Figure 1.
Differential methylation of the human fibrinogen locus in fibrinogen expressing and non-expressing cells.
The reference genes (FGB, FGA and FGG) and the liver enhancers CNC12 and PFE2 are represented under each graph. Methylation percentages of the individual CpGs analyzed in this study are depicted as bars using the Integrative Genomics viewer IGV 2.1 [52] (NCBI36/hg18). A line is drawn arbitrarily at 15% methylation; values above are depicted as blue bars and values below as red bars. The peaks represent the mean methylation of 3 or 4 (hepatocytes) biological replicates. (A) Methylation profiles of human cell-lines: a non fibrinogen-expressing human cell line (HEK293T) and two fibrinogen-expressing hepatoma cell lines (HepG2 and HuH7). (B) Methylation profiles of human primary cells: non-fibrinogen expressing human umbilical vein endothelial cells (HUVEC) and fibrinogen expressing primary human hepatocytes. Average methylation percentages plus SEM across the whole locus are indicated.
Figure 2.
Methylation of the FGA promoter reduces expression of a reporter gene in fibrinogen expressing cells.
(A) Linear constructs containing the differentially methylated FGA promoter (FGAP) and the firefly (ff) luciferase gene (luc). The differentially methylated promoter constructs were ligated to the luciferase gene and directly transfected without cloning steps. (B) Luciferase assays performed in fibrinogen expressing HuH7 cells and HEK293T control cells. Signals for the ligated fragments are normalized to the signal of the unligated luciferase gene. Bars show mean signals plus SEM, n=3. Statistical testing was performed using an unpaired t test (*P< .05, **P< .01, ***P< .001).
Figure 3.
Changes in fibrinogen expression and fibrinogen DNA methylation during mouse and zebrafish development.
(A) Experimental approach used to compare the changes in expression of the three fibrinogen genes with DNA methylation of the fibrinogen locus. The same mouse tissues were used to perform both experiments, except for E12.5 tissues, where two tissue samples were pooled per experiment (see Materials and Methods). For qPCR and DNA methylation sequencing on zebrafish tissues, the livers and caudal fins (for 3 month old females) or trunks (for 8dpf larvae) were dissected and DNA and RNA were extracted from the same samples. The experiments were performed with three biological replicates. (B) Absolute quantification of mouse and zebrafish fibrinogen RNA in the liver. The number of cDNA copies per µg RNA was determined by interpolation of the qPCR results on a standard curve of known concentrations of plasmids containing the respective cDNAs. The mean number of cDNA copies plus SEM are shown on a logarithmic scale, n=3. Statistical testing was performed using an unpaired t test (*P< .05, **P< .01). (C) Methylation percentages of individual CpGs analyzed are depicted as bars at their genome position using the Integrative Genomics viewer IGV [52] (for mouse: NCBI37/mm9, for zebrafish: Zv9/danRer7). A line is drawn at 35% methylation (mouse) or 40% (zebrafish); values above are depicted as blue bars and values below as red bars. Reference genes are shown below each graph and scales and genomic localization (mouse chromosome 3 reverse strand or zebrafish chromosome 1 forward strand) are shown above. Note that in zebrafish fgg is located ~15MB downstream of fga. The peaks represent the mean methylation of three biological replicates, the average methylation percentage plus SEM across the whole locus (for zebrafish the two loci shown) is given for each tissue.
Figure 4.
Active chromatin mark H3K4me3 on the fibrinogen promoters during mouse and zebrafish development.
(A) Overview of the samples used for Chromatin immunoprecipitation (ChIP). For mouse experiments, the 8 week tissues are the same as those used for qPCR and DNA methylation sequencing, the E17.5 and E12.5 mice are from the same crossing as the embryos used for Figure 3. For zebrafish experiments, ChIP was performed on whole embryos at different stages. All experiments were performed with three biological replicates, except for mouse E12.5 liver, where two pools of five livers were used. (B) Enrichment of H3K4me3 on mouse and zebrafish fibrinogen promoters and control regions. After ChIP on tissues or whole zebrafish embryos the signal relative to the total input was determined by qPCR, using primers situated around the transcription start site or in the first intron of each fibrinogen gene (Table S1). Primers binding to the first intron of the ubiquitously expressed Rpl30 gene serve as a positive control for enrichment. As a negative control for the H3K4me3 enrichment, gene free regions on chromosome 12 (mouse) and 14 (zebrafish) were chosen. The mean enrichment plus SEM is shown. Anti-rabbit IgG was used as a negative control in the ChIP experiments and the signals per total input were <0.0007 for all genes examined.
Figure 5.
A model of changes in epigenetic marks and expression of fibrinogen genes during development.
A single fibrinogen gene is shown here to represent the three vertebrate fibrinogen chain genes and the model represents an average situation in a population of cells. An upstream promoter (P) and an intergenic liver enhancer (E) sequence are also represented. Prior to hepatocyte differentiation, CpGs in the fibrinogen genes are methylated (“me”) both in regulatory sequences and across the gene bodies. Embryonic hepatocytes show similar CpG methylation patterns, but with the expression of hepatocyte transcription factors (hep TFs) regulatory sequences are employed and fibrinogen mRNA expression is measurable. The H3K4me3 histone mark is also detectable near the gene promoters (grey cylinders). In later embryonic hepatocytes, expression of fibrinogen is clearly increased (red arrow), the H3K4me3 mark is more abundant at promoters and CpG methylation is moderately reduced, perhaps as a result of the loss of repressive CpG methylation upon cell division. In adult hepatocytes expression is marginally higher still, but CpG methylation is markedly reduced, in gene bodies and regulatory elements.