点击显示 收起
Department of Pharmacology, Center of Biomolecular Medicine and Pharmacology, Medical University of Vienna, Vienna, Austria
Abstract
Presynaptic inhibition of transmitter release is commonly mediated by a direct interaction between G protein subunits and voltage-activated Ca2+ channels. To search for an alternative pathway, the mechanisms by which presynaptic bradykinin receptors mediate an inhibition of noradrenaline release from rat superior cervical ganglion neurons were investigated. The peptide reduced noradrenaline release triggered by K+-depolarization but not that evoked by ATP, with Ca2+ channels being blocked by Cd2+. Bradykinin also reduced Ca2+ current amplitudes measured at neuronal somata, and this effect was pertussis toxin-insensitive, voltage-independent, and developed slowly within 1 min. The inhibition of Ca2+ currents was abolished by a phospholipase C inhibitor, but it was not altered by a phospholipase A2 inhibitor, by the depletion of intracellular Ca2+ stores, or by the inactivation of protein kinase C or Rho proteins. In whole-cell recordings, the reduction of Ca2+ currents was irreversible but became reversible when 4 mM ATP or 0.2 mM dioctanoyl phosphatidylinositol-4,5-bisphosphate was included in the pipette solution. In contrast, the effect of bradykinin was entirely reversible in perforated-patch recordings but became irreversible when the resynthesis of phosphatidylinositol-4,5-bisphosphate was blocked. Thus, the inhibition of Ca2+ currents by bradykinin involved a consumption of phosphatidylinositol-4,5-bisphosphate by phospholipase C but no downstream effectors of this enzyme. The reduction of noradrenaline release by bradykinin was also abolished by the inhibition of phospholipase C or of the resynthesis of phosphatidylinositol-4,5-bisphosphate. These results show that the presynaptic inhibition was mediated by a closure of voltage-gated Ca2+ channels through depletion of membrane phosphatidylinositol bisphosphates via phospholipase C.
Via changes in the strength of synaptic transmission, the nervous system can adapt to alterations in the environment, a phenomenon that is generally referred to as neuromodulation. In this respect, the modulation of transmitter release via presynaptic receptors is of utmost importance, and a plethora of neuromodulators act via presynaptic G protein-coupled receptors (GPCR). In most, if not all, types of synapses, the activation of GPCRs was found to lead to a presynaptic inhibition of transmitter release, because activated G protein subunits directly interacted with voltage-activated Ca2+ channels (VACCs) and thereby reduced the Ca2+ influx required for vesicle exocytosis (Stevens, 2004). The precise mechanisms underlying the modulation of VACCs via GPCRs have been investigated in greatest detail in sympathetic neurons (Hille, 1994); there, the receptor-dependent activation of G proteins leads to an inhibition of Ca2+ currents (ICa) either via a direct, membrane-delimited and voltage-dependent interaction of G protein subunits with VACCs or via a second messenger system (Hille, 1994; Ikeda and Dunlap, 1999; Elmslie, 2003). In the experiments presented below, we used sympathetic neurons to delineate an example of presynaptic inhibition that relies on the modulation of VACCs through a second messenger system, but independently of a membrane-delimited action of G protein subunits.
In rat superior cervical ganglion (SCG) neurons, a large number of GPCRs including the prototypic 2-adrenoceptors mediate the voltage-dependent, membrane-delimited, -mediated inhibition of ICa, but only M1 muscarinic and AT1 angiotensin receptors were reported to reduce ICa in a voltage-independent manner via diffusible second messengers (Hille, 1994; Ikeda and Dunlap, 1999; Elmslie et al., 2003). These two latter receptors also use second messengers to inhibit M-type K+ (KM) channels (Hille, 1994). The underlying signal cascade remained obscure for decades but was recently evidenced to involve a phospholipase C (PLC)-dependent regulation of the membrane levels of phosphatidylinositol 4,5-bisphosphates (PIP2; Suh and Hille, 2002; Zhang et al., 2003; Winks et al., 2005). While the present work was in progress, the same signaling pathway was reported to mediate the inhibition of VACCs in SCG neurons via M1 receptors (Gamper et al., 2004). An inhibition of KM channels in SCG neurons has also been observed when B2 bradykinin receptors were activated (Jones et al., 1995), and this effect involved both the reduction of membrane PIP2 and inositol trisphophate-dependent increases in intracellular Ca2+ (Cruzblanca et al., 1998; Bofill-Cardona et al., 2000; Winks et al., 2005). Most recently, we found that bradykinin also caused a reduction of transmitter release from SCG neurons via presynaptic B2 receptors and an inhibition of ICa (Edelbauer et al., 2005). However, the signaling cascade mediating the modulation of VACCs was not elucidated, and it remained obscure whether the inhibition of VACCs also occurred at presynaptic sites and thus was the basis for the reduction of transmitter release. Here, we first demonstrate that the inhibition of presynaptic VACCs is involved in the repression of transmitter release from rat SCG neurons by bradykinin and then provide evidence that the inhibition of both VACCs and transmitter release involves a PLC-dependent reduction in membrane PIP2.
Materials and Methods
Primary Cultures of Rat Superior Cervical Ganglion Neurons. Primary cultures of dissociated SCG neurons from neonatal rats were prepared as described previously (Boehm, 1999). Ganglia were dissected from 2- to 6-day-old Sprague-Dawley rat pups that had been killed by decapitation in accordance with the rules of the university animal welfare committee. After incubation in collagenase (1.5 mg/ml; Sigma-Aldrich, Vienna, Austria) and dispase (3.0 mg/ml; Roche Diagnostics, Mannheim, Germany) for 45 min at 36°C, ganglia were trypsinized (0.25% trypsin; Worthington Biochemicals, Lakewood, NJ) for 20 min at 36°C, dissociated by trituration, and resuspended in Dulbecco's modified Eagle's medium (Invitrogen, Vienna, Austria) containing 2.2 g/l glucose, 10 mg/l insulin, 25,000 IU/l penicillin, 25 mg/l streptomycin (Invitrogen), 50 μg/l nerve growth factor (R&D Systems, Minneapolis, MN), and 5% fetal calf serum (Invitrogen). Cells were plated either onto 5-mm discs (approximately 40,000 cells per disc) for [3H]noradrenaline release or onto 35-mm culture dishes for electrophysiology. All tissue culture plastic was coated with rat tail collagen (Biomedical Technologies, Stoughton, MA). Cells were kept in a humidified 5% CO2 atmosphere at 36°C for up to 7 days, and one half of the medium was exchanged twice during this culture period. At 1 to 2 days before experiments, fresh medium without serum was added.
Determination of [3H]Noradrenaline Release. The release of [3H]noradrenaline was determined as described before (Boehm, 1999). Cultures were labeled with 0.05 μM [3H]noradrenaline (specific activity, 71.7 Ci/mmol) in culture medium supplemented with 1 mM ascorbic acid at 36°C for 1 h. After labeling, culture discs were transferred to small chambers and superfused with a buffer containing 120 mM NaCl, 6.0 mM KCl, 2.0 mM CaCl2, 2.0 mM MgCl2, 20 mM glucose, 10 mM HEPES, 0.5 mM fumaric acid, 5.0 mM sodium pyruvate, 0.57 mM ascorbic acid, and 0.001 mM desipramine, adjusted to pH 7.4 with NaOH. Superfusion was performed at 25°C at a rate of approximately 1.0 ml/min. The collection of 4-min superfusate fractions was started after a 60-min washout period. Tritium overflow was evoked during two consecutive stimulation periods (S1 and S2) by the inclusion of either 0.3 mM ATP or 40 mM KCl (NaCl was reduced accordingly to maintain isotonicity) in the buffer for 60 s. Radioactivity released in response to electrical field stimulation from rat sympathetic neurons after labeling with tritiated noradrenaline and under conditions similar to those of the present study had been shown previously to consist predominantly of the authentic transmitter and to contain only small amounts (15%) of metabolites (Schwartz and Malik, 1993). Hence, the outflow of tritium measured in this study was assumed to reflect primarily the release of noradrenaline and not that of metabolites.
Tetrodotoxin (TTX; 0.1 μM), CdCl2 (100 μM), and thapsigargin (0.3 μM), if appropriate, were added to the superfusion buffer after 50 min of superfusion (i.e., 10 min before the start of sample collection). Bradykinin (1 μM) and UK 14304 (1 μM) were added to the superfusion buffer 2 min before and phenylarsine oxide (10 μM) and dithiothreitol (1 mM) 4 min before the second stimulation period. At the end of experiments, the radioactivity remaining in the cells was extracted by immersion of the discs in 1.2 ml of 2% (v/v) perchloric acid followed by sonication. Radioactivity in extracts and collected fractions was determined by liquid scintillation counting (Packard Tri-Carb 2100 TR; PerkinElmer Life and Analytical Sciences, Boston, MA).
Electrophysiology. ICa of sympathetic neurons was determined as described previously (Boehm et al., 1996). Currents were recorded at room temperature (20–24°C) from single SCG neurons in vitro using an Axopatch 200B amplifier and the Pclamp 6.0 hard- and software (Axon Instruments Inc., Union City, CA). Currents were low-pass filtered at 5 kHz, digitized at 10 to 50 kHz, and stored on an IBM-compatible computer. Traces were analyzed offline by the Clampfit 8.1 program (Axon Instruments). Patch electrodes were pulled (Flaming-Brown puller; Sutter Instruments, Novato, CA) from borosilicate glass capillaries (Science Products, Frankfurt/Main, Germany). For perforated-patch recordings, electrodes were front-filled with a solution consisting of 130 mM CsCl, 20 mM tetraethylammonium chloride, 0.24 mM CaCl2, 10 mM glucose, 10 mM HEPES, and 5 mM EGTA, adjusted to pH 7.3 with KOH, and were then back-filled with the same solution containing 200 μg/ml amphotericin B (in 0.8% DMSO), which yielded tip resistances of 2 to 3 M. For whole-cell recordings, electrodes were filled with the solution used for front-filling, which additionally contained 2 mM magnesium ATP and 2 mM sodium GTP. Unless stated otherwise, all experiments were performed in the perforated-patch configuration. The external solution contained 120 mM NaCl, 20 mM tetraethylammonium chloride, 3 mM KCl, 2 mM MgCl2, 5 mM CaCl2, 20 mM glucose, and 10 mM HEPES, adjusted to pH 7.3 with KOH. This combination of solutions results in small liquid junction potentials of approximately +2 mV, which, however, were neglected. In a few experiments, 5 mM Ba2+ was used instead of Ca2+ as the charge carrier. Drugs were applied via a DAD-12 drug application device (Adams and List, Westbury, NY), which permits a complete exchange of solutions surrounding the cells under investigation within less than 100 ms (Boehm, 1999).
Unless stated otherwise, ICa was elicited every 15 s by 30-ms depolarizations from a holding potential of -80 to +10 mV. Leakage currents were corrected for by an online leak subtraction protocol, which applies four hyperpolarizing pulses before the depolarization to +10 mV to determine the extent of leakage. The extent of ICa inhibition by bradykinin was quantified according to the equation % inhibition = 100 - 100 x (B1 + B2)/(C1 + C2), where B1 and B2 are the peak current amplitudes determined 45 and 60 s after the start of bradykinin application, and C1 and C2 are the amplitudes of control currents measured directly before the bradykinin application. Recovery from inhibition was calculated using the equation: % recovery = 100 x [(W1 + W2) - (B1 + B2)]/[(C1 + C2) - (B1 + B2)], where W1 and W2 are the current amplitudes measured 135 and 150 s after the start of bradykinin washout (B1,2, C1,2, and W1,2 are indicated in Fig. 2A).
To determine the voltage dependence of inhibition, currents were elicited by a double-pulse voltage protocol (illustrated in Fig. 2D): cells were clamped at -80 mV; every 15 s, a 35-ms depolarization to +10 mV (ICa,-PP) was applied, followed by a 3-s waiting period at -80 mV, a 20-ms prepulse to +80 mV, a 10-ms repolarization to -80 mV, and finally another 35-ms depolarization to +10 mV (ICa,+PP). Facilitation was calculated as the ratio of ICa amplitudes measured before and after the prepulse, respectively, by using the equation: prepulse facilitation = ICa,+PP/ICa,-PP.
Statistics. All data represent arithmetic means ± S.E.M.; n represents the number of culture discs in [3H]noradrenaline-release experiments and the number of single neurons in electrophysiological experiments. Statistical significances between data points were evaluated by the nonparametric Mann-Whitney test.
Materials. (-)-[Ring-2,5,6-3H]noradrenaline was obtained from PerkinElmer Life and Analytical Sciences (Boston, MA); bradykinin, TTX, pertussis toxin (PTX), U73122 , U73343 , thapsigargin, Clostridium difficile toxin B, 7,7-dimethyl-5,8-eicosadienoic acid (DEDA), LY 294,002, wortmannin, and UK 14304 were from Sigma-Aldrich; phorbol-12-myristate-13-acetate (PMA) and GF 109203X were from Calbiochem (Bad Soden, Germany); BAPTA-AM was from Molecular Probes (Eugene, OR); and dioctanoyl phosphatidyl-4,5-bisphosphate (diC8-PIP2) was from Cayman Chemicals (Ann Arbor, MI).
Results
Bradykinin Reduces the Release of [3H]Noradrenaline Evoked by K+ but Not That Evoked by ATP in the Presence of Cd2+. SCG neurons labeled with [3H]noradrenaline steadily released small amounts of tritium into the superfusion buffer when excess radioactivity had been removed during a 60-min washout period. The buffer contained 0.1 μM TTX to isolate drug effects that occurred at nerve terminals from those at remote sites, such as axons or neuronal somata, and to thereby prevent the release-stimulating action of bradykinin (Boehm and Huck, 1997; Edelbauer et al., 2005). The spontaneous release of radioactivity into the superfusion buffer per 4-min collection period amounted to 1.03 ± 0.19% of total tritium in the cultures (n = 30), which corresponded to 0.22 ± 0.04 nCi.
Depolarization of SCG neurons by high K+ concentrations induces the release of previously incorporated [3H]noradrenaline by eliciting transmembrane Ca2+ influx via VACCs, which can be abolished by 100 μMCd2+. Application of 0.3 mM ATP in the continuing presence of 100 μMCd2+,in contrast, causes Ca2+ entry via P2X receptors and thereby triggers [3H]noradrenaline release independently of VACCs (Boehm, 1999). Exposure to either 40 mM KCl (NaCl was reduced to maintain isotonicity) or 0.3 mM ATP in the continuing presence of Cd2+ during two consecutive periods of stimulation (S1 and S2) caused reproducible increases in tritium release (Fig. 1A), which lacked in the absence of extracellular Ca2+ (data not shown; Boehm, 1999). When bradykinin (1 μM) was present during the second period of K+ stimulation (S2), the amount of released tritium was reduced (Fig. 1A). Therefore, the ratio of the amount of tritium efflux triggered during the two stimulation periods (S2/S1) was decreased from 0.84 ± 0.03 (n = 9) in the absence of the peptide to 0.46 ± 0.06 (n = 9; p < 0.01) in its presence. In contrast, when ATP was used to stimulate tritium efflux, bradykinin failed to cause an inhibition, and the S2/S1 ratio remained unchanged [0.59 ± 0.04 in the absence (n = 12) and 0.58 ± 0.07 in the presence of bradykinin (n = 9); Fig. 1B].
2-Adrenoceptors are well known to reduce noradrenaline release from sympathetic neurons through an inhibition of VACCs (Boehm and Kubista, 2002), and the 2-adrenergic agonist UK 14304 causes presynaptic inhibition in SCG neurons only when VACCs are involved in excitation-secretion coupling (Boehm and Huck, 1995). In accordance with that observation, 1 μM UK 14304 diminished the S2/S1 ratio of K+-evoked but not that of ATP-evoked tritium efflux (Fig. 1B). Hence, activation of both presynaptic bradykinin B2 and 2-adrenergic receptors only reduced [3H]noradrenaline release, when transmembrane Ca2+ entry occurred via VACCs. The presynaptic inhibition of transmitter release was presumably mediated by an inhibition of VACCs.
Bradykinin Induces a Slow, Voltage-Independent, and PTX-Resistant Inhibition of ICa. The above results demonstrate that the inhibition of VACCs is the mechanism by which bradykinin reduces sympathetic transmitter release. We therefore investigated the signal cascade by which bradykinin inhibits VACCs. To this end, we first determined some of the basic parameters of the inhibitory action of bradykinin on ICa and compared them with those of the 2-adrenergic agonist UK 14304. Evidence has been provided that GPCRs may use different signaling pathways to regulate ICa in rat SCG neurons, depending on the recording technique (Filippov et al., 2003). Therefore, the initial set of experiments was performed in the whole-cell and perforatedpatch configuration. In whole-cell, 1 μM bradykinin reduced ICa by 45.04 ± 5.16% (n = 11), and the reduction of current amplitudes was maximal after 45 to 60 s (Fig. 2A). For comparison, UK 14304 (10 μM) reduced whole-cell ICa by 47.95 ± 5.23% (n = 13), and this effect was maximal after <15 s (data not shown). In the perforated-patch configuration, the reduction of ICa by 1 μM bradykinin was again maximal after 45 to 60 s and amounted to 44.28 ± 11.2% (n = 8) inhibition (Fig. 2A). In such recordings, UK 14304 reduced ICa by 49.87 ± 11.13% (n = 6), and the effect was maximal after <15 s (data not shown). With respect to the recovery from the inhibition by bradykinin, however, there were significant differences between the two recording techniques: during a 150-s washout period, ICa only partially recovered from the inhibition in whole-cell recordings (24.95 ± 14.78% recovery), whereas the inhibition was fully reversible in perforated-patch recordings (102.39 ± 11.85% recovery) (Fig. 2, A and B). For comparison, the inhibition by UK 14304 was entirely reversible within 30 s in both types of recordings (data not shown).
The inhibition of ICa in SCG neurons via the membrane-delimited pathway is voltage-dependent and thus decreases the more the cell is depolarized (Hille, 1994). In the present experiments, however, the reduction of ICa amplitudes by bradykinin was approximately the same at voltages between -30 and +50 mV (Fig. 2C). The voltage-dependent membrane-delimited inhibition of ICa is also characterized by a slowing of activation kinetics, which is attenuated by brief depolarizing prepulses, a phenomenon called "prepulse facilitation" (Hille, 1994). The prepulse facilitation of the bradykinin inhibition was assessed by a double-pulse protocol and compared with that of the 2-adrenergic inhibition caused by UK 14304. The bradykinin inhibition was neither accompanied by a slowing of activation kinetics nor attenuated by the prepulse (Fig. 2D). Therefore, the values of prepulse facilitation in the presence of bradykinin were not different from those in the absence of the peptide. In contrast, in the presence of 10 μM UK 14304, a marked prepulse facilitation was observed. The extent of prepulse facilitation in the absence or presence of receptor agonists was not different between whole-cell and perforated-patch recordings (Fig. 2E). When neurons had been treated for 24 h with 100 ng/ml PTX, the inhibition by UK 14304 was completely abolished (0.94 ± 1.18% inhibition; n = 7), but the inhibition by bradykinin remained unaffected (44.52 ± 6.45% inhibition; n = 7).
PLC, but None of Its Downstream Effector Systems, Is Required for the Inhibition of ICa. In rat SCG neurons, the actions of bradykinin are commonly mediated by Gq/11 proteins and PLC- (Haley et al., 2000; Scholze et al., 2002). The inhibition of PLC in SCG neurons by U73122 prevents the formation of inositol phosphates in response to bradykinin, an effect that is not observed with the inactive analog U73343 (Bofill-Cardona et al., 2000). Here, the inhibition of ICa, as determined in perforated-patch recordings, was largely attenuated when neurons had been incubated in 3 μM U73122 for 30 min. An equivalent incubation in U73343 , however, did not alter the ICa inhibition by bradykinin (Fig. 3).
Via PLC-, bradykinin triggers the synthesis of inositol trisphosphate (IP3) and diacylglycerol, which in turn cause Ca2+ release from the endoplasmic reticulum and activation of protein kinase C (PKC) (Cruzblanca et al., 1998; Scholze et al., 2002), respectively. However, when neurons had been treated with the Ca2+ ATPase inhibitor thapsigargin (0.3 μM) for 30 min to deplete the intracellular Ca2+ stores (Foucart et al., 1995), the peptide reduced ICa to the same extent as under control conditions (Fig. 3). Likewise, in neurons incubated for at least 30 min in 3 μM concentration of the cell-permeant Ca2+ chelator BAPTA-AM (followed by a 30-min incubation in regular buffer to permit hydrolysis of the acetoxymethylester), bradykinin diminished current amplitudes again by approximately 50% (Fig. 3). However, these two latter manipulations did abolish the inhibition of IM by bradykinin (data not shown; Bofill-Cardona et al., 2000). These results indicate that the ICa inhibition did not require the release of Ca2+ from its stores into the cytosol. To investigate whether Ca2+ entry via VACCs was essential, whole-cell recordings were performed with Ba2+ instead of Ca2+ as charge carrier. In these experiments, bradykinin reduced current amplitudes by only 15.50 ± 3.02% (n = 9) compared with the 36.37 ± 5.05% inhibition with Ca2+ as charge carrier (n = 7; p < 0.01). Hence, the flow of Ca2+ ions through VACCs was required to permit maximal current inhibition by bradykinin.
Exposure of the neurons to the PKC inhibitor GF 109203X (0.3 μM for 30 min), which abrogates the excitatory actions of bradykinin in SCG neurons (Scholze et al., 2002), failed to alter the reduction of ICa. Therefore, a pretreatment with 1 μM PMA for 24 h, which eliminates classic and novel PKC isoforms from SCG neurons (Scholze et al., 2002), did not attenuate the inhibitory action of bradykinin (Fig. 3).
In neuroblastoma-glioma hybrid (NG108-15) cells, bradykinin inhibits ICa via the monomeric G proteins Rac1 and/or Cdc42 (Wilk-Blaszczak et al., 1997). To test for a role of this signaling pathway, cultures were treated for at least 6 h with 50 ng/ml C. difficile toxin B, which inactivates members of the Rho protein family, such as Rac1 and Cdc42, by glycosylation (Just et al., 1995). However, the inhibitory action of bradykinin remained unaffected (Fig. 3), although the toxin led to a shape change and to detachment of the neurons after prolonged exposure (>7 h).
The inhibition of ICa in SCG neurons via M1 muscarinic receptors involves phospholipase A2 (PLA2) (Liu et al., 2004), and this enzyme has also been reported to mediate cellular effects of bradykinin (Burch and Axelrod, 1987). However, the inhibition of ICa by bradykinin was not altered in the presence of 50 μM DEDA (Fig. 3), which efficiently reduces the enzymatic activity of PLA2 (Lister et al., 1989).
Recovery from the Inhibition of ICa Requires Lipid Kinase Activity or Intracellular PIP2. Because none of the typical downstream effector systems of PLC- seemed to be involved in the actions of bradykinin, we reasoned that a loss of membrane PIP2 might be responsible for the reduction of ICa. If that was the case, PIP2 resynthesis should be a prerequisite for the recovery from inhibition, and this requires ATP and lipid kinases, as demonstrated for the recovery of KM channels from the inhibition via muscarinic receptors in SCG neurons (Suh and Hille, 2002; Zhang et al., 2003; Winks et al., 2005). Therefore, we investigated the effects of the lipid kinase inhibitors wortmannin (Nakanishi et al., 1995) and phenylarsine oxide (PAO) (Wiedemann et al., 1996) on the recovery from inhibition in perforated-patch recordings. When bradykinin was applied in the continuous presence of wortmannin (50 μM) or PAO (30 μM), ICa no longer recovered from the inhibition (Fig. 4, A, B, and E). To test for unspecific effects of PAO, this agent was incubated and applied together with 1 mM dithiothreitol (DTT), which traps PAO in stable inactive complexes (Schaefer et al., 1994). DTT entirely prevented the inhibitory action of PAO on ICa recovery (Fig. 4, C and E). Besides inhibiting phosphatidylinositol 4-kinase (PI4-kinase), wortmannin also inhibits phosphatidylinositol 3-kinase (PI3-kinase) (Nakanishi et al., 1995). To verify that the loss of recovery in the presence of wortmannin was not caused by the inhibition of PI3-kinase, we investigated the effect of LY 294,002, a selective PI3-kinase inhibitor (Vlahos et al., 1994). Cultures were pretreated with 100 μM LY 294,002 for 1 h, and then bradykinin was applied in the continuous presence of the inhibitor. Under these conditions, ICa displayed clear-cut recovery (Fig. 4, D and E). Because PAO was shown to inhibit PI4-kinase but not phosphatidylinositol 5-kinase activity (Wiedemann et al., 1998), the coincident inhibition of recovery by wortmannin and PAO but not by LY 294,002 suggests a major role for PI4-kinase.
The enzyme activity of PI4-kinase requires high intracellular ATP concentrations (Balla, 1998), and the recovery of KM channels from the inhibition caused by PIP2 depletion depended on the ATP concentration in the intracellular recording solution (Suh and Hille, 2002). When ATP in our pipette solution for whole-cell recordings was increased from the 2 mM standard concentration to 4 mM, the recovery of ICa from the bradykinin inhibition increased from 20% to >80% (Fig. 5A). The inner leaflet of the plasma membrane of SCG neurons had been calculated to contain 261 μM PIP2 under resting conditions (Winks et al., 2005). The addition of a similar concentration (200 μM) of the more soluble diC8-PIP2 to the whole-cell intracellular solution also led to a significant increase in recovery (Fig. 5B). Thus, the supply of either PI4-kinase substrate or of the product of phosphatidylinositol-4-phosphate kinase is sufficient to re-establish the recovery of whole-cell ICa from the inhibition by bradykinin.
The above manipulations altered the recovery of ICa from the inhibition by bradykinin. In contrast, the extent of ICa inhibition caused by the peptide was not significantly changed under any of these conditions (Table 1). Hence, a continuous supply of PIP2 to the membrane does not seem to be a prerequisite for the inhibition of ICa by bradykinin under the present electrophysiological recording conditions.
The Inhibition of [3H]Noradrenaline Release Involves PLC Activity and Changes in PIP2. The data shown above indicate that PLC-dependent changes in PIP2 mediate the inhibition of ICa by bradykinin. Because the inhibition of VACCs was the basis for the inhibition of noradrenaline release (Fig. 1), the same mechanisms should also be involved in the reduction of transmitter release by bradykinin. To test for this hypothesis, the inhibition of K+-evoked tritium efflux was investigated in SCG neurons loaded with [3H]noradrenaline and treated either with the PLC inhibitor U73122 (3 μM) or with its inactive analog U73343 (3 μM), both for 1 h. These two agents cannot be used during the determination of [3H]noradrenaline release because they cause large increases in spontaneous tritium outflow (Scholze et al., 2002). However, U73122 has been found to block bradykinin effects in SCG neurons mediated by PLC in an irreversible manner (Bofill-Cardona et al., 2000). In U73122 -treated cultures, bradykinin (1 μM) failed to diminish the S2/S1 ratio of tritium efflux, but the peptide caused a significant inhibition in cultures treated with U73343 (Fig. 6A). For comparison, the inhibition of 3H efflux by the 2-adrenergic agonist UK 14304 (1 μM) was also investigated in cultures treated with either U73122 (3 μM) or its inactive analog. In that case, the results obtained after the two treatment procedures were not different from each other (Fig. 6A). Hence, only the presynaptic inhibition by bradykinin, but not that by an 2-adrenergic agonist, involved an activation of PLC.
To test for a role of PLC products in the inhibition of noradrenaline release by bradykinin, cultures were continuously superfused with 0.3 μM thapsigargin, a concentration that abolishes the inhibition of KM channels by the peptide (Bofill-Cardona et al., 2000). However, bradykinin reduced K+-evoked tritium efflux in the presence of thapsigargin to approximately the same extent as in its absence (Fig. 6B). We found previously that the inhibition of noradrenaline release by bradykinin was not attenuated when protein kinase C had been inhibited (Edelbauer et al., 2005). Hence, there was no evidence for a role of PLC products in the presynaptic inhibition caused by bradykinin.
To investigate whether changes in PIP2 might be involved in the reduction of [3H]noradrenaline release by bradykinin, PAO (10 μM) was applied before and during the second K+ stimulation, either alone or together with bradykinin. PAO per se reduced the S2/S1 ratio of K+-evoked tritium efflux and prevented an additional inhibitory effect of bradykinin (Fig. 6C). The effects of PAO on ICa were prevented by DTT. Therefore, when PAO was applied together with 1 mM DTT, it neither caused a significant reduction of K+-evoked tritium efflux nor prevented the inhibitory action of bradykinin (Fig. 6C). Once again, UK 14304 (1 μM) was used instead of bradykinin for comparison, and the inhibitory effect of the 2-adrenergic agonist turned out to be additive to that of PAO (Fig. 6C).
Discussion
In SCG and other postganglionic sympathetic neurons, a large number of GPCRs mediate presynaptic inhibition of noradrenaline release, on one hand, and a G protein-dependent inhibition of VACCs, on the other hand. All of the receptor subtypes that mediate both effects are linked to VACCs via a membrane-delimited interaction between G protein subunits and Ca2+-channel proteins (Koh and Hille, 1997; Boehm and Kubista, 2002). Most recently, B2 bradykinin receptors were added to this list of inhibitory presynaptic receptors (Edelbauer et al., 2005), and the present results demonstrate that the associated signaling cascade is a novel one and is definitely distinct from those described before.
Presynaptic P2X receptors of rat SCG neurons are highly Ca2+ permeable and thereby provide a route for transmembrane Ca2+ entry to trigger transmitter release that is independent of VACCs (Boehm, 1999). This type of stimulated noradrenaline release was altered neither by bradykinin nor by an 2-adrenergic agonist, but both agents did reduce noradrenaline release elicited by depolarizing K+ concentrations. This confirms that 2-adrenergic receptors mediate a presynaptic inhibition of transmitter release from SCG neurons via an inhibition of VACCs (Boehm and Huck, 1995) and demonstrates that the same signaling mechanism is used by the inhibitory presynaptic B2 receptors (Edelbauer et al., 2005). Nevertheless, the signal cascade that linked B2 receptors to VACCs was clearly different from that of 2-adrenoceptors: the reduction of ICa by bradykinin was not altered by large depolarizing prepulses nor by a treatment of the neurons with pertussis toxin, two manipulations that attenuated or abolished the reduction of ICa by an 2-adrenergic agonist. Hence, bradykinin controls VACCs via mechanisms different from those of 2-adrenoceptors which do so through a direct interaction of G protein subunits with channel proteins (Ikeda and Dunlap, 1999; Elmslie et al., 2003).
In sympathetic neurons, bradykinin inhibits not only VACCs, as described here, but also KM channels (Jones et al., 1995). This latter effect involves PLC-dependent decreases in membrane PIP2 and IP3-dependent increases in intracellular Ca2+ (Cruzblanca et al., 1998; Bofill-Cardona et al., 2000; Winks et al., 2005). Only the former mechanism, the decrease in membrane PIP2, was responsible for the bradykinin-dependent inhibition of VACCs, as revealed by the following results: 1) the PLC inhibitor U73122 almost abolished the inhibition of ICa, but the inactive analog U73343 had no effect; 2) the depletion of intracellular Ca2+ stores by thapsigargin or the chelation of intracellular Ca2+ ions by BAPTA failed to alter the bradykinin inhibition of ICa. Nevertheless, Ca2+ influx via VACCs was required for an efficient inhibition of the channels, because Ba2+ currents were less affected by the peptide than Ca2+ currents; 3) the inhibition of protein kinase C, whether by GF 109203X or by long-term phorbol ester treatment, did not attenuate the inhibitory effect of bradykinin on ICa; 4) the inactivation of Rho proteins, which mediate the bradykinin inhibition of VACCs in neuroblastoma cells (Wilk-Blaszczak et al., 1997), did not affect the reduction of ICa in SCG neurons; and 5) the inhibition of PLA2 by DEDA failed to alter the ICa reduction by bradykinin, although this enzyme is involved in the modulation of VACCs in SCG neurons via M1 muscarinic acetylcholine receptors (Liu et al., 2004). Taken together, bradykinin inhibited ICa via PLC, but it did so independently of downstream effectors or alternative signaling cascades.
The reduction of membrane PIP2 by bradykinin has been demonstrated most recently in SCG neurons (Gamper et al., 2004; Winks et al., 2005). The kinetics of this effect have been studied in detail in neuroblastoma cells (Xu et al., 2003). There, the peptide caused a short (<10 s) transient increase in PIP2 and a subsequent decrease that was maximal after 30 s to 1 min. This was followed by a slow resynthesis that brought PIP2 levels back to control within up to 3 min. This time course of PIP2 depletion was much slower than the increase in IP3, which was maximal within less than 10 s (Xu et al., 2003). Hence, the kinetics of the bradykinin-induced changes in membrane PIP2, but not the changes in PLC products, paralleled the time course of the inhibition of ICa as observed here. In SCG neurons, the concentration of membrane PIP2 is directly correlated with the conductance of KM channels (Winks et al., 2005), and the time course of KM channel inhibition by bradykinin (Jones et al., 1995) is almost congruent with the present time course of ICa inhibition. Recombinant VACCs, in particular P/Q- (Wu et al., 2002) and N-type (Gamper et al., 2004) channels, were also found to be regulated by the membrane PIP2 concentration. Sympathetic neurons express N-type, rather than P/Q-type, Ca2+ channels, and the inhibition of ICa in sympathetic neurons by luteinizing hormone-releasing hormone was reported to be PLC-dependent (Wu et al., 2002). In addition, evidence has been presented that activation of M1 muscarinic receptors inhibited VACCs in rat SCG neurons through a depletion of membrane PIP2 (Gamper et al., 2004). Moreover, bradykinin was found to enhance currents through VR1 receptors via a depletion of membrane PIP2 (Chuang et al., 2001). Taken together, there is experimental evidence showing that bradykinin receptors control the membrane levels of PIP2 and that a number of ion channels, including N-type VACCs, are regulated by PIP2. In the present experiments, the recovery of ICa from the inhibition by bradykinin required replenishment of membrane PIP2, as evidenced by the following results: 1) in whole-cell recordings, high intracellular ATP concentrations were required for full recovery of ICa, and they are necessary for maximal PI4-kinase activity (Balla, 1998); 2) independently of the ATP concentration, the addition of dic8-PIP2 to the whole-cell intracellular solution enabled the recovery of ICa; and 3) inhibition of PI4-kinase activity by wortmannin or PAO prevented the recovery of ICa from the inhibition by bradykinin, which was otherwise seen in perforated-patch recordings. Thus, together with the previous observations, the present results indicate that a PLC-induced reduction of membrane PIP2 mediates the bradykinin inhibition of VACCs.
Bradykinin was recently reported to inhibit VACCs in SCG neurons only when the resynthesis of PIP2 was blocked (Gamper et al., 2004). In the present experiments, however, the peptide reduced ICa whether lipid kinases were inhibited or not, and the PI4-kinase inhibitors affected only the recovery of ICa, but not the extent of inhibition elicited by bradykinin. Although these results are somehow contradictory, they both do confirm that VACCs are controlled by the relation of PIP2 catalysis and synthesis. Nevertheless, it remains elusive why the efficiency of bradykinin receptors in mediating the depletion of membrane PIP2 may differ. Causative factors for the observed discrepancies might be 1) differences in the composition of culture media and sera because growth factors also control membrane PIP2 (Chuang et al., 2001); 2) differences in the buffering of the intracellular Ca2+ concentration as the PIP2 synthesis may be altered by changes in intracellular Ca2+ (Gamper et al., 2004; Winks et al., 2005); 3) differences in intracellular Mg2+, millimolar concentrations of which are required for maximal G protein signaling (Suh et al., 2004); and 4) differences in the voltage protocols used to elicit ICa because repeated depolarizations are known to reduce membrane PIP2 (Micheva et al., 2001).
The inhibition of VACCs was found to be a prerequisite for the reduction of transmitter release by bradykinin. Therefore, the PLC-mediated depletion of membrane PIP2, as involved in the inhibition of ICa at neuronal somata, should also play a role in the presynaptic inhibition. In accordance with this expectation, the active PLC inhibitor U73122 abolished the reduction of noradrenaline release by bradykinin without affecting the inhibitory action of the 2 agonist. However, inhibition of protein kinase C (Edelbauer et al., 2005) or depletion of intracellular Ca2+ stores by thapsigargin left the inhibition of noradrenaline release by bradykinin unaltered. In contrast, inhibition of lipid kinases by PAO reduced noradrenaline release and prevented a further reduction by bradykinin. The 2-adrenergic agonist, however, did reduce release even in the presence of PAO. All of the effects of PAO on noradrenaline release were abolished by DTT, which also prevents the inhibition of lipid kinases (Schaefer et al., 1994). These results indicate that not the products of PLC but rather changes in the substrate, PIP2, are involved in the presynaptic inhibition by bradykinin. Nevertheless, these results are not congruent with those obtained for the inhibition of ICa: PAO did reduce noradrenaline release and prevented a further reduction by bradykinin, two effects not observed with ICa. Several facts may explain these apparent discrepancies. First, lipid kinases that are inhibited by PAO are associated not only with the plasma membrane (Micheva et al., 2001) but also with vesicle membranes (Wiedemann et al., 1996). Therefore, the inhibition of both pools of enzymes may contribute to the inhibition of release, but only one pool can be involved in the regulation of ICa. Second, repeated or prolonged depolarizing stimuli are known to strongly reduce the PIP2 contents of presynaptic membranes, and its resynthesis is blocked by PAO (Micheva et al., 2001). Therefore, our 1-min K+ stimulation to trigger noradrenaline release will decrease PIP2 in the presynaptic membrane, but the 30-ms depolarizations used to evoke ICa are unlikely to do so. As a consequence, bradykinin may be able to reduce ICa through PIP2 depletion in current recordings from neuronal somata but fails to further reduce the PIP2 associated with presynaptic VACCs during the stimulation of noradrenaline release. Third, VACCs are highly concentrated at the sites of vesicle exocytosis (Stevens, 2004), and there they are clustered in lipid microdomains (Taverna et al., 2004). It thus seems reasonable to assume that the quantitative relation between membrane PIP2 and VACCs is different between neuronal somata and presynaptic nerve terminals. Nevertheless, in both locations, bradykinin controls the function of VACCs through PLC-dependent changes in PIP2, as indicated by the present results. In conclusion, our results show that bradykinin inhibits VACCs of sympathetic neurons through a PLC-mediated depletion of membrane PIP2 and demonstrate that this effect provides a novel mechanism for the presynaptic inhibition of transmitter release via GPCRs.
Footnotes
This study was supported by the "Virologiefonds" of the Medical University of Vienna and by the Austrian Science Fund (Fonds zur Frderung der Wissenschaftlichen Forschung P15797 and P17611 ).
ABBREVIATIONS: GPCR, G protein-coupled receptor; ICa, Ca2+ current; DEDA, 7,7-dimethyl-5,8-eicosadienoic acid; diC8-PIP2, dioctanoyl phosphatidyl-4,5-bisphosphate; GF 109203X, bisindolylmaleimide I; IP3, inositol trisphosphate; KM channel, M-type K+ channel; LY 294,002, 2-(4-morpholinyl)-8-phenyl-4H-1-benzopyran-4-one; PIP2, phosphatidylinositol 4,5-bisphosphate; PI3-kinase, phosphatidylinositol 3-kinase; PI4-kinase, phosphatidylinositol 4-kinase; PKC, protein kinase C; DTT, dithiothreitol; PAO, phenylarsine oxide; DMSO, dimethyl sulfoxide; PLA2, phospholipase A2; PLC, phospholipase C; PMA, phorbol-12-myristate-13-acetate; PTX, pertussis toxin; SCG, superior cervical ganglion; TTX, tetrodotoxin; U73122 , 1-[6-[[(17)-3-methoxyestra-1,3,5(10)-trien-17-yl]amino]hexyl]-1H-pyrrole-2,5-dione; U73343 , 1-[6-[[(17)-3-methoxyestra-1,3,5(10)-trien-17-yl]amino]hexyl]-2,5-pyrollidinedione; VACC, voltage-activated Ca2+ channel; UK 14304, 5-bromo-N-(4,5-dihydro-1H-imidazol-2-yl)-6-quinoxalinamine; BAPTA-AM, 1,2-bis(2-aminophenoxy)ethane-N,N,N,N-tetraacetic acid acetoxymethyl ester.
References
Balla T (1998) Phosphatidylinositol 4-kinases. Biochim Biophys Acta 1436: 69-85.
Boehm S (1999) ATP stimulates sympathetic transmitter release via presynaptic P2X purinoceptors. J Neurosci 19: 737-746.
Boehm S and Huck S (1995) Alpha2-adrenoreceptor-mediated inhibition of acetylcholine-induced noradrenaline release from rat sympathetic neurons: an action at voltage-gated Ca2+ channels. Neuroscience 69: 221-231.
Boehm S and Huck S (1997) Noradrenaline release from rat sympathetic neurones triggered by activation of B2 bradykinin receptors. Br J Pharmacol 122: 455-462.
Boehm S, Huck S, and Freissmuth M (1996) Involvement of a phorbol esterinsensitive protein kinase C in the alpha2-adrenergic inhibition of voltage-gated calcium current in chick sympathetic neurons. J Neurosci 16: 4596-4603.
Boehm S and Kubista H (2002) Fine tuning of sympathetic transmitter release via ionotropic and metabotropic presynaptic receptors. Pharmacol Rev 54: 43-99.
Bofill-Cardona E, Vartian N, Nanoff C, Freissmuth M, and Boehm S (2000) Two different signaling mechanisms involved in the excitation of rat sympathetic neurons by uridine nucleotides. Mol Pharmacol 57: 1165-1172.
Burch RM and Axelrod J (1987) Dissociation of bradykinin-induced prostaglandin formation from phosphatidylinositol turnover in Swiss 3T3 fibroblasts: evidence for G protein regulation of phospholipase A2. Proc Natl Acad Sci USA 84: 6374-6378.
Chuang HH, Prescott ED, Kong H, Shields S, Jordt SE, Basbaum AI, Chao MV, and Julius D (2001) Bradykinin and nerve growth factor release the capsaicin receptor from PtdIns(4,5)P2-mediated inhibition. Nature (Lond) 411: 957-962.
Cruzblanca H, Koh DS, and Hille B (1998) Bradykinin inhibits M current via phospholipase C and Ca2+ release from IP3-sensitive Ca2+ stores in rat sympathetic neurons. Proc Natl Acad Sci USA 95: 7151-7156.
Edelbauer H, Lechner SG, Mayer M, Scholze T, and Boehm S (2005) Presynaptic inhibition of transmitter release from rat sympathetic neurons by bradykinin. J Neurochem 93: 1110-1121.
Elmslie KS (2003) Neurotransmitter modulation of neuronal calcium channels. J Bioenerg Biomembr 35: 477-489.
Filippov AK, Simon J, Barnard EA, and Brown DA (2003) Coupling of the nucleotide P2Y4 receptor to neuronal ion channels. Br J Pharmacol 138: 400-406.
Foucart S, Gibbons SJ, Brorson JR, and Miller R (1995) Increases in [Ca2+]i by CCh in adult rat sympathetic neurons are not dependent on intracellular Ca2+ pools. Am J Physiol 268: C829-C837.
Gamper N, Reznikov V, Yamada Y, Yang J, and Shapiro MS (2004) Phosphotidylinositol 4,5-bisphosphate signals underlie receptor-specific Gq/11-mediated modulation of N-type Ca2+ channels. J Neurosci 24: 10980-10992.
Haley JE, Abogadie FC, Fernandez-Fernandez JM, Dayrell M, Vallis Y, Buckley NJ, and Brown DA (2000) Bradykinin, but not muscarinic, inhibition of M-current in rat sympathetic ganglion neurons involves phospholipase C-beta 4. J Neurosci 20: RC105.
Hille B (1994) Modulation of ion channel function by G protein-coupled receptors. Trends Neurosci 17: 531-535.
Ikeda SR and Dunlap K (1999) Voltage-dependent modulation of N-type calcium channels: role of G protein subunits. Adv Second Messenger Phosphoprotein Res 33: 131-151.
Jones S, Brown DA, Milligan G, Willer E, Buckley NJ, and Caulfield MP (1995) Bradykinin excites rat sympathetic neurons by inhibition of M current through a mechanism involving B2 receptors and Gq/11. Neuron 14: 400-405.
Just I, Selzer J, Wilm M, von Eichel-Streiber C, Mann M, and Aktories K (1995) Glucosylation of Rho proteins by Clostridium difficile toxin B. Nature (Lond) 375: 500-503.
Koh DS and Hille B (1997) Modulation by neurotransmitters of catecholamine secretion from sympathetic ganglion neurons detected by amperometry. Proc Natl Acad Sci USA 94: 1506-1511.
Lister MD, Glaser KB, Ulevitch RJ, and Dennis EA (1989) Inhibition studies on the membrane-associated phospholipase A2 in vitro and prostaglandin E2 production in vivo of the macrophage-like P388D1 cell. Effects of manoalide, 7,7-dimethyl-5,8-eicosadienoic acid and p-bromophenacyl bromide. J Biol Chem 264: 8520-8528.
Liu L, Roberts ML, and Rittenhouse AR (2004) Phospholipid metabolism is required for M1 muscarinic inhibition of N-type calcium current in sympathetic neurons. Eur Biophys J 33: 255-264.
Micheva KD, Holz RW, and Smith SJ (2001) Regulation of presynaptic phosphatidylinositol 4,5-biphosphate by neuronal activity. J Cell Biol 154: 355-368.
Nakanishi S, Catt KJ, and Balla T (1995) A wortmannin-sensitive phosphatidylinositol 4-kinase that regulates hormone-sensitive pools of inositolphospholipids. Proc Natl Acad Sci USA 92: 5317-5321.
Schaefer T, Wiedemann C, Gitler C, and Burger MM (1994) Effects of arsenicals on the secretory process in chromaffin cells. Ann NY Acad Sci 710: 356-367.
Scholze T, Moskvina E, Mayer M, Just H, Kubista H, and Boehm S (2002) Sympathoexcitation by bradykinin involves Ca2+-independent protein kinase C. J Neurosci 22: 5823-5832.
Schwartz DD and Malik KU (1993) Cyclic AMP modulates but does not mediate the inhibition of [3H]norepinephrine release by activation of alpha-2 adrenergic receptors in cultured rat ganglion cells. Neuroscience 52: 107-113.
Stevens CF (2004) Presynaptic function. Curr Opin Neurobiol 14: 341-345.
Suh BC and Hille B (2002) Recovery from muscarinic modulation of M current channels requires phosphatidylinositol 4,5-bisphosphate synthesis. Neuron 35: 507-520.
Suh BC, Horowitz LF, Hirdes W, Mackie K, and Hille B (2004) Regulation of KCNQ2/KCNQ3 current by G protein cycling: the kinetics of receptor-mediated signaling by Gq. J Gen Physiol 123: 663-683.
Taverna E, Saba E, Rowe J, Francolini M, Clementi F, and Rosa P (2004) Role of lipid microdomains in P/Q-type calcium channel (Cav2.1) clustering and function in presynaptic membranes. J Biol Chem 279: 5127-5134.
Vlahos CJ, Matter WF, Hui KY, and Brown RF (1994) A specific inhibitor of phosphatidylinositol 3-kinase, 2-(4-morpholinyl)-8-phenyl-4H-1-benzopyran-4-one (LY 294,002). J Biol Chem 269: 5241-5248.
Wiedemann C, Schafer T, and Burger MM (1996) Chromaffin granule-associated phosphatidylinositol 4-kinase activity is required for stimulated secretion. EMBO (Eur Mol Biol Organ) J 15: 2094-2101.
Wiedemann C, Schafer T, Burger MM, and Sihra TS (1998) An essential role for a small synaptic vesicle-associated phosphatidylinositol 4-kinase in neurotransmitter release. J Neurosci 18: 5594-5602.
Wilk-Blaszczak MA, Singer WD, Quill T, Miller B, Frost JA, Sternweis PC, and Belardetti F (1997) The monomeric G-proteins Rac1 and/or Cdc42 are required for the inhibition of voltage-dependent calcium current by bradykinin. J Neurosci 17: 4094-4100.
Winks JS, Hughes S, Filippov AK, Tatulian L, Abogadie FC, Brown DA, and Marsh SJ (2005) Relationship between membrane phosphatidylinositol-4,5-bisphosphate and receptor-mediated inhibition of native neuronal M channels. J Neurosci 25: 3400-3413.
Wu L, Bauer CS, Zhen XG, Xie C, and Yang J (2002) Dual regulation of voltage-gated calcium channels by PtdIns(4,5)P2. Nature (Lond) 419: 947-952.
Xu C, Watras J, and Loew LM (2003) Kinetic analysis of receptor-activated phosphoinositide turnover. J Cell Biol 161: 779-791.
Zhang H, Craciun LC, Mirshahi T, Rohacs T, Lopes CM, Jin T, and Logothetis DE (2003) PIP2 activates KCNQ channels and its hydrolysis underlies receptor-mediated inhibition of M currents. Neuron 37: 963-975.