Release of substance P by low oxygen in the rabbit carotid body: evidence for the involvement of calcium channels

Brain Research 892 (2001) 359–369www.elsevier.com/ locate /bres

Research report

Release of substance P by low oxygen in the rabbit carotid body:evidence for the involvement of calcium channels

a a b bDong-Kyu Kim , Eui K. Oh , Beth A. Summers , Nanduri R. Prabhakar ,a ,*Ganesh K. Kumar

aDepartment of Biochemistry, School of Medicine, Case Western Reserve University, Cleveland, OH 44106, USAbDepartment of Biophysics and Physiology, School of Medicine, Case Western Reserve University, Cleveland, OH 44106, USA

Accepted 21 November 2000

Abstract

Carotid bodies from diverse species contain substance P (SP), an 11-residue peptide that belongs to the tachykinin peptide family.Previous studies indicated that SP is excitatory to the carotid body and is associated with sensory response to hypoxia. However, releaseof SP from the carotid body during hypoxia has not been documented. In the present study, we determined whether hypoxia releases SPfrom the carotid body and further characterized the mechanism(s) associated with SP release by low oxygen. The release of SP fromsuperfused rabbit carotid body was determined by an enzyme immunoassay (EIA). SP-like immunoreactivity was localized to manyglomus cells and nerve fibers and the concentration of SP in the rabbit carotid body was 1.560.1 ng/mg protein. For release studies,carotid bodies (n556) were superfused with a modified Tyrode medium containing Hepes buffer, pH 7.4, saturated with either room air(normoxia) or hypoxic gas mixtures. The basal release of SP during normoxia was 51.061.5 fmol /min per mg protein. Hypoxia increasedSP release from the carotid body and the magnitude of release is dependent on the severity of hypoxic stimulus. Moderate hypoxia (pO ,2

7964 mmHg) stimulated SP release by |50%, whereas SP release during severe hypoxia (pO , 1166 mmHg) was 2-fold higher than the2

normoxic control. A similar pattern of SP release was also observed when superfusion medium containing CO –HCO buffer, pH 7.4,2 3

was used for release studies. To examine the mechanism(s) associated with hypoxia-induced SP release from the carotid body, moderatelevel of hypoxia (12% O 1N ) was used. Omission of calcium in the superfusion medium markedly attenuated hypoxia-induced SP2 2

21 21release (.95%), whereas the basal release of SP was unaffected. Cd (100 mM), a voltage-dependent Ca -channel blocker, abolished21hypoxia-induced SP release. About 85% of SP release by hypoxia was inhibited by v-conotoxin GVIA (1 mM), an N-type Ca channel

21blocker, whereas nitrendipine (1.5 mM), an inhibitor of L-type Ca channel partially attenuated (|65%) hypoxia-induced SP release. By21contrast, v-agatoxin TK (50 nM), a P/Q-type Ca -channel inhibitor, had no significant effect (P.0.05, n56). These results suggest that

SP is released from the rabbit carotid body by hypoxia that depends on the severity of the hypoxic stimulus. Further, SP release by21hypoxia is a calcium-dependent process and is primarily mediated by N- and L-type Ca channels. 2001 Elsevier Science B.V. All

rights reserved.

Theme: Excitable membranes and synaptic transmission

Topic: Mechanisms of neurotransmitter release

Keywords: Carotid body; Arterial chemoreceptor; Substance P; Release; Tachykinin; Hypoxia; Voltage-gated calcium channel; N-Type calcium channel;L-Type calcium channel; Enzyme immunoassay

1. Introduction arterial blood and transduce these changes into actionpotential encoded nerve impulses that are relayed to brain

Carotid bodies are the major peripheral chemoreceptors stem neurons for proper ventilatory adjustments. Thethat detect changes in the partial pressure of oxygen in the mechanism(s) by which carotid body senses low oxygen

remains to be an active area of investigation. It has beenproposed that sensing of low oxygen by carotid body may*Corresponding author. Tel.: 11-216-368-3349; fax: 11-216-368-occur in two discrete but related steps characterized by an3419.

E-mail address: [email protected] (G.K. Kumar). initial transduction process that is coupled to a second step

0006-8993/01/$ – see front matter 2001 Elsevier Science B.V. All rights reserved.PI I : S0006-8993( 00 )03272-8

360 D.-K. Kim et al. / Brain Research 892 (2001) 359 –369

involving transmission of the hypoxic stimulus [7,14,37]. carotid body preparation was used for the release studiesIn the transduction step, hypoxia causes membrane depo- and SP release was determined using an enzyme immuno-larization [5,6] via mechanisms involving a heme protein assay. Our results showed that hypoxia augments SP

1 21[25] and/or inhibition of a K channel [26,27]. In the release in a Ca -dependent manner and that N- and21ensuing transmission step, activation of voltage-dependent L-type Ca channels play a major role in hypoxia-in-

21Ca channel(s) facilitates a transient elevation in the duced SP release in the carotid body.cytosolic calcium and subsequent release of neurotrans-mitter(s) leading to an increase in sensory discharge[7,14,37]. Several candidate neurotransmitter molecules 2. Materials and methodshave been identified in the carotid body and their roles inthe sensory response of the carotid body have been 2.1. Materialsinvestigated. These neurochemicals include acetylcholine,catecholamines, neuropeptides and gas molecules [7,37]. Substance P (amide form) and cadmium chloride wereExperimental evidences suggest that in most species, with obtained from Sigma. Trifluoroacetic acid and acetonitrilefew exception, acetylcholine [10,11] and the neuropeptide were from Pierce. v-Agatoxin TK and v-conotoxin GVIAsubstance P (SP) [36–38,40] augment the sensory dis- were purchased from Alomone Laboratories. Sep-Pakcharge of the carotid body, whereas dopamine [7,14,37], (C18) cartridge column was from Peninsula Lab. Nitren-atrial natriuretic peptide [48], and NO [37] inhibit carotid dipine was obtained from Calbiochem. Rat anti-substancebody activity. P antibody was purchased from Chemicon. Substance P

Several lines of evidence suggest that SP-like peptide enzyme immunoassay kit was from Cayman Chemicalmay play important neurotransmitter /neuromodulator roles (Ann Arbor, MI).in the sensory response of the carotid body to hypoxia.First, among various species studied so far, SP-like 2.2. Isolation of carotid bodiesimmunoreactivity has been reported to occur either in thetype I glomus cells [45] or in the afferent nerve fibers Experiments were performed in adult male New Zealand[17,24] or associated with both structures [1,4,40,49]. White rabbits (n556 CBs) weighing 2.0–2.5 kg. TheSecond, like hypoxia, intracarotid administration of SP animals were killed with CO and the bifurcations at the2

close to the carotid body increases the sensory activity of common carotid arteries were rapidly excised followingthe carotid body in cat [43] and rat [3] and ventilation in protocols approved by the Institutional Animal Care andrabbit [42] but not in goat [35]. More importantly, exogen- Use Committee. Carotid bodies were isolated from theous SP augments the sensory response of the carotid body bifurcation under the microscope and placed in oxygenated

21 21to hypoxia [4,21,28,36,39,40]. Third, spantide and CP- Ca /Mg -free modified Tyrode solution at 48C.96,345, a peptidyl and non-peptidyl SP receptor antago-nists, respectively, either attenuate or abolish the hypoxic 2.3. Release of substance P (SP) from carotid bodies inresponse of the cat carotid body [21,38,43] without affect- vitroing the hypercapnic response [41]. Similarly, spantidegreatly attenuated the hypoxic response of the rat carotid The experimental set up for the release of SP from thebody in vivo [3]. Taken together, these observations carotid body in vitro is shown in Fig. 1. For a givensuggest a crucial role for SP in the sensory response of the experiment, freshly isolated carotid bodies (n54) werecarotid body to hypoxia. Fourth, carboxypeptidase-E, a placed in a polypropylene chamber (volume of 800 ml)protease that is involved in the processing of the precursor fitted with a porous filter at the bottom to retain the tissues.form of SP into its biologically active form [19] as well as The carotid bodies were superfused with a modifiedneutral endopeptidase, an enzyme that mediates the hydro- Tyrode solution having the following composition (mM):lytic inactivation of SP, have been identified in the carotid 140 NaCl, 5 KCl, 1.8 CaCl , 1.0 MgCl , 5 glucose, and 102 2

body [20,23]. These results suggest that the machinery Hepes; extracellular pH 7.4. The superfusion medium alsoresponsible for synthesis and degradation of SP occurs in contained the following: 0.1% bovine serum albumin tothe carotid body. Thus, on the basis of the above observa- minimize adsorption of SP to surfaces; 1 mM phenyl-tions, SP appears to fulfil several criteria required for its methylsulfonyl fluoride to reduce proteolytic degradationpurported role in the transmission of the hypoxic stimulus of released peptides; and essential and non-essential aminoin the carotid body. However, the release of SP in the acids to sustain carotid body metabolism during SP releasecarotid body in response to hypoxia has not yet been studies. The temperature of the superfusing medium wasdemonstrated. continuously monitored and maintained at 3760.58C, and

This study was undertaken to determine whether SP is the rate of superfusion maintained at 0.3 ml /min. Thereleased in the carotid body during hypoxia and to chamber containing the carotid bodies was sealed with ancharacterize the cellular mechanisms associated with hypo- airtight plug to minimize exposure to atmospheric air.xia-induced SP release. An in vitro superfused rabbit Gas-impermeable Tygon silicone tubing (Cole-Parmer)

D.-K. Kim et al. / Brain Research 892 (2001) 359 –369 361

Fig. 1. Diagram illustrating the experimental set up for the measurement of substance P (SP) release from the carotid body.

was used to connect the chamber containing the carotid 2.5. Sep-Pak column purification of SP released into thebody to the superfusion medium reservoirs. The superfu- superfusion medium or extracted from the carotid bodysion medium was equilibrated with either room air (nor-moxia) or hypoxic gas mixtures for at least 30 min prior to Either the frozen superfusate or the peptide extract ofinitiation of the experiment. The carotid bodies were the carotid body was thawed and applied to a Sep-Paksuperfused in succession with medium saturated with (C18) cartridge column that was previously equilibratedappropriate gas mixtures for a defined period as outlined with 0.1% (v/v) aqueous trifluoroacetic acid (buffer 1).under protocols. The superfusate during each gas challenge Following absorption of peptides the column was elutedwas collected at an interval of 5 min in polypropylene with buffer 1 to remove salts (from the superfusiontubes and immediately acidified with glacial acetic acid to medium) and any non-peptide materials from the samples.give a final concentration of 2 M. The acidified superfu- Peptides were eluted from the Sep-Pak column using 0.1%sates containing the peptides released during various gas (v /v) trifluoroacetic acid in 60% (w/v) acetonitrile andchallenges were frozen and stored at 2208C until further concentrated in a centrifugal concentrator (SpeedVac). Theanalysis. lyophilized samples were reconstituted in 200 ml of

enzyme immunoassay buffer. The efficiency of SP re-covery from Sep-Pak column was calculated by applying a

2.4. Extraction of SP from carotid bodies known amount of synthetic SP and determining theconcentration of the peptide in the eluted fraction from the

After completion of the superfusion protocols carotid column by enzyme immunoassay as described below. On21 21bodies were washed twice with Ca /Mg -free modified average, the efficiency of peptide recovery was 7063%

Tyrode medium and homogenized in 450 ml of 2 M acetic (n56).acid containing 1 mM EDTA at 48C. The homogenate wascentrifuged at 32 0003g for 15 min at 48C and the 2.6. Enzyme immunoassay for SPsupernatant containing the peptides was collected. Theextraction was repeated and the supernatants from the two The enzyme immunoassay (EIA) for SP was performedextractions were pooled and stored at 2208C until further in a 96-well microplate and is based on the competitionanalysis. between free SP, derived from the unknown sample, and a


covalent conjugate of SP with electric eel acetylcho- incubated in 0.5% (v/v) H O in 100% methanol for 152 2

linesterase (AChE tracer) for a fixed amount of anti-SP min to reduce endogenous peroxidase activity. The sec-antibody available for binding. For the quantitative de- tions were rinsed with PBS for 5 min and then incubatedtermination of concentration of SP in the carotid body with 0.05 M sodium phosphate buffer, pH 7.4, containingtissue as well as in the release medium, a commercial EIA 0.5 M NaCl, 0.3% (V/V) Triton X-100, and 2.0% BSAkit from Cayman Chemicals was used and the EIA (buffer A) for 60 min. Non-specific binding sites on tissueprotocol described by the manufacturer was followed. The sections were blocked by incubating the sections sequen-details of the procedure were as follows. First, the non- tially with avidin (15 min), PBS (5 min), biotin (15 min),specific binding sites on the 96-well microplate, pre-coated PBS (5 min) and buffer A (10 min). Thereafter, thewith mouse monoclonal anti-rabbit IgG (the capture anti- sections were incubated with rat anti-substance P mono-body), were saturated with a blocking solution supplied by clonal antibody (1:250 dilution; Chemicon International)the manufacturer. AChE tracer, SP antiserum and either for 18 h. Subsequently, the sections were rinsed withstandard SP or unknown samples were added sequentially buffer A and PBS for 10 min each and incubated withand incubated at 48C for 18 h. After washing with EIA affinity purified biotinylated goat anti-rat IgG (1:200buffer to remove any unbound reagents, an aqueous dilution; Vector Laboratories, CA) for 60 min. The carotidsolution containing acetylthiocholine, the substrate for body sections were washed in PBS, three times, 10 minAChE and 5,59-dithio-bis-(2-nitrobenzoic acid) (DTNB) each followed by incubation with Cy3-conjugated strep-was added to each well. AChE hydrolyzed acetyl-thioch- tavidin in PBS (0.3 mg/ml; Jackson ImmunoResearcholine into thiocholine that reacts with DTNB to produce Laboratories, PA) for 30 min. After washing with PBS5-thio-2-nitrobenzoic acid. This product has an intense three times, 5 min each the sections were incubated withyellow color with a characteristic absorbance at 412 nm. phenyldiamine (1 mg/ml) in PBS. FollowingThe absorbance was determined using either a Shimadzu phenyldiamine treatment, the sections were rinsed twiceSpectrophotometer or a Microplate Reader (Molecular with PBS, covered with a solution containing glycerin andDevices, CA). The intensity of the color developed is PBS in a ratio of 2 to 1 and sealed with a cover slip. Theinversely proportional to the amount of free SP present in sections were imaged using a NikonE Eclipse E600the well during the incubation. The sensitivity of the assay fluorescence microscope equipped with a computer on-linewas determined by serial dilution and was found to be |17 data acquisition system and SPOTE software (Diagnosticpg of SP/ml. The cross-reactivity of SP related peptides in Instrument). Control for SP immunolabeling studies in-the assay was as follows: SP, 100%; SP (2–11), 93%; SP cludes treatment of sections with rat pre-immune antibody(4–11), 97%; SP (7–11), 30%; SP (8–11), ,0.01%; preparation in place of the primary antibody.neurokinin A, 2.7%; and neurokinin B, 0.04%. The amountof SP released into the medium was expressed in fmol /min 2.8. Experimental protocolsper mg protein. The total SP is defined as the sum of SPreleased into the medium during various gas challenges In the following series of experiments, both carotidand the residual SP in the tissue. Protein concentration bodies obtained from each animal were used.either in the medium or in the tissue extract was de-termined by a dye binding method involving colloidal gold 2.8.1. Series 1[2] and bovine serum albumin was used as the standard. In the first series of experiments (n54 CBs), the cellular

distribution of SP in the carotid body was determined by2.7. Immunocytochemical studies immunocytochemistry using a rat anti-SP antibody. The

carotid bodies were isolated from anaesthetized rabbits,For immunocytochemical localization of SP, the carotid breathing room air. Tissues were fixed by immersion

bodies (n54) were isolated from the bifurcation of the fixation and immunolabeling of the carotid body sectionsrabbit carotid artery and fixed by immersing the tissues in was performed using procedures as described above.a fixative medium containing 4% (w/v) paraformaldehydein 0.1 M phosphate-buffered physiological saline (PBS) 2.8.2. Series 2for 36 h at 48C. After a brief washing in PBS, the tissues In the next series of experiments (n58 CBs), the effectwere subsequently placed in a 30% (w/v) sucrose solution of hypoxia (12% O in N ) on the release of SP from the2 2

at 48C, overnight. Thin sections (12–14 mm) of the carotid carotid body was determined by enzyme immunoassay.bodies were obtained using a cryostat and collected on Prior to initiation of release studies, the superfusion

21slides coated with polylysine. The sections were stored at medium containing nominal Ca (1.8 mM) was bubbled2808C. with either room air (pO , 13866 mmHg; basal release) or2

For immunocytochemical analysis, avidin–biotin conju- 12% O in N (pO , 7964 mmHg; hypoxia) for at least 302 2 2

gation method was used. Unless otherwise specified, all the min and the pO of the medium was measured using a2

procedures were performed at 258C. The frozen carotid blood gas analyzer. Carotid bodies were superfused se-body sections were hydrated in PBS for 5 min and quentially with media equilibrated with the following gas


mixtures for the duration indicated: room air for 30 min significance was evaluated by a paired t-test or ANOVA(basal), 12% O in N for 20 min (hypoxia), and room air for repeated measures, and, if a significant interaction was2 2

for 20 min (recovery to basal level). The superfusion indicated, the results were further compared using Tukey’smedium was collected at 5-min intervals during gas test. P values ,0.05 were considered significant.challenges. Each fraction containing the superfusate wasindividually concentrated by Sep-Pak (C18) cartridgecolumn and the concentration of SP was determined using 3. Resultsenzyme immunoassay as described above.

In parallel experiments (n58 CBs each), the effects of 3.1. Tissue content and cellular distribution of SPsevere hypoxia (i.e., medium equilibrated with 100% N ;2

pO , 1166 mmHg) as well as hyperoxia (i.e., medium2 On average, the wet weight of a single rabbit carotidequilibrated with 100% O ; pO , 690611 mmHg) were2 2 body was 1.260.1 mg with the total protein content ofassessed using the protocol described earlier. 49.965.5 mg (Table 1). The tissue concentration of SP in

the rabbit carotid body was 1.560.1 ng/mg protein and2.8.3. Series 3 was in good agreement with the previously reported value

In another series of experiments (n54 CBs), the in- for SP in the rabbit carotid body [15]. The distribution offluence of CO /HCO buffer on hypoxia-induced SP2 3 SP in the rabbit carotid body following immersion fixationrelease from the carotid body was assessed. The superfu- was examined using rat anti-SP monoclonal antibody. Assion medium containing Hepes buffer, pH 7.4 (extracellu- shown in Fig. 2, both glomus cells as well as nerve fiberslar), was replaced with a medium containing CO –HCO2 3 were positive for SP-like immunoreactivity (SP-LI). Basedbuffer, pH 7.4 (extracellular). Prior to initiation of release on the immunostaining pattern, two populations of glomus21studies, the superfusion medium containing Ca (1.8 cells could be distinguished; those that show strongmM) was equilibrated with either 21% O 15% CO in N2 2 2 immunoreactivity and the others with weak SP-LI. Mor-(basal) or 12% O 15% CO in N (hypoxia). Other2 2 2 phologically, these two populations of glomus cells remainexperimental details were as described in ‘series 2’. indistinguishable as there are no significant differences in

the size of these cells.2.8.4. Series 4

21In experiments assessing the role of extracellular Ca3.2. Characteristics of SP release during hypoxia(n58 CBs), the carotid bodies were superfused with a

21 21low-Ca medium (,0.1 mM; i.e., omission of Ca andaddition of 10 mM EGTA) equilibrated with room air 3.2.1. Effect of hypoxia(normoxia) or 12% O in N (hypoxia). The results were We have used a superfusion set up for the in vitro2 2

compared with carotid bodies superfused with nominal studies on the release of SP from the carotid body (see Fig.21Ca medium (i.e., 1.8 mM; ‘Series 2’). The sequence and 1). The release of SP in the superfusion medium equili-

duration of gas challenges were as described in the brated either with room air (normoxia) or 12% O in2

experiments of ‘Series 2’. N (hypoxia) was determined and expressed in fmol /min2

per mg protein. With superfusion medium containing2.8.5. Series 5 Hepes buffer, extracellular pH 7.4, the basal, normoxic

In this series of experiments, the effects of inorganic and release of SP was 51.061.5 fmol /min per mg protein,21organic Ca channel blockers on the basal and hypoxia- whereas after hypoxic exposure SP release was increased

induced SP release from the carotid body were investi- to 76.263.3 fmol /min per mg protein (Table 2 and Fig. 3).gated. The effects of calcium channel blockers on SP Thus, on average, SP release from the carotid body duringrelease were assessed in individual experiments (n54 CBs hypoxia was increased by |50% compared to the normox-each) by adding the blockers to the superfusion medium ic, basal release suggesting that hypoxia increases theequilibrated with either room air (normoxia) or 12% O in release of SP from the carotid body.2

N (hypoxia) at the indicated final concentration. The2

following calcium channel blockers were used: cadmiumTable 1chloride (100 mM), nitrendipine (1.5 mM; for L-type), aWet weight, protein and substance P content of the rabbit carotid body

v-agatoxin TK (50 nM; for P/Q-type), and v-conotoxinWet weight 1.260.1 mg per CBGVIA (1 mM; for N-type). The processing of superfusatesTotal protein 49.965.5 mg per CBand the EIA measurement of SP concentration in the bTissue SP 77.6613.2 pg per CB

bsuperfusate were as described under ‘Series 2’. Tissue SP 1.560.1 ng/mg proteinbTissue SP 1.160.1 pmol /mg protein

2.9. Data analysis a The data represent mean6S.E.M. (n556 carotid bodies).b Tissue level of SP was determined following Sep-Pak cartridge column

All data are expressed as means6S.E.M. Statistical purification.


Fig. 3. Representative example of an experiment illustrating the releaseof substance P from the carotid body in response to sequential superfu-sion with medium equilibrated with room air (normoxia), hypoxic gasmixture (12% O 1N ), and room air (normoxic recovery) is shown.2 2

superfusion of the carotid bodies with hypoxic medium(12% O 1N ) for 20 min, there was a sharp increase in2 2

SP release in the first 5 min of hypoxic challenge (see Fig.3). In the ensuing periods of hypoxia, there was a slightdecrease in SP release that was nearly maintained for theremaining duration of hypoxia. Following the removal ofhypoxic challenge, there was a lag of 2065 min in thereturn of SP to pre-stimulus control value.

3.2.3. Effect of buffer ionsIn another series of experiments, we assessed the effect

of buffer ion in the superfusion medium on SP releaseFig. 2. Light micrographs of substance P-like immunoreactivity (SP-LI) from the carotid body. With medium containing CO –2in the rabbit carotid body. Top panel: SP-LI was seen in numerous HCO buffer, extracellular pH 7.4, the basal and hypoxia-3glomus cells (short arrows) and nerve fibers (long arrow). Bottom panel:

induced SP releases were 56.462.8 and 76.562.2 fmol /control section with omission of primary antibody staining. Bar: 50 mm.min per mg protein, respectively, and were comparable tothose observed with medium containing Hepes buffer,3.2.2. Effect of duration of hypoxiaextracellular pH 7.4 (Table 2). These observations suggestThe magnitude of SP release during hypoxia appears tothat both the basal and hypoxia-induced SP release wasbe dependent on the duration of hypoxic challenge. Withinsensitive to buffer ion composition of the superfusionmedium. Therefore, in the subsequent experiments, we

Table 2 have used medium containing Hepes buffer, extracellularEffects of salt composition of the superfusion medium on the basal and pH 7.4 for further characterization of SP release from thehypoxia (12% O 1N )-induced substance P release from the carotid2 2 carotid body.abody

Gas challenges Substance P release (fmol /min per mg protein) 3.2.4. Effect of partial pressure of oxygen in the2Hepes buffer, pH 7.4 CO /HCO buffer, pH 7.4 superfusion medium2 3

b The magnitude of SP release in the carotid bodyNormoxia 51.061.5 (n56) 56.462.8 (n56)b appeared to be dependent on the partial pressure of oxygenHypoxia 76.263.3 (n56) 76.562.2 (n56)

a in the superfusion medium. As shown in Fig. 4 the basalData represent mean6S.E.M.b release of SP, during hyperoxic challenge was 40.662.0SP release is not significantly different from that observed in Hepesbuffer. fmol /min per mg protein. SP release increased by |25,


Fig. 5. Removal of extracellular calcium attenuates hypoxia-induced butnot the basal SP release from the carotid body. Data are mean6S.E.M.

21from six experiments. **Significant effect of Ca removal on SP releaseduring hypoxia (12% O 1N ; P,0.01); n.s. indicates absence of2 2

21significant effect of Ca removal on the basal SP release duringFig. 4. Effects of partial pressure of oxygen in the superfusion medium normoxia (P.0.05).on substance P release from the carotid body as measured by enzymeimmunoassay. Data are mean6S.E.M. from three to four independent

(P,0.01, n58). The basal release of SP during normoxia,experiments along with triplicate measurements. Desired level of oxygen21is achieved by balancing it with nitrogen. however, was unaffected by the low extracellular Ca

(P.0.05; n58). The above results suggest that entry of21|90 and |150% during superfusion of the carotid body extracellular Ca is coupled to hypoxia-induced SP

with medium equilibrated with room air, hypoxic gas release in the carotid body.mixtures of 12% O 1N and 100% N , respectively.2 2 2

21Taken together, these observations provide evidence that 3.3.2. Effects of voltage-gated Ca channel blockers21hypoxia stimulates SP release from the carotid body and To further understand the mechanism by which Ca

that with in a narrow range of partial pressure of oxygen modulates SP release during hypoxia, the effect of voltage-21(i.e., from 140 to 10 mmHg) there is a robust release of SP gated Ca channel blockers in hypoxia-induced SP

in the carotid body. release from the carotid body was investigated. When thecarotid bodies were superfused with medium containing

21 21 213.3. Involvement of Ca in hypoxia-induced SP release Cd (100 mM), an inorganic voltage-gated Ca channelin the carotid body blocker, SP release during normoxia was significantly

attenuated (P,0.01; n56), whereas hypoxia-induced SPCalcium has been shown to play important roles in the release was completely abolished (P,0.005; n56; Fig. 6).

secretion of neurotransmitters /neuromodulators from brain Previously, it has been shown that carotid body expres-21cells, synaptosomes, motor nerve endings and endocrine ses four types of voltage-gated Ca channels that include

cells [12,13,18,34]. In the following experiments, we L-, N-, P/Q-, and resistant types [9,29,32]. The in-21 21examined the role of Ca in hypoxia-induced SP release volvement of one or more of Ca channels in hypoxia-

from the carotid body. induced SP release from the carotid body was assessed bythe addition of blockers specific to L-, or N-, or P/Q-type

21 213.3.1. Extracellular Ca Ca channels individually to the superfusion medium21Carotid bodies were superfused with a low Ca - prior to challenge of the carotid bodies with appropriate

medium (i.e., medium containing 10 mM EGTA with gas mixtures. The release of SP during hypoxia, in the21 21omission of 1.8 mM Ca ) equilibrated with either room presence of Ca blockers was compared with that mea-

air or 12% O in N (hypoxia). Carotid bodies superfused sured in their absence. As shown in Fig. 7, nitrendipine2 221 21with medium containing Ca (1.8 mM), equilibrated with (1.5 mM), an inhibitor of L-type Ca channel, attenuated

either room air or hypoxic gas mixture served as controls. hypoxia-induced SP release by |65% (n58; P,0.05). On21As shown in Fig. 5, removal of extracellular Ca marked- the other hand, v-agatoxin TK (50 nM), a specific blocker

21ly attenuated or abolished hypoxia-induced SP release of P/Q-type Ca channel [30,47], attenuated SP release


tions suggest that activation of N- and L-types of voltage-21gated Ca -channels is involved in hypoxia-induced SP

release.

4. Discussion

In the present study, our objectives were to determinewhether hypoxia stimulates the release of SP in the carotidbody and to assess the role of calcium in this response. Ourresults demonstrated that hypoxia facilitates the release ofSP in the carotid body and that the magnitude of SPrelease depends on the severity of the hypoxic stimulus.

21Further, we showed that extracellular Ca is essential for21hypoxia-induced SP release and that voltage-gated Ca

channels, especially N- and L-type channels, are coupledto hypoxia-induced SP release in the carotid body.

We have developed an in vitro superfusion method for21 21Fig. 6. Influence of inorganic voltage-gated Ca channel blocker, Cd the analysis of SP release by hypoxia using rabbit carotid

on hypoxia-induced SP release from the carotid body. Data are21 bodies and a sensitive EIA assay specific for SP. We chosemean6S.E.M. from six experiments. Note that Cd significantly reduced

rabbit carotid body because it contains the highest con-the basal release, whereas it completely abolished hypoxia (12% O 1221 21N )-induced SP release. **Significant effects of Cd (Cd treatment centration of SP among the species studied so far [15,36].2

versus untreated control; P,0.01). The in vitro superfusion model provides a direct methodfor the determination of SP release as the release medium

during hypoxia but this effect was not statistically signifi- can be collected at regular intervals for the analysis of SP.cant (P.0.05; n58; Fig. 7). In contrast, v-conotoxin The EIA assay used in the present study is based on a rat

21GVIA (1 mM), a specific N-type Ca channel blocker anti-SP monoclonal antibody that is highly specific for SP[30,47] markedly attenuated hypoxia-induced SP release since cross-reactivity toward other tachykinin peptides likefrom the carotid body (P,0.001; n58). These observa- neurokinin-A or neurokinin B is either absent or negli-

gible. However, this antibody exhibits significant cross-reactivity toward C-terminally derived SP peptides thatinclude 2–11, 4–11, and 7–11 SP fragments. In ourprevious in vitro hydrolysis studies, we have found noevidence for the accumulation of C-terminally derived SPfragments in the carotid body [20,23]. Therefore, it isreasonable to assume that EIA-based analysis used in thisstudy provides a fairly accurate estimation of the contentof intact SP in the carotid body as well as in the releasemedium.

4.1. Hypoxia and SP release in the carotid body

A major finding of the present study is that hypoxiaenhances SP release in the carotid body and that themagnitude of the release is dependent on the intensity ofthe hypoxic stimulus. For instance, under hyperoxic con-dition (100% O ), SP release from the carotid body was2

40.662.0 fmol /min per mg protein, whereas during hyp-oxic challenge (i.e., 12% O in N ) SP release increased to2 2

76.263.3 fmol /min per mg protein. Notably, superfusionFig. 7. Influence of L-, P/Q-, and N-type calcium channel blockers onhypoxia-induced SP release from the carotid body. Data represent of the carotid bodies with an anoxic solution (i.e., 100%mean6S.E.M. from eight experiments. Note the marked reduction in N ) further increased SP release to 100.463.6 fmol /min2hypoxia (12% O 1N )-induced SP release in the presence of N-type2 2 per mg protein in SP release. Previously, using a radioim-calcium channel blocker. **Significant difference between the untreated

21 munoassay, Hanson et al. [15] noted a decrease in tissuecontrol and Ca channel blockers (P,0.05); n.s. denotes absence of21 content of both SP and met-enkephalin in the rabbit carotidsignificant effect between untreated control and Ca channel blockers

(P.0.05). body following intermittent hypoxia in vivo. These authors


21interpreted this decrease in peptide content as evidence for observation that divalent metal ion, Cd markedly at-the release of SP and met-enkephalin by hypoxia. Tissue tenuated hypoxia-induced SP release suggests that activa-concentration of neuropeptides, under a given physiologi- tion of voltage-gated Ca channels is coupled to SP release.cal condition, is dependent on their synthesis, precursor Thus far, four types of voltage-gated calcium channelsprocessing, storage, release and degradation. Thus, a have been described in rabbit carotid body. They are L-,decrease in tissue level of neuropeptides, in response to a N-, P/Q-, and resistant types, respectively [9,29,32]. Ourstimulus like hypoxia, may be attributed to changes in one observation that v-conotoxin markedly attenuated hypoxia-or more of the processes associated with peptide metabo- induced SP release suggests the involvement of N-type of

21lism per se. These potential complications in the interpreta- Ca channel in this response. Further, our results also21tion of results arising from the measurement of tissue indicate additional contribution from L-type of Ca

peptide concentration are avoided in our present study by channel to SP release during hypoxia. Notably, SP releasedetermining directly SP in the release medium. Thus, observed in the carotid body, however, differs from

21results presented in this study provide a direct evidence for catecholamine release during hypoxia wherein L-type CaSP release from the carotid body. channels are involved [14,29]. Taken together, these

observations suggest that although hypoxia stimulates the4.2. Factors affecting hypoxia-induced SF release in the release of both SP and catecholamines in the carotid body,carotid body the mechanism of release differs among them via differen-

tial recruitment of one or more of voltage-gated calciumSeveral factors seem to influence SP release in the channel sub-types.

carotid body during hypoxia. The magnitude of hypoxia-induced SP release depends on the severity of the hypoxic 4.4. Site(s) of SP release and SP-binding site(s) in the

2challenge, whereas extracellular CO –HCO has minimal carotid body2 3

effect. Previous studies have shown that extracellular2CO –HCO greatly improves the hypoxic response of the In the present study, we made the observation that2 3

cat carotid body by mechanism(s) involving intracellular SP-like immunoreactivity (SPLI) was associated with type2 2acidification mediated by the Cl –HCO exchanger I cell as well as afferent nerve fibers in the carotid body.3

2[16,46]. However, extracellular CO –HCO per se was This dual localization of SP-LI in the rabbit carotid body is2 3

not essential for the sensory response of the carotid body in contrast to a previous report wherein SP-LI is primarilyto hypoxia [31]. Our observation that hypoxia-induced SP localized to nerve fibers [24]. The apparent discrepancy inrelease remains the same both in the absence (Hepes the localization of SP-LI could be due to differences in

2buffer, pH 7.4) and presence of extracellular CO –HCO tissue fixation conditions employed in these studies prior to2 3

suggests that SP release is insensitive to potential changes immunocytochemical analysis. In the present study, wein intracellular pH. By contrast, catecholamine release in post-fixed the carotid body via immersion fixation, where-the carotid body appears to be dependent on extracellular as in the study of Kusakabe and his co-workers [24] thebicarbonate [33]. animals were perfused with fixative solution prior to

Extracellular calcium seems to be absolutely essential surgical removal of the carotid body. Our observations arefor hypoxia-induced SP release. Thus, omission of calcium in accord with the previous report showing significantand addition 10 mM EGTA to the superfusion medium number of type I cells in the cat carotid body are positiveeither markedly attenuated or abolished hypoxia-induced for SP-LI following fixation of the tissue via immersionSP release. Likewise, the release of catecholamines from fixation [49]. By contrast, when tissues were fixed bythe rabbit carotid bodies [8] as well as in PC 12 cells [22] perfusion technique, afferent nerve fibers and a lowduring hypoxia was markedly attenuated by low extracellu- percentage of type I cells showed positive SP-like im-lar calcium. Taken together, these observations suggest munoreactivity [4,40].that in the carotid body, calcium plays a central role in the It remains possible that hypoxia may facilitate therelease of neurochemicals especially SP and catechol- release of SP either from the type I cell or from theamines by hypoxia. afferent nerves that innervate type I cells or from both.

Based on our present findings, it was not possible to214.3. Ca -signaling mechanism(s) in hypoxia-induced SP determine from which of the two sources SP is released

release in the carotid body during hypoxia. Irrespective of the source of SP, thereleased SP may have diverse roles in carotid body

21Several lines of evidence support the notion that Ca functions. For instance, SP release from the type I cell mayplays a central role in hypoxia-induced SP release in the stimulate nerve discharge via possible activation of neuro-carotid body. As mentioned earlier, removal of extracellu- kinin receptors localized to the afferent nerve fiber.

21lar Ca blocked hypoxia-induced SP release but not the Although autoradiographic analysis showed the occurrencebasal release during normoxia, suggesting a low oxygen- of SP-receptors in the cat carotid body [36] it remains to

21mediated influx of Ca that precedes SP release. The be determined whether SP-receptors are expressed in


[10] R.S. Fitzgerald, Oxygen and carotid body chemotransduction: theafferent nerve terminals in the rabbit carotid body. On thecholinergic hypothesis — a brief history and new evaluation, Respir.other hand, SP released from the afferent nerve fibers mayPhysiol. 120 (2000) 89–104.

be involved in the regulation of secretory processes of type [11] R.S. Fitzgerald, M. Shirahata, H.Y. Wang, Acetylcholine releaseI cells in response to hypoxia. For instance, it may be from cat carotid bodies, Brain Res. 841 (1999) 53–61.involved in the modulation of release of neurochemicals, [12] G.E. Gibson, T. Manger, L. Toral-Barza, G. Freeman, Cytosolic-free

calcium and neurotransmitter release with decreased availability ofsuch as acetylcholine, as shown in the central nervousglucose or oxygen, Neurochem. Res. 14 (1989) 437–443.system [44].

[13] G. Gibson, L. Toral-Barza, H.M. Huang, Cytosolic free calciumIn summary, in the present study, we demonstrated thatconcentrations in synaptosomes during histotoxic hypoxia, Neuro-

SP-like immunoreactivity is localized in numerous type I chem. Res. 16 (1991) 461–467.cells as well as nerve fibers of the rabbit carotid body and [14] C. Gonzalez, L. Almaraz, A. Obeso, R. Rigual, Carotid body

chemoreceptors: from natural stimuli to sensory discharges, Physiol.that low oxygen stimulated SP release in the carotid body.Rev. 74 (1994) 829–898.Further, we showed that SP release during hypoxia is

21 [15] G. Hanson, L. Jones, S. Fidone, Physiological chemoreceptorcoupled to entry of extracellular Ca , via activation of N-stimulation decreases enkephalin and substance P in the carotid

and L-type voltage-gated calcium channels. Our present body, Peptides 7 (1986) 767–769.result that hypoxia augments SP release in the carotid body [16] R. Iturriaga, S. Lahiri, Carotid body chemoreception in the absence

and presence of CO -HCO , Brain Res. 568 (1991) 253–260.together with the previously reported physiological actions 2 3

[17] D.M. Jacobowitz, C.J. Helke, Localization of substance P immuno-of SP [36] provide a compelling evidence for the notionreactive nerves in the carotid body, Brain Res. Bull. 5 (1980)that SP plays important modulatory role(s) in the transmis-195–197.

sion of the sensory response of the carotid body to [18] B. Katz, R. Miledi, Spontaneous and evoked activity of motor nervehypoxia. endings in calcium Ringer, J. Physiol. (London) 203 (1969) 689–

706.[19] G.K. Kumar, Neuropeptide processing enzymes of the carotid body.

Biochemical and immunological characterization of carboxypep-Acknowledgements tidase activity, Adv. Exp. Med. Biol. 410 (1996) 319–323.

[20] G.K. Kumar, Peptidases of the peripheral chemoreceptors: bio-chemical, immunological, in vitro hydrolytic studies and electronThis study was supported by National Heart, Lung, andmicroscopic analysis of neutral endopeptidase-like activity of theBlood Institute grants HL-25830 (NRP) and HL-46462carotid body, Brain Res. 748 (1997) 39–50.

(GKK). [21] G.K. Kumar, Y.R. Kou, J.L. Overholt, N.R. Prabhakar, Involvementof substance P in neutral endopeptidase modulation of carotid bodysensory responses to hypoxia, J. Appl. Physiol. 88 (2000) 195–202.

[22] G.K. Kumar, J.L. Overholt, G.R. Bright, K.Y. Hui, H. Lu, M. Gratzl,ReferencesN.R. Prabhakar, Release of dopamine and norepinephrine byhypoxia from PC-12 cells, Am. J. Physiol. 274 (1998) C1592–

[1] I.L. Chen, R.D. Yates, J.T. Hansen, Substance P-like immuno- C1600.reactivity in rat and cat carotid bodies: light and electron microscopic [23] G.K. Kumar, M. Runold, R.D. Ghai, N.S. Cherniack, N.R.studies, Histol. Histopathol. 1 (1986) 203–212. Prabhakar, Occurrence of neutral endopeptidase activity in the cat

[2] T. Ciesiolka, H.J. Gabius, An 8- to 10-fold enhancement in carotid body and its significance in chemoreception, Brain Res. 517sensitivity for quantitation of proteins by modified application of (1990) 341–343.colloidal gold, Anal. Biochem. 168 (1988) 280–283. [24] T. Kusakabe, T. Kawakami, Y. Tanabe, S. Fujii, T. Takenaka,

[3] P.A. Cragg, M. Runold, Y.R. Kou, N.R. Prabhakar, Tachykinin Distribution of substance P-containing and catecholaminergic nerveantagonists in carotid body responses to hypoxia and substance P in fibers in the rabbit carotid body: an immunohistochemical study inthe rat, Respir. Physiol. 95 (1994) 295–310. combination with catecholamine fluorescent histochemistry, Arch.

[4] A.C. Cuello, D.S. McQueen, Substance P: a carotid body peptide, Histol. Cytol. 57 (1994) 193–199.Neurosci. Lett. 17 (1980) 215–219. [25] S. Lahiri, H. Acker, Redox-dependent binding of CO to heme

[5] F. De Castro, Sur la Structure et l’innervation de le glande protein controls PO -sensitive chemoreceptor discharge of the rat2

intercarotidienne (glomus caroticum) de l’homme et des mammi- carotid body, Respir. Physiol. 115 (1999) 169–177.feres, et sur un nouveau system d’innervation autonome dunerf [26] J. Lopez-Barneo, Oxygen-sensing by ion channels and the regulationglossopharyngien, Trab. Lab. Invest. Biol. Univ. Madrid 25 (1926) of cellular functions, Trends. Neurosci. 19 (1996) 435–440.365–432. [27] J. Lopez-Barneo, J.R. Lopez-Lopez, J. Urena, C. Gonzalez, Chemot-

1[6] C. Eyzaguirre, S.J. Fidone, Transduction mechanisms in carotid ransduction in the carotid body: K current modulated by PO in2

body: glomus cells, putative neurotransmitters and nerve endings, type I chemoreceptor cells, Science 241 (1988) 580–582.Am. J. Physiol. 239 (1980) C135–152. [28] D.S. McQueen, Effects of substance P on carotid chemoreceptor

[7] S. Fidone, C. Gonzalez, Initiation and control of chemoreceptor activity in the cat, J. Physiol. (London) 302 (1980) 31–47.activity in the carotid body, in: N.S. Cherniack, J.G. Widdicombe [29] A. Obeso, A. Rocher, S. Fidone, C. Gonzalez, The role of

21(Eds.), Section 3: The Respiratory System, Handbook of Physiology, dihydropyridine-sensitive Ca channels in stimulus-evoked cat-Vol. II, American Physiological Society, Bethesda, MD, 1986, pp. echolamine release from chemoreceptor cells of the carotid body,267–312. Neuroscience 47 (1992) 463–472.

[8] S. Fidone, C. Gonzalez, K. Yoshizaki, Effects of low oxygen on the [30] B.M. Olivera, G.P. Miljanich, J. Ramachandran, M.E. Adams,release of dopamine from the rabbit carotid body in vitro, J. Physiol. Calcium channel diversity and neurotransmitter release: the v-(London) 333 (1982) 93–110. conotoxins and v-agatoxins, Annu. Rev. Biochem. 63 (1994) 823–

[9] L.A. Fieber, E.W. McCleskey, L-type calcium channels in type I 867.cells of the rat carotid body, J. Neurophysiol. 70 (1993) 1378–1384. [31] S. Osanal, C. Rozanov, A. Mokashi, D.G. Buerk, S. Lahiri, CO


1 2interact with intracellular [H ] with and without CO -HCO in the [41] N.R. Prabhakar, J. Mitra, N.S. Cherniack, Role of substance P in2 3

cat carotid chemosensory discharge, Brain Res 764 (1997) 221–224. hypercapnic excitation of carotid chemoreceptors, J. Appl. Physiol.21[32] J.L. Overholt, N.R. Prabhakar, Ca current in rabbit carotid body 63 (1987) 2418–2425.

glomus cells is conducted by multiple types of high-voltage-acti- [42] N.R. Prabhakar, M. Runold, Y. Yamamoto, H. Lagercrantz, N.S.21vated Ca channels, J. Neurophysiol. 78 (1997) 2467–2474. Cherniack, C. von Euler, Role of the vagal afferents in substance

[33] J.M. Panisello, D.F. Donnelly, Catecholamine secretion from glomus P-induced respiratory responses in anaesthetized rabbits, Acta.cells is dependent on extracellular bicarbonate, Adv. Exp. Med. Physiol. Scand. 131 (1987) 63–71.Biol. 410 (1996) 267–273. [43] N.R. Prabhakar, M. Runold, Y. Yamamoto, H. Lagercrantz, C. von

[34] C. Peterson, G.E. Gibson, Synaptosomal calcium metabolism during Euler, Effect of substance P antagonist on the hypoxia-inducedhypoxia and 3,4-diaminopyridine treatment, J. Neurochem. 42 carotid chemoreceptor activity, Acta. Physiol. Scand. 121 (1984)(1984) 248–253. 301–303.

[35] J. Pizarro, M.L. Ryan, M.S. Hedrick, D.H. Xue, I.M. Keith, G.E. [44] Z. Preston, K. Lee, L. Widdowson, P.J. Richardson, R.D. Pinnock,3Bisgard, Intracarotid substance P infusion inhibits ventilation in the Tachykinins increase [ H]acetylcholine release in mouse striatum

goat, Respir. Physiol. 101 (1995) 11–22. through multiple receptor subtypes, Neuroscience 95 (2000) 367–[36] N.R. Prabhakar, Neurotransmitters in the carotid body, Adv. Exp. 376.

Med. Biol. 360 (1994) 57–69. [45] M. Scraggs, P. Smith, D. Heath, Glomic cells and their peptides in[37] N.R. Prabhakar, Oxygen sensing by the carotid body chemorecep- the carotid body of the human fetus, Pediatr. Pathol. 12 (1992)

tors, J. Appl. Physiol. 88 (2000) 2287–2295. 823–834.2[38] N.R. Prabhakar, H. Cao, J.A.d. Lowe, R.M. Snider, Selective [46] M. Shirahata, R.S. Fitzgerald, The presence of CO /HCO is2 3

inhibition of the carotid body sensory response to hypoxia by the essential for hypoxic chemotransduction in the in vivo perfusedsubstance P receptor antagonist CP-96,345, Proc. Natl. Acad. Sci. carotid body, Brain Res. 545 (1991) 297–300.USA 90 (1993) 10041–10045. [47] C.C. Toner, J.A. Stamford, Involvement of N- and P/Q- but not L-

[39] N.R. Prabhakar, Y.R. Kou, M. Runold, Chemoreceptor responses to or T-type voltage-gated calcium channels in ischaemia-inducedsubstance P, physalaemin and eledoisin: evidence for neurokinin-1 striatal dopamine release in vitro, Brain Res. 748 (1997) 85–92.receptors in the cat carotid body, Neurosci. Lett. 120 (1990) 183– [48] Z.Z. Wang, L. He, L.J. Stensaas, B.G. Dinger, S.J. Fidone, Localiza-186. tion and in vitro actions of atrial natriuretic peptide in the cat carotid

[40] N.R. Prabhakar, S.C. Landis, G.K. Kumar, D. Mullikin-Kilpatrick, body, J. Appl. Physiol. 70 (1991) 942–946.N.S. Cherniack, S. Leeman, Substance P and neurokinin A in the cat [49] Z.Z. Wang, L.J. Stensaas, B. Dinger, S.J. Fidone, The co-existencecarotid body: localization, exogenous effects and changes in content of biogenic amines and neuropeptides in the type I cells of the catin response to arterial PO , Brain Res. 481 (1989) 205–214. carotid body, Neuroscience 47 (1992) 473–480.2

Documents

Release of substance P by low oxygen in the rabbit carotid body: evidence for the involvement of calcium channels