Aversive Stimuli Drive Drug Seeking in a State of Low Dopamine Tone (2014)

Biol Psychiatry. 2014 Sep 22. pii: S0006-3223(14)00703-3. doi: 10.1016/j.biopsych.2014.09.004. Twining RC1, Wheeler DS2, Ebben AL2, Jacobsen AJ2, Robble MA2, Mantsch JR2, Wheeler RA2.



Stressors negatively impact emotional state and drive drug seeking, in part, by modulating the activity of the mesolimbic dopamine system. Unfortunately, the rapid regulation of dopamine signaling by the aversive stimuli that cause drug seeking is not well characterized. In a series of experiments, we scrutinized the subsecond regulation of dopamine signaling by the aversive stimulus, quinine, and tested its ability to cause cocaine seeking. Additionally, we examined the midbrain regulation of both dopamine signaling and cocaine seeking by the stress-sensitive peptide, corticotropin releasing factor (CRF).


Combining fast-scan cyclic voltammetry with behavioral pharmacology, we examined the effect of intraoral quinine administration on nucleus accumbens dopamine signaling and hedonic expression in 21 male Sprague-Dawley rats. We tested the role of CRF in modulating aversion-induced changes in dopamine concentration and cocaine seeking by bilaterally infusing the CRF antagonist, CP-376395, into the ventral tegmental area (VTA).


We found that quinine rapidly reduced dopamine signaling on two distinct time scales. We determined that CRF acted in the VTA to mediate this reduction on only one of these time scales. Further, we found that the reduction of dopamine tone and quinine-induced cocaine seeking were eliminated by blocking the actions of CRF in the VTA during the experience of the aversive stimulus.


These data demonstrate that stress-induced drug seeking can occur in a terminal environment of low dopamine tone that is dependent on a CRF-induced decrease in midbrain dopamine activity.


Stressful life events are potent modulators of mood and can trigger a variety of destructive behaviors, including drug abuse (1). While addiction is a multifaceted disorder, it has been suggested that aversive life events can promote relapse in addicts by inducing negative affect and craving (2, 3, 4, 5). Likewise, drug-associated stimuli evoke a negative affective state in abstinent cocaine users that is predictive of relapse (2, 4, 6). Ultimately these stimuli are thought to promote a spiral of maladaptive behaviors in which substance abusers, attempting to remain abstinent, are prompted to correct an environmentally induced negative affective state through the resumption of drug use (7, 8, 9, 10, 11).

Aversive events and their attendant emotional states most likely drive drug seeking by impinging upon the mesolimbic dopamine system, but the manner by which they do this is poorly understood. In fact, while the evidence is mounting that negative affect is a critical determinant of the resumption of drug taking following periods of abstinence, the literature is conflicted on the basic question of the directionality of the dopamine response to aversive stimuli (12, 13). Electrophysiological and electrochemical studies that measure dopamine neuron activity and terminal dopamine release, respectively, commensurate with the immediate sensation and perception of aversive stimuli routinely characterize rapid reductions in dopamine signaling in response to aversive stimuli and their predictors (14, 15, 16, 17, 18, 19). This reduction in dopaminergic activity is reportedly induced, in part, by stress-sensitive neuromodulators such as corticotropin-releasing factor (CRF) (20, 21). Unfortunately, electrophysiological recordings of dopamine neurons indicate that neither the aversion-induced decrease in dopamine neuron activity nor the CRF regulation of that response is uniform (22, 23, 24, 25), necessitating an approach that examines rapid terminal signaling in dopamine neuronal projection targets.

Little is known about the nature of rapid, aversion-induced dopamine release patterns in relevant terminal regions. It is unclear how such stimuli could cause reductions in dopamine signaling and how decreased dopamine may promote stress-mediated maladaptive behaviors, like drug seeking. In the nucleus accumbens (NAc), a critical locus of the reward circuit, increases and decreases in dopamine concentration selectively activate D1- and D2-receptor–expressing medium spiny neurons (MSNs), respectively, which have opposing effects on motivated behavior (26, 27). Activation of these distinct circuits has long been known to differentially regulate a diverse array of motivated behaviors, including responses to drugs of abuse (28, 29, 30, 31, 32, 33). Therefore, characterizing whether aversive stimuli increase or decrease NAc dopamine concentration is likely essential to determining how stressful life events activate specific striatal circuitry to cause relapse to drug use. Previously, we observed that cocaine-predictive stimuli can induce a negative affective state, while simultaneously reducing dopamine signaling in the NAc (19). However, the behavioral impact of either of these observations remains to be tested. Critical questions of how aversive stimuli negatively regulate dopamine signaling and whether this mechanism is one that can lead to drug-seeking-like behaviors in rodents must be addressed. In these studies, we scrutinized the precise temporal dynamics of aversion-induced reductions in dopamine signaling, the regulation by stress-induced CRF release into the ventral tegmental area (VTA), and the behavioral impact on hedonic processing and drug seeking. Overall, our findings reveal temporal complexity in dopamine signaling and the ability of CRF to regulate dopamine tone and promote drug seeking

Methods And Materials


Twenty-one male Sprague-Dawley rats (275–300 g; Harlan Laboratories, St. Louis, Missouri) were individually housed in a temperature- and humidity-controlled, Association for Assessment and Accreditation of Laboratory Animal Care accredited vivarium. Rats were maintained on a 12/12-hour reversed cycle (lights off at 7 am) and had ad libitum access (unless otherwise noted) to water and food (Teklad; Harlan Laboratories). All experimental protocols were approved by the Institutional Animal Care and Use Committee at Marquette University in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.


All surgical procedures were conducted under ketamine/xylazine (100 mg/kg/20 mg/kg, intraperitoneal) anesthesia. Intraoral and intrajugular catheter implantations were conducted as previously described (11). Guide cannulas for microinjections (26-gauge; Plastics One, Roanoke, Virginia) were implanted bilaterally immediately above the VTA (anterior-posterior: −5.6; medial-lateral: ±2.2 at 11º angle; dorsal-ventral: −7.0). To prepare for voltammetric recordings, electrode guide cannulas were implanted above the NAc shell unilaterally (anterior-posterior: +1.3; medial-lateral: ±1.3), and a silver/silver chloride reference electrode was placed contralateral to the guide cannula. Additionally, a combined bipolar stimulating electrode/microinjection guide cannula (Plastics One) was placed immediately above the ipsilateral VTA, and a guide cannula was placed above the contralateral VTA. For all surgical procedures, rats were treated with the anti-inflammatory med-cam (1% oral suspension) the day of and for 2 days following the surgery to reduce inflammation and postoperative pain. To maintain patency, the intraoral and intrajugular catheters were flushed daily with distilled water (intraoral) or heparinized saline and the antibiotic cephazolin (intravenous [IV]), respectively.


Microinjectors extended .5 mm from the end of the guide cannula. Artificial cerebrospinal fluid (aCSF) (.3 µL/min) or the selective CRF receptor antagonist CP-376395 (.3 µg/.3 µL/min) was bilaterally injected into the VTA (n = 6 aCSF, n = 6 CP-376395). CP-376395 is a selective CRF-R1 antagonist, but interactions with R2 are likely at this dose. Microinjectors were left in place for 2 minutes after the injection to allow for diffusion. In both procedures, quinine delivery was (re)initiated immediately after the injection.

Voltammetric Recordings

After recovering from surgery, rats were habituated for 2 hours in the voltammetric recording environment, consisting of a clear Plexiglas chamber (Med Associates, St. Albans, Vermont) housed in a custom-designed Faraday cage. The VTA stimulating electrode was harnessed to a rotating commutator (Crist Instrument Co., Hagerstown, Maryland), and one intraoral cannula was harnessed to a fluid swivel (Instech Laboratory, Plymouth Meeting, Pennsylvania) that could receive fluid from a syringe pump (Razel, St. Albans, Vermont). On the following day, voltammetric recordings were conducted as previously described (16). Details of the recording procedure and analysis are described in Supplement 1. Briefly, a carbon fiber electrode was lowered into the NAc shell, a fluid line was attached to the intraoral cannula, and the behavioral session was initiated. The experiment consisted of a 30-minute baseline dopamine monitoring phase (phase 1); a 30-minute quinine delivery period (phase 2); bilateral VTA microinjections; and a 50-minute postinjection quinine delivery period (phase 3). Throughout the quinine delivery phases, a 6-second infusion of .2 mL quinine (.001 mmol/L) was delivered approximately every minute.

Voltammetry Data Analysis

Analyte identification details are described in Supplement 1. Data from each trial (−20 sec before and 30 sec postinfusion onset) were background subtracted using a 1-second block at the local minima in the 20 seconds before infusion onset. For each rat, data were averaged across the quinine infusion trials in the 10 seconds following the initiation of the quinine infusion period (quinine) compared with the previous 10-second period (prequinine) and the next 10-second period (postquinine). The resultant current changes over time were analyzed for dopamine changes using principle component regression. For all rats (n = 12), reductions in naturally occurring (non-time-locked) dopamine tone were quantified and analyzed by comparing the first 5 trials (early) with trials 11 to 15 (middle) and the last 5 trials (late) in the prequinine period, 10 seconds before quinine infusion, using a repeated measures analysis of variance (ANOVA). Significant changes in dopamine concentration over time, time-locked to the quinine infusion, were evaluated using two within-subjects repeated measures ANOVAs varying phase (baseline, quinine, and quinine + drug [aCSF or CP-376395]) × period (prequinine, quinine, postquinine). When significant main or interactive effects were detected, all pairwise comparisons were made with Tukey’s post hoc tests for multiple comparisons with alpha set at .05.

Dopamine release events occurred independent of any applied stimuli or experimenter controlled behavioral action in the baseline period. To determine how aversive stimuli affected the likelihood of high concentration dopamine release events, every 100-msec sample on every trial for each rat was time-stamped if its concentration was 40 nmol/L or higher. This threshold is within the range of affinities for high-affinity D1 receptors and is the approximate average value of spontaneous dopamine release events (34, 35). From this characterization, transient frequency and amplitude were quantified and analyzed. A two-way ANOVA was used to identify main effects of period (quinine versus postquinine) and drug (aCSF vs. CP-376395). Tukey’s post hoc tests for multiple comparisons were used to identify significant differences within period and drug. In all cases, the alpha level for significance was .05. Statistical comparisons were made using commercially available software (Statistica; StatSoft, Tulsa, Oklahoma).

Taste Reactivity Data Analysis

Taste reactivity was analyzed in a frame-by-frame analysis using digital video recorded on the test day in aCSF- and CP-376395-injected rats (n = 5 in each group). Appetitive and aversive taste reactivity was counted in the prequinine and quinine periods using the technique of Grill and Norgren (36). Mouth movements that matched the triangle shape for a duration exceeding 90 msec were counted as aversive. These criteria excluded all neutral and ingestive mouth movements, which were counted separately. Instances in which the tongue protruded and crossed the midline were counted as appetitive. The remaining licking behavior was counted as neutral licking. Statistical analyses of all behavioral data were performed using commercially available software (Statistica).

Self-Administration and Reinstatement

Mildly food-restricted rats (15–18 g/day) were trained to press a lever for sucrose pellets. Upon acquisition of lever pressing (~3–5 days), intraoral and intravenous catheters were implanted as described above. After recovery, rats were food restricted again and trained to self-administer cocaine (.3 mg/.2 mL/infusion, IV) on a fixed-ratio 1 schedule in computer-interfaced operant conditioning chambers enclosed in sound-attenuating cubicles (Med Associates). When the cocaine session began, a house light illuminated the chamber, and a cue light located above the active lever signaled cocaine availability. Each cocaine infusion was accompanied by turning off the house light and cue light, and a time-out period lasting 20 sec, during which the lever remained extended and responses were recorded but yielded no reinforcement. Responding on a second inactive lever was also recorded. After the time-out period, the house light and cue light were turned on and signaled cocaine availability. Self-administration sessions occurred in a series of four experimenter-controlled 6-day cycles consisting of 3 days of cocaine self-administration and 3 days without cocaine in the home cage. After the third cycle, all rats received VTA cannulation surgery and began their fourth cycle after 2 weeks of recovery. Each daily cocaine session ended when rats achieved a fixed maximum number of cocaine infusions (25 infusions for the first 9 days of access before VTA cannulation and 30 infusions for the last three cocaine sessions following VTA cannulation). Extinction consisted of daily 2-hour sessions during which each lever press resulted in a saline infusion but no cue light signaling or cocaine delivery. Once the extinction criterion was met (<15 active lever responses for the terminal 2-day average; Table S1 in Supplement 1), each rat was tested for quinine-induced reinstatement. To prevent the potential confound of spontaneous recovery, reinstatement testing was conducted for each animal the day after extinction criteria were met. Before each reinstatement session, rats received intra-VTA microinjections of aCSF (n = 4) or CP-376395 (n = 5). Reinstatement sessions began with 15 intraoral infusions of quinine delivered in the cocaine self-administration chamber in the same manner as in the previous experiment for 15 minutes. Five minutes after quinine delivery, the levers were extended and responses were recorded for 1 hour.

Reinstatement Data Analysis

Changes in lever pressing behavior in the first hour of each session were analyzed using a two-way ANOVA varying the between-subjects factor of drug (aCSF, CP-376395) × the within-subjects factor of day (extinction, reinstatement, posttest). Extinction responding was defined as the last day of extinction training, and posttest responding was a final session tested under extinction conditions without quinine administration. Significant differences in drug-seeking behavior were identified when appropriate by Tukey’s post hoc tests for multiple comparisons with alpha set at .05.


After the completion of experimental procedures, all subjects were euthanized with carbon dioxide. To verify placement of recording electrodes, small electrolytic lesions were created by running a current (250 µA) through a stainless steel electrode placed at the depth at which the recording took place. Brains were then removed and submerged in 10% formaldehyde for 14 days. They were then sliced into 40-µm sections, mounted, stained with .25% thionin, and coverslipped. Depictions of the cannula and electrode placements from the voltammetry and reinstatement experiments are presented in Figures S1 and S2 in Supplement 1, respectively (37).


To probe the temporal dynamics of dopamine reductions triggered by aversive events, we employed fast-scan cyclic voltammetry in freely moving rats exposed to brief intraoral infusions of the unpalatable, bitter taste of quinine. This design allows for the simultaneous monitoring of an animal’s affective response coincident with the assessment of terminal dopamine release in the NAc on a subsecond time frame (16, 19). As expected, across the 30-minute test session (1 infusion/min), quinine exposure evoked the expression of aversive taste reactivity time-locked to reductions in terminal dopamine release events (Figure 1B; phase 2, prequinine compared with quinine/postquinine periods, left and right; Figure S3 in Supplement 1). Intriguingly, dopamine reductions displayed two discrete temporal signatures: an immediately apparent, transient drop during each exposure to quinine, as well as a longer-lasting reduction in naturally occurring dopamine tone that emerged only after repeated exposure to quinine. This latter effect was quantified as a significant reduction in the middle (trials 11–15) and late (trials 26–30) trials, compared with the early (first 5) trials in the prequinine period 10 seconds before quinine infusion (Figure 1B, right). These data confirm the ability of aversive stimuli to lower terminal dopamine concentration and reveal a temporal complexity to this response.

Thumbnail image of Figure 1. Opens large image    

Figure 1

Corticotropin-releasing factor regulation of dopamine signaling during the experience of an unavoidable aversive stimulus. (A) Representative fluctuations in naturally occurring dopamine concentration in the shell of the nucleus accumbens in a behaving control (left) and experimental (right) rat in the baseline phase (phase 1). (B) Altered dopamine signaling in response to the intraoral administration of quinine (phase 2). Reductions can be observed both acutely in response to quinine (x axis) and also broadly across trials (y axis, prequinine [Pre Q] period). (B) (far right) Intraoral delivery of quinine reduced tonic dopamine concentration measured across trials in the prequinine period of phase 2 (analysis of variance main effect: trials F2,22 = 11.73, p < .01; Tukey’s post hoc, *p < .05, significant reduction in middle and late trials compared with early trials). (C) Quinine-induced reductions in dopamine signaling were attenuated by intraventral tegmental area injections of the corticotropin-releasing factor antagonist, CP-376395 (phase 3). aCSF, artificial cerebrospinal fluid; DA, dopamine; Post Q, postquinine.

We next asked whether this aversion-induced drop in terminal dopamine is influenced by CRF signaling in the VTA (21). In phase 3, animals received intra-VTA microinjections of the CRF antagonist CP-376395 (.3 µg/.3 µL/min) or aCSF (.3 µL/min), while intraoral quinine delivery and fast-scan cyclic voltammetry recordings continued (Figure 1C, right; Figure S1 in Supplement 1). CRF antagonism in the VTA had no effect on the ability of quinine to cause a rapid, transient decrease in dopamine concentration during the quinine intraoral infusion period (Figure 1C). In contrast, CRF antagonism in the VTA abolished the inhibitory effect of quinine on non-time-locked dopamine tone during the prequinine and postquinine periods (Figure 1C). By averaging across trials, a time-averaged dopamine concentration can be visualized (Figure 2A,B), along with the acute reduction that results from quinine infusion. An attenuation of this response can be visualized following CRF antagonism (Figure 2D) and quantified following chemometric analysis (Figure 3A,B).

Thumbnail image of Figure 2. Opens large image    

Figure 2

Time-averaged dopamine concentration change during quinine infusion and following corticotropin-releasing factor receptor blockade. Two-dimensional color representations of cyclic voltammetric data collected for 50 seconds around quinine infusions, averaged across trials for each phase of the experiment. The ordinate is the applied voltage (Eapp) and the abscissa is time (seconds [s]). Changes in current at the carbon fiber electrode are indicated in color. In phase 2, quinine infusion reduced the time-averaged dopamine concentration in control (A) and experimental (B) animals. (C) This reduction persisted in rats that received artificial cerebrospinal fluid (aCSF) infusions bilaterally into the ventral tegmental area. (D) Bilateral infusions of the corticotropin-releasing factor antagonist CP-376395 attenuated this reduction. Vertical dashed lines indicate time points in which cyclic voltammograms are plotted to illustrate the presence of dopamine (left), its reduction by quinine (center), and the pH change following intraoral infusion (right).

Thumbnail image of Figure 3. Opens large image    

Figure 3

Intraoral delivery of the aversive taste, quinine, reduced dopamine concentration in a corticotropin-releasing factor dependent manner. Changes in dopamine (DA) concentration, determined via principal component analysis, are plotted in (A) and (B). (A) Quinine reduced dopamine concentration significantly from baseline (phase 1) in artificial cerebrospinal fluid (aCSF)-injected rats (analysis of variance period × drug interaction; F4,20 = 10.683, p < .001; Tukey’s post hoc, *p < .05). (B) The quinine-induced dopamine reduction was attenuated in CP-376395-injected rats (analysis of variance period × drug interaction; F4,20 = 6.77, p < .01; Tukey’s post hoc, *p < .05, significant reduction in CP-376395-treated animals only in quinine period). The dopamine reduction was reversed by intraventral tegmental area injections of CP-376395 but only in the prequinine (Pre-Q) and postquinine (Post-Q) periods in which quinine was not present. Data are presented as mean + SEM.

Changes in terminal dopamine concentration in behaving animals could be driven by alterations in either the frequency or amplitude of dopamine release events (38). Here, we observed that quinine reduced dopamine tone by selectively reducing release frequency, and this effect was reversed by blocking CRF receptors in the VTA (Figure 4A). Combined, these data indicate that upon aversive stimulation, CRF signaling in the VTA suppresses dopamine tone in the NAc by modulating the frequency of dopamine release events.

Thumbnail image of Figure 4. Opens large image    

Figure 4

Quinine reduced the frequency of dopamine release events. (A) The aversive quinine stimulus reduced dopamine transient frequency in the period following the intraoral infusion, and this effect was reversed by the corticotropin-releasing factor antagonist [artificial cerebrospinal fluid (aCSF) baseline compared with aCSF postquinine (F1,10 = 10.21, Tukey post hoc, *p < .05]. (B) The quinine infusion had no effect on release amplitude during this same period (F1,10 = .75, p > .05). Data are presented as mean + SEM.

Aversive stimuli potently regulate not only affective state but also the maladaptive motivated behavior of drug seeking (3, 9, 39, 40), which is intimately tied to midbrain dopamine signaling (41, 42) and regulated by CRF (43, 44, 45, 46). We therefore tested whether quinine exposure and its attendant drop in NAc dopamine are sufficient to drive drug seeking in a reinstatement paradigm. Rats were trained to press a lever for an IV cocaine infusion. After a period of stable self-administration, the lever-pressing behavior was extinguished by discontinuing cocaine availability. After extinction, rats received inescapable intraoral quinine infusions (1 infusion/min for 15 minutes) followed by the opportunity to press the lever that previously provided cocaine. Quinine administration increased lever pressing only on the active lever (Figure 5A; Figure S4 in Supplement 1), demonstrating that an aversive stimulus that suppresses dopamine tone can also reinstate drug-seeking behavior. Moreover, reinstatement behavior was fully prevented by blocking CRF receptors in the VTA (Figure 5A; Figure S2 in Supplement 1). Intriguingly, although CRF antagonism blocked reinstatement behavior, it spared the perceived aversive properties of quinine, as indicated by the persistent expression of aversive taste reactivity (Figure 5B). Taken together, these data demonstrate that an aversive stimulus that suppresses dopamine tone can also reinstate drug-seeking behavior and both of these responses are preventable by blocking CRF receptors in the VTA.

Thumbnail image of Figure 5. Opens large image    

Figure 5

Motivational processes, but not hedonic expression, were regulated by corticotropin-releasing factor. (A) Following extinction (Ext), intraoral infusions of quinine caused cocaine seeking in the reinstatement test (Rein) in artificial cerebrospinal fluid (aCSF)-treated rats, an effect that was reversed by intraventral tegmental area injections of CP-376395 (analysis of variance drug × day interaction; F2,16 = 5.83, p < .05; Tukey’s post hoc, *p < .05). (B) Intraoral delivery of the aversive taste, quinine, caused the expression of aversive taste reactivity in aCSF-treated rats. This effect was not altered by intraventral tegmental area injections of CP-376395 (t1,9 = .98, p > .05). Data are presented as mean + SEM. Post, postquinine.


This report highlights a mechanism by which aversive stimuli can drive the motivational circuit to induce drug-seeking behavior. Intraoral infusions of the aversive tastant, quinine, caused both phasic and tonic reductions in terminal dopamine signaling (i.e., reductions across seconds and across minutes). Previous reports have described rapid, phasic, aversion-induced decreases in dopamine release (16). However, in the absence of direct aversive stimulation, the transient release events that comprise a tonic signal also can be time-averaged over minutes (47). Using this approach, we found that cocaine-predictive stimuli can induce negative affect while simultaneously reducing both phasic and tonic dopamine signaling in the NAc (19). Interestingly, we observed that the tonic, but not the phasic, reduction was reversed by blocking CRF receptors in the VTA. While this manipulation did not affect the perceived aversive properties of quinine, it did reverse quinine-induced cocaine seeking, demonstrating that aversive stimuli can drive drug seeking in a state of low dopamine tone. Additional studies will be necessary to characterize both the mechanism and potential behavioral significance of the phasic decrease in dopamine in response to aversive stimuli.

These findings underscore the need to scrutinize the apparently complex manner by which aversive stimuli act on reward circuitry to motivate behavior, elevating dopamine signaling in some situations and decreasing dopamine signaling in others. For example, aversive electric footshock has been shown to increase CRF activity in the VTA (45, 46), which, in turn, can increase dopamine neuron activity (25, 44, 45, 48) and reinstate drug seeking (43, 45, 46). While these findings appear to be at odds with the current report, they are consistent with terminal measures of dopamine signaling using microdialysis that typically report elevations in dopamine concentration for several minutes during and after aversive stimulation that promotes dug seeking (49, 50, 51, 52). The current data are provocative because they demonstrate that aversive stimuli that decrease dopamine signaling can also drive drug seeking and that both phenomena are under the control of CRF.

One possible explanation for how increases and decreases in dopamine signaling could both lead to drug seeking can be found in the cellular organization of dopamine target regions.

Phenotypically distinct striatal neuron populations are tuned to be differentially sensitive to either increases or decreases in dopamine concentration.

  • In the dorsal striatum, low-affinity D1-receptor-expressing MSNs that comprise the direct motor output pathway are activated by elevations of dopamine that promote voluntary movement.
  • Correspondingly, high-affinity D2-receptor-expressing MSNs, comprising the indirect motor output pathway, are inhibited by high dopamine tone but are sensitive to, and activated by, phasic pauses in dopamine that suppress behavior [for review, see (26)].

There is mounting evidence that this organization is paralleled to a significant degree in the ventral striatum. In the NAc, phasic increases in dopamine signaling activate low-affinity dopamine receptor-expressing MSNs that promote reward learning.

Conversely, decreases in dopamine signaling activate high-affinity dopamine receptor-expressing MSNs and promote aversion (27, 30, 53).

In the current studies, quinine likely engaged the latter circuit, serving as an aversive environmental stressor that decreased dopamine signaling in a CRF-dependent manner and promoted drug seeking. Other ethologically relevant environmental stimuli may increase dopamine signaling in the NAc and could lead to the same behavioral result by engaging different circuitries.

NAc dopamine signaling has been heavily implicated in the mechanisms that promote addiction. NAc dopamine is essential for reward-related learning (54) and the proper responses to incentive cues (55), supporting the idea that dopamine signaling provides incentive for or stamps in the motivational value of reinforcing stimuli (42) and contributes significantly toward compulsive drug seeking (41, 56).

Accepting this, it may be intuitive to imagine dopamine-elevating stimuli causing drug seeking but less so to imagine how aversive stimuli that decrease dopamine signaling accomplish this.

However, some of the earliest theories of substance abuse suggested that drug withdrawal acts through negative reinforcement mechanisms to promote relapse in substance abusers attempting to remain abstinent (57, 58, 59). Although subsequent tests of these theories questioned whether acute withdrawal could contribute to a disorder characterized by chronic relapse following extended periods of drug abstinence (60, 61, 62, 63), negative reinforcement mechanisms clearly have a role.

The host of neuroadaptations that accompany chronic drug use and promote reward insensitivity and tolerance (7, 56, 64, 65, 66) could make the sensitivity to environmental stressors an even more important factor in promoting relapse in drug-abstinent populations.

In fact, even moment-to-moment cocaine self-administration appears to involve negative reinforcement learning mediated by striatal dopamine signaling. As dopamine concentration falls, self-administration reliably resumes and animals titrate cocaine intake to maintain the desired brain dopamine concentration (67, 68). During self-administration, animals might learn to respond to avoid a state of lowered dopamine, and the product of this negative reinforcement learning could be subsequently engaged by aversive stimuli that lower dopamine tone. The current report identifies a potential dopaminergic mechanism of this aversive motivation that involves CRF, a stress-activated neuromodulator.

Aversive stimuli reduce dopamine signaling, engaging this mechanism and its attendant emotional states (e.g., negative affect or craving) in the absence of drug availability. Indeed, the current findings suggest that substance abusers learn to correct stress-induced decreases in dopamine signaling in the most efficient manner by self-administering cocaine.

Acknowledgments And Disclosures

This work was funded by the US National Institutes of Health ( R01-DA015758 , JRM, and R00-DA025679 , RAW).

We thank M. Gilmartin, M. Blackmore, P. Gasser, J. Evans, and E. Hebron for contributions to the manuscript.

The authors declare no biomedical financial interests or potential conflicts of interest.

Appendix A. Supplementary Materials


  1. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 5th ed. American Psychiatric Association, Arlington, VA; 2013
  2. Paliwal, P., Hyman, S.M., and Sinha, R. Craving predicts time to cocaine relapse: Further validation of the Now and Brief versions of the cocaine craving questionnaire. Drug Alcohol Depend. 2008; 93: 252–259
  3. View in Article
  4. | Abstract
  5. | Full Text
  6. | Full Text PDF
  7. | PubMed
  8. | Scopus (56)
  9. View in Article
  10. | CrossRef
  11. | PubMed
  12. | Scopus (240)
  13. View in Article
  14. | CrossRef
  15. | PubMed
  16. | Scopus (247)
  17. View in Article
  18. | CrossRef
  19. | PubMed
  20. | Scopus (176)
  21. View in Article
  22. | Abstract
  23. | Full Text
  24. | Full Text PDF
  25. | PubMed
  26. | Scopus (29)
  27. View in Article
  28. | CrossRef
  29. | PubMed
  30. | Scopus (389)
  31. View in Article
  32. | CrossRef
  33. | PubMed
  34. | Scopus (1)
  35. View in Article
  36. | CrossRef
  37. | PubMed
  38. | Scopus (521)
  39. View in Article
  40. | CrossRef
  41. | PubMed
  42. | Scopus (62)
  43. View in Article
  44. | Abstract
  45. | Full Text
  46. | Full Text PDF
  47. | PubMed
  48. | Scopus (67)
  49. View in Article
  50. | CrossRef
  51. | PubMed
  52. | Scopus (50)
  53. View in Article
  54. | CrossRef
  55. | PubMed
  56. | Scopus (28)
  57. View in Article
  58. | CrossRef
  59. | PubMed
  60. | Scopus (452)
  61. View in Article
  62. | CrossRef
  63. | PubMed
  64. | Scopus (344)
  65. View in Article
  66. | CrossRef
  67. | PubMed
  68. | Scopus (116)
  69. View in Article
  70. | CrossRef
  71. | PubMed
  72. | Scopus (28)
  73. View in Article
  74. | CrossRef
  75. | PubMed
  76. | Scopus (26)
  77. View in Article
  78. | Abstract
  79. | Full Text
  80. | Full Text PDF
  81. | PubMed
  82. | Scopus (35)
  83. View in Article
  84. | CrossRef
  85. | PubMed
  86. | Scopus (23)
  87. View in Article
  88. | CrossRef
  89. | PubMed
  90. | Scopus (9)
  91. View in Article
  92. | CrossRef
  93. | PubMed
  94. | Scopus (215)
  95. View in Article
  96. | CrossRef
  97. | PubMed
  98. | Scopus (331)
  99. View in Article
  100. | CrossRef
  101. | PubMed
  102. | Scopus (67)
  103. View in Article
  104. | CrossRef
  105. | PubMed
  106. | Scopus (92)
  107. View in Article
  108. | Abstract
  109. | Full Text
  110. | Full Text PDF
  111. | PubMed
  112. | Scopus (341)
  113. View in Article
  114. | CrossRef
  115. | PubMed
  116. | Scopus (3)
  117. View in Article
  118. | CrossRef
  119. | PubMed
  120. | Scopus (146)
  121. View in Article
  122. | CrossRef
  123. | PubMed
  124. | Scopus (99)
  125. View in Article
  126. | CrossRef
  127. | PubMed
  128. | Scopus (188)
  129. View in Article
  130. | CrossRef
  131. | PubMed
  132. | Scopus (41)
  133. View in Article
  134. | CrossRef
  135. | PubMed
  136. View in Article
  137. | CrossRef
  138. | PubMed
  139. | Scopus (3)
  140. View in Article
  141. | CrossRef
  142. | PubMed
  143. | Scopus (187)
  144. View in Article
  145. | Abstract
  146. | Full Text
  147. | Full Text PDF
  148. | PubMed
  149. | Scopus (80)
  150. View in Article
  151. | CrossRef
  152. | PubMed
  153. | Scopus (522)
  154. View in Article
  155. View in Article
  156. | CrossRef
  157. | PubMed
  158. | Scopus (80)
  159. View in Article
  160. | CrossRef
  161. | PubMed
  162. | Scopus (1293)
  163. View in Article
  164. | CrossRef
  165. | PubMed
  166. | Scopus (777)
  167. View in Article
  168. | CrossRef
  169. | PubMed
  170. | Scopus (497)
  171. View in Article
  172. | CrossRef
  173. | PubMed
  174. View in Article
  175. | CrossRef
  176. | PubMed
  177. | Scopus (25)
  178. View in Article
  179. | CrossRef
  180. | PubMed
  181. | Scopus (32)
  182. View in Article
  183. | CrossRef
  184. | PubMed
  185. | Scopus (172)
  186. View in Article
  187. | CrossRef
  188. | PubMed
  189. | Scopus (93)
  190. View in Article
  191. | CrossRef
  192. | PubMed
  193. | Scopus (42)
  194. View in Article
  195. | Abstract
  196. | Full Text
  197. | Full Text PDF
  198. | PubMed
  199. | Scopus (150)
  200. View in Article
  201. | PubMed
  202. View in Article
  203. | CrossRef
  204. | PubMed
  205. View in Article
  206. | CrossRef
  207. | PubMed
  208. View in Article
  209. | CrossRef
  210. | PubMed
  211. | Scopus (280)
  212. View in Article
  213. | CrossRef
  214. | PubMed
  215. | Scopus (12)
  216. View in Article
  217. | PubMed
  218. View in Article
  219. | CrossRef
  220. | PubMed
  221. | Scopus (1727)
  222. View in Article
  223. | CrossRef
  224. | PubMed
  225. | Scopus (5)
  226. View in Article
  227. | CrossRef
  228. | PubMed
  229. View in Article
  230. | CrossRef
  231. | PubMed
  232. | Scopus (605)
  233. View in Article
  234. | PubMed
  235. View in Article
  236. | CrossRef
  237. | PubMed
  238. | Scopus (3491)
  239. View in Article
  240. | CrossRef
  241. | PubMed
  242. | Scopus (78)
  243. View in Article
  244. | CrossRef
  245. | PubMed
  246. | Scopus (665)
  247. View in Article
  248. | CrossRef
  249. | PubMed
  250. | Scopus (90)
  251. View in Article
  252. | CrossRef
  253. | PubMed
  254. | Scopus (120)
  255. View in Article
  256. | CrossRef
  257. | PubMed
  258. | Scopus (7)
  259. View in Article
  260. | CrossRef
  261. | PubMed
  262. | Scopus (10)
  263. View in Article
  264. | CrossRef
  265. | PubMed
  266. | Scopus (77)
  267. View in Article
  268. | CrossRef
  269. | PubMed
  270. | Scopus (262)
  271. Sinha, R., Catapano, D., and O’Malley, S. Stress-induced craving and stress response in cocaine dependent individuals. Psychopharmacology (Berl). 1999; 142: 343–351
  272. Sinha, R., Fuse, T., Aubin, L.R., and O’Malley, S.S. Psychological stress, drug-related cues and cocaine craving. Psychopharmacology (Berl). 2000; 152: 140–148
  273. Sinha, R., Talih, M., Malison, R., Cooney, N., Anderson, G.M., and Kreek, M.J. Hypothalamic-pituitary-adrenal axis and sympatho-adreno-medullary responses during stress-induced and drug cue-induced cocaine craving states. Psychopharmacology (Berl). 2003; 170: 62–72
  274. Robbins, S.J., Ehrman, R.N., Childress, A.R., Cornish, J.W., and O’Brien, C.P. Mood state and recent cocaine use are not associated with levels of cocaine cue reactivity. Drug Alcohol Depend. 2000; 59: 33–42
  275. Koob, G.F. and Le Moal, M. Addiction and the brain antireward system. Annu Rev Psychol. 2008; 59: 29–53
  276. Mantsch, J.R., Vranjkovic, O., Twining, R.C., Gasser, P.J., McReynolds, J.R., and Blacktop, J.M. Neurobiological mechanisms that contribute to stress-related cocaine use. Neuropharmacology. 2014; 76: 383–394
  277. Baker, T.B., Piper, M.E., McCarthy, D.E., Majeskie, M.R., and Fiore, M.C. Addiction motivation reformulated: An affective processing model of negative reinforcement. Psychol Rev. 2004; 111: 33–51
  278. Fox, H.C., Hong, K.I., Siedlarz, K., and Sinha, R. Enhanced sensitivity to stress and drug/alcohol craving in abstinent cocaine-dependent individuals compared to social drinkers. Neuropsychopharmacology. 2008; 33: 796–805
  279. Wheeler, R.A., Twining, R.C., Jones, J.L., Slater, J.M., Grigson, P.S., and Carelli, R.M. Behavioral and electrophysiological indices of negative affect predict cocaine self-administration. Neuron. 2008; 57: 774–785
  280. Cabib, S. and Puglisi-Allegra, S. The mesoaccumbens dopamine in coping with stress. Neurosci Biobehav Rev. 2012; 36: 79–89
  281. McCutcheon, J.E., Ebner, S.R., Loriaux, A.L., and Roitman, M.F. Encoding of aversion by dopamine and the nucleus accumbens. Front Neurosci. 2012; 6: 137
  282. Mirenowicz, J. and Schultz, W. Preferential activation of midbrain dopamine neurons by appetitive rather than aversive stimuli. Nature. 1996; 379: 449–451
  283. Ungless, M.A., Magill, P.J., and Bolam, J.P. Uniform inhibition of dopamine neurons in the ventral tegmental area by aversive stimuli. Science. 2004; 303: 2040–2042
  284. Roitman, M.F., Wheeler, R.A., Wightman, R.M., and Carelli, R.M. Real-time chemical responses in the nucleus accumbens differentiate rewarding and aversive stimuli. Nat Neurosci. 2008; 11: 1376–1377
  285. Oleson, E.B., Gentry, R.N., Chioma, V.C., and Cheer, J.F. Subsecond dopamine release in the nucleus accumbens predicts conditioned punishment and its successful avoidance. J Neurosci. 2012; 32: 14804–14808
  286. Badrinarayan, A., Wescott, S.A., Vander Weele, C.M., Saunders, B.T., Couturier, B.E., Maren, S., and Aragona, B.J. Aversive stimuli differentially modulate real-time dopamine transmission dynamics within the nucleus accumbens core and shell. J Neurosci. 2012; 32: 15779–15790
  287. Wheeler, R.A., Aragona, B.J., Fuhrmann, K.A., Jones, J.L., Day, J.J., Cacciapaglia, F. et al. Cocaine cues drive opposing context-dependent shifts in reward processing and emotional state. Biol Psychiatry. 2011; 69: 1067–1074
  288. Beckstead, M.J., Gantz, S.C., Ford, C.P., Stenzel-Poore, M.P., Phillips, P.E., Mark, G.P., and Williams, J.T. CRF enhancement of GIRK channel-mediated transmission in dopamine neurons. Neuropsychopharmacology. 2009; 34: 1926–1935
  289. Wanat, M.J., Bonci, A., and Phillips, P.E. CRF acts in the midbrain to attenuate accumbens dopamine release to rewards but not their predictors. Nat Neurosci. 2013; 16: 383–385
  290. Brischoux, F., Chakraborty, S., Brierley, D.I., and Ungless, M.A. Phasic excitation of dopamine neurons in ventral VTA by noxious stimuli. Proc Natl Acad Sci U S A. 2009; 106: 4894–4899
  291. Matsumoto, M. and Hikosaka, O. Two types of dopamine neuron distinctly convey positive and negative motivational signals. Nature. 2009; 459: 837–841
  292. Anstrom, K.K., Miczek, K.A., and Budygin, E.A. Increased phasic dopamine signaling in the mesolimbic pathway during social defeat in rats. Neuroscience. 2009; 161: 3–12
  293. Wanat, M.J., Hopf, F.W., Stuber, G.D., Phillips, P.E., and Bonci, A. Corticotropin-releasing factor increases mouse ventral tegmental area dopamine neuron firing through a protein kinase C-dependent enhancement of Ih. J Physiol. 2008; 586: 2157–2170
  294. Bromberg-Martin, E.S., Matsumoto, M., and Hikosaka, O. Dopamine in motivational control: Rewarding, aversive, and alerting. Neuron. 2010; 68: 815–834
  295. Danjo, T., Yoshimi, K., Funabiki, K., Yawata, S., and Nakanishi, S. Aversive behavior induced by optogenetic inactivation of ventral tegmental area dopamine neurons is mediated by dopamine D2 receptors in the nucleus accumbens. Proc Natl Acad Sci U S A. 2014; 111: 6455–6460
  296. Aragona, B.J., Liu, Y., Yu, Y.J., Curtis, J.T., Detwiler, J.M., Insel, T.R., and Wang, Z. Nucleus accumbens dopamine differentially mediates the formation and maintenance of monogamous pair bonds. Nat Neurosci. 2006; 9: 133–139
  297. Ferguson, S.M., Eskenazi, D., Ishikawa, M., Wanat, M.J., Phillips, P.E., Dong, Y. et al. Transient neuronal inhibition reveals opposing roles of indirect and direct pathways in sensitization. Nat Neurosci. 2011; 14: 22–24
  298. Lobo, M.K., Covington, H.E. 3rd, Chaudhury, D., Friedman, A.K., Sun, H., Damez-Werno, D. et al. Cell type-specific loss of BDNF signaling mimics optogenetic control of cocaine reward. Science. 2010; 330: 385–390
  299. Lobo, M.K. and Nestler, E.J. The striatal balancing act in drug addiction: Distinct roles of direct and indirect pathway medium spiny neurons. Front Neuroanat. 2011; 5: 41
  300. Self, D.W. and Stein, L. Receptor subtypes in opioid and stimulant reward. Pharmacol Toxicol. 1992; 70: 87–94
  301. Porter-Stransky, K.A., Seiler, J.L., Day, J.J., and Aragona, B.J. Development of behavioral preferences for the optimal choice following unexpected reward omission is mediated by a reduction of D2-like receptor tone in the nucleus accumbens. Eur J Neurosci. 2013; 38: 2572–2588
  302. Richfield, E.K., Penney, J.B., and Young, A.B. Anatomical and affinity state comparisons between dopamine D1 and D2 receptors in the rat central nervous system. Neuroscience. 1989; 30: 767–777
  303. Stuber, G.D., Wightman, R.M., and Carelli, R.M. Extinction of cocaine self-administration reveals functionally and temporally distinct dopaminergic signals in the nucleus accumbens. Neuron. 2005; 46: 661–669
  304. Grill, H.J. and Norgren, R. The taste reactivity test. I. Mimetic responses to gustatory stimuli in neurologically normal rats. Brain Res. 1978; 143: 263–279
  305. Swanson, L.W. Brain Maps. 3rd ed. Elsevier Academic Press, Oxford, UK; 2004
  306. Aragona, B.J., Cleaveland, N.A., Stuber, G.D., Day, J.J., Carelli, R.M., and Wightman, R.M. Preferential enhancement of dopamine transmission within the nucleus accumbens shell by cocaine is attributable to a direct increase in phasic dopamine release events. J Neurosci. 2008; 28: 8821–8831
  307. Koob, G.F. and Le Moal, M. Drug abuse: Hedonic homeostatic dysregulation. Science. 1997; 278: 52–58
  308. Shaham, Y., Shalev, U., Lu, L., De Wit, H., and Stewart, J. The reinstatement model of drug relapse: History, methodology and major findings. Psychopharmacology (Berl). 2003; 168: 3–20
  309. Phillips, P.E., Stuber, G.D., Heien, M.L., Wightman, R.M., and Carelli, R.M. Subsecond dopamine release promotes cocaine seeking. Nature. 2003; 422: 614–618
  310. Wise, R.A. Dopamine, learning and motivation. Nat Rev Neurosci. 2004; 5: 483–494
  311. Blacktop, J.M., Seubert, C., Baker, D.A., Ferda, N., Lee, G., Graf, E.N., and Mantsch, J.R. Augmented cocaine seeking in response to stress or CRF delivered into the ventral tegmental area following long-access self-administration is mediated by CRF receptor type 1 but not CRF receptor type 2. J Neurosci. 2011; 31: 11396–11403
  312. Hahn, J., Hopf, F.W., and Bonci, A. Chronic cocaine enhances corticotropin-releasing factor-dependent potentiation of excitatory transmission in ventral tegmental area dopamine neurons. J Neurosci. 2009; 29: 6535–6544
  313. Wang, B., Shaham, Y., Zitzman, D., Azari, S., Wise, R.A., and You, Z.B. Cocaine experience establishes control of midbrain glutamate and dopamine by corticotropin-releasing factor: A role in stress-induced relapse to drug seeking. J Neurosci. 2005; 25: 5389–5396
  314. Wang, B., You, Z.B., Rice, K.C., and Wise, R.A. Stress-induced relapse to cocaine seeking: Roles for the CRF(2) receptor and CRF-binding protein in the ventral tegmental area of the rat. Psychopharmacology (Berl). 2007; 193: 283–294
  315. Owesson-White, C.A., Roitman, M.F., Sombers, L.A., Belle, A.M., Keithley, R.B., Peele, J.L. et al. Sources contributing to the average extracellular concentration of dopamine in the nucleus accumbens. J Neurochem. 2012; 121: 252–262
  316. Ungless, M.A., Singh, V., Crowder, T.L., Yaka, R., Ron, D., and Bonci, A. Corticotropin-releasing factor requires CRF binding protein to potentiate NMDA receptors via CRF receptor 2 in dopamine neurons. Neuron. 2003; 39: 401–407
  317. Kalivas, P.W. and Duffy, P. Time course of extracellular dopamine and behavioral sensitization to cocaine. I. Dopamine axon terminals. J Neurosci. 1993; 13: 266–275
  318. Thierry, A.M., Tassin, J.P., Blanc, G., and Glowinski, J. Selective activation of mesocortical DA system by stress. Nature. 1976; 263: 242–244
  319. Abercrombie, E.D., Keefe, K.A., DiFrischia, D.S., and Zigmond, M.J. Differential effect of stress on in vivo dopamine release in striatum, nucleus accumbens, and medial frontal cortex. J Neurochem. 1989; 52: 1655–1658
  320. McFarland, K., Davidge, S.B., Lapish, C.C., and Kalivas, P.W. Limbic and motor circuitry underlying footshock-induced reinstatement of cocaine-seeking behavior. J Neurosci. 2004; 24: 1551–1560
  321. Hikida, T., Yawata, S., Yamaguchi, T., Danjo, T., Sasaoka, T., Wang, Y., and Nakanishi, S. Pathway-specific modulation of nucleus accumbens in reward and aversive behavior via selective transmitter receptors. Proc Natl Acad Sci U S A. 2013; 110: 342–347
  322. Smith-Roe, S.L. and Kelley, A.E. Coincident activation of NMDA and dopamine D1 receptors within the nucleus accumbens core is required for appetitive instrumental learning. J Neurosci. 2000; 20: 7737–7742
  323. Berridge, K.C. and Robinson, T.E. What is the role of dopamine in reward: Hedonic impact, reward learning, or incentive salience?. Brain Res Brain Res Rev. 1998; 28: 309–369
  324. Wise, R.A. and Koob, G.F. The development and maintenance of drug addiction. Neuropsychopharmacology. 2014; 39: 254–262
  325. Solomon, R.L. and Corbit, J.D. An opponent-process theory of motivation. II. Cigarette addiction. J Abnorm Psychol. 1973; 81: 158–171
  326. Solomon, R.L. and Corbit, J.D. An opponent-process theory of motivation. I. Temporal dynamics of affect. Psychol Rev. 1974; 81: 119–145
  327. Wikler, A. Recent progress in research on the neurophysiologic basis of morphine addiction. Am J Psychiatry. 1948; 105: 329–338
  328. Robinson, T.E. and Berridge, K.C. The neural basis of drug craving: An incentive-sensitization theory of addiction. Brain Res Brain Res Rev. 1993; 18: 247–291
  329. Stewart, J. Conditioned and unconditioned drug effects in relapse to opiate and stimulant drug self-adminstration. Prog Neuropsychopharmacol Biol Psychiatry. 1983; 7: 591–597
  330. Stewart, J., de Wit, H., and Eikelboom, R. Role of unconditioned and conditioned drug effects in the self-administration of opiates and stimulants. Psychol Rev. 1984; 91: 251–268
  331. Stewart, J. and Wise, R.A. Reinstatement of heroin self-administration habits: Morphine prompts and naltrexone discourages renewed responding after extinction. Psychopharmacology (Berl). 1992; 108: 79–84
  332. Kenny, P.J., Chen, S.A., Kitamura, O., Markou, A., and Koob, G.F. Conditioned withdrawal drives heroin consumption and decreases reward sensitivity. J Neurosci. 2006; 26: 5894–5900
  333. Willuhn, I., Burgeno, L.M., Groblewski, P.A., and Phillips, P.E. Excessive cocaine use results from decreased phasic dopamine signaling in the striatum. Nat Neurosci. 2014; 17: 704–709
  334. Koob, G.F. Negative reinforcement in drug addiction: The darkness within. Curr Opin Neurobiol. 2013; 23: 559–563
  335. Tsibulsky, V.L. and Norman, A.B. Satiety threshold: A quantitative model of maintained cocaine self-administration. Brain Res. 1999; 839: 85–93
  336. Wise, R.A., Newton, P., Leeb, K., Burnette, B., Pocock, D., and Justice, J.B. Jr. Fluctuations in nucleus accumbens dopamine concentration during intravenous cocaine self-administration in rats. Psychopharmacology (Berl). 1995; 120: 10–20