Role of matrix metalloproteinases in the acquisition and reconsolidation of cocaine-induced conditioned place preference

Abstract

Persistent drug seeking/taking behavior involves the consolidation of memory. With each drug use, the memory may be reactivated and reconsolidated to maintain the original memory. During reactivation, the memory may become labile and susceptible to disruption; thus, molecules involved in plasticity should influence acquisition and/or reconsolidation. Recently, matrix metalloproteinases (MMPs) have been shown to influence neuronal plasticity, presumably by their regulation of extracellular matrix (ECM) molecules involved in synaptic reorganization during learning. We hypothesized that inhibition of MMP activity would impair the acquisition and/or reconsolidation of cocaine-conditioned place preference (CPP) in rats. Intracerebral ventricular (i.c.v.) microinjection of a broad spectrum MMP inhibitor, FN-439, prior to cocaine training suppressed acquisition of CPP and attenuated cocaine-primed reinstatement in extinguished animals. In a separate experiment, the cocaine memory was reactivated on two consecutive days with a cocaine priming injection. On these two days, artificial cerebral spinal fluid (aCSF) or FN-439 was administered either 30 min prior to or 1 min after cocaine-primed reinstatement sessions. Infusion of FN-439 partially impaired retrieval of the cocaine-associated context when given 30 min prior to cocaine. In both groups, however, FN-439 suppressed reinstatement compared with controls on the third consecutive test for cocaine-primed reinstatement, when no FN-439 was given. Control experiments demonstrated that two injections of FN-439 + cocaine given in the home cage, or of FN-439 + saline priming injections in the CPP chambers did not disrupt subsequent cocaine-primed reinstatement. These results show for the first time that (1) MMPs play a critical role in acquisition and reconsolidation of cocaine-induced CPP, and (2) rats demonstrate apparent disruption of reconsolidation by an MMP inhibitor after extinction and while they are under the influence of cocaine during reinstatement.

Figure 1.

MMP enzyme activity and inhibition by FN-439. (A) Inhibition of MMP-9 catalytic domain-mediated activity and inhibition by FN-439. Data are mean ± SEM of relative fluorescence units (RFU). N = 3 per dose. The IC₅₀ is 223 μM. (B) Effect of FN-439 on total MMP activity in dorsal hippocampal tissue. Hippocampal tissue was chosen because it expresses relatively high levels of MMP-2 and MMP-9 (the gelatinases tested in C, below). Data are mean ± SEM of RFU, normalized for protein content. N = 3–6 per dose. The IC₅₀ is 3.9 mM. (C) In situ zymography demonstrating basal gelatinase activity by MMP-2 and MMP-9 in the dentate gyrus in the presence of PBS (left panel) and in the presence of FN-439 (right panel). MMP activity is indicated by green fluorescence, and nuclei are indicated by blue fluorescence. Note the similar appearance of the cell nuclei between control and FN-treated sections, indicating a lack of overt cellular toxicity. The basal activity of MMP-2 and MMP-9 was not detectable after FN-439 treatment.

Figure 2.

MMP inhibition by FN-439 infusions given 30 min prior to training sessions suppresses acquisition and subsequent reinstatement of cocaine CPP. (A) Comparison among the three treatment groups. Data are mean ± SEM of time spent in the cocaine-paired compartment. Microinjection (i.c.v.) of aCSF or FN-439 was given on days 1, 3, 5, and 7, 30 min prior to each cocaine injection on the training days (which are not shown above). Initial preference indicates initial preference for cocaine-paired chamber prior to training; test, test day for place preference in cocaine-paired chamber in which no cocaine was administered; last extinction, last extinction day; cocaine prime, cocaine-primed reinstatement using 10 mg/kg i.p. cocaine. Rats were tested immediately after cocaine injection for reinstatement for a 15-min period. Only the last extinction day is shown for clarity. Pooled saline indicates rats given only saline paired with each CPP compartment and given either aCSF or FN-439. Line represents the average of all three groups on the initial preference day. N = 12 for pooled saline group; N = 6 for aCSF group; N = 8 for FN-439 group. There was a significant day effect (F_(8,184) = 8.73, P < 0.0001) and a significant treatment × day interaction (F_(16,184) = 8.08, P < 0.0001). *P < 0.05, compared within groups to their initial preference day; + P < 0.05, compared with aCSF group on the same day. (B–D) Treatment groups are shown separately to indicate time spent in each of the three CPP compartments (cocaine-paired, middle, saline-paired) during each phase of CPP. (B) There was a significant chamber effect (F_(2,22) = 6.73, P < 0.005) and a significant chamber × day interaction (F_(6,66) = 3.08, P < 0.010). (C) There was a significant chamber effect (F_(2,10) = 5.16, P < 0.029) and a significant chamber × day interaction (F_(6,30) = 26.5, P < 0.0001). (D) There was a significant chamber × day interaction (F_(6,42) = 2.70, P < 0.026). *P < 0.05, compared to saline-paired side on the same day. Note that for clarity, only significant differences from the saline-paired side are indicated in B–D.

Figure 3.

MMP inhibition by FN-439 infusions given 30 min prior to daily extinction sessions does not suppress extinction or subsequent cocaine-primed reinstatement. Data are mean ± SEM of time spent in the cocaine-paired compartment. Microinjection (i.c.v.) of aCSF or FN-439 was given 30 min prior to the test and extinction days 1–5 (arrows). See Figure 2 for full explanation of terminology. Three consecutive reinstatement days were done (cocaine prime 1–3). N = 10 for pooled saline group; N = 9 for aCSF group; N = 10 for FN-439 group. There was a significant treatment effect (F_(2,26) = 18.37, P < 0.0001), a significant day effect (F_(11,286) = 26.88, P < 0.0001) and a significant treatment × day interaction (F_(22,286) = 8.00, P < 0.0001). *P < 0.05, compared within groups to their initial preference day; + P < 0.05, compared with aCSF group on the same day.

Figure 4.

MMP inhibition by FN-439 infusions given 30 min prior to reinstatement sessions blocks reinstatement of cocaine CPP. Data are mean ± SEM of time spent in the cocaine-paired compartment. Three consecutive reinstatement days were done (cocaine prime 1–3). Microinjection (i.c.v.) of aCSF or FN-439 (indicated by arrows) was given 30 min prior to each of the first two reinstatement days in the CPP cage. See Figure 2 for full explanation of terminology. N = 11 for aCSF group; N = 10 for FN-439 group. There was a significant treatment effect (F_(1,19) = 10.30, P < 0.004), a significant day effect (F_(5,95) = 12.54, P < 0.0001) and a significant treatment × day interaction (F_(5,95) = 8.08, P < 0.029). *P < 0.05, compared within groups to their initial preference day; + P < 0.05, compared with aCSF group on the same day.

Figure 5.

MMP inhibition by FN-439 infusions given 1 min after reinstatement sessions were completed suppresses reinstatement of cocaine CPP. Data are mean ± SEM of time spent in the cocaine-paired compartment. Three consecutive reinstatement days were done (cocaine prime 1–3). Microinjection (i.c.v.) of aCSF or FN-439 (indicated by arrows) was given 1 min after each of the first two reinstatement days in the CPP cage. See Figure 2 for full explanation of terminology. N = 13 per group. There was a significant treatment × day effect (F_(5,120) = 2.88, P < 0.017). *P < 0.05, compared within groups to their initial preference day; + P < 0.05, compared with aCSF group on the same day.

Figure 6.

MMP inhibition by FN-439 infusions given 30 min prior to reinstatement sessions in the home cage does not alter reinstatement of cocaine CPP. Data are mean ± SEM of time spent in the cocaine-paired compartment. Three consecutive reinstatement days were done (cocaine prime 1–3), but drugs were administered in the home cage on the first 2 d instead of in the CPP box. Microinjection (i.c.v.) of aCSF or FN-439 (indicated by arrows) was given 30 min prior to cocaine injection in the home cage or in the CPP cage (cocaine prime 3). See Figure 2 for full explanation of terminology. N = 10 for aCSF group; N = 11 for FN-439 group. There was a significant day effect (F_(3,57) = 14.02, P < 0.0001). *P < 0.05, compared within groups to their initial preference day.

Figure 7.

MMP inhibition by FN-439 infusions given 30 min prior to saline priming injections does not alter subsequent cocaine-primed reinstatement. Data are mean ± SEM of time spent in the cocaine-paired compartment. Three consecutive reinstatement days were done, but on the first 2 d only saline injections were given (saline prime 1 and 2). A cocaine injection was given 1 d later (cocaine prime). Microinjection (i.c.v.) of aCSF or FN-439 (indicated by arrows) was given 30 min prior to saline injections. See Figure 2 for full explanation of terminology. N = 6 for aCSF group; N = 7 for FN-439 group. There was a significant day effect (F_(5,55) = 5.31, P < 0.0005). *P < 0.05, compared within groups to their initial preference day.