Cytosine methylation changes in enhancer regions of core pro-fibrotic genes characterize kidney fibrosis development
© Ko et al.; licensee BioMed Central Ltd. 2013
Received: 9 April 2013
Accepted: 7 October 2013
Published: 10 December 2013
One in eleven people is affected by chronic kidney disease, a condition characterized by kidney fibrosis and progressive loss of kidney function. Epidemiological studies indicate that adverse intrauterine and postnatal environments have a long-lasting role in chronic kidney disease development. Epigenetic information represents a plausible carrier for mediating this programming effect. Here we demonstrate that genome-wide cytosine methylation patterns of healthy and chronic kidney disease tubule samples obtained from patients show significant differences.
We identify differentially methylated regions and validate these in a large replication dataset. The differentially methylated regions are rarely observed on promoters, but mostly overlap with putative enhancer regions, and they are enriched in consensus binding sequences for important renal transcription factors. This indicates their importance in gene expression regulation. A core set of genes that are known to be related to kidney fibrosis, including genes encoding collagens, show cytosine methylation changes correlating with downstream transcript levels.
Our report raises the possibility that epigenetic dysregulation plays a role in chronic kidney disease development via influencing core pro-fibrotic pathways and can aid the development of novel biomarkers and future therapeutics.
Clinical retrospective data indicate that altered nutrient availability during development could have a long lasting effect on the development of adult diseases, a phenomenon called 'programming'. Hypertension and chronic kidney disease (CKD) show one of the highest sensitivities to intrauterine programming . Epigenetic changes caused by altered intrauterine nutrient availability have been proposed as the mechanistic link for hypertension and CKD development . Epigenetic modifications are inherited during cell division, thus solidifying 'the memory or programming' effects of the environment . The epigenome, which includes the covalent modifications of DNA and its associated proteins and defines DNA accessibility to the transcriptional machinery, is the key determinant of outcome after transcription factor binding. At the root of the epigenetic modifications is the direct chemical modification of cytosines by methylation . In different cancer types, hypermethylation of tumor suppressor gene promoters has been observed . Increased promoter methylation can interfere with transcription factor binding, causing loss of tumor suppressor expression, thereby contributing to the malignant transformation [6, 7]. Agents that reduce cytosine methylation (for example, azacytidine) are now in clinical use and are associated with improvements in clinical outcome, especially for patients with myelodysplastic syndrome . In addition, mutations of different chromatin-modifying enzymes have been described in various cancer types, contributing to alterations in the cancer epigenome .
Not much is known about the epigenome of chronic human diseases other than cancer. Most previous studies have been performed on cultured cells, animal models, or surrogate cell types (mostly circulating mononuclear cells) . As the epigenome is cell type-specific, little mechanistic information can be drawn from cultured cells and surrogate cell types . To understand whether or not epigenetic changes occur and thereby potentially contribute to CKD development in patients, we performed genome-wide cytosine methylation profiling of tubule epithelial cells obtained from CKD and control kidneys. We found that core fibrosis-related genes show cytosine methylation changes in their gene regulatory regions. In vitro studies indicate that cytosine methylation differences play a role in regulating transcript expression. Examining the CKD epigenome can be an important first step in understanding the role of epigenetics outside the cancer field .
CKD kidneys show distinct cytosine methylation profiles
Demographic, clinical and histological characteristics of the samples
Age (years) mean ± SD
68.0 ± 10.81
61.14 ± 11.2
Asian, Pacific Islander
Other and unknown
Height (cm) mean ± SD
165 ± 8.69
166.5 ± 8.63
Weight (kg) mean ± SD
78.0 ± 22.02
BMI (kg/m2) mean ± SD
27.85 ± 6.41
31.25 ± 5.58
3.0 ± 1.83
0.36 ± 0.81
Serum BUN (mg/dL) mean ± SD
35.0 ± 14.7
Serum creatinine (mg/dL) mean ± SD
3.0 ± 1.61
1.08 ± 0.18
eGFR (ml/minute/1.73 m2) mean ± SD
29.0 ± 13.68
70.94 ± 8.35
31.0 ± 31.35
Mesangial matrix expansion
1 ± 0.91
0.17 ± 0.39
Tubular atrophy (%)
34.0 ± 24.94
9.82 ± 15.76
Interstitial fibrosis (%)
34.0 ± 25.15
5.68 ± 5.07
2.0 ± 0.78
0.9 ± 1.1
2.0 ± 0.78
0.29 ± 0.62
Gene ontology annotation showed that genes around the DMRs are enriched for cell adhesion and development related functions including: collagen, fibronectin, transforming growth factor beta (TGFβ) and Smad proteins (Figure 1C), many of these genes are known to play a critical role in CKD development. In summary, microdissected kidney tubule cells showed distinct differences in their cytosine methylation patterns in CKD.
Validation and external replication of the results
Internal validation of the results was performed using site-specific primer-based amplification of bisulfite-converted genomic DNA and MassArray Epityper (Sequenom) quantification of modified cytosines . This method is based on mass spectrometry that allows us to determine absolute methylation levels. We correlated these (mass array based) absolute methylation levels with the HpaII/MspI relative ratios (Additional file 4).
Significant methylation differences were detected for 1,061 genes (corresponding to 98% of the genes in the original dataset; Figure 2A). The complete list of DMRs in the original and the replication dataset can be found in Additional file 5.
Locus-specific validation was performed for six different genes, including COLIVA1. COLIV4A1/A2 are critical basement membrane proteins synthesized by epithelial cells. Increased expression is known to be responsible for increasing the thickness of the basement membrane and it is considered to be an early change in progressive kidney fibrosis . The COLIVA1 and COLIVA2 transcripts are transcribed from a single promoter (Figure 2B). This locus showed significantly lower cytosine methylation of CKD samples (Figure 2C). We examined the absolute methylation level of COLIV4A1/2 by MassArray Epityper analysis (Figure 2D) in control and CKD samples and confirmed the methylation differences between healthy and diseased tubule epithelial cells. Next we examined COLIVA1/2 methylation in the validation dataset (Infinium arrays from 66 control and 21 DKD samples). Using this dataset we also confirmed the predominant (2 to 12%) hypomethylation of this locus (Figure 2E). The methylation differences correlated with increased COLIVA1 transcript (Figure 2F) and protein levels (Figure 2G). Using the MassArray Epityper we also validated the methylation status of additional loci (Figure S3A,B in Additional file 6). In summary, the methylation differences appear to be highly consistent between the original and validation experiments using multiple different methods.
Differentially methylated loci are enriched in kidney-specific gene regulatory regions
To further understand the functional significance of the DMRs, we generated genome-wide chromatin annotation maps using cultured human proximal tubular epithelial cells (HKC8). First, we performed chromatin immunoprecipitation followed by next-generation sequencing (ChIP-seq) for a panel of important histone modifications: H3K4me1, H3K4me2, H3K4me3, H3K27ac, H3K27me3, and H3K36me3. Next, we generated gene regulatory annotation maps from the panel of ChIP-seq data using the hidden Markov model-based ChromHMM chromatin segmentation program [17, 18]. Consistent with the RefSeq-based annotation, there are very few DMRs localized to ChromHMM-annotated kidney promoter regions (Figure 3B). The analysis indicated that 30% of the DMRs localized to enhancer regions, which was the most significant enrichment. Similar results were obtained when we generated adult kidney cortex ChromHMM maps (from published ChIP-seq data; Figure 3B) . Next, we compared CKD-specific DMRs with chromatin annotation maps of other, different cell types using the publicly available ENCODE database (Figure 3C). We found that CKD-specific DMRs localized mostly to repressed chromatin regions, while transcription and enhancer regions showed the second highest enrichment. The result indicates that DMR in CKD are enriched in kidney-specific gene regulatory regions, mainly (intronic) enhancers.
Differential methylated regions are functionally relevant and correlate with transcript levels
Differentially methylated region methylation drives gene expression in vitro
To further dissect the relationship between cytosine methylation and transcript level changes, we analyzed gene expression and cytosine methylation patterns of tubule epithelial cells at both baseline and 9 days after treatment with a DNA methyltransferase inhibitor, decitabine (5-aza-2-dexoycytidine). We used AffymetrixST1.0 arrays to compare gene expression changes and the Infinium 450K arrays to examine cytosine methylation changes in control (n = 3) and decitabine-treated cells (n = 4).
Discussion and conclusion
While epigenetic dysregulation has been suggested as a mechanism for the development of many diseases, little is known about the epigenome of normal and diseased human cells and organs. Here we describe cytosine methylation differences in tubule cells obtained from patients with CKD. We found that CKD DMRs have many special features. First, most loci showed consistent cytosine methylation differences in different forms of CKD. These changes were smaller compared with what has described in the cancer literature previously. While the absolute differences were modest, the identified loci showed highly consistent changes even across different datasets and platforms. Unexpectedly, we found that most methylation differences localized outside of promoter areas, with promoter regions markedly spared from cytosine methylation differences. Our results indicate that the differentially methylated regions were located mainly at candidate enhancers. We found that the DMRs contain consensus-binding motifs for key renal transcription factors (HNF, TCFAP, SIX2). Furthermore, cytosine methylation levels correlated with baseline gene expression changes. These epigenetically distinct but morphologically similar cells also showed differences in their cytokine response. We illustrated our findings in a model hypothesizing that enhancer DMRs might modify transcription factor binding and thereby downstream transcript levels.
Based on our results, we propose that cytosine methylation changes are causally linked to transcript levels and phenotype development. As hypertensive and diabetic tubule samples showed similarities (both in cytosine methylation and gene expression changes), the observed changes are likely to be part of a common mechanism of progression. This may be expected, as phenotypically the tubulointerstitial fibrosis of DKD and hypertensive CKD is similar. In addition, we found that DMRs were enriched for genes related to development, many of them no longer expressed in the adult kidney. The DMR regions also contained binding sites for key kidney developmental factors (such as SIX2, HNF, and TCFAP). One possible interpretation of our findings is that the epigenetic differences are established during development. This is the time when the cell type-specific epigenome is established and when these genes and transcription factors play functional roles. Therefore they can possibly provide the mechanistic link between fetal programming and CKD development - the Brenner-Barker hypothesis put forward many decades ago [26, 27], proposing that nutrient availability during development has a long lasting programming role in hypertension and CKD development. In addition, reactivation of the developmental pathways is also needed during organ injury repair . We can also speculate that the altered developmental wiring of these pathways could continue to play a role later on as alterations observed after repair. Indeed, control and CKD kidney epithelial cells showed not only cytosine methylation differences but also different responses to cytokine treatment.
A limitation of our results remains that our samples were collected in a single center. Furthermore, base pair resolution results will likely help to refine the more precise location of DMRs and the methylation differences in the future. Furthermore, while microdissection is an excellent separation method to generate a homogenous tubular epithelial cell population from the kidney, the potential risk for increased cell type heterogeneity in CKD remains. As isolated and cultured cells continued to show many of the epigenetic and transcriptional differences, it is more likely that the observed differences are not related to cell type heterogeneity.
In summary, while it has long been speculated that epigenetic dysregulation might occur in non-cancerous diseases, including CKD, here we provide experimental evidence for cytosine methylation changes in human kidney tissue samples, opening the possibility that they play a role in CKD development.
Materials and methods
The clinical study used the cross-sectional design. Kidney samples were obtained from routine surgical nephrectomies. Samples were de-identified and the corresponding clinical information was collected by an individual who was not involved in the research protocol. The study was approved by the Institutional Review Boards of the Albert Einstein College of Medicine Montefiore Medical Center (IRB#2002-202) and the University of Pennsylvania. Histological analysis was performed by an expert pathologist (IRB#815796).
Tissue handling and microdissection
Tissue was placed into RNALater and manually microdissected at 4°C for glomerular and tubular compartments as described earlier. Dissected tissue was homogenized and RNA was prepared using RNAeasy mini columns (Qiagen, Valencia, CA, USA) according to the manufacturer’s instructions. RNA quality and quantity was determined using Lab-on-Chip Total RNA PicoKit (Agilent BioAnalyzer, Santa Clara, CA,USA). Only samples without evidence of degradation were used. Genomic DNA was extracted by phenol chloroform protocol for HELP analysis and the DNAeasy kit was used for the Infinium platform.
DNA methylation analysis by HELP
The HELP assay was carried out as previously published . Intact DNA of high molecular weight was corroborated by electrophoresis on 1% agarose gels in all cases. One microgram of genomic DNA was digested overnight with either HpaII or MspI (NEB, Ipswich, MA, USA). The digested DNA was used to set up an overnight ligation of the HpaII adapter using T4 DNA ligase. The adapter-ligated DNA was used to carry out the PCR amplification of the HpaII- and MspI-digested DNA as previously described . Both amplified fractions were submitted to Roche-NimbleGen, Inc. (Madison, WI, USA) for labeling and hybridization onto a human hg18 high-density custom-designed oligonucleotide array (50-mers) containing 2.6 million loci. HpaII amplifiable fragments are defined as genomic sequences contained between two flanking HpaII sites found within 200 to 2,000 bp of each other. All microarray hybridizations were subjected to extensive quality control using the following strategies. First, uniformity of hybridization was evaluated using a modified version of a previously published algorithm  adapted for the NimbleGen platform, and any hybridization with strong regional artifacts was discarded and repeated. The raw data can be accessed under GSE49557.
HELP data processing and analysis
Signal intensities at each HpaII amplifiable fragment were calculated as a robust (25% trimmed) mean of their component probe-level signal intensities. Any fragments found within the level of background MspI signal intensity, measured as 2.5 mean absolute differences (MAD) above the median of random probe signals, were categorized as 'failed'. These 'failed' loci therefore represent the population of fragments that did not amplify by PCR, whatever the biological (for example, genomic deletions and other sequence errors) or experimental cause. On the other hand, 'methylated' loci were so designated when the level of HpaII signal intensity was similarly indistinguishable from background. PCR-amplifying fragments (those not flagged as either 'methylated' or 'failed') were normalized using an intra-array quantile approach wherein HpaII/MspI ratios are aligned across density-dependent sliding windows of fragment size-sorted data. The log2 (HpaII/MspI) was used as a representative for methylation and analyzed as a continuous variable. For most loci, each fragment was categorized as either methylated, if the centered log HpaII/MspI ratio was less than zero, or hypomethylated if the log ratio was greater than zero.
Statistical analysis of HELP data was performed using the statistical software R version 2.13.1 . A two-sample t-test was used for each gene or locus to summarize methylation differences between the two clinical groups (cases and controls). Genes were ranked on the basis of the magnitude of this test statistic and a set of differentially methylated loci with P-value <0.01 and a fold change >0.5 was identified.
Quantitative DNA methylation analysis by MassArray epityping
Validation of HELP microarray findings was carried out by matrix-assisted laser desorption/ionisation-time of flight (MALDI-TOF) mass spectrometry using EpiTyper by MassArray (Sequenom, San Diego, CA, USA) on bisulfite-converted DNA as previously described . MassArray primers were designed to cover the flanking HpaII sites for a given HpaII-amplifiable fragments (HAF), as well as any other HpaII sites found up to 2,000 bp upstream of the downstream site and up to 2,000 bp downstream of the upstream site, in order to cover all possible alternative sites of digestion. HAF is defined by those fragments where two HpaII sites are located 200–2000 bp apart with at least some unique sequence between them and selected those located at gene promoters and imprinted regions.
Gene expression analysis using Affymetrix arrays
Transcript levels were analyzed using Affymetrix U133A and 1.0ST arrays. Probes were prepared using an Affymetrix 3′ IVT kit. After hybridization and scanning, raw data files were imported into Genespring GX software (Agilent Technologies). Raw expression levels were normalized using the RMA16 summarization algorithm. Genespring GX software was then used for statistical analysis; the data were above the 20th percentile when filtered by expression. We used a Benjamini-Hochberg multiple testing correction with a P-value <0.05. Both heatmap of methylation data and gene expression data were generated using an unsupervised hierarchical clustering method calculated by squared Euclidean distances. Methylation data used in clustering have a P-value <0.00015 and a fold change ≥0.5. The raw data can be accessed through accession GSE48944.
Gene ontology and transcription factor binding sites
The Database for Annotation, Visualization and Integrated Discovery (DAVID) bioinformatics package was used for gene ontology and pathway analysis. In addition, Ingenuity Pathway Analysis (IPA, Redwood City, CA, USA) was used to generate networks.
Sequences of DMRs (n = 4,751) were lifted over from hg18 to hg19 using UCSC Genome Browser Utilities. The regions were then intersected with fetal kidney or human kidney epithelial-specific DHS peaks (data from GEO GSM530655); a total of 364 overlapping regions were used. Motif weight matrices overrepresented in the overlapped sequences were identified using MEME version 4.8.0  on the 364 regions with parameter -oc -nmotifs 10 -minw 8 -maxw 50.
Adult kidney ChIP-seq data were downloaded from the Roadmap database (GEO accessions GSM670025 for adult kidney and GSM621638 for adult kidney input). The overlap was set to be a minimum of 1 bp in length.
We compared de novo motifs to motifs available as part of various databases, including Transfac, version 2011.1, Jaspar Core, and UniPROBE using TOMTOM software , version 4.8.1. TOMTOM parameters were set to their default values during motif comparisons. When partitioning the de novo motifs, assigning each to a single category, the order of match assignment preference was to Transfac, Jaspar Core, UniPROBE, and then to the novel motif category.
HKC8 cells were kindly provided by Lorainne Racusen (Johns Hopkins University) and were cultured in DMEM/F12 medium supplemented with 2.5% fetal bovine serum, antibiotics and insulin, transferrin and selenium. Cells were incubated with 0.5 μM decitabine on days 2, 4, 6, and 8 and harvested on day 9. RNA was isolated using a Qiagen RNeasy kit labeled using an Ovation transcript labeling kit and hybridized onto Affymetrix Human ST1.0 arrays.
Chromatin immunoprecipitation sequencing
HKC8 cells were harvested and crosslinked with 1% formaldehyde when they reached 80% confluency on culture plates. Chromatin was sheared using a Bioruptor and immunoprecipitated using H3K4me1 (Abcam ab8895, Cambridge, MA, USA), H3K4me2 (Abcam ab11946), H3K4me3 (Abcam ab8580), H3K36me3 (Abcam ab9050), H3K27ac (Abcam ab4729) and H3K27me3 (Millipore 07–499, Billerica, MA, USA) marks. ChIP was performed as described in the manual of MAGnify™ Chromatin Immunoprecipitation System (Invitrogen, Grand Island, NY, USA). Quantitative real-time PCR was performed to ensure the quality of the ChIP product. The ChIP product was assessed for size, purity, and quantity using an Agilent 2100 Bioanalyzer (Agilent Technologies). Library preparation and sequencing were performed at the Einstein Epigenome Center. Sequence reads (100 bp) were generated from llumina HiSeq 2000 . Reads were aligned to the reference genomes (NCBI build 37, hg19) using Bowtie (v 0.12.7). Repetitively mapped and duplicate reads were excluded. The data can be accessed using accession GSE49637.
ChIP-seq data analysis
We used the MACS version 1.4.1 (model-based analysis of ChIP-Seq) peak-finding algorithm to identify regions of ChIP-Seq enrichment over background . A false discovery rate threshold of enrichment of 0.01 was used for all data sets. The resulting genomic coordinates in bed format were further used in ChromHMM v1.06 for chromatin annotation. The following parameters were used: -Xmx1600M -jar ChromHMM.jar BinarizeBed hg19 -Xmx2000M -jar ChromHMM.jar LearnModel 10 hg19.
DNase I hypersensitive site analysis
Human kidney DHS sequencing data (GEO GSM530655) was analyzed with MACS (v.1.4.1). The resulting peaks were overlapped with the differentially methylated regions. The control random genomic loci were generated using Regulatory Sequence Analysis Tools. Based on the data property of differentially methylated regions, we used the same number of fragments (4,751) and the same average fragment size (443 bp) as parameters for the random loci.
Illumina infinium 450K BeadChip arrays
Genomic DNA (200 ng) was purified using the DNeasy Blood and Tissue Kit (Qiagen) following the manufacturer’s protocol. Purified DNA quality and concentration were assessed with a NanoDrop ND-1000 (Thermo Scientific, Waltham, MA, USA) and by Quant-iT™ PicoGreen® dsDNA Assay Kit (Life Technologies) prior to bisulfite conversion. Purified genomic DNA was bisulfite converted using the EZ DNA Methylation Kit (Zymo Research, Orange, CA, USA) following the manufacturer’s protocol. Bisulfite DNA quality and concentration were assessed, following the Illumina 450K array protocol, bisulfite converted sample was whole-genome amplified, enzymatically digested, and hybridized to the array, and then single nucleotide extension was performed.
Chips were scanned using an Illumina HiScan on a two-color channel to detect Cy3-labeled probes on the green channel and Cy5-labeled probes on the red channel. Illumina GenomeStudio Software 2011.1 Methylation Module 1.8 was used to read the array output and conduct background normalization. The level of DNAm for 428,216 probes in our sample dataset was intersected with the expanded annotation for further analyses. All samples were run together to eliminate the batch effect according to the pipelines established by Illumina Genome Studio. The full dataset can be accessed in GEO under GSE50874.
chronic kidney disease
DNase I hypersensitivity
differentially methylated region
Gene Expression Omnibus
HpaII fragment enrichment by ligation-mediated PCR
transforming growth factor beta
This work was supported by the NIH 5R01DK087635-03 to JMG and KS. MCI was supported by the Ministry of Spain FI08/00476. We would like to thank Drs Anil Gaikwad and Yiting Yu for their help. We would like to thank the staff members of the Montefiore Pathology department and Einstein’s Center for Epigenomics for their support.
- Jensen BL: Reduced nephron number, renal development and 'programming’ of adult hypertension. J Hypertens. 2004, 22: 2065-2066. 10.1097/00004872-200411000-00006.View ArticlePubMedGoogle Scholar
- Woroniecki R, Gaikwad AB, Susztak K: Fetal environment, epigenetics, and pediatric renal disease. Pediatr Nephrol. 2011, 26: 705-711. 10.1007/s00467-010-1714-8.View ArticlePubMedGoogle Scholar
- Pujadas E, Feinberg AP: Regulated noise in the epigenetic landscape of development and disease. Cell. 2012, 148: 1123-1131. 10.1016/j.cell.2012.02.045.View ArticlePubMedPubMed CentralGoogle Scholar
- Laird PW: Principles and challenges of genomewide DNA methylation analysis. Nat Rev Genet. 2010, 11: 191-203.View ArticlePubMedGoogle Scholar
- Weber W: Cancer epigenetics. Prog Mol Biol Transl Sci. 2010, 95: 299-349.View ArticlePubMedGoogle Scholar
- Jones PA, Laird PW: Cancer epigenetics comes of age. Nat Genet. 1999, 21: 163-167. 10.1038/5947.View ArticlePubMedGoogle Scholar
- You JS, Jones PA: Cancer genetics and epigenetics: two sides of the same coin?. Cancer Cell. 2012, 22: 9-20. 10.1016/j.ccr.2012.06.008.View ArticlePubMedPubMed CentralGoogle Scholar
- O’Dwyer K, Maslak P: Azacitidine and the beginnings of therapeutic epigenetic modulation. Expert Opin Pharmacother. 2008, 9: 1981-1986. 10.1517/146565126.96.36.1991.View ArticlePubMedGoogle Scholar
- Ryan RJ, Bernstein BE: Molecular biology, Genetic events that shape the cancer epigenome. Science. 2012, 336: 1513-1514. 10.1126/science.1223730.View ArticlePubMedGoogle Scholar
- Villeneuve LM, Natarajan R: The role of epigenetics in the pathology of diabetic complications. Am J Physiol Renal Physiol. 2010, 299: F14-F25. 10.1152/ajprenal.00200.2010.View ArticlePubMedPubMed CentralGoogle Scholar
- Labrie V, Pai S, Petronis A: Epigenetics of major psychosis: progress, problems and perspectives. Trends Genet. 2012, 28: 427-435. 10.1016/j.tig.2012.04.002.View ArticlePubMedPubMed CentralGoogle Scholar
- Stenvinkel P, Carrero JJ, Axelsson J, Lindholm B, Heimbürger O, Massy Z: Emerging biomarkers for evaluating cardiovascular risk in the chronic kidney disease patient: how do new pieces fit into the uremic puzzle?. Clin J Am Soc Nephrol. 2008, 3: 505-521. 10.2215/CJN.03670807.View ArticlePubMedGoogle Scholar
- Woroniecka KI, Park AS, Mohtat D, Thomas DB, Pullman JM, Susztak K: Transcriptome analysis of human diabetic kidney disease. Diabetes. 2011, 60: 2354-2369. 10.2337/db10-1181. doi: 10.2337/db10-1181. Epub 2011 Jul 13View ArticlePubMedPubMed CentralGoogle Scholar
- Suzuki M, Greally JM: DNA methylation profiling using HpaII tiny fragment enrichment by ligation-mediated PCR (HELP). Methods. 2010, 52: 218-222. 10.1016/j.ymeth.2010.04.013.View ArticlePubMedPubMed CentralGoogle Scholar
- Alvarez H, Opalinska J, Zhou L, Sohal D, Fazzari MJ, Yu Y, Montagna C, Montgomery EA, Canto M, Dunbar KB, Wang J, Roa JC, Mo Y, Bhagat T, Ramesh KH, Cannizzaro L, Mollenhauer J, Thompson RF, Suzuki M, Meltzer SJ, Melnick A, Greally JM, Maitra A, Verma A: Widespread hypomethylation occurs early and synergizes with gene amplification during esophageal carcinogenesis. PLoS Genet. 2011, 7: e1001356-10.1371/journal.pgen.1001356.View ArticlePubMedPubMed CentralGoogle Scholar
- Djavani M, Yenice S, Kirkali G, Guner G, Sessiz HT: Alterations of collagen content in kidney of diabetic rabbits. Biochem Soc Trans. 1993, 21: 275S-View ArticlePubMedGoogle Scholar
- Ernst J, Kellis M: ChromHMM: automating chromatin-state discovery and characterization. Nat Methods. 2012, 9: 215-216. 10.1038/nmeth.1906.View ArticlePubMedPubMed CentralGoogle Scholar
- Maher B: ENCODE: the human encyclopaedia. Nature. 2012, 489: 46-48. 10.1038/489046a.View ArticlePubMedGoogle Scholar
- Aiden AP, Rivera MN, Rheinbay E, Ku M, Coffman EJ, Truong TT, Vargas SO, Lander ES, Haber DA, Bernstein BE: Wilms tumor chromatin profiles highlight stem cell properties and a renal developmental network. Cell Stem Cell. 2010, 6: 591-602. 10.1016/j.stem.2010.03.016.View ArticlePubMedPubMed CentralGoogle Scholar
- Thurman RE, Rynes E, Humbert R, Vierstra J, Maurano MT, Haugen E, Sheffield NC, Stergachis AB, Wang H, Vernot B, Garg K, John S, Sandstrom R, Bates D, Boatman L, Canfield TK, Diegel M, Dunn D, Ebersol AK, Frum T, Giste E, Johnson AK, Johnson EM, Kutyavin T, Lajoie B, Lee BK, Lee K, London D, Lotakis D, Neph S, et al: The accessible chromatin landscape of the human genome. Nature. 2012, 489: 75-82. 10.1038/nature11232.View ArticlePubMedPubMed CentralGoogle Scholar
- Bailey TL: Discovering novel sequence motifs with MEME. Curr Protoc Bioinformatics. 2002, Chapter 2: Unit 2.4-PubMedGoogle Scholar
- Tanaka E, Bailey T, Grant CE, Noble WS, Keich U: Improved similarity scores for comparing motifs. Bioinformatics. 2011, 27: 1603-1609. 10.1093/bioinformatics/btr257.View ArticlePubMedPubMed CentralGoogle Scholar
- Inazaki K, Kanamaru Y, Kojima Y, Sueyoshi N, Okumura K, Kaneko K, Yamashiro Y, Ogawa H, Nakao A: Smad3 deficiency attenuates renal fibrosis, inflammation, and apoptosis after unilateral ureteral obstruction. Kidney Int. 2004, 66: 597-604. 10.1111/j.1523-1755.2004.00779.x.View ArticlePubMedGoogle Scholar
- Schiffer M, von Gersdorff G, Bitzer M, Susztak K, Bottinger EP: Smad proteins and transforming growth factor-beta signaling. Kidney Int Suppl. 2000, 77: S45-S52.View ArticlePubMedGoogle Scholar
- Yu J, Valerius MT, Duah M, Staser K, Hansard JK, Guo JJ, McMahon J, Vaughan J, Faria D, Georgas K, Rumballe B, Ren Q, Krautzberger AM, Junker JP, Thiagarajan RD, Machanick P, Gray PA, van Oudenaarden A, Rowitch DH, Stiles CD, Ma Q, Grimmond SM, Bailey TL, Little MH, McMahon AP: Identification of molecular compartments and genetic circuitry in the developing mammalian kidney. Development. 2012, 139: 1863-1873. 10.1242/dev.074005.View ArticlePubMedPubMed CentralGoogle Scholar
- Godfrey KM, Forrester T, Barker DJ, Jackson AA, Landman JP, Hall JS, Cox V, Osmond C: Maternal nutritional status in pregnancy and blood pressure in childhood. Br J Obstet Gynaecol. 1994, 101: 398-403. 10.1111/j.1471-0528.1994.tb11911.x.View ArticlePubMedGoogle Scholar
- Roseboom TJ, van der Meulen JH, Ravelli AC, van Montfrans GA, Osmond C, Barker DJ, Bleker OP: Blood pressure in adults after prenatal exposure to famine. J Hypertens. 1999, 17: 325-330. 10.1097/00004872-199917030-00004.View ArticlePubMedGoogle Scholar
- Bielesz B, Sirin Y, Si H, Niranjan T, Gruenwald A, Ahn S, Kato H, Pullman J, Gessler M, Haase VH, Susztak K: Epithelial Notch signaling regulates interstitial fibrosis development in the kidneys of mice and humans. J Clin Invest. 2010, 120: 4040-4054. 10.1172/JCI43025.View ArticlePubMedPubMed CentralGoogle Scholar
- Khulan B, Thompson RF, Ye K, Fazzari MJ, Suzuki M, Stasiek E, Figueroa ME, Glass JL, Chen Q, Montagna C, Hatchwell E, Selzer RR, Richmond TA, Green RD, Melnick A, Greally JM: Comparative isoschizomer profiling of cytosine methylation: the HELP assay. Genome Res. 2006, 16: 1046-1055. 10.1101/gr.5273806.View ArticlePubMedPubMed CentralGoogle Scholar
- Thompson RF, Suzuki M, Lau KW, Greally JM: A pipeline for the quantitative analysis of CG dinucleotide methylation using mass spectrometry. Bioinformatics. 2009, 25: 2164-2170. 10.1093/bioinformatics/btp382.View ArticlePubMedPubMed CentralGoogle Scholar
- Zhou L, Opalinska J, Sohal D, Yu Y, Mo Y, Bhagat T, Abdel-Wahab O, Fazzari M, Figueroa M, Alencar C, Zhang J, Kambhampati S, Parmar S, Nischal S, Hueck C, Suzuki M, Freidman E, Pellagatti A, Boultwood J, Steidl U, Sauthararajah Y, Yajnik V, McMahon C, Gore SD, Platanias LC, Levine R, Melnick A, Wickrema A, Greally JM, Verma A: Aberrant epigenetic and genetic marks are seen in myelodysplastic leukocytes and reveal Dock4 as a candidate pathogenic gene on chromosome 7q. J Biol Chem. 2011, 286: 25211-25223. 10.1074/jbc.M111.235028.View ArticlePubMedPubMed CentralGoogle Scholar
- Feng J, Liu T, Qin B, Zhang Y, Liu XS: Identifying ChIP-seq enrichment using MACS. Nat Protoc. 2012, 7: 1728-1740. 10.1038/nprot.2012.101.View ArticlePubMedGoogle Scholar
- Thomas-Chollier M, Defrance M, Medina-Rivera A, Sand O, Herrmann C, Thieffry D, van Helden J: RSAT 2011: regulatory sequence analysis tools. Nucleic Acids Res. 2011, 39: W86-W91. 10.1093/nar/gkr377.View ArticlePubMedPubMed CentralGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.