Diabetes Publish Ahead of Print, published online September 5, 2008 Lymphocytes from Patients with Type 1 diabetes Display a Distinct Profile of Chromatin Histone H3 Lysine 9 Dimethylation: An Epigenetic Study in Diabetes * † * * * * Feng Miao , David D. Smith , Lingxiao Zhang , Andrew Min , Wei Feng , Rama Natarajan * †, Departments of Diabetes and Biomedical Informatics Beckman Research Institute of City of Hope, 1500 East Duarte Road, Duarte, CA 91010 Address for Correspondence: Rama Natarajan, PhD, FAHA, FASN Professor, Department of Diabetes Beckman Research Institute of City of Hope 1500 East Duarte Road Duarte, CA 91010 Email: [email protected] Submitted 14 May 2008 and accepted 21 August 2008. Additional information for this article can be found in an online appendix at http://diabetes.diabetesjournals.org This is an uncopyedited electronic version of an article accepted for publication in Diabetes. The American Diabetes Association, publisher of Diabetes, is not responsible for any errors or omissions in this version of the manuscript or any version derived from it by third parties. The definitive publisher-authenticated version will be available in a future issue of Diabetes in print and online at http://diabetes.diabetesjournals.org. 1 Copyright American Diabetes Association, Inc., 2008 ABSTRACT OBJECTIVE: The complexity of interactions between genes and the environment is a major challenge for Type I diabetes (T1D) studies. Nuclear chromatin is the interface between genetics and environment, and the principal carrier of epigenetic information. Since histone tail modifications in chromatin are linked to gene transcription, we hypothesized that histone methylation patterns in cells from T1D patients can provide novel epigenetic insights into T1D and its complications. RESEARCH DESIGN AND METHODS: We used Chromatin immunoprecipitation linked to microarray (ChIP-chip) approach to compare genome-wide histone H3 lysine 9 dimethylation (H3K9me2) patterns in blood lymphocytes and monocytes from T1D patients versus healthy control subjects. Bioinformatics evaluation of methylated candidates was performed by Ingenuity Pathway Analysis (IPA) tools. RESULTS: A subset of genes in the T1D cohort showed significant increase in H3K9me2 in lymphocytes, but not monocytes. CLTA4, a T1D susceptibility gene, was one of the candidates displaying increased promoter H3K9me2 in T1D. IPA identified two high scoring networks that encompassed genes showing altered H3K9me2. Many of them were associated with autoimmune and inflammation-related pathways such as TGF-β, NF-κB, p38MARK, TLR and IL6. IPA also revealed biological relationships between these networks and known T1D candidate genes. CONCLUSIONS: The concerted and synergistic alteration of histone methylation within the identified network in lymphocytes might have an effect on the etiology of T1D and its complications. These studies provide evidence of a novel association between T1D and altered histone methylation of key genes that are components of T1D related biological pathways and also a new understanding of the pathology of T1D. 2 ype I diabetes (T1D), is an autoimmune a key chromatin mark, histone H3 disease resulting from complex dimethylated at lysine 9 (H3K9me2). This T interactions between genetic and mark, broadly spread within the human environmental factors. It is characterized by T- genome, is generally associated with gene lymphocyte mediated destruction of the repression, and could either be a repressive insulin-producing β-cells of pancreatic islets. mark in euchromatin or a hallmark feature of More than 30 genome loci have been linked to heterochromatin (11-13). T1D susceptibility through genome linkage Chromatin immunoprecipitation coupled to analysis (1). Although numerous genetic DNA microarray analysis, or ChIP-chip, is a studies, including the recent genome-wide widely-used approach for acquiring genome- association study (2), have provided a wealth wide information on histone modifications of knowledge about genetic factors associated (14-18). We recently implemented this with T1D, the underlying mechanisms and approach to profile and compare the variations pathways causing T1D remain only partly in histone H3K4me2 and H3K9me2 in human understood (3). Genetic approaches in general gene coding and CpG island regions in THP-1 cannot fully take into consideration the monocytes cultured in normal and high interplay between genes and the environment glucose (19). We observed that the treatment in T1D. As such, a study of the role of of monocytes with high glucose to mimic epigenetics in T1D can provide valuable new diabetic conditions could lead to key variations insights. in H3K9me2 in the promoter and coding Nuclear chromatin is a crucial interface regions of several genes including those between the effects of genetics and relevant to the pathogenesis of diabetes (19). environment, and the principal carrier of In the present study, we have used three types epigenetic information. The chromatin is of microarrays, (CpG, cDNA and promoter constantly affected by environmental stimuli, tiling arrays) to compare, for the first time, the such as diet, chemicals and pathogens (4). The H3K9me2 profiles of blood lymphocytes and basic repeat unit of chromatin is the monocytes obtained directly from T1D nucleosome, which consists of two copies each patients versus healthy control subjects. of histones H2A, H2B, H3 and H4, and is Analyses of the data revealed a group of genes wrapped by 147 base pairs of DNA (5). showing a striking alteration in histone Notably, the post-translational modifications H3K9me2 in the lymphocytes of the T1D (PTMs) of histone tails in chromatin have been patients relative to healthy controls and their linked to gene transcription (6; 7). The strong associations to T1D. In addition, by “histone code” hypothesis has changed our applying the bioinformatics software Ingenuity view of histones as being mainly a DNA Pathway Analysis (IPA), we uncovered key scaffold to a key regulatory layer of gene biological relationships among a subset of transcription. By interacting with various genes in the differentially methylated group chromatin factors and regulatory proteins, and their relevance to T1D. histone PTMs can alter the architecture of chromatin and gene expression. Abnormal RESEARCH DESIGN AND METHODS alterations in these interactions that cause Human Subject Enrollment. An informed relatively stable epigenetic changes at the consent form was obtained from all volunteers chromatin level could lead to dysregulated before blood samples were drawn with an gene transcription, metabolic memory (8-10) approved IRB protocol at the City of Hope and disease progression. We sought to obtain General Clinical Research Center. We evidence for this by genome-wide mapping of prospectively enrolled 16 volunteers into two 3 groups, the first with 9 patients having a Analysis of Microarray data. Microarray diagnosis of Type 1 diabetes for more than ten images were scanned with a GenePix 4000B years (median 16 years, range11 to 52 years), scanner and quantified with GenePix Pro and the second with 7 healthy volunteers. No version 4.1.1.31 (Molecular Devices, Inc., history of autoimmunity was reported in Sunnyvale, CA). We then imported the raw healthy controls and their autoantibody status data from GenePix into R/ Bioconductor was not tested. Patient demographics are statistical software (21-24). We powered our shown in Fig. 1C. There were no statistically study to detect large differences in up- or significant differences between age or gender down-regulation across all probes. Based on proportion in the two comparison groups. our previous results (20), we found that the Blood lymphocytes and monocytes from these standard deviation for lymphocyte ChIP-chip volunteers were separated using the Ficoll data in this population was approximately 0.5 method as described (20) (details provided in on the log2 scale in the healthy the Supplemental Information online). The volunteers. Assuming a two-sided test at 0.05 purity was about 85% for both monocyte and and 80% power, approximately 8 subjects in lymphocyte fractions. Both cell types were each group allowed us to detect log2 effect used for ChIP experiments. sizes of 0.75 or greater (fold change of 1.68 or DNA Microarrays. Human 12K cDNA arrays greater). We excluded one normal subject after were from the University of Pennsylvania ChIP-chip due to technical problems with the Functional Genomics Core, and human 12K hybridization seen in Fig 1B. We also CpG island arrays (containing 12192 CpG excluded one T1D patient ChIP-chip data due island clones) from the Universal Health to a history of prostate cancer. Full details of Network Microarray Center (Toronto) as the statistical analysis of microarray data are described previously (19, 20). Human provided in the Supplemental Information promoter tiling arrays containing 24659 well- online. characterized RefSeq genes were from Roche Real-Time Quantitative PCRs. Real-Time Nimblegen. Quantitative PCRs were performed as Chromatin Immunoprecipitation (ChIP) described (19). Student t-tests were used for and ChIP-chip experiments. Conventional statistical analyses of the data. Sequences of ChIPs and the Chip-chips were performed as primers are provided in the Supplemental described previously (15). Purified blood Information online. monocytes and lymphocytes from T1D and Ingenuity Pathway Analysis (IPA). Core healthy controls were crosslinked, and bioinformatics analyses were