Lesions of the Patch Compartment of Dorsolateral Striatum Disrupt Stimulus-Response Learning

Abstract

The striatum mediates habit formation and reward association. The striatum can be divided into the patch and matrix compartment, which are two distinct regions that sub-serve different aspects of behavior. The patch compartment may mediate reward-related behaviors, while the matrix compartment may mediate adaptive motor functions. Previous studies indicate that enhanced relative activation of the patch versus matrix compartment is associated with inflexible behaviors, such as stereotypy. Habitual behaviors are also inflexible in nature, but whether enhanced activation of the patch compartment contributes to habitual behavior is not known. The goal of the current study was to examine the role of patch compartment in the development of habit formation. We used dermorphin-saporin to ablate neurons of the patch compartment in the dorsolateral striatum prior to training animals to self-administer sucrose on a random interval schedule of reinforcement. Our data showed that patch compartment lesions in the dorsolateral striatum reduced the reinstatement of sucrose self-administration after sucrose devaluation, indicating that destruction of this region prevented the development of habitual behavior. Additionally, in animals with patch compartment lesions in the DLS that did not develop habitual behavior, activation of the dorsolateral striatum and sensorimotor cortex was diminished, while activity in the dorsomedial striatum and prefrontal cortex was increased, suggesting less engagement of regions that mediate habitual behaviors and heightened engagement of regions that mediate goal-directed behaviors occurs with reduced habit formation. These data indicate that the dorsolateral patch compartment may mediate habit formation by altering information flow through basal ganglia circuits.

Keywords: basal ganglia; cortex; immediate early gene; self-administration; striosome; sucrose.

Figure 1.

Representative images showing of striatum showing the effects of DERM-SAP infusion on mu opioid receptor immunoreactivity in the striatum. Rats were infused with the neurotoxin DERM-SAP (17 ng/ml) in the DLS (+1.7 mm AP, ±3.0 mm ML, −5.0 mm DV; A). Vehicle-infused rats maintained mu opioid receptor staining in the DLS (B), while DERM-SAP infused rats showed a decrease in mu opioid receptor staining in the DLS (C), indicating a loss of patch compartment neurons in this region.

Figure 2.

Schematic diagram of the regions of rat brain used for analysis of c-Fos immunoreactivity. A 400 x 400 pixel area was analyzed in the mPFC (within the prelimbic subregion) and SMC (at +4.2 mm from bregma; A) and the DMS and DLS (at +1.7 mm from bregma; B).

Figure 3.

Effects of patch compartment lesions in the DLS on stimulus-response-associated learning. A.) Response rates (lever presses/min) of SAP- and DERM-SAP-pretreated animals over eight days of continuous reinforcement (CRF) training and random interval schedule of reinforcement training. The data are expressed as the mean lever presses/minute (±SEM) from the average lever presses for each animal across the two sessions per day for two days of CRF, RI15, RI30 and RI60 schedules of reinforcement during instrumental training. B.) Development of lithium chloride (LiCl)-induced conditioned taste aversion over a three-day period following CRF and RI training. Data are expressed as the mean (±SEM) milliliters of sucrose consumed in the home cage during a 30 min access period, after which time animals were injected with either LiCl (0.15 M) or saline. **significantly different from sucrose consumption in LiCl-treated animals, p<0.05. C.) Effect of outcome devaluation due to LiCl-induced CTA on extinction responding in SAP- and DERM-SAP-pretreated animals trained on RI schedules of reinforcement. The data is expressed as the response rate during the 10-min extinction test after outcome devaluation as a percentage of the response rate on the last day of RI60 training (±SEM). *significantly different from all other groups, p<0.05.

Figure 4.

Effects of patch compartment lesions in the DLS on c-Fos immunoreactivity in the DMS and mPFC (within the prelimbic subregion) following reinforcer devaluation and extinction testing. Photomicrographs showing c-Fos immunoreactivity in the DMS (A) and mPFC (C). Scale bar=100 μM. Quantitative analysis of c-Fos immunoreactivity in the DMS (B) and mPFC (D) in rats infused with SAP or DERM-SAP (17 ng/μl) in the DLS, prior to RI training, CTA and extinction testing. Data are presented as the mean (±SEM) number of c-Fos immunoreactive particles per area. *significantly different from DERM-SAP-pretreated, saline-treated rats, p<0.05; +significantly different from SAP-pretreated, LiCl-treated rats, p<0.05.

Figure 5.

Effects of patch compartment lesions in the DLS on c-Fos immunoreactivity in the DLS and SMC following reinforcer devaluation and extinction testing. Photomicrographs showing c-Fos immunoreactivity in the DLS (A) and SMC (C). Scale bar=100 μM. Quantitative analysis of c-Fos immunoreactivity in the DLS (B) and SMC (D) in rats infused in the DLS with SAP or DERM-SAP (17 ng/μl), prior to RI training, CTA and extinction testing. Data are presented as the mean (±SEM) number of c-Fos immunoreactive particles per area. *significantly different from DERM-SAP-pretreated, saline-treated rats, p<0.05; +significantly different from SAP-pretreated, LiCl-treated rats, p<0.05.