Transcript PowerPoint-Präsentation

Combined analysis of DNA-methylation
and transcription profiles in different immune cells
identifies hot spots of gene regulation
by DNA methylation
Marc Bonin1, Stephan Flemming2, Stefan Günther 2, Andreas Grützkau3, Thomas Häupl1
1
Department of Rheumatology and Clinical Immunology, Charité University Hospital, Berlin, Germany, 2 Institute of Pharmaceutical Sciences , University of
Freiburg, Germany, 3 German Arthritis Research Center, Berlin, Germany
Background:
Methods:
Methylation of DNA may contribute to the regulation of gene expression. Chip
technology enables to analyse for methylation of CpG sites but requires a preselection of potential hot spots. Such a selection of potential sites is represented on
the HumanMethylation450 array (Illumina). However, it is still necessary to
validate, experimentally, which CpG selection is functionally relevant.
Cells from 4 healthy donors were sorted by FACS technology for naive and memory
T-cells (CD4m, CD4n, CD8m, Cd8n), B-cells (CD19m, CD19n), NK-cells (CD56),
monocytes (CD14), and granulocytes (CD15). Genome-wide DNA methylation was
assessed using the Illumina HumanMethylation450 BeadChip platform. Analysis of
data was performed using Genome-Studio (Illumina). Gene expression data were
collected from Affymetrix HG-U133Plus 2.0 transcriptomes analysed in the BioRetis
database. Mapping of CpG sites with genes was performed using the ensemble
genome assembly GRCh37 genomic location map.
Objective:
In order to test these CpG sites for possible functional effects, we assumed that cell
type specific gene expression in different immune cells like T-cells and monocytes
are influenced and maintained by DNA methylation.
Figure 1
Human Methylation450 BeadChip
Figure 2
Comparison of naive and
memory
CD4+
T-cells
revealed a shift from :
The number of differentially expressed genes or methylated CpG sites were highest
between very different cell types like CD14 monocytes and CD4 T-cells (4624 genes;
19261 sites) and lower between naive and memory cells of the same lymphocyte
subtype (CD4: 638 genes; 9412 sites). There was a tendency towards more
methylation in memory (CD4m: 5433 sites ≈ 2694 genes) compared to naive cells
(CD4n: 3979 sites ≈ 2258 genes) for more than 2-fold change while the overall
change was dominated by a decrease from naive to memory status. Overlap of
differential expression with corresponding changes in methylation was found in
only 629 (279) of 1951 increased (2673 decreased) expressed genes for CD14
versus CD4 comparison and 57 (53) of 332 (306) genes for CD4m versus CD4n cells.
Of all CpG sites annotated to these identified genes, only about 10% were
concordant with expression. These CpG site were within or immediately upstream
of the annotated start of the gene with a maximum distance of ≈1500 nucleotides,
indicating that overlap with the promoter site is most likely. A common sequence
motif around these CpG sites was not immediately detectable but requires more
detailed analysis.
Conclusion:
Microarray based comparative analysis of transcriptional and epigenetic differences
suggests a detailed picture of methylation associated gene regulation and enables
to generate an epigenetic map of relevant CpG site for genes expressed and
regulated in immune cell types. As many of the microarray based suspected CpG
sites of a defined gene did not match with differential gene expression, epigenetic
profiling with microarrays has to be interpreted carefully.
n=3979
Figure 4
% methylation in CD14 cells
n=5344
n=307198
% methylation in CD4 cells
CD4 naive
1) methylated CpG in naive
to unmethylated CpG in
memory T-cell genes (total
n=307198 sites; >2-fold
change n=3979)
2) unmethylated CPG in
naive to methylated CpG in
memory T-cell genes (total
n=178378 sites, >2-fold
change n=5344)
Results:
nucleotides upstream of gene start
nucleotides upstream of gene start
n=178378
CD4 memory
Figure 3
all
629
top 50
% methylation in CD14 cells
distance and methylation for CpGs of individual genes
nucleotides upstream of gene start
Increased DNA methylation in CD4+ naive cells compared to CD14+ monocytes when
focussing on monocyte (CD14+) but not naive T-cell (CD4+) related genes: distribution of
methylation frequency for A) all CpG sites; B) 629 genes with increased expression in CD14+
compared to CD4+ cells; C) 50 top candidates with increased expression in CD14+ cells.
Contacts:
Methylation of CpGs in CD14+ monocytes and CD4+ naive T-cells for the top 10 genes
increased in CD14+ but not CD4+ cells. X-axis: nucleotides upstream of the gene start; Y-axis:
percentage of methylation. The closer the CpG to the start of the gene, the lower the
methylation level in actively transcriped genes. Only a minor fraction of all CpGs measured
for a defined gene are indicative for activation or silencing of the corresponding gene.
Marc Bonin
Department of Rheumatology
and Clinical Immunology
Charité University Hospital
Charitéplatz 1
D-10117 Berlin Germany
Tel: +49(0) 30 450 513 296
Fax: +49(0) 30 450 513 968
E-Mail: [email protected]
Web: www.charite-bioinformatik.de
www.charite-bioinformatik.de