- Email: [email protected]
Role of parabrachial nucleus in submandibular salivary secretion induced by bitter taste stimulation in rats
Autonomic Neuroscience: Basic and Clinical 88 Ž2001. 61–73 www.elsevier.comrlocaterautneu
Role of parabrachial nucleus in submandibular salivary secretion induced by bitter taste stimulation in rats Ryuji Matsuo a,) , Yoji Yamauchi a , Motoi Kobashi a , Makoto Funahashi a , Yoshihiro Mitoh a , Akira Adachi b b
a Department of Oral Physiology, Okayama UniÕersity Dental School, 2-5-1 Shikata-cho, Okayama 700-8525, Japan Department of Clinical Welfare SerÕice, Kyushu UniÕersity of Health and Welfare, Nobeoka, Miyazaki 882-8508, Japan
Received 11 December 2000; received in revised form 26 January 2001; accepted 29 January 2001
Abstract When rats lick a bitter taste solution such as quinine-hydrochloride, they secrete profuse amounts of saliva. The salivation has a higher flow rate than that induced by other qualities of taste stimulation: sweet, salty, and sour. The present study is aimed to clarify the neural mechanism of the quinine-evoked salivation by means of behavioral, neuroanatomical, and electrophysiological experiments. Behaviorally, submandibular salivary secretion and rejection behavior Žgaping. were observed in normal rats, as well as in rats chronically decerebrated at the precollicular level. In chronically decerebrate rats, these quinine-evoked reactions were strongly suppressed by destruction of the medial part of the parabrachial nucleus, including the so-called taste area, and ventral part of the parabrachial nucleus, including the pontine reticular formation. Neuroanatomical study using a retrograde tracer, Fluoro-gold, revealed that the neurons sending their axons to the superior salivatory nucleus, parasympathetic secretory center, were located mainly in the pontine reticular formation ventral to the parabrachial nucleus, not in the parabrachial taste area. Extracellular neural activity was recorded from the parabrachial region in decerebrate rats, and responsiveness to taste stimulation, jaw movements, and electrical stimulation of the superior salivatory nucleus was examined. Neurons responsive to both taste stimulation and antidromic stimulation of the superior salivatory nucleus were found in the pontine reticular formation ventral to the parabrachial nucleus, which responded well to quinine and HCl taste stimuli. Neurons in the parabrachial taste area could respond to four qualities of taste stimulation, but not to antidromic stimulation of the salivary center. These results suggest that aversive taste information from the parabrachial taste area reaches the salivary secretory center via the reticular formation ventral to the parabrachial nucleus. q 2001 Elsevier Science B.V. All rights reserved. Keywords: Salivary secretion; Taste; Rejection; Parabrachial nucleus; Salivatory nucleus
1. Introduction When rats lick a bitter taste substance, e.g., quinine-hydrochloride, they exhibit stereotypic rejection behavior consisting of gaping, face washing, chin rubbing, head shaking, forelimb flailing and paw pushing ŽGrill and Norgren, 1978a.. Matsuo et al. Ž1994. have found that the taste rejection behavior is always accompanied by vigorous salivation, about 30 mlrmin from the unilateral submandibular gland, which is a far higher flow rate than that evoked by licking of a sweet, salty or sour taste solution Žless than about 5 mlrmin.. It is well recognized that the
) Corresponding author. Tel.: q81-86-235-6640; fax: q81-86-2356644. E-mail address: [email protected] ŽR. Matsuo..
gustatory–salivary reflex Že.g., Matsuo, 1999a. and taste rejection behavior occur in decerebrate animals ŽGrill and Norgren, 1978b; Travers et al., 1999., and the fundamental neural structures for the reactions are situated in the lower brainstem. However, precise neural mechanisms underlying the quinine-evoked salivation have not been elucidated yet. One of the problems in studies on the reflex salivation is effects of anesthesia. When anesthetized, decerebrate animals do not profusely secrete saliva on bitter taste stimulation, and secrete only a little saliva on various kinds of taste stimulation Žless than 6 mlrmin from rat submandibular gland. Že.g., Matsuo et al., 1989.. Such relatively weak salivation is abolished by destruction of the rostral part of the solitary nucleus Žthe first order of the taste relay area., but not affected by destruction of the parabrachial taste area Žthe second order of the taste relay area.. This
1566-0702r01r$ - see front matter q 2001 Elsevier Science B.V. All rights reserved. PII: S 1 5 6 6 - 0 7 0 2 Ž 0 1 . 0 0 2 3 4 - X
62
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73
result suggests that the reflex salivation is evoked mainly by taste information mediated through the solitary nucleus. Nevertheless, electrical and chemical stimulation of the parabrachial nucleus induced salivary secretion in anesthetized animals ŽMatsuo et al., 1989; Dick et al., 1992.. From this finding, we speculate that there is a gustatory– salivary reflex arc which involves the parabrachial taste area, but this reflex arc may be susceptible to anesthesia partly because the parabrachial-mediated reflex arc is more polysynaptic than the solitary-mediated one. Moreover, there is a possibility that the parabrachial-mediated reflex arc is responsible for the vigorous salivation during taste rejection, since rats with lesions in the parabrachial nucleus generally displayed smaller aversive responses to bitter taste stimulation in reference to gapes, chin rubs, and intake of bitter taste solution ŽHill and Almli, 1983; Flynn et al., 1991; Spector, 1995.. To test this possibility, it is necessary to examine salivation using decerebrate animals while minimizing the effects of anesthesia. Considering the above-mentioned findings and possibilities, we attempted to investigate the role of the lower brainstem, especially the parabrachial nucleus, in salivation and oromotor reactions evoked by aversive taste stimulation. To this end, we conducted behavioral, neuroanatomical, and electrophysiological experiments. First, we measured submandibular salivary secretion in chronically decerebrate rats and confirmed that vigorous salivation occurs during taste rejection. Second, the neurons sending their axons from the parabrachial region to the superior salivatory nucleus Žthe parasympathetic secretory center. were examined by using a retrograde tracer, Fluoro-gold. Third, with the results of the above neuroanatomical experiment in mind, we examined changes in the rejection behavior and salivation after lesioning of the medial part of the parabrachial region, including the taste relay area, the pontine reticular formation and supra trigeminal area, which project to the superior salivatory nucleus Žrevealed by the present study., and the lateral part of the parabrachial region, which receives general visceral inputs ŽHerbert et al., 1990. and relates to the rejection response to bitter taste ŽYamamoto et al., 1994; Travers et al., 1999.. Fourth, neural responses to taste stimulation, jaw movements, and electrical stimulation of the superior salivatory nucleus were analyzed in the parabrachial region. Through these experiments, we show that the taste rejection behavior is accompanied by salivation, and the aversive taste information related to salivation reaches the salivary center via the parabrachial taste area and pontine reticular formation ventral to the parabrachial nucleus.
2. Materials and methods Sixty-eight adult male Wistar rats weighing 280–350 g ŽCharles River Breeders, Osaka, Japan. were used; 46 for the behavioral study on submandibular salivary secretion
and rejection behavior, five for the neuroanatomical study using Fluoro-gold and 17 for the electrophysiological study on the parabrachial neurons. Animal protocols were in accordance with the Guiding Principles for the Care and Use of Animals, approved by the Council of the Physiological Society of Japan. 2.1. BehaÕioral study Submandibular salivary secretion and jaw muscle activity were recorded from decerebrate Ž n s 9., decerebrate plus parabrachial lesioned Ž n s 26., and sham-operated rats Ž n s 10.. Since decerebrate animals are aphagic and adipsic, Grill and Norgren Ž1978b. maintained decerebrate rats by tube feeding of a liquid diet, and observed variations in behavioral and oro-facial responses to taste stimulation dependent on the postsurgical period. In the present study, decerebration and the sham-operation were performed 3–4 days before the recording of behavioral responses, because the feeding of a liquid diet for several days results in atrophy of the salivary glands ŽNilsson et al., 1990; Scott and Gunn, 1991.. For the surgical operation, the rats were deeply anesthetized with an i.p. injection of sodium pentobarbitone Ž50–75 mgrkg. and prophylactically treated with a penicillin suspension Ž30,000 units, i.m... They were mounted on a stereotaxic apparatus, and the skull was exposed through a midline incision leveled between bregma and lambda. A small slit was drilled in the frontal plane of the skull and the brainstem was cut at the precollicular level with an L-shaped spatula ŽGrill and Norgren, 1978b.. Ten rats were used for a control group. Their skulls were drilled and the surface of the brain was exposed, but the brainstem was not cut. After the surgical operation, animals were returned to their home cage. The control rats were given laboratory food pellets and water. The decerebrate rats were fed 5–10 ml of liquid diet ŽControl Diet, Oriental, Japan. by gavage three times a day Žbetween 9 AM and 7 PM.. At least 2 days after decerebration, 26 rats were anesthetized as described and received bilateral electrolytic lesions in the medial part, lateral part or ventral part of the parabrachial region. Each electrolytic lesion was produced by four to six electrolytic burns by passing 0.3 mA of anodal current for 5 s through a glass-coated elgiloy wire with 50 mm exposed at the tip Žthe tip diameter was about 15 mm.. A burr hole was drilled in the skull over the parabrachial region, the exposed dura was reflected, and the transverse sinus was hocked anteriorly to avoid bleeding. The obex was exposed, and the electrode was introduced into the medial part Ž5.0 mm anterior to the obex, 1.6 and 1.9 mm lateral to midline, 5.0 and 5.5 mm below the brain surface., lateral part Ž5.0 mm anterior to the obex, 2.0 and 2.4 mm lateral to midline, 5.0 and 5.5 mm below the brain surface. and ventral part Ž5.0 mm anterior to the obex, 1.6, 2.0 and 2.4 mm lateral to midline, 6.0 and 6.5 mm below the brain surface. of the parabrachial re-
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73
gion, which stereotaxic coordinates were taken from the atlas of Paxinos and Watson Ž1986.. Following electrolytic burns, the holes of skull were packed with Gelfoam. For recording salivation, the duct of the left submandibular gland was exposed near its orifice at the floor of the mouth through an incision made at the corner of the mouth. The duct was cannulated with polyethylene tubing Ž3-mm-long. of the largest possible bore. The tubing was fixed to the duct and the dorsal edge of the mandibular bone, between incisor and molar teeth, with a butyl-2cyanoacrylate adhesive and dental acrylic resin. The free end of the tubing was connected to a second polyethylene tube Ž0.8-mm inner diameter, 1.1-mm outer diameter, about 7-cm-long. that was bent subcutaneously backwards over the masseter and temporalis muscles, and led out through an incision made on the top of the skull. For oral injection of taste solution, an intraoral fistula was made on the ipsilateral side of the salivary tubing by a similar method to that described by Phillips and Norgren Ž1970.. Briefly, a curved surgical needle was inserted in the buccal mucosa lateral to the upper molar teeth and out through the incision in the skull. The needle was used to guide polyethylene tubing Ž0.8-mm inner diameter, 1.1-mm outer diameter, about 7-cm-long.. The tubing was slipped through the needle, and then the needle was removed. The salivary and intraoral tubings were cemented to the skull with dental acrylic resin. For recording electromyographic ŽEMG. activity of the jaw-closing and -opening muscles, a pair of Teflon-coated stainless steel electrodes Ž50-mm diameter, about 3-mm interpolar distance. with bared ends Ž1 mm. were inserted into the left masseter and anterior digastric muscles, respectively. The EMG electrodes were passed under the skin to a connector on the skull. The connector was also cemented to the skull with dental acrylic resin. The recording experiment was started after recovery from anesthesia Ž24–36 h after the surgical operation.. The animals were moved to a test box Ža clear plastic cylinder 30 cm in diameter and 35-cm high.. The implanted salivary tubing close to the skull was connected to tubing of the same bore Žfilled with saline., and led out the test box. The other end of the tube was connected to a small bottle via a clamp, and then to a pressure-sensitive transducer ŽMPU-0.5 A, Nihon Khoden, Japan.. The flow of saliva was measured as an increase in liquid pressure in the bottle using the transducer, and this was recorded on a pen-recorder. Since the inner space from the salivary duct to the bottle was closed up, increased pressure was released by opening the clamp. When measuring salivary flow, if the volume of saliva reached 13 ml, the clamp was opened to release the pressure, and closed again for the succeeding recording. The system was calibrated by measuring the deflection of the pen induced by injection of a set amount of saline Ž10 ml. into the bottle. The EMG electrodes were connected to a slip-ring connector ŽAPCL-15, Air Precision. via a mating plug and shielded cables. From the
63
connector, EMG activity was fed to a bandpass amplifier Ž30–3000 Hz., monitored by an oscilloscope. The EMG activity and salivary flow data was stored on magnetic tape. For evoking salivation, an aliquot Ž0.1 ml. of taste solution was injected into the mouth through the intraoral fistula. The taste stimuli consisted of 0.1 M NaCl, 0.5 M sucrose, 0.01 M HCl and 0.01–10 mM quinine-hydrochloride. The stimulations were repeated twice, and the order of applying the stimuli was varied each time, and in different animals. At least 15-min rest was taken after termination of the previous salivary and EMG responses, and two water rinses preceded and followed each stimulus. All the chemical stimuli and distilled water were applied at room temperature Ž23–268C.. The onset of taste stimulation and any other obvious behavior such as grooming or taste rejection was noted on magnetic tape by voice commentary. After the recording, the animals which received electrolytic lesions of the parabrachial region were killed with an overdose of sodium pentobarbitone Ž) 100 mgrkg i.p.. and perfused intracardially with isotonic saline and 10% buffered paraformaldehyde. The brains were removed, stored in 10% paraformaldehyde containing 30% sucrose for a few days, then cut in 50 mm coronal sections on a freezing microtome. The sections were mounted onto chrome alum-coated slides and stained with 0.5% cresyl violet. 2.2. Neuroanatomical study with Fluoro-gold For injection of the retrograde tracer Fluoro-gold into the superior salivatory nucleus, the rats were deeply anesthetized, prophylactically treated, and placed in a stereotaxic apparatus as for decerebration. Stereotaxic coordinates for the superior salivatory nucleus were 1.5 mm lateral- and 4.2 mm anterior to the obex and 7.5–7.7 mm ventral to the brain surface. A small burr hole was made in the skull above the nucleus, and an aliquot Ž0.2 ml. of a 1% solution of Fluoro-gold ŽFluorochrome, Englewood, USA. dissolved in saline was injected through a glass micropipette with a tip diameter of 30 mm connected to a 5-ml Hamilton microsyringe. The pipette was left in the brain for 10 min, to prevent tracer leakage along the injection tract. The hole of the skull was packed with Gelfoam and the incision was sutured. Three to five days after the injection, rats were deeply anesthetized as previously, and transcardially perfused with 100 ml of saline followed by 500 ml of a fixative containing 4% paraformaldehyde in 0.1M phosphate buffer ŽpH 7.4.. The brainstems were removed, further fixed in the same fixative for 90 min, and stored in a 20% sucrose solution Ž0.1 M phosphate buffer, pH 7.4. overnight at 48C. Subsequently, the brainstems were cut transversely, 50-mm-thick, on a freezing microtome and mounted onto chrome alum-coated slides; the sections were divided into
64
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73
two series of alternate serial sections. One set was air-dried overnight, coverslipped, and examined under ultraviolet illumination to determine the boundaries of the injection sites and the locations of the retrogradely labeled neurons. The other set was stained with 0.5% cresyl violet, coverslipped, and examined under brightfield illumination to identify the cytoarchitectonic structures and boundaries of nuclei in the brainstem. The boundaries of the subdivisions of the parabrachial nucleus were defined by Fulwiler and Saper Ž1984.. 2.3. Electrophysiological study Neural activity was recorded in the parabrachial region of decerebrate rats, and responsiveness was tested by applying taste stimulation, electrical stimulation of the superior salivatory nucleus, and passive or spontaneous movement of the jaw. At least 3 days after decerebration, the rats were fitted with the intraoral fistula and mounted on the stereotaxic frame, the same as for the behavioral study. The trachea was cannulated and a small burr hole was drilled in the skull for implantation of a stimulating electrode into the left superior salivatory nucleus. A stainless steel coaxial electrode with a tip diameter of 150 mm was inserted into the nucleus; the electrode was oriented 208 to the vertical Žwith the tip anterior. to keep space enough for neural recording. It was fixed to the skull with a screw and dental acrylic resin. Electrical stimuli were square wave pulses of 50-ms duration and 0.2–0.4-mA intensity, with the tip of the electrode cathodal. For recording neural activity, the surface of the cerebellum over the left parabrachial region was exposed by the
same procedure used for electrolytic destruction. Extracellular neuronal activity was recorded using glass micropipettes Žtip diameter of about 2 mm. filled with 2% brilliant blue dissolved in 0.5 M sodium acetate. With the obex as a reference point, microelectrode penetrations were made in mediolateral steps of 0.2 mm in the region of the parabrachial nucleus in rostrocaudal planes running from 4.5 to 5.5 mm rostral to the obex. Neural activity was differentially amplified ŽDAM-6A, WPI. with respect to an indifferent electrode Ža chloridized AgrAgCl plate. positioned on the skull. The amplified potentials were displayed on an oscilloscope ŽTektronix 5110., and stored on magnetic tape. The location of the tip of the micropipette was marked by electrophoretic deposition of brilliant blue Ž5 mA DC current for 5–10 min.. Recording of neural activity started at least 4 h after the surgical operation. During the recording sessions, whenever spontaneous movements appeared, maintenance doses of ketamine were administered intraperitoneally. Rectal temperature was monitored and maintained at 37–388C by an electric heating blanket. Under these experimental conditions, quinine taste stimulation evoked salivary secretion and rhythmical jaw movements; for monitoring these responses, some animals received surgery for salivary tubing and EMG recording from the anterior digastric muscle. Taste stimulation was performed by injection of an aliquot Ž0.5 ml. of taste solution Ž0.1 M NaCl, 0.5 M sucrose, 0.01 M HCl and 10 mM quinine-hydrochloride. into the mouth through the intraoral fistula. After about 30 s, an aliquot of distilled water was injected several times until the discharge rates of neurons returned to their spontaneous rates. At the end of experiments, the stimulation site of the brain
Fig. 1. Submandibular salivary secretion and electromyographic activity in a decerebrate rat. ŽA–D. Each set of recordings shows salivary flow rate ŽSaliva., and electromyographic activities of the masseter ŽMass. and anterior digastric ŽDigast. muscles. Oral infusion Žarrows. of 0.1 M NaCl ŽA., 0.5 M sucrose ŽB., and 0.01 M HCl ŽC. induced licking of solutions and salivation at low flow rates; 10 mM quinine ŽD. induced vigorous salivation and gapes. ŽE. Electromyographic activity of the anterior digastric muscle at a faster time scale showing licks and gapes.
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73
65
was coagulated by passing direct current Ž10 A for 30 s. through the stimulating electrode. The locations of stimulating and recording sites were verified histologically by the same procedure used in the behavioral study. 2.4. Statistics Results ŽFigs. 2, 7 and 11. are expressed as the means " SE. The differences between data ŽFigs. 2 and 7. were tested across animal groups using a repeated measures analysis of variance ŽANOVA. followed by a post hoc Scheffe’s test when justified. Probabilities less than 0.05 were considered significant.
3. Results 3.1. Effects of decerebration on saliÕation and taste rejection behaÕior Submandibular salivary secretion and behavioral reaction in response to taste stimulation were compared in normal and decerebrate Ž3–4 days after decerebration. rats. The behavioral responses differed between normal and decerebrate animals, but salivary secretion was similar between both groups. Sweet Ž0.5 M sucrose., salty Ž0.1 M NaCl., sour Ž0.01 M HCl., and weak bitter Ž0.01 mM quinine. taste stimulation elicited lapping in normal and decerebrate rats, although decerebrate animals showed smaller magnitude jaw and tongue movements than normal. When a strong bitter stimulation Žsay 10 mM quinine. was infused into the mouth, normal rats showed rejection behavior consisting of gaping, face washing, chin rubbing, head shaking, forelimb flailing and paw pushing ŽGrill and Norgren, 1978a., whereas decerebrate rats showed gaping and face washing in all animals Ž n s 9. and chin rubbing and head shaking only in three rats. Forelimb flailing or paw pushing was not observed in decerebrate rats, as reported by Grill and Norgren Ž1978b.. In spite of the behavioral difference, both normal and decerebrate rats secreted 2–5 ml of submandibular saliva on infusion of 0.5 M sucrose, 0.1 M NaCl, and 0.01 M HCl ŽFig. 1A–C, obtained from a decerebrate rat., and vigorous salivation was evoked by 1–10 mM quinine ŽFig. 1D.. The rejection behavior lasted for 3–5 min and gaping was most frequently observed in both normal and decerebrate rats. In EMG activity ŽFig. 1E., rhythmical responses of gapes could be differentiated from those of licks primarily by amplitude and duration of the anterior digastric muscle ŽYamamoto et al., 1982; Travers and Norgren, 1986; Dinardo and Travers, 1994.. When the number of gapes was estimated by counting the rhythmical activity of the anterior digastric muscle, it was found that they increased with the quinine concentration ŽFig. 2A.. Statistical analysis of the concentration–response functions showed no difference between normal Ž n s 10. and decer-
Fig. 2. Number of gapes ŽA. and rate of salivary flow ŽB. for 3 min as a function of quinine concentration in normal and decerebrate rats. Each value was the mean"SE obtained from normal Žopen circles, ns10. and decerebrate Žsolid circles, ns9. rats. No significant difference between the groups was evident in the concentration–response function to number of gapes or salivary flow rate.
ebrate Ž n s 9. rats. The salivary flow rate also increased as a function of quinine concentration in normal and decerebrate rats ŽFig. 2B., and there was no difference between the groups. This result indicates that the fundamental neural structures for the quinine-evoked rejection behavior Žgaping. and salivation were situated in the lower brainstem. 3.2. Efferent connection from the parabrachial region to the superior saliÕatory nucleus Next, we examined the neuroanatomy of the gustatory–salivary reflex arc in the lower brainstem based on the retrograde axonal transport of Fluoro-gold. After injec-
66
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73
Fig. 3. Injection site of Fluoro-gold and labeled cells in the caudal parabrachial region. ŽA. Photomicrograph of injection site stained with cresyl violet, which was taken under brightfield illumination. ŽB. Enlarged photomicrograph of the injection site Žsquare area in A. taken under ultraviolet illumination. ŽC. Labeled neurons scattered in the supratrigeminal area and reticular formation. The photomicrographs of B and C are shown as negative film images for visualizing background structures.
tion of Fluoro-gold into the superior salivatory nucleus ŽFig. 3A,B., we observed the retrogradely labeled neurons in the lower brainstem from the level of the obex to the rostral parabrachial region ŽFig. 4.. The labeled neurons
were not found in the solitary nucleus, the rostral half of which contains the taste relay neurons ŽFig. 4D., and only a few neurons were detected in the parabrachial taste area ŽFig. 4B,b.. In the rostral parabrachial area ŽFig. 4A,a.,
Fig. 4. Distribution of Fluoro-gold-labeled cells and extent of Fluoro-gold injection site Žgrey area in panel C.. Drawings are arranged rostrocaudally from A to E, and panels a and b show the parabrachial areas in A and B, respectively, ipsilateral to the injection site. The arrow in b indicates the waist area. bc s Brachium conjunctivum; g7 s genu facial nerve; Mo5 s trigeminal motor nucleus; NTS s nucleus of the solitary tract; PCRts parvicellular reticular nucleus; Pr5 s principal sensory trigeminal nucleus; py s pyramidal tract; Sp5C s spinal trigeminal nucleus, caudal division; Sp5I s spinal trigeminal nucleus, interpolar division; Sp5Os spinal trigeminal nucleus, oral division; Su5 s supratrigeminal nucleus; 7 s facial nucleus.
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73
most labeled cells were distributed in the pontine reticular formation dorsal to the trigeminal motor nucleus and parabrachial nucleus, except for the external lateral subdivision. At the caudal parabrachial level, labeled cells were mainly distributed in the supratrigeminal nucleus, pontine reticular formation just dorsal to the supratrigeminal nucleus, and external medial subnucleus of the parabrachial nucleus. Many labeled cells were also found in the lateral reticular formation from the level of the obex to the superior salivatory nucleus ŽFig. 4C–E.. Fig. 3C shows that the labeled cells appeared at the level of the parabrachial taste area. It is suggested that the taste relay neurons in the lower brainstem do not directly project to the superior salivatory neurons, and the pontine reticular formation may be included in the taste pathway from the parabrachial taste area to the superior salivatory nucleus. 3.3. Effect of parabrachial lesion on saliÕation and rejection behaÕior On the basis of the anatomical results shown above and of the c-fos immunohistochemical studies described in
67
Section 1 ŽYamamoto et al., 1994; Travers et al., 1999., we made bilateral electrolytic lesions in the parabrachial region in decerebrate rats, and compared differences in the number of gapes and rate of salivary flow. The photomicrographs in Fig. 5 show representative lesions made in the medial ŽA., lateral ŽB. and ventral ŽC. part of the parabrachial region, and Fig. 6 shows the extent of the lesions in all subjects used for the behavioral study Žeight rats for each site.. The solid areas show overlapping lesions in all animals, which centered at the waist ŽA. and lateral ŽB. of the brachium conjunctivum, and the pontine reticular formation between the brachium conjunctivum and trigeminal motor nucleus ŽC.. Fig. 7 shows the quinine concentration–response functions of the parabrachial lesioned and control rats. The control data are from the decerebrate rats in Fig. 2. As to the number of gapes ŽFig. 7A., the parabrachial lesions significantly decreased in gapes at quinine concentrations more than 1 mM ŽScheffe’s test, p - 0.05 following ANOVA: F3,29 s 23.91, P 0.001.. The effects of the lesion appeared most evident in the medial lesioned group; 15.5% of controls at 10 mM quinine. As to the salivary secretion ŽFig. 7B., the lateral
Fig. 5. Representative coronal sections with electrolytic lesions in the parabrachial region. Electrolytic lesions were made in the medial ŽA., lateral ŽB., and ventral ŽC. part of the parabrachial region.
68
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73
Fig. 6. Extent of the lesions in the parabrachial region. The medial ŽA., lateral ŽB., and ventral ŽC. lesions for eight rats are superimposed on camera-lucida drawings of the representative sections taken from the caudal level of the parabrachial nucleus Ž5.0-mm rostral to the obex.. Encircled areas are lesions for each rat Ž ns8., and solid areas are overlapped by all lesions: bc s brachium conjunctivum; Mo5s trigeminal motor nucleus.
lesions did not decrease on salivation at any of the stimulus concentrations ŽScheffe’s test, p ) 0.05., whereas the medial and ventral lesions significantly decreased on salivation at more than 1 mM quinine ŽScheffe’s test, p 0.05.. The medial and ventral lesioned rats showed 10.6% and 22.7% of control salivation, respectively, at 10 mM quinine. Under our experimental conditions, the parabrachial lesioning did not entirely abolish the quinine-induced salivation; the flow rate of the medial lesioned group was 7.2 " 1.2 mlr3 min at 10 mM. Moreover, the parabrachial lesioning affected little, if at all, salivation induced by other modalities of taste stimulation; oral-injection of 0.1 M NaCl, 0.5 M sucrose and 0.01 M HCl evoked 2–5 ml of salivation. These results suggest that the parabrachial region, especially the taste area and pontine reticular formation, are involved in the neural pathway subserving the vigorous salivation during taste rejection.
Fig. 7. Number of gapes ŽA. and rate of salivary flow ŽB. for 3 min as a function of quinine concentration in decerebrate and decerebrate plus parabrachial lesioned rats. Each value was the mean"SE obtained from decerebrate Žsolid circles, ns9., medial-parabrachial Žopen squares, n s 8., lateral-parabrachial Žopen triangles, n s 8., and ventralparabrachial Žopen rhombes, ns8. lesioned rats.
tion, and supratrigeminal nucleus. We defined a response as an increase in impulse frequency during taste or water Table 1 Number of neurons responsive to taste and movement of the jaw in relation to electrical stimulation of the superior salivatory nucleus Natural stimulation
3.4. Electrophysiological study of the parabrachial region Neural activity of the parabrachial region was examined by monitoring submandibular salivary secretion in lightly anesthetized decerebrate rats. Under the experimental conditions, we could record 48 neurons responsive to taste and water stimulation, and 35 responsive to jaw movements from the parabrachial nucleus, surrounding reticular forma-
Taste NaCl-best Sucrose-best HCl-best Quinine-best Water-best Movement of the jaw
Antidromic stimulation Žq.
Antidromic stimulation Žy.
0 0 2 8 0 8
15 3 13 3 4 27
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73
69
Fig. 8. Discharge pattern of a taste responsive and antidromically activated neuron recorded from the pontine reticular formation. ŽA. Taste stimuli were applied at the point indicated by the arrow, and quinine induced the best response. ŽBa. Spontaneous orthodromic impulse Žopen circle. and antidromic impulse Žsolid circle. evoked by electrical stimulation of the superior salivatory nucleus Žarrow.. ŽBb. The spontaneous and evoked impulses collided at shorter intervals between the spontaneous impulse and electrical stimulus.
stimulation greater than the mean " 2SD of spontaneous activity. Responses to jaw movement were brisk impulse discharges synchronized with passive jaw stretches, or
with spontaneous rhythmical jaw openings. Of 83 neurons recorded, 18 antidromically responded to electrical stimulation of the superior salivatory nucleus, suggesting they
Fig. 9. Distribution of neurons responsive to taste stimulation ŽA. and jaw movements ŽB.. Solid circles indicate neurons which did not respond to electrical stimulation of the superior salivatory nucleus. Open circles indicate neurons antidromically activated by the electrical stimulation.
70
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73
Fig. 10. Histologically confirmed recording sites by electrophoretic deposition of brilliant blue. This preparation contains recording sites of non-antidromic Žarrow 1. and antidromic Žarrow 2. taste responsive neurons.
project to the nucleus. Antidromicity was tested by the collision test and invariant impulse latency to electrical stimuli. Table 1 summarizes the number of neurons responsive to taste and jaw movements in relation to antidromic stimulation. Among the taste responsive neurons, the antidromic neurons responded best to quinine Žeight neurons. or HCl Žtwo neurons. ŽTable 1 and Fig. 8.. Such neurons were mainly located in the pontine reticular formation just dorsal to the supratrigeminal area ŽFig. 9A..
The non-antidromic neurons were distributed in the parabrachial taste area, which responded well to one of the four qualities of taste and water. A photomicrograph in Fig. 10 shows recording sites of the antidromic and nonantidromic taste responsive neurons. Both the antidromic and non-antidromic taste responsive neurons showed a tonic discharge pattern irrespective of movement of the jaw. The average impulse frequency was lower in the antidromic neurons than the non-antidromic neurons ŽFig. 11.. We could not record taste responsive neurons from the external lateral subdivision or rostral part of the parabrachial nucleus. The jaw-movement responsive neurons were mainly distributed in the supratrigeminal area, and the antidromic and non-antidromic neurons were intermingled ŽFig. 9B.. These results suggest that the pontine reticular formation is related to the gustatory–salivary reflex evoked by an aversive taste stimulus.
4. Discussion
Fig. 11. Mean response rate for the taste responsive neurons activated Žsolid columns. and non-activated Žopen columns. by antidromic stimulation of the superior salivatory nucleus. The response rate is spikes per s"SE measured over 10 s. Ss 0.5 M sucrose; Ns 0.1 M NaCl; H s 0.01 M HCl; Qs10 mM quinine; Wsdistilled water; SP sspontaneous discharge rate.
The taste rejection behavior of rats consists of various stereotyped oro-facial and body reactions. Among the reactions, gapes are rhythmical jaw and tongue movements, which can be quantified by calculating the magnitude, duration, and number of muscle contractions monitored by electromyographs ŽYamamoto et al., 1982; Travers and Norgren, 1986; Dinardo and Travers, 1994.. Decerebrate animals, at more than 10 days after decerebration, showed similar gapes in terms of number and duration of burst
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73
myographic activities of jaw muscles ŽGrill and Norgren, 1978b; Travers et al., 1999.. In the present study, using day two to three decerebrated rats to prevent atrophy of salivary glands, gaping, as well as profuse salivation similar to normal rats were observed. These results suggest that the fundamental neural structures for the quinine-evoked motor and autonomic reactions are situated in the lower brainstem. The gaping and salivation occurred at the same time in normal and decerebrate animals, but when the lateral part of the parabrachial nucleus was destroyed ŽFig. 5., the number of gapes was reduced without a marked effect on salivation. This separation of the reactions suggests that the salivation is evoked mainly by bitter taste stimulation, not by oral sensory inputs due to movement of the oral structures. The lower brainstem has the first- and second-order taste relay neurons, the solitary and parabrachial taste area, respectively, both of which are involved in the neural mechanism of salivation. Electrical stimulation of the solitary nucleus, and electrical and chemical stimulation of the parabrachial nucleus, produces salivary secretion. Moreover, a neuroanatomical study using the transneuronal transport of a virus showed that the parasympathetic nerve of rat submandibular gland polysynaptically connects with the solitary and parabrachial taste areas ŽJansen et al., 1992.. These previous findings suggest that taste information via both taste relay areas activates salivary secretion. The present lesioning study revealed that the parabrachial route is more crucial for the quinine-evoked vigorous salivation than the solitary route. Salivation after destruction of the parabrachial taste area, which is thought to be activated via the solitary route, had a much lower flow rate. However, the flow rates of the small salivation correspond to those when rats licked palatable taste solutions when conscious ŽMatsuo et al., 1994.. Therefore, it is conceivable that the parabrachial route is activated only during the taste rejection behavior. On the basis of the following present results, it is concluded that a feature of the parabrachial route is that bitter taste information from the parabrachial taste area is selectively inputed to the superior salivatory nucleus through the pontine reticular formation ventral to the parabrachial nucleus: Ž1. salivation during rejection was markedly reduced by lesioning of the parabrachial taste area or its ventral pontine reticular formation; Ž2. the Fluoro-gold labeled cells that may directly input to the superior salivatory nucleus were dominantly found in the pontine reticular formation and supratrigeminal area, not in the parabrachial taste area; Ž3. neurons in the pontine reticular formation, which may have a monosynaptic input to the superior salivatory nucleus, responded well to bitter taste stimulation. Our proposed parabrachial–salivary route contains the pontine reticular formation just dorsal to the supratrigeminal nucleus, capping the trigeminal motor nucleus dorsally. The present Fluoro-gold study showed the labeled neurons
71
in the supratrigeminal nucleus and its dorsal reticular formation, but not in the parabrachial taste area consisting of the ventral lateral and medial subnuclei. This result is very similar to that obtained by Herbert et al. Ž1990. after injection of horseradish peroxidase into the parvicellular reticular area just rostral to the solitary nucleus, which possibly includes the inferior salivatory nucleus. These anatomical studies suggest that the supratrigeminal nucleus and pontine reticular formation have projections to the salivatory nuclei. However, the present electrophysiological study showed that only the pontine reticular formation contains taste-responsive neurons innervating the superior salivatory nucleus. On the other hand, the supratrigeminal neurons were responsive to movement of the jaw, but not to taste stimulation. This nucleus receives inputs from trigeminal mesencephalic neurons arising from muscle spindles in the jaw-closing muscles Že.g. Fay and Norgren, 1997. and oral mechanoreceptors Že.g., Bae et al., 1997.. It is conceivable that the supratrigeminal nucleus is related to the so-called masticatory–salivary reflex. The present results imply the existence of a short circuit from the parabrachial taste area to the pontine reticular formation, and in this circuit, bitter and sour taste information of the parabrachial taste area was selectively conducted to the reticular formation in a tonic firing pattern Ževen when the jaw rhythmically moved. at a low frequency of about 7 Hz. Our observed impulse frequency of quinine response in the parabrachial taste area is similar to that observed by Di Lorenzo Ž1988. in decerebrate anesthetized rats. She also observed that decerebration decreased the impulse frequency of NaCl and HCl responses, but not that of the quinine response. Thus, the short circuit may have a synaptic connection to make a tonic firing pattern at low frequency. We speculate that such a low firing frequency is enough to evoke vigorous salivation, because electrical stimulation of the parasympathetic preganglionic nerve Žaxons of the superior salivatory neurons. at about 10 Hz evoked submandibular salivary secretion, whose flow rate was similar to that observed during taste rejection behavior ŽMatsuo 1999a,b.. Recently, an immunohistochemical technique to detect the expression of the Fos protein product of the immediate early gene c-fos has been applied to examine activated neurons in the central nervous system. The c-fos expression was examined in the brainstem after taste rejection behavior evoked by quinine stimulation in both normal ŽHarrer and Travers, 1996; DiNardo and Travers, 1997; Yamamoto and Sawa, 2000a,b. and decerebrate rats ŽTravers et al., 1999.. However, the authors of these studies make no mention of quinine-evoked c-fos expression in the superior salivatory nucleus, pontine reticular formation, or supratrigeminal nucleus. It is likely that this method is not sensitive enough to detect neurons subserving reflex salivation, partly because of their low firing activity. In respect to oro-facial movements, when taste solutions were infused into the mouth, c-fos was exten-
72
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73
sively expressed in the lateral and intermediate zone of the medullary reticular formation more caudal to the superior salivatory nucleus, which includes premotor neurons of masticatory, facial and lingual motor systems ŽDiNardo and Travers, 1997; Travers et al., 1997.. This area overlaps sites labeled with Fluoro-gold ŽFig. 7D,E., and receives inputs from the taste relay areas of the solitary nucleus ŽShammah-Lagnado et al., 1992; Halsell et al., 1996. and parabrachial nucleus ŽHerbert et al., 1990; Karimnamazi and Travers, 1998.. These findings suggest that the medullary reticular formation conducts taste- and motor-related information to the superior salivatory nucleus. Thus, there is a possibility that the medullary reticular formation is involved in the solitary route of the gustatory–salivary reflex. It is also speculated that the medullary reticular formation has a role in blood flow changes in the tongue associated with oro-facial movements andror taste stimulation; the superior salivatory nucleus also contains neurons innervating the parasympathetic ganglia in the tongue and regulates blood flow of the tongue ŽYu and Srinivasan, 1980; Matsuo and Kang, 1998.. In conclusion, we showed that the quinine-evoked salivation during rejection behavior is controlled by a neural axis of the parabrachial taste area—pontine reticular formation—salivatory nucleus. This salivary reflex arc is susceptible to anesthesia, and the pontine reticular formation showed a low firing frequency. It may be due to these characteristics, that neither electrophysiological taste response nor c-fos expression has been detected in the pontine reticular formation in previous studies.
Acknowledgements We thank to Profs. Y. Shigenaga and T. Yamamoto for discussion of the data. This study was supported by Grants-in-Aid for Scientific Research ŽNo. 11470391. from the Ministry of Education, Science and Culture of Japan.
References Bae, Y.C., Park, K.P., Yoshida, A., Nakagawa, S., Kurata, S., Chen, K., Takemura, M., Shigenaga, Y., 1997. Identification of gamma-aminobutyric acid-immunoreactive axon endings associated with mesencephalic periodontal afferent terminals and morphometry of the two types of terminals in the cat supratrigeminal nucleus. J. Comp. Neurol. 389, 127–138. Dick, T.E., Haxhiu, M.A., Cherniack, N.S., 1992. Salivary secretion elicited by activation of the parabrachial nuclei in the cat. J. Auton. Nerv. Syst. 39, 19–28. Di Lorenzo, P.M., 1988. Taste responses in the parabrachial pons of decerebrate rats. J. Neurophysiol. 59, 1871–1887. DiNardo, L.A., Travers, J.B., 1994. Hypoglossal neural activity during ingestion and rejection in the awake rat. J. Neurophysiol. 72, 1181– 1191. DiNardo, L.A., Travers, J.B., 1997. Distribution of Fos-like immunoreactivity in the medullary reticular formation of the rat after gusta-
tory elicited ingestion and rejection behaviors. J. Neurosci. 17, 3826– 3839. Fay, R.A., Norgren, R., 1997. Identification of rat brainstem multisynaptic connections to the oral motor nuclei using pseudorabies virus: I. Masticatory muscle motor systems. Brain Res. Rev. 25, 255–272. Flynn, F.W., Grill, H.J., Schwartz, G.J., Norgren, R., 1991. Central gustatory lesions: I. Preference and taste reactivity tests. Behav. Neurosci. 105, 933–943. Fulwiler, C.E., Saper, C.B., 1984. Subnuclear organization of the efferent connections of the parabrachial nucleus in the rat. Brain Res. Rev. 7, 229–259. Grill, H.J., Norgren, R., 1978a. The taste reactivity test: I. Mimetic responses to gustatory stimuli in neurologically normal rats. Brain Res. 143, 263–279. Grill, H.J., Norgren, R., 1978b. The taste reactivity test: II. Mimetic responses to gustatory stimuli in chronic thalamic and chronic decerebrate rats. Brain Res. 143, 281–297. Halsell, C.B., Travers, S.P., Travers, J.B., 1996. Ascending and descending projections from the rostral nucleus of the solitary tract originate from separate neuronal populations. Neuroscience 72, 185–197. Harrer, M.I., Travers, S.P., 1996. Topographic organization of Fos-like immunoreactivity in the rostral nucleus of the solitary tract evoked by gustatory stimulation with sucrose and quinine. Brain Res. 711, 125–137. Herbert, H., Moga, M.M., Saper, C.B., 1990. Connections of the parabrachial nucleus with the nucleus of the solitary tract and the medullary reticular formation in the rat. J. Comp. Neurol. 293, 540–580. Hill, D.L., Almli, C.R., 1983. Parabrachial nuclei damage in infant rats produces residual deficits in gustatory preferencesraversions and sodium appetite. Dev. Psychobiol. 16, 519–533. Jansen, A.S.P., Ter Horst, G.J., Mettenleiter, T.C., Loewy, A.D., 1992. CNS cell groups projecting to the submandibular parasympathetic preganglionic neurons in the rat: a retrograde transneuronal viral cell body labeling study. Brain Res. 572, 253–260. Karimnamazi, H., Travers, J.B., 1998. Differential projections from gustatory responsive regions of the parabrachial nucleus to the medulla and forebrain. Brain Res. 813, 283–302. Matsuo, R., 1999a. Interrelation of taste and saliva. In: Garrett, J.R., Ekstrom, ¨ J., Anderson, L.C. ŽEds.., Neural Mechanisms of Salivary Gland Secretion. Karger, Basel, pp. 185–195. Matsuo, R., 1999b. Discharge pattern of the superior salivatory nucleus in rats: in vivo and in vitro studies. In: Nakamura, Y., Sessle, B.J. ŽEds.., Neurobiology of Mastication. Elsevier, Amsterdam, pp. 139– 148. Matsuo, R., Kang, Y., 1998. Two types of parasympathetic preganglionic neurones in the superior salivatory nucleus characterized electrophysiologically in slice preparations of neonatal rats. J. Physiol. 513, 157–170. Matsuo, R., Yamamoto, T., Kiyomitsu, Y., Morimoto, T., 1989. Neural substrates for reflex salivation induced by taste, mechanical, and thermal stimulation of the oral region in decerebrate rats. Jpn. J. Physiol. 39, 349–357. Matsuo, R., Yamamoto, T., Ikehara, A., Nakamura, O., 1994. Effect of salivation on neural taste responses in freely moving rats: analyses of salivary secretion and taste responses of the chorda tympani nerve. Brain Res. 649, 136–146. Nilsson, B.-O., Rosengren, E., Ekstrom, ¨ J., 1990. Polyamines and longterm disuse of rat parotid glands. Acta Physiol. Scand. 140, 105–109. Paxinos, G., Watson, C., 1986. The Rat Brain in Stereotaxic Coordinates. Academic Press, New York. Phillips, M.I., Norgren, R.E., 1970. A rapid method for permanent implantation of an intraoral fistula in rats. Behav. Res. Methods Instr. 2, 124. Scott, J., Gunn, D.L., 1991. A comparative quantitative histological investigation of atrophic changes in the major salivary glands of liquid-fed rats. Arch. Oral Biol. 36, 855–857.
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73 Shammah-Lagnado, S.J., Costa, M.S.M.O., Ricardo, J.A., 1992. Afferent connections of the parvocellular reticular formation: a horseradish peroxidase study in the rat. Neuroscience 50, 403–425. Spector, A.C., 1995. Gustatory parabrachial lesions disrupt taste-guided quinine responsiveness in rats. Behav. Neurosci. 109, 79–90. Travers, J.B., Norgren, R., 1986. Electromyographic analysis of the ingestion and rejection of sapid stimuli in the rat. Behav. Neurosci. 100, 544–555. Travers, J.B., DiNardo, L.A., Karimnamazi, H., 1997. Motor and premotor mechanisms of licking. Neurosci. Biobehav. Rev. 21, 631–647. Travers, J.B., Urbanek, K., Grill, H.J., 1999. Fos-like immunoreactivity in the brain stem following oral quinine stimulation in decerebrate rats. Am. J. Physiol. 277, R384–R394. Yamamoto, T., Sawa, K., 2000a. c-Fos-like immunoreactivity in the
73
brainstem following gastric loads of various chemical solutions in rats. Brain Res. 866, 135–143. Yamamoto, T., Sawa, K., 2000b. Comparison of c-Fos-like immunoreactivity in the brainstem following intraoral and intragastric infusions of chemical solutions in rats. Brain Res. 866, 144–151. Yamamoto, T., Matsuo, R., Fujiwara, T., Kawamura, Y., 1982. EMG activities of masticatory muscles during licking in rats. Physiol. Behav. 29, 905–913. Yamamoto, T., Shimura, T., Sakai, N., Ozaki, N., 1994. Representation of hedonics and quality of taste stimuli in the parabrachial nucleus of the rat. Physiol. Behav. 56, 1197–1202. Yu, W.H., Srinivasan, R., 1980. Origin of the preganglionic visceral efferent fibres to the glands in the rat tongue as demonstrated by the horseradish peroxidase method. Neurosci. Lett. 19, 143–148.
Role of parabrachial nucleus in submandibular salivary secretion induced by bitter taste stimulation in rats Ryuji Matsuo a,) , Yoji Yamauchi a , Motoi Kobashi a , Makoto Funahashi a , Yoshihiro Mitoh a , Akira Adachi b b
a Department of Oral Physiology, Okayama UniÕersity Dental School, 2-5-1 Shikata-cho, Okayama 700-8525, Japan Department of Clinical Welfare SerÕice, Kyushu UniÕersity of Health and Welfare, Nobeoka, Miyazaki 882-8508, Japan
Received 11 December 2000; received in revised form 26 January 2001; accepted 29 January 2001
Abstract When rats lick a bitter taste solution such as quinine-hydrochloride, they secrete profuse amounts of saliva. The salivation has a higher flow rate than that induced by other qualities of taste stimulation: sweet, salty, and sour. The present study is aimed to clarify the neural mechanism of the quinine-evoked salivation by means of behavioral, neuroanatomical, and electrophysiological experiments. Behaviorally, submandibular salivary secretion and rejection behavior Žgaping. were observed in normal rats, as well as in rats chronically decerebrated at the precollicular level. In chronically decerebrate rats, these quinine-evoked reactions were strongly suppressed by destruction of the medial part of the parabrachial nucleus, including the so-called taste area, and ventral part of the parabrachial nucleus, including the pontine reticular formation. Neuroanatomical study using a retrograde tracer, Fluoro-gold, revealed that the neurons sending their axons to the superior salivatory nucleus, parasympathetic secretory center, were located mainly in the pontine reticular formation ventral to the parabrachial nucleus, not in the parabrachial taste area. Extracellular neural activity was recorded from the parabrachial region in decerebrate rats, and responsiveness to taste stimulation, jaw movements, and electrical stimulation of the superior salivatory nucleus was examined. Neurons responsive to both taste stimulation and antidromic stimulation of the superior salivatory nucleus were found in the pontine reticular formation ventral to the parabrachial nucleus, which responded well to quinine and HCl taste stimuli. Neurons in the parabrachial taste area could respond to four qualities of taste stimulation, but not to antidromic stimulation of the salivary center. These results suggest that aversive taste information from the parabrachial taste area reaches the salivary secretory center via the reticular formation ventral to the parabrachial nucleus. q 2001 Elsevier Science B.V. All rights reserved. Keywords: Salivary secretion; Taste; Rejection; Parabrachial nucleus; Salivatory nucleus
1. Introduction When rats lick a bitter taste substance, e.g., quinine-hydrochloride, they exhibit stereotypic rejection behavior consisting of gaping, face washing, chin rubbing, head shaking, forelimb flailing and paw pushing ŽGrill and Norgren, 1978a.. Matsuo et al. Ž1994. have found that the taste rejection behavior is always accompanied by vigorous salivation, about 30 mlrmin from the unilateral submandibular gland, which is a far higher flow rate than that evoked by licking of a sweet, salty or sour taste solution Žless than about 5 mlrmin.. It is well recognized that the
) Corresponding author. Tel.: q81-86-235-6640; fax: q81-86-2356644. E-mail address: [email protected] ŽR. Matsuo..
gustatory–salivary reflex Že.g., Matsuo, 1999a. and taste rejection behavior occur in decerebrate animals ŽGrill and Norgren, 1978b; Travers et al., 1999., and the fundamental neural structures for the reactions are situated in the lower brainstem. However, precise neural mechanisms underlying the quinine-evoked salivation have not been elucidated yet. One of the problems in studies on the reflex salivation is effects of anesthesia. When anesthetized, decerebrate animals do not profusely secrete saliva on bitter taste stimulation, and secrete only a little saliva on various kinds of taste stimulation Žless than 6 mlrmin from rat submandibular gland. Že.g., Matsuo et al., 1989.. Such relatively weak salivation is abolished by destruction of the rostral part of the solitary nucleus Žthe first order of the taste relay area., but not affected by destruction of the parabrachial taste area Žthe second order of the taste relay area.. This
1566-0702r01r$ - see front matter q 2001 Elsevier Science B.V. All rights reserved. PII: S 1 5 6 6 - 0 7 0 2 Ž 0 1 . 0 0 2 3 4 - X
62
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73
result suggests that the reflex salivation is evoked mainly by taste information mediated through the solitary nucleus. Nevertheless, electrical and chemical stimulation of the parabrachial nucleus induced salivary secretion in anesthetized animals ŽMatsuo et al., 1989; Dick et al., 1992.. From this finding, we speculate that there is a gustatory– salivary reflex arc which involves the parabrachial taste area, but this reflex arc may be susceptible to anesthesia partly because the parabrachial-mediated reflex arc is more polysynaptic than the solitary-mediated one. Moreover, there is a possibility that the parabrachial-mediated reflex arc is responsible for the vigorous salivation during taste rejection, since rats with lesions in the parabrachial nucleus generally displayed smaller aversive responses to bitter taste stimulation in reference to gapes, chin rubs, and intake of bitter taste solution ŽHill and Almli, 1983; Flynn et al., 1991; Spector, 1995.. To test this possibility, it is necessary to examine salivation using decerebrate animals while minimizing the effects of anesthesia. Considering the above-mentioned findings and possibilities, we attempted to investigate the role of the lower brainstem, especially the parabrachial nucleus, in salivation and oromotor reactions evoked by aversive taste stimulation. To this end, we conducted behavioral, neuroanatomical, and electrophysiological experiments. First, we measured submandibular salivary secretion in chronically decerebrate rats and confirmed that vigorous salivation occurs during taste rejection. Second, the neurons sending their axons from the parabrachial region to the superior salivatory nucleus Žthe parasympathetic secretory center. were examined by using a retrograde tracer, Fluoro-gold. Third, with the results of the above neuroanatomical experiment in mind, we examined changes in the rejection behavior and salivation after lesioning of the medial part of the parabrachial region, including the taste relay area, the pontine reticular formation and supra trigeminal area, which project to the superior salivatory nucleus Žrevealed by the present study., and the lateral part of the parabrachial region, which receives general visceral inputs ŽHerbert et al., 1990. and relates to the rejection response to bitter taste ŽYamamoto et al., 1994; Travers et al., 1999.. Fourth, neural responses to taste stimulation, jaw movements, and electrical stimulation of the superior salivatory nucleus were analyzed in the parabrachial region. Through these experiments, we show that the taste rejection behavior is accompanied by salivation, and the aversive taste information related to salivation reaches the salivary center via the parabrachial taste area and pontine reticular formation ventral to the parabrachial nucleus.
2. Materials and methods Sixty-eight adult male Wistar rats weighing 280–350 g ŽCharles River Breeders, Osaka, Japan. were used; 46 for the behavioral study on submandibular salivary secretion
and rejection behavior, five for the neuroanatomical study using Fluoro-gold and 17 for the electrophysiological study on the parabrachial neurons. Animal protocols were in accordance with the Guiding Principles for the Care and Use of Animals, approved by the Council of the Physiological Society of Japan. 2.1. BehaÕioral study Submandibular salivary secretion and jaw muscle activity were recorded from decerebrate Ž n s 9., decerebrate plus parabrachial lesioned Ž n s 26., and sham-operated rats Ž n s 10.. Since decerebrate animals are aphagic and adipsic, Grill and Norgren Ž1978b. maintained decerebrate rats by tube feeding of a liquid diet, and observed variations in behavioral and oro-facial responses to taste stimulation dependent on the postsurgical period. In the present study, decerebration and the sham-operation were performed 3–4 days before the recording of behavioral responses, because the feeding of a liquid diet for several days results in atrophy of the salivary glands ŽNilsson et al., 1990; Scott and Gunn, 1991.. For the surgical operation, the rats were deeply anesthetized with an i.p. injection of sodium pentobarbitone Ž50–75 mgrkg. and prophylactically treated with a penicillin suspension Ž30,000 units, i.m... They were mounted on a stereotaxic apparatus, and the skull was exposed through a midline incision leveled between bregma and lambda. A small slit was drilled in the frontal plane of the skull and the brainstem was cut at the precollicular level with an L-shaped spatula ŽGrill and Norgren, 1978b.. Ten rats were used for a control group. Their skulls were drilled and the surface of the brain was exposed, but the brainstem was not cut. After the surgical operation, animals were returned to their home cage. The control rats were given laboratory food pellets and water. The decerebrate rats were fed 5–10 ml of liquid diet ŽControl Diet, Oriental, Japan. by gavage three times a day Žbetween 9 AM and 7 PM.. At least 2 days after decerebration, 26 rats were anesthetized as described and received bilateral electrolytic lesions in the medial part, lateral part or ventral part of the parabrachial region. Each electrolytic lesion was produced by four to six electrolytic burns by passing 0.3 mA of anodal current for 5 s through a glass-coated elgiloy wire with 50 mm exposed at the tip Žthe tip diameter was about 15 mm.. A burr hole was drilled in the skull over the parabrachial region, the exposed dura was reflected, and the transverse sinus was hocked anteriorly to avoid bleeding. The obex was exposed, and the electrode was introduced into the medial part Ž5.0 mm anterior to the obex, 1.6 and 1.9 mm lateral to midline, 5.0 and 5.5 mm below the brain surface., lateral part Ž5.0 mm anterior to the obex, 2.0 and 2.4 mm lateral to midline, 5.0 and 5.5 mm below the brain surface. and ventral part Ž5.0 mm anterior to the obex, 1.6, 2.0 and 2.4 mm lateral to midline, 6.0 and 6.5 mm below the brain surface. of the parabrachial re-
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73
gion, which stereotaxic coordinates were taken from the atlas of Paxinos and Watson Ž1986.. Following electrolytic burns, the holes of skull were packed with Gelfoam. For recording salivation, the duct of the left submandibular gland was exposed near its orifice at the floor of the mouth through an incision made at the corner of the mouth. The duct was cannulated with polyethylene tubing Ž3-mm-long. of the largest possible bore. The tubing was fixed to the duct and the dorsal edge of the mandibular bone, between incisor and molar teeth, with a butyl-2cyanoacrylate adhesive and dental acrylic resin. The free end of the tubing was connected to a second polyethylene tube Ž0.8-mm inner diameter, 1.1-mm outer diameter, about 7-cm-long. that was bent subcutaneously backwards over the masseter and temporalis muscles, and led out through an incision made on the top of the skull. For oral injection of taste solution, an intraoral fistula was made on the ipsilateral side of the salivary tubing by a similar method to that described by Phillips and Norgren Ž1970.. Briefly, a curved surgical needle was inserted in the buccal mucosa lateral to the upper molar teeth and out through the incision in the skull. The needle was used to guide polyethylene tubing Ž0.8-mm inner diameter, 1.1-mm outer diameter, about 7-cm-long.. The tubing was slipped through the needle, and then the needle was removed. The salivary and intraoral tubings were cemented to the skull with dental acrylic resin. For recording electromyographic ŽEMG. activity of the jaw-closing and -opening muscles, a pair of Teflon-coated stainless steel electrodes Ž50-mm diameter, about 3-mm interpolar distance. with bared ends Ž1 mm. were inserted into the left masseter and anterior digastric muscles, respectively. The EMG electrodes were passed under the skin to a connector on the skull. The connector was also cemented to the skull with dental acrylic resin. The recording experiment was started after recovery from anesthesia Ž24–36 h after the surgical operation.. The animals were moved to a test box Ža clear plastic cylinder 30 cm in diameter and 35-cm high.. The implanted salivary tubing close to the skull was connected to tubing of the same bore Žfilled with saline., and led out the test box. The other end of the tube was connected to a small bottle via a clamp, and then to a pressure-sensitive transducer ŽMPU-0.5 A, Nihon Khoden, Japan.. The flow of saliva was measured as an increase in liquid pressure in the bottle using the transducer, and this was recorded on a pen-recorder. Since the inner space from the salivary duct to the bottle was closed up, increased pressure was released by opening the clamp. When measuring salivary flow, if the volume of saliva reached 13 ml, the clamp was opened to release the pressure, and closed again for the succeeding recording. The system was calibrated by measuring the deflection of the pen induced by injection of a set amount of saline Ž10 ml. into the bottle. The EMG electrodes were connected to a slip-ring connector ŽAPCL-15, Air Precision. via a mating plug and shielded cables. From the
63
connector, EMG activity was fed to a bandpass amplifier Ž30–3000 Hz., monitored by an oscilloscope. The EMG activity and salivary flow data was stored on magnetic tape. For evoking salivation, an aliquot Ž0.1 ml. of taste solution was injected into the mouth through the intraoral fistula. The taste stimuli consisted of 0.1 M NaCl, 0.5 M sucrose, 0.01 M HCl and 0.01–10 mM quinine-hydrochloride. The stimulations were repeated twice, and the order of applying the stimuli was varied each time, and in different animals. At least 15-min rest was taken after termination of the previous salivary and EMG responses, and two water rinses preceded and followed each stimulus. All the chemical stimuli and distilled water were applied at room temperature Ž23–268C.. The onset of taste stimulation and any other obvious behavior such as grooming or taste rejection was noted on magnetic tape by voice commentary. After the recording, the animals which received electrolytic lesions of the parabrachial region were killed with an overdose of sodium pentobarbitone Ž) 100 mgrkg i.p.. and perfused intracardially with isotonic saline and 10% buffered paraformaldehyde. The brains were removed, stored in 10% paraformaldehyde containing 30% sucrose for a few days, then cut in 50 mm coronal sections on a freezing microtome. The sections were mounted onto chrome alum-coated slides and stained with 0.5% cresyl violet. 2.2. Neuroanatomical study with Fluoro-gold For injection of the retrograde tracer Fluoro-gold into the superior salivatory nucleus, the rats were deeply anesthetized, prophylactically treated, and placed in a stereotaxic apparatus as for decerebration. Stereotaxic coordinates for the superior salivatory nucleus were 1.5 mm lateral- and 4.2 mm anterior to the obex and 7.5–7.7 mm ventral to the brain surface. A small burr hole was made in the skull above the nucleus, and an aliquot Ž0.2 ml. of a 1% solution of Fluoro-gold ŽFluorochrome, Englewood, USA. dissolved in saline was injected through a glass micropipette with a tip diameter of 30 mm connected to a 5-ml Hamilton microsyringe. The pipette was left in the brain for 10 min, to prevent tracer leakage along the injection tract. The hole of the skull was packed with Gelfoam and the incision was sutured. Three to five days after the injection, rats were deeply anesthetized as previously, and transcardially perfused with 100 ml of saline followed by 500 ml of a fixative containing 4% paraformaldehyde in 0.1M phosphate buffer ŽpH 7.4.. The brainstems were removed, further fixed in the same fixative for 90 min, and stored in a 20% sucrose solution Ž0.1 M phosphate buffer, pH 7.4. overnight at 48C. Subsequently, the brainstems were cut transversely, 50-mm-thick, on a freezing microtome and mounted onto chrome alum-coated slides; the sections were divided into
64
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73
two series of alternate serial sections. One set was air-dried overnight, coverslipped, and examined under ultraviolet illumination to determine the boundaries of the injection sites and the locations of the retrogradely labeled neurons. The other set was stained with 0.5% cresyl violet, coverslipped, and examined under brightfield illumination to identify the cytoarchitectonic structures and boundaries of nuclei in the brainstem. The boundaries of the subdivisions of the parabrachial nucleus were defined by Fulwiler and Saper Ž1984.. 2.3. Electrophysiological study Neural activity was recorded in the parabrachial region of decerebrate rats, and responsiveness was tested by applying taste stimulation, electrical stimulation of the superior salivatory nucleus, and passive or spontaneous movement of the jaw. At least 3 days after decerebration, the rats were fitted with the intraoral fistula and mounted on the stereotaxic frame, the same as for the behavioral study. The trachea was cannulated and a small burr hole was drilled in the skull for implantation of a stimulating electrode into the left superior salivatory nucleus. A stainless steel coaxial electrode with a tip diameter of 150 mm was inserted into the nucleus; the electrode was oriented 208 to the vertical Žwith the tip anterior. to keep space enough for neural recording. It was fixed to the skull with a screw and dental acrylic resin. Electrical stimuli were square wave pulses of 50-ms duration and 0.2–0.4-mA intensity, with the tip of the electrode cathodal. For recording neural activity, the surface of the cerebellum over the left parabrachial region was exposed by the
same procedure used for electrolytic destruction. Extracellular neuronal activity was recorded using glass micropipettes Žtip diameter of about 2 mm. filled with 2% brilliant blue dissolved in 0.5 M sodium acetate. With the obex as a reference point, microelectrode penetrations were made in mediolateral steps of 0.2 mm in the region of the parabrachial nucleus in rostrocaudal planes running from 4.5 to 5.5 mm rostral to the obex. Neural activity was differentially amplified ŽDAM-6A, WPI. with respect to an indifferent electrode Ža chloridized AgrAgCl plate. positioned on the skull. The amplified potentials were displayed on an oscilloscope ŽTektronix 5110., and stored on magnetic tape. The location of the tip of the micropipette was marked by electrophoretic deposition of brilliant blue Ž5 mA DC current for 5–10 min.. Recording of neural activity started at least 4 h after the surgical operation. During the recording sessions, whenever spontaneous movements appeared, maintenance doses of ketamine were administered intraperitoneally. Rectal temperature was monitored and maintained at 37–388C by an electric heating blanket. Under these experimental conditions, quinine taste stimulation evoked salivary secretion and rhythmical jaw movements; for monitoring these responses, some animals received surgery for salivary tubing and EMG recording from the anterior digastric muscle. Taste stimulation was performed by injection of an aliquot Ž0.5 ml. of taste solution Ž0.1 M NaCl, 0.5 M sucrose, 0.01 M HCl and 10 mM quinine-hydrochloride. into the mouth through the intraoral fistula. After about 30 s, an aliquot of distilled water was injected several times until the discharge rates of neurons returned to their spontaneous rates. At the end of experiments, the stimulation site of the brain
Fig. 1. Submandibular salivary secretion and electromyographic activity in a decerebrate rat. ŽA–D. Each set of recordings shows salivary flow rate ŽSaliva., and electromyographic activities of the masseter ŽMass. and anterior digastric ŽDigast. muscles. Oral infusion Žarrows. of 0.1 M NaCl ŽA., 0.5 M sucrose ŽB., and 0.01 M HCl ŽC. induced licking of solutions and salivation at low flow rates; 10 mM quinine ŽD. induced vigorous salivation and gapes. ŽE. Electromyographic activity of the anterior digastric muscle at a faster time scale showing licks and gapes.
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73
65
was coagulated by passing direct current Ž10 A for 30 s. through the stimulating electrode. The locations of stimulating and recording sites were verified histologically by the same procedure used in the behavioral study. 2.4. Statistics Results ŽFigs. 2, 7 and 11. are expressed as the means " SE. The differences between data ŽFigs. 2 and 7. were tested across animal groups using a repeated measures analysis of variance ŽANOVA. followed by a post hoc Scheffe’s test when justified. Probabilities less than 0.05 were considered significant.
3. Results 3.1. Effects of decerebration on saliÕation and taste rejection behaÕior Submandibular salivary secretion and behavioral reaction in response to taste stimulation were compared in normal and decerebrate Ž3–4 days after decerebration. rats. The behavioral responses differed between normal and decerebrate animals, but salivary secretion was similar between both groups. Sweet Ž0.5 M sucrose., salty Ž0.1 M NaCl., sour Ž0.01 M HCl., and weak bitter Ž0.01 mM quinine. taste stimulation elicited lapping in normal and decerebrate rats, although decerebrate animals showed smaller magnitude jaw and tongue movements than normal. When a strong bitter stimulation Žsay 10 mM quinine. was infused into the mouth, normal rats showed rejection behavior consisting of gaping, face washing, chin rubbing, head shaking, forelimb flailing and paw pushing ŽGrill and Norgren, 1978a., whereas decerebrate rats showed gaping and face washing in all animals Ž n s 9. and chin rubbing and head shaking only in three rats. Forelimb flailing or paw pushing was not observed in decerebrate rats, as reported by Grill and Norgren Ž1978b.. In spite of the behavioral difference, both normal and decerebrate rats secreted 2–5 ml of submandibular saliva on infusion of 0.5 M sucrose, 0.1 M NaCl, and 0.01 M HCl ŽFig. 1A–C, obtained from a decerebrate rat., and vigorous salivation was evoked by 1–10 mM quinine ŽFig. 1D.. The rejection behavior lasted for 3–5 min and gaping was most frequently observed in both normal and decerebrate rats. In EMG activity ŽFig. 1E., rhythmical responses of gapes could be differentiated from those of licks primarily by amplitude and duration of the anterior digastric muscle ŽYamamoto et al., 1982; Travers and Norgren, 1986; Dinardo and Travers, 1994.. When the number of gapes was estimated by counting the rhythmical activity of the anterior digastric muscle, it was found that they increased with the quinine concentration ŽFig. 2A.. Statistical analysis of the concentration–response functions showed no difference between normal Ž n s 10. and decer-
Fig. 2. Number of gapes ŽA. and rate of salivary flow ŽB. for 3 min as a function of quinine concentration in normal and decerebrate rats. Each value was the mean"SE obtained from normal Žopen circles, ns10. and decerebrate Žsolid circles, ns9. rats. No significant difference between the groups was evident in the concentration–response function to number of gapes or salivary flow rate.
ebrate Ž n s 9. rats. The salivary flow rate also increased as a function of quinine concentration in normal and decerebrate rats ŽFig. 2B., and there was no difference between the groups. This result indicates that the fundamental neural structures for the quinine-evoked rejection behavior Žgaping. and salivation were situated in the lower brainstem. 3.2. Efferent connection from the parabrachial region to the superior saliÕatory nucleus Next, we examined the neuroanatomy of the gustatory–salivary reflex arc in the lower brainstem based on the retrograde axonal transport of Fluoro-gold. After injec-
66
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73
Fig. 3. Injection site of Fluoro-gold and labeled cells in the caudal parabrachial region. ŽA. Photomicrograph of injection site stained with cresyl violet, which was taken under brightfield illumination. ŽB. Enlarged photomicrograph of the injection site Žsquare area in A. taken under ultraviolet illumination. ŽC. Labeled neurons scattered in the supratrigeminal area and reticular formation. The photomicrographs of B and C are shown as negative film images for visualizing background structures.
tion of Fluoro-gold into the superior salivatory nucleus ŽFig. 3A,B., we observed the retrogradely labeled neurons in the lower brainstem from the level of the obex to the rostral parabrachial region ŽFig. 4.. The labeled neurons
were not found in the solitary nucleus, the rostral half of which contains the taste relay neurons ŽFig. 4D., and only a few neurons were detected in the parabrachial taste area ŽFig. 4B,b.. In the rostral parabrachial area ŽFig. 4A,a.,
Fig. 4. Distribution of Fluoro-gold-labeled cells and extent of Fluoro-gold injection site Žgrey area in panel C.. Drawings are arranged rostrocaudally from A to E, and panels a and b show the parabrachial areas in A and B, respectively, ipsilateral to the injection site. The arrow in b indicates the waist area. bc s Brachium conjunctivum; g7 s genu facial nerve; Mo5 s trigeminal motor nucleus; NTS s nucleus of the solitary tract; PCRts parvicellular reticular nucleus; Pr5 s principal sensory trigeminal nucleus; py s pyramidal tract; Sp5C s spinal trigeminal nucleus, caudal division; Sp5I s spinal trigeminal nucleus, interpolar division; Sp5Os spinal trigeminal nucleus, oral division; Su5 s supratrigeminal nucleus; 7 s facial nucleus.
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73
most labeled cells were distributed in the pontine reticular formation dorsal to the trigeminal motor nucleus and parabrachial nucleus, except for the external lateral subdivision. At the caudal parabrachial level, labeled cells were mainly distributed in the supratrigeminal nucleus, pontine reticular formation just dorsal to the supratrigeminal nucleus, and external medial subnucleus of the parabrachial nucleus. Many labeled cells were also found in the lateral reticular formation from the level of the obex to the superior salivatory nucleus ŽFig. 4C–E.. Fig. 3C shows that the labeled cells appeared at the level of the parabrachial taste area. It is suggested that the taste relay neurons in the lower brainstem do not directly project to the superior salivatory neurons, and the pontine reticular formation may be included in the taste pathway from the parabrachial taste area to the superior salivatory nucleus. 3.3. Effect of parabrachial lesion on saliÕation and rejection behaÕior On the basis of the anatomical results shown above and of the c-fos immunohistochemical studies described in
67
Section 1 ŽYamamoto et al., 1994; Travers et al., 1999., we made bilateral electrolytic lesions in the parabrachial region in decerebrate rats, and compared differences in the number of gapes and rate of salivary flow. The photomicrographs in Fig. 5 show representative lesions made in the medial ŽA., lateral ŽB. and ventral ŽC. part of the parabrachial region, and Fig. 6 shows the extent of the lesions in all subjects used for the behavioral study Žeight rats for each site.. The solid areas show overlapping lesions in all animals, which centered at the waist ŽA. and lateral ŽB. of the brachium conjunctivum, and the pontine reticular formation between the brachium conjunctivum and trigeminal motor nucleus ŽC.. Fig. 7 shows the quinine concentration–response functions of the parabrachial lesioned and control rats. The control data are from the decerebrate rats in Fig. 2. As to the number of gapes ŽFig. 7A., the parabrachial lesions significantly decreased in gapes at quinine concentrations more than 1 mM ŽScheffe’s test, p - 0.05 following ANOVA: F3,29 s 23.91, P 0.001.. The effects of the lesion appeared most evident in the medial lesioned group; 15.5% of controls at 10 mM quinine. As to the salivary secretion ŽFig. 7B., the lateral
Fig. 5. Representative coronal sections with electrolytic lesions in the parabrachial region. Electrolytic lesions were made in the medial ŽA., lateral ŽB., and ventral ŽC. part of the parabrachial region.
68
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73
Fig. 6. Extent of the lesions in the parabrachial region. The medial ŽA., lateral ŽB., and ventral ŽC. lesions for eight rats are superimposed on camera-lucida drawings of the representative sections taken from the caudal level of the parabrachial nucleus Ž5.0-mm rostral to the obex.. Encircled areas are lesions for each rat Ž ns8., and solid areas are overlapped by all lesions: bc s brachium conjunctivum; Mo5s trigeminal motor nucleus.
lesions did not decrease on salivation at any of the stimulus concentrations ŽScheffe’s test, p ) 0.05., whereas the medial and ventral lesions significantly decreased on salivation at more than 1 mM quinine ŽScheffe’s test, p 0.05.. The medial and ventral lesioned rats showed 10.6% and 22.7% of control salivation, respectively, at 10 mM quinine. Under our experimental conditions, the parabrachial lesioning did not entirely abolish the quinine-induced salivation; the flow rate of the medial lesioned group was 7.2 " 1.2 mlr3 min at 10 mM. Moreover, the parabrachial lesioning affected little, if at all, salivation induced by other modalities of taste stimulation; oral-injection of 0.1 M NaCl, 0.5 M sucrose and 0.01 M HCl evoked 2–5 ml of salivation. These results suggest that the parabrachial region, especially the taste area and pontine reticular formation, are involved in the neural pathway subserving the vigorous salivation during taste rejection.
Fig. 7. Number of gapes ŽA. and rate of salivary flow ŽB. for 3 min as a function of quinine concentration in decerebrate and decerebrate plus parabrachial lesioned rats. Each value was the mean"SE obtained from decerebrate Žsolid circles, ns9., medial-parabrachial Žopen squares, n s 8., lateral-parabrachial Žopen triangles, n s 8., and ventralparabrachial Žopen rhombes, ns8. lesioned rats.
tion, and supratrigeminal nucleus. We defined a response as an increase in impulse frequency during taste or water Table 1 Number of neurons responsive to taste and movement of the jaw in relation to electrical stimulation of the superior salivatory nucleus Natural stimulation
3.4. Electrophysiological study of the parabrachial region Neural activity of the parabrachial region was examined by monitoring submandibular salivary secretion in lightly anesthetized decerebrate rats. Under the experimental conditions, we could record 48 neurons responsive to taste and water stimulation, and 35 responsive to jaw movements from the parabrachial nucleus, surrounding reticular forma-
Taste NaCl-best Sucrose-best HCl-best Quinine-best Water-best Movement of the jaw
Antidromic stimulation Žq.
Antidromic stimulation Žy.
0 0 2 8 0 8
15 3 13 3 4 27
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73
69
Fig. 8. Discharge pattern of a taste responsive and antidromically activated neuron recorded from the pontine reticular formation. ŽA. Taste stimuli were applied at the point indicated by the arrow, and quinine induced the best response. ŽBa. Spontaneous orthodromic impulse Žopen circle. and antidromic impulse Žsolid circle. evoked by electrical stimulation of the superior salivatory nucleus Žarrow.. ŽBb. The spontaneous and evoked impulses collided at shorter intervals between the spontaneous impulse and electrical stimulus.
stimulation greater than the mean " 2SD of spontaneous activity. Responses to jaw movement were brisk impulse discharges synchronized with passive jaw stretches, or
with spontaneous rhythmical jaw openings. Of 83 neurons recorded, 18 antidromically responded to electrical stimulation of the superior salivatory nucleus, suggesting they
Fig. 9. Distribution of neurons responsive to taste stimulation ŽA. and jaw movements ŽB.. Solid circles indicate neurons which did not respond to electrical stimulation of the superior salivatory nucleus. Open circles indicate neurons antidromically activated by the electrical stimulation.
70
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73
Fig. 10. Histologically confirmed recording sites by electrophoretic deposition of brilliant blue. This preparation contains recording sites of non-antidromic Žarrow 1. and antidromic Žarrow 2. taste responsive neurons.
project to the nucleus. Antidromicity was tested by the collision test and invariant impulse latency to electrical stimuli. Table 1 summarizes the number of neurons responsive to taste and jaw movements in relation to antidromic stimulation. Among the taste responsive neurons, the antidromic neurons responded best to quinine Žeight neurons. or HCl Žtwo neurons. ŽTable 1 and Fig. 8.. Such neurons were mainly located in the pontine reticular formation just dorsal to the supratrigeminal area ŽFig. 9A..
The non-antidromic neurons were distributed in the parabrachial taste area, which responded well to one of the four qualities of taste and water. A photomicrograph in Fig. 10 shows recording sites of the antidromic and nonantidromic taste responsive neurons. Both the antidromic and non-antidromic taste responsive neurons showed a tonic discharge pattern irrespective of movement of the jaw. The average impulse frequency was lower in the antidromic neurons than the non-antidromic neurons ŽFig. 11.. We could not record taste responsive neurons from the external lateral subdivision or rostral part of the parabrachial nucleus. The jaw-movement responsive neurons were mainly distributed in the supratrigeminal area, and the antidromic and non-antidromic neurons were intermingled ŽFig. 9B.. These results suggest that the pontine reticular formation is related to the gustatory–salivary reflex evoked by an aversive taste stimulus.
4. Discussion
Fig. 11. Mean response rate for the taste responsive neurons activated Žsolid columns. and non-activated Žopen columns. by antidromic stimulation of the superior salivatory nucleus. The response rate is spikes per s"SE measured over 10 s. Ss 0.5 M sucrose; Ns 0.1 M NaCl; H s 0.01 M HCl; Qs10 mM quinine; Wsdistilled water; SP sspontaneous discharge rate.
The taste rejection behavior of rats consists of various stereotyped oro-facial and body reactions. Among the reactions, gapes are rhythmical jaw and tongue movements, which can be quantified by calculating the magnitude, duration, and number of muscle contractions monitored by electromyographs ŽYamamoto et al., 1982; Travers and Norgren, 1986; Dinardo and Travers, 1994.. Decerebrate animals, at more than 10 days after decerebration, showed similar gapes in terms of number and duration of burst
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73
myographic activities of jaw muscles ŽGrill and Norgren, 1978b; Travers et al., 1999.. In the present study, using day two to three decerebrated rats to prevent atrophy of salivary glands, gaping, as well as profuse salivation similar to normal rats were observed. These results suggest that the fundamental neural structures for the quinine-evoked motor and autonomic reactions are situated in the lower brainstem. The gaping and salivation occurred at the same time in normal and decerebrate animals, but when the lateral part of the parabrachial nucleus was destroyed ŽFig. 5., the number of gapes was reduced without a marked effect on salivation. This separation of the reactions suggests that the salivation is evoked mainly by bitter taste stimulation, not by oral sensory inputs due to movement of the oral structures. The lower brainstem has the first- and second-order taste relay neurons, the solitary and parabrachial taste area, respectively, both of which are involved in the neural mechanism of salivation. Electrical stimulation of the solitary nucleus, and electrical and chemical stimulation of the parabrachial nucleus, produces salivary secretion. Moreover, a neuroanatomical study using the transneuronal transport of a virus showed that the parasympathetic nerve of rat submandibular gland polysynaptically connects with the solitary and parabrachial taste areas ŽJansen et al., 1992.. These previous findings suggest that taste information via both taste relay areas activates salivary secretion. The present lesioning study revealed that the parabrachial route is more crucial for the quinine-evoked vigorous salivation than the solitary route. Salivation after destruction of the parabrachial taste area, which is thought to be activated via the solitary route, had a much lower flow rate. However, the flow rates of the small salivation correspond to those when rats licked palatable taste solutions when conscious ŽMatsuo et al., 1994.. Therefore, it is conceivable that the parabrachial route is activated only during the taste rejection behavior. On the basis of the following present results, it is concluded that a feature of the parabrachial route is that bitter taste information from the parabrachial taste area is selectively inputed to the superior salivatory nucleus through the pontine reticular formation ventral to the parabrachial nucleus: Ž1. salivation during rejection was markedly reduced by lesioning of the parabrachial taste area or its ventral pontine reticular formation; Ž2. the Fluoro-gold labeled cells that may directly input to the superior salivatory nucleus were dominantly found in the pontine reticular formation and supratrigeminal area, not in the parabrachial taste area; Ž3. neurons in the pontine reticular formation, which may have a monosynaptic input to the superior salivatory nucleus, responded well to bitter taste stimulation. Our proposed parabrachial–salivary route contains the pontine reticular formation just dorsal to the supratrigeminal nucleus, capping the trigeminal motor nucleus dorsally. The present Fluoro-gold study showed the labeled neurons
71
in the supratrigeminal nucleus and its dorsal reticular formation, but not in the parabrachial taste area consisting of the ventral lateral and medial subnuclei. This result is very similar to that obtained by Herbert et al. Ž1990. after injection of horseradish peroxidase into the parvicellular reticular area just rostral to the solitary nucleus, which possibly includes the inferior salivatory nucleus. These anatomical studies suggest that the supratrigeminal nucleus and pontine reticular formation have projections to the salivatory nuclei. However, the present electrophysiological study showed that only the pontine reticular formation contains taste-responsive neurons innervating the superior salivatory nucleus. On the other hand, the supratrigeminal neurons were responsive to movement of the jaw, but not to taste stimulation. This nucleus receives inputs from trigeminal mesencephalic neurons arising from muscle spindles in the jaw-closing muscles Že.g. Fay and Norgren, 1997. and oral mechanoreceptors Že.g., Bae et al., 1997.. It is conceivable that the supratrigeminal nucleus is related to the so-called masticatory–salivary reflex. The present results imply the existence of a short circuit from the parabrachial taste area to the pontine reticular formation, and in this circuit, bitter and sour taste information of the parabrachial taste area was selectively conducted to the reticular formation in a tonic firing pattern Ževen when the jaw rhythmically moved. at a low frequency of about 7 Hz. Our observed impulse frequency of quinine response in the parabrachial taste area is similar to that observed by Di Lorenzo Ž1988. in decerebrate anesthetized rats. She also observed that decerebration decreased the impulse frequency of NaCl and HCl responses, but not that of the quinine response. Thus, the short circuit may have a synaptic connection to make a tonic firing pattern at low frequency. We speculate that such a low firing frequency is enough to evoke vigorous salivation, because electrical stimulation of the parasympathetic preganglionic nerve Žaxons of the superior salivatory neurons. at about 10 Hz evoked submandibular salivary secretion, whose flow rate was similar to that observed during taste rejection behavior ŽMatsuo 1999a,b.. Recently, an immunohistochemical technique to detect the expression of the Fos protein product of the immediate early gene c-fos has been applied to examine activated neurons in the central nervous system. The c-fos expression was examined in the brainstem after taste rejection behavior evoked by quinine stimulation in both normal ŽHarrer and Travers, 1996; DiNardo and Travers, 1997; Yamamoto and Sawa, 2000a,b. and decerebrate rats ŽTravers et al., 1999.. However, the authors of these studies make no mention of quinine-evoked c-fos expression in the superior salivatory nucleus, pontine reticular formation, or supratrigeminal nucleus. It is likely that this method is not sensitive enough to detect neurons subserving reflex salivation, partly because of their low firing activity. In respect to oro-facial movements, when taste solutions were infused into the mouth, c-fos was exten-
72
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73
sively expressed in the lateral and intermediate zone of the medullary reticular formation more caudal to the superior salivatory nucleus, which includes premotor neurons of masticatory, facial and lingual motor systems ŽDiNardo and Travers, 1997; Travers et al., 1997.. This area overlaps sites labeled with Fluoro-gold ŽFig. 7D,E., and receives inputs from the taste relay areas of the solitary nucleus ŽShammah-Lagnado et al., 1992; Halsell et al., 1996. and parabrachial nucleus ŽHerbert et al., 1990; Karimnamazi and Travers, 1998.. These findings suggest that the medullary reticular formation conducts taste- and motor-related information to the superior salivatory nucleus. Thus, there is a possibility that the medullary reticular formation is involved in the solitary route of the gustatory–salivary reflex. It is also speculated that the medullary reticular formation has a role in blood flow changes in the tongue associated with oro-facial movements andror taste stimulation; the superior salivatory nucleus also contains neurons innervating the parasympathetic ganglia in the tongue and regulates blood flow of the tongue ŽYu and Srinivasan, 1980; Matsuo and Kang, 1998.. In conclusion, we showed that the quinine-evoked salivation during rejection behavior is controlled by a neural axis of the parabrachial taste area—pontine reticular formation—salivatory nucleus. This salivary reflex arc is susceptible to anesthesia, and the pontine reticular formation showed a low firing frequency. It may be due to these characteristics, that neither electrophysiological taste response nor c-fos expression has been detected in the pontine reticular formation in previous studies.
Acknowledgements We thank to Profs. Y. Shigenaga and T. Yamamoto for discussion of the data. This study was supported by Grants-in-Aid for Scientific Research ŽNo. 11470391. from the Ministry of Education, Science and Culture of Japan.
References Bae, Y.C., Park, K.P., Yoshida, A., Nakagawa, S., Kurata, S., Chen, K., Takemura, M., Shigenaga, Y., 1997. Identification of gamma-aminobutyric acid-immunoreactive axon endings associated with mesencephalic periodontal afferent terminals and morphometry of the two types of terminals in the cat supratrigeminal nucleus. J. Comp. Neurol. 389, 127–138. Dick, T.E., Haxhiu, M.A., Cherniack, N.S., 1992. Salivary secretion elicited by activation of the parabrachial nuclei in the cat. J. Auton. Nerv. Syst. 39, 19–28. Di Lorenzo, P.M., 1988. Taste responses in the parabrachial pons of decerebrate rats. J. Neurophysiol. 59, 1871–1887. DiNardo, L.A., Travers, J.B., 1994. Hypoglossal neural activity during ingestion and rejection in the awake rat. J. Neurophysiol. 72, 1181– 1191. DiNardo, L.A., Travers, J.B., 1997. Distribution of Fos-like immunoreactivity in the medullary reticular formation of the rat after gusta-
tory elicited ingestion and rejection behaviors. J. Neurosci. 17, 3826– 3839. Fay, R.A., Norgren, R., 1997. Identification of rat brainstem multisynaptic connections to the oral motor nuclei using pseudorabies virus: I. Masticatory muscle motor systems. Brain Res. Rev. 25, 255–272. Flynn, F.W., Grill, H.J., Schwartz, G.J., Norgren, R., 1991. Central gustatory lesions: I. Preference and taste reactivity tests. Behav. Neurosci. 105, 933–943. Fulwiler, C.E., Saper, C.B., 1984. Subnuclear organization of the efferent connections of the parabrachial nucleus in the rat. Brain Res. Rev. 7, 229–259. Grill, H.J., Norgren, R., 1978a. The taste reactivity test: I. Mimetic responses to gustatory stimuli in neurologically normal rats. Brain Res. 143, 263–279. Grill, H.J., Norgren, R., 1978b. The taste reactivity test: II. Mimetic responses to gustatory stimuli in chronic thalamic and chronic decerebrate rats. Brain Res. 143, 281–297. Halsell, C.B., Travers, S.P., Travers, J.B., 1996. Ascending and descending projections from the rostral nucleus of the solitary tract originate from separate neuronal populations. Neuroscience 72, 185–197. Harrer, M.I., Travers, S.P., 1996. Topographic organization of Fos-like immunoreactivity in the rostral nucleus of the solitary tract evoked by gustatory stimulation with sucrose and quinine. Brain Res. 711, 125–137. Herbert, H., Moga, M.M., Saper, C.B., 1990. Connections of the parabrachial nucleus with the nucleus of the solitary tract and the medullary reticular formation in the rat. J. Comp. Neurol. 293, 540–580. Hill, D.L., Almli, C.R., 1983. Parabrachial nuclei damage in infant rats produces residual deficits in gustatory preferencesraversions and sodium appetite. Dev. Psychobiol. 16, 519–533. Jansen, A.S.P., Ter Horst, G.J., Mettenleiter, T.C., Loewy, A.D., 1992. CNS cell groups projecting to the submandibular parasympathetic preganglionic neurons in the rat: a retrograde transneuronal viral cell body labeling study. Brain Res. 572, 253–260. Karimnamazi, H., Travers, J.B., 1998. Differential projections from gustatory responsive regions of the parabrachial nucleus to the medulla and forebrain. Brain Res. 813, 283–302. Matsuo, R., 1999a. Interrelation of taste and saliva. In: Garrett, J.R., Ekstrom, ¨ J., Anderson, L.C. ŽEds.., Neural Mechanisms of Salivary Gland Secretion. Karger, Basel, pp. 185–195. Matsuo, R., 1999b. Discharge pattern of the superior salivatory nucleus in rats: in vivo and in vitro studies. In: Nakamura, Y., Sessle, B.J. ŽEds.., Neurobiology of Mastication. Elsevier, Amsterdam, pp. 139– 148. Matsuo, R., Kang, Y., 1998. Two types of parasympathetic preganglionic neurones in the superior salivatory nucleus characterized electrophysiologically in slice preparations of neonatal rats. J. Physiol. 513, 157–170. Matsuo, R., Yamamoto, T., Kiyomitsu, Y., Morimoto, T., 1989. Neural substrates for reflex salivation induced by taste, mechanical, and thermal stimulation of the oral region in decerebrate rats. Jpn. J. Physiol. 39, 349–357. Matsuo, R., Yamamoto, T., Ikehara, A., Nakamura, O., 1994. Effect of salivation on neural taste responses in freely moving rats: analyses of salivary secretion and taste responses of the chorda tympani nerve. Brain Res. 649, 136–146. Nilsson, B.-O., Rosengren, E., Ekstrom, ¨ J., 1990. Polyamines and longterm disuse of rat parotid glands. Acta Physiol. Scand. 140, 105–109. Paxinos, G., Watson, C., 1986. The Rat Brain in Stereotaxic Coordinates. Academic Press, New York. Phillips, M.I., Norgren, R.E., 1970. A rapid method for permanent implantation of an intraoral fistula in rats. Behav. Res. Methods Instr. 2, 124. Scott, J., Gunn, D.L., 1991. A comparative quantitative histological investigation of atrophic changes in the major salivary glands of liquid-fed rats. Arch. Oral Biol. 36, 855–857.
R. Matsuo et al.r Autonomic Neuroscience: Basic and Clinical 88 (2001) 61–73 Shammah-Lagnado, S.J., Costa, M.S.M.O., Ricardo, J.A., 1992. Afferent connections of the parvocellular reticular formation: a horseradish peroxidase study in the rat. Neuroscience 50, 403–425. Spector, A.C., 1995. Gustatory parabrachial lesions disrupt taste-guided quinine responsiveness in rats. Behav. Neurosci. 109, 79–90. Travers, J.B., Norgren, R., 1986. Electromyographic analysis of the ingestion and rejection of sapid stimuli in the rat. Behav. Neurosci. 100, 544–555. Travers, J.B., DiNardo, L.A., Karimnamazi, H., 1997. Motor and premotor mechanisms of licking. Neurosci. Biobehav. Rev. 21, 631–647. Travers, J.B., Urbanek, K., Grill, H.J., 1999. Fos-like immunoreactivity in the brain stem following oral quinine stimulation in decerebrate rats. Am. J. Physiol. 277, R384–R394. Yamamoto, T., Sawa, K., 2000a. c-Fos-like immunoreactivity in the
73
brainstem following gastric loads of various chemical solutions in rats. Brain Res. 866, 135–143. Yamamoto, T., Sawa, K., 2000b. Comparison of c-Fos-like immunoreactivity in the brainstem following intraoral and intragastric infusions of chemical solutions in rats. Brain Res. 866, 144–151. Yamamoto, T., Matsuo, R., Fujiwara, T., Kawamura, Y., 1982. EMG activities of masticatory muscles during licking in rats. Physiol. Behav. 29, 905–913. Yamamoto, T., Shimura, T., Sakai, N., Ozaki, N., 1994. Representation of hedonics and quality of taste stimuli in the parabrachial nucleus of the rat. Physiol. Behav. 56, 1197–1202. Yu, W.H., Srinivasan, R., 1980. Origin of the preganglionic visceral efferent fibres to the glands in the rat tongue as demonstrated by the horseradish peroxidase method. Neurosci. Lett. 19, 143–148.