Infusion of 0.05 μg/0.5 μl/side TCN-201 (TCN), or 2 μg/0.5 μl/side Ro-25-6981 (Ro) into the prelimbic cortex blocked BDNF’s suppressive effect on cocaine-seeking during the post-abstinence (PA) relapse test and cue-induced reinstatement test. (A) Left: The average number of inactive lever presses during the last 3d of cocaine SA was similar among groups. Right: During the last 3d of cocaine SA, the average number of active lever presses was similar among all rats subsequently assigned to one of six infusion groups (n=17–19 in Veh-P or Veh-B; n=6–10 in other groups). (B) Left: The average number of inactive lever presses during the PA test was similar among groups. Right: TCN or Ro effects on PA test responding after one week of abstinence. Veh-BDNF rats showed significantly less pressing on the previously active levers than Veh-PBS controls (***p < 0.001), but rats infused with TCN or Ro before BDNF pressed significantly more on the active levers than rats infused with vehicle-BDNF (##p < 0.01; #p < 0.05) (n=17–19 in vehicle-infused groups; n=6–10 in other groups). (C) Left: The average number of inactive lever presses during the last 2d of extinction training was similar among groups. Right: During the last 2d of extinction training, the average number of active lever presses was similar among groups (n=17–18 in vehicle-infused groups; n=6–9 in other groups). (D) Left: The average number of inactive lever presses during the cue-induced reinstatement was similar among groups. Right: TCN or Ro effects on cue-induced reinstatement after extinction training. Vehicle-BDNF- treated rats exhibited significantly less active lever pressing than Vehicle-PBS controls (***p<0.001). Infusion of TCN or Ro before BDNF significantly inhibited the ability of BDNF to suppress cocaine-seeking compared to the Vehicle-BDNF group (##p< 0.01) (n=17–18 in vehicle-infused groups; n=6–9 in other groups). The bar graphs indicate the mean ± SEM; Veh= vehicle, TCN=TCN-201, Ro=Ro-25-6981, P=PBS, B= BDNF.