INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

DURING INFLAMMATORY REACTION of the airways, various autacoids (e.g., histamine, thromboxanes, and prostaglandins) are locally released by a variety of cells, including airway epithelia, leukocytes, and mast cells. PGE₂ is a particularly abundant prostanoid found in the lungs and airways during inflammation. For example, PGE₂ levels in the bronchoalveolar lavage of human patients can increase 2- to 10-fold during an asthmatic episode (18, 22, 28). Furthermore, in guinea pig nodose ganglia, ovalbumin sensitization can elicit a fivefold increase in the release of PGE₂ from mast cells within the ganglia (37).

Once released, PGE₂ binds to specific prostanoid receptors on target cells and is then rapidly removed by prostaglandin transporters on a variety of cells (33). To date, four subtypes of prostanoid receptors that exhibit the highest affinity to PGE₂ are known: EP₁, EP₂, EP₃, and EP₄ (10, 31). All are members of the superfamily of G protein-coupled receptors that possess a seven-transmembrane domain topology and are linked to heterotrimeric G protein molecules. The subtypes differ in their distribution in various tissues and signal transduction mechanism (31). The prostaglandin receptors EP₂, EP_3C, and EP₄ are coupled to G_s protein and provoke increases in cAMP (31). Studies in dorsal root ganglion nociceptive neurons have shown that the sensitizing effects of PGE₂ are mediated through a cAMP-protein kinase A signal transduction mechanism, suggesting that activation of the EP₂, EP_3C, or EP₄ receptor may be involved (12, 24, 29, 34).

The lungs and airways are extensively innervated by vagal sensory fibers that are sensitive to circulating autacoids and inhaled irritants. The foremost of these chemosensitive neurons innervating the respiratory tract is the C-fiber afferent (2). The somata of these neurons reside in the nodose and jugular ganglia and send projections to the nucleus tractus solitarius, thus relaying information from the lungs and airways to medullary respiratory centers. The primary function of these neurons under normal physiological conditions is protective: when the sensory terminals are activated by irritants, cholinergic and tachykininergic reflexes are initiated, resulting in bronchoconstriction, mucous secretion, and cough, which limit further infiltration of the irritants deeper into the airways. Under certain pathological states, however, these afferent nerves develop hypersensitivity, and the exaggerated reflex responses become deleterious to normal airway function. For example, during airway inflammatory reaction (e.g., in asthma), the airways become exquisitely sensitive, and irritants can elicit a far more severe bronchoconstriction and more copious mucous secretion.

A previous study in our laboratory showed that PGE₂ infusion markedly enhances the apneic response elicited by chemical stimuli in intact rats (26). A subsequent study demonstrated that PGE₂ increased the single-unit vagal C-fiber activity elicited by capsaicin and other chemical stimulants (16). These two studies clearly showed that PGE₂ enhances the sensitivity of vagal C fibers. However, the effects of PGE₂ on other types of cells (e.g., smooth muscles) in the lung are well documented, and whether PGE₂ acts directly on the afferent fiber is unclear. In addition, the subtype of prostanoid receptor mediating these potentiating effects is unknown. Therefore, the purpose of this study was to determine, in acutely isolated cultured neurons, whether PGE₂ could directly modulate the function of vagal bronchopulmonary C neurons. Furthermore, we hypothesized that selective activation of the EP₂ prostanoid receptor could sensitize the vagal pulmonary neurons; the availability of a selective agonist of the EP₂ receptor, butaprost, made it feasible to test this hypothesis (23).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Labeling respiratory vagal neurons with DiI. Sensory neurons innervating the lungs and airways were identified by retrograde labeling from the lungs by using the fluorescent tracer, 1,1'-dioctadecyl-3,3,3',3'-tetramethylindocarbocyanine perchlorate (DiI). Young adult (150-200 g) male Sprague-Dawley rats were anesthetized with an intraperitoneal injection of pentobarbital (50-60 mg/kg) and intubated with a polyethylene tube such that the tip rested in the trachea above the thoracic inlet. With the rat tilted head up at ~30°, DiI (0.25 mg/ml; 1% ethanol concentration) was initially dissolved and sonicated in ethanol, diluted in saline, and then instilled into the lungs (2 × 0.25 ml). After 7-11 days, an interval previously determined to be sufficient for the dye to diffuse to the cell body, the nodose and jugular ganglia were extracted.

Cell culture. The animals were anesthetized with halothane (100%) and immediately decapitated. The head was immersed in ice-cold Hanks' balanced salt solution. Within 10 min, nodose and jugular ganglia with attached nerve trunks were extracted under a dissecting microscope and placed in ice-cold DMEM-F12 solution.

Electrophysiological recording. Borosilicate glass was pulled by using a micropipette puller and fire polished to a tip resistance of 2-4 M. The micropipettes were coated with Sylgard (Dow Corning) to minimize pipette capacitance. The micropipette electrode was a silver-silver chloride electrode. The silver-silver chloride pellet reference electrode was placed in a dish of pipette solution, and a 1% agar salt bridge closed the circuit to the bath solution. The liquid junction potentials and agar bridge and bath potentials were symmetrical and produced opposing junction effects. Calculations for liquid junction potential were performed by using the junction potential module within the data-acquisition software, based on JPCalc software (5).

1

Experimental design. The experiments were carried out in two study series. The first series aimed to test whether PGE₂ pretreatment modulates the sensitivity of cultured pulmonary C neurons to various chemical and electrical stimuli. The second series aimed to determine the effect of selective activation of the EP₂ prostanoid receptor on the sensitivity of cultured C neurons. The stimulus was applied before and then after pretreatment of the cell with either PGE₂, butaprost, or vehicle for 5 min and after a 1-min washout. In some cells, the recovery responses were studied after washout of PGE₂ or butaprost for >20 min. Although each neuron was identified as originating from either the nodose or jugular ganglion, there was no detectable difference in the responses between the two populations; their data were pooled for statistical analysis.

Chemicals. All chemicals used were purchased from Sigma Chemical (St. Louis, MO) except DiI (Molecular Probes, Eugene, OR); pentobarbital sodium (Butler, Columbus, OH); halothane (Halocarbon Laboratories, River Edge, NJ); PGE₂ and butaprost (Cayman Chemicals, Ann Arbor, MI); and DMEM-F12, fetal bovine serum, NGF 2.5S, trypsin-EDTA, and penicillin/streptomycin solution (GIBCO/Invitrogen, Carlsbad, CA).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

DiI-labeled vagal bronchopulmonary neurons. The composite photomicrograph in Fig. 1 represents the distribution of DiI fluorescence in the respiratory tract and the selective labeling of pulmonary neurons cultured from the vagal ganglia. The airway epithelium of the trachea showed prominent DiI fluorescence (Fig. 1A), which was evenly distributed throughout the lung, including bronchioles and alveoli (Fig. 1B). Because the jugular and nodose ganglia innervate multiple visceral organs, we examined a variety of tissues in the nervous and other organ systems, besides the lungs and trachea, after they were fixated with 4% paraformaldehyde and sliced in ~50-µm sections, for possible contamination by the DiI labeling: spinal cord and dorsal root ganglia (thoracic and lumbar), heart, blood, kidney, liver, esophagus, stomach, and intestine. DiI fluorescence was found only in nodose (e.g., Fig. 1C) and jugular ganglia, lungs, and airways. Because DiI labeling was sequestered in the lungs and airways, the labeled nodose and jugular ganglion cells represented neurons selectively innervating the respiratory tract. We found that 10.6% of the cultured nodose ganglion cells and 2.9% of the jugular ganglion cells were labeled.

View larger version (79K): [in this window] [in a new window]   Fig. 1.   1,1'-Dioctadecyl-3,3,3',3'-tetramethylindocarbocyanine perchlorate (DiI) labeling of pulmonary tissue and vagal ganglia neurons innervating the lungs and airways. The composite photomicrograph depicts DiI fluorescent labeling of trachea (A), lung periphery (B), and cultured nodose ganglion neurons (C) in both transmitted light (left) and epifluorescence (right). C: the small-diameter neurons (e.g., arrow) were used for perforated patch-clamp recordings. The DiI fluorescence was imaged with a TRITC filter set. The scale bars are 50 (A), 200 (B), and 100 µm (C).

PGE₂ increases the sensitivity of cultured bronchopulmonary vagal sensory neurons. In voltage-clamp mode, in pulmonary nodose and jugular ganglion neurons, capsaicin in the range of 0.01-0.3 µM elicited incremental whole cell currents in a dose-dependent manner (e.g., Fig. 2A). PGE₂ pretreatment increased both the peak and the duration of the inward current response elicited by capsaicin; an example is shown in Fig. 2B. In this neuron, the potentiating effects of PGE₂ recovered within 15 min, and both the peak and duration of the response returned to control levels. Whereas the vehicle for PGE₂ had no effect on the cellular response to capsaicin (Fig. 2D), PGE₂ pretreatment increased the whole cell current density elicited by capsaicin over that of control in a separate group of neurons (Fig. 2D; control: 23.1 ± 7.5 pA/pF; PGE₂: 42.4 ± 14.8 pA/pF; P < 0.05; n = 7). In four of these neurons, we were able to maintain the gigaohm seal long enough to record a recovery from the PGE₂ pretreatment (control: 33.8 ± 9.4 pA/pF; recovery: 31.6 ± 8.2 pA/pF; P > 0.05). A representative sample of pulmonary vagal neurons displayed an average resting membrane potential of 60.8 ± 1.3 mV and a whole cell capacitance of 17.9 ± 1.0 pF (n = 30) under normal ionic conditions; the latter was directly proportional to the surface area of the cell membrane (15) and was measured electronically with the Axon 200B amplifier.

View larger version (12K): [in this window] [in a new window]   Fig. 2.   PGE2 pretreatment potentiates the whole cell inward current elicited by chemical stimuli. A: in voltage-clamp mode (Vhold = 60 mV), capsaicin (Cap) delivered at varying doses elicited whole cell inward currents in a dose-dependent manner in a small pulmonary jugular ganglion neuron (17.4 pF). Bar, duration of Cap challenge, 4 s; 10-min interval between these and all subsequent challenges. B: experimental record in a small pulmonary nodose ganglion neuron (14.4 pF) showing the response to Cap (0.1 µM, 4 s) was potentiated after PGE2 pretreatment (1 µM, 5 min; middle trace). In this neuron, the gigaohm seal was maintained long enough to record a recovery response (right trace; ~15 min after termination of PGE2 perfusion) that was similar to that of control (Con) levels (left trace). C: in a different pulmonary nodose neuron (14.6 pF), the same dose of PGE2 potentiated the inward current elicited by phenylbiguanide (PBG; 1 µM, 4 s; right trace). Recovery was not tested in this neuron. D: left: responses of individual neurons (lines) and group means (circles) of the whole cell current densities elicited by Cap (0.1 µM, 4 s) before (Con) and after PGE2 pretreatment (1 µM, 5 min). * Significantly different from Con (P < 0.05; n = 7). Right: pretreatment with the vehicle (Veh) for PGE2 did not alter the response to the same dose of Cap in a separate group of neurons (P > 0.05; n = 7).

View larger version (14K): [in this window] [in a new window]   Fig. 3.   PGE2 and butaprost (Buta) enhance the no. of Cap-evoked action potentials. A: experimental record in current-clamp mode shows that pretreatment with PGE2 (1 µM, 5 min) increased the no. of action potentials elicited by Cap (0.1 µM, 4 s; middle trace) in an isolated cultured pulmonary nodose ganglion neuron (18.0 pF). The response during recovery (right trace) returned toward that of Con (left trace) after ~2 h. B: pretreatment with Buta (3 µM, 5 min) also potentiated the response evoked by Cap (0.1 µM, 4 s; middle trace) over that of Con (left trace) in a pulmonary nodose neuron (19.4 pF). In this cell, the recovery response (right trace) returned toward Con levels after ~45 min. C: composite graph showing that the response to Cap (0.1 µM, 4 s) increased in 4 out of 5 pulmonary vagal ganglion neurons (lines) after PGE2 pretreatment (1 µM, 5 min) compared with that of Con. The responses in all 3 neurons tested for recovery (Rec) returned toward Con levels 30-120 min later. D: in 5 other pulmonary vagal ganglion neurons (lines), the response to the same dose of Cap increased after Buta pretreatment (3-10 µM, 5 min) in 4 of them; all 3 of the neurons tested for a Rec response returned completely to that of Con levels.

View larger version (20K): [in this window] [in a new window]   Fig. 4.   PGE2 sensitizes pulmonary vagal ganglion neurons to electrical stimulation. A: in current-clamp mode, incremental levels of injected current (40, 70, and 100 pA; bar, 280 ms) evoked incremental nos. of action potentials in a labeled nodose ganglion neuron (32.6 pF); 10 levels of current (10-100 pA, increase = 10 pA) were tested, but only responses to 3 are shown here to avoid clustering. After PGE2 pretreatment (1 µM, 5 min; middle), the identical stimulus evoked a greater no. of action potentials compared with that of Con (top); the response eventually recovered toward Con levels (bottom) within 1 h. B: averaged data in vagal pulmonary neurons showing that, after PGE2 pretreatment, there was a significant increase in the no. of action potentials evoked by all 3 current injection levels compared with that of Con. Dotted lines, responses of individual neurons; solid circles, group means. * Significantly different from Con (P < 0.05; n = 6).

View larger version (25K): [in this window] [in a new window]   Fig. 5.   PGE2 modulates voltage-sensitive channels in pulmonary vagal ganglion neurons. A: superimposed transient inward and sustained outward current traces elicited by voltage steps applied in 5-mV increments of 50-ms duration to test potentials between 80 and +75 mV. The whole cell currents were recorded from a pulmonary nodose ganglion neuron (28.4 pA; Vhold = 60 mV) in reduced-Na+ extracellular solution (see METHODS). The transient inward currents enclosed in the box (bottom) are expanded in B. B: comparison of the whole cell transient inward current traces recorded during Con (top) and after PGE2 pretreatment (1 µM, 5 min; bottom). C: current-voltage relationship illustrating modulation by the PGE2 pretreatment and its recovery. The plot was fitted with a modified Boltzmann equation (see METHODS). , Con; , after PGE2 pretreatment; , recovery. D: tabulated data showing that PGE2 pretreatment caused a hyperpolarizing shift in the half-activation potential (V1/2) and a steeper slope factor k. Gmax, maximal macroscopic conductance; Erev, reversal potential. * Significantly different from Con (P < 0.05; n = 4). Values are means ± SE.

Butaprost increases the sensitivity of cultured adult bronchopulmonary vagal sensory neurons. Butaprost pretreatment (3-10 µM) increased the whole cell current density elicited by capsaicin compared with that of control (Fig. 6B; control: 5.6 ± 7.6 pA/pF; butaprost: 7.7 ± 1.2 pA/pF; P < 0.05; n = 5). The response to capsaicin returned toward control levels during recovery in three of the neurons tested (e.g., Fig. 6A; control: 6.1 ± 1.2 pA/pF; recovery: 6.1 ± 1.5 pA/pF; P > 0.05). In current-clamp mode, the same dose of butaprost also significantly increased the number of action potentials elicited by capsaicin in five other cells tested (P < 0.05; Fig. 3, B and D). A complete recovery was recorded in all three cells tested.

View larger version (7K): [in this window] [in a new window]   Fig. 6.   Buta enhances the whole cell inward current elicited by Cap. A: selective activation of the EP2 prostanoid receptor by Buta (3 µM; 5 min) caused an increase in the whole cell current response to Cap (0.1 µM, 4 s; Vhold = 60 mV; middle trace) in a pulmonary nodose ganglion neuron (26.6 pF). The response during recovery (right trace) was similar to that of Con (left trace). B: group data from both nodose and jugular ganglion neurons showing the significant increase in the Cap-evoked whole cell current density after Buta pretreatment (3-10 µM, 5 min) compared with that of Con. Lines, responses from individual neurons; , group means. * Significantly different from Con (P < 0.05; n = 5).

View larger version (22K): [in this window] [in a new window]   Fig. 7.   Buta sensitizes pulmonary vagal ganglion neurons to electrical stimulation. A: with the use of the identical current injection protocol as in Fig. 4, Buta pretreatment (3 µM, 5 min) increased the no. of action potentials evoked at the 2 higher levels of current injection (bar, 280 ms; bottom) compared with that of Con (top) in a pulmonary jugular neuron (12.1 pF). Recovery was not tested in this neuron. B: group data reveal that, after Buta pretreatment (3-10 µM, 5 min), the no. of action potentials elicited by the medium current injection (70 pA) increased compared with that of Con in vagal pulmonary neurons. Dotted lines, responses of individual neurons; solid circles, group means. * Significantly different from Con (P < 0.05; n = 10). See Fig. 4 legend for further explanation.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Our study showed that, in cultured adult rat pulmonary vagal C neurons, PGE₂ potentiates the response to chemical and electrical stimuli. This potentiation is probably mediated, at least in part, through the EP₂ prostanoid receptor. Our results clearly demonstrate that PGE₂ and butaprost exert direct sensitizing effects on cultured vagal bronchopulmonary C neurons. Specifically, in voltage-clamp mode, PGE₂ increased the whole cell current density evoked by capsaicin. In current-clamp mode, PGE₂ increased the number of action potentials elicited by capsaicin and current injection. Very similar responses were observed when the EP₂ prostanoid receptor was selectively activated by butaprost. Together, these data suggest that activation of the EP₂ receptor, either by PGE₂ or butaprost, has the ability to modulate the function of both ligand-gated and voltage-gated ion channels in isolated pulmonary chemosensitive neurons.

To our knowledge, this is the first report of the direct sensitizing effects of PGE₂ and butaprost in isolated vagal neurons innervating the lungs and airways. Our results are consistent with and build on previous experiments that showed PGE₂-induced potentiation of the pulmonary chemoreflex response and the single-unit C-fiber sensitivity (16, 26). In those studies in intact animals, we were unable to determine whether PGE₂ was acting on C fibers directly or through an intermediary effect on other cells (e.g., degranulation of mast cells). We should point out that these direct and indirect actions of PGE₂ are not mutually exclusive. In addition, we are making electrophysiological recordings from the perikaryon with the assumption that the channels, receptors, and intracellular signaling mechanisms present on the nerve terminals are also present on the cell body. However, it is possible that functional expression of these proteins is not present or is present at a significantly lower density in the soma. Therefore, the current densities elicited by capsaicin recorded at the cell bodies may not exactly match those elicited at the terminals. Furthermore, the magnitude of elicited currents that would have evoked action potentials at the sensory terminal may not evoke them at the cell soma. In addition, because we were recording the responses of these neurons at the temperature and partial pressures of O₂ and CO₂ of room air, the magnitude of the responses may not be the same as that of in vivo neurons. Despite these potential limitations, our data demonstrate expression of functional capsaicin, 5-HT₃, and EP₂ receptors on the cell body and a distinct sensitizing effect of PGE₂ on the isolated neurons.

The fact that butaprost also enhances the excitability of pulmonary C neurons (Figs. 3, 6, and 7), in a manner similar to that observed with PGE₂, suggests that PGE₂ may act through the EP₂ receptor in these cultured neurons. Butaprost is highly selective for the EP₂ prostanoid receptor subtype and does not bind to EP₁, EP₃, or EP₄ prostanoid receptors at the concentrations used in this study (23). However, definitive tests ultimately confirming the involvement of the EP₂ receptor in pulmonary C neurons require the use of a selective EP₂-receptor antagonist, which is, however, presently not yet available. Whether other prostanoid-receptor subtypes (either singly or in combination) play a role in modulating pulmonary C neurons is not known. It is possible that other members of the EP_x family could also modulate the sensitivity of these neurons, such as that suggested in dorsal root ganglion neurons with the EP_3C and EP₄ receptors (35). Additionally, perhaps other classes of prostanoid receptors coexpress and interact. For example, the prostacyclin receptor (IP) mRNA is coexpressed with EP₁, EP₃, or EP₄ mRNA in some dorsal root ganglion neurons (32). The potential contribution of other EP_x receptors in modulation of pulmonary C neurons should be further explored when selective pharmacological agents become available.

This study revealed that activation of the EP₂ receptor can modulate the function of a capsaicin receptor, the 5-HT₃ receptor, and one or more voltage-gated ion channels. The effects of EP₂-receptor activation are thought to be mediated through activation of protein kinase A. In a parallel study measuring calcium transients with fura 2 in cultured vagal chemosensitive neurons, Gu et al. (13) found that forskolin mimicked the potentiating effects of PGE₂, whereas H89, a membrane-permeant inhibitor of protein kinase A, inhibited them. Possible end-targets of EP₂ activation include type 1 vanilloid receptor, which has been identified on nodose neurons and has several predicted protein kinase A phosphorylation sites; TTX-resistant and TTX-sensitive sodium channels (e.g., SNS/PN3); calcium-dependent potassium currents (e.g., SK); and hyperpolarization-activated cation current (3, 6, 14, 20, 21, 38). It is likely that activation of the EP₂ receptor by PGE₂ results in the protein kinase A-mediated phosphorylation of both ligand-gated and voltage-sensitive channels, thus increasing their probability of opening (15).

Our results clearly demonstrate the labeling of neurons innervating the bronchopulmonary region by DiI; we did not detect the fluorescent dye in any of the tissues outside of the pulmonary regions. In contrast, the dye coverage throughout the respiratory tree was extensive: we found labeling from trachea to alveoli and in all lobes of the lung. When pooled, both vagal ganglia showed 7.3% of the cells labeled. This figure is slightly lower than that reported in the cat, which revealed that ~15% of vagal afferent fibers innervate the respiratory tree (2), and higher than the ~4% of nodose ganglion neurons that innervate the guinea pig trachea (8). The discrepancies could be due to species difference, or, alternatively, it is possible that part of the lungs and nerve endings may not have been exposed to a sufficient amount of DiI, and, therefore, the number of pulmonary neurons is underestimated. When we analyzed the nodose and jugular ganglion separately, we were surprised to find a fivefold increase in the percentage of labeled nodose neurons vs. jugular neurons. In contrast, Hunter and Undem (19) found that the guinea pig tracheal epithelium, when labeled with rhodamine microspheres, receives sensory innervation almost exclusively from the jugular ganglia. That study found that, when fast blue dye was used, which is able to penetrate and label nerves innervating deeper structures (e.g., submucosa), the dye was more evenly distributed between the two ganglia. It is also possible that, in the rat, the more distal regions of the respiratory tree receive innervation from a larger percentage of nodose ganglion neurons relative to that of the jugular ganglion neurons. Therefore, the differences between our two studies could be due to a combination of species difference and a difference in the distribution of sensory innervation between trachea and lung periphery. It must be noted that, despite these quantitative differences in labeling percentage, the precision with which respiratory structures are identified remains exceedingly high in this study.

In these experiments, we selected small-diameter neurons for study. It has been shown that the smaller diameter cultured isolated neurons correspond to the substance P-containing, neurofilament-negative, chemosensitive C fibers (9, 39). However, we cannot rule out the possibility that a fraction of the chemosensitive cells that we studied is in fact A-neurons. A-afferents are a heterogeneous population of thin myelinated fibers, a subset of which is characterized by their polymodal sensitivities and chemosensitive properties (17, 25). Although these A-neurons share some similarities in their physiological properties with C neurons, the proportion of capsaicin-sensitive pulmonary A-fibers to C fibers is miniscule in the rat, according to the study by Ho and coworkers (17). Therefore, although A-neurons may be present in the population of small chemosensitive neurons that we studied, the vast majority of them are C neurons.

The pulmonary C-fiber afferents exhibit polymodal sensitivities: known stimulators include capsaicin, bradykinin, histamine, hypertonic saline, adenosine, adenosine triphosphate, acid, heat, wood smoke, and mechanical stimuli (11, 27). Indeed, receptors for these compounds have been identified on various C-fiber neurons across species, including ferret, rabbit, guinea pig, rat, and humans (11, 27). These C-fiber terminals have extensive innervation throughout the respiratory tree. A dense plexus of nerve terminals exists below and between the lung and airway epithelial cells, in close proximity to the airway lumen (1, 4). From such a vantage point, these nerve terminals are able to monitor the airways for inhaled irritants as well as local and circulating autacoids. During inflammation, many circulating and locally released autacoids, particularly PGE₂, sensitize these C fibers, which lead to an exaggerated bronchopulmonary reflex response. In healthy human subjects, inhalation of aerosolized PGE₂ elicits the sensation of dyspnea during exercise and enhances the cough reflex elicited by capsaicin and phenylbiguanide (7, 36). Results of this study have established direct evidence of the PGE₂-induced hypersensitivity of bronchopulmonary C neurons, lending further support to these previous studies.

In summary, data from this study clearly show that PGE₂ enhances the excitability of cultured pulmonary vagal chemosensitive neurons to both chemical and electrical stimuli. This study has, therefore, provided direct evidence in support of the previous finding of a sensitizing effect of PGE₂ on vagal pulmonary C-fiber afferents in vivo. These results further reveal that the sensitizing effect of PGE₂ is probably mediated, at least in part, through the activation of the EP₂ prostanoid receptor, which, in turn, modulates the functions of a variety of ligand-gated and voltage-sensitive ion channels on the neuronal somata.