Characterization of the genetic determinants of context-specific DNA methylation in primary monocytes

doi:10.1016/j.xgen.2024.100541

. 2024 May 8;4(5):100541.

doi: 10.1016/j.xgen.2024.100541. Epub 2024 Apr 24.

Characterization of the genetic determinants of context-specific DNA methylation in primary monocytes

Affiliations

¹ Department of Paediatrics, University of Oxford, Oxford OX3 9DU, UK; MRC Weatherall Institute of Molecular Medicine, University of Oxford, Oxford OX3 9DS, UK.
² Centre for Human Genetics, University of Oxford, Oxford OX3 7BN, UK.
³ Ludwig Cancer Research Oxford, University of Oxford, Oxford OX3 7DQ, UK.
⁴ MRC Weatherall Institute of Molecular Medicine, University of Oxford, Oxford OX3 9DS, UK; Centre for Human Genetics, University of Oxford, Oxford OX3 7BN, UK.
⁵ Nuffield Department of Orthopaedics, Rheumatology, and Musculoskeletal Sciences, University of Oxford, Oxford OX3 7LD, UK.
⁶ MRC Weatherall Institute of Molecular Medicine, University of Oxford, Oxford OX3 9DS, UK.
⁷ Oxford Centre for Diabetes, Endocrinology, and Metabolism, Rheumatology and Musculoskeletal Sciences, University of Oxford, Oxford OX3 7LE, UK.
⁸ MRC Weatherall Institute of Molecular Medicine, University of Oxford, Oxford OX3 9DS, UK; Department of Oncology, University of Oxford, Oxford OX3 9DS, UK. Electronic address: benjamin.fairfax@oncology.ox.ac.uk.

PMID: 38663408
PMCID: PMC11099345
DOI: 10.1016/j.xgen.2024.100541

Characterization of the genetic determinants of context-specific DNA methylation in primary monocytes

James J Gilchrist et al. Cell Genom. 2024.

. 2024 May 8;4(5):100541.

doi: 10.1016/j.xgen.2024.100541. Epub 2024 Apr 24.

Authors

Affiliations

¹ Department of Paediatrics, University of Oxford, Oxford OX3 9DU, UK; MRC Weatherall Institute of Molecular Medicine, University of Oxford, Oxford OX3 9DS, UK.
² Centre for Human Genetics, University of Oxford, Oxford OX3 7BN, UK.
³ Ludwig Cancer Research Oxford, University of Oxford, Oxford OX3 7DQ, UK.
⁴ MRC Weatherall Institute of Molecular Medicine, University of Oxford, Oxford OX3 9DS, UK; Centre for Human Genetics, University of Oxford, Oxford OX3 7BN, UK.
⁵ Nuffield Department of Orthopaedics, Rheumatology, and Musculoskeletal Sciences, University of Oxford, Oxford OX3 7LD, UK.
⁶ MRC Weatherall Institute of Molecular Medicine, University of Oxford, Oxford OX3 9DS, UK.
⁷ Oxford Centre for Diabetes, Endocrinology, and Metabolism, Rheumatology and Musculoskeletal Sciences, University of Oxford, Oxford OX3 7LE, UK.
⁸ MRC Weatherall Institute of Molecular Medicine, University of Oxford, Oxford OX3 9DS, UK; Department of Oncology, University of Oxford, Oxford OX3 9DS, UK. Electronic address: benjamin.fairfax@oncology.ox.ac.uk.

PMID: 38663408
PMCID: PMC11099345
DOI: 10.1016/j.xgen.2024.100541

Abstract

To better understand inter-individual variation in sensitivity of DNA methylation (DNAm) to immune activity, we characterized effects of inflammatory stimuli on primary monocyte DNAm (n = 190). We find that monocyte DNAm is site-dependently sensitive to lipopolysaccharide (LPS), with LPS-induced demethylation occurring following hydroxymethylation. We identify 7,359 high-confidence immune-modulated CpGs (imCpGs) that differ in genomic localization and transcription factor usage according to whether they represent a gain or loss in DNAm. Demethylated imCpGs are profoundly enriched for enhancers and colocalize to genes enriched for disease associations, especially cancer. DNAm is age associated, and we find that 24-h LPS exposure triggers approximately 6 months of gain in epigenetic age, directly linking epigenetic aging with innate immune activity. By integrating LPS-induced changes in DNAm with genetic variation, we identify 234 imCpGs under local genetic control. Exploring shared causal loci between LPS-induced DNAm responses and human disease traits highlights examples of disease-associated loci that modulate imCpG formation.

Keywords: DNA methylation; LPS; cancer; epigenetic aging; genetics; innate immune activation; mQTL; monocytes.

PubMed Disclaimer

Conflict of interest statement

Declaration of interests The authors declare no competing interests.

Figures

**Figure 1**
Monocyte DNAm is influenced at distinct sites by divergent immune stimuli, with LPS-induced changes involving hydroxymethylation formation (A) Principal-component analysis of 2,371 differentially methylated CpGs across any treatment (FDR $< 0.05$ ) shows that samples cluster according to treatment. (B) Examples of CpGs showing significant differential methylation with divergent treatments, including cg01882871 6.2 kb upstream of CCL2 (PAM3CysK4-specific demethylation), cg12762413 within the gene body of RHOU (IFNγ-specific demethylation), cg15912732 within the gene body of AKT1 (PAM3CysK4 and LPS methylation), and cg27409514 intronic to SBNO2 (pan-treatment demethylation). (C) Differentially methylated CpGs seen for each treatment (n = 11 individuals) summarized in Venn diagrams. (D) Density plot of median UT beta of CpGs from pilot samples (n = 27) comparing CpGs that are stable (n = 405,580) versus those that are either methylated (n = 384) or demethylated (n = 1,987) across any condition. (E) Variance of beta values for a CpG in UT samples according to the number of treatments under which it is observed to be modulated. CpGs not observed to change with treatment (n = 405,580) have reduced variance across the cohort compared to those modulated in one condition (n = 1,726, $p < 2.2 \times 10^{- 16}$ , Wilcoxon rank-sum test), 1 vs. 2 (n = 570) conditions ( $p = 0.003$ ), and 2 vs. 3 (n = 75) conditions ( $p = 0.006$ ). (F and G) Boxplots of all CpGs (F) significantly methylated with either 6- or 24-h LPS or (G) significantly demethylated at either 6 or 24 h of LPS. ∗∗ $p = 4.4 \times 10^{- 4}$ , ∗∗∗ $p < 2.2 \times 10^{- 16}$ (Wilcoxon signed-rank tests of either UT vs. 6 h LPS or 6 h LPS versus 24 h LPS). (H) Baseline percentage 5hmC methylation at demethylated immune-modulated CpGs, methylated immune-modulated CpGs, and across all sequenced CpGs (background), as assayed with paired bisulfite (BS) and oxidative BS (OxBS) sequencing. ∗∗∗ $p < 2.2 \times 10^{- 16}$ ; ns, not significant (Wilcoxon rank-sum tests). (I) Percentage total (left), 5mC (middle), and 5hmC (right) methylation at LPS-demethylated immune-modulated CpGs in the UT state and following 6 and 24 h of LPS stimulation, as assayed with paired BS and OxBS sequencing. ∗ $p < 0.05$ , ∗∗∗ $p < 2.2 \times 10^{- 16}$ (Wilcoxon signed-rank tests). In box-and-whisker plots, boxes depict the upper and lower quartiles of the data, and whiskers depict the range of the data excluding outliers (outliers are defined as data points $> 1.5 \times$ the interquartile range from the upper or lower quartiles).

**Figure 2**
Characterization of pan-cohort LPS-induced changes in monocyte DNAm denotes genomic enrichment, disease associations, and epigenetic age-acceleration (A) Manhattan plots of differentially methylated CpGs in response to LPS stimulation; shown are all CpGs (top), demethylated CpGs (blue, middle), and CpGs with gains of methylation (red, bottom). The p values represent moderated t tests of pairwise linear models. (B) Enrichment of significantly demethylated imCpGs (blue, top) and CpGs with significant gains of methylation in response to LPS stimulation (red, bottom) with ENCODE chromatin state segmentations. (C) Enrichment of demethylated imCpGs within promoter-captured enhancer DNA regions across 17 cell types. (D) Enrichment of imCpGs within TFBSs in K562 cells. (E and F) Reactome Immune System (E) and Disease Ontology (F) pathway analyses of genes proximal (within 5 kb) to LPS-demethylated CpGs in monocytes. All depicted pathways are significantly enriched (adjusted $p < 0.05$ ), with the most signficantly enriched pathways from each ontology plotted. Reactome Immune System, 15 of 57; Disease Ontology, 15 of 67. nOverlap, number of overlapping genes from a pathway. Enrichment is calculated using Fisher’s exact tests. (G) Correlation of methylation age vs. chronological age for monocytes (UT) from 92 individuals (male, red; female, blue). The p value represents Pearson’s product moment correlation. (H) Effect of LPS stimulation on methylation age (left). Shown are the relationship between baseline methylation age vs. chronological age (individuals dichotomized as methylation increased or reduced with respect to chronological age) and change in methylation age on LPS stimulation (right). The p values are calculated with Wilcoxon signed-rank tests. In box-and-whisker plots, boxes depict the upper and lower quartiles of the data, and whiskers depict the range of the data excluding outliers (outliers are defined as data points $> 1.5 \times$ the interquartile range from the upper or lower quartiles). Violin plots are trimmed to the range of the data.

**Figure 3**
Identification of correlates of differential CpG methylation in response to LPS (A) Relationship between baseline methylation and change in methylation induced by LPS for all imCpGs. ImCpGs are colored according to evidence of correlation between baseline methylation and LPS-induced methylation change: baseline-correlated (BC) imCpGs, blue; baseline-independent (BI) imCpGs, red. (B) Examples of BC and BI imCpGs for LPS-methylated and LPS-demethylated loci. (C) BC and BI imCpGS have distinct baseline methylation distributions (left) and distinct methylation responses to LPS stimulation (right). Distributions are compared with Kolmogorov-Smirnov tests. (D) Enrichment for overlap with TFBSs in K562 cells at BC and BI methylated and demethylated imCpGs. Significantly enriched TFBSs ( $F D R < 0.05$ ) are colored according to enrichment p-value (two-tailed binomial test). TFs not significantly associated with an imCpG group are colored gray. The top three enriched TFs for each imCpG group are labeled. (E) Enrichment for overlap with ENCODE chromatin state segmentations for BI and BC imCpGs. The p values are calculated by two-tailed binomial tests. Significantly enriched ( $F D R < 0.05$ ) chromatin states for each imCpG group are colored: BI, red; BC, blue.

**Figure 4**
Integration of genetic variation reveals local regulatory determinants of LPS-induced imCpGs (A) Manhattan plot of immune-modulated methylation quantitative trait loci (im-mQTLs) in LPS-stimulated monocytes (n = 234). Shown is QTL mapping of normalized differential methylation following LPS stimulation (delta, methylation beta LPS-treated monocytes minus methylation beta UT monocytes) in *cis* (within 100 kb of the CpG) at 7,359 imCpGs. im-mQTLs are colored according to the effect of LPS on CpG methylation (demethylation, n = 217, blue; increased methylation, n = 17, red). (B) *De novo* im-mQTL at cg02724909; correlation of rs869191 genotype with untreated (UT, green), LPS-stimulated (LPS, red), and LPS-induced differential (delta, purple) methylation at cg02724909. (C) Regional association plot of cg02724909 im-mQTL. Protein-coding genes are highlighted (blue). Bottom: recombination rate. (D) Baseline enhanced im-mQTL at cg17462560; correlation of rs6875879 genotype with UT (green), LPS-stimulated (LPS, red), and LPS-induced differential (delta, purple) methylation at cg17462560. (E) Regional association plot of cg17462560 im-mQTL. The p values are calculated by linear regression. For regional association and colocalization plots, SNPs are colored according to strength of linkage disequilibrium (LD) (Utah residents with ancestry from northern and western Europe [CEU] population) to the peak im-mQTL SNP. In box-and-whisker plots, boxes depict the upper and lower quartiles of the data, and whiskers depict the range of the data excluding outliers (outliers are defined as data points $> 1.5 \times$ the interquartile range from the upper or lower quartiles). The p values are calculated by linear regression.

**Figure 5**
LPS-induced im-mQTLs colocalize with hematological trait-associated loci (A) Baseline enhanced im-mQTL at cg19906672; correlation of rs6446553 genotype with UT (green) and LPS-stimulated (LPS, red) and LPS-induced differential (delta, purple) cg19906672 methylation. (B) Regional association plot of cg19906672 im-mQTL. Protein-coding genes are highlighted (blue). Bottom: recombination rate. (C) Colocalization plots of the cg19906672 im-mQTL with white blood cell count (WBCC), neutrophil count (Neut), neutrophil percentage (Neut%), lymphocyte percentage (Lymph%), mean corpuscular volume (MCV), and mean sphered corpuscular volume (MSCV) GWAS. PP4, posterior probability of a shared causal locus as calculated with Coloc. (D) Forest plot depicting effect estimates and 95% confidence intervals of rs6446553:G carriage WBCC, neutrophil count and Neut%, Lymph%, MCV, and MSCV. (E) Correlation of rs6446553 genotype with *TBC1D14* expression in unstimulated monocytes (UT, green) and 24 h of LPS stimulation (LPS, 24 h, red). The p values are calculated by linear regression. For regional association and colocalization plots, SNPs are colored according to strength of LD (CEU population) to the peak im-mQTL SNP. In box-and-whisker plots, boxes depict the upper and lower quartiles of the data, and whiskers depict the range of the data excluding outliers (outliers are defined as data points $> 1.5 \times$ the interquartile range from the upper or lower quartiles).

**Figure 6**
Identification of *trans* influences on LPS-induced imCpGs putatively mediated by expression of the insulator protein *ZFP57* (A) Baseline enhanced im-mQTL at cg16885113; correlation of rs3129058 genotype with UT (green) and LPS-stimulated (LPS, red) and LPS-induced differential (delta, purple) cg16885113 methylation. (B) Regional association plot of cg16885113 im-mQTL. Protein-coding genes are highlighted (blue). Bottom: recombination rate. (C) Colocalization plot of the cg16885113 im-mQTL with lung cancer. (D) Correlation of rs6446553 genotype with *ZFP57* expression in unstimulated monocytes (UT, green) and following 24 h of LPS stimulation (LPS, 24 h, red). (E) Regional association plot of *ZFP57* eQTL in UT monocytes (n = 414). Protein-coding genes are highlighted (blue). Bottom: recombination rate. (F) Circos plot depicting CpGs with methylation levels significantly modified by *ZFP57* eQTL genotype. Known imprinted regions are highlighted (orange). (G) Correlation of baseline CpG methylation levels and the effect of *ZFP57* eQTL genotype (rs417764) on methylation at CpGs significantly affected by rs417764 genotype (Pearson’s product moment correlation, n = 151, $r = - 0.33$ , $p = 3.0 \times 10^{- 5}$ ). (H) Distribution of methylation levels in UT monocytes at *ZFP57* eQTL-associated CpGs according to genotype at rs417764 (n = 151). (I) Correlation of *ZFP57* eQTL genotype (rs417764) with change in methylation at significantly rs417764-associated CpGs (n = 7). The p values are calculated by linear regression. For regional association and colocalization plots, SNPs are colored according to strength of LD (CEU population) to the peak im-mQTL/eQTL SNP. In box-and-whisker plots, boxes depict the upper and lower quartiles of the data, and whiskers depict the range of the data excluding outliers (outliers are defined as data points $> 1.5 \times$ the inter quartile range from the upper or lower quartiles).

See this image and copyright information in PMC

References

1. Furman D., Campisi J., Verdin E., Carrera-Bastos P., Targ S., Franceschi C., Ferrucci L., Gilroy D.W., Fasano A., Miller G.W., et al. Chronic inflammation in the etiology of disease across the life span. Nat. Med. 2019;25:1822–1832. doi: 10.1038/s41591-019-0675-0. - DOI - PMC - PubMed
1. Auffray C., Sieweke M.H., Geissmann F. Blood monocytes: Development, heterogeneity, and relationship with dendritic cells. Annu. Rev. Immunol. 2009;27:669–692. doi: 10.1146/annurev.immunol.021908.132557. - DOI - PubMed
1. Fairfax B.P., Humburg P., Makino S., Naranbhai V., Wong D., Lau E., Jostins L., Plant K., Andrews R., McGee C., Knight J.C. Innate immune activity conditions the effect of regulatory variants upon monocyte gene expression. Science. 2014;343 doi: 10.1126/science.1246949. - DOI - PMC - PubMed
1. Kim-Hellmuth S., Bechheim M., Pütz B., Mohammadi P., Nédélec Y., Giangreco N., Becker J., Kaiser V., Fricker N., Beier E., et al. Genetic regulatory effects modified by immune activation contribute to autoimmune disease associations. Nat. Commun. 2017;8:266. doi: 10.1038/s41467-017-00366-1. - DOI - PMC - PubMed
1. Quach H., Rotival M., Pothlichet J., Loh Y.-H.E., Dannemann M., Zidane N., Laval G., Patin E., Harmant C., Lopez M., et al. Genetic adaptation and neandertal admixture shaped the immune system of human populations. Cell. 2016;167:643–656.e17. doi: 10.1016/j.cell.2016.09.024. - DOI - PMC - PubMed

Publication types

Actions
Actions

MeSH terms

Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions

Substances

Actions

LinkOut - more resources

Full Text Sources

[1] Furman D., Campisi J., Verdin E., Carrera-Bastos P., Targ S., Franceschi C., Ferrucci L., Gilroy D.W., Fasano A., Miller G.W., et al. Chronic inflammation in the etiology of disease across the life span. Nat. Med. 2019;25:1822–1832. doi: 10.1038/s41591-019-0675-0. - DOI - PMC - PubMed

[2] Furman D., Campisi J., Verdin E., Carrera-Bastos P., Targ S., Franceschi C., Ferrucci L., Gilroy D.W., Fasano A., Miller G.W., et al. Chronic inflammation in the etiology of disease across the life span. Nat. Med. 2019;25:1822–1832. doi: 10.1038/s41591-019-0675-0. - DOI - PMC - PubMed

[3] Auffray C., Sieweke M.H., Geissmann F. Blood monocytes: Development, heterogeneity, and relationship with dendritic cells. Annu. Rev. Immunol. 2009;27:669–692. doi: 10.1146/annurev.immunol.021908.132557. - DOI - PubMed

[4] Auffray C., Sieweke M.H., Geissmann F. Blood monocytes: Development, heterogeneity, and relationship with dendritic cells. Annu. Rev. Immunol. 2009;27:669–692. doi: 10.1146/annurev.immunol.021908.132557. - DOI - PubMed

[5] Fairfax B.P., Humburg P., Makino S., Naranbhai V., Wong D., Lau E., Jostins L., Plant K., Andrews R., McGee C., Knight J.C. Innate immune activity conditions the effect of regulatory variants upon monocyte gene expression. Science. 2014;343 doi: 10.1126/science.1246949. - DOI - PMC - PubMed

[6] Fairfax B.P., Humburg P., Makino S., Naranbhai V., Wong D., Lau E., Jostins L., Plant K., Andrews R., McGee C., Knight J.C. Innate immune activity conditions the effect of regulatory variants upon monocyte gene expression. Science. 2014;343 doi: 10.1126/science.1246949. - DOI - PMC - PubMed

[7] Kim-Hellmuth S., Bechheim M., Pütz B., Mohammadi P., Nédélec Y., Giangreco N., Becker J., Kaiser V., Fricker N., Beier E., et al. Genetic regulatory effects modified by immune activation contribute to autoimmune disease associations. Nat. Commun. 2017;8:266. doi: 10.1038/s41467-017-00366-1. - DOI - PMC - PubMed

[8] Kim-Hellmuth S., Bechheim M., Pütz B., Mohammadi P., Nédélec Y., Giangreco N., Becker J., Kaiser V., Fricker N., Beier E., et al. Genetic regulatory effects modified by immune activation contribute to autoimmune disease associations. Nat. Commun. 2017;8:266. doi: 10.1038/s41467-017-00366-1. - DOI - PMC - PubMed

[9] Quach H., Rotival M., Pothlichet J., Loh Y.-H.E., Dannemann M., Zidane N., Laval G., Patin E., Harmant C., Lopez M., et al. Genetic adaptation and neandertal admixture shaped the immune system of human populations. Cell. 2016;167:643–656.e17. doi: 10.1016/j.cell.2016.09.024. - DOI - PMC - PubMed

[10] Quach H., Rotival M., Pothlichet J., Loh Y.-H.E., Dannemann M., Zidane N., Laval G., Patin E., Harmant C., Lopez M., et al. Genetic adaptation and neandertal admixture shaped the immune system of human populations. Cell. 2016;167:643–656.e17. doi: 10.1016/j.cell.2016.09.024. - DOI - PMC - PubMed

Save citation to file

Email citation

Add to Collections

Add to My Bibliography

Your saved search

Create a file for external citation management software

Your RSS Feed

Characterization of the genetic determinants of context-specific DNA methylation in primary monocytes

Affiliations

Characterization of the genetic determinants of context-specific DNA methylation in primary monocytes

Authors

Affiliations

Abstract

Conflict of interest statement

Figures

Similar articles

References

Publication types

MeSH terms

Substances

LinkOut - more resources

Full Text Sources

Abstract

Conflict of interest statement

Figures

Similar articles

References

Publication types

MeSH terms

Substances

Related information

LinkOut - more resources

Full Text Sources