Alterations in the cholinergic system of brain stem neurons in a mouse model of Rett syndrome

Abstract

Rett syndrome is an autism-spectrum disorder resulting from mutations to the X-linked gene, methyl-CpG binding protein 2 (MeCP2), which causes abnormalities in many systems. It is possible that the body may develop certain compensatory mechanisms to alleviate the abnormalities. The norepinephrine system originating mainly in the locus coeruleus (LC) is defective in Rett syndrome and Mecp2-null mice. LC neurons are subject to modulation by GABA, glutamate, and acetylcholine (ACh), providing an ideal system to test the compensatory hypothesis. Here we show evidence for potential compensatory modulation of LC neurons by post- and presynaptic ACh inputs. We found that the postsynaptic currents of nicotinic ACh receptors (nAChR) were smaller in amplitude and longer in decay time in the Mecp2-null mice than in the wild type. Single-cell PCR analysis showed a decrease in the expression of α3-, α4-, α7-, and β3-subunits and an increase in the α5- and α6-subunits in the mutant mice. The α5-subunit was present in many of the LC neurons with slow-decay nAChR currents. The nicotinic modulation of spontaneous GABAA-ergic inhibitory postsynaptic currents in LC neurons was enhanced in Mecp2-null mice. In contrast, the nAChR manipulation of glutamatergic input to LC neurons was unaffected in both groups of mice. Our current-clamp studies showed that the modulation of LC neurons by ACh input was reduced moderately in Mecp2-null mice, despite the major decrease in nAChR currents, suggesting possible compensatory processes may take place, thus reducing the defects to a lesser extent in LC neurons.

Keywords: Mecp2; Rett syndrome; acetylcholine; compensatory mechanisms; locus coeruleus; nicotinic acetylcholine receptor.

Fig. 1.

The nicotinic acetylcholine (ACh) receptor (nAChR) currents were altered in locus coeruleus (LC) neurons from methyl-CpG binding protein 2 (Mecp2)^−/Y mice. A: a 100 μM ACh pulse (1 s, 44 nl) elicited a large transient current with a short decay time (τ) in a LC neuron from a wild type (WT) mouse. B: a 100 μM ACh pulse (1 s, 44 nl) elicited currents with a small amplitude and a long τ in a LC neuron from a Mecp2^−/Y mouse. C: the current amplitude was smaller in Mecp2^−/Y mice than in WT mice. D: the current density was smaller in Mecp2^−/Y mice than in WT mice. E: the τ was longer in Mecp2^−/Y mice than in WT mice. The τ was determined using a single-exponential equation. F: the area under the current curve was calculated by multiplying the amplitude by the time and was similar between both groups. Values are means ± SE; n = 31 WT cells and n = 11 Mecp2^−/Y cells. *P < 0.05, ***P < 0.001 (Mann-Whitney).

Fig. 2.

Nicotine (Nic) elicited a similar current to ACh in WT and Mecp2^−/Y mice. A: a 100 μM Nic pulse (1 s, 44 nl) elicited a large transient current with a short τ in a LC neuron from a WT mouse. B: a 100 μM Nic pulse (1 s, 44 nl) elicited a small-amplitude current with a long τ in a LC neuron from a Mecp2^−/Y mouse. C: the current amplitude was smaller in Mecp2^−/Y mice than in WT mice. D: the current density was smaller in Mecp2^−/Y mice than in WT mice. E: the τ is longer in Mecp2^−/Y mice than in WT mice. The τ was determined using a single-exponential equation. F: the area under the current curve calculated by multiplying the amplitude by the time was not significantly different between both groups. Values are means ± SE; n = 10 WT cells and n = 12 Mecp2^−/Y cells. **P < 0.01, ***P < 0.001 (Mann-Whitney).

Fig. 3.

Receptor subunit expression in identified LC neurons. Cytoplasm was extracted from the LC neuron after local perfusion with either ACh or Nic. A: nAChR expression from one WT LC neuron. B: a second LC neuron showing a different expression pattern than in A. C: nAChR expression in a LC neuron from a Mecp2^−/Y mouse. Black bar is equal to 200 base pairs. D: number of cells that contained each nAChR subunit (n = 19 WT cells, and n = 18 Mecp2^−/Y cells).

Fig. 4.

LC neurons from Mecp2^−/Y mice expressing the α₅-subunit had longer τ values than β₃-expressing neurons. Open bars are data from WT mice, and solid bars are data from Mecp2^−/Y mice. A: the nAChR current in LC neurons from Mecp2^−/Y mice displayed longer τ values than that from WT mice; n = 19 cells and n = 18 cells (Mann-Whitney). B: α₅-expressing LC neurons from Mecp2^−/Y mice had a long τ. β₃-expressing LC neurons from WT and Mecp2^−/Y mice had short τ values (n = 6 Mecp2^−/Y cells with α₅; n = 14 WT cells with β₃, n = 6 Mecp2^−/Y cells with β₃; one-way ANOVA) C: LC neurons from Mecp2^−/Y mice that expressed α₉α₅-receptors had longer τ values than neurons that expressed α₉β₃-receptors (n = 5 α₉α₅ and n = 5 α₉β₃; Mann-Whitney). Values are means ± SE. *P < 0.05. **P < 0.01.

Fig. 5.

Presence of the α₇-subunit affects the current amplitude and τ in LC neurons from WT mice. A: cells expressing the α₇-subunit have a larger whole cell nAChR current amplitude in WT mice than cells not expressing α₇. B: cells expressing the α₇-subunit have a shorter whole cell nAChR current τ in WT mice than cells not expressing α₇. Values are means ± SE; n = 14 cells with α₇ and n = 5 cells with no α₇. *P < 0.05, **P < 0.01 (Student's t-test).

Fig. 6.

Cholinergic modulation of GABAergic input to LC neurons with nAChR agonist in Mecp2^−/Y mice is enhanced compared with WT mice. A: LC neurons of WT mice were patch clamped, and GABAergic spontaneous inhibitory postsynaptic currents (sIPSCs) were recorded during the 5-min baseline (BL) and during a 5-min 10 μM 1,1-dimethyl-4-phenylpiperazinium iodide (DMPP) treatment. B: LC neurons of Mecp2^−/Y mice were patch clamped, and recordings were taken during the 5-min BL recording of GABAergic sIPSCs and during a 5-min 10 μM DMPP treatment. C: the frequency increase seen in the WT mice was enhanced in Mecp2^−/Y mice. **P < 0.01 (Mann-Whitney). D: the amplitude did not show a significant difference during DMPP treatment. There was no difference between WT and Mecp2^−/Y mice as well. E: cumulative fraction of the interevent interval before and during 10 μM DMPP treatment. Dashed lines are from BL, and the solid lines are from the 10 μM DMPP treatment. Shaded lines are from WT LC neurons, and black lines are from Mecp2^−/Y LC neurons. Shift to the left during DMPP treatment in Mecp2^−/Y mice was greater than in WT mice. F: cumulative fraction of amplitude before and during 10 μM DMPP treatment. Dashed lines are from BL, and the solid lines are from the 10 μM DMPP treatment. Shaded lines are from WT LC neurons, and black lines are from Mecp2^−/Y LC neurons. n = 8 WT cells and n = 8 Mecp2^−/Y cells.

Fig. 7.

Cholinergic modulation of GABA inputs to the LC neurons with a nAChR antagonist is enhanced in Mecp2^−/Y mice. A: LC neurons of WT mice were patch clamped, and GABAergic sIPSCs were recorded during the 5-min BL and during a 5-min 30 μM mecamylamine (Mec) treatment. B: LC neurons of Mecp2^−/Y mice were patch clamped, and recordings were taken during the 5-min BL recording of GABAergic sIPSCs and during a 5-min exposure to 30 μM Mec. C: the frequency decrease seen in the WT mice was enhanced in Mecp2^−/Y mice. **P < 0.01 (Mann-Whitney). D: the amplitude showed no significant difference during DMPP treatment. There was no difference between WT and Mecp2^−/Y mice. Values are means ± SE; n = 9 WT cells and n = 17 Mecp2^−/Y cells. E: cumulative fraction of the interevent interval before and during 30 μM Mec treatment. Dashed lines are from BL, and the solid lines are from the 30 μM Mec treatment. Shaded lines are from recordings of WT LC neurons, and black lines are from Mecp2^−/Y LC neurons. The shift to the right during Mec treatment in Mecp2^−/Y mice was greater than in WT mice. F: cumulative fraction of amplitude before and during 30 μM Mec treatment. Dashed lines are from BL, and the solid lines indicate the 30 μM Mec treatment. Shaded lines are from WT LC neurons, and black lines are from Mecp2^−/Y LC neurons. There is no shift between the BL and drug treatment for both groups.

Fig. 8.

There is a small but insignificant difference in nicotinic modulation of LC neurons from WT and Mecp2^−/Y mice. A: patch-clamp recordings of LC neuronal firing activity in a WT mouse before, during, and after 10 μM DMPP treatment. Each inset is for the hyperpolarizing current indicating the input resistance. B: patch-clamp recordings of LC neuronal firing activity in a Mecp2^−/Y mouse before, during, and after 10 μM DMPP treatment. Each inset is for the hyperpolarizing current, indicating the input resistance. C: frequency was analyzed from the recording made in A. This showed that 10 μM DMPP increased the frequency in the WT mouse. D: frequency was analyzed from the recording in B, and 10 μM DMPP increased the frequency in Mecp2^−/Y mice. E: there was no significant difference in the increase in firing frequency between WT and Mecp2^−/Y mice. F: the difference in input resistance modulation by 10 μM DMPP between WT and Mecp2^−/Y mice was insignificant. G: 10 μM DMPP depolarized the LC neuron in WT and Mecp2^−/Y mice. There was no significant difference between groups. Values are means ± SE; n = 11 WT cells and n = 13 Mecp2^−/Y cells. ns, Not significant (Mann-Whitney).