Skip to main content
NCBI home page
Neuropsychopharmacology logoLink to Neuropsychopharmacology
. 2016 Aug 24;42(3):707–715. doi: 10.1038/npp.2016.138

Selective Ablation of GIRK Channels in Dopamine Neurons Alters Behavioral Effects of Cocaine in Mice

Nora M McCall 1,5, Lydia Kotecki 2,5, Sergio Dominguez-Lopez 3, Ezequiel Marron Fernandez de Velasco 2, Nicholas Carlblom 2, Amanda L Sharpe 3,4, Michael J Beckstead 3, Kevin Wickman 2,*
PMCID: PMC5240170  PMID: 27468917

Abstract

The increase in dopamine (DA) neurotransmission stimulated by in vivo cocaine exposure is tempered by G protein-dependent inhibitory feedback mechanisms in DA neurons of the ventral tegmental area (VTA). G protein-gated inwardly rectifying K+ (GIRK/Kir3) channels mediate the direct inhibitory effect of GABAB receptor (GABABR) and D2 DA receptor (D2R) activation in VTA DA neurons. Here we examined the effect of the DA neuron-specific loss of GIRK channels on D2R-dependent regulation of VTA DA neuron excitability and on cocaine-induced, reward-related behaviors. Selective ablation of Girk2 in DA neurons did not alter the baseline excitability of VTA DA neurons but significantly reduced the magnitude of D2R-dependent inhibitory somatodendritic currents and blunted the impact of D2R activation on spontaneous activity and neuronal excitability. Mice lacking GIRK channels in DA neurons exhibited increased locomotor activation in response to acute cocaine administration and an altered locomotor sensitization profile, as well as increased responding for and intake of cocaine in an intravenous self-administration test. These mice, however, showed unaltered cocaine-induced conditioned place preference. Collectively, our data suggest that feedback inhibition to VTA DA neurons, mediated by GIRK channel activation, tempers the locomotor stimulatory effect of cocaine while also modulating the reinforcing effect of cocaine in an operant-based self-administration task.

INTRODUCTION

Dopamine (DA) neurons of the ventral tegmental area (VTA) are an integral part of the mesocorticolimbic system, a network of brain regions involved in reward-related behavior. Most drugs of abuse share the ability to increase extracellular levels of DA within this circuit (Nestler, 2005). Cocaine enhances DA neurotransmission by inhibiting transporters that remove DA from the extracellular space, allowing levels of DA to rise in downstream targets of DA neurons (Di Chiara and Imperato, 1988). Elevated DA signaling triggered by cocaine is implicated in behavioral effects, including locomotor stimulation and sensitization, and conditioned place preference (CPP; Pierce and Kalivas, 1997; Zweifel et al, 2008).

In addition to enhancing DA levels in downstream targets of VTA DA neurons, cocaine also increases DA within the VTA (Groves et al, 1975; Beart et al, 1979). The cocaine-induced increase in VTA DA levels activates autoreceptors (D2R) that, together with GABABR-dependent feedback (Waddington and Cross, 1978; Wolf et al, 1978), temper VTA DA neuron excitability (Einhorn et al, 1988). Pharmacological blockade or genetic suppression of G protein-dependent inhibitory feedback pathways in midbrain DA neurons alters behavioral effects of cocaine, including locomotor activation and self-administration (Steketee and Kalivas, 1991; Bello et al, 2011; Holroyd et al, 2015; de Jong et al, 2015). Moreover, inhibitory G protein signaling mediated by GABABR and D2R is decreased following cocaine administration (Ackerman and White, 1990; Kushner and Unterwald, 2001; Arora et al, 2011), highlighting the reciprocal relationship between cocaine and inhibitory G protein signaling in DA neurons.

The direct inhibitory influence of GABABR and D2R activation on VTA DA neurons is mediated primarily by activation of G protein-gated inwardly rectifying K+ (GIRK/Kir3) channels found in the somatodendritic compartment (Beckstead et al, 2004; Cruz et al, 2004). Although GIRK1/GIRK2 heterotetramers are considered to be the prototypical neuronal GIRK channel (Lujan et al, 2014), VTA DA neurons express a GIRK2/GIRK3 heteromer (Cruz et al, 2004). Girk2 ablation eliminates all GIRK channel activity in VTA DA neurons (Beckstead et al, 2004; Cruz et al, 2004).

Multiple lines of evidence suggest that GIRK channels modulate DA-dependent behaviors. Constitutive Girk2–/– mice are hyperactive, a phenotype normalized by D1 DA receptor (D1R) blockade (Blednov et al, 2002). Moreover, Girk2–/– mice exhibit enhanced locomotor activation in response to morphine and cocaine (Arora et al, 2010; Kotecki et al, 2015). The many phenotypes and adaptations associated with global Girk2 ablation, however, confound interpretation of these data (Lujan et al, 2014). For example, VTA DA neurons, NAc medium spiny neurons (MSNs), and layer 5/6 medial prefrontal cortex (mPFC) pyramidal neurons from Girk2–/– mice exhibit elevated AMPA receptor-mediated neurotransmission (Arora et al, 2010; Hearing et al, 2013). The recent availability of mice lacking GIRK2 in DA neurons permits a more precise evaluation of the role of GIRK-dependent signaling in modulating DA neuron excitability and cocaine-induced behaviors (Kotecki et al, 2015). Here we report that GIRK channel ablation in DA neurons reduces the autoreceptor-mediated inhibition of VTA DA neurons and alters behavioral sensitivity to cocaine.

MATERIALS AND METHODS

Animals

All studies were approved by the Institutional Animal Care and Use Committees at the University of Minnesota and University of Texas Health Science Center, San Antonio. The generation of Girk2flox/flox, DATCre(+/−):Girk2flox/flox, and Pitx3-eGFP(+)/DATCre(+/−):Girk2flox/flox mice was described previously (Kotecki et al, 2015). DATCre (B6.SJL-Slc6a3tm1.1(cre)Bkmn) and Drd1a-tdTomato (B6.Cg-Tg(Drd1a-tdTomato)6Calak/J) lines were purchased from The Jackson Laboratory (Bar Harbor, ME), and the Drd2-eGFP strain (Tg(Drd2-EGFP)S118Gsat) was obtained from the Mutant Mouse Regional Resource Center. Mice were maintained on a 12 h light/dark cycle (lights on at 0700 hours), with food and water available ad libitum.

Drugs

Quinpirole, sulpiride, tetrodotoxin (TTX), and picrotoxin were purchased from Sigma (St Louis, MO). Cocaine hydrochloride was purchased from Sigma or provided by the National Institute on Drug Abuse drug supply program (RTI International, Research Triangle Park, NC).

Slice Electrophysiology

Horizontal slices (225 μm) of the mouse VTA (5–7 weeks) were prepared as described (Kotecki et al, 2015). Neurons medial to the medial terminal nucleus of the accessory optic tract, and identified via GFP expression driven by the Pitx3 promoter, were targeted for analysis. DA neurons in the most medial aspect of the VTA were avoided as they were reported to exhibit low GIRK2 and D2R expression (Lammel et al, 2008). Whole-cell data were acquired using a Multiclamp 700 A amplifier and the pCLAMPv.9.2 software (Molecular Devices; Sunnyvale, CA). Ih amplitude was assessed using a 1-s voltage ramp (−60 to −120 mV). Somatodendritic currents (Vhold=−60 mV) were measured in the presence of TTX (0.5 μM). Spontaneous activity was measured in current-clamp mode (I=0) for 1 min. Neurons exhibiting no or irregular spontaneous activities were not evaluated. Rheobase and current/spike relationships were measured by injecting currents from −60 to 220 pA, increasing in 20 pA increments (1 s/step). Rheobase was defined as the minimum current evoking one or more action potentials. To assess the effect of D2/3R activation on excitability, neurons were voltage-clamped (Vhold=−60 mV) while quinpirole was applied to the bath. At the peak of the quinpirole response, spontaneous activity and excitability were reassessed. All command potentials factored in a junction potential of −15 mV. Series and membrane resistances were tracked throughout the experiment. If series resistance was high (>20 MΩ) or unstable (>20% variation), the experiment was excluded from analysis.

Locomotor Activity

Cocaine-induced activity was assessed in open field activity chambers (Med-Associates, St Albans, VT), as described (Pravetoni and Wickman, 2008). Subjects (7–10 weeks) were acclimated over 3 days, during which the animals were handled and exposed to i.p. injection (saline) and the open field. Distance traveled during the 60-min period following saline injection on the last acclimation day was taken as baseline activity. For acute cocaine-induced activity studies, subjects were given 1 of the 3 cocaine doses (3, 15, or 30 mg/kg i.p.). For cocaine sensitization, subjects received cocaine (15 mg/kg i.p.) for 5 days, followed by cocaine challenge (15 mg/kg i.p.) 10–11 days later.

Conditioned Place Preference

CPP testing was performed in two-compartment chambers (Med Associates) housed within sound-attenuating cubicles. One cohort was evaluated with 0 (saline) or 15 mg/kg cocaine using a three conditioning session design (Mirkovic et al, 2012). A second cohort was evaluated using a modified design involving lower cocaine doses (0.5 and 3 mg/kg) (Wydeven et al, 2014). Side preference in this study was evaluated twice, after the second and fourth cocaine conditioning sessions. CPP was calculated as the difference in time spent in drug- and saline-paired chamber during the posttest.

Self-Administration

Male mice (15–20 weeks) were group-housed (3–5/cage) on a 12/12 h reverse light–dark cycle (lights off at 0900 hours), with ad libitum access to food and water throughout the study. Following jugular catheterization, mice were housed individually and allowed ⩾7 days to recover. Operant sessions (2 h) were conducted as described (Sharpe et al, 2014), with minor modifications. During training, responses in the correct nose poke hole were rewarded on a fixed ratio 1 (FR1) schedule of reinforcement for infusions 1–5, an FR2 for infusions 6–8, and FR3 thereafter. Upon completing the response requirement, a green stimulus light in the correct nose poke hole was turned off, a 30-s time out was initiated, and cocaine was delivered (0.5 mg/kg/infusion over 2 s), accompanied by a sound stimulus of 2 kHz. After the timeout, the correct nose poke hole was re-illuminated. Although responding in both holes was recorded during the timeout, no responses were reinforced. Self-administration was considered acquired when infusion number was ⩾8 in two consecutive sessions and the number of nose pokes in the correct hole represented 70% of the total. Mice then advanced to 7 days of training on an FR3 schedule (0.5 mg/kg/infusion) to ensure stable responding prior to the dose-response assessment. For the dose–response study, cocaine dose per infusion was increased daily (0.03, 0.1, 0.3, 1.0, and 3.0 mg/kg/infusion). Self-administration across this dose range was evaluated over three rounds, with the first round considered training. The average of data for each subject over the last two rounds was used for analysis. Although all doses were calculated based on a typical weight of a young adult mouse (28 g), final intake values were corrected for actual bodyweight. Catheters were flushed before and after sessions to assess patency. Two GIRK2DAWT and four GIRK2DAKO subjects were excluded from the study owing to inconsistent self-administration or failed catheter patency.

Statistical Analysis

Data are presented throughout as the mean±SEM. Statistical analyses were performed using Prism (GraphPad Software, La Jolla, CA) and SigmaPlot (Systat Software, San Jose, CA). With the exception of mEPSC and self-administration studies, which included males only, all studies involved balanced groups of male and female mice. Data were analyzed first for effects of sex and genotype using two-way ANOVA when those were the only variables. For studies with additional variables (eg, multiple testing days), the interaction of sex and genotype was assessed using two-way ANOVAs at each level; the effect of sex within each genotype was determined using a repeated-measures two-way ANOVA. When an effect of sex or a sex interaction was observed, data from each sex were analyzed separately. When sex differences were not observed, male and female data were pooled. Pairwise comparisons were performed using Student's t or Mann–Whitney U tests, or Bonferroni post-hoc test, as appropriate. Differences were considered significant if P<0.05.

RESULTS

D2/3R-Dependent Signaling and Excitability in VTA DA Neurons from GIRK2DAKO Mice

We recently demonstrated that VTA DA neurons in DATCre(+):Girk2flox/flox mice (GIRK2DAKO mice) exhibited diminished inhibitory somatodendritic current responses to GABABR activation (Kotecki et al, 2015). To test whether autoreceptor-mediated signaling in VTA DA neurons from GIRK2DAKO mice was similarly blunted, we measured currents evoked by a saturating concentration (20 μM) of the D2/3R agonist quinpirole. GIRK2DAKO mice were crossed with mice expressing GFP under the control of a DA neuron-specific promoter (Pitx3) to permit targeted characterization of VTA DA neurons. VTA DA neurons from DATCre(−):Girk2flox/flox mice (GIRK2DAWT mice) exhibited outward currents that were reversed by the D2/3R antagonist sulpiride (5 μM; Figure 1a). Although DA neurons from GIRK2DAKO mice also showed quinpirole-induced responses, amplitudes were smaller than their wild-type counterparts (Figure 1a and b; Mann–Whitney U=85.5, ***P<0.001). There was no effect of genotype on other properties of VTA DA neurons (Supplementary Table S1).

Figure 1.

Figure 1

Open in a new tab

D2/3R-dependent somatodendritic currents in VTA DA neurons from GIRK2DAWT and GIRK2DAKO mice. (a) Representative currents evoked by quinpirole (20 μM) and reversed by sulpiride (5 μM) in VTA DA neurons from GIRK2DAWT and GIRK2DAKO mice. Currents were recorded with TTX (0.5 μM) in the bath to prevent synaptic activity. (b) Summary of quinpirole-induced currents in VTA DA neurons from GIRK2DAWT (n=23) and GIRK2DAKO (n=19) mice. Circles represent individual data points. ***P<0.001 vs GIRK2DAWT.

We next measured spontaneous activity and rheobase in the absence or presence of quinpirole (20 μM). At baseline, VTA DA neurons from GIRK2DAWT and GIRK2DAKO mice exhibited no difference in spontaneous activity (Figure 2a and b), rheobase (Figure 2c and d), current–spike relationship (Figure 2e), or other properties (Supplementary Table S2). In the presence of quinpirole, spontaneous activity of VTA DA neurons from GIRK2DAWT mice was completely eliminated (16/16 neurons; Figure 2a and b). In contrast, quinpirole eliminated spontaneous activity in only 6/12 neurons GIRK2DAKO VTA DA neurons, with the remainder showing an incomplete suppression of activity (Mann–Whitney U=48.0, **P<0.01). Quinpirole markedly decreased excitability (increased rheobase) in GIRK2DAWT VTA DA neurons but had less of an effect on GIRK2DAKO VTA DA neurons (Figure 2c and d; Mann–Whitney U=17.5, ***P<0.001). Finally, the current/spike relationship was more prominently suppressed by quinpirole in VTA DA neurons from GIRK2DAWT as compared with GIRK2DAKO mice (Figure 2e; F14,336=24.5, P<0.001, current × genotype interaction). These data indicate that, while loss of GIRK-dependent signaling does not impact baseline excitability of VTA DA neurons, it does dampen the inhibitory influence of autoreceptor-dependent signaling on these neurons.

Figure 2.

Figure 2

Open in a new tab

Excitability of GIRK2DAWT and GIRK2DAKO VTA DA neurons in the absence and presence of quinpirole. (a) Representative traces showing spontaneous activity in GIRK2DAWT and GIRK2DAKO VTA DA neurons at baseline and in the presence of quinpirole (20 μM). (b) Summary of spontaneous activity in GIRK2DAWT (n=16) and GIRK2DAKO (n=12) VTA DA neurons. (c) Typical responses of GIRK2DAWT and GIRK2DAKO VTA DA neurons to depolarizing current injection (80 pA), at baseline, and in the presence of quinpirole (20 μM). (d) Rheobase summary for GIRK2DAWT (n=16) and GIRK2DAKO (n=12) VTA DA neurons at baseline and in the presence of quinpirole (20 μM). (e) Plots showing the number of action potentials elicited by a 1-s current injection (−60 to 220 pA) in GIRK2DAWT (n=16) and GIRK2DAKO (n=12) VTA DA neurons at baseline and in the presence of quinpirole (20 μM). **P<0.01 and ***P<0.001 vs GIRK2DAWT (within current step).

Excitatory Neurotransmission in the NAc of GIRK2DAKO Mice

The amplitude and frequency of miniature excitatory postsynaptic currents (mEPSCs) were elevated in MSNs of the nucleus accumbens (NAc) shell from constitutive Girk2–/– mice, observations paralleling an increase in AMPA receptor levels at excitatory synapses in these neurons (Arora et al, 2010). To discern whether this adaptation is driven by loss of GIRK-dependent signaling in VTA DA neurons, we compared mEPSCs in NAc core and shell MSNs from GIRK2DAWT and GIRK2DAKO mice. To facilitate the targeted evaluation of neuron subpopulations in the NAc, we crossed GIRK2DAKO mice with transgenic mice expressing fluorescent proteins in D1R-expressing (Drd1a-tdTomato) or D2R-expressing (Drd2-GFP) MSNs. No genotype difference in mEPSC amplitude or frequency was observed in D1R- or D2R-expressing MSNs in the NAc core or shell (Supplementary Figure S1). Thus the increased excitatory neurotransmission observed in NAc MSNs in Girk2–/– mice is driven by loss of GIRK2 in a non-DA neuron population(s).

Cocaine-Induced Locomotor Activity in GIRK2DAKO Mice

We next evaluated GIRK2DAWT and GIRK2DAKO mice in an open field activity test (Figure 3). We observed a significant effect of sex (F1,99=7.7, P<0.01) and genotype (F1,99=15.0, P<0.001) on baseline activity (saline day 2, S2), with GIRK2DAKO mice exhibiting a small but significantly higher level of activity than GIRK2DAWT mice, and females showing higher activity than males. Significant main effects of genotype (male F1,42=30.8, P<0.001; female F1,49=8.2, P<0.01) and dose (male F2,42=74.0, P<0.001; female F2,49=43.8, P<0.001), as well as genotype × dose interactions (male F2,42=4.6, P<0.05; female F2,49=6.7, P<0.01), were observed in both sexes. Notably, mice of both genotypes and sexes increased locomotor activity in response to cocaine, with the 15 mg/kg dose revealing elevated responses in male (Figure 3a) and female (Figure 3b) GIRK2DAKO mice. Interestingly, there was no difference between female GIRK2DAWT and GIRK2DAKO at the 30 mg/kg dose. No difference in baseline activity or the locomotor-stimulatory effect of 15 mg/kg cocaine was observed between DATCre(+) and DATCre(−) littermates (Supplementary Figure S2), arguing that activity differences between GIRK2DAWT and GIRK2DAKO mice are attributable to the loss of GIRK channels from DA neurons.

Figure 3.

Figure 3

Open in a new tab

Cocaine-induced locomotor activity and sensitization in GIRK2DAWT and GIRK2DAKO mice. Distance traveled (in m) by GIRK2DAWT and GIRK2DAKO mice in an open field test. The test included a handling day (H), followed by 2 days of saline injections (S1 and S2), and then a final day with injection of 3, 15, or 30 mg/kg cocaine. Distance traveled on S2 was taken as baseline activity. (a) Baseline (n=20–28 per genotype) and acute cocaine-induced locomotor activity (n=6–12 per genotype and dose) in male GIRK2DAWT and GIRK2DAKO mice. **P<0.01 and ***P 0.001 vs GIRK2DAWT (within dose). (b) Baseline (n=25–30 per genotype) and acute cocaine-induced locomotor activity (n=7–11 per genotype and dose) in female GIRK2DAWT and GIRK2DAKO mice. Symbols: ***P<0.001 vs GIRK2DAWT (within dose). (c) Total distance traveled by male GIRK2DAWT (n=8) and GIRK2DAKO (n=8) mice during the locomotor sensitization test. The repeated dosing protocol included a handling day (H) followed by 2 days of saline injections (S1 and S2), and then 5 days of cocaine injections (C1–5; 15 mg/kg); mice were challenged again with 15 mg/kg cocaine test (T) 10–11 days after the fifth cocaine injection. Symbols: *P<0.05, **P<0.01, and ***P<0.001 vs GIRK2DAWT (within injection). (d) Distance traveled by female GIRK2DAWT (n=11) and GIRK2DAKO (n=9) mice during the locomotor sensitization test. *P<0.05 and ***P<0.001 vs GIRK2DAWT (within injection).

Repeated cocaine leads to locomotor sensitization, the enhanced response to subsequent cocaine exposures that persists after prolonged withdrawal (Robinson and Berridge, 2001). To test whether locomotor sensitization differed between GIRK2DAWT and GIRK2DAKO mice, we used a repeated dosing procedure involving 5 days of cocaine injections (15 mg/kg) and a cocaine challenge test. We observed a significant interaction of sex and injection number in GIRK2DAWT mice (F8,136=2.5; P<0.05), and thus male (Figure 3c) and female (Figure 3d) subjects were analyzed separately. Significant main effects of genotype (male F1,112=10.6, P<0.01; female F1,144=20.2, P<0.001) and injection number (male F8,112=110.2, P<0.001; female F8,144=89.0, P<0.001), as well as genotype × injection number interactions (male F8,112=3.8, P<0.001; female F8,144=8.3, P<0.001), were observed for male and female subjects. Male and female GIRK2DAKO mice exhibited enhanced locomotor activity relative to GIRK2DAWT counterparts in response to cocaine on all days (Figure 3c and d). Moreover, all groups exhibited significantly greater locomotor activity on the challenge test relative to activity measured after the initial cocaine exposure (not shown). Notably, while the activity of GIRK2DAWT mice increased with each cocaine injection, the activity of GIRK2DAKO mice plateaued between the first and third injections. These findings suggest that GIRK2DAKO mice exhibit a ‘presensitized' phenotype similar to that induced by genetic suppression of GIRK-dependent signaling in the mPFC (Hearing et al, 2013).

Cocaine-Induced CPP in GIRK2DAKO Mice

We next assessed cocaine reward in GIRK2DAKO mice using a CPP test. As male and female GIRK2DAKO mice exhibited increased locomotor-stimulatory effect of 15 mg/kg cocaine relative to GIRK2DAWT counterparts, we used this dose with three drug/side pairings. A separate group of mice was conditioned with saline (0 mg/kg). Surprisingly, cocaine CPP was indistinguishable in GIRK2DAWT and GIRK2DAKO mice (Figure 4a). Previous CPP studies have shown a shallow or non-existent dose–response relationship for cocaine in the 4–12 mg/kg dose range in mice and a significant effect of conditioning session number on CPP magnitude (Brabant et al, 2005). Thus we also evaluated CPP utilizing lower cocaine doses (0.5 or 3 mg/kg) and after both two and four conditioning sessions. Again, no significant difference in cocaine CPP was detected between genotypes (Figure 4b and c).

Figure 4.

Figure 4

Open in a new tab

Cocaine-induced CPP in GIRK2DAWT and GIRK2DAKO mice. (a) Cocaine-induced CPP in GIRK2DAWT (n=10–12 per dose) and GIRK2DAKO (n=12–14 per dose) mice, measured as the difference in time spent in the drug (CS+) and saline (CS−) paired sides on test day. (b) Cocaine-induced CPP in a separate cohort of GIRK2DAWT (n=9–11 per dose) and GIRK2DAKO (n=10 per dose) mice, measured after two drug conditioning sessions. (c) Cocaine-induced CPP in the same mice as in panel (b), measured after four drug conditioning sessions.

Cocaine Self-Administration in GIRK2DAKO Mice

Finally, we asked whether intravenous cocaine self-administration was impacted by the loss of GIRK channels in DA neurons. No difference between GIRK2DAWT and GIRK2DAKO mice was observed with respect to acquisition of self-administration (Figure 5a–c), defined as earning eight infusions during two consecutive sessions, with nose pokes at the correct hole accounting for >70% of the total (Supplementary Figure S3A and B). After meeting acquisition criteria, mice were transitioned to an FR3 schedule of reinforcement (0.5 mg/kg/infusion), during which GIRK2DAWT (14.3±1.2) and GIRK2DAKO (12.8±0.8; t138=1.1, P=0.29) earned comparable infusions per session.

Figure 5.

Figure 5

Open in a new tab

Cocaine self-administration in male GIRK2DAWT and GIRK2DAKO mice. (a) Depiction of the self-administration procedure, beginning with jugular catheterization surgery and a ⩾7-day recovery period. Acquisition of cocaine self-administration was achieved in daily 2-h sessions where the response requirement was raised from an FR1 to an FR3 (0.5 mg/kg/infusion). GIRK2DAWT (n=9) and GIRK2DAKO (n=11) meeting acquisition criteria were maintained at an FR3 schedule for 7 days (baseline) prior to assessing intake and infusions as a function of cocaine dose. (b) The number of infusions earned during the first 11 days of cocaine self-administration (training). (c) Number of training sessions needed to meet acquisition criteria (t18=0.65, P=0.54). (d) Number of infusions, with FR3 schedule of reinforcement, earned as a function of unit dose of cocaine (0.03–3.0 mg/kg/infusion). **P<0.01 and ***P<0.001 vs GIRK2DAWT (within dose). (e) Cocaine intake as a function of unit dose, with the two lowest doses expanded in the inset. *P<0.05 (main effect of genotype).

After completing baseline FR3 training, self-administration behavior as a function of cocaine dose was assessed using an FR3 schedule of reinforcement. Although patterns of responding were comparable within and across subjects at the various cocaine doses (Supplementary Figure S4), an interaction between genotype and dose with respect to infusions earned was observed (F4,72=3.0, P<0.05); GIRK2DAKO mice earned more infusions of cocaine than GIRK2DAWT controls, most evident during sessions involving the lowest three cocaine doses (Figure 5d). Importantly, a strong preference for responding in the correct nose poke hole was observed both GIRK2DAWT and GIRK2DAKO throughout the dose–response study (Supplementary Figure S3C and D). Furthermore, a main effect of genotype was detected for total cocaine intake (F1,72=6.1, P<0.05), with higher levels observed for GIRK2DAKO mice relative to littermate controls (Figure 5e).

DISCUSSION

We reported previously that GABABR-GIRK signaling was diminished in VTA DA neurons, but not in VTA GABA neurons, from GIRK2DAKO mice (Kotecki et al, 2015). Here we show that autoreceptor-mediated signaling is also diminished in GIRK2DAKO mice, while baseline measures of neuronal excitability are unchanged. Thus the GIRK2DAKO mouse is a selective model of decreased inhibitory feedback (both GABABR and autoreceptor mediated) to DA neurons. Behavioral analysis of these mice revealed the critical influence of inhibitory feedback to DA neurons in both non-contingent (locomotor activity) and response-contingent (self-administration) cocaine-related behaviors.

GIRK-Dependent Inhibitory Feedback and Cocaine-Induced Locomotor Activity

GIRK2DAKO mice, similar to constitutive Girk2–/– mice (Arora et al, 2010), are more sensitive to the locomotor-stimulatory effect of cocaine. This unconditioned behavioral response to cocaine is DA dependent and tempered by GABABR- and autoreceptor-mediated inhibitory feedback to VTA DA neurons. Indeed, cocaine-induced locomotor activity was enhanced by intra-VTA infusion of pertussis toxin (Steketee and Kalivas, 1991), which inhibits the G proteins that mediate GABABR and autoreceptor-dependent signaling. In addition, intra-VTA infusion of the GABABR agonist baclofen or the D2/3R-antagonist sulpiride blocked and potentiated, respectively, the locomotor-stimulatory effect of cocaine (Steketee and Kalivas, 1991; Chen and Reith, 1994). In terms of cocaine-induced locomotor activity, GIRK2DAKO mice behave comparably to mice lacking D2R in DA neurons (Bello et al, 2011). RNAi-dependent suppression of D2R in VTA in rats also yielded enhanced cocaine-induced locomotor activity (de Jong et al, 2015). As GIRK2-containing channels mediate most of the direct inhibitory effect of GABABR and autoreceptor activation on VTA DA neurons (Beckstead et al, 2004; Cruz et al, 2004), these behavioral insights highlight the key influence of the GIRK component of the inhibitory G protein-mediated feedback pathways on acute cocaine-induced locomotor stimulation. Our data further show that while GIRK-dependent signaling in DA neurons is not required for the development of locomotor sensitization, they do provide an opposing influence on this addiction-related phenomenon.

GIRK-Dependent Inhibitory Feedback and Cocaine Reinforcement

Although available evidence shows that inhibitory G protein-dependent feedback to VTA DA neurons modulates the reinforcing effect of cocaine, the nature of this influence differs across studies. For example, we show here that GIRK2DAKO mice display normal acquisition of cocaine-induced self-administration but an overall higher intake of cocaine and, specifically, enhanced responding when presented with low cocaine doses per infusion. Constitutive Girk2–/– mice were also able to acquire cocaine self-administration but displayed decreased responding at lower cocaine doses (Morgan et al, 2003). The divergent phenotypes of constitutive Girk2–/– and GIRK2DAKO mice are likely attributable to widespread deficits in GIRK-dependent signaling and alterations in both excitatory and inhibitory ionotropic neurotransmission in the former model (Lujan et al, 2014). Interestingly, GIRK2DAWT and GIRK2DAKO mice exhibited different responding for cocaine at a dose (0.3 mg/kg/infusion) comparable to that used during training, wherein no genotype difference was detected. This may reflect an impact of the order of cocaine dose presentation on self-administration behavior or a manifestation of differential adaptations triggered by cumulative cocaine exposure between the groups.

GIRK2DAKO mice also differ from other genetic models of autoreceptor ablation in terms of cocaine reinforcement. For example, the RNAi-mediated suppression of D2R in the rat VTA enhanced motivation to work for cocaine as assessed using progressive ratio scheduling, whereas not altering acquisition of cocaine self-administration or fixed ratio responding (de Jong et al, 2015). Moreover, mice lacking D2R in DA neurons acquired cocaine self-administration more quickly than control subjects, while other parameters, including intake, motivation, and sensitivity, were normal (Holroyd et al, 2015). Interestingly, these mice also exhibited enhanced cocaine CPP at low doses (Bello et al, 2011), whereas we detected no CPP phenotype in GIRK2DAKO mice. Apparent behavioral discrepancies across these genetic models could reflect procedural and/or species differences or point to the overlapping but distinct impact of the genetic lesions. With respect to the latter point, it is noteworthy that GIRK2DAKO mice exhibit deficits in both GABABR- and autoreceptor-mediated inhibitory feedback to DA neurons. GABABR activation in the VTA decreases cocaine self-administration in rats (Brebner et al, 2000; Backes and Hemby, 2008), suggesting that GABABR-GIRK signaling in VTA DA neurons may also influence sensitivity to cocaine in a response-contingent procedure. In addition, whereas GIRK channels contribute to the somatodendritic inhibitory impact of autoreceptor activation on VTA DA neurons, D2R are also expressed on axon terminals of DA neurons in terminal regions where they can modulate neurotransmitter release (Sesack et al, 1994). Finally, the residual quinpirole-induced current seen in VTA DA neurons from GIRK2DAKO mice shows that the inhibitory influence of autoreceptor activation on these neurons is mediated by multiple effectors.

Notably, our self-administration studies involving GIRK2DAKO mice did not attempt to dissociate motivation to gain access to cocaine from intake, and thus we cannot speak of the potential impact of GIRK ablation in DA neurons on motivation separate from intake. Furthermore, it is important to note that increased levels of responding may reflect a decreased sensitivity to the reinforcing effect of the drug. Given that GIRK2DAKO mice exhibit increased sensitivity to the locomotor-stimulatory effect of cocaine, however, our current working hypothesis is that GIRK2DAKO mice are more sensitive to cocaine.

The Unique GIRK Channel in VTA DA Neurons

The VTA DA neuron GIRK channel (GIRK2/GIRK3) is unique, as most neuronal GIRK channels contain GIRK1 (Cruz et al, 2004; Koyrakh et al, 2005; Labouebe et al, 2007). Interestingly, Girk2 ablation yields a complete loss of VTA DA neuron GIRK channel activity (Beckstead et al, 2004; Cruz et al, 2004), whereas Girk3 ablation yields increased sensitivity of the residual channel (GIRK2 homomer) to receptor activation (Labouebe et al, 2007; Lunn et al, 2007). The opposing contributions of GIRK2 and GIRK3 are also sensed at the behavioral level, where loss of GIRK2 in DA neurons correlates with increased locomotor effects of cocaine and morphine, while constitutive Girk3 ablation correlates with decreased locomotor-stimulatory effect of morphine (Arora et al, 2010; Kotecki et al, 2015). The GIRK channel in VTA DA neurons undergoes activity-dependent bidirectional modulation; burst firing increases GIRK-dependent signaling in VTA DA neurons, whereas tonic firing suppresses channel activity (Lalive et al, 2014). The latter phenomenon is reminiscent of the cocaine-induced suppression of GIRK-dependent signaling in VTA DA neurons, attributable to a subcellular redistribution of GIRK2-containing channels (Arora et al, 2011). The plasticity of GIRK channels provides a plausible explanation for how experience could impact the sensitivity of the mesocorticolimbic system to subsequent experiences, drug or otherwise.

CONCLUDING REMARKS

The unique subunit composition of the VTA DA neuron GIRK channel, and its importance in modulating sensitivity to cocaine and other drugs of abuse, suggests that direct modulation of this target could be useful for treating aspects of addiction. Indeed, GABABR activation reduced self-administration and addiction-related behaviors linked to cocaine and other drugs of abuse in rodents (Brebner et al, 2000; Xi and Stein, 2000; Ranaldi and Poeggel, 2002; Leite-Morris et al, 2004; Liang et al, 2006; Filip et al, 2007). As most GIRK channels contain GIRK1, a GIRK2/GIRK3-selective activator should selectively enhance GIRK-dependent signaling in VTA DA neurons. Although GIRK2/GIRK3 channel activators have not yet been reported, the development of potent modulators of GIRK1-containing GIRK channels suggests that other channel subtype-selective modulators will be forthcoming (Kaufmann et al, 2013; Wydeven et al, 2014).

FUNDING AND DISCLOSURE

This work was supported by National Institute of Health grants to NMM (DA007234 and DA041767), LK (DA007097), MJB (DA032701), and KW (DA034696 and MH061933). The authors declare no conflict of interest.

Acknowledgments

We thank Zhilian Xia for exceptional care of the mouse colony at the University of Minnesota and Dr Ramaswamy Sharma for genotyping mice bred in San Antonio.

Footnotes

Supplementary Information accompanies the paper on the Neuropsychopharmacology website (http://www.nature.com/npp)

Supplementary Material

Supplementary Information
Click here for additional data file. (3.1MB, pdf)

References

  1. Ackerman JM, White FJ (1990). A10 somatodendritic dopamine autoreceptor sensitivity following withdrawal from repeated cocaine treatment. Neurosci Lett 117: 181–187. [DOI] [PubMed] [Google Scholar]
  2. Arora D, Haluk DM, Kourrich S, Pravetoni M, Fernandez-Alacid L, Nicolau JC et al (2010). Altered neurotransmission in the mesolimbic reward system of Girk mice. J Neurochem 114: 1487–1497. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Arora D, Hearing M, Haluk DM, Mirkovic K, Fajardo-Serrano A, Wessendorf MW et al (2011). Acute cocaine exposure weakens GABA(B) receptor-dependent G-protein-gated inwardly rectifying K+ signaling in dopamine neurons of the ventral tegmental area. J Neurosci 31: 12251–12257. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Backes EN, Hemby SE (2008). Contribution of ventral tegmental GABA receptors to cocaine self-administration in rats. Neurochem Res 33: 459–467. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Beart PM, McDonald D, Gundlach AL (1979). Mesolimbic dopaminergic neurones and somatodendritic mechanisms. Neurosci Lett 15: 165–170. [DOI] [PubMed] [Google Scholar]
  6. Beckstead MJ, Grandy DK, Wickman K, Williams JT (2004). Vesicular dopamine release elicits an inhibitory postsynaptic current in midbrain dopamine neurons. Neuron 42: 939–946. [DOI] [PubMed] [Google Scholar]
  7. Bello EP, Mateo Y, Gelman DM, Noain D, Shin JH, Low MJ et al (2011). Cocaine supersensitivity and enhanced motivation for reward in mice lacking dopamine D2 autoreceptors. Nat Neurosci 14: 1033–1038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Blednov YA, Stoffel M, Cooper R, Wallace D, Mane N, Harris RA (2002). Hyperactivity and dopamine D1 receptor activation in mice lacking Girk2 channels. Psychopharmacology (Berl) 159: 370–378. [DOI] [PubMed] [Google Scholar]
  9. Brabant C, Quertemont E, Tirelli E (2005). Influence of the dose and the number of drug-context pairings on the magnitude and the long-lasting retention of cocaine-induced conditioned place preference in C57BL/6J mice. Psychopharmacology (Berl) 180: 33–40. [DOI] [PubMed] [Google Scholar]
  10. Brebner K, Phelan R, Roberts DC (2000). Intra-VTA baclofen attenuates cocaine self-administration on a progressive ratio schedule of reinforcement. Pharmacol Biochem Behav 66: 857–862. [DOI] [PubMed] [Google Scholar]
  11. Chen NH, Reith ME (1994). Autoregulation and monoamine interactions in the ventral tegmental area in the absence and presence of cocaine: a microdialysis study in freely moving rats. J Pharmacol Exp Ther 271: 1597–1610. [PubMed] [Google Scholar]
  12. Cruz HG, Ivanova T, Lunn M-L, Stoffel M, Slesinger PA, Lüscher C (2004). Bi-directional effects of GABAB receptor agonists on the mesolimbic dopamine system. Nat Neurosci 7: 153–159. [DOI] [PubMed] [Google Scholar]
  13. de Jong JW, Roelofs TJ, Mol FM, Hillen AE, Meijboom KE, Luijendijk MC et al (2015). Reducing ventral tegmental dopamine D2 receptor expression selectively boosts incentive motivation. Neuropsychopharmacology 40: 2085–2095. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Di Chiara G, Imperato A (1988). Drugs abused by humans preferentially increase synaptic dopamine concentrations in the mesolimbic system of freely moving rats. Proc Natl Acad Sci USA 85: 5274–5278. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Einhorn LC, Johansen PA, White FJ (1988). Electrophysiological effects of cocaine in the mesoaccumbens dopamine system: studies in the ventral tegmental area. J Neurosci 8: 100–112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Filip M, Frankowska M, Przegalinski E (2007). Effects of GABA(B) receptor antagonist, agonists and allosteric positive modulator on the cocaine-induced self-administration and drug discrimination. Eur J Pharmacol 574: 148–157. [DOI] [PubMed] [Google Scholar]
  17. Groves PM, Wilson CJ, Young SJ, Rebec GV (1975). Self-inhibition by dopaminergic neurons. Science 190: 522–528. [DOI] [PubMed] [Google Scholar]
  18. Hearing M, Kotecki L, Marron Fernandez de Velasco E, Fajardo-Serrano A, Chung HJ, Lujan R et al (2013). Repeated cocaine weakens GABA(B)-GIRK signaling in layer 5/6 pyramidal neurons in the prelimbic cortex. Neuron 80: 159–170. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Holroyd KB, Adrover MF, Fuino RL, Bock R, Kaplan AR, Gremel CM et al (2015). Loss of feedback inhibition via D2 autoreceptors enhances acquisition of cocaine taking and reactivity to drug-paired cues. Neuropsychopharmacology 40: 1495–1509. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kaufmann K, Romaine I, Days E, Pascual C, Malik A, Yang L et al (2013). ML297 (VU0456810), the first potent and selective activator of the GIRK potassium channel, displays antiepileptic properties in mice. ACS Chem Neurosci 4: 1278–1286. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kotecki L, Hearing M, McCall NM, Marron Fernandez de Velasco E, Pravetoni M, Arora D et al (2015). GIRK channels modulate opioid-induced motor activity in a cell type- and subunit-dependent manner. J Neurosci 35: 7131–7142. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Koyrakh L, Lujan R, Colon J, Karschin C, Kurachi Y, Karschin A et al (2005). Molecular and cellular diversity of neuronal G-protein-gated potassium channels. J Neurosci 25: 11468–11478. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Kushner SA, Unterwald EM (2001). Chronic cocaine administration decreases the functional coupling of GABA(B) receptors in the rat ventral tegmental area as measured by baclofen-stimulated 35S-GTPγS binding. Life Sci 69: 1093–1102. [DOI] [PubMed] [Google Scholar]
  24. Labouebe G, Lomazzi M, Cruz HG, Creton C, Lujan R, Li M et al (2007). RGS2 modulates coupling between GABAB receptors and GIRK channels in dopamine neurons of the ventral tegmental area. Nat Neurosci 10: 1559–1568. [DOI] [PubMed] [Google Scholar]
  25. Lalive AL, Munoz MB, Bellone C, Slesinger PA, Luscher C, Tan KR (2014). Firing modes of dopamine neurons drive bidirectional GIRK channel plasticity. J Neurosci 34: 5107–5114. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Lammel S, Hetzel A, Hackel O, Jones I, Liss B, Roeper J (2008). Unique properties of mesoprefrontal neurons within a dual mesocorticolimbic dopamine system. Neuron 57: 760–773. [DOI] [PubMed] [Google Scholar]
  27. Leite-Morris KA, Fukudome EY, Shoeb MH, Kaplan GB (2004). GABA(B) receptor activation in the ventral tegmental area inhibits the acquisition and expression of opiate-induced motor sensitization. J Pharmacol Exp Ther 308: 667–678. [DOI] [PubMed] [Google Scholar]
  28. Liang JH, Chen F, Krstew E, Cowen MS, Carroll FY, Crawford D et al (2006). The GABA(B) receptor allosteric modulator CGP7930, like baclofen, reduces operant self-administration of ethanol in alcohol-preferring rats. Neuropharmacology 50: 632–639. [DOI] [PubMed] [Google Scholar]
  29. Lujan R, Marron Fernandez de Velasco E, Aguado C, Wickman K (2014). New insights into the therapeutic potential of GIRK channels. Trends Neurosci 37: 20–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Lunn ML, Nassirpour R, Arrabit C, Tan J, McLeod I, Arias CM et al (2007). A unique sorting nexin regulates trafficking of potassium channels via a PDZ domain interaction. Nat Neurosci 10: 1249–1259. [DOI] [PubMed] [Google Scholar]
  31. Mirkovic K, Palmersheim J, Lesage F, Wickman K (2012). Behavioral characterization of mice lacking Trek channels. Front Behav Neurosci 6: 60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Morgan AD, Carroll ME, Loth AK, Stoffel M, Wickman K (2003). Decreased cocaine self-administration in Kir3 potassium channel subunit knockout mice. Neuropsychopharmacology 28: 932–938. [DOI] [PubMed] [Google Scholar]
  33. Nestler EJ (2005). Is there a common molecular pathway for addiction? Nat Neurosci 8: 1445–1449. [DOI] [PubMed] [Google Scholar]
  34. Pierce RC, Kalivas PW (1997). A circuitry model of the expression of behavioral sensitization to amphetamine-like psychostimulants. Brain Res Brain Res Rev 25: 192–216. [DOI] [PubMed] [Google Scholar]
  35. Pravetoni M, Wickman K (2008). Behavioral characterization of mice lacking GIRK/Kir3 channel subunits. Genes Brain Behav 7: 523–531. [DOI] [PubMed] [Google Scholar]
  36. Ranaldi R, Poeggel K (2002). Baclofen decreases methamphetamine self-administration in rats. Neuroreport 13: 1107–1110. [DOI] [PubMed] [Google Scholar]
  37. Robinson TE, Berridge KC (2001). Incentive-sensitization and addiction. Addiction 96: 103–114. [DOI] [PubMed] [Google Scholar]
  38. Sesack SR, Aoki C, Pickel VM (1994). Ultrastructural localization of D2 receptor-like immunoreactivity in midbrain dopamine neurons and their striatal targets. J Neurosci 14: 88–106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Sharpe AL, Varela E, Bettinger L, Beckstead MJ (2014). Methamphetamine self-administration in mice decreases GIRK channel-mediated currents in midbrain dopamine neurons. Int J Neuropsychopharmacol 18: pii: pyu073. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Steketee JD, Kalivas PW (1991). Sensitization to psychostimulants and stress after injection of pertussis toxin into the A10 dopamine region. J Pharmacol Exp Ther 259: 916–924. [PubMed] [Google Scholar]
  41. Waddington JL, Cross AJ (1978). Neurochemical changes following kainic acid lesions of the nucleus accumbens: implications for a GABAergic accumbal-ventral tegmental pathway. Life Sci 22: 1011–1014. [DOI] [PubMed] [Google Scholar]
  42. Wolf P, Olpe HR, Avrith D, Haas HL (1978). GABAergic inhibition of neurons in the ventral tegmental area. Experientia 34: 73–74. [DOI] [PubMed] [Google Scholar]
  43. Wydeven N, de Velasco EMF, Du Y, Benneyworth MA, Hearing MC, Fischer RA et al (2014). Mechanisms underlying the activation of G-protein-gated inwardly rectifying K+ (GIRK) channels by the novel anxiolytic drug, ML297. Proc Natl Acad Sci USA 111: 10755–10760. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Xi ZX, Stein EA (2000). Increased mesolimbic GABA concentration blocks heroin self-administration in the rat. J Pharmacol Exp Ther 294: 613–619. [PubMed] [Google Scholar]
  45. Zweifel LS, Argilli E, Bonci A, Palmiter RD (2008). Role of NMDA receptors in dopamine neurons for plasticity and addictive behaviors. Neuron 59: 486–496. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Information
Click here for additional data file. (3.1MB, pdf)

Articles from Neuropsychopharmacology are provided here courtesy of Nature Publishing Group

RESOURCES