As one of the most stable and deeply studied regulatory mechanisms in epigenetics, DNA methylation plays a central role in the regulation of the temporal and spatial expression of the human genome, the determination of cell fate, and the occurrence and development of diseases. It precisely regulates the opening and closing of genes through chemical modification of specific CpG loci in the genome without changing DNA sequence, and then affects a series of life activities such as cell proliferation, differentiation, and apoptosis. With the rapid development of high-throughput sequencing technology, a comprehensive analysis of genome methylation profiles has become a key breakthrough to reveal the molecular mechanism of diseases and explore new biomarkers.

Under this background, Illumina Infinium Methylation Epic Beadchip IP (935K chip) relies on its high coverage of more than 935,000 methylation sites in the human genome-covering 99% of the RefSeq gene, 95% of CpG islands, a large number of enhancers, CTCF binding sites, and other regulatory regions. As well as the quantitative detection ability of single-base resolution, it has become an indispensable core tool in the research of epigenome association (EWAS), disease mechanism exploration, and clinical transformation. Its excellent accuracy, repeatability, and wide sample compatibility (including FFPE tissue, blood, saliva, etc.) not only provide strong support for analyzing the epigenetic regulatory network of complex diseases, but also build a key technical bridge from basic research to clinical application.

The article presents cases where the Illumina 935K DNA Methylation Array is applied in studying epidemic diseases and retinal macular disease by identifying gene-epigenome interactions in age-related macular degeneration.

Evaluating the Relationship Between Phenotypes of Diseases and DNA Methylation

Title: Blood-based epigenome-wide analyses of 19 common disease states: A longitudinal, population-based linked cohort study of 18,413 Scottish individuals

Publish Magazine: PLoS Med

Impact Factors: 10.5

Publication Time: 2023.07.06

DOI: https://doi.org/10.1371/journal.pmed.1004247

As a dynamic epigenetic modification, DNA methylation regulates gene expression by adding methyl groups at CpG sites, which is the key medium for the interaction between the environment and genome. EWAS provides an important perspective to analyze the biological mechanism of common diseases by analyzing the relationship between the methylation level of CpG loci and health outcomes. However, in the past, EWAS focused on a single disease, with a limited sample size and a lack of systematic analysis across diseases, which made it difficult to reveal the commonalities and specificities of blood methylation as a peripheral marker in many diseases. In addition, the existing research exhibits significant heterogeneity in phenotypic definition, statistical models, and result reporting, which leads to low repeatability of the results.

Based on this, this study takes 18,413 Scottish people as the object, and systematically explores the relationship between blood DNA methylation and the prevalence and incidence of 19 common diseases through epigenome analysis of whole blood samples, aiming at discovering new disease-related methylation sites, and evaluating the consistency of existing EWAS results through literature review, so as to provide a basis for the application of blood methylation as a disease marker.

Study design for epigenome-wide analyses on prevalent and incident disease states in Generation Scotland (Hillary et al., 2023)

Demographics and Disease Counts in Generation Scotland

The research is based on the Generation Scotland queue, which is a large and representative crowd queue in Scotland. The research team conducted an in-depth analysis of the whole blood samples of 18,413 participants, with an average age of 47.5 years, of whom 58.8% were women. The gender and age distribution fully considered the representative requirements of epidemiological research. In terms of technical means, the Illumina Infinium Methylation Epic chip is used. As an industry-leading methylation detection tool, the chip can accurately detect the methylation level of 752,722 CpG loci, providing a rich and high-quality database for subsequent analysis.

The design of this study adopts the two-dimensional analysis strategy of combining cross-sections with longitudinal direction. In the cross-sectional analysis, with the help of baseline self-reported data, the researchers systematically evaluated the association between blood methylation and 14 common chronic diseases, including breast cancer, type 2 diabetes, coronary heart disease, asthma, and rheumatoid arthritis, in order to reveal the methylation characteristics of diseases in the current state.

The longitudinal analysis links Scotland's comprehensive medical record system (data as of October 2020) and tracks the participants for a long time, covering 19 disease types, especially including COVID-19 hospitalization, long-term COVID, and other emerging health problems. There is a significant difference in the follow-up period, with the average follow-up time ranging from 5.0 to 11.7 years, which ensures that the dynamic changes of methylation level during the occurrence and development of diseases can be captured and provides a key basis for disease prediction and early intervention.

Epigenome-wide association studies on 14 prevalent disease states in GenerationScotland (Hillary et al., 2023)

Epigenome-wide Analyses of Prevalent Disease

In the initial basic model, through the in-depth analysis of Scottish population samples, 1,340 loci with potential association with 10 common diseases were successfully identified, which covered endocrine, tumor, and other medical fields. Subsequently, a fully adjusted model was used for more rigorous screening, which comprehensively considered the confounding factors such as age, gender, and lifestyle, and finally retained 78 strong association loci involving 8 diseases. It is noteworthy that 69 association loci exist stably in both the basic model and the fully adjusted model, and they mainly involve four diseases, such as type 2 diabetes and breast cancer. From the point of genome distribution, most of these loci are located in the "open sea" region outside CpG islands, which implies that non-traditional regulatory regions may play an important role in the occurrence and development of diseases.

Epigenome-wide association studies on 19 incident disease states in GenerationScotland (Hillary et al., 2023)

Epigenome-wide Analyses on Incident Disease

Under the basic model, the research team identified 14,237 potential associations with 11 diseases, showing the complexity of the association between blood methylation and diseases. After the optimization of the fully adjusted model, 79 statistically significant association loci were retained, corresponding to 5 diseases. Among them, 64 overlapping association loci are highly concentrated in chronic obstructive pulmonary disease (COPD) and type 2 diabetes, involving key genes such as ALPG and TXNIP. These genes have been proven to be involved in important physiological processes such as inflammatory reaction and glucose metabolism regulation, suggesting that blood methylation may affect the risk of diseases by regulating the expression of these genes.

The research team used two key indicators, odds ratio (OR) and risk ratio (HR), to quantitatively evaluate the correlation between blood methylation sites and disease risk. Taking breast cancer as an example, the methylation level of locus cg06072257 is negatively correlated with the risk of breast cancer; that is, the higher the methylation level of this locus, the lower the possibility of breast cancer, which provides a potential epigenetic marker for early warning of breast cancer.

Blood CpGs associated with prevalent or incident disease states showing effect sizeon interpretable scale (Hillary et al., 2023)

Structured Review on Existing Epigenome-wide Analyses

The researchers systematically combed 69 prevalence-related studies and 64 incidence-related studies, and found that only 11 prevalence-related studies and 8 incidence-related studies were verified by other independent studies, and the cross-study repeatability was generally low. Among them, the repeatability of lung cancer-related research is the highest, only reaching 16.8%, which indicates that there is significant heterogeneity between different cohorts in the current research on methylation and disease, and it is urgent to verify this with larger and standardized research.

In addition, the research team used co-location analysis technology to explore whether methylation sites and diseases share genetic variation mechanisms. The results show that there is no strong evidence to support the common genetic basis between them, suggesting that the influence of methylation on diseases may be independent of traditional genetic factors. In order to ensure the reliability of the research results, the sensitivity analysis was further carried out, and the robustness of the core conclusion was verified by correcting the genetic relationship of the population and adjusting the potential confounding factors, which provided a solid data foundation for the follow-up research.

Look-up and replication analyses within EWAS on common disease states (Hillary et al., 2023)

This study found more than 100 new associations between blood CpG and common disease states, which were independent of the main confounding risk factors. It further shows that blood DNA methylation can be used as a marker of common diseases, which lays a foundation for the future study of biomarkers.

Discovering the Interaction Between DNA Methylation and Macular Disease

Title: QTL mapping of human retina DNA methylation identifies 87 gene-epigenome interactions in age-related macular degeneration

Publish Magazine: Nat Commun

Impact Factors: 14.7

Publication Time: 2024.03.04

DOI: https://doi.org/10.1038/s41467-024-46063-8

Age-related macular degeneration (AMD) is the main cause of irreversible vision loss in the elderly, and its pathogenesis involves the complex interaction of genetic and environmental factors. Many AMD-related genetic variants have been found in GWAS, but most of them are located in non-coding regions. The specific mechanism of their effects on diseases through epigenetic regulation is not clear. As a key epigenetic marker linking genetic variation with environmental factors, DNA methylation may play a central role in the pathogenesis of AMD. However, the previous studies on epigenetics of retinal tissue were limited, and there was a lack of systematic analysis on the relationship between methylation, gene expression, and genetic variation.

The purpose of this study is to reveal how genetic variation affects gene expression by regulating methylation, and then participate in the pathogenesis of AMD, and provide new insights into the molecular mechanism of AMD. In this study, 160 human retinal samples (with an average age of 73 years and balanced gender distribution) were included, and the methylation level of 749,158 CpG loci was systematically detected by Illumina Infinium Methylation Epic chip. At the same time, RNA-seq data and more than 8 million genetic variation information were integrated to construct a complete integration analysis process.

Graphic summary of datasets generated, integrated and analyses performed in thepresent study and robust identification of retina mOTLs (Advani et al., 2024)

Genomic Features and Biological Processes Enriched in Retina

Based on the systematic analysis of cis-methylation quantitative trait loci (cis-mQTL) and cis-expression quantitative trait loci (cis-eQTL), 37,453 mQTLs and 12,505 eQTLs were identified, of which about one-third showed retinal-specific expression characteristics. The distribution of these loci in the genome has a significant preference, mainly concentrated in the upstream and downstream regions of transcription initiation sites (TSS200, TSS1500) and genome regions. Further functional element annotation shows that mQTL is statistically significantly enriched in functional regulatory elements such as open chromatin regions and enhancers.

Taking PARK7 gene as an example, the encoded product of the gene is involved in the regulation of mitochondrial function, and five mQTL loci were detected in its genome region, among which the single nucleotide polymorphism of rs7517357 was significantly correlated with the DNA methylation level of the corresponding loci (r²=0.68, p = 2.3× 10), suggesting that the genetic variation may regulate the oxidative stress response pathway of retinal cells through epigenetic modification mechanism.

Characterization and distribution of retina eQTMs (Advani et al., 2024)

Relationships Between DNAm Regulation and Gene Expression

Secondly, through in-depth analysis of methylation and gene expression association (eQTM), a total of 13,747 statistically significant association sites were identified. Among them, 54.5% showed negative correlation (such as cg24846343 and GSTT2B gene), while 45.5% showed positive correlation (such as cg24307499 and NLRP2 gene). Further spatial distribution analysis showed that these eQTM loci were significantly clustered near the gene transcription initiation site (TSS), with a median distance of 1.07 kb from TSS. Chromosome mapping studies showed that the above-mentioned associated loci were significantly enriched on chromosomes 16 and 19. Functional enrichment analysis indicated that these loci are involved in autophagy regulation, protein degradation, and other important biological pathways.

Associations between retina DNA methylation and gene expression throughgenotypes (Advani et al., 2024)

SMR and Colocalization of Retina mQTLs and eQTLs

The complex regulatory network of "genetic variation-methylation-gene expression-AMD" was systematically analyzed by the frontier statistical methods such as Mendel randomization (SMR) and co-location analysis (coloc, MOOC). The methylation site cg24506221 of the GSTM1 gene was negatively correlated with gene expression (R=-0.74), and the correlation was verified by strict SMR (p = 4.6×10), suggesting that epigenetic regulation of the glutathione metabolic pathway may play an important role in the pathological process of AMD.

Further co-location analysis found that methylation and expression changes of 87 genes may be used as intermediaries to mediate the influence of genetic variation on AMD. These genes are widely involved in immune response, glycolysis, and other important biological pathways. In the immune response pathway, abnormal methylation modification may affect the activation of immune cells and the expression of inflammatory factors, and aggravate the inflammatory microenvironment of the retina. The epigenetic changes of genes related to the glycolysis pathway may interfere with the energy metabolism of retinal cells and affect their function and viability. These findings provide a new perspective and potential therapeutic target for understanding the pathogenesis of AMD.

Associations between retina DNA methylation, gene expression and AMD GWAS through genotypes (Advani et al., 2024)

Colocalization Analysis of Retina mQTLs, eQTLs and GWAS

Finally, the regulatory relationship was verified by integrating retinal Hi-C data (chromatin ring, super enhancer, etc.). The results showed that the mQTL (methylation quantitative trait loci) /eQTL (expression quantitative trait loci) of ALDH2 and GSTP1 genes could be physically linked to the target genes through chromatin rings. Using Hi-C thermogram and chromosome conformation capture data, the chromatin ring structure formed between ALDH2 and GSTP1 gene sites and downstream target genes was clearly presented, and the cyclization region was highly overlapped with mQTL/eQTL sites, which directly verified the molecular mechanism of remote regulation of regulatory elements through three-dimensional conformation. This result not only provides a structural biological basis for understanding the epigenetic regulatory network in the pathogenesis of AMD, but also provides a potential target for the subsequent drug development targeting the three-dimensional structure of chromatin.

Colocalization analysis among AMD GWAS and retina mOTL and eQTL using Coloc and Moloc (Advani et al., 2024)

Integrative Analysis with Hi-C Retina Map

In this study, the molecular mechanism of genetic variation participating in the pathological process of age-related macular degeneration (AMD) through epigenetic regulation network was first clarified through systematic exploration. The research results not only confirmed the application value of blood DNA methylation profile as a non-invasive biomarker in early warning of AMD, but also provided key theoretical basis and experimental support for the development of precise medical strategies targeting epigenetic modification.

Hi-C retina data enables target gene and variant prioritization (Advani et al., 2024)

Conclusion

To sum up, the Illumina 935K methylation chip accurately analyzes the DNA methylation pattern from the whole genome level with its extensive coverage of over 935,000 CpG sites. Both in detecting the methylation difference between tumor and normal samples and in genome association analysis of complex diseases, high-resolution and high-reliability data are provided.

Looking forward to the future, with the continuous innovation of technology, Illumina 935K methylation chip is expected to be deeply applied in more disease fields and more sample types, further expanding our knowledge of DNA methylation regulation network, pushing life science research and precision medicine practice to a new height, and contributing to overcoming more medical problems.

The DNA Methylation Array (Illumina 935K) introduced by CD Genomics is also based on Qualcomm design, which can accurately quantify the level of DNA methylation and support the analysis of multiple sample types, such as blood and tissues, and provide reliable data support for researchers, but it does not involve clinical diagnosis or treatment services.

References

• Hillary RF, McCartney DL, Smith HM, et al. "Blood-based epigenome-wide analyses of 19 common disease states: A longitudinal, population-based linked cohort study of 18,413 Scottish individuals." PLoS Med. 2023 20(7): e1004247 https://doi.org/10.1371/journal.pmed.1004247
• Advani J, Mehta PA, Hamel AR, et al. "QTL mapping of human retina DNA methylation identifies 87 gene-epigenome interactions in age-related macular degeneration." Nat Commun. 2024 15(1): 1972 https://doi.org/10.1038/s41467-024-46063-8