3.1. Behavioral experiments
3.1.1. Maintenance of cocaine self-administration
Figure 2a shows the effects of the acute WAY administration in OBX and SHAM rats on the active and inactive lever presses and on the number of cocaine infusions. A three-way ANOVA for the surgical condition × pretreatment × lever did not show differences between the groups treated with WAY (F(2,74) = 0.10 p = 0.91), and the analysis inside each experimental group treated with WAY did not indicate a difference in the cocaine reinforcement (a two-way ANOVA: F(2,34) = 1.50 p = 0.24 and F(2,40) = 1.10 p = 0.34, respectively).
Acute treatment with WAY did not alter the number of cocaine infusions in both animal groups, as shown by a two-way ANOVA for the surgical condition × pre-treatment interaction in OBX (F(2,17) = 1.60 p = 0.23) and SHAM groups (F(2,20) = 0.93 p = 0.40); the observed effect was independent of surgical manipulation (F(3,44) = 0.24 p = 0.86).
The numbers of active and inactive lever presses and the number of cocaine infusions for OBX and SHAM animals after RO administration are shown in Fig. 2b. A three-way ANOVA for the surgical condition × pretreatment × lever interaction did not show a difference between the types following RO treatment (F(2,66) = 0.48 p = 0.62), what means that the administration of RO had a similar effect on the number of presses on the active lever in both groups of animals. Pre-treatment with RO in a dose-dependent manner decreased the number of presses of the active, but not the inactive lever in both groups of rats. Individual two-way ANOVA analyses for OBX and SHAM groups indicted a significant effect for the pre-treatment × lever interaction (F(2,30) = 5.54 p < 0.01 and F(2,36) = 16.66 p < 0.001, respectively).
Removal of the olfactory bulbs did not affect the number of cocaine infusions after acute administration of RO (F(2,33) = 1.21 p = 0.34). A one-way ANOVA for each group showed a significant effect of RO on the cocaine reinforcement in OBX (F(4,28) = 3.85 p < 0.05, 3–10 mg/kg) and SHAM rats (F(2,18) = 21.67 p < 0.001), but only at a dose of 10 mg/kg (Fig. 2b).
The lack of significant effects after acute administration of MIR in OBX and SHAM rats was demonstrated with a three-way ANOVA (F(2,182) = 0.69 p = 0.57; Fig. 2c). The inner analysis for each experimental group treated with MIR did not indicate a difference in the cocaine reinforcement. MIR (5–20 mg/kg) did not alter the numbers of lever responses in OBX or SHAM rats (a two-way ANOVA: F(3,88) = 2.30 p = 0.08 and F(3,94) = 0.14 p = 0.92, respectively).
Acute treatment with MIR did not alter the number of cocaine infusions in both animal groups, as shown by a two-way ANOVA for the surgical condition × pre-treatment interaction in OBX (F(3,47) = 0.37 p = 0.77) and SHAM groups (F(3,44) = 1.60 p = 0.33). The observed response was independent of surgical manipulation (F(3,91) = 0.89 p = 0.45).
3.1.2. Extinction training
After cocaine self-administration, an extinction training was introduced to all animals, without the drug or a drug-associated cue. Figure 3 shows the effects of the repeated administration of WAY, RO or MIR, in OBX and SHAM rats on the active and inactive lever presses during extinction training.
A mixed repeated measures ANOVA did not show a significant effect of the repeated administration of WAY (1 mg/kg) or VEH during extinction training in OBX and SHAM rats (the surgical condition × session × pre-treatment × lever interaction (F(8,464) = 0.68 p = 0.71). Similarly, a three-way repeated measures ANOVA for each group separately did not indicate a significant effect for the session × pre-treatment × lever interaction in OBX (F(8,256) = 1.30 p = 0.24) and SHAM rats (F(8,208) = 3.40 p = 0.92)(Fig. 3a).
Repeated administration of RO (1 mg/kg) during extinction training in SHAM and OBX rats did not cause changes in the number of lever presses (F(8,256) = 0.52 p = 0.84 and F(8,208) = 0.19 p = 0.99, respectively). Moreover, the effect of the session × pre-treatment × lever interaction was the same in both groups of animals (F(8,454) = 0.48 p = 0.87) (Fig. 3b).
In like manner, the repeated administration of MIR (10 mg/kg) or VEH did not significantly vary lever responses between experimental groups during the extinction training F(8,608) = 0.74 p = 0.65). The inner three-way repeated measures ANOVA for each group did not indicate a significant effect for the session × pre-treatment × lever interaction in OBX (F(8,272) = 1.18 p = 0.31) and SHAM rats (F(8,304) = 0.99 p = 0.44)(Fig. 3c).
3.1.3. Cue-induced reinstatement of seeking behavior – acute administration
The results for the WAY experiment following presentation of the cocaine-associated cue (tone + light) to OBX and SHAM groups are shown in Fig. 4a. A two-way ANOVA for the self-administration procedure × lever interaction showed a decrease in the number of active lever presses after 10 days extinction training in SHAM and OBX groups, as compared to cocaine self-administration (F(1,26) = 193.02 p < 0.001 and F(1,28) = 66.56 p < 0.001, respectively), regardless of the surgical condition (F(1,54) = 1.01 p = 0.31).
A significant main effect was observed for the reinstatement (cue) × lever interaction (F(1,54) = 0.63 p = 0.43), and a post hoc analysis revealed an increase (p < 0.001) in the number of active lever presses during the reinstatement of the drug-seeking behavior in rats that had previously self-administered cocaine (0.5 mg/kg/infusion).
A three-way ANOVA did not reveal an effect for the surgical condition × pre-treatment × lever interaction (F(3,104) = 0.42 p = 0.74), what indicates that there was no significant difference between OBX and SHAM rats in the effects of WAY. An individual two-way ANOVA analysis for SHAM and OBX rats showed that pre-treatment with WAY (0.1–1 mg/kg) significantly decreased the cue-induced reinstatement in SHAM (pre-treatment × lever interaction: F(3,54) = 3.04 p < 0.05) and OBX (F(3,50) = 5.90 p < 0.05) animals, what proves that pretreatment with WAY effectively altered the reinstatement of the drug seeking behavior (Fig. 4a).
For the RO experiment, the results for the cue-induced reinstatement of cocaine seeking in OBX and SHAM animals are shown in Fig. 4b. A two-way ANOVA showed a significant reduction in active lever presses after 10 days of extinction training under both surgical conditions (OBX: F(1,24) = 37.67 p < 0.001; SHAM: F(1,24) = 74.61 p < 0.001). A two-way ANOVA revealed a significant effect for the reinstatement (cue) × lever interaction in both SHAM (F(1,24) = 13.60 p < 0.01) and OBX (F(1,24) = 39.89 p < 0.001) groups, regardless of the surgical condition (F(1,46) = 1.21 p = 0.27).
A three-way ANOVA (the surgical condition × pre-treatment × lever interaction) did not show an effect for acute pre-treatment with RO (0.1-1 mg/kg) (F(3,90) = 0.37 p = 0.76). An individual analysis with a two-way ANOVA in OBX or SHAM rats showed that pre-treatment with RO significantly decreased the cue-induced reinstatement in OBX (pre-treatment × lever: F(3,44) = 4.15 p < 0.01) and SHAM (pre-treatment × lever: F(3,46) = 5.90, p < 0.01) animals, what indicates that pre-treatment with RO effectively changed the reinstatement of the drug seeking behavior (Fig. 4b).
The effects of the MIR experiment, following presentation of the cocaine-associated cue, are shown in Fig. 4c. A two-way ANOVA revealed a significant effect for the self-administration procedure × lever interaction in both OBX (F(1,28) = 24.93 p < 0.001) and SHAM (F(1,28) = 15.34 p < 0.001) rats; a reduction in active lever presses after 10 days of extinction was found in both types of rats.
The conditional cue induced an increase (p < 0.001) in the number of active lever presses during the reinstatement of the drug-seeking behavior in OBX (a two-way ANOVA for reinstatement (cue) × lever interaction: F(1,28) = 28.51 p < 0.001) and SHAM rats (F(1,28) = 16.18 p < 0.001); the observed effect was independent of the surgical condition (F(1,68) = 0.22 p = 0.64).
During cue induced reinstatement a significant reduction in the drug-seeking behavior in OBX and SHAM rats was observed, following acute administration of MIR (a two-way ANOVA: F(3,56) = 24.37 p < 0.001 and F(3,56) = 7.36 p < 0.001). Post hoc analyses showed that a significant reduction in the drug seeking behavior occurred in both groups following administration of MIR at doses 2.5–10 mg/kg in OBX rats and only for 5 and 10 mg/kg in SHAM rats. The effect of pretreatment with MIR on OBX and SHAM rats was regardless of the surgical condition (F(3,112) = 1.33 p = 0.23).
3.1.4. Cocaine-induced reinstatement of seeking behavior – acute administration
Figure 5a shows the effects of acute WAY administration on cocaine-induced reinstatement in OBX and SHAM groups. A 10-day extinction training led to the reduction in active lever presses (p < 0.001), more than inactive lever presses, compared to the last cocaine self-administration session in both experimental groups (Fig. 5a) (a two-way ANOVA for the self-administration procedure × lever interaction: OBX F(1,20) = 39.68 p < 0.001; SHAM F(1,24) = 17.45 p < 0.001). A reduction in active lever presses after 10 days of extinction was found in both group.
A three-way ANOVA did not indicate an effect for the surgical condition × pretreatment × lever interaction (F(3,84) = 2.32 p = 0.08), what indicates that there was no significant difference between OBX and SHAM rats in the effects of WAY. A two-way ANOVA analysis demonstrated a significant effect for the pre-treatment × lever interaction in SHAM and OBX rats (F(3,44) = 8.02 p < 0.001 and F(3,40) = 2.92 p < 0.05, respectively). Post hoc analyses showed that a significant reduction in the drug seeking behavior occurred in both groups following administration of WAY, for all doses (0.3-3 mg/kg) in OBX rats and in SHAM rats only for the dose of 3 mg/kg.
Similarly to the RO experiment, a two-way ANOVA showed a significant reduction in the active lever presses after 10 days of extinction training for both types of surgery condition (OBX: F(1,24) = 37.68 p < 0.001; SHAM: F(1,24) = 74.61 p = < 0.001), regardless of the surgical condition (F(1.48) = 0.54 p = 0.46) (Fig. 5b). Administration of cocaine at a dose of 10 mg/kg (i.p.) resulted in an increase in the responses to the active lever in OBX (F(1,24) = 53.98 p < 0.001) and SHAM groups (F(1,24) = 19.26 p = < 0.001). A three-way ANOVA did not indicate an effect for the surgical condition × pretreatment × lever interaction (F(3,93) = 0.78 p = 0.51), what shows that there was no significant difference between OBX and SHAM rats in the effects of RO. A two-way ANOVA analysis demonstrated a significant effect for the pre-treatment x lever interaction in SHAM and OBX rats (F(3,48) = 7.07 p < 0.001 or F(3,45) = 7.29 p < 0.001, respectively) (Fig. 5b).
For the acute MIR administration experiment, the effect of the 5-HT2C agonist on cocaine-induced reinstatement in OBX and SHAM groups is shown in Fig. 5c. A two-way ANOVA for the self-administration procedure × lever interaction showed a significant reduction in active lever presses after 10 days of extinction in both experimental groups (OBX F(1,32) = 23.86 p < 0.001; SHAM F(1,36) = 95.16 p < 0.001); regardless of the surgical condition (F(1,68) = 0.12 p = 0.73).
Regarding the cocaine-induced seeking behavior, measured as increases of active lever-responses (p < 0.001), a two-way ANOVA revealed an effect for the cue-induced reinstatement of seeking behavior in both experimental groups (OBX: F(1,30) = 31.37 p < 0.001 and SHAM: F(1,36) = 20.00 p < 0.001).
When MIR (2.5–20 mg/kg) was administered before placing rats in the experimental cage, a marked reduction in the number of active lever-responses induced by cocaine was observed (a two-way ANOVA: OBX F(4,74) = 3.81 p < 0.001; SHAM F(4,78) = 4.08 p < 0.01). A post hoc test demonstrated a reduction in active lever presses in rats treated with 10 and 20 mg/kg doses of MIR for OBX and in SHAM rats only for 20 mg/kg. A three-way ANOVA did not indicate an effect for the surgical condition × pretreatment × lever interaction (F(4,150) = 0.14 p = 0.97), what demonstrates that there was no significant difference between OBX and SHAM rats in the effects of MIR.
3.1.5. Cue-induced reinstatement of cocaine seeking behavior – repeated drug administration during extinction training
After10-daily administration of vehicle or WAY (1 mg/kg), during the extinction-training period, rats were tested for the response reinstatement induced by a cocaine-associated cue (Fig. 6a). As shown by a mixed ANOVA for the surgical condition surgical condition × pretreatment × reinstatement (cue) × lever interaction, repeated WAY administration did not alter the cue-induced reinstatement for either type of surgery condition (F(1,116) = 0.17 p = 0.68; Fig. 6a).
In the SHAM-operated animals, rats after repeated WAY administrations during extinction training reacted differently on cue-induced reinstatement (F(1,52) = 15.09 p < 0.001). A post hoc analysis showed a significant increase in the number of active lever presses (p < 0.001) in the group receiving the vehicle, while in the group chronically treated with WAY a significant reduction in the cue-induced relapse (p < 0.001) was observed, as compared to the vehicle group. In OBX group, the differences were also observed in the response on cue-induced reinstatement between rats receiving repeated doses of WAY or VEH (F(1,64) = 14.24 p < 0.05) (Fig. 6a). As demonstrated by the post hoc analysis, WAY administered chronically significantly attenuated the cocaine-induced relapse (p < 0.001) (Fig. 6a).
In like manner, a mixed ANOVA for both types of surgery condition did not show a significant effect of the surgical condition × pretreatment × reinstatement (cue) × lever interaction (F(1,116) = 0.01 p = 0.94) (Fig. 6b), what proves that pretreatment with RO (1 mg/kg) had the same effect for both surgery conditions. However, for these conditions, a three-way ANOVA showed a significant effect of the reinstatement (cue) × lever × pre-treatment interaction in SHAM (F(1,52) = 12.16 p < 0.001) and OBX (F(1,64) = 5.49 p < 0.05) rats. A post hoc analysis showed a significant reduction in the cue-induced relapse for both surgery conditions (p < 0.001) after chronic administration of RO.
As shown by a mixed ANOVA for the surgical condition surgical condition × pretreatment × reinstatement (cue) × lever interaction (F(1,145 = 1.26 p = 0.26), the reinstatement of seeking behavior induced by a cue was relevant following either VEH or MIR (10 mg/kg) pretreatment in both OBX or SHAM rats (Fig. 6c). Moreover, inner analyses for each group did not reveal a significant difference between treatment with VEH or MIR, suggesting that rats responded similarly to the cue (OBX: F(1,68) = 2.28 p = 0.14; SHAM: F(1,76) = 0.25 p = 0.62).
3.1.6. Cocaine-induced reinstatement of cocaine seeking behavior – repeated drug administration during extinction training
Following 10 days of VEH or WAY administration, during the extinction training, SHAM and OBX rats were tested for the response reinstatement induced by cocaine (10 mg/kg, i.p., Fig. 6a). A mixed ANOVA for both surgery conditions did not reveal a significant effect of repeated administration of WAY during extinction training for the pretreatment × surgical condition × cocaine × lever interaction (F(1,116) = 0.02 p = 0.90), what indicates that the surgery condition, OBX and SHAM, did not correlate with a behavioral response to a priming cocaine injection (Fig. 6a). A separate three-way ANOVA for SHAM (F(1,51) = 2.31 p = 0.13) and OBX (F(1,64) = 2.73 p = 0.10) rats showed no significant effect of WAY on the cocaine-induced reinstatement (Fig. 6a).
After 10 days of repeated treatment with the vehicle or RO, during the extinction training, SHAM and OBX rats were tested for the response reinstatement induced by cocaine (10 mg/kg, i.p.). For both types of surgery condition, a mixed ANOVA did not reveal a significant effect for the pretreatment × surgical condition × cocaine × lever interaction (F(1,116) = 0.14 p = 0.71), what provides evidence that OBX and SHAM rats responded similarly to RO pretreatment. A separate analysis for SHAM (F(1,52) = 1.37 p = 0.25) and OBX (F(1,64) = 2.87 p = 0.09) rats indicated no significant effect of RO on the cocaine-induced reinstatement (Fig. 6b).
Repeated administration of VEH or MIR did not change behavioral responses during reinstatement of seeking behavior, represented as the increase of lever presses in experimental groups (a mixed ANOVA for the pretreatment × surgical condition × reinstatement (cocaine) × lever interaction: F(1,144) = 0.06 p = 0.80)(Fig. 6c). Further, an individual analysis revealed that the cocaine-induced reinstatement was similar in both groups (OBX: F(1,76) = 0.12 p = 0.73; SHAM: F(1,68) = 0.57 p = 0.45).
3.2. The expression of the 5-HT2C receptor
Cocaine self-administration per se did not change the 5-HT2C receptor expression in examined brain structures in OBX and SHAM rats as compared to saline SHAM animals (Fig. 7a and Table 1). Removal of the olfactory bulbs induced an increase in the 5-HT2C receptor expression only in the ventral hippocampus in yoked saline SHAM rats (t=-2,43; df = 10; p < 0,05) (Fig. 7a).
Table 1
Results from two-way ANOVA analyses on 5-HT2C receptor expression in the several brain structures in bulbectomized and SHAM rats following cocaine self-administration and saline yoked delivery.
Brain structure | PFCXPL | PFCXIL | FCX | vHIP | dHIP | DLS | DMS | NAC | BLA | CER |
F(1,20)= | 0.01 | 2.40 | 0.28 | 1.20 | 3.34 | 1.82 | 0.47 | 0.11 | 0.61 | 0.90 |
p= | 0.91 | 0.14 | 0.60 | 0.28 | 0.08 | 0.19 | 0.49 | 0.11 | 0.61 | 0.90 |
PFCXIL – infralimbic prefrontal cortex, PFCXPL – prelimbic prefrontal cortex, FCX – frontal cortex, vHIP – ventral hippocampus, dHIP – dorsal hippocampus, DLS – dorsolateral striatum, DMS – dorsomedial striatum, NAC – nucleus accumbens, BLA – basolateral amygdala, CER – cerebellum. |
A two-way ANOVA showed significant changes in the 5-HT2C receptor expression in SHAM and OBX groups after 10 days extinction in rats self-administering cocaine or the vehicle group in the frontal cortex (F(1,20) = 5,42; p < 0,05) and dorsomedial striatum (F(1,20) = 39,88; p < 0,001). In OBX vehicle group, an increase in the 5-HT2C receptor expression was shown in the frontal cortex (t=-2,64; df = 10; p < 0,05), as compared to SHAM vehicle group. In OBX cocaine group, there was a decrease in the 5-HT2C receptor expression in the frontal cortex (t = 3,35; df = 10; p < 0,01) and an increase in the dorsomedial striatum (t=-7,01; df = 10; p < 0,001), compared to OBX vehicle group (Fig. 7b).