Ž .Brain Research 800 1998 312–318

Short communication

Coexistence of S100b and putative transmitter agents in vagal andglossopharyngeal sensory neurons of the rat

Hiroyuki Ichikawa a, Cinda J. Helke b,)

a Second Department of Oral Anatomy, Okayama UniÕersity Dental School, Okayama, Japanb Department of Pharmacology and Neuroscience Program, Uniformed SerÕices UniÕersity of the Health Sciences, Bethesda, MD 20814, USA

Accepted 5 May 1998

Abstract

Ž . Ž . Ž .The coexistence of S100b with calcitonin gene-related peptide CGRP , substance P SP , somatostatin SOM , nicotinamideŽ . Ž .adenosine dinucleotide phosphate-diaphorase NADPH-d , and tyrosine hydroxylase TH was examined in the glossopharyngeal andŽ .vagal sensory ganglia. S100b immunoreactive -ir neurons in the jugular and petrosal ganglia frequently colocalized CGRP- or SP-ir,

whereas S100b-ir neurons in the nodose ganglion infrequently contained CGRP- or SP-ir. No S100b-ir neurons in the jugular andpetrosal ganglia showed SOM-ir while the small number of SOM-ir neurons in the nodose ganglion colocalized S100b-ir. Many neuronsin the nodose ganglion colocalized S100b-ir and NADPH-d activity, whereas S100b-ir neurons in the jugular and nodose gangliainfrequently contained NADPH-d activity. S100b- and TH-ir were frequently colocalized in nodose ganglion but not in petrosal orjugular ganglion neurons. These findings suggest relationships between S100b and specific putative transmitters in functions ofsubpopulations of vagal and glossopharyngeal sensory neurons. q 1998 Elsevier Science B.V. All rights reserved.

Keywords: Calcitonin gene-related peptide; Jugular ganglion; NADPH-diaphorase; Nodose ganglion; Petrosal ganglion; Somatostatin; Substance P;Tyrosine hydroxylase

Vagal and glossopharyngeal sensory neurons relay in-formation from visceral and somatic receptors of various

w xmodalities to the CNS 3,5,20,28 . The general visceralsensory neurons for the vagus and glossopharyngeal nervesare located in the nodose and petrosal ganglia, respectivelyw x3 . These neurons are associated with chemoreceptor,

w xbaroreceptor, gastrointestinal and pulmonary reflexes 20 .The respective general somatic sensory neurons for thesetwo cranial nerves are located in the superior ganglia of

Ž .the vagus jugular ganglion and glossopharyngeal nervesw x3 . Primary sensory neurons in the nodose, petrosal, andjugular ganglia contain putative transmitter agents or syn-thetic enzymes such as calcitonin gene-related peptideŽ . Ž . Ž .CGRP , substance P SP , somatostatin SOM , nicotin-amide adenosine dinucleotide phosphate-diaphoraseŽ .NADPH-d; a marker of nitric oxide producing neurons

Ž . w xand tyrosine hydroxylase TH 7,8,12,14,16,30 . Thesesubstances have been characterized in each ganglion ac-

) Corresponding author. Fax: q1-301-295-3220; E-mail:chelke@usuhs.mil

cording to their distributions, peripheral projections, andw xpossible functions 30 .

S100b is a member of the S100 family of smallŽ .calcium-binding proteins CaBPs , members of this family

contain two helix–loop–helix calcium-binding motifs andw xinteract with a wide range of proteins 6,24 . S100b is

w xpresent in neuronal perikarya and glia 13,15,21,29 , and ispresent in neuronal perikarya in the nodose, petrosal, and

w xjugular ganglia 13 . S100b can stimulate neuronal cal-cium flux, cause astrocytosis and neurite proliferation,

w xaffect microtubule assembly 24 , and has neurotrophicactions on dorsal root ganglion primary sensory neuronsw x25 . In an effort to gain additional insights into thefunction of S100b in sensory neurons, we studied itspresence in neurochemically-identified subpopulations of

Ž .both placode-derived nodose and petrosal ganglia andŽ .neural crest- derived jugular ganglion sensory neurons by

examining the possible coexistence of S100b with CGRP,SP, SOM, NADPH-d and TH in the nodose, petrosal andjugular ganglia of the adult rat.

Ž .Eight male Sprague–Dawley rats 200–300 g wereŽ .deeply anesthetized with chloral hydrate 400 mgrkg IP

Ž .and perfused transvascularly via the ascending aorta with

0006-8993r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved.Ž .PII: S0006-8993 98 00511-3

( )H. Ichikawa, C.J. HelkerBrain Research 800 1998 312–318 313

Ž .4% paraformaldehyde in phosphate buffer pH 7.4 . Thenodose, petrosal and jugular ganglia were removed andplaced overnight in phosphate buffered saline containing20% sucrose at 48C. The ganglia were frozen, sectioned at12 mm in a cryostat, mounted on gelatin-coated slides and

w xstained with a double immunofluorescence method 2,13The sections were incubated for 24 h at 48C with a mixture

Žof mouse monoclonal anti-S100b antibody aS2657, clone.SH-B4; 1:1000, Sigma, St. Louis, MO and one of the

Žfollowing; rabbit anti-CGRP serum 1:1000, Cambridge.Research Biochemicals, Atlantic Beach, NY , rabbit anti-

Ž .SP serum 1:1000, ImmunoNuclear, Stillwater, MN , rab-Ž .bit anti-SOM serum 1:1000, ImmunoNuclear or rabbit

Žanti-TH serum 1:200, Eugene Tech International, Allen-.dale, NJ . Subsequently, the sections were incubated with a

mixture of fluorescein isothiocyanate-conjugated donkeyŽanti-mouse IgG 1:100, Jackson ImmunoResearch Labs,.West Grove, PA and lissamine rhodamine B chloride-con-

Žjugated donkey anti-rabbit IgG 1:500, Jackson Im-.munoResearch Labs . Subsequent to analysis and photog-

raphy of the immunolabeling, the sections immunostainedfor S100b- and SOM-ir were processed for NADPH-dhistochemistry by incubation in 0.1 N phosphate bufferŽ .pH 7.4 containing 0.1 mgrml nitroblue tetrazoliumŽ . Ž .Sigma and 1.0 mgrml b-NADPH Sigma . In controlexperiments, immunolabeling with the monoclonal S100b

antibody was abolished by preabsorption of the antibodyŽwith the S100b protein 20 mgrml, SWant, Bellinzona,

.Switzerland . The specificities of other antibodies werew xpreviously described elsewhere 7,8 .

w xAs described previously 7,8,12–14,16,29 , S100b-im-Ž .munoreactive -ir , CGRP-ir, SP-ir, SOM-ir, TH-ir and

NADPH-d positive neuronal perikarya were found in thenodose, petrosal, and jugular ganglia. S100b-ir was ob-served in scattered neurons throughout the ganglia. CGRP-and SP-ir were observed mainly in neurons clustered in therostral portions of the nodose and petrosal ganglia andscattered throughout the jugular ganglion. SOM-ir wasobserved in scattered neurons throughout the petrosal andjugular ganglia, and in a few neurons in the nodoseganglion. TH-ir was found in scattered neurons throughoutthe nodose ganglion, in neurons clustered in the caudal

Ž .portion of the petrosal ganglion, and in few if anyneurons in the jugular ganglion. NADPH-d activity wasdetected in scattered neurons throughout the nodose gan-glion, and in few neurons, in the jugular and petrosalganglia.

The coexistence of S100b-ir with CGRP-ir, SP-ir,SOM-ir, TH-ir or NADPH-d activity was found in neuronsof the nodose, petrosal andror jugular ganglia. Whereas

Ž .few -1% , if any, of the numerous S100b-ir in theŽ .nodose ganglion contained CGRP or SP Fig. 1E–H ,

modest percentages of the S100b-ir neurons in the petrosalŽand jugular ganglia contained CGRP or SP Fig. 1A–D,I–

.L, Fig. 3A . Of the CGRP- or SP-ir neurons, considerableŽ .percentages 12–31% were also immunoreactive for

ŽS100b in the jugular and petrosal ganglia Fig. lA–D,.I–L; Fig. 3B , and of the small number of CGRP-ir

neurons in the nodose ganglion, 29% of colocalizedŽ .S100b-ir Fig. 1E–H, Fig. 3B . As reflected by the larger

number of CGRP-ir than SP-ir neurons in each ganglionw x7,8,14 , the percentage of CGRP-ir neurons which con-tained S100b-ir was always larger than that of SP-ir

Ž .neurons which contained S100b-ir Fig. 3 .Ž . ŽFew -1% in nodose ganglion or no petrosal and

. Žjugular ganglia S100b-ir neurons contained SOM-ir Fig..2A–F, Fig. 3A . However, of the few SOM-ir neurons

Ž .present in the nodose ganglion, all 8r8 containedŽ .S100b-ir Fig. 2E–F, Fig. 3B .

In the nodose ganglion, a large number of neuronsŽcolocalized S100b-ir and NADPH-d activity Fig. 2E,G,

.Fig. 3 , 41% of S100b-ir neurons contained NADPH-dactivity, and 68% of NADPH-d positive neurons were

Ž .immunoreactive for S100b Fig. 2E,G, Fig. 3 . However,the nodose neurons which colocalized S100b- and SOM-ir

Ž . Ž .lacked NADPH-d activity Fig. 2E–G . Few -1% of theS100b-ir neurons of the petrosal or jugular ganglion con-

Ž .tained NADPH-d activity Fig. 3A .Ž .In the nodose ganglion, a small percentage 7% of the

S100b-ir neurons showed TH-ir, however nearly one-halfŽof the TH-ir neurons were also S100b-ir Fig. 2J–K, Fig.

.3 . Very few S100b-ir neurons contained TH-ir in theŽ .petrosal and jugular ganglia Fig. 2H–I, Fig. 3 .

The present study demonstrates the coexistence ofS100b with CGRP, SP, SOM, NADPH-d or TH in thenodose, petrosal and jugular ganglia of the adult rat. Thisrepresents the first report of the neuronal colocalization ofS100b with any of these agents. With the exception of areport that S100b is colocalized with GABA in sensory

w xneurons of the chick inner ear 4 , no other studies havereported on the presence of S100b in neurochemically-identified sensory neurons. However, the robustness of thecoexistence of S100b varies with the putative neurotrans-mitter agents and with the ganglia. The most extensivecolocalization noted was for S100b and NADPH-d in thenodose ganglion. Perhaps because of the much lowernumbers of NADPH-d neurons in the petrosal and jugularganglia, little colocalization of S100b and NADPH-d wasnoted in those ganglia. Similarly, the considerable coexis-tence of S100b and either CGRP or SP noted in thepetrosal and jugular ganglia was not apparent in the no-dose ganglion as this ganglion contains relatively fewCGRP-ir neurons. However, the petrosal ganglion whichcontains similar numbers of TH-ir neurons as does thenodose ganglion, showed much less colocalization ofS100b and TH than did the nodose ganglion. Likewise,SOM-ir neurons colocalized S100b in the nodose gan-glion but not in the jugular or petrosal ganglia. Whetherthese are associated with the differences in embryologicorigins, andror sensory modalities transmitted is unclear.

The source of the S100b -ir in these neurons is notknown. Some neurons are known to express S100b mRNA

( )H. Ichikawa, C.J. HelkerBrain Research 800 1998 312–318314

Ž . Ž . Ž . ŽFig. 1. Representative double immunofluorescent photomicrographs for S100b A, C, E, G, I, K , CGRP B, D, F, H and SP J, L in the jugular A, B, I,. Ž . Ž .J , petrosal C, D, K, L and nodose ganglia E-H . Figs. A and B, C and D, E and F, G and H, I and J, and K and L show the same fields of view,

respectively. Arrows in A-F and I-L indicate neurons which colocalize S100b-ir with CGRP-or SP-ir. Arrowheads in G and H indicate a CGRP-ir neuronwhich lacks S100b-ir in the nodose ganglion. Calibration bar in A indicates 50 mm. A-L are at the same magnification.

w x21 , thus, it is possible that primary sensory neurons of thejugular, petrosal and nodose ganglia synthesize S100b.However, S100b is mainly synthesized by glial cellsw x6,23,24 . There is evidence that S100b may be secreted

w xand may act extracellularly 6,17,23,24,26 . Thus, it ispossible that S100b is synthesized in satellite cells, re-leased into extracellular spaces and binds to neurons. Thepresent immunohistochemical detection of S100b in vagal

( )H. Ichikawa, C.J. HelkerBrain Research 800 1998 312–318 315

Ž . Ž . Ž . Ž . Ž . ŽFig. 2. Representative photomicrographs for S100b A, C, E, H, J , SOM B, D, F , NADPH-d G and TH-ir I, K in the jugular A, B , petrosal C, D,. Ž .H, I and nodose ganglia E-G, J-K . Figs. A and B, C and D, E, F and G, H and I, J and K show the same fields of view, respectively. SOM-ir neurons in

Ž . Žthe jugular and petrosal ganglia lacked S100b-ir arrowheads in A-D . In the nodose ganglion, S100b-ir neurons colocalize either SOM-ir large arrows in. Ž .E-G or NADPH-d activity small arrows in E-G . Arrowheads in E-G indicate a NADPH-d positive neuron which lacks both S100b and SOM-ir. TH-ir

Ž . Ž .neurons colocalize S100b-ir in the nodose ganglion arrows in J, K but not in the petrosal ganglion arrowheads in H, I . Calibration bar in A indicates 50mm. A-K are at the same magnification.

and glossopharyngeal neurons could result from eitherneuronal synthesis andror the transport and accumulationof this protein from satellite cells.

The numerous reported actions of S100b include thoseon the development, maintenance, and pathology of the

w xnervous system 24 . S100b has trophic actions on neu-rons, including an effect to promote neurite outgrowth of

w xdorsal root ganglion sensory neurons 25 . Whereas thetrophic effects of S100b on the sensory neurons of the

vagus and glossopharyngeal nerves are unknown, theseneurons are responsive to certain trophic influences includ-ing those on the synthesis of putative transmittersw x9,10,18,27,30 . The neuronal coexistence of S100b withputative transmitter agents may suggest the action of S100b

on the synthesis of these agents in vagal and glossopharyn-geal neurons. S100b also stimulates calcium fluxes in

w xneurons 1 and can stimulate nitrite production by anŽ .action on inducible nitric oxide synthase NOS activity in

( )H. Ichikawa, C.J. HelkerBrain Research 800 1998 312–318316

Fig. 3. Bar graphs showing the colocalization of S100b with putative neurotransmitter agents in nodose, petrosal, and jugular ganglia neurons. Data areŽ .expressed in percentages with the actual cell counts of the immunoreactive neurons shown in parentheses above or at the top each bar. ) indicates that

less than 1% colocalization was found. The data were obtained from multiple sections from 3 ganglia. The data were reproducible among ganglia. Forexample, the mean and standard error values for the nodose ganglia colocalization of CGRP in S100b neurons was 1" .2 CGRPq S100b neurons and208"23 S100b neurons, and for the nodose ganglia colocalization of NADPH-d in S100b neurons was 89"18 CGRPq NADPH-d neurons and220"27 S100b neurons.

w xastrocytes 11 . The possibility that S100b can also acti-Ž .vate NOS as reflected by NADPH-d activity in nodose

ganglion sensory neurons is intriguing.

Whereas nodose ganglion neurons provide visceral af-ferent innervation to the abdominal, thoracic and extratho-racic visceral structures, only a few published reports have

( )H. Ichikawa, C.J. HelkerBrain Research 800 1998 312–318 317

described the peripheral projections of NADPH-d positive,w xSOM-ir or TH-ir neurons in the nodose ganglion 7,8,12 .

Thus, the projection of S100b-ir nodose neurons thatcolocalize these substances remains unclear. The petrosalganglion contains visceral sensory neurons of the glos-sopharyngeal nerve and the carotid sinus nerve. Theseneurons innervate the carotid body and taste buds in the

w xcircumvallate papilla of the tongue 13,16,19,22 . All TH-irneurons in the petrosal ganglion have been thought to

w xinnervate the carotid body 16 . Thus, the small numbers ofneurons in the petrosal ganglion found in this study tocolocalize TH- and S100b-ir suggest that TH- and S100b-irare not prominently colocalized in the carotid sinus nerveand carotid body. On the other hand, the abundance ofS100b-ir petrosal neurons which colocalize CGRP- orSP-ir suggests the coexistence in their lingual projections.This is supported by previous findings that the taste budcontains numerous CGRP-, SP- or S100b-ir nerve fibersw x13,19,22 . Recently, we demonstrated that 70% of cal-bindin D-28k-ir neurons colocalized S100b-ir in the pet-

w xrosal ganglion 13 . In the taste bud, most calbindin D-28k-ir nerve fibers colocalized S100b-ir, whereas S100b-irnerve fibers were more numerous than calbindin D-28k-ir

w xnerve fibers 13 . However, the number of neurons in thepetrosal ganglion shown to colocalize calbindin D-28k-

w xand CGRP-ir was very small 12 . Thus, it is unlikely thatnerve fibers which colocalize calbindin D-28- and S100b-iralso contain CGRP-ir. The jugular ganglion predominantlycontains somatic sensory neurons of the pharyngeal and

w xarticular branches of the vagus 3 . The pharynx containsnumerous CGRP-ir nerve fibers which are thought to be

w xderived from the jugular ganglion 22 . Thus, the abun-dance of CGRP-ir jugular neurons which colocalizeS100b-ir may suggest that these CGRP-ir nerve fibers alsocontain S100b-ir. Additional studies on the origin, projec-tion and sensory modalities of S100b in vagal and glos-sopharyngeal neurons will be important for understandingthe significance of this protein.

Acknowledgements

This work was supported by NIH grant R01-NS20991to C.J.H.

References

w x1 S.W. Barger, L.J. Van Eldik, S100 beta stimulates calcium fluxes inŽ .glial and neuronal cells, J. Biol. Chem. 267 1992 9689–9694.

w x Ž .2 Coons, A.H., Fluorescence antibody methods. In J.F. Danielli Ed. ,General Cytochemical Methods, Academic Press, New York, 1958,pp. 399–422.

w x3 F.S. Dubois, J.O. Foley, Quantitative studies of the vagus nerve inthe cat: II. The ratio of jugular to nodose fibers, J. Comp. Neurol. 67Ž .1937 69–87.

w x4 C.D. Fermin, D.S. Martin, Expression of S100 beta in sensory and

Žsecretory cells of the vertebrate inner ear, Cell. Mol. Biol. Noisy-. Ž .le-grand 41 1995 213–225.

w x5 G. Gabella, The vagus nerve. In Structure of the Autonomic NervousSystem, Halsted, New York, 1976, pp. 133–145.

w x6 C.W. Heizmann, Calcium-binding proteins: basic concepts and clini-Ž .cal implications, Gen. Physiol. Biophys. 11 1992 411–425.

w x7 C.J. Helke, K.M. Hill, Immunohistochemical study of neuropeptidesin vagus and glossopharyngeal afferent neurons in the rat, Neuro-

Ž .science 26 1988 539–551.w x8 C.J. Helke, A.J. Niederer, Studies on the coexistence of substance P

with other neurotransmitters in the nodose and petrosal ganglia,Ž .Synapse 5 1990 144–151.

w x9 T. Hertzberg, J.C.W. Finley, D.N. Katz, Trophic regulation ofŽ .carotid body afferent development, Adv. Exp. Med. Biol. 360 1993

305–307.w x10 T. Hertzberg, G. Fan, J.C.W. Finley, J.T. Erickson, D.M. Katz,

BDNF supports mammalian chemoafferent neurons in vitro andŽ .following peripheral target removal in vivo, Dev. Biol. 166 1994

801–811.w x11 J. Hu, F. Castets, J.L. Guevara, L.J. Van Eldik, S100 beta stimulates

inducible nitric oxide synthase activity and mRNA levels in ratŽ .cortical astrocytes, J. Biol. Chem. 27 1996 2543–2547.

w x12 H. Ichikawa, C.J. Helke, Coexistence of calbindin D-28k andNADPH-diaphorase in vagal and glossopharyngeal sensory neurons

Ž .of the rat, Brain Res. 735 1996 325–329.w x13 H. Ichikawa, C.J. Helke, Coexistence of calcium-binding proteins in

vagal and glossopharyngeal sensory neurons of the rat, Brain Res.Ž .768 1997 349–353.

w x14 H. Ichikawa, D.M. Jacobowitz, L. Winsky, C.J. Helke, Calretinin-immunoreactivity in vagal and glossopharyngeal sensory neurons ofthe rat; distribution and coexistence with putative transmitter agents,

Ž .Brain Res. 557 1991 312–316.w x15 H. Ichikawa, D.M. Jacobowitz, T. Sugimoto, S100 protein-im-

munoreactive primary sensory neurons in the trigeminal and dorsalŽ .root ganglia of the rat, Brain Res. 748 1997 253–257.

w x16 H. Ichikawa, A. Rabchevsky, C.J. Helke, Presence and coexistenceof putative transmitters in carotid sinus baro- and chemoreceptor

Ž .afferent neurons, Brain Res. 611 1993 67–74.w x17 D. Kligman, D. Marshak, Purification and characterization of a

neurite extension factor from bovine brain, Proc. Natl. Acad. Sci.Ž .U.S.A 82 1985 7136–7139.

w x18 D.B. Maclean, S.F. Lewis, F.B. Wheeler, Substance P content incultured neonatal rat vagal sensory neurons; the effect of nerve

Ž .growth factor, Brain Res. 457 1988 53–62.w x19 J.I. Nagy, H. Goedert, S.B. Hunt, A. Bond, The nature of the

substance P-containing nerve fibers in taste papillae of the ratŽ .tongue, Neuroscience 7 1982 3137–3151.

w x20 A.S. Paintal, Vagal sensory receptors and their reflex effects, Phys-Ž .iol. Rev. 53 1973 159–227.

w x21 N. Rickmann, J.R. Wolff, Sl00 protein expression in subpopulationsŽ .of neurons of rat brain, Neuroscience 67 1995 977–991.

w x22 G. Terenghi, J.N. Polak, J. Rodrigo, B.K. Mulderry, S.R. Bloom,Calcitonin gene-related peptide-immunoreactive nerves in the tongue,epiglottis and pharynx of the rat: occurrence, distribution and origin,

Ž .Brain Res. 365 1986 1–14.w x23 R.H. Selinfreund, S.W. Barger, W.J. Pledger, L.J. Van Eldik, Neu-

rotrophic protein S-l00b stimulates glial cell proliferation, Proc.Ž .Natl. Acad. Sci. U.S.A. 88 1991 3554–3558.

w x24 B.W. Schafer, C.W. Heizmann, The Sl00 family of EF-hand cal-cium-binding proteins: functions and pathology, Trends Biochem.

Ž .Sci. 21 1996 134–139.w x25 L.J. Van Eldik, B. Christie-Pope, L.M. Bolin, E.M. Shooter, W.O.

Whetsell Jr., Neurotrophic activity of S-100 beta in cultures ofdorsal root ganglia from embryonic chick and fetal rat, Brain Res.

Ž .542 1991 280–285.w x26 L.J. Van Eldik, D.B. Zimmer, Secretion of Sl00 from rat C6 glioma

Ž .cells, Brain Res. 436 1987 367–370.

( )H. Ichikawa, C.J. HelkerBrain Research 800 1998 312–318318

w x27 V.N. Verge, P.M. Richardson, Z. Wiesenfeld-Halin, T. Hokfelt,¨Differential influence of nerve growth factor on neuropeptide ex-pression in vivo: a novel role in peptide suppression in adult sensory

Ž .neurons, J. Neurosci. 15 1995 2081–2096.w x Ž .28 J.G. Widdicombe, Enteroreceptors. In J.I. Hubbard Ed. , The Pe-

ripheral Nervous System, Plenum, New York, 1974, pp. 455–485.

w x29 Q. Yang, X. Hou, A. Hamberger, S. Wang, A. Dahlstrom, K.G.Haglid, S-l00b immunoreactivity in neurones of the rat peripheral

Ž .sensory ganglia, NeuroReport 6 1995 2005–2009.w x30 H. Zhuo, H. Ichikawa, C.J. Helke, Neurochemistry of the nodose

Ž .ganglion, Prog. Neurobiol. 52 1997 79–107.