Profiling prefrontal cortex protein expression in rats exhibiting an incubation of cocaine craving following short-access self-administration procedures

Abstract

Introduction: Incubation of drug-craving refers to a time-dependent increase in drug cue-elicited craving that occurs during protracted withdrawal. Historically, rat models of incubated cocaine craving employed extended-access (typically 6 h/day) intravenous drug self-administration (IV-SA) procedures, although incubated cocaine craving is reported to occur following shorter-access IV-SA paradigms. The notoriously low-throughput of extended-access IV-SA prompted us to determine whether two different short-access IV-SA procedures akin to those in the literature result in qualitatively similar changes in glutamate receptor expression and the activation of downstream signaling molecules within prefrontal cortex (PFC) subregions as those reported previously by our group under 6h-access conditions.

Methods: For this, adult, male Sprague-Dawley rats were trained to intravenously self-administer cocaine for 2 h/day for 10 consecutive days (2-h model) or for 6 h on day 1 and 2 h/day for the remaining 9 days of training (Mixed model). A sham control group was also included that did not self-administer cocaine.

Results: On withdrawal day 3 or 30, rats were subjected to a 2-h test of cue-reinforced responding in the absence of cocaine and a time-dependent increase in drug-seeking was observed under both IV-SA procedures. Immunoblotting of brain tissue collected immediately following the cue test session indicated elevated phospho-Akt1, phospho-CaMKII and Homer2a/b expression within the prelimbic subregion of the PFC of cocaine-incubated rats. However, we failed to detect incubation-related changes in Group 1 metabotropic glutamate receptor or ionotropic glutamate receptor subunit expression in either subregion.

Discussion: These results highlight further a role for Akt1-related signaling within the prelimbic cortex in driving incubated cocaine craving, and provide novel evidence supporting a potential role also for CaMKII-dependent signaling through glutamate receptors in this behavioral phenomenon.

Keywords: Akt; Homer proteins; glutamate receptors; incubation of craving; infralimbic cortex; prelimbic cortex.

Figure 1

Cocaine self-administration and test for incubated craving. (A) Experimental timeline of the study. On day 1 of self-administration training, Mixed rats that underwent a 6-h cocaine IV-SA session exhibited more active lever-presses (B), received more reinforcers (C), and exhibited more inactive lever-presses (D), compared to both Sham controls and 2-h rats that underwent a 2-h cocaine IV-SA session. By the last 3 days of self-administration training, rats in both IV-SA groups exhibited a comparable number of active lever-presses (E), reinforcers earned (F), and inactive lever-presses (G), which were higher than Sham controls. (H) On the test for incubated cocaine craving, both cocaine IV-SA groups showed a time-dependent increase in cue-elicited active lever-pressing from WD3 to WD30. (I) There were no time-dependent changes in inactive lever-pressing for any groups. Data is represented as means ± SEMs of the number of rats indicated. *p < 0.05, compared to Sham, **p < 0.05, compared to other two groups, +p < 0.05, compared to WD3 (incubation).

Figure 2

Changes in mGlu1/5 and Homer proteins in the PL. We detected no changes in the expression of mGlu1 (A) or the mGlu5 monomer (B) in the PL. (C) Irrespective of cocaine history, rats tested on WD30 exhibited lower PL expression of the mGlu5 dimer, compared to those tested on WD3. (D) Relative to Sham controls, only rats that underwent the Mixed IV-SA procedure exhibited a significant time-dependent increase in Homer2a/b within the PL. (E) No group differences in Homer1b/c expression were observed within the PL. Data are normalized to the average protein densities of the Sham controls tested on WD3 and represented as means ± SEMs of the number of rats indicated. *p < 0.05, compared to Sham, +p < 0.05, compared to WD3.

Figure 3

Changes in iGluR subunits in the PL. There were no differences between Sham, 2-h and Mixed rats with respect to the expression of the AMPA receptor subunits GluA1 (A) and GluA2 (B) within the PL, nor did we detect group differences in the NMDA subunits, GluN1 (C), GluN2a, (D) or GluN2b (E). Data are normalized to the average protein densities of the Sham controls tested on WD3 and represented as means ± SEMs of the number of rats indicated.

Figure 4

Changes in Akt1 and CaMKII expression and phosphorylation in the PL. While we did not detect any time-dependent or group differences in the PL expression of total Akt1 (A), both 2-h and Mixed rats exhibited a time-dependent increase total p(Ser473)-Akt1 expression (B) and the 2-h rats exhibited a time-dependent increase in the relative expression of p(Ser473)-Akt1 (C). (D) We detected no group or time-dependent changes in total CaMKII expression within the PL. However, the Mixed group demonstrated a significant time-dependent increase in both total (E) and relative (F) p(Thr286)-CaMKII expression. Data are normalized to the average protein densities of the Sham controls tested on WD3 and represented as means ± SEMs of the number of rats indicated. +p < 0.05, compared to WD3.

Figure 5

Changes in glutamate receptors and Homer proteins in the IL. Within the IL, we detected a time-dependent reduction in the mGlu1 monomer (A), mGlu5 monomer (B), mGlu5 dimer (C) and GluN1 (D), irrespective of the cocaine history of the rats, although both IV-SA groups exhibited higher GluN1 expression than Sham controls. Overall, mixed rats exhibited lower levels of both total (E) and (F) relative p(Thr286)-CaMKII expression within the IL, relative to both 2-h rats and Sham controls. Data are normalized to the average protein densities of the Sham controls tested on WD3 and represented as means ± SEMs of the number of rats indicated. *p < 0.05, compared to Sham; +p < 0.05, compared to WD3; **p < 0.05, compared to other two groups.