close
Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Sep 15:6:33222.
doi: 10.1038/srep33222.

DNA damage and Repair Modify DNA methylation and Chromatin Domain of the Targeted Locus: Mechanism of allele methylation polymorphism

Giusi Russo  1 , Rosaria Landi  1 , Antonio Pezone  1 , Annalisa Morano  1 , Candida Zuchegna  2 , Antonella Romano  2 , Mark T Muller  3 , Max E Gottesman  4 , Antonio Porcellini  2 , Enrico V Avvedimento  1
Affiliations

DNA damage and Repair Modify DNA methylation and Chromatin Domain of the Targeted Locus: Mechanism of allele methylation polymorphism

Giusi Russo et al. Sci Rep. .

Abstract

We characterize the changes in chromatin structure, DNA methylation and transcription during and after homologous DNA repair (HR). We find that HR modifies the DNA methylation pattern of the repaired segment. HR also alters local histone H3 methylation as well chromatin structure by inducing DNA-chromatin loops connecting the 5' and 3' ends of the repaired gene. During a two-week period after repair, transcription-associated demethylation promoted by Base Excision Repair enzymes further modifies methylation of the repaired DNA. Subsequently, the repaired genes display stable but diverse methylation profiles. These profiles govern the levels of expression in each clone. Our data argue that DNA methylation and chromatin remodelling induced by HR may be a source of permanent variation of gene expression in somatic cells.

PubMed Disclaimer

Figures

Figure 1
Figure 1. Spatial and temporal changes of histone H3 PTM after DSB at the GFP locus.
(A) Schematic diagram of DRGFP plasmid and a reference locus (TSHR exon 9). Structure of the DRGFP plasmid integrated as single copy in different locations in Hela cells. Primers a and c cannot be used in cells transiently transfected with I-SceI plasmid because they are also present in intron 1 of the I-SceI expression vector. (B) H3K4m2/3 and H3K9m2/3 content of GFP and the reference gene. DRGFP cells were transfected with I-SceI and characterized 24 h, 48 h and 7 days later. Cells were fixed and the chromatin analyzed by ChIP with the indicated antibodies. qPCR on each immunoprecipitate was carried out with the primers indicated in A. The specific antibodies are indicated at the top of each column. Each panel is identified by a numbered box in the upper left side.*P < 0.01 (t test) as compared with untreated control or basal. (C) H3K4m2/3 and H3K9m2/3 content in cells sorted 7 days after transfection. CTRL are cells transfected with a control plasmid; UnRec were GFP− cells sorted and separated from GFP+ Rec cells after I-SceI transfection. *P < 0.01 (t test) as compared with UnRec or CTRL. The detailed statistical analysis of the data shown in panels (B,C) is reported in Supplemental Statistical Tables 1 and 2.
Figure 2
Figure 2. Spatial and temporal changes of histone H3 K4-K9 methylation after homologous repair at the GFP locus in HR cells.
(A) DRGFP Hela cells were transfected with the I-SceI vector and sorted for GFP expression 14 days later as described in Methods. The panels from the left to the right show the gating strategy used to sort Rec H and Rec L cells. The percent of viable cells was 90.6. The fraction of the total GFP+ cells (middle panel) or of Rec H and Rec L (left panel) is indicated. (B) H3K4m2/3 and H3K9m2/3 levels at GFP chromatin in purified Rec H and Rec L cells, 14 days after I-SceI transfection. The specific primers are indicated at the top of each column as shown in Fig. 1A. The data are normalized to total H3 content. *P < 0.01 (t test) as compared with cells transfected with control plasmid (CTRL). (C) H3K9m2/3 content of GFP in chromatin derived from sorted Rec H or Rec L clones, untreated or treated with 5-azadC (10 μM for 3 days and analyzed 4 days later). ChIP analysis was performed with the indicated antibodies. qPCR on each immunoprecipitate was carried out with primers r/h, as shown in Fig. 1A. *p < 0.01 (Wilcoxon rank-sum test) as compared with untreated control or (primers a/c). (D) Histone H3 and DNA methylation changes following the DSB and HR, A schematic cartoon illustrating the major chromatin changes of H3K4me2/3 or H3K9me2/3 and DNA methylation (Δ) following the DSB and the HR or NHEJ in recombinant (HR) and non recombinant (NHEJ) cells. The detailed statistical analysis of the data shown in panels (B,C) is reported in Supplemental Statistical Tables 3 and 4.
Figure 3
Figure 3. Chromatin-DNA domains induced by damage-repair.
(A) Left. GFP chromatin loops in recombinant (Rec, black) and unrecombinant (UnRec, white) cells 7 days after DSB. qPCR in 3C, performed with the primers indicated (HaeIII map), show the frequency of ligation (mean ± SD) of the DRGFP HaeIII fragments amplified with the specific primers. The ends of the loops are shown by the green (loop A) and blue (loop C) lines, corresponding to the regions indicated by arrows and boxes. Ligation efficiency is relative to the DRGFP plasmid digested with HaeIII, ligated and amplified by qPCR. Distance, in bp, is relative to I-SceI site (vertical arrow), *p < 0.01 (t test) Rec vs UnRec. Right. A representative gel of the ligated fragments (loops A and C) in cells exposed or not for 48 h to I-SceI. (B) GFP loops in Rec H and Rec L cells. The 5′ end of the loop A (green) includes an alternative GFP transcription start site, identified by primer d. Differences between non-recombinant, H or L cells *p < 0.01 (t test); °p < 0.01 (t test) H vs L cells. (C) Inhibition of methylation (panels 1, 3) or transcription (panels 2, 4) alters the chromatin loops induced by HR. Panels 1 and 3. Sorted GFP− (UnRec) or Rec H or Rec L cells were exposed to 5 or 50 μM 5-azadC for 24 h + 24 h in standard medium. Loops A and C were monitored as described above. The results derive from at least 3 experiments in triplicate. *p < 0.01 as compared to untreated samples (Wilcoxon rank-sum test). Panels 2 and 4. Sorted GFP− (UnRec) or Rec H or Rec L cells were exposed to 5 μM actinomycin D for 24 h, washed and analysed 48 h later. *p < 0.01 (Wilcoxon rank-sum test) compared to untreated cells. (D) Model summarizing the features of Loops A and C. The red and green arrows indicate decrease or increase of loop formation, respectively, while the black arrows indicate no change. The two GFP transcription start sites are indicated by black and blue arrows.
Figure 4
Figure 4. Inhibition of BER early after repair reduces transcription and increases methylation of the repaired DNA.
(A) Chronology of BER-silencing experiments. Time 0 indicates I-SceI transfection and arrows the time of analysis or treatments. GFP+ cells were 10% ± 2% in all treatments. At days 12 and 32, 5-azadC (10 μg/ml, blue arrows) was added for 24 h, removed and 24 h later cells were analyzed. (B,C) Analysis was performed at day 7 (red) or 14 (black) after treatments. GFP index is the product of GFP intensity and reciprocal cell fraction in the fluorescence gate to normalize frequency of GFP+ cells in H and L gates to intensity of signal, to compare different experiments. The data shown derive from 20 independent experiments. (B,C) Show GFP index in Rec H and Rec L cells, respectively. *p < 0.01 (Matched t test) compared to scrambled control. (Supplemental Fig. S3 and Table S2). (D) GFP mRNA levels 7 days after APE1 and TDG-silencing by qPCR for recombinant GFP. TDG protein levels are in Supplemental Fig. S3C. *P < 0.01 (Wilcoxon rank-sum test) compared to scrambled control. (E,F) Methylation analysis of GFP in mass cultures of I-SceI-transfected cells, in which APE1 levels were modified after HR (48 h after I-SceI and analyzed 7 days later, panel A). DNA was subjected to bisulfite analysis and sequenced with Myseq Illumina (Supplemental Table S3). Panel E shows the average methylation of the recombinant GFP. All the cells were exposed to I-SceI and these vectors: SCR shRNA (white shaded) scrambled shRNA; APE1 shRNA (green) shRNA APE1; APE1 shRNA + WT (purple) shRNA APE1 and APE1 expression vector. The percent of methylation of recombinant GFP in all samples is normalized to the recombinant GFP cassette as shown in Supplemental Table S3; data were expressed as the mean ± SEM. *p < 0.01 (t test) compared APE1 shRNA vs SCR shRNA or vs APE1 shRNA + WT. Panel F shows the percent of CpG methylation in recombinant GFP in cells in which the levels of APE1 were modified early after repair. The position of the I-SceI/BcgI site corresponds to CpG 7 and 8. *p < 0.01 (Pearson’s chi-squared test) compared APE1 shRNA vs SCR shRNA (+SceI) or vs APE1 shRNA + WT.
Figure 5
Figure 5. Methylated GFP molecules are polymorphic.
(A) Qualitative analysis of GFP methylation profiles in UnRec, Rec H and Rec L clones. We compared the location of methylated CpGs in the most abundant GFP molecules (above 5–10%) derived from recombinant (Rec H and Rec L) and non-recombinant (UnRec,) cells. The sequence at the I-SceI site in UnRec molecules was edited to BcgI to eliminate the differences in the sequence between Rec and UnRec molecules and to permit the comparison of Rec and UnRec GFP molecules only on the basis of methylation. Cluster analysis (ClustalW) shows three main families of methylated molecules: I, represented essentially by Rec L clones; II and III represented by Rec H and Rec L clones. Colored circles indicate the number of methylated CpGs/molecule. (B) Specific CpGs are methylated in Rec H and Rec L clones following HR. Molecules containing 3 to 6 mCpGs were sorted from Rec H or Rec L cells pools and compared. The location of methylated CpGs at the 5′ and 3′ ends of the DSB is shown relative to the DSB (black-yellow box centered at the 6–8 CpG). Methylation of CpG from 1 to 5 is not modified by DSB or HR (ref. 11). The histograms show the percentage of methylation of the specific CpGs in Rec H (green) and Rec L (red) clones.
Figure 6
Figure 6. Homologous targeting of GFP to the mouse Rosa26 locus generates ES clones with variable DNA methylation and GFP expression.
Three independent ES clones in which the EGFP gene was targeted to the mouse Rosa26 locus, (kindly provided by A. Simeone and D. Acampora, Intern. Inst. Gen. Biophys., IGB, Naples, Italy). These clones, containing a single copy integrated GFP, were purified 7 days after transfection with the targeting vector, amplified and analyzed by cytofluorimetry as described in Methods. The same clones were exposed to 5-azadC (0.5 μM) for 4 days and analyzed 48 h later. Panel (A) shows: 1. the structure of the targeting vector; the two homologous regions are shown in red; 2. the structure of the mouse Rosa26 locus and; 3. the structure of the targeted locus. A segment at the 5′ end, containing a CpG island, is shown (lines). The primers for MEDIP analysis are located in regions I and II. Panel (B) shows the cytofluorimetric analysis of the 3 clones exposed or not to 0.5 μM 5-azadC for 4 days and analyzed 48 h later. Dot Plot scans are shown to illustrate the composition of GFP+ or GFP− cells. The arrows indicate the shift of the L population after 5-azadC treatment. Differences in GFP expression between control and 5-azadC treated cells were tested for statistical significance using the Chi Square test, T(X), (Population Comparison module of the FlowJo software from Tree Star). Cl 44, untreated vs 5-azadC T(X) = 460, p > 0.001; Cl 55, untreated vs 5-azadC T(X) = 246, p > 0.001; Cl 59, untreated vs 5-azadC T(X) = 61, p < 0.001. (C) The panel on the left shows the overlapping profiles of the three clones without treatment to compare the relative GFP expression levels. Cl 44 vs Cl55 T(X) = 143, p < 0.001; Cl 44 vs Cl59 T(X) = 658, p > 0.001; Cl 55 vs Cl59 T(X) = 255, p > 0.001. The central and the left panels show the quantitative analysis of GFP expression in Rec H and Rec L cells before or after 5-azadC treatment as % of GFP+ cells and mean of fluorescence intensity. In the left panel, Rec H and Rec L clones were analyzed together. Differences between treatments were tested for statistical significance using matched pairs t test: *p < 0.001. Panel (D) shows MEDIP analysis of region I and II, respectively in the 3 clones. *p < 0.001.
See this image and copyright information in PMC

References

    1. Horvath S. DNA methylation age of human tissues and cell types. Genome Biology 14, R115 (2013). - PMC - PubMed
    1. Huh I., Zeng J., Park T. & Yi S. V. DNA methylation and transcriptional noise. Epigenetics Chromatin 6, 9 (2013). - PMC - PubMed
    1. Li E., Beard C. & Jaenisch R. Role for DNA methylation in genomic imprinting. Nature 366, 362–365 (1993). - PubMed
    1. Fu A. Q., Genereux D. P., Stöger R., Laird C. D. & Stephens M. Statistical inference of transmission fidelity of DNA methylation patterns over somatic cell divisions in mammals. Ann Appl Stat 4, 871–892 (2010). - PMC - PubMed
    1. Baylin S. B. & Ohm J. E. Epigenetic gene silencing in cancer - a mechanism for early oncogenic pathway addiction? Nat Rev Cancer 6, 107–116 (2006). - PubMed

Publication types

LinkOut - more resources