DNA damage and repair permanently alters chromatin and DNA: Mechanisms of DNA methylation polymorphism
Antonio Pezone, PhD, Dept Molecular Medicine
Antonio Porcellini, MD, PhD, Dept Biology
Enrico Vittorio Avvedimento, MD, Dept Molecular Medicine
University “Federico II” Naples, Italy
DNA methylation is a biological clock that ticks with age, and numerous studies have been able to construct age-prediction models based on methylation status. However in somatic cells, DNA methylation is highly polymorphic – though this is not always appreciated in genome wide studies because coverage is often limited and variants below 10% usually escape detection. In an effort to clarify the origin of somatic DNA methylation heterogeneity and the impact it has on chromatin domains and gene expression, our team has asked the following questions: 1. Does local chromatin modify DNA methylation or does DNA methylation modify chromatin? 2. What is the origin of methylation polymorphism in somatic cells? 3. Does the extent and pattern of methylation impart gene expression variation in cell populations with an identical genotype?
To answer these questions we used a reductionist approach to assess the deterministic and stochastic factors that contribute to the final methylation status of GFP epialleles. We have previously shown that DNA damage and homologous repair (HR) modify the methylation profiles of the repaired gene with a recognizable signature spanning the exchanged DNA segment.1,2 In our current study, we synchronized a DNA damage event (1 double strand break/genome) and analyzed the histone modification and and DNA methylation of the repaired region overtime. Cells harboring the repaired gene were sorted and separately analyzed. A time course analysis of chromatin changes after the double strand break (DSB) reveals that cells transiently (within 24hrs) lose activating H3K4 methylation and gain repressive H3K9 methylation at the DSB area in order to temporarily block local transcription and permit repair. Eventually (approximately 7 days after DSB) only a fraction of cells in which the DNA has been repaired by homologous recombination (HR), retain the repressive H3K9 methylation (dimethyl or trimethyl lysine 9). Treatment with demethylating drugs are able to eliminate H3K9me2-3 from repaired chromatin in L cells, suggesting that stabilization and inheritance of this repressive marker requires de novo DNA methylation. Furthermore, a chromatin-DNA loop connecting the 5’ with the 3’ end of the repaired gene was induced by HR and stabilized by the methylation signature (see the figure) demonstrating that these epigenetic changes permanently modify chromatin structure.
During these studies we found that the profiles of methylated alleles of the repaired gene (epialleles) were not homogenous, but varied from cell to cell during a critical period (21 days) after HR. This editing of methylation was due to local transcription after repair, associated with active cytosine demethylation. 2-3 weeks after HR, DNA methylation profiles stabilized. After years of continuous culture, methylation profiles of HR alleles remained polymorphic but stable.3
Now we are left with the remaining question of whether somatic DNA methylation is deterministic or stochastic? Some data suggest that it is deterministic; for example, methylation of the INK4-ARF suppressor gene can be induced by a specific Ki-Ras oncogene transcriptional program.4 Locus-specific targeting of DNMTs is sufficient to induce and maintain DNA methylation. The target is determined by specific affinities of transcription factors and chromatin modelers. Eventually, the preference of DNMT1 for hemi-methylated DNA stabilizes the methylation profiles. This deterministic model may account for clustering of methylated sites in the same DNA region. However it still fails to explain the extreme polymorphism of methylated alleles found upon deeper sequence coverage of the genome.3,5
In our system we can quantify the deterministic and the stochastic factors that contribute to the final methylation status of GFP epialleles. Our data suggests that both deterministic and stochastic factors govern stable DNA methylation profiles. Repair factors that are recruited to a DSB determine the location and specific strand that will be methylated.2,3 However, HR and stochastic editing of methylation by transcription and BER enzymes account for methylated polymorphisms in those regions. We wish to note that also in embryonic stem cells, homologous targeting of GFP generates clones with various levels of GFP expression and DNA methylation.3
In conclusion, we propose that multiple mechanisms contribute to the final methylation status of DNA in each cell. Our findings are especially important in the field of genome editing, which is largely impacted by HR events. The final penetrance of a repaired gene will be directed by the events related to DNA methylation revision described here.
Figure: DNA damage and homologous repair permanently modify DNA and chromatin. A single double strand break (DSB indicated by lighting bolt) induced in the genome of Hela cells was repaired by homologous recombination (HR) or non-homologous end joining (NHEJ). The DNA and the chromatin of the repaired locus are shown at the left or right side of the figure, respectively. LEFT. DNA-chromatin loops, detected by chromatin-conformation–capture technique (3C), were analyzed in cells not subjected to damage, exposed to the DSB and repaired by HR. The chromatin of the HR repaired gene is found in two configurations:1. loop A in high expresser cells, H, (gene hypo-methylated) and loop C in low-expresser cells, L, (gene hyper-methylated). The red loop indicates the chromatin structure of the reference gene contiguous to GFP repaired by HR. This gene has not been subjected to DSB or repair of any sort. RIGHT. Chromatin of the targeted locus: histone H3 methylation code at lysine 4 (K4) or lysine 9 (K9); I-SceI and Bcg I indicate non-recombinant (I-SceI) or recombinant (BcgI) DNA. On the left side is shown the methylation code of the GFP locus 24, 48h and 7days after the DSB (I-SceI). The triangle indicates methylated CpG. Note the different DNA and chromatin status 7days after damage-HR between HR and NHEJ clones. These changes are stable and permanent. HR cells after 3 years of continuous culture display the same chromatin and DNA configurations.
Russo, G., Landi, R., Pezone, A., Morano, A., Zuchegna, C., Romano, A., Muller, M., Gottesman, M., Porcellini, A., & Avvedimento, E. (2016). DNA damage and Repair Modify DNA methylation and Chromatin Domain of the Targeted Locus: Mechanism of allele methylation polymorphism Scientific Reports, 6 DOI: 10.1038/srep33222
1. Cuozzo, C., Porcellini, A., Angrisano, T., Morano, A., Lee, B., Pardo, A., Messina, S., Iuliano, R., Fusco, A., Santillo, M., Muller, M., Chiariotti, L., Gottesman, M., & Avvedimento, E. (2007). DNA Damage, Homology-Directed Repair, and DNA Methylation PLoS Genetics, 3 (7) DOI: 10.1371/journal.pgen.0030110
2. Morano, A., Angrisano, T., Russo, G., Landi, R., Pezone, A., Bartollino, S., Zuchegna, C., Babbio, F., Bonapace, I., Allen, B., Muller, M., Chiariotti, L., Gottesman, M., Porcellini, A., & Avvedimento, E. (2013). Targeted DNA methylation by homology-directed repair in mammalian cells. Transcription reshapes methylation on the repaired gene Nucleic Acids Research, 42 (2), 804-821 DOI: 10.1093/nar/gkt920
3. Russo, G., Landi, R., Pezone, A., Morano, A., Zuchegna, C., Romano, A., Muller, M., Gottesman, M., Porcellini, A., & Avvedimento, E. (2016). DNA damage and Repair Modify DNA methylation and Chromatin Domain of the Targeted Locus: Mechanism of allele methylation polymorphism Scientific Reports, 6 DOI: 10.1038/srep33222
5. Landan, G., Cohen, N., Mukamel, Z., Bar, A., Molchadsky, A., Brosh, R., Horn-Saban, S., Zalcenstein, D., Goldfinger, N., Zundelevich, A., Gal-Yam, E., Rotter, V., & Tanay, A. (2012). Epigenetic polymorphism and the stochastic formation of differentially methylated regions in normal and cancerous tissues Nature Genetics, 44 (11), 1207-1214 DOI: 10.1038/ng.2442