Dorsolateral Striatum Research Articles (Page 1)

Sex differences in responsiveness to social stress in adulthood are highly conserved across species, with females more sensitive to isolation. Here, we show that Arginine vasopressin receptor 1a (AVPR1A) in the central nucleus of the amygdala (CeA) mediates the enhanced susceptibility of females to post-pubertal chronic social isolation stress (CSIS) in mice. Chemogenetic activation of AVPR1ACeA circuits induces anxiety-related behaviors in both sexes. However, genetic, pharmacological, chemogenetic and optogenetic loss of function approaches support the idea that it is only endogenously engaged in females in the context of CSIS. Using a combination of virus-based tools, we identified a major source of AVP ligand in the posterodorsal part of the medial amygdala (MePD) as well as an important downstream target of AVPR1ACeA neurons, the dorsolateral striatum (DLS). Loss of function approaches identified three nodes in the circuit that provide sex-specificity in the effects of CSIS on anxiety-related behaviors: 1) ERα signaling in AVPMePD neurons; 2) engagement of the AVPR1A pathway in the CeA; and 3) number of AVPR1ACeA projections to the DLS. These data support new therapeutic applications for AVPR1A antagonists in women experiencing social isolation or loneliness.

Dopamine (DA) midbrain neurons are involved in a wide array of key brain functions including movement control and reward-based learning. They are also critical for major brain disorders such as Parkinson’s disease or schizophrenia. DA neurons projecting to distinct striatal territories are diverse with regard to their molecular makeup and cellular physiology, which are likely to contribute to the observed differences in temporal DA dynamics. Among these regions, the dorsolateral striatum (DLS) displays the fastest DA dynamics, which might control the moment-to-moment vigor and variability of voluntary movements. However, the underlying mechanisms for these DLS-specific fast DA fluctuations are unresolved. Here, we show that DLS-projecting DA neurons in the substantia nigra (SN) possess a unique biophysical profile allowing immediate 10-fold accelerations in discharge frequency via rebound bursting. By using a combination of in vitro patch-clamp recordings in projection-defined DA SN subpopulations from adult male mice and developing matching projection-specific computational models, we demonstrate that a strong interaction of Cav3 and SK channels specific for DLS-projecting Aldh1a1–positive DA SN (DLS-DA) neurons controls the gain of fast rebound bursting, while Kv4 and HCN channels mediate timing of rebound excitability. In addition, GIRK channels activated by D2 and GABAB receptors prevent rebound bursting in these DLS-DA neurons. Furthermore, our in vivo patch-clamp recordings and matching in vivo computational models provide evidence that these unique rebound properties might be preserved in the intact brain, where they might endow specific computational properties well suited for the generation of fast DA dynamics present in DLS.

The density and overlap of cortical axons in the dorsolateral striatum (DLS) have suggested that striatal neurons integrate widespread information from cortical regions that are functionally related. However, in vivo, DLS neuronal responses to sensory stimuli have shown unexpectedly high selectivity, raising questions about the actual degree of convergence of functional corticostriatal projections on individual striatal cells. Here, we investigated this question by focusing on the projections from different whisker cortical columns in mice, as they overlap in the striatum and are co-active during behavior. Using ex vivo patch-clamp recordings in the DLS and glutamate uncaging for focal stimulations in the barrel cortex, we were able to map the location of presynaptic neurons to individual striatal projection neurons (SPNs). We found that each SPN was innervated by cells located in a small number of whisker cortical columns scattered across the barrel field in the slice. Connectivity of single SPNs with cortical neurons was thus highly discontinuous horizontally, despite the presence of more potential connections. Moreover, connectivity patterns were specific to each cell, with neighboring SPNs sharing few common clusters of presynaptic cells in the cortex. Despite this sparse and distinct innervation of individual SPNs, the projection was topographically organized at the population level. Finally, we found similar innervation patterns for D1- and D2-type SPNs, but observed differences in synaptic strength in their connections with certain cortical layers, notably the associative layer 2/3. Our results suggest that the high convergence of somatosensory inputs to the striatum, enabled by diffuse and overlapping cortical innervation, is accomplished through sparse yet complementary connectivity to individual SPNs.

The density and overlap of cortical axons in the dorsolateral striatum (DLS) have suggested that striatal neurons integrate widespread information from cortical regions that are functionally related. However, in vivo, DLS neuronal responses to sensory stimuli have shown unexpectedly high selectivity, raising questions about the actual degree of convergence of functional corticostriatal projections on individual striatal cells. Here, we investigated this question by focusing on the projections from different whisker cortical columns in mice, as they overlap in the striatum and are co-active during behavior. Using ex vivo patch-clamp recordings in the DLS and glutamate uncaging for focal stimulations in the barrel cortex, we were able to map the location of presynaptic neurons to individual striatal projection neurons (SPNs). We found that each SPN was innervated by cells located in a small number of whisker cortical columns scattered across the barrel field in the slice. Connectivity of single SPNs with cortical neurons was thus highly discontinuous horizontally, despite the presence of more potential connections. Moreover, connectivity patterns were specific to each cell, with neighboring SPNs sharing few common clusters of presynaptic cells in the cortex. Despite this sparse and distinct innervation of individual SPNs, the projection was topographically organized at the population level. Finally, we found similar innervation patterns for D1- and D2-type SPNs, but observed differences in synaptic strength in their connections with certain cortical layers, notably the associative layer 2/3. Our results suggest that the high convergence of somatosensory inputs to the striatum, enabled by diffuse and overlapping cortical innervation, is accomplished through sparse yet complementary connectivity to individual SPNs.

The density and overlap of cortical axons in the dorsolateral striatum (DLS) have suggested that striatal neurons integrate widespread information from cortical regions that are functionally related. However, in vivo, DLS neuronal responses to sensory stimuli have shown unexpectedly high selectivity, raising questions about the actual degree of convergence of functional corticostriatal projections on individual striatal cells. Here, we investigated this question by focusing on the projections from different whisker cortical columns in mice, as they overlap in the striatum and are co-active during behavior. Using ex vivo patch-clamp recordings in the DLS and glutamate uncaging for focal stimulations in the barrel cortex, we were able to map the location of presynaptic neurons to individual striatal projection neurons (SPNs). We found that each SPN was innervated by cells located in a small number of whisker cortical columns scattered across the barrel field in the slice. Connectivity of single SPNs with cortical neurons was thus highly discontinuous horizontally, despite the presence of more potential connections. Moreover, connectivity patterns were specific to each cell, with neighboring SPNs sharing few common clusters of presynaptic cells in the cortex. Despite this sparse and distinct innervation of individual SPNs, the projection was topographically organized at the population level. Finally, we found similar innervation patterns for D1- and D2-type SPNs, but observed differences in synaptic strength in their connections with certain cortical layers, notably the associative layer 2/3. Our results suggest that the high convergence of somatosensory inputs to the striatum, enabled by diffuse and overlapping cortical innervation, is accomplished through sparse yet complementary connectivity to individual SPNs.

Cognitive inflexibility covaries with substance use disorder (SUD) risk. To determine if there is a neural relationship between these phenomena, glutamate and dopamine release in the dorsomedial (DMS) and dorsolateral (DLS) striatum were measured as rats performed a discrimination and strategy switching test. Elevations in glutamate release, with reductions in dopamine, at trial initiation (DLS) and prior to choice (DMS and DLS) predicted fast strategy switching and punishment sensitive cocaine seeking. Elevations in DLS and DMS dopamine release at these respective timestamps predicted slow switching and punishment resistance. Orbitofrontal cortex and intralaminar thalamus were significant contributors to DLS and DMS glutamate release, but their relative contributions differed between rats that were fast or slow strategy switchers, and in how they affected behavior. As such, these data describe a neural signature of flexibility and associated circuitry that could be used to predict and treat SUDs in humans.

SummaryAnimals can flexibly initiate actions guided by external cues or by internal drive. Disease states often disrupt cue-driven and self-paced actions in distinct ways, underscoring separable neural mechanisms. Such differences could arise from specialized circuits dedicated to each action mode or shared neuronal populations that shift their dynamics across contexts. To distinguish between these possibilities, we developed a task in which mice performed a lever press either spontaneously or in response to a cue, enabling direct comparison of internally and externally triggered movements. Two-photon calcium imaging in dorsolateral striatum revealed subpopulations of neurons tuned to the cue, movement, or post-movement periods. One cluster was consistently active around movement regardless of context, yet population dynamics diverged prior to action. Support vector machine decoding and subspace analyses revealed distinct “context” and “action” components within the same population. Both D1- and D2-SPNs contributed to both subspaces, with D1-SPNs more active at the time of the sensory stimulus. These results show that context shapes dynamics in shared action-encoding neurons within striatal circuits, suggesting that different initiation contexts are funneled into a common action space that flexibly supports movement execution.

Nicotine is one of the most widely used addictive substances, yet its primary reinforcing effects are relatively weak. Nicotine's ability to potentiate responding for conditioned reinforcers is thought to drive persistent drug use. Here, we show that nicotine failed to alter responding maintained by a primary sucrose reinforcer (under a variable-ratio [VR] 11 schedule) across a broad dose range (0.01-1.0 mg/kg). Yet, nicotine enhanced operant behavior under a second-order reinforcement schedule in which responses produced sucrose-associated cues and were only intermittently reinforced by sucrose itself. Together, we demonstrate that nicotine selectively augments behavior maintained by conditioned reinforcers (sucrose-associated cues) in a rate-dependent manner, increasing responding only in mice with low rates of reinforcement behavior at baseline. Using fiber photometry, we demonstrate that nicotine selectively amplified cue-evoked dopamine release in the dorsolateral striatum-but only in low baseline responders-while having no effect on dopamine signaling in the nucleus accumbens. These effects were blocked by the nicotinic receptor antagonist mecamylamine. Further, nicotine's influence on behavior was abolished when the contingency between action and conditioned stimuli was disrupted, indicating that nicotine strengthens cue control of behavior rather than increasing motivation generally. Collectively, these findings reveal that nicotine's behavioral actions emerge through an interaction between pharmacological mechanisms, behavioral contingencies, and individual differences in baseline behavioral control. SIGNIFICANCE STATEMENT: Nicotine strengthens the impact of environmental cues on behavior by amplifying dopamine signals in specific projection targets, but only in individuals with specific behavioral traits. This reveals how nicotine hijacks learning processes to promote persistent, cue-driven actions.

BackgroundThe dorsolateral striatum (DLS) is predominantly associated with movement and habitual behaviors while the dorsomedial striatum (DMS) is primarily recognized for regulating goal-directed reward-seeking behaviors. Notably, activating striatal direct-pathway medium spiny neurons (dMSN) enhances reinforcement learning, whereas stimulating striatal indirect-pathway MSN (iMSN) induces avoidance behaviors and inhibits reinforcement. Although extensive research has compared the functions of DMS versus DLS or iMSN versus dMSN in reward-seeking, it remains unclear whether iMSN may exhibit distinct roles in suppressing compulsive alcohol-seeking within each specific striatal subregion.Aims & ObjectivesIn this study, we sought to address this by selectively ablating iMSNDMS->GPe and iMSNDLS->GPe and trained mice to exhibit goal-directed and habitual reward-seeking behaviors using random ratio (RR) and random interval (RI) operant conditioning, respectively. We also aim to investigate whether circuit-specific partial ablation alters compulsive alcohol-seeking and drinking behaviors.MethodTo ablate the circuit in a targeted fashion, we employed a genetically engineered caspase 3, in which activation results in apoptosis. We also investigated compulsive ethanol-seeking behaviors using nose-poke operant alcohol-drinking experiments in a two-bottle choice (water versus alcohol) paradigm with or without quinine adulteration.ResultsInterestingly, caspase 3-mediated ablation of iMSNDMS->GPe resulted in an insensitivity to satiety-based reward devaluation in RR-trained mice, consistent with a shift toward habitual behavior. The ablation of iMSNDLS->GPe did not influence locomotor function in the open field test and accelerated rotor rod or anxiety-related behaviors, nor did it affect the expression of habitual or goal-directed seeking behavior during the devaluation of a random interval or random ratio operant schedule. However, when subjected to quinine adulteration, mice with iMSNDLS->GPe ablation exhibited increased compulsive alcohol-seeking behaviors in habitual and goal-directed operant conditioning paradigms. Consistently, in a separate cohort of mice, we found that the iMSNDLS->GPe ablated mice showed higher preference and consumption of alcohol than control mice in a two-bottle choice experiment, with increasing quinine concentration and exhibiting more compulsive-like behavior. The iMSNDLS->GPe ablated mice prefer more ethanol than control mice while still developing an aversion to seeking the reward.Discussion & ConclusionsAltogether, our findings reveal an essential role of iMSNDLS->GPe in behavioral flexibility and inhibiting compulsive-like behavior, which may be a potential therapeutic target for alcohol and other addictive disorders. Our findings reveal a novel role of iMSNDLS->GPe in behavioral flexibility and constraining habitual reward-seeking, which may be a potential therapeutic target for alcohol use disorder and other compulsive disorders.

The sensorimotor arm of the basal ganglia is a major part of the mammalian motor control network, yet whether it supports all movements or is specialized for task-oriented behaviors remains unclear. To examine this, we probed the contributions of the rat sensorimotor striatum (dorsolateral striatum (DLS)) in two behavioral domains: free exploration, in which naturalistic behaviors are expressed, and during a motor task. In contrast to prior work, which showed the DLS being essential for generating task-specific learned movements, DLS lesions had no effect on naturalistic behaviors like rearing, grooming or walking. To explore the neural basis of this functional dissociation, we compared DLS activity across the two domains. Although neural activity reflected movement kinematics in both, the kinematic codes differed starkly. These findings suggest that sensorimotor basal ganglia are not essential parts of mammalian motor control but, rather, shift their output into a motor-potent space to shape task-specific behaviors.

It has long been known that the dorsomedial (DMS) and dorsolateral striatum (DLS) mediate distinct forms of action control, with DMS mediating goal-directed actions and DLS mediating habits. Recent evidence suggests that, in accord with its role in goal-directed control, unilateral stimulation of dorsomedial striatum (DMS) enhances actions contralateral to the stimulation in a manner that scales with the prior reward history of that action. In the current study, we assessed whether the effects of unilateral stimulation of the dorsolateral striatum (DLS) induce a response bias that reflects enhanced habitual control, as measured by the effect of stimulation on ongoing goal-directed control. Rats were first trained to press two levers for distinct outcomes in a manner likely to induce goal-directed control of these actions. We then assessed the effects of unilateral hM3D DREADDs-induced stimulation of the DMS or DLS, applied during an outcome devaluation choice test and outcome-mediated reinstatement, both known to depend on DMS activity. DMS stimulation had no impact on the choice of either the action ipsilateral or contralateral to the stimulation, either during devaluation or reinstatement. In contrast, stimulation of the DLS abolished goal-directed control in both tests, reducing sensitivity to outcome devaluation both on the ipsilateral and contralateral lever. In addition, stimulation both attenuated reinstatement and induced a response bias away from the contralateral lever during reinstated responding. These data suggest that, rather than driving motor output per se, the DLS is important for selecting motor programs triggered by stimulus-response associations.

The control of instrumental actions engages distinct behavioral strategies whose contributions are regulated with experience. Instrumental performance, which depends on the causal relationship between actions and their outcomes (A–O), relies on flexible, goal-directed control of behavior. Actions can become less sensitive to changes in action–outcome (A–O) contingencies with repetition, resulting in more inflexible, habitual behaviors. The loss of flexibility with repetition requires plasticity at corticostriatal circuits. However, the underlying molecular mechanisms are not yet established, and how these mechanisms specifically relate to the inability to adapt to new contingencies is unknown. In mice, we find that inflexible behavioral performance following overtraining of an appetitive instrumental task is associated with a reduced capacity of mGluR5 receptors in the dorsolateral striatum (DLS) to engage intracellular signaling in response to changes in action–outcome contingency. We also observed dichotomous modulation of timing-dependent synaptic depression (tLTD) at striatal projection neurons of the indirect (iSPNs) and direct (dSPNs) pathways. Preventing overstimulation of mGluR5 signaling through a homotypic process preserved behavioral sensitivity to changes in A–O contingencies despite overtraining, and averted the related biochemical and synaptic changes. Furthermore, mGluR5 couples to different signaling pathways to regulate tLTD in iSPNs and dSPNs. Our findings demonstrate that decreased signaling capacity of mGluR1/5, accompanied by cell-type-specific modulation of corticostriatal synapses in the DLS, represents a key molecular mechanism underlying overtraining-induced behavioral inflexibility.

Spaceflight induces multifaceted physiological adaptations, yet the molecular mechanisms underlying microgravity-associated neurological dysfunction remain poorly defined. Although microgravity is known to influence the dopaminergic system, most existing studies have relied on surrogate markers such as the expression of dopamine (DA) biosynthetic enzymes and transport proteins rather than direct measurements of neurotransmission. To address this gap, we employ a 14-day hindlimb unloading (HU) mouse model to simulate microgravity conditions and utilize fast-scan cyclic voltammetry (FSCV) to directly quantify stimulus-evoked DA release in the dorsolateral striatum. Our results reveal a significant reduction in DA levels under HU conditions. Immunofluorescence analysis further indicates that the observed deficits are associated with a decrease in the expression of tyrosine hydroxylase (TH), the rate-limiting enzyme in DA synthesis. To elucidate the underlying mechanisms, we use high spatiotemporal resolution single-cell amperometry (SCA) to analyze exocytotic kinetics at the vesicular level and observe marked impairments in neurotransmitter release dynamics including reduced quantal size, narrowed initial fusion pore diameter, and delayed fusion pore closure. These alterations in vesicle fusion behavior are correlated to behavioral deficits in motor coordination and cognitive performance. This finding essentially establishes a direct mechanistic link between simulated microgravity-induced dopaminergic dysfunction and neurobehavioral impairments, guiding the development of targeted neuroprotective strategies for spaceflight missions.

Behavioral flexibility allows organisms to modify actions based on new information, such as shifts in reward value or availability, and is promoted by the dorsomedial striatum (DMS). In contrast, behavioral inflexibility provides efficiency and automaticity in familiar contexts, and is promoted by the dorsolateral striatum (DLS). Importantly, chronic elevation of the primary stress hormone, corticosterone (CORT) in rodents or cortisol in humans, impairs behavioral flexibility through dendritic atrophy in the DMS, and promotes inflexible behavioral response strategies through dendritic outgrowth in the DLS. However, understanding of the molecular mechanisms that underlie the structural changes promoting behavioral inflexibility is lacking. We used a food-motivated operant task in male and female mice to define synaptic plasticity gene regulation supporting a decreased DMS activity and increased DLS activity in the shift to inflexible behavior with CORT. We discovered that CORT-accelerated loss of behavioral flexibility is accompanied by decreased DMS- and increased DLS-specific synaptic plasticity gene expression, and that distinct genes are either differentially expressed or spliced in the transition to inflexible behavior. Splicing analysis suggests that repressed activity in the DMS during the transition to inflexible behavior reflects both reduced expression and increased degradation of plasticity-related mRNA transcripts. Finally, given the ability of CORT to influence histone acetylation, we defined CORT-mediated H3K9ac enrichment profiles associated with synaptic plasticity gene regulation stratified by sex and striatal subregion. This study is the first to define CORT-driven epigenetic regulation in the DMS and DLS during the transition from flexible to inflexible behavior in male and female mice.

Coordinated motor behavior emerges from information flow across brain regions. How long-range inputs drive cell-type-specific activity within motor circuits remains unclear. The dorsolateral striatum (DLS) contains direct- and indirect-pathway medium spiny neurons (dMSNs and iMSNs) with distinct roles in movement control. In mice performing skilled locomotion, we recorded from dMSNs, iMSNs, and their cortical and thalamic inputs identified by monosynaptic rabies tracing. An RNN classifier and clustering analysis revealed functionally heterogeneous subpopulations in each population, with dMSNs preferentially activated at movement onset and offset, and iMSNs during execution. Cortical and thalamic inputs were preferentially activated during onset/offset and execution, respectively, though dMSN- and iMSN-projecting neurons in each region showed similar patterns. Locomotion phase-specific rhythmic activity was found in a subset of thalamic dMSN-projecting neurons and dMSNs. Cortex or thalamus inactivation reduced MSN activity. These findings suggest that corticostriatal and thalamostriatal inputs convey complementary motor signals via shared and cell-type-specific pathways.

Alcohol use disorder is characterized by persistent drinking in the face of negative consequences. Such inflexible drinking requires dorsolateral striatum fast-spiking interneurons, which comprise roughly 1% of all striatal neurons. How chronic ethanol exposure affects fast-spiking interneuron physiology is poorly understood. We discover in mice that chronic ethanol exposure induced a dramatic loss of GABAergic, but not glutamatergic, synapses onto dorsolateral striatum fast-spiking interneuron somata and proximal dendrites where perineuronal nets, a subdivision of the extracellular matrix, are enriched. We found that chronic ethanol exposure degraded these perineuronal nets and that enzymatically degrading perineuronal nets similarly reduced GABAergic transmission onto dorsolateral striatum fast-spiking interneurons. Modeling the effect of alcohol, we find that silencing extrinsic GABAergic projections to the dorsolateral striatum increased voluntary ethanol consumption. Taken together, these data suggest chronic alcohol exposure remodels perineuronal nets and inhibitory synapses on fast-spiking interneurons to facilitate alcohol drinking.

The striatum, the central hub of cortico-basal ganglia loops, contains functionally heterogeneous subregions distinguished by the topographic patterns of structural connectivity. These subregions mediate various processes of procedural learning. However, it remains unclear when and how striatal subregions engage in the acquisition of sensory stimulus-based decision-making. A neuroimaging of regional brain activity shows that the anterior dorsolateral striatum (aDLS) and posterior ventrolateral striatum (pVLS) in rats are activated in a different temporal pattern during the acquisition phase of auditory discrimination. Chronic and transient pharmacologic manipulations show that the aDLS promotes the behavioral strategy driven by the stimulus-response association while suppressing that by the response-outcome association, and that the pVLS contributes to forming and maintaining the stimulus-response strategy. Electrophysiological recording indicates that subpopulations of aDLS neurons predominantly represent the outcome of specific behaviors at the initial period of discrimination learning, and that pVLS subpopulations encode the beginning and ending of each behavior according to the progress of learning. In addition, other subpopulations of striatal neurons indicate sustained activation after obtaining reward with distinct patterns reflecting the stimulus-response associations. Our findings demonstrate that aDLS and pVLS neurons integrate the new learning of auditory discrimination in spatiotemporally and functionally different manners.

The striatum, the central hub of cortico-basal ganglia loops, contains functionally heterogeneous subregions distinguished by the topographic patterns of structural connectivity. These subregions mediate various processes of procedural learning. However, it remains unclear when and how striatal subregions engage in the acquisition of sensory stimulus-based decision-making. A neuroimaging of regional brain activity shows that the anterior dorsolateral striatum (aDLS) and posterior ventrolateral striatum (pVLS) in rats are activated in a different temporal pattern during the acquisition phase of auditory discrimination. Chronic and transient pharmacologic manipulations show that the aDLS promotes the behavioral strategy driven by the stimulus-response association while suppressing that by the response-outcome association, and that the pVLS contributes to forming and maintaining the stimulus-response strategy. Electrophysiological recording indicates that subpopulations of aDLS neurons predominantly represent the outcome of specific behaviors at the initial period of discrimination learning, and that pVLS subpopulations encode the beginning and ending of each behavior according to the progress of learning. In addition, other subpopulations of striatal neurons indicate sustained activation after obtaining reward with distinct patterns reflecting the stimulus-response associations. Our findings demonstrate that aDLS and pVLS neurons integrate the new learning of auditory discrimination in spatiotemporally and functionally different manners.

The striatum, the central hub of cortico-basal ganglia loops, contains functionally heterogeneous subregions distinguished by the topographic patterns of structural connectivity. These subregions mediate various processes of procedural learning. However, it remains unclear when and how striatal subregions engage in the acquisition of sensory stimulus-based decision-making. A neuroimaging of regional brain activity shows that the anterior dorsolateral striatum (aDLS) and posterior ventrolateral striatum (pVLS) in rats are activated in a different temporal pattern during the acquisition phase of auditory discrimination. Chronic and transient pharmacologic manipulations show that the aDLS promotes the behavioral strategy driven by the stimulus-response association while suppressing that by the response-outcome association, and that the pVLS contributes to forming and maintaining the stimulus-response strategy. Electrophysiological recording indicates that subpopulations of aDLS neurons predominantly represent the outcome of specific behaviors at the initial period of discrimination learning, and that pVLS subpopulations encode the beginning and ending of each behavior according to the progress of learning. In addition, other subpopulations of striatal neurons indicate sustained activation after obtaining reward with distinct patterns reflecting the stimulus-response associations. Our findings demonstrate that aDLS and pVLS neurons integrate the new learning of auditory discrimination in spatiotemporally and functionally different manners.

Sex- and site-specific effects of GPER-1 activation on saccharin vs cocaine preference in male and female rats.