Glutamate receptor interacting protein acts within the prefrontal cortex to blunt cocaine seeking

Abstract

Glutamate receptor interacting protein (GRIP) is a neuronal scaffolding protein that anchors GluA2-containing AMPA receptors to the cell membrane. GRIP plays a critical role in activity-dependent synaptic plasticity, including that which occurs after drug exposure. Given that cocaine administration alters glutamate receptor trafficking within the prefrontal cortex (PFC), a better understanding of the role of receptor trafficking proteins could lead to a more complete understanding of addictive phenotypes. AMPA receptor trafficking in general, and GRIP specifically, is known to play a role in cocaine seeking and conditioned reward in the nucleus accumbens, but its role in the PFC has not been characterized. The current study demonstrates that conditional deletion of GRIP1 in the medial prefrontal cortex increases the motivation for cocaine and potentiates cue-induced reinstatement of cocaine seeking in male and female mice. As no effects of PFC GRIP1 deletion were seen in reinstatement of food seeking, strategy set-shifting, or reversal learning the effects on cocaine seeking are not related to generalized alterations in cognitive function. While disrupting GRIP1 might be expected to lead to decreased AMPA transmission, our electrophysiological data indicate an increase in sEPSC amplitude in the prefrontal cortex and a corresponding decrease in paired pulse facilitation in the nucleus accumbens. Taken together this suggests a strengthening of the PFC to NAc input following prefrontal GRIP1 deletion that may mediate the enhanced drug seeking behavior.

Keywords: AMPA; Cocaine; Glutamate receptor interacting protein; Prefrontal cortex; Reinstatement.

Figure 1.. Cre recombinase injection into the mPFC leads to a significant decrease in GRIP1 protein levels.

Green areas indicate the location of the bilateral injections of 0.4 µg of GFP or AAV-Cre recombinase into the medial prefrontal cortex (a). Coronal section of the mouse brain showing the viral expression of GFP within the mPFC (b). Quantification of western blot reveals a significant decrease in GRIP1 protein within the mPFC following AAV-Cre injection, as normalized to glyceraldehyde 3-phosphate dehydrogenase [GAPDH; t(35)=4.79, ***p<.0001, n=15–24; c]. Representative western blots showed GRIP1 knockout in the mPFC (d). There was no effect of AAV-Cre injection on either GAPDH expression in the PFC (e) or GRIP1 expression in the nucleus accumbens (NAc; f). Bars represent average ± SEM.

Figure 2.. GRIP1 knockout in the mPFC does not alter operant learning during food self-administration.

Over the 10 days of food self-administration, there were no significant differences between GFP control mice (n=39) and Cre GRIP1 KO mice (n=46) in the number of pellets consumed (a), number of responses on the active wheel (b), or percent of active responses during food training (c). Boxes represent average ± SEM.

Figure 3.. GRIP1 knockout in the mPFC does not alter cocaine self-administration on a fixed ratio schedule of reinforcement.

Over 10 days of cocaine self-administration, there were no significant differences between GFP control mice and Cre GRIP1 KO mice in the number of cocaine infusions or the number of responses on the active wheel (a, b; n=27–34). Boxes represent average ± SEM.

Figure 4.. Prefrontal GRIP1 knockout increases progressive ratio breakpoint and reinstatement responding for cocaine but not sucrose.

Following prefrontal GRIP1 knockout, mice exhibited an increased breakpoint on a progressive ratio schedule for cocaine [a; t(38) = 2.18, *p = 0.04; n=18–24] but not for food (b; n=24–26). Breakpoint is defined as the final ratio the mice achieved before timing out of the session, i.e. the number active responses required to move onto the next step. Further, PFC GRIP1 knockout mice exhibit greater cocaine seeking during cue-induced reinstatement [c; t(31) = 2.27, *p = 0.03; n=14–20] but not greater reinstatement of food seeking (d; n=13–14). Bars represent average ± SEM.

Figure 5.. Prefrontal GRIP1 knockout does not impact cognitive flexibility.

No differences in were seen following prefrontal GRIP1 knockout in set-shifting performance, either trials to criterion or errors to criterion (a, b; n=40–45). Further, no differences were seen between the groups in trials to criterion or errors to criterion in the reversal learning task (c,d; n=40–45). Bars represent average ± SEM.

Figure 6.. Disrupting GRIP1 in the PFC augments glutamate transmission in the PFC and nucleus accumbens.

Quantification of sEPSC amplitude reveals an increase in prefrontal GRIP1 knockout mice compared to GFP controls [a; t(15)=3.24, **p<0.01; n=15–20]. No differences were seen between the groups in sEPSC frequency (b; n=15–19). Prefrontal GRIP1 knockout led to a decrease in the paired-pulse ratio (40ms IPI) in both naïve and cocaine-experienced mice [c; F(1,33)=4.35, p=0.04; n=12–13]. Bars represent average ± SEM.