In the development of addiction, drug seeking becomes habitual and controlled by drug-associated cues, and the neural locus of control over behaviour shifts from the ventral to the dorsolateral striatum. The neural mechanisms underlying this functional transition from recreational drug use to drug-seeking habits are unknown. Here we combined functional disconnections and electrophysiological recordings of the amygdalo-striatal networks in rats trained to seek cocaine to demonstrate that functional shifts within the striatum are driven by transitions from the basolateral (BLA) to the central (CeN) amygdala. Thus, while the recruitment of dorsolateral striatum dopamine-dependent control over cocaine seeking is triggered by the BLA, its long-term maintenance depends instead on the CeN. These data demonstrate that limbic cortical areas both tune the function of cognitive territories of the striatum and thereby underpin maladaptive cocaine-seeking habits.
Cocaine addiction is increasingly considered to stem from progressive neural adaptations to chronic drug exposure1,2,3 subserving a loss of control over drug-seeking behaviour that has become established as habitual4 and predominantly controlled by drug-associated conditioned stimuli in the environment5. However, the mechanisms whereby maladaptive cocaine-seeking habits are established are not yet understood.
The initiation of cocaine seeking as a goal-directed behaviour6,7 depends on limbic cortico-striatal networks including the nucleus accumbens (Acb) and its functional interactions with the basolateral amygdala (BLA)8. While the core of the Acb (AcbC) is suggested to translate drug-related motivation into drug-seeking actions4,9, the BLA encodes Pavlovian incentive associations9,10 and through its projections to the AcbC11 underlies the performance of prolonged sequences of drug seeking reinforced by drug-associated conditioned stimuli (CS) acting as conditioned reinforcers12, bridging delays to drug taking4. Thus, the BLA and its interactions with the AcbC are necessary for the acquisition of voluntary cue-controlled cocaine-seeking behaviour.
However, cocaine addiction is not simply a problem of voluntary, or ‘recreational’, drug use, but is instead characterized by a compulsive drug-seeking habit4, the transition to which has been shown neurally to depend on functional shifts in the control over behaviour from the ventral to the anterior dorsolateral striatum (aDLS)5,13,14,15,16. Thus protracted exposure to cocaine recruits neurobiological adaptations in the striatum that progressively spread from the AcbC to the aDLS14,17,18,19,20 finally to consolidate aDLS6,21, dopamine-dependent cocaine-seeking habits13,21 and compulsivity22,23, a hallmark of addiction24.
This intrastriatal functional shift has been suggested to depend on the recruitment of the ascending striato-nigro-striatal spiralling dopamine circuitry25,26, which allows control by the AcbC over dorsal striatal dopaminergic processes5,13,27, thereby linking aberrant drug-related incentive motivational processes to a more rigid habit system5. However, the neural mechanisms that underlie this striatal plasticity in the control over drug seeking are poorly understood.
Despite having no direct projections to the aDLS28, as opposed to the posterior dorsomedial striatum28, the BLA is both a key determinant of AcbC activity11,29, which is in turn necessary to recruit DLS-dopamine-dependent cocaine-seeking habits13, and also of the recruitment of aDLS-dependent functions such as response learning in a spatial navigation task30,31. In addition, synaptic plasticity at the BLA–AcbC pathway is involved in both appetitive learning32, reward seeking29 and the strengthening of drug conditioned reinforcement (incubation of cocaine ‘craving’) in rats33, that may presage the association between cue-evoked cocaine craving and increased dopamine transmission in the aDLS observed in humans34. We therefore hypothesized that the BLA, through its functional interactions with the ventral striatum, is the neural locus that triggers the recruitment of DLS-dependent cocaine-seeking habits.
Behaviourally, habitual cocaine seeking has been operationalized as instrumental responding in over-trained second-order schedules of reinforcement35,36 (SOCR) procedures5,15,37,38 whereby drug seeking during prolonged time periods is under the direct control of the conditioned reinforcing properties of cocaine CSs39.
We therefore tested the hypothesis that functionally disconnecting the BLA from dopaminergic transmission in the aDLS may prevent the establishment of cue-controlled cocaine-seeking habits. Thus, we combined a pretraining unilateral excitotoxic lesion of the BLA (see Supplementary Fig. 1 for the timeline of the experiments), which we predicted would prevent the cocaine-induced functional recruitment of aDLS-dopamine-dependent control over cocaine seeking on this side of the brain, with dopamine receptor blockade in the contralateral aDLS. Such an asymmetric manipulation bilaterally disrupts—or disconnects—the functional circuitry linking the BLA to dopaminergic transmission in the DLS recruited via the AcbC13.
As predicted, we show that the BLA, which controls the activity of the aDLS through AcbC glutamatergic mechanisms, is required to recruit aDLS control over cocaine seeking. We further demonstrate that a functional shift occurs in the amygdala from the BLA to the central nucleus (CeN) over the course of cocaine seeking such that it is the CeN, and not the BLA, that maintains aDLS-dopamine-dependent control over behaviour when cocaine seeking has become habitual.
BLA recruits aDLS control of cocaine-seeking habits
Permanent pretraining unilateral BLA lesions (see Supplementary Fig. 2 for histological assessment) did not impair the acquisition of cocaine self-administration nor the acquisition of cue-controlled cocaine-seeking behaviour under a SOCR (Supplementary Fig. 3A). Goal-directed6, cue-controlled cocaine seeking, measured at an early stage of training (Supplementary Fig. 1), was sensitive neither to bilateral aDLS-dopamine receptor blockade nor to the BLA–aDLS disconnection manipulation (Fig. 1a,b left panel). Thus disconnecting the BLA from aDLS-dopamine transmission does not impair cue-controlled cocaine seeking when we have previously shown it to be goal-directed and dependent on the AcbC and the posterior dorsomedial striatum40, but not the aDLS. However, when rats had an intermediate history of training in cocaine seeking, associated with a moderate increase in active lever presses (from 47 to 73 active lever presses in 15 min) (Fig. 1a,b middle panel), behavioural control over cocaine seeking began to devolve to the aDLS as revealed by a limited (27%) but significative effect on cocaine seeking of both bilateral aDLS-dopamine receptor blockade or BLA–aDLS disconnection.
This initial functional recruitment of the aDLS becomes progressively more established over several weeks of training, as shown in our previous work40,41, reflecting the emergence of a DLS-dependent cocaine-seeking habit under the control of cocaine-associated CSs5. Thus, as previously described13,21,40, at this later stage of training, cocaine seeking was dose-dependently decreased by bilateral aDLS infusions of the non-selective dopamine receptor antagonist α-flupenthixol (Fig. 1a right panel), an effect that was dependent on both D1 and D2 dopamine receptors as it was also observed after aDLS infusions of SCH23390 or raclopride, selective antagonists of D1 and D2 receptors, respectively (Supplementary Fig. 4). Moreover, cocaine seeking was similarly dose-dependently decreased by disconnecting the BLA from aDLS dopaminergic mechanisms (Fig. 1b right panel). Thus a pretraining unilateral BLA lesion prevented the functional recruitment of the aDLS ipsilaterally, such that otherwise ineffective unilateral dopamine receptor blockade in the contralateral aDLS now resulted in a decrease in cocaine seeking to the same extent as that seen following bilateral aDLS manipulations in non lesioned rats.
The effect of causal manipulations of the amygdalo-striatal circuitry on cocaine seeking was attributable to a functional disconnection between the BLA and the DLS because neither unilateral DLS dopamine receptor blockade, nor a double ipsilateral manipulation, impaired instrumental performance (Supplementary Fig. 5). Thus, the BLA is required to recruit intrastriatal functional shifts subserving the development of cocaine-seeking habits.
CeN maintains aDLS control over cocaine-seeking habits
Recent evidence that appetitive memories are associated with long-term potentiation of the BLA–AcbC pathway and concomitant long-term depression of the BLA–CeN pathway32 suggests that BLA control over striatal mechanisms as observed here is functionally independent of its connections with, and circuitry involving, the CeN, and thus does not rely on serial processing (BLA-to-CeN) in the amygdala. However, the BLA and CeN have also been shown to be involved in parallel and independent processes42; for example, the CeN is functionally linked to the aDLS through its direct projections to the substantia nigra dopaminergic neurons43,44, and via this pathway mediates the invigoration of instrumental responding in the presence of conditioned Pavlovian excitors45,46 and also biases motivation towards specific rewards47. Thus the CeN projections to the substantia nigra may also be involved in intrastriatal functional shifts. This alternative amygdalo-striatal route, shown to be involved in habitual control over behaviour for natural reinforcers48, may serve as a regulatory route, transmitting arousal caused by presentation of cocaine CSs49 to the aDLS-dependent habit system. We therefore investigated whether disconnecting the CeN from the aDLS could also influence the recruitment of a aDLS-dependent cocaine-seeking habit.
As predicted, when cocaine seeking did not depend on the aDLS, that is, early after training40,41, a disconnection combining a selective lesion of the CeN with contralateral aDLS-dopamine receptor blockade had no effect on cocaine seeking (Fig. 1c left panel). In marked contrast to the effects of a bilateral aDLS manipulation or the BLA–aDLS disconnection, the CeN–DLS disconnection also had no effect on cocaine seeking after intermediate training (Fig. 1c middle panel). However, in contrast, this disconnection markedly decreased well-established, habitual, seeking responses by 70% (Fig. 1c right panel). The influence of the asymmetric manipulation of the CeN and dopaminergic transmission in the aDLS is attributable to a functional disconnection, since a unilateral CeN lesion or a double ipsilateral manipulation had no effect on the acquisition of cocaine self-administration, cocaine seeking (Supplementary Fig. 3) or the expression of cocaine-seeking habits (Supplementary Fig. 5C). Thus the CeN is involved in the long-term maintenance of cocaine-seeking habits by the aDLS.
The lack of effect of the CeN–aDLS disconnection at an intermediate stage of training suggests that the CeN is functionally recruited later to exert an influence on the aDLS control over cocaine-seeking behaviour, consolidating a process initiated by the BLA.
We therefore sought further to investigate whether the BLA and the CeN were indeed controlling successive, non-overlapping, stages of intrastriatal, dopamine-dependent, functional shifts subserving cocaine-seeking responses5. Therefore, we tested the influence of acute bilateral inactivation or acute functional disconnection of either amygdala territory from aDLS dopaminergic mechanisms.
BLA-to-CeN shift in the control over cocaine-seeking habits
Transient inactivation of the BLA by bilateral infusion of a mixture of GABA receptor agonists (baclofen/muscimol (B/M))50 decreased early, but not well-established cocaine-seeking behaviour (Fig. 2a), the latter being dependent on aDLS dopaminergic processes (Fig. 2b). These drug-seeking habits were not influenced by an acute disconnection of the BLA and aDLS dopaminergic mechanisms, nor were they decreased by a double ipsilateral manipulation of the two structures (Fig. 2b). In marked contrast, bilateral inactivation of the CeN, which had no effect on early cocaine seeking (Fig. 2a), dramatically decreased well-established cocaine seeking to the same extent as bilateral blockade of aDLS-dopamine receptors (Fig. 2c). This habitual cocaine seeking was also dose-dependently decreased by a functional disconnection of the CeN and aDLS, while a double ipsilateral manipulation of these two structures had no effect on instrumental responding (Fig. 2c).
Taken together these results strongly support the hypothesis that the BLA is required to recruit aDLS-dependent cue-controlled cocaine-seeking habits5, whereas the CeN is necessary to maintain them.
BLA functionally regulates neural activity in the DLS
While functional interactions between the CeN and the DLS have previously been shown to support habitual control over behaviour48, demonstration of the functional recruitment by the BLA of aDLS-dopamine-dependent mechanisms is unprecedented. We therefore sought further to investigate the nature of the control exerted by the BLA over aDLS function.
We hypothesized that, for the BLA to trigger the instantiation of dopamine-dependent aDLS control over cocaine seeking in drug naïve animals (in which the amygdalo-striatal system is in the same state as that in experiment 1 when rats received pretraining lesions of the BLA), the BLA should influence the electrophysiological activity of aDLS neurons and this in turn should depend on antecedent glutamatergic mechanisms in the AcbC, the primary striatal target of the network originating from the BLA8,11,29.
For this we used extracellular electrophysiological recordings of putative medium spiny neurons (pMSNs) in the aDLS of anaesthetized rats and measured the influence of electrical stimulation of the ipsilateral BLA51 at various intervals (from 1 to 1,000 ms, see Fig. 3a) on the response probability of aDLS pMSNs to the stimulation of their primary glutamatergic input from the motor cortex 1 (M1)52,53 (Fig. 3a). As there are no direct projections between the BLA and the DLS54, we predicted that if the BLA indeed exerts control over aDLS pMSNs activity independently of the CeN, which is not involved in the early stages of intrastriatal shifts, then its gating of pMSN responses would require a multisynaptic route5 and would be dependent on antecedent glutamatergic mechanisms in the AcbC, its primary striatal target11 which has no direct connections with the CeN55.
Of the ∼50 aDLS pMSNs recorded, spike probability was significantly influenced by BLA stimulation (Fig. 3a, see Methods) at long, but not short, intervals prior to M1 stimulation (Fig. 3a–d). A cluster analysis (Fig. 3b, c, see Methods for further details) revealed that one subset of neurons showed a facilitation of cortico-striatal responses (upregulated, 39.6% of the population), a second subset (33.3% of the population) showed a robust inhibition while a third subpopulation showed no modulation of their cortico-striatal responses following BLA stimulation (27%) (Fig. 3d). These neurons displayed similar M1-stimulation-evoked action potentials, with similar spike latencies and duration parameters that were not influenced by BLA stimulation (all Fs<1) (Fig. 3c, Supplementary Fig. 6A,B and 7). As predicted by a lack of direct projection from the BLA to the aDLS, BLA stimulation alone triggered no AP in aDLS pMSNs (Fig. 3d) nor did BLA stimulation 1, 5 or 50 ms prior to M1 stimulation. This demonstrates that the BLA exerts no influence over aDLS pMSN activity through neural pathways that require few synapses (see ref. 51), such as the one involving the CeN.
However, when the BLA was stimulated 100, 200 or 300 ms prior to M1 stimulation, it induced a marked increase, or decrease, in the spike probability of upregulated or downregulated neurons, respectively (Fig. 3d). This differential effect was not observed for longer latencies and did not result in recurrent network effects as post-challenge M1 stimulation alone resulted in aDLS pMSN spike probabilities that were similar to pre BLA stimulation baseline levels (Fig. 3d).
BLA controls aDLS pMSNs via AcbC glutamatergic mechanisms
These observations demonstrate that the BLA is functionally associated with the aDLS despite the absence of direct neuronal connectivity reported previously28. It is thereby able to exert control over the physiological activity of pMSNs in the aDLS through a multisynaptic route by modulating the spike probability of these neurons. The long BLA–M1 interstimulation interval (IsI) at which the effect is observed, alongside the differential nature of the BLA control over aDLS pMSNs, that is, an up- or downregulation of the firing probability of different populations, suggest that this effect may depend on the ascending striato-nigro-striatal dopamine-dependent loops that allow the activity of the AcbC to regulate dopamine transmission in the aDLS14. This hypothesis was further supported by the demonstration that the BLA control over aDLS pMSNs was abolished by glutamate receptor blockade in the AbcC (Fig. 3e-f). Indeed, an infusion of an AP5 (2-amino-5-phosphonopentanoic acid)–CNQX (6-cyano-7-nitroquinoxaline-2,3-dione) mixture into the AcbC prior to BLA–M1 stimulation in the range of 100–300 ms IsI (Fig. 3e) completely abolished the increase and decrease in spike probability observed in up- and downregulated neurons, respectively (Fig. 3f). Thus all neurons then displayed a spike probability of ∼50%, being no longer different from the one set prior to any BLA–M1 co-stimulation. This result thereby demonstrates that the control exerted by the BLA over the function of aDLS pMSNS is dependent on antecedent glumatergic transmission in the AcbC.
Together these results demonstrate that associative processing in two domains of the amygdala, namely the BLA and the CeN, are both necessary remotely to engage, through parallel processes, functional shifts in the striatal locus supporting well-established cocaine-seeking behaviour.
The influence exerted by the BLA over aDLS pMSNs is bidirectional, potentially reflecting a dopamine-dependent modulation of D1 and D2 dopamine receptor containing neurons. Because this influence requires antecedent glutamatergic transmission in the AcbC, it may reflect the recruitment of the striato-nigro-striatal serial, or loop, circuitry13,25, providing an informational route whereby conditioned incentive processing in the BLA reaches, and controls, aDLS-dopamine-dependent instrumental response mechanisms. This observation warrants further research to understand the routes by which the BLA and CeN control the activity of aDLS MSNs in the course of the development of cocaine-seeking habits in behaving rats.
These results also demonstrate that a shift occurs within the amygdala from the BLA to the CeN to support well-established cocaine seeking. Thus, while the BLA is necessary to recruit aDLS-dopamine-dependent mechanisms, the CeN comes online later to maintain aDLS-dependent drug-seeking habits at a point when BLA manipulations are without effect. Thus, this parallel processing in the amygdala56 may provide two different informational routes whereby conditioned incentive processes control the aDLS-dopamine-dependent habit system5.
Future studies will be necessary to define the precise circuitry supporting these amygdalo-striatal interactions57, as well as the mechanisms influencing the shift from BLA-to-CeN control over aDLS mechanisms in the development of cocaine-seeking habits, which may depend on top-down influences of structures such as the infralimbic or insula cortex58,59.
However, the present studies provide clear evidence for independent amydgalo-striatal functional networks linking associative processes in the amygdala, that underlie the motivational salience of drug CSs, to the rigid habit system, resulting in the emergence and persistent performance of maladaptive cocaine-seeking habits5.
Male Lister Hooded rats (Charles River Laboratories, Kent, UK) weighing 396±5 g were singly housed under a 12-h reverse light/dark cycle (off 0700 hours). Following recovery from surgery and throughout the experiment, rats were fed 20 g chow per day with water available ad libitum. Experiments were performed 5–7 days per week during the dark phase. All procedures were conducted in accordance with the United Kingdom 1986 Animals (Scientific Procedures) Act, Project License 80/2234. The experiments were carried out with approval from the Home Office.
Cocaine hydrochloride (Macfarlan-Smith, Edinburgh, UK) and α-flupenthixol (Sigma-Aldrich, Poole, UK) were prepared as previously described13. Quinolinic acid (Sigma-Aldrich) was dissolved in sterile phosphate-buffered saline (PBS) and infused at 0.09 M, pH=7.4. Ibotenic acid (Abcam Biochemicals, Cambridge, UK) was dissolved in PBS at a concentration of 10 μg μl−1, pH=7.4 . The mixture of the GABA-B and GABA-A receptor agonists, baclofen, muscimol (B/M) (Sigma-Aldrich) was dissolved in PBS at the final concentration of 0.6 and 0.06 mM50. The selective D1 and D2 receptor antagonists SCH23390 and raclopride (Sigma-Aldrich) were dissolved in PBS at the concentration of 0.5, 1 or 1.5 μg μl−1 and 1, 2 and 3 μg μl−1, respectively. Drug doses are reported in the salt form.
Experiments were conducted using 12 standard operant conditioning chambers (Med Associates, St Albans, VT, USA) each fitted with a syringe pump and equipped with 2 cue lights above 2 levers and a houselight on the opposite wall, and were housed in sound- and light-attenuating cubicles as described previously41. Personal computers with Whisker software (Cardinal and Aitken, http://www.whiskercontrol.com) controlled infusions and light presentations and recorded lever presses.
The three behavioural experiments illustrated in Supplementary Fig. 1 aimed to test the effect of functionally disconnecting either the BLA or the CeN from dopaminergic transmission in the aDLS. The rationale for using such causal manipulations is provided in Supplementary Fig. 1.
All rats were anaesthetized with a mixture of ketamine hydrochloride (100 mg kg−1; Ketaset; Fort Dodge Animal Health Ltd, Southampton, UK) and xylazine (12 mg kg−1; Rompun; Bayer, Wuppertal, Germany) and implanted with an intravenous jugular catheter (Camcaths, Ely, UK) as described previously41,60. Rats were then immediately positioned in a stereotaxic frame (David Kopf Instruments, Tujunga, CA, USA) with the incisor bar set at −3.3 mm (ref. 61).
For experiment 1, rats were given either a unilateral BLA lesion, a unilateral CeN lesion or a sham lesion (couterbalanced). BLA lesions were made using two infusions (anteroposterior (AP)−2.3/−3, mediolateral (ML)+/−4.6, dorsoventral (DV)−7.3 (ref. 62)) of quinolinic acid (0.3 μl per 2 min followed by a 2-min diffusion time). CeN lesions were made using one infusion (AP−2.3, ML+/−4.2, DV−7.7 (ref. 63)) of ibotenic acid (0.25 μl per 6 min−1 followed by a 2-min diffusion time). AP and ML coordinates were measured from bregma, DV for BLA and CeN lesions were measured from dura and skull, respectively. All rats were subsequently implanted bilaterally with 22-gauge guide cannulae (Plastics One, Roanoke, VA, USA) positioned to lie 2 mm above the aDLS (AP+1.2, ML±3, DV−3)41 infusion target.
For experiment 2, rats were implanted with four 22-gauge guide cannulae, all bilaterally targeting the aDLS (AP+1.2, ML±3, DV−3) and then either bilaterally targeting the CeN (AP−2.3, ML±4, DV−3.7) or the BLA (AP−2.6, ML±4.6, DV−3.6). AP and ML coordinates were measured from bregma, DV from skull.
For Experiment 3, rats were then implanted bilaterally with 22-gauge guide cannulae positioned to lie 2 mm above the aDLS (AP+1.2, ML±3, DV−3 (ref. 40)).
Cannulae and catheters were secured and maintained as described previously40,41. Rats were treated daily from the day before to 7 days after surgery with 10 mg kg−1 s.c. of the antibiotic, Baytril (Bayer).
Intrastriatal and intra-amygdala infusions
Intrastriatal infusions (0.5 μl per side; bilateral for control group; unilateral for BLA-lesioned and CeN-lesioned groups—see Supplementary Fig. 2) were made via 28-gauge steel hypodermic injectors (Plastics One) lowered to the injection sites 2 mm (aDLS), 4 mm (CeN) or 5 mm (BLA) ventral to the end of the guide cannulae.
α-flupenthixol, SCH23390 and raclopride were, respectively, administered at the doses of 10 or 15 μg per infusion; 0.25, 0.5 and.75 μg per side and at 0.5, 1 and 1.5 mg per side. B/M was administered at 0.3/0.03 nmol per infusion, respectively. Infusions were made over 90 s using a syringe pump (Harvard Apparatus, Holliston, MA, USA) and were followed by a 60-s period to allow diffusion of the infused drug or vehicle before injectors were removed and obturators were replaced. Test sessions (see Procedures) began 5 min later.
For all pharmacological challenges, intracranial infusions were delivered according to a Latin-square design and were separated by at least two baseline sessions to control that stable levels of cocaine seeking or cocaine self-administration were maintained throughout.
For all the experimental groups, cocaine self-administration training sessions began 7 days following surgery (Supplementary Fig. 1). Cocaine (0.25 mg per infusion; 0.1 ml per 5 s−1) was available under a continuous reinforcement (that is, fixed-ratio 1 (FR1)) schedule. One active lever press resulted in an infusion and initiated a 20-s time out. Each cocaine infusion was accompanied by a 20-s illumination of the active lever-specific cue-light (CS). The houselight was extinguished and both levers were retracted during the time out. Pressing on the inactive lever was recorded but had no programmed consequence. The maximum number of available cocaine infusions during this stage was 30. Active and inactive lever assignment was counterbalanced. All test conditions were administered in a counterbalanced, Latin-square order of treatment, and sessions were conducted prior to, and were thus unaffected by, self-administered cocaine on that day (Fig. 1). Each test was immediate followed by the appropriate training session.
For Experiment 1 (Supplementary Fig. 1 top panel), following 10 training sessions under FR1, the effects of aDLS-dopamine receptor blockade on early-stage cocaine seeking were tested. Infusions (bilateral for control group, unilateral and contralateral to the lesion in BLA and CeN groups) of α-flupenthixol (0 and 10 μg per infusion) were made into the aDLS. During each 15-min test session, every active lever press resulted in a 1-s light CS presentation, and cocaine was only delivered on the first lever press following the 15-min interval (that is, FI15 (FR1:S)). Thus, the early-performance tests were conducted prior to, and were thus unaffected by, self-administered cocaine on that day (Fig. 1). Each test session was immediately followed by a FR1 cocaine self-administration training session (that is, up to 30 reinforcers in a 2-h session).
Following the early-performance cocaine-seeking tests, the response requirement was increased across the daily training sessions from FR1 to FR3, FR5(FR2:S), FR10(FR2:S) and ultimately FR10(FR4:S)41. Under each intermediate second-order schedule, completion of the unit schedule (given within parentheses) resulted in a 1-s CS light presentation; cocaine infusion and the 20-s time out were given only on completion of the overall schedule. Therefore, for the intermediate stage assessments, rats had been trained under conditions in which contingent presentations of the cocaine-associated CS occurred after 4 responses (FR4:S); cocaine was delivered on completion of the tenth set of 4 lever presses, that is, 40 lever presses. After reaching the FR10(FR4:S) schedule, rats began the transition-stage cocaine-seeking tests. During each 15-min test session with aDLS α-flupenthixol infusions (0 and 10 μg per infusion), every four active lever presses resulted in a 1-s light CS presentation, and cocaine was only delivered on the fourth lever press following the 15-min interval (that is, FI15(FR4:S)). Thus, the transition-stage performance tests were again conducted prior to, and were unaffected by, daily self-administered cocaine. Each test session was immediately followed an FR10(FR4:S) cocaine self-administration training session (30 reinforcers over 2 h).
The response requirements were then again increased through daily training sessions across the following reinforcement SOCR:FR10(FR6:S), FR10(FR10:S) and finally to an overall schedule of FI15(FR10:S)13,40,41 (Supplementary Fig. 1) during which responding was maintained by contingent presentation of the cocaine-associated Cs every tenth lever press (FR10:S), and each of the five daily cocaine infusions was delivered after the tenth lever press following completion of each 15-min fixed interval. Following 15 sessions, the habitual-stage tests were conducted in which the effects of aDLS α-flupenthixol infusions (0, 5, 10 and 15 μg per infusion) were assessed. The first 15-min interval of the schedule provides a time period in which no cocaine has been administered, yet rats are actively seeking cocaine.
Finally, on completion of contralateral testing rats were assessed for effects of ipsilateral aDLS-dopamine receptor blockade (0 and 10 μg per infusion) (Supplementary Fig. 5). Rats in the BLA and CeN groups were given infusions in the aDLS on the same side of the brain as their unilateral amygdala lesion; rats in the control group were given unilateral aDLS infusions.
For Experiment 2 (Supplementary Fig. 1 middle panel), following five training sessions under FR1 the effects of amygdala inactivation on early-stage cocaine seeking were tested. Bilateral infusions into the amygdala (either the CeN or the BLA depending on the group) of B/M22 or PBS (counterbalanced) were made. Dopamine receptor blockade in the aDLS was not assessed at this stage as it has already been established that it has no effect on cocaine-seeking at this point40,41.
The response requirement was then increased across the daily training sessions as described for Experiment 1 all the way to FI15(FR10:S). Following 15 sessions of training under this schedule, habitual-stage tests were conducted. Rats received bilateral inactivation (via B/M) of the amygdala (CeN or BLA), bilateral dopamine receptor blockade (α-flupenthixol, 10 μg per side) in the aDLS, a disconnection procedure involving unilateral amygdala inactivation with contralateral aDLS-dopamine receptor blockade and an ipsilateral procedure in which both the amygdala and aDLS were treated in the same hemisphere. For each of these tests, PBS control infusions were intermixed.
For Experiment 3 (Supplementary Fig. 1 bottom panel), following 10 sessions under FR1 rats were trained under fixed interval SOCR with daily increases in the interval duration, from 1 to 15 min as previously described13. Rats were then trained for 20 sessions under the FI15(FR10:S) schedule prior to being challenged with bilateral intra-aDLS infusions of vehicle, SCH23390 or raclopride.
Despite the different time course in the functional control over the recruitment of aDLS-dependent cue-controlled cocaine cocaine-seeking behaviour between the BLA and the CeN and their dissociable neural targets, we wanted to identify better the neural pathway whereby the BLA controls aDLS function. Such remote functional connectivity having never been tested before, we decided to use extracellular electrophysiological recordings of aDLS medium spiny neurons to measure the nature and the characteristics of the influence of BLA stimulation over the spike probability of these neurons.
Male Sprague–Dawley rats (Charles River, France) weighing 300 g at experiment start were housed two per cage under conditions similar to the behavioural experiments.
Rats were anesthetized by a single i.p. injection of urethane (1.7 g kg−1, in distilled water). Rats were positioned in a stereotaxic frame (m2e-Unimecanique, France) and maintained at 37 °C with a homeothermic blanket. Burr holes were drilled in the cleaned skull above the electrode targets, and the overlying dura was carefully resected. The coordinates are relative to bregma in the right hemisphere.
A concentric electrode (SNEX-100, Phymep) was implanted in M1 (AP+4.15; ML+3 ; DV−2.5). A current pulse (0.5-ms long) was applied by an isolated stimulator (DS3 digitimer) triggered by a 1401 Plus system (Cambridge Electronic Design, Cambridge, UK). Because the target aDLS MSNs are silent, cortical stimulation was continuously applied at 0.3 Hz. A bipolar electrode (SNEX-200, Phymep) was lowered into the BLA (AP−2.5; ML+5; DV−7.3). The two poles of the electrodes are oriented in the same sagittal plan.
Electrophysiological recordings and neuron selection
The recording electrode was lowered to the aDLS (AP+1.8; ML+3, from DV−3.3 to DV−4.4) ipsilateral to the stimulating electrodes. The recording electrodes were pulled from borosilicate glass capillaries (GC150F, Harvard Apparatus, UK) using a P97 micropipette puller (Sutter instrument, CA, USA). The tip was broken back under microscopic control to achieve an impedance of 10–15 MΩ as measured in situ with an axoclamp2B (Axon instruments, Foster City, CA) by bridge balance. The electrodes were filled with 0.4 M NaCl. Through the electrode, the extracellular potential was recorded with an axoclamp2B amplifier in the bridge mode versus a reference electrode maintained in contact with the skull skin by a sponge moistened with 0.9% NaCl. The signal was amplified 10 × by the axoclamp2B, further amplified 100 × and band pass filtered (low-pass filter at 300 Hz and high-pass filter at 20 kHz) via a differential AC amplifier (model 1700; AM systems, Carlsborg, WA). Spike occurrence was continuously recorded by a 1401 Plus Cambridge Electronic Design system running Spike 2.
The neurons of interest were isolated by slowly lowering the recording electrode in the aDLS while a single electrical stimulation pulse was delivered to M1 at an overall rate of 0.3 Hz. A total of ∼50 striatal neurons that were excited by stimulation of M1 were recorded. Because MSNs compose ∼95% of striatal neurons64,65 and because putative interneurons were not tested, almost all of the neurons recorded in this study were likely to be the medium spiny subtype identified after stimulation of the primary glutamatergic input to the DLS, namely M1. Neurons were considered as pMSNs based on their electrophysiological properties: a spike duration above 1 ms over 20 trials and the shape of their action potential, as previously described51,66,67 (Supplementary Fig. 6A).
Neurons with sustained spontaneous firing, which produced more than two spikes following a single electrical stimulation, with a spike duration <1 ms were considered to be putative interneurons, and were not analysed.
BLA stimulation protocol
Once a pMSN was found, the effect of the pre-stimulation of the BLA on the M1-evoked spike probability was tested. M1 stimulation current intensity was adjusted to produce about 50% evoked spike probability (the current intensity was adjusted for each neuron). To obtain a stable baseline spiking probability, M1 stimulation was delivered for 20 trials at a rate of 0.3 Hz.
To test the effect of the BLA stimulation alone on the spike probability of the selected pMSNs, M1 stimulation was turned off and BLA stimulation was applied alone for at least 40 trials. Following these 20 trials, a single BLA stimulus pulse (.5 ms, .5 mA) was delivered before the M1 stimulation pulse for 20 trials with IsI ranging from 1 to 1,000 ms (IsI tested: 1, 5, 50, 100, 200, 300, 500, 1,000 ms). These IsI, which were applied following a Latin-square design, were chosen to test mono- and poly-synaptic delays.
AcbC glutamate receptor blockade
AcbC glutamatergic receptors were blocked by an infusion of a mixture of the AMPA (α-amino-3-hydroxy-5-methylisoxazole-4-proprionic acid) and NMDA (N-methyl-d-aspartate) receptor antagonists—CNQX and AP568,69. An injection electrode filled with the CNQX/APV solution (4 mg ml−1 and 8 mg ml−1, respectively) was lowered into the AcbC (AP+1.6; ML+1.4; DV−7) ipsilateral to the stimulating and recording electrodes. The injection electrode was pulled from borosilicate glass capillaries (GC100F, Harvard Apparatus) using a P97 micropipette puller. A scale system on the pipette allowed us to accurately measure each injected volume by visual observation of the meniscus. After the injection, the absence of meniscus movement showed that no spontaneous leakage occurred. The pulled tip of the injection pipette was broken under a microscopic control at an external diameter of ∼30–40 μm. A volume of 0.5 μl of CNQX/AP5 was injected into the AcbC11 with a picoinjector (UC-8 controller, Bioscience tools). BLA–M1 co-stimulations were thus applied and we measured the effect of an infusion of the CNQX/APV mixture into the AcbC on the influence exerted by BLA pre-stimulation over spike probability of aDLS pMSNs in response to M1 stimulation for the specific 100–200–300 ms IsI, those at which the recorded neurons in the aDLS had been shown to be sensitive to BLA stimulation prior to M1 stimulation.
Data and statistical analyses
Data are presented as mean±s.e.m. except when individual points are displayed (Fig. 2b). Analyses were performed using Statsoft STATISTICA 10 software.
Assumptions about normality of distribution and homogeneity of variance were assessed using the Kolmogorov–Smirnov and Levene tests, respectively. In case of violation of at least one of these assumptions, data sets were log-transformed (Fig. 2b,c).
For behavioural studies, lever presses during each 15-min drug-seeking tests were analysed using two-way analyses of variance (ANOVAs) with lever, dose and training level as within-subject factors.
For electrophysiological studies, spike probability evoked by M1 stimulation was calculated as the ratio between the number of evoked APs and the number of stimulation cycles. Change in probability of spike discharge evoked by BLA stimulation was calculated by comparing the probability of spike discharge in M1 stimulation-only situation and the spike probability on each pre-stimulation condition.
A cluster analysis on average spike probability of pMSNs, obtained for the BLA–M1 IsI of 100, 200 and 300 ms, identified three neuronal populations according to the nature of BLA pre-stimulation influence on the spike probability of discharge evoked by the M1 stimulation.
Latency to AP and AP duration were calculated as previously described66. Briefly, the AP duration was measured as the time between start of depolarization and the depolarization of each AP evoked by M1 stimulation. AP waveforms were averaged using the spike 2 software and extracted as the shape of the APs of the recorded neuron.
When no AP was recorded in some downregulated neurons (on only four occasions), provided there was no effect of the stimulation on the AP duration and latency to first AP overall, the value was computed as the average of the other values (50, 100, 200 or 300 ms IsI). Variations in spike probability, latency to first AP and AP duration were analysed by ANOVA with neuronal type as between-subject factor and stimulation condition (including IsI) as within-subject factors.
Significance was set at α=0.05. Significant interactions were analysed further using the Dunnet post hoc or Newman–Keuls post hoc tests where appropriate. Tests used are stated in either the main text or figure legends. Effect sizes are reported using partial η2 values70.
How to cite this article: Murray, J. E. et al. Basolateral and central amygdala differentially recruit and maintain dorsolateral striatum-dependent cocaine-seeking habits. Nat. Commun. 6:10088 doi: 10.1038/ncomms10088 (2015).
Wise, R. & Koob, G. The development and maintenance of drug addiction. Neuropsychopharmacology 39, 254–262 (2014).
Everitt, B. J. Neural and psychological mechanisms underlying compulsive drug-seeking habits and drug memories--indications for novel treatments of addiction. Eur. J. Neurosci. 40, 2163–2182 (2014).
Volkow, N., Wang, G., Tomasi, D. & Baler, R. Unbalanced neuronal circuits in addiction. Curr. Opin. Neurobiol. 23, 639–648 (2013).
Everitt, B. J. & Robbins, T. W. Neural systems of reinforcement for drug addiction: from actions to habits to compulsion. Nat. Neurosci. 8, 1481–1489 (2005).
Belin, D., Belin-Rauscent, A., Murray, J. E. & Everitt, B. J. Addiction: failure of control over maladaptive incentive habits. Curr. Opin. Neurobiol. 23, 564–572 (2013).
Zapata, A., Minney, V. & Shippenberg, T. Shift from goal-directed to habitual cocaine seeking after prolonged experience in rats. J. Neurosci. 30, 15457–15463 (2010).
Olmstead, M., Lafond, M., Everitt, B. & Dickinson, A. Cocaine seeking by rats is a goal-directed action. Behav. Neurosci. 115, 394–402 (2001).
Di Ciano, P. & Everitt, B. J. Direct interactions between the basolateral amygdala and nucleus accumbens core underlie cocaine-seeking behavior by rats. J. Neurosci. 24, 7167–7173 (2004).
Everitt, B., Cardinal, R., Hall, J., Parkinson, J. & Robbins, T in The Amygdala: a Functional Analysis ed. Aggleton J.P. 353–390Oxford Univ. Press 2000.
Cador, M., Robbins, T. W. & Everitt, B. J. Involvement of the amygdala in stimulus-reward associations: interaction with the ventral striatum. Neuroscience 30, 77–86 (1989).
Ambroggi, F., Ishikawa, A., Fields, H. & Nicola, S. Basolateral amygdala neurons facilitate reward-seeking behavior by exciting nucleus accumbens neurons. Neuron 59, 648–661 (2008).
Whitelaw, R. B., Markou, A., Robbins, T. W. & Everitt, B. J. Excitotoxic lesions of the basolateral amygdala impair the acquisition of cocaine-seeking behaviour under a second-order schedule of reinforcement. Psychopharmacology (Berl) 127, 213–224 (1996).
Belin, D. & Everitt, B. J. Cocaine-seeking habits depend upon dopamine-dependent serial connectivity linking the ventral with the dorsal striatum. Neuron 57, 432–441 (2008).
Willuhn, I., Burgeno, L., Everitt, B. & Phillips, P. Hierarchical recruitment of phasic dopamine signaling in the striatum during the progression of cocaine use. Proc. Natl Acad. Sci. USA 109, 20703–20708 (2012).
Belin, D., Jonkman, S., Dickinson, A., Robbins, T. W. & Everitt, B. J. Parallel and interactive learning processes within the basal ganglia: Relevance for the understanding of addiction. Behav. Brain Res. 199, 89–102 (2009).
Everitt, B. J., Belin, D., Dalley, J. & Robbins, T. W. in Dopamine Handbook eds Iversen S., Iversen L., Dunnett S., Bjorklund A. Oxford Univ. Press (2008).
Besson, M. et al. Cocaine modulation of frontostriatal expression of Zif268, D2, and 5-HT2c receptors in high and low impulsive rats. Neuropsychopharmacology 38, 1963–1973 (2013).
Porrino, L. J., Smith, H. R., Nader, M. A. & Beveridge, T. J. The effects of cocaine: a shifting target over the course of addiction. Prog. Neuropsychopharmacol. Biol. Psychiatry 31, 1593–1600 (2007).
Porrino, L. Cocaine self-administration produces a progressive involvement of limbic, association, and sensorimotor striatal domains. J. Neurosci. 24, 3554–3562 (2004).
Letchworth, S. R., Nader, M. A., Smith, H. R., Friedman, D. P. & Porrino, L. Progression of changes in dopamine transporter binding site density as a result of cocaine self-administration in rhesus monkeys. J. Neurosci. 21, 2799–2807 (2001).
Vanderschuren, L., Di Ciano, P. & Everitt, B. Involvement of the dorsal striatum in cue-controlled cocaine seeking. J. Neurosci. 25, 8665–8670 (2005).
Jonkman, S., Pelloux, Y. & Everitt, B. Differential roles of the dorsolateral and midlateral striatum in punished cocaine seeking. J. Neurosci. 32, 4645–4650 (2012).
Vollstadt-Klein, S. et al. Initial, habitual and compulsive alcohol use is characterized by a shift of cue processing from ventral to dorsal striatum. Addiction 105, 1741–1749 (2010).
Amercian Psychiatric Association. The Diagnostic and Statistical Manual of Mental Disorders: DSM 5 bookpointUS (2013).
Haber, S., Fudge, J. & McFarland, N. Striatonigrostriatal pathways in primates form an ascending spiral from the shell to the dorsolateral striatum. J. Neurosci. 20, 2369–2382 (2000).
Ikemoto, S. Dopamine reward circuitry: Two projection systems from the ventral midbrain to the nucleus accumbens–olfactory tubercle complex. Brain Res. Rev. 56, 27–78 (2007).
Belin, D. & Everitt, B. J. in Handbook of Basal Ganglia Structure and Function eds Steiner Heinz, tseng Kuei Ch. 13, 571–592 Elsevier, Academic Press (2010).
Kelley, A. E., Domesick, V. B. & Nauta, W. J. The amygdalostriatal projection in the rat--an anatomical study by anterograde and retrograde tracing methods. Neuroscience 7, 615–630 (1982).
Stuber, G. et al. Excitatory transmission from the amygdala to nucleus accumbens facilitates reward seeking. Nature 475, 377–380 (2011).
Wingard, J. & Packard, M. The amygdala and emotional modulation of competition between cognitive and habit memory. Behav. Brain Res. 193, 126–131 (2008).
Packard, M. G. & Wingard, J. C. Amygdala and ‘emotional’ modulation of the relative use of multiple memory systems. Neurobiol. Learn. Mem. 82, 243–252 (2004).
Namburi, P. et al. A circuit mechanism for differentiating positive and negative associations. Nature 520, 675–678 (2015).
Lee, B. et al. Maturation of silent synapses in amygdala-accumbens projection contributes to incubation of cocaine craving. Nat. Neurosci. 16, 1644–1651 (2013).
Volkow, N. et al. Cocaine cues and dopamine in dorsal striatum: mechanism of craving in cocaine addiction. J. Neurosci. 26, 6583–6588 (2006).
Schindler, C., Panlilio, L. & Goldberg, S. Second-order schedules of drug self-administration in animals. Psychopharmacology (Berl) 163, 327–344 (2002).
Goldberg, S. R., Kelleher, R. T. & Morse, W. H. Second-order schedules of drug injection. Fed. Proc. 34, 1771–1776 (1975).
Belin-Rauscent, A., Fouyssac, M., Bonci, A. & Belin, D. How preclinical models evolved to resemble the diagnostic criteria of drug addiction. Biol. Psychiatry doi:10.1016/j.biopsych.2015.01.004 (2015).
Belin-Rauscent, A. & Belin, D. ‘Addictions—From Pathophysiology to Treatment’ 3–20InTech (2012).
Everitt, B. & Robbins, T. Second-order schedules of drug reinforcement in rats and monkeys: measurement of reinforcing efficacy and drug-seeking behaviour. Psychopharmacology 153, 17–30 (2000).
Murray, J. E., Belin, D. & Everitt, B. J. Double dissociation of the dorsomedial and dorsolateral striatal control over the acquisition and performance of cocaine seeking. Neuropsychopharmacology 37, 2456–2466 (2012).
Murray, J. E. et al. Increased impulsivity retards the transition to dorsolateral striatal dopamine control of cocaine seeking. Biol. Psychiatry 76, 15–22 (2014).
Killcross, S., Robbins, T. W. & Everitt, B. J. Different types of fear-conditioned behaviour mediated by separate nuclei within amygdala. Nature 388, 377–380 (1997).
El-Amamy, H. & Holland, P. Dissociable effects of disconnecting amygdala central nucleus from the ventral tegmental area or substantia nigra on learned orienting and incentive motivation. Eur. J. Neurosci. 25, 1557–1567 (2007).
Ferreira, T. L. et al. The indirect amygdala-dorsal striatum pathway mediates conditioned freezing: insights on emotional memory networks. Neuroscience 153, 84–94 (2008).
Holland, P. & Gallagher, M. Double dissociation of the effects of lesions of basolateral and central amygdala on conditioned stimulus-potentiated feeding and Pavlovian-instrumental transfer. Eur. J. Neurosci. 17, 1680–1694 (2003).
Corbit, L. H. & Balleine, B. Double dissociation of basolateral and central amygdala lesions on the general and outcome-specific forms of Pavlovian-instrumental transfer. J. Neurosci. 25, 962–970 (2005).
Robinson, M. J., Warlow, S. M. & Berridge, K. C. Optogenetic excitation of central amygdala amplifies and narrows incentive motivation to pursue one reward above another. J. Neurosci. 34, 16567–16580 (2014).
Lingawi, N. W. & Balleine, B. W. Amygdala central nucleus interacts with dorsolateral striatum to regulate the acquisition of habits. J. Neurosci. 32, 1073–1081 (2012).
Koob, G. et al. Addiction as a stress surfeit disorder. Neuropharmacology 76, (Pt B): 370–382 (2013).
McFarland, K. & Kalivas, P. W. The circuitry mediating cocaine-induced reinstatement of drug-seeking behavior. J. Neurosci. 21, 8655–8663 (2001).
McGinty, V. & Grace, A. Selective activation of medial prefrontal-to-accumbens projection neurons by amygdala stimulation and Pavlovian conditioned stimuli. Cereb. Cortex 18, 1961–1972 (2008).
Ebrahimi, A., Pochet, R. & Roger, M. Topographical organization of the projections from physiologically identified areas of the motor cortex to the striatum in the rat. Neurosci. Res. 14, 39–60 (1992).
McGeorge, A. & Faull, R. The organization of the projection from the cerebral cortex to the striatum in the rat. Neuroscience 29, 503–537 (1989).
Zorrilla, E. & Koob, G. Amygdalostriatal projections in the neurocircuitry for motivation: a neuroanatomical thread through the career of Ann Kelley. Neurosci. Biobehav. Rev. 37, (9 Pt A): 1932–1945 (2012).
McDonald, A. J. Topographical organization of amygdaloid projections to the caudatoputamen, nucleus accumbens, and related striatal-like areas of the rat brain. Neuroscience 44, 15–33 (1991).
Balleine, B. W. & Killcross, S. Parallel incentive processing: an integrated view of amygdala function. Trends Neurosci. 29, 272–279 (2006).
Tye, K. M. & Deisseroth, K. Optogenetic investigation of neural circuits underlying brain disease in animal models. Nat. Rev. Neurosci. 13, 251–266 (2012).
Coutureau, E. & Killcross, S. Inactivation of the infralimbic prefrontal cortex reinstates goal-directed responding in overtrained rats. Behav. Brain Res. 146, 167–174 (2003).
Parkes, S. L. & Balleine, B. W. Incentive memory: evidence the basolateral amygdala encodes and the insular cortex retrieves outcome values to guide choice between goal-directed actions. J. Neurosci. 33, 8753–8763 (2013).
Ducret, E. et al. N-acetylcysteine facilitates self-imposed abstinence after escalation of cocaine intake. Biol. Psychiatry doi:10.1016/j.biopsych.2015.09.019 (2015).
Paxinos, G. & Watson, C. The Rat Brain in Stereotaxic Coordinates Academic Press (1986).
Ito, R., Everitt, B. J. & Robbins, T. W. The hippocampus and appetitive Pavlovian conditioning: effects of excitotoxic hippocampal lesions on conditioned locomotor activity and autoshaping. Hippocampus 15, 713–721 (2005).
Han, J. S., McMahan, R. W., Holland, P. & Gallagher, M. The role of an amygdalo-nigrostriatal pathway in associative learning. J. Neurosci. 17, 3913–3919 (1997).
Jiang, Z. & North, R. Membrane properties and synaptic responses of rat striatal neurones in vitro. J. Physiol. 443, 533–553 (1991).
Kemp, J. & Powell, T. The synaptic organization of the caudate nucleus. Philos. Trans. R. Soc. Lond. B Biol. Sci. 262, 403–412 (1971).
Mallet, N., Le Moine, C., Charpier, S. & Gonon, F. Feedforward inhibition of projection neurons by fast-spiking GABA interneurons in the rat striatum in vivo. J. Neurosci. 25, 3857–3869 (2005).
Mallet, N., Ballion, B., Le Moine, C. & Gonon, F. Cortical inputs and GABA interneurons imbalance projection neurons in the striatum of parkinsonian rats. J. Neurosci. 26, 3875–3884 (2006).
Nicola, S., Kombian, S. & Malenka, R. Psychostimulants depress excitatory synaptic transmission in the nucleus accumbens via presynaptic D1-like dopamine receptors. J. Neurosci. 16, 1591–1604 (1996).
Nicola, S., Surmeier, D. J. & Malenka, R. C. Dopaminergic modulation of neuronal excitability in the striatum and nucleus accumbens. Annu. Rev. Neurosci. 23, 185–215 (2000).
Levine, T. R. & Hullett, C. R. Eta squared, partial eta squared, and misreporting of effect size in communication research. Hum. Commun. Res. 28, 612–625 (2002).
This work was supported by the Fondation pour la Recherche Médicale (FRM), the United Kingdom Medical Research Council (MRC) Grant 9536855 to B.J.E., the AXA research fund to A.B.-R., an INSERM Avenir and an Agence Nationale de la Recherche (ANR) grant ANR12 SAMA00201 to D.B. Research was conducted within both the MRC/Wellcome Trust Behavioural and Clinical Neuroscience Institute of Cambridge University and the Inserm team ‘Psychobiology of Compulsive Disorders’, the University of Poitiers. We thank Alan Lyon for his technical assistance, Laurence Turi for her help in the animal facility of Poitiers and Vince McGinty for his help with extracellular electrophysiology.
The authors declare no competing financial interests.
About this article
Cite this article
Murray, J., Belin-Rauscent, A., Simon, M. et al. Basolateral and central amygdala differentially recruit and maintain dorsolateral striatum-dependent cocaine-seeking habits. Nat Commun 6, 10088 (2015). https://doi.org/10.1038/ncomms10088
Nature Reviews Neuroscience (2020)
Nature Reviews Neuroscience (2020)
Cellular and Molecular Life Sciences (2020)
Random access to palatable food stimulates similar addiction-like responses as a fixed schedule, but only a fixed schedule elicits anticipatory activation
Scientific Reports (2019)