Multiplex epigenome editing of MECP2 to rescue Rett syndrome neurons

Abstract

Rett syndrome (RTT) is an X-linked neurodevelopmental disorder caused by loss-of-function heterozygous mutations of methyl CpG-binding protein 2 (MECP2) on the X chromosome in young females. Reactivation of the silent wild-type MECP2 allele from the inactive X chromosome (Xi) represents a promising therapeutic opportunity for female patients with RTT. Here, we applied a multiplex epigenome editing approach to reactivate MECP2 from Xi in RTT human embryonic stem cells (hESCs) and derived neurons. Demethylation of the MECP2 promoter by dCas9-Tet1 with target single-guide RNA reactivated MECP2 from Xi in RTT hESCs without detectable off-target effects at the transcriptional level. Neurons derived from methylation-edited RTT hESCs maintained MECP2 reactivation and reversed the smaller soma size and electrophysiological abnormalities, two hallmarks of RTT. In RTT neurons, insulation of the methylation-edited MECP2 locus by dCpf1-CTCF (a catalytically dead Cpf1 fused with CCCTC-binding factor) with target CRISPR RNA enhanced MECP2 reactivation and rescued RTT-related neuronal defects, providing a proof-of-concept study for epigenome editing to treat RTT and potentially other dominant X-linked diseases.

Fig. 1.. Reactivation of the MECP2 reporter on Xi by DNA methylation editing.

(A) Scheme of genetically engineered MECP2 dual-color reporter hESC lines derived from a wild-type female hESC (NIH registration code: WIBR-1, #29) after methylation editing. For the 29-R cell line, GFP was inserted after MECP2 exon 3 in frame, followed by polyA termination signal on the inactive X chromosome (Xi), and tdTomato was inserted after MECP2 exon 3 in frame followed by poly A termination signal on the active X chromosome (Xa). For 29-G cell line, GFP is on Xa and tdTomato is on Xi. (B) Illustration of the differentially methylated region (DMR) between female and male hESCs at the MECP2 promoter. Ten sgRNA were designed to target this DMR, and four pyrosequencing (Pyro-seq) assays were designed to measure the DNA methylation of this DMR. TSS, transcription start site. (C) 29-R hESCs were infected with lentiviruses expressing dCas9-Tet1-P2A-BFP (dC-T) or dCas9-dead Tet1-P2A-BFP (dC-dT) with 10 sgRNAs with mCherry as a fluorescent marker targeting this DMR at the MECP2 promoter as illustrated in (B). The infection-positive cells (BFP⁺;mCherry⁺) were isolated by FACS and subjected to Pyro-seq analysis. Shown is the mean percentage ± SD of three biological replicates. (D) Immunofluorescence (IF) staining of cells described in (C) with antibodies against GFP and Cas9. Scale bars, 300 μm. (E) A Dox-inducible dCas9-Tet1 expression cassette was inserted into the 29-R hESCs via the PiggyBac transposon system (labeled as 29-R_138-dC-T), and dCas9-Tet1 expression was examined by reverse transcription qPCR (RT-qPCR) in response to Dox treatment. Shown is the mean ± SD of three biological replicates. (F) Cells in (E) were infected with individual target sgRNA (labeled as 1 to 10) or the mixture of 10 sgRNAs (labeled as 1–10) targeting the MECP2 promoter and then subjected to methylation analysis by Pyro-seq in the presence of Dox. Shown is the average DNA methylation of CpGs within the targeted MECP2 promoter region ± SD of three biological replicates. (G) RT-qPCR analysis of GFP reporter for MECP2 on Xi in cells in (F). GFP expression was normalized to 29-G cells with GFP reporter for MECP2 on Xa. Shown is the mean percentage ± SD of at least two biological replicates. *P < 0.05, one-way analysis of variance (ANOVA) with Bonferroni correction.

Fig. 2.. DNA methylation editing efficacy and off-target effects.

(A) A Manhattan plot showing 27 genome-wide binding sites of dCas9-Tet1 with the MECP2 target sgRNA-3 in 29-R cells identified by anti-Cas9 ChIP-seq. (B) DNA methylation of these 27 binding sites measured by anti-Cas9 ChIP-bisulfite-seq of 29-R cells expressing dCas9-Tet1 with sgRNA-3 or dCas9-dTet1 with sgRNA-3. MECP2 is labeled in black; five binding sites with a change of methylation larger than 16% are labeled in red. The diameter of a circle is in proportion to the number of matched base pairs to the sgRNA-3 target sites. The dashed lines mark the 16% methylation difference between samples. (C) Transcriptomes of cells in (B) by RNA-seq. Red dots highlight the genes associated with the 26 dCas9-Tet1 binding sites identified in (A). MECP2-GFP reporter on Xi is labeled with a black dot. Red dashed lines mark the twofold difference between the samples. (D to F) Mock 29-R, mock 29-G, or methylation-edited 29-R hESCs were differentiated into neuronal precursor cells (NPCs) and neurons for gene expression analysis by qPCR. Shown is the mean ± SD of three biological replicates. (G) Illustration showing the location of genes on the X chromosome prone to erosion of X chromosome inactivation in high passage of female hESC/iPSC. (H to N) Gene expression analysis of cells in (D) by qPCR: HPRT in (H), PRPS1 in (I), COL4A5 in (J), CASK in (K), RPGR in (L), PDHA1 in (M), and HCCS in (N). Shown is the mean ± SD of two biological replicates. *P < 0.05, one-way ANOVA with Bonferroni correction. The differences between 29-R and 29-R_dC-T + sgRNA in (H) to (N) were not significant (ns; P > 0.05).

Fig. 3.. Functional rescue of RTT neurons derived from edited RTT hESCs.

(A) Scheme of an RTT-like hESC line (#860) genetically engineered from a WT female human ESCs (NIH registration code: WIBR-3) after methylation editing. In this RTT-like cell line, the WT allele of MECP2 is on Xi, and the MECP2 null function allele is on Xa. (B) RTT-like hESCs (labeled as RTT) were infected with lentiviral dCas9-Tet1-P2A-BFP with either sgRNA-3 alone or 10 sgRNAs together. Infection-positive cells were isolated by FACS and subjected to RT-qPCR analysis of MECP2 expression with primers targeting the exon 4 region that will only be expressed from the WT allele on Xi but not the null function allele on Xa. The expression of MECP2 mRNA in these samples was normalized to WIBR-3 (labeled as WT). Shown is the mean ± SD of three biological replicates. (C) Western blot analysis of the neurons derived from the cells is described in (B). Protein abundance of MeCP2 was quantified by ImageJ and is shown as the mean of relative percentages as compared with WT neurons ± SD of two biological replicates. (D) Neurons in (C) were grown on mouse astrocytes to promote neuronal maturation and then IF-stained with anti-MeCP2 and anti-Map2 antibodies. Scale bar, 30 μm. (E) Soma sizes of neurons in (D) were quantified by ImageJ. (F) Neurons in (C) were grown on the MEA plate for measurement of electrophysiological activities along the neuronal maturation process. Shown is the mean ± SD of biological replicates for each group of neurons. (G) Neuronal activities of neurons in (F) on the day of maturation (day 58). (H to J) Representative trace images showing spontaneous synaptic events of neurons in (D). (K to M) The mEPSC frequency (K), mEPSC amplitude (L), and membrane capacitance (M) of neurons in (D). Shown is the mean ± SD of at least two biological repeats, with more than 20 neurons for each condition. *P < 0.05 and **P < 0.01, one-way ANOVA with Bonferroni correction.

Fig. 4.. Direct editing and reactivation of MECP2 in RTT neurons.

(A) Neurons derived from 29-R hESCs were infected with lentiviral dCas9-Tet1 (dC-T) and sgRNA-3 and then subjected to qPCR analysis on day 5 (D5), D10, D15, and D20 after infection. GFP expression was normalized to 29-G neurons with GFP reporter for MECP2 on Xa. Shown is the mean ± SD of biological triplicates for each time point. (B) Neurons derived from WT WIBR-3 hESCs (labeled as WT), mock RTT-like hESCs (labeled as RTT), and RTT-like hESCs edited by dCas9-Tet1/sgRNA-3 (labeled as RTT_ESC + d-T), or RTT neurons infected by lentiviral dCas9-Tet1 and sgRNA-3 (labeled as RTT_neuron + dC-T) were grown on the MEA plate for measurement of firing rates along the neuronal maturation process. Shown is the mean ± SD of biological triplicates for each condition. (C) MECP2 mRNA expression of the neurons in (B) measured by qPCR. (D) DNA methylation of the MECP2 promoter in neurons in (B) was measured by Pyro-seq on D60. Shown is the average DNA methylation of CpGs within the targeted MECP2 promoter region ± SD of three biological replicates. (E) Neurons derived from RTT-like hESCs were infected with lentiviral dCas9-Tet1 and sgRNA-3 (labeled dC-T) or dCas9-deadTet1 and sgRNA-3 (labeled as dC-dT) and were grown on mouse astrocytes to promote neuronal maturation and then IF-stained with anti-MeCP2 and anti-Tuj1 antibodies on D60. Scale bar, 30 μm. (F) Percentages of neurons (Tuj1⁺) expressing MeCP2 protein within each group of samples described in (E). (G) Quantification of MeCP2 protein abundance in the MeCP2-positive neurons in (E). (H) Quantification of soma sizes of the neurons in (E). Quantifications were done using ImageJ and are shown as the mean of relative percentages as compared with WT neurons ± SD of at least two biological replicates. *P < 0.05 and **P < 0.01, one-way ANOVA with Bonferroni correction.

Fig. 5.. Multiplex epigenome editing of MECP2 to rescue RTT neurons.

(A) Enrichment of CTCF binding around the MECP2 locus in the neurons derived from mock (labeled as RTT neuron_mock) or neurons derived from DNA methylation–edited RTT hESCs (labeled as RTT neuron_edited) and primed or native RTT hESCs. 4C-seq was performed to reveal the genomic interactions of the MECP2 promoter. The two CTCF anchor sites are labeled by rectangles in black. (B) Top illustrates the all-in-one lentiviral construct to express Dox-inducible dCpf1-CTCF-HA (human influenza hemagglutinin tag) and target crRNAs. Bottom shows an anti-Cpf1 ChIP-seq using human embryonic kidney 293T cells transfected with empty vector or the dCpf1-CTCF construct with two crRNAs targeting the anchor sites in (A). LTR, long terminal repeat. (C) A Manhattan plot showing 28 genome-wide binding sites of dCpf1-CTCF with the MECP2 target crRNAs identified by anti-Cpf1 and anti-CTCF ChIP-seq. (D) Transcriptomes of cells transfected with dCpf1-CTCF + MECP2 target crRNAs or dCpf1-CTCF + scrambled crRNA examined by RNA-seq. Red dots highlight the genes associated with the 28 dCpf1-CTCF binding sites identified in (C). MECP2 is labeled with a black dot. The red dashed lines mark the twofold difference between the samples. (E) WT neurons, mock RTT neurons, or RTT neurons infected by lentiviral dCas9-Tet1/sgRNA-3 (labeled as dC-T), or dCpf1-CTCF (dC-C), or both (dC-T + dC-C) were grown on the MEA plate for measurement of electrophysiological activities in a time course experiment. Shown is the mean ± SD of biological triplicates for each group of neurons. (F) MECP2 mRNA quantity of the neurons in (E) was measured by qPCR on day 57. Shown is the mean ± SD of biological triplicates for each group of neurons. (G) DNA methylation of the MECP2 promoter in neurons in (E) was measured by Pyro-seq on day 57. Shown is the average DNA methylation of CpGs within the targeted MECP2 promoter region ± SD of three biological replicates. (H) Neurons in (E) were grown on mouse astrocytes to promote neuronal maturation and then IF-stained with anti-MeCP2 and anti-Tuj1 antibodies. Scale bar, 50 μm. (I) Percentages of neurons (Tuj1⁺) expressing MeCP2 protein within each group of samples described in (H). (J) Quantification of MeCP2 protein amounts in the MeCP2-positive neurons in (H). (K) Quantification of soma sizes of the neurons in (H). Quantifications were done using ImageJ and are shown as the mean of relative percentages as compared with WT neurons ± SD of two biological replicates. (L to N) The mEPSC frequency (L), mEPSC amplitude (M), and membrane capacitance (N) of neurons in (F). Shown is the mean ± SD of at least two biological repeats, with more than 20 neurons for each condition. *P < 0.05 and **P < 0.01, one-way ANOVA with Bonferroni correction.