J Neurophysiol 94: 2195–2206, 2005.
First published June 22, 2005; 10.1152/jn.00035.2005.
Direct Excitation of Hypocretin/Orexin Cells by Extracellular ATP
at P2X Receptors
Guido Wollmann, Claudio Acuna-Goycolea, and Anthony N. van den Pol
Department of Neurosurgery, Yale University School of Medicine, New Haven, Connecticut
Submitted 10 January 2005; accepted in final form 8 June 2005
INTRODUCTION
The lateral hypothalamus harbors a variety of functionally
distinct neuron populations. One of these cell groups, the hypocretin/orexin neurons (de Lecea et al. 1998; Sakurai et al. 1998),
has gained great attention because their loss seems to be the
causative factor for narcolepsy, a condition characterized by
excessive daytime sleepiness and cataplexy (Peyron et al. 2000;
Thannickal et al. 2000; van den Pol 2000). Dogs with mutations
in their hypocretin receptor-2 gene (Lin et al. 1999) and hypocretin-knockout mice (Chemelli et al. 1999) present a narcoleptic
phenotype. Hypocretin has also been postulated to play a crucial
role in energy homeostasis and feeding behavior (Hara et al. 2001;
Sakurai et al. 1998; Yamanaka et al. 2003). Hypocretin cells
project widely throughout the brain and the spinal cord (Chen et
al. 1999; Peyron et al. 1998; van den Pol 1999).
Address for reprint requests and other correspondence: A. N. van den Pol,
Dept. of Neurosurgery, Yale Univ. School of Medicine, 333 Cedar St., New
Haven, CT 06520 (E-mail anthony.vandenpol@yale.edu).
www.jn.org
Extracellular ATP can modulate synaptic transmission and
neuronal excitability in the CNS (Cunha and Ribeiro 2000;
Robertson et al. 2001). The effects of ATP can be mediated
either through fast-responding ionotropic P2X receptors or
through slower metabotropic G protein– coupled P2Y receptors
(Edwards 1994; Edwards and Gibb 1993). ATP can also
directly inhibit N-methyl-D-aspartate (NMDA) receptor–mediated excitatory currents (Ortinau et al. 2003) or glutamate
release (Inoue 1998) in cultured hippocampal neurons. In
contrast, the ATP actions in locus coeruleus, cerebellar Purkinje cells, and somatosensory cortex appear excitatory
(Brockhaus et al. 2004; Nieber et al. 1997; Pankratov et al.
2003). Additionally, ATP can act intracellularly to modulate or
activate a number of ion channels (e.g., the ATP-activated
K⫹-channel) (Ballanyi 2004; Liss and Roeper 2001).
Purinergic receptors have been reported in a number of loci
in the mammalian brain (Kanjhan et al. 1999; Norenberg and
Illes 2000) including hippocampus, hypothalamus, cerebral
cortex, medial habenula, locus coeruleus, and the dorsal horn
of spinal cord (Bardoni et al. 1997; Edwards et al. 1997; Jo and
Role 2002; Jo and Schlichter 1999; Pankratov et al. 1998;
Sorimachi et al. 2001), suggesting a role for external ATP in
modulating synaptic transmission in these brain regions. ATP
can be released from axon terminals, and there is evidence for
synaptic colocalization and corelease of ATP in cholinergic,
glutamatergic, and GABAergic neurons (Jo and Role 2002;
Mori et al. 2001; Silinsky and Redman 1996). In addition, ATP
is involved in intercellular signaling from glia to other glia and
to neurons (Bowser and Khakh 2004; Zhang et al. 2003). ATP
is also released in a number of pathological conditions, including stroke, trauma, and epilepsy (Ciccarelli et al. 2001; Hansson and Ronnback 2003; James and Butt 2002).
P2X receptors are expressed in the hypothalamus (Xiang et al.
1998), where they may play a role in ATP modulation of electrical
activity (Jo and Role 2002; Sorimachi et al. 2001). We studied the
effect of extracellular ATP on a single neuronal phenotype, the
hypocretin-producing cells, using transgenic mice expressing
green fluorescent protein (GFP) exclusively in hypocretin cells.
We found substantial direct excitation of hypocretin cells through
activation of ionotropic P2X receptors.
METHODS
Animals
Transgenic mice (kindly provided by Dr. T. Sakurai, University of
Tsukuba) that express enhanced GFP under control of the hypocretin
The costs of publication of this article were defrayed in part by the payment
of page charges. The article must therefore be hereby marked “advertisement”
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
0022-3077/05 $8.00 Copyright © 2005 The American Physiological Society
2195
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on October 1, 2016
Wollmann, Guido, Claudio Acuna-Goycolea, and Anthony N. van
den Pol. Direct excitation of hypocretin/orexin cells by extracellular ATP
at P2X receptors. J Neurophysiol 94: 2195–2206, 2005. First published
June 22, 2005; 10.1152/jn.00035.2005. Hypocretin/orexin (hcrt) neurons play an important role in hypothalamic arousal and energy
homeostasis. ATP may be released by neurons or glia or by pathological conditions. Here we studied the effect of extracellular ATP on
hypocretin cells using whole cell patch-clamp recording in hypothalamic slices of transgenic mice expressing green fluorescent protein
(GFP) exclusively in hcrt-producing cells. Local application of ATP
induced a dose-dependent increase in spike frequency. In the presence
of TTX, ATP (100 ␮M) depolarized the cells by 7.8 ⫾ 1.2 mV. In
voltage clamp under blockade of synaptic activity with the GABAA
receptor antagonist bicuculline, and ionotropic glutamate receptor
antagonists DL-2-amino-5-phosphonopentanoic acid (AP-5) and
6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), ATP (100 ␮M)
evoked an 18 pA inward current. The inward current was blocked by
extracellular choline substitution for Na⫹, had a reversal potential of
⫺27 mV, and was not affected by nominally Ca2⫹-free external
buffer, suggesting that ATP activated a nonselective cation current.
All excitatory effects of ATP showed rapid attenuation. ATP-induced
excitatory actions were mimicked by nonhydrolyzable ATP-␥-S but
not by ␣,␤-MeATP and inhibited by the purinoceptor antagonists
suramin and pyridoxal phosphate-6-azo(benzene-2,4-disulfonic acid)
tetrasodium salt (PPADS). The current was potentiated by a decrease
in bath pH, suggesting P2X2 subunit involvement. Frequency and
amplitude of spontaneous and miniature synaptic events were not
altered by ATP. Suramin, but not PPADS, caused a small suppression
of evoked excitatory synaptic potentials. Together, these results show
a depolarizing response to extracellular ATP that would lead to an
increased activity of the hypocretin arousal system.
2196
G. WOLLMANN, C. ACUNA-GOYCOLEA, AND A. N. VAN DEN POL
promoter were used in this study. Selectivity and specificity of
reporter gene expression has been previously described (Li et al. 2002;
Yamanaka et al. 2003). Mice were kept in a 12:12-h light-dark cycle.
Experiments in this study were performed in accordance with institutional guidelines of the Yale University Committee on Animal Use.
Hypothalamic slice preparation
Patch-clamp recordings
Voltage- and current-clamp experiments were performed in the
whole cell configuration using low-resistance (4 – 6 M⍀) borosilicate
glass pipettes (World Precision Instruments, Sarasota, FL). Pipettes
were pulled with a PP-83 vertical puller (Narishige, Tokyo, Japan)
and filled with a pipette solution containing (in mM) 130 KMeSO4
[KCl for inhibitory postsynaptic current (IPSC) and miniature
(m)IPSC experiments], 1 MgCl2, 10 HEPES, 1.1 EGTA, 2 Mg-ATP,
0.5 Na2-GTP, and 10 Na2-phosphocreatin, pH 7.3 with KOH. Hypocretin cells were identified through fluorescence and approached
under visual control. Whole cell configuration was established after
stable gigaohm seal formation using gentle negative pressure. All
recordings were made with an EPC9 amplifier controlled by Pulse
acquisition software (HEKA Elektronik, Lambrecht/Pfalz, Germany).
Capacitance artifacts were automatically compensated by the acquisition software. Access resistance was monitored continuously to
assess stable recording conditions. If not otherwise mentioned, cells
were held at ⫺60 mV in voltage-clamp experiments, and currentclamp experiments were done with zero current injection through the
recording pipette (at resting potential). To study evoked excitatory
potentials, bipolar stimulation electrodes (World Precision Instruments) were placed within the lateral hypothalamus medial or lateral
to the recorded cell. Stimulatory pulse characteristics (10 –150 ␮A,
0.5 ms, 0.2 Hz) were controlled by an Isostim stimulator (A320,
World Precision Instruments). All experiments were performed at
32°C controlled by a dual sensor flowheater (Warner Instruments).
Analysis and statistics
PulseFit (HEKA Electronik), Axograph (Axon Instruments, Foster
City, CA), Igor Pro (WaveMetrics, Lake Oswego, OR), and Kaleidagraph (Synergy, Reading, PA) software were used for analysis. Adobe
Photoshop software was used for figure layout. For studying miniature
and spontaneous synaptic currents, an event-defining algorithm was
used in Axograph (Axon Instruments), and the amplitude threshold
was set to ⬎5 pA (Bekkers and Stevens 1995; Gao and van den Pol
2001). Data are expressed as mean ⫾ SE. For statistical evaluation,
ANOVA was used either in one-way mode or in conjunction with a
Bonferroni post hoc test for multiple comparisons, when indicated.
The cumulative distributions of mPSCs under different conditions
were compared using the nonparametric Kolmogorov-Smirnov test.
For all tests, P ⬍ 0.05 was considered statistically significant. Higher
J Neurophysiol • VOL
Preparation and application of reagents
Most drugs were prepared as a 1,000 times stock solution. In most
cases, drugs were locally applied by using a flow pipette with a
250-␮m tip diameter aimed at the recorded cell. During control
recordings, normal ACSF was applied through the flow pipette.
DL-2-amino-5-phosphonopentanoic acid (AP-5), 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), bicuculline methiodide (BIC), ATP,
adenosine 5⬘[␥-thio]triphosphate tetralithium salt (ATP-␥-S), 8-(3benzamido-4-methylbenzamido)-naphthalene-1,3,5-trisulfonic acid
(suramin), and pyridoxal phosphate-6-azo(benzene-2,4-disulfonic
acid) tetrasodium salt (PPADS) were purchased from Sigma (St.
Louis, MO), TTX was obtained from Alomone Labs (Jerusalem,
Israel). Stock solutions for ATP, ATP-␥-S, suramin, and PPADS were
prepared before each experiment.
RESULTS
ATP excites hypocretin cells
To address the effect of extracellular ATP on hypocretin
cells, we used acute hypothalamic slices of transgenic mice
expressing GFP exclusively in hypocretin-producing cells. Local application of ATP (100 ␮M) consistently induced a rapid
increase in spike frequency by 50.2 ⫾ 8.2% of control (Fig.
1A; range, 8 –113%; P ⬍ 0.05; n ⫽ 14; ANOVA). This
activation significantly decreased within 30 s. after onset of
ATP application, even in the continued presence of ATP, as
shown in the mean time course of 14 experiments (Fig. 1B).
Dose–response analysis revealed significant excitatory effects
for 3, 10, and 100 ␮M, respectively (P ⬍ 0.05; n ⱖ 4;
ANOVA). Doses of 0.3 (P ⫽ 0.42) and 1.0 ␮M (P ⫽ 0.12) did
not evoke a statistically significant increase in spike frequency
(Fig. 1C). To determine whether this excitatory action was
caused by direct postsynaptic mechanisms, we bath-applied
TTX (0.5 ␮M) to block spike-mediated transmitter release.
Under these conditions, current-clamp recordings revealed a
rapid ATP-induced depolarization of the cell membrane (Fig.
1D), indicating that ATP exerts direct postsynaptic effects on
hypocretin neurons, with most cells presenting signs of a decay
of the ATP effect after 30 s of ATP application. The average
peak value for this depolarization in TTX was 7.77 ⫾ 1.23 mV
(range, 2.8 –13.1 mV; P ⬍ 0.001; n ⫽ 7; ANOVA). In
addition, we evaluated the ATP actions on hypocretin cells in
voltage-clamp (⫺60 mV holding potential). In normal ACSF
bath solution, ATP (100 ␮M) induced an inward current of
15.2 ⫾ 1.7 pA (range, 10.8 –27.8 pA; P ⬍ 0.01; n ⫽ 10;
ANOVA). To determine whether this ATP-induced inward
current results from direct postsynaptic mechanisms, TTX (0.5
␮M), AP5 (50 ␮M), CNQX (10 ␮M), and BIC (30 ␮M) were
bath-applied to prevent glutamate- or GABA-mediated synaptic transmission. These experiments revealed an inward current
of 17.9 ⫾ 4.1 pA (range, 3.6 –34.0 pA; P ⬍ 0.01; n ⫽ 7;
ANOVA) elicited by ATP (100 ␮M; Fig. 1E), further substantiating the direct nature of the ATP action on hypocretin cells.
The effect of ATP on the whole cell input resistance in
hypocretin neurons was examined. In these experiments, a
series of short (300 ms) negative current steps from 0 to ⫺140
pA were injected to the cells through the recording pipette in
20-pA increments. To compare the variability of the basic
94 • SEPTEMBER 2005 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on October 1, 2016
Two- to 3-wk-old mice were deeply anesthetized with pentobarbital
sodium (100 mg/kg), and the brain was carefully removed and placed
in ice-cold oxygenated (95% O2-5% CO2) high-sucrose solution
containing (in mM) 220 sucrose, 2.5 KCl, 6 MgCl2, 1 CaCl2, 1.25
NaH2PO4, 26 NaHCO3, and 10 glucose, pH 7.4 with NaOH. A tissue
block containing the hypothalamus was excised and transferred to a
vibratome. Slices (250 ␮m thick; 350 ␮m for evoked potentials) were
carefully moved to an equilibration chamber filled with oxygenated
(95% O2-5% CO2) normal artificial cerebrospinal fluid (ACSF) containing (in mM) 124 NaCl, 3 KCl, 2 MgCl2, 2 CaCl2, 1.23 NaH2PO4,
26 NaHCO3, and 10 glucose, pH 7.4 with NaOH. After a 1- to 2-h
resting period, slices were transferred to a recording chamber of
an upright Olympus BX51 WI microscope with infrared differential interference contrast condenser (IR-DIC) and fluorescence capabilities.
grades of significances (P ⬍ 0.01; P ⬍ 0.001) are indicated where
applicable.
ATP-P2X RECEPTORS IN HYPOCRETIN/OREXIN NEURONS
input resistance between cells, resulting voltage amplitudes
were normalized to the maximum value at the biggest current
step for each cell. The slope for this normalized I/V curve was
calculated with a linear curve-fitting function and changed
from 0.71 ⫾ 0.01 to 0.64 ⫾ 0.02 in the presence of ATP and
recovered to 0.72 ⫾ 0.03 after ATP washout, indicating a
2197
significant ATP-induced decrease of the normalized input resistance by 10.6 ⫾ 3.3% (P ⬍ 0.05; n ⫽ 8; ANOVA; Fig. 1F),
suggesting ATP opened ion channels. The slope correlates to
absolute values for input resistance of 629, 598, and 618 M⍀
for control, ATP (100 ␮M), and recovery, respectively. All
effects of ATP were reversible, with values returning to control
levels 0.5–3 min after ATP washout.
P2X receptors mediate ATP actions in hypocretin cells
FIG. 1. Extracellular ATP excites hypocretin neurons. A: representative
trace of the excitatory action of local ATP (100 ␮M) application on the
spontaneous spike frequency. B: mean time-course of the ATP-induced spike
frequency increment of 14 hypocretin cells recorded. A reduction of the
response to ATP is evident after 30 s of continuous drug application. C:
dose–response curve for the ATP-induced increase in firing rate. D: representative trace showing the membrane depolarization on ATP application in the
presence of 0.5 ␮⌴ TTX. E: in voltage clamp (at – 60 mV holding potential),
ATP induced a substantial inward current and an increase in the thickness of
baseline; representative trace. To isolate the ATP effect from synaptic modulation, TTX (0.5 ␮⌴), DL-2-amino-5-phosphonopentanoic acid (AP-5; 50
␮⌴), 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX; 10 ␮⌴), and bicuculline
methiodide (BIC; 30 ␮⌴) were bath-applied. F: current–voltage relationship
shows a moderate decrease in input resistance on ATP application. Responses
were normalized to the voltage resulting from a ⫺140-pA pulse. A–F: ATP
100 ␮M if not otherwise noted.
J Neurophysiol • VOL
94 • SEPTEMBER 2005 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on October 1, 2016
Many brain regions express a variety of ATP receptors that
underlie purinergic signaling. Recent reports have highlighted
the hypothalamus as a locus for both strong P2X receptor
expression as well as functional purinergic transmission (Jo
and Role 2002; Xiang et al. 1998). Here, we tested whether the
direct excitatory effects of ATP on hypocretin cells were
mediated by activation of P2X receptors. In the presence of the
P2X receptor antagonist suramin (100 ␮M), the ATP-induced
increase in spike frequency was substantially reduced (Fig.
2A). In these experiments ATP responses were measured in the
same cell before and during suramin application. On four cells
tested, ATP (100 ␮M) induced an increase in spike frequency
by 45.0 ⫾ 7.9% (P ⬍ 0.001) in control conditions and by
15.9 ⫾ 5.2% in the presence of suramin (100 ␮M). This marks
a net suppressive effect of suramin of 59.9 ⫾ 13.9%, a
statistically significant inhibition (Fig. 2B; P ⬍ 0.05; ANOVA
followed by Bonferroni post hoc test for multiple comparisons). After suramin washout, ATP induced excitatory responses similar to those before suramin application (n ⫽ 3;
data not shown). The application of suramin (100 ␮M) alone
did not significantly alter the spike frequency (increase by
4.7 ⫾ 5.4%; P ⫽ 0.48; n ⫽ 4; ANOVA).
Alhough suramin is widely used to study purinergic transmission, some reports have suggested that suramin might also
exert an inhibitory effect on glutamatergic transmission (Gu et
al. 1998). We therefore tested the P2X receptor antagonist
PPADS in additional experiments. In four cells tested, ATP
increased spike frequency by 40.2 ⫾ 9.1% (P ⬍ 0.01) under
control conditions, and this was reduced to 9.2 ⫾ 3.6% in the
presence of PPADS (50 ␮M; Fig. 2, C and D), a significant
inhibition of 78.5 ⫾ 9.0% (P ⬍ 0.05; ANOVA with Bonferroni
post hoc test). Application of PPADS alone had no detectable
effect on spike frequency (increase by 4.6 ⫾ 7.6%; P ⫽ 0.56;
ANOVA). In control experiments (n ⫽ 5), repeated application
of ATP resulted in consistent peak excitatory responses without signs of a significant peak run-down (data not shown),
confirming that the inhibition of the ATP-induced spike fre-
2198
G. WOLLMANN, C. ACUNA-GOYCOLEA, AND A. N. VAN DEN POL
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on October 1, 2016
FIG. 2. P2X receptor antagonists inhibit ATP action on hypocretin cells. A: in the presence of suramin (100 ␮M), the excitatory effect of ATP (100 ␮M) on
the firing rate was substantially inhibited; representative trace. B: bar graph summarizing the effect of P2X receptor blockage with suramin on the ATP-induced
excitation; n ⫽ 4. C: in the presence of pyridoxal phosphate-6-azo(benzene-2,4-disulfonic acid) tetrasodium salt (PPADS; 50 ␮M), ATP (100 ␮M) was unable
to excite hypocretin cells. D: bar graph summarizing the effect of ATP in the absence and presence of PPADS on 4 cells. Statistically significant: *P ⬍ 0.05,
***P ⬍ 0.001. E: PPADS (50 ␮M) effectively inhibited the ATP-induced (100 ␮M) inward current, typical trace. F: summary of the PPADS effect on
ATP-induced inward current. *P ⬍ 0.05.
quency increment by suramin and PPADS was in fact caused
by the block of purinergic P2X receptors. The presence of
PPADS also reduced the ATP-induced inward current of
hypocretin cells held at – 60 mV in voltage clamp (Fig. 2E).
J Neurophysiol • VOL
In 6 cells tested, ATP (100 ␮M) induced an inward current
of 14.6 ⫾ 2.5 pA, which was significantly reduced to 6.8 ⫾
1.4 pA in the presence of 50 ␮M PPADS (P ⬍ 0.05,
ANOVA; Fig. 2F). A recovery of the ATP-induced inward
94 • SEPTEMBER 2005 •
www.jn.org
ATP-P2X RECEPTORS IN HYPOCRETIN/OREXIN NEURONS
the view that external ATP excites hypocretin cells through
purinergic P2X receptors and suggest the involvement of P2X2
subunits.
Effect of the nonhydrolysable ATP agonist ATP-␥-S
on hypocretin cells
Extracellular ATP is subject to rapid hydrolysis by surfacelocated ectonucleotidases (reviewed in Zimmermann 1996); in
addition, ATP can have significant effects intracellularly. We
examined the effect of ATP-␥-S, a nonhydrolysable ATP
analog, on the activity of hypocretin neurons. Experiments
were carried out with local application of ATP-␥-S (100 ␮M)
followed by ATP (100 ␮M) after a 2- to 3-min wash-out
interval. In all cells recorded, ATP-␥-S induced a robust
excitatory response comparable to ATP application. Under
current clamp, spike frequency increased by 23.8 ⫾ 7.3% after
100 ␮M ATP-␥-S application. In the same group of cells, the
frequency of action potentials was increased by 25.8 ⫾ 9.2%
by ATP (100 ␮M), a nonsignificant difference between ATP
and ATP-␥-S (P ⫽ 0.87; n ⫽ 5; Fig. 4, A1 and A3). Although
the peak value for this spike frequency increment was slightly
higher after ATP application, the mean increase over the whole
1-min application period was slightly higher after ATP-␥-S
application (12.1 ⫾ 2.6 vs. 9.7 ⫾ 5.9% for ATP; Fig. 4A3). A
comparison of a typical time-course is shown in Fig. 4A2. The
action of ATP-␥-S was also subject to a considerable rundown,
suggesting that ATP hydrolysis does not account for the full
attenuation. We further compared the inward current induced
by ATP-␥-S and ATP in voltage clamp (Fig. 4B1) in normal
ACSF solution. Application of ATP-␥-S generated an inward
current of 15.1 ⫾ 1.8 pA and ATP generated an 11.3 ⫾ 2.1-pA
current (P ⫽ 0.2; n ⫽ 6; ANOVA; Fig. 4B2).
FIG. 3. P2X subunit identification. A: selective P2X1 and P2X3 agonist ␣,␤-MeATP did not mimic ATP actions on hypocretin cells. A1: representative trace
of inward current induction by ATP, but not by ␣,␤-MeATP. A2: bar graph summarizing the inward current induced by ATP (100 ␮M) before (ATP 1) and after
(ATP 2) ␣,␤-MeATP (100 ␮M) application, which did not induce a sizable response. **P ⬍ 0.01. A3: increase in spike frequency induced by ATP (100 ␮M)
but not by ␣,␤-MeATP (100 ␮M). **P ⬍ 0.01. B: decrease in bath pH potentiates ATP current, a characteristic feature of P2X2 subunits. B1: typical trace of
a cell held at – 60 mV in voltage clamp with 2 consecutive ATP applications (100 ␮M) in normal (pH 7.4) and acidified (pH 6.8) artificial cerebrospinal fluid
(ACSF). B2: bar graph shows increase of the ATP-induced inward current under low pH conditions compared with normalized current under control conditions.
**P ⬍ 0.01. Inset: individual current increase of each of 5 cells tested in absolute values.
J Neurophysiol • VOL
94 • SEPTEMBER 2005 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on October 1, 2016
current was seen in four cells tested after PPADS wash-out
(⬎5 min).
Although suramin and PPADS block P2X receptors, they
may also inhibit metabotropic P2Y receptors. To rule out that
the ATP-induced activation of hypocretin cells is mediated
through P2Y receptors and to examine the subtype composition
of the P2X receptors involved, we studied the effect of ␣,␤MeATP, a selective P2X1 and P2X3 agonist (Lambrecht 2000)
and the effect of bath acidification, which potentiates currents
through P2X2 receptors (North 2002; Stoop et al. 1997). The
effect of ␣,␤-MeATP was assessed in both voltage and current
clamp. In six cells tested, ␣,␤-MeATP (100 ␮M) failed to
induce a sizable inward current (1.6 ⫾ 0.6 pA, n ⫽ 6; Fig. 3,
A1 and A2). The same cells responded to ATP (100 ␮M) with
significant inward currents of 9.9 ⫾ 2.1 pA when applied 2–3
min before ␣,␤-MeATP and 13.2 ⫾ 3.9 pA when applied 2
min after ␣,␤-MeATP wash-out (Fig. 3A2, column ATP 1 and
ATP 2, respectively). The effects of ATP and ␣,␤-MeATP
application on inward currents differed statistically (P ⬍ 0.01;
n ⫽ 6; ANOVA with Bonferroni post hoc test). In current
clamp and holding the cells at resting membrane potential,
application of ␣,␤-MeATP (100 ␮M) did not significantly
change the spike frequency (4.2 ⫾ 3.5%; n ⫽ 7) compared
with a significant increase by 28.7% ⫾ 5.2 after ATP (100 ␮M)
application (P ⬍ 0.01; ANOVA with Bonferroni post hoc test;
Fig. 3A3).
We further tested whether the ATP-induced activation of
hypocretin cells was pH-dependent. In five cells examined, the
mean inward current amplitude on ATP application increased
from 12.6 ⫾ 2.5 pA under normal pH (7.4) conditions to
19.2 ⫾ 4.0 pA in acidified ACSF medium (pH 6.8; Fig. 3B1),
a significant mean increase by 57.3 ⫾ 14.0% (P ⬍ 0.01; n ⫽
5; ANOVA; Fig. 3B2; inset). Together these data substantiate
2199
2200
G. WOLLMANN, C. ACUNA-GOYCOLEA, AND A. N. VAN DEN POL
In addition, we tested whether ectonucleotidase activity
may inhibit any intrinsic ATP action on hypocretin cells. In
seven cells examined, local application of the ectonucleotidase
inhibitor 6-N,N-diethyl-␤-␥-dibromomethylene-D-adrenosineS-triphosphatase 67156 (100 ␮M) did not change the spontaneous spike frequency (103.9 ⫾ 4.5%; n ⫽ 7; not significant;
ANOVA; data not shown). The same cells responded to subsequent ATP application (100 ␮M) by a mean increase in spike
frequency of 28.5 ⫾ 5.9%. This may suggest that under these
experimental conditions, the intrinsic tone of ATP-mediated
purinergic transmission on hypocretin cells is low.
Synaptic input and membrane activity in hypocretin cells
GABA appears to be the primary inhibitory transmitter of
the hypothalamus (Decavel and van den Pol 1990). To study
the effect of ATP application on GABA-mediated synaptic
transmission, we used a KCl pipette solution and bath-applied
AP5 (50 ␮M) and CNQX (10 ␮M) to block the excitatory
ionotropic glutamate receptor– dependent synaptic activity.
Under these conditions, voltage-clamp experiments at – 60-mV
holding potential revealed well-defined brief inward currents
characteristic of IPSCs. Application of BIC (30 ␮M) completely abolished these IPSCs, confirming that they were attributable to GABAA receptor subtype activation (n ⫽ 5; data
not shown). To determine whether ATP modulates GABAergic
transmission in hypocretin cells, we analyzed the frequency of
IPSCs before, during, and after ATP application. Representative traces of a typical experiment are shown in Fig. 5A1. ATP
(100 ␮M) did not change the frequency of IPSCs in hypocretin
cells (97.8 ⫾ 9.6% of control; 107.1 ⫾ 13.5% recovery; P ⫽
0.78; n ⫽ 9; ANOVA with Bonferroni post hoc test; Fig. 5A3).
Figure 5A2 shows the mean time-course of this IPSC frequency
J Neurophysiol • VOL
analysis. We further studied mIPSCs in the presence of TTX
(0.5 ␮M) in the external solution. Representative traces are
shown in Fig. 5B1. ATP did not significantly reduce the mIPSC
frequency (88.1 ⫾ 8.6% of control; 98.4 ⫾ 21.2% recovery;
P ⫽ 0.79; n ⫽ 6; ANOVA; Fig. 5B2). In addition, the mIPSC
amplitude pattern before, during, and after ATP application
was compared. The mean amplitude of all events was 37.3 ⫾
6.4, 40.3 ⫾ 6.1, and 44.5 ⫾ 8.2 pA for control, ATP, and
washout, respectively (P ⫽ 0.77; not significant; n ⫽ 5;
ANOVA). The cumulative distribution of the mIPSC amplitude revealed also no significant effect of ATP application
(P ⬎ 0.05; n ⫽ 5; Kolmorogov-Smirnov). Figure 5B3 shows a
typical result of such a comparison.
The decay kinetics of mIPSCs were compared before, during, and after ATP application. Detected mIPSCs were averaged and subject to an exponential curve fitting, which revealed the time constant for the event decay. Figure 5C1 shows
a typical example of such a comparison with the averaged trace
in black, superimposed on the detected raw events in gray. In
five cells tested, we found no change of the decay kinetics of
mIPSCs with ATP (100 ␮M) application (decay time constant
␶ ⫽ 15.1 ⫾ 1.2, 14.9 ⫾ 1.0, and 14.5 ⫾ 1.1 ms for control,
ATP, and wash-out, respectively; not significant, P ⫽ 0.923;
n ⫽ 5; ANOVA; Fig. 5C2). In addition to the analysis of
spontaneous inhibitory synaptic inputs, we tested whether
electrically evoked potentials have a BIC-insensitive component that may be attributed to a purinergic effect. This has been
previously reported in cultured developing hypothalamic neurons from chickens (Jo and Role 2002). However, in our
postnatal brain slice preparation model in the presence of 10
␮M CNQX and 50 ␮M AP5 to block glutamatergic excitatory
transmission, evoked potentials were completely abolished in
the presence of 30 ␮M BIC, suggesting that they were entirely
94 • SEPTEMBER 2005 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on October 1, 2016
FIG. 4. Nonhydrolysable ATP analogue ATP-␥-S excites hypocretin cells. A1: representative traces showing the increase of spike frequency by ATP-␥-S (100
␮M) and ATP (100 ␮M) at the same cell. A2: time-course of the excitatory action of ATP and ATP-␥-S corresponding to A1. Effect of ATP on spike frequency
lasted slightly longer than the one evoked by native drug. A3: bar graph comparing peak values with 1-min mean values for the increase in firing rate of 5 cells.
B1: in voltage clamp, ATP-␥-S application induced an inward current similar, but longer-lasting, than the one induced by ATP; representative traces of current
responses in the same cell. B2: bar graph showing mean inward currents induced by ATP-␥-S and ATP; n ⫽ 6; n.s., not significant.
ATP-P2X RECEPTORS IN HYPOCRETIN/OREXIN NEURONS
2201
GABA-mediated (n ⫽ 5; data not shown), confirming observations from previous studies on hypocretin cells (AcunaGoycolea et al. 2004; Fu et al. 2004) that evoked potentials are
abolished by antagonists of glutamatergic and GABAergic
transmission. Together, these results suggest that ATP does not
modulate GABAergic transmission to hypocretin neurons.
When ATP was applied to hypocretin cells, an increase in
membrane electrical activity was found. To test whether this
was caused by a direct effect on the membrane or by excitatory
synaptic input, we used KMeSO4 pipette solution and bathapplied BIC (30 ␮M) to block GABA-mediated inhibitory
events. Under these conditions, excitatory postsynaptic currents (EPSCs) were detected as fast inward currents at – 60 mV
holding potentials (Fig. 6A). ATP (100 ␮M) caused a substantial increase in whole cell channel activity (Fig. 6A), and this
remained when TTX was added (Fig. 6B), indicating a spikeindependent nature of this ATP effect. To further determine
whether this ATP effect of increased whole cell channel
activity involved modulation of synaptic currents, we first
blocked both GABA and glutamate mediated synaptic activity
with BIC (30 ␮M), AP-5 (50 ␮M), and CNQX (10 ␮M). In the
presence of these antagonists and TTX (0.5 ␮M), hypocretin
cells did not show any synaptic activity. Application of ATP
J Neurophysiol • VOL
still resulted in a similar increase in background activity as
seen above, indicating that this ATP effect is independent of
activation of ionotropic GABA and glutamate receptors (Fig.
6C). Finally, inhibiting synaptic transmitter release by introduction of a nominally Ca2⫹-free extracellular solution did also
not abolish the ATP-induced increase in background channel
activity (Fig. 6D), supporting the view that a direct activation
of purinergic P2X receptors causes this increase in membrane
activity. As the large increase in membrane electrical activity
obscured the response to ATP of EPSCs, these were not further
analyzed.
We found no indication that exogenously applied ATP alters
the inhibitory synaptic tone at hypocretin cells. These data
further substantiate the view that the excitatory actions of ATP
on these cells are based on direct activation of P2X receptors
at these cells.
Na⫹-dependence of the ATP-induced current in
hypocretin cells
To study the ionic mechanisms underlying the ATP-mediated depolarization, we first tested the involvement of extracellular Na⫹. To this end, the ATP response was recorded in
94 • SEPTEMBER 2005 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on October 1, 2016
FIG. 5. ATP does not affect inhibitory synaptic inputs to hypocretin cells. A1: example traces of a voltage-clamp experiment showing the unaltered occurrence
of inhibitory postsynaptic currents (IPSCs) before, during, and after ATP (100 ␮M) application. A pipette solution containing KCl was used in these experiments;
holding potential, – 60 mV. A2: mean time-course of IPSCs from 9 cells recorded. ATP application did not alter the frequency of these events. A3: bar graph
showing lack of IPSC modulation by ATP. B1: in the presence of 0.5 ␮M TTX, miniature (m)IPSCs were recorded. Representative traces showing no significant
effect of ATP application on mIPSC occurrence. IPSCs and mIPSCs were BIC sensitive, indicating that they were GABA-mediated currents. B2: summary of
mIPSC frequency analysis, no effect of ATP application. B3: cumulative distribution of the mIPSC amplitude was not affected by ATP. Typical result of 1 of
5 cells tested (P ⬍ 0.05). C1: unchanged mIPSC decay kinetics after ATP application. Averaged traces (black) superimposed on raw traces, representative
example with decay time constant. C2: summary of decay kinetic analysis of 5 cells. n.s., not significant.
2202
G. WOLLMANN, C. ACUNA-GOYCOLEA, AND A. N. VAN DEN POL
FIG. 6. ATP actions on background channel activity in hypocretin cells. A:
typical trace showing increase in background channel activity on ATP application, obscuring the analysis of synaptic currents. This experiment was done
in the presence of BIC (30 ␮M) to block inhibitory synaptic input and clamped
at – 60 mV. B: ATP-induced effect on whole cell channel activity persisted in
the presence of TTX (0.5 ␮M) in the bath, suggesting that it was not
attributable to spike-dependent activity. C: when glutamate- and GABAmediated activity was blocked (using AP5/CNQX and BIC, respectively), ATP
still induced an increase in membrane noise, suggesting mechanisms independent from excitatory postsynaptic current (EPSC) modulation. Traces show a
representative experiment. D: ATP-mediated increase in background channel
activity persisted under nominally Ca2⫹-free/high magnesium ACSF (conditions that attenuate release of neurotransmitter), further substantiating the
independent nature of this ATP effect from GABA or glutamate release.
Representative trace from a series of 8 cells recorded. A–D: ATP 100 ␮M if not
otherwise noted.
voltage clamp before and during choline chloride substitution
for extracellular NaCl. Sodium replacement became effective
⬃10 min after switching the bath inflow and was detected by
the fading and then absence of action potentials. Figure 7A
displays a representative trace before and after exchange of the
bath solution. Under these conditions, the mean inward current
induced by local ATP application dropped significantly from
20.6 ⫾ 4.1 (in Na⫹) to 4.9 ⫾ 1.6 pA (in choline) (P ⬍ 0.01;
n ⫽ 9; ANOVA; Fig. 7B), indicating that Na⫹-ions are involved in this ATP-induced inward current. To determine the
reversal potential of this current, we applied slow voltage
J Neurophysiol • VOL
7. Na⫹-dependence and reversal potential of ATP-induced current. A:
ATP-induced inward current in hypocretin neurons (trace 1) was nearly
abolished after exchange of the NaCl bath solution for choline chloride, traces
of a representative experiment. B: bar graph summarizing effect of choline
substitution on ATP-induced inward current of 9 cells. In nominally Ca2⫹-free
external solution, amplitude of ATP-mediated inward current was not significantly decreased. C: reversal potential for net ATP-induced current was near
⫺30 mV in this representative experiment. Voltage-ramp protocols (from ⫺60
to 0 mV during 5 s) were used to determine reversal potential of ATP-mediated
current in hypocretin cells. Together, these results are consistent with ATP
activation of nonselective cation channels in hypocretin cells. *P ⬍ 0.001.
FIG.
94 • SEPTEMBER 2005 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on October 1, 2016
ramps (5 s) from ⫺60 to 0 mV to identified hypocretin
neurons. A Cs-based pipette solution and 0.5 ␮M TTX in the
bath were used in these experiments to prevent the activation of
potassium and sodium channels with depolarizing protocols.
The reversal potential for the net ATP-mediated current was
⫺27.6 ⫾ 3.5 mV (range: ⫺36.4 to ⫺14.6 mV; n ⫽ 6). This
value was negative to the expected Na⫹ reversal potential, but
positive to the K⫹ reversal potential, and was suggestive of a
nonselective cation current. A typical experiment showing the
reversal potential for the ATP-induced current is presented in
Fig. 7C.
In addition, studies in Ca2⫹-free conditions revealed that the
ATP-induced inward current is not altered under nominal 0
mM extracellular Ca2⫹, consistent with the characteristics of a
Na⫹-dependent nonselective cation channel (Liu et al. 2002).
In Ca2⫹-free extracellular conditions and high extracellular
Mg2⫹, ATP (100 ␮M) induced a mean inward current of
16.1 ⫾ 4.1 pA (n ⫽ 7; Fig. 7B), which was not significantly
different (P ⫽ 0.454; ANOVA) from inward currents under
ATP-P2X RECEPTORS IN HYPOCRETIN/OREXIN NEURONS
2203
normal ACSF conditions. Together, these results suggest that
ATP activates a nonselective cation current. Such a current has
been previously reported with ATP receptor activation in other
regions of the brain (Edwards 1994).
Is there a purinergic component in evoked excitatory
potentials on hypocretin cells?
To evaluate the possibility that purinergic transmission
might be involved in electrically evoked excitatory synaptic
potentials, cells were current-clamped with a slightly negative
current corresponding to a resulting membrane voltage of – 65
to ⫺75 mV. A bipolar stimulation electrode placed in the
lateral hypothalamus was used to generate controlled electrical
stimuli (10 –150 ␮A, 0.5 ms, 0.2 Hz) as described previously
(Acuna-Goycolea et al. 2004). BIC (30 ␮M) was bath-applied
to abolish inhibitory GABA-mediated evoked potentials. Local
application of suramin (100 ␮M) reduced the time-voltage
integral (area) of the evoked potentials of 10 cells to 70.3 ⫾
6.9% of control (83.9 ⫾ 8.3% recovery), a statistically significant effect (P ⬍ 0.01: n ⫽ 10; ANOVA; Fig. 8B), suggesting
ATP might be release from axonal terminals. As a control, we
examined ionotropic glutamate receptor antagonists under the
same conditions. Application of NMDA and AMPA receptor
antagonists AP-5 (50 ␮M) and CNQX (10 ␮M) completely
blocked evoked potentials (Fig. 8A), consistent with the view
that they are attributable to glutamate release as was shown
previously (Acuna-Goycolea et al. 2004). The inhibition of
evoked potentials by suramin might therefore be attributable to
suramin effects on glutamate transmission, as has been reported in spinal dorsal horn neurons (Gu et al. 1998; Lambrecht 2000). We therefore used a different receptor antagonist,
PPADS, to study the effect on evoked excitatory potentials.
Application of 50 ␮M PPADS did not change the time-voltage
integral of evoked potentials in hypocretin cells (97.6 ⫾ 8.2%
of control; P ⫽ 0.83; n ⫽ 11; ANOVA; Fig. 8C). Together,
these data indicate that the primary transmitter released during
an excitatory evoked potential was probably glutamate. The
partial blockade of the evoked excitatory postsynaptic potential
(EPSP) by suramin but only a modest statistically insignificant
reduction by PPADS suggests that, although a small amount of
ATP may be released, the block could also be caused by a
nonselective effect of suramin on glutamate receptors.
J Neurophysiol • VOL
DISCUSSION
In this study, we studied the effect of extracellular ATP on
the activity of GFP-expressing hypocretin cells in mouse hypothalamic slices using whole cell patch-clamp recordings.
Local application of ATP induced a substantial excitatory
action on hypocretin cells with a positive response in 154 of
170 recorded cells (89.6% response rate), showed by membrane depolarization, an increase in spike frequency, a decrease
in input resistance, and an induction of a Na⫹-dependent
inward current that reversed near ⫺27 mV. This excitatory
effect was direct and postsynaptic in nature and sensitive to the
ATP receptor antagonists PPADS and suramin. Application of
the nonhydrolyzable ATP agonist ATP-␥-S mimicked these
excitatory action, further substantiating activation of extracellular receptors. Insensitivity to ␣,␤-MeATP and pH dependence of the inward current suggest involvement of purinergic
P2X receptors that include P2X2 subunits.
Direct excitation of hypocretin cells by ATP through
activation of P2X receptors
In the CNS, ATP can modulate synaptic transmission
through the activation of ionotropic (P2X family) or metabotropic (P2Y family) receptor subtypes (Edwards and Gibb
1993). ATP can also exert an intracellular effect on membrane
K⫹ channel activity (Ballanyi 2004; Liss and Roeper 2001).
P2X receptors show a prominent expression throughout the
hypothalamus (Kanjhan et al. 1999; Xiang et al. 1998). Here
we found that local ATP application increased the spike frequency of hypocretin cells in a dose-dependent fashion. This
excitatory reaction was independent of synaptic input, because
even in the presence of TTX, ATP produced depolarization and
an inward current. In addition, these effects correlated with a
decrease in input resistance suggestive of an opening of ion
channels in the postsynaptic membrane.
When choline was exchanged for sodium in the external
solution, the ATP-induced inward current was substantially
suppressed. A small inward current remained under nominally
Na⫹-free conditions and might reflect a Ca2⫹-component of
the ATP-induced current, as previously reported in dorsomedial hypothalamic neurons (Matsumoto et al. 2004). The reversal potential for this inward current was near ⫺27 mV, part
94 • SEPTEMBER 2005 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on October 1, 2016
FIG. 8. Suramin, but not PPADS, attenuates evoked excitatory potentials. A: in this representative neuron, suramin (100 ␮M) reversibly decreased evoked
potentials in a hypocretin cells. At the end of this experiment, the evoked potentials were almost completely blocked by ionotropic glutamate receptor antagonists
AP5 (50 ␮M) and CNQX (10 ␮M) in the bath, suggesting that they were attributable to synaptic release of glutamate onto the recorded cell. B: bar graph
summarizing inhibitory effect of 100 ␮M suramin on evoked potentials in hypocretin cells. C: PPADS (50 ␮M), another P2X receptor antagonist, did not
significantly reduce the evoked potentials (n ⫽ 11), as shown in this bar graph.
2204
G. WOLLMANN, C. ACUNA-GOYCOLEA, AND A. N. VAN DEN POL
J Neurophysiol • VOL
those subunits (North 2002; Stoop et al. 1997). Our data show
clearly an increase in ATP current amplitude when the extracellular medium is acidified to a pH of 6.8, thereby indicating
ionotropic P2X2 receptor involvement and excluding mediation through metabotropic P2Y receptors.
Additional support for the nature of P2X2 receptor expression can be found in the desensitization kinetics of the ATP
effects. Compared with P2X1 and P2X3, which desensitize
rapidly (time constant, ␶ ⬍ 100 ms), receptors composed of
P2X2 subunits are considered slow desensitizing, with ␶ ⬎ 10 s
(North 2002). Also, P2X1 and P2X3 subunits require a considerable recovery time after ATP application for a reproducible
response (⬎15 min), whereas P2X2 subunits can mediate ATP
currents on repetitive agonist application in the frequency
range of seconds, even responding with an increased current
amplitude (Khakh et al. 2001; North 2002). Figures 1, 2, and 4
clearly show the slow but consistent desensitization of the ATP
response on hypocretin cells. In addition, repetitive ATP application throughout our studies resulted in reproducible current responses as exemplified in Fig. 3A2. Consistent with the
idea of P2X2 receptor involvement, the amplitude of the
second ATP application was increased compared with the
initial response. Taken together, the data suggest that hypocretin cells express predominantly P2X2 subunits. However, the
expression of other subunits as part of heteromeric P2X channels cannot be ruled out and may account for the incomplete
inhibition of the ATP action by PPADS and suramin (Jo and
Role 2002). Our data are consistent with previous reports that
at least three different ionotropic purinoceptors are expressed
in the hypothalamus: P2X1, P2X2, and P2X3 (Gurin et al. 2003;
Kanjhan et al. 1999; Xiang et al. 1998). We previously showed
that identified hypothalamic melanin-concentrating hormone
(MCH) cells are also excited by ATP (van den Pol et al. 2004).
However, the subunit composition of ATP receptors in those
cells was not addressed.
In other parts of the brain, ATP has both direct and indirect
effects on synaptic activity (Bowser and Khakh 2004). In
contrast to the strong direct excitatory action of ATP on
hypocretin cells, analysis of both spontaneous and miniature
synaptic activity did not reveal any detectable effect on the
synaptic input to hypocretin cells by ATP. Hypocretin neurons
do show an increase in electrical activity on ATP stimulation,
but because this activity was blocked by choline substitution
for sodium but not by GABA and glutamate receptor antagonists, it was most likely caused by an increase in somatic ion
channel activity independent of synaptic activity.
Source of ATP and functional considerations
As described above, hypocretin cells show robust excitatory
responses to extracellular ATP. What might be potential
sources of ATP that could stimulate hypocretin neurons? An
interesting report showed that ATP may be released by developing GABAergic axons of lateral hypothalamic cells during a
period when GABA has excitatory effects caused by the
elevated Cl-reversal potential (Jo and Role 2002). In the
presence of the ATP receptor antagonist suramin, evoked
potentials recorded in hypocretin neurons were diminished in
amplitude; however, when a different ATP receptor antagonist,
PPADS, was used, only a minor depression was noted.
Whereas this could suggest a low level of ATP was released
94 • SEPTEMBER 2005 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on October 1, 2016
way between the values suggested by the Nernst equation for
reversal of Na⫹ or K⫹ currents, suggesting a mixed current,
probably a nonselective cation current, a type that has been
reported elsewhere for ATP actions (Edwards 1994; Sorimachi
et al. 2001). In parallel, G protein– coupled nonselective cation
currents underlie the response of hypocretin cells to glucagonlike peptide 1 (Acuna-Goycolea and van den Pol 2004). Finally, use of a nominally Ca2⫹-free extracellular buffer did not
block the current, suggesting that it was not primarily caused
by a calcium dependent mechanism, for instance a sodium/
calcium exchanger (Liu et al. 2002).
The response of hypocretin neurons to ATP was reduced
during the course of continuous application (see Fig. 1B). One
mechanism that might account for this effect is the action of
ectonucleotidases, which can rapidly degrade extracellular
ATP (Zimmermann 1996). The nonhydrolysable ATP analog,
ATP-␥-S, induced a depolarization and an inward current that
lasted slightly longer than the one evoked by native ATP.
However, even in the ongoing presence of ATP-␥-S we found
a reduction in the postsynaptic responses, suggesting that
receptor desensitization mechanisms also contribute to this
effect (Ralevic and Burnstock 1998). Activity of ectonucleotidases has also been shown to diminish the action of endogenously released ATP (Westfall et al. 1996). However, using
the ectonucleotidase inhibitor ARL 67156, we did not find any
effect on the spontaneous spike frequency, suggesting a low
intrinsic tone of extracellular ATP on hypocretin cells.
The question of what receptor class and subunit composition
may mediate the ATP actions on hypocretin cells was addressed pharmacologically by using a number of purinergic
agonists and antagonists. P2X channels form homomultimers
or heteromers and different subtypes can be coexpressed in the
same cell. Among the seven P2X subunits, P2X1–7, that have
been described, CNS receptors seem to be predominantly
comprised of P2X1, P2X2, P2X3, P2X4, and P2X4/6 (Illes and
Ribeiro 2004; Norenberg and Illes 2000; Robertson et al.
2001). On the other hand, 10 members of the G protein–
coupled P2Y receptor family have been described, of which a
number have been detected in the CNS (Illes and Ribeiro
2004). Our results show that postsynaptic ATP actions could
be blocked by two antagonists of purinergic transmission,
suramin and PPADS. Although these antagonists appear to be
nonselective in terms of P2X and P2Y receptor specificity
(Ralevic and Burnstock 1998), a number of P2X and P2Y
subunits, i.e., P2X4, P2X6, P2X7, P2X4/6, P2Y2, P2Y4, P2Y6,
and P2Y11, can be excluded based on their insensitivity toward
these drugs (Lambrecht 2000; North and Suprenant 2000).
Molliver et al. (2002) showed a suramin-sensitive P2Y2 receptor mediated depolarization in sensory neurons.
In this regard, additional tests were used to study purinergic
receptors on hypocretin cells. The nonhydrolyzable ATP agonist ATP-␥-S has been shown to be active at homomeric
P2X1– 6 and heteromeric P2X2/3 and P2X1/5 subunits as well as
on P2Y1, P2Y2, and P2Y11. In contrast, ␣,␤-MeATP is a potent
agonist only at P2X1 and P2X3 subunits and heteromeric
combinations thereof (Lambrecht 2000). Because ␣,␤-MeATP
was ineffective on hypocretin cells, those subunits can also be
excluded, suggesting an involvement of either P2X2 or P2Y1.
Although no specific agonists or antagonists for P2X2 are
available, the potentiation of the ATP current in the presence of
an acidic environment seems to be a characteristic feature of
ATP-P2X RECEPTORS IN HYPOCRETIN/OREXIN NEURONS
ACKNOWLEDGMENTS
We thank V. Rougulin for technical assistance with transgenic mice.
GRANTS
This work was supported by National Institute of Neurological Disorders
and Stroke Grants NS-34887 and NS-41454.
REFERENCES
Acuna-Goycolea C, Li Y, and van den Pol AN. Group III metabotropic
glutamate receptors maintain tonic inhibition of excitatory synaptic input to
hypocretin/orexin neurons. J Neurosci 24: 3013–3022, 2004.
J Neurophysiol • VOL
Acuna-Goycolea C and van den Pol A. Glucagon-like peptide 1 excites
hypocretin/orexin neurons by direct and indirect mechanisms: implications
for viscera-mediated arousal. J Neurosci 24: 8141– 8152, 2004.
Ballanyi K. Protective role of neuronal KATP channels in brain hypoxia. J
Exp Biol 207: 3201–3212, 2004.
Bardoni R, Goldstein PA, Lee CJ, Gu JG, and MacDermott AB. ATP P2X
receptors mediate fast synaptic transmission in the dorsal horn of the rat
spinal cord. J Neurosci 17: 5297–5304, 1997.
Bekkers JM and Stevens CF. Quantal analysis of EPSCs recorded from small
numbers of synapses in hippocampal cultures. J Neurophysiol 73: 1145–
1156, 1995.
Bowser DN and Khakh BS. ATP excites interneurons and astrocytes to
increase synaptic inhibition in neuronal networks. J Neurosci 24: 8606 –
8620, 2004.
Brockhaus J, Dressel D, Herold S, and Deitmer JW. Purinergic modulation
of synaptic input to Purkinje neurons in rat cerebellar brain slices. Eur
J Neurosci 19: 2221–2230, 2004.
Chemelli RM, Willie JT, Sinton CM, Elmquist JK, Scammell T, Lee C,
Richardson JA, Williams SC, Xiong Y, Kisanuki Y, Fitch TE, Nakazato
M, Hammer RE, Saper CB, and Yanagisawa M. Narcolepsy in orexin
knockout mice: molecular genetics of sleep regulation. Cell 98: 437– 451,
1999.
Chen CT, Dun SL, Kwok EH, Dun NJ, and Chang JK. Orexin A-like
immunoreactivity in the rat brain. Neurosci Lett 260: 161–164, 1999.
Ciccarelli R, Ballerini P, Sabatino G, Rathbone MP, D’Onofrio M, Caciagli F, and Di Iorio P. Involvement of astrocytes in purine-mediated
reparative processes in the brain. Int J Dev Neurosci 19: 395– 414, 2001.
Cunha RA and Ribeiro JA. ATP as a presynaptic modulator. Life Sci 68:
119 –137, 2000.
Decavel C and van den Pol AN. GABA: a dominant neurotransmitter in the
hypothalamus. J Comp Neurol 302: 1019 –1037, 1990.
de Lecea L, Kilduff TS, Peyron C, Gao X, Foye PE, Danielson PE,
Fukuhara C, Battenberg EL, Gautvik VT, Bartlett FS 2nd, Frankel
WN, van den Pol AN, Bloom FE, Gautvik KM, and Sutcliffe JG. The
hypocretins: hypothalamus-specific peptides with neuroexcitatory activity.
Proc Natl Acad Sci USA 95: 322–327, 1998.
Edwards FA. ATP receptors. Curr Opin Neurobiol 4: 347–352, 1994.
Edwards FA and Gibb AJ. ATP–a fast neurotransmitter. FEBS Lett 325:
86 – 89, 1993.
Edwards FA, Gibb AJ, and Colquhoun D. ATP receptor-mediated synaptic
currents in the central nervous system. Nature 359: 144 –147, 1992.
Edwards FA, Robertson SJ, and Gibb AJ. Properties of ATP receptormediated synaptic transmission in the rat medial habenula. Neuropharmacology 36: 1253–1268, 1997.
Fields RD and Stevens B. ATP: an extracellular signaling molecule between
neurons and glia. Trends Neurosci 23: 625– 633, 2000.
Fu LY, Acuna-Goycolea C, and van den Pol AN. Neuropeptide Y inhibits
hypocretin/orexin neurons by multiple presynaptic and postsynaptic mechanisms: tonic depression of the hypothalamic arousal system. J Neurosci 24:
8741– 8751, 2004.
Gao XB and van den Pol AN. Melanin concentrating hormone depresses
synaptic activity of glutamate and GABA neurons from rat lateral hypothalamus. J Physiol 533: 237–252, 2001.
Gerashchenko D and Shiromani PJ. Effects of inflammation produced by
chronic lipopolysaccharide administration on the survival of hypocretin
neurons and sleep. Brain Res 1019: 162–169, 2004.
Gu JG, Bardoni R, Magherini PC, and MacDermott AB. Effects of the
P2-purinoceptor antagonists suramin and pyridoxal-phosphate-6-azophenyl2⬘,4⬘-disulfonic acid on glutamatergic synaptic transmission in rat dorsal
horn neurons of the spinal cord. Neurosci Lett 253: 167–170, 1998.
Gurin VN, Gurin AV, Melenchuk EV, and Spyer KM. The effects of
activation and blockade of central P2X receptors on body temperature.
Neurosci Behav Physiol 33: 845– 851, 2003.
Hansson E and Ronnback L. Glial neuronal signaling in the central nervous
system. FASEB J 17: 341–348, 2003.
Hara J, Beuckmann CT, Nambu T, Willie JT, Chemelli RM, Sinton CM,
Sugiyama F, Yagami K, Goto K, Yanagisawa M, and Sakurai T. Genetic
ablation of orexin neurons in mice results in narcolepsy, hypophagia, and
obesity. Neuron 30: 345–354, 2001.
Haydon PG. GLIA: listening and talking to the synapse. Nat Rev Neurosci 2:
185–193, 2001.
Illes P and Ribeiro JA. Molecular physiology of P2 receptors in the central
nervous system. Eur J Pharmacol 483: 5–17, 2004.
94 • SEPTEMBER 2005 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on October 1, 2016
from stimulated axons, an alternative possibility is that the
suramin attenuation of the evoked potential was not caused by
blockade of the ATP receptor but could be caused by a reported
nonselective attenuation of ionotropic glutamate receptors (Gu
et al. 1998; Lambrecht 2000). This would be consistent with
our finding both here and previously (Acuna-Goycolea et al.
2004; Li et al. 2002) that selective ionotropic glutamate receptor antagonists eliminate both evoked and spontaneous EPSPs
in hypocretin neurons. It is noteworthy that Jo and Role found
ATP release from cultured lateral hypothalamus cells, but not
slices. The lack of detection in hypothalamic slices (here and in
Jo and Role 2002) could be caused by developmental differences, to endogenous extracellular enzymes that break ATP
down, or to different parameters regulating ATP release. Along
similar lines, in spinal cord slices, ⬍5% of the neurons showed
clear ATP synaptic responses (Bardoni et al. 1997). ATP
concentration is highest in neuronal vesicles and other cellular
organelles, reaching ⱕ100 mM; free cytosolic ATP levels can
reach high millimolar levels (Zimmermann 1996). The latter is
true not only in neurons, but also in glia, from which ATP can
be released to act as an intercellular signaling molecule (James
and Butt 2002). ATP released by neurons into the synaptic cleft
may reach local concentrations of ⱕ100 ␮M. Local stimulation
of retinal astrocytes evoked ATP release with extracellular
concentrations reaching 78 ␮M (Newman 2001). ATP can
initiate glial Ca2⫹ and ATP waves able to travel several
hundred microns (Hansson and Ronnback 2003; Wang et al.
2000). This glia–neuron communication, potentially mediated
by ATP, can modulate or underlie homo- and heterosynaptic
suppression at the astroglia-nerve terminal interface in other
brain regions (Zhang et al. 2003), interneuron excitation
(Bowser and Khakh 2004), and extrasynaptic neuron-glial
signaling (Fields and Stevens 2000). Acute pathological conditions in the brain, such as seizures, hypoxia, ischemia, or
trauma, can induce a dramatic increase in extracellular ATP,
reaching millimolar levels (Ciccarelli et al. 2001; Hansson and
Ronnback 2003; James and Butt 2002). In these conditions,
elevated extracellular ATP would lead to P2X receptor activation, which in turn would excite hypocretin cells, resulting in
enhanced secretion of hypocretin from axon terminals. This
putative link between local cell injury and activation of the
hypocretin system gets additional support from the finding that
hypocretin cells appear to express mainly P2X2 subunits.
Those subunit typically potentiate their response at a lower pH,
and could do so in an acidified milieu of local brain ischemia
or injury. Thus an increase in ATP within the lateral hypothalamus, either from neurons, glia, or from pathological conditions, may play a role in activation of the hypothalamic arousal
system.
2205
2206
G. WOLLMANN, C. ACUNA-GOYCOLEA, AND A. N. VAN DEN POL
J Neurophysiol • VOL
Peyron C, Faraco J, Rogers W, Ripley B, Overeem S, Charnay Y,
Nevsimalova S, Aldrich M, Reynolds D, Albin R, Li R, Hungs M,
Pedrazzoli M, Padigaru M, Kucherlapati M, Fan J, Maki R, Lammers
GJ, Bouras C, Kucherlapati R, Nishino S, and Mignot E. A mutation in
a case of early onset narcolepsy and a generalized absence of hypocretin
peptides in human narcoleptic brains. Nat Med 6: 991–997, 2000.
Peyron C, Tighe DK, van den Pol AN, de Lecea L, Heller HC, Sutcliffe JG,
and Kilduff TS. Neurons containing hypocretin (orexin) project to multiple
neuronal systems. J Neurosci 18: 9996 –10015, 1998.
Phillis JW, O’Regan MH, and Perkins LM. Adenosine 5⬘-triphosphate
release from the normoxic and hypoxic in vivo rat cerebral cortex. Neurosci
Lett 151: 94 –96, 1993.
Ralevic V and Burnstock G. Receptors for purines and pyrimidines. Pharmacol Rev 50: 413– 492, 1998.
Robertson SJ, Ennion SJ, Evans RJ, and Edwards FA. Synaptic P2X
receptors. Curr Opin Neurobiol 11: 378 –386, 2001.
Sakurai T, Amemiya A, Ishii M, Matsuzaki I, Chemelli RM, Tanaka H,
Williams SC, Richardson JA, Kozlowski GP, Wilson S, Arch JR,
Buckingham RE, Haynes AC, Carr SA, Annan RS, McNulty DE, Liu
WS, Terrett JA, Elshourbagy NA, Bergsma DJ, and Yanagisawa M.
Orexins and orexin receptors: a family of hypothalamic neuropeptides and G
protein-coupled receptors that regulate feeding behavior. Cell 92: 573–585,
1998.
Silinsky EM and Redman RS. Synchronous release of ATP and neurotransmitter within milliseconds of a motor nerve impulse in the frog. J Physiol
492: 815– 822, 1996.
Sorimachi M, Ishibashi H, Moritoyo T, and Akaike N. Excitatory effect of
ATP on acutely dissociated ventromedial hypothalamic neurons of the rat.
Neuroscience 105: 393– 401, 2001.
Spyer KM, Dale N, and Gourine AV. ATP is a key mediator of central and
peripheral chemosensory transduction. Exp Physiol 89: 53–59, 2004.
Stoop R, Surprenant A, and North RA. Different sensitivities to pH of
ATP-induced currents at four cloned P2X receptors. J Neurophysiol 78:
1837–1840, 1997.
Thannickal TC, Moore RY, Nienhuis R, Ramanathan L, Gulyani S,
Aldrich M, Cornford M, and Siegel JM. Reduced number of hypocretin
neurons in human narcolepsy. Neuron 27: 469 – 474, 2000.
van den Pol AN. Hypothalamic hypocretin (orexin): robust innervation of the
spinal cord. J Neurosci 19: 3171–3182, 1999.
van den Pol AN. Narcolepsy: a neurodegenerative disease of the hypocretin
system? Neuron 27: 415– 418, 2000.
van den Pol AN, Acuna-Goycolea C, Clark KR, and Ghosh PK. Physiological properties of hypothalamic MCH neurons identified with selective
expression of reporter gene after recombinant virus infection. Neuron 42:
635– 652, 2004.
Wakamori M and Sorimachi M. Properties of native P2X receptors in large
multipolar neurons dissociated from rat hypothalamic arcuate nucleus. Brain
Res 1005: 51–59, 2004.
Wang Z, Haydon PG, and Yeung ES. Direct observation of calciumindependent intercellular ATP signaling in astrocytes. Anal Chem 72:
2001–2007, 2000.
Westfall TD, Kennedy C, and Sneddon P. Enhancement of sympathetic
purinergic neurotransmission in the guinea-pig isolated vas deferens by the
novel ecto-ATPase inhibitor ARL 67156. Br J Pharmacol 117: 867– 872,
1996.
Xiang Z, Bo X, Oglesby I, Ford A, and Burnstock G. Localization of
ATP-gated P2X2 receptor immunoreactivity in the rat hypothalamus. Brain
Res 813: 390 –397, 1998.
Yamanaka A, Beuckmann CT, Willie JT, Hara J, Tsujino N, Mieda M,
Tominaga M, Yagami K, Sugiyama F, Goto K, Yanagisawa M, and
Sakurai T. Hypothalamic orexin neurons regulate arousal according to
energy balance in mice. Neuron 38: 701–713, 2003.
Zhang JM, Wang HK, Ye CQ, Ge W, Chen Y, Jiang ZL, Wu CP, Poo
MM, and Duan S. ATP released by astrocytes mediates glutamatergic
activity-dependent heterosynaptic suppression. Neuron 40: 971–982, 2003.
Zimmermann H. Biochemistry, localization and functional roles of ectonucleotidases in the nervous system. Prog Neurobiol 49: 589 – 618, 1996.
94 • SEPTEMBER 2005 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on October 1, 2016
Inoue K. ATP receptors for the protection of hippocampal functions. Jpn
J Pharmacol 78: 405– 410, 1998.
Inoue K, Koizumi S, and Ueno S. Implication of ATP receptors in brain
functions. Prog Neurobiol 50: 483– 492, 1996.
James G and Butt AM. P2Y and P2X purinoceptor mediated Ca2⫹ signalling
in glial cell pathology in the central nervous system. Eur J Pharmacol 447:
247–260, 2002.
Jo YH and Role LW. Coordinate release of ATP and GABA at in vitro
synapses of lateral hypothalamic neurons. J Neurosci 22: 4794 – 4804, 2002.
Jo YH and Schlichter R. Synaptic corelease of ATP and GABA in cultured
spinal neurons. Nat Neurosci 2: 241–245, 1999.
Kanjhan R, Housley GD, Burton LD, Christie DL, Kippenberger A,
Thorne PR, Luo L, and Ryan AF. Distribution of the P2X2 receptor
subunit of the ATP-gated ion channels in the rat central nervous system.
J Comp Neurol 407: 11–32, 1999.
Khakh BS, Smith WB, Chiu CS, Ju D, Davidson N, and Lester HA.
Activation-dependent changes in receptor distribution and dendritic morphology in hippocampal neurons expressing P2X2-green fluorescent protein
receptors. Proc Natl Acad Sci USA 98: 5288 –5293, 2001.
Lambrecht G. Agonists and antagonists acting at P2X receptors: selectivity
profiles and functional implications. Naunyn Schmiedebergs Arch Pharmacol 362: 340 –350, 2000.
Li Y, Gao XB, Sakurai T, and van den Pol AN. Hypocretin/orexin excites
hypocretin neurons via a local glutamate neuron—a potential mechanism for
orchestrating the hypothalamic arousal system. Neuron 36: 1169 –1181,
2002.
Lin L, Faraco J, Li R, Kadotani H, Rogers W, Lin X, Qiu X, de Jong PJ,
Nishino S, and Mignot E. The sleep disorder canine narcolepsy is caused
by a mutation in the hypocretin (orexin) receptor 2 gene. Cell 98: 365–376,
1999.
Liu RJ, van den Pol AN, and Aghajanian GK. Hypocretins (orexins)
regulate serotonin neurons in the dorsal raphe nucleus by excitatory direct
and inhibitory indirect actions. J Neurosci 22: 9453–9464, 2002.
Liss B and Roeper J. Molecular physiology of neuronal K-ATP channels
(review). Mol Membr Biol 18: 117–127, 2001.
Matsumoto N, Sorimachi M, and Akaike N. Excitatory effects of ATP on rat
dorsomedial hypothalamic neurons. Brain Res 1009: 234 –237, 2004.
Molliver DC, Cook SP, Carlsten JA, Wright DE, and McCleskey EW. ATP
and UTP excite sensory neurons and induce CREB phosphorylation through
the metabotropic receptor, P2Y2. Eur J Neurosci 16: 1850 –1860, 2002.
Mori M, Heuss C, Gahwiler BH, and Gerber U. Fast synaptic transmission
mediated by P2X receptors in CA3 pyramidal cells of rat hippocampal slice
cultures. J Physiol 535: 115–123, 2001.
Neary JT, Rathbone MP, Cattabeni F, Abbracchio MP, and Burnstock G.
Trophic actions of extracellular nucleotides and nucleosides on glial and
neuronal cells. Trends Neurosci 19: 13–18, 1996.
Newman EA. Propagation of intercellular calcium waves in retinal astrocytes
and Muller cells. J Neurosci 21: 2215–2223, 2001.
Nieber K, Poelchen W, and Illes P. Role of ATP in fast excitatory synaptic
potentials in locus coeruleus neurones of the rat. Br J Pharmacol 122:
423– 430, 1997.
Norenberg W and Illes P. Neuronal P2X receptors: localisation and functional properties. Naunyn Schmiedebergs Arch Pharmacol 362: 324 –339,
2000.
North RA. Molecular physiology of P2X receptors. Physiol Rev 82: 1013–
1067, 2002.
North RA and Surprenant A. Pharmacology of cloned P2X receptors. Annu
Rev Pharmacol Toxicol 40: 563–580, 2000.
Ortinau S, Laube B, and Zimmermann H. ATP inhibits NMDA receptors
after heterologous expression and in cultured hippocampal neurons and
attenuates NMDA-mediated neurotoxicity. J Neurosci 23: 4996 –5003,
2003.
Pankratov Y, Castro E, Miras-Portugal MT, and Krishtal O. A purinergic
component of the excitatory postsynaptic current mediated by P2X receptors
in the CA1 neurons of the rat hippocampus. Eur J Neurosci 10: 3898 –3902,
1998.
Pankratov Y, Lalo U, Krishtal O, and Verkhratsky A. P2X receptormediated excitatory synaptic currents in somatosensory cortex. Mol Cell
Neurosci 24: 842– 849, 2003.