Directory UMM :Data Elmu:jurnal:B:Brain Research:Vol884.Issue1-2.2000:

www.elsevier.com / locate / bres

Interactive report

Neurotrophins alter the numbers of neurotransmitter-ir mature

1

vagal / glossopharyngeal visceral afferent neurons in vitro

*

Cinda J. Helke , Dominik Verdier-Pinard

Department of Pharmacology and Neuroscience Program, Uniformed Services University of the Health Science, 4301 Jones Bridge Road, Bethesda,

MD20814-4799, USA Accepted 27 September 2000

Abstract

Mature nodose and petrosal ganglia neurons (placodally derived afferent neurons of the vagal and glossopharyngeal nerves) contain TrkA and TrkC, and transport specific neurotrophins [nerve growth factor (NGF), neurotrophin-3 (NT-3), neurotrophin-4 (NT-4)]. This study evaluated neurotrophin influences on the presence of neuropeptides and / or neurotransmitter enzymes in these visceral sensory neurons. NGF, NT-3 and NT-4 (10–100 ng / ml) were applied (5 days) to dissociated, enriched, cultures of mature nodose / petrosal ganglia neurons, and the neurons processed for tyrosine hydroxylase (TH), vasoactive intestinal peptide (VIP), calcitonin gene-related peptide (CGRP) and neurofilament (NF-200) immunocytochemistry. Addition of NGF to nodose / petrosal ganglia neuron-enriched cultures significantly increased the number of TH-immunoreactive (ir) neurons, decreased the number of VIP-ir neurons in the cultures, and did not affect the numbers of CGRP-ir neurons. The addition of an NGF neutralizing antibody attenuated the effects of NGF on TH and VIP-ir neurons. NT-3 increased the number of VIP-ir neurons in the nodose / petrosal ganglia cultures and did not alter the numbers of TH-, or CGRP-ir neurons. The addition of an NT-3 neutralizing antibody attenuated the effects of NT-3 on VIP-ir neurons. NT-4 had no significant effects on the numbers of TH, VIP and CGRP-ir neurons. The absence of neurotrophin-induced changes in the numbers of NF-200-ir neurons in culture showed the lack of neurotrophin-mediated changes in survival of mature vagal afferent neurons. These data demonstrate that specific neurotrophins influence the numbers of neurons labeled for specific neurochemicals in nodose / petrosal ganglia cultures. These data, coupled with previous evidence for the presence of TrkA and TrkC mRNA and of the retrograde transport of NGF and NT-3, suggest important roles for NGF and NT-3 in the maintenance of transmitter phenotype of these mature visceral afferent neurons.  2000 Elsevier Science B.V. All rights reserved.

Theme: Sensory systems

Topic: Somatic and visceral afferents

Keywords: Nodose ganglion; Petrosal ganglion; Nerve growth factor; Neurotrophin-3; Neurotrophin-4; Vasoactive intestinal peptide; Tyrosine hydroxylase

1. Introduction rons, NGF regulates the expression of substance P and calcitonin gene-related peptide (CGRP) immunoreactive An established action of neurotrophins [nerve growth (ir) neurons in vitro and in vivo [2,16]. The presence of a factor (NGF), brain-derived neurotrophic factor (BDNF), target-derived factor other than NGF has been proposed to neurotrophin-3 (NT-3) and neurotrophin-4 (NT-4)] is the induce the gene expression of VIP and galanin in DRG maintenance of normal neurotransmitter and neuropeptide [10,24]. In sympathetic neurons, NGF effects the expres-phenotype expression in mature neurons [12,13,18,30]. In sion of TH [19,28].

mature somatic dorsal root ganglion (DRG) sensory neu- In contrast, little is known about the responsiveness to and functions of neurotrophins in other types of adult sensory neurons such as the visceral afferent neurons of

1

Published on the World Wide Web on 23 October 2000.

the nodose and petrosal ganglia. These neurons are critical

*Corresponding author. Tel.: 11-301-2953-238; fax: 1

1-301-2953-to the maintenance of cardiovascular, respira1-301-2953-tory and

220.

E-mail address: chelke@usuhs.mil (C.J. Helke). gastrointestinal functions. The placode-derived visceral

0006-8993 / 00 / $ – see front matter  2000 Elsevier Science B.V. All rights reserved. P I I : S 0 0 0 6 - 8 9 9 3 ( 0 0 ) 0 2 9 8 8 - 7

afferent neurons of the nodose and petrosal ganglia trans- petrosal ganglia (from 15 rats) were pooled in 2 ml of F14 mit visceral sensory information from specialized sensory growth medium supplemented with 10% heat-inactivated endings of the vagus nerve to the nucleus tractus solitarius horse serum (F-14 complete medium, F-14CM), freed of [1,7,31]. Differences in embryologic origin, transmitter nerve trunks and capsular connective tissue, treated twice contents, and presence of specific neurotrophin receptors for 1.5 h at 378C with 0.25% collagenase, and treated for [36] predict distinct differences in the neurotrophin respon- 45 min with 0.25% trypsin in F-14. Following trituration siveness of mature vagal afferent and efferent neurons by 8–10 passages through the barrel of fire-polished compared to somatic sensory neurons or sympathetic pasteur pipet, a single-cell suspension was obtained, and neurons. phase-bright neurons (15–50mm diameter) were counted Neurons of the nodose and petrosal ganglia do show with a hemocytometer. Centrifugation and resuspension of plasticity in the phenotype of neurotransmitters and neuro- the neuronal pellet removed cellular debris.

peptides in response to injury or environmental change Neuronal viability was tested with Trypan blue exclu-[6,8,26,38]. The ability of inhibition of the axoplasmic sion. Neuronal enrichment was achieved by a differential transport in the vagus nerve, to alter the numbers of VIP-, adhesion procedure wherein the cell suspension was pre-TH- and CGRP-ir and mRNA containing neurons in the plated overnight in a 35 mm polyornithine (500 mg / ml)-nodose ganglion of the adult rat [37] suggests that axonal coated culture dish. After 12–14 h, the non-neuronal cells transport of a target-derived regulatory influence, such as a were firmly attached to the culture dish, the weakly neurotrophin, alters the expression of these neurochemi- attached neurons were gently dislodged with media. Fur-cals. ther neuronal enrichment was achieved by centrifugation, Specific neurotrophin receptors (p75, TrkA, TrkC) are the pellet of viable neurons was resuspended in 0.8–1 ml present on mature nodose and petrosal ganglia neurons F14CM. Cell counts and viability testing with Trypan blue [4,34,35]. Moreover, NGF, NT-3 and NT-4 (but not were done. Enriched neurons from 30 pairs of ganglia (15 BDNF) are transported retrogradely by mature afferent rats) were seeded (1000 neurons / well in 600 ml in vagal neurons to the nodose ganglion [4]. These data and F14CM) in four-well (|20 wells / experiment) Nunc

Lab-the competition profiles for Lab-the transport of NT-3 and NGF Tek chamber slides (Nalge Nunc, Rochester, NY, USA) [4] are consistent with the presence of TrkA and TrkC and (double-coated with 500mg / ml polyornithine and 5mg / ml the absence of TrkB in the nodose ganglion. However, laminin), and incubated with 800 ml of Ham’s F-14CM little is known about neurotrophic factors and the regula- maintained at 378C in 3.5% CO –96.5% air. Media was2

tion of expression of neurochemicals in mature placode- changed at 2–3 day intervals.

derived visceral sensory neurons of the nodose ganglion. Human recombinant NGF and NT-4 were supplied by To determine the effects of neurotrophins on TH, VIP Regeneron (Tarrytown, NY, USA). Human recombinant and CGRP in visceral afferent neurons, human recombi- NT-3 was provided by Amgen (Thousand Oaks, CA, nant NGF, NT-3 and NT-4 were applied to dissociated, USA). Neurotrophins (10, 50 and 100 ng / ml) were added enriched, cultures of mature nodose and petrosal ganglia on the day of plating and for the entire time in culture (5 neurons. The numbers of TH-, VIP- and CGRP-immuno- days). In some experiments, neutralizing antibodies to labeled cells were assessed after 5 days in culture in the NGF (monoclonal antibody to mouse-NGF, catalog No. presence of three doses of each neurotrophin. Neurofila- 1087 754, used at 0.5 mg / ml, Boehringer Mannheim) or ment 200-ir labeling was used to label sensory neurons [3] 3 (chicken polyclonal IgY to human recombinant NT-and assess the survival of neurons after 5 days in culture 3, catalog No. G1651, used at 2 mg / ml; Promega, with and without exogenous neurotrophins. Madison, WI, USA) were added to the wells on the day of

plating and for the entire time in culture (5 days). Cultured nodose and petrosal ganglia cells were fixed on

2. Materials and methods the Lab-Tek chamber slides with 4% paraformaldehyde (pH 7.4; 800 ml / well for 20 min at room temperature), Adult male Sprague–Dawley rats weighing 225–250 g rinsed with phosphate buffered saline (pH 7.4), incubated were obtained from Taconic Farms (Germantown, NY, (60 min at room temperature) with blocking serum [0.3% USA), and given food and water ad libitum on a 12 h Triton X-100 and 5% normal goat serum in phosphate-light–12 h dark cycle. Rats were anesthetized with buffered saline (PBS]). After rinsing, primary antibodies halothane (2% in oxygen) and decapitated. The bilateral (diluted in 0.3% Triton X-100 in PBS) were added for nodose and petrosal ganglia were rapidly and aseptically overnight incubation at room temperature. The primary dissected and separated from the jugular ganglion. Because antibodies used were: rabbit polyclonal anti-TH (1:1000; separate preparations of nodose ganglia and petrosal Eugene Tech, Avendale, NJ, USA); rabbit polyclonal anti-ganglia yielded too few neurons, the nodose and petrosal VIP serum (1:1000, Incstar, Stillwater, MN, USA); rabbit ganglia (both placode-derived visceral afferent ganglia) polyclonal anti-CGRP (1:1000, Cambridge Research Bio-were pooled for these experiments. For each preparation of chemicals, Atlantic Beach, NJ, USA), polyclonal rabbit cultured adult ganglia neurons, 30 pairs of nodose and anti-NPY antiserum (1:1000 Peninsula Labs., Belmont,

CA, USA); rabbit anti-neurofilament 200 (NF-200) IgG control using one-way analysis of variance (ANOVA) and (1:500, Sigma, St. Louis, MO, USA). The antisera were multiple comparison test.

previously tested and characterized for specificity [3,5,6,9]. Immunolabeling was visualized using the appropriate

secondary antibodies, avidin–biotin–peroxidase reagent 3. Results

(VectaStain Elite ABC kit), and the chromagen, Vip

(Vector Labs.). After chambers and gaskets were removed Preliminary studies showed that dissociated neuron-en-from the slides, the slides were rinsed, dried, and cover riched cultures of mature rat nodose and petrosal ganglia slipped with DPX mountant (Fluka, New York, NY, USA), were viable (Fig. 1). Survival of neurons was apparent for and viewed with a Leitz Diaplan microscope equipped for at least 7 days in culture, although by 7 days in the absence photomicroscopy and digital image capture. of anti-mitotic agents there was an increased number of The average numbers and percentages of immuno- non-neuronal cells thus neurons were routinely harvested reactive neurons labeled for each agent were counted for and processed for immunocytochemistry after 5 days in each sample well. The total number of immunolabeled culture. TH-, VIP-, CGRP- and NF-200-ir were present and neurons with visible nuclei per well were counted. The readily detectable in nodose / petrosal ganglia neurons after percentage of immunolabeled neurons (number of im- 5 days in culture (Fig. 1).

munoreactive neurons with visible nuclei divided by the

total number of neurons with visible nuclei3100) in 6–8 3.1. NGF randomly selected fields of view within each culture well

were determined and averaged to obtain a single value for Each concentration of NGF (10, 50 and 100 ng / ml) that well. Data are presented as means6standard error of applied to nodose / petrosal ganglia neuron-enriched cul-the mean (S.E.M.). Each value presented is averaged from tures and present throughout the 5 days in culture, similar-a minimum of three culture chsimilar-amber wells from similar-at lesimilar-ast ly elevated the numbers and percentages of TH-ir neurons two distinct preparations of cultured adult ganglia neurons. (Fig. 2A). In the presence of 100 ng / ml NGF, the numbers Statistical significance was set at P,0.05 compared to (219624 versus 303628 TH-ir neurons /|1000 neurons

Fig. 1. Photomicrographs of immunolabeled dissociated neuron-enriched cultures of nodose / petrosal ganglia after 5 days in culture. (A) TH-ir neurons, (B) VIP-ir neurons, (C) CGRP-ir neurons, (D) NF-200-ir neurons. The calibration bar represents 50mm.

plated, control and NGF, respectively) and the percentage (19% versus 27% TH-ir neurons as % of neurons in culture well, control and NGF, respectively) of cultured neurons that were immunolabeled for TH-ir increased more than a third. The addition of an NGF neutralizing antibody (0.5mg / ml) completely attenuated the effects of 100 ng / ml NGF on TH-ir (Fig. 2A).

In the presence of 100 ng / ml NGF, a reduction in the numbers (265622 versus 182617 VIP-ir neurons /|1000

neurons plated, control and NGF, respectively) and per-centage (24% versus 16% VIP-ir neurons in culture well, control and NGF, respectively) of VIP-ir nodose / petrosal ganglia neurons was noted (Fig. 2A). A smaller but significant decline in VIP-ir neurons was noted with 10 ng / ml but not 50 ng / ml NGF. The addition of an NGF neutralizing antibody (0.5 mg / ml) partially attenuated the effects of 100 ng / ml NGF on VIP-ir neurons (Fig. 2A).

The number and percentages of CGRP-ir neurons were unchanged by any concentration of NGF (Fig. 2A). The number of NF-200-ir nodose / petrosal ganglia neurons did not significantly change in the presence of any dose of NGF over 5 days in culture (data for 100 ng / ml shown in Fig. 4).

3.2. NT-3

Human recombinant NT-3 (10, 50, 100 ng / ml) did not alter the numbers or percentages of nodose / petrosal gang-lion neurons immunolabeled for TH, or CGRP after 5 days in culture (Fig. 2B). Nor did the addition of an NT-3

Fig. 2. Bar graphs showing the percent of nodose / petrosal ganglia

neutralizing antibody affect the numbers of TH-ir neurons

neurons immunoreactive for TH, VIP and CGRP after 5 days in cultures

(Fig. 2B).

treated with (A) NGF at 0, 50, 100 ng / ml or 100 ng / ml in the presence of

Each dose of NT-3 (10, 50 100 ng / ml) significantly

an NGF neutralizing antibody (1AB), (B) NT-3 at 0, 10, 50, 100 ng / ml

or 100 ng / ml in the presence of an NT-3 neutralizing antibody (1AB). increased the numbers and percentages of VIP-ir nodose /

Neurotrophins and neutralizing antibodies were present from the time of _{petrosal ganglia neurons in the cultures at 5 days. The 100} plating, throughout the 5-day culture period. a5P,0.05 compared to

ng / ml concentration of NT-3, increased the numbers of

control, b5P,0.05 compared to corresponding 100 ng / ml data.

VIP-ir neurons from 244618 to 347627 (neurons /|1000

neurons plated, control and NT-3 treated, respectively) and the percentage of VIP-ir neurons from 19% to 26% (of total neurons in culture well, control and NT-3 treated, respectively) (Fig. 2B). The addition of an NT-3 neutraliz-ing antibody (2mg / ml) completely attenuated the effects of 100 ng / ml NT-3 on VIP-ir neurons (Fig. 2B). The number and percentages of CGRP-ir neurons (Fig. 2B) or NF-200-ir neurons (Fig. 4) did not change in the presence of any dose of NT-3 over 5 days in culture.

3.3. NT-4

Addition of human recombinant NT-4 (10, 50 100 ng / ml) to cultures of nodose / petrosal ganglia neurons throughout the 5-day culture time, had no significant

Fig. 3. Bar graphs showing the percent of nodose / petrosal ganglia

effects on the numbers or percentages of TH, VIP, CGRP

neurons immunoreactive for TH, VIP and CGRP after 5 days in cultures

decrease in TH and an increase in VIP neurons [6,37,38], the current data are consistent with a role for endogenous NGF in vivo to maintain normal neurotransmitter pheno-type in nodose / petrosal ganglia neurons. These data, coupled with the presence of TrkA mRNA and of the

125

retrograde transport of I-NGF by vagal afferent neurons, indicate an important role for NGF in the functions (including maintenance of normal transmitter phenotype) of these visceral afferent neurons. Moreover, these data coupled with the well established insensitivity of develop-ing nodose ganglion neurons to the survival promotdevelop-ing actions of NGF at a time when the neurons do require trophic support [14], suggest a developmental change in the responsiveness of these neurons. Our data also demon-strate differences in the response between mature

placode-Fig. 4. Bar graphs showing the numbers of NF-200-ir neurons (per 1000 derived visceral afferent and neural crest-derived DRG

cells plated in each well) in dissociated, enriched 5-day cultures of _{neurons. In distinction to the NGF-induced alterations in} nodose / petrosal ganglia in the absence of neurotrophin (control), and the

TH- and VIP-ir in nodose / petrosal ganglia neurons; in

presence of 100 ng / ml of NGF, NT-3 or NT-4.

adult DRG neurons in culture, NGF (50 ng / ml) increases SP- and CGRP-ir but does not alter VIP-ir in culture (mature DRG neurons do not have TH) [17,23]. Verge et al. [33] showed however that NGF infusion reduced the

4. Discussion numbers of VIP-ir neurons in the DRG after injury. In sympathetic neurons of the superior cervical ganglion, The data presented here demonstrate that specific neuro- NGF increases the expression of TH and but does not alter trophins (NGF and NT-3) influence the numbers of VIP [19,39].

neurons labeled for TH and VIP in nodose / petrosal The mechanism through which NGF alters the numbers ganglia cultures. That the effects are mediated by actions of TH and VIP neurons is not known. One possibility is of the neurotrophin is verified by the ability of the that NGF is acting directly through TrkA receptors on precipitating antibodies to prevent or attenuate the actions specific cultured nodose and petrosal ganglion neurons. of the neurotrophins. The neurotrophin-induced changes Although we have preliminary evidence for the presence of are likely due to changes in the content of the neurochemi- TrkA mRNA in cultured NG / PG neurons (Zhuo, Verdier-cals within neurons and are unlikely to be due to altera- Pinard and Helke, unpublished data), we do not know if tions in neuronal survival. This is based on our data the TrkA mRNA containing neurons are those in which the demonstrating an absence of effect of any of the neuro- content of TH or VIP is altered. Moreover, we did not find trophins tested on the numbers of NF-200 immunoreactive TH-positive neurons of intact nodose and petrosal ganglia neurons present after 5 days in culture. Moreover, it was that co-expressed TrkA [9]. NGF is also a ligand for p75 previously shown that the mature visceral sensory neurons and nearly all nodose ganglion neurons contain p75 do not require exogenous neurotrophins for survival [32,34,36]. Thus, the roles of neuronal TrkA and p75, [11,15,16]. However, we cannot rule out the possibility remain to be defined in this system. Likewise, the possi-that specific survival effects on a small subset of neurons bility that a non-neuronal cell type (e.g., fibroblasts) expressing a specific agent contributes to an observed remaining in these neuronally-enriched cultures responds change in immunoreactive neurons. The proposed changes to the addition of NGF with the secretion of a factor that in content of immunoreactive agents within the neurons of secondarily alters neuronal phenotype requires additional the cultured nodose and petrosal ganglia may reflex studies.

neurotrophin-induced changes in TH and / or VIP mRNA Whereas NT-3 (up to 150 ng / ml) does not affect peptide content and subsequent synthesis of TH and / or VIP. expression in newborn DRG [23], we found that NT-3 did However, it is also possible that the neurotrophins either affect the numbers of VIP-ir neurons in the mature nodose / directly or indirectly affect post translational processing, petrosal ganglia cultures. In contrast, in DRG neurons and / or stability of the products within the neurons. NT-3 was shown to upregulate neuropeptide Y but not VIP In this study, exogenous NGF increased the TH-ir and in an in vivo model of neural injury where VIP was already decreased VIP-ir neurons in the nodose / petrosal ganglia significantly upregulated [27].

cultures. Given that in vivo NGF is a retrogradely trans- The opposite actions of NGF and NT-3 on the numbers ported target-derived neurotrophin, and that loss of contact of VIP-ir neurons are intriguing and potentially significant. with target (either through vagotomy or inhibition of If additional studies verify that these actions result from axonal transport in the cervical vagus nerve) results in a changes in VIP mRNA, it would represent directionally

opposed regulation of a gene (VIP) by two distinct References

neurotrophins. This may be relevant to potentially distinct

effects of NGF versus NT-3 in specific therapeutic situa- [1] R.J. Contreras, R.M. Beckstead, R. Norgren, The central projections of the trigeminal, facial, glossopharyngeal and vagus nerves: and

tions, and will become more clear as the relative loss of, or

autoradiographic study in the rat, J. Auton. Nerv. Syst. 6 (1982)

need for, NT-3 versus NGF trophic support after specific

303–322.

types of perturbations (e.g., injury) to the neuronal

en-[2] J. Donnerer, R. Schuligoi, C. Stein, Increased content and transport

vironment. Perhaps locally derived (not target derived) _{of substance P and calcitonin gene-related peptide in sensory nerves} NT-3 is involved in the elevated VIP seen in vivo after innervating inflamed tissue: evidence for a regulatory function of

injury or inhibition of axonal transport. The induction of nerve growth factor in vivo, Neuroscience 49 (1992) 693–698. [3] M.E. Goldstein, S.B. House, H. Gainer, NF-L and peripherin

NT-3 mRNA was noted in non-neuronal cells of the vagus

immunoreactivities define distinct classes of rat sensory ganglion

nerve trunk immediately proximal and distal to a nerve

cells, J. Neurosci. Res. 30 (1991) 92–104.

lesion within 1 day after injury (Lee, Zhuo and Helke, _{[4] C.J. Helke, K.M. Adryan, J. Fedorowicz, H. Zhuo, J.S. Park, R.} submitted). In vivo this non-neuronally-derived NT-3 may _{Curtis, H.E. Radley, P.S. DiStefano, Axonal transport of} neuro-have access to the injured neuron and, coupled with the trophins by visceral afferent and efferent neurons of the vagus nerve

of the rat, J. Comp. Neurol. 393 (1998) 102–117.

loss of NGF, be involved in the elevation of neuronal VIP.

[5] C.J. Helke, A.J. Niederer, Studies on the coexistence of substance P

Neurotrophic influences on these visceral sensory

neu-with other putative transmitters in the nodose and petrosal ganglia,

rons are significant because of the importance of these

Synapse 5 (1990) 144–151.

neurons and the associated reflexes in maintaining homeo- _{[6] C.J. Helke, A. Rabchevsky, Axotomy alters putative} neurotrans-stasis of autonomic, respiratory, and endocrine visceral _{mitters in visceral sensory neurons of the nodose and petrosal} adaptive systems [25]. The functions of these neurons are ganglia, Brain Res. 551 (1991) 44–51.

[7] G.D. Housley, R.L. Martin-Body, N.J. Dawson, J.D. Sinclair, Brain

altered in chronic disease states such as hypertension

stem projections of the glossopharyngeal nerve and its carotid sinus

(chronic overloading of baroreceptors) and congestive

branch in the rat, Neuroscience 22 (1987) 237–250.

heart failure (chronic overloading of cardiac stretch

re-[8] F.-L. Huang, H. Zhuo, C. Sinclair, M.E. Goldstein, J.T. McCabe,

ceptors) [20,40]. Injury to visceral afferent nerves can _{C.J. Helke, Peripheral deafferentation alters calcitonin gene-related} occur in many ways, including trauma, tumors, and disease peptide mRNA in visceral sensory neurons of the nodose and

(e.g., diabetes mellitus, Guillain–Barre syndrome) petrosal ganglia, Mol. Brain Res. 22 (1994) 290–298.

[9] H. Ichikawa, C.J. Helke, The coexistence of TrkA with putative

[21,22,29]. Thus, a better understanding of the influences

transmitter agents and calcium-binding proteins in the vagal and

of neurotrophins on visceral afferent neurons may lead to

glossopharyngeal sensory neurons of the adult rat, Brain Res. 846

their use in the prevention and / or alleviation of clinical _{(1999) 268–273.}

problems associated with their dysfunction. [10] H. Kashiba, E. Senba, Y. Kawai, Y. Ueda, M. Tohyama, Axonal blockade induces the expression of vasoactive intestinal polypeptide and galanin in rat dorsal root ganglion neurons, Brain Res. 577 (1992) 19–28.

[11] H. Leal-Cardoso, G.M. Koschorke, G. Taylor, D. Weinreich,

Elec-5. Disclaimer _{trophysiological properties and chemosensitivity of acutely isolated}

nodose ganglion neurons of the rabbit, J. Auton. Nerv. Syst. 45

The opinions or assertions contained herein are the (1993) 29–39.

[12] R.M. Lindsay, Role of neurotrophins and trk receptors in the

private ones of the authors and are not to be construed as

development and maintenance of sensory neurons: an overview,

official or reflecting the views of the DoD or the USUHS.

Phil. Trans. Royal Soc. London 351 (1996) 365–373.

The experiments reported herein were conducted according _{[13] R.M. Lindsay, Therapeutic potential of the neurotrophins and} to the principles set forth in the ‘Guide for the Care and neurotrophin–CNTF combinations in peripheral neuropathies and

Use of Laboratory Animals’, Institute of Animal Re- motor neuron diseases, Ciba Found. Symp. 196 (1996) 39–48. [14] R.M. Lindsay, Y.A. Barde, A.M. Davies, H. Rohrer, Differences and

sources, National Research Council, DHEW Publ. No.

similarities in the neurotrophic growth factor requirements of

(NIH) 74-23.

sensory neurons derived from neural crest and neural placode, J. Cell Sci. 3 (Suppl.) (1985) 115–129.

[15] R.M. Lindsay, C.J. Evison, J. Winter, Culture of adult mammalian peripheral neurons, in: J. Chad, H. Wheal (Eds.), Cellular Neuro-biology – A Practical Approach, IRL Press, New York, 1991, pp.

Acknowledgements

3–17.

[16] R.M. Lindsay, A.J. Harmer, Nerve growth factor regulates

expres-We thank Drs. Ann Acheson (Regeneron, Tarrytown, _{sion of neuropeptide genes in adult sensory neurons, Nature} NY, USA) and Dr. James Coulombe (USUHS) for advice (London) 337 (1989) 362–364.

and guidance in setting up the cultures of mature nodose / [17] R.M. Lindsay, C. Lockett, J. Sternberg, J. Winter, Neuropeptide expression in cultures of adult sensory neurons: modulation of

petrosal ganglia neurons. Drs. Huang Zhuo and Clara

substance P and calcitonin gene-related peptide levels by nerve

Carobi (USUHS) are appreciated for their assistance with

growth factor, Neuroscience 33 (1989) 53–65.

and discussion of various aspects of the study. This work _{[18] R.M. Lindsay, S.J. Wiegand, C.A. Altar, P.S. DiStefano,} Neuro-was supported by NIH grants NS20991 and NS38845, and trophic factors: from molecule to man, TINS 17 (1994) 182–189.

regulation of neuronal gene expression by nerve growth factor, J. [32] V.M.K. Verge, J.-P. Merlio, J. Grondin, P. Ernfors, H. Persson, R.J. ¨

Cell Biol. 117 (1992) 135–141. Riopelle, T. Hokfelt, P.M. Richardson, Colocalization of NGF [20] J.W. McCubbin, C.M. Ferrario, Baroreceptor reflexes and hyperten- binding sites, trk mRNA, and low-affinity NGF receptor mRNA in sion, in: J. Genest et al. (Ed.), Hypertension, McGraw Hill, New primary sensory neurons: responses to injury and infusion of NGF, York, 1977, pp. 128–133. J. Neurosci. 12 (1992) 4011–4022.

¨ [21] A.J. McDougall, J.G. McLeod, Autonomic neuropathy. I: Clinical [33] V.M.K. Verge, P.M. Richardson, Z. Wiesenfeld-Hallin, T. Hokfelt,

features, investigation, pathophysiology, and treatment, J. Neurol. Differential influence of nerve growth factor on neuropeptide Sci. 137 (1996) 79–88. expression in vivo: a novel role in peptide suppression in adult [22] A.J. McDougall, J.G. McLeod, Autonomic neuropathy. II: Specific sensory neurons, J. Neurosci. 15 (1995) 2081–2096.

neuropathies, J. Neurol. Sci. 138 (1996) 1–13. [34] C. Wetmore, L. Olson, Neuronal and nonneuronal expression of [23] P.K. Mulderry, Neuropeptide expression by newborn and adult rat neurotrophins and their receptors in sensory and sympathetic ganglia sensory neurons in culture: effects of nerve growth factor and other suggest new intercellular trophic interactions, J. Comp. Neurol. 353 neurotrophic factors, Neuroscience 59 (1994) 673–688. (1995) 143–159.

[24] P.K. Mulderry, R.M. Lindsay, Rat dorsal root ganglion neurons in [35] H. Zhuo, C.J. Helke, Presence and localization of neurotrophin culture express vasoactive intestinal polypeptide (VIP) independent- receptor tyrosine kinase (TrkA. TrkB, TrkC) mRNAs in visceral ly of nerve growth factor, Neurosci. Lett. 108 (1990) 314–320. afferent neurons of the nodose and petrosal ganglia, Mol. Brain Res. [25] A.S. Paintal, Vagal sensory receptors and their reflex effects, Physiol. 38 (1996) 63–70.

Rev. 53 (1973) 159–227. [36] H. Zhuo, H. Ichikawa, C.J. Helke, Neurochemistry of the nodose [26] M. Reimer, M. Kanje, Peripheral but not central axotomy promotes ganglion, Prog. Neurobiol. 52 (1997) 79–107.

axonal outgrowth and induces alterations in neuropeptide synthesis [37] H. Zhuo, A.C. Lewin, E.T. Phillips, C. Sinclair, C.J. Helke, in the nodose ganglion of the rat, Eur. J. Neurosci. 11 (1999) Inhibition of the axoplasmic transport in the vagus nerve alters the

3415–3423. numbers of neuropeptide and tyrosine hydroxylase

mRNA-con-[27] G.D. Sterne, R.A. Brown, C.J. Green, G. Terenghi, NT-3 modulates taining and immunoreactive visceral afferent neurons of the nodose NPY expression in primary sensory neurons following peripheral ganglion, Neuroscience 66 (1995) 175–187.

nerve injury, J. Anat. 193 (1998) 273–281. [38] H. Zhuo, C. Sinclair, C.J. Helke, Plasticity of tyrosine hydroxylase [28] H. Thoenen, P.V. Angeletti, R. Levi-Montalcini, R. Kettler, Selective and vasoactive intestinal peptide mRNAs in visceral afferent neu-induction by nerve growth factor of tyrosine hydroxylase and rons of the nodose ganglion upon axotomy-induced deafferentation, dopamine-B-hydroxylase in the rat superior cervical ganglia, Proc. Neuroscience 63 (1994) 617–626.

Natl. Acad. Sci. USA 68 (1971) 1598–1692. [39] R.E. Zigmond, S.H. Hyatt, C. Baldwin, X. Qu, Y. Sun, T.W. McCain, [29] P.K. Thomas, C.J. Mathias, Diseases of the ninth, tenth, eleventh, R.C. Schreiber, U. Vaidyanathan, Phenotypic plasticity in adult and twelfth cranial nerves, in: P.J. Dyck, P.D. Thomas (Eds.), sympathetic neurons: changes in neuropeptide expression in organ Peripheral Neuropathy, 3rd Edition, Saunders, Philadelphia, PA, culture, Proc. Natl. Acad. Sci. USA 89 (1992) 1507–1511. 1993, pp. 869–885. [40] I.H. Zucker, J.P. Gilmore, Atrial receptor modulation of renal [30] D.R. Tomlinson, P. Fernyhough, L. Mohiuddin, J.D. Delcroix, M. function in heart failure, in: F.M. Abboud et al. (Ed.), Disturbances Malcangio, Neurotrophic factors – regulation of neuronal pheno- in Neurogenic Control of the Circulation, American Physiology type, Neurosci. Res. Commun. 21 (1997) 57–66. Society, Bethesda, MD, 1981, pp. 1–16.

[31] A. Torvik, Afferent connections to the sensory trigeminal nuclei, the nucleus of the solitary tract and adjacent structures, J. Comp. Neurol. 106 (1956) 51–141.

afferent neurons of the nodose and petrosal ganglia trans- petrosal ganglia (from 15 rats) were pooled in 2 ml of F14 mit visceral sensory information from specialized sensory growth medium supplemented with 10% heat-inactivated endings of the vagus nerve to the nucleus tractus solitarius horse serum (F-14 complete medium, F-14CM), freed of [1,7,31]. Differences in embryologic origin, transmitter nerve trunks and capsular connective tissue, treated twice contents, and presence of specific neurotrophin receptors for 1.5 h at 378C with 0.25% collagenase, and treated for [36] predict distinct differences in the neurotrophin respon- 45 min with 0.25% trypsin in F-14. Following trituration siveness of mature vagal afferent and efferent neurons by 8–10 passages through the barrel of fire-polished compared to somatic sensory neurons or sympathetic pasteur pipet, a single-cell suspension was obtained, and

neurons. phase-bright neurons (15–50mm diameter) were counted

Neurons of the nodose and petrosal ganglia do show with a hemocytometer. Centrifugation and resuspension of plasticity in the phenotype of neurotransmitters and neuro- the neuronal pellet removed cellular debris.

peptides in response to injury or environmental change Neuronal viability was tested with Trypan blue exclu-[6,8,26,38]. The ability of inhibition of the axoplasmic sion. Neuronal enrichment was achieved by a differential transport in the vagus nerve, to alter the numbers of VIP-, adhesion procedure wherein the cell suspension was pre-TH- and CGRP-ir and mRNA containing neurons in the plated overnight in a 35 mm polyornithine (500 mg / ml)-nodose ganglion of the adult rat [37] suggests that axonal coated culture dish. After 12–14 h, the non-neuronal cells transport of a target-derived regulatory influence, such as a were firmly attached to the culture dish, the weakly neurotrophin, alters the expression of these neurochemi- attached neurons were gently dislodged with media.

Fur-cals. ther neuronal enrichment was achieved by centrifugation,

Specific neurotrophin receptors (p75, TrkA, TrkC) are the pellet of viable neurons was resuspended in 0.8–1 ml present on mature nodose and petrosal ganglia neurons F14CM. Cell counts and viability testing with Trypan blue [4,34,35]. Moreover, NGF, NT-3 and NT-4 (but not were done. Enriched neurons from 30 pairs of ganglia (15 BDNF) are transported retrogradely by mature afferent rats) were seeded (1000 neurons / well in 600 ml in vagal neurons to the nodose ganglion [4]. These data and F14CM) in four-well (|20 wells / experiment) Nunc

Lab-the competition profiles for Lab-the transport of NT-3 and NGF Tek chamber slides (Nalge Nunc, Rochester, NY, USA) [4] are consistent with the presence of TrkA and TrkC and (double-coated with 500mg / ml polyornithine and 5mg / ml the absence of TrkB in the nodose ganglion. However, laminin), and incubated with 800 ml of Ham’s F-14CM little is known about neurotrophic factors and the regula- maintained at 378C in 3.5% CO –96.5% air. Media was₂ tion of expression of neurochemicals in mature placode- changed at 2–3 day intervals.

derived visceral sensory neurons of the nodose ganglion. Human recombinant NGF and NT-4 were supplied by To determine the effects of neurotrophins on TH, VIP Regeneron (Tarrytown, NY, USA). Human recombinant and CGRP in visceral afferent neurons, human recombi- NT-3 was provided by Amgen (Thousand Oaks, CA, nant NGF, NT-3 and NT-4 were applied to dissociated, USA). Neurotrophins (10, 50 and 100 ng / ml) were added enriched, cultures of mature nodose and petrosal ganglia on the day of plating and for the entire time in culture (5 neurons. The numbers of TH-, VIP- and CGRP-immuno- days). In some experiments, neutralizing antibodies to labeled cells were assessed after 5 days in culture in the NGF (monoclonal antibody to mouse-NGF, catalog No. presence of three doses of each neurotrophin. Neurofila- 1087 754, used at 0.5 mg / ml, Boehringer Mannheim) or ment 200-ir labeling was used to label sensory neurons [3] 3 (chicken polyclonal IgY to human recombinant NT-and assess the survival of neurons after 5 days in culture 3, catalog No. G1651, used at 2 mg / ml; Promega, with and without exogenous neurotrophins. Madison, WI, USA) were added to the wells on the day of

plating and for the entire time in culture (5 days). Cultured nodose and petrosal ganglia cells were fixed on 2. Materials and methods the Lab-Tek chamber slides with 4% paraformaldehyde (pH 7.4; 800 ml / well for 20 min at room temperature), Adult male Sprague–Dawley rats weighing 225–250 g rinsed with phosphate buffered saline (pH 7.4), incubated were obtained from Taconic Farms (Germantown, NY, (60 min at room temperature) with blocking serum [0.3% USA), and given food and water ad libitum on a 12 h Triton X-100 and 5% normal goat serum in phosphate-light–12 h dark cycle. Rats were anesthetized with buffered saline (PBS]). After rinsing, primary antibodies halothane (2% in oxygen) and decapitated. The bilateral (diluted in 0.3% Triton X-100 in PBS) were added for nodose and petrosal ganglia were rapidly and aseptically overnight incubation at room temperature. The primary dissected and separated from the jugular ganglion. Because antibodies used were: rabbit polyclonal anti-TH (1:1000; separate preparations of nodose ganglia and petrosal Eugene Tech, Avendale, NJ, USA); rabbit polyclonal anti-ganglia yielded too few neurons, the nodose and petrosal VIP serum (1:1000, Incstar, Stillwater, MN, USA); rabbit ganglia (both placode-derived visceral afferent ganglia) polyclonal anti-CGRP (1:1000, Cambridge Research Bio-were pooled for these experiments. For each preparation of chemicals, Atlantic Beach, NJ, USA), polyclonal rabbit cultured adult ganglia neurons, 30 pairs of nodose and anti-NPY antiserum (1:1000 Peninsula Labs., Belmont,

CA, USA); rabbit anti-neurofilament 200 (NF-200) IgG control using one-way analysis of variance (ANOVA) and (1:500, Sigma, St. Louis, MO, USA). The antisera were multiple comparison test.

previously tested and characterized for specificity [3,5,6,9]. Immunolabeling was visualized using the appropriate

secondary antibodies, avidin–biotin–peroxidase reagent 3. Results (VectaStain Elite ABC kit), and the chromagen, Vip

(Vector Labs.). After chambers and gaskets were removed Preliminary studies showed that dissociated neuron-en-from the slides, the slides were rinsed, dried, and cover riched cultures of mature rat nodose and petrosal ganglia slipped with DPX mountant (Fluka, New York, NY, USA), were viable (Fig. 1). Survival of neurons was apparent for and viewed with a Leitz Diaplan microscope equipped for at least 7 days in culture, although by 7 days in the absence photomicroscopy and digital image capture. of anti-mitotic agents there was an increased number of The average numbers and percentages of immuno- non-neuronal cells thus neurons were routinely harvested reactive neurons labeled for each agent were counted for and processed for immunocytochemistry after 5 days in each sample well. The total number of immunolabeled culture. TH-, VIP-, CGRP- and NF-200-ir were present and neurons with visible nuclei per well were counted. The readily detectable in nodose / petrosal ganglia neurons after percentage of immunolabeled neurons (number of im- 5 days in culture (Fig. 1).

munoreactive neurons with visible nuclei divided by the

total number of neurons with visible nuclei3100) in 6–8 3.1. NGF randomly selected fields of view within each culture well

were determined and averaged to obtain a single value for Each concentration of NGF (10, 50 and 100 ng / ml) that well. Data are presented as means6standard error of applied to nodose / petrosal ganglia neuron-enriched cul-the mean (S.E.M.). Each value presented is averaged from tures and present throughout the 5 days in culture, similar-a minimum of three culture chsimilar-amber wells from similar-at lesimilar-ast ly elevated the numbers and percentages of TH-ir neurons two distinct preparations of cultured adult ganglia neurons. (Fig. 2A). In the presence of 100 ng / ml NGF, the numbers Statistical significance was set at P,0.05 compared to (219624 versus 303628 TH-ir neurons /|1000 neurons

Fig. 1. Photomicrographs of immunolabeled dissociated neuron-enriched cultures of nodose / petrosal ganglia after 5 days in culture. (A) TH-ir neurons, (B) VIP-ir neurons, (C) CGRP-ir neurons, (D) NF-200-ir neurons. The calibration bar represents 50mm.

plated, control and NGF, respectively) and the percentage (19% versus 27% TH-ir neurons as % of neurons in culture well, control and NGF, respectively) of cultured neurons that were immunolabeled for TH-ir increased more than a third. The addition of an NGF neutralizing antibody (0.5mg / ml) completely attenuated the effects of 100 ng / ml NGF on TH-ir (Fig. 2A).

In the presence of 100 ng / ml NGF, a reduction in the numbers (265622 versus 182617 VIP-ir neurons /|1000

neurons plated, control and NGF, respectively) and per-centage (24% versus 16% VIP-ir neurons in culture well, control and NGF, respectively) of VIP-ir nodose / petrosal ganglia neurons was noted (Fig. 2A). A smaller but significant decline in VIP-ir neurons was noted with 10 ng / ml but not 50 ng / ml NGF. The addition of an NGF neutralizing antibody (0.5 mg / ml) partially attenuated the effects of 100 ng / ml NGF on VIP-ir neurons (Fig. 2A).

The number and percentages of CGRP-ir neurons were unchanged by any concentration of NGF (Fig. 2A). The number of NF-200-ir nodose / petrosal ganglia neurons did not significantly change in the presence of any dose of NGF over 5 days in culture (data for 100 ng / ml shown in Fig. 4).

3.2. NT-3

Human recombinant NT-3 (10, 50, 100 ng / ml) did not alter the numbers or percentages of nodose / petrosal gang-lion neurons immunolabeled for TH, or CGRP after 5 days in culture (Fig. 2B). Nor did the addition of an NT-3

Fig. 2. Bar graphs showing the percent of nodose / petrosal ganglia

neutralizing antibody affect the numbers of TH-ir neurons

neurons immunoreactive for TH, VIP and CGRP after 5 days in cultures

(Fig. 2B).

treated with (A) NGF at 0, 50, 100 ng / ml or 100 ng / ml in the presence of

Each dose of NT-3 (10, 50 100 ng / ml) significantly

an NGF neutralizing antibody (1AB), (B) NT-3 at 0, 10, 50, 100 ng / ml

or 100 ng / ml in the presence of an NT-3 neutralizing antibody (1AB). increased the numbers and percentages of VIP-ir nodose /

Neurotrophins and neutralizing antibodies were present from the time of _{petrosal ganglia neurons in the cultures at 5 days. The 100} plating, throughout the 5-day culture period. a5P,0.05 compared to

ng / ml concentration of NT-3, increased the numbers of

control, b5P,0.05 compared to corresponding 100 ng / ml data.

VIP-ir neurons from 244618 to 347627 (neurons /|1000

neurons plated, control and NT-3 treated, respectively) and the percentage of VIP-ir neurons from 19% to 26% (of total neurons in culture well, control and NT-3 treated, respectively) (Fig. 2B). The addition of an NT-3 neutraliz-ing antibody (2mg / ml) completely attenuated the effects of 100 ng / ml NT-3 on VIP-ir neurons (Fig. 2B). The number and percentages of CGRP-ir neurons (Fig. 2B) or NF-200-ir neurons (Fig. 4) did not change in the presence of any dose of NT-3 over 5 days in culture.

3.3. NT-4

Addition of human recombinant NT-4 (10, 50 100 ng / ml) to cultures of nodose / petrosal ganglia neurons throughout the 5-day culture time, had no significant

Fig. 3. Bar graphs showing the percent of nodose / petrosal ganglia

effects on the numbers or percentages of TH, VIP, CGRP

neurons immunoreactive for TH, VIP and CGRP after 5 days in cultures

decrease in TH and an increase in VIP neurons [6,37,38], the current data are consistent with a role for endogenous NGF in vivo to maintain normal neurotransmitter pheno-type in nodose / petrosal ganglia neurons. These data, coupled with the presence of TrkA mRNA and of the

125

retrograde transport of I-NGF by vagal afferent neurons, indicate an important role for NGF in the functions (including maintenance of normal transmitter phenotype) of these visceral afferent neurons. Moreover, these data coupled with the well established insensitivity of develop-ing nodose ganglion neurons to the survival promotdevelop-ing actions of NGF at a time when the neurons do require trophic support [14], suggest a developmental change in the responsiveness of these neurons. Our data also demon-strate differences in the response between mature

placode-Fig. 4. Bar graphs showing the numbers of NF-200-ir neurons (per 1000 derived visceral afferent and neural crest-derived DRG

cells plated in each well) in dissociated, enriched 5-day cultures of _{neurons. In distinction to the NGF-induced alterations in} nodose / petrosal ganglia in the absence of neurotrophin (control), and the

TH- and VIP-ir in nodose / petrosal ganglia neurons; in

presence of 100 ng / ml of NGF, NT-3 or NT-4.

adult DRG neurons in culture, NGF (50 ng / ml) increases SP- and CGRP-ir but does not alter VIP-ir in culture (mature DRG neurons do not have TH) [17,23]. Verge et al. [33] showed however that NGF infusion reduced the 4. Discussion numbers of VIP-ir neurons in the DRG after injury. In sympathetic neurons of the superior cervical ganglion, The data presented here demonstrate that specific neuro- NGF increases the expression of TH and but does not alter trophins (NGF and NT-3) influence the numbers of VIP [19,39].

neurons labeled for TH and VIP in nodose / petrosal The mechanism through which NGF alters the numbers ganglia cultures. That the effects are mediated by actions of TH and VIP neurons is not known. One possibility is of the neurotrophin is verified by the ability of the that NGF is acting directly through TrkA receptors on precipitating antibodies to prevent or attenuate the actions specific cultured nodose and petrosal ganglion neurons. of the neurotrophins. The neurotrophin-induced changes Although we have preliminary evidence for the presence of are likely due to changes in the content of the neurochemi- TrkA mRNA in cultured NG / PG neurons (Zhuo, Verdier-cals within neurons and are unlikely to be due to altera- Pinard and Helke, unpublished data), we do not know if tions in neuronal survival. This is based on our data the TrkA mRNA containing neurons are those in which the demonstrating an absence of effect of any of the neuro- content of TH or VIP is altered. Moreover, we did not find trophins tested on the numbers of NF-200 immunoreactive TH-positive neurons of intact nodose and petrosal ganglia neurons present after 5 days in culture. Moreover, it was that co-expressed TrkA [9]. NGF is also a ligand for p75 previously shown that the mature visceral sensory neurons and nearly all nodose ganglion neurons contain p75 do not require exogenous neurotrophins for survival [32,34,36]. Thus, the roles of neuronal TrkA and p75, [11,15,16]. However, we cannot rule out the possibility remain to be defined in this system. Likewise, the possi-that specific survival effects on a small subset of neurons bility that a non-neuronal cell type (e.g., fibroblasts) expressing a specific agent contributes to an observed remaining in these neuronally-enriched cultures responds change in immunoreactive neurons. The proposed changes to the addition of NGF with the secretion of a factor that in content of immunoreactive agents within the neurons of secondarily alters neuronal phenotype requires additional the cultured nodose and petrosal ganglia may reflex studies.

neurotrophin-induced changes in TH and / or VIP mRNA Whereas NT-3 (up to 150 ng / ml) does not affect peptide content and subsequent synthesis of TH and / or VIP. expression in newborn DRG [23], we found that NT-3 did However, it is also possible that the neurotrophins either affect the numbers of VIP-ir neurons in the mature nodose / directly or indirectly affect post translational processing, petrosal ganglia cultures. In contrast, in DRG neurons and / or stability of the products within the neurons. NT-3 was shown to upregulate neuropeptide Y but not VIP In this study, exogenous NGF increased the TH-ir and in an in vivo model of neural injury where VIP was already decreased VIP-ir neurons in the nodose / petrosal ganglia significantly upregulated [27].

cultures. Given that in vivo NGF is a retrogradely trans- The opposite actions of NGF and NT-3 on the numbers ported target-derived neurotrophin, and that loss of contact of VIP-ir neurons are intriguing and potentially significant. with target (either through vagotomy or inhibition of If additional studies verify that these actions result from axonal transport in the cervical vagus nerve) results in a changes in VIP mRNA, it would represent directionally

opposed regulation of a gene (VIP) by two distinct References neurotrophins. This may be relevant to potentially distinct

effects of NGF versus NT-3 in specific therapeutic situa- [1] R.J. Contreras, R.M. Beckstead, R. Norgren, The central projections of the trigeminal, facial, glossopharyngeal and vagus nerves: and

tions, and will become more clear as the relative loss of, or

autoradiographic study in the rat, J. Auton. Nerv. Syst. 6 (1982)

need for, NT-3 versus NGF trophic support after specific

303–322.

types of perturbations (e.g., injury) to the neuronal

en-[2] J. Donnerer, R. Schuligoi, C. Stein, Increased content and transport

vironment. Perhaps locally derived (not target derived) _{of substance P and calcitonin gene-related peptide in sensory nerves} NT-3 is involved in the elevated VIP seen in vivo after innervating inflamed tissue: evidence for a regulatory function of

injury or inhibition of axonal transport. The induction of nerve growth factor in vivo, Neuroscience 49 (1992) 693–698. [3] M.E. Goldstein, S.B. House, H. Gainer, NF-L and peripherin

NT-3 mRNA was noted in non-neuronal cells of the vagus

immunoreactivities define distinct classes of rat sensory ganglion

nerve trunk immediately proximal and distal to a nerve

cells, J. Neurosci. Res. 30 (1991) 92–104.

lesion within 1 day after injury (Lee, Zhuo and Helke, _{[4] C.J. Helke, K.M. Adryan, J. Fedorowicz, H. Zhuo, J.S. Park, R.} submitted). In vivo this non-neuronally-derived NT-3 may _{Curtis, H.E. Radley, P.S. DiStefano, Axonal transport of} neuro-have access to the injured neuron and, coupled with the trophins by visceral afferent and efferent neurons of the vagus nerve

of the rat, J. Comp. Neurol. 393 (1998) 102–117.

loss of NGF, be involved in the elevation of neuronal VIP.

[5] C.J. Helke, A.J. Niederer, Studies on the coexistence of substance P

Neurotrophic influences on these visceral sensory

neu-with other putative transmitters in the nodose and petrosal ganglia,

rons are significant because of the importance of these

Synapse 5 (1990) 144–151.

neurons and the associated reflexes in maintaining homeo- _{[6] C.J. Helke, A. Rabchevsky, Axotomy alters putative} neurotrans-stasis of autonomic, respiratory, and endocrine visceral _{mitters in visceral sensory neurons of the nodose and petrosal} adaptive systems [25]. The functions of these neurons are ganglia, Brain Res. 551 (1991) 44–51.

[7] G.D. Housley, R.L. Martin-Body, N.J. Dawson, J.D. Sinclair, Brain

altered in chronic disease states such as hypertension

stem projections of the glossopharyngeal nerve and its carotid sinus

(chronic overloading of baroreceptors) and congestive

branch in the rat, Neuroscience 22 (1987) 237–250.

heart failure (chronic overloading of cardiac stretch

re-[8] F.-L. Huang, H. Zhuo, C. Sinclair, M.E. Goldstein, J.T. McCabe,

ceptors) [20,40]. Injury to visceral afferent nerves can _{C.J. Helke, Peripheral deafferentation alters calcitonin gene-related} occur in many ways, including trauma, tumors, and disease peptide mRNA in visceral sensory neurons of the nodose and

(e.g., diabetes mellitus, Guillain–Barre syndrome) petrosal ganglia, Mol. Brain Res. 22 (1994) 290–298.

[9] H. Ichikawa, C.J. Helke, The coexistence of TrkA with putative

[21,22,29]. Thus, a better understanding of the influences

transmitter agents and calcium-binding proteins in the vagal and

of neurotrophins on visceral afferent neurons may lead to

glossopharyngeal sensory neurons of the adult rat, Brain Res. 846

their use in the prevention and / or alleviation of clinical _{(1999) 268–273.}

problems associated with their dysfunction. [10] H. Kashiba, E. Senba, Y. Kawai, Y. Ueda, M. Tohyama, Axonal blockade induces the expression of vasoactive intestinal polypeptide and galanin in rat dorsal root ganglion neurons, Brain Res. 577 (1992) 19–28.

[11] H. Leal-Cardoso, G.M. Koschorke, G. Taylor, D. Weinreich,

Elec-5. Disclaimer _{trophysiological properties and chemosensitivity of acutely isolated}

nodose ganglion neurons of the rabbit, J. Auton. Nerv. Syst. 45

The opinions or assertions contained herein are the (1993) 29–39.

[12] R.M. Lindsay, Role of neurotrophins and trk receptors in the

private ones of the authors and are not to be construed as

development and maintenance of sensory neurons: an overview,

official or reflecting the views of the DoD or the USUHS.

Phil. Trans. Royal Soc. London 351 (1996) 365–373.

The experiments reported herein were conducted according _{[13] R.M. Lindsay, Therapeutic potential of the neurotrophins and} to the principles set forth in the ‘Guide for the Care and neurotrophin–CNTF combinations in peripheral neuropathies and

Use of Laboratory Animals’, Institute of Animal Re- motor neuron diseases, Ciba Found. Symp. 196 (1996) 39–48. [14] R.M. Lindsay, Y.A. Barde, A.M. Davies, H. Rohrer, Differences and

sources, National Research Council, DHEW Publ. No.

similarities in the neurotrophic growth factor requirements of

(NIH) 74-23.

sensory neurons derived from neural crest and neural placode, J. Cell Sci. 3 (Suppl.) (1985) 115–129.

[15] R.M. Lindsay, C.J. Evison, J. Winter, Culture of adult mammalian peripheral neurons, in: J. Chad, H. Wheal (Eds.), Cellular Neuro-biology – A Practical Approach, IRL Press, New York, 1991, pp.

Acknowledgements

3–17.

[16] R.M. Lindsay, A.J. Harmer, Nerve growth factor regulates

expres-We thank Drs. Ann Acheson (Regeneron, Tarrytown, _{sion of neuropeptide genes in adult sensory neurons, Nature} NY, USA) and Dr. James Coulombe (USUHS) for advice (London) 337 (1989) 362–364.

and guidance in setting up the cultures of mature nodose / [17] R.M. Lindsay, C. Lockett, J. Sternberg, J. Winter, Neuropeptide expression in cultures of adult sensory neurons: modulation of

petrosal ganglia neurons. Drs. Huang Zhuo and Clara

substance P and calcitonin gene-related peptide levels by nerve

Carobi (USUHS) are appreciated for their assistance with

growth factor, Neuroscience 33 (1989) 53–65.

and discussion of various aspects of the study. This work _{[18] R.M. Lindsay, S.J. Wiegand, C.A. Altar, P.S. DiStefano,} Neuro-was supported by NIH grants NS20991 and NS38845, and trophic factors: from molecule to man, TINS 17 (1994) 182–189.

regulation of neuronal gene expression by nerve growth factor, J. [32] V.M.K. Verge, J.-P. Merlio, J. Grondin, P. Ernfors, H. Persson, R.J. ¨

Cell Biol. 117 (1992) 135–141. Riopelle, T. Hokfelt, P.M. Richardson, Colocalization of NGF [20] J.W. McCubbin, C.M. Ferrario, Baroreceptor reflexes and hyperten- binding sites, trk mRNA, and low-affinity NGF receptor mRNA in sion, in: J. Genest et al. (Ed.), Hypertension, McGraw Hill, New primary sensory neurons: responses to injury and infusion of NGF,

York, 1977, pp. 128–133. J. Neurosci. 12 (1992) 4011–4022.

¨ [21] A.J. McDougall, J.G. McLeod, Autonomic neuropathy. I: Clinical [33] V.M.K. Verge, P.M. Richardson, Z. Wiesenfeld-Hallin, T. Hokfelt,

features, investigation, pathophysiology, and treatment, J. Neurol. Differential influence of nerve growth factor on neuropeptide

Sci. 137 (1996) 79–88. expression in vivo: a novel role in peptide suppression in adult

[22] A.J. McDougall, J.G. McLeod, Autonomic neuropathy. II: Specific sensory neurons, J. Neurosci. 15 (1995) 2081–2096.

neuropathies, J. Neurol. Sci. 138 (1996) 1–13. [34] C. Wetmore, L. Olson, Neuronal and nonneuronal expression of [23] P.K. Mulderry, Neuropeptide expression by newborn and adult rat neurotrophins and their receptors in sensory and sympathetic ganglia sensory neurons in culture: effects of nerve growth factor and other suggest new intercellular trophic interactions, J. Comp. Neurol. 353 neurotrophic factors, Neuroscience 59 (1994) 673–688. (1995) 143–159.

[24] P.K. Mulderry, R.M. Lindsay, Rat dorsal root ganglion neurons in [35] H. Zhuo, C.J. Helke, Presence and localization of neurotrophin culture express vasoactive intestinal polypeptide (VIP) independent- receptor tyrosine kinase (TrkA. TrkB, TrkC) mRNAs in visceral ly of nerve growth factor, Neurosci. Lett. 108 (1990) 314–320. afferent neurons of the nodose and petrosal ganglia, Mol. Brain Res. [25] A.S. Paintal, Vagal sensory receptors and their reflex effects, Physiol. 38 (1996) 63–70.

Rev. 53 (1973) 159–227. [36] H. Zhuo, H. Ichikawa, C.J. Helke, Neurochemistry of the nodose [26] M. Reimer, M. Kanje, Peripheral but not central axotomy promotes ganglion, Prog. Neurobiol. 52 (1997) 79–107.

axonal outgrowth and induces alterations in neuropeptide synthesis [37] H. Zhuo, A.C. Lewin, E.T. Phillips, C. Sinclair, C.J. Helke, in the nodose ganglion of the rat, Eur. J. Neurosci. 11 (1999) Inhibition of the axoplasmic transport in the vagus nerve alters the

3415–3423. numbers of neuropeptide and tyrosine hydroxylase

mRNA-con-[27] G.D. Sterne, R.A. Brown, C.J. Green, G. Terenghi, NT-3 modulates taining and immunoreactive visceral afferent neurons of the nodose NPY expression in primary sensory neurons following peripheral ganglion, Neuroscience 66 (1995) 175–187.

nerve injury, J. Anat. 193 (1998) 273–281. [38] H. Zhuo, C. Sinclair, C.J. Helke, Plasticity of tyrosine hydroxylase [28] H. Thoenen, P.V. Angeletti, R. Levi-Montalcini, R. Kettler, Selective and vasoactive intestinal peptide mRNAs in visceral afferent neu-induction by nerve growth factor of tyrosine hydroxylase and rons of the nodose ganglion upon axotomy-induced deafferentation, dopamine-B-hydroxylase in the rat superior cervical ganglia, Proc. Neuroscience 63 (1994) 617–626.

Natl. Acad. Sci. USA 68 (1971) 1598–1692. [39] R.E. Zigmond, S.H. Hyatt, C. Baldwin, X. Qu, Y. Sun, T.W. McCain, [29] P.K. Thomas, C.J. Mathias, Diseases of the ninth, tenth, eleventh, R.C. Schreiber, U. Vaidyanathan, Phenotypic plasticity in adult and twelfth cranial nerves, in: P.J. Dyck, P.D. Thomas (Eds.), sympathetic neurons: changes in neuropeptide expression in organ Peripheral Neuropathy, 3rd Edition, Saunders, Philadelphia, PA, culture, Proc. Natl. Acad. Sci. USA 89 (1992) 1507–1511.

1993, pp. 869–885. [40] I.H. Zucker, J.P. Gilmore, Atrial receptor modulation of renal

[30] D.R. Tomlinson, P. Fernyhough, L. Mohiuddin, J.D. Delcroix, M. function in heart failure, in: F.M. Abboud et al. (Ed.), Disturbances Malcangio, Neurotrophic factors – regulation of neuronal pheno- in Neurogenic Control of the Circulation, American Physiology type, Neurosci. Res. Commun. 21 (1997) 57–66. Society, Bethesda, MD, 1981, pp. 1–16.

[31] A. Torvik, Afferent connections to the sensory trigeminal nuclei, the nucleus of the solitary tract and adjacent structures, J. Comp. Neurol. 106 (1956) 51–141.