In order to determine whether neural responsiveness from single cells changed in the NST following CT transection, comparisons between sides were made using several analyses. Initially, mean spontaneous rates and responses to taste mix and individual tastants for whole mouth and individual receptive fields for all cells isolated were compared between sides using two-way ANOVAs
78 followed by Independent f-tests. Independent f-tests were performed to
determine significant differences between CT-intact and cut groups.
Significance levels were set at P<0.05. Additionally, Pearson correlation
coefficients were calculated to determine similarities between responses evoked
by pairs of different tastants. If the similarity between sodium and ammonium
chloride increased between sides, this would be consistent with behavioral
differences (Sollars and Bemstein, 1996).
Neuron types
Finally, hierarchical cluster analysis (Pearson correlation coefficients and average-linkage method) was performed to determine whether there was any evidence for altered neuron types in the NST between CT-intact and cut sides.
Cluster analysis has been used by numerous investigators to classify chemosensitive response profiles in the periphery, NST and parabrachial nucleus of rodents (e.g., Frank, et al., 1983; Frank, 1991; Smith, et al., 1983;
Halsell and Travers, 1997). A change in response profiles could account for the increased preference for ammonium chloride if changes in types were found.
For example, if cells that normally responded best to sodium chloride were no longer found on the denervated side, but replaced by cells which responded best to ammonium chloride.
79 Taste pores and fungiform papillae
The number of taste pores and fungiform papillae on the anterior tongue
was determined for CT-intact and cut sides using a similar procedure described
by Spector and Grill (1992). The anterior 5mm of each tongue was stained with
0.5% methylene blue and observed under light microscopy. An observer blinded
to which side was denervated counted the number of taste pores and fungiform
papillae per side. The number of pores and papillae were averaged per side and
compared by paired f-tests (significance set at P<0.05).
Results
Orotopic representation of multi-unit responses
Based on multi-unit activity in five rats, the orotopic representation of taste
and tactile multi-unit responses did not change after long-term CT transection. A
total of 139 tracks made in the vicinity of the rostral NST were classified based
upon taste and tactile responsiveness to anterior (AO, includes the anterior tongue and nasoincisor duct) and posterior oral cavity (PC, includes the foliate
papillae and soft palate) stimulation. Of these, 109 tracks passed through the
NST as determined by the location of the electrolytic lesions (Figure 3.1). Since some recording sites responded to more than one modality, a summary of 153
responses segregated by modality appears in Figure 3.2. A total of 84 responses were identified on the CT-intact side and 69 on the cut side. Anterior oral cavity taste and tactile responsive sites were found rostral and lateral to
80 posterior oral cavity taste and tactile responsive sites on both sides (f-tests,
P<0.05). Mean anterior-posterior and medio-lateral locations for anterior oral cavity taste-responsive sites were similar between sides (f-tests, P>0.05). Also, mean locations of posterior oral cavity taste-responsive sites were similar between sides (f-tests, P>0.05). Anterior and posterior oral cavity tactile- responsive sites compared separately were also found at similar locations on the two sides (f-tests, P>0.05). Even though the overall orotopic representation of taste and tactile responses appears similar, differences in the number of anterior oral cavity taste and jaw stretch responses were found after denervation.
Of anterior oral cavity taste responses, only nasoincisor duct but not anterior tongue taste responses were found on the CT-cut side. Thus there were fewer total anterior oral cavity taste (anterior tongue plus nasoincisor duct) responses on the denen/ated side in the rostral NST (7/69 v. 24/84, Chl- square=7.96, df=1, P<0.01 ). However, regardless of side almost all of the anterior oral cavity taste responses were found in the rostral 40% of the NST (29 of 31 combining the two sides). In addition to taste responsiveness, tactile responses were also examined. Similar to taste responses, all of the anterior oral cavity tactile responses were found in the rostral 40% of the nucleus.
In contrast to the rostral representation of anterior oral cavity taste responses, sites that responded to posterior oral cavity taste stimulation were entirely restricted to the caudal 70% of the rostral NST on either side. There were similar numbers of posterior oral cavity taste responses on the two sides (8-
81 cut side and 9-Intact side). The rostral extent of posterior oral cavity tactile
responses matched taste responses on the CT-cut side but actually extended
further rostrally (rostral 20%) on the intact side. Thus, contrary to our
hypothesis, neither posterior oral cavity taste or tactile responses were
represented further rostrally in the NST. As an example, Figure 3.3 represents a
case where tracks were made at the same anterior-posterior level of the NST.
Although a response to posterior oral cavity taste stimulation was found on the
CT intact side, such a response was not found on the cut side this far rostrally.
However, caudal to this level, posterior oral cavity taste responses were found on
both sides.
Instead of a rostral migration of posterior oral cavity taste responses, the
rostral pole of the NST (rostral 40%) on the cut side was less responsive to oral taste and tactile stimulation (Chi-square=5.88, d ^ 1 , P<0.05). At locations homologous to those yielding robust anterior tongue taste responses on the intact side, a high proportion of tracks characterized only by responses to depressing the mandible were noted on the cut side. On some of these tracks a clear unresponsive region was noted just dorsal to the jaw-stretch response.
There were six such “unresponsive” tracks on the cut side and these appear as large asterisks in Figure 3.1. After passing through the vestibular nucleus, identified by high amplitude spontaneous activity, an area 50-100 pm in depth that was unresponsive to taste or tactile stimulation was encountered. Further ventrally, jaw stretch responses increased in magnitude and these were assumed to be located in the reticular formation ventral to NST. For example, in
82 one animal, an unresponsive area measured 100pm in depth
eiectrophysiologically. This area was later confirmed histologically to be located
within NST, because the lesion was 134pm ventral to the dorsal border of the
nucleus. Based on responses at similar locations on the intact side and
experiments in non-denervated animals (Travers and Norgren, 1995; Dinkins
and Travers, in press), one would have expected to find anterior oral cavity taste
responses in this unresponsive region. On other tracks where the only sensory
response was elicited by jaw stretch stimulation, there appeared to be an immediate transition between the vestibular nucleus and proprioceptive responses; i.e., there was no intervening unresponsive region.
Like unresponsive tracks, these “jaw stretch-only" tracks were more frequently encountered near the rostral pole of NST on the cut side. Because these data were derived from multi-unit activity, we cannot be certain whether these responses arose from NST neurons or cells ventral to the NST in the reticular formation. However, based on single neuron responses, jaw stretch responses were always found ventral to or in the ventral subdivision of the NST, on both the CT-intact and cut sides. Thus, although jaw stretch responses were noted more frequently on the cut side, there is no evidence that their position changed as a result of CT transection.
General description and orotopic organization of single cells
Taste responses were obtained from single cells in the NST to further explore whether CT denervation led to changes in orotopy and neural
83 responsiveness. Data were collected from 94 taste-responsive cells; fifty-one
were recorded on the CT-intact and 43 on the cut side. Histologic reconstruction
of 78 cells confirmed their location In NST. SIxty-sIx were fully characterized for
receptive field responsiveness, whereas the remaining twelve were not, but they
were plotted and marked with asterisks. Twelve other cells were also within the
NST, but were only responsive to whole mouth stimulation and not plotted. The
location of the remaining four cells were not histologically confirmed to be within the NST because they were not marked with electrolytic lesions.
In general, similar numbers of taste-responsive cells per side were found
(2.00/Intact side and 1.83/cut side per animal; f-test, P>0.05). Also, a similar number of taste-responsive cells were Isolated per taste-responsive track (0.39 cells/track Intact side versus 0.38 cells/track cut side; f-test, P>0.05). However, a greater number of total tracks were made on the CT-cut side to find the sam e number of taste-responsive cells (x=7.5 tracks/animal on the Intact side versus x=10.5 tracks/animal on the cut side; f-test, P=0.001).
Since the number of cells is approximately half the number of tracks used for multi-unit mapping, the orotopic organization Is not as clear. However, In general the slngle-unlt results agree with the multi-unit results. That is, there was no change in the well-described orotopic representation of taste-responsive cells in the NST (Travers and Norgren, 1995). Figure 3.4 demonstrates the location of single cells identified by anterior and/or posterior oral cavity taste responsiveness for each side. The quantitative analysis of the histological data only included
84 cells fully characterized by receptive field input (n=66). In addition, four posterior oral cavity responsive cells that were found in the caudal 25% of the rostral NST on the intact side were removed from this comparison since comparable tracks were not made on the cut side (see Figure 3.4). On the intact side, cells that responded best to anterior oral cavity stimulation were rostral and lateral to those that responded best to posterior oral cavity stimulation (f-tests, P<0.01 for both).
On the cut side, cells that responded best to anterior oral cavity stimulation were found rostral to cells that responded best to posterior oral cavity stimulation, although they were at similar medio-lateral locations (f-tests, P<0.05, P>0.05, respectively). Cells that responded best to anterior oral cavity stimulation were found at similar anterior-posterior and medio-lateral locations between sides ( t- tests, P>0.05 for both). However, although cells that responded best to posterior oral cavity stimulation were found at similar anterior-posterior locations they were at different medio-lateral locations between sides (t-tests, P>0.05, P<0.05, respectively). Figure 3.5 summarizes the number of cells which responded to anterior and/or posterior oral cavity taste stimulation in each of the rostral five divisions plus the caudal half of the rostral NST. In agreement with multi-unit data, posterior oral cavity taste responsive cells did not extend further rostral. In fact, on the intact side, a cell was found in the rostral 10% of the NST that responded to both anterior and posterior oral cavity stimulation.
85 Receptive field organization
Responses from multi- and single-unit sites revealed a lack of orotopic
reorganization in the NST. However, reorganization of afferent input could also
be evident by changes in the proportion of cells that responded to a particular
receptive field or a change in the amount of convergence. To address these
possibilities, we examined the responses to stimulation of a variety of individual
receptive fields within the oral cavity and found that there were differences
between sides (summarized in Figure 3.6). Of the 77 cells in which all of the
individual receptive fields were tested, nineteen of 42 (45%) on the intact side responded only to stimulation of the taste buds on the anterior tongue. In contrast, the most common type of cell on the cut side responded to stimulation of the nasoincisor duct, eleven of 35 (31%) responded in this way. Not unexpectedly, twenty-eight cells on the intact side responded to anterior tongue stimulation whereas only one cell whose response barely met criteria did on the cut side (Chi-square=33.11, df=1, P<0.001). Contrary to our hypothesis, an increase in the incidence of glossopharyngeal-mediated taste-responsive cells was not observed. In fact, more foliate papillae-responsive cells were found on the intact side although this difference was not significant (11 versus 6, respectively: Chi-square=0.91, d^1, P>0.05). However, an increase in the incidence of palatal responses was found on the cut side. Twenty nasoincisor duct-responsive cells were recorded on the cut side and only 11 such cells were found on the intact side (Chi-square=6.42, dh=1,P<0.05). In addition, 3 cells on the intact side responded to soft palate stimulation whereas 13 cells on the cut
86 side did (Chi-square=10.44, df=1, P<0.01). Thus, in our sample of cells on the denervated side, the number of palatal, but not foliate papillae responses increased to compensate for the missing anterior tongue responses. Since cells on the CT-cut side could not respond to anterior tongue stimulation, we removed cells only responsive to anterior tongue stimulation to specifically compare non- anterior tongue responses. Unexpectedly, there were still significantly more soft palate-responsive cells on the cut side (Chi-square=4.04, df=1, P<0.05). On the other hand, the number of foliate papillae-responsive cells significantly decreased on the cut side (Chi-square=6.31, d^^=^, P<0.05). The difference in nasoincisor duct-responsive cells was no longer significant (Chi-square=0.14, dh=1, P>0.05). Therefore, it appears that only soft palate-responsive cells increased in proportion to other receptive field-responsive cells after CT denervation.
Although the proportion of cells responding to particular receptive fields differed between sides, the proportion of cells which responded to multiple receptive fields did not. Twenty-seven of 42 (64%) cells on the intact side and
19 of 35 (54%) on the cut side responded to only one receptive field. Twelve of
42 (29%) cells on the intact side and 11 of 35 (31%) on the cut side responded to multiple receptive fields. Also, if the cells responsive to anterior tongue stimulation are removed from the analysis, 3 of 14 (21%) cells on the intact side and 10 of 34 (29%) cells on the cut side responded to more than one receptive field (Chi-square=0.32, d ^ 1 , P>0.05). Therefore, the overall degree of
87 convergence remained similar between sides which suggests a lack of
reorganization of afferent input in single cells in the NST after CT denervation.
Altered neural responses in the NST
The preceding analyses investigated the possibility of central
reorganization by categorizing neurons according to suprathreshold responses to
various receptive fields without regard to response magnitude. The proceeding
analyses explore the possibility of more subtle functional changes in the NST
with respect to spontaneous or evoked activity by analyzing changes in firing
rates.
a. Spontaneous activity
Since previous work has shown that the spontaneous rate of cells that
receive CT input is greater than cells that do not (Dinkins and Travers, in press),
a change in the characteristic spontaneous activity of a particular cell type could
suggest altered functional input. The overall mean spontaneous rate of cells on
the CT-intact versus the cut side was lower but not statistically significant (Figure
3.7)(x=2.08± 0.39 (SEM) spikes/s intact versus x=2.95± 0.67 spikes/s cut side; t-
test, P=0.26). However, when anterior tongue-responsive cells were removed so that we could directly compare cells without CT input, the mean spontaneous
rate was significantly higher for cells on the CT-cut side (x=2.97± 0.69 spikes/s
cut side versus x=0.87± 0.28 spikes/s intact side; f-test, P<0.01). This effect was
88 further explored by examining subsets of neurons with optimal inputs from
various receptive fields and found to be restricted to cells that responded best to
nasoincisor duct stimulation. The mean spontaneous rate of cells that
responded best to nasoincisor duct stimulation on the CT-intact side exhibited a
much lower mean spontaneous rate than on the CT-cut side (x=1,68± 0.61
spikes/s versus x=4.75± 1.06 spikes/s, respectively; Mest, P<0.05). There was
no apparent change in mean spontaneous activity for cells which responded best to foliate or soft palate stimulation (f-tests, P>0.05 for both).
b. Taste responses to mixture stimulation
In addition to spontaneous activity, changes in evoked responses to taste mixture stimulation could also suggest altered functional input. Mean taste responses tended to increase on the CT-cut side for cells responsive to receptive fields other than the anterior tongue. Across all cells there was a significant interaction between receptive field and side (ANOVA, P<0.001)(see Figure 3.8).
Specifically, mean responses to anterior tongue stimulation were robust on the
CT-intact side but virtually obliterated on the cut side reflecting the efficacy of our denervation procedure. In contrast, mean responses to stimulation of the whole mouth and foliate papillae were not different between sides, whereas responses to stimulation of the nasoincisor duct and soft palate were greater on the CT-cut side. The responses to stimulation of the individual receptive fields with taste mixture for each cell on the CT-intact and cut sides are depicted in Figure 3.9.
89 Differences in mean responses to taste mixture stimulation of individual
receptive fields on the two sides may arise simply from the loss of CT input or from compensatory mechanisms that are a result of CT denervation (e.g.,
unmasking). Many neurons on the intact side responded only to anterior tongue stimulation. On the CT-cut side, by default, neurons with other receptive fields made up a larger proportion of that population. Specifically, there were increased proportions of neurons responsive to soft palate and nasoincisor duct stimulation (see Receptive field organization). Thus, the increased mean responses to palatal stimulation on the denervated side can be partially explained by the increased proportions of cells responsive to these receptor subpopulations. However, if comparisons are restricted to cells exhibiting suprathreshold responses to a given receptive field, observed differences are more likely due to compensatory changes within the NST. Our analyses for these comparisons have limited statistical power because of the more restricted sample sizes, but they show a consistent trend. Figure 3.10 depicts mean, median, and maximal responses elicited by stimulating the nasoincisor duct, foliate papillae and soft palate, restricted to cells that received suprathreshold input from that particular receptive field. Although only the foliate comparison reached statistical significance, in each case, the mean response on the CT-cut side is greater. In addition, the median and maximal responses were greater on the cut side for all three receptive fields. Thus, there appears to be a tendency for the magnitude of palatal and foliate papillae responses to increase after denervation, independent of the differences in the proportion of cells that receive
90 various receptive field Input on the two sides. These results suggest that NST
cells compensate for lost CT input (also see Figure 3.9).
c. Taste responses to individual tastant stimulation
Quality is an important aspect of taste, therefore we investigated the
possibility of changes in neural responses to individual tastant stimulation. With the exception of ammonium chloride, the testants that we used are considered representative of the four “pure” taste qualities in humans (Miller and Bartoshuk,
1991). Whole mouth-elicited sucrose responses tended to increase whereas sodium and ammonium chloride responses decreased on the CT-cut side
(significant interaction between testants and side, ANOVA, P<0.05; f-tests,
P=0.08 for sucrose, P<0.01 for both sodium and ammonium chloride). Because chemosensitivity is related to receptive field location, differences between the sides could be affected by the different proportions of cells responding to the various receptive fields. Since few individual receptive fields were tested with individual tastants, analyses similar to those performed above for taste mixture were not possible. Another approach to directly compare non-anterior tongue responses on the two sides was to simply remove the anterior tongue-responsive neurons from the data set and analyze whole mouth responses. This approach was less comprehensive since neurons receiving convergent input from the anterior tongue and other receptive fields could not be used but it resulted in a reasonable sample size. Using this strategy, there was no significant interaction between individual tastant responses and side suggesting that the relative
91 Chemical responsiveness of residual taste responses was not altered by
denervation (ANOVA, P>0.05). Figure 3.11 depicts mean responses to whole
mouth stimulation with Individual tastants of cells on the CT-lntact side, cells on
the Intact side that did not respond to anterior tongue stimulation, and cells on
the cut side.
In addition to comparing response magnitudes between sides, correlation
coefficients were calculated between responses evoked by pairs of stimuli.
Correlation coefficients have been proposed to reflect the perceived similarity of
two stimuli (Doetsch and Erickson, 1970; Smith, et al., 1983). Because many CT
fibers respond selectively to sodium (Frank, et al., 1983), we predicted an
Increase In the similarity of responses evoked by sodium and ammonium
chloride on the cut side. However, In direct opposition to what we predicted, the
responses to sodium and ammonium chloride were correlated less to each other on the CT-cut side (Pearson correlation coefficients, r=.853 Intact side, r=.560 cut side). That Is, cells on the cut side respond well to sodium or ammonium chloride but not well to both, whereas on the Intact side, there Is a larger number of cells that respond similarly to these tastants. Importantly, this comparison remains the same even when the anterior tongue responsive cells are removed
(r=.857 Intact side, r=.553 cut side).
92 Neuron types
Finally, we examined the chemosensitive response profiles of individual
neurons for evidence that neuron types changed after CT denervation. A
difference in these profiles could suggest that afferent inputs were reorganized
after denervation. Using hierarchical cluster analysis, we found four groups on
the intact side that responded best to sucrose, ammonium chloride, hydrochloric
acid or quinine; whereas on the cut side we found five groups that responded
best to either sucrose, sodium chloride, ammonium chloride, hydrochloric acid or
quinine. Figure 3.12 depicts the response profiles of each of these groups by
side. Note the lack of responsivity of cells that responded best to hydrochloric
acid and quinine on both sides and to sodium chloride on the cut side. The additional group on the cut side contains four cells that respond specifically to sodium chloride. However, their mean response to sodium chloride is very low, therefore they are not typical of NST cells that respond specifically to sodium.
Cells that responded best to ammonium chloride on average, had quite variable responses to sodium chloride individually. That is, some cells responded much better to sodium whereas others responded much better to ammonium, while many others fell somewhere between these two extremes. Although it was uncommon, there were two cells on the denervated side that responded only to ammonium chloride. For example, the cell depicted in Figure 3.13 responds well to ammonium chloride, but not to any other tastant including sodium chloride.
However, this was not unique to the cut side, since a cell on the intact side also
9 3 responded only to ammonium chloride. Therefore, we did not find evidence for
novel neuron types on the cut side.
Taste pores and fungiform papillae
The number of taste pores and fungiform papillae decreased significantly
on the CT-cut side of the tongue (f-tests, P<0.001 for both)(Figure 3.14). During
the procedure a section of the nerve was removed which appears to be sufficient
to prevent nerve regeneration. These results confirm the efficacy of the CT
transection procedure and provide evidence that this procedure results in long
term denervation of the tongue.
Discussion
Orotopic representation
This study was designed to determine whether neonatal CT nerve transection leads to changes in the orotopic representation of taste and other alterations of single unit gustatory responses in NST. The orderly organization of anterior and posterior oral cavity taste responses demonstrated by Travers and Norgren (1995) appears unaltered following unilateral CT nerve transection in young rats. Specifically, posterior oral cavity taste-responsive neurons were not found further rostrally on the CT-cut side compared to the intact side.
Instead, posterior oral cavity responses were found in symmetrical anterior- posterior locations based on multi- and single-unit sites. Therefore, large-scale
94 functional changes that would be expected to be caused by anatomical changes
(e.g., axonal sprouting or increased dendritic arborization) do not appear to occur
in the NST following CT transection in 10 day old rat pups.
The rostral pole of the NST (rostral 20%) appears to be less responsive to
taste stimulation on the CT-cut side. In the multi-unit experiments 2 of 16 tracks
on the CT-cut side were taste-responsive, compared to 12 of 17 on the intact
side. Likewise, only 5 taste-responsive single-units were isolated on the cut side versus 10 on the intact side. Despite the lack of taste responsivity, commonly we were able to detect weak multi-unit responses to jaw stretch stimulation at locations we presume were in NST, based on their depth. Jaw stretch responses on the intact side were detected less frequently probably due to the increased spontaneous activity derived from the CT nerve. By cutting the CT, the decreased spontaneous activity allowed for their detection. Single cells responsive to jaw stretch stimulation were not found in the taste-responsive area on the cut side of the NST based on the location of several “jaw-stretch” cells
(n=6) marked with lesions. Therefore, it seems unlikely that CT transection resulted in the appearance of novel jaw stretch responses.
In contrast to the results of this study In the gustatory system, changes in topographic representation after peripheral denervation have been well documented In other sensory systems, such as the somatosensory system.
Median nerve transection results In altered responses In the primary somatosensory cortex In monkeys. After denervation neurons that previously responded to the denervated receptive field respond to an adjacent receptive
95 field innervated by an intact nerve (e.g., the ulnar nerve) (Merzenich, et al.,
1983a; reviewed in Merzenich, et al., 1988; Buonomano and Merzenich, 1998).
Apparently these changes are at least partially due to “unmasking” since they can be seen immediately after denervation (Merzenich, et al., 1983b). Studies over the last decade have shown that changes in the somatotopic representation that occur in the primary somatosensory cortex of monkeys after peripheral deafferentation (e.g., nerve section and limb amputation) are reflected in changes that have occurred at lower levels of the neuraxis, that is in the spinal cord, medulla and thalamus (Florence and Kaas, 1995; Garraghty and Kaas,
1991).
Increased convergence would also be indicative of functional changes in the NST. However, the degree of convergent input from intact nerves in the NST did not change after CT transection. A considerable amount of convergence normally occurs between individual receptive fields in the oral cavity, especially between anterior tongue and nasoincisor duct receptive fields (Travers, et al.,
1986; Travers and Norgren, 1991; Travers and Norgren, 1995; Dinkins and
Travers, in press). Since convergence between the anterior tongue and other receptive fields could not occur on the cut side, it is somewhat surprising that the degree of convergence remains the same. This finding is probably due to the greater number of palatal-responsive cells on the cut side, because these cells displayed frequent convergence between nasoincisor duct and soft palate receptive fields. Therefore, cells that responded to anterior tongue and other receptive fields on the intact side were replaced with cells that responded to both
96 palatal receptive fields. These results reflect the increased incidence of soft
palate-responsive cells on the cut side. This is significant because there were
only 3 soft palate-responsive cells on the intact side and all of them received
convergent input. Whereas on the cut side, 9 of 13 soft palate-responsive cells
received convergent input. Although we did not find an increase in convergence
in the NST after CT denervation, an increase in the number of convergent cells was found in the spinal trigeminal nucleus after tooth pulpotomy in cats (pars oralis) (Hu, et al., 1986).
The taste-responsive area of the NST is very difficult to locate in CT- denervated animals. The advantage of testing unilateral CT-cut rats was the ability to locate taste responses on the intact side first and then use the corresponding coordinates to locate taste responses on the cut side.
Conceivably, unilateral transactions could have affected taste responses on the contralateral side (CT-intact side). However, this seems unlikely since the CT nerve projects exclusively to the ipsilateral NST in rodents (Hamilton and
Norgren, 1984; Whitehead and Frank, 1983). One could argue that although the ascending input to the NST is Ipsilateral, descending corticofugal projections are not (Di Lorenzo and Monroe, 1995). It is possible that altered corticofugal input affected contralateral NST neural activity. If this occurred, one would expect to see differences in spontaneous and evoked activity on the intact side. However, significant differences were not found when the spontaneous rate and whole mouth-elicited taste mixture response were compared to single cell responses in rats without CT transection under similar recording conditions (t-tests, P>0.05 for
97 both) (Dinkins and Travers, in press). Therefore, we do not think that unilateral
CT transection had a significant impact on contralateral NST responses.
Neural responses
a. Spontaneous activity
Although a change in orotopic representation of anterior and posterior oral
cavity taste-responsive neurons was not found, the spontaneous rate of NST
cells were different between sides. Interestingly, we found an increase in mean
spontaneous rate of nasoincisor duct-responsive cells after CT transection in this
study, however in a previous study (Dinkins and Travers, in press), we found a
decrease in mean spontaneous rate for nasoincisor duct -responsive cells during
CT anesthesia (Figure 3.15). Because there was no difference in the
spontaneous rate of cells on the intact side of this study and those from the
anesthesia study (before anesthesia), we can rule out potential variables
between studies (x=2.08± 0.39 spikes/s intact side versus x=1.92± 0.54 spikes/s
CT anesthesia study; f-test, P=0.81). Although it is tempting to speculate that
chronic denervation led to reactive changes in the central nervous system to
account for the difference in spontaneous rates of acute and chronic CT- denervated animals, this difference could also be due to the denervation methods that we used (i.e., anesthesia versus transection). On one hand, CT anesthesia may result in decreased spontaneous rate (short-term change) simply by decreasing peripheral input (mono- or multi-synaptic). On the other hand, CT transection (long-term change) may trigger neuroma formation in the
98 proximal nerve. In the somatosensory system, damaged peripheral nerves can form neuromas (Wall and Gutnick, 1974). The spontaneous activity in these nerves has been implicated as a cause of phantom limb pain (Nystrom and
Hagbarth, 1981). If this also occurs in the CT, then increased spontaneous activity in the nerve could be reflected as an increase in spontaneous rate of
NST cells with CT input. Unfortunately, this possibility was not considered at the time of the experiments, hence it cannot be ruled out.
b. Taste responses
Changes in taste response magnitudes elicited by stimulation of receptive fields innervated by intact nerves could suggest increased synaptic efficacy of these nerves in the NST. In contrast to the CT anesthesia study (Dinkins and
Travers, in press) in which increased mean taste response magnitudes were not observed during CT anesthesia, the present study found several increases on the denervated side. Across all cells, taste mixture responses to nasoincisor duct and soft palate stimulation increased on the denervated side. Also, the proportion of cells that responded to the soft palate increased on the cut side.
When only cells responsive to particular taste bud subpopulations were compared, mean responses to nasoincisor duct and soft palate stimulation were not significantly different, although there was still a trend for increased responses on the denervated side. In addition, cells responsive to the foliate papillae exhibited significantly increased taste mixture responses when the foliate papillae were stimulated. These results suggest that g lossophary ngeal-med iated
99 and possibly greater superficial petrosal-mediated responses increased in
magnitude potentially from an increase in synaptic efficacy after CT denervation.
However, a greater number of cells responsive to non-anterior tongue receptive
fields on the intact side is needed to definitively support this notion.
Sodium and ammonium chloride responses
The increased behavioral preference for ammonium chloride after CT
transection in 10-day old rats was studied by Sollars and Bernstein (1996), yet the physiologic changes that account for the behavior remain unknown. We expected an increase in the magnitude of ammonium chloride responsiveness in the NST. However, the mean response to whole mouth ammonium chloride stimulation was actually reduced on the denervated side. Because cells which responded to anterior tongue stimulation in the rostral third of the nucleus were not replaced by cells that responded to the glossopharyngeal nerve, the simple hypothesis of a switch from sodium chloride best cells to ammonium chloride best cells was not proven. In fact, there was no hint of such a change.
Additionally, increased convergent input from intact nerves that could result in different neuron types that would plausibly explain the change was not found.
For example. If more cells that responded well to sucrose and ammonium chloride (two stimuli that don’t normally co-activate cells) were found, then their activity could cause a behavioral change. Yet, there was no significant change in neuron types either (see Figure 3.9). It is important to note that in addition to a decrease in ammonium chloride responsiveness, an even larger decrease in
100 sodium chloride responsiveness was found on the cut side for cells categorized
as ammonium chloride best (see Figure 3.9). These results lend evidence for the
importance of the CT nerve in sodium responsivity in the CNS as suggested by
several other authors (Specter and Grill, 1992; Hill, et al., 1990; Markison, et al.,
1995; St John, et al., 1997a).
Although we did not find an increase in ammonium chloride responses,
we did note a change in response to this stimulus that may have behavioral
consequences. The Pearson correlation coefficient between sodium and
ammonium chloride decreased on the CT-cut side. This is just the opposite to
what we originally predicted. Although this difference may result in behavioral
changes, a decrease in response similarity of sodium and ammonium chloride
does not appear to explain the results of Sollars and Bernstein (1996) In a
straightforward way. Nevertheless, the decrease in response similarity on the
cut side may allow the animal to better discriminate between sodium and
ammonium chloride. The difference in correlation coefficients remained the
same when comparing cells on the intact side without CT input to the cut side.
This suggests that the change in across-neuron correlation coefficients between
sodium and ammonium chloride responsiveness of NST cells could be due to
plasticity rather than simply reflecting a difference in anterior tongue/non-anterior tongue responsive cells.
The fact that neonatally CT-transected rats drank more ammonium chloride than water is still puzzling. Similar to a phantom taste where there is no apparent physical stimulus yet the subject experiences a taste, maybe water
101 elicits a taste phantom. Water stimulation commonly reduces the spontaneous rate of cells responsive to anterior tongue stimulation (unpublished observations). Yet the same effect is not as obvious for cells responsive to nasoincisor duct stimulation. However, in this study, water commonly elicited a decrease in the spontaneous rate of nasoincisor duct-responsive cells on the denervated side, although this effect was not statistically significant. Possibly a greater number of cells would prove this to be true. If the decrease in spontaneous rate to water stimulation Is perceived as a taste phantom (e.g., bitter), the CT-denervated animals may prefer ammonium chloride over water because it is less aversive. Recording taste-responsive cells in the NST of adult
CT-transected rats may illuminate the reason why neonatal CT-cut rats prefer ammonium over sodium chloride.
1 0 2 Figure 3.1
Horizontal schematic of the NST depicting the location of 109 multi-unit taste and
tactile-responsive sites on intact and cut sides.
Sites were classified as anterior (anterior tongue and/or nasoincisor duct
receptive fields) and/or posterior oral cavity (soft palate and/or foliate receptive
fields) taste and tactile-responsive based on qualitative assessment of receptive field responses. The following abbreviations are used: AO-anterior oral cavity,
PO-posterior oral cavity, AO/PO-both anterior and posterior oral cavity, G-taste
responsive, M-tactile responsive, JS-jaw-stretch responsive, NR/JS- unresponsive dorsal and jaw stretch responsive ventral.
103 CT intact side CT cut side
4 AO/PO.G O
□ PO:M
0 AO/PO;M
* NR/JS
B- 0.2 mm B- lateral ^ ■ k lateral
0
B
Figure 3.1 Figure 3.2
The number of sites that responded to AO, PC or AO/PO taste or tactile stimulation on each side.
List of abbreviations: AO=anterior oral cavity, PO=posterior oral cavity,
AO/PO=anterior and posterior oral cavity, G=taste responsive, M=tactile responsive, JS=jaw stretch responsive, AP%=anterior posterior percentile. The anterior posterior percentile is based upon dividing the rostral NST into 10 equal segments starting with the rostral pole and ending caudally at the IV ventricle.
The intact side was the unoperated side and the cut side was the chorda tympani transected side of the animal.
105 NEOPLAS RECORDING SITES- CT INTACT SIDE AP% AO:G AO:M PO:G PO:M AO/PO:G AO/PO:M JS TOTAL 0-10 7 5 3 16 11-20 5 2 2 1 0 2 12 21-30 5 1 1 1 8 31-40 5 5 2 4 7 3 26 41-50 2 2 4 1 2 11 51-100 5 6 1 12 TOTAL 24 13 9 16 1 9 12 84 8 NEC»PLAS RECORDING SITES- CT CUT SIDE AP% AO:G AO:M PO:G PO:M AO/PO:G AO/PO:M JS TOTAL 0-10 2 3 7 12 11-20 2 4 6 21-30 3 2 3 3 11 31-40 2 2 3 4 1 3 2 17 41-50 1 4 2 1 8 51-100 4 4 3 3 1 15 TOTAL 7 9 8 12 4 11 18 69 Figure 3.2 Figure 3.3
Photomicrographs of symmetrically located recording sites made on both sides of the NST.
Responses were recorded from sites that were 200-300 pm caudal to the rostral pole on both sides of the NST. Taste responses were found on the intact side (top) after anterior and posterior tongue stimulation (lesion made 300 pm ventral to site). Taste responses were not found on the cut side (bottom) at this anterior-posterior level (lesion made 200 pm ventral to site). Thus, in this case, posterior tongue taste responses were actually found further rostral on the CT- intact side. Scale bar=200 pm
107 Figure 3.3
108 Figure 3.4
Horizontal schematic of the NST depicting the location of 78 single cells responsive to taste stimulation on intact and cut sides.
Cells were classified as anterior (anterior tongue and/or nasoincisor duct receptive fields) and/or posterior oral cavity (soft palate and/or foliate receptive fields) taste-responsive (AO;G and/or PO;G, respectively) based on quantitative assessment of receptive field responses. Cells that responded to anterior and posterior oral cavity taste stimulation (AO/PO:G) were classified as anterior or posterior oral cavity best if one response was 50% greater than the other.
109 CT intact side CT cut side AO:G
PO;G 00 AO/PO:G üf------Cl/. AO/PO:G o / AO BEST
AO/PO:G PC BEST
ALLRFs not tested o
caudal
■Ù 0.2
IV ) < — IV
Figure 3.4 Figure 3.5
The number of cells that responded to AO, PC, AO/PO taste stimulation.
List of abbreviations: AO=anterior oral cavity, PO=posterior oral cavity,
AO/PO=anterior and posterior oral cavity, G=taste responsive, AP%=anterior
posterior percentile.
The anterior posterior percentile is based upon dividing the rostral NST into 10 equal segments starting with the rostral pole and ending caudally at the IV ventricle. The intact side was the unoperated side and the cut side was the chorda tympani transected side of the animal.
111 The number of cells which responded to AO, PC or AO/PO taste stimulation AO:G PO:G AO/PO:G AP% Intact Cut Intact Cut Intact Cut 0-10 3 1 0 0 1 0 11-20 5 2 0 1 1 1 21-30 4 3 2 3 1 1 31-40 12 10 0 3 1 5 41-50 2 2 0 2 3 1 51-100 1 0 3 2 2 0 Total 27 18 5 11 9 8
Figure 3.5
112 Figure 3.6
The number of cells on each side that responded to particular receptive fields.
Each cell is classified based upon a response to a particular receptive field following stimulation with taste mixture. Only cells which were fully characterized by receptive field response are included (n=77). Abbreviations used: AT-anterior tongue, NID-nasoincisor duct, FOL-foliate papillae, SP-soft palate, CV- circumvallate papillae, WM-whole mouth, CT-chorda tympani nerve, RF- receptive field.
113 The number of cells on each side that responded to particular receptive fields RF CT intact CT cut Total AT 19 0 19 AT/NID 3 1 4 AT/FOL 4 0 4 AT/NID/FOL 1 0 1 AT/NID/SP 1 0 1
NID 4 11 15 NID/FOL 1 0 1 NID/SP 1 7 8 NID/FOL/SP 0 1 1
FOL 4 3 7 FOL/SP 1 1 2 FOL/CV 0 1 1
SP 0 4 4
CV 0 1 1
WM 3 5 8 Total 42 35 77 Figure 3.6
114 Figure 3.7
The mean (± SEM) spontaneous rates for all cells and subsets of cells that responded best to a particular receptive field for each side.
The spontaneous rate of nasoincisor duct-responsive (NID) cells was significantly greater on the CT-cut side (Mest, P<0.05). The following abbreviations are used: AT-anterior tongue, FOL-foliate papillae, SP-soft palate.
115 Spontaneous Activity
CT intact aWMMI CT cut
AT NID FOL Best receptive field
Figure 3.7
116 Taste mixture responses of all cells on each side to individual receptive fields Receptive field Intact side Cut side WM-n 51 42 mean 10.7 8.2 median 6.9 4.0 maximum 47.9 30.9 SEM 1.6 1.4 AT -n 45 40 mean 5.9 0.4* median 3.4 0.05 maximum 23.5 4.5 SEM 1.0 0.2 NID -n 44 40 mean 2.2 6.8* median 0.2 2.5 maximum 23.4 39.7 SEM 0.6 1.5 FOL -n 43 38 mean 0.8 1.6 median 0.2 0.1 maximum 6.4 16.4 SEM 0.2 0.6 SP -n 42 34 mean 0.2 1.7* median 0.1 0.4 maximum 4.8 12.7 SEM 0.2 0.6
The following abbreviations are used: WM-
whole mouth, AT-anterior tongue, NID-
nasoincisor duct, FOL-foliate papillae, SP-soft
palate.
Figure 3.8
117 Figure 3.9
Individual responses to individual receptive field stimulation with taste mixture.
Responses due to whole mouth (WM), anterior tongue (AT), nasoincisor ducts (NID), foliate papillae (FOL), and soft palate (SP) stimulation as well as spontaneous rate (SPON) are shown for each cell on the CT intact (left panels) and CT cut sides (right panels). The responses for each cell are stacked vertically. Suprathreshold responses are indicated by filled bars, whereas subthreshold responses are indicated by open bars.
118 CT intact side CT cut side 50 WM 40 •
30 .
20 .
10 .
Immn
10 20 30 40 50 60
20 AT 2 0 . AT
10 10 ■ llllllii lllllllllllllllln.. il
------u u""------■ n, IL.^nn -„■■■■■■■■ ■■ ------ninl
0 10 20 30 40 50 60 20 FOL
10 -
- • .n III.. i ,nl 1 in u
0 10 20 30 40 50 60
10 SP
flnn - In 1 I *u U "
0 10 20 30 40 50 60
10 SPON SPON
ll l l l l L U . . L . 1. 1,1 _l 11 i-H 0 10 20 30 40 50 60 0 10 20 30 40 50 Figure 3.9 Cell
119 Figure 3.10 Individual receptive field responses for cells which responded to that receptive field
Receptive field Intact side Cut side NID -n 11 26 mean 8.2 10.2 median 5.6 7.4 maximum 23.4 39.7 SEM 2.2 2.0 FOL -n 11 6 mean 2.7 8.3* median 2.1 7.4 maximum 6.4 16.4 SEM 0.5 2.6 SP -n 3 13 mean 2.4 4.2 median 2.3 3.0 maximum 2.8 12.7 SEM 0.2 1.2
The following abbreviations are used: NID-
nasoincisor duct, FOL-foliate papillae, SP-soft
palate.
Figure 3.10
120 Figure 3.11
Mean (± SEM) responses to whole mouth stimulation with individual tastants.
The following tastants were used; 0.3M sucrose (8), 0.3M sodium chloride
(N), 0.3M ammonium chloride (NH4), 0.01 M hydrochlorous (H) or 0.003M
quinine hydrochloride (Q). Mean responses to sodium and ammonium chloride were less on the cut side compared to the intact side (f-tests, P<0.05, indicated
by asterisks for CT intact versus cut only). Mean responses for cells responsive to non-anterior tongue receptive fields on the Intact side are also Included.
121 Responses to Individual Tastants
CT intact 14 - CT intact (non AT- responsive cells) CT cut
12 -
8 « 5. CO
S N NH4 HQ Tastant
Figure 3.11
122 Figure 3.12
Response profiles of neuron types found In NST between CT-lntact and cut sides.
Hierarchical cluster analysis of single cell responses was used to compare neuron types in NST between CT-intact and cut sides. Cells were categorized into four groups on the intact side (left panels) and five groups on the cut side
(right panels). The following abbreviations are used: S' for sucrose, NH4' for ammonium chloride, N' for sodium chloride, H" for hydrochloric acid and ‘Q’ for quinine hydrochloride. Cells that responded best to ammonium chloride stimulation comprised a heterogenous group on both sides. Note the significant loss of robust sodium chloride responsiveness on the cut side.
123 CT intact side CT cut side S' BEST (N=8) S' BEST (N=13) 15
10 ■
20 20 •NH4' BEST •NH4’ BEST (N=28) (N=11) 15 15
10 10
O Q) JO 0 lo 2 Q. ■N' BEST (N=4) CO 0 à 10 10 H' BEST (N=3) ■H* BEST (N=2)
L J , mim i
10 •Q’ BEST (N=2) 10 Q' BEST (N=1)
5
0
S N NH4 H Q S N NH4 H Q Tastant Figure 3.12
124 Figure 3.13
Neural trace of single cell on the CT-cut side responsive to ammonium chloride but not to sodium chloride.
The whole mouth was stimulated with water, tastant and water.
125 Water Ammonium chloride Water
4 A A lU 12 14 16 ts 20 22 24 26 zt JO )2 34 16 31 40 42 44 46 4* 50 <2 54 <6 5A W Time (s)
Water Sodium chloride Water
■'1 ;
0 2 4 6 A in 12 U 16 1A 20 22 24 26 2A 10 12 U 36 )A 40 42 «4 46 4t V* 52 vi ^ (*, mi Time (s)
Figure 3.13
126 Figure 3.14
Mean (± SEM) number of taste pores and fungiform papillae.
Number of taste pores and fungiform papillae were determined for each
side of the anterior tongue in 29 rats. Significant differences were determined by paired t-tests (P<0.05) and indicated with asterisks.
127 Taste Pores and Fungiform Papillae (n=29 rats)
60 CT intact 1 CT cut 50 -
40
E 30 3
20 -
10 -
Taste pores Fungiform papillae
Figure 3.14
128 Figure 3.15
Mean (± SEM) spontaneous rates of nasoincisor duct-responsive (NID) cells from the two studies.
In contrast to an increased mean spontaneous rate found during CT anesthesia (Study 1), a decreased mean spontaneous rate was found after CT transection (Study 2)(t-tests, P<0.05, indicated by asterisks).
129 Spontaneous Activity of NID-responsive Cells
5 -
4 - o (D « 3 - f a. CO
1 -
Intact Cut Before During CT transection study CT anesthesia study
Figure 3.15
130 CHAPTER 4
GENERAL DISCUSSION
Morphological (e.g., axonal sprouting) and functional (e.g., increased synaptic efficacy) changes have been found centrally in the somatosensory system after peripheral deafferentation (e.g., Woolf, et al., 1992; Merzenich, et al., 1983a; Calford and Tweedale, 1988; reviewed in Wilson and Kitchener,
1996; Weinberger, 1995; Buonomano and Merzenich, 1998). Unmasking of tactile responses, once thought to be primarily a cortically-mediated phenomenon, has been documented throughout the neuraxis. Novel and/or altered receptive field responses have been found in spinal cord, brainstem and thalamus (e.g., Xu and Wall, 1997; Garraghty and Kaas, 1991; Dostrovsky, et al., 1976). Given the substantial degree of plasticity found throughout the somatosensory system, it was surprising that more dramatic results were not found in the gustatory system after peripheral deafferentation.
Chorda tympani anesthesia
In the first set of experiments (Dinkins and Travers, in press), the possibility of disinhibition in the taste system was studied in an anesthesia
131 model. Using an anesthesia model in the gustatory system has special
advantages, because the anesthetic technique is atraumatic and rapidly
reversible. In the somatosensory system, local anesthetics are injected into the
cutaneous receptive field (e.g., Pettit and Schwark, 1993; Panetsos, et al.,
1995). The latter technique causes inflammation from mechanical trauma
(needle) and pH changes (anesthetic) in the tissue. Inflammation alone could
affect the response properties of central neurons especially in the
somatosensory system where “silent" neurons have been found (Neugebauer
and Schaible, 1990). In contrast, the CT is anesthetized by perfusing the middle
ear with anesthetic (lidocaine) delivered through a polyethylene tube minimizing
inflammation at the delivery site. Additionally, the current model eliminates
tissue inflammation as a possible cause of unmasking of receptive field
responses because the site of nerve block (middle ear) does not involve the
receptive field (oral cavity).
Neural responses to taste stimulation were determined before, during and
after recovery from chorda tympani (CT) anesthesia in the first-order central
taste relay, the nucleus of the solitary tract (NST). Mean taste responses did
not increase during anesthesia. Also, novel receptive field responses were not found. However, responses to taste stimulation at 10% of individual NST sites
(multi- and single-unit) increased during anesthesia, although the average
increase in response magnitude was only 33%. Therefore, disinhibition of taste
responses in the NST occurred in a small proportion of cells to a small degree.
Such a small effect is unlikely to result in behavioral consequences that fully
132 account for taste compensation after CT damage. However, we did note a
marked change in the spontaneous activity of NST neurons that received input
from the anterior tongue and hard palate. Specifically, during anesthesia, taste
responses elicited by palatal (nasoincisor duct) stimulation were about the same
as before anesthesia, but the spontaneous rates decreased significantly. We
proposed that this change resulted in an increase in signal-to-noise ratio of these
cells which may be a mechanism responsible for taste constancy. In fact, the
response to whole mouth stimulation when factoring in the spontaneous activity was the same during anesthesia. Therefore, both disinhibition and an increase in signal-to-noise ratio may be significant mechanisms that are responsible for taste compensation during CT anesthesia. If these mechanisms also occur after
CT denervation, then they may allow a patient suffering from nerve damage to experience “real world” taste as they did before denervation.
Chorda tympani transection
In the second set of experiments, orotopic representation and changes in neural responsiveness were studied in a chronically CT-denervated model. The
CT nerve was transected in 10 day old rats unilaterally. Neural responses to taste stimulation were compared in the NST between CT-intact and cut sides of adult animals. Contrary to our original hypothesis, the orotopic representation of taste responses in the NST did not change after CT transection. Moreover, increased convergent input from intact nerves was not found on the denervated side. These results suggest a lack of reorganization in the NST after
133 denervation. However, for each of 3 non-anterior tongue receptive fields stimulated there was an increase in the mean response elicited by mixture stimulation although only the increase for the foliate papillae reached statistical significance. Also, the mean spontaneous rate of cells that received palatal input was significantly increased on the denervated side. These results suggest possible increases in functional input (e.g., increased synaptic efficacy) of NST cells from intact nerves. Indeed, the increased responses from stimulation of receptive fields other than the anterior tongue appear even more significant when they are compared to the lack of changes in responses found before and during CT anesthesia in the first study (see Figure 4.1, compared to Figure 4.2).
A greater number of cells responsive to non-anterior tongue receptive fields on the intact side might have provided conclusive evidence that CT denervation resulted in increased functional input in NST cells.
Factors influencing our results
The lack of more dramatic changes in neural responses indicative of anatomical and functional alterations could be due to numerous reasons including: site studied, anesthetic regimen, time of observation, experimental method and parameters assessed. Although many studies in other sensory systems have documented particularly dramatic changes in the cortex, studying taste responses in the insular cortex was impractical since a clear orotopic organization of taste has not been found (Ogawa, et al., 1992; reviewed in
Yamamoto, 1984). However, the orotopic organization and response properties
134 of neurons in the NST and second order taste relay, the parabrachial nucleus
(PBN) have been well studied in rats (e.g. Doetsch and Erickson, 1970; Halpem and Nelson, 1965; Travers and Norgren, 1995; Norgren and Pfaffmann, 1975; reviewed in Travers, et al., 1987; Norgren, 1984; Norgren, 1993). This information was critical In determining whether denervation results in reorganization. The NST was chosen because taste responsive neurons are found in one continuous area whereas taste responsive neurons in the PBN are found in two discrete areas (Halsell and Travers, 1997). Ethyl carbamate and sodium pentobarbital were used as sedatives during electrophysiologic recording sessions. Barbiturates are known to depress cortical activity (Clark and Rosner,
1973). Cortical depression could have an effect on taste responses in the brainstem because the cortex projects to the NST. The degree of such an effect is unknown, but NST cells recorded in awake rats have a higher mean spontaneous rate than those recorded from anesthetized rats (Nakamura and
Norgren, 1991). Therefore, using awake rats may yield different results than those observed in anesthetized rats. CT anesthesia (acute) and CT-transected
(chronic) preparations were chosen to encompass potential changes that occurred over time. The observation times and experimental procedures were based on two previous studies mentioned above (Halpem and Nelson, 1965;
Sollars and Bernstein, 1996). Because both of these studies showed physiological and behavioral changes after CT anesthesia or transection, respectively, its unlikely that the time of observation and procedures had a significant impact on our results.
135 Lastly, the parameters that were assessed in these studies are critical In
the determination of whether CT denervation leads to central reorganization In
the NST. Determining changes In orotopic organization and receptive field
organization of taste responses was a major thrust of this research. These
parameters were assessed because of physiological, psychophysical and
behavioral observations made In animals and humans after peripheral deafferentatlon. Halpern and Nelson (1965) found that NST cells which responded to stimulation of the anterior tongue and another receptive field
Increased response to non-anterior tongue stimulation during CT anesthesia In rats. They assumed that the non-anterior tongue response was from the posterior tongue. Psychophyslcally, Lehman et al. (1995) found an Increase In the perceived Intensity of quinine when the circumvallate papillae were stimulated during CT anesthesia. Behavlorally, patients with CT damage do not commonly express a loss of “real-world” taste (Miller and Bartoshuk, 1991). To account for these findings. It was hypothesized that the CT normally Inhibits responses mediated by the glossopharyngeal nerve (Miller and Bartoshuk, 1991;
Lehman, et al., 1995; Yanaglsawa, et al., 1998). Given this Information and the reported receptive field changes In the somatosensory system. It seemed obvious to look for receptive field changes In the taste system after CT anesthesia or transection. However, the taste system Is not as discretely organized as the somatosensory system, although orotopic organization Is present at the level of the NST (Travers and Norgren, 1995). The taste system
Is assumed to be more concerned with taste quality rather than stimulus location.
136 If this is true, then one would expect changes in chemosensitive response
profiles rather than orotopic alterations. Individual tastants were seldom used in
the CT anesthesia study, because we were unable to keep most of the cells
isolated for a long enough time to test responses of individual receptive fields
and tastants in the same cell. However, one cell tested in this way did respond
differently to individual tastants compared to taste mixture during anesthesia.
That is, the response to whole mouth stimulation with taste mixture did not
change, whereas the response to quinine stimulation decreased and the
response to hydrochloric acid stimulation increased during CT anesthesia. In the
CT-transected rats, individual tastants were commonly used. Sucrose and
quinine responses tended to increase on the CT-cut side but these changes were not significant. With the increased statistical power of the CT anesthesia
model, it would be interesting to replicate the anesthesia study using individual tastants in awake rats. Of course, the alternative conclusion is that the central gustatory system is resistant to large-scale changes after deafferentatlon.
Model
Although robust changes were not seen after CT anesthesia or denervation, subtle changes were found that could have important behavioral consequences. How do these changes relate to or explain the psychophysical and clinical observations gathered thus far? A model can help in our understanding of this question. As more evidence emerges concerning the circuitry, complexity of synaptic arrangements, neuromodulators/
137 neurotransmitters and receptors in NST, we will be able to expand and revise this model. First, allow me to give a very brief overview of what is known concerning gustatory processing at the level of the NST. Afferent input from facial, glossopharyngeal and vagal nerves synapse on NST cells primarily in an excitatory manner (Smith, et al., 1997; Grabauskas and Bradley, 1996; Wang and Bradley, 1995; Li and Smith, 1997). Some NST cells project rostrally (e.g.,
Norgren, 1978) and through a series of synapses affect cortical cells (reviewed in
Norgren, 1993). Descending cortical input in NST is excitatory or inhibitory (Di
Lorenzo and Monroe, 1995). Anatomically, synaptic contacts have been found that resemble excitatory and inhibitory synapses in other systems (Whitehead,
1993). NMDA and GABAergic receptors are present in the NST (Smith, et al.,
1997; Liu, et al., 1993; Wang and Bradley, 1995).
Given this limited amount of information, I would like to propose a very simple model of gustatory processing in the NST. CT anesthesia (acute denervation) leads to a decrease in spontaneous activity for many NST cells
(Dinkins and Travers, in press). The decrease in spontaneous activity (noise) may result in an increased signal-to-noise ratio for cells that still respond to taste stimulation. This change in signal-to-noise ratio could potentially account for
Immediate and short-term taste compensation. However, over time cortical influences or synaptic changes in the NST may allow for long-term compensation after chronic CT denervation. The eventual recovery of spontaneous activity after CT transection (Study 2) may coincide with an increase in taste-elicited responses (signal) by increased synaptic efficacy from intact primary afferent
138 nerves. This effect may be partially due to increased facilatory corticofugal input potentially mediated through NMDA receptors in the NST. If so, then the role of
NMDA receptors could be similar in the NST as that found in the hippocampus and visual cortex (Kirkwood and Bear, 1994; Kirkwood, et al., 1993). That is, long-term potentiation could occur in the NST via NMDA receptors, thereby allowing the cortex to modulate the activity of NST cells over long periods of time. However, it should be noted that we do not know if synapses from corticofugal projections involve NMDA receptors. Based upon the results from the current studies, since anesthetizing the CT nerve resulted in decreased spontaneous rate of NST cells (Dinkins and Travers, in press), decreased CT input may reduce the amount of excitatory input to the cortex. Intrinsic mechanisms in the cortex may react to a loss of excitatory input by increasing its facilatory descending output to the NST. The increased neuromodulatory role of the cortex via NMDA receptors could favor excitation in the NST. Hence the potential for afferent input to result in increased firing of NST cells to taste stimulation in chronically denervated rats. Specifically, increased facilitatory input in NST may decrease action potential threshold resulting in increased responses from normally weak input. Additionally, increased synaptic strength could be due to a Hebbian mechanism in which coincidental action potentials in the pre and postsynaptic cells increase the ability of the presynaptic cell to influence the postsynaptic cell (Hebb, 1949; reviewed in Cruikshank and
Weinberger, 1996). Both increased cortical facilitation and synaptic efficacy
139 could allow previously weak glossopharyngeal and greater superficial petrosal
input to influence NST cells to a greater degree in an excitatory manner.
A change in cortical modulation is not necessary for NST responses to
increase in magnitude after CT transection. Increased synaptic efficacy of intact
nerves may be sufficient to cause increased NST responsivity.
Glossopharyngeal or greater superficial petrosal input (synapses) may
strengthen in the absence of CT input based on a Hebbian mechanism. Since
CT input would no longer be present after CT transection, only afferent input
from intact nerves could influence NST cell firing. One would predict that the
activity of intact nerves and NST cells would coincide more frequently due to the
loss of CT input. Take for instance, a cell that receives equally strong
convergent input from the CT (anterior tongue receptive field) and greater
superficial petrosal nerve (nasoincisor duct receptive field). A cell such as this
may respond well to sodium chloride stimulation on the anterior tongue and
sucrose stimulation on the nasoincisor duct (Travers, et al., 1986). When the
oral cavity is stimulated with sodium chloride, it’s the activity of the CT which drives the NST cell. Conversely, when sucrose is applied to the oral cavity, input from the greater superficial petrosal is responsible for driving the cell. CT transection abolishes the possibility of that cell to be driven by one of its inputs.
Therefore, the cell would only be activated by its remaining input, the greater superficial petrosal nerve. Based on a Hebbian mechanism, intact nerves
(glossopharyngeal and greater superficial petrosal) might be able to increase their synaptic strength on NST cells that received convergent input because of
140 the differences in receptive field chemosensitivities. Granted this example has
been grossly oversimplified, but if the dominant input reflects this simple circuit,
then the same result may occur in a more complex circuit.
Taste phantoms
This model may account for the physiological changes seen in these
experiments as well as certain psychophysical and clinical observations that
have already been discussed. Additionally, this model may partially explain
another clinically relevant problem, that is taste phantoms. During
psychophysical testing, taste phantoms were found during CT anesthesia
(Yanagisawa, et al., 1998). When human subjects were anesthetized, some
reported “sour”, “salty”, “sweet”, “bitter” or “metallic” tastes which were commonly
located on the contralateral posterior tongue. Topical anesthetic applied to the
site of the taste phantom usually abolished it. The authors have proposed that
taste phantoms arise from a release of inhibition (disinhibition) of unanesthetized
or intact nen/es (Yanagisawa, et al., 1998). Based on the results from the CT
anesthesia study in rats (Dinkins and Travers, in press), we propose that taste
phantoms may be due to a decrease in spontaneous activity in the NST. Both explanations are plausible and could together account for the changes seen after
CT anesthesia or damage. A decrease in the spontaneous activity lowers noise
in the taste system so that any signal from saliva could be perceived as a taste phantom. Contrary to a release of inhibition phenomenon, our hypothesis maintains that increased input does not necessarily have to occur to result in
141 altered taste perceptions. Since denervation decreases the background level of activity (noise) of CT-innervated cells, the resultant activity in non-CT-innervated cells is increased relative to background activity prior to denervation. The percept of a taste phantom does not necessarily occur in a subset of cells, but includes the entire population of gustatory responsive neurons. This relative change in activity may be perceived as a taste". This effect may also explain why taste phantoms usually disappear when topical anesthetic is applied to the location of the phantom. Topical anesthesia may reduce spontaneous activity of receptive fields not innervated by the CT leading to a loss in signal. Moreover, taste phantoms that arise from nerve damage may also be explained by this model. For example, a taste phantom was identified in a patient after CT transection that became worse after topical anesthetic was applied to the mouth
(Bartoshuk, et al., 1994). If damage to gustatory nerves lead to the formation of neuromas, spontaneous activity from the neuroma may result in neurogenic tastes. Application of topical anesthetic in the mouth may decrease spontaneous activity of unaffected receptive fields, but not change the spontaneous activity (signal) derived from the neuroma. The net effect would be an increase in the signal-to-noise ratio in the CNS due to the continued activity of the neuroma. Hence, the taste phantom would become worse with the application of topical anesthetic. The majority of taste phantoms caused by unilateral CT damage improve over time (Bull, 1965). Sixty percent of patients that reported symptoms consistent with a taste phantom improved over a several
142 month period (average 3-4 months). Based on our physiologic data, this could be due to a recovery of spontaneous activity of affected cells in the NST (Study
2).
143 Figure 4.1
Individual responses to individual receptive field stimulation with taste mixture
(study 2).
Responses due to whole mouth (WM), anterior tongue (AT), nasoincisor ducts (NID), foliate papillae (FOL), and soft palate (SP) stimulation as well as spontaneous rate (SPON) are shown for each cell on the CT intact (left panels) and CT cut sides (right panels). The responses for each cell are stacked vertically. Suprathreshold responses are indicated by filled bars, whereas subthreshold responses are indicated by open bars.
144 CT TRANSECTION STUDY CT Intact side CT cut side WM 40 •
30 .
2 0 .
10 .
10 20 30 40 50 60 0 10 2 0 30 40 50
2 0 AT 2 0 AT
1 0 1 0 - llllllll
0 llllllllllfllllln.. h n U UU u u
0 10 20 30 40 SO 60 0 10 20 30 40 50 V 40 40 NID NID 1 30 30 • Q. 20 - 20 . CO 10 10
- nil n b i l l .. l l l l l n - , , .0 in in
10 20 30 40 50 FOL
_ m m mnH n_nw IJ - Ü lin 1
10 20 30 40 50 60 10 20 30 40 50 10 - SP 10 1 11 - u “ u u ...... «- j ■ -u* k . . . Hill.
10 20 30 40 50 60 10 20 30 40 50
10 • SPON 10 . SPON 1 1 illli llJ—|..li lilfi .1 !.. . 1 . 0 il. ilfl 0 ■ LllJlil.l_m._..l y.l. I» ■ 10 20 30 40 50 60 10 20 30 40 50 Figure 4.1 Cell
145 Figure 4.2
Individual responses to individual receptive field stimulation with taste mixture
(study 1).
Responses due to whole mouth (WM), anterior tongue (AT), nasoincisor ducts (NID) and foliate papillae (FOL) stimulation as well as spontaneous rate
(SPON) are shown for each cell before CT anesthesia (left panels) and during
CT anesthesia (right panels). The responses for each cell are stacked vertically.
Suprathreshold responses are indicated by filled bars, whereas subthreshold responses are indicated by open bars.
146 CT ANESTHESIA STUDY
BEFORE ANESTHESIA DURING ANESTHESIA WM ■
, 1 1 jlll illl.i. - 10 15 20 25 30
SO •
40
0 30 0) i 2 20 1 „ Q. CO 0 III, li
10 15 20 25 30 0 10 15 20 25 30 NID NID
10 ■
1 III . .
10 15 20 25 30 5 10 15 20 25 30
30 FOL 30 FOL
20
lihw 0 l l l l . . .
-10
10 15 20 25 30 0 5 10 15 20 25 30
10 SPON 10 • SPON
10 IS 20 25 30 0 5 10 15 20 25 30 Cell Figure 4.2
147 LIST OF REFERENCES
Bartoshuk, L., Duffy, V., Kveton, J., Catalanotto. F., Weiffenbach, J. 1994. Taste phantoms: diagnosis via topical anesthetics. Chem Senses 19:440.
Bull, T. R. 1965. Taste and the chorda tympani. J. Laryngol. Otol. 79:479-93.
Buonomano, P., Merzenich, M. 1998. Cortical plasticity, from synapses to maps. Annu Rev Neurosci 21:149-86.
Calford, M. B., Tweedale, R. 1988. Immediate and chronic changes in responses of somatosensory cortex in adult flying-fox after digit amputation. Nature 332(31/March):446-48. 31/March.
Catalanotto, P., Bartoshuk, L., Moestrom, K., Gent, J., Fast, K. 1993. Effects of anesthesia of the facial nerve on taste. Chem. Senses 18(5):461-70.
Clark, P., Rosner, B. 1973. Neurophysiologic effects of general anesthetics: I. The Electroencephalogram and sensory evoked responses in man. Anesthesiol. 38(6, June):564-82. June.
Cruikshank, S. J., Weinberger, N. M. 1996. Evidence for Hebbian hypothesis in experience-dependent physiological plasticity of neocortex: a critical view. Brain Research Reviews 22:191-228.
Pahl, M., Erickson, R., Simon, S. 1997. Neural responses to bitter compounds in rats. Brain Res. 756:22-34.
Pickman, J. P., Smith, P. 1989. Topographic distribution of taste responsiveness in the hamster medulla. Chem Senses 14:231-247.
Pi Lorenzo, P. M., Monroe, S. 1995. Corticofugal influence on taste responses in the nucleus of the solitary tract in the rat. J. Neurophysiol. 74:258-72.
Pinkins, M., Travers, S. in press. Effects of chorda tympani nerve anesthesia on taste responses on the NST. Chem Senses.
148 Doetsch, G. S., Erickson, R. P. 1970. Synaptic processing of taste-quality information in the nucleus tractus solitarius of the rat. J. Neurophysiol. 33:490-507.
Donoghue, J. P. 1995. Plasticity of adult sensorimotor representations. Curr. Opin. Neurobioi. 5:749-54.
Dostrovsky, J., Millar, J., Wall, P. 1976. The immediate shift of afferent drive of dorsal column nucleus cells following deafferentation: a comparison of acute and chronic deafferentation in gracile nucleus and spinal cord. Experimental Neurology 52:480-95.
Duchamp-Viret, P., Coronas, V., Delaleu, J., Moyse, E., Duchamp. A. 1997. Dopaminergic modulation of mitral cell activity in the frog olfactory bulb: a combined radioligand binding-electrophysiological study. Neurosci. 79(1)1203-16.
Elbert, T., Pantev, 0., Weinbruch, C., Rockstroh, B., Taub, E. 1995. Increased cortical representation of the fingers of the left hand in string players. Science 270(13/October):305-07. 13/October.
Flor, H., Elbert, T. M., W., Pantev, 0., Weinbruch, 0., Taub, E. 1998. Cortical reorganization and phantom phenomena in congenital and traumatic upper extremity amputees. Exp Brain Res 119:205-12.
Flor, H., Elbert, T., Knecht, S., Weinbruch, C., Pantev, C. 1995. Phantom limb pain as a perceptual correlate of massive cortical reorganization in upper extremity amputees. Nature 375:482-84.
Florence, S., Kaas, J. 1995. Large-scale reorganization at multiple levels of the somatosensory pathway follows therapeutic amputation of the hand in monkeys. The Joumal of Neuroscience 15(12):8083-95.
Frank, M. E. 1991. Taste-responsive neurons of the glossopharyngeal nerve of the rat. J. Neurophysiol. 65:1452-63.
Frank, M., Contreras, R., Hettinger, T. 1983. Nerve fibers sensitive to ionic taste stimuli in chorda tympani of the rat. J. Neurophysiol. 50:941-60.
Garraghty, P. E., Kaas, J. H. 1991. Functional reorganization in adult monkey thalamus after peripheral nerve injury. NeuroReport 2:747-50.
149 Garraghty, P., LaChica, E., Kaas, J. 1991. Injury-induced reorganization of somatosensory cortex is accompanied by reductions in GABA staining. Somatosensory and Motor Research 8(4):347-54.
Gilbert, C., Wiesel, T. 1992. Receptive field dynamics in adult primary visual cortex. A/afurs 356:150-52.
Grabauskas, G., Bradley, R. 1996. Synaptic interactions due to convergent input from gustatory afferent fibers in the rostral nucleus of the solitary tract. J Neurophysiol 76(5):2919-27.
Halpern, B. P., Nelson, L. M. 1965. Bulbar gustatory responses to anterior and to posterior tongue stimulation in the rat. Am. J. Physiol. 209:105-10.
Halsell, C. B., Frank, M. E. 1992. Organization of taste-evoked activity in the hamster parabrachial nucleus. Brain Res. 572:1-5.
Halsell, C. B., Travers, J. B., Travers, S. P. 1993. Gustatory and tactile stimulation of the posterior tongue activate overlapping but distinctive regions within the nucleus of the solitary tract. Brain Res. 632:161-73.
Halsell, C., Travers, S. 1997. Anterior and posterior oral cavity responsive neurons are differentially distributed among parabrachial subnuclei in rat. J. Neurophys. 78:920-38.
Hamilton, R. B., Norgren, R. 1984. Central projections of gustatory nerves in the rat. J. Comp. Neurol. 222:560-77.
Hayama, T., Ito, S., Ogawa, H. 1985. Responses of solitary tract nucleus neurons to taste and mechanical stimulations of the oral cavity in decerebrate rats. Exp. Brain Res. 60:235-42.
Hebb. D.O. 1949. The Organization of Behavior. A Neuropsychological Theory. New York: Wiley.
Hill, D., Formaker, B., White, K. 1990. Perceptual characteristics of the amiloride-suppressed sodium chloride taste response in the rat. Behav Neurosci 104(5):734-41.
Hill, D., Mistretta, 0. 1990. Developmental neurobiology of salt taste sensation. Trends in Neurosci. 13:188-95.
150 Hill, D., Mistretta, C., Bradley, R. 1982. Developmental changes in taste response characteristics of rat single chorda tympani fibers. J. Neurosci. 2:782-90.
Hill, □., Przekop, P., Jr. 1988. Influences of dietary sodium on functional taste receptor development: A sensitive period. Science 241:1826-28.
Hill, T., Yirrell, Blyth, W. 1986. Infection of the adrenal gland as a route to the central nervous system after viremia with herpes simplex virus in the mouse. J. General Virology 67:309-20.
Hu, J., Dostrovsky, J., Lenz, Y., Ball, G., Sessle, B. J. 1986. Tooth pulp deafferentation is associated with functional alterations in the properties of neurons in the trigeminal spinal tract nucleus. J Neurophysiol. 56:1650- 1668.
Hubei, D., Wiesel, T. 1970. The period of susceptibility to the physiological effects of unilateral eye closure in kittens. J Physiol 206:419-36.
Jones, E. 1993. GABAergic neurons and their role in cortical plasticity in primates. Cerebral Cortex 3(Sept/Oct):361-72.
Kaas, J. H., Merzenich, M. M., Killackey, H. P. 1983. The reorganization of somatosensory cortex following peripheral nerve damage in adult and developing mammals. Annual Reviews in Neuroscience 6:325-56.
Kaas, J. 1991. Plasticity of sensory and motor maps in adult mammals. Annu. Rev. o f Neurosci. 14:137-67.
Kalaska, J., Pomeranz, B. 1979. Chronic paw degeneration causes an age-dependent appearance of novel responses from forearm in 'paw cortex' of kittens and adult cats. J Neurophysiology 42i2):6^ 8-33.
Kirkwood, A., Bear. M.F. 1994. Hebbian synapses in visual cortex. The Joumal o f Neuroscience 14(3):1634-45.
Kirkwood, A., Dudek, S., Gold, J. A., C D., Bear, M. 1993. Common forms of synaptic plasticity in the hippocampus and neocortex in vitro. Science 260:1518-21.
Lasiter, P. S. 1992. Postnatal development of gustatory recipient zones within the nucleus of the solitary tract. Brain Res. Bull. 28:667-77.
151 Lasiter, P. S., Kacheie, D. L. 1990. Effects of early postnatal receptor damage on development of gustatory recipient zones within the nucleus of the solitary tract. Dev Brain Res. 55(1):57-71.
Lasiter, P. 1993. Axonal migratory pattems and target findings in gustatory recipient zones of the solitary nucleus during post natal development. In; NIDCD monograph: Development, growth, and senescence in the chemical senses. Vol.3. U.S. Department of Health Services; Public Health Service; National Institute of Health; NIH Publication No. 93-3483; 47-55.
Lehman, C. D., Bartoshuk, L. M., Catalanotto, F. 0., Kveton, J. F., Lowlicht, R. A. 1995. Effect of anesthesia of the chorda tympani nerve on taste perception in humans. Physiol. Behav. 57:943-51.
Li, C., Smith, D. 1997. Glutamate receptor antagonists block gustatory afferent input to the nucleus of the solitary tract. J. Neurophysiol. 77:1514-25.
Liu, H., Behbehani, M. M.. Smith, D. V. 1993. The influence of GABA on cells in the gustatory region of the hamster solitary nucleus. Chem. Senses 18:285-305.
Markison, S., St John, S., Spector, A. 1995. Glossopharyngeal nerve transection does not compromise the specificity of taste-guided sodium appetite in rats. Am J Physiol Regul Integr Comp Physiol 269(1 ):R215-21.
Melzer, P., Smith, C. 1998a. Plasticity of cerebral metabolic whisker maps in adult mice after whisker follicle removal - 1. Modification in barrel cortex coincide with reorganization of follicular innervation. Neuroscience 83(1):27-41.
Melzer, P., Smith, 0. 1998b. Plasticity of cerebral metabolic whisker maps in adult mice after whisker follicle removal - II. Modifications in the subcortical somatosensory system. Neuroscience 83(1):43-61.
Merzenich, M. M., Kaas, J. H., Wall, J. T., Sur, M., Nelson, R. J., et al. 1983b. Progression of change following median nerve section in the cortical representation of the hand in areas 3b and 1 in adult owl and squirrel monkeys. Neuroscience 10(3):639-65.
Merzenich, M. M., Kaas, J. H., Wall, J., Nelson, R. J., Sur, M., et al. 1983a. Topographic reorganization of somatosensory cortical areas 3b and 1 in adult monkeys following restricted deafferentation. Neuroscience 8(1):33-55.
152 Merzenich, M. M., Recanzone, G., Jenkins, W., Allard, T., Nudo, R. 1988. Cortical representational plasticity. In Neurobiology of Neocortex, ed. P. Rakie, W. Singer, pp. 42-67. Berlin: Wiley.
Miller, I. J., Jr., Bartoshuk, L. M. 1991. Taste perception, taste bud distribution, and spatial relationships. In Smell and Taste in Health and Disease, ed. T. Getchell, R. Doty, L. Bartoshuk, J. Snow, Jr., pp. 205-33. N.Y.: Raven Press.
Miller, I. J., Gomez, M. M., Lubarsky, E. H. 1978. Distribution of the facial nerve to taste receptors in the rat. Chem. Senses and Flav. 3:397-410.
Miyaoka, Y., Shingai, T., Takahashi, Y., Yamada, Y. 1997 Jul. Responses of parabrachial nucleus neurons to chemical stimulation of posterior tongue in chorda tympani-sectioned rats. Neurosci Res 28{3):20A-07.
Nakamura, K., Norgren, R. 1991. Gustatory responses of neurons in the nucleus of the solitary tract of behaving rats. J. Neurophysiol. 66:1232-48.
Neugebauer, V., Schaible, H.-G. 1990. Evidence for a central component in the sensitization of spinal neurons with joint input during development of acute arthritis in cat's knee. J Neurophysiol 64(1):299-99.
Norgren, R. 1984. Central neural mechanisms of taste. In Handbook of Physiology, Section 1: The Nervous System, ed. J. M. Brookhart, V. B. Mountcastle, I. Darian-Smith. Bethesda Maryland: American Physiological Society.
Norgren, R. 1993. Gustatory system. In The Rat Nervous System, ed. G. Paxinos, pp. 1-47. San Diego, California: Academic Press.
Norgren, R. 1978. Projections from the nucleus of the solitary tract in rat. Neurosci. 3:207-18.
Norgren, R., Grill, H. 1976. Efferent distribution from the cortical gustatory area in rats. Sac. Neurosci. Absts. 2:124.
Norgren, R., Pfaffmann, C. 1975. The pontine taste area in the rat. Brain Res. 91:99-117.
Nystrom, B., Hagbarth, K. 1981. Microelectrode recordings from transected nerves in amputees with phantom nerve pain. Neuroscience Letters 27:211-16.
153 O'Leary, D., Ruff, N., Dyck, R. 1994. Development, critical period plasticity, and adult reorganizations of mammalian somatosensory systems. Current Opinion in Neurobiology 4:535-44.
Ogawa, H., Murayama, N., Hasegawa, K. 1992. Difference in receptive field features of taste neurons in rat granular and dysgranular insular cortices. Exp. Brain Res. 91:408-14.
Panetsos, F., Nunez, A., Avendano, C. 1995. Local anesthesia induces immediate receptive field changes in nucleus gracilis and cortex. NeuroReport 7:150-52.
Pettit, M., Schwark, H. 1993. Receptive field reorganization in dorsal column nuclei during temporary denervation. Science 262:2052-56.
Pons, T. P., Garraghty, P. E., Ommaya, A. K., Kaas, J. H., Taub, E.. et al. 1991. Massive cortical reorganization after sensory deafferentation in adult macaques. Sc/ence 252(28/June): 1857-9. 28/June.
Rasmusson, D. D., Turnbull, B. G. 1983. Immediate effects of digit amputation on SI cortex in the raccoon: unmasking of inhibitory fields. Brain Research 288:368-70.
Rice, J. 1963. The chorda tympani in stapedectomy. J. Laryngol. Otol. 77:943-44.
Schwaber, J., Graves, E., Paton, J. 1993. Computational modeling of neuronal dynamics for systems analysis: Application to neurons of the cardiorespiratory NTS in the rat. Brain Res 604(1-2):126-41.
Shaw, 0. A., Lanius, R. A., van den Doe!, K. 1994. The origin of synaptic neuroplasticity: crucial molecules or a dynamical cascade? Brain Research Reviews 19:241-63.
Smith, D. V., Van Buskirk, R. L., Travers, J. B., Bieber, S. L. 1983. Gustatory neuron types in hamster brain stem. J. Neurophysiol. 50:522-40.
Smith, D., Li, C.-S., Davis, B. 1997. Excitatory and inhibitory modulation of taste responses in the hamster brainstem. In Olfaction and taste XII, ed. C. Murphy, 0. Greer. New York: New York Academy of Sciences.
Sollars, S., Bernstein, I. 1996. Neonatal chorda tympani transection alters adult preference for ammonium chloride in the rat. Behav Neurosci 110:551-58.
154 Spector, A. C., Grill, H. J. 1992. Sait taste discrimination after bilateral section of chorda tympani or glossopharyngeal nen/es. Am. J. Physiol. 263:R169-76.
Spector, A., Schwartz, G., Grill, H. 1990. Chemospecific deficits in taste detection after selective gustatory deafferentation in rats. Am. J. Physiol. 258:R820-826.
Spector, A., Travers, S., Norgren, R. 1993. Taste receptors on the anterior tongue and nasoincisor ducts of rats contribute synergistically to behavioral responses to sucrose. Behav. Neurosci. 107(4):694-702.
St John, S., Markison, S., Guagliardo, N., Hackenberg, T., Spector, A. 1997a. Chorda tympani transection and selective desalivation differentially disrupt two-lever salt discrimination performance in rats. Behavioral Neuroscience 111:450-59.
St John, S., Markison, S., Spector, A. 1997b. Chorda tympani nerve transection disrupts taste aversion learning to potassium chloride, but not sodium chloride. Behavioral Neuroscience 111:188-94.
St. John, S., Garcea, M., Spector, A. 1994. Combined, but not single, gustatory nerve transection substantially alters taste-guided licking behavior to quinine in rats. Behav. Neurosci. 108(1):131-40.
Sweazey, R. D., Smith, D. V. 1987. Convergence onto hamster medullary taste neurons. Brain Res. 408:173-85.
Travers, J. B., Grill, H. J., Norgren, R. 1987. The effects of glossopharyngeal and chorda tympani nerve cuts on the ingestion and rejection of sapid stimuli: an electromyographic analysis in the rat. Behavioral Brain Res. 25:233-46.
Travers, S. P., Nicklas, K. 1990. Taste bud distribution in the rat pharynx and larynx. Anat. Rec. 227:373-79.
Travers, S. P., Norgren, R. 1991. Coding the sweet taste in the nucleus of the solitary tract: Differential roles for anterior tongue and nasoincisor duct gustatory receptors in the rat. J. Neurophysiol. 65:1372-80.
Travers, S. P., Norgren, R. 1995. Organization of orosensory responses in the nucleus of the solitary tract of the rat. J. Neurophysiol. 73:2144-62.
155 Travers, S. P., Pfaffmann, C., Norgren, R. 1986. Convergence of lingual and palatal gustatory neural activity in the nucleus of the solitary tract. Brain Res. 365:305-20.
Travers, S. P., Smith, D. V. 1984. Responsiveness of neurons in the hamster parabrachial nuclei to taste mixtures. J. Gen. Physiol. 84:221-50. van der Kooy, D., Koda, L Y., McGinty, J. F., Gerfen, C. R., Bloom, F. E. 1984. The organization of projections from the cortex, amygdala, and hypothalamus to the nucleus of the solitary tract in rat. J. Comp. Neurol. 224:1-24.
Van der Loos, H., Woolsey, T. 1973. Somatosensory cortex: structural alterations following early injury to sense organs. Science 179:395-97.
Wall, P., Gutnick, M. 1974. Ongoing activity in peripheral nerves: The physiology and pharmacology of Impulses originating from a neuroma. Experimental Neurology 43:580-93.
Wang, L., Bradley, R. 1995. In vitro study of afferent synaptic transmission in the rostral gustatory zone of the rat nucleus of the solitary tract. Brain Res 702(1-2):188-98.
Weinberger, N. 1995. Dynamic regulation of receptive fields and maps in the adult sensory cortex. Annu Rev Neurosci 18:129-58.
Whitehead, M. 0. 1993. Distribution of synapses on identified cell types in a gustatory subdivision of the nucleus of the solitary tract. J. Comp. Neurol. 332:326-40.
Whitehead, M. C., Frank, M. E. 1983. Anatomy of the gustatory system in the hamster: Central projections of the chorda tympani and the lingual nerve. J. Comp. Neurol. 220:378-95.
Whitehead, M., McGlathery, S., Manion, B. 1995. Transganglionic degeneration in the gustatory system consequent to chorda tympani damage. Exp Neurol 132(2):239-50.
Whiteside, B. 1927. Nerve overlap in the gustatory apparatus of the rat. J. Comp. Neurol. 44:363-77.
Wiesel, T. N., Hubei, D. H. 1963a. Effects of visual deprivation on morphology and physiology of cells in the cat’s lateral geniculate body. J Neurophysiol. 26:978-993.
156 Wiesel, T. N., Hubel, D. H. 1963b. SIngle-cell responses in striate cortex of kittens deprived of vision in one eye. J Neurophysiol. 26:1003-1017.
Wilson. P., Kitchener, P. 1996. Plasticity of cutaneous primary afferent projections to the spinal dorsal horn. Progress in Neurob/o/ogy 48:105-29.
Woolf, C., Shortland, P., Coggeshall, R. 1992. Peripheral nerve injury triggers central sprouting on myelinated afférents. Nature 355:75-78.
Xu, J., Wall, J. 1997. Rapid changes on brainstem maps of adult primates after peripheral injury. Brain Research 774:211-15.
Yamamoto, T. 1984. Taste Responses of Cortical Neurons. Progress in Neurobiology 23:273-315.
Yamamoto, T., Kawamura, Y. 1975. Cortical responses to electrical and gustatory stimuli in the rabbit. Brain Res. 94:447-63.
Yanagisawa, K., Bartoshuk, L, Catalanotto, F., Karrer. T.A., Kveton, J. 1998. Anesthesia of the chorda tympani nerve and taste phantoms. Physiology and Behavior 63:329-35.
157 APPENDIX
Overview of the gustatory system
Taste buds are located In several regions in the rat oral cavity. Taste bud subpopulations are located on the anterior tongue (fungiform papillae), posterior tongue (foliate and circumvallate papillae), hard palate (nasoincisor duct) and soft palate. The anterior tongue and hard palate (nasoincisor duct) are considered in the anterior oral cavity (AO), whereas the posterior tongue
(fungiform and circumvallate papillae) and soft palate are in the posterior oral cavity (PC). Taste receptor cells on the anterior tongue are innervated by the chorda tympani nerve (CT), on the posterior tongue by the lingual-tonsillar branch of the glossopharyngeal nerve (GL)(CN IX), and on the palate by the greater superficial petrosal nerve (GSR). Both the CT and GSR are branches of the facial nerve (CN VII).
Central projections of these nerves terminate in the first order central taste relay, the nucleus of the solitary tract (NST) located in the medulla. The CT and
GSR terminal fields are overlapping in the rostral part of the rostral NST. The GL terminal field is located more medial and caudal to facial nerve terminal fields.
These terminal field patterns result in the orotopic organization of taste responsive neurons in the NST, that is anterior oral cavity responsive neurons
158 are primarily located rostral and lateral to neurons that respond to the posterior oral cavity (see Travers and Norgren, 1995).
159 IMAGE EVALUATION TEST TARGET (Q A -3 ) /
/&
/
%
12.8 1.0 U à ML: m 12.0 i.i 1.8
1.25 1.4 1.6
150mm
V v : /APPLIED A IIVMGE . Inc ^ S i 1653 East Main street Rochester. NY 14609 USA ■A Phone: 716/4820300 ------Fax: 716/288-5989
0 1993. Applied Image. Inc.. All Rights Reserved