Skip to main content
PMC home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2001 Dec 15;537(Pt 3):667–677. doi: 10.1111/j.1469-7793.2001.00667.x

Expression of P2X2 and P2X3 receptor subunits in rat carotid body afferent neurones: role in chemosensory signalling

Mona Prasad 1, Ian M Fearon 1, Min Zhang 1, Michael Laing 1, Cathy Vollmer 1, Colin A Nurse 1
PMCID: PMC2278999  PMID: 11744746

Abstract

  1. Hypoxic chemotransmission in the rat carotid body (CB) is mediated in part by ATP acting on suramin-sensitive P2X purinoceptors. Here, we use RT-PCR, cloning and sequencing techniques to show P2X2 and P2X3 receptor expression in petrosal neurones, some of which develop functional chemosensory units with CB receptor clusters in co-culture.

  2. Single-cell RT-PCR revealed that hypoxia-responsive neurones, identified electrophysiologically in co-culture, expressed both P2X2 and P2X3 mRNA.

  3. Isohydric hypercapnia (10% CO2; pH 7.4) caused excitation of chemosensory units in co-culture. This excitation depended on chemical transmission, with ATP acting as a co-transmitter, since it was inhibited by reduction of the extracellular Ca2+:Mg2+ ratio and by the purinoceptor blocker suramin (50–100 μm).

  4. Hypoxia and isohydric hypercapnia could separately excite the same chemosensory unit, and together the two stimuli interacted synergistically.

  5. Using confocal immunofluorescence, co-localization of P2X2 and P2X3 protein was demonstrated in many petrosal somas and CB afferent terminals in situ. Taken together, these data indicate that ATP and P2X2–P2X3 purinoceptors play important roles in the peripheral control of respiration by carotid body chemoreceptors.

Over the past several years purinergic receptors have occupied centre stage as important mediators of a broad range of intercellular signals in various tissues (Burnstock, 1997; North & Barnard, 1997). These receptors are composed of two main groups: the ionotropic P2X receptors and the metabotropic (G-protein coupled) P2Y receptors. In the nervous system, purinergic excitatory synapses use ATP to mediate fast synaptic transmission via activation of P2X receptor cation channels (Surprenant et al. 1995). Gene cloning procedures have revealed that P2X receptors are members of a multigene family of ATP-gated ion channels and consist of at least seven receptor subunits: P2X1-P2X7 (Buell et al. 1996; Collo et al. 1996). In neural tissues, the expression of P2X2, P2X3, P2X4 and P2X6 subunits is especially common (Lewis et al. 1995; Vulchanova et al. 1997; Khakh et al. 1999), and though homomeric channels can account for ATP responses in some cases, heteropolymerization of P2X subunits may occur in both the central and peripheral nervous system (Lewis et al. 1995; Collo et al. 1996; Torres et al. 1999). To date four functional heteromultimers including P2X2-P2X3, P2X4-P2X6, P2X1-P2X5 and P2X2-P2X6 have been identified (King et al. 2000).

Recent work in this laboratory demonstrated a role for ATP as a co-transmitter during hypoxic signalling in a peripheral chemosensory organ: the rat carotid body (Zhang et al. 2000). This organ helps maintain blood homeostasis via the initiation of corrective alterations in ventilation following changes in partial pressure of blood gases (PO2 and PCO2) and pH (Gonzalez et al. 1994). Our studies utilized a co-culture model in which clusters of carotid body chemoreceptors (glomus or type 1 cells) developed functional synaptic connections with dissociated neurones from the petrosal ganglion, which supplies the main chemoafferent innervation to the organ. With this model of a chemosensory unit we showed that a hypoxic stimulus was transduced at the level of the type 1 cells, and that the information was transmitted to nearby petrosal neurones via co-release of ATP and ACh (Zhang et al. 2000). The purinergic receptors mediating the ATP component of the response are likely to be P2X2-P2X3 heteromultimers (Zhang et al. 2000). This hypothesis is based largely on the observation that physiologically identified hypoxia-responsive neurones in co-culture show slow desensitization kinetics following ATP application, and are activated by α,β-methylene ATP (α,β-meATP). In addition, immunoreactive P2X2 subunits are localized to petrosal chemoafferent neurones and carotid body nerve endings in situ (Zhang et al. 2000).

In the present study we extended the use of the co-culture model, combined with RT-PCR, electrophysiological and confocal immunofluorescence techniques, to test for co-expression of both P2X2 and P2X3 subunits in hypoxia-responsive chemosensory neurones. In addition, we addressed whether ATP also acted as a co-transmitter during transmission of other carotid body stimuli, e.g. elevated PCO2 or hypercapnia. Furthermore, we examined whether or not an individual chemosensory unit could process both sensory modalities, hypoxia and hypercapnia, and if so, whether the two stimuli interacted synergistically when applied together (Lahiri & DeLaney, 1975; Pepper et al. 1995). Recent studies showed that ATP receptors, probably P2X2-P2X6 heteromultimers, were the mediators of central respiratory drive in the rat during both CO2- and pH-dependent changes in respiration (King et al. 2000; Thomas & Spyer, 2000). However, it remains to be demonstrated whether or not ATP and P2X receptors play a similar role in the periphery, during hypercapnic chemoexcitation of the carotid body. This study revealed that both P2X2 and P2X3 subunits were co-expressed in individual chemosensory neurones, and contributed to the ATP-mediated chemoexcitation of the afferent fibres following exposure of carotid body receptors to both hypoxia and hypercapnia.

METHODS

Reverse transcriptase-polymerase chain reaction (RT-PCR)

Total RNA was obtained from between 50 and 150 mechanically isolated neurones following combined enzymatic and mechanical dissociation of rat petrosal ganglia as previously described (Zhong et al. 1997; Zhang et al. 2000). Postnatal rat pups (Wistar; Charles River, Quebec, Canada), 9-13 days old, were first rendered unconscious by a blow to the back of the head (produced by rapid deceleration) and killed immediately by decapitation. This was followed by bilateral excision of the carotid bifurcations and dissection of the petrosal ganglia. All procedures were carried out according to the guidelines of the Canadian Council on Animal Care (CCAC). After dissociation of the tissue, isolation of individual neurones was carried out under a dissecting microscope with the aid of a glass microelectrode (tip diameter ∼40-60 μm; Flaming/Brown Micropipette Puller, Model P-97; Sutter Instruments Co., Novato, CA, USA), to which suction was applied via an electrode holder and attached polyethylene tubing. The isolated neurones were pooled and total RNA extracted using the Qiagen RNeasy kit (Qiagen Inc., Missisauga, Ontario, Canada) according to the manufacturer's instructions. To remove contaminating genomic DNA, isolated RNA was on-column treated with RNase-free DNase (Qiagen) or treated with 0.1 U μl−1 RQ1 DNase (Promega Corporation, Madison, WI, USA) for 30 min at 37 °C prior to the RT reaction.

RNA was reverse transcribed using Superscript II reverse transcriptase (Gibco BRL Life Technologies, Burlington, Ontario, Canada). The RT mixture consisted of the following (mm unless stated): 50 Tris-HCl, 75 KCl, 3 MgCl2, 0.025 μg μl−1 oligo(dT)12-18, 20 DTT, 0.5 each deoxyribonucleoside triphosphate (dNTP), 2 U μl−1 RNaseout ribonuclease inhibitor (Gibco BRL Life Technologies) and 10 U μl−1 reverse transcriptase. The reaction mixture had a total volume of 20 μl. Reactions were run at 42 °C for 60 min then at 70 °C for 15 min in a GeneAmp PCR System 9700 thermal cycler (Perkin-Elmer, Norwalk, CT, USA). Reaction products were stored at 4 °C prior to running the PCR reaction. Control reactions were performed where RT was omitted from the reaction mix.

Following reverse transcription, DNA was amplified in a single round of PCR. The PCR reaction consisted of the following (mm unless stated): 20 Tris-HCl, 50 KCl, 1.5 MgCl2, 0.2 each dNTP, 0.2 μm each upstream and downstream primers, 5 μl DNA template and 2.5 U μl−1 Platinum Taq polymerase (Gibco BRL Life Technologies). Total reaction volume was 25 μl with DNase-free water. Gene-specific primers were designed using Jellyfish software (Biowire.com, Mountain View, CA, USA) and synthesized by The Central Facility of the Institute for Molecular Biology and Biotechnology (MOBIX), McMaster University, Hamilton, Ontario, Canada. The following primers were used and are listed as sequence amplified, 5′ to 3′ upstream primer sequence, 5′ to 3′ downstream primer sequence, and size of product(s) (bp):

graphic file with name tjp0537-0667-mu1.jpg

P2X2 primers were designed against a region common to multiple splice variants and thus identify multiple isoforms of this receptor. The PCR was held at 94 °C for 2 min and cycled 35 times. Each cycle consisted of 94 °C for 30 s (denaturation), 55 °C (P2X3 and β-actin) or 52 °C (P2X2) for 30 s (annealing), and 72 °C for 1 min (extension). This was followed by a 10 min final extension at 72 °C. PCR products were visualized on an ethidium bromide-stained 2 % agarose gel under ultraviolet (UV) illumination. The 100 bp DNA ladder, used as markers (M), was obtained from Gibco BRL Life Technologies.

Cloning and sequencing of P2X2 and P2X3 PCR products

The QIAquick Gel Extraction Kit (Qiagen Inc.) was used to extract PCR fragments from the agarose gel according to the manufacturer's instructions. The extracted DNA band was eluted with sterile RNase/DNase-free water. The putative P2X DNA fragments were ligated into a pCR 4-TOPO vector (Invitrogen Corp., Carlsbad, CA, USA) and transformed into the competent TOPO 10 cells provided by the TOPO-TA cloning kit for sequencing. The transformed cells were plated onto LB/Ampicillin plates and incubated overnight at 37 °C. Isolated colonies were picked from the plates and grown in 5 ml of LB/Amp media overnight on a 37 °C shaker. Mini Prep Alkaline Lysis protocol (Sambrook et al. 1989) and an isopropanol precipitation reaction were used to extract DNA from 1 ml of cultured cells. The plasmid DNA from several colonies was analysed by restriction enzyme digest (1 h at 37 °C) with EcoR1 (Gibco BRL Life Technologies), and the bands were separated on a 1.2 % agarose gel. Positive colonies containing the correct insert size were cultured in 500 ml of LB/Amp and the plasmid DNA was purified by a Maxi Prep Alkaline Lysis protocol (Sambrook et al. 1989) followed by equilibrium centrifugation in a CsCl-ethidium bromide gradient. The DNA sample was then sequenced (MOBIX, McMaster University, Hamilton, Ontario, Canada) using an ABI Prism automated sequencer (with T7 polymerase). The sequencing results were analysed by BLAST 2, an NIH computer software for identification of gene sequences. The sequences were matched to the Rattus norvegicus P2X2 and P2X3 receptor mRNA coding sequences (Genbank accession numbers U14414 and X91167, respectively).

Single-cell RT-PCR on identified neurones

Chemosensory units were identified in co-culture by conventional whole-cell recording (see following) from individual petrosal neurones juxtaposed to a type 1 cell cluster. To remove RNases, patch pipettes were heated overnight at 200 °C. The pipette solution was autoclaved before use, prior to the addition of 5 units μl−1 RNaseout ribonuclease inhibitor (Gibco BRL Life Technologies). After identifying a ‘functional’ petrosal neurone in co-culture, i.e. one which depolarized and/or fired action potentials during a brief exposure to hypoxia (PO2∼5 mmHg), the cell contents were aspirated under visual control using a syringe attached to the recording patch pipette. The gigaseal remained intact during aspiration of cell contents, which were then expelled into a 0.5 ml PCR tube containing the RT mixture. Typically, the volume of intracellular recording solution in the patch pipette was 20 μl. Negative controls were performed by aspirating a similar volume of extracellular perfusate from above the cells in the recording chamber and the sample obtained was similarly processed. RT was carried out as described previously, except the total reaction volume was 40 μl. PCR was also carried out as described previously except the reaction was cycled 40 times. Total reaction volume was made up to 50 μl with RNase/DNase-free water.

Electrophysiological recording

The procedures for recording chemosensory responses from co-cultured petrosal neurones were similar to those described in detail elsewhere (Zhong et al. 1997; Zhang et al. 2000). The selection of neurones that were juxtaposed within a few micrometres to a type 1 cluster allowed for the recording of both subthreshold and/or suprathreshold postsynaptic responses during chemosensory stimulation. The nystatin perforated-patch, whole-cell technique, which preserves cytoplasmic integrity (Horn & Marty, 1988), was used when long-term stable recordings were desirable, e.g. during pharmacological interventions or during testing of the same neurone for responsiveness to more than one stimulus (e.g. hypoxia and hypercapnia). In these experiments the patch pipette contained intracellular solution plus 150-300 μg ml−1 nystatin (Stea & Nurse, 1991; Zhong et al. 1997). However, for single-cell RT-PCR experiments (see previously), where there was exchange of cell contents and pipette solution, nystatin was omitted and conventional whole recording was used to test for hypoxic chemotransmission. For control conditions, recordings were carried out at ∼35 °C in bicarbonate/CO2 buffered extracellular fluid of the following composition (mm): 115 NaCl, 24 NaHCO3, 5 KCl, 2 CaCl2, 1 MgCl2, 10 glucose and 12 sucrose, at pH 7.4 maintained by bubbling 95 % air-5 % CO2. Hypoxia (PO2∼5 mmHg) was generated by bubbling nitrogen into the extracellular fluid, whereas for (isohydric) hypercapnia, a 10 % CO2 gas mixture was bubbled into a modified extracellular solution containing 91 mm NaCl and 48 mm NaHCO3 to maintain extracellular pH constant at 7.4. Both hypoxic and hypercapnic stimuli were applied with the aid of a rapid perfusion system utilizing a double-barrelled pipette as described previously (Zhong et al. 1997). Data from electrophysiological experiments were analysed by one-way ANOVA.

Confocal immunofluorescence

Rat pups (2-3 weeks old) were first anaesthetized by intraperitoneal administration of Somnotol (65 mg kg−1), before perfusion via the aorta with phosphate-buffered saline (PBS) followed by PBS containing 4 % paraformaldehyde. The carotid bifurcations were excised bilaterally and immersed in 4 % paraformaldehyde for 1 h at room temperature. Following incubation overnight in 20 % sucrose at 4 °C, sections (thickness, ∼18 μm) were cut in a cryostat and collected on glass slides coated with 2 % silane (Sigma Chemical Co., St Louis, MO, USA). After air drying onto glass slides, the sections were washed (3 × 5 min each) in PBS before exposure to a blocking solution containing 2 % bovine serum albumin (BSA) in PBS (BSA-PBS) for ∼45 min at room temperature, followed by overnight incubation at 4 °C in a cocktail of the primary antisera. The primary antibodies were: (i) anti-P2X2 (1:800 dilution), a rabbit polyclonal antibody raised against a highly purified peptide corresponding to amino acid residues 457-472 of rat P2X2 (Alomone Laboratories, Jerusalem, Israel); and (ii) anti-P2X3 (1:500 dilution), a guinea-pig polyclonal antiserum raised against amino acid residues 383-397 of rat P2X3 (Neuromics, Minneapolis, MN, USA). The secondary antibodies (Jackson Immunoresearch Laboratories, West Grove, PA, USA) consisted of N-hydroxysuccinimidyl fluorophore (Cy3)-conjugated goat anti-rabbit IgG (1:500 dilution) and fluorescein isothiocyanate (FITC)-conjugated goat anti-guinea-pig IgG (1:20 dilution), respectively. Secondary antibodies were diluted in PBS containing 1 % BSA, 10 % normal goat serum and 0.3 % Triton X-100. Samples were covered with Vectashield Mounting Medium (Vector Laboratories, Burlington, Ontario, Canada) before viewing under a Bio-Rad Microradiance 2000 confocal microscope, equipped with argon (two lines, 488 and 514 nm) and helium/neon (543 nm). Lasersharp software was used for image acquisition. Digitized images of positive immunofluorescence were taken at 1 μm step intervals following excitation of the two fluorochromes. In control experiments pre-incubation of the primary P2X2 and the P2X3 antisera for 1 h with excess of each of their blocking peptides resulted in complete abolition of staining.

RESULTS

RT-PCR identification of P2X mRNA in isolated petrosal neurones

The RT-PCR technique was employed in order to establish whether or not petrosal neurones expressed mRNAs encoding both the P2X2 and P2X3 subunits. In each experimental series, a relatively pure neuronal population was obtained by isolating 50-150 individual neurones as described in Methods. A combination of RT-PCR, cloning and sequencing techniques was used to identify the target sequences for P2X2 and P2X3 subunits. RT-PCR was employed utilizing gene-specific primers for P2X2 and P2X3 (see Methods) to amplify products with sizes of 357 bp and 326 bp, the expected sizes for their respective target sequences (Fig. 1). In each case β-actin was amplified to provide a positive control; its product gave a single band of 327 bp (Fig. 1). To confirm that the RNA samples were devoid of contaminating genomic DNA, negative control RT-PCR reactions were carried out by omitting the reverse transcriptase enzyme (Fig. 1). The RT-PCR reactions were carried out on samples isolated from five separate preparations of P9-P13 petrosal neurones, and in each case results similar to those displayed in Fig. 1 were obtained.

Figure 1. Detection of mRNA for P2X2, P2X3 and β-actin in isolated petrosal neurones.

Figure 1

Open in a new tab

RT-PCR was carried out on groups of isolated petrosal neurones using gene-specific primers for P2X2 and P2X3 receptors, and β-actin. Expected product sizes are (bp): P2X2, 357; P2X3, 326; β actin, 327. The marker lane (M) shows bands at 100 bp increments with the 600 bp fragment at increased intensity. In negative control reactions without RT (-) no PCR products were observed. PCR products were visualized with 2 % agarose gel stained with ethidium bromide and viewed under UV illumination.

The putative P2X2 and P2X3 target sequences amplified by RT-PCR were extracted from agarose gels, ligated into a pCR 4-TOPO vector, and subsequently sequenced as described in Methods. BLAST analysis of the sequence data confirmed that the cloned P2X2 and P2X3 subunit PCR products were 100 % identical to published known P2X2 and P2X3 subunit sequences.

Single-cell RT-PCR on physiologically identified neurones

The above data showing expression of P2X2 and P2X3 mRNA in populations of 50-150 petrosal neurones did not address the issue as to whether or not single carotid body (CB) chemoafferents expressed mRNA for both receptors, as expected if heteromultimeric P2X2-P2X3 receptors are indeed involved in synaptic transmission (Zhang et al. 2000). In order to clarify this issue, we used co-cultures of petrosal neurones and clusters of CB chemoreceptors (type 1 or glomus cells), combined with whole-cell recording, to identify single neurones that were members of a functional chemosensory unit, i.e. those that depolarized and/or increased their firing rate during application of a hypoxic stimulus (PO2∼5 mmHg). Single-cell RT-PCR analysis was subsequently carried out on the same functional neurone to test for the expression of both P2X2 and P2X3 subunits. In the present experiments, individual neurones were tested within 3-5 days after plating on a pre-existing monolayer of CB cells.

A phase-contrast micrograph showing the typical arrangement of a type 1 or glomus cell (GC) cluster and a ‘juxtaposed’ petrosal neurone (PN) is shown in Fig. 2A. This unit configuration, as opposed to others where the neurone was situated further away, was representative of those used in the majority of experiments. Such a configuration increased the probability of detecting functional connections as well as subthreshold postsynaptic events (Zhong et al. 1997; Nurse & Zhang, 1999; Zhang et al. 2000). In each of seven cases, single-cell RT-PCR on individual hypoxia-responsive neurones revealed that both the P2X2 and P2X3 mRNAs were expressed. A typical example is shown in Fig. 2, where a ‘juxtaposed’ neurone in co-culture was studied electrophysiologically before its cytoplasmic contents were analysed by RT-PCR (see Methods section). In Fig. 2Ba, the hypoxic stimulus (PO2∼5 mmHg) produced a reversible subthreshold depolarisation in the neurone, attributable in part to co-release of ATP acting at suramin-sensitive P2X receptors (Zhang et al. 2000). Subsequently, PCR products of the expected sizes, which corresponded to the target sequences of P2X2 and P2X3, were detected in the same neurone (Fig. 2Bb).

Figure 2. Whole-cell current-clamp recording and single-cell RT-PCR products.

Figure 2

Open in a new tab

A, phase-contrast micrograph of co-cultured type 1 or glomus cell (GC) cluster and juxtaposed petrosal neurone (PN); this configuration was used to test for functional neurones in electrophysiological experiments. Scale bar represents 10 μm. Ba, whole-cell current-clamp recording from a single petrosal neurone juxtaposed to a type 1 cell cluster in co-culture (similar to A). For the period indicated by the horizontal bar, hypoxia (PO2∼ 5 mmHg) was applied by the switching of the extracellular perfusate for one equilibrated with 100 % N2. Following recording, the cell contents were aspirated and subjected to RT-PCR analysis as described in Methods. Bb, 2 % agarose gel showing RT-PCR products from the cell in Ba, typical of results obtained in seven cells examined. Expected product sizes are (bp): P2X2, 357; P2X3, 326; β-actin, 327. The marker lane (M) shows bands at 100 bp increments with the 600 bp fragment at increased intensity. Lane C shows a negative control where extracellular perfusate, aspirated from above the cells in the recording chamber, was subjected to RT-PCR analysis as described using primers for P2X2, P2X3 and β-actin.

PCO2 signalling is chemically mediated with ATP acting as a co-transmitter

The synaptic mechanisms mediating CB chemoexcitation during stimulation by increased PCO2 or hypercapnia are unknown. As in the case of hypoxia, however, increased PCO2 causes membrane depolarization, increased firing frequency and elevation of intracellular Ca2+ in type 1 cells, suggesting the latter are the sites of transduction (Buckler & Vaughan-Jones 1994a, b). In the present study, we used co-cultures to investigate for the first time whether or not hypercapnic chemotransmission was chemically mediated, and whether ATP also acted as a co-transmitter.

Isohydric hypercapnia (10 % CO2; pH 7.4) induced increased firing and/or membrane depolarization in several ‘postsynaptic’ petrosal neurones (n = 10; Fig. 3 and 4). The increased neuronal firing appeared to be the result of chemical transmission since it was reversibly abolished by reduction of the extracellular Ca2+:Mg2+ ratio (n = 4), as exemplified in Fig. 3. In the presence of normal extracellular Ca2+ (2 mm) and Mg2+ (1 mm), isohydric hypercapnia markedly increased firing in a co-cultured neurone that was spontaneously active (Fig. 3A). Substitution of the extracellular perfusate for one containing low Ca2+ (0.1 mm)-high Mg2+ (6 mm) resulted in the abolition of spontaneous firing, as well as the responses to isohydric hypercapnia (Fig. 3B). Moreover, in the same neurone, when the low Ca2+-high Mg2+ perfusate was present only during the time of application of the hypercapnic stimulus, the sensory response was completely abolished (Fig. 3C), and the effect was reversible (Fig. 3D).

Figure 3. Processing of hypercapnic stimuli by chemosensory units in co-culture is mediated by chemical transmission.

Figure 3

Open in a new tab

All recordings were obtained from a spontaneously active petrosal neurone (near a type 1 cell cluster) during perfusion of a normocapnic (5 % CO2; pH 7.4) solution, except during the period indicated by lower horizontal bars, when the hypercapnic stimulus (10 % CO2; pH 7.4) was applied by rapid perfusion. The initial increase in neuronal discharge during isohydric hypercapnia is shown in A, where the perfusion fluid contained normal extracellular Ca2+ (2 mm) and Mg2+ (1 mm) concentrations. The spontaneous firing and hypercapnic response was markedly inhibited or abolished when the perfusate was switched to one containing reduced extracellular Ca2+ (0.1 mm) and increased extracellular Mg2+ (6 mm) concentrations (B and C); in contrast to B, the low Ca2+-high Mg2+ solution was present only when the hypercapnic stimulus was present in C. In D, the hypercapnic response was restored after the manipulations in B and C. Resting potential = -74 mV.

Figure 4. Inhibition of the hypercapnic response in carotid body co-cultures by the P2X purinoceptor blocker suramin.

Figure 4

Open in a new tab

A, current-clamp recordings from the same petrosal neurone, juxtaposed to a type 1 cluster in co-culture, before and after switching from normocapnia (5 % CO2; pH 7.4) to isohydric hypercapnia (10 % CO2; pH 7.4; period indicated by upper horizontal bars), followed by a return to normocapnia. Note the stimulus-induced depolarization was partially inhibited by 100 μm suramin (middle trace) and the effect was reversible (right trace). In B, the recording conditions were similar to A, but the interaction was stronger with the hypercapnic stimulus giving rise to a burst of action potentials (left and right traces); 100 μm suramin markedly inhibited the hypercapnic response, abolishing all spike activity (middle trace). The resting potential was -61 mV in A and -54 mV in B, and the neurones were in co-culture for 3-4 days. C, mean (±s.e.m.) subthreshold depolarization induced by the hypercapnic stimulus for a group of six co-cultured neurones as in A, before, during and after (wash) perfusion with 100 μm suramin. The latter caused a significant reduction (∼65 %; **P < 0.005, one-way ANOVA) in the magnitude of the depolarization from 5.4 ± 1.3 to 1.9 ± 0.83 mV, suggesting ATP acted as a co-transmitter. D, spike frequency plotted before, during and after suramin for a different group of four co-cultured neurones that responded with a burst of spike activity during isohydric hypercapnia as in B; the spike discharge was markedly inhibited by suramin (***P < 0.001, ANOVA).

Consistent with the occurrence of chemical transmission during hypercapnic excitation, release of ATP from type 1 cells appeared to be involved since the postsynaptic response was reduced in a dose-dependent manner by the P2X purinoceptor blocker, suramin. For example, in cases where the postsynaptic response was subthreshold (n = 6; Fig. 4A), 100 μm suramin caused a significant reduction (∼65 %) in the magnitude of the postsynaptic depolarization, from 5.4 ± 1.3 to 1.9 ± 0.9 mV (P < 0.005, one-way ANOVA; Fig. 4C). In this same group of cells, 50 μm suramin caused a ∼48 % reduction in the magnitude of the postsynaptic response (2.8 ± 1.9 mV; P < 0.01). In other cases where the synaptic interaction was strong (n = 4), 10 % CO2 caused a marked increase in neuronal firing, which was reversibly inhibited by 100 μm suramin (Fig. 4B and D). A residual depolarization often persisted in neurones (juxtaposed to type 1 clusters) during application of 100 μm suramin (Fig. 4A and B; middle traces), a dose expected to block the effects of ATP on the P2X purinoceptors in these neurones (Zhang et al. 2000). Preliminary experiments indicated that this residual response was due largely to co-release of ACh (unpublished observations). Taken together, these data suggest that as in the case of hypoxia (Zhang et al.2000), ATP also acts as a co-transmitter during hypercapnic excitation of the rat carotid body.

Excitation of same chemosensory unit by hypoxia and hypercapnia

The fact that both hypoxic and hypercapnic stimuli could be transduced and transmitted in these co-cultures allowed us to investigate whether or not the same chemosensory unit (i.e. neurone and adjacent type 1 cluster) could process both sensory modalities. This was the case as illustrated in Fig. 5. In these experiments, a total of 20 hypoxia-responsive neurones were first identified in co-culture and tested further for sensitivity to isohydric hypercapnia (10 % CO2; pH 7.4). In 16 cases, hypoxia (PO2∼5 mmHg) under normocapnic conditions (5 % CO2; pH 7.4) elicited a subthreshold depolarization (Fig. 5A and C; mean 3.9 ± 0.3 mV). Furthermore, in a subset of these cases (11/16, ∼69 %), the hypercapnic stimulus also triggered a depolarization that was significantly larger than that triggered by hypoxia (Fig. 5A and C; mean 8.6 ± 0.4 mV; P < 0.001). Interestingly, combined application of the two stimuli produced a depolarization that was significantly larger than either stimulus elicited on its own (Fig. 5A, right trace; mean 11.4 ± 0.4 mV, P < 0.001, Fig. 5C). In four other cases, the initial response to hypoxia induced neuronal firing, and in each case the hypercapnic stimulus evoked an even more robust discharge (Fig. 5B; left and middle traces). Again, combined application of the two stimuli caused the neurone to fire at a significantly higher frequency (Fig. 5B, right trace; Fig. 5D) than that produced by either stimulus acting alone. Indeed, though not evident from the average of pooled data in Fig. 5D, in individual cases (n = 2) the O2-CO2 interaction was multiplicative, where the two stimuli together produced a larger response than the sum of the separate responses. For example in Fig. 5B, the mean neuronal firing frequency in hypoxia was 1.7 ± 0.6 Hz (three trials), compared to 10.1 ± 3.4 Hz in hypercapnia, and 23.2 ± 3.6 Hz in hypoxia plus hypercapnia. A multiplicative O2-CO2 interaction is characteristic of the mature carotid body (Lahiri & DeLaney, 1975; Pepper et al. 1995; Bamford et al. 1999), and there is evidence it may occur at the level of the type 1 cell (Dasso et al. 2000). Thus, a single chemosensory unit may transmit signals induced by both hypoxic and hypercapnic stimuli, and the two stimuli may interact in a multiplicative manner.

Figure 5. Interactions between hypoxia and hypercapnia at the same chemosensory unit in co-culture.

Figure 5

Open in a new tab

Recordings were obtained from petrosal neurones that were adjacent to a type 1 cluster. In A, both hypoxic (PO2∼5 mmHg) and hypercapnic (10 % O2; pH 7.4) stimuli gave rise to a subthreshold depolarization in the same neurone when applied separately; when applied together the response was larger than that of either stimulus alone. A similar situation is seen in B, except that the responses were suprathreshold, resulting in a burst of action potentials during stimulus application. In B, the O2-CO2 interaction was multiplicative, where the response to the combined stimuli was larger than the sum of the separate responses; the mean firing frequency was 1.7 ± 0.6 Hz in hypoxia (three trials), compared to 10.1 ± 3.4 Hz in hypercapnia, and 23.2 ± 3.6 Hz in hypoxia plus hypercapnia. C, mean (±s.e.m.) depolarization from 11 neurones (similar to A) exposed to the three stimuli; the response to hypercapnia was significantly stronger than that to hypoxia (***P < 0.001, one-way ANOVA), and combined application of the two stimuli was significantly stronger than that to hypercapnia alone (***P < 0.001, ANOVA). D, mean (±s.e.m.) spike frequency from four neurones which fired action potentials upon stimulus application (e.g. B); the response to hypercapnia was significantly stronger than that to hypoxia (**P < 0.005, one-way ANOVA), and combined application of the two stimuli was significantly stronger than that to hypercapnia alone (*P < 0.01, ANOVA).

Localization of P2X2 and P2X3 subunits by confocal immunofluorescence

The observation that the P2X2 and P2X3 mRNAs were expressed in individual petrosal chemoafferent neurones (Fig. 2), and that functional P2X receptor channels mediated both hypoxic and hypercapnic chemotransmission in co-culture (Zhang et al. 2000; Fig. 4), led us to investigate the localization of the corresponding proteins in situ. P2X2 and P2X3 subunits were localized using double-label immunofluorescence and confocal microscopy on tissue sections of 12- to 15-day-old rat petrosal ganglia and carotid bodies (CBs). Many neurones in the petrosal ganglia were strongly positive for both P2X2 and P2X3 subunits (Fig. 6A–C). Examination of several immunostained sections (n = 8) revealed that though the majority of neurones were positive for both subunits (> 60 %), a significant fraction (15-20 %) was positive for only one or the other subunit. Since the petrosal ganglion contains neurones that project to targets other than the CB, it was of interest to investigate whether both subunits were expressed in CB afferent terminals in situ. Importantly, nerve terminals found in association with CB chemoreceptor lobules were positive for both P2X2 and P2X3 immunoreactivity (Fig. 6D–F). For these experiments, three sections from each of four different CBs were immunostained, and in all cases there was complete overlap of P2X2 and P2X3 immunofluorescence. This was also the case during routine examination of serial confocal sections, at 1 μm steps, through the ∼18 μm thick tissue section. Thus, CB chemoafferent neurones express not only P2X2 and P2X3 mRNA in their soma, but also express the P2X2 and P2X3 protein subunits at their peripheral synaptic terminals apposed to type 1 chemoreceptors in situ.

Figure 6. Expression of P2X2 and P2X3 subunits in tissue sections of petrosal ganglia and carotid body as revealed by confocal immunofluorescence.

Figure 6

Open in a new tab

A–C, represent the same tissue section from the petrosal ganglion of a 15-day-old rat after immunostaining with P2X2 and P2X3 antibodies. Localization of P2X2 (A) and P2X3 (B) subunits is revealed by Cy3 and FITC fluorescence, respectively; dual exposure in C shows overlap of P2X2 and P2X3 staining. D–F, the same section from the carotid body of a 14-day-old rat after similar immunostaining with P2X2 and P2X3 antibodies as in A–C. Note complete overlap of P2X2 and P2X3 immunostaining in nerve terminals, which are apposed to lobules of chemoreceptor cells. Calibration bar represents 40 μm in A–C, and 50 μm in D–F.

DISCUSSION

In this study, we demonstrated for the first time that mRNAs for both P2X2 and P2X3 purinoceptor subunits were co-expressed in petrosal neurones that convey afferent chemosensory information from peripheral chemoreceptors (glomus or type 1 cells) located in the carotid body (CB). These experiments utilized RT-PCR, cloning and sequencing techniques on populations of freshly isolated petrosal neurones. Furthermore, we employed single-cell RT-PCR on physiologically identified neurones that developed functional connections with type 1 clusters in co-culture. A recent study from this laboratory demonstrated that similar functional neurones are activated by a hypoxic stimulus via co-release of ATP and ACh from type 1 cells (Zhang et al. 2000), and we proposed that the purinoceptors involved were likely to consist of heteromultimers of P2X2 and P2X3 subunits. Consistent with this proposal, mRNA for both P2X2 and P2X3 subunits was present in the same chemosensory neurone that formed part of a functional unit with a chemoreceptor cell cluster. Moreover, P2X2 and P2X3 proteins co-localized in CB afferent nerve terminals apposed to chemoreceptor lobules in situ, as demonstrated by serial confocal immunofluorescence. While these data alone do not exclude the possibility that homomeric P2X2 and P2X3 receptors may function within the same chemosensory neurone, the native receptors on identified functional neurones showed a slowly desensitizing response to α,β-methylene ATP (Zhang et al. 2000), a fact not readily explained by independent association of the two subunits into homomeric receptor complexes (Lewis et al. 1995; Radford et al. 1997). In response to the agonist α,β-methylene ATP, homomeric P2X3 receptors display rapid desensitization, in contrast to the slowly desensitizing properties of P2X2-P2X3 heteromultimers (Khakh et al. 1995; Lewis et al. 1995; Radford et al. 1997), and of the native receptors on functional chemosensory neurones (Zhang et al. 2000). On the other hand, though homomeric P2X2 receptors are slowly desensitizing, they are relatively insensitive to α,β-methylene ATP (Lewis et al. 1995; Radford et al. 1997).

However in a recent study (Jiang et al. 2001), wild-type homomeric P2X2 receptors were reported to be activated by high concentrations of α,β-methylene ATP (EC50∼1 mm), and a single point mutation (F44C) in the first membrane spanning domain was sufficient to produce a dramatic increase (∼100-fold) in receptor sensitivity. Since our previous study (Zhang et al. 2000) identified that hypoxia-responsive neurones in co-culture were quite sensitive to α,β-methylene ATP (EC50∼2 μm), it would appear that the heteromultimeric receptors represent a major type expressed by these chemosensory neurones. Nonetheless, co-expression of P2X2 and P2X3 receptor subunits in varying amounts is now known to generate heterogeneous populations of P2X receptors (Liu et al. 2001), and this may well contribute to broad variability in responses at CB afferent nerve terminals in situ.

Consistent with our immunofluorescence data showing positive staining of occasional petrosal neurones for P2X3 (but not P2X2) subunits, we did observe rapidly desensitizing currents evoked by ATP in several neurones sampled at random in petrosal cultures (M. Zhang & C. A. Nurse, unpublished observations). Such neurones probably expressed P2X3 homomeric receptors, but their peripheral targets are as yet unknown. In other studies, intravenous injection of α,β-methylene ATP in adult rats caused reflex hyperventilation and potent stimulation of chemosensory fibres in the cut carotid sinus nerve (McQueen et al. 1996), presumably due to direct activation of heteromultimeric P2X2-P2X3 receptors in the sensory terminals.

In this study, we also demonstrated for the first time that transmission of another natural CB stimulus, hypercapnia or increased CO2, was chemically mediated since it was sensitive to reduction of the extracellular Ca2+:Mg2+ ratio and involved co-release of ATP.

Recently ATP, acting via putative P2X receptors, was implicated as a mediator of mammalian central CO2 chemoreception (Thomas & Spyer, 2000; see also, King et al. 2000). Specifically, the CO2-evoked increases in respiration involved actions attributable to pH-sensitive P2 purinoceptors located within the ventral respiratory group. Taken together with the present findings, it therefore appears that ATP and P2X purinoceptors play pivotal roles in CO2 chemoreception generally, in both the central and peripheral nervous system. Acidity is also known to excite CB chemoreceptors (Gonzalez et al. 1994), but further studies are required to determine whether or not acidic pH enhances transmission at the heteromultimeric P2X2-P2X3 receptors in these co-cultures.

As was the case for hypoxia, preliminary indications are that co-release of ATP and ACh is the principal mechanism mediating hypercapnic chemoexcitation in the rat CB (M. Zhang & C. A. Nurse, unpublished observations). Different CB chemoexcitants have been reported to cause release of different proportions of neurotransmitters (Gomez-Nino et al. 1990), suggesting there could be different downstream signalling pathways leading to a rise in intracellular Ca2+ and type 1 cell secretion. We also found that the same chemosensory unit, consisting of a single petrosal neurone and a type 1 cell cluster, could be excited by both sensory modalities, i.e. hypoxia and hypercapnia. Though previous single fibre recordings from the carotid sinus nerve in the cat suggest that both stimuli can be processed by the same chemosensory fibre (Lahiri & DeLaney, 1975), it could not be excluded that terminal branching in the CB resulted in the innervation of more than one type 1 cluster, with each cluster processing a different sensory modality. Importantly, our data further support the notion that a single type 1 cluster can sense and respond to both hypoxia and hypercapnia (Bamford et al. 1999). Indeed, recent data based on fluorescence imaging of intracellular calcium indicate this property may be intrinsic to single type 1 cells (Dasso et al. 2000).

Interestingly, combined application of hypoxic and hypercapnic stimuli in co-culture produced a neuronal response that was consistently greater than that of either stimulus alone, suggesting synergy between the two stimuli. Interactions between these stimuli have been well documented in the intact carotid body and indeed the interaction can be multiplicative (Lahiri & Delaney, 1975; Pepper et al. 1995), where the two stimuli together produce a larger response than the sum of the separate responses. This multiplicative interaction appears to depend on the developmental stage or postnatal maturation of the CB and its afferent input (Pepper et al. 1995; Bamford et al. 1999). For example, based on fluorescence measurements of intracellular Ca2+ in rat type 1 cell clusters, no multiplicative O2-CO2 interaction was seen at any age (Bamford et al. 1999), though in another study using similar techniques evidence for stimulus interaction was seen in single type 1 cells (Dasso et al. 2000). With respect to the carotid sinus nerve discharge in situ, the multiplicative O2-CO2 interaction was seen at 2-3 weeks postnatally, but not in younger pups (Bamford et al. 1999).

In the present study where neurones (from ∼2-week-old pups) were sampled after 3-5 days in co-culture, we obtained clear evidence for synergy between the two stimuli but only preliminary evidence that a multiplicative O2-CO2 interaction (two cases) may occur with respect to the afferent discharge (cf. Dasso et al. 2000).

In conclusion, this study provides the first demonstration that single carotid body chemoafferent neurones express mRNA for both P2X2 and P2X3 purinoceptor subunits, which co-localize in synaptic terminals apposed to chemoreceptor cells. This expression pattern is consistent with the idea that P2X2-P2X3 heteromultimeric receptors are the functional receptors, activated by ATP released as a co-transmitter from the chemoreceptors. This release of ATP occurs along with other excitatory (e.g. ACh), and possibly inhibitory, neurotransmitters, which help shape the postsynaptic response in sensory afferent terminals during carotid body chemoexcitation by natural stimuli, e.g. hypoxia and hypercapnia.

Acknowledgments

This work was supported by grants from the Natural Sciences and Engineering Research Council of Canada and the Canadian Institutes for Health Research (MOP 12037) to C.A.N., and a Wellcome International Prize Traveling Research Fellowship (Ref: 06154/B/00/Z) to I.M.F. We thank Drs William Muller and Chris Wood for use of equipment.

References

  1. Bamford OS, Sterni LM, Wasiko MJ, Montrose MH, Carroll JL. Postnatal maturation of carotid body and type 1 cell chemoreception in the rat. American Journal of Physiology. 1999;276:L875–884. doi: 10.1152/ajplung.1999.276.5.L875. [DOI] [PubMed] [Google Scholar]
  2. Buckler KJ, Vaughan-Jones RD. Effects of hypoxia on membrane potential and intracellular calcium in rat neonatal carotid body type 1 cells. Journal of Physiology. 1994a;476:423–428. doi: 10.1113/jphysiol.1994.sp020143. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Buckler KJ, Vaughan-Jones RD. Effects of hypercapnia on membrane potential and intracellular calcium in rat carotid body type 1 cells. Journal of Physiology. 1994b;478:157–171. doi: 10.1113/jphysiol.1994.sp020239. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Buell G, Collo G, Rassendren F. P2X receptors: an emerging channel family. European Journal of Neuroscience. 1996;8:2221–2228. doi: 10.1111/j.1460-9568.1996.tb00745.x. [DOI] [PubMed] [Google Scholar]
  5. Burnstock G. The past, present and future of purine nucleotides as signalling molecules. Neuropharmacology. 1997;36:1127–1139. doi: 10.1016/s0028-3908(97)00125-1. [DOI] [PubMed] [Google Scholar]
  6. Collo G, North RA, Kawashima E, Merlo-Pich E, Neidhart S, Surprenant A, Buell G. Cloning of P2X5 and P2X6 receptors and the distribution and properties of an extended family of ATP-gated ion channels. Journal of Neuroscience. 1996;1(6):2495–2507. doi: 10.1523/JNEUROSCI.16-08-02495.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Dasso LL, Buckler KJ, Vaughan-Jones RD. Interactions between hypoxia and hypercapnic acidosis on calcium signaling in carotid body type 1 cells. American Journal of Physiology - Lung Cellular and Molecular Physiology. 2000;279:L36–42. doi: 10.1152/ajplung.2000.279.1.L36. [DOI] [PubMed] [Google Scholar]
  8. Gomez-Nino A, Dinger B, Gonzalez C, Fidone SJ. Differential stimulus coupling to dopamine and norepinephrine stores in rabbit carotid body type 1 cells. Brain Research. 1990;525:160–164. doi: 10.1016/0006-8993(90)91334-d. [DOI] [PubMed] [Google Scholar]
  9. Gonzalez C, Almaraz L, Obeso A, Rigual R. Carotid body chemoreceptors: from natural stimuli to sensory discharges. Physiological Reviews. 1994;74:829–898. doi: 10.1152/physrev.1994.74.4.829. [DOI] [PubMed] [Google Scholar]
  10. Horn R, Marty A. Muscarinic activation of ionic currents measured by a new whole-cell recording method. Journal of General Physiology. 1988;92:145–159. doi: 10.1085/jgp.92.2.145. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Jiang L-H, Rassendren F, Spelta V, Surprenant A, North RA. Amino acid residues involved in gating identified in the first membrane-spanning domain of the rat P2X2 receptor. Journal of Biological Chemistry. 2001;276:14902–14908. doi: 10.1074/jbc.M011327200. [DOI] [PubMed] [Google Scholar]
  12. Khakh BS, Humphrey PPA, Surprenant A. Electrophysiological properties of P2X-purinoceptors in rat superior cervical, nodose and guinea-pig coeliac neurones. Journal of Physiology. 1995;484:385–395. doi: 10.1113/jphysiol.1995.sp020672. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. King BF, Townsend-Nicholson A, Wildman SS, Thomas T, Spyer KM, Burnstock G. Coexpression of rat P2X2 and P2X6 subunits in Xenopus oocytes. Journal of Neuroscience. 2000;20:4871–4877. doi: 10.1523/JNEUROSCI.20-13-04871.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Lahiri S, DeLaney RG. Stimulus interaction in the responses of carotid body chemoreceptor single afferent fibres. Respiration Physiology. 1975;24:249–266. doi: 10.1016/0034-5687(75)90017-1. [DOI] [PubMed] [Google Scholar]
  15. Lewis C, Neldhart S, Holy C, North RA, Buell G, Surprenant A. Coexpression of P2X2 and P2X3 receptor subunits can account for ATP-gated currents in sensory neurones. Nature. 1995;377:432–435. doi: 10.1038/377432a0. [DOI] [PubMed] [Google Scholar]
  16. Liu M, King BF, Dunn PM, Rong W, Townsend-Nicholson A, Burnstock G. Coexpression of P2X3 and P2X2 receptor subunits in varying amounts generates heterogeneous populations of P2X receptors that evoke a spectrum of agonist responses comparable to that seen in sensory neurones. Journal of Pharmacology and Experimental Therapeutics. 2001;296:1043–1050. [PubMed] [Google Scholar]
  17. McQueen DS, Bond SM, Moores C, Chessell I, Humphrey PPA, Dowd E. Activation of P2X receptors for adenosine triphosphate evokes cardiorespiratory reflexes in anesthetized rats. Journal of Physiology. 1996;507:843–855. doi: 10.1111/j.1469-7793.1998.843bs.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. North RA, Barnard EA. Nucleotide receptors. Current Opinions in Neurobiology. 1997;7:346–357. doi: 10.1016/s0959-4388(97)80062-1. [DOI] [PubMed] [Google Scholar]
  19. Nurse CA, Zhang M. Acetylcholine contributes to hypoxic chemotransmission in co-cultures of rat type 1 cells and petrosal neurones. Respiration Physiology. 1999;115:189–199. doi: 10.1016/s0034-5687(99)00017-1. [DOI] [PubMed] [Google Scholar]
  20. Pepper DR, Landauer RC, Kumar P. Postnatal development of CO2-O2 interaction in rat carotid body in vitro. Journal of Physiology. 1995;485:531–541. doi: 10.1113/jphysiol.1995.sp020749. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Radford KM, Virginio C, Surprenant A, North RA, Kawashima E. Baculovirus expression provides direct evidence for heteromeric assembly of P2X2 and P2X3 receptors. Journal of Neuroscience. 1997;17:6529–6533. doi: 10.1523/JNEUROSCI.17-17-06529.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Sambrook J, Fritsch EF, Maniatis T. Molecular Cloning: A Laboratory Manual. 2. NY, USA: Cold Spring Harbor Laboratory Press; 1989. [Google Scholar]
  23. Stea A, Nurse CA. Whole-cell and perforated-patch recordings from O2-sensitive rat carotid body cells grown in short and long-term culture. Pflügers Archiv. 1991;418:93–101. doi: 10.1007/BF00370457. [DOI] [PubMed] [Google Scholar]
  24. Surprenant A, Buell GN, North RA. P2X receptors bring new structure to ligand-gated ion channels. Trends in Neurosciences. 1995;18:224–229. doi: 10.1016/0166-2236(95)93907-f. [DOI] [PubMed] [Google Scholar]
  25. Thomas T, Spyer KM. ATP as a mediator of mammalian chemoreception. Journal of Physiology. 2000;523:441–447. doi: 10.1111/j.1469-7793.2000.00441.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Torres GE, Egan TM, Voigt MM. Hetero-oligomeric assembly of P2X receptor subunits. Journal of Biological Chemistry. 1999;274:6653–6659. doi: 10.1074/jbc.274.10.6653. [DOI] [PubMed] [Google Scholar]
  27. Vulchanova L, Riedl MS, Shuster SJ, Buell G, Surprenant A, North RA, Elde R. Immunohistochemical study of the P2X2 and P2X3 receptor subunits in rat and monkey sensory neurones and their central terminals. Neuropharmacology. 1997;36:1229–1242. doi: 10.1016/s0028-3908(97)00126-3. [DOI] [PubMed] [Google Scholar]
  28. Zhang M, Zhong H, Vollmer C, Nurse CA. Co-release of ATP and ACh mediates hypoxic signalling at rat carotid body chemoreceptors. Journal of Physiology. 2000;525:143–158. doi: 10.1111/j.1469-7793.2000.t01-1-00143.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Zhong H, Zhang M, Nurse CA. Synapse formation and hypoxic signalling in co-cultures of rat petrosal neurones and carotid body type 1 cells. Journal of Physiology. 1997;503:599–612. doi: 10.1111/j.1469-7793.1997.599bg.x. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES