UBC Faculty Research and Publications

CAGE-defined promoter regions of the genes implicated in Rett Syndrome Vitezic, Morana; Bertin, Nicolas; Andersson, Robin; Lipovich, Leonard; Kawaji, Hideya; Lassmann, Timo; Sandelin, Albin; Heutink, Peter; Goldowitz, Dan; Ha, Thomas; Zhang, Peter; Patrizi, Annarita; Fagiolini, Michela; Forrest, Alistair R; Carninci, Piero; Saxena, Alka Dec 24, 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12864_2013_Article_7082.pdf [ 1.6MB ]
JSON: 52383-1.0221584.json
JSON-LD: 52383-1.0221584-ld.json
RDF/XML (Pretty): 52383-1.0221584-rdf.xml
RDF/JSON: 52383-1.0221584-rdf.json
Turtle: 52383-1.0221584-turtle.txt
N-Triples: 52383-1.0221584-rdf-ntriples.txt
Original Record: 52383-1.0221584-source.json
Full Text

Full Text

RESEARCH ARTICLE Open AccessCAGE-defined promoter remeoomAes(Mrerereities in the clinical profile that overlap with RTT. ClassicRTT patients with MECP2 mutations have a normalbinds methylated and unmethylated DNA [6-9] and func-tions as a repressor and activator of genes [10-13]. EvenVitezic et al. BMC Genomics 2014, 15:1177http://www.biomedcentral.com/1471-2164/15/1177Genomics Core Facility, Guy’s Hospital, London, UKFull list of author information is available at the end of the articleperiod of development followed by regression of acquiredskills, deceleration of head circumference, epilepsy, handstereotypies, breathing abnormalities, inability to walk ortalk and intellectual disability while patients with atypicalRTT may show some but not all features of classic RettthoughMECP2 is expressed ubiquitously [14], MECP2 mu-tations and copy number variations in humans lead toneurological phenotypes such as classic or atypical Rettsyndrome and in rare cases Angelman Syndrome, X-linkedmental retardation and Autism (reviewed in [15]) suggest-ing a distinct role for MeCP2 protein in the brain [16,17].The level of MeCP2 protein in neurons increases withneuronal maturity [18] and it is abundantly expressed inthe mature brain, almost equivalent to Histone H1 levels* Correspondence: alka.saxena@gstt.nhs.uk1Omics Science Center, RIKEN Yokohama Institute, Omics Science Center(OSC), 1-17-22 Suehiro cho, Tsurumi ku, Yokohama, Japan14Currently at: Biomedical Research Centre at Guy’s and St Thomas’ Trust,regions of the three genes.Results: Our investigations reveal the predominantly used transcription start sites (TSSs) for each gene includingnovel transcription start sites for FOXG1. We show that FOXG1 expression is poorly correlated with the expressionof MECP2 and CDKL5. We identify promoter shapes for each TSS, the predicted location of enhancers for each geneand the common transcription factors likely to regulate the three genes. Our data imply Polycomb RepressiveComplex 2 (PRC2) mediated silencing of Foxg1 in cerebellum.Conclusions: Our analyses provide a comprehensive picture of the regulatory regions of the three genes involvedin Rett Syndrome.Keywords: Rett Syndrome, CAGE, Transcriptomics, Promoter architectureBackgroundRett Syndrome (RTT) is a disorder caused by mutations inMethyl CpG binding protein 2 (MECP2), Forkhead boxG1 (FOXG1) or Cyclin-dependent kinase-like 5 (CDKL5)genes [1-3] Although the phenotype of patients with mu-tations in MECP2 differs from the phenotype of patientswith FOXG1 or CDKL5 mutations, there are some similar-syndrome [4]. Mutations in FOXG1 are known to causethe congenital variant of Rett syndrome where theinitial normal developmental window is absent [2].CDKL5 mutations are found in patients with severeepilepsy during early childhood that later show fea-tures that resemble atypical RTT syndrome [5].MeCP2 is an X-linked methyl CpG binding protein whichimplicated in Rett SyndroMorana Vitezic1,2,3, Nicolas Bertin1,4,5, Robin Andersson3, LAlbin Sandelin3, Peter Heutink4,10,11, Dan Goldowitz12, ThMichela Fagiolini13, Alistair RR Forrest1,4, Piero Carninci1,4,AbstractBackground: Mutations in three functionally diverse genForkhead box G1 (FOXG1), Methyl CpG binding protein 2been studied individually, not much is known about theirand regulatory regions. Here we analyzed data from hundFANTOM5 project, to identify transcript initiation sites, exp© 2014 Vitezic et al.; licensee BioMed Central.Commons Attribution License (http://creativecreproduction in any medium, provided the orDedication waiver (http://creativecommons.orunless otherwise stated.gions of the genesenard Lipovich6,7, Hideya Kawaji1,4,8, Timo Lassmann1,4,9,as Ha12, Peter Zhang12, Annarita Patrizi13,lka Saxena1,14* and The FANTOM Consortiumcause Rett Syndrome. Although the functions ofECP2) and Cyclin-dependent kinase-like 5 (CDKL5) havelation to each other with respect to expression levelsds of mouse and human samples included in thession levels, expression correlations and regulatoryThis is an Open Access article distributed under the terms of the Creativeommons.org/licenses/by/2.0), which permits unrestricted use, distribution, andiginal work is properly credited. The Creative Commons Public Domaing/publicdomain/zero/1.0/) applies to the data made available in this article,Vitezic et al. BMC Genomics 2014, 15:1177 Page 2 of 16http://www.biomedcentral.com/1471-2164/15/1177[19], but the level of MECP2 mRNA in cells is reported tonot correlate with the level of MeCP2 protein in cells [18].FOXG1 protein is a brain specific member of theforkhead transcription factor family with a role intranscriptional repression. Similar to other membersof the forkhead family, FOXG1 has a defined bindingsequence motif [20], which bears sequence similarityto other forkhead protein binding sites.CDKL5 protein is a serine threonine kinase, whoseexpression is low in embryonic stages, but increases inpostnatal stages up to postnatal day 15 [21] CDKL5mRNA is expressed in brain and all other tissues [22,23].CDKL5 protein levels are known to coincide with itsmRNA levels [24].Even though these three genes have different expressionpatterns, distinct functions and specific regulatory targets,their paths appear to intersect. Both MeCP2 and FOXG1proteins regulate transcription via DNA binding andassociation with other transcriptional regulators [6-13][25,26]. The functions of MeCP2 and CDKL5 proteinsalso appear to be interconnected. MeCP2 has multiplephosphorylation sites [27] and is a target of CDKL5phosphorylation. Additionally, there are contradictoryreports on the expression level of CDKL5 protein andmRNA in the absence of MeCP2 [28-30]. Altogether,these observations suggest that the overlapping fea-tures in Rett syndrome may be caused by impairmentof common or intersecting biological pathways down-stream of expression in the brain. Alternately, thesegenes may be interdependent on each other for expressionor regulation, which may lead to the overlap in phenotypicfeatures.Although we have some knowledge of their downstreamintersecting functions, we are yet unaware of the commongenomic features between these three genes, whichmay provide insights into their regulation. Importantly,although both MECP2 and CDKL5 genes are expressedubiquitously, their mutations cause a brain specific pheno-type suggesting that their expression level, transcriptionregulation, or function in brain may be distinct from thatin other tissues. We tried to resolve these questionsthrough bioinformatics analyses using the FANTOM5dataset [31].Data from FANTOM5 provide the unprecedentedopportunity to identify the transcription start sites(TSSs) of these genes and study their expression profile inhundreds of mouse and human samples using CapAnalysis of Gene Expression method (CAGE) [31]. Inconjunction with the recently released ENCODE dataset[32]. FANTOM5 data also enable the identification ofregulatory histone marks at TSSs. Since the RTT pheno-type is reflected in the Mecp2 KO mouse model [33] andstudies on this disorder are conducted in mouse tissuesand cells, we also included mouse samples in our analyses.We analyzed the TSS expression data from the FANTOM5project using over 1000 human and over 450 mousesamples to identify common and diverse features of thegenomic architecture of the three genes implicated in RTT(for a complete list of samples Additional file 1: Table S1).For our investigation, we divided the human and mousesamples into tissues, primary cells and cell lines to studythe expression levels of the TSS of the three genes invarious samples. Our data reveal the precise initiationsites for the three genes, including previously unknownTSSs for FOXG1 in mouse and humans. We show thateach of these genes use the same TSS in most tissues andprovide information on the expression level of the threegenes over development in multiple human and mousesamples. Although we did not find a significant correlationbetween the expression levels of the three genes in thebrain, our genome wide analyses uncovered commontranscription factors regulating the three genes, suggestingan additional molecular layer in the pathogenesis of RettSyndrome. The FANTOM5 CAGE dataset also allowed usto locate putative enhancers regulating the three genes inhuman (methods described in Anderson et al., [34]) andusing mouse ENCODE ChIP-seq data, we identifiedgenomic regions bearing promoter and enhancer marks.This work is part of the FANTOM5 project. Data down-loads, genomic tools and co-published manuscripts aresummarized here: http://fantom.gsc.riken.jp/5/.MethodsFANTOM5 samplesSingle molecule CAGE profiles were generated fromRNA obtained from a collection of 573 human primary cellsamples (~3 donors for most cell types) covering mostmammalian cell steady states. This data set was comple-mented with profiles of 250 different cancer cell lines, 152human post-mortem tissues and 456 mouse samples(detailed sample list is available in Additional file 1: Table S1and origin of each sample is available as SupplementaryMaterial in Forrest et al. 2014 [31]). Primary cells forneurons and astrocytes discussed in this manuscriptwere obtained from ScienCell Research Laboratories.Human neurons were isolated from the human brain,cryopreserved at primary cultures and delivered frozen.Human astrocytes were isolated from cerebral cortex andcerebellum. Both were cryopreserved at passage one anddelivered frozen.All human samples used in the project were eitherexempted material (available in public collections or com-mercially available), or provided under informed consent.All non-exempt material is covered under RIKENYokohama Ethics applications (H17-34 and H21-14). Mousetissue samples were collected as per RIKEN Yokohamainstitutional guidelines. Mouse primary cells were collectedas per our collaborators Institutional guidelines andVitezic et al. BMC Genomics 2014, 15:1177 Page 3 of 16http://www.biomedcentral.com/1471-2164/15/1177shipped as either purified RNA or as guanidiniumisothyocyanate lysates (Trizol, Isogen or Qiazol) whichwere then purified using the miRNeasy kit (QIAGEN).More detailed information for each specific sample isavailable in Additional file: 1 Table S1 of [31].All the data published by the Fantom5 project and bythis study are available through the Fantom5 portalhttp://fantom.gsc.riken.jp/5/data/. All CAGE data hasbeen deposited at DDBJ DRA under accession numberDRA000991.Identifying CAGE derived transcription start sitesWe used the FANTOM5 database ([31]) to identifytranscription start sites (TSS) for our genes, using thedecomposition peak identification (DPI) clustering andnomenclature developed for the FANTOM5 project [31].We selected robust CAGE defined DPI clusters fallinginside the RefSeq regions known to be associated to thethree genes. To select for genuine TSSs we used theFANTOM5 TSS classifier and restricted our TSS selectionto those with a value of 0.1 and above [31]. The TSSs wereannotated using the names assigned to clusters in theFANTOM5 Resource browser (SSTAR, Semantic catalogueof, samples, transcription initiations, and regulations,http://fantom.gsc.riken.jp/5/sstar/). Annotation files werebuilt in the context of the FANTOM5 project with respectto Gencode v10 gene model (human), RefSeq (mouse),CpG islands and TATA box in bed format.TSS expressionWe extracted expression information for each TSSsusing the FANTOM5 expression dataset for tissues,cell lines and primary cells in human and mouse (seeAdditional file 1: Table S1 for a full list of samplesand TPM expressions). The expression values are shownin tags per million (TPM) calculated on a per-library totalexpression. We discarded all the TSSs that did nothave over 5 TPM expression in any of the samples.All expression level figures, heatmaps and correlationswere calculated using R (http://www.r-project.org/).Mecp2 and histone expression comparisonWe extracted the CAGE defined promoters associatedto the genes whose products form the Histone1 tran-scripts (HIST1H1A, HIST1H1B, HIST1H1C, HIST1H1D,HIST1H1E, H1F0, H1FX). All the values for differentgenes were added together and compared to expressionlevels of MECP2.Identifying TSS overlaps with ChIP seq data from humanand mouse ENCODE TSSsTSSs identified were expanded by 500 nucleotides oneither side (±500bp). ChIP seq data from Human andmouse ENCODE were downloaded as bed files and inter-sected with our expanded TSS using intersectBed [35].Defining human enhancersTo identify enhancers associated with the human Rettgenes, we used the CAGE derived enhancer database fromAndersson et al. [34]. In short, the identified enhancersfrom Andersson et al. within 500kb distance from the iden-tified Rett genes promoters were selected [34]. The expres-sion of pairs of enhancers and promoters was thencompared in all the human samples using a Pearson correl-ation test. The resulting comparisons were then correctedfor multiple testing using Bonferroni correction. Onlyenhancers significantly correlated (corrected P < 0.05)with any of the three Rett genes promoters wereincluded.Transcription factor binding site (TFBS) analysisWe downloaded the whole-genome alignment of thehuman genome with 45 other vertebrate genomes,and of the mouse genome with 29 other vertebrategenomes, from the UCSC Genome Browser database[36]. From these alignments, we retained the alignmentsbetween the human, macaque, mouse, rat, cow, horse,dog, opossum, and chicken genomes only, and used theT-Coffee alignment tool [37] on 1000 bp segments ofthe genome to optimize the alignment for the nineselected genomes. We then ran MotEvo [38] on thesewhole-genome alignments using a background priorprobability of 0.98, a prior for the UFE (unidentifiedfunctional element) model of 200 relative to the weightmatrices, a UFE motif length of 8 base pairs, and auniform background sequence probability. A posteriorprobability calculated by MotEvo for a putative TFBSwas retained if it was at least 0.2. We used the centerposition for a given CAGE promoter on the genome as areference point, and summed the posterior probabilitiesfor the putative binding sites for each transcription factorwithin a distance of 500 basepairs of the reference pointto obtain the estimated number of binding sites for eachtranscription factor. To evaluate the statistical significanceof this number, for each transcription factor we estimatedthe number of binding sites in exactly the same way for all184,827 (human) or 116,227 (mouse) promoters inthe FANTOM5 data sets, and ranked the promotersaccordingly. The tail probability was then obtained bydividing the rank of the promoter of interest by 184,827(human) or 116,227 (mouse).ResultsFOXG1 expression in mouse and humansAnalyses of TSS from 1193 human samples and 457mouse samples comprised of tissues, primary cells andcell lines (only one mouse cell line was investigated inVitezic et al. BMC Genomics 2014, 15:1177 Page 4 of 16http://www.biomedcentral.com/1471-2164/15/1177FANTOM5) identified 8 TSSs for human FOXG1 and 6TSSs for mouse Foxg1 (Additional file 2: Table S2).FOXG1 expression above 1 TPM was found in 23% (231)and 30% (140) of human and mouse samples respectively,suggesting that the expression of this gene was limited toselected tissues (Additional file 1: Table S1). Transcriptionstart sites were defined as novel if they were found at adistance of over 500 bp from the known RefSeq TSSs.Our data show 3 TSSs in mouse highly expressed in brainsub-regions and cells, two of which are novel. The expres-sion levels of different TSSs of FOXG1 were variable inhuman and mouse samples, with the highest expressionseen in specific regions of the brain (Figures 1a, b,Additional file 3: Figures S1a,b). The top three initiationsites were located at (in order of their expression levels)chr12:50484904..50484950,+; (pA@Foxg1, novel promoter,located more than 1000 bases downstream of the RefSeqannotated TSSs) chr12:50483639..50483654,+; (p1@Foxg1,200 bp upstream of the two annotated Foxg1 TSSs) andchr12:50485112..50485144,+;(pB@Foxg1, novel promoter,1200 bp downstream from the annotated RefSeq initiationsites) (Figure 2a). Expression of mouse Foxg1 was alsorestricted to brain tissue and brain related cells, butsurprisingly the two novel TSSs of mouse Foxg1 werealso found highly expressed in the single mouse cellline sequenced in the FANTOM5 project (fibroblastcell line) suggesting that other than the brain, fibroblastcell lines may be useful for in vitro analysis of Foxg1 inmouse (Figure 1a).In human, there is RefSeq annotation support for asingle FOXG1 isoform and therefore a single TSS. Wefound 8 TSSs for human FOXG1 expressed over 5 TPMand the 3 TSSs with the highest expression in human brainwere located at chr14:29235961..29236008,+ (p1@FOXG1);chr14:29234581..29234601,+ (p2@FOXG1; novel); andchr14:29236269..29236285,+ (p3@FOXG1), with distancesof 317, 1697 and 9 bases upstream of the RefSeq annotatedTSS, respectively (Additional file 2: Table S2, Figure 2d,Additional file 3: Figure S1). Thus, our analyses reveal thatin human brain the highest used TSS for FOXG1 is lo-cated 317 bases upstream of the annotated start site.Contrary to mouse, where we found expression leveldifferences of over 10-fold between TSSs, in humansamples, the difference in expression between the threeTSSs was less than 2-fold (Figure 1a, 1b, Additional file 1:Table S1).Intriguingly, we did not find FOXG1 expression inmouse or human cerebellum suggesting silencing ofFOXG1 in cerebellum. Inability to detect TSS expressionmay result from technical artifacts such as low expressionlevels not discernible at the conducted depth of sequencingor the use of an alternate tissue specific start site. To ruleout technical artifacts, we referred to the ENCODE datasetto investigate signs of transcriptional activity in thechromosomal location of Foxg1 and up to 10 kb upstreamin mouse cerebellum. We analyzed ENCODE data forDNAse-I hypersensitive sites (DNAse-I HSS), which areknown to faithfully recognize active transcription initiationsites [39], in mouse cerebellum, cerebrum and whole brain.Our analyses revealed an absence of DNAse-I HSS inmouse cerebellum, while DNAse-I HSS were present incerebrum and whole brain samples (Additional file 4:Figure S2). Since DNAse-I HSS usually coincide withthe active promoter specific histone mark of trimethylatedHistone 3 lysine 4 (H3K4me3) [39] and transcriptionallyactive enhancers may also bear the specific histone markof acetylated Histone 3 lysine 27 (H3K27ac), we lookedfor these two marks in mouse cerebellum and mousecortex. We found that H3K4me3 and H3K27ac wereenriched at the locus in 8-week old cortex samples, butnot in 8-week old cerebellum samples. Surprisingly,our investigations revealed trimethylated Histone 3 lysine27 (H3K27me3) enrichment at this chromosomal locus inmouse cerebellum, suggesting silencing of Foxg1 by Poly-comb Repressive Complex 2 (PRC2) (Additional file 4:Figure S2). Since ChIP data for active and repressivehistone marks in human brain is not available fromENCODE at this time, we were unable to confirmsimilar chromatin signatures for PRC2 silencing ofFOXG1 in the human cerebellum.Silencing by chromatin remodeling proteins such asPRC2 requires a non-coding RNA to mediate chromatinmodification [40]. Therefore, we searched for potentialcis-regulatory ncRNAs that may mediate Foxg1 silencing.We found one ncRNA downstream of Foxg1 (RefSeqNR_026733), however its expression was not entirelydiscordant with that of Foxg1 (data not shown). Analysisof the ncRNA database and manual annotation of UCSCGenome Browser revealed several ncRNAs within agenomic window of 1.5 MB around Foxg1, but noneof the listed ncRNAs were detectable in the FANTOM5CAGE dataset.MECP2 expression in mouse and humansIn humans and mouse we identified two TSSs for MECP2,less than 100 bases upstream of the RefSeq annotated startsites of which p1@MECP2/Mecp2 was expressed predom-inantly in most tissues and p2@MECP2/Mecp2 displayeda stable low level expression in all tissues (expression lessthan 10 TPM) (Additional file 2: Table S2, Figures 1c, 1d,2b and 2e). We found an additional intronic promoter(p5@MECP2) in humans alone, expressed exclusively inblood primary cells, particularly in CD14 monocytes(Figure 1d and Figure 2e). Expression of p1@MECP2in humans and mouse was found above 5 TPM in mosttissues, primary cells and cell lines, suggesting that tran-scripts arising from this promoter were ubiquitouslyexpressed. Surprisingly our analysis of human tissues andFigure 1 (See legend on next page.)Vitezic et al. BMC Genomics 2014, 15:1177 Page 5 of 16http://www.biomedcentral.com/1471-2164/15/1177cells revealed that the highest expression of MECP2was seen in non-neuronal tissues (vagina and ovary)and cell lines (Breast Carcinoma, Krukenberg tumour, lensbirth (P00) but remained unstable up to the age of P30(Additional file 5: Figure S3a-c).(See figure on previous page.)Figure 1 Expression levels of the identified TSS for the three genes. Dot plots showing the expression level of each promoter in TPM valuesin all brain regions, and selected other samples (based on expression level). The novel promoter pA@Foxg1 is the most highly expressed Foxg1TSS in mouse primary cells and brain tissue (a), with the highest expression in cortical neurons (1018 TPM) and neonate hippocampus (435 TPM).Among mouse cells, we find high levels of p1@ Foxg1 expressed in hippocampal neurons and fibroblast cell line. In human samples (panel b)the highest expression of FOXG1 is seen from p1@FOXG1 in fetal temporal lobe (292 TPM), among primary cells in neurons (149 TPM) andamong cell lines in medulloblastoma cell line (184 TPM). For mouse Mecp2, the highest expression of p1@Mecp2 is in striatal neurons (77 TPM)and cerebellar granule cells (70 TPM) and among mouse tissues (panel c) the maximum expression is seen in neonate corpus striatum (65 TPM)and adult cerebellum (52 TPM). For human, the highest expression of p1@MECP2 is found in cancer cell lines including breast carcinoma cell line(119 TPM) (panel d). In human brain the highest expression of p1@MECP2 is found in the temporal lobe (63 TPM). The two promoters of Cdkl5 inmouse are co-expressed with highest expression in adult cortex in the brain and raphe neurons among primary cells (panel e). In humans (panel f)the two promoters are expressed differentially with transcripts arising from p1 over-represented. p1@CDKL5 expression is highest in the newbornmedial frontal gyrus and in neurons. In human cancer cell lines, CDKL5 is generally expressed at low levels (less than 10 TPM) from either of thepromoters (p1 > p2), with a few exceptions (Additional file 1: Table S1, Additional file 3: Figure S1f).Vitezic et al. BMC Genomics 2014, 15:1177 Page 6 of 16http://www.biomedcentral.com/1471-2164/15/1177epithelial and lung adenocarcinoma) (Additional file 3:Figure S1d). In agreement with previous reports in mouse,we found that at mRNA level, the expression of p1@Mecp2in astrocytes (15 TPM) was much lower than in neurons(77 to 41 TPM) (Additional file 3: Figure S1c) but amonghuman primary cells, the expression of p1@MECP2 inneurons (12 TPM) was lower than the p1@MECP2expression in astrocytes (34 TPM). In contrast withthe brain, the expression levels of MeCP2 protein inthe heart are reportedly higher in embryonic stagesthan in postnatal heart [41]. Therefore, we investigatedthe expression levels of Mecp2 during development inheart, liver and kidney. Our analyses showed that in heart,Mecp2 expression fluctuated during embryonic stages andwas higher than at postnatal day 25 (P25) and P30. Inkidney, the expression of Mecp2 declined after P20 and inliver the expression ofMecp2 appeared to be induced afterFigure 2 Locations of the TSSs identified for the three genes. GenomeFOXG1 (panels a and d), MECP2 (panels b and e) and CDKL5 (panels c anIn each panel the top two tracks show RefSeq genes and mRNAs from GenbaCpG islands. Red arrows mark the key TSSs for each gene. We found 6 TSSs foare identified by asterisks. We also found a CD14 specific intronic TSS p5 for MCDKL5 expression in Humans and MouseThe RefSeq database annotates one TSS for CDKL5 inmouse and two TSSs in human (Additional file 2: Table S2).Our analyses identified two TSSs within 100 bp of theannotated TSS in mouse and the upstream TSS inhuman samples (p1@CDKL5/Cdkl5 and p2@CDKL5/Cdkl5, Figures 2c and 2f ). In both human and mousesamples, CDKL5 expression was higher in brain tissuesthan in primary cells or cell lines (Figure 1e and 1f). Thetwo TSSs of CDKL5 were co-expressed ubiquitously inhuman and mouse, however p1@CDKL5 was expressedmore than p2@CDKL5 in most tissues in humans suggest-ing that transcripts arising from p1@CDKL5 may beover-represented in humans (Figure 1e and 1f). In mousep1@Cdkl5 and p2@Cdkl5 were expressed at similar levelsin some brain sub-regions. We tracked the expression ofCdkl5 in mouse heart, liver and kidney over developmentbrowser images showing all the TSSs identified in this study ford f) in mouse (panels a,b and c) and humans (panels d, e and f).nk. The third track shows FANTOM5 TSS and the bottom track showsr Foxg1 in mouse (panel a) and 8 TSSs in humans (panel d). Novel TSSsECP2 in human cells.Vitezic et al. BMC Genomics 2014, 15:1177 Page 7 of 16http://www.biomedcentral.com/1471-2164/15/1177from embryonic day 11 to postnatal day 30. These tissueswere previously reported to have undetectable levels ofCdkl5 [22,23]. In heart, the expression levels of Cdkl5fluctuated up to P30 (Additional file 5: Figure S3dp1@Cdkl5 p2@Cdkl5 heart). In liver and kidney,Cdkl5 expression from both TSSs was lower in adult(P25 and P30) than embryonic tissues (Additional file5: Figure S3e,f ). This observation was in contrast tothe brain where the expression of Cdkl5 was generallyhigher in postnatal brain in both mouse and humans(Additional file 6: Figure S5e,f). In agreement with pub-lished data [21] we found restricted expression of Cdkl5 inmouse astrocytes (maximum 2 TPM, Additional file 3:Figure S1e).Developmental profile for the three genes in brain subregionsThe expression levels of FOXG1, MECP2 and CDKL5 aredevelopmentally regulated in the brain. FOXG1 expressionis reported to be highest during early embryogenesis[2,42,43]. CDKL5 is weakly expressed during embryogen-esis in the cortex and its expression increases in postnatalstages until P14 after which CDKL5 expression is dimin-ished [21,24]. MeCP2 protein levels in the brain increase asdevelopment proceeds stabilizing around postnatal day 5[19]. We investigated if the reported developmental expres-sion profile for the three proteins was reflected at the TSSlevel. For mouse, we investigated the developmental TSSexpression of Mecp2 and Cdkl5 in the cerebellum (n = 3 ateach age), pituitary cortex (n = 1 at each age) and visualcortex (n = 4 at P15 and n= 3 at P30 and P60). Our data re-veal that p1@Mecp2 expression fluctuates in embryoniccerebellum samples but is clearly induced after postnatal day9 (Additional file 7: Figure S4). The expression of p1@Cdkl5and p2@Cdkl5 in mouse closely resemble the pattern ofexpression of p1@Mecp2 but at lower levels. These data arein agreement with protein expression levels of Cdkl5 andMecp2 reported previously in cerebellum by Rusconi et al.[21]. In visual cortex samples, where we investigated abroader time course we found a striking resemblance ofexpression pattern between p1@Mecp2 and p2@Cdkl5 withboth genes showing an increase from P14 to P30 and stabil-izing from P30 to P60, while the expression of p1@Cdkl5remained steady. In contrast, the expression of pA@Foxg1decreased as visual cortex matured (Additional file 7:Figure S4b). Similarly, in the pituitary gland, we found theexpression of Cdkl5 and Mecp2 to fluctuate during embry-onic stages, while Foxg1 displayed high expression duringembryonic stages with lowest expression in adult.In broad time-course samples of fetal, neonate andadult human brain sub-regions we found that FOXG1expression was generally higher in fetal samples(Additional file 6: Figure S5a) while the expressionof both promoters of CDKL5 and p1@MECP2 in fetalsamples were lower than their expression level in adults(Additional file 6: Figure S5b,c).Comparison between Mecp2 and Histone H1 expressionlevel in neuronsSkene et al. previously showed that in wild-type mouseneurons, the density of MeCP2 protein is one moleculeper two nucleosomes - equal to that of Histone H1,which is also one molecule per two nucleosomes [19].We investigated whether the similarity between Mecp2and Histone H1 protein density was reflected at themRNA level in human and mouse neurons. We alsocompared the expression level of Histone H1 and Mecp2TSS in brain sub regions even though Skene et al. [19]had not reported a correlation between the levels of thetwo proteins in whole brain tissue (they reported a correl-ation only in neuronal nuclei). We extracted the expressionfor all H1 transcripts and compared their individual andcollective levels to p1@MECP2 expression in both humanand mouse. The primary cells in mouse included neuronsfrom various brain sites as well as astrocytes and microgliacells. Our data show that the combined as well as individualexpression levels of all histone H1 transcripts were muchhigher than the expression levels of p1@Mecp2 at all agesin all samples (Additional file 8: Figure S6). In rapheneurons, substantia nigra neurons (E14), ventral spinal cordneurons (E14) and hippocampal astrocytes, we found theexpression level of a few histone transcripts closer to theexpression level of Mecp2 (Additional file 9: Table S3).Overall, this is interesting, because in order to reconcile thiswith the findings of Skene et al., substantial changesin either protein production or decay of MeCP2 and/orthe Histone 1 proteins must occur to offset the mRNAsteady state.Intra gene and Inter gene expression correlationsbetween FOXG1, CDKL5 and MECP2 in brain sub-regionsTo investigate the relationship between the expressionof all TSSs of each gene, we conducted intra gene corre-lations and found a high degree of correlation betweenp1@FOXG1 and p3@FOXG1 in human samples andbetween the promoters pA@Foxg1, pB@Foxg1, p1@Foxg1and p2@Foxg1 in mouse (Table 1, Additional file 10: TableS4, Additional file 11: Figure S7). The two TSSs ofMECP2were moderately but positively correlated with each otherin humans and mouse, while the two TSSs of Cdkl5were highly correlated in mouse (correlation coefficient,Spearman’s rank correlation 0.85) confirming our earlierobservation of similarities in expression of the two Cdkl5promoters in mouse.Since mutations in the MECP2, FOXG1 and CDKL5genes result in overlapping neurological phenotypes, weadditionally investigated the inter gene expression correla-tions of the three genes in the brain. We first generatedooxOVitezic et al. BMC Genomics 2014, 15:1177 Page 8 of 16http://www.biomedcentral.com/1471-2164/15/1177heatmaps from all brain sub-regions and brain related pri-mary cells (Figure 3a and b) for the three genes. We foundthat in humans and mouse the expression of p1@FOXG1and pA@Foxg1 in brain was strikingly discordant with theexpression of p1@MECP2 and p1@Mecp2 respectively(Figure 3a, 3b). Next, we investigated the correlationbetween the highly expressed promoters of the threegenes using Spearman’s rank test and Pearson’s correlationTable 1 Spearman Rank Correlations between the key promA: Correlation between promoters in mouse samplespA@Foxg1 pB@Foxg1 p1@FpA@Foxg1 1.00 0.77 0.76pB@Foxg1 0.77 1.00 0.83p1@Foxg1 0.76 0.83 1.00p1@Mecp2 0.35 0.32 0.30p1@Cdkl5 0.24 0.22 0.21P2@Cdkl5 0.29 0.26 0.24B: Correlation between promoters in human samplesP1@FOXG1 P2@FOXG1 P3@FP1@FOXG1 1.00 0.75 0.81P2@FOXG1 0.75 1.00 0.77P3@FOXG1 0.81 0.77 1.00p1@MECP2 0.06 0.08 0.06p1@CDKL5 0.23 0.30 0.24P2@CDKL5 0.20 0.26 0.22See Additional file 9: Table S4 for a complete list.coefficients. Based on our earlier heatmap visualization ofcontrasting expression of MECP2 and FOXG1 in braintissue, we expected to see a negative correlation betweenthe promoters of these genes in both species. Our datagenerated from all brain tissues, neurons and astrocytesshowed that in mouse the correlation between pA@Foxg1and p1@Mecp2 expression was poor (0.3) and in humanswe found negative correlation of −0.1, suggesting slightdiscordance of expression of the two genes in brain. Thus,our analyses failed to find mathematically significantevidence of contrasting expression between FOXG1 andMECP2. The two promoters of CDKL5 were also poorlycorrelated with the FOXG1 promoter expression in brain,while there was a positive correlation (23-49%) betweenexpression of MECP2 and CDKL5 in both species(Figures 3c - 3f, Additional file 11: Figure S7).Identification of regulatory regions of the three genesTo identify the regulatory regions associated with the TSSof the three genes in mouse, we extended our TSS co-ordinates by 500bp on either side and intersected themwith active histone regulatory marks of H3K27ac andH3K4me3 from ENCODE datasets. Based on current lit-erature [44-47], on chromatin modifications markingactive enhancers and promoters, we defined the cri-teria for active enhancers as those regions carrying theH3K27ac mark and active promoters as those regions car-rying the H3K4me3 without complete overlap with theH3K27ac mark.Since all three genes were found highly expressed in thecortex, we used ENCODE ChIP tracks for 8 week oldcortex for this analysis. Based on our results we derivedters of the three genes in mouse (A) and human (B)g1 p1@Mecp2 p1@Cdkl5 P2@Cdkl50.35 0.24 0.290.32 0.22 0.260.30 0.21 0.241.00 0.49 0.560.49 1.00 0.850.56 0.85 1.00XG1 p1@MECP2 p1@CDKL5 P2@CDKL50.06 0.23 0.200.08 0.30 0.260.06 0.24 0.221.00 0.23 0.270.23 1.00 0.730.27 0.73 1.00gene models for the three genes in mouse cortex (Figure 4).Our data revealed that the investigated regulatory marksfor mouse Foxg1, were distinct and non-overlapping(Additional file 12: Table S5). We found that the main TSSspA@Foxg1 and pB@Foxg1 were located between an en-hancer specific histone mark upstream and a promoter spe-cific histone mark downstream (Figure 4a and Additionalfile 12: Table S5). In contrast, for Mecp2, we found theenhancer and promoter specific histone marks to co-incide in this tissue (Figure 4b, Additional file 12:Table S5). The Mecp2 TSSs p1@Mecp2 and p2@Mecp2were upstream but within 500 bp of the histone spe-cific marks for enhancer and promoter. For Cdkl5, wefound a partial overlap between enhancer and promoterspecific histone marks. The p1@Cdkl5 was located withinthe promoter specific histone mark while the enhancerspecific histone mark was found upstream (Figure 4c,Additional file 12: Table S5). The TSS p2@Cdkl5 was alsolocated within 500 bp of these marks.As a complementary analysis, we identified human en-hancers for the three genes using the database provided in[34] that predicts active enhancers based on the expressionof balanced bi-directional low expressed enhancer RNAtranscripts. For FOXG1, MECP2 and CDKL5, we foundVitezic et al. BMC Genomics 2014, 15:1177 Page 9 of 16http://www.biomedcentral.com/1471-2164/15/11774, 14 and 1 significantly correlated cis-enhancers re-spectively (Additional file 13: Table S6, Additional file 14:Figure S8). In contrast to the mouse cortex data, the pre-dicted enhancers in human samples were found kilo basesaway from each gene suggesting long-range complex regu-lation of the three genes in humans. For FOXG1 the mosthighly correlated enhancer (r = 0.78) was located 7kb up-stream of the gene. Many predicted enhancers for MECP2had an average correlation of 0.37, the closest enhancer(53kb) had an expression correlation of 0.43, while thehighest correlated enhancer (r = 0.55) was over 408kb dis-tant. For CDKL5, the only identified active enhancer had alow correlation of 0.2 and was located over 245 kb up-stream of the gene (Additional file 13: Table S6). Interest-ingly, our data revealed that in humans, the only enhancerdisplaying the expected high correlation with gene expres-sion was for the tissue specific gene FOXG1.Figure 3 Expression correlations between the three genes. Heat maps shorelated primary cells in mouse (a) and humans (b). Expression of Mecp2 p1 appmouse (pA) and in humans (p1). The trees above the heatmaps show clusteringthe three genes in mouse (panels c and e) and human (panels d and f) as labMECP2 and FOXG1 could not be confirmed in either species across all samples (and Mecp2 in both species (panels e and f).CpG/TATA regulation of the three Rett genes and theirpromoter shapesTo investigate the regulation of the promoters of thethree genes, we analyzed computationally, the presenceof CpG islands and TATA boxes in the vicinity of thepromoters of the three genes. Intersections of CpG andTATA UCSC bed files, with our extracted list of TSSs,revealed that the three genes had TATA-less promoters inboth species. Our data showed both promoters of MECP2and CDKL5 within CpG islands in both species. ForFOXG1, 4 TSSs in mouse and 3 TSSs in human samplesappeared to be regulated by CpG islands (Table 2).It is known that promoters regulated by TATA boxesare ‘sharp’ where transcript initiation occurs at a well de-fined dominant site, no more than 4 consecutive nucleo-tides long, while promoters regulated by CpG islands are‘broad’ where multiple start sites can be detected in awing the TPM expression of all promoters in sub-regions of brain and brainears to be in contrast with the expression of the main promoter of Foxg1 inaccording to expression. Plots in panels c to f show correlation betweeneled. The expected negative correlation based on the heatmap betweenpanels c and d). We found positive correlation (23-49%) between Cdkl5Vitezic et al. BMC Genomics 2014, 15:1177 Page 10 of 16http://www.biomedcentral.com/1471-2164/15/1177broad genomic region [48]. We analyzed the extractedTSSs for sharp or broad shapes by aligning their expressionlevels across the genomic locus. For our investigation, wedefined sharp promoters as those where the majority of thetranscripts start from a single dominant TSS or frommultiple TSSs within 5 nucleotides, while promoters wereclassified as broad when they had multiple dominating ini-tiation sites within a defined TSS cluster (maximum 50 bpgenomic window). We analyzed in humans and mouse,the 3 main promoters for FOXG1 and two promoters eachfor MECP2 and CDKL5. Our analyses revealed that themain promoters for FOXG1 in both species (p1@FOXG1and pA@Foxg1) were broad in keeping with the CpGislands in their vicinity (Figure 5a, 5f). The second highestFigure 4 Mouse gene models derived from FANTOM5 TSS and ENCODCdkl5 (panel c) were drawn for the main TSS for each gene and the ENCODmark was 1 kb upstream and the promoter mark was 1.1 kb downstream of tand enhancer mark. For Cdkl5, the TSS was within the promoter and the enhaexpressed FOXG1 promoters in human and mouse(p2@FOXG1 and pB@Foxg1) appeared to have species-specific shapes and regulation. While p2@FOXG1 inhumans was found to be sharp with no TATA-box orCpG island, pB@Foxg1 in mouse was broad and CpGregulated (Figure 5b, 5g). In each species we found forFOXG1, one sharp promoter (p3@FOXG1 and p1@foxg1)devoid of TATA box or CpG island (Additional file 15:Figures S9a,f).The two main promoters of MECP2 were broad, inagreement with the CpG islands near their TSS (Figure 5c,hand Additional file 15: Figures S9e,l). In both species,the promoter p1@CDKL5 was broad in shape whilep2@CDKL5 was sharp despite the presence of a CpGE ChIP data. Gene models for Foxg1 (panel a), Mecp2 (panel b) andE histone ChIP marks for 8 week mouse cortex. For Foxg1, the enhancerhe TSS. For Mecp2, the TSS was upstream of the overlapping promoterncer was upstream of the TSS.Vitezic et al. BMC Genomics 2014, 15:1177 Page 11 of 16http://www.biomedcentral.com/1471-2164/15/1177Table 2 List of all transcript initiation sites for the 3genes in mouse (A) and human (B) samples with theirshapes and association with TATA-box and CpG islandsA Mouse promotersPromoter TATA CpG ShapepA@Foxg1 TATA-less CpG BroadpB@Foxg1 TATA-less CpG-less Broadp1@Foxg1 TATA-less CpG-less Sharpp2@Foxg1 TATA-less CpG Broadp3@Foxg1 TATA-less CpG Broadp4@Foxg1 TATA-less CpG Broadp1@Mecp2 TATA-less CpG Broadp2@Mecp2 TATA-less CpG Broadisland in its vicinity (Figure 5d, e ,i and j). A comprehen-sive list of promoters, their regulation and shapes is shownin Table 2 and Additional file 15: Figure S9.Transcriptional regulation of the three genesTo identify transcription factors binding to the three genesin both species, we analyzed the genomic sequence within500 bp from the promoters of the three genes using theSwissRegulon database of sequence motifs associated withtranscription factors [49] (see Methods for details). Wefound a putative binding site with a posterior probabilitygreater than 0.7 in the human FOXG1, MECP2 andCDKL5 promoter regions for 23, 11 and 14 TFs respect-ively (Additional file 16: Table S7A). Of these, bindingsites for the three transcription factors RREB1, FOXP1and NFY were found in all three genes, suggesting thatthe three genes implicated in Rett syndrome may bep1@Cdkl5 TATA-less CpG Broadp2@Cdkl5 TATA-less CpG SharpB Human promotersPromoter TATA CpG Shapep1@FOXG1 TATA-less CpG Broadp2@ FOXG1 TATA-less CpG-less Sharpp3@ FOXG1 TATA-less CpG-less Sharpp4@ FOXG1 TATA-less CpG Broadp5@ FOXG1 TATA-less CpG-less Broadp6 @FOXG1 TATA-less CpG-less Broadp7@ FOXG1 TATA-less CpG Broadp9@ FOXG1 TATA-less CpG-less Sharpp1@MECP2 TATA-less CpG Broadp2@ MECP2 TATA-less CpG Broadp5@MECP2 TATA-less CpG-less Broadp1@CDKL5 TATA-less CpG Broadp2@CDKL5 TATA-less CpG SharpPromoters were listed as TATA-less if a TATA-box was absent 500 bp upstreamof the promoter. Similarly in the absence of a CpG island within 500 bp of theTSS, the promoter was classified as CpG-less.regulated by the same TFs in humans. We then summedthe posterior probabilities over each promoter region toestimate the number of binding sites for each transcrip-tion factor and evaluated its statistical significance(Table 3). The data reveal that in human, the sequencearound the main promoter of FOXG1 in human was sig-nificantly enriched in binding sites for the RREB1 (p =0.01), FOXP1 (p = 0.03), and NFY (p = 0.01) transcriptionfactors. NFY was also predicted to regulate MECP2 (p= 0.01) and possibly CDKL5 (p = 0.09). Similar ana-lyses in the mouse genome revealed motifs for 21, 5and 3 TFs within 500 bp of the Foxg1, Mecp2 andthe Cdkl5 promoters respectively (Additional file 15:Table S7B). In mouse although all TFs with bindingsites in the Mecp2 and Cdkl5 promoter regions alsoappeared to have binding sites in the Foxg1 promoter re-gion, only 2 TFs (Sp1 and NFY) were common to allthree genes. Calculating the statistical significance ofthe estimated number of binding sites revealed thatin mouse for all three genes the promoter regionswere enriched for motifs associated with transcriptionfactor NFY, as well as Sp1 (Table 3).DiscussionOur analyses of the FANTOM5 CAGE data revealmultiple sites for transcript initiation and identify thepredominantly used TSSs of the three genes implicated inRTT. Mutation testing for RTT is currently performedsolely on known coding exons, even though it has beensuggested that the non-coding regulatory regions may playa role in the pathogenesis of RTT [50,51]. Our data showthat the highly used TSSs lie upstream of currentlyannotated start sites and we propose that these regions beincluded in testing to ensure accurate representation ofgenes in diagnosis.In our investigation we found the expression of FOXG1strikingly in contrast with the expression of MECP2 in thebrain, but we could not get firm negative correlationfor this observation of discordance in expression. Thisdiscrepancy may be due to the high expression levelof FOXG1 transcripts and the variable but comparativelylow-level expression of MECP2 mRNA in the brain.Alternately, our visual observation may have resulted fromthe fact that some brain regions in mouse (cerebellum andmedulla oblongata) and humans (locus coeruleus, pinealgland, cerebellum, medulla oblongata and substantia nigra)are clearly devoid of FOXG1 expression at any developmen-tal stage. We further confirmed our observation of theabsence of Foxg1 expression in mouse cerebellum throughanalyses of chromatin signatures from mouse ENCODE.Our investigation revealed enrichment of H3K27me3 in theFoxg1 genomic region, suggesting PRC2 mediated silencingof Foxg1 in the cerebellum. Although H3K27me3 has alsobeen reported to be present at transcriptionally active orpeoseshVitezic et al. BMC Genomics 2014, 15:1177 Page 12 of 16http://www.biomedcentral.com/1471-2164/15/1177poised loci [52], the absence of active chromatin marks inthe Foxg1 promoter region in cerebellum but not in thecortex, strongly suggest specific repression of Foxg1 in thecerebellum. A similar examination in liver also revealedH3K27me3 enrichment at the Foxg1 promoter region(Additional file 4: Figure S2). It is tempting to proposePRC2 mediated silencing as a universal mechanism to re-strict Foxg1 expression to brain. It is known that PRC2 me-diated silencing is facilitated through long ncRNAs [40],but our screening did not reveal potential regulatory longFigure 5 Shapes of key promoters of the three genes. Promoter shaas labeled) based on the location of the first nucleotide in all tissues in mseen across the two species in all promoters except pB of Foxg1 in mouhigh correlation between p1 and p2 of CDKL5, we find variation in theirncRNAs in the vicinity of Foxg1 suggesting such regulationmight be mediated by ncRNAs located outside our windowof investigation. It would be interesting to identify the longncRNAs involved in Foxg1 silencing and investigate theircontribution to the disease phenotype.Despite the known discrepancy in mRNA and proteinlevels of MECP2 [18,53]), we found that similar toMeCP2 protein [54], MECP2 mRNA expression was lowin embryonic stages and high in adult stages in most brainregions except the cerebellum, where its expression wasTable 3 Transcription factor binding sites analyses at the proTF p1@FOXG1 p2@FOXG1 p1@MECPRREB1 1.5 (p = 0.01)* 0.0 (p = 1.0) 0.0 (p = 1.0FOXP1 1.3 (p = 0.03)* 0.7 (p = 0.1) 0.0 (p = 1.0NFY 1.9 (p = 0.01)* 0.0 (p = 1.0) 1.3 (p = 0.0TF pA@Foxg1 pB@Foxg1 p1@MecpRREB1 0.0 (p = 1.0) 0.9 (p = 0.06) 0.0 (p = 1.0FOXP1 0.0 (p = 1.0) 0.0 (p = 1.0) 0.0 (p = 1.0NFY 1.9 (p = 0.007)* 1.9 (p = 0.007)* 1.3 (p = 0.0SP1 1.7 (p = 0.04)* 1.4 (p = 0.06) 2.5 (p = 0.0*Denotes significant values.comparatively high in embryonic tissues as well. We alsoexamined the relation between Histone H1 and MECP2 atthe mRNA level. Our data show that in each brain relatedsample, Histone H1 transcript expression is 10–1000 foldgreater than MECP2 transcript expression. Therefore, forthese gene transcripts to produce equal amounts ofprotein, as suggested in Skene et al., massive up-regulationof protein translation is required from MECP2 transcriptsor massive down-regulation of protein translation isneeded from Histone transcripts. Thus our data point tos were drawn for the key promoters of the three genes (panels a to j,use (panels a to e) and humans (panels f to j). Shape conservation isand p2 of FOXG1 in humans. Despite the closeness in location andapes suggesting differential regulation in both species.another layer of regulatory control between transcriptionand translation to equalize the protein output from lowexpressed MECP2 transcripts and abundantly expressedHistone H1 transcripts. The presence of inverted SINEelements in the vicinity of promoters have been reportedto up-regulate protein translation [55] but we did not finda similar configuration of SINE near theMECP2 promoter.The MECP2 gene gives rise to two mRNA isoformswith same transcription start site [56,57] and despite thefact that our analyses revealed two TSSs for MECP2 inmoters of the three genes2 p2@MECP2 p1@CDKL5 p2@CDKL5) 0.0 (p = 1.0) 0.7 (p = 0.08) 0.7 (p = 0.08)) 0.0 (p = 1.0) 0.0 (p = 1.0) 0.0 (p = 1.0)2)* 1.9 (p = 0.01)* 0.9 (p = 0.09) 0.9 (p = 0.09)2 p2@Mecp2 p1@Cdkl5 p2@Cdkl5) 0.34 (p = 0.6) 0.0 (p = 1.0) 0.0 (p = 1.0)) 0.2 (p = 0.7) 0.0 (p = 1.0) 0.0 (p = 1.0)2)* 1.3 (p = 0.02)* 1.0 (p = 0.03)* 1.0 (p = 0.03)*1)* 2.5 (p = 0.01)* 2.6 (p = 0.01)* 2.6 (p = 0.01)*the three genes shown per sample in sheets 1 and 2, and averaged TPMexpression across replicates shown in sheets 3 and 4.are shown. The expression of the other key promoters in these samples isVitezic et al. BMC Genomics 2014, 15:1177 Page 13 of 16http://www.biomedcentral.com/1471-2164/15/1177humans and mouse in all tissues, we could not allocatetwo distinct start sites for the two isoforms of MECP2.Based on our data, we were unable to conclude whetherp2@MECP2/Mecp2 represented an independent poorlyexpressed protein coding isoform, a shorter non-codingregulatory ncRNA transcript arising from the vicinity ofthe main promoter p1@MECP2/Mecp2 or a tissue spe-cific enhancer RNA (eRNA) [54] for MECP2. Almost25% of all enhancers are expected to transcribe shortbi-directional capped transcripts called e-RNAs [54].Our observations of stable low expression level ofp2@MECP2/Mecp2 sometimes below 5 TPM irrespectiveof the expression level of p1@MECP2/Mecp2, its poorcorrelation with p1@MECP2/Mecp2 expression and theabsence bi-directional transcripts at p2@MECP2/Mecp2,do not support its identification as an e-RNA for MECP2.The two TSSs for CDKL5 are highly correlated with eachother in mouse as well as humans. Based on their similarexpression levels and distinct promoter shapes, wepropose that they represent two independently regulatedtranscripts despite their proximity.The comparison between corresponding promoters inhuman and mouse samples, including the novel promoterp1@FOXG1 in human and pA@Foxg1 in mouse, revealedremarkably similar shapes, suggesting evolutionary con-servation in their regulation. The only exceptions were thehuman p2@FOXG1 and pB@Foxg1 mouse, which due totheir distinctive promoter shapes appear to be regulatedin a species-specific manner.The recently released ENCODE Histone ChIP seq data[39], allowed us to distinguish, among our identifiedTSS, active enhancers from active promoters [44,47] inmouse. Despite the presence of enhancer specific histonemark of H3K27ac, we could not find evidence of low-levelantisense transcripts at the enhancer marks in mousesuggesting that enhancers at close range that do notgenerate e-RNAs may regulate the three genes inmouse. For human samples the histone ChIP datawere not available, but we found correlated e-RNAsat distal locations from the TSSs. Our data suggestthat in humans the three genes may be regulated bye-RNA producing enhancers at long range. It is unclear atthis stage whether this discrepancy reflects true species-specific differences or if it reflects differences in dataanalyses (Histone marks with no evidence of e-RNAs inmouse vs e-RNAs alone in humans). Further experimentalvalidation is required to confirm whether these regionsidentified in our study play a regulatory role in theexpression of the respective genes.Almost 20% patients of atypical RTT do not have mu-tations in the three genes. We conducted genome wideTFBS analyses with the aim to discover the commontranscription factors likely to regulate the three genesand thus identify shared pathways upstream. Mutationsalso shown (p2@FOXG1, p3@FOXG1, p2@MECP2 and p2@CDKL5 in humanand pB@Foxg1, p2@Mecp2 and p2@Cdkl5 in mouse).Additional file 4: Figure S2. Silencing of Foxg1 in mouse. UCSCBrowser image of the genomic locus for Foxg1 showing ENCODE tracksfor DNAse-I hypersensitive sites, active enhancer specific histone mark(H3K27ac), active promoter specific histone mark (H3K4me3) and PRC2mediated repressor mark (H3K27me3) in mouse cerebellum, cerebrum,whole brain and liver as labeled. Cerebellum samples lack the DNAse-Ihypersensitive sites visible in cerebrum and whole brain samples.Cerebellum samples also lack the active promoter mark H3K4me3 seen incortex, but contain PRC2 repressive histone mark H3K27me3 not seen incortex at the locus.Additional file 5: Figure S3. Expression levels of Mecp2 and Cdkl5during development in heart kidney and liver. The line plots show thefluctuations in expression for the two promoters for Mecp2 and Cdkl5 inheart, (a and d), kidney (b and e) and liver (c and f) in mouse.Additional file 6: Figure S5. Developmental profile for the 3 genes inhuman brain. Human FOXG1 (a), MECP2 (b) and CDKL5 (c) expression inTPM across a set of adult, newborn and fetal brain regions is shown aslabeled. FOXG1 shows the highest overall expression as well as havingAdditional file 2: Table S2. List of RefSeq and FANTOM5 detectedtranscription start sites in human and mouse.Additional file 3: Figure S1. Top 15 samples in expression for each ofthe three genes. Panels a, c and e represent mouse, while panels b,d andf are human samples showing promoter expression in TPM, on X-axis, invarious tissues, as labeled on Y-axis. For each gene, the samples with thehighest expression of the main promoter (p1@FOXG1, p1@MECP2 andp1@CDKL5 in human and pA@Foxg1, p1@Mecp2 and p1@Cdkl5 in mouse)or functional impairment of such common TFs mayaffect the expression of the three genes, which may resultin disease phenotype. Our data predict that TFs NFY andSP1 are likely to regulate FOXG1 and MECP2 but notCDKL5 in humans and NFY is likely to regulate all threegenes in mouse. Further investigation will be needed toexperimentally verify these findings nevertheless, it will beof interest to study the expression level and presence ofmutations in the common TFs in mutation negative RTTpatients.Our investigations failed to demonstrate brain specificpromoter usage or particularly high levels of expression ofMECP2 in brain or neurons, which could have explainedthe predominantly neurological phenotype seen in patientswith mutations in this ubiquitously expressed gene.ConclusionOur comprehensive analyses of data from the FANTOM5project reveal novel insights into the common and distinctgenomic features of the three genes, which are related notonly by disease phenotype, but also in their regulation in aspecies-specific manner.Additional filesAdditional file 1: Table S1. List of all tissues, cells and cell lines withthe TPM expression of the FANTOM5 defined transcription start sites ofhigher expression in fetal than in adult samples as opposed to theexpression of MECP2 and CDKL5 in the same samples.3. Weaving LS, Christodoulou J, Williamson SL, Friend KL, McKenzie OL, Archer H,Vitezic et al. BMC Genomics 2014, 15:1177 Page 14 of 16http://www.biomedcentral.com/1471-2164/15/1177Additional file 7: Figure S4. Expression profile of the three genes inmouse in developing brain tissues. Line plots showing expression ofselected promoters of Foxg1, Mecp2 and Cdkl5 during development inmouse cerebellum (panel a), mouse visual cortex (panel b) and mousepituitary gland (panel c). Refer main text for details.Additional file 8: Figure S6. Comparison of mRNA levels of Histone H1and MECP2. Bar charts showing TPM expression of the key promoter ofMECP2 and collective total expression of Histone H1 TSSs in brain relatedcells (panels a and c) and tissues (panels b and d) in mouse (panels a and b)and humans (panels c and d). Histone expression levels appear to be over100 fold higher than MECP2 in brain related cells suggesting a massiveup-regulation of MeCP2 at the level of protein translation.Additional file 9: Table S3. Comparison of the expression of theHistone H1 genes promoters and MECP2 in both human and mouse.Additional file 10: Table S4. Pearson and Spearman correlations for allTSSs in human and mouse.Additional file 11: Figure S7. Intra and inter gene expressioncorrelations between the three genes. Expression correlation plots for allother promoter combinations not present in Figure 3. Plots a-g aremouse promoters, while plots h-n are human promoters as labeled.Additional file 12: Table S5. Location of enhancer and promoterspecific Histone marks in relation to TSSs in mouse.Additional file 13: Table S6. Locations and correlations of humanenhancers to the three Rett genes.Additional file 14: Figure S8. Locations of active enhancers correlatedto the three genes in human samples. UCSC snapshot showing thepositions of all eRNA producing human enhancers that are correlated tothe expression of the three genes: FOXG1 (a), MECP2 (b) and CDKL5 (c).Additional file 15: Figure S9. Promoter shapes for all the otherpromoters. The shapes of all the individual promoters in mouse (a-e) andhuman (f-m) are shown as labeled. The shapes are drawn from the firstnucleotide of the first mapped CAGE tag to the first nucleotide of the lastmapped CAGE tag, the y-axis shows the counts in TPM for each position.Additional file 16: Table S7. List of transcription factors with highbinding probability of 0.7 and above to the promoters of the three genesin mouse (A) and human (B) genome. Transcription factors common tothe three genes are shown in red.Competing interestsThe authors declare that they have no competing interests.Authors’ contributionsMV, NB, RA, LL, TL, KH, AS, and A.Saxena analyzed the data, PH, DG, TH, PZ,PA, MF provided human and mouse brain samples, MV and A.Saxena wrotethe manuscript and MV, PC, ARRF and AS planned the project. All authorsread and approved the final manuscript.AcknowledgmentsWe would like to thank all members of the FANTOM5 consortium forcontributing to samples and data analyses and thank GeNAS for dataproduction.RIKEN Omics Science Center ceased to exist as of April 1st, 2013, due toRIKEN reorganization.FundingFANTOM5 was funded by a Research Grant for RIKEN Omics Science Centerfrom MEXT to Yoshihide Hayashizaki and a Grant for Innovative Cell Biologyby Innovative Technology (Cell Innovation Program) from MEXT, to YH, by agrant from MEXT to RIKEN Center for Life Science Technologies and by agrant from MEXT to RIKEN Preventive Medicine and Diagnosis InnovationProgram. MV is supported by an International Program Associate scholarshipfrom RIKEN and a grant from the Frankopani Fund. RA and AS, as well as MV’swork in AS group, was funded by the Lundbeck and Novo NordiskFoundations, and the RiMod-FTD JPND EU joint program. Work by DG, PZ, andTH was funded by Genome BC and NSERC. MF is supported by SFARI SimonsFoundation, AP is supported by Rett Syndrome Foundation. A.Saxena issupported by a Funding program for next generation world leading ResearchersEvans J, Clarke A, Pelka GJ, Tam PP, Watson C, Lahooti H, Ellaway CJ, Bennetts B,Leonard H, Gecz J: Mutations of CDKL5 cause a severe neurodevelopmentaldisorder with infantile spasms and mental retardation. Am J Hum Genet 2004,75(6):1079–1093.4. Neul JL, Kaufmann WE, Glaze DG, Christodoulou J, Clarke AJ, Bahi-Buisson N,Leonard H, Bailey ME, Schanen NC, Zappella M, Renieri A, Huppke P, Percy AK,RettSearch Consortium: Rett syndrome: revised diagnostic criteria andnomenclature. Ann Neurol 2010, 68(6):944–950.5. Archer HL, Evans J, Edwards S, Colley J, Newbury-Ecob R, O'Callaghan F,Huyton M, O'Regan M, Tolmie J, Sampson J, Clarke A, Osborne J: CDKL5mutations cause infantile spasms, early onset seizures, and severemental retardation in female patients. J Med Genet 2006, 43(9):729–734.6. Georgel PT, Horowitz-Scherer RA, Adkins N, Woodcock CL, Wade PA,Hansen JC: Chromatin compaction by human MeCP2. Assembly of novelsecondary chromatin structures in the absence of DNA methylation.J Biol Chem 2003, 278(34):32181–32188.7. Jones PL, Veenstra GJ, Wade PA, Vermaak D, Kass SU, Landsberger N,Strouboulis J, Wolffe AP: Methylated DNA and MeCP2 recruit histonedeacetylase to repress transcription. Nat Genet 1998, 19(2):187–191.8. Nan X, Tate P, Li E, Bird A: DNA methylation specifies chromosomallocalization of MeCP2. Mol Cell Biol 1996, 16(1):414–421.9. Nikitina T, Shi X, Ghosh RP, Horowitz-Scherer RA, Hansen JC, Woodcock CL:Multiple modes of interaction between the methylated DNA bindingprotein MeCP2 and chromatin. Mol Cell Biol 2007, 27(3):864–877.10. Chahrour M, Jung SY, Shaw C, Zhou X, Wong ST, Qin J, Zoghbi HY: MeCP2,a key contributor to neurological disease, activates and repressestranscription. Science 2008, 320(5880):1224–1229.11. Nan X, Ng HH, Johnson CA, Laherty CD, Turner BM, Eisenman RN, Bird A:Transcriptional repression by the methyl-CpG-binding protein MeCP2by MEXT, HFSP (RGP0014/2012) to PC, a JSPS International Fellowship (P09745)and currently by the National Institute of Health Research funded BiomedicalResearch Centre at Guy’s and St. Thomas’ Trust, London, UK.Author details1Omics Science Center, RIKEN Yokohama Institute, Omics Science Center(OSC), 1-17-22 Suehiro cho, Tsurumi ku, Yokohama, Japan. 2Department ofCell and Molecular Biology (CMB), Karolinska Institutet, Stockholm, Sweden.3The Bioinformatics Center, Department of Biology and Biotech Research andInnovation Center, University of Copenhagen, Copenhagen, Denmark. 4RIKENCenter for Life Science Technologies, Division of Genomic Technologies(DGT), Yokohama, Japan. 5Cancer Science Institute of Singapore, NationalUniversity of Singapore, Singapore, Singapore. 6Center for MolecularMedicine and Genetics, Wayne State University, Detroit, MI, USA.7Department of Neurology, School of Medicine, Wayne State University,Detroit, MI, USA. 8RIKEN Preventive Medicine and Diagnosis InnovationProgram (PMI), Wako, Japan. 9Telethon Kids Institute, The University ofWestern Australia, Perth, Australia. 10German Center for NeurodegenerativeDiseases (DZNE), Tübingen, Germany. 11Eberhard Karls University, Tübingen,Germany. 12Centre for Molecular Medicine and Therapeutics, Child andFamily Research Institute, Dept of Medical Genetics, University of BritishColumbia, Vancouver, Canada. 13FM Kirby Neurobiology Center, Departmentof Neurology, Boston Children’s Hospital, Harvard Medical School, Boston,MA, USA. 14Currently at: Biomedical Research Centre at Guy’s and St Thomas’Trust, Genomics Core Facility, Guy’s Hospital, London, UK.Received: 17 November 2013 Accepted: 4 December 2014Published: 24 December 2014References1. Amir RE, Van den Veyver IB, Wan M, Tran CQ, Francke U, Zoghbi HY: Rettsyndrome is caused by mutations in X-linked MECP2, encodingmethyl-CpG-binding protein 2. Nat Genet 1999, 23(2):185–188.2. Ariani F, Hayek G, Rondinella D, Artuso R, Mencarelli MA, Spanhol-Rosseto A,Pollazzon M, Buoni S, Spiga O, Ricciardi S, Meloni I, Longo I, Mari F, Broccoli V,Zappella M, Renieri A: FOXG1 is responsible for the congenital variant of Rettsyndrome. Am J Hum Genet 2008, 83(1):89–93.involves a histone deacetylase complex. Nature 1998, 393(6683):386–389.12. Yasui DH, Peddada S, Bieda MC, Vallero RO, Hogart A, Nagarajan RP,Thatcher KN, Farnham PJ, Lasalle JM: Integrated epigenomic analyses ofVitezic et al. BMC Genomics 2014, 15:1177 Page 15 of 16http://www.biomedcentral.com/1471-2164/15/1177neuronal MeCP2 reveal a role for long-range interaction with activegenes. Proc Natl Acad Sci U S A 2007, 104(49):19416–19421.13. Nan X, Campoy FJ, Bird A: MeCP2 is a transcriptional repressor withabundant binding sites in genomic chromatin. Cell 1997, 88(4):471–481.14. Meehan RR, Lewis JD, Bird AP: Characterization of MeCP2, a vertebrateDNA binding protein with affinity for methylated DNA. Nucleic Acids Res1992, 20(19):5085–5092.15. Chahrour M, Zoghbi HY: The story of Rett syndrome: from clinic toneurobiology. Neuron 2007, 56(3):422–437.16. Ramocki MB, Peters SU, Tavyev YJ, Zhang F, Carvalho CM, Schaaf CP,Richman R, Fang P, Glaze DG, Lupski JR, Zoghbi HY: Autism and otherneuropsychiatric symptoms are prevalent in individuals with MeCP2duplication syndrome. Ann Neurol 2009, 66(6):771–782.17. Ramocki MB, Tavyev YJ, Peters SU: The MECP2 duplication syndrome. Am JMed Genet A 2012, 152A(5):1079–1088.18. Shahbazian MD, Antalffy B, Armstrong DL, Zoghbi HY: Insight into Rettsyndrome: MeCP2 levels display tissue- and cell-specific differences andcorrelate with neuronal maturation. Hum Mol Genet 2002, 11(2):115–124.19. Skene PJ, Illingworth RS, Webb S, Kerr AR, James KD, Turner DJ, Andrews R,Bird AP: Neuronal MeCP2 is expressed at near histone-octamer levelsand globally alters the chromatin state. Mol Cell 2010, 37(4):457–468.20. Sonderegger CK, Vogt PK: Binding of the corepressor TLE1 to Qinenhances Qin-mediated transformation of chicken embryo fibroblasts.Oncogene 2003, 22(12):1749–1757.21. Rusconi L, Salvatoni L, Giudici L, Bertani I, Kilstrup-Nielsen C, Broccoli V,Landsberger N: CDKL5 expression is modulated during neuronaldevelopment and its subcellular distribution is tightly regulated by theC-terminal tail. J Biol Chem 2008, 283(44):30101–30111.22. Williamson SL, Giudici L, Kilstrup Nielsen C, Gold W, Pelka GJ, Tam PP,Grimm A, Prodi D, Landsberger N, Christodoulou J: A novel transcript ofcyclin-dependent kinase-like 5 (CDKL5) has an alternative C-terminus andis the predominant transcript in brain. Hum Genet 2012, 131(2):187–200.23. Montini E, Andolfi G, Caruso A, Buchner G, Walpole SM, Mariani M, ConsalezG, Trump D, Ballabio A, Franco B: Identification and characterization of anovel serine-threonine kinase gene from the Xp22 region. Genomics1998, 51(3):427–433.24. Kilstrup-Nielsen C, Rusconi L, La Montanara P, Ciceri D, Bergo A, Bedogni F,Landsberger N: What we know and would like to know about CDKL5 andits involvement in epileptic encephalopathy. Neural Plast 2012, 2012:728267.25. Roth M, Bonev B, Lindsay J, Lea R, Panagiotaki N, Houart C, Papalopulu N:FoxG1 and TLE2 act cooperatively to regulate ventral telencephalonformation. Development 2010, 137(9):1553–1562.26. Yao J, Lai E, Stifani S: The winged-helix protein brain factor 1 interactswith groucho and hes proteins to repress transcription. Mol Cell Biol 2001,21(6):1962–1972.27. Tao J, Hu K, Chang Q, Wu H, Sherman NE, Martinowich K, Klose RJ, Schanen C,Jaenisch R, Wang W, Sun YE: Phosphorylation of MeCP2 at Serine 80regulates its chromatin association and neurological function. Proc NatlAcad Sci U S A 2009, 106(12):4882–4887.28. Carouge D, Host L, Aunis D, Zwiller J, Anglard P: CDKL5 is a brain MeCP2target gene regulated by DNA methylation. Neurobiol Dis 2010,38(3):414–424.29. Kalscheuer VM, Tao J, Donnelly A, Hollway G, Schwinger E, Kubart S, Menzel C,Hoeltzenbein M, Tommerup N, Eyre H, Harbord M, Haan E, Sutherland GR,Ropers HH, Gécz J: Disruption of the serine/threonine kinase 9 gene causessevere X-linked infantile spasms and mental retardation. Am J Hum Genet2003, 72(6):1401–1411.30. Mari F, Azimonti S, Bertani I, Bolognese F, Colombo E, Caselli R, Scala E,Longo I, Grosso S, Pescucci C, Ariani F, Hayek G, Balestri P, Bergo A,Badaracco G, Zappella M, Broccoli V, Renieri A, Kilstrup-Nielsen C,Landsberger N: CDKL5 belongs to the same molecular pathway of MeCP2and it is responsible for the early-onset seizure variant of Rett syndrome.Hum Mol Genet 2005, 14(14):1935–1946.31. FANTOM Consortium and the RIKEN PMI and CLST (DGT), Forrest AR, Kawaji H,Rehli M, Baillie JK, de Hoon MJ, Lassmann T, Itoh M, Summers KM, Suzuki H,Daub CO, Kawai J, Heutink P, Hide W, Freeman TC, Lenhard B, Bajic VB, Taylor MS,Makeev VJ, Sandelin A, Hume DA, Carninci P, Hayashizaki Y: A promoter-levelmammalian expression atlas. Nature 2014, 507(7493):462–470.32. Stamatoyannopoulos JA, Snyder M, Hardison R, Ren B, Gingeras T, Gilbert DM,Groudine M, Bender M, Kaul R, Canfield T, Giste E, Johnson A, Zhang M,Balasundaram G, Byron R, Roach V, Sabo PJ, Sandstrom R, Stehling AS,Thurman RE, Weissman SM, Cayting P, Hariharan M, Lian J, Cheng Y, Landt SG,Ma Z, Wold BJ, Dekker J, Crawford GE, et al: An encyclopedia of mouse DNAelements (Mouse ENCODE). Genome Biol 2012, 13(8):418.33. Shahbazian M, Young J, Yuva-Paylor L, Spencer C, Antalffy B, Noebels J,Armstrong D, Paylor R, Zoghbi H: Mice with truncated MeCP2 recapitulatemany Rett syndrome features and display hyperacetylation of histoneH3. Neuron 2002, 35(2):243–254.34. Andersson R, Gebhard C, Miguel-Escalada I, Hoof I, Bornholdt J, Boyd M, ChenY, Zhao X, Schmidl C, Suzuki T, Ntini E, Arner E, Valen E, Li K, Schwarzfischer L,Glatz D, Raithel J, Lilje B, Rapin N, Bagger FO, Jørgensen M, Andersen PR, BertinN, Rackham O, Burroughs AM, Baillie JK, Ishizu Y, Shimizu Y, Furuhata E, MaedaS, et al: An atlas of active enhancers across human cell types and tissues.Nature 2014, 507(7493):455–461.35. Quinlan AR, Hall IM: BEDTools: a flexible suite of utilities for comparinggenomic features. Bioinformatics 2012, 26(6):841–842.36. Meyer LR, Zweig AS, Hinrichs AS, Karolchik D, Kuhn RM, Wong M, Sloan CA,Rosenbloom KR, Roe G, Rhead B, Raney BJ, Pohl A, Malladi VS, Li CH, Lee BT,Learned K, Kirkup V, Hsu F, Heitner S, Harte RA, Haeussler M, Guruvadoo L,Goldman M, Giardine BM, Fujita PA, Dreszer TR, Diekhans M, Cline MS,Clawson H, Barber GP: The UCSC Genome Browser database: extensionsand updates 2013. Nucleic Acids Res 2013, 41(Database issue):D64–D69.37. Notredame C, Higgins DG, Heringa J: T-Coffee: A novel method for fast andaccurate multiple sequence alignment. J Mol Biol 2000, 302(1):205–217.38. Arnold P, Erb I, Pachkov M, Molina N, van Nimwegen E:MotEvo: integratedBayesian probabilistic methods for inferring regulatory sites and motifs onmultiple alignments of DNA sequences. Bioinformatics 2012, 28(4):487–494.39. Dong X, Greven MC, Kundaje A, Djebali S, Brown JB, Cheng C, Gingeras TR,Gerstein M, Guigo R, Birney E, Weng Z: Modeling gene expression usingchromatin features in various cellular contexts. Genome Biol 2012, 13(9):R53.40. Saxena A, Carninci P: Long non-coding RNA modifies chromatin: epigeneticsilencing by long non-coding RNAs. Bioessays 2011, 33(11):830–839.41. Alvarez-Saavedra M, Carrasco L, Sura-Trueba S, Demarchi Aiello V, Walz K,Neto JX, Young JI: Elevated expression of MeCP2 in cardiac and skeletaltissues is detrimental for normal development. Hum Mol Genet 2010,19(11):2177–2190.42. Hanashima C, Shen L, Li SC, Lai E: Brain factor-1 controls the proliferationand differentiation of neocortical progenitor cells through independentmechanisms. J Neurosci 2002, 22(15):6526–6536.43. Seoane J, Le HV, Shen L, Anderson SA, Massague J: Integration of Smadand forkhead pathways in the control of neuroepithelial andglioblastoma cell proliferation. Cell 2004, 117(2):211–223.44. Creyghton MP, Cheng AW, Welstead GG, Kooistra T, Carey BW, Steine EJ,Hanna J, Lodato MA, Frampton GM, Sharp PA, Boyer LA, Young RA, Jaenisch R:Histone H3K27ac separates active from poised enhancers and predictsdevelopmental state. Proc Natl Acad Sci U S A 2010, 107(50):21931–21936.45. Ernst J, Kheradpour P, Mikkelsen TS, Shoresh N, Ward LD, Epstein CB, Zhang X,Wang L, Issner R, Coyne M, Ku M, Durham T, Kellis M, Bernstein BE: Mappingand analysis of chromatin state dynamics in nine human cell types.Nature 2011, 473(7345):43–49.46. Heintzman ND, Hon GC, Hawkins RD, Kheradpour P, Stark A, Harp LF, Ye Z,Lee LK, Stuart RK, Ching CW, Ching KA, Antosiewicz-Bourget JE, Liu H,Zhang X, Green RD, Lobanenkov VV, Stewart R, Thomson JA, Crawford GE,Kellis M, Ren B: Histone modifications at human enhancers reflect globalcell-type-specific gene expression. Nature 2009, 459(7243):108–112.47. Rada-Iglesias A, Bajpai R, Swigut T, Brugmann SA, Flynn RA, Wysocka J: Aunique chromatin signature uncovers early developmental enhancers inhumans. Nature 2011, 470(7333):279–283.48. Carninci P, Sandelin A, Lenhard B, Katayama S, Shimokawa K, Ponjavic J,Semple CA, Taylor MS, Engstrom PG, Frith MC, Forrest AR, Alkema WB,Tan SL, Plessy C, Kodzius R, Ravasi T, Kasukawa T, Fukuda S,Kanamori-Katayama M, Kitazume Y, Kawaji H, Kai C, Nakamura M, Konno H,Nakano K, Mottagui-Tabar S, Arner P, Chesi A, Gustincich S, Persichetti F, et al:Genome-wide analysis of mammalian promoter architecture andevolution. Nat Genet 2006, 38(6):626–635.49. Pachkov M, Balwierz PJ, Arnold P, Ozonov E, van Nimwegen E: SwissRegulon,a database of genome-wide annotations of regulatory sites: recent up-dates. Nucleic Acids Res 2013, 41(Database issue):D214–D220.50. Ellaway CJ, Ho G, Bettella E, Knapman A, Collins F, Hackett A, McKenzie F,Darmanian A, Peters GB, Fagan K, Christodoulou J: 14q12 microdeletionsexcluding FOXG1 give rise to a congenital variant Rett syndrome-likephenotype. Eur J Hum Genet 2013, 21(5):522–527.51. Saxena A, de Lagarde D, Leonard H, Williamson SL, Vasudevan V,Christodoulou J, Thompson E, MacLeod P, Ravine D: Lost in translation:translational interference from a recurrent mutation in exon 1 of MECP2.J Med Genet 2006, 43(6):470–477.52. Young MD, Willson TA, Wakefield MJ, Trounson E, Hilton DJ, Blewitt ME,Oshlack A, Majewski IJ: ChIP-seq analysis reveals distinct H3K27me3profiles that correlate with transcriptional activity. Nucleic Acids Res 2011,39(17):7415–7427.53. Balmer D, Goldstine J, Rao YM, LaSalle JM: Elevated methyl-CpG-bindingprotein 2 expression is acquired during postnatal human braindevelopment and is correlated with alternative polyadenylation. J MolMed 2003, 81(1):61–68.54. Kim TK, Hemberg M, Gray JM, Costa AM, Bear DM, Wu J, Harmin DA,Laptewicz M, Barbara-Haley K, Kuersten S, Markenscoff-Papadimitriou E,Kuhl D, Bito H, Worley PF, Kreiman G, Greenberg ME:Widespread transcriptionat neuronal activity-regulated enhancers. Nature 2010, 465(7295):182–187.55. Carrieri C, Cimatti L, Biagioli M, Beugnet A, Zucchelli S, Fedele S, Pesce E,Ferrer I, Collavin L, Santoro C, Forrest AR, Carninci P, Biffo S, Stupka E,Gustincich S: Long non-coding antisense RNA controls Uchl1 translationthrough an embedded SINEB2 repeat. Nature 2012, 491(7424):454–457.56. Kriaucionis S, Bird A: The major form of MeCP2 has a novel N-terminusgenerated by alternative splicing. Nucleic Acids Res 2004, 32(5):1818–1823.57. Mnatzakanian GN, Lohi H, Munteanu I, Alfred SE, Yamada T, MacLeod PJ,Jones JR, Scherer SW, Schanen NC, Friez MJ, Vincent JB, Minassian BA:A previously unidentified MECP2 open reading frame defines a newprotein isoform relevant to Rett syndrome. Nat Genet 2004, 36(4):339–341.doi:10.1186/1471-2164-15-1177Cite this article as: Vitezic et al.: CAGE-defined promoter regions of thegenes implicated in Rett Syndrome. BMC Genomics 2014 15:1177.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionVitezic et al. BMC Genomics 2014, 15:1177 Page 16 of 16http://www.biomedcentral.com/1471-2164/15/1177Submit your manuscript at www.biomedcentral.com/submit


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items