Research ArticleNeuroscience

Ligand- and voltage-gated Ca2+ channels differentially regulate the mode of vesicular neuropeptide release in mammalian sensory neurons

See allHide authors and affiliations

Sci. Signal.  20 Jun 2017:
Vol. 10, Issue 484, eaal1683
DOI: 10.1126/scisignal.aal1683

Kiss-and-run or a full commitment?

Communication between sensory neurons underlies our sense of touch and temperature and is mediated by the release of neuropeptides from exocytic vesicles. In response to calcium influx, these vesicles can either fuse completely with the synaptic membrane and release all of their contents (called full-fusion release) or fuse transiently and release only some of their contents (called “kiss-and-run” release). Using single-vesicle imaging in rodent sensory neurons, Wang et al. discovered that the release mode used by a neuron was determined by the type of calcium channel that was activated. Activation of voltage-gated calcium channels (VGCCs) promoted greater calcium influx at the plasma membrane, which inhibited a protein that limits fusion pore size, thus enabling full-fusion release. Activation of ligand-gated TRPV1 calcium channels promoted partial but pulsed and thus more prolonged neuropeptide release. The findings provide insight into how calcium channels influence sensory neurotransmission.


Neuropeptides released from dorsal root ganglion (DRG) neurons play essential roles in the neurotransmission of sensory inputs, including those underlying nociception and pathological pain. Neuropeptides are released from intracellular vesicles through two modes: a partial release mode called “kiss-and-run” (KAR) and a full release mode called “full fusion–like” (FFL). Using total internal reflection fluorescence (TIRF) microscopy, we traced the release of pH-sensitive green fluorescent protein–tagged neuropeptide Y (pHluorin-NPY) from individual dense-core vesicles in the soma and axon of single DRG neurons after Ca2+ influx through either voltage-gated Ca2+ channels (VGCCs) or ligand-gated transient receptor potential vanilloid 1 (TRPV1) channels. We found that Ca2+ influx through VGCCs stimulated FFL and a greater single release of neuropeptides. In contrast, Ca2+ influx through TRPV1 channels stimulated KAR and a pulsed but prolonged release of neuropeptides that was partially mediated by Dynamin 1, which limits fusion pore expansion. Suppressing the Ca2+ gradient to an extent similar to that seen after TRPV1 activation abolished the VGCC preference for FFL. The findings suggest that by generating a steeper Ca2+ gradient, VGCCs promote a more robust fusion pore opening that facilitates FFL. Thus, KAR and FFL release modes are differentially regulated by the two principal types of Ca2+-permeable channels in DRG neurons.


The release of neurotransmitters—neuropeptides, monoamines, and neurotrophins—by dense-core vesicle (DCV) exocytosis plays essential roles in neuronal growth and activity, synaptic transmission and plasticity, and most neuron-based behaviors (1, 2). Triggered by an evoked rise in Ca2+ concentration in neurons, neuropeptide release from primary sensory dorsal root ganglion (DRG) neurons is of particular importance because these DCV transmitters are critically involved in nerve injury, inflammation, and therapies for spinal cord damage/disease (1). Neuropeptide Y (NPY), for example, is critical for the regulation of vasoconstriction and blood pressure, food intake and storage, anxiety, stress, and epileptic seizures (3, 4).

Specifically, Ca2+ influx through the two principal types of Ca2+ channels—voltage-gated Ca2+ channels (VGCCs) and transient receptor potential (TRP) channels (5, 6)—plays essential roles in DCV exocytosis in primary sensory DRG neurons in response to various sensory stimuli. VGCCs mediate neurotransmitter release in response to action potential (AP), whereas TRP vanilloid 1 (TRPV1) channels mediate the response to thermal and chemical stimuli in DRG neurons (7, 8). In addition to physiological sensation, both VGCCs and TRPV1 channels play essential roles in mediating nociception and pathological pain by triggering neuropeptide release (57). However, whether and how Ca2+ influx through these two principal types of Ca2+ channels differentially triggers DCV-mediated neuropeptide release remain to be determined. Although the mechanisms of neurotransmitter release have been extensively studied, analysis of DCVs has largely measured cumulative release from groups of neurons (911), but studies using total internal reflection fluorescence (TIRF) imaging in hippocampal neurons show that analysis of release profiles and kinetics of single-DCV events is possible (1215). Here, we used TIRF imaging of primary sensory DRG neurons transfected with a pH-sensitive green fluorescent protein–tagged construct of NPY (NPY-pHluorin) to investigate the modes, probability, and fusion pore kinetics of single-DCV release after different stimuli.


DCV exocytosis undergoes full-fusion and kiss-and-run modes in a single neuron

To image single-vesicle release, NPY-pHluorin was transiently expressed in DRG neurons. Immunofluorescence and confocal imaging revealed a punctate pattern of NPY-pHluorin abundance, most of which (>80%) colocalized with the DCV markers secretogranin II and calcitonin gene-related peptide (CGRP) (Fig. 1A). The DCV targeting and pH dependence of NPY-pHluorin were also confirmed by the fluorescence increase in response to application of an NH4Cl-containing solution (fig. S1). These findings demonstrated that the expressed NPY-pHluorin is sorted into DCVs in DRG neurons.

Fig. 1 VGCCs trigger more FFL release events than “KAR.”

(A) Representative confocal images showing the colocalization of NPY-pHluorin (NPY-pH) (green) with the endogenous DCV marker secretogranin II (SgII) (left, red; 84.9 ± 2.9%, n = 16 cells) and CGRP (right, red; 82.9 ± 1.9%, n = 22 cells) in DRG neurons. Scale bars, 5 μm. DIC, differential interference contrast. (B to F) Representative fluorescence time course, TIRF image frames (2.4 μm × 2.4 μm; 50-ms interval), and cartoons showing FFL (B and E) and KAR (C and F) release of NPY-pHlourin from DRG neurons. (D) Area selection for fluorescence calculation. Center diameter was 1.92 μm, whereas annulus was 2.4 μm each. Fluorescence intensity values of the 0.5-s baseline before the peak value were averaged and used as F0. (G) Histogram of release events per 100 μm2 from DRG neurons during a 10-s incubation with 2Ca70K (2 mM Ca2+- and 70 mM K+-containing solution) before [control (Ctrl)] and after a 90-s incubation with nifedipine (Nif; 10 μM) or ω-conotoxin GVIA (1 μM). Data are means ± SEM from n = 7 cells. *P < 0.05, paired Student’s t test. (H) Mean percentage of FFL and KAR events in DRG neurons induced by 70 mM K+ solution (n = 18 cells, ***P < 0.001, paired Student’s t test). (I) Mean fluorescence plots and the distribution of HHD in FFL and KAR events. Black (FFL) and gray (KAR) lines are fitted curves of the two release modes. n = 56 events for KAR, n = 73 events for FFL. Norm. intensity, normalized intensity. Red (FFL) and green (KAR) lines in the right panel are fitted curves of the two release modes. (J) A representative fluorescence trace showing the measurements of HHD and decay time constant τ fitted (red) to an exponential decay function (left). Distribution of decay time constant τ (n = 56 for KAR, n = 73 for FFL; right).

DCV exocytosis was simultaneously recorded in real time at 20 frames/s in single DRG neurons using a TIRF microscope (movie S1). Release events were identified by a transient fluorescence increase at the center of NPY-pHluorin puncta, followed by a decay to baseline (Fig. 1, B to F), which was abolished by the application of acidic solution (pH 5.5) but reversed rapidly in normal pH 7.4 solution (fig. S1C) (16, 17). Cell depolarization–evoked NPY exocytosis with 70 mM KCl–containing bath solution was greatly inhibited when Ca2+ influx was blocked by the VGCC blockers nifedipine (L-type) and ω-conotoxin GVIA (N-type) (Fig. 1G). Thus, Ca2+ influx through VGCCs triggers vesicular neuropeptide release from DRG neurons.

When the single NPY release events were characterized, we observed two types of fusion modes: full fusion–like (FFL; full release) and kiss-and-run (KAR; partial release) (Fig. 1, B to F), as reported in other neuronal and glial cells by us and others using amperometry, FM1-43 fluorescence imaging, and NPY-TIRF imaging (14, 15, 1823). In FFL (full content release) events, corresponding to spreading or discharge events reported earlier (19, 23), a robust fluorescence increase occurred both at the center and in the annular area of NPY-pHluorin puncta (Fig. 1, B and E), indicating either full collapse of a vesicle to the plasma membrane or a dilated fusion pore for full cargo release. In contrast, KAR (partial content release) events showed a brief appearance and brightening of the puncta, but no or only a limited fluorescence increase in the annular area (Fig. 1, C and F), representing a transient opening and reclosure of a restricted fusion pore that limits the release of NPY. Markedly, >80% of the total DCV fusion events were FFL when KCl was applied to DRG neurons (Fig. 1H). For physiological considerations of depolarizing stimulation using either physiological APs or high K+ solution (70 mM KCl), the DCV exocytosis events were similar after either high K+ or 20-Hz AP stimulation (fig. S2), suggesting that high K+ mimics “painful” high-frequency AP activity.

We next characterized the kinetics of single fusion events in DRG neurons (Fig. 1, I and J). The half-height duration (HHD) distribution differed markedly between FFL (~0.1 s) and KAR (~0.3 s) events (Fig. 1I, right panel). In contrast, the decay (τ) was similar between the two types (Fig. 1J, right panel), although they represent different processes during vesicle fusion. The fluorescence decay in FFL events is probably due to the rapid diffusion of the fluorescent cargo into the medium. In KAR events, it probably represents a combination of release, reacidification of the vesicle lumen, and vesicle retrieval back into the cytosol. Collectively, the data indicate that after stimulation by membrane depolarization, which activates VGCCs, FFL is the dominant release mode in DRG neurons.

TRPV1 favors the KAR release mode

DRG neurons respond to sensory stimuli upon activation of specialized channels (7). TRPV1 channels are activated by heat and environmental or ingested chemicals, such as H+ and capsaicin, respectively, and are modulated by inflammatory mediators that contribute to both nociception and inflammatory pain (7, 24). The TRPV1 agonist capsaicin (300 nM) was used to trigger TRPV1-mediated DCV exocytosis in DRG neurons (Fig. 2). Consistent with previous findings (7, 25), capsaicin-induced Ca2+ influx was mediated through plasma membrane channels, because it was abolished in Ca2+-free solution (fig. S3A). Unbiased screening revealed that 30.8% of DRG neurons were TRPV1-positive (fig. S3, B and C). In contrast, none of the DRG neurons from TRPV1-knockout (hereafter, TRPV1-KO) mice showed a rise in intracellular Ca2+ concentration ([Ca2+]i) in response to capsaicin (fig. S3, B and C). TIRF imaging revealed a substantial release of NPY in response to capsaicin, and this diminished in TRPV1-KO neurons (fig. S3, B to D), indicating that the capsaicin-induced [Ca2+]i rise and DCV release are mediated specifically by TRPV1 activation.

Fig. 2 Compared with VGCCs, TRPV1 activation triggers more KAR release events.

(A to C) Number of FFL events (A) and KAR events (B) and percentage of KAR events (C) induced in DRG neurons by 70 mM KCl (K+) and 300 nM capsaicin (Cap). Data are means ± SEM from n = 37 cells for KCl and n = 18 cells for capsaicin. *P < 0.05, unpaired Student’s t test. (D) Quantification and statistical analysis of release events per 100 μm2 in DRG neurons stimulated by 70 mM KCl or 0.3 μM capsaicin. Not significant by unpaired Student’s t test. (E) Capsaicin prolonged the fusion pore opening in KAR mode. Average fluorescence traces (left), cumulative plots of HHD (middle), and decay (τ; right) of KAR events stimulated by 70 mM KCl (n = 105) or 0.3 μM capsaicin (n = 149) are shown. ***P < 0.001, Kolmogorov-Smirnov test. Cum. frequency, cumulative frequency.

Next, we determined whether the DCV release modes were also sensitive to different types of Ca2+ channels. Unexpectedly, capsaicin (300 nM) evoked many more KAR events and fewer FFL events than high K+ (Fig. 2, A to C), although the total numbers of release events were about the same (Fig. 2D). The kinetics of single KAR events induced by capsaicin was also slower, with larger HHD and decay than those induced by high K+ (Fig. 2E), indicating that capsaicin induces a more prolonged and restricted opening of fusion pore. Collectively, these findings showed that the vesicle release modes are dependent on native stimulus modalities: Nerve activity/depolarization–activated VGCCs favor FFL exocytosis that leads to maximum DCV transmitter release per fusion, whereas TRPV1 channels activated by environmental factors (temperature, acid, and capsaicin) favor KAR exocytosis that leads to partial DCV transmitter release per fusion in DRG neurons.

Microdomain Ca2+ acts as a regulator of fusion mode switching

According to the release rate, the local [Ca2+]i gradient near the release site is much higher than the global level (2628), and this might contribute to selection of the vesicle release mode (20, 29). We next investigated the [Ca2+] gradients underneath the plasma membrane using TIRF imaging of Lck-GCaMP5g (Addgene plasmid 34924) in the plasma membrane of DRG neurons. In contrast to TRPV1 channels, which mediated weak and restricted local Ca2+ signals, activation of VGCCs triggered a global and much more robust Ca2+ gradient beneath the membrane (Fig. 3, A to C, and movie S2). Considering the observations that the total Ca2+ integrals were similar for the two channels (Fig. 3C, middle panel) and that the KCl-induced [Ca2+]i showed a faster rise time than that induced by capsaicin (fig. S4), we propose that subsurface [Ca2+] gradients are responsible for switching the release mode between KAR and FFL. To test this hypothesis, we preloaded DRG neurons with EGTA-AM (EGTA-acetoxymethyl ester) (200 μM) as an intracellular Ca2+ buffer, which limited the VGCC-mediated Ca2+ gradient to a level similar to that induced by TRPV1 activation (Fig. 3, A to C). Pretreatment with EGTA diminished the preference of VGCCs for FFL events (Fig. 3D), supporting our hypothesis that, with comparable global Ca2+ influxes (Fig. 3C, middle panel), these two principal types of Ca2+ channels mediate different levels of subsurface [Ca2+] gradient signals and thus trigger different modes (FFL and KAR) of neuropeptide release in DRG neurons.

Fig. 3 Microdomain [Ca2+]i is responsible for the switch between KAR and FFL release modes.

(A) Representative TIRF images of control (top) or EGTA-AM–treated (200 μM) (bottom) DRG neurons expressing Lck-GCaMP5g in the plasma membrane and stimulated with 300 nM capsaicin or 70 mM KCl (see also movie S2). Dashed outlines delineate the Ca2+ signal (hotspot) within seconds after 5 s of stimulation with capsaicin or KCl. Scale bars, 10 μm. (B) Representative fluorescence traces of the hotspot Ca2+ signals, indicated as regions of interest (ROIs) 1 to 4 in (A). The control trace [background (BK)] was selected from an extracellular region. (C) Quantification and statistical analysis of local [Ca2+]i (right) and global Ca2+ influx (middle) in DRG neurons induced by 300 nM capsaicin or 70 mM KCl with or without EGTA-AM treatment. Left: Total area of ROIs. a.u., arbitrary units. (D) Ratio of KAR- to FFL-mediated release of NPY-pHluorin from DRG neurons. The number of cells analyzed in each experiment is indicated within the bar graphs. Data are means ± SEM. *P < 0.05 and **P < 0.01, by one-way analysis of variance (ANOVA) (post hoc test).

To further confirm the critical role of local Ca2+ gradients in mediating different release modes upon VGCC and TRPV1 activation, we performed two-color TIRF imaging to directly visualize the distances between vesicles being released and Ca2+ channels (fig. S5A). Markedly, upon KCl stimulation, the KAR release events mainly occurred at the margins of the cell body, whereas FFL release events occurred closer to VGCCs (fig. S5, B and C), confirming the critical role of the local [Ca2+] in determining the release mode. In contrast, both the FFL and KAR events occurred near TRPV1 channels (with a lower local Ca2+ transient) upon capsaicin stimulation (fig. S5C). Thus, the differences in release mode are most likely due to differences in local Ca2+ integrals near the release sites. In physiological solutions, the fractional Ca2+ current (Pf) through TRPV1 channels, Pf(TRPV1), was only 5.3%, whereas Pf(VGCC) was 100% (fig. S6), indicating that the Ca2+ permeability of VGCCs is ~18 times that of TRPV1 channels under physiological conditions. This explains, in part, why the Ca2+ gradient for VGCCs is higher than that for TRPV1 channels. These results strongly support our hypothesis that VGCCs favor FFL release by mediating a more robust local Ca2+ gradient beneath the plasma membrane.

We next explored the possible fusion pore–regulating proteins involved in the selection of release mode. Dynamin (Dyn) functions to mediate the fission of endocytotic vesicles (30) and regulate fusion modes by limiting the fusion pore expansion during exocytosis, and these activities are modulated by cycles of Ca2+/calmodulin/calcineurin-dependent phosphorylation-dephosphorylation cycles (31). Our previous work also reported a potent role of Dyn in the fusion mode switching between FFL and KAR (21). We thus tested its possible contribution to fusion mode selection by the two principal types of Ca2+ channels (Fig. 4, A and B, and fig. S7). We found that, after VGCC activation by application of KCl, the release modes (the ratio of KAR to total events) and the release frequency remained similar in wild-type and Dyn1-knockout (Dyn1-KO) DRG neurons. However, upon capsaicin simulation, the ratio of KAR release events decreased markedly in Dyn1-KO neurons (Fig. 4, A and B, and fig. S7), indicating that Dyn1 functions as a fusion pore regulator to mediate DCV release upon TRPV1 activation. These are consistent with the finding that KCl stimulation evokes a more robust local Ca2+ signal and mediates more FFL events by the Ca2+-dependent dephosphorylation and inhibition of Dyn1, whereas the phosphorylated Dyn1 induces a switch from FFL to KAR upon TRPV1 activation (31).

Fig. 4 Dyn1 functions as a fusion pore regulator in TRPV1-mediated DCV release and a model of the release modes.

(A) Cartoon of the presumed role of Dyn1 in the regulation of fusion pore expansion and fusion mode preference. (B) Quantification and statistical analysis of the ratio of KAR with KCl (K+; 70 mM) or capsaicin (300 nM), as indicated in wild-type (WT) and Dyn1-KO DRG neurons. Data are means ± SEM from 7 to 11 cells in each treatment. *P < 0.05 and ***P < 0.001, by one-way ANOVA (post hoc test). (C) Extracellular CGRP concentrations in the medium of primary DRG cultures after a 30-s incubation with normal external solution (Ctrl) (see Materials and Methods) or the same solution containing 300 nM capsaicin or 70 mM KCl (30 s). Data are means ± SEM from n = 5 independent experiments for each condition. *P < 0.05 and ***P < 0.001, paired Student’s t test for Ctrl versus capsaicin or KCl and unpaired Student’s t test for capsaicin versus KCl. (D) Model of the channel-specific release modes in DRG neurons.

To determine whether native neuropeptide release is also differentially regulated by VGCCs versus TRPV1 channels, we measured CGRP release from cultured DRG neurons after stimulation with capsaicin or KCl that evoked comparable Ca2+ influxes (Fig. 3C). Consistent with the NPY-pHluorin release revealed by TIRF assays, KCl induced a greater CGRP release than did capsaicin (Fig. 4C), supporting a differential regulation of native CGRP release by VGCCs versus TRPV1 channels and confirming the specific coupling of release mode and stimulus modality. Furthermore, peripheral APs propagating along the skin–axon–spinal cord pathway triggered the release of DCV peptides (including CGRP) (fig. S2), which in turn activated autoreceptors and provided feedback to regulate the APs (32). This suggests a physiological relevance of somatic and proximal axonal release in DRG neurons (33, 34). Collectively, these results suggest that VGCCs and TRPV1 channels mediate different levels of subsurface Ca2+ signals and thus trigger different modes of neuropeptide release in DRG neurons (Fig. 4D).


Using real-time TIRF imaging of single-vesicle release after activation of VGCCs or TRPV1 Ca2+ channels in single DRG neurons, we uncovered channel-specific DCV release modes in neurons. We had previously reported that vesicle release modes are regulated either by Gi-βγ subunit in chromaffin cells (21) or by [Ca2+]i in astrocytes (20, 35). Our findings in this study indicate that although both TRPV1 channels and VGCCs trigger DCV release, they are preferentially coupled with two types of release modes: TRPV1 with KAR and VGCC with FFL (Fig. 4D). In astrocytes, the amount of transmitter release per vesicle fusion differs greatly between the two release modes (KAR/FFL, 1:20) (20). This illustrates the physiological relevance of our findings regarding the specific coupling of Ca2+ channels with release modes.

Regarding the mechanisms of the coupling between the two Ca2+ channels and the two release modes, we demonstrated that TRPV1 channels had a lower Ca2+ permeability [see also (36)] and preferentially triggered the KAR release mode, whereas VGCCs had an ~18 times greater Ca2+ permeability and preferentially triggered the FFL release mode. Thus, to determine the release mode of DCVs, the microdomain [Ca2+]i gradient around a channel is more critical than the total Ca2+ influx. This hypothesis was confirmed here by our direct measurements of microdomain [Ca2+]i after the activation of VGCCs and TRPV1 channels. Our present findings in DRG neurons are consistent with previous work in astrocytes, in which greater [Ca2+]i triggers FFL release whereas lower [Ca2+]i triggers KAR release of lysosomes (20, 37). The VGCC is exclusively selective for Ca2+ (38) and can produce up to 100 μM [Ca2+]i in nanodomains (that is the distance between a VGCC and the vesicle release site) (39). It remains to be investigated, but we propose that it is likely that this previously unknown mechanism regulates native neuropeptide (specifically CGRP) release underlying acute pain sensation.


Animals, cell culture, transfection, and plasmids

The use and care of animals were approved and directed by the Institutional Animal Care and Use Committee of Peking University and the Association for Assessment and Accreditation of Laboratory Animal Care. TRPV1-KO mice were provided by Z. Zhu (Third Military Medical University, Chongqing, China). Dyn1-KO mice were gifted by P. De Camilli (Yale University, New Haven, CT). Sprague-Dawley rats (~7 days old) were euthanized by an intraperitoneal injection of 0.15 ml of 10% chloral hydrate, and hypothermic anesthesia was used for mice. The DRGs of all spinal segments were isolated in ice-cold L15 medium (Gibco) and enzymatically dissociated in trypsin (0.2 mg/ml) and collagenase type 1A (1 mg/ml) containing Dulbecco’s modified Eagle’s medium/F12 for 40 min at 37°C. Cells were then dissociated by trituration and transfected with 3 μg of an NPY-pHluorin–expressing plasmid using a Neon (100-μl system) electroporation system (Invitrogen, MPK10096) according to the manufacturer’s instructions. The transfected cells were plated on polyethyleneimine-coated coverslips and cultured for 18 to 28 hours in a humidified incubator (37°C, 5% CO2) in Neurobasal-A medium supplemented with 2% B27 and 0.5 mM GlutaMAX-I (all from Gibco). NPY-pHluorin plasmid was constructed from NPY-Venus (a gift from N. Gamper, University of Leeds) and synapto-pHluorin (a gift from G. Miesenböck, University of Oxford). All chemicals were from Sigma, unless otherwise indicated.


DRG neurons transfected with NPY-pHluorin were fixed in 4% paraformaldehyde in phosphate-buffered saline (PBS) for 15 min at room temperature and permeabilized with 0.3% Triton X-100 for 5 min. Cells were blocked with 2% bovine serum albumin (BSA) in PBS for 1 hour at room temperature and then incubated overnight with primary antibodies diluted in PBS with 2% BSA at 4°C. After washing out the primary antibodies with blocking buffer, cells were incubated with Alexa Fluor 594–conjugated goat anti-rabbit immunoglobulin G (H+L) (Invitrogen, A11037) diluted in PBS with 2% BSA for 1 hour at room temperature. The nuclei were then stained with DAPI (4′,6-diamidino-2-phenylindole) and mounted with DAKO. Antibodies against CGRP (Peninsula, IHC6006) and secretogranin II (Abcam, ab12241) were used. Images were captured on an LSM 710 inverted confocal microscope (Carl Zeiss). For colocalization analysis, all intracellular puncta within 1-μm optical sections were selected and analyzed using the JACoP plugin of McMaster Biophotonics Facility ImageJ software (National Institutes of Health).

TIRF imaging, stimulation, and analysis

TIRF imaging was performed on an inverted microscope with a 100× TIRF objective lens (numerical aperture, 1.45; Olympus IX-81). Images were captured by an Andor electron-multiplying charge-coupled device using Andor iQ software with an exposure time of 50 ms. The standard bath solution contained the following: 150 mM NaCl, 5 mM KCl, 2.5 mM CaCl2, 1 mM MgCl2, 10 mM H-Hepes, and 10 mM d-glucose (pH 7.4). The temperature was kept at ~35°C throughout all TIRF experiments using a laboratory-made heater. Exocytotic events were defined as abrupt fluorescence increases, immediately followed by a fluorescence decrease or diffusion of NPY-pHluorin puncta in the vicinity. For the analysis of single release events, each event was selected and marked with 1.92-μm-diameter (center) and 2.4-μm-diameter (annulus) circular areas. Fluorescence intensity was calculated and analyzed using ImageJ; the intensity values during the 0.5-s baseline before the peak value were averaged and used as F0. In FFL (or spreading) events, a robust fluorescence increase occurred at both the center and the annular area of NPY-pHluorin puncta, representing the release and spread of NPY-pHluorin. KAR events showed a brief brightening of the puncta, but no or only a very limited fluorescence increase in the annular area, representing a transient opening and reclosure of a restricted fusion pore that limited the release of NPY. High K+ [85 mM NaCl, 70 mM KCl, 2.5 mM CaCl2, 1 mM MgCl2, 10 mM H-Hepes, and 10 mM d-glucose (pH 7.4)] and capsaicin (300 nM) were applied using a gravity-fed perfusion system. Electrical field stimulation (1 ms, 15 V) was applied through a pair of platinum wires using an electronic stimulator (Nihon Kohden, SEN-3201); the negative pole was placed in the vicinity of the cell under examination. For two-color TIRF imaging, the Ca2+ channels were labeled with mCherry (Cav2.2-mCherry and TRPV1-mCherry), and fusion events were visualized by NPY-pHluorin.

Fluorescence and fractional Ca2+ measurements

Intracellular calcium ([Ca2+]i) was measured using a Ca2+ imaging system (TILL Photonics). Fura-2 potassium salt (1.0 mM) was loaded into the cell via a patch pipette in the whole-cell configuration. The fluorescence was sampled at 2 or 10 Hz.

Fractional Ca2+ current, Pf, is defined as the percentage of Ca2+ current in the total current passing through a cation channel (Im in this case). According to the original definition (40),Pf=ICadtImdt=ΔFdFmax×Imdt(1)where Im is the total whole-cell current and ICa is the proposed fractional Im current carried by Ca2+. ΔFd is the change of Fd, which is the “modified Ca2+-sensitive fura-2 signal” before (Fdt0) and after (Fdt1) the voltage pulse or ligand-induced Ca2+ influx. Fd = F340 − F380, ΔFd = Fdt1 − Fdt0, and Fmax is a constant, which was determined by measuring the Ca2+ influx through VGCCs in the solutions specified above. Under physiological conditions, all ions contributing to the current through VGCCs are Ca2+, namely, Pf = 100%. From Eq. 1, Fmax = ΔFd/∫ICadt, where ICa = Im (which is the current through VGCCs). According to Eq. 1, after determining the Fmax by measuring the fura-2 signal that is evoked after activation of TRPV1, the Pf of each channel can be determined.

CGRP immunoassay

Basal and stimulated extracellular CGRP concentrations were evaluated in freshly isolated DRG neurons using an enzyme immunometric assay kit (Bachem), following the manufacturer’s instructions. Cells were washed three times with normal external solution and then incubated in the same solution for 30 s at room temperature, followed by another 30-s incubation in this solution containing 70 mM KCl or 300 nM capsaicin. The incubation solutions were collected for subsequent analysis of basal and stimulation-coupled CGRP levels. All samples were centrifuged at 13,000 rpm for 5 min, and the supernatants were processed for CGRP measurement. Samples were analyzed at 450 nm using a microplate reader (BioTek Synergy 4). CGRP concentrations (in picograms per milliliter) were extrapolated from a best-fit line calculated from serial dilutions of a CGRP standard. All data points were measured in triplicate.

Electrophysiological recordings

We used an EPC10/2 amplifier with Pulse software (HEKA Elektronik) to obtain whole-cell patch-clamp recordings as described previously (41, 42). Pipette resistance was controlled between 3 and 4 megohms when filled with an internal solution containing 153 mM CsCl, 1 mM MgCl2, 10 mM Hepes, and 4 mM Mg–adenosine 5′-triphosphate (pH 7.2). Normal external solution contained 150 mM NaCl, 5 mM KCl, 2.5 mM CaCl2, 1 mM MgCl2, 10 mM Hepes, and 10 mM glucose (pH 7.4). Igor software (WaveMetrics) was used for all offline data analyses. All experiments were performed at room temperature, unless otherwise indicated.

Calcium imaging

Changes of the [Ca2+]i in DRG neurons were measured as the Fluo-4/Fura-Red fluorescence ratio (43). Cells were loaded with 2.5 μM Fluo-4 AM and 5 μM Fura-Red AM (Invitrogen) dissolved in 0.2% dimethyl sulfoxide and 0.04% Pluronic F-127 in a standard bath solution at 37°C for 20 min. Cells were then washed and imaged on an inverted confocal microscope (Zeiss LSM 710). The fluorescent Ca2+ indicators were excited using a 488-nm laser, and the light emitted from the Fluo-4 and Fura-Red was recorded on separate channels at 500 to 540 nm and 600 to 680 nm, respectively. Images (512 × 512 pixels) were acquired at 1 Hz under a 40× oil objective lens (Zeiss). The Ca2+ level was determined from ROIs using the ratio of intensity traces recorded on the Fluo-4 and Fura-Red channels.


All experiments were replicated at least three times. Data were analyzed offline using ImageJ and Igor software. Data are means ± SEM. Statistical comparisons were performed using two-tailed unpaired Student’s t test, Kolmogorov-Smirnov test, or Mann-Whitney U test, as indicated. All tests were conducted using SPSS 13.0 (Statistical Package for the Social Sciences). Significance threshold was set at P < 0.05.


Fig. S1. NH4Cl dequenching of NPY-pHluorin and the pH-dependent release.

Fig. S2. KCl (70 mM) mimics 20-Hz AP stimulation in triggering release from DRG neurons.

Fig. S3. TRPV1 mediates Ca2+ influx and DCV exocytosis in DRG neurons.

Fig. S4. The delay to peak amplitude of the Ca2+ signal in response to capsaicin or KCl.

Fig. S5. Distance between Ca2+ channels and vesicle release sites revealed by two-color TIRF imaging of DRG neurons.

Fig. S6. Pf of TRPV1 channels in DRG neurons.

Fig. S7. Dyn1 acts as a fusion pore regulator in TRPV1-mediated vesicle release in DRG neurons.

Movie S1. Imaging KCl-induced NPY-pHluorin release from a DRG neuron.

Movie S2. Imaging TRPV1- and VGCC-mediated Ca2+ rise beneath the plasma membrane.


Acknowledgments: We thank Z. Zhu (Third Military Medical University, China) for the TRPV1-KO mice, P. De Camilli (Yale University, USA) for the Dyn1-KO mice, N. Gamper (Leeds University, UK) for help in culturing DRG neurons, C. X. Zhang for help in the initial experiments, and I. Bruce (Peking University) for reading of the manuscript. Funding: This work was supported by grants from the National Basic Research Program of China (2012CB518006 and 2016YFA0500401) and the National Natural Science Foundation of China (31228010, 31171026, 31100597, 31327901, 31221002, 31330024, 31400708, 31521062, and 31670843). C.W. was supported in part by a postdoctoral fellowship from the Peking-Tsinghua Center for Life Sciences. Author contributions: Yeshi W., Q.W., C.W., M.H., B.L., Z.C., R.H., Yuan W., H.X., L. Zhou, and L. Zheng carried out the experiments. Yeshi W., Q.W., and C.W. analyzed the data. Z.Z., C.W., and Yeshi W. contributed to planning the work. Z.Z., Yeshi W., Q.W., and C.W. wrote the manuscript. Competing interests: The authors declare that they have no competing interests.
View Abstract

Navigate This Article