Striatal Dopamine Release
Is Triggered by Synchronized Activity
in Cholinergic Interneurons
Sarah Threlfell,1,2Tatjana Lalic,1Nicola J. Platt,1Katie A. Jennings,1Karl Deisseroth,3and Stephanie J. Cragg1,2,*
1Department of Physiology, Anatomy, and Genetics, Sherrington Building, University of Oxford, Oxford OX1 3PT, UK
2Oxford Parkinson’s Disease Centre, University of Oxford, Oxford OX1 3QX, UK
3Departments of Bioengineering, Psychiatry, and Behavioral Sciences, and Howard Hughes Medical Institute, Stanford University, Stanford,
CA 94305, USA
Striatal dopamine plays key roles in our normal and
pathological goal-directed actions. To understand
dopamine function, much attention has focused
on how midbrain dopamine neurons modulate
their firing patterns. However, we identify a presyn-
aptic mechanism that triggers dopamine release
directly, bypassing activity in dopamine neurons.
We paired electrophysiological recordings of striatal
neurons with simultaneous detection of dopamine
release at carbon-fiber microelectrodes in striatal
slices. We reveal that activation of cholinergic inter-
neurons by light flashes that cause only single action
on dopamine axons. This event overrides ascending
activity from dopamine neurons and, furthermore,
is reproduced by activating ChR2-expressing thala-
mostriatal inputs, which synchronize cholinergic
interneurons in vivo. These findings indicate that
synchronized activity in cholinergic interneurons
directly generates striatal dopamine signals whose
functions will extend beyond those encoded by
dopamine neuron activity.
Striatal dopamine (DA) is critical to the regulation of motivation
and movement. Disruptions to DA signaling underlie a variety
of psychomotor disorders, including Parkinson’s disease (PD)
and addiction disorders. To understand striatal DA function,
there has been intense study of when and how midbrain DA
neurons change their firing rate, from tonic firing frequencies to
intermittent bursts of action potentials at high frequencies.
Current hypotheses posit that switches to phasic bursts of DA
neuron activity and subsequent DA release encode motivational
value and/or salience (Bromberg-Martin et al., 2010; Jin and
Costa, 2010; Phillips et al., 2003; Redgrave et al., 2008; Schultz,
2010; Tsai et al., 2009) and regulate long-term changes in striatal
synaptic plasticity (Owesson-White et al., 2008; Surmeier et al.,
2009) that underpin action selection.
Action potentials in DA neurons have been assumed to be the
principal trigger for DA transmission from striatal axons. How
temporal or rate codes in DA neuron firing are relayed into DA
release has been shown also to be modulated by presynaptic
filters in DA axons that dynamically gate action potential-
dependent DA release (Cragg, 2003; Montague et al., 2004).
Although few in number, striatal cholinergic interneurons (ChIs)
are thought to provide one such critical presynaptic mechanism
through extensive striatal arborization (Contant et al., 1996) that
supplies ACh to nicotinic receptors (nAChRs, b2-subunit con-
taining) on DA axons (Jones et al., 2001). ChIs exhibit burst-
and-pause changes that coincide with changes in DA neuron
activityonpresentation ofsalientstimuli (Dingetal.,2010;Morris
et al., 2004). ChI pauses have been suggested to reduce DA
release probability but promote the gain on DA signals when
action potential frequency in DA neurons increases (Cragg,
2006; Rice and Cragg, 2004; Threlfell and Cragg, 2011; Zhang
and Sulzer, 2004).
However, ChIs have been suggested to drive DA release from
DA axons directly without requiring ascending activity in DA
neurons (Ding et al., 2010). If physiological ACh release from
ChIs can be demonstrated to evoke DA exocytosis, it would
require us to radically reassess whether activity in DA neurons
versus ChIs is the primary basis of DA function, to reappraise
the outcome of coincident changes in activity in these neurons,
and more generally to rethink the roles of inputs to neuronal
axons versus soma. Here, we reveal such a mechanism, indi-
cating that DA function can be independent of action potentials
in DA soma; rather, activity in ChIs and their inputs that generate
depolarization locally in DA axons have unexpectedly privileged
importance in driving DA signals.
RESULTS AND DISCUSSION
To identify the effects of activation of striatal ChIs on DA trans-
mission, we incorporated the light-activated ion channel chan-
was restricted to ChIs by injecting an adeno-associated virus
58 Neuron 75, 58–64, July 12, 2012 ª2012 Elsevier Inc.
(AAV) carrying a Cre-inducible ChR2 gene (fused inframe with
the coding sequence for enhanced yellow fluorescent protein
[eYFP]) into the striatum of transgenic mice expressing Cre-
recombinase under the control of the promoter for choline
acetyltransferase (ChAT) (Figure 1A) (also see Supplemental
Information available online). In coronal slices that contain DA
axons without DA soma, single blue laser flashes (1–2 ms;
473 nm) of ChR2-expressing terminals (15- to 60-mm-diameter
spot) in dorsal or ventral striatum evoked the transient release
(FCV) at carbon-fiber microelectrodes (see Supplemental Infor-
mation) (n = 29 animals) (Figure 1B). Extracellular DA concentra-
tions reached values similar to those evoked by local electrical
stimuli (Figure 1B), indicating DA release from a population of
axons. Light-evoked DA release was reproducible for several
hours (sampling interval ?2.5 min) and required ACh activation
of nAChRs. The b2-nAChR antagonist DHbE abolished DA
release (Figure 1C; n = 10, p < 0.001) but not spiking in ChIs (Fig-
ure S1E, n = 3) indicating nAChRs postsynaptic to ChIs. ChI-
driven DA release did not require muscarinic AChRs (mAChRs,
Figure 1D, n = 11), glutamate receptors, or GABA receptors
(Figure 1E, n = 9) but was modulated by mechanisms that nor-
mally gate ACh and/or DA exocytosis; it was abolished by
Nav+-block by tetrodotoxin (TTX) (n = 10, p < 0.001), zero extra-
cellular Ca2+(n = 10, p < 0.001), D2receptor activation with quin-
pirole, (n = 8, p < 0.001), or mAChR activation with oxotremorine
(n = 10, p < 0.001), which limits ACh release from ChIs (Threlfell
et al., 2010) (Figure 1E). These observations indicate that endog-
enous ACh released from ChIs triggers DA release by activating
axonal nAChRs, bypassing action potentials in DA soma.
To understand the neuronal events required for ChI-driven DA
release, we paired recording of laser-evoked DA using FCV with
whole-cell patch-clamp recording of ChR2-expressing, eYFP-
tagged ChIs (see Supplemental Information) (Figure 2A). ChR2-
expressing ChIs had normal resting membrane potential and
TTX-sensitive action potentials (Figure S1; Table S1, n = 11).
Laser-evoked DA release was seen after action potentials were
evoked in local ChIs (latency 2.0 ± 0.5 ms, Figure 2B, n = 11).
1 p stim
0 mM Ca2+
D-AP5 GYKI MCPG
[DA]o (normalised to control)
Figure 1. Activation of ChIs Drives Striatal Dopamine Release via nAChRs
(A) Schematic of Cre-dependent AAV ChR2(H134R)-eYFP; the gene is doubly flanked by two incompatible sets of loxP sites. Upon delivery into Cre-transgenics,
ChR2-eYFP is inverted to enable transcription from the EF-1a promoter. Fluorescence shows ChR2-eYFP expression (green) in a population of ChIs (red). Scale
bar represents 20 mm. (B) A local electrical or laser pulse (473 nm, 2 ms) in striatal slices evokes release of DA in CPu. Representative traces are shown. Inset:
cyclic voltammograms identifying DA. (C and D) Mean DA release profiles (±SEM) in CPu after single laser pulse are prevented by antagonist of nAChRs (DHbE,
1 mM) (C) but not mAChRs (atropine, 2 mM) (D), n = 10–11. (E) Ionic and receptor dependence of ChI-driven DA release. TTX (1 mM); DHbE (1 mM); mAChR agonist
receptor antagonists: bicuculline (bic, 10 mM), saclofen (sac, 50 mM). Data are means ± SEM.
Striatal ACh Triggers Dopamine Release
Neuron 75, 58–64, July 12, 2012 ª2012 Elsevier Inc. 59
Short laser pulses that generated only one action potential in
any recorded ChI were sufficient to evoke DA release (Figure 2B,
n = 11); however, when we used current injection through the
patch pipette (steps or ramps) to stimulate those same neurons
individually to generate a single action potential, ongoing activity
(1–2 Hz), or brief bursts, DA release was not evoked (Figure S1,
n = 11) in ChAT-Cre or wild-type animals. One critical difference
between current injection and a laser pulse is the number of
neurons activated: the laser beam will synchronously activate
a population of ChIs and/or their axons owing to the extensive
overlapping arborization of ChI axons and dendrites (Contant
et al., 1996). These data therefore suggest that ChI-driven DA
release occurs during synchronization of activity in ChIs. The
requirement for synchronization was confirmed by showing
that laser stimuli that minimize synchrony in ChIs did not evoke
DA release. To achieve this, we recruited activity gradually in
a population of ChIs by slowly ramping laser intensity during
continuous exposure until threshold for spiking was reached in
a given recorded ChI. Using this protocol, outcome on activity
in each ChI was variable (e.g., threshold intensity, see variation
in spike frequency in Figure 2C, n = 6) and this protocol did not
evoke DA release (Figure 2C, n = 6). Multiple spikes in a given
- 60 mV
100 nM [DA]o
100 nM [DA]o
Figure 2. Synchronous Activity in Cholin-
ergic Interneuron Population Evokes DA
(A) Schematic and IR-DIC image of recording
configuration during combined recordings of ChI
activity at patch electrode and DA at carbon-fiber
microelectrode (CFM) in striatal slices from ChR2-
eYFP-expressing ChAT-Cre mice. Bottom: eYFP-
fluorescent ChIs were biocytin filled and identified
using Alexa Fluor 488 streptavidin (SA, green) and
coimmunoreactivity for ChAT (red). Scale bar re-
presents 10 mm. (B) Representative example of
simultaneous recordings of a single action poten-
tial in ChIs and DA release after a single laser
pulse (2 ms, 10 mW/mm2). Note FCV waveform
(FCV sweep) generates 8 Hz artifact at patch
electrode. Extracellular DA concentration versus
time (bottom trace) is determined from DA oxida-
tion current in background-subtracted voltam-
metric current (Faradaic signal), expanded in
bottom inset.Topinset: short latency tospike after
laser pulse (<2 ms; see Table S1). (C) Gradual
increase in laser intensity (0–2 mW/mm2) induces
variable activity in different representative ChIs
(cells 1 and 2) but no DA release. (D) Long laser
pulse (1 s, 10 mW/mm2) induces a burst of action
potentials in ChI and DA release. See also Fig-
ure S1 and Table S1.
ChI per se did not preclude DA release
since longer duration laser pulses above
threshold that evoked burst firing in ChIs
were accompanied by DA release (Fig-
ure 2D, n = 11). These data show that
synchronous recruitment of activity in
a population of ChIs and/or axons evokes
We also noted that multiple action potentials in a given ChI
induced by long laser pulses did not evoke more DA release
than a single action potential (compare Figures 2D and 2B), sug-
gesting that ChI-driven DA release does not convey frequency
information from individual ChIs. This weak relationship between
frequency and DA release is also seen with striatal electrical
stimulation when DA axons and ChIs are simultaneously depo-
larized (Rice and Cragg, 2004; Zhang and Sulzer, 2004), but
not with stimulation of medial forebrain bundle when DA axons
are activated (Chergui et al., 1994). These observations sug-
gest that ChI-driven DA release does not report frequency and
moreover that it may limit how frequency information in
ascending DA axons is transduced into DA release.
We therefore explored the relationships between frequency of
axons only, or both in combination. Trains of four laser pulses at
a range of frequencies in ChR2-expressing ChAT-Cre striatum
reliably evoked four action potentials in ChIs at corresponding
frequencies (Figure 3A), but the consequent DA release was
invariant, reaching only DA levels seen with a single light pulse
(and single action potentials) (Figures 3B and 3D, n = 8). This
refractoriness (or depression) of rerelease after release by single
Striatal ACh Triggers Dopamine Release
60 Neuron 75, 58–64, July 12, 2012 ª2012 Elsevier Inc.
synchronized spikes in ChIs was therefore not due to spike
attenuation in ChIs (and was also not due to activation of
mAChRs or D2receptors causing ACh terminal inhibition, data
not shown). These data show that ChI-driven DA release is not
a direct readout of the frequency of activity in a given ChI. By
contrast, when DA release was evoked by laser activation of
ChR2-expressing DA axons in striatum of DAT-Cre mice (Fig-
ure 3C; also see Supplemental Information; TTX sensitive, Ca2+
dependent, nAChR independent, Figure S2), DA release was
sensitive to laser frequency (Figures 3B and 3D, n = 6–9, p <
0.001). As shown previously (Rice and Cragg, 2004; Zhang and
Sulzer, 2004), when activation of DA axons occurs concurrently
with nAChR activity, as occurs here using local electrical stimu-
lation to evoke release of DA and ACh, the dominant outcome
was frequency-insensitive DA release (in all genotypes) (Fig-
ure 3E, n = 6). Frequency sensitivity was restored with nAChR-
that the frequency insensitivity of ChI-driven DA release domi-
natesoverascendingactivityin DAaxons: ChI-drivenDA release
shunts the efficacy of concurrent activity in DA axons in evoking
role for dynamic changes in the plasticity of ACh or DA release or
the nAChR effector mechanism, e.g., nAChR desensitization.
Our findings have several implications. First, the roles of excit-
ability in axons versus soma in determining neurotransmitter
release need to be reappraised. Activity in DA soma is not an
exclusive trigger for axonal DA release; striatal ACh acting at
nAChRs on DA axons bypasses midbrain DA neurons to trigger
DA release directly. It has been suggested previously that
nAChRs modulate the gain on action potential-elicited release
(Rice and Cragg, 2004), but it has also been speculated from
the effects of applied ACh or nicotine (Lambe et al., 2003; Le ´na
et al., 1993; Wonnacott, 1997) that preterminal nAChRs might
1 p4p 5 Hz
4p 10 Hz
4p 25 Hz
- 0.7 V
[DA]o (normalised to 1p)
DA + nAChRs
[DA]o (normalised to control 1p)
Figure 3. ChI-Driven DA Release Is Frequency Insensitive and Shunts DA Release Evoked by Ascending Activity
(A) Representative trace of ChR2-expressing ChIs following light pulses at various frequencies with corresponding action potentials. (B) Mean DA release profiles
(±SEM) inCPuof ChR2-expressing ChAT-Cre (black) or DAT-Cre(red) miceafterlaser pulses(blue lines, 2ms, 10 mW/mm2)of different frequency,n =6–9. Inset:
represents 20 mm. (D) Mean peak DA release in CPu is sensitive to frequency in ChR2-expressing DAT-Cre (red) (p < 0.001), but not ChR2-expressing ChAT-Cre
(black), n = 6–9. (E) Effect on mean peak [DA]oof activating ChIs only (laser control), ChIs plus DA axons (elec control), or DA axons (elec DHbE) by single pulses
with number of pulses after activating DA axons (p < 0.001) but not ChIs, with or without DA axon stimulation, n = 6. See also Figure S2. Data are means ± SEM.
Striatal ACh Triggers Dopamine Release
Neuron 75, 58–64, July 12, 2012 ª2012 Elsevier Inc. 61
trigger ectopic action potentials in axons. Our data now show
that endogenous ACh released by single action potentials
synchronized among ChIs does trigger DA release, via a direct
preterminal action. These data also add to an accumulating
body of evidence (Ding et al., 2010; Witten et al., 2010) suggest-
opposition is outmoded and oversimplistic.
Second, these data indicate that circuits that activate striatal
ChIs will have privileged roles as triggers of DA signals. What
are the likely triggers and corresponding functions? Our data
show that this ChI-driven DA signal is not a readout of activity
in individual ChIs. But mechanisms that increase activity in
ChIs in vivo should enhance the likelihood of synchronous
activity in a subpopulation and bring this mechanism to
threshold. Thus, ChI-driven DA release will reflect ChI population
activity as a coincidence detector. Inputs that drive excitability
and/or synchrony in ChIs could in turn be powerful triggers of
DA signals. In vivo, ChI activity is strongly driven and synchro-
laminar nuclei, that provide a rich innervation of networks of ChIs
(as well as MSNs) (Ding et al., 2010; Goldberg and Reynolds,
2011; Morris et al., 2004; Raz et al., 1996; Smith et al., 2004)
and that show stereotyped burst activity on presentation of
salient stimuli (Aosaki et al., 1994; Matsumoto et al., 2001). We
directly tested the intriguing possibility that activation of intrala-
minar thalamic glutamate inputs to striatum might also drive DA
release via a striatal nAChR-dependent mechanism. Indeed,
laser activation of ChR2-eYFP-expressing thalamostriatal axons
arising from intralaminar thalamus in CaMKII-Cre mice evoked
DA release in coronal striatal slices, and this was prevented by
nAChR inhibition and, necessarily, glutamate receptor antago-
nists but not GABA receptor antagonists (Figure 4; n = 4 animals,
TTX-sensitive, Ca2+-dependent). ACh-dependent DA signals
can therefore be driven by the thalamic inputs that synchronize
activity in ChIs in vivo. It is interesting in this regard that the rela-
tively ‘‘digital’’ nature of the stereotyped burst activity in the
thalamostriatal network that is associated with salient event
detection parallels the lack of simple frequency dependence in
the ChI activation of DA release seen here. In any event, these
data suggest that DA may be important for conveying salience-
or attention-related signals mediated not through changes in DA
Third, we would expect that a ChI-driven DA signal will have
key outcomes for DA functions that are encoded by dynamic
patterns of activity in DA neurons themselves. The outcome
will depend entirely on the timing of activity in DA neurons rela-
tive to ChIs. Pauses in ChIs have been suggested previously to
remove a low-pass filter on DA release during concurrent
changes in DA neuron activity (Cragg, 2006). Prior ChI-driven
DA release could shunt (limit) the impact of subsequent changes
in DA neuron activity, while alternatively, postpause ‘‘rebound’’
facilitation in ChI activity (Aosaki et al., 1995; Apicella, 2007;
Morris et al., 2004), which probably corresponds to increased
synchrony in the population, could critically supplement pre-
ceding DA signals and promote, for example, the selection of
a behavior. In addition, discrete functions for DA could be
driven by synchronous activity in ChIs despite an absence of
accompanying phasic changes in DA neuron activity, which
otherwise would be taken as evidence for functions not requiring
phasic DA. Furthermore, what might be the outcome for nicotine
action? By desensitizing nAChRs on DA axons, nicotine would
the control of DA release to activity in DA neurons without modu-
lation by ChIs. In this case, DA release might be a more direct
reporter of activity in DA neurons than with nAChRs active
(Rice and Cragg, 2004). Hypo- or hypercholinergic states impli-
cated in basal ganglia disorders including PD, Huntington’s,
Tourette’s, and dystonia and in the actions of addictive drugs
could correspondingly result in the behavioral dysfunctions
that underlie each of these disorders.
In summary, we show that endogenous striatal ACh release by
synchronized activity in ChIs is sufficient to evoke DA release
and thereby uncouple DA release from its relationship to activity
in DA neurons. This mechanism may clamp or reinforce DA
release triggered by ascending activity from DA axons depend-
ing on timing and could endow ChIs and DA with key functions
that go beyond those identified from DA neuron recordings in
the processes underpinning action selection.
Virus Injections and Slice Preparation
To generate expression of ChR2 in ChIs, DA neurons, or thalamos-
triatal glutamate inputs, we used a Cre-loxP approach by injecting a
Cre-inducible recombinant AAV vector containing ChR2 (pAAV-double
floxed-hChR2(H134R)-EYFP-WPRE-pA) in mice expressing Cre-recombinase
in choline acetyltransferase (ChAT)-, dopamine transporter (DAT)-, or
-0.7 V +1.3 V
Figure 4. Activation of Thalamostriatal Axons Evokes ACh-Depen-
dent DA Release
(A) Cartoon scheme of striatal connectivity between thalamostriatal afferents,
ChIs,and DA axons. (B–E) Single laser/LED activation (473 nm,2 ms) of ChR2-
expressed thalamostriatal axons in striatal slices evokes release of DA in CPu
that was prevented by inhibition of nAChRs (DHbE, 1 mM) (B), NMDA and
AMPA glutamate receptors (D-AP5, 50mM; GYKI 52466, 10 mM) (C), as well as
Na+v-channels with TTX (1 mM) (D), but not by inhibition of GABAAand GABAB
receptors (bicuculine, 10 mM; saclofen 50 mM) (E). Mean ± SEM, n = 5. Inset in
(B): cyclic voltammogram identifying DA.
Striatal ACh Triggers Dopamine Release
62 Neuron 75, 58–64, July 12, 2012 ª2012 Elsevier Inc.
Ca2+-calmodulin-dependent kinase II (CaMKII)-positive neurons, respec-
tively. Transgenic mice were bred from homozygotes for ChAT-internal
ribosome entry site (IRES)-Cre, DAT-IRES-Cre, or CaMKII-Cre obtained from
Jackson Laboratories (B6.129S6-Chattm1(cre)Lowl/J, stock 006410; B6.SJL-
Slc6a3tm1.1(cre)Bkmn/J, stock 006660; B6.Cg-Tg(Camk2a-cre)T29-1Stl/J, stock
005359). The experimental data presented in this paper are from ChAT-Cre
homozygote (and heterozygote, data not shown), DAT-Cre heterozygote, or
CaMKII-Cre homozygote mice aged 2–8 months. Mice were anaesthetised
Bilateral intracerebral injections of a Cre-inducible recombinant AAV (1 ml per
site for ChAT-Cre and DAT-Cre mice, 300 nl per site for CaMKII-Cre mice)
were made with a 2.5 ml, 33 gauge Hamilton syringe using a microinjector at
0.2 ml/min. In ChAT-Cre mice, injections were made in dorsal CPu
(AP +1.0 mm, ML ±1.8 mm, DV ?2.2 mm) and in contralateral NAc core
(AP +1.0 mm, ML ±1.0 mm, DV ?4.0 mm). In DAT-Cre mice, injections were
made in SNc (AP ?3.5 mm, ML ±1.2 mm, DV ?4.0 mm) and in contralateral
VTA (AP ?3.1 mm, ML ±0.5 mm, DV ?4.4 mm). In CaMKII-Cre mice, injections
were made in the intralaminar nucleus of the thalamus (AP ?2.3, ML ±0.5,
DV ?3.4 mm). Wild-type C57BL/6 mice used in some experiments were aged
postnatal days (P) 14–P22.
On days 12–76 postinjection, mice were decapitated after cervical disloca-
tion or halothane anesthesia (for combined patch-clamp/FCV recordings).
Coronal slices, 300 mm thick, containing CPu and NAc were prepared as
described previously in ice-cold HEPES-buffered artificial cerebrospinal fluid
(aCSF) or high-sucrose aCSF (for electrophysiology, see below) saturated
with 95% O2/5% CO2. Slices were then maintained in a bicarbonate-buffered
aCSF at room temperature prior to recording. During recordings, neurons
were visualized on an upright microscope (Olympus BX51WI) equipped with
IR-DIC,fluorescenceopticsforvisualizing eYFP,and acharge-coupleddevice
Fast-Scan Cyclic Voltammetry
Slices were superfused with a bicarbonate-buffered aCSF maintained at
30?C–32?C as described previously (Rice and Cragg, 2004; Threlfell et al.,
2010). Extracellular DA concentration ([DA]o) was monitored using fast-scan
cyclic voltammetry (FCV) with 7-mm-diameter carbon fiber microelectrodes
(CFMs; tip length 50–100 mm) and a Millar voltammeter (Julian Millar, Barts
and the London School of Medicine and Dentistry) as described previously
(Threlfell et al., 2010). In brief, the scanning voltage was a triangular waveform
(?0.7V to +1.3V range versus Ag/AgCl) at a scan rate of 800V/s and sampling
frequency of 8 Hz. Electrodes were calibrated in 1–2 mM DA in each experi-
mental media.Forfurther details, seeSupplemental ExperimentalProcedures.
For whole-cell patch-clamp studies (in isolation or in combination with FCV),
300 mm coronal slices containing CPu and NAc were prepared in ice-cold
high-sucrose aCSF containing 85 mM NaCl, 25 mM NaHCO3, 2.5 mM KCl,
1.25 mM NaH2PO4, 0.5 mM CaCl2, 7 mM MgCl2, 10 mM glucose, and
75 mM sucrose after decapitation under halothane anesthesia. Slices were
then transferred to oxygenated aCSF (95% O2/5% CO2) containing 130 mM
NaCl, 25 mM NaHCO3, 2.5 mM KCl, 1.25 mM NaH2PO4, 2 mM CaCl2, 2 mM
MgCl2, and 10 mM glucose at 35?C for 30–45 min and then maintained at
room temperature until recording. During recordings, slices were superfused
with aCSF saturated with 95% O2/5% CO2at 32?C. Whole-cell patch-clamp
electrodes (4–7 MU) were filled with an intracellular solution containing
120 mM K-gluconate, 10 mM KCl, 10 mM HEPES, 4 mM MgATP, 0.3 mM
NaGTP, 10 mM Na-phosphocreatine, and 0.5% biocytin. ChIs in the striatum
were identified by their distinctive morphological features (Figure S1A) (large
somas and thick primary dendrites) and their characteristic electrophysio-
logical properties, prominent Ih, AHP, and broad action potential (Figures
700B amplifier and digitized at 10–20 kHz using Digidata 1440A acquisition
board. While performing current-clamp recordings, a small amount of holding
current (typically <?25 pA) was injected when necessary to keep the cell close
to its initial resting membrane potential (?60mV). Biocytin was included in the
intracellular solution to allow post hoc visualization and confirmation of cell
identity. All data were analyzed offline with Clampfit (pClamp 10), Neuromatic
(http://neuromatic.thinkrandom.com), and custom-written software running
within IgorPro environment.
Light and Electrical Stimulation
ChR2-expressing fibers were activated using a 473 nm diode laser (DL-473,
Rapp Optoelectronic) coupled to the microscope with a fiber optic cable
(200 mm multimode, NA 0.22), which illuminated a 15- to 60-mm-diameter
spot (403/103 water-immersion objectives) or, in CaMKII experiments, an
LED system (OptoLED, CAIRN) (see Supplemental Experimental Procedures).
TTL-driven laser pulses (1–2 msduration, 2–40 mW/mm2at specimen) or elec-
trical pulses (0.6–0.7 mA, 200 ms) were delivered at a variety of frequencies
designed to mimic physiological firing frequencies. Light power at microscope
objective exit was 2–40 mW/mm2(see Figure S2). Electrical stimulation was
delivered evoked by a local bipolar concentric electrode (25 mm diameter,
Pt/Ir; FHC). Both light and electrical stimuli were delivered locally; the laser
spot was out of field of view of the CFM (?200–300 mm from CFM) and
stimulating electrode was placed ?150 mm from the CFM. Mean peak light-
evoked [DA]oin dorsal CPu from ChAT-Cre (1.4 ± 0.2 mM) or DAT-Cre (1.0 ±
0.1 mM) was not significantly different (n = 24, p > 0.05). Data presented here
is from dorsal CPu; however, we made similar observations in NAc (data not
Data were acquired and analyzed using Axoscope 10.2 (Molecular Devices)
and locally written programs. Data are represented as means ± SEM, and
‘‘n’’ refers to the number of observations. The number of animals in each
data set is R3. Data are expressed as extracellular concentration of dopamine
([DA]o), or as [DA]onormalized to a single pulse in control. Comparisons for
statistical significance were assessed by one- or two-way ANOVA and post
hoc multiple-comparison t tests or unpaired t tests using GraphPad Prism.
Levels of DA indicated either after current-induced activity in ChIs (Figures
S1F–S1H) or while gradually increasing laser power from 0 mW/mm2until
spike threshold is reached in single ChIs (Figure 2C) were indistinguishable
D(-)-2-Amino-5-phosphonovaleric acid (D-AP5), 4-(8-methyl-9H-1,3-dioxolo
hydrochloride), (S)-a-methyl-4-carboxyphenylglycine [(S)-MCPG], oxotre-
morine-M (Oxo-M), bicuculline methiodide, and saclofen were purchased
from Tocris Bioscience or Ascent Scientific. Atropine, dihydro-b-erythroidine
(DHbE), and all other reagents were purchased from Sigma-Aldrich. Drugs
were dissolved in distilled water, aqueous alkali [(S)-MCPG], or aqueous acid
(GYKI 52466 hydrochloride) to make stock aliquots at 1,000–10,0003 final
concentrations and stored at ?20?C until required. Stock aliquots were diluted
with oxygenated aCSF to final concentration immediately before use.
To determine the specificity of ChR2 expression in ChAT-Cre or DAT-Cre
mice, we fixed acute striatal (ChAT) or midbrain slices (DAT) containing
ChR2-eYFP positive neurons postrecording and processed them for ChAT
and/or TH and/or biocytin immunoreactivity. Immunoreactivity was visualized
using fluorescent secondary antibodies (see Supplemental Experimental
Supplemental Information includes two figures, one table, and Supplemental
Experimental Procedures and can be found with this article online at http://
We thank Neil Blackledge, Rob Klose, Diogo Pimentel, Ole Paulsen, Dennis
Kaetzel, Gero Miesenbock, P. Wendy Tynan, and Oxford Biomedical Services
for their invaluable input. This work was supported by a Parkinson’s UK
Striatal ACh Triggers Dopamine Release
Neuron 75, 58–64, July 12, 2012 ª2012 Elsevier Inc. 63
Monument Trust Discovery Award and Parkinson’s UK Grants G-0808 and
G-0803, the Royal Society, and the MRC.
Accepted: April 25, 2012
Published: July 11, 2012
Aosaki, T., Graybiel, A.M., and Kimura, M. (1994). Effect of the nigrostriatal
dopamine system on acquired neural responses in the striatum of behaving
monkeys. Science 265, 412–415.
Aosaki, T., Kimura, M., and Graybiel, A.M. (1995). Temporal and spatial
characteristics of tonically active neurons of the primate’s striatum.
J. Neurophysiol. 73, 1234–1252.
Apicella, P. (2007). Leading tonically active neurons of the striatum from
reward detection to context recognition. Trends Neurosci. 30, 299–306.
Bromberg-Martin, E.S., Matsumoto, M., and Hikosaka, O. (2010). Dopamine in
motivational control: rewarding, aversive, and alerting. Neuron 68, 815–834.
between impulse flow, dopamine release and dopamine elimination in the rat
brain in vivo. Neuroscience 62, 641–645.
Contant,C., Umbriaco, D.,Garcia,S.,Watkins,K.C.,andDescarries,L.(1996).
Ultrastructural characterization of the acetylcholine innervation in adult rat
neostriatum. Neuroscience 71, 937–947.
Cragg, S.J. (2003). Variable dopamine release probability and short-term plas-
ticity between functional domains of the primate striatum. J. Neurosci. 23,
Cragg, S.J. (2006). Meaningful silences: how dopamine listens to the ACh
pause. Trends Neurosci. 29, 125–131.
Ding, J.B., Guzman, J.N., Peterson, J.D., Goldberg, J.A., and Surmeier, D.J.
(2010). Thalamic gating of corticostriatal signaling by cholinergic interneurons.
Neuron 67, 294–307.
Goldberg, J.A., and Reynolds, J.N. (2011). Spontaneous firing and evoked
pauses in the tonically active cholinergic interneurons of the striatum.
Neuroscience 198, 27–43.
Jin, X., and Costa, R.M. (2010). Start/stop signals emerge in nigrostriatal
circuits during sequence learning. Nature 466, 457–462.
Jones, I.W., Bolam, J.P., and Wonnacott, S. (2001). Presynaptic localisation of
the nicotinic acetylcholine receptor beta2 subunit immunoreactivity in rat
nigrostriatal dopaminergic neurones. J. Comp. Neurol. 439, 235–247.
Lambe, E.K., Picciotto, M.R., and Aghajanian, G.K. (2003). Nicotine induces
glutamate release from thalamocortical terminals in prefrontal cortex.
Neuropsychopharmacology 28, 216–225.
Le ´na, C., Changeux, J.P., and Mulle, C. (1993). Evidence for ‘‘preterminal’’
nicotinic receptors on GABAergic axons in the rat interpeduncular nucleus.
J. Neurosci. 13, 2680–2688.
Matsumoto, N., Minamimoto, T., Graybiel, A.M., and Kimura, M. (2001).
Neurons in the thalamic CM-Pf complex supply striatal neurons with infor-
mation about behaviorally significant sensory events. J. Neurophysiol. 85,
Montague, P.R., McClure, S.M., Baldwin, P.R., Phillips, P.E., Budygin, E.A.,
Stuber, G.D., Kilpatrick, M.R., and Wightman, R.M. (2004). Dynamic gain
control of dopamine delivery in freely moving animals. J. Neurosci. 24,
Morris, G., Arkadir, D., Nevet, A., Vaadia, E., and Bergman, H. (2004).
Coincident but distinct messages of midbrain dopamine and striatal tonically
active neurons. Neuron 43, 133–143.
Owesson-White, C.A., Cheer, J.F., Beyene, M., Carelli, R.M., and Wightman,
R.M. (2008). Dynamic changes in accumbens dopamine correlate with
learning during intracranial self-stimulation. Proc. Natl. Acad. Sci. USA 105,
Phillips, P.E., Stuber, G.D., Heien, M.L., Wightman, R.M., and Carelli, R.M.
(2003). Subsecond dopamine release promotes cocaine seeking. Nature
Raz, A., Feingold, A., Zelanskaya, V., Vaadia, E., and Bergman, H. (1996).
Neuronal synchronization of tonically active neurons in the striatum of normal
and parkinsonian primates. J. Neurophysiol. 76, 2083–2088.
Redgrave, P., Gurney, K., and Reynolds, J. (2008). What is reinforced by
phasic dopamine signals? Brain Res. Brain Res. Rev. 58, 322–339.
Rice, M.E., and Cragg, S.J. (2004). Nicotine amplifies reward-related dopa-
mine signals in striatum. Nat. Neurosci. 7, 583–584.
Schultz, W. (2010). Dopamine signals for reward value and risk: basic and
recent data. Behav. Brain Funct. 6, 24.
Smith, Y., Raju, D.V., Pare, J.F., and Sidibe, M. (2004). The thalamostriatal
system: a highly specific network of the basal ganglia circuitry. Trends
Neurosci. 27, 520–527.
Surmeier, D.J., Plotkin, J., and Shen, W. (2009). Dopamine and synaptic plas-
ticity in dorsal striatal circuits controlling action selection. Curr. Opin.
Neurobiol. 19, 621–628.
Threlfell, S., and Cragg, S.J. (2011). Dopamine signaling in dorsal versus
ventral striatum: the dynamic role of cholinergic interneurons. Front Syst.
Neurosci. 5, 11.
Threlfell, S., Clements, M.A., Khodai, T., Pienaar, I.S., Exley, R., Wess, J., and
Cragg, S.J. (2010). Striatal muscarinic receptors promote activity dependence
of dopamine transmission via distinct receptor subtypes on cholinergic inter-
neurons in ventral versus dorsal striatum. J. Neurosci. 30, 3398–3408.
Tsai, H.C., Zhang, F., Adamantidis, A., Stuber, G.D., Bonci, A., de Lecea, L.,
and Deisseroth, K. (2009). Phasic firing in dopaminergic neurons is sufficient
for behavioral conditioning. Science 324, 1080–1084.
Witten, I.B., Lin, S.C., Brodsky, M., Prakash, R., Diester, I., Anikeeva, P.,
Gradinaru, V., Ramakrishnan, C., and Deisseroth, K. (2010). Cholinergic inter-
neurons control local circuit activity and cocaine conditioning. Science 330,
Wonnacott, S. (1997). Presynaptic nicotinic ACh receptors. Trends Neurosci.
Zhang, H., and Sulzer, D. (2004). Frequency-dependent modulation of dopa-
mine release by nicotine. Nat. Neurosci. 7, 581–582.
Striatal ACh Triggers Dopamine Release
64 Neuron 75, 58–64, July 12, 2012 ª2012 Elsevier Inc.