Abstract
Voluntary movement, motivation, and reinforcement learning depend on the activity of ventral midbrain neurons that extend axons to release dopamine (DA) in the striatum. These neurons exhibit two patterns of action potential activity: a low-frequency tonic activity that is intrinsically generated, and superimposed high-frequency phasic bursts that are driven by synaptic inputs. Ex vivo acute striatal brain preparations are widely employed to study the regulation of evoked DA release but exhibit very different DA release kinetics than in vivo recordings. To investigate the relationship between phasic and tonic neuronal activity, we stimulated the slice in patterns intended to mimic tonic activity, which were interrupted by a series of burst stimuli. Conditioning the striatal slice with low-frequency activity altered DA release triggered by high-frequency bursts, and produced kinetic parameters that resemble those in vivo. In the absence of applied tonic activity, nicotinic acetylcholine receptor and D2 dopamine receptor antagonists had no significant effect on neurotransmitter release driven by repeated burst activity in the striatal brain slice. In contrast, in tonically stimulated slices, D2 receptor blockade decreased the amount of DA released during a single burst and facilitated DA release in subsequent bursts. This experimental system provides a means to reconcile the difference in the kinetics of DA release ex vivo and in vivo and provides a novel approach to more accurately emulate pre- and post-synaptic mechanisms that control axonal DA release in the acute striatal brain slice.
Introduction
Basal ganglia dopaminergic neurons located in the substantia nigra pars compacta (SNpc) and ventral tegmental area of the midbrain send axonal projections to the striatum, where the release of dopamine (DA) plays pivotal roles in reinforcement learning, motivation, and motor control (Graybiel et al., 1994; Lanciego et al., 2012; Salamone, 2007). In vivo, these neurons predominantly exhibit two firing patterns: a tonic self-autonomous single-spike activity at 1-10 Hz (Grace & Bunney, 1984), and phasic burst activity at 13-20 Hz in anesthetized rodents and up to 80 Hz in awake animals or humans (Duda et al., 2016) that is driven by synaptic inputs. Tonic and phasic firing of DA neurons have been suggested to play distinct roles in animal behaviors: tonic neurotransmitter release is thought to be required for motivation and motor control, whereas phasic activity in response to environmental cues encodes reward prediction error and is important for learning (Eshel et al., 2024; Zweifel et al., 2009). Both tonic and phasic activity are regulated by pharmacological agents, including antipsychotic drugs, psychostimulants, and drugs of abuse including opiates, nicotine, and alcohol, as well as stress and pathological conditions. A dysregulation of firing patterns is associated with neurological disorders including Parkinson’s disease, addiction, and schizophrenia (Juárez Olguín et al., 2016).
Many studies have analyzed the kinetics of DA exocytosis in acute striatal slices, providing fundamental insights into the regulation of evoked neurotransmitter release by drugs and local synaptic circuitry and demonstrating regional variations in DA release probability and short-term plasticity within the striatum (Cragg, 2003, 2006; Exley et al., 2008; Wu et al., 2002; H. Zhang & Sulzer, 2003). These ex vivo studies have shown that striatal DA release is modulated by many mechanisms, including changes in vesicular DA content, inhibition of the dopaminergic axons by D2 receptors (D2R) (Kehr et al., 1972; Lindgren et al., 2001; O’Hara et al., 1996; Schmitz et al., 2002), control by locally released acetylcholine (ACh) (Brimblecombe et al., 2018; Sulzer et al., 2016), and regulation of fusion pore kinetics (Staal et al., 2004). Additionally, the dynamic activity-dependent equilibrium between different pools of synaptic vesicles within the release sites – the readily releasable pool (RRP), the recycling pool and the reserve pool – is an important factor in the regulation of DA vesicle release probability and the refilling of recycling synaptic vesicles with neurotransmitter (Somayaji, Cataldi, Choi, et al., 2020; Somayaji, Cataldi, Edwards, et al., 2020; Yavich & MacDonald, 2000). Due to these mechanisms, phasic DA release has been suggested to be inversely dependent on the frequency of tonic activity, due to a larger readily releasable pool of synaptic vesicles and lower D2R-mediated autoinhibition in inactive dopaminergic axons (Eshel et al., 2024; Grace, 1991), although experimental evidence for this hypothesis has not been reported.
While the acute striatal brain slice provides strong experimental advantages due to the ease of pharmacological or optogenetic manipulations of the local circuitry, limitations of this model system are highlighted by very different DA release kinetics in vivo (Shashaank et al., 2023; Somayaji, Cataldi, Choi, et al., 2020). Typically, kinetic studies of electrochemical recordings of evoked DA release in vivo are performed in anesthetized animals with the recording electrode implanted in the striatum and stimulation electrode in the ventral midbrain or the medium forebrain bundle. In contrast to slice recordings where single electric pulse evokes large release of DA (Schmitz et al., 2003; Stamford et al., 1991), a single stimulus pulse is typically insufficient to evoke detectable release of DA in vivo (Chergui et al., 1994; Somayaji, Cataldi, Choi, et al., 2020; Venton et al., 2003). Moreover, consecutive phasic bursts in striatal slices produce a profound release depression that contrasts with a facilitation of DA release observed in vivo (Somayaji, Cataldi, Choi, et al., 2020).
A notable difference between the in vivo and ex vivo recordings is that tonic activity is present in vivo but absent in coronal striatal slices that lack the cell bodies of dopaminergic neurons, where tonic activity is generated. To investigate whether tonic activity underlies the difference between ex vivo and in vivo kinetics of DA release and plasticity, we applied local electrical stimulation to the slice at tonic frequencies. The tonic conditioning in the striatal slice was followed by a series of high frequency bursts to determine changes in release facilitation or depression. Following characterization of these kinetics, we investigated the effect of nicotinic ACh receptor and D2 dopamine receptor antagonists on slice burst kinetics in the presence and absence of tonic activity. Our results indicate that superimposition of phasic firing patterns on tonic activity provides a means to use an ex vivo preparation for an improved analysis of the mechanisms that govern the release of monoamine neurotransmitters in vivo.
Results
To verify previously reported differences between in vivo and ex vivo striatal recordings, we compared evoked DA release in anesthetized mice and in acute striatal slices after a single stimulus followed by three high-frequency bursts of 30 pulses at 50 Hz (Fig 1A). In vivo, electrical stimuli were applied to the SNpc, while DA release was recorded in the dorsal striatum with fast scan cyclic voltammetry (FSCV). DA release in the striatal slice was stimulated locally in the dorsal striatum.
As expected, in vivo DA release was below the detection limit with a single pulse stimulation and was only detectable after a burst of stimuli (Fig 1A). In contrast, a single electrical stimulus produced a large and easily resolved FSCV signal in the slice, with burst stimulation further increasing the amount of released neurotransmitter. Importantly, facilitation of DA release during consecutive stimulation bursts was observed in vivo while a depression of release occurred ex vivo (Fig 1B).
To investigate whether lack of tonic activity in the slice is responsible for the differences in DA release kinetics in vivo and ex vivo, we stimulated slices at frequencies ranging from 0.5 to 5 Hz that were intended to emulate pacemaking activity of DA neurons (Fig 2A). We note that synaptic vesicle fusion in DA axons is expected to have a higher release probability with local electrical stimulation than with action potentials generated distally in the cell bodies, due to electric charge loss along lightly myelinated dopaminergic axons (Formisano et al., 2020; Liu et al., 2018). Introduction of artificial tonic activity in the striatal slice rapidly decreased DA release in individual pulses to undetectable levels, with more rapid depletion at higher stimulation frequencies (Fig 2B-E). No evoked DA peaks were observed at any tonic frequency stimulus by the end of a 5-minute train application.
Tonic stimulation was followed by three bursts, each consisting of 30 pulses at 50 Hz, to mimic phasic activity; and identical stimulus series was applied before the start of the tonic activity as an internal control for the effect of tonic stimuli (Fig 2A). An increase in tonic frequency from 0.5 to 2 Hz produced a progressive reduction of DA release following a single phasic burst (Fig 2E-F), while no release was detected at 5 Hz, likely due to the depletion of DA stored in synaptic vesicles (Fig 2D). Interestingly, neurotransmitter release during the 2nd and 3rd high-frequency bursts was much less affected by preceding tonic stimulation (Fig 2E-F). This resulted in a progressive relief of DA release depression during a series of burst stimuli that correlated with higher tonic activity (Fig 2G): at 2 Hz tonic stimulation, there was no significant reduction in the amplitude of DA evoked by sequential bursts, suggesting that the depression of release in ex vivo preparations is dependent on the lack of tonic activity.
Control of DA release by nAChR
Striatal cholinergic interneurons (ChI) constitute only 2%–5% of striatal neurons, but establish an extensive network of axons that provides a profound control over striatal DA neurotransmission via pre- and postsynaptic cholinergic receptors (Giorguieff et al., 1976; Rice & Cragg, 2004; H. Zhang & Sulzer, 2004; Zhou et al., 2001). Similar to previously published results (McGuirt et al., 2021; Threlfell & Cragg, 2011), application of the nAChR antagonist dihydro-β-erythroidine hydrobromide (DhβE, 1 μM) decreased DA release after a single stimulus but not after a burst stimulation, thus significantly increasing burst-to-single pulse ratio (Fig 3A-C). However, inhibition of nAChR in the presence of artificial tonic activity exerted no additional effect on a single burst release (Fig 3D) or on neurotransmitter release evoked by a series of bursts (Fig 3E), suggesting that nAChRs play a limited role in DA release modulation in the presence of artificial tonic activity.
Role of D2R
The modulation of dopamine release within the striatum originates from a dynamic interplay between dopamine receptors and a local striatal circuitry. D2 auto-receptors expressed on the axons of DA neurons provide feedback control of striatal DA release via activation of potassium GIRK channels that hyperpolarize the cells (Marcott et al., 2014), while D2 receptors on ChIs regulate ACh and DA release via postsynaptic mechanisms (Kharkwal et al., 2016).
Acute treatment of striatal slice with D2R antagonist sulpiride (10 μM) significantly increased DA released by a single pulse stimulus but did not change the amount of DA overflow following burst stimulation (Fig 3A-C). Tonic slice stimuli in the presence of sulpiride caused further decrease in DA released by a single burst (Fig 3D), consistent with a disinhibition of dopaminergic axons that produced a larger depletion of the RRP of dopaminergic vesicles. Furthermore, blockade of D2R combined with tonic stimulation caused more DA to be released in consecutive bursts, leading to a facilitation of neurotransmitter release (Fig 3F, Table 1).
Discussion
We introduce a simple experimental approach that can be employed to study the role of tonic activity in modulating phasic DA release in an acute striatal slice preparation. While the highest tested tonic frequency (5 Hz) caused complete cessation of DA neurotransmission (Fig 2D), likely due to the depletion of vesicular transmitter pools, at lower frequencies we uncovered several key characteristics of DA release that have not been previously reported in ex vivo preparations (Table 1).
First, tonic slice conditioning greatly diminished DA release following a single pulse stimulation, accompanied by a much smaller decrease in phasic neurotransmitter release, thus leading to an enhanced burst-to-single stimulus ratio. Such ‘high frequency filtering’ of DA neurotransmission, also observed in vivo, would lead to enhanced signal/noise during phasic DA release in response to environmental cues critical for learning (Bromberg-Martin et al., 2010). Interestingly, similar alteration in DA release kinetics has been reported in acute striatal slices treated with nAChR blockers (Brimblecombe et al., 2018; Rice & Cragg, 2004; Sulzer et al., 2016) (Fig 3A,B), suggesting that tonic DA and ACh tones have opposite effects on the signal/noise of DA neurotransmission. Nicotinic receptors are ligand-gated non-selective cation channels that potently modulate DA release in the striatum (Exley et al., 2008; T. Zhang et al., 2009; Zhou et al., 2001) and their activation on DA neurons drives depolarization and local action potentials (Liu et al., 2022). It is reported that tonic ACh tone facilitates DA release at lower firing frequencies but has no effect or caused depression of dopamine release at higher frequencies (>2 Hz) (McGuirt et al., 2021; Paladini & Roeper, 2014; Rice & Cragg, 2004; T. Zhang et al., 2009). The similarity between the effects of nAChR antagonists and the presence of tonic slice stimuli that both decrease single pulse-evoked DA release without affecting neurotransmitter release during bursts may be explained by the desensitization of nAChR (Rice & Cragg, 2004). During local tonic electrical stimulation of the slice, basal ACh concentration likely also increases, leading to nAChR desensitization. Since recovery from desensitization of nAChR may take up to a minute (Giniatullin et al., 2005), at the time of the burst stimulus (10 sec after the end of the tonic train) the effect of the tonic conditioning would be similar to nAChR pharmacological blockade (Table 1), although the contribution of this and other mechanisms requires further investigation.
Another mechanism that exerts control over striatal DA neurotransmission is via activation of D2R expressed on both dopaminergic axons and ChI cell bodies (Kharkwal et al., 2016; Kita et al., 2007; Schultz, 1998). In acute striatal slices, tonically released DA is expected to bind D2R and induce chronic depression of DA axons and ChI, thus decreasing the probability of DA release. Conversely, D2R blockade, not only allows more DA to be released from slices without tonic activity (Fig 3A), but also causes higher apparent synaptic vesicle depletion after tonic activity (Fig 3D).
As long reported, in naïve slices, sequential bursts of stimuli cause a profound depression of DA release (Phillips et al., 2002; Schmitz et al., 2002), but we now show that tonic stimulation of the slice leads to less dependence of released DA on the burst number, especially at higher tonic frequencies (Fig 2G). Furthermore, when tonic slice stimulation was combined with the pharmacological blockade of D2R, repeated stimulus bursts facilitated DA release, a response that has not been previously observed ex vivo (Fig 3F). A possible explanation for this phenomenon is that DA release from different vesicle pools possesses different calcium sensitivities. For example, if release from the RRP has higher Ca2+ affinity (i.e., can be triggered at lower cytosolic Ca2+ concentrations) than release from the recycling and the reserve pools, then depletion of the RRP during tonic activity would reveal synaptic facilitation due to Ca2+ buildup during high-frequency bursts. Consistently, we find that burst-to-burst depression of DA release was inversely proportional to tonic frequency, while release facilitation occurred in slices treated with a D2R antagonist, which as discussed above, drives a still stronger depletion of the RRP. These phenomena may further depend on muscarinic ACh receptors (Threlfell & Cragg, 2011), GABA receptors (Lopes et al., 2019) and possibly additional receptor-mediated mechanisms.
While the experimental system we introduce is relatively simple, it features many parameters that can be varied depending on the aims of the study, including the intensity and the frequency of tonic and phasic stimuli, the duration of tonic stimulation, and the delay before the burst stimulation and between the bursts. Furthermore, optogenetic stimulation of DA axons and/or ChI cell bodies, which should cause more targeted neuronal activation than electrical stimuli albeit with non-physiological activation mechanisms, as well as assessment of basal DA and ACh levels with electrochemical or optical techniques promise to further decipher the roles of tonic dopaminergic and cholinergic activities in modulating phasic DA neurotransmission.
In summary, we present a proof of principle study that demonstrates a critical role of tonic firing in shaping dopamine release dynamics and highlights the importance of studying neurotransmitter modulation in physiologically relevant settings. Bridging the gap between ex vivo and in vivo data will pave the way for understanding the mechanisms underlying dopamine signaling and its implications for neurological disorders and pharmacological interventions.
Material and Methods
Experimental model and reagents
Male C57Bl6/J mice (3-6 months, The Jackson Laboratory, Bar Harbor, Maine) were used for both in vivo and ex vivo experiments. All animals were maintained on a 12:12 hour light/dark cycle with food and water ad libitum. All animals were maintained according to the National Institutes of Health guidelines in Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC) accredited facilities. All experimental procedures were approved by the Columbia University Institutional Animal Care and Use Committee and followed guidelines established in the NIH Guide for the Care and Use of Laboratory Animals. Sulpiride (Cat#: 0894) and dihydro-β-erythroidine hydrobromide (Cat#: 2349) were from Tocris Bioscience.
Ex vivo striatal slices
Animals were euthanized by cervical dislocation, the brain was removed and placed in ice-cold sucrose cutting solution, containing (in mM): 10 NaCl, 2.5 KCl, 25 NaHCO3, 0.5 CaCl2, 7 MgCl2, 1.25 NaH2PO4, 180 sucrose and 10 glucose, oxygenated with 95% O2/5% CO2 to pH 7.4. Coronal slices (250 μm) that included the striatum were collected and allowed to rest at 34°C for 30 min in artificial cerebrospinal fluid (ACSF; in mM): 125 NaCl, 2.5 KCl, 25 NaHCO3, 1.5 CaCl2, 1 MgCl2, 1.25 NaH2PO4 and 10 glucose, oxygenated with 95% O2/5% CO2 to pH 7.4.
Fast-scan cyclic voltammetry (FSCV) ex vivo
FSCV recordings of evoked DA release by FSCV were performed as detailed previously (Lieberman et al., 2018; McGuirt et al., 2021). Striatal slices were incubated in oxygenated ACSF at room temperature for 30 min and transferred to a recording chamber with ACSF perfused at 2 mL/min and 34°C. A carbon fiber electrode was placed in the dorsolateral striatum ∼50 μm into the slice. A triangular voltage ramps from a holding potential of - 450 mV to +800 mV over 8.5 ms (ramp of 294 V/s) were applied to the carbon fiber electrode every 100 ms. Current was recorded with an Axopatch 200B amplifier (Molecular Devices, Foster City, CA) filtered with a 10-kHz low-pass Bessel filter and digitized at 25 kHz (ITC-18 board, InstruTECH, Great Neck, NY). FSCV data acquisition and analysis were controlled by custom-built routines in Igor Pro (WaveMetrics Inc., Lake Oswego, OR). Slices were stimulated with a sharpened bipolar concentric electrode (400 μm max outer diameter; Pt/Ir; WPI) placed ∼150 μm from the recording electrode, using an Iso-Flex stimulus isolator (AMPI) triggered by TTL pulses generated in the same custom-built Igor Pro routine and synchronized with FSCV pulses. Duration of each stimulus pulse was 0.2 ms. Carbon fiber electrodes were calibrated by quantifying background-subtracted voltammograms in standard solutions of DA in ACSF prepared fresh each recording day. Concentration traces were made in Igor Pro by collecting current over time at the peak DA oxidation potential normalized to the standard DA calibration factor.
FSCV recordings in vivo
We used previously published methods for in vivo FSCV recordings (Shashaank et al., 2023; Somayaji, Cataldi, Choi, et al., 2020). Briefly, male mice were anesthetized with iso-flurane (induction 4%, maintenance 0.8–1.4%), placed on a heating pad, and head-fixed on a stereotaxic frame. A craniotomy was performed to insert a 22G bipolar stimulating electrode (P1 Technologies, Roanoke, VA) in the ventral midbrain to trigger electrical pulse trains and a carbon-fiber microelectrode (5 μm diameter, ∼150 μm length) in the dorsal striatum to measure evoked DA release. The depth of the stimulating electrode was adjusted between 4 and 4.5 mm for maximal DA release. FSCV recordings and generation of the stimulation patterns were performed as described above. The electrical pulse trains were delivered to the stimulating electrode at a constant current of 400 μA using an Iso-Flex stimulus isolator; duration of each stimulus pulse was 4 ms. The carbon-fiber microelectrodes were calibrated in artificial cerebrospinal fluid using known concentrations of DA.
Statistical Analysis
Data was acquired and analyzed using the Igor Pro software and imported into Microsoft Excel (Microsoft Corp., Redmond, WA). All statistical analysis was performed in GraphPad Prism (version 10; La Jolla., San Diego, CA). All bar graphs show the mean ± SEM. Data comparing two variables were analyzed with a two-way ANOVA, followed by Bonferroni post-hoc test. Data comparing one variable among more than two groups were analyzed with one-way ANOVA and Tukey’s post-hoc test.
SAFETY
We have not found any unexpected, new, and/or significant hazards or risks associated with the reported work.
AUTHOR CONTRIBUTIONS
SB conducted all ex vivo FSCV experiments; EAM performed in vivo FSCV recordings in anesthetized mice. Data analysis and presentation was done by SB and EVM. All other authors contributed to experimental design, critical data analysis and writing of the manuscript.
Created with BioRender
ACKNOWLEDGMENT
This research was supported by NIDA R0107418 (DS) and the JPB Foundation (DS). We thank Avery McGuirt for advice and training and Dr. Mark Wightman for critical discussions of the experimental approach.
ABBREVIATIONS
- DA
- Dopamine
- D2R
- Dopamine D2 receptor
- DHβE
- Dihydro-β-erythroidine hydrobromide
- FSCV
- Fast-scan cyclic voltammetry
- GABA
- Gamma-aminobutyric acid
- GIRK
- G-protein-coupled inward rectifier potassium
- MFB
- Medium forebrain bundle
- nAChR
- Nicotinic acetylcholine receptors
- RRP
- Readily releasable pool
- SNpc
- Substantia nigra pars compacta