Epac2 in midbrain dopamine neurons contributes to cocaine reinforcement via enhancement of dopamine release
Curation statements for this article:-
Curated by eLife
Evaluation Summary:
This manuscript reports that Epac2, a downstream effector of cAMP, positively regulates cocaine reward by altering dopamine release properties in the striatum. These results provide insight into Epac2 as a potential presynaptic molecular target through which dopamine signaling and drug taking might be manipulated and is of interest to scientists studying dopamine transmission and substance use disorders.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1 agreed to share their name with the authors.)
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
- Neuroscience (eLife)
Abstract
Repeated exposure to drugs of abuse results in an upregulation of cAMP signaling in the mesolimbic dopamine system, a molecular adaptation thought to be critically involved in the development of drug dependence. Exchange protein directly activated by cAMP (Epac2) is a major cAMP effector abundantly expressed in the brain. However, it remains unknown whether Epac2 contributes to cocaine reinforcement. Here, we report that Epac2 in the mesolimbic dopamine system promotes cocaine reinforcement via enhancement of dopamine release. Conditional knockout of Epac2 from midbrain dopamine neurons (Epac2-cKO) and the selective Epac2 inhibitor ESI-05 decreased cocaine self-administration in mice under both fixed-ratio and progressive-ratio reinforcement schedules and across a broad range of cocaine doses. In addition, Epac2-cKO led to reduced evoked dopamine release, whereas Epac2 agonism robustly enhanced dopamine release in the nucleus accumbens in vitro. This mechanism is central to the behavioral effects of Epac2 disruption, as chemogenetic stimulation of ventral tegmental area (VTA) dopamine neurons via deschloroclozapine (DCZ)-induced activation of Gs-DREADD increased dopamine release and reversed the impairment of cocaine self-administration in Epac2-cKO mice. Conversely, chemogenetic inhibition of VTA dopamine neurons with Gi-DREADD reduced dopamine release and cocaine self-administration in wild-type mice. Epac2-mediated enhancement of dopamine release may therefore represent a novel and powerful mechanism that contributes to cocaine reinforcement.
Article activity feed
-
-
Author Response
Reviewer #3 (Public Review):
Liu et al. investigated the role of Epac2, the "other" less studied cAMP effector (compared to the classical PKA) in dopamine release and cocaine reinforcement using slice electrochemistry, behavior, and in vivo imaging in dopamine neuron-specific Epac2 conditional knockout mice (confirmed by elegant single-cell RT-PCR). Epac2 genetic deletion (Epac2 cKO) or pharmacological inhibition (using the Epac2 antagonist ESI-05, i.p.) reduced cocaine (under both fixed and progressive ratio schedules) but not sucrose, self-administration, supporting an essential role for Epac2 in cocaine reinforcement but not natural reward. Cyclic voltammetry on striatal slices demonstrated that evoked DA release was reduced in Epac2 cKO mice and enhanced by the Epac2 activator S-220 or the PKA activator 6-Bnz …
Author Response
Reviewer #3 (Public Review):
Liu et al. investigated the role of Epac2, the "other" less studied cAMP effector (compared to the classical PKA) in dopamine release and cocaine reinforcement using slice electrochemistry, behavior, and in vivo imaging in dopamine neuron-specific Epac2 conditional knockout mice (confirmed by elegant single-cell RT-PCR). Epac2 genetic deletion (Epac2 cKO) or pharmacological inhibition (using the Epac2 antagonist ESI-05, i.p.) reduced cocaine (under both fixed and progressive ratio schedules) but not sucrose, self-administration, supporting an essential role for Epac2 in cocaine reinforcement but not natural reward. Cyclic voltammetry on striatal slices demonstrated that evoked DA release was reduced in Epac2 cKO mice and enhanced by the Epac2 activator S-220 or the PKA activator 6-Bnz independently. Using in vivo chemogenetics and fiber photometry (with the DA fluorescent sensor GRABDA2M), authors showed that DCZ activation of VTA DA neurons expressing rM3D(Gs) increased NAc DA release and cocaine SA in Epac2 cKO mice (rescuing), whereas inhibition of VTA DA neurons expressing hM4D(Gi) decreased DA release and cocaine SA in WT mice (mimicking). Based on these experiments, the authors concluded that Epac2 in midbrain DA neurons contributes to cocaine reinforcement via enhancement of DA release.
The experiments are generally rigorous and the conclusions are mostly well supported by data, but some aspects of behavioral experiments and data analysis need to be clarified or extended.
- The chemogenetic rescue experiments in Fig. 7 suggested that enhancing DA release in Epac2 cKO mice rescued cocaine SA in mutant mice, but did not necessarily demonstrate that Epac2 mediates this process, thus a causal mechanistic link is missing. This is an important point to clarify because the central theme of the work is that Epac2 regulates cocaine SA via DA release. In addition, it's unclear if chemogenetic activation of DA neurons also enhances sucrose reward. A potentially positive result would not affect the conclusion that enhancing DA release can rescue cocaine SA in mutant mice but will affect the interpretation and specificity of the rescue data.
The reviewer’s viewpoint is well taken. We agree that Gs-DREADD activation may restore the Epac2-cKO-induced decrease in dopamine release, but not other deficits caused by Epac2 deletion. We acknowledge the limitations of our DREADD experiments (see our response to Reviewer 2 above). Please also see our response to question 2 below.
In the revised manuscript, we provided representative temporal patterns of FR1 sucrose self-administration in WT and Epac2-cKO mice, which did not display significant differences between genotypes (see newly added Figure 3 – figure supplement 2). To prevent excessive sucrose intake, sessions ended if the maximum number (64) of reinforcers were earned during the 1-hour training session. Almost all wild-type and Epac2-cKO mice had approached this maximum level near the end of the 10-day training. While testing if chemogenetic activation of VTA dopamine neurons enhances sucrose self-administration is, in principle, a good idea, such enhancement would likely lead to a ceiling effect, making the detection of potential differences between genotypes difficult.
- Relatedly, chemogenetic inhibition experiments in Fig 8 showed that inhibiting DA neurons reduced DA release and cocaine SA in WT mice, which suggested that the strength of DA transmission was a regulator of cocaine SA. This is expected given the essential role of DA transmission in reward in general, but it did not provide strong insights regarding the specific roles of Epac2 in the process.
An ideal experiment would be to examine whether viral expression of Epac2 in VTA dopamine neurons in Epac2-cKO mice could restore cocaine self-administration to the level of WT mice. However, our lab is not equipped to do this type of study at its current capacity, but we are very interested in exploring this exciting experiment in the future.
- Fig 7B. DCZ-induced DA releases enhancement in the fiber photometry recording seems to only last for ~30 min, well short of the duration of a cocaine SA session (3 hrs). It's unclear how this transient DA release enhancement could cause the prolonged cocaine SA behavior.
We appreciate the insight from the Reviewer. We have included the time course of dopamine transients following DCZ injection (now Fig. 6B,C). Although the DCZ-induced enhancement of DA transients was most robust during the first 30 min, an enhancement persisted for the duration of fiber photometry recording (1 hour after DCZ injection). In the original study in which DCZ was developed as a DREADD ligand (Nagai et al., 2020), in vivo two-photon imaging of somatosensory cortex neurons that co-expressed Gq-DREADD (hM3Dq) and GCaMP6 revealed that i.p. injection of DCZ led to a rapid increase in GCaMP6 activity in mice that peaked at about 10 min and plateaued for at least 150 min (see Fig. 4 in that paper). Although Gs-DREADDs may respond to DCZ differently, it appears that DCZ induces long-lasting activation of DREADDs expressed in the brain. We have added a brief discussion in the Results section of the revised manuscript (page 12, lines 262-265).
- Fig. 9. working hypothesis: hM4D(Gi) and hM3D(Gs) are shown to inhibit and enhance synaptic vesicle docking, which is not accurate. These DREADDS presumably regulate neuronal excitability, which in turn affects SV release.
We agree with the reviewer and have removed synaptic vesicle docking from the model (now Figure 8).
-
Evaluation Summary:
This manuscript reports that Epac2, a downstream effector of cAMP, positively regulates cocaine reward by altering dopamine release properties in the striatum. These results provide insight into Epac2 as a potential presynaptic molecular target through which dopamine signaling and drug taking might be manipulated and is of interest to scientists studying dopamine transmission and substance use disorders.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1 agreed to share their name with the authors.)
-
Reviewer #1 (Public Review):
The manuscript by Liu et al. outlines the role of exchange protein directly activated by cAMP (Epac2) in dopamine neurons and how this relates to cocaine effects on dopamine release and associated behaviors. Through a series of manipulations, they show that Epac2 expression increases cocaine reinforcement and dopamine release while decreases in Epac2 have the opposite effect. The manuscript is interesting and important, the design is rigorous, and it of broad impact on the field. There are only minor issues with the wording of the operant schedule (I am not sure that it is actually FR1) and some other wording issues (in some places it just states Epac2, rather than denoting these are its effects in dopamine neurons), but overall this is an excellent manuscript.
-
Reviewer #2 (Public Review):
The study used a healthy range of techniques, with multidisciplinary results thoroughly characterizing the cellular and behavioral roles of Epac2 in mice. Despite some minor concerns, the study provides strong evidence, which not only supports the dopamine hypothesis of cocaine reinforcement but also a potentially novel behavioral role of an important signaling protein.
-
Reviewer #3 (Public Review):
Liu et al. investigated the role of Epac2, the "other" less studied cAMP effector (compared to the classical PKA) in dopamine release and cocaine reinforcement using slice electrochemistry, behavior, and in vivo imaging in dopamine neuron-specific Epac2 conditional knockout mice (confirmed by elegant single-cell RT-PCR). Epac2 genetic deletion (Epac2 cKO) or pharmacological inhibition (using the Epac2 antagonist ESI-05, i.p.) reduced cocaine (under both fixed and progressive ratio schedules) but not sucrose, self-administration, supporting an essential role for Epac2 in cocaine reinforcement but not natural reward. Cyclic voltammetry on striatal slices demonstrated that evoked DA release was reduced in Epac2 cKO mice and enhanced by the Epac2 activator S-220 or the PKA activator 6-Bnz independently. Using in …
Reviewer #3 (Public Review):
Liu et al. investigated the role of Epac2, the "other" less studied cAMP effector (compared to the classical PKA) in dopamine release and cocaine reinforcement using slice electrochemistry, behavior, and in vivo imaging in dopamine neuron-specific Epac2 conditional knockout mice (confirmed by elegant single-cell RT-PCR). Epac2 genetic deletion (Epac2 cKO) or pharmacological inhibition (using the Epac2 antagonist ESI-05, i.p.) reduced cocaine (under both fixed and progressive ratio schedules) but not sucrose, self-administration, supporting an essential role for Epac2 in cocaine reinforcement but not natural reward. Cyclic voltammetry on striatal slices demonstrated that evoked DA release was reduced in Epac2 cKO mice and enhanced by the Epac2 activator S-220 or the PKA activator 6-Bnz independently. Using in vivo chemogenetics and fiber photometry (with the DA fluorescent sensor GRABDA2M), authors showed that DCZ activation of VTA DA neurons expressing rM3D(Gs) increased NAc DA release and cocaine SA in Epac2 cKO mice (rescuing), whereas inhibition of VTA DA neurons expressing hM4D(Gi) decreased DA release and cocaine SA in WT mice (mimicking). Based on these experiments, the authors concluded that Epac2 in midbrain DA neurons contributes to cocaine reinforcement via enhancement of DA release.
The experiments are generally rigorous and the conclusions are mostly well supported by data, but some aspects of behavioral experiments and data analysis need to be clarified or extended.
1. The chemogenetic rescue experiments in Fig. 7 suggested that enhancing DA release in Epac2 cKO mice rescued cocaine SA in mutant mice, but did not necessarily demonstrate that Epac2 mediates this process, thus a causal mechanistic link is missing. This is an important point to clarify because the central theme of the work is that Epac2 regulates cocaine SA via DA release. In addition, it's unclear if chemogenetic activation of DA neurons also enhances sucrose reward. A potentially positive result would not affect the conclusion that enhancing DA release can rescue cocaine SA in mutant mice, but will affect the interpretation and specificity of the rescue data.
2. Relatedly, chemogenetic inhibition experiments in Fig 8 showed that inhibiting DA neurons reduced DA release and cocaine SA in WT mice, which suggested that the strength of DA transmission was a regulator of cocaine SA. This is expected given the essential role of DA transmission in reward in general, but it did not provide strong insights regarding the specific roles of Epac2 in the process.
3. Fig 7B. DCZ-induced DA releases enhancement in the fiber photometry recording seems to only last for ~30 min, well short of the duration of a cocaine SA session (3 hrs). It's unclear how this transient DA release enhancement could cause the prolonged cocaine SA behavior.
4. Fig. 9. working hypothesis: hM4D(Gi) and hM3D(Gs) are shown to inhibit and enhance synaptic vesicle docking, which is not accurate. These DREADDS presumably regulate neuronal excitability, which in turn affects SV release. -
