Differential Impact of Inhibitory G-Protein Signaling Pathways in Ventral Tegmental Area Dopamine Neurons on Behavioral Sensitivity to Cocaine and Morphine

Abstract Drugs of abuse engage overlapping but distinct molecular and cellular mechanisms to enhance dopamine (DA) signaling in the mesocorticolimbic circuitry. DA neurons of the ventral tegmental area (VTA) are key substrates of drugs of abuse and have been implicated in addiction-related behaviors. Enhanced VTA DA neurotransmission evoked by drugs of abuse can engage inhibitory G-protein-dependent feedback pathways, mediated by GABAB receptors (GABABRs) and D2 DA receptors (D2Rs). Chemogenetic inhibition of VTA DA neurons potently suppressed baseline motor activity, as well as the motor-stimulatory effect of cocaine and morphine, confirming the critical influence of VTA DA neurons and inhibitory G-protein signaling in these neurons on this addiction-related behavior. To resolve the relative influence of GABABR-dependent and D2R-dependent signaling pathways in VTA DA neurons on behavioral sensitivity to drugs of abuse, we developed a neuron-specific viral CRISPR/Cas9 approach to ablate D2R and GABABR in VTA DA neurons. Ablation of GABABR or D2R did not impact baseline physiological properties or excitability of VTA DA neurons, but it did preclude the direct somatodendritic inhibitory influence of GABABR or D2R activation. D2R ablation potentiated the motor-stimulatory effect of cocaine in male and female mice, whereas GABABR ablation selectively potentiated cocaine-induced activity in male subjects only. Neither D2R nor GABABR ablation impacted morphine-induced motor activity. Collectively, our data show that cocaine and morphine differ in the extent to which they engage inhibitory G-protein-dependent feedback pathways in VTA DA neurons and highlight key sex differences that may impact susceptibility to various facets of addiction.

Pharmacological and genetic approaches have implicated GABA B R and D 2 R-dependent signaling in the VTA in drug-induced behaviors. For example, pharmacological activation of D 2 R in the VTA suppressed cocaine-induced activity (Koulchitsky et al., 2016) and cocaine-reinstated drug seeking behavior (Xue et al., 2011), while pharmacological inhibition of D 2 R in the VTA increased psychostimulant-induced locomotor activity (Chen and Reith, 1994;Tanabe et al., 2004). Similarly, intra-VTA infusion of the GABA B R agonist baclofen blocked the locomotor-stimulatory effect of psychostimulants and opioids Steketee and Kalivas, 1991;Chen and Reith, 1994;Leite-Morris et al., 2002 and attenuated selfadministration of cocaine, opioids, and other drugs of abuse (Xi and Stein, 1999;Brebner et al., 2000;Leite-Morris et al., 2004;Backes and Hemby, 2008). RNAimediated suppression of D 2 R in the rat VTA (cell non-selective) enhanced cocaine-related behavior (de Jong et al., 2015;Chen et al., 2018), enhanced choice impulsivity (Bernosky-Smith et al., 2018), and increased functional brain activity (Martin et al., 2020). Ablation of D 2 R in DA neurons throughout the mouse brain correlated with enhanced cocaine-induced activity (Bello et al., 2011), as well as acquisition of cocaine self-administration and reactivity to drug-paired cues (Holroyd et al., 2015). Interestingly, partial suppression of GABA B R in VTA DA neurons unmasked cocaine-induced activity normally absent in BALB/c mice, but did not impact morphine-induced activity (Edwards et al., 2017).
Collectively, available data suggests that D 2 R and GABA B R-dependent signaling pathways in VTA DA neurons may exert a differential influence on behavioral sensitivity to drugs of abuse. Published studies investigating D 2 R and GABA B R signaling in VTA DA neurons, however, have used approaches that provide either anatomic or cellular specificity, or do not completely suppress inhibitory signaling, or focus only on one signaling pathway or drug of abuse. Accordingly, the goal of this study was to compare the impact of D 2 R or GABA B R ablation in VTA DA neurons on behavioral sensitivity to cocaine and morphine. To this end, we developed a neuron-specific viral CRISPR/Cas9 approach to ablate D 2 R and GABA B R selectively in VTA DA neurons of adult mice. Our findings show that D 2 R and GABA B R-dependent signaling exert drug-specific and sex-specific influences on motor activity in mice.

Animals
All studies were approved by the Institutional Animal Care and Use Committee at the University of Minnesota. The B6.SJL-Slc6a3 tm1.1(cre)Bkmn /J (stock #006660, The Jackson Laboratory) knock-in line was used in this study; heterozygous subjects, referred to throughout as DATCre (1) mice, were generated by crossing with C57BL/6J subjects. DATCre(1) mice were also crossed in multiple rounds with a Cre-dependent Cas9GFP knock-in line (B6;129-Gt(ROSA)26Sor tm1(CAG-cas9 * ,-EGFP)Fezh /J, The Jackson Laboratory, stock #026179), to generate DATCre(1) subjects homozygous for the Cas9GFP(1) mutation; these mice are referred to throughout as DATCre(1):Cas9GFP(1) mice. All mice used in experiments were bred in-house. Mice were group housed, maintained on a 14/10 h light/dark cycle and were provided ad libitum access to food and water.

Intracranial manipulations
Mice (.45 d) were placed in a stereotaxic frame (David Kopf Instruments) under isoflurane anesthesia. Microinjectors, made by affixing a 33-gauge stainless steel hypodermic tube within a shorter 26-gauge stainless steel hypodermic tube, were attached to polyethylene-20 tubing affixed to 10-ml Hamilton syringes, and were lowered through burr holes in the skull to the VTA (from bregma: À2.75 mm A/P, 60.55-0.7 mm M/L, À5 mm D/V); 300-500 nl of virus was injected per side at a rate of 100 nl/min. The optimized coordinates and viral load ensured full coverage of the VTA along anterior/ posterior and rostral-caudal axes, with minimal spread into the adjacent substantia nigra pars compacta. Microinjectors were left in place for 10 min following infusion to reduce solution backflow along the infusion track. Slice electrophysiology and behavioral experiments were performed three to four and five to six weeks after viral infusion for chemogenetic and CRISPR experiments, respectively.

Slice electrophysiology
Electrophysiological properties of VTA DA neurons were evaluated in behaviorally naive adult mice (66-93 d). Horizontal slices (225 mm) containing the VTA were prepared in ice-cold sucrose substituted ACSF, and allowed to recover at room temperature in ASCF containing the following: 125 mM NaCl, 2.5 mM KCl, 1.25 mM NaH 2 PO 4 , 25 mM NaHCO 3 , 11 mM glucose, 1 mM MgCl 2 , and 2 mM CaCl 2 , pH 7.4, for at least 1 h, as described (McCall et al., 2017). Neurons in the lateral VTA exhibiting appropriate fluorescence were targeted for analysis as this sub-region of the VTA receives prominent input from the NAc that mediates GABA B R-dependent feedback (Edwards et al., 2017). Whole-cell data were acquired using a Multiclamp 700A amplifier and pCLAMPv.9.2 software (Molecular Devices, LLC), using recording conditions described in previous publications (McCall et al., 2017). Input/membrane resistance (R M ) and apparent capacitance (C M ) were determined using a 5 mV/10 ms voltage step, with current responses filtered at 5 Hz. Immediately after establishing whole-cell access, I h conductance was measured using a 200-ms voltage step to À120 mV; the difference in current from beginning to end of the À120-mV step was taken as I h amplitude. Subsequently, spontaneous activity was measured in current-clamp mode (I = 0) for 1 min. Neurons exhibiting no spontaneous activity were not evaluated. Action potential half-width (AP 50 ) was determined by averaging five AP 50 values. For rheobase assessments, cells were held in current-clamp mode at À80 pA to prevent spontaneous activity, and then given 1-s current pulses, beginning at À100 pA and progressing in 20-pA increments. Rheobase was defined as the minimum current step that evoked one or more action potentials. In chemogenetic experiments, the change in rheobase measured before and after bath application of CNO application was determined. Somatodendritic holding currents were measured in voltage-clamp mode (V hold = À60 mV) following bath application of CNO (10 mM), baclofen (200 mM), or quinpirole (10 mM). All command potentials factored in a junction potential of À15 mV predicted using JPCalc software (Molecular Devices, LLC). Series and input resistances were tracked throughout the experiment. If series resistance was unstable or exceeded 20 MV, the experiment was excluded from analysis.

Locomotor activity
Locomotor activity was assessed in open-field activity chambers housed in sound-attenuating cubicles (Med-Associates). Each cubicle was equipped with three 16beam infrared arrays permitting automated measurements of distance traveled (Activity Monitor 5; Med-Associates). Animals were habituated to the testing room for at least 30 min before testing. Subjects were acclimated over 3 d; on day 1, animals were handled and placed in the open field for 60 min. On days 2 and 3, animals were given an intraperitoneal injection of saline and placed in the open field for 60 min. For DREADD experiments, mice were injected with CNO (2 mg/kg, i.p.) 30 min before either saline, cocaine (15 mg/kg, i.p.), or morphine (10 mg/kg, i.p.) injection on day 4. Distance traveled during the 60-min period following saline or drug injection was measured; separate cohorts of mice received saline, cocaine and morphine injections. For CRISPR/Cas9 experiments, mice were placed in the open field for 30 min before injection each day to acclimate to the chamber. Activity was measured on day 3 following saline injection, and again on day 4 following injection of cocaine (15 mg/ kg, i.p.) or morphine (10 mg/kg, i.p.), separate cohorts of mice were used for cocaine and morphine studies.
Thigmotaxis was quantified by dividing the distance traveled in the periphery by the total distance traveled, as described (Pravetoni and Wickman, 2008). After behavioral testing, the scope and accuracy of viral targeting was assessed by fluorescence microscopy; 225-mm horizontal slices of the midbrain were obtained using a vibratome and images were acquired on an Olympus IX-80 microscope using MetaMorph Advanced Acquisition v.7.7.7.0 software (Molecular Devices, LLC). Only data from animals with bilateral viral-driven fluorescence, where the majority of fluorescence was confined to VTA (with minimal spread to the adjacent substantia nigra), were included in the final analysis.

Statistical analysis
Data are presented throughout as the mean 6 SEM. Statistical analyses were performed using Prism v.9 software (GraphPad Software). All studies included balanced numbers of male and female mice, and data were analyzed first for sex effects. If no sex differences were observed, data from male and female subjects were pooled. If any data point fell outside the range of 2 SDs from the mean, it was excluded as an outlier. Across the entire study, this outlier detection approach led to the removal of one point from the hM4Di/morphine activity dataset and 1 point from the hM4Di control/morphine activity dataset. Differences were considered significant if p , 0.05.

G-protein-dependent inhibition of VTA DA neurons suppresses motor activity
Motor activation is an unconditioned DA-dependent response in mice to systemic administration of cocaine and morphine (van Rossum et al., 1962;Delfs et al., 1990;Kalivas and Duffy, 1993) that can be recapitulated by direct chemogenetic or optogenetic stimulation of VTA DA neurons (Kim et al., 2012;Tye et al., 2013;Boender et al., 2014;Guo et al., 2014). To test whether inhibitory G-protein-dependent signaling in VTA DA neurons can suppress the locomotor stimulatory effect of systemic cocaine and morphine, we used DATCre(1) mice and a Cre-dependent viral inhibitory chemogenetic approach. Cre-dependent AAV vectors harboring either hM4Di (mCherry) or mCherry control were infused into the VTA of adult male and female DATCre(1) mice to permit selective chemogenetic inhibition of VTA DA neurons (Fig. 1A).
Consistent with a previous report (Runegaard et al., 2018), chemogenetic inhibition of VTA DA neurons suppressed motor activity evoked by injection of saline, as well as cocaine and morphine (Fig. 1F). These results confirm that VTA DA neurons are key regulators of motor activity in mice and show that activation of inhibitory Gprotein signaling in these neurons can potently suppress baseline activity and the motor-stimulatory effects of cocaine and morphine.
CRISPR/Cas9 ablation of D 2 R and GABA B R in VTA DA neurons To assess the impact of D 2 R and GABA B R-dependent signaling pathways in VTA DA neurons on behavioral sensitivity to cocaine and morphine, we developed a DA neuron-specific, viral CRISPR/Cas9 ablation approach. DATCre(1) mice were crossed with a Cre-dependent Cas9GFP line to generate DATCre(1):Cas9GFP(1) mice. AAV vectors harboring gRNAs targeting LacZ (control), D 2 R, or GABA B R1 were generated and infused into the VTA of male and female DATCre(1):Cas9GFP(1) mice ( Fig. 2A).
To assess the efficacy and selectivity of the viral vectors, somatodendritic currents evoked by bath application of the D 2/3 R agonist quinpirole and the GABA B R agonist baclofen were measured in VTA DA neurons five to six weeks after viral infusion. In slices from LacZ gRNAtreated subjects, somatodendritic outward inhibitory currents were reliably evoked by quinpirole and baclofen (Fig. 2B), consistent with previous studies (Beckstead et al., 2004;Cruz et al., 2004;Labouèbe et al., 2007;Arora et al., 2010;McCall et al., 2017). In VTA DA neurons from mice treated with D 2 R gRNA, quinpirole-induced currents were completely absent (Fig. 2B,C), while baclofen (applied after quinpirole) evoked normal responses (Fig. 2B,  D). Interestingly, a sex difference in the amplitude of quinpirole-induced responses was observed, with currents in VTA DA neurons from female subjects larger than those in males (Fig. 2C). No sex differences were observed in the amplitude of baclofen-induced currents.
In VTA DA neurons from mice treated with GABA B R1 gRNA, somatodendritic responses to baclofen were completely absent (Fig. 2E,F). Responses to quinpirole (applied after baclofen application) were larger than those seen in VTA DA neurons from LacZ-treated controls (Fig.  2G). As this difference may reflect the impact of prior activation by GABA B R of a shared effector in control cells, we conducted a separate study where only quinpirole was applied to the slice. In these experiments, quinpirole-induced somatodendritic currents were not significantly different between GABA B R-lacking and control VTA DA neurons (Fig. 2H).
We also examined the impact of D 2 R and GABA B R ablation on VTA DA neuron excitability, as measured by spontaneous activity and rheobase. No sex differences were observed with respect to either excitability measure and, as such, data from male and female subjects were pooled to increase the power of the study. Ablation of D 2 R or GABA B R did not impact spontaneous activity of VTA DA neurons (Fig. 2I) or rheobase (Fig. 2J). Similarly, no impact of D 2 R or GABA B R ablation was detected on I h current amplitude, cell capacitance, action potential halfwidth, or input resistance (Table 1). Collectively, these data show that we can selectively ablate GABA B R-dependent or D 2 R-dependent signaling in VTA DA neurons, and that loss of these signaling pathways does not impact baseline electrophysiological characteristics of VTA DA neurons Impact of D 2 R and GABA B R in VTA DA neurons on baseline and cocaine-induced activity We next examined the impact of D 2 R or GABA B R ablation on open-field motor activity measured after injection of saline or cocaine. Given the observed sex difference in the strength of D 2 R-dependent signaling in VTA DA neurons, and our prior report of a sex difference in cocaine-induced motor activity in mice (McCall et al., 2017), we analyzed data from male and female subjects separately.
In support of this approach, we found that while male and female LacZ control mice showed no difference in total distance traveled after cocaine injection, the temporal profile of cocaine-induced activity was notably different for male and female subjects (Extended Data Fig. 3-1A), female subjects showed sharper time-to-peak and decay phases relative to male subjects.
In males as in females, we observed main effects of drug and viral treatment, and a significant interaction between drug and viral treatment on total distance traveled. Loss of D 2 R or GABA B R did not impact saline-induced activity (Fig. 3D, left). In contrast to our observations in females, however, ablation of D 2 R or GABA B R in males yielded comparably enhanced cocaine-induced activity over the 60-min postinjection interval (Fig. 3D, right). Interestingly, loss of D 2 R in males correlated with higher levels of activity seen shortly after cocaine injection (Fig.  3E), yielding a temporal profile that was qualitatively similar to that observed in females (Extended Data Fig. 3-1B). As was the case with female subjects, thigmotaxis was significantly impacted by drug but not viral treatment (Fig.  3C). Thus, D 2 R-dependent signaling in VTA DA neurons tempers behavioral sensitivity to cocaine in male and female mice, whereas GABA B R-dependent signaling exerts an influence on behavioral sensitivity to cocaine in male mice only.

Impact of D 2 R and GABA B R in VTA DA neurons on baseline and morphine-induced activity
As we previously reported that the loss of GIRK channel activity in DA neurons in mice correlated with enhance motor stimulation in response to systemic administration of 10 mg/kg, morphine (Kotecki et al., 2015), we next compared the relative contribution of D 2 R and GABA B R to this morphine-induced behavior. In female subjects, we observed a main effect of drug on motor activity, but no main effect of viral treatment or interaction between drug and viral treatment. Notably, no significant impact of D 2 R or GABA B R ablation on morphine-induced activity was observed in females (Fig. 4A, right). Although D 2 R ablation correlated with elevated activity levels at all time points following morphine injection, activity levels were not significantly different from controls (Fig. 4B). Consistent with prior reports (Mickley et al., 1990), we observed a marked increase in thigmotaxis in all subjects after morphine injection, but there was no difference between viral treatment groups (Fig. 4C).
In males, we also observed a main effect of drug, but no main effect of viral treatment or interaction between drug and viral treatment. As was the case in females, loss of either D 2 R or GABA B R in males had no significant impact on morphine-induced activity during the 60-min interval (Fig. 4D, right) or on the temporal activity profile following morphine injection (Fig. 4E). We also observed a similar increase in thigmotaxis after injection of morphine as seen in females, with no difference between viral treatment groups (Fig. 4F). Thus, neither D 2 R nor GABA B R ablation in VTA DA neurons exerts a significant impact on morphine-induced motor activity in mice.

Discussion
Here, we used a neuron-specific viral CRISPR/Cas9 approach to compare the impact of D 2 R or GABA B R ablation in VTA DA neurons on baseline activity and behavioral sensitivity to cocaine and morphine. Loss of GABA B R or  D 2 R in VTA DA neurons had no significant impact on baseline activity, paralleling the lack of effect of these manipulations on excitability and other electrophysiological properties of VTA DA neurons. These behavioral findings are consistent with studies involving genetic suppression or ablation of D 2 R or GABA B R in the rodent VTA (de Jong et al., 2015;Edwards et al., 2017). Thus, D 2 R-dependent and GABA B R-dependent signaling pathways in VTA DA neurons exert minimal influence on baseline DA neuron excitability. Notably, mice lacking D 2 R in DA neurons displayed hyperlocomotion (Bello et al., 2011), indicating that D 2 R-dependent signaling in DA neuron populations outside of the VTA may regulate baseline motor activity. D 2 R ablation in VTA DA neurons potentiated cocaine-induced activity in male and female mice, consistent with studies involving DA neuron-specific ablation of D 2 R in mice (Bello et al., 2011), and an RNAi-based approach targeting D 2 R in the rat VTA (de Jong et al., 2015). GABA B R ablation also potentiated cocaine-induced activity, but the influence of this signaling pathway was restricted to males. Enhanced D 2 R-dependent signaling in VTA DA neurons from female mice may compensate for the loss of GABA B R signaling, explaining the weak influence of GABA B R on cocaine-induced activity in females. Alternatively, there may be reduced GABAergic feedback to VTA DA neurons in females, rendering the loss of GABA B R less effective (Zachry et al., 2021).
in their influence on VTA DA neurons. While cocaine hyperpolarizes VTA DA neurons in a D 2 R-dependent manner (Beckstead et al., 2004), opioids increase VTA DA neuron firing by suppressing GABAergic input from local GABA neurons and/or RMTg GABA neurons (Johnson and North, 1992;Jhou et al., 2009;Jalabert et al., 2011), as well as disinhibiting glutamatergic input to VTA DA neurons (Yang et al., 2020). Thus, any inhibitory influence of somatodendritic D 2 R activation triggered by the opioid-induced rise in VTA DA is likely offset by the excitatory influence of disinhibition.
Since morphine increases NAc DA levels (Spielewoy et al., 2000;Chefer et al., 2003;Vander Weele et al., 2014), and DA neurotransmission in the NAc drives GABA B R-dependent feedback to VTA DA neurons (Edwards et al., 2017), it is also surprising that GABA B R ablation does not impact behavioral sensitivity to morphine. This outcome might reflect differences in the amplitude of the DA increase evoked by cocaine and morphine. Indeed, NAc DA levels in freely moving rats increased more in response to intravenous cocaine than morphine (Pontieri et al., 1995). The spatiotemporal pattern of DA increases in the NAc may also differ for cocaine and morphine. Consistent with this premise, cocaine elicited a more pronounced increase in DA in the NAc shell as compared with core (Aragona et al., 2008(Aragona et al., , 2009, whereas morphine evoked similar DA increases in NAc core and shell (Vander Weele et al., 2014). Moreover, the morphine-induced increase in NAc DA was relatively transient, a phenomenon potentially linked to a simultaneous increase in NAc GABA levels evoked by morphine (Vander Weele et al., 2014).
The direct inhibitory influence of GABA B R and D 2 R activation on VTA DA neurons is mediated primarily by activation of GIRK channels (Brodie and Dunwiddie, 1990;Ackerman et al., 1993;Mercuri et al., 1997;Chen and Pan, 2000;Bunney et al., 2001;Beckstead et al., 2004;Cruz et al., 2004;Ford, 2014), though other effectors are modulated as well (Philippart and Khaliq, 2018;Su et al., 2019). Genetic ablation of Girk2 globally or selectively in DA neurons correlated with increased motor-stimulatory effect of cocaine (Pravetoni and Wickman, 2008;Kotecki et al., 2015;McCall et al., 2017McCall et al., , 2019 and morphine (Kotecki et al., 2015). The contribution of GIRK channels to cocaine-induced activity was further localized to VTA DA neurons (McCall et al., 2019). Given that GIRK channels mediate the D 2 R-dependent and GABA B R-dependent inhibition of VTA DA neurons and that DA neuronspecific loss of GIRK channels enhances the motor-stimulatory effect of cocaine and morphine, the lack of impact of D 2 R or GABA B R ablation on morphine-induced activity was unexpected. It is possible that GABA B R-dependent and D 2 R-dependent signaling pathways are functionally redundant and that both need to be eliminated to see an influence on morphine-induced activity. GIRK channels in non-VTA DA neurons, perhaps in the adjacent substantia nigra pars compacta (Koyrakh et al., 2005), may also explain the impact of GIRK channel ablation on morphine.
Females are more susceptible to various facets of addiction (Fattore et al., 2008;Anker and Carroll, 2011;Becker and Koob, 2016), fueling interest in identifying relevant sex differences at molecular and cellular levels. Here, we found multiple sex differences related to inhibitory G-protein signaling in VTA DA neurons. Loss of GABA B R, for example, had minimal impact on behavioral sensitivity to cocaine in females, but significantly enhanced the motor-stimulatory effect of cocaine in males. D 2 R-dependent somatodendritic inhibitory currents were also larger in VTA DA neurons from females. This difference could reflect elevated D 2 R expression and/or function in VTA DA neurons from females and is predicted to decrease cocaine sensitivity. Indeed, females exhibit elevated D 2 R-dependent signaling at baseline as compared with males (Walker et al., 2006;Zachry et al., 2021).
Cocaine-induced activity was characterized by a more rapid and sharp peak in females as compared with males, which could be explained by a tighter regulation of synaptic DA by DAT in females compared with males (Zachry et al., 2021). D 2 R ablation in VTA DA neurons in male mice yielded a temporal profile comparable to that seen in female mice. While this suggests that D 2 R-dependent signaling in VTA DA neurons is critical in tempering the early behavioral response to cocaine, the shift in the temporal profile observed in males is somewhat counterintuitive given that somatodendritic D 2 R-dependent signaling is weaker in VTA DA neurons from males. This apparent discrepancy is perhaps explained by sex differences in the influence of presynaptic/terminal D 2 R (which should also be eliminated by the CRISPR/Cas9 ablation approach) on DA dynamics in the NAc. Presynaptic D 2 R-dependent signaling in VTA DA neurons in males may be stronger than that seen in females. Alternatively, while D 2 R-dependent somatodendritic response amplitudes are larger in VTA DA neurons from female mice, VTA DA neurons in males may be more sensitive to D 2 R-dependent inhibition.
While D 2 R-dependent and GABA B R-dependent signaling pathways in the VTA can suppress behavioral sensitivity to cocaine, psychostimulant exposure can weaken these inhibitory feedback pathways. For example, a transient decrease in D 2 R influence on VTA DA neuron firing has been reported after repeated cocaine treatment or self-administration (Henry et al., 1989;Ackerman and White, 1990). This adaptation, and corresponding potentiation of cocaine-induced motor activity, was reproduced by repeated quinpirole treatment (Henry et al., 1998). Cocaine self-administration in rats also increased firing rate and burst activity of midbrain DA neurons, paralleled by decreased ability of quinpirole to inhibit DA neuron firing rate (Marinelli et al., 2003). Self-administration of amphetamine in rats reduced the ability of D 2 R to suppress evoked DA release in the NAc, an effect mediated in part by an RGS2-dependent disruption of D 2/3 R/Ga i2 functional coupling (Calipari et al., 2014;Sun et al., 2015). Repeated cocaine in male rats also decreased GABA B R/ G-protein coupling (Kushner and Unterwald, 2001). Finally, cocaine suppressed GABA B R-dependent somatodendritic signaling in putative VTA DA neurons (Arora et al., 2011), and methamphetamine self-administration suppressed D 2 R-dependent and GABA B R-dependent somatodendritic responses in VTA DA neurons (Sharpe et al., 2014).
Notably, D 2 R activation appears critical for many of these forms of plasticity, which likely contribute to the hyperexcitability of meso-accumbens DA pathway following psychostimulant exposure (Henry et al., 1998;Francis et al., 2019). The reciprocal relationship between psychostimulants and inhibitory G-protein-dependent feedback pathways in VTA DA neurons may be dependent on age and/or species, or methodological variables. Indeed, repeated methamphetamine injections correlated with enhanced D 2 R-dependent hyperpolarization in young (8-10 d) rats (Amano et al., 2003). In mice, repeated methamphetamine injections suppressed somatodendritic GABA B R-dependent (but not D 2 Rdependent) signaling in VTA DA neurons, but only when methamphetamine was given in novel environment (Munoz et al., 2016).
Our work highlights innate signaling mechanisms that regulate behavioral sensitivity to cocaine. Knowledge of the molecular players and neuron populations regulating behavioral sensitivity to cocaine and other drugs of abuse may help in identifying individuals at risk for developing addiction. Further investigation of mechanisms regulating the strength and sensitivity of inhibitory G-protein signaling pathways in VTA DA neurons may suggest opportunities for selective therapeutic interventions tailored to specific drugs of abuse.