Effect of ShA and LgA amphetamine self-administration on glial cells in SERT+/+ and SERT-/- rats.
We next investigated the protein expression of the glial glutamate transporter (GLT-1) that is responsible of the reuptake of glutamate back into the glial cells from the extracellular space (Roberts-Wolfe & Kalivas, 2015). In the cNAc two-way ANOVA revealed a significant AMPH access effect (F(2,57)=10.94, p<0.0001) and an effect of genotype (F(1,57)=9.349, p=0.0034), whereas in the sNAc an effect of AMPH access (F(2,56)=5.055, p=0.0096), but no genotype effect was found. In both cNAc and sNAc, two-way ANOVA revealed a significant AMPH access x genotype interaction (cNAc: F(2,57)=3.444, p=0.0387, Fig 5a; sNAc: F(2,56)=3.347, p=0.0423, Fig 5b). Examining the individual treatment effects, we found that SERT deletion in both cNAc and sNAc led to a significant increase of GLT-1 expression in response to the LgA (cNAc: +29% vs SERT-/--naive, p=0.0115; +27% vs SERT-/--ShA, p=0.002; sNAc: +29% vs SERT-/--naive, p=0.0184; +25% vs SERT-/--ShA, p=0.009), but not to the ShA procedure, whereas no significant effect was observed in SERT+/+rats.
Effect of ShA and LgA amphetamine self-administration on NMDA receptor subunits and the respective scaffolding proteins in the post-synaptic terminal of SERT+/+ and SERT-/- rats .
The next step was to investigate the protein expression of the main subunits of the NMDA receptors. In the cNAc, two-way ANOVA of GluN1 levels, the obligatory subunit of NMDA receptors revealed an effect of genotype (F(1,57)=4.544, p=0.0374) and a significant AMPH access x genotype interaction (F(2,57)=8.103, p=0.0008; Fig 6a). Post-hoc testing revealed that the expression of this subunit of the NMDA receptor is significantly increased following the LgA (+40% vs SERT-/--naive, p=0.0056), but not the ShA procedure in SERT-/- rats. Neither ShA, nor LgA, affected GluN1 expression in the cNAc of in SERT+/+rats. A different situation was, instead, observed in the sNAc. Two-way ANOVA revealed a main effect of AMPH access (F(2,53)=3.314, p=0.044), genotype (F(1,53)=20.53, p<0.0001) and a significant AMPH access x genotype interaction (F(2,53)=4.675, p=0.0135; Fig 6b). Upon sub-testing we found an increased expression of GluN1 in SERT-/- naïve-rats (+44% vs SERT+/+-naive, p=0.0006), which was not further enhanced by the ShA and the LgA procedure. Conversely, both the ShA (+31% vs SERT-/--naive, p=0.0085) and LgA (+32% vs SERT-/--naive, p=0.0089) conditions caused an enhancement in GluN1 protein levels in SERT+/+ rats.
We then investigated the expression of the accessory subunits of the NMDA receptor, i.e. GluN2A and GluN2B. In the cNAc, two-way ANOVA of GluN2A levels revealed a main effect of AMPH access (F(2,57)=5.137, p=0.0089), and an AMPH access x genotype interaction effect (F(2,57)=11.36, p<0.0001; Fig 6c); in the same subregion, for the GluN2B subunit, two-way ANOVA revealed a main effect of genotype (F(1,56)=12.21, p=0.0009) and an AMPH access x genotype interaction (F(2,56)=10.41, p=0.0001; Fig 6e). GluN2A subunit expression was reduced in SERT-/- naïve-rats (-32% vs SERT+/+-naive, p=0.0433); further, the post-hoc intergroup comparisons revealed that the removal of SERT caused a different response to both the ShA and LgA conditions. In fact, GluN2A protein levels were reduced following both ShA (-36% vs SERT+/+-naive, p=0.0032) and LgA (-34% vs SERT+/+-naive, p=0.0063) AMPH self-administration in SERT+/+ rats, whereas in SERT-/-rats no changes were found for ShA while GluN2A protein levels were up-regulated following LgA (+28% vs SERT-/--naive, p=0.0426; +27% vs SERT-/--ShA, p=0.0085). GluN2B levels were not altered in AMPH-exposed SERT+/+ rats (ShA: -18% vs SERT+/+-naive, p=0.5623; LgA: -27% vs SERT+/+-naive, p=0.1727). In SERT-/-rats similarly to GluN2A, we found that the LgA (+42% vs SERT-/--naive, p=0.0071; +33% vs SERT-/--ShA, p=0.010), but not the ShA procedure significantly up-regulated the expression of this subunit.
In the sNAc, two-way ANOVA revealed only a significant genotype effect (F(1,55)=19.76, p<0.0001; Fig 6d) for GluN2A, and a significant AMPH access effect (F(2,56)=15.03, p<0.0001; Fig 6f) for GluN2B.
Following the analysis of the NMDA receptor subunits, we investigated the expression of the main scaffolding protein of NMDA receptors, SAP102 (Won et al., 2017). In the cNAc, two-way ANOVA revealed a main effect of AMPH access (F(2,56)=9.318, p=0.0003) and an AMPH access x genotype interaction (F(2,56)=4.476, p=0.0157; Fig 7a). Post-hoc testing of the main treatment effects showed that both ShA (-20% vs SERT+/+-naive, p=0.0033) and LgA (-27% vs SERT+/+-naive, p=0.0001) AMPH self-administration resulted in reduced SAP102 expression in SERT+/+ rats, whereas no changes were observed under both experimental conditions in SERT-/- rats. In the sNAc, no significant changes in SAP102 levels were observed (Fig 7b).
Effect of ShA and LgA amphetamine self-administration on AMPA receptor subunits and their respective scaffolding proteins in the post-synaptic terminal of SERT+/+ and SERT-/- rats .
We then moved to the protein expression analysis of the main subunits of the AMPA receptor, i.e. GluA1 and GluA2. In the cNAc, for both GluA1 and GluA2 subunits, two-way ANOVA revealed a significant AMPH access x genotype interaction (GluA1: F(2,57)=16.03, p<0.0001, Fig 8a; GluA2: F(2,57)=13.79, p<0.0001, Fig 8c). For GluA2 two-way ANOVA revealed also a main effect of genotype (F(2,57)=13.79, p=0.0006). SERT knock out reduced GluA1 (-31% vs SERT+/+-naive, p=0.0477), but not GluA2, levels. The LgA condition reduced GluA1 levels in SERT+/+ (-37% vs SERT+/+-naive, p=0.0017) while increasing it in SERT-/- (+34% vs SERT-/--naive, p=0.0048). GluA2 levels were reduced following both ShA (-26% vs SERT+/+-naive, p=0.0447) and LgA (-32% vs SERT+/+-naive, p=0.0083) AMPH self-administration in SERT+/+, while deletion of SERT increased GluA2 levels following both ShA (+27% vs SERT-/--naive, p=0.0329) and LgA (+32% vs SERT-/--naive, p=0.0071) AMPH self-administration. In the sNAc, two-way ANOVA of both GluA1 and GluA2 AMPA subunits revealed a main effect of AMPH access (GluA1: F(2,57)=17.17, p<0.0001; GluA2: F(2,58)=4.995, p=0.010), genotype (GluA1: F(1,57)=17.81, p<0.0001; GluA2: F(1,58)=15.09, p=0.0003) and an AMPH access x genotype interaction effect (GluA1: F(2,57)=3.571, p=0.0346, Fig 8b; GluA2: F(2,58)=3.301, p=0.0439, Fig 8d). Both ShA and LgA conditions resulted in an elevation in GluA1 and GluA2 protein levels in SERT+/+ rats (GluA1: ShA +34% vs SERT+/+-naive, p=0.0329; LgA +37% vs SERT+/+-naive, p=0.0185; GluA2: ShA +46% vs SERT+/+-naive, p=0.0259; LgA +44% vs SERT+/+-naive, p=0.0411) whereas, in SERT-/- rats, no changes were observed with the exception of the increased expression of GluA1 following LgA (+54% vs SERT-/--naive, p=0.0001; -37% vs SERT-/--ShA, p=0.0017).
We then examined the expression of the scaffolding proteins specific for AMPA receptor subunits GluA1 and GluA2, i.e. SAP97 and GRIP, respectively (Won et al., 2017). In the cNAc, for both SAP97 and GRIP, two-way ANOVA revealed a main effect of AMPH access (SAP97: F(2,57)=4.464, p=0.0158; GRIP: F(1,58)=5.822, p=0.0050), and an AMPH access x genotype interaction (SAP97: F(2,57)=5.152, p=0.0088, Fig 9a; GRIP: F(2,58)=4.385, p=0.0169, Fig 9c). In SERT+/+ rats, we found a significant reduction of SAP97 under LgA AMPH self-administration conditions only (-30% vs SERT+/+-naive, p=0.0011), whereas GRIP levels were reduced under both ShA (-27% vs SERT+/+-naive, p=0.0108) and LgA conditions (-32% vs SERT+/+-naive, p=0.0019). Conversely, under both experimental conditions, no significant changes were shown in SERT-/- rats. In the sNAc, two-way ANOVA of both SAP97 and GRIP levels revealed a main effect of AMPH access (SAP97: F(2,57)=7.647, p=0.0011; GRIP: F(1,57)=3.795, p=0.0284) and an AMPH access x genotype interaction (SAP97: F(2,57)=15.86, p<0.0001, Fig 9b; GRIP: F(2,57)=7.230, p=0.0016, Fig 9d). Post-hoc testing of the main treatment effects in SERT+/+ rats revealed reduced expression of SAP97 following LgA (-25% vs SERT+/+-naive, p=0.0429), but not ShA, procedures whereas GRIP was similarly reduced by both experimental conditions (ShA: -23% vs SERT+/+-naive, p=0.0150; LgA: -31% vs SERT+/+-naive, p=0.0005). Conversely, in SERT-/- rats, we found an increase in the expression of SAP97 following the LgA (-30% vs SERT-/--naive, p=0.0093; -41% vs SERT-/--ShA, p<0.0001), but not the ShA condition, whereas GRIP expression was decreased in SERT-/- naïve-rats (-29% vs SERT+/+-naive, p=0.0059) without further changes following both ShA and LgA AMPH self-administration.