Overview of Striatal Circuits
The striatum is a major brain nuclei in the basal ganglia (BG) system. The BG consists of set of corticobasal ganglia-cortical loops, which are a series of parallel projection loops that convey limbic, associative, and sensorimotor information. In this circuit, cortical neurons send input to striatum, which conveys output through various BG nuclei, relaying information to thalamus and then ultimately back to cortex. The striatum consists of the dorsal striatum (dStr, caudate and putamen in humans), which regulates actions and habits, and the ventral striatum (a.k.a. nucleus accumbens [NAc]), which is involved in motivation and reinforcement. These striatal areas have distinct projections through the BG output nuclei, consisting of two distinct pathways (often referred to as the direct and indirect pathways) and they were originally proposed to play antagonistic but balancing roles on BG output and behavior. The two pathways can be resolved at a cellular level in the main projection neurons of the striatum. The projection neurons, which comprise 90%–95% of all neurons in the striatum, are medium spiny neurons (MSNs), which are divided into two morphologically identical and heterogeneously distributed cell types. The MSNs in striatum are subdivided into two subtypes based on their axonal targets. MSNs that are considered part of the direct pathway project to globus pallidus internal (GPi), ventral pallidum (VP), and midbrain regions including substantia nigra (SN) and ventral tegmental area (VTA); whereas the indirect pathway MSNs project to the globus pallidus external (GPe) and VP ( Fig. 9.1 ). However, it is important to note that MSN projections from dStr appear more segregated from those in NAc. The dStr MSN subtypes have distinct projections with minimal overlap to BG nuclei, whereas the NAc MSN subtypes both send input to VP. Thus this ventral BG circuit does not quite represent the classical direct and indirect pathways (see Fig. 9.1 ). Due to this overlap in NAc MSN subtype projections, we refer to these two neuron subtypes based on their enrichment of dopamine receptors 1 versus 2, with D1-MSNs being part of the classical direct pathway and D2-MSNs part of the indirect pathway. Although both D1-MSNs and D2-MSNs in NAc project to VP, the NAc D1-MSNs also send projections to classical direct pathway nuclei including GPi, SN, and VTA (see Fig. 9.1 ).
Along with their enrichment of D1 versus D2 receptors, the two MSN subtypes are further distinguished by their differential expression of several other genes, most notably G-protein–coupled receptors and neuropeptides. D1-MSNs express muscarinic receptor 4, substance P, and dynorphin, whereas D2-MSNs express adenosine receptor 2a, G-protein–coupled receptor 6, and enkephalin ( Fig. 9.2 ). Through the two BG pathways the D1-MSNs versus D2-MSNs have been demonstrated to display differential behavioral output. Activity in the D1-MSNs is implicated in movement initiation, reinforcement, and reward seeking, whereas activity in the D2-MSNs antagonizes the D1-MSN pathway, thus inhibiting movement, promoting punishment or avoidance, and inhibiting reward seeking. a
a References 23, 25, 39, 46, 48, 52, 55.However, there are some studies that support a role for coordinated activity in these two neurons in actions and natural reward behaviors. Studies on animal models of addiction and depression have demonstrated distinct roles of these MSN subtypes in striatal circuits in these motivational diseases. This chapter discusses these current findings and the overlap between these striatal circuits in addiction and depression.
Striatal Circuit Activity in Animal Models of Addiction and Depression
Striatal MSN Subtype Activity in Addictive Drug Exposure and Behavior
Much of the evidence for the differential roles of D1-MSNs and D2-MSNs in addiction is based on studies examining cocaine-induced behaviors in rodents, using neuron-subtype–specific techniques to activate or inhibit these MSN subtypes. Enhanced activity in D1-MSNs underlies the reinforcing and sensitizing effects of cocaine. Likewise, blocking activity in D2-MSNs results in similar outcomes. b
b References 8, 11, 23, 39, 52, 70.The first insight into MSN-subtype participation in psychostimulant-mediated behavior involved NAc D2-MSN ablation. Ablating these MSNs increased psychostimulant-induced conditioned place preference without altering normal locomotion. Subsequent studies demonstrated an opposite role for D1-MSNs versus D2-MSNs in psychostimulant-mediated behavior. Optogenetic stimulation, using the blue light–activated channelrhodopsin-2 (ChR2) of NAc D1-MSNs enhances the rewarding properties of cocaine, and NAc D2-MSN optogenetic stimulation reduces this outcome. In addition, after repeated exposure to cocaine the optogenetic activation of NAc D1-MSNs resulted in enhanced locomotor activity. This implicates that cocaine primes these MSN subtypes to display a sensitized response to other stimuli, in this case artificial activation. The selective blockade of neurotransmission in D1-MSNs reduces cocaine-induced locomotor sensitization and conditioned place preference. Conversely, using optogenetics or chemogenetics, the latter using designer receptor activated by designer drugs (DREADDs), inhibition of D1-MSNs or activation of D2-MSNs reduces psychostimulant-induced locomotor sensitization, while the inhibition of D2-MSNs increases this behavior. Furthermore, chemogenetic inhibition of D2-MSNs, in cocaine self-administration, enhanced the motivation to obtain cocaine, whereas optogenetic activation of D2-MSNs suppressed cocaine self-administration. Finally, a recent study using in vivo fiber photometry with the calcium indicator, gCamp6f, confirmed the MSN subtype activity manipulation studies described above. This study showed that acute cocaine exposure enhanced D1-MSN and suppressed D2-MSN activity, and that cocaine-induced D1-MSN activity is required for formation of cocaine–context associations. In addition, MSN subtype–specific signaling encodes contextual information about the cocaine environment such that increased D1-MSN activity precedes entry into a cocaine-paired environment, while decreased D2-MSN activity occurred after entering the cocaine-paired environment. Finally, inhibiting this D1-MSN calcium signal by DREADD inhibition blocked the cocaine-conditioned preference. Altogether, these findings show that a circuit imbalance of these D1-MSN versus D2-MSN pathways occurs upon cocaine exposure, leading to an enhanced D1-MSN pathway, thus promoting cocaine-seeking, intake, and sensitization behaviors ( Fig. 9.3 ).
Electrophysiology studies examining psychostimulant-induced plasticity in the MSN subtypes corroborate with the activity studies described earlier. Excitatory synaptic potentiation occurs at D1-MSNs after repeated cocaine exposure or cocaine self-administration. Of interest, mice that display poor cocaine intake display enhanced excitatory synaptic input at D2-MSNs. Consistent with this, increased dendritic spine remodeling occurs in D1-MSNs after repeated injections (i.p) of cocaine. Evidence demonstrates that the increased spines in D1-MSNs are thin or immature spines, characterized as silent synapses, since they consist of N -methyl- d -aspartate (NMDAR) receptors but lack α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPAR) receptors. The silent synapses, which are typical throughout the immature brain, can either retract or develop into fully functional synapses to induce new neural circuits, after periods of cocaine withdrawal. It is likely that these new neural circuits mediate enduring behaviors in response to cocaine, such as relapse behavior. Future studies examining MSN subtypes in relapse behavior will be important for understanding their role in the long-term effects of cocaine and the transition from early drug taking to the addictive state. Finally, examination into MSN subtype output in the VP, the one region receiving dense innervation from both MSN subtypes, demonstrates potentiated output of D1-MSNs but weakened output of D2-MSNs after repeated cocaine exposure. This study further showed that optogenetic depotentiation of D1-MSN output to the VP abolished cocaine locomotor sensitization; however, restoring D2-MSN transmission to VP did not alter this behavior.
As described in the preceding text, much of the data examining the striatal MSN circuits in drug abuse are from studies performed with cocaine. However, a small number of studies examine striatal circuits in morphine-mediated behaviors. Similar to the cocaine studies, optogenetic activation of NAc D1-MSNs enhanced morphine-conditioned place preference, whereas optogenetic activation of NAc D2-MSNs blunted this behavior. Of interest, examination of plasticity in these MSNs reveals a different outcome compared to cocaine, since silent synapses are induced in D2-MSNs after repeated morphine exposure. Finally, examination of analgesic tolerance demonstrated that optogenetic activation of D1-MSNs facilitates the development of morphine tolerance, whereas activation of D2-MSNs did not affect the development of tolerance. Additional studies examining these MSN subtypes in opiate-mediated behaviors are needed to uncover the mechanisms accounting for differences between cocaine and morphine.
Striatal MSN Subtype Activity in Depression-Like Behavior
There are sparse studies examining activity in MSN subtypes in animal models of depression. In contrast to the cocaine studies described earlier (which show enhanced excitatory synaptic input onto D1-MSNs and reduced input onto D2-MSNs), stress models, showing depression-like behavior, display reduced excitatory input onto D1-MSNs and/or enhanced input onto D2-MSNs. The D2-MSN data are in line with those of previous studies demonstrating enhanced excitatory input onto MSNs, which correlates with increased mushroom-shaped spines in MSNs, using an animal model of stress-induced depression, chronic social defeat stress. Use of optogenetics or DREADDs in mice that underwent chronic social defeat stress uncovered a bidirectional role for MSNs in depression-like behavior. Repeated high-frequency optogenetic activation of D1-MSNs, in mice that display depression-like behavior to chronic social defeat stress, resulted in an antidepressant phenotype. In contrast, repeated DREADD inhibition of D1-MSNs in mice displaying resilient behavior (lack of depression-like behavior) after chronic social defeat stress shifted these mice to a susceptible, depression-like state. Altering activity in D2-MSNs after stress did not alter behavioral outcomes to chronic social defeat stress. However, priming D2-MSNs with repeated activity prior to stress induced a depression-like outcome to a subthreshold social defeat stress. These data are in line with the BG model of activity in D1-MSNs promoting reward, while activity in D2-MSNs promotes avoidance or punishment.
Molecular Mechanisms in Striatal Circuits in Addiction
MSN Subtype Signaling Mechanisms in Addictive Drug Exposure and Behavior
D1-MSNs and D2-MSNs display different molecular adaptations in response to cocaine. This potentially occurs via differential signaling through dopamine receptors. Enhanced dopamine levels, occurring with exposure to drugs of abuse, can positively modulate excitatory glutamatergic input in D1-MSNs through activation of D1-receptor signaling via G s or G olf , which stimulate adenylyl cyclase, leading to increased protein kinase A (PKA) activity. In contrast, dopamine negatively modulates D2-MSNs through D2-receptor signaling via G i and G o , which inhibit adenylyl cyclase causing decreased PKA activity. This can lead to differential phosphorylation of the dopamine- and cAMP-regulated neuronal phosphoprotein (DARPP-32) in MSN subtypes after cocaine exposure. As a result, the deletion of DARPP-32 from D1-MSNs decreases cocaine-induced locomotion, while its deletion from D2-MSNs increases locomotion. In addition, brain-derived neurotrophic factor (BDNF) signaling has been shown to exert opposing roles on MSN subtypes. Deletion of the BDNF receptor, tropomyosin receptor kinase B (TrkB), from D1-MSNs increases cocaine-conditioned place preference and locomotor sensitization, while TrkB deletion from D2-MSNs reduces these behaviors. Of interest, the D2-MSN results are consistent with those of previous studies that used non-cell-type specific deletion of TrkB from NAc, demonstrating that the main effects of cocaine on BDNF might be occurring through D2-MSNs. However, assessment of morphine-conditioned place preference in these TrkB MSN subtype lines showed enhanced morphine place preference with deletion in D1-MSNs but no altered behavior with deletion in D2-MSNs. Investigation of dopamine- and BDNF-signaling targets, with repeated cocaine exposure, demonstrated activate extracellular signal-regulated kinase (pERK) associated with a downregulation of its direct nuclear target mitogen- and stress-activated kinase-1 (pMSK1) in D1-MSNs exclusively. Finally, the cell-type-specific silencing of p11 (S100A10), a protein linked with the transport of neurotransmitters and receptors to the plasma membrane, on D1-MSNs increases cocaine-conditioned place preference.
Transcription Factors in MSN Subtypes in Addictive Drug Exposure and Behavior
Overall, molecular adaptations occur in both MSN subtypes in response to drugs of abuse, such as cocaine. However, many experiments highlight major molecular alterations in the D1-MSN pathway, confirming its predominant role in cocaine-mediated behaviors. This predominant role of D1-MSNs has been well documented with immediate early gene transcription factors. Early studies examining immediate early genes provided the first insight into how the MSN subtypes respond to psychostimulants. Previous studies, in rats, demonstrate c-Fos induction in both MSN subtypes when a psychostimulant is given in a novel environment. Using D1-GFP and D2-GFP reporter mice, researchers demonstrate that c-Fos induction by cocaine in a novel environment occurs primarily in D1-GFP MSNs throughout striatum with a small induction in D2-GFP MSNs in dorsal striatum. Isolation and molecular profiling of active striatal neurons in context-dependent cocaine locomotor sensitization, using a c-Fos reporter rat line, demonstrated that these neuronal ensembles express both D1-MSN and D2-MSN markers. However, they express higher levels of a D1-MSN enriched gene, dynorphin, and lower levels of D2-MSN enriched genes, D2 and adenosine 2A receptor, suggesting a greater number of D1-MSNs in this population. c-Fos deletion in D1-MSNs, blunted cocaine-induced locomotor sensitization and MSN dendritic spine formation. Of interest, c-Fos deletion in D1 neurons did not alter cocaine-conditioned place preference but it did prevent the extinction of this contextual association. These data illustrate a dynamic role for c-Fos induction in D1-MSNs; however, one cannot rule out the differential behavioral effects as being mediated by other brain regions that express the D1 receptor.
The immediate early gene (IEG), FosB, has been well studied in MSN subtypes in addiction. FBJ murine osteosarcoma viral oncogene homolog B (FosB) is induced in striatum by acute cocaine, but the long-lasting ΔFosB, generated from the FosB primary transcript, persistently accumulates after chronic psychostimulant exposure. This long-lasting induction of ΔFosB by cocaine is dependent on D1-receptor signaling, and use of a D1-GFP reporter lines confirmed that ΔFosB induction occurs primarily in D1-MSNs after chronic cocaine. Consistent with these findings, FosB messenger RNA (mRNA) was induced in D1-MSNs with acute and chronic injection (i.p.) of cocaine using a ribosomal tagging approach.
Initial studies using a transgenic line with preferential overexpression of ΔFosB D1-MSNs resulted in enhanced locomotor and conditioned place preference responses to cocaine. In addition, this D1-MSN ΔFosB line shows facilitated acquisition to cocaine self-administration at low-threshold doses and enhanced effort to maintain self-administration of higher doses on a progressive ratio schedule of reinforcement. These behaviors are occurring potentially through enhanced structural plasticity in D1-MSNs, since adenoassociated virus (AAV)–mediated ΔFosB overexpression in NAc enhances MSN structural plasticity. Use of Cre-inducible herpes simplex virus (HSV) to overexpress ΔFosB in D1-MSNs in the NAc of D1-Cre mice confirmed the enhanced cocaine-mediated behavioral responses and showed that ΔFosB alone can enhance immature spine formation and reduce AMPAR/NMDAR ratios in D1-MSNs. These structural and synaptic plasticity changes by ΔFosB are an indication of enhanced silent synapses, which are characteristic of cocaine effects on D1-MSNs. . Thus, ΔFosB may set the stage for long-term cocaine abuse by regulating the establishment of silent synapses in D1-MSNs during the initial stage of drug exposure. Finally, investigation of ΔFosB overexpression in D2-MSNs had no effect on cocaine-induced behaviors or spine formation but did enhance AMPAR/NMDAR ratios, suggesting that ΔFosB in these MSNs might play a role in mature spine formation. A mechanistic role of ΔFosB in promoting behavioral and structural plasticity after cocaine has been examined. The D1-MSN ΔFosB line displayed enhanced expression of GluR2 in NAc, and GluR2 overexpression in NAc enhances cocaine conditioned place preference. In addition, ΔFosB increased CAMKIIα gene expression in NAc of the D1-MSN ΔFosB line and the enhanced cocaine-mediated behavioral and structural plasticity effects of ΔFosB in NAc are CAMKIIα dependent. ΔFosB also transcriptionally regulates a number of genes in NAc by chronic cocaine. Future studies using neuronal subtype chromatin immunoprecipitation to examine FosB enrichment on target genes can provide improved understanding into the MSN subtype transcriptional role of ΔFosB in cocaine action.
ΔFosB induction has been examined in other drugs of abuse including THC, ethanol, and opioids. Similar to the cocaine studies, repeated THC and ethanol leads to increased ΔFosB in D1-MSNs. Of interest, chronic morphine and heroin self-administration resulted in increased ΔFosB in both MSN subtypes. This could reflect induction in D1-MSNs in response to the rewarding effects of morphine and induction in D2-MSNs during the aversive, withdrawal phase of opioids. However, the D1-MSN-specific ΔFosB line displayed enhanced place preference for morphine, reduced morphine analgesia, and accelerated morphine tolerance, whereas a D2-MSN-specific ΔFosB line did not show any altered behavioral responses to morphine.
Another transcription factor examined is the early growth response (Egr) family member, Egr3. A modest decrease in Egr3 in total NAc tissue was observed after repeated cocaine exposure and cocaine self-administration. However, use of the RiboTag methodology to isolate ribosome-associated mRNA from each MSN subtype, demonstrated an enrichment of Egr3 mRNA in D1-MSNs, with a decrease occurring in D2-MSNs. Overexpressing Egr3 in D1-MSNs and knocking down Egr3 in D2-MSNs enhanced cocaine-conditioned place preference and locomotor sensitization, while reducing Egr3 in D1-MSNs and enhancing it in D2-MSNs blunted these behaviors, confirming the opposing role of Egr3 in both MSN subtypes. These results further support the predominant role for D1-MSNs in cocaine-mediated behaviors; however, the cell-type-specific study demonstrated that the molecular changes in D2-MSNs also account for critical aspects of the responses to cocaine. Taken together, the above studies show that changes in transcription factor regulation are pivotal in cocaine-related behaviors.
MSN Subtype Epigenetic and Posttranscriptional Modifications in Addictive Drug Exposure and Behavior
In recent years, a growing number of studies have evaluated epigenetic changes induced by cocaine. Repeated cocaine exposure can induce stable changes in gene expression that may underlie addiction. However, only a few studies examined cell-type-specific epigenetic changes after cocaine exposure. For instance, in D1-GFP versus D2-GFP mice, an increase in phosphorylation of histone 3 on Ser-10 was found after acute and chronic cocaine injections (i.p). Using ribosome-associated mRNA profiling, a recent study found cocaine-induced decrease of G9a (a repressive histone methyltransferase) in both D1- and D2-MSNs. However, developmental knockout of G9a from D1-MSNs decreased cocaine-conditioned place preference and locomotor sensitization, while knockout from D2-MSNs had the opposite effect. Surprisingly, the G9a knockout from D2-MSNs induced a partial-phenotypic switch, making D2-MSNs more similar to D1-MSNs, providing insight on the epigenetic mechanisms, as well as potential developmental mechanisms contributing to cocaine abuse. Recently the histone arginine methylation enzyme, protein-R-methyltransferase-6 (Prmt6), was examined in MSN subtypes after repeated cocaine exposure. Ribosome-associated mRNA profiling revealed a downregulation of Prmt6 in D2-MSNs after repeated cocaine, which was consistent with reduced Prmt6 levels in total NAc in this condition. In contrast, Prmt6 was upregulated in D1-MSNs. The decreased Prmt6 levels led to a reduction of the repressive mark H3R2me2a on the Src kinase signaling inhibitor 1 ( Srcin1 ) gene, which resulted in increased Scrin1 protein in NAc after repeated cocaine. Overexpression of Prmt6 in D2-MSNs or total NAc enhanced cocaine-conditioned place preference, while overexpression in D1-MSNs reduced this behavior. Consistent with reduced Prmt6 resulting in increased Srcin1, the overexpression of Srcin1 in D2-MSNs or total NAc reduced cocaine-conditioned place preference, with opposite effects observed with D1-MSN overexpression. These results suggest that the effects of reduced Prmt6 in D2-MSNs counteracts the rewarding effects of cocaine through enhancement of Srcin1 in these neurons. Srcin1 is an endogenous inhibitor that constrains the activity of the Src family of protein tyrosine kinases. Further examination in this pathway in D2-MSNs could uncover improved information into the role of D2-MSNs in cocaine action. Other work has shown cell-type-specific and time-dependent epigenetic modifications after cocaine. For instance, H3K5 acetylation was steadily increased in D1-MSNs while only transiently in D2-MSNs, whereas H3K14 increased after acute cocaine in D1-MSNs and after chronic cocaine in D2-MSNs. This type of study further points out the importance of examining cell-type-specific patterns of histone modifications, since epigenetic changes may differ with drug-exposure time and have distinct effects on gene transcription.
In addition to transcriptional and epigenetic studies in cocaine abuse, researchers are beginning to examine posttranscriptional adaptations. Reduction in Argonaute 2 (Ago 2), which plays a role in micoRNA (miRNA) generation and miRNA gene silencing, in D2-MSNs reduces the motivation to self-administer cocaine. Furthermore, this study demonstrated a number of miRNAs enriched in D2-MSNs after cocaine exposure that are also downregulated in Ago 2–deficient striatum. Collectively, identifying transcriptional and posttranscriptional changes, such as chromatin modifications and miRNA functions, in striatal circuits in cocaine addiction will be important for better understanding of the complex molecular networks underlying addiction.