Maladaptive circuit rearrangements in the nucleus accumbens (NAc)-embedded reward network underlie the development of various psychiatric disease states, including addiction (Joffe et al., 2014; Turner et al., 2018) and depression (Russo and Nestler, 2013). The NAc directs reward-related motivational behavior by integrating glutamatergic input from assorted corticolimbic structures (Kalivas, 2009). The strength of these inputs onto medium spiny projection neurons (MSNs) is scaled by plasticity mechanisms at pre- and postsynaptic loci that often require the activity of G_i/o-coupled G-protein-coupled receptors (GPCRs) (Johnson and Lovinger, 2016). Numerous neuromodulatory systems implicated in addiction-related behavior engage presynaptic G_i/o-coupled GPCRs at glutamatergic synapses in the NAc, including the GABA_B receptor (GABA_BR) (Li and Slesinger, 2022), adenosine receptors (Ballesteros-Yáñez et al., 2018), cannabinoid receptor type-1 (CB₁R) (Wiskerke et al., 2008), 5-HT_1B/D serotonin receptors (Cao et al., 2013), opioid receptors (Reeves et al., 2022), and the histamine H3 receptor (H₃R) (Manz et al., 2021a). Although GPCRs hold promise as targets for therapeutic intervention, the precise effector pathways of these receptors at glutamatergic synapses in the NAc remain poorly understood.

Presynaptic G_i/o-coupled GPCRs decrease neurotransmitter release probability by engaging Gβγ-dependent signaling pathways (Atwood et al., 2014). Mobilization of Gβγ subunits at the presynaptic active zone can decrease terminal Ca²⁺ influx by inhibiting voltage gated Ca²⁺ channels (VGCCs) (Bean, 1989; Ikeda, 1996). Gβγ signaling can also promote inwardly rectifying K⁺ channel (Kir) opening, hyperpolarizing cells below spike threshold and limiting VGCC activity (Kofuji et al., 1995; Hibino et al., 2010). However, recent studies from our lab and others indicate that G_i/o-coupled GPCR signaling can directly inhibit vesicular release by binding synaptosomal-associated protein of 25 kDa (SNAP25), a member of the soluble N-ethylmaleimide attachment protein receptor (SNARE) complex (Zurawski et al., 2016, 2017; Manz et al., 2019). While mechanisms may be recruited in parallel to exert manifold control over synaptic function, it is unknown which G_i/o-coupled GPCR systems in the NAc utilize the Gβγ-SNARE interaction.

To address this gap, we employed whole-cell patch-clamp electrophysiology with high-affinity and biased agonist pharmacology in acute brain slices from transgenic mice containing a C-terminus 3-residue deletion of SNAP25 (SNAP25Δ3). Gβγ binds the botulinum toxin type-A cleavage site of SNAP25, disrupting the Ca²⁺-sensitive interaction with vesicular SNARE (v-SNARE) protein, synaptotagmin-1 (Wells et al., 2012). We previously demonstrated that Gβγ has a reduced ability to inhibit the interaction between SNAP25Δ3 and synaptotagmin-1 by approximately 47% compared to WT SNAP25 protein (Zurawski et al., 2019). Therefore, this model would allow us to explore the contributions of the Gβγ-SNARE complex interaction to G_i/o-coupled GPCR systems in the NAc. We report that basal glutamate release probability is decreased at synapses in the NAc of SNAP25Δ3 mice. Screening G_i/o-coupled GPCRs canonically confined to presynaptic domains at glutamatergic synapses in the NAc, we show that κ opioid, CB₁, adenosine A1, and histamine H3 receptors inhibit glutamatergic transmission onto MSNs independent of the Gβγ-SNARE interaction. In contrast, GABA_B, 5-HT_1B/D, and μ opioid receptor signaling is Gβγ-SNAP25-dependent, with plasticity elicited by μ opioid receptors completely abolished in SNAP25Δ3 mice. These findings offer insight into the diversity with which G_i/o-coupled GPCRs modulate glutamatergic transmission in the NAc and yield additional intracellular targets for the treatment of NAc-dependent pathologies.

The NAc receives robust glutamatergic input from limbic and paralimbic centers guiding motivational decision-making. Experience shifts NAc circuit activity by scaling the strength of these inputs through GPCR-dependent plasticity mechanisms, including G_i/o-coupled GPCRs implicated in addiction and depression (Lüscher and Malenka, 2011; Grueter et al., 2012). G_i/o-coupled GPCRs, including those mediating endogenous opioid, cannabinoid, and GABA signaling, canonically inhibit presynaptic release probability through Gαi/o and Gβγ-dependent pathways involving VGCCs, Kirs, and others. However, G_i/o-coupled GPCRs in the NAc, hippocampus, and BNST have recently been shown to inhibit vesicular neurotransmitter release through a Gβγ-dependent assembly with SNARE proteins such as SNAP25 (Zurawski et al., 2019). Here, we obtained a rigorous functional inventory of key G_i/o-coupled GPCRs in the NAc that reduce glutamatergic synaptic efficacy through the Gβγ-SNARE interaction. We report that select GPCR systems, each with temporally distinct plasticity patterns, engage SNAP25-dependent effector systems at glutamatergic synapses onto MSNs in the NAc.

First, we aimed to determine the consequences of the SNAP25Δ3 mutation on glutamatergic transmission in the nucleus accumbens. We found that the SNAP25Δ3 mice have a enhanced PPR and reduced sEPSC frequency when compared to WT animals (Figures 1C–F), which are both indicative of reduced vesicular release probability. This result is surprising in that the Gβγ-SNARE interaction is known to decrease vesicular release probability, and this interaction is compromised in the SNAP25Δ3 mice. One possibility is that because the SNAP25Δ3 mutation is present developmentally, compensatory mechanisms are reducing the probability of vesicle fusion to compensate for the loss of inhibition via the Gβγ-SNARE interaction. The basal changes in glutamatergic transmission are an important caveat in this study, however, further studies are needed to elucidate which specific mechanisms are driving the decrease in glutamatergic vesicular release probability at these synapses. Our data suggest that GABA_B, 5-HT1_B/D, and μ opioid heteroreceptors differentially engage SNAP25. These findings share electrophysiological similarities in that each GPCR system has been shown in early studies to reduce the frequency of action potential-independent quantal synaptic events recorded as miniature EPSCs (mEPSCs) in MSNs of the NAc (Muramatsu et al., 1998; Ma et al., 2012; Manz et al., 2019). Although these data corroborate a presynaptic localization of function, a VGCC-independent control mechanism has been invoked to support an effect observed with mEPSCs. Studies from our group and others suggest that GABA_B, a G_i/o-coupled GPCR well-known for its Gβγ-VGCC interaction (Hamid et al., 2014; Church et al., 2022), inhibits glutamatergic transmission in the NAc independent of N- and P/Q-type VGCCs (Manz et al., 2019; Zurawski et al., 2019). Similar to data obtained here, we report that GABA_B instead engages a SNAP25-dependent effect on vesicular release. It’s of interest that in hippocampal CA1 neurons, GABA_B inhibits glutamate release independent of Gβγ-SNARE interaction, indicating tissue-specific effects (Alten et al., 2022). To our surprise, the μ opioid receptor-induced decrease in glutamatergic transmission was completely abolished in SNAP25Δ3 mice, highlighting a GPCR system entirely contingent on the Gβγ-SNARE interaction.

Despite a pharmacological screen revealing SNAP25-dependent GPCR systems at glutamatergic synapses in the NAc, a broader group of GPCRs were functionally indistinguishable in SNAP25Δ3 and WT controls, including κ opioid, CB1, adenosine A1, histamine H3, and group II metabotropic glutamate receptors. It is tempting to suggest that these data capture the behavior of the selected GPCRs within the NAc. However, glutamatergic transmission was evoked using local field stimulation of heterogenous afferents, each with distinct GPCR profiles, receptor densities, and presynaptic microenvironments. The adenosine A1 receptor, for example, may recruit input-specific effector systems that are inadequately sampled in our experimental configuration (Fritz et al., 2021). Further studies are needed to delineate whether each GPCR system obeys input- or cell-type-specific signaling heuristics in the NAc.

We hypothesized that by employing the biased agonist BnOCPA, which preferentially engages G-protein signaling as opposed to β-arrestin signaling, we would amplify the Gβγ-signaling component provoked by adenosine A1 receptor agonism. Previous studies have shown that in addition to receptor internalization, β-arrestin can promote diverse signaling cascades that can recruit synaptic plasticity (Hall et al., 1999; Lefkowitz and Shenoy, 2005; Dhami and Ferguson, 2006), we therefore aimed to determine whether Gβγ-SNARE signaling may be engaged by A1 receptor but perhaps occluded by simultaneous β-arrestin signaling by using the G-protein signaling biased A1R agonist BnOCPA. We did not observe differences in EPSC amplitude between WT and SNAP25Δ3 when we applied BnOCPA (Figure 4G), providing us with greater confidence that the Gβγ-SNARE signaling interaction is not necessary for adenosine 1 receptor depression of excitatory synaptic transmission in the NAc. G-protein biased agonists are not available for all of the receptors assessed but are a future avenue for exploring whether some of these receptor systems may still engage the Gβγ-SNARE motif for their synaptic depression.

Gβγ subunits have been shown to have different affinities for SNAP25 (Smrcka, 2008; Yim et al., 2021). An intriguing question is whether the Gβγ-SNARE signaling interaction is directed by GPCRs coupled to Gβγ complexes comprised of specific subunits. Indeed, we have evidence of this GPCR specificity for certain Gβγ subunits that then impact which Gβγ subunits bind SNARE (Yim et al., 2019, 2021). Thus, only those Gβγ subunits with sufficient affinity for SNARE can elicit a presynaptic effect on release probability. Although beyond the scope of this study, careful examination of the Gβγ subunits bound to GPCRs surveyed here would offer additional mechanistic insight into the specificity of our findings. An alternative hypothesis is that SNAP25-dependent GPCR signaling is spatially restricted to those GPCR classes positioned at or near the active zone. As an extension of this hypothesis is the possibility of scaffolding factors which are important for determining localization to the active zone. Nanodomain resolution would be required to resolve whether differential localization of these GPCRs predicts the efficacy with which the Gβγ-SNARE interaction can influence synaptic transmission. An important limitation of our studies is the inability to isolate the Gβγ signaling component by these GPCRs from simultaneously activated pathways such as Gα and β-arrestin (DeWire et al., 2007). Therefore, it is possible that some of the GPCR systems we examined may still engage the Gβγ-SNARE interaction to exert their depression, but that the contributions of this interaction may have been occluded by pathways activated in parallel.

Several of these GPCR systems have been the targets of therapeutics for addiction, though activation of these receptors has been shown to have different behavioral outcomes in addiction models. For example, the GABA_B receptor agonist baclofen has been found to reduce drug-craving, drug-seeking, and drug relapse in both humans and rodent models (Hotsenpiller and Wolf, 2003; Kahn et al., 2009), but cannabinoid 1 receptor agonists are able to increase cocaine and heroin relapse (De Vries et al., 2001; Parolaro et al., 2007). Conversely, blocking cannabinoid 1 receptors in the NAc inhibits the reinstatement of drug seeking of heroin (Alvarez-Jaimes et al., 2008), morphine (Yuan et al., 2013), and cocaine (Xi et al., 2006). A deeper understanding of the mechanistic function by these GPCRs may provide important insight for targeting of downstream effector systems for treating addiction. We have successfully identified GPCR systems which, in naïve mice, mediate depression of EPSCs through the SNAP25-Gβγ interaction. Our results provide impetus for the exploration of this effect in additional GPCR systems, inhibitory transmission, brain regions, cell-types, and how these systems are altered in addiction.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The animal study was reviewed and approved by the Vanderbilt University IACUC through Office of Animal Welfare Assurance.

KM, JZ, HH, and BG designed the experiments and interpreted the results. KM and JZ performed the experiments. KM, JZ, and BG wrote the manuscript. All authors contributed to the article and approved the submitted version.