Asynchronous Release of Glutamate Only from TRPV1+ Afferent Axons

We examined afferent synaptic transmission by stimulating the ST with suprathreshold electrical shocks and recording EPSCs at second order NTS neurons (supra,Figure 1A). Increasing shock intensity evoked large monosynaptic EPSCs of constant amplitude (Figure 1A, inset) consistent with all-or-none, unitary responses of a single ST axon (Bailey et al., 2008). Evoked synchronous EPSCs were either completely blocked by the TRPV1 agonist, capsaicin (100 nM), (TRPV1+, n = 21 neurons); or were entirely resistant (TRPV1−, n = 20 neurons) (Figure S1) (Doyle et al., 2002). Bursts of five stimuli at physiological frequencies (50 Hz) triggered rapidly depressing EPSCs. In TRPV1+ afferents this was followed by a long period of asynchronous EPSCs (Figures 1A, B). Peak asynchronous release rates were significantly increased compared to the equivalent subthreshold period from TRPV1+ (n = 21, p < 0.001), but not TRPV1− afferents (n = 20, p = 0.32) (Figure 1A–C). Synchronous and asynchronous release had identical intensity-response relationships within neurons indicating that both required activation of the same TRPV1+ afferent (Figure 1A, insets). Repeated trains of afferent activation episodically and reliably increased the rate of asynchronous events (Figure S2) in a pattern that was stable for >20 min. Bursts of asynchronous events were maximal in these neurons between 200–500 msec after stimulus onset and decayed to control over 5 sec (Figure 1B). On average (n = 11), the burst of asynchronous events decayed bi-exponentially with ~90% of the total release accounted for during the initial process (τ₁ = 0.34 ± 0.06 sec) and the remaining occurring more slowly (τ₂ = 1.56 ± 0.18 sec) (Figure 1B). In the prolonged asynchronous release period, the number of quanta released during the synchronous burst of five EPSCs was similar to the number released asynchronously (23 ± 4 quanta/trial vs. 24 ± 2 quanta/trial, respectively, n = 21, p = 0.74). Neurons receiving TRPV1− afferent innervation failed to release glutamate asynchronously following ST activation (Figure 1B, C). Previous work comparing synchronous release properties between TRPV1+ and TRPV1− afferents, including variance-mean analysis of ST-evoked EPSCs, found no significant differences (Andresen and Peters, 2008;Bailey et al., 2006;Peters et al., 2008). Thus, despite sharing similar evoked EPSC characteristics, only TRPV1+ afferents released glutamate asynchronously at NTS.

NTS neurons receiving TRPV1+ afferents had greater basal frequencies of spontaneous EPSCs with larger quantal amplitudes compared to TRPV1− afferents (Figure 1C). The decay kinetics of asynchronous EPSCs were indistinguishable from spontaneous EPSCs recorded at subthreshold shock intensities, suggesting that basal spontaneous and asynchronous events likely arise from the same synaptic contacts (Best and Regehr, 2009). Neurons with the largest evoked EPSCs had greater asynchronous responses (Figure 1D). Across neurons asynchronous responses increased in proportion to EPSC1 amplitude for TRPV1+ but not TRPV1− neurons (Figure 1D). Since the release probabilities and quantal size for afferent evoked EPSCs are similar across NTS neurons (Andresen and Peters, 2008;Bailey et al., 2006;Peters et al., 2008), this scaling relationship suggests that the number of asynchronous release sites increases in proportion with the number of active zones producing synchronous release. Asynchronous events were not observed even in TRPV1− neurons receiving the largest amplitude EPSC1 (Figure 1D). These findings suggest a fundamental difference in the regulation of asynchronous vesicle release between TRPV1+ and TRPV1− afferents.

Activity Dependent Increases in Asynchronous Release

Variability in the relative amounts of synchronous and asynchronous release at different synapses is well recognized (Best and Regehr, 2009). However, the sharp contrast in asynchronous release observed between TRPV1+ and TRPV1− afferent terminals was surprising. The dependence of asynchronous release on the degree of synchronous release in TRPV+ terminals (Figure 1D) led us to investigate the relationship between afferent activity and the asynchronous release rate. Increasing the number of shocks to 10 at a fixed frequency increased the asynchronous release in TRPV+ neurons but still failed to elicit any asynchronous release in TRPV− neurons (Figure 2A). Physiologically relevant frequencies of afferent activation produce substantial frequency dependent depression (FDD) of synchronous glutamate release that is attributed to depletion of the readily releasable pool of vesicles (RRP) (Bailey et al., 2006;Doyle and Andresen, 2001). Extending the trains at a fixed frequency (50 Hz) from 1–20 stimulus shocks induced substantial FDD depression that was near maximal at 10 shocks (synchronous EPSCs decreasing by ~80%) and was uniform across both TRPV1+ and TRPV1− afferent subtypes (Figure 2B). Remarkably, despite this substantial net depression of the synchronous release process, asynchronous release from TRPV1+ afferents increased in proportion to the number of stimuli up to a maximum at ~10 afferent shocks (Figure 2C). This activity-dependent relationship for TRPV1+ neurons was saturable and frequency independent between 25 and 100 Hz (Figure S3). Successive action potentials should increase terminal calcium levels and this process is thought to be essential for the asynchronous release of additional vesicles (Cummings et al., 1996;Hefft and Jonas, 2005;Neher and Sakaba, 2008). However, even extended trains of terminal activation failed to trigger additional asynchronous events from TRPV1− afferents (Figure 2C) indicating a fundamental difference across the two phenotypes of terminal. Such differences could reflect differential regulation of a common pool of vesicles or the presence of a unique pool in the TRPV1+ terminals.

Synchronous and Asynchronous Vesicle Pools

If synchronous and asynchronous releases arise from a single common pool of vesicles, the continued asynchronous release of glutamate from TRPV1+ afferents would be expected to slow or reduce the recovery of synchronous release from FDD. To test this hypothesis we compared recovery from FDD between TRPV1+ and TRPV1− neurons by delivering single test shocks at various time points following the initial stimulus train (Figure 3A). While synchronous EPSCs were depressed deeply at times during which asynchronous release reached its peak (Figure 3A), the recovery relationships were not different between test-evoked EPSC amplitude and poststimulus delivery time for TRPV1+ and TRPV1− afferents (p = 0.47,Figure 3B). These results suggest that vesicular mobilization into the readily releasable pool (RRP) following depletion was independent of the occurrence of asynchronous release. Thus, either evoked and asynchronous EPSCs draw from separate pools of vesicles or the absolute rate of asynchronous release is too small to impact recovery from FDD (Otsu et al., 2004).

Common Calcium Sensitivity of Synchronous and Asynchronous Release

It has been proposed that asynchronous release represents exocytosis of vesicles with a low affinity sensor for intracellular calcium whereas synchronous release reflects fusion of vesicles triggered by a higher affinity calcium sensor (Cummings et al., 1996;Sun et al., 2007). Alternatively, an allosteric form of vesicle fusion consisting of calcium-dependent and -independent steps may account for differences in synchronous and asynchronous release (Wolfel et al., 2007). In both schemes, activation of VACC generates large calcium transients localized nearest to synchronous release sites. However, following the initial transient, calcium spreads throughout the terminal activating distinct synaptotagmins with higher affinities that mobilize additional vesicles asynchronously (Sun et al., 2007). Variability in the relative amounts of synchronous and asynchronous release at different synapses may reflect variation in the distance between VACC and synaptic vesicles (Best and Regehr, 2009). In our experiments, both the afferent-evoked EPSCs and the subsequent asynchronous EPSC frequency were steeply sensitive to extracellular calcium concentration (Figure 4). Reducing external calcium concentration reduced asynchronous release of glutamate in proportion to the reduction in synchronous EPSCs (Figure 4D).

Differences in VACCs might lead to systematic differences in the intracellular calcium signal across these two phenotypes of afferent terminals (Hefft and Jonas, 2005). Evoked synchronous glutamate release at afferent terminals in NTS is dominated by the activation of N-type VACC (Mendelowitz et al., 1995). We asked if a differential contribution of N-type VACC between terminal subgroups might explain the differences in asynchronous release. Blockade of the N-type VACC, with ω-conotoxin GVIa (1 μM) similarly reduced the amplitude of the synchronous EPSC1 in TRPV1+ (n = 4) and TRPV1− (n = 4) as well as reducing the frequency of asynchronous release in TRPV1+ to <25% of control (Figure 5A–C). While necessary for both forms of release, VACC are likely to be active only transiently while asynchronous release occurs over a prolonged time course during which the N-type channels will be closed. Thus, differences in VACC are unlikely to account for the presence of asynchronous release only from TRPV1+ terminals.

At other central synapses, calcium buffers have distinguished different calcium sensitivities for synchronous and asynchronous release and been used to estimate the calcium diffusion distance within the terminal (Sun et al., 2007). In TRPV1+ NTS neurons, the low-affinity calcium chelator EGTA-AM reduced the synchronous as well as asynchronous release in similar proportion (Figure 5D–F). Cell-to-cell variability in the effectiveness of EGTA-AM was considerable (n = 5,Figure 5F). The higher affinity buffer, BAPTA-AM, produced similar reductions (Figure 5D–F, n = 4). The modest attenuation of glutamate release by these calcium buffers suggests an intrinsically high calcium sensitivity of the glutamate release process and a relatively short diffusion length compared with other synapses (Cummings et al., 1996). Overall, the co-attenuation of synchronous and asynchronous vesicular release suggests a common dependence on ST evoked rises in terminal calcium. This common path of calcium dependent activation leads to release of synchronous and asynchronous pools of vesicles within the TRPV1+ terminals. Thus, action potential invasion and calcium entry through N-type VACC are required for activity-dependent augmentation of asynchronous neurotransmission. Since the evoked EPSC release process is indistinguishable between TRPV1+ and TRPV1− afferents, then an additional signaling mechanism must be present to account for asynchronous release only from TRPV1+ terminals.

TRPV1 Activity Required for Asynchronous Release

Capsaicin sensitivity of ST evoked EPSCs reliably predicted the presence of asynchronous glutamate release in NTS neurons. TRPV1 delivers sufficient calcium to these terminals to trigger glutamate release following blockade of VACC and sodium channels (Jin et al., 2004). To determine if TRPV1 itself participates in triggering activity-dependent asynchronous glutamate release; we tested the cinnamide TRPV1 receptor antagonist SB366791 (Gunthorpe et al., 2004). SB366791 selectively and competitively binds to the intracellular activation domain of TRPV1 responsible for sensitivity to capsaicin, heat, and endocannabinoids (Gunthorpe et al., 2004). SB366791 (10 μM) rapidly and reversibly decreased the frequency of asynchronous events (42 ± 5% of control, n = 5) without altering the amplitude of the evoked synchronous EPSCs in TRPV1+ neurons (Figure 6). The threshold concentration for SB366791 was 1 μM for a significant reduction in asynchronous release and evoked EPSC amplitude was unaltered (Figure 6C). Similar results were observed in TRPV1+ neurons in which the pyridinylpiperazine TRPV1 antagonist JNJ 17203212 (Swanson et al., 2005) selectively attenuated asynchronous release (49 ± 9% of control, n = 3, p = 0.02) but not the amplitude of the evoked EPSCs (93 ± 2% of control, n = 3, p = 0.15) (data not shown). TRPV1 antagonism depressed the relationship between EPSC1 amplitude and the integrated asynchronous response (Figure 6D) to a low slope relationship that resembled that in TRPV1− neurons (Figure 1D).

Asynchronous release clearly increased with evoked synchronous activity (Figures 1–2,S2–3). However if TRPV1 is active under basal conditions, then the tight coupling between TRPV1 and asynchronous release leads to the prediction that some of the release seen in the presence of tetrodotoxin (TTX) will be sensitive to TRPV1 antagonism. We tested this hypothesis in a separate series of experiments, by examining the rate of miniature EPSCs (TTX 1 μM) in TRPV1+ neurons following application of SB366791. SB366791 (10 μM) decreased mEPSC frequency to 59 ± 4% of control (n = 4, p = 0.03) which suggests that the basal release of asynchronous vesicles is regulated by TRPV1 even in the absence of afferent activity (Figure S4).

Thermosensitivity of Asynchronous Release

TRPV1 is activated in peripheral neurons by temperatures >40°C. We next asked whether temperature might modify the asynchronous release process. In neurons receiving TRPV1+ afferent contacts, elevating temperature from 25°C to 33°C more than tripled the rate of asynchronous release (356 ± 78%, n = 14, p < 0.01) while increasing the amplitude of synchronous EPSCs by about one third (138 ± 5% of control, n = 14, p = 0.03) (Figure 7). Increasing temperature to 38°C augmented asynchronous release further (410 ± 60% of control, n = 14, p < 0.001) without significantly altering the evoked EPSC1 amplitude. Within the more physiological range (33–38°C), evoked EPSCs changed little in latency, jitter or waveform. The asynchronous kinetic profile to a train of 5 shocks (50 Hz) became more accentuated in warming from 33 and 38°C (Figure 7C) despite little change in the synchronous volley. In contrast, at 25°C, the asynchronous profile was profoundly depressed. In a subgroup of neurons (n = 5), addition of SB366791 (10 μM) decreased the temperature-induced increase in asynchronous release but did not alter synchronous release (Figure 7D) indicating that this effect was mediated by TRPV1. Across neurons, the relationship between EPSC1 amplitude and the integrated asynchronous response had a depressed slope at 25°C (Figure 7E). Thus, physiological temperatures support a tonic drive to spontaneously release glutamate and facilitate the activity-dependent increase in asynchronous release of glutamate from these central terminals.

Asynchronous Release Prolongs Synaptic Activity

The physiological role of asynchronous release remains uncertain at most central synapses because it has been difficult to clearly separate synchronous and asynchronous pathways. NTS contains afferents that both possess and lack asynchronous release allowing for a direct comparison of these processes. In current-clamp recordings of TRPV1+ neurons, suprathreshold ST stimulation (5 shocks at 50 Hz) produced an initial burst of tightly synchronized action potentials that were followed by a prolonged period of asynchronous excitatory postsynaptic potentials (EPSPs) which often triggered action potentials (Figure 8A). Across 50 trials, an average of 217 ± 72 (n = 5 neurons) more asynchronous action potentials were generated compared to equivalent trials with subthreshold ST shocks. In contrast, no additional action-potentials occurred in TRPV1− neurons (n = 4) (Figure 8A). Because SB366791 attenuates the activity evoked asynchronous glutamate release from TRPV1+ afferents, we tested whether such additional release could be selectively removed and how that would impacted postsynaptic activity. SB366791 (10 μM) did not alter the number of synchronous action potentials across 50 trials (control: 91 ± 17, SB366791: 82 ± 18, p = 0.37) but almost eliminated asynchronous action potentials and EPSPs (Figure 8A, right). These results are consistent with a novel form of synaptic plasticity, called TRPV1-mediated facilitation (TMF), which substantially expands the postsynaptic excitatory period.

Together our findings suggest that two afferent phenotypes in NTS display common mechanisms regulating synchronous release of glutamate but that the TRPV1+ terminals possess an additional, independently regulated pool of vesicles for asynchronous release that is steeply graded by temperatures in the physiological/pathophysiological range. Activity dependent release from the asynchronous pool strongly depended on TRPV1 receptor activation and caused prolongation of spiking at postsynaptic NTS neurons. This TMF provides a new mode of synaptic potentiation at certain ST-NTS synapses (Figure 8C).

Synchronous and Asynchronous Release Arise from the Same Afferent Terminal

Graded stimulus recruitment protocols detected precise thresholds for activation of single afferents contacting the recorded cell (Bailey et al., 2006;McDougall et al., 2009). Although recurrent excitatory pathways are common in NTS (Fortin and Champagnat, 1993;Smith et al., 1998), such inputs are triggered with unique additional thresholds (McDougall et al., 2009). Asynchronous release only occurred in response to successful afferent activation and had the same threshold as synchronous release making it highly likely that it arose from the same terminals. Basal rates of spontaneous EPSCs were greater at TRPV1+ neurons than at TRPV1− neurons. Even in TTX, TRPV1 antagonism reduced this mEPSC rate – observations that are consistent with tonic TRPV1 activity promoting release from the vesicle pool even in the absence of afferent activation. Likewise, lowering temperature or TRPV1 selective antagonists attenuated the augmentation of asynchronous glutamate release following ST activation demonstrating an active role of TRPV1 in translating acute afferent activity into long lasting glutamate release. Since spontaneous events from the pre- and post-stimulus periods have identical decay kinetics, they likely arise from similar if not the same afferent contacts (Best and Regehr, 2009). Evoked ST EPSC amplitude varies widely across both TRPV1+ and TRPV1− inputs and primarily reflects different numbers of active release sites for individual afferents (Bailey et al., 2006;Peters et al., 2008). Interestingly, the magnitude of the asynchronous response increased in direct proportion to the amplitude of the evoked EPSC within TRPV1+ neurons. We propose this scaling of asynchronous release may reflect the variable arborizations and number of contacts for individual afferents and the resulting number of asynchronous release sites. In contrast, even the largest TRPV1− contacts failed to release glutamate asynchronously despite similar release site numbers and afferent architecture compared to TRPV+ terminals. Our findings are consistent with synchronous and asynchronous vesicles arising from the same afferent release sites.

Synchronous and Asynchronous Vesicle Pools

The time course of asynchronous release at ST TRPV1+ synapses was an order of magnitude longer than most central synapses (Atluri and Regehr, 1998;Chen and Regehr, 1999;Goda and Stevens, 1994). As a result, the total number of asynchronous quanta was nearly equal to that released synchronously by a volley of five shocks. Although precise action potential timing is lost during asynchronous release, the continued excitatory drive becomes relatively more influential since synchronous release is largely depressed. FDD is similar between TRPV1+ and TRPV1− afferents consistent with similar synchronous release mechanisms and vesicle shuttling (Silver et al., 1998;Smith et al., 2008). Recovery from FDD was nearly identical between TRPV1+ and TRPV1− afferents. Surprisingly, the number of shocks and not the frequency of stimulation determined the amount of asynchronous release; although the process was equally saturable at high shock counts despite a fourfold difference in stimulation frequency (Figure S3A, B). In contrast, FDD increased with stimulation frequency (Figure S3C) while asynchronous release was maximal indicating the vesicle pools involved may be independent. However, the average peak rate of asynchronous release (6.3 ± 5.8 quanta per 100 msec) was low compared to the rate of synchronous released sustained by a typical ST-NTS neuron (18 quanta per 100 msec; assuming 20 release sites, release probability 90% (Bailey et al., 2006), and 80% FDD at 50 Hz). Consequently even if both fast and slow release were from the same vesicle pool, asynchronous release may have had little impact on recovery from FDD. Thus, a common pool of vesicles may be so large that neither form of release is significant enough under the present experimental conditions to interfere with vesicle distribution. In TRPV1+ afferents, capsaicin exposure dramatically increases asynchronous vesicle release followed over time by the loss of synchronous EPSCs (Figure S1C) (Doyle et al., 2002). Following disappearance of ST-evoked EPSCs, no facilitation of asynchronous release with ST shocks can be discerned in the face of sustained high levels of spontaneous EPSCs, but the difference in time course makes it less likely that depletion of a common vesicle pool leads to severe attenuation of synchronous release. Likewise, increasing temperatures from 33°C to 38°C augmented the asynchronous release without decreasing the amplitude of synchronous EPSCs even under near maximal release conditions (Bailey et al., 2006;Peters et al., 2008). Further, the temperature dependent increase of asynchronous release was selectively attenuated by TRPV1 antagonism with no effect on synchronous release, again suggesting independent vesicle pools. In either case the subcellular mechanism by which TRPV1 targets vesicles for asynchronous release separately from synchronous release remains unknown.

Calcium and the Role of TRPV1

In TRPV1+ afferents, lowering bath calcium decreased the amplitude of EPSC1 and reduced the magnitude and duration of the asynchronous component. The relative sensitivities to extracellular calcium were similar; although the asynchronous component was consistently reduced somewhat more than the synchronous. Reliable synchronous release of neurotransmitter depends on the tight coupling of VACC and the subcellular release machinery (Matthews, 1996). Calcium sensitive synaptotagmins coordinate terminal depolarization and subsequent calcium influx with the concerted release of vesicles from the active zone (Chapman, 2008). During prolonged stimulation trains, coordination between local calcium influx and vesicles could be disrupted (Atluri and Regehr, 1998;Lu and Trussell, 2000). The identity of the VACC associated with particular release sites may determine the fidelity of this pairing and subsequent vesicle release. Terminals possessing N-type channels exhibit proportionately more asynchronous release than terminals in which exocytosis depended on P/Q-type VACCs (Hefft and Jonas, 2005). Previous work at the ST-NTS syna pse demonstrated synchronous release was largely dependent on the N-type VACC (Cav2.2) although determination of afferent subtypes was not performed (Mendelowitz et al., 1995). We found that both TRPV1+ and TRPV1− afferents relied on N-type channels for 80% of their synchronous release. Although differences in VACC expression did not explain the presence or absence of asynchronous release at ST-NTS synapses; voltage dependent calcium influx was common and requisite for both release pathways.

At synapses with relatively brief asynchronous components, diffusion of presynaptic calcium transients acting at high affinity synaptotagmins may account for the continued release (Hui et al., 2005;Sudhof, 2002). This process is limited by diffusion rate constants and calcium buffering capacity (Meinrenken et al., 2002) which are not known to differ between ST afferent subtypes. Bath application of the membrane-permeant calcium chelators EGTA-AM and BAPTA-AM attenuated both synchr onous and asynchronous components indicating similar average distance between the calcium source and sensors for the two vesicle pools (Meinrenken et al., 2002). Nevertheless, even following long bursts of afferent activation, when terminal calcium would be greatest, TRPV1− afferents failed to release vesicles asynchronously.

TRPV1+ was predictive of asynchronous release and TRPV1 itself may provide additional calcium drive responsible for the prolonged vesicle mobilization. The TRPV1 antagonists, SB366791 and JNJ 17203212, as well as lower temperature reduced the asynchronous component triggered by afferent stimuli without altering synchronous events. SB366791 decreased baseline spontaneous event rates and mEPSCs in TTX in TRPV1+ neurons as well as their sensitivity to temperature changes. Considering SB366791 competitively binds to the intracellular activation domain of TRPV1 (Gunthorpe et al., 2004) an endogenous ligand may be generated during synchronous release which subsequently activates TRPV1. Sustained activation of TRPV1 would prolong calcium influx and explain the time course of asynchronous release. The question as to whether this could arise due to slow rates of TRPV1 deactivation or requires a prolonged TRPV1 agonist has already been addressed. Rapid temperature jumps in a heterologous expression system suggest that TRPV1 deactivation occurs rapidly (<10 ms) following cessation of an infrared pulse (Yao et al., 2009) – too rapidly for TRPV1 channel deactivation kinetics to delay release by the observed amount.

Other mechanisms, up- or downstream of TRPV1 activation, could enhance presynaptic calcium and thereby trigger asynchronous release. These include: calcium entry via TRPV1 and presynaptic accumulation following repeated spikes; action potential broadening that increases calcium entry through TRPV1 with repeated activity (Li et al., 2007); voltage- and/or pH-dependence of presynaptic voltage-dependent channels; increased calcium entry via VACC due to altered action potential shape resulting from TRPV1 recruitment during a train of action potentials; increased calcium entry via TRPV1 due to activity dependent release of a TRPV1 agonist; slow rates of exocytosis for an asynchronous vesicle pool; and altered calcium buffering in the TRPV1+ terminals. While our studies show that TRPV1 is an important determinant of asynchronous release, N-type channels contribute importantly to translating afferent activity into asynchronous transmission. We do not yet fully understand how TMF results from the coupling between activity, global calcium and TRPV1 but the temperature-sensitivity indicates thermal activation of asynchronous release is substantial under normal basal conditions.

The specificity with which calcium entry via TRPV1 selectively mobilizes asynchronous vesicles without interfering with synchronous release is intriguing. One possibility is that TRPV1 may be physically remote from synchronous release mechanisms; consistent with a secondary release process targeting a separate pool of vesicles for asynchronous release. However, exogenous activation of TRPV1 with capsaicin dramatically increased spontaneous rates and eliminated evoked EPSCs from TRPV1+ afferents consistent with a common vesicle pool (Doyle et al., 2002). Another possibility is that the two forms of release occur because of two separate intracellular calcium signals that trigger fast and slow forms of exocytosis. In the CA3 region of the hippocampus, calcium entry through presynaptic nicotinic receptors recruited intracellular calcium stores and precipitated prolonged, action-potential independent release of glutamate and postsynaptic spiking (Sharma and Vijayaraghavan, 2003). While store calcium could be involved in TMF, the absence of any difference of action of EGTA and BAPTA argues against this. Further experiments are required to test the hypotheses that both types of release arise from an allosteric form of vesicle fusion (Lou et al., 2005) or from more than one intracellular calcium sensor or source (Sun et al., 2007). Overall, the evidence at ST-NTS TRPV1+ synapses favors separate vesicle pools. It is unclear if separate asynchronous and synchronous pools are similar to the spontaneous and evoked pools identified in hippocampal neurons (Sara et al., 2005). Our studies also suggest that tonic TRPV1 activity at physiological temperatures is a critical determinant of activity independent spontaneous glutamate release consistent with separate vesicle pools for synchronous and asynchronous processes. In light of the clear role of TRPV1 as a trigger for asynchronous release at NTS neurons and the distinct heterogeneity in the amount of asynchronous release between TRPV1+ and TRPV1− synapses, we speculate that TRPV1 or other non-selective cation channels may contribute to asynchronous release at other central synapses (Best and Regehr, 2009).

Functional Consequences of Asynchronous Release

TMF importantly expands the signaling repertoire and transfer of information of TRPV1+ ST-NTS synapses. TMF doubles the amount of glutamate released for a given action potential volley which translates into continued postsynaptic spiking for many seconds following the stimulation period. This substantial potentiation will shape NTS output to target nuclei throughout the brain. TPRV1 dependent asynchronous release is strongly temperature dependent and augments activity by ~10%/°C in the CNS temperature range. This result contradicts the common extrapolation of peripheral TRPV1 threshold temperature (>40°C) to CNS TRPV1 (Julius and Basbaum, 2001). Afferent activation of these pathways through NTS is critically important during cardiovascular and respiratory crises in which C-fiber afferents are recruited (Coleridge and Coleridge, 1984). For example, cardiovascular effects from modest activation of unmyelinated baroreceptors required 20-fold higher activation frequencies in myelinated baroreceptors (Fan et al., 1999;Fan and Andresen, 1998). TMF represents a novel signaling process that supports heightened reflexes from C-fiber activation. The magnitude and prolonged time period of asynchronous release dramatically potentiates the relative synaptic strength. This additional glutamate release may be upregulated under different physiological or pathophysiological states (Kline et al., 2005). Processes which target TRPV1 activity, either endogenous or exogenous, will have profound effects on information transfer across the synapse. Considering the broad distribution of TRPV1, this additional mechanism of glutamate release should be considered at other central synapses (Kauer and Gibson, 2009). In NTS, the strict phenotypic association of asynchronous release with TRPV1+ afferents identifies a previously unrecognized control of neurotransmitter release at this synapse and provides a novel signaling role for thermosensitive TRPV1 receptors at central terminals that integrates neuronal activity.

Reagents

Capsaicin, ω-conotoxin GVIa, SB366791 and JNJ 17203212 were purchased from Tocris (Ellisville, Missouri) and EGTA-AM and BAPTA-AM from Invitrogen (Carlsbad, CA).

Horizontal Brainstem Slice Preparation

Brain stem slices (250 μm) were prepared from adult male Sprague Dawley rats (>160 g) under isoflurane anesthesia as previously described (Doyle and Andresen, 2001;Peters et al., 2008). Briefly, the medulla was removed and placed in cooled aCSF containing (mM): 125 NaCl, 3 KCl, 1.2 KH₂PO₄, 1.2 MgSO₄, 25 NaHCO₃, 10 dextrose, and 2 CaCl₂, bubbled with 95% O₂–5% CO₂. The tissue was trimmed rostrally and caudally to yield a tissue block centered on obex. To orient the ST axons with the NTS in a common plane for cutting, the ventral surface of the brain stem block was cut. Slices were cut with a sapphire knife (Delaware Diamond Knives, Wilmington, DE) mounted in a vibrating microtome (VT1000S; Leica Microsystems Inc., Bannockburn, IL). Slices were secured with a fine polyethylene mesh and perfused with aCSF constantly bubbled with 95% O₂ 5% CO₂ at 34–35°C and 300 mOsm. All animal procedures were performed with the approval of the Institutional Animal Care and Use Committee at Oregon Health & Science University and conform to the guidelines of the National Institutes of Health publication “Guide for the Care and Use of Laboratory Animals”.

Whole-cell recordings

Patch electrodes were visually guided to neurons using infrared illumination and differential interference contrast optics (DIC) (40X water immersion lens) on an Axioskop 2 microscope (Zeiss, Thornwood, NJ) with digital camera (Hamamatsu Photonic Systems, Bridgewater, NJ). Recording electrodes (2.1 – 3.2MΩ) were filled with a low Cl⁻ (10mM, E_Cl = −69 mV), intracellular solution which contained (mM): 6 NaCl, 4 NaOH, 130 K-gluconate, 11 EGTA, 1 CaCl₂, 1 MgCl₂, 10 HEPES, 2 Na₂ATP, and 0.2 Na₂GTP. The intracellular solution was pH 7.3 and 296 mOsm. For most experiments, neurons were studied under voltage clamp conditions with a MultiClamp 700B amplifier (Axon Instruments, Foster City, CA) and held at V_H = −60 mV using pipettes in open, whole cell patch configuration. Liquid junction potentials were not corrected. Signals were filtered at 10 kHz and sampled at 30 kHz using p-Clamp software (version 8.2, Axon Instruments). Neurons were recorded at 32°C for all experiments. When necessary changes in bath temperature were accomplished using an inline pre-heater (Cell MicroControls, Norfolk, VA).

Identification of second- order neurons

A concentric bipolar stimulating electrode (200 μm outer tip diameter; Frederick Haer Co., Bowdoinham, ME) was placed on distal portions of the visible ST rostral (>1 mm) to the recording region. Current shocks were delivered to the ST every 6 sec (shock duration 100 μsec) using a Master-8 isolated programmable stimulator (A.M.P.I., Jerusalem, Israel). Latency was measured between the ST shock artifact and the onset of the resulting EPSC. Synaptic jitter = standard deviation of 30–40 ST-EPSC latencies within each neuron. Jitters of <200 μsec identify monosynaptic afferent contacts. At the end of each recording, afferent evoked EPSCs were tested with capsaicin (100 nM) to determine vanilloid sensitivity - TRPV1+ or TRPV1−. The location of each neuron studied was noted and plotted on a schematic representation of the NTS slice (adapted from Paxinos and Watson) (Figure 8). Interestingly, neurons receiving TRPV1+ contacts were dispersed throughout the medial NTS whereas those with TRPV1− contacts tended to align immediately adjacent to the ST.

Statistical Analysis

Digitized waveforms were analyzed using an event detection and analysis program (MiniAnalysis, Synaptosoft, Decatur, GA) for all quantal synaptic currents and Clampfit 10 (Axon Instruments, Foster City, CA) for all ST evoked EPSCs. To accurately measure the amplitudes and kinetics of quantal events, EPSCs smaller than 10 pA were excluded. To determine the number of additional events released asynchronously the frequencies of quantal events were collected in 100 msec bins and summed across 50 stimulation trials. The average rate of baseline events measured for 1 sec before each shock train commenced was subtracted from the rate calculated for each 100 msec bin during the period of 4 sec following the last ST shock to yield a net asynchronous event rate. The resulting asynchronous profile was integrated to represent the total additional release across trials. Across trials the average frequency of events during the first 1000 msec following the last ST stimulus shock was a measure of the peak asynchronous frequency. For statistical comparisons data was tested for Normal distributions with and the appropriate parametric or non-parametric statistics were used, including ANOVA, Fisher s PLSD post hoc analysis.

1. Andresen MC, Doyle MW, Bailey TW, Jin Y-H. Differentiation of autonomic reflex control begins with cellular mechanisms at the first synapse within the nucleus tractus solitarius. Braz J Med Biol Res. 2004;37:549–558. doi: 10.1590/s0100-879x2004000400012. [DOI] [PubMed] [Google Scholar]
2. Andresen MC, Peters JH. Comparison of baroreceptive to other afferent synaptic transmission to the solitary tract nucleus. Am J Physiol Heart Circ Physiol. 2008;295:H2032–H2042. doi: 10.1152/ajpheart.00568.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
3. Atluri PP, Regehr WG. Delayed release of neurotransmitter from cerebellar granule cells. J Neurosci. 1998;18:8214–8227. doi: 10.1523/JNEUROSCI.18-20-08214.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
4. Bailey TW, Appleyard SM, Jin YH, Andresen MC. Organization and properties of GABAergic neurons in solitary tract nucleus (NTS) J Neurophysiol. 2008;99:1712–1722. doi: 10.1152/jn.00038.2008. [DOI] [PubMed] [Google Scholar]
5. Bailey TW, Jin Y-H, Doyle MW, Smith SM, Andresen MC. Vasopressin inhibits glutamate release via two distinct modes in the brainstem. J Neurosci. 2006;26:6131–6142. doi: 10.1523/JNEUROSCI.5176-05.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
6. Best AR, Regehr WG. Inhibitory regulation of electrically coupled neurons in the inferior olive is mediated by asynchronous release of GABA. Neuron. 2009;62:555–565. doi: 10.1016/j.neuron.2009.04.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
7. Chapman ER. How does synaptotagmin trigger neurotransmitter release? Annu Rev Biochem. 2008;77:615–641. doi: 10.1146/annurev.biochem.77.062005.101135. [DOI] [PubMed] [Google Scholar]
8. Chen C, Regehr WG. Contributions of residual calcium to fast synaptic transmission. J Neurosci. 1999;19:6257–6266. doi: 10.1523/JNEUROSCI.19-15-06257.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
9. Coleridge JCG, Coleridge HM. Afferent vagal C fibre innervation of the lungs and airways and its functional significance. Rev Physiol Biochem Pharmacol. 1984;99:2–110. doi: 10.1007/BFb0027715. [DOI] [PubMed] [Google Scholar]
10. Cummings DD, Wilcox KS, Dichter MA. Calcium-dependent paired-pulse facilitation of miniature EPSC frequency accompanies depression of EPSCs at hippocampal synapses in culture. J Neurosci. 1996;16:5312–5323. doi: 10.1523/JNEUROSCI.16-17-05312.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
11. Doyle MW, Andresen MC. Reliability of monosynaptic transmission in brain stem neurons in vitro. J Neurophysiol. 2001;85:2213–2223. doi: 10.1152/jn.2001.85.5.2213. [DOI] [PubMed] [Google Scholar]
12. Doyle MW, Bailey TW, Jin Y-H, Andresen MC. Vanilloid receptors presynaptically modulate visceral afferent synaptic transmission in nucleus tractus solitarius. J Neurosci. 2002;22:8222–8229. doi: 10.1523/JNEUROSCI.22-18-08222.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
13. Fan W, Andresen MC. Differential frequency-dependent reflex integration of myelinated and nonmyelinated rat aortic baroreceptors. Am J Physiol Heart Circ Physiol. 1998;275:H632–H640. doi: 10.1152/ajpheart.1998.275.2.H632. [DOI] [PubMed] [Google Scholar]
14. Fan W, Schild JH, Andresen MC. Graded and dynamic reflex summation of myelinated and unmyelinated rat aortic baroreceptors. Am J Physiol Regul Integr Comp Physiol. 1999;277:R748–R756. doi: 10.1152/ajpregu.1999.277.3.R748. [DOI] [PubMed] [Google Scholar]
15. Fortin G, Champagnat J. Spontaneous synaptic activities in rat nucleus tractus solitarius neurons in vitro: Evidence for re-excitatory processing. Brain Res. 1993;630:125–135. doi: 10.1016/0006-8993(93)90650-c. [DOI] [PubMed] [Google Scholar]
16. Gibson HE, Edwards JG, Page RS, Van Hook MJ, Kauer JA. TRPV1 channels mediate long-term depression at synapses on hippocampal interneurons. Neuron. 2008;57:746–759. doi: 10.1016/j.neuron.2007.12.027. [DOI] [PMC free article] [PubMed] [Google Scholar]
17. Goda Y, Stevens CF. Two components of transmitter release at a central synapse. Proc Natl Acad Sci U S A. 1994;91:12942–12946. doi: 10.1073/pnas.91.26.12942. [DOI] [PMC free article] [PubMed] [Google Scholar]
18. Gunthorpe MJ, Rami HK, Jerman JC, Smart D, Gill CH, Soffin EM, Luis HS, Lappin SC, Egerton J, Smith GD, Worby A, Howett L, Owen D, Nasir S, Davies CH, Thompson M, Wyman PA, Randall AD, Davis JB. Identification and characterisation of SB-366791, a potent and selective vanilloid receptor (VR1/TRPV1) antagonist. Neuropharmacology. 2004;46:133–149. doi: 10.1016/s0028-3908(03)00305-8. [DOI] [PubMed] [Google Scholar]
19. Hefft S, Jonas P. Asynchronous GABA release generates long-lasting inhibition at a hippocampal interneuron-principal neuron synapse. Nat Neurosci. 2005;8:1319–1328. doi: 10.1038/nn1542. [DOI] [PubMed] [Google Scholar]
20. Hui E, Bai J, Wang P, Sugimori M, Llinas RR, Chapman ER. Three distinct kinetic groupings of the synaptotagmin family: candidate sensors for rapid and delayed exocytosis. Proc Natl Acad Sci U S A. 2005;102:5210–5214. doi: 10.1073/pnas.0500941102. [DOI] [PMC free article] [PubMed] [Google Scholar]
21. Jin Y-H, Bailey TW, Li BY, Schild JH, Andresen MC. Purinergic and vanilloid receptor activation releases glutamate from separate cranial afferent terminals. J Neurosci. 2004;24:4709–4717. doi: 10.1523/JNEUROSCI.0753-04.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
22. Julius D, Basbaum AI. Molecular mechanisms of nociception. Nature. 2001;413:203–210. doi: 10.1038/35093019. [DOI] [PubMed] [Google Scholar]
23. Kauer JA, Gibson HE. Hot flash: TRPV channels in the brain. Trends Neurosci. 2009;32:215–224. doi: 10.1016/j.tins.2008.12.006. [DOI] [PubMed] [Google Scholar]
24. Kerr AM, Reisinger E, Jonas P. Differential dependence of phasic transmitter release on synaptotagmin 1 at GABAergic and glutamatergic hippocampal synapses. Proc Natl Acad Sci U S A. 2008;105:15581–15586. doi: 10.1073/pnas.0800621105. [DOI] [PMC free article] [PubMed] [Google Scholar]
25. Kline DD, Buniel MC, Glazebrook P, Peng YJ, Ramirez-Navarro A, Prabhakar NR, Kunze DL. Kv1.1 deletion augments the afferent hypoxic chemosensory pathway and respiration. J Neurosci. 2005;25:3389–3399. doi: 10.1523/JNEUROSCI.4556-04.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
26. Kline DD, Hendricks G, Hermann G, Rogers RC, Kunze DL. Dopamine inhibits N-type channels in visceral afferents to reduce synaptic transmitter release under normoxic and chronic intermittent hypoxic conditions. J Neurophysiol. 2009;101:2270–2278. doi: 10.1152/jn.91304.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
27. Lawson SN. Phenotype and function of somatic primary afferent nociceptive neurones with C-, Adelta- or Aalpha/beta-fibres. Exp Physiol. 2002;87:239–244. [PubMed] [Google Scholar]
28. Li BY, Feng B, Tsu HY, Schild JH. Unmyelinated visceral afferents exhibit frequency dependent action potential broadening while myelinated visceral afferents do not. Neurosci Lett. 2007;421:62–66. doi: 10.1016/j.neulet.2007.05.036. [DOI] [PubMed] [Google Scholar]
29. Loewy AD. Central autonomic pathways. In: Loewy AD, Spyer KM, editors. Central regulation of autonomic functions. New York: Oxford; 1990. pp. 88–103. [Google Scholar]
30. Lou X, Scheuss V, Schneggenburger R. Allosteric modulation of the presynaptic Ca2+ sensor for vesicle fusion. Nature. 2005;435:497–501. doi: 10.1038/nature03568. [DOI] [PubMed] [Google Scholar]
31. Lu T, Trussell LO. Inhibitory transmission mediated by asynchronous transmitter release. Neuron. 2000;26:683–694. doi: 10.1016/s0896-6273(00)81204-0. [DOI] [PubMed] [Google Scholar]
32. Lumpkin EA, Caterina MJ. Mechanisms of sensory transduction in the skin. Nature. 2007;445:858–865. doi: 10.1038/nature05662. [DOI] [PubMed] [Google Scholar]
33. Matthews G. Neurotransmitter release. Annu Rev Neurosci. 1996;19:219–233. doi: 10.1146/annurev.ne.19.030196.001251. [DOI] [PubMed] [Google Scholar]
34. McDougall SJ, Peters JH, Andresen MC. Convergence of cranial visceral afferents within the solitary tract nucleus. J Neurosci. 2009;29:12886–12895. doi: 10.1523/JNEUROSCI.3491-09.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
35. Meinrenken CJ, Borst JG, Sakmann B. Calcium secretion coupling at calyx of held governed by nonuniform channel-vesicle topography. J Neurosci. 2002;22:1648–1667. doi: 10.1523/JNEUROSCI.22-05-01648.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
36. Mendelowitz D, Yang M, Reynolds PJ, Andresen MC. Heterogeneous functional expression of calcium channels at sensory and synaptic regions in nodose neurons. J Neurophysiol. 1995;73:872–875. doi: 10.1152/jn.1995.73.2.872. [DOI] [PubMed] [Google Scholar]
37. Neher E, Sakaba T. Multiple roles of calcium ions in the regulation of neurotransmitter release. Neuron. 2008;59:861–872. doi: 10.1016/j.neuron.2008.08.019. [DOI] [PubMed] [Google Scholar]
38. Otsu Y, Shahrezaei V, Li B, Raymond LA, Delaney KR, Murphy TH. Competition between phasic and asynchronous release for recovered synaptic vesicles at developing hippocampal autaptic synapses. J Neurosci. 2004;24:420–433. doi: 10.1523/JNEUROSCI.4452-03.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
39. Patapoutian A, Tate S, Woolf CJ. Transient receptor potential channels: targeting pain at the source. Nat Rev Drug Discov. 2009;8:55–68. doi: 10.1038/nrd2757. [DOI] [PMC free article] [PubMed] [Google Scholar]
40. Peters JH, McDougall SJ, Kellett DO, Jordan D, Llewellyn-Smith IJ, Andresen MC. Oxytocin enhances cranial visceral afferent synaptic transmission to the solitary tract nucleus. J Neurosci. 2008;28:11731–11740. doi: 10.1523/JNEUROSCI.3419-08.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
41. Saper CB. The central autonomic nervous system: Conscious visceral perception and autonomic pattern generation. Annu Rev Neurosci. 2002;25:433–469. doi: 10.1146/annurev.neuro.25.032502.111311. [DOI] [PubMed] [Google Scholar]
42. Sara Y, Virmani T, Deak F, Liu X, Kavalali ET. An isolated pool of vesicles recycles at rest and drives spontaneous neurotransmission. Neuron. 2005;45:563–573. doi: 10.1016/j.neuron.2004.12.056. [DOI] [PubMed] [Google Scholar]
43. Sharma G, Vijayaraghavan S. Modulation of presynaptic store calcium induces release of glutamate and postsynaptic firing. Neuron. 2003;38:929–939. doi: 10.1016/s0896-6273(03)00322-2. [DOI] [PubMed] [Google Scholar]
44. Silver RA, Momiyama A, Cull-Candy SG. Locus of frequency-dependent depression identified with multiple-probability fluctuation analysis at rat climbing fibre Purkinje cell synapses. J Physiol. 1998;510:881–902. doi: 10.1111/j.1469-7793.1998.881bj.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
45. Smith BN, Dou P, Barber WD, Dudek FE. Vagally evoked synaptic currents in the immature rat nucleus tractus solitarii in an intact in vitro preparation. J Physiol. 1998;512:149–162. doi: 10.1111/j.1469-7793.1998.149bf.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
46. Smith SM, Renden R, von GH. Synaptic vesicle endocytosis: fast and slow modes of membrane retrieval. Trends Neurosci. 2008;31:559–568. doi: 10.1016/j.tins.2008.08.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
47. Sudhof TC. Synaptotagmins: why so many? J Biol Chem. 2002;277:7629–7632. doi: 10.1074/jbc.R100052200. [DOI] [PubMed] [Google Scholar]
48. Sun J, Pang ZP, Qin D, Fahim AT, Adachi R, Sudhof TC. A dual-Ca2+-sensor model for neurotransmitter release in a central synapse. Nature. 2007;450:676–682. doi: 10.1038/nature06308. [DOI] [PMC free article] [PubMed] [Google Scholar]
49. Swanson DM, Dubin AE, Shah C, Nasser N, Chang L, Dax SL, Jetter M, Breitenbucher JG, Liu C, Mazur C, Lord B, Gonzales L, Hoey K, Rizzolio M, Bogenstaetter M, Codd EE, Lee DH, Zhang SP, Chaplan SR, Carruthers NI. Identification and biological evaluation of 4-(3-trifluoromethylpyridin-2-yl)piperazine-1-carboxylic acid (5-trifluoromethylpyridin-2-yl)amide, a high affinity TRPV1 (VR1) vanilloid receptor antagonist. J Med Chem. 2005;48:1857–1872. doi: 10.1021/jm0495071. [DOI] [PubMed] [Google Scholar]
50. Wolfel M, Lou X, Schneggenburger R. A mechanism intrinsic to the vesicle fusion machinery determines fast and slow transmitter release at a large CNS synapse. J Neurosci. 2007;27:3198–3210. doi: 10.1523/JNEUROSCI.4471-06.2007. [DOI] [PMC free article] [PubMed] [Google Scholar]
51. Yao J, Liu B, Qin F. Rapid temperature jump by infrared diode laser irradiation for patch-clamp studies. Biophys J. 2009;96:3611–3619. doi: 10.1016/j.bpj.2009.02.016. [DOI] [PMC free article] [PubMed] [Google Scholar]

Supplementary Materials