MeCP2 phosphorylation limits psychostimulant-induced behavioral and neuronal plasticity

Abstract

The methyl-DNA binding protein MeCP2 is emerging as an important regulator of drug reinforcement processes. Psychostimulants induce phosphorylation of MeCP2 at Ser421; however, the functional significance of this posttranslational modification for addictive-like behaviors was unknown. Here we show that MeCP2 Ser421Ala knock-in mice display both a reduced threshold for the induction of locomotor sensitization by investigator-administered amphetamine and enhanced behavioral sensitivity to the reinforcing properties of self-administered cocaine. These behavioral differences were accompanied in the knock-in mice by changes in medium spiny neuron intrinsic excitability and nucleus accumbens gene expression typically observed in association with repeated exposure to these drugs. These data show that phosphorylation of MeCP2 at Ser421 functions to limit the circuit plasticities in the nucleus accumbens that underlie addictive-like behaviors.

Keywords: GABAergic interneurons; MeCP2; cocaine self-administration; intrinsic excitability; nucleus accumbens; psychostimulants.

Figure 1.

MeCP2 Ser421Ala KI mice have a reduced threshold for behavioral sensitization to AMPH. a, Quantification of glutamatergic (VGLUT1-immunoreactive) and GABAergic (VGAT-immunoreactive) synaptic termini in the NAc of adult MeCP2 WT and Ser421Ala KI mice. Fields for quantification were taken from the medial part of the rostral NAc and spanned both core and shell. Scale bar, 10 μm. n = 8–9 WT, 12 KI mice per group. b, c, Open-field locomotor activities 60 min before [preinjection (Pre-Inj)] or 60 min after a single dose of investigator-administered AMPH. b, *p < 0.05, WT 3 mg/kg versus WT Pre-Inj; ^#p < 0.05, KI 3 mg/kg versus KI Pre-Inj or cocaine; n = 6–9 per group. c, *p < 0.05, WT 20 mg/kg versus WT Pre-Inj; ^#p < 0.05, KI 20 mg/kg versus KI Pre-Inj; n = 7 per group. d, Persistent locomotor sensitization induced by 5 d of repeated daily investigator-administered 3 mg/kg AMPH (i.p.) in MeCP2 WT and Ser421Ala KI mice. n = 10 per genotype. *p = 0.028, WT versus KI on Day 3; ^#p = 0.031 for KI on Day1 compared to Day 5; p̂ = 0.026 for WT on Day 1 compared to Day 5. e, Locomotor sensitization after 2 d of investigator-administered AMPH in MeCP2 Ser421Ala KI, but not WT, mice. n = 6 WT and 8 KI; *p < 0.05, KI versus WT on Days 2, 9, and 23; ^#p < 0.05, Day 9 versus Day 1 and Day 23 versus Day 1 in KI group.

Figure 2.

Self-administered cocaine drives MeCP2 Ser421 phosphorylation in select neurons of the NAc. C57BL/6J mice that were food trained in operant chambers were cannulated and then returned to the operant chambers for daily 2 h intravenous SA sessions on an FR1-TO20 schedule with either saline or cocaine. Two hours after the final SA session, mice were perfused, and coronal sections of brain tissue were cut for immunolabeling with antibodies against pMeCP2. a, Representative brain sections immunolabeled for pMeCP2 in mice that self-administered saline or cocaine. The anterior commissure (AC) is at the top left corner of each NAc section, medial is to the right, and the border between the core and shell of the NAc is indicated by the dotted white line. b, Quantification of pMeCP2 immunofluorescence intensity in NAc, dorsolateral striatum (DStr), and the prelimbic region of the prefrontal cortex (PrL). n = 5 WT, 6 KI; *p < 0.05, cocaine versus saline. The NAc sections were taken from the medial side of the rostral NAc and comprised both core and shell as shown in a. c, High-magnification image of a pMeCP2-positive neuron in the NAc shell after cocaine SA. Triple immunofluorescence is shown for pMeCP2 (red), the fast-spiking GABAergic interneuron marker GAD67 (blue), and the MSN marker DARPP32 (green). Scale bars: a, 60 μm; c, 10 μm.

Figure 3.

MeCP2 Ser421Ala KI mice show enhanced intake of self-administered cocaine. a, MeCP2 Ser421Ala KI and WT mice both rapidly acquire food training in the operant chambers. Food training was performed under FR1, FR2, and FR5 schedules with a 20 s time-out period after each press on the active lever. The criterion for success at each schedule was defined as obtaining 10 rewards in 1 h. The percentage of individuals reaching this criterion is shown each day (n = 12/genotype). b, c, Number of active and inactive lever presses (b) and number of rewards received (c) for MeCP2 Ser421Ala KI mice and their WT littermates during each daily self-administration session at the training dose of cocaine; n = 5 WT, 7 KI. d, Self-administration dose–response curve. Mean rewards received over the last 3 d are shown for each dose. *p = 0.05, KI versus WT at 0.1 mg/kg per injection; ^#p < 0.05, KI versus WT at 0.3 mg/kg per injection; n = 8/group.

Figure 4.

Depressed MSN excitability in MeCP2 Ser421Ala KI mice exposed to 2 d of repeated AMPH. a, Representative spike traces from MSNs in the NAc shell from striatal slices of AMPH-sensitized MeCP2 WT and Ser421Ala KI littermates at the current injections indicated. Calibration: 10 ms, 20 mV. b, Mean spike number generated for a given magnitude of current injection in AMPH-sensitized MeCP2 WT and Ser421Ala KI littermates. The mice used for these recordings represent a subset of the mice whose locomotor activities are shown in Figure 1e. c, Rheobase current in AMPH-sensitized MeCP2 Ser421Ala KI mice and their WT littermates. n = 18 cells from 3 WT mice; 19 cells from 3 KI mice; *p = 0.026 KI versus WT. d, Mean spike number generated for a given magnitude of current injection in psychostimulant-naive MeCP2 Ser421Ala KI mice and WT littermates. e, Rheobase current in psychostimulant-naive MeCP2 Ser421Ala KI mice and their WT littermates. p = 0.71 KI versus WT; n = 21 cells from 3 WT mice; 21 cells from 3 KI mice.

Figure 5.

Reduced CREB expression in the NAc MeCP2 Ser421Ala KI mice after cocaine self-administration. a, Western blot of CREB expression after chronic cocaine SA in the NAc of MeCP2 Ser421Ala WT and KI mice. Actin was used as a loading control to normalize CREB expression. n = 6 WT and 6 KI mice; *p = 0.016. b, Western blot of CREB expression in the NAc of psychostimulant-naive MeCP2 Ser421Ala WT and KI mice. Actin was used as a loading control to normalize CREB expression. n = 5 WT and 7 KI mice.

Figure 6.

Decreased Fos expression after repeated AMPH exposure in GAD67-positive neurons of the NAc. a, Representative images of Fos (red) and GAD67 (green) immunoreactivity in the NAc (core and shell) of MeCP2 WT mice and their Ser421Ala KI littermates 2 h after a single dose of either saline (Veh) or 3 mg/kg AMPH in the open field. AC, Anterior commissure. The dotted boxes show the regions enlarged in the overlay images to the right. Arrows indicate Fos and GAD67 double-positive neurons. Scale bars: left, 100 μm; right, 20 μm. b, Quantification of Fos immunoreactivity in GAD67-positive neurons of the NAc (core and shell) 2 h after either a single dose or 3 mg/kg AMPH (i.p.), or the second in a 2 d series of AMPH injections in the open field. The animals used for this study represent an independent cohort from those used for the quantification of behavioral sensitization in Figure 1e. Data are graphed as the percentage of integrated Fos intensity in each condition relative to the induction of Fos in GAD67-positive neurons of the WT mice on Day 1. n = 6 WT and 8 KI mice; *p = 0.008, WT versus KI on Day 2. c, Quantification of Fos immunoreactivity in all cells of the NAc from the same mice shown in b.