Integrated epigenomic analyses of neuronal MeCP2 reveal a role for long-range interaction with active genes

Yasui et al. 10.1073/pnas.0707442104.

Supporting Information

Files in this Data Supplement:

SI Figure 6
SI Table 1
SI Table 2
SI Figure 7
SI Figure 8
SI Figure 9
SI Figure 10
SI Figure 11
SI Figure 12
SI Figure 13




SI Figure 6

Fig. 6. Specificity controls for custom anti-MeCP2 antibody. (A) SH-SY5Y nuclear extracts were resolved by SDS-PAGE, transferred to nitrocellulose, and probed with the custom anti-MeCP2 that detects a single band at 75 kDa. (B) MeCP2 was immunoprecipitated from SH-SY5Y nuclear extracts, resolved by SDS-PAGE, transferred to nitrocellulose, and probed with a commercial rabbit anti-MeCP2 antibody (Affinity). IP with a control IgY antibody is shown for comparison. (C) Mass spectrometric analysis was performed on trypsin-digested peptides from proteins immunoprecipitated from wild-type mouse brain nuclear extracts with anti-MeCP2 antibodies. Peptides matching MeCP2 are shown. (D) Chromatin was prepared from wild-type (left bar) and Mecp2 deficient mouse brain (right bar) and immunoprecipitated with our custom anti-MeCP2 antibody and an IgY control. Enrichment of Snrpn using anti MeCP2 was assayed by quantitative PCR and normalized to levels using control IgY set to one.

SI Figure 7

Fig. 7. L1, L3, and L4 validation by conventional ChIP. Amplicons from differentiated SH-SY5Y cells were amplified with 30 cycles of conventional PCR and resolved by electrophoresis. L1, L3, and L4 sites are labeled with their corresponding chromosomal band. For each site amplified total reference DNA (TR) is shown first followed by DNA enriched by MeCP2 ChIP (IP). Band intensities were determined by densitometry and the ratio of IP to total reference signal for each primer pair is shown below chromosomal band location. Images were compiled from separate gels.

SI Figure 8

Fig. 8. ChIP-chip analysis of the imprinted regions 15q11-13 and 11p15.5. (A) A total of 62 L3 MeCP2 binding sites were identified over 13 Mb. Four sites were observed between UBE3A and GABRB3, genes affected by MeCP2 deficiency and between the two known MeCP2 binding sites in SNRPN and GABRB3 (pink boxes): one was between the imprinted genes UBE3A and ATP10A, two were within ATP10A introns, and one was between ATP10A and GABRB3 (middle bracket). Two other sites were observed between GABRB3 and OCA2. A region centromeric to SNRPN (left bracket) contained four sites; two of which were located between OR4M2 and CYFIP1, two between NIPA1 and MKRN3, and one between MKRN3 and NDN. The region between TJP1 and RYR3 contains an unusually dense cluster of 39 L3 sites (right bracket). This is an average of one site per 98 Kb versus an average of one site per 209 Kb in the total loci tested. Sites in this subregion including two that were validated (black stars) contain the neurotransmitter receptor genes CHRNA7 and duplicate CHRFAM7A and the neurologically expressed genes SGNE1 and RYR3. (B) An example of abundant MeCP2 sites flanking the cholinergic receptor gene CHRNA7 within 15q11-13. (C) Seven MeCP2 binding sites were identified in the transcriptionally active H19-IGF2 locus. One site upstream of H19 (black box) corresponds to the imprinting control region (ICR) that controls expression of H19 and IGF2. MeCP2 has also been shown previously to bind to the murine H19/Igf2 ICR 20. Another site was observed between and upstream of both LOC492304 and IGF2 and validated. Five sites (black boxes) were found flanking INS and TH encoding insulin and tyrosine hydroxylase.

SI Figure 9

Fig. 9. ChIP-chip analysis of ID1-4. (First row) For ID1, a previously identified MeCP2 L3 site located 7 kb upstream of the first exon was confirmed by ChIP-chip and validated (pink box). (Second row) In ID2, an L3 site was identified upstream and validated (black box), and a site in the second intron was confirmed and validated (pink box) by ChIP-chip. (Third row) In ID3, three MeCP2 L3 sites flanking the gene were observed in ChIP-chip, one of which was observed previously and validated (pink box). Two other sites were observed in ChIP-chip and validated (black boxes). (Fourth row) Two entirely novel MeCP2 L3 sites were revealed »7 kb and 12 kb (black boxes) downstream of ID4 and validated.

SI Figure 10

Fig. 10. ChIP-chip promoter validation by conventional ChIP. Randomly selected human promoters were selected for validation by conventional PCR. Primers corresponding to the listed promoters were used to amplify ChIP DNA from SH-SY5Y cells by using 30 cycles of conventional PCR. PCR products were resolved by gel electrophoresis. For each site, bands from amplified total reference DNA (TR) is shown first, followed by bands from DNA enriched by MeCP2 ChIP (IP). The ratio of IP to total reference signal for each primer pair is shown below chromosomal band location. Images were compiled from separate gels.

SI Figure 11

Fig. 11. Statistical estimation of MeCP2 promoter hits. Histogram of signal intensities (x axis) plotted for 24,275 human gene promoters (y axis). Solid line represents a Guassian noise fit, and the filled circles represent mean log2 signal values. Analysis of data based on signal intensity and a 10% false positive correction factor suggest that between 4,300 and 2,600 promoters are MeCP2 bound. Data from one promoter array hybridization is representative.

SI Figure 12

Fig. 12. MeCP2 is bound to promoters of highly expressed genes. A comparison is shown of the top 500, 1,000, and 2,000 genes for RNA expression to the top 500, 1,000, and 2,000 MeCP2-bound promoters respectively. In each case, the number of genes in common on the lists (solid bars) exceeds the number expected at random (unfilled bars) by at least 2-fold. See Methods for random calculation and comparison methods.

SI Figure 13

Fig. 13. Validation of MeDIP assay. (A) PCR amplification of genomic DNA fragments purified by pull-down with anti-5 methylcytosine antibody. MECP2 promoter was used as a positive control as an X chromosome inactivation control in female SH-SY5Y cells. (Top) Enrichment of MeCP2 promoter is observed in both replicates, consistent with CpG methylation. GAPDH promoter was chosen as a negative control as this gene is bi-allelically expressed at high levels. PCR amplification levels of this promoter at background levels (IgG lane) is consistent with a lack of CpG methylation. As an additional control, anti-methyl-cytosine purified DNA was amplified with primers to a genomic sequence lacking CpG dinucleotides. (B) To further validate MeDIP analysis and methylation patterns of SH-SY-5Y cells compared with a nontransformed fibroblast cell line, the MeDIP log2 values previously reported (1) were compared with our MeDIP analysis for 12 randomly chosen genes (HOXA9, KCNA1, IPO13, SFRS5, NME6, SURF1, MGC23280, POU5F1, AQP2, LDHC, OXT, GPR109A). Only HOXA9, a tissue-specific gene expressed in fibroblasts not neurons (data point circled in red), was discordant between the two analyses, as expected. The correlation between the two studies for the remaining 11 genes was r² = 0.901.

1. Weber M, Hellmann I, Stadler MB, Ramos L, Paabo S, Rebhan M, Schubeler D (2007) Nat Genet 39:457-66.

Table 1. Summary of MeCP2 binding sites

L1-L4 MeCP2 binding sites revealed by ChIP-chip were categorized into two groups on the basis of location: intergenic (between genes) and intragenic (within genes). Intergenic MeCP2 sites were subdivided into groups on the basis of distance from transcriptional start sites (TSS) or transcriptional end sites (TES). Intragenic sites were subdivided according to overlap with exons, introns, 5' ends, or 3' ends. In some cases, subtotals do not add up to totals (bold), because some sites overlap both introns and exons. The number of MeCP2 binding sites that overlap with inter- or intragenic CpG islands is also shown.

Table 2. Expression analysis of genes with MeCP2-bound CpG islands

MeCP2 binds to only 10 CpG islands in 26.3 Mb and significantly affects the expression of only one linked gene, RNASEH2A. Analysis of Affymetrix HU133A expression chip signals by dChip software was used to calculate fold change and P values from previously published study (1). NS, not significant; UD, undifferentiated SH-SY5Y neurons; D, 48-h PMA-differentiated SH-SY5Y neurons; MD, MeCP2 decoy-treated differentiated SH-SY5Y neurons. Probes for MORG1 were not present on the Affymetrix array (NA).

1. Peddada S, Yasui DY, LaSalle JM (2006) Hum Mol Genet 15:2003-2014.

• Early Edition
• Archives
• Online Submission

• Feature Articles

  (Expanded research articles of exceptional breadth)

• Commentaries

  (Expert commentary on leading research)

• Letters

  (Online-only letters to the editor)

• Inaugural Articles

  (Articles by recently elected NAS members)

• PNAS Profiles

  (Biographical profiles of Academy members)

• Sustainability Science

  (Interactions of human and environmental systems)

• Collected Papers

  (Editorials, Perspectives, Reviews, Colloquia, etc.)

• Special Features

  (Solicited articles on exceptional research topics)

• Sackler Colloquia

  (Scientific reports held under NAS auspices)