A role for lateral hypothalamic orexin neurons in
Glenda C. Harris1, Mathieu Wimmer1& Gary Aston-Jones1
The lateral hypothalamus is a brain region historically implicated
in reward and motivation1–4, but the identity of the neurotrans-
mitters involved are unknown. The orexins (or hypocretins) are
neuropeptides recently identified as neurotransmitters in lateral
hypothalamus neurons5,6. Although knockout and transgenic
overexpression studies have implicated orexin neurons in arousal
and sleep7, these cells also project to reward-associated brain
regions, including the nucleus accumbens and ventral tegmental
area8,9. This indicates a possible role for these neurons in reward
function and motivation3,10, consistent with previous studies
implicating these neurons in feeding6. Here we show that acti-
vation of lateral hypothalamus orexin neurons is strongly linked
to preferences for cues associated with drug and food reward. In
addition, we show that chemical activation of lateral hypothala-
mus orexin neurons reinstates an extinguished drug-seeking
behaviour. This reinstatement effect was completely blocked by
prior administration of an orexin A antagonist. Moreover, admin-
istration of the orexin A peptide directly into the ventral teg-
mental area also reinstated drug-seeking. These data reveal a new
role for lateral hypothalamus orexin neurons in reward-seeking,
drug relapse and addiction.
We used a two-chamber, nonbiased, conditioned place-preference
(CCP) model to measure the rewarding properties of morphine,
cocaine or food11–13. In this model, one chamber becomes associated
with drug or food reward through repeated pairings, whereas the
other chamber is associated with no reward. Preference for reward is
measured by the amount of time animals spend in the reward-
associated chamber minus the time it spends in the non-rewarded
Orexin-expressing neurons are located in three contiguous hypo-
thalamic regions: lateral hypothalamus, perifornical area (PFA) and
dorsomedial hypothalamus (DMH)14. To determine whether orexin
neurons were stimulated during the expression of preference for
different rewards, we used double-label immunohistochemistry for
both orexin and the immediate early gene protein, Fos (a marker of
neuronal stimulation15). Conditioned animals displayed reward-
seeking by spending significantly more time in the reward-paired
chamber than non-conditioned animals (Fig. 1a). Only conditioned
animals that exhibited a preference for the reward-paired chamber
showedincreased Fos activation in lateral hypothalamus orexincells.
Significant percentages (48–52%) of orexin neurons in the lateral
hypothalamus were Fos-activated in these animals after preference
testing for morphine, cocaine or food reward, when compared to
non-conditioned animals (17%, P , 0.01) (Table 1; Fig. 1b, see also
Supplementary Fig. 1). The percentage of Fos activation in lateral
hypothalamus orexin neurons in non-conditioned animals was not
statistically different from that found in naı ¨ve untreated animals
(15%, P ¼ 0.10). Our findings in naı ¨ve animals are similar to base-
line levels of Fos activation reported previously in awake rats16.
Enhanced activation of lateral hypothalamus orexin neurons after
conditioning indicates a potential involvement of these neurons in
the preference for reward-related cues. In support of this idea, the
amount of Fos activation in these neurons was correlated with the
intensity of reward seeking. Significant positive correlations were
found between the percentages of orexin neurons that were Fos-
positive in the lateral hypothalamus and preferences shown by
corresponding conditioned animals (P , 0.01; Table 1). Thus, as
preference scores increased, percentages of lateral hypothalamus
for all three rewards. In contrast, numbers of Fos-activated non-
orexin neurons in the lateral hypothalamus did not correlate with
preference. Moreover, despite the fact that substantial numbers of
Fos-positive orexin neurons were found in PFA and DMH, no
correlations were found in these areas between preference scores
and the percentage of Fos-positive orexin neurons (P . 0.20;
Table 1). This indicates that the relationship between reward-seeking
and neuronal stimulation in the hypothalamus occurs specifically in
Table 1 | Orexin and Fos double labelling
Morphine-conditioned n ¼ 12Orx LH
48 ^ 2*
55 ^ 6
62 ^ 2
67 ^ 4
50 ^ 3*
47 ^ 5
42 ^ 3
47 ^ 6
52 ^ 5*
78 ^ 7
67 ^ 3
74 ^ 3
17 ^ 2
43 ^ 6
52 ^ 4
59 ^ 4
15 ^ 1
29 ^ 8
52 ^ 3
57 ^ 6
18 ^ 2
50 ^ 1
56 ^ 3
63 ^ 5
0.72, P , 0.01*
Food-conditioned n ¼ 80.87, P , .01*
Cocaine-conditioned n ¼ 80.90, P , 0.01*
Non-conditioned n ¼ 15
Naı ¨ve n ¼ 6
Novelty-conditioned n ¼ 6
The percentages of orexin-positive cells that were also Fos-positive are indicated. Correlation
coefficients for the comparisons between these percentages and the corresponding
preference score in each animal. Abbreviations used: Orx, orexin-positive neurons; NonOrx,
orexin-negative neurons; LH, lateral hypothalamus; PFA, perifornical area; and DMH,
dorsomedial hypothalamus. The non-orexin Fos-positive neurons in the lateral hypothalamus
are given as total counts not percentages. Lateral hypothalamus by group ANOVA
F3,39¼ 33, P , 0.01.
*Significantly different from other groups, P , 0.05.
1Laboratory of Neuromodulation and Behavior, Department of Psychiatry, University of Pennsylvania, 705 Stellar Chance/6100 422 Curie Blvd, Philadelphia, Pennsylvania
Vol 437|22 September 2005|doi:10.1038/nature04071
© 2005 Nature Publishing Group
orexin neurons, but not in non-orexin neurons or in orexin neurons
in adjacent hypothalamic areas.
A separate group of animals (n ¼ 11) was given morphine CPP
training followed by a systemic injection of the selective orexin A
antagonist (SB 334867; ref. 17) on a second consecutive day of CPP
testing. Administration of the orexin antagonist produced a signifi-
cant reduction in preference (206 ^ 23s versus 87 ^ 30s; t12¼ 2.3;
P , 0.05) that was not found in a different group of animals given a
vehicle injection instead (182 ^ 21s versus 145 ^ 27s; t12¼ 0.9;
P ¼ 0.39). There were no differences in the original preferences for
morphine between these two groups (P ¼ 0.46). These data indicate
that orexin has a role in the amount of preference expressed.
Not all reward-seeking behaviours were associated with enhanced
stimulation of lateral hypothalamus orexin neurons. In a separate
group of animals conditioned with novel object reward18, we found
no enhanced Fos activation in lateral hypothalamus orexin neurons.
In this model, exposure to a novel object is paired with one
conditioning chamber, whereas no object is paired with the other
with the novel object (n ¼ 6 rats; preference score, 228 ^ 34s), and
similar activity levelsas inthe drugand food conditioned animalson
the test day. Despite this robust preference there was no significant
Fos induction in the lateral hypothalamus orexin neurons during
preference testing (18 ^ 2% of orexin neurons were Fos positive;
Table 1; see also Supplementary Fig. 2). Measures of orexin versus
non-orexin neurons that were Fos-activated were not statistically
differentfromnon-conditionedanimals(P . 0.20)andtherewasno
correlation with preference scores (Table 1). Novelty conditioning
entailed more conditioning days than food or drug conditioning;
however, it seems unlikely that these extra conditioning days
decreased orexin neuron stimulation. These data indicate that not
all rewards work through the activation of the lateral hypothalamus
orexin system, and that this system may be specifically engaged with
consummatory rewards (that is, food and drugs).
These results show that lateral hypothalamus orexin neurons are
activated in proportion to preference for certain rewards. We
reasoned that if activation of these neurons drives reward seeking,
then stimulation of these cells should reinstate extinguished pref-
erence. To test this, we submitted rats to morphine place condition-
ing and then extinguished the morphine preference by repeatedly
exposing the animals to the chambers without morphine adminis-
tration19(Fig. 2a). To activate orexin neurons, we microinfused the
Y4 agonist rPP (rat pancreatic polypeptide) directly into the lateral
hypothalamus orexin neuronal area. We expected rPP to stimulate
orexin neurons because these cells express the Y4 receptor, and
activation of this receptor by rPP potently induces Fos activation
in lateral hypothalamus orexin neurons20. We found that micro-
injecting rPP into the lateral hypothalamus robustly reinstated an
extinguished morphine place preference (Fig. 2a). Similar control
injections of rPP dorsal, ventral or medial to lateral hypothalamus
orexin cells did not reinstate preference (Fig. 2a, b). To insure that
vehicle injections alone did not produce reinstatement, five animals
from the lateral hypothalamus-injected group were given vehicle
injections 3days before the rPP injection. In these cases, the vehicle
Figure 1 | Morphine conditioned animals had significantly greater place
preferences and Fos activated orexin neurons in the lateral hypothalamus
than non-conditioned animals. a, Conditioned animals showed significant
preferences for the reward-paired chamber compared withnon-conditioned
animals (F3,39¼ 15, P , 0.01). Preference scores for the morphine
(8mgkg21), cocaine (15mgkg21) and food (lucky charms cereal)-paired
environments expressed as the time spent in the reward-paired side minus
the time spent on the non-rewarded side on the test day (mean ^ s.e.m.).
b, High-power photomicrograph of the lateral hypothalamus showing the
doublelabelling oforexin (browncytoplasm)andFos protein(blacknuclei)
in morphine-conditioned and non-conditioned animals. Black arrows
indicate double-labelled cells.
Figure 2 | Activation of lateral hypothalamus orexin neurons by rPP
reinstated an extinguished preference for morphine. a, Preference scores
are shown for both rPP- (150nM) and vehicle-injected groups
(mean ^ s.e.m. in morphine-paired side minus saline-paired side) during
the initial conditioning test, after extinction and during the reinstatement
test. b, Effective (filled circles, n ¼ 12) and ineffective (open squares, n ¼ 9)
sites of rPP administration. c, The selective orexin A antagonist, SB 334867
(20–30mgkg21),blockedreinstatementby rPP(n ¼ 8).Datawereincluded
only if rPP injection into lateral hypothalamus on the following day
(without the antagonist pretreatment) produced reinstatement of
preference. d, Plot of correlation between reinstatement scores and
percentages of lateral hypothalamus orexin neurons that were Fos activated
in rPP reinstated animals. Abbreviations used were: ZI, zona incerta; Sub I,
subincertal nucleus; Me, medial amygdala nucleus; VMH, ventromedial
hypothalamic nucleus; Arc, arcuate nucleus; PFA, perifornical area; DMH,
dorsomedial hypothalamus; LH lateral hypothalamus; 3V, third ventricle;
and f, fornix.
NATURE|Vol 437|22 September 2005
© 2005 Nature Publishing Group
injections did not reinstate the CPP behaviour, whereas the later rPP
injections produced robust reinstatement (Fig. 2a; reinstatement
score: 247 ^ 71s for vehicle versus 465 ^ 100s for rPP). Systemic
injections of morphine are a reliable method to reinstate extin-
guished responses to morphine cues in this model19. We found that
reinstatement produced by systemic morphine (8mgkg21, intra-
peritoneal; morphine reinstatement score was 424 ^ 103s, n ¼ 8
rats and rPP reinstatement score was 353 ^ 52s, n ¼ 12, P . 0.5).
The rPP-induced reinstatement was completely blocked by prior
confirming a role for orexin receptors in rPP-induced reinstatement.
One day later, the same animals reinstated preference when injected
with rPP in the lateral hypothalamus without antagonist pretreat-
ment. This confirmed that antagonist pretreatment blocked an
otherwise effective reinstatement by rPP. In addition, we measured
Fos staining in both orexin and non-orexin lateral hypothalamus
cells in seven rats with significant reinstatement after rPP injections,
scores were 279 ^ 40s and 251 ^ 46s, respectively; P , 0.01). In
the animals for which rPP reinstated preference, the percentage of
Fos-activated orexin cells in the lateral hypothalamus was signifi-
cantly greater than in control animals, for which rPP did not cause
reinstatement (P , 0.01; lateral hypothalamus orexin neurons that
wereFos-activated forreinstatedrats,66 ^ 8%;andforcontrolnon-
reinstated animals, 11 ^ 5%). Moreover, a highly significant corre-
lation was found between the preference expressed during rPP-
induced reinstatement and the percentage of lateral hypothalamus
orexinneuronsthatwereFos-activated(Fig.2d;R ¼ 0.99;P . 0.01).
Thus, the greater the number of lateral hypothalamus orexin cells
that were stimulated, the greater the intensity of reinstatement that
resulted. No differences were found in the number of non-orexin
Fos-positivecellsbetweenthegroups(P ¼ 0.77),andnocorrelations
were found between the numbers of these non-orexin Fos-activated
neurons and reinstatement scores (R ¼ 0.23; P ¼ 0.72).
The above evidence strongly indicates that stimulation of orexin
lateral hypothalamus neurons can drive reinstatement behaviour for
cues associated with drug reward. The effect of rPP on morphine
reinstatement could be due to a stress/arousal effect, or to the fact
that rPP mimics the motivational effects of morphine. Our prelimi-
nary results indicate that an acute injection of morphine (but not
saline) in the CPP box strongly induces Fos in lateral hypothalamus
orexin neurons, whereas, exposure to footshock (using parameters
that we and others find reinstates morphine CPP)19only activated
orexin neurons in the DMH and PFA but not in the lateral
hypothalamus. These data support the latter interpretation (see
Supplementary Table 1).
To determine whether the orexin peptide could produce reinstate-
drug reinstatement21and orexin has an excitatory action on GABA-
and dopamine-containing neurons in this area22. We found that
orexin caused a significant reinstatement response for morphine
reward only when microinjected into VTA, but not when injected
into areas surrounding VTA (Fig. 3a, b). Similar vehicle injections
had no effect.
Taken together these data strongly indicate that orexin neurons in
the lateral hypothalamus become activated by cues associated with
consummatory rewards such as food and drugs. This view is
consistent with an early proposal for orexin function in feeding6
and recent implications in morphine dependence and withdrawal23.
Orexin receptors are expressed at high levels in reward-associated
areas, including the VTA and nucleus accumbens24,25. As both food
and drugs of abuse work through common brain substrates10,26, it
is not surprising that lateral hypothalamus orexin neurons may
mediate responses to cues associated with both food and drugs,
but not non-consummatory rewards (for example, novelty). This is
consistent with a recent report showing that treatments which
increase consummatory behaviour were associated with Fos acti-
vation of lateral hypothalamus orexin cells more than exposure to
orexin neurons may have different functions. A previous report
indicated that Fos activation of orexin neurons in PFA and DMH
showed diurnal changes consistent with a role in production or
maintenance of arousal; lateral hypothalamus orexin neurons
showed no such diurnal property28. These results, together with
our findings, suggest the hypothesis that orexin neurons in PFA and
DMH regulate arousal whereas those in lateral hypothalamus regu-
late reward processing. Our data further indicate that activation of
lateral hypothalamus orexin neurons may be important in the
reinstatement of drug-seeking behaviours and thereby have an
important role in modulating rewarding behaviour and addiction.
Although orexin administration directly into VTA reinstated mor-
phine preference, VTA may not be the only location for such orexin
effects as orexin-containing fibres are found throughout areas
associated with reward processing, such as the extended amygdala10
and mesocorticolimbic system9. We conclude that the lateral
hypothalamus orexin system is an integral part of circuitry that
stimulation of these neurons can drive relapse of drug-seeking
Subjects. Male Sprague–Dawley rats (200–250g; Harlan) were group-housed in
accordance with NIH guidelines on a 12-h light/dark cycle with food and water
available ad libitum. The Institutional Animal Care and Use Committee of the
University of Pennsylvania approved all animal procedures.
Drugs. Morphine sulphate powder, provided by the National Institute on Drug
Abuse, was dissolved in sterile saline and administered via intraperitoneal
injection (8mgkg21). Cocaine (Sigma-Aldrich) was dissolved in sterile saline
(15mgkg21, intraperitoneal). SB-334867 (Tocris) was dissolved in a 10% (w/v)
encapsin/2% DMSO solution inwater (intraperitoneal, 20mgkg21(n ¼ 2) and
30mgkg21(n ¼ 18);refs29,30).OrexinA(140nM;Tocris)andrPP(150nM20;
Anaspec) were dissolved in artificial cerebral spinal fluid (vehicle).
Surgery and histology. Rats were anaesthetized with a ketamine/xylazine
cocktail and implanted with bilateral chronic indwelling guide cannulae
Figure 3 | Orexin administration into the VTA reinstated an extinguished
preferenceformorphine. a,Reinstatementscores(timeismean ^ s.e.m.in
morphine-paired side minus saline-paired side) after orexin administration
(140nM) in the VTA. Orexin in the VTA (n ¼ 8) was significantly different
from vehicle in the VTA (n ¼ 6) or orexin outside the VTA (n ¼ 7) in
reinstating an extinguished morphine preference (F2,18¼ 11, P , 0.01).
b, Effective (filled circles) and ineffective (open squares) sites of orexin
administration. Gray circles are sites of vehicle administration. Drawings
were adapted from ref. 31 (plates 36–38). Abbreviations used were: SN,
substantia nigra; MM, medial mammillary nucleus; and RN, red nucleus.
NATURE|Vol 437|22 September 2005
© 2005 Nature Publishing Group
aimed 2mm above the lateral hypothalamus (anterior/posterior (AP) 23.3,
medial/lateral (ML) þ2.5 and dorsal/ventral (DV) 28.0, 58 angle from bregma)
or VTA (AP 25.3, ML þ 2.5, DV 27.5). Surgeries were similar to previous
reports12. Control injections were made ,2mm above the lateral hypothalamus
or VTA. Animals were given 1week to recover from surgery before place
of sodium pentobarbital (100mgkg21intraperitoneal) and cannulae locations
were confirmed in histological analyses.
Conditioned place preference procedure. We used a standard two-chamber
balanced design, with two conditioning sessions (for food and drugs) occurring
between morning and afternoon sessions. Conditioning and testing were
identical to our previously published reports for morphine11, cocaine12and
food13. Food conditioned animals were not food deprived and were pre-exposed
to the sweet flavoured cereal one week before conditioning. Non-conditioned
animals consisted of morphine-treated animals for which the environment was
explicitly unpaired with morphine (n ¼ 7)11, and cocaine-treated animals in
which conditioning failed (animals who were conditioned but whose preference
scores were less than 70s for the cocaine environment, n ¼ 8). No significant
differences(P . 0.08)werefoundbetweenthesetwonon-conditionedgroupsin
any measures taken, and their data were pooled. Novel object conditioning18
included 10-min pairings of a novel object versus no object in distinct
compartments for 8d, with 1h between pairings (n ¼ 6). The novel objects
were paper bows, a plastic ball, a piece of cotton, a newspaper, halves of socks,
pieces of egg carton, a sponge and wood chips. The order of pairings (object
versus no object) was alternated each day.
Extinction and reinstatement procedure. After conditioning, animals were
tested daily until preference levels for the morphine-paired chamber dropped to
less than 70s difference for the two chambers for two consecutive days. rPP or
vehicle was injected bilaterally into the lateral hypothalamus (or, orexin
administered into the VTA) 30min before the reinstatement test. For the orexin
antagonist experiments, SB-334867 was injected 30min before rPP. For mor-
phine reinstatement, morphine 8mgkg21(intraperitoneal) was injected
immediately before placing animals in the conditioning boxes.
Double label immunohistochemistry. Two hours after CPP testing rats were
deeply anaesthetized and perfused transcardially with paraformaldehyde as
ing with nickel ammonium sulphate intensification of 3
(DAB) as we have previously described11, and then processed for orexin (orexin
A antibody, Santa Cruz Biotechnology; 1:1,000; biotinylated secondary, 1:500).
cover slip. Double-labelled cells were readily identified because orexin stained
the cytoplasm brown and Fos immunoreactive nuclei were black. For counting,
two sections from each animal were chosen at the same levels with equivalent
numbers of orexin-positive neurons. Colour photographs of the lateral hypo-
thalamus and DMH were taken and the numbers of orexin neurons, and Fos-
positive/orexin doubly labelled neurons were counted and averaged for each
animal. The number of double-labelled cells were determined by both blinded
and non-blind observers; there was a 94% agreement between these two sets of
be in the lateral hypothalamus. All orexin labelled neurons located dorsal and
0.4mm medial to the fornix were considered to be in the PFA. All remaining
orexin labelled neurons from the medial edge of the PFA region to the third
ventricle, were considered to be in the DMH.
Data analyses. Place conditioning data were analysed by calculating the time
spent in the reward-paired chamber minus the time spent in the other chamber.
The resulting difference score was compared between groups using analysis of
variance (ANOVA). The percentages of double-labelled cells were compared
between groups using ANOVA. Pearsons R correlation was used to compare
preference or reinstatement scores with the number of double labelled cells.
Where necessary, post-hoc analysis was carried out with a Newman–Keuls test.
Received 31 May; accepted 26 July 2005.
Published online 14 August 2005.
1.Anand, B. K. & Brobeck, J. R. Hypothalamic control of food intake in rats and
cats. Yale J. Biol. Med. 24, 123– -140 (1951).
Olds, J. & Milner, P. Positive reinforcement produced by electrical stimulation
of septal area and other regions of rat brain. J. Comp. Physiol. Psychol. 47,
419– -427 (1954).
DiLeone, R. J., Georgescu, D. & Nestler, E. J. Lateral hypothalmic neuropeptides
in reward and drug addiction. Life Sci. 73, 759– -768 (2003).
Petrovich, G. D. & Gallagher, M. Amygdala subsystems and control of feeding
behaviour by learned cues. Ann NY Acad. Sci. 985, 251– -262 (2003).
de Lecea, L. et al. The hypocretins: hypothalamus-specific peptides with
neuroexcitatory activity. Proc. Natl Acad. Sci. USA 95, 322– -327 (1998).
Sakurai, T. et al. Orexins and orexin receptors: a family of hypothalamic
neuropeptides and G protein-coupled receptors that regulate feeding
behaviour. Cell 92, 573– -585 (1998).
Chemelli, R. M. et al. Narcolepsy in orexin knockout mice: molecular genetics
of sleep regulation. Cell 98, 437– -451 (1999).
Peyron, C. et al. Neurons containing hypocretin (orexin) project to multiple
neuronal systems. J. Neurosci. 18, 9996– -10015 (1998).
Fadel, J. & Deutch, A. Y. Anatomical substrates of orexin-dopamine
interactions: lateral hypothalamic projections to the ventral tegmental area.
Neuroscience 111, 379– -387 (2002).
10. Baldo, B. A., Daniel, R. A., Berridge, C. W. & Kelley, A. E. Overlapping
distributions of orexin/hypocretin- and dopamine-beta-hydroxylase
immunoreactive fibers in rat brain regions mediating arousal, motivation, and
stress. J. Comp. Neurol. 464, 220– -237 (2003).
11. Harris, G. C. & Aston-Jones, G. Enhanced morphine preference following
prolonged abstinence: association with increased Fos expression in the
extended amygdala. Neuropsychopharmacology 28, 292– -299 (2003).
12. Harris, G. C. & Aston-Jones, G. Critical role for ventral tegmental glutamate in
preference for a cocaine-conditioned environment. Neuropsychopharmacology
28, 73– -76 (2003).
13. Harris, G. C. & Aston-Jones, G. Altered motivation and learning following
opiate withdrawal: evidence for prolonged dysregulation of reward processing.
Neuropsychopharmacology 28, 865– -871 (2003).
14. Siegel, J. M. Hypocretin (orexin): role in normal behaviour and neuropathology.
Annu. Rev. Psychol. 55, 125– -148 (2004).
15. Herrera, D. G. & Robertson, H. A. Activation of c-fos in the brain. Prog.
Neurobiol. 50, 83– -107 (1996).
16. Espana, R. A., Valentino, R. J. & Berridge, C. W. Fos immunoreactivity in
hypocretin-synthesizing and hypocretin-1 receptor-expressing neurons: effects
of diurnal and nocturnal spontaneous waking, stress and hypocretin-1
administration. Neuroscience 121, 201– -217 (2003).
17. Smart, D. et al. SB-334867-A: the first selective orexin-1 receptor antagonist.
Br. J. Pharmacol. 132, 1179– -1182 (2001).
18. Bevins, R. A. et al. Novel-object place conditioning: behavioural and
dopaminergic processes in expression of novelty reward. Behav. Brain Res. 129,
41– -50 (2002).
19. Wang, B., Luo, F., Zhang, W. T. & Han, J. S. Stress or drug priming induces
reinstatement of extinguished conditioned place preference. Neuroreport 11,
2781– -2784 (2000).
20. Campbell, R. E. et al. Orexin neurons express a functional pancreatic
polypeptide Y4 receptor. J. Neurosci. 23, 1487– -1497 (2003).
21. Shalev, U., Grimm, J. W. & Shaham, Y. Neurobiology of relapse to heroin and
cocaine seeking: a review. Pharmacol. Rev. 54, 1– -42 (2002).
22. Korotkova, T. M., Sergeeva, O. A., Eriksson, K. S., Haas, H. L. & Brown, R. E.
Excitation of ventral tegmental area dopaminergic and nondopaminergic
neurons by orexin/hypocretins. J. Neurosci. 23, 7– -11 (2003).
23. Georgescu, D. et al. Involvement of the lateral hypothalamic peptide orexin in
morphine dependence and withdrawal. J. Neurosci. 23, 3106– -3111 (2003).
24. Trivedi, P., Yu, H., MacNeil, D. J., Van der Ploeg, L. H. & Guan, X. M.
Distribution of orexin receptor mRNA in the rat brain. FEBS Lett. 438, 71– -75
25. Marcus, J. N. et al. Differential expression of orexin receptors 1 and 2 in the rat
brain. J. Comp. Neurol. 435, 6– -25 (2001).
26. Carr, K. D. Augmentation of drug reward by chronic food restriction:
behavioural evidence and underlying mechanisms. Physiol. Behav. 76, 353– -364
27. Baldo, B. A. et al. Activation of a subpopulation of orexin/hypocretin-containing
hypothalamic neurons by GABAA receptor-mediated inhibition of the nucleus
accumbens shell, but not by exposure to a novel environment. Eur. J. Neurosci.
19, 376– -386 (2004).
28. Estabrooke, I. V. et al. Fos expression in orexin neurons varies with behavioural
state. J. Neurosci. 21, 1656– -1662 (2001).
29. Rodgers, R. J. et al. SB-334867, a selective orexin-1 receptor antagonist,
enhances behavioural satiety and blocks the hyperphagic effect of orexin-A in
rats. Eur. J. Neurosci. 13, 1444– -1452 (2001).
30. Haynes, A. C. et al. A selective orexin-1 receptor antagonist reduces food
consumption in male and female rats. Regul. Pept. 96, 45– -51 (2000).
31. Swanson, L. W. Brain Maps: Structure Of The Rat Brain (Elsevier, Amsterdam,
Supplementary Information is linked to the online version of the paper at
Acknowledgements We thank R. Shiekhattar and R. Smith for comments on the
manuscript; and Y. Zhu and S. Aston-Jones for assistance with illustrations and
photography. This work was supported by NIH.
Author Information Reprints and permissions information is available at
npg.nature.com/reprintsandpermissions. The authors declare no competing
financial interests. Correspondence and requests for materials should be
addressed to G.C.H. (firstname.lastname@example.org).
NATURE|Vol 437|22 September 2005
© 2005 Nature Publishing Group