D1 dopamine receptor-mediated LTP at GABA synapses encodes motivation to self-administer cocaine in rats

Abstract

Enhanced motivation to take drugs is a central characteristic of addiction, yet the neural underpinning of this maladaptive behavior is still largely unknown. Here, we report a D1-like dopamine receptor (DRD1)-mediated long-term potentiation of GABAA-IPSCs (D1-LTPGABA) in the oval bed nucleus of the stria terminalis that was positively correlated with motivation to self-administer cocaine in rats. Likewise, in vivo intra-oval bed nucleus of the stria terminalis DRD1 pharmacological blockade reduced lever pressing for cocaine more effectively in rats showing enhanced motivation toward cocaine. D1-LTPGABA resulted from enhanced function and expression of G-protein-independent DRD1 coupled to c-Src tyrosine kinases and required local release of neurotensin. There was no D1-LTPGABA in rats that self-administered sucrose, in those with limited cocaine self-administration experience, or in those that received cocaine passively (yoked). Therefore, our study reveals a novel neurophysiological mechanism contributing to individual motivation to self-administer cocaine, a critical psychobiological element of compulsive drug use and addiction.

Figure 1.

Effect of intra-ovBNST DRD1 blockade on sucrose and cocaine self-administration. a, Schedule for oral sucrose or intravenous cocaine self-administration (0.75 mg/ml/injection). S-A, self-administration. b, Effect of intra-ovBNST or ICV microinjections of the DRD1 antagonist SCH-23390 on sucrose or cocaine self-administration on final ratio pressing (top), time to reach breaking point (BP, middle), and time to reach ½ BP (bottom). Asterisks indicate significant difference compared with vehicle, p < 0.05. c, Representative experiments showing the effect of intra-ovBNST SCH-23390 (1.6 μg/side, solid lines) compared with baseline (dotted lines) on sucrose (left) and cocaine (right) self-administration. Oblique tick marks indicate reward deliveries. BPs defined as described in the Materials and Methods are circled for each presented conditions. d, Left: Effect of SCH (1.6 μg/side) compared with vehicle on final ratio pressing for cocaine in each rat tested. Right: Correlation between SCH (1.6 μg/side)-induced reduction in final ratio pressing and magnitude of final ratio pressing for cocaine. Each dot represents data from one rat. Dotted lines indicate 95% confidence interval. e, Sum of fluorescent microspheres spread (dark gray areas) summarizing successful placements from which data were used in b. f, Representative histological location of a typical microinjection of fluorescent microspheres seen in bright- or dark-field microscopy. Schematics of coronal brain sections are adapted from Paxinos and Watson (2005). Lv, lateral ventricles; ac, anterior commissure; ic, internal capsule.

Figure 2.

Effect of DRD1 activation on ovBNST GABA_A-IPSC. a, Schedule of acquisition and maintenance of sucrose and intravenous cocaine self-administration (0.75 mg/ml/injection). S-A, self-administration. b, Representative traces showing the effect of DA (1 μm) in the presence of sulpiride on GABA_A-IPSC amplitude and PPRs. S1 and S2 represent the maximum GABA_A-IPSC amplitude in response to two electrical stimulations given at 20 Hz to measure changes in PPR summarized in f. c, Effect of a 5 min bath application of SKF (1 μm; black horizontal bar) on the amplitude of electrically evoked ovBNST GABA_A-IPSC (0.1 Hz) as a function of time in 5 experimental groups of rats. Evoked events were binned (1 min, 6 events) and data points and error bars represent means ± SEM across all recorded neurons within each experimental group. Numbers in parentheses indicate sample size (neurons/rats). d, Effect of a 5 min bath application of DA (1 μm; black horizontal bar) in the presence of sulpiride (10 μm) on the amplitude of electrically evoked ovBNST GABA_A-IPSC (0.1 Hz) as a function of time in 5 experimental groups of rats. e, Effect of DRD1 activation with DA/sulpiride (black horizontal bar) on the amplitude of electrically evoked GABA_A-IPSC as a function of time in ovBNST neurons of Cocaine Maintenance rats in the presence of the DRD1 antagonist SCH-23390 (10 μm) applied throughout (right, gray bar) the recordings or at the peak (middle, hatched bar) of the DRD1-mediated response. Each data point represents the amplitude of one electrically evoked (0.1 Hz) GABA_A-IPSC. Inset: Bar chart summarizing the effects of bath application of the DRD1 antagonist SCH-23390 on DRD1-mediated increase in GABA_A-IPSC amplitude calculated at time 20–25 min (averages of 30 events). f, Summary of PPR calculated before (0′) and 25 min (30′) after DRD1 activation in ovBNST neurons of Cocaine Maintenance rats. Each dot represents individual neurons and white symbols are means ± SEM of PPRs at times 0 and 30′. g, CV analysis of the effects of DRD1 activation on the amplitude of evoked GABA_A-IPSC in ovBNST neurons from Cocaine rats. Each symbol (SKF, squares; DA, circles) represents the data from one neuron (n = 23, combined recordings reported in c and d). Baseline data were calculated from 30 events preceding DRD1 activation (time −5 to 0) and DRD1 data from at time 25–30 min. Dot plot shows an index of the between evoked GABA_A-IPSC variability, r = [(CV_baseline²)/(CV_DRD1²)] as a function of DRD1-induced change in GABA_A-IPSC amplitude, Π = Peak amplitude_DRD1/Peak amplitude_baseline. A positive linear relationship shows a reduction in between evoked GABA_A-IPSC variability with increased amplitude, suggesting enhanced efficacy in neurotransmitter release, indicative of a presynaptic effect. h, DRD1-induced change in GABA_A-IPSC amplitude (time 25–30 min) as a function of final ratio pressing for cocaine. Each symbol (SKF, squares; DA, circles) represents the average of all neurons from a single rat (n = 12, same dataset as in Fig. 2c,d). Dotted lines indicate 95% confidence interval. i, DRD1-induced change in GABA_A-IPSC amplitude (time 25–30 min) as a function of the number of days that individual rats spent on the PR schedule of cocaine reinforcement. Each symbol (SKF, squares; DA, circles) represents the average of all neurons from a single rat (n = 12, same dataset as in Fig. 2c,d). Dotted lines indicate 95% confidence interval.

Figure 3.

Effect of cocaine self-administration on ovBNST DRD1 expression. Immunohistochemical detection of DRD1 in the ovBNST of Control (a,b) and Cocaine Maintenance rats (c,d). The representative weak DRD1 signal in ovBNST soma of Control rats (b) becomes intense in both the soma and proximal dendrites of ovBNST neurons from a Cocaine Maintenance rat (d). Scale bars in a, c, 100 μm; b, d, 10 μm. e, Number of ovBNST DRD1-positive neurons in Control (n = 5) or Cocaine Maintenance (n = 6) rats. f, Perimembranous ovBNST neuron optical density measurement of the DRD1 immunostaining in the same groups. Results represent the means ± SEM. Asterisks indicate significant difference compared with controls, p < 0.05. ov, ovBNST; ac, anterior commissure; st, striatum.

Figure 4.

Role of neurotensin in ovBNST D1-LTP_GABA. a, Effect of neurotensin (black horizontal bar, 1 μm) on the amplitude of evoked GABA_A-IPSC in ovBNST neurons of Control and Cocaine Maintenance rats as a function of time. Evoked events were binned (1 min, 6 events) and data point and error bars represent means ± SEM across all recorded neurons within each experimental group. Numbers in parentheses indicate sample size (neurons/rats). b, Representative traces showing the effect of neurotensin on the amplitude and PPR of GABA_A-IPSC measured in ovBNST neurons of Cocaine Maintenance rats. S1 and S2 represent the maximum GABA_A-IPSC amplitudes in response to two electrical stimulations given at 20 Hz to measure changes in PPR summarized in c. c, Summary of the effects of neurotensin on the PPR of GABA_A-IPSC in Cocaine Maintenance rats. Each black circle represents individual neurons and white circles are means ± SEM of PPRs at times 0 and 10′. d, Effect of postsynaptic depolarization (−70 to 0 mV, 100 ms, 2 Hz, 5 min) on the amplitude of evoked GABA_A-IPSC as a function of time in ovBNST neurons of Cocaine Maintenance rats in the absence (circles) or presence (triangles) of the NTS antagonist SR-14298 (10 μm). Evoked events were binned (1 min, 6 events) and data point and error bars represent means ± SEM across all recorded neurons. Numbers in parentheses indicate sample size (neurons/rats). e, Representative traces showing the effect of postsynaptic depolarization on the amplitude and PPR of GABA_A-IPSC measured in ovBNST neurons of Cocaine Maintenance rats. f, Summary of the effects of neurotensin on the PPR of GABA_A-IPSC in Cocaine Maintenance rats. Each dot represents individual neurons and white circles are means ± SEM of PPRs at times 0 and 30′. g, Effect of neurotensin (black horizontal bar, 1 μm) on the amplitude of evoked GABA_A-IPSC as a function of time in ovBNST neurons of Cocaine Maintenance rats in the presence of the neurotensin receptor antagonist SR-14298 (10 μm) applied throughout (gray bar) the recordings or at the peak (chaser, hatched bar) of the DRD1-mediated response. Each data point represents an electrically evoked (0.1 Hz) GABA_A-IPSC. h, Bar chart summarizing the effects of the neurotensin antagonist SR-14298 (1 μm) on DRD1-LTP_GABA in ovBNST neurons of Cocaine Maintenance rats calculated at time 25–30 min.

Figure 5.

Intracellular signal underlying ovBNST D1-LTP_GABA. a, Effect of DRD1 activation (black horizontal bar) on the amplitude of evoked ovBNST GABA_A-IPSCs in brain slices of Cocaine Maintenance rats. H89 (PKA inhibitor), U73122 (PLC inhibitor), genistein (TK inhibitor), daidzein (inactive analog of genistein), PP2 (c-Src inhibitor), PP3 (inactive analog of PP2), GDP-βs (G-protein blocker) were added intracellularly and were thus present during the complete duration of the recordings. Evoked events were binned (1 min, 6 events) and data point and error bars represent means ± SEM across all recorded neurons within each experimental group. Numbers in parentheses indicate the sample sizes (neurons/rats). b, Bar charts summarizing the effects of signaling molecules (in μm) and enzymes inhibitors on ovBNST D1-LTP_GABA. Data were calculated from the last 5 min of recording.

Figure 6.

DA modulation of ovBNST GABA_A-IPSC before and after cocaine self-administration. In the ovBNST of Control rats, DA binds to presynaptic DRD2 and decreases GABA release, resulting in a reversible reduction in GABA_A-IPSC amplitude (Krawczyk et al., 2011a, 2011b). With a prolonged cocaine self-administration experience, presynaptic DRD2 receptors are substituted with postsynaptic DRD1 and DRD2 (Krawczyk et al., 2011b). The resulting effect of postsynaptic activation of DRD1 and DRD2 is DRD1-LTP_GABA, which involves c-Src-dependent release of neurotensin that increases presynaptic GABA release.