Mesolimbic dopamine projections mediate cue-motivated reward seeking but not reward retrieval in rats
- University of California, Irvine, United States
- University of California, Los Angeles, United States
Figures
Figure 1
Microstructural organization of instrumental behavior.
(A) Hungry rats were trained to perform a self-paced ‘reward seeking’ task, in which pressing a lever was intermittently reinforced with food pellets (RI-60s schedule). Press-contingent food-cup approaches were taken as a measure of attempted ‘reward retrieval’. (B) Probability of food-cup approaches as a function of time surrounding reinforced (purple) and nonreinforced (gray) lever presses. (C) Representative pattern of food-cup approach behavior for an individual rat surrounding reinforced and nonreinforced lever presses. Individual reinforced trials are separately presented across the y-axis aligned at the point at which the lever became activated (i.e., primed for reinforcement). (D, E) Effects of manipulating instrumental reinforcement contingency on the organization of reward-seeking and -retrieval responses. Total lever presses (D) or presses followed by an approach (E) during tests in which lever pressing was intermittently reinforced (RI-60s) either with food pellets and associated cues (Food and Cues) or with pellet dispenser cues but no actual food delivery (Cues Only). Rats were also tested without any reinforcement (No Food or Cues). (F) The proportion of lever presses that were followed by food-cup approach was higher for reinforced presses than for nonreinforced presses, regardless of whether pressing was reinforced with Food and Cues, or Cues Only. Rats also continued to sporadically check the food cup after nonreinforced lever presses, albeit at a much lower level than after reinforced presses.
-
Figure 1—source data 1
This spreadsheet contains the behavioral responses for individual rats in Figure 1.
- https://doi.org/10.7554/eLife.43551.003
Figure 2
DREADD expression in Th:Cre +rats.
(A) Th:Cre+ rats received bilateral injections of AAV-hSyn-DIO-hM4Di-mCherry or AAV-hSyn-DIO-mCherry in the VTA. (B) Representative expression of the mCherry-tagged inhibitory DREADD hM4Di (red) in VTA Th positive neurons (green) of Th:Cre+ rats, as well as in neuronal terminals (C) projecting to the nucleus accumbens (NAc) and medial prefrontal cortex (mPFC). Scale bar is 500 µm.
Figure 3 with 3 supplements
Chemogenetic inhibition of dopamine neurons on Pavlovian to instrumental transfer (PIT) performance.
(A) Experimental design: Following viral vectors injections and recovery, rats received Pavlovian training, during which they learned to associate an auditory cue (CS+) with food pellet delivery. During instrumental conditioning, rats performed the same lever-press task used in Experiment 1. Lever pressing was extinguished (Ext) before rats were submitted to a PIT test, which included separate noncontingent presentations of the CS+ and an unpaired control cue (CS-). (B) Chemogenetic inhibition of VTA dopamine neurons disrupted cue-motivated reward seeking. Total lever presses during PIT trials for rats expressing the inhibitory DREADD hM4Di or mCherry following vehicle (left) or CNO (5 mg/kg, right) treatment prior to test. Presses during pre-CS (gray) and CS periods (red) are plotted separately. (C) PIT expression is specifically impaired in hM4Di expressing Th:Cre+ rats. PIT scores (total presses: CS+ - pre-CS+) show that the CS+ increased lever pressing after vehicle treatment for both groups, but that CNO suppressed this effect in the hM4Di group but not the mCherry group. **p<0.01. (D) The CS+ increased the proportion of lever presses that were followed by a food-cup approach during PIT testing. Inhibiting VTA dopamine neurons did not disrupt expression of this effect. Instead, rats in both groups showed a modest increase in their likelihood of checking the food cup after lever pressing when treated with CNO. (E) Representative organization of the effects of the CS+ and CS- on attempts to seek out and retrieve reward during PIT. Data show lever presses and food-cup approaches (press-contingent or noncontingent) for two control rats (Th:Cre+ rats expressing mCherry and receiving vehicle).
-
Figure 3—source data 1
This spreadsheet contains the behavioral responses for individual rats in Figure 3.
- https://doi.org/10.7554/eLife.43551.009
Figure 3—figure supplement 1
Probability of food-cup approaches as a function of time surrounding individual lever-press responses during PIT testing, plotted separately for CS+ (blue), CS- (red), and pre-CS (yellow) periods.
This analysis was restricted to vehicle tests completed by rats in Experiments 2 and 3A. Shewhart process control chart analyses were used to determine the times when food-cup approach behavior was elevated with respect to constant background rates. This value approximated 2.5 s following each lever press, as shown by the shaded box.
Figure 3—figure supplement 2
Frequency of lever presses that were followed by a food-cup approach during PIT testing by rats expressing the inhibitory DREADD hM4Di or mCherry following vehicle or CNO (5 mg/kg) treatment in Experiment 2.
Data from pre-CS (gray) and CS (blue) periods are plotted separately. Error bars represent ±1 standard error of the estimated marginal means from the corresponding fitted generalized linear mixed-effects model. We found that the CS+ but not the CS- strongly increased the frequency of these press-approach sequences (CS Period * CS Type interaction, t(240) = 7.87, p<0.001), which did not vary as a function of Group or Drug treatment (interactions involving these factors, ps ≥. 18). However, while CNO administration did not disrupt the ability of the CS+ to elicit these press-approach sequences, it did result in a general, hM4Di-independent reduction in the frequency of these sequences (Drug effect, t(240) = −4.52, p<0.001; Drug * Group interaction, t(240) = −1.19, p=0.234).
Figure 3—figure supplement 3
Noncontingent (press-independent) food-cup approaches during PIT testing by rats expressing the inhibitory DREADD hM4Di or mCherry following vehicle or CNO (5 mg/kg) treatment in Experiment 2.
Data from pre-CS (gray) and CS (green) periods are plotted separately. Error bars represent ±1 standard error of the estimated marginal means from the corresponding fitted generalized linear mixed-effects model. This behavior was selectively increased by the CS+ relative to the CS- (CS Period * CS Type interaction, t(240) = 11.72, p<0.001). CNO produced a nonspecific (hM4Di-independent) suppression in this approach response (Drug effect, t(240) = −1.99, p=0.047), which was more pronounced for the mCherry group (Group * Drug * CS Type * CS Period interaction, t(240) = 2.73, p=0.0067).
Figure 4 with 5 supplements
Pathway specific chemogenetic inhibition of dopamine on PIT performance.
(A) Th:Cre+ rats initially received VTA AAV-hSyn-DIO-hM4Di-mCherry injections and were implanted with guide cannulas aimed at the medial prefrontal cortex (mPFC) or nucleus accumbens (NAc) for microinjection of CNO (1 mM) or vehicle to inhibit dopamine terminals at test. (B) Following surgery, rats underwent training and testing for PIT, as described above. We analyzed the microstructural organization of behavior (Lever presses: seeking, and presses followed by a food-cup approach: retrieval) at test. (C) Pathway specific inhibition of dopamine terminals in the NAc but not the mPFC disrupted cue-motivated reward seeking. Total lever presses during PIT trials for rats expressing the inhibitory DREADD hM4Di and receiving CNO or vehicle microinfusions in either the mPFC or NAc prior to test. Presses during pre-CS (gray) and CS periods (red) are plotted separately. (D) PIT expression was specifically impaired following NAc CNO treatment. PIT scores (total presses: CS+ - pre-CS+) show that the CS+ increased lever pressing following vehicle treatment in both groups, but that CNO suppressed this effect when injected into the NAc but not the mPFC. *p<0.05. (E) The CS+ increased the proportion of lever presses that were followed by a food-cup approach during PIT testing. This effect did not significantly vary as a function of drug treatment or group. (F) Scatter plots show the relationship between individual differences in the effect of the CS+ on lever presses that were not followed by food-cup approach in the vehicle condition (PIT Score for presses without approach) and the suppressive effect of CNO on CS+ evoked lever pressing (PIT Score for CNO test - PIT Score for vehicle test). Data points are from individual rats receiving intra-mPFC (left panel) or intra-NAc (right panel) microinjections.
-
Figure 4—source data 1
This spreadsheet contains the behavioral responses for individual rats in Figure 4.
- https://doi.org/10.7554/eLife.43551.016
Figure 4—figure supplement 1
Cannulae placements for Experiment 3A hM4Di expressing rats.
Individual placements in nucleus accumbens (NAc) and medial prefrontal cortex (mPFC).
Figure 4—figure supplement 2
Frequency of lever presses during PIT testing by rats expressing mCherry following microinjection of vehicle (A) or CNO (B) into the mPFC or NAc in Experiment 3B.
Data from pre-CS (gray) and CS (red) periods are plotted separately. Error bars represent ±1 standard error of the estimated marginal means from the corresponding fitted generalized linear mixed-effects model. The CS+ induced a cue-specific increase in lever pressing (CS Type * CS Period interaction, t(176) = 4.51, p<0.001). Rats in the NAc group exhibited a marginally weaker CS+ specific increase in lever pressing than rats in the mPFC group (Site * CS Type * CS Period interaction, t(176) = −1.89, p=0.060). CNO appeared to slightly attenuate CS+ elicited lever pressing, particularly when injected into the mPFC, though this effect was not significant (effect was not significant (Drug * CS Period * CS Type interaction, t(176) = −0.0022, p=0.998; Site * Drug * CS Period * CS Type interaction, t(176) = 1.83, p=0.068). (C) A focused analysis of CS+ elicited lever pressing (PIT Score; ±1 between-subjects SEM) confirmed that CNO injections did not significantly disrupt this effect in either the mPFC (t(5) = 1.63, p=0.165) or NAc (t(5) = 0.33, p=0.753) group. These results indicate that the tendency for intra-NAc CNO injections to disrupt CS+ elicited lever pressing in hM4Di expressing rats in Experiment 3A (see Figure 4, main text) was due to dopamine terminal inhibition and not a nonspecific CNO effect.
Figure 4—figure supplement 3
Frequency of presses that were followed by a food-cup approach during PIT testing by rats expressing the inhibitory DREADD hM4Di following microinjection of CNO or vehicle into the mPFC or NAc in Experiment 3A.
Data from pre-CS (gray) and CS (blue) periods are plotted separately. Error bars represent ±1 standard error of the estimated marginal means from the corresponding fitted generalized linear mixed-effects model. The cue-induced increase in the frequency of these press-approach sequences was specific to the CS+ (CS Type * CS Period interaction, t(240) = 7.34, p<0.001), which did not vary as a function of drug and/or injection site (ps >0.689), nor were there any main effects of these treatment variables (Drug effect, t(240) = 1.05, p=0.294; Site effect, t(240) = −0.26, p=0.798; Drug * Site interaction, t(240) = 1.28, p=0.203).
Figure 4—figure supplement 4
Noncontingent (press-independent) food-cup approaches during PIT testing in rats expressing the inhibitory DREADD hM4Di following microinjection of CNO or vehicle into the mPFC or NAc in Experiment 3A.
Data from pre-CS (gray) and CS (green) periods are plotted separately. Error bars represent ±1 standard error of the estimated marginal means from the corresponding fitted generalized linear mixed-effects model. Noncontingent approach behavior was selectively elicited by the CS+ but not the CS- (CS Period * CS Type interaction, t(240) = 12.98, p<0.001). CNO administration led to a very modest but reliable enhancement in the tendency for the CS+ to increase this behavior over baseline levels (Drug * CS Type * CS Period interaction, t(240) = 2.98, p=0.0032), which did not depend on injection site (Drug * Site * CS Type * CS Period interaction, t(240) = −1.45, p=0.147). CNO did not have any general (cue-independent) effects on this response (Drug effect, t(240) = −0.41, p=0.680; Site effect, t(240) = 0.75, p=0.456; Drug * Site interaction, t(240) = −0.96, p=0.338).
Figure 4—figure supplement 5
Scatter plots show the relationship between individual differences in the effect of the CS+ on lever presses that were not followed by food-cup approach in the vehicle condition (PIT score for presses without approach) and the suppressive effect of CNO on CS+ evoked lever pressing (PIT Score for CNO test - PIT Score for vehicle test).
Data points are from individual rats expressing mCherry (left panel) or hM4Di (right panel).
Figure 5 with 1 supplement
Chemogenetic inhibition of dopamine neurons on reward devaluation performance.
(A) Th:Cre+ rats received VTA injections of AAV-hSyn-DIO-hM4Di-mCherry or AAV-hSyn-DIO-mCherry. Following recovery, rats were trained on two distinct lever-press actions for two different rewards (Instrumental Learning). Rats then underwent reward-specific devaluation testing following treatment with CNO (5 mg/kg) or vehicle. (B) Chemogenetic VTA dopamine inhibition did not alter the impact of reward devaluation on reward seeking. Total lever presses on the valued (red bars) and devalued (gray) levers in hM4Di or mCherry expressing Th:Cre+ rats, following CNO (5 mg/kg) or vehicle treatments. (C) Proportion of valued (blue) and devalued (gray) lever-press actions that were followed by a food-cup approach. Rats were more likely to attempt to retrieve reward after performing the devalued lever-press action. This effect was not altered by VTA dopamine neuron inhibition. (D). Lever presses performed without a subsequent food-cup approach response (red) were more sensitive to reward devaluation than presses that were followed by an approach (blue).
-
Figure 5—source data 1
This spreadsheet contains the behavioral responses for individual rats in Figure 5.
- https://doi.org/10.7554/eLife.43551.019
Figure 5—figure supplement 1
Data from the reinforced phase of reward devaluation testing for rats expressing the inhibitory DREADD hM4Di or mCherry following vehicle or CNO treatment in Experiment 4.
(A) Significantly fewer lever presses were performed on the devalued lever compared to the valued lever, t(148) = −5.55, p<0.001, which did not interact with Group or Drug, ps ≥. 095. (B) Similarly, the frequency of presses that were followed by a food-cup approach was lower on the devalued versus the valued lever, t(148) = −5.46, p<0.001, which also did not depend on Group or Drug, ps ≥0. 128. In A-B, error bars represent ±1 standard error of the estimated marginal means from the corresponding fitted generalized linear mixed-effects model. (C) The proportion of lever presses that were followed by a food-cup approach was significantly higher for the devalued versus the valued lever, t(131) = 4.11, p<0.001, which also did not depend on Group or Drug conditions, ps ≥. 679.
Additional files
-
Supplementary file 1
Generalized linear mixed-effects model outputs.
- https://doi.org/10.7554/eLife.43551.020
-
Transparent reporting form
- https://doi.org/10.7554/eLife.43551.021
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Mesolimbic dopamine projections mediate cue-motivated reward seeking but not reward retrieval in rats
eLife 8:e43551.
https://doi.org/10.7554/eLife.43551
