Loss of MeCP2 from forebrain excitatory neurons leads to cortical hyperexcitation and seizures

Abstract

Mutations of MECP2 cause Rett syndrome (RTT), a neurodevelopmental disorder leading to loss of motor and cognitive functions, impaired social interactions, and seizure at young ages. Defects of neuronal circuit development and function are thought to be responsible for the symptoms of RTT. The majority of RTT patients show recurrent seizures, indicating that neuronal hyperexcitation is a common feature of RTT. However, mechanisms underlying hyperexcitation in RTT are poorly understood. Here we show that deletion of Mecp2 from cortical excitatory neurons but not forebrain inhibitory neurons in the mouse leads to spontaneous seizures. Selective deletion of Mecp2 from excitatory but not inhibitory neurons in the forebrain reduces GABAergic transmission in layer 5 pyramidal neurons in the prefrontal and somatosensory cortices. Loss of MeCP2 from cortical excitatory neurons reduces the number of GABAergic synapses in the cortex, and enhances the excitability of layer 5 pyramidal neurons. Using single-cell deletion of Mecp2 in layer 2/3 pyramidal neurons, we show that GABAergic transmission is reduced in neurons without MeCP2, but is normal in neighboring neurons with MeCP2. Together, these results suggest that MeCP2 in cortical excitatory neurons plays a critical role in the regulation of GABAergic transmission and cortical excitability.

Keywords: GABA; Rett syndrome; hyperexcitation; neocortex; pyramidal neuron; seizure.

Figure 1.

Conditional deletion of Mecp2 with Emx1-Cre leads to spike-wave discharges. A, Seizure activity in an Emx1-Mecp2 mutant mouse at 7 weeks of age. EEG was recorded from a wake Emx1-Mecp2 mouse using four electrodes placed over the right frontal (RF), left frontal (LF), right back (RB), and left back (LB) surface of the neocortex. Each trace is the differential between signals recorded from two of the electrodes. B, EEG recording from a control mouse (Mecp2^wt/y; Emx1-Cre) at 7 weeks of age. C, EEG recording from a Dlx6a-Mecp2 mutant mouse at 7 weeks of age.

Figure 2.

Deletion of Mecp2 with Emx1-Cre caused reduction in the frequency of sIPSCs and mIPSCs in layer 5 pyramidal neurons in the mPFC at P17–P20. A, sIPSCs recorded from a control (top trace) and mutant (bottom trace) neurons at P18. B, Distributions of sIPSC intervals for mutant neurons (Mecp2^F/y; Emx1-Cre, line in gray, n = 25 from 4 mice) and control neurons (Mecp2^F/y, dashed line, n = 15 from 2 mice; Mecp2^wt/y; Emx1-Cre, line in black, n = 27 from 4 mice). For each group, the distribution was established using 200 consecutive events from each of the recorded cells; p = 0.0007, K–S test. C, The mean frequency of sIPSCs for the three groups (4.9 ± 0.3 Hz for Mecp2^F/y;Emx1-Cre; 6.7 ± 0.4 Hz for Mecp2^wt/y;Emx1-Cre; 7.1 ± 0.3 Hz for Mecp2^F/y; ***p < 0.001). D, The mean peak amplitude of sIPSCs for the three groups (34.2 ± 1.1 pA for Mecp2^F/y;Emx1-Cre; 34.6 ± 1.4 pA for Mecp2^wt/y;Emx1-Cre; and 36.7 ± 2.2 pA for Mecp2^F/y; p > 0.3). The mean decay constant of sIPSC was 5.6 ± 0.3 ms for mutant neurons (Mecp2^F/y; Emx1-Cre), 5.5 ± 0.2 ms for Mecp2^F/y control, and 5.8 ± 0.2 ms for Mecp2^wt/y; Emx1-Cre control (p = 0.53, one-way ANOVA). E, Distributions of mIPSC intervals for mutant neurons (Mecp2^F/y; Emx1-Cre, gray) and control neurons (Mecp2^wt/y; Emx1-Cre, black). p = 0.0003, K–S test. The mean frequency of mIPSC was 4.5 ± 0.2 Hz (n = 22 from 3 mice) for mutant neurons, and 6.4 ± 0.5 Hz (n = 24 from 3 mice) for control neurons (p = 0.001). F, Distributions of mIPSC amplitude for mutant neurons (Mecp2^F/y; Emx1-Cre, gray) and control neurons (Mecp2^wt/y; Emx1-Cre, black). The mean peak amplitude of mIPSC was 28.1 ± 1.2 pA for mutant neurons, and 28.5 ± 1.3 pA for control neurons (p = 0.68). G, Averaged mIPSCs for mutant and control neurons. The mean decay constant of mIPSC was 6.1 ± 0.2 ms for mutant neurons, and 5.7 ± 0.2 ms for control neurons (p = 0.08).

Figure 3.

Deletion of Mecp2 with Emx1-Cre had no effect on excitatory transmission in the prefrontal cortex. A, mEPSCs recorded from a control (top trace, black) and mutant (bottom trace, gray) layer 5 pyramidal neurons in mPFC at P18 in presence of TTX (0.3 μm) and picrotoxin (100 μm). B, Distributions of mEPSC intervals for mutant (black) and mutant (gray) neurons. The mean frequency was 6.3 ± 0.6 Hz for control neurons (n = 24 from 3 mice) and 6.4 ± 0.6 Hz for mutant neurons (n = 21 from 3 mice; p = 0.71). C, Distributions of mEPSC peak amplitude for control (black) and mutant (gray) neurons. The mean peak amplitude was 11.8 ± 0.4 pA for control and 12.0 ± 0.3 pA for mutant neurons (p = 0.99). Inset, Averaged mEPSCs for control (black) and mutant (gray) neurons. Scale bars, 2 pA and 2 ms. The mean decay constant of mEPSCs was 3.7 ± 0.1 ms for control and 3.8 ± 0.1 ms for mutant neurons (p = 0.89).

Figure 4.

Deletion of Mecp2 with Emx1-Cre reduced GABAergic transmission in the somatosensory cortex. A, sIPSCs recorded from a control (top trace, black) and mutant (bottom trace, gray) layer 5 pyramidal neurons in S1 at P18. B, Distributions of sIPSC intervals for mutant neurons (Mecp2^F/y; Emx1-Cre, gray) and control neurons (Mecp2^wt/y; Emx1-Cre, black); **p < 0.001, K–S test. The mean frequency of sIPSCs for control (8.7 ± 0.4 Hz, n = 17 cells from 3 mice) and mutant neurons (6.5 ± 0.6 Hz, n = 19 cells from 3 mice; **p = 0.004). C, Distributions of sIPSC amplitude for mutant neurons (Mecp2^F/y; Emx1-Cre, gray) and control neurons (Mecp2^wt/y; Emx1-Cre, black). The peak amplitude of sIPSCs for control (37.8 ± 2.3 pA) and mutant neurons (34.3 ± 1.5 pA; p = 0.13). D, Averaged sIPSCs for control and mutant neurons. The mean decay constant of sIPSCs was 5.6 ± 0.2 ms for control neurons, and 4.9 ± 0.1 ms for mutant neurons (p = 0.04).

Figure 5.

Selective deletion of Mecp2 from forebrain GABAergic neurons had no effect on spontaneous GABAergic transmission. A, tdTomato signal and GABA immunostaining in the mPFC of a Dlx6a-Cre; Ai14 mouse. Scale bar, 20 μm. B, tdTomato signal and MeCP2 immunostaining in the mPFC of a Mecp2^F/y; Dlx6a-Cre; Ai14 mouse. Scale bar, 20 μm. C, sIPSCs recorded from a control (Mecp2^F/y) and mutant (Mecp2^F/y; Dlx6a-Cre) neurons at P18. D, Distributions of sIPSC intervals for control (black) and mutant (gray) neurons. The mean frequency of sIPSCs was 6.3 ± 0.4 Hz (n = 15 from 3 mice) for control neurons, and 6.0 ± 0.3 Hz (n = 19 from 3 mice) for mutant neurons (p = 0.82). E, Distributions of sIPSC amplitude for control and mutant neurons. Inset, Averaged sIPSCs from control and mutant neurons. Scale bars, 10 ms and 10 pA. The mean amplitude of sIPSCs was 38.1 ± 3.0 pA for control neurons, and 36.9 ± 1.9 pA for mutant neurons (p = 0.8). The mean decay constant was 6.2 ± 0.3 ms for control neurons, and 5.7 ± 0.2 ms for mutant neurons (p = 0.2).

Figure 6.

Mosaic deletion of Mecp2 in the cortex selectively disrupts GABAergic transmission in neurons deficient in MeCP2. A, confocal image of the S1 in coronal section of a SERT-Ai14 mouse at P21. Neurons were labeled with anti-NeuN (green); Cre⁺ cells were identified with tdTomato signal (red). The lines and numbers on the side indicate the locations of the six cortical layers. Scale bar, 200 μm. B, Layer 1 and upper layer 2/3 of S1 at a higher-magnification. Scale bar, 100 μm. Approximately 19% of neurons in upper layer 2/3 were Cre⁺. C, Layer 2/3 of S1 immunostained for GABA (green). The vast majority of Cre⁺ neurons (red) in layer 2/3 were pyramidal neurons and negative for GABA. Scale bar, 20 μm. D, sIPSCs recorded from a Cre⁻ (top trace, black) and a Cre⁺ (bottom trace, red) pyramidal neurons in S1 layer 2/3 of a SERT-Mecp2-Ai14 mutant mouse at P18. E, Mean frequency of sIPSCs from Cre⁻ and Cre⁺ layer 2/3 pyramidal neurons in SERT-Mecp2-Ai14 mutant and SERT-Ai14 control mice at P17–P18. For mutant mice, the mean frequency was 7.6 ± 0.4 Hz (n = 18, for Cre⁻ neurons, black) and 3.6 ± 0.3 Hz (n = 17, for Cre⁺ neurons, red; ***p < 0.0005). For control mice, the mean frequency was 8.3 ± 0.7 Hz (n = 17, for Cre⁻ neurons, blue) and 7.8 ± 0.7 Hz (n = 17, for Cre⁺ neurons, purple). F, Mean peak amplitude of sIPSCs for the four groups. The mean peak amplitude was 31.6 ± 1.6 pA (Cre⁻ neurons in mutant mice), 25.0 ± 1.6 pA (Cre⁺ neurons in mutant mice), 33.3 ± 1.8 pA (Cre⁻ neurons in control mice), and 34.1 ± 1.9 pA (Cre⁺ neurons in control mice); ***p < 0.001. G, Averaged sIPSCs for Cre⁻ (in black) and Cre⁺ (in red) neurons in mutant mice. The decay time constant of sIPSCs was 7.5 ± 0.2 ms (Cre⁻ neurons in mutant mice), 8.4 ± 0.2 ms (Cre⁺ neurons in mutant mice), 7.4 ± 0.1 ms (Cre⁻ neurons in control mice), and 7.6 ± 0.2 ms (Cre⁺ neurons in control mice). The decay time of Cre⁺ neurons in mutant mice was significantly slower than the other three groups (**p = 0.002).

Figure 7.

Evoked GABAergic transmission and the number of GABAergic synapses were reduced in the mPFC of Emx1-Mecp2 mutant mice. A, Monosynaptic IPSCs recorded from control and mutant neurons evoked by local stimulation. B, The input–output curve of IPSCs for control (n = 13 from 3 mice) and mutant (n = 14 from 3 mice) neurons. C, Paired pulse responses for IPSCs from control and mutant neurons. The averaged IPSCs from control and mutant neurons were normalized to the peak of the first IPSC. The paired pulse ratio at 100 ms interval was 0.89 ± 0.02 (n = 10 from 3 mice) for mutant neurons, and 0.85 ± 0.03 (n = 11 from 3 mice) for control neurons (p = 0.14). D, Double immunostaining of VGAT (red) and NeuN (green) in the mPFC of control and mutant mice at P18 (top) and 7 weeks of age (bottom). Scale bar, 10 μm. E, Numbers of VGAT puncta per 1000 μm³ volume in layer 5 of mPFC of mutant and control mice at P18–P19 and 7–8 weeks of age. Data were obtained from four pairs of mice at P18–P19 and 7–8 weeks of age, respectively; three measurements were performed for each mouse. P values were obtained with Mann–Whitney test.

Figure 8.

Hyperexcitation of cortical neurons in Emx1-Mecp2 mutant mice. A, Neuronal activity at the resting potential (−58, −59 mV) and at −53 mV in a control (Mecp2^wt/y; Emx1-Cre) and mutant (Mecp2^F/y; Emx1-Cre) neurons. B, Frequency of action potential of control and mutant neurons at the resting potential and at −53 mV. The mean firing frequency at −53 mV was 0.3 ± 0.2 Hz for control (n = 21 from 3 mice), and 1.6 ± 0.4 Hz for mutant (n = 26 from 3 mice; ***p < 0.001). C, Spontaneous IPSCs recorded from wild-type neurons without and with picrotoxin (0.5 mm) in the patch pipette. These recordings were obtained at 0 mV with the pipette solution containing Cs-methylsulfate ([Cl^-]_i of 35 mm). D, Firing recorded from a control and mutant neurons at −53 mV with picrotoxin in the patch pipettes. E, Firing frequency of control and mutant neurons at −60 and −53 mV with intracellular picrotoxin. The mean firing frequency at −53 mV was 3.0 ± 0.6 Hz for control (n = 24 from 3 mice), and 2.7 ± 0.5 Hz for mutant (n = 23 from 3 mice; p = 0.86). F, Membrane potential changes in response to current steps in control and mutant neurons. Synaptic transmission was blocked by bath perfusion of DNQX, kynurenic acid, and picrotoxin. G, Plot of spike frequency versus injected current for control (n = 27 from 3 mice) and mutant neurons (n = 31 from 3 mice).