MECP2e1 isoform mutation affects the form and function of neurons derived from Rett syndrome patient iPS cells

Abstract

MECP2 mutations cause the X-linked neurodevelopmental disorder Rett Syndrome (RTT) by consistently altering the protein encoded by the MECP2e1 alternative transcript. While mutations that simultaneously affect both MECP2e1 and MECP2e2 isoforms have been widely studied, the consequence of MECP2e1 deficiency on human neurons remains unknown. Here we report the first isoform-specific patient induced pluripotent stem cell (iPSC) model of RTT. RTTe1 patient iPS cell-derived neurons retain an inactive X-chromosome and express only the mutant allele. Single-cell mRNA analysis demonstrated they have a molecular signature of cortical neurons. Mutant neurons exhibited a decrease in soma size, reduced dendritic complexity and decreased cell capacitance, consistent with impaired neuronal maturation. The soma size phenotype was rescued cell-autonomously by MECP2e1 transduction in a level-dependent manner but not by MECP2e2 gene transfer. Importantly, MECP2e1 mutant neurons showed a dysfunction in action potential generation, voltage-gated Na(+) currents, and miniature excitatory synaptic current frequency and amplitude. We conclude that MECP2e1 mutation affects soma size, information encoding properties and synaptic connectivity in human neurons that are defective in RTT.

Keywords: MeCP2; Rett syndrome; iPS cell disease models.

Figure 1. Generation and pluripotency characterization of RTTe1-hiPSCs

(A) Androgen receptor (AR) methylation screen of RTTe1-hiPSCs demonstrates skewed XCI ratios in all examined lines. Bar graph depicts XCI ratio of two X chromosomes (X1 and X2). Fib., fibroblasts. (B) qRT-PCR analyses of pMXs reprogramming retroviral transgenes. (C) qRT-PCR of endogenous pluripotency genes in RTTe1-hiPSC lines. Data are expressed as mean ± SEM. (D) Bisulfite sequencing of retroviral pMXs-LTR reprogramming vectors in RTTe1-hiPSC lines. Open and closed CpG sites indicate unmethylated and methylated CpG sites, respectively. (E) RTTe1-hiPSCs differentiate into the three germ layers in vitro and in vivo (representative images of RTTe1-hiPSC #39 teratomas). Scale bars, 50 μm (immunocytochemistry) and 100 μm (histology).

Figure 2. RTTe1-hiPSCs and -neurons retain an Xi and exclusively express the mutant MECP2e1 allele

(A) cDNA sequencing of MECP2e1 transcripts in selected RTTe1-hiPSCs reveals that the expressing X-chromosome contains the 11 bp RTTe1 deletion (left panel), with representative chromatogram (right panel). (B) Differentiated RTTe1-neurons maintain expression of the mutant MECP2 allele (left panel) and share the same inactive X chromosome revealed by the AR assay (right panel).

Figure 3. RTTe1 neurons maintain normal balance of neuronal identities

(A) Transcription of both MECP2 isoforms increases over the course of neuronal differentiation. (B) MeP-MECP2e1 and EF1α-MECP2e1 lentiviral constructs are expressed in mature RTTe1#27 neurons. Data are expressed as mean ± SEM. (C) Western blot analysis of transduced RTTe1#27 cells shows that MeP-MECP2e1 rescue results in a moderate increase of total MECP2 protein, and detectable levels of the MYC tagged MECP2e1 with unchanged levels of MECP2e2 protein. Transduction with EF1α-MECP2e1 construct leads to overexpression of total MECP2 protein and higher levels of MYC tagged MECP2e1. Densitometry bar graph is shown with normalization to loading control, histone H3. (D) Bar graphs show comparable percentage of cells expressing majority of neuronal markers as determined by Fluidigm arrays in RTTe1-mock (cumulative data from all 4 RTTe1 lines) and MeP rescued RTTe1#27 cells. Data are expressed as mean ± SEM; * p-value<0.05. (E) Similar percentages of cortical layer and neurotransmitter neurons are produced from both RTTe1-mock and MeP rescue cells. Data are expressed as mean ± SEM; * p-value<0.05.

Figure 4. RTTe1-neurons exhibit a soma size defect that can be rescued by MECP2e1 in a cell-autonomous manner

(A) Immunocytochemistry for MAP2 and MECP2 in WT Δ3-4#37 and RTTe1#27 neurons. Bar graph represents soma size analysis of neurons derived from individual RTTe1 hiPS cell lines compared to WT neurons derived by three differentiation protocols utilized in the study. (B) Immunocytochemistry for MAP2 and MYC in RTTe1#27-neurons with or without MECP2e1-vectors. Bar graph shows cumulative soma size analysis of all four RTTe1-neurons, and RTTe1#27 neurons with MECP2e1-vectors compared to WT-neurons. Total number of measured neurons for each analyzed genotype is indicated within the appropriate bar. Data are expressed as mean ± SEM (*P < 0.001; Student’s t-test; n = at least 3 independent differentiations). Scale bars, 44 μm for large image, 10 μm for inset. Arrows, MYC-positive neurons. Arrowheads, MYC-negative neurons.

Figure 5. Soma-size rescue is MECP2e1 level-dependent

(A) Immunocytochemistry for MAP2 and MECP2 or MAP2 and MYC in RTTe1-neurons with or without MECP2e2-vectors and WT-neurons. Scale bars, 44 μm for large image. Arrows, MYC-positive neurons. Arrowheads, MYC-negative neurons. (B) Soma size analysis of RTTe1-neurons with or without MECP2e1 (control) or MECP2e2 vectors compared to WT-neurons. Number of measured neurons for each analyzed genotype is indicated within the appropriate bar. (C) qRT-PCR of total MECP2 and the lenti-derived MECP2 transcripts in cells transduced with either MECP2e1 or MECP2e2 vectors. (D) Soma size analysis in cells expressing low or high levels of MECP2, based on immunostaining intensities of the MYC signal. Data are expressed as mean ± SEM (*P < 0.001; Student’s t-test; n = 3 independent differentiations).

Figure 6. RTTe1-neurons exhibit alterations in intrinsic membrane properties

(A) Histogram shows average input resistance in WT- and RTTe1-neurons. (B) Bar graph showing average cell capacitance in WT-neurons compared with RTTe1-neurons. (C) Representative traces show evoked action potentials triggered by injecting a series of current steps from −5 pA to +50 pA in WT- (Upper) and RTTe1- (Bottom) neurons. (D) A plot showing the numbers of action potentials evoked by injecting a series of current steps from +5 pA to +50 pA in WT- and RTTe1-neurons. (E) A plot depicting current-voltage relationships between WT- and RTTe1-neurons. Peak average inward currents (at −20 mV) were compared between WT- and RTTe1-neurons. *P<0.05, **P<0.01, ***P<0.001.

Figure 7. RTTe1-neurons exhibit decreased mEPSC frequency and amplitude

(A) Representative traces showing mEPSCs in WT- (Upper) and RTTe1- (Bottom) neurons, and the inset showing averaged mEPSCs in the WT- (Upper) and RTTe1- (Bottom) neurons, respectively. (B) Bar graph shows average mEPSC amplitude in WT-neurons compared with RTTe1-neurons. (C) Histogram showing average mEPSC frequency in WT-neurons compared with RTTe1-neurons. *P<0.05, ***P<0.001.