Operant conditioning maintained by cocaine
Then, we investigated the effects of the heterozygous deletion ofPlcb1 in cocaine behavioral responses related to its addictive
properties. For this purpose, Plcb1+/- mice and their WT
littermates were trained for cocaine operant self-administration
(0.5mg/kg/infusion) during FR1 and FR3, progressive ratio, extinction
and cue-induced reinstatement (Figure 2A). Control mice trained with
saline were included for both genotypes. Results showed that the
percentage of mice reaching the criteria of operant conditioning was
100% for both genotypes trained with cocaine and 33% for mice trained
with saline: [Chi-square test=9.50; P <0.01,Plcb1+/- cocaine vs. Plcb1+/- saline] and
[Chi-square test=14.00; P <0.001, WT cocaine
vs. WT saline], as expected.
The primary reinforcing effects and the motivation for cocaine were
similar in both genotypes (Figure 2B-C). During FR1, both genotypes
similarly increased the number of cocaine infusions across sessions,
whereas mice trained with saline remained steady (repeated measures
ANOVA, interaction between genotype x drug x sessions,P <0.05, Figure 2B and Supplementary Table S1). Thus,
the evolution of operant responding was different in mice trained with
cocaine and saline, with an increased responding over nearly all
sessions in cocaine-trained mice . This enhancement in cocaine
responding was more pronounced in mutants than WT mice. Indeed, WT mice
showed a decrease in cocaine intake in the second session. The maximum
number of infusions was reached on session 5 in both genotypes and was
slightly superior in Plcb1+/- mutants (36.35±1.47) than in WT
(29.61±2.23) but post hoc analyses demonstrated that this difference in
cocaine intake was not significant (Figure 2B). Similar findings were
observed for the total number of nose-pokes in which operant responding
increased in both genotypes with cocaine but remained stable with saline
(Supplementary Figure S2A). When the effort to obtain one dose of
cocaine increased to FR3, the number of infusions was stable across
sessions in WT and Plcb1 +/- mice. Similarly, operant responding
was higher for all groups than in FR1 with stable higher levels of
responding (Supplementary Figure S2A) and a higher number of infusions
(repeated measures ANOVA, the main effect of the drug,P <0.001, Figure 2B) in mice trained with cocaine than
with saline, independently of the genotype.
Motivation for cocaine was evaluated in a progressive ratio schedule,
and no significant differences were obtained between genotypes (Figure
2C). The levels of extinction of the operant behavior were similar
between genotypes and decreased progressively across sessions (repeated
measures ANOVA, interaction genotype x sessions,P <0.001, Figure 2D).
The percentage of mice reaching cocaine-seeking extinction criteria was
similar in Plcb1+/- (73%) and WT (78%) mice.
Importantly, Plcb1+/- mice showed significantly reduced
cue-induced reinstatement of cocaine-seeking compared to WT mice
(U Mann-Whitney, P <0.05, Figure 2E), with
27.59% less active nose-pokes compared to WT mice trained with cocaine.
Furthermore, mice trained with saline from both genotypes exhibited
55.84% reduction of active nose-pokes than WT mice trained with cocaine
and 39.20% less than mutants trained with cocaine. No significant
differences were obtained between genotypes in inactive nose-pokes
during operant conditioning maintained by cocaine nor during extinction
(Supplementary Figure S2B-C). Both genotypes trained with cocaine
acquired the reinstatement criterion (double nose pokes in the active
hole than the number of nose pokes during the 3 consecutive days when
the mice acquired the extinction criteria) showing their capability to
maintain this conditioning learning task. These data showed thatPlcb1+/- resulted in a phenotype of resistance to cue-induced
reinstatement with reduced cocaine-seeking (Figure 2E).