Research organism

elife

eLife

2050-084X

eLife Sciences Publications, Ltd

58359

10.7554/eLife.58359

Short Report

Cell Biology

Neuroscience

Synaptotagmin-1 is the Ca²⁺ sensor for fast striatal dopamine release

Banerjee

Aditi

https://orcid.org/0000-0003-2016-0717

1Lee

Jinoh

https://orcid.org/0000-0002-2158-8507

1Nemcova

Paulina

https://orcid.org/0000-0002-0323-8079

1†Liu

Changliang

1Kaeser

Pascal S

https://orcid.org/0000-0002-1558-1958

kaeser@hms.harvard.edu11Department of Neurobiology, Harvard Medical School

Boston

United States

Chen

Reviewing EditorStanford UniversityUnited StatesWestbrook

Gary L

Senior EditorOregon Health and Science UniversityUnited States

†

Department of Neuronal Plasticity, Max Planck Institute of Psychiatry, Munich, Germany

03062020

2020

e58359

2804202002062020

2020

Banerjee et al

http://creativecommons.org/licenses/by/4.0/

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Dopamine powerfully controls neural circuits through neuromodulation. In the vertebrate striatum, dopamine adjusts cellular functions to regulate behaviors across broad time scales, but how the dopamine secretory system is built to support fast and slow neuromodulation is not known. Here, we set out to identify Ca²⁺-triggering mechanisms for dopamine release. We find that synchronous dopamine secretion is abolished in acute brain slices of conditional knockout mice in which Synaptotagmin-1 is removed from dopamine neurons. This indicates that Synaptotagmin-1 is the Ca²⁺ sensor for fast dopamine release. Remarkably, dopamine release induced by strong depolarization and asynchronous release during stimulus trains are unaffected by Synaptotagmin-1 knockout. Microdialysis further reveals that these modes and action potential-independent release provide significant amounts of extracellular dopamine in vivo. We propose that the molecular machinery for dopamine secretion has evolved to support fast and slow signaling modes, with fast release requiring the Ca²⁺ sensor Synaptotagmin-1.

dopaminesecretionneuromodulationstriatumcalcium sensor

Research organismMouse

http://dx.doi.org/10.13039/100000065

National Institute of Neurological Disorders and Stroke

R01NS103484

Kaeser

Pascal S

http://dx.doi.org/10.13039/100006691

Harvard Medical School

Dean's Initiative Award

Kaeser

Pascal S

http://dx.doi.org/10.13039/100006691

Harvard Medical School

Hearst Fellowship

Banerjee

Aditi

http://dx.doi.org/10.13039/100006691

Harvard Medical School

Brooks Fellowship

Banerjee

Aditi

http://dx.doi.org/10.13039/100006691

Harvard Medical School

Gordon Fellowship

Liu

Changliang

http://dx.doi.org/10.13039/100008328

Marshallplan-Jubiläumsstiftung

Exchange Scholarship

Nemcova

Paulina

http://dx.doi.org/10.13039/100006691

Harvard Medical School

Harvard-MIT Joint research grant

Kaeser

Pascal S

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Author impact statement

Genetic and electrophysiological analyses reveal calcium-triggering mechanisms for dopamine release in the striatum that may enable fast and slow dopamine coding.

Introduction

Dopamine is an important neuromodulator in the vertebrate brain, but the secretory biology of dopamine is not well understood. A key dopamine pathway arises from midbrain dopamine neurons located in the substantia nigra pars compacta. Their axons send projections to the dorsal striatum, where dopamine neuromodulation controls initiation and execution of movement. A prominent model is that dopamine operates slowly and on distant receptors through volume transmission (Agnati et al., 1995; Liu and Kaeser, 2019; Sulzer et al., 2016). Recent studies, however, have started to suggest that dopamine can modulate the neuronal membrane potential (Beckstead et al., 2004), structural synaptic plasticity (Yagishita et al., 2014) and behavior (Howe and Dombeck, 2016; Menegas et al., 2018) with temporal precision in the range of tens of milliseconds, suggesting the presence of molecular machines for rapid dopamine coding.

A requirement for Ca²⁺-triggering of secretion is the presence of Ca²⁺ sensors. Various Ca²⁺ binding proteins are used as Ca²⁺ sensors for vesicular exocytosis (Kaeser and Regehr, 2014; Pang and Südhof, 2010), and each could be a candidate for dopamine release. Fast synaptic transmission relies on Synaptotagmin-1, –2 or –9 (Fernández-Chacón et al., 2001; Sun et al., 2007; Xu et al., 2007). Synapses without these fast Synaptotagmins have prominent asynchronous release (Turecek and Regehr, 2019). At these asynchronous and other synapses, the higher affinity Ca²⁺ sensors Synaptotagmin-7 and Doc2, and possibly additional sensors, mediate asynchronous release (Bacaj et al., 2013; Kaeser and Regehr, 2014; Wen et al., 2010; Yao et al., 2011). At inner ear ribbon synapses, release is triggered by otoferlin, a Ca²⁺ sensor with different Ca²⁺ binding properties and kinetics (Michalski et al., 2017; Roux et al., 2006). In chromaffin cells, which release catecholamines, a major release component is left after Synaptotagmin-1 deletion, and this component is likely mediated by Synaptotagmin-7 (Schonn et al., 2008; Voets et al., 2001; de Wit et al., 2009). Knockdown of Synaptotagmin-1, –4 or –7 resulted in partial impairments of [³H]-dopamine released into the supernatant in response to KCl depolarization of cultured midbrain neurons, and BDNF release is also modulated by Synaptotagmin-4, but at least Synaptotagmin-4 is unlikely to operate as a Ca²⁺ sensor in these experiments (Dai et al., 2004; Dean et al., 2009; Mendez et al., 2011; Wang and Chapman, 2010). In this study, we find that fast dopamine secretion is abolished in the striatum of mouse mutants that lack Synaptotagmin-1 in dopamine neurons, and conclude that Synaptotagmin-1 is the Ca²⁺ sensor for fast dopamine release.

Results and discussion

We here set out to identify Ca²⁺-triggering mechanisms for rapid dopamine signaling. First, we analyzed the dependence of striatal dopamine release on the extracellular Ca²⁺ concentration ([Ca²⁺]_ex) in acute brain slices. We generated mice that express channelrhodopsin-2 (ChR2) selectively in dopamine neurons using mouse genetics (Figure 1A) and measured optogenetically evoked dopamine transients in slices of the dorsal striatum using carbon fiber amperometry (Figure 1—figure supplement 1). Similar to electrical stimulation paradigms (Brimblecombe et al., 2015; Ford et al., 2010), optogenetically triggered dopamine release was steeply [Ca²⁺]_ex dependent below 2 mM [Ca²⁺]_ex (Figure 1B and C). At 2 mM [Ca²⁺]_ex, the 20–80% rise time was 1.91 ± 0.13 ms, which includes the diffusion of dopamine from the release site to the electrode, and rise times slowed down in low [Ca²⁺]_ex (Figure 1D). Together, these data establish that action potential-triggered dopamine release is mostly synchronous, and suggest the presence of a fast, low-affinity Ca²⁺ sensor.

Figure 1.Synaptotagmin-1 is required for synchronous dopamine release.

(A) Schematic of the experimental setup for Cre-dependent expression of channelrhodopsin variants using mutant mice (B–D) and AAVs (F–I). (B–D) Sample traces (B, average of four sweeps), and quantification of peak amplitudes (C) and 20–80% rise times (D) of dopamine release evoked by optogenetic activation (1 ms light pulse at 470 nm) at different [Ca²⁺]_ex, n = 7 slices/4 mice at each [Ca²⁺]_ex. (E) Schematic of the generation of dopamine neuron specific Synaptotagmin-1 knockout (Syt1-cKO^DA) mice. (F, G) Sample traces (F, average of four sweeps) and quantification of peak amplitudes (G) of dopamine release evoked by a 1 ms light pulse, Syt-1 control n = 8 slices/4 mice, Syt-1 cKO^DAn = 8/4. (H, I) Sample traces (H, average of four sweeps) and quantification (I) of dopamine release evoked by ten 1 ms light pulses at 10 Hz. Amplitudes are normalized to the average first amplitude in Syt-1 control, Syt-1 control n = 8/4, Syt-1 cKO^DA n = 8/4. All data are shown as mean ± SEM, *p<0.05, ***p<0.001. Recordings are performed in 2 mM [Ca²⁺]_ex unless noted otherwise, statistical significance was determined by one-way ANOVA followed by Dunnett’s multiple comparisons test in C and D, by Mann-Whitney test in G, and by two-way ANOVA followed by Sidak’s multiple comparisons test in I (*** for genotype, stimulus number and interaction, post-tests: *** for stimuli 1 and 2, ** for stimulus three and p>0.05 for stimuli 4–10).

Figure 1—figure supplement 1.Carbon fiber recordings.

(A) Sample traces and summary plot of the calibration of a carbon fiber electrode (CFE) with different concentrations of dopamine applied to the electrode by a puff pipette. The standard curve generated by linear regression confirms a linear relationship. Calibrations were done for each electrode, each electrode was only used for one day, and on each day of recording a control and test littermate animal were compared with the same electrode. (B) Sample traces of CFE currents in response to 20 µM dopamine puffs at different [Ca²⁺]_ex. The sensitivity of the CFE is not dependent on changes in [Ca²⁺]_ex. (C, D) Sample traces (average of four sweeps) and quantification of peak dopamine release evoked by optogenetic stimulation performed as in Figure 1A–D, but at 2 mM vs. 4 mM [Ca²⁺]_ex, n = 7 slices/3 mice. The modest enhancement of peak dopamine detected in response to increasing [Ca²⁺]_ex from 2 mM to 4 mM is likely due to the high initial release probability of dopamine secretion at 2 mM [Ca²⁺]_ex, which we described before (Liu et al., 2018). Data in D are shown as mean ± SEM, *p<0.05, statistical significance was determined by paired Student’s t-test in D.

Figure 1—figure supplement 2.Synaptotagmin-1 is present in TH-positive striatal synaptosomes.

(A) Schematic of the striatal synaptosome preparation. (B, C) Sample confocal images (B) and quantification (C) of striatal synaptosomes stained for Synaptotagmin-1, the release site marker Bassoon and the dopamine neuron marker TH. Consistent with experiments from cultured midbrain neurons (Mendez et al., 2011), Syt-1 is often present in synaptosomes that are positive for Bassoon and TH (solid arrowhead). It is infrequently present in TH-positive particles that do not contain Bassoon (arrow) and is sometimes present in TH negative synaptosomes that contain Bassoon and are from non-dopaminergic axons (hollow arrowheads, striatal synapses rely in part on Syt-9 instead of Syt-1 [Xu et al., 2007]), n = 21 images/2 mice, each circle represents the average result of an image containing 300–600 synaptosomes. Data are mean ± SEM, ***p<0.001, statistical significance was determined by one-way ANOVA followed by Dunnett’s multiple comparisons test.

Figure 1—figure supplement 3.Action potential firing of dopamine axons persists upon Synaptotagmin-1 knockout.

(A) Schematic of the extracellular recording. Dopamine axonal firing was induced optogenetically in the dorsal striatum and field potentials were recorded with an extracellular recording pipette filled with ACSF. Recordings were performed as described in the Materials and methods, and it was previously established that these potentials reflect action potential firing of dopamine axons (Liu et al., 2018). (B–E) Sample traces (B, C, average of 100 sweeps) and analyses (D, E) of extracellular potentials recorded during a 10 Hz train in Syt-1 control and Syt-1 cKO^DA slices. The response to the 1^st stimulus in B is magnified in C and shown as before (black) and after (grey) application of the sodium channel blocker TTX (1 µM). Quantification of the amplitude before and after TTX is shown in D, quantification of all responses in the 10 stimulus 10 Hz train normalized to the average first amplitude in Syt-1 control is shown in E, Syt-1 control n = 6 slices/3 mice, Syt-1 cKO^DA n = 6/3 in D and E. (F, G) Same experiment as in B and E but at a stimulation frequency of 40 Hz, the first 10 of a total of 40 stimuli are shown. The higher stimulation frequency leads to action potential failures and depression of the normalized amplitudes after the first stimulus, Syt-1 control n = 5/4, Syt-1 cKO^DA n = 4/4. Data are mean ± SEM, **p<0.01, statistical significance was determined by one-way ANOVA followed by Sidak’s multiple comparisons test in D, and two-way ANOVA followed by Sidak’s multiple comparisons test in E (ns, p>0.05 for genotype, stimulus number and interaction) and G (***p<0.001 for stimulus number, p>0.05 for genotype and interaction).

Figure 1—figure supplement 4.Release evoked by electrical stimulation is abolished upon Synaptotagmin-1 knockout.

(A) Schematic of the experiment. Electrical stimulation activates dopamine fibers and cholinergic interneurons, and cholinergic innervation of dopamine axons accounts for as much as ~90% of the extracellular dopamine detected upon electrical stimulation and leads to enhanced depression of dopamine release during stimulus trains (Liu et al., 2018; Threlfell et al., 2012; Zhou et al., 2001). (B) Sample traces (single sweep) and (C) quantification of peak dopamine release in Syt-1 control and Syt-1 cKO^DAslices evoked by single electrical stimuli at increasing intensities (10–90 µA), Syt-1 control n = 7 slices/3 mice, Syt-1 cKO^DA n = 7/3. (D, E) Sample traces (D, average of four sweeps) and (E) all responses normalized to the average first amplitude in Syt-1 control, evoked by 10 Hz 10-stimulus train using electrical stimulation. Syt-1 control n = 3/3 mice, Syt-1 cKO^DAn = 5/3. Data are mean ± SEM, ***p<0.001, statistical significance in C and E were determined by two-way ANOVA (C, E: *** for genotype, stimulus number and interaction) followed by Sidak’s multiple comparisons test (C: *** each for 40–90 μA; E: *** for first stimulus).

The Allen Brain Atlas and single cell sequencing data (Lein et al., 2007; Saunders et al., 2018) suggest that of the putative Ca²⁺ sensors, Synaptotagmin-1 and –7 expression levels are high in midbrain dopamine neurons, while expression of the other candidates, Synaptotagmin-2 and −9, Doc2 and otoferlin, appears to be low. Using subcellular fractionation, we found that Synaptotagmin-1 is present in striatal synaptosomes that were positive for the dopamine cell marker tyrosine hydroxylase (TH) and the active zone protein Bassoon (Figure 1—figure supplement 2). We hypothesized that Synaptotagmin-1 is the main Ca²⁺ sensor for dopamine release. We generated conditional knockout mice in which we deleted Synaptotagmin-1 from dopamine neurons (Syt-1 cKO^DA mice, Figure 1E) by crossing mice with ‘floxed’ conditional alleles for Synaptotagmin-1 (Skarnes et al., 2011; Zhou et al., 2015) to DAT^IRES-Cre mice (Bäckman et al., 2006). In these mice, we expressed oChiEF-citrine, a fast channelrhodopsin, selectively in dopamine neurons using AAVs (Figure 1A) to optogenetically evoke dopamine release through triggering of axonal action potentials (Liu et al., 2018). A 1 ms light pulse triggered dopamine release with a rise time of 1.81 ± 0.23 ms in control mice, but in Syt-1 cKO^DA mice dopamine release was effectively abolished (Figure 1F and G). During short stimulus trains (10 stimuli at 10 Hz), dopamine release strongly depressed in Syt-1 control mice (Figure 1H and I), which was not due to action potential failures (Figure 1—figure supplement 3), but likely a consequence of the high initial release probability (Figure 1—figure supplement 1C and D, and Liu et al., 2018). In Syt-1 cKO^DA mice, stimulus trains failed to evoke measurable dopamine release. Hence, Synaptotagmin-1 is likely the main Ca²⁺ sensor that mediates synchronous dopamine release.

Striatal dopamine release is not only triggered by ascending action potentials, but also by cholinergic interneurons that innervate dopamine axons and trigger release via activation of axonal nicotinic acetylcholine receptors (nAChRs) (Threlfell et al., 2012; Zhou et al., 2001). This mechanism dominates in response to electrical stimulation in the slice preparation used here and accounts for as much as 90% of the released dopamine (Liu et al., 2018). We used electrical stimulation to assess whether Synaptotagmin-1 triggers dopamine release initiated by nAChR activation. Electrically evoked dopamine release was also abolished in Syt-1 cKO^DA mice (Figure 1—figure supplement 4), indicating that Synaptotagmin-1 mediates both release evoked by ascending action potentials and release triggered by nAChR activation.

To assess whether loss of Synaptotagmin-1 and synchronous dopamine release has effects on striatal structure, we used 3D-structured illumination superresolution microscopy (Gustafsson et al., 2008; Liu et al., 2018). Striatal dopamine axons were labeled by TH, and their length and density were unchanged in Syt-1 cKO^DA mice (Figure 2A–C). Dopamine release sites can be marked by Bassoon (Liu et al., 2018), and Bassoon clustering inside TH axons was not strongly affected in Syt-1 cKO^DA mice (Figure 2D and E). Bassoon cluster volumes were unchanged, and there was only a very mild increase in Bassoon cluster densities in Syt-1 cKO^DA. Local shuffling of Bassoon objects decreased Bassoon density but increased the volume of the clusters artificially localized inside of TH axons in both genotypes. This confirms that Bassoon clusters are more frequent within TH axons than in areas surrounding these axons and that the Bassoon clusters in dopamine axons are smaller than the nearby Bassoon clusters (Liu et al., 2018), and indicates signal specificity. These experiments indicate that dopamine axons and release sites develop mostly normally in Syt-1 cKO^DA mice despite the strong impairments in dopamine release.

Figure 2.Dopamine axon structure and depolarization induced dopamine release are intact after ablation of Synaptotagmin-1.

(A) Sample 3D-SIM images of striatal brain sections of Syt-1 control and Syt-1 cKO^DA mice stained for the dopamine axon marker TH and the release site marker Bassoon. Volume rendered reconstructions (10 × 10 × 2 µm³, top, all Bassoon is shown), surface rendering of the same volumes (middle, all Bassoon is shown), and zoomed-in volumes (bottom, 5 × 3 × 2 µm³ in front view and rotated by +90° along the x-axis, only including Bassoon clusters with >40% volume overlap with TH) are shown. (B–E) Quantification of the fraction of the image volume covered by TH (B), TH axon length (C), Bassoon cluster densities (D) and Bassoon cluster volumes (E). For the shuffled controls in D and E, each Bassoon object was randomly relocated 1000 times within a volume of 1 × 1 × 1 µm³, and the actual Bassoon densities and volumes were compared to the average of the shuffled controls. Syt-1 control: n = 43 images/5 slices/3 mice, Syt-1 cKO^DA n = 41/5/3. (F–I) Schematic of the experiment (F), sample traces (G) and analyses of peak amplitudes (H) and total dopamine (I, quantified as area under the curve) of dopamine release measured in response to puffing of 100 mM KCl onto the recording area. Syt-1 control: n = 7 slices/3 mice, Syt-1 cKO^DA n = 7/3. All data are shown as mean ± SEM, *p<0.05, ***p<0.001, statistical significance was determined by unpaired t tests in B and C, one-way ANOVA followed by Sidak’s multiple comparisons test in D and E, and Mann-Whitney tests in H and I .

We next tested whether strong depolarization could lead to dopamine release in the absence of Synaptotagmin-1. Local puffing of KCl onto brain slices causes a strong depolarization of dopamine axons and surrounding neurons, for example cholinergic interneurons, which triggers massive dopamine release that requires the active zone protein RIM in dopamine axons (Liu et al., 2018). Compellingly, puffing KCl directly onto the recording site evoked an amperometric response that was indistinguishable between control and Syt-1 cKO^DA mice (Figure 2F–I). This establishes that dopamine is produced, loaded into vesicles, and released by strong depolarization throughout a structurally largely normal dopamine axon in Syt-1 cKO^DA mice. While the exact mechanism of KCl depolarization induced dopamine release is not known, the data further suggest that depolarization-induced, likely massive Ca²⁺ entry may trigger vesicular dopamine release via one or multiple alternative Ca²⁺ sensors in the absence of Synaptotagmin-1.

These additional sensors may for example mediate asynchronous release, a form of vesicular exocytosis at synapses that is triggered with a longer, variable delay in response to action potentials and Ca²⁺ entry (Kaeser and Regehr, 2014; Pang and Südhof, 2010). However, if asynchronous dopamine release is present in Syt-1 cKO^DA mice, the amount of dopamine after one or 10 action potentials is too small to be detected (Figure 1F–I). It is possible that the dopamine transporter (DAT) is sufficient to mediate re-uptake of dopamine before extracellular accumulation is observed. To test this hypothesis, we repeated the optogenetic experiments with 10 Hz stimulus trains before and after wash-in of the DAT-blocker nomifensine. In control slices, nomifensine resulted in a reduction of the first amplitude in the train (Figure 3A and B), which may be due to suppression mediated by tonic axonal D2-receptor activation (Benoit-Marand et al., 2001; Ford, 2014). When we assessed release throughout the train (measured as the area under the curve), a robust enhancement was observed in Syt-1 control and Syt-1 cKO^DA mice (Figure 3C), and the total increase in extracellular dopamine upon DAT blockade was similar between Syt-1 control and Syt-1 cKO^DA (Figure 3D). When normalized to the measured extracellular dopamine before DAT blockade, it amounted to 1.7-fold and 31.5-fold enhancements in Syt-1 control and cKO^DA mice, respectively (Figure 3E). These experiments indicate that asynchronous release is present and persists in Syt-1 cKO^DA mice, and this can be detected upon blockade of dopamine re-uptake. While sensors for asynchronous dopamine release are not known, the presence of Synaptotagmin-7 in substantia nigra dopamine neurons and its role in asynchronous release at fast synapses (Bacaj et al., 2015; Lein et al., 2007; Mendez et al., 2011; Saunders et al., 2018; Wen et al., 2010) makes Synaptotagmin-7 a candidate sensor protein.

Figure 3.Asynchronous dopamine release sustains extracellular dopamine in vivo after ablation of Synaptotagmin-1.

(A–E) Sample traces (A, average of four sweeps) and quantification of dopamine release (B–E) evoked by a 10 Hz stimulus train induced as described in Figure 1H before (black traces) and after addition of the DAT blocker nomifensine (10 µM, purple traces) in Syt-1 control and Syt-1 cKO^DA slices. Amplitudes of the first response (B, Syt-1 control only), total dopamine (area under the curve for 2.935 s after the 1^st stimulus) before and after nomifensine (C), subtracted area (D) and fold increase after nomifensine (E) are shown, Syt-1 control n = 7 slices/5 mice, Syt-1 cKO^DA n = 7/5. (F, G) Schematic of the experiment (F) and summary plot (G) of in vivo dopamine measurements using microdialysis in the dorsal striatum (microdialysates were collected over periods of 15 min and values measured in these microdialysates at the end of each period are plotted) of Syt-1 control and Syt-1 cKO^DA mice. The quantification in G shows dopamine levels normalized to the average concentration from the 76^th - 120^th min in Syt-1 control, and reverse dialysis of 10 µM TTX to block action potential firing started at 121 min, Syt-1 control n = 5 mice, Syt-1 cKO^DA n = 5 mice. All data are shown as mean ± SEM, **p<0.01, ***p<0.001, statistical significance was determined by Wilcoxon matched pairs signed rank tests in B and C, Mann Whitney tests in D and E, and by two-way ANOVA followed by Sidak’s multiple comparisons test in G. In G, the data followed a lognormal distribution and statistical testing was done after the data were converted to a log_e scale, *** for time and ** for interaction, post-tests for genotypes are shown.

In vivo, asynchronous release may significantly contribute to extracellular dopamine. To test this hypothesis, we performed microdialysis in anesthetized mice. In Syt-1 control mice, extracellular dopamine levels were reduced to approximately one third upon reverse dialysis of the sodium channel blocker tetrodotoxin (TTX), which inhibits action potential firing (Figure 3F and G). Remarkably, extracellular dopamine levels before TTX reverse dialysis were only mildly reduced in Syt-1 cKO^DA mice and action potential blockade robustly reduced extracellular dopamine in these mice. Hence, asynchronous release may provide a substantial amount of extracellular dopamine in vivo. These data suggest that three modes of dopamine release exist: synchronous and asynchronous release in response to action potentials, and action potential independent release that is likely mediated by spontaneous exocytotic events. Remarkably, each component appears to contribute significantly to the extracellular dopamine measured by microdialysis in anesthetized mice, suggesting that asynchronous and action potential-independent release may be prominent. It is not known whether Synaptotagmin-1 knockout affects spontaneous dopamine release, but literature from conventional synapses establishes that Synaptotagmin-1 knockout strongly enhances miniature synaptic vesicle release (Broadie et al., 1994; Xu et al., 2009). Hence, it is possible that Syt-1 cKO^DA leads to increased spontaneous dopamine release in the striatum. But the observation that extracellular dopamine levels after TTX are not increased in Syt-1 cKO^DA mice suggests that there is no dramatic enhancement of miniature dopamine release, that microdialysis is not sufficiently sensitive to detect such a change, or that an enhancement is counteracted by dopamine clearance.

Here, we find that synchronous striatal dopamine release requires the fast Ca²⁺ sensor Synaptotagmin-1. Given the prevailing model of volume transmission, which postulates that dopamine neurotransmission is slow and imprecise, it is surprising that dopamine neurons have evolved to employ a fast Ca²⁺ sensor. This finding, however, is in line with a recent description of sparse secretory hotspots in striatal dopamine axons (Liu et al., 2018). We propose that at these sparse sites, dopamine is rapidly and synchronously released with a high vesicular release probability to generate an extracellular dopamine signal that is spatially restricted and has rapid kinetics. While the dopamine receptor distribution relative to the dopamine release hotspots is not known, our data generally suggest that dopamine transmission may be fast and compartmentalized within a target cell. These mechanisms may support fast dopamine coding functions (Howe and Dombeck, 2016; Menegas et al., 2018; Yagishita et al., 2014). Our work further suggests that slower signaling modes exist. Substantial amounts of extracellular dopamine may come from asynchronous and action potential-independent dopamine release. Future studies should address which dopamine functions rely on fast dopamine signaling machinery, and whether some functions are supported by slower signaling mechanisms.

Materials and methods

Key resources table

Reagent type (species) or resource	Designation	Source or reference	Identifiers	Additional information
Genetic reagent (M. musculus)	B6.SJL-Slc6a3^{tm1.1(cre)Bkmm}/J; DAT^IRES-Cre	Bäckman et al., 2006	JAX 006660, RRID:IMSR_JAX:006660
Genetic reagent (M. musculus)	B6.129S-Gt(ROSA)26Sor^{tm32(CAG-COP4*H134R/EYFP)Hze}; Ai32	Madisen et al., 2012	JAX 012569, RRID:IMSR_JAX:012569
Genetic reagent (M. musculus)	C57BL/6NTac-^{Syt1tm1a(EUCOMM)Wtsi}/WtsiCnrm; Syt-1 floxed; Syt1^lox/lox	Skarnes et al., 2011; Zhou et al., 2015; obtained from Dr. T.C. Südhof	EUCOMM (EM:06829), RRID:IMSR_EM:06829	Conditional Synaptotagmin-1 floxed mice
Cell line (H. sapiens)	HEK293T	ATCC	Cat#: CRL-3216, RRID:CVCL_0063
Recombinant DNA reagent	AAV-flex-oChIEF-citrine	Addgene; Lin et al., 2009	Plasmid# 50973, RRID:Addgene_50973
Chemical compound, drug	Nomifensine	Tocris	Cat. No. 1992
Chemical compound, drug	Tetrodotoxin	Tocris	Cat. No. 1078
Antibody	Mouse monoclonal IgG2a anti-Bassoon SAP7F407 (A85)	Enzo Life Sciences	Cat# ADI-VAM-PS003-F, RRID:AB_11181058	IHC (1:500)
Antibody	Guinea pig polyclonal anti-Tyrosine hydroxylase (A111)	Synaptic Systems	Cat# 213 104, RRID:AB_2619897	IHC (1:1000)
Antibody	Rabbit polyclonal anti-Synaptotagmin-1 antiserum (A24)	gift from Dr. T.C. Südhof	V216	ICC (1:1000)
Other	Microdialysis probe	Harvard Apparatus	Item# CMA8309581
Other	Carbon fiber filaments	Goodfellow	Item# C 005722
Software, algorithm	Fiji	Schindelin et al., 2012	RRID:SCR_002285, https://imagej.net/Fiji/Downloads	Used for confocal synaptosome and 3D-SIM slice experiments
Software, algorithm	SoftWoRX	GE Healthcare	http://incelldownload.gehealthcare.com/bin/download_data/SoftWoRx/7.0.0/SoftWoRx.htm	Used for 3D reconstruction
Software, algorithm	Custom MATLAB code	Liu et al., 2018	https://github.com/hmslcl/3D_SIM_analysis_HMS_Kaeser-lab_CL (Liu, 2017)	Used for analysis of 3D-SIM and synaptosome images
Software, algorithm	Prism8	GraphPad	RRID:SCR_002798, https://www.graphpad.com/scientific-software/prism	Used for statistical analysis

Mice

DAT^IRES-Cre mice (Bäckman et al., 2006) express Cre recombinase under the dopamine transporter (DAT) gene locus, and were obtained from the Jackson laboratories (RRID:IMSR_JAX: 006660, B6.SJL-Slc6a3^{tm1.1(cre)Bkmm}/J). Mice for the Cre-dependent expression of ChR2 (Madisen et al., 2012) (Ai32, RRID:IMSR_JAX:012569, B6.129S-Gt(ROSA)26Sor^{tm32(CAG-COP4*H134R/EYFP)Hze}), obtained from the Jackson Laboratories, were crossed to DAT^IRES-Cre mice, and mice used for experiments were heterozygote for both DAT^IRES-Cre and ChR2. The conditional Synaptotagmin-1 knockout mice were generated by EUCOMM (EM:RRID:IMSR_EM:06829, C57BL/6NTac-^{Syt1tm1a(EUCOMM)Wtsi}/WtsiCnrm) (Skarnes et al., 2011), described in Zhou et al., 2015, and obtained from Dr. T.C. Südhof. Syt-1 cKO^DA mice were generated by crossing Syt-1 floxed mice with DAT^IRES-Cre mice, and for all experiments, Syt-1 cKO^DA mice were mice homozygote for the Syt-1 floxed allele and heterozygote for DAT^IRES-Cre. Syt-1 control mice were siblings of Syt-1 cKO^DAmice with two wild type Syt-1 alleles and a heterozygote DAT^IRES-Cre allele, except for KCl puffing experiments in Figure 2F–I and electrical stimulation experiments in Figure 1—figure supplement 4. In these experiments, Syt-1 controls were either heterozygote for the Syt-1 floxed allele and for DAT^IRES-Cre, or homozygote for the Syt-1 floxed allele without a DAT^IRES-Cre allele. Mice were group housed in a 12 hr light-dark cycle with free access to water and food, and experiments were done in male and female mice. All animal experiments were done in accordance with approved protocols of the Harvard University Animal Care and Use Committee.

Production of AAV viruses and stereotaxic surgeries

In Syt-1 control and Syt-1 cKO^DA mice striatal dopamine fibers were activated after transduction of dopamine neurons with AAVs for Cre-dependent expression of oChIEF-citrine (Lin et al., 2009), a fast channelrhodopsin variant (p867, RRID:Addgene_50973). AAVs (serotype AAV2/5) were generated in HEK293T cells (purchased as identified, mycoplasma free cell line from ATCC, Cat#: CRL-3216, RRID:CVCL_0063) using calcium phosphate transfection. 72 hr after transfection, cells were collected, lysed, and viral particles were extracted and purified from the 40% layer after iodixanol gradient ultracentrifugation. Quantitative rtPCR was used to measure the genomic titer (2.31–2.75 × 10¹² genome copies/ml). For stereotaxic surgeries, mice were anesthetized using 5% isoflurane and mounted on a stereotaxic frame. 1.5–2% isoflurane was used to maintain a stable anesthesia during the surgery. 1 μl of viral solution was injected unilaterally into the right substantia nigra pars compacta (SNc – 0.6 mm anterior, 1.3 mm lateral of Lambda and 4.2 mm below pia) of Syt-1 control and Syt-1 cKO^DA mice at P25-29 using a microinjector (PHD ULTRA syringe pump, Harvard Apparatus) at the rate of 100 nl/min. After surgery, the mice obtained analgesia and were allowed to recover for at least 21 d prior to recording. Stereotaxic surgeries were performed according to protocols approved by the Harvard University Animal Care and Use Committee.

Electrophysiological recordings

Male and female mice (42–113 days old) were deeply anesthetized with isoflurane and decapitated. 250 µm thick sagittal brain sections containing the striatum were cut using a vibratome (Leica, VT1200s) in ice-cold cutting solution with (in mM): 75 NaCl, 75 sucrose, 2.5 KCl, 7.5 MgSO₄, 26.2 NaHCO₃, 1 NaH₂PO₄, 12 glucose, 1 sodium ascorbate, 1 myo-inositol, 3 sodium pyruvate, (pH 7.4, 300–310 mOsm). Slices were incubated at room temperature for 1 hr in incubation solution bubbled with 95% O₂ and 5% CO₂ containing (in mM): 126 NaCl, 2.5 KCl, 1.3 MgSO₄, 2 CaCl₂, 26.2 NaHCO₃, 1 NaH₂PO₄, 12 glucose, 1 sodium ascorbate, 1 myo-inositol, 3 sodium pyruvate (pH 7.4, 305–310 mOsm). Recording was done at 34–36°C, and slices were continuously perfused with artificial cerebrospinal fluid (ACSF) at 3–4 ml/min bubbled with 95% O₂ and 5% CO₂. ACSF contained (in mM): 126 NaCl, 2.5 KCl, 2 CaCl₂ (unless noted otherwise), 1.3 MgSO₄, 1 NaH₂PO₄, 12 glucose, 26.2 NaHCO₃, pH 7.4, 300–310 mOsm. Recordings were completed within 5 hr of slicing. In Figure 3A–E, 10 µM nomifensine (Tocris, Catalogue No.#1992) was applied to block the dopamine transporter (DAT). For all genotype comparisons, each littermate pair was recorded on the same day with interleafed recordings, and the experimenter was blind to genotype throughout recording and data analyses. All data acquisition and analyses for electrophysiology was done using pClamp10 (Clampex, Axon Instruments).

For all carbon fiber amperometry, carbon fiber microelectrodes (CFEs, 7 µm diameter, 100–150 µm long) were made from carbon fiber filaments (Goodfellow). Each CFE was calibrated by puffing freshly made dopamine solutions of increasing concentrations (0, 1, 5, 10, 20 µM) in ACSF for 10 s (Figure 1—figure supplement 1A). The currents for each concentration of dopamine were plotted against the dopamine concentration and only CFEs with a linear relationship were used. On each day, a new CFE was calibrated and dopamine release was measured with the same CFE for a Syt-1 control and Syt-1 cKO^DA littermate pair. CFEs were held at 600 mV and placed 20–60 µm below the slice surface in the dorsolateral striatum. Signals were sampled at 10 kHz and low-pass filtered at 400 Hz. Dopamine release was evoked by electrical or optogenetic stimulation every 2 min.

Optogenetic stimulation was performed with channelrhodopsin expression limited to dopamine neurons by AAV-mediated expression of Cre-dependent oChIEF-citrine (Figure 1F–I, Figure 3A–E, Figure 1—figure supplement 3B–3G) or transgenic mice for Cre-dependent ChR2 expression (Madisen et al., 2012; Figure 1B–D, Figure 1—figure supplement 1C and D). For stimulation, brief 1 ms pulses of 470 nm light were delivered at the recording site in the dorsolateral striatum through a 60 x objective by a light-emitting diode (Cool LED pE4000). Optogenetic stimulation was applied as a single stimulus (Figure 1B–D, Figure 1—figure supplement 1C and D using ChR2, Figure 1F and G using oChIEF) or as stimulus trains (10 stimuli at 10 Hz in Figures 1H, I and 3A–E, Figure 1—figure supplement 3B–3E or 40 stimuli at 40 Hz in Figure 1—figure supplement 3F and G using oChIEF). Optogenetic stimulation was applied every 2 min for all dopamine release measurements, or every 10 s for the field recordings shown in Figure 1—figure supplement 3. Total dopamine release during stimulation trains (Figure 3A–E) was measured as area under the curve from the start of the 1^st stimulus for 2.935 s and expressed as μM x s after removal of the stimulus artefacts. The total increase (Figure 3D) was calculated by subtracting the area (area_nomifensine – area_baseline), and the fold increase (Figure 3E) was calculated by division (area_nomifensine/area_baseline).

Electrical stimulation was performed with an ACSF filled glass pipette (tip diameter 3–5 µm) connected to a linear stimulus isolator (A395, World Precision Instruments) to deliver monopolar electrical stimulation (10–90 µA). The stimulation pipette was placed 20–30 µm below the slice surface in the dorsolateral striatum and 100–120 µm away from the tip of the CFE. A biphasic wave (0.25 µs in each phase) was applied to evoke dopamine release. Electrical stimulation was delivered either as a single stimulus or a 10-stimulus 10 Hz train (Figure 1—figure supplement 4).

For KCl stimulation in Figure 2F–I, KCl solution containing (in mM) 100 KCl, 50 NaCl, 1.3 MgSO₄, 2 CaCl₂, 12 glucose, 10 HEPES, pH 7.3, 300–310 mOsm) was puffed onto the recording site in the dorsolateral striatum for 10 s at 9 µl/s using a syringe pump (World Precision Instruments). Only one KCl puff was applied per slice. The peak amplitude of the dopamine response and the area under the curve were quantified from start of application of KCl to 200 s after the puff, at which time the levels returned to baseline.

For experiments in variable [Ca²⁺]_ex, CFEs were first tested with a 20 µM dopamine puff at different [Ca²⁺]_ex (0.5, 1, 2 and 4 mM) to ensure that the CFE correctly reports [dopamine] across variable amounts of [Ca²⁺]_ex (Figure 1—figure supplement 1B). Slices were either recorded with decreasing [Ca²⁺]_ex (2, 1, and 0.5 mM, Figure 1B–D) or with increasing [Ca²⁺]_ex (2, 4 mM, Figure 1—figure supplement 1C and D) in separate experiments. Extracellular magnesium was adjusted for recordings in variable [Ca²⁺]_ex (solutions contained Ca²⁺/Mg²⁺ in mM: 0.5/2.8; 1/2.3; 2/1.3 and 4/0).

Extracellular recordings in Figure 1—figure supplement 3 were performed with an ACSF filled glass pipette (2–3 µm tip diameter) that was placed 20–60 µm below the slice surface in areas of the dorsolateral striatum with uniform citrine fluorescence. Optogenetic stimulation was applied as a 10 Hz train (Figure 1—figure supplement 3B–3E) or a 40 Hz train (Figure 1—figure supplement 3F and G) every 10 s and 100 sweeps were averaged for quantification (Figure 1—figure supplement 3B–3G). Sodium channels were blocked using 1 µM TTX (Tocris, Catalogue No.# 1078) and extracellular potentials evoked by 10 Hz trains were recorded before and after TTX. To quantify the reduction by TTX, the amplitude evoked by the 1^st stimulus in the 10 Hz train before and after TTX was analyzed in Syt-1 control and Syt-1 cKO^DA mice (Figure 1—figure supplement 3C and D).

Immunostaining of brain sections

Male and female Syt-1 control and cKO^DA littermate mice (99–111 days old) were deeply anesthetized with 5% isoflurane and perfused transcardially with ice-cold 30–50 ml phosphate buffer saline (PBS), followed by 50 ml of 4% paraformaldehyde (PFA) in PBS at 4°C. Brains were then left in 4% PFA for 12–16 hr followed by incubation in 30% sucrose + 0.1% sodium azide in PBS overnight or until they sank to the bottom of the tube. Coronal striatal sections (20 µm thick) were cut using a vibratome (Leica, VT1000s) in ice-cold PBS. Antigen retrieval was performed on slices overnight at 60⁰C in 150 mM NaCl, 1 mM EDTA, 0.05% Tween 20, 10 mM Tris Base, pH 9.0. After antigen retrieval, slices were washed in PBS, and incubated in Image-iT FX signal enhancer (Invitrogen, I36933) for 30 min at room temperature. Slices were washed in PBS for 10 min and non-specific binding was blocked in 10% goat serum in 0.25% Triton X-100 in PBS (PBST) for 1 hr at room temperature. Slices were stained with primary antibodies for 12 hr at 4°C, and the following primary antibodies were used: mouse monoclonal IgG2a anti-Bassoon (1:500, A85, RRID:AB_11181058) and guinea pig polyclonal anti-TH (1:1000, A111, RRID:AB_2619897). Next, slices were washed three times in PBST for 10 min and incubated in secondary antibodies (1:500, goat anti-mouse IgG2a Alexa 488, S8, RRID:AB_2535771, and goat-anti guinea pig Alexa 568, S27, RRID:AB_2534119) for 2 hr at room temperature in PBST. Sections were washed three times in PBST for 10 min and then mounted on Poly-D-lysine coated #1.5 cover glasses (GG-18–1.5-pdl, neuVitro) with H-1000 mounting medium (Vectashield). At all times, the experimenter was blind to the genotype of the mice.

3D-SIM image acquisition and analysis

Image acquisition and analyses were done essentially as described before (Liu et al., 2018) using a DeltaVision OMX V4 Blaze structured illumination microscope (GE Healthcare) with a 60 x, 1.42 N.A. oil immersion objective and Edge 5.5 sCMOS cameras (PCO) for each channel. Z stacks were acquired with 125 nm step size and 15 raw images per plane (five phases, three angles). Immersion oil matching was used to minimize spherical aberration. Lateral shift between green and red channels was measured using a control slide to generate a calibration image and all images were reconstructed using this calibration to remove lateral shifts. All raw images were aligned and reconstructed to obtain superresolved images using the image registration function in softWoRx. Image volumes (40 × 40 × 6 µm³) were acquired from 7 to 8 regions within the dorsolateral striatum in each section. For image analysis, regions of interest (ROIs) ranging from 20 × 20 × 2.5 µm³ to 25 × 25 × 2.5 µm³ were selected manually in each image stack. ROIs were analyzed to characterize TH and Bassoon signals and to determine their overlap using a custom written MATLAB code (Liu et al., 2018) (available at https://github.com/hmslcl/3D_SIM_analysis_HMS_Kaeser-lab_CL; Liu, 2017). Briefly, intensity thresholding using Otsu and size thresholding (0.04–20 µm³ for TH axons, 0.003–0.04 µm³ for Bassoon) were applied to each ROI. The volume occupied by TH was quantified and was divided by the total image volume (Figure 2B). The TH signals were skeletonized to determine TH axon length by 3D Gaussian filtering and a homotypic thinning algorithm. TH axon length was divided by the total image volume in Syt-1 control and Syt-1 cKO^DA mice (Figure 2C). The volume of Bassoon clusters and percentage overlap of Bassoon with TH was calculated, and >40% overlap of Bassoon with TH was considered to be a positive association (Liu et al., 2018; Figure 2D and E). For generating controls using local shuffling, each Bassoon object was randomly shuffled within 1 × 1 × 1 µm³ for 1000 rounds of shuffling and for each ROI, and the overlap between the average of shuffled Bassoon and TH was calculated. Sample images in Figure 2A were generated using Imaris 9.0.2 (Oxford Instruments) from masked images of either ‘Bassoon + TH’ or ‘Bassoon within TH’ derived from the custom written MATLAB code. Adjustments of contrast, intensity and surface rendering were done identically for each condition for illustration, but after quantification. For all 3D-SIM data acquisition and analyses, the experimenter was blind to the genotype of the mice.

Striatal synaptosome preparation and immunostaining

Striatal synaptosome preparations were performed as previously described (Liu et al., 2018). Wild type mice (P21-70) were deeply anesthetized, decapitated, and brains were harvested into ice-cold PBS. Dorsal striata were dissected and placed into a pre-cooled detergent-free glass tube and 1 ml of ice-cold homogenizing buffer containing (in mM): 4 4-(2-hydroxyethyl)−1-piperazineethanesulfonic acid (HEPES), 320 sucrose, pH 7.4, and 1x of a mammalian protease inhibitor cocktail was added. A detergent-free ice-cold glass-teflon homogenizer was used to homogenize the tissue using 12 strokes. The striatal homogenate was added to 1 ml of homogenizing buffer and centrifuged at 1,000 g for 10 min at 4°C. The supernatant (S1) was collected and centrifuged at 12,500 g for 15 min at 4°C. The supernatant (S2) was removed and the pellet (P2) was re-homogenized in 1 ml homogenizing buffer with six strokes. A sucrose density gradient was prepared with 5 ml of both 0.8 M and 1.2 M sucrose in thin wall ultracentrifugation tubes (Beckman Coulter, Cat # 344059). The P2 homogenate was mixed with 1 ml of homogenizing buffer, and 1.5 ml was added to the top of the sucrose gradient and was centrifuged at 69,150 x g for 70 min at 4°C (SW 41 Ti Swinging-Bucket Rotor, Beckman Coulter, Cat. # 331362). The synaptosome layer (1–1.5 ml) was collected from the interface of the two sucrose layers. Synaptosomes were then diluted 20–30 times in homogenizing buffer and spun (4000 x g, 10 min) onto Poly-D-lysine coated #1.5 coverslips at 4°C. Excess homogenizing buffer was pipetted out and synaptosomes were fixed using 4% PFA in PBS for 20 min at 4⁰C. Non-specific binding block and permeabilization was done in 3% bovine serum albumin + 0.1% Triton X-100 in PBS at room temperature for 45 mins. Primary antibody staining was done for 12 hr at 4°C, followed by three washes for 15 mins each. Secondary antibody staining was done for 2 hr at room temperature in blocking solution followed by three washes each for 15 mins. The primary antibodies used were: mouse monoclonal IgG2a anti-Bassoon (1:1000, A85, RRID:AB_11181058), guinea pig polyclonal anti-TH (1:1000, A111, RRID:AB_2619897), and rabbit polyclonal anti-Synaptotagmin-1 antiserum (1:1000, A24, V216, a gift from Dr. T.C. Südhof). The secondary antibodies were: goat anti-mouse IgG2a Alexa 488 (1:500, S8, RRID:AB_2535771), goat anti-rabbit Alexa 555 (1:500, S22, RRID:AB_2535849), and goat anti-guinea pig Alexa 633 (1:500, S34, RRID:AB_2535757).

Confocal microscopy and image analysis of striatal synaptosomes

Single optical sections of synaptosomes (105 × 105 µm²) stained for Bassoon (detected via Alexa 488), Synaptotagmin-1 (detected via Alexa-555) and TH (detected via Alexa 633) were imaged with an oil immersion 60 x objective and 1.5 x optical zoom using an Olympus FV1000 confocal microscope. For quantification, images were processed for background subtraction in Fiji using the ‘rolling ball’ algorithm with a radius of 1 µm for each channel. Each background subtracted image was analyzed in a custom MATLAB program (available at https://github.com/hmslcl/3D_SIM_analysis_HMS_Kaeser-lab_CL; Liu, 2017). Otsu intensity thresholds, size thresholds (0.2–1 µm²) and shape thresholds (ratio of x to y axis <1.5) were applied for object detection. These threshold settings were identical for each image and allowed for unbiased and automated detection of Bassoon-positive (Bassoon⁺), TH-positive (TH⁺) and Synaptotagmin-1-positive (Syt-1⁺) ROIs in each image, with a total of 300–600 synaptosome objects detected per image. These ROIs were then used to generate the Bassoon⁺TH⁺, Bassoon^-TH⁺, Bassoon⁺TH^- ROIs displayed in Figure 1—figure supplement 2C. The extent of overlap of Synaptotagmin-1⁺ ROIs with the various Bassoon/TH ROIs was quantified and a 20–100% overlap criterion was applied to define Synaptotagmin-1 positivity for the various Bassoon/TH ROIs. Sample images in Figure 1—figure supplement 2B were generated in Fiji with adjustments of brightness and contrast.

Microdialysis

Microdialysis was performed according to previously established methods (Liu et al., 2018). The probes (6 kDa MW cut-off, CMA 11, Harvard Apparatus, Catalogue# CMA8309581) were calibrated with freshly made dopamine solutions (0, 4 and 8 µM) dissolved in ACSF before each experiment. A fresh probe was used for each mouse. The microdialysis probe was continuously perfused with ACSF containing (in mM): 155 NaCl, 1.2 MgCl_2, 2.5 KCl, 1.2 CaCl₂, and 5 glucose at a speed of 1 µl/min. After probe calibration, the probe was inserted into dorsal striatum (coordinates: 1.0 mm anterior, 2.0 mm lateral of bregma, and 3.3 mm below pia) of anesthetized male and female mice using stereotaxy (73–103 days old). Striatal dialysates were collected every 15 min and the concentration of dopamine was measured using an HPLC (HTEC-510, Amuza Inc) connected to an electrochemical detector (Eicom). The data during the first 75 min were not plotted because during this time window dopamine levels stabilize after surgery. Average dopamine levels from the 76^th - 120^th min of Syt-1 control mice were used to normalize all dopamine values. 10 µM TTX dissolved in ACSF was applied using reverse dialysis starting at 121 min to inhibit firing of dopamine axons as described before (Liu et al., 2018). For all microdialysis data acquisition and analyses, the experimenter was blind to the genotype of the mice.

Statistical analyses

Data are expressed as mean ± SEM. All statistical analyses were performed in Graphpad Prism. Student’s unpaired t-tests were used in Figure 2B and C, paired t-test in Figure 1—figure supplement 1D, Wilcoxon tests in Figure 3B and C, Mann Whitney tests in Figures 1G, 2H, I, 3D and E, one-way ANOVA followed by Dunnett’s multiple comparisons in Figure 1C and D and Figure 1—figure supplement 2C, one-way ANOVA followed by Sidak’s multiple comparisons tests in Figure 2D and E and Figure 1—figure supplement 3D, two-way ANOVA followed by Sidak’s multiple comparisons tests in Figures 1I and 3G, Figure 1—figure supplement 3E and Figure 1—figure supplement 4C, and two-way ANOVA mixed-effects analysis followed by Sidak’s multiple comparisons tests in Figure 1—figure supplement 3G and Figure 1—figure supplement 4E. For all genotype comparisons, the experimenter was blind to genotype during data acquisition and analyses.

Acknowledgements

This work was supported by the National Institutes of Health (R01NS103484 to PSK), the Dean’s Initiative Award for Innovation (to PSK), a Harvard-MIT Joint Research Grant (to PSK), William Randolph Hearst (to A B), Alice Joseph Brooks (to A B) and Gordon family (to CL) postdoctoral fellowships, and a Marshall Plan Foundation fellowship (to PN). We thank J Wang for technical assistance, L Kershberg for help with setting up synaptosome preparations, Drs. K Balakrishnan and W Regehr for insightful discussions, Dr. TC Südhof for providing the conditional Syt-1 knockout mice, and the Cell Biology Microscopy Facility and the Neurobiology Imaging Facility (supported by a NINDS P30 Core Center grant, NS072030) for availability of microscopes and advice.

Additional information

Competing interests

No competing interests declared

Author contributions

Conceptualization, Formal analysis, Investigation, Visualization, Methodology, Writing - original draft, Writing - review and editing

Formal analysis, Investigation, Visualization, Writing - review and editing

Formal analysis, Investigation, Writing - review and editing

Resources, Methodology, Writing - review and editing

Conceptualization, Formal analysis, Supervision, Funding acquisition, Writing - original draft, Project administration, Writing - review and editing

Ethics

Animal experimentation: All animal experiments were performed according to institutional guidelines of Harvard University, and were in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The animals were handled according to protocols (protocol number IS00000049) approved by the institutional animal care and use committee (IACUC).

Additional files

Transparent reporting form

Data availability

All data generated in the study are included in the figures, including individual data points whenever possible.

References

Agnati

Zoli

Strömberg

Fuxe

1995

Intercellular communication in the brain: wiring versus volume transmission

Neuroscience69711726

10.1016/0306-4522(95)00308-6

8596642

Bacaj

Yang

Morishita

Zhou

Malenka

Südhof

2013

Synaptotagmin-1 and synaptotagmin-7 trigger synchronous and asynchronous phases of neurotransmitter release

Neuron80947959

10.1016/j.neuron.2013.10.026

24267651

Bacaj

Burré

Malenka

Liu

Südhof

2015

Synaptotagmin-1 and -7 are redundantly essential for maintaining the capacity of the Readily-Releasable pool of synaptic vesicles

PLOS Biology13

e1002267

10.1371/journal.pbio.1002267

26437117

Bäckman

Malik

Zhang

Shan

Grinberg

Hoffer

Westphal

Tomac

2006

Characterization of a mouse strain expressing cre recombinase from the 3' untranslated region of the dopamine transporter locus

Genesis44383390

10.1002/dvg.20228

16865686

Beckstead

Grandy

Wickman

Williams

2004

Vesicular dopamine release elicits an inhibitory postsynaptic current in midbrain dopamine neurons

Neuron42939946

10.1016/j.neuron.2004.05.019

15207238

Benoit-Marand

Borrelli

Gonon

2001

Inhibition of dopamine release via presynaptic D2 receptors: time course and functional characteristics in vivo

The Journal of Neuroscience2191349141

10.1523/JNEUROSCI.21-23-09134.2001

11717346

Brimblecombe

Gracie

Platt

Cragg

2015

Gating of dopamine transmission by calcium and axonal N-, Q-, T- and L-type voltage-gated calcium channels differs between striatal domains

The Journal of Physiology593929946

10.1113/jphysiol.2014.285890

25533038

Broadie

Bellen

DiAntonio

Littleton

Schwarz

1994

Absence of synaptotagmin disrupts excitation-secretion coupling during synaptic transmission

PNAS911072710731

10.1073/pnas.91.22.10727

Dai

Shin

Machius

Tomchick

Südhof

Rizo

2004

Structural basis for the evolutionary inactivation of Ca2+ binding to synaptotagmin 4

Nature Structural & Molecular Biology11844849

10.1038/nsmb817

15311271

de Wit

Walter

Milosevic

Gulyás-Kovács

Riedel

Sørensen

Verhage

2009

Synaptotagmin-1 docks secretory vesicles to syntaxin-1/SNAP-25 acceptor complexes

Cell138935946

10.1016/j.cell.2009.07.027

19716167

Dean

Liu

Dunning

Chang

Jackson

Chapman

2009

Synaptotagmin-IV modulates synaptic function and long-term potentiation by regulating BDNF release

Nature Neuroscience12767776

10.1038/nn.2315

19448629

Fernández-Chacón

Königstorfer

Gerber

García

Matos

Stevens

Brose

Rizo

Rosenmund

Südhof

2001

Synaptotagmin I functions as a calcium regulator of release probability

Nature4104149

10.1038/35065004

11242035

Ford

Gantz

Phillips

Williams

2010

Control of extracellular dopamine at dendrite and axon terminals

Journal of Neuroscience3069756983

10.1523/JNEUROSCI.1020-10.2010

20484639

Ford

2014

The role of D2-autoreceptors in regulating dopamine neuron activity and transmission

Neuroscience2821322

10.1016/j.neuroscience.2014.01.025

24463000

Gustafsson

MGL

Shao

Carlton

Wang

CJR

Golubovskaya

Cande

Agard

Sedat

2008

Three-Dimensional Resolution Doubling in Wide-Field Fluorescence Microscopy by Structured Illumination

Biophysical Journal9449574970

10.1529/biophysj.107.120345

Howe

Dombeck

2016

Rapid signalling in distinct dopaminergic axons during locomotion and reward

Nature535505510

10.1038/nature18942

27398617

Kaeser

Regehr

2014

Molecular mechanisms for synchronous, asynchronous, and spontaneous neurotransmitter release

Annual Review of Physiology76333363

10.1146/annurev-physiol-021113-170338

24274737

Lein

Hawrylycz

Ayres

Bensinger

Bernard

Boe

Boguski

Brockway

Byrnes

Chen

Chin

Chong

Crook

Czaplinska

Dang

Datta

Dee

Desaki

Desta

Diep

Dolbeare

Donelan

Dong

Dougherty

Duncan

Ebbert

Eichele

Estin

Faber

Facer

Fields

Fischer

Fliss

Frensley

Gates

Glattfelder

Halverson

Hart

Hohmann

Howell

Jeung

Johnson

Karr

Kawal

Kidney

Knapik

Kuan

Lake

Laramee

Larsen

Lau

Lemon

Liang

Liu

Luong

Michaels

Morgan

Mortrud

Mosqueda

Orta

Overly

Pak

Parry

Pathak

Pearson

Puchalski

Riley

Rockett

Rowland

Royall

Ruiz

Sarno

Schaffnit

Shapovalova

Sivisay

Slaughterbeck

Smith

Sodt

Stewart

Stumpf

Sunkin

Sutram

Tam

Teemer

Thaller

Thompson

Varnam

Visel

Whitlock

Wohnoutka

Wolkey

Wong

Wood

Yaylaoglu

Young

Youngstrom

Yuan

Zhang

Zwingman

Jones

2007

Genome-wide atlas of gene expression in the adult mouse brain

Nature445168176

10.1038/nature05453

17151600

Lin

Steinbach

Tsien

2009

Characterization of engineered channelrhodopsin variants with improved properties and kinetics

Biophysical Journal9618031814

10.1016/j.bpj.2008.11.034

19254539

Liu

2017

3D_SIM_analysis_HMS_Kaeser-lab_CL

GitHub3

https://github.com/hmslcl/3D_SIM_analysis_HMS_Kaeser-lab_CL

Liu

Kershberg

Wang

Schneeberger

Kaeser

2018

Dopamine secretion is mediated by sparse active Zone-like release sites

Cell172706718

10.1016/j.cell.2018.01.008

29398114

Liu

Kaeser

2019

Mechanisms and regulation of dopamine release

Current Opinion in Neurobiology574653

10.1016/j.conb.2019.01.001

30769276

Madisen

Mao

Koch

Zhuo

Berenyi

Fujisawa

Hsu

Garcia

Zanella

Kidney

Mao

Hooks

Boyden

Buzsáki

Ramirez

Jones

Svoboda

Han

Turner

Zeng

2012

A toolbox of Cre-dependent optogenetic transgenic mice for light-induced activation and silencing

Nature Neuroscience15793802

10.1038/nn.3078

22446880

Mendez

Bourque

Fasano

Kortleven

Trudeau

2011

Somatodendritic dopamine release requires synaptotagmin 4 and 7 and the participation of voltage-gated calcium channels

Journal of Biological Chemistry2862392823937

10.1074/jbc.M111.218032

21576241

Menegas

Akiti

Amo

Uchida

Watabe-Uchida

2018

Dopamine neurons projecting to the posterior striatum reinforce avoidance of threatening stimuli

Nature Neuroscience2114211430

10.1038/s41593-018-0222-1

30177795

Michalski

Goutman

Auclair

Boutet de Monvel

Tertrais

Emptoz

Parrin

Nouaille

Guillon

Sachse

Ciric

Bahloul

Hardelin

Sutton

Avan

Krishnakumar

Rothman

Dulon

Safieddine

Petit

2017

Otoferlin acts as a Ca²⁺ sensor for vesicle fusion and vesicle pool replenishment at auditory hair cell ribbon synapses

eLife6

e31013

10.7554/eLife.31013

29111973

Pang

Südhof

2010

Cell biology of Ca2+-triggered exocytosis

Current Opinion in Cell Biology22496505

10.1016/j.ceb.2010.05.001

20561775

Roux

Safieddine

Nouvian

Grati

Simmler

Bahloul

Perfettini

Le Gall

Rostaing

Hamard

Triller

Avan

Moser

Petit

2006

Otoferlin, defective in a human deafness form, is essential for exocytosis at the auditory ribbon synapse

Cell127277289

10.1016/j.cell.2006.08.040

17055430

Saunders

Macosko

Wysoker

Goldman

Krienen

de Rivera

Bien

Baum

Bortolin

Wang

Goeva

Nemesh

Kamitaki

Brumbaugh

Kulp

McCarroll

2018

Molecular diversity and specializations among the cells of the adult mouse brain

Cell17410151030

10.1016/j.cell.2018.07.028

30096299

Schindelin

Arganda-Carreras

Frise

Kaynig

Longair

Pietzsch

Preibisch

Rueden

Saalfeld

Schmid

Tinevez

White

Hartenstein

Eliceiri

Tomancak

Cardona

2012

Fiji: an open-source platform for biological-image analysis

Nature Methods9676682

10.1038/nmeth.2019

22743772

Schonn

Maximov

Lao

Südhof

Sørensen

2008

Synaptotagmin-1 and -7 are functionally overlapping Ca2+ sensors for exocytosis in adrenal chromaffin cells

PNAS10539984003

10.1073/pnas.0712373105

18308932

Skarnes

Rosen

West

Koutsourakis

Bushell

Iyer

Mujica

Thomas

Harrow

Cox

Jackson

Severin

Biggs

Nefedov

de Jong

Stewart

Bradley

2011

A conditional knockout resource for the genome-wide study of mouse gene function

Nature474337342

10.1038/nature10163

21677750

Sulzer

Cragg

Rice

2016

Striatal dopamine neurotransmission: regulation of release and uptake

Basal Ganglia6123148

10.1016/j.baga.2016.02.001

27141430

Sun

Pang

Qin

Fahim

Adachi

Südhof

2007

A dual-Ca2+-sensor model for neurotransmitter release in a central synapse

Nature450676682

10.1038/nature06308

18046404

Threlfell

Lalic

Platt

Jennings

Deisseroth

Cragg

2012

Striatal dopamine release is triggered by synchronized activity in cholinergic interneurons

Neuron755864

10.1016/j.neuron.2012.04.038

22794260

Turecek

Regehr

2019

Neuronal regulation of fast synaptotagmin isoforms controls the relative contributions of synchronous and asynchronous release

Neuron101938949

10.1016/j.neuron.2019.01.013

30733150

Voets

Moser

Lund

Chow

Geppert

Südhof

Neher

2001

Intracellular calcium dependence of large dense-core vesicle exocytosis in the absence of synaptotagmin I

PNAS981168011685

10.1073/pnas.201398798

11562488

Wang

Chapman

2010

Rat and Drosophila synaptotagmin 4 have opposite effects during SNARE-catalyzed membrane fusion

Journal of Biological Chemistry2853075930766

10.1074/jbc.M110.137745

20688915

Wen

Linhoff

McGinley

Corson

Mandel

Brehm

2010

Distinct roles for two synaptotagmin isoforms in synchronous and asynchronous transmitter release at zebrafish neuromuscular junction

PNAS1071390613911

10.1073/pnas.1008598107

20643933

Mashimo

Südhof

2007

Synaptotagmin-1, -2, and -9: ca(2+) sensors for fast release that specify distinct presynaptic properties in subsets of neurons

Neuron54567581

10.1016/j.neuron.2007.05.004

17521570

Pang

Shin

Südhof

2009

Synaptotagmin-1 functions as a Ca2+ sensor for spontaneous release

Nature Neuroscience12759766

10.1038/nn.2320

19412166

Yagishita

Hayashi-Takagi

Ellis-Davies

Urakubo

Ishii

Kasai

2014

A critical time window for dopamine actions on the structural plasticity of dendritic spines

Science34516161620

10.1126/science.1255514

25258080

Yao

Gaffaney

Kwon

Chapman

2011

Doc2 is a Ca2+ sensor required for asynchronous neurotransmitter release

Cell147666677

10.1016/j.cell.2011.09.046

22036572

Zhou

Liang

Dani

2001

Endogenous nicotinic cholinergic activity regulates dopamine release in the striatum

Nature Neuroscience412241229

10.1038/nn769

11713470

Zhou

Lai

Bacaj

Zhao

Lyubimov

Uervirojnangkoorn

Zeldin

Brewster

Sauter

Cohen

Soltis

Alonso-Mori

Chollet

Lemke

Pfuetzner

Choi

Weis

Diao

Südhof

Brunger

2015

Architecture of the synaptotagmin-SNARE machinery for neuronal exocytosis

Nature5256267

10.1038/nature14975

26280336

10.7554/eLife.58359.sa1

Decision letter

Chen

Reviewing EditorStanford UniversityUnited States

Pang

Zhiping

ReviewerCastillo

Pablo E

ReviewerAlbert Einstein College of MedicineUnited States

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Acceptance summary:

Using unequivocal mouse genetics, elegant electrophysiology, optogenetics, carbon fiber amperometry, high resolution imaging and in vivo microdialysis, your work convincingly established synaptotagmin-1 as the Ca²⁺ sensor for rapid synchronous synaptic release of dopamine in striatal neurons. It has been a pleasure to read about and handle the review process for your work.

Decision letter after peer review:

Thank you for submitting your article "Synaptotagmin-1 is the Ca²⁺ sensor for fast striatal dopamine release" for consideration by eLife. Your article has been reviewed by three peer reviewers, and the evaluation has been overseen by a Reviewing Editor and Gary Westbrook as the Senior Editor. The following individuals involved in review of your submission have agreed to reveal their identity: Zhiping Pang (Reviewer #2); Pablo Castillo (Reviewer #3).

The reviewers have discussed the reviews with one another, and the Editors drafted this decision to help you prepare a revised submission in which we ask for some minor revisions. We would like to draw your attention to changes in our revision policy that we have made in response to COVID-19 (https://elifesciences.org/articles/57162). Specifically, we ask editors to accept without delay manuscripts, like yours, that they judge can stand as eLife papers without additional data. Thus the revisions requested below only address clarity and presentation.

Summary:

Banerjee et al. aimed to determine the Ca²⁺ sensor for fast dopamine release in striatal neurons. Despite the obvious importance of dopamine, the molecular machinery that governs release of dopamine from nerve terminals remains enigmatic. The authors used interdisciplinary techniques including unequivocal mouse genetics, elegant electrophysiology, optogenetics, carbon fiber amperometry, high resolution imaging and in vivo microdialysis to show that synaptotagmin-1 acts as a Ca²⁺ sensor for rapid exocytosis of dopaminergic synaptic vesicles in striatal neurons. Syt-1 deletion completely abolished fast synchronous release of dopamine. These terminals were capable of Syt-1-independent asynchronous release that accounted for roughly two thirds of extracellular dopamine levels. Overall, this is a straightforward study with convincing evidence establishing Syt-1 as the main Ca²⁺ sensor for fast dopamine release in striatal neurons. All three reviewers were enthusiastic about this work and had only the following minor concerns. The editors will assess your response at the time of the revision.

1) How does Syt-1 deletion affect miniature synaptic responses in these neurons?

2) How does external Ca concentration affect asynchronous dopamine release? Recordings in the Syt-1 DAT- cKO with dopamine re-uptake blocker provides an unique opportunity to explore this a bit further.

3) The authors mention that Syt-7 levels are also high in midbrain dopaminergic neurons. Could Syt-7 be the Ca²⁺ sensor for asynchronous dopamine release?

4) Puffing of 100mM KCl in Figure 2 F-H. The authors showed that puffing high concentrations of KCl induced similar amount of dopamine release in dopamine neuronal conditional Synaptotagmin -1 knockout and in control. Can the authors elaborate on the nature of dopamine release induced by high KCl^-mediated depolarization. See also point 9.

5) For each manipulation of extracellular Ca²⁺, was extracellular Mg²⁺ appropriately adjusted?

6) Regarding the Syt-1 cKO DA slices, the traces in Figures 1H and 3A and the normalization in 1I indicate a slow accumulation of dopamine over time as a result of optical stimulation. The Syt-1 cKO DA dopamine traces and normalization yielded from electrical stimulation (Figure 1—figure supplement 4 D,E) show a much more stable dopamine level closer to 0. What is the potential explanation for this discrepancy?

7) Figure 3C: The legend indicates that these measurements approximate total dopamine accumulated by integrating the dopamine curves between 0 and 2.9 s following the 1st stimulus. It is therefore unclear why the baseline DA level in Syt-1 cKO DA slices is exactly at 0 when the traces in Figures 1H and 3A, as well as the normalization in 1I, show a slight accumulation of dopamine over time in these slices.

8) Figure 3G: In the pre-TTX condition, the average dopamine level in Syt-1 cKO DA animals is significantly smaller than that in Syt-1 control animals at only 1 of 3 time points. Can the author provide an explanation?

9) The authors' explanation for how KCl but not ChR2 is able to cause DA release in Syt-1 cKO DA slices is not clear. Is this high KCl^-induced release calcium-dependent?

10) The third paragraph of the main section seems to have been misplaced. The information at the start of the paragraph would help clarify rationale if presented sooner.

10.7554/eLife.58359.sa2

Author response

All three reviewers were enthusiastic about this work and had only the following minor concerns. The editors will assess your response at the time of the revision.

1) How does Syt-1 deletion affect miniature synaptic responses in these neurons?

This is an important question in at least two ways. First, it has been shown that for conventional synaptic vesicle exocytosis, knockout of the fast Ca²⁺ sensor often increases miniature release (for key examples, see Broadie et al., 1994; Xu et al., 2009). Second, and as we discuss in the paper, there appears to be a significant amount of extracellular dopamine in the extracellular space even after removal of action-potential triggered dopamine release. This is true for knockouts for synaptotagmin, for knockouts of RIM, and for wild type mice after reverse dialysis of TTX (Figures 1 and 3 in this manuscript and Liu et al., 2018). In microdialysis experiments in the striatum, however, there is no robust increase in extracellular dopamine in Syt1 cKO^DA mice after TTX, suggesting that miniature release is not enhanced, or that dopamine clearance is efficient enough to counteract such an increase. Unfortunately, there is currently no reliable direct measurement for miniature dopamine release in the striatum. Notably, in the midbrain, individual spontaneous events can be detected via measuring GIRK-IPSCs (Gantz et al., 2013) and these events are unaffected after knockout of RIM (Robinson et al., 2019).

We have revised the manuscript to better explain what is known about miniature release in the dopamine system. We now state in the manuscript: “It is not known whether synaptotagmin-1 knockout affects spontaneous dopamine release, but literature from conventional synapses establishes that synaptotagmin-1 knockout strongly enhances miniature synaptic vesicle release (Broadie et al., 1994; Xu et al., 2009). Hence, it is possible that Syt-1 cKO^DA leads to increased spontaneous dopamine release in the striatum. But the observation that extracellular dopamine levels after TTX are not strongly increased suggests that there is no dramatic enhancement of miniature dopamine release, or that an enhancement is counteracted by dopamine clearance.”

We thank the reviewers for bringing up the [Ca²⁺ ]_ex dependence of asynchronous dopamine release, and realize that we did not properly introduce the topic of asynchronous release in the text. While there is a large body of literature that establishes complex but positive correlations between [Ca²⁺ ]_ex and asynchronous release (reviewed in Kaeser and Regehr, 2014), we currently do not know what the [Ca²⁺ ]_ex dependence of asynchronous dopamine release is. The strong prediction from conventional synapses is that asynchronous dopamine release is positively correlated with [Ca²⁺ ]_ex, but characterization beyond this point will require better measurements than accumulation during stimulus trains upon DAT blockade. We hope that recent developments in measurements of dopamine release will allow in the future to detect individual secretory events and their timing relative to action potential firing during trains. To better introduce asynchronous release, we added the following statement: “These additional sensors may for example mediate asynchronous release, a form of vesicular exocytosis at synapses that is triggered with a longer, variable delay in response to action potentials and Ca²⁺ entry (Kaeser and Regehr, 2014; Pang and Sudhof, 2010).”

3) The authors mention that Syt-7 levels are also high in midbrain dopaminergic neurons. Could Syt-7 be the Ca²⁺ sensor for asynchronous dopamine release?

We agree with the reviewers that, in the long-term, it will be important to know which other Ca²⁺ sensors contribute to dopamine release. Synaptotagmin-7 is indeed a leading candidate because it is expressed in these neurons. But the properties of asynchronous dopamine release are not understood, and better measurements need to be established before mechanisms of asynchronous release can be studied. Furthermore, while Synaptotagmin-7 is expressed in these neurons, it is uncertain whether Synaptotagmin-7 is present in their axonal arbor in the striatum and previous literature suggests that Synaptotagmin-7 mediates somatodendritic release (Mendez et al., 2011). We now state in the revised manuscript that synaptotagmin-7 is a candidate sensor for asynchronous release based on the literature: “While sensors for asynchronous dopamine release are not known, the presence of Synaptotagmin-7 in substantia nigra dopamine neurons and its role in asynchronous release at fast synapses (Bacaj et al., 2015; Lein et al., 2007; Mendez et al., 2011; Saunders et al., 2018; Wen et al., 2010) makes Synaptotagmin-7 a candidate sensor protein.”

We thank the reviewers for pointing this out and now elaborate in the manuscript on mechanisms of KCl depolarization-induced dopamine release. The KCl puff causes massive depolarization in all neurons in the area of the puff, including dopamine axons and cholinergic interneurons (which stimulate dopamine release via nAChRs on dopamine axons). Most likely, this depolarization triggers Ca²⁺ entry into dopamine axons, which results in dopamine release. While we do not know how Ca²⁺ triggers KCl-mediated dopamine release, we hypothesize that other Ca²⁺ sensors account for it in Syt-1cKO^DA mice. We are confident that KCl triggers dopamine release through vesicular exocytosis at active zone-like release sites, because in mutants in which we knockout RIM from dopamine neurons, KCl-triggered dopamine release is abolished (Liu et al., 2018). We have added the following statement to the paper: “Local puffing of KCl onto brain slices causes a strong depolarization of dopamine axons and surrounding neurons, for example cholinergic interneurons, which triggers massive dopamine release that requires the active zone protein RIM in dopamine axons (Liu et al., 2018).”

Additional explanations that are related to this point are provided further below in response to point 9.

5) For each manipulation of extracellular Ca²⁺, was extracellular Mg²⁺ appropriately adjusted?

Yes, extracellular Mg²⁺ was adjusted accordingly in all experiments. The revised Materials and methods now state: “Extracellular magnesium was adjusted for recordings in variable [Ca²⁺]_ex (solutions contained Ca²⁺/Mg²⁺ in mM: 0.5/2.8; 1/2.3; 2/1.3 and 4/0).”

We note that we have done experiments increasing [Ca²⁺]_ex before without adjusting Mg²⁺ (Liu et al., 2018), and that the increase in dopamine release was similar to what we report here.

The key difference between the two experiments is that in the main figures (Figures 1H and 3A), optogenetic stimulation was used while in the figure supplement (Figure 1—figure supplement 4D and 4E) electrical stimulation was applied. Optogenetic stimulation only recruits dopamine axons and leads to depression. In contrast, electrical stimulation recruits dopamine axons and cholinergic interneurons, and stimulating cholinergic interneurons strongly enhances depression (Liu et al., 2018; Threlfell et al., 2012; Zhou et al., 2001). Hence, the discrepancy is explained by recruitment of the cholinergic mechanism and is consistent with previous papers published on this topic. We now state in the figure legend of Figure 1—figure supplement 4: “Electrical stimulation activates dopamine fibers and cholinergic interneurons, and cholinergic innervation of dopamine axons accounts for as much as ~90% of the extracellular dopamine detected upon electrical stimulation and leads to enhanced depression of dopamine release during stimulus trains (Liu et al., 2018; Threlfell et al., 2012; Zhou et al., 2001).”

We thank the reviewers for bringing this up and would like to note that the baseline DA level in Syt1 cKO^DA was not exactly zero, but the scale of the y-axis was such that it was difficult to see this. We have now split up the graph in Figure 3C such that Syt1 control and Syt1 cKO^DA values are shown on separate scales and the low amount of accumulation at baseline can be visualized. We thank the reviewers for pointing this out.

In our view, the pre-TTX effect of Syt1 cKO^DA in Figure 3G is surprisingly mild given the strong reduction in synchronous release. Because the pre-TTX effect of Syt1 cKO^DA is mild and there is some variability in microdialysis measurements, it is difficult to detect. This is particularly striking because in mutants that lack RIM in dopamine neurons in which action potential triggered release is also absent, pre-TTX dopamine levels in microdialysis are very strongly reduced (Liu et al., 2018).

We describe the effect of synaptotagmin-1 ablation on pre-TTX dopamine in vivo as follows : “Remarkably, extracellular dopamine levels before TTX reverse dialysis were only mildly reduced in Syt-1 cKO^DA…”.

While we don’t have a definitive answer as to why this is the case, we think it is most likely that in vivo, asynchronous release contributes robustly to extracellular dopamine because dopamine levels in microdialysis drop significantly after action potential blockade with TTX in Syt-1 cKO^DA mice, and changes in the clearance of dopamine could also contribute. We also note that there is a good match across different mutants between dopamine in the microdialysate and KCl depolarization induced release: in Syt1 cKO^DA animals, amperometric currents in response to action potentials are strongly reduced, but KCl depolarization and in vivo microdialysis reports significant extracellular dopamine. This is different from dopamine neuron RIM knockouts (RIM cKO^DA), in which KCl triggered release is abolished and pre-TTX extracellular dopamine in microdialysis is strongly reduced (Liu et al., 2018). To summarize these effects, we now state the following: “These data suggest that three modes of dopamine release exist: synchronous and asynchronous release in response to action potentials, and action potential independent release that is likely mediated by spontaneous exocytotic events. Remarkably, each component appears to contribute significantly to the extracellular dopamine measured by microdialysis in anesthetized mice, suggesting that asynchronous and action potential-independent release may be prominent.”

We hope that these statements, together with the explanations above, address this concern.

9) The authors' explanation for how KCl but not ChR2 is able to cause DA release in Syt-1 cKO DA slices is not clear. Is this high KCl^-induced release calcium-dependent?

We thank the reviewers to bring up this important point, and it relates to point 4 as well. This point has two aspects: (1) how channelrhodopsin triggers release, and (2) how KCl triggers release.

1) While it is possible that Ca²⁺ enters through channelrhodopsin directly under some circumstances, this is not the case here in a way that Ca²⁺ is sufficient to trigger dopamine release. The experiments in Syt1 cKO^DA animals are performed with oChiEF, not ChR2, and we have directly tested whether sodium channels are necessary for this form of release. We found that TTX entirely blocks oChiEF induced release (Figures 6A-6C in Liu et al., 2018). Hence, this method triggers release by evoking action potentials in dopamine axons. We now state: “In these mice, we expressed oChiEF-citrine, a fast channelrhodopsin, selectively in dopamine neurons using AAVs (Figure 1A) to optogenetically evoke dopamine release through triggering of axonal action potentials (Liu et al., 2018).”

2) As outlined above, KCl triggers release most likely through at least two mechanisms, massive depolarization of (i) dopamine axons and (ii) of cholinergic interneurons, which in turn release acetylcholine and then trigger dopamine release through an unknown mechanism after activation of dopamine axonal nAChRs. Notably, this form of dopamine release requires the presence of vesicular release machinery, for example RIM, and hence we concluded in a previously published study that it is vesicular (Liu et al., 2018). Because release triggered through nAChRs activation (Figure 1—figure supplement 4) is abolished in Syt1 cKO^DA mice, we conclude that KCl triggers vesicular exocytosis of dopamine through strong depolarization of dopamine axons.

We now specifically describe the mechanisms of KCl mediated release: “Local puffing of KCl onto brain slices causes a strong depolarization of dopamine axons and surrounding neurons, for example cholinergic interneurons, which triggers massive dopamine release that requires the active zone protein RIM in dopamine axons (Liu et al., 2018).”

The best experiment to directly test the requirement for extracellular Ca²⁺ for this form of release would be to remove extracellular Ca²⁺ entirely. For optogenetic experiments, we have tried to fully remove extracellular Ca²⁺ from striatal slices, but have found that this is challenging and we do not have good experiments to evaluate the requirement of extracellular Ca²⁺ for KCl^-triggered dopamine release. Hence, although there is a long-term body of literature to suggest that KCl triggers release through depolarization-dependent opening of voltage gated Ca²⁺ channels followed by flooding of the nerve terminal with Ca²⁺, we do not have the capability to prove this at this point in striatal brain slices. To account for this in the manuscript, we now state: “While the exact mechanism of KCl depolarization induced dopamine release is not known, the data further suggest that depolarization-induced, likely massive Ca²⁺ entry may trigger vesicular dopamine release via one or multiple alternative Ca²⁺ sensors in the absence of Synaptotagmin-1.”

10) The third paragraph of the main section seems to have been misplaced. The information at the start of the paragraph would help clarify rationale if presented sooner.

We thank the reviewers for this stylistic suggestion and have switched the paragraphs as suggested to present some of the rationale sooner.

References

Gantz, S.C., Bunzow, J.R., and Williams, J.T. (2013). Spontaneous Inhibitory Synaptic Currents Mediated by a G Protein-Coupled Receptor. Neuron 78, 807–812.

Robinson, B.G., Cai, X., Wang, J., Bunzow, J.R., Williams, J.T., and Kaeser, P.S. (2019). RIM is essential for stimulated but not spontaneous somatodendritic dopamine release in the midbrain. eLife8.