Transcriptional regulation of the MET receptor tyrosine kinase gene by MeCP2 and sex-specific expression in autism and Rett syndrome

Abstract

Single nucleotide variants (SNV) in the gene encoding the MET receptor tyrosine kinase have been associated with an increased risk for autism spectrum disorders (ASD). The MET promoter SNV rs1858830 C 'low activity' allele is enriched in ASD, associated with reduced protein expression, and impacts functional and structural circuit connectivity in humans. To gain insight into the transcriptional regulation of MET on ASD-risk etiology, we examined an interaction between the methyl CpG-binding protein 2 (MeCP2) and the MET 5' promoter region. Mutations in MeCP2 cause Rett syndrome (RTT), a predominantly female neurodevelopmental disorder sharing some ASD clinical symptoms. MeCP2 binds to a region of the MET promoter containing the ASD-risk SNV, and displays rs1858830 genotype-specific binding in human neural progenitor cells derived from the olfactory neuroepithelium. MeCP2 binding enhances MET expression in the presence of the rs1858830 C allele, but MET transcription is attenuated by RTT-specific mutations in MeCP2. In the postmortem temporal cortex, a region normally enriched in MET, gene expression is reduced dramatically in females with RTT, although not due to enrichment of the rs1858830 C 'low activity' allele. We newly identified a sex-based reduction in MET expression, with male ASD cases, but not female ASD cases compared with sex-matched controls. The experimental data reveal a prominent allele-specific regulation of MET transcription by MeCP2. The mechanisms underlying the pronounced reduction of MET in ASD and RTT temporal cortex are distinct and likely related to factors unique to each disorder, including a noted sex bias.

Figure 1

MeCP2 directly binds to the MET promoter. (a) Schematic of the 5′ promoter region of MET drawn to scale; Human Genome Browser (hg19), chr7: 116098419–116099867. Horizontal lines indicate the relative locations for the CpG island (green), primers used in ChIP assays (gray), functional promoter variant (rs1858830), MET transcriptional start site (TSS). (b) Anti-MeCP2 antibody directly pulls down the MET promoter sequence using primers 2 and 7 (red bars). Anti-acetyl histone 3 (H3) and anti-immunoglobulin G (IgG) antibodies were used as positive and negative controls, respectively. (c) qPCR analysis of ChIP by MeCP2 of the MET promoter sequence using primer 7. *P<0.001.

Figure 2

Functional characterization of MeCP2 and MECP2 mutations on MET transcriptional activation. (a) Luciferase reporter assays demonstrate differential activation of the MET promoter by MeCP2. MET luciferase reporter constructs containing rs1858830 G or C were transiently transfected into HEK cells with or without addition of MECP2 cDNA. *P<0.001. (b) Schematic of the MeCP2 protein structure with common RTT-causing mutations (MBD; TRD, transcription repressor domain; NLS, nuclear localization signal). (c) Luciferase assays of MECP2 mutations cotransfected with MET promoter luciferase showed altered transcription compared with wild-type MECP2. (red line-rs1858830 C allele+WT MECP2; blue line-rs1858830 G allele+WT MECP2). *P<0.05 compared with rs1858830 allele (G or C respectively)+ WT MECP2. Fold of luciferase activation was calculated after normalization against an empty luciferase control vector. All transfections were performed in triplicate.

Figure 3

Differential binding of MeCP2 in the presence of rs1858830 ASD risk (C) and non-risk (G) alleles. Human CNON cells were assayed by ChIP using an anti-MeCP2 antibody. qPCR analysis using primer 7 showed increased MeCP2 binding to the MET promoter in CNON cells homozygous for the risk (CC) genotype compared with CNON cells heterozygous (GC) or homozygous for the non-risk (GG) allele. Results reflect the mean ±s.e.m. across three CNON cells for each genotype. P<0.037.

Figure 4

Sex-specific MET expression in postmortem brain of individuals with ASD. Samples are represented by open circles and group means are represented by horizontal bars. (a) MET expression in temporal cortex of females with RTT and controls (CTL). **P<0.001 (b) MET expression in temporal cortex of individuals with ASD and controls. (c) MET expression in the temporal cortex of individuals with ASD and controls as shown in panel b, separated by sex. No significant difference in MET expression was detected between males (M) and females (F) among controls. *P<0.05.